INDUSTRIAL COMBUSTION TESTING
© 2011 by Taylor and Francis Group, LLC
INDUSTRIAL COMBUSTION TESTING Charles E. Baukal...
196 downloads
1809 Views
24MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
INDUSTRIAL COMBUSTION TESTING
© 2011 by Taylor and Francis Group, LLC
INDUSTRIAL COMBUSTION TESTING Charles E. Baukal, Jr.
Boca Raton London New York
CRC Press is an imprint of the Taylor & Francis Group, an informa business
© 2011 by Taylor and Francis Group, LLC
CRC Press Taylor & Francis Group 6000 Broken Sound Parkway NW, Suite 300 Boca Raton, FL 33487-2742 © 2011 by Taylor and Francis Group, LLC CRC Press is an imprint of Taylor & Francis Group, an Informa business No claim to original U.S. Government works Printed in the United States of America on acid-free paper 10 9 8 7 6 5 4 3 2 1 International Standard Book Number: 978-1-4200-8528-0 (Hardback) This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to publish reliable data and information, but the author and publisher cannot assume responsibility for the validity of all materials or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint. Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging-in-Publication Data Industrial combustion testing / edited by Charles E. Baukal, Jr. p. cm. “A CRC title.” Includes bibliographical references and index. ISBN 978-1-4200-8528-0 (alk. paper) 1. Furnaces--Testing. 2. Furnaces--Combustion. 3. Furnaces--Industrial applications. I. Baukal, Charles E. TH7140.I47 2010 621.402’3--dc22 Visit the Taylor & Francis Web site at http://www.taylorandfrancis.com and the CRC Press Web site at http://www.crcpress.com
© 2011 by Taylor and Francis Group, LLC
2010014467
Contents Preface..........................................................................................................................................................................................ix Editor...........................................................................................................................................................................................xi Contributors............................................................................................................................................................................. xiii
Section I General 1. Introduction......................................................................................................................................................................... 3 Charles E. Baukal, Jr. 2. Testing Safety.................................................................................................................................................................... 41 Charles E. Baukal, Jr. 3. Experimental Design....................................................................................................................................................... 63 Joseph Colannino 4. Fluid Flow.......................................................................................................................................................................... 77 Wes Bussman and Joseph Colannino 5. Temperature....................................................................................................................................................................... 97 Charles E. Baukal, Jr. 6. Heat Flux............................................................................................................................................................................117 Charles E. Baukal, Jr. 7. Pollution Emissions........................................................................................................................................................141 Charles E. Baukal, Jr. 8. Combustion Noise.......................................................................................................................................................... 183 Mahmoud M. Fleifil, Carl-Christian Hantschk, and Edwin Schorer 9. Flame Impingement Measurements............................................................................................................................211 Charles E. Baukal, Jr. 10. Physical Modeling in Combustion Systems............................................................................................................. 241 Christopher Q. Jian 11. Virtual Testing................................................................................................................................................................ 251 Eddy Chui, Allan M. Runstedtler, and Adrian J. Majeski
Section I I Advanced Diagnostics 12. Laser Measurements...................................................................................................................................................... 269 Michele Marrocco and Guido Troiani 13. CARS Temperature Measurements in Flames in Industrial Burners................................................................. 289 Patrick M. Hughes, Thangam Parameswaran, and Richard J. Lacelle
v © 2011 by Taylor and Francis Group, LLC
vi
Contents
14. Diode Laser Temperature Measurements..................................................................................................................311 Thomas P. Jenkins and John L. Bergmans 15. Image-Based Techniques for the Monitoring of Flames........................................................................................ 337 Javier Ballester and Ricardo Hernández 16. High Temperature Cameras......................................................................................................................................... 355 William J. Lang 17. Liquid Fuel Atomization Testing................................................................................................................................ 369 Khaled A. Sallam
Section II I Burner Testing 18. Process Burners............................................................................................................................................................... 377 Jeffrey Lewallen, Thomas M. Korb, Jaime A. Erazo, Jr., and Erwin Platvoet 19. Commercial Boiler Burners.......................................................................................................................................... 395 Yaroslav Chudnovsky and Mikhail Gotovsky 20. Power Burners..................................................................................................................................................................411 Vit Kermes, Petr Beˇ lohradský, Petr Stehlík, and Pavel Skryja 21. Regenerative Combustion Using High Temperature Air Combustion Technology (HiTAC)........................ 429 Ashwani K. Gupta, Susumu Mochida, and Tsutomu Yasuda 22. Characterization of Ribbon Burners.......................................................................................................................... 449 Colleen Stroud Alexander and Melvyn C. Branch 23. Flameless Burners.......................................................................................................................................................... 471 Joachim G. Wünning and Ambrogio Milani 24. Radiant Tube Burners.................................................................................................................................................... 487 Michael Flamme, Ambrogio Milani, Joachim G. Wünning, Wlodzimierz Blasiak, Weihong Yang, Dariusz Szewczyk, Jun Sudo, and Susumu Mochida 25. Metallic Mat Gas Combustion..................................................................................................................................... 505 Giuseppe Toniato, Andrea Zambon, and Andrea D’Anna 26. Performance Prediction of Duct Burner Systems via Modeling and Testing.................................................... 517 Steve Londerville 27. Oxy-Fuel and Oxygen-Enhanced Burner Testing.................................................................................................... 529 Lawrence E. Bool, III, Nicolas Docquier, Chendhil Periasamy, and Lee J. Rosen
Section I V Flare Testing 28. Large-Scale Flare Testing.............................................................................................................................................. 553 Charles E. Baukal, Jr., Jianhui Hong, Roger Poe, and Robert Schwartz 29. Flare Experimental Modeling...................................................................................................................................... 571 Chendhil Periasamy and Subramanyam R. Gollahalli
© 2011 by Taylor and Francis Group, LLC
Contents
vii
30. Flare Radiation................................................................................................................................................................ 595 Wes Bussman and Jianhui Hong
Section V Testing in Combustors 31. Cement Kilns....................................................................................................................................................................615 Eugen Dan Cristea and Givanni Cinti 32. Glass Furnaces................................................................................................................................................................. 671 R. Robert Hayes and Charles E. Baukal, Jr. 33. Thermal Oxidizer Testing............................................................................................................................................ 691 Bruce C. Johnson and Nate Petersen 34. Utility Boilers.................................................................................................................................................................. 705 Giuseppe Toniato and Silvio Rudi Stella Appendix A: F-Distribution (99%, 95%, & 90% Confidence)........................................................................................ 729 Appendix B: EPA Sample Methods.................................................................................................................................... 733 Appendix C: Common Conversions.................................................................................................................................. 735 Index......................................................................................................................................................................................... 737
© 2011 by Taylor and Francis Group, LLC
Preface This book is intended to fill a gap in the literature for books on industrial combustion testing. It should be of interest to anyone working in or with the field of industrial combustion. This includes burner and furnace designers, researchers, end users, government regulators, and funding agencies. It can also serve as a reference work for those teaching and studying combustion. The book covers a wide range of testing techniques used in a broad array of applications in the metals, minerals, thermal oxidation, hydrocarbon/petrochemical, and power generation industries. There are 61 authors from 10 countries representing 33 prominent combustion organizations, and these authors have hundreds of years of combined experience with industrial combustion testing. The book contains 34 chapters divided into five sections. Section I is a general section with 11 chapters: Introduction, Testing Safety, Experimental Design, Fluid Flow, Temperature, Heat Flux, Pollution Emissions, Combustion Noise, Flame Impingement Measurements, Physical Modeling in Combustion Systems, and Virtual Testing. It is designed to provide some of the basic information referenced in succeeding chapters. Section II contains six chapters on advanced diagnostics: Laser Measurements, CARS Temperature Measurements in Flames in Industrial Burners, Diode Laser Temperature Measurements, Image-Based Techniques for the Moni toring of Flames, High Temperature Cameras, and Liquid Fuel Atomization Testing. Section III has ten chapters on burner testing: Process Burners, Commercial Boiler Burners, Power Burners, Regenerative Combustion Using High Temperature Air Combustion Technology (HiTAC), Characterization of Ribbon Burners, Flameless Burners, Radiant Tube Burners, Metallic Mat Gas Combustion, Performance Prediction of Duct Burner Systems Via Modeling and Testing, and Oxy-Fuel and Oxygen-Enhanced Burner Testing. Section IV has three chapters on flare testing: Large-Scale Flare Testing, Flare Experimental Modeling, and Flare Radiation. Section V has four chapters on testing in combustors: Cement Kilns, Glass Furnaces, Thermal Oxidizer Testing, and Utility Boilers. The purpose of this work is to compile testing techniques utilized in industrial combustion for use by practitioners. No such book currently exists, which means that those in this field must consult a range of sources such as journals, magazines, and conference proceedings to get this kind of information. This is generally impractical because those practicing in the field usually do not have the time or the resources to
extensively research this topic. While academics have access to the information, they generally do not work at the large scales associated with industrial combustion and therefore may not be familiar with how techniques are applied in production applications. This book is designed to help practitioners both in the field and in academics. Besides providing a single-source reference, this book also provides information for specific applications. This means that someone practicing in a particular area can immediately go to their application, without necessarily having to read through other chapters. They can determine for themselves what is useful for them. The reader can save time and more quickly use the information provided by experts in each area. Nearly 1300 references and over 800 figures, and 140 tables are provided for those that need further information on a particular topic. The book provides case studies and examples to show how to apply the information for particular applications. This includes identifying potential problems that could be very costly if not avoided. For example, failure to properly measure pollution emissions could lead to large fines from regulatory agencies. The book is designed to be more hands-on and less theoretical so the information can be easily applied to real situations in a variety of industries. This book tells the reader how to make measurements and conduct tests in industrial combustion systems including full-scale burners, furnaces, heaters, boilers, flares, and thermal oxidizers. There are some topics that are not covered and some that are not treated extensively. Since the majority of industrial applications use gaseous fuels, there is more treatment of that type of fuel, with less discussion of liquid and solid fuels. This book concerns atmospheric pressure combustion, which is the predominant type used in most industrial applications. There are some burner designs, combustors, and applications that are not considered. As with any book of this type, there are sure to be author preferences and biases, but the coverage is fairly extensive and comprehensive. There are also generous discussions of many common industrial applications to help the reader better understand the requirements for different types of tests. Particularly because of the increasing emphasis on the environment, most industrial tests include some type of pollution emission measurements. While industrial combustion testing is a dynamic area of continuing research, the principles considered here are expected to be applicable well into the foreseeable future. ix
© 2011 by Taylor and Francis Group, LLC
Editor Charles E. Baukal, Jr., PhD, is the Director of the John Zink Institute for the John Zink Co., LLC (Tulsa, Oklahoma) where he has been since 1998. He has also been the Director of Research and Development and the Director of the Research and Development Test Center at Zink, which is a leading supplier of industrial combustion equipment to a variety of industries. Previously, Dr. Baukal worked for 13 years at Air Products and Chemicals, Inc. (Allentown, Pennsylvania) in the areas of oxygen-enhanced combustion and rapid gas quenching in the ferrous and nonferrous metals, minerals, and waste incineration industries. He worked for Marsden, Inc. (a burner supplier in Pennsauken, New Jersey) for five years in the paper, printing, and textile industries, and Selas Corp. (a burner supplier in Dresher, Pennsylvania) in the metals industry, both in the area of industrial combustion equipment. He has 30 years of experience in the fields of industrial combustion, pollution control, and heat transfer and has authored more than 100 publications in those areas. Dr. Baukal is an adjunct instructor for Oral Roberts University and the University of Tulsa, both in Tulsa, Oklahoma. He is the author or editor of seven books in the field
of industrial combustion including: Oxygen-Enhanced Combustion (1998), Heat Transfer in Industrial Combustion (2000), Computational Fluid Dynamics in Industrial Combustion (2001), The John Zink Combustion Handbook (2001), Industrial Combustion Pollution and Control (2004), Handbook of Industrial Burners (2004), and Heat Transfer from Flame Impingement Normal to a Plane Surface (2009). Dr. Baukal has a PhD in mechanical engineering from the University of Pennsylvania (Philadelphia, Pennsylvania) and is a licensed Professional Engineer in the state of Pennsylvania, a Board Certified Environ mental Engineer, and a Qualified Environmental Professional. He has served as an expert witness in the field of combustion, has 11 U.S. patents, and is a member of numerous honorary societies and Who’s Who compilations. He is a member of the American Society of Mechanical Engineers, the Air and Waste Management Association, the Combustion Institute, and the American Society for Engineering Education. He serves on several advisory boards, holds offices in the Air and Waste Management Association and the American Society for Engineering Education, and is a reviewer for combustion, heat transfer, environmental, and energy journals.
xi © 2011 by Taylor and Francis Group, LLC
Contributors Colleen Stroud Alexander, PhD, received her doctorate in mechanical engineering from the University of Colorado at Boulder. Her research focused on the heat transfer, fluid flow, and chemical kinetics involved in methane-air combustion flame treatment processes. Her PhD research efforts resulted in five technical publications in multiple peer-reviewed journals. She carried out her postdoctoral work as a guest researcher at the National Institute for Standards and Technology in Gaithersburg, Maryland, performing both experimental and numerical analysis in the study of controlled combustion reactions within reacting flows. She most recently worked as a research engineer studying the performance of advanced fuels in various combustion regimes at the National Renewable Energy Laboratory in Golden, Colorado. Prior to receiving her degree, Dr. Alexander also worked in the information technologies sector as a project manager and technical support engineer at CSG Systems (Englewood, Colorado). She also worked as a pneumatics engineer supporting the assembly of the Atlas Centaur Rocket at Lockheed Martin in Denver, Colorado. Javier Ballester, PhD, is currently a professor in fluid mechanics at the University of Zaragoza (Spain), where he has been since 1997. He received his degree in electrical engineering from the University of Zaragoza in 1992. Previously, he was hired as a researcher at the Laboratory of Research on Combustion Technologies by the Technological Institute of Aragon (1991–1992) and by the Spanish Council of Scientific Research (1992–1997). His areas of expertise are fluid mechanics and combustion, and his current research interests include the combustion of solid fuels, advanced monitoring and control of industrial flames, and combustion instabilities. He has three patents and has authored over 90 papers in international journals and conferences. He has participated, in most cases as principal investigator, in more than 90 research projects, including contracts with private companies and projects funded by the Spanish and European administrations. Petr Beˇlohradský, MS, is currently a postgraduate student and he works as a technician at the Institute of Process and Environmental Engineering at Brno University of Technology (Czech Republic). He holds his degree in mathematical engineering from Brno University of Technology. His work is directed toward the research of combustion on gaseous fuels with special focus on modeling by using statistical methods and computational fluid dynamic methods. He is an author or co-author of several papers related to combustion modeling presented at international conferences. John L. Bergmans, MEng, is the principal engineer and owner of Bergmans Mechatronics LLC (Newport Beach, California). Bergmans received his degree in mechanical engineering from Carleton University (Ottawa, Canada) in 1995. He founded Bergmans Mechatronics in 2003 and has since developed data acquisition and control systems for several rocket motor test stands and an oxyfuel combustion system. Bergmans is also active in the development and testing of tunable-diode, laser-based instrumentation for large-scale combustion applications. Prior to founding BML, Bergmans was employed for eight years by CFD Research Corp. (Huntsville, Alabama), where he developed closed-loop pressure controllers for solid-propellant rocket and air-breathing propulsion systems. Wlodzimierz Blasiak, PhD, is head and professor in the Division of Energy and Furnace Technology, Royal Institute of Technology, Sweden. He has his degree of applied thermodynamics from Technical University of Czestochowa (Poland). He has carried out research on heat and mass transfer processes in boilers and furnaces and published around 200 papers since 1993. For the last ten years the main research themes of his work are high performance industrial furnaces, high temperature air combustion-HiTAC/flameless combustion for gaseous and solid fuel, high temperature air/steam gasification of biomass and wastes-HTAG, oxyfuel, and flameless oxyfuel combustion. He also carried out and managed many research projects financed by Swedish and international agencies in cooperation with European and Japanese industry. He has four patents (three of them are PCT, and two of them are U.S. provisional-pending) on solid fuel thermal conversions. Lawrence E. Bool III, PhD, is a senior development associate in the combustion research and development group for Praxair, Inc. (Tonawanda, New York) where he has been since 1997. Dr. Bool received his doctorate in chemical engineering from the University of Arizona in 1993. His work focuses on using basic science to develop new xiii © 2011 by Taylor and Francis Group, LLC
xiv
Contributors
oxyfuel applications for industry. Recent examples include a novel process to reduce pollutants from power plants and a process to produce activated carbon. Dr. Bool holds 20 patents and has authored several peer-reviewed publications. Melvyn C. Branch, MS, PhD, is the Joseph Negler Professor of Mechanical Engineering, Emeritus at the University of Colorado at Boulder. He received his degrees in mechanical engineering from the University of California at Berkeley. He has previously served as associate dean of Engineering for Research and Administration, associate dean of the graduate school, and Director of the Center for Combustion Research. He has taught graduate and undergraduate courses on combustion fundamentals, fluid mechanics, heat transfer, applied thermodynamics, and fuel technology. His research activity in these areas includes experimental and theoretical studies of combustion-generated air pollutants, fuel efficiency, flame processing, metal burning, and aircraft and rocket combustion. His recent consulting activity includes the 3M Company, the Combustion Research Division of Sandia National Laboratories, the Air Pollution Control Division of the State of Colorado and the U.S. Federal Trade Commission. Dr. Branch has served as a member and Chair of the Colorado Air Quality Control Commission, the state agency responsible for promulgating state regulations relating to air quality, and as a member of the Research Committee of the Health Effects Institute. He is a member of the Combustion Institute, Tau Beta Pi, Pi Tau Sigma, and a Fellow of the American Society of Mechanical Engineers. He is a past chairman of the Western States Section‑Combustion Institute. He has been honored with the Society of Automotive Engineers Ralph Teetor Award for engineering educators and the University of Colorado Teacher Recognition Award for outstanding teacher during the year. His research awards include the American Society of Mechanical Engineers Gustus L. Larson Award, the Fulbright Fellowship, the University of Colorado Faculty Fellowship, and the Associated Western Universities Faculty Fellowship. He has authored over 90 technical articles and supervised 15 students to completion of their PhD. Wes Bussman, PhD, is a senior research and development engineer for the John Zink Co., LLC (Tulsa, Oklahoma) where he has been since 1981. He received his degree in mechanical engineering from the University of Tulsa (Tulsa, Oklahoma). Dr. Bussman has 19 years of basic scientific research work, industrial technology research and development, and combustion design engineering. He holds ten patents, has authored several published articles and conference papers, and has been a contributing author to several combustion-related books. He has taught engineering courses at several universities and is a member of Kappa Mu Epsilon Mathematical Society and Sigma Xi Research Society. Yaroslav Chudnovsky, MS, PhD, is a senior staff member of research and development at the Gas Technology Institute (Des Plaines, Illinois). Dr. Chudnovsky received his degrees in 1982 and 1990, respectively, from Bauman Technical University (Moscow, Russia). He conducts research and development of advanced, low-emissions, highefficiency, and high heat transfer combustion systems and technologies for industrial applications. He has over 25 years of combined basic and applied research and development experience in engineering, design, and laboratory/ field evaluation of advanced energy exchange and combustion systems and technologies. Prior to joining the Gas Technology Institute in 1995, he worked as a head of the research laboratory at Power Machinery Research Institute (Moscow, Russia) where he developed solutions for energy, space, and military applications. His areas of interest include: heat transfer enhancement and waste heat recovery, convective heat transfer and heat exchangers, advanced combustion and environmental technologies, and smart thermal management. He has over 100 publications and six patents. He is the editor of the Heat Exchanger Design Handbook and the Journal of Enhanced Heat Transfer. Eddy Chui, PhD, is currently a senior research scientist with CanmetENERGY, the clean energy research and technology development centre of Natural Resources Canada of the Canadian Government (Ottawa, Canada). Dr. Chui received his degree in mechanical engineering from the University of Waterloo (Ontario, Canada) in 1990. Prior to joining CanmetENERGY in 1993, he had worked for Bechtel Canada in project engineering, University of Alberta in acoustic research, and Advanced Scientific Computing Ltd. (presently ANSYS Canada) in numerical modeling. At CanmetENERGY, he is responsible for directing and conducting research on various aspects of combustion modeling technology and utilizing the model to assist industries in practical applications. Past achievements include the development of new modeling strategies to predict NOx formation in coal flames and natural gas flames, determination of the sensitivity of combustion performance to coal blending, design of a new generation of furnace model for process simulation, evaluation of combustion performance in appliances using biomass, development of strategies for industrial processes to convert to a lower-carbon fuel, and the successful implementation of the model to resolve combustion-related problems in full-scale units like utility boilers, coke ovens, refinery furnaces, blast
© 2011 by Taylor and Francis Group, LLC
Contributors
xv
furnaces, and industrial furnaces for metal processing. Current research efforts are focused on developing clean coal technologies: oxy-coal combustion and coal gasification for CO2 capture, modeling CO2 storage in subsurface environments, new computational fluid dynamic tools for nonexpert users, and assisting the power sector in China and Canada to burn coals more cleanly and efficiently through the use of simulation. Also, a new capability of microscale modeling has been developed under his supervision, presently being implemented on investigating solid oxide fuel cells. He has authored and co-authored over 150 publications in international journals, conference proceedings, industrial reports, and government departmental reports. Giovanni Cinti is the technology department manager for the Technical Centre of Italcementi Group (CTG), located in Bergamo (Italy). He received his certificate degree at Politecnico di Milano in 1973 as a chemical engineer. In 1975 he started his professional activity in Italcementi SpA in the central headquarters as a member of the combustion department, dealing with all the aspects of cement kiln burners and related combustion pollutants. He is a member of the International Flame Research Foundation, holding the chairmanship of the Italian Association for four years (2004–2008). He has represented the company in the Associazione Tecnico-Economica del Cemento (AITEC) and in the European Cement Association (Cembureau) and was the cement expert of the Italian delegation in the meeting for the definition of Best Available Technologies for Cement Manufacturing in 2001 and 2007. Joseph Colannino, BS, MS, is director of engineering for John Zink Co., LLC, where he has worked for the last 12 years. He received his degree in chemical engineering with minors in materials and chemistry from the California Polytechnic University at Pomona and a degree in knowledge management with emphasis in organizational dynamics from the University of Oklahoma. He is a registered professional engineer in the state of California. He has been engaged in combustion research for more than 20 years and has authored many papers and presentations. His book, Modeling of Combustion Systems: A Practical Approach (Taylor & Francis), was published in 2006. Besides the John Zink Handbook, Joseph has also contributed book chapters in other volumes including the Industrial Combustion Handbook (CRC Press), and the Air and Pollution Control Equipment Selection Guide, (Lewis). He is an adjunct faculty member at Tulsa University and Oral Roberts University (both in Tulsa, Oklahoma) teaching combustion and engineering courses. Colannino is listed in several Who’s Who compilations. Eugen Dan Cristea, PhD, MEng, has worked since 1987 in the cement and lime industry in the positions of technical director of Cimprogetti SpA, an engineering company located in Bergamo (Italy) and today as a process function manager of Italcementi Group in Bergamo. He received his doctorate degree in thermal sciences from the Politehnica University of Timisoara (Romania) and his engineering degree in power generation engineering from Politehnica University of Bucharest (Romania). He did postdoctoral work as a visiting adjunct assistant professor at Montana State University (Bozeman, Montana) performing numerical simulation combustion for MHD combustor fired on natural gas. He has served as head of Combustion and MHD laboratory of Scientific Research Division (formerly the Power Institute of Romania Academy) of the Institute of Scientific Research and Technological Engineering for Power Equipment in Bucharest. He conducted some fundamental and mainly applied research works in all areas of thermal sciences including combustion science and combustion engineering, heat and mass transfer, fluid mechanics, thermodynamics and chemical thermodynamics, and direct energy conversion, with particular emphasis on experimental as well as computational approaches. He is a member of the American Society of Mechanical Engineers, of the International Flame Research Foundation at Pisa (Italy), and has served on the Italian Flame Research Committee. Dr. Cristea has authored and co-authored two combustion-related books, over 20 journal articles, over 20 conference papers and holds four patents for novel burner development. He has delivered seminar lectures at the International Flame Research Foundation. Andrea D’Anna, PhD, is an associate professor of chemical engineering at Università “Federico II” di Napoli (Napoli, Italy) where he has been since 2001. He has a degree in chemical engineering from Università “Federico II” di Napoli (Napoli, Italy). He was a researcher at Istituto Ricerche Combustione, CNR and at Fertimont, Montedison SpA. His research interests include combustion chemistry, chemical kinetics, combustion-formed particles and their effects on health and climate, nano-material synthesis, characterization and modeling, transport properties of nano-materials, and filtration procedures. He is the author of over 100 technical publications. Nicolas Docquier, PhD, is the general manager of ACI (Atlanta, Georgia), a division of Air Liquide Advanced Technologies US, specializing in combustion equipment for steel, nonferrous, and glass industries. He has a doctoral degree in energy sciences from the Ecole Centrale de Paris (France), a degree in fluid mechanics from the von Karman
© 2011 by Taylor and Francis Group, LLC
xvi
Contributors
Institute (Belgium), and an engineering degree from Université de Liège (Belgium). He has been a combustion specialist with Air Liquide since 2003, in France and in the United States. He also worked at IFP Powertrain Engineering (Paris, France) and for Rolls-Royce Industrial and Marine Gas Turbines (Coventry, United Kingdom). He has a strong industrial and research background in combustion, industrial heating and melting processes, fluid mechanics, heat transfer, fuels and emissions, and has developed several combustion test platforms. His experience includes oxyburners, furnaces and melting processes, safety systems and practices, emission measurements and sensors, optical diagnostics, internal combustion engines, and turbomachinery. He is the author or co-author of over 20 publications on these topics, has nine patents for novel burner and sensor development and has taught graduate courses on fluid mechanics. Jaime A. Erazo, Jr., MS, is a design/test engineer at the John Zink Co., LLC, Tulsa, Oklahoma. He has worked for the John Zink Company Process Burners group for one year. He graduated with a degree in mechanical engineering from the University of Oklahoma in 2008. He authored five combustion related technical publications and presentations during his time at Oklahoma University. Michael Flamme, PhD, is internationally known for his work on gas-fired technology over a period of more than 20 years with Gaswärme-Institut (Essen, Germany). In 1989 he received his degree from Bochum University, Germany for his work focused on high temperature processes using high preheated combustion air. He has particular expertise and knowledge of high temperature industrial processes, combustion technologies for gas turbines and boilers, and waste and biomass conversion to energy. He authored over 110 publications in national and international journals and conference proceedings. His scientific achievements were rewarded by the Wilhelm Jost Medal of the German Section of the Pittsburgh Combustion Institute in 1993. He currently manages his own independent energy consultancy (FlammeConsulting) in Essen, Germany. Mahmoud M. Fleifil, BS, MS, PhD, is a senior thermoacoustic and vibration engineer in the research and development department of John Zink Co., LLC (Tulsa, Oklahoma). He has been with the company since 1999. Dr. Fleifil graduated from Ain Shams University, Cairo, Egypt with his degrees in mechanical engineering, and his doctoral degree in mechanical engineering from a co-supervisory program between Ain Shams University and MIT. His areas of expertise are fluid dynamics, combustion instabilities, and noise control. He published eight journal articles and over 20 conference papers. He has over 13 years of experience in advanced techniques of acoustically driven combustion instability and noise control. He is a member of ASME and AIAA. He is an honored member of several Who’s Who compilations. Subramanyam R. Gollahalli, MASc, PhD, is a professor and holds the Lesch Centennial Chair in the School of Aerospace and Mechanical Engineering at the University of Oklahoma (Norman, Oklahoma) where he has been since 1976. He received his Master’s degree in 1970 and his doctoral degree in 1973 both in mechanical engineering from the University of Waterloo (Waterloo, Canada). He has held the positions of assistant professor, associate professor, professor, and Lesch Centennial professor, and academic director. He also worked as a research assistant at the University of Waterloo (Waterloo, Canada) and as a lecturer at the Indian Institute of Science (Bangalore, India). He has developed and taught courses in the combustion and energy areas at both undergraduate and graduate levels. He investigated the combustion of emulsified fuels and synthetic fuels, the publications based on his research, which are cited frequently. He served on the following technical committees: Combustion and Fuels Committee, Gas Turbine Division, ASME; Terrestrial Energy Committee, AIAA; Propellants and Combustion Committee-AIAA; Fuels and Combustion Technologies Committee, ASME; Emerging Energy Technologies Committee, ASME; and the Technical Program Committee, Combustion Institute. He has been recognized with the following awards: Regents Award for Superior Teaching at the University of Oklahoma, Energy Systems Award-AIAA, George Westinghouse Gold Medal-ASME, AIAA Sustained Service Award Robert Angus Medal, Engineering Institute of Canada, and Three Best Paper Awards, ASME Fellow, AIAA Associate Fellow. He was the associate editor of the Journal of Energy Resources Technology (1994–2000) and the associate editor of the Journal of Engineering for Gas Turbines and Power (2000–2006). Mikhail Gotovsky, MS, PhD, ScD, is a leading researcher at NPO CKTI (1963 to present) and a professor at GTURP (State University of Plant Polymer Technologies). He received his first degree in 1963 from St-Petersburg State Polytechnic University (Russia) and his other two degrees in 1970 and 2000, respectively, from Central Boiler-Turbine Institute (Russia). After graduating from St-Petersburg State Polytechnic University, Dr. Gotovsky
© 2011 by Taylor and Francis Group, LLC
Contributors
xvii
joined the team of highly experienced and motivated professionals at NPO CKTI and grew from a junior engineer to the leading research and development professional in the area of heat and mass transfer for industrial applications. His areas of interest include: a variety of heat transfer problems (liquid metals, forced convection, high heat flux boiling), two-phase flow hydrodynamics and heat transfer, heat transfer enhancement, and thermal problems of nuclear waste transportation and storage. He has over 120 publications and 12 patents. Ashwani K. Gupta, PhD, DSc, is a distinguished university professor at the University of Maryland (College Park, Maryland) where he has been a professor of mechanical engineering. Prior to this he was a member of the research staff at MIT, and an independent research worker at the University of Sheffield, United Kingdom. His main research interests have been in the fields of combustion, air pollution, propulsion, high temperature air combustion, swirl flows, diagnostics, fuel sprays, fuel reforming, sensors, microscale combustion, and wastes to clean energy conversion. He has co-authored three books on swirl flows, and flowfield modeling and diagnostics, and high temperature air combustion: from energy conservation to pollution reduction. In addition he has authored nine book chapters and published over 450 archival papers in journals, refereed symposia, and conference proceedings. His honors and awards include: AIAA Energy Systems Award, AIAA Propellants and Combustion Award, ASME George Westinghouse Gold Medal, ASME James Harry Potter Gold Medal Award, ASME James N. Landis Medal Award, ASME Worcester Reed Warner Medal Award. Dr. Gupta received the University of Maryland President Kirwan Research Award and College of Engineering Research Award. He received eight Best Paper Awards from ASME and AIAA for his research contributions. He is the founding co-editor of the Energy Engineering and Environment Series published by CRC Publishers. He is an associate editor of the Journal of Propulsion and Power, Journal of Applied Energy, International Journal of Sprays and Combustion Dynamics, and International Journal of Reacting Systems. He has served as chair of AIAA Terrestrial Energy Systems Technical Committee, chair of Propellants and Combustion Technical Committee, deputy director of Energy Group, and director of Propulsion and Energy Group. At ASME he served as chair for Fuels and Combustion Technology Division, and Computers in Engineering Division. He is cited in Who’s Who in America, Engineering, Technology, American Education, and Aviation in the United States, and The Men of Achievement in the United Kingdom. Carl-Christian Hantschk, PhD, has been working as a consulting engineer in industrial acoustics for MüllerBBM GmbH (Munich, Germany) since 2001. He was promoted to managing director in 2009. He works on industrial acoustics in general, including theoretical and applied acoustics, environmental acoustics, aero-acoustics and numerical acoustics, with special focus on the interdisciplinary field between combustion and acoustics. He holds a diploma in mechanical engineering and received his doctorate in thermodynamics from the Technical University Munich, Germany. His research focused on combustion-driven acoustic oscillations in burners and combustionacoustic interactions. He gave lectures on chemical thermodynamics, thermal radiation, and heat transfer and acoustics at his university, international conferences, and for industrial clients. His work resulted in 30 publications and four invention disclosures. As one of his main research projects, he codeveloped an active acoustic feedback control for industrial combustion systems. R. Robert Hayes, MS, is the vice president of El Dorado Engineering, Inc. (Salt Lake City, Utah) where he has been since 2006. He also worked at the National Renewable Energy Laboratory and the John Zink Co., LLC. He received his degree in mechanical engineering from Brigham Young University. He has a strong technical background in combustion, heat transfer, fuels, and emissions formation/reduction, with experience in both industrial and research facilities. He led a world-class alternative fuels and advanced vehicle technology research facility at the National Renewable Energy Laboratory. His experience includes internal combustion systems, alternative fuels, furnaces, burners, air/exhaust handling systems, pollution control systems, emissions measurements, instrumentation, safety systems, and combustion of energetic materials (propellants and explosives). He is the author or co-author of over 20 publications on these topics. He has three patents for novel burner development and has taught professional courses on burners, formation and control of combustion emissions, heat transfer, and fluid mechanics. Ricardo Hernández, graduated in 2003 with a degree in physics at the University of Zaragoza (Zaragoza, Spain), is currently a PhD student there and works in the Laboratory of Research on Combustion Technologies. He works on research of advanced monitoring and control of industrial flames and combustion instabilities, has participated in some research projects and is co-author of several papers in international journals and conferences.
© 2011 by Taylor and Francis Group, LLC
xviii
Contributors
Jianhui Hong, BS, PhD, is a flare process engineer at John Zink Co., LLC (Tulsa, Oklahoma). He is the principal investigator and lead inventor of the ultra-stable WindProof™ pilot, the triple-redundancy InstaFireTM flare pilot, and the ultra-efficient, steam-assisted Steamizer-XP™ flare. He received a degree from Tsinghua University, Beijing, China and his doctorate from Brigham Young University, Provo, Utah, both in chemical engineering. He has other U.S. and foreign patents including low NOx incinerator apparatus and control method and air-assisted flare. He also worked as a research and development engineer at John Zink. His other areas of expertise include ground flare design and optimization, kinetic simulation involving NOx, SOx, and soot; global optimization of steel stack structure considering structural and process constraints; phased array of thermal radiometers for measuring the flame epicenter and radiant fraction of industrial flares; flare smoke control method; flare control for over-steaming/over-aeration prevention. He has authored and co-authored over 15 journal articles and book chapters. His personal interests include heli-plane design, aircraft emergency landing system, and personal aerial vehicle design. Patrick M. Hughes, MSc, is the group leader for measurement systems and combustion kinetics at Canmet ENERGY Ottawa (Ottawa, Canada). He has been a research scientist with CanmetENERGY since 1982. Before that he was a defense scientist for six years with National Defence Canada. He has his degree in mechanical engineering from the University of Waterloo (Waterloo, Canada). Throughout his career he has developed advanced measurement techniques to study combusting flow fields. His research with Natural Resources Canada has involved the use of laser-based and other optical techniques to study industrial burner technology. He is also involved in the development of techniques to characterize coal combustion kinetics and deposition in power boilers. His publications cover optical measurement techniques and their application to industrial burners, data packages for evaluation of computational fluid dynamic models, advanced characterization techniques for coal combustion and deposition, and rocket motor instabilities. He is currently the editor-in-chief of the Combustion Journal of the International Flame Research Foundation. Thomas P. Jenkins, PhD, is a senior scientist at MetroLaser, Inc., where he has worked since 2000. He received his degree in mechanical engineering from the University of California at Davis. He has been principal investigator on eight research programs for the DoE, Air Force, NASA, Army, and private industry to develop laser-based diagnostics for studying combustion and fluid mechanics. He has demonstrated several first-of-a-kind measurements, including quantitative nonintrusive measurements of soot concentration in an aircraft engine exhaust, nonintrusive temperature and H2O concentration in an industrial glass furnace, and a large area flow velocimetry system for studying parachutes. He is a member of the American Institute of Aeronautics and Astronautics (AIAA), and is an active member of the AIAA Aerodynamic Measurement Technologies committee. He has been the primary author on more than 30 journal articles and conference papers. Prior to coming to MetroLaser, Dr. Jenkins worked for three years as a research associate at Stanford University, where he developed soot diagnostics for advanced propulsion systems. Christopher Q. Jian, PhD, is the director of the Simulation Technology Solutions Group (STS) at John Zink Co., LLC (Tulsa, Oklahoma) where he has been since 2004. He received his degree in mechanical engineering from the University of Maryland at College Park (College Park, Maryland). Prior to joining John Zink, he was the business research manager at Owens Corning responsible for asset utilization and customer profitability analyses and mergers and acquisitions. He was the research and development manager at Vortec Corporation before he joined Owens Corning in 1995. He served as past chair of the production efficiency subcommittee of the Glass Manufacturing Industry Council and a member of its executive advisory committee. Dr. Jian’s research areas include fossil fuel combustion, glass melting and delivering, computational fluid dynamics and physical modeling, as well as low level radioactive material vitrification. He holds four U.S. patents and has authored and/or co-authored over 80 technical publications. Bruce C. Johnson, MSc., PE, is the technology development leader for the Thermal Oxidation Systems Group at the John Zink Co., LLC (Tulsa, Oklahoma). He received his degree in chemical engineering from the University of North Dakota under a Bureau of Mines research fellowship. He has spent much of his career in research and development and has been employed by the Calgon Corporation, Department of Energy, Combustion Engineering Co., and several thermal oxidation companies. His qualifications include process design, new product development, testing, and project management of governmental and corporate research and development groups. He has five patents and has authored numerous technical reports and papers.
© 2011 by Taylor and Francis Group, LLC
Contributors
xix
Vit Kermes, PhD, currently works as a lecturer at Brno University of Technology (Czech Republic). He holds his degree in process engineering from Brno University. His work is directed at applied and industrial research of reduction of NOx emissions in combustion of gaseous fuels and industrial research of nonstandard liquid combustion such as liquid wastes and renewable liquid fuels. He is an author and co-author of about 20 papers related to combustion modeling presented at international journals and international conferences. Thomas M. Korb, PhD, PE, is a technical leader in the Process Burner Group at the John Zink Co., LLC (Tulsa, Oklahoma). He received his degree in mechanical engineering from Arizona State University (Tempe, Arizona). Dr. Korb has 15 years of experience in combustion and thermal sciences. His work has included design and testing of combustion equipment for the refining and petrochemical industries as well as failure analysis engineering of accidental fires and explosions. He has also worked in the development of both gas turbine and diesel engines. He is a registered professional engineer and is a member of Tau Beta Pi National Engineering Honor Society, American Society of Mechanical Engineers, Society of Automotive Engineering and the Experimental Aircraft Association. His research focuses on fundamental ignition mechanisms with a particular emphasis on hot surface ignition of hydrocarbon fuels and the impact of hot surface material and surface oxide structure. He is a recipient of the Darryl E. Metzger Scholarship and Dean’s Graduate Scholars Award at Arizona State University. Richard J. Lacelle, CET, LSO, is an electro-optics technologist working at CanmetENERGY, Natural Resources CANADA (Ottawa, Ontario). He is a graduate of Algonquin College of Applied Arts and Technologies (Ottawa, Ontario, Canada) in the field of electronics engineering technology and electronics engineering techniques. His initial research project was the commissioning of the coherent anti-Stokes Raman spectroscopy (CARS) system for the measurement of high temperature combustion flames in industrial burners. He is also the laser safety officer who oversees all laser operations at the CanmetENERGY Bells Corners Complex located west of Ottawa. Lacelle has been a member of the CARS team at CanmetENERGY since its inception. He is responsible for the development, assembly, bench testing, operation, and data acquisition of the CARS system. For the past 26 years he has been working on research projects like very high speed electronic triggering circuits for the various laser applications, Schlieren photography, high speed camera imaging, laser sheet visualization, infrared imaging and analysis, laser doppler velocimetry, Canmet flame identification control system, tuneable diode laser absorption spectroscopy, and laser induced breakdown spectroscopy. He has authored over 20 technical documents, as laser safety manuals based on ANSI Z136.1 standards for the safe use of lasers, operator’s manual, and standard operating procedures for the above technologies. William J. Lang, BS, is vice president and co-owners of Lenox Instrument Co., Inc. (Trevose, Pennsylvania), manufacturer of the FireSight high-temperature video camera system, along with a full line of other remote visual inspection equipment. A graduate of La Salle College, he began his career in the Lenox shop fabricating high-temperature lenses and optics, and he later pioneered the use of the portable FireSight camera system that is in wide use in fossil-fuel power plants. He has 42 years of application engineering experience visual inspection and process monitoring. He has written articles that have appeared in a variety of technical publications. One such article, “Furnace Cameras Assist in NOx Reduction” appeared in Power Engineering magazine in November 2002. His extensive background in applications include: installing a furnace camera in the combustion chamber of an operating gas turbine, monitoring nuclear waste encapsulation in glass, chemical and biological warfare incineration, and thousands of boiler and furnace installations around the world. He has experience with all fossil fuels. Jeffery Lewallen, BSME, PE, is the applications sales manager of the burner division of Callidus Technologies by Honeywell (Tulsa, Oklahoma). He is a University of Tulsa graduate with a degree in mechanical engineering and a professional engineer licensed in Oklahoma. He has over 17 years of combustion related experience including design engineering, production testing, technical field support, sales, and project management for global projects in the refining and petrochemical industry. He is a contributing author in the books Industrial Burners Handbook and the John Zink Combustion Handbook. Steve Londerville, BSME, is currently director of design engineering at Coen Company (Foster City, California). He received his degree from San Jose State University (San Jose, California) in 1977. Previous positions, since 1978, at Coen were chief technical officer research and development, vice president research and development, director research and development, and chief engineer. During the last 31 years he has been involved with all aspects of product
© 2011 by Taylor and Francis Group, LLC
xx
Contributors
development at Coen. He holds seven patents and has authored 15 publications. He is a member of ASME, ACHIE, Combustion Institute, Tau Beta PI, and past officer and board of directors for Institute for Liquid Atomization and Spray Systems. He was recognized as Engineer of the Year by ASME, Santa Clara Valley section. He also received a Best Paper Award from ASME. Adrian J. Majeski, MSc, is a research engineer at Natural Resources Canada’s CanmetENERGY research center in Ottawa, Canada. He received his degree from the University of Alberta where he participated in the Flare Research Project. Since joining in 2001, he has worked on computational fluid dynamic models of both pilot- and industrialscale combustion systems including utility boilers and equipment related to metal processing and petroleum refining. His current research includes model development for clean-coal technologies, such as gasification and oxyfuel combustion. Before joining CanmetENERGY, Majeski worked on low-swirl burner technology at Lawrence Berkeley National Laboratory. Michele Marrocco, PhD, is a researcher in laser spectroscopy at ENEA (Rome, Italy) (1999 to present). He received his degree in physics from the University of Rome in 1994. He was employed as a postdoctorate at the Max-Planck Institute for Quantum Optics (Munich, Germany), as a researcher at the Quantum Optics Labs at the University of Rome (Rome, Italy), and as an optics researcher by the army. His research activities include: traditional and innovative spectroscopic techniques for diagnosis of combustion and nanoscopic systems studied by means of optical microscopy. The techniques used include: adsorption, laser induced fluorescence, spontaneous Raman, stimulated Raman gain, stimulated Raman loss, coherent anti-Stokes Raman, degenerate four wave mixing, polarization spectroscopy, laser induced breakdown, laser induced incandescence, laser induced thermal gratings. He has over 30 technical publications. Ambrogio Milani, DrIng, is a consultant for WS GmbH (Germany). He received a degree from the Politecnico di Milano (Milan, Italy) in 1965. He has 40 years of experience in combustion technology in energy intensive industrial sectors (steel and power generation). He was the head of the former CSM Experimental Station on combustion devoted to research and development for products and industrial processes. He works with the International Flame Research Foundation and the Combustion Institute. He is the manager of ECSC-funded research and development projects and of educational/training programs and courses. His interests include: combustion research (mainly iron and steel making), burner development, heat recovery, high efficiency, low emissions, and flameless oxidation. He has a number of technical publications including co-authoring the Handbook of Burner Technology for Industrial Furnaces. Susumu Mochida is the director and general manager of Technology & Engineering Division in Nippon Furnace Co., Ltd. (Yokohama, Japan) where he has been since 1982. He has participated in a number of projects and has been an active member of the high temperature air combustion (HiTAC) project team. The HiTAC technology has been widely adapted in industrial furnaces to save energy, reduce size of the equipment, and reduce pollution emission. He serves as chairman on the Japanese Flame Research Committee and a member of the Executive Committee of International Flame Research Foundation from 2006. He has authored over 25 technical publications, contributed to 20 patents, and made several presentations at meetings and conferences. He has been honored with AIAA Best Paper Award in 1999 and ASME George Westinghouse Silver Medal Award in 2001. Thangam Parameswaran, PhD, is a research scientist at CanmetENERGY, Natural Resources Canada (Ottawa, Canada). She obtained her degree from Northwestern University (Evanston, Illinois). Her early research was focused on the theoretical aspects of the optical properties of organic and transition metal complexes. During later years her research activities at Carleton University (Ottawa, Canada) involved theoretical and experimental aspects of laser Raman spectroscopy of transition metal compounds. Subsequently she worked for the National Research Council of Canada toward the development of coherent anti-Stokes Raman spectroscopy (CARS) for combustion diagnostics. Dr. Parameswaran has been a member of the CARS teams at NRC, Canada and CanmetENERGY for many years. During this period she was responsible for developing and applying theoretical calculations and analysis methods for retrieving information from CARS spectra. For the past 11 years she has been working as a research scientist at CanmetENERGY, Natural Resources Canada. She has also developed methods for applying flame emission spectroscopy for flame performance monitoring in industrial burners. Recently this approach was tested in an industrial boiler and has the potential to be implemented in a flame advisory system. Other optical methods she has initiated at CanmetENERGY are tunable diode laser absorption spectroscopy for stack gas measurements in pilot-scale facilities
© 2011 by Taylor and Francis Group, LLC
Contributors
xxi
and laser induced breakdown spectroscopy for trace metal detection in combustion emissions. She has authored over 60 technical documents, as journal publications, conference proceedings, presentations, and contract reports in the fields of optical spectroscopy and combustion applications of spectroscopic methods. In 1993 she received a Joint Staff Performance Award as a member of the Advanced Combustion Diagnostics Technique (Coherent antiStokes Raman Spectroscopy) team, of the Institute for Chemical Processes and Environmental Technology, National Research Council of Canada. Chendhil Periasamy, PhD, BS, MS, a research scientist at Air Liquide Delaware Research and Technology Center (Newark, Delaware) since 2007. He received his doctorate in 2007 in mechanical engineering from the University of Oklahoma with Professor S. R. Gollahalli. He has degrees in mechanical engineering from the Indian Institute of Technology Madras (India) and Anna University (India). He specializes in developing and testing cleaner and energy-efficient oxy-combustion burners for glass, nonferrous, and steel industry applications. He has developed test platforms for evaluating oxy-burner performance and conducted several customer field trials. His research interests include oxy-combustion, industrial furnaces, energy systems, burner testing, combustion diagnostics, combustion in porous media, and oxygen safety. He is the author or co-author of over 25 peer-reviewed journal and conference publications in combustion and energy related topics. He received the Outstanding Graduate Student Award in 2007 for his porous media combustion research and undergraduate teaching activities. Nate Petersen, PE, BS, MS, is currently a process engineer at John Zink Co., LLC (Tulsa, Oklahoma) where he has been since 2005. He has degrees in chemistry and chemical engineering along with a degree in chemical engineering from the University of Utah (Salt Lake City, Utah). He has served in various engineering roles in the process burner group, flare group, and thermal oxidizer group consisting of process and mechanical design and equipment testing. He is a licensed professional engineer in the state of Oklahoma. Erwin Platvoet, MSc, is the director of process burner engineering at John Zink Co., LLC (Tulsa, Oklahoma) where he has been since 2009. He has a degree in chemical engineering from Twente University of Technology (Enschede, The Netherlands). He was a cracking furnace specialist at Total Petrochemicals (Feluy, Belgium) from 2004 to 2009. He was at ABB companies in The Netherlands, USA, and Switzerland from 1993 to 2009 and held a variety of positions including thermal engineer, principal development engineer, and research and development engineer at a variety of locations around the world. He worked at NRF Thermal Engineering (Uden, The Netherlands) from 1991 to 1993. He has authored a number of technical publications and has eight patents. Roger L. Poe, BS, is a research associate at John Zink Co., LLC (Tulsa, Oklahoma) where he has worked since 1999. He received a degree in mechanical engineering from Fairmont State University (Fairmont, West Virginia). Previously, he served as manager of the Callidus Technologies (Tulsa, Oklahoma) Test and Research Center from 1995 to 1999 where he was responsible for the design and testing of specialty burners, as well as the development of new burner equipment for the refinery and petrochemical industry. From 1989 to 1995 he managed the facilities and personnel for the Penn State University Energy and Fuels Research Center (State College, Pennsylvania). He served as a working manager and researcher for the Donlee Technologies Research and Development Group (York, Pennsylvania) from 1985 to 1989. During his career he has been involved with low NOx boiler burner technologies as they relate to both liquid and gaseous fuels, coal gasification in pilot-scale, fluidized bed reactors, and the development and testing of fluidized bed combustion units while working with Donlee Technologies. Further work, when on staff at Penn State University, was done with fluidized bed combustion, coal gasification, micronized and pulverized coal applications, coal slurry formulation and combustion, as well as low NOx gas and oil development. His most recent work has been concentrated in low NOx process type burners and large-scale flaring equipment while working with both the John Zink Co. and Callidus Technologies. He has published more than 24 articles and holds numerous patents relating to burners, flares, and pilots. His areas of interest are centered on fluid mechanics, combustion, thermodynamics, combustion testing, and manufacturing. Over the course of his career he has jointly worked with the Department of Energy, the Department of Defense, NORAD, the Institute of Gas Technology, the Gas Research Institute, Sandia National Labs, and the Natick Naval Test labs. Lee J. Rosen, DSc, is a senior manager of the combustion research and development group for Praxair, Inc. (Tonawanda, New York) where he has been since 2003. He received his doctorate in mechanical engineering from Washington University in St. Louis, Missouri. Dr. Rosen has 19 years of basic scientific research work, industrial technology research and development, and combustion design engineering. His experience includes oxyfuel
© 2011 by Taylor and Francis Group, LLC
xxii
Contributors
combustion, flame stability, pulsed combustion, and flame synthesis of ultrafine particles. He holds three U.S. patents and has authored several published articles and conference papers. Silvio Rudy Stella, PhD, is the marketing director of Reway SrL (Possagno, Italy) that manufactures cogeneration systems where he has been since 2006. He has a degree in electrical engineering from the Ministry of Scientific Research and Technology (Rome, Italy). He worked in a variety of roles at Riello Energy Group for Life from 2002 to 2005, at Thermital SpA from 2000 to 2002, at Calortecnica SpA from 1992 to 2000, at Gemmo SpA from 1991 to 1992 and at Padova Ricerche from 1989 to 1990. He has participated in a number of organizations and associations, held offices in some of those organizations including president of the Italian cogeneration federated ANIMA called ItalCogen, has given many presentations, has been an expert witness, and is an inventor on five patents. He won a prize in memory of Antonio Sarpi (University of Padua) as the best graduate of the Faculty of Engineering in 1989. Allan M. Runstedtler, MASc, is a research scientist with CanmetENERGY, the energy research and technology centre of Natural Resources Canada of the Canadian Government (Ottawa, Canada). He received his degree in mechanical engineering from the University of Waterloo (Waterloo, Canada) in 2000. He has been the technical lead on the investigation of industrial problems in refinery heaters and metal ore processing. Other work has led to the development of a simple boiler model for process modeling of utility boilers and the conceptual design of an ultra-low NOx burner for heat recovery from microturbines. He is interested in fundamental physical, chemical, and mathematical issues related to energy systems and has authored or co-authored journal papers on radiant properties of combustion gases and diffusion in micro-scale pores. Among his current interests are a fluid dynamic theory of turbulence and the use of density functional theory to study the relationship of material properties and reactivity to atomic structure. Khaled A. Sallam, PhD, is an associate professor at Oklahoma State University (Stillwater, Oklahoma) since 2003. He received his degree in aerospace engineering from the University of Michigan (Ann Arbor, Michigan) where he worked in the spray dynamics lab in 2002. From 2003 to 2009 he worked as an assistant professor of mechanical and aerospace engineering at Oklahoma State University. In the summer of 2009, he was tenured and promoted to associate professor. In 2008 he was selected as a summer faculty fellow for the Air Force Summer Fellowship Program at Wright-Patterson Air Force Base. He was awarded the 2007 Halliburton Excellent Young Teacher Award from Oklahoma State University and the 2006 W.R. Marshall Award from ILASS-Americas—Institute for Liquid Atomization and Spray Systems, North and South America. He is a member of the American Society of Mechanical Engineers, the American Institute of Aeronautics and Astronautics (and a member of AIAA Fluid Dynamics Technical Committee), the American Physical Society, and the Institute for Liquid Atomization and Spray Systems. He published 12 journal articles and 29 conference papers and supervised two PhD students and seven master students. Edwin Schorer, PhD, has been working as a consulting engineer in industrial acoustics for Müller-BBM GmbH (Munich, Germany) since 1989. He received his degree in electrical engineering and his doctorate in psychoacoustics from the Technical University Munich, Germany. He was promoted to managing director in 2006. He works in industrial acoustics in general, including theoretical and applied acoustics, with special focus on noise predictions for flare noise and fan noise, fluid mechanics, ship acoustics, and acoustic optimization of postal automation systems. His research work resulted in 15 publications on psychoacoustics as well as industrial and technical acoustics. Dr. Schorer is a member of the German Institute for Standardization, the Noise Control and Vibration Engineering Standards Committee, and the German Acoustical Society. His research focuses on a functional schematic of just noticeable frequency and amplitude variations. He worked as temporary academic counsel at his university, lecturing electroacoustics and technical acoustics. He acted as supervising tutor for the student’s diploma theses and practical trainings. Robert E. Schwartz, PE, BS, MS, is a senior technical specialist at John Zink Co., LLC (Tulsa, Oklahoma). He has received his degrees in mechanical engineering from the University of Missouri. He has worked in the fields of combustion, flares, pressure relieving systems, fluid flow, and heat transfer for more than 40 years including 42 years with Zink where he has provided technical and business leadership in all product areas and has extensive international experience. He has 51 U.S. patents for inventions of apparatus and methods that are in use throughout the John Zink Co. He is the associate editor of The John Zink Combustion Handbook. His areas of technical expertise include: development, design, fabrication, and operation of combustion equipment including flares, incinerators, process
© 2011 by Taylor and Francis Group, LLC
Contributors
xxiii
burners, boiler burners, and vapor control; reduction of NOx and other emissions from combustion processes; fluid flow and heat transfer in process and combustion equipment; noise elimination and control; vapor emissions control using recovery processes; hazardous waste site remediation; and permitting and operation of hazardous waste storage and disposal sites. His professional organizations and awards include: member of the American Society of Mechanical Engineers, the American Institute of Chemical Engineers and Sigma Xi, the Scientific Research Society; Registered Professional Engineer in the state of Oklahoma; recipient of the University of Missouri Honor Award for Distinguished Service in Engineering and election to the University of Tulsa Engineering Hall of Fame. Pavel Skryja, MS, currently works as a consultant for industrial power burners and combustion equipment. He holds a degree in process engineering from Brno University of Technology. He has more than seven years of experience as a project manager and designer in research and development of industrial power burners designed for refineries and special installations. He cooperates with Brno University of Technology (Czech Republic) on research of renewable liquid fuels and liquid waste combustion. Petr Stehlík is a professor and director of the Institute of Process and Environmental Engineering at Brno University of Technology in the Czech Republic. He currently holds a position of vice president of the Czech Society of Chemical Engineers. He has several years of experience in engineering industrial practice before joining the university, and at present he is also a director of the research and development team for a certified engineering and contracting company with activities focusing on waste and biomass processing. Some of his main activities include: executive editor of Heat Transfer Engineering and guest editor of international journals, coordinator or contractor of international research projects, author or co-author of more than 200 papers in journals and proceedings, and plenary or keynote speaker at various international conferences. His research and development as well as application activities involve waste and biomass processing, waste to energy systems, applied heat transfer, energy saving, and environmental protection. Jun Sudo is a consultant for Nippon Furnace Co., Ltd. (Japan). He has 40 years of experience in combustion technology in the fields of steel, petro-refinery, boiler, and cement industries. He has taken a leading role for the development of regenerative combustion system in Nippon Furnace. He has numerous technical publications and several patents. He co-received the Technical Award from the Combustion Society of Japan in 1995 for The Regenerative Combustion System. Dariusz Szewczyk, PhD, specializes in high temperature combustion and innovative methods of combustion related to traditional fuels, biofuels, industrial, and waste gases. He received his degree at the Poznan University of Technology (Poznan, Poland). His particular interests are in oxygen deficient combustion technologies centered on the idea of lowering fuel consumption and pollutants emission. He worked at the Royal Institute of Technology (Stockholm, Sweden) where he published, either as author or co-author, a number of papers concerning high temperature air combustion as well as high temperature air gasification. From 2004 to 2007, Dr. Szewczyk worked for VTS AB, a Swedish engineering company working in the field of industrial combustion systems. Currently he is a general manager and co-owner of ICS Industrial Combustion System Sp z o.o., Poland, an engineering company working in the field of industrial combustion systems. In VTS as well as in ICS he is responsible for projects utilizing NFK HRS/HTB/HiTAC technology in Europe. Giuseppe Toniato, DrIng, is business innovation manager of Riello Group (Legnago, Italy) where he has been since 1999. He received his degree in mechanical engineering from the University of Padova (Italy) in 1991. He has been director of engineering for Riello Burner Division for eight years. Prior to this he was projects manager in Magneti Marelli Engine Control Division (Fiat Group). He has been engaged in combustion research for more than 14 years. He holds ten patents and has authored or co-authored over ten publications on metallic mat combustion, catalytic combustion, and burner controls. Guido Troiani, PhD, is a researcher in combustion and fluid-mechanics at ENEA (Rome, Italy) where he has been since 2006 as a postdoctorate. He received his degree in fluid mechanics from the Faculty of Engineering of the University of Rome “La Sapienza” (2004). He began his postdoctoral research at the University of Rome “La Sapienza” in cooperation with the Italian Ship Model Basin (INSEAN), performing experiments and theoretical analysis on free-surface turbulence and on the transition to turbulence of laminar flows. Successively he was employed as postdoctorate at the ENEA research center in the field of turbulent combustion. His main topics
© 2011 by Taylor and Francis Group, LLC
xxiv
Contributors
of interest are the interactions between turbulence and combustion, flame chemiluminescence emissions, fractal aspects of flame fronts and heavy particle dynamics from a theoretical and experimental point of view. He carried out his experiments by means of intrusive devices such as Pitot and hot-wire probes, acoustic doppler anemometers, and also spectroscopical techniques: LDA, PDA, PIV, and laser induced fluorescence. He is the author of five peer-reviewed publications and of 15 conference proceedings. Joachim G. Wünning, DrIng, is the general manager and co-owner of WS GmbH (Germany). He received his degree from the RWTH Aachen (Technical University of Aachen, Germany). His special interests include: combustion research, burner development, heat recovery, high efficiency, low emissions, flameless oxidation. He has numerous technical publications and several patents. Weihong Yang, PhD, is an associate professor in the Division of Energy and Furnace Technology, Royal Institute of Technology, Sweden, where he has been since 2000. He has a degree in thermal energy from Central South University of Technology (China). His major research areas include: high-temperature air combustion (flameless, or MILD combustion), flameless oxyfuel combustion, gasification with high-temperature air/steam, improvement of combustion systems in boilers and incinerators and rotary kiln in the process industries. He has published about 40 papers in international journals, and presented over 60 papers at international conferences. He also carried out and managed many research projects financed by the Swedish nation, EU, and industries from Sweden, United States, Japan, China, France, Poland. Tsutomu Yasuda, BS, is auditor (Supervisory Board Member) of NFK-Holdings Co., Ltd., and auditor of Nippon Furnace Co., Ltd (Yokohama, Japan; wholly owned subsidiary of above company). He received a degree of electric and electronic engineering, from Tokyo Institute of Technology (Tokyo, Japan). He has been engaged as director in research and development of high temperature air combustion technology since he joined the company in 1993. He moved in October 1998 to Japan Industrial Furnace Manufactures Association, which is a nonprofit organization attached to the Japanese Ministry of Economy, Trade and Industries. He assumed director of Japan Industrial Furnace in charge of HiTAC industrial furnaces and took responsibility of administrating the national project funded by Japanese Ministry of Economy, Trade and Industries. In March 2000, he returned to Nippon Furnace Kogyo Co. and assumed director of project promotion mainly related to high temperature air combustion. He has 20 patents and authored 15 papers all regarding HiTAC and high temperature steam gasification. He has presented the papers in Germany, France, Italy, Sweden, Poland, Indonesia, Malaysia, Thailand, Vietnam, Taiwan, and P.R. China. Andrea Zambon, PhD, is an associate professor at the University of Padova (Padova, Italy) where he has been a researcher since 1990 and as an associate professor in the Faculty of Engineering since 1998. He received his degree in metallurgical engineering from the Polythecnic of Turin (Italy) in 1986. He has taught courses in metallic materials, technology of the metallic materials, science and technology of composite materials, materials science and metallurgy, and of materials selection and design. He is also a professor in the PhD course of mechatronics and industrial systems of the PhD School in Industrial Engineering by the University of Padova. He is a member of the Steering Committee at the university. He is a member of the Italian Association of Metallurgy (AIM-Milan) and of both the Technical Committee of the Center for Physical Metallurgy and Materials Science and of the Technical Committee of the Center for Welding and Permanent Joints of the Italian Association of Metallurgy. He has authored over 100 scientific or technical publications on: development of mathematical models for the analysis and the forecast of thermal fields in laser welding and processing of metallic materials; application of nonconventional techniques (plasma torches, laser, high velocity oxygen-fuel) in the production of coatings and in the modification of the surfaces for the improvement of the wear resistance; methods of mechanical characterization and study of the tribological behavior of metal matrix composites; and powder metallurgy: processes of production by gas atomization, analysis of the cooling regimes, production of bulk spray-formed and sintered samples, microstructural and mechanical characterization of the spray-formed or sintered products.
© 2011 by Taylor and Francis Group, LLC
Section I
General
© 2011 by Taylor and Francis Group, LLC
1 Introduction Charles E. Baukal, Jr. Contents 1.1 Introduction........................................................................................................................................................................ 4 1.2 Industrial Combustion Applications.............................................................................................................................. 8 1.2.1 Metals Production.................................................................................................................................................. 8 1.2.2 Minerals Production.............................................................................................................................................. 9 1.2.3 Chemicals Production......................................................................................................................................... 10 1.2.4 Thermal Oxidation.............................................................................................................................................. 11 1.2.5 Industrial Boilers and Power Generation......................................................................................................... 11 1.2.6 Drying................................................................................................................................................................... 12 1.3 Combustion System Components................................................................................................................................. 13 1.3.1 Burners.................................................................................................................................................................. 13 1.3.1.1 Burner Design Factors.......................................................................................................................... 13 1.3.1.2 General Burner Types........................................................................................................................... 17 1.3.1.3 Burner Components.............................................................................................................................. 22 1.3.2 Combustors........................................................................................................................................................... 23 1.3.2.1 Design Considerations......................................................................................................................... 23 1.3.2.2 General Classifications......................................................................................................................... 23 1.3.3 Heat Load.............................................................................................................................................................. 25 1.3.3.1 Opaque Materials.................................................................................................................................. 26 1.3.3.2 Transparent Materials.......................................................................................................................... 26 1.3.4 Heat Recovery Devices........................................................................................................................................ 26 1.3.4.1 Recuperators.......................................................................................................................................... 26 1.3.4.2 Regenerators.......................................................................................................................................... 27 1.4 Testing............................................................................................................................................................................... 27 1.4.1 Purposes................................................................................................................................................................ 27 1.4.2 Probe Types........................................................................................................................................................... 29 1.4.3 Configurations...................................................................................................................................................... 30 1.4.3.1 Bench Scale Testing............................................................................................................................... 30 1.4.3.2 Pilot Scale Testing................................................................................................................................. 31 1.4.3.3 Large-Scale Testing .............................................................................................................................. 31 1.4.3.4 Field Testing........................................................................................................................................... 31 1.4.4 Important Input Parameters............................................................................................................................... 31 1.4.4.1 Fuel Composition.................................................................................................................................. 32 1.4.4.2 Air/Fuel Ratio........................................................................................................................................ 32 1.4.4.3 Geometry................................................................................................................................................ 32 1.4.4.4 Furnace Temperature and Pressure................................................................................................... 32 1.5 Experimental Errors........................................................................................................................................................ 33 1.5.1 Sources of Error.................................................................................................................................................... 33 1.5.2 Minimizing Errors............................................................................................................................................... 33 1.5.3 Uncertainty Analysis........................................................................................................................................... 35 1.6 Combustion Testing Resources...................................................................................................................................... 36 1.6.1 General References.............................................................................................................................................. 36
3 © 2011 by Taylor and Francis Group, LLC
4
Industrial Combustion Testing
1.6.2 Test Facilities......................................................................................................................................................... 37 1.6.2.1 Academic................................................................................................................................................ 37 1.6.2.2 Institutional............................................................................................................................................ 38 1.7 Future................................................................................................................................................................................ 38 References................................................................................................................................................................................... 38
1.1 Introduction Combustion systems are among the most challenging technologies to study. There are not only high temperatures, but usually very high temperature gradients ranging from the incoming reactants at ambient temperature up to flame temperatures. The fluid flow is typically turbulent and may include swirl. The heat transfer includes conduction, convection, and radiation. The radiation is further complicated by the spectral nature of gaseous combustion products. The chemistry is extremely complicated, where the combustion of a relatively simple fuel like methane can involve hundreds of chemical reactions and dozens of species. Trace combustion products such as carbon monoxide and nitrogen oxides are critical because they are typically regulated. For liquid and solid fuels, multiple phases are present. The fuel composition can vary widely and may contain multiple components, waste products, and sometimes multiple phases, depending on the process. The length scales in industrial combustion processes may vary by orders of magnitude, ranging from millimeters for fuel injection ports up to meters for the combustor itself. The combustion system may also include heat recuperation equipment such as air preheaters and waste heat boilers, and pollution control equipment such as scrubbers and catalytic treatment reactors. The materials being heated may be solids, liquids, or gases and have a wide range of properties. For example, molten aluminum can be highly reflective, molten glass is spectrally absorptive, and cement is highly absorptive. The field of industrial combustion is very broad and touches, directly or indirectly, nearly all aspects of our lives. The electronic devices we use are generally powered by fossil fuel fired power plants. The cars we drive use internal combustion engines. The planes we fly in use jet fuel powered turbine engines. Most of the materials we use have been made through some type of heating process. While this book is concerned specifically with industrial combustion, all of the above combustion processes have many features in common. Industrial combustion is complicated by many factors. First, the science of combustion is still developing and has a long way to go until we completely understand it so it can be better applied and controlled. While fire has been with us since the beginning of time, much remains to be learned about it. As the science of combustion combines heat transfer, thermodynamics, chemical
© 2011 by Taylor and Francis Group, LLC
kinetics, and multiphase turbulent fluid flow to name a few areas of physics, the study of industrial combustion is interdisciplinary by necessity. Combustion has been the foundation of worldwide industrial development for the past 200 years [1]. Industry relies heavily on the combustion process as shown in Table 1.1. The major uses for combustion in industry are shown in Table 1.2. Hewitt et al. (1994) have listed some of the common heating applications used in industry, as shown in Table 1.3 [2]. As can be seen in Figure 1.1, the worldwide demand for energy continues to increase. Most of the energy (86%) is produced by the combustion of fossil fuels like petroleum, natural gas, and coal (see Figure 1.2). According to the U.S. Department of Energy, the demand in the industrial sector is projected to increase by 0.8% per year to the year 2020 [3]. Figure 1.3 shows that the industrial sector is one of the largest energy consumers in the United States. Figure 1.4 shows the projected energy source and end use for the United States in 2008. This again highlights the importance of industrial combustion. The combustion community has identified a number of driving forces that will shape industrial combustion for the foreseeable future: markets and economics; environmental quality and greenhouse gases; process improvement; policy and politics; fuel/oxidant choices; energy efficiency; enabling technologies; health, safety, and reliability; and research and education [1]. Many of these are complex issues that dynamically change over time. For example, pollution regulations vary by Table 1.1 The Importance of Combustion to Industry % Total Energy from (at the Point of Use) Industry Petroleum refining Forest products Steel Chemicals Glass Metal casting Aluminum
Steam
Heat
Combustion
29.6 84.4 22.6 49.9 4.8 2.4 1.3
62.6 6.0 67.0 32.7 75.2 67.2 17.6
92.2 90.4 89.6 82.6 80.0 69.6 18.9
Source: U.S. Department of Energy (DOE). Industrial Combustion Vision: A Vision by and for the Industrial Combustion Community. Washington, DC: U.S. DOE, 1998.
5
Introduction
Table 1.2 Major Process Heating Operations Metal melting • Steel making • Iron and steel melting • Nonferrous melting Metal heating • Steel soaking, reheat, ladle preheating • Forging • Nonferrous heating Metal heat treating • Annealing • Stress relief • Tempering • Solution heat treating • Aging • Precipitation hardening Curing and forming • Glass annealing, tempering, forming • Plastics fabrication • Gypsum production Fluid heating • Oil and natural gas production • Chemical/petroleum feedstock preheating • Distillation, visbreaking, hydrotreating, hydrocracking, delayed coking
Bonding • Sintering, brazing Drying • Surface film drying • Rubber, plastic, wood, glass products drying • Coal drying • Food processing • Animal food processing Calcining • Cement, lime, soda ash • Alumina, gypsum Clay firing • Structural products • Refractories Agglomeration • Iron, lead, zinc Smelting • Iron, copper, lead Nonmetallic materials melting • Glass Other heating • Ore roasting • Textile manufacturing • Food production • Aluminum anode baking
Source: U.S. Department of Energy (DOE). Industrial Combustion Vision: A Vision by and for the Industrial Combustion Community. Washington, DC: U.S. DOE, 1998.
Table 1.3 Examples of Processes in the Process Industries Requiring Industrial Combustion Process Industry Steel making Chemicals Nonmetallic minerals (bricks, glass, cement, and other refractories) Metal manufacture (iron and steel, and nonferrous metals) Paper and printing
Examples of Processes Using Heat Smelting of ores, melting, annealing Chemical reactions, pyrolysis, drying Firing, kilning, drying, calcining, melting, forming Blast furnaces and cupolas, soaking and heat treatment, melting, sintering, annealing Drying
Source: Adapted from U.S. Department of Energy, Energy Information Administration. Annual Energy Outlook 2008, report DOE/EIA-0384, 2008, Washington, DC, released June 26, 2009.
location and continue to get more rigorous. Technology continues to improve as the emission requirements get more stringent. The U.S. Department of Energy sponsored a workshop to develop a roadmap for industrial combustion technology. The resulting report [4] identified the following priority research and development (R&D) needs: • Through a system approach, create burner designs that effectively transfer heat to the load. • Improve heat recovery processes to capture waste gases cost-effectively. • Create innovative new furnace designs through better fundamental understanding of combustion and scale-up.
© 2011 by Taylor and Francis Group, LLC
• Improve boiler technology and combustion cycles through fundamental design innovations such as gasification and fluidized beds. • Determine advanced methods to maintain a stable flame and achieve low emissions while using different fuels. • Develop a burner capable of adjusting operating parameters in real time. • Develop real-time sensors and process controls that are more reliable and robust in harsh environments. • Develop computational tools that are more accurate in a wide variety of applications through the collection of physical data and model validation.
6
Other Nuclear electric power Hydroelectric power Coal
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
Natural gas Petroleum
1982
500 450 400 350 300 250 200 150 100 50 0
1980
Quadrillion Btu
Industrial Combustion Testing
Year Figure 1.1 Historical and projected (2008) world energy consumption. (Energy Information Administration).
Natural gas, 23%
Coal, 27%
Hydroelectric power, 6% Nuclear electric power, 6% Petroleum, 36% Other, 2%
Figure 1.2 2006 world energy consumption. (Energy Information Administration).
120 Transportation Industrial Commercial Residential
Quadrillion Btu
100 80 60 40
Year Figure 1.3 U.S. energy consumption by industry sector. (Energy Information Administration).
© 2011 by Taylor and Francis Group, LLC
2009
2006
2003
2000
1997
1994
1991
1988
1985
1982
1979
1976
1973
1970
1967
1964
1961
1958
1955
1952
0
1949
20
7
Introduction
Stock change and other6 0.18 Coal 23.86
Exports 7.06
Natural gas 21.15
Fossil fuels 57.94
Crude oil1 10.52
Coal 22.42 Domestic production 73.71
Petroleum4 27.56
8
Supply 106.55
NGPL22.41 Nuclear electric power 8.46
Renewable energy3 7.32
Petroleum 3.77 Other exports7 3.30
Natural gas 23.84
Fossil fuels10 83.44
Petroleum9 37.14
Imports 32.84
Nuclear electric power 8.46 3
Renewable energy 7.30
Residential12 21.64 Commercial12 18.54
Consumption11 99.30
Industrial12 31.21
Transportation 27.92
Other imports5 5.28
Figure 1.4 2008 projected energy flow (quadrillion Btu) by source and end use in the United States. (Energy Information Administration.)
• Create a pathway to demonstrate and commercialize new technologies and enable information sharing throughout industry about new technologies. • Develop robust design tools that are more userfriendly and accurate, especially with complex phenomena such as turbulence and create a unified code to allow sharing of information more easily and speed development. • Improve integration of boiler systems with the rest of the plant process with new “smart” control systems and designs that capture unused waste heat. Table 1.4 lists the priority R&D needs that were identified for burners, boilers, and furnaces. Essentially all of these needs require some amount of testing. These needs were generated from the following end-use requirements: increased system efficiency; reduced NOx, CO, CO2, and particulate emissions; increased fuel flexibility; more robust and flexible process control and operations; better safety, reliability, and maintenance; lower capital and operational costs; faster, low-cost technology development; and enhanced system integration. Coupled with these needs are some barriers to improvement: financial risk, inability to accurately predict the performance of new systems, lack of industry standards, and the wide gap that often exists between the research done at a small scale that needs to be applied to industrial-scale systems. Testing is often required to address some of these barriers. As shown in Figure 1.5, three elements are required to sustain combustion processes: fuel, oxidizer, and an
© 2011 by Taylor and Francis Group, LLC
ignition source usually in the form of heat. Industrial combustion is defined here as the rapid oxidation of hydrocarbon fuels to generate large quantities of energy for use in industrial heating and melting processes. Industrial fuels may be solids (e.g., coal), liquids (e.g., oil), or gases (e.g., natural gas). The fuels are commonly oxidized by atmospheric air (which is approximately 21% O2 by volume) although it is possible in certain applications to have an oxidizer (sometime referred to as an “oxidant” or “comburent”) containing less than 21% O2 (e.g., turbine exhaust gas [TEG] [5]) or more than 21% O2 (e.g., oxy/fuel combustion [6]). The fuel and oxidizer are typically mixed in a device referred to as a burner that is discussed in more detail below. An industrial heating process may have one or many burners depending on the specific application and heating requirements. Many theoretical books have been written on the subject of combustion, but they have little if any discussion of industrial combustion processes [7–12]. Edwards (1974) has a brief chapter on applications including both stationary (boilers and incinerators primarily) and mobile sources (primarily internal combustion engines) [13]. Barnard and Bradley (1985) have a brief chapter on industrial applications, but have little on pollution from those processes [14]. A book by Turns (2000), which is designed for undergraduate and graduate combustion courses, contains more discussions of practical combustion equipment than most similar books [15]. There have also been many books written on the more practical aspects of combustion. Griswold’s (1946) book has a substantial treatment of the theory of combustion,
8
Industrial Combustion Testing
Table 1.4 Priority Research and Development Needs Burners Burners capable of adjusting operating parameters in real time Sensors and controls for emissions and heat distribution Fuel and oxidant physical and chemical characteristics Fuel and oxidant mixing within the burner Process interactions within the burner system Robust design tools/unified code Fundamental understanding of physical and chemical science Burner system design models (using fundamentals) Performance targets for models More effective heat transfer to load Radiative heat transfer from the flame to the load Heat delivery to the load Physical modeling Sensors and controls Advanced combustion stabilization methods Use of multiple fuels and their characteristics New flame stabilization techniques Boilers New boiler technologies Sensors and controls Alternatives to burners for solid and liquid fuels Fuel and oxidant flexibility Boiler system efficiency improvement
System integration Sensors and controls Application of combined heat and power systems (CHP) Condensate systems integration Boiler personnel training Technology transfer Test and demonstration programs Speed of the commercialization process Furnaces New furnace designs Fundamental knowledge of physical and chemical science Materials of construction Model validation and technology transfer End user product quality Process characterization and development Advanced sensors and process control Best practices (from other industries) Advanced materials of construction Application of sensors and controls Cost-effective heat recovery Design, application, and performance of heat recovery systems Advanced materials of construction Flue gas characterization Integrated computational design tools Design tool characteristics Technology transfer
Source: Energetics, Inc. Industrial Combustion Technology Roadmap: A Technology Roadmap by and for the Industrial Combustion Community, October 2002, www1.eere.energy.gov/industry/combustion/pdfs/combustion_roadmap2002.pdf, 2002.
Fuel
Oxygen
Source of ignition Figure 1.5 Combustion triangle. (From Baukal, C. E., ed., Oxygen-Enhanced Combustion, Boca Raton, FL: CRC Press LLC, 1998; courtesy of CRC Press.)
but is also very practically oriented and includes chapters on gas burners, oil burners, stokers and pulverized coal burners, heat transfer, furnace refractories, tube heaters, process furnaces, and kilns [16]. Stambuleanu’s (1976) book on industrial combustion has information on actual furnaces and on aerospace applications, particularly rockets [17]. There is much data in the book on flame lengths, flame shapes, velocity profiles, species concentrations, liquid, and solid fuel combustion. A book on industrial combustion has significant discussions on flame chemistry, although little on pollution from flames [18]. A book by Borman and Ragland (1998) attempts to bridge the gap between the theoretical and practical
© 2011 by Taylor and Francis Group, LLC
books on combustion [19]. Deshmukh (2005) discusses a range of topics of interest in industrial combustion including fuels, burners, and refractories, where the emphasis is more on metal treatment [20]. However, the book has little discussion about the types of industrial applications considered here. Keating’s (2007) book on applied combustion is aimed at engines and has no treatment of industrial combustion processes [21]. Mullinger and Jenkins (2008) have written a very practical book that discusses a wide range of industrial furnaces and processes [22]. Surprisingly, many handbooks on combustion applications have little if anything on industrial combustion systems [23–27].
1.2 Industrial Combustion Applications Some of the most common industrial applications are briefly discussed next. Many of them are discussed in more detail in other chapters in this book. 1.2.1 Metals Production Metals are used in nearly all aspects of our lives and play a very important role in society. The use of metals
9
Introduction
has been around for thousands of years. There are two predominant classifications of metals: ferrous (ironbearing) and nonferrous (e.g., aluminum, copper, and lead). Ferrous metal production is often high temperature because of higher metal melting points compared to nonferrous metals. Many metals production processes are done in batch, compared to most other industrial combustion processes considered here (e.g., glass production) that are typically continuous. Another fairly unique aspect of metal production is the very high use of recycled materials. This often lends itself to batch production because of the somewhat unknown composition of the incoming scrap materials that may contain trace impurities that could be very detrimental to the final product if not removed. The metals are typically melted in some type of vessel and then sampled to determine the chemistry so that the appropriate chemicals can be either added or removed to achieve the desired grade of material. Another unique aspect of the metals industry is that transfer vessels are preheated prior to the introduction of molten metal into the vessel to minimize the thermal shock to the refractory. Figure 1.6 shows an example of preheating a transfer ladle used to move molten metal around a plant. Since metals melt at higher temperatures, higher intensity burners are often used in these applications. This includes, for example, oxygen-enhanced combustion (OEC) [28,29] (see Chapter 27) and air preheating (see Chapter 21) to increase the flame temperatures and metal melting capability. These higher intensity burners have the potential to produce high pollutant emissions
Firewall
Ladle seal Ladle
Figure 1.6 Ladle preheater. (From Baukal, C. E., ed., Industrial Burners Handbook, Boca Raton, FL: CRC Press, 2004.)
© 2011 by Taylor and Francis Group, LLC
so the burner design is important to minimize these emissions. Another somewhat unique aspect of metals production is that supplemental heating may be required to reheat the metals for further processing. For example, ingots may be produced in one location and then transported to another location to be made into the desired shape (e.g., wheel castings are often made from remelting aluminum ingots or sows). While this process may be economically efficient, it is energy and pollutant inefficient due to the additional heating. Burners are used in the original melting process as well as in the reheating process (see Chapter 9). This is something that has begun to attract more attention in recent years where the entire life cycle of a product is considered rather than just its unit cost and initial energy requirements. For example, aluminum has a low life cycle cost compared to many other metals because of its high recycle ratio. While the energy consumption to make aluminum from raw ore is fairly high, remelting scrap aluminum takes only a fraction of that energy that also means less overall pollution as well. Burners commonly used in the metals industry include high velocity burners, regenerative burners (Chapter 21), radiant tube burners (Chapter 24), air-oxy/ fuel burners and oxy/fuel burners (Chapter 27). 1.2.2 Minerals Production Some common minerals processes include the production of glass, cement, bricks, refractories, and ceramics. These are typically high temperature heating and melting applications that require a significant amount of energy per unit of production. They also tend to have fairly high pollutant emissions as a result of the high temperatures and unit energy requirements. Most of the minerals applications are continuous processes, but there is a wide range of combustors. Large glass furnaces are typically rectangularly shaped and have multiple burners (see Chapter 32). On the other hand, cement kilns are long refractory-lined rotating cylinders that are slightly inclined so that the materials flow gradually down hill (see Figure 1.7; see Chapter 31). A typical cement plant is shown in Figure 1.8. Many of the minerals applications employ some type of heat recovery in the form of air preheating to improve To off gas treatment Feed materials
Burner
Processed clinker
Figure 1.7 Countercurrent rotary cement kiln schematic. (From Baukal, C. E., ed., Industrial Burners Handbook, Boca Raton, FL: CRC Press, 2004.)
10
Industrial Combustion Testing
Figure 1.8 Cement plant. (From Baukal, C. E., ed., Industrial Burners Handbook, Boca Raton, FL: CRC Press, 2004.)
Figure 1.9 Refinery. (From Baukal, C. E., ed., Industrial Burners Handbook, Boca Raton, FL: CRC Press, 2004.)
energy efficiency. However, the heat recovery typically significantly increases NOx emissions. While recycling of used glass (referred to as cullet) is practiced in some applications, there is generally must less recycling in the minerals industry compared to the metals industry. 1.2.3 Chemicals Production This is a very broad classification that encompasses many different types of production processes that have been loosely subcategorized into chemicals (organic and inorganic) and petrochemicals (organic) applications. A typical refinery is shown in Figure 1.9. There is some overlap in terms of the types of heating equipment used where many of the incoming feed materials are in liquid form (e.g., crude oil) that are processed in heaters with tubes running inside them. These are generally lower temperature applications (< 2300°F or < 1300°C) that incorporate heat recovery to preheat the incoming feed materials. Nearly all of the chemicals heating
© 2011 by Taylor and Francis Group, LLC
applications employ multiple burners but in much more diverse configurations compared to many other industries. Burners may be fired horizontally, vertically up, vertically down, or at some angles in-between depending on the specific process (see Chapter 18). There are numerous configurations for fired process heaters (a typical example is shown in Figure 1.10). There are some aspects that make this industry unique compared to others. The first and one of the most important is the wide range of fuel compositions used to fire the heaters. These are mostly gaseous fuels that are byproducts of the production process. These gaseous fuels often contain significant quantities of hydrogen, methane, and propane and may include large quantities of inert gases such as nitrogen and carbon dioxide. A given heater may need to be able to fire on multiple fuels that may be present during various times in the production process. Another unique aspect of this industry is that many of the heaters are fired with natural draft burners (see Chapter 18) where no blower is used to supply the
11
Introduction
Stack Damper
Convection section
Heat loss
Radiant section
Burner Figure 1.10 Process refinery heater. (From Baukal, C. E., ed., The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001; courtesy of John Zink Co. LLC.)
combustion air. These burners are designed differently than conventional forced draft burners and are more susceptible to variations in ambient conditions such as air temperature, humidity, and wind speed. A specific type of burner used in petrochemical applications is a radiant wall burner (see Chapter 18). 1.2.4 Thermal Oxidation The objective of thermal oxidation (also known as waste incineration) processes is to reduce or eliminate waste products that involves combusting those materials (see Chapter 33). Not only is the thermal oxidizer fired with burners, but the waste material itself is often part of the fuel that generates heat in the process. However, the waste usually has a very low heating value, hence the need for supplemental fuel (e.g., Figure 1.11). Thermal oxidation is a more complicated and dynamic process compared to most other industrial combustion processes by nature of the variability of the feed material. Figure 1.12 shows a schematic of a municipal solid waste incinerator. The waste may be very wet after a rain storm that may put a huge extra heat load on the incinerator. In some locations where waste materials are separated for recycling, the waste actually fed into the thermal oxidizer may have a much higher heating value compared to other incinerators where there is no separation of the waste. A complicating factor with incinerators is that the end product, for example the noncombustible waste, must also be disposed of which means that one of the goals of most incineration processes is to produce minimal
© 2011 by Taylor and Francis Group, LLC
Figure 1.11 Municipal waste incinerator. (From Baukal, C. E., ed., Industrial Burners Handbook, Boca Raton, FL: CRC Press, 2004.)
waste output. Because of the waste material variability, other pollutants may be generated that are not normally associated with industrial combustion processes. An example is the burning of plastics that can produce dioxins and furans. The types of thermal oxidizers can vary greatly depending upon a variety of factors. In some cases, waste materials to be destroyed may be fed through the burners. This is particularly true of waste hydrocarbon liquids. Some of the burners used in incineration include air-oxy/fuel and oxy/fuel (Chapter 27). 1.2.5 Industrial Boilers and Power Generation Boilers are used for a variety of purposes in an assortment of applications. Common uses include producing hot water or steam for heating, producing steam for use within a plant such as atomizing oil for oil-fired burners, and producing steam to generate power in large power plants (see Chapter 34). Applications range from small single-burner uses in hospitals, schools, and small businesses up to large multiburner boilers in power plants. The burners used in boilers are typically regulated because of their proliferation and widespread use in applications involving the general public. The burners are normally required to have a full complement of safety controls to ensure safe operation. These burners are often highly regulated to minimize pollutant emissions, particularly in large power plants because of the size of the source. Boiler burners used in larger applications requiring multiple burners are considered in Chapter 20. A special category of burners sometimes used in large power generating plants with gas turbines are called duct burners (see Figure 1.13). A schematic of the typical
12
Industrial Combustion Testing
To steam generator and bag house
Feed hopper
Moving grate
Combustion air
Ash and metal
Figure 1.12 Schematic of municipal solid waste incinerator. (From Baukal, C. E., ed., Industrial Burners Handbook, Boca Raton, FL: CRC Press, 2004.)
Stack Steam drums
Duct burner
Gas turbine Figure 1.13 (See color insert following page 424.) Duct burner flame. (From Baukal, C. E., ed., The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001; courtesy of John Zink Co. LLC.)
location of duct burners downstream of the turbine is shown in Figure 1.14. These burners are unique because they use the combustion products from a turbine as their combustion “air.” The TEG is at an elevated temperature and contains significant quantities of carbon dioxide and water that are the products of the upstream combustion process. The TEG is also at a depleted oxygen level (< 21%) so duct burners are designed to operate under these conditions. They are treated in detail in Chapter 26. 1.2.6 Drying Burners are used in a wide variety of lower temperature drying applications to remove water from
© 2011 by Taylor and Francis Group, LLC
Steam generator
Figure 1.14 Duct burner process schematic. (From Baukal, C. E., ed., The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001; courtesy of John Zink Co. LLC.)
products that was added during the manufacturing process. These are lower temperature applications that include paper manufacturing, printing and publishing, textile manufacturing, and food processing. Drying is defined as “a process in which a wet solid is heated or contacted with a hot gas stream, causing some or all of the liquid wetting the solid to evaporate” [30]. Kudra and Mujumdar (2002) have written a book [31] on advanced drying technologies that cover a wide range of industries. One example is the drying of paper produced from a wet slurry. A typical paper mill is shown in Figure 1.15. In many drying processes, moisture is removed from webs that may be traveling at high speeds. Radiant heating is often used to supplement steam-heated
13
Introduction
cylinders or high velocity hot air dryers [32]. The radiant heaters are either electric or fired with a fuel gas such as natural gas. Thermal radiation burners used in many of these applications are discussed in Chapter 25.
1.3 Combustion System Components There are six components that may be important in industrial combustion processes (see Figure 1.16). One component is the burner that combusts the fuel with an oxidizer to release heat. Another component is the load itself that can greatly affect how the heat is transferred from the flame. In most cases, the flame and the load are located inside of a combustor, which may be a furnace, heater, dryer, or kiln that is the third component in the system. In some cases, there may be some type of heat recovery device to increase the thermal efficiency of the overall combustion system, which is the fourth component of the system. The fifth component is the flow control system used to meter the fuel and the oxidant to the burners. The sixth and last component is the air pollution control system used to minimize the pollutants emitted from the exhaust stack into the atmosphere. The first four system components are considered next. 1.3.1 Burners 1.3.1.1 Burner Design Factors
Figure 1.15 Paper mill. (From Baukal, C. E., ed., Industrial Burners Handbook, Boca Raton, FL: CRC Press, 2004.)
The burner is the device that is used to combust the fuel with an oxidizer to convert the chemical energy in the fuel into thermal energy. A given combustion system may have a single burner or many burners, depending
To atmosphere Pollution control system
Heat exchanger
Exhaust fan
Flue gases Combustion air blower
Combustion air
Burner Load Flow control system
Fuel
Furnace
Figure 1.16 Schematic of an industrial combustion process. (From Baukal, C. E., ed., Industrial Burners Handbook, Boca Raton, FL: CRC Press, 2004.)
© 2011 by Taylor and Francis Group, LLC
14
Industrial Combustion Testing
on the size and type of application. For example, in a rotary kiln a single burner is located in the center of the wall on one end of a cylindrically shaped furnace (see Figure 1.17). The heat from the burner radiates in all directions and is efficiently absorbed by the load. However, the cylindrical geometry has some limitations concerning size and load type that makes its use limited to certain applications such as melting scrap aluminum or producing cement clinker (see Chapter 31). A more common combustion system has multiple burners in a rectangular geometry (see Figure 1.18). This type of system is generally more difficult to simulate with testing because of the multiplicity of heat sources and because of the interactions between the flames and their associated products of combustion. There are many factors that go into the design of a burner. This section briefly considers some of the important factors that are taken into account for a particular type of burner. These factors affect things like
Air Rotary furnace
Burner
Fuel
Figure 1.17 Rotary kiln with single burner. (From Baukal, C. E., Heat Transfer in Industrial Combustion, Boca Raton, FL: CRC Press, 2000; courtesy of CRC Press.)
heat transfer [32] and pollutant emissions [46]. There have many changes in the traditional designs that have been used in burners, primarily because of the continued interest in reducing pollutant emissions. In the past, the burner designer was primarily concerned with efficiently combusting the fuel and transferring the energy to a heat load. New and increasingly more stringent environmental regulations have added the requirement to consider the pollutant emissions produced by the burner. In many cases, reducing pollutant emissions and maximizing combustion efficiency are at odds with each other. For example, a well-accepted technique for reducing NOx emissions is known as staging (discussed below), where the primary flame zone is deficient of either fuel or oxidizer [33]. The balance of the fuel or oxidizer may be injected into the burner in a secondary flame zone or, in a more extreme case, may be injected somewhere else in the combustion chamber. Staging reduces the peak temperatures in the primary flame zone and also alters the chemistry in a way that reduces NOx emissions because fuel rich or fuel lean zones are less conducive to NOx formation than near stoichiometric zones. Figure 1.19 shows how the NOx emissions are affected by the exhaust product temperature. Since thermal NOx is exponentially dependent on the gas temperature, even small reductions in the peak flame temperature can dramatically reduce NOx emissions. However, lower flame temperatures often reduce the radiant heat transfer from the flame since radiation is dependent on the fourth power of the absolute temperature of the gases. Another potential problem with staging is that it may increase CO emissions, which is an indication of incomplete combustion and reduced combustion efficiency. However, it is also possible that staged combustion may produce soot in the flame that Gas temperature (K) 200
Batch
Burner off
Burner off
Burner off
Figure 1.18 Plan view of multiple burners in a glass furnace. (From Baukal, C. E., Heat Transfer in Industrial Combustion, Boca Raton, FL: CRC Press, 2000; courtesy of CRC Press.)
1000
1500
2000
200
180
180
160
160
140
140
120
120
100
100
80
80
60
60
40
40
20
20
0 Flue
© 2011 by Taylor and Francis Group, LLC
Species concentration (ppmvw)
Glass
500
0
Species concentration (ppmvw)
Burner firing Burner firing Burner firing
0 400 800 1200 1600 2000 2400 2800 3200 3600 Gas temperature (°F)
Figure 1.19 NOx as a function of gas temperature. (From Baukal, C. E., Heat Transfer in Industrial Combustion, Boca Raton, FL: CRC Press, 2000. courtesy of CRC Press.)
15
Introduction
1.0
1.5
2.0
2.5
3.0
3.5
4.0 3500 3000 2500
CO (vol. %) NO (ppmvw)
2000 1500 1000 500
1.0
1.5
2.0
2.5
3.0
3.5
NO concentration (ppmvw)
CO concentration (vol. %)
Equivalence ratio 0.5 13 12 11 10 9 8 7 6 5 4 3 2 1 0 0.5
0 4.0
Equivalence ratio Figure 1.20 NOx and CO as a function of stoichiometry. (From Baukal, C. E., Heat Transfer in Industrial Combustion, Boca Raton, FL: CRC Press, 2000; courtesy of CRC Press.)
can increase flame radiation. The actual impact of staging on the heat transfer from the flame is highly dependent on the actual burner design. In the past, the challenge for the burner designer was often to maximize the mixing between the fuel and the oxidizer to ensure complete combustion, especially if the fuel was difficult to burn, as in the case of low heating value fuels such as waste liquid fuels or process gases from chemicals production. Now the burner designer must balance the mixing of the fuel and the oxidizer to maximize combustion efficiency while simultaneously minimizing all types of pollutant emissions. This is no easy task as, for example, NOx and CO emissions often go in opposite directions as shown in Figure 1.20. When CO is low, NOx may be high and vice versa. Modern burners must be environmentally friendly, while simultaneously efficiently transferring heat to the load. There are many types of burner designs that exist due to the wide variety of fuels, oxidizers, combustion chamber geometries, environmental regulations, thermal input sizes, and heat transfer requirements that includes things like flame temperature, flame momentum, and heat distribution. Some of these design factors are briefly considered here. Other important design factors such as heat flux (Chapter 6) and noise (Chapter 8) are discussed elsewhere in the book. Some of the tools used to optimize a burner design include computational fluid dynamic (CFD) modeling (Chapter 11), testing (the subject of this book), and physical modeling (Chapter 10). 1.3.1.1.1 Fuel Depending upon many factors, certain types of fuels may be preferred for certain geographic locations due to cost and availability considerations. Gaseous fuels,
© 2011 by Taylor and Francis Group, LLC
particularly natural gas, are commonly used in most industrial heating applications in the United States and Europe. In Asia and South America, heavy fuel oils are generally preferred although the use of gaseous fuels is on the rise. Fuels also vary depending on the application. For example, in incineration processes, waste fuels are commonly used either by themselves or with other supplemental fuels like natural gas. In the petrochemical industry, fuel gases often consist of a blend of several components, including gases like hydrogen, methane, propane, butane, and propylene. The fuel choice has an important influence on the heat transfer from a flame. In general, solid fuels like coal and liquid fuels like oil produce very luminous flames that contain soot particles that radiate like blackbodies to the heat load. Gaseous fuels like natural gas often produce nonluminous flames because they burn so cleanly and completely without producing soot particles. A fuel like hydrogen is completely nonluminous as there is no carbon available to produce soot. In cases where highly radiant flames are required, a luminous flame is preferred. In cases where convection heat transfer is preferred, a nonluminous flame may be preferred to minimize the possibility of contaminating the heat load with soot particles from a luminous flame. Where natural gas is the preferred fuel and highly radiant flames are desired, new technologies are being developed to produce more luminous flames. These include things like pyrolyzing the fuel in a partial oxidation process [34], using a plasma to produce soot in the fuel [35], and generally controlling the mixing of the fuel and oxidizer to produce fuel rich flame zones that generate soot particles [36]. Therefore, the fuel itself has a significant impact on the heat transfer mechanisms between the flame and the load. In most cases, the fuel choice is dictated by the customer as part of the specifications for the system and is not chosen by the burner designer. The designer must make the best of whatever fuel has been selected. In most cases, the burner design is optimized based on the choice for the fuel. The fuel also has a large impact on the pollutant emissions. For example, gaseous fuels generally contain little or no sulfur so SOx emissions are usually small. However, heavy oils often contain significant quantities of sulfur and therefore SOx emissions are of concern and need to be controlled. Another example is particulate emissions. Gaseous fuels generally burn very cleanly and produce negligible particulates. However, heavy liquid oil fuels can generate high levels of particulate emissions. Therefore, the burner design is important in minimizing pollutant emissions depending on the fuel. In some cases, the burner may have more than one type of fuel. An example of a duel fuel burner is shown in Figure 1.21 [37]. Dual-fuel burners are designed to operate typically on either gaseous or liquid fuels.
16
Industrial Combustion Testing
Staged air is mixed with the combustion products from the primary zone, which lowers the peak flame temperature.
Sub-stoichiometric conditions in primary zone increase the amount of reducing agents (H2 and CO).
Tertiary air Secondary air Primary air Oil gun
Figure 1.21 Dual fuel burner. (From Baukal, C. E., ed., The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001; courtesy of John Zink Co. LLC.)
These burners are used where the customer may need to switch between a gaseous fuel like natural gas and a liquid fuel like oil, usually for economic reasons. These burners normally operate on one fuel or the other, and occasionally on both fuels simultaneously. Another application where multiple fuels may be used is in waste incineration. One method of disposing of waste liquids contaminated with hydrocarbons is to combust them by direct injection through a burner. The waste liquids are fed through the burner that is powered by a traditional fuel such as natural gas or oil. The waste liquids often have very low heating values and are difficult to combust without auxiliary fuel. This further complicates the burner design where the waste liquid must be vaporized and combusted concurrently with the normal fuel used in the burner. 1.3.1.1.2 Oxidizer The predominant oxidizer used in most industrial heating processes is atmospheric air. This can present challenges in some applications where highly accurate control is required due to the daily variations in the barometric pressure and humidity of ambient air. The combustion air is sometimes preheated and sometimes blended with some of the products of combustion, which is usually referred to as flue gas recirculation (FlGR). In certain cases, preheated air is used to increase the overall thermal efficiency of a process (see Chapter 21). The FlGR is often used to both increase thermal efficiency and to reduce NOx emissions. The thermal efficiency is increased by capturing some of the energy in the exhaust gases that are used to preheat the incoming combustion oxidizer. The NOx emissions may also be reduced because the peak
© 2011 by Taylor and Francis Group, LLC
flame temperatures are reduced, which can reduce the NOx emissions that are highly temperature dependent. There are also many high temperature combustion processes that use an oxidizer that contains a higher proportion of oxygen than the 21% (by volume) that is found in normal atmospheric air. This is referred to as OEC and has many benefits, which include increased productivity and thermal efficiency while reducing the exhaust gas volume and pollutant emissions [6]. A simplified global chemical reaction for the stoichiometric combustion of methane with air is given as follows:
CH4 + 2O2 + 7.52N2 → CO2
+ 2H2O + 7.52N2, trace species.
(1.1)
This compares to the same reaction where the oxidizer is pure O2 instead of air:
CH4 + 2O2 → CO2 + 2H2O + trace species.
(1.2)
The volume of exhaust gases is significantly reduced by the elimination of N2. In general, a stoichiometric oxygen-enhanced methane combustion process may be represented by: CH4 + 2O2 + xN2 → CO2 + 2H2O + xN2 + trace species,
(1.3)
where 0 ≤ x ≤ 7.52, depending on the oxidizer. The N2 c ontained in air acts as ballast that may inhibit the combustion process and have negative consequences. The benefits of using OEC must be weighed against the added cost of the oxidizer, which in the case of air is essentially free except for the minor cost of the air handling equipment and power for the blower. The use of a higher purity oxidizer has many consequences with regard to heat transfer from the flame and pollutant emissions generated. Oxygen-enhanced combustion testing is considered in more detail in Chapter 27. 1.3.1.1.3 Gas Recirculation A common technique used in combustion systems is to design the burner to induce furnace gases to be drawn into the burner to dilute the flame, usually referred to as furnace gas recirculation (FuGR). Even though the furnace gases are hot, they are still much cooler than the flame itself. This dilution may accomplish several purposes. One is to minimize NOx emissions by reducing the peak temperatures in the flame, as in FlGR (see Figure 1.22). However, FuGR may be preferred to FlGR because no external high temperature ductwork or fans are needed to bring the product gases into the flame zone. Another reason to use FuGR may be to increase the convective heating from the flame because
17
Introduction
To atmosphere
Fuel
Recirculated Combustion products
Figure 1.23 Schematic of a premixed burner. (From Baukal, C. E., Heat Transfer in Industrial Combustion, Boca Raton, FL: CRC Press, 2000; courtesy of CRC Press.)
Air
ID fan Fuel
Air
Burner Combustor
Figure 1.22 Schematic of flue gas recirculation. (From Baukal, C. E., Heat Transfer in Industrial Combustion, Boca Raton, FL: CRC Press, 2000; courtesy of CRC Press.)
of the added gas volume and momentum. An example of FlGR recirculation into the burner is given in Fioravanti, Zelson, and Baukal [38]. A specific type of burner incorporating FuGR are so-called regenerative burners (see Chapter 21). 1.3.1.2 General Burner Types There are numerous ways that burners can be classified. Some of the common methods are discussed below, with a brief consideration as to how the burner performance is impacted. 1.3.1.2.1 Mixing Type One common method for classifying burners is according to how the fuel and the oxidizer are mixed. In premixed burners, shown in a cartoon in Figure 1.23 and in a drawing in Figure 1.24, the fuel and the oxidizer are completely mixed before combustion begins. Thermal radiation burners (Chapter 25) usually are of the premixed type. Premixed burners often produce shorter and more intense flames, compared to diffusion flames. This can produce high temperature regions in the flame leading to nonuniform heating of the load and higher NOx emissions, although this is very dependent on the specific design. However, in flame impingement heating (see Chapter 9), premixed burners are useful because the higher temperatures and shorter flames can enhance the heating rates. In diffusion-mixed burners, shown schematically in Figure 1.25 and in a drawing in Figure 1.26, the fuel and the oxidizer are separated and unmixed prior to combustion, which begins where the oxidizer/fuel mixture is within the flammability range (assuming the
© 2011 by Taylor and Francis Group, LLC
Figure 1.24 Drawing of a premixed burner. (Courtesy of John Zink Co. LLC.)
Air Fuel Air Figure 1.25 Schematic of a diffusion-mixed burner. (From Baukal, C. E., Heat Transfer in Industrial Combustion, Boca Raton, FL: CRC Press, 2000; courtesy of CRC Press.)
temperature is high enough for ignition). Oxygen/fuel burners (see Chapter 27) are usually diffusion burners, primarily for safety reasons, to prevent flashback. Diffusion gas burners are sometimes referred to as “raw gas” or “nozzle mix” burners as the fuel gas exits the burner essentially intact with no oxidant mixed with it until it reaches the end of the burner nozzle. Diffusion burners typically have longer flames than premixed burners, have lower peak flame temperatures, and usually have a more uniform temperature and heat flux distribution. They may also have lower NOx emissions although again this is design dependent.
18
Figure 1.26 Drawing of a diffusion-mixed burner. (Courtesy of John Zink Co. LLC.)
Air Fuel Air Fuel Air Figure 1.27 Schematic of a partially premixed burner. (From Baukal, C. E., Heat Transfer in Industrial Combustion, Boca Raton, FL: CRC Press, 2000; courtesy of CRC Press.)
It is also possible to have partially premixed burners, shown schematically in Figure 1.27 and in a drawing in Figure 1.28, where a portion of the fuel is mixed with the oxidizer prior to reaching the end of the burner. This is often done for stability and safety reasons where the partial premixing helps anchor the flame, while not fully premixing lessons the chance for flashback. This type of burner often has a flame length and temperature and heat flux distribution that is in-between the fully premixed and diffusion flames. Another burner classification based on mixing is known as staging: staged air and staged fuel. A staged air burner is shown schematically in Figure 1.29 and in a drawing in Figure 1.30. A staged fuel burner is shown schematically in Figure 1.31 and in a drawing in Figure 1.32. Secondary and sometimes tertiary injectors in the burner are used to inject a portion of the fuel and/ or the oxidizer into the flame, downstream of the root
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
Figure 1.28 Drawing of a partially premixed burner. (Courtesy of John Zink Co. LLC.)
Air Fuel Air Fuel Air Figure 1.29 Schematic of a staged-air burner. (From Baukal, C. E., Heat Transfer in Industrial Combustion, Boca Raton, FL: CRC Press, 2000; courtesy of CRC Press.)
of the flame. Staging is often done to reduce NOx emissions and produce longer flames. These longer flames typically have a lower peak flame temperature and more uniform heat flux distribution than nonstaged flames. However, an additional challenge is that multiple longer flames may interact with each other and produce unpredictable consequences compared to single flames, including increasing NOx rather than reducing it. 1.3.1.2.2 Fuel Type Burners may also be classified according to the fuel type. Gaseous fuel burners are the predominant type used in most of the applications considered here. In general, natural gas is the predominant gaseous fuel used because of its low cost and availability. However, a wide range of gaseous fuels are used, for example, in
19
Introduction
Regen tile Gas risers (for combination firing)
Oil gun Tertiary air control
Gas pilot
Air inlet plenum Primary air control Secondary air control Gas riser manifold (for combination firing)
Figure 1.30 Drawing of a staged-air process burner. (From Baukal, C. E., ed., The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001; courtesy of John Zink Co. LLC.)
the chemicals industry [39]. These fuels contain multiple components such as methane, hydrogen, propane, nitrogen, and carbon dioxide and are sometimes referred to as refinery fuel gases. Figure 1.33 shows an example of a typical nonluminous gaseous flame from a burner used in the petrochemical industry. Gaseous fuels are among the easiest to control because no fuel vaporization is required as in liquid and solid fuels. They are also often simpler to control to minimize pollution emissions because they are more easily staged compared to liquid and solid fuels. Liquid fuel burners are used in some limited applications. One of the specific challenges of using oils is vaporizing the liquid into small enough droplets to burn completely. Improper atomization produces high unburned hydrocarbon emissions and reduces fuel efficiency. Steam and compressed air are commonly used to atomize liquid fuels. The atomization requirements often reduce the options for modifying the burner design to reduce pollutant emissions. Another challenge is that liquid fuel oils often contain impurities like nitrogen and sulfur that can increase pollution emissions. In the
Fuel Air Fuel Air Fuel Figure 1.31 Schematic of a staged fuel burner. (From Baukal, C. E., Heat Transfer in Industrial Combustion, Boca Raton, FL: CRC Press, 2000; courtesy of CRC Press.)
Secondary gas nozzle
Primary gas nozzle Flame holder Burner tile
Air plenum
Air register Air register handle Figure 1.32 Drawing of a staged fuel burner. (From Baukal, C. E., ed., The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001; courtesy of CRC Press.)
© 2011 by Taylor and Francis Group, LLC
Figure 1.33 Typical nonluminous flame. (From Baukal, C. E., ed., Heat Transfer in Industrial Combustion, Boca Raton, FL: CRC Press, 2000; courtesy of John Zink Co. LLC.)
20
case of fuel bound nitrogen, so-called fuel NOx emissions increase. In the case of sulfur, essentially all of the sulfur in the fuel converts to SOx emissions. Solid fuels are not commonly used in most industrial combustion applications except in utility boilers (see Chapter 34). The most common solid fuels are coal and coke. Coal is used in power generation and coke is used in some primary metals production processes. Another type of pseudosolid fuel is sludge that is processed in incinerators. Solid fuels also often contain impurities such as nitrogen and sulfur that can significantly increase pollutant emissions. Some solid fuels may also contain hazardous chemicals that can produce carcinogenic pollution emissions. Because solid fuels are not used frequently in the applications considered, they are only discussed in those specific cases. There are some applications that require the burner to be able to fire on a gaseous fuel-like natural gas, a liquid fuel-like fuel oil, or both simultaneously. This is generally due to the economics of the fuel costs. In some locations, a more favorable fuel cost rate can be obtained, for example on natural gas, if the supply may be interrupted with sufficient notice. The back-up fuel is typically fuel oil. These dual-fuel burners (see Figure 1.21) have special challenges because of the significant differences in designs for gaseous and liquid burners. 1.3.1.2.3 Oxidizer Type Burners and flames are often classified according to the type of oxidizer that is used. The majority of industrial burners use air for combustion. In many of the higher temperature heating and melting applications, such as glass production, the oxidizer is pure oxygen. In other applications, the oxidizer is a combination of air and oxygen, often referred to as oxygen-enriched air combustion. Both of these burner types are discussed in Chapter 27. Figure 1.34 shows a schematic of an air/fuel burner that is the most commonly used type in industrial combustion applications. In most cases the combustion is supplied by a fan or blower although there are many applications in the petrochemical industry where
Air Fuel
Industrial Combustion Testing
natural draft burners are commonly used. There are numerous variations of air/fuel burners that are discussed throughout the book. Figure 1.35 shows a method of using OEC commonly referred to as an oxy/fuel burner. In nearly all cases, the fuel and the oxygen remain separated inside the burner and do not mix until reaching the outlet of the burner. This is commonly referred to as a nozzle-mix burner that produces a diffusion flame. There is no premixing of the gases for safety reasons. Due to the extremely high reactivity of pure O2, there is the potential for a flashback if the gases are premixed. In this method, high purity oxygen ( > 90% O2 by volume) is used to combust the fuel. There are several ways of generating the O2. In an oxy/fuel system, the actual purity of the oxidizer will depend on what method has been chosen to generate the O2. Oxy/fuel combustion has the greatest potential for improving a process compared to air/fuel, but it also may have the highest operating cost. Figure 1.36 shows an air/fuel process where the air is enriched with O2. This may be referred to as low-level O2 enrichment or premix enrichment. Many conventional air/fuel burners can be adapted for this technology [40]. The O2 is injected into the incoming combustion air supply, usually through a diffuser to ensure adequate mixing. This is usually an inexpensive retrofit that can provide substantial benefits. Typically, the added O2 will shorten and intensify the flame. However, there may
Fuel Oxygen Figure 1.35 Schematic of an oxy/fuel burner. (From Baukal, C. E., ed., OxygenEnhanced Combustion, Boca Raton, FL: CRC Press LLC, 1998; courtesy of CRC Press.)
Air Fuel Oxygen
Figure 1.34 Schematic of an air/fuel burner. (From Baukal, C. E., ed., OxygenEnhanced Combustion, Boca Raton, FL: CRC Press, 1998.)
© 2011 by Taylor and Francis Group, LLC
Figure 1.36 Schematic of an air-oxy/fuel burner. (From Baukal, C. E., ed., OxygenEnhanced Combustion, Boca Raton, FL: CRC Press LLC, 1998; courtesy of CRC Press.)
21
Introduction
be some concerns if too much O2 is added to a burner designed for air/fuel. The flame shape may become unacceptably short. The higher flame temperature may damage the burner or burner block. The air piping may need to be modified for safety reasons to handle higher levels of O2. 1.3.1.2.4 Draft Type Most industrial burners are known as forced-draft burners. This means that the oxidizer is supplied to the burner under pressure. For example, in a forced-draft air burner, the air used for combustion is supplied to the burner by a blower. In natural-draft burners (see Chapter 18), the air used for combustion is induced into the burner by the negative draft produced in the combustor and by the motive force of the incoming fuel that may be at a significant pressure. A schematic is shown in Figure 1.37 and an example is shown in Figure 1.38. In this type of burner, the pressure drop and combustor stack height are critical in producing enough suction to induce enough combustion air into the burners. This type of burner is commonly used in the chemical and petrochemical industries in fluid heaters. The main consequence of the draft type on burner performance is that the natural-draft flames are usually longer than the forced-draft flames so that the heat flux from the flame is distributed over a longer distance and the peak temperature in the flame is often lower. 1.3.1.2.5 Heating Type Burners are often classified as to whether they are direct (see Figure 1.39) or indirect heating (see Figure 1.40). In direct heating, there is no intermediate heat exchange
surface between the flame and the load. In indirect heating, such as radiant tube burners (see Chapter 24), there is an intermediate surface between the flame and the load. This is usually done because the combustion
Figure 1.38 Photograph of a natural draft burner. (From Baukal, C. E., Heat Transfer in Industrial Combustion, Boca Raton, FL: CRC Press, 2000; courtesy of CRC Press.)
Burner
Secondary air
Air
Load
Figure 1.39 Direct fired process. (From Baukal, C. E., ed., Industrial Burners Handbook, Boca Raton, FL: CRC Press, 2004.)
Burner
Muffle Load
Primary air
Air Gas
Pilot gas
Figure 1.37 Schematic of a natural draft burner. (Courtesy of John Zink Co. LLC.)
© 2011 by Taylor and Francis Group, LLC
Figure 1.40 Indirect fired process. (From Baukal, C. E., ed., Industrial Burners Handbook, Boca Raton, FL: CRC Press, 2004.)
22
Industrial Combustion Testing
products cannot come in contact with the load because of possible contamination. Radiation heat transfer from the flame to the product is the primary mode used in many industrial combustion systems. There are a variety of burner designs that rely primarily on this mechanism. Radiant wall burners are discussed in Chapter 18. Radiant tube burners are discussed in Chapter 24. Thermal radiation burners are discussed in Chapter 25. Forced convection is the other predominant mechanism for transferring heat from flames to a load. For example, high velocity burners are particularly useful in applications where primarily radiant heating may overheat the surface with much less energy getting inside the load. An example would be heating a pile of scrap metal. Highly radiant heating could melt the outside of the pile and cause excessive oxidation leading to high metal yield losses. Convective heating can penetrate inside the load to cause more uniform heating. In certain applications, high velocity burners may not be preferred because the materials being heated may contain fine particles that can easily become airborne. An example is glass manufacturing where the incoming batch materials contain fine powders. 1.3.1.2.6 Geometry There are two primary shapes for the outlet nozzle of industrial burners: round or rectangular. Figure 1.41 shows identical heaters with the same number of burners, but with different burner shapes: round flame and flat flame. Round flames are the predominant shape used in industry. Most of the burners discussed in this book are predominantly round. This is often due to the lower cost of making round shapes compared to making rectangular shapes. It is also often due to the burner tile where round shapes generally require less maintenance (b)
(a)
Burner
Burner
Burner
Burner
Burner
Burner
Figure 1.41 (a) Round and (b) rectangular burner shapes in identical combustors. (From Baukal, C. E., ed., Industrial Burners Handbook, Boca Raton, FL: CRC Press, 2004.)
© 2011 by Taylor and Francis Group, LLC
compared to rectangular tiles that have corners more susceptible to cracking. Another reason may be due to more preferred flow patterns inside round burner plenums compared to rectangular shapes. Rectangular shapes are sometimes preferred in certain applications depending upon the geometry of the combustor and the load. Burners with a fairly high aspect ratio (length to width) are sometimes referred to as “flat” flame burners because the flame shape appears to be flat. One example is in ethylene cracking furnaces where flat-shaped burners fire up along a refractory wall to heat the wall to radiate to tubes opposite that wall. Another example is in glass furnaces where flat-shaped flames fire over the molten glass where these flat shapes often give better flame coverage, more uniform heating, and better thermal efficiency. 1.3.1.3 Burner Components There are several important components that are briefly considered here that impact the burner design. The ignition system is an important component in the burner system to ensure safe and reliable operation. The ignition system is often built into the burner, but in some cases it may be separate from the burner. The system may be fully automatic or completely manual. Different types of ignitors are available. In many cases a pilot is used to ignite the main flame. This may be continuous or interruptible depending on the system design. The pilot may be permanent or removable. The pilot may be ignited by something like a spark-ignitor or by an external torch. Pilots require a separate fuel supply and are typically premixed. Plenums are used to homogenize the incoming gas flows to evenly distribute them to the outlet of the burner. This is important to ensure proper operation of the burner over the entire range of operating conditions, especially at turndown. These gases may include combustion air, premixed fuel and air, or partially premixed fuel and air. If the plenum is too large, then the flows may be unevenly distributed across the burner nozzle outlet. If the plenum is too small, then the pressure drop through the plenum may be excessive. The burner tile, sometimes referred to as a block or quarl, is an important component because it helps shape the flame and protects the internal parts from overheating. In the majority of designs, the burner tile is made of some type of ceramic that often contains alumina and silica depending upon the temperature requirements. The burner tile may also play an important role in the ignition and fluid dynamics of the combustion process. The tile may have bluff body components that enhance flame stability. There may be holes through the tile to enhance mixing of furnace gases with the gases fed into the burner. Advances in ceramics and manufacturing processes have led to increasingly more complicated burner tiles.
23
Introduction
There may be controls on the burner. For example, there is often a damper built into natural draft burners (see Chapter 18) to control the incoming air flow and the furnace draft. Other controls on a burner may be for adjusting the distribution of fuels or air throughout the burner. For example, if a burner has multiple fuel injectors particularly for fuel staging, controls on the burner may be used to control how much fuel goes to each injector. The flame safety system is critical to the safe operation of the combustion system. This may include some type of flame scanner or flame rod to ensure either the burner or the pilot is operating. These are connected to the fuel supply system so that the fuel flow will be stopped if the flame goes out to prevent a possible explosion for unreacted fuel gases contacting a hot surface somewhere in the combustor. 1.3.2 Combustors This section briefly introduces the combustors that are commonly used in industrial heating and melting applications. 1.3.2.1 Design Considerations There are many important factors that must be considered when designing a combustor. This section only briefly considers a few of the more important factors. 1.3.2.1.1 Load Handling A primary consideration for any combustor is the type of material that will be processed. The various types of loads are considered later in this section. One obvious factor of importance in handling the load and transporting it through the combustor is its physical state, whether it is a solid, liquid, or gas. Another factor is the transport properties of the load. For example, the solid may be granular or it might be in the form of a sheet (web). Related to that is how the solid will be fed into the combustor. A granular solid could be fed into a combustor continuously with a screw conveyor or it could be fed in with discrete charges from a front-end loader. The shape of the furnace will vary according to how the material will be transported through it. For example, limestone is fed continuously into a rotating and slightly downwardly inclined cylinder (see Figure 1.17). 1.3.2.1.2 Temperature Industrial heating applications may be divided into two broad categories: higher and lower temperatures. The division between the two is somewhat arbitrary but mainly concerns the different types of applications used in each. For example, most of the metal and glass
© 2011 by Taylor and Francis Group, LLC
melting applications fall into the higher temperature categories as the furnace temperatures are often well over 2000°F (1400 K). They use technologies like air preheating and oxygen-enrichment (see Chapter 27) to achieve those higher temperatures. Lower temperature applications include dryers, process heaters, and heat treating and are typically below about 2000°F (1400 K). Although many of these processes may use air preheating, it is primarily to improve the thermal efficiency and not to get higher flame temperatures. Those processes rarely use oxygen-enrichment, which are usually only economical for higher temperature processes. Both the combustors and the burners are designed differently for higher and lower temperature processes. The heat transfer and pollutant generation mechanisms are often different as well. In higher temperature processes, the primary mode is often radiation, while in lower temperature applications convection often plays a significant role. The NOx emissions, for example, are much more significant at higher temperatures compared to lower temperatures. 1.3.2.1.3 Heat Recovery When heat recovery is used in an industrial combustion process, it is an integral part of the system. The two most popular methods are regenerative and recuperative that are discussed briefly below. The heat recovery system is important in the design of the combustor as it determines the thermal efficiency of the process and the flame temperatures in the system. It also influences the heat transfer modes as it may increase both the radiation and convection because of higher flame temperatures. Another type of heat recovery that is used in some processes is furnace or FlGR where the exhaust products are recirculated back through the flame. This also influences the heat transfer and furnace design as it can moderate the flame temperature but increase the volume flow of gases through the combustion chamber. 1.3.2.2 General Classifications There are several ways that a combustor can be classified that are briefly discussed in this section. 1.3.2.2.1 Load Processing Method Furnaces are often classified as to whether they are batch or continuous. In a batch furnace, the load is charged into the furnace at discrete intervals. There may be multiple load charges, depending on the application. Normally, the firing rate of the burners is reduced or ceased during the charging cycle. On some furnaces, a door may also need to be opened during charging. These significantly impact the thermal efficiency in the system as the heat losses during the charge cycle are very large. The
24
Industrial Combustion Testing
radiation losses through open doors are high and the reduced firing rate may not be enough to maintain the furnace temperature. In some cases, the temperature on the inside of the refractory wall, closest to the load, may actually be lower than the temperature of the refractory at some distance from the inside, due to the heat losses during charging. The heating process and heat transfer are dynamic and constantly changing as a result of the cyclical nature of the load charging. The burners may cycle on and off or between high fire and low fire. This makes testing of these systems more complicated because of the transient nature of the process. In a continuous furnace, the load is fed into and out of the combustor constantly. The feed rate may change sometimes due to conditions upstream or downstream of the combustor or due to the production needs of the plant, but the process is nearly steady state. This makes continuous processes simpler to test as there is no need to include time as a variable in the analysis. It is often easier to make meaningful measurements in continuous processes due to their steady-state nature. There are some furnaces that are semicontinuous where the load may be charged in a nearly continuous fashion, but the finished product may be removed from the furnace at discrete intervals. An example is an aluminum reverberatory furnace that is charged using an automatic side well feed mechanism (see Figure 1.42). In that process, shredded scrap is added in nearly continuous increments to a circulating bath of molten aluminum. When the correct alloy composition has been reached and the furnace has a full load, some or all of that load is then tapped out of the furnace. The effect
Pump well Charge well Figure 1.42 Aluminum reverberatory furnace. (From Baukal, C. E., Heat Transfer in Industrial Combustion, Boca Raton, FL: CRC Press, 2000; courtesy of CRC Press.)
© 2011 by Taylor and Francis Group, LLC
on heat transfer is somewhere between that for batch and continuous furnaces. 1.3.2.2.2 Heating Type As described above for burners, combustors are often classified as direct (see Figure 1.39) or indirect (see Figure 1.40) heating. In indirect heating, there is some type of intermediate heat transfer medium between the flames and the load that keeps the combustion products separate from the load. One example is a muffle furnace where there is a high temperature ceramic muffle between the flames and the load. The flames transfer their heat to the muffle, which then radiates to the load that is usually some type of metal. The limitation of indirect heating processes is the temperature threshold of the intermediate material. Although ceramic materials have fairly high temperature limits, other issues such as structural integrity over long distance spans and thermal cycling can still reduce the recommended operating temperatures. Another example of indirect heating is in process heaters where fluids are transported through metal tubes that are heated by hot gases. Indirect heating processes often have fairly uniform heat flux distributions because the heat exchange medium tends to homogenize the energy distribution from the flames to the load. The heat transfer from the heat exchange surface to the load is often fairly simple and straightforward to compute because of the absence of chemical reactions inbetween. However, the heat transfer from the flames to the heat exchange surface and the subsequent thermal conduction through that surface are as complicated as if the flame was radiating directly to the load. Indirect heating may also have an advantage for reducing pollutant emissions when contact of the high temperature exhaust gases with the load could generate pollutants. As a result of the temperature limits of the heat exchange materials, most higher temperature processes are of the direct heating type where the flames can directly heat the load. 1.3.2.2.3 Geometry Another common way of classifying combustors is according to their geometry that includes their shape and orientation. The two most common shapes are rectangular and cylindrical. The two most common orientations are horizontal and vertical, although inclined furnaces are commonly used in certain applications such as rotary cement furnaces. An example of using the shape and orientation of the furnace as a means of classification would be a vertical cylindrical heater (sometimes referred to as a VC) used to heat fluids in the petrochemical industry. Both the furnace shape and orientation have important effects on the heat transfer in the system. They also determine the type of test that will be done.
25
Introduction
For example, in a VC heater it is often possible to model only a slice of the heater due to its angular symmetry, in which case cylindrical coordinates would be used. On the other hand, it is usually not reasonable to model a horizontal rectangular furnace using cylindrical coordinates, especially if buoyancy effects are important. Some furnaces are classified by what they look like. One example is a shaft furnace used to make iron. The raw materials are loaded into the top of a tall thin vertically oriented cylinder. Hot combustion gases generated at the bottom through the combustion of coke flow up through the raw materials that get heated. The melted final product is tapped out of the bottom. The furnace looks and acts almost like a shaft because of the way the raw materials are fed in through the top and exit at the bottom. A transfer chamber used to move molten metal around in a steel mill is often referred to as a ladle because of its function and appearance. These ladles are preheated using burners (see Figure 1.6) before the molten metal is poured into them to prevent the refractorylined vessels from thermally shocking. Another aspect of the geometry that is important in some applications is whether the furnace is moving or not. For example, in a rotary furnace for melting scrap aluminum, the furnace rotates to enhance mixing and heat transfer distribution. This again affects the type of testing that would be appropriate for that system and can add some complexity to the computations. The burner orientation with respect to the combustor is also sometimes used to classify the combustor. For example, a wall-fired furnace has burners located in and firing along the wall. 1.3.2.2.4 Heat Recuperation In many heat processing systems, energy recuperation is an integral part of the combustion system. Sometimes the heat recuperation equipment is a separate component of the system and sometimes it is part of the burners. Depending on the method used to recover the energy, the combustors are commonly referred to as either recuperative or regenerative (see discussion below). The heat transfer in these systems is a function of the energy recovery system. For example, the higher the combustion air preheat temperature, the hotter the flame and the more radiant heat that can be produced by that flame. The convective heat transfer may also be increased due to the higher gas temperature and also due to the higher thermal expansion of the gases that increases the average gas velocity through the combustor.
process heaters are used to heat petroleum products for further processing. The fluids are transported through the process heaters in process tubes. These heaters often have a radiant section and a convection section. In the radiant section, the process tubes are heated primarily by radiation. In the convection section, the process tubes are heated primarily by convection. The design of the radiant section is especially important as flame impingement on the tubes can cause premature failure of the tubes or cause the hydrocarbon fluids to coke inside the tubes that reduces the heat transfer to the fluids. In some applications, heaters and burners are used to heat or dry moving substrates or webs. An example is shown in Figure 1.43. One common application is the use of gas-fired infrared burners (see Chapter 25) to remove moisture from paper during the forming process [41]. These paper webs can travel at speeds over 300 m/s (1000 ft/s) and are normally dried by traveling over and contacting steam-heated cylinders. The IR (infrared) heaters are often used to selectively dry certain portions of the web that may be wetter than others. For example, if the target moisture content for the paper is 5%, then the entire width of the paper must have no more than 5% moisture. Streaks of higher moisture areas often occur in sections along the width of the paper. Without selectively drying those areas, those streaks would be dried to the target moisture level, which means that the rest of the sheet would be dried to even lower moisture levels. This creates at least three important problems. The first is lost revenue because paper is usually sold on a weight basis. Any water unnecessarily removed from the paper decreases its weight and therefore results in lost income. The second problem is a reduction in the quality of the paper. If areas of the paper are too dry, they do not handle as well in devices like copiers and printers and are not nearly as desirable as paper of uniform moisture content. The third problem is increased operating costs where more energy is consumed than necessary to dry sections of the paper that are already below the target moisture content. Therefore, selective drying of the paper only removes the minimum amount of water from the substrate. The challenge in this application is IR burner
Wet web Roller
1.3.3 Heat Load This section is a brief introduction to some of the important issues concerning the heat load in a furnace or combustor. In petrochemical production processes,
© 2011 by Taylor and Francis Group, LLC
Dry web
IR burner
Figure 1.43 Elevation view of infrared burner heating a moving web. (From Baukal, C. E., Heat Transfer in Industrial Combustion, Boca Raton, FL: CRC Press, 2000; courtesy of CRC Press.)
26
to measure the moisture content profile across the width of a sheet that may be several meters wide and moving at high speeds. The moisture content information must then be fed to the control system for the IR heaters, which then must be able to react almost instantaneously. This is possible today because of advances in measurement and controls systems. Another example of a moving substrate application is using IR burners to remove water during the production of fabrics in textile manufacturing [42]. Moving substrates present unique challenges for burners. Often the material being heated can easily be set on fire if there is a line stoppage and the burners are not turned off quickly enough. This means that the burner control system must be interlocked with the web handling equipment so that the burners can be turned off immediately in the event of a line stoppage. If the burners have substantial thermal mass, then the burners may need to be retracted away from the substrate during a stoppage or heat shields may need to be inserted between the burners and the substrate to prevent overheating. Convection dryers are also used to heat and dry substrates. Typically, high velocity heated air is blown at the substrate from both sides so that the substrate is elevated between the nozzles. In many cases, the heated air is used for both heat and mass transfer, to volatilize any liquids on or in the substrate such as water, and then carry the vapor away from the substrate. An important aspect of heating webs is how the energy is transferred into the material. For example, dry paper is known to be a good insulator. When steam cylinders are used to heat and dry paper, they become less and less effective as the paper becomes drier because the heat from the cylinder does not conduct through the paper as well as when it is moist since the thermal conductivity of the paper increases with moisture content. Infrared burners are effective for drying paper because the radiant energy transfers into the paper and is absorbed by the water. The radiant penetration into the paper actually increases as the paper becomes drier, unlike steam cylinders that become less effective. Another important aspect related to burner design is the ability to cool down rapidly to prevent causing a fire, for example, when a paper web stops unexpectedly during processing due to some type of problem. An alternative is to shield the burner from the web in the event of a sudden line stoppage. 1.3.3.1 Opaque Materials This type of load encompasses a wide range of materials including granular solids like limestone and liquids like molten metal. For this type of load, the heat transfers to the surface of the load and must conduct down into the material. This process can be enhanced by proper
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
mixing of the materials so that new, cooler material is constantly exposed to the surface as in rotary kilns or in aluminum reverberatory furnaces that have molten metal pumps to continuously recirculate the metal through the heating zone. The potential problems with this method include overheating the surface materials or having lower thermal efficiencies by limiting the heat transfer to the surface to prevent overheating. 1.3.3.2 Transparent Materials The primary example of this type of load is glass that has selective radiant transmission properties. In glass melting processes, the primary mode of heat transfer is by radiation. Flames have specific types of radiant outputs that vary as a function of wavelength (see Chapter 6). If the flame is nonluminous, the flame usually has higher radiant outputs in the preferred wavelengths for water and carbon dioxide bands. If the flame is luminous, it has a broader, more graybody-type of spectral radiant profile. Luminous flames are preferred in melting glass because of the selective transmission properties of molten glass. This allows a significant portion of the radiation received at the surface of the glass to penetrate into the glass, which enhances heat transfer rates and reduces the chances of overheating the surface that would reduce product quality. 1.3.4 Heat Recovery Devices Heat recovery devices are often used to improve the efficiency of combustion systems. Some of these devices are incorporated into the burners, but more commonly they are another component in the combustion system, separate from the burners. These heat recovery devices incorporate some type of heat exchanger, depending on the application. The two most common types have been recuperators and regenerators that are briefly discussed next. Reed (1987) predicts an increasing importance for heat recovery devices in industrial combustion systems for increasing heat transfer and thermal efficiencies [43]. 1.3.4.1 Recuperators A recuperator is a low- to medium-temperature (up to about 1300°F or 700°C), continuous heat exchanger that uses the sensible energy from hot combustion products to preheat the incoming combustion air. These heat exchangers are commonly counterflow where the highest temperatures for both the combustion products and the combustion air are at one end of the exchanger with the coldest temperatures at the other end. Lower temperature recuperators are normally made of metal, while higher temperature recuperators may be made of
Introduction
ceramics. Recuperators are typically used in lower temperature applications because of the limitations of the metals used to construct these heat exchangers. 1.3.4.2 Regenerators A regenerator is a higher temperature, transient heat exchanger that is used to improve the energy efficiency of high temperature heating and melting processes, particularly in the high temperature processing industries like glass production. In a regenerator, energy from the hot combustion products is temporarily stored in a unit constructed of firebricks. This energy is then used to heat the incoming combustion air during a given part of the firing cycle up to temperatures in excess of 2000°F (1000°C). Regenerators are normally operated in pairs. During one part of the cycle, the hot combustion gases are flowing through one of the regenerators and heating up the refractory bricks, while the combustion air is flowing through and cooling down the refractory bricks in the second regenerator. Both the exhaust gases and the combustion air directly contact the bricks in the regenerators, although not both at the same time since they are in a different regenerator at any given time. After a sufficient amount of time (usually from 5 to 30 min.), the cycle is reversed so that the cooler bricks in the second regenerator are then reheated while the hotter bricks in the first regenerator exchange their heat with the incoming combustion air. A reversing valve is used to change the flow from one gas to another in each regenerator. The burners used in these systems must be capable of not only handling the high temperature preheated air, but also the constant thermal cycling.
1.4 Testing This section discusses some of the common purposes for testing, the general classification of probes used in testing, a brief discussion of the testing configurations in industrial combustion processes, and some of the important parameters in typical tests. 1.4.1 Purposes There are numerous reasons why tests are conducted in industrial combustion processes. Initial new product development is sometimes done at a reduced scale to minimize costs and risks. When the product is close to commercialization, large-scale (preferably full-scale) tests are required to validate the technology. End users are not likely to use a new technology until it
© 2011 by Taylor and Francis Group, LLC
27
has been validated at or near full scale, although they may be willing to do validation testing if the risk is low enough and the reward is large enough. One way this is sometimes done, for example, is to replace a single old burner with a new burner design in a combustor with many burners. While the data collected from such a test is of limited value, the new technology can be demonstrated to be feasible. This may be done, for example, when a new low NOx burner design is being validated. The key is that there are no operational issues identified during this testing that could cause an unscheduled process shutdown or unexpectedly reduce throughput. Testing may be done to generate data that can be used to validate computer models. There is sometimes the tendency to believe anything generated by a computer. However, the adage “garbage in = garbage out” is appropriate because the model is only as good as the physics in the model and the accuracy of the input data. Nearly all models involve some level of simplification because it is normally prohibitive in terms of time and cost to model the exact physics and because the exact physics are not always known. Computational fluid dynamic (CFD) modeling is a tool that has become popular for simulating all types of combustion problems [44] (see Chapter 11). While it is often possible to model almost anything, this does not mean all predictions are equally valid. Experimental validation must be done to show how well models match actual measurements. However, this means the variability in the experimental measurements must be accurately quantified, so the comparisons with modeling results can be fairly judged. It also means that enough experimental data must be collected to compare the model results over a range of conditions. Failure to do so may limit the reliable range of the model. It may also mean that deficiencies in the model may not be identified because the range of conditions experimentally tested was too narrow. This is one reason why an adequate design of experiments should be completed prior to starting testing (see Chapter 3). Testing can be used to validate new experimental techniques. For example, the optical techniques for measuring temperature discussed in Chapter 5 have been tested against more traditional and well-established techniques such as bare wire thermocouples and suction pyrometers. Some of these optical techniques (e.g., tunable diode lasers discussed in Chapter 14) used more traditional methods such as bare wire thermocouples for calibration. In either case, testing is required to validate new experimental techniques. In many cases, a well known and fully tested configuration with sufficient measurements using proven techniques will be used to demonstrate the validity of a new measurement technique.
28
Industrial Combustion Testing
Testing is done to determine equipment performance. Performance is a broad term that can encompass many different elements. For burners, some important measurements may include pollution emissions (e.g., NOx, CO), turndown range (minimum and maximum firing rates), fuel variability (e.g., hydrogen content), noise, flame shape (length and cross section), heat flux profile, waste destruction capability for waste fuels, and flame stability (a subjective parameter). For flares, some important measurements include the smokeless capacity, the hydraulic capacity, thermal radiation, and noise (see Chapter 28). All of these are determined experimentally to ensure that the equipment meets any guarantees and warrantees that may have been made. For example, Figure 1.44 shows a process burner with a very large muffler. The noise specification imposed by the end user was no more than 64 dBA, which had to be tested to demonstrate the burner met that very stringent performance criterion. Testing may be done to identify and solve problems. For example, process burners (see Chapter 18) are often tested under a range of conditions that rarely exist in actual operation. The tests are done to determine how the burner would behave under those conditions if they ever should arise. One test condition is referred to as “CO breakthrough” where the excess air in the burner is continuously reduced until there is a spike in CO emissions. In general, the CO emissions begin to rise at a faster pace as the excess O2 approaches zero. The purpose of this test condition is to make sure the burner remains stable even at low excess O2 levels. While an actual production process heater is not normally run at those levels, it is possible they could be seen due to an upset or an unexpected transient such as a rapid increase in the fuel heating value. Related to that, process burners are tested under a wide range of fuel compositions and firing rates to ensure stability. If the burner becomes unstable under any of the test conditions, the burner vendor modifies the design to correct the problem so the burner safely operates under the range of conditions that may be encountered. This does not mean the Burner Muffler 4 ft (1.2 m)
8 ft (2.4 m) Figure 1.44 Process burner with very large muffler.
© 2011 by Taylor and Francis Group, LLC
burner will optimally perform (e.g., low emissions) over the entire range of conditions tested, but it must be safe over the entire range. There is normally a design case where optimal performance is guaranteed. Testing may be done to measure the variability in a process. This is particularly important in industrial production processes where the material being heated may have tight tolerances and specific performance requirements. If there is considerable variation in the combustion process, this could adversely impact the product quality. This may mean that appropriate control equipment must be used to ensure any combustion variability is within acceptable tolerances. An example will illustrate this aspect of testing. Changes in the ambient air can significantly impact the pollution emissions from a combustion process [45]. These normal variations in air temperature and humidity, atmospheric pressure, and wind speed and direction must be considered to properly adjust the operating conditions. Either more frequent manual adjustments or automatic controls can be used to combat these environmental fluctuations. Another example is the variation that occurs in the fuel gas composition in, for example, most process heaters [39]. Testing can be used to determine if variations in the combustion process adversely affect the product, and if they do, how effective are the corrective measures. Testing is done to generate performance data that could be related to the burners, the combustor, or to the load. One example of performance data for a burner is the heat flux profile (see Chapter 6) it produces under specific operating conditions. For example, ethylene cracking furnaces are designed to operate with burners producing a specific heat flux profile. Flux rates are typically highest near the middle of the furnace and lower near the floor and ceiling. However, the variation between the highest and lowest flux rates is typically specified by the furnace designer. If there is too much variation, then either production rates may have to be reduced or the run length, before performing maintenance, may have to be reduced. In either case, this reduction in performance is very costly to the end user so it is very important the burners produce the desired profile. This is usually determined through testing in pilotscale furnaces (see Chapter 18) that have only a fraction of the number of burners that a production furnace has, but the burners are normally full scale. The fuel composition, draft conditions, and furnace temperature are matched to that of the production furnace to ensure the testing is realistic. Testing is sometimes done to identify opportunities for increasing process efficiency that often reduces fuel consumption. For example, it is well known that the excess O2 has a big impact on the thermal efficiency of a combustion process [45]. If excess O2 is too high, then the extra air absorbs energy, which it carries out the
29
Introduction
exhaust, thus reducing efficiency. If excess O2 is too low, then high CO emissions may result, which also reduces efficiency because CO is a fuel with heating value that is not being utilized. Testing may be done in a test furnace or in the field in an actual production furnace to determine how low the excess O2 can be before significant levels of CO are generated. For manually controlled furnaces, some margin for error is often built into the minimum O2 level to account for variations that occur such as changes in air temperature, humidity, and barometric conditions as well as other weather-related factors such as high winds and rain. Other examples of testing to improve process efficiency might be using a different type of burner, changing the firing rate profile in the furnace, or moving burners to different locations in the combustor to try to improve the heating efficiency. Testing may be necessary for environmental compliance. There continues to be great interest in reducing pollution emissions from combustion equipment [46]. Many industrial combustion processes have permitted limits on how much pollution they are allowed to emit. Testing is nearly always required to prove the equipment meets the permitted levels. There may be periodic testing (e.g., annually) or there may be continuous testing such as with continuous emissions monitoring systems or CEMS (see Chapter 7). Environmental testing is usually done in the actual production equipment, but in some cases it may be done in a test facility, depending on the requirements of the permitting agency. Testing may be used to enhance combustion safety including determining equipment operational limits. There are safety devices such as ultraviolet flame scanners, flame rods (ionization scanners), and thermocouples that are used to detect the presence (and by default the absence) of a flame. These are not traditional testing devices as defined here. However, there are tests that are conducted to determine the operating limits of combustion equipment such as burners and flares. Nearly all burners have a lower and upper firing rate range [47]. Premix burners
can flashback if the firing rate is too low. Flame liftoff can occur in all types of burners if the maximum design firing rate is exceeded. Although in some limited cases it may be possible to determine operating limits with computer modeling, in most cases these limits are determined through testing. Testing is sometimes used to explain complex phenomenon. While this is often done with CFD modeling, it is also done experimentally. In some cases, it is not yet possible to model all of the physics for a complete industrial combustion system. Better understanding of the physics behind a process can lead to both improvements of existing technologies as well as leading to new developments. For example, understanding how and where NOx is formed in a flame can help in the development of lower NOx technologies. Besides the importance to new product development, test data is often used in the patent process to demonstrate why a particular technology is in fact new and novel and should be granted a patent. Testing is sometimes used to develop empirical models that can be used to predict equipment performance. Testing and CFD modeling can both be expensive and time-consuming. For existing combustion equipment, it is often more cost and time effective to have empirical models to predict performance. For example, a given technology may come in a variety of sizes. Rather than testing every size, a common approach is to test at least the extremes and possibly one or more sizes inbetween depending on how big the range is between those extremes. Then, the performance for any size can be extrapolated from the empirical data, without the need to test or model every possible size. 1.4.2 Probe Types Becker (1993) developed a classification system for physical probes based on the duty conditions as shown in Table 1.5 [48]. While optical probes continue to increase
Table 1.5 Classification of Duty Conditions for Physical Probes Duty Class A B C D E
Typical Case Lean premixed gas flame Turbulent gas diffusion flame in open air Turbulent gas diffusion flame in a combustor Turbulent pulverized-fuel flame in a combustor Fluidized-bed combustion
Radiative Boundary
Gradients
Cold Cold
Very high Medium
Hot
Medium
Hot
Medium
Hot
Small
Probe Issues Miniaturization important Modest if any cooling requirements Modest if any cooling requirements High particulate loading High particular loading but modest cooling requirements
Source: Adapted from Becker, H. A. “Physical Probes.” In Instrumentation for Flows with Combustion, edited by A.M.K.P. Taylor, 53–112. London, UK: Academic Press, 1993.
© 2011 by Taylor and Francis Group, LLC
30
Industrial Combustion Testing
in popularity, there are several technical limitations that need to be considered (Becker 1993) [48]:
1. Fairly short reach due to the reduction of radiant density flux with distance. 2. Radiant signal degradation by passage through variable refractive index fields. 3. Inadequate signal-to-noise ratio in fields with high particle loading. 4. Practical implementation problems in largescale systems such as industrial combustion.
Optical probes have become very popular for measuring a wide range of quantities in laboratory flames. They have several important advantages over physical probes. One is that they are nonintrusive so they do not disturb the flows. This can be very important in regions of high gradients such as inside flames, where a physical probe could affect the measurement being made. Another important advantage is that optical probes often have very fast response times so many measurements can be made in a short time period. This can be critical in transient environments such as turbulent flows. Optical probes often are not subject to temperature limits because the sending and receiving optics can be located outside of high temperature regions. While it is usually possible to provide adequate cooling for physical probes so they can be located in high temperature environment, the increased probe size further exacerbates the problem of intrusiveness. Some optical probes can also generate instantaneous two-dimensional profiles of a given parameter such as temperature or species concentration. Not only would these profiles take much longer with a physical probe that makes point measurements, but there is also the problem that the compilation of point measurements is only an average over some period of time. In highly turbulent flows, these average measurements may not adequately represent the actual transient nature of the fluid dynamics.
1.4.3.1 Bench Scale Testing This refers to small size tests. While the majority of tests done in academic institutions are generally at this scale, this small size can also be useful for certain aspects of industrial combustion testing. One common example of this testing category is usually referred to as physical modeling (see Chapter 10). Scale models may be made for combustion systems such as a large scale power boiler. Plexiglas construction permits easy visualization of the combustion air flow to ensure the flow to each burner is uniform and that there are no undesirable flow recirculations. Figure 1.45 shows an example of a Plexiglas bench scale model. These small scale models are often very detailed to capture the proper fluid dynamics. Bench scale testing may also be used when testing a particular component in the system. For example, testing a burner pilot or a new type of refractory can often be done effectively at a very small scale. Testing with bench scale models is much more cost effective than testing full-scale systems. Fewer people are usually required to
1.4.3 Configurations There are a range of configurations that are used in industrial combustion testing. The Gas Research Institute (GRI) sponsored a program called Scaling 400 where measurements were made on a range of combustors at multiple locations where the highest firing rate tested was 400 times larger than the smallest rate [49]. These measurements ranged from laboratory to large scale furnaces, although no field measurements were included. For convenience here, the range of typical test configurations has been divided into four general categories that are briefly discussed. While these categories are somewhat arbitrary and not precisely defined, they represent the types of combustion tests that have been reported in the literature.
© 2011 by Taylor and Francis Group, LLC
Figure 1.45 Plexiglas bench scale model of a corner-fired utility boiler for flow visualization. (Courtesy of John Zink Co. LLC.)
31
Introduction
run the tests. The initial setup is generally much faster and can usually be relatively quickly and easily modified. However, there may be scaling issues depending on the system. To deal with that problem, physical modeling may be done in conjunction with virtual modeling (see Chapter 11) to ensure that any scaling issues are properly handled. Tests may be conducted over a range of different size combustors [49] to determine what scaling factors should be used when testing in a bench scale unit. Another important limitation of bench scale testing is that industrial end users are unlikely to adopt new technology based solely on it. They normally want to see technology demonstrated at or near full scale. 1.4.3.2 Pilot Scale Testing These are somewhat larger than bench scale tests, but not as large as full-scale tests. Pilot scale may refer to all of the equipment scaled to some fraction of the full size, or it may refer to a full-scale setup but of only part of the actual equipment. In the former case, it may be preferred to test equipment at, for example, half of the full size because of the associated time and cost savings, where scaling up to full size is well defined based on past experience. In the latter case, a pilot scale furnace may have full size burners, but fewer of them compared to an actual production furnace. For example, it is common to test single full size process burners that will be used in a production furnace that may have dozens of those burners in it. 1.4.3.3 Large-Scale Testing This refers to full size or nearly full size testing in some type of test facility where the operating conditions can be easily changed and controlled. An example would be testing full size flares (see Chapter 28). These are very difficult to test in a production plant because it is almost impossible to vary the waste gas flow rates and compositions going to the flare. Due to the critical safety nature of flares, they need to be tested prior to installation in a plant because they usually need to operate continually until the next plant turn around that could be as long as five years or more. Premature failure of the flare could mean shutting down an entire unit or possibly even the whole plant, which would be very expensive and disruptive. In a flare test facility, the flare tip is often at or near full size, but the flare stack is often much shorter than in an actual plant. If the proper measurements are made, such as flare radiation (see Chapter 30), then the flare stack can be properly designed using experimental data from a much shorter stack. Full-scale tests are generally the closest to an actual production environment that helps minimize or eliminate scaling issues. They have an important advantage over field testing in
© 2011 by Taylor and Francis Group, LLC
a production plant because the conditions can be better controlled, varied, and measured. Conditions that may rarely occur in a plant can be easily simulated in a test facility to ensure proper performance. However, large-scale testing can be very expensive, especially if a new apparatus must be built or significant modifications made to an existing test rig, to simulate the wide range of equipment designs that exist in the field. In most practical applications, the large-scale test rig will be close to, but probably not exactly like the actual production equipment in the field. 1.4.3.4 Field Testing This is often the ultimate goal, but usually not until a technology has been satisfactorily demonstrated in a test facility. Few production plants are willing to test unproven technologies because of the associated risk of lost product. Field testing may be done incrementally. For example, a new burner design could be tested by replacing a small proportion of the burners in a heater first before making a complete replacement. This greatly reduces the risk, although it also may mask the benefits as well if too few burners are replaced to see any measurable improvement. There are many potential problems related to field testing. It is generally very difficult to vary the conditions enough to satisfy the researcher who may only be able to test under a very limited set of normal conditions. It may not be practical to simulate anomalous conditions that rarely occur in the plant. Field combustion systems typically have much less instrumentation than test furnaces so it may only be possible to collect a limited amount of data. Field equipment may not be calibrated as often as test equipment so data accuracy may be an issue. Field equipment may be much larger than test equipment, which means instrumentation access may be a problem. For example, getting a gas temperature profile across the top of a large furnace may prove challenging because of both the high elevation and the long distance across the furnace. Field experiments usually must be conducted while minimizing the impact on production, which means staying out of the way of the operators. Despite the limitations of field testing, it offers the ultimate performance measure for a technology since that is where the technology will ultimate be used. A technology that works in the lab but not in the field may be of academic interest but will not be of industrial interest. 1.4.4 Important Input Parameters There are many input parameters that may be of interest in industrial combustion testing, depending on the nature of the technology. However, there are a relatively small number of parameters that are typically important
32
in most tests. Those parameters are briefly discussed here. The output parameters of interest, for example temperature and heat flux, are discussed in detail in other chapters in this book and are not discussed here. 1.4.4.1 Fuel Composition This is an important input parameter in certain tests, while in other cases it varies very little and is therefore not an important input parameter. For example, the fuel composition can vary significantly for process burners (see Chapter 18) and thermal oxidizers (see Chapter 33). Process burners normally use the waste gases generated in the plant as their fuel source. These waste gases may have a dozen or more components in them that may include both hydrocarbons (e.g., methane, propane, and butane) and nonhydrocarbons (e.g., hydrogen, carbon monoxide, nitrogen, and carbon dioxide). This can be challenging to simulate in a test facility if the exact blend composition must be matched. Fortunately, calculations and experience have shown that a small number of components can reasonably accurately simulate a blend with many components, many of which are often in very small quantities and can be safely ignored or combined with other species. One of the key aspects of process burner testing is demonstrating performance over the range of fuels the burner is expected to use. 1.4.4.2 Air/Fuel Ratio This is sometimes referred to as stoichiometry or equivalence ratio and is often an important input test parameter. While a combustion system will likely operate over a relatively narrow range of air/fuel ratios, it is important to demonstrate the combustion system’s capability over a wider range of ratios. The air/fuel ratio is related to the excess O2 in the combustion products (see Chapter 7). If the air/fuel ratio is increasing above stoichiometric levels, then excess O2 increases. If the air/fuel ratio is decreasing to or below stoichiometric levels, then excess O2 decreases to levels approaching zero. As previously discussed, one common test for a process burner (see Chapter 18) is called CO breakthrough, where the excess O2 is continuously reduced until high levels of CO are generated (which normally occurs at excess O2 levels close to zero). This test demonstrates safe performance if the burners should run out of air. While that condition is normally unlikely to occur, if it does occur the burner needs to safely operate, although the flame may not be pretty and the emissions may be high. 1.4.4.3 Geometry There are several aspects of geometry that may be important in industrial combustion testing. One important geometrical factor is the burner design. Burner
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
vendors have extensive experimental facilities to test a wide range of burner concepts. Numerous patents are generated each year regarding new burner designs. A primary focus for some time has been on reducing pollution emissions, so today’s burners produce much less pollution than the designs of older burners. Some of these burners have components that can be adjusted, depending on the needs of the application. The various geometries need to be tested to determine the optimal design for a given set of operating conditions. Furnace design is sometimes an important geometrical parameter. Some burner manufacturers have a variety of test furnaces to simulate the kinds of geometries their burners will see in the field. If an end user is witnessing a performance test of a burner they have purchased, they will want to see it tested in a furnace geometry similar to what they have in their plant. Furnace designers may test various geometries to try to improve performance (e.g., better thermal efficiency or increased production). Geometrical similarity between an industrial combustion test and the actual production system may be important. This means having the right scale burner firing in a test chamber that is representative of the combustor in the field. 1.4.4.4 Furnace Temperature and Pressure These are often important input parameters in combustion testing. Furnace temperature has a direct effect on a number of output parameters such as material processing throughput rate, NOx emissions [50], and heat flux. Realistic tests then normally need to match the furnace temperature(s) that will be seen in the field. This means there needs to be some method for adjusting the temperature in the test furnace. This is typically done by varying the heat load. One method that is commonly used is water-cooled tubes or bayonets. Figure 1.46 shows an
Figure 1.46 (See color insert following page 424.) Water-cooled bayonets inserted into floor of a test furnace. (Adapted from Bussman, W., Baukal, C., and French, K., “Variable Test Furnace Cooling,” presented at the 2005 Summer Heat Transfer Conference, paper HT200572012, San Francisco, CA, July 17–22, 2005.)
33
Introduction
Experimental error is the difference between the true value and the measured value. For example, if the true temperature is 1000 K (1340°F) and the measured value is 950 K (1250°F), then the experimental error would be 50 K (90°F). This section considers the possible sources for experimental errors, minimizing those errors, and estimating the uncertainty in measurements.
that is generally lower than the actual value. The raw measurement needs to be corrected using an appropriate calculation to account for the convection, conduction, and radiation to and from the thermocouple junction inside the hot chamber. A suction pyrometer is another instrument used to measure gas temperature inside a furnace that automatically corrects for these heat transfer effects. These are discussed in more detail in Chapter 5. Another type of error is referred to as random error. If repeated measurements are made under the same conditions, the measured values will vary somewhat. In general, the variations will be normally distributed about some mean value [53]. If the random error is high, then the standard deviation of these repeated measurements will be relatively large. If the random error is low, then the standard deviation will be small with relatively little variability. This random error is more difficult to identify. If the experimenter knew the error, then it could be corrected and would not be random. Therefore, the term “uncertainty” is used because the sources of these random errors are unknown. However, as will be shown later in this section, this uncertainty can be estimated.
1.5.1 Sources of Error
1.5.2 Minimizing Errors
It is important to quantify the uncertainty in measurements made during testing so the data validity is understood. Errors are associated with all testing, regardless of the care taken by the experimenter [52]. One type of error is due to some obvious mistakes by the experimenter. For example, bad data might be generated by failing to calibrate an instrument, such as a NOx analyzer, prior to the start of the experiment, so this data should be discarded. This does not mean that any data can be thrown out just because they do not support the experimenter’s theory. There must be adequate justification for eliminating outliers or other suspicious data. There may be certain fixed errors, also called systematic errors, which cause repeated measurements to be in error by approximately the same amount. For example, if an instrument is incorrectly calibrated, then every measurement may be off by approximately the same amount. Other examples of systematic errors include different experimenters taking measurements differently or some change to the environment such as a valve leaking or left open that should have been closed. There are some measurement techniques that have an inherent error in the measured value compared to the actual value. This is sometimes called a fixed error, although this is somewhat of a misnomer as the amount of the error usually varies with the reading. An example of this type of error is using a thermocouple inserted through the wall of a furnace to measure the gas temperature inside the furnace. This produces a reading
There are several ways that can be used to minimize experimental error. The most important is simply to be very deliberate and careful in the design and setup of the experiment. While this may seem obvious, it is sometimes overlooked due to time pressures and budget constraints. For example, if a Gardon gauge is being used to measure heat flux (see Chapter 6), then the gauge must either be supplied with a calibration curve from the supplier that shows the heat flux as a function of the electrical signal, or it must be calibrated by the experimenter. Even if the gauge came with a calibration curve, it may be prudent to have it recalibrated if the gauge has not been used in some time. It is also recommended that the experimenter thoroughly analyze what is being collected to ensure the necessary data will be available for the analysis that will be done. For example, if the thermal efficiency of a combustion process will be calculated, the experimenter needs to measure both the composition and the temperature of the exhaust products, among other variables such as the fuel flow rate and composition, the combustion air flow rate, the furnace pressure, and the furnace skin temperature. If the actual water content in the exhaust products is not measured, which is often the case, it can be calculated knowing the other components in the stream and the fuel composition and flow rate. The furnace air leakage can be calculated based on the flue gas composition and combustion air flow rate. The point is that the experimenter should carefully check
example of four water-cooled bayonets inserted into the floor of a test furnace [51]. There are several variables that can be adjusted to change the heat load. The insertion depth into the furnace, the number of bayonets inserted, and the water-cooling flow rate to each bayonet can all be varied to change the furnace temperature. Another method used to control test furnace temperature is to vary the amount of insulation in front of a water-cooled wall or tube. Adding insulation increases the furnace temperature and removing insulation decreases it.
1.5 Experimental Errors
© 2011 by Taylor and Francis Group, LLC
34
the analysis that will be done to ensure that the necessary data will be collected. Another fairly obvious method that should be used to minimize experimental error is frequent calibration. For example, if an analyzer is used to measure the exhaust gas composition, the analyzer should be calibrated at least daily with a suitable span gas cylinder. It would even be better to calibrate the analyzer more than once a day. For example, it could be calibrated at the beginning of the day, in the middle of the day, and at the end of the day. If these multiple calibrations consistently show negligible drift, then one calibration per day may be justified. Related to the frequent calibration is using the appropriate measurement scale. Again using the example of a gas analyzer, many of these instruments have multiple scales for measuring gas composition. For example, an oxygen analyzer might have multiple scales such as 0–1%, 0–5%, 0–10%, 0–25% and 0–100%. The appropriate scale should be used when making measurements to minimize the error that is often a percent of the fullscale reading. In most typical industrial combustion tests, the excess O2 in the exhaust gases is generally 2–4%, depending on the type of fuel, the burner design, and several other parameters. Then, the appropriate scale to use would be 0–5%. If the 0–100% scale were used instead and the error in the reading is 2% of the full scale, then the error in using the 0–100% scale would be 2%. This means that a 3% excess O2 reading on a 0–100% scale with a 2% error could actually be anywhere from 1 to 5%, which is unacceptable accuracy. However, if the 0–5% scale were used, the error would be 2 of 5%, which is 0.1%. This means that a 3% excess O2 reading with a 2% error on a 0–5% scale could actually be anywhere from 2.9 to 3.1%, which is much better accuracy than 1–5% for the 0–100% scale. It is important that a suitable calibration standard be used for the measurement technique. As an example, for gas analyzer calibration, it is generally recommended that the calibration gas be 80–90% of the full-scale reading. For example, if the 0–5% scale will be used to measure O2, then the calibration gas should have approximately 80–90% of the full-scale reading of 5%, which means the calibration gas should have 4.0–4.5% O2. One reason to specify slightly less than the full scale is that the span gas supplier may be slightly off in their gas blending and normally has some allowable tolerance in the gas composition they supply. For example, if the supplier has a 5% tolerance in the actual blend they supply and the experimenter requests a blend with 5% O2, the actual span gas supplied could be as high as 5.25% O2, which is above the range of the 0–5% scale and therefore could not be used to span that scale. Wherever possible, instruments should be calibrated in place when they are connected to the measurement
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
system. For example, if a thermocouple is being used to measure temperature, ideally it should be calibrated with a portable calibrator while it is connected to the data acquisition system. If necessary, the thermocouple can be temporarily removed from its measurement location to insert it into the calibrator. The important factor is that it should be electrically connected to the data acquisition system. This will help identify obvious mistakes such as mistakenly setting up an actual Type R thermocouple as a Type K thermocouple in the data acquisition system. If the thermocouple was disconnected, removed from the system and calibrated somewhere else, this mistake may not be immediately recognized, which would result in bad data. This is a systematic error that could be corrected after the fact with a proper conversion. Some other errors may not be as easy to correct after the experiments are completed. Another way to minimize experimental error is to use multiple methods to measure the same variable. Suppose the experimenter is trying to measure the draft (pressure) in a process heater. There are several devices commonly used to measure draft. Figure 1.47 shows an inclined manometer, which is typically filled with red gauge oil that is manually read off a scale. Figure 1.48 shows a mechanical gauge where a needle deflects and the reading is manually read off a gauge. Figure 1.49 shows an electronic manometer where a transducer sends an electronic signal that can be displayed on a computer screen. The experimenter might use two or three of these devices to measure the draft, or maybe only one of them that is periodically compared against another one to ensure accuracy. For example, the electronic manometer might be used for ease of data collection, but manual readings might be taken periodically to corroborate the electronic readings. When possible, it is good practice to make repeated readings of the same experimental conditions. For
Figure 1.47 Inclined manometer.
35
Introduction
Figure 1.48 Capsuhelic gauge. (From Baukal, C. E., Industrial Combustion Pollution and Control, New York: Marcel Dekker, 2004.)
example, assume that test condition #1 is the base case. If only one day’s worth of testing will be done, then the base case conditions should be tested at least twice but not consecutively. Maybe the first and last tests of the day are the base case. More replications would be even better but time and budget constraints may preclude more than one replication. If the testing will take place over multiple days, then the base case may be replicated on several of those days. This data will give the experimenter an idea of the repeatability of the tests. If the same test done at two different times on the same day gives wildly different results, then the experimenter will have little confidence in the data and should check for sources of experimental error that can be fixed or appropriate corrections can be made if the actual test conditions are not exactly the same. For example, it has been shown that ambient air humidity can significantly impact NOx measurements [54]. If the base case conditions were tested in the morning when the humidity was low and then again late in the afternoon when the humidity was high, then the NOx readings are likely to be somewhat different. Therefore, the experimenter should measure the humidity to account for this effect on NOx. While the experimenter cannot control the humidity, the measured humidity can be used to compare NOx measurements made under different humidity conditions. Chapter 3 discusses general principles of experimental design that can be used to efficiently collect data with a minimum number of tests, with sufficient replications to validate the data measurement and collection techniques.
© 2011 by Taylor and Francis Group, LLC
Figure 1.49 Electronic pressure gauge. (From Baukal, C. E., Industrial Combustion Pollution and Control, New York: Marcel Dekker, 2004.)
1.5.3 Uncertainty Analysis Kline and McClintock (1953) developed a method to estimate the uncertainty in experimental data [55]. In some cases, some primary measurements like temperature, pressure, and flow are combined to calculate another result:
R = R ( x1 , x2 , x3 ,…, xn ) ,
(1.4)
where R = the result of interest and xi are the variables involved in the calculation. Then, the uncertainty in the result R (δR) is calculated using: 2 2 2 ∂R ∂R ∂R δR = δx1 + δx 2 + … + δx n ∂x 2 ∂x n ∂x1
12
(1.5)
where ∂R/∂xi is the partial derivative of the equation for R with respect to the variable xi and δxi is the uncertainty in the parameter xi. An example will be used to demonstrate the technique. In the study by Baukal (2009), the heat flux from an impinging flame was measured using a disk calorimeter segmented into six concentric sections as shown
36
Industrial Combustion Testing
in Figure 1.50 [56]. The total heat flux (qi”) to each calorimeter section, i, was determined as follows:
qi′′=
i c p (tout i − tin i ) m , Ai
(1.6)
i was the cooling water mass flow rate, cp was where m the specific heat of the cooling water, tout was the outi let temperature of the cooling water, tin was the inlet i temperature of the cooling water, and Ai was the ring impingement surface area. The total heat flux was calculated using Equation 1.6. The uncertainty was calculated using: 2 2 ∂q ′′ ∂q ′′ + δq ′′ = × δc p × δm ∂c p ∂m
the lowest firing rate (qf = 5 kW) and oxidizer composition (O2 fraction in the oxidizer = Ω = 0.21 for air), the maximum axial spacing (nondimensional distance L = 6) and radial spacing (nondimensional radius Reff = 0.59. There, a high water cooling flow rate was used to reduce the high surface temperatures as much as possible. The uncertainty was so high because the temperature difference between the inlet and outlet cooling water temperatures was only 0.58 K. For the same firing rate and radial location, only a slightly higher oxidizer composition (Ω = 0.35) and a closer axial spacing (L = 0.5), the uncertainty dropped to 15%. The maximum uncertainty for all other calorimeters was 12%. Again that occurred at the lowest firing rate, where the temperature differences were smaller. For the most commonly used firing rate (qf = 15 kW), the maximum uncertainty for any calorimeter was 9.2%. Typical uncertainties were 6–8%.
2 12
∂q ′′ ∂q ′′ ∂q ′′ × δA , + × δtout + × δtin + ∂A ∂tout ∂tin (1.7) 2
2
where δm = 0.3%, δcp = ±0.1%, δtin and δtout were determined from calibrations, and δA was determined using:
2 2 ∂A ∂A δA = × δdinner + × δdouter ∂douter ∂dinner
12
, (1.8)
where dinner and douter were the inner and outer diameters of each calorimeter ring, respectively. The maximum uncertainty in the heat flux was 58%. This occurred at
1
2
3
4
5
6
Figure 1.50 Disk calorimeter segmented into six concentric sections. (From Baukal, C. E., Heat Transfer from Flame Impingement Normal to a Plane Surface, Saarbrücken, Germany: VDM Verlagsservicegesellschaft mbH, 2009.)
© 2011 by Taylor and Francis Group, LLC
1.6 Combustion Testing Resources 1.6.1 General References Specific references are provided with the various techniques discussed in other chapters in this book. Those references give more details for the interested reader. Several excellent general references have been published concerning combustion diagnostics. Surugue (1961) edited a manual (in both English and French) for the Advisory Group for Aeronautical Research and Development (part of NATO) that contains papers on making experimental measurements in combustion [57]. The book by Beér and Chigier (1972) gives detailed descriptions and diagrams of different probes (e.g., surface and suction pyrometers, ellipsoidal radiometer, gas sampling, pitot tube) used to make measurements in industrial-scale flames [58]. This book is particularly helpful for those planning to use physical probes as the authors describe in detail the design of various probes and how to make specific measurements including calibration and error correction if appropriate. Zinn (1977) edited a comprehensive book on experimental diagnostics in gas phase combustion systems, which includes both probe and optical techniques [59]. Gupta and Lilley (1985) describe how to make flow field measurements, including flow visualization, in combustion [60]. The report by Okoh and Brown (1988) [61] reviews and compares many techniques for the measurement of a wide range of variables, in combustion systems. It also includes equipment specifications and suppliers. The books edited by Durão et al. (1992) [62] and Taylor (1993) [63] contain chapters on the use of both physical probes and optical techniques. Fristrom (1995) has
37
Introduction
several chapters devoted to probe and optical measurements in flames [64]. The papers and chapters by Fristrom (1976) [65], Bowman (1977) [66], Gouldin (1980) [67], and Becker (1993) [48] review probe measurements in flames. Newbold et al. (1996) gave a good example of making probe measurements of velocity, species, radiation, and gas temperature in an industrial, gas-fired, flat-glass furnace [68]. Mullinger and Jenkins (2008) [22] describe many types of physical probes and present representative results of measurements made in industrial processes. There are many advanced, mostly laser-based, techniques that have been successfully used in laboratory flames. These techniques include, for example, Raman and Rayleigh scattering, coherent antistokes Raman scattering (CARS), degenerate four wave mixing (DFWM), Mie scattering, laser-induced fluorescence (LIF), laser-induced incandescence (LII), resonantly enhanced multiphoton ionization (REMPI), tunable diode laser (TDL) absorption, and laser Doppler velocimetry (LDV). Solomon et al. (1986) reviewed Fourier Transform Infrared (FT-IR) optical techniques for measuring particles, species concentrations, and temperature in combustion processes [69]. The books by McCay and Roux (1984) [70], Eckbreth (1988) [71], Chigier (1991) [72], and Kohse-Höinghaus and Jeffries (2002) [73] focus specifically on optical techniques for diagnosing flames. Papers are available on the use of laser diagnostics in combustion systems [74–79]. Advanced optical techniques permit spatially, temporally, and spectrally resolved measurements that are difficult if not impossible to make with traditional probes. Many of these advanced techniques have proven to be a challenge in industrial-scale flames and are generally beyond practical implementation by most industrial combustion researchers. The cost of implementing these techniques on a large scale is often prohibitive. The environmental conditions such as high temperatures and vibration also make implementation more difficult in large-scale combustors. Therefore, most of the techniques described in this book concern physical probes that have been proven in large-scale combustion systems, although there are some chapters dedicated to optical techniques as well. 1.6.2 Test Facilities There are relatively few industrial-scale combustion test facilities outside of equipment manufacturers. A report prepared from a workshop sponsored by the U.S. Department of Energy states, “For the most part, the size and type of laboratory test equipment available are inadequate, and the costs are prohibitive” ([4], p. 11). This section is intended to provide a sample of some of those industrial-scale test facilities, although there are others as well. They are broadly categorized
© 2011 by Taylor and Francis Group, LLC
as academic and institutional. While this is somewhat arbitrary, academic refers to universities (web addresses generally have an “edu” suffix) and institutional refers to organizations that are not universities (although some are affiliated with universities) and are typically not-for-profit (web addresses generally have an “org” suffix). Some of the authors in this book work for some of these organizations that are listed in alphabetical order in each category. 1.6.2.1 Academic There are not many universities that conduct industrial-scale research, mostly because of the cost of building and operating such facilities. Adelaide University (Adelaide, Australia) (www.adelaide.edu.au) has done some research on large-scale flares in its Turbulence, Energy & Combustion (TEC) group. The TEC consists of members from the mechanical and chemical engineering schools. Brigham Young University (BYU) in Provo, Utah (www.byu.edu) and the University of Utah (U of U) in Salt Lake City, Utah (www.utah.edu) formed a cooperative consortium called the Advanced Combustion Engineering Research Center (ACERC; www-acerc.byu. edu). The mission of ACERC is “to develop, apply, and transfer advanced combustion technology to industry through fundamental engineering research and educational programs aimed at the solution of critical combustion problems.” The ACERC has an international reputation for conducting applied research in a wide range of areas such as coal, biomass, black liquor, forest fires, and gas turbines. The University of Utah has a number of industrial-scale combustors used for making a wide range of measurements. Cranfield Institute of Technology (Cranfield, United Kingdom; www.cranfield.ac.uk) has a gas turbine engineering lab in an off-campus site with multiple test cells for conducting large-scale research as part of its energy program. Osaka Gas in Osaka, Japan (www.osakagas.co.jp) and Tokyo Gas in Tokyo, Japan (www.tokyo-gas.co.jp) both conduct a wide range of research for industrial combustion applications. Some examples include cogeneration, advanced fuel-air mixing, gas turbines, oxy-natural gas combustion, and biogas. Università di Pisa in Pisa, Italy (www.unipi.it) is associated with the IFRF that is discussed below in the next section. University of California at Irvine in Irvine, California (www.uci.edu) has done research on industrial burners using natural gas in their UCI Combustion Laboratory (UCICL; www.ucicl.uci.edu). The UCICL has also done research on active control, emissions, gas turbine combustion, and liquid fuels and sprays.
38
1.6.2.2 Institutional There are a number of well-known institutions that conduct industrial-scale combustion testing. CANMET Energy Technology Centre is located in Ottawa, Ontario (Canada; canmetenergy-canmetenergie.nrcan-rncan. gc.ca) and is part of Natural Resources Canada. CANMET is a leader in clean energy research and technology development and develops technologies in many industrial areas including, for example, furnaces, boilers, oxy-fuel combustion, process integration, and NOx reduction. Chapters 11 and 13 discuss some of the modeling and experimental data generated by CANMET. The ENEA headquartered in Rome, Italy (www.enea. it) is the Italian National Agency for New Technologies, Energy and the Environment. ENEA has labs located in multiple locations in Italy and conducts a wide range of industrial combustion ranging in many application areas with particular emphasis on the environment, climate change, and sustainability. Chapter 12 discusses advanced laser measurements made at ENEA. Gas Technology Institute (GTI) located in Des Plaines, Illinois (www.gastechnology.org) has been a leader in the development of technology solutions in the natural gas industry for more than 65 years. The institute has a wide range of industrial-scale test capabilities and has done testing in essentially all major application areas. They have developed numerous technologies such as a forced internal recirculation burner, oxygen-enriched air staging, methane de NOx®, a high-luminosity oxygas burner, and the super boiler. Some experimental results from GTI are discussed in Chapter 19. The International Flame Research Foundation (IFRF; www.ifrf.net) in Livorno, Italy has long been recognized as one of the premier organizations for industrial combustion research. The IFRF has published numerous research studies over the many decades of its existence. The foundation has conducted research in its own facilities and in conjunction with many of its members in a wide range of industries including, for example, ferrous and nonferrous metals, cement, lime, biomass, and boilers. The IFRF has an online handbook that may be of interest to researchers.
Industrial Combustion Testing
are expected to continue to reduce over time. Energy efficiency is also expected to be of interest as well. A wide range of testing configurations will continue to be used to simulate realistic operating conditions. It is expected that field testing in particular, with all of its limitations, will continue to be an important component of a combustion technology development program, especially since most end users will be unwilling to use a technology that has not been verified on a large scale, preferably in an operating plant. Laboratory testing will also continue to be important because of its convenience, costeffectiveness, controllability, and accuracy. Initial testing can be done more quickly, less expensively, with better control, and with more accuracy than field testing. New technology development typically needs to be proven in the laboratory first before proceeding to field testing. The use of optical instrumentation is expected to increase in the future because it has some important advantages compared to physical probes. The main barrier to using many of the optical techniques has been adapting them for use in an industrial environment. More experience continues to be gained using these sophisticated techniques in the harsh environments often present in industry. Industrial combustion testing should continue to be important well into the foreseeable future. While computer modeling continues to grow in popularity, current models can not replicate the complete physics of an industrial combustion system because of computer hardware limitations and because not all of the physics are completely known. In nearly all models, simplifications need to be made to get results in a reasonable amount of time. In addition, those models still need experimental data for validation. Testing is needed because of the complexity of combustion problems. It is expected that computer modeling and industrial scale testing will be used together as complementary tools to further improve industrial combustion technologies.
References
1.7 Future Industrial combustion is expected to continue to play a prominent role in the future. It is directly related to the vast majority of energy consumption in the world and indirectly related to many of the products people use on a daily basis. The increasing attention on the environment means that industrial combustion pollution emissions
© 2011 by Taylor and Francis Group, LLC
1. U.S. Department of Energy (DOE). Industrial Combustion Vision: A Vision by and for the Industrial Combustion Community. Washington, DC: U.S. DOE, 1998. 2. Hewitt, G. F., Shires, G. L., and Bott, T. R. Process Heat Transfer. Boca Raton, FL: CRC Press LLC, 1994. 3. U.S. Department of Energy, Energy Information Administration. Annual Energy Outlook 2008, report DOE/ EIA-0384 (2008), Washington, DC, released June 26, 2009. 4. Energetics, Inc. Industrial Combustion Technology Roadmap: A Technology Roadmap by and for the Industrial Combustion Community, October 2002, www1.eere.energy.gov/ industry/combustion/pdfs/combustion_roadmap2002. pdf (accessed July 29, 2009).
Introduction
5. Barry, P., and Somers, S. “Duct Burners.” In The John Zink Combustion Handbook, edited by C. E. Baukal, 523–44. Boca Raton, FL: CRC Press, LLC, 2001. 6. Baukal, C. E., ed. Oxygen-Enhanced Combustion. Boca Raton, FL: CRC Press LLC, 1998. 7. Strehlow, R. A. Fundamentals of Combustion. Scranton, PA: Inter. Textbook Co., 1968. 8. Williams, F. A. Combustion Theory. Menlo Park, CA: Benjamin/Cummings Publishing, 1985. 9. Lewis, B., and von Elbe, G. Combustion, Flames and Explo sions of Gases. 3rd ed. New York: Academic Press, 1987. 10. Bartok, W., and Sarofim, A. F., eds. Fossil Fuel Combustion. New York: John Wiley & Sons, Inc., 1991. 11. Fristrom, R. M. Flame Structure and Processes. New York: Oxford University Press, 1995. 12. Glassman, I. Combustion. 3rd ed. New York: Academic Press, 1996. 13. Edwards, J. B. Combustion: The Formation and Emission of Trace Species. Ann Arbor, MI: Ann Arbor Science Publishers, 1974. 14. Barnard, J. B., and Bradley, J. N. Flame and Combustion. 2nd ed. London: Chapman and Hall, 1985. 15. Turns, S. R. An Introduction to Combustion. 2nd ed. New York: McGraw-Hill, 2000. 16. Griswold, J. Fuels, Combustion and Furnaces. New York: McGraw-Hill, 1946. 17. Stambuleanu, A. Flame Combustion Processes in Industry. Tunbridge Wells, UK: Abacus Press, 1976. 18. Perthuis, E. La Combustion Industrielle. Paris: Éditions Technip, 1983. 19. Borman, G., and Ragland, K. Combustion Engineering. New York: McGraw-Hill, 1998. 20. Deshmukh, Y. V. Industrial Heating. Boca Raton, FL: CRC Press, 2005. 21. Keating, E. L. Applied Combustion. Boca Raton, FL: CRC Press, 2007. 22. Mullinger, P., and Jenkins, B. Industrial and Process Furnaces. Oxford, UK: Butterworth-Heinemann, 2008. 23. Segeler, C. G., ed. Gas Engineers Handbook. New York: Industrial Press, 1965. 24. Reed, R. D. Furnace Operations. 3rd ed. Houston, TX: Gulf Publishing, 1981. 25. Pritchard, R., Guy, J. J., and Connor, N. E. Handbook of Industrial Gas Utilization New York: Van Nostrand Reinhold, 1977. 26. Reed, R. J. North American Combustion Handbook. Vol. I, 3rd ed. Cleveland, OH: North American Mfg. Co., 1986. 27. IHEA. Combustion Technology Manual. 5th ed. Arlington, VA: Industrial Heating Equipment Assoc., 1994. 28. Kistler, M. D., and Becker, J. S. “Ferrous Metals.” In Oxygen-Enhanced Combustion, edited by C. E. Baukal. Boca Raton, FL: CRC Press, 1998. 29. Saha, D., and Baukal, C. E. “Non-Ferrous Metals.” In Oxygen-Enhanced Combustion, edited by C. E. Baukal. Boca Raton, FL: CRC Press, 1998. 30. Fedler, R. M., and Rousseau, R. W. Elementary Principles of Chemical Process. 3rd ed. New York: John Wiley & Sons, Inc., 2000. 31. Kudra, T., and Mujumdar, A. S. Advanced Drying Technologies. New York: Marcel Dekker, 2002.
© 2011 by Taylor and Francis Group, LLC
39
32. Baukal, C. E. Heat Transfer in Industrial Combustion. Boca Raton, FL: CRC Press, 2000. 33. Reese, J. L., Moilanen, G. L., Borkowicz, R., Baukal, C., Czerniak, D., and Batten, R. “State-Of-The-Art of NOx Emission Control Technology.” ASME paper 94-JPGCEC-15, Proceedings of International Joint Power Generation Conference, Phoenix, AZ, October 3–5, 1994. 34. Joshi, M. L., Tester, M. E., Neff, G. C., and Panahi, S. K. “Flame Particle Seeding with Oxygen Enrichment for NOx Reduction and Increased Efficiency.” Glass 68, no. 6 (1990): 212–13. 35. Ruiz, R., and Hilliard, J. C. “Luminosity Enhancement of Natural Gas Flames.” Proceedings of 1989 International Gas Research Conference, edited by T. L. Cramer, 1345–53. Rockville, MD: Govt. Institutes, 1990. 36. Slavejkov, A. G., Gosling, T. M., and Knorr, R. E. “LowNOx staged combustion device for controlled radiative heating in high temperature furnaces.” U.S. patent 5,611,682, March 18, 1997. 37. API Publication 535. Burner for Fired Heaters in General Refinery Services. 1st ed. Washington, DC: American Petroleum Institute, July 1995. 38. Fioravanti, K. J., Zelson, L. S., and Baukal, C. E. “Flame Stabilized Oxy-Fuel Recirculating Burner.” U.S. Patent 4,954,076, September 4, 1990. 39. Baukal, C. E., ed. The John Zink Combustion Handbook. Boca Raton, FL: CRC Press, 2001. 40. Joshi, S. V., Becker, J. S., and Lytle, G. C. “Effects of Oxygen Enrichment on the Performance of Air-Fuel Burners.: In Industrial Combustion Technologies, edited by M. A. Lukasiewicz, 165. Materials Park, OH: American Society of Metals, 1986. 41. Longacre, S. “Using Infrared to Dry Paper and Its Coatings.” Process Heating 4, no. 2 (1997): 45–49. 42. Smith, T. M., and Baukal, C. E. “Space-Age Refractory Fibers Improve Gas-Fired Infrared Generators for Heat Processing Textile Webs.” Journal of Coated Fabrics 12, no. 3 (January 1983): 160–73. 43. Reed, R. J. “Future Consequences of Compact, Highly Effective Heat Recovery Devices.” In Heat Transfer in Furnaces, edited by C. Presser and D. G. Lilley, 23–28, Vol. 74. New York: ASME HTD, 1987. 44. Baukal, C. E., Gershtein, V. Y., and Li, X. M., eds. Computational Fluid Dynamics in Industrial Combustion. Boca Raton, FL: CRC Press, 2001. 45. Bussman, W., and Baukal, C. “Ambient Condition Effects on Process Heater Emissions.” Proceedings of the International Mechanical Engineering Congress & Exhibition, Paper IMECE2008-68284, Boston, MA, November 2008. 46. Baukal, C. E. Industrial Combustion Pollution and Control. New York: Marcel Dekker, 2004. 47. Baukal, C. E., ed. Industrial Burners Handbook. Boca Raton, FL: CRC Press, 2004. 48. Becker, H. A. “Physical Probes.” In Instrumentation for Flows with Combustion, edited by A.M.K.P. Taylor, 53–112. London, UK: Academic Press, 1993. 49. Weber, R. “Scaling Characteristics of Aerodynamics, Heat Transfer, and Pollutant Emissions in Industrial Flames.” 26th Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, pp. 3343–54, 1996.
40
50. Bussman, W., and Baukal, C. “The Effect of Firebox Temperature on NOx Emissions.” Presented at 2004 Air & Waste Management Association annual meeting, paper #04-A-664-AWMA, Indianapolis, IN, June 2004. 51. Bussman, W., Baukal, C., and French, K. “Variable Test Furnace Cooling.” Presented at the 2005 Summer Heat Transfer Conference, paper HT2005-72012, San Francisco, CA, July 17–22, 2005. 52. Holman, J. P. Experimental Methods for Engineers. 5th ed. New York: McGraw-Hill, 1989. 53. Harrison, S. J., and Dieck, R. H. “Measurement Errors and Uncertainty.” In The Engineering Handbook, edited by R. C. Dorf, 2nd ed. Boca Raton, FL: CRC Press, 2005. 54. Bussman, W., and Baukal, C. “Ambient Conditions Impact CO and NOx Emissions: Part I.” Petroleum Technology Quarterly 14, no.3 (2009): 93–99. 55. Kline, S. J., and McClintock, F. A. “Describing Uncertainties in Single-Sample Experiments.” Mechanical Engineering (January 1953): 3–8. 56. Baukal, C. E. Heat Transfer from Flame Impingement Normal to a Plane Surface. Saarbrücken, Germany: VDM Verlagsservicegesellschaft mbH, 2009. 57. Surugue, J., ed. Experimental Methods in Combustion Research: A Manual. New York: Pergamon Press, 1961. 58. Beér, J. M., and Chigier, N. A. Combustion Aerodynamics. London: Applied Science Publishers, 1972. 59. Zinn, B. T., ed. Experimental Diagnostics in Gas Phase Combustion Systems. Washington, DC: American Institute of Aeronautics and Astronautics, 1977. 60. Gupta, A. K., and Lilley, D. G. Flowfield Modeling and Diagnostics. Tunbridge Wells, UK: Abacus Press, 1985. 61. Okoh, C. I., and Brown, R. A. Combustion Experimentation Handbook. Chicago, IL: Gas Research Institute report GRI-88/0143, 1988. 62. Durão, D. F. G., Heitor, M. V., Whitelaw, J. H., and Witze, P. O. Combusting Flow Diagnostics. Dordrecht, The Netherlands: Kluwer Academic Publishers, 1992. 63. Taylor, A.M.K.P. Instrumentation for Flows with Combustion. London: Academic Press, 1993. 64. Fristrom, R. M. Flame Structure and Processes. New York: Oxford University Press, 1995. 65. Fristrom, R. M. “Probe Measurements in Laminar Combustion Systems.” In Combustion Measurements: Modern Techniques and Instrumentation, edited by R. Goulard. New York: Academic Press, 1976. 66. Bowman, C. T. “Probe Measurements in Flames.” Progress in Astronautics and Aeronautics 53, (1977): 1–24. 67. Gouldin, F. C. “Probe Measurements in Multi-Dimensional Reacting Flows.” In Testing and Measurement Techniques in Heat Transfer and Combustion, AGARD CP-281, paper #4, 1980.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
68. Newbold, J., McQuay, M. Q., Webb, B. W., and Huber, A. M. “The Experimental Characterization of the Combustion Process in an Industrial, Gas-Fired, FlatGlass Furnace.” 29th International ISATA Conference: Automotive Technology & Automation, Florence, Italy, Automotive Association Ltd., Vol. 2, 967–76, June 1996. 69. Solomon, P. R., Best, P. E., Carangelo, R. M., Markham, J. R., Chien, P-L, Santoro R. J., and Semerjian, H. J. “FT-IR Emission/Transmission Spectroscopy for In Situ Combustion Diagnostics.” 21st Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 1763–71, 1986. 70. McCay, T. D., and Roux, J. A. Combustion Diagnostics by Nonintrusive Methods. New York: American Institute of Aeronautics and Astronautics, 1984. 71. Eckbreth, A. C. Laser Diagnostics for Combustion Temperature and Species. Cambridge, MA: Abacus Press, 1988. 72. Chigier, N. Combustion Measurements. New York: Hemisphere Publishing Corp., 1991. 73. Kohse-Höinghaus, K., and Jeffries, J. B., eds. Applied Combustion Diagnostics. New York: Taylor & Francis, 2002. 74. Eckbreth, A. C. “Recent Advances in Laser Diagnostics for Temperature and Species Concentrations in Combustion.” 18th Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 1471–88, 1980. 75. Penner, S. S., Wang, C. P., and Bahadori, M. Y. “Laser Diagnostics Applied to Combustion Systems.” 20th Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 1149–76, 1984. 76. Hanson, R. K. “Combustion Diagnostics: Planar Imaging Techniques.” 21st Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 1677–91, 1986. 77. Kohse-Höinghaus, K. “Laser Techniques for the Quantitative Detection of Reactive Intermediates in Combustion Systems.” Progress in Energy and Combustion Science 20 (1994): 203–79. 78. Fornaciari, N. R., Schefer, R. W., Walsh, P. M., and Claytor, L. E. “Application of Laser-Based Diagnostics to Industrial Scale Burners.” Proceedings of 1995 International Gas Research Conference, edited by D. A. Dolenc, 2398– 2405. Rockville, MD: Govt. Institutes, 1996. 79. Perrin, M., Imbach, J., Albert, S., Mariasine, J., and Quinqueneau, A. “Application of Advanced Instantaneous In-Flame Measurements Techniques in an Industrial Flame with Preheated Air.” Proceedings of 1995 International Gas Research Conference, edited by D. A. Dolenc, 2406–15. Rockville, MD: Govt. Institutes, 1996.
2 Testing Safety Charles E. Baukal, Jr. Contents 2.1 Introduction...................................................................................................................................................................... 41 2.2 Safety Review................................................................................................................................................................... 42 2.3 Hazards............................................................................................................................................................................. 44 2.3.1 Excessive Temperature........................................................................................................................................ 44 2.3.2 Thermal Radiation............................................................................................................................................... 45 2.3.3 Noise...................................................................................................................................................................... 46 2.3.4 High Pressure....................................................................................................................................................... 51 2.3.5 Explosion............................................................................................................................................................... 52 2.3.6 Flame Instability.................................................................................................................................................. 53 2.3.7 Environmental...................................................................................................................................................... 54 2.4 Accident Prevention........................................................................................................................................................ 55 2.4.1 Ignition Control.................................................................................................................................................... 55 2.4.2 General.................................................................................................................................................................. 56 2.5 Accident Mitigation......................................................................................................................................................... 58 2.6 Recommendations........................................................................................................................................................... 60 References................................................................................................................................................................................... 60
2.1 Introduction Industrial combustion testing can be dangerous for many reasons. As stated in Chapter 1, one of the reasons for testing is to try new concepts as part of research and development. This often means that there is some uncertainty in the result, hence the need for testing. In some cases, this uncertainty could mean potentially dangerous conditions. For example, flameless combustion (see Chapter 23) could be dangerous because the combustor must be above the autoignition temperature of the air–fuel mixture or an explosion could occur. It is also potentially dangerous because flameless combustion is usually difficult to detect with traditional optical flame detectors. This means an alternative method, such as furnace temperature, may be needed to shut off the fuel supply to the furnace if the flame goes out. Fires and explosions are a major concern in industrial combustion processes and account for as much as 95% of the losses in accidents in the process industries [1]. Figure 2.1 shows the bulging walls in a test heater that resulted from an overpressure event caused by a rapid deflagration. The consequences of a fire or an explosion
in a chemical or petrochemical plant, for example, can be very severe and very public because of the high volume of flammable liquids and gases handled in those plants [2,3]. Burning large quantities of fuel means appropriate precautions must be taken to prevent equipment damage and personnel injury. There are many factors that can contribute to an accident [4]: human error [5,6], equipment malfunction, upset conditions, fire or explosion near the apparatus [7], improper procedures, and severe weather conditions. There are also many potential dangers caused particularly by fires and explosions such as flying shrapnel, pressure waves from a blast, high heat loads from flame radiation [8–10], and high temperatures. All of these can have severe consequences to both people and equipment and may need to be considered in minimizing the potential impact of an incident. Kletz (1998) [11] lists four circumstances that are frequent causes of accidents or dangerous conditions including performing or preparing for maintenance, making modifications to furnace design, human error, and labeling errors or omissions. When preparing for maintenance, it is important to remove hazards from the maintenance area, isolate the area and/or equipment
41 © 2011 by Taylor and Francis Group, LLC
42
Industrial Combustion Testing
rocedures, and training documentation. Feedback p from each of these documentation elements are linked together as part of a plant’s overall process safety program. This chapter is not intended to be exhaustive, but is designed to highlight some of the more common safety concerns for industrial combustion processes. Additional references are available on combustion safety for the interested reader [12–18]. The National Fire Protection Association (NFPA) is a good source of information including producing some important standards related to industrial combustion safety (e.g., [19–22]). Other examples of sources include the European Committee for Standardization (CEN) EN746-1 standard “Industrial Thermoprocessing Equipment—Common Safety Requirements for Industrial Thermoprocessing Equipment,” the Canadian Standards Association (CSA) B139-00 standard “Installation Code for Oil Burning Equipment,” and the Japanese Standards Association (JSA) JIS B 8415 standard “General Safety Code for Industrial Combustion Furnaces.” This chapter contains the following sections: safety review, hazards, accident prevention and mitigation, and recommendations.
2.2 Safety Review Figure 2.1 Test heater that has been overpressured.
from operational equipment, and follow maintenance procedures carefully. When modifying furnace design, even when the modification seems minor, the proposed modification should go through design procedures similar to the procedures used for the original installation of the equipment. Without appropriate design, it is not always possible to anticipate how a change in one piece of equipment will affect other equipment. Human error is sometimes caused by inattention or poor training, but is frequently caused by a deliberate attempt to shortcut a cumbersome procedure or to make an inconvenient piece of equipment more convenient to use. Accidents caused by labeling are frequently the result of out-of-date labeling, incorrect labeling, or no labeling at all resulting in the incorrect operation of equipment. Safety documentation and operator training provide the backbone of a strong safety program, and are abso lutely essential to maintain a safe combustion working environment. Safety documentation for combustion related processes includes design information, process hazard analysis (PHA) reports, standard operating
© 2011 by Taylor and Francis Group, LLC
A safety review is recommended before any type of combustion experiment, no matter how routine. The extent of the review will depend on how common the test is and the potential for damage or injury. For tests that are commonly done with relatively little chance for damage or injury, a short informal review may be all that is required. For less common tests with a high potential for damage or injury, a more extensive formal review is recommended that could take a significant amount of time to complete. Some type of checklist should be used to ensure that all appropriate factors have been considered in the review. There are various types of analyses that are used for a process hazard analysis (PHA) of the equipment design and test procedures, including the effects of human error. Qualitative methods include checklists, What-If, and Hazard and Operability (HAZOP) studies. Quantitative methods include Event Trees, Fault Trees, and Failure Modes and Effect Analysis (FMEA). All of these methods require rigorous documentation and implementation to ensure that all potential safety problems are identified and the associated recommendations are addressed. The review should also consider what personal protective equipment (PPE) is needed to protect workers from injuries.
Testing Safety
The review team should consist of those with appropriate experience in all of the relevant areas involved in the test. The engineer or scientist conducting the experiment should be part of the team, although not necessarily the team leader. At least one experienced technician or operator, preferably one who will be involved in the test, should participate in the review. Many organizations have someone responsible for safety who is normally part of such a review team and may lead the team as appropriate. If a formal review will be conducted, someone trained in the particular review process being used will often lead the process. Some specialists may be needed on the team. For example, if new advanced sensors or controls will be used in the test, then someone familiar with those technologies should be included. Often, someone involved in the management of the test facility will be involved in these safety reviews. Depending on the extent of the review and the potential for damage or injury, it is generally recommended that at least one person on the review team should not be part of the test. This person should be independent of the budgeting and scheduling so they can focus on safety and not be unduly influenced by other factors. Prior to conducting the review, the team leader should gather all appropriate documentation such as drawings and procedures and make sure they are up-to-date. Some new documentation may need to be created if it does not already exist. Each component should have a unique label that can be used in the review to identify various scenarios. For example, a fuel supply ball valve might have a label such as BV101. The review can then use this label to refer to conditions where BV101 might be open or closed. The documentation should also include performance specifications for the equipment such as the maximum flow rate, temperature, pressure, and so forth. If appropriate, vendor literature may also be needed to determine the limits of operation for a given piece of equipment. Formal reviews normally require appropriate documentation. For example, Excel spreadsheets or Word templates are available for conducting safety reviews to ensure that a specific process has been used in the analysis and to document the results of the review. The documentation should include the team members, the dates and times of the review, the functional area of each team member, the purpose of the review, the equipment and processes being analyzed, key specifications for the equipment (e.g., maximum flow rate), what codes and standards apply to the review if any, and a discussion of past incidents that may be relevant to what is being reviewed. The review should consider the equipment and materials in the vicinity of what is being reviewed. For example, fuel storage tanks or high voltage electrical equipment in
© 2011 by Taylor and Francis Group, LLC
43
the vicinity should be considered. All potential deviations (e.g., high flow or no flow when there should be flow) should be analyzed including possible causes for the deviation (e.g., human error or equipment failure), the likelihood (e.g., frequency) the event might occur, the consequences if it did occur, and the safeguards to help mitigate the consequences or prevent the event from occurring. This often means educated estimates as many of the potential events may never have occurred before. If necessary, if the current equipment or procedures are deemed inadequate, then appropriate recommendations should be made. In some cases, a recommendation might need further analysis to study a particular issue. Each recommendation should be assigned to someone, not necessarily a team member, with a completion date. The recommendations should be tracked to ensure they have been completed on a timely basis. The review team may recommend that certain recommendations be completed prior to conducting the experiment, while other recommendations may not need to be completed before testing can begin. Safety reviews should be scheduled far enough in advance that any recommendations from the reviews can be implemented in time so they do not adversely affect the testing schedule, especially if any equipment needs to be ordered or any new apparatus needs to be built. Waiting until the last minute for the review may unnecessarily pressure the review team into making hasty decisions or cause them to overlook possible incidents that could occur. This may mean that alternative team members may need to be used if primary team members are unavailable in the time frame of the review. Another aspect of a safety review that may need to be considered (depending on the nature of the test) is environmental permitting. For example, if a new type of fuel will be used that the test facility is not currently permitted to use, then the appropriate approvals will need to be obtained before testing can begin. This is often the case if a fuel contains hazardous components, such as benzene or chlorine-containing compounds. A related scenario is when the exhaust products may contain components that the facility is not currently permitted to emit. For example, if furans or dioxins may be potential effluents, the facility environmental permit may need to be checked to see if these are allowed and if so at what levels. Getting a new permit or a variance of an existing permit can be a lengthy process, so be sure to allow enough time for this. If additional treatment equipment is required to reduce pollution emissions, this can further lengthen the process. For example, obtaining equipment to remove furans or dioxins from the exhaust gases can be both expensive and lengthy, so plan accordingly.
44
Industrial Combustion Testing
Combustion typically involves high temperatures that can cause equipment damage and personnel injuries if not properly handled. For production equipment in operating plants, there is often a maximum surface temperature limit dictated by government regulations. For example, the U.S. Occupational Safety and Health Administration (OSHA) guidelines limit the maximum external temperature of a surface exposed to personnel to 140°F (60°C). If the temperature of equipment in the vicinity of personnel exceeds that temperature, then some type of shielding (see Figure 2.2) is usually required to prevent people from accidentally touching hot surfaces that could burn their skin. An alternative is
to add more insulation to the hot surface to reduce the skin temperature (see Figure 2.3). For any experiments that might have high external surface temperatures that could be contacted by personnel, appropriate precautions must be taken to prevent burn injuries. Equipment could be damaged by excessive temperatures. This includes equipment outside the combustion chamber as well as inside. For example, external equipment such as valves for controlling fuel and air flows may have Teflon® seals that could be damaged by high temperatures. Test setups should be analyzed to ensure the equipment can either handle the temperatures that might be encountered or that shields are in place to protect any equipment that could be adversely affected by the heat. Equipment inside the combustion chamber must be capable of handling the temperatures that may be generated. For example, metal furnace walls are normally protected by insulation rated to a suitable temperature. Any uninsulated metal that will be directly exposed to the heat may need to be cooled. The tubes in a process heater are cooled by the hydrocarbon fluids flowing through them. The tubes in a boiler are cooled
Figure 2.2 Metal mesh shielding personnel from hot exhaust stack.
Figure 2.3 Insulated temporary ductwork.
2.3 Hazards There are many potential hazards that may need to be considered in industrial combustion testing. Some of the common ones are briefly considered next. 2.3.1 Excessive Temperature
© 2011 by Taylor and Francis Group, LLC
45
Testing Safety
by water flowing through them. Instrumentation such as metal gas sampling probes is often water cooled to protect the metal and to quench the gas sample as part of the measurement process (see Chapter 7). In some cases, the temperatures may not be known a priori in the case of experimental conditions that have not been previously tested. While calculations may give an estimate of the expected temperatures, it is generally advisable to include some instrumentation, such as thermocouples and infrared detectors, to measure the actual temperatures. If there are any areas that are a particular concern for high temperatures, warnings and alarms could be added to alert operators. Automatic shutdowns could even be added to prevent equipment damage. 2.3.2 Thermal Radiation Thermal radiation is related to excessive temperature, but no direct contact is necessary for damage or injury to occur that distinguishes it here from the hazard caused by contacting high temperature surfaces. Thermal radiation is a necessary and important phenomenon in nearly all industrial combustion processes [23]. Figure 2.4 shows thermal radiation coming from the viewport of a hot furnace. It is typically the predominant form of heat transfer from the flame to the combustor walls and heat load. Most combustion engineers design systems to optimize the radiation heat transfer to get the proper temperature distribution in the combustor. Some burners are deliberately designed to maximize the radiation from the flame by creating luminous flames [24]. Excessive external thermal radiation can cause equipment damage and injure personnel. Electronic equipment in particular is susceptible to damage from high radiant loadings. External radiation can melt plastic parts such as valve seals, dry out lubricated parts such as motors, and make equipment operation more difficult where, for example, operators may need to wear gloves to open or close valves. Figure 2.5 shows a corrugated
Figure 2.4 Radiation from a viewport. (From Baukal, C.E., Industrial Combustion Pollution and Control, New York: Marcel Dekker, 2004.)
stainless steel fence designed to shield combustion flow control equipment, especially the electronics, from thermal radiation during flare testing (see Chapter 30). An even more serious concern is premature ignition caused by thermal radiation of the air–fuel mixture prior to reaching the burner outlet, where the fuel and oxidant are premixed upstream of the burner outlet. Personnel injury from excessive thermal radiation should be considered in most industrial combustion applications. Figure 2.6 shows an instrument for measuring thermal radiation that can be used to determine potentially dangerous levels. Heat stress from high thermal radiation loading can cause illnesses ranging from behavioral disorders to heat stroke and even death [25]. Heat stress is the mildest form and is temporary in nature. It could include heat rashes sometimes referred to as “prickly heat” and dehydration. High humidity ambient conditions can exacerbate the problem for personnel in the vicinity of high thermal radiation conditions. Skin can be damaged from excessive heat. Even lower thermal
Figure 2.5 Stainless steel fence shielding flow control equipment from thermal radiation during flare testing.
© 2011 by Taylor and Francis Group, LLC
46
Figure 2.6 Radiometer for measuring thermal radiation. (From Baukal, C.E., Industrial Combustion Pollution and Control, New York: Marcel Dekker, 2004.)
Figure 2.7 Viewport with shutter. (From Baukal, C.E., Industrial Combustion Pollution and Control, New York: Marcel Dekker, 2004.)
radiation loadings can cause worker fatigue and reduce performance. Older, overweight, or out-of-shape workers are particularly at risk from high thermal radiation loads. There are two general abatement strategies used to mitigate and control external thermal radiation loading: reducing the source of the radiation or shielding personnel and equipment from the source. Source reduction involves reducing the external radiation source. One way is to insulate the radiation source, for example, a furnace wall, to reduce the external surface temperature and hence the radiation. Another way is to reduce the energy source heating the radiating surface. The easiest way to do that is to reduce the firing rate; however, this will also reduce the production rate, which is normally not
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
Figure 2.8 Heat resistant suit. (From Baukal, C.E., Industrial Combustion Pollution and Control, New York: Marcel Dekker, 2004.)
desirable. Another way to reduce the radiation source is to cool it, for example, by water cooling the external furnace walls. Viewports should have some type of shutter to minimize the heat escaping through them when they are not in use (see Figure 2.7). Furnace leaks should be repaired, not only to reduce external thermal radiation, but also to improve the thermal efficiency. The second abatement technique is shielding and cooling, which involves shielding personnel and equipment from the external radiation source. This can be done with some type of physical barrier such as a wall (e.g., Figure 2.5). It can be done with screens around the radiation source to prevent personnel from getting too close. Individual pieces of equipment can be shielded with insulation, a reflective surface, or some type of solid material. Personnel can be shielded by wearing appropriate protective clothing designed for high heat environments (see Figure 2.8). 2.3.3 Noise Noise is often defined as unwanted sound. There are many possible sources of noise in industrial combustion testing (see Chapter 8). Some of these include combustion roar, jet noise for flow through orifices, flow noise for fluids flowing through piping, and equipment noise such as from the combustion air fan. Acoustic resonance can exacerbate the problem by magnifying the noise. There are several strategies that can be used to reduce combustion noise (see Bussman and Jayakaran [26] and Baukal [27] for more details). One strategy is either to move the source of the undesirable sound away from the people or to move the people away from the sound. However, this may not be practical for many industrial applications. Another strategy is to put some type of
47
Testing Safety
sound barrier between the noise and the people. The barrier can be either reflective or absorptive to minimize the noise. The noise source might be surrounded by an enclosure or the operators may be located inside a soundproofed enclosure. In some cases, it may be possible to use a silencer, which would act as a barrier, to reduce the noise. For example, the exhaust from a car is reduced by the muffler, which acts like a silencer. The barrier could also be in the form of earplugs, ear phones, or some other sound-reducing safety device worn by people in the vicinity of the noise. Another technique is to reduce the exposure time to the noise since noise has a cumulative effect on human hearing. In some cases, it may be possible to replace noisy equipment with new equipment that has been specifically designed to produce less noise, or to retrofit existing equipment to produce less noise. For example, old combustion air blowers and fans could be replaced by new, quieter blowers and fans. Another way to reduce noise is to increase the pipe size and reduce the number of bends in the pipe to reduce the fluid noise caused by the fluids flowing through the pipe. Resonance and instabilities can usually be designed out of a system if they are a problem. Noisy burners can be replaced by quieter burners. The burner noise is a function of the burner design, fuel, firing rate, air–fuel stoichiometry, combustion intensity, and aerodynamics of the combustion chamber. These strategies are discussed in some detail next. The primary source of noise in most industrial combustion systems is from the burner(s). The design of the burner nozzle and the burner tile or quarl are important factors in noise generation in combustion processes. One method to reduce noise from a burner is to use larger exit ports that produce lower gas velocities. However, there are limits to how low gas velocities can be, depending on the fuels used and the burner type. For example, the exit gas velocities in a premix burner must be greater than the flame speed of the oxidizer/fuel mixture or else flashback may result. Fuels containing high concentrations of hydrogen will necessarily require higher exit velocities because of the high flame speed of hydrogen. Another method for reducing the sound from a combustion process is to add a pipe or tube, often referred to as a quarter wave tube, to the resonance system to cancel out the harmonics causing the noise. In this technique, a specially designed tube is attached to the chamber where sound of a predominant frequency is causing resonance. The quarter wave tube is designed to cancel out this resonance and therefore reduce the noise levels. Figure 2.9 shows a quarter wave tube installed on the side of an exhaust stack for a gas-fired, propylene vaporizer that previously exhibited a low frequency rumble during operation, producing excessive noise levels. The quarter wave tube was built with some adjustability to determine the best geometry to maximize noise reduction. The
© 2011 by Taylor and Francis Group, LLC
total noise level for the vaporizer without the tube was 95.3 dBA. The total noise level with the tube dropped to 83.8 dBA for a noise reduction of more than 10 dBA. This brought the noise levels within acceptable limits. Note that adding a quarter wave tube is not always a practical option because, depending on the combustion chamber geometry and the frequency of the harmonic, the tube diameter and length may be excessive. Other techniques to mitigate noise caused by combustion instability include modifying the [28]: • • • •
Furnace stack height Internal volume of the furnace Acoustical properties of the furnace lining Pressure drop through the burner by varying the damper positions • Location of the pilot • Flame stabilization techniques Figure 2.10 shows some of the common silencers used to reduce the source of noise from industrial processes [29]. Figure 2.11 shows a natural draft burner without an air inlet muffler. These burners are often quiet enough that no muffler is needed. Figure 2.12 shows a common type of muffler used on a natural draft burner used in the floor of refinery heaters where the air inlet has a baffle lined with sound-deadening insulation. Figure 2.13 shows a comparable standard muffler for a radiant wall burner
Exhaust stack
1⁄
Figure 2.9 Quarter wave tube on a propylene vaporizer.
4
Wave tube
48
Absorptive silencer. This silencer is the most common type and takes the form of a duct lined on the interior with a sound-absorptive material.
Industrial Combustion Testing
Absorptive silencer
Reactive expansion chamber. This type reflects sound energy back toward the source to cancel some of the oncoming sound energy. Reactive expansion chamber Reactive resonator. This type functions in approximately the same way as the reactive expansion chamber type. Plenum chamber. This device allows the sound to enter a small opening in the chamber: that sound which has not been absorbed by the chamber’s acoustical lining leaves by a second small opening, generally at the opposite end of the chamber. Lined bend. Sound energy flowing down a passage is forced to turn a corner, the walls of which are lined with acoustical material. The sound energy is thus forced to impinge directly on a sound-absorbing surface as it reflects its way around the corner; each successive impingement takes sound energy from the traveling wave. Diffuser. This device does not actually reduce noise. In effect, it prevents the generation of noise by disrupting high-velocity gas streams.
Reactive resonator
Plenum chamber
Figure 2.11 Natural draft burner with no air inlet muffler. (From Baukal, C.E., Industrial Combustion Pollution and Control, New York: Marcel Dekker, 2004.)
Lined bend
Diffuser
Figure 2.10 Silencers. (From Liu, D. H. F., “6.7 Noise Control in the Transmission Path,” in Environmental Engineers’ Handbook, edited by D. H. F. Liu and B. G. Lipták, 2nd ed., Boca Raton, FL: Lewis Publishers, 1997.)
used in the side of ethylene cracking furnaces. Figure 2.14 shows a nonstandard muffler used on a floor-fired natural draft burner where the muffler is larger than the burner because of the very low noise requirements for the particular application. Figure 2.15 shows a nonstandard muffler for low noise requirements on wall-fired burners. Sound transmission mitigation is a technique that involves mitigating the transmission of the sound from the source to the receiver. This can be done in a variety of ways. One rather simple but not always practical method is simply to increase the distance between the source and the receiver, which reduces the sound levels at the receiver. Another strategy is to put some type of barrier between the source and the receiver. For example, a concrete wall could be built around the source. People commonly plant trees and shrubs on
© 2011 by Taylor and Francis Group, LLC
Figure 2.12 Natural draft burner with an air inlet damper. (From Baukal, C.E., Industrial Combustion Pollution and Control, New York: Marcel Dekker, 2004.)
their property to mitigate some of the sound from their neighbors and from road traffic noise. Figure 2.16 shows enclosed flares with walls around the enclosures to help mitigate the sound produced by the flares. Some type of sound-absorptive material could be placed between the source and the receiver. Different materials have different sound-absorbing characteristics, depending on both the composition and configuration [30]. Another way could be to use a medium that is less transmissive for sound. For example, water is less transmissive than air. Noise generated by the burners in a combustion system may be greatly mitigated by the combustion chamber,
49
Testing Safety
Insulation
Muffler
Figure 2.13 Typical muffler for a radiant wall-fired natural draft burner. (From Baukal, C.E., Industrial Combustion Pollution and Control, New York: Marcel Dekker, 2004.)
Figure 2.15 Large mufflers on two radiant wall burners. (From Baukal, C.E., Industrial Combustion Pollution and Control, New York: Marcel Dekker, 2004.) Enclosed flare helps reduce noise
Wall at bottom of enclosure helps reduce noise
Figure 2.14 Large mufflers on two natural draft burners. (From Baukal, C.E., Industrial Combustion Pollution and Control, New York: Marcel Dekker, 2004.)
which is usually a furnace of some type. The refractory linings in most furnaces generally significantly reduce the noise emitted from the burners. Noise is not commonly considered in many industrial heating applications for a variety of reasons. This is evidenced by the general lack of information available on the subject. It is difficult to predict the noise levels before installing the equipment due to the wide variety of factors that influence noise. Often, there are many other pieces of machinery that are much noisier than the combustion system so that the workers are already required to wear hearing protection. In the future, noise reduction may become more important. In combustion testing, noise
© 2011 by Taylor and Francis Group, LLC
Figure 2.16 Two enclosed flares. (From Baukal, C.E., Industrial Combustion Pollution and Control, New York: Marcel Dekker, 2004.)
may be less of a concern because of the temporary nature of many tests. Mufflers are commonly used on burners to reduce noise levels (see Figure 2.17). Figure 2.18 shows how effective a muffler is at mitigating the noise produced by a burner. These mufflers typically go on the combustion air inlet to the burner and usually have some type of noise-reducing insulation on one or more sides. Reed gives an example of a natural draft burner producing 107 dB of noise before any mitigation [31]. The addition of a primary muffler reduced the noise to 93 dB and the addition of a secondary muffler further reduced the noise to 84 dB. Figure 2.19 shows a custom-made cylindrical muffler on the air outlet of a velocity thermocouple, often called a suction pyrometer (see Chapter 5), used to measure higher temperatures. This type of temperature
50
Industrial Combustion Testing
Figure 2.17 Typical muffler used on a natural draft burner. (From Baukal, C.E., Industrial Combustion Pollution and Control, New York: Marcel Dekker, 2004.)
Unmuffled A-weighted burner noise
80 75 70 65 60 55 50
8000
4000
16000
Frequency, Hz
2000
1000
500
250
63
40
125
Muffled A-weighted burner noise
45 31.5
Sound pressure level, dBA
85
Figure 2.18 Sound pressure versus frequency with and without a muffler. (From Baukal, C.E., ed., The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001.)
Figure 2.19 Cylindrical muffler on a suction pyrometer. (From Baukal, C.E., Industrial Combustion Pollution and Control, New York: Marcel Dekker, 2004.)
© 2011 by Taylor and Francis Group, LLC
measuring device relies on high air flow rates across a venturi to induce furnace gases to flow through the pyrometer and across a shielded thermocouple. This minimizes the effects of thermal radiation on the thermocouple and produces a more accurate measure of the true temperature. Measurements using bare wire thermocouples can be as much as 200°F or more too low. Compressed air is often used as the motive gas to induce furnace gas inspiration. The high exit gas flow rates into the atmosphere can produce relatively high noise levels that are a concern for workers in the vicinity. The cylindrical muffler shown in Figure 2.19 reduced noise levels by more than 10 dBA down to an acceptable level. There are two levels of protection commonly used by industrial workers to reduce noise levels [32]: plugs (see Figure 2.20) and muffs (see Figure 2.21). One or both types may be used, depending on the noise levels. A third type of protection device is a helmet, commonly used by motorcycle drivers, which provides relatively little hearing protection and is rarely used in industry for hearing protection. Therefore, this is not considered further here. Plugs can reduce noise levels by 5–45 dB, depending on the plug type and sound frequency. They come in a variety of forms including disposable and reusable. Disposable plugs are typically made of some type of moldable material (e.g., foam) that can be inserted into a variety of ear sizes and shapes. These are very inexpensive and are typically bought in large quantities. They are especially convenient for visitors who do not have their own ear plugs. Reusable ear plugs can be cleaned and used multiple times. Custom molded ear plugs are available that are made to fit exactly in a specific person’s ears and are designed to be reusable. Ear muffs are designed to cover the entire ear and typically reduce levels by 5–50 dB, depending on the muff type and sound frequency. These can be used in lieu of or in addition to ear plugs. When both plugs and muffs are worn, noise protection is greater than either individually but is not additive. Muffs may be more comfortable to some compared to ear plugs. However, they may also interfere with other PPE such as hard hats. Special
51
Testing Safety
Figure 2.20 Typical ear plugs. (From Baukal, C.E., Industrial Combustion Pollution and Control, New York: Marcel Dekker, 2004.)
Figure 2.22 Ear muffs designed to be used with hard hats. (From Baukal, C.E., Industrial Combustion Pollution and Control, New York: Marcel Dekker, 2004.)
2.3.4 High Pressure
Figure 2.21 Typical ear muffs. (From Baukal, C.E., Industrial Combustion Pollution and Control, New York: Marcel Dekker, 2004.)
muffs are made to attach to certain types of hard hats where the muffs can be folded down or up as needed (see Figure 2.22). Convenience and comfort are important factors when choosing appropriate noise protection for a given environment. If too much effort is required to use or maintain the hearing protection devices, or if they are uncomfortable, then they are less likely to be used. Proper training and education is essential to maximize the effectiveness of any hearing conservation program [33]. While not possible in some cases, a simple way to protect workers is either to increase their distance from the sound source or to put them in a soundproofed enclosure such as a control room or building. However, it is almost impossible to keep all workers away from high noise sources all the time, so hearing protection will probably be necessary.
© 2011 by Taylor and Francis Group, LLC
There are multiple ways that equipment and personnel might be exposed to high pressures in combustion testing. High pressure gases are sometimes used, for example, as span gases to calibrate gas analyzers. Specialty fuels may sometimes be supplied from high pressure cylinders. Proper handling techniques should be used when moving these cylinders around. The cylinders must also be properly stored. For example, oxidizers and flammables are not normally stored together. High pressure cylinders should be stored away from high temperature sources. If possible, cylinders should be stored outside, so that if a leak should occur it would not be confined in a building (see Figure 2.23). Some combustion experiments may have the potential to produce high pressures. This can happen if liquids or gases are being heated, such as water in a boiler or hydrocarbon fluids in a process heater. The high pressures may be caused by high temperatures, by fluids being blocked in, by some type of blockage in the system that causes a restriction, or possibly by fluid vaporization from a liquid to a gas that causes a large and rapid increase in volume. In most commercial systems, adequate provisions are designed into the system for pressure relief in case the fluid pressures exceed the design limits. This may not be the case for experimental systems, so the experimenter must ensure that adequate precautions are taken so the pressures do not exceed the design limits. This may include the addition of a properly designed relief system and some type of warning system to signal the operators before potentially excessive pressures are
52
Industrial Combustion Testing
Figure 2.23 Pressurized gas cylinders located outside a building where gases are used in a lab inside the building.
reached. Adequate sensors should be provided so the operators can monitor the pressures so corrective action can be taken if necessary. One type of hazardous overpressure condition is a rapid deflagration, which is considered separately next. 2.3.5 Explosion Danger of explosion may come from many sources, but explosions most often occur when the equipment involved is in a state of change such as startup, shutdown, or maintenance. Since furnaces are made primarily of metals, maintenance often involves welding. The welding process provides an ignition source for any combustible gases or liquids that might remain in the work area. Typically, tests for the presence of flammable gases or liquids are conducted before any maintenance is allowed to commence. However, thorough testing to assure the absence of flammable materials can be complex and difficult. Two principles should be followed when testing for combustibles in a planned, enclosed maintenance area. First, testing should be done not only in the immediate work area but also in any connecting plumbing or equipment. Second, testing should be done immediately before the work begins and periodically during the time the ignition source is in use. Stacks are designed to vent exhaust gases from i ndustrial processes. If the gases are combustible, flare stacks may be used to burn the exhaust gases before they are vented into the atmosphere. In the case of combustible gases, operational procedures should be used that prevent the infiltration of oxygen into the stack where it could mix with the exhaust gases and cause an explosion. Stacks should be built with welded seams rather than bolted joints. Bolted joints, especially joints
© 2011 by Taylor and Francis Group, LLC
having surfaces that are not machined, can leak air into the stack in areas of negative pressure. Operational procedures should be used that assure a continuous flow of gas is moving up the stack. A continuous flow of gas will sweep away small leaks of air and prevent air from moving down into the stack from the top. The most common source of furnace explosions is the use of an improper lighting procedure. Even procedures that have been used for years without incident can cause explosions if conditions change. For example, if the fuel source is not completely isolated from the furnace before the ignition source is ignited, a leaky valve can allow enough gas into the furnace to cause an explosion. A safe furnace lighting procedure for a furnace containing piloted burners should include the following generalized steps.
1. Confirm that the furnace fuel lines are completely isolated from the furnace (either by disconnection, blind flange, or a double block and bleed valve assembly). 2. Confirm that all auxiliary furnace equipment is functioning properly, including all instrumentation and measurement devices; open both the burner air inlet register/damper and the furnace stack damper to the fully open position. 3. Purge the furnace of any combustible or flammable substances; follow the NFPA recommendation of purging the furnace with four furnace volumes of fresh air or inert gas (N2 or CO2) [22]. 4. Test the atmosphere inside the furnace, just before ignition will be attempted, to assure that there are no combustibles present. 5. Connect the pilot fuel line to the burner; activate the permanent pilot igniter or insert a portable pilot igniter or premixed ignition torch.
53
Testing Safety
6. Slowly open the pilot fuel control valve to the manufacturer-specified pilot fuel pressure; visually confirm stable ignition of the pilot flame. 7. Reestablish the main burner fuel supply (either by connecting the main fuel line, removing a blind flange, or reversing the double block and bleed valve assembly) to provide a fuel source to the furnace. 8. Slowly open the main burner fuel control valve to supply the burner with the manufacturerspecified ignition fuel pressure; visually confirm stable ignition of the main burner flame.
If the burner ignition attempt is unsuccessful after a relatively short time, or if the procedure is aborted prior to successful burner ignition, the burner fuel supply should be immediately disconnected. Subsequent attempts to successfully light the burner must begin at step 1. Another source of explosions in furnaces is improper air management. If the furnace is starved for air, a pulsating huffing sound may result. The flame will be unsteady, changing from long to short or wide to narrow. The variations are the result of the available air being completely consumed without burning all of the fuel. The air-starved flame will then be reduced in volume until more air is available. It will then increase in size until the available air is again consumed. This alternating cycle causes the huffing sound. The correct action, when a furnace is huffing, is to reduce the flow of fuel until there is enough air for full combustion. If the operator incorrectly increases the air without first reducing the fuel flow, the increased air supply may mix with the large volume of unburned fuel already in the furnace causing an explosion. Obviously, the best plan is to prevent an explosion from occurring, but appropriate plans should be made in the event one should occur. For example, it is recommended that all personnel be located away from the furnace during light-off in case there is an incident. It may also be advisable to have some type of relief system built into the furnace, for example pressure relief doors (sometimes referred to as explosion doors) that can open in the case of overpressurization.
they recombine into more stable species such as CO2 and H2O. This chemistry must be allowed to proceed to completion because any interruptions could extinguish the flame or cause the flame to become unstable. For example, halon fire extinguishers disturb the chemistry in a flame and cause it to go out. The velocity of the oxidizer and fuel are critical to flame stability. If these velocities are too slow or too fast, the flame can become unstable. If the fuel and oxidizer mixing is inadequate, this can produce an unstable flame. The more common causes of flame instability are briefly discussed here. If the exit velocity from a premixed burner is below the flame speed for the oxidizer-fuel mixture, flashback (sometimes referred to as blowback) can occur. This is where the flame travels back inside the burner. The flame can only travel as far as the oxidizer-fuel mixture is flammable. The burner design also dictates how far the flame can travel as shapes inside the burner can act as flame holders or flame arrestors. Figure 2.24 shows a radiant wall burner that has been damaged from a flashback. Some design features that prevent flashback include: • making the outlet velocity profile as uniform as possible • maximizing the oxidizer-fuel exit velocity from the burner • making the oxidizer-fuel mixture fuel lean, which reduces the flame speed • making the oxidizer-fuel mixture inflammable (either above the higher flammability limit or below the lower flammability limit) • promoting laminar flow out of the burner because the laminar flame speed is much slower than the turbulent flame speed
2.3.6 Flame Instability There are multiple potential problems that could cause flame instability. The fire triangle (see Figure 1.5) shows that fuel, oxygen, and an ignition source are required for a fire. However, these are not enough to guarantee a stable flame. Combustion chemistry is very complicated. The combustion of methane with air includes over 350 chemical reactions and dozens of species, most of which only survive for fractions of a second in the flame before
© 2011 by Taylor and Francis Group, LLC
Figure 2.24 Burner damaged from a flashback.
54
Industrial Combustion Testing
Figure 2.26 Coen iScan® flame detector. Figure 2.25 Premix burner lifting off.
Flashback is normally easy to visually and aurally detect. The general response to flashback is to either shut off the fuel or increase the firing rate so the exit velocity is above the mixture flame speed. Experimenters should pay particularly close attention to any conditions that might result in flashback. If flashback conditions are likely to occur, appropriate precautions should be taken. These might include using stronger and higher temperature materials in the burner and shielding the equipment in case the burner fails. If the oxidizer-fuel mixture exiting the burner is much higher than the flame speed, liftoff (sometimes referred to as blowoff) may occur. Figure 2.25 shows an example of a premix radiant wall burner lifting off. In this type of burner design, the flame exits the burner at a 90° angle and impinges on the refractory. When it is operating correctly, the refractory is uniformly heated and radiates to the process tubes in the furnace. In Figure 2.25, the burner is being overfired and the flame is partially lifting off, causing the flower-shaped heating pattern. Liftoff typically occurs when the oxidizer-fuel mixture velocity exiting the burner is much higher than the flame speed and there is inadequate flame stabilization in the burner to anchor the flame. There are some flames that are designed to be lifted, but most industrial flames are not lifted because of the possibility of the flame blowing out and reigniting causing an explosion. One indicator of this type of instability is a bouncing or pulsing flame. This phenomenon is commonly referred to as “huffing.” This type of low frequency instability can cause damage to the furnace, supporting structure, and associated equipment due to vibration. The normal procedure to handle huffing is to slowly reduce the fuel flow until the huffing stops and then to make whatever adjustments (e.g., repairing or cleaning the burner) are necessary to
© 2011 by Taylor and Francis Group, LLC
prevent this from occurring again. If the cause is due to overfiring the burner, then the burner design may need to be modified to reduce the outlet velocities and prevent liftoff. If the flame were to completely lift off and go out, the fuel should immediately be shut off to prevent an explosion or limit the damage that could occur from an explosion. This is typically done with some type of optical flame detector (see Figure 2.26), such as an ultraviolet flame scanner. Some burner design features that minimize the chance for liftoff include: • reducing the oxidizer-fuel outlet velocities (e.g., reducing the firing rate) • designing flameholders into the burner to anchor the flame and reduce the outlet velocities • improving the oxidizer-fuel mixing (e.g., changing the fuel injection angles) • increasing the oxidizer-fuel turbulence that increases the flame speed 2.3.7 Environmental There are many potential environmental problems from combustion processes that can be safety issues [27]. Incomplete combustion of carbon-containing fuels can produce carbon monoxide (CO) and smoke. Carbon monoxide is usually a regulated pollutant that can asphyxiate people in high enough concentrations. Smoke is also a commonly regulated pollutant that can damage the lungs and block or impair vision. These pollutants are typically more prevalent with liquid and solid fuels. Both can generally be easily controlled with sufficient oxygen to complete combustion, proper mixing of the oxygen with the fuel, and adequate temperatures for the reactions to go to completion. Incomplete combustion of hydrocarbon fuels can produce a wide range of products of incomplete combustion (PICs), including in the extreme allowing some fuel to go through the combustor
55
Testing Safety
unchanged. These PICs are also generally easily controlled with sufficient oxygen, temperature, and mixing. Fuels that contain potentially harmful chemicals may produce effluents that contain hazardous materials. For example, fuels containing benzene could produce benzene in the exhaust under certain conditions. Since benzene is a hydrocarbon, it is generally easy to completely destroy if there is adequate oxygen, temperature, and mixing. Fuels containing chlorine or fluorine can produce dioxins and furans that are toxic. These are not as easy to eliminate from the exhaust, so some type of posttreatment system is often required to minimize these pollutants to acceptable levels. Appropriate analyzers should be used to detect the emissions from combustion processes. There are two general types of analysis systems (see Chapter 7): extractive (see Figure 2.27) and in situ (see Figure 2.28). The extractive system pulls a small sample from the exhaust and sends it first to a conditioning system that cools, cleans, and dries the sample, before sending the conditioned sample to the analyzers. These systems have some lag time because they are remotely located from the stack, usually in some type of conditioned building such as the control room. In situ analyzers measure the gas composition at the sample location, where the sample is often hot and contains water. These systems provide very rapid response as there is very little delay in getting the measurement.
Figure 2.27 Typical continuous emission measurement system. (From Baukal, C.E., Industrial Combustion Pollution and Control, New York: Marcel Dekker, 2004.)
2.4 Accident Prevention The best safety strategy is to prevent accidents from occurring in the first place, thus preventing equipment damage and personnel injury. Quintana, Camet, and Deliwala (2001) describe the application of a predictive risk analysis model to a large-scale combustion test facility [34]. The model uses continuous data sampling and analysis of an experiment to predict potential hazards that could lead to an incident, a system malfunction or unacceptable risk conditions. Operators are warned of a potential problem so they can take appropriate corrective action. If the proper corrective action is not implemented quickly enough, then the system can be designed to take appropriate action. This type of automated safety analysis system is particularly beneficial in combustion testing because of the fast and dynamic environment that can be present. Some of the more common strategies for preventing accidents are briefly discussed in this section. 2.4.1 Ignition Control Ignition is the process through which combustion is initiated, and occurs when a flammable mixture of fuel
© 2011 by Taylor and Francis Group, LLC
Figure 2.28 Typical in situ analyzer.
and oxidizer comes in contact with a suitable ignition source. Direct contact with a spark or flame is a very common energy source often used for the intentional ignition of industrial combustion equipment. For many types of burners, the ignition source may be in the form of a small premixed pilot burner, a portable electrostatic ignitor, or a portable premixed gas torch. Static electricity is a common potential ignition source in chemical processing plants. An electrostatic charge is
56
formed whenever two dissimilar surfaces move relative to each other. A relevant example is liquid flowing through a pipeline, moving past the walls of the pipe. In this example, one charge is formed on the pipe surface, while another equal but opposite charge is formed on the surface of the moving liquid. When the voltage becomes strong enough, the static electricity will discharge in the form of an electrical spark. The spark can ignite any combustible and flammable materials present. The NFPA (1993) [35] presents detailed explanations of design fundamentals for the prevention of fires and explosions due to electrostatic discharge. Some methods used to prevent unintended ignition include proper grounding of electrical equipment, using nonsparking tools, preventing smoking, and keeping electrical devices that are not intrinsically safe (e.g., cell phones) out of areas in the plant where flammables could be present.
Figure 2.29 Rubber mat over tripping hazard.
Figure 2.30 Safety tape around test apparatus.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
2.4.2 General There are many general precautions that can be taken to prevent an accident during combustion testing. A sample of such precautions is given here. Others may be implemented as appropriate. Some test rigs are temporary for one-time experiments, which may require specific precautions to prevent equipment damage and personnel injury. Figure 2.29 shows a photo of a rubber mat over hoses being used for a temporary combustion test. For test rigs that will be permanent, cables and hoses should be located so they will not be tripping hazards. For temporary rigs that are commonly used in testing, appropriate precautions should be taken to prevent injuries that could be caused by these types of setups. Figure 2.30 shows safety tape around a portion of a combustion test setup to prevent
57
Testing Safety
unauthorized personnel from accidentally entering an area where they could possibly be injured. Monitoring and control equipment can be used to rapidly stop a test if a problem is detected. The conventional method used in commercial combustion systems is some type of burner management system that includes a flame detector (e.g., Figure 2.26). A furnace camera (Figure 2.31; see Chapter 16) may be useful for remote monitoring of the burner, flame, and furnace. These cameras are specially designed for use in furnaces and may include air cooling, water cooling, and air purging of the lens. The camera monitor can be located in the control room so the testing can be rapidly and remotely stopped. Figure 2.32 shows some emergency stop push buttons in a control room that can be used to rapidly stop a test if necessary. It may also be advisable to have emergency stop push buttons at or near the test rig. Figure 2.33 shows one of these push buttons located next to a sight port on a test furnace so someone watching the testing through the sight port can rapidly stop the test if a problem is detected. Another example of a precaution that should be taken is to prevent liquid fuel in storage tanks from leaking into the ground and possibly contacting an ignition source that could lead to a fire or explosion. Figure 2.34 shows an example of a metal containment system at the base of elevated diesel storage tanks. The containment needs to be capable of holding the entire volume
of the tanks plus include some additional capacity for rain and any other materials that could possibly enter the containment system. Some type of monitoring system to detect leaks may be advisable depending on the circumstances, particularly in enclosed locations. As an example, Figure 2.35 shows gas analyzers mounted in a combustion laboratory wall inside a building.
Figure 2.33 Emergency stop push button next to a sight port on a test furnace.
Figure 2.31 Furnace camera.
Figure 2.32 Emergency stop push buttons in a control room.
© 2011 by Taylor and Francis Group, LLC
Figure 2.34 Liquid fuel containment for diesel storage tanks.
58
Industrial Combustion Testing
Figure 2.35 NO, CO, O2, and combustibles analyzers in a combustion laboratory.
Figure 2.37 Fire alarm system in a combustion test facility.
Figure 2.36 Example of safety signs on the outside of a building.
Personnel must be properly trained to be able to identify potential hazards, to operate the equipment, and to respond in the event of an accident. Appropriate safety signs (see Figure 2.36) should be used to alert personnel of potential hazards in a particular area.
2.5 Accident Mitigation While every effort should be made to prevent an accident, appropriate plans should also be in place in case there is an accident to minimize the consequences if one should occur. If a fire should occur, a fire alarm system (see Figure 2.37) should be in place to alert test personnel and to notify the local fire department. If the fire is not too large and personnel have been trained to extinguish fires, appropriate fire extinguishers (see Figure 2.38) should be located in the vicinity of combustion testing.
© 2011 by Taylor and Francis Group, LLC
Figure 2.38 Large portable fire extinguisher.
59
Testing Safety
Figure 2.40 Examples of safety and medical kits.
Figure 2.39 Emergency pull ring to shut down the entire facility.
There may be multiple levels of shutdown depending on the severity of an incident. In a multiburner test, the lowest level of shutdown might be to shut down a single burner, for example, in the event that burner flames out. Another level might be to shut off all the burners to a single test heater. Depending on how the test facility is designed, it might be desirable to shut off an entire section of the facility. For example, if a building in the facility contains multiple test furnaces, it might be desirable to rapidly shut off all the furnaces in the building in the event of a fuel leak in the building. The highest level of shutdown might be to shut down the entire test facility in the event of a very serious incident. Figure 2.39 shows an emergency pull ring designed to shut down an entire combustion test facility in case of a serious incident. These shutdown systems are designed to minimize the impact of an incident that could become much worse if it adversely affects nearby equipment and personnel. Figure 2.40 shows some examples of safety and medical kits to treat personnel who might be injured in an incident. A well-equipped medical kit should be readily available and personnel properly trained and certified
© 2011 by Taylor and Francis Group, LLC
Figure 2.41 Emergency shower.
in first aid to treat minor injuries and provide temporary treatment until emergency medical personnel arrive in the event of a serious injury. It may also be advisable to have an emergency shower (see Figure 2.41) available in case personnel are exposed to chemicals or hot fluids. Appropriate PPE should be used during industrial combustion testing. This may include steel-toed shoes, safety
60
Industrial Combustion Testing
Figure 2.42 Wind sock.
glasses, a hard hat, gloves, and flame-retardant clothes. It may also include a face shield, special glasses for viewing bright objects such as flames in furnaces, and possibly even a special heat resistant suit (see Figure 2.8). The PPE can prevent or minimize personnel injury that could result during testing. Figure 2.42 shows a photo of a wind sock located near fuel gas piping. This sock can be used to determine what direction leaking chemicals might be taken by the wind in the event of a leak. This is particularly important for toxics that could harm personnel and flammable materials that could start or feed a fire. In general, personnel would normally want to go upwind of leaking equipment unless that path would be more hazardous (e.g., they would have to go through a fire to get upwind).
References
2.6 Recommendations Industrial combustion processes have the potential to be very dangerous because of the large quantities of fuels being combusted. Industrial combustion testing can be even more dangerous due to the potential of using unproven equipment designs and configurations. Rigorous safety reviews are strongly recommended to identify potential hazards so appropriate precautions can be taken to prevent equipment damage and personnel injury. There are many potential hazards that could be present during industrial combustion testing, some of which have been discussed in this chapter. Appropriate measures should be taken to prevent accidents, but plans should also be made to minimize the consequences of an accident if one should occur. Safety should be the primary concern for any industrial combustion experiments.
© 2011 by Taylor and Francis Group, LLC
1. Faisal. I., and Abbasi, S. A. “Major accidents in process industries and an analysis of causes and consequences.” Journal of Loss Prevention in the Process Industries 12, no. 5 (1999): 361–78. 2. Center for Chemical Process Safety. Explosions in the Process Industries. New York: American Institute of Chemical Engineers, 1994. 3. Sanders, R. E. Chemical Process Safety: Learning from Case Histories. Boston: Butterworth-Heinemann, 1999. 4. Kletz, T. Learning from Accidents. 2nd ed. Oxford: Butterworth-Heinemann, 1994. 5. Kletz, T. An Engineer’s View of Human Error. 2nd ed. Rugby, UK: Institution of Chemical Engineers, 1991. 6. Center for Chemical Process Safety. Guidelines for Preventing Human Error in Process Safety. New York: American Institute of Chemical Engineers, 1994. 7. Eckhoff, R. K. Explosion Hazards in the Process Industries. Houston, TX: Gulf Publishing, 2005. 8. Crocker, W. P., and Napier, D. H. “Thermal Radiation Hazards of Liquid Pool Fires and Tank Fires,” IChemE Symp. Series No. 97.159–84. Oxford, UK: Institute of Chemical Engineers, Pergamon Press, 1986. 9. Center for Chemical Process Safety. Thermal Radiation 1: Sources and Transmission. New York: American Institute of Chemical Engineers, 1989. 10. Center for Chemical Process Safety. Thermal Radiation 2: The Physiological and Pathological Effects. New York: American Institute of Chemical Engineers, 1996. 11. Kletz, T. What Went Wrong: Case Histories of Process Plant Disasters. 4th ed. Houston, TX: Gulf Publishing Company, 1998. 12. Lewis, B., and Von Elbe, G. Combustion, Flames and Explosions of Gases. 3rd ed. New York: Academic Press, 1987. 13. Cox, G. Fire Safety Science. UK: Routledge, 1990. 14. Jones, J. C., and Williams, A. Topics in Environmental and Safety Aspects of Combustion Technology. Scotland: Whittles Publishing, 1997.
Testing Safety
15. Marshall, V. C., and Ruhemann, S. Fundamentals of Process Safety. UK: Institute of Chemical Engineers, 2000. 16. Korver, W.O.E. Electrical Safety in Flammable Gas/Vapor Laden Atmospheres. Norwich, CT: William Andrew Publishing, 2001. 17. Thomson, N. G. Fire Hazards in Industry. Oxford, UK: Butterworth-Heinemann, 2002. 18. BP Safety Group. Safe Furnace and Boiler Firing. 4th illustrated ed. Rugby, UK: Institute of Chemical Engineers, 2005. 19. NFPA 30: Flammable and Combustible Liquids Code. Quincy, MA: National Fire Protection Association, 2008. 20. NFPA 54: National Fuel Gas Code. Quincy, MA: National Fire Protection Association, 2009. 21. NFPA 85: Boiler and Combustion Systems Hazard Code. Quincy, MA: National Fire Protection Association, 2007. 22. NFPA 86: Standard for Ovens and Furnaces. Quincy, MA: National Fire Protection Association, 2007. 23. Baukal, C. E. Industrial Combustion Heat Transfer. Boca Raton, FL: CRC Press, 2000. 24. Slavejkov, A. G., Baukal, C. E., Joshi, M. L., and Nabors, J. K. “Advanced Oxygen/Natural Gas Burner for Glass Melting.” Proceedings of 1992 International Gas Research Conference, Orlando, FL, November 16–19, 1992. Rockville, MD: Government Institutes, Inc., 1993. 25. Owens, P. D. “Health Hazards Associated with Pollution Control and Waste Minimization.” In Process Engineering for Pollution Control and Waste Minimization, edited by D. L. Wise and D. J. Trantolo. New York: Marcel Dekker, 1994.
© 2011 by Taylor and Francis Group, LLC
61
26. Sams, G., and Jordan, J., “How to design low-noise burners.” Hydrocarbon Processing 75, no.12 (1996): 101–108. 27. Baukal, C. E. Industrial Combustion Pollution and Control. New York: Marcel Dekker, 2004. 28. Bussman, W., and Jaykaran, J. D. “Noise.” In The John Zink Combustion Handbook, edited by C. E. Baukal. Boca Raton, FL: CRC Press, 2001. 29. Liu, D. H. F. “6.7 Noise Control in the Transmission Path.” In Environmental Engineers’ Handbook, edited by D. H. F. Liu and B. G. Lipták, 2nd ed. Boca Raton, FL: Lewis Publishers, 1997. 30. Moulder, R. “Sound-Absorptive Materials.” In Handbook of Acoustical Measurements and Noise Control, edited by C. M. Harris, 3rd ed. Woodbury, NY: Acoustical Society of America, 1998. 31. Reed, R. D. Furnace Operations. 3rd ed. Houston, TX: Gulf Publishing, 1981. 32. Nixon, C. W., and Berger, E. H. “Hearing Protection Devices.” In Handbook of Acoustical Measurements and Noise Control, edited by C. M. Harris, 3rd ed. Woodbury, NY: Acoustical Society of America, 1998. 33. Royster, L. H., and Royster, J. D. “Hearing Conservation Programs.” In Handbook of Acoustical Measurements and Noise Control, edited by C. M. Harris, 3rd ed. Woodbury, NY: Acoustical Society of America, 1998. 34. Quintana, R., Camet, M., and Deliwala, B. “Application of a Predictive Safety Model in a Combustion Testing Environment.” Safety Science 38 (2001): 183–209. 35. National Fire Protection Association. NFPA 77: Recommended Practice on Static Electricity. Quincy, MA: National Fire Protection Association, 1993.
3 Experimental Design Joseph Colannino Contents 3.1 Factor Coding................................................................................................................................................................. 63 3.2 Using Matrix Equations................................................................................................................................................ 64 3.3 Factorial Designs............................................................................................................................................................ 65 3.4 About Experimental Error............................................................................................................................................ 66 3.5 Sums of Squares............................................................................................................................................................. 67 3.6 Degrees of Freedom....................................................................................................................................................... 67 3.7 The ANOVA.................................................................................................................................................................... 67 3.8 Error................................................................................................................................................................................. 68 3.9 Using Center Points and Replicates............................................................................................................................ 69 3.10 Fractional Factorials...................................................................................................................................................... 71 3.11 Generating the Fractional Design................................................................................................................................ 73 3.12 Calculating the Confounding Pattern........................................................................................................................ 73 3.13 Central Composite Designs.......................................................................................................................................... 73 3.14 Blocking........................................................................................................................................................................... 75 The goal of experimental design is efficiency: collecting the maximum amount of information with the minimum of experimental cost. We do this by making use of symmetry in the experimental design. For example, consider Figure 3.1. The design has three major symmetry planes, denoted by the dotted lines—namely, horizontal (x1), vertical (x2), and diagonal (x1x2)—and admits the following equation: y = a0 + a1x1 + a2x2 + a12x1x2 + ε, where y is some response of interest a0 through a12 are constants to be determined x1 and x2 are factors of interest ε is the experimental error.
(3.1)
3.1 Factor Coding To simplify the solutions of Equation 3.1, we shall code the factor extrema to ±1. This is easily done by means of the following linear transforms:
+ ξ k ,min ξ ξ k − k ,max 2 xk = ξ k ,max − ξ k ,min 2
(3.2)
where xk is the kth coded factor, ξk is the factor in its natural dimension, ξk,max is the maximum value of the factor, and ξk,min is the minimum value of the factor. The quantity in parenthesis will be recognized as the average factor value while the denominator represents the half-range. We illustrate the use of the transform with an example. Suppose ξ1 is the oxygen fraction (from 1 to 3%) and ξ2 is the furnace temperature (from 1450 to 1600°F). Then the transforms become x1 = (ξ1 – 2%)/1%, x2 = (ξ1 – 1525°F)/75°F, and the reader may easily verify that the transform results in factors of having a dimensionless range of ±1. With such coding, we shall show that the solutions for the coefficients in Equation 3.1 are as follows: a0 =
y 00 + y11 + y12 + y 21 + y 22 ( y + y22 ) − ( y11 + y21 ) , , a1 = 12 4 5
a2 =
( y21 + y22 ) − ( y11 + y12 ) , a = ( y11 + y22 ) − ( y12 + y21 ) 12 4
4
(3.3)
where y00 through y22 are defined per Figure 3.1. That is, a0 is the mean value of all observations and the 63
© 2011 by Taylor and Francis Group, LLC
64
Industrial Combustion Testing
Or, making use of matrix algebra, we may write Equation 3.4 in the following way:
Vertical symmetry plane y22
y12 x2 direction y0,0
Horizontal symmetry plane
x1 direction
y 00 1 y 1 11 y12 = 1 y 21 1 y 1 22
0 −1
0 −1
−1 1 1
1 −1 1
0 −0.4 57 57.4 −1 0.1 40 7.5 + 0.1 = 50 −1 12.55 −1 0.1 60 2.5 0.1 80 1
y21
y11
Noting that in matrix multiplication, columns are multiplied by rows, the reader should satisfy himself that Equations 3.5 and 3.4 are equivalent. This is standard linear algebra.* Equation 3.5 may be further condensed by making use of matrix notation.
Diagonal symmetry plane (1 of 2 equivalent planes)
Figure 3.1 An experimental design. The design comprises five points and three main symmetry planes: horizontal, vertical, and diagonal.
remaining coefficients are contrasts along various axes of symmetry. For example, if
y 00 57 y 40 11 y12 = 50 y 60 21 y 80 22
a0 577.5 a 7.5 then 1 = a2 12.5 a 2.5 12
3.2 Using Matrix Equations We may recast Equation 3.1 as follows: + + + + +
7.5 ( 0 ) 7.5 ( −1) 7.5 ( −1) 7.5 ( +1) 7.5 ( +11)
+ + + + +
12.5 ( 0 ) 12.5 ( −1) 12.5 ( +1) 12.5 ( −1) 12.5 ( +1)
+ + + + +
2.5 ( 0 )( 0 ) 2.55 ( −1)( −1) 2.5 ( −1)( +1) 2.5 ( +1)( −1) 2.5 ( +1)( +1)
−0.4 57 0.1 40 + 0.1 = 50 0.1 60 0.1 80
© 2011 by Taylor and Francis Group, LLC
(3.4)
y = Xa + ε
(3.6)
In this chapter, we use the convention that bold lower case letters represent column vectors (i.e., n × 1 matrices), while bold upper case letters represent matrices (i.e., general m × n matrices). For row vectors (i.e., 1 × n matrices) we use the transpose operator, which switches rows and columns for any matrix or vector (e.g., x T). Equation 3.7 demonstrates the principle:
In fact, these are the least squares solutions as we shall demonstrate.
y 00 57.4 y 57.4 11 y12 = 57.4 y 57. 4 21 y 57.4 22
(3.5)
x0 ,1 x 0 ,2 X= x 0 ,n x T = ( x1
x1,1 x1,2 x1,n x2
xm ,1 x1 x xm , 2 , x = 2 , x xm , n n
(3.7)
xn )
where xj,k represents the matrix element in the jth c olumn and the kth row. For purposes of convenience, we index the columns beginning with 0 and the rows beginning with 1. Returning now to Equation 3.6, we premultiply by the transpose of X. This assures that the matrix to be solved will be square (equal number of rows and columns) and therefore a candidate† for a unique solution.
XT y = XTX a + XTε
(3.8)
As a principle of the least squares process, a is chosen such that XTε is driven to 0. This leaves XT y = XTX a, which is known as the normal equation. To find a with * For a review of matrix algebraic concepts, please see J. Colannino, Modeling of Combustion Systems, A Practical Approach, Chapter 1, Taylor & Francis/CRC Press, Boca Raton, FL, 2006, or any standard text on linear algebra. † All matrices having general solutions are square; however, not all square matrices have solutions. Notwithstanding, the experimental designs we present will always be soluble in the way shown.
65
Experimental Design
this property, we multiply by the inverse matrix (XTX)−1. This gives (XTX)−1XT y = (XTX)−1(XTX)a. Since by definition (XTX)−1(XTX) = I (the identity matrix) and Ia = a, we have our final result: (3.9)
a = (XTX)−1XT y
This is a general result for any experimental design whatsoever. However, owing to symmetry in statistically cognizant designs, all nondiagonal elements in XTX and (XTX)−1 are zero. This results in diagonal matrices that are not only easy to solve (e.g., Equation 3.3), but also results in coefficient solutions that are independent and maximally efficient. To see this, we begin by expanding XTX: x0 ,1 x 1,1 XTX = x m ,1
x0 , 2 x1,2 xm , 2
x0 , n x0 ,1 x1,n x0 ,2 xm ,n x0 ,n
∑x
2 0,k
k =1
∑x
∑x
x
2 1, k
∑x
n
n
x0 , k xm , k
k =1
∑x
x
1, k m , k
x
0,k m,k
∑
k =1
k =1
∑
x
0 , k 1, k
n
0 , k 1, k
∑
k =2
(3.10)
where k indexes the rows, m is the total number of columns (factors), and n is the total number of rows (observations). For the designs we consider, ∑ nk =1 x02, k = n, the number of observations (always), because xk,1 = 1 for every k (see Equation 3.5). In all cases, XTX is symmetrical as reflected about the principal diagonal; that is, XTX is a symmetrical matrix. Now for symmetrical experimental designs centered at zero (also known as balanced designs), all nondiagonal elements are zero (the reader should verify this using the matrix of Equation 3.5, for example). This results in the following matrix:
n XTX =
x0 ,1 x 1, 1 XT y = x m ,1
n
∑x
2 1, k
k =1
© 2011 by Taylor and Francis Group, LLC
n
∑ k =1
xm2 , k
(3.11)
x0 , n y 1 x1,n y 2 xm ,n y n
x0 , 2 x1,2 xm , 2
=
yk k =1 n x1, k y k k =1 n xm , k y k k =1 (3.12) n
∑x
0,k
∑ ∑
Since x0,k = 1 for all values of k, then ∑ nk =1 x0 , k y k = ∑ nk =1 y k . Substituting Equations 3.11 and 3.12 into Equation 3.8 and dropping the subscripts, we obtain:
n
k =1
n
∑x
k =1 n x1, k xm , k k =2 n xm2 , k k =1
n
n
xm ,1 xm , 2 = xm ,n
x1,1 x1,2 x1,n
Solving for XT y, we obtain
∑ y n ∑ x y = ∑ x
2 1
1
xm y
∑
∑
a 0 a1 . (3.13) xm2 am
Since the matrix is diagonal, the solutions are independent and easy to solve: a0 = ∑y/n and aj = ∑xjy/∑x2j. Indeed, this is the basis for the solutions in Equation 3.3, which the reader should now be able to verify. Spreadsheet software is generally able to perform matrix operations including the inverse, multiplication, and transpose operations.
3.3 Factorial Designs The type of design shown in Figure 3.1 is known as a full factorial two-level experimental design with a center point. Not including center points, full factorial twolevel designs require 2f points, directed in f dimensions. In two dimensions, the design points occupy the vertices of a square, in three dimensions—a cube, in four or more dimensions—a hypercube. Although difficult to visualize, higher dimensionality presents no mathematical challenge beyond what has already been presented. These designs can fit a 2f -coefficient model of the following type: f −1
f
∑
y = a0 + a j
j=1
x j + a jk
f − 2 f −1
f
∑∑
x j xk + ahjk
j< k k =1
f
∑ ∑ ∑ x x x + . h j k
h< j
j< k k =1
(3.14)
66
Industrial Combustion Testing
Table 3.1
Table 3.2
A Full Factorial Two-Level Design in Three Factors
23 Factorial Design Using Hypothetical Data
Point
x1
x2
x3
Point
1 2 3 4 5 6 7 8
− − − − + + + +
− − + + − − + +
− + − + − + − +
1 2 3 4 5 6 7 8
Oxygen x1
Temperature x2
H2 in Fuel x3
NOx, ppm y
− − − − + + + +
− − + + − − + +
− + − + − + − +
26.4 30.9 38.1 45.3 29.9 35.3 56.3 57.7
No model can fit more coefficients than observations, and often, the model is truncated far short of 2f coefficients as it is rarely necessary to fit anything greater than a second-order model. Therefore, full factorial designs are typically unnecessary for f > 3 or so. To accommodate fewer experimental points, fractional factorial designs have been developed. We shall discuss each in turn. A two-level full factorial is an experimental design that investigates all possible factor combinations at two levels: high and low. These levels are coded to ±1 using the transforms presented in Equation 3.2. Since the designs use ±1, generally, only the sign of the factor is reported. To construct a full factorial in f factors, prepare f columns of design points, with minus and plus signs alternating in blocks of 2f−k where k is the factor subscript. For example, for f = 3 there will be 2f = 8 design points. Then x1 alternates in blocks of four (23−1), x2 alternates in blocks of two (23−2), and x3 alternates in blocks of one (23−3). Table 3.1 gives the design. Once the design is selected, the experiments are run, the responses are measured, and the coefficients calculated. As an example, the reader should duplicate the least squares solutions calculated earlier. However, something very important is missing from the analysis—an analysis of experimental error.
If the first four experimental points were run before the lunch break where cooler temperatures prevailed and the second four experimental points were run in the afternoon where warmer temperatures prevailed, then x1 would be confounded with ambient temperature. If a1 were judged to be significant, it would be impossible to know whether the effect was due to variations in x1 or variations in the temperature. Even worse, since we are unaware that ambient temperature affects the results, we would ascribe the results to a1 even though there may be no such association! Therefore, randomizing the run order is an important prophylactic against serial correlation. The presumption in Equation 3.1 is that there is only one error structure to worry about. That is, it presumes that all experiments are run in such a way that they differ only with respect to changes associated with various levels of the factors. This would be likely with randomized runs occurring over a short span of time but unlikely if, for example, half the experiments were run in January and the other half in October. Unknown factor variations in feedstock, ambient conditions, differences in equipment used, and so on could all affect the results. If, however, we can presume that Equation 3.1 is an accurate model, then we can estimate the statistical significance of each coefficient. To do this, let us consider the example given in Table 3.2. We postulate the following model
3.4 About Experimental Error
Cognizance of experimental error is critical for understanding the meaning of the coefficients and conducting the experiments themselves. As a first principle, factorial experimental designs should be run in fully randomized order. This is necessary in order to break any serial correlations that may exist. For example, suppose the design of Table 3.1 were run in the order presented. Further suppose that ambient temperature is an important but unrealized factor affecting the responses.
We have truncated the design at second-order. This is equivalent to presuming that the x1x2x3 interaction is insignificant. Presuming higher order interactions are negligible is usually a good assumption. However, if we would like an independent error assessment, we can duplicate some points. We discuss this later. For now, we shall continue on the basis of our assumption. Equation 3.16 shows the normal equation and the solution to the
3
© 2011 by Taylor and Francis Group, LLC
y = a0 +
∑ k =1
2
ak x k +
3
∑∑ a x x + ε jk
j k
(3.15)
j< k k =1
67
Experimental Design
coefficients. Table 3.2 results in the following normal equation and coefficient estimates. 318.5 8 120.9 82.6 57.0 = 38.9 − 3.1 0.9
8 8 8 8 8
a0 a0 39.8 a a 15.1 1 1 a2 a2 10.3 a ; a = 7.1 . 3 3 a12 a12 4.9 a13 a13 −0.4 8 a23 a23 0.1
(3.16)
The problem now is to decide if the model as a whole is significant, and if so, which coefficients are significant. This is most easily done in an analysis of variance table (ANOVA).
Therefore, any of the quantities in Equation 3.19 may be used interchangeably for yˆ T yˆ in Equation 3.18. Note also the following equalities that come in handy when constructing the ANOVA. SSM = ( yˆ T yˆ − y T y ) = a 1T… ( X T y )1 ,
SSR = ( y T y − yˆ T yˆ ) = y T y − a T ( X T y ) ,
(3.20)
SST = ( y T y − y T y ) = y T y − a T0 ( X T y )0 where T is the coefficient row vector, starting from 1 a 1 (the second element) rather than 0 (the first element), T ( X y )1 is the XTy column vector starting from 1 (the second element) rather 0 (the first element), a T0 is a0, the first element of the coefficient vector, ( X T y )0 is the first element of the XTy column vector (the one associated with a0).
3.5 Sums of Squares ANOVA is constructed using the following equation:
∑ ( yˆ k
− y) + 2
k
∑(y
k
2 − yˆ ) =
k
∑(y
− y ) , 2
k
(3.17)
k
where k indexes the observations, yk is the kth observation, y is the average of all observations, ∑kyk/n, and, yˆk is the kth predicted observation (least squares solution). The first term is known as the sum of squares, model (SSM). The second term is known as the sum of squares, residual (SSR), and the final term is known as the sum of squares, total (SST). Equation 3.16 is true for any least squares solution whatsoever. However, SSM can be apportioned by component (i.e., a0, a1, a2, …) only for so-called orthogonal models or data sets—that is, those that generate diagonal XTX matrices (as is the case for factorial designs). Equation 3.16 also has a matrix formulation:
( yˆ T yˆ − y T y ) + ( y T y − yˆ T yˆ ) = ( y T y − y T y ) .
(3.18)
Also, since y = Xa, by the rules of matrix algebra, Equation 3.19 also holds
yˆ T yˆ = a T ( X T X ) a = a T ( X T yˆ ) = a T ( X T y )
© 2011 by Taylor and Francis Group, LLC
(3.19)
3.6 Degrees of Freedom The sums of squares are an important component of the ANOVA table that we shall construct. Another important component is known as the degrees of freedom (DF). The total degrees of freedom (DFT) is given by n – 1, because if we know y and n – 1 other of the yk values, we can calculate the remaining yk value. Likewise, there are m – 1 degrees of freedom for the model (DFM) because yˆk is determined from m parameters (i.e., the model coefficients) and we must subtract 1 DF for y. Finally, there are n – m degrees of freedom in the residual (DFR) because the last term comprise n values of yk and m values to determine yˆk. We are now ready to build our ANOVA (Table 3.3).
3.7 The ANOVA Making use of these terms, the general ANOVA table is organized as shown in Table 3.3. Applying this to the data of Table 3.2 and making use of Equation 3.15 gives the ANOVA shown in Table 3.4. We shall take this opportunity to point some salient features of the ANOVA table. The first three columns of cell entries are defined by appending the left-hand column entry (M, R, or T) to the top row
68
Industrial Combustion Testing
Table 3.3 The General ANOVA for Model and Residual Terms Parameter M
F
SS
DF
MS
2 ∑ ( yˆ k − y )
m−1
2 ∑ ( yˆ k − y )
k
k
2 ∑ ( yˆ k − y )
m−1 k 2 ∑ ( y k − yˆ ) N − m
m−1
centered at zero by letting z = y – μ/σ. Unfortunately, we typically do not know µ and σ precisely. However, if our sample is sufficiently large, we can estimate them with good confidence. In that case, we use y to estimate the mean and s to estimate the standard error. These are given by the following formulas:
k
R
2 ∑ ( y k − yˆ )
∑ ( y k − yˆ )
n−m
k
k
∑ ( yk − y )
T
2
2
N−m
y=
n−1
k
Table 3.4
ANOVA for Table 3.2 and Equation 3.15 Parameter
SS
M R T
DF
3275.7 5.3 3281.0
6 1 7
F
MS 546.0 5.3
103.4
entry (SS, DF, or MS). For example, SSR = 5.7, DFM = 6, and MSM = 546.0. The mean square is MS and F is Snedecor’s ratio. We shall have more to say about both of these presently. But first, we digress to a discussion of error.
3.8 Error If the model accounts for all of the major influences in the response then the residual cumulates all the minor errors. The central limit theorem states that these errors will be distributed as a normal distribution, characterized by a mean (µ) and a standard error (σ) according to Equation 3.21.
PN ( y ) ∼
1 e 2 πσ 2
1 y−µ − 2 σ
2
, PN ( z) ∼
y−µ and z = σ
1 e 2π
z2 − 2
=
, 2 2 πe ( z ) (3.21)
where PN(y) is the probability that the normal distribution will have the value at y. In Equation 3.21, the parameter µ indicates where the distribution will be centered (central tendency) and the parameter σ indicates the dispersion (how narrow or broadly the distribution will spread with y). This is the familiar bell-shaped curve shown in so many statistics books. Generally the curve is normalized in terms of standard error units (σ) and
© 2011 by Taylor and Francis Group, LLC
N
∑(y k
k
k
k
,
(3.22)
− y)
2
.
n−1
(3.23)
Note that in Equation 3.23, the denominator is reduced by 1 in order to subtract the degree of freedom used to estimate the mean, as discussed previously. Equation 3.22 presumes that each observation differs only by random error, ε. Since the long run average of ε is 0, then averaging with a sufficiently large sample will give a good estimate for µ by way of y. However, an arithmetic average may not be used to estimate σ because the result will always tend to zero. Rather, we shall use a root mean square (RMS) procedure to derive the dispersion parameter. If the various yk differ 2 only by experimental error, then ( y k − y ) is a measure of σ2, known as the variance. The long run average of these measures will tend toward σ2. However, a short run average will be biased if we divide by n. This is because we have used the data themselves to estimate y. Therefore, we must subtract one degree of freedom in order not to double count one data point and unfairly shrink our estimate of σ2. The true variance is estimated by s2:
1
s=
∑y
s2 =
∑(y k
k
− y)
n−1
2
.
(3.24)
Returning to the ANOVA, let us consider what would happen if the model were not statistically significant. In that case there would be no difference between yˆ and y, except for experimental error; that is, y − yˆ = ε. In that case, both MSM and MSR would estimate σ2. In such a case, the ratio MSM/MSR ~ 1. Indeed, there would be no reason to use two separate estimates of the same quantity and we would simply pool both estimates. By inspection, the reader may see that adding SSM + SSR and dividing by DFM + DFR, we arrive at Equation 3.24. If, on the other hand, the model were statistically significant, it would be improper to pool the estimates and Equation 3.24
69
Experimental Design
would not be appropriate for estimating σ2. In such a case, MSR would be the appropriate estimate for σ2 because it alone averages the random error. Moreover, MSM/MSR >> 1 in such a case. The question now arises. How much g reater than 1 should MSM/MSR be in order to decide that the model is significant? To decide, we must know the probability distribution for MSM/MSR. One thing is certain, it will not be distributed as PN because the range of PN is –∞ < PN < ∞, while σ2 can never take on negative values. In fact, the ratio of two variances is distributed as Snedecor’s F distribution. If both belong to the same population then F ~ 1. If MSM and MSR belong to different populations (indicating that MSM and MSR are not derived from the same distribution of σ2) then F >> 1. Tabulations exist for critical values (Fcrit) of the F distributions. These are shown in Appendix A. To determine Fcrit from the table, we need to know three things. First, how many degrees of freedom are in the model (DFM)? Second, how many degrees of freedom are in the residual (DFM)? Third, since we can never be perfectly confident that the values do not differ by some chance event, what probability of a false positive (α) are we willing to tolerate? Generally α is set to 0.05 (sometimes noted as p = 0.05; i.e., 95% confidence that if F exceeds Fcrit, the model accounts for significantly more variance than could occur merely by chance). From Appendix A we see that Fcrit (6, 1, 0.05) = 234.0; however, F = 103.4. Since F fails to equal or exceed Fcrit, the model is not statistically significant as a whole. However, perhaps some parameters are statistically significant. An advantage of factorial designs is that they are orthogonal and therefore permit separate assessment of each model parameter. We can reconstruct the ANOVA table (see Table 3.5) using the kth sum of squares as ak(XT y)k and noting that Table 3.5 ANOVA with an Accounting of Individual Parameters Parameter
F
SS
DF
MS
a1
a1(XTy)1
1
a1(XTy)1/1
a1 (X T y)1 / ∑ ( y k − yˆ ) / 1
a2
a2(XTy)2
1
a2(XTy)2/1
a2 (X T y)2 / ∑ ( y k − yˆ ) / 1
a3
a3(XTy)3
1
a3(XTy)3/1
2 a3 (X T y)3 / ∑ ( y k − yˆ ) / 1
a12
a12(X y)12
1
a12(X y)12/1
2 a12 (X y)12 / ∑ ( y k − yˆ ) / 1
T
T
2
k
2
k
k
1
a13(XTy)13/1
a13 (X y)13 / ∑ ( y k − yˆ ) / 1
a23
a23(XTy)23
1
a23(XTy)23/1
a23 (X T y)23 / ∑ ( y k − yˆ ) / 1
R
∑ ( y k − yˆ )
1
∑ ( y k − yˆ ) / 1
T
3281.0
2
k
7
© 2011 by Taylor and Francis Group, LLC
Parameter a0 = 39.8 a1 = 15.1 a2 = 10.3 a3 = 7.1 a12 = 4.9 a13 = −0.4 a23 = 0.1 R T
SS
DF
MS
(presumed significant) 1827.2 1 1827.2 851.9 1 851.9 406.3 1 406.3 189.0 1 189.0 1.2 1 1.2 0.1 1 0.1 5.3 1 5.3 3281.0 7
F * 345.9* 161.3* 76.9* 35.8* 0.2 0.0
* Significant effects.
Table 3.7 ANOVA for Table 3.2 and Equation 3.15 with a Reduced Model: y = a0 + a1x1 + a2x2 + a3x3 + a12x1x2 + ε Parameter M R T
SS
DF
MS
F
3274.5 6.6 3281.0
3 3 6
1091.5 2.2
498.5*
* Significant effects.
SSM = ∑ mk =1 ak ( X T y )k ; that is, excluding a 0(XT y)0. Note we are asking the following statistical question. Do any parameters besides a 0 add some significant advantage to the model? Even if there are good theoretic reasons to exclude a 0 in our model, we shall always compare the final model with a model comprising a single parameter—the mean—which for factorial designs is a 0. Therefore, a 0 is excluded from the analysis. Table 3.6 shows the values. From Table 3.6 we see that a1 through a12 are statistically significant as indicated by an asterisk in the F column; a 0, in the absence of other parameters, is equivalent to y the hypothesis against which we are testing all other effects. However, a13 and a23 are not statistically significant. As such, they can be pooled into the residual. Since a1 through a12 are significant, they could be pooled into the model. If this were done, the reduced model would now show itself to be significant (see Table 3.7).
k
a13(XTy)13
2
ANOVA for Table 3.2 and Equation 3.15 with an Accounting of Individual Parameters
T
a13
k
Table 3.6
2
T
k
2
k
3.9 Using Center Points and Replicates The residual error may be divided into two components—random error (also called pure error, the
70
Industrial Combustion Testing
accumulation of manifold minor sources that act in no organized way) and bias error (a systematic error that is not random). Bias error is also called lack of fit. Now if we replicate a point exactly and randomize the run order, any difference in responses to the replicate points should be due to random error alone. If we subtract this from the residual, then what remains is an estimate for bias or lack of fit. We can test the bias component in the same way as any other and see whether it is indeed significant. If it is significant, we can augment the model with additional points to eliminate any bias caused by an incorrect model specification. If the bias component is not significant then we will have additional confidence that our model is adequate and its terms are significant. We want to add replicate points in a way that will not disturb the orthogonal properties of the XTX matrix. One way to do this is to replicate all points in a design. Another way is to add several center points. Replicating center points helps distinguish bias error in two ways—replicating them allows for an independent assessment of pure error and their location allows us to detect any curvature or systematic bias in our model. Additionally, replicating center points is typically less expensive experimentally as we do not need to double or triple the existing number of runs. Table 3.8 shows the design augmented with three additional points. We must now add two additional rows (four additional entries) to our ANOVA. We need a row for bias comprising the appropriate entries for sums of squares, bias (SSB) and the associated degrees of freedom (DFB). Additionally, we need the pure error terms (SSE
Table 3.8 23 Factorial Design Plus 3 Center Points (Hypothetical Data) Oxygen x1
Temperature x2
H2 in Fuel x3
NOx, ppm y
1 2 3 4 5 6 7 8
− − − − + + + +
− − + + − − + +
− + − + − + − +
12.7 25.8 21.7 38.6 32.3 47.2 64.1 76.1
9 10 11
0 0 0
0 0 0
0 0 0
39.5 41.4 39.3
Point
© 2011 by Taylor and Francis Group, LLC
and DFE). Regarding the bias, we have the following formulations: n
SSB =
∑(y
k
2 − yˆ k ) = y k T y k − yˆ T yˆ , DFB = u − m (3.25)
k =1
where k indexes each response, y k is the mean of all the runs performed at the design coordinate associated with yk. In the case of nonreplicated points, y k = yk because a single value is its own mean. The number of unique points is u (i.e., number of coordinates or unique positions in factor space), and m is the number of model parameters (includes a0). Regarding the pure error, we may account for it thus: n
SSE =
∑(y
− y k ) = y T y − y k T y k , DFB = n − u (3.26) 2
k
k =1
The total number of points is n, and u is the number of unique points (i.e., number of coordinates or unique positions in factor space). Constructing the ANOVA from Table 3.8 and Equation 3.15 we note that n = 11; that is, there are a total of 11 runs, u = 9; that is, there are nine unique coordinate points (8 factorial coordinates and 1 center point coordinate); therefore, u = 9, and m = 7; that is, there are seven model parameters—a 0, a1, a2, a3, a12, a13, a23. If we chose to fit the reduced model—a 0, a1, a2, a3, a12—then m = 4. Table 3.9 shows the generic ANOVA table for individual model parameters and with residual split into bias and pure error components. Table 3.10 gives the actual numbers. For convenience, we have appended the values of F to the ANOVA table and added an asterisk for statistically significant effects. From the table in Appendix A, we have Fcrit(0.05, 1, 1) = 18.5 and Fcrit(0.05, 2, 2) = 19.0. As before, a0 through a12 are significant; but notably, the bias is not significant since 2.7 < 19.0. In that case, our original decision to use the x123 effect as an estimate of error seems justified. In such a case, we can pool the bias with the pure error and x13 and x23 effects, thus obtaining Table 3.11. Note that the critical threshold has fallen from 18.5 to 6.0 owing to the additional degrees of freedom Fcrit(0.05, 1, 6). Thus, we can assess statistical significance with better precision. Finally, since MSR is our best estimate of σ2, we estimate σ ~ √2.7 ≈ 1.6 ppm (having the same units as the response).
71
Experimental Design
Table 3.9 ANOVA for Factorial with Replicate Points SS
DF
MS
F
a1(XTy)1
1
a1 ( X T y )1
a1 ( X T y )1 n − u 1 2 ∑ ( yk − yk )
Parameter Model, M SSM = a1…(XTy)1… = 2 ∑ ( yˆ k − y )
a1
1
k
DFM = m – 1
k
a2 ( X T y ) 2 n − u 1 2 ∑ ( yk − yk )
a2
a2(X y)2
1
am–1
am–1 (XTy)m–1
1
am − 1 ( X T y ) m − 1
Bias, B
n
2 ∑ ( y k − yˆ )
Pure error, E Total, T
k =1
am − 1 ( X T y ) m − 1 n − u 1 n 2 ∑ ( yk − yk )
1
u–m
k =1
∑ ( y k − yˆ k )
2
∑ ( yk − yk )
2
k =1 n
n− u u − m
k =1
n
n
∑ ( yk − yk )
2
n–u
k =1
∑ ( yk − y )
n
1
k =1
DFR = n – m
k =1
a2 ( X T y ) 2
T
n
Residual, R 2 SSR = ∑ ( y k − yˆ )
n
∑ ( yk − yk )
k =1
2
n− u
n–1
2
k
Table 3.10 ANOVA for Factorial with Replicate Points with Table 3.8 and Equation 3.15 as the Basis Parameter a0 = 39.9* a1 = 15.1* a2 = 10.3* a3 = 7.1* a12 = 4.9* a13 = –0.4 a23 = 0.1 B E T
SS
DF
MS
(presumed significant) 1827.2 1 1827.2 851.9 1 851.9 406.3 1 406.3 189.0 1 189.0 1.2 1 1.2 0.1 1 0.1 5.4 2 2.7 2.6 2 1.3 3283.8 10
3.10 Fractional Factorials
F
Fcrit
* 1827.2* 851.9* 406.3* 189.0* 1.2 0.1 1.4
18.5 18.5 18.5 18.5 18.5 18.5 19.0
Full factorials are a good way to fit models such as Equation 3.15. However, they quickly become unwieldy for f > 5. Moreover, in principle, one needs only as many unique data points as there are coefficients to fit, plus additional replicates as necessary. In such cases, full factorials may have many more runs than are necessary. For example, consider that we wish to investigate performance in response to the following factors: • ξ1: oxygen concentration in the furnace, from 1 to 3% • ξ2: hydrogen concentration in the fuel, from 0 to 25% • ξ3: propane concentration in the fuel, from 0 to 25% • ξ4: furnace temperature, from 1400 to 1800°F • ξ5: fuel injection angle, from 5 to 15°
* Statistically significant, α = 0.05.
Table 3.11 ANOVA for Factorial with Replicate Points with Table 3.8 and Equation 3.15 as the Basis Parameter a0 = 39.9* a1 = 15.1* a2 = 10.3* a3 = 7.1* a12 = 4.9* R T
SS
DF
MS
(presumed significant) 1827.2 1 1827.2 851.9 1 851.9 406.3 1 406.3 189.0 1 189.0 9.3 6 2.7 3283.8 10
* Statistically significant, α = 0.05.
© 2011 by Taylor and Francis Group, LLC
F
Fcrit
* 1827.2* 851.9* 406.3* 189.0*
6.0 6.0 6.0 6.0
Suppose further that we are only interested in the first-order terms per Equation 3.27: 5
yˆ = a0 +
∑a x
k k
k =1
.
(3.27)
72
Industrial Combustion Testing
The formula has 1 zero-order coefficient (a0), 5 first-order coefficients (a1 → a5), and needs at least one degree of freedom to estimate of error. Then the minimum number of runs required is 7. However, a full factorial would have 25 = 32 experimental points and be able to fit the following 32 coefficients associated with the respective factors and interactions.
a0 a5 a23 a45 a135 a345 a2345
a1 a12 a24 a123 a145 a1234 a12345
a2 a13 a25 a124 a234 a1235
a3 a14 a34 a125 a235 a1245
a4 a15 a35 a134 a245 a1345
This seems excessive. Can we reduce the number of required experiments? The answer is “Yes”, if we are willing to give up on higher-order effects. Fractional factorial coefficients offer one method for doing this and require 1/2 k · 2 f or 2f−k runs where 1/2k represents the fraction. For example, if we run the half fraction (k = 1), this will require only 16 runs. We do this by first preparing a 24 factor design and then we add a fifth column. We shall describe presently how the sign order is determined in this last column. For now, please examine Table 3.12. This design has all the desirable orthogonal properties of the full factorial. However, we have not gotten something for nothing. Although we will be able to determine up to 16 coefficients, they will all be confounded with others. We show the confounding pattern presently. Rather than list the coefficients, we merely list
Table 3.12 A ½ 25 Fractional Factorial Point
x1
x2
x3
x4
x5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
− − − − − − − − + + + + + + + +
− − − − + + + + − − − − + + + +
− − + + − − + + − − + + − − + +
− + − + − + − + − + − + − + − +
+ − − + − + + − − + + − + − − +
© 2011 by Taylor and Francis Group, LLC
the factor numbers. For example, the x1x2x5 interaction requiring determination of a125 coefficient will be represented merely by the factor list 125. 0 = 12345 1 = 2345 2 = 1345 3 = 1245 4 = 1235 5 = 1234 12 = 345 13 = 245 14 = 235 15 = 234 23 = 145 24 = 135 25 = 134 34 = 125 35 = 124 45 = 123 From this list, we see that every coefficient is confounded with another. For example, a 34 is confounded with a125. This means though it will be possible to determine the a 34 coefficient (or the a125 coefficient— both are identical), it will not be possible to separately know the influence of x34 from x125 or the individually associated coefficients. Both effects will be merged. Notwithstanding, third- and higher-order effects are rarely important. Therefore, this is not a significant loss, and this is what makes fractional factorials so powerful. In this design, we see that no two-factor interaction is confounded with another two-factor interaction. Why not fractionate further? Let us examine the confounding pattern for the ¼ 25 factorial (see Table 3.13). Confounding pattern for Table 3.13: 0 = 145 = 235 = 1234 1 = 45 = 234 = 1235 2 = 35 = 134 = 1245 3 = 25 = 124 = 1345 4 = 15 = 123 = 2345 5 = 14 = 23 = 12345 12 = 34 = 135 = 245 13 = 24 = 125 = 345 We see from the confounding pattern that main and two-factor effects are confounded with other two-factor effects. Since two-factor interactions are often significant, highly fractionated designs like this are typically used for screening. In other words, we believe that firstorder effects will likely show up if significant; however, their values are subject to error. The reason for using highly fractionated designs is to weed out less important factors and reduce the design to fewer factors that can be investigated in greater detail, perhaps with a full factorial. Table 3.13 A ¼ 25 Fractional Factorial Point
x1
x2
x3
x4
x5
1 2 3 4 5 6 7 8
− − − − + + + +
− − + + − − + +
− + − + − + − +
− + + − + − − +
+ − − + + − − +
73
Experimental Design
y02
3.11 Generating the Fractional Design So how do we generate the fractional factorial? As stated before, we first prepare a 2f−k and determine the k remaining columns by selecting k defining contrasts. A defining contrast is any factor or factor interaction that occurs in the full factorial (e.g., 123, 1234, 245, etc.). We choose the defining contrasts to be as large as possible without significant overlap. For example, the ½ 25 factorial has k = 1. So we need to select 1 defining contrast; we do not need to worry about a second defining contrast or its overlap with the first. In such a case, we shall use 12345, the largest possible contrast in the 25 factorial. That is how column 5 in Table 3.13 was derived: by multiplying 1234 (that is x5 = x1x2x3x4). For example, the value for x5 in row 12 is found by x5 = (+)(−)(+)(+) = (−). This is equivalent to setting 12345 = + 1 in all rows. (We could have just as easily set 12345 to −1 as well. In that case, all the signs in column x5 would be reversed.)
3.12 Calculating the Confounding Pattern The method for calculating the confounding pattern also involves the defining contrast. The rule is that we shall multiply the contrast with the effect of interest and eliminate the squared components. For example (125)(12345) = 12223452 = 34. Therefore 125 = 34, and the reader may verify that multiplying either the x1x2x5 or x3x4 columns together gives an identical factor pattern. We repeat this for every possible factor combination so the entire confounding pattern may be found. In order to find the ¼ fraction, k = 2, and f – k = 5 − 2 = 3, we start with a 23 full factorial and we choose two defining contrasts. We attempt to choose them with as little overlap as possible because with multiple contrasts we need to worry not only about the defining contrasts but the implicit contrasts created by their interaction. For example, if we choose 1234 and 235 we also have, tacitly, 1223245 = 145, and this is how Table 3.12 was generated. Other choices fare no better (124 and 235 give 1345) and some fare much worse (e.g., 124 and 125 give 5, meaning that we will be confounding a5 with a0 – a very bad idea).
3.13 Central Composite Designs Generally, for large experimental design problems it is useful to tackle the job sequentially. For example, for 11
© 2011 by Taylor and Francis Group, LLC
y22
y12
y0,0
y10
y20
y11
y21
y01 Figure 3.2 A central composite design. In two dimensions, the design comprises eight points and as many centerpoint replicates as required.
factors, one uses a highly fractionated factorial. Suppose such a screening design identifies two factors as important. Then we investigate them with a full factorial having a few center points. But when we analyze the data via ANOVA, we find significant lack of fit (bias). In such a case we may wish to augment the design with additional points that will allow us to fit a full second-order design to include pure quadratics. A very good design for doing this is known as a central composite. It comprises a full or fractional factorial augmented with axial points. Figure 3.2 shows a central composite in two factors. The design comprises four factorial points (connected in the figure for illustration by a dotted square), four axial points, and as many center point replicates as necessary (typically three to six). The central composite design allows one to fit a full second-order model of the following kind: f −1
f
y = a0 +
∑ k =1
ak x k +
f
∑∑ j< k k =1
f
a jk x j xk +
∑a
x + ε. (3.28)
2 kk k
k =1
Note the last summand indexes pure quadratic terms; that is, a11x12 + a22 x22 + . Central composites interrogate the factor space at five levels along each axis. The question becomes what axial levels should we use? We have two main choices.
1. Make the design strictly rotatable (e.g., circular, spherical, or hyperspherical) by using α = f ,
74
Industrial Combustion Testing
where all axial points lie along an axis ±α from the center. 2. Make the design strictly orthogonal by choosing α= nn f − n f / 2 , where n is the total number of points in the design, and nf is the number of factorial design points.
Typically the first option is chosen and the quadratics are assessed as a single orthogonal entry by fitting the model: f −1
f
yˆ = a0 +
∑
ak x k +
k =1
f
∑∑
f
a jk x j xk +
j< k k =1
∑a
kk
( xk2 − qk ) (3.29)
k =1
where qk = ∑ nk =1 xk2 /n. For example, consider the central composite design of Table 3.14. For this design, we shall fit the model of Equation 3.30: yˆ = a0 + a1 x1 + a2 x2 + a12 x1 x2 + [ a11 ( x12 − q1 ) + a22 ( x22 − q2 )]
(3.30)
where q1 = q2 = 2/3. Subtracting the mean square from x1 and x2 keeps them orthogonal to a0 and the rest of the model terms. Although the bracketed terms are orthogonal to all other model terms, a11 is not orthogonal with respect to a22. Therefore, we shall assess the entire bracketed quantity separately in the ANOVA. If it is significant we shall not know if one or both quadratic terms are significant. There is also a small chance that both terms could be significant but of exactly opposite effect so that they do not appear significant when pooled in the ANOVA. This is not very likely but it is possible.
Table 3.14
Table 3.14 generates the coefficients and ANOVA given in Table 3.15. From the ANOVA, all the parameters are significant including the entry for the quadratics. However, the quadratic coefficients are widely divergent with a11 = 0.2 and a22 = 3.9. It appears that a22 could be the only reason that the quadratic entry is significant. It is also apparent that there is no significant bias, meaning the model fits the data adequately. Therefore, we can pool the bias and pure error into a single residual, MSR. This gives Table 3.16. In this case MSR = 0.08 and s = √0.08 = 0.29. To assess the quadratic effects separately, we can compare each parameter with its standard error. The standard error is defined by Equation 3.31:
sk = s 2 ( X T X ) k , k −1
(3.31)
where sk is the standard error for the kth effect, s is the standard error given by the best estimate of error (√MSR if the bias and error can be pooled or √MSE if it cannot), −1 and ( X T X )k , k is the kth diagonal element of the inverse matrix.
Table 3.15 ANOVA with Coefficients for Table 3.14 Data Parameter a1 = 30.1* a2 = 15.0* a12 = 5.1* a11 = 0.2* + a22 = 3.9 B E T
SS
DF
MS
F
Fcrit
1800.12 799.62 103.02 100.85 0.06 0.43 2805.79
1 1 1 2 3 3 11
1800.12 799.62 103.02 50.43 0.02 0.14
12,559* 5579* 719* 352* 0.1
10.13 10.13 10.13 9.55 9.28
* Significant at p = 0.05.
A Central Composite Design in Two Factors Point
x1
x2
Y
1 2 3 4 5 6 7 8
− − + + 0 0
−
−√2 √2
+ −√2 √2 0 0
11.6 21.5 31.4 61.6 21.0 49.2 6.4 48.9
9 10 11 12
0 0 0 0
0 0 0 0
27.0 27.8 27.2 27.0
© 2011 by Taylor and Francis Group, LLC
+ −
Table 3.16 ANOVA with Coefficients for Table 3.14 Data Having a Pooled Residual Parameter a1 = 30.1* a2 = 15.0* a12 = 5.1* a11 = 0.2* + a22 = 3.9 R T * Significant at p = 0.05.
SS
DF
MS
F
Fcrit
1800.12 799.62 103.02 100.85 0.49 2805.79
1 1 1 2 6 11
1800.12 799.62 103.02 50.43 0.08
12,559* 5579* 719* 352*
5.99 5.99 5.99 5.14
75
Experimental Design
Table 3.17
Table 3.18
Analysis of Standard Error of Effects
A 24 Factorial Design in Two Blocks
Coefficient a0 = 30.1* a1 = 15.0* a2 = 10.0* a12 = 5.1* a11 = 0.2 a22 = 4.0*
sk 0.08 0.10 0.10 0.14 0.11 0.11
t Ratio
tcrit
Point
x1
x2
x3
x4
Block
365.68* 149.04* 99.34* 35.66* 2.11 35.21*
2.45 2.45 2.45 2.45 2.45 2.45
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
− − − − − − − − + + + + + + + +
− − − − + + + + − − − − + + + +
− − + + − − + + − − + + − − + +
− + − + − + − + − + − + − + − +
II I I II I II II I I II II I II I I II
* Significant at p = 0.05.
Equation 3.32 shows the inverse matrix, (XTX)−1. Note the matrix is not quite diagonal with the quadratic entries mutually biasing one another: 1 12 −1 T (X X ) =
18 18
14 5 32 1 32
. 1 32 5 32 (3.32)
Table 3.17 presents the results from an analysis of the standard errors of each effect. Dividing the coefficient value by the standard error for each effect gives the t ratio. The critical t ratio is given by tcrit = √Fcrit(p, 1, DFR) ratio. Here we see that a22 exceeds tcrit while a11 does not. This method is not foolproof since a11 and a22 are mutually biased. However, for central composite designs the degree of bias is often slight.
3.14 Blocking Designs having large numbers of experimental runs or experiments run sequentially often cannot be run under homogeneous conditions. This is because during intermittent periods, important factor changes may occur. These include ambient factors, time-based factors such as settling or aging of raw materials, and batch effects. If we presume that such effects will alter the mean response but not dramatically affect other coefficient values, we can eliminate bias in the analysis by a technique known as blocking. In general one should “block whenever possible and randomize the remainder.” For example, suppose we wish to test a burner using a full factorial design in four factors. However, we only
© 2011 by Taylor and Francis Group, LLC
have a limited time to run the tests. Perhaps the burner is only available for two days separated by some intervening gap of days or weeks. During the intervening time many things could change. Perhaps the oil fuel quality changes over time, or unbeknown to us, some ambient conditions are important and these conditions have changed. Anything that is not specifically accounted for in the declared model will accumulate in the error term, inflating it. That is, in a fully randomized experiment, the error term will contain not only random error but also systematic bias of which we are unaware. This kind of bias cannot be measured because we do not even know that it exists. Notwithstanding, our ignorance will not prevent this influential step change from affecting our results. Table 3.18 shows a suggested way of blocking the design. In the first set of tests we run points 2, 3, 5, 8, 9, 12, 14, 15 in randomized order. In the second set of tests we run the remainder, also in a randomized order. Before describing how we arrived at such a blocking arrangement, let us examine a very poor blocking arrangement—namely, Block I = Points 1–7, Block II = Points 8–16. If we use this blocking arrangement, and we fail to randomize runs, we will have committed two sins and completely confounded any block effect with the factor effect, x1. The block effect will be ascribed falsely to x1 because x1 is run at its low level in tests 1–7 and at its high level in runs 8–16. In short, x1 changes when the block changes and the effects are hopelessly entangled. What we desire is a block effect that will be orthogonal to the effects of interest. However, we tacitly learned how to block orthogonally when we learned how to fractionate a design. We
76
Industrial Combustion Testing
Table 3.19
Table 3.20
A 2 Full Factorial Design in Four Blocks with Eight Center Points
t Tests for Factor Effects for Table 19 Data
4
Point
x1
x2
x3
x4
b1
b2
1 2 3 4 5 6
+ − + − 0 0
+ − − + 0 0
+ + − − 0 0
− − + + 0 0
− − − − 0 0
− − − − 0 0
7 8 9 10 11 12
− + + − 0 0
+ − + − 0 0
+ + − − 0 0
+ + − − 0 0
− − − − 0 0
+ + + + 0 0
13 14 15 16 17 18
− + − + 0 0
− − + + 0 0
− + + − 0 0
+ − − + 0 0
+ + + + 0 0
− − − − 0 0
19 20 21 22 23 24
+ + − − 0 0
+ − + − 0 0
+ − − + 0 0
+ − − + 0 0
+ + + + 0 0
+ + + + 0 0
Coefficient
Block
y
I
19.5 16.7 26.4 12.1 17.5 18.1
II
27.2 41.0 7.0 4.5 20.7 19.4
III
16.5 31.9 15.6 21.1 20.5 20.2
IV
35.3 17.5 1.5 30.3 20.2 21.0
should assign our blocks in the same way: use a defining contrast. Since there is only one block, we use the largest possible contrast—1234B where B stands for the block effect. For this reason, the blocking pattern takes on a similar pattern to x5 in the fractional factorial of Table 3.12. We can also add some center points to each block and randomize them along with the rest of the runs in order to assess pure error within each block. The axial and factorial points of a central composite design are also orthogonal to one another, so one can run those designs sequentially and subtract the block effects from the error, thus making the F tests more sensitive. What if we need four blocks instead of two? Then, we would add two blocking factors (b1, b2) in the same way that we would add two additional factors in a fractional factorial. We would then assign the four combinations to a block. For example, (− −) = I, (− +) = II, + −) = III, and (+ +) = IV. Table 3.19 gives an example with defining contrasts being 124b1, 34b2, and 123b1b2. Therefore, 124, 34, and 123 are confounded with block effects. Let us presume that we can tolerate no more than six runs at any time and that there is a possibility
© 2011 by Taylor and Francis Group, LLC
sk
Value
a0* a1* a2* a3* a4* a12* a13 a14 a23 a24* a34 a123 a124 a134 a234 a1234 Block I Block II Block III Block IV
20.07 4.71 −2.84 6.93 5.98 −1.39 0.03 0.01 0.06 0.53 −0.34 0.16 0.18 −0.04 0.06 0.13 −1.90 0.35 0.65 0.90
0.16 0.19 0.19 0.19 0.19 0.19 0.19 0.19 0.19 0.19 0.34 0.34 0.34 0.19 0.19 0.19 0.48 0.48 0.48 0.48
t 126.1* 24.1* −14.6* 35.6* 30.7* −7.1* 0.2 0.0 0.3 2.7 −1.0 0.5 0.5 −0.2 0.3 0.7 −4.0 0.7 1.4 1.9
* Significant at p = 0.05, tcrit = 2.57.
Table 3.21 ANOVA for Table 3.19 Data Coefficient
DF
SS
MS
F
Fcrit
a1 a2 a3 a4 a12 a13 a14 a23 a24 a34 a123 a124 a134 a234 a1234 Blocks R T
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 3 5 23
354.38 129.39 768.68 572.41 31.08 0.02 0.00 0.05 4.52 0.63 0.13 0.18 0.03 0.05 0.28 9.93 3.04 1874.78
354.38 129.39 768.68 572.41 31.08 0.02 0.00 0.05 4.52 0.63 0.13 0.18 0.03 0.05 0.28 3.31 0.61
582.82 212.80 1264.18 941.39 51.12 0.03 0.00 0.08 7.43 1.04 0.21 0.29 0.05 0.08 0.45 5.44 0.09
6.61 6.61 6.61 6.61 6.61 6.61 6.61 6.61 6.61 6.61 6.61 6.61 6.61 6.61 6.61 5.41
that significant changes that affect our response may occur between blocks. Table 3.20 gives the factor effects. Tables 3.20 and 3.21 show the associated t and F tests. For these data, lack of fit turns out not to be significant. Therefore, we may use the pooled residual to perform our tests.
4 Fluid Flow Wes Bussman and Joseph Colannino Contents 4.1 Introduction...................................................................................................................................................................... 77 4.2 Gas Properties.................................................................................................................................................................. 78 4.2.1 Density.................................................................................................................................................................. 78 4.2.2 Ratio of Specific Heats and Related Concepts................................................................................................. 78 4.2.3 Viscosity................................................................................................................................................................ 78 4.2.4 Fuel Heating Value.............................................................................................................................................. 78 4.3 Basic Fluid Dynamic Concepts...................................................................................................................................... 79 4.3.1 Definition and Units of Pressure....................................................................................................................... 79 4.3.2 Gauge and Absolute Pressure............................................................................................................................ 79 4.3.3 Draft....................................................................................................................................................................... 79 4.3.4 Compressible Flow through an Orifice............................................................................................................ 80 4.3.5 Pressure Drop in Pipes and Fittings................................................................................................................. 81 4.4 Pressure Measurement Techniques.............................................................................................................................. 82 4.4.1 Manometer............................................................................................................................................................ 82 4.4.1.1 U-Tube Manometer............................................................................................................................... 82 4.4.1.2 Inclined Manometer............................................................................................................................. 82 4.4.2 Bourdon Tube Gauge........................................................................................................................................... 83 4.4.2.1 Design of the Bourdon Tube Gauge................................................................................................... 83 4.4.2.2 Common Failure Mechanisms............................................................................................................ 84 4.4.2.3 Calibration of Pressure Gauges........................................................................................................... 85 4.4.2.4 Selection.................................................................................................................................................. 85 4.4.2.5 Installation............................................................................................................................................. 85 4.5 Differential Producing Flow Meters............................................................................................................................. 86 4.5.1 Orifice Meter......................................................................................................................................................... 86 4.5.2 Venturi Meter....................................................................................................................................................... 89 4.5.3 Turbine Flow Meter............................................................................................................................................. 89 4.5.4 Vortex Flow Meter............................................................................................................................................... 89 4.5.5 Magnetic Flow Meter.......................................................................................................................................... 90 4.5.6 Ultrasonic Flow Meter......................................................................................................................................... 90 4.5.7 Thermal Mass Meter........................................................................................................................................... 91 4.5.8 Positive Displacement Meter.............................................................................................................................. 91 4.5.9 Pitot Tube.............................................................................................................................................................. 92 4.5.10 Annubar................................................................................................................................................................ 94 References................................................................................................................................................................................... 95
4.1 Introduction The hydrocarbon and petrochemical industries employ the flow of many fluids. For example, combustion equipment requires the flow of air and fuel. Then there are the process streams themselves comprising hydrocarbon
liquids and gases flowing under pressure, through pipes and fittings, to their delivery point. Calculation of fluid flow, therefore, is of fundamental importance. The purpose of this chapter is to give the reader a fundamental understanding of some of the fluid dynamic concepts and equipment that are important in testing combustion systems. 77
© 2011 by Taylor and Francis Group, LLC
78
Industrial Combustion Testing
4.2.3 Viscosity
4.2 Gas Properties 4.2.1 Density Density (ρ) is defined as mass (m) divided by volume (V). Accordingly, units of density are g/cm3, kg/m3, lbm/ft3, and so forth. When the volume is macroscopic then, technically, the density is an average density. The general way to define the density at a point is to use the limit at infinitesimal volume.
ρ=
dm ; dV
(4.1)
however, this definition breaks down as we near the molecular level. Therefore, we shall presume that in all cases that matter this may be considered a continuum.
Shear stress (τ) is the force per unit area exerted due to friction of a flowing fluid. If a stationary wall is exposed to a flowing fluid, the velocity at the wall will be zero and the velocity far from the wall will be the free stream velocity. Between these two points a velocity gradient exists. The viscosity (µ) is the constant of proportionality between the velocity gradient (dv/dy) and the shear stress. Figure 4.1 shows the coordinate directions and the variation in velocity from zero at the plate to the free stream velocity, vs. The velocity is flowing in the x direction and the velocity gradient varies in the y direction. The shear stress for ideal flowing gases is given by the following equation and gauges the resistance to establishing a rate of strain (dislocation of the fluid) τ=µ
4.2.2 Ratio of Specific Heats and Related Concepts The specific heat represents the amount of energy required to raise a unit mass one unit in temperature. For gases, the specific heat differs depending on whether the gas is allowed to do work by expanding against an atmosphere (constant pressure definition, cp) or is constrained within a volume (constant volume definition, cv). Examples of units of cp and cv are J/g K or Btu/lbm °F. If we wish to express this on a molar basis we shall use the upper case: that is, Cp or Cv having units of J/mol K or Btu/lbmol °F. The universal gas constant (R) is defined for an ideal gas as follows:
R = PV/nT,
(4.2)
where P is the pressure, V is the volume, n is the number of moles, and T is the absolute temperature.
Cp − Cv = R.
The fuel heating value is the total amount of energy that can be liberated by combustion of a fuel. The heating value may be normalized by mass, moles, or volume; accordingly, it will have units of kJ/kg, J/mol, or J/Nm3. Customary units include Btu/lbm, Btu/lbmol, and Btu/scf. In the case of volume, some standard temperature and pressure must be chosen, such as 25°C and 1 atm. Two examples of unit volumes referenced to such conditions are the normal meter cubed (Nm3) and the standard cubic foot (scf). For hydrocarbons, combustion always produces water and carbon dioxide according to the following equation: CxHy + (x + y/4)O2 = xCO2 + y/2 H2O.
k=
c p Cp = . c v Cv
vs y
(4.3)
© 2011 by Taylor and Francis Group, LLC
(4.4)
(4.6)
For combustion reactions producing water, we may consider two different standard states for calculating the heating value: the standard state may be taken to be either a liquid or vapor at the standard temperature and pressure. If the standard state considers water as a vapor, the heating value so determined is called the
x
It also turns out that the dimensionless ratio of heat capacities (k) is an important thermodynamic parameter:
(4.5)
4.2.4 Fuel Heating Value
By ideal gas we mean a gas whose atomic and molecular volumes and attraction entities exert no significant influence upon the bulk volume or pressure. All real gases deviate from ideality at very low temperatures and very high pressures. In many cases (and especially so for combustion) the ideal gas law is sufficiently accurate for engineering purposes. Remarkably, the universal gas constant and heat capacities are related:
dy . dx
Figure 4.1 Fluid flow shear stress.
79
Fluid Flow
lower heating value (LHV) or net heating value. If the standard state considers water as a liquid, then the heating value so determined is called the higher heating value (HHV) or gross heating value. It is important to understand that this determination is set by convention and has nothing to do with the actual state of the water. The difference between the higher and LHVs is the standard enthalpy of vaporization for water. Some combustion produces no water (e.g., combustion of CO) and therefore such higher and LHVs are identical.
Bernoulli’s equation, as written above, considers no losses due to friction or microscopic recirculation. In order to calculate the total draft for a furnace of a given height (h) and temperature, one may equate the potential energy of the gas owing to its mass in a gravitational field with that of the ambient fluid outside the furnace; that is, air in the usual case. This gives ΔP/Δρ = gh or ΔP = Δρgh where Δρ is the difference in densities between the two fluids. Noting that Δρ = PM/ RΔT where Δρ is the density, P is the ambient pressure, M is the molecular weight, R is the universal gas constant, and ∆T is the absolute temperature difference, we may recast the equation as
4.3 Basic Fluid Dynamic Concepts
∆P =
4.3.1 Definition and Units of Pressure Pressure (P) is defined as a normal force per unit area. For example, the pressure of a standard atmosphere at sea level is ~14.7 lbf/in2 (psi). Other pressure units include kilopascals (kPa), bar (101.3 kPa), and atmospheres (atm).
Problem statement: What is the maximum draft pressure at the floor of a 10-m furnace operating at 800°C? Presume that the control system adjusts the pressure at the top of the furnace to a draft of 3 mm w.c. via a stack damper. Solution: From Equation 2.110, at 1 atm of pressure (101 kPa) we have g 101[kPa ]29 mol ∆P = 3 8.314 m ⋅ Pa mol ⋅ K 1 1 m (800 + 273)[K ] − ( 25 + 273)[K ] 9.8 s 2 10 [m]
4.3.3 Draft
∆P v 2 = . 2 ρ
© 2011 by Taylor and Francis Group, LLC
(4.7)
(4.8)
Example 4.1: Calculation of Draft Pressure
Although the pressure at sea level is 14.7 psi, pressure gauges are calibrated to read 0 at 1 atmospheric pressure. Therefore, when one measures pressure in an automobile tire, say 32 psig, one is measuring the pressure above atmospheric. In this example, the total pressure (also called the absolute pressure) in the tire is actually 46.7 psia because the pressure at sea level is 14.7 psia. In order to distinguish between gauge (g) and absolute (a) pressure, the respective letter is appended to the psi designation.
,
where T∞ is the absolute ambient temperature and Tf is the absolute furnace temperature.
4.3.2 Gauge and Absolute Pressure
Draft is a differential pressure between the outside atmosphere and the internal pressure in a furnace. Differential pressure such as units of draft are sometimes referenced as psid. However, draft pressures are ordinarily so low that smaller units are needed. Two common units are mm w.c. or in w.c. (also w.c. or i.w.c.); that is, millimeters, water column or inches, water column, respectively. For example, in order to displace a fluid from a glass to one’s mouth three inches above the liquid surface via a straw, one must cause a differential pressure of 3 in. w.c. Draft pressures are often smaller. For example, a typical natural draft refinery furnace operates anywhere from 0.25 in. w.c. to 0.50 in. w.c. This seemingly small pressure difference can cause great volumes of air flow. Gas pressure and velocity are related by Bernoulli’s equation:
PMgh 1 1 − R T∞ T f
= −54.6[Pa ][mm w.c.]
The negative sign reminds us that the pressure is below atmospheric. Adding the −3 mm w.c. pressure (draft) we have at the bridgewall gives a total of −8.6 mm w.c. pressure at the floor (or 8.6 mm w.c. draft).
We can also write an equation for the mass flow (m ˙ ) as a function of velocity and density:
= ρAv , m
(4.9)
where A is the flow area. Coupling these two equations one arrives at mass flow as a function of draft pres = A 2ρ ∆P . For purposes of the sure through an area: m calculation, the absolute value of the pressure is used with the understanding that the flow direction is always defined as flow from the higher to the lower pressure. However, as previously noted, Bernoulli’s equation does not consider friction loss. We can account for friction by premultiplying by a loss coefficient, Co. In that case, the
80
Industrial Combustion Testing
mass flow rate as a function of pressure and density becomes = Co A 2ρ ∆P . m
(4.10)
For British units, one should note that 32.174 lbm.ft/lbf−s2 = 1 = gc, where lbm is the pound mass unit and lbf is the pound force unit. Force and mass are related by Newton’s second law, F = mg, (4.11) where F is the force, m is the mass, and g is the gravitational acceleration (9.8 m/s2 or 32.174 ft/s2). By definition, 1 lbm at 1 g of gravity exerts 1 lbf. Therefore, 1 lbf = 32.174 lbm ft/s2. Since gc = 1 by definition, the incompressible flow equation is often written as = Co A 2 gcρ ∆P , m
The incompressible flow equation used here is accurate for natural draft devices where the resulting air velocity is much lower than the speed of sound. However, it is not a good assumption for fuel flow from a fuel orifice as fuel pressures are normally 25 psig or so and the exit velocity is sonic. For these cases, we must make use of a more inclusive form of the Bernoulli equation.
(4.12)
when using British units. This would not be necessary except that engineers before Newton did not recognize the difference between mass and force and developed an inconsistent set of units. SI units, or any other consistent set of units, do not require the use of gc . Example 4.2: Calculation of Mass Flow Through Hole in a Furnace Problem Statement: A furnace has a leak equating to a total flow area of 4 in2 at an elevation in the heater corresponding to −0.25 in. w.c. of draft. Presuming an orifice coefficient of 0.80, what is the mass flow of air that this represents? Presume that the ambient air density is 0.075 lbm/ft3. = Co A 2gc ρ| ∆P | with Solution: We use the equation m the following information: Co = 0.80 A = 4 in2 ρ = 0.075 lbm/ft3 ∆P = −0.25″ w.c.
4.3.4 Compressible Flow through an Orifice We shall find it convenient to define a pressure ratio (ro) that is a function of the absolute pressure upstream of the orifice (Po) and the absolute pressure surrounding the orifice into which the fluid is exhausting (P∞):
ro =
Po . P∞
(4.13)
As the pressure increases, the velocity through the rifice also increases until it reaches a natural limit o known as the speed of sound. This limit occurs at a critical pressure ratio (rc). This critical ratio is given by the following equation:
( )
k+1 rc = 2
k k −1
.
Considering compressible flow through an orifice the following equation gives the mass flow. An exact derivation is available at http://www.combustion-modeling.com
( )
1− k 2k k r 1 − o M k−1 2 C A P β o o o 2 RT roγ − β 4 m = 2k M k+1 Co Ao Poβ 2 RT k + 1 k −k 1 − β4 2
( )
( )
( )
Therefore, we have
lbm lbm ft = (0.80 )( 4 )[in2 ] 2(32.174 ) m 0.075 3 0.25[in w.c.]. ft lbf s 2
We also note that there are 144 in2 per 1 ft 2 and 27.69 in w.c. per 1 lbf/in2. With these conversions we have lbf 1 2 2 ft lbm 1 in2 ft lbm [ ] in 0.075 0.25[in w.c.] = (0.80 )( 4 )[in2 ] 144 2 m 2(32.174 ) ft 3 27.69[in w.c.] 144 [in2 ] ft lbf s 2 = 0.056 lbm/s (or 200.4 lbm/h).
© 2011 by Taylor and Francis Group, LLC
(4.14)
ro < rc , (4.15) ro ≥ rc
81
Fluid Flow
where β is the ratio of the orifice diameter to the upstream pipe or duct diameter, Ao is the area of the orifice, and T is the upstream temperature. For monoatomic gases such as helium or argon, k ≈ 5/3. For diatomic gases such as air, oxygen, or nitrogen, k ≈ 7/5. For polyatomic gases k approaches 1 as the number of atoms increases to infinity; however, for common polyatomic gases having 3 to 5 atoms, 9/7 < k < 8/6 and an average value of 1.3 is commonly used. Although k is not a strong function of temperature (much weaker than either Cp or Cv alone) at higher temperatures more energy states become available and this also tends to drive k lower. However, for any particular gas or temperature, exact values of k are available.
Example 4.3: Calculation of Pressure Drop Through a Pipe
4.3.5 Pressure Drop in Pipes and Fittings
lbm ft 50 3 50 2 100 [ ft ] ft s ∆P = 0.023 + 4.5 2 0.5[ ft ] lbm − ft 2(32.174 ) (144) in2 2 ft − lbf s
L ∆P = f + D
ρv 2
∑ K 2 g
,
What is the pressure drop in psi? Solution: Applying the pressure drop equation gives the following:
(4.16)
= 122.8 lbf / in2. Figure 4.2 gives the Moody Friction Factor Chart. This chart allows f to be read as a function of pipe roughness, ε, divided by pipe diameter (ε/D, the so called relative roughness) and the Reynolds number (Re = Dvρ/µ), where µ is the viscosity of the fluid. One can also solve the Colebrook equation iteratively to find f:
c
where L is the length of the duct, D is its diameter, and ∑K is the sum of correction factors for bends and fittings. 0.1 0.09 0.08 0.07 0.06
ρ = 50 lbm/ft3 f = 0.023 L = 100 ft D = 0.5 ft ΣK = 4.5
2
Although velocity through the final gas orifice is often sonic, for industrial combustion the usual practice is to make the upstream diameter of the duct or pipe many times larger than the orifice and to keep the flow much below sonic. Also, liquids, even under high pressure, may be generally regarded as incompressible. Therefore, we may treat pressure drop in pipes and fittings as incompressible flow. A friction factor ( f ) is used to correct for friction. The general relation is
Problem Statement: Consider a liquid hydrocarbon mixture flowing at 50 ft/s and having the following properties.
(
f = 64/Re
z/D 0.05 0.04 0.03 0.02 0.015 0.01
0.05 Friction factor f
)
εD 2.51 1 . = −2log10 + 3.7 Re f f
0.04
0.006 0.004
0.03
0.002 0.001 0.000 6 0.000 4 0.000 2 0.000 1 0.000 05
0.02
0 (smooth)
0.01 0.009
0.000 005
1000
Figure 4.2 U-tube manometer.
© 2011 by Taylor and Francis Group, LLC
2 34 68
104
2 34 68
2 34 68
106 105 Reynolds number
2 34 68
107
2 3 4 6 8 0.000 01
108
(4.17)
82
Industrial Combustion Testing
The Haaland equation provides a useful approximation that does not require iteration:
( )
1 εD = −1.8log10 f 3. 7
1.11
6. 9 + . Re
Clear tube Scale (in. WC)
(4.18)
4 3
Example 4.4: Calculation of Friction Factor Problem Statement: Use the Colebrook and Haaland equations to calculate the friction factor with ε/D = 0.008 and Re = 8 × 105.
2 1 0 1
Pressure differential
2 3
Solution: We begin with the Haaland equation, as it does not require any iteration. This gives
(
1 0.0008 = −1.8log10 f 3. 7
)
1.11
+
4
6. 9 = 7.18, and 2 × 106 Liquid-filled
f = 0.0194. The Colebrook equation yields
(
Figure 4.3 U-tube manometer.
)
εD 2.51 1 = 0.0188. = −2log10 + 3.7 Re f f
The specific weight of water and mercury is 62.4 lbf/ft3 and 847lbf/ft3, respectively.
Thus, the Haaland equation is a bit more than 3% too high. This is well within the accuracy of the calculation; indeed, the relative roughness is never known with certainty and can change with use.
Example: A U-tube manometer, filled with mercury, is connected to a pressurized vessel. The difference in the height of mercury reads 60-inch. (a) What is the pressure in units of pounds per square inch? (b) What would the difference in height be if the manometer was filled with water? (a) Pd = γh = 847
4.4 Pressure Measurement Techniques
lbf 1ft 1ft 2 × 60 in × = 29.4 psi, ft 3 12in 144in 2
Pd
4.4.1 Manometer One of the simplest and most useful pressure measuring devices is the manometer. There are several variations of the manometer. In this section we will discuss two types: U-tube and inclined manometer. 4.4.1.1 U-Tube Manometer In its most basic form, the manometer takes the form of a U-shaped tube that is partly filled with some liquid, such as water, oil, or mercury, as shown in Figure 4.3. The differential pressure in a U-tube manometer can be expressed as
Oil-filled Water-filled
Pd = γh,
where Pd ≡ differential pressure (lbf/ft2), γ ≡ specific weight of the fluid (lbf/ft3), h ≡ height of the liquid (ft).
© 2011 by Taylor and Francis Group, LLC
(4.19)
(b)
hmercury γ mercury = Pd hwater γ water hwater = hmercury
γ mercury 847 = 60 in × = 814.4 in. γ water 64.2
This example illustrates that mercury-filled, U-tube manometers are reasonable to use in pressure ranges as high as 30 psig; at this pressure, a 5-foot (60 inches) manometer is required. A water-filled manometer, however, is not reasonable at this pressure because it would require a U-tube manometer over 68-feet (815 inches) long. 4.4.1.2 Inclined Manometer In combustion applications it is often important to be able to measure very small pressure levels accurately.
83
Fluid Flow
For example, it is common to be able to measure the draft inside of a heater with an accuracy of 1/100th of an inch of water column (0.00036 psi). With U-tube monometers, it is not possible to read pressures this accurate; instead, inclined manometers are commonly used. Inclined manometers essentially expand the scale of a U-tube manometer by orienting it at an angle, relative to the horizontal, as illustrated in Figure 4.4. Although the actual vertical rise of the liquid in the U-tube and inclined manometer are identical, the scale on the inclined manometer is spread out over a much larger horizontal distance; this expanded scale increases the instruments accuracy. Inclined manometers consist of a clear tube positioned at a slight angle relative to the horizontal as shown in Figure 4.5. The tube is typically filled with red oil with a specific gravity of approximately 0.8. Located at the top of the manometer are two pressure taps: a high pressure side and low pressure side. The low pressure side tap contains the oil reservoir. For accurate measurement it is important that the manometer is level with the horizontal. Typically inclined manometers are designed with a bubble level to assist in accurate leveling.
0
Expanded scale 0.2
1 cm
0.4
0.6
0.8
1.0
1 cm
Inclined manometer
U-tube manometer
The units typically used in the combustion industry to measure furnace draft are expressed in inches or millimeters of water column (in WC or mm WC). A column of water 27.68 inches high will create a hydrostatic pressure of 1 pound per square inch. 4.4.2 Bourdon Tube Gauge In 1849 Eugene Bourdon, a French watchmaker and engineer, patented an instrument used to measure pressure. This instrument is called the Bourdon tube gauge and is one of the most common instruments used in industrial plants. The purpose of this section is to discuss the design, failure, and calibration of these gauges. 4.4.2.1 Design of the Bourdon Tube Gauge Bourdon tube gauges are available in a variety of sizes (see Figure 4.6). The diameter of the dial faces can vary from less than 1 inch (2.5 centimeters) to as large as 3 feet (0.9 meter), with the largest typically used for extreme accuracy [1]. Bourdon tube gauges are also available in a variety of pressure ranges: from 5 psig full-scale to as much as 10,000 psig full-scale [2]. The Bourdon gauge consists of a flattened, thin-walled metal tube bent into the form of an arc or C shape as illustrated in Figures 4.7 and 4.8 [3]. One end of the tube is fixed and securely sealed and bonded to a threaded connection called a socket. The other end is tightly sealed and free to move. At atmospheric pressure (zero gauge pressure) the tube is undeflected and for this condition the gauge pointer is calibrated to read zero pressure [4]. However, when pressure is applied into the fixed end of the tube, the effect of the forces tends
Figure 4.4 Showing how an inclined manometer expands the measurements scale. Low pressure-side tap
High pressure-side tap
Clear tube
Bubble level
Oil reservoir
Figure 4.5 Basic components of a Bourdon pressure gauge.
© 2011 by Taylor and Francis Group, LLC
Figure 4.6 Basic components of a Bourdon pressure gauge.
84
Industrial Combustion Testing
Figure 4.7 Internal view of the Bourdon pressure gauge.
Tube movement Bourdon tube
and ½-inch National Pipe Thread (NPT) tapered pipe threads. For dial sizes smaller than 4.5 inches, a ¼-inch NPT is common; however, in some cases a 1/8-inch NPT will be used. For pressure-gauge dial sizes smaller than 2 inches, a 1/8 NPT is common. The movement mechanism can be made from a variety of material such as nickel–silver, brass, or stainless steel; brass is a very common material. It is important that the material used allows for a friction-free assembly. The movement mechanism is protected within an enclosure commonly made of brass, aluminum, steel, or plastic. Most lenses, protecting the indicating needle and dial are made of plastic instead of glass for safety reasons due to breakage. 4.4.2.2 Common Failure Mechanisms
Spiral spring
Gear and linkage system (movement)
Indicating needle
Socket Fluid in
Figure 4.8 Basic components of a Bourdon tube pressure gauge. (Adapted from Hydraulic Energy Transmission and Control, Hill Learning Systems, 2001–2007, http://www.hetacfluidpower.com/screen_shot_gallery. htm)
to uncoil or straighten the tube causing the end to move in a slightly curved path. The end of the Bourdon tube does not move a great distance within its pressure range: typically 0.125 to 0.25 inch (0.31 to 0.63 centimeter) [5]. Using a gear and linkage system (called the movement) the motion of the tube is magnified and converted into rotary movement of the indicating needle. The indicating needle rotates around a circular dial calibrated for pressure [6]. The indicating needle is commonly fixed to a spiral spring that tightens up as more pressure is applied. The purpose of the spiral spring is to assist the tube and pointer to quickly respond to a drop in pressure after reading some higher value. The most important part of the gauge is the Bourdon tube. Bourdon tubes are made of many materials: beryllium copper, phosphor bronze, and various alloys of steel and stainless steel [7]. Beryllium copper is typically used for high pressure applications. Most gauges in air, light oil, or water applications use phosphor bronze. Stainless steel alloys usually add cost to the gauge if specific corrosion resistance is not required. The socket is usually made of brass, steel, or stainless steel. The most common socket size is usually ¼-inch
© 2011 by Taylor and Francis Group, LLC
Since a typical plant will use many pressure gauges, they do not always receive the proper maintenance causing them to be unreliable. Three common reasons for gauge failure are harsh environments, vibrations, and water condensation. Harsh environments such as corrosive and dirty fluids can cause a pressure gauge to fail. Bourdon tubes can be made from a variety of materials and are most often manufactured from 316 stainless steel, phosphor bronze, Monel, and Inconel [8]. The selection of material depends on the fluid for the application. If the application consists of a highly corrosive fluid then a diaphragm seal can be used with the gauge; the seal prevents the fluid from coming in contact with the Bourdon tube. Diaphragm seals not only protect a pressure gauge from corrosion attack, it can also prevent dirt and scale from clogging the tube. The diaphragm seal will typically reduce the accuracy of the gauge. If a gauge is used in a vibrating environment, it can cause excessive motion of the movement mechanism causing it to wear and fail prematurely. Vibrating environments also makes it difficult to read the pressure gauge accurately because the indicating needle oscillates. To help prevent this problem, pressure gauges are often filled with a dampening fluid such as glycerin or a silicon fluid. The fluid helps cushion the tube and movement mechanism against damage from sudden impact and vibration. Many industrial plants have adopted liquid-filled gauges as a standard. Excessive motion of the movement mechanism can also occur if rapid pressure changes occur within a process. Pressure snubbers are commonly used to protect pressure gauges from the system pressure shock and surges and can prolong the life of a gauge. The pressure snubber is placed inline between the gauge and the process as shown in Figure 4.9 [9] and are typically designed with an orifice or a porous metal disk to help dampen pressure fluctuations.
85
Fluid Flow
Female pipe thread
Standard masses
Platform
Oil reservoir Displacer pump
Test gauge
Piston and cylinder
Check valve
Pressure snubber
Male pipe thread Process
Oil pressure
Orifice or porous metal disk
Figure 4.9 Pressure snubber. (Adapted from O’Keefe Controls Co., Trumbull, CT, 2003, http://www.okcc.com/PDF/Pressure%20Snubber%20pg.42. pdf)
Figure 4.11 Basic components of a deadweight tester. (Adapted from Miller, R. W., Flow Measurement Engineering Handbook, New York: McGraw-Hill, Inc., 1989.)
enough pressure to raise the weights, the force exerted by the oil pressure over the piston area is balanced by the force of the weights. The pressure is calculated as follows:
Pgauge = F/Apiston.
(4.20)
When using a deadweight tester, it is important to apply corrections for local gravity, elevation, buoyancy, or thermal expansion of the piston, since these errors may be significant [10]. Other sources of error include weights that are dirty, corroded, or chipped or using a tester that is not properly leveled [11]. 4.4.2.4 Selection Figure 4.10 An oil-type deadweight tester.
4.4.2.3 Calibration of Pressure Gauges To insure accuracy, it is customary to calibrate pressure gauges periodically. Depending on their use and importance to a given process, gauges might be calibrated monthly, quarterly, yearly, or every several years. Deadweight testers are the primary standard for accurate pressure calibration. A typical oil-type deadweight tester is shown in Figure 4.10. The basic components of an oil-type deadweight tester are shown in Figure 4.11. These testers consist of an accurately honed piston of a known cross-sectional area inserted into a cylinder. Known weights (standard masses) are placed on a freely rotating platform attached to the top of the piston; the platform is designed to freely rotate to reduce friction between the piston and cylinder. When the oil pump supplies
© 2011 by Taylor and Francis Group, LLC
When selecting a pressure gauge, it is recommended to use one that has a range of twice the normal working pressure. The maximum operating pressure, in all cases, should not exceed 75% of the maximum range of the gauge. If gauges are used in applications where pulsations are encountered, the pressure should be limited to 67% of the maximum range of the gauge. The ambient temperature range for most standard dry or silicone filled gauges is about −40°F to + 140°F (−40°C to 60°C). The ambient temperature range for glycerine filled gauges is −4°F to + 140°F (−20°C to 60°C). The error caused by temperature change is + /−0.3% per 18°F rise or fall, respectively, based on the temperature of the gauge. 4.4.2.5 Installation Bourdon gauges are commonly used to measure the static pressure of a fluid flowing in a pipe. To reduce the potential for error, it is important that static pressure taps are
86
Industrial Combustion Testing
designed correctly. Below are several important design considerations for static pressure taps: • The hole should be drilled perpendicular to the pipe. • The hole should be free of burrs. • The edge of the hole should be square with the inner pipe surface with no rounded corners or no concave or convex surface surrounding the hole. • The drilling of the hole should be done after all fittings for attaching the pressure gauge have been welded to the pipe. Ferron [12] states that countless static pressure measurements are in error because of poor pressure tap construction. Brunkalla [13] suggests a ¼ inch (6.4 mm) diameter hole for pipe sizes greater than 2 inches (50 mm). Departure from the recommended hole size, inclination, or edge condition results in errors of −0.5 to + 1.1 percent of the velocity or dynamic pressure. [10]
Figure 4.12 Orifice plate. P1
Differential pressure (1c p) P1–P2
P2
Static pressure profile
4.5 Differential Producing Flow Meters The differential producing flow meter is a device used to create varying static pressure, within a flow stream that can be used to determine the flow rate of the fluid. These devises have been used for over 1000 years [10]. Today, differential flow meters are the most common and reliable flow meters used in industry. This section will discuss three common types of differential producing flow meters: Orifice meter, Venturi meter, and Nozzle meter. 4.5.1 Orifice Meter Orifice meters have been in commercial use since the early 1900s and are the most common differential producing flow meter used today [14,15]. The orifice meter consists of a flat, thin plate with a circular hole machined into the center of it as shown in Figure 4.12 and is held in place with a holding device such as a pipe flange. When a fluid flows through an orifice plate, the static pressure within the pipe varies as illustrated in Figure 4.13. Notice that just upstream of the orifice plate the static pressure reaches a maximum value. When the fluid passes through the bore of the plate it accelerates. As the fluid jet exits the other side of the plate it continues to accelerate and decrease in cross-sectional area; hence, the minimum flow area is actually smaller than the area of the orifice. The location where the jet reaches its minimum cross-sectional area is referred to as the vena contracta and is where the velocity is the highest and the
© 2011 by Taylor and Francis Group, LLC
Flow
Orifice plate
Figure 4.13 Static pressure trend across an orifice plate. (Adapted from Smith Metering, Inc., “Fundamentals of Orifice Metering,” http://www. afms.org/Docs/gas/Fundamenatls_of_Orifice.pdf)
static pressure the lowest. As the fluid continues to flow downstream, it de-accelerates, where some, but not all of the static pressure is recovered. Pressure taps are required on each side of the orifice plate to allow the measurement of the pressure drop across the plate when the fluid is flowing. Knowing the differential pressure, the ratio of the orifice plate bore to pipe inside diameter (beta ratio) and fluid density, the flow rate can be calculated. The differential pressure depends on the location of the pressure taps. Pressure taps are located in four general arrangements as illustrated in Figure 4.14; flange taps, D and D/2 taps, corner taps, and pipe taps. Flange taps are typically placed on the pipe flange and are located 1 inch upstream and 1 inch downstream from the face of the plate to centerline of the tap. The D and D/2 taps are located one pipe inner diameter upstream and one-half downstream from the face to the flange. Corner taps are located at the face of the orifice plate. Pipe taps are located 2.5 pipe inner diameters upstream
87
Fluid Flow
1 in 1 in
Flow direction
D
Tap d
Tap
±0.04 in for d/D <= 0.6 ±0.02 in for d/D <= 0.6
Taps
Flow direciton
D
D±0.1D
Flange taps
d
±0.02D for d/D <= 0.6 ±0.01D for d/D > 0.6
0.5D
D and D/2 taps
Taps
2.5D Flow direction
D
Flow direction
Corner taps
Tap
8D
Tap
D
Pipe taps
Figure 4.14 Common pressure-tap arrangements.
and 8 diameters downstream from the face of the orifice plate; these arrangements measure the pressure drop at the point of full pressure recovery. When fluid flows through elbows, tees, and valves a swirling flow with an uneven velocity profile develops. For accurate flow measurement, the flow upstream of a differential pressure meter should be free of these disturbances. In order to eliminate these disturbances, a long length of straight pipe is required. For example, swirl flow inside a pipe can require 50 to 100 diameters of straight pipe to eliminate the spin [16]. Swirl-free conditions are defined as where the angle of the swirl is less than 2° [16]. An acceptable velocity profile condition exits when the velocity at each point within the pipe cross-section agrees to within ± 5% of the velocity of a very long, straight length (over 100 diameters) of similar pipe. Flow conditioners are commonly placed upstream of the orifice metering device to help straighten the flow profile and reduce swirl, thus allowing shorter lengths of upstream pipe. In general, there are three standard types of flow conditioners: tube bundle conditioner, Zanker straightener, and Sprenkle straightener [16]. The tube bundle conditioner consists of a number of parallel, thin-walled tubes connected together. The tube bundles do a good job of eliminating swirl, but do a poor
© 2011 by Taylor and Francis Group, LLC
job of straightening the velocity profile. The Zanker straightener, consists of a plate with holes of a certain size located just upstream of a number of parallel, thinwalled square channels connected together; one channel for each hole. The Sprenkle straightener consists of the three perforated plates, in series, spaced one pipe diameter apart. The plates are held together by studs or bolts around the perimeter of the plates. Research and development efforts in this area are ongoing in order to gain better insight into the best design and location of flow conditioners upstream of a differential metering device. The equation for the mass flow rate through an orifice plate starts with the Bernoulli equation:
Po Vo2 P V2 + + zo = 1 + 1 + z1 , ρg 2 g ρg 2 g
(4.21)
where the subscripts “o” represent the location at the plane of the orifice bore and “1” represents the location upstream of the orifice plate. Setting zo = z1, ΔP = P1 − Po, and solving Equation 4.21 for Vo, gives 1
2 × ∆P 2 Vo = V12 + . ρ
(4.22a)
88
Industrial Combustion Testing
Now, V1 is eliminated by means of the continuity equation V1 A1 = VoAo; and solving for Vo gives
Vo =
2 × ∆P . Ao 2 ρ 1 − A1
(4.22b)
= ρVo Ao . m
(4.23a)
Substituting Equation 4.22b into Equation 4.23a, letting β = do/d1, and rearranging gives = Ao m
2 × ∆P × ρ . 1 − β4
(4.23b)
Equation 4.23b represents the theoretical or ideal mass flow rate through the orifice; that is, an incompressible, inviscid flow. To determine the actual mass flow rate, this equation needs to be multiplied by two correction terms: gas expansion factor (Y1), and discharge coefficient (C). When a gas flows through an orifice plate it takes a pressure drop causing the gas to expand and reduce the density. To correct for the density change the equations are multiplied by a gas expansion factor (Y1). The gas expansion factor is based on laboratory tests and is defined as follows:
Y1 = 1 − ( 0.41 + 0.35β 4 )
∆P , k P1
(4.24)
where k is the isentropic coefficient. The discharge coefficient corrects the theoretical equation for the effects of the approaching velocity profile, the assumption of no energy loss between the pressure taps and pressure tap location and is defined as
C=
true flow rate . theoretical flow rate
(4.25)
The discharge coefficient is determined from laboratory experiments and varies with tap arrangement and diameter of pipe and orifice plate. For example, the discharge coefficient, C, for an orifice plate with corner taps is
C = C∞ +
© 2011 by Taylor and Francis Group, LLC
b RDn
RD =
The mass flow rate can be written as
where C∞ = 0.5959 + 0.0312β2.1−0.184β8 b = 91.71β2.5 n = 0.75 RD ≡ Reynolds number based on the pipe inner diameter and upstream velocity
(4.26)
V1d . υ
For a complete listing of discharge coefficient equations see Miller [10]. The final form of the actual mass flow rate through an orifice is the theoretical mass flow rate multiplied by the gas expansion factor and orifice discharge coefficient and is written as
= Ao × Y1 × C × m
2 × ∆ P × ρ1 . (1 − β 4 )
(4.27)
Notice that the discharge coefficient, C, is a function of the upstream velocity in the pipe (V1). Since the upstream velocity is not known, the solution to this equation requires an iterative procedure. For applications where the fluid temperature is very high or low, it is important to include another correction factor. Pipe and orifice plate bore diameters are typically provided by the manufacturer and are based on room temperature. As the temperature of the orifice plate and pipe vary from room temperature, they expand and contract. To correct for the expansion and contraction of the material, a thermal-expansion factor is applied to the dimensions; these new dimensions are then used in the calculation. The thermal-expansion correction factor for a temperature relative to 68°F is
Fa = 1 +
2 (α PE − β 4α P )(TF − 68) , 1 − β4
(4.28)
where αPE and αF are the thermal-expansion factor of the orifice plate and pipe, respectively, Tf is the material temperature in °F, and β is the diameter ratio of the orifice bore to the inner diameter of the pipe. The thermal expansion factor for several materials is given in Table 4.1 [17]. Table 4.1 Thermal-Expansion Factors Material Austenitic stainless steel Martensitic stainless steel Aluminum Copper
(10−6 in/in-°F) 10.2 to 9 6.5 to 5.5 16.3 to 14.4 9.8 to 7.7
89
Fluid Flow
Pressure sensing ports
Throat section
∆P
Figure 4.15 Typical Venturi meter. (Adapted from Roberson, J. A., and Crowe, T., Engineering Fluid Mechanics, p. 45, Dallas, TX: Houghton Mifflin Co., 1980.)
4.5.2 Venturi Meter The orifice meter is a simple and accurate device for measuring flow rates; however, the pressure drop for an orifice meter can be quite large [4]. A meter that operates on the same principle as the orifice meter, but with a much smaller pressure drop, is the Venturi meter. The Venturi meter is designed with gradual upstream converging cone with an angle of 15° to 20° as illustrated in Figure 4.15. Downstream of the converging cone the fluid enters into a throat section. The flow rate is determined based on the pressure difference between the upstream side of the converging cone and the throat section. After leaving the throat, the fluid enters into a diverging section designed with an angle of 5° to 7°. Due to the gradual upstream contraction there is no vena contracta (like for the flow through an orifice plate) resulting in a low pressure drop; again, this makes the Venturi meter suitable where only small pressure heads are available. The equation for the mass flow rate through a Venturi meter is the same as for the orifice meter, Equation 4.27. The discharge coefficient for a Venturi meter is larger and typically varies from 0.95 to 0.975; for the orifice meter the discharge coefficient typically varies from 0.6 to 0.7. For more information on the design and discharge coefficients of a Venturi meter see ASME MFC-3M-1989 [16].
The rotational speed of the rotor, or turbine, increases linearly with flow velocity: within ± 0.5% over a wide flow range from 10:1 to 20:1 [10]. The turbine speed is typically measured by detecting the pulse of each blade electrically, mechanically, or optically; each pulse represents a certain volume of fluid. The flow rate of the fluid is determined by integrating the total number of pulses over a period of time. The accuracy of these meters is better than 1% over a wide range of flow rates [4]. Accuracy is compromised, however, with blade wear, bearing friction due to wear, and when a vapor enters the line for liquid flows [10]. 4.5.4 Vortex Flow Meter When a fluid flows past a bluff body, the wake downstream will form rows of vortices that shed continuously from each side of the body as illustated in Figure 4.16. These repeating patterns of swirling vorticies are referred to as Karman vortex streets: named after the fluid dynamicist Theodore von Karman. Vortex shedding is a common flow phenomenon that causes car antennas to vibrate at certain wind speeds and also lead to the collapse of the famous Tacoma Narrows Bridge in 1940. Each time a vortex is shed from the bluff body it creates a sideways force causing the body to vibrate. The frequency of vibration is linearly proportional to the velocity of the approching fluid stream and is independent of the fluid density. The frequency of vortex shedding or vibration of the bluff body is determined as follows: f=
© 2011 by Taylor and Francis Group, LLC
(4.29)
where St is the Strouhal number (dimensionless), Vaverage is the average upstream velocity, and dbody is diameter of the bluff body. For flow over a cylindrical body, the Strouhal number is constant for Reynolds numbers from 10,000 to 1,000,000: based on the diameter of the bluff body. As the Reynolds number increases, the wake becomes more complex and turbulent, but the alternate
4.5.3 Turbine Flow Meter Turbine flow meters, when installed and calibrated properly, are capable of providing the highest accuracies attainable by any currently available flow sensor for both liquid and gas volumetric flow measurement [18]. The turbine meter consists of a cylindrical housing containing a balanced, free-spinning rotor designed with a series of curved blades that mounts on the centerline of the pipe. The number of blades typically varies from about 6 to 20 or more depending on the design with a pitch angle of 30° to 45° [18]. As fluid passes over the blades, it forces the rotor to spin.
St × Vaverage , dbody
Karman vortex streets
Bluff body Cycle Figure 4.16 Pair of vortices shed downstream of a bluff body.
90
Industrial Combustion Testing
shedding can still be detected at a Reynolds number of 10,000,000 [19]. Knowing the Strouhal number, frequency of vibration, and the diameter of the bluff body, the average velocity can be determined. Vortex flow meters are designed to partially obstruct the fluid stream using a bluff body in order to create vortex shedding. The frequency of the vortex shedding is measured by various methods and converted to an average velocity. Today, most Vortex meters operate accurately over a wide range of flows and are available for line sizes from ½ inch to 16 inches. The main advantage of this meter is that it has no moving parts, but it does create pressure drop comparable to other obstruction-type meters [10]: turbine meter, orifice plate, and so forth. 4.5.5 Magnetic Flow Meter All the flow meters mentioned so far require that some sort of obstruction, such as a turbine or orifice, be placed within the flow stream. Magnetic flow meters, however, do not obstruct the flow of the fluid. The magnetic flow meter operates on the basic principle of magnetic induction; that is, when a conducting material moves through a magnetic field, it produces an electromagnetic force as illustrated in Figure 4.17. It is interesting to note that the basic principle of the magnetic flow meter was investigated by Faraday in 1832; however, it was not practically used until a century later [10]. When a current is applied to the coil, a magnetic field is produced that is at right angles to the pipe. As the conductive liquid flows through the pipe, an electrical voltage is induced and measured by two electrodes mounted flush to the pipe wall. This induced voltage is proportional to the average velocity of flow and therefore to the volume flow rate. In order for a magnetic flow meter to
Coil
work, the fluid must be conductive; therefore, they are not suitable for gas flow [20]. It is important that the strength of the magnetic field remain constant; this is accomplished with a microprocessor built into the transmitter controls. The microprocessor also detects the voltage of the electrodes and converters the signal to a fluid flow rate. The main advantages of the magnetic flow meter are that the output signal varies linearly with the flow rate and that the meter does not obstruct the flow. The major disadvantages are its relative high cost and that it is not suited for gas flow [4]. 4.5.6 Ultrasonic Flow Meter There are two types of ultrasonic flow meters: time-offlight and Doppler. The time-of-flight meter sends pulses of high frequency sound waves diagonally across the pipe as illustrated in Figure 4.18 [21]. The time required for the sound wave to reach the opposite wall depends on whether it is moving with the flow or against the flow; if the wave is moving with the flow it will travel faster than the wave moving against the flow. The time difference is a measure of the fluid flow rate. The ultra sonic flow meter illustrated in Figure 4.18 is referred to as a single-path meter because only one beam path crosses the pipe. Some ultrasonic flow meters send sound waves along several paths to improve accuracy as illustrated in Figure 4.19. The Doppler flow meter sends out sound waves similar to the time-of-flight meter, but the waves are reflected back to a detector by small particles or bubbles moving with the fluid. The sound wave reflected back from the fluid will have a slightly lower frequency than the transmitted sound wave due to the Doppler effect. The difference between transmitted frequency and the reflected frequency is used to determine the flow rate.
Magnetic field Electrode plates
V Voltage applied to produce magnetic field
Flow
Pulses of high frequency sound
Flow of conductive fluid V Electrode voltage Figure 4.17 Principle of magnetic induction. (Adapted from “Precision Measuring PD 340 Flow Transmitter,” http://www.instruments-gauges.co.uk/ pd340.htm)
© 2011 by Taylor and Francis Group, LLC
Figure 4.18 Time-of-flight ultrasonic flow meter: single-path type.
91
Fluid Flow
Flow
Flow
Flow
Figure 4.19 Time-of-flight ultrasonic flow meter: multipath type. (Adapted from Miller, R. W., Flow Measurement Engineering Handbook, New York: McGrawHill, Inc., 1989.)
Commercially available Doppler flow meters requires that the liquid contain at least 100 parts per million of 100 micron or larger suspended particles or bubbles [22]. 4.5.7 Thermal Mass Meter A typical thermal mass meter is shown in Figure 4.20. The operation of these meters is based on the principle that the rate of heat absorbed by a flowing fluid is directly proportional to the mass flow rate [23]. In general there are three types of thermal mass meters: constant temperature, constant power, and energy balance. Energy balance thermal mass meters require one heating element located between two temperature sensors as illustrated in Figure 4.21 [24]. Although several design variations exist, their operation is basically similar. As the fluid flows past the heating element, it absorbs heat. This heat is carried downstream where it is transferred to the downstream temperature sensor. The temperature difference between the upstream and downstream sensor is detected. This output signal is then converted into a mass flow rate. These meters typically have a turn down ratio of 10:1 while the constant temperature and constant power meters have a turndown ratio of 1000:1 and 100:1, respectively.
Figure 4.20 Thermal mass meter.
Temperature sensors Heating element
Flow
4.5.8 Positive Displacement Meter Positive displacement flow meters, or PD meters, are used more than all other flow measurement devices [10]; a typical PD flow meter is shown in Figure 4.22. Millions of PD
© 2011 by Taylor and Francis Group, LLC
Figure 4.21 Energy balance type thermal mass meter. (Adapted from “Flow Measurement,” http://en.wikipedia.org/wiki/Flow_Measurement.)
92
Industrial Combustion Testing
Figure 4.22 Positive displacement meter.
Flow
Oval gear
Flow
Lobe
As the flow rate of the fluid increases, these parts turn or reciprocate proportionally faster and are sensed by mechanical registers or electronic transmitters. The rate of revolution or reciprocation determines the volume of fluid passing through the meter. It is important that these mechanical units have tight tolerances to prevent fluid from passing through the flow meter without being measured; this is referred to as slippage. Due to these tight tolerances, filters are typically installed upstream to prevent dirt from entering the meter; this helps reduce mechanical wear and prevent plugging. Typically, filters are used to remove particles larger then 100 µm as well as gas bubbles from the liquid flow [26]. Mechanical wear can result in substantial flow error; therefore, PD meters should be periodically calibrated. Since PD meters only measure the volume of fluid that passes through, they are rarely used as flow rate meters; however the average flow rate can be determined by measuring the volume of fluid that passes through in a given amount of time. It should also be mentioned that PD meters create a considerable pressure drop that should be considered for any application. 4.5.9 Pitot Tube
Oval gear
Rotating lobe Impeller Flow
Retractable rotor blade
Flow
Rotating impeller
A Pitot tube is an instrument used to measure the velocity of a flowing fluid. The Pitot tube was invented by French engineer Henri Pitot in the early 1700s. Today, it is widely used in industrial applications and is manufactured in a variety of sizes as shown in Figure 4.24.
12-inch ruler
Rotating vane
Figure 4.23 Types of positive displacement meters. (Adapted from Universal Flow Monitors (UFM), “Positive Displacement Flowmeters,” Hazel Park, MI, http://www.flowmeters.com/ufm/index.cfm?task = positive_ displacement)
meters are used daily to meter gasoline, natural gas, and water into our homes. Although there are many different PD meter designs commercially available, they all share the same principle of operation: they measure the volume of fluid by separating the flow into known volumes and then counting, repeatedly, the filling and discharging of volumes passing through the meter over time. The pumping of individual parcels of fluid is accomplished using rotating parts, pistons, or diaphragms that form moving seals between each other and/or the flow meter body [25]. A few of these mechanical units designed with rotating parts is illustrated in Figure 4.23.
© 2011 by Taylor and Francis Group, LLC
Figure 4.24 Various sizes of Pitot-static tubes.
93
Fluid Flow
Static taps around circumference
Streamlines
Streamlines
Stagnation point
Stagnation point
Velocity
Velocity Simple pitot tube
Pitot-static tube
Total pressure Velocity pressure U-tube manometer Figure 4.25 The simple Pitot tube.
The simple Pitot tube consists of a tube bent at a right angle as illustrated in Figure 4.25. One end of the tube is placed directly into a flowing fluid stream, parallel to the flow direction or streamlines. As the fluid impacts the open end of the tube it is brought to rest, with zero velocity, directly in front of the opening of the Pitot tube. The pressure at this point is the total or stagnation pressure. By attaching a pressure sensing device on the opposite end of the Pitot tube one can measure the total pressure in the flowing stream directly. The velocity pressure cannot be determined unless the static pressure is known within the flow stream. Recall that the total pressure is written as Total Pressure = Static Pressure + Velocity Pressure. The static pressure is generally measured using the Pitot-static tube design as shown in Figures 4.26 and 4.27. Pitot-static designs consist of a coaxial tube placed around the outside of the simple Pitot tube. Several pressure ports located around the perimeter of the outside coaxial tube are used to measure the static pressure within the flow stream; these holes are called the static taps. By attaching a pressure sensing device and measuring the differential pressure between the total and static pressure allows one to determine the velocity pressure. Knowing the velocity pressure, the velocity of the fluid at the point of measurement can be determined. This is illustrated in the following example. Example: Pitot-static tube is placed within duct flowing air with a temperature of 200°F. The static pressure within the duct is 1 psig and the atmospheric pressure is 14.3 psia. A U-tube manometer measures a differential pressure of 4 inches of WC between the total pressure
© 2011 by Taylor and Francis Group, LLC
U-tube manometer Figure 4.26 Illustration of the Pitot-static tube. Total pressure side
Static tap
Static pressure side
Total pressure tap or impact hole
Fittings for used mounting
Figure 4.27 Photograph of the Pitot-static tube.
tap and static tap on the Pitot tube. Determine the velocity of the air at the measurement point. First, determine the density of the air flowing within the duct. lb 0.0765 m3 ft
ρair,duct =
×
density of air at standard temperature and pressure (59 ° F and 14.696 psia)
×
59 F + 460 460 200 F+
14.3 psia + 1 psig 14.696 psia correcting for the actual air pressure e inside the duct
×
slugs 1 slug = 0.001944 . ft 3 lbm 32.2
correcting for the actual air converting pounds temperature inside the duct mass to slugs
Next, convert the differential pressure, dP, to units of lbf/ft2. lb f 144 inch 2 × lb f 2 ft 2 dP = 4 inches of WC × inch = 20.81 2 . ft 27.68 inches of WC
94
Industrial Combustion Testing
Finally, the velocity can be calculated as follows: slug-ft lb f 1 s 2 1 2 × 20.81 2 × 2 × dP 2 ft 1 lbf V = = ρ slugs 0.001944 2 ft
1
2 ft = 146.3 . s
The static reading on a Pitot tube is accurate to 0.5% for Mach numbers up to 0.5. For Mach numbers between the values of 0.5 to 0.7 the accuracy increases to 1.5%. Above a Mach number of 0.7 the error can increase as much as 10% due to the formation of shock waves on and around the tip of the probe. Above a Mach number of 1.0 both the total and static readings vary significantly for the actual values [27]. The Mach number in the above example is calculated as follows: Ma =
V c
c ≡speed of sound = ( k × R × T )
(4.30)
1 2
where
Although Pitot-static tubes are not extremely sensitive to their angle with respect to the flow streamline (angle of attack), it is good practice to position them as parallel as possible to the flow stream. Errors as high as 2% can result if the angle of attack varies from 5° to 30°. If the probe is positioned too close to a wall, the fluid can accelerate between the probe and wall decreasing the static pressure on one side. It is recommended that the probe be positioned at least 10 probe diameters away from the wall. If the volume or mass flow rates are required in a round or rectangular ducts then one must take measurements at various locations across the duct. It is important that the Pitot traverse be conducted at least 8 duct diameters downstream or 2 diameters upstream of an obstruction such as fans, elbows, dampers, expansions, and so on. Figure 4.28 shows the locations for a Pitot tube traverse in a round or rectangular duct, based on centroids of equal area, for determining the volume or mass flow rate [28]. After measuring the velocity at each point in the duct using the locations provided in Figure 4.28, the total volume and mass flow rate of the fluid can be determined as follows:
k ≡ratio of specific heat = 1.4 for air veral gas constant univ R= mole weight ft-lb f ft − lb f lbm -mol-R = 53.3 lbm lbm − R 29 lbm − mol
1545.32 =
Q = Aduct
1 n
n
∑V i
i=1
(4.31)
= ρ×Q m where Q ≡ volume flow rate in the duct Aduct ≡ duct cross-sectional area n ≡ total number of points traversed Vi ≡ velocity at each measurement point m ˙ ≡ mass flow rate in the duct.
T = 200 F + 460 = 660 R ft − lb f c = 1.4 × 53.3 lbm − R
4.5.10 Annubar × 660 R 1
slug-ft 2 1 32.2 lbm s 2 = 1259.3 ft × × 1 slug 1 lb f s ft s = 0.116. Ma = ft 1259.3 s 146.3
Since the Mach number in this example is less than 0.7 we would not expect any error from formation of shock waves around the probe.
© 2011 by Taylor and Francis Group, LLC
The annubar is widely used as a flow measurement device in industry. In general, an annubar consists of a hollow tube or strut, with various cross-sectional designs, which spans across a flow stream as illustrated in Figure 4.29. The strut is designed with several holes strategically placed along its length so that the average of the total and static pressure is measured. The difference in the average, total, and static pressure provides the mean velocity across the pipe; the volume flow rate of the fluid is determined by multiplying the mean velocity by the cross-sectional area of the pipe. A single Pitot tube measurement will give a similar reading if it is located at a point in the pipe cross section where the flowing velocity is near the average velocity [29]. The strut can consist of various designs such as a cylinder, diamond, or triangular cross section. The
95
Fluid Flow
CL y
Round ducts # Point/row 3 4 5 6
x
Distance from centerline (CL), x/W or y/H 0.204 0.353 0.457 CL 0.177 0.306 0.395 0.468 0.158 0.274 0.354 0.418 0.474 0.323 0.382 0.433 0.479 0.144 0.25
Example of 6 points/row CL Rectangular ducts # Points/row 4 5 6 7
x
y
Distance from CL, x/W or y/H ±0.25 ±0.375 CL ±0.20 ±0.40 0 ±0.083 ±0.25 ±0.417 ±0.143 ±0.286 ±0.429 0
H
W Example of 4 points/row Figure 4.28 Pitot tube traverse locations for round and rectangular ducts based on centroids of equal areas. (Adapted from Flow Kinetics LLC, “Using a Pitot Static Tube for Velocity and Flow Rate Measurement.,” http://www.flowmeterdirectory.com/flowmeter_artc/flowmeter_artc_02111201.html)
Various strut designs
References
Flow
Flow
Static pressure
Total pressure Figure 4.29 Annubar.
position of the pressure ports is critical to the accuracy of the device. If the ports are not located in the proper location then the average velocity profile will be in error. It is also important that the velocity profile is symmetric; if skewed to one side significant errors can result. Typically, as the pipe or duct size increases, the number of total pressure ports and static ports increases. Annubars are commonly used to measure the flow rate in ducts and large pipes. The advantages of the annubar is that they are easy to install, relatively inexpensive, and do not take a large pressure drop [10]. The disadvantage is their poor performance with dirty fluids.
© 2011 by Taylor and Francis Group, LLC
1. Betts, D. “Pressure Gauge. How Products Are Made.” Gale Research Inc. Encyclopedia.com November 20, 2009, http://www.encyclopedia.com/doc/1G2-2896500079. html 2. Zigrang, D. J. “Elements of Engineering Measurements.” Department of Mechanical Engineering, University of Tulsa, Oklahoma, 1985. 3. Hydraulic Energy Transmission and Control. Hill Learning Systems. 2001–2007. http://www.hetacfluidpower.com/screen_shot_gallery.htm 4. Roberson, J. A., and Crowe, T. Engineering Fluid Mechanics, p. 45. Dallas, TX: Houghton Mifflin Co., 1980. 5. Betts, D. “Pressure Gauge. How Products Are Made.” Gale Research Inc. Encyclopedia.com November 20, 2009, http://www.encyclopedia.com/doc/1G2-2896500079. html 6. Cornforth, J. R. Combustion Engineering and Gas Utilization. London: British Gas plc, 1992. 7. Betts, D. “Pressure Gauge. How Products Are Made.” Gale Research Inc. Encyclopedia.com November 20, 2009, http://www.encyclopedia.com/doc/1G2-2896500079. html 8. Jankura, R. Seven Steps to Pressure Gage Selection (Tech Check). Dresser Instrument: Encyclopedia.com July 1, 2002, http://www.encyclopedia.com/doc/1G1-90331703.html 9. O’Keefe Controls Co., Trumbull, CT, 2003, http://www. okcc.com/PDF/Pressure%20Snubber%20pg.42.pdf
96
10. Miller, R. W. Flow Measurement Engineering Handbook. New York: McGraw-Hill, Inc., 1989. 11. Deadweight Testers. “Improving Deadweight Tester Accuracy.” www.transcat.com 2007, http://www.transcat.com/PDF/DeadweightTesters.pdf 12. Ferron, A. G. “Construction of Static Pressure Taps.” Technical Memo. Alden, MA: Alden Research Lab, September 1986. 13. Brunkalla, R. L. “Effects of Fabrication Technique on the Discharge Coefficient for Throat Tap Nozzles,” ASME Paper 84-JPGC-PTC-10, October 1985. 14. Buckley, B. “Fundamentals of Orifice Metering.” Houston, TX: Daniel Measurement and Control. http:// help.intellisitesuite.com/ASGMT%20White%20Papers/ papers/002.pdf 15. Smith Metering, Inc., “Fundamentals of Orifice Metering,” http://www.afms.org/Docs/gas/Fundamenatls_of_ Orifice.pdf 16. “Measurement of Fluid Flow in Pipes Using Orifice, Nozzle, and Venturi,” ASME MFC-3M-1989, The American Society of Mechanical Engineers, 345 East 47th Street, New York. 17. “Tool Design,” http://www.wisetool.com/designation/ te.htm 18. Wadlow, D. “Turbine and Vane Flowmeters.” in The Measurement, Instrumentation and Sensors Handbook, edited by J. G. Webster. Boca Raton, FL: CRC Press, 1998. http://www.sensors-research.com/articles/turbines. htm
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
19. White, F. M. Viscous Fluid Flow, p. 10. St. Louis, MO: McGraw-Hill, 1991. 20. “Precision Measuring PD 340 Flow Transmitter,” http:// www.instruments-gauges.co.uk/pd340.htm 21. “Why and How to Measure Flare Gas, Roxar,” http:// www.flowmeterdirectory.com/flowmeter_artc/flowmeter_artc_02021401.html 22. “Introduction to Ultrasonic Doppler Flow Meters,” http://www.omega.com/prodinfo/ultrasonicflowmeters. html 23. “A Flow Measurement Primer: Thermal Mass Flow Meters, Cross Instrumentation,” http://www.crossinstrumentation.com/ga/mfg/Flow/tmf_meters.htm 24. “Flow Measurement,” http://en.wikipedia.org/wiki/ Flow_measurement 25. Universal Flow Monitors (UFM). “Positive Displacement Flowmeters.” Hazel Park, MI. http://www.flowmeters. com/ufm/index.cfm?task = positive_displacement 26. eFunda. “Positive Displacement Flowmeters.” 2009. http://www.efunda.com/designstandards/sensors/ flowmeters/flowmeter_pd.cfm 27. United Sensor. “Pitot-Static Pressure Probes: For Measuring Total and Static Pressures of a Moving Fluid,” Bulletin 1, 6–83 Amherst, NH. www.unitedsensorcorp.com 28. Flow Kinetics LLC. “Using a Pitot Static Tube for Velocity and Flow Rate Measurement.” http://www.flowmeterdirectory.com/flowmeter_artc/flowmeter_artc_02111201. html 29. Annubar. 2009. http://en.wikipedia.org/wiki/Annubar
5 Temperature Charles E. Baukal, Jr. Contents 5.1 Introduction...................................................................................................................................................................... 97 5.2 Temperature Measurement Examples.......................................................................................................................... 98 5.3 Thermocouples................................................................................................................................................................. 98 5.4 Gas Temperature............................................................................................................................................................ 101 5.4.1 Suction Pyrometer............................................................................................................................................. 101 5.4.2 Optical Techniques............................................................................................................................................ 102 5.4.2.1 Light Scattering................................................................................................................................... 102 5.4.2.2 Emission............................................................................................................................................... 102 5.4.3 Fine Wire Thermocouples................................................................................................................................. 103 5.4.3.1 Coatings................................................................................................................................................ 103 5.4.3.2 Corrections........................................................................................................................................... 104 5.4.3.3 Example Uses....................................................................................................................................... 107 5.4.4 Line Reversal...................................................................................................................................................... 107 5.4.5 Acoustic Pyrometry........................................................................................................................................... 108 5.5 Surface Temperature..................................................................................................................................................... 108 5.5.1 Embedded Thermocouple................................................................................................................................ 108 5.5.2 Infrared Detectors.............................................................................................................................................. 109 5.5.3 Other Techniques................................................................................................................................................111 5.6 Future...............................................................................................................................................................................111 References................................................................................................................................................................................. 112
5.1 Introduction Temperature is a particularly important variable in industrial combustion applications because it directly or indirectly affects a number of other important variables. The product temperature is often a critical parameter in most processes. While there is usually a minimum temperature that must be reached for adequate processing, there may also be a maximum temperature above which product quality is reduced. Higher than necessary product temperatures not only increase fuel costs, but they may also increase cooling costs after the product exits the combustion process. Temperature affects the heat transfer in a furnace [1]. Thermal NOx emissions are exponentially dependent on flame temperatures [2]. Combustion chemistry is very complicated and dependent on temperature. High exhaust gas temperatures mean reduced thermal efficiency [3].
Low combustion chamber temperatures may be safety issues for some technologies like flameless combustion (see Chapter 23). Low combustion chamber temperatures during startup can significantly impact pollutant emissions [4]. Excessively high combustion chamber temperatures could cause damage to the refractory lining and to equipment inside the chamber such as process tubes. Figure 5.1 shows impinging flames on the convection tubes in a process heater. Prolonged impingement overheats the tubes and may eventually damage these tubes, possibly leading to a rupture. Figure 5.2 shows a composite image consisting of multiple infrared scans of the process tubes in a heater. Many plants with process heaters take regular scans to ensure there are no hot spots that could lead to damage and premature shutdown. Bussman and Baukal have shown that even the ambient air temperature can have a significant impact on both thermal efficiency [5] and pollution emissions in industrial combustion processes [6].
97 © 2011 by Taylor and Francis Group, LLC
98
Industrial Combustion Testing
5.2 Temperature Measurement Examples
Flames into roof tubes
Figure 5.1 (See color insert following page 424.) Flame impingement on the process tubes in the convection section of a process heater.
Outlets
Some examples of temperature measurements for industrial combustion processes will serve to illustrate the various methods that have been used that are described in some detail in this chapter. Hayes et al. [7] describe glass surface temperatures made in a production regenerative flat glass furnace (see Chapter 32). The optical measurements were made with a water-cooled, twocolor pyrometer inserted through holes in the crown (roof) at six locations along the length of the furnace. Both average and time-resolved glass surface temperatures were measured during the transient reversal cycles in the furnace. Average temperatures ranged from about 1700 K to a peak of about 1900 K depending on the time in the cycle and the location in the furnace. These measurements are discussed in more detail in Chapter 32 of this book. Honner, Vesely, and Svantner [8] made temperature measurements in a continuous pusher-type metal reheating furnace using thermocouples imbedded in a billet used specifically as a “sensor”. These were particularly challenging measurements, not only because of the harsh high temperature environment inside a reheat furnace, but also because the sensor billet was moving, which means the electrical wiring and batterypowered data acquisition system also moved through the furnace. Rafidi and Blasiak [9] measured local temperature in a test furnace using four high temperature air combustion (HiTAC) regenerative burners (see Chapter 21). A suction pyrometer was used to measure gas temperature and a thin-wire thermocouple (80 μm diameter) with a short response time (time constant = 25 ms) was used to measure temperature fluctuations. Fast response was particularly important in this application because of the dynamic nature of regenerative burners that cycle between firing and exhausting. An important objective of this study was to determine spatial temperature uniformity and comparing it to conventional nonregenerative burners.
Thermal probe
5.3 Thermocouples View port
Tubeskin
Tubeskin
Figure 5.2 (See color insert following page 424.) Example of an infrared scan of process tubes in a heater.
© 2011 by Taylor and Francis Group, LLC
Many different types of devices are used to measure temperature, depending on the temperature range and specific application. The most common device is the thermocouple. There are many references on using thermocouples in combustion applications. Some are briefly discussed here. Eckert and Goldstein [10], Goldstein and Chiang [11], Holman [12], and Deshmukh [13] are
99
Temperature
some examples of references that discuss general temperature measurement techniques. Bowman [14] discusses the use of various types of temperature probes in flames. Cornforth [15] devotes a chapter to instrumentation and measurement, including temperature, in combustion applications. The report by Okoh and Brown [16] reviews and compares many techniques for the measurement of temperature in combustion systems. The report also includes equipment specifications and suppliers. Thermocouples are the primary device used to make temperature measurements in industrial combustion applications. A variety of other techniques have been used to measure both gas and surface temperatures that are briefly discussed later in this chapter. Due to the importance of thermocouples, they are considered in some detail in this section. Temperature measurement using thermocouples is based on the thermoelectric effect. Two dissimilar metals are joined together at a junction where an electromotive force (emf) is generated according to the Seebeck effect. The emf level depends on the junction temperature. The Peltier effect causes an emf to be generated when the dissimilar metals are connected to an electrical circuit.
A third emf is produced if there is a temperature gradient along either of the materials. The emf generated by the Peltier effect at the junction can be used to measure temperature, knowing the thermoelectric properties of the metals. There are many different types of thermocouples, depending on the temperature range needed. Table 5.1 shows the common thermocouple types and the alloys used in each. Each combination generates a specific millivolt output as a function of temperature. These millivolt reference tables are based on some reference junction temperature which is usually 0°C. Most electronic data acquisition systems have these tables built into the software, usually through a curve fit of the data, so the millivolt readings are automatically converted to temperature. For best accuracy, a thermocouple with the appropriate range should be used for the expected temperatures. For example, type T thermocouples are typically used for lower temperature measurements such as the water temperature in a cooling system. In industrial combustion, Type J and K thermocouples are commonly used to measure intermediate temperatures, such as the temperature inside a refractory wall or temperatures inside a burner. Type
TABLE 5.1 Application Characteristics of Some Common Thermocouple Alloys Max T °F
Max T °C
Allowable Atmos. (Hot)
5072
2800
5000
2760
4000
2210
Inert, H2, vacuum Inert, H2, vacuum Inert, H2
3720
1800
Oxidizingb
2900
1600
Oxidizingb
2800
1540
Oxidizingb
2372
1300
Oxidizingb,c
2300
1260
Oxidizing
1800 1600 750
980 875 400
Reducinga Reducing Reducing
a b c d e f g
Material Names Tungsten/tungsten 26% rhenium Tungsten 5% rhenium/tungsten 26% rhenium Tungsten 3% rhenium/tungsten 35% rhenium Platinum 30% rhodium/ platinum 6% rhodium Platinum 13% rhodium/ platinum Platinum 10% rhodium/ platinum Platinel II (5355)/Platinel II (7674) Chromel/Alumel,d Tophel/ Nial,e Advance T1/T2,f Thermo-Kanathal P/Ng Chromel/constantan Iron/constantan Copper/constantan
Color Code
Output mV/100°F
—
—
0.86
—
—
—
—
0.76
—
—
—
—
0.74
—
—
B
—
0.43
1/2
1/4
R
—
0.64
1/4
1/4
S
—
0.57
1/4
1/4
—
—
2.20
5/8
—
K
Yellow red
2.20
4°F, or 3/4%
2°F, or 3/8%
E J T
Purple red White red Blue red
4.20 3.00 2.50
1/2 4°F, or 3/4% 3/4
3/8 2°F, or 3/8% 3/8
Per ANSI C96.1 Standard. Avoid contact with carbon, hydrogen, metallic vapors, silica, reducing atmosphere. @ Engelhard Corp. @ Hoskins Mfg. Co. Wilber B. Driver Co. Driver-Harris Co. The Kanthal Corp.
© 2011 by Taylor and Francis Group, LLC
Accuracy, %
ANSI Typea
Standarda
Precisiona
100
Industrial Combustion Testing
Flue gas temperature thermocouple
O2 probe
Figure 5.3 Example of a thermocouple at the top of the radiant section in a process heater. (Courtesy of John Zink Co. LLC, Tulsa, OK.)
R, S, and B (platinum–rhodium) thermocouples are commonly used to measure higher temperatures, such as hot face temperatures and gas temperatures inside a furnace. Type L thermocouples can be used for even higher temperatures, although they are not as readily available as the platinum–rhodium thermocouples. Another important factor to consider when selecting a thermocouple is the environment where it will be used. Some thermocouples can handle oxidizing and reducing conditions, while others cannot. For example, Type R, S, and B thermocouples can handle the high temperature oxidizing conditions that are often present in industrial combustion processes. While a Type C thermocouple (tungsten–rhenium) can measure higher temperatures than platinum–rhodium thermocouples, Type C thermocouples cannot be used in oxidizing environments— which severely limits their use in industrial combustion applications. Thermocouple installation is important to ensure the proper temperature will be measured. Figure 5.3 shows a thermocouple located at the top of the radiant section in a process heater used to measure the temperature of the heater. As will be discussed later, this type of thermocouple measurement needs to be corrected to get the actual temperature. The location of this thermocouple is important because if it is located too close to the wall, then the temperature will be lower due to the lower temperature tubes that are cooled by process fluid. That lower temperature would not be representative of the average heater temperature. Figure 5.4 shows a photo of the thermocouples used to measure the water outlet temperatures from calorimeters in a flame impingement heating study [17]. The thermocouples were positioned so that water would have to flow over the junctions, regardless of the flow rate. If the thermocouples were positioned, for example, perpendicular to a vertically downward flow of water, there is a good chance the junction would not
© 2011 by Taylor and Francis Group, LLC
Figure 5.4 Example of measuring the outlet temperature from calorimeters used to measure the heat flux in flame impingement heating. (From Baukal, C. E., Heat Transfer from Flame Impingement Normal to a Plane Surface, Saarbrücken, Germany: VDM Verlag, 2009.)
Figure 5.5 Portable thermocouple calibrator set to 150°F with a thermocouple inserted into it.
always contact the water flow, particularly at low flow rates. This would lead to measurement error in determining the actual water temperature. As with any instrument, calibration is important for thermocouples. Although manufacturers supply accuracy data for typical thermocouples, calibration can be used to check that these data apply to the specific instrument being used in an experiment. The accuracy can be improved by calibrating the specific thermocouple being used. More importantly, a thermocouple calibrator (see Figure 5.5) can be used to check for any wiring problems or any setup mistakes in the data acquisition system. A thermocouple that has been wired and connected to the data acquisition system can be inserted into the calibrator to check for problems. For example, if a Type R thermocouple is actually being used but it
101
Temperature
was set up as a Type K thermocouple in the data acquisition system, then systematic error will result as all readings will at least be in error by the difference between the calibration curves for the two different types of thermocouples.
5.4 Gas Temperature Many techniques have been used for measuring the spatial gas temperature distribution. They may be generally classified as contact (thermocouples) and optical, such as Rayleigh and Raman scattering [18–21]. At this time, the light-scattering methods are very expensive and difficult to use [22]. Their primary use is in turbulent flows, and where probe survivability is a problem. Thermocouples are preferable because of their well-established properties, low cost, and ease of use. Some optical techniques, such as line reversal, may be difficult to use because of the need to seed the flame with particles. Those seeds could plug burners with small passage sizes. 5.4.1 Suction Pyrometer
Thermocouple junction
(b)
Thermocouple wires
Hot gases
Outer tube
A suction pyrometer is designed to measure closer to the true gas temperature than a bare thermocouple measures in a hot environment such as a combustion chamber. There are several types of heat transfer to and (a)
from the junction of a thermocouple that cause it to read lower than the actual temperature. This is discussed in more detail later, including how to calculate a correction for a bare thermocouple. A properly designed and operated suction pyrometer needs very little if any correction. At the desired measurement location, gases are extracted from the flame through a sampling tube in the suction pyrometer. A thermocouple is positioned just inside this tube, typically made of a ceramic or high temperature metal, which acts as a radiation shield. In some designs, multiple shields are used (see Figure 5.6). Examples of suction pyrometers are shown in Figure 5.7. This technique is particularly useful when measuring gas temperatures in hot-surfaced enclosures. Surface radiation from the walls to an unshielded thermocouple
Middle tube Center tube
Hot gas eductor
Air or steam
Figure 5.6 Drawing of a multishield suction pyrometer.
(c)
Figure 5.7 Examples of suction pyrometers: (a) new, (b) in the sidewall of a furnace, and (c) in the roof of a furnace. (From Baukal, C. E., Heat Transfer in Industrial Combustion, Boca Raton, FL: CRC Press, 2000.)
© 2011 by Taylor and Francis Group, LLC
102
may introduce large errors in the gas temperature measurement. Chedaille and Braud [23] and Goldman [24] specifically discuss the use of a suction pyrometer to measure gas temperatures in combustors. This device has been successfully used to measure gas temperatures in excess of 2200 K (3500°F). In flame impingement studies, Beér and Chigier [25], Vizioz and Lowes [26], Smith and Lowes [27], and Rajani, Payne, and Michelfelder [28] measured gas temperatures with a suction pyrometer. Grey [29] describes a technique related to suction pyrometry known as a twin sonic orifice probe. In this technique, the gases are extracted at a very high velocity, through two converging–diverging nozzles inside the probe. One limitation of the suction pyrometer is that active chemical reactions may also occur inside the probe passages for highly dissociated gases. With proper corrections, this effect may be overcome by the dual sonic orifice probe. Rajani [28] and Ivernel and Vernotte [30] used a twin sonic orifice probe to measure gas temperatures in oxygen/fuel flames up to 3060 K and 2800 K (5050°F and 4600°F), respectively. 5.4.2 Optical Techniques A variety of optical techniques have been used to measure gas temperatures in combustion applications, particularly in flames. There are potentially some important advantages of optical techniques compared to contact techniques such as suction pyrometers (see Figure 5.7). Optical measurement techniques do not disturb the flow, where thermocouples may have a significant impact on the fluid dynamics. Optical techniques can potentially measure higher temperatures as there are not the materials issues compared to thermocouples. For some optical techniques, temperature profiles can be measured at one point in time without the need to make multiple individual measurements over some length of time. Optical techniques often have a much faster response time compared to contact methods. This is particularly important in turbulent and transient flows. The techniques are categorized here as light scattering (where a light source is needed) and emission (no light source needed). 5.4.2.1 Light Scattering Laurendeau [31] gives a brief review of temperature measurements using light-scattering techniques. The techniques discussed included spontaneous vibrational and rotational Raman scattering, Rayleigh scattering and laser-induced fluorescence. Fluorescence methods
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
were recommended for temporally resolved planar thermometry. See Chapter 12 for a discussion of some of these techniques. Kondic [32] developed a method to measure the density and temperature fields in fluids along a light path where the light source and detector were located external to the system. Although a laser is the preferred light source, a conventional light source can also be used. The gas temperature is related to the density field and the index of refraction of the fluid. Posillico [33] made detailed spatial gas temperature measurements, using Raman spectroscopy [20], for air/CH4 flames impinging on a water-cooled flat plate. The highest reported gas temperature was 3200 K (5300°F), even though the adiabatic flame temperature for a stoichiometric air/CH4 flame is 2223 K (3542°F). No explanation was given for this apparent discrepancy. Hughes, Lacelle, and Parameswaran (1995) compared a suction pyrometer and CARS (coherent anti-Stokes Raman spectroscopy) in an industrial-scale flame [34] (See Chapter 13). Hofmann and Leipertz [35] presented experimental measurements of the temperature field in a sooting, methane/air flame using a new technique referred to as filtered Rayleigh scattering (FRS) [35]. Rumminger et al. [36] have used laser-induced fluorescence (LIF) to measure gas temperatures at the exit of porous radiant burners. The measurements were slightly higher than those made with a thermocouple after all corrections had been made to both techniques. The overall uncertainty in the LIF measurements was estimated to be ± 110°F (60 K). 5.4.2.2 Emission Chao and Goulard [37] presented a nonlinear iterative inversion technique for calculating flame temperatures from a single-line-of-sight multi-frequency set of radiance measurements. Hommert, Viskanta, and Mellor [38] measured flame temperatures using spectral remote sensing (SRS). A detector measures the line-of-sight spectral radiation coming from hot combustion products. Some advantages include: no external radiation source needed, noncontact, remote, and relatively simple equipment. Some disadvantages include: strong infrared emitter needed, average rather than point measurement, single data point rather than a profile, and accurate model needed to convert the data. The technique was tested on a flat air/CH4 diffusion flame, which had relatively high concentration and temperature gradients. The 4.3 μm band for CO2 was selected because it is strong with relatively little overlap with other species. An inversion algorithm was used to convert the spectral radiation measurements to temperatures. Temperature
103
Temperature
measurements were made with a Type S thermocouple, which varied from about 600 to 1800 K (600 to 2800°F) to compare against the SRS measurements. CO2 was added to some of the flames to vary the concentration. Kim and Song used SRS to make gas temperature measurements in furnaces [39–41]. In the first study [39], they investigated SRS in a 2 m (6.6 ft) long by 0.12 m (4.7 in.) inner diameter stainless steel tube where kerosene was the fuel and where the measurement was made along the length of the tube. The SRS measurements ranged from about 900 to 1400 K (1200 to 2100°F) and were within 4% of thermocouple readings. In the second study [40], a longer 3.4 m (11 ft) by 0.12 m (4.7 in.) inner diameter stainless steel tube, kerosene was again the fuel, and temperatures were similar to the first study. In the third study [41], measurements were made in three furnaces: 0.8 m long, 2.0 m long and 3.4 m long. Reuss [42] used holographic interferometry in fuel lean propane–air flames and compared the measurements with those made with a thermocouple [42]. The interferometry readings were within 50 K of those made with the thermocouple. This technique was developed to measure gas temperatures inside engine cylinders. Char and Yeh [43] describe an optical technique of measuring the flame temperature using a combination of measuring the infrared radiation at a selected wavelength (4.3 µm, which has the highest intensity for CO2 and H2O) and making some iterative calculations to determine the flame emissivity. The measurements were made on a fuel lean (equivalent ratio = ϕ = 0.51) open propane flame. The results compared favorably with thermocouple measurements. A correlation was developed for the gas emissivity for this flame as a function of the gas temperature position from the burner outlet. Correia, Ferrao, and Caldeira-Pires [44] discuss threedimensional emission tomography for use inside combustors with sooty flames. Previous applications of this technique did not account for radiation absorption (extinction) inside the flame. Correia et al. developed a reconstruction algorithm to account for that effect. The basic advantage of emission tomography compared to transmission tomography is that no background radiation source is needed, which makes it more amenable to industrial combustion environments. This technique was applied to both axisymmetric and nonaxisymmetric semi-industrial flames with uncertainties below ±10% where soot was available in the flame. Temperatures in the range of approximately 700 to 1200°C (1300°F– 2200°F) were reported and compared favorably with thermocouple measurements in laboratory propane flames. Temperatures of approximately 1000 to 1700°C (1800°F–3100°F) were measured in a large pilot-scale, oil-fired test furnace.
© 2011 by Taylor and Francis Group, LLC
Zhou et al. [45] used a flame image processing technique to measure the gas temperatures in a 200 MW pulverized-coal corner-fired boiler. Eight flame image CCD detectors were used to capture monochromatic radiation intensity images under the visible wavelengths of red, green, and blue. A blackbody furnace was used to calibrate the color images. A Monte Carlo method was used to calculate the radiation intensity to establish the relationship between the images and the temperature in the furnace. Three dimensional profiles were constructed from the measurements. The computed temperatures were within 5% of suction pyrometer measurements. 5.4.3 Fine Wire Thermocouples Fine wire thermocouples have a small diameter junction that minimizes both response time and the heat transfer effects that must be corrected for to get the actual temperature that is discussed later. In some cases, the junction may be coated to minimize chemical reactions at the junction that could be catalyzed by the bare metals and which could impact the measurements. Coatings are discussed first in this section. Measurements with fine wire thermocouples typically need to be corrected. Several methods for making these calculations are discussed in the second part of this section. Fine wire thermocouples have been used in numerous combustion applications. Some examples are given at the end of this section. 5.4.3.1 Coatings Heitor and Moreira [22] provided an extensive review of using thermocouples in combustion environments. They stated that flow disturbances and measurement errors can be made negligible in laminar flows. They suggested a coating consisting of 90% Al2O3 and 10% MgO, which minimizes catalytic effects and is not toxic. According to Cookson, Dunham, and Kilham (1964) [46], thermocouple junctions should be coated. The junctions should also be as small as possible, to prevent surface recombination effects. Kent [47] lists the following general requirements for a proper thermocouple coating: • noncatalytic to the flame • inert to the thermocouple material • impermeable to gases to protect the thermo couple • poor electrical conductor to prevent electrical shorts • stable within the desired temperature range • capable of being evenly applied
104
Industrial Combustion Testing
A coating made from a mixture of approximately 10% BeO/90%Y2O3 was recommended. Most high temperature applications require some type of protective coating on the thermocouple, unless the atmosphere is inert. Kaskan [48] described a procedure for applying a silica coating to thermocouples. The junction was immersed in a flame containing a small amount of hexamethyldisiloxane. This produced a uniform coating of fused quartz. It was found that the coating failed at about 1900 K (3000°F), due to softening of the silica. Cookson and Kilham [49], Kilham and Dunham [50], and Milson and Chigier [51] used silica coatings to prevent catalyic hydrogen atom recombination on the platinum in the thermocouple [64]. Madson and Theby [52] showed that uncoated platinum alloy thermocouples significantly overpredicted temperature, compared to silicacoated thermocouples in an O2/Ar/CH4 flame. Pollock [53] has shown that silica coatings may contaminate platinum-alloy thermocouple wires, in chemically reducing environments. Kremer, Buhr, and Haupt [61] used Al 2O3 to prevent radical recombination. Hoogendoorn, Popiel, and van der Meer [54], Popiel, van der Meer, and Hoogendoorn [55], and Van der Meer [56] used an yttrium–beryllium oxide (Yt 2O3/ BeO), as recommended by Kent [47]. This eliminates silica contamination. Unfortunately, beryllia is toxic and extremely poisonous. 5.4.3.2 Corrections When using thermocouples to measure high temperatures, the measurements must be corrected for the errors due to radiation, convection, and wire conduction. Sato et al. [57] present a calculation procedure for making these corrections. Moffat [58] notes the general equations for estimating these types of errors:
Econv = ( 1 − α )
Erad =
Econd =
v2 , 2 g c Jc p
(5.1)
σε 4 4 (Tprobe − Tsurr ), h tgas − tprobe mount hAc cosh l kAcond
where Econv = convection error Erad = radiation error Econd = conduction error α = recovery factor
© 2011 by Taylor and Francis Group, LLC
,
(5.2)
(5.3)
v = gas velocity gc = universal gravitational constant J = Joules constant cp = gas specific heat σ = Stefan-Boltzmann constant ε = probe emissivity h = convection coefficient for gas flowing over probe Tprobe = absolute temperature of the probe Tsurr = absolute temperature of the surroundings tgas = temperature of the gas flowing past the probe (usually what is being measured) tprobe mount = temperature of the probe mount l = length of the exposed junction Ac = area for convection heat transfer k = probe thermal conductivity Acond = area for conduction heat transfer. Note that these are iterative calculations as the temperature of the gas is the quantity being measured. Also, some of the quantities may not be known with a high degree of precision. Some of these include the convection coefficient, the gas velocity, the probe emissivity, the mount temperature, and the exposed junction length. Therefore, there will still be some uncertainty even after the measurements have been corrected for these errors. In general, the errors increase as the probe size increases. Several methods are commonly used to compensate for these errors. One is electrical compensation [59]. However, this method cannot be used at very high temperatures, where the thermocouple may already be near its melting point. Any additional heating caused by the electrical heating may cause the thermocouple to fail. As examples, Buhr, Haupt, and Kremer [60] and Kremer, Buhr, and Haupt [61] used electrical compensation for radiation and conduction heat losses. Another compensation method is to make measurements using diminishingly smaller diameter thermocouples [62]. The thermocouple measurements are plotted against the cross-sectional area of the thermocouple junction. The temperature curves are then extrapolated to a zero diameter thermocouple. This eliminates the need for correcting the measurements for the losses from the junction. In the study by Baukal [17], two different types of thermocouples were used to measure flame temperatures. Diminishing-diameter uncoated Type B thermocouples were used to measure the gas temperatures in oxygen-enhanced-air/natural gas flames. A coated Type L thermocouple was used to measure the gas temperatures in oxygen-enhanced/ natural gas flames (see Figure 5.8). The raw, uncorrected measurements for the Type B thermocouples are shown in Figure 5.9. Three different wire diameters were tested. These are referred to by their wire gauge numbers. The 20, 24, and 30 gauge wires had
105
Temperature
1800
1600
tj (K)
1400
1200 L = 0.5 L = 1.0 L = 2.0 L = 4.0 L = 6.0
1000
Figure 5.8 Type L fine wire thermocouple measuring the gas temperature in a natural gas flame.
1600
1400 tj (K)
0.0
0.5
1.0
1.5
2.0
dT/C2 (mm2)
2.5
3.0
3.5
4.0
Figure 5.10 Gas temperature (tj) measurements, along the centerline (Rp = 0) of an air/natural gas flame (q f = 5.00 kW), with 20, 24, and 30 gauge Type B thermocouples, extrapolated to zero diameter. (From Baukal, C. E., Heat Transfer from Flame Impingement Normal to a Plane Surface, Saarbrücken, Germany: VDM Verlag, 2009.)
1800
1200
1000 20 gauge 24 gauge 30 gauge
800
600
800
0
1
2
3
Lp
4
5
6
7
Figure 5.9 Uncorrected gas temperature (tj) measurements, along the centerline (Rp = 0) of an air/natural gas flame (q f = 5.00 kW), with 20, 24, and 30 gauge Type B thermocouples. (From Baukal, C. E., Heat Transfer from Flame Impingement Normal to a Plane Surface, Saarbrücken, Germany: VDM Verlag, 2009.)
diameters of 0.79, 0.48, and 0.23 mm, respectively. Using the method recommended by Nichols (1900), the measured temperatures, at a given axial location L, were plotted as a function of the square of the wire diameter. This is shown in Figure 5.10. The data was then extrapolated to a zero wire diameter. The extrapolations have been curve-fitted with a second-order
© 2011 by Taylor and Francis Group, LLC
polynomial. No extrapolation was possible at L = 0.5. There, a hollow inner core could be seen in the flame. The temperature measurements there were very unsteady. Also, the thermocouple wires were hotter at a distance from the junction, than at the junction. This confirmed the hollow inner core. Figure 5.11 shows a comparison of the extrapolated gas temperatures and the corrected gas temperatures, at L = 1, 2, 4, and 6. The corrected values were consistently higher than the extrapolated values. One possible explanation is the danger in extrapolating only three data points to zero diameter. Nichols [63] used four different wire diameters, to determine an extrapolation to zero diameter. In that study, extrapolated gas temperature curves increased much more rapidly, as the data was extended to zero, than is shown in Figure 5.10. In Nichols’s study, the wires ranged from 0.082 to 0.20 mm in diameter. In Baukal’s study, the wire diameters ranged from 0.23 to 0.79 mm in diameter. The rapid increase in Nichols’s data was for thermocouples that were much smaller than those used here. This rapid increase may not have been detected in Baukal’s study, since the thermocouples were much larger. Therefore, due to the uncertainty in the extrapolation, and due to the consistency of the corrected temperatures for all three wire sizes, the corrected temperatures (discussed next) were subsequently used in Baukal’s study.
106
Industrial Combustion Testing
qcatalytic + qconduction + qradiation + qconvection = 0.
(5.4)
Hayhurst and Kittelson [59] have shown that only surface reactions contribute to catalytic heating. This can generally be neglected when the thermocouple is coated. Bradley and Matthews [62] have shown that conduction may be neglected for wires over approximately 1 mm (0.04 in.) long. When conduction effects have been neglected, the energy balance then simplifies to: 4 t j = tT/C + σε T/C tT/C / h,
(5.5)
where tT/C is the uncorrected temperature as measured with the thermocouple and tj is the corrected temperature. The radiation from the environment to the junction has been neglected. For example, an emissivity of 0.14 has been recommended for uncoated platinum Type B thermocouple [65]. An emissivity of 0.60 is recommended for a coated Type L thermocouple [66]. A convection coefficient can be calculated using the equation recommended by Hinze [67]:
Nu = 0.42 Pr 0.2 + 0.57 Pr 0.33 Re0.5 .
(5.6)
An iterative procedure is required to calculate tj because the gas properties used to calculate the convection coefficient are dependent on tj. The gas temperature tj may also be required to calculate the gas velocity when, for example, Pitot-static probe measurements are used to calculate the velocity. Son, Queiroz, and Wood [68] presented a mathematical method for compensating for the thermal inertia effects of thermocouples. These heat transfer effects can cause errors if not properly corrected. A Type L (iridium) thermocouple was used to measure the gas temperatures in oxygen-enhanced/natural gas flames [17]. The raw, uncorrected, measurements are shown in Figure 5.12. These have been corrected using the calculation method discussed above. At q f = 15.0 kW,
Ω = 0.30 (30 vol% O2 in the oxidizer) was the lowest oxidizer composition that produced an acceptable flame. Several measurements were made at q f = 15.0 kW and Ω = 0.40 (40 vol% O2 in the oxidizer), before the Type L thermocouple failed. Table 5.2 shows a comparison of the maximum measured flame temperatures from Baukal’s study, after correction, with the adiabatic flame temperatures. The maximum measured temperatures, after corrections, were 340–488 K less than the adiabatic temperatures. This was due to heat losses from the flames, which were not adiabatic. These measured temperatures compare favorably with the gas temperatures reported in previous air and air/O2 flame jet studies [69]. Using calculations to correct for fine wire thermocouple errors has been the most popular method for correction. For example, Horsley, Purvis, and Tariq [70], Kataoka, Shundoh, and Matsuo [71], You [72], Hargrave, 1900
1800
1700 tj (K)
A third method to correct thermocouple measurements is to compute the correction [64]. An energy balance at the thermocouple junction may be given as:
1600
1500 30 gauge, corrected 24 gauge, corrected 20 gauge, corrected Extrapolated
1400
1300
0
1
2
3
4
5
Figure 5.11 Comparison of the extrapolated and the corrected Type B thermocouple gas temperature (tj) measurements along the centerline (Rp = 0) of an air/natural gas flame (q f = 5 kW). (From Baukal, C. E., Heat Transfer from Flame Impingement Normal to a Plane Surface, Saarbrücken, Germany: VDM Verlag, 2009.)
Table 5.2 Comparison of the Peak Measured (after Correction) Flame Temperatures with Adiabatic Flame Temperatures qf (Kw) 5 15 15
Ω
T/C Type
tj,max (K)
0.21 0.30 0.35
B L L
1880 2170 2170
Adiabatic Flame Temperature (K) 2220 2546 2658
Source: From Baukal, C. E., Heat Transfer from Flame Impingement Normal to a Plane Surface, Saarbrücken, Germany: VDM Verlag, 2009.
© 2011 by Taylor and Francis Group, LLC
6
L
107
Temperature
Fairweather, and Kilham [73], Hustad, Røkke, and Sønju [74], Hustad, and Sønju [75], and Hustad, Jacobsen, and Sønju [76] calculated corrections for the heat losses from the thermocouple junction.
2200
2000
5.4.3.3 Example Uses
© 2011 by Taylor and Francis Group, LLC
1800 tj (K)
Many researchers have used fine wire thermocouples (T/Cs), with and without coatings. Costa et al. [77] used a bare 300 µm diameter wire uncoated Pt/Pt13%Rh thermocouple to measure gas temperatures inside an industrial glass-melting furnace. Mital et al. [78] used three different diameter (76, 125, and 200 µm) fine wire Type R thermocouples to measure the gas temperature inside porous ceramic radiant burners by extrapolating the measurements to zero diameter. Rumminger et al. [36] used fine wire thermocouples to measure the gas temperature above the surface of a porous radiant burner. At 5 mm (0.2 in.) above the surface, they estimated a temperature correction of 270 K (490°F) including the effects of re-radiation from the burner surface to the thermocouple bead. Such a large correction resulted from a relatively large bead size (460 µm), high bead emissivity (0.6), and relatively slow gas flow (Red = 0.9, Nu = 0.7). Fontes, Costa, and Azevedo [79] used 300 μm diameter Type R thermocouple in an alumina sheath with an external diameter of 4 mm and placed inside a 3 m or 5 m long water-cooled stainless steel probe to measure gas temperatures in a black liquor recovery boiler. Calculations showed the “true” temperature in the regions of highest temperature did not exceed measurements by more than 8%. Measurements were made at multiple elevations, multiple locations at each elevation, and at multiple insertion depths into the boiler. A traversing mechanism was used to insert the probe, which was checked frequently for deposition of black liquor particles and sulfate deposits. The temperatures ranged from approximately 700°C to 1000°C. In a flame impingement study, two types of thermocouples were used to measure the gas temperatures [17]. For the lower temperature flames, uncoated bare wire Type B thermocouples (Pt-60% Rh/Pt-6% Rh) were used. The probes were designed according to the method recommended by Peterson [80]. However, those thermocouples melted at Ω > 0.25. For the higher temperature flames, an Ir / 60% Ir-40% Rh thermocouple was used. Candler [81] has referred to this thermocouple as a Type L. According to Blackburn and Caldwell [82], this thermocouple may be used to measure temperatures up to 2400 K (3900°F), within ±20 K (±40°F), using the appropriate reference table. This accuracy only refers to the measured temperature. The actual gas temperature still requires correction for the losses from the thermocouple junction. The Type L thermocouple wires were 0.25 mm
1600
1400
Ω = 0.30, Uncorrected Ω = 0.35, Uncorrected Ω = 0.30, Corrected Ω = 0.35, Corrected
1200 0
1
2
3
Lp
4
5
6
7
Figure 5.12 Gas temperature measurements (tj) along the centerline (Rp = 0) of flames (q f = 15.0 kW), as a function of the oxidizer composition (Ω = 0.30 and 0.35), using a Type L thermocouple. (From Baukal, C. E., Heat Transfer from Flame Impingement Normal to a Plane Surface, Saarbrücken, Germany: VDM Verlag, 2009.)
(0.01 in.) in diameter. Figure 5.8 shows this type L thermocouple measuring the temperature in an oxygenenhanced natural gas flame. Fine wire thermocouples are of low cost and easy to use. They have relatively short response times and good spatial resolution. However, they are somewhat intrusive to the flame. Corrections are also required due to heat transfer effects at the junction bead. These include radiation from the bead to the environment and the heat conduction along the thermocouple wires. Often the junction must be coated to prevent catalytic reactions. There are material limitations, due to high temperature oxidizing and reducing conditions. They can only make so many point measurements over a span of time to get a temperature profile. 5.4.4 Line Reversal Many researchers have used this technique, originally suggested by Féry [83], to measure flame temperatures. The flame is commonly seeded with sodium salt. The salt vaporizes in the flame and dissociates into sodium atoms and other products. These excited atoms emit at a specific wavelength. This radiation is then compared to a calibrated reference source, such as a tungsten filament lamp. The temperature of this source is adjusted, until it matches the seed radiation. To the naked eye, the background source appears to disappear when it is at
108
Industrial Combustion Testing
the same temperature as the flame. Strong et al. [84] give a general discussion of this technique to measure the temperatures in flames [84]. Thomas [85,86] discusses some of the problems associated with this technique as well as a method for automatically remotely operating a system for using the method. Kilham [87] used a spectrometer to compare the flame with the filament. Fells and Harker [88] estimated the accuracy of their line reversal measurements to be 20K. Nawaz [89] and Fairweather and colleagues [90,91] introduced sodium chloride smoke, by passing N2 over molten salt, through a tube in the center of their burner. Therefore, only the temperatures along the central axis of the flame were measured. This reduced the influence of ambient air entrainment cooling effects, at the flame periphery. Nawaz used a high pressure xenon arc lamp (maximum temperature of 5000 K or 8500°F) as the comparison radiator, since the temperature levels were out of the range of the conventional tungsten lamps that were available at that time. Measured temperatures were 50–100 K (90°F–180°F) below the calculated adiabatic flame temperature. Fairweather [91] measured a 33 K (59°F) and 40 K (72°F) reduction, across the equilibrium zone between the reaction zone and the stagnation point, for laminar and turbulent flames, respectively. Atmospheric air entrainment into the flame was thought to be a possible explanation of this effect. This method is only slightly intrusive, due to the seeding. It is of relatively low cost. It may be effectively used for very high temperatures. However, the seeded field must be uniform. This may require the addition of a carrier gas to transport the seeds into the flame. Only an average gas temperature, across the flame, is measured. A calibrated reference light source is required and the adjustment for the source to vanish is somewhat subjective. 5.4.5 Acoustic Pyrometry The speed of sound in a gas is dependent on temperature. For an ideal gas, the speed of sound is:
cideal = γRT ,
(5.7)
where γ is the ratio of specific heats at constant pressure and volume (Cp/Cv), R is the specific gas constant, and T is the absolute temperature. The product γR is sometimes referred to as the acoustic constant, which is dependent on the gas composition [92]. If the speed is known, then the temperature can be calculated from the equation, assuming the gas composition is known. The error in not knowing the exact gas composition of combustion products can be significant, but can usually be minimized with proper knowledge of the fuel and operating conditions [92]. This principle can be used to measure temperature in a gas by measuring the flight time for
© 2011 by Taylor and Francis Group, LLC
sound to travel between a source and a receiver that are a known distance apart. The distance traveled divided by the flight time gives the speed. This technique is referred to as acoustic pyrometry and has been applied in industrial combustors, particularly boilers [93–95] and process heaters [96,97]. The average temperature is measured across the flight path. By using an array of sources and receivers and a deconvolution algorithm [98], a temperature profile in a combustor can be developed. A common method used to generate the sound waves is using blasts of air through a nozzle. One source can be used with multiple receivers. In some cases, Pitot tubes and thermocouples are used in conjunction with the acoustical pyrometry system to improve accuracy. The practical frequency range for these pyrometers is between 500 and 2000 Hz. A study funded by the Electric Power Research Institute (EPRI) showed acoustic pyrometry is easy to use, accurate, nonintrusive, and real-time, which makes it useful for monitoring and control [93]. It can be used as a diagnostic tool to identify burner and furnace problems.
5.5 Surface Temperature There are many surface temperatures that may be important in industrial combustion applications. The product temperature (usually measured at the surface) is typically an important variable that often must be maintained within a relatively narrow band. In some cases, the product temperature is part of the control system, where the firing rate is adjusted to maintain a target product temperature. The surface temperature of equipment inside the combustion chamber, such as process tubes, is often important. Figure 5.13 shows an example of an infrared scan of the process tubes inside a process heater. This particular image was taken while the heater tubes were being de-coked, where coke build-up inside the process tubes is burned out. The location of the de-coking can be seen by the elevated temperatures near the middle of the photo. Furnace refractory temperatures may be of interest to ensure they do not exceed the material limits. Burner temperatures may be important to ensure there are no hot spots that could reduce the operating life. There are two common ways to measure the surface temperature of an object that are discussed next. 5.5.1 Embedded Thermocouple In the flame impingement study by Baukal [17], a Type K thermocouple was imbedded in the impingement rings of copper calorimeters (see Figure 5.14). The
109
Temperature
1000
1267.5°F
Reff = 0.16 Reff = 0.37 Reff = 0.59 Reff = 0.82 Reff = 1.04
900 1200
1100
tw (K)
800
1000
700
600
947.8°F
500
Figure 5.13 Infrared scan of tube temperatures inside a process heater during a de-coking cycle. (Courtesy of Quest Tru-Tec.)
400
0
1
2
3
4
5
6
L Figure 5.15 Surface temperature (tw) for O2/natural gas flames (q f = 15.0 kW) impinging on an untreated stainless surface. (From Baukal, C. E., Heat Transfer from Flame Impingement Normal to a Plane Surface, Saarbrücken, Germany: VDM Verlag, 2009.)
The hot face temperatures were then calculated using the one-dimensional, steady-state conduction equation. Using the calculated heat flux to each ring, the hot face temperature, tw, was calculated using:
tw = tmiddle +
q ′′ l1 2 , k1
(5.8)
where tmiddle was the measured temperature at the middle of the ring, q” was the calculated heat flux to the ring, l1 was the thickness of the impingement ring, and k1 was the thermal conductivity of the impingement ring. A typical result is shown in Figure 5.15. This graph shows that the wall temperatures decreased as the axial distance between the burner and the target decreased. However, for L > 2, the differences became smaller. Figure 5.14 Thermocouple embedded in a ring calorimeter to measure surface temperature. (From Baukal, C. E., Heat Transfer from Flame Impingement Normal to a Plane Surface, Saarbrücken, Germany: VDM Verlag, 2009.)
thermocouples had a 0.5 mm (0.02 in.) o.d. inconel sheath and the clearance hole in the side of each impingement ring was 0.64 mm (0.025 in.) i.d. A high conductivity antiseize sealant was applied to the end of each thermocouple, before it was inserted into a ring, to ensure good thermal contact. These thermocouples measured the temperatures at the middle of each impingement ring.
© 2011 by Taylor and Francis Group, LLC
5.5.2 Infrared Detectors Thermal imaging systems have been used for years in a variety of applications including military and intelligence surveillance activities. Within about the last decade, the cost of these systems has declined enough that they are now being routinely used to measure temperatures in combustion applications. In combustion systems, these IR imaging systems are typically used to measure the temperature of the furnace or heater walls, the heat load which might be a material like scrap metal or a tube used for process fluid heating, or the
110
Industrial Combustion Testing
burner itself. Imaging systems are primarily used as a diagnostic to detect hot spots which could indicate potential problem areas. Infrared detectors have a number of potential advantages over thermocouples [99]: • • • •
faster response times very high accuracies can measure temperatures on moving objects can measure higher temperatures than conventional thermocouples • can give two-dimensional temperature profiles • do not disturb the surface being measured • are not dependent on how they are attached to the surface, as thermocouples are, to get an accurate surface temperature measurement There are also some potential disadvantages compared to thermocouples: • the surface emissivity is often required, except for two-color pyrometers • measurements can be impaired by anything, including combustion products, in between the pyrometer and the surface, which could absorb any of the radiation emitted by the surface • the front lens of the pyrometer must be kept clean • the electronics in the pyrometer must be kept cool • calibration is typically more involved than for a thermocouple The theory of radiation pyrometry is available elsewhere [100]. The original devices used for combustion applications were infrared pyrometers that give a temperature based on the infrared energy emitted by the object and an input emissivity for the object. An example of an infrared pyrometer being used to measure burner surface temperature is shown in Figure 5.16. Newer devices use imaging technology to produce thermal images of the temperature profile of an object (see Figure 5.13) [101]. These devices are capable of producing real-time images of dynamically changing temperature profiles. A typical system consists of a high resolution infrared camera, a computer, and thermal image analysis software. Besides displaying the temperature profile for an object, the software provides capabilities for analyzing the data. Due to the fast response of the new systems, it is also possible to use them for real-time control of the heating system to produce specific temperature profiles in the load.
© 2011 by Taylor and Francis Group, LLC
Figure 5.16 Example of an infrared pyrometer measuring the surface temperature of a porous refractory matrix burner. (From Baukal, C.E., Heat Transfer in Industrial Combustion, Boca Raton, FL; CRC Press, 2000.)
Lappe [102] gave some recommendations for the installation and use of single-color IR thermometers. Single color means the sensor is tuned to operate at one specific wavelength. The most important factor in measuring the correct surface temperature is to know the emissivity of the body being measured. The emissivity of a graybody is constant for all wavelengths and usually does not change significantly with temperature unless there is a change in the material or in its surface characteristics. For example, if a solid material melts and becomes liquid there may be a significant change in its emissivity. Another example is if a metal oxidizes or becomes roughened from wear, its emissivity also may change dramatically. Lappe recommends using the shortest wavelength instrument possible to measure the surface temperature, for two reasons. The first is that the difference in the emissivity of common surfaces such as metals is usually significantly less at shorter wavelengths, with the difference increasing with wavelength. Therefore, there will be less error if the emissivity used on the pyrometer is not properly adjusted for changes in the surface conditions of the body whose temperature is being measured. The second reason is that lower wavelength instruments typically have less error in the temperature reading compared to higher wavelength units. Lappe also gave some recommendations for the installation of these units: • make sure the unit is focused properly on the object of interest and that the spot size is the correct size • make sure there are no obstructions (e.g., sightports walls, smoke, steam, etc.) between the pyrometer and the surface being measured
111
Temperature
• fiber optic systems may be appropriate for certain applications • ensure that the interconnecting cable is not overheated or installed with other wires that may carry high voltages that could adversely affect the signal • make sure the sensing head is kept clean and cool but not overcooled, which could cause condensation in the unit • keep the lens clean with a purge of clean dry air • have the instrument calibrated at least annually Two-color pyrometers avoid the problem of determining the emissivity of the surface. Those pyrometers make measurements at two different wavelengths and with appropriate computations determine temperature without knowing the emissivity. Figure 5.17 shows infrared temperature measurements made on a burner [103]. By selecting an appropriate wavelength for the intensity measurement (in this case, the wavelength was 3.9 μm), the effects of CO2 and H2O absorption between the emitting surface and the infrared camera were negated because those gases are essentially transparent at that wavelength. Surface temperatures can then be readily measured “through” a flame. It is very difficult to make reliable gas temperature measurements by measuring the infrared emission because the emission from gases depends on the temperature of the gas volume, the composition of the gas volume, as well as the dimension of the gas volume. Since most real applications involve nonisothermal gas
*> 900.0°C 895.2 875.4 855.1 834.3 812.5 789.0 762.6 730.8 691.3 648.9 610.4 573.2 530.5 *< 530.1°C Figure 5.17 (See color insert following page 424.) Infrared thermal image of a flame in a furnace. (From Singh, P., Henneke, M., Jayakaran, J. D., Hayes, R., and Baukal, C. E., The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001.)
© 2011 by Taylor and Francis Group, LLC
volumes (such as a flame in a furnace), IR measurements are not feasible. Hernberg, Stenberg, and Zethraeus [104] used a twocolor infrared pyrometer coupled to a fiberoptic probe to measure burning fuel particle temperatures in a fluidized bed furnace. The probe consisted of a 1 mm diameter quartz fiberguide encapsulated in an uncooled stainless steel tube. The probe was operated in temperatures of about 850°C and pressures up to 20 bar. Twocolor (wavelengths of 0.65 μm and 1.05 μm) allowed determination of both temperature and a parameter related to particle emissivity, particle size, and geometrical configuration factor. The instrument was calibrated against a blackbody standard. A statistical method was developed to determine particle size from these measurements in addition to temperature. 5.5.3 Other Techniques Stephan et al. [105,106] used a technique referred to as microwave radiometry where low cost X-band microwaves are used to measure surface temperatures in industrial combustion environments with suspended particles such as water vapor and dust. In most applications, IR pyrometry is used to inexpensively measure product temperatures. However, in heavily particle-laden environments, an IR pyrometer’s view is obscured making it difficult to get an accurate temperature measurement. In microwave radiometry, the longer wavelengths are not scattered by the particulates. The temperature range of one version of this device was from room temperature up to about 500°C (900°F). This device was used in a demonstration that showed the device could accurately measure the surface temperature of a layer of carbon adsorbent through an artificially created, water-based fog [105]. A higher temperature version of the device was used to measure the wall temperatures inside a dusty cement kiln that ranged up to approximately 1500°C (2700°F).
5.6 Future Measuring temperatures inside industrial combustors and in flames continues to be challenging. More tools are being developed and applied to these conditions. Some of those tools are not yet commercially viable, which means they may be too expensive, require specialized knowledge to be properly used, or have not yet been ruggedized enough to handle the challenges of an industrial environment. This is likely to change in the future as the equipment costs decline and more
112
experience is gained using the techniques under industrial operating conditions. One might argue that the rapid expansion in the use of computational fluid dynamics [107], may cause a decline in experimental measurements. However, the old adage “garbage in = garbage out” is still relevant because a computer model that has not been validated is problematic. Given the extreme operating conditions, environmental regulations, and safety considerations associated with industrial combustion applications, relying solely on computer models in the absence of validating measurements seems to be imprudent. Therefore, making temperature measurements in industrial combustion applications is expected to continue into the foreseeable future.
References
1. Baukal, C. E. Heat Transfer in Industrial Combustion. Boca Raton, FL: CRC Press, 2000. 2. Baukal, C. E. Industrial Combustion Pollution and Control. New York: Marcel Dekker, 2004. 3. Reed, R. J. North American Combustion Handbook. Vol. 1, 3rd ed. Cleveland, OH: North American Mfg. Co., 1986. 4. Baukal, C., Hong, J., Bussman, W., and Waibel, R. “Cold Furnace Startup Emissions.” Chemical Engineering Progress 103, no. 2 (2007): 42–46. 5. Bussman, W., and Baukal, C. “Ambient Condition Effects on Process Heater Efficiency.” Energy, 34, no. 10 (2009). 1624–35. 6. Bussman, W., and Baukal, C. “Ambient Condition Effects on Process Heater Emissions.” Paper IMECE2008-68284, Proceedings of the International Mechanical Engineering Congress & Exhibition, Boston, MA, November 2008. 7. Hayes, R. R., Wang, J., McQuay, M. Q., and Webb, B. W. “Predicted and Measured Glass Surface Temperatures in an Industrial, Regeneratively Gas-Fired Flat Glass Furnace.” Glass Science and Technology-Glastechnische Berichte 72, no. 12 (1999): 367–77. 8. Honner, M., Vesely, Z., and Svantner, M. “Temperature and Heat Transfer Measurement in Continuous Reheating Furnaces.” Scandinavian Journal of Metallurgy 32, no. 5 (2003): 225–32. 9. Rafidi, N., and Blasiak, W. “Heat Transfer Characteristics of HiTAC Heating Furnace Using Regenerative Burners.” Applied Thermal Engineering 26, no. 16 (2006): 2027–34. 10. Eckert, E. R. G., and Goldstein, R. J. Measurements in Heat Transfer. 2nd ed. Washington: Hemisphere, 1976. 11. Goldstein, R. J., and Chiang, H. D. “Measurement of Temperature and Heat Transfer.” In Handbook of Heat Transfer Applications, edited by W. M. Rohsenow, J. P. Hartnett, and E. N. Ganic´, 2nd ed. New York: McGrawHill, 1985. 12. Holman, J. P. Experimental Methods for Engineers. 5th ed. New York: McGraw-Hill, 1989.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
13. Deshmukh, Y. V. Industrial Heating. Boca Raton, FL: CRC Press, 2005. 14. Bowman, C. T. “Probe Measurements in Flames.” Progress in Astronautics and Aeronautics 53 (1977): 1–24. 15. Cornforth, J. R. Combustion Engineering and Gas Utilisation (sic). 3rd ed. London: E&FN Spon, 1992. 16. Okoh, C. I., and Brown, R. A. Combustion Experimentation Handbook. Gas Research Institute report GRI-88/0143, Chicago, IL, 1988. 17. Baukal, C. E. Heat Transfer from Flame Impingement Normal to a Plane Surface. Saarbrücken, Germany: VDM Verlag, 2009. 18. Durão, D. F. G., Heitor, M. V., Whitelaw, J. H., and Witze, P. O. Combusting Flow Diagnostics. Dordrecht: Kluwer Academic Publishers, 1992. 19. Taylor, A.M.K.P. Instrumentation for Flows with Combustion. London: Academic Press, 1993. 20. Eckbreth, A. C. Laser Diagnostics for Combustion Temperature and Species. Cambridge, MA: Abacus Press, 1988. 21. Chigier, N. Combustion Measurements. New York: Hemisphere Publishing Corp., 1991. 22. Heitor, M. V., and Moreira, A. L. N. “Probe Measurements of Scalar Properties in Reacting Flows.” In Combusting Flow Diagnostics, edited by D. F. G. Durão, M. V. Heitor, J. H. Whitelaw, and P. O. Witze, 79–136. Netherlands: Kluwer Academic Publishers, 1992. 23. Chedaille, J., and Braud, Y. Vol. 1: Measurements in Flames. New York: Crane, Russak & Co., 1972. 24. Goldman, Y. “Gas Temperature Measurement in Combustors by Use of Suction Pyrometry.” in Heat Transfer in Furnaces, edited by C. Presser and D. G. Lilley, 19–22. New York: ASME HTD-Vol. 74, 1987. 25. Beér, J. M., and Chigier, N. A. Combustion Aerodynamics London: Applied Science Publishers, 1972. 26. Vizioz, J.-P., and Lowes, T. M. “Convective Heat Transfer from Impinging Flame Jets.” International Flame Research Foundation report F 35/a/6, IJmuiden, the Netherlands, 1971. 27. Smith, R. B., and Lowes, T. M. “Convective Heat Transfer from Impinging Tunnel Burner Flames: A Short Report on the NG-4 Trials.” International Flame Research Foundation report F 35/a/9, IJmuiden, the Netherlands, 1974. 28. Rajani, J. B., Payne, R., and Michelfelder, S. “Convective Heat Transfer from Impinging Oxygen-Natural Gas Flames: Experimental Results from the NG5 Trials.” International Flame Research Foundation report F 35/a/12, IJmuiden, the Netherlands, 1978. 29. Grey, J. “Thermodynamic Methods of High Temperature Measurement.” ISA Transactions 4 (1965): 102–15. 30. Ivernel, A., and Vernotte, P. “Etude expérimentale de l’amélioration des transferts convectis dans les fours par suroxygénation du comburant.” Revue Generale de Thermique 18, nos. 210–211 (1979): 375–91. 31. Laurendeau, N. M. “Temperature Measurements by Light-Scattering Methods.” In Developments in Experimental Techniques in Heat Transfer and Combustion, edited by R. O. Warrington, M. M. Chen, J. D. Felske, and W. L. Grosshandler, 45–65. New York: ASME HTD-Vol. 71, 1987.
Temperature
32. Kondic, N. N. “Temperature Field Measurement in Flames by External Means.” In Heat Transfer in Flames, edited by N. H. Afgan and J. M. Beer, 353–63. Washington, DC: Scripta Book Company, 1974. 33. Posillico, C. J. “Raman Spectroscopic and LDV Measurements of a Methane Jet Impinging Normally on a Flat Water-Cooled Boundary.” PhD thesis, Polytechnic Institute of New York, 1986. 34. Hughes, P. M. J., Lacelle, R. J., and Parameswaran, T. “A Comparison of Suction Pyrometer and CARS Derived Temperatures in an Industrial Scale Flame.” Combustion Science and Technology 105 (1995): 131–45. 35. Hofmann, D., and Leipertz, A. “Temperature Field Measurements in a Sooting Flame by Filtered Rayleigh Scattering (FRS).” 26th Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 945–50, 1996. 36. Rumminger, M. D., Dibble, R. W., Heberle, N. H., and Crosley, D. R. “Gas Temperature Above a Porous Radiant Burner: Comparison of Measurements and Model Predictions.” 26th Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 1755–62, 1996. 37. Chao, C. M., and Goulard, R. “Nonlinear Inversion Techniques in Flame Temperature Measurements.” In Heat Transfer in Flames, edited by N. H. Afgan and J. M. Beer, 295–337 Washington, DC: Scripta Book Company, 1974. 38. Hommert, P. J., Viskanta, R., and Mellor, A. M. “Flame Temperature Measurements by Spectral Remote Sensing.” Combustion and Flame 30 (1977): 295–308. 39. Kim, H. K., and Song, T.-H. “Measurement of Gas Temperature Distributions in a Test Furnace Using Spectral Remote Sensing.” Journal of Quantitative Spectroscopy & Radiative Transfer 73 (2002): 517–28. 40. Kim, H. K., and Song, T.-H. “Determination of the Gas Temperature Profile in a Large-Scale Furnace Using a Fast/Efficient Inversion Scheme for the SRS Technique.” Journal of Quantitative Spectroscopy & Radiative Transfer 93 (2005): 369–81. 41. Song, T.-H. “Spectral Remote Sensing for Furnaces and Flames,” Heat Transfer Engineering 29, no. 4 (2008): 417–28. 42. Reuss, D. L. “Temperature Measurements in a Radially Symmetric Flame Using Holographic Interferometry.” Combustion and Flame 49 (1983): 207–19. 43. Char, J.-M., and Yeh, J.-H. “The Measurement of Open Propane Flame Temperature Using Infrared Technique,” Journal of Quantitative Spectroscopy & Radiative Transfer 56, no. 1 (1996): 133–44. 44. Correia, D. P., Ferrao, P., and Caldeira-Pires, A. “Flame Three-Dimensional Tomography Sensor for In-Furnace Diagnostics.” Proceedings of the Combustion Institute 28 (2000): 431–38. 45. Zhou, H.-C., Lou, C., Cheng, Q., Jiang, Z., He, J., Huang, B., Pei, Z., and Lu, C. “Experimental Investigations on Visualization of Three-Dimensional Temperature Distributions in a Large-Scale Pulverized-Coal-Fired Boiler Furnace.” Proceedings of the Combustion Institute 30 (2005): 1699–1706.
© 2011 by Taylor and Francis Group, LLC
113
46. Cookson, R. A., Dunham, P. G., and Kilham, J. K. “NonCatalytic Coatings for Thermocouples.” Combustion and Flame 8 (1964): 168–70. 47. Kent, J. H. “A Noncatalytic Coating for PlatinumRhodium Thermocouples.” Combustion and Flame 14 (1970): 279–82. 48. Kaskan, W. E. “The Dependence of Flame Temperatures on Mass Burning Velocity.” Sixth Symposium (International) on Combustion, 134–43. Reinbold, NY, 1957. 49. Cookson, R. A., and Kilham, J. K. “Energy Transfer from Hydrogen-Air Flames.” Ninth Symposium (International) on Combustion, 257–63. New York: Academic Press, 1963. 50. Kilham, J. K., and Dunham, P. G. “Energy Transfer from Carbon Monoxide Flames.” 11th Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 899–905, 1967. 51. Milson, A., and Chigier, N. A. “Studies of Methane-Air Flames Impinging on a Cold Plate.” Combustion and Flame 21 (1973): 295–305. 52. Madson, J. M., and Theby, E. A. “SiO2 Coated Thermocouples.” Combustion Science and Technology 36 (1984): 205–9. 53. Pollock, D. D. “Thermocouples in High-Temperature Reactive Atmospheres.” Combustion Science and Technology 42 (1984): 111–13. 54. Hoogendoorn, C. J., Popiel, C. O., and van der Meer, T. H. “Turbulent Heat Transfer on a Plane Surface in Impingement Round Premixed Flame Jets.” Proceedings of 6th International Heat Transfer Conference, Toronto, Vol. 4, pp. 107–12, 1978. 55. Popiel, C. O., van der Meer, T. H., and Hoogendoorn, C. J. “Convective Heat Transfer on a Plate in an Impinging Round Hot Gas Jet of Low Reynolds Number.” International Journal of Heat and Mass Transfer 23 (1980): 1055–68. 56. van der Meer, T. H. “Stagnation Point Heat Transfer from Turbulent Low Reynolds Number Jets and Flame Jets.” Experimental Thermal and Fluid Science 4 (1991): 115–26. 57. Sato, A., Hashiba, K., Hasatani, M., Sugiyama, S., and Kimura, J. “A Correctional Calculation Method for Thermocouple Measurements of Temperatures in Flames.” Combustion and Flame 24 (1975): 35–41. 58. Moffat, R. J. “Temperature and Heat Transfer Measurements.” In The CRC Handbook of Mechanical Engineering, edited by F. Kreith, 4-182–4-205. Boca Raton, FL: CRC Press, 1998. 59. Hayhurst, A. N., and Kittelson, D. B. “Heat and Mass Transfer Considerations in the Use of Electrically Heated Thermocouples of Iridium versus an Iridium/Rhodium Alloy in Atmospheric Pressure Flames.” Combustion and Flame 28 (1977): 301–17. 60. Buhr, E., Haupt, G., and Kremer, H. “Heat Transfer from Impinging Turbulent Jet Flames to Plane Surfaces.” In Combustion Institute European Symposium 1973, edited by F. J. Weinberg, 607–12. New York: Academic Press, 1973. 61. Kremer, H., Buhr, E., and Haupt, R. “Heat Transfer from Turbulent Free-Jet Flames to Plane Surfaces.” In Heat Transfer in Flames, edited by N. H. Afgan and J. M. Beér, 463–72. Washington, DC: Scripta Book Company, 1974.
114
62. Bradley, D., and Matthews, K. J. “Measurements of High Gas Temperatures with Fine Wire Thermocouples.” Journal of Mechanical Engineering Science 10, no. 4 (1968): 299–305. 63. Nichols, E. L. “On the Temperature of the Acetylene Flame.” Physical Review 10 (1900): 234–52. 64. Fristrom, R. M., and Westenberg, A. A. Flame Structure. New York: McGraw-Hill, 1965. 65. Sparrow, E. M., and Cess, R. D. Radiation Heat Transfer. New York: McGraw-Hill, 1978. 66. Peterson, R. C., and Laurendeau, N. M. “The Emittance of Yttrium-Beryllium Oxide Thermocouple Coating.” Combustion and Flame 60 (1985): 279–84. 67. Hinze, J. O. Turbulence. New York: McGraw-Hill, 1959. 68. Son, S. F., Queiroz, M., and Wood, C. G. “Compensation of Thermocouples for Thermal Inertia Effects Using a Digital Deconvolution.” In Heat Transfer Phenomena in Radiation, Combustion, and Fires, edited by R. K. Shah, 515–22. New York: ASME HTD-Vol. 106, 1989. 69. Baukal, C. E., and Gebhart, B. “A Review of Flame Impingement Heat Transfer Studies—Part 2: Measurements.” Combustion Science & Technology 104, nos. 4–6 (1995): 359–85. 70. Horsley, M. E., Purvis, M. R. I., and Tariq, A. S. “Convective Heat Transfer From Laminar and Turbulent Premixed Flames.” Heat Transfer 1982, vol. 3, edited by U. Grigull, E. Hahne, K. Stephan, and J. Straub, 409–15. Washington, DC: Hemisphere 1982. 71. Kataoka, K., Shundoh, H., and Matsuo, H. “Convective Heat Transfer Between a Flat Plate and a Jet of Hot Gas Impinging On It.” In Drying ’84, edited by A. S. Mujumdar, 218–27. New York: Hemisphere/SpringerVerlag 1984. 72. You, H.-Z. “An Investigation of Fire-Plume Impingement on a Horizontal Ceiling: 2—Impingement and CeilingJet Regions.” Fire & Materials 9, no. L (1985): 46–56. 73. Hargrave, G. K., Fairweather, M., and Kilham, J. K. “Forced Convective Heat Transfer from Premixed Flames—Part 1: Flame Structure.” International Journal of Heat and Fluid Flow 8, no. 1 (1987): 55–63. 74. Hustad, J. E., Røkke, N. A., and Sønju, O. K. “Heat Transfer to Pipes Submerged in Lifted Buoyant Diffusion Flames.” In Experimental Heat Transfer, Fluid Mechanics, and Thermodynamics, 1991, edited by J. F. Keffer, 567–74. New York: Elsevier, 1991. 75. Hustad, J. E., and Sønju, O. K. “Heat Transfer to Pipes Submerged in Turbulent Jet Diffusion Flames.” In Heat Transfer in Radiating and Combusting Systems, 474–90. Berlin: Springer-Verlag, 1991. 76. Hustad, J. E., Jacobsen, M., and Sønju, O. K. Radiation and Heat Transfer in Oil/Propane Jet Diffusion Flames, Institution of Chemical Engineers Symposium Series 10, no. 129 (1992): 657–63. 77. Costa, M., Mourão, M., Baltasar, J., and Carvalho, M. G. “Combustion Measurements in an Industrial GlassMelting Furnace.” Journal of the Institute of Energy 69, no. 479 (1996): 80–86. 78. Mital, R., Gore, J. P., Viskanta, R., and Singh, S. “Radiation Efficiency and Structure of Flames Stabilized Inside Radiant Porous Ceramic Burners.” In Combustion and
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
Fire, edited by M. C. McQuay, W. Schreiber, E. Bigzadeh, K. Annamalai, D. Choudhury, and A. Runchal, 131–37. New York: ASME Proceedings of the 31st National Heat Transfer Conference, Vol. 6, ASME HTD-Vol. 328, 1996. 79. Fontes, P., Costa, M., and Azevedo, J. L. T. “Measurements in and Modeling of a Black Liquor Recovery Boiler.” Combustion Science and Technology 180, no. 3 (2008): 494–508. 80. Peterson, R. C. “Kinetics of Hydrogen-Oxygen-Argon and Hydrogen-Oxygen-Argon-Pyridine Combustion Using a Flat Flame Burner.” PhD thesis, Purdue University, 1981. 81. Candler, E. M. Thermocouple Reference Tables, Air Force Flight Dynamics Laboratory report no. AFFDL-TR-66178, Wright-Patterson Air Force Base, Ohio, 1967. 82. Blackburn, G. F., and Caldwell, F. R. Reference Tables for Thermocouples of Iridium-Rhodium Alloys Versus Iridium, J Research of National Bureau of Standards—C. Engineering and Instrumentation 68C, no. 1 (1964): 41–59. 83. Féry, C. “Sur la température des flammes.” Comptes Rendus de l’Académie des Science, no. 22 (1903): 909–12. 84. Strong, H. M., Bundy, F. P., and Larson, D. A. “Temperature Measurement on Complex Flames by Sodium Line Reversal and Sodium D Line Intensity Contour Studies” Third Symposium (International) on Combustion, 641–47. Baltimore, MD: The Williams & Wilkins Co., 1949. 85. Thomas, D. L. “Problems in Applying the Line Reversal Method of Temperature Measurement to Flames.” Combustion and Flame 12 (1968): 541–49. 86. Thomas, D. L. “An Automatic Remotely Operated Sodium D-Line Reversal Temperature Sensing Technique.” Combustion and Flame 12 (1968): 569–74. 87. Kilham, J. K. “Energy Transfer from Flame Gases to Solids.” Third Symposium on Combustion and Flame and Explosion Phenomena, 733–40. Baltimore, MD: The Williams and Wilkins Co., 1949. 88. Fells, I., and Harker, J. H. “An Investigation Into Heat Transfer from Unseeded Propane-Air Flames Augmented with D.C. Electrical Power.” Combustion and Flame 12 (1968): 587–96. 89. Nawaz, S. “Heat Transfer from Oxygen Enriched Methane Flames.” PhD thesis, The University of Leeds, Leeds, U.K., 1973. 90. Fairweather, M., Kilham, J. K., and Nawaz, S. “Stagnation Point Heat Transfer from Laminar, High Temperature Methane Flames.” International Journal of Heat and Fluid Flow 5, no. 1 (1984): 21–27. 91. Fairweather, M., Kilham, J. K., and Mohebi-Ashtiani, A. “Stagnation Point Heat Transfer from Turbulent MethaneAir Flames.” Combustion Science Technology 35 (1984): 225–38. 92. Young, K. J., Ireland, S. N., Melendez-Cervates, M. C., and Stones, R. “On the Systematic Error Associated with the Measurement of Temperature Using Acoustic Pyrometry in Combustion Products of Unknown Mixture.” Measurement Science and Technology 9 (1998): 1–5. 93. Muzio, L. J., Eskinazi, D., and Green, S. F. “Acoustic Pyrometry: New Boiler Diagnostic Tool.” Power Engineering 93, no. 11 (1989): 49–52.
Temperature
94. Kleppe, J. A. “Measurement of Combustion Gas Temperature Using Acoustic Pyrometry.” Proceedings of Instrumentation, Control, and Automation in the Power Industry 33 (1990): 25–40. 95. Bramanti, M., Salerno, E. A., Tonazzini, A., Pasini, S., and Gray, A. “An Acoustic Pyrometer System for Tomographic Thermal Imaging in Power Plant Boilers.” IEEE Transactions on Instrumentation and Measurement 45, no. 1 (1996): 159–67. 96. DiSimone, G. “Combustion Control.” Hydrocarbon Engineering 11, no. 10 (2006): 53–55. 97. DiSimone, G. “Furnace Combustion Management.” Petroleum Technology Quarterly 11, no. 4 (2006): 103–11. 98. Lu, J., Wakai, K., Takahashi, S., and Shimizu, S. “Acoustic Computer Tomographic Pyrometry for Two-Dimensional Measurement of Gases Taking into Account the Effect of Refraction of Sound Wave Paths.” Measurement Science and Technology 11 (2000): 692–97. 99. Young, A. M. “Infrared Temperature Measurement Essentials: Performance Characteristics, Calibration and Testing.” Industrial Heating LXV, no. 8 (1998): 45–60. 100. Dewitt, D. P., and Nutter, G. D., eds. Theory & Practice of Radiation Thermometry. New York: John Wiley & Sons, 1988.
© 2011 by Taylor and Francis Group, LLC
115
101. Patrick, D. K. “Infrared Thermal Image Recording and Analysis for Thermal Processes.” Industrial Heating LXV, no. 1 (1998): 33–35. 102. Lappe, V. “Installation and Maintenance of Infrared Ther mometers.” Industrial Heating LXIV, no. 11 (1997): 45–49. 103. Singh, P., Henneke, M., Jayakaran, J. D., Hayes, R., and Baukal, C. E. “Heat Transfer.” In The John Zink Combustion Handbook, edited by C. E. Baukal. Boca Raton, FL: CRC Press, 2001. 104. Hernberg, R., Stenberg, J., and Zethraeus, B. “Simultaneous In Situ Measurement of Temperature and Size of Burning Char Particles in a Fluidized Bed Furnace by Means of Fiberoptic Pyrometry.” Combustion and Flame 95, nos. 1–2 (1993): 191–205. 105. Stephan, K. D., Pearce, J. A., Wang, L., and Ryza, E. “Pros pects for Industrial Remote Temperature Sensing Using Microwave Radiometry.” Microwave Symposium Digest, 2004 IEEE MTT-S International, Vol. 2, pp. 651–54, 2004. 106. Stephan, K. D., Pearce, J. A., Wang, L., and Ryza, E. “Cement Kiln Temperature Measurements Using Microwave Radiometry.” IEEE MTT-S International Symposium, pp. 151–54, 2005. 107. Baukal, C. E., Gershtein, V. Y., and Li, X., eds. Computational Fluid Dynamics in Industrial Combustion. Boca Raton, FL: CRC Press, 2001.
6 Heat Flux Charles E. Baukal, Jr. Contents 6.1 Introduction.....................................................................................................................................................................117 6.2 Total Heat Flux............................................................................................................................................................... 120 6.2.1 Steady-State Uncooled Solids........................................................................................................................... 120 6.2.2 Steady-State Cooled Solids............................................................................................................................... 120 6.2.2.1 Single Cooling Circuit........................................................................................................................ 121 6.2.2.2 Multiple Cooling Circuits.................................................................................................................. 121 6.2.2.3 Surface Probe....................................................................................................................................... 121 6.2.3 Steady-State Cooled Gauges............................................................................................................................. 122 6.2.3.1 Gradient through Thin Solid Rod.................................................................................................... 122 6.2.3.2 Thin Disk Calorimeter....................................................................................................................... 123 6.2.3.3 Heat Flux Transducer......................................................................................................................... 124 6.2.4 Transient Uncooled Targets.............................................................................................................................. 124 6.2.5 Transient Uncooled Gauges.............................................................................................................................. 124 6.2.5.1 Slug Calorimeter................................................................................................................................. 125 6.2.5.2 Heat Flux Transducer......................................................................................................................... 125 6.3 Radiant Heat Flux.......................................................................................................................................................... 126 6.3.1 Heat Flux Gauge................................................................................................................................................. 126 6.3.2 Ellipsoidal Radiometer...................................................................................................................................... 130 6.3.3 Spectral Radiometer.......................................................................................................................................... 131 6.3.4 Other Techniques............................................................................................................................................... 132 6.4 Convective Heat Flux.................................................................................................................................................... 136 References................................................................................................................................................................................. 136
6.1 Introduction For most industrial heating applications, heat flux is one of the most important parameters in the system design [1]. The heat flux distribution in many furnaces is not uniform because of the combustion chamber geometry or asymmetrical firing of the burners [2]. The temperatures are generally over 1000°C (1800°F), which means that radiation is often an important if not dominant mode of heat transfer. The various types of fuels may impact the heat flux in the furnace. For example, the combustion of H2 produces very little gaseous radiation, while the combustion of oil or coal produces high levels of flame radiation due to soot formation. This means that in petrochemical applications where the fuel composition is usually highly variable, flame radiation may also be highly variable [3]. Some industrial applications, such as regeneratively fired glass furnaces, are highly transient in nature, which impacts the temporal heat flux in the
furnace. While many analytical and numerical models are available to predict heat flux, actual measurements are needed to validate those predictions. Measuring the heat flux in an industrial combustion process is both important and necessary. Heat flux to the load is closely related to the material processing rate, which is directly related to the firing rate (see Figure 6.1) [4]. Colannino [5] showed that heat flux can be correlated with furnace elevation (see Figure 6.2) using appropriate normalization techniques. An empirical model was developed that correlates heat flux as a function of fuel pressure, air preheat temperature, furnace bridgewall temperature, excess O2 in the exhaust, firing rate, H2 content in the fuel, furnace height, furnace width, heat absorption surface area, fuel port geometry, burner tile (quarl) geometry, amount of flue gas recirculation, number of fuel stages, and burner type. This data is critical for designing ethylene cracking furnaces, which illustrates the importance of measuring heat flux in industrial combustion applications. 117
© 2011 by Taylor and Francis Group, LLC
118
Industrial Combustion Testing
9
30
25
7 6
Furnace elevation (ft)
Peak heat flux (Btu/hr-ft2)
8
5 4 3 2
20
15 75% NG, 25% H2 25% NG, 25% H2
10
1 0
5 0
5 10 Heat release (MMBtuh)
15
Figure 6.1 Heat flux versus heat release for a typical burner used in ethylene furnaces. (From Colannino, J., Mathematical Models for Characterizing and Predicting Heat Flux Profiles from Ethylene Cracking Units, Proceedings of the 2007 AIChE Spring National Meeting, Houston, TX, April 23–26, 2007.) 1.0
Normalized heat flux
0.5 0.0 –0.5 –1.0 –1.5 –1.5
–1.0
–0.5
0.0
0.5
1.0
Normalized elevation Figure 6.2 Normalized heat flux versus elevation for nearly 1000 measurements in a pilot-scale test furnace. (From Colannino, J., Modeling of Combustion Systems: A Practical Approach, Boca Raton, FL, CRC Press, 2006.)
Some examples of heat flux measurements in industrial combustion processes will serve to illustrate the various methods that have been used that are described in some detail in this chapter. Hayes et al. [6] describe heat flux measurements for full-scale burners fired vertically upward along a radiant wall in pilot scale furnaces that simulate ethylene cracking furnaces. A water-cooled heat flux probe was inserted into the
© 2011 by Taylor and Francis Group, LLC
0
65
70
75 80 85 90 % of maximum heat flux
95
100
Figure 6.3 Effect of fuel composition on measured heat flux profiles in a f urnace for a burner firing at 3.5 × 106 Btu/hr. (From Hayes, R., Singh, P., Foote, D., and Baukal, C. E., Heat Flux from Process Burners, Proceedings of the ASME IMECE, New York, November 11–16, 2001, Heat Transfer Division, Vol. 369, No. 4, pp. 161–64, 2001.)
furnace at different vertical locations, at the centerline of the wall closest to the water-cooled tubes, between the central pair of tubes. The probe had an air-cooled window on the front to both protect the heat flux sensor from contamination and to exclude convective heat flux. Figure 6.3 shows heat flux profiles as a function of fuel composition for a natural draft burner mounted in the floor of the furnace and firing vertically upward along a wall at 3.5 × 106 (1.0 MW) Btu/hr with 2% O2 (dry) in the exhaust products. Analysis shows that the effect on the measured radiant heat flux profile is relatively small due to small variances in fuel compositions tested. Figure 6.4 shows the effect of the firing rate on heat flux profiles for a fuel blend with 78% natural gas (Tulsa) and 22% H2, with 2% O2 (dry) in the exhaust. All of the heat fluxes have been normalized to the maximum heat flux that occurred at 8.1 × 106 Btu/hr (2.4 MW) at an elevation slightly over 15 ft (4.6 m). As expected, there was a significant increase in the overall measured incident radiant heat flux at higher firing rates. Figure 6.5 shows the heat flux profiles as a function of excess O2 for a fuel blend with 83% H2 and 17% natural gas (Tulsa). Over the entire furnace, the higher excess O2 case produced lower heat flux compared to the lower excess O2 case. This is as expected since higher levels of excess air absorb additional heat, which reduces temperatures that decrease the overall radiant heat transfer from the hot gases and surfaces in the furnace.
119
Heat Flux
40
Furnace elevation (ft)
35 30 25 20 15 10.2 × 106 Btu/hr 11.8 × 106 Btu/hr
10 5 0
55
60
65
70 75 80 85 % of maximum heat flux
90
95
100
Figure 6.4 Effect of firing rate on measured heat flux profiles in a furnace. (From Hayes, R., Singh, P., Foote, D., and Baukal, C. E., Heat Flux from Process Burners, Proceedings of the ASME IMECE, New York, November 11–16, 2001, Heat Transfer Division, Vol. 369, No. 4, pp. 161–64, 2001.) 40
Furnace elevation (ft)
35 30 25 20 15 2% O2 4% O2
10 5 0
50
55
60
65 70 75 80 85 % of maximum heat flux
90
95
100
Figure 6.5 Effect of excess O2 on measured heat flux profiles in a furnace. (From Hayes, R., Singh, P., Foote, D., and Baukal, C. E., Heat Flux from Process Burners, Proceedings of the ASME IMECE, New York, November 11–16, 2001, Heat Transfer Division, Vol. 369, No. 4, pp. 161–64, 2001.)
Rafidi and Blasiak [7] measured local heat flux in a test furnace using four high temperature air combustion (HiTAC) regenerative burners (see Chapter 21). An ellipsoidal radiometer was used to measure irradiation (radiation heat flux incident per unit area falling on a surface). A plug type total heat flux meter was used to measure the rate of total heat flow (convection plus radiation). The convection heat flux was calculated by using an energy balance on the total heat flux measurements:
q = qr + qc = (αsGr − εσTs4) + hc(Tg − Ts), T
© 2011 by Taylor and Francis Group, LLC
(6.1)
where qT is the total heat flux received at the meter, qr is the net radiative heat flux, qc is the net convective heat flux, αs is the absorptivity of the plug surface, Gr is the measured irradiance, ε is the emissivity of the plug surface, σ is the Stefan–Boltzmann constant, Tg is the measured gas temperature, hc is the unknown convection coefficient, and Ts is the calculated surface temperature of the plug. The convection coefficient is the only unknown, which can be calculated using Equation 6.1. Hayes et al. [8] describe making radiation heat flux measurements in a production regenerative, flat-glass furnace. The measurements were taken at the crown (roof) during both the firing and nonfiring cycles, using both a hemispherical ellipsoidal radiometer and a circular foil heat flux gauge radiometer. Measured peak heat fluxes increased from 425 up to 710 kW/m2 during the firing cycle. These measurements showed the importance of transient conditions in certain applications. These measurements and some related studies [9,10] are discussed in more detail in Chapter 32 of this book. There are usually two types of heat flux that are important in industrial furnaces: radiative and total, which is the combination of radiation and convection. In some high temperature, low gas velocity combustors, the total and radiative fluxes may be nearly the same because of the predominance of radiation over convection. In lower temperature, high gas velocity combustors, the radiative flux may be much lower than the total flux where radiation and convection may be comparable in those applications. However, if any heat flux is measured, it is rare that more than one type would be measured. This type of information can be very valuable in understanding the dynamics of a given combustion system and is useful for making design improvements. Total, radiative, and convective heat flux measurements are discussed here. Hoogendoorn, Ballintijn, and Dorresteijn [11] describe the development and application of a heat flow meter capable of separately determining radiation and convection. The probe was used in a large-scale furnace fired with either gas or oil. The results showed a significant difference in the vertical heat flux profile with these two types of fuels where the profile was more uniform with gas and had a higher peak closer to the burners with oil. Goldstein and Chiang [12], Arai, Matsunami, and Churchill [1], and Okoh and Brown [13] discuss the theory of operation of the heat flux techniques discussed here. Bachmann, Chambers, and Giedt [14] experimentally studied how the installation of heat flux sensors embedded in a surface affects the actual heat transfer in the body. They concluded that proper sensor design can minimize the heat flux measurement errors caused by probe thermal resistance that can disturb the heat
120
transfer through the surface. Beasley and Figliola [15] mathematically analyzed heat flux probes embedded in surfaces for a range of probe and wall materials to study probe sensitivity and design. Ducharme, Frim, and Tikuisis [16] analytically and experimentally studied the measurement errors caused by the thermal resistance of the heat flux sensors and showed the errors can be very large in certain circumstances.
6.2 Total Heat Flux While total heat flux includes conduction, convection, and radiation, in most industrial applications the total heat flux is primarily radiation plus convection. Total heat flux has been measured using both steady-state and transient methods. Three different steady-state methods have been used. The first is to calculate the flux from an energy balance on an uncooled solid. The second is to measure the sensible energy gain of the coolant, for a cooled solid. The third is to directly measure the flux with a calibrated heat flux gauge imbedded in a cooled target. Two different transient methods have been used. In the first method, an uncooled solid is inserted into the flame for either a given amount of time or until a certain temperature is reached. The flux is calculated from the sensible energy gain of the solid. In the second method, the flux is measured using a gauge imbedded in the surface of an uncooled solid. Both of these methods have very short testing times. Both steady-state and transient methods are discussed next. 6.2.1 Steady-State Uncooled Solids In one version of this method, the heat flux is determined from an energy balance on the solid. It was used in two early flame impingement studies [17,18]. Refractory cylinders were coated with different oxides. These were heated until steady-state conditions were reached. Radiation from the cylinders, along with the gas and cylinder surface temperatures, were measured. The total heat flux to the cylinder was then calculated from an energy balance on the cylinder surface. The heat gained by the cylinder was assumed to be by convection and radiation from the flame. The energy lost by the cylinder was assumed to be solely due to radiation. A thermopile was used to measure radiation from the flame, both with and without the cylinder present. The surface radiation from the cylinder was calculated by subtracting the first measurement from the second. This value was assumed to be equal to the total heat flux from the flame to the cylinder.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
Another version of this technique is to measure the temperature gradient through a surface and use the steady-state, one-dimensional conduction equation to calculate the heat flux:
q ′′ = k
th − tl , l
(6.2)
where q” = heat flux per unit area (kW/m2), k = thermal conductivity (W/m-K), th = high temperature (K), tl = low temperature (K), and l = distance between th and tl. This equation assumes the thermal conductivity of the wall is constant. By measuring the temperature at two locations in the material and measuring the distance between the locations, the total heat flux can be simply calculated. It is recommended that a third temperature be measured in between to ensure that steady-state conditions have been achieved. Special thermocouple assemblies can be made that have multiple thermocouples in a single sheath that can be used to measure multiple temperatures at different locations through a material. The thermocouples can also be of different types. For example, a type R or S can be located in the hottest part and a type K or T could be located in the coolest part (see Chapter 5). One specific use of this arrangement is to measure the heat loss through a refractory wall in a furnace, where a single hole can be drilled through the wall, rather than multiple holes. This technique is very simple and of low cost. However, only the average heat flux over the entire solid is determined. It also relies on an accurate value of the target emissivity. It has only limited practical relevance and it is limited by the permissible maximum material temperature level of the solid. Also, the solid should be uniformly heated to obtain an accurate average measurement. Nonuniform heating complicates the energy balance calculations. In Kilham’s studies [17,18], the cylinders were rotated to minimize surface temperature gradients. However, only a single thermocouple, mounted on the inside diameter of the hollow cylinder to measure the inside cylinder temperature, was used to calculate the surface temperature. 6.2.2 Steady-State Cooled Solids In this method, a cooled solid is used as a calorimeter. It is commonly internally cooled with water or glycol. The flow rate and the temperature rise of the coolant are measured after steady-state conditions have been reached. The heat flux to the solid is calculated from
121
Heat Flux
the rate of sensible energy gain of the coolant. The three arrangements that have been used are: a single cooling circuit, multiple cooling circuits, and a surface probe. 6.2.2.1 Single Cooling Circuit In this technique, the average heat flux over the entire solid surface is calculated from the sensible energy gain of a single cooling circuit. Davies [19] and Fells and Harker [20] studied the effects of electrical boost on enhancing heat flux from impinging flames boosted by an AC discharge that produced as much as 3.6 times more heat flux than unboosted flames. The flames impinged on water-cooled stainless steel and copper pipes. Vizioz and Lowes [21] studied large-scale flames impinging on a large, circular, water-cooled plate with a refractory surface to simulate a hot target. The surface temperature was maintained using an imbedded spiral water-cooled channel. Davies [22] used water-cooled pipes to study heat flux from impinging flames on a cylinder as a function of equivalence ratio, firing rate, oxidizer/fuel mixing, and oxygen enrichment ratio. Posillico [23] studied flames impinging on a flat plate with a single internal cooling passage. 6.2.2.2 Multiple Cooling Circuits This technique has been used for flames impinging normal to the surface of a large metal disk, located inside a furnace. The surface temperature of the disk was maintained with concentric cooling channels. The heat flux to each annulus of the target was calculated from the sensible energy gain for each of the individual cooling circuits. Vizioz and Lowes [21] used a steel plate to simulate a cold target. The surface temperature was maintained below 373 K (212°F), with eight equally spaced concentric cooling channels. The temperature difference between each water inlet and outlet was only about 20–25 K (36–45°F) to reduce the radial heat transfer between the adjacent channels. Smith and Lowes [24] used a similar plate design. Air gaps between the channels were provided to further reduce radial conduction. Experiments included both air and water cooling. Rajani, Payne, and Michelfelder [25] used a smaller version of this plate. Baukal and Gebhart (1998) used a disk segmented into six concentric sections, as shown in Figure 6.6 [26]. The space between sections was insulated to minimize radial heat transfer between sections. Figure 6.7 shows contours of heat flux for impinging oxygen-natural gas flames as functions of firing rate and position (L = normalized axial distance from the burner exit and R = normalized radial distance from the centerline). As expected, the highest measured fluxes were at the highest firing rate, at the centerline, and close to the burner outlet.
© 2011 by Taylor and Francis Group, LLC
Figure 6.6 Circular target (before insulation) used in flame impingement studies. (From Baukal, C. E., and Gebhart, B., “Surface Condition Effects on Flame Impingement Heat Transfer,” Thermal and Fluid Science 15 (1997): 323–35.)
Figure 6.8 shows heat flux as a function of firing rate and axial distance from the burner. As expected, the highest fluxes were measured at the highest firing rate for the shortest distance from the burner. 6.2.2.3 Surface Probe The American Society for Testing and Materials (ASTM) standard E422 provides details for the construction of a single circuit, water-cooled calorimeter probe to measure heat flux [27]. Anderson and Stresino [28], Reed [29], and Fay [30] studied flames impinging on a so-called surface probe. In those studies, two identical, water-cooled copper plates were connected in series and separated by a thermal barrier. A single water-cooling circuit passed across both sections. The flame was traversed from the first to the second section. The water temperature rise through each plate was measured as a function of distance from the flame centerline. The data was inverted to determine the spatial heat flux distribution. See Reed [29] for details of the inversion calculation. This method is simple and of relatively low cost. It needs no calibration. The heat flux through the surface is measured directly by the sensible energy gain of the internal coolant. It has high accuracy. The spatial resolution depends on the number of separate cooling
122
Industrial Combustion Testing
1.0
Reff
0.8 0.6
1.0
Reff
500
0.6 0.4
700
0.2 1.0
300
0.6
0.2
500
400 15 kW
500
500
0.8 800
0.2 0
1
400
500
500
600 600
700
500
500
600
600
700
800 2
400
500
600
700
20 kW
400
500
600
700
900
400
500
600
600
0.4
300
400
500
400
500
1.0
0.6
400
500
800 900
0.2
300
600
600 700
0.4
400 10 kW
400
500
0.6
300
400
500
1.0 0.8
300
400 500
700
800
200
500
600
0.4
200
300
400
400 500
0.8
5 kW
300
400
600
100
200
300
300
0.8
Reff
200
400
0.2
Reff
200
300
0.4
Reff
100
200
3 L
25 kW 4
5
6
Figure 6.7 Contours of the total heat flux, q” (kW/m2), from oxygen-natural gas flames with various firing rates (q f = 5 to 25 kW) impinging on the surface of an untreated stainless surface. (From Baukal, C. E., and Gebhart, B., Experimental Thermal and Fluid Science, 16, no. 3, 247–59, 1998.)
channels used. However, finer spatial resolution requires more controls. The choices of target material and coolant are limited at higher surface temperatures. 6.2.3 Steady-State Cooled Gauges In this method, the local heat flux is determined using a small gauge imbedded in a much larger solid. The heat flux gauges are commonly made of a high thermal conductivity material, such as copper. The hot end of the gauge is exposed to the flame, while the cold end is water-cooled. Three different variations of this method have been used: a gradient through a thin solid rod, a thin disk calorimeter, and a heat flux transducer.
© 2011 by Taylor and Francis Group, LLC
6.2.3.1 Gradient through Thin Solid Rod The hot end of a thin solid rod is made flush with the target surface. Thermocouples, located along the probe axis, are used to determine the axial internal temperature gradient that is assumed to be linear. The heat flux is calculated using a one-dimensional conduction equation. In most cases, the probe has shields to minimize heat flux from the sides. Figure 6.9 shows an example of a portable, water-cooled total heat flux probe used for inserting into high temperature furnace environments. This type of probe, used by both Cookson and Kilham [31] and Kilham and Dunham [32] is described
123
Heat Flux
in detail by Cookson, Dunham, and Kilham [33]. It was a 4.8 mm (0.19 in.) o.d. copper rod, imbedded in a larger diameter cooled hemi-nosed brass cylinder. Beér and Chigier [34] used a 2.5 cm (1.0 in.) o.d. uncoated stainless steel probe. Buhr, Haupt, and Kremer [35] and Kremer,
1000 900 800
q" (kW/m2)
700 600 500 L = 0.5 L = 1.0 L = 2.0 L = 4.0 L = 6.0
400 300 200 100
Buhr, and Haupt [36] used a 3 mm (0.12 in.) o.d. copper rod imbedded in an internally water-cooled flat plate. Hoogendoorn, Popiel, and van der Meer [37] and Popiel van der Meer, and Hoogendoorn [38] used a 5.85 mm (0.23 in.) o.d. copper rod imbedded in a water-cooled flat plate. Hargrave and Kilham [39] and Hargrave, Fairweather, and Kilham [40] used a 3.2 mm (0.125 in.) o.d. copper rod internally cooled by glycol. The rod was imbedded inside a 22 mm o.d. hemi-nosed cylinder. Kataoka, Shundoh, and Matsuo [41] used a 10.21 mm (0.402 in.) o.d. copper rod imbedded in a water-cooled flat plate. Cassiano, Heitor, and Silva [42] used a probe of this type to measure the total heat flux in an end-port regenerative glass furnace. The probe consisted of cylindrical block inside a water-cooled housing that was required because of the high temperatures in the furnace. The front surface was serrated and blackened to maximize the radiant absorptivity. The probe was calibrated using a blackbody furnace. The response time of the probe was ~10 minutes with an accuracy of approximately 5%.
6.2.3.2 Thin Disk Calorimeter 5
10
15 q (kW) f
20
25
Figure 6.8 Total heat flux (q”) from oxygen-natural gas flames of various firing rates (q f) impinging on the stagnation point (Reff = 0.16) of an untreated stainless surface. (From Baukal, C. E., and Gebhart, B., Experimental Thermal and Fluid Science, 16, no. 3, 247–59, 1998.)
In this variation, a thin water-cooled disk, usually copper, is imbedded in a much larger target body. The heat flux to the calorimeter is calculated using the flow rate and the temperature rise of the water. Shorin and Pechurkin [43] used a 20 mm (0.8 in.) o.d. copper disk imbedded in a larger flat plate. The gauge and target were both independently water-cooled.
Water jacket Operating length Water outlet
Water inlet
Thermocouples
Guard rings Stainless steel plug
Cooling water
Cooling water
Figure 6.9 Water-cooled total heat flux probe. (From Baukal, C. E., The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001.)
© 2011 by Taylor and Francis Group, LLC
124
This technique is relatively low cost and needs no calibration. Accuracy depends on at least two temperature measurements and a coolant flow measurement. Response time may be slow depending on the design. 6.2.3.3 Heat Flux Transducer In this technique, an electrical signal is generated that is proportional to the heat flux. The transducer is often imbedded flush with a solid surface, such as the target in flame impingement heating or a furnace wall. One type of transducer uses an array of thermocouples to measure the heat flux across a thin distance. Schulte and Kohl [44] describe the water-cooled heat flux transducer used in Schulte’s [45] study. It was designed for surface flux rates up to 1000 kW/m2 (300,000 Btu/hr-ft2), using a single germanium crystal as the sensing element. Smith and Lowes [24] used a stainless steel disk with 15 spaced heat flux gauges, at eight equidistant radial locations. You [46] used water-cooled transducers whose surfaces were coated with either a gold or black foil, to determine either the convective or total heat flux, respectively. The transducers were imbedded flush with the surface of a water-cooled plate. Hager et al. [47] describe a thin film heat flux probe that can be attached to a surface, which minimizes the error caused by the thermal resistance of the sensor disturbing the heat flux through the surface. Another type of transducer is known as the Gardon [48] gauge. The basic configuration consists of a thin metal disk. Thermocouples are connected on the back of the disk, at the center, and at the perimeter. When heat flux is received at the front of the disk, a linear voltage is produced by the radial conduction from the center of the disk to the perimeter. This voltage is proportional to the heat flux. This gauge was originally developed for measuring high radiant fluxes. Its use has been extended to mixed convection and to radiation environments, using proper corrections. Grey [49] gives details on the construction of these probes. Ash [50] discusses the theory of the fast response of Gardon gauges. Borell and Diller [51] describe an apparatus for calibrating Gardon gauges in convective air flows. Van der Meer [52] used this type of gauge, imbedded flush with the surface of a water-cooled flat plate. The technique is simple and of relatively low cost. It has good accuracy and spatial resolution. The response time is fast enough that it can be used for feedback control systems [53]. There are some potential concerns with heat flux gauges. Calibration is required. This may be complicated in a mixed radiation and convection environment, since calibration typically requires a blackbody source. The maximum allowable temperature and heat flux, for some of the commercial transducers, may limit their use, for example, in high intensity flame impingement.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
6.2.4 Transient Uncooled Targets In this technique, the transient heat flux to a relatively large uncooled body is determined indirectly. The temperature profiles in the target are measured by an array of imbedded thermocouples, usually at or near the impingement surface. The flux is then calculated from the measured temperature responses, using inverse conduction heat transfer computational techniques [54]. Giedt, Cobb, and Russ [55] measured the transient heating in an ingot iron plate, whose surface was parallel to the flame. Measurements were made until the plate reached 1170 K (1650°F). This usually took about 21 seconds. Woodruff and Giedt [56] used a blunt-nosed molybdenum plate to simulate an air foil. Test durations were about 50 seconds. Milson and Chigier [57] used a flat, uncooled plate. The back of the plate was coated with lamp black, to create a surface boundary condition having a well-defined emissivity. The heat flow through the plate was equated to the natural convection and radiation losses from the back side. Veldman, Kubota, and Zukoski [58] used a thin steel disk target with an array of thermocouples welded onto the back side. Matsuo et al. [59] heated rectangular slabs of varying thicknesses, located inside a furnace. Rauenzahn [60] used a thick copper plate, which remained nearly isothermal during the heating process. Gatowski, Smith, and Alkidas [61] describe the use of a thermocouple embedded in the wall of an internal combustion engine to measure the heat flux to the wall by using the transient surface temperature measurements. This technique is very simple and of low cost. It does not need calibration. It commonly suffers from poor spatial resolution. It requires very rapid instrument response times. This may cause difficulties if many measurements must be frequently recorded in a very short time period. This technique also relies on accurate thermal conductivity values, as a function of temperature.
6.2.5 Transient Uncooled Gauges In this technique, a small heat flux gauge is imbedded flush with the target. The gauge is commonly made of a high thermal conductivity material. It is assumed to be at a uniform temperature level at all times. The sensible heat gained by the gauge may be simply calculated. Hornbaker and Rall [62] list the ideal characteristics for this kind of gauge in high temperature applications as:
1. the measured flux should equal that transferred to the target 2. the flux that would have been received by the target, with no gauge, should not be changed 3. the gauge must have a rapid temporal response
125
Heat Flux
4. the gauge output should be directly proportional to the flux 5. the gauge should be small, simple, and ruggedly constructed
Bachmann, Chambers, and Giedt [14] discuss the errors using gauges caused by (1) conduction between the gauge and target and (2) by perturbation of the boundary layer resulting from the temperature discontinuity at the gauge-target interface. Bachmann used copper and nickel slug calorimeters to measure the heat flux from O2/C2H2 flames flowing parallel to a flat plate. The gauge heat flux was 40–140% higher than to a calibrated reference calorimeter, as a result of the second error type above. Two different variations of this method have been used: the slug calorimeter and heat flux transducer. 6.2.5.1 Slug Calorimeter In this technique, the slug thermal capacity is assumed to be internally isothermal, at any instant in time. This occurs when the thermal conductivity is large compared to the convection heat transfer coefficient. The heat flux to the slug is calculated from the sensible energy gain of the slug. The average flux is usually calculated over a temperature range during which the energy gain was linear. Kilham and Purvis [63], Nawaz [64], Fairweather, Kilham, and Nawaz [65], and Fairweather, Kilham, and Mohebi-Ashtiani [66] used a 2.18 mm (0.086 in.) o.d. stainless steel conductivity plug, located inside a hemi-nosed cylinder. The target was inserted into the flame using a solenoid coil. The insertion time was estimated by Nawaz to be 0.04 seconds. This was compared to the residence time in the flame, which was estimated to be 0.15–0.25 seconds. Nawaz used a hollow stainless steel cylinder. The conductivity plug was a 0.26 mm thick plate, used to close one end of the cylinder. A thermocouple was welded on the inside surface of the plug, to measure the energy gain. The experimental results of Fairweather, Kilham, and Mohebi-Ashtiani [66] showed large heat fluxes near the laminar flame reaction zone due to the large and nonequilibrium concentration of reactive species. Lower peaks in flux were measured for turbulent flames. This was due to the more diffuse mixing in the reaction zone. Conolly and Davies [67] used an assembly of a cylindrical plug (made of a copper/1% beryllium alloy) and a rear-mounted thermocouple. Kremer, Buhr, and Haupt [36] used a 3 mm o.d. by 1.5 mm (0.059 in.) thick copper disk. The disk was insulated from the rest of the target with a ceramic, to reduce radial losses. It was mounted flush to the surface of a flat plate. Kilham and Purvis [68] used 12 mm (0.47 in.) o.d. high thermal conductivity disks, with thicknesses of 1, 2, or 3 mm (0.04, 0.08 or 0.12 in.). These were mounted flush with the surface
© 2011 by Taylor and Francis Group, LLC
of a large flat plate. They were made of either silicon carbide or platinum, to test the importance of hydrogen atom recombination. No difference in heat flux was measured using the two materials. Ivernel and Vernotte [69] used 15 mm (0.59 in.) o.d. by 2 mm (0.08 in.) thick platinum disks. The temperature rise of the disk was measured with a thermocouple attached to the back. This rise was used to calculate the total heat flux, based on the sensible energy gain of the disk. Horsley, Purvis, and Tariq [70] and Hemeson et al. [71] used a calorimeter consisting of a small oxygen-free high conductivity (OFHC) copper disk with a thermocouple attached on the back face. Horsley estimated that the exposure time for the probe varied from 0.32 to 2.0 seconds. The mean surface temperature was 360 K (190°F). The sensor probe was located in a cylinder that was inserted by a piston normal to the surface of a water-cooled plate. Hemeson used a 4.44 mm (0.175 in.) o.d. by 1.57 mm thick (0.0618 in.) copper disk imbedded flush and contoured to the surface of both cylinders and hemispheres. Hargrave, Fairweather, and Kilham [40] used a 3.2 mm (0.125 in.) o.d. by 1.0 mm (0.04 in.) thick copper slug, located at the stagnation point of a solid brass cylinder. Hustad, Røkke, and Sønju [72], Hustad and Sønju [73], and Hustad, Jacobsen, and Sønju [74] used a probe consisting of a 25 mm (1 in.) o.d. by 3 mm (0.12 in.) thick flat copper disk. It was imbedded flush with the surface of a round steel pipe, at the forward stagnation point of the flow. This technique is relatively simple in principle and needs no calibration. However, fast-acting insertion equipment and accurate temporal temperature measurements and material properties are needed for this technique. 6.2.5.2 Heat Flux Transducer In this method, an uncooled transducer is used to measure heat flux. Hornbaker and Rall [62] measured the heat flux from O2/C2H2 flames, parallel to a flat plate, using both the transient uncooled target and the uncooled gauge techniques. The first technique was assumed to indicate the actual heat flux. Comparison with the second technique showed that the gauges overpredicted, matched, and underpredicted the true flux, respectively, at surface temperatures of about 300–550, 550–600, and 600 K–900 K (80–530, 530–620, 620°F –1200°F). The transient uncooled gauge technique is relatively simple. It simulates many industrial heating processes, such as rapid metal heating. However, the instruments must have very fast response times. Some mechanism must be provided for rapidly inserting the target into the flame. For high temperature flames, the response time must be fractions of a second, to prevent the transducer from melting.
126
Industrial Combustion Testing
6.3 Radiant Heat Flux Gray and Müller [75] note there are four components for most instruments used to measure radiation: (1) a detector that converts the radiation to an electrical signal, (2) an optical system that directs the radiation to the detector that may include some type of focusing element, (3) a filtering system that ensures only the proper wavelength band of radiation required for the detector is received by the detector, and (4) an amplifier to enhance the signal and some type of recorder/indicator to display the output. In the following sections, only components one and three are considered. It has been assumed the optical system and amplifier can be properly designed for a given detector. Note that radiation from flares is covered in Chapter 30 of this book and is therefore not included here. Many of the devices used to measure radiation have a highly absorptive black coating. This coating is usually very thin to minimize its effect on reducing the conduction through the coating that would affect the measured heat flux. One of the challenges in calibrating these absorptive sensors is determining the thickness of the blackened coatings. Stradomskiy, Maksimov, and Malyarov [76] describe a method for determining the coating thickness for these probes so they can be properly calibrated. 6.3.1 Heat Flux Gauge There are three general types of radiant emission that may arise in combustion processes. One is nonluminous gaseous radiation (see Figure 6.10). This results from the well-characterized spectral emission characteristics of both H2O (see Figure 6.11) and CO2 (see Figure 6.12) [77]. These are two of the main constituents in the products of combustion, for hydrocarbon fuels. Even at high flame temperatures, nonluminous radiation is usually small compared to convection, since the emissivity of the gases is very low [78]. The second type of radiation involves luminous emission from particles. Such sooting flames (see Figure 6.13) often generate very high radiation fluxes. This effect is usually only important for liquid and solid fuels, but not for gases that normally produce very little if any soot unless the flames are very fuel rich. The third effect is radiation from solid surfaces, such as from refractory walls inside a furnace (see Figure 6.14). Such radiation depends on the geometry, temperature, and radiant properties of the walls. In some studies, several of these types of radiation have been important. The techniques used to measure thermal radiation are discussed below. One method that has been used to measure nonluminous radiation is to directly measure the radiation from free jet flames with a radiometer. An elevation and plan view of a radiometer aimed at a flame are shown in Figure 6.15 and Figure 6.16, respectively. In one study, a
© 2011 by Taylor and Francis Group, LLC
Figure 6.10 (See color insert following page 424.) Example of a nonluminous flame. (Courtesy of John Zink Co. LLC.)
Medtherm Corp. (Huntsville, AL) model 40-5-18T heat flux transducer was used to measure the radiation [79]. The transducer was a Gardon gauge [48] with a diameter of 6.35 mm (0.25 in.). The sensor absorptance was 92%, in the spectral range of 0.6 to 15.0 µm. A model 40VRW-7C water-cooled view restrictor was used to restrict the field of view to θradm = 6.45°. A narrow angle radiometer was used to ensure that only radiation from the flame, not from the ambient environment, was measured. This was representative of the case for an impinging flame, where the target was engulfed by the hot combustion gases. The radiometer did not have any windows or gas purging, since the ambient environment was clean. Figure 6.17 shows the measured flame radiation as a function of axial distance from the burner outlet and firing rate. The graph shows that radiation initially increases with distance and then decreases at differing rates depending on the firing rate Figure 6.18 shows the radiation from a pure O2/natural gas flame was approximately double that of an oxygen-enriched air/natural gas flame. Gore et al. [80] used a Medtherm wide angle (150°) radiometer to measure luminous radiation from methane/heptane and methane/crude oil flames. The measured radiant fluxes for the methane/heptane flames
127
Heat Flux
1
L = 2000 cm L = 200 cm L = 20 cm
T = 600 K p = 1 atm pH O = 0.5 atm 2
L = 2 cm 0 1 Absorptivity
L = 2000 cm L = 200 cm L = 20 cm L = 2 cm
T = 1200 K p = 1 atm pH O = 0.5 atm 2
0 L = 2000 cm 1 L = 2 cm
Relative blackbody radiance
0
1000
2000
L = 20 cm
3000
T = 2400 K p = 1 atm pH O = 0.5 atm 2
L = 200 cm
6000 5000 4000 Wavenumber (cm–1)
7000
8000
9000
10000
1 2400 K 1200 K 600 K
0
Figure 6.11 Calculated spectral emissivity of H2O as a function of wavelength and path length. (From Ludwig, C. B., Malkmus, W., Reardon, J. E., and Thomson, J. A. L., Handbook of Infrared Radiation from Combustion Gases, Washington, DC: National Aeronautics and Space Administration report SP-3080, 1973.)
and for the methane/crude oil flames ranged from 0.34 to 0.64 and 0.53 to 0.68 W/cm2 (1100–2000 and 1700–2200 Btu/hr-ft2). The modeling results significantly underpredicted the experimental data due to lower estimates of temperature, the complexity of predicting soot in turbulent flow, and turbulent radiation interactions. Shorin and Pechurkin [43] used a steady-state cooled gauge to measure the total heat flux from impinging flames. The gauge was imbedded flush with the surface of a large plane surface. To determine the importance of radiation, gauges with polished, oxidized, and blackened surfaces were tested. The lowest flux was expected to arise for the polished gauge, which reflects most of the radiation. Higher fluxes were expected with the oxidized and blackened gauges. These are better absorbers. However, no difference in measured heat flux was found.
© 2011 by Taylor and Francis Group, LLC
It was concluded that radiation was not important for their experimental conditions. You [46] measured flame radiation to the surroundings, using commercial heat flux gauges. The radiant flux to a plane surface was also indirectly measured. For these measurements, commercial heat flux gauges were coated with either gold or black foil, to measure the convective or total heat flux, respectively. The radiant flux was determined by subtracting the convective flux from the total. Thermal radiation was 13–26% of the total flux, near the stagnation point. It was negligible at the edge of the plate (R = 7.3). Heat flux gauges are often used as radiometers by putting a glass window in front of them to prevent convective heat transfer to the gauge. A variety of materials may be used for the infrared detectors, each having a specific spectral characteristic as shown in Figure 6.19 [81].
128
Industrial Combustion Testing
L = 200 cm
L = 2000 cm
1
T = 600 K
L = 20 cm
p = 1 atm pCO = 0.5 atm 2
L = 2 cm
L = 2 cm
L = 2000 cm
Absorptivity
1
L = 200 cm
0
T = 1200 K p = 1 atm L = 20 cm L = 2 cm
pCO = 0.5 atm
L = 2000 cm
T = 2400 K
2
0 1
L = 2 cm
L = 200 cm
p = 1 atm
L = 20 cm
pCO = 0.5 atm 2
L = 2 cm
Relative blackbody radiance
0
1000
2000
3000
5000 6000 4000 Wavenumber (cm–1)
7000
8000
9000
10000
1 2400 K 1200 K 600 K
0
Figure 6.12 Calculated spectral emissivity of CO2 as a function of wavelength and path length. (From Ludwig, C. B., Malkmus, W., Reardon, J. E., and Thomson, J. A. L., Handbook of Infrared Radiation from Combustion Gases, Washington, DC: National Aeronautics and Space Administration report SP-3080, 1973.)
Matthews, Harris, and Garcia [82] describe a multihead transpiration radiometer used to make radiative heat flux measurements in a sooty pool fire. The transpiration radiometer [83] uses a transducer to measure the radiation and the transpiration is used to eliminate the effects of convection. The multihead version measured radiation from five different directions in the flame and worked successfully in a sooty environment. The main drawback was insensitivity at low flux rates. Brajuskovic, Matovic, and Afgan [84] and Brajuskovic, and Afgan [85] describe a blow-off heat flux sensor that uses gas flowing through a porous sintered metal disc to measure hemispherical radiation heat flux in a dirty environment that includes high particulate loads. The gas acts to blow off the boundary layer at the surface. It also cools the instrument, prevents fouling of the porous
© 2011 by Taylor and Francis Group, LLC
surface that can particularly be a problem when gaseous or liquid fuels are used that can produce soot, and is used to distinguish between convection and radiation heat flux. Martins et al. [86,87] describe the calibration of such a blow-off heat flux sensor. Ploteau, Glouannec, and Noel (2007) developed a thermoelectric heat flux meter for use in measuring radiation in industrial furnaces [88]. Three versions were developed: water-cooled for a constant reference temperature, uncooled with a paraffin buffer to maintain a constant reference temperature for lower temperature environments, and floating reference temperature with two thermoelectric modules. The cooled version can handle the highest temperatures. Heat flux gauges need to be properly calibrated to ensure the measurements are accurate. Gauges
129
Heat Flux
Flame Radiometer
lradm Burner
Figure 6.15 Elevation view of the radiometer and the flame. (From Baukal, C. E., and Gebhart, B., International Journal of Heat and Mass Transfer, 40, no. 11, 2539–47, 1997.)
Flame Radiometer
θradm dradm df
View restrictor
Figure 6.13 (See color insert following page 424.) Example of a luminous flame. (Courtesy of John Zink Co. LLC.)
Figure 6.16 Plan view of a radiometer and a flame. (From Baukal, C. E., and Gebhart, B., International Journal of Heat and Mass Transfer, 40, no. 11, 2539–47, 1997.)
70
q”rad (kW/m2)
60
50
40
q = 5.00 kW f qf = 15.0 kW
30
qf = 25.0 kW Figure 6.14 (See color insert following page 424.) Surface radiation in a furnace.
purchased from vendors normally come with a calibration curve. It is helpful to have a blackbody calibrator (see Figure 6.20) to ensure the calibration curve is still valid after the gauge has been in use for some time or something has happened to the gauge (e.g., it has been
© 2011 by Taylor and Francis Group, LLC
20
0
1
2
3
Lradm
4
5
6
7
Figure 6.17 Thermal radiation (q″rad) versus nondimensional axial distance from the burner exit (Lr) and firing rate (q f) (5.0, 15.0, 25.0 kW) for stoichiometric oxygen-natural gas flames. (From Baukal, C. E., and Gebhart, B., International Journal of Heat and Mass Transfer, 40, no. 11, 2539–47, 1997.)
130
Industrial Combustion Testing
70
q”rad (kW/m2)
60 50 40 30 20 10
Ω = 0.35 Ω = 1.00 1
0
2
3
Lradm
4
5
6
7
Figure 6.18 Thermal radiation (q″rad) versus axial distance from the burner outlet (Lr) and oxidizer O2 content of 35% and 100% (Ω = 0.35 and 1.00) for a firing rate natural gas at 15.0 kW. (From Baukal, C. E., and Gebhart, B., International Journal of Heat and Mass Transfer, 40, no. 11, 2539–47, 1997.) Figure 6.20 Blackbody calibration furnace.
1014
The curve can be adjusted if necessary to match the blackbody heat flux from the furnace. If no calibrator is available, the gauge can be sent back to the vendor for recalibration.
Ge 77 K 1013
1012
D*(λ)
Ideal photovoltaic (BLIP limit)
Ge–30°C
Ideal photoconductor (BLIP limit) Ge 22°C
1011
InAs 77 K
InSb 77 K HgCdTe–55°C
1010
6.3.2 Ellipsoidal Radiometer
HgCdTe 77 K HgCdTe 77 K
InAS–30°C
HgCdTe 77 K
Doped Ge 4 K
109 InAS 22°C 8
10
1
2
3
4
HgCdTe–75°C 5 6 7 8 10 Wavelength (μm)
12
20
30
Figure 6.19 Spectral characteristics of infrared detectors. (From Lide, D. R., CRC Handbook of Chemistry and Physics, 79th ed., Boca Raton, FL: CRC Press, 10–216, 1998.)
dropped) that may affect the calibration. The calibrator can be set to different temperatures, which means the radiation is known because the furnace acts like a blackbody. The probe is inserted into the calibrator and the known blackbody radiation from the furnace can be compared against the calibration curve for the gauge.
© 2011 by Taylor and Francis Group, LLC
This probe (see Figure 6.21) consists of an ellipsoidal cavity, internally coated with a highly reflective metal, such as polished gold. Radiation enters the cavity through a small hole, located at one of the foci of the ellipsoid cavity. The radiation is reflected inside the ellipsoid, onto a thermopile, located at the other focal point. The thermopile converts the radiant flux to an electrical signal. The radiometer is commonly calibrated against a blackbody furnace (see Figure 6.20) having a known internal surface temperature. Several flame impingement studies ([21,25,34]) measured total radiant emission with hollow, gold-plated, ellipsoidal radiometers. These were developed by the International Flame Research Foundation (IFRF) in IJmuiden, the Netherlands [89]. Measurements at the IFRF indicated that radiation may account for as much as half of the total heat flux to a target located in a hot furnace. The radiation was primarily from the hot refractory walls. Vizioz attempted to isolate this radiation from the hot combustion gases. The radiation reflected and emitted from a flat plate target was measured with a narrow angle radiometer.
131
Heat Flux
Water jacket Operating length
Water outlet
Water inlet
Nitrogen purge Sensing pellet
Cooling water Constantan wires
Mirrored ellipsoidal cavity
Cooling water
Figure 6.21 Drawing of an ellipsoidal radiometer. (From Baukal, C. E., The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001.)
This consisted of a thermopile with a window in front of it, to eliminate forced convection. The window and thermopile were arranged so that only radiation from certain angles is intercepted. The radiated flux from the plate was assumed to be a constant fraction of the incident radiation. The absorption and emission of the gas layer, between the plate and the radiometer, were estimated. Smith and Lowes [24] were unable to obtain consistent results using this ellipsoidal radiometer. Butler and Webb [90] reported wall radiant heat flux measurements, ranging from 100 to 500 kW/m2 (30,000 to 160,000 Btu/ft2-hr), in an 80 MWe (2.7 × 108 Btu/hr) industrial coal-fired boiler using an ellipsoidal radiometer. Average wall radiant heat fluxes were measured at six elevations in the boiler. In experiments located outdoors at ambient conditions, Hustad, Røkke, and Sønju [72] measured radiation perpendicular to the flame axis using a wide angle radiometer. The measurements were made at the middle of the flame height, where soot concentration and, therefore thermal radiation, were at their peaks. It was found that the radiation was higher than the model predictions. It was concluded that the measurement location was not representative of the entire flame. The calculated values were used, instead of the actual measurements, in subsequent analyses. Hustad [74] found that radiation perpendicular to the flame axis was 26%, 40%, and 44% of the total heat flux for CH , C H , and C H , respectively at Ma = 1. For C H , 4 4 10 3 8 3 8 at Ma = 0.6, radiation was 26% of the total flux. These results show that luminous radiation may be important, even for gaseous fuels. Beltagui, Kenbar, and Maccallum [91] used both a total heat flux probe and calorimetry to
© 2011 by Taylor and Francis Group, LLC
measure the total heat flux in a furnace and a radiometer to measure the radiant heat flux in the furnace. They studied the characteristics of swirling flames fired vertically into a water-cooled chamber 3 m (10 ft) long and 1 m (3 ft) i.d., divided into six sections for calorimetry. The firing rate was 400 kW (1.4 × 106 Btu/hr) on natural gas with equivalence ratios ranging from 0.70 to 1.35 and 5% excess air. The total heat flux probe measurements were in good agreement with the calorimetry calculations for each of the six sections. The radiation measurements showed that the radiant heat flux was at least half of the total flux and in some cases nearly accounted for all of the heat flux. The measurements also showed that the heat flux distribution was much more uniform for the case with no swirl and that the peak flux 70 kW/m2 (22,000 Btu/ft2-hr) that occurred in the second section downstream from the burner for the case with high swirl (S = 2.25) was almost double the peak flux 40 kW/m2 (13,000 Btu/ft2-hr) for the case without swirl. Radiometers are generally low cost and simple to use. They have good spatial resolution. They operate successfully at high temperatures. However, they must be calibrated. Ellipsoidal radiometers are difficult to use in very sooty or in particle-laden environments that may foul the polished reflectors. 6.3.3 Spectral Radiometer Ji and Baukal [92] studied the spectral radiation from oxygen-enhanced flames. Spectral radiation was measured (see Figure 6.22) with a spectral radiometer from Monolight Instruments (now known as Irradian Ltd. in East Lothian, U.K.). Three scanning monochromator
132
Industrial Combustion Testing
3.5 m
Spectrometers 0.5 m Burner Side view
UV-VIS 0.5 m
NIR MIR Spectrometers Figure 6.22 Experimental apparatus for measuring spectral flame radiation.
Top view
Figure 6.23 Side view and top view of multispectrometer setup.
7
Irradiance (W m–2 μm–1)
6 φ = 1.00
5
φ = 2.00
φ = 0.67
4 3 2 1 0 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 4.00 4.25 4.50 4.75 5.00 Wavelength λ (μm)
Figure 6.24 Spectral flame radiation measurements for natural gas (5 kW) combusted with pure O2 contents for different equivalence ratios. (From Ji, B., and Baukal, C. E., “Spectral Radiation Properties of Oxygen-Enhanced/Natural Gas Flames,” Proceedings of 1998 International Gas Research Conference, Vol. 5, 422–33, November 8–11, 1998, San Diego, CA, 1998.)
modules with appropriate sets of order-sorting filters were used to cover the spectral range from 0.25 µm to 5.0 µm (see Figure 6.23). The spectral irradiance calibration was made with a quartz tungsten halogen lamp (Oriel Model 6333, Stratford, Connecticut) in the 0.25–1.7 µm range and with an IR emitter (Oriel Model 6580) in the 1.7–5.0 µm range. Figure 6.24 shows how the measured spectral radiation for an oxy-natural gas flame varied with equivalence ratio. Stoichiometric flames (ϕ = 1.00) had the most pronounced peaks. Fuel rich flames (ϕ = 2.00) had lower peaks but more broadband
© 2011 by Taylor and Francis Group, LLC
radiation. Fuel lean flames (ϕ = 0.67) were generally in between. Figure 6.25 shows that flame radiation increased as the O2 content in the oxidizer increased, where oxy-fuel produced significantly more radiation than air–fuel combustion. 6.3.4 Other Techniques In the Ji and Baukal [92] study referred to above, spectral radiometers and the properties of molten glass were used to determine the optimal wavelength range for
133
Heat Flux
8 7
Irradiance (W m–2 µm–1)
6 5 100% O2
4
60% O2
3 2 21% O2
1
0 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 4.00 4.25 4.50 4.75 5.00 5.25 Wavelength λ (µm)
Figure 6.25 Spectral flame radiation measurements for natural gas (5 kW) stoichiometrically combusted with oxidizers with variable O2 contents (21%, 60%, 100%). (From Ji, B., and Baukal, C. E., “Spectral Radiation Properties of Oxygen-Enhanced/Natural Gas Flames,” Proceedings of 1998 International Gas Research Conference, edited by D. Dolenc, Vol. 5, 422–33, November 8–11, 1998, San Diego, CA, 1998.)
thermal radiation to penetrate into the glass. However, that method is time-consuming and expensive. A more convenient and reliable way to determine the penetrating radiation is desirable. While the scanning spectral radiometer employed in that work provides quantitative measurements of the penetrating radiation, it requires spectral irradiance calibration, is time consuming, and may not be as conveniently applied to large-scale flames in industrial furnaces. A more convenient way is to use total radiometers with appropriate spectral filters. Total (heat flux) radiometers have a flat spectral response in the 0.6–15.0 µm region. Without any window or filter in front of it, a heat flux meter gives the integrated total radiation intensity over the entire 0.6–15.0 µm range. Since one of the most important heat transfer mechanisms in a glass furnace is absorption of the flame radiation by the glass melt, one must consider how flame radiation propagates through the glass melt over the entire spectral range. Several studies on high temperature glass melt absorption spectra [93–95] show that the spectral absorption coefficient increases dramatically at about 2.7 µm. The glass melt essentially becomes opaque at longer wavelengths due to strong absorption from the O–H and Si–O bonds in silicate. This is a universal feature of molten glass. The spectral absorption at wavelengths shorter than 2.7 µm varies with the impurities and/or additives in the glass melt, the temperature, and the furnace atmosphere [93–95]. For low iron (FeO) content molten glass without other impurities or additives, the spectral absorption at wavelengths
© 2011 by Taylor and Francis Group, LLC
shorter than 2.7 µm is relatively small and constant [96]. The spectral absorption data of such low iron content glass melts [96] has been used for the case study in this work. Figure 6.26 shows the spectral absorption coefficient K of low iron content molten glass as reported by Berg [96]. Also plotted in Figure 6.26 are the spectral radiation intensity of a flame with Ω = 1.0, ϕ = 2.0, q f = 5 kW, and the intensity after the radiation propagates through 1 cm (0.4 in.) of molten glass, respectively. It is evident from Figure 6.26 that only radiation at wavelengths λ < 2.7 µm has significant penetration into molten glass. Since glass is a poor heat conductor, the radiation energy at λ > 2.7 µm only heats the top surface of the glass, which can cause overheating that may reduce the glass quality. Therefore, only the penetrating (λ < 2.7 µm) radiation is the effective radiation for molten glass. In order to improve heat transfer efficiency and reduce fuel consumption in glass furnaces, it is important that the penetrating is maximized, rather than the total radiation intensity. To compare the penetrating radiation intensity among the various flames, the spectral radiation was numerically integrated for λ < 2.7 µm. The results are summarized in Table 6.1. For the penetrating radiation, the pure oxygen/natural gas flame produced five times more radiation than the air-fired flame, for the same fuel consumption and stoichiometry. Taking this into account, the fuel savings using oxygen enrichment is more pronounced for glass furnaces than the savings based only in terms of the total radiation. For oxygen/natural gas flames
134
Industrial Combustion Testing
7
18
Radiation intensity (W m–2 µm–1)
6
16
5
I at melt surface
14 12
4
10
I at 1 cm below surface
3 2
K
8 6
K
4
1
2
Molten glass spectral absorption coefficient K (cm–1)
20
0 0 0.25 0.5 0.75 1 1.25 1.5 1.75 2 2.25 2.5 2.75 3 3.25 3.5 3.75 4 4.25 4.5 4.75 5 5.25 Wavelength λ (mm)
Figure 6.26 Interaction of flame radiation with high temperature glass melt. (From Ji, B., and Baukal, C. E., “Spectral Radiation Properties of OxygenEnhanced/Natural Gas Flames,” Proceedings of 1998 International Gas Research Conference, edited by D. Dolenc, Vol. 5, 422–33, November 8–11, 1998, San Diego, CA, 1998.)
Table 6.1 Flame Radiation Intensity Measurements Ω 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.60 0.21
φ
Inrm(total)
Inrm(pen.)
Inrm(BK7)
γ
0.67 0.80 1.00 1.33 1.50 1.60 1.79 1.89 2.00 1.00 1.00
0.90 ± 0.06 0.99 ± 0.07 1.00*
0.89 ± 0.06 0.98 ± 0.07 1.00*
0.82 ± 0.06 0.95 ± 0.07 1.00*
0.93 ± 0.07 0.91 ± 0.06 0.84 ± 0.06 0.80 ± 0.06 0.75 ± 0.05 0.80 ± 0.06 0.73 ± 0.05 0.26 ± 0.02
0.98 ± 0.07 0.93 ± 0.07 0.83 ± 0.06 0.83 ± 0.06 0.87 ± 0.06 1.12 ± 0.08 0.70 ± 0.05 0.19 ± 0.01
0.99 ± 0.07 0.87 ± 0.06 0.88 ± 0.06 0.83 ± 0.06 0.96 ± 0.07 1.10 ± 0.08 0.75 ± 0.05 0.19 ± 0.01
0.15 ± 0.01 0.13 ± 0.01 0.14 ± 0.01 0.17 ± 0.01 0.20 ± 0.01 0.21 ± 0.01 0.37 ± 0.03 0.55 ± 0.04 1.02 ± 0.07 0.12 ± 0.01 0.005 ± 0.001
Source: From Ji, B., and Baukal, C. E., “Spectral Radiation Properties of Oxygen-Enhanced/Natural Gas Flames,” Proceedings of 1998 International Gas Research Conference, Vol. 5, 422–33, November 8–11, 1998, San Diego, CA, 1998. * The Ω = 1.0, and ϕ = 1.0 intensities were used as the reference in the normalization.
with different fuel equivalence ratios, a fuel rich flame (ϕ = 2.00) apparently produced less total radiation than a stoichiometric flame (ϕ = 1.00), due to incomplete combustion (see Table 6.1). However, the fuel rich flame emits more penetrating radiation than a stoichiometric flame even though not all the fuel was consumed. By using an appropriate spectral filter to block out the undesirable longer wavelength (λ > 2.7 µm) radiation, a heat flux meter can then be used to conveniently give the integrated radiation intensity that would be most penetrating into molten glass. Ji and Baukal tested
© 2011 by Taylor and Francis Group, LLC
several optical filters and compared the results against a spectral radiometer. Among the filters tested, a Schott BK7 optical glass filter exhibited a sharp cut-off transmission at λ = 2.7 µm. It became opaque at longer wavelengths for filter thickness greater than 6 mm (0.25 in.). Figure 6.27 shows a comparison of the measured flame radiation spectrum after it passed through a room temperature BK7 glass window, and the calculated flame radiation spectrum after it passed through 1 cm thick, high temperature, low iron content molten glass (with the same spectral absorption coefficient used in Figure 6.26). The BK7 transmitted spectra closely matched that of the molten glass. This technique is a simple, reliable way of determining the penetrating portion of the radiation from a flame. This method was used to measure the penetrating radiation from q f = 5.0 kW flames with different fuel equivalence and oxygen enrichment ratios. Similar to the analysis of spectral radiometer results, the radiation intensities were normalized to that of a flame with q f = 5.0 kW, Ω = 1.0, and ϕ = 1.0. The agreement between the heat flux meter with a BK7 filter measurement and the numerically integrated spectral radiometer measurement was satisfactory as shown by the qualitative trends for columns Inrm(pen.) and Inrm(BK7) in Table 6.1. This validated the use of a Vatell heat flux meter with a Schott BK7 optical glass filter to determine the penetrating radiation. To estimate the radiation intensity beyond 5.0 µm, a sapphire window was placed in front of the total heat flux meter. The transmission efficiency of a sapphire window is nearly constant at about 84% for λ = 0.25–5.0 µm, but quickly drops beyond 5.0 µm and cuts off at
135
Heat Flux
5.5
Flame radiation intensity (W m–2 µm–1)
5 4.5 4 3.5 3
I (w/BK7, measured)
2.5
I (w/o BK7, measured)
2
I (through 1 cm glass, calc.)
1.5 1 0.5 0 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 4.00 4.25 4.50 4.75 5.00 Wavelength λ (mm)
Figure 6.27 Transmission of flame radiation through a room temperature BK7 optical glass filter and through high temperature molten glass. (From Ji, B., and Baukal, C. E., “Spectral Radiation Properties of Oxygen-Enhanced/Natural Gas Flames,” Proceedings of 1998 International Gas Research Conference, edited by D. Dolenc, Vol. 5, 422–33, November 8–11, 1998, San Diego, CA, 1998.)
Table 6.2 Estimation of the Radiation Intensity Beyond 5.0 µm Ω 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.60 0.21
φ
I(w/sap.)
I(λ < 5.0µm)
I(w/o)
I(λ > 5.0µm)
(λ > 5.0 µm)%
0.67 0.80 1.00 1.33 1.50 1.60 1.79 1.89 2.00 1.00 1.00
0.0476 0.0504 0.0532 0.0504 0.0482 0.0465 0.0439 0.0429 0.0443 0.0434 0.0170
0.0567 0.0600 0.0633 0.0600 0.0574 0.0554 0.0523 0.0511 0.0527 0.0517 0.0202
0.0672 0.0723 0.0751 0.0727 0.0684 0.0674 0.0636 0.0609 0.0620 0.0631 0.0269
0.0105 0.0123 0.0118 0.0127 0.0110 0.0120 0.0113 0.0098 0.0093 0.0114 0.0067
16 17 16 17 16 18 18 16 15 18 25
Source: From Ji, B., and Baukal, C. E., “Spectral Radiation Properties of Oxygen-Enhanced/Natural Gas Flames,” Proceedings of 1998 International Gas Research Conference, Vol. 5, 422–33, November 8–11, 1998, San Diego, CA, 1998.
6.0 µm. The difference between the total radiometer readings with and without the sapphire window can be used to estimate radiation intensity beyond 5.0 µm. The results are listed in Table 6.2. For Ω = 1.00 and Ω = 0.60 flames, radiation beyond 5.0 µm accounted for about 17% of the total flame radiation. For Ω = 0.21 flames, the I(λ > 5.0µm) was about 25% of the total radiation. Combustion gas radiation at wavelengths longer than 5.0 µm is dominated by an H2O band at 6.3 µm [97,98]. The decrease in the relative contribution from shorter wavelength vibration band intensity in Ω = 0.21 flames can be attributed to less population in the highly excited
© 2011 by Taylor and Francis Group, LLC
vibrational levels in H2O molecules. This can be caused by two factors in Ω = 0.21 flames: less violent combustion reactions lead to weaker (if any) nonequilibrium chemiluminescence and a cooler flame temperature leads to fewer equilibrium thermal populations in the upper vibrational levels. Although the estimation given was only approximate, it gave a qualitative indication of the relative contribution from long wavelength (λ > 5.0 µm) radiation. Kilham [17] and Jackson and Kilham [18] used a thermopile normal to the surface of heated refractory cylinders. The combustion products from the flame, flowing around the cylinder, were also in the field of view. Therefore, the sum of the radiation from the gases, and from the cylinder surface, was measured. Two screens with narrow slit openings were placed between the flame and the detector, to limit the field of view to a small region around the heated cylinder. Giedt, Cobb, and Russ [55] measured radiation for a H/H2/CO mixture flame. From available data and radiation measurements, the gas emissivity was estimated to be 0.001. Nonluminous gas radiation was calculated to be, at most, only 2% of the total heat flux. Woodruff and Giedt [56] measured gas and surface radiation, with two Leeds and Northrup m irror-type rayotubes. Radiation was found to be a negligible effect. Veldman, Kubota, and Zukoski [58] concluded that the measured radiation, from an air/C3H8 flame, was negligible. No details of the measurement techniques were given. Baukal and Gebhart [99] used surfaces of different emissivities to study flame radiation. These removable
136
Industrial Combustion Testing
850 Polished (ε ~ 0.028) Untreated (ε ~ 0.05 – 0.60) Blackened (ε ~ 0.95)
800
q” (kW/m2)
750
6.4 Convective Heat Flux
700 650 600 550 500 450
nm and two wavelength emission pyrometry at 900 and 1000 nm.
0
1
2
3
L
4
5
6
Figure 6.28 Total heat flux (q”) versus axial distance between the burner outlet and the impingement surface for O2/natural gas flames (firing rate = 15.0 kW) impinging on the stagnation point (Reff = 0.16) of brass surfaces with various surface radiant characteristics. (From Baukal, C. E., and Gebhart, B., Thermal and Fluid Science, 15, 323–25, 1997.)
surfaces were attached to copper calorimeters. In order to minimize contact resistance [100], a high thermal conductivity copper-based antiseize lubricant was used at the interface between the various surfaces tested and the copper calorimeters. Polished metal surfaces were used as the low emissivity surfaces. Surfaces coated with a high temperature flat black paint were used as the high emissivity surfaces. One of the problems with the polished surfaces was very rapid degradation of both the brass and the copper surfaces. Within minutes, those surfaces lost much of their reflectivity, due to the contact with the flame gases. The polished stainless steel surfaces remained reflective for over an hour. There was no noticeable degradation of the blackened surfaces, for all three target materials, after hours of operation. Figure 6.28 shows that the blackened surface had the highest heat flux and the polished surface had the lowest flux. The difference between the two fluxes was relatively small and represents the radiation portion of the total heat flux to the surface. The majority of the heat flux was by convection. Sivathanu, Gore, and Dolinar [101] measured the soot volume fractions in highly turbulent and strongly radiating flames that are common industrial-type flames. They used laser extinction measurements made with a three-line optical probe that had laser extinction at 632
© 2011 by Taylor and Francis Group, LLC
In low temperature processes, radiation may be insignificant and can be ignored, which means that the convective heat flux can be measured with a standard heat flux gauge. However, in high temperature combustion processes where radiation is very significant, it is difficult to directly measure the convective heat flux. The technique that is commonly used is to measure the total heat flux and the radiant flux and then to calculate the convective heat flux by subtracting the two. This obviously only works if there are no other modes of heat transfer that contribute to the total heat flux. A related method to determine the convective heat flux is by using other measurements along with a calculation. To do this, the temperatures of the surface where the convection is occurring and of the gas flowing over the surface, along with the gas velocity are needed. Using the appropriate forced convection heat transfer equation for the given geometry, the convective heat flux can be calculated. This method is useful as a check against other methods that may be used. Another method of determining the convective heat flux by direct measurement involves adding a radiant shield between the heat source and the heat flux gauge to exclude radiation from being received by the gauge. The problem is that the shield often disturbs the flow and the measured flux is not representative of the actual convective flow. Another problem is that the surface temperature under the shield will be lower than it would be without the shield, so the measurement would need to be done quickly before the wall temperature under the shield drops too much. In addition, although radiation from the primary source (burners) may be shielded, the rest of the furnace walls may still contribute radiation to the gauge. A third technique that could be used is to make a perfectly reflective heat flux gauge so that any radiation received by the gauge is reflected away and not measured by the gauge. Then (in theory) the gauge would only measure convection, assuming no other heat transfer mechanisms were important. Although in theory this sounds plausible, the reality is that it is very difficult to have a perfect reflector in an industrial combustion environment. A fourth technique that could be used is to use a heat flux gauge embedded in the wall where convection is to be measured and then to turn off the radiant source and then quickly take measurements with the gauge. The radiation from the hot furnace walls can be calculated
Heat Flux
and subtracted from the total flux. Again, this method suffers from similar problems as the other methods above in that it relies on a transient condition and accurate calculations.
References
1. Arai, N., Matsunami, A., and Churchill, S. W. “A Review of Measurements of Heat Flux Density Applicable to the Field of Combustion.” Experimental Thermal and Fluid Science 12, no. 4 (1996): 452–60. 2. Baukal, C. E. Heat Transfer in Industrial Combustion. Boca Raton, FL: CRC Press, 2004. 3. Baukal, C. E., ed. The John Zink Combustion Handbook. Boca Raton, FL: CRC Press, 2001. 4. Colannino, J. “Mathematical Models for Characterizing and Predicting Heat Flux Profiles from Ethylene Cracking Units.” Proceedings of the 2007 AIChE Spring National Meeting, Houston, TX, April 23–26, 2007. 5. Colannino, J. Modeling of Combustion Systems: A Practical Approach. Boca Raton, FL: CRC Press, 2006. 6. Hayes, R., Singh, P., Foote, D., and Baukal, C. E. “Heat Flux from Process Burners.” Proceedings of the ASME IMECE, New York, November 11–16, 2001, Heat Transfer Division, Vol. 369, No. 4, pp. 161–64, 2001. 7. Rafidi, N., and Blasiak, W. “Heat Transfer Character istics of HiTAC Heating Furnace Using Regenerative Burners.” Applied Thermal Engineering 26, no. 16 (2006): 2027–34. 8. Hayes, R. R., Brewster, S., Webb, B. W., McQuay, M. Q., and Huber, A. M. “Crown Incident Radiant Heat Flux Measurements in an Industrial, Regenerative, Gas-Fired, Flat-Glass Furnace.” Experimental Thermal and Fluid Science 24 (2001): 35–46. 9. Newbold, J., Webb, B. W., McQuay, M. Q., and Huber, A. M. “Combustion Measurements in an Industrial GasFired Flat-Glass Furnace.” Journal of the Institute of Energy 70 (1997): 71–81. 10. McQuay, M. Q., and Webb, B. W. “Effect of Rebuild on the Combustion Performance of an Industrial Gas-Fired Flat Glass Furnace.” Combustion Science and Technology 150, no. 1 (2000): 77–97. 11. Hoogendoorn, C. J., Ballintijn, C. M., and Dorresteijn, W. R. “Heat-Flux Studies in Vertical Tube Furnaces.” Journal of the Institute of Fuel 43 (1970): 511–16. 12. Goldstein, R. J., and Chiang, H. D. “Measurement of Temperature and Heat Transfer.” In Handbook of Heat Transfer Applications, edited by W. M. Rohsenow, J. P. Hartnett, and E. N. Ganic´ , 2nd ed. New York: McGrawHill, 12-1–12-94, 1985. 13. Okoh, C. I., and Brown, R. A. Combustion Experimentation Handbook. Gas Research Institute report GRI-88/0143, Chicago, IL, 1988. 14. Bachmann, R. C., Chambers, J. T., and Giedt, J. T. “Investigation of Surface Heat-Flux Measurements with Calorimeters.” ISA Transactions 4 (1965): 143–51.
© 2011 by Taylor and Francis Group, LLC
137
15. Beasley, D. E., and Figliola, R. S. “A Generalised (sic) Analysis of a Local Heat Flux Probe.” Journal of Physics E-Scientific Instruments 21 (1988): 316–22. 16. Ducharme, M. B., Frim, J., and Tikuisis, P. “Errors in Heat Flux Measurements Due to the Thermal Resistance of Heat Flux Disks.” Journal of Applied Physiology 69, no. 2 (1990): 776–84. 17. Kilham, J. K. “Energy Transfer from Flame Gases to Solids.” Third Symposium on Combustion and Flame and Explosion Phenomena, 733–40. Baltimore, MD: The Williams and Wilkins Co., 1949. 18. Jackson, E. G., and Kilham, J. K. “Heat Transfer from Combustion Products by Forced Convection.” Industrial & Engineering Chemistry 48, no. 11 (1956): 2077–79. 19. Davies, R. M. “Heat Transfer Measurements on Electrically-Boosted Flames.” 10th Symposium (Inter national) on Combustion, The Combustion Institute, Pittsburgh, PA, 755–66, 1965. 20. Fells, I., and Harker, J. H. “An Investigation Into Heat Transfer from Unseeded Propane-Air Flames Augmented with D.C. Electrical Power.” Combustion and Flame 12 (1968): 587–96. 21. Vizioz, J.-P., and Lowes, T. M. “Convective Heat Transfer from Impinging Flame Jets.” International Flame Research Foundation report F 35/a/6, IJmuiden, the Netherlands, 1971. 22. Davies, D. R. “Heat Transfer from Working Flame Burners.” BS thesis, University of Salford, UK, 1979. 23. Posillico, C. J. “Raman Spectroscopic and LDV Measure ments of a Methane Jet Impinging Normally on a Flat Water-Cooled Boundary.” PhD thesis, Polytechnic Institute of New York, 1986. 24. Smith, R. B., and Lowes, T. M. “Convective Heat Transfer from Impinging Tunnel Burner Flames—A Short Report on the NG-4 Trials.” International Flame Research Founda tion report F 35/a/9, IJmuiden, the Netherlands, 1974. 25. Rajani, J. B., Payne, R., and Michelfelder, S. “Convective Heat Transfer from Impinging Oxygen-Natural Gas Flames—Experimental Results from the NG5 Trials.” International Flame Research Foundation report F 35/a/12, IJmuiden, the Netherlands, 1978. 26. Baukal, C. E., and Gebhart, B. “Heat Transfer from Oxygen-Enhanced/Natural Gas Flames Impinging Normal to a Plane Surface.” Experimental Thermal and Fluid Science 16, no. 3 (1998): 247–59. 27. American Society for Testing and Materials Standard E 422 – 83, Standard Method for Measuring Heat Flux Using a Water-Cooled Calorimeter, Philadelphia, PA, 1983. 28. Anderson, J. E., and Stresino, E. F. “Heat Transfer from Flames Impinging on Flat and Cylindrical Surfaces.” Journal of Heat Transfer 85, no. 1 (1963): 49–54. 29. Reed, T. B. “Heat-Transfer Intensity from Induction Plasma Flames and Oxy-Hydrogen Flames.” Journal of Applied Physics 34, no. 8 (1963): 2266–69. 30. Fay, R. H. “Heat Transfer from Fuel Gas Flames.” Welding Journal, Research Supplement, (1967): 380s–83s. 31. Cookson, R. A., and Kilham, J. K. “Energy Transfer from Hydrogen-Air Flames.” Ninth Symposium (International) on Combustion, 257–63. New York: Academic Press, 1963.
138
32. Kilham, J. K., and Dunham, P. G. “Energy Transfer from Carbon Monoxide Flames.” 11th Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 899–905, 1967. 33. Cookson, R. A., Dunham, P. G., and Kilham, J. K. “Stagnation Point Heat Flow Meter.” Journal of Scientific Instruments 42 (1965): 260–62. 34. Beér, J. M., and Chigier, N. A. “Impinging Jet Flames.” Combustion and Flame 12 (1968): 575–86. 35. Buhr, E., Haupt, G., and Kremer, H. “Heat Transfer from Impinging Turbulent Jet Flames to Plane Surfaces.” In Combustion Institute European Symposium 1973, edited by F. J. Weinberg, 607–12. New York: Academic Press, 1973. 36. Kremer, H., Buhr, E., and Haupt, R. “Heat Transfer from Turbulent Free-Jet Flames to Plane Surfaces.” In Heat Transfer in Flames, edited by N. H. Afgan and J. M. Beér, 463–72. Washington, DC: Scripta Book Company, 1974. 37. Hoogendoorn, C. J., Popiel, C. O., and van der Meer, T. H. “Turbulent Heat Transfer on a Plane Surface in Impingement Round Premixed Flame Jets.” Proceedings of 6th International Heat Transfer Conference, 107–12. Toronto, Vol. 4, August 7–11, 1978. 38. Popiel, C. O., van der Meer, T. H., and Hoogendoorn, C. J. “Convective Heat Transfer on a Plate in an Impinging Round Hot Gas Jet of Low Reynolds Number.” International Journal of Heat and Mass Transfer 23 (1980): 1055–68. 39. Hargrave, G. K., and Kilham, J. K. “The Effect of Turbulence Intensity on Convective Heat Transfer From Premixed Methane-Air Flames.” Institution of Chemical Engineers Symposium Series 2, no. 86 (1984): 1025–34, 1984. 40. Hargrave, G. K., Fairweather, M., and Kilham, J. K. “Forced Convective Heat Transfer from Premixed Flames—Part 2: Impingement Heat Transfer.” International Journal of Heat and Fluid Flow 8, no. 2 (1987): 132–38. 41. Kataoka, K., Shundoh, H., and Matsuo, H. “Convective Heat Transfer Between a Flat Plate and a Jet of Hot Gas Impinging On It.” In Drying ’84, edited by A. S. Mujumdar, 218–27. New York: Hemisphere/SpringerVerlag, 1984. 42. Cassiano, J., Heitor, M. V., and Silva, T. F. “Combustion Tests on an Industrial Glass-Melting Furnace.” Fuel 73, no. 10 (1994): 1638–42. 43. Shorin, S. N., and Pechurkin, V. A. “Effectivnost’ teploperenosa na poverkhnost’ plity ot vysokotemperaturnoi strui produktov sjoraniya razlichnykh gazov” (The Effectiveness of Heat Transfer to the Surface of a Plate from a High-Temperature Jet of Combustion Products of Various Gases). Teoriya i Praktika Szhiganiya Gaza 4 (1968): 134–43. 44. Schulte, E. M., and Kohl, R. F. “A Transducer for Measuring High Heat Transfer Rates.” Review of Scientific Instruments 41, no. 12 (1970): 1732–40. 45. Schulte, E. M. “Impingement Heat Transfer Rates from Torch Flames.” Journal of Heat Transfer 94 (1972): 231–33. 46. You, H.-Z. “An Investigation of Fire-Plume Impingement on a Horizontal Ceiling: 2—Impingement and Ceiling-Jet Regions.” Fire & Materials 9, no. l (1985): 46–56.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
47. Hager, J. M., Langley, L. W., Onishi, S., and Diller, T. E. “Microsensors for High Heat Flux Measurements.” Journal of Thermophysics 7, no. 3 (1993): 531–34. 48. Gardon, R. “A Transducer for the Measurement of Heat Flow Rate.” Journal of Heat Transfer 82 (1960): 396–98. 49. Grey, J. “Probe Measurements in High Temperature Gases and Dense Plasmas.” In Measurements in Heat Transfer, edited by E. R. G. Eckert and R. J. Goldstein, 2nd ed. Washington, DC: Hemisphere Publishing, 337–374, 1976. 50. Ash, R. L. “Response Characteristics of Thin Foil Heat Flux Sensors.” AIAA Journal 7, no. 12 (1969): 2332–35. 51. Borell, G. J., and Diller, T. E. “A Convection Calibration Method for Local Heat Flux Gages.” Journal of Heat Transfer 109 (1987): 83–89. 52. van der Meer, T. H. “Stagnation Point Heat Transfer from Turbulent Low Reynolds Number Jets and Flame Jets.” Experimental Thermal and Fluid Science 4 (1991): 115–26. 53. Barnes, A. “Heat Flux Sensors, Part 2: Applications.” Sensors 16, no. 2 (1999): 54–57. 54. Gebhart, B. Heat Transfer. New York: McGraw-Hill, 1971. 55. Giedt, W. H., Cobb, L. L., and Russ, E. J. “Effect of Hydrogen Recombination on Turbulent Flow Heat Transfer.” ASME Paper 60-WA-256, New York, 1960. 56. Woodruff, L. W., and Giedt, W. H. “Heat Transfer Measurements from a Partially Dissociated Gas with High Lewis Number.” Journal of Heat Transfer 88 (1966): 415–20. 57. Milson, A., and Chigier, N. A. “Studies of Methane-Air Flames Impinging on a Cold Plate.” Combustion and Flame 21 (1973): 295–305. 58. Veldman, C. C., Kubota, T., and Zukoski, E. E. “An Experimental Investigation of the Heat Transfer from a Buoyant Gas Plume to a Horizontal Ceiling—Part 1. Unobstructed Ceiling.” National Bureau of Standards report NBS-GCR-77-97, Washington, DC, 1975. 59. Matsuo, M., Hattori, M., Ohta, T., and Kishimoto, S. “The Experimental Results of the Heat Transfer by Flame Impingement.” International Flame Research Foundation report F 29/1a/1, IJmuiden, the Netherlands, 1978. 60. Rauenzahn, R. M. “Analysis of Rock Mechanics and Gas Dynamics of Flame-Jet Thermal Spallation Drilling.” PhD thesis, MIT, Cambridge, MA, 1986. 61. Gatowski, J. A., Smith, M. K., and Alkidas, A. C. “An Experimental Investigation of Surface Thermometry and Heat Flux.” Experimental Thermal and Fluid Science 2, no. 3 (1989): 280–92. 62. Hornbaker, D. R., and Rall, D. L. “Thermal Perturbations Caused by Heat-Flux Transducers and Their Effect on the Accuracy of Heating-Rate Measurements.” ISA Transactions 3 (1964): 123–30. 63. Kilham, J. K., and Purvis, M. R. I. “Heat Transfer from Hydrocarbon-Oxygen Flames.” Combustion and Flame 16 (1971): 47–54. 64. Nawaz, S. “Heat Transfer from Oxygen Enriched Methane Flames.” PhD thesis, The University of Leeds, U.K., 1973. 65. Fairweather, M., Kilham, J. K., and Nawaz, S. “Stagnation Point Heat Transfer from Laminar, High Temperature Methane Flames.” International Journal of Heat and Fluid Flow 5, no. 1 (1984): 21–27.
Heat Flux
66. Fairweather, M., Kilham, J. K., and Mohebi-Ashtiani, A. “Stagnation Point Heat Transfer from Turbulent Methane-Air Flames.” Combustion Science and Technology 35 (1984): 225–38. 67. Conolly, R., and Davies, R. M. “A Study of Convective Heat Transfer from Flames.” International Journal of Heat and Fluid Flow 15 (1972): 2155–172. 68. Kilham, J. K., and Purvis, M. R. I. “Heat Transfer from Normally Impinging Flames.” Combustion Science and Technology 18 (1978): 81–90. 69. Ivernel, A., and Vernotte, P. “Etude expérimentale de l’amélioration des transferts convectis dans les fours par suroxygénation du comburant.” Revue Generale de Thermique 18, nos. 210–211 (1979): 375–91. 70. Horsley, M. E., Purvis, M. R. I., and Tariq, A. S. “Convective Heat Transfer From Laminar and Turbulent Premixed Flames.” Heat Transfer 1982, edited by U. Grigull, E. Hahne, K. Stephan, and J. Straub, 409–15. Washington, DC: Hemisphere, Vol. 3, 1982. 71. Hemeson, A. O., Horsley, M. E., Purvis, M. R. I., and Tariq, A. S. “Heat Transfer from Flames to Convex Surfaces.” Institution of Chemical Engineers Symposium Series 2, no. 86 (1984), Rugby, UK: Institution of Chemical Engineers, 969–78. 72. Hustad, J. E., Røkke, N. A., and Sønju, O. K. “Heat Transfer to Pipes Submerged in Lifted Buoyant Diffusion Flames.” In Experimental Heat Transfer, Fluid Mechanics, and Thermodynamics, 1991, edited by J. F. Keffer, 567–74. New York: Elsevier, 1991. 73. Hustad, J. E., and Sønju, O. K. “Heat Transfer to Pipes Submerged in Turbulent Jet Diffusion Flames.” In Heat Transfer in Radiating and Combusting Systems. Berlin: Springer-Verlag, 474–90. 74. Hustad, J. E., Jacobsen, M., and Sønju, O. K. “Radiation and Heat Transfer in Oil/Propane Jet Diffusion Flames.” Institution of Chemical Engineers Symposium Series 10, no. 129 (1992): 657–63. 75. Gray, W. A., and Müller, R. Engineering Calculations in Radiative Heat Transfer. Oxford, UK: Pergamon, 1974. 76. Stradomskiy, M. V., Maksimov, Y. A., and Malyarov, V. S. “Calibration of High-Temperature Heat Flux Sensors.” Heat Transfer–Soviet Research 20, no. 4 (1988): 562–68. 77. Ludwig, C. B., Malkmus, W., Reardon, J. E., and Thomson, J. A. L. Handbook of Infrared Radiation from Combustion Gases. Washington, DC: National Aeronautics and Space Administration report SP-3080, 1973. 78. Siegel, R., and Howell, J. R. Thermal Radiation Heat Transfer. 2nd ed. Washington, DC: Hemisphere, 1981. 79. Baukal, C. E., and Gebhart, B. “Oxygen-Enhanced Natural Gas Flame Radiation.” International Journal of Heat and Mass Transfer 40, no. 11 (1997): 2539–47. 80. Gore, J. P., Skinner, S. M., Stroup, D. W., Madrzykowski, D., and Evans, D. D. “Structure and Radiation Properties of Large Two-Phase Flames.” In Heat Transfer in Combustion Systems, edited by N. Ashgriz, J. G. Quintiere, H. G. Semerjian, and S. E. Slezak, 77–86. New York: American Society of Mechanical Engineers, HTDVol. 122, 1989. 81. Lide, D. R., ed. CRC Handbook of Chemistry and Physics, 79th ed. Boca Raton, FL: CRC Press, 1998.
© 2011 by Taylor and Francis Group, LLC
139
82. Matthews, L., Harris, A., and Garcia, G. “Radiative Flux Measurements in a Sooty Pool Fire Using a Multihead Transpiration Radiometer.” In Heat Transfer Phenomena in Radiation, Combustion, and Fires, edited by R. K. Shah, 375–80. New York: ASME HTD-Vol. 106, 1989. 83. Moffat, R. J., Hunn, B. D., and Ayers, J. F. “Development of a Transpiration Radiometer.” Instrument Society of America, Conference Proceedings, Paper No. 613, pp. 613.1-613.7, 1971. 84. Brajuskovic, B., Matovic, M., and Afgan, N. “A Heat Flux-Meter for Ash Deposit Monitoring Systems – I. Ash Deposit Prevention.” International Journal of Heat and Mass Transfer 34, no. 9 (1991): 2291–-2301. 85. Brajuskovic, B., and Afgan, N. “A Heat Flux-Meter for Ash Deposit Monitoring Systems—II. ‘Clean’ Heat FluxMeter Characteristics.” International Journal of Heat and Mass Transfer 34, no. 9 (1991): 2303–15. 86. Martins, N., Carvalho, M. G., Afgan, N. H., and Leontiev, A. L. “Experimental Verification and Calibration of the Blow-Off Heat Flux Sensor.” Applied Thermal Engineering 18, no. 6 (1998): 481–89. 87. Martins, N., Carvalho, M. G., Afgan, N. H., and Leontiev, A. I. “Design and Sensitivity Analysis of a New Gauge for Radiation Heat Flux Assessment.” Heat & Technology 16, no. 2 (1998): 77–84. 88. Ploteau, J. P., Glouannec, P., and Noel, H. “Conception of Thermoelectric Flux Meters for Infrared Radiation Measurements in Industrial Furnaces.” Applied Thermal Engineering 27, nos. 2–3 (2007): 674–81. 89. Chedaille, J., and Braud, Y. Vol. 1: Measurements in Flames. New York: Crane, Russak & Co., 1972. 90. Butler, B. W., and Webb, B. W. “Measurements of Local Temperature and Wall Radiant Heat Flux in an Industrial Coal-Fired Boiler.” In Heat Transfer in Combustion Systems—1990, edited by B. Farouk, W. L. Grosshandler, D. G. Lilley, and C. Presser, 49–56. New York: ASME HTD-Vol. 142, 1990. 91. Beltagui, S. A., Kenbar, A. M. A., and Maccallum, N. R. L. “Heat Transfer and Emission Studies in a Gas Fired Furnace.” In Boilers & Furnaces. Proceedings of International Symposium on Combustion & Emissions Control, Cardiff, UK, September 1993, Institute of Energy, pp. 275–96, 1993. 92. Ji, B., and Baukal, C. E. “Spectral Radiation Properties of Oxygen-Enhanced/Natural Gas Flames.” Proceedings of 1998 International Gas Research Conference, edited by D. Dolenc, Vol. 5, 422–33. November 8–11, 1998, San Diego, CA, 1998. 93. Franz, H. “Infrared Absorption of Molten Soda-LimeSilica Glasses Containing Transition Metal Oxides.” International Congress on Glass, Science and Technology 9, no. 1 (1971): 243–60. 94. Endrys, J., Geotti-Bianchini, F., and De Riu, L. “Study of the High-Temperature Spectral Behavior of Container Glass.” Glastechnische Berichte Glass Science and Technology 70, no. 5 (1997): 126–36. 95. Banner, D., and Klarsfeld, S. “High Temperature Infrared Spectra of Silicate Melts.” In Physics of Non-crystalline Solids, edited by L. D. Pye, W. C. La Course, and H. J. Stevens, 371–75, London, 1992.
140
96. Berg, J. I. “Near Infrared Absorption Coefficient of Molten Glass by Emission Spectroscopy.” International Journal of Thermophysics 2, no. 4 (1981): 381–94. 97. Faeth, G. M., Gore, J. P., Chuech, S. G., and Jeng, S.-M. “Radiation from Turbulent Diffusion Flames.” Annual Review of Fluid Mechanics & Heat Transfer 2 (1989): 1–38. 98. Ludwig, C. B., Malkmus, W., Reardon, J. E., and Thomson, J. A. L. Handbook of Infrared Radiation from Combustion Gases. Washington, DC: National Aeronautics and Space Administration report SP-3080, 1973.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
99. Baukal, C. E., and Gebhart, B. “Surface Condition Effects on Flame Impingement Heat Transfer.” Thermal and Fluid Science 15 (1997): 323–35. 100. Holman, J. P. Heat Transfer. 10th ed. New York: McGrawHill, 2010. 101. Sivathanu, Y. R., Gore, J. P., and Dolinar, J. “Transient Scalar Properties of Strongly Radiating Jet Flames.” In Heat and Mass Transfer in Fires and Combustion Systems, edited by W. L. Grosshandler and H. G. Semerjian, 45–56. New York: ASME HTD-Vol. 148, ASME, 1990.
7 Pollution Emissions Charles E. Baukal, Jr. Contents 7.1 Introduction.....................................................................................................................................................................141 7.2 Some Statistics................................................................................................................................................................ 142 7.3 Sampling......................................................................................................................................................................... 144 7.3.1 Extractive Sampling........................................................................................................................................... 148 7.3.2 In-Situ Sampling................................................................................................................................................ 152 7.3.3 Ambient Condition Effects............................................................................................................................... 152 7.4 Testing Equipment......................................................................................................................................................... 160 7.4.1 Exhaust Gas Flow.............................................................................................................................................. 160 7.4.1.1 Pitot Tube.............................................................................................................................................. 160 7.4.1.2 Direct Measurement............................................................................................................................161 7.4.1.3 Calculation............................................................................................................................................162 7.4.1.4 Non-Contact..........................................................................................................................................162 7.4.2 Gas Composition.................................................................................................................................................162 7.4.3 Particulates......................................................................................................................................................... 166 7.5 Source Testing.................................................................................................................................................................167 7.6 Continuous Monitoring................................................................................................................................................ 169 7.7 Spot Checking................................................................................................................................................................ 170 7.8 Example Measurements................................................................................................................................................ 170 7.8.1 Oxygen-Enhanced Combustion...................................................................................................................... 170 7.8.1.1 Experimental Setup............................................................................................................................ 170 7.8.1.2 Results................................................................................................................................................... 172 7.8.1.3 Discussion............................................................................................................................................ 173 7.8.1.4 Conclusions and Recommendations................................................................................................ 173 7.8.2 Fuel Composition Effects.................................................................................................................................. 173 7.8.2.1 Test Setup..............................................................................................................................................174 7.8.2.2 Sample Results..................................................................................................................................... 175 7.8.2.3 Conclusions...........................................................................................................................................176 7.8.3 Ghost NOx...........................................................................................................................................................176 7.8.3.1 Introduction......................................................................................................................................... 177 7.8.3.2 Test Description................................................................................................................................... 177 7.8.3.3 Results................................................................................................................................................... 178 7.8.3.4 Conclusions.......................................................................................................................................... 181 References................................................................................................................................................................................. 181
7.1 Introduction There are a number of reasons for sampling the exhaust gases from an industrial combustion process [1]. One is to determine whether the emissions are within regulatory limits. Another is to determine the efficiency and effectiveness of the pollution control technology being
used in a given process. Stricter limits or ineffective pollution control technologies may lead to design changes in the equipment. Sampling is also useful for predicting any potential problems in the development of a new process or modification of the existing process. The U.S. Environmental Protection Agency has established sampling methods for different pollutants in 40 CFR 60 [2]. These are available electronically over the Internet [3]. 141
© 2011 by Taylor and Francis Group, LLC
142
Industrial Combustion Testing
There are many useful general references on experimental measurements [4–8] as well as specific references on pollution measurement techniques (e.g., [9–15]). Crawford (1976) lists the following steps in the process of measuring pollutants in an air stream [16]:
1. Collection of the air sample containing the pollutant 2. Conditioning of the sample 3. Separation of the pollutants from the air in the sample 4. Measurement of the desired properties of the pollutants 5. Measuring the sample flow rate 6. Disposal of the sample gas after completing the measurements
are briefly discussed here without derivation or proof. The reader is referred to environmental statistics books for those details [18–29]. The first concept is the mean or average value of a set of numbers: 1 x= n
n
∑ x , i
(7.1)
i=1
where x is the average value, xi is the ith value, i is the index going from 1 to n, and n is the total number of observations. The range of readings is: Range = xmin → xmax.
It is noted that not all of these steps are required in particular circumstances. The last step is normally not a factor for most industrial combustion processes where the gas sample is vented to the atmosphere. This is only a factor where the pollutants in the sample may be poisonous, radioactive, or toxic in the concentrations present. The objective of this chapter is to discuss the general principles in measuring pollutants in industrial combustion exhaust gas streams [17].
7.2 Some Statistics There are some limited statistics that may be needed for measuring and reporting pollution emissions. These
(7.2)
Some operating permits may be written on a notto-exceed basis, which means that a given pollutant concentration cannot exceed a given value. In that case, only the maximum value over a given time period may be of interest. Some permits are written so that the average concentration cannot exceed a specific value, so it is important to know how the permit is written. Another statistical concept that may be used in reporting pollutant emissions is referred to as a moving average. Some regulations are written based on the average of a set of readings taken over a specific time period that is moving. For example, a pollutant may be limited to a certain maximum concentration over an eight-hour moving average. The moving average takes the most recent data for the specified period. Figure 7.1 shows a set of daily averages for a 10-day time span with a line through the data, which is a linear regression fit. Note that
Average NOx (ppmvd) at 3% O2
44
42
40
38
1
2
3
4
5
Day
6
7
8
9
10
Figure 7.1 Average emissions over a 10-day time span with a linear regression fit. (From Baukal, C. E., Industrial Combustion Pollution and Control, Marcel Dekker, New York, 2004.)
© 2011 by Taylor and Francis Group, LLC
143
Pollution Emissions
the vertical axis has been exaggerated to show the relatively small shift in the averages during that time span. The graph shows that average daily NOx appears to be increasing. This may indicate some type of problem that needs to be corrected before regulatory limits are exceeded. An example might be that the removal efficiency of a downstream NOx posttreatment system may be declining and the equipment is in need of service. Another statistical measure that is useful for reporting data is referred to as the standard deviation. This is a measure of the dispersion of the data from the average value. Figure 7.2 shows two sets of data taken on two different days over the same hours of the day. The data have been artificially chosen so that the average over the given period is the same for both days, but there is considerably more “scatter” on Day 2, which means that the data are farther from the average value compared to Day 1. The Day 2 data then have a higher standard deviation than that of the Day 1 data. The unbiased or “n-1” standard deviation is defined as: n
∑ (x − x ) i
σu =
i=1
2
n−1
.
(7.3)
The biased or “n” standard deviation is defined as: n
∑ (x − x ) i
σb =
2
i=1
n
,
(7.4)
where the difference between the two is the divisor. For a large sample size where n is large, these two will be essentially the same. The unbiased standard deviation is typically used for emissions reporting and will be used here for illustration purposes. The Day 1 data are assumed to be normally distributed about the mean value as shown in Figure 7.3. This means that if enough data were collected and the data were truly normally distributed about the mean, then they would fit the curve shown in the figure. The graph also shows the limits of one standard deviation from the mean both to the left and to the right. As can be seen, the majority of the data would fall in the region bounded by one standard deviation to the left and to the right of the mean. Therefore, it is common to report data in the following form: x = x ± σ.
(7.5)
The standard deviation shows how tightly distributed the data are about the mean. The smaller the standard deviation, the more closely the data are to the mean value. Figure 7.2 showed that the Day 2 data were more scattered compared to the Day 1 data. Figure 7.4 shows a comparison of the normal distribution curves for the Day 1 and Day 2 data, both of which have the same mean value. The curve for Day 1 is taller and thinner, which shows that most of the data are closer to the mean. The curve for Day 2 is shorter and fatter, which shows that most of the data are farther from the mean. A high relative standard deviation could indicate that there may be a problem with the measurement system
NOx (ppmvd) at 3% O2
50 Day 1 Day 2 Average
45
40
35
30 12:00 PM
1:00 PM
2:00 PM
3:00 PM
4:00 PM
5:00 6:00 7:00 PM PM PM Time of the day
8:00 PM
9:00 PM
10:00 11:00 PM PM
Figure 7.2 Distribution about a mean value of data taken on two consecutive days over the same time span. (From Baukal, C. E., Industrial Combustion Pollution and Control, Marcel Dekker, New York, 2004.)
© 2011 by Taylor and Francis Group, LLC
144
Industrial Combustion Testing
16% Day 1 Average -1 Std. Dev. +1 Std. Dev.
14%
Frequency (%)
12% 10% 8% 6% 4% 2% 0%
30
32
34
36 38 40 42 NOx (ppmvd) at 3% O2
44
46
48
Figure 7.3 Normal distribution of Day 1 data about the mean of 38.7 ppmvd NOx at 3% O2. 16%
Day 1 Day 2 Average
14%
Frequency (%)
12% 10% 8% 6% 4% 2% 0%
20
25
30
35 40 45 NOx (ppmvd) at 3% O2
50
55
60
Figure 7.4 Comparison of normal distributions for Day 1 with a smaller standard deviation and Day 2 with a larger standard deviation, both with the same mean value. (From Baukal, C. E., Industrial Combustion Pollution and Control, Marcel Dekker, New York, 2004.)
where there is not high repeatability if the data are not truly changing. A high standard deviation could also indicate that there are wider swings in the data that could be caused by changing conditions. For example, if the fuel composition is changing with time, which is sometimes the case in petrochemical applications, this could cause the NOx to change significantly. Another example is a changing ambient environment where there may be wide swings in the temperature and relative humidity of the incoming combustion air that can affect NOx emissions. A high relative standard deviation is not generally desirable because more cushion will be needed between the permitted or regulated emission limits and the measured emissions because of the higher possibility of significant excursions above the mean value.
© 2011 by Taylor and Francis Group, LLC
7.3 Sampling An intersociety committee put together a guidebook for air sampling, primarily in the ambient atmosphere [30]. The following general physical precautions relevant to extractive gas sampling were noted: ensure that the sample is homogeneous, be aware of any absorption and diffusion effects in the sampling system, check for mechanical defects in the sampling system, and calibrate at the design sampling flow rate. It is also noted that any interferences between chemicals in a sampling system must also be known and corrected for in order to ensure accurate and reliable measurements. The U.S. EPA has developed an extensive manual to aid in estimating the costs of pollution control equipment
145
Pollution Emissions
[31]. A chapter in the manual is devoted to emissions monitoring [32]. It is important that the position of the sample probe is known accurately if there is a considerable variation in the gas composition profile across the duct [33]. It is preferred that the measurements be made sufficiently downstream of flow disturbances like elbows or dampers to minimize the variation across the duct. If this is not possible as is the case when physical restrictions limit how the exhaust ducting can be run and where the measurements can be made, then more traverses are recommended. The EPA Method 1 (see Appendix B for a complete list of EPA Methods) recommends that both sample and velocity measurements be made at least 8 internal diameters downstream of any flow disturbance (e.g., elbow, duct contraction, or expansion) and at least 2 internal diameters upstream from any flow disturbance. If this is not possible, then measurements may be made as close as 2 internal diameters downstream and 0.5 internal diameters upstream of disturbances, but more measurements across the duct cross section are needed. Figure 7.5 shows the recommended minimum number of traverse (sampling) points for measuring either the gas sample composition or the gas velocity where no particulates are in the gas stream. The figure shows that at least 8 or 9 measurement points are recommended for internal stack diameters between 12 and 24 in. where the measurement location is at least 8 internal diameters downstream and 2 internal diameters upstream from any upstream disturbances. For internal stack diameters greater than 24 in. in internal diameter, at least 12 sampling points are recommended if the measurement location is at least 7
Minimum number of traverse points
50
internal diameters downstream and 1.75 internal diameters upstream of any flow disturbance. A minimum of 16 sample points are recommended for internal stack diameters 12 in. in internal diameter or greater where the sampling location is between 2 and 7 (8 for internal stack diameters between 12 and 24 in.) internal diameters downstream and between 0.5 and 1.75 (2 for internal stack diameters between 12 and 24 in.) internal diameters upstream from a flow disturbance. Figure 7.6 shows the recommended minimum number of traverse points if there are particulates in the flow. As can be seen, more traverses are generally needed when particulates are present. The minimum distance to a disturbance determines the minimum number of recommended traverse points. For example, if the stack diameter is 18 in. in internal diameter, the gas stream contains no particulates, and the sample location is 10 internal diameters downstream from the closest downstream disturbance but only 1 internal diameter upstream from the closest upstream disturbance, then the distance to the upstream disturbance determines the minimum number of traverses. According to Figure 7.5, only 8 or 9 traverses would be needed based on the downstream distance (10 diameters) to a disturbance but at least 16 traverses are recommended because the sample location is only 1 internal diameter upstream from the location. For rectangular ducts, an equivalent diameter is calculated using the following De =
Duct diameters upstream from flow disturbance* (distance A) 1.0 1.5 2.0
0.5
aHigher number is for rectangular stacks or ducts
40
2 LW , L+W
(7.6)
2.5
Disturbance A
Measurement site
B
30
Disturbance 20
16
12
a 8 or 9 * From point of any type of disturbance (bend, expansion, contraction, etc.) Stack diameter = 0.30 to 0.61 m (12-24 in.)
10
0
Stack diameter > 0.61 m (24 in.)
2
3
4
5
6
7
8
9
10
Duct diameters downstream from flow disturbance* (distance B) Figure 7.5 Minimum number of traverse points for sample or velocity measurements in when particulates are not present. (From U.S. EPA, 40 CFR 60: Standards of Performance for New Stationary Sources, Washington, DC: U.S. Environmental Protection Agency, 2001.)
© 2011 by Taylor and Francis Group, LLC
146
Industrial Combustion Testing
Minimum number of traverse points
50
Duct diameters upstream from flow disturbance* (distance A) 1.0 1.5 2.0
0.5 a
40 30
Higher number is for rectangular stacks or ducts
Measurement site
B Disturbance
20
16
* From point of any type of disturbance (bend, expansion, contraction, etc)
10 0
Disturbance
A
24 or 25a
20
2.5
Stack diameter > 0.61 m (24 in.) 12 a 8 or 9
Stack diameter = 0.30 to 0.61 m (12-24 in.) 2
3 4 5 6 7 8 9 Duct diameters downstream from flow disturbance* (distance B)
10
Figure 7.6 Minimum number of traverse points for sample or velocity measurements when particulates are present. (From U.S. EPA, 40 CFR 60: Standards of Performance for New Stationary Sources, Washington, DC: U.S. Environmental Protection Agency, 2001.)
Table 7.1
Typical centroid measurement location
Recommended Measurement Layout for Rectangular Cross-Sectional Area Ducts. Cross Section Layout for Rectangular Stacks (40 CFR 60: Appendices, p. 11) # Traverse Points in Layout 9 12 16 20 25 30 36 42 49
Recommended Matrix 3 × 3 4 × 3 4 × 4 5 × 4 5 × 5 6 × 5 6 × 6 7 × 6 7 × 7
Source: U.S. EPA, 40 CFR 60: Standards of Performance for New Stationary Sources, Washington, DC: U.S. Environmental Protection Agency, 2001.
where L is the length of the rectangular duct cross section, W is the width of the rectangular duct cross section, and De is the equivalent diameter. Note that L = W for square ducts and that all of these dimensions are internal. Table 7.1 shows the recommended layout for rectangular ducts based on the number of traverse points from the above figures. For example, if 12 traverse points are recommended, then the table suggests a 4 × 3 layout. Figure 7.7 shows an example of a rectangular duct divided into 12 equal subareas in a 4 × 3 matrix. Then, the samples are measured in the centroid of each subarea as shown in the figure. Table 7.2 shows the recommended layout for circular ducts based on the number of traverse points from the above figures. For example, if 12 traverse points are
© 2011 by Taylor and Francis Group, LLC
Figure 7.7 Rectangular cross-sectional duct divided into 12 equal subareas. (From Baukal, C. E., Industrial Combustion Pollution and Control, Marcel Dekker, New York, 2004.)
recommended, then the table suggests the location of the sample points. Figure 7.8 shows an example of a circular duct divided into 12 equal subareas in a symmetric matrix. Then the samples are measured in the centroid of each subarea as shown in the figure. The distances to each centroid are given as a percentage of the total diameter. The EPA Method 1A applies to ducts with equivalent internal diameters of less than 12 in. but greater than or equal to 4 in. The method does not apply to swirling or cyclonic flows. If particulates are present, one or two traverse points should be located at least 8 internal diameters downstream and 10 internal diameters upstream from any flow disturbances. Note that velocity measurements should be made 8 internal diameters downstream from the gas sampling location for this case. Figure 7.9 shows the recommended sampling arrangement for small ducts. Again, Figure 7.5 should be used to determine the number of sample points for flows without particulates and Figure 7.6 should be used for flows with
147
Pollution Emissions
Table 7.2 Recommended Measurement Layout for Circular Cross-Sectional Area Ducts. Location of Traverse Points in Circular Stacks in Percentage of Stack Diameter from Inside Wall to Traverse Point (40 CFR 60: Appendices, pp. 12–13) Traverse Point on a Diameter 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Number of Traverse Points on a Diameter 2
4
6
8
10
12
14
16
18
20
22
14.6 85.4
6.7 25.0 75.0 93.3
4.4 14.6 29.6 70.4 85.4 95.6
3.2 10.5 19.4 32.3 67.7 80.6 89.5 96.8
2.6 8.2 14.6 22.6 34.2 65.8 77.4 85.4 91.8 97.4
2.1 6.7 11.8 17.7 25.0 35.6 64.4 75.0 82.3 88.2 93.3 97.9
1.8 5.7 9.9 14.6 20.1 26.9 36.6 63.4 73.1 79.9 85.4 90.1 94.3 98.2
1.6 4.9 8.5 12.5 16.9 22.0 28.3 37.5 62.5 71.7 78.0 83.1 87.5 91.5 95.1 98.4
1.4 4.4 7.5 10.9 14.6 18.8 23.6 29.6 38.2 61.8 70.4 76.4 81.2 85.4 89.1 92.5 95.6 98.6
1.3 3.9 6.7 9.7 12.9 16.5 20.4 25.0 30.6 38.8 61.2 69.4 75.0 79.6 83.5 87.1 90.3 93.3 96.1 98.7
1.1 3.5 6.0 8.7 11.6 14.6 18.0 21.8 26.2 31.5 39.3 60.7 68.5 73.8 78.2 82.0 85.4 88.4 91.3 94.0 96.5 98.9
24 1.1 3.2 5.5 7.9 10.5 13.2 16.1 19.4 23.0 27.2 32.3 39.8 60.2 67.7 72.8 77.0 80.6 83.9 86.8 89.5 92.1 94.5 96.8 99.9
Source: U.S. EPA, 40 CFR 60: Standards of Performance for New Stationary Sources, Washington, DC: U.S. Environmental Protection Agency, 2001. Typical centroid measurement location
Nearest upstream disturbance
Gas sample location
Velocity measurement location
Gas flow direction >8D 4
1
2
5
6
3
Figure 7.8 Circular cross-sectional duct divided into 12 equal subareas. (From Baukal, C. E., Industrial Combustion Pollution and Control, Marcel Dekker, New York, 2004.)
particulates. In either case, the minimum number of sample points is 8 for circular ducts and 9 for rectangular ducts. For circular ducts, the number of sample points should be divisible by 4. The main difference is that the
© 2011 by Taylor and Francis Group, LLC
Nearest downstream disturbance
D >8D
>2D
Figure 7.9 Recommended sampling arrangement for small ducts. (From Baukal, C. E., Industrial Combustion Pollution and Control, Marcel Dekker, New York, 2004.)
distances from the gas sampling location and from the velocity measurement location to the nearest upstream and downstream disturbances should be calculated (four total equivalent internal diameters). The distance requiring the most number of traverse points should be used as the minimum number of traverse points. There are two techniques for sampling exhaust products that are referred to as extractive and in-situ. Mullinger and Jenkins [34] compare both techniques in Section 8.3 of their book. These two techniques are discussed here in the next two sections.
148
Industrial Combustion Testing
7.3.1 Extractive Sampling Isokinetic sampling pertains to extractive gas sampling where the sample is withdrawn out of the gas stream at the same velocity as the exhaust gas stream velocity. The EPA defines isokinetic sampling as “sampling in which the linear velocity of the gas entering the sampling nozzle is equal to that of the undisturbed gas stream sample point” [35]. This is particularly i mportant when measuring larger particulates, especially those greater than about 5 µm. Figure 7.10 shows the three possibilities for withdrawing a sample from (a)
the exhaust gas stream. When the sample velocity is less than the gas stream velocity, some particles in the gas stream flow around the probe and the measured particulate concentration is lower than the actual particulate concentration. When the sample velocity is greater than the gas stream velocity, then some particles in the gas stream are preferentially drawn into the sample probe and the measured particulate concentration is greater than the actual particulate concentration. The sample flow rate is normally easily varied by either controlling the sample pump suction
Stream lines Concentration of all sizes the same here as here
(b)
Stream lines
No bend of stream lines
Stream lines bent on entering probe
Sample
le
mp
Sa
Sample-tube not in line nonisokinetic Too few large particles are collected
Illustration of isokinetic sampling
(c)
Stream lines
(d)
Stream lines bent at sample point
Stream lines
Stream lines bent at sample point
Sample
Sample
Sample velocity too low nonisokinetic Too many large particles collected
Sample velocity too high nonisokinetic Too few large particles collected
Figure 7.10 Schematics of isokinetic and nonisokinetic sampling. (From Lodge, J. P., ed., Methods of Air Sampling and Analysis, 3rd ed., Chelsea, MI: Lewis Publishers, 1977. With permission.)
© 2011 by Taylor and Francis Group, LLC
149
Pollution Emissions
rate or more typically by adjusting a control valve in the sampling system. The challenge in isokinetic sampling is measuring both the sample gas velocity and the exhaust gas stream velocity. In most cases, the exhaust gas stream velocity varies across the exhaust duct and also varies as a function of time as changes in the combustion system cause the exhaust gas volume to vary due to changing process requirements. It is not sufficient to measure the exhaust gas velocity at the location where the sample probe will be located with, for example, a Pitot tube and assume that the exhaust gas velocity will be constant at all times and under all conditions. Isokinetic conditions are approximated when the static pressure is equal inside and outside a properly designed sample probe (see Figure 7.11). The combined Pitot-sample probe is used to measure the gas velocity in the vicinity of the probe and then to adjust the sample flow rate so that the average velocity at the sample probe inlet is equal to the gas velocity as measured by the Pitot probe. The gas velocity is measured using the Pitot probe by determining the pressure difference across the probe as shown in Figure 7.12. This can be done by simply using a differential pressure gauge (Figure 7.13), manometer (Figure 7.14), or differential pressure transmitter (Figure 7.15), and controlling the sample flow rate until the measured differential pressure is zero.
Another type of isokinetic sampling probe is shown in Figure 7.16. While this probe is more difficult to build because of the small passage sizes, it is easier to operate than the probe recommended by the U.S. EPA. Two static pressure measurements are made: one for the outer stream and one for the inner stream. These are connected to a differential pressure measuring device (see discussion above) and the gas sample flow is adjusted until the static pressure differential is zero, which means that the static pressure is the same inside and outside the probe. The static pressure holes on both the inside and outside of the probe go around the entire circumference at equally spaced distances apart. These holes are connected to the appropriate manifold inside the probe so that the static pressure measurements are actually averaged values around the circumference of the probe in case there are any significant flow deviations. The probe tip is aerodynamically designed to minimize disturbances to the out exhaust gas stream flow. Another important aspect of extractive sampling is proper conditioning of the sample stream. If the exhaust gas stream contains particulates, these can clog up the sample lines. Some type of high temperature filter (see Figure 7.17) or shield (see Figure 7.18) is often attached to the sample probe to prevent particulates from entering the sampling system. Figure 7.19 shows a schematic of a typical sampling system. In order to prevent further
Dt Type S pitot tube X ≥ 1.90 cm (2/4 in.) for D = 1.3 cm (1/2 in.) Sampling nozzle
Dn
A. Bottom view; showing minimum pitot tube-nozzle separation. Sampling nozzle
Sampling probe
Dt
Type S pitot tube
Static pressure opening plane
Impact pressure opening plane
Nozzle entry plane
B. Side view, to prevent pitot tube from interfering with gas flow streamlines approaching the nozzle. the impact pressure opening plane of the pitot tube shall be even with or above the nozzle entry plane. Figure 7.11 Isokinetic sample probe. (From U.S. EPA, 40 CFR 60: Standards of Performance for New Stationary Sources, Washington, DC: U.S. Environmental Protection Agency, 2001.)
© 2011 by Taylor and Francis Group, LLC
150
Industrial Combustion Testing
1.90 - 2.54 cm* (0.75 - 1.0 in.)
7.62 cm (3 in.)*
Temperature sensor
Leak-free connections
Type S pitot tube
*Suggested (interference free) pitot tube/thermocouple spacing
Manometer
Figure 7.12 Type S Pitot tube manometer assembly for measuring gas velocity. (From U.S. EPA, 40 CFR 60: Standards of Performance for New Stationary Sources, Washington, DC: U.S. Environmental Protection Agency, 2001.)
Figure 7.13 Differential pressure gauge. (From Baukal, C. E., Industrial Combustion Pollution and Control, Marcel Dekker, New York, 2004.)
reactions in the sampling system, the gas sample is typically quenched as quickly as possible to “freeze” the gas chemistry or the sample is maintained above the dew point to prevent water from condensing out until it reaches the water knockout unit. This is sometimes referred to as hot-wet sampling where the sample is maintained above the dew point until the sample reaches the condenser. Figure 7.20 shows an example of a heated sample probe to prevent catalytic reactions inside the probe itself. The sample is not diluted after it has been withdrawn from the exhaust gas stream. One common type of water knockout unit is an electronic water condenser (see Figure 7.21). Another type uses a permeation dryer where nitrogen on one side of
© 2011 by Taylor and Francis Group, LLC
Figure 7.14 Manometers for measuring differential pressure. (From Baukal, C. E., Industrial Combustion Pollution and Control, Marcel Dekker, New York, 2004.)
a membrane removes water from the gas sample on the other side of the membrane. The gas sample conditioning system can be purchased prepackaged as shown in Figure 7.22. It is also necessary to measure the sample flow rate to ensure that the proper flows are going to the analyzers. Figure 7.23 shows a sampling system that includes flow meters for measuring and controlling the sample flow rates to the analyzers. Another alternative to removing water from the gas sample prior to sending it to the analyzers is to dilute the sample so that the effective moisture content is not a problem for the analyzers. This is sometimes referred
151
Pollution Emissions
Flue gas stream
Outer stream pressure tap Outer stream static pressure To gas analyzers Inner stream static pressure
Gas sample
Inner stream pressure tap Figure 7.16 Nozzle for isokinetic sampling probe (not to scale). (From Baukal, C. E., Industrial Combustion Pollution and Control, Marcel Dekker, New York, 2004.)
Figure 7.17 Sample frit filter. (Courtesy of Millennium Instruments, Spring Grove, IL.)
Figure 7.15 Differential pressure transmitter. (From Baukal, C. E., Industrial Combustion Pollution and Control, Marcel Dekker, New York, 2004.)
Heated sample
Condenser
Water drain pump
Sample pump
Figure 7.18 Sample shield. (Courtesy of Millennium Instruments, Spring Grove, IL.)
Filter
Flowmeter
To analyzers To drain
Figure 7.19 Gas sample flow schematic.
to as cool-dry sampling where the sample is greatly diluted immediately after extraction from the exhaust gas stack. Specially designed probes are used to sample particle-laden flows according to EPA Method 5, as shown in Figure 7.24. Once the gas sample has been properly conditioned, it can be analyzed in one or more analyzers, depending on what gases are being measured. Figure 7.25 shows a single analyzer used to measure the oxygen concentration
© 2011 by Taylor and Francis Group, LLC
in the sample. A multigas analyzer is used to measure gases such as nitrogen oxides, sulfur oxides, oxygen, carbon dioxide, and carbon monoxide. There are a number of potential sampling prob lems [32]: • probes and lines clogging with contamination • heated lines failing in cold climates causing water to freeze and block lines
152
SP2000-H heated gas sample probe
Industrial Combustion Testing
• adsorption of pollutant to the wall, filter, tubing, or other components • adsorption of pollutant to the water, which is removed by a conditioning system 7.3.2 In-Situ Sampling
Figure 7.20 Heated gas sample probe. (Courtesy of M&C, Ratingen, Germany.)
In-situ sampling involves analyzing the exhaust gases in the exhaust stack without extracting them. In-situ analyzers are available for SO2, CO, O2, NOx, and CO2. They are also available for measuring opacity, particulates, and gas flow rate. In-situ sampling eliminates the need for sample transport and conditioning that are required in extractive sampling. There are two types of in-situ sampling: point and path. Point sampling measures at a specific point or location in the exhaust stack while path measures across the stack, giving an averaged value. Another important advantage of in-situ sampling is rapid response since the transport time in extractive sampling can be significant. However, there are some additional challenges of using in-situ sensors due to the harsh conditions that usually include high temperatures and may include particulates and high gas flow rates. Another potential problem is that the analytical equipment may be located in inconvenient locations, possibly at elevated locations on tall exhaust stacks, which can make maintenance and troubleshooting more difficult compared to extractive systems that are typically located in conditioned enclosures or rooms at ground level. In-situ sampling systems can also become plugged or clogged in exhaust gas streams with high particulate levels. One common type of in-situ sensor is made of zirconia (ZrO2), which is used for wet O2 measurements (e.g., see Figure 2.28). Light absorption techniques involving UV or IR spectrometers are also used to measure gases like CO, CO2, SO2, and NO. 7.3.3 Ambient Condition Effects
Figure 7.21 Electronic water condenser. (Courtesy of PERMA PURE LLC, Toms River, NJ.)
• probe filter causing loss of pollutant as it passes through the probe media (scrubbing) • dilution probe causing temperature, pressure, gas density effects, and water droplet evaporation when dilution air is added to the sample gas • water entrainment • leaks in the tubing or elsewhere in the system
© 2011 by Taylor and Francis Group, LLC
Bussman and Baukal [36] have shown that the ambient environment can have a significant impact on the flue gas composition, specifically the pollutants NOx and CO. The environment variables include ambient air temperature, relative humidity, barometric pressure, and wind velocity. Climatic changes in ambient air temperature and humidity can significantly impact the excess O2 in a heater or furnace. If the excess air is not properly controlled, it can ultimately lead to dramatic changes in NOx and CO emissions. When the temperature of the ambient air changes, its density also changes. For example, dry air at 60°F (16°C) and 14.7 psia has a density of 0.0765 lbm/ft3 (1.23 kg/m3). At 100°F (38°C) the density drops to 0.0710 lbm/ft3 (1.14
153
Pollution Emissions
SS-5 panel-mount sample conditioning system Figure 7.22 Prepackaged sample conditioner. (Courtesy of M&C, Ratingen, Germany.)
CSS-3 rack-mount compact sampling system
Figure 7.23 Sampling system. (Courtesy of M&C, Ratingen, Germany.)
Figure 7.24 Method 5 sampling probe. (Courtesy of Environmental Supply, Durham, NC.)
© 2011 by Taylor and Francis Group, LLC
154
9
400
8
350
7
300
6
Relative humidity 0% 10 % 80 % 0 6 40%
5 4 3 2
kg/m3). If a natural draft burner, operating at a given draft, experiences an increase in air density due to a decrease in temperature, it will pull more combustion air into the heater on a mass basis, resulting in higher excess O2. Conversely, if the air temperature increases, excess O2 would decrease. The presence of water vapor in the combustion air can also affect heater excess O2 levels. As the moisture in the air increases, the excess O2 drops because part of the combustion air is displaced by the water vapor. The amount of water vapor that air can hold depends on the temperature and pressure. When the air can’t hold all the water, then the moisture condenses as dew. Relative humidity is the term commonly used to describe the amount of water vapor in the air and is defined as the amount of moisture in the air compared to the maximum amount the air could hold at that temperature. Figure 7.26 shows the effects of temperature and relative humidity on the amount of water vapor in air. For example, at 100°F (38°C) and 100% relative humidity, air holds about 6.5% water vapor by volume, or about 275 grains of H2O per pound of dry air (7000 grains = 1 lbm = 0.454 kg). Figure 7.27 shows theoretical results for how the O2 concentration decreases with an increase in air temperature and relative humidity. These calculations are based on a constant furnace draft level with 3% excess O2 corresponding to air at 60°F (16°C) and 0% relative humidity (baseline point). The results show that the excess O2 drops off as the air temperature increases. Also notice that the variation of O2 is more dramatic with changes in air temperature as the relative humidity increases. Next, a couple of scenarios are given to demonstrate how these variations can result in dramatic changes in NOx and CO emissions. The American Petroleum Institute (API) Publication 535 [37] states that “NOx concentrations will increase as the excess oxygen increases in raw gas burners … this is
© 2011 by Taylor and Francis Group, LLC
150 100 50
30
40
50
60 70 80 90 100 Air temperature (degrees F)
110
120
0
Figure 7.26 Effects of temperature and relative humidity on the amount of water vapor in air at an atmospheric pressure of 14.7 psia (101 kPa). (From Bussman, W., and Baukal, C., “Ambient Condition Effects on Process Heater Emissions,” Proceedings of the Int’l Mechanical Engineering Congress & Exhibition, Paper IMECE2008-68284, Boston, MA, November 2008. With permission.) 3.5 3.0 Percent O2, dry volume
Figure 7.25 Oxygen analyzer. (Courtesy of AMETEK, Inc., Paoli, PA.)
200
20%
1 0
250
Humidity (grains per lb dry air)
Volume % water vapor in air
Industrial Combustion Testing
Relative humidity 0% 20% 40% 60 % 80 % 10 0%
2.5 2.0 1.5 1.0 0.5 0.0
30
40
50
60 70 80 90 100 Air temperature (degrees F)
110
120
Figure 7.27 Percent O2 as a function of ambient air temperature and relative humidity firing methane (or propane). (From Bussman, W., and Baukal, C., “Ambient Condition Effects on Process Heater Emissions,” Proceedings of the Int’l Mechanical Engineering Congress & Exhibition, Paper IMECE2008-68284, Boston, MA, November 2008.)
true for typical refinery heater excess oxygen (1–5% O2, wet basis) rates.” Figure 7.28 shows the API 535 curve (shaded area) demonstrating the effect of excess O2 on NOx emissions from raw gas burners. The NOx covers a fairly broad range at a given excess O2. This variation in data may be due to burner type, fuel composition, burner design, and so on. The API 531 curve does not characterize the effect of excess O2 on NOx emissions for all operating conditions. For example, Figure 7.28 also shows test results of a diffusion-type wall burner firing at 1.65 × 106 Btu/hr (0.483 MW) with two fuel compositions: (1) 100% natural gas
155
Pollution Emissions
Ratio of NOx at new condition to baseline condition
4.0
Diffusion wall burner 60% NG (vol.), 40% H2
3.5 3.0
Diffusion wall burner 100% NG
2.5 2.0
API 535
1.5
Pre-mix wall burner - No secondary air - H2 (10 to 70% vol.), bal. NG
1.0 0.5
0
1
2 3 4 5 6 7 Furnace O2 (percent dry on volume basis)
8
9
Figure 7.28 Effects of excess O2 on NOx. (From Bussman, W., and Baukal, C., “Ambient Condition Effects on Process Heater Emissions,” Proceedings of the Int’l Mechanical Engineering Congress & Exhibition, Paper IMECE2008-68284, Boston, MA, November 2008.)
(NG), and (2) 60% NG, 40% hydrogen (by volume). The data for each fuel are corrected for dilution to 3% O2 and a furnace temperature of 1600°F (870°C). When firing 100% NG, the effects of O2 on NOx follow the API curve fairly closely. However, when firing with the hydrogen mixture, the data deviate significantly from the API curve. These results show that increasing the hydrogen content increases the sensitivity of excess O2 on NOx production for this particular burner. For example, suppose this burner is operating at 3% O2 with NOx emissions of 20 ppmvd firing both the NG and high-hydrogen fuel cases. If the excess O2 were to increase to 6% at a constant furnace temperature, the NOx would increase to 27 and 32 ppmvd for the NG and high-hydrogen case, respectively. Experimental data clearly show that increasing the excess O2 results in an increase in NOx for diffusion-style burners. The increase is attributed to the combination of high flame temperatures and excess O2 available to combine with N2 to form NOx. Consider a hypothetical example demonstrating changes in NOx. Suppose during a plant start-up, operators tune a heater full of diffusion-style burners to 2% excess O2. The time of year is midsummer with an air temperature of 100°F (38°C) and 80% relative humidity. At night in the late summer months, the temperature drops to 50°F (10°C) and 0% relative humidity. If the operators fail to trim the O2 during this climate change, Figure 7.29 shows the excess O2 will increase from 2% to approximately 4%. Using the data from Figure 7.28, NOx will increase by 40% for the NG fuel case and 58% for the high-hydrogen fuel case. This demonstrates that dramatic changes in NOx can occur if excess O2 is not trimmed.
© 2011 by Taylor and Francis Group, LLC
Carbon monoxide is produced by the incomplete combustion of fuels containing carbon. The CO burns by reacting with O2 to form carbon dioxide (CO2). The CO is generally easy to convert into CO2 given sufficient temperature, oxygen, and mixing; however, it can be formed if there is insufficient oxygen available for complete combustion. Figure 7.30 shows the effects of excess O2 on CO emissions for a diffusion-style wall burner, where excess O2 is measured at the top of the radiant section. Data show that as the excess O2 decreases to about 1%, CO emissions start to increase. As the excess O2 approaches 0.5%, the CO production rapidly accelerates, which is often referred to as “CO breakthrough.” Consider a hypothetical example demonstrating changes in CO. Suppose plant operators working the nightshift tune a heater with diffusion-style burners to 2% excess O2. During adjustments, the ambient temperature is 60°F (16°C) and 0% relative humidity. As midday approaches, the weather conditions reach 100°F (38°C) and 80% relative humidity. If the dayshift operators fail to adjust the heater, the excess O2 will drop to nearly 0% (see Figure 7.31). Using the data from the diffusionstyle burner in Figure 7.30, CO breakthrough will occur resulting in high CO emissions out the stack. Operating a heater under these conditions poses a safety risk and can cause long-term heater problems. Another weather-related variable that can impact heater excess O2 is the atmospheric (barometric) pressure. Atmospheric pressure is closely approximated by the hydrostatic pressure caused by the weight of air above the measurement point. Low pressure areas have less air mass above their location, whereas high pressure areas have more air mass above their location. As the atmospheric pressure changes, the air density also changes due to the compression effect of the air mass. Natural variations in the atmospheric pressure occur as a consequence of changing weather conditions. For example, an incoming storm can cause a large drop in atmospheric pressure. As the atmospheric pressure drops, the density of the air (at a constant temperature) also drops. For a heater operating at a given draft, this would result in a reduction in excess O2 because the reduction in air density results in less combustion air going through the burner. Figure 7.32 shows the effects of atmospheric pressure on excess O2 where the baseline point is 3% O2 with ambient conditions of 60°F (16°C), 14.7 psia (101 kPa). Results show that as the atmospheric pressure decreases, the O2 also decreases. The rate of change of excess O2, with atmospheric pressure, is not significantly dependent on the air temperature. Weather data show that atmospheric pressure, at a given location, can vary as much as 0.6 psia (4 kPa) in a short period of time [38]. This corresponds to a change
156
Industrial Combustion Testing
4.0 Weather changes to these conditions: - Temperature = 50° F - Relative humidity = 0% - Atmospheric pressure = 14.7 psia
Percent O2, dry volume
5 4
Diffusion wall burner 60% NG (vol.), 40% H2
Ratio of NOx at new condition to baseline
6
3.5 3.0 2.5
3 2
30
40
50
60 70 80 90 Air temperature (degrees F)
Diffusion wall burner 100%NG
1.5
Heater tuned at these conditions: - Temperature = 100° F - Relative humidity = 80% - Atmospheric pressure = 14.7 psia
1 0
58% increase in NOx
2.0
40% increase in NOx
1.0
100
110
120
0.5
0
1
2 3 4 5 6 7 Furnace O2 (percent dry on volume basis)
8
9
Figure 7.29 Example demonstrating how variations in ambient air temperature and humidity can result in dramatic changes in NOx emissions. (From Bussman, W., and Baukal, C., “Ambient Condition Effects on Process Heater Emissions,” Proceedings of the Int’l Mechanical Engineering Congress & Exhibition, Paper IMECE2008-68284, Boston, MA, November 2008. With permission.) 3500
CO corrected to 3% O2
3000
Region of CO breakthrough
2500 2000 1500 1000 500 0
0
0.5
1
1.5
2
2.5
3
Heater excess O2 (dry volume) Figure 7.30 Experimental data for a diffusion-style burner showing the CO breakthrough. (From Bussman, W., and Baukal, C., “Ambient Condition Effects on Process Heater Emissions,” Proceedings of the Int’l Mechanical Engineering Congress & Exhibition, Paper IMECE200868284, Boston, MA, November 2008. With permission.)
in heater excess O2 of about 0.5% based on the results in Figure 7.32. A drop of 0.5% in excess O2 can be significant, especially, if a burner is operating near the CO breakthrough point. Varying winds, especially storm force winds, can cause dramatic changes in pressure inside the stack and at the burner inlet. This can significantly impact heater draft resulting in swings in excess O2. When wind blows over the top of a stack, as illustrated in Figure 7.33, it can contribute to either a back pressure or a suction pressure inside the stack. If the momentum of the flue gas is low relative to the wind, then a negative pressure inside of
© 2011 by Taylor and Francis Group, LLC
the stack can be formed. Since the heater box is in direct dynamic communication with the stack, this contributes to an increase in heater draft. On the other hand, if the flue gas momentum is dominant, the wind can create a back pressure resulting in a reduction in heater draft. Figure 7.34 shows experimental results for the effects of wind on heater draft and excess O2. Data were collected over a 10-minute period for a diffusion-style burner firing NG at a heat release of 10 × 106 Btu/hr (2.9 MW). Over that period, the burner damper was full-open and the stack damper was set at 58% open. The stack damper was exposed to the full force of the wind while the burner inlet was somewhat isolated from the wind due to equipment and structures located in the area. For the high-wind case, the speed ranged from about 8 to 35 mph (3.6 to 16 m/s) causing variations in draft of about 0.25 in. WC (0.062 kPa) and swings in excess O2 of about 1%. However, for the low-wind case, the speed varied from about 0 to 10 mph (0 to 4.5 m/s) causing variations in draft of 0.05 inches WC (0.012 kPa) and swings in excess O2 of only about 0.2%. Clearly, the wind blowing over a heater stack can have a significant impact on the draft and excess O2. Next, consider the effects of wind blowing past the air intake of a burner. Wind blowing past a burner intake can have a significant impact on the air flow rate into the heater. The impact depends on the burner intake design, and wind velocity and direction. Figure 7.35 shows several burners with a variety of intake designs. Due to the diverse air intake designs, burners can be affected by the wind differently. For example, the burner depicted in Figure 7.35b will be less susceptible to a headwind as compared to the burner in Figure 7.35c because the front plate acts as a barrier to protect it from the force of the wind.
157
Pollution Emissions
4000
Heater tuned at these conditions - Temperature air = 60° F - Relative humidity = 0% - Atmospheric pressure = 14.7 psia
2.5 2.0 1.5 1.0
Weather changes to these conditions - Temperature air = 100° F - Relative humidity = 80% - Atmospheric pressure = 14.1 psia
0.5 0.0
30
40
50
60
70
80
Region of CO breakthrough
3500 CO corrected to 3% O2
Heater excess O2 (percent dry volume)
3.0
3000 2500 2000 1500 1000 500
90
100
110
120
0
0
0.5
1
1.5
2
2.5
3
Heater excess O2 (percent dry volume)
Air temperature (degrees F)
Figure 7.31 Example demonstrating how variations in ambient temperature and humidity can result in dramatic changes in CO emissions. (From Bussman, W., and Baukal, C., “Ambient Condition Effects on Process Heater Emissions,” Proceedings of the Int’l Mechanical Engineering Congress & Exhibition, Paper IMECE2008-68284, Boston, MA, November 2008. With permission.) 4.0 Percent O2, dry volume
3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 13.0
Wind flowing over the top of a stack con contribute to either a heater back pressure or a suction pressure depending on flow rates
Baseline point at 3% O2 60°F, 14.7 psia = 32°F rature e p m Air te 0°F re = 6 eratu p m e Air t °F = 100 ature r e p m Air te
13.5 14.0 14.5 Atmospheric pressure (psia)
Flue gas Wind
15.0
Figure 7.32 Calculated percent O2 as a function of atmospheric pressure firing methane (or propane). (From Bussman, W., and Baukal, C., “Ambient Condition Effects on Process Heater Emissions,” Proceedings of the Int’l Mechanical Engineering Congress & Exhibition, Paper IMECE2008-68284, Boston, MA, November 2008. With permission.)
A diffusion-style burner, firing NG at its design maximum heat release of 10 × 106 Btu/hr (2.9 MW), was tested in a crosswind direction with the damper set at 100% open. Wind speeds up to 35 mph (16 m/s) were simulated using a variable speed blower positioned perpendicular to the burner intake as illustrated in Figure 7.36. During all tests, the ambient wind speed remained less than 2 mph (0.9 m/s) and therefore, had an insignificant affect on the test results. Experimental data show that for this particular burner design, a crosswind of 20 mph (8.9 m/s) did not have a significant impact on the excess O2; however, at 35 mph (16 m/s) it did. At a wind speed of 35 mph (16 m/s), the excess O2 varied from 0.5 to 2.3% over a period of five minutes with an average of about
© 2011 by Taylor and Francis Group, LLC
Wind flowing past a burner inlet can create either a positive or negative pressure depending on burner orientation
Heater Wind
Figure 7.33 Illustration showing ways the wind can impact heater excess O2. (From Bussman, W., and Baukal, C., “Ambient Condition Effects on Process Heater Emissions,” Proceedings of the Int’l Mechanical Engineering Congress & Exhibition, Paper IMECE2008-68284, Boston, MA, November 2008. With permission.)
1.5% excess O2. During this time period, the CO emissions varied from 0 to about 200 ppmvd (uncorrected for O2). One would expect that as the wind speed further increased, the heater would operate in a fuel rich condition. Figure 7.37 demonstrates the effects of air humidity on NOx emissions for two burner styles. The circular
158
Industrial Combustion Testing
Low wind condition Wind speed (mph)
30 20 10
30 20 10 0
0.75
0.75 Floor draft (inches W.C.)
0
0.50
0.50
0.25
0.25
0
0
3 2 1 0
High wind condition
40
Excess O2 (dry volume)
Excess O2 (dry volume)
Floor draft (inches W.C.)
Wind speed (mph)
40
3 2 1 0
Figure 7.34 Data trends showing wind effects on heater draft and excess O2 at high and low wind speeds over a 10-minute period. (From Bussman, W., and Baukal, C., “Ambient Condition Effects on Process Heater Emissions,” Proceedings of the Int’l Mechanical Engineering Congress & Exhibition, Paper IMECE2008-68284, Boston, MA, November 2008. With permission.) (a)
(b)
(c) Air intake
Air intake
Air intake Air intake
(d)
(e)
(f )
Air intake Air intake
Air intake Figure 7.35 Illustrations showing burner designs with various air intake configurations. (From Bussman, W., and Baukal, C., “Ambient Condition Effects on Process Heater Emissions,” Proceedings of the Int’l Mechanical Engineering Congress & Exhibition, Paper IMECE2008-68284, Boston, MA, November 2008. With permission.)
symbols represent data from a large industrial heater full of premix, radiant wall burners (data for both high and low NOx burners) [39]. The triangular symbols represent experimental data firing a single, diffusionstyle, wall burner. All of the data are normalized to a NOx value corresponding to zero humidity. The NOx value, at zero humidity, is obtained by extrapolating
© 2011 by Taylor and Francis Group, LLC
a best-fit exponential curve through the experimental data. Experimental data clearly show that for both the premix and diffusion-style burners, NOx decreases with an increase in air humidity. For example, a variation in air humidity from 0 to 120 grains corresponds to a 45% change in NOx for the high NOx burner. As previously discussed, this represents a change in the
159
Pollution Emissions
Average excess O2 (% dry)
4.0
Excess O2 = 0.5 to 2.3% CO = 0 to 200 ppmvd
3.5 3.0
Air intake
2.5 2.0 1.5 1.0 0.5 0.0
Wind direction 0
10
20 30 Average wind speed (mph)
40
Figure 7.36 Experimental data showing the effects of a crosswind past a burner intake on excess O2. (From Bussman, W., and Baukal, C., “Ambient Condition Effects on Process Heater Emissions,” Proceedings of the Int’l Mechanical Engineering Congress & Exhibition, Paper IMECE2008-68284, Boston, MA, November 2008. With permission.)
Figure 7.38 Control motor on each burner. (From Bussman, W., and Baukal, C., “Ambient Condition Effects on Process Heater Emissions,” Proceedings of the Int’l Mechanical Engineering Congress & Exhibition, Paper IMECE2008-68284, Boston, MA, November 2008. With permission.)
NOx, normalized to zero humidity
1.1 1.0 0.9 0.8 Shaft connecting burners
0.7 0.6
Field data - Low NOx, premix, radiant wall burners Field data - High NOx, premix, radiant wall burners Test data - Wall-fired, diffusion burner Exponential fit of Low NOx, premix data Exponential fit of High NOx, premix data Exponential fit of Wall-fired, diffusion data
0.5 0.4
0
20
40 60 80 100 Humidity, (grains per lb dry air)
120
140
Figure 7.37 Experimental data and model results showing the effects of air humidity on NOx emissions. (Adapted from Bussman, W., and Baukal, C., “Ambient Condition Effects on Process Heater Emissions,” Proceedings of the Int’l Mechanical Engineering Congress & Exhibition, Paper IMECE2008-68284, Boston, MA, November 2008. With permission.)
theoretical flame temperature of only about 2.3%. The data also show the percentage NOx reduction is more pronounced for the high NOx burner as compared to the low NOx burner. Based on theoretical results previously discussed, the flame temperature from a low NOx burner is less sensitive to variations in air humidity and hence, changes in NOx. This section has shown that ambient conditions can significantly impact NOx and CO emissions. Tight control of burner operating conditions is essential to minimize these emissions. This can be done in two ways: manual or automatic control. Manual control means that operators must frequently adjust dampers (air registers) on every burner as ambient conditions change. It is not enough to adjust only some of the burners to
© 2011 by Taylor and Francis Group, LLC
Burners connected by shaft Figure 7.39 Multiple burners mechanically connected by a shaft to a single control motor. (From Bussman, W., and Baukal, C., “Ambient Condition Effects on Process Heater Emissions,” Proceedings of the Int’l Mechanical Engineering Congress & Exhibition, Paper IMECE200868284, Boston, MA, November 2008. With permission.)
achieve the target O2 level. While this could be done, the flames would not be uniform, which could lead to multiple operational problems such as uneven heating and flame impingement on process tubes. Some heaters may have dozens of burners and each operator is normally responsible for multiple heaters. Because they already have many other responsibilities, this is not a practical option for most plants. The other method is automatic control where the excess O2 is continuously adjusted to a target level. The two ways this is commonly done are using a control motor on each burner (see Figure 7.38) or with some type of mechanical linkage connecting multiple burners to a single control motor (see Figure 7.39). The motors may
160
Industrial Combustion Testing
7.4 Testing Equipment This section briefly considers the equipment commonly used to measure variables of importance when quantifying pollution emissions. These variables include the exhaust gas flow rate, the gas composition, and particulate emissions (solids in the gas stream). Note that there are many other variations that may be used in a given installation and new types of equipment are constantly being developed to make these measurements easier, more repeatable, and more accurate.
7.4.1.1 Pitot Tube The EPA Method 2 discusses how to determine the volumetric stack gas flow rate using an S-Type Pitot tube that includes a temperature sensor (see Figure 7.40). The method only applies to flows that are not swirling or cyclonic. This method combines measurements and calculations to determine the gas flow rate. Specific guidelines are given for what materials should be used and what dimensions are critical. Each Pitot probe must be calibrated with a known coefficient that should be engraved on the probe. The static and dynamic measuring holes on the probe must not be plugged, which can be a problem in particulate-laden flows. The pressure “head” or difference between the static and dynamic measurements is typically very small and can be measured with an inclined manometer (see Figure 7.14) or a similar device like a magnehelic gauge. Both of these need to be calibrated to ensure accuracy. The equation governing the response of a Pitot probe is commonly given as (Becker and Brown 1974) [40]:
7.4.1 Exhaust Gas Flow
Qi = vi Ai ,
(7.7)
where Qi is the gas flow rate through sub area i, vi is the gas velocity through the sub area, and Ai is the area. The sum of all the subareas equals the total cross-sectional area of the duct: n
A=
∑A , i
(7.8)
i=1
where A is the total duct cross-sectional area and n is the total number of subareas. Then, the total exhaust gas flow rate equals the sum of the individual gas flow rates: n
Q=
∑Q . i
i=1
© 2011 by Taylor and Francis Group, LLC
(7.9)
v=
2( pt − pst ) C pρ
(7.10)
where pt is the total or impact pressure, pst is the static pressure, and Cp is the probe calibration constant. According to EPA Method 2, the average stack gas velocity can be calculated for an S-Type Pitot tube using:
vs = K pCp
( pt − pst )Ts , ps Ms
(7.11)
where vs = average stack gas velocity (ft/s) Kp = velocity equation constant Cp = Pitot tube probe constant
D
10 D
The exhaust gas flow is usually not directly measured but is often determined through a combination of measurements, calculations, and assumptions. For example, the gas velocity is measured at one or more points across the exhaust duct. Since it is not known a priori if the gas sample is homogeneous, it must be assumed that there is some variation across the exhaust gas duct until proven otherwise. This is done by making multiple traverses across the duct. The cross-sectional area of the duct is subdivided into smaller areas for the purposes of making measurements. The gas velocity in each subarea is assumed to be constant. Then the gas flow rate through that subarea is simply:
4 static holes 3/8 D Impact opening 1/2 D
4D
be electrically or pneumatically actuated. Quick disconnects are normally included so individual burners can be quickly and easily manually controlled or taken out of service if necessary.
Figure 7.40 Standard hemispherically nosed Pitot tube from 40 CFR 60 Method 2C. (From U.S. EPA, 40 CFR 60: Standards of Performance for New Stationary Sources, Washington, DC: U.S. Environmental Protection Agency, 2001.)
161
Pollution Emissions
pt – pst = difference between total and static pressures measured with Pitot tube (in. H2O) Ts = absolute stack temperature (°R) Ps = absolute stack pressure (in. Hg)
Ms = Md (1 − Bws ) + 18Bws ,
(7.12)
where Ms = molecular weight of wet stack gas (lb/lb-mole) Md = molecular weight of dry stack gas (lb/lb-mole) Bws = fraction by volume of water in the gas stream (determined by Methods 4 or 5). The EPA Method 2C concerns determination of the exhaust gas flow rate using a standard hemispherically nosed Pitot probe as shown in Figure 7.40. The previous equations for the S-Type Pitot probe apply to the standard Pitot tube, although the probe constant will be different. Becker and Brown [40] studied the response of Pitot probes in turbulent nonreacting gas streams. The following terminology is used to specify the recommendations of their study: d1 = total pressure hole i.d. d2 = probe o.d. d3 = static pressure hole i.d. l1 = internal length before any change in diameter or direction of total impact tube l2 = external length before any change in diameter or direction of probe l3 = length from probe tip to centerline of static pressure hole In that study, the following conditions were assumed:
1. The probe is long enough (l2/d2 > 6) so that the downstream geometry has a negligible effect on the response. 2. The internal geometry (l1/d1 > 3) has a negligible effect on the response. 3. Compressibility effects are small (Mach number < 0.3).
Cho and Becker [41] recommend a single static pressure inlet. A thermocouple or similar device can be used to measure the gas temperature that is needed to compute the gas density. A barometer is needed to measure the atmospheric pressure, also needed to compute the gas density. In lieu of a barometer, the barometric pressure can be obtained from the National Weather Service. According to EPA Method 2, the average dry stack gas volumetric flow rate is calculated as follows:
© 2011 by Taylor and Francis Group, LLC
T P Q = 3600(1 − Bws )vs A std s Ts Pstd
(7.13)
where Q = is the volumetric flow rate in dry standard cubic feet per hour (dscf/hr) Bws = fraction by volume of water in the gas stream (determined by Methods 4 or 5) vs = average stack gas velocity (ft/s) A = cross-sectional area of the stack (ft2) Tstd = standard absolute temperature = 528°R Ps = absolute stack pressure (in. Hg) Ts = absolute stack temperature (°R) Pstd = standard absolute pressure (29.92 in. Hg). The EPA Methods 2G and 2F discuss the use of twoand three-dimensional Pitot probes, respectively, which are useful for swirling or cyclonic flows, but are used infrequently in industrial combustion applications and are therefore not discussed further here. The calibration procedures are considerably more difficult than those for one-dimensional Pitot probes. The EPA Method 2H discusses how to determine a correction for the velocity decay near the wall in stacks greater than 1.0 m (3.3 ft) in diameter. Two procedures are discussed: one uses velocity measurements and the other uses a generic adjustment factor. 7.4.1.2 Direct Measurement The EPA Method 2A discusses the direct measurement of the gas volume flow rate through pipes and small ducts. These devices include, for example, positive displacement meters and turbine meters. This is a much simpler and faster technique than measuring multiple traverse points in the exhaust gas stack and then computing the flow rate based on several different types of measurements (pressure difference, gas temperature, barometric conditions, and moisture in the stack gases). Direct measuring devices also need to be properly calibrated to ensure accuracy. These usually cost much more than using Pitot tubes and manometers but require far less labor to operate. They can also mitigate the problem of varying conditions across the exhaust duct. However, they are typically only used on smaller diameter ducts and they may introduce a significant pressure drop in the exhaust system that may require either additional fan power or a reduction in the exhaust gas volume flow rate, which is usually not desirable. The EPA Method 2D discusses the measurement of the gas volume flow rates in small pipes and ducts using rotameters, orifice plates, or other similar devices to measure flow rate or pressure drop. A rotameter consists
162
Figure 7.41 Orifice plate flow measurement system. (From Baukal, C. E., Industrial Combustion Pollution and Control, Marcel Dekker, New York, 2004.)
of a glass vertical cylinder with a slight taper inside and some type of float. There are graduated markings along the length of the cylinder. The gas flow comes in from the bottom and causes the float to rise proportionally to the flow rate. Appropriate curves are then used for the particular gas flowing through the rotameters to determine the flow rate. Orifice plates (see Figure 7.41) measure the pressure drop across a known and precise hole or orifice. Using the upstream pressure and the gas temperature and properties, the gas flow rate can then be calculated. 7.4.1.3 Calculation This is a simple and straightforward method for calculating the average exhaust gas flow rate, which is often of sufficient accuracy for many situations. As will be shown, this method uses a combination of measurements, assumptions, and calculations to determine the gas flow rate. There are also some variations of the technique as will be shown. However, in general, the technique involves calculating the gas flow rate based on either the measured input flow rates, on the measured stack gas composition, or a combination of the two as a check to make sure that the results make sense. In nearly all industrial combustion systems, the fuel flow rate is measured. In many cases it must be measured in order to calculate emissions according to regulatory requirements. For example, some air permits are based on a given maximum fuel flow rate. For some applications that do not have continuous emissions monitoring systems, an emission factor may be assumed using EPA AP-42 [42]. These factors
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
are usually calculated based on an emission rate per unit fuel input (e.g., lb NO2/106 Btu-input). If there is a total emission limit on the plant, then the maximum total fuel flow rate can be calculated. This type of permit often requires regular reports on the fuel consumption that is used to calculate the estimated emissions. The combustion air flow rate is also sometimes measured. Together with the fuel flow rate, the estimated exhaust gas flow rate can be calculated by using an O2 measurement in the exhaust stack to estimate air infiltration into the furnace (assuming that the measured O2 is higher than the calculated O2). While measuring the combustion air flow may not be necessary to calculate the exhaust gas flow, measuring only the O2 in the stack (either wet or dry) only tells how much air has gotten into the furnace. It does not tell where it came from so it is possible that the burners are running fuel rich with a large amount of air infiltration. Therefore, it is usually desirable to measure the combustion air flow to ensure the operating conditions of the burners. 7.4.1.4 Non-Contact Optical sensing systems are available for measuring the exhaust gas flow rate. These are similar to direct measurement techniques except that those are normally thermal or mechanical, as opposed to optical. A schematic of an optical flow measurement system is shown Figure 7.42. A photograph of an optical sensor is shown in Figure 7.43. Figure 7.44 shows an ultrasonic time of flight instrument for non-intrusively measuring flows in small and large stacks and ducts. The velocity measurement is independent of the flue gas temperature, pressure, and density. 7.4.2 Gas Composition The EPA Method 3 is used to determine the dry molecular weight of an exhaust gas stream. This is needed, for example, to calculate the exhaust gas flow rate used in calculating the pollutant emission rates. The method specifically refers to determining the CO2 and O2 concentrations in a gas stream. The technique uses single-point grab sampling (batch extractive sample), single-point integrated sampling, or multipoint integrated sampling. The sample is typically analyzed using an Orsat analyzer. This method is not commonly used in industrial combustion applications where instrument analyzers are typically used. The Method 3A concerns the determination of O2 and CO2 using instrument analyzers. A typical O2 analyzer is shown in Figure 7.45. This method discusses test and calibration procedures. Method 3B is specific to
163
Pollution Emissions
Airflow
Stack or duct
Control panel
Power Customer data monitoring
Output
40 CFR part 75 requires measurement of gas flow to obtain mass emissions. The optical flow sensor makes a driftfree measurement across the entire stack or duct and calculates an average reading without contacting the gas stream. Thus giving a true cross-stack flow measurement of the process.
Figure 7.42 Schematic of an optical exhaust gas flow sensing system. (Courtesy of OSI, Gaithersburg, MD.)
Figure 7.43 Optical flow sensor for measuring stack gas flow rates. (Courtesy of OSI, Gaithersburg, MD.)
gas streams that contain significant quantities of gases other than O2, CO2, CO, and N2 (excluding water, which can be easily removed from the stream) that could affect the gas analysis results. For example, significant quantities of SO2 or HCl can interfere with CO2 measurements using an Orsat analysis. Method 4 is used to determine the moisture content in an exhaust gas stream. This is important in determining the total exhaust gas volume flow rate. A gas sample is removed from the exhaust gas stream at a constant flow rate. Moisture is removed from the sample and determined either volumetrically or gravimetrically. The sample probe and transport lines are
© 2011 by Taylor and Francis Group, LLC
Figure 7.44 In-situ exhaust gas flow meter. (Courtesy of Teledyne Monitor Labs, Denver, CO.)
Figure 7.45 O2 instrument analyzer. (From Baukal, C. E., Industrial Combustion Pollution and Control, Marcel Dekker, New York, 2004.)
164
Industrial Combustion Testing
heat traced to prevent water from condensing out of the sample until it reaches the condenser, which might be an electronic device (see Figure 7.21) or an ice bath for example. Figure 7.46 shows a schematic of the recommended sampling train used to measure the moisture content in an exhaust gas stream. The EPA Method 6 provides procedures for measuring sulfur dioxide emissions from stationary sources where the gas sample is extracted from the exhaust stack. Ammonia, water-soluble cations, and fluorides cause interferences with SOx measurements. Method 6A concerns sulfur dioxide, moisture, and carbon dioxide measurements from fossil fuel combustion sources by chemically separating the SO2 and CO2 components, where different reagent chemicals are used. Method 6C discusses the use of instrument analyzers to measure Filter (either in stack or out of stack)
Stack wall
sulfur dioxide emissions from stationary sources. This is the most commonly used method in industrial combustion processes. These analyzers typically use ultraviolet (UV), nondispersive infrared (NDIR), or fluorescence techniques. A schematic of an accepted sampling system is shown in Figure 7.47. Method 8 discusses determining sulfuric acid and sulfur dioxide emissions from stationary sources. The EPA Method 7 concerns the determination of nitrogen oxide emissions from stationary sources. This method is based on chemical separation using a reagent from a grab sample taken from the exhaust stack. Methods 7A and 7B give procedures for the use of ion chromatography and ultraviolet spectrophotometry, respectively, for measuring nitrogen oxide emissions, also from a grab sample. Methods 7C and 7D discuss
Condenser-ice bath system including silica gel tube Probe
Check valve Vacuum line
Temperature sensors
Vacuum gauge
Orifice
Dry gas meter
By-pass valve
Main valve Air-tight pump
Figure 7.46 EPA Method 4 moisture sampling train: reference method. (From U.S. EPA, 40 CFR 60: Standards of Performance for New Stationary Sources, Washington, DC: U.S. Environmental Protection Agency, 2001.) Excess sample vent Needle valve Sample by-pass vent
Rotameter Midget impingers
3% H2O2 (15 ml each)
Drying tube
Dry gas meter Ice bath
Figure 7.47 Method 6C sampling system schematic for SO2 measurements. (From U.S. EPA, 40 CFR 60: Standards of Performance for New Stationary Sources, Washington, DC: U.S. Environmental Protection Agency, 2001.)
© 2011 by Taylor and Francis Group, LLC
165
Pollution Emissions
the use of the alkaline permanganate/calorimetric and alkaline permanganate/ion chromatographic methods, respectively, with extractive samples from the exhaust stack. Method 7E discusses the use of instrument analyzers (see Figure 7.48) to measure NOx emissions from gas samples continuously extracted from the exhaust stack. This is the most commonly used technique in industrial combustion applications. The EPA Method 10 discusses measuring carbon monoxide emissions from stationary sources from continuous samples extracted from an exhaust stack where the sample is measured with nondispersive infrared (NDIR) analyzer. Possible interferences include water, carbon dioxide, and carbon monoxide. Method 10A tells how to make certified carbon monoxide measurements from continuous emission monitoring systems (CEMS) at petroleum refineries. The EPA Method 18 discusses measurement techniques for determining gaseous organic compound emissions from exhaust stacks. The technique involves separating the major organic compounds from the
Figure 7.48 Typical NOx analyzer. (From Baukal, C. E., Industrial Combustion Pollution and Control, Marcel Dekker, New York, 2004.)
sample and then individually quantifying the constituents using flame ionization, photoionization, electron capture, or other appropriate techniques. Diluting the sample is also used in certain cases to mitigate the effects of moisture. Method 25 is a similar process for determining the total gaseous nonmethane organic (volatile organic compounds [VOCs]) emissions from a stack. These measurements are made on a continuous gas sample with a flame ionization detector (FID). Carbon dioxide and water vapor can potentially interfere with the measurements. Figure 7.49 shows a schematic of the recommended sampling train to determining VOC emissions. Method 25A concerns measuring VOCs with a flame ionization analyzer (FIA) for determining the concentrations of samples containing primarily alkanes, alkenes, and/or arenas. Method 25B concerns the use of an (NDIR) analyzer to measure VOCs consisting primarily of alkanes. The EPA Method 23 discusses how to measure the emissions of dioxins and furans from stationary sources. A sample is withdrawn isokinetically and collected on a glass fiber filter. The sample is then separated using high resolution gas chromatography and measured by high resolution mass spectrometry. A schematic of an approved gas sampling train is shown in Figure 7.50. The condenser and adsorbent trap must be specially designed for this application. Specific filters, reagents, adsorbents, and sample recovery are required. The sample analysis is also extensive. These types of measurements should be done by qualified professionals to ensure high accuracy, especially because of the toxic nature of the constituents and the strict regulatory requirements. Regulating valve
Dual range rotameter Thermocouples Temperature controller
Manometer Vaccum pump
Purge valve
Flow control valve
Thermocouple
Sample tank valve
Sample valve Rotameter
Stainless steel filter holder Stainless steel probe
Condensate trap
Heated box Dry ice
Sample tank
Stack wall Figure 7.49 EPA Method 25 sampling train schematic for measuring volatile organic compounds. (From U.S. EPA, 40 CFR 60: Standards of Performance for New Stationary Sources, Washington, DC: U.S. Environmental Protection Agency, 2001.)
© 2011 by Taylor and Francis Group, LLC
166
Industrial Combustion Testing
Stack wall
Temperature sensor
Condenser
Temperature sensor
Heated glass liner Heated area
XAD-2 trap
Gas flow
“S” type pilot tube Inclined manometer
Temperature sensor
Filter holder Recirculation pump
Ice 100 ml HPLC water Empty
Gas exit Inclined manometer
Dry gas meter Calibrated orifice
Fine
Silica gel Empty coarse
Vacuum gauge P
Vacuum pump
Figure 7.50 EPA Method 23 sampling train schematic for measuring dioxins and furans. (From U.S. EPA, 40 CFR 60: Standards of Performance for New Stationary Sources, Washington, DC: U.S. Environmental Protection Agency, 2001.)
7.4.3 Particulates The EPA Method 5 is used to determine particulate emissions from stationary sources. Method 5B refers to the determination of nonsulfuric acid particulate matter emissions from stationary sources. Method 5D concerns determining particulate matter emissions from positive pressure fabric filters, also known as baghouses, where this method has particular emphasis on the sample location. Method 5E describes the procedure for determining particulate emissions from fiberglass wool manufacturing. Method 5F concerns nonsulfate particulate emissions from stationary sources, which includes a modified method for separating the sample from the filter. Method 5I concerns the determination of low concentrations of particulate emissions from stationary sources. This method differs from Method 5 with the following modifications: improved sample handling procedures, use of a lightweight sample filter assembly, and the use of low residue grade acetone to recover the particulate matter collected on the filter. For particulate measurements, the gas sample must be withdrawn isokinetically (see Section 7.3.1) to ensure that a representative sample of particles is withdrawn from the gas stream. If the extraction rate is too high, then the measured particulate concentration will be too
© 2011 by Taylor and Francis Group, LLC
high. If the extraction rate is too low, then the measured particulate concentration will be too low. The construction of the sampling system is also important to ensure water and particulates do not come out of the extracted sample until it reaches the measurement location in the system. For example, the probe should be lined with borosilicate or quartz glass tubing and heated to prevent condensation and to prevent the particulates from coating the walls. The extracted sample containing particulates passes through a glass fiber filter maintained at a temperature of 120 ± 14°C (248 ± 25°F). A schematic of a typical particulate sampling train is shown in Figure 7.51. The sample is extracted for a known period of time, long enough to collect a sample of sufficient mass. Moisture is removed from the particulate sample captured on the filter. The dried sample and filter are then weighed. This is compared to the weight of the clean filter prior to sampling. The difference in the weights then gives the weight of particulates captured. Using the measured sample extraction time, sample flow rate, and total exhaust gas flow rate, the particulate matter emission rate can then be calculated. The EPA Method 9 concerns the visual determination of opacity from emissions from stationary sources. The opacity may be caused by particulates and/or by steam. This method requires a certified observer trained in
167
Pollution Emissions
Temperature sensor Probe
Impinger train optional, may be replaced by an equivalent condenser
Type S pitot tube Temperature sensor Gooseneck nozzle
Temperature sensor
Temperature sensor
Heat traced glase-lined probe
Impingers Check valve Glass filter holder
Type S pitot tube
Ice water bath
Heated area Stack wall
Manometer Temperature Water sensor
Empty
Vacuum line Silica gel
Orifice By-pass valve Dry gas meter
Vacuum gauge
Main valve Air-tight pump
Figure 7.51 Method 5 particulate sampling train schematic. (From U.S. EPA, 40 CFR 60: Standards of Performance for New Stationary Sources, Washington, DC: U.S. Environmental Protection Agency, 2001.)
this measurement technique. An alternative technique known as LIDAR based on light backscattering can be used to determine opacity remotely as described in Alternate Method 1. Method 18 discusses a particulate matter measurement technique that can be used in lieu of Method 5 when it is known that there are no temperature effects so that the measurements can be made directly in the exhaust stack. Method 22 discusses the visual determination of smoke emissions from flares. This method does not require a certified observer as in Method 9 as the opacity is not quantified. It is merely a visual determination if smoke is present from flaring (see Chapter 28).
7.5 Source Testing Source testing involves making “official” pollution emission measurements used either to determine permit levels or to determine compliance with an air permit. Often an outside, third-party company specializing in making these measurements is hired. These companies
© 2011 by Taylor and Francis Group, LLC
usually must be certified by a recognized agency to make these measurements. The company will usually bring a truck to the site equipped with the appropriate measurement and calibration equipment. This source testing must be done in applications where there is no certified continuous emission monitoring system. It may also be done even where there is a CEMS as an independent verification. It is also used by equipment vendors trying to certify new technology. For example, a burner manufacturer may hire a firm to make official measurements to demonstrate reduced pollution emissions compared to other technologies. Figure 7.52 shows an example of measurements being made from an exhaust stack. The frequency of source testing depends on a number of factors. Measurements are usually made when an air permit is being established for the first time and whenever significant modifications are made to an existing permitted system. Annual verification tests may also be required, especially where no emissions monitoring system exists. These verification tests may not require “official” third-party measurements, depending on the rules of the given governing agency. An inspector may make unannounced spot checks with a portable
168
Industrial Combustion Testing
analyzer to verify compliance, which is discussed below in Section 7.7. An alternative to source testing is to measure the emissions of the burner in a test facility (see Figure 7.53). Figure 7.54 shows a pilot-scale test furnace that could be used for regulatory source monitoring. While this is not always permissible, depending on the rules of the local regulatory agency, it may be used to establish permit limits. It may also be used in lieu of expensive field source tests that may not be economically practical for smaller emission sources.
Figure 7.52 Measuring emissions from an exhaust stack.
Figure 7.53 John Zink R&D Test Center. (Courtesy of John Zink Co., LLC.)
© 2011 by Taylor and Francis Group, LLC
Figure 7.54 Test furnace to simulate a section of an ethylene cracking furnace. (Courtesy of John Zink Co., LLC.)
Pollution Emissions
169
7.6 Continuous Monitoring Depending on regulations and the size of the emission source, some industrial combustion processes may be required to have a continuous emissions monitoring system. For example, some air permits may be written in such a way as to limit emissions over some time interval, such as a 24-hour rolling average. Another common limit is that emissions of a particular pollutant may not exceed a given concentration at any time. There are many other variations as well. In order to determine whether a plant is in compliance, a certified continuous emission monitoring system may be required. These are not only fairly expensive to purchase, but also have rigorous and ongoing maintenance and calibration requirements to ensure high accuracy and reliability. This is compared to other pollutant emission sources that may only require an annual test. In some cases, only an initial test may be required after the equipment is first started or after any significant modifications have been made. The U.S. EPA has developed a series of performance specifications (PS) for continuous emissions monitoring systems (see Figure 7.55) for measuring a wide range of pollutant emissions [43]. These specifications are listed in Table 7.3. Note that two of the specifications (4 and 4A) have identical titles. PS-4 is for the general measurement of CO emissions while PS-4A is for measuring lower concentrations (< 200 ppmv) of CO emissions. These are to be used in conjunction with the EPA methods discussed above. The performance specifications include discussions of calibration procedures and relative accuracy requirements. The EPA has also developed quality assurance procedures to be used for compliance determination [44]. Parker (1995) lists the following factors as important in a continuous monitoring program [45]:
1. knowledge of the process to be monitored
2. knowledge of the relevant regulatory requir ements
3. using the correct monitors
4. using appropriately trained personnel with the proper calibration, quality control, and quality assurance procedures
5. relevant data collection, review and repor ting
6. ongoing maintenance program
7. comprehensive cost tracking, control, and budgeting
© 2011 by Taylor and Francis Group, LLC
Figure 7.55 Continuous emission monitoring system. (Courtesy of Horiba Instruments, Inc., Irvine, CA.)
The three major components of a continuous emissions monitoring system are: (1) the sampling and conditioning system, (2) the gas analyzers and monitors, and (3) the data acquisition system (DAS). The first two have been discussed above. The DAS may be as simple as manually recording emissions data at scheduled and regular time intervals. It may be as sophisticated as a completely automated computerrecorded data management system. It may also be some combination of the two where some of the data is manually recorded while other data is automatically recorded by a computer or some similar type of electronic device. For example, some of the emissions such as particulate emissions may not be easily collected on a continuous basis while other emissions such as NOx are easily collected through continuous monitoring.
170
Industrial Combustion Testing
Table 7.3 EPA Performance Specifications for Continuous Emission Monitoring System (40 CFR 60: Appendix B) Specification # 1 2 3 4 4A 4B 5 6 7 8 8A 9 15
Specification Title Specifications and test procedures for continuous opacity monitoring systems in stationary sources Specifications and test procedures for SO2 and NOx continuous emission monitoring systems for stationary sources Specifications and test procedures for O2 and CO2 continuous emission monitoring systems for stationary sources Specifications and test procedures for carbon monoxide continuous emission monitoring systems for stationary sources Specifications and test procedures for carbon monoxide continuous emission monitoring systems for stationary sources Specifications and test procedures for carbon monoxide and oxygen continuous emission monitoring systems for stationary sources Specifications and test procedures for TRS (total reduced sulfur) continuous emission monitoring systems for stationary sources Specifications and test procedures for continuous emission rate monitoring systems for stationary sources Specifications and test procedures for hydrogen sulfide continuous emission monitoring systems for stationary sources Specifications and test procedures for volatile organic compound continuous emission monitoring systems for stationary sources Specifications and test procedures for total hydrocarbon continuous emission monitoring systems for stationary sources Specifications and test procedures for gas chromatographic continuous emission monitoring systems for stationary sources Specifications and test procedures for extractive FTIR (Fourier transform infrared) continuous emission monitoring systems for stationary sources
Source: U.S. EPA, 40 CFR 60: Standards of Performance for New Stationary Sources, Washington, DC: U.S. Environmental Protection Agency, 2001.
7.7 Spot Checking Spot checks may be made by the firm owning the industrial combustion equipment or by inspectors checking compliance with air permits. Portable analyzers are usually used to make spot check measurements of exhaust gas emissions. Figure 7.56 shows an example of a portable kit used for making field measurements. The kit includes the analyzer, the sample probe, a sample conditioner, and the appropriate tubing and connectors for the system. This can be easily transported to a site and quickly assembled for portable measurements. Figure 7.57 shows an example of a similar portable measurement analysis system. A more sophisticated “portable” analyzer will generally have higher accuracy and repeatability compared to other smaller analyzer systems. However, it is not nearly as portable and is less convenient, for example, for carrying on airplanes to remote sites. Figure 7.56 Portable gas analyzer (Testo 330). (Courtesy of ETA, Garner, NC.)
7.8 Example Measurements 7.8.1 Oxygen-Enhanced Combustion 7.8.1.1 Experimental Setup Experiments using different probes were conducted in Air Products’ combustion laboratory furnace shown in Figure 7.58 [46]. The furnace design has been described
© 2011 by Taylor and Francis Group, LLC
elsewhere [47]. Briefly, the furnace can accommodate one burner firing up to 10 × 106 Btu/hr. The furnace maximum operating temperature is 1800 K (2800°F) and can be somewhat controlled by means of steam generation in sidewall water panels. There is a damper at the furnace exit allowing furnace pressure control. For the probe emission tests, the furnace was kept at a positive
171
Pollution Emissions
Oxygen
Fuel
Fuel
Front view
Side view
Figure 7.59 K-Tech burner. (Slavejkov, A. G., and Baukal, C. E., Transport Phenomena in Combustion, Taylor & Francis, Boca Raton, Florida, 1996.)
Inconel tube
Wall = 18 in Sampling port position varied between –14 and 12 inches with respect to hot face of the wall
Figure 7.57 Portable gas analyzer. (Courtesy of ECOM, Gainsville, GA.)
Quarts tube
Furnace
Refractory plug Enerac 2000 Figure 7.60 Sampling setup with Enerac 2000 analyzer. (Slavejkov, A. G., and Baukal, C. E., Transport Phenomena in Combustion, Taylor & Francis, Boca Raton, FL, 1996.)
Figure 7.58 Air Products’ combustion laboratory furnace. (Slavejkov, A. G., and Baukal, C. E., Transport Phenomena in Combustion, Taylor & Francis, Boca Raton, FL, 1996.)
pressure to minimize air infiltration. The average temperature was about 1380 K (2030ºF). A K-Tech burner, shown in Figure 7.59, was fired at about 0.64 MW (2.2 × 106) Btu/hr. During the experiments the oxygen/natural gas volumetric ratio was varied between 1.8 (fuel rich) and 2.2 (fuel lean). A permanent CEMS and a portable handheld unit Enerac 2000 were used in these experiments. The CEMS is a permanently installed sampling system for monitoring the exhaust gas composition at the exit of the furnace. It consists of a water-cooled stainless steel probe, heattraced transport lines, water removal system, and gas analyzers. NO is measured with a chemiluminescent and
© 2011 by Taylor and Francis Group, LLC
nondispersive infrared (NDIR) analyzers. O2 is measured with a paramagnetic analyzer. CO is measured with an NDIR analyzer and a gas chromatograph. The Enerac 2000 analyzer is a portable unit for field measurements of combustion gases. It is capable of measuring concentrations of NO, CO, O2, SO2, and combustibles in gas stream temperatures up to 1400 K (2000ºF). Electrochemical cells are used to determine the gas concentrations. This type of instrument may not be accepted for official regulatory monitoring. However, it may be used in observing trends and establishing relationships between different combustion variables. The original probe for the Enerac 2000 was a 9.5 mm (3/8 in.) o.d., 1 m (3 ft) long tube made of Inconel. This tube had a thermocouple mounted at the end that enables monitoring of the sampling gas temperature as well as temperature of that end of the Inconel tube. In addition to this tube, a quartz tube of similar dimensions was used for the experiments. The Enerac 2000 analyzer probes were inserted in the furnace through a port on the 46 cm (18 in.) thick wall (below and to the right of the K-Tech burner). The experiment was designed to allow quick changes between the tubes. The insertion length of the tubes was varied during the experiments allowing rough probe temperature control, as shown in Figure 7.60.
172
Industrial Combustion Testing
In parallel with the Enerac 2000 measurements, the flue gas was monitored using the standard CEMS technique. This allowed comparison between results obtained with different instruments. 7.8.1.2 Results The results of the measurements are given in Table 7.4 through Table 7.6. Note that the probe position indicates the distance of the probe tip from the hot face of the furnace wall. Thus, the positions greater than 0 show that the tip extends inside the furnace. The measurements
shown in the tables were made in the sequence from left to right. For the fuel lean experiments, it may be seen from Table 7.4 that the Enerac NO and O2 readings were by about 30% higher than the CEMS readings. There was practically no variation between the results obtained using different tubes. The Enerac CO measurements showed zero except for the first reading, which was well below the 575 ppm CO measured with the CEMS. For the fuel rich experiments shown in Table 7.5, the NO readings with the Inconel probe were much lower than the other readings. Also, there is a trend between
Table 7.4 Measurements at Oxygen: Natural Gas Ratio of 2.2 (Fuel Lean)* Analysis System Probe material Position (inches) O2 (%) NO (ppmvd) CO (ppmvd) Probe temp (ºF)
Enerac 2000 Inconel 0 5.4 425 200 > 2000
Inconel –8 4.4 430 0 1972
Inconel 0 4.5 445 0 > 2000
Inconel –4 5.0 445 0 > 2000
CEMS Quartz 6 6.4 421 0 > 2000
Quartz 12 7.5 410 0 > 2000
Stainless in the stack 3.8 300 575 water-cooled
*Tfurnace average = 2030ºF. Firing rate = 2.2 × 106 Btu/hr. Source: From Slavejkov, A. G., and Baukal, C. E., Transport Phenomena in Combustion, Taylor & Francis, Boca Raton, FL, 1996.
Table 7.5 Measurements at Oxygen: Natural Gas Ratio of 1.8 (Fuel Rich)* Analysis System Probe material Position (inches) O2 (%) NO (ppmvd) CO (ppmvd) Probe temp (ºF)
Enerac 2000 Inconel –4 0 11 over > 2000
Inconel 12 0 7 over > 2000
Inconel –10 0 30 over 1515
Inconel –14 0 107 over 1310
CEMS Quartz 12 0 214 over > 2000
Quartz 12 0 170 over > 2000
Stainless in the stack 1.5 210 42,000 water cooled
*Tfurnace average = 2040ºF. Firing Rate = 2.2 × 106 Btu/hr. Source: From Slavejkov, A. G., and Baukal, C. E., Transport Phenomena in Combustion, Taylor & Francis, Boca Raton, FL, 1996.
Table 7.6 Transitional Measurements* Analysis System
Enerac
Enerac
CEMS
Enerac
Enerac
CEMS
Probe Material Position (inches) O2 (%) NO (ppmvd) CO (ppmvd) Probe temp (ºF)
Inconel 12 0.5 240 0 > 2000
Quartz 12 1.2 327 0 > 2000
Stainless in the stack
Inconel 12 0.4 233 0 > 2000
Quartz 12 1.1 334 0 > 2000
Stainless in the stack 3.5
2800 > 2000
2000 water cooled
*Tfurnace average = 2040ºF. Firing Rate = 2.2 × 106 Btu/h. Transition from oxygen: natural gas ratio of 1.8 (fuel rich) to 2.2 (fuel lean). Source: From Slavejkov, A. G., and Baukal, C. E., Transport Phenomena in Combustion, Taylor & Francis, Boca Raton, FL, 1996.
© 2011 by Taylor and Francis Group, LLC
173
Pollution Emissions
7.8.1.3 Discussion The discrepancy between the Enerac 2000 readings with the Inconel and quartz tubes, as well as its temperature dependence, clearly indicates that the probe material plays a key role. From the above data, it may be concluded that Inconel is a good catalyst of the following reactions at temperatures above 980 K (1300ºF).
CO + NO → CO2 + N
(7.14)
CO + O2 → CO2 + O
(7.15)
The quartz tube has also catalyzed these reactions but with lower efficiency. For the fuel lean case in Table 7.4, there was much more O2 and NO than CO. In other words the excess quantities of oxidizers in reactions 7.14 and 7.15 resulted in complete consumption of CO. The difference in the NO and O2 readings between the Enerac 2000 and the CEMS was probably due to the calibration, instrument accuracy or the difference in probe location in the furnace. Here, the NO readings with the Enerac 2000 were consistently higher than those obtained with the CEMS. The fuel rich case shown in Table 7.5, indicates that the 4.2% CO furnace concentration was able to completely eliminate NO when sampled through the Inconel tube. However, the NO started increasing after the Inconel tube temperature was lowered by pulling it out of the furnace. This temperature dependence is shown in Figure 7.61. The higher NO readings at lower temperatures, especially below 1100 K (1500ºF), indicate a decrease in the
© 2011 by Taylor and Francis Group, LLC
120 100 NO readings, ppmvd
the Inconel probe temperature and the NO readings. That is, at Tprobe < 1400 K (2000ºF) the NO readings were significantly greater than at Tprobe > 1400 K (2000ºF). For the quartz probe, however, the NO readings were much higher even though the probe was inside the furnace for both measurements. The Enerac O2 readings were all at zero and the CO was off the scale ( > 5000 ppm). At the same time, the CEMS showed 1.5% O2, about the same NO as the quartz tube and 4.2% CO. Following these two sets of measurements, the oxygen/natural gas ratio was set to 2.2 (fuel lean). The data shown in Table 7.6 were taken during the transitional period when the furnace CO was being reduced from the initial very high reading of 42,000 ppm. The readings shown in the table were again taken in the sequence from left to right within five minutes from adjusting the flow controls. It may be seen from Table 7.6 that a reproducible difference in the NO and O2 readings was detected for the Inconel and quartz tubes. The Inconel tube always showed lower NO and O2 readings. The NO readings with the CEMS were not made for these transitional measurements because of the time limitation.
80 60 40 20 0 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 Temperature, deg F
Figure 7.61 Temperature dependence of NO readings for Enerac 2000 inconel probe at reducing furnace conditions.
Inconel catalytic efficiency for the reactions 7.14 and 7.15. However, even at about 980 K (1300ºF), the Enerac NO readings were still much lower than those obtained with the CEMS. The quartz tube showed about equal readings of NO as the CEMS. However, in view of the first data set that showed NO consistently higher, it appears that the reactions 7.14 and 7.15 were weakly catalyzed in the quartz tube too, which lowered the NO readings. The last transitional data set with fuel lean conditions, in Table 7.6, showed a reproducible indication of this catalytic activity. The 2000–2800 ppm of CO was not enough to consume all of O2 and NO. This time, the oxidizers were in excess. 7.8.1.4 Conclusions and Recommendations The above example has demonstrated some of the problems one can expect in using a hot probe for sampling gases in high temperature furnaces away from the flame. For reducing conditions inside a furnace, the measurements may show much lower NO readings. Similarly, if the furnace is operating under oxidizing conditions the measurements will show much lower CO readings. The experiments have shown that both metal and quartz uncooled probes can affect the readings provided the probe surface temperature exceeds 980 K (1300ºF). To avoid the surface catalytic effects, a water-cooled probe must be used when sampling high temperature combustion products. This is of particular importance to oxyfuel combustion where measured species concentrations are much higher due to the elimination of nitrogen. Recommended probe materials and cooling requirements to avoid reaction of the various gases inside the probe are given elsewhere [48]. 7.8.2 Fuel Composition Effects The composition of the fuel supplied to a combustion system has a significant impact on the NOx emissions. In the petrochemical and chemical process industries,
174
Industrial Combustion Testing
there is a very wide range of fuel blends used for process heating. These fuels are often by-products from a refining process. They typically contain hydrocarbons ranging from C1 to C4, hydrogen, and inert gases like N2 and CO2. In a given plant or refinery, burners used in process heaters may need to be capable of firing on multiple fuels that are present at different times (e.g., start-up, normal operation, upset conditions, etc.). In many cases, the NOx emissions from the heaters may be limited by regulations regardless of what fuel composition is being fired. Therefore, it is critical that the effects of the fuel composition on NOx emissions be understood and quantified to ensure that permitted values are not exceeded. This study [49] investigated the effects of fuel composition on NOx emissions from an industrial-scale burner. The acquired data provided additional insight into the effects on NOx over the range of fuel compositions consisting of various fractions of three primary components: H2, C3H8, and CH4. The results quantify 3-way fuel component effects on NOx emissions and provide insight into fundamental parameters involved and how they interact, which provides information on which to base future research, design, and development work.
layers of refractory lining to achieve the desired furnace temperature. A velocity thermocouple (also known as a suction thermocouple or suction pyrometer; see Chapter 5) was used to measure the furnace and stack gas temperatures. The furnace draft was measured with an automatic, temperature-compensated, pressure transducer as well as an inclined manometer connected to a pressure tap in the furnace floor. Fuel flow rates were measured using
7.8.2.1 Test Setup To demonstrate the strong dependence of NOx on fuel gas composition, experiments were conducted using an industrial burner fired into a pilot-scale furnace. Tests were conducted using a conventional-type burner (see Figure 7.62) with a single fuel gas tip and flameholder. The burner was fired vertically upward in a rectangular furnace (see Figure 7.63). The test furnace was a rectangular heater with internal dimensions of 2.4 m (8 ft) wide, 3.7 m (12 ft) long, and 4.6 m (15 ft) tall. The furnace was cooled by a water jacket on all four walls. The interior of the water-cooled walls was covered with varying (a)
Figure 7.63 Test Furnace 1 at the John Zink Co., R&D test center. (From Baukal, C. E., The John Zink Combustion Handbook. Boca Raton, FL, CRC Press, 2001.)
(b)
Figure 7.62 Raw gas burner: (a) side view, (b) top view. (From Baukal, C. E., The John Zink Combustion Handbook. Boca Raton, FL, CRC Press, 2001.)
© 2011 by Taylor and Francis Group, LLC
175
Pollution Emissions
calibrated orifice meters (see Figure 7.41), fully corrected for temperature and pressure. Emission levels were measured using state-of-the-art continuous emissions monitors (CEMs) to measure emissions species concentrations of NOx, CO, and O2. The experimental matrix consisted of firing the burner at a constant heat release (2.2 MW or 7.5 × 106 Btu/hr) and excess air level (15%) with 15 different fuel blends comprised of varying amounts of H2, C3H8, and Tulsa Natural Gas (TNG).* For testing and analysis purposes TNG was treated as a single fuel component for convenience. The TNG, which is comprised of approximately 93% CH4, is a more economical choice than pure CH4 for experimental work and the analysis is simplified by treating it as a single component. All 15 fuel compositions were tested on each of six different fuel gas tips, which differed in port diameter sizes, to enable the acquisition of additional information regarding effects resulting from differing fuel pressures. 7.8.2.2 Sample Results Figure 7.64 shows the variation in relative measured NOx emissions resulting from different concentrations (volume basis) of H2 in a fuel blend composed with a balance of TNG for each of the six different fuel gas tips
100%
90% 80% 70% 60% 50% 40%
0%
20%
40% 60% H2 in TNG (vol. %)
80%
100%
90% 80% 70% 60% 50% 40%
Tip 1
Tip 2
Tip 3
Tip 4
Tip 5
Tip 6
Figure 7.64 Measured NOx (percentage of the maximum ppmvd value) as a function of the fuel blend composition for H2/TNG blends combusted with 15% excess air where both the fuel and the air were at ambient temperature and pressure. (From Baukal, C. E., The John Zink Combustion Handbook. Boca Raton, FL, CRC Press, 2001.) * The nominal composition by volume of TNG is: 93.4% CH4, 2.7% C2H6, 0.60% C3H8, 0.20% C4H10, 0.70% CO2, and 2.4% N2.
© 2011 by Taylor and Francis Group, LLC
Relative NOx (fraction of max., ppmv basis)
Relative NOx (% of max., ppmv basis)
100%
tested. The plot, which illustrates NOx levels on a concentration basis, clearly shows the correlation between increased H2 content and higher NOx emission levels. The slope of the profile is exponentially increasing, qualitatively similar to that predicted by the plotted theoretical calculations shown in reference [49]. The effect of H2 is significant, with the sharpest increase in NOx levels taking place as concentration levels of H2 in the fuel mixture rise from 75% to 100%. The variation in relative measured NOx emissions resulting from different concentrations (volume basis) of C3H8 in a fuel blend composed with a balance of TNG is shown in Figure 7.65. The slope of the increase in NOx levels corresponding to increased concentrations of C3H8 is shown to be relatively constant or slightly declining over the gradient in C3H8 concentration, in contrast with the exponentially increasing profile of the H2-TNG plot-in. The profile showing the effect of C3H8 content is also seen to be similar to the corresponding calculated trends shown in Hayes and colleagues research [49]. Figure 7.66 shows the final two-component fuel blend results being examined, which describe the variation in relative measured NOx emissions resulting from different concentrations (volume basis) of H2 in a fuel blend composed with a balance of C3H8. The upper plot, which shows measured relative NOx on a volume concentration basis, illustrates that for a given tip, geometry, and port size, the measured NOx concentrations actually decrease
0%
20%
40%
60%
80%
100%
C3H8 in TNG (vol. %) Tip 1
Tip 2
Tip 3
Tip 4
Tip 5
Tip 6
Figure 7.65 Measured NOx (percentage of the maximum ppmvd value) as a function of the fuel blend composition for C3H8/TNG blends combusted with 15% excess air where both the fuel and the air were at ambient temperature and pressure. (From Baukal, C. E., The John Zink Combustion Handbook. Boca Raton, FL, CRC Press, 2001.)
176
Industrial Combustion Testing
Relative NOx (% of max., ppmv basis)
100% 95% 90% 85% 80% 75% 70% 65%
Relative NOx (% of max., lb/106 Btu basis)
60%
0%
20%
0%
20%
40%
60%
80%
100%
40%
60%
80%
100%
100% 95% 90% 85% 80% 75% 70% 65% 60%
H2 in C3H8 (vol %) Tip 1
Tip 2
Tip 3
Tip 4
Tip 5
Tip 6
Figure 7.66 Measured NOx (percentage of the maximum value in both ppmv and lb/106 Btu) as a function of the fuel blend composition for H2/C3H8 blends combusted with 15% excess air where both the fuel and the air were at ambient temperature and pressure.
slightly with increasing H2 content up to 75% H2 content, then sharply increase with H2 concentration. Due to the decrease in total dry products of combustion from the burning of H2 expressing NOx in terms of concentration (ppmvd) does not fully represent the actual mass rate of NOx emissions produced. The lower plot, which shows the variation in measured NOx levels on a mass per unit heat release basis, illustrates that the overall emissions of NOx on a mass basis decrease with increasing fuel hydrogen content and continue to decrease or remain relatively flat even in the high-hydrogen content region that produced a sharp increase in NOx levels on a volume concentration basis.
lower fuel pressures. On a mass per heat release basis however, the highest relative NOx levels were achieved for fuel compositions containing large fractions of C3H8. This appears to result from some combined characteristics of a high-propane mixture including: very low fuel pressure for a given heat release in comparison with the other fuels; somewhat higher adiabatic flame temperature than CH4; and a substantially larger amount of total dry products of combustion produced for a given heat release when compared with H2. In summary, the results of this work provide both quantitative and qualitative information to improve emission performance prediction and design of burners with application to a wide variation of fuel compositions.
7.8.2.3 Conclusions This study showed experimentally that NOx is highly dependent on the fuel blend composition. Adiabatic flame temperature and fuel pressure are both identified as significant fundamental parameters affecting NOx emission levels when considering the effect of fuel composition on NOx levels. For a conventional burner, with NOx on a concentration basis, the adiabatic flame temperature is dominant, with fuel pressure remaining significant in affecting NOx emission levels. The highest NOx levels on a volume concentration basis occurred at the highest hydrogen content fuel compositions at
© 2011 by Taylor and Francis Group, LLC
7.8.3 Ghost NOx Field NOx measurements have been made in a number of process heaters, at locations before and after the convection sections. In nearly all cases, NOx readings after the convection section have been higher than before the convection section. In some cases, the increase in NOx has been as much as 50%. This phenomenon has been lightheartedly referred to as “ghost NOx” because of its unknown origin. It indicates that either additional NOx is being formed in the convection section, that there is some systematic measurement problem, or some
177
Pollution Emissions
combination of the two. Extensive pilot scale measurements have also been made to study this phenomenon. These suggest that the nitrogen oxide (NOx = NO + NO2) readings can be significantly different depending on the sample location within the furnace. This section discusses the field and laboratory measurements, discusses possible sources of the ghost NOx, and recommends further work to study this unusual phenomenon. 7.8.3.1 Introduction Environmental regulations continue to reduce the allowable NOx emissions from process heaters [17]. This has led to the development of ultra low NOx burners [50,51]. One important issue has become the measurement of emissions as the levels get lower as slight errors are significant, because they are now a much higher proportion of the overall emission level. This section considers the potential error associated with the overall NOx (NO + NO2) emissions from process heaters. Field data suggests that the nitrogen oxide (NOx = NO + NO2) readings can be significantly different depending on the sample location within the furnace. For example, field data shows that the NOx emissions in the furnace stack can be as much as 50% higher than for a sample extracted before the convection section. This anomaly is referred to here as ghost NOx. The purpose of this section is to discuss possible reasons why this apparent anomaly exists and to present results of data collected in the field at an ethylene cracking refinery, and in a pilot-scale test furnace that attempt to simulate this occurrence. Although not the subject of this section, this phenomenon has also been observed in industrial boilers during the development of ultra low NO burner technology. There are many possible hypotheses for why the ghost NOx anomaly might occur.
1. NOx meter does not measure NO2, only NO 2. NO2 readily absorbs into water during sampling 3. NO2 catalytically decomposes in the convection section 4. Thermal NOx formation increases due to longer residence time 5. NOx emissions are not uniformly distributed in the radiant section 6. NO2 to NO reduction converter in analyzer is not efficient 7. NOx formation chemistry produces additional NOx in the convection section
One or a combination of these could be a culprit in the NOx anomaly. Each of these hypotheses is discussed elsewhere [52].
© 2011 by Taylor and Francis Group, LLC
7.8.3.2 Test Description Lab tests were performed in a pilot-scale test furnace with the following arrangement: • Two (2) floor burners and two (2) radiant wall burners • Total heat release ~ 4.30 MW (14.7 × 106 Btu/hr) • Percentage O2 in the exhaust products, variable • Fuel composition, Tulsa natural gas (TNG) • Inside furnace dimensions, H 14 m × L 2.7 m × W 1.8 m (45′ × 9′ × 6′) • Convection tubes, carbon steel Figure 7.67 is a schematic showing the test arrangement. The setup allowed for a sample probe to be inserted at three locations just before the convection section or at three locations in the middle of the convection section as shown in the figure. The probe depth could also be adjusted to various depths within the convection section. The flue gas sample was pulled through a heat traced Teflon tube to a remotely located chemiluminescent NOx analyzer. Flue gas samples were also collected from a sample port located approximately 3 m (10 ft) below the top of the furnace and stack exit plane. Those samples were also pulled through a heat traced Teflon tube to the same NOx analyzer. Tests were performed using state-of-the-art continuous emissions monitors (CEMs) to measure emissions species concentrations of NOx, CO, and O2. Stack samples were continuously extracted at high flow rates to minimize response time. A continuous sample was fed to a condenser/conditioner, which cools the sample to 1°C dew point. Once the moisture is removed, the dry sample flows to the CEM for species analysis. The reduction converter efficiency was determined using NO2 span gas with 16.5 ppm. With this span gas, the analyzer read a value of 14.6 ppm, for a conversion efficiency of 88%. This compared very favorably with the 87% conversion efficiency reported in Bussman, Baukal, and Waibel [52] for MnO2. Field tests were performed in a full-scale ethylene cracking furnace comprised of multiple floor and wall burners operating at a total heat release of approximately 68.9 MW (235 × 106 Btu/hr). The setup allowed for a sample probe to be inserted at various locations through site ports within the furnace radiant section and crossover section located just before the convection section. The probe length was approximately 3 m (10 ft) long and could be adjusted to various depths within the furnace. The flue gas sample was pulled through a heat traced Teflon tube to a remotely located CEM. Tests were performed using state-of-the-art CEMS. Stack samples were continuously extracted at high flow
178
Industrial Combustion Testing
29 in
Stack
17_5 in Top view of convection section showing sample probe locations measurement depths at d = 6, 12, 18 and 24 inches
29_25 in 40 in
3
6
40 in
2
5
29_25 in
1
4
Sample 15 ft probe 5 ft
27 in 6 ft L × 9 ft W
d Convection section with carbon steel tubes
45 ft H
Radiant wall burners
Floor burners Figure 7.67 Schematic showing test furnace and sample probe locations. (From Bussman, W., Baukal, C., Waibel, R. “Ghost NOx.” Presented at the Air & Waste Management Association’s 98th Annual Conference & Exhibition, Control #28, Minneapolis, MN, 2005.)
rates to measure emissions species concentrations of NOx, CO, and O2. A continuous sample was fed to a condenser/ conditioner, which cools the sample to 1°C dew point. Once the moisture was removed, the dry sample flows to the CEM for species analysis. The reduction converter efficiency of 96% was determined using NO2 span gas. 7.8.3.3 Results Lab Tests. Figures 7.68 and 7.69 are plots of the NO at locations before and in the middle of the convection section at various depths. This data was collected at a total furnace heat release of 4.3 MW (14.7 × 106 Btu/hr) burning 100% Tulsa natural gas ( > 90% CH4) at 3% O2 in the dry exhaust products. The data clearly show that the
© 2011 by Taylor and Francis Group, LLC
NOx concentration varies by a maximum of 15% over a distance of 46 cm (18 in.) through the convection section, demonstrating that the NOx was highly stratified within the convection section. The stratification of NOx might be due to recirculating flow patterns inside the furnace. It is speculated that the flow pattern of furnace flue gases might look similar to that as illustrated in Figure 7.70. As the hot combustion products turn 90º into the convection section, they create a recirculation zone in the upper corner of the furnace. Also in Figure 7.70 is a streak photograph showing small particles flowing through two 90º elbows [53]. Notice in this photograph that there is a small recirculation zone in the upper corner of the first 90º elbow. This recirculation zone, containing relatively cool gas, might
179
Pollution Emissions
In convection section 58
Ave. NO (ppmdv)
56
46 48 50 52 54 56 58
54 52 50
(in
on
iti
po s
8
Figure 7.68 (See color insert following page 424.) NO emissions at locations before and in the middle of the convection section at various depths. (From Bussman, W., Baukal, C., Waibel, R. “Ghost NOx.” Presented at the Air & Waste Management Association’s 98th Annual Conference & Exhibition, Control #28, Minneapolis, MN, 2005.)
Hot combustion products
Re-circulation zone
58 NO + NO2 emissions (ppmvd)
Hot combustion products results in high NOx concentration levels in this region
est
22 20 18 16 14 Depth 12 10 (in.)
W
46
100 90 80 70 60 50 40 30 6 Before convection section
.)
48
Re-circulation zone
57
Average in stack
56 55
After convection section
54
Figure 7.70 Illustration showing theorized flow pattern within test furnace. (From Bussman, W., Baukal, C., Waibel, R. “Ghost NOx.” Presented at the Air & Waste Management Association’s 98th Annual Conference & Exhibition, Control #28, Minneapolis, MN, 2005.) Also shown is a streak photograph of particles flowing through two 90º turns.
53 52 51 50
Middle of convection section
0
5
10 15 20 Probe depth (inches)
25
30
Figure 7.69 Top surface is in middle of convection section, bottom surface before convection section. (From Bussman, W., Baukal, C., Waibel, R. “Ghost NOx.” Presented at the Air & Waste Management Association’s 98th Annual Conference & Exhibition, Control #28, Minneapolis, MN, 2005.)
extend into the convection section of the furnace and create the NOx stratification observed. Also notice in Figure 7.70 that the NO appears to be slightly lower at a location before the convection section than at a location in the middle of the convection section for all depths tested. This suggests that the carbon steel convection tubes might be acting as a catalyst and
© 2011 by Taylor and Francis Group, LLC
converting the NO2 to NO, one of the possible culprits for the ghost NOx anomaly. Figures 7.71 and 7.72 are plots of NOx (NO + NO2) at locations before and in the middle of the convection section at various depths. The data clearly show that NOx varied by a maximum of 11% over a distance of 46 cm (18 in.). Notice that the NOx appeared to be lower at a location before the convection section than at a location in the middle of the convection section, for all depths tested. It is unclear why this occurs because the total NOx (NO + NO2) should be conserved between these two locations. Perhaps the recirculation pattern in the upper corner of the firebox creates this anomaly. At a probe depth of 61 cm (24 in.), the average NOx in the stack appears to be in the center of the data scatter for NOx collected at various locations in the convection section.
180
Industrial Combustion Testing
58 Average in stack
56 54
Middle of convection section
52 50 48 46
0
5
10 15 20 Probe depth (inches)
25
30
Figure 7.71 NOx (NO + NO2) at locations before and in the middle of the convection section at various depths. (From Bussman, W., Baukal, C., Waibel, R. “Ghost NOx.” Presented at the Air & Waste Management Association’s 98th Annual Conference & Exhibition, Control #28, Minneapolis, MN, 2005.) In convection section ) (ppmdv) Ave. NOx (NO + NO2
58
51 52 53 54 55 56 57 58
57 56 55 54 53
(in
ion
sit
po
8
est
22 20 18 16 14 Depth 12 10 (in.)
W
51
100 90 80 70 60 50 40 30 6 Before convection section
.)
52
Figure 7.72 (See color insert following page 424.) Top surface is in middle of convection section, bottom surface before convection section. (Bussman, W., Baukal, C., Waibel, R. “Ghost NOx.” Presented at the Air & Waste Management Association’s 98th Annual Conference & Exhibition, Control #28, Minneapolis, MN, June, 2005.)
The convection tube skin temperature in the test furnace might not have been high enough to provide substantial conversion of NO2 to NO. In the first few rows of the convection section of a typical ethylene cracking furnace, the tube skin temperature can reach values above 1300 K (1800°F). Field Tests. Figure 7.73 is a plot of the NOx (NO + NO2), corrected to 3% O2, at various locations in the upper 3 m (10 ft) of the radiant section. This data was collected at a total furnace heat release of 68.9 MW (235 × 106 Btu/hr)
© 2011 by Taylor and Francis Group, LLC
30 25 20
Stack (NO2 + NO) – Average = 29.4 ppm
Probe (NO2 + NO) – Average = 23.3 ppm
15 10 5 0
Before convection section
1
2
3 Sample number
4
5
Figure 7.73 NOx (NO + NO2) at various locations in the upper 3 m (10 ft) of the radiant section of the field test furnace. (From Bussman, W., Baukal, C., Waibel, R. “Ghost NOx.” Presented at the Air & Waste Management Association’s 98th Annual Conference & Exhibition, Control #28, Minneapolis, MN, 2005.) 30 Stack NO – Average = 27.4 ppm Emissions corrected to 3% O2
NO emissions (ppmvd)
NO2 + NO corrected to 3% O2
35
25 Probe NO – Average = 18.6 ppm
20 15 10
Probe NO2 – Average = 4.8 ppm
5 0
Stack NO2 – Average = 1.9 ppm
1
2
3 Sample number
4
5
Figure 7.74 NO and NO2 at various locations in the upper 3 m (10 ft) of the radiant section of the field test furnace. (From Bussman, W., Baukal, C., Waibel, R. “Ghost NOx.” Presented at the Air & Waste Management Association’s 98th Annual Conference & Exhibition, Control #28, Minneapolis, MN, 2005.)
burning a typical refinery fuel at approximately 5.7% O2 in the stack. The data show that the overall average NOx value is 26% lower in the upper zone of the radiant section than at the stack exit. The data also show that variations in probe readings within the radiant section vary significantly (approximately 40%), but are fairly constant at the stack exit. Figure 7.74 is a plot of the NO and NO2 at various locations in the upper 3 m (10 ft) of the radiant section; the data corresponds to the plotted values in Figure 7.73. The data show that the overall average NO2 value is 152% lower in the stack than in the upper zone of the radiant section. However, the NO is 147% higher in the stack than in the upper zone of the radiant section. This trend appears to suggest NO2 is converted to NO as the flue gas passes through the convection section.
Pollution Emissions
181
7.8.3.4 Conclusions
15. Friedrich, R., and Reis, S., eds. Emissions of Air Pollutants. New York: Springer, 2004. 16. Crawford, M. Air Pollution Control Theory. New York: McGraw-Hill, 1976. 17. Baukal, C. E. Industrial Combustion Pollution and Control. New York: Taylor & Francis, 2004. 18. United Nations, ed. Concepts and Methods of Environment Statistics: Statistics of the Natural Environment. New York: United Nations Publications, 1992. 19. Newman, O., and Foster, A., eds. Environmental Statistics Handbook: Europe. Farmington Hills, MI: Gale Group, 1993. 20. Patil, G. P., and Rao, C. R., eds. Multivariate Environmental Statistics. New York: Elsevier Science, 1994. 21. Rao, C. R., and Patil, G. P., eds. Environmental Statistics. New York: Elsevier Science, 1994. 22. Ott, W. R. Environmental Statistics and Data Analysis. Boca Raton, FL: CRC Press, 1995. 23. Hoshmand, A. R. Statistical Methods for Environmental and Agricultural Sciences. 2nd ed. Boca Raton, FL: CRC Press, 1997. 24. Kottegoda, N. T., and Russo, R. Statistics, Probability and Reliability Methods for Civil and Environmental Engineers. New York: McGraw-Hill, 1997. 25. Piegorsch, W. W., Cox, L. H., and Nychka, D., eds. Case Studies in Environmental Statistics. New York: SpringerVerlag, 1998. 26. Novartis Foundation. Environmental Statistics: Analysing (sic) Data for Environmental Policy. New York: John Wiley & Sons, 1999. 27. Manly, B. F. Statistics for Environmental Science and Management. Boca Raton, FL: CRC Press, 2000. 28. Berthouex, P. M., and Brown, L. C. Statistics for Environmental Engineers. 2nd ed. Boca Raton, FL: Lewis Publishers, 2002. 29. Townsend, J. Practical Statistics for Environmental and Biological Scientists. New York: John Wiley & Sons, 2002. 30. Lodge, J. P., ed. Methods of Air Sampling and Analysis. 3rd ed. Chelsea, MI: Lewis Publishers, 1977. 31. Mussatti, D. C., ed. EPA Air Pollution Control Cost Manual. 6th ed. Report EPA/452/B-02-001. Washington, DC: U.S. Environmental Protection Agency, January 2002. 32. Mussatti, D. C., Groeber, M., Maloney, D., Koucky, W., and Hemmer, P. M. “Section 2: Generic Equipment and Devices, Chapter 4: Monitors.” In Air Pollution Control Cost Manual, edited by D. C. Mussatti. 6th ed. .Report EPA/452/B-02-001. Washington, DC: U.S. Environmental Protection Agency, January 2002. 33. American National Standards Institute. ISO 10780: 1994, Stationary source emissions: Measurement of velocity and volume flowrate of gas streams in ducts. Washington, DC: American National Standards Institute, 2007. 34. Mullinger, P., and Jenkins, B. Industrial and Process Furnaces. Oxford, UK: Butterworth-Heinemann, 2008. 35. Office of the Federal Register. 60.2 Definitions. U.S. Code of Federal Regulations Title 40 Part 60. Washington, DC: U.S. Government Printing Office, 2001. 36. Bussman, W., and Baukal, C. “Ambient Condition Effects on Process Heater Emissions.” Proceedings of the Int’l Mechanical Engineering Congress & Exhibition, Paper IMECE2008-68284, Boston, MA, November 2008.
Due to the NOx stratification observed in the convection section of the lab furnace, one cannot conclude with certainty that the ghost NOx anomaly was observed. The field tests, however, appear to indicate the ghost NOx anomaly. This is at least in part likely due to catalytic decomposition of NO2 through the convection section. Other causes are also possible. It is recommended that additional tests be performed. One test might include testing in another furnace configuration. Another could be with different burner types. Measurements of other species (e.g., N2O) may also produce useful information that can be used to isolate the source of the additional NOx.
References
1. Yeh, J. T. “Modeling Atmospheric Dispersion of Pollutants.” In). Air Pollution Control and Design for Industry, edited by P. N. Cheremisinoff. New York: Marcel Dekker, 1993. 2. U.S. EPA. 40 CFR 60: Standards of Performance for New Stationary Sources. Washington, DC: U.S. Environmental Protection Agency, 2001. 3. U.S EPA. Title 40-Protection of Environment, Part 60-Standards of Performance for New Stationary Sources. http:// ecfr.gpoaccess.gov/cgi/t/text/text-idx?c=ecfr&tpl=/ ecfrbrowse/Title40/cfr60_main_02.tpl/ (accessed 12 February 2010). 4. Wheeler, A. J., and Ganji, A. R. Introduction to Engineering Experimentation. 3rd ed. New York: Prentice Hall, 2010. 5. Cobb, G. W. Introduction to Design and Analysis of Experiments. New York: Springer-Verlag, 1998. 6. Holman, J. P. Experimental Methods for Engineers. 7th ed. New York: McGraw-Hill, 2000. 7. Montgomery, D. C. Design and Analysis of Experiments. 5th ed. New York: John Wiley & Sons, 2000. 8. Diamond, W. J. Practical Experiment Designs: For Engineers and Scientists. 3rd ed. New York: John Wiley & Sons, 2001. 9. Finlayson-Pitts, B. J., and Pitts, J. N. Atmospheric Chemistry: Fundamentals and Experimental Techniques. New York: John Wiley & Sons, 1986. 10. Clement, R., and Kagel, R., eds. Emissions from Combustion Processes: Origin, Measurement, Control. Boca Raton, FL: CRC Press, 1990. 11. Cornforth, J. R. ed. “Instrumentation and Measurement.” In Combustion Engineering and Gas Utilization. 3rd ed. London: E&FN Spon, 1992. 12. Baumbach, G. Air Quality Control. Berlin: Springer, 1996. 13. Liu, D. H. F., and Lipták, B. G., eds. Environmental Engineers’ Handbook. Boca Raton, FL: Lewis, 1997. 14. Schnelle, K. B., and Brown, C. A. Air Pollution Control Technology Handbook. Boca Raton, FL: CRC Press, 2002.
© 2011 by Taylor and Francis Group, LLC
182
37. American Petroleum Institute. Burners for Fired Heaters in General Refinery Services, API Recommended Practice 535. 2nd ed. Washington, DC: American Petroleum Institute, 2006. 38. Farnsworth, R., and Rowling, H. “The Reduction of Pollution from Industrial and Domestic Equipment: The Oil Industry’s Contribution.” In Pollution Prevention, edited by Peter Hepple. London, UK: The Institute of Petroleum, The Elsevier Publishing Co., 1968. 39. Chellappan, C., and Waibel, R. “Reducing NOx Emissions from Ethylene Furnaces: Practical Application of Lean Premix Combustion.” American Flame Research Committee Meeting, October 2004. 40. Becker, H. A., and Brown, A. P. G. “Response of Pitot Probes in Turbulent Streams. “Journal of Fluid Mechanics 62 (1974): 85–114. 41. Cho, H. S., and Becker, H. A. “Response of Static Pressure Probes in Turbulent Streams.” Experiments in Fluids 3 (1985): 93–102. 42. U.S. EPA. Compilation of Air Pollutant Emission Factors, Vol. I: Stationary Point and Area Sources. 5th ed. Washington, DC: U.S. Environmental Protection Agency report AP-42, 1995. 43. U.S. EPA. Appendix B: Performance Specifications. Code of Federal Regulations 40 Part 60. Washington, DC: U.S. Environmental Protection Agency, 2001. 44. U.S. EPA. Appendix F: Quality Assurance Procedures. Code of Federal Regulations 40 Part 60. Washington, DC: U.S. Environmental Protection Agency, 2001. 45. Parker, J. M. “Continuous Emission Monitoring (CEM).” In Handbook of Air Pollution Control and Engineering,
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
edited by J. C. Mycock, J. D. McKenna, L. Theodore. Boca Raton, FL: Lewis Publishers, 1995. 46. Slavejkov, A. G., and Baukal, C. E. “Flue Gas Sampling Challenges in Oxygen-Fuel Combustion Processes.” In Transport Phenomena in Combustion, edited by S. H. Chan, vol. 2, 1230–37. Boca Raton, FL: Taylor & Francis, 1996. 47. Baukal, C. E., and Dalton, A. I. “Nitric Oxide Measurements in Oxygen Enriched Air-Natural Gas Combustion Systems.” Proceedings of the Fossil Fuel Combustion Symposium, edited by S. N. Singh, New York: ASME, 1990. 48. ANSI/ASME. Flue and Exhaust Gas Analysis: Instruments and Apparatus. New York: American Society of Mechanical Engineers standard PTC 19.10, 1981. 49. Hayes, R., Baukal, C. E., Singh, P., and Wright, D. “Fuel Composition Effects on NOx.” Proceedings of the Air & Waste Management Association’s 94th Annual Conference and Exhibition, Orlando, FL, June 24–28, 2001. 50. Baukal, C. E., ed. The John Zink Combustion Handbook. Boca Raton, FL: CRC Press, 2001. 51. Baukal, C. E., ed. Industrial Burners Handbook. Boca Raton, FL: CRC Press, 2004. 52. Bussman, W., Baukal, C., Waibel, R. “Ghost NOx.” Presented at the Air & Waste Management Association’s 98th Annual Conference & Exhibition, Control #28, Minneapolis, MN, June 21–24, 2005. 53. Idelchik, I. E. Handbook of Hydraulic Resistance. New York: Hemisphere Publishing Corporation, 1986.
8 Combustion Noise Mahmoud M. Fleifil, Carl-Christian Hantschk, and Edwin Schorer Contents 8.1 Fundamentals of Sound................................................................................................................................................ 184 8.1.1 Introduction........................................................................................................................................................ 184 8.1.2 Basics of Sound................................................................................................................................................... 184 8.1.2.1 Sound Pressure Level and Frequency.............................................................................................. 185 8.1.2.2 Decibel.................................................................................................................................................. 186 8.1.2.3 Sound Power Level............................................................................................................................. 186 8.1.2.4 Threshold of Hearing......................................................................................................................... 187 8.1.2.5 Threshold of Pain................................................................................................................................ 188 8.1.2.6 Correction Scales................................................................................................................................. 188 8.1.3 Measurements.................................................................................................................................................... 189 8.1.3.1 Overall Sound Level and How to Add dB Values.......................................................................... 190 8.1.3.2 Atmospheric Attenuation.................................................................................................................. 193 8.2 Industrial Noise Pollution............................................................................................................................................ 193 8.2.1 OSHA Requirements......................................................................................................................................... 194 8.2.2 International Requirements............................................................................................................................. 194 8.2.3 Noise Sources and Environment Interaction................................................................................................. 194 8.3 Mechanisms of Industrial Combustion Equipment Noise...................................................................................... 195 8.3.1 Combustion Roar and Combustion Instability Noise.................................................................................. 195 8.3.1.1 Flare Combustion Roar...................................................................................................................... 195 8.3.1.2 Flare Combustion Instability Noise................................................................................................. 198 8.3.1.3 Burner Combustion Noise................................................................................................................. 199 8.3.1.4 Burner Combustion Instability Noise.............................................................................................. 199 8.3.2 Fan Noise............................................................................................................................................................ 200 8.3.3 Gas Jet Noise....................................................................................................................................................... 200 8.3.3.1 Gas Jet Mixing Noise.......................................................................................................................... 200 8.3.3.2 Shock-Associated Noise..................................................................................................................... 201 8.3.4 Valve and Piping Noise..................................................................................................................................... 202 8.4 Noise Abatement Techniques...................................................................................................................................... 202 8.4.1 Flare Noise Abatement Techniques................................................................................................................ 202 8.4.2 Burner Noise Abatement Techniques............................................................................................................. 204 8.4.3 Valve and Piping Noise Abatement Techniques........................................................................................... 205 8.4.4 Fan Noise Abatement Techniques................................................................................................................... 206 8.5 Analysis of Combustion Equipment Noise................................................................................................................ 206 8.5.1 Multiple Burner Interaction.............................................................................................................................. 206 8.5.2 High Pressure Flare........................................................................................................................................... 207 8.5.3 Atmospheric Attenuation Example................................................................................................................. 207 Glossary.................................................................................................................................................................................... 208 References................................................................................................................................................................................. 209 Bibliography............................................................................................................................................................................ 210
183 © 2011 by Taylor and Francis Group, LLC
184
Industrial Combustion Testing
8.1 Fundamentals of Sound 8.1.1 Introduction The sensation of sound is a thing sui generis, not comparable with any of our other sensations. No one can express the relation between a sound and a color or smell. Directly or indirectly, all questions connected with this subject must come for decision to the ear, as the organ of hearing; and from it there can be no appeal. But we are not therefore to infer that all acoustical investigations are conducted with the unassisted ear. When once we have discovered the physical phenomena which constitute the foundation of sound, our explorations are in great measure transferred to another field lying within the dominion of the principles of Mechanics. Important laws are in this way arrived at, to which the sensations of the ear cannot but conform. —The Theory of Sound by Lord Rayleigh III
Noise is commonly referred to as unwanted sound.* Noise is a common by-product of our mechanized civilization and is an insidious danger in industrial environments. Noise pollution is usually a local problem and so is not viewed on the same scale of importance as the more notorious industrial emissions like NOx, CO, and particulates. Nonetheless, it is an environmental pollutant of significant impact. Serious concern is merited when a pollutant can result in either environmental damage or human discomfort. Considering the impact on people, noise is most often a source of annoyance, but it can also have much more detrimental effects, such as causing actual physical injury. Noise-related injuries range from short-term discomfort to permanent hearing loss. According to recent statistics, more than 20 million Americans are exposed to hazardous sound levels on a regular basis. There are approximately 28 million Americans who have some degree of hearing loss: about one-third of these—more than nine million—have been affected, at least in part, by exposure to excessive noise. The sense of hearing is a fragile and vital function of the human body. It is similar to vision, more so than the other senses, because permanent and complete damage can be sustained quite commonly in an industrial environment. So it follows that noise pollution has been recognized as a safety concern for a long time, and has been appropriately regulated. Although personnel safety may be the most important, noise pollution has several other side effects that * A glossary of terms concerning noise is given at the end of this chapter before the references.
© 2011 by Taylor and Francis Group, LLC
are significant. Combustion equipment designers are often asked why they would want to constrain the combustion process in order to reduce noise. Frequently these questions come from persons working in plants situated in remote areas. However, given the age and economic drivers of the petroleum refining and chemical industries, currently it is common to find plants located in densely populated areas. With industry that is situated close to residential areas or busy commercial facilities, high levels of noise become objectionable to people in the neighborhood. This eventually leads to government regulations to control noise emissions. On the other hand, within the industrial site itself, the immediate issue with noise is one of employee safety. Furthermore, it is not surprising to find that employee moral and performance improves when noise is reduced. Equipment too is affected by noise. In most cases these effects are addressed in the realm of vibration control and are beyond the scope of this chapter. Suffice it to say, that noise is indeed a form of vibration and eventually contributes to fatigue, which reduces equipment life. Usually the effects of fatigue are long term and are possibly accepted as normal wear and tear if the equipment life cycle spans a reasonable duration. In extreme cases, the effects of vibration may be more rapidly manifested, such as in the case of cracking and falling of the hard refractory linings in furnaces. This chapter has been written as a practical guide as well as a reference on noise for engineers involved in design, operation, or maintenance of combustion equipment, be it burners, furnaces, flares, or thermal oxidizers. In addition, since this chapter provides a comprehensive coverage of the fundamentals of sound, the creative engineer will be able to extend his or her knowledge to analysis of other noise producing industrial equipment as well. 8.1.2 Basics of Sound If a tree falls in the forest and nobody is around to hear it, does it still make a sound?
Webster’s dictionary defines sound as “That which is heard.” Obviously, an engineer will find this definition woefully inadequate for his or her purposes. So we resort to the definition provided by the Handbook of Noise Measurement: “Sound is the vibration of particles in a gas liquid or solid” [1]. Sound is propagated through any medium in waves that take the form of pressure peaks (compressions) and troughs (rarefactions) as illustrated in Figure 8.1. The pressure wave travels through the medium at the speed of sound. The auditory system in humans and most animals senses the impingement of these pressure waves
185
Combustion Noise
level and frequency. A tuning fork is a good example of a pure tone generator. Such naturally occurring pure tone generators are rare. Even musical instruments create notes that have significant pressure levels at two or more multiples, or harmonics, of the fundamental frequency of the note.
Sound wave - a pressure wave moving at the speed of sound
8.1.2.1 Sound Pressure Level and Frequency + Pressure –
Figure 8.1 Pressure peaks and troughs.
Eardrum
Nerve that sends electrical signals to the brain Cochlea
Ear canal Fluid filled membrane
Hair cells Figure 8.2 Cross-section of human ear.
on a tissue membrane and converts them to electrical impulses that are then sent to the brain and interpreted there. Figure 8.2 is a cross-section of the human ear. Sound is collected and funneled into the ear canal by the outer ear. At the end of the ear canal, the sound impinges upon a membrane called the ear drum. The bones of the middle ear convey the ear drum’s vibration to the inner ear. The inner ear consists of a fluid filled Basilar membrane that has tiny hair cells on the inside. The hair cells sense the vibration conveyed to the Basilar membrane and convert this into electrical signals that are then conveyed to the brain. In reality, most naturally occurring sounds are composites of different pressure levels at various frequencies. On the other hand, a pure tone is a sound at only one frequency. Any pure tone can be uniquely identified by two of its properties, namely, pressure
© 2011 by Taylor and Francis Group, LLC
Pressure level defines the loudness of the sound, while frequency defines the pitch or tone of the sound. Pressure level is the amplitude of the compression, or rarefaction, of the pressure wave. The common unit of pressure level is Decibel, abbreviated to “dB.” Frequency is the number of pressure waves that pass by an arbitrary point of reference in a given unit of time. As such, the measure of sound frequency can be cycles per second (CPS), and as with electricity, the commonly used unit is the Hertz. One Hertz is one cycle per second. Another important quantity in the description of sound waves is the speed of sound, c, which describes the velocity at which a sound wave propagates through the ambient air or other fluid (e.g., water). In other words, the speed of sound describes the distance a sound wave propagates per unit of time. As a third quantity that characterizes a sound wave, the wavelength λ must be mentioned. It describes the distance between one wave crest and the next (i.e., between one point of maximum compression in the sound wave to the next). The frequency f, the speed of sound c, and the wavelength λ are related as follows:
c = λ × f.
(8.1)
The speed of sound depends mainly on the fluid in which the sound wave propagates and the temperature. Table 8.1 gives some examples for the speed of sound in different media and at different temperatures. The typical range of human hearing extends from 20 Hz to 20 kHz. Young children can hear frequencies slightly higher than 20 kHz but this ability diminishes with age. This trend of reduced high frequency sensitivity continues with advancing age. Loss of hearing in humans in the later stages of life typically manifests itself as diminished sensitivity to frequencies from 10 to 20 kHz. Mechanically, this is due to the deterioration of the fine hair cells in the Basilar membrane. It is important to note that the ear is not equally sensitive over the entire range from 20 Hz to 20 kHz. This is vital to understanding how noise affects us and how noise control is implemented. The human ear is much less sensitive to sound at the extremes of low and high frequencies and we will discuss this in more detail later in the chapter.
186
Industrial Combustion Testing
Table 8.1
Temperature [°C]
Speed of Sound c [m/s]
Air, dry
−40 20 0 20 40
307 320 332 344 356
Hydrogen Methane Ethylene Water
20 20 20 0 40 80 100
1316 448 331 1403 1529 1555 1543
Steel Soft rubber
20 20
5180 54
Fluid
Table 8.2 The Ten Octave Bands Full Octave Band Standards Octave Band, Hz
Center Frequency, Hz
22–44 44–88 88–177 177–355 355–710 710–1420 1420–2840 2840–5680 5680–11,360 11,360–22,720
31.5 63 125 250 500 1000 2000 4000 8000 16,000
The wide range of frequencies in our hearing range may be conveniently handled by breaking it up into octave bands. Each octave band represents a doubling in frequency. Table 8.2 shows the 10 octave bands that cover the hearing range and the center frequencies that can be used to represent the octave band. Each octave band extends over seven fundamental musical notes. 8.1.2.2 Decibel The unit of sound level, the decibel, is difficult to visualize and merits some explanation. While it is possible to quantify sound in units of either power or pressure, neither unit is convenient to use. This is because, in practice, we have to deal everyday with sounds that extend over a very large range of power
© 2011 by Taylor and Francis Group, LLC
Power, W
Speed of Sound in Different Media and at Different Temperatures
1×101 1×100 1×10–1 1×10–2 1×10–3 1×10–4 1×10–5 1×10–6 1×10–7 1×10–8 1×10–9 1×10–10 1×10–11 1×10–12
0 10 20 30 40 50 60 70 80 90 100 110 120 130 dB
Figure 8.3 Relationship of decibels to watts.
or pressure values. For example, the sound power of a whisper is 10−9 watts, while the sound power of a jet plane is 103 watts. The range of these two sound sources spans 1012 ratio. The decibel, a dimensionless unit, was invented in order to represent these large ranges conveniently. In the 1960s Bell Laboratories coined the term decibel. The “deci” stands for the base 10 Log scale on which the decibel is based. See Figure 8.3 for how decibel relates to watts. The “bel,” of course, stood for Bell Labs. In Figure 8.3, the Y-axis is power in watts given in scientific notation. The Y-axis follows a base-10 scale. The X-axis gives dB values and the line provides the relationship: 120 dB is equal to 1 watt. As an illustration of the log10 relationship, note that 110 dB is equal to onetenth of a watt (0.1 watt) and 100 dB is equal to 100th of a watt. 8.1.2.3 Sound Power Level There is a subtle but important difference between the terms Sound Power Level (PWL), and Sound Pressure Level (SPL). Sound power level is used to indicate the total energy emitting ability of a sound source. In other words, sound power is an attribute of the source itself. Sound pressure level, on the other hand, is used to indicate the intensity of sound received at any point of interest, from one or more sources. The illustration in Figure 8.4 shows the formula to calculate the SPL to be expected at a distance r from a spherically radiating source of power level Lw:
Lp = Lw − 10 log10 (4πr2) + 10.5,
where Lp = sound pressure level in dB Lw = Sound power level in dB r = Distance from source in ft.
(8.2)
187
Combustion Noise
Equation 8.2 is only valid exactly if the following conditions are fulfilled:
1. Noise radiation from the source is uniform and equal in every direction. 2. The source is small compared to the distance r. 3. There is no relevant noise contribution from other sources at the point of interest. 4. There is no relevant influence on the noise at the point of interest from sound reflecting surfaces nearby.
If the above conditions are sufficiently fulfilled, then Equation 8.2 can not only be used to calculate the SPL expected at a certain distance r from the source if the sound power level Lw is known, but also to backcalculate the PWL of a source from a measurement of the sound pressure level at a known distance r from the source. The latter procedure is a technique that is standardized in ISO 3744 [2]. If the above conditions are not fulfilled, then the SPL predicted at a certain distance from a sound source can differ significantly from the level actually measured there, or the PWL of a source back-calculated from a measurement of the SPL can differ significantly from the real SPL of the source. In this context, it is important to keep in mind that the PWL is an intrinsic property of the source that is independent from the environment while the SPL not only depends on the distance to the source, but is also influenced by other sound sources, the environment (reflecting surfaces. etc.), and the sound propagation conditions (temperature, atmospheric pressure, wind, barriers, etc.).
The following are useful equations that may be used to calculate sound pressure and power levels, in dB, from the equivalent pressure and power units.
Lp (dB) = 20 log10 P/(2 × 10−5),
Lw (dB) = 10 log10 W/(1 × 10−12),
where Lp = Lw = P = W =
sound pressure level in dB, sound power level in dB, pressure in N/m2, sound power level in watts.
8.1.2.4 Threshold of Hearing Figure 8.5 shows a curve, which is a map of the threshold of hearing in humans. The Y-Axis is SPL and the X-Axis is frequency. Any SPL at a frequency that falls below the curve will be inaudible to humans. For example a SPL of 30 dB at 63 Hz, being below the curve, will be inaudible. Whereas, a SPL of 70 dB at the same 63 Hz will be audible. Humans are most sensitive to sounds in the so called midfrequencies from 1 kHz to about 5 kHz. This is generally the region in which most of our everyday hearing activities take place. The human voice, most musical instruments, and so on produce sounds predominantly in the midfrequencies. Additionally, at a constant level, sound with a low or very high frequency will not have the same loudness sensation as that in the medium frequency range. For example, a 100 Hz tone at Lp = 50 dB gives the same loudness as a 1000 Hz tone at Lp = 40 dB [3].
130
Figure 8.4 Calculating sound pressure level at a distance r.
© 2011 by Taylor and Francis Group, LLC
Human threshold of hearing
50 30 10
Frequency, Hz Figure 8.5 Threshold of hearing in humans.
20000
16000
8000
4000
2000
1000
500
250
–30
125
–10 63
Lp = Lw + 10 log10 1 +10.5 4π r 2 where r is in feet.
70
31.5
Noise source with a given sound power level (Lw) radiating outward.
90
16
The sound pressure level at this point can be calculated as
r
Sound pressure level, dB
110 Sound pressure wave at a distance r feet from the source.
188
Industrial Combustion Testing
80
Figure 8.6 also shows the threshold of pain superimposed. Fortunately for us, the threshold of pain is relatively flat. Generally a SPL of over 120 dB at any frequency will cause pain. An important observation that can be derived from the two curves is that if a sound is audible at very low or very high frequencies, persons subject to this sound are very close to experiencing pain.
60
0
8000
16,000
8000
16,000
4000
2000
1000
250
125
31.5
A-scale correction factor 63
–40
Frequency, Hz Figure 8.7 A-weighted scale.
85 80 75 70 65 60
A-weighted
55 50
4000
2000
1000
500
250
125
40
63
45 31.5
Sound pressure level, dB
Sound meters have the capability to measure with equal sensitivity over the entire audible range. However, since humans do not hear with equal sensitivity at all frequencies, the sound meter measurement needs to be modified to quantify what really affects us. This can be done using a correction curve. The most common correction is the A-scale correction curve, which resembles an idealized inverse of the threshold of hearing curve (refer to Figure 8.7). An A-weighted sound level correlates reasonably well with hearing-damage risk in industry and with subjective annoyance for a wide category of industrial and community noises. After applying the A-scale correction, dBA is usually used as the unit of SPL. Figure 8.8 shows a typical burner noise curve as measured by the noise meter (flat scale) and the result after applying A-scale correction. The other, less used correction scales are named, as may be expected B, C, and D. Referring to Figure 8.9 we can see that the C-scale is essentially flat over the range of interest and the B-scale lies somewhere between the A and C scales. With the understanding we have today of the influence of low-frequency sounds, we find that the B and C scales, although meant to be used with lowfrequency sounds, do not apply adequate correction in
Threshold of hearing
20
–20
8.1.2.6 Correction Scales
Frequency, Hz Figure 8.8 Typical burner noise curve.
+20
130
+10
110
0
70
Relative response, dB
Threshold of pain
90
Threshold of hearing
50 30 10
–10 –20
© 2011 by Taylor and Francis Group, LLC
20,000
16,000
8000
4000
2000
1000
500
250
125
63
–60
31.5
–30
Figure 8.6 Threshold of hearing and threshold of pain for humans.
B C
–40 –50
Frequency, Hz
A
–30
–10 16
Sound pressure level, dB
40
500
Sound pressure level, dB
8.1.2.5 Threshold of Pain
–70 10
D
2
Figure 8.9 Weighting scales.
5
102
103 Frequency, Hz 2
5
2
5
104
2
189
Combustion Noise
the lower frequencies. Lastly, the D-scale is different from the others, in that it has a pronounced correction in the range of 2 kHz to 5 kHz. The D-scale was devised for the aircraft industry and is rarely used otherwise. 8.1.3 Measurements A simple schematic of a noise meter is shown in Figure 8.10. The microphone is a transducer that transforms pressure variations in air to a corresponding electrical signal. Since the electrical signal generated by the microphone is relatively small in magnitude, a preamplifier is needed to boost the signal before it can be analyzed, measured, or displayed. Special weighting networks are used to shape the signal spectrum and apply the various correction scales discussed above. The weighted signal then passes through a second output amplifier into a meter. The meter and associated electronic circuits detect the approximate rms-value of the signal and display it in dB. Noise meters range from the simplest—microphone and needle gauge—to sophisticated digital signal processing (DSP) equipped analyzers. The more sophisticated analyzers are equipped with fast Fourier transform (FFT) capabilities that aid accurate narrow band analysis. In general spectrum analyzers, allow the user to map the SPL at different frequencies, or in other words, generate a curve of the sound over different frequencies. However, there is a great deal of difference between instruments than make one measurement per octave band and those that slice the octave band up into several intervals and make a measurement at each interval. Typically, instruments are capable of:
1. Octave band measurements 2. One-third octave band measurements 3. Narrow band measurements
Table 8.3 shows the usual octave and one-third octave bands. As the name suggests, a one-third octave band instrument makes three measurements in each octave as opposed to the single measurement of the octave band instrument. A narrow band instrument, on the other hand, uses DSP to implement analysis FFT, and in the current state of the art, FFT analysis allows the analyzed frequency range to be sliced up into a large number of smaller intervals, limited in number only by the measured time interval length and the available computer power. Figure 8.11 is a comparison of the same sound spectrum as analyzed using three different band intervals—octave band, one-third octave, and narrow band. This comparison shows that the additional resolution provided by narrower band methods can be of vital importance. In this example the level at 1 kHz, as recorded by the octave band instrument is 90 dB, on the one-third octave instrument it is 85 dB, and on the narrow band instrument it is 70 dB. The lower resolution measurements produce higher measurements due to the spillover influence of the nearby peak at 1.8 kHz. In addition, in implementing noise control for this source, it is very valuable to know that it is the narrow peak at 1.8 kHz that is driving the maximum noise. This knowledge helps to zero in on the source, which, for example, may be a fan at 3600 rpm and with 30 blades. This results in a tonal frequency (blade passing frequency) of 3600/60 × 30 = 1800 Hz (see Section 8.3.2 for the calculation of the tonal frequency of fan noise). However, as with all things, there is a cost associated with high performance. For most purposes, a one-third octave analysis is usually quite adequate. The advantage of making broad band analyses using octave or one-third octave band filter sets are that less time is
Weighting networks A Microphone
Amplifier
C D Flat
Band filters Figure 8.10 Schematic of a noise meter.
© 2011 by Taylor and Francis Group, LLC
Rectifier
B
Fast Meter
Amplifier
Output
Slow
190
Industrial Combustion Testing
Table 8.3 Octave and One-Third Octave Bands Octave Band 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43
Lower Band Limit
Center
One-Third Octave Upper Band Limit
11
16
22
22
31.5
44
44
63
88
88
125
177
177
250
355
355
500
710
710
1,000
1,420
1,420
2,000
2,840
2,840
4,000
5,680
5,680
8,000
11,360
11,360
16,000
22,720
Lower Band Limit
Center
14.1 17.8 22.4 28.2 35.5 44.7 56.2 70.8 89.1 112 141 178 224 282 355 447 562 708 891 1,122 1,413 1,778 2,239 2,818 3,548 4,467 5,623 7,079 8,913 11,220 14,130 17,780
16 20 25 31.5 40 50 63 80 100 125 160 200 250 315 400 500 630 800 1,000 1,250 1,600 2,000 2,500 3,150 4,000 5,000 6,300 8,000 10,000 12,500 16,000 20,000
Upper Band Limit 17.8 22.4 28.2 35.5 44.7 56.2 70.8 89.1 112 141 178 224 282 355 447 562 708 891 1,122 1,413 1,788 2,239 2,818 3,548 4,467 5,623 7,079 8,913 11,220 14,130 17,780 22,390
Note: The advantages of making broad analyses of sound using octave or one-third octave band filter sets are that less time is needed to obtain data and the instrumentation required to measure the date is less expensive. The main disadvantage is the loss of detailed information about the sound which is available from narrow band (FFT) analyzers.
needed to obtain data and the instrumentation required to measure the data is less expensive. In making sound measurements, several factors regarding the nature of the source should be considered. Whether the source is a true point source in space radiating spherically, or whether it is a hemispherical source close to one flat surface or a quarter sphere between two flat surfaces, and so on, it will make a difference in how the measurement needs to be performed. However, a detailed discussion of measurement issues is beyond the scope of this chapter and the reader may use some of the more comprehensive
© 2011 by Taylor and Francis Group, LLC
works in the list of references at the end of this chapter. The American Petroleum Institute has issued a recommended practice for measuring noise from fired process heaters [4]. 8.1.3.1 Overall Sound Level and How to Add dB Values As mentioned before, most sounds are composites of several different levels at different frequencies. This is especially true of industrial noise. A typical burner noise curve is shown in Figure 8.12. As can be seen, there are significant levels in two frequency zones, both of which
191
Combustion Noise
(a)
Table 8.4
100 90
A-Weighting of the Burner Sound Curve from Figure 8.12
80
Frequency Hz 31.5 63 125 250 500 1000 2000 4000 8000 16,000
70 60 50 (b)
Octave-band spectrum
100 90 80 70 60 50
(c)
Third-octave band spectrum
100 90 80 70 60 50
10
100 1K Narrow-band spectrum
10K
Figure 8.11 Comparison of same sound spectrum analyzed using three different band intervals.
80 75 70 65 60 55 50
Frequency, Hz Figure 8.12 Typical burner noise curve.
© 2011 by Taylor and Francis Group, LLC
16000
8000
4000
2000
1000
500
250
125
31.5
40
A-Scale CF dB
SPL dBA
72 75 79 79 72 69 68 78 83 80
−39 −26 −16 −9 −3 0 1 1 −1 −7
33 49 63 70 69 69 69 79 82 73
will contribute to the apparent intensity experienced by a person working in the vicinity of the burner. It is difficult to describe this sound without using either a diagram like the one shown or a table listing various SPLs occurring in the different octave bands. The overall sound level, a single number, has been devised to represent such composite sound curves conveniently. If a single number is to be used to represent the whole curve, then it should adequately represent the peaks in the curve, since the peaks have the most influence on the listener. Consequently, it is not practical to use the average of the various levels in the octave bands, since this number would be less than the levels at the peaks. Therefore, one must not confuse the average with the overall sound level. The overall sound level is calculated by adding the individual levels in the various octave bands. In columns 1 and 2 of Table 8.4, the burner sound curve has been split up into its component levels in each octave band. In column 3, the A-weighted correction has similarly been split up and listed. Column 4 gives the A-corrected values for the sound curve by simply adding column 3 to column 2. Now, the values in column 4 must be added to obtain the A-weighted overall sound level. Since the decibel is based on a Log10 scale, simple addition cannot be used. For example, if two values of equal magnitude are added, say 100 dB and 100 dB, the result is 103 dB. The formula to be used is as follows:
45 63
Sound pressure level, dB
85
SPL dB
Ltotal = 10 log( ∑ i =1 to n 100.1Li ),
where Ltotal = the total level Li = each individual level n = the number of levels to be added.
192
Industrial Combustion Testing
Table 8.5 Addition Rules Frequency Hz 31.5 63 125 250 500 1,000 2,000 4,000 8,000 16,000
SPL dB
A-Scale CF dB
SPL dBA
72 75 79 79 72 69 68 78 83 80
−39 −26 −16 −9 −3 0 1 1 −1 −7
33 49 63 70 69 69 69 79 82 73
49 63 71 73 75 76 80 84 82
And subtraction can be performed by using
Ldiff = 10 log(100.1L2 − 100.1L1 ).
However, some simple rules of thumb may be used to perform quick estimates. They are as follows:
1. When adding dB values of equal magnitude or differ by 1, the sum is 3 dB added to the greater number. 2. When the two values are different by 2 or 3 dB, then the sum is 2 dB added to the greater number. 3. When adding two values that differ by 4–9 dB, then the sum is 1 dB added to the greater number. 4. For values that differ by 10 dB or more, the sum is just the larger number. 5. Always start with the smallest number in the list and add it to the next larger number.
To understand why these rules work refer to the chart in Figure 8.3. From the chart it can be seen that one watt is equal to 120 dB.
1 watts = 120 dB 1 watts = 120 dB 2 watts = 123 dB
On the chart, 2 watts registers 123 dB on the line. Similarly, the reason numbers that are 10 dB or more in difference are neglected is because:
1.0 watts = 120 dB 0.1 watts = 110 dB 1.1 watts = 120 dB
© 2011 by Taylor and Francis Group, LLC
ΔL 0
20
0.5
10 9 8
1
7
6
1.5
5
4
2
3 L1–L2
2
2.5
3
1
0
Figure 8.13 Nomogram for level addition.
Since the 110 dB contributes only a 10th of a watt of power, we neglect it in approximation. The example becomes more vivid when we add two numbers that differ by 20 dB or more.
1.00 watts = 120 dB 0.01 watts = 100 dB 1.01 watts = 120 dB
Rule number 5 is especially necessary when adding a list that contains several numbers that are almost equal in value and one or more that are 10 dB greater, such as in a list that contains six values of 90 dB and one value of 100 dB. If we begin to add from the 100 dB value we will arrive at a wrong result. Reader beware, the rules provided are approximations. For exact calculations, the formulas should be used. Table 8.5 shows the effect of applying the addition rules to the values generated by breaking up the burner noise curve. At the end of the addition list, 1 dB has been added to compensate for any errors due to approximation. As an alternative to using the above rules of thumb for level addition, the nomogram in Figure 8.13 can be used to determine the sum of the two levels L1 and L2: For a certain difference between the two levels L1 and L2 in the lower scale of the nomogram, the corresponding
8.2 Industrial Noise Pollution Thus far, sound has been discussed. So what is noise? An all-encompassing definition would be that noise is any undesirable sound. By saying this, the concept is introduced that what is considered to be noise is somewhat relative and depends on several temporal and circumstantial factors. For example, it is not unusual for a person to encounter SPLs of 100 to 110 dB at a sporting event, in a stadium full of cheering fans, and yet not be perturbed by it. On the contrary, the barely 45 dB sound of a dripping faucet may cause considerable annoyance in the quiet of the night. Table 8.6 gives some typical noise levels for various scenarios. Industrial noise pollution is a major concern for society as a whole. In a recent survey, the effects of exposure to noise in refinery workers was studied extensively. A cross-section of workers in different divisions/units was chosen. It was found that noise levels averaged
© 2011 by Taylor and Francis Group, LLC
30
40
355/710 Hz, 4th octave band, GMF 500 Hz
20 0
0 10
710/1400 Hz, 5th octave band, GMF 1000 Hz
0 20
1400/2800 Hz, 6th octave band, GMF 2000 Hz
Relative 40 humidity, % 10 20 20 30 50 0 70 90
60 40
10
20 0 10 20
–10
30 40 50 60 70 80 Temperature, °F 0
Temperature°C 10 20
30
2800/5600 Hz 7th octave band GMF 4000 Hz
30
90 100
40 100
10
20 0
10 20
–10 50
30 40 50 60 70 80 90 100 Temperature, °F 0
Temperature,°C 10 20 30
160 Relative humidity, % 10 20 30 50 60 70 90
30 20
0
40
5600/11,200 Hz 8th octave band GMF 8000 Hz
40
140 120 100 80 60 40
10 0
0
Relative humidity, % 80 10 20 60 30 50 70 40 90
20
0
40
dB/1000 m
Temperature°C 10 20
dB/1000 m
0
dB/1000 m
Atmospheric attenuation, dB/1000 ft
10
0
Atmospheric attenuation, dB/1000 ft
When a sound wave travels through still air it is absorbed or attenuated by the atmosphere. Over a couple of hundred feet, the atmosphere does not significantly attenuate the sound, however, over a few thousand feet the sound level can be substantially reduced. The amount of sound that is attenuated in still air depends largely on the atmospheric temperature and relative humidity. Figure 8.14 are plots showing the atmospheric attenuation for aircraft-to-ground propagation in SPL per 1000 feet (300 m) distance for center frequencies of 500, 1000, 2000, 4000, and 8000 Hz. Notice that the atmospheric attenuation is more significant at higher frequencies than lower ones. For example, suppose that we are 1000 feet (300 m) away from a noise source and that the atmospheric temperature and relative humidity is 80°F (27°C) and 10%. The plots in Figure 8.14 show that the atmospheric attenuation for 500 Hz is approximately 2 dB whereas for 8000 hertz the attenuation is 55 dB. For an example, see Section 8.5.3. Atmospheric attenuation, outdoors, can also be affected by turbulence, fog, rain, or snow. Typically, the more turbulence in the air, the more the sound is attenuated. There appears to be conflicting evidence as to whether or not fog attenuates sound. It is recommended that no excess attenuation be assigned to fog or light precipitation.
Atmospheric attenuation, dB/1000 ft
8.1.3.2 Atmospheric Attenuation
–10
dB/1000 m
level increase ΔL to be added to the higher one of the two levels can be found in the upper part of the nomogram.
dB/1000 m
193
Combustion Noise
20 0
10 20 30 40 50 60 70 80 90 100 Temperature, °F
0
Figure 8.14 Atmospheric attenuation.
87 to 88 dBA in the aromatic and paraffin facilities and 89 dBA in alkylation facilities. In comparison, workers in the warehouse, health clinics, laboratories, and offices were, generally, found to be exposed to much lower levels. Noise can damage hearing, and can cause physical or mental stress (increased pulse rate, high blood pressure,
194
Industrial Combustion Testing
Table 8.6
Table 8.7
Sound Levels of Various Sources
OSHA Permissible Noise Exposure
Threshold of hearing Rustle of leaves Normal conversation (at 1 m) Min. level in Chicago at night City street, very busy traffic Noisiest spot at Niagara Falls Threshold of pain Jet engine (at 50 m) Rocket (at 50 m)
Duration per Day (Hours)
0 dBA 10 dBA 30 dBA 40 dBA 70 dBA 85 dBA 120 dBA 130 dBA 200 dBA
8.0 6.0 4.0 3.0 2.0 1.5 1.0 0.5 0.25 or less
nervousness, sleep disorders, lack of concentration, and irritability). Irreparable damage can be caused by single transient sound events with peak levels exceeding 140 dBA (e.g., shots or explosions). Long duration noise exceeding 85 dBA can lead to short-term reversible hearing impairment and long-term exposure to higher levels than 85 dBA can cause permanent hearing loss. The following is a mathematical model based on empirical data (ISO 1999) used to calculate the maximum permissible continuous noise level at the work place that will not lead to permanent hearing loss:
L A,m < 85 + 10 log(24/Tn) dBA,
(8.3)
where Tn = daily noise exposure time in hours. Wearing ear protection devices at continuous noise levels greater than 85 dBA can prevent or reduce the danger of permanent hearing damage. 8.2.1 OSHA Requirements Title 29 CFR, Section 1910.95 of the Occupational Safety and Health Act (OSHA) pertains to the protection of workers from potentially hazardous noise. Table 8.7 shows OSHA permissible noise exposure levels. Required by OSHA, the employer must provide protection against the effects of noise exposure when the sound levels exceed those shown in Table 8.7. When the daily noise exposure consists of two or more periods of noise exposure at different levels, their combined effect should be considered rather than the individual effects of each. According to OSHA the exposure factor (EF) is defined as [5]:
EF = C1/T1 + C2/T2 + C3/T3 + … + Cn/Tn,
(8.4)
where Cn is the total time of exposure at a specific noise level and Tn is the total time of exposure permitted at that level and shown in Table 8.7. If the EF exceeds 1.0, the employee’s exposure is above OSHA limits. If OSHA identifies such a situation, a citation may be issued and a
© 2011 by Taylor and Francis Group, LLC
Sound Pressure Level, dBA (Slow Response) 90 92 95 97 100 102 105 110 115
Note: Exposure to impulsive or impact noise should not exceed 140 dBA.
grace period defined in which the employer must correct the violation or face penalties of even $10,000 per day. 8.2.2 International Requirements Regulations aimed at protecting individuals from industrial noise pollution have been enforced in almost all industrialized countries. The noise caused in industries and the work place is generally taken as a serious issue. Most countries have adopted 85 dBA as the limit for the permissible noise. At any workplace with sound levels exceeding 85 dBA, ear protection devices must be worn, and workers exposed to this level should have their hearing level checked periodically. 8.2.3 Noise Sources and Environment Interaction The main individual sources of noise in chemical and petrochemical plants are: burners (process furnaces, steam boilers, and flares), fans, compressors, blowers, pumps, electric motors, steam turbines, gears, valves, exhausts to open air, conveyors and silos, airborne splash noise from cooling towers, coal mills, and loading and unloading of raw and finished materials. Although noise pollution caused by industrial sectors are minor as compared to the noise caused by road and rail traffic, industrial noise receives more attention due to public representation. The ISO 1996 provides a set of international regulations for noise protection in residential neighborhoods located near industrial areas. National or local authorities must enforce noiselimiting values that should not be exceeded in the neighborhood. The magnitude of limiting value, additional charges for tonality and impulsive noise, and the legalities change not only from country to country, but sometimes within different states and regions in the same country. In general, nighttime noise limits are 10 to 15 dB lower than that for the daytime.
195
Combustion Noise
There are four major mechanisms of noise production in combustion equipment. They can be categorized as predominantly high frequency or low frequency sources. They are: Low Frequency Noise Sources Combustion roar and instability Fan noise High Frequency Noise Sources Gas jet noise Piping and valve noise 8.3.1 Combustion Roar and Combustion Instability Noise
5m
5m
In order to understand combustion roar, the mixing process taking place between the fuel and the oxidant on a very minute scale must be considered. It is known that a well-blended mixture of fuel and air will combust very rapidly if the mixture is within the flammability limits for that fuel. On the other hand, a raw fuel stream that depends on turbulence and momentum to mix in the ambient fluid, to create a flammable mixture, tends to create a slower combustion process. In either case, when regions in the mixing process achieve a flammable mixture and encounter a source of ignition, combustion takes place. The closer a mixture is to stoichiometry when it encounters the ignition source, the more rapid will be the combustion. Combustion occurring close to stoichiometry converts more of the energy release into noise. Figure 8.15 shows an example of a test flare
Figure 8.15 Test flare at John Zink test site in Tulsa. Combustion of identical fuel flow rates with different degrees of mixing. Total noise emissions of the flame in the left picture are about 5 dB higher than for the flame in the right picture.
© 2011 by Taylor and Francis Group, LLC
8.3.1.1 Flare Combustion Roar It has been recognized for a long time that the noise emitted from a normal operating flare has two mechanisms at work, namely, combustion roar and gas jet noise. Combustion roar typically resides in the lower frequency region of the audible frequency spectrum while gas jet noise occurs in the higher frequencies as illustrated in Figure 8.16.
Noise level
8.3 Mechanisms of Industrial Combustion Equipment Noise
operated at the same flow rate of fuel, but with different degrees of premixing. In the left picture, the fuel flow is well mixed with air and thus quickly burns in a crisp smokeless flame. In the picture on the right, the same fuel flow rate is burned with very poor premixing and a slowly burning large flame is generated. The difference in overall PWL of the noise generated by the combustion process itself is significant: For the operational mode shown in the left picture, it is about 5 dB higher than for the operational mode in the right picture. See the discussion on thermoacoustic efficiency (TAE) later in this section for more details. The noise coming from each small region of rapidly combusting mixture adds up to create what we call combustion roar. Therefore, combustion roar is largely a function of how rapidly the fuel is being burned. In addition, in the context of combustion equipment like burners and flares, usually the larger the fuel release, the more the turbulence in the combustion process. Since turbulence directly influences the mixing rate, high turbulence processes also produce more combustion roar. So it is more accurate to state that the level of combustion roar generated from a combustion process is a function of the amount of fuel being burned and how rapidly we arrange to burn it.
Region of combustion roar
Region of gas jet noise
Noise frequency (Hertz) Figure 8.16 Illustration showing typical noise signature emitted from a flare operating under normal operating conditions.
196
© 2011 by Taylor and Francis Group, LLC
Table 8.8 Calculation of the Typical Combustion Noise Spectrum of a Stable Burning Flare from the Overall Sound Pressure Level OASPL Frequency (Hz)
Resultant Noise Spectrum (dB)
31.5 63 125 250 500 1,000 2,000 4,000 8,000
OASPL – 5 OASPL – 4 OASPL – 9 OASPL – 15 OASPL – 20 OASPL – 21 OASPL – 24 OASPL – 28 OASPL – 34 EF EF-SSA
175 165
EF-SSS
155
Lw A , dB
As mentioned before, the amount of combustion roar emitted from a flare, generally, depends on how fast the waste gas stream mixes with the ambient air. A waste gas stream that exits a flare tip with a low velocity and low levels of turbulence will mix slowly with the ambient air and burn relatively quietly. These types of flames are called buoyancy-dominated flames. A waste gas stream that exits a flare tip with a high velocity and high levels of turbulence, however, will burn much faster and create substantially more combustion noise for the same heat release rate. These high velocity flames are momentum-dominated flames. Increasing the rate at which the waste gas burns results in “bigger explosions” of the air–fuel mixture. These bigger explosions create larger disturbances in the atmosphere resulting in higher levels of combustion roar. The combustion roar emitted from a flare flame is not highly directional and is considered to be a monopole source. That is, it is analogous to a spherical balloon whose surface is expanding and shrinking at various frequencies and emitting uniform spherical waves. High levels of turbulence in a flare flame are usually desirable because it helps reduce radiation and increase the smokeless capacity of the flare. Unfortunately, it increases the combustion roar. Unlike the solution for flare radiation reduction, it is not practical to increase the height of a flare stack or boom to reduce combustion noise. This is because even doubling the flue stack height will only reduce the SPL at the flare base by about 6 dB (see Equation 8.2). In addition, combustion roar is low frequency sound and so can travel a great distance without being substantially attenuated by the atmosphere. The signature of low frequency combustion roar noise typically consists of a broadband spectrum with a single peak. The combustion roar emitted from a stable burning flare typically peaks at a frequency of about 63 Hz. The combustion noise spectrum can be estimated from the overall sound pressure level (OASPL) by using the values in Table 8.8 [6]. It can be noted that, at frequencies above about 500 Hz, the noise contribution from combustion is relatively insignificant. A typical method for estimating the PWL emitted from a flare flame is to relate the energy released from the combustion of the waste gas stream (chemical energy) to the noise energy liberated by the combustion. The ratio of noise energy to chemical energy released from the combustion is called the TAE. For a stable-burning flare, the TAE typically varies between 1 × 10−9 to 3 × 10−6. The value of the TAE largely depends on the turbulent mixing of the waste gas with ambient air and is usually determined experimentally. In Figure 8.17, the sound power emitted by a number of industrial flares at different loads is shown. The
Industrial Combustion Testing
EF-SSS-NC
145 135
GF-SSS-NC
125 115 10
100 1000 Qcombust , megawatts
10.000
TAE = 1.9E-8 TAE = 4.5E-5
Figure 8.17 Sound power level Lw calculated from measured noise data, plotted versus heat release rate Q combust for different types of industrial flares under various operating conditions. (From Müller-BBM GmbH. “Noise Emissions of Different Flare Systems—Field Measurements Taken in Various Refineries and Petrochemical Plants.” Proprietary data, 1900–2004.) EF = elevated (single-point) flare, GF = (enclosed) ground flare, SSA = smoke suppression by air, SSS = smoke suppression by steam, NC = equipped with advanced noise control.
data are obtained from field measurements on several types of flares under various operating conditions [7]. The noise intensity is given as the sound power level Lw according to Equation 8.5 and is plotted versus the heat release rate of the flare Q combust, for which a logarithmic scale is used:
Wacoust . Lw = 10 ⋅ log 1 × 10−12 Watts
(8.5)
It is important to emphasize that in this context “operating condition” does not only refer to the heat release rate of the flare alone but also to all other parameters characterizing its operation, as for example, flow rates of auxiliary equipment (steam or air injection, etc.). Accordingly, flares shown in Figure 8.17 can operate under different conditions at the same heat release rate.
197
Combustion Noise
In Figure 8.17 lines of constant TAE are straight lines. The corresponding lines for the maximum and minimum TAE observed in the measurements—that is TAE = 1.9 × 10−8 and 4.5 × 10−5—are shown as the dashed and dashed-dotted lines. In designing flares, the combustion noise emitted from flares operating under various conditions is usually measured in order to determine the TAE. This information can then be used in computer programs that are used to model the level of combustion roar emitted from a flare. Figure 8.18 is a photograph of a John Zink engineer collecting noise levels from a flare using a real-time noise level meter. A flare flame that is highly turbulent, such as the high-pressure flare as shown in Figure 8.18, may have a TAE in the order of 1 × 10−6. However, a flame with low levels of turbulence, such as the butane cigarette lighter as shown in Figure 8.19 may have a TAE in the order of 1 × 10−9. For every order of magnitude that the TAE changes, the SPL will change by 10 dB. Example: Consider a flare burning a waste gas stream with a heat release of 5000 × 106 BTU/hr (1465 million watts). A noise measurement shows that the sound pressure level 400 feet (120 m) from the flame is 100 dB. What is the TAE of the flare flame? The sound power emitted from the flame, W, can be determined as follows:
Figure 8.18 Photograph of an engineer collecting flare noise data.
W ( watts)
Lp + 10 log 10 ( 4 πr 2 ) − 10.5 (8.6) = 1 × 10−12 anti log 10 , 10
where Lp is the sound pressure level in dB (100 dB for the example) and r is the distance from the flame in feet (400 feet or 120 m for the example). Substituting these values into Equation 8.6 gives W = 1792 watts. The TAE is then calculated to be TAE =
Acoustical Power 1792 watts = 1.2 × 10−6. = Thermal Power 14665 × 106 watts (8.7)
Since the TAE is in the order of magnitude of 1 × 10−6, one would expect that the flame would be highly turbulent and momentum-dominated. A central problem of the approach described above certainly lies in applying it to a whole flare system instead of the combustion process alone. Doing so means that TAE becomes a “lumped parameter” incorporating all effects that have an impact on the acoustical behavior of the flare (i.e., noise emissions from valves, injectors, smoke suppression devices etc.) as well as any noise control measures installed. Other
© 2011 by Taylor and Francis Group, LLC
Figure 8.19 Shadow photograph of a burning butane lighter.
approaches that are similar in that respect can be found in the literature [8]. Another problem with using the TAE concept to predict combustion noise is that it does not take the frequency characteristics of the noise emissions into account. Frequency characteristics are very important for the design of adequate noise control measures and will also determine the human perception of the noise emissions since the ear’s sensitivity for noise is frequency-dependent (see Section 8.1).
198
Industrial Combustion Testing
POI@100 m
POI@100 m
POI@100 m
POI@50 m
POI@50 m
POI@50 m
POI@30 m
POI@30 m
POI@30 m
POI@1 m
POI@1 m
POI@1 m
Point source Vert. area source Barrier Receiver Calculation area
> 60.0 dB(A) > 65.0 dB(A) > 70.0 dB(A) > 75.0 dB(A) > 80.0 dB(A) > 85.0 dB(A) > 90.0 dB(A) > 95.0 dB(A) > 100.0 dB(A) > 105.0 dB(A) > 110.0 dB(A) > 115.0 dB(A)
Figure 8.20 (See color insert following page 424.) Predicted sound pressure field contour plots for a multipoint LRGO flare system. Left: first stage in operation; Middle: stages 1 and 2 in operation; Right: all stages in operation.
Additional challenges arise for spatially extended flare systems that can no longer be treated as point sources. Prediction of the noise emissions of a grade-mounted multipoint flare system requires the correct modeling not only of the individual burners and their combustion and jet noise emissions, but also of the arrangement of the burners in the flare pit and the effect of the wind fence. Figure 8.20 shows the calculated sound pressure field contour plots for an LRGO flare system at different operating conditions. In summary, it can be concluded that the TAE concept applied to flare systems will usually only allow a very rough estimate of the actual noise emissions and the associated effect of these emissions on persons in the neighborhood of the flare. In some cases, the results can even be completely off the real situation. A key issue in developing reliable noise prediction tools for flares that are more generally applicable in a broad range of operating conditions lies in a proper treatment of the individual sources that contribute to the overall noise emissions of a flare system.
© 2011 by Taylor and Francis Group, LLC
8.3.1.2 Flare Combustion Instability Noise If a flame lifts too far above a flare tip, it can become unstable. An unstable flame will periodically lift and then reattach to the flare tip and create a low frequency rumbling noise. Typically, this low rumbling noise occurs in the frequency range of 5–10 Hz and is usually called combustion instability. Being as low in frequency as it is, combustion instability noise is usually inaudible and can travel over several miles without being substantially attenuated by the atmospheric air. When there are reports of shaking the walls and windows of buildings in the vicinity of a flare, it is usually due to combustion instability. Combustion instability noise can occur if too much steam or air is added in the base of the flame. So, overaerating the waste gas stream in a flare, causes the flame to periodically lift from the flare tip. This periodic lifting and reattachment of the flame from the flare tip is the mechanism that drives the low frequency rumbling noise. Usually, combustion instability noise can
199
Combustion Noise
100 Frequency, Hz
1000
Figure 8.21 Graph showing an example of a typical steam-assisted flare operating at a normal and an oversteamed condition.
be reduced by lowering the steam flow rate to a steamassisted flare or by lowering the blower airflow rate to an air-assisted flare. Figure 8.21 is a graph showing an example of a typical steam assisted flare operating at both a normal condition as well as an oversteamed condition [9]. Note that the combustion noise frequency shifts substantially to a lower region and the level dramatically increases when the flare is oversteamed. 8.3.1.3 Burner Combustion Noise Like flares, burner combustion noise is an unwanted sound associated with combustion roar and combustion instability. In many situations, the combustion noise can be the dominant source of noise emitted from a burner. Combustion roar and combustion instability are quite complex by nature. The literature contains a variety of combustion noise and combustion instability prediction techniques for burners operating in a furnace. Most of these prediction techniques are based on experimental studies that attempt to correlate the acoustic power radiated by the burner/furnace geometry, laminar burning velocity of the air–fuel mixture, and various turbulence parameters such as the turbulent length scale and intensity. This section will not attempt to discuss these prediction techniques in detail but will give a broad and general discussion of combustion roar and combustion instability noise using some of the results from these studies. Figure 8.12 is a plot showing a typical noise spectrum emitted from a burner operating under normal conditions in a furnace. Notice that the noise spectrum has two peak frequencies associated with it. The high frequency noise contribution is from the fuel gas jets while the low frequency contribution is from the combustion roar. Like the combustion roar emitted from flares, burner combustion roar is associated with a smooth
© 2011 by Taylor and Francis Group, LLC
8.3.1.4 Burner Combustion Instability Noise Combustion instability within a furnace is characterized as a high amplitude, low frequency noise often resembling the puffing sound of a steam locomotive. This type of noise can create significant pressure fluctuations within a furnace that can cause damage to the structure and radiate high noise levels to the surroundings. Figure 8.22 is a plot showing the SPL for a gas burner operating under normal conditions and with instability. It is obvious that the SPL increases substantially when the operation is accompanied by instability. Combustion instability noise has a high efficiency of conversion of chemical energy to noise. Typically the TAE from burner combustion instability is in the range of 1 × 10−4 [10]. The oscillations caused by combustion instability are naturally damped by pressure drop and other losses through the burner and furnace and therefore, can not be sustained unless energy is provided. These 115
With instability
110 105 100 95 90 85 80
Normal operation
75 70
Frequency, Hz
16,000
10
8000
70
4000
75
2000
Normal operating steam flare
80
1000
85
500
90
250
95
125
Noise level, dB
100
63
105
broadband spectrum having relatively low conversion efficiency from chemical energy to noise: in the range of 1 × 10−9–1 × 10−6. However, the combustion noise spectrum associated with a burner and a flare are not similar. The reason is that a flame burning in the open atmosphere will behave differently compared to a flame that is burning in an enclosed chamber such as a furnace. The combustion roar associated with flares typically peaks at a frequency of approximately 63 Hz while combustion roar associated with burners can vary in the 200–500 Hz range. Burner noise can have a spectrum shape and amplitude that can vary with many factors. Several of these factors include the internal shape of the furnace, the design of the burner muffler, plenum and tile, the acoustic properties of the furnace lining, the transmission of the noise into the fuel supply piping, and the transmissive and reflective characteristics of the furnace walls and stack.
31.5
Over-steamed flare
Sound pressure level, dB
110
Figure 8.22 SPL for a gas burner operating under normal conditions and with instability.
200
Industrial Combustion Testing
steady oscillations are sustained from energy extracted from the rapid expansion of the air–fuel mixture upon reaction. Over the years, furnace operators have used several techniques to try to eliminate combustion instability. Some of these techniques include modifying the (1) f urnace stack height, (2) internal volume of the furnace, (3) acoustical properties of the furnace lining, (4) pressure drop through the burner by varying the damper position, (5) fuel port diameter, (6) location of the pilot, and (7) flame stabilization techniques. Locations where orderly patterns are developing
8.3.2 Fan Noise The noise emitted from industrial fans typically consists of two noise components: broadband and discrete tones. Vortex shedding of the moving blades and the interaction of the turbulence with the solid construction parts of the fan creates the broadband noise. This broadband noise is of the dipole type, meaning that the noise is directional. On the other hand, the discrete tones are created by the periodic interactions of the rotating blades and nearby upstream and downstream surfaces. Discrete tonal noise is usually the loudest at the frequency at which a blade passes a given point. The tonal frequency is easily calculated by multiplying the number of blades times the impeller rotation speed in revolutions per second. The broadband and discrete tonal noise emitted from fans can radiate from both the suction and pressure side of a fan and through the fan casing. The noise can radiate downstream through the ducting and discharge into the environment at an outlet. Fan and duct systems should provide provisions to control this noise if residential areas are located nearby. Installation of mufflers and silencers on the suction and the discharge sides of the fan as well as wrapping of the casing and the ducts are common methods for reducing fan noise. 8.3.3 Gas Jet Noise Gas jet noise is very common in the combustion industry and in many instances it can be the dominant noise source within a combustion system. The noise that is created when a high-speed gas jet exits into an ambient gas usually consists of two principal components: gas jet mixing noise and shock-associated noise [11]. 8.3.3.1 Gas Jet Mixing Noise Studies have shown that a high-speed gas jet, exiting a nozzle, will develop a large-scale orderly pattern as shown in Figure 8.23. This orderly structure is known as the “global instability” or “preferred mode” of the jet. The presence of both the small-scale turbulent eddies within the jet and the large-scale structure is responsible for the gas jet mixing noise.
© 2011 by Taylor and Francis Group, LLC
Figure 8.23 High speed gas jet.
The source of gas jet mixing noise begins near the nozzle exit and extends several nozzle diameters downstream. Near the nozzle exit the scale of the turbulent eddies is small and predominantly responsible for the high frequency component of the jet mixing noise. The lower frequencies are generated farther downstream of the nozzle exit where the large-scale orderly pattern of the gas jet exists. Gas jet mixing noise consists of a broadband frequency spectrum. The frequency at which the spectrum peaks depends on several factors such as the diameter of the nozzle, Mach number of the gas jet, the angle of the observer’s position relative to the exit plane of the jet, and temperature ratio of the fully expanded jet to the ambient gas. In the flare and burner industry, gas jet mixing noise typically peaks somewhere between 2000 and 16,000 Hz. The characteristic shape is the same for all temperatures and angles, although there is a significant dependence on temperature and Mach number of the gas jet when the observer is positioned at an angle of less than 50° off axis from the centerline of the gas jet velocity vector [11]. The OA SPL created by gas jet mixing depends on several variables. These variables include the distance from the gas jet, the angle of the observer relative to the gas jet centerline velocity vector, Mach number, fully expanded gas jet area, and density ratio of the fully expanded jet to the ambient gas. In the Mach number M range of 0.7 < M < 1.6 the sound power W of a turbulent open jet increases with the flow speed to the power of 8 [12]; in the range M > 2 it increases to the power of 3 [13]. For the sound power, the following approximation is valid:
5 × 10−5 ρSu3 M 5 for M ≤ 1.82 W = −3 , 10 ρSu3 for M > 1.82
(8.8)
where ρ = density of the flowing medium at the jet opening in kg/m3, S = discharge opening cross-sectional area in m2,
201
Combustion Noise
Region of maximum jet mixing noise
Nozzle exit plane
30°
15°
Nozzle
Jet centerline
0°
Figure 8.24 Illustration showing the region of maximum jet mixing noise.
Shock waves Figure 8.25 Shock waves downstream of an air jet.
ht eec Scr
Noise level
The maximum overall SPL of gas jet mixing noise occurs at an angle between approximately 15°–30° relative to the centerline of the gas jet velocity vector as illustrated in Figure 8.24 [11]. As one moves in either direction from this angle, the noise level, in some cases, can drop off significantly. For example, the overall SPL created by gas jet mixing can be reduced as much as 25 dB when one moves from an angle of maximum noise level to an angle directly behind the nozzle (180°).
one s
u = jet discharge velocity in m/s, c = sonic speed of the flowing medium in m/s, M = u/c Mach number.
Broa
d dban
Shock noise from 4th shock cell
Shock associated noise spectrum Mixing noise spectrum
Nozzle exit Shock cell Noise frequency
8.3.3.2 Shock-Associated Noise
Figure 8.26 Photograph showing the location of where screech tones are emitted.
When a flare or burner operates above a certain fuel pressure a marked change occurs in the structure of the gas jet. Above a certain pressure, called the critical pressure, the gas jet develops a structure of shock waves downstream of the nozzle as shown in the photographs in Figure 8.25. The critical pressure of a gas jet typically occurs at a pressure of 12–15 psig (0.8 to 1 barg), depending on the gas composition and temperature. These shock cells consist of compression and expansion waves that repeatedly compress and expand the gas as it moves downstream. Using Schlieren photography, several investigators have seen as many as seven shock cells downstream of a nozzle. These shock cells are responsible for creating two additional components of gas jet noise: screech tones and broadband shockassociated noise. Screech tones are distinct narrowband frequency sound that can be described as a “whistle” or “screech.” The literature reports that these tones are emitted from the fourth and fifth shock cell downstream of the nozzle exit as shown in Figure 8.26 [14]. The sound waves from these shock cells propagate upstream where they interact with the shear layer at
the nozzle exit. This interaction then creates oscillating instability waves within the gas jet. When these instability waves propagate downstream they interfere with the fourth and fifth shock cell causing them to emit the screech tones. Screech tone noise is not highly directional, unlike gas jet mixing noise. The overall SPL is a function of the fully expanded jet Mach number and area, distance from gas jet and the ratio of the total temperature of the gas to the total ambient temperature. The peak frequency associated with screech tones depends on the nozzle diameter, fully expanded jet Mach number and velocity, and the angle of the observer relative to the jet centerline velocity vector. Broadband shock-associated noise occurs when the turbulent eddies within the gas jet pass through shock waves. The shock waves appear to suddenly distort the turbulent eddies that creates a noise that can range over several octave bands. The broadband shock-associated peak frequency noise typically occurs at a higher frequency than the screech tone peak frequency.
© 2011 by Taylor and Francis Group, LLC
202
Industrial Combustion Testing
8.3.4 Valve and Piping Noise When a gas flowing steadily in a pipe encounters a valve, a change in the flow pattern and pressure will occur that can create turbulence and shock waves downstream of the valve. Typically, when valves are partially closed, creating a reduction in flow area, the small flow passage behaves much like an orifice and produces jet noise. As discussed above, turbulence and shock waves create mixing noise and shock-associated noise. This noise can radiate downstream through the pipe and exhaust into the environment at an outlet and/or radiate through the pipe wall not only into the space near the valve itself as illustrated in Figure 8.27, but also at relatively large distances from the valve. Usually butterfly valves and ball valves are noisier than globe valves. Butterfly valves and ball valves typically have a smaller vena contracta than a globe valve operating at the same pressure drop that results in higher levels of mixing and shock-associated noise. As a general guideline when the pressure ratio across a valve is less than approximately three, the mixing noise and shock-associated noise are within about the same order of magnitude. Although, with pressure ratios greater than three, shock noise usually dominates over the mixing noise [15]. There are several methods used for reducing the noise emitted from a valve. These include sound absorptive-wrapping of the pipes and the valve casings and the installation of silencers between a valve and the connected pipes. In addition, special low-noise valve designs with multiple pressure drop stages exist that generate less noise than standard designs. Sound radiating from flare tip to surroundings
Sound radiating from valve to surroundings
Pressure relief or steam valve
High pressure side
Low pressure side
Figure 8.27 Illustration showing noise radiating from a valve to the surroundings.
© 2011 by Taylor and Francis Group, LLC
8.4 Noise Abatement Techniques There are three places noise can be reduced: at the source, in the path between the source and personnel, and on the personnel [16]. The ideal place to stop noise is at the source. There are several techniques used in the flare and burner industry to reduce the noise at the source, however, these techniques are limited. Ear protection can reduce noise relative to the personnel using it, unfortunately a plant operator cannot ask a surrounding community or workers within a nearby office building to wear ear protection when the noise levels become a problem. The most common method for reducing noise is in the path between the source and personnel using silencers, plenums, and mufflers. The purpose of this section is to discuss the most common and effective noise abatement techniques utilized in the flare and burner industry. 8.4.1 Flare Noise Abatement Techniques The following individual sound sources can contribute to the overall noise emissions of flares:
1. Combustion process 2. Gas jet and flow noise of the released gases 3. Smoke suppression equipment • Jet and flow noise by steam and air injection • Combustion air fans 4. Pilot burners 5. Valves (gas and steam side) and connected piping
As previously discussed, the two main sources of noise emitted from industrial flares is combustion roar and gas jet noise. Inhibiting the rate at which the air and fuel streams mix can reduce the level of combustion roar; however, this noise abatement technique generally tends to reduce the smokeless performance, increase thermal radiation, and flame length. Reducing the mixing rate of the air and fuel stream in order to lower combustion roar levels usually cannot justify the accompanying sacrifices in the performance of a flare. In such cases, enclosed flares may provide one solution. Enclosed flares are designed to completely hide a flare flame in order to reduce noise and thermal radiation levels. The design of these flare systems typically consist of an insulated enclosure with a wind fence around the perimeter as shown in the photograph in Figure 8.28. These types of flares can substantially reduce noise emissions as compared to open elevated flares.
203
Sound pressure level (dB)
Combustion Noise
Enclosed flare helps reduce noise
100 95 90 85 80 75 70 100
Without muffler With muffler 1000
10,000
100,000
Frequency (Hz) Wall at bottom of enclosure helps reduce noise
Figure 8.30 Steam jet noise emitted from a steam-assisted flare for conditions with and without a muffler.
110 105
Figure 8.28 Photograph of two enclosed flares.
Noise level (dB)
100 95 90 85 80 75 70 10
No water injection, 116 dB 114 dBA Optimum water injection, 113 dB, 101 dBA 100
1000 Frequency (Hz)
10,000
100,000
Figure 8.31 Plot of full-scale flare testing showing the noise spectrum from a high-pressure flare operating with and without water injection. Steam nozzle
Muffler
Figure 8.29 Illustration showing a steam-assisted flare with a muffler located around the lower steam nozzles.
There are several abatement techniques commonly used to reduce the gas jet noise emitted from flares. Several of these techniques include mufflers, water injection, and modifications to the nozzle geometry. Mufflers are most commonly used on steam-assisted flares to abate the high-pressure steam jet noise as shown in the illustration in Figure 8.29. In most flare systems, steam is supplied to nozzles at a pressure of 100–150 psig (7–10 barg). These highpressure steam jets produce high frequency mixing and shock-associated noise. A number of flare muffler styles have been used in the industry with varying degrees of
© 2011 by Taylor and Francis Group, LLC
noise abatement performance. Many of these mufflers are designed with a fiber material placed on the inside with a thickness of several inches. Usually mufflers do a good job of absorbing the high frequency steam jet noise as demonstrated by the data in Figure 8.30. This plot shows the noise spectrum emitted from a steam-assisted flare operating with and without a muffler on the lower steam jets. Clearly the data shows that mufflers are more efficient at absorbing the higher noise frequencies than the lower ones. In high-pressure flaring applications gas jet noise can be the major source of noise. In the year 2000, John Zink Company developed a unique method for reducing gas jet noise using water injection [17]. This method injects water into the waste gas stream near the nozzle exit. The water injection method substantially reduces the shockassociated noise as shown in Figure 8.31. This plot shows the noise spectrum emitted from a John Zink high-pressure flare operating with and without water injection. Schlieren photography shows that water injection does not eliminate the downstream shock cell structure but does appear to alter its appearance. This suggests that water injection suppresses the feedback mechanism responsible for growth of the gas jet instability that leads to screech tones.
204
Industrial Combustion Testing
Gas jet noise reduction using water injection is more pronounced when flaring high molecular weight gases as compared to low molecular weight gases at the same operating pressure. Test data and computer modeling prove that high molecular weight gases are more dominated by screech tone noise than low molecular weight gases operating at the same pressure. Which explains why gas jet noise reduction using water injection is more pronounced when flaring high molecular weight gases. It is very common in the flare industry to design a flare using several small diameter nozzles in order to reduce the A-weighted gas jet noise level. Usually gas jet noise emitted from high-pressure flares peak at a frequency between approximately 2000 and 16,000 Hz. The peak frequency is a function of several variables but is most affected by the diameter of the nozzle. For example, a one-inch gas jet nozzle will peak at a frequency between 2000 and 4000 Hz, whereas a quarter-inch gas jet nozzle will peak between 8000 and 16,000 Hz. To the human ear, a group of several smaller diameter gas jet nozzles will appear quieter than a single larger nozzle operating at the same pressure and mass flow rate. The primary reason is because the group of smaller nozzles will peak at a higher frequency, where the human ear is less sensitive. Designing a flare with many small diameter nozzles is not always practical or economical to build. Some large capacity flare designs would require several thousand nozzles to substantially reduce the gas jet noise. The analogue principle can also be applied to the steam injector nozzles in steam-assisted flares. Figure 8.32 From measurements
140
8.4.2 Burner Noise Abatement Techniques Typical sound sources contributing to the overall noise emissions of burners used in industrial heaters and furnaces are:
Lw A, dBA
100
63
125
250
500
1k
2k
4k
8k
Frequency, Hz Injector-system Original Optimized
Figure 8.32 Example for the acoustic optimization of a steam injector for a steam-assisted flare at the John Zink Company test facilities in Tulsa. A-weighted sound power level LwA for identical steam pressure and steam mass flow rate.
© 2011 by Taylor and Francis Group, LLC
1. Combustion process 2. Gas jet noise of the fuel gas at the burner 3. Pilot burners 4. Combustion air fan 5. Steam or air injection for atomization 6. Control valves and connected pipes
The most important of the above sound sources, typically, are the combustion roar from the combustion process itself, which resides in the frequency range of approximately 100 to 1000 Hz, and the gas jet noise, which typically ranges between 4000 and 16,000 Hz. The mid-to-high frequency noise is the most annoying and damaging to the ear. Several techniques have been used to suppress the noise emitted within the mid-to-high frequencies. Four common techniques used to reduce noise in industrial burners are the following:
120
80
shows an example for the acoustic optimization of a steam injector that was performed at the John Zink Company test facilities in Tulsa: By an optimized nozzle design, the overall noise emission from the injector could be reduced by more than 10 dB.
1. Sound insulation in the burner plenum 2. Mufflers at air inlets of natural draft burners 3. Acoustically optimized furnace wall construction 4. Acoustical treatment of the air ducts in forced draft burners
Figure 8.33 shows a plot of the SPL versus frequency for a burner operating with and without muffler. Clearly, without the muffler, the noise level is higher especially in the higher frequency region. For burner mufflers, not only their size but also an optimized arrangement can be crucial to obtain the best performance. Figure 8.34 shows a simple example of a muffler for the air inlet of a natural draft burner attached to the intake in two different ways. Due to the direct ray path from the burner plenum to the outside in the arrangement on the right in Figure 8.34, the effect of the muffler may be significantly less than in the arrangement on the left side even at the same muffler size. Where individual mufflers for the numerous burners in a furnace are a problem, for example, due to space limitations, an arrangement as shown in Figure 8.35,
205
Combustion Noise
80 75
Unmuffled A-weighted burner noise Furnace
70 65
Floor burners
60 55
Silencer
50
16,000
8000
4000
2000
1000
500
250
125
31.5
Frequency, Hz
Figure 8.33 Sound-pressure level versus frequency for a burner with and without a muffler. Often unfavorable
Better
Air
Plenum chamber
Muffled A-weighted burner noise
45 40
Silencer
Air
63
Sound pressure level, dBA
85
Air
Air
Figure 8.35 Schematic representation of a common plenum chamber for floor burners in a furnace with mufflers at the common air intakes. From measurements
110 Valve noise
Gas nozzle
Silencer
LwA, dBA
100
90 Figure 8.34 Example for the effect of burner air inlet muffler arrangements. Direct noise ray path in the arrangement on the right side leads to reduced efficiency of the muffler compared to the arrangement on the left.
with a common plenum chamber for groups of burners, can be an advantageous solution. Properly designed plenum chambers can bring about a reduction of the burner noise emissions in addition to the reduction achieved by the mufflers. 8.4.3 Valve and Piping Noise Abatement Techniques Valve and piping noise abatements include sound absorptive wrapping of the pipes and valve casings, installation of silencers between the valves and the connecting pipes, and the use of low-noise valve designs with multiple pressure stages. Acoustical pipe lagging is similar to thermal pipe insulation. All acoustical pipe lagging also provides excellent thermal insulation, but many thermal insulations provide poor noise control. Rigid insulations for cold service (such as foam glass installed on smaller diameter pipes) can actually aggravate the noise situation by conducting the noise to the outer surface easily. Although acoustical energy radiated per unit area of insulated and jacketed pipe is less than for the same noninsulated pipe, the surface area of an insulated and jacketed pipe
© 2011 by Taylor and Francis Group, LLC
80
63
125
250
500 1k Frequency, Hz
2k
4k
8k
Valve + piping With noise control No noise control Figure 8.36 Example for the noise emissions of a steam control valve of a steamassisted elevated flare. A-weighted sound power level LwA as in Figure 8.32 of the flare with standard control valve and uninsulated piping and with low noise valve and acoustically insulated steam pipes.
is greater. The product of these two factors can cause larger diameter jacketed pipe to radiate more noise than bare pipe [18]. Piping requiring acoustical treatment in a typical petrochemical plant is often in cold service. These lagging systems have to be both thermal and acoustical insulators. For that reason, fibrous insulation followed by an outer leaded aluminum jacket is commonly used. Sometimes, very noisy pipes need a layer of impregnated vinyl sandwiched between layers of fibrous insulation, called a Septum system [18]. Figure 8.36 shows an example for the noise emissions of a steam control valve located at the base of a
206
Industrial Combustion Testing
s team-assisted elevated flare. In the original configuration, a standard butterfly valve has been used and neither the valve itself nor the connected steam piping have been insulated. It is seen that the noise emissions of the flare (at very low waste gas flow rates) are clearly dominated by the valve noise with its peak at 4000 Hz. After installation of a low-noise valve and acoustical insulation for the steam pipes, the noise emissions at frequencies of 1000 Hz and above—where valve and piping noise, typically, is most prominent—are significantly reduced. Noise emissions at 500 Hz and below remain unchanged as these are dominated by the combustion noise coming from the flare tip. 8.4.4 Fan Noise Abatement Techniques Fan noise can usually be addressed similar to valve and piping noise:
1. Silencers can be installed at the suction and pressure sides of the fan particularly for fans communicating to the atmosphere on either the suction or the pressure side and thereby cut down on noise coming out of these portals. 2. Acoustically enclose the fan casing to address noise radiated from or transmitted through the casing surface. 3. Acoustically isolate the ductwork leading to and from a fan.
At the design stage one may consider the use of low noise motors (85 dBA or less), utilizing impellers with more blades and reduced tip speed, and so on.
Furnace wall
8.5.1 Multiple Burner Interaction A burner manufacturer will typically guarantee a burner noise level at a location 3 feet (1 m) directly in front of the muffler. When several burners are installed in a furnace, however, the noise level 3 feet (1 m) from the burner may be higher than for a single burner due to the noise contribution from surrounding burners. The purpose of this section is to give an example that illustrates the noise level increase due to noise emitted from surrounding burners. As an illustration, assume a furnace with a simple burner configuration as illustrated in Figure 8.37. If burner B is operating alone and the noise level is 85 dB at location 2, how is the noise level determined at location 2 when all burners are operating? First, find the
© 2011 by Taylor and Francis Group, LLC
Burner
Burner
A
B
C
1A
1B 3 ft
1C 5 ft
2
Figure 8.37 Illustration used for burner noise example.
sound power level, Lw, emitted from each burner assuming that the noise is emitted at the muffler exit at points 1A, 1B, and 1C. Assume that the noise spreads over a uniform sphere from each of these points. The SPL can be calculated as follows:
1 − 10.5, Lw = Lp − 10 log 10 4 πr 2
(8.9)
where Lp is the sound pressure level and r is the distance from the source in feet. The noise level 3 feet (1 m) from burner B (location 2) is 85 dB when it is operating alone. From Equation 8.9 we find that the LwB = 95.03 dB. Assuming that all burners are operating at the same conditions we know that the PWL must be 95.03 dB for each one. The SPL contribution, Lp, can now be calculated at location 2 when burner A is operating alone by solving Equation 8.9 for Lp:
8.5 Analysis of Combustion Equipment Noise
Burner
1 + 10.5. Lp = Lw + 10 log 10 4 πr 2
(8.10)
For this case LwA = 95.03 and r = (52 + 32)0.5 = 5.83 feet. Substituting these values into Equation 8.10 gives LpA = 79.2 dB. This is the SPL contribution emitted from burner A measured at location 2. Since the distance from burner C to location 2 is the same distance we know that the SPL contribution from burner C at location 2, LpC, is also 79.2 dB. The total SPL at location 2 can be determined by adding the SPL contribution from each burner (79.2 dB + 79.2 dB + 85 dB). The SPLs can be added by using the following equation: Lptotal = 10 log 10 ( 10LpA /10 + 10LpB /10 + 10LpC/10 ) = 86.8 dB. (8.11) For this example, the noise level will be approximately 1.6 dB higher when all the burners are operating than if burner B is operating alone.
207
Combustion Noise
8.5.2 High Pressure Flare
Table 8.9
Figure 8.38 is a plot showing the SPL spectrum of a high pressure flaring event burning Tulsa natural gas at a flow rate of approximately 23,000 lb/hr (10,400 kg/hr) in a 3.5 inch (8.9 cm) diameter flare tip (corresponding to an upstream pressure of approximately 22.5 psig or 1.53 barg). The symbols and the lines represent the noise spectrum gathered using a real-time, sound-level meter and mathematical modeling results, respectively. The SPL spectrum consists of two major peaks; a low frequency peak that corresponds to the combustion roar and a high frequency peak that corresponds to the gas jet noise. The intermediate peak is a result of piping and valve noise. Notice that the combustion roar peaks at a frequency of approximately 63 Hz, which is typical for a stable burning open flare. The mathematical modeling results of the combustion roar are determined using the correction values specified by Leite as previously discussed [6]. Figure 8.39 is a plot showing the noise contributions separately based on the mathematical model. Notice that the gas jet mixing noise is a broadband frequency spectrum while the screech noise occurs over a fairly narrow bandwidth.
The Overall Sound Pressure Level (OASPL) Determined Experimentally and Using the Mathematical Model
Noise level (dB)
105 100 90 85 80
10
100
10000
1000
100000
Frequency (Hz)
Figure 8.38 Sound pressure level spectrum for high pressure flaring. 110 105 Noise level (dB)
105.1 105.3
113.0 97.4
114.3 108.6
− −
− −
113.7 109.2
Total
Noise spectrum at a distance 3000 ft With atmospheric attenuation @ 3000 ft 68 F, 90% relative humidity 86 dB 80 dBA 80 dB 74 dBA
8000
2000
500
79 dB 64 dBA 125
90 80 70 60 50 40 30
Frequency, Hz Figure 8.40 (See color insert following page 424.) Effect of distance on flare noise.
95
75 70
105.7 105.2
Combustion Roar
Noise spectrum at a distance 1500 ft
31.5
Experiment without water injection, 115 dB, 109 dBA Model without water injection, 114 dB, 110 dBA
115 110
Screech Noise
Experiment dB − dBA −
Sound power level, dB
120
Model dB dBA
Jet Mixing Noise
100 95 90 85
eec
Co
80
h to
Jet mix
nes
ing
mb ro ustio ar n
75 70
Scr
Total
10
100
1000 Frequency (Hz)
10,000
100,000
Figure 8.39 Plot showing the noise contributions separately based on the mathematical model.
© 2011 by Taylor and Francis Group, LLC
The screech noise would not exist if the flare operated below the critical gas pressure. Below the critical gas pressure shock waves do not form, which causes screech noise. The summation of the combustion roar, gas jet mixing noise, and screech noise provides the total SPL prediction emitted from the flare. The overall SPL determined experimentally and calculated using the mathematical model is summarized in Table 8.9. Notice that in this particular example the overall SPL, on a dBA-scale, is dominated by the gas jet noise. If this 3.5 inch (8.9 cm) diameter flare were designed with several smaller diameter ports, having the same total exit area, then the gas jet noise would shift to higher frequencies. If the diameter of these ports were small enough to substantially shift the frequency of the gas jet noise then the combustion noise would dominate on the dBA scale. 8.5.3 Atmospheric Attenuation Example Figure 8.40 shows noise measurements emitted from a flare. Notice at a distance of 1500 feet (460 m)
208
from the flare, the noise level peaks at about 80 dBA, while at 3000 feet (910 m) the peak reduces to about 74 dBA. When the atmospheric attenuation is taken into account, depending on the ambient temperature and humidity level at the time of measurement (see Section 8.1.3.2), we see further reduction in noise levels. It is important to note that the contribution in each case is significant. Given the particular atmospheric conditions in this example, the attenuation has created a significant difference. The 10 dB attenuation, from 74 dBA to 64 dBA, amounts to the sound intensity reduction equal to one-eighth of its i ntensity at 3000 feet (910 m) without atmospheric attenuation. Hence, it should be noted that measurements may vary significantly on different days for the same equipment, if the atmospheric conditions are significantly different.
Glossary Absorption: conversion of sound energy into another form of energy—usually heat—when passing through an acoustical medium. Absorption coefficient: ratio of sound absorbing effectiveness, at a specific frequency, of a unit area of acoustical absorbent to a unit area of perfectly absorptive material. Acoustics: science of the production, control, transmission, reception, and effects of sound and of the phenomenon of hearing. Ambient noise: all-pervasive noise associated with a given environment. Anechoic room: room whose boundaries effectively absorb all incident sound over the frequency range of interest, thereby creating essentially free field conditions. Audibility threshold: sound pressures level, for a specified frequency, at which persons with normal hearing begin to respond. Background noise: ambient noise level above which signals must be presented or noise sources measured. Decibel scale: linear numbering scale used to define a logarithmic amplitude scale, thereby compressing a wide range of amplitude values to a small set of numbers. Diffraction: scattering of radiation at an object smaller than one wavelength and the subsequent interference of the scattered wavefronts. Diffuse field: sound field in which the sound pressure level is the same everywhere and the flow of energy is equally probable in all directions.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
Diffuse sound: sound that is completely random in phase, sound which appears to have no single source. Directivity factor: ratio of the mean-square pressure (or intensity) on the axis of a transducer at a certain distance to the mean-square pressure (or intensity) that a spherical source radiating the same power would produce at that point. Far field: distribution of acoustic energy at a very much greater distance from a source than the linear dimensions of the source itself; the region of acoustic radiation used to the source and in which the sound waves can be considered planar. See also diffraction. Free field: an environment in which there are no reflective surfaces within the frequency region of interest. Hearing loss: an increase in the threshold of audibility due to disease, injury, age, or exposure to intense noise. Hertz: unit of frequency measurement, representing cycles per second. Infrasound: sound at frequencies below the audible range (i.e., below about 16 Hz). Isolation: resistance to the transmission of sound by materials and structures. Loudness: subjective impression of the intensity of a sound. Masking: process by which threshold of audibility of one sound is raised by the presence of another (masking) sound. Near field: that part of a sound field, usually within about two wavelengths from a noise source, where there is no simple relationship between sound level and distance. Noise emission level: dB(A) level measured at a specified distance and direction from a noise source, in an open environment, above a specified type of surface. Generally follows the recommendation of a national or industry standard. Noise reduction coefficient (NRC): arithmetic average of the sound absorption coefficients of a material at 250, 500, 1000, and 2000 Hz. Phone: loudness level of a sound, numerically equal to the sound pressure level of a 1 kHz free progressive wave, which is judged by reliable listeners to be as loud as the unknown sound. Pink noise: broadband noise whose energy content is inversely proportional to frequency (-3dB per octave or –10 dB per decade). Power spectrum level: level of the power in a band one hertz wide referred to a given reference power (also: power spectral density). Reverberation: persistence of sound in an enclosure after a sound source has been stopped. Reverberation
209
Combustion Noise
time is the time (in seconds) required for sound pressure at a specific frequency to decay 60 dB after a sound source is stopped. Root mean square (RMS): the square root of the arithmetic average of a set of squared instantaneous values. Sabine: measure of sound absorption of a surface. One metric sabine is equivalent to 1 m2 of perfectly absorptive surface. Sound: energy that is transmitted by pressure waves in air or other materials and is the objective cause of the sensation of hearing. Commonly called noise if it is unwanted. Sound intensity: rate of sound energy transmission per unit area in a specified direction. Sound level: level of sound measured with a sound level meter and one of its weighting networks. When A-weighting is used, the sound level is given in dBA. Sound level meter: an electronic instrument for measuring RMS of sound in accordance with an accepted national or international standard. Sound power: total sound energy radiated by a source per unit time. Sound power level: fundamental measure of sound power defined as Lw = 10 log P/P0 dB, where P is the RMS value of sound power in watts and P0 is 1 pwatt. Sound pressure: dynamic variation in atmospheric pressure. The pressure at a point in space minus the static pressure at that point. Sound pressure level: fundamental measure of sound pressure defined as Lp = 20 log p/po dB, where p is the RMS value (unless otherwise stated) of sound pressure in pascals and po is 20 μPa. Sound transmission loss: ratio of the sound energy emitted by an acoustical material or structure to the energy incident upon the opposite side. Standing wave: a periodic wave having a fixed distribution in space that is the result of interference of progressive waves of the same frequency and kind. Characterized by the existence of maxima and minima amplitudes that are fixed in space. Thermoacoustic efficiency: a value used to characterize the amount of combustion noise emitted from a flame. Defined as the ratio of the acoustical power emitted from the flame to the total heat release rate of the flame. Ultrasound: sound at frequencies above the audible range (i.e., above about 20 kHz). Wavelength: distance measured perpendicular to the wavefront in the direction of propagation between two successive points in the wave,
© 2011 by Taylor and Francis Group, LLC
which are separated by one period. Equals the ratio of the speed of sound in the medium to the fundamental frequency. Weighting network: an electronic filter in a sound level meter, which approximates under defined conditions the frequency response of the human ear. The A-weighting network is most commonly used. White noise: broadband noise having constant energy per unit of frequency.
References
1. Peterson, A. P. G. Handbook of Noise Measurement. 9th ed. West Concord, MA: GenRad, 1980. 2. ISO 3744: “Acoustics - Determination of Sound Power Levels of Noise Sources Using Sound Pressure— Engineering Method in an Essentially Free Field Over a Reflecting Plane.” May 1, 1994. 3. Daiminger, W., Fritz, K. R., Schorer, E., and Stüber, B. Ullmann’s Encyclopedia of Industrial Chemistry B7, pp. 384–401. Weinheim, Germany: VCH, 1995. 4. API Recommended Practice 531M. Measurement of Noise from Fired Process Heaters. Reaffirmed ed. Washington, DC: American Petroleum Institute, August 1995. 5. Thumann, A., and Miller, R. K. Secrets of Noise Control. Lilburn, GA: Fairmont Press, 1974. 6. Leite, O. C. “Predict Flare Noise and Spectrum.” Hydrocarbon Processing 68 (1988): 55. 7. Müller-BBM GmbH. “Noise Emissions of Different Flare Systems—Field Measurements Taken in Various Refineries and Petrochemical Plants.” Proprietary data, 1900–2004. 8. Verein Deutscher Ingenieure. “VDI 3732—Characteristic Noise Emission Values of Technical Sound Sources— Flares.” VDI guideline, 1999. 9. Bussman, W., and White, J. “Steam-Assisted Flare Testing,” John Zink Co. Internal Report, September 1996. 10. Putnam, A. A. “Combustion Noise in the Handheld Industry.” Battelle, Columbus Laboratories. 11. Beranek, L. L., and Ve’r, I. L. Noise and Vibration Control Engineering. New York: John Wiley and Sons, Inc., 1992. 12. Lighthill, M. J. “On Sound Generated Aerodynamically.” Proceedings of the Royal Society, A 211 (1952): 564–87, A 222 (1954): 1–32. 13. Ffowcs Williams, J. E. “The Noise from Turbulence Convected at High Speeds.” Philosophical Transactions of the Royal Society of London, Series A 225 (1963): 469–503. 14. Shen, H., and Tam, C. K. W. “Numerical Simulation of the Generation of Axisymmetric Mode Jet Screech Tones,” AIAA Journal 36, no. 10 (October 1998): 1801 15. Beranek, L. L. Noise and Vibration Control. New York: McGraw Hill Book Co., 1971.
210
16. Allied Witan Co., Noise Facts and Control, 1976. 17. Bussman, W. R., and Knott, D. “Unique Concept for Noise and Radiation Reduction in High-Pressure Flaring,” OTC Conference, Houston Texas, 2000. 18. Frank, L. D., and Dembicki, D. R. “Lower Plant Noise with Lagging.” Hydrocarbon Processing 71, no. 8 (1992): 83–85.
Bibliography Alberta Energy and Utilities Board. Calgary, Alberta, 1998. American Petroleum Institute, 50 (1972): 125–46. Brief, R. S., and Confer, R. G. “Interpreting Noise Dosimeter Results Based on Different Noise Standards.” American Industrial Hygiene Journal 36, no. 9 (1975): 677–82. Crow, S. C., and Champagne, F. H. “Orderly Structure in Jet Turbulence.” Journal of Fluid Mechanics 48, no. 3 (1971): 547–91. Diserens, A. H. “Personal Noise Dosimetry in Refinery and Chemical Plants.” Journal of Occupational Medicine 16, no. 4 (1974): 255–57. Gharabegian, A., and Peat, J. E. “Saudi Petrochemical Plant Noise Control.” Journal of Environmental Engineering 112, no. 6 (1986): 1026–40. HFP Acoustical Consultants. “Effect of Flow Parameters on Flare Stack Generator Noise.” Proceedings of the Spring Environmental Noise Conference: Innovations in Noise Control for the Energy Industry, Alberta, April 19–22, 1998. International Electrochemical Commission, IEC Standard, Publication 651, Sound Level Meters, 1979. ISO 1683 (E), Acoustics-Preferred Reference Quantities for Acoustic Levels, 1983. ISO 532 (E), Acoustics Method for Calculating Loudness Level, 1975. ISO 1996-1 (E), Acoustics: Description and Measurement of Environmental Noise. ISO 3744 (E), Acoustics: Determination of Sound Power Levels of Noise Sources.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
Engineering Methods for Free Field Conditions Over a Reflecting Plane, 1081. ISO/DIS 8297 Acoustics: Determination of Sound Power Levels of Multi-Source Industrial Plants for the Evaluation of Sound Pressure Levels in the Environment-Engineering Method, 1988. ISO 9614-1 (E), Determination of Sound Power Levels of Noise Sources Using Sound Intensity, Part -I: Measurement at Discrete Points, Part II: Measurement at Planned Points, 1993. Lang, W. W., ed. “A Commentary on Noise Dosimetry and Standards.” Proceedings of Noise Congress-75, Gaithersburg, MD, September 15–17, 1975. Maling, Jr., J. C., ed. “Start-Up ilencers for a Petrochemical Complex,” Proceedings of International Conference of Noise Control Engineering, December 3–5, 1984. Powell, A. “On the Noise Emanating from a Two Dimensional Jet Above the Critical Pressure.” The Aeronautical Quarterly 4 (1953): 103. Putnam, A. A. “Combustion Noise in the Hand Glass Industry.” 10th Annual Symposium on the Reduction Cost in the Hand Operated Glass Plants, 1979. Reed, R. Furnace Operations. Houston: Gulf Publishing Co., 1981. Ribner, H. S. “Perspectives on Jet Noise.” AIAA Journal 19, no. 12 (1981): 1513. Roberts, J. P. PhD thesis, London University, 1971. Seebold, G., and Hersh, A. S. “Control Flare Steam Noise.” Hydrocarbon Processing 51 (1971): 140. Selle, G. K. “Steam-Assisted Flare Eliminates Environmental Concerns of Smoke and Noise.” Hydrocarbon Processing, 73, no. 12, (1994): 77–78. Shivashankara, B. N., Strahle, W. C., and Henkley, J. C. “Combustion Noise Radiation by Open Turbulent Flames.” Presented as Paper 73-1025 at AIAA AeroAcoustics Conference, Seattle, WA, 1973. Straitz, J. F. “Improve Flare Design,” Hydrocarbon Processing 73, no. 10 (1994): 61–66. Thomas, A., and Williams, G. T. “Flame Noise: Sound Emis sion from Spark-Ignited Bubbles of Combustion Gas.” Proceedings of the Royal Society, A 294 (1966): 449. Zwicker, E, and Fastl, H. Psychoacoustics-Facts and Models. Berlin: Springer Verlag, 1990.
9 Flame Impingement Measurements Charles E. Baukal, Jr. Contents 9.1 Introduction.....................................................................................................................................................................211 9.2 Experimental Conditions...............................................................................................................................................214 9.2.1 Configurations.....................................................................................................................................................214 9.2.1.1 Flame Normal to a Cylinder in Crossflow...................................................................................... 215 9.2.1.2 Flame Normal to a Hemispherically Nosed Cylinder................................................................... 215 9.2.1.3 Flame Normal to a Plane Surface..................................................................................................... 215 9.2.1.4 Flame Parallel to a Plane Surface...................................................................................................... 215 9.2.2 Operating Conditions....................................................................................................................................... 215 9.2.2.1 Oxidizers.............................................................................................................................................. 217 9.2.2.2 Fuels...................................................................................................................................................... 218 9.2.2.3 Equivalence Ratios.............................................................................................................................. 222 9.2.2.4 Firing Rates.......................................................................................................................................... 222 9.2.2.5 Reynolds Number............................................................................................................................... 224 9.2.2.6 Burner Designs.................................................................................................................................... 224 9.2.2.7 Number of Burners............................................................................................................................. 225 9.2.2.8 Burner Nozzle Diameter.................................................................................................................... 225 9.2.2.9 Location................................................................................................................................................ 225 9.2.3 Stagnation Targets............................................................................................................................................. 228 9.2.3.1 Size........................................................................................................................................................ 228 9.2.3.2 Target Materials................................................................................................................................... 231 9.2.3.3 Surface Preparation............................................................................................................................. 231 9.2.3.4 Surface Temperatures......................................................................................................................... 233 9.2.4 Measurements.................................................................................................................................................... 233 References................................................................................................................................................................................. 235
9.1 Introduction Impinging flame jets have been extensively studied because of their importance in a wide range of applications [1]. Figure 9.1 shows a flame impinging normal to a water-cooled plate [2]. Early experiments were used to simulate the extremely high heat fluxes encountered by space vehicles reentering Earth’s atmosphere. These heat levels are caused by the hypersonic flow impact velocities that ionize the highly shocked atmospheric gases [3]. Subsequent studies have since investigated the heat fluxes attainable using high intensity combustion, with pure oxygen instead of air, to increase metal heating and melting rates [4]. High intensity impinging jet flames have been used in recent years to produce
synthetic diamond coatings, by chemical vapor deposition [5]. Supersonic velocity, high intensity flames have been used in a process known as thermal spallation. In this process the impinging jet bores through rock by causing it to fragment, due to the large thermal stresses arising from the high heat fluxes on a cold surface [6,7]. This may be more rapid and economical, than traditional mechanical rock drilling, depending on the rock type. High velocity flames impinging on structural elements have been used to simulate largescale fires, caused by ruptured piping in the chemical process industry [8]. Low intensity impinging flame jets have been used in safety research (e.g., Johnson et al. [9]), to quantify the heating rate caused by buoyant fires impinging on walls and ceilings [10]. They have also been used to study heating in domestic gas-fired
211 © 2011 by Taylor and Francis Group, LLC
212
Industrial Combustion Testing
stoves [11]. Direct flame impingement is used to heat metals in the production of strip, bars, blooms, and ingots [12–15]. Figure 9.2 shows an example of impinging flames used to heat the edge of metal strip. Flame impingement is used to harden metal parts [16] and modify the surface of polymer films [17]. Eibeck et al. (1993) studied the impact of pulse impinging flames on the heat transfer to a target [18]. There are several general reviews of flame impingement [19–23]. Most of the previous research has concer
Coolant out
Coolant in
Cold side Target r
Hot side dj
tw
te lj
ned air/fuel combustion. In these lower intensity flames, the predominant mechanism is forced convection. Much work has also concerned fuels combusted with pure oxygen. These high intensity flames produce significant amounts of dissociated species (e.g., H, O, OH, etc.) and uncombusted fuel (e.g., CO, H2, etc.). These reactive gases then impact on a relatively low temperature target surface. As these species cool, they exothermically combine into products such as CO2 and H2O, which are more thermodynamically stable at lower temperatures. This chemical heat release is sometimes referred to as “convection vivre,” or live convection [24]. This mechanism may be comparable in magnitude to the forced convection heat transfer at the surface [25]. Such high intensity flames require much more complicated analysis. Several experiments have used fuels combusted with an oxidizer having an oxygen content between that of air and pure oxygen. This process is sometimes referred to as oxygen-enriched air combustion. These medium intensity flames have received much less attention. They are currently not used in industrial applications. Six heat transfer mechanisms have been identified in previous flame impingement studies:
Burner
1. convection (forced and natural) 2. conduction (steady-state and transient) 3. radiation (surface, luminous, and nonluminous)
dn Figure 9.1 Flame impingement normal to a cooled target. (From Baukal, C. E., Farmer, L. K., Gebhart, B., and Chan, I., “Heat Transfer Mechanisms in Flame Impingement Heating,” In 1995 International Gas Research Conference, Vol. II, edited by D. A. Dolenc, 2277–87, Rockville, MD: Government Institutes, 1996.)
Burner
Boiling Rad. Conv. Target
Conden TCHR Conduction
Convection + Boiling
H2O in Figure 9.2 Flame impingement used to heat the edge of metal strip.
© 2011 by Taylor and Francis Group, LLC
H2O out
Figure 9.3 Heat transfer mechanisms in flame impingement on a water-cooled target.
213
Flame Impingement Measurements
(b)
Furnace Target
(a) Target
Target Target
Target
Figure 9.6 Radiation heat transfer mechanisms in flame impingement: (a) luminous flame, (b) nonluminous flame, and (c) surface radiation.
Figure 9.4 Convective heat transfer mechanisms in flame impingement: (a) forced, and (b) natural. (a)
(b) tflame
tflame
Time > 0 Flame
Coolant
tcoolant
Target
Flame
Time = 0
tinitial
Target
Figure 9.5 Conduction heat transfer mechanisms in flame impingement: (a) steady-state. and (b) transient.
4. thermochemical heat release (equilibrium, catalytic, and mixed) 5. water vapor condensation 6. boiling (internal and external) [2]
These are shown schematically for a water-cooled target in Figure 9.3. All of the mechanisms are not usually present simultaneously and depend on the specific problem.
© 2011 by Taylor and Francis Group, LLC
Figure 9.4 shows a schematic of forced and natural convection in flame impingement. In natural convection flame impingement, the flame is often far from the surface so that the buoyant hot combustion products blended with cooler ambient air are impinging on the target. Figure 9.5 shows a schematic of steady-state conduction for a water-cooled target and transient conduction for an uncooled target. In the water-cooled case, both the hot and cold sides of the target are approximately constant at steady-state conditions. In the uncooled target, there is a nonlinear temperature gradient through the target during transient heating. Figure 9.6 depicts radiation heat transfer in flame impingement for luminous flames, nonluminous flames, and surface radiation from hot furnace walls. The luminous flame contains particles that glow and radiate heat to the target. The nonluminous flame contains no radiating particles, so that only gaseous radiation from CO2 and H2O are commonly present. Figure 9.7 shows TCHR (thermochemical heat release) for catalytic, equilibrium, and mixed mechanisms for CO and CO2 as an example. Similar reactions occur for H2 and H2O, as well as other dissociated species. Equilibrium TCHR occurs in the gas phase between the burner and the target, outside the boundary layer, which has been greatly exaggerated. The CO combines with radicals to produce CO2. Catalytic TCHR occurs when the gases contact the surface, which catalytically promotes the reaction of CO to CO2 in the presence of radical species. Mixed TCHR is a combination of equilibrium and catalytic TCHR. Internal and external boiling are schematically shown in Figure 9.8. Internal boiling occurs on the cold side of a water-cooled target where the coolant is heated above its boiling point by contacting the hot target. External boiling can occur if an initially cold target that has
214
Industrial Combustion Testing
Flame
Burner
Equilibrium
Stagnation region Partially catalytic surface
Target
ttarget < dew point of gases
CO, CO2 Catalytic
Liquid H2O in CO + M = CO2 Mixed = Equil. + Catal.
Liquid H2O out
Figure 9.9 Water vapor condensation in flame impingement on a water-cooled target.
CO + Surface = CO2
9.2 Experimental Conditions
Target
Figure 9.7 Thermochemical heat release mechanisms in flame impingement.
Flame
Flame Target
Target
Liquid H2O in
Liquid + Vapor H2O out
ttarget > 100°C
Figure 9.8 Boiling heat transfer mechanisms in flame impingement: (a) internal, and (b) external.
Many important parameters arise in flame jet impingement processes [26]. The first and most important aspect is the overall geometric configuration. This includes the target shape and its orientation relative to the burner. The operating conditions strongly influence the heat transfer intensity. They also determine what mechanisms will be most important. These conditions include the oxidizer composition, the fuel composition, the equivalence ratio, and the Reynolds number at the nozzle exit. Other factors commonly have secondary influences on the heat transfer processes. These include the burner design and the position relative to the target. The characteristics of the target also influence the heat transfer. These include the dimensions, material composition, surface treatments or coatings, and the surface temperature. The above parameters are tabulated and discussed below. While some flame impingement experiments (e.g., Glumac and Goodwin [27] and Roy et al. [28]) have been conducted under vacuum conditions commonly encountered in chemical vapor deposition processes, all of the studies discussed here have been at approximately atmospheric pressure. Some studies (e.g., Malikov et al. [e.g. 12,29]) have been conducted inside a furnace. 9.2.1 Configurations
condensed water from the products of combustion of the flame on the target surface then heats above the boiling point of water. Water vapor condensation is shown in Figure 9.9. This can occur when the water vapor, in the products of combustion from flame, condenses onto a target whose temperature is below the dewpoint of the water vapor.
© 2011 by Taylor and Francis Group, LLC
This specifies the relative orientation of the target surface and the burner. This is usually the most important consideration for the designer or researcher, when reviewing previous experimental results. The four most common geometric configurations in flame jet experiments have been flames impinging (1) normal to a cylinder in crossflow, (2) normal to a hemi-nosed cylinder,
215
Flame Impingement Measurements
Burner
Target
analytic solutions (e.g., Sibulkin [38]) derived for aerospace applications, such as rockets and missiles. In this configuration, the heat flux has been measured only at the forward stagnation point. Most such studies have concerned laminar flames. 9.2.1.3 Flame Normal to a Plane Surface
Figure 9.10 Flame impinging normal to a cylinder in crossflow. (From Baukal, C. E., Heat Transfer from Flame Impingement Normal to a Plane Surface, Saarbrücken, Germany: VDM Verlag/Dr. Müller, 2009.)
(3) normal to a plane surface, and (4) parallel to a plane surface. Other configurations have been tested, including flames at an angle to a plane surface and flames normal to and around the circumference of a large cylindrical furnace [30]. These are not common geometries and little information is available on them, so they are not considered in detail here. 9.2.1.1 Flame Normal to a Cylinder in Crossflow In this configuration, shown in Figure 9.10, the cylinder axis is perpendicular to the burner axis. This configuration has been widely studied as shown in Table 9.1. It applies to processes such as heating round metal billets, heating cylinders for surface treatment of polymers [17], and fires impinging on pipes in chemical plants [8]. Some of the earliest studies investigated impingement on refractory cylinders [31,32]. A pipe was the most common target for this geometry (e.g., Anderson and Stresino [33]). In some studies, the local heat flux was measured at the forward stagnation point of the cylinder (e.g., Hustad, Jacobsen, and Sønju [8,34–36]; Hustad and Sønju [34]; Hustad, Røkke, and Sønju [35]; and Chander and Ray [36]). In one study [36], the angle of the cylinder with respect to the impinging flame was also varied. In many studies the pipe was cooled with water circulating through it (e.g., Davies [37]). The average heat flux, over the entire surface, was calculated from the sensible energy gain of the cooling water. 9.2.1.2 Flame Normal to a Hemispherically Nosed Cylinder For this geometry, shown in Figure 9.11, the cylinder axis is parallel to the burner axis. The flame impinges on the end of the cylinder, which is hemispherical. These tests have been very important in aerospace applications. This relatively uncommon geometry (see Table 9.2), for industrial application, is important because the heat transfer results may be directly compared to some of the
© 2011 by Taylor and Francis Group, LLC
This configuration, shown in Figure 9.12, has received the most attention (see Table 9.3), since it has been widely used in many industrial processes. It has also been used for the widest range of operating and target surface conditions. An example of a flame impinging normal to a plane surface is shown in Figure 9.13. Zhang and Bray [39] noted that the flame configuration for certain types of flames impinging on a flat plate can be very dependent on and sensitive to the burner nozzle exit velocity and the spacing between the burner and the plate. They identified the following five combustion modes: ring flame, conical flame, disc flame, envelope flame, and cool central core flame. Shorin and Pechurkin [40], Kremer, Buhr, and Haupt [41], and Dong, Leung, and Cheung [42] also investigated the effect of flame impingement at an angle between parallel and normal, onto plane surfaces. Only the results for flame impingement normal to a plane surface are included here. The reader is referred to the literature for further information on angled jets. 9.2.1.4 Flame Parallel to a Plane Surface This configuration, shown in Figure 9.14, has been the least studied (see Table 9.4). However, it is very important for flight applications. It simulates the heat transfer processes on airfoil surfaces. The studies by Giedt and coworkers [3,43] investigated flow across the top and bottom of a plane surface. Neither of these studies investigated the heat flux at the leading edge of the target. Beér and Chigier [44] studied impingement only on one side of a plane surface, which was the hearth of a furnace. The flame was inclined at 20° above the horizontal (see Figure 9.14b). Mohr, SeyedYagoobi, and Page [45]; Mohr, Seyed-Yagoobi, and Page [46–48]; and Wu, Seyed-Yagoobi, and Page [49] studied a special configuration for flames parallel to a plane surface that they called radial jet reattachment flames. This type of flame promises to give more uniform heating of the surface, compared to flames impinging normal to a surface. 9.2.2 Operating Conditions The operating conditions have been found to strongly influence the heat transfer intensity. The effects include the oxidizer and fuel composition, the flame
216
Industrial Combustion Testing
Table 9.1 Experimental Conditions for Flame Impingement Normal to Cylindrical Surfaces Oxidizer
Fuel
ϕ
qf (kW)
Air
CH4
1.25–1.67
0.71–1.2
0.8–1.2
1.2–10
b
tube
∞
37–150
∞
L
R
Reference
laminarb
1.1–4.5
0–0.7
16
2000–8000
3.1–11
0
tube
5–40
200–3600
30–200
0
60–1600
tube
5–40
26,000
80
0
0.8–1.3
0.29–0.76
tube
6.35–12.7
600–1300
1–5
0
1.25
3.5–9.4
18
not given
1.7–6.9
0
∞
34.3–740
tube
5–40
200–3600
20–200
0
∞
60–1600
tube
5–40
120, 150
0
C4H10
∞
60–1600
tube
5–40
54,000, b 330,000 b 66,000
130
0
CO H2
1.0, 1.19 1.0
0.51–0.75 2.1, 4.3
4 × 50 not given
laminara laminara
0.6–1.3
0.56–1.7
not given
not given
2.5–7.5 not given (c)
0 0
natural gas
0
1.05
not given
22.4, 41.0
12,600
0–4.5
0
1.0–1.05
250–600
slot annular port with precombustor premix, slotshaped torch flame retention head array of tubes
Anderson & Stresino, 1963 [33] Hargrave et al., 1987 [109] Hustad et al., 1991 [34,35] Hustad et al., 1992 [8] Chander & Ray, 2007 [36] Fells & Harker, 1968 [92] Hustad et al., 1991 [34,35] Hustad et al., 1992 [8] Hustad et al., 1992 [8] Kilham, 1949 [31] Jackson & Kilham, 1956 [32] Davies, 1979 [37]
4
0
town gas natural gas C2H2
lean-richa
7.1–28
0.5–1.3
0.56–1.7
2.34
1.3–4.1
C3H8
1.0, 1.43
77, 3.3
CO
1.0, 2.0
2.4
H2
1.0
6.1
1.0
670
0.5–1.16
1.7–2.8
1.0
12–16
C3H8
Air/O2 O2
natural gas town gas a b c
Burner partial premix
tunnel
tunnel premix, slotshaped torch partial premix tunnel, partial premix annular port with precombustor annular port with precombustor tunnel premix, slotshaped torch tunnel
dn (mm) 12.7–34.9
Ren
b
13, 26
not given
0.5
0
Hemeson et al., 1984 [97] Malikov et al., 2001 [12] Davies, 1965 [61]
not given
not given
(c)
0
Davies, 1979 [37]
1.2, 1.8
laminarb
4.3–7.5
0–5.4
9.5, 1.8
laminarb
0, 8.6
not given
laminara
not given
laminara
9.5
laminarb
not given not given 1.3
0–0.53 0–6.1 0
not given
not given
(c)
0
Anderson & Stresino, 1963 [33] Anderson & Stresino, 1963 [33] Jackson & Kilham, 1956 [32] Jackson & Kilham, 1956 [32] Anderson & Stresino, 1963 [33] Davies, 1979 [37]
13
not given
0.5
0
Davies, 1965 [61]
4
0 0–0.67
According to author (only qualitative). Calculated from data given in reference. Target located at point of maximum heat flux.
equivalence ratio (ϕ) and firing rate (qf), the Reynolds number at the nozzle exit (Ren), the burner type, the nozzle diameter (d n), and the location of the target with respect to the burner. Most studies were steadystate. Some earlier studies (e.g., Milson, and Chigier
© 2011 by Taylor and Francis Group, LLC
[24]) used a transient method to determine the heat flux. Tuttle and colleagues [50,51] made high resolution temporal measurements of air/methane flames impinging on a water-cooled polished aluminum flat plate.
217
Flame Impingement Measurements
9.2.2.1 Oxidizers Burner
The most important variable, after the physical configuration, is the oxidizer composition. The oxygen mole fraction in the oxidizer, Ω, has a very large influence on heat transfer intensity. Almost all of the previous studies used either air (Ω = 0.21) or pure oxygen (Ω = 1.0) as the oxidizer. This affects both the flame temperature and the amount of dissociation in the combustion products. As an example, the adiabatic flame temperature
Target
Figure 9.11 Flame impinging normal to a hemi-nosed cylinder. (From Baukal, C. E., Heat Transfer from Flame Impingement Normal to a Plane Surface, Saarbrücken, Germany: VDM Verlag/Dr. Müller, 2009.)
Table 9.2
Experimental Conditions for Flame Impingement Normal to Hemi-Nosed Cylinders Oxidizer
Fuel
ϕ
qf (kW)
Air
CH4
1.0
1–8
0.943, 1.171 0.8–1.2
not given b 1.2–10
CO
0.7–1.3
H2
0.5–0.63 1.0
Air/O2
natural gas #2 natural gas CH4
O2
natural gas CH4
1.05 0.62–1.14 0.84–1.14 0.95 0.95–1.31 1.0 0.83–1.70 0.83–1.70
C2H4
1.0
C3H8
1.45–1.83 1.0
a b
CO
1.0
H2
1.0
natural gas
0.95
dn (mm)
Ren
L
tube
16
1.6–3.8
0
converging nozzle tube
12
4.6–11
0
Hargrave & Kilham, 1984 [79] Fairweather et al., 1984 [54]
1.9–15
0
Hargrave et al., 1987 [109]
not given
porous bronze
51
1200– 10,000 laminar & turbulenta 2000– 12,000 laminarb
0.02–0.3
0
not given not given not given 3.2 not given 210, b 270 not given not given 3.8 not given not given not given not given not given not given b 210
porous bronze
51
laminarb
0.02–0.2
0
bundle of tubes
not given 22.4, 41.0
50–500
(1.3 cm)
0
Kilham & Dunham, 1967 [101] (cf. Dunham, 1963 [102]) Cookson & Kilham, 1963 [73] (cf. Cookson, 1963 [72]) Conolly & Davies, 1972 [4]
12,600
0–4.5
0
Hemeson et al., 1984 [97]
b
According to author (only qualitative). Calculated from data given in reference.
© 2011 by Taylor and Francis Group, LLC
Burner
flame retention head bundle of tubes bundle of tubes
16
R
Reference
10 10
laminarb laminara
0.3–1.4 0.3–1.4
0 0
Nawaz, 1973 [53] Fairweather et al., 1984 [54]
multiannular with swirl glass torch
not given 10
not given
0
Ivernel & Vernotte, 1979 [55]
0
bundle of tubes
not given 10 10
50–500
(40–140 cm) not given (1.3 cm)
0
Kilham & Purvis, 1971 [76] (cf. Purvis, 1974 [77]) Conolly & Davies, 1972 [4]
laminarb laminara
0.3–1.4 0.3–1.4
0 0
Nawaz, 1973 [53] Fairweather et al., 1984 [54]
not given 10
50–500
(1.3 cm)
0
Conolly & Davies, 1972 [4]
laminara
0
not given not given not given not given
50–500
not given (1.3 cm)
0
Kilham & Purvis, 1971 [76] (cf. Purvis, 1974 [77]) Conolly & Davies, 1972 [4]
50–500
(1.3 cm)
0
Conolly & Davies, 1972 [4]
50–500
(1.3 cm)
0
Conolly & Davies, 1972 [4]
not given
(40–140 cm)
0
Ivernel & Vernotte, 1979 [55]
bundle of tubes bundle of tubes bundle of tubes glass torch bundle of tubes bundle of tubes bundle of tubes multiannular with swirl
laminara
218
Industrial Combustion Testing
tw r Target surface
dj lj
x Burner dn Figure 9.12 Flame impinging normal to a plane surface. (From Baukal, C. E., and Gebhart, B., “A Review of Empirical Flame Impingement Heat Transfer Correlations,” International Journal of Heat and Fluid Flow 17, no. 4 (1996): 386–96.)
for methane combusted stoichiometrically with air and with pure oxygen is 2220 K and 3054 K (3537°F and 5038°F), respectively. The products of combustion for a stoichiometric air/CH4 adiabatic flame contain essentially no unreacted fuel or dissociated species, except at very high flame temperatures approaching adiabatic conditions. However, the combustion products of an O2/CH4 flame contain nearly 23 vol.% unreacted fuel (CO and H2), and over 18 vol.% dissociated species (H, O, and OH). As these products cool down in the boundary layer along the target surface, they exothermically release heat. This heat release rate from TCHR may be greater in magnitude than that of the forced convection heat transfer mechanism. A few studies have used oxygen concentration levels between those of air and pure oxygen. Figure 9.15 shows natural gas flames impinging normal to a plane surface with different oxidizers. Beér and Chigier [44] injected a small, unspecified, amount of O2 into an air/ coke oven gas flame, to raise the temperature slightly. Vizioz and Lowes [52] studied oxygen-enriched air for Ω = 0.30. In related studies, Nawaz [53] and Fair weather, Kilham, and Nawaz [54] studied oxygenenriched air flames for Ω ranging from 0.46 to 0.61. Davies [37] studied flame impingement heat transfer to a water-cooled tube in cross flow for Ω = 0.23 to 0.35. Ivernel and Vernotte [55] studied oxygen-enriched air/ natural gas flames, with Ω ranging from 0.25 to 1.0. Kataoka, Shundoh, and Matsuo [56] blended O2 and N2 to an equivalent Ω of 0.39. Baukal and Gebhart [57] studied natural gas flames combusted with air, oxygenenriched air, and pure oxygen. They found that the heat transfer to the target was highly dependent on the oxidizer composition.
© 2011 by Taylor and Francis Group, LLC
Some studies used preheated air for combustion. Matsuo et al. [58] used combustion air heated to about 500 K (440°F) before reaching the burner. Malikov et al. [12] used both ambient air and air preheated up to 740 K (870°F). 9.2.2.2 Fuels Another parameter of interest is the fuel composition. Only experiments using gaseous fuels are considered here. Tables 9.1 through 9.4 indicate that natural gas and methane, the main constituent in natural gas, have been the most widely used. Where specified, the specific natural gas compositions are noted in Table 9.5. An example of another fuel used is acetylene by Schulte [59]. The study by Fay [60] was the only one in which MAPP® gas was used. This is a stabilized mixture of methyl acetylene and propadiene (equivalent to C3.1H5.4) that is produced by Dow Chemical. Beér and Chigier [44] used another uncommon fuel, coke oven gas that is a by-product of converting coal to coke. Davies [61] and Shorin and Pechurkin [40] used town gas that is also obtained from coal. Both coke oven gas and town gas are roughly half H2 and a third CH4, with a mixture of other gaseous hydrocarbons. Both fuels have similar heating values and adiabatic flame temperatures. In one set of tests, Shorin and Pechurkin [40] used a mixture of propane and butane in unspecified proportions. Ng, Leung, and Cheung [86] and Zhen, Leung, and Cheung [87] used a blend of 30% propane and 70% butane. Dong, Cheung, and Leung extensively studied butane [42,62–66,69,70]. Saha, Ganguly, and Datta [71] compared the heat transfer and emissions from methane (nonluminous
Experimental Conditions for Flame Impingement Normal to Plane Surfaces Oxidizer Air
Fuel
ϕ
qf (kW)
coke oven gas CH4
not given 1.33, 1.67
606 0.71
2.58–2.76
1.0–160
4.8, ∞ 2.75
46–220c 1.0–160
0.51
0.7
0.78
22
1.0–8.9 1.0
0.5–20 0.75
0.8–1.2 1.54–2.57 1.0 2.57 1.0
0.33–0.72 0.5–0.8 not given 0.5–0.9 1.2, 1.5
0.7–1.2 1.0 1.0 1.0
0.4–1.2 0.33–0.72 0.44 0.63
0.95–1.3
0.40–0.93
1.0 1.0–2.4 0.9–1.2 1.0 1.0 1.8 0.8–2.1 1.0–1.8
0.37–0.56 0.6–2 0.2–0.8 0.41 0.32–0.69 0.1 0.7–4.9 not given
C2H2 C2H4 C3H8 C4H10
c
dn (mm)
Ren
L
R
100 13, 16
turbulent laminarc
5–9 2.4
0–7 0–1.3
partial premix
7.8
17,680–22,700
13–65
0–4
tube partial premix
12.7 7.8
7000–35,300 17,680–22,700
10, 16 13–64
0–60 0–10
long tube with precombustor premix premix, pulsed tube 3 tubes in a triangular pattern tube tube premix torch tube premix tube with stainless steel mesh tube slot 2 slots in a row 3 tubes in a row tube slot 2 tubes in a row inverse diffusion tube 3 slots in a row tube tube inverse diffusion array with swirl
9.5
3100
7.5
0–3
Buhr et al., 1973 [82] (cf. Buhr, 1969 [83]) Milson & Chigier, 1973 [24] Kremer et al., 1974 [41] (cf. Buhr, 1969 [83]) Posillico, 1986 [100]
50
2, 3, 4
0–5
Eibeck et al., 1993 [18]
6.6 6.35
6630 9700 1500–6000 800
5–35 2–7
0–27 0–12
Tuttle et al., 2005 [50,51] Chander & Ray, 2007 [94]
8, 9.7, 12 12.8 2.0 12.8 25
800–1700 515–310 laminara 515–310 c 1300
1.6–14 1–5 1.3–13 1–5 23, 32
0 0 0–2.5 0 0–17
Chander & Ray, 2008 [95] Saha et al., 2008 [71] Schulte, 1972 [59] Saha et al., 2008 [71] Veldman et al., 1975 [80]
10 13.7 × 4.6 9 × 3 5 8.7 13.4 × 4.3 5 13.9 9 13.4 × 4.3 8.6 2.54 10.4 15.3
600–2500 800–1700 800 900
1.0–8.0 2.0–13.0 1–6 2–8
0–6.5 0–7 0–6 0–12
Dong et al., 2001 [62] Dong et al., 2002 [63] Dong et al., 2003 [64] Dong et al, 2003 [65]
1000–1700
2
0–3.5
Kwok et al., 2003 [88]
800–1200 2000–3000 500–1800 1000 800–1700 500 3000–8000 500–2500
1–7 0–12 3–7 2–6 1.5–4.0 0 1.5–16.5 1.3–2.0
0–10 0–2 0–7 0–7 0–3.5 0 0–12 0
Dong et al., 2004 [66] Sze et al., 2004 [85] Zhao et al., 2004 [67] Kwok et al., 2005 [93] Huang et al., 2006 [89] Qi et al., 2006 [68] Dong et al., 2007 [67,70] Zhao et al., 2009 [90]
Burner tunnel with air preheat partial premix
c
Reference Matsuo et al., 1978 [58] Anderson & Stresino, 1963 [33]
Flame Impingement Measurements
Table 9.3
(Continued)
219
© 2011 by Taylor and Francis Group, LLC
220
Table 9.3 Experimental Conditions for Flame Impingement Normal to Plane Surfaces (Continued) Oxidizer Air
ϕ
qf (kW)
0.8–1.8 1.0–2.0
0.42–1.4 4.7–9.4
natural gas #1 natural gas natural gas #3 natural gas #2 natural gas #3 natural gas #5 natural gas
0.95 1.0 1.0 1.0 1.0 ∞ 1.05
2100 1000–1500 not given 1000, 1500 not given 1.67, 8.51 0.3, 0.4c
natural gas #4 natural gas
1.0 0.65–2.79
not given 39.7
natural gas
∞
natural gas
1.0
10.5 5.1, 10.5 46
natural gas #6 natural gas natural gas #7
1.0 1.0–1.05 1.0
5 ? 60
town gas +
0.67–0.95
not given
© 2011 by Taylor and Francis Group, LLC
Burner inverse diffusion inverse diffusion with swirl variable swirl tunnel tunnel tunnel tunnel tube flame retention head tunnel coaxial radial jet reattachment flame tube 2 coaxial radial jet reattachment flames flame working torch array of tubes 3 coaxial radial jet reattachment flames water-cooled tube with precombustor
dn (mm)
Ren
L
R
9.3 12
2000–3000 6000–10000
4–10 0–11
0–8.5 0–15
131.3–152.0 172 13.8 not given 13.8 55 5.6–15.3
turbulentc turbulentc 1050, 1860 not given 1050, 1860 70, 356 7050–16,200
5.7–11 0–8.4 2–20 (60–100 cm) 2–20 7.3 16–27
0–4.6 0–3.5 0–5 (0–37 cm) 0–10 0–7.3 0
13.8, 27.6 63.5
1700–4250 3640–24,600
1–12 0.93
0 0–8.11
3.81 6.35 63.5
5000 1450, 3000 6700
0–170 0–100 0.24
0–100 0–60 0–1.1
Mohr et al., 1996 [46], 1997 [48]
38.5 4 63.5
2910 38,000–51,000 8855
0.5–6.0 25–75 1.18
0–1.0 0–15 0–1
Baukal and Gebhart, 1998 [57] Malikov et al., 1999 [29] Wu et al., 2001 [49]
25, 46
1500–19,000
0–14
0–5
Shorin & Pechurkin, 1968 [40]
Reference Ng et al., 2007 [85] Zhen et al., 2009 [87] Vizioz & Lowes, 1971 [52] Smith & Lowes, 1974 [74] Hoogendoorn et al., 1978 [105] Rajani et al., 1978 [75] Popiel et al., 1980 [106] You, 1985 [10] Horsley et al., 1982 [98] (cf. Tariq, 1982 [99]) van der Meer, 1991 [107] Mohr et al., 1995 [45], 1996 [47] Rigby and Webb, 1995 [96]
Industrial Combustion Testing
Fuel 30% C3H8, 70% C4H10
O2
CH4
0.61
not given
natural gas #1 natural gas #6 CH4
0.95 1.0 0.95–1.31
2100, 3000 5–25 not given
C2H2
2.34 0.9–2.5 1.0 1.43 1.0–2.0 1.45–1.83
4.1 2.9 2.7 78 2.9 not given
1.0 0.83–1.0
61, 94 34
1.0 0.85 1.1 0.85–2.2 1.0 1.0 1.0
105 5.4 2.8 2.9 not given 1000–1750 5–25
C3H8
C3H8 + C4H10 H2
MAPP® gas natural gas natural gas #2 natural gas #6
converging nozzle with precombustor variable swirl flame working torch premixed multiport partial premix premix welding torch tunnel, partial premix premix torch premixed multiport premix welding torch water-cooled tube with precombustor tunnel (2) glass working torches premix welding torch premix torch (2) commercial burners flame working torch
10
857–2000
2–20
0–3
115.4 38.5 9.5
turbulentc 1850–9200 489–549
8.7 0.5–6.0 0.74, 2.1
0–4.8 0–1.0 0
1.8 not given 1.8 9.5 not given 9.5
laminarc turbulentc laminarc turbulentc 551–692
4.3 (b) 1 8.6 (b) 0.74, 2.1
0–36 (0–1.3 cm) 0–13 0–29 (0–1.3 cm) 0
25 25
turbulentc 300–1200
6, 8 0–8
0 0
9.5 27 5.6 not given 0.53, 1.7 not given 38.5
laminarc 2000c 4600c turbulentc laminara not given 810–4050
1.3 2 9 (b) 1.2–29 (60–100 cm) 0.5–6.0
0–12 0–2.5 0–10 (0–1.3 cm) 0–9.5 (0–37 cm) 0–1.0
Kataoka et al., 1984 [56] Vizioz and Lowes, 1971 [52] Baukal and Gebhart, 1998 [57] Kilham & Purvis, 1978 [81] (cf. Purvis, 1974 [77]) Anderson & Stresino, 1963 [33] Fay, 1967 [60] Anderson & Stresino, 1963 [33] Fay, 1967 [60] Kilham & Purvis, 1978 [81] (cf. Purvis, 1974 [77]) Rauenzahn, 1986 [6] Shorin & Pechurkin, 1968 [40]
Flame Impingement Measurements
Air/O2
Anderson & Stresino, 1963 [33] Reed, 1963 [84] Fay, 1967 [60] Schulte, 1972 [59] Rajani et al., 1978 [75] Baukal and Gebhart, 1998 [57]
According to author (only qualitative). Tip of inner flame cone just touching target. c Calculated from data given in reference. a
b
221
© 2011 by Taylor and Francis Group, LLC
222
Industrial Combustion Testing
flames) and ethylene (luminous flames) impinging on a water-cooled plate. The combination of fuel type and the equivalence ratio ϕ determines the tendency to produce soot and, therefore, luminous gas radiant emission. This tendency is higher in fuel rich mixtures (ϕ > 1). It also increases with higher carbon-to-hydrogen weight ratios in the fuel. For example C4H10, which has a C/H weight ratio of 4.8, has a higher propensity to produce soot than CH4, which has a C/H weight ratio of 3. 9.2.2.3 Equivalence Ratios
Figure 9.13 Example of a natural gas flame impinging normal to a water-cooled disk. (a) Target
Burner (b)
20°
Burne
r
Target
This ratio directly affects both the sooting tendency and the level of dissociation in the combustion products. Fuel rich flames (ϕ > 1) produce a combination of both luminous and nonluminous thermal radiation. The combustion products of these flames may also contain unreacted fuel components, due to insufficient oxygen. Fuel lean flames (ϕ < 1) normally do not produce luminous thermal radiation, due to the absence of soot particles. These flames seldom produce significant quantities of unreacted fuel species, unless the flame temperature is high enough to produce dissociation. Flames at or near stoichiometric (ϕ = 1) produce the highest flame temperatures, due to complete combustion. They also generally produce only nonluminous radiation, since no soot is generated. Most of the studies used stoichiometric mixtures. Cookson and Kilham [72,73] and Davies [61] used equivalence ratios as fuel lean as ϕ = 0.5. Milson and Chigier [24]; You [10]; and Hustad and co-workers [8,34,35] used pure fuel jets (ϕ = ∞). Many studies have investigated a range of equivalence ratios, in order to determine the resulting effects on heat transfer to the target. Davies [61] used lean and rich mixtures. The actual equivalence ratios were not specified. 9.2.2.4 Firing Rates
Figure 9.14 Flames impinging (a) parallel, and (b) oblique to a plane surface. (From Baukal, C. E., Heat Transfer from Flame Impingement Normal to a Plane Surface, Saarbrücken, Germany: VDM Verlag/Dr. Müller, 2009.)
The firing rate or gross heat release (qf) of the flames has ranged from 0.3 to 3000 kW (103 to 107 Btu/hr). Figure 9.16 shows oxy/natural gas flames impinging
Table 9.4 Experimental Conditions for Flame Impingement Parallel to Plane Surfaces Oxidizer Air O2
Fuel
ϕ
qf (kW)
Burner
dn (mm)
Ren
L
R
coke oven gas C2H2
1.0
1470
concentric tubes
150
9
0–10
2.5
not given not given
(30) multihead torch tips multihead torch tips
not given
4800– 270,000 turbulenta
(1.3–14 cm)
not given
laminara
not given not given
2.5
According to author (only qualitative).
a
© 2011 by Taylor and Francis Group, LLC
(1.3–14 cm)
Reference Beér & Chigier, 1968 [44] Giedt et al., 1960 [3] Woodruff & Giedt, 1966 [43]
223
Flame Impingement Measurements
(a)
(b)
5 kW, Ω = 0.21
(c)
5 kW, Ω = 1.00
15 kW, Ω = 0.30
Figure 9.15 Flames impinging normal to a plane surface with natural gas and variable oxidants: (a) air (Ω = 0.21, 5 kW), (b) Ω = 0.30 (15 kW), and (c) pure oxygen (Ω = 1.0, 5 kW). (a)
(b)
(c)
5 kW, Ω = 1.00
15 kW, Ω = 1.00
Figure 9.16 Flames impinging normal to a plane surface with natural gas combusted with pure O2 (Ω = 1.0) at different firing rates: (a) 5 kW, (b) 15 kW, and (c) 35 kW.
Table 9.5 Natural Gas Compositions (Percentage by Volume) No.
CH4
1 2 3 4 5 6 7
81.3 84.9 81.3 81.3 94.9 95.3 90.361
C2H6 2.9 5.64 2.85 2.85 3.75 2.54 5.749
C3H8
C4H10
CmHn
CO2
0.4
0.1
0.1 2.0 0.60
0.8 0.18 0.89
0.12 0.12 0.012
0.42 0.87 2.528
0.27 0.54 0.498
Note: Where CmHn = other hydrocarbons.
© 2011 by Taylor and Francis Group, LLC
0.9 0.2 0.058
O2
0.01
0.002
N2
Inerts
14.4 7.27 14.35 14.35 0.41 0.42 3.312
224
on a water-cooled disk at different firing rates. Large industrial-scale flames were used by Beér and Chigier [44]; Vizioz and Lowes [52]; Smith and Lowes (1974) [74]; and Rajani, Payne, and Michelfelder (1978) [75] at the International Flame Research Foundation (IFRF) in IJmuiden, the Netherlands. Beér’s flame impinged downward on the hearth of a furnace. In the other three studies, the flames impinged on water-cooled targets located inside a furnace environment. Many of the studies (e.g., Kilham, and Purvis [76]; Purvis [77]) considered here used torch tips, firing at under 50 kW (170,000 Btu/hr). 9.2.2.5 Reynolds Number The Reynolds number at the burner nozzle, Ren, varied from 50 to 330,000. Both laminar (e.g., Fairweather, Kilham, and Nawaz [54]) and turbulent (e.g., Fairweather, Kilham, and Mohebi-Ashtiani [78]) flow conditions arose. As shown in Tables 9.1 through 9.4, the Reynolds number was not always given. For some studies, the flows were indicated to be either laminar or turbulent. In other studies, the nozzle Reynolds number was estimated from other information, such as nozzle diameter and gas flow rates. The Reynolds number varies directly with the burner diameter. It is also influenced by the burner design. In partially or fully premixed flames, the combustion products leave the burner at an elevated temperature. However, in diffusion flames, the gases leave the burner at essentially ambient conditions. Since the gas viscosity increases with temperature, Ren is generally lower if the gases have been heated at the burner exit. In some studies, turbulence effects were analyzed (e.g., Hargrave and Kilham [79]). However, in most studies, the turbulence effects were not assessed. 9.2.2.6 Burner Designs Many different types have been used. These range from fully premixed to diffusion mixing, downstream of the burner exit. In fully premixed burners, the fuel and oxidizer mix prior to reaching the nozzle exit. An example was the study by Veldman, Kubota, and Zukoski (1975) where premixed gases were fired in a burner consisting of a tube with a stainless steel mesh [80]. As another example, Kilham and Purvis [81] used a premix multiport burner. The resulting flame in premix burners may have either a uniform or a nonuniform velocity profile, depending on the nozzle design. It also depends on the distance between the ignition point and the exit. In most studies using this burner type, both the temperature and the composition of the combustion products, at the burner exit, were approximately uniform.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
This circumstance is often simpler to analyze, since the flame conditions are well established at the nozzle exit. There are then few subsequent chemical reactions ahead of the target surface. The tunnel burner is a common fully premixed burner. The gases are mixed and ignited inside the burner. They then travel through a refractory-lined chamber before leaving the burner. The combustion products may equilibrate inside the chamber. The temperature and composition are then uniform at the exit. However, the velocity profile may not be uniform. It may be approximately developed pipe flow, depending on the downstream length of the equilibration chamber. In partially premixed burners (e.g., Buhr, Haupt, and Kremer [82]; Buhr [83]), the fuel and oxidizer mix prior to reaching the nozzle exit. However, only a portion of the stoichiometric amount of oxygen is supplied through the burner. The rest is provided by mixing with the surrounding ambient air, entrained into the flame. At the nozzle exit, the velocity profile is commonly nonuniform. Both uniform and nonuniform outlet temperature profiles and compositions have been reported. In diffusion-mixing burners, the fuel and oxidizer begin to mix at the nozzle exit, where the velocity is often nonuniform. Two examples of diffusion burners used in flame impingement testing are shown in Figure 9.17. An example is the glass working torches used by Reed [84]. In diffusion burners, the exit temperature field is commonly homogeneous and equal to ambient conditions. The gas composition at the exit is pure fuel and pure oxidizer with no combustion products. If the oxidizer is not supplied through the burner, a pure diffusion flame results. The oxygen is provided for combustion by ambient air entrainment into the flame. Most diffusion burners have the fuel in the center surrounded by the oxidizer. Some studies [67,70,85–87] used an inverse diffusion burner, where the oxidizer is in the center and diffuses into the outer fuel jets. Most burners tested have been round, although some were rectangular (referred to as “slot”). Kwok, Leung, and Cheung [88] compared a round burner against a slot burner. Many had some type of straightening material such as glass or metal beads, steel mesh, or a bundle of tubes to make the outgoing gas flow uniform. Some burners had swirl (e.g., Huang et al. [89]; Zhao et al. [90]), a precombustion chamber (e.g., Kataoka, Shundoh, and Matsuo [56]), and a converging nozzle (e.g., Fairweather, Kilham, and Nawaz [54]). Some were similar to torches used in metal working. Fernandes and Leandro (2006) studied acoustically driven oscillating air/propane flames impinging on a flat plate [91]. One study augmented the flame with electrical power [92].
225
Flame Impingement Measurements
9.2.2.7 Number of Burners Nearly all previous studies considered single burners. Some recent studies have studied the effects of multiple burners. Malikov et al. [12] studied an array of high velocity burners impinging on a flat plate inside a furnace. Dong, Leung, and Cheung [64–66] tested two slot, two round, and three butane/air burners in a row at various spacings impinging on a flat plate. Kwok, Leung, and Cheung [93] studied three butane/air slot burners in a row at various spacings impinging on a flat plate. Chander and Ray [94] tested three interacting methane/air flames arranged in a triangle with various spacings impinging on a flat surface. 9.2.2.8 Burner Nozzle Diameter In the configurations considered here, the exiting flame-shape was round, except for those used by Kilham [31] and Davies [37], which were slot-shaped. The burner nozzle diameter, dn, ranged from 0.53 to 152 mm (0.02 to 6.0 inch). In many studies (e.g., Conolly, (a)
and Davies [4]), the burner outlet consisted of a nested bundle of small tubes or orifices, arranged in a circular pattern. This is common practice for nozzles used on cutting and welding torches. In some studies, (e.g., Chander and Ray [36,95]), a range of nozzle diameters was tested. 9.2.2.9 Location The dimensionless axial distance between the burner exit and the target, L, varied from 0 to 200. In most studies it was less than about 20. Figure 9.18 shows two different spacings used in some flame impingement experiments. This distance has a strong influence on the resulting heat transfer to the target. Shorter distances result in higher flux rates. One cause is less ambient air entrainment. Another reason is that the flame widens at longer distances. This diffuses the heat flux over a wider cross-sectional area. Fay [60] adjusted the axial position of a premixed torch flame so that the inner flame cone tip touched the target. The specific axial locations were (b)
Figure 9.17 Examples of different diffusion burner designs. (a)
(b)
Figure 9.18 Examples of different burner spacings used in natural gas flames impinging on a flat disk.
© 2011 by Taylor and Francis Group, LLC
226
Industrial Combustion Testing
Table 9.6 Flame Jet Impingement Studies with the Target in a Furnace Furnace Wall Temperature (K)
Furnace Dimensions (m)
Beér and Chigier, 1968 [44] Vizioz and Lowes, 1971 [52] Smith and Lowes, 1974 [74] Matsuo et al., 1978 [58]
1200–1500 not given
2 W × 2 H × 6.25 L 2 W × 2 H × 6.25 L
not given
2 W × 2 H × 6.25 L
1300–1500
Rajani et al., 1978 [75] Ivernel and Vernotte, 1979 [55] Malikov et al., 1999 [29]
770–1870 1650–2000
2.16 W × 1.80 H × 1.36 L 1.0 dia. × 4.5 L 0.6 dia. × 2.05 L
Malikov et al., 2001 [12]
1200–1400
Reference
1100–1300
0.6 W × 0.6 H × 1.7 L 0.3 dia. × 1.2 L
not given. Davies [37] positioned water-cooled tubes at the location in the flame that produced the highest heating rate. This location varied with the oxidizer composition and the equivalence ratio. Again, no dimensions were reported. For flames parallel to plane surfaces, L is taken as the distance from the burner to the leading edge of the target. The dimensionless radial distance, R, from the nozzle centerline to the location on the target where the heat flux was measured, is an important parameter. The heat flux at the stagnation point (R = 0), was measured in every study. In most of the experiments of jet flames impinging normal to plane surfaces, the radial heat flux profile was measured. The radial distance, R, ranged from 0 to as high as 60. In general, the heat flux decreased with R. However, in several studies the peak heat flux did not occur at the stagnation point of the target. For example,
Table 9.7 Stagnation Targets for Flames Impinging Normal to Cylindrical Surfaces Reference Kilham, 1949 [31] Jackson & Kilham, 1956 [32] Anderson & Stresino, 1963 [33] Davies, 1965 [61] Fells & Harker, 1968 [92] Davies, 1979 [37] Hemeson et al., 1984 [97] Hargrave et al., 1987 [109] Hustad et al., 1991 [34,35] Hustad et al., 1992 [8] Malikov et al., 2001 [12] Chander & Ray, 2007 [36]
db (mm)
Material
3
sillimanite
3.2
sillimanite
0.91, 1.5 1.5 3.2 50 2.7 32.6–59.8 22 50 50 108 118
copper stainless steel copper stainless steel stainless steel not given brass steel steel not given brass
Surface Coating (s) or Treatment Al2O3, FeO, Cr2O3, uranium oxide Al2O3, FeO, Cr2O3, uranium oxide none none none none none none none none none not given uncoated
tw (K) 1220–1630 950–1445 <373 <373 <373 <373 300 not given 378 520–620 520–570 not given 311–313
Table 9.8 Stagnation Targets for Flames Impinging Parallel to Hemi-Nosed Cylinders Reference Cookson & Kilham, 1963 [72] Kilham & Dunham, 1967 [101] Kilham & Purvis, 1971 [76] Conolly & Davies, 1972 [4] (cf. Conolly, 1971 [103]) Nawaz, 1973 [53] Ivernel & Vernotte, 1979 [55] Fairweather et al., 1984 [54] Fairweather et al., 1984 [78] Hargrave & Kilham, 1984 [79] Hemeson et al., 1984 [97] Hargrave et al., 1987 [109]
© 2011 by Taylor and Francis Group, LLC
Db (mm)
Material
14 22.2 9.5 12.7
brass brass stainless steel stainless steel
none none, A418, Pt none none, SiO2, Pt
brass inconel brass brass brass not given brass
none none none none none none none
9.5 46 9.5 9.5 22 50–150 22
Surface Coating(s) or Treatment
tw (K) >373 >373 290–470 400 290–420 600–1400 340–1600 ∼378 393 not given 418
227
Flame Impingement Measurements
Table 9.9 Targets for Flames Impinging Normal to Plane Surfaces Reference Anderson & Stresino, 1963 [33] Reed, 1963 [84] Fay, 1967 [60] Shorin & Pechurkin, 1968 [40] Vizioz & Lowes, 1971 [52] Schulte, 1972 [59] Buhr et al., 1973 [82] (cf. Buhr, 1969 [83]) Milson & Chigier, 1973 [24] Kremer et al., 1974 [41] Smith & Lowes, 1974 [74] Veldman et al., 1975 [80] Hoogendoorn et al., 1978 [105] Kilham & Purvis, 1978 [81] Matsuo et al., 1978 [58] Rajani et al., 1978 [75] Popiel et al., 1980 [106] Horsley et al., 1982 [98] (cf. Tariq, 1982 [99]) Kataoka et al., 1984 [56] You, 1985 [10]
Dimensions (cm)
Material
tw (K) <373 <373 <373 <373
20 × 20 × 0.64 13 × 25 × 0.16 13 × 25 × 0.32 16 dia. 50 × 30 120 dia. × 1 120 dia. × 8 not given 32 dia.
copper copper copper not given
none none none none
steel refractory not given not given
none none not given none
<373 1280–1420
183 × 183 32 dia. not given 120 dia. × 0.1
steel not given iron steel stainless steel steel copper alumina steel steel copper copper
none none none none none none polished none none none polished none
300–650
copper not given
none
91 dia. × 0.16 32 × 57 15.2 × 15.2 × 5.6 140 × 80 × 2.5, 10 74 dia. × 0.8 57.3 × 32.0 0.56 dia. 42 dia. 100 dia.
Posillico, 1986 [100] Rauenzahn, 1986 [6] van der Meer, 1991 [107] Eibeck et al., 1993 [18] Mohr et al., 1995 [45], 1996 [47] Rigby & Webb, 1995 [96] Mohr et al., 1996 [46], 1997 [48] Baukal & Gebhart, 1998 [57] Malikov et al., 1999 [29] Dong et al., 2001 [62], 2002 [63], 2003 [64,65], 2004 [66], 2007 [67,70] Wu et al., 2001 [49] Kwok et al., 2003 [88] Sze et al., 2004 [85] Zhao et al., 2004 [67]
46 dia. × 0.64 38 dia × 13 not given 61 × 61 × 3.81 103 × 103 76.2 dia. 103 × 103 10.5 dia. not given 20 × 20 × 0.8
aluminum copper copper alumina copper copper & aluminum copper brass, copper, stainless stainless steel copper
91 × 91 × 0.635 20 × 20 × 0.8 50 × 50 × 0.8 40 × 40 × 0.5
Kwok et al., 2005 [93] Tuttle et al., 2005 [50,51] Huang et al., 2006 [89] Ng et al., 2006 [86] Chander & Ray, 2007 [94,95] Ng et al., 2007 [85] Saha et al., 2008 [71] Zhao et al., 2009 [90] Zhen et al., 2009 [87]
50 × 50 × 0.8 71 × 71 × 1.27 40 × 40 × 0.8 20 × 20 × 0.8 30 dia. 50 × 50 × 0.8 6.4 dia. × 0.5 15 × 15 × 7.5 60 × 60 × 1.2
copper copper copper brass stainless copper aluminum copper copper copper copper copper not given copper
© 2011 by Taylor and Francis Group, LLC
Surface Coating (s) or Treatment
high emissivity paint none none polished none none none copper various polished none none none none none none polished none none none none none none none
<370 <373
<373 350–1000 373 970 293–347 303 490 300–1500 470–520 <373 360 306 ~300 <310 290–480 330 300–1205 305–338 323 320–338 323–1400 420–460 not given not given 311 311 313–357 316–378 311 313–323 311 311 308–318 311 308–368 not given 311
228
Industrial Combustion Testing
Milson and Chigier [24] measured a peak flux at R ranging from 6 to 29, depending on the nozzle exit Reynolds number. The turbulent mixing of ambient air with the pure jets of CH4 (ϕ = ∞) was given as the cause. In the study by Rigby and Webb [96], the heat flux to a wide range of both axial (L = 0 – 170) and radial (R = 1 – 100) locations were investigated. Table 9.6 indicates the studies in which the target was located inside a furnace. This influences the heat transfer in two important ways. The radiation from the hot refractory walls to the target is a significant portion of the total heat flux to the target. Vizioz and Lowes [52] and Smith and Lowes [74] showed that this effect was comparable in magnitude to the forced convection effect at the target. Ivernel and Vernotte [55] calculated this effect to be up to 42% of the total heat flux to a hemi-nosed cylinder. Also, any gases entrained into the jet are hot combustion products, instead of ambient air, as in most studies. Therefore, the flame jet is not cooled as much as in tests done at ambient conditions. (a)
9.2.3 Stagnation Targets Tables 9.7 through 9.10 tabulate the important features of the targets. These include the dimensions, compositions, surface conditions, and temperature level for each of the four configurations. These properties varied widely among the studies reviewed here. In some studies, these were varied to examine the effect on the heat transfer mechanisms. These are discussed further below. 9.2.3.1 Size The targets ranged in size from a 0.56 cm (0.22 inch) o.d. disk to a 200 × 625 cm (6.6 × 20.5 feet) furnace hearth. Most of the cylindrical targets have been hollow pipes. Kilham [31] and Jackson and Kilham [32] used solid refractory rods. Hustad and colleagues [8,34,35] used solid steel cylinders. All of the hemi-nosed cylinders have been under 22 mm (0.87 inch) o.d., except for Hemeson et al. [97], which were from 50 to 150 mm (2 to 6 inch). Most of the plane surfaces in Table 9.9 have been disks of over 16 mm (0.63 inch) in diameter, except
(b)
(c)
Figure 9.19 Examples of different target materials used in flame impingement tests: (a) copper, (b) brass, and (c) stainless steel. (From Baukal, C. E., and Gebhart, B., “Surface Condition Effects on Flame Impingement Heat Transfer,” Experimental Thermal & Fluid Science 15 (1997): 323–35.)
(a)
Table 9.10
(b)
Stagnation Targets for Flames Impinging Parallel to Plane Surfaces
Reference Giedt et al., 1960 [3] Woodruff & Giedt, 1966 [43] Beér & Chigier, 1968 [44]
Surface Coating (s) or Treatment
Dimensions (cm)
Material
7.6 × 7.6 × 0.16
iron
none, porcelain
294–1170
7.6 × 15 × 0.32
molybdenum
none
1100–1900
200 × 625
refractory
© 2011 by Taylor and Francis Group, LLC
none
tw (K)
1210–1470
Figure 9.20 Examples of different target coatings used in flame impingement tests: (a) alumina-coated, and (b) platinum-coated. (From Baukal, C. E., and Gebhart, B., “Surface Condition Effects on Flame Impingement Heat Transfer,” Experimental Thermal & Fluid Science 15 (1997): 323–35.)
Experimental Methods and Ranges for Total Heat Flux, Gas Temperature and Velocity for Air/Fuel Studies Total Heat Flux, q” Year
Author(s)
Gas Temperature, tj
Method
(kW/m2)
Method
(K)
line reversal (6) decreasing dia. uncoated T/C adiabatic equilibrium calculation (7) decreasing dia. coated T/ Cs line reversal (7) decreasing dia. coated T/ Cs suction pyrometer line reversal uncoated T/C
1213–1633 1179–1659
calculated calculated
5–30* 8–20*
not given
calculated
0.3–38
1360–1450
not measured
–
2260 1624–1910
not measured not measured
– –
1370–1950 2500–2750 not given
Pitot tube not measured uncooled Pitot-static probe 5 hole Pitot-static probe cooled Pitot tube
1949 1956
Kilham [31] Jackson & Kilham [32]
steady-state uncooled target steady-state uncooled target
70–130* 20–107*
1963
Anderson & Stresino [33]
steady-state cooled target
32–1100
Cookson & Kilham [72]
steady-state cooled gauge
73–180
1965 1967
Davies [61] Kilham & Dunham [101]
steady-state cooled target steady-state cooled gauge
570–950 41–95
1968
Beér & Chigier [44] Fells & Harker [92] Shorin & Pechurkin [40]
steady-state cooled gauge steady-state cooled target steady-state cooled gauge
15–54 250–840 nondimen.
1971
Vizioz & Lowes [52]
steady-state cooled target
24–280
1972
Conolly & Davies [4] (cf. Conolly, 1971 [103]) Schulte [59] Buhr et al. [82] Milson & Chigier [24] Kremer et al. [41]
transient uncooled gauge
nondimen.
Veldman et al. [80] Hoogendoorn et al. [105]
steady-state cooled gauge steady-state cooled gauge transient uncooled target steady-state cooled gauge transient uncooled gauge steady-state cooled target steady-state cooled gauge transient uncooled target steady-state cooled gauge
79–620 110–440 1.3–13 120–450 120–470 180–450 57–150 0.086–1.8 0–560
Matsuo et al. [58] Rajani et al. [75]
transient uncooled target steady-state cooled target
1979
Davies [37]
1980 1982
1973 1974
suction pyrometer line reversal
1450–1850 2200
Method
not measured uncoated T/C coated T/C coated T/C
– 573–1923 430–2200 not given
not measured not measured not measured not measured
suction pyrometer
1170–2020
coated T/C coated T/C
292–419 370–2100
23–350 262–421
uncoated T/C suction pyrometer
1200–1600 460–1950
steady-state cooled target
330–650
Popiel et al. [106]
steady-state cooled gauge
100–630
adiabatic equilibrium calculation coated decreasing dia. T/Cs
1 & 5 hole Pitot-static probe not measured cooled Pitot tube & LDV Pitot tube 1 & 5 hole Pitot-static probe not measured
370–2100
Horsley et al. [98] (cf. Tariq, 1982 [99])
transient uncooled gauge
220–440
uncoated T/C
1643–2165
Smith & Lowes [74] 1975 1978
Gas Velocity, vj
2225
cooled Pitot tube & LDV not measured
(m/s)
Flame Impingement Measurements
Table 9.11
not given – not given 2.8–46 not given – – – – 6–195 – 2–54 20–82 0.91–87 – 2–54 – (Continued)
229
© 2011 by Taylor and Francis Group, LLC
230
Table 9.11 Experimental Methods and Ranges for Total Heat Flux, Gas Temperature and Velocity for Air/Fuel Studies (Continued) Total Heat Flux, q” Year 1984
Author(s)
Method
Gas Temperature, tj (kW/m2)
Method
(K)
line reversal line reversal & uncoated T/C not measured uncoated T/C
not given 1900–2100 – 293–1500
Raman spectroscopy uncoated T/C
600–3200 400–2350
LDV LDV not measured Pitot-static probe & LDV LDV LDV
uncoated T/C coated decreasing dia. T/Cs uncoated T/C not measured
400–1700 up to 2130 not given –
Pitot tube LDV not measured not measured
5–30 up to 50 – –
–
not measured
–
300–1650 not given 350–1900 not given not given – – – – not given – – not given 330–720
not measured not measured not measured not measured not measured not measured not measured not measured not measured not measured not measured not measured not measured not measured
– – – – – – – – – – – – – –
– 300–2100 – – – – 300–1900
not measured not measured not measured not measured not measured not measured not measured
– – – – – – –
transient uncooled gauge steady-state cooled gauge transient uncooled gauge steady-state cooled gauge
240–580 120–400 100–430 0.2–11*
1986 1987
Posillico [100] Hargrave et al. [109,110]
1991
Hustad et al. [34,35] van der Meer [107] Hustad et al. [8] Mohr et al. [45] Mohr et al. [47] Mohr et al. [46] Mohr et al. [48] Malikov et al. [29] Dong et al. [62] Malikov et al. [12] Wu et al. [49] Dong et al. [63] Dong et al. [64,65] Kwok et al. [88] Dong et al. [66] Sze et al. [85] Zhao et al. [67] Kwok et al. [93] Tuttle et al. [50,51] Huang et al. [89] Ng et al. [86]
steady-state cooled target steady-state cooled gauge transient uncooled gauge transient uncooled gauge steady-state cooled gauge transient uncooled gauge steady-state cooled target
21 100–460 150–410 0–160 nondimen. 35–180 8.4–20
steady-state cooled gauge
30–210
not measured
steady-state cooled target steady-state cooled gauge steady-state cooled target steady-state cooled gauge steady-state cooled gauge steady-state cooled gauge steady-state cooled gauge steady-state cooled gauge steady-state cooled gauge steady-state cooled gauge steady-state cooled gauge steady-state cooled gauge steady-state cooled gauge none
140–270 0–200 250–500 0–190 0–190 0–240 0–380 0–170 0–300 0–310 0–210 0–85 0–190 –
steady-state cooled gauge steady-state cooled gauge steady-state cooled gauge steady-state cooled gauge steady-state cooled target steady-state cooled target steady-state cooled gauge
0–382 0–330 0–430 0–521 40–155 not given 0–285
silicon-coated T/C uncoated T/C silicon-coated T/C uncoated T/C uncoated T/C not measured uncoated T/C not measured not measured uncoated T/C not measured not measured uncoated T/C reference-beam interferometry not measured uncoated T/C not measured not measured not measured not measured uncoated T/C
1997 1999 2001
2002 2003 2004
2005 2006
2007
2008 2009
Chander & Ray [36,94] Dong et al. [67,70] Ng et al., 2007 [85] Chander & Ray [95] Saha et al., 2008 [71] Zhao et al. [90] Zhen et al. [87]
*Calculated from data given in reference.
© 2011 by Taylor and Francis Group, LLC
Method
(m/s) not given 2–3 – 0.3–2 6.7–13 3.2–14.3
Industrial Combustion Testing
1985
Fairweather et al. [78] Hargrave & Kilham [79] Hemeson et al. [97] You [10]
1992 1995 1996
Gas Velocity, vj
231
Flame Impingement Measurements
for the 0.56 mm (0.022 inch) o.d. disks used by Horsley, Purvis, and Tariq [98] and Tariq [99]. In addition to disks, many studies used square and rectangular plates as the target surfaces.
quickly conducted away. However, these materials also have relatively low melting points. For that reason, they have not been used at very high surface temperatures. Refractories, including alumina (Al2O3) and sillimanite (Al2SiO5), were also used.
9.2.3.2 Target Materials The most commonly used materials were aluminum, copper, brass, and stainless steel. Figure 9.19 shows some examples of target materials used in flame impingement experiments. For example, Posillico (1986) studied flames impinging normal to uncoated aluminum disks [100]. Aluminum, copper, and brass have very high internal thermal conductivities. Therefore, they are easier to water cool, since the heat from the impinging flame is
Figure 9.21 Thermocouple being used to measure gas temperature in a flame.
9.2.3.3 Surface Preparation Most of the target surfaces were untreated. However, in some experiments the surfaces were treated or coated, to study a specific surface effect. Kilham [31] and Jackson and Kilham [32] coated the surface of their refractory cylinders with different oxides. The objective was to estimate the emissivities of the coatings. The surface temperature was approximately determined using a thermocouple imbedded inside the cylinder body. Radiation from the surface was measured with a thermopile. Using the temperature level and the radiation, along with an energy balance on the cylinder, the emissivity of the coatings was calculated as a function of temperature. Some studies used coatings to determine the effects of surface catalysis on the heat transfer from an impinging flame. Giedt, Cobb, and Russ [3] used both uncoated and porcelain-coated iron plates. These surfaces were considered to be catalytic and noncatalytic, respectively. The measured heat flux for these two surface conditions was about the same. It was concluded that surface chemical recombination effects were negligible for those test parameters. Kilham and Dunham [101] and Dunham [102] used three different surface conditions to study the effects on air/CO flames impinging on a hemi-nosed cylinder. The results indicated that the lowest heat flux was for the National Bureau of Standards Coating A418. This coating is a mixture of
Table 9.12 Experimental Methods and Ranges for Total Heat Flux, Gas Temperature and Velocity for Air/O2/Fuel Studies Total Heat Flux, q” Year
Author(s)
1971 1973
Vizioz & Lowes [52] Nawaz [53]
1979
Davies [37]
1984
1998
Ivernel & Vernotte [55] Fairweather et al. [54] Kataoka et al. [56] Baukal & Gebhart [57]
Method
(kW/m2)
steady-state cooled target transient uncooled gauge steady-state cooled target transient uncooled gauge transient uncooled gauge steady-state cooled gauge steady-state cooled target
26.5
© 2011 by Taylor and Francis Group, LLC
Gas Temperature, tj Method
Gas Velocity, vj (K)
Method
suction pyrometer
1850–1950
not measured
1400–2600
line reversal
2600–2763
calculated
600–1600
2380–2875
not measured
1640–2750
1400–2600
adiabatic equilibrium calculation twin sonic orifice probe line reversal
2710–2760
cooled Pitot-static probe calculated
nondimen.
uncoated T/C
820–1720
250–850
not measured
–
50–1200
uncooled Pitot tube not measured
(m/s) – not given – 5–51 32–36 7.4–45.1 –
232
Table 9.13 Experimental Methods and Ranges for Total Heat Flux, Gas Temperature and Velocity for O2/Fuel Studies Total Heat Flux, q” Year
Method
(kW/m )
1956
Jackson & Kilham [32]
steady-state uncooled target
27–58c
1960 1963
Giedt et al. [3] Anderson & Stresino [33]
transient uncooled target steady-state cooled target
200–670 32–41000
Reed [84]
steady-state cooled target
0–1450
1965 1966 1967 1968
Davies [61] Woodruff & Giedt [43] Fay [60] Shorin & Pechurkin [40]
steady-state cooled target transient uncooled target steady-state cooled target steady-state cooled gauge
1971 1972
Kilham & Purvis [76] Conolly & Davies [4] (cf. Conolly, 1971 [103]) Schulte [59] Nawaz [53] Kilham & Purvis [81] Rajani et al. [75]
1973 1978
1979
Method
Gas Velocity, vj (K)
Method
1394–1468
calculated
– not given
not measured calculated
3500–5500 nondimen. 0–5600 nondimen.
(6) decreasing dia. uncoated T/Cs not measured adiabatic equilibrium calculation adiabatic equilibrium calculation line reversal line reversal not measured uncoated T/C
transient uncooled gauge transient uncooled gauge
1990–2590 410–2400
line reversal line reversal
2755–2925 2543–3144
steady-state cooled gauge transient uncooled gauge transient uncooled gauge steady-state cooled target
140–3500 1700–3600 950–1500 349–1290
not measured line reversal line reversal twin sonic orifice probe
– 2795–2996 2755–2925 323–3060
Davies [37]
steady-state cooled target
1350–3400
3050
Ivernel & Vernotte [55]
transient uncooled gauge
340–1300
adiabatic equilibrium calculation twin sonic orifice probe
2300–2800
Fairweather et al. [54] Rauenzahn [6] Baukal & Gebhart [57]
transient uncooled gauge transient uncooled target steady-state cooled target
1700–3700 645–1210 150–950
line reversal not measured not measured
2930–2990 – –
Calculated from data given in reference.
c
© 2011 by Taylor and Francis Group, LLC
3080 2260 2900–3200 – not given
calculated
(m/s) 3–6c – 43–1400 7.5, 10
not measured cooled Pitot tube not measured uncooled Pitot-static probe calculated cooled Pitot tube
– 40–64 – not given
not measured calculated calculated cooled 1 & 5 hole Pitot-static probe not measured
– not given 39.5–48.6 2–135
cooled Pitot-static probe calculated not measured not measured
11–66
21–32 9–17
–
35–38 – –
Industrial Combustion Testing
1984 1986 1998
Author(s)
Gas Temperature, tj 2
233
Flame Impingement Measurements
Table 9.14 Reported Ranges for Each Heat Flux Technique Total Heat Flux
Gas Temperature
(kW/m2)
(K)
20–130
1179–1659
0–41000
323–3200
0–3500
293–2350
0.086–1210
292–3200
0–3700
400–3144
Technique 1. Steady-state uncooled target 2. Steady-state cooled target 3. Steady-state cooled gauge 4. Transient uncooled target 5. Transient uncooled gauge
Table 9.15 Reported Ranges for Each Temperature Technique Gas Temperature Technique
(K)
Fine wire thermocouple Line reversal Raman spectroscopy Reference-beam interferometry Suction pyrometer Twin sonic orifice
292–2350 1213–3200 600–3200 330–720 460–2020 323–3060
Table 9.16
Line Reversal Measurements
eight oxides. It was assumed to be noncatalytic. On the other hand, the highest flux was found for a platinum coating, which was considered to be perfectly catalytic. The heat flux to an uncoated copper heat flux gauge surface was intermediate to the results for the A418 and platinum coatings. However, the differences were small. It was concluded the bulk of the chemical reactions occurred in the boundary layer. In addition, as L increased, the heat flux for all three surface conditions converged to the same value. Davies and Conolly [4,103] found no difference in heat flux, between a surface coated with highly catalytic platinum and a surface coated with SiO2, which was expected to be noncatalytic. This again led to the conclusion that the recombination of radical species occurs in the boundary layer, prior to reaching the surface. Baukal and Gebhart [104] also found no difference in heat flux to surfaces coated with alumina or platinum. Figure 9.20 shows some examples of targets with different coatings that were used in that study. You [10] used a high emissivity (0.96) paint to maximize the amount of radiant heat absorbed by the target. Three interrelated studies [105–107] used polished copper surfaces to reduce the effects of flame radiation. In some studies the heat flux gauge surface was coated or treated. 9.2.3.4 Surface Temperatures These ranged from 290 to 1900 K (63°F–2960°F). For many measurements, the temperature, tw, was maintained below 373 K (212°F), using water-cooled targets. In some studies (e.g., [78,108]) the surface temperature was slightly above 373 K (212°F). This eliminated the possibility of combustion products condensing on the target. Ethylene glycol was used as the target coolant because it has a higher boiling temperature than water. Beér and Chigier [44] and Vizioz and Lowes [52] measured surface temperatures above 1200 K (1700°F), for refractory targets. In some studies, the surface temperature level was actually for the heat flux gauge, and not the target. For example, Fairweather, Kilham, and Nawaz [54] reported a maximum surface temperature of 1600 K (2400°F). However, the target was made of brass, which melts at about 1300 K (1900°F). A stainless steel heat flux gauge, imbedded in the brass target, was used to measure the heat flux. Stainless steels have a melting point of about 1700 K (2600°F).
Seeding
Comparison Light Source
Sodium
not given
3200
Sodium
tungsten filament
2100
Sodium
tungsten filament tungsten filament tungsten filament xenon arc carbon arc
2260
Woodruff & Giedt, 1966 [43] Hargrave & Kilham, 1984 [79] (cf. Hargrave, 1984 [108]) Davies, 1965 [61]
2925
Kilham & Purvis, 1971 [76]
3144
Conolly & Davies, 1972 [4]
2925 2750
Kilham & Purvis, 1978 [81] Fells & Harker, 1968 [92]
1633
Kilham, 1949 [31]
9.2.4 Measurements
2990
Fairweather et al., 1984 [54,78] Nawaz, 1973 [53]
Table 9.11 shows the measured heat fluxes, gas temperatures, and gas velocities for the air/fuel flame impingement studies, for all geometries. Figure 9.21 shows an example of a thin-wire thermocouple used to measure
Sodium Sodium Sodium Sodium carbonate Sodium chloride Sodium chloride Sodium chloride
tungsten filament tungsten filament xenon arc
max. tj (K)
2996
© 2011 by Taylor and Francis Group, LLC
Reference
234
Industrial Combustion Testing
Table 9.17 Fine Wire Thermocouple Measurements T/C Type
Coating
max. tj (K)
o.d. (mm)
Chromel/alumel
K
Iron/constantan Pt/Pt-Rh
J
Pt/Pt-10%Rh
S
S, B
None None Teflon None None Al2O3 None None None Yt2O3/BeO
not given 1700 419 1720 1923 not given 1500 1659 1970 2100
not given not given 0.13 0.1 0.1 0.1 0.05 0.061–0.193 not given 0.05–0.18
Hustad et al., 1992 [8] Hustad et al., 1991 [34,35] Veldman et al., 1975 [80] Kataoka et al., 1984 [56] Buhr et al., 1973 [82] Kremer et al., 1974 [41] You, 1985 [10] Jackson & Kilham, 1956 [32] Wu et al., 2001 [49] Hoogendoorn et al., 1978 [105]
R
Yt2O3/BeO Yt2O3/BeO None
2100 2130 not given
0.05–0.18 0.05–0.18 not given
None None SiO2 SiO2 SiO2 silicon silicon None None None None None None None None None
2165 2350 1450 1910 2200 1650 1900 not given not given not given not given not given not given 2100 1900 1600
0.5 0.24 0.127 0.127 0.127 not given not given 0.3 not given not given not given not given 0.25 not given not given not given
Popiel et al., 1980 [106] van der Meer, 1991 [107] Hargrave & Kilham, 1984 [79] (cf. Hargrave, 1984) Horsley et al., 1982 [98] Hargrave et al., 1987 [109] Cookson & Kilham, 1967 [72] Kilham & Dunham, 1967 [101] Milson & Chigier, 1973 [24] Malikov et al., 1999 [29] Malikov et al., 2001 [12] Shorin & Pechurkin, 1968 [40] Dong et al, 2001 [62] Dong et al, 2002 [63] Kwok et al., 2003 [88] Zhao et al., 2004 [67] Huang et al., 2006 [89] Dong et al., 2007 [67,70] Zhen et al., 2009 [87] Matsuo et al., 1978 [58]
T/C Alloys
Pt/Pt-10%Rh & Pt-30%Rh/Pt-6%Rh
Pt/Pt-13%Rh
Pt-30%Rh/Pt-6%Rh
B
Not given
Reference
Table 9.18 Pitot Measurements Probe Type Pitot tube
Pitot-static
5-hole
Ren laminar 50–500 200–3600 857–2000 1050, 1860 turbulent turbulent 4800–270,000 not given 70, 356 1500–19,000 not given turbulent turbulent not given
© 2011 by Taylor and Francis Group, LLC
max. tj (K)
Coolant
Material
o.d. (mm)
3200 2200 1700 1720 2100
water water not given none water
not given copper not given quartz tantalum
not given 3.6 not given 1.4 1.2
2020 1600 1950 3060 1500 not given 2800 1950 2020 3060
not given not given not given water not given none water not given not given water
not given not given not given tantalum not given alundum not given not given not given tantalum
not given not given not given 12 not given 1.8 not given 8 not given 12
Reference Woodruff & Giedt, 1966 [43] Conolly & Davies, 1972 [4] Hustad et al., 1991 [34,35] Kataoka et al., 1984 [56] Hoogendoorn et al., 1978 [105]; Popiel et al., 1980 [106] Smith & Lowes, 1974 [74] Matsuo et al., 1978 [58] Beér & Chigier, 1968 [44] Rajani et al., 1978 [75] You, 1985 [10] Shorin & Pechurkin, 1968 [40] Ivernel & Vernotte, 1979 [55] Vizioz & Lowes, 1971 [52] Smith & Lowes, 1974 [74] Rajani et al., 1978 [75]
235
Flame Impingement Measurements
Table 9.19 Analyzers used to Measure Gas Compositions Reference
CH4
CO
CO2
Buhr, 1969 [83] Conolly, 1971 [103] Vizioz & Lowes, 1971 [52] Nawaz, 1973 [53] Smith & Lowes, 1974 [74] Rajani et al., 1978 [75] Ivernel & Vernotte, 1979 [55] You, 1985 [10] Posillico, 1986 [100] Malikov et al., 1999 [29] Malikov et al., 2001 [12] Wu et al., 2001 [49] Sze et al., 2004 [85] Dong et al., 2007 [67,70] Saha et al., 2008 [71]
GC
GC
GC
GC
GC GC GC IR GC GC
GC GC GC IR GC GC RS GC GC IR IR ?
GC
GC GC GC
GC GC IR EC ? ?
HC
H2 GC calc. GC GC GC GC GC
H2O
N2
NO
NO2
GC
calc.
GC
GC GC
GC CL
CL
CL EC CL ?
CL
GC GC RS
GC GC
FID
O2
OH
GC
CL ?
GC GC GC Para. GC GC RS GC GC Para. EC ? ?
LA
Note: Key; CL = chemiluminescent; EC = electro-chemical; FID = flame ionization detector; GC = gas chromatograph; HC = hydrocarbons; IR = nondispersive infrared; LA = line absorption; Para. = paramagnetic; RS = Raman spectroscopy.
the gas temperature in a flame. Table 9.12 is similar except for air/O2/fuel studies and Table 9.13 is for O2/fuel studies. Table 9.14 is a compilation of the maximum reported ranges for the total heat flux at the given gas temperatures using the five methods of measuring heat flux that have been used in the flame impingement studies. Some studies (e.g., Hargrave, Fairweather, and Kilham [109,110]) used multiple methods to measure heat flux. Baukal and Gebhart [111] reviewed empirical correlations for the heat flux from impinging flames. Table 9.15 gives the maximum temperature ranges for studies for the various methods used to measure gas temperature. Remie et al. [112] used phosphor thermometry to measure the temperature distribution on a quartz plate with a flame impinging on it. Table 9.16 lists all of the flame impingement studies that used the line reversal method to measure gas temperature. Table 9.17 gives the details of experiments that used fine wire thermocouples to measure gas temperatures. Table 9.18 gives details of the experiments that used Pitot probes to measure gas velocity. Table 9.19 shows the studies that measured gas composition, which gases were measured, and what techniques were used to measure them. Various techniques have been used to visualize impinging flames. Many studies (e.g., Chander and Ray [36,94) used photography to study flame structure. Some studies (e.g., Dong, Leung, and Cheung [66]) measured the static pressure distribution at the target surface. Some studies (e.g., Fernandes and Leandro [91]) used microphones to study high speed pressure oscillations due to pulsating flame impingement. High speed video cameras have been used in some studies. Foat, Yap, and Zhang (2001) [113] investigated turbulent flames impinging on
© 2011 by Taylor and Francis Group, LLC
a flat disk and identified eight flame modes: (1) blown ring, (2) disc flames, (3) ring flame, (4) conical flames, (5) envelope flame, (6) cool central core flame, (7) detached conic flame, and (8) complex flame. Hsieh and Lin [114] used a high speed video camera to study flame stability for a wide range of air/methane flames impinging on a water-cooled flat disk. Fernandes and Leandro [91] used a high speed video camera to identify six types of structures for oscillating flames impinging on a flat plate: (1) conical, (2) envelope, (3) disc, (4) cool central core, (5) ring, and (6) side lifted. Schuller, Durox, and Candel [115] studied the noise produced by acoustically excited air/methane flames impinging on a water-cooled flat plate.
References
1. Baukal, C. E. Heat Transfer from Flame Impingement Normal to a Plane Surface. Saarbrücken, Germany: VDM Verlag/ Dr. Müller, 2009. 2. Baukal, C. E., Farmer, L. K., Gebhart, B., and Chan, I. “Heat Transfer Mechanisms in Flame Impingement Heating.” In 1995 International Gas Research Conference, Vol. II, edited by D. A. Dolenc, 2277–87. Rockville, MD: Government Institutes, 1996. 3. Giedt, W. H., Cobb, L. L., and Russ, E. J. “Effect of Hydrogen Recombination on Turbulent Flow Heat Transfer.” New York: ASME Paper 60-WA-256, 1960. 4. Conolly, R., and Davies, R. M. “A Study of Convective Heat Transfer from Flames.” International Journal of Heat and Mass Transfer 15 (1972): 2155–72.
236
5. Cappelli, M. A., and Paul, P. H. “An Investigation of Diamond Film Deposition in a Premixed Oxyacetylene Flame.” Journal of Applied Physics 67, no. 5 (1990): 2596–2602. 6. Rauenzahn, R. M. “Analysis of Rock Mechanics and Gas Dynamics of Flame-Jet Thermal Spallation Drilling.” PhD thesis, Massachusetts Institute of Technology, Cambridge, MA, 1986. 7. Augustine, C., Potter, J., Potter, R., and Tester, J. W. “Feasibility of Spallation Drilling in a High Pressure, High-Density, Aqueous Environment: Characterization of Heat Transfer from an H2-O2 Flame Jet.” Transactions: Geothermal Resources Council, Vol. 31, pp. 241–45, Annual Meeting of the Geothermal Resources Council 2007 “Renewables Baseload Energy: Geothermal Heat Pumps to Engineered Reservoirs,” Reno, NV, September 30–October 3, 2007. 8. Hustad, J. E., Jacobsen, M., and Sønju, O. K. “Radiation and Heat Transfer in Oil/Propane Jet Diffusion Flames.” Institute of Chemical Engineers Symposium Series 10, no. 129 (1992): 657–63. 9. Johnson, A. D., Ebbinhaus, A., Imanari, T., Lennon, S. P., and Marie, N. “Large-Scale Free and Impinging Turbulent Flame Jets: Numerical Modelling and Experiments.” Process Safety & Environmental Protection 75, no. 3 (1997): 145–51. 10. You, H.-Z. “Investigation of Fire Impingement on a Horizontal Ceiling. 2—Impingement and Ceiling-Jet Regions.” Fire & Materials 9, no. 1 (1985): 46–56. 11. Hou, S.-S., and Ko, Y.-C., “Effects of Heating Height on Flame Appearance, Temperature Field and Efficiency of an Impinging Laminar Jet Flame used in Domestic Gas Stoves.” Energy Conversion and Management 45, nos. 9–10 (2004): 1583–95. 12. Malikov, G. K., Lobanov, D. L., Malikov, K. Y., Lisienko, V. G., Viskanta, R., and Fedorov, A. G. “Direct Flame Impingement Heating for Rapid Thermal Materials Processing.” International Journal of Heat and Mass Transfer 44 (2001): 1751–58. 13. Larnicol, M., Crahay, J., and Renard, M. “Rapid Heating of Low Carbon Strips by Impinging Burners: Recent Developments and Synthesis.” Revue de Metallurgie 102, no. 5 (2005): 379–84. 14. Wagner, J., Kurek, H., Chudnovsky, Y., Malikov, G., and Lisienko, V. “Direct Flame Impingement for the Efficient and Rapid Heating of Ferrous and Nonferrous Shapes.” Proceedings of Materials Science and Technology 2005 Conference, Pittsburgh, PA, September 25–28, 2005, Vol. 3, pp. 23–33. 15. Eichelkraut, H., Heiler, H.-J., Domels, H. P., and Högner, W. “DFI Oxyfuel Process for Saving Energy and Improving the Performance and Quality of Continuous Strip Lines.” ThyssenKrupp Techforum no. 2 (2007): 10–15. 16. Sirrine, M. M. “Direct Flame Impingement: A Versatile, Reliable Heating Technique.” Heat Treating Progress 6, no. 1 (2006): 42–45. 17. Sullivan, N. P., Branch, M. C., Strobel, M., and Ulsh, M. “Transport Issues when Impinging Laminar Premixed Flame Jets on a Rotating Cylinder.” Proceedings of the Combustion Institute 28 (2000): 1405–11.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
18. Eibeck, P. A., Keller, J. O., Bramlette, T. T., and Sailor, D. J. “Pulse Combustion: Impinging Jet Heat Transfer Enhancement.” Combustion Science and Technology 94 (1993): 147–65. 19. Viskanta, R. “Enhancement of Heat Transfer in Industrial Combustion Systems: Problems and Future Challenges.” Proceedings of ASME/JSME Thermal Engineering Joint Conference, edited by J. R. Lloyd and Y. Kurosaki, Vol. 5, 161–73, New York: ASME, 1991. 20. Viskanta, R. “Heat Transfer to Impinging Isothermal Gas and Flame Jets.” Experimental Thermal and Fluid Science 6 (1993): 111–34. 21. Viskanta, R. “Convective and Radiative Flame Jet Impingement Heat Transfer.” International Journal of Transport Phenomena 1 (1998): 1–15. 22. Viskanta, R. “Overview of Flame Impingement Heat Transfer: Fundamentals and Applications.” Proceedings of the Fourth Baltic Heat Transfer Conference, August 25–27, 2003. In Advances in Heat Transfer Engineering, edited by B. Sunden and J. Vilemas, 143–58, New York: Begell House, 2003. 23. Chander, S., and Ray, A. “Flame Impingement Heat Transfer: A Review.” Energy Conversion and Management 46 (2005): 2803–37. 24. Milson, A., and Chigier, N. A. “Studies of Methane-Air Flames Impinging on a Cold Plate.” Combustion and Flame 21 (1973): 295–305. 25. Baukal, C. E., and Gebhart, B. “A Review of Semi-Analytic Solutions for Flame Impingement Heat Transfer.” International Journal of Heat and Mass Transfer 39, no. 14 (1996): 2989–3002. 26. Baukal, C. E., and Gebhart, B. “A Review of Flame Impingement Heat Transfer Studies—Part 1: Experi mental Conditions.” Combustion Science and Technology 104 (1995): 339–57. 27. Glumac, N. G., and Goodwin, D. G. “Diagnostic and Modeling of Strained Fuel-Rich Acetylene/Oxygen Flames Used for Diamond Deposition.” Combustion and Flame 105 (1996): 321–31. 28. Roy, S., Kulatilaka, W. D., Lucht, R. P., Glumac, N. G., and Hu, T. “Temperature Profile Measurements in the Near-Substrate of Low-Pressure Diamond-Forming Flames.” Combustion and Flame 130 (2002): 261–76. 29. Malikov, G. K., Lobanov, D. L., Malikov, Y. K., Lisienko, V. G., Viskanta, R., and Fedorov, A. G. “Experimental and Numerical Study of Heat Transfer in a Flame Jet Impingement System.” Journal of the Energy Institute 72 (1999): 2–9. 30. Hemsath, K. H. “A Novel Gas Fired Heating System for Indirect Heating.” In Fossil Fuel Combustion Symposium 1990, edited by S. Singh, 155–59. New York: ASME PD-Vol. 30, 1990. 31. Kilham, J. K. “Energy Transfer from Flame Gases to Solids.” Third Symposium on Combustion and Flame and Explosion Phenomena, 733–40. Baltimore, MD: The Williams and Wilkins Co., 1949. 32. Jackson, E. G., and Kilham, J. K. “Heat Transfer from Combustion Products by Forced Convection.” Industrial and Engineering Chemistry Process Design and Development 48, no. 11 (1956): 2077–79.
Flame Impingement Measurements
33. Anderson, J. E., and Stresino, E. F. “Heat Transfer from Flames Impinging on Flat and Cylindrical Surfaces.” Journal of Heat Transfer 85, no. 1 (1963): 49–54. 34. Hustad, J. E., and Sønju, O. K. “Heat Transfer to Pipes Submerged in Turbulent Jet Diffusion Flames.” In Heat Transfer in Radiating and Combusting Systems, 474–90. Berlin: Springer-Verlag, 1991. 35. Hustad, J. E., Røkke, N. A., and Sønju, O. K. “Heat Transfer to Pipes Submerged in Lifted Buoyant Diffusion Flames.” In Experimental Heat Transfer, Fluid Mechanics, and Thermodynamics, 1991, edited by J. F. Keffer et al., 567–74. New York: Elsevier, 1991. 36. Chander, S., and Ray, A. “Heat transfer characteristics of laminar methane/air flame impinging normal to a cylindrical surface.” Experimental Thermal and Fluid Science 32 (2007): 707–21. 37. Davies, D. R. “Heat Transfer from Working Flame Burners.” BS thesis, University of Salford, U.K., 1979. 38. Sibulkin, M. “Heat Transfer Near the Forward Stagnation Point of a Body of Revolution.” Journal of Aeronautical Science and Technologies 19 (1952): 570–71. 39. Zhang, Y., and Bray, K. N. C. “Characterization of Impinging Jet Flames.” Combustion and Flame 116 (1999): 671–74. 40. Shorin, S. N., and Pechurkin, V. A. “Effectivnost’ teploperenosa na poverkhnost’ plity ot vysokotemperaturnoi strui produktov sjoraniya razlichnykh gazov.” Teoriya i Praktika Szhiganiya Gaza 4 (1968): 134–43. 41. Kremer, H., Buhr, E., and Haupt, R. “Heat Transfer from Turbulent Free-Jet Flames to Plane Surfaces.” In Heat Transfer in Flames, edited by N. H. Afgan and J. M. Beér, 463–72. Washington, DC: Scripta Book Company, 1974. 42. Dong, L. L., Leung, C. W., and Cheung, C. S. “Heat Transfer Characteristics of Premixed Butane/Air Flame Jet on an Inclined Flat Surface.” Heat & Mass Transfer 39 (2002): 19–26. 43. Woodruff, L. W., and Giedt, W. H. “Heat Transfer Measurements From a Partially Dissociated Gas With High Lewis Number.” Journal of Heat Transfer 88 (1966): 415–20. 44. Beér, J. M., and Chigier, N. A. “Impinging Jet Flames.” Combustion and Flame 12 (1968): 575–86. 45. Mohr, J. W., Seyed-Yagoobi, J., and Page, R. H. “Tempe rature, Heat Transfer and Pressure Measurements from an Impinging Radial Jet Reattachment Flame.” Proceedings of ASME/JSME Thermal Engineering Conference, Vol. 3, 127–34, 1995. 46. Mohr, J. W., Seyed-Yagoobi, J., and Page, R. H. “Heat Transfer from a Pair of Radial Jet Reattachment Flames.” In Combustion and Fire, ASME Proceedings of the 31st National Heat Transfer Conference, Vol. 6, edited by M. McQuay, K. Annamalai, W. Schreiber, D. Choudhury, E. Bigzadeh, and A. Runchal, HTD, Vol. 328, 11–17, New York, 1996. 47. Mohr, J. W., Seyed-Yagoobi, J., and Page, R. H. “Combustion Measurements from an Impinging Radial Jet Reattachment Flame.” Combustion and Flame 106 (1996): 69–80.
© 2011 by Taylor and Francis Group, LLC
237
48. Mohr, J. W., Seyed-Yagoobi, J., and Page, R. H. “Heat Transfer from a Pair of Radial Jet Reattachment Flames.” Journal of Heat Transfer 119 (1997): 633–35. 49. Wu, J., Seyed-Yagoobi, J., and Page, R. H. “Heat Transfer and Combustion Characteristics of an Array of Radial Jet Reattachment Flames.” Combustion and Flame 125 (2001): 955–64. 50. Tuttle, S. G., Webb, B. W., and McQuay, M. Q. “Convective Heat Transfer from a Partially Premixed Impinging Flame Jet. Part I: Time-Averaged Results.” International Journal of Heat and Mass Transfer 48 (2005): 1236–51. 51. Tuttle, S. G., Webb, B. W., and McQuay, M. Q. “Convective Heat Transfer from a Partially Premixed Impinging Flame Jet. Part II: Time-Resolved Results.” International Journal of Heat and Mass Transfer 48 (2005): 1252–66. 52. Vizioz, J.-P., and Lowes, T. M. “Convective Heat Transfer from Impinging Flame Jets.” International Flame Research Foundation report F 35/a/6, IJmuiden, the Netherlands, 1971. 53. Nawaz, S. “Heat Transfer from Oxygen Enriched Methane Flames.” PhD thesis, The University of Leeds, UK, 1973. 54. Fairweather, M., Kilham, J. K., and Nawaz, S. “Stagnation Point Heat Transfer from Laminar, High Temperature Methane Flames.” International Journal of Heat and Fluid Flow 5, no. 1 (1984): 21–27. 55. Ivernel, A., and Vernotte, P. “Etude expérimentale de l’amélioration des transferts convectis dans les fours par suroxygénation du comburant.” Revue Generale de Thermique 18, nos. 210–211 (1979): 375–91. 56. Kataoka, K., Shundoh, H., and Matsuo, H. “Convective Heat Transfer Between a Flat Plate and a Jet of Hot Gas Impinging On It. In Drying ’84, edited by A. S. Majumdar, 218–27. New York: Hemisphere/ Springer-Verlag, 1984. 57. Baukal, C. E., and Gebhart, B. “Heat Transfer from Oxygen-Enhanced/Natural Gas Flames Impinging Normal to a Plane Surface.” Experimental Thermal & Fluid Science 16, no. 3 (1998): 247–59. 58. Matsuo, M., Hattori, M., Ohta, T., and Kishimoto, S. “The Experimental Results of the Heat Transfer by Flame Impingement.” International Flame Research Foundation report F 29/1a/1, IJmuiden, the Netherlands, 1978. 59. Schulte, E. M. “Impingement Heat Transfer Rates from Torch Flames.” Journal of Heat Transfer 94 (1972): 231–33. 60. Fay, R. H. “Heat Transfer from Fuel Gas Flames.” Welding Journal, Research Supplement (1967): 380s–83s. 61. Davies, R. M. “Heat Transfer Measurements on Electrically-Boosted Flames.” 10th Symposium (Inter national) on Combustion, The Combustion Institute, Pittsburgh, PA, 755–66. 62. Dong, L. L., Cheung, C. S., and Leung, C. W. “Heat Transfer Characteristics of an Impinging Butane/Air Flame Jet of Low Reynolds Number.” Experimental Heat Transfer 14 (2001): 265–82. 63. Dong, L. L., Cheung, C. S., and Leung, C. W. “Heat Transfer from an Impinging Premixed Butane/Air Slot Flame Jet.” International Journal of Heat and Mass Transfer 45 (2002): 979–92.
238
64. Dong, L. L., Leung, C. W., and Cheung, C. S. “Heat Transfer Characteristics of a Pair of Impinging Rectangular Flame Jets.” Journal of Heat Transfer 125 (2003): 1140–46. 65. Dong, L. L., Leung, C. W., and Cheung, C. S. “Heat Transfer of a Row of Three Butane/Air Flame Jets Impinging on a Flat Plate.” International Journal of Heat and Mass Transfer 46 (2003): 113–25. 66. Dong, L. L., Leung, C. W., and Cheung, C. S. “Heat Transfer and Wall Pressure Characteristics of a Twin Premixed Butane/Air Flame Jets. International Journal of Heat and Mass Transfer 47 (2004): 489–500. 67. Zhao, Z., Wong, T. T., and Leung, C. W. “Impinging Premixed Butane/Air Circular Laminar Flame Jet: Influence of Impingement Plate on Heat Transfer Characteristics. International Journal of Heat and Mass Transfer 47 (2004): 5021–31. 68. Qi, J. A., Leung, C. W., Wong, W. O., and Probert, S. D. “Temperature-Field Measurements of a Premixed Butane/Air Circular Impinging-Flame Using ReferenceBeam Interferometry.” Applied Energy 83 (2006): 1307–16. 69. Dong, L. L., Cheung, C. S., and Leung, C. W. “Heat Transfer Characteristics of an Impinging Inversion Diffusion Flame Jet. Part I: Free Flame Structure.” International Journal of Heat and Mass Transfer 50 (2007): 5108–23. 70. Dong, L. L., Cheung, C. S., and Leung, C. W. “Heat Transfer Characteristics of an Impinging Inversion Diffusion Flame Jet. Part II: Impinging Flame Structure and Impingement Heat Transfer.” International Journal of Heat and Mass Transfer 50 (2007): 5124–38. 71. Saha, C., Ganguly, R., and Datta, A. “Heat Transfer and Emission Characteristics of Impinging Rich Methane and Ethylene Jet Flames.” Experimental Heat Transfer 21, no. 3 (2008): 169–87. 72. Cookson, R. A. “An Investigation of Heat Transfer from Flames.” PhD thesis, The University of Leeds, UK, 1960. 73. Cookson, R. A., and Kilham, J. K. “Energy Transfer from Hydrogen-Air Flames.” Ninth Symposium (International) on Combustion, 257–63. New York: Academic Press, 1963. 74. Smith, R. B., and Lowes, T. M. “Convective Heat Transfer from Impinging Tunnel Burner Flames: A Short Report on the NG-4 Trials.” International Flame Research Foun dation report F 35/a/9, IJmuiden, the Netherlands, 1974. 75. Rajani, J. B., Payne, R., and Michelfelder, S. “Convective Heat Transfer from Impinging Oxygen-Natural Gas Flames: Experimental Results from the NG5 Trials.” International Flame Research Foundation report F 35/a/12, IJmuiden, the Netherlands, 1978. 76. Kilham, J. K., and Purvis, M. R. I. “Heat Transfer from Hydrocarbon-Oxygen Flames.” Combustion and Flame 16 (1971): 47–54. 77. Purvis, M. R. I. “Heat Transfer from Normally Impinging Hydrocarbon Oxygen Flames to Surfaces at Elevated Temperatures.” PhD thesis, The University of Leeds, UK, 1974. 78. Fairweather, M., Kilham, J. K., and Mohebi-Ashtiani, A. “Stagnation Point Heat Transfer from Turbulent Methane-Air Flames.” Combustion and Science Technology 35 (1984): 225–38.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
79. Hargrave, G. K., and Kilham, J. K. “The Effect of Turbulence Intensity on Convective Heat Transfer From Premixed Methane-Air Flames.” Institute of Chemical Engineers Symposium Series 2, no. 86 (1984): 1025–34. 80. Veldman, C. C., Kubota, T., and Zukoski, E. E. “An Experimental Investigation of the Heat Transfer from a Buoyant Gas Plume to a Horizontal Ceiling: Part 1. Unobstructed Ceiling.” Washington, DC: National Bureau of Standards report NBS-GCR-77-97, 1975. 81. Kilham, J. K., and Purvis, M. R. I. “Heat Transfer from Normally Impinging Flames.” Combustion and Science Technology 18 (1978): 81–90. 82. Buhr, E., Haupt, G., and Kremer, H. “Heat Transfer from Impinging Turbulent Jet Flames to Plane Surfaces.” In Combustion Institute European Symposium 1973, edited by F. J. Weinberg, 607–12. New York: Academic Press, 1973. 83. Buhr, E. “Über den Wärmeflub in Staupunkten von turbulenter Freistrahl-Flammen an gekühlten Platten.” PhDthesis, University of Trier-Kaiserslautern, Kaiserslautern, Germany, 1969. 84. Reed, T. B. “Heat-Transfer Intensity from Induction Plasma Flames and Oxy-Hydrogen Flames.” Journal of Applied Physics 34, no. 8 (1963): 2266–69. 85. Sze, L. K., Cheung, C. S., and Leung, C. W. “Temperature Distribution and Heat Transfer Characteristics of an Inverse Diffusion Flame with Circumferentially Arran ged Fuel Ports.” International Journal of Heat and Mass Transfer 47 (2004): 3119–29. 86. Ng, T. K., Leung, C. W., and Cheung, C. S. “Experimental Investigation on the Heat Transfer of an Impinging Inverse Diffusion Flame.” International Journal of Heat and Mass Transfer 50, nos. 17–18 (2007): 3366–75. 87. Zhen, H. S., Leung, C. W., and Cheung, C. S. “Heat Transfer from a Turbulent Swirling Inverse Diffusion Flame to a Flat Surface.” International Journal of Heat and Mass Transfer 52, nos. 11–12 (2009): 2740–48. 88. Kwok, L. C., Leung, C. W., and Cheung, C. S. “Heat Transfer Characteristics of Slot and Round Premixed Impinging Flame Jets.” Experimental Heat Transfer 16 (2003): 111–37. 89. Huang, X. Q., Leung, C. W., Chan, C. K., and Probert, S. D. “Thermal Characteristics of a Premixed Impinging Circular Laminar-Flame Jet with Induced Swirl.” Applied Energy 83, no. 4 (2006): 401–11. 90. Zhao, Z., Yuen, D. W., Leung, C. W., and Wong, T. T. “Thermal Performance of a Premixed Impinging Circular Flame Jet Array with Induced-Swirl.” Applied Thermal Engineering 29 (2009): 159–66. 91. Fernandes, E. C., and Leandro, R. E. “Modeling and Experimental Validation of Unsteady Impinging Flames.” Combustion and Flame 146 (2006): 674–86. 92. Fells, I., and Harker, J. H. “An Investigation into Heat Transfer from Unseeded Propane-Air Flames Augmented with D.C. Electrical Power.” Combustion and Flame 12 (1968): 587–96. 93. Kwok, L. C., Leung, C. W., and Cheung, C. S. “Heat Transfer Characteristics of an Array of Impinging PreMixed Slot Flame Jets.” International Journal of Heat and Mass Transfer 48 (2005): 1727–38.
Flame Impingement Measurements
94. Chander, S., and Ray, A. “Heat Transfer Characteristics of Three Interacting Methane/Air Flame Jets Impinging on a Flat Surface.” International Journal of Heat and Mass Transfer 50 (2007): 640–53. 95. Chander, S., and Ray, A. “An Experimental and Numerical Study of Stagnation Point Heat Transfer for Methane/Air Laminar Flame Impinging on a Flat Surface.” International Journal of Heat and Mass Transfer 51 (2008): 3595–3607. 96. Rigby, J. R., and Webb, B. W. “An Experimental Investi gation of Diffusion Flame Jet Impingement Heat Trans fer.” Proceedings of the ASME/JSME Thermal Engineering Conference, Vol. 3, 117–26. New York: ASME, 1995. 97. Hemeson, A. O., Horsley, M. E., Purvis, M. R. I., and Tariq, A. S. 1984, “Heat Transfer from Flames to Convex Surfaces.” Institute of Chemical Engineers Symposium Series 2, no. 86 (1984): 969–78. Rugby, UK: Published by the Institute of Chemical Engineers. 98. Horsley, M. E., Purvis, M. R. I., and Tariq, A. S. “Convective Heat Transfer From Laminar and Turbulent Premixed Flames.” In Heat Transfer 1982, edited by U. Grigull, E. Hahne, K. Stephan, and J. Straub, Vol. 3, 409–15. Washington, DC: Hemisphere, 1982. 99. Tariq, A. S. “Impingement Heat Transfer From Turbulent and Laminar Flames.” PhD thesis, Portsmouth Polytechnic, Hampshire, UK, 1982. 100. Posillico, C. J. “Raman Spectroscopic and LDV Measurements of a Methane Jet Impinging Normally on a Flat Water-Cooled Boundary.” PhD thesis, Polytechnic Institute of New York, New York, 1986. 101. Kilham, J. K., and Dunham, P. G. “Energy Transfer from Carbon Monoxide Flames.” 11th Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 899–905. 102. Dunham, P. G. “Convective Heat Transfer from Carbon Monoxide Flames.” PhD thesis, The University of Leeds, UK, 1963. 103. Conolly, R. “A Study of Convective Heat Transfer from High Temperature Combustion Products.” PhD thesis, University of Aston, Birmingham, UK, 1971. 104. Baukal, C. E., and Gebhart, B. “Surface Condition Effects on Flame Impingement Heat Transfer.” Experimental Thermal & Fluid Science 15 (1997): 323–35.
© 2011 by Taylor and Francis Group, LLC
239
105. Hoogendoorn, C. J., Popiel, C. O., and van der Meer, T. H. “Turbulent Heat Transfer on a Plane Surface in Impingement Round Premixed Flame Jets.” Proceedings of 6th International Heat Transfer Conference, Toronto, Vol. 4, 107–12, 1978. 106. Popiel, C. O., van der Meer, T. H., and Hoogendoorn, C. J. “Convective Heat Transfer on a Plate in an Impinging Round Hot Gas Jet of Low Reynolds Number.” International Journal of Heat and Mass Transfer 23 (1980): 1055–68. 107. Van der Meer, T. H. “Stagnation Point Heat Transfer from Turbulent Low Reynolds Number Jets and Flame Jets.” Experimental Thermal and Fluid Science 4 (1991): 115–26. 108. Hargrave, G. K. “A Study of Forced Convective Heat Transfer from Turbulent Flames.” PhD thesis, University of Leeds, UK, 1984. 109. Hargrave, G. K., Fairweather, M., and Kilham, J. K. “Forced Convective Heat Transfer from Premixed Flames: Part 1: Flame Structure.” International Journal of Heat and Fluid Flow 8, no. 1 (1987): 55–63. 110. Hargrave, G. K., Fairweather, M., and Kilham, J. K. “Forced Convective Heat Transfer from Premixed Flames: Part 2: Impingement Heat Transfer.” International Journal of Heat and Fluid Flow 8, no. 2 (1987): 132–38. 111. Baukal, C. E., and Gebhart, B. “A Review of Empirical Flame Impingement Heat Transfer Correlations.” International Journal of Heat and Fluid Flow 17, no. 4 (1996): 386–96. 112. Remie, M. J., Särner, G., Cremers, M. F. G., Omrane, A., Schreel, K. R. A. M. Aldén, L. E. M., and de Goey, L. P. H. “Heat-Transfer Distribution for an Impinging Laminar Flame Jet to a Plate.” International Journal of Heat and Mass Transfer 51 (2008): 3144–52. 113. Foat, T., Yap, K. P., and Zhang, Y. “The Visualization and Mapping of Turbulent Premixed Impinging Flames.” Combustion and Flame 125 (2001): 839–51. 114. Hsieh, W.-D., and Lin, T.-H. “Methane Flame Stability in a Jet Impinging onto a Wall.” Energy Conversion and Management 46, no. 5 (2005): 727–39. 115. Schuller, T., Durox, D., and Candel, S. “Dynamics of and Noise Radiated by a Perturbed Impinging Premixed Jet Flame.” Combustion and Flame 128 (2002): 88–110.
10 Physical Modeling in Combustion Systems Christopher Q. Jian Contents 10.1 Introduction.................................................................................................................................................................. 241 10.2 Basics of Similitude Theory........................................................................................................................................ 242 10.2.1 Geometric Similarity..................................................................................................................................... 242 10.2.2 Kinematic Similarity..................................................................................................................................... 242 10.2.3 Dynamic Similarity....................................................................................................................................... 242 10.2.3.1 Buckingham Π Theorem (1914)................................................................................................... 243 10.2.4 Principle of Relaxation and Self-Similar Flow Regime............................................................................ 243 10.3 Physical Model Flow Measurement and Visualization Techniques.................................................................... 244 10.3.1 Flow Measurement........................................................................................................................................ 244 10.3.1.1 Velocity Measurement................................................................................................................. 245 10.3.1.2 Pressure Measurement................................................................................................................ 245 10.3.1.3 Temperature Measurement......................................................................................................... 245 10.3.2 Flow Visualization......................................................................................................................................... 245 10.3.3 Mixing Measurement Using Thermal Energy Balance Method............................................................. 246 10.4 A John Zink COOLflowTM Physical Modeling Case Study.................................................................................... 247 10.4.1 Introduction.................................................................................................................................................... 247 10.4.2 Boiler Technical Data: (Basis for the Model).............................................................................................. 248 10.4.3 Modeling Criteria.......................................................................................................................................... 248 10.4.4 Scale Model Description............................................................................................................................... 248 10.4.5 Instrumentation............................................................................................................................................. 248 10.4.6 Modeling Procedure...................................................................................................................................... 248 10.4.7 Physical Modeling Results............................................................................................................................ 248 10.4.7.1 Mass Flow Distribution............................................................................................................... 248 10.4.7.2 Primary Air Velocity Distribution............................................................................................. 249 10.4.7.3 Burner Exit Peripheral Air Velocity Distribution.................................................................... 250 10.4.8 Summary of Case Study............................................................................................................................... 250 References................................................................................................................................................................................. 250
10.1 Introduction From geological studies to aerospace engineering, physical modeling has been widely used in the industry to study complex fluid dynamics where engineering calculations or computational fluid dynamics are deemed either unreliable (the former) or uneconomical (the latter). In the field of combustion, physical modeling is employed in studying flow distribution involving combustion air, over-fire air (OFA), and flue gas recirculation (FGR) as well as isothermal flows in combustion chambers of furnaces, boilers, heat recovery and steam
generators (HRSG), and so on. Physical modeling is often used to study flow patterns prior to the commission of new furnaces and boilers to gain a better understanding of the flow characteristics and the interactions between various flow streams inside the combustion chamber to fine tune operating strategies and parameter settings. For burners with common windbox or furnaces with a large number of burners (e.g., 520 burners) connected with extensive ductwork, physical modeling is routinely used to identify flow maldistributions and to engineer flow solutions through the use of internals such as turning vanes, baffles, splitters, kickers, and so on to ensure desired flow distribution and flow patterns.
241 © 2011 by Taylor and Francis Group, LLC
242
Industrial Combustion Testing
Examples of applications of physical modeling in combustion systems include:
1. Air distribution in a single burner windbox 2. Air distribution in multiple burners sharing a common windbox 3. Distribution of OFA 4. Distribution and mixing of FGR 5. Determine and optimize primary, secondary, and tertiary air distribution 6. Air mass flow distribution among multiple burners with individual ductwork 7. Gas turbine exhaust flow distribution and duct burner auxiliary firing patterns 8. Combustion aerodynamics and partial load performance in furnaces and boilers 9. Combustion air and FGR pressure loss minimization 10. Air flow management in natural draft combustion systems
Construction of a scale model must be accompanied with an analysis to determine test conditions that ensure the test results from the scale model are representative of the processes in the prototype. In combustion applications, although most of the processes are inherently at elevated temperatures, physical modeling is usually carried out under isothermal conditions. Isothermal physical modeling technique is based on the principle of relaxation. Under this principle, the variables that are important for the phenomena under study are stressed. The variables that are stressed are duplicated as necessary to obtain a representative result. No scale physical model can be an exact model of the reality unless an exact full-scale prototype is made. However, by using accurate correlations the modeling work can provide a good qualitative understanding of the fluid dynamics in the prototype. This chapter attempts to answer the question: How does one ensure that the scale model test results are representative of the actual processes in the prototype?
Wp
Lp
Hp Sp
Wm
Lm
αp
Sm
Hm αm
Figure 10.1 Illustration of geometric similarity.
10.2.1 Geometric Similarity Geometric similarity requires that the scale physical model is dimensionally similar to the prototype. Such similarity exists between the scale model and the prototype if the ratio of all corresponding dimensions and all angles in the model and prototype are equal. Figure 10.1 illustrates the geometric similarity between a prototype and a scale model. Mathematically, geometric similarity can be expressed as:
Lp Lm Lp Lm Lp Lm = , = , = , and α p = α m . (10.1) Wp Wm Hp Hm Sp Sm
10.2.2 Kinematic Similarity Kinematic similarity is the similarity of fluid flow behavior in terms of time within the similar geometries. Kinematic similarity requires that the motion of fluids of both the scale model and prototype undergo similar rate of change (velocity, acceleration, etc.). This similarity criterion ensures that streamlines in both the scale model and prototype are geometrically similar and spatial distributions of velocity are also similar. Mathematically, kinematic similarity can be expressed as:
Vp Lp/Tp = Velocity ratio, Vm Lm/Tm ap Lp/Tp2 = Acceleration ratio. am Lm/Tm2
(10.2)
(10.3)
10.2.3 Dynamic Similarity
10.2 Basics of Similitude Theory The theoretical basis of physical modeling is the similitude theory or similarity theory [1]. The basic requirements to achieve similitude are: (1) geometric similarity, (2) kinematic similarity, and (3) dynamic similarity.
© 2011 by Taylor and Francis Group, LLC
Dynamic similarity ensures that the ratios of all forces, on the fluid flow and boundaries, in the prototype and scale model are the same and can be expressed as constants. Ratios of forces in fluid flows are often expressed in terms of dimensionless numbers. These dimensionless numbers are derived using what is called dimensional analysis using the Buckingham Π theorem.
Physical Modeling in Combustion Systems
243
10.2.3.1 Buckingham Π Theorem (1914)*
heat exchangers. These flows are usually subsonic with a Mach number of less than 0.3 and can be considered incompressible flows [2]. The dimensionless parameter that characterizes such flows is known as the Reynolds number (Re). The physical interpretation of a Reynolds Number is that it is the ratio of two forces in the fluid flow, namely, the inertia force and the viscous force or friction force. In mathematical terms, the Reynolds number is defined as
The Buckingham Π theorem is a key theorem in dimensional analysis. The theorem states that if we have a physically meaningful equation involving a certain number of physical variables (e.g., n), and these variables are expressible in terms of k independent fundamental physical variables (such as length, mass, time, etc.), then the original equation is equivalent to an equation involving at set of p = n − k dimensionless variables constructed from the original variables. In mathematical terms, if we have a physically meaningful equation such as
f (q1 , q2 ,..., qn ) = 0,
g(π1, π2, …, πn) = 0,
(10.5)
where the πi are dimensionless parameters constructed from the qi by p = n − k equations of the form
π i = q1m1 q2m2 ... qnmn ,
(10.6)
where the exponents mi are rational numbers (they can always be taken to be integers: just raise it to a power to clear denominators). The use of the πi as the dimensionless parameters was introduced by Edgar Buckingham in his original 1914 paper on the subject from which the theorem draws its name. The theorem provides a method for computing sets of dimensionless parameters from the given variables, even if the form of the equation is still unknown. However, the choice of dimensionless parameters, using this nondimensionalization scheme, is not unique: Buckingham’s theorem only provides a way of generating sets of dimensionless parameters, and will not choose the most physically meaningful dimensionless parameters. 10.2.4 Principle of Relaxation and Self-Similar Flow Regime In industrial combustion systems, fluid flows typically involve combustion air, fuel, FGR, and products of combustion (POC) in the combustion chamber and various * Edgar Buckingham (1867–1940) was educated at Harvard and Leipzig, and worked at the U.S. National Bureau of Standards (now the National Institute of Standards and Technology or NIST) from 1905 to 1937. His fields of expertise included soil physics, gas properties, acoustics, fluid mechanics, and blackbody radiation.
© 2011 by Taylor and Francis Group, LLC
Re =
inertia force F = i , viscous force Fv
(10.7)
(10.4)
where the qi are the n physical variables, and they are expressed in terms of k independent physical units, then the above equation can be restated as
where the inertia force and viscous force can be derived from dimensional analysis and the Buckingham Π theorem: mV 2 = ρD2V 2 , D
Inertial force: Fi =
Viscous force: Fv = τD2 =
µVD2 = µVD. D
(10.8) (10.9)
Substituting Equations 10.8 and 10.9 into Equation 10.7, the Reynolds number becomes
Re =
Fi ρVD VD = = , Fv µ υ
(10.10)
where m = mass D = length dimension (or hydraulic diameter) V = flow velocity ρ = fluid density µ = dynamic viscosity υ = kinematic viscosity τ = shear stress. The dynamic similarity requires that in two geometrically and kinematically similar systems, the Reynolds number must be the same. With the constraints on length scale (satisfying geometric similarity) and velocity ratio (satisfying kinematic similarity), an identical Reynolds number can be obtained only by a dramatic change in the properties of the fluid. Since the scale model study typically uses air as the working medium and the properties of air do not change from the scale model to the prototype, the prototype Reynolds number cannot be achieved in the scale model. The solution to this complete similarity requirement is to apply a modeling technique based on the principle of relaxation. Under this principle, the variables that are important for the phenomena under study
244
Industrial Combustion Testing
1. Assure desired flow patterns such as flow uniformity and minimization of flow recirculation. 2. Improve mixing performance of two or more flow streams. 3. Achieve equal flow mass distribution among multiple outlets. 4. Reduce total system pressure drop to reduce energy consumption or increase capacity.
Flows that develop a state that depends only on the local flow quantities, such as the local value of the mean velocity and the flow resistance, are said to be self-similar or self-preserving. This state of flow is present in the turbulent flow regime when sufficiently high Reynolds numbers are achieved. A majority of industrial combustion systems operate in this flow regime. When the scale model and prototype are both operating in the selfsimilar flow regime, they will manifest the same flow patterns and pressure drop coefficient despite different absolute local flow quantities. The flow pressure drop coefficient (Cp) is defined as the ratio between the static pressure and the dynamic pressure: Cp =
static pressure p − p2 = 1 , 1 dynamic pressure 2 ρV V 2
(10.11)
static pressure = p1 − p2 1 dynamic pressure = ρV 2. 2
3 2.5 Euler number (Eu)
are stressed. The variables that are stressed are duplicated as necessary to obtain a representative result. To ensure that the scale model results are representative of the prototype, the concept of self-similar or self-preserving flow regime is applied [3]. The objectives of performing scale model studies can be summarized as follows:
2 1.5 Recr = 10.5 × 104
1 0.5 0
0
5
10 15 Reynolds number (Re) × 10–4
20
Figure 10.2 Typical relationship of Eu = f(Re).
When the flow is in the self-similar regime, the Euler number becomes more or less a constant independent of the Reynolds. Figure 10.2 shows the relationship between the Reynolds number and Euler number for a particular flow category. In this particular system, if the Reynolds number is greater than the critical Reynolds number Recr = 10.5 × 104, the Euler number becomes a constant of approximately 0.65. Since the flow pressure drop coefficient is twice the Euler number (Equation 10.12), a constant Euler number indicates that the pressure drop coefficient remains a constant at and beyond the critical Reynolds number. However, the critical Reynolds number that ensures selfsimilar flow state is system specific. As a general guide, for combustion air ductwork and FGR ducts, Recr should be maintained at > 5 × 104. For burners and combustion chambers, the recommended Recr is > 2 × 105. In summary, in two geometrically and kinematically similar systems, when the flows are in the self-similar state, the flow characteristics such as flow patterns and streamlines are similar and the pressure drop coefficients in both the scale model and prototype are the same.
The pressure drop coefficient can also be expressed as
Cp =
p1 − p2 = 2 Eu, 1 ρV 2 2
(10.12)
where
10.3 Physical Model Flow Measurement and Visualization Techniques 10.3.1 Flow Measurement
Eu =
p1 − p2 = f (Re) ρV 2
(10.13)
is the Euler number, which is the ratio between pressure drop (flow friction) and twice the dynamic pressure. The Euler number is a function of the Reynolds number.
© 2011 by Taylor and Francis Group, LLC
One of the advantages of physical modeling is the direct measurements and observation of the fluid flow behavior inside the scale model that represent a prototype’s field operation. Typical flow measurements include velocity, pressure, and temperature. In some cases, species monitoring is also used to assess mixing performance of the scale model.
245
Physical Modeling in Combustion Systems
10.3.1.1 Velocity Measurement A variety of instruments are available for measuring velocity. For isothermal physical modeling, velocity is usually measured using: • • • •
Pitot tube and a wide range of its variants Precision vane anemometer Hot wire anemometer Laser-Doppler velocimetry
The laser-doppler velocimetry (LDV) is also applicable for velocity measurements at elevated temperature such as the region inside a flame zone. The LDV requires that the flow field be seeded with tiny particles. For high temperature measurements, high melting point solid particles such as high alumina oxide particles on the order of 1 ~ 2 microns are used. Figure 10.3 shows the velocity measurements using a Pitot tube and a digital manometer during a scale physical model study campaign. A detailed description of the various measurement apparatus is out of the scope of this chapter. 10.3.1.2 Pressure Measurement Many scale model measurements include quantitative measurement of pressure drops throughout the system. Maximum allowable system pressure drop is frequently a major constraint from the equipment manufacturer or end user when flow correction solutions are sought. System pressure drop directly relates to the operating cost due to blower/fan power consumption and the hydraulic head available for downstream power or steam generation. For scale model testing, pressure
easurements are usually performed using the followm ing instruments: • Pressure gauge • Manometer • Pressure transducer 10.3.1.3 Temperature Measurement Temperature measurements are required when testing scale model mixing performance using the thermal mixing technique (see Section 10.3.3 for more information). Temperature measurements typically involve the following instruments: • Thermometer • • • •
Thermocouple Thermistor Infrared thermometer Infrared or charge coupled device (CCD) camera
It should be pointed out that the application of a CCD camera for temperature measurement is a recent development. A CCD camera is an apparatus that is designed to convert optical brightness into electrical amplitude signals using a variety of CCDs, and then reproduces the image of a subject using the electric signals without time restriction. Coupled with robust image analysis software, the advantage of a CCD camera based temperature measurement system is that it can take unlimited point of measurements in the line of sight. Where most temperature measuring devices can only capture point data, the CCD camera system is capable of mapping an entire surface at the same instance. 10.3.2 Flow Visualization
Figure 10.3 Velocity measurement using a Pitot tube and manometer.
© 2011 by Taylor and Francis Group, LLC
Physical models in combustion related applications are typically built with transparent synthetic thermoplastic resin materials such as Plexiglas or Lexan. They offer the advantage of flow visualization for direct observation of complex flow patterns that are difficult, if not impossible, to visualize in the prototype. Efforts in prototype flow visualization have been made with limited success to gain a better understanding of flow patterns in critical areas. These flow visualization techniques typically involve setting up arrays of Tufts yarns as flow indicators and the use of sparks generated with burning sawdust. In many areas (such as pipes and ducts, heat exchangers, fan inlets and outlets, etc.), there are simply no observation stations that can provide the view of the entire flow field or can not perform flow visualization. The technique most commonly used for visualizing flow patterns in physical models is to inject smoke into
246
Industrial Combustion Testing
Center of recirculation
m1, T1
C1, C2, TA A
m2, T2
Up draft from main flow
Figure 10.5 Illustration of mixing.
Figure 10.4 Airflow/helium bubble analog flow visualization.
the model. While the smoke serves as a good tracer to establish the flow field, after a short period, the entire flow passage is filled with smoke that makes it difficult to track the streamlines in the flow domain. Furthermore, the smoke also tends to build up in the viewing area, which further hampers continued visual observation. At John Zink Company’s COOLflowTM Physical Modeling Facility, a different flow visualization technique has been successfully applied to scale model studies. In this technique, small quantities of minute helium bubbles are suspended in the airflow and used as tracer spheres. The helium bubbles are neutrally buoyant with a nominal density close to that of the air at ambient temperatures. Figure 10.4 shows the helium bubble flow visualization technique applied to a scale model. In order to visualize the movement of a large number of helium bubbles, collimated light sources are used to illuminate the bubbles at various planes. These collimated light sources are positioned perpendicular to the plane being viewed and the helium bubbles provide maximum light deflection at about 90º to the incident rays. Thus, the gross flow characteristics can be determined qualitatively and possible flow corrective measures evaluated. 10.3.3 Mixing Measurement Using Thermal Energy Balance Method Study of the mixing performance of multiple streams is often an integral part of the combustion process to improve combustion efficiency, reduce emission, redistribute energy absorption, and so on. Common gaseous streams include primary, secondary, and tertiary air,
© 2011 by Taylor and Francis Group, LLC
OFA, and FGR. The thermal energy balance method is the simplest way to determine the effectiveness of mixing. Under the steady-state condition, the temperature of the mixture is a function of the temperature and mass fraction of each individual stream. Take the two-stream mixing point A as an example (see Figure 10.5), we have C1 =
∆ mA 1 ∆ mA 2 , C2 = , and ∆mA = ∆mA1 + ∆mA 2 , ∆ mA ∆ mA
where ΔmA1, ΔmA2 represent the mass of stream m1 and m2 at point A and ΔmA is the total mass at point A. Under the steady-state condition, the energy balance at point A can be expressed as
c p 1T1 C1 + c p 2T2C2 = c pTA .
(10.14)
Equation 10.14 has two unknowns, namely, C1 and C2, hence one additional equation is needed to solve for the two unknowns. This can easily be achieved by varying the temperature of one or more individual streams. In the current example, T1, T2 is changed to T′1, T2′. Under this condition, the energy balance Equation 10.14 becomes
c ′p 1T1′C1 + c ′p 2T2′C2 = c ′pTA′ .
(10.15)
Solving Equations 10.14 and 10.15, the mass fraction of the two streams can be determined as follows:
C1 =
c pTA c p′ 2T2′ − c p 2T2 c p′ TA′ , c p 1T1c p′ 2T2′ − c p 2T2 c ′p 1T1′
C2 =
c p 1T1c p′ TA′ − c pTA c p′ 1T1′ . c p 1T1c ′p 2T2′ − c p 2T2 c ′p 1T1′
(10.16)
The temperatures of the individual streams and at point A are measured by thermal couples. For
247
Physical Modeling in Combustion Systems
combustion applications, air is the predominant medium used in scale model testing. The specific heats for air at various temperatures are readily available in most databases or handbooks. The thermal energy balance method is simple and fairly accurate in scale model testing. With an array of thermal couples, the thermal energy balance method can be further expanded to study the mixing intensity (dC/dx, dC/dy, dC/dz) in the combustion process. Based on the same working principle, mixing can also be studied using streams of different gaseous species. The downside of this method is that when more than two streams are present, species sampling becomes more difficult and time consuming. Figure 10.7 Scale physical model of the combustion air system.
10.4 A John Zink COOLflow TM Physical Modeling Case Study 10.4.1 Introduction The prototype was a field erected boiler with two natural gas firing burners. The burners shared a common windbox. The purpose of the study was to ensure that each burner receives the same amount of combustion air with even peripheral velocity distributions through the engineering of a baffle system based on John Zink’s COOLflowTM physical modeling study. Figures 10.6 and 10.7 show the burner being installed in the prototype and the scale physical model of the air supply duct and burner/windbox system. The objective of this scale model study was to develop a set of windbox baffles designed to provide an even combustion air mass flow distribution between the two burners, uniform peripheral flow entering each burner,
Figure 10.6 Burner to be installed on the prototype.
© 2011 by Taylor and Francis Group, LLC
and the elimination of tangential velocities within each burner. The major constraint in achieving these objectives was the air distribution arrangement and the internal dimensions of the windbox. Considering that air in the combustion process accounts for approximately 94% of the mass flow, numerous observations on boiler combustion systems have shown that the correct air distribution and peripheral entry condition is a key factor in the achievement of high performance (low NOx, low excess O2, and low CO). The purpose of each objective relates to a specific burner performance parameter is described next. Mass flow deviations are minimized to enable lower postcombustion O2, CO, and NOx. The lowest postcombustion O2 concentration possible is constrained by the burner most starved for air. This starved burner will generate a high CO concentration and consequently the total O2 must be raised to minimize the formation of CO in that burner. By equalizing the air flow to each burner and assuring that the fuel flow is equal, the O2 can be lowered until the CO starts to increase equally for all burners. Lower O2 has additional benefits of lower NOx formation and higher thermal efficiency. The equalization of the peripheral air velocity at the burner inlet will result in equal mass flow of air around and through the periphery of the swirler. The result of this equal air mass flow distribution through the swirler will be a fully developed and balanced air vortex at the center of the outlet of the swirler [4]. Flame stability and turndown of the burner depend on the condition of this vortex. Unequal air distribution results in an asymmetrical vortex leading to a flame that is more sensitive to pressure variations, limited in turndown ratio, sensitive to FGR on flame instability at lower loads, difficult in light-off by the igniter, and sensitive to flame scanning operation. The creation of swirling air is a fundamental requirement for many burners. Louvered burners are designed
248
Industrial Combustion Testing
to create a controlled swirl by rotating the entire air mass. Unfortunately, if the air has tangential velocities entering the burner, these velocities may act counter to the design of the burner, either working against the designed swirl of the burner (opposite swirl direction), or providing excessive swirl (same swirl direction). Uncontrolled swirl may create a problem at high turndown rates. At low loads (10%, for example), excess O2 is typically 11–13%. By having the entire mass swirling improperly, the fuel may be diluted to the point where flame stability becomes marginal.
velocity profile in the duct of the model. However, to measure the peripheral distribution of velocities around the inlet of burners, a custom made miniature Pitot tube was used. Since this Pitot tube was not calibrated, absolute values were not recorded. These readings are only used to compare velocities from one point to another and no units are given for the results. Since the geometry of the model burner is well known, the total flow through each burner can be determined by integrating the velocity profile over the measurement area. A straight-blade spinner was used to determine the swirl at the exit of burner.
10.4.2 Boiler Technical Data: (Basis for the Model) Boiler type: Number of burners: Maximum combustion air flow: (based on natural gas with 10% excess air) Nominal combustion air temperature: Model scale: Model air temperature: Model airflow:
Field erected 2 134,366 kg/hr 40°C 1/8 32°C 1937 kg/hr
10.4.3 Modeling Criteria The following criteria were applied to ensure optimum performance of the burners: • Mass flow differences for each burner shall be within ±2% of mean, in the model. • Peripheral air distribution velocity differences shall be less than ±10% of the mean. • Swirls shall be eliminated if possible or kept to a practical minimum. 10.4.4 Scale Model Description As shown in Figure 10.7, the model was built using acrylic sheets at a 1/8th scale. This included the combustion air ducting from the force draft fan outlet through the windbox and the burners. Cold flow modeling was performed using air at ambient temperature as the working fluid for the combustion air. Quantitative airflow measurements were taken using a Pitot tube. Observation of the flow field was performed using John Zink’s analog helium bubble flow visualization technique. 10.4.5 Instrumentation Extensive quantitative data were obtained by airflow modeling. A miniature calibrated Pitot tube, combined with an electronic manometer was used to measure the
© 2011 by Taylor and Francis Group, LLC
10.4.6 Modeling Procedure The model was initially run at MCR for a baseline condition. The velocity profiles were set to the design conditions at the model entrance. The mass flow readings, peripheral distribution, and swirl were then recorded. Baffle plates were then installed strategically within the model and the mass flow, peripheral distribution, and swirl were then remeasured. 10.4.7 Physical Modeling Results 10.4.7.1 Mass Flow Distribution Table 10.1 shows the mass flow data recorded during the baseline and after modification testing. This table also shows a significant disparity of mass flow rates between the two burners. Burner 1 severely starves for air with a 57.6% mass deficit. This can potentially lead to a number of operating issues including, for example, combustion stability, high concentration of unburned hydrocarbon and CO in the exhaust flue gases, and elongated flame that may cause hot spots on heat transfer tubes. On the other hand, Burner 2 has a high percentage of excess air. The availability of abundant oxygen in a high combustion zone with the presence of a large amount of nitrogen can lead to high NOx emissions. Table 10.1 Mass Flow Measurements Baseline Burner # 1 2
% of Average 42.4 157.6
After Installing Baffle Solutions Burner # 1 2
% of Average 100.3 99.7
249
Physical Modeling in Combustion Systems
The mass flow rates with the implementation of a baffle system are also shown in Table 10.1. The variations of mass flow between the two burners are corrected to within ± 0.3%.
Burner exit air velocity distribution burner #2 — baseline
8000 7000 6000 5000 4000 3000 2000 1000 0 –1000
10.4.7.2 Primary Air Velocity Distribution Pitot tube measurements of the axial velocities exiting the burner are shown in Figures 10.8 through 10.11. The numbers corresponding to the measurements are directly from an uncalibrated Pitot tube and do not represent the true local velocity magnitude. However, they do represent the relative velocity profiles across the burner exits. Burner exit air velocity distribution of burner #1 — baseline
9000 8000 7000 6000 5000 4000 3000 2000 1000 0 –1000
8000–9000 7000–8000 6000–7000 5000–6000 4000–5000 3000–4000 2000–3000 1000–2000 0–1000 –1000–0
Figure 10.10 (See color insert following page 424.) Exit air velocity distribution of burner #2—before modification. Burner exit air velocity distribution burner #2 — after modification
6000 5000 4000 3000
Figure 10.8 (See color insert following page 424.) Exit air velocity distribution burner #1—before modification.
2000
5000–6000 4000–5000 3000–4000 2000–3000 1000–2000 0–1000
1000 0
Burner exit air velocity distribution burner #1 — after modification
5000–6000 6000 5000 4000 3000 2000 1000
4000–5000 3000–4000 2000–3000 1000–2000 0–1000
0
Figure 10.9 (See color insert following page 424.) Exit air velocity distribution of burner #1—after modification.
© 2011 by Taylor and Francis Group, LLC
7000–8000 6000–7000 5000–6000 4000–5000 3000–4000 2000–3000 1000–2000 0–1000 –1000–0
Figure 10.11 (See color insert following page 424.) Exit air velocity distribution of burner #2—after modification.
Figures 10.8 and 10.10 show that the axial velocities exiting both burners are severely skewed toward one side of the burner exit. Such air flow distribution will cause parts of the burner to starve for air while other parts to operate with high excess air. The results will be similar to the mass flow maldistribution discussed earlier but on a more localized scale for each individual burner. Such uneven peripheral air distribution should be corrected to allow the burners to function properly. Figures 10.9 and 10.11 show the burner exit axial velocities with the baffle solution implemented. More uniform velocity profiles similar to that of a plug flow can be seen.
250
Industrial Combustion Testing
10.4.7.3 Burner Exit Peripheral Air Velocity Distribution
Model test results 12:00 16,000
Measurements of relative axial velocity distributions along the peripheries of the burners and swirl numbers are shown in Figures 10.12 and 10.13. Peripheral velocity distributions (the soild lines) at burner exits vary from + 21.1% from the mean velocity to −19.8%. With the baffle solution implement, the range of axial peripheral velocity deviation from the mean velocity is reduced + 2.7 to −3.9%. The baffle solution also eliminated the swirl in both burners.
14,000 10:30
12,000
1:30
10,000 8000 6000 4000 2000 0
9:00
3:00
10.4.8 Summary of Case Study This case study demonstrated the importance of air supply system designs in achieving optimal operations of the burners, and hence the boiler. Physical modeling is a powerful and effective technique for understanding
Baseline Solution Max pos deviation: 15.1% 2.3% Max neg deviation: –12.8% –3.3% Swirl number: 0.07 0.00
Model test results
Figure 10.13 Peripheral air velocity distribution of burner #2.
12,000 10,000
1:30
8000 6000 4000 2000 0
9:00
7:30
6:00 Peripheral velocity distribution Before correction After correction
12:00 14,000 10:30
4:30
7:30
3:00
the fluid flow and for developing solutions to the flow related issues. Based on the Buckingham theorem and the self-similar nature of turbulent flows, the solutions developed in the scale model can be directly translated to prototype design and operation. Even in today’s business world where computer simulations are widely used, physical modeling continues to demonstrate its value and power in improving the energy efficiency, environment performance, and product quality of a variety of industrial processes.
4:30
References 6:00 Peripheral velocity distribution Before correction After correction Figure 10.12 Peripheral air velocity distribution of burner #1.
© 2011 by Taylor and Francis Group, LLC
1. Schlichting, H. Boundary Layer Theory. New York: McGraw-Hill, 1979. 2. Panton, R. L. Incompressible Flow. New York: John Wiley & Sons, 1984. 3. Hinze, J. O. Turbulence. New York: McGraw-Hill, 1959. 4. Batchelor, G. K. An Introduction to Fluid Dynamics. Cambridge, UK: Cambridge University Press, 1983.
11 Virtual Testing Eddy Chui, Allan M. Runstedtler, and Adrian J. Majeski Contents 11.1 Introduction.................................................................................................................................................................. 251 11.2 Idea of Virtual Testing................................................................................................................................................. 252 11.3 Example 1: Virtual Testing of Burner Performance................................................................................................ 252 11.3.1 Context............................................................................................................................................................ 252 11.3.2 Test Facility..................................................................................................................................................... 252 11.3.3 Experimental Findings.................................................................................................................................. 253 11.3.4 Overview of Virtual Test Approach for Oxy-Coal Combustion............................................................. 254 11.3.5 Virtual Test Results........................................................................................................................................ 255 11.3.6 Additional Information Provided by Virtual Testing.............................................................................. 255 11.3.7 Value of Virtual Testing................................................................................................................................ 255 11.4 Example 2: Virtual Testing for Burner and Combustor Design Concepts.......................................................... 255 11.4.1 Context............................................................................................................................................................ 255 11.4.2 Scope of Virtual Investigation...................................................................................................................... 257 11.4.3 Burner C1: Level 1 Improvement................................................................................................................. 257 11.4.4 Impact of Combustor Size............................................................................................................................ 257 11.4.5 Burner C2: Level 2 Improvement................................................................................................................. 258 11.4.6 Demonstration of Virtual Test Results........................................................................................................ 259 11.5 Example 3: Virtual Testing to Identify Heater Operational Problem................................................................... 260 11.5.1 Context............................................................................................................................................................ 260 11.5.2 Virtual Test Approach................................................................................................................................... 260 11.5.3 Cause of Uneven Heat Distribution............................................................................................................ 260 11.5.4 Value of Virtual Testing................................................................................................................................ 261 11.6 Example 4: Virtual Testing to Guide Retrofit of Refinery Furnace....................................................................... 261 11.6.1 Context............................................................................................................................................................ 261 11.6.2 The Furnace.................................................................................................................................................... 261 11.6.3 Virtual Test Approach................................................................................................................................... 262 11.6.4 Virtual Test Results........................................................................................................................................ 262 11.6.5 Value of Virtual Test...................................................................................................................................... 263 11.7 Summary Remarks...................................................................................................................................................... 264 References................................................................................................................................................................................. 265
11.1 Introduction With rising energy cost and the pressing need to reduce greenhouse gas emissions to combat climate change, industrial processes involving combustion of fossil fuels are facing fundamental changes. Unconventional firing like oxy-firing is being seriously considered to generate exhaust streams high in CO2 concentration to facilitate potential capture and storage technologies. The CO2
neutral biofuels are viewed as options for usage even in large industrial processes (e.g., blast furnaces in steelmaking). Utilization of other opportunity fuels (e.g., petroleum coke) is desired to replace or blend with conventional fuels like coal, oil, and natural gas to reduce production costs and undesirable emissions simultaneously for economic and environmental reasons. These attempts involve unconventional burning of fuels and may require systematic changes to the design and operation of combustion equipment like burners, fireboxes, furnaces, and combustors.
251 © 2011 by Taylor and Francis Group, LLC
252
11.2 Idea of Virtual Testing Successful systematic changes to combustion equipment must be based on new knowledge and understanding to be generated through industrial combustion testing. While experimental testing from bench scale to industrial scale continues to be essential, parallel testing in the “virtual” or “computer-generated” environment should not be overlooked for cost and time effectiveness, especially with the significant advances in computer hardware and computer-based simulation technologies. Detailed descriptions of computational fluid dynamics (CFD) based simulation technologies for turbulent reacting flows in industrial combustion have been well documented in Baukal, Gershtein, and Li [1] and Henneke et al. [2] and will not be repeated here. In essence, CFD-based combustion modeling aims to solve the fundamental continuity and transport equations of momentum, enthalpy, and associated species involved in the chemical reactions of a given fuel and oxidant system. The process involves generating a computational grid to represent the actual system geometry in a discrete fashion. With modern computer aided design (CAD) tools, grids can be created to emulate the actual system with high fidelity using hundreds of thousands to millions of computational “control volumes.” The aforementioned fundamental equations are solved numerically in every control volume to provide information of turbulent fluid flow, chemical reactions, thermal characteristics, and even pollutant generation. To date, direct simulation (i.e., solving these fundamental equations without modeling) is possible only for simple fuel systems with limited geometry size and complexity. Typically for most practical industrial combustion systems, modeling is required to account for the interaction between turbulence and chemistry and the associated chemical reactions and heat transfer. Baukal, Gershtein, and Li [1] and Henneke et al. [2] address the various CFD modeling techniques and industrial applications with description of their assumptions and expected performance. Presently, CFD-based simulation technologies are by no means perfect in capturing all the physics in practical combustion systems. However, when they are properly used, especially in conjunction with careful experimental testing, they can provide highly valuable information not easily obtainable by measurements. Also, they are readily available via licensable commercial software packages from engineering software developers (for example, ANSYS supplying the FLUENT and CFX codes well known in the industrial combustion applications). CanmetENERGY, an energy research facility of the Canadian Government, has been applying the combined experimental-numerical approach to investigate industrial combustion systems for decades. Based on the experience of the authors at CanmetENERGY, the following
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
sections provide examples of how virtual testing can be implemented in (a) the pilot-scale study of oxy-firing, one of the advanced combustion technologies complementary to the major greenhouse gas reduction strategy of CO2 capture and storage (CCS), and (b) industrial heaters in petroleum refining. The focus is not on the findings in oxy-firing system or petroleum refining, but rather on the role of virtual testing in these particular investigations.
11.3 Example 1: Virtual Testing of Burner Performance 11.3.1 Context To produce a CO2 rich flue gas stream amenable for use or storage, oxy-fuel combustion typically involves replacing combustion air with pure oxygen, either partially or completely, and usually in combination with recycled flue gases in order to temper the flame. In this particular investigation, the concept of flue gas recycling was tested in a pilot-scale test facility [3]. 11.3.2 Test Facility Figure 11.1 provides a schematic of the CanmetENERGY pilot plant configuration, which consisted of four major systems: fuel delivery, combustion, flue gas treatment, and instrumentation and control [4]. The combustion system was a cylindrical, down-fired vertical combustor (0.61 m I.D. and 8.3 m long). It had a rated firing capacity of up to 0.3 MWth and could burn solid, gaseous, and liquid fuels. The combustor was lined with refractory to conserve heat in order to provide a realistic time–temperature history for burning coal particles. The burner (burner B, Figure 11.2) had a single register and was swirl stabilized with the primary stream being admitted to an annulus located within the inner diameter of the swirling secondary stream. The primary stream carried the dry pulverized coal to the flame and was made up of air or CO2 from a bulk storage vessel. Small amounts of O2 could be added to the primary stream to simulate real flue gas recycle conditions. The secondary stream could comprise ambient air alone or in combination with carbon dioxide from the bulk storage vessel, nitrogen, oxygen, or real recycled flue gases. The amount of oxygen mixed into the secondary stream was restricted to 28% (dry volume basis) for safety reasons. Pure oxygen was admitted directly at the burner through a series of holes arranged in the annulus located within the inner diameter of the secondary stream. To heat up the combustor prior to oxy-coal combustion tests, natural gas was fired down the center of the burner. The burner was expected to provide stable flames over a wide range of firing conditions [5].
253
Virtual Testing
Flue gas recycle
Heater
Oxygen cell O2 Mixing chamber
O2 storage
Direct oxygen
Oxygen cell O2
Primary stream
Mixing chamber
CO2 storage Crushed dry coal Fine coal silo Pulverizer
Heater
Air or CO2
Heater
Hot flue gas
Natural gas
Air fan
Coarse coal silo
Vent to stack
Secondary stream
Baghouse
CEM analyzers
Vertical combustor
Heater Air or CO2
Flue gas cooler O2 Oxygen cell
Electrostatic precipitator (ESP)
Condensing heat exchanger (CHX)
Cyclone
Coal feeder
CEM analyzers
Eductor
Heater
Flue gas fan
Flue gas recycle
Figure 11.1 Schematic of Canmet ENERGY Vertical Combustor Research Facility (VCRF). (From Chui, E. H., Douglas, M. A., and Tan, Y., Fuel, 82, 1201–10, 2003. With permission.)
Oxygen
Secondary stream
Natural Gas Natural Gas
Pulverized coal in primary stream
O2 Coal/Air/CO2 CO2/O2/N2/Air or Recycled Flue Gas Swirl generator Burner front view
Burner side view Burner B
Figure 11.2 Schematics of burner B configuration for the VCRF. (From Chui, E. H., Douglas, M. A., and Tan, Y., Fuel, 82, 1201–10, 2003. With permission.)
11.3.3 Experimental Findings Figure 11.3 shows the discrete measured values of temperature, oxygen, CO, and NO along the centerline of the combustor. The data indicated that high temperature was achieved relatively quickly and that there was a steady drop of O2 along the length of the combustor to the expected excess level with low CO levels throughout. This
© 2011 by Taylor and Francis Group, LLC
burner could provide stable, efficient oxy-coal combustion with the given flue gas recycling condition. However, the experimental data could not explain why the centerline O2 distribution had double peaks. Also, it was counterintuitive that this oxy-firing condition, devoid of molecular N2 and hence, thermal NO, could generate more NO (149 ng/J) than the baseline air-firing case (110 ng/J).
254
Industrial Combustion Testing
Burner B: dry recycle case; centerline temperature
O2 (vol. %DRY)
2000 1500 1000 500
0
1
2 3 4 Axial distance (m)
5
80000 60000 40000 20000 0
0
1
2
3
4
5
20 10 0
1
6
Axial distance (m)
2 3 4 Axial distance (m)
5
6
Burner B: dry recycle case; center-line NOx
4000
Measured Predicted
100000
Measured Predicted
30
0
6
Burner B: dry recycle case; center-line CO
120000 CO (ppmDRY)
40
Measured Predicted
NO (ppmDRY)
Temperature (°C )
2500
Burner B: dry recycle case; center-line O2
Measured Predicted
3000 2000 1000 0
0
1
2
3
4
5
6
Axial distance (m)
Figure 11.3 Comparison of measured and predicted temperature, oxygen, CO, and NO along combustor centerline for dry recycle case with burner B. (From Chui, E. H., Douglas, M. A., and Tan, Y., Fuel, 82, 1201–10, 2003. With permission.)
11.3.4 Overview of Virtual Test Approach for Oxy-Coal Combustion The virtual test approach of pulverised coal combustion was based on the works of Lockwood and colleagues [7,8,9]. Virtual test results (or model predictions) were obtained by solving the time-averaged conservation equations for the gas and coal particle phases. An Eulerian approach was used for the gas phase but the particle phase was treated in a Lagrangian manner. Based on the measured particle size distribution, coal particles of different sizes were injected into the virtual test domain from the primary gas stream outlet. Each particle trajectory, representative of a specific size group from the firing location, was tracked and its devolatilization and combustion history was recorded. The flow/particle interactions were handled by the “particle source in cell” method [10] where the exchanges in mass, momentum, and energy between phases were represented by appropriate sink/source terms in the gas/particulate equations. The standard k-ε method was used for turbulent fluid flow calculations. Devolatilization was simulated by a first-order single reaction model [11]. Volatile combustion was assumed to be controlled by the mixing rate of reactants (the eddybreakup model of Magnussen and Hjertager [12]). Char burning was governed by the chemical kinetic rate and
© 2011 by Taylor and Francis Group, LLC
the external diffusion rate of oxygen to the char surface [7]. The NOx formation was calculated using the approach proposed by Chui and Hughes [13]. The majority of NOx production in turbulent coal flames was formed through the oxidation of fuel-bound nitrogen (fuel NO) and was modeled by the eddy-dissipation concept. The oxidation of molecular nitrogen (thermal and prompt NO) and the heterogeneous reduction of NO by char were estimated based on their established chemical kinetic rates using the presumed probability density function (pdf) method. This approach of modeling or virtual testing coal combustion and NOx formation characteristics in air was computationally inexpensive enough to be readily applicable to large industrial scale combustion units and it has been shown to provide reasonably accurate predictions of the performance of utility boilers by CanmetENERGY [14]. This virtual test extended the above methodology to oxy-coal combustion atmo spheres (with elevated O2 and CO2 levels and low or nonexistent N2). The approach naturally adjusted to give the proper radiative properties of the new medium based on the concentrations of CO, H2O, and the substantially higher amount of CO2. Also, the thermal and prompt NOx formation mechanism was automatically deactivated in the absence of molecular nitrogen.
255
Virtual Testing
11.3.5 Virtual Test Results Figure 11.3 also shows the predicted values of temperature, oxygen, and CO (solid lines) together with the measured data. The close overall matching of prediction with experimental data validated the virtual test. The amount of NO in the flue gas stream going to the stack was predicted to be 141 ng/J, which compared very well with the measured value, 149 ng/J. Even the double peaks on the centerline O2 measurement were reproduced by the model.
11.3.7 Value of Virtual Testing
11.3.6 Additional Information Provided by Virtual Testing The simulation also provided more detailed description of the coal particle trajectories, temperature, and oxygen distributions inside the combustor (Figure 11.4). The model results clearly indicated that the coal particle distribution near the burner exit was highly nonuniform (Figure 11.4a). This peculiar fuel distribution resulted in a flame attachment toward one side of the combustor as shown by the temperature distribution across the midsection of the combustor (Figure 11.4b), and an asymmetric distribution of oxygen: high in the near burner region and low at the central core of the combustor where the flame was most intense (Figure 11.4c). This explains the double peaks of the O2 distribution along the centerline of the combustor as measured and predicted in Figure 11.3. Figure 11.5 shows the fast rate of volatile and fuel bound N evolution from the coal particles in the near burner region, with a substantial portion in the O2-rich portion of the combustor adjacent to the burner face, (a)
prior to the surface of equivalence ratio of 1 separating the O2-rich and fuel-rich regions in the combustor (Figure 11.5b). Figure 11.6 shows that a significant amount of NO was formed prior to the fuel-rich zone (equivalence ratio > 1), and it propagated along one side of the combustor toward the outlet. This explains the relatively high NO generation in this case of oxy-firing with flue gas recycling. The virtual test results provided explanations to both apparent anomalies raised in the experimental findings.
The virtual test results complemented the experimental findings to give a more complete understanding of the various physical phenomena in this oxy-firing investigation. The simulation also revealed that the absence of molecular nitrogen in oxy-firing could not guarantee lower NOx formation within the combustor. Furthermore, it highlighted the need for further research in oxygen management and staging techniques for the optimization of burner configurations applicable to oxyfiring with flue gas recycle.
11.4 Example 2: Virtual Testing for Burner and Combustor Design Concepts 11.4.1 Context The NOx reduction is an important consideration in oxyfuel combustion for CCS because any trace amount of NO
(b)
(c)
Temperature (°C)
O2 (Vol. %DRY)
1800
10
1700
9
1600
8
1500
7
1400
6
1300
5
1200
4
1100
3
1000
2
900
1
800
0
Figure 11.4 (See color insert following page 424.) Flame characteristics for dry recycle case with burner B: (a) coal particle trajectories, (b) temperature, and (c) oxygen distributions at combustor midsection. (From Chui, E. H., Douglas, M. A., and Tan, Y., Fuel, 82, 1201–10, 2003. With permission.)
© 2011 by Taylor and Francis Group, LLC
256
Industrial Combustion Testing
Figure 11.5 (See color insert following page 424.) Isosurface of volatile matter evolution at 10% of the maximum level for dry recycle case with burner B (left) and overlaid with the boundary of fuel-rich zone (shown on right with color red for equivalence ratio = 1). (From Chui, E. H., Douglas, M. A., and Tan, Y., Fuel, 82, 1201–10, 2003. With permission.)
(a)
(b) Equivalence ratio
NO (ppmDRY)
1.0
2000
0.9
1800
0.8
1600
0.7
1400
0.6
1200
0.5
1000
0.4
800
0.3
600
0.2
400
0.1
200
0.0
0
Figure 11.6 (See color insert following page 424.) Equivalence ratio (left) and NO (right) distributions across combustor midsection for dry recycle case using burner B. (From Chui, E. H., Douglas, M. A., and Tan, Y., Fuel, 82, 1201–10, 2003. With permission.)
© 2011 by Taylor and Francis Group, LLC
257
Virtual Testing
in the flue gas stream tends to accumulate in the vent portion of the product recovery process, which involves compression, dehydration, and refrigeration of the flue gas stream prior to pumping it into a reservoir. It will typically need to be discharged into the atmosphere and should be minimized. The investigation described below was a follow-up study from Example 1, using the same virtual testing tool to evaluate new burner and combustor design concepts to minimize NOx in oxy-coal combustion [6]. 11.4.2 Scope of Virtual Investigation Two versions of new burner design concepts were developed. Due to proprietary reasons, details of the new burners cannot be disclosed here. Each design was incorporated into a computational grid that also included the details of the combustor. Simulations of oxy-coal combustion and NOx formation were performed to determine the flame and NOx characteristics for each design. In addition, two different combustor sizes were also investigated using the model to quantify wall effects.
Down firing
11.4.3 Burner C1: Level 1 Improvement Figure 11.7 shows the contrast in temperature distribution over a cross section of the combustor between burner B (the burner from Example 1), and the new conceptual burner C1. The temperature plot, covering a 4.75 m long region starting from the end of the quarl section, indicates that the flame from burner C1 was more central, axisymmetric, and did not attach to one side of the combustor. Burner C1 also facilitated the creation of a large, distinct fuel-rich zone in the central core of the combustor with a low O2 concentration (< 3 vol.% dry as shown in Figure 11.8). As coal from the primary gas stream passed through this fuel-rich core in the middle of the furnace, fuel-bound nitrogen released during devolatilization did not transform to NO. Figure 11.9 shows a significant drop in NO throughout the f urnace and especially along the centerline when burner C1 was used. At the exit of the modeled furnace, the NO level was 55 ng/J, compared to 229 ng/J measured at the same location when burner B was used. (Please note that the 149 ng/J as reported in Example 1 above for burner B was measured after the recycle stream was removed.) With burner C1, the mean CO level at axial distance 5 m from quarl outlet was 361 ppm, which was lower than with burner B (725 ppm), indicating that burner C1 promoted better overall mixing of fuel and oxidant in addition to lowering NOx production. 11.4.4 Impact of Combustor Size At the beginning of the study, the size of the combustor was 0.61 m ID and could be too small to properly evaluate
© 2011 by Taylor and Francis Group, LLC
Burner: Furnace ID Temperature (°C) 800
1000
B : 0.6 m
1200
1400
C1 : 0.6 m
1600
1800
Figure 11.7 (See color insert following page 424.) Temperature distribution on furnace cross section for burners B and C1. (From Chui, E. H., Majeski, A. J., Douglas, M. A., Tan, Y., and Thambimuthu, K. V., Energy, 29, 1285–96, 2004. With permission.)
the performance of burner design concepts. To gain insights on wall effects, burner C1 was virtually tested via numerical modeling in a much larger 1.5 m ID chamber with identical input conditions. The flame changed shape from being central, axisymmetric, and steady as in Figure 11.7 to long and lazy, and it impinged on the furnace wall (Figure 11.10). In a less confining environment, the momentum of the flame from burner C1 was greatly reduced and the burner could no longer generate a favorable flow field for proper mixing of fuel and oxidant. Then a stacked chamber (1.0 m ID in the first 1.2 m; 0.61 m ID for the remaining portion) was chosen to test burner C1. The intent was to minimize the wall effects in the near burner region without unduly reducing the flame momentum. Figure 11.10 (middle picture) shows the resulting temperature distribution for C1 in the stacked furnace. This flame still attached to the furnace wall, indicating that the flame characteristics from burner
258
Industrial Combustion Testing
0
Axial distance (m)
1
2
3
Centerline distribution Burner : Furnace ID
4
B : 0.6 m C1 : 0.6 m B : 0.6 m 0
C1 : 0.6 m
0
8 16 24 32 40
20 O2 (Vol. % dry)
Figure 11.8 (See color insert following page 424.) Oxygen distribution on furnace cross section and centerline for burners B and C1. (From Chui, E. H., Majeski, A. J., Douglas, M. A., Tan, Y., and Thambimuthu, K. V., Energy, 29, 1285–96, 2004. With permission.) 0
Axial distance (m)
1
Peak 3950 ppm
2
C1 : 1.0 m
C2 : 1.0 m
Burner : furnace Temperature (°C): 800
1000 1200 1400 1600 1800
Figure 11.10 (See color insert following page 424.) Cross-sectional temperature distribution for burners C1 and C2 in 1.0 m and 1.5 m ID furnaces. (From Chui, E. H., Majeski, A. J., Douglas, M. A., Tan, Y., and Thambimuthu, K. V., Energy, 29, 1285–96, 2004. With permission.)
C1 might not be as desirable as shown in Figures 11.7 through 11.9 when wall effects in the near burner region were reduced. A second level of design refinements was required to develop a better burner concept. 11.4.5 Burner C2: Level 2 Improvement
3
Centerline distribution Burner : Furnace ID
4
B : 0.6 m C1 : 0.6 m B : 0.6 m
C1 : 1.5 m
C1 : 0.6 m
50 300 550 800 1050 1300
0
1000 2000 NO (ppm dry)
Figure 11.9 (See color insert following page 424.) NO distribution on fur nace cross section and centerline for burners B and C1. (From Chui, E. H., Majeski, A. J., Douglas, M. A., Tan, Y., and Thambimuthu, K. V., Energy, 29, 1285–96, 2004. With permission.)
© 2011 by Taylor and Francis Group, LLC
Burner C2 was a modified version of C1 with features less sensitive to wall effects. With C2 the flame no longer attached to the wall of the larger-diameter, stacked furnace as seen in Figure 11.10. Also, the peak flame temperature dropped (from 1820°C in C1 to 1620°C in C2) and the furnace cross section had a fairly uniform temperature distribution, suggesting improved mixing. The CO plot (Figure 11.11) further substantiated better mixing of fuel and oxidant with burner C2. The average CO level at 5 m downstream of the quarl outlet was 24 ppm, two orders of magnitude less than 2540 ppm, the average CO level at the same location when burner C1 was used. Figure 11.12 shows that burner C1 still succeeded in generating a distinct fuel-rich zone in the near burner region of the 1.0 m ID stacked furnace (as in the
259
0
0
1
1
Axial distance (m)
Axial distance (m)
Virtual Testing
2 Centerline distribution burner : furnace ID
3
2
3
C1 : 1.0 m C2 : 1.0 m
4
C1 : 1.0 m 10
100
C1 : 1.0 m C2 : 1.0 m
C2 : 1.0 m
101
1000 10000 100000
102 103 104 CO (ppm dry)
105
0
Axial distance (m)
1
2
3
Centerline distribution burner : furnace ID C1 : 1.0 m C2 : 1.0 m
4
0
C1 : 1.0 m
C2 : 1.0 m
50 300 550 800 1050 1300
Figure 11.11 CO distribution on stacked furnace cross section and centerline for burners C1 and C2 (note logarithmic scale). (From Chui, E. H., Majeski, A. J., Douglas, M. A., Tan, Y., and Thambimuthu, K. V., Energy, 29, 1285–96, 2004. With permission.)
C1 : 1.0 m
Centerline distribution burner : furnace ID
4
C2 : 1.0 m
8 16 24 32 40
0
20 O2 (Vol. % dry)
Figure 11.12 (See color insert following page 424.) O2 distribution on stacked furnace cross section and centerline for burners C1 and C2. (From Chui, E. H., Majeski, A. J., Douglas, M. A., Tan, Y., and Thambimuthu, K. V., Energy, 29, 1285–96, 2004. With permission.)
© 2011 by Taylor and Francis Group, LLC
0
1000 2000 NO (ppm dry)
Figure 11.13 (See color insert following page 424.) NO distribution on stacked furnace cross section and centerline for burners C1 and C2. (From Chui, E. H., Majeski, A. J., Douglas, M. A., Tan, Y., and Thambimuthu, K. V., Energy, 29, 1285–96, 2004. With permission.)
original 0.61 m ID unit, Figure 11.8), with a low O2 (fuelrich) central core to inhibit NOx formation. However, in using burner C1, the high O2 streams near the wall in the larger stacked combustor did not migrate as quickly and uniformly (as in the smaller existing unit) to the central core downstream of the furnace to effectively react with the volatile matter and CO. This was the basis for the high CO level observed even at about 5 m from the burner exit. In contrast, burner C2 created a less stratified distribution of O2 in the near burner region (Figure 11.12) and yet maintained a low O2, fuel-rich region in the central core to reduce NOx. As a result of these improvements, both CO and NOx could be minimized. Figure 11.13 compares the distribution of NO between burners C1 and C2. At the furnace exit, C2 yielded 62 ng/J of NO (328 ppm average), higher than that from using C1 (40 ng/J of NO, 214 ppm average) but with much better combustion characteristics (i.e., much lower CO emissions). 11.4.6 Demonstration of Virtual Test Results Based on the findings of the numerical investigation, burner C2 was built and the combustor was enlarged to 1.0 m ID in the near burner region as suggested by virtual testing. Actual experiments were completed to test the technical viability of the new burner–combustor
260
Industrial Combustion Testing
TABLE 11.1 Operating Conditions and Flue Gas Analyses for the Experimental and Numerical Investigations of Burner C2-Stacked Furnace Combination in Recycled Flue Gas Mode
Coal feed (kg/h) Primary stream CO2 (kg/h) Recycled flue gas (kg/h) Oxygen input (kg/h) Flue gas analysis O2 (vol.% dry) CO (ppm, dry) NOx (ppm, dry) CO2 (vol.% dry by balance)
Numerical Study
Measurements from Commissioning Runs
35.5
35.7–36.7
67.4 86.7 59.3
65.0–79.9 58.3–112.1 50.6–55.5
2.0 24 328 98.0
1.9–3.4 24–73 226–800 96.6–98.1
Source: From Chui, E. H., Majeski, A. J., Douglas, M. A., Tan, Y., and Thambimuthu, K. V., Energy, 29, 1285–96, 2004. With permission.
combination. The experimental results indicated that stable flames could be achieved with burner C2 mounted on the stacked furnace with input and output conditions staying fairly constant over each test. Table 11.1 shows the operating conditions covered in the actual test runs and the corresponding flue gas analyses. Although the test conditions between the experimental and numerical investigations were similar (Table 11.1), none of the experimental test cases had a set of operating conditions perfectly matching the numerical study due to some minor difficulties experienced in setting the various flow rates into the unit. Generally, the experiments confirmed that NOx production dropped significantly with the new burner–combustor combination (226 to 800 ppm versus 1230 ppm with burner B and the original combustor) while combustion characteristics remained satisfactory for a coal burner (24 to 73 ppm CO at exit). In fact, the lowest measured NOx at furnace exit (226 ppm average) was registered when the experimental operating conditions were closest to those assumed for the numerical study, providing a validation of the predicted performance of burner C2 (328 ppm of NO under similar operating conditions).
11.5 Example 3: Virtual Testing to Identify Heater Operational Problem 11.5.1 Context The nonuniform heat distribution in the radiant section of a nine-burner refinery process fluid heater
© 2011 by Taylor and Francis Group, LLC
Process fluid
37´ Burners
Figure 11.14 Picture of refinery furnace and the schematic of its cross section. (From Chui, E. H., Applications of CFD Modeling in Canadian Industries, Fourth International Conference on CFD in the Oil and Gas, Metallurgical & Process Industries SINTEF/NTNU, Trondheim, Norway, June 2005.)
(Figure 11.14) was a concern because the two streams of process fluid going through the unit received uneven heat transfer and reached the next process at unacceptably different temperatures, compromising production and safety. Repeated field efforts to balance the fuel input to the furnace did not seem to fully correct the problem. The main goal of the virtual test was to use simulation technology to investigate the cause of nonuniform heat distribution [15]. 11.5.2 Virtual Test Approach To represent the partially premixed turbulent combustion of a refinery gas in the heater, a combination of the flamelet formulations for premixed and nonpremixed combustion was used [16]. The standard k–ε model was used for turbulent flow calculations. The effect of turbulence on the mixture fraction was accounted for by integrating a beta-PDF derived from the local mixture fraction and mixture fraction variance, which were in turn obtained by solving their respective transport equations. A relatively simple approach was used to compute radiant heat transfer—a diffusion model with a constant absorption coefficient (0.1 m–1). 11.5.3 Cause of Uneven Heat Distribution Even with identical fuel and airflow rates to each of the six outer burners on the two sides of the heater, the virtual test results showed that the temperature distribution in the upper part of the heater was far from symmetric (Figure 11.15), just as observed in the actual field test. A thorough investigation of the virtual test results of the unit revealed that the asymmetric temperature distribution was caused by the unstable flow characteristics inside the heater. By design, each burner in the center of the heater was given
261
Virtual Testing
a fuel input double of that for each side burner because the center burners would impact on both process fluid streams going through the heater, while the side burners would impact on only one of the two streams (Figure 11.14). As a result, the jet flames from the three burners in the center of the unit possessed a higher momentum than those from the six outer burners. If the conditions on the two sides of these high momentum jets were just slightly different, they would hit the heater ceiling off-center and generate an unbalanced flow pattern and hence, an uneven heat distribution to the two streams of process fluid. Perfect symmetry could have been achieved only if the fuel and airflow rates on the outer burners were balanced and the flame holders have identical shape and orientation. It would have been unrealistic to expect these conditions to be fulfilled in real practice. 11.5.4 Value of Virtual Testing The virtual test results provided a wealth of information on the detailed flow and thermal characteristics inside the unit, which would have been very difficult to obtain experimentally. It was based on such detailed information that the cause of the operational problem was uncovered. The plant engineers were then able to exploit the new understanding from the virtual test to make modifications on the burners. The temperature differential between the two process streams dropped from hundreds to tens of °F. Energy consumption of the furnace was reduced due to improved efficiency. More importantly, the unit could now handle an elevated throughput, translating into a revenue increase of almost CDN $300,000 per year.
11.6 Example 4: Virtual Testing to Guide Retrofit of Refinery Furnace 11.6.1 Context A refinery gas furnace that supplied heat for a process reaction would need to be retrofitted to meet the new heat requirement of an anticipated upgrade of the associated process equipment to increase overall production. The heating was achieved by passing the process fluid through pipes suspended in the furnace and exposed to flame radiation and convection from the hot gases inside the furnace. The maximum allowable pipe surface temperature typically governed the maximum firing rate so local “hot spots” were undesirable. Therefore, the distribution of heat flux to the pipes should be relatively uniform over all pipe surfaces. Recent thermography measurements had confirmed the operators’ suspicion that the pipe surface temperatures were not uniform, indicating that the heat flux was also not uniform. The purpose of the virtual test was to understand the reasons for the nonuniformity and to provide guidance on retrofitting the furnace to accommodate the associated process equipment upgrade [17]. 11.6.2 The Furnace The furnace (Figure 11.16) was divided into four “cells,” in each of which a row of pipes carrying the process fluid was suspended, providing heat to the catalytic reaction process in four stages. Taking advantage of
Temp Hot
Dissimilar flow
Cool Figure 11.15 (See color insert following page 424.) Furnace cross section colored by gas temperature, with flow pattern superimposed. (From Chui, E. H., Applications of CFD Modeling in Canadian Industries, Fourth International Conference on CFD in the Oil and Gas, Metallurgical & Process Industries SINTEF/NTNU, Trondheim, Norway, June 2005.)
© 2011 by Taylor and Francis Group, LLC
Figure 11.16 Picture of a refinery furnace with multiple burners. (From Runstedtler, A., Chui, E. H., Majeski, A., Trottier, S., Brunet, D., and Van Doormaal, M. A., Modeling of a Refinery Furnace, American Flame Research Committee International Symposium, Livermore, CA, October 2003.)
262
Figure 11.17 Virtual test domain representing one-half the geometry. Left to right: Cells B, A, C, D. (From Runstedtler, A., Chui, E. H., Majeski, A., Trottier, S., Brunet, D., and Van Doormaal, M. A., Modeling of a Refinery Furnace, American Flame Research Committee International Symposium, Livermore, CA, October 2003.)
symmetry, only half of the unit was considered in the virtual test. Figure 11.17 shows the view through the symmetry plane, identifying cells A, B, C, and D, partially separated by dividing walls. In total, the heat release for the furnace was 99.5 MW, with cells A, B, C, and D releasing 31.4, 34.7, 20.5, and 12.9 MW, respectively. All the burners were identical in design (with partially premixed flames) and firing rate. They were wall-mounted with double units in the center for cells A, B, and C, but floor-mounted with single units for cell D. The overall combustion reaction was based on a representative refinery gas fuel with a lower heating value of roughly 46,048 kJ/kg and an air/fuel ratio of 19.1 (by mass) with 4% excess oxygen (by volume) in the flue gas. 11.6.3 Virtual Test Approach The virtual test approach of the furnace was similar to the one used in Example 3 to treat multiple burners with partially premixed turbulent combustion of refinery gas. In actual furnace operation, fuel was adjusted using a valve controlled by the process fluid temperature while air was drawn into the furnace by an induced draft. The damper located at each burner was adjusted until the desired excess oxygen level at the furnace outlet was achieved. Except for the known fuel flow rate, the detailed distribution of fluid flow, temperature, turbulence, and mixture fraction at each burner needed to be defined via a single burner simulation. These profiles
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
Figure 11.18 (See color insert following page 424.) Left: Visualization of burners in cell D based on virtual test results. Right: Photograph of burner in cell D. (From Boisvert, P. G., Majeski, A., Runstedtler, A., and Chui, E. H., Flame Visualization Based on Computational Fluid Dynamics (CFD) Results, The 2008 International Conference on Modelling, Simulation & Visualization Methods, Las Vegas, NV, July 2008.)
once determined were applied to all burners of the furnace as inlet boundary condition. A sample burner virtual test result is shown in Figure 11.18 alongside a photograph of an actual burner in operation. The calculated flame characteristics are seen to be very similar to the actual ones. For this virtual test of the furnace, the flow of process fluid inside the pipes was not considered so the outer surface temperature had to be specified based on thermography data and an assumed emissivity of 0.8. To represent the heating of the process fluid traveling through the pipes and its resulting affect on heat transfer, the inlet side was made colder than the outlet side according to process fluid temperature data provided by furnace operators. The four furnace outlets (see Figure 11.17) were assumed to have the same pressure. The refractory wall was assumed to be adiabatic with an emissivity of 0.6. 11.6.4 Virtual Test Results The predicted distribution of net heat flux at the pipe surfaces is illustrated in Figure 11.19. It is interesting to note that the process fluid temperature distribution had a noticeable effect on the heat transfer. Cells A and B had the largest heat fluxes at the bottom, with considerably less heating at the top, whereas cells C and D had relatively uniform heat flux. It is also evident that, in cells A, B, and C, the peak heat flux occurred near the center of the tube bank (near the symmetry plane in the model). A summary of firing rates and heat transfer predictions is provided in Table 11.2. The calculated total
263
Virtual Testing
Figure 11.19 (See color insert following page 424.) Net heat flux [W/m2] at pipe surfaces. (From Runstedtler, A., Chui, E. H., Majeski, A., Trottier, S., Brunet, D., and Van Doormaal, M. A., Modeling of a Refinery Furnace, American Flame Research Committee International Symposium, Livermore, CA, October 2003.)
Figure 11.20 (See color insert following page 424.) Incident radiant heat flux [kW/m2] at pipe surfaces. (From Runstedtler, A., Chui, E. H., Majeski, A., Trottier, S., Brunet, D., and Van Doormaal, M. A., Modeling of a Refinery Furnace, American Flame Research Committee Inter national Symposium, Livermore, CA, October 2003.)
TABLE 11.2 Summary of Firing Rates and Heat Transfer Predictions from Virtual Test of Refinery Furnace Input Measured Predicted
Predicted distribution
Cell
A
B
C
D
Firing rate [MW] Firing rate per burner [MW] Heat transfer rate [MW] Heat transfer rate [MW] % Radiation % Convection Average heat flux [kW/m2] Maximum heat flux [kW/m2] Minimum heat flux [kW/m2] Root mean square deviation [kW/m2]
31.4 1.97 10.0 11.0 86.2 13.8 51.8 86.2 23.9 11.1
34.7 1.45 16.5 12.3 88.1 11.9 48.3 84.4 21.6 11.2
20.5 1.28 8.91 9.94 88.1 11.9 46.9 67.0 27.5 6.24
12.9 1.29 4.81 4.45 87.6 12.4 44.7 54.1 22.8 4.20
heat transfer to the pipes (37.9% of the heat release rate for the furnace) agrees closely with the measured value (40.5%), the main discrepancy being that the heat transfer in cell B is underpredicted by about 25%. This could be explained by the pipe surface temperature boundary condition, which was crudely estimated from thermography pictures. Table 11.2 also provides the convective and radiant heat transfer as a percentage of the total heat transfer for each cell. The heating in the bottom of cells A and B is driven by radiant energy from the flame, as demonstrated by a plot of the incident radiant flux in Figure 11.20. However, from this plot it can be deduced that peak heating near the center of the tube bank is not caused by radiation. It must, instead, be the result of convection. The net convective heat flux plot and the isosurface plot
© 2011 by Taylor and Francis Group, LLC
in Figure 11.21 confirm this, where it is apparent that the interaction of opposing jets causes the hot gases to turn and pass through the tube bank, thereby coming into contact with the pipes. 11.6.5 Value of Virtual Test The virtual test results showed that the cells with the greatest nonuniformity (A and B) had the largest heat fluxes at the bottom and considerably less heating at the top. This was caused by radiation from the flames, which were concentrated in the bottom of the furnace. Further, the peak heat fluxes that occured near the center of the tube banks were caused by the interaction of opposing burner jets, which met in the center of the furnace. The hot gases from the burner jets must turn
264
Figure 11.21 (See color insert following page 424.) Left: Net convective heat flux [W/m2] at pipe surfaces for Cells B and A. Right: Isosurface of temperature (1970°F) in the furnace. (From Runstedtler, A., Chui, E. H., Majeski, A., Trottier, S., Brunet, D., and Van Doormaal, M. A., Modeling of a Refinery Furnace, American Flame Research Commit tee International Symposium, Livermore, CA, October 2003.)
and pass through the tube bank, thereby coming into contact with the pipes. The knowledge gained from the virtual test was used to improve the uniformity of heat flux. Given that the incident radiant heat flux was concentrated toward the bottom of cells A and B, where the burners were located, it would seem logical to spread out the heat release by repositioning burners or adding additional burners higher up. Also, the interaction of the opposing burner jets most likely caused the peak heat flux near the center of the tube bank. Slowing down the burner jets by adjusting the firing rate would be desirable. These proposals were passed to the refinery personnel to formulate the retrofit plan with equipment suppliers. Subsequently, an additional level of burners was added with firing conditions based on the knowledge obtained from the virtual test. Field tests indicated that the retrofit was a success and increased production was achieved without incidence. In essence, the virtual test results provided a priori information that guided the development of a viable furnace retrofit necessary to accommodate equipment upgrade, and spared the refinery from determining the retrofit strategy with a potentially more costly, trial-and-error approach.
11.7 Summary Remarks The idea of virtual testing, based on CFD combustion modeling techniques, was introduced in the context of
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
industrial combustion testing. The first example highlighted how virtual testing could be used to complement experimental testing to gain better understanding of an advanced combustion process pertinent to CCS (i.e., oxy-firing). The second example illustrated the effectiveness in utilizing virtual testing to explore burner and combustor design concepts prior to actual construction. Subsequent experimental testing of the burner and combustor designs, optimized by modeling, confirmed the predicted superior performance. These examples highlighted the role that virtual testing could play in research and development of burner understanding and design concepts. The third and fourth examples demonstrated how virtual testing could be used to meet actual needs in practical industrial applications. The virtual test results facilitated the diagnosis of equipment operational problems and more importantly, the identification of practical solution strategies that resulted in increased production and revenue. Because the last two examples dealt with actual industrial size units with complexity in the physics of the combustion process and in the interactions between multiple burners, the accuracy of the virtual tests should be addressed. Basically, the objective of these virtual tests was to produce some useful insights on the units. Therefore, the accuracy required should be consistent with the features of interest to be resolved and did not necessarily have to be at an extremely high level, considering the differences between industrial units and laboratory experiments. Typically, the operation of an industrial furnace is not as rigorously controlled as a laboratory experiment and hence, the operating parameters or conditions will vary, either intentionally or unintentionally. For instance, the fuel for the refinery furnaces in Examples 3 and 4 could change greatly on a daily basis, depending upon what was available in the plant. The intent of the virtual tests was to determine the major operating features of the units using a time-averaged fuel representation. It would have been unnecessary and too costly to perform virtual tests on all possible fuel variations unless the fuel property was a focus of investigation. Also, there was the frequent lack of knowledge or control of certain operating parameters. For example, the operators would usually know the total airflow rate going into a furnace, but they might not know how much air went into individual nozzles because it was not feasible to measure all the flows. For the virtual test, an estimate would have to be made about the relative airflows, resulting in some uncertainty. Hence, extremely sophisticated (and computationally expensive) virtual testing approaches would have been inappropriate in these situations. In fact, engineering virtual testing approaches used in Examples 3 and 4 were found quite adequate in generating valuable outcomes.
Virtual Testing
References
1. Baukal, C. E., Gershtein, V. Y., and Li, Y., (eds.). Computational Fluid Dynamics in Industrial Combustion. Boca Raton, FL: CRC Press, 2001. 2. Henneke, M., Smith, J. D., Lorra, M., and Jayakaran, D. “CFD Based Combustion Modeling.” In The John Zink Combustion Handbook, edited by C. E. Baukal, Jr. and R. E. Schwartz, 287–325. Boca Raton, FL: CRC Press, 2001. 3. Chui, E. H., Douglas, M. A., and Tan, Y. “Modeling of Oxy-Fuel Combustion for a Western Canadian SubBituminous Coal.” Fuel 82 (2003): 1201–10. 4. Douglas, M. A., Chui, E. H., Tan, Y., Lee, G. K., Croiset, E., and Thambimuthu, K. V. “Oxy-Fuel Combustion at the CANMET Vertical Combustor Research Facility.” The First National Conference on Carbon Sequestration, Washington, DC, May 14–17, 2001. 5. Tan, Y., Douglas, M. A., and Chui, E. H. “A Review of Experimental Findings in Oxy-Fuel Combustion at the CANMET Vertical Combustor Research Facility.” American Flame Research Committee Spring Meeting, Ottawa, Canada, May 8–10, 2002. 6. Chui, E. H., Majeski, A. J., Douglas, M. A., Tan, Y., and Thambimuthu, K. V. “Numerical Investigation of OxyCoal Combustion to Evaluate Burner and Combustor Design Concepts.” Energy 29 (2004): 1285–96. 7. Lockwood, F. C., Rizivi, S. M. A., Lee, G. K., and Whaley, H. “Coal Combustion Model Validation Using Cylind rical Furnace Data.” 20th Symposium (International) on Combustion. The Combustion Institute, Pittsburgh, PA, 513–22, 1984. 8. Lockwood, F. C., and Mahmud, T. “The Prediction of Swirl Burner Pulverized Coal Flames.” 22nd Symposium (International) on Combustion. The Combustion Institute, Pittsburgh, PA, 165–73, 1988. 9. Lockwood, F. C., Mahmud, T., and Yehia, M. A. “Mathematical Modelling of Some Semi-Industrial Pulverized Coal Flames.” The Eighth Annual International Pittsburgh Coal Conference, 41–46, 1991.
© 2011 by Taylor and Francis Group, LLC
265
10. Migdal, D., and Agosta, V. D. “A Source Flow Model for Continuum Gas-Particle Flow.” Journal of Applied Mechanics 35 (1967): 860–65. 11. Badzioch, S., and Hawsksley, P. G. W. “Kinetics of Thermal Decomposition of Pulverized Coal Particles.” Industrial & Engineering Chemistry Process Design and Development 9, no. 4 (1970): 521–30. 12. Magnussen, B. F., and Hjertager, B. H. “On Mathematical Modelling of Turbulent Combustion with Special Emphasis on Soot Formation and Combustion.” 16th Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 719–29, 1976. 13. Chui, E. H., and Hughes, P. M. J. “Validation of NOx and NOx Precursor Predictions in Coal Flames.” Combustion Science and Technology 119 (1996): 51–75. 14. Chui, E. H., McDonald, M. M., and Smith, D. W. “Application of Numerical Modelling to Predict Combustion and NOx characteristics in Industrial Combustion Devices.” Air and Waste Management Associations 90th Annual Meeting and Exhibition, Toronto, June, 1997. 15. Chui, E. H. “Applications of CFD Modeling in Canadian Industries.” Fourth International Conference on CFD in the Oil and Gas, Metallurgical & Process Industries SINTEF/NTNU, Trondheim, Norway, June, 2005. 16. Mueller, C. M., Breitbach, H., and Peters, N. “Partially Premixed Turbulent Flame Propagation in Jet Flames.” 20th Symposium (International) on Combustion, The Combustion Institute, Irvine, CA, 1994. 17. Runstedtler, A., Chui, E. H., Majeski, A., Trottier, S., Brunet, D., and Van Doormaal, M. A. “Modeling of a Refinery Furnace.” American Flame Research Committee International Symposium, Livermore, CA, October, 2003. 18. Boisvert, P. G., Majeski, A., Runstedtler, A., and Chui, E. H., “Flame Visualization Based on Computational Fluid Dynamics (CFD) Results.” The 2008 International Conference on Modelling, Simulation & Visualization Methods, Las Vegas, NV, July, 2008.
Section II
Advanced Diagnostics
© 2011 by Taylor and Francis Group, LLC
12 Laser Measurements Michele Marrocco and Guido Troiani Contents 12.1 Purpose of Laser Measurements............................................................................................................................... 269 12.2 Materials for Laser Measurements............................................................................................................................ 270 12.2.1 Lasers.............................................................................................................................................................. 270 12.2.2 Optical Elements............................................................................................................................................ 271 12.2.3 Spectral Dispersion and Detection............................................................................................................. 272 12.3 Laser Measurements for Chemical Species............................................................................................................. 272 12.3.1 Incoherent Techniques.................................................................................................................................. 273 12.3.1.1 Laser Induced Fluorescence (LIF)...............................................................................................274 12.3.1.2 Spontaneous Raman Scattering (SpRS)......................................................................................274 12.3.1.3 Other Incoherent Techniques for Detection of Chemical Species......................................... 275 12.3.2 Coherent Techniques.................................................................................................................................... 276 12.3.2.1 Coherent Anti-Stokes Raman Scattering (CARS).................................................................... 276 12.3.2.2 Degenerate Four Wave Mixing (DFWM).................................................................................. 277 12.3.2.3 Other Coherent Techniques for the Detection of Chemical Species..................................... 278 12.4 Laser Measurements for Velocimetry....................................................................................................................... 278 12.4.1 Laser Doppler Velocimetry (LDV).............................................................................................................. 279 12.4.2 Particle Image Velocimetry (PIV)................................................................................................................ 279 12.5 Laser Measurements for Particle Sizing................................................................................................................... 280 12.5.1 Laser Doppler Velocimetry (LDV).............................................................................................................. 280 12.5.2 Phase Doppler Anemometry (PDA) or Velocimetry (PDV).................................................................... 281 12.5.3 Other Laser Techniques for Particle Sizing............................................................................................... 281 12.6 Laser Measurements for Thermometry.................................................................................................................... 281 12.6.1 Rayleigh Thermometry................................................................................................................................. 282 12.6.2 LIF Thermometry.......................................................................................................................................... 283 12.6.2.1 LIF Thermometry with Excitation Scans.................................................................................. 283 12.6.2.2 LIF Thermometry with Two-Color Approach.......................................................................... 283 12.6.2.3 LIF Thermometry with Thermal Assistance............................................................................ 283 12.6.3 Raman-Based Techniques for Thermometry............................................................................................ 284 12.6.3.1 Thermometry with SpRS............................................................................................................. 284 12.6.3.2 Thermometry with CARS........................................................................................................... 284 12.6.4 Other Thermometric Laser Techniques..................................................................................................... 285 12.7 Combined Laser Measurements................................................................................................................................ 285 12.7.1 Examples of Combined Measurements...................................................................................................... 286 References................................................................................................................................................................................. 286
12.1 Purpose of Laser Measurements It is undeniable that one of the scientific disciplines undergoing an unprecedented level of development is laser spectroscopy, which could be defined as the countless collection of techniques aiming at the extraction of
physical and chemical information from the interaction between the matter in any phase (solid, liquid, or gas) and the electromagnetic fields created by one or more laser beams [1–5]. Certainly, the advent of efficient, powerful, reliable, and commercially available laser systems has boosted many scientific and technological achievements encompassing basic and industrial applications 269
© 2011 by Taylor and Francis Group, LLC
270
as well as research at the cutting edge of science. It is then not surprising that the understanding of atomic and molecular physics is going through a flourishing time thanks to the vital role of laser measurements [1–5]. Needless to say, the same applies to combustion science [6–9]. In this context, the introduction of lasers has powered the understanding of the multifaceted processes happening in various combustion environments and often involving the ability to deal with the interdisciplinary subtleties of chemistry, molecular physics, and fluid-dynamics that challenge the scientist as well as the practitioner. Not only are such accomplishments distinctive of fundamental combustion research, but also industrial combustion benefits from them [10–22]. In retrospect, we can say that combustion science would have failed many advances without the assistance of laser spectroscopy necessary to remotely generate and interpret data about chemical compositions, temperatures, velocities, and so on. It is then obvious to reach the conclusion that laser measurements in combustion science are planned according to important diagnostic criteria that will be reviewed here. In agreement with the intent of this chapter, we will try to give a brief description of the spectroscopic techniques that are currently available in combustion research. Some of them, the most common, will be discussed at a very basic level in order to help the nonspecialist to get hold of the main facts that orientate the choice of one technique in place of another. However, for a more detailed and accurate analysis, the reader is encouraged to look at the specialized literature that has grown beyond measure during the past decades (see, for instance, the numerous references in [6–9]). There, one can not only learn in much greater detail about the techniques broadly summarized here, but also realize that there exist many other minor spectroscopic techniques that, for the sake of conciseness, cannot be treated in the following pages. The chapter is organized around four central topics dictated by the demand for experimental information on chemical and physical variables acting during the combustion. As a consequence, we will outline laser measurements for chemical composition (Section 12.3), fluid-dynamics (Section 12.4), particle sizing (Section 12.5), and temperature (Section 12.6). The compendium of the most meaningful features reported throughout the chapter will not be sufficient to form a complete overview on these four fundamental aspects of combustion diagnostic. Regrettably, the whole matter is so specialized that many notions and concepts of laser spectroscopy are required to cope with the diversity of the problems arising when laser measurements are carried out. In an effort to reduce the dependence on this specialized knowledge, the scope of an introductory approach to the subject is at the heart of what follows.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
For this reason, it seems imperative to start with the concise description of fundamental optical tools and instrumentation needed to perform laser measurements.
12.2 Materials for Laser Measurements Diagnostic applications based on lasers would amount to nothing without the handling of several optical components that complement the choice of the laser system. Apart from the technology developed to build increasingly sophisticated lasers and beyond the scientific knowledge to understand the optical physics of molecules interacting with electromagnetic fields, the success of a diagnostic strategy contemplating laser methods depends on the correct use and arrangement of numerous optical elements. These include mirrors, lenses, beam splitters, prisms, polarizers, nonlinear crystals, filters, optical fibers, gratings, monochromators or spectrographs, detectors, and so on. In addition to this variety of objects (and rather often all the mentioned optical components appear in an experimental set-up), it must be remarked that they are not unique in their own genre and it is enough to browse through one of the different commercial catalogs to realize that the choice is incredibly vast. This is why the assembling of an experiment is more demanding than the mere laser selection. It is outside the scope of this chapter to examine the detail of the pieces of an advanced laser equipment; however, before reviewing the main spectroscopic techniques that are fit for use in combustion science, it is appropriate to get acquainted with the fundamental optical elements that are essential in any measurement of laser spectroscopy. Clearly, it is not possible to go through all the elements and some of them are intentionally left out of the discussion. For a more detailed description, the reader can refer to the book by Eckbreth [7]. All the same, in an attempt to give a brief summary of those optical components that are decisive for a successful result, it is compulsory to begin with the apparatus that plays the major role in the experimental methods considered later: the laser system. 12.2.1 Lasers The laser is a fascinating light source that offers the advantage of high luminous power concentrated in a beam of high directionality, negligible divergence, characterized by some time and spectral properties. These characteristics can vary extensively, because the laser technology has achieved an impressive level of development and, currently, it is possible to find a laser system tailored for almost any experimental requirement. In
271
Laser Measurements
practice, power, frequency range, time duration, and repetition of the laser pulse can be chosen at will. But, it might happen that one laser system cannot meet all the requirements and the combination of more than one single laser is indispensable. Actually, this is rather common for advanced diagnostics conceived to tackle the complexity of the information stored in molecules and particles. Generally speaking, any laser can be conceptualized as an optical cavity terminated at the two ends by mirrors [23]. Somewhere in the middle, the cavity contains an active medium, namely the medium that is responsible for the laser effect, the onset of which is triggered by some sort of optical excitation. The radiation that is released by the optically excited medium starts to bounce back and forth between the two mirrors and it gains more and more energy at any passage through the medium. The phenomenon is known as stimulated emission and competes against spontaneous emission that introduces energy losses detrimental to the laser process [24]. When the former emission overcomes the latter (this condition is called population inversion), the final balance is an energy gain. The word “laser” is, indeed, an acronym for light amplification by stimulated emission of radiation [23,24]. The amplified light does not stay permanently in the optical cavity. A portion of it leaks out at each roundtrip and builds up the laser beam. There are many types of laser systems but they all obey the principles just sketched. One important criterion of differentiation is, instead, the physical state of the optical medium. For instance, the laser with continuous emission of radiation (i.e., continuous wave or CW regime) and customarily employed in combustion diagnostic, accommodates a gaseous medium (more specifically, Ar ions in a plasma tube). That is the socalled Ar-ion laser that generates a powerful beam with two main spectral lines in the visible spectrum at 488 and 514 nm. Another gas laser, used mostly for optical alignment, is made of a mixture of He and Ne atoms (He–Ne laser). But, the laser that is recurrently shown in experimental schemes of combustion measurements is a solid-state laser engineered with a glassy material called Nd-doped yttrium aluminum garnet (Nd-YAG laser). This laser can provide intense laser pulses of less than 10 ns duration that are very useful in combustion diagnostics, but the repetition rate of few tens of pulses per second represents a severe limit to study dynamical processes. Other than this drawback, the Nd-YAG laser displays relevant spectral versatility. It can be coupled with nonlinear crystals for harmonic generation [4] to create secondary laser lines lying in an important region of the spectrum. Indeed, the fundamental wavelength is at 1064 nm, but efficient second-, third- and fourthharmonic generation shifts the wavelength down to 532,
© 2011 by Taylor and Francis Group, LLC
355, and 266 nm, respectively. This spectral versatility has a key role in the popularity of such a laser system and, notably, the vast majority of scientific works regarding combustion applications exploits to some extent the very favorable spectral and lasing characteristics of the Nd-YAG laser. To conclude this very incomplete list of laser systems, we should add dye lasers that occupy a special place in molecular spectroscopy. Their optical medium is made of a nonaqueous solution of an organic dye or a mixture of dyes (regular solvents are ethanol and methanol) that, when excited, emits light over a broad range of wavelengths. What is more, the choice of dyes is wide and covers many bands in the entire visible spectrum. This means that dye lasers are very versatile and widely employed in combustion diagnostic for their continuous tunability within large spectral intervals. Moreover, they can be operated either in CW or pulsed regime. In the latter configuration, the dye laser is customarily pumped by a Nd-YAG laser and, as a matter of fact, the coupling of these two laser systems features repeatedly in an uncountable number of scientific works on laser diagnostic. 12.2.2 Optical Elements The availability of a laser beam serves no purpose if we are not able to control it. Any experimental set-up is not simply reducible to the laser system per se and a couple of pieces of equipment for the detection of the light obtained in response to the laser excitation. There are many other things in between. For example, as mentioned earlier, it is fairly common to shift the main laser line to another more convenient wavelength. This can be done by means of nonlinear crystals that convert a portion of the input beam from the fundamental wavelength to a fraction of its initial value (1/2 for second harmonic generation, 1/3 for third harmonic generation, and so on). Continuing with the optical elements that influence the characteristics of the laser beams, polarizers might also become necessary to change the state of polarization (the direction of oscillation of the electric field). The typical polarizer for laser work is the Glan–Thomson prism, made of two anisotropic quartz prisms with triangular section. Once the laser beam is prepared in the state that is right for the experimental purposes, there are other optical operations without which the application of the diagnostic technique would be senseless. For example, should the laser beam be split into two intensity components, optical elements, purposely named beam splitters, are fitting. Mirrors are, instead, employed to deliver the laser beam wherever is required. Similarly, simple prisms in place of mirrors can be functional to further manipulate the direction of laser beams or lift
272
(through refraction) the spatial degeneracy of different wavelengths. Dichroism is also a property of certain optical surfaces that allow for the transmittance of certain wavelengths and the reflection of others that lie far apart in the spectrum (this feature is of value when the rejection of unwanted laser wavelengths must be accomplished). Furthermore, lenses of different kinds (spherical, cylindrical, convergent, divergent, etc.) are extremely useful to focus the laser light and then collect the optical signal that gives meaning to the measurement. This part of the experiment is rather delicate, because in most cases it defines completely the spatial resolution in point-wise or two-dimensional laser excitation. Optical filters are also frequent components of a set-up intended for laser spectroscopy. They are utilized to select the signal and reject other wavelengths belonging to spurious sources (stray light, residual laser light, emission from indiscriminate molecules, etc.). The filters can show dichroism. In other terms, they can be low- or high-pass (transmission of wavelengths lower or higher than a cut-off reference) whereas interference filters are manufactured with a narrow spectral window that is chosen in correspondence of the signal wavelength. 12.2.3 Spectral Dispersion and Detection A laser measurement, more often than not, entails dispersing the spectral information contained in the measured signal. The common attempt to meet this demand is made by means of monochromators (or spectrometers or spectrograph). These are grating instruments that put to good use the phenomenon of diffraction of a grating. The typical monochromator adopted in laser diagnostic is designed according to the Czerny–Turner geometry, where a rotating grating is symmetrically placed between two arms equipped with spherical mirrors facing an entrance slit on one side and an exit aperture (or slit) on the other side. Finally, the actual measurements are made by detectors. These have in common a photo-sensitive material that converts the light into an electric signal. This can be achieved in different ways. For example, some detectors are based on the photoelectric effect (extraction of electrons from an illuminated surface). The photomultiplier and vacuum photodiodes belong to this category. Another type of detectors banks on the electrons promoted in the conduction band of semiconductors. The PIN diodes belong to this category. Charged coupled devices (or CCD cameras) are also made of a photosensitive semiconductor that can be arranged in pixels for imaging. The dimension of the pixels is very small, on the order of 10 µm and cameras with an array of 1024 × 1024 pixels are now commercially available.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
12.3 Laser Measurements for Chemical Species A thorough analysis of combustion environments cannot disregard the information on local concentrations of the molecular species taking part in the chemical reactions characterizing the combustion processes [6–9]. Admittedly, the task is incredibly difficult because an ordinary event of combustion is made of tens (if not hundreds) of significant reactions running at the same time and involving many different molecular constituents. To track all of them is impossible. Nonetheless, there are a few species that are relatively easy to detect and, fortunately enough, are of great interest for their intimate relationship with the combustion as a whole. In particular, diatomics (molecules made of two atoms) can be measured under different circumstances because they possess not too complex schemes of energy levels. For instance, this is the case of nitrogen and oxygen molecules that have a primary importance in air-fed combustion, or we may also think of some of the most important polluting species (i.e., nitrogen and carbon monoxides). The situation becomes very problematic with more complex molecules. However, some triatomics (molecules made of three atoms) are still detectable, even though their optical response is complicated by the more intricate molecular photo-dynamics. The typical example is water that constitutes one of the major combustion products or we could name pollutants, such as nitrogen and carbon dioxides. Other than di- and tri-atomics, there are just very few molecules that can be detected with a certain accuracy. Likely, the main hydrocarbon fuels, such as methane and ethane, or some interesting intermediate species (formaldehyde) can be spectrally identified with acceptable difficulty. Despite the limited number of molecules that are technically detectable, the above-mentioned scenario is more than enough. Usually, one or two molecular species are observed for local concentration measurements (either spatially resolved through point-like acquisitions or imaged with field techniques). For instance, images of the distribution of hydroxil radical (OH) can be very rich in information. Rather often, this type of data is associated with other diagnostic means capable of complementary data (e.g., temperature or velocity) so that the researchers can reach a comprehensive understanding of the processes they are looking at. In principle, there exist several techniques that are specifically conceived to obtain measurements of chemical species [6–9]. Their physical origin has a direct consequence for the experimental strategy and, as will become apparent shortly, it is preferable to classify these laser methods in reference to the characteristics of their
273
Laser Measurements
signals. According to this criterion, we distinguish two groups (see Table 12.1). One group is made of incoherent techniques. The term “incoherent” refers to the fact that the molecules exposed to one (or more than one) laser beam react independently from each other and, for this reason, the final signal is proportional to the number of radiating molecules. Furthermore, the incoherence results in an optical response that is usually isotropic (i.e., the signal is dispersed over the whole solid angle) and, consequently, the typical experimental set-up is prepared in a 90° geometry as in Figure 12.1a (i.e., laser propagation is along one direction, whereas detection is made in a direction orthogonal to the excitation). The most popular examples of incoherent techniques used in combustion science are laser induced fluorescence (LIF) and spontaneous Raman scattering (SpRS). We will discuss these techniques in the next subsection. In opposition to incoherent spectroscopies, we find coherent techniques. This second group is made of those laser methods that emphasize the collective optical response of the excited molecules. The coherence manifests itself in a nonlinear behavior of the signal (this means that the emitted radiation is not proportional to the number of radiating molecules) and, additionally, the signal is not distributed over the whole solid angle because it shows directional properties (phase matching). This is the reason why experimental set-ups for coherent techniques require line-of-sight optical access as shown in Figure 12.1b (i.e., direction of detection coincident with direction of propagation of the laser beams). Common examples of coherent techniques used Table 12.1 General Classification of Popular Laser Techniques Used in Combustion Science Incoherent Techniques
Coherent Techniques
LIF SpRS
CARS DFWM
(a) Spectrograph and detector
in combustion science for detection of chemical species are degenerate four wave mixing (DFWM) and coherent anti-Stokes Raman scattering (CARS). We briefly comment on these. The classification based on the presence or absence of coherence is just one simple way to approach laser measurements. Other criteria of classification could be adopted. For instance, we could distinguish between techniques probing electronic and nonelectronic transitions, or we could separate techniques of linear optics from those involving optical nonlinearities, or we could contrast techniques limited to point-wise measurements against techniques with the potential of two-dimensional imaging, and so on. Assuming one of these criteria, say the distinction between approaches of linear and nonlinear optics, we should mention that methods of laser absorption (falling into the category of linear optics) receive much attention in industrial applications [10,11,14,25,26]. 12.3.1 Incoherent Techniques The incoherence translates the independence of an excited molecule from the surrounding excited molecules (the same is for atoms, but we limit the discussion to molecules, because they are most commonly found in combustion). In other terms, each radiating molecule acts on its own and, caused by the random orientation of the molecules filling the interaction region, the scattering of the emitted light occurs in all directions. This means that the most intuitive experimental arrangement is based on the idea that it is better to avoid detection in the direction of propagation of the powerful laser beam (Figure 12.1a). The latter would cause stray light in the detection equipment with troublesome consequences for the reliability of what is measured. This detection geometry is seen in experiments of LIF and SpRS, which are the prevalent spectroscopic methods in combustion research [7–9].
(b)
Entrance optics Filters Collection lens
Focusing lens
Secondary laser system
Dichroic Beam splitter mirror
Flame
Laser system
Mirror
Mirror
Laser system
Figure 12.1 Experimental geometry for (a) incoherent techniques and (b) coherent techniques.
© 2011 by Taylor and Francis Group, LLC
Spectrograph and detector Dispersion Filters Collection prism Flame lens Entrance Optical optics trap Focusing lens
274
12.3.1.1 Laser Induced Fluorescence (LIF) The LIF is probably the most common laser approach to measurements of chemical species [7–9,27,28]. Its success is due to the physical meaning of fluorescence. This term refers to the radiation emitted spontaneously by atoms or molecules after laser excitation of one of the outer electrons, which is promoted from its ground level (or energy) up to an excited level. The electron persists in the excited state for a very short time and, during the downward transition to the initial ground level, the negatively charged electron emits electromagnetic radiation, called fluorescence. For this reason, it is said that LIF involves electronic transitions. These are very strong if compared with other atomic and molecular processes. Based on this, LIF is often employed to detect species at very low concentrations, say ppm or sub-ppm level. In particular, radicals (such as OH, CH, NH, CN, C2) and pollutants (such as NO, CO), which are present in very small quantities and yet are extremely important, can be detected via LIF. Besides, the excitation is made using pulsed lasers so that the data are taken at precise time instants and within short time windows that help in discriminating against the background luminosity. Despite the convenience of LIF measurements, there are some constraints to do quantitative analysis on a given molecule. First of all, the fluorescence spectrum should be known to exclude effects of laser-induced dissociations and ionizations. Moreover, the molecular photo-dynamics should be considered in that it determines the fluorescence power. This is a function of the radiative decay and excited state losses. Among the latter, losses caused by collisional disexcitations, otherwise known as quenching, are very important, because they do not produce measurable radiative contributions. As a consequence, quenched molecules that regain their ground state after laser excitation and do not fluoresce, cannot be seen or measured. It becomes evident that one fundamental aspect of LIF measurements is related to how quenching is dealt with or avoided. There is a number of ways that guarantee proper accounting of quenching. They refer to either theoretical methods in computer-corrected algorithms or experimental methods with solutions contemplating alternatively the use of very short laser pulses, excitation of predissociative states, photoionization control, fluorescence saturation, or time-resolved acquisitions. The interested reader can learn in the specialized literature the principles of operation of all these approaches to quenching [7,27,28]. One of the final remarks on LIF is about the possibility of two-dimensional imaging (planar LIF or PLIF) [7,27,28]. This version differs from traditional LIF in terms of laser excitation (a sheet of laser light is arranged by means of a proper optical system of lenses)
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
and detection (usually made by means of an intensified CCD equipped with a specific interference filter). In effect, most of laser imaging as to detection of chemical species of combustion interest is based on two-dimensional fluorescence maps [7,27,28]. The reason for this success is soon explained. As mentioned earlier, the LIF originates from electronic transitions that are more strongly coupled with the laser fields than other types of molecular transitions (i.e., Raman transitions). The PLIF signals are then sufficiently intense to offer the prospect of good image quality. This potential is very advantageous and PLIF has held promise as a spectroscopic tool of unrivaled application for two-dimensional visualization of molecules. As a matter of fact, beyond the numerous applications in laboratories, two-dimensional diagnostic on radical and pollutant species produced in combustion systems of industrial interest seems feasible [17,18,22]. We conclude these very short comments on LIF by adding that, among the characteristics enumerated about LIF detection, it appears that the choice of spectroscopic targets is limited to light molecules. This is not true. For instance, detection of metal compounds of concern for industrial combustion testing includes a variety of examples (see Monkhouse [14] and references therein). 12.3.1.2 Spontaneous Raman Scattering (SpRS) The SpRS is the basic process of a large class of Raman effects summarized in the literature [5,7–9]. These refer to the interaction between the internal motion of molecules and a traveling electromagnetic field. This polarizes the molecules and the resultant oscillating polarization (e.g., the molecular electric dipole), and produces a new electromagnetic wave that is shifted in frequency. Indeed, during the passage of the external field, some energy is exchanged with the molecules and, if these lose some energy, the polarization shifts toward lower frequency values (Stokes shifts). Conversely, if the molecules gain some energy during the interaction, the polarization shifts toward higher frequency values (anti-Stokes shifts). A SpRS spectrum can thus be composed of different parts depending on the energy exchange between the traveling laser field and the perturbed molecules. These rotate and vibrate during the interaction with the laser and the changes in the corresponding molecular states of motion reflect the energy content that the molecules exchange with the field. Therefore, pure rotational spectral lines (determined by a change of the molecular rotation only) are very close to the exciting laser wavelength, whereas rovibrational spectra (determined by a combined change of rotational and vibrational states) are separated from the laser wavelength and they appear as spectral lines
275
Laser Measurements
at significantly higher or lower wavelengths (Stokes and anti-Stokes lines, respectively). A spectrum of this kind is very rich in information and contains the manifest signature of any given molecule with sufficient Raman activity. Moreover, the energy content of the Raman transitions is on the same order of the Boltzmann factor KBT suggesting that SpRS can also be adjusted to thermometric purposes (see the corresponding section on SpRS thermometry). One of the most remarkable advantages of SpRS is its independence from the laser frequency. In other words, any laser frequency can be injected into the Raman medium, although the dependence of the signal on the fourth power of the scattered frequency would indicate a preference for the shorter wavelengths. Reversing the reasoning, the independence from the laser frequency holds for any molecule and, as a consequence, simultaneous measurements of more than one chemical species is within the realistic potential of SpRS (as long as the Raman cross section is sufficiently large). This is not the sole valuable aspect of SpRS, because another striking characteristic is that, unlike LIF, collisional quenching does not matter here. This is very advantageous in that the experimental complications of a technique sensitive to the unavoidable collisional environment disappear in the present case. But, beyond the undoubted interest in SpRS as an excellent way to diagnose combustion, there is a substantial weakness. The cross section, characterizing the strength of the Raman effect, is usually much smaller than the strength of a LIF scattering process. This deficiency is experimentally circumvented by long integration times (possible with CW lasers or exciting pulses with long duration). In this manner, different diatomics and polyatomic molecules can be monitored with SpRS. Although SpRS is very common in basic combustion research, where the extreme richness of the Raman processes is well recognized [7–9], industrial applications are problematic due to the larger background luminosity that accentuates the intrinsic weakness of the Raman spectroscopic strength. Nonetheless, attempts on SpRS detection in systems of industrial interest cannot be neglected. For example, natural gas diffusion flames were studied in a prototype of industrial burners [19]. Rabenstein et al. [29] applied the technique to a diesel engine. In this case, the polarization filtering of the Raman signal and a less stringent spatial resolution contributed to the attenuation of the role played by background luminosity. Ideally, overcoming the environmental limitations, Raman probes could be devised to favor industrial process monitoring. An example of this is given by Khijwania et al. [30], who built a fiber optics Raman sensor for the detection of ethanol and methanol in gasoline solutions.
© 2011 by Taylor and Francis Group, LLC
12.3.1.3 Other Incoherent Techniques for Detection of Chemical Species If the recourse to LIF and SpRS is a common practice in advanced combustion research aimed at the detection of chemical species, many other incoherent techniques are available to the experimentalist. These include Rayleigh scattering (RS), many versions of absorption spectroscopy, resonantly enhanced multiphoton ionization, laser breakdown spectroscopy, resonant Raman scattering, laser induced incandescence (LII), and so on [6–9]. A general description of all these techniques is beyond the scope of this chapter, but the reader should be made aware of these additional spectroscopic options that depend on the optical independence of each individual molecular response to the laser field (incoherence). Some of these techniques can be applied more easily than LIF and SpRS, but their signals are less informative. For example, RS could be regarded as the simplest laser technique [7,14] (see below for the section on thermometry) but its signal is sensitive to the average number density of the gas mixture and does not distinguish among the chemical species. Other techniques considered above, despite their complications, could be more suited for particular combustion environments. For example, LII is particularly indicated for soot diagnostic, where LIF and SpRS are of little use [9]. Other examples of incoherent techniques applicable to industrial testing are laser absorption and laser induced breakdown spectroscopy (LIBS). Optical absorption is one of the basic atomic and molecular processes caused by the electromagnetic promotion of a bound electron from its initial state to one excited state [1–3,24]. When absorption is performed with a laser that crosses an absorbing medium, the transfer of some energy from the laser to the atom or molecule provokes the electronic excitation and the simultaneous disappearance of some energy from the laser beam (provided that the laser frequency is in resonance with an atomic or molecular electronic transition). One evident physical result of this combination of optical events is that the energy loss induces attenuation of the laser intensity. If this is measured, an exponential decay is found (Beer-Lambert law) on condition that the laser intensity is well below the saturation limit. Note that direct measurements of the transmitted laser light implies line-of-sight optical access and, in this respect, laser absorption is an incoherent method that differs from more popular techniques of combustion science, such as LIF and SpRS. Under these circumstances, various laser techniques could be imagined as soon as one looks for the adsorbed signal. Conventional absorption spectroscopy is recognized in industrial systems [9–11,14,22,25,26]. For this type of applications, diode lasers are primarily used. They are
276
of compact shapes, possess optimal spectral windows for absorption measurements of many molecular species and, furthermore, their commercial cost is easily affordable. If these lasers are tuned, their frequency can be scanned over the absorption lines giving rise to the so-called tunable diode laser absorption spectroscopy (TDLAS) [9–11,14,22]. More sophisticated approaches, such as photoacoustic and photothermal spectroscopies or cavity ring-down spectroscopy [7,9,14] are also applicable. The LIBS could be thought as an absorption dependent spectroscopic tool that is appreciated in industrial contexts [16,22,32]. The idea is to use a very powerful laser (e.g., a pulsed laser like Nd-YAG) that heats the target up to a point that a plasma is produced. This is possible when the laser is focused within a small volume and the increased intensity determines temperatures that can be as high as 10,000 K or more. In this manner, the original molecular constituents are broken into their isolated atomic components, which are identified thanks to the characteristic light they emit when they are produced during the plasma formation. 12.3.2 Coherent Techniques The coherence is one of the fundamental properties of light beams. This feature can be tested in experiments of interferometry [24], where the phase relationship (either spatial or temporal) is measured. Indeed, a stable phase difference (or a slowly variable phase difference) is at the core of the phenomenon called interference taught in basic courses of optics [33]. The degree of coherence revealed by interferometric tests informs about the relationship among the various sources that contribute to the formation of the light beam. Atomic or molecular sources, such as the molecules of combustion interest, make no exception. For this reason, we define as coherent techniques those spectroscopic approaches that make use of the coherent response of atoms and molecules. This response is usually mediated by strong signals, even though the optical process that generates the coherence is intrinsically weak. This aspect can be understood by pondering that the coherence is a collective phenomenon, in which molecules (or atoms) participate in coordination with each other. The coherence of the emitted macroscopic field is then the resultant cooperation of many microscopic fields, each one associated with the response of single molecules to the exiting laser fields. This fact leads to three important characteristics of the coherent techniques. They are nonlinear, fairly intense, and show directional sensitivity (phase matching). These aspects can be understood by considering that the macroscopic field stems from a vectorial sum of many microscopic fields. If these cooperate, the squared modulus of the resultant macroscopic field (note that
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
this is the quantity measured by photo-sensitive detectors) depends nonlinearly on the number of the microscopic sources, which is the number of molecules. The vectorial structure joined up with the coherent nature of the resultant field is also responsible for the directional properties. Finally, the constructive interference, typical of coherent phenomena, determines the increased strength of the signals found in coherent laser spectroscopies. The foregoing favorable characteristics are, however, tainted with critical disadvantages. The coherence is generated and controlled by more than one laser beam. Typically, three laser fields are needed for the purposes of the main popular combustion diagnostics. For this reason, it is commonly observed that coherent techniques are synonym of dielectric nonlinearity (i.e., dependence on two or more electric fields). The feature is not without consequences. On the experimental side, the minimum requirement of two laser systems and the crucial sensitivity to the optical alignment renders the measurements difficult. On the theoretical side, the data interpretation is very elaborate and much more sophisticated than the theoretical analysis of incoherent techniques. To review some of the applicative problems related to the use of coherent methods for the detection of chemical species, we concentrate on two specific examples belonging to the realm of third-order dielectric nonlinearities. One is the so-called CARS and the other is the degenerate four-wave mixing (DFWM). This choice is also suggested by the thermometric uses of these two techniques (see the corresponding section below). 12.3.2.1 Coherent Anti-Stokes Raman Scattering (CARS) The CARS is a very well-known spectroscopic method based on the mixing of electric fields traveling through a Raman medium [7–9]. As discussed before, three electric fields are necessary to produce the CARS response and, in turn, this implies that three laser beams are combined together to generate the fourth electric field, the anti-Stokes field, which gives rise to the CARS signal. Despite the difficulty of handling three interacting laser fields, the potential of CARS as an important approach for high-resolution molecular spectroscopy are recognized [1,7]. The essential features of CARS can be summarized as follows. A typical CARS set-up involves the use of a pump laser beam at frequency ω1 and a red-shifted laser beam at frequency ωS (Stokes beam). These beams are crossed within the Raman medium and, when their frequency difference equals a vibration–rotation or a purely rotational transition in the molecule, a third external laser beam at frequency ω2 can probe (in a fashion
277
Laser Measurements
similar to SpRS) the resonantly enhanced molecular transitions at the frequency difference ω1−ωS. The resultant oscillating polarization of the medium generates the measured anti-Stokes signal at ω1 + ω2− ωS. The CARS measurements can be planned in two different ways depending on the spectral characteristics of the Stokes beam. If this is narrowband, its frequency can be scanned across the Raman transitions. The final spectrum will be given by the piecewise reconstruction of the whole scan of the Stokes frequency. Alternatively, if the Stokes beam is broadband, the spectrum can be obtained within the duration of a single laser pulse. In the former experimental choice, the narrowband approach constitutes a measure to guarantee high spectral resolution. In the broadband approach, the instantaneous spectral acquisition is, instead, the preferred criterion. This applies very often to the nonstationary phenomena of combustion and, in general, the simplest experimental set-up for combustion diagnostic consists of two pump frequencies from a single powerful laser (ω1 = ω2), while the Stokes frequency ωS is taken from a broadband laser (degenerate multiplex or broadband CARS). Unfortunately, one constituent at a time can be monitored in this case and, to overcome such a disadvantage, more complex set-ups, based on multicolor approaches have been introduced in the past [7]. The detailed interpretation of spectral shapes of CARS measurements is a known subject in the specialized literature (see the references in the book of Eckbreth [7]). For instance, the fundamental importance of the optical alignment of the laser beams is at the base of various geometric schemes suggested over the years to fulfill the phase-matching condition. Another relevant aspect is connected to the role of CARS background, which originates from the spectrally featureless nonresonant Raman interaction. Indeed, the background can alter dramatically the shape of a spectrum because it mixes coherently with the resonant components of the total CARS field. For this reason, the background cannot be cavalierly dismissed and, actually, it is sometimes used for diagnostic of chemical species. But, the spectral synthesis of CARS spectra is also strongly dependent on molecular parameters (vibrational and rotational energy levels, Raman linewidths and their broadening coefficients, etc.) and the situation is even worse when the pump lasers have spectral widths comparable to the Raman linewidths. However, despite the problematic modeling of CARS measurements, spectral signatures of important species of combustion interest, diatomics such as H2, N2, O2, CO, NO, or the triatomics H2O, CO2 can be worked out. Finally, industrial applications of this valuable technique are known [12,13] and chiefly considered for thermometry. More insight into this aspect of CARS can be gained in the specific
© 2011 by Taylor and Francis Group, LLC
chapter by Hughes, Parameswaran, and Lacelle of this book. 12.3.2.2 Degenerate Four Wave Mixing (DFWM) The DFWM is a coherent technique that is similar to CARS as to the requirement of three laser fields mixed together inside the optical medium under examination [7–9]. But, unlike CARS, the three input fields have the same frequency. This is chosen in such a manner to equal the frequency of an electronic transition in a given molecule, which responds with an output signal at the same frequency and wave vector controlled by the phase matching, just as it happens for other coherent techniques (i.e., the vectorial sum of the four wave vectors, three relative to the incident fields and one relative to the generated field, must vanish). This very rapid summary of the underlying physics of DFWM clarifies that this spectroscopic technique combines successfully the benefit of the coherence with the strong interactions established by electronic transitions as in LIF. In principle, this fact makes DFWM very attractive for combustion research. However, the DFWM nonlinearity is more intricate than the CARS nonlinearity and its understanding has prevented DFWM from the widespread application that is distinctive of CARS. The spectroscopic basics were formulated in the 1990s [7,9] and, according to these findings, the role of collisions, atomic motion, internal molecular states, crossed polarization of the laser fields, and phase matching geometry can be properly explained. In rarefied gases, the saturable absorption is the dominant mechanism and, in particular, the dependence on the collisional environment (typical of LIF and CARS data) can be softened by operating with very intense pump laser beams. Experimentally, the detection efficiency can also be enhanced by the simple expedient of working with an intense probe laser beam, even though this solution on the one hand enhances the DFWM scattering and on the other complicates the theoretical analysis. Other experimental aspects come into play for a correct application of DFWM. These refer to the relative comparison between the laser linewidths and the Doppler and collisional broadenings. As a matter of fact, the two limits of narrowband and broadband laser line have important consequences for the quantitative analysis of chemical species. The effect of absorption, caused by the frequency degeneracy, is also problematic in quantitative DFWM and has to be evaluated in some instances. To sum up, an all-embracing interpretation of DFWM measurements is not available as it is, by contrast, for CARS. Nevertheless, careful choice of the experimental conditions and rigorous analysis of the corresponding theoretical approximations can lead to reasonable results for both thermometry and concentration measurements, as
278
long as laboratory flames are investigated. Applications to practical devices are very recent and limited to NO detection in spark-ignition engines [9,34,35]. Finally, it must be added that the electronic resonances shared with LIF, demonstrate their usefulness when attempting one-dimensional or two-dimensional DFWM mapping [7,9].
Industrial Combustion Testing
with a strong (either linearly or circularly polarized) pump beam. Both beams are chosen in resonance with a given molecular transition. As a consequence, the strong beam is responsible for an induced optical anisotropy (birefringence) that the probe beam senses.
12.3.2.3 Other Coherent Techniques for the Detection of Chemical Species
12.4 Laser Measurements for Velocimetry
There are many coherent approaches besides the thirdorder wave mixing leading to CARS and DFWM. Some examples are related to the family of Raman effects and can be considered, to some extent, companion techniques of CARS (stimulated Raman scattering, stimulated Raman gain spectroscopy, stimulated Raman loss or inverse Raman scattering, Raman induced Kerr effect spectroscopy, photoacoustic Raman spectroscopy, electronic resonance CARS, etc.) [7,9]. The key point of these techniques is, obviously, the Raman interaction that has been discussed for SpRS and CARS. The differences are, instead, related to the different connection between the laser excitation of Raman transitions and their interplay with the mechanism of generation of the detected signal. For example, if CARS emphasizes the detection of the anti-Stokes signal, the Stokes counterpart is at the heart of the stimulated Raman gain spectroscopy. Wave mixing of two electric fields can give rise to second-order effects of nonlinear optics [4]. One of these is the harmonic generation that converts the fundamental wavelength of a laser into its half (see Section 12.2.2). But, if electric fields at different frequencies are used, the response of a medium with sufficient second-order dielectric susceptibility can be frequency shifted to the sum and the difference of the two laser frequencies [4]. In particular, sum frequency generation (SFG) is often used to study surfaces and has found applications to examine catalytic combustion [9,36] In direct connection with DFWM, laser induced thermal gratings spectroscopy (LITGS) is another example of minor nonlinear techniques applied to combustion at high pressure corresponding to the regime of dominant collisional quenching [7,9]. Thermal gratings can arise in this condition inasmuch as the interference between pump and probe laser pulses results in a spatial modulation of the absorption followed by collisional quenching that delivers heat to the medium. The density fluctuation, subsequent to the heat wave, determines a periodic variation of the scattering of a CW laser beam. Based on this phenomenology, thermometry and concentration measurements have been demonstrated. Another coherent technique sensitive to combustion conditions is polarization spectroscopy (PS) [7,9]. Here, a weak linearly polarized probe laser beam is crossed
Any experiment of fluid-dynamics faces the problem of measuring dynamical variables without perturbing them. Clearly, this general rule is violated by the introduction of measuring devices, whose lifetime might be seriously shortened by the harsh environments found in combustion systems. The viable alternative is represented by laser-based techniques that have reached an unmatched level of accuracy as to the measurement of fluid velocities [6,37,38]. These techniques, however, cannot measure the fluid-dynamical state on the spatial scale of the molecular constituents of the fluid and, therefore, seeding with solid particles is the inescapable operative condition for the application of laser methods. Alternatively, if droplets are present (like in combustion of atomized liquid fuels or in seeding with oil droplets), these can function as laser light scatterers and the recourse to seeding with solid particles can be avoided. Mie scattering is the optical phenomenon at the base of laser measurements of particle velocities. This is a type of elastic scattering (i.e., scattered frequency equal to the incident frequency) that takes place whenever the particle dimension d is much greater than the incident wavelength λL [6,7,37]. As a result, the minimum size d of the particles should obey the condition d >> λ L. The relationship holds typically for wavelengths belonging to the visible spectrum, where laser wavelengths are definitely smaller than the particles (e.g., λL = 532 nm corresponding to the second harmonic wavelength of a standard Nd-YAG laser or λL = 488 and λL = 514 nm for the two main lines of an Ar-ion laser). The conventional approaches to anemometry are laser Doppler velocimetry (LDV) and particle image velocimetry (PIV) [6,37,38]. The former is employed to obtain point-like measurements of one velocity component, whereas a combination of two or three LDV systems allows for the measurement of the vectorial structure of the velocities. The PIV is, instead, conceived for twodimensional acquisition of velocity fields. One important industrial application of these techniques is in laser diagnostic of gas turbines and engines [9]. For instance, atomization of liquid fuels into droplets is typical of modern IC engines and one can study the fuel–air mixing that is an essential factor in efficient combustion. In
© 2011 by Taylor and Francis Group, LLC
279
Laser Measurements
the following sections we present an overview of these techniques and their possible sources of error. 12.4.1 Laser Doppler Velocimetry (LDV) The LDV is a nonintrusive method for the measurement of fluid velocities that could range from few millimeters per second up to supersonic flows. Fluids that are transparent to the laser beam and properly preconditioned with reflecting tracing particles qualify for LDV measurements [6,37]. These are grounded on the Doppler effect consisting in a frequency shift (Doppler frequency) of the laser wavelength when reflected by moving particles. Being the light frequency very high (on the order of 1014 Hz), the direct measurement of the tiny Doppler-shifted frequency is a difficult task, but the use of two laser beams intersecting each other within a small volume makes it possible to quantify the Doppler shift independently from the laser frequency (differential Doppler anemometry). The pictorial view on this type of anemometry is provided by the so-called fringe model exemplified in Figure 12.2. When two coherent and monochromatic laser beams are made to cross, their wave fronts interact creating planes of light and darkness in correspondence of the creation of constructive and destructive interference. The resulting Doppler frequency is: fD = 2 sin(θ/2)U/λ L ,
(12.1)
where θ is the angle between the propagation directions of the two laser beams and U is the projection of the particle velocity on the normal to the fringe planes. When a particle travels across the stack of fringe planes, its scattered light reproduces the oscillation between the zones of interference and the velocity U becomes a measurable quantity by tracing with an oscilloscope the time behavior of the signal. It is important to underline that, since the angle θ and the laser wavelength λ L are known constants, LDV measurements are free from calibration needs. Turning to some practical aspects of LDV, the prescribed experimental arrangement consists of a laser Beam 2 t1
e fron
Wav
Beam 1
Interference fringes
Wave
front
Figure 12.2 Fringe pattern resulting from two intersecting laser beams.
© 2011 by Taylor and Francis Group, LLC
2
source, a beam splitter, a convergent lens and a detector (i.e., a photomultiplier that converts the scattered light into a current signal). The detector position with respect to the measurement volume does not modify the Doppler frequency but it influences the scattering amplitude, which is a function of the point of view. Thus, the detector position cannot be completely arbitrary. In point of fact, the scattering is maximum in the forward direction (i.e., along the propagation direction of the laser beam) and is substantially smaller in the opposite direction (backward scattering). At first sight, forward scattering may look like a better choice, but backward scattering is easier to arrange (one single optical access) and, on the plus side, lens and detector can be enclosed in a common case. The principles of operation described so far have an important limitation. While it is possible to measure absolute values of the velocity component along the normal to the fringe pattern, its direction remains unknown. This happens because the Doppler frequency associated with the fringe pattern is invariant to the sign of the velocities. The snag is corrected by moving the interference fringes with the help of an electromechanical device, called Bragg cell, which shifts the wavelength of one of the two laser beams. Another limitation of LDV is relative to the scalar nature of the velocity measurement. Indeed, in the experimental geometry of Figure 12.2, nothing can be said about those velocity components that lie parallel to the fringe planes. The use of additional laser beams, in a fashion similar to Figure 12.2 but with rotated fringe planes, introduces velocity measurements along other directions. This secondary system of laser beams is taken at a different wavelength for spectral discrimination and, for this reason, the Ar-ion laser is the best candidate for two-dimensional LDV measurements. Extension to three-dimensional data are possible with a third set of fringes. 12.4.2 Particle Image Velocimetry (PIV) The PIV differs from LDV in many respects [6,38]. The laser beam is pulsed (the Nd-YAG laser is the standard choice) and focused with cylindrical lenses in order to illuminate the particles with a sheet of light of considerable size. This means that PIV is an imaging technique according to which the information about the velocity comes from a comparison between two successive images of the scattered light captured by a CCD camera that freezes the particles at the positions of the two subsequent (and known) instants of the laser pulses. Each image is then divided into subregions, usually called interrogation windows (IWs) and a cross-correlation algorithm between two corresponding IWs yields a map of vector displacement, each representing the average
280
Industrial Combustion Testing
displacement of the particles contained in the IWs. The knowledge of the pulse to pulse delay or, equivalently, the lag between the recorded images leads to the definition of a vector velocity map. If the particle inertia is negligible compared to flow velocity fluctuations, then the particle velocity can be assumed as the flow velocity (lagrangian particle). The PIV may be prone to several inaccuracies. Among them, the failure of the above-mentioned negligibility of particle inertia. If so, an erroneous estimate of the fluid velocity can be expected. To evaluate whether or not the particles follow the fluid, we resort to the Stokes number St. This is the ratio between the particle relaxation time (time taken by the particle velocity to conform with the velocity of the surrounding fluid) and characteristic time scale of the flow. A vanishing St indicates a lagrangian particle, while large values (larger than one) suggest very massive particles. The calculation of St is based on the Stokes expression of the relaxation time τp = ρp dp2/18μ (where dp is the diameter of a particle of density ρp and μ is the fluid viscosity that is, in turn, related to the fluid density ρf) divided by a flow characteristic time, for instance the Kolmogorov time τη = ν0.5 ε–0.5 (with ν the kinematic viscosity and ε the turbulent energy dissipation rate). In this case, the Stokes number reads
Stη = τ p /τ η = ρ p d 2 /(18ρ f ν3/2 ε −1/2 ),
(12.2)
which is proportional to T–5/4 with the temperature. This dependence underlines that a particle crossing the flame front sees a reduced Stokes number, hence most stringent inertia limitations should act on the cold reactant side. But this conclusion should be refined in light of the thermal expansion that lowers the Stokes number. These two effects are in competition with each other and particular care has to be taken in the selection of the particle dimension. Usually, a good choice is in the range of 1 to 10 μm. Another source of worries in PIV measurements is the thermophoretic effect, which could be significant in presence of strong thermal gradients and light tracers. A sudden temperature difference between nearby regions leads to a net migration of particles from the hot side toward the cold side. For instance, the thermophoretic diffusivity Dtp is estimated to be approximately 0.5 ν for alumina particles with diameters between 2 nm and 0.4 μm [39,40]. The thermophoretic velocity, vtp = –Dtp ΔT/T, for typical alumina particles, is vtp ≅ 0.5 ν ΔT/(δth T), which takes values on the order of 0.1 m/s, with δth the thermal thickness of the flame front. When the flow velocity is on the order of 10 m/s, this source of error, usually localized within the flame front, can be considered negligible.
© 2011 by Taylor and Francis Group, LLC
Further concerns regard other technical aspects. For example, out-of-plane motions of the particles caused by normal-to-laser-sheet velocity fluctuations, or the maximum parallax angle related to perspective distortions that grow from the center to the edges of the images.
12.5 Laser Measurements for Particle Sizing The stereotypical combustion is thought to be caused by the reaction of gaseous mixtures. This is not always true. Liquid fuels are often used and their injection into the combustion chamber through atomizers results in a spatial distribution of droplets (spray). The distribution and the dimension of the droplets are important parameters that can alter the combustion and, for this reason, it is important to gain insight into the problems connected to the measurements of such parameters. To this end, we will summarize very briefly the laser methods that are suited to the purpose. Much more extensive reviews can be found in textbooks and specific papers [6,9,37]. There, one can learn that applications to industrial combustion testing are found in studies about turbines and engines, but laser methods of particle sizing in the analysis of industrial processes are of more general applicability and their uses are reviewed in the detailed work by Black, McQuay, and Bonin [21]. 12.5.1 Laser Doppler Velocimetry (LDV) This method is an extension of the method LDV outlined before and used to acquire information about particle velocities (see Section 12.4.1) [6,37]. In this method, two laser beams created by the same laser are focused within the particle field. Since the two beams are optically coherent, their intersection is characterized by interference fringes (Figure 12.2). A particle traversing the fringes will scatter light according to the fringe pattern and the Doppler signal will result in a periodic structure, whose visibility (the difference between maximum and minimum values of the signal divided by their sum) is a function of the particle diameter. Unfortunately, this method has several inconveniences that make it difficult to apply. To begin with, one must ensure that one particle at a time is present in the interaction volume. The latter is ordinarily on the millimetric scale and most sprays have denser concentrations. Furthermore, the particles cannot be smaller than the resolution limit and, on the other hand, too large particles may exceed the fringe spacing. Another restriction is imposed by the spatial profile of the laser beam. Indeed, it is fundamental that the particle passes through the center of the beam.
281
Laser Measurements
12.5.2 Phase Doppler Anemometry (PDA) or Velocimetry (PDV) The objective of PDA is a simultaneous measurement of the velocity and size of the particles [6,37]. This method is an extension of the LDV principles. With a similar experimental set-up and interaction region (Figure 12.2), the analysis of the scattered light is this time more rigorous. Indeed, under the approximation of geometrical optics, the light scattered by a spherical particle consists of different contributions, due to the contemporary phenomena of reflection and refraction. In other words, part of the incident light is reflected from the surface of the particle (assumed of spherical shape), whereas another contribution arises out of the refracted light that penetrates the particle and emerges afterward as though it was reflected from the center of the particle. The initially reflected radiation travels a shorter distance compared to the distance covered by the refracted radiation. This results in a phase difference of the two signals measured as if they were obtained from two LDV measurements recorded at different scattering angles (chosen in such a manner to mutually offset one signal against the other). The final information about the particle dimension is extracted from the dependence of the measured phase difference on the particle dimension itself. The PDA works well regardless of the dimensions of the particles for sizes between a few microns up to millimeters. At larger diameters (larger than the laser beam) the measure will depend on the local curvature of the particles. However, for large particles still below the limit of curvature effects, the phase shifts could exceed 2π and, consequently, the sizing would be indeterminate. In this case, to obviate the problem, a third detector is introduced and two phase shifts are derived from the comparison between pairs of signals. One phase shift will surely be less sensitive to the phase variation with the resultant effect of lifting the phase degeneracy.
Particle sizing is not only peculiar to the scattering of light. It can also be achieved by means of LII in the context of soot diagnostic [9]. The idea is to heat the soot particles with a powerful laser pulse. After the absorption, the particles start to radiate the excess of thermal energy that is measured as a temporal decay informative of the particle size.
12.6 Laser Measurements for Thermometry Thermometric measurements are of primary importance in the analysis and the control of combustion processes. As a matter of fact, the chemical routes, established by specific chemical reactions, depend very much on the temperature at which they take place. For this reason, the temperature is one of the most important parameters that dramatically influence the whole combustion efficiency as well as heat release and formation of the pollutants. It is then not surprising that many laser measurements, designed for diagnostic purposes, are meant to obtain an experimental evaluation of the high temperatures developed during the combustion. Various techniques can offer a solution to the problem. The most known are Rayleigh, LIF, SpRS, and CARS. They are compared in the following Table 12.2. Table 12.2 Advantages and Disadvantages of Various Thermometric Laser Techniques Advantages Rayleigh
• Independence from the laser wavelength • Spectral simplicity • Strong signal
LIF
• Strong interaction (electronic transitions) • Species specific • Linear dependence • Species specific • No calibration • Linear dependence
12.5.3 Other Laser Techniques for Particle Sizing The techniques described above can be considered point-like. This means that the information is limited to the probe volume defined by the focused laser beams. Imaging methods of particle sizing would be of note. Holography is one of these [6]. It has the advantage of three-dimensional resolution, but it is limited to large droplets and the data handling is somewhat difficult. Another possibility of imaging is the laser sheet dropsizing (LSD) that originates from the comparison of a LIF image with the Mie counterpart of the same object [9]. Since the dependences of LIF and Mie scattering on the particle diameter are different, their ratio can be examined pixel by pixel for size distribution.
© 2011 by Taylor and Francis Group, LLC
SpRS
CARS
• Species specific • Strong collimated signal • No calibration
Disadvantages • Not sensitive to specific molecules • Signal dispersed over solid angle • Perturbation of Mie scattering and stray light • Fluorescence interference (especially at short laser wavelengths) • Calibration • Signal dispersed over solid angle • Calibration • Weak interaction • Signal dispersed over solid angle • Fluorescence interference (especially at short laser wavelengths) • Complex spectra • Nonlinearity • Experimental complexity (three laser beams)
282
As shown before, Rayleigh, LIF, and SpRS are incoherent techniques, whose thermometric potential was known long ago [41]. One common character is that the measured signals are linearly dependent on the concentration of the molecular species. However, strictly speaking, the RS is elastic (the measured signal is at the same wavelength of the laser) and does not distinguish among the different molecular constituents. By contrast, LIF and SpRS are species dependent as they probe specific electronic (LIF) and rovibrational (SpRS) transitions that are firm signature of any given molecule. Similar to SpRS, CARS is also based on the detection of rovibrational transitions. Nonetheless, it is a nonlinear and coherent technique, whose usefulness in thermometry for combustion research is indisputable [7]. Understandably, the strong and highly collimated CARS signal is a clear advantage when background luminosity is troublesome. Based on this, it would seem that CARS thermometry is by far the method of choice, but this resolution is too simplistic and a more wellinformed decision might point to another spectroscopic alternative. In an effort to let the general reader take the problem in stride, we will highlight virtues and pitfalls of each technique. 12.6.1 Rayleigh Thermometry When a laser beam crosses a medium, any single molecule (and larger particles) scatters light at different angles and with the same wavelength of the incoming laser (elastic scattering). The phenomenon that gives rise to the elastically scattered radiation is termed Rayleigh scattering (RS) when the size of the scatterer is considerably smaller than the laser wavelength. This definition is important to distinguish RS from Mie scattering associated with the light scattered by larger particles, whose size is comparable with (or greater than) the laser wavelength. The distinction is not only nominal, because Mie scattering is insensitive to temperature or particle number density and, hence, cannot be turned into a diagnostic tool of thermometric interest apart from its uses for velocimetry and particle sizing (see Section 12.4 and 12.5). The RS signal is, instead, proportional to the average density of the medium and, assuming constant pressure, the ideal gas law establishes an inversely proportional relationship between RS measurements and temperature [7,9,31,41]. To this end, however, it is necessary to know the differential RS cross section. This quantity is tabulated for many molecular species, but gas mixtures are characterized by an average RS cross section obtained as linear combination of the cross sections of the individual molecular components. The coefficients of the combination are the mole fractions and this suggests that RS thermometry becomes possible only if the
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
composition of the gas mixture is somehow determined. A reasonable option is to accompany RS measurements with SpRS data that are used for the simultaneous detection of various chemical species. A first drawback of RS thermometry consists in the requirement of a stable composition of the gas mixture. If this varies between reactants and products, the RS cross section changes accordingly and so does the final outcome. In principle, it is possible to calculate correction factors and, in general, the variation of the cross section is not greater than 10% in typical combustion environments. Another problem with RS measurements is that the corresponding thermometry is subject to calibration. This implies careful evaluation of the experimental parameters (quantum efficiency of the detector, collection efficiency, laser energy, total number density, solid angle of the collection optics, optical path length, and so on), but a typical procedure relies on the ratio between the measured RS signal and a reference signal obtained from a gas of known RS cross section and temperature. There are other sources of uncertainties. One is the fluorescence interference that could be problematic when energetic laser frequencies are employed (especially for UV or near UV beams). These could be extremely useful, because the RS cross section rescales with the fourth power of the laser frequency and, ideally, a laser beam at 250 nm induces a 16-fold increase of the RS signal found for a laser beam at 500 nm. Nonetheless, the higher frequencies (or lower wavelengths) might be more easily adsorbed by the molecules that, if not collisionally deactivated, regain their fundamental state through a LIF process. The resulting fluorescence could, eventually, alter the Rayleigh measurement. Mie scattering could also be a problem if large particles are present in the flow field. Along the same lines, stray light caused by spurious reflection of the laser beam could sensibly alter the RS measurement in a way that more scattered light is measured simulating a denser gas mixture. To sum up, we can state that the best RS thermometry is realized in flames of low luminosity and free of large particles (nonsooting flames fed with dust-free air). In addition, care must be put in the mechanical enclosure (if it is present) to attenuate background laser light. Beyond all the limiting factors, RS thermometry can boast two main advantages. One is the extreme simplicity of the experimental set-up. This involves one laser of any wavelength in the visible or UV spectrum and orthogonal optical collection of the signal. The other great advantage of RS thermometry is due to the intense optical response that can be used for planar imaging without too many problems. Actually, two-dimensional RS thermometry can be considered a
283
Laser Measurements
common practice in combustion research, even though it is prone to inaccuracies on the order of 10% for singlepulse measurements. 12.6.2 LIF Thermometry The principles of LIF thermometry are common to the spectroscopic use described before for the detection of chemical species (Section 12.3.1.1). For this reason, the general concepts will not be repeated here. Instead, in the current context, we describe the modifications needed to adapt LIF for temperature determination. They could also be retrieved from the specialized literature that offers extensive reviews of the subject [7,9,27,28,31,41]. An important general remark before considering specific aspects of LIF thermometry is that the technique is not of first choice. There are some limitations due to the minimum temperature required to detect those molecules that are relevant for the measurement. For example, OH radicals, particularly suitable for LIF detection, are quantitatively significant at temperatures above 1500 K. Pollutants, such as CO and NO, or oxygen have also been considered for LIF thermometry but they suffer from a similar limitation imposed by the concentration. In other terms, temperature measurements are guaranteed only if there is a sufficient amount of the species being probed with the LIF technique. An alternative is to seed the gas mixture with an atomic or molecular species, whose LIF signal is spectrally well understood. The most significant feature of LIF thermometry resides in its incomparable intensity of the measured signal. This advantage stems from the electronic nature of the absorption process subsequent to the fine adjustment of the laser line to the transition of a molecular electron from its ground state to the excited state. The advantage is such that thermometric imaging is feasible. For example, two-dimensional OH LIF images can be manipulated to provide temperature maps in those situations where stray laser light or fluorescence interferences are detrimental to RS thermometry. There are three main approaches to LIF thermometry. All of them are keyed to the measurement of the thermal (or Boltzmann) distribution of the molecular population in known rovibrational states. For this reason, a measurement can be considered reliable on the condition that the molecular target is well characterized in terms of its spectroscopy. The differences among the approaches are briefly summarized in the next sections. 12.6.2.1 LIF Thermometry with Excitation Scans This method is very often applied to steady-state combustion. The reason is soon understood. The Boltzmann
© 2011 by Taylor and Francis Group, LLC
distribution is probed during the excitation by means of a narrowband laser whose frequency is scanned over several rovibrational transitions available to the molecular electron. This means that the whole scan takes some time to be completed and its use is meaningless if the LIF molecule does not experience a stationary environment. The thermometric information can be derived from Boltzmann plots (rovibrational energies in the vertical axis, logarithmic scale of a composite variable of spectroscopic parameters in the horizontal axis), where the resulting straight lines have negative slopes with values inversely proportional to the temperature. In this method, it is essential to know the dependence of spontaneous emission on the various molecular levels. Additionally, one should determine the importance of nonradiative disexcitation channels opened by molecular collisions (quenching). These two remarks clearly tie in the success of excitation scans with the deep knowledge of the spectroscopic behavior of the LIF molecule with respect to its surrounding combustive environment. 12.6.2.2 LIF Thermometry with Two-Color Approach The troubles of excitation scans can be overcome in the two-line approach. Here, only two transitions are probed. But the key point is that the excited molecular level is the same in both excitations that occur in a sequence where one transition is probed while the detection is made for the other transition. In this manner, the spurious dependences on quenching and other molecular parameters are canceled when the ratio between the fluorescence signals at the two distinct emitted wavelengths is measured. 12.6.2.3 LIF Thermometry with Thermal Assistance A disadvantage of the two-color approach is that two different and tunable laser sources are required. Plus, sequencing of the two laser pulses can be disadvantageous when single-pulse measurements are in demand. A solution is thermally assisted LIF. In this method, which can be regarded as a monochromatic approach to thermometry, one laser source is used to pump the ground-state electron up to a level belonging to a manifold that is sensitive to collisions with rapid transfer processes. This means that the equilibrium could be reached in a very short time after the laser excitation and the consequent fluorescence spectrum should encode the collisionally equilibrated rovibrational population. If so, the extraction of temperatures from population distributions becomes feasible. Of course, the method, although simple in its hardware, is very delicate because the whole reasoning is based on the premise of rapid collision transfers and the precise
284
knowledge of the collisional dynamics in the excited electronic manifold. 12.6.3 Raman-Based Techniques for Thermometry The whole branch of Raman diagnostics is characterized by the interrogation of the internal molecular state as configured by the vibration and rotation of molecules (rovibrational state) [1–3,5,7–9]. In other words, a Raman spectrum, irrespective of the specific technique used to generate it, incorporates spectral lines that identify transitions between two distinct rovibrational states (purely rotational lines are also possible). The energy spacing among the lines is on the order of the Boltzmann factor KBT and this suggests that the temperature T controls the molecular population among the states, which changes according to the known Boltzmann distribution. This is the key point that makes Raman-based techniques so attractive for thermometry. We discuss here the most popular approaches, which are SpRS and CARS. These have been previously described in reference to the detection of chemical species. In the present context, we point out the main differences that explain the thermometric capacity. 12.6.3.1 Thermometry with SpRS The phenomenon at the base of SpRS is an inelastic process of light scattering that takes in the interaction between the laser beam and the internal molecular state. A brief description of this interaction has been given in the application of SpRS to concentration measurements (Section 12.3.1.2) and will not be repeated here. We concentrate, instead, on the thermometric value of the technique. The extraction of temperatures from SpRS spectra can happen by following a number of different strategies. One is based on the deduction of the number density of all the main molecular species resident in the interaction volume and responsible for rovibrational lines or pure rotational spectra. Another strategy focuses on the details of the whole spectral shape that usually contains a considerable number of lines. Finally, a third approach is rooted in the ratio of integrated line intensities of Stokes and anti-Stokes transitions. It is clearly understood that, independently from the strategy considered, the application of SpRS thermometry is legitimate as long as one can bank on a proper spectral knowledge of the scattering molecules. For example, spectral convolution with the laser linewidth, influence of the Raman linewidths (including various broadening mechanisms), and slit function of the spectrograph must be taken into consideration when fitting the theoretical models to the experimental spectral shape. In doing so, the measured temperature
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
results from the minimization of a least-squares fitting routine where the temperature is one of the free parameters. The main trouble with SpRS thermometry is associated with the small Raman cross section. This fact is partly counterbalanced by choosing the SpRS spectrum of the most abundant and stable species. Usually, nitrogen is the best candidate in air-fed combustion realized in laboratory burners [7,9] and the choice proves to be valid even for industrial flames and furnaces [18–20], but this is not enough to make SpRS thermometry evolve into a common resource. For instance, the RS cross section is about three orders of magnitude greater and twodimensional thermometry is dominated by RS instead of the more accurate SpRS. 12.6.3.2 Thermometry with CARS This type of laser thermometry is described to a greater extent by Hughes, Parameswaran, and Lacelle, who report on some industrial applications of CARS in Chapter 13 of this book. For this reason, we limit ourselves to a summary of the main aspects of CARS thermometry. On the other hand, books [7,9,42] and reviews [8,43–46] unravel the physics behind CARS and the reader should refer to this further literature to grasp the extraordinary thermometric potential of such a technique. A first important note is that CARS is currently considered to be the most mature coherent technique for temperature measurements in hostile environments. However, accurate thermometry is accomplished after very complex spectral analysis that requires heavy computational work. This includes a variety of phenomena happening during CARS interaction. For instance, the coherent nature of CARS introduces constructive and destructive interferences. Spectral line mixing and collisional narrowing are observed and one must take them into account during the elaboration of data at moderate and high pressures. What is more, precise knowledge of the rovibrational energies is critical and yet insufficient if Raman linewidths have not been determined a priori. On the experimental side, the nonlinearity of CARS increases the sensitivity to practical problems such as the laser stability, beam profiles, optical alignment, background, spectral shape of the lasers, and so on. In addition, saturation of the Raman transitions, laser mode statistics, or more importantly, average effects in the presence of temperature gradients within the interaction volume are other practical troubles that worry the experimentalist. Despite the above-mentioned difficulties, CARS thermometry is widespread. The reason for this success is due to several factors. The signal is collimated and
285
Laser Measurements
therefore is exceptionally strong in comparison with other accurate laser-based thermometric diagnostics (LIF and SpRS). Limitation to temperature ranges, such as in LIF thermometry from radicals, does not matter here. Besides, there is sufficient knowledge of some of the main Raman species produced in flames, furnaces, or other combustion systems. Indeed, in analogy with SpRS thermometry, the typical CARS molecule is nitrogen that is well studied and is ubiquitous in airfed combustion environments. For other applications where nitrogen is not one of the main constituents, there are other suitable molecules (hydrogen, oxygen, carbon and nitrogen oxides, water vapor) whose CARS spectrum is well established and can be used for thermometric purposes. Another factor that promotes CARS thermometry is its versatility in view of the fact that different experimental strategies are available. Broadband degenerate CARS is typical (the two pump frequencies are from a common narrowband laser, whereas the Stokes laser is spectrally broad), but this is not the only option. For instance, multicolor approaches with different combinations of pump and Stokes lasers have been demonstrated in the past for simultaneous spectral acquisition of temperature and concentrations. One important drawback of CARS thermometry is its point-wise resolution. Although attempts at one-dimensional temperature distributions are present in the literature (these are described in Marrocco [46]), they are rather difficult to implement in real combustors. Therefore, RS and LIF outdo CARS when it comes to multidimensional thermometric information. 12.6.4 Other Thermometric Laser Techniques The CARS process emphasizes the role of the anti-Stokes signal. Under the same experimental conditions of three laser fields mixed within a Raman medium, analogous emphasis can be put in the Stokes signal or the pump fields. In the first case, if the Stokes signal is somehow detected, we have a new laser method, termed as stimulated Raman gain scattering (StRGS). In the second case, where the emphasis is on the pump signal, we define the so-called stimulate Raman loss scattering (StRLS) or inverse Raman scattering (IRS). In both cases, the dependence of the optical alignment of the laser beams is much less stringent. Unlike CARS, the beams can be arbitrarily oriented. Despite this remarkable advantage, there are a series of reasons that have prevented StRGS and StRLS from being as popular as CARS. For example, StRGS cannot be broadband and this precludes single-shot measurements. Moreover, these additional techniques work on the modulation of the pump or Stokes beam and in practical combustors the presence of particulate or optical effects (absorption, refraction,
© 2011 by Taylor and Francis Group, LLC
spurious scattering of optical surfaces) can easily alter the intensity. Another nonlinear technique that is potentially applicable to thermometric measurements is DFWM [7,9]. For instance, a Boltzmann plot constructed out of the relative line intensities of a DFWM spectrum can lead to temperature predictions that can be as accurate as CARS in some cases. An alternative method is to fit theoretical simulations to the experimental spectrum. Nonetheless, the versatility of CARS is not equaled by DFWM. In effect, single pulse measurements seems to be limited to some radical species and mode fluctuations of conventional lasers perturb the data severely. To avoid troubles with such laser intensity fluctuations, saturated DFWM is often employed, but the difficulties of spectral synthesis remain a serious hindrance to a major role of DFWM thermometry. Thermometry can also be realized with techniques that are less popular within the combustion research community. For example, LITGS and PS, mentioned before for the detection of chemical species, can be useful for temperature measurements. The principles of operation do not differ from earlier remarks. For example, the temperature in LITGS can be extracted from the measured oscillating temporal decay of the signal fitted with the theoretical model having the temperature as free parameter.
12.7 Combined Laser Measurements The incredible list of spectroscopic tools seen so far shows that several data could be collected by means of laser methods. Their results have helped us clarify the mechanism acting during the combustion and, apparently, it is imaginable to combine some of them in an effort to retrieve information about mutually interacting variables (e.g., simultaneous measurements of temperature, concentrations, and velocities). As a matter of fact, the tendency to integrate various sources of information is recognizable in the recent research on laser diagnostics applied to combustion where the complexity of the processes is such that composite knowledge is mandatory. Promising applications concern turbulent combustion, where the diagnostic of the interaction between the chaotic fluid-dynamic field and thermochemical reactions of combustion is of primary importance [9,47,48]. However, the possibility of measurements of different combustion parameters taken at the same time is a general and intriguing perspective, which could be accomplished by a suitable coupling of two or more techniques. In the next section, we discuss some examples.
286
Industrial Combustion Testing
2500
50
0.007 0.006
2000
T (K)
0.004 0.003
1000
40 y (mm)
1500
OH
0.005
0.002 500
0
30
0.001
0
2 x (mm)
4
6
0
Figure 12.3 Temperature and OH distribution in a monodimensional laminar premixed CH4/air flame (x = 0 corresponds to the pure reactants). Gray line, temperature; black line, OxH distribution. OH distribution has a steep gradient in the thermal thickness and decreases slowly in the combustion products zone.
12.7.1 Examples of Combined Measurements As anticipated in the previous sections, turbulent heat and mass transfer has a key role in nonpremixed combustion (fuel and oxidizer have to be thoroughly mixed before they react) as well as in turbulent premixed combustion, where flame fronts can be deeply altered by velocity fluctuations. This brings out the necessity of a combined measurement of scalars (concentration/temperature) and velocity field. Thermal and mass fraction distribution and derived quantities can be measured by means of Rayleigh and Raman scattering [49–51]. However, direct thermometry is spoiled by the strong Mie scattering of the tracing particles used for velocimetry. On the other hand, premixed flame front can be characterized not only by a steep rise of temperature, but also by the presence of radical molecules, such as OH and CH, whose spatial distribution across the flame front is known and distinguished from that of nonpremixed processes. The radical CH has a symmetric distribution and is highly peaked in the proximity of the front inner layer. The radical OH has, instead, a strong gradient in the thermal flame front but it decreases very slowly in the region of combustion products (see Figure 12.3). Most importantly, the thermometric variation of Figure 12.3 (vertical axis on the left side) almost overlaps with the OH distribution for small spatial displacements x and this fact is useful to identify approximately the steep temperature gradient of the flame front when LIF images of the radical species are acquired and contrasted with PIV data as in Figure 12.4. Comparisons of this kind are abundant
© 2011 by Taylor and Francis Group, LLC
10
20 x (mm)
30
Figure 12.4 (See color insert following page 424.) Combined measurements of OH-LIF distribution (image) and velocity field by PIV (arrows).
in textbooks [7,9], reviews [8,52,53] and many research articles cited therein. The combination of laser methods is not only peculiar to joint velocity and concentration measurements. For example, the laser thermometry can be coupled with other types of spectroscopic tools of experimental investigation. Examples are many, spectroscopic thermometry based on RS, or SpRS, or CARS routinely assists the complementary diagnostic activity obtained with other spectroscopic means [7–9,47,52,53].
References
1. Demtröder, W. Laser Spectroscopy. Berlin: Springer, 2003. 2. Levine, I. N. Molecular Spectroscopy. New York: John Wiley & Sons, 1975. 3. Hollas, J. M. Modern Spectroscopy. Chichester, UK: John Wiley & Sons, 2004. 4. Shen, Y. R. The Principles of Nonlinear Optics. Somerset, NJ: John Wiley & Sons, 2003. 5. McCreery, R. L. Raman Spectroscopy for Chemical Analysis. New York: John Wiley & Sons, 2000. 6. Chigier, N., ed. Combustion Measurements. New York: Hemisphere Publishing Corporation, 1991. 7. Eckbreth, A. C. Laser Diagnostics for Combustion Temperature and Species. Amsterdam: Gordon and Breach Publishers, 1996. 8. Wolfrum, J. “Lasers in Combustion: from Basic Theory to Practical Devices.” Twenty-Seventh Symposium (International) on Combustion, The Combustion Insti tute, 1–41, 1998.
Laser Measurements
9. Hohse-Höinghaus, K., and Jeffries, J. B., eds. Applied Combustion Diagnostics. New York: Taylor & Francis, 2002. 10. Jenkins, T. P., and Bergmans, J. L. “Diode Laser Temperature Measurements.” In Industrial Combustion Testing, edited by C. E. Baukal. Boca Raton, FL: Taylor & Francis, 2010. 11. Linnerud, I., Kaspersen, P., and Jæger, T. “Gas Monitoring in the Process Industry Using Diode Laser Spectroscopy.” Applied Physics B 67, no. 3 (1998): 297–305. 12. Hughes, P. M., Lacelle, R. J., and Parameswaran, T. “A Comparison of Suction Pyrometer and CARS Derived Temperatures in an Industrial Scale Flame.” Combustion Science and Technology 105 (1995): 131–45. 13. Hughes, P. M., Parameswaran, T., and Lacelle, R. J. “CARS Temperature Measurements in Flames in Industrial Burners.” In Industrial Combustion Testing, edited by C. E. Baukal. Boca Raton, FL: Taylor & Francis, 2010. 14. Monkhouse, P. “On-Line Diagnostic Methods for Metal Species in Industrial Process Gas.” Progress In Energy and Combustion Science 28 (2002): 331–81. 15. Boesl, U. “Laser Mass Spectrometry for Environmental and Industrial Chemical Trace Analysis.” Journal of Mass Spectrometry 35 (2000): 289–304. 16. Blevins, L. G., Shaddix, C. R., Sickafoose, S. M., and Walsh, P. M. “Laser-Induced Breakdown Spectroscopy at High Temperatures in Industrial Boilers and Furnaces.” Applied Optics 42 (2003): 6107–18. 17. A. Arnold, et al., “2D-Diagnostics in Industrial Devices.” Berichte der Bunsengesellshaft-Physical Chemistry Chemical Physics 97 (1993): 1650–1661 18. Müller, D. et al., “Two-Dimensional Concentration and Temperature Measurements in Extended Flames of Industrial Burners Using PLIF.” Proceedings of SPIE - The International Society for Optical Engineering 5191 (2003): 66–74. 19. Keck, O., Meier, W., Stricker, W., and Aigner, M. “Establishment of a Confined Swirling Natural Gas/Air Flame as a Standard Flame: Temperature and Species Distributions from Laser Raman Measurements.” Combustion Science and Technology 174, no. 8 (2002): 117–51. 20. Zikratov, G., Yueh, F-Y, Singh, J. P., Norton, O. P., Kumar, R. A., and Cook, R. L. “Spontaneous anti-Stokes Raman Probe for Gas Temperature Measurements in Industrial Furnaces.” Applied Optics 38 (1999): 1467–80. 21. Black, D. L., McQuay, M. Q., and Bonin, M. P. “Laser-Based Techniques for Particle-Size Measurement: A Review of Sizing Methods and their Industrial Applications.” Progress in Energy Combustion Science 22 (1996): 267–306. 22. Deguchi, Y., Noda, M., Fukuda, Y., Ichinose, Y., Endo, Y., Inada, M., Abe, Y., and Isasaki, S. “Industrial Applications of Temperature and Species Concentration Monitoring Using Laser Diagnostics.” Measurement Science and Technology 13 (2002): R103–R115. 23. Siegman, A. E. Lasers. Mill Valley, CA: University Science Book, 1986. 24. Loudon, R. The Quantum Theory of Light. Oxford: Oxford University Press, 2000. 25. Allen, M. G. “Diode Laser Absorption Sensors for GasDynamic and Combustion Flows.” Measurement Science and Technology 9 (1998): 545–62.
© 2011 by Taylor and Francis Group, LLC
287
26. Willer, U., Saraji, M., Khorsandi, Geiser, P., Schade, W. “Near- and Mid-Infrared Laser Monitoring of Industrial Processes, Environment and Security Applications. ” Optics and Lasers in Engineering 44 (2006): 699–710. 27. Kohse-Höinghaus, K. “Laser Techniques for the Quantitative Detection of Reactive Intermediates in Combustion Systems.” Progress in Energy Combustion Science 20 (1994): 203–79. 28. Daily, J. W. “Laser Induced Fluorescence Spectroscopy in Flames.” Progress in Energy Combustion Science 23 (1997): 133–99. 29. Rabenstein, F., Egermann, J., Leipertz, A., and d’Alfonso, N. “Vapor-Phase Structures of Diesel-Type Fuel Sprays: an Experimental Analysis” SAE Paper 982543, 1998. 30. Khijwania, S. K., Tiwari, V. S., Yueh, F.-Y., and Singh, J. P. “A Fiber Optic Raman Sensor for Hydrocarbon Detection.” Sensors and Actuators B-Chemical 125 (2007): 563–68. 31. Miles, R. B., Lempert, W. R., and Forkey, J. N. “Laser Rayleigh Scattering.” Measurement Science and Technology 12 (2001): R33–R51. 32. Lee, W.-B., Wu, J., Lee, Y. I., and Sneddon, J. “Recent Applications of Laser-Induced Breakdown Spectrometry: A Review of Material Approaches.” Applied Spectroscy Reviews 39 (2004): 27–97. 33. Hecht, E. Optics. San Francisco, CA: Addison Wesley, 2002. 34. Grant, A. J., Ewart, P., and Stone, C. R. “Detection of NO in a Spark-Ignition Research Engine Using Degenerate Four-Wave Mixing.” Applied Physics B 74 (2002): 105–10. 35. Stevens, R., Ewart, P., and Ma, H. “Measurement of Nitric Oxide Concentration in a Spark-Ignition Engine Using Degenerate Four-Wave Mixing.” Combustion and Flame 148 (2007): 223–33. 36. Pfefferle, W. C., and Pfefferle, L. D. “Catalytically Stabilized Combustion.” Progress in Energy Combustion Science 12 (1986): 25–41. 37. Albrecht, H. E., Borys, M., Damaschke, N., and Tropea, C. Laser Doppler and Phase Doppler Measurement Techniques. Berlin: Springer, 2003. 38. Raffael, M., Willert, C., and Kompenhans, J. Particle Image Velocimetry: A Practical Guide. Berlin: Springer, 1998. 39. Frank, J., Kalt, P., and Bilger, R. “Measurements of Conditional Velocities in Turbulent Premixed Flames by Simultaneous OH PLF and PIV.” Combustion and Flame 116 (1999): 220–32. 40. Filatyev, S., Driscoll, J., Carter, C., and Donbar, J. “Measured Properties of Turbulent Premixed Flames for Model Assessment, Including Burning Velocities, Stretch Rates, and Surface Densities.” Combustion and Flame 141 (2005): 1–21. 41. Laurendeau, N. M. “Temperature Measurements by Light-Scattering Methods.” Progress in Energy Combustion Science 14 (1988): 147–70. 42. Levenson, M. D., and Kano, S. S. Introduction to Nonlinear Laser Spectroscopy. Boston, MA: Academic Press, 1988. 43. Hall, R. J. “CARS Spectra of Combustion Gases.” Combustion and Flame 35 (1979): 47–60. 44. Druet, S. A. J., and Taran, J.-P. E. “CARS Spectroscopy.” Progress in Quantum Electronics 7 (1981): 1–72.
288
45. Greenhalgh, D. A. “Quantitative CARS Spectroscopy.” in Advances in Nonlinear Spectroscopy, edited by R. J. H. Clark and R. E. Hester, New York: John Wiley & Sons, 1988. 46. Marrocco, M. “CARS Spectroscopy.”, in Handbook of Combustion Vol 2: Combustion Diagnostics and Pollutants, edited by M. Lackner, F. Winter and A. K. Agarwal, Weinheim: Wiley-VCH, 2010. 47. Masri, A. R., Dibble, R. W., and Barlow, R. S. “The Structure of Turbulent Nonpremixed Flames Revealed by Raman-Rayleigh-LIF Measurements.” Progress in Energy Combustion Science 22 (1996): 307–62. 48. Beeck, M.-A., and Hentschel, W. “Laser Metrology - A Diagnostic Tool in Automotive Development Processes.” Optics and Lasers in Engineering 34 (2000): 101–120. 49. Veynante, D., and Vervisch, L. “Turbulent Combustion Modeling.” Progress in Energy Combustion Science 28 (2002): 193–266.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
50. Wang, G.-H., Clemens, N. T., Varghese, P. L., and Barlow, R. S. “Turbulent time Scales in a Nonpremixed Turbulent Jet Flame by Using High-Repetition Rate Thermometry.” Combustion and Flame 152 (2008): 317–35. 51. Wang, G., and Barlow, R. S. “Spatial Resolution Effects on the Measurement of Scalar Variance and Scalar Gradient in Turbulent Nonpremixed Jet Flames.” Experiments in Fluids 44 (2008): 633–45. 52. Hassel, E. P., and Linow, S. “Laser Diagnostics for Studies of Turbulent Combustion.” Measurement Science and Technology 11 (2000): R37–R57. 53. Kohse-Höinghaus, K., Barlow, R. S., Aldén, M., and Wolfrum, J. “Combustion at the Focus: Laser Diagnostics and Control.” Proceedings of the Combustion Institute 30 (2005): 89–123.
13 CARS Temperature Measurements in Flames in Industrial Burners Patrick M. Hughes, Thangam Parameswaran, and Richard J. Lacelle Contents 13.1 Introduction................................................................................................................................................................ 289 13.2 CARS Theory.............................................................................................................................................................. 291 13.3 Phase Matching in the CARS Configuration......................................................................................................... 292 13.4 Experimental Setup................................................................................................................................................... 293 13.5 Laser Considerations................................................................................................................................................. 296 13.6 Data Acquisition........................................................................................................................................................ 299 13.7 Analysis of CARS Spectra........................................................................................................................................ 300 13.8 CARS Measurements made in Large, Practical Flames....................................................................................... 300 13.9 Measurements in Coal and Oil Flames.................................................................................................................. 301 13.10 Measurements in Natural Gas Flames................................................................................................................... 303 13.11 Summary..................................................................................................................................................................... 306 Acknowledgments.................................................................................................................................................................. 307 References................................................................................................................................................................................. 307
13.1 Introduction Temperature measurement plays a major role in the understanding of combustion (see Chapter 5). In the design of a burner for an industrial process, the distribution of the gas temperature is very important. For most industrial combustion processes the dominant methods of heat transfer include radiation and convection, which are both dependant on the gas temperature. The burner designer will modify his design to achieve the temperature distribution that suits the industrial process. In recent times the use of computational fluid dynamic (CFD) computer models are being used to design burners and furnaces. Along with other measurements, the gas temperature inside the furnace is used by the modeler to explore the regions of reaction and heat release in the flame and to evaluate the model predictions. This chapter deals with a noninvasive method of measuring the gas temperature in furnace flames. Conventional techniques used for measuring temperature require that a probe be inserted at the test point in the measurement field. They are therefore termed intrusive techniques. The influence of the temperature
of the gases on the probe tip, usually a thermocouple, is monitored. The very presence of the probe in the measurement field causes a distortion of the flow field, which influences the mixing processes. In a reacting flow this distortion may result in changes in the chemical composition and temperature in the vicinity of the probe. Intrusive techniques require that the probe be protected with a coolant jacket and, as a result, heat is extracted from the flame, which further disturbs the temperature field. These disturbances to the temperature and velocity fields and to the mixing processes can lead to significant errors in temperature measurement. These errors, coupled with those inherent in the measurement technique, can result in significant differences between the measured and the actual values [1,2]. Further, as a result of limitations of the intrusive measurement technique, the measured values are usually limited to timeaveraged quantities. Unlike intrusive thermocouple probes, optical techniques of temperature measurement do not disturb the combustion process (see Chapter 12). Sodium linereversal is an optical technique first used for measuring flame temperature by Kurlbaum [3] as early as 1902. Although this technique has been improved through the years, line-reversal methods use line of sight optics
© Her Majesty the Queen in Right of Canada, as represented by the Minister of Natural Resources, 2009.
289 © 2011 by Taylor and Francis Group, LLC
290
Industrial Combustion Testing
and measure only the average temperature along a line through the flame. Measurements based on this technique also have poor temporal resolution and are not very useful in turbulent flames. With the advent of high-energy lasers that are coherent light sources, a new tool was made available to investigators studying gaseous flow fields. Pulsed, high-powered lasers make it possible to make light scattering and absorption measurements in a gas with high spatial and temporal resolution. Laser-based measurements are nonintrusive and many techniques such as Rayleigh scattering, Raman scattering, laser induced fluorescence, and Coherent anti-Stokes Raman spectroscopy (CARS) have been described and their application to combustion diagnostics well reviewed in literature [4,5]. This chapter deals with the technique termed CARS for the measurement of flame temperature. The CARS has been found to have better spatial and temporal resolution than other nonintrusive techniques [6,7,8]. The CARS technique involves the cofocusing of two laser beams at a point in the test section (Figure 13.1). At the point of intersection of the laser beams, a threephoton interaction involving two photons at the pump frequency and one at the Stokes frequency produces a coherent beam at the anti-Stokes frequency. This radiation emerges from the test section as a collimated, laser-like beam. With a pulsed laser, spontaneous Raman measurements can be performed with a short time resolution; however, the measured signal is incoherent and not directional. That is to say the measured signal is captured from light that is scattered over a wide region in space. Using the Raman scattering techniques the spatial resolution can only be improved at the expense of the signal-to-noise ratio. Unlike the Raman process, which uses a single laser and produces a scattered beam that is isotropic in space and weaker in intensity, the CARS beam is highly directional and easier to collect. The CARS signal is at a higher frequency (anti-Stokes side) than the pump signal and hence it is free from interferences caused by fluorescence. As a result, the signal-to-noise ratio for CARS experiments is much better than that for other light scattering techniques. Until the advent of the high-powered laser sources, CARS remained in the realm of theoretical, non-linear Test section
CARS ωas
spectroscopy. Whereas the underlying Raman process was first predicted by Smekal [9] in 1923 and subsequently discovered by Raman and Krishnan in 1928 [10], the CARS phenomenon was first discovered by Maker and Terhune in 1965 [11] and assigned its name by Begley [12] in 1974. It was the work of Taran and Regnier in 1973 [13] that initiated the use of Raman spectroscopy in gas phase diagnostics. Since its introduction there have been many improvements added to the CARS technique such as the USED-CARS and BOX-CARS [4] configuration and the use of more than two lasers. Also a variety of novel applications have been found that exploit the advantages of the CARS technique and thus give the combustion scientist expanded capabilities. For combustion measurements, the advantages of CARS over other techniques are varied. The attainable spatial resolution of CARS, typically 100 μm by 1.0 to 5.0 mm [14], is much better than any of the intrusive techniques and most of the other optical techniques. The CARS signal strengths can be as much as 106 –1010 times that from other scattering techniques [15]. Another advantage of the CARS technique is that because the generated signal resides in the anti-Stokes region, the measured signal is free of fluorescence interferences. Since the duration of the laser pulses is on the order of 10 ns, the CARS measurement has the advantage of being almost instantaneous in time. Further, as the lasers continue pulsing, many CARS signals can be collected to define such statistics as the probability distribution function (PDF) of the measurement. This type of information about temperature in a flame is not possible using classical measurement techniques. Over the years, the literature shows that the CARS technique has been used to take measurements in harsh environments in many different applications, some of which are listed below:
1. The combustion chamber of an automobile engine [16–19] 2. Ballistic measurements for solid propellants [20] 3. Chemical process measurements [21–23] 4. Turbo-machinery [24–28] 5. Fundamental combustion studies [29–35] 6. Industrial furnace measurements [36–43]
ωp
ωs Figure 13.1 CARS interaction at a test point.
© 2011 by Taylor and Francis Group, LLC
Among the sample applications listed above, the last is of particular interest in this chapter that reviews the basic theory and discusses the practical aspects of a CARS setup for the measurement of gas temperature in a large research furnace.
291
CARS Temperature Measurements in Flames in Industrial Burners
13.2 CARS Theory The CARS measurement system consists of two high powered pulsed lasers, optics and detection equipment, and specialized software for a system dedicated to the acquisition and analysis of CARS spectra for the time resolved point measurements of temperature in a large scale flame. The theory of CARS has been dealt with extensively [14,15,30,44,45] and only a brief outline is provided here. The CARS is a non-linear, optical phenomenon, and involves three-wave mixing produces a resonant. When two laser beams of frequencies (ωp and ωs) interact in a medium, three-wave mixing, it produces a resultant coherent beam with the frequency 2ωp – ωs, which is the CARS beam as shown in the energy level diagram in Figure 13.2. The mixing occurs for all samples but the intensity of the CARS signal is greatly enhanced when ωp – ωs approaches a Raman frequency (ωv) of the test molecules in the medium. The interaction of laser radiation with the medium occurs through the third-order, non-linear, electric susceptibility denoted by χ(3) and gives rise to an induced polarization field, which acts as a source term in Maxwell’s wave equation. On solving the wave equation, one arrives at the following expression for the intensity of the CARS signal: 2
I (ω as ) ∝ I p2 I s χ( 3) .
(13.1)
In this equation the CARS intensity I(ωas) varies l inearly with the Stokes intensity Is, quadratically with the pump intensity Ip, and is also proportional to the square of the third-order susceptibility χ(3), which has resonant and non-resonant contributions: χ( 3) = χ R + χ NR .
(13.2)
Energy level diagram Strokes ωp
ωp
The non-resonant component χNR has only a weak dependence on frequency and varies linearly with number density. The resonant component χR for a transition with index j, is expressed as
ωv
Energy conservation: ωas = 2ωp – ωs Figure 13.2 CARS energy level diagram.
© 2011 by Taylor and Francis Group, LLC
Anti-stokes CARS signal
(13.3)
Kj =
4 πNc 4 d σ −1 ∆j Γj , hω s4 dΩ j
(13.4)
and sum over all such ωj where the j index refers to the vibrational and rotational quantum numbers, the total CARS susceptibility χ(3) is expressed as:
χ( 3 ) =
K jΓ j
∑ 2∆ω − iΓ j
j
+ χ NR .
(13.5)
j
The Raman cross-section dσ/dΩ is a function of the properties of the vibration-rotation transitions being summed. In the above analysis, the pump frequency (ωp) and the Stokes frequency (ωs) have been assumed to be ideally monochromatic, which is applicable when each vibration-rotation line is scanned. In a broadband CARS system as described here the Stokes (dye) laser has a broad spectral profile so that the multiplexed CARS spectral profile of the probed species, is generated with each laser pulse. In practice, the pump laser has a narrow but finite spectral width and CARS the intensity I(ωas) can be obtained by convolving over the laser line widths and is given by Yuratich [46] as the integral
Pump
∆j 4 πNc 4 dσ , 4 hω s dΩ j 2 ∆ω j − iΓ j
where N is the number density, ωj is the Raman frequency, Δj is the population difference between the upper and lower vibration-rotation states for the transition, Γj is the Raman line width, and (dσ/dΩ)j is the Raman-scattering cross section and Δωj is the frequency detuning. If we define
ωas
ωs
χR =
∫
2
I (ω as ) = I p I s (ω (po ) − δ)I p (ω as − δ) χ(δ) dδ , (13.6)
where: ω(o) p is the center frequency of the pump laser. Usually a Gaussian distribution may be assumed for the spectral profiles of the pump and the Stokes laser. In actual practice the convolution over the dye profile is eliminated by dividing the experimental CARS spectrum with a non-resonant reference spectrum collected with a suitable reference gas as the medium. When a multimode pump laser is used the Yuratich approximation above will lead to errors in temperature
292
Industrial Combustion Testing
+ 1/2 χ R ( ω as − ω ′) 2
(13.7)
0.6 0.4
0.0 21070 21080
+ 1/2 χ R ( ω as − ω ′ ) χ R ( ω as − ω ″ ) ,
21090
21100 21110 21120 Wavenumber cm–1
21130
21140
Figure 13.3 Effect of temperature on the CARS spectral profile for Q-branched nitrogen.
where:
2400°K 2100°K 1800°K 1500°K 1200°K
0.8
0.2
I (ω as ) = χ 2 NR 1 + 2 χ NR Re χ R (ω as − ω ′)
1.0
Normalized intensity
and concentration estimates if the laser width is comparable to the Raman-line width. One way to avoid this error it is to use a single-mode pump laser. When a multimode pump laser is used it gives rise to cross coherence terms that can be accounted for by using the Kataoka, Maeda, and Hirose [47] and Teets [48] convolution scheme. When cross coherence terms are included, the CARS intensity (without dye laser convolution) is given as
F =
∫ dω′I (ω′) ∫ dω″ I (ω″ ) F (ω′, ω″ ). p
p
(13.8)
In addition to the effects of the laser widths on the CARS signal it is necessary to include the detector response of the spectrometer and other instrument factors. This is usually achieved by a further convolution of the calculated CARS signal with an appropriate Voigt [49] slit function. From Equation 13.5 it is clear that the resonant CARS intensity varies with temperature through its dependence on the number density N as well as population factor Δj and Raman line width Γj, which are functions of temperature. The total CARS intensity also depends on the non-resonant contribution, which in turn varies with species concentration and temperature. These features are the basis of temperature and species concentration measurements using the CARS technique. In principle any molecule present in the test section may be used to generate the CARS spectrum and measure the temperature. The CARS signal strength is quadratic in number density of the generating molecule. Thus in air-fed combustion, nitrogen, which is the major component, is used as the test molecule. The variation of the theoretical CARS spectrum of nitrogen with temperature is shown in Figure 13.3. Two main structures are observed in these theoretically generated spectral profiles. The largest peak near a wavelength of 21,125 cm−1 is the fundamental peak and corresponds to the v = 0 to v = 1 Q-branch vibration transition. The second peak is called the “hot band” and results from the v = 1 to v = 2 Q-branch vibration transitions. The fine structure on the spectral profile results from individual Q-branch rotational transitions. As the temperature of the molecule or the resolution
© 2011 by Taylor and Francis Group, LLC
of the spectrometer is reduced, this structure is less resolved.
13.3 Phase Matching in the CARS Configuration The difference between scanning and broadband arrangements for CARS spectrum generation is shown in Figure 13.4. The broadband CARS arrangement is usually used for turbulent combustion flow fields in order to make, almost instantaneous, time resolved measurements of temperature. With each shot of the lasers, a history of the fluctuation of the measurement can be stored. With this information, a PDF of the measurement can be created. One disadvantage of broadband CARS is that the CARS signal strength is less than that generated with a truly monochromatic CARS setup. With modern sensitive and enhanced detection equipment this should not be a big concern. The CARS is a coherent process in which the signal is generated from the coherent addition of complex amplitudes that have magnitude and phase. On the other hand Raman and laser induced fluorescence arise from the incoherent addition of intensities from single molecules. In order to optimize the signal generation in the CARS interaction it is important that the incident beams are aligned so that the incoming and outgoing waves are properly matched in phase. This phase matching between the Stokes beam and the pump beam determines the direction of emission of the CARS beam from the interaction point as illustrated in Figure 13.5 for a two beam, three wave mixing configuration. If kp, ks, and
293
CARS Temperature Measurements in Flames in Industrial Burners
Scanned
ωs
Test section
Stokes CARS
ωas Pump
Broadband
ωp Figure 13.6 Laser beam arrangement for the BOX-CARS configuration.
ωs
ωp
ωas Test point
Figure 13.4 Comparison between the scanned and broadband CARS techniques. Stokes Pump
Phase matching diagram kas = 2kp – ks
kp
k = ωn = 2π c λ
→
kp
θ
φ
Figure 13.5 Phase matching diagram for the CARS interaction.
kas, represent the wave vectors for the pump, Stokes and CARS beams, respectively, the general phase matching requirement is: 2 k p = k s + k as .
(13.9)
The wave vector k is equal in magnitude to ωn/c, where ω is the frequency of the radiation, n is the refractive index, and c is the speed of light. With the beam requirements and frequency relationships given above, it is obvious that the phase matching condition can be fulfilled when the laser beams are co-linear. The disadvantage of co-linear CARS is that the spatial resolution is very poor since the signal is generated along the full length of the beam overlap in the test section. A certain amount of phase mismatch can be tolerated in the CARS interaction. As a result of this allowable mismatch, optical arrangements have been developed with such names as BOX-CARS, folded BOX-CARS, and USED-CARS. The BOX-CARS optical arrangement is such that the pump beam is split into two beams and converges on the test section at a specific angle with the Stokes beam (see Figure 13.6). Folded BOX-CARS is similar except that the three beams are arranged three dimensionally. In order to use either of these two
© 2011 by Taylor and Francis Group, LLC
Pump
Stokes
Figure 13.7 Relative positions of the CARS, pump, and Stokes beams for the USED-CARS configuration.
ks
Combining optic
CARS
CARS system Orientation of pump and stokes probe laser beams with respect to the measured CARS beam
kas
Lens
configurations it is necessary that the specific phase matching requirements be satisfied [7,8]. The USED-CARS arrangement can be employed with CARS systems where an unstable resonator is used to generate the pump beam. In this arrangement the pump beam has a doughnut shaped spatial profile with no illumination along the centerline. It is then possible to align the Stokes beam such that it travels down the central hole in the pump beam (see Figure 13.7) and results in quasi-collinear phase matching [50,51]. Over the last two decades many CARS measurements have been reported in literature with other configurations such as dual Stokes, dual pump [52–54] laser beams. A more recent example of pure rotational CARS spectroscopy at high pressures can be found in Vestin et al. [55].
13.4 Experimental Setup The essential elements of a CARS setup, the details of the component requirements, and potential concerns about CARS measurements are discussed in this section. A typical optical arrangement for the CARS system is shown in Figure 13.8. The heart of the CARS system is the neodymium-yttrium aluminum garnet (Nd-YAG)
294
Industrial Combustion Testing
HBS 2
HBS 1
1064 nm
1064 nm
P
HG
KG3
P
M P
M 532 nm Dye laser
2× Nd: YAG laser
532 nm
λ/2
KG3 532 nm L
532 nm 1064 nm
532 nm
OC
532 nm
Fiber optic to spectrometer
M
DC 532 nm 532 nm BS
A
L
GL
M
P
DC
L
Beam expander
532 nm
D
473 nm
Pump power control
M
–λ/2
D
TF LM
A
607 nm
M
M
M
D
D
Test point
473 nm
L
L
S
532 nm
D
Figure 13.8 Optical arrangement for a USED-CARS measurement system.
laser that provides the excitation for the dye-laser oscillator and amplifier as well as the pump photons for the CARS interaction. The Nd-YAG laser is pumped by ultraviolet flash lamps oscillating at a rate of 10 flashes per second. The optical cavity is an unstable resonator and, as stated above, the Nd-YAG beam has a doughnut-shaped spatial profile. Since the cavity is Q-switch dumped, the laser pulses have a temporal profile from 8 to 10 ns in duration. The Nd-YAG laser produces a beam of light at a wavelength of 1.06 μm. The frequency of the Nd-YAG laser is doubled to take advantage of the quadratic dependence of the CARS generation efficiency on the pump frequency and of the optimum, dye-laser pumping efficiency at the doubled frequency. This produces a laser beam at 532 nm and residual laser energy at 1.06 μm. A second laser is required to produce the Stokes photons for the CARS interaction. This can be achieved with a separate Stokes laser with its own pulse control system or, as is shown in Figure 13.8, some of the laser energy from the pump laser can be used to optically pump a laser suited to the CARS system. The synchronization of the timing between the pump and Stokes laser systems must be accurate to within 1 ms to ensure that the two laser beams arrive at the same place at the same time. The advantage of the latter laser system is that only one timing circuit is needed to keep the pump
© 2011 by Taylor and Francis Group, LLC
Table 13.1 Stokes and Anti-Stokes Wavelengths for Typical Combustion Species Species N2 H2 O2 CO2
Stokes Wavelength (nm)
Anti-Stokes Wavelength (nm)
607 683 580 574
473 436 491 496
and Stokes laser beams synchronized in time. With a separate Stokes laser it is necessary to synchronize the two laser control systems and the CARS signal detection system. The setup discussed here (Figure 13.8) requires synchronizing only the pump laser and the CARS detection systems. The wavelength of the Stokes laser depends on the species to be probed with the CARS system. For nitrogen thermometry, a rhodamine dye laser is used that produces laser energy at about 607 nm and the CARS or anti-Stokes wavelength is 473 nm. Table 13.1 lists the Stokes and anti-Stokes wavelengths for species of interest in air breathing combustion assuming a 532 nm pump laser. The authors have had experience with Nd-YAG lasers that produce about 436 mJ and 1.0 J per pulse at the
CARS Temperature Measurements in Flames in Industrial Burners
doubled wavelength (532 nm). With the lower power Nd-YAG laser, it is possible to use a harmonic beam splitter (HBS) to split off the residual 1.06 μm laser energy after the doubler and double that to obtain a second laser beam at 532 nm. This second 532 nm beam can be used to pump the oscillator of a homemade Stokes laser system as shown in Figure 13.8. The advantage of using the higher power Nd-YAG laser is that there is no stray 1.06 μm radiation on the table. The Nd-YAG laser energy is sufficient for all of the components of the CARS laser system. Any extraneous 1.06 μm radiation can be split off and sent to a beam dump. The residual 1.06 μm laser beam (for the lower power Nd-YAG laser configuration), after being separated from the main laser beam, is directed through a half-wave plate (λ/2) to ensure that the polarization of the beam is horizontal on entering the second doubling crystal (HD Figure 13.8). The beam frequency is doubled and any residual 1.06 μm energy is absorbed in the KG-3 filter. The redoubled laser beam (now at 532 nm) is then used to pump the dye oscillator (Stokes laser). The lens arrangement in the dye-laser pump beam (Figure 13.8) reduces the pump beam to the appropriate size for the dye-laser beam. The dye cell (DC) is maintained at Brewster’s angle to ensure that the dye-laser beam is horizontally polarized. The dye-laser oscillator cavity extends from the total reflecting mirror (TM) to the output coupler (OC). The interference filter (TF) in the dyelaser oscillator cavity is a tuning filter used to shift the dye-laser spectral profile so that the spectral envelope covers the necessary Stokes wavelengths. For broadband nitrogen CARS the spectral width (FWHM) of the dye laser is about 9 nm and the tuning filter has a range of about 1 nm. This allows the placement of the peak of the dye spectral profile over the weakest region of the nitrogen CARS spectral profile (Figure 13.9). 606
607
609
610
Dye tuning
5000 Dye intensity (arb)
Dye wavelength (nm) 608
High dye Low Dye N2 cars
4000 3000 2000 1000 0 472
473
475 474 CARS wavelength (nm)
476
477
Figure 13.9 The Stokes and Antistokes (CARS) spectral profiles showing tuning of the dye laser wavelength.
© 2011 by Taylor and Francis Group, LLC
295
The originally doubled, Nd-YAG (2 × Nd-YAG) laser beam at 532.0 nm is directed through a KG-3 filter to remove any 1.06-um energy. Part of this laser beam is used to excite the dye in the dye-laser amplifier cell. Usually about 30% of the originally doubled ND-YAG beam is used for dye pumping and the residual 70% is directed to the CARS test section. The dye amplifier cell (DC) is at Brewster’s angle to maintain the horizontal polarization of the dye-laser beam. The laser beams are then combined to produce the CARS signals using one of the phase matching configurations described earlier. The 607 nm and the 532 nm beams are combined after they have been sized to fit together. The 607 nm beam is made to travel along with the 532 nm beam through the hole in the 532 nm beam doughnut profile by means of the beam-combining optic (D). This optic transmits the 607 nm beam and reflects the 532 nm beam (see Figure 13.7). These beams are then focused through a lens (L) before entering the test section (TS). In the region of the focal spot, the 532 nm beam collapses onto the 607 nm beam and through an interaction with the nitrogen molecule the CARS signal is generated. This home built system was specifically designed for CARS experiments. As discussed above, it is possible to use a more powerful, doubled Nd-YAG laser system to get around redoubling the residual 1.06 μm laser beams. With a more powerful doubled Nd-YAG laser the necessary laser energy can be split off and used to pump and amplify the dye-laser beams eliminating the necessity of external doubling crystals. In either case, a separate leg is used to control the power of the pump laser beam. This leg is shown in Figure 13.8 where the pump laser is directed through a λ/2 wave plate and then a Glanlaser polarizer (Pump Power Control region). By rotating the wave plate it is possible to control the power of the pump laser beam. Adjusting the pump beam power at the laser source (the Nd-YAG laser itself) is not recommended as it can seriously affect the timing between the pump and Stokes laser beams. The CARS beam, which emerges with the 607 and the 532 nm beams, is a coherent beam located in an annular region around the dye-laser beam (Figure 13.7). The location of the CARS beam is dependent on the physical characteristics and wavelengths of the probe beams [50,56,57]. The interaction length depends on the focal length of the focusing lens and the diameter of the beams entering the lens. To evaluate the gas temperature, the probe beams (pump and Stokes) must be removed from the CARS signal. This is done by using a beam stop at the center of the beam as it leaves the collimator to dump the 607 nm beam. The pump beam is removed at the interference filters (F) that reflect only at the CARS wavelength and transmit the 532 nm energy into beam dumps. Before the CARS signal enters the
296
spectrometer, part of the beam is split off and sent to a photomultiplier tube (PMT). This aids in tweaking the probe beams to ensure optimum overlap in the test section. The output of the PMT is sent to a boxcar averager that computes and displays the average of a number of CARS signal pulses and thus monitors the CARS signal strength while adjustments are made to the mirrors directing the probe beams. Optimization of the CARS signal is necessary because of the quadratic dependence of the CARS signal strength on both the pump beam intensity and the number density of the probe molecule. Minor fluctuations in either one result in a large variation in the CARS signal. The boxcar averager smoothes out variations caused by extraneous sources and allows the operator to follow trends as adjustments are made to the position of the two probe beams. Once the CARS measurement geometry has been optimized, the CARS signal is ready to be dispersed and analyzed in the holographic grating spectrometer (HGS). The CARS signal is carried to the HGS by means of a fiber optic. The fiber is sized to accommodate the optical properties of the CARS beam. The authors have used a fiber with an 80-um core and a 125-um cladding, made of fused silica and designed to have an f number of 4.1 [58]. After removing the probe beams from the beams leaving the test section, a lens (L) (Figure 13.8) focuses the CARS beam onto the input end of the fiber optic. The fiber then carries the optical signal to the HGS for analysis. This fiber-optic link enables the separation of the spectral analysis equipment from the detectionside optics. The study of the CARS signal can then be carried out in an area remote from the combustion environment. It is important to have the pump and Stokes beams arrive at the test point at the same time. The “on” time for the doubled Nd-YAG laser beam is about 9 ns. This means that the pump laser beam is about 2.7 m long and there is a gap of about 100 ms (less the 9 ns) between each pulse. There is a period of time required to build up the population inversion in the dye laser to begin producing dye-laser (Stokes laser) photons. Due to this build up delay, it is possible that the Stokes photons can be produced while the pump laser beam is off, resulting in the two laser beams arriving at the test point at different times. The delay is dependant on the dye laser and the pump energy. For this reason it is necessary to ensure that the energy used to pump and amplify the dye laser is constant and that the delay is measured using fast detectors. This is usually handled by using fast photo detectors and a fast oscilloscope. Measurement accuracy of the time difference between the two beams must be within 1.0 ns so that the influence of the cables used with the photo detectors on the delay and response times will be important. It is also important to note that the delay in
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
producing Stokes photons at the output of the dye laser is dependent on the amount of energy dumped into the cavity by the pump beams. For example, the delay can usually be cut in half by doubling the pump energy. As a result, the synchronization experiments must be done at the pump laser energy levels used in the CARS experiments. The pump laser energy must never be changed from this value otherwise the timing between the pump and dye-laser beams will be affected. With this delay measured, it is possible to establish the length of a pump beam delay leg so that the beams arrive at the test point at the same time. Typically this leg is about 140 cm but is quite dependent on the properties of the dye-laser system. This delay leg can also accommodate a wave plate-Glan Laser polarizer combination (discussed above) to control the pump laser power. If two separate laser systems are used for the Stokes and pump beams the need for a delay leg can be determined from synchronization tests. The requirements of the spectrometer used for the CARS experiment are not necessarily specific to CARS spectrometry and as such it is possible to purchase a spectrometer to suit. The typical nitrogen CARS spectral profile is about 100 cm−1 or about 1.5 nm wide. Since the CARS signal can be quite weak, it is necessary to have a sensitive detector array that can be gated. It is necessary, as well, to have the sourceside (laser beams) and detection-side (spectrometer) optics synchronized such that the laser beams arrive at the same time and the detector is open when the CARS signal is generated. This should all happen for every laser pulse.
13.5 Laser Considerations For the CARS system the pump photons in the CARS interaction are provided by a doubled Nd-YAG (2X-NdYAG) laser. This type of laser is used, as opposed to a ruby laser for example, to take advantage of the quadratic dependence of the CARS signal generation efficiency on pump frequency and the optimum dye laser pumping efficiency at the 2X-Nd-YAG frequency. The amount of optical energy supplied by 2X-Nd-YAG lasers can be significant (360 mJ/pulse) and as a result, care must be taken that during the CARS measurement the energy density at the test section is not excessive. In the event that pump energies are too high, stimulated Raman effects can occur that will cause an upset in the distribution of molecules in the various states of excitation, which equilibrium allows [59]. This disturbance in the distribution can usually be monitored by ensuring that there are no “hot band” transitions in the room
297
CARS Temperature Measurements in Flames in Industrial Burners
temperature CARS spectrum. At room temperature this band should be almost nonexistent. Snelling, Sawchuck, and Mueller [60] indicates that simply avoiding tight focal spots by rotating the focusing lens before the test section can alleviate this problem. Too tight focusing of the probe beams can also cause breakdown of the gases in the region of the test point. The light energy emitted during breakdown will mask the CARS signal and must be avoided. Depending on how the 2X-Nd-YAG laser is run, the pump spectral width can vary from 0.02 cm−1 to 1.0 cm−1 (fwhm). The effect of the pump spectral profile on CARS measurements will depend on the Raman line widths, which themselves vary with temperature and Raman frequency. In practice, the spectral profile of the pump laser has been shown [60–63] to be important in the analysis of CARS spectra. During CARS experiments it is found that the spectral width of the 2X-Nd-YAG laser can vary by as much as a factor of three from one laser to another for the same model of laser. It is therefore necessary to measure the pump spectral profile and include a pump convolution [46,47] in the generation of the theoretical CARS spectra. The CARS arrangement described here is of the broadband type and for each laser pulse the dye laser must provide the complete range of the Stokes photons to interact with the pump photons to create the complete CARS spectral profile of the probe molecule. The spectral profile of the dye laser output is shown in Figure 13.9. This broad Gaussian profile is a 10-shot average spectral profile of the dye-laser beam. The CARS spectrum drawn on the same graph indicates the spectral region of the dye profile that is used to generate the nitrogen CARS signal. The dye-laser spectral profile is centered at 607.5 nm and has the Gaussian fwhm of 6.7 nm. This figure shows the utility of a tuning filter in the dye-laser cavity to achieve the correct positioning of the dye spectral profile for the generation of the best CARS spectral profile. It has not been necessary to pay close attention to the synchronization of the two pump beams used to generate and amplify the dye-laser beam. The rhodamine molecules in the amplifier dye cell tend to remain in an excited state long enough for the unamplified dye-laser beam to arrive at the cell and stimulate the emission of more dye-laser photons. The only problem that has arisen is that the rhodamine dye is so willing to give off photons, that any two partially reflecting surfaces in the optical path of the amplifier cell will define an optical cavity. The end result is that the population inversion built up by the optical pumping in the amplifier cell can be depleted by stimulated emission before the arrival of the unamplified dye-laser beam at the amplifier cell. This problem can be identified by judicious blocking of laser beams to determine the extent of the false optical
© 2011 by Taylor and Francis Group, LLC
cavity. In order to alleviate this problem it is necessary to slightly detune the offending optics. The temporal profile of the pump beam can be modulated by as much as 100% during the eight ns lifetime of each laser pulse. The modulation in the Stokes beam tends to follow that of the pump beam; however, the extent of the modulation is not found to be as large. Along with the variations in the temporal profile of the probe beams, there will be modulations in the spectral profile as well. Figure 13.10 shows schematically the temporal and spectral makeup of a single shot of a typical pair of CARS probe beams. This figure shows that for a particular laser shot, the temporal profile at a specific frequency has its own individual modulation. The particular form of these modulations depends on the competing modes that exist in the laser for the population inversion in the laser cavity. The form of the modulation in the temporal profiles of the probe beams will vary from shot to shot; however, an average temporal profile resulting from the sum of many individual laser shots will still maintain what is termed a noisy profile. As these noisy laser beams are convolved through the CARS interaction the CARS spectral profile will develop a noise component itself. Researchers [60–66] discuss the sources and effects of the temporal and spectral noise in the probe beams. What is indicated is that the dye laser is responsible for the bulk of the probe beam generated noise in the CARS signal. It has been demonstrated [62] that, in broadband CARS experiments, the noise in individual non-resonant CARS spectra can be as much as 9% (expressing the standard deviation as a percentage). This type of noise can be reduced to a minimum of 5% by using what is termed a single-mode pump laser. This type of laser has special equipment installed in the Nd-YAG laser cavity that favors one mode in the cavity ωs ωp
ωas
(ω)
(t)
Figure 13.10 (See color insert following page 424.) Typical temporal and spectral profiles of the probe beams and the CARS measured signal.
298
for the emission of laser radiation. The effect of this “line-narrowing” equipment on the output of Nd-YAG lasers is to reduce or eliminate the modulations in the temporal and spectral profile of the pump laser. The lower limit of 5% of the noise in the non-resonant spectral profile is believed to result from the inherent 5% noise in the dye-laser spectral profile. It has been demonstrated that the contrary is true for the resonant CARS signal. That is to say that if a single-mode pump laser is used then the noise in the resonant CARS signal (the signal of interest in CARS thermometry) will be greater than the 5% minimum if a multimode pump laser is used. Snelling et al. [62] speculate that the reason for the reduction in the measured noise in the resonant CARS signal with increased pump spectral width comes from a beneficial averaging of the pump-Stokes convolution over a greater number of Stokes laser modes. The spectral averaging is of particular importance in flame spectra where Raman line widths are of the order of 0.03 to 0.04 cm−1. At lower temperatures and higher pressures this beneficial averaging effect will tend to be reduced because of the increased Raman line widths. Noise contribution from the many modes in the Stokes laser can be reduced by the use of a modeless dye laser invented by Ewart [67]. Fortunately nitrogen CARS spectra are less affected by this due to the large number of spectral lines that can be probed together [68]. The detector will also contribute to the noise in the measured CARS resonant signal. Depending on the type of device used in these experiments to digitize the dispersed CARS signal, the detector itself will have a noise component to contribute to the measured signal. The contribution of detector noise to the total noise in a CARS measurement is discussed by Snelling et al. [62,69]. The calculation of weighting coefficients from dark noise, shot noise, and noise from shot to variation and the effect of weighting on CARS temperature fitting, are also discussed [62]. Coupled with these noise characteristics is the fact that some linear array detectors have been demonstrated [31,69,70,71] to have a non-linear behavior with respect to the illuminating intensity. The end result is that the computed temperature from measured CARS spectra may be higher than the actual gas temperature at the test point. The reason for this is that, in the event of detector non-linearity the shapes of the fundamental and the hot bands will be distorted. This distortion will favor the height of the hot band resulting in higher measured temperatures. Depending on the amount of non-linearity in the detector, the errors in the measurement of temperature can be as great as 300 K. This nonlinearity has been demonstrated to result from tight focusing of the dispersed signal on the detector elements. It is recommended [60] to ensure that the image height fills the complete detector at the detection plane.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
The effects of this type of non-linearity can be included in the postprocessing of the collected spectra. The nonlinearity problems encountered with these early IPDA detectors do not appear to be an issue if more advanced CCD arrays are used for CARS spectral detection. Another potential source of error for USED-CARS measurement systems is that the two photons provided by the pump laser come from the same source. This results in an enhancement of the non-resonant signal arising from the effect of rapid intensity fluctuations of the two correlated pump beams [72]. With Gaussian beams, resonant intensities increase in a similar manner and hence there is no problem. For non Gaussian pump fields only the non-resonant component is enhanced and this will introduce errors in the fitting process and lead to an incorrect estimate of temperatures and concentrations. One way to overcome this problem is to insert a polarization analyzing filter in the CARS beam after the test section with the filter axis aligned at 90° to the polarization of the Stokes beam [73]. This will suppress the non-resonant signal and the error caused by inaccuracies in the non-resonant susceptibility is avoided. The disadvantage with this technique is that the overall signal can be reduced by as much as a factor of 16. Non-resonant signal enhancement in a CARS measurement can be reduced [74] by introducing a relative delay between the two pump beams. Lang and Wolfrum [74] conclude that effective de-correlation can be achieved with a delay equal to several times the coherence length of the pump laser. This is implemented in USED-CARS by inserting two quartz rods co-linear with the pump beam such that two opposite quadrants of the beam are covered by the rods. A photograph of these rods is shown in Figure 13.11
Figure 13.11 Picture of quartz rods for suppression of the enhancement of the non-resonant susceptibility.
CARS Temperature Measurements in Flames in Industrial Burners
Pump beam
Uncorrelated quadrants Quartz rods Figure 13.12 Schematic of the arrangement of quartz rods placed in the pump beam for the suppression of the enhancement of the nonresonant susceptibility.
and a schematic of the optical arrangement is shown in Figure 13.12. The optimum location in the optical path for these rods would be where the pump beam has been expanded. The length of the rods (typically about 3 cm) is calculated after Lang and Wolfrum [74] and should be long enough to have a sufficient delay to ensure that the pump photons from the quadrants appear to come from different laser sources. This technique can result in a slight loss in CARS signal strength but may be compensated for by increasing the pump laser strength. Another approach the authors have found to work well is to have the ratio of the nitrogen concentration to the non-resonant susceptibility as an additional fit parameter in the CARS signal analysis. This does not interfere with the CARS signal strength and gives accurate temperature measurements; however, the fitted non-resonant susceptibility is incorrect and the technique is not recommended for simultaneous measurements of concentration [13]. There are many factors to keep in mind when building a CARS measurement system. The first of these is equipment. The most important component in the CARS experiment is the pump laser. Due to signal generation efficiency considerations a 2X-Nd-YAG laser is recommended. Line narrowing devices are not necessary (and in fact not recommended) for the pump laser. Broad spectral profile pump lasers require complicated cross-coherence effects in the generation of the theoretical profiles; however, these are well understood [75] and can be implemented in the calculation of theoretical CARS spectra once the spectral width of the pump is taken into account in the signal analysis.
© 2011 by Taylor and Francis Group, LLC
299
The repetition rate for this type of laser is usually 10–20 Hz; however, higher repetition rate lasers radiating in the same wavelength region do exist. The problem with using these high repetition rate lasers is that the CARS data acquisition hardware is limited by the rate at which it can store the spectra. If the number of points per spectrum was reduced and/or the data capture rate limit is increased then lasers with higher repetition rate could be used. The next most important component in the CARS setup is the detection system. As indicated above, a sensitive detector with the least susceptibility to noise and non-linearities will result in measurements with the least machine dependant statistics. Commercially available dye lasers can be used for the Stokes laser; however, for reasons of expediency, simplicity, and economy, a home built dye laser is sufficient.
13.6 Data Acquisition To determine the temperature at a point in the test section, each CARS spectral profile is recorded and is then compared with a theoretically calculated spectral profile. Any molecule in the test section can be used to determine the temperature; however, as stated above, nitrogen is commonly used for various reasons. The spectral wavelengths and line-width parameters for nitrogen are well known. Furthermore, the strength of the CARS signal varies as the square of the number density of the probe molecule. Nitrogen is used for measuring temperature because it is in abundant supply in air-breathing combustion. It has been found that the CARS spectral profile is relatively insensitive to the concentration of the nitrogen molecule above 50% [8]. The spectral profile is, however, quite sensitive to changes in temperature. The calibration measurements that must be made before a temperature can be determined from the CARS spectral profile can be divided into two basic groups: variable and long-term. The long-term calibration measurements consist of the slit function and the dispersion of the spectrometer and the spectral profile characteristics of the pump laser. The slit function of the spectrometer is the response of the spectrometer to monochromatic irradiation. It can be thought of as the broadening of the spectral profile because of the finite width of the entrance slit. The long-term calibration values have been shown not to change over time. A recalibration is usually performed when changes are made to the optics involved. The variable calibration parameters involve the measurement of the dye-laser spectral profile and the determination of the diode wavelength
300
relationship for the detector. These parameters have been shown to change from day to day. Therefore, they form part of the series of measurements taken before each experiment. The laser beams are directed into the test section to the sample point. Depending on the type of measurement being performed, the laser beams are scanned throughout the test section to collect many single-point CARS spectra or left at one test point for a multishot average spectrum. As the spectra are collected, they are sent to the computer for analysis. The analysis involves the manipulation of large, vector arrays and these computations cannot be carried out in the time between the firing of the laser (0.1 s). Therefore, the CARS spectra, along with the short-term calibration spectra, are stored for future analysis.
13.7 Analysis of CARS Spectra Experimental CARS spectra are recorded as arrays of CARS intensities as a function of pixel numbers collected by the array detector in the spectrometer. After taking into account the dispersion of the detector array, the instrument function, and the spectral shape of the dye (Stokes) laser, the individual CARS shots are ready to be analyzed. As already stated, the N2, Q-branch, CARS spectrum is used for estimating temperatures in air-fed combustion. The simplest approach to doing this would be to fit the spectrum with theoretical spectra according to Equation 13.6 by a least-mean-square procedure. However the generation of theoretical spectra for every iteration cycle in the fitting process, would be time-consuming. A more efficient scheme is to use a pre-calculated library of theoretical, N2 CARS spectra generated at 50°K intervals and interpolate the spectra for intermediate temperatures. A suitable software package can be used for the calculation of the theoretical CARS spectra, post-processing of the stored experimental CARS spectra, and to perform the non-linear fitting process. In generating the theoretical CARS spectra for nitrogen, collisional narrowing effects [76,77], which arise from the broadening and overlap of the Raman lines should be included. These narrowing effects can be significant even at atmospheric pressures. Cross-coherence effects in the generation of theoretical CARS spectra are also necessary. The fitting is t ypically done by applying the Marquardt non-linear, leastmean-square scheme discussed by Marquardt [78]. The fitting scheme should also include the weighting coefficients calculated as discussed above and described by Snelling et al. [62].
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
In a typical CARS experiment, a background spectrum is collected first to be able to compensate for stray illumination in the spectrometer. Then a non-resonant CARS spectrum, generated in a cell filled with CO2 or propane, is collected. This is followed by the acquisition of an averaged room temperature N2 CARS spectrum and later the required number of flame CARS spectra are acquired. In the analysis scheme, the dye convolution of the theoretical CARS spectra is eliminated by dividing the experimental N2 CARS spectrum with the stored non-resonant CARS spectrum. The resulting spectrum will be the resonant spectrum. The parameters for the slit function are derived by fitting a Voigt type function to the averaged room temperature CARS spectrum. This is easily done since the N2 concentration is known. This slit function can then be used to convolve the theoretical CARS spectral profiles to make them ready for interpolation to calculate the experimental temperatures. Temperature, major species concentration, and frequency shift between the theoretical and experimental CARS spectra are typical fit parameters. Suppressing the non-resonant contribution by suitable polarization arrangements as discussed above can simplify the CARS fitting process, but this can greatly reduce the resonant CARS signals [79] and not very useful for broadband CARS. The CARS analysis demands intensive calculations and during the early days of CARS development so called fast fitting methods [24] were used to overcome computer storage and speed limitations. These techniques involved using the variation of parameters such as the ratio of area of the fundamental peak to the fundamental peak height, with temperature and creating a look up table. The great advancement in the capabilities of computers in the last decade has eliminated the need to use approximating fast fitting methods. An intelligent first guess of the temperature with a library generated with suitable temperature intervals can speed up the fitting process. Spatial averaging is a potential problem in CARS when varying temperatures and composition are present at spatial scales smaller than the focal volume, a situation that can occur in turbulent flames. The role of spatial averaging in CARS and methods to quantify and correct for these errors using simulated coherently added average spectra are discussed elsewhere [80–83].
13.8 CARS Measurements made in Large, Practical Flames The authors have used the USED-CARS technique to make gas temperature measurements in industrial-scale
301
CARS Temperature Measurements in Flames in Industrial Burners
research flames. The optical arrangement for these measurements is similar to that shown in Figure 13.8. Figures 13.13 and 13.14 are photographs of the test setup at the research furnace. All of the measurements were made in a research furnace with an interior diameter of about 1.0 m with a burner fired along the central axis. Figure 13.15 is a schematic of the research furnace. The burner design was different for each of the tests and the CARS measurements were used to investigate the properties of the gas temperature in the flames and to gain an understanding of the properties of the flames produced by these burners.
One segment of furnace calorimetric section - 28 in total
Stack sample port To flow meter Coolant out
Coolant in Burner
Observation port
Probe ports Calorimetric section
Adiabatic section
Figure 13.15 Schematic of research furnace.
13.9 Measurements in Coal and Oil Flames
Furnace
Figure 13.13 Photograph of the optical arrangement for CARS measurements in a pilot-scale furnace.
Pump and stokes beams
USEDCARS system
Measurement port
Figure 13.14 USED-CARS system making temperature measurements in a pilotscale furnace.
© 2011 by Taylor and Francis Group, LLC
A series of tests were conducted in the furnace to compare CARS temperature measurements with those acquired with a suction pyrometer [84]. A suction pyrometer is an intrusive probe to measure gas temperature in a flame. The principle of the device is to insert the water cooled probe to the measurement point and draw furnace gases over a thermocouple located at the tip of the probe in an enclosure shielded from flame radiation. The gases are drawn at sufficient velocity to enhance the convective heat transfer to the thermocouple bead. Typically, the flow rate of furnace gases through the probe tip is increased until the thermocouple temperature no longer increases. At this point the thermocouple is assumed to measure the true gas temperature. The disturbance by such a measurement technique can be considerable in the near burner region where chemical reactions are occurring and there is heat release. Combustion tests were conducted with oil and coal as a fuel at firing rates between 300 and 400 kW. The fitting technique described above was used to determine the gas temperature from the collected CARS spectral profiles. A typical comparison between the measured and the theoretical, fitted CARS spectral profiles is shown in Figure 13.16. For these tests about 1,000 individual temperature measurements were made at several points in the near burner region of the flame over a period of about 1.67 min. Using these single shot temperature measurements, the CARS measurement technique can show the frequency distribution of the gas temperature as measured at a point in the
302
Industrial Combustion Testing
Area normalized CARS spectral plot comparison of measured single shot with theory
0.08 0.07
Coal flame Record no. 993 Exp’t Theory
0.06
Intensity
0.05 0.04 0.03 0.02 0.01 0.00 21060
21070
21080
21090
21100 21110 21120 Frequency (cm–1)
21130
21140
21150
Figure 13.16 Sample comparison of measured and theoretical CARS spectral profile for a coal flame.
Table 13.2
Temperature PDF in an oil flame
Frequency of occurrence
120 100 80 60
Average temperature
Temperature of average CARS spectrum
Temperature standard deviation = 137 K
40 20 0 800
1000
1200 1400 1600 Temperature (K)
1800
2000
Figure 13.17 Typical frequency distribution of CARS temperature measurements in an oil flame.
flame. A typical distribution for these tests is shown in Figure 13.17. As will be seen later, this frequency distribution can be used to derive information about the combustion and mixing in the flame. The arithmetic mean of the measured temperatures is shown on Figure 13.17 as well as the temperature of the multishot CARS spectrum. The arithmetic mean temperature is the mean of 1000 temperatures derived from individual CARS spectral profiles each captured during the 9 ns “on” time of the probe lasers. The temperature of the multishot average CARS spectrum is derived from a fit of a spectrum that is a pixel by pixel
© 2011 by Taylor and Francis Group, LLC
Comparison of CARS and Suction Pyrometer Temperatures Fuel
TCARS (K)
TSD (K)
TSP (K)
Coal Oil
1375 1525
208.5 291.1
1394 1565
sum of the many shots acquired during the measurements at the measurement point (1.67 min in duration). Note that these two “averages are not equal. Due to the turbulent nature of the flow field, the gas temperatures will fluctuate as is shown in the frequency distribution graph. As the gas temperature increases the relationship between strength of the fundamental band (ν = 1 to 0 transition) and the hot band (ν = 2 to 1 transition) portions of the signal change and do not sum linearly with temperature. As a result a multishot average CARS signal or a time averaged CARS signal must not be used to measure accurate gas temperatures in turbulent combustion. Table 13.2 shows the comparison between the temperatures measured with the CARS and suction pyrometer techniques as measured in the flame about 1.0 m from the burner. This point was chosen to have a minimal effect on the flow field by the suction pyrometer. The temperatures for the two measurement techniques compare well in the two flames. The CARS technique also provides statistical information such as the standard deviation of the gas temperature at the measurement point. It is interesting to note that,
303
CARS Temperature Measurements in Flames in Industrial Burners
Temperature variation at test point 1550 Oil flame Coal flame
1500 Temperature (K)
Suct’n pyro. 1450 1400 1350 Suct’n pyro.
1300 1250
0
10
20 30 Elapsed time (min.)
40
50
Figure 13.18 Typical mean temperature variation with time at the test point. The duration of the suction pyrometer measurement is also shown.
© 2011 by Taylor and Francis Group, LLC
CARS spectral profile in a coal flame
0.10
Measured Theory
0.09 Intensity (area normalized)
whereas the CARS technique gives an instantaneous measurement of temperature, the suction pyrometer can take as much as 15 minutes to arrive at the correct suction value for the temperature measurement. This is illustrated in Figure 13.18 [84] where the CARS temperatures, measured at different time intervals along with the duration of the suction pyrometer measurement, are shown. The CARS temperatures were measured immediately after removing the suction pyrometer from the test point. Making CARS measurements in coal and oil flames can be challenging and there are some problems that can arise in these types of flames. The first is the influence of particles or droplets at the measurement point. The intense laser beams will cause breakdown when particles or droplets are at the test point and the spectral profile will not resemble the expected shape. These “bad” shots can be removed through several techniques using a combination of the area under the spectrum and abnormal “fit” temperatures. Another problem that can arise when making measurements in the near burner region of a high carbon fuel is the absorption of the CARS spectrum by C2 along the path of the exiting CARS beam. This is identified and characterized by Bengtsson et al. [85]. A typical example of this is shown in Figure 13.19. The C2 absorption occurs between about 2315 cm−1 and 2325 cm−1 (Raman shifted). The tendency is to try to fit the theoretical curve to the experimental by choosing regions above and or below the region of absorption; however, the extent of absorption is never quite known and the “fitted” temperature is quite sensitive
0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00 2275
2285
2295
2305 2315 2325 Raman shift (cm–1)
2335
2345
2355
Figure 13.19 Comparison of measured and theoretical CARS spectral profile showing C2 absorption in the measured spectrum.
to the relationship between the fundamental and the hot bands. The spectra that have been affected by C2 absorption should not be used.
13.10 Measurements in Natural Gas Flames The CARS measurements can be used for the evaluation of CFD codes [86–90], but they can also be used to gain an insight into the combustion and flow in the near burner region. Hughes et al. [91] uses a similar CARS optical arrangement to make temperature measurements in two natural gas flames. One of the burners produces a high temperature jet-like flame. The other has tertiary air supplied off the central axis on either side of the flame jet. Details of the burners can be found with Hughes et al. [91]. For these burners, the CARS temperature was measured at several axial and radial locations to establish the flame properties. In the jet-like flame the temperature frequency distribution was shown to have three basic shapes as shown in Figure 13.20; symmetric and skewed to either lower or higher temperatures. The distribution of these shapes was used to define a flame boundary or the region of highest heat release in the flame. The locations of the different temperature distribution shapes are shown in Figure 13.21. The inferred flame boundary is also indicated on this graph. The distribution of the mean temperature and the associated standard deviation at the measurement points in the burner region are shown in Figure 13.22. These data along with the temperature distributions
Industrial Combustion Testing
0.10 0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00
Histogram Tmean = 1484 K
Flame boundary
Skewed lower
0.10 m
Skewed higher
Symmetric on axis
Skewed higher –0.10 m Skewed lower
–0.25 m
1000
1500 2000 Temperature (K)
0.11 0.10 0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00
2500
Symmetric
3000
Histogram Tmean = 1659 K
Axial
0.323 m
0.680 m
Figure 13.21 Schematic of flame boundary inferred from skew shape of temperature frequency distribution. CARS temperatures; mean and std. deviation Mean temperature
1500
2000
2500
3000
Temperature (K) 0.10 0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00
Histogram Tmean = 1836 K
x = 0.680 m
1600 1400 1200 1000
–0.0
0.1
0.2
Temperature (K)
1500 2000 Temperature (K)
2500
can be used to evaluate the mean temperature field and the temperature PDFs used in CFD modeling codes for these types of flames. The burner with tertiary air supplied [91] resulted in a much different frequency distribution. A bimodal frequency distribution was found to exist at certain locations in the flame. A typical example of such a distribution is shown in Figure 13.23. As indicated in the reference, a frequency distribution such as this
© 2011 by Taylor and Francis Group, LLC
100 –0.0
x = 0.443 m
1400 1200 –0.0
0.1
0.2
150 100 –0.0
0.1
0.2
0.3
Radial dist. (m) 300
x = 0.323 m
1400 1200 1000
0.3
x = 0.443 m
200
50
0.3
2000 1600
0.2
250
Radial dist. (m)
1800
0.1
300
1600
3000
Figure 13.20 Skewed temperature frequency distribution in a jet-type flame.
150
Radial dist. (m)
1800
1000
1000
x = 0.680 m
200
Radial dist. (m) 2000
500
Temperature standard deviation
250
50
0.3
Temp. std. dev. (K)
1000
–0.0 0.1 0.2 Radial dist. (m)
0.3
Temp. std. dev. (K)
500
1800
Temp. std. dev. (K)
300
2000 Temperature (K)
Normalized frequency
Symmetric
0.25 m
500
Normalized frequency
Radial
Temperature (K)
Normalized frequency
304
250 200
x = 0.323 m
150 100 50
–0.0
0.1 0.2 0.3 Radial dist. (m)
Figure 13.22 Variation of the mean and standard deviation of gas temperature in the near burner region.
305
CARS Temperature Measurements in Flames in Industrial Burners
Temperature histogram; Bi-modal distribution, X = 0.678 m, Y = 0.2 m Histogram Tmean = 1233 K
0.14 0.12
Std. dev. = 216.8 K
0.10 0.08 0.06 0.04 0.02
Suction pyrometer temp. Cars temperature
1700 1600 Temperature (K)
Normalized frequency
0.16
Suction pyrometer and CARS temperatures
1500 1400
at X = 0.8 m
1300 1200 1100
0.00
1000
400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 Temperature (K)
900
Figure 13.23 Example of a bimodal temperature distribution.
© 2011 by Taylor and Francis Group, LLC
–0.2
–0.1 –0.0 0.1 0.2 Radial distance (m)
0.3
1700 1600 Temperature (K)
indicates that during the time of the measurement at this point (approximately 1.67 minute duration) two gas streams were passing through the measurement point (1.0 mm by 3.0 cm long). The possible explanation for this could be a lazy, high temperature jet swinging in and out of the measurement point. When the high temperature jet is absent from the measurement point the gas stream has a specific mean and frequency distribution due to the turbulent nature of the flow. When the high temperature jet passes through the measurement point, a different mean and frequency distribution is captured. These distributions have their own separate frequency of occurrence and thus they are superposed. It is not possible to separate the distributions one from the other; however, it is apparent the there are two mean temperatures at this measurement point. With no suggestion of accuracy it is possible to assign two temperatures to the locations where bimodal frequency distributions are measured. From Hughes et al. [91], Figure 13.24 is drawn showing the CARS mean temperatures extracted from the frequency distributions. Locations with two temperatures may not have accurate temperatures but indicate the variation in the mean temperature due to the gross movement of jets with different temperature characteristics. Also drawn in this graph is the temperature as measured by the suction pyrometer. It is interesting to note that for the most part the suction pyrometer measurements lie between the bimodal temperatures. As well, because of the time to arrive at the suction pyrometer measurement (about 15 minutes), the temperature as measured with this intrusive probe may not reflect the correct temperature. In another series of tests [92], the CARS measurement technique was used to study the performance of a self-regenerative burner compared to that of a conventional, low NOx, natural gas burner. A similar furnace
–0.3
1500 1400
at X = 0.68 m
1300 1200 1100 1000 900
–0.3
–0.2
–0.1 –0.0 0.1 0.2 Radial distance (m)
0.3
Figure 13.24 Comparison between the CARS and suction pyrometer temperatures. Both bimodal CARS temperatures are shown.
and optical arrangement was used. The CARS measurements were very useful in understanding why the regenerative burner produced very little NOx. Figure 13.25 shows a comparison between the two burners with respect to the mean and standard deviation of the gas temperatures in the near burner region. The low NOx burner is shown to have a relatively high mean temperature; however, the standard deviation of these temperatures indicates that, in parts of the flame the gas temperature is well above the 1800 K limit for thermal NOx production. The regenerative burner, on the other hand, has low mean and standard deviation temperatures. The self-regenerative burner operates in a cyclic fashion where the furnace gases leave the furnace through part of the burner while air is entering the furnace through another part of the burner. The function of these parts of the burner change every 20 seconds so that the heat captured in the burner while the gases are leaving is transferred to the air entering through the burner. The CARS temperature measurements were used to characterize
306
Industrial Combustion Testing
Regen burner; CARS temperature measurements X = 20 cm
80 40
–20
–10 –0 10 Radial distance (cm)
20
Low no burner; CARS temperature measurements * X = 54 cm
142.5
950
95.0
475 0
47.5
Avg. temp. Std. dev. –20
–10
–0
10
20
160
950
120 80
475
Temp. Temp. std dev –20
–10 –0 10 Radial distance (cm)
40 20
0
Regen burner; CARS temperature measurements X = 54 cm
190.0
1425
200
1425
0
0
0.0
Temperature (K)
Avg. temp. Std. dev.
475
240
1900
190.0
1425
142.5
950
95.0
475 0
Radial distance (cm)
47.5
Temp. Temp. std dev –20
–10
–0
10
20
Temperature std. dev. (K)
120
Temperature (K)
Temperature std. dev. (K)
160
950
1900
Temperature (K)
200
1425
0
1900
240
Temperature std. dev. (K)
Temperature (K)
1900
Temperature std. dev. (K)
Low no burner; CARS temperature measurements * X = 21 cm
0.0
Radial distance (cm)
Figure 13.25 Comparison of the mean and standard deviation of the CARS temperature measurements in two burners.
due to the regenerative process. By windowing the CARS temperature measurements (Figure 13.26) a more realistic estimate of the turbulent property of the gas temperature is found to be about 75.4 K.
Comparison of CARS and regenerator temperatures X = 20 cm; Y = 20 cm
1500
940 930
1400 1300
920
1200 910
1100 1000
900
900 800
Regen temperature (C)
1600
CARS temperature (K)
950
CARS Regenerator Window
890 0
10
20 Time (s)
30
40
Figure 13.26 CARS gas temperature measurement compared to the regenerator air temperature.
this temperature performance. Figure 13.26 shows the variation of the furnace gas temperature along with the thermocouple temperature in the adjacent air entry port (regenerator). For this gas temperature the standard deviation is 131.8 K but that includes the cyclic variation
© 2011 by Taylor and Francis Group, LLC
13.11 Summary This chapter describes the use of the CARS technique for gas temperature measurements in flames. The description includes the optical arrangement, a brief summary of the theory of CARS signal generation, data acquisition, and analysis. The establishment of a CARS measurement system for practical flames can be challenging and it is hoped that the practical information included in this chapter will be of assistance to a novice attempting to get measurable and accurate CARS temperatures in large flames. Since the establishment of CARS as a practical measurement technique for temperature measurements in flames, literature shows that a lot has been learned about this non-linear interaction resulting
CARS Temperature Measurements in Flames in Industrial Burners
in a coherent optical signal, the sources of error, and data analysis issues. There have also been significant improvements in source and detection equipment. With the theoretical and experimental advances in the field and the availability of more powerful lasers, efficient CCD detectors and much faster computers it can be said that the CARS technique can make very accurate, spatially and temporally resolved gas temperature measurements in flames. The results presented here demonstrate the application of USED-CARS as a powerful tool to study pilot-scale flames and show that the technique can be used to do more than simply measure gas temperature. Details regarding the nature of the aerodynamics of the flow field, the geometry of the flame and an understanding of the flame chemistry can also be extracted from the CARS measurements.
Acknowledgments The authors gratefully acknowledge Natural Resources Canada, Program of Energy Research and Development (PERD) for their financial support in the development and application of the CARS measurement system to large-scale flames. As well we would like to thank Dr. D. R. Snelling, Principal Research Officer at the National Research Council of Canada for his advice and consultations over the years.
References
1. Tine, G. Gas Sampling and Chemical Analysis in Combustion Processes. Oxford, UK: Pergamon Press, 1961. 2. Bilger, R. W. “Experimental Diagnostics in Gas Phase Combustion Systems.” Zinn, B. T. (ed.) Progress in Astronautics and Aeronautics 53 (1977): 49. 3. Kurlbaum, F. “A Simple Method for Determining the Temperature of Luminous Flames” (Translated from German). Physikalische Zeitschrift 3 (1902): 18788. 4. Eckbreth, A. C. Laser Diagnostics for Combustion Temperature and Species Combustion Science and Technology. 2nd ed. Boca Raton, FL: CRC Press, 1996. 5. Kohse-Höinghaus, K., and Jeffries, J. B., eds. Applied Combustion Diagnostics. New York: Taylor & Francis, 1989. 6. Eckbreth, A. C. “CARS Thermometry in Practical Combustors.” Combust and Flame 39 (1980): 133–47. 7. Eckbreth, A. C., and Schreiber, P. W. “Coherent AntiStokes Raman Spectroscopy (CARS): Application to Combustion and Gas-Phase Diagnostics.” Chemical Applications of Nonlinear Raman Spectroscopy in Laser Applications, Vol. 1. New York: Academic Press, 1981.
© 2011 by Taylor and Francis Group, LLC
307
8. Hall, R. J., and Eckbreth, A. C. “Coherent Anti-Stokes Raman Spectroscopy (CARS): Application to Combustion Diagnostics.” Laser Applications Vol. 5. New York: Academic Press, 1984. 9. Smekal, A. “Zuschriften und vorlaufige mitteilungen.“ Naturwissenschaften 11 (1923): 873. 10. Raman, C. V., and Krishnan, K. S. “A New Type of Secondary Radiation.” Nature 121 (1928): 501. 11. Maker, P. D., and Terhune, R. W. “Study of Optical Effects Due to an Induced Polarization Third Order in Electric Field Strength.” Physical Review 137 (February 1965): A801–A818. 12. Begley, R. Coherent Anti-Stokes Raman Spectroscopy Applied Physics Letters 25, no. 7 (1974): 387–90. 13. Regnier, P. R., and Taran, J. P. E. “On the Possibility of Measuring Gas Concentrations by Stimulated AntiStokes Scattering.” Applied Physics Letters 23 (September 1973): 240–42. 14. Druet, S. A. J., and Taran, J. P. E. “CARS Spectroscopy.” Progress in Quantum Electronics 2 (1981): 1–72. 15. Anderson, H. C., and Hudson, B. S. “Coherent Anti-Stokes Raman Scattering.” Molecular Spectroscopy, edited by D. A. Long, 142–201. New York: Chemical Society, 1978. 16. Klick, D., Marko, K. A., and Rimai, L. “Broadband SinglePulse CARS Spectra in a Fired Internal Combustion Engine.” Applied Optics 20, no. 7 (1981): 1178. 17. Alessandretti, G. C. and Violino, P. “Thermometry by CARS in an Automobile Engine.” Journal of Physics D 16, no. 9 (1983): 1583–94. 18. Stenhouse, I. A., Williams, D. R., Cole, J. B., and Swords, M. D. “CARS Measurements in an Internal Combustion Engine.” Applied Optics 18, no 22 (1979): 3819. 19. Kataoka, H., Maeda, S., Hirose, C., and Kajiyama, K. “A Study for N2 Coherent Anti-Stokes Raman Spectroscopy Thermometry at High Pressure.” Applied Spectroscopy 37, no. 6 (1983): 508–12. 20. Harris, L. E., and McIlwain, M. E. “Coherent AntiStokes Raman (CARS) Temperature Measurements in a Propellant Flame.” Combustion and Flame 48 (1982): 97–100. 21. Greenhalgh, D. A. “Laser Diagnostics of Combustion Devices and Chemical Reactors Using Coherent Anti-Stokes Raman Spectroscopy.” 183–94. From the Proceedings of the 2nd International Conference on Lasers in Manufacturing, March 26–28, 1985. 22. England, W. A., Glass, D. H. W., Brennan, G., and Greenhalgh, D. A. “Study of a Tube Wall Methanation Reactor Using CARS Spectroscopy.” AERE Harwell report no AERE-R-11449, AERE Harwell, UK, October 1984. 23. England, W. A., and Greenhalgh, D. A. “The Application of Coherent Anti-Stokes Raman Spectroscopy (CARS) to Chemical Reactors.” AERE Harwell report no AERE-R10383, AERE Harwell, UK, November 1981. 24. Eckbreth, A. C., Dobbs, G. M., Stufflebeam, J. H., and Tellex, P. A. “CARS Temperature and Species Measurements in Augmented Jet Engine Exhausts.” Applied Optics 23 (1984): 1328. 25. Eckbreth, A. C. “CARS Thermometry in Practical Combustors.” Combustion and Flame 39 (1980): 133–47.
308
26. Switzer, G. L., Goss, L. P., Trump, D. D., Reeves, C. M., Stutrud, J. S., Bradley, R. P., and Roquemore, W. M. “CARS Measurements in the Near-Wake Region of an Axisymetric Bluff-Body Combustor.” AIAA Journal 24, no. 7 (July 1986): 1122. 27. Greenhalgh, D. A., and Porter, F. M. “The Application of Coherent Anti-Stokes Raman Scattering to Turbulent Combustion Thermometry.” Combustion and Flame 49 (1983): 171–81. 28. Switzer, G. L., Goss, L. P., Roquemore, W. M., Bradley, R. P., Schreiber, P. W., and Roh, W. B. “Application of CARS to Simulated Practical Combustion Systems.” Journal of Energy 4, no. 5 (September/October 1980): 209. 29. Anderson, T. J., Dobbs, G. M., and Eckbreth, A. C. “Mobile CARS Instrument for Combustion and Plasma Diagnostics.” Applied Optics 2, no. 22 (November 1986). 30. Eckbreth, A. C., and Hall, R. J. “CARS Thermometry in a Sooting Flame.” Combustion and Flame 36 (1979): 87–98. 31. Goss, L. P., Trump, D. D., MacDonald, B. G., and Switzer, G. L. “10 Hz Coherent Anti-Stokes Raman Spectroscopy Apparatus for Turbulent Combustion Studies.” American Institute of Physics, Review of Scientific Instruments 54 (1983): 563. 32. Furuno, S., Akihama, K., Hanabusa, M., Iguchi, S., and Inoue, T. “Nitrogen CARS Thermometry for a Study of Temperature Profiles Through Flame Fronts.” Combustion and Flame 54 (1983): 149–54. 33. Goss, L. P., Switzer, G. L., and Schreiber, P. W. “Flame Studies with the Coherent Anti-Stokes Raman Spectro scopy Technique.” AIAA paper no AIAA-80-1543. Presented to the AIAA 15th Thermophysics Conference in Snowmass, Colorado; July 14–16, 1980. 34. Fuji, S., Gomi, M., and Jin, Y. “Instantaneous CARS Thermometry in Turbulent Flames.” Combustion and Flame 48 (1982): 233. 35. Alden, M., Edner, H., and Svanberg, S. “Coherent Anti-Stokes Raman Spectroscopy (CARS) Applied in Combustion Probing.” Physica Scripta 27, no. 1 (1983). 36. Alden, M., and Wallin, S. “CARS Experiments in a Full-Scale (10X10) Industrial Furnace.” Applied Optics 24, no. 21 (1985): 3434. 37. Beiting, E. J. “CARS Temperature Measurements in a Coal Fired MHD Environment.” IFRF Document No. K 20/a/206. Presented at the AFRC Fall Symposium, October 4–6, Akron, OH, 1983. 38. Beiting, E. J. “Multiplex CARS Temperature Measurements in a Coal-Fired MHD Environment.” Applied Optics 2, no. 10 (May 1986). 39. Hartford, A. “Laser-Based Analysis of Coal Gasification Streams.” Laser Focus/Electro-Optics (April 1984): 63. 40. Ferrario, A., Garbi, M., and Malvicini, C. “Real Time CARS Spectroscopy in a Semi-Industrial Furnace.” Presented to the CLEO 83 Conference on Lasers and Electro-Optics, Baltimore, MD, May 17–20, 1983. 41. Porter, F. M., and Greenhalgh, D. A. “Application of the Laser Optical Technique CARS to Heat Transfer and Combustion.” AERE Harwell report no AERE-R-11824, AERE Harwell, UK, May 1985.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
42. Singh, J. P., Yueh, F. Y., Cook, R. L., Lee, J. J., and Linberry, J. T. “Comparison of CARS Temperature Measurements with Flow Field Model Calculations at the CFFF Diffuser.” Applied Spectroscopy 46 (1992): 1649. 43. Hancock, R. D., Hedman, P. O., and Kramer, S. K. “Coherent Anti-Stokes Raman Spectroscopy (CARS) Measurements in Coal Seeded Flames.” Combustion and Flame 87 (1991): 77. 44. Tolles, W. M., Nibler, J. W., McDonald, J. R., and Harvey, A. B. “A Review of the Theory and Application of Coherent Anti-Stokes Raman Spectroscopy (CARS).” Applied Spectroscopy 31, no. 4 (1977): 253–71. 45. Dewitt, R. N., Harvey, A. B., and Tolles, W. M. “Theoretical Development of Third-Order Susceptibility as Related to Coherent Anti-Stokes Raman Spectroscopy (CARS).” Naval Research Laboratory Report NRL-MR3260, 1976. 46. Yuratich, M. A. “Effects of Laser Linewidth on CARS.” MoZec Physics 38 (1979): 625. 47. Kataoka, H., Maeda, S., and Hirose, C. “Effects of Laser Line Width on the Coherent Anti-Stokes CARS Spectroscopy Profiles.” Applied Spectroscopy 36 (1982): 565. 48. Teets, R. E. “Accurate Convolutions of CARS Spectra.” Optics Letters 9 (1984): 226. 49. Whiting, E. E. “An Empirical Approximation to the Voigt Slit Function.” Journal of Quantitative Spectroscopy and Radiative Transfer 8 (1968): 1379. 50. Eckbreth, A. C., Dobbs, G. M., Stufflebeam, J. H., and Tellex, P. A. “CARS Temperature and Species Measurements in Augmented Jet Engine Exhausts.” Applied Optics 23 (1984): 1328. 51. Davis, L. C., Marko, K. A., and Rimai, L. “Angular Distribution of Coherent Raman Emission in Degenerate Four-Wave Mixing with Pumping by a Single Diffraction Coupled Laser Beam: Configurations for High Spatial Resolution.” Applied Optics 20, no. 9 (1981): 1685–90. 52. Eckbreth, A. C., and Anderson, T. J. “Dual Broadband CARS for Simultaneous, Multiple Species Measurements.” Applied Optics 24 (1985): 2731. 53. Boyack, K. W., and Hedman, P. O. “Dual-Stokes CARS System for Simultaneous Measurement of Temperature and Multiple Species in Turbulent Flames.” 23rd Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 1990. 54. Hancock, R.D., Schauer, R. F., Lucht, R. P., and Farrow, R. L. “Dual-Pump Coherent Anti-Stokes Raman Scattering Measurements of Nitrogen and Oxygen in a Laminar Jet Diffusion Flame.” Applied Optics 36 (1997): 3217. 55. Vestin, F., Sedarsky, D., Collin, R., Alden, M., Linne, M., and Bengtsson, P. “Rotational Coherent Anti-Stokes Raman Spectroscopy (CARS) Applied to Thermometry in High Pressure Flames.” Combustion and Flame 154 (2008): 143. 56. Marko, K. A., and Rimai, L. “Space and Time-Resolved Coherent Anti-Stokes Raman Spectroscopy for Combus tion Diagnostics.” Optic Society of America 4, no. 7 (1979): 211–13.
CARS Temperature Measurements in Flames in Industrial Burners
57. Bjorklund, G. C. “Effects of Focussing on Third-Order Non-Linear Processes in Isotropic Media.” IEEE Journal of Quantum Electronics 11, no. 6 (1975): 287–96. 58. Snelling, D. R., Sawchuk, R. A., and Parameswaran, T. “An Improved CARS Spectrometer for Single Shot Measurements in Turbulent Combustion.” Review of Scientific Instruments 63 (1992): 5556. 59. Gierulski, A., Noda, M., Yamamato, T., Markowski, G., and Slenczka, A. “Pump-Induced Population Changes in Broadband Coherent Anti-Stokes Raman Scattering.” Optics Letters 12, no. 8 (1987): 608. 60. Snelling, D. R., Sawchuck, R. A., and Mueller, R. E. “Single Pulse CARS Noise: A Comparison Between Single-Mode and Multimode Pump Lasers.” Applied Optics 24, no. 17 (1985): 2771. 61. Barton, S., and Garneau, J. M. “Effect of Pump-Laser Linewidth on Noise in Single-Pulse Coherent Anti-Stokes Raman Spectroscopy Temperature Measurements.” Optics Letters 12, no. 7 (1987): 486. 62. Snelling, D. R., Smallwood, G. J., Sawchuck, R. A., and Parameswaran, T. “Precision of Multiplex CARS Temperatures Using Both Single-Mode and Multimode Pump Lasers.” Applied Optics 26, no. 1 (1987). 63. Kroll, S., Alden, M., Berglind, T., and Hall, R. J. “Noise Characteristics of Single Shot Broadband RamanResonant CARS with Single- and Multimode Lasers.” Applied Optics 26, no. 6 (1987). 64. Snelling, D. R., Parameswaran, T., and Smallwood, G. J. “Noise Characteristics of Single-Shot Broadband CARS Signals.” Applied Optics 26, no. 19 (1987): 4298. 65. Westling, L. A., Raymer, M. G., and Snyder, J. J. “SingleShot Spectral Measurements and Mode Correlations in a Multimode Pulsed Dye Laser.” Journal of the Optical Society of America B 1, no. 2 (1984): 150. 66. Greenhalgh, D. A., and Whittley, S. T. “Mode Noise in Broadband CARS Spectroscopy.” Applied Optics 24, no. 6 (1985): 907. 67. Ewart, P. “A Modeless Variable Bandwidth Tunable Laser.” Optics Communications 55 (1985): 124. 68. Kaminski, C. P., and Ewart, P. “Multiples H2 Coherent Anti-Stokes Raman Scattering Thermometry with a Modeless Dye Laser.” Applied Optics 38 (1997): 731. 69. Snelling, D. R., Sawchuck, R. A., and Smallwood, G. J. “Multichannel Light Detectors and Their Use for CARS Spectroscopy.” Applied Optics 23, (1984): 4083. 70. Antcliff, R. R., Hillard, M. E., and Jarrett, O. “Intensified Silicon Photodiode Array Detector Linearity: Application to Coherent Anti-Stokes Raman Spectroscopy.” Applied Optics 14 (1984): 2369. 71. Snelling, D. R., Smallwood, G. J., Sawchuck, R. A., and Parameswaran, T. “Nonlinearity and Single Shot Noise Problems in CARS Spectroscopy.” Paper No. 4; AGARD Conference Proceedings No. 399, Advanced Instrumentation for Aero Engine Components, held in Philadelphia, PA, May 19–23, 1986. 72. Rahn, L. A., Farrow, R. L, and Lucht, R. P. “Effect of Laser Field Statistics on Coherent Anti-Stokes Raman Spectroscopy Intensities.” Optics Letters 9 (1984): 223.
© 2011 by Taylor and Francis Group, LLC
309
73. Eckbreth, A. C., and Hall, R. J. “CARS Concentration Sensitivity with and without Non-Resonant Background Suppression.” Combustion Science and Technology 25 (1981): 175. 74. Lang, B., and Wolfrum, J. “The Impact of Laser Field Statistics in the Determination of Temperature and Concentration by Multiplex USED CARS.” Applied Physics B 51 (1990): 53. 75. Greenhalgh, D. A., and Hall, R. J. “A Closed Form Solution for the CARS Intensity Convolution.” Optics Communications 57 (1986): 12. 76. Hall, R. J., Verdieck, J. F., and Eckbreth, A. C. “Pressure Induced Narrowing of the CARS Spectrum of N2.” Optics Communications 35 (1980): 69. 77. Koszykowski, M. L., Farrow, R. L., and Palmer, R. E. “Calculation of Collisionally Narrowed Coherent AntiStokes Raman Spectroscopy Spectra.” Optics Letters 10 (1985): 478. 78. Marquardt, D. J. “An Algorith for Least Squares Estimation of Nonlinear Parameters.” Journal of the Society Industrial Applied Mathematics 11 (1963): 43. 79. Attal-Tretout, B., Bouchardy, P., Magre, P., Pealat, M., and Taran, J. P. “CARS in Combustion: Prospects and Problems.” Applied Physics B 51 (1980): 17. 80. Parameswaran, T., and Snelling, D. R. “Effect of Spatial Averaging on CARS Derived Temperatures.” Combustion and Flame 106 (1996): 511. 81. Parameswaran, T., and Snelling, D. R. “Estimation of Spatial Averaging of Temperatures from Coherent AntiStokes Raman Spectroscopy.” Applied Optics 35 (1996): 5461. 82. Thumann, A., Seeger, T., and Leipertz, A. “Evaluation of Two Different Gas Temperatures and Their Volumetric Fraction from Broadband N2 Coherent Anti-Stokes Raman Spectroscopy Spectra.” Applied Optics 34 (1995): 331. 83. Garman, J. D., and Dunn-Rankin, D. “Spatial Averaging Effects in CARS Thermometry of a Non-Premixed Flame.” Combustion and Flame 115 (1998): 481. 84. Hughes, P. M., Lacelle, R. J., and Parameswaran, T. “A Comparison of Suction Pyrometer and CARS Derived Temperatures in an Industrial Scale Flame.” Combustion Science and Technology 105 (1995): 131–45. 85. Bengtsson, P. E., Alden, M., Kroll, S., and Nilsson, D. “Vibrational CARS Thermometry in Sooty Flames: Quantitative Evaluation of C2 Absorption Interference.” Combustion and Flame 82 (1990): 199. 86. Mantzaras, J., and Van Der Meer, T. H. “Coherent Anti-Stokes Raman Spectroscopy Measurements of Temperature Fluctuations in Turbulent Natural GasFueled Piloted Jet Diffusion Flames.” Combustion and Flame 110 (1997): 39–53. 87. Grisch, F., Attal-Tretout, B., Bresson, A., Bouchardy, P., Katta, V. R., and Roquemore, W. M. “Investigation of a Dynamic Diffusion Flame of H2 in Air with Laser Diagnostics and Numerical Modeling.” Combustion and Flame 139 (2004): 28–38. 88. Antcliff, R., Jarrett, O., and Rogers, R. “CARS System for Turbulent Flame Measurements.” Journal of Propulsion 1, no. 3 (1985): 205.
310
89. Hildenbrand, S., Staudacher, S., Brüggemann, D., Beyrau, F., Weikl, M. C., Seeger, T., and Leipertz, A. “Numerical and Experimental Study of the Vaporization Cooling in Gasoline Direct Injection Sprays.” Proceedings of the Combustion Institute 31, no. II (2007): 3067–73. 90. Anderson, T. J., and Eckbreth, A. C. “Simultaneous CoherentAnti-Stokes Raman Spectroscopy Measurements in Hydrogen-Fueled Supersonic Combustion.” Journal of Propulsion 8, no. 1 (1992): 7.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
91. Hughes, P. M., Lacelle, R. J., Wong, J., and Moffatt, B. “CARS and Sample Probe Measurements in Two PilotScale Natural Gas Flames.” Combustion Science and Technology 175 (2003): 393–420. 92. Hughes, P. M., Lacelle, R. J., Idris, M., Legere, M., Percy, D., Wong, J., and Parameswaran, T. “CARS and Heat Flux Measurements in Regenerative and Conventional Industrial-Scale Burners.” Combustion Journal, International Flame Research Foundation, 200901 (March 2009). http:// www.journal.ifrf.net/200901hughes.html
14 Diode Laser Temperature Measurements Thomas P. Jenkins and John L. Bergmans Contents 14.1 Introduction.................................................................................................................................................................. 312 14.2 Principles of Tunable Diode Laser Absorption Spectroscopy............................................................................... 312 14.2.1 Overview........................................................................................................................................................ 312 14.2.2 Quantitative Measurements........................................................................................................................ 313 14.2.2.1 Species Concentration.................................................................................................................. 313 14.2.2.2 Temperature...................................................................................................................................314 14.2.3 Practical Implementations............................................................................................................................ 315 14.2.3.1 Calibration Techniques................................................................................................................ 315 14.2.3.2 Environmental Interferences...................................................................................................... 315 14.2.3.3 Sensitive Detection Techniques...................................................................................................316 14.3 History of TDL Sensors in Industrial Applications................................................................................................ 317 14.3.1 Lead-Salt TDL Sensors.................................................................................................................................. 317 14.3.2 Room Temperature Diode Lasers............................................................................................................... 317 14.3.3 Field Tests in Large-Scale Combustion Systems........................................................................................318 14.4 Current and Recent Industrial Systems.................................................................................................................... 319 14.4.1 Air Liquide..................................................................................................................................................... 319 14.4.1.1 Electric Arc Furnace Application............................................................................................... 319 14.4.1.2 TDL Instrument Description...................................................................................................... 320 14.4.1.3 Results............................................................................................................................................ 322 14.4.1.4 Air Liquide Summary.................................................................................................................. 322 14.4.2 MetroLaser/Bergmans Mechatronics LTS-100.......................................................................................... 322 14.4.2.1 System Description....................................................................................................................... 323 14.4.2.2 Large-Scale Testing ..................................................................................................................... 323 14.4.2.3 Future Work................................................................................................................................... 327 14.4.2.4 LTS-100 Summary......................................................................................................................... 328 14.4.3 University of Heidelberg.............................................................................................................................. 328 14.4.3.1 TDL Instrument Description...................................................................................................... 328 14.4.3.2 Results............................................................................................................................................ 329 14.4.3.3 University Heidelberg Summary............................................................................................... 330 14.4.4 Zolo Technologies ZoloBOSS....................................................................................................................... 330 14.4.4.1 System Overview.......................................................................................................................... 330 14.4.4.2 Application-Specific Features..................................................................................................... 330 14.4.4.3 Temperature Measurement Details........................................................................................... 330 14.4.4.4 Accuracy Investigations............................................................................................................... 330 14.4.4.5 Temperature Measurement Applications................................................................................. 332 14.4.4.6 ZoloBOSS Summary..................................................................................................................... 333 14.5 Conclusions................................................................................................................................................................... 333 References................................................................................................................................................................................. 334
311 © 2011 by Taylor and Francis Group, LLC
312
14.1 Introduction As the global supply of fossil fuels diminishes and the total mass of pollutants emitted from combustion systems increases, engineers are coming under growing pressure to increase the efficiencies and reduce pollutants of combustion-based power generation and manufacturing processes. Active control strategies are being developed for optimizing and improving combustion processes, and these strategies often rely on accurate in situ measurements of temperature in the combustion zone. Several existing instruments, including suction pyro meters, IR spectrometers, and acoustic pyrometers are already in regular use for combustion gas temperature measurement. However, conventional sensors are often subject to errors due to disturbances caused by the presence of the probe, ill-defined spatial resolution, and slow response time. Tunable Diode Laser (TDL) absorption spectroscopy seeks to overcome limitations sometimes associated with these traditional instruments. This technique is of particular interest for industrial combustion applications since it provides the following features: • capability to measure temperatures in excess of 3000°F • nonintrusive and in situ • optional capability to simultaneously measure concentrations of species such as H2O, CO, CO2, and O2 • fast response time • well-defined measurement zone geometry Section 14.2 presents a description of the TDL technique for temperature measurement, including the theory and basic principles behind the method. Section 14.3 provides a historical overview of the development of this technique to provide a sense of the considerable amount of research that has been performed in this field to date. Section 14.4 provides details of four TDL-based systems capable of temperature measurement in full-scale environments that are thought to be of particular interest to the industrial combustion community. Finally, Section 14.5 contains some concluding remarks about the current state of TDL sensors and possible outcomes of future developments.
Industrial Combustion Testing
as it passes through a gaseous region to measure temperature and/or species concentrations. The laser wavelength is scanned over a small range that encompasses at least one absorption line of the molecule of interest. Transmitted light intensities are measured using a photodetector and the signals are analyzed to obtain the desired parameters. A basic TDL sensor optical layout is shown in Figure 14.1. The beam from a diode laser is typically highly divergent, and thus needs to be collimated using a lens or a parabolic mirror so that it can traverse a path long enough for the given application. As it transits the measurement zone, the beam gets absorbed by the molecules of interest, and the remaining intensity is measured by the detector. Since molecular absorption is typically a function of wavelength, the laser wavelength is scanned while acquiring the data, producing a transmission spectrum. Each molecule has a unique energy storage structure that determines its absorption spectrum. Molecules in the gas phase store internal energy in three modes: electronic, vibrational, and rotational. A molecule can change from a lower energy state to a higher one by absorbing a photon, whose energy is determined by its wavelength. When the photon energy matches the difference between molecular states, the photon is absorbed. The specific wavelengths absorbed depend upon the molecule’s unique energy structure, and it is this uniqueness that is used to identify the molecule, or to obtain information about a specific molecule. Energy can also be transferred to other molecules via collisions. Figure 14.2 shows an example of the H2O absorption spectrum near 1484 nm at temperatures of 300 K, 1000 K, and 1800 K, representing conditions typical of ambient, flue gas, and flame zone, respectively. It can be seen that the absorption is high at some wavelengths, corresponding to energy transitions of the molecule and low at other wavelengths. The number of molecules at a particular energy state is determined by the Boltzmann distribution. In the absorption process, the number of molecules in the lower state determines how many photons get absorbed. L
Diode laser
14.2 Principles of Tunable Diode Laser Absorption Spectroscopy 14.2.1 Overview Tunable diode laser absorption spectroscopy involves measuring the attenuation of a beam from a diode laser
© 2011 by Taylor and Francis Group, LLC
I0
Collimating lens
I Gas
Detector
Collection lens
Figure 14.1 A typical optical layout for Tunable Diode Laser (TDL) absorption spectroscopy.
313
Diode Laser Temperature Measurements
0.15 T= 0.10
300 K 1000 K 1800 K
α(λ)
intensity, I0. The absorbance due to absorption line j of molecular species i is given the symbol αj, and is defined in terms of the transmission ratio by the Beer–Lambert Law,
P = 1 atm XH2O = 0.50 L = 20 cm λ0 = 1483.6 nm
I = e −α j , I0
(14.1)
0.05
αj is a function of wavelength, temperature, and pressure and is related to the parameters of interest by 0.00 –0.3
–0.2
–0.1
0.0 λ – λ0 (nm)
0.1
0.2
0.3
Figure 14.2 Simulated H2O absorption spectrum near 1484 nm at three different temperatures.
Different transitions will, in general, have different lower state energies and thus different temperature dependencies. In the spectrum of Figure 14.2 there are two strong transitions, or “lines”: one at a relative wavelength, λ – λ0, of approximately –0.2 nm, and one at about λ – λ0 = 0.19 nm, each with a different temperature dependence. Going from 1000 K to 1800 K, the feature on the left decreases while the feature on the right increases. It is the different behavior between these two features with temperature that enables TDL temperature measurements. Also, it can be seen that for both transitions the absorbance at 300 K is small compared to that at higher temperatures. This characteristic is useful for in situ measurements since it helps to minimize interference from ambient H2O absorption. TDL spectroscopy typically involves vibrational and rotational transitions. The fundamental transitions occur at wavelengths in the mid-infrared (mid-IR), from 3 to 12 µm, and overtone and combination bands occur in the near-infrared (near-IR) between 1.3 and 1.6 µm with somewhat weaker absorption. Since diode lasers are mass produced in the near-IR, these wavelengths are more often used for industrial sensor applications, even though the linestrengths in this region are two to three orders of magnitude less than in the fundamental bands. Near-IR TDL sensor systems are currently finding applications in a growing number of industries, including power generation, process monitoring, and combustion research. 14.2.2 Quantitative Measurements Quantitative data are obtained from TDL spectroscopy as follows. The transmission ratio is defined as the ratio of measured intensity at the detector, I, to the incident
© 2011 by Taylor and Francis Group, LLC
αj(λ,Τ,P) = PXiSj(T)φj(λ,Τ,P)L,
(14.2)
where P is the pressure, Xi is the mole fraction of species i, Sj(T) is the linestrength of line j, and L is the path length of the laser beam through the measurement zone. The Sj(T) is typically a strong function of temperature, which enables temperature measurements by TDL absorption. All of the wavelength dependence of αj(λ,T,P) is contained in the lineshape function, φj(λ,T,P), which also depends on pressure due to collisional line broadening and on temperature due to Doppler line broadening. The lineshape function is defined such that when integrated over all wavelengths it equals unity. 14.2.2.1 Species Concentration Equation 14.2 can be used to obtain measurements of mole fraction, Xi, from the measured αi at a particular wavelength if the spectroscopic parameters Sj(T) and φj(λ,T,P) are known for line j. However, in practice it is often difficult to know precisely what wavelength the laser is tuned to, since the wavelength of a diode laser is usually sensitive to temperature and is subject to drift. In addition, the line broadening characteristics for a given line are often not known to a high accuracy. A more accurate approach is to scan the wavelength completely over the line, in which case φj(λ,T,P) integrates to unity and thus drops out of the equation. This can be seen by multiplying both sides of Equation 14.2 by dλ and integrating the wavelength from zero to infinity. Doing this and solving for Xi leads to
Xi =
Aj , PSj (T )L
(14.3)
where Aj is the area under the αj versus λ curve. An experimental scan of laser transmission provides Aj, and Sj(T) is usually obtained from a spectroscopic database if the temperature is known. Often, P and L are constants and can usually be measured independently.
314
Industrial Combustion Testing
14.2.2.2 Temperature
25
Sj (T ) = Sj (T0 )
Q(T0 ) T0 Q(T ) T
hcν0 , j 1 − exp − hcE ″ kT 1 j 1 , exp − − k T T0 hcν0 , j 1 − exp − kT0
10 5 0
0
500
1000
1500 T (K)
2000
2500
3000
Figure 14.3 Linestrengths of two H2O transitions near 1484 nm as a function of temperature, normalized by the linestrength of one of them (Line 2) at 296 K.
Gas temperature can be measured by taking the ratio of areas under two absorption lines. Let S1(T) and S2(T) be the linestrengths of Line 1 and Line 2, respectively. From Equation 14.3, the area is seen to be proportional to the linestrength. Taking the ratio of areas, R = A1/A2, we find that Xi, P, and L cancel out, and we can use Equation 14.4) to obtain
R=
© 2011 by Taylor and Francis Group, LLC
Line 2
15
(14.4) where Q(T) is the molecule’s partition function, and Ej ″ and ν0,j are the lower state energy and linecenter frequency, respectively, of the transition. In addition to the linestrengths, Ej ″ and ν0,j are also tabulated in the HITRAN database, and Q(T) can be approximated by a polynomial (Rothman et al. 2009). The constants h, c, and k are Planck’s constant (6.63 × 10 –34 J·s), the speed of light in a vacuum (3.00 × 108 m/s), and Boltmann’s constant (1.38 × 1023 J/K), respectively. Each of the two prominent features in Figure 14.2 is actually made up of several overlapping lines. However, to simplify the data processing, we treat them each as a single line by finding an effective linestrength and lower state energy for each “line” that adequately describes its behavior with respect to temperature. We will refer to these effective lines as Line 1 and Line 2 for the features at λ – λ0 = 0.19 nm and λ – λ0 = − 0.20 nm, respectively. There are also many weaker lines that cause the small ripples observed in the baseline, but which do not have a large affect on Line 1 or Line 2. This particular portion of the H2O spectrum is employed in a TDL system for monitoring combustion gas in industrial furnaces, the MetroLaser/Bergmans Mechatronics LTS-100, as described in Section 14.4.2. In that system, the laser wavelength is scanned throughout the range shown, covering about 0.6 nm.
Line 1
20 S/S2,296 K
In combustion environments, the temperature is often unknown since it is difficult to measure using traditional instruments. The TDL spectroscopy offers a means to measure gas temperatures nonintrusively using the fact that the linestrength depends on the Boltzmann fraction of molecules in the absorbing state, which is an exponential function of temperature. Spectroscopic databases such as HITRAN (Rothman et al. 2009) and the PNNL spectral library (Sharpe et al. 2004) contain tabulations of the linestrength for each transition for a number of different molecules at a reference temperature (usually 296 K). These linestrengths can be scaled to other temperatures using the relation,
hc ( E1′′− E2′′) 1 1 A1 S1 (T ) S1 (T0 ) exp − = = T − T . k A2 S2 (T ) S2 (T0 ) 0 (14.5)
The factor in Equation 14.4 containing exp(–hcν0,j/kT) accounts for stimulated emission and is negligible at wavelengths below 2.5 µm and temperatures below 2500 K, and therefore it has been neglected. As Equation 14.5 shows, the sensitivity of R to T depends on the difference in the lower state energies, E1″ – E2″. The greater this difference, the greater the sensitivity of the ratio of areas to temperature. Figure 14.3 depicts a plot of the effective linestrengths of the two H2O transitions of Figure 14.2, which have different effective lower state energies. Good sensitivity is obtained because the difference in lower state energy is sufficiently high. This line pair is well suited for combustion environments because the magnitudes of the two linestrengths are comparable at temperatures from 1000 K to 3000 K, and because they are significantly lower at room temperature, which helps to minimize unwanted absorption by room air. The ratio of the two linestrengths, which is equal to the ratio of areas under the αj versus λ curve, is plotted in Figure 14.4. A monotonic relationship between R and T can be seen, which is the basis for TDL temperature measurements.
315
Diode Laser Temperature Measurements
2
Collimator
Flat flame burner Detector
R = A1/A2
1.5
1 Thermocouple
Coax cable
Laptop computer
0.5 Fiber optic 0
0
500
1000
1500 T (K)
2000
2500
3000
Figure 14.4 Ratio of simulated lineshape areas of two H2O lines near 1484 nm as a function of temperature.
14.2.3 Practical Implementations 14.2.3.1 Calibration Techniques In practice, the actual area ratio versus temperature curve will be somewhat different than that shown in Figure 14.4 due to inaccuracies in the database parameters and simplifications in the analytical model, such as the ignoring of the multitude of weak lines within the laser scanning range that also may contribute to the measured absorbance spectrum. These differences from the model can be accounted for by calibrating the TDL system in a controlled environment, such as a burner that produces a stable well-defined flame, and measuring temperatures using a reliable conventional means, such as a thermocouple. A premixed one-dimensional flame produced by a flat flame burner of the McKenna design is often used for temperature calibrations at combustion temperatures. Figure 14.5 shows a typical setup, in which the beam projection optics and the detector are mounted on either side of the burner. Multiple passes can also be used to increase the absorbance by placing mirrors on each side of the burner (not shown). A fine wire S-type thermocouple is mounted near the flame, at the same height as the beam. Care should be taken to align the portion of the thermocouple leads near the bead so that they are within the flame zone. This will prevent strong temperature gradients from occurring near the bead, which can cause errors in the measured temperature. Experience has shown that the indicated thermocouple temperatures are typically stable to within about ±2°F, and the measured absorbance ratios are typically stable to within about ±10°F. At flame temperatures, the thermocouple measurements need to be corrected for radiation losses to the surroundings. The greatest uncertainty in this correction is in the emissivity of the
© 2011 by Taylor and Francis Group, LLC
Lasers and drivers Figure 14.5 Calibration setup for a TDL temperature sensor for combustion zones.
thermocouple bead since it can be at various stages of oxidation. Uncertainties in the measured thermocouple temperatures will depend on the temperature and the size of the bead. The calibration procedure involves collecting both thermocouple data and TDL spectra at different temperatures in the flame, such that a curve similar to that of Figure 14.4 can be constructed. This relationship can then be used for converting measurements of absorption line area ratio to temperature in a practical application, such as in a boiler. 14.2.3.2 Environmental Interferences Nonresonant attenuation over large portions of the spectrum is often encountered in practical environments due to effects that can include scattering from particulates, reflections from windows, or beam steering. These effects can reduce signal levels often as much as several orders of magnitude in applications such as coal-fired power plants. Resulting decreases in signalto-noise ratio adversely affect accuracy and precision. This attenuation by itself is not necessarily a problem, since it can be compensated for using more laser power or more gain. However, the time-varying nature of these attenuations can be particularly challenging, because it can result in such a large dynamic range of signals that detector saturation can readily occur. Methods have been developed to address this problem using techniques such as variable beam power, as demonstrated by Air Liquide (Section 14.4.1), and automatic dynamic range adaptation of the detector signal, as demonstrated by Teichart at al. (2003) and Ebert et al. (2005). Resonant attenuation from molecules in ambient air, for example water vapor in a humid environment, can be a problem as well. For combustion zone measurements, ambient air effects can sometimes be avoided by selecting lines that have little or no absorbance at room
316
Industrial Combustion Testing
(a)
2
P (atm)
0 763.6
763.7
763.8 Wavelength (nm)
763.9
(b)
764
Figure 14.6 Simulated absorbance spectrum of oxygen in air at three pressures.
temperature, minimizing path length outside the measurement zone, and purging with nitrogen. Pressure broadening, or collisional broadening, refers to changes in the shape of absorption features with changes in pressure. A simulated spectrum of oxygen in air at three different pressures is shown in Figure 14.6, illustrating this effect. At one atmosphere the two lines are well separated, while at higher pressures they overlap significantly. The overlapping at high pressure can be a problem because it requires a broader scanning range of the laser to cover a given line, and because it can prevent the fitting of data to a model function for a given line due to interference from nearby lines. Therefore, the sensitivity of absorption features to pressure broadening over the expected range of pressures of a particular application must be considered. If the sensitivity is high, active pressure measurement and compensation in the processing algorithm may be necessary. 14.2.3.3 Sensitive Detection Techniques Measurements of relative absorbance of 10 –3 or greater can be made fairly easily by directly measuring the change in transmitted intensity. At absorbance levels lower than this, changes in the measured intensity may be on the order of the laser or detector noise. In many practical applications, absorbance levels as low as 10 –6 may be required, especially for wavelengths in the near-IR where the linestrengths are orders of magnitude less than at the fundamental bands. There are two sensitive detection techniques in common use for TDL spectroscopy systems to enable measurements of small absorbances: Wavelength Modulation Spectroscopy (WMS), and Balanced Ratiometric Detection (BRD). Wavelength modulation spectroscopy involves modulating the laser current at a frequency, f, typically
© 2011 by Taylor and Francis Group, LLC
Direct absorbance λ0 = 1392 nm x = 10.5 cm Tmax = 1356 K ∆tscan = 3 ms
0.00
1 3 6
1
0.05
Absorbance
T = 296 K XO2 = 0.21 L = 1 cm
2f signal (A.U.)
Absorbance × 104
3
2f absorbance 0.01
0.00 –0.1
0.0 λ−λ0 (nm)
0.1
Figure 14.7 Measured H2O lineshapes in a turbulent industrial spray flame. (Reprinted from Jenkins, T. P., DeBarber, P. A., and Oljaca, M., “A Rugged Low-Cost Diode Laser Sensor for H2O and Temperature Applied to a Spray Flame,” AIAA 2003-585, Paper presented at the 41st AIAA Aerospace Sciences Meeting and Exhibit, Reno, NV, January 6–9, 2003. With permission.)
on the order of tens to hundreds of kHz and lock-in detecting at twice this frequency, 2f (Reid and Labrie 1981). This process is very effective in reducing laser noise, which typically is proportional to 1/f, since it shifts the detection band out to frequencies at which the laser amplitude noise is negligible. The second harmonic is used rather than the fundamental because the 2f signal has an inherently zero background level. This latter feature is especially helpful for discriminating against a rapidly varying background signal in turbulent environments. An example of the benefits of WMS is presented in Figure 14.7, which shows H2O absorbance lineshapes from a TDL sensor applied to an industrial torch used for chemical vapor deposition (Jenkins, DeBarber, and Oljaca 2003a). The system could be used in either direct absorbance or WMS mode. Measurements were made in this highly turbulent flame while the laser current (and thus the wavelength) was scanned with a triangle wave at 200 Hz, corresponding to a single-scan time of about 3 ms. A low amplitude 30-kHz sine wave modulation was also applied to the laser current to enable WMS. The raw signal was lock-in detected at 2f for WMS processing, and was also low-pass filtered for direct absorbance processing. During the scan, the turbulent environment caused large variations in the background emission that manifested themselves as ripples in the direct absorbance lineshape seen at the top of Figure 14.7. In contrast, the 2f WMS lineshape seen at the bottom of Figure 14.7 is much smoother and enabled more accurate
317
Diode Laser Temperature Measurements
measurements of temperature and species concentration in this application. Balanced ratiometric detection involves an electrical circuit that subtracts the photocurrents of two detectors, one of which measures a reference beam from the laser while the other measures the signal beam, to reject common mode laser noise (Haller and Hobbs 1991). The technique is capable of rejecting the laser amplitude noise to a high degree, and has been shown to enable shotnoise-limited measurements of absorption, with relative absorbances as low as 10 –7. Another way to improve detection sensitivity is to increase the path length using a multipass cell. In this method, a sample of the gas being measured is extracted into a cell containing a mirror at each end. The beam is sent into the cell and reflects up to hundreds of times between the two mirrors before exiting and being sent into a detector. Path lengths of tens of meters can typically be produced within a small volume. A commonly used design is the Herriot cell (Herriot, Kogelnik, and Kompfner 1964), which uses curved mirrors and produces an elliptical beam pattern on the mirrors. In addition to providing a long pathlength to increase the absorbance, the pressure inside the cell can be reduced to better isolate individual lines. A disadvantage to using extractive sampling is that the gases can undergo changes in composition due to chemical reactions in the sampling lines, especially in the case of high temperature gases going into lower temperature sampling lines.
14.3 History of TDL Sensors in Industrial Applications 14.3.1 Lead-Salt TDL Sensors The first TDL instruments involved lead-salt diode lasers, which produced wavelengths in the mid-IR and thus were ideally suited for accessing the strong fundamental vibrational bands of most molecules. These sensor systems tended to have very good sensitivity, especially when used with multipass cells, and were found to be useful for atmospheric trace gas detection (Ku, Hinkley, and Sample 1975), as well as for in situ concentration measurements of species in combustion gases (Hanson, Kuntz, and Kruger 1977). Although early works such as these demonstrated in situ measurements at kilohertz rates with excellent sensitivity, the lasers themselves were typically subject to mode hops and had poor reliability. In addition, the lasers needed to be operated at liquid nitrogen temperatures, about 77 K, which required complex cryogenic cooling systems leading to
© 2011 by Taylor and Francis Group, LLC
rather large, expensive, and fragile instruments. Thus, early TDL sensors were not well suited for industrial environments. Developments during the 1980s resulted in lead-salt diode lasers with better reliability that could be operated at temperatures up to 200 K, simplifying the hardware required. The WMS and BRD detection techniques were developed during this time period, demonstrating detection limits at the subparts per billion level for some species (Cassidy and Reid 1982). By 1990, several commercially available mid-IR TDL sensor systems were available using lead-salt diode lasers. However, these systems were designed primarily for the atmospheric sensing market, and were in general not rugged enough for applications such as process monitoring or power plants. 14.3.2 Room Temperature Diode Lasers In addition to improved lead-salt lasers, lasers made from AlGaAs and InGaAsP materials were developed during the 1980s that could be operated at room temperature (Camparo 1985; Wieman and Hollberg 1990). These lasers produced outputs at shorter wavelengths, from 0.7 to 1.6 µm, and thus could not access the fundamental bands in the mid-IR. However, they could reach many overtone and combination bands that have linestrengths two or three orders of magnitude weaker than the fundamental bands. By the mid-1980s, near-IR diode lasers were being produced in large numbers for commercial markets that included telecommunications, bar code scanners, rangefinders, compact disk players, and laser pointers. In 1988, Cassidy demonstrated a sensor based on a single mode fiber-coupled telecommunications laser at 1.3 µm to access the overtone bands of atmospheric trace gases. He was able to demonstrate a minimum detectable absorbance of 5 × 10 –5 using WMS, corresponding to a detection limit of 0.5 ppm for H2O with a 1-m path length (Cassidy 1988). Ron Hanson’s group at Stanford University has been particularly active in developing and refining direct absorbance and WMS TDL sensor methods that could be applied to a large variety of industrial applications. They were the first to demonstrate the two-line ratio technique for temperature in flames, first with CO using a cryogenically cooled lead-salt laser (Hanson and Falcone 1978), and later for H2O using a room temperature near-IR laser (Arroyo and Hanson 1993). Subsequent developments by the group produced increased accuracy using multiple fiber-coupled lasers simultaneously to access H2O line pairs that had more desirable characteristics (Baer et al. 1994, 1996). These efforts led to a demonstration of a wavelength-multiplexed TDL sensor in a closed-loop control system that enabled optimization of the combustion efficiency of a 50 kW naval
318
Industrial Combustion Testing
Temperature (K)
1500
1000
500
x = 10.5 cm Tmax = 1368 K
0 –20 Below CL
–10
Thermocouple TDLAS, deconvolved
0
10
Vertical coordinate, z (mm)
20 Above CL
Figure 14.8 Temperature profiles measured in an axisymmetric turbulent spray flame torch used for chemical vapor deposition. (Reprinted from Jenkins, T. P., DeBarber, P. A., and Oljaca, M., “A Rugged Low-Cost Diode Laser Sensor for H2O and Temperature Applied to a Spray Flame,” AIAA 2003-585, Paper presented at the 41st AIAA Aerospace Sciences Meeting and Exhibit, Reno, NV, January 6–9, 2003. With permission.)
waste incinerator (Furlong et al. 1999). Although this environment was highly turbulent, disturbances to the direct absorbance lineshapes were avoided by scanning the laser fast enough (10 kHz) to effectively “freeze” the flow during each scan. Additionally, over the past decade the Stanford group has implemented various forms of wavelength modulation absorption, which have increased speed and sensitivity of TDL sensing, while reducing sensitivity to noise. Their more recent work has extended the realm of TDL measurements to higher pressures (Reiker et al. 2007) and greater sensitivity using new room temperature lasers at longer wavelengths (Farooq et al. 2008), also providing access to additional species including hydrocarbons, CO, CO2, and NO. As already mentioned in Section 14.2.3.3, Jenkins and coworkers (2003a) applied a TDL system using WMS detection to an industrial torch used for depositing thin films for the microelectronics industry. They demonstrated that TDL sensors could be used for obtaining accurate temperature profiles in a highly turbulent axisymmetric flame. Figure 14.8 shows some results from this study in which TDL temperature profiles across the flame are compared to thermocouple measurements, showing excellent correlation. 14.3.3 Field Tests in Large-Scale Combustion Systems Many control strategies for optimizing processes invol ved in power generation and manufacturing depend on accurate measurements of combustion gas properties,
© 2011 by Taylor and Francis Group, LLC
which can be used in an active feedback loop. TDL sensors have the potential for making in situ measurements of temperature and species in large scale combustion systems, such as power plant boilers, steel furnaces, glass furnaces, and waste incinerators. However, the harsh environments of large industrial furnaces can cause complex sensitive instruments to malfunction or fail. Beginning in the 1990s, TDL sensors have been tested in large-scale combustion facilities with varying degrees of success. Here we touch on some examples. Steel making is a process that uses large amounts of energy, and the electric arc furnaces (EAFs) used are typically only about 55% efficient. Most of the energy is lost in the exhaust gases in the form of thermal and chemical energy. Thus, there exists large potential cost savings if the efficiency can be improved through feedback and control of the combustion process. Researchers at Sandia National Labs (Allendorf et al. 2003) experimented with a TDL sensor system for measuring temperature, CO, and CO2 in a commercial EAF. A challenge in this application was the high level of particulates in the furnace offgases, where they were measuring. They chose long wavelength lasers, near 4.6 µm, to enable higher transmission levels through the particle-laden flow. The lead-salt diode lasers used were difficult to maintain, and eventually succumbed to the harsh environment before they could obtain the quality of data they had hoped for. Researchers at the University of Toronto developed a sensor for EAF applications using a room temperature diode laser at 1.56 µm for temperature and CO, for which they determined accuracies of 30 K and 0.8% for temperature and CO concentration, respectively (Wu, Thomson, and Chanda 2005). However, when they tried to apply this sensor to an actual EAF, they were unable to make measurements because particulate loading was too great to obtain adequate transmission levels. Air Liquide developed and tested a system using similar wavelengths that overcame this problem through the use of a laser amplifier and shortening the path length using beam tubes extending into the exhaust gas region. This system proved to be rugged enough for a commercial EAF environment, and demonstrated high accuracies for both temperature and CO concentration (Von Drasek et al. 2006). Further details are given in a Section 14.4.1. Linnerud, Kaspersen, and Jaeger (1998) developed near-IR TDL systems employing WMS that they applied to various industrial environments, including sensors for O2, CO, and HCl in a waste incinerator, a sensor for HF in an aluminum smelter, and a sensor for NH3 in a coal-fired power plant. Temperature was measured independently using conventional sensors so that the linestrengths could be corrected. Also, they noted that the WMS method they used was susceptible to errors due to line broadening. Nevertheless, they have
319
Diode Laser Temperature Measurements
reported results that demonstrate the fast time response and the value of in situ measurements for several species in industrial environments. Jenkins and coworkers also developed direct absorbance TDL systems that were demonstrated in a number of industrial applications. In one experiment (Jenkins et al. 2003b), a 1500 nm laser was used for acquiring 100 µs scans covering two H2O lines in the combustion region of a 120 kW low NOx burner, the Coen QLN, which was designed for applications such as power plants, petroleum refineries, petrochemical industries, and incinerators. The TDL measurements revealed large fluctuations in temperature that preceded blow-off, and correlated well with chemiluminescence measurements in the same flame. These results suggested that TDL sensors can be used in active control systems to detect signs of blow-off during lean combustion in low-NOx burners. A variation of this sensor system was applied to a fullscale, coal-fired power plant (Jenkins et al. 2004), which produced measurements of boiler temperature showing some dependence on the burner tilt angle and suggested that the sensor could help an operator to find optimum conditions. Although the system proved itself rugged enough to operate for several hours in this harsh environment, the manual alignment procedure was found to be difficult and time consuming. An automatic alignment system was subsequently developed that was tested in a coal-fired power plant (Bergmans and Jenkins 2007). The alignment system worked well, but some temperature measurement accuracy issues of the sensor were revealed that, as of this writing, are still being investigated. The same sensor was also applied to the regenerator section of a float glass furnace (Jenkins and Bergmans 2005). Excellent signal-to-noise ratios were obtained from the high signal levels and long path lengths obtained through the regenerator. Data were acquired continuously for two hours, producing the first ever in situ measurements of temperature and H2O concentration during the cyclic operation of a glass furnace.
14.4 Current and Recent Industrial Systems Details of four TDL sensors capable of measuring combustion gas temperatures in full-scale industrial facilities are presented here. The first system was developed by Air Liquide, and was tested successfully in a number of industrial applications including, most notably, an EAF environment. A second instrument is the LTS-100 Laser Temperature Sensor developed by MetroLaser, Inc. and Bergmans Mechatronics LLC, which was tested in a glass plant, an ethylene cracking furnace simulator, and two coal-fired power plants. Third is a prototype
© 2011 by Taylor and Francis Group, LLC
instrument developed by the University of Heidelberg, which was the first instrument to be successfully tested in a coal-fired power plant. Finally, the ZoloBOSS instrument by Zolo Technologies, Inc. represents perhaps the ultimate embodiment of TDL-based instrumentation for large-scale applications. This instrument features up to 15 measurement beams, the capability to measure temperature and multiple species concentrations, and tomographic mapping software. 14.4.1 Air Liquide Air Liquide, in collaboration with Physical Sciences Inc., developed a TDL instrument for temperature and species concentration measurements in heavy industrial applications. This work, sponsored by the U.S. Department of Energy (DoE), was performed between 2000 and 2006, and involved: (1) the development of a prototype instrument, (2) validation tests in a small-scale Air Liquide furnace, and (3) field tests in three separate industrial facilities. The three industrial facilities were: (1) a steel reheat furnace, (2) a reverberatory aluminum recycling furnace, and (3) an EAF. Documentation on this project is provided in a publicly available final report (Von Drasek et al. 2006). The EAF testing is a highlight of the project since it illustrates the feasibility of using TDL measurement techniques in an unusually harsh environment. Von Drasek and colleagues (2006) describe the challenges of introducing instrumentation to the EAF environment as follows. Instrumentation in this environment must be: able to withstand exposure to high temperatures, immune to strong electromagnetic fields, resistant to mechanical vibration, and protected against flying debris. The challenge for conducting measurements is the dynamic process conditions that are characterized as having large changes in temperature and composition with varying particulate levels in the gas stream. This combination of the environment and monitoring conditions requires special design considerations for successful operation of the sensor on the EAF.
Due to the importance of the successful i mplementation of TDL measurements in the EAF environment, Air Liquide’s instrument as configured for this application is described here. Descriptions of the lab-scale tests and other industrial tests can be found in the DoE report. 14.4.1.1 Electric Arc Furnace Application Air Liquide installed its TDL sensor on an EAF at a Gerdau Ameristeel US, Inc. plant in Wilton, Iowa. This
320
Industrial Combustion Testing
Break flange
Off-gas
Measurement location
Table 14.1 Challenges and Solutions for Off-Gas Measurement in EAF using TDL Instrumentation
Drop-out chamber
Challenges • Time-varying and high level obscurities in beam path (particle densities in excess of 100 g/Nm3)
Molten metal
Electric arc furnace Figure 14.9 Location of TDL measurements in electric arc furnace.
furnace had a single 25 MW AC furnace with four sidewall burners. Two of the burners were natural gas oxygen burners. Measurements of temperature and CO concentrations of the off-gas were made using the TDL sensor along a horizontal line of sight at a “break-flange” mounted on the debris collection drop-out chamber downstream of the EAF (Figure 14.9). The motivation for these measurements in the EAF is for postcombustion control to recover energy from unburned CO in the exhaust using O2 injection. 14.4.1.2 TDL Instrument Description The Air Liquide TDL instrument used in the EAF application employed a TDL operating at a nominal wavelength of 1.56 μm for measurement of CO, H2O, and temperature. The instrument used a pair of H2O absorption features for H2O and temperature measurement and a CO absorption feature for CO measurement. The instrument design incorporates the features listed in Table 14.1 to address the many challenging conditions in this installation. An overview of the system, is described here, followed by descriptions of Air Liquide’s solutions to the challenges of measurements in the EAF environment. A schematic of the system is shown in Figure 14.10. The processing unit contained a TDL source and associated laser control equipment; a rackmount industrial PC with data acquisition hardware and data processing software, and other ancillary components. The processing unit was located in an office next to the control room. Laser light emerged from the TDL at a power level of approximately 20 mW and was amplified to approximately 300 mW. A beam splitter directed 99% of the amplified laser power via a Variable Optical Attenuator (VOA) and a 350–400 foot long fiber optic cable to the launch module. The beam was transmitted across the
© 2011 by Taylor and Francis Group, LLC
• Elevated ambient temperature (Average: 44°C, Peak: 90°C) • High off-gas temperatures (Range: 1000°C to 1600°C) • Electromagnetic fields from EAF
• Module movement
Solutions • ODPC unit incorporating laser amplifier and dynamically adjusted VOA • Pathlength control probes • Nitrogen purge on probes and windows • Water-cooled modules, probes • Total of six thermocouples to monitor module internal temperatures and water cooling lines • All signals between optics modules and processing unit transmitted via fiber optic cable • Capability to remotely adjust receiver optics alignment
measurement location and was received by a photodiode detector within the receiver module. The balance of the amplified laser light from the beam splitter was transmitted down a separate fiber optic cable directly to a detector within the receiver module. The output of both detectors was input into a BRD. The electrical output of the BRD was converted to a light signal and transmitted back via fiber optic cable to a light-to-voltage converter within the processing unit. The signal was then converted to an electrical signal and digitized by an A/D converter. Processing software on the rackmount industrial PC converted the absorption data to water and CO concentration levels and temperature. The use of the BRD in the circuit reduced common mode laser noise in the reference and measurement beams. In addition, the fact that the electrical output of the BRD is directly proportional to absorbance reduces errors that can occur due to baseline fitting to measurement detector data. 14.4.1.2.1 Time-Varying and High Levels of Obscurities The On-Demand Power Control (ODPC) unit addressed the time-varying levels of obscurities present at the break flange. This component used an amplifier to increase the TDL power output to levels sufficient to penetrate the high level obscurities that can occur in the beam path. This solution is suitable when high levels of obscurities are present; however, when obscurity levels drop, the increased laser light transmitted across the measurement location can damage the detector. To accommodate the time-varying nature of the obscurities, the VOA was included in the beam path and used
321
Diode Laser Temperature Measurements
A/D converter and PC
Laser ~20 mW source
Laser 300 mW Splitter amplifier
Detector output
L/V
Reference beam
1% 99%
Variable optical attenuator ~350–400 ft
On-demand power control (ODPC) unit Processing unit (in sensor room)
Measurement beam N2 in Enclosure N2 in (high-flow, Probe water in (continuous) periodic) water in
Flange water in (x2)
N2 in Probe (High-flow, N2 in Enclosure water in periodic) (continuous) water in
~180 cm 60 cm Launch optics
T
Detector Measurement location
Window Water-cooled enclosure Enclosure External water out shell T
T Probe Pathlength water out control probes
Launch module
Flange water out (x2)
Break flange
T
T Probe water out
Detector BRD
V/L
Window Enclosure Water-cooled water out enclosure
T
External shell Receiver module
Legend Fiber Free-space beam T
Thermocouple
L/V
Light to voltage converter
V/L
Voltage to light converter
Figure 14.10 Schematic of Air Liquide TDL instrument installed on Gerdeau EAF.
as part of a closed-loop control system that sought to maintain a constant ratio of reference-to-signal at the BRD. The pathlength control probes also addressed the high levels of obscurities by reducing the length of the beampath exposed to high particulate levels from approximately 180 cm to 60 cm. Two nitrogen purge lines were installed on each side of the measurement location to prevent dust buildup. One provides a continuous flow to the module window and probe and the other provides a high flow purge to the probe automatically every 30 minutes.
© 2011 by Taylor and Francis Group, LLC
14.4.1.2.2 High Ambient and Off-Gas Temperatures Water cooling was employed for the launch and receiver modules, the pathlength control probes, and the break flange. A total of six thermocouples were used during the first several months of operation to monitor the thermal conditions of the optics modules and probes. Thermocouple data was recorded on a separate data logger located near the measurement location. Each module had an internal thermocouple, a thermocouple to measure enclosure water outlet temperature, and a thermocouple to monitor the water outlet temperature of each probe.
322
14.4.1.2.3 Electromagnetic Fields from EAF Of concern in this application of the instrument was the effect of electromagnetic fields generated by the EAF on electrical signals transmitted between the processing unit in the sensor room and optics modules. To address this concern, fiber optic links are used to transmit data between these two locations. 14.4.1.2.4 Module Movement The EAF was operated on a nearly continuous basis, with typically only an eight hour time window available for maintenance every week. To ensure that the alignment of the beam could be maintained in the event of excessive module movement, the optics within the receiver module could be controlled remotely from the processing unit. Long-term testing revealed that the alignment of the modules was more stable than originally anticipated and this feature was not used. 14.4.1.3 Results 14.4.1.3.1 Functionality The instrument operated successfully for a period of over six months (from July to December 2005) with no user intervention for alignment or cleaning of any internal optical components. The only maintenance that occurred during this period was brushing material buildup from the ends of the probes during furnace maintenance shutdowns. This procedure required five minutes of effort. The maximum allowable time between cleanings was not determined; however, the system was operated successfully for periods of up to four weeks without cleaning. One maintenance anomaly occurred in January 2006. Laser power was observed to be decreasing due to a buildup film of what appeared to be water on the receiver entrance window. Laser power recovered after the window was cleaned. Another anomaly was the periodic malfunction of the data acquisition board, which was traced back to elevated ambient temperatures in the nonair-conditioned office in which the processing unit was located. As of the writing of the DoE report in May 2006, the instrument was still in operation at the Gerdau facility. 14.4.1.3.2 Measurements Temperatures measured by the instrument in the EAF averaged approximately 2000°F. Although simultaneous temperature measurements were not made by a separate instrument, this level is consistent with expected values. An estimate of the temperature measurement accuracy of the instrument is provided by tests performed in Air Liquide’s 500 kW oxy/air-fuel pilot furnace.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
Measurements were made at the flue of the pilot furnace with the TDL and with Air Liquide’s Integrated Sampling Probe that incorporates a suction pyrometer. The TDL measurements are made over a pathlength of 30 cm. During these tests, the temperatures measured by the TDL and suction pyrometer matched to within less than 1°F when averaged over a two minute period. Matching occurred both with and without particles added to the flowstream in the test furnace (2590°F and 2507°F, respectively). A direct comparison of the TDL and extractive probe measurements of CO in the EAF is difficult due to the transient nature of the process and the different transient responses of the two instruments. The DoE report does however show that CO measurements made by both instruments in the EAF match to within approximately 5 to 10% CO over a range of 0 to 30% CO, with a band of noise in the TDL data of approximately ±2.5% CO. In an effort to improve signal quality by reducing baseline noise, a water-cooled particle impinger bar was installed immediately upstream of the path control probes. This bar directs a portion of the particulates around the beam path and reduces the interaction of particles with the beam. In this configuration, the TDL and extractive CO measurements agreed to within approximately 2% CO over a range of approximately 0 to 15% CO. The output of both instruments contains a noise band of approximately ±1% CO. 14.4.1.4 Air Liquide Summary This sensor development and testing effort by Air Liquide is significant in that high accuracy was demonstrated in temperature measurements in a pilot furnace simulating EAF conditions and good CO measurement accuracy was achieved in an actual EAF. Two unique innovations enabled successful long-term operation in the adverse EAF environment: a laser amplifier to penetrate high dust loadings, and a VOA to enable operation with varying dust loadings. The use of fiber optic data transfer and rugged optical enclosures permitted the instrument to survive this environment. 14.4.2 MetroLaser/Bergmans Mechatronics LTS-100 MetroLaser, Inc. and Bergmans Mechatronics LLC have developed the LTS-100 Laser Temperature Sensor for measurement of combustion gas temperatures up to 4000°F (2500K). This instrument employs a near-IR laser operating at approximately 1500 nm to make temperature measurements using direct absorption of two closely spaced H2O lines. Testing of this instrument has been performed in large-scale facilities such as coal-fired power plants, a
323
Diode Laser Temperature Measurements
glass plant, and an ethylene cracking furnace simulator. This testing has shown that the instrument is capable of measuring temperatures in these environments, but has also revealed that some additional work is required to improve measurement accuracy. This section describes the current version of the instrument, test results, and items for future work. 14.4.2.1 System Description A schematic of the current version of the system is shown in Figure 14.11. Laser light is transmitted from a TDL within the processing unit (Figure 14.12) to the flange-mounted launch optics enclosure via a fiber optic cable. The laser light passes through the boiler or furnace gasses and is received at the flange-mounted detector enclosure (Figure 14.12). Detector signals are transmitted back to the processing unit via coax cable. Not shown in Figure 14.11 are the power cable to Detector enclosure
Boiler or furnace
Launch optics enclosure
Coax cable
Fiber optic cable
Detector signal
Laser light Processing unit
Terminal room
Figure 14.11 Schematic of typical LTS-100 installation.
Figure 14.12 LTS-100 Processing Unit (L) and detector optics enclosure (R).
© 2011 by Taylor and Francis Group, LLC
provide power from the processing unit to the detector enclosure, and a power and control cable to power the launch optics enclosure and control the laser scanner mirrors that are part of the Laser Alignment System (LAS) (described below). Details on the theory behind the TDL temperature measurement technique used in the LTS-100 are presented in Section 14.2.2. The system features an automatic LAS, which employs two computer controlled mirrors to adjust the pan and tilt angles of the beam. These mirrors are located within the launch optics enclosure. The LAS performs the initial alignment of the beam across the measurement location and maintains alignment, even in the presence of movement of the optical enclosures. This system has been demonstrated to be capable of automatic continuous beam alignment over a pathlength of 48 feet in a coal-fired power plant boiler (Bergmans and Jenkins 2007). In a TDL sensor, long-term variations within the laser can cause changes in the relationship between laser input control current and output wavelength. These changes can create apparent wavelength shifts in the absorption spectrum that in turn result in absorption measurement errors. To address this potential problem, the LTS-100 features a closed-loop absorption peak control system that automatically introduces adjustments to the laser current to ensure fixed positioning of the absorption peaks. 14.4.2.2 Large-Scale Testing 14.4.2.2.1 Glass Plant An early configuration of the LTS-100 was tested in January 2005 in an oil-fired float glass furnace that used a cross-fired furnace with regenerative heat recovery (Jenkins and Bergmans 2005). Figure 14.13 shows a
324
Industrial Combustion Testing
iagram of the furnace, which consists of a tank cond taining molten glass into which the raw materials are fed, and two regenerator chambers, one on each side of the molten glass chamber. The furnace operates in a cycle, reversing the direction of the combustion gases approximately every 20 minutes. As air flows up one of the regenerators it is preheated, then gets combined with a fuel spray to produce flames directed over the surface of the molten glass. Hot gases exit through another regenerator on the other side. The exhaust gases heat a brickwork matrix in the other Laser collimator
Fiber optic
Pool of molten glass
Ports
Regenerator
Ports
18.3 m
LTS-100
Fuel jets
Detector
Raw materials in
Coax cable Top view
Exhausting
2000
1500
H2O Mole fraction
Temperature (K)
Figure 14.13 Float glass furnace schematic showing LTS-100 measurement beam path. (From Jenkins, T. P., and Bergmans, J. L. “Measurements of Temperature and H2O Mole Fraction in a Glass Furnace Using Diode Laser Absorption.” Paper presented at the IEEE Sensors 2005 Conference, Irvine, CA, 2005. With permission.)
regenerator before exiting through a stack. By cycling the inlet air between the regenerators, high temperatures are maintained in the brickwork matrix, resulting in the incoming air being at a temperature of about 1590 K (2403°F) before combustion. For these tests, the optical components were mounted on tripods and located on either end of one of the regenerators. The beam was manually aligned and the absorption peak locations were manually controlled. The path length through the regenerator was 18.3 m (60 feet). Figure 14.14 shows plots of temperature and H2O concentration as a function of measurement time during four hours of furnace operation. Two sets of data were acquired, each lasting about two hours, with a 15-minute break between. The operating characteristics of the furnace are evident in the test data of Figure 14.14. During firing, the temperature and H2O concentration are seen to be about 1590 K (2403°F) and 1%, respectively. These values represent the condition of the air drawn into the furnace just upstream of the combustion zone. The H2O concentration of 1% is close to that of the ambient air, and the temperature of 1590 K is due to preheating by the brickwork matrix that serves as a thermal reservoir. During exhausting, both the temperature and H2O concentration are significantly higher due to combustion, about 2030 K (3195°F) and 10%, respectively. According to the plant engineers, the measured values of both temperature and H2O concentration throughout the cycle were consistent with the design specifications of the furnace. Interesting details of the furnace operation are present in the temperature traces of Figure 14.14. As the exhausting part of the cycle begins, the temperature rapidly jumps to about 1950 K (3051°F). Then it slowly rises, reaching its maximum of about 2030 K at the end
Firing
0.15 0.10 0.05 0.00
0
60
120
t (minutes)
180
240
300
Figure 14.14 LTS-100 temperature and water mole fraction measurements in float glass furnace. (From Jenkins, T. P., and Bergmans, J. L. “Measurements of Temperature and H2O Mole Fraction in a Glass Furnace Using Diode Laser Absorption.” Paper presented at the IEEE Sensors 2005 Conference, Irvine, CA, 2005. With permission.)
© 2011 by Taylor and Francis Group, LLC
325
Diode Laser Temperature Measurements
of the half cycle. Prior to the application of this sensor, these characteristics were not known, since it is not possible to measure these temperatures using existing sensors. During the firing cycle, the temperature trace looks similar but inverted, with an initial rapid drop in temperature followed by a slow decrease. The cooling slope during this part of the cycle would depend upon the thermal mass of the brickwork. Potential applications of the LTS-100 in this type of furnace could include monitoring the performance of the brickwork matrix as a thermal reservoir and optimizing the furnace reversal cycle time. 14.4.2.2.2 Ethylene Cracking Furnace Simulator Initial investigations into the accuracy of the LTS-100 in a full-scale environment were performed in 2005 on an ethylene cracking furnace simulator at John Zink Company, LLC in Tulsa, Oklahoma (Bergmans, Jenkins, and Baukal 2005). For these tests, the LTS-100 employed an early version of the LAS, absorption peak locations were set manually, and the optical components were tripod-mounted. The exterior dimensions of the furnace (Figure 14.15) were 11.5 feet × 8.5 feet with a wall thickness of 1 foot. The exterior height of the furnace was nearly 46 feet. Measurements were made through windows on the furnace, which were offset slightly by tubes on the outside of the furnace walls. The total pathlength between furnace windows was 13.2 feet. The beam path transited the furnace directly above the floor-mounted burners, as shown in Figure 14.15.
The LTS-100 and suction pyrometer tests were performed on both Levels 2 and 4 of the furnace. The results presented here were performed on Level 2 using methane as the fuel with a nominal heat release rate of 9.2 MMBtu/hr. Figure 14.16 shows the LTS-100 temperature and water concentration measured on Level 2 over a period of five minutes. The suction pyrometer traverse was performed along the LTS-100 line-of-sight immediately after the LTS-100 data were acquired. The suction pyrometer data, Figure 14.17, consists of the minimum and maximum temperature readings that occurred in a two minute period at each of the 10 stations across the furnace. The time-averaged LTS-100 temperature data at this location was 1726°F. In comparison, the path averaged temperature calculated using the average of the minimum and maximum suction pyrometer temperatures is 1694°F. Thus, the two instruments agreed with one another to within 32ºF, which is approximately the uncertainty of the suction pyrometer. Tests on Level 4, performed under similar conditions to those performed on Level 2 are described in (Bergmans, Jenkins, and Baukal 2005). The difference between LTS-100 and suction pyrometer data on Level 4 was greater than on Level 2. A contributing factor to the discrepancy on Level 4 may be the larger temporal variability in furnace temperature, which appears to be present at this elevation. Development and testing of the LTS-100 performed subsequent to this work suggests that measurement errors may have been introduced by the attenuation
North Level 4
Collimator /scanner tripod Level 2
Burners Detector tripod
Beam path
West
East
Process tubes (8) South
Figure 14.15 John Zink Company ethylene cracking furnace simulator. (L) exterior view showing measurement locations, (R) internal top view of furnace showing beam path.
© 2011 by Taylor and Francis Group, LLC
326
Industrial Combustion Testing
1900
0.22
Tmax = 1897°F
Tave = 1726°F
0.2
1800
Temperature (°F)
1700
0.18
1600 1500
Tmin = 1511°F
1400
T(F) Tave H2O mole frac H2O ave
XH2O ave = 0.121
1300 1200
0.16
0.14
H2O mole fraction
2000
0.12
1100 1000 12:49:00
12:50:00
12:51:00
12:52:00 Time
12:53:00
0.1 12:55:00
12:54:00
Figure 14.16 Temperature and water concentration data acquired by LTS-100 in ethylene cracking furnace simulator.
2000
Max traverse temp: 1867°F
1900 1800
Temperature (°F)
1700
Delta: 245°F
Average of all min and max points: 1694°F
1600 Min traverse temp: 1465°F
1500 1400
Min temp Max temp Average Ave of min and max
1300 1200 1100 1000 0.0
West window location 2.0
East window location (13.2 ft)
4.0 6.0 8.0 Position from west window (feet)
10.0
12.0
Figure 14.17 Suction pyrometer traverse test data from ethylene cracking furnace simulator.
characteristics of the particular coax cable interfacing the detector and processing unit. Therefore, the agreement of the LTS-100 and suction pyrometer data on Level 2 should be considered tentative until the results are confirmed by repeating these tests. 14.4.2.2.3 Coal-Fired Power Plant The LTS-100 was tested at the Ameren Sioux Station coal-fired power plant during campaigns in 2006 and
© 2011 by Taylor and Francis Group, LLC
2007. The plant has two 485 MW cyclone boilers and each boiler is approximately 138 feet high. During these tests, the LTS-100 launch and detector enclosures were mounted on flanges on the sides of the boiler approximately 103 feet above the bottom of the boiler. The pathlength of the LTS-100 measurement beam was 48 feet. The processing unit was located in the plant terminal room 122 feet below the measurement location.
327
Diode Laser Temperature Measurements
400
Temperature change (°F)
300
discrepancy is the method used by the LTS-100 to determine the incident laser power based on the detector signal. Despite the differences between LTS-100 and expected temperature data, it is significant to note that the LTS100 is capable of resolving changes in internal boiler conditions during load changes. For example, following the downward drop in load at 10:45 pm on September 3, 2007, there was a short upward spike in load. The LTS100 data shows a corresponding steady drop in temperature and an upward spike in temperature during this period. Similarly, the LTS-100 recorded an upward temperature spike that matches the upward load spike during the morning of September 4. 14.4.2.3 Future Work Planned future LTS-100 work currently consists of continued instrument development and improved techniques for independent calibration and full-scale temperature measurement. One system improvement under consideration is the use of a BRD to more accurately measure the H2O absorption spectrum. Another improvement being considered is to locate a remote A/D converter for detector signals within the detector enclosure. The output of this converter would then be transmitted back to the processing unit in digital form, preferably via fiber optic cable. This approach would reduce the sensitivity of the instrument to high frequency signal attenuation effects and potential introduction of electrical noise that can occur with long coaxial cables. During flat-flame calibration testing of the LTS-100, the use of a diminishing diameter thermocouple array 500
Typical high temp (T = 2637°F at 19:03:33 on 9/3/07)
450
200
400
100
350
0
300
–100
250
–200
200
–300
150
–400
–600 09/02/07 18:00
100
LTS-100 based temperature change Estimated boiler temperature change Load
–500 09/03/07 00:00
09/03/07 06:00
09/03/07 09/03/07 12:00 18:00 Date/time
09/04/07 00:00
Load (MW)
For the 2006 campaign, the instrument featured the LAS hardware described above, but no automatic absorption peak control system. In the 2007 campaign, the instrument employed the same LAS hardware, with updated software and featured the peak controller system. An overview of key results of the 2007 campaign is presented here and details of the campaign are described in (Bergmans and Jenkins 2007). The system operated continuously and autonomously over a period of 37 hours between 8 pm on September 2, 2007 and 9 am on September 4, 2007. Figure 14.18 shows the power output, or load, of the generating unit and two temperature values: (1) the change in temperature measured by the LTS-100 relative to the start of the test period, and (2) the change in estimated boiler temperature relative to the start of the test period. As presented in Figure 14.18, during the test period, the generating unit was cycled twice between high (approximately 440 MW) and low (approximately 330 MW) load levels. LTS-100 temperature readings were approximately 2600°F during high load operation and 2300°F during low load operation, resulting in 300°F temperature variation during the test period. By comparison, during this same period, the estimated boiler temperature, which was computed based on plant data for inlet temperature to the economizer, exhibited variations of approximately 200°F to 320°F during load changes. The LTS-100 temperature reading of approximately 2600°F at high load levels are higher than the expected value of approximately 2200°F under these conditions. This expected value is based on thermodynamic modeling, high velocity thermocouple measurements and passive IR spectrometers. One suspected cause of this
50 09/04/07 06:00
0
Figure 14.18 Relative LTS-100 and estimated boiler temperature changes and unit load during testing in coal-fired power plant.
© 2011 by Taylor and Francis Group, LLC
328
Industrial Combustion Testing
is planned. Since a diminishing diameter thermocouple array can provide an accurate estimate of the true gas temperature (Mullikin and Osborn 1941), this technique should provide more accurate measurements of the actual flame temperature during calibration. Prior to performing additional testing in a coal-fired power plant, verification of the instrument accuracy through additional testing in gas-fired furnaces with path lengths of no more that approximately 10 feet is recommended. Independent temperature measurements along the beam path may be made by both suction pyrometers and diminishing diameter thermocouple arrays to ensure an accurate measurement of the actual furnace temperature. 14.4.2.4 LTS-100 Summary The LTS-100 results presented here from three important industrial applications show encouraging performance in each case. The combination of this good performance and the simple design of the instrument make the LTS100 promising for cost-effective, nonintrusive industrial temperature sensing. 14.4.3 University of Heidelberg The research group of Professor Volker Ebert at the University of Heidelberg Physical Chemistry Institute (PCI) performed, to our knowledge, the earliest TDL testing in the combustion zone of large-scale waste incinerators (Ebert et al. 1998), gas-fired power plants
(Ebert et al. 2000), and coal-fired power plants (Teichert, Fernholz, and Ebert 2003). Here we will summarize their efforts to use TDL techniques for the in situ measurement of temperature, as well as CO and H2O concentrations, in a 600 MW coal-fired power plant, as documented by Teichert and coworkers (2003). The significance of this work is that it demonstrates the basic feasibility of long-term, TDL-based measurements in an adverse, full-scale industrial environment. 14.4.3.1 TDL Instrument Description An overview of the instrument, configured for testing at the plant, is shown in Figure 14.19. The 600 MW boiler was fired with pulverized coal and had a 20 m × 20 m cross section. The boiler was approximately 100 m high and TDL measurements were made at an elevation of approximately 65 m, immediately below the first heat exchangers. The instrument employed two TDL. One laser operating at a nominal wavelength of 0.813 μm and output power of 40 mW interrogated three closely spaced H2O absorption features. A second laser operating nominally at 1.56 μm with a power level of 15 mW interacted with a single CO absorption feature. Both lasers resided within the launch enclosure mounted to a flange on the boiler and were modulated over the absorption features of interest at a rate of 5000 Hz. A small fraction of the CO beam was diverted to a reference CO cell to verify the laser overlap with the relevant CO line.
100% CO cell Motor. mirror
Det. Laser (H2O)
Laser drivers
Laser (CO)
Radiative pyrometers
m
20 m
13
Signal generator
Sph. mirror Combustion chamber 600 MWth
Filter Del (CO)
Beamsplitter Filter
Det. (H2O)
Computer ADC Amplifiers
Figure 14.19 University of Heidelberg system configuration in coal-fired power plant. (Reprinted from Teichert, H., Fernholz, T., and Ebert, V., “Simultaneous In Situ Measurement of CO, H2O, and Gas Temperatures in a Full-Sized Coal-Fired Power Plant by Near-Infrared Diode Lasers,” Applied Optics 42, 2043–51, 2003. With permission.)
© 2011 by Taylor and Francis Group, LLC
329
Diode Laser Temperature Measurements
1.2
1440
1.1 1420
1.0 0.9 06:00 PM
A
B
C
D
08:00 PM 10:00 PM Time (hh:mm)
1400
computing species concentrations as shown in Section 14.2.2. In this work, however, the gas temperature data was generated independently by a pair of infrared spectrometers, or radiative pyrometers, at the same elevation in the boiler as the measurement beam. This independent temperature measurement was used in place of a TDL-based temperature measurement since the instrument had only been calibrated for temperature measurements up to 1270 K (1827°F). An extractive, nondispersive infrared (NDIR)-based CO instrument was located in the flue gas duct approximately 100 m downstream of the TDL measurement location. This measurement was used for comparison to the TDL-based CO measurement. 14.4.3.2 Results Measurements were made with the instrument for a total of two 60 hour periods, one on the 13 m path and the other on the 20 m path. Time histories of H2O absorption line ratios and spectrometer temperature data acquired over a six-hour period on the 13 m path are shown in Figure 14.20. We extracted several data points from these time histories and, based on this data, created the plot of temperature versus absorption ratio shown in Figure 14.20. The fact that this plot shows a monotonic relationship illustrates good tracking between the average temperature and average absorption ratio for this instrument. Although extractive CO concentrations were found to be lower than the TDL-based measurements, analysis revealed a close correlation between these two sets of data. The analysis also displayed a time lag between the two measurements, which is to be expected since the extractive measurements were done 100 m downstream of the TDL measurements. 1448 Pyrometric temperature (K)
1450
Pyrometric temperature (K)
Absorption line area ratio (A/C)
The measurement beams were combined, reflected off a motorized alignment mirror, and transmitted through the boiler. The results presented by Teichert and coworkers (2003) described here correspond to the configuration in which the beam traveled along a diagonal 13 m long path across the boiler. They report that subsequent measurements were also made along a 20 m path across the center of the boiler. Upon transiting the boiler, the beams entered the detector optics enclosure, also mounted to a flange on the boiler. Here the beams were separated using a beam splitter and passed through interference filters to reduce background radiation and crosstalk between the two channels. The 1.56 and 0.813 μm beams were then directed to a thermoelectrically cooled InGaAs detector and uncooled silicon photodiode, respectively. Each of the two lasers was modulated with a 5 kHz triangle wave, synchronized to the other and to the data acquisition system. The detector outputs were amplified and digitized at 5 MHz. For each detector, 200 scans were acquired over three seconds and were averaged together and saved to disk. Ten sets of these scans, representing data acquired over a 30 second period, were averaged together. Background emissions still present in the signal after optical filtering were subtracted from the total signal. A baseline was formed by fitting a third-order polynomial to the data in the averaged scans adjacent to the absorption lines of interest. Linear multiline Voigt fits were used to model the absorption line shapes and derive the resulting line areas. Concentrations of CO and H2O were calculated using an analysis similar to that described in Section 14.2.2. In theory, it is possible to use ratios of areas under the H2O lines to measure the gas temperature used for
1446 A
1444 1442
C
B
1440 1438 1436 1434 1432 0.94
D 0.96
2nd order polynomial fit 0.98 1 1.02 Absorption line area ratio
1.04
1.06
Figure 14.20 Absorption ratio and pyrometer temperature measurements. Time histories (L) (Lighter Traces: 30 second average, Darker Traces: 30 minute moving average) (Reprinted from Teichert, H., Fernholz, T., and Ebert, V., “Simultaneous In Situ Measurement of CO, H2O, and Gas Temperatures in a Full-Sized Coal-Fired Power Plant by Near-Infrared Diode Lasers,” Applied Optics 42, 2043–51, 2003. With permission.) and temperature versus absorption rate (R).
© 2011 by Taylor and Francis Group, LLC
330
High levels of beam attenuation were observed during the 13 m path tests. The most probable beam transmission level for the 813 nm laser (Teichert et al. 2003) was determined to be 4 × 10–4, corresponding to 99.96% attenuation of the laser light. Background emissions from the boiler were intermittently high relative to the laser signal. Emission levels in the detector signal were approximately four times the levels of the laser on the CO channel on average, with peaks of over 10 times observed. The automatic beam alignment system was found to perform well, being capable of aligning the beam within minutes. A system was shown earlier by the same group (Ebert et al. 2000) to enable continuous dynamic alignment control. 14.4.3.3 University Heidelberg Summary This work by Heidelberg has been significant in that it demonstrated for the first time that TDL sensors can provide temporally resolved in situ measurements of temperature and species concentrations in the combustion zone of a coal-fired power plant. These measurements are not attainable using conventional sensors. Perhaps even more importantly, the body of work by Heidelberg demonstrates that TDL sensors can be made rugged enough to survive harsh industrial environments, including waste incinerators and power plants. 14.4.4 Zolo Technologies ZoloBOSS The ZoloBOSS sensor, developed by Zolo Technologies, Inc., is a commercially available instrument developed primarily for combustion optimization of coal-fired power plant boilers (Sappey 2007). To achieve this capability the instrument makes measurements of H2O, O2, CO, and CO2 concentrations as well as temperature on up to 15 individual lines of sight. Tomographic mapping software combines measurements from multiple orthogonal beams at a particular elevation into twodimensional temperature or species distribution maps. As of 2007, eight ZoloBOSS systems had been installed in coal-fired power plants. The measurement techniques of the ZoloBOSS system were modeled after those developed by Stanford University (Hanson et al. 2005). 14.4.4.1 System Overview In a typical ZoloBOSS installation, a control rack located in or near the plant control room contains multiple laser sources and detectors, diode laser drivers, and a system computer. The light from the individual lasers, ranging in nominal wavelengths between 0.67 and 1.7 μm and tuned to CO, CO2, O2, and H2O absorption features are multiplexed into one single-mode fiber using Zolo’s Zmux multiplexer/demultiplexer device.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
The beam travels via fiber to a Matrix Distribution Cabinet (MDC) located near the boiler. A fiber optic switch within this enclosure routes the laser light to the transmitter head on a particular measurement path. The beam transits the measurement location and is focused into a fiber within the receiver head. From here, the beam is routed back by fiber to a fiber optic switch in the MDC and then back to the control rack. Another Zmux in the control rack demultiplexes the incoming beam into multiple beams of different wavelengths that are then received by multiple photodiode detectors. Signals from the individual detectors are processed by the system computer to produce the species concentrations and temperature along the current beam path. Measurements along each beam path require five seconds to acquire (Sappey 2009). Thus, a complete view of temperature and species concentrations within the boiler can be achieved within approximately 75 seconds. 14.4.4.2 Application-Specific Features The ZoloBOSS includes two features to address the challenges of in situ measurements in coal-fired power plants (Zolo 2009a). The first feature is the SensAlign automatic alignment system, which is used to maintain alignment of the optical components in the flangemounted transmitter and receiver heads in the presence of movement of the mounting locations that can occur due to changes in boiler conditions. The second feature is the ClearView port rodding system, which is designed to automatically remove slag, ash, or dust that can block the ports of instrument heads located within the combustion zone. Images of the optical heads are shown in Figure 14.21. 14.4.4.3 Temperature Measurement Details As described in Section 14.2.2, TDL-based measurement of temperature along a beam path is possible in theory by using the ratio of two H2O absorption features. The ZoloBOSS, however, measures temperature using six H2O features. This large number of features is used to improve temperature measurement accuracy (Sappey and Zimmerman 2008). The ZoloBOSS system measures absorption at a particular wavelength by comparing the ratio of the laser signal passing through the measurement location with a reference beam that does not pass through the measurement zone. 14.4.4.4 Accuracy Investigations Zolo has performed testing in their own laboratory calibration facility to investigate the accuracy of the
331
Diode Laser Temperature Measurements
Figure 14.21 ZoloBOSS optical housing. Housing mounted to flange (L). (Reprinted from Zolo Technologies, Inc., “Introducing ZoloBOSS Boiler Optimization Spectroscopy Sensor,” Zolo Technologies, Inc., Presentation, 2009, http://www.zolotech.com/pdfs/2008-ZoloBOSS.pdf [accessed April 24, 2009]. With permission.) Reverse view showing port rodder details (R). (Reprinted from Zolo Technologies, Inc., “SensAlign Auto-Alignment,” Zolo Technologies, Inc., webpage, 2009, http://www.zolotech.com/sub/power/sensalign.php [accessed April 28, 2009]. With permission.)
Table 14.2 ZoloBOSS Measurement Specifications Constituents
Temperature Range °F
Accuracy
Minimum Detectable Limit
Maximum Detectable Limit
Long-Term Repeatability
Temperature O2 H2O CO CO2
— 500–3000 500–3000 500–3000 500–800
± 5% of measured value (omv) ± (0.2% abs + 5% omv) ± (0.1 + 5% omv) ± (500 ppm + 5% omv) ± (0.15% + 5% omv)
500°F 0 0.001 500 ppm 0.003
3000°F 0.15 0.3 1 1
± 0.1% omv ± 0.02% ± 0.02% ± 100 ppm ± 0.02%
Source: From Sappey, A. D., and Zimmerman, R. J., Accuracy verification of ZoloBOSS measurements, Zolo Technologies, Inc., White Paper, 2008, http://www.zolotech.com/pdfs/zoloboss-accuracy-whitepaper.pdf (accessed April 28, 2009). With permission. Note: omv = of measured value.
ZoloBOSS temperature and species concentration measurement capabilities. The firm has also performed full-scale temperature measurement accuracy testing in power plant conditions (Sappey and Zimmerman 2008). These tests are described next. 14.4.4.4.1 Lab Testing The Zolo laboratory calibration facility is a test cell with a path length of approximately 45 cm and can reach temperatures of up to 2060°F. Controlled levels of target gases can be introduced into the cell. Temperatures are measured using thermocouples and concentration levels are measured using independent conventional laboratory instruments. The facility is able to simulate phenomenon that occur when making measurements in a full scale boiler, including background emissions, low signal to noise ratios, and laser transmission noise. Testing in the Zolo laboratory under these simulated boiler conditions has confirmed that the instrument performs to within the specifications of the instrument, which are shown in Table 14.2.
© 2011 by Taylor and Francis Group, LLC
14.4.4.4.2 Full-Scale Testing The standard instrument for temperature measurement in the coal-fired power plant industry is the High Velocity Thermocouple (HVT) probe (Babcock & Wilcox Company 1972, Mullikin and Osborn 1941). This device is a water-cooled tube, up to approximately 20 feet in length that is inserted into the combustion gases. Combustion gases are drawn into the tube and pass over a thermocouple located near the inlet to the tube. One of the objectives of locating the thermocouple in the tube is to reduce errors due to radiation losses from the thermocouple to the cooler zone within the combustion environment, such as the walls of the boiler. To determine how the ZoloBOSS temperature measurement capability compares to HVT data, Zolo measured gas temperatures with an HVT probe at a location in a full-scale, coal-fired boiler where nine ZoloBOSS beams were being used for temperature measurements. For these tests, one ZoloBOSS beam was removed and HVT measurements were made at several stations along this beam path.
332
Industrial Combustion Testing
As shown in Figure 14.22, the path average of the HVT measurements was found to be 2421°F. Two sets of ZoloBOSS data can be compared with this value. One is the average temperature data from the ZoloBOSS, acquired before and after the HVT tests, along the path that was temporarily removed. This temperature reading was 2431°F. The second set of ZoloBOSS data
is the time averaged data of the eight remaining beams acquired during the HVT traverse. This temperature reading is 2461°F. Since the ZoloBOSS temperatures were within 40ºF of the HVT temperatures, these tests illustrate that ZoloBOSS temperature measurements are consistent with HVT traverse measurements. 14.4.4.5 Temperature Measurement Applications Temperature measurement in the combustion zone of the boiler of a coal-fired power plant offers practical benefits to the power industry, as shown here with two examples. The capabilities of the ZoloBOSS for measurement of species concentrations and tomographic mapping are also of significance for industrial applications. However, consideration of these aspects of the ZoloBOSS are outside the scope of this chapter and thus they are not discussed here.
HVT-ZoloBOSS comparison
2600
Temperature (°F)
2550 2500
ZoloBOSS path
2450
HVT average
2400
HVT individual points
2350 2300
0
100
200
300 Distance (in)
400
500
Figure 14.22 Comparison of ZoloBOSS and HVT measurements in a coal-fired power plant boiler. (From Sappey, A. D., and Zimmerman, R. J., Accuracy verification of ZoloBOSS measurements, Zolo Technologies, Inc., White Paper, 2008, http://www.zolotech.com/pdfs/zolobossaccuracy-whitepaper.pdf [accessed April 28, 2009]. With permission.)
14.4.4.5.1 Avoiding Outages from Slagging Zolo reports an interesting example that illustrates the potential benefit of measuring temperatures in a coal-fired power plant boiler (Zolo 2009c). In this case, a nine-path ZoloBOSS system was installed on an 830 MW tangentially fired plant. As shown in Figure 14.23, six beams were placed in a 3 × 3 matrix below the boiler nose, and three parallel beams were installed at stations in the superheater section in an orientation perpendicular to the flow.
Superheat paths 2800
Average temperature, combustion zone
Temperature °F
2600
Combustion Paths ZoloBOSS configuration:
ZoloBoss off
2400 2200
Path 11 (Superheat)
2000 1800
Clinker falls and unit trips
1600 1400
1
5
10
15
20
25
30 35 Days
40
45
50
55
9 path system temperature, O2, CO, and H2O OPC data connection Figure 14.23 ZoloBOSS measurements during clinker fall. Measurement locations (L) and temperature time histories (R). (Reprinted from Zolo Technologies, Inc., “Avoiding Outages from Slagging,” Zolo Technologies, Inc., White Paper, 2009, http://www.zolotech.com/pdfs/case-study-avoidingslagging.pdf [accessed April 24, 2009]. With permission.)
© 2011 by Taylor and Francis Group, LLC
333
Diode Laser Temperature Measurements
During the course of commissioning the ZoloBOSS at this plant, coal ash gradually formed a deposit of slag in the superheater area of the boiler. The slag then fell in the form of a large “clinker” to the bottom of the furnace and caused damage, which resulted in a two-day outage to repair the unit. The ZoloBOSS temperature data captured this event as a distinctive upward shift in temperature in the superheat region of over 100°F between Day 35 and Day 45 of the commissioning process (Figure 14.23). In contrast, the average temperature measured at the furnace exit plane by the six-beam grid remained at a constant level. The increase in superheat temperature while the combustion zone temperature remained constant can be explained by the fact that slag buildup acts as an insulator and reduces heat transfer to the internal surfaces of the boiler. Therefore, while the boiler temperature remained constant during this event, superheater gas temperatures increased since the heat being removed from the boiler gasses was reduced. This temperature trend suggests that it may be possible to use the ZoloBOSS to detect unexpected changes in heat transfer due to slag and thus provide an early warning to operators about the presence of slag accumulation. To illustrate the value of this technology in monetary terms, Zolo estimates that a two-day outage for an 830 MW plant would result in lost revenue of $1.75 million. The investment in the instrument could therefore be justified if it could be shown that the ZoloBOSS enabled the plant to avoid even one such incident. 14.4.4.5.2 Support for Intelligent Soot Blowing Another example of TDL-based temperature measurement in a coal-fired power plant shows how the ZoloBOSS instrument can be used to replace existing instrument technologies (Zolo 2009d). In this case, ZoloBOSS data were used to replace that from an optical pyrometer in an Intelligent Soot Blowing (ISB) system. The ISB systems are used in power plants to automatically control cleaning of the heat transfer surfaces within the boiler. These systems employ several types of sensors at different locations within and downstream of the boiler to estimate the locations of slag formation. One such instrument is a temperature sensor, such as an optical pyrometer, located near the top of the boiler for Furnace Exit Gas Temperature (FEGT) measurements. A 274 MW power plant in which Zolo had installed a ZoloBOSS used an optical pyrometer for FEGT measurement input to the ISB system. The ZoloBOSS system at this plant was a 15 path configuration with 9 paths in a 5 × 4 grid located below the nose of the boiler, at a location 13 feet below the pyrometer. Following a failure of the optical pyrometer, the ZoloBOSS temperature measurement was first validated
© 2011 by Taylor and Francis Group, LLC
2450°F 2400°F
Zolo average furnace temperature
2350°F 2300°F Optical pyrometer 2250°F 5/2/2008 8:03:03 am
4.72 hours
5/2/2008 12:46:33 pm
Figure 14.24 Comparison of ZoloBOSS and optical pyrometer measurements in 274 MW coal-fired boiler. (Reprinted from Zolo Technologies, Inc., “ZoloBOSS Enables Intelligent Soot Blowing,” Zolo Technologies, Inc., White Paper, 2009, http://www.zolotech.com/pdfs/case-studyintelligent-soot-blowing.pdf [accessed April 24, 2009]. With permission.)
using a comparison with historical pyrometer outputs (Figure 14.24) and then used as a replacement FEGT input measurement to the ISB. An important advantage of the ZoloBOSS multipath measurement over the optical pyrometer for FEGT measurement is that the multipath measurement provides an improved representation of the boiler exit conditions. The improvement comes about because the measurement zone of the ZoloBOSS is well defined and covers a large area, while the measurement zone of an optical pyrometer may be difficult to quantify. 14.4.4.6 ZoloBOSS Summary The unique measurement capabilities, good temperature measurement accuracy, and robust mechanical design of the ZoloBOSS instrument are notable, and these attributes appear to meet industrial requirements as evidenced by a number of existing industrial installations. Future developments and applications of the ZoloBOSS will be of interest to those in the combustion community who require insight into their combustion processes.
14.5 Conclusions In this chapter we have attempted to provide an overview of the state of the art TDL-based sensors for temperature and species concentrations in industrial applications. To paint an accurate picture of current capabilities, we have discussed the limitations as well as the strengths of this class of sensors. As the work outlined here shows, TDL sensors offer an unprecedented capability for rapid in situ measurements with high accuracy. However, recent TDL temperature sensor installations in combustion environments have revealed difficulties
334
that require innovative solutions. Challenges are related to the robustness of the sensor hardware, interferences from harsh environments, and spectroscopic uncertainties. In at least one case, however, some of the biggest obstacles have already been overcome, as illustrated by the encouraging results from the ZoloBOSS system for the power generation industry. We believe that the full potential of TDL-based sensor systems for industrial combustion monitoring has not yet been realized, and it is only a matter of time before diode laser sensors enable new levels of efficiency through active control of critical industrial processes such as power generation, waste incineration, and glass and steel manufacturing, just to name a few.
References Allendorf, S. W., Ottesen, D. K., Green, R. W., Hardesty, D. R, Kolarik, R., Goodfellow, H., Evenson, E., et al. “Optical Sensors for Post Combustion Control in Electric Arc Furnace Steelmaking.” Final Report, AISI/DOE Technology Roadmap Program, Cooperative Agreement No. DE-FC36-97ID13554. Prepared for U.S. Department of Energy, December 31, 2003. Arroyo, M. P., and Hanson, R. K. “Absorption Measurements of Water-Vapor Concentration, Temperature, and LineShape Parameters Using a Tunable InGaAsP Diode Laser.” Applied Optics 32 (1993): 6104–115. Babcock & Wilcox Company. Steam: Its Generation and Use. 38th ed. Whitefish, MT: Kessinger Publishing, 1972. Baer, D. S., Hanson, R. K., Newfield, M. E., and Gopaul, N. K. L. M. “Multiplexed Diode-Laser Sensor System for Simultaneous H2O, O2, and Temperature Measurements.” Applied Optics 33 (1994): 3296–3306. Baer, D. S., Nagali, V., Furlong, E. R., Hanson, R. K., and Newfield, M. E. “Scanned- and Fixed-Wavelength Absorption Diagnostics for Combustion Measurements Using a Multiplexed Diode-Laser Sensor System.” AIAA Journal 34 (1996): 489–93. Bergmans, J. L., and Jenkins, T. P. “Progress in the Development of a TDLAS Sensor for Industrial Applications.” Paper presented at the American-Japanese Flame Research Committees International Symposium, Waikaloa, HI, 2007. Bergmans, J. L., Jenkins, T. P., and Baukal, C. E. “Accuracy of a Tunable Diode Laser Sensor in Large Scale Furnaces: Initial Test Results.” Paper presented at the American Flame Research Committee/Georgia Institute of Technology, Department of Aerospace Engineering, Fall 2005 Symposium, Atlanta, Ga. Camparo, J. C. “The Diode Laser in Atomic Physics.” Contemporary Physics 26 (1985): 443–77.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
Cassidy, D. T. “Trace Gas Detection Using 1.3-gm InGaAsP Diode Laser Transmitter Modules.” Applied Optics 27 (1988): 610–14. Cassidy, D. T., and Reid, J. “Atmospheric Pressure Monitoring of Trace Gases Using Tunable Diode Lasers.” Applied Optics 21 (1982): 1185–90. Ebert, V., Fernholz, T., Giesemann, C., Pitz, H., Teichert, H., Wolfrum, J., and Jaritz, H. “Simultaneous DiodeLaser-Based In-Situ-Detection of Multiple Species and Temperature in a Gas-Fired Power-Plant.” Proceedings of the Combustion Institute 28 (2000): 423–30. Ebert, V., Fitzer, J., Gerstenberg, I., Pleban, K.-U., Pitz, H., Wolfrum, Brown, T. M., et al. “Simultaneous LaserBased In-Situ-Detection of Oxygen and Water in a Waste Incinerator for Active Combustion Control Purposes.” Proceedings of the Combustion Institute 27 (1998): 1301–8. Ebert, V., Teichert H., Strauch P., Kolb T., Seifert H., Wolfrum, J. “Sensitive in situ detection of CO and O2 in a rotary kilnbased hazardous waste incinerator using 760 nm and new 2.3 mm diode lasers.” Proceedings of the Combustion Institute 30 (2005) 1611–18. Farooq, A., Jeffries, J. B., and Hanson, R. K. “In Situ Combustion Measurements of H2O and Temperature near 2.5 μm using Tunable Diode Laser Absorption.” Measurement Science and Technology 19 (2008): 075604. Furlong, E. R., Mihalcea, R. M., Webber, M. E., Baer, D. S., and Hanson, R. K. “Diode-Laser Sensors for Real-Time Control of Pulsed Combustion Systems.” AIAA Journal 37 (1999): 732–37. Haller, K. L., and Hobbs, P. C. D. “Double Beam Laser Absorption Spectroscopy: Shot Noise-Limited Performance at Baseband with a Novel Electronic Noise Canceller.” SPIE 1435 (1991): 298–309. Hanson, R. K., and Falcone, P. K. “Temperature Measurement Technique for High-Temperature Gases Using a Tunable Diode Laser.” Applied Optics 17 (1978): 2477–80. Hanson, R. K., Jeffries, J. B., Zhou, X., Liu, X., Li, H., Mattison, D., Klingbeil, A. “Smart Sensors for Advanced Combustion Systems.” Global Climate and Energy Project Technical Report 2005. http://gcep.stanford.edu/pdfs/ib0KJnIw5S-8G1RS8EqnPg/hanson_tech05web.pdf (accessed May 3, 2009). Hanson, R. K., Kuntz, P. A., and Kruger, C. H. “High-Resolution Spectroscopy of Combustion Gases Using a Tunable IR Diode Laser.” Applied Optics 16 (1977): 2045–48. Herriot, D., Kogelnik, H., and Kompfner, R. “Off-Axis Paths in Spherical Mirror Interferometers.” Applied Optics 3 (1964): 523–26. Jenkins, T. P., and Bergmans, J. L. “Measurements of Temperature and H2O Mole Fraction in a Glass Furnace Using Diode Laser Absorption.” Paper presented at the IEEE Sensors 2005 Conference, Irvine, CA, 2005. Jenkins, T. P., Bergmans, J. L., DeBarber, P. A., Coffey, M. R., and Starnes, G. M. C. “In Situ Measurements of Temperature in a Coal-Fired Power Plant Using Tunable Diode Laser Absorption Spectroscopy.” Paper presented at AFRC/JFRC 2004 Joint International Symposium, Maui, HI, 2004.
Diode Laser Temperature Measurements
Jenkins, T. P., DeBarber, P. A., and Oljaca, M. “A Rugged Low-Cost Diode Laser Sensor for H2O and Temperature Applied to a Spray Flame.” AIAA 2003-585. Paper presented at the 41st AIAA Aerospace Sciences Meeting and Exhibit, January 6–9, Reno, NV, 2003a. Jenkins, T. P., DeBarber, P. A., Shen, J., and McDonell, V. G. 2003B. “Time-Resolved In-Situ Temperature Measure ments in a Model Industrial Burner Using a Tunable Diode Laser.” Paper presented at the 2003 Western States Section/Combustion Institute, October 20–21, University of California, LA, 2003b. Ku, R. T., Hinkley, E. D., and Sample, J. O. “Long-Path Moni toring of Atmospheric Carbon Monoxide with a Tunable Diode Laser System.” Applied Optics 14 (1975): 854–61. Linnerud, I., Kaspersen, P., and Jaeger, T. “Gas Monitoring in the Process Industry Using Diode Laser Spectroscopy.” Applied Physics B 67 (1998): 297–305. Mullikin, H. F., and Osborn, W. J. “Accuracy Tests of the High Velocity Thermocouple.” In Temperature, Its Measurement and Control in Science and Industry, 805–29. New York: Reinhold Publishing, 1941. Reid, J., and Labrie, D. “Second-Harmonic Detection with Tunable Diode Lasers - Comparison of Experiment and Theory.” Applied Physics B 26 (1981): 203–10. Rieker, G. B., Lil, H., Liu, X., Jeffries, J. B., Hanson, R. K., Allen, M. G., Wehe, S. D., Mulhall, P. A., and Kindle, H. S. “A Diode Laser Sensor for Rapid, Sensitive Measurements of Gas Temperature and Water Vapour Concentration at High Temperatures and Pressures.” Measurement Science and Technology 18 (2007): 1195–1204. Rothman, L. S., Gordon, I. E., Barbe, A., Benner, D. C., Bernath, P. F., Birk, M., Boudon, V., et al. “The HITRAN 2008 Molecular Spectroscopic Database.” Journal of Quantitative Spectroscopy & Radiative Transfer 110 (2009): 533–72. Sappey, A. D. “Wavelength-Multiplexed Diode Laser Spe ctroscopy for Closed Loop Combustion Control and Optimization.” Paper presented at the American-Japanese Flame Research Committees International Symposium, Waikaloa, HI, 2007. Sappey, A. D. E-mail message to J. L. Bergmans, April 29, 2009.
© 2011 by Taylor and Francis Group, LLC
335
Sappey, A. D., and Zimmerman, R. J. Accuracy verification of ZoloBOSS measurements, Zolo Technologies, Inc. White Paper, 2008. http://www.zolotech.com/pdfs/zolobossaccuracy-whitepaper.pdf (accessed April 28, 2009). Sharpe, S. W., Johnson, T. J., Sams R. L., Chu, P. M., Rhoderick, G. C., and Johnson, P. A. “Gas-Phase Databases for Quantitative Infrared Spectroscopy.” Applied Spectroscopy 58 (2004): 1452–61. Teichert, H., Fernholz, T., and Ebert, V. “Simultaneous In Situ Measurement of CO, H2O, and Gas Temperatures in a Full-Sized Coal-Fired Power Plant by Near-Infrared Diode Lasers.” Applied Optics 42 (2003): 2043–51. Von Drasek, W., Pubill-Melsió, A., Fauve, E., Mulderink, K., Allen, M., Mulhall, P., Frish, M., et al. “Tunable Diode Laser Sensors for Monitoring and Control of Harsh Combustion Environments.” Department of Energy Final Report, Cooperative Agreement No. DE-FC36-00CH11030, 2006. Wieman, C. E., and Hollberg, L. “Using Diode Lasers for Atomic Physics.” Review of Scientific Instruments 62 (1990): 1–20. Wu, Q., Thomson, M. J., and Chanda, A. 2005. “Tunable diode laser measurements of CO, H2O, and Temperature Near 1.56 µm for Steelmaking Furnace Pollution Control and Energy Efficiency.” Metallurgical and Materials Transactions B 36B (2005): 53–57. Zolo Technologies, Inc. “SensAlign Auto-Alignment.” Zolo Technologies, Inc. webpage, 2009a. http://www.zolotech.com/sub/power/sensalign.php (accessed April 28, 2009). Zolo Technologies, Inc. “Introducing ZoloBOSS Boiler Optimization Spectroscopy Sensor.” Zolo Technologies, Inc., Presentation, 2009b. http://www.zolotech.com/ pdfs/2008-ZoloBOSS.pdf (accessed April 24, 2009). Zolo Technologies, Inc. “Avoiding Outages from Slagging.” Zolo Technologies, Inc. White Paper, 2009c. http://www. zolotech.com/pdfs/case-study-avoiding-slagging.pdf (accessed April 24, 2009). Zolo Technologies, Inc. “ZoloBOSS Enables Intelligent Soot Blowing.” Zolo Technologies, Inc. White Paper, 2009d. http://www.zolotech.com/pdfs/case-study-intelligentsoot-blowing.pdf (accessed April 24, 2009).
15 Image-Based Techniques for the Monitoring of Flames Javier Ballester and Ricardo Hernández Contents 15.1 Introduction................................................................................................................................................................. 337 15.2 Overview of Vision-Based Methods for Flame Characterization........................................................................ 338 15.3 Image Processing for Flame Identification: General Approach and Testing Procedure................................... 340 15.4 Flame Identification with Self-Organizing Feature Maps..................................................................................... 342 15.4.1 Image Preprocessing..................................................................................................................................... 343 15.4.2 Configuration and Training of the SOFM................................................................................................. 343 15.4.3 Topological Classification............................................................................................................................. 343 15.4.4 Estimation of NOx Emissions Using Flame Images................................................................................. 345 15.4.5 Application Situations.................................................................................................................................. 346 15.5 A Pattern-Recognition Technique for Flame Identification.................................................................................. 347 15.5.1 Image Preprocessing..................................................................................................................................... 347 15.5.2 Probability Calculation and Pattern Selection.......................................................................................... 348 15.5.3 Classification Tests........................................................................................................................................ 349 15.5.4 Application to Premixed Flames................................................................................................................. 351 15.6 Conclusions.................................................................................................................................................................. 351 Acknowledgments.................................................................................................................................................................. 353 References................................................................................................................................................................................. 353
15.1 Introduction Industrial flames are an outstanding example of complex reacting flows. They involve many physical and chemical magnitudes (velocity, temperature, radiation, chemical species, reaction rates, etc.), usually displaying large spatial and temporal variations. An enormous amount of information is, therefore, necessary to describe a flame in detail. Scientists have developed many diagnostic methods (including intrusive [1] and optical techniques [2]), which have contributed significantly to the progress of the combustion science. However, their use is still mostly restricted to research studies, with scarce applications for the supervision of practical flames. In general, most industrial combustors use conventional instrumentation, whose capabilities for flame monitoring are notably poor; in many cases, only input flow rates and flue gas composition are measured. There is an obvious need for new monitoring techniques, suitable for industrial use and, preferably, capable of providing direct information on the properties of the flame, which is the core of a combustion process. New developments in this field might open new possibilities for the supervision of the process
and, ultimately, would greatly facilitate the development of advanced combustion controls. In this context, vision-based techniques are emerging as a viable alternative for the monitoring of practical combustion equipment. These methods are based on the capture of radiation spontaneously emitted by a flame (as opposed to laser-based techniques, requiring external illumination) and offer some important advantages: (1) they are nonintrusive, (2) flame images can be recorded with rugged, relatively cheap sensors, (3) they provide direct information on the flame, and (4) include a large amount of data in the form of 2-D maps. This, together with the fast progress in sensor technology, probably explains the growing interest in the use of CCD (charge coupled device) cameras for the monitoring of industrial flames. A significant shortcoming is, however, the difficulty of converting flame images into meaningful results. Video cameras installed in furnaces or boilers are normally used to provide visual information to the operators, who finally are responsible for the interpretation of the images (usually by comparing with previous experience). However, the automatic conversion of visual data into practical information or in the form of physical combustion parameters is still a challenging objective. In 337
© 2011 by Taylor and Francis Group, LLC
338
the authors’ opinion, this is the most critical aspect that needs to be solved in order to develop advanced monitoring and control methods based on flame images. Widely different approaches for the capture and processing of flame images have been developed and tested. A review of published works in this field is presented in Section 15.2. One of the alternatives is to develop image classifiers, capable of automatically identifying a flame with respect to situations previously characterized. This chapter presents two novel processing techniques that have been developed with this purpose. Section 15.3 summarizes the general approach and the combustion facilities used to evaluate the performance of the two methods, described in Sections 15.4 and 15.5, respectively. One of them uses self-organizing feature maps (SOFM) and yields as output the most probable combustion regime, among those previously characterized (Section 15.4). The other one (Section 15.5) is an adaptation of a speech recognition method and informs about the probability of an unknown flame to correspond to each of the different combustion regimes. Finally, in Section 15.6 the main conclusions are summarized.
15.2 Overview of Vision-Based Methods for Flame Characterization In many cases, the performance of a combustion system can be quantified in terms of a few parameters, such as oxygen concentration and temperature of flue gases, pollutant emissions, steam production, and so on. However, these data may not be enough to describe the actual state of the combustion process. A clear example is the case of multiburner chambers, where global values do not represent necessarily the conditions at individual flames. Even for a single burner, bulk parameters (e.g., flue gas composition) can only afford a limited characterization of the flame. As it was pointed out in the introduction, a detailed description would require a vast amount of information in terms of the distributions in space (and maybe time) of many physicochemical variables. Imaging techniques are most valuable tools in this respect, since information on spatial patterns can be very helpful to describe, or even understand, important characteristics of a flame. A wide range of imaging methods is currently available. Active techniques (i.e., requiring external illumination, usually a laser) are used in research studies to determine the spatial distribution of species concentrations, temperature or velocities (see for example, Kohse-Höinghaus and colleagues [2]). Practical reasons still prevent, however, their use for the monitoring of industrial flames: lack of ruggedness, high cost, difficult optical access, and
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
so on. Much more feasible is the use of passive techniques (i.e., recording of radiation naturally emitted by the flame) and is the approach usually selected for the monitoring of practical flames. The radiation spontaneously emitted in flames includes different contributions (see for example, Gaydon [3]). Hot solid particles produce a continuous spectrum, in accordance with Plank’s law. The radiation emitted by gaseous species, on the contrary, occurs at discrete wavelengths and is due to different effects. Some chemical reactions produce radicals in excited states, a fraction of which lose their excess energy by emitting a photon, giving rise to chemiluminescence; for example, OH*, CH*, and C2* can pass to their fundamental state through the emission of light in specific regions of the spectrum, with peaks at 307, 431, and 516 nm. The formation rate of those excited radicals is closely related to the nature of the combustion reactions, and the light collected at those wavelengths has been related to local, instantaneous heat release rate or equivalence ratio (ER). The emission of thermally excited molecules is also a significant component of flame emission; hot H2O and CO2 molecules irradiate in the mid and far infrared and are the main contributors to heat transfer in blue flames. The radiation emitted is, therefore, closely related to the physicochemical properties of a flame. However, deriving physical parameters from a recorded image is not straightforward at all. The superposition of different emission mechanisms is a first difficulty, although this can be alleviated by bandfiltering. Since passive imaging is a line-of-sight technique, radiation from points with widely different conditions are combined at each pixel; however, deriving quantitative information is not obvious at all, especially for nonsymmetric and turbulent flames. Also, the laws relating the emitted energy with the variable of interest depend on a number of local conditions (optical properties, temperature, concentrations of major species and radicals, reaction rates), very difficult to know with the required accuracy. As a result of these difficulties, no standard method for the interpretation and/or utilization of flame images exists and a number of different approaches can be found in the literature. Pyrometry is thought to be the most feasible method to extract quantitative information from radiation spontaneously emitted from flames. Temperature distributions over large regions of the flame can be derived by applying the Plank law to every pixel of a CCD sensor. Twocolor techniques are usually applied, due to its reduced sensitivity to uncertainties in emissivity or attenuation. Infrared cameras are used in most cases, as radiation levels are highest in this region of the spectrum [4–7]. Two channels of RGB video cameras have been also used in some works [8]; although less accurate, this has the
Image-Based Techniques for the Monitoring of Flames
advantage of using cheaper, conventional CCD cameras. Tomographic reconstruction has also been applied to images captured simultaneously from several cameras in order to obtain 3-D temperature maps in sooting gas flames [7] and in coal-fired boilers [8]. Alternatively, chemiluminescence imaging can be applied to gain some insight into the location of the flame front or the spatial distribution of heat release rates. The wavelengths associated to the chemiluminescence of OH* and CH* are usually selected for this purpose. This requires collecting bandfiltered radiation in UV (at ~307 nm for OH*) or blue regions (~431 nm for CH*) as well as the use of intensified CCD cameras to detect the very small amounts of energy emitted in the bands selected. Some works [9–12] are oriented to the diagnostic/control of industrial flames based on the analysis of images of narrow band radiation at characteristic chemiluminescence wavelengths. For example, this kind of information can be used to estimate the stoichiometry of a flame. In fact, methods based on chemiluminescence of excited radicals are probably among the most mature technologies for online monitoring in practical premixed combustors (e.g., [13–15]). Nevertheless, even for this application, which has been analyzed in many works and has a sound physical basis, no universal relationships are available and the technique needs calibration for specific fuels and facilities [14–16]. The flame maps obtained from pyrometry or chemiluminescence imaging can provide a most valuable insight into the characteristics of the combustion process. However, it should be noted that those data are rarely meaningful by themselves and their physical interpretation or their use for monitoring or optimization purposes require the participation of experienced researchers or operators as well as some prior knowledge on the values (e.g., peak temperatures) or patterns associated to different combustion regimes and/or to optimal operation. Alternatively, flame images can be interpreted as a characteristic “signature” of a particular combustion state. This approach has been applied in several works, always aimed at developing novel monitoring techniques (and, ultimately, advanced control strategies) for practical applications. Whereas this approach can be applied to images informing on meaningful flame variables (e.g., chemiluminescent radiation, pyrometry-derived temperature maps, PLIF images, etc.), it is also compatible with nonfiltered images recorded with video cameras. An important advantage of this last option is that their hardware requirements are relatively little demanding: conventional CCD cameras and frame-grabbers are, in principle, enough to collect the data needed; most of the difficulties with the hardware are associated with installation in the combustion chamber and protection of instruments from high temperatures and fouling. The main challenges are thought to be related with data
© 2011 by Taylor and Francis Group, LLC
339
analysis and, above all, the conversion of raw images into usable information. The most common approach is feature extraction (i.e., the derivation of significant parameters from a flame image). Such a procedure greatly reduces the amount of information needed to represent an image. The features analyzed consist of geometrical parameters [12,17,18], different variables related to the level and spatial distribution of luminosity [19,20] or color [21–23], as well as combinations of several of those properties [6,24]. Characteristic frequencies have also been derived from high-speed image records, as a relevant feature of the process either by themselves or combined with other image properties [6,20,25]. The existence of relationships among combustion conditions and selected image features has been demonstrated by parametric studies reported in a number of works [6,18,19,24–26]. But, in most cases, those parameters do not have an intrinsic physical meaning, and serve mainly as indicators of a particular combustion state. Therefore, any attempt to exploit flame images for practical purposes involves as a necessary subsequent task that of relating those variables with known situations or with real combustion parameters. A wide range of approaches and processing methods have been reported. Kurihara et al. [17] proposed an empirical index for the prediction of NOx emissions as a function of some selected geometrical parameters. Shimoda et al. [4] applied a highly simplified physical model to relate unburnt carbon in a coal-fired plant with luminosity features derived from flame images collected at two wavelengths. Baek et al. [21] found a linear relationship between a chromatic parameter calculated from flame images and the emissions of NOx and unburnt carbon in a pilot coal-fired furnace. Rahman, Gibbins, and Forrest [20] proposed two quality indices, combining parameters related to luminosity and fluctuation frequency derived from flame images, as performance indicators for utility boilers. Yu and MacGregor [23] used partial least squares to predict NOx and SO2 emissions and heat losses in an industrial boiler from a set of image features (related to its geometry, luminosity, and color). Artificial neural networks (ANN) have been used as fitting tools to relate image features with relevant combustion parameters. Lu et al. [12] used a feed-forward ANN to predict successfully the combustion regime from five geometrical/luminous parameters of the flame. The ANN and fuzzy logic were combined in [10,27] to adjust air and fuel flow rates, based on flame brightness and length. Wang et al. [28] applied a back-propagation ANN to relate a set of image parameters (average and deviation of temperature, ignition distance) with NOx emissions. Bae et al. [29] used a neural network to distinguish flame on/off conditions from the analysis of luminosity distributions.
340
A different, relatively little explored approach consists in treating the image as a data set that can be processed as a whole in order to classify or identify an image as representative of a particular combustion state. Allen et al. [11] reported notably good results using artificial neural networks to classify different combustion states using the full image as input data. Finally, the use of flame images for control or optimization purposes has been examined in a few works (all in laboratory rigs). Allen et al. [11] and Lu et al. [12] coupled an ANN-based state predictor and a closedloop control to bring the system to the desired state. Burkhardt et al. [10,27] combined ANN and a fuzzy controller to reach optimal conditions. Soot formation was avoided by Tuntrakoon and Kuntanapreeda [22] using a fuzzy controller whose inputs were the levels of blue and orange in flame images. It should be noted that all control applications mentioned require a state identification stage, before a control action can be defined.
15.3 Image Processing for Flame Identification: General Approach and Testing Procedure The two methods described in the following sections were designed to identify combustion states using the flame image as a characteristic signature. Therefore, it is implicitly assumed that, due to the huge amount of information contained in an image (intensity at each pixel), the probability that different combustion states produce the same set of data is very low. As a fundamental advantage, this type of method does not require interpreting the highly complex physicochemical phenomena involved. At the same time, its empirical nature can also be an important weakness. Whether this is a serious problem or not will largely depend on how the empirical knowledge is developed and applied. For a method to become a feasible monitoring technique, it should be of a generalizable nature in the sense that implementation in new applications does not require a costly design/adaptation procedure and includes as few as possible ad-hoc models or parameters. The processing techniques presented here make use of algorithms conceived to classify a particular data set with respect to patterns previously characterized. Instead of selecting a number of image features, all pixels forming the image were used as input data; this requires processing a large amount of data but avoids the significant information loss inherent to feature-based methods. One of the methods is based on the use of SOFM, capable of automatically classifying complex sets of data according to their degree of similarity. The second method is an
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
adaptation of a speech-recognition algorithm, originally designed to identify the speaker by analyzing sound records and comparing them with those contained in a database. In both cases, the objective is to identify an unknown flame among a set of regimes representative of the operating range of a particular burner. Besides its interesting applications for monitoring and control purposes, this approach can be useful, for example, for the detection of malfunctions or for the estimation of the conditions of individual flames in multiburner chambers (e.g., air-to-fuel ratio, pollutant emissions). It is important to remark that the aim was not to develop and fine-tune a method specifically adapted to a particular situation, which would be of limited interest, but to assess the potential of the approach for a relatively general case. In order to approach this objective (1) all results presented were obtained by applying the methods as initially conceived, without any ad-hoc work (apart from some details regarding preprocessing of raw images), (2) the methods were applied to different burners, again with no ad-hoc adaptation (for brevity, only a part of the results are described in detail), and (3) the use of knowledge or rules specific to the particular burners used was avoided. The methods were thoroughly tested in two different facilities. One of them is a water-cooled experimental furnace, equipped with an industry-type natural gas diffusion burner (Figure 15.1). The burner includes a number of adjustable settings, enabling the study of a wide range of flame types. Combustion air enters through two concentric air injections, each with a variable swirl generator (moveable blocks). A broad range of conditions, from conventional, fast-mixing flames to lowNOx, air-staged combustion, can be achieved by means of a proper adjustment of air distribution and the two swirl settings. Further details on this facility and a detailed characterization of the different combustion regimes can be found in reference [30]. Three of the burner settings were varied in the experiments: air distribution between primary and secondary ducts (air ratio, AR, expressed as fraction of secondary air over the total air flow rate) and the swirl numbers of both air streams (S1 and S2, respectively; both reported here as a percentage with respect to maximum swirl level). Detailed in-flame measurements revealed significant changes in the distribution of species and temperatures inside the flame when AR, S1, or S2 were varied [30]. Flame images were recorded for 49 different combinations, including combustion regimes ranging from fast-mixing flames (AR ~ 50%, [NOx] up to 44 ppm) to deep air-staging conditions (AR ~ 100%, [NOx] down to 15 ppm), including highly stable, unstable and oscillating flames. The mass-flow rate of natural gas was fixed to 5 Nm3/h throughout the tests and the ER was kept at 0.88 (3% O2 in flue gases, by vol. db).
341
Image-Based Techniques for the Monitoring of Flames
Natural gas Primary air
Moveable-block swirlers
Secondary air Radiation sensors
CCD camera
Si photodiode
Microphone M1
800
CCD Camera Microphone M2
P transducer
400
ADC Fuel
250
Combustion chamber
810
Swirl vanes
Exhaust duct
Figure 15.1 Experimental furnace with air-staged burner and instrumentation. (From Hernandez, R. and Ballester, J., Combustion and Flame, 155, 509–28, 2008. With permission from Elsevier.)
The second rig was the premixed combustor shown in Figure 15.2. A quartz tube (L = 250 mm, i.d. = 120 mm) configures the combustion chamber and allows for unrestricted optical access to the flame. The fuel–air mixture is injected as a swirling annular jet (o.d. = 40 mm, i.d. = 25 mm, 60° axial vanes). The fuel issues through 12 radial orifices, located 320 mm upstream of the chamber. All tests reported had a thermal input of 32 kW. In this case, the ER was varied in the range from 0.57 to 1.0 by regulating the air flow rate at a constant fuel feeding. This led to widely different results. High ER values were characterized by high NOx emissions (up to 40 ppm), highly stable attached flames and low CO levels. As the mixture became leaner, NOx emissions were dramatically reduced (down to 5 ppm) and, when it approached the lean blow-off limit (ER < 0.65), CO emissions increased suddenly and detached, pulsating flames were obtained. A more detailed description can be found in Ballester et al. [31].
© 2011 by Taylor and Francis Group, LLC
Air
Figure 15.2 Premixed combustor.
Therefore, the test cases included a notably broad variety of flames (diffusion/premixed, fast-/delayedmixing, attached/detached, stable/pulsating), useful to test the processing methods in a wide range of situations. It should be noted that exploiting the knowledge gained from detailed measurements for flame monitoring from images was avoided on purpose; whereas this might add some physics and even lead to improved performance, the procedures so developed would lose much of their generality. Hence, the monitoring methods presented here do not use any qualitative or quantitative piece of information not contained in the very flame images. Flame images were captured using a CCD camera (Panasonic WV-CL350) housed inside a water-cooled jacket, installed in the furnace wall, and oriented perpendicularly to the burner axis, so that a full picture of the flame can be obtained. The use of a pinhole lens together with a purge of compressed air maintained the optics clean and cool. Monochrome images (8-bit grayscale, 320 × 240 pixels, exposure time 1/60 s) were recorded in a process computer through a frame-grabber board (Data Translation DT3155). Since the objective is the exploration of feasible methods for the monitoring of practical flames, an essential feature of this study is the use of conventional video cameras. This imposes some limitations on the information that may be recorded. The radiation collected by the CCD corresponds basically to the visible range; for blue flames, this includes chemiluminescent
342
emission from different radicals: discrete peaks for CH*, C2*, broadband radiation from CO2*, and other minor contributions [3]. The camera performs a line-of-sight integration, so that each pixel accumulates contributions from points at different distances. As explained below, representative images were obtained by averaging several frames; hence, fluctuations due to turbulence or (much more important in some cases) flame oscillations are smoothed out. Therefore, although there must be some connection between the images recorded and the physics of the flame, the lack of resolution in wavelength, space, and time makes the derivation of some kind of combustion variable from them very difficult. Temporal or spatial averaging can be a serious drawback for monitoring methods based on physical parameters (e.g., derived by pyrometry or chemiluminescence), but they are not necessarily when the image is treated as a signature. By no means does this imply that low resolution is a positive feature. On the contrary, a better resolution in wavelength, space, or time might produce images more representative of the process; but this would require much more expensive and delicate instrumentation (e.g., intensified CCD, filters, several cameras for 3-D reconstruction, etc.). Furthermore, whereas a wide range of conventional, inexpensive CCD cameras are currently available that might provide images with better definition, the good results reported with monochrome, 340 × 240, 8-bit images serve to demonstrate the low requirements in this respect of the methods studied. Similarly to other works, the option of using selected image features for flame identification was explored. Flame images were processed to extract a number of geometrical and luminous magnitudes [32]. Significant variations in those parameters (e.g., flame length, area, location of centers of mass, etc.) were observed among the different flames. However, the relationships were clearly nonlinear in most cases and correlations among image features and operating conditions by multiple linear regression yielded poor results. Therefore, other types of functional forms, specifically adapted to each feature, should be used to represent adequately the relationships observed. This is thought to be a significant shortcoming if the objective is to develop generalizable diagnostic methods. Some trials were also performed using artificial neural networks as the fitting algorithm, due to their ability to handle multivariate, nonlinear relationships. Although this method provided reasonable results with some combinations of image features, the capabilities for state identification were poorer than with the approaches explained in the following sections. The cause is thought to be the information loss associated to the replacement of the full image by a few representative parameters.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
15.4 Flame Identification with SelfOrganizing Feature Maps Artificial neural networks (ANN) are mathematical algorithms inspired on the structure of biological neural systems and the way information is processed by the brain, with good capabilities to handle complex information with unknown and highly nonlinear functional relationships among the different variables. Haykin [33] describes fundamental and application aspects of these methods. Images are an outstanding case of complex data sets, whose processing and/or interpretation has motivated great efforts during the last decades. As described by Egmont-Petersen, de Ridder, and Handels [34], ANNbased algorithms can be very effective for most of the tasks usually faced in image processing. Object recognition based on features is one of them; in fact, ANNs have been successfully used in a number of works specifically oriented to identify combustion states (or, equivalently, to estimate performance parameters like pollutant or unburnt emissions) using as inputs some selected features previously extracted [10,12,28,29]. This procedure was also tested in this work, although its performance was not as good as with the methods reported. Alternatively, all pixels forming an image can be directly used as input data (i.e., the number of representative features equals the number of pixels). This requires processing a much larger amount of data, but avoids information loss due to data reduction. The only experience found in the open literature is the work by Allen et al. [11] where combustion states were successfully identified using a back-propagation ANN fed with the intensity levels in 1024 pixels. As noted by Egmont-Petersen, de Ridder, and Handels [34], SOFM are suitable for object recognition based on pixel data, and was the method selected in this work. The so-called Kohonen maps are used here, which have been widely applied for classification purposes, especially in pattern recognition [35,36]. A SOFM consists of neurons organized on a regular low-dimensional grid. The neurons are connected to adjacent ones by a neighborhood relation dictated by the topology of the map. Each neuron in the map contains a weight vector with the same dimension than the input vectors and represents a pattern flame image. During training the input vectors are compared with that of each neuron by calculating a distance (Euclidean in this case) and the weight vectors of the closest neuron (winner neuron) and its neighbors are modified according to the so-called Kohonen learning rule. This process is iterated so as to gradually reduce the distances between the input vectors and the winner neurons as well as those among neighbor neurons; the
343
Image-Based Techniques for the Monitoring of Flames
final result is a set of pattern images, not identical to but representative of the range of data used for training that are ordered in the map according to their visual features. The interested reader can find more detailed descriptions in the references quoted. Once the SOFM has been trained, an unknown flame representing an unknown state is classified by the winner neuron when it is shown to the SOFM; identification with respect to known states consists in selecting, among the flames included in the training database, the case yielding the same winner neuron. The following sections describe in detail the method and its application to air-staged diffusion flames (rig shown in Figure 15.1). A final subsection summarizes the main results obtained with other burners. 15.4.1 Image Preprocessing Different values of the three burner settings (S1, S2, AR) were combined to yield a total of 49 different combustion states. Sequences of 100 images (grayscale, 320 × 240 pixels) were captured for each of the tests. Firstly, the images were preprocessed in order to reduce the size of the input vector as much as possible without losing information as well as to strengthen some relevant characteristics. The steps followed were:
1. Average: All images recorded for a single test were averaged. This step is not only very useful for data reduction (99%) but also to filter out fluctuations among different frames, resulting in a more representative image of the flame. This was ensured by averaging over 100 images, although most probably a much lower number would suffice. 2. Contrast: Luminosity values were normalized, so that the darkest pixel of the image was set to 255 (white in grayscale) and the brightest to 0 (black). This operation shouldn’t be interpreted as an information loss because absolute luminosity depends on many factors (CCD c alibration, lens, attenuation along the optical path, etc.) and its consideration could make the procedure highly dependent on the particular setup. 3. Segmentation (thresholding): The flame area was delimited by setting to 0 all pixels with intensity <60. 4. Trimming: After completing the previous operations, large areas of the image consisted of black pixels. In order to discard useless data, the flame images were trimmed by a rectangle defined as the disjunction of the smallest rectangles enclosing nonblack parts of the individual images. Such rectangular area was defined
© 2011 by Taylor and Francis Group, LLC
at the training stage and used afterward to trim any new image. As a result, images were reduced from 320 × 240 to 170 × 200 pixels. 15.4.2 Configuration and Training of the SOFM Initially, the topology and number of cells of the SOFM need to be defined. For this application, a 3 by 10 map might be suitable to represent the matrix of test cases: three adjustable burner settings and a maximum of 10 different values for each of them. However, a larger map was preferred to favor a clear distinction among different groups of flames, and a 10 by 10 map with hexagonal topology was finally used. As a first step, the ANN was trained with images captured in 49 different combustion regimes. The SOFM was trained for 1000 epochs. In each of these epochs, the distance between a particular flame image and every neuron in the SOFM (defined by a set of weights) was calculated. The neuron yielding the shortest distance was marked as the winner and the weights of all neurons in the neighborhood of the winner were updated using the Kohonen rule. As a result of the training process, each of the 100 neurons ends with an associated pattern flame image. 15.4.3 Topological Classification Figure 15.3 displays the pattern images associated to the different neurons in a 10 by 10 map. This representation also serves to illustrate the way in which the SOFM operates: images are grouped according to their degree of similitude. It should be noted that although the neurons are represented equidistant in the network, they actually are not. The white lines have been drawn on the map to indicate the longest distances between neighbor neurons. Each subgroup so obtained includes images with evident similarities among them and is characterized by patterns that are appreciably different from those of other groups. A basic assumption behind the use of images for the diagnostic of combustion states is that geometrical/ luminous properties maintain some relationship with operating conditions or system performance. The type of classification provided by Kohonen maps can be particularly useful to evaluate this hypothesis. With this purpose, numerical values were assigned to each of the 100 neurons, according to the following procedure. The parameters (burner settings, emissions) characterizing each of the 49 combustion tests were transferred to the corresponding winner neuron. An interpolation procedure was applied to assign numerical values to those neurons without an associated real combustion regime, using the two coordinates, V and H, defining its location in the map (Figure 15.3) as independent variables.
344
Industrial Combustion Testing
V
H
Figure 15.4 Swirl level in primary stream, S1, associated to the neurons of the SOFM (grayscale: white for S1 = 100%, and black for 0%). (From Hernandez, R. and Ballester, J., Combustion and Flame, 155, 509–28, 2008. With permission from Elsevier.)
Figure 15.3 Pattern flame images associated to the 100 neurons of the SOFM. (From Hernandez, R. and Ballester, J., Combustion and Flame, 155, 509–28, 2008. With permission from Elsevier.)
Figures 15.4 and 15.5 display the results obtained in terms of the swirl levels in the primary air stream and NOx emissions, respectively. The following example describes how this assignation procedure was implemented. The location of the neurons in Figures 15.3 through 15.4, and 15.5 will be described with the coordinates V and H (both in the range 1 to 10). Let’s take the image at V = 2, H = 4 in Figure 15.3; like the other images shown in Figure 15.3, it does not correspond to a real flame but was created internally by the SOFM algorithm from the set of 49 images used for training. In a subsequent stage, the SOFM was used to assign a neuron to each of the 49 flames. Table 15.1 displays the results: some of the neurons were associated to a real flame image (or, in a few cases, to several, like neuron V = 1, H = 1), but others were not (e.g., V = 2, H = 4). Those with an associated real flame were assigned its measured NOx emission. For the other neurons, a NOx value was calculated by linear interpolation from actual measured emissions assigned to neighbor neurons, using the coordinates V and H as independent variables. As a result, a NOx concentration of 37.6 ppm was calculated for the neuron at V = 2, H = 4; this method enabled assigning a NOx value to all 100 neurons as shown in Figure 15.5. An equivalent procedure was applied to build Figure 15.4 for S1, where it can be seen that the different degrees of swirl in the primary air become
© 2011 by Taylor and Francis Group, LLC
Figure 15.5 NOx emissions associated to each of the neurons of the SOFM (grayscale: white indicates the highest value, 44 ppm, and black the lowest, 14 ppm). (From Hernandez, R. and Ballester, J., Combustion and Flame, 155, 509–28, 2008. With permission from Elsevier.)
automatically grouped in the SOFM. Since other burner settings (S2 and AR) were also varied, flames with the same S1 can display widely different aspects, which explains that similar S1 values can be found in different regions of Figure 15.4. This automatic grouping of cases with similar burner settings is, obviously, due to the fact that those precise settings determine the aerodynamics
345
Image-Based Techniques for the Monitoring of Flames
Table 15.1 Measured NOx Emission for the Neurons in the Bottom-Left Zone of Figure 15.3 With an Associated Real Flame Neuron Coordinates V
H
Measured NOx (ppm)
1 1 1 1 1 2 3 3 4 4
1 1 1 3 4 5 1 2 3 5
37.6 37.9 38.2 35.8 34.4 33.4 41.5 35.8 22.7 22.8
Source: From Hernandez, R. and Ballester, J., Combustion and Flame, 155, 509–28, 2008. With permission from Elsevier.
Figure 15.6 Average flame image recorded with the burner adjusted at AR = 50%, S1 = 100%, S2 = 50%; measured NOx emission was 37.7 ppm. (From Hernandez, R. and Ballester, J., Combustion and Flame, 155, 509–28, 2008. With permission from Elsevier.)
and mixing pattern in the flame and therefore its visual appearance.
50 45
An alternative viewpoint is to evaluate whether images and NOx emissions (two different results of the combustion process) display some correlation. This can be observed in Figure 15.5, where the NOx emissions assigned to the 100 neurons of the SOFM are displayed. The different emission levels are noticeably grouped over the map, with gradual transitions between one neuron and its neighbors. Since this distribution is directly obtained from the distribution generated by the Kohonen map according to image properties, the close relationship among both aspects becomes apparent. Such strong correlation might be exploited to infer the NOx emission of a particular flame using the image as the only input. With that purpose, a set of 42 new image records (different from those used for training) were classified using the SOFM developed; that is, each of them was assigned to one of the neurons in the 10 by 10 map. The NOx emission estimated for that flame was the value previously assigned to its winner neuron, as shown in Figure 15.5. For example, the flame image shown in Figure 15.6 was recorded for AR = 50%, S1 = 100%, S2 = 50%, and the measured NOx emission was 37.7 ppm. Classification with SOFM yielded as winner the neuron at V = 2, H = 4. As explained above, no measured NOx was available for this neuron, and the value assigned by interpolation was 37.6 ppm. Therefore this was the value of NOx predicted from the image in this case, very close to the actual value of 37.7 ppm.
40
© 2011 by Taylor and Francis Group, LLC
NOx estimation (ppm)
15.4.4 Estimation of NOx Emissions Using Flame Images
35 30 25 20 15 15
20
40 25 30 35 NOx experimental (ppm)
45
50
Figure 15.7 Comparison between predicted and measured NOx emissions. Predictions based on classification with SOFM. (From Hernandez, R. and Ballester, J., Combustion and Flame, 155, 509–28, 2008. With permission from Elsevier.)
Actual NOx emissions and estimations with this method are compared in Figure 15.7. A fairly good agreement is obtained, with deviations <3 ppm in most cases; in particular, high- and low-NOx regimes are perfectly distinguished. This is thought to be a most relevant result for practical applications, as the good correlation obtained
346
among images and NOx emissions might enable the development of advanced diagnostic methods by using flame images to estimate NOx emissions (i.e., a virtual NOx sensor). This could open new possibilities for the monitoring of emissions from individual flames in multiburner furnaces, or to greatly reduce the response time of gas analyzers down to the levels acceptable for control purposes. Alternatively, the Kohonen map can just be used to identify the combustion state with respect to others previously known (used at the training stage). The process of assigning a winner neuron to a new input image yields a particular location in the map that can coincide with a particular combustion regime, or can be used to find the most similar regimes as those located at short distances. The comparison between estimated and actual NOx emissions serves also as an indication of the good accuracy of the SOFM as a flame identification tool. 15.4.5 Application to Other Combustion Situations The potential of monitoring methods should not be evaluated from the results in a particular application but strongly depends on their capabilities to handle different situations. Moreover, ideally, they should be applicable without needing a costly process of ad-hoc design and adjustment and should avoid the use of specific parameters and interpretation criteria as much as possible, which can be adequate for one particular application but not for many others. Due to the nature of the methods reported here, it is not possible to demonstrate their generality on physical grounds; although there is a relationship among visual and physical characteristics of the flames, it involves many different physicochemical processes that are not well understood yet and whose role strongly depends on the particular system. However, some insight in this respect can be gained from experiences with other types of flames. A preliminary version of the SOFM method was also applied to a commercial oil-fired burner with a nominal capacity of 60 kW. The flames in this case were yellow and highly radiating, with characteristics (physical and visual) very different from those of the blue flames described. In particular, the nature and origin of emission spectra are expected to be deeply different: dominated by blackbody radiation from soot particles, with a much smaller contribution due to chemiluminescence of excited radicals. An exercise similar to that reported in Section 15.4.4 was performed to estimate NOx concentration from flame images. The NOx emissions varied in the range of 53 to 94 ppm for the flames analyzed; the estimation error was within ±5 ppm in practically all cases, very similar in relative terms to the results shown in Figure 15.7.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
This identification procedure was also evaluated for a set of combustion regimes studied in the premixed burner shown in Figure 15.2. In this case, modifications in equivalence ratio (ER = 0.57-1) led to widely different flames, in terms of their pollutant emissions and stability as well as visual appearance. Although other map topologies (7 by 7 or 12 by 3 neurons) provided similar (or even better) results, a one-dimensional SOFM with 18 by 1 neurons will be analyzed here, as this might seem a more intuitive configuration for tests with only one variable setting. The method was applied exactly as described above, initially developed for the diffusion flame, without any adaptation specific to this facility. The visual and physical characteristics of the flames were, obviously, very different from those of the airstaged burner. Figure 15.8 shows the set of 18 flame images created by training the SOFM with real images captured for 16 different values of ER. Those 18 images are supposed to represent the range of flame geometries characterizing the combustion regimes studied. Figure 15.8 reveals evident similarities between neighbor neurons, with shapes changing gradually between both ends of the map; this suggests that operating conditions (i.e., ER) also vary monotonically along the map. This was confirmed by applying the interpolation procedure described in previous sections. Figure 15.9 shows the ER values associated to each of the 18 neurons, indicating that the extreme left and right images correspond, respectively, to the richest and leaner flames, with values varying gradually between both limits. Figure 15.9 also includes interpolated values for NOx and CO emissions, 1
2
3
4
5
6
7
8
9
...
...
10
11
12
13
14
15
16
17
18
Figure 15.8 Pattern images associated to the 18 by 1 SOFM, trained with premixed flame images recorded at different ER values. ER
NOx
CO
Figure 15.9 ER, NOx, and CO associated to the neurons in the 18 by 1 map. White denotes peak values (ER = 1.0, Nox = 40.8 ppm, CO = 3537.8 ppm) and black the lowest (ER = 0.5, Nox = 1.8 ppm, CO = 34.2 ppm).
347
Image-Based Techniques for the Monitoring of Flames
fully consistent with the corresponding ER values (NOx decreases monotonically with ER, and CO displays a minimum at intermediate values of ER and increases gradually toward richer mixtures and more steeply as the flame approaches the lean blow-off limit). The accuracy of this identification procedure was assessed by comparing measured values with those estimated by applying the interpolation procedure described previously. As shown in Figure 15.10, a good accuracy was achieved in practically all cases. Signifi cant discrepancies are only observed for tests with high CO levels (where even experimental repeatability is relatively low). The SOFM developed in this exercise, with only one variable parameter (ER), might be interpreted as an empirical correlation between ER and flame images. It has, therefore, some similarity with feature-based methods, where numerical values of selected image characteristics are correlated with operating conditions or flue gas properties. However, both approaches differ in some fundamental aspects. First, the use of the full image avoids information losses; at the same time, this makes the method independent of the success in selecting the most adequate image features for a particular application. Second, multivariable and/or nonlinear dependences are automatically captured by the SOFM, which avoids the usually costly task of finding suitable variable combinations and functional forms. As a result, 1
Estimated values
0.8
0.6
a third and very important difference should be noted: this is a relatively general procedure, conceived to be applied without any specific adaptation (except, perhaps, aspects related to initial image conditioning) to a new combustion situation.
15.5 A Pattern-Recognition Technique for Flame Identification The result of most identification methods (e.g., SOFM) is the state that most probably corresponds to the unknown state being processed among those previously known. A shortcoming is that the reliability of this result is not qualified. It would be most desirable, for example, to obtain some information about the degree of coincidence with respect to the state assigned or, if the unknown flame displays some similarity with several states, the relative probabilities associated to each of them. This would enable, among others, the detection of off-design operations, evaluating how reliable the identification is or combining the result with other sources of information. A different approach has been applied for the classification of flame images that provides information on the probability of coincidence with each of the combustion states previously known. This procedure is inspired on the cepstral analysis techniques, commonly used for speech recognition [37]. Although sound records are of a type different from image data, both can be equally transformed into covariance matrices, as required in this kind of method. As in the previous section, the results with the airstaged burner are used to explain the method in detail, whereas its application to the premixed combustor is summarized in the last subsection. 15.5.1 Image Preprocessing
0.4
0.2
0
0
0.2
0.4 0.6 Measured values ER
NOx
0.8
1
CO
Figure 15.10 Comparison between predictions and measurements. NOx and CO emissions have been normalized by characteristic maximum values (50 and 3000 ppm, respectively).
© 2011 by Taylor and Francis Group, LLC
Ten monochrome images (320 × 240 pixels) were recorded for each of the combustion regimes studied with the airstaged burner (Figure 15.1). Each of them is obtained by averaging 10 individual frames in the sequence captured. As a first step, Principal Component Analysis (PCA) was applied for data reduction. Given the large amount of data needed to represent the images, minimizing the size of the input vectors entails significant benefits, especially when real-time performance is required. The PCA is a convenient method for image processing, because it allows reducing the dimension of the input space while maintaining much of the original structure of the input images. It has been used extensively in pattern recognition [38] and has shown to be an effective method for the analysis of flame images [23,26].
348
Industrial Combustion Testing
Images are represented by a matrix I of dimension 320 × 240. Previous to PCA, I is transformed by rearranging image blocks into columns. First, I is divided into p square blocks (if necessary, padded with zeros) of s*s pixels; second, a new matrix X with s2 rows and p columns is formed by filling each column with the values in one of the blocks. Finally, PCA is applied to derive a lower-dimensional representation. Given a matrix X, and the desired new dimension, q, the PCA algorithm yields the matrix M (q rows and s2 columns) and the vector B (q components) needed to obtain a matrix Y (q rows and p columns), according to: s2
Yij =
∑M X ik
kj
+ Bi i = 1… p.
(15.1)
k =1
N2
P ( xi ~ y j ) =
∑ P(x
i ,n
s
( Z _ COV )ij = ∑ Zik Zt kj .
(15.2)
k =1
A library of covariance matrices was created by calculating Z_COV for nine of the images stored for each of the N combustion regimes tested; therefore, each combustion state is characterized by N1 = 9 different data sets. The degree of similarity among different images was compared by calculating the arithmetic harmonic sphericity (AHS) distance between their respective covariance matrices. The AHS distance between two generic matrices A and B, of dimension M, is defined as: D( A, B) = log 10 (Tr ( A × B−1 ) × Tr ( B × A −1 )) − 2 × log 10 m. (15.3) The important data reduction rate achieved should be noted, since only q2 = 144 data contained in Z_COV are finally used to evaluate an image, instead of 76,800 values in the original matrix I. 15.5.2 Probability Calculation and Pattern Selection Classification is performed by calculating the probability of an unknown test flame (xi) to be like each of the N pattern flames in the library (yj), P(xi ~ yj). In general, each test flame is characterized by different test images, xi,n
© 2011 by Taylor and Francis Group, LLC
~ yj )
n= 1
.
N2
(15.4)
P(xi,n ~ yj) is calculated as the probability for a particular test image, xi,n, of test flame xi to have a null AHS distance from the matrices characterizing a pattern flame yj. Since each pattern flame is defined by N1 pattern images yj,m, N1 different distances can be calculated. The average of the AHS distances between the test image (xi,n) and each of the N1 pattern images was taken as the representative value: N1
Y, M, and B can be taken, therefore, as an alternative representation of the original matrix X. A parametric study on the influence of s (in the range of 5 to 18) and q (from 5 to 20) yielded as the optimal combination s = 14, q = 12. Finally, Y, M, and B were put together in a new matrix Z, with q rows and p + s2 + 1 columns. Cepstral analysis techniques were applied on the matrices Z. The corresponding covariance matrix, Z_COV, is calculated as:
with n = 1 … N2, so the probability is averaged over all of them:
D ( xi , n , y j ) =
∑ D( x
i ,n
m= 1
N1
, y j ,m ) .
(15.5)
So, the set of N1 pattern images is treated as an ensemble of data sets providing a statistical representation of a specific combustion regime; this is a relevant feature of the method and provides the robustness required to handle noisy data (associated, for example, to turbulent fluctuations in industrial flames). This is represented statistically by a Gaussian probability distribution, f yj(d), for the AHS distance: different occurrences of the same combustion regime are expected to provide an ensemble of AHS distances with ~0 mean and a variance associated to this particular regime. The width of the distribution is estimated from the standard deviation, σyj, of the N1*(N1-1)/2 distances calculated among pairs of the N1 pattern images available. The probability of a test image, xi,n, to belong to a particular pattern, yj, is estimated as f yj(d) for d = D(xi,n, yj). This is illustrated in Figure 15.11 for two hypothetical patterns y1 and y2; if the functions f y1(d) and f y2(d) (with respective widths σy1 and σy2) are centered, respectively, at D(xi,n, y1) and D(xi,n, y2), the probability sought is directly obtained as the intercept with the ordinates axis. This example clearly shows that not only the mean value must be considered, but also the width of the distribution. Even though the average AHS distance is shorter for pattern y1, the most probable combustion regime is y2. Pattern y1 is represented by a very narrow distribution, and the probability decays rapidly even at short AHS distances; on the contrary, pattern y2 is characterized by a wider range of covariance matrices, and the probability of actually representing an unknown state maintains significant values up to relatively high AHS distances. Figure 15.12 displays the classification tests performed for two of the flames. The test case on the left (#5)
349
Image-Based Techniques for the Monitoring of Flames
fy1(d)
6000 5000 4000 3000 2000
fy2(d)
P(xn~y2)
1000
P(xn~y1) D(xn, y1)
0
D(xn, y2)
0
1
2
3 4 5 AHS distance - # 5
6
7
8 × 10–3
0
1
2
3 4 5 AHS distance - # 30
6
7
8 × 10–3
AHS distance Figure 15.11 Evaluation of the probability that a test flame xn corresponds with pattern flames y1 and y2. (From Hernandez, R. and Ballester, J., Combustion and Flame, 155, 509–28, 2008. With permission from Elsevier.)
corresponds to the image shown in Figure 15.6, already mentioned in a previous section. The intercept of the Gaussians to the reference patterns, fyj(0), yields a measure of the probability of the test image to belong to that particular combustion regime. In some cases (Figure 15.12), several of the reference cases give a nonnegligible probability. To facilitate the interpretation, the results were normalized so that the sum for all patterns is 100%:
P ( xi , n ~ y j ) =
fy , j ( 0) N
∑f
y ,m
( 0)
5000 4000 3000 2000 1000 0
.
(15.6)
m= 1
For example, test image #5 yielded nonzero probabilities for three of the flames (see Figure 15.12 and Table 15.2). Since the highest probability was obtained for the correct pattern, the identification was successful in this case. It should also be noted that flames #6 and #7 were also very similar, because variations in swirl number of secondary air in the range of 0 to 50% had little influence on the aerodynamics of the flame for S1 = 100%, as evidenced by the very similar NOx emissions measured for the three flames. 15.5.3 Classification Tests This identification method was applied to the same N = 49 combustion regimes considered in the previous section. The N1 = 9 images were used to generate a set of covariance matrices for each of those pattern flames; an additional image (N2 = 1), recorded in the same operating
© 2011 by Taylor and Francis Group, LLC
6000
Figure 15.12 Classification tests for two of the test flames (arbitrary units). (From Hernandez, R. and Ballester, J., Combustion and Flame, 155, 509–28, 2008. With permission from Elsevier.)
Table 15.2 Results of the Classification Procedure for Test Image #5 Pattern Flame #5 #6 #7
AR (%)
S1 (%)
S2 (%)
NOx (ppm)
Probability
50 50 50
100 100 100
50 25 0
37.6 37.9 38.2
49.9% 25.2% 24.9%
Source: From Hernandez, R. and Ballester, J., Combustion and Flame, 155, 509–28, 2008. With permission from Elsevier.
conditions, was stored for each of the 49 cases, and then used as test images to evaluate the performance of the method. Figure 15.13 displays the calculated probabilities, P(xi ~ yj), that a particular test image, xi, corresponds to pattern yj. A fully successful identification would be obtained if the probability is 0% if i ≠ j and 100% when
350
Industrial Combustion Testing
50 100
45
90 80
NOx estimation (ppm)
40
) Probability (%
70 60
35
50 40
30
30 20
C43 C36 C29 C22 C15
11
16 21 Patt ern 26 31 flam es 36 41
fla
6
C8 46
C1
Figure 15.13 P(xi ~ yj) estimated by image identification for the 49 flames. (From Hernandez, R. and Ballester, J., Combustion and Flame, 155, 509–28, 2008. With permission from Elsevier.)
i = j. Figure 15.13 can be interpreted in two complementary ways: • The outcome of the identification process is that test image xi corresponds to the combustion regime, yj, for which P(xi ~ yj) is maximum. The results shown in Figure 15.13 indicate that identification has been successful in 32 out of 49 cases. Nevertheless, it should be noted that all wrong identifications occurred among very similar flames, in most cases with probabilities for the winner pattern that are only slightly higher than those obtained for the correct one. This is an important advantage of this method, as the output not only includes the most probable pattern but also informs about similarities with other flames. • The absolute value of the probability obtained for the winner pattern can be taken as an indication of the reliability of this result. Low probabilities can be due, as shown in Figure 15.13, to the similarity between some reference cases. Otherwise, a low value could indicate, for example, that the unknown state is out of the range represented by the pattern flames, or that some deviation has occurred in the combustion equipment. Similarly to the exercise reported with the SOFM, this identification tool might be used to estimate
© 2011 by Taylor and Francis Group, LLC
20
st
1
25
Te
0
me s
10
15
15
20
25 30 35 40 NOx experimental (ppm)
45
50
Figure 15.14 Comparison between predicted and measured NOx emissions. (From Hernandez, R. and Ballester, J., Combustion and Flame, 155, 509–28, 2008. With permission from Elsevier.)
some relevant parameters of the combustion process. Figure 15.14 displays the NOx estimations obtained by assigning the value measured for the corresponding winner patterns to each of the test cases. The coincidence is almost perfect in most cases; even in those with wrong identifications, the deviation is very small, further confirming the similarities between the flames with wrong or ambiguous identifications. Significant differences are only observed in the regimes with lowest NOx emissions, corresponding to the highest air-staging ratios (AR = 90–100%). As previously noted, those flames displayed thermoacoustic oscillations, which can result in wider variability among flame images and, hence, in less certain identifications. As already noted, an advantage of this approach is that the result obtained is not limited to a yes/no answer, but also quantifies the degree of similarity with respect to all the patterns available. This can be further exploited, for example, to combine this type of diagnostic with other potential sources of information. This possibility has been explored in Hernández and Ballester [32], where NOx emission was used as an additional representative feature of the different combustion states. The combination of probabilities calculated separately from the images (using the cepstral method) and from the NOx emissions (comparing the value measured for the test image with those stored in the flame library) increased the success rate and the confidence level of the predictions, as NOx values helped to disambiguate
351
Image-Based Techniques for the Monitoring of Flames
some subgroups including flames with similar visual appearance. 15.5.4 Application to Premixed Flames In order to evaluate this procedure in a different situation, an identification exercise was performed on the set of premixed flames already mentioned in Section 15.4.5. In this case, a total of N = 18 combustion regimes were studied (ER = 0.57–1) and the probabilities were evaluated with N1 = N2 = 10 images. Figure 15.15 compares measured values of ER, NOx, and CO emissions with predictions based exclusively on flame images. As it happened with the SOFM method (Section 15.4.5), significant uncertainties are only observed when estimating high CO emissions. In general, the quality of the predictions is considered satisfactory, suggesting that this technique also offers a good potential for the monitoring of premixed flames. As it has been discussed previously, the capabilities of identification procedures might be dependent on the nature of the images being analyzed. Those used in this work correspond to spontaneous flame emission and have a low resolution in space (in particular, due to line-of-sight integration), time (effective exposure times are longer than characteristic fluctuation times), 1
Estimated values
0.8
0.6
0.4
and wavelength (visible radiation without bandfiltering). Since other imaging techniques might produce different images for the same flames, it is not obvious whether they would be compatible with the same processing methods. A limited exploration in this respect was accomplished by changing the spatial resolution of the images. Tomographic deconvolution (Abel inversion) was applied to the images in order to estimate the actual radial distribution of light emission. Figure 15.16 shows one of the original flame images and its associated tomographic reconstruction, which display a notably different pattern. In particular, the deconvoluted image reveals that light emission comes mainly from a relatively thin, conical layer surrounding the internal recirculation zone, with a negligible contribution from inner regions. The cepstral analysis was applied in order to evaluate if this method is also suitable for images with in-depth resolution. A total of N = 9 combustion regimes (ER = 0.56–1) were studied. N2 = 18 frames were used to evaluate an unknown state by comparing with each pattern flame, characterized by N1 = 18 images. Figure 15.17a displays the probabilities calculated with the original images. Five out of the nine cases were correctly predicted. Variations in ER result in slight visual changes, and the larger uncertainties are observed for ER > 0.8 for this set of flames. The results of the application of the same algorithm to deconvoluted images are shown in Figure 15.17b. The probabilities along the diagonal were sensibly higher in this case, with correct identifications in eight of the nine combustion regimes. These results suggest that the identification procedure can also be applied to spatially resolved images. For this particular case, deconvolution of line-of-sight images led to even better results, probably due to an enhanced definition of flame geometry and luminosity distribution. Although experimental confirmation would be needed, the method is expected to be also valid for spectrally resolved images, which might even lead to improved performance since they display stronger gradients and sensitivity to operating conditions than the luminosity in the visible range.
0.2
0
0
0.2
0.4 0.6 Measured values ER
NOx
0.8
1
CO
Figure 15.15 Comparison between predictions and measurements. NOx and CO emissions have been normalized by characteristic maximum values (50 and 3000 ppm, respectively). (From Hernandez, R. and Ballester, J., Combustion and Flame, 155, 509–28, 2008. With permission from Elsevier.)
© 2011 by Taylor and Francis Group, LLC
15.6 Conclusions In recent years there has been a growing interest in the development of image-based techniques for the monitoring of practical flames. On the one hand, this is motivated by the need for advanced diagnostic and optimization methods for combustion applications, which might entail important benefits in terms of efficiency, pollutant reduction, reliability, and flexibility of
352
Industrial Combustion Testing
Figure 15.16 Original image (left) and its tomographic reconstruction (right) of a premixed flame with ER = 0.8. (a)
(b)
50
80
45
70
40
60
) Probability (%
50
15 10
me
0.98
f la
1
0.8
0.7 0.65 Patt 0.6 ern flam 0.59 es 0.57
20 10 0
1
0.98
Te st
0
s
5
0.56 0.57 0.59 0.6 0.65 0.7 0.8 0.98
30
0.56
1
0.56 0.57 0.59 0.6 0.65 0.7 0.8 0.98 s
20
40
me
25
0.8 0.7 0.65 Patt 0.6 ern flam es 0.59 0.57
f la
30
Te st
(%) Probability
35
0.56
1
Figure 15.17 P(xi ~ yj) estimated using (a) original images and (b) their tomographic reconstructions.
combustion equipment. On the other hand, the continuous increase in the performance-to-cost ratio of sensors and data-processing devices enables the development of powerful, yet affordable vision-based monitoring systems, suitable for industrial environments. Flame images contain a large amount of information that might be used to characterize a particular combustion state. However, the main difficulty in this field is thought to be the conversion of visual data into practical information or in the form of physical combustion parameters. Widely different approaches can be found in the literature. Some authors work with bandfiltered radiation (infrared, chemiluminescence bands) whereas others capture broadband light (usually, in the visible range). Nevertheless, in the end, feature-extraction methods are applied in most cases; that is, the images collected are postprocessed to calculate a few selected magnitudes.
© 2011 by Taylor and Francis Group, LLC
These can have a direct physical meaning (e.g., temperature) or constitute a set of representative parameters related to the geometry, luminosity, or color of the flame. A notable advantage of feature-extraction methods is the dramatic reduction in the amount of information needed to characterize a flame. However, they necessarily entail some information loss and, when facing a new application, a costly development stage may be required in order to select the most representative features as well as functional forms suitable to correlate visual data with combustion states. The two processing methods described in this work follow a different approach, as they treat the image as a whole instead of reducing it to a few representative parameters. The first one exploits the good capabilities of SOFM for pattern classification (and, therefore, identification). Although these methods are conceived to
353
Image-Based Techniques for the Monitoring of Flames
provide qualitative information, values for the relevant parameters can be readily obtained through an assignation/interpolation procedure. The second method uses an approach similar to those used for speaker recognition. In this case, PCA demonstrated to be an effective data-reduction technique as well as to convert the image data into a suitable format. A probabilistic approach was followed to quantify the degree of similarity of an unknown image with all reference patterns; the representation of patterns by a set of images is thought to be a salient feature of this method, as it provides the robustness required to account for the variability inherent to each combustion state. Both methods showed a notably good performance as flame identification tools (or, equivalently, for the estimation of relevant variables like NOx emissions). Also very relevant is that both methods are conceived as general procedures; none of the operations performed for image analysis (except, perhaps, a part of the preprocessing) was specific to one particular combustion rig. It is important to note that these techniques have been applied, without specific adaptations for the different cases, to widely different flames: oil-/gas-fired, stable/ pulsating, diffusion/premixed, fast-mixing/staged, and attached/detached. The good results obtained in all cases are thought to strongly support the generalizable nature of the approaches presented. This must be considered, however, an empirical conclusion, as it is not possible, on physical grounds, to justify the existence of a defined relationship among images of visible radiation and flame characteristics for the general case. This kind of technique is thought to enable the development of advanced monitoring methods in practical combustion systems, either for the identification of flame state (e.g., control applications, anomaly detection, etc.) or as a virtual sensor for the prediction of relevant parameters such as NOx emissions (e.g., for reduced response time or to determine emissions from individual flames in multiburner chambers). The sensor can be a conventional CCD camera, but also any other imaging method such as planar laser techniques. Moreover, the performance might still improve (as observed in the analysis presented for tomography-reconstructed images), since the information generated with those methods is usually more intimately related to flame properties and displays stronger gradients and sensitivity to operating conditions than the luminosity in the visible range.
Education and Science (grant ENE2007-63641) and by the European Commission (contract ENK5-CT2002000662, “Alternative fuels for industrial gas turbines”— AFTUR). The authors gratefully acknowledge the help of R. Ichaso, A. Sanz, L. Ojeda, and A. Sobrino.
References
Acknowledgments The results presented were obtained in the framework of research projects funded by the Spanish Ministry of
© 2011 by Taylor and Francis Group, LLC
1. Heitor, M. V., and Moreira, A. L. N. “Thermocouples and Sample Probes for Combustion Studies.” Progress in Energy and Combustion Science 19 (1993): 259–78. 2. Kohse-Höinghaus, K., Barlow, R. S., Aldén, M., and Wolfrum, J. “Combustion at the Focus: Laser Diagnostics and Control.” Proceedings of the Combustion Institute 30 (2005): 89–123. 3. Gaydon, A. G. The Spectroscopy of Flames. London: Chapman & Hall, 1974. 4. Shimoda, M., Sugano, A., Kimura, T., Watanabe, Y., and Ishiyama, K. “Prediction Method of Unburnt Carbon for Coal Fired Utility Boiler Using Image Processing Technique of Combustion Flame.” IEEE Transactions on Energy Conversion 5 (1990): 640–45. 5. Huang, Y., Yan, Y., and Riley, G. “Vision-Based Measurement of Temperature Distribution in a 500-kW Model Furnace Using the Two-Colour Method.” Measurement 28 (2000): 175–83. 6. Lu, G., Gilabert, G., and Yang, Y. “Vision Based Monitoring and Characterisation of Combustion Flames.” Journal of Physics: Conference Series 15 (2005): 194–200. 7. Correia, D. P., Ferrao, P., and Caldeira-Pires, A. “Flame Three-Dimensional Tomography Sensor for In-Furnace Diagnostics.” Proceedings of the Combustion Institute 28 (2000): 431–38. 8. Lou, C., and Zhou, H. C. “Deduction of the TwoDimensional Distribution of Temperature in a Cross Section of a Boiler Furnace from Images of Flame Radiation.” Combustion and Flame 143 (2005): 97–105. 9. Ruão, M., Costa, M., and Carvalho, M. G. “A NOx Diagnostic System Based on a Spectral Ultraviolet/Visible Imaging Device.” Fuel 78 (1999): 1283–92. 10. Tao, W., and Burkhardt, H. “Vision-Guided Flame Control Using Fuzzy-Logic and Neural Networks.” Particle and Particle System Characterization 12 (1995): 87–94. 11. Allen, M. G., Butler, C. T., Johnson, S. A., Lo, E. Y., and Russo, F. “An Imaging Neural Network Combustion Control System for Utility Boiler Applications.” Combustion and Flame 94 (1993): 205–14. 12. Lu, G., Yan, Y., Huang, Y., and Reed, A. “An Intelligent Vision System for Monitoring and Control of Combustion Flames.” Measurement Control 32 (1999): 164–68. 13. Docquier, N., Lacas, F., and Candel, S. “Closed-Loop Equivalence Ratio Control of Premixed Combustors Using Spectrally Resolved Chemiluminescence Measurements.” Proceedings of the Combustion Institute 29 (2002): 139–45.
354
14. Muruganandam, T. M., Kim, B.-H., Morrell, M. R., Nori, V., Patel, M., Romig, B. W., and Seitzman, J. M. “Optical Equivalence Ratio Sensors for Gas Turbine Combustors.” Proceedings of the Combustion Institute 30 (2005): 1601–9. 15. Hardalupas, Y., Orain, M., Panoutsos, C. S., Taylor, A. M. K. P., Olofsson, J., Seyfried, H., Richter, M., Hult, J., and Aldén, M. “Chemiluminescence Sensor for Local Equivalence Ratio of Reacting Mixtures of Liquid Fuel Vapour in Air.” In Proceedings of the First International Conference on Industrial Gas Turbine Technologies, edited by D. Pollard and P. A. Pilavachi, July 10–11, 2003, Brussels, Belgium. 16. Nori, V. N., and Seitzman, J. M. “Chemiluminescence Measurements and Modelling in Syngas, Methane and Jet-A Fuelled Combustors.” In 45th Aerospace Sciences Meeting and Exhibit, AIAA-2007-0466, January 2007. 17. Kurihara, N., Nishikawa, M., Watanabe, A., Satoh, Y., Ohtsuka, K., Miyagaki, H., Higashi, T., and Masai, T. “A Combustion Diagnosis Method for Pulverized Coal Boilers Using Flame-Image Recognition Technology.” IEEE Transactions on Energy Conversion EC-1 (1986): 99–103. 18. Wu, J., Zhang, M., Fan, H., Fan, W., and Zhou, Y. “A Study on Fractal Characteristics of Aerodynamic Field in low-NOx Coaxial Swirling Burner.” Chemical Engineering Science 59 (2004): 1473–79. 19. Marques, J. S., and Jorge, P. M. “Visual Inspection of a Combustion Process in a Thermoelectric Plant.” Signal Process 80 (2000): 1577–89. 20. Rahman, M. G. A., Gibbins, J. R., and Forrest, A. K. “Combustion in Power Station Boilers—Advanced Monitoring Using Imaging.” Report No. Coal R264 DTI/ Pub URN 04/1797, 2004. 21. Baek, W. B., Lee, S. J., Baeg, S. Y., and Cho, C. H. “Flame Image Processing and Analysis for Optimal Coal Firing of Thermal Power Plant.” In IEEE International Symposium on Industrial Electronics Proceedings, 2001, 928–31. 22. Tuntrakoon, A., and Kuntanapreeda, S. “Image-Based Flame Control of a Premixed Gas Burner Using Fuzzy Logics.” Foundations of Intelligent Systems, 673–77. Berlin: Springer-Verlag, Heidelberg, 2003, 2871/2003. 23. Yu, H., and MacGregor, J. F. “Monitoring Flames in an Industrial Boiler Using Multivariate Image Analysis.” American Institute of Chemical Engineering Journal 50 (2004): 1474–83. 24. Yan, Y., Lu, G., and Colechin, M. “Advanced Monitoring and Characterisation of Pulverised Coal Flames.” Fuel 81 (2002): 647–56.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
25. Huang, Y., Yan, Y., Lu, G., and Reed, A. “On-Line Flicker Measurement of Gaseous Flames by Image Processing and Spectral Analysis.” Measurement Science Technology 10 (1999): 726–33. 26. Sbarbaro, D., Farias, O., and Zawadsky, Z. Combustion and Flame 132 (2003): 591–95. 27. Burkhardt, H., Oest, L., and Tao, W. “Vision-Guided Flame Control.” SENSOR 95 Kongreß, 1995. 28. Wang, F., Wang, X. J., Ma, Z. Y., Yan, J. H., Chi, Y., Wei, C. Y., Ni, M. J., and Cen, K. F. “The Research on the Estimation for the NOx Emissive Concentration of the Pulverized Coal Boiler by the Flame Image Processing Technique.” Fuel 81 (2002): 2113–20. 29. Bae, H., Kim, S., Wang, B.-H., Lee, M. H., and Hrashima, F. “Flame Detection for the Steam Boiler Using Neural Networks and Image Information in the Ulsan Steam Power Generation Plant.” IEEE Transactions on Industrial Electronics 53 (2006): 338–48. 30. Sanz, A., Ballester, J., and González, M. A. “Investigation of the Characteristics and Stability of Air-Staged Flames.” Experimental Thermal Fluid Science 32 (2008): 776–90. 31. Ballester, J., Hernández, R., Sanz, A., Smolarz, A., Barroso, J., and Pina, A. “Chemiluminescence Monitoring in Premixed Flames of Natural Gas and its Blends with Hydrogen.” Proceedings of the Combustion Institute 32 (2009): 2983–91. 32. Hernández, R., and Ballester, J. “Flame Imaging as a Diagnostic Tool for Industrial Combustion.” Combustion and Flame 155 (2008): 509–28. 33. Haykin, S. Neural Networks: A Comprehensive Foundation. New York: Macmillan/IEEE Press, 1994. 34. Egmont-Petersen, M., de Ridder, D., and Handels, H. “Image Processing with Neural Networks: A Review,” Pattern Recognition 35 (2002): 2279–2301. 35. Molz, R. F., Engel, P. M., Moraes, F. G., Torres, L., and Robert, M. “A Fast Prototyping Neural Network Model for Image Classification.” In Proceedings of the XV Conference on Design of Circuit and Integrated Systems (DCIS), 1 (2000): 836–41. 36. Franco, L., and Treves, A. “A Neural Network Facial Expression Recognition System Using Unsupervised Local Processing.” in Proceedings of the 2nd International Symposium on Image and Signal Processing and Analysis (ISPA 01), 2001: 628–32. 37. Campbell, J. P. “Speaker Recognition: A Tutorial.” Proceedings of IEEE 85 (1997): 1437–62. 38. Vicente, M. A., Reinoso, O., Pérez, C., Sabater, J. A., and Azorín, J. A. “Recognition of 3D Objects by PCA Analysis” (in Spanish). In XXIII Jornadas de Automática, Tenerife, 2002.
16 High Temperature Cameras William J. Lang Contents 16.1 Introduction.................................................................................................................................................................. 356 16.2 History of Furnace Cameras...................................................................................................................................... 356 16.3 Technology of Furnace Cameras............................................................................................................................... 356 16.3.1 Furnace Lens.................................................................................................................................................. 356 16.3.2 Fields of View................................................................................................................................................. 357 16.3.3 Direction of View.......................................................................................................................................... 357 16.3.4 Light Volume Control................................................................................................................................... 357 16.3.5 Wall Box Mounting....................................................................................................................................... 357 16.3.6 Lens Coolant.................................................................................................................................................. 357 16.3.7 Camera............................................................................................................................................................ 359 16.3.8 Camera Housing............................................................................................................................................ 359 16.3.9 Automatic Retraction Systems..................................................................................................................... 359 16.3.10 Portable Systems............................................................................................................................................ 359 16.4 Applications of Furnace Cameras............................................................................................................................. 360 16.4.1 Wide Range of Process Applications.......................................................................................................... 360 16.4.2 Refineries........................................................................................................................................................ 360 16.4.3 Trash-to-Steam (Waste-to-Energy) Plants.................................................................................................. 361 16.4.4 Boilers.............................................................................................................................................................. 362 16.4.5 Cement Kilns.................................................................................................................................................. 362 16.4.6 Biomass........................................................................................................................................................... 363 16.4.7 Steel Industry................................................................................................................................................. 363 16.4.8 Reheat Furnaces............................................................................................................................................. 363 16.4.9 Vacuum Degassers........................................................................................................................................ 365 16.4.10 Remelt/Reverberatory Furnaces................................................................................................................. 365 16.4.11 U.S. Navy........................................................................................................................................................ 365 16.4.12 Pulp and Paper Industry.............................................................................................................................. 365 16.4.13 Electric Utilities.............................................................................................................................................. 365 16.4.14 Glass Furnaces............................................................................................................................................... 365 16.4.15 Chemical Incineration................................................................................................................................... 365 16.5 Summary: Observing Combustion Problems......................................................................................................... 365 16.5.1 Flame Impingement...................................................................................................................................... 366 16.5.2 Boiler or Ash Hopper Slag........................................................................................................................... 366 16.5.3 Defective Ignition.......................................................................................................................................... 366 16.5.4 Unbalanced Combustion.............................................................................................................................. 366 16.5.5 NOx Control.................................................................................................................................................... 366 16.5.6 Steam Leaks.................................................................................................................................................... 366 16.5.7 Waste Hot Spots............................................................................................................................................. 366 16.5.8 Smoking Conditions..................................................................................................................................... 366 16.5.9 Flame Lift-Off................................................................................................................................................ 367
355 © 2011 by Taylor and Francis Group, LLC
356
Industrial Combustion Testing
16.5.10 Pulsating Flame............................................................................................................................................. 367 16.5.11 Flashback........................................................................................................................................................ 367 16.6 The Future..................................................................................................................................................................... 367 16.7 Conclusion.................................................................................................................................................................... 368 References................................................................................................................................................................................. 368
16.1 Introduction Temperatures in boilers, kilns, and furnaces in the oil refining, steel, glass, trash-to-steam, cement, power, pulp and paper, and other industries can reach 4200°F. It is critical that operators monitor the combustion in order to know, in real time, what is happening inside the “firebox,” and then take corrective action where necessary. Are burners igniting correctly? Are flames impinging on the inside walls, potentially piercing them? Is the fire smoking, a sign of incomplete combustion? Watching the flames 24/7 provides certainty that the chamber is running with maximum cost efficiency and safety. You can fly an airplane without windows, but why would you want to? Likewise, why would you want to operate a combustor of any kind without being able to observe the quality of your combustion? It is also critical that operators monitor the process— the flow of material through the furnace—to be sure everything is happening correctly. Are steel billets or glass materials, for instance, moving uniformly without causing a pileup? Though continuous monitoring is essential, thousands of installations still follow the inefficient procedure in which a person walks up to the furnace at inspection periods several times a day and looks through existing openings in the sidewall, roof, or floor to see what is going on. Sometimes a maintenance person wearing a protective mask will even open an access port, which can be deadly if pressures suddenly turn positive inside the furnace and flames burst out of the port. Manual inspection of furnaces through a viewing port offers very limited information since the person’s field of view is usually only about 10° due to the thickness of the refractory wall. Many locations are scrapping that procedure in favor of high temperature furnace cameras that operate continuously and provide a very wide field of view—potentially everything within the firebox.
and view nearby flames. Employing a lens, this system, basically a flame detector, used a black-and-white vidicon tube-type camera. The manufacturer asked John Lang, who headed Lenox Instrument Co., to supply a periscope device that could penetrate the boiler wall, protect the optics from the heat, and spread the field of view over the target area. Lenox developed a 12-inchlong furnace periscope with a quartz objective lens, air cooling, and an adjustable iris for light control, and supplied it over several decades to the manufacturer and other original equipment manufacturers (OEMs). The quartz lens, at the tip of the tube, sent images of the flames through a series of achromatic relay lenses, back to a camera at the base of the periscope. Before the tube was utilized, inspectors would check furnace interiors by placing cameras in viewing doors that were typically 4 inches wide and 12 inches high. Their field of view was limited by that aperture. The furnace lens was a major advance. It spread the field of view, with the potential of viewing everything inside the furnace. Penetrating the portal, it provided a larger view, a more useful, complete image of everything occurring in the firebox. Lenox sold the furnace lens tube to OEMs until about 1990, when it developed its own complete system replacing the vidicon with a charged couple device (CCD) solid state color camera in a protective housing behind the lens tube. This system, called FireSight, can view very bright images without being scarred, whereas tube-type camera devices would be damaged over time in the furnace environment. The closed circuit CCD camera televises continuous color images of the furnace interior, including burner performance and flame patterns, to monitors in a control room.
16.3 Technology of Furnace Cameras 16.3.1 Furnace Lens
16.2 History of Furnace Cameras Furnace cameras have been around a long time. Back in 1950, a major boiler manufacturer was utilizing a large camera system that could look through the boiler door
© 2011 by Taylor and Francis Group, LLC
The furnace lens (lens tube) is the single most important component of today’s furnace camera systems. Cooled by compressed air or water, it can be 12 inches or even 12 feet long. The length of the probe depends on the application. In a refinery, for instance, it might be as long as 3 feet. Older glass-optic systems could not operate above 2500°F, provided hazy, relatively low-quality
357
High Temperature Cameras
images, required regular maintenance, and were often discarded after relatively short times. In contrast, the objective lens at the tip of today’s probes is usually quartz. This lens gathers the image and then presents it to the achromatic relay lenses or transfer elements (which could be thought of as amplifiers) that move the signal down an elongated tube, bringing it back to the solid-state camera sensor. The last element presents the image to the sensor in the proper size format for that particular camera. The image then goes to the control room via coaxial cable. The furnace lens provides clear color images at temperatures even higher than 3500°F, is self-maintaining, and very reliable over years of use. A lens at the tip of a 12-inch long probe at an installation like a refinery is air-cooled, with a 24-inch or 36-inch length and either direct or right-angle viewing. An air filtration system removes aerosols, vapor, oil, and particles as small as 0.03 microns from cooling air that circulates under positive pressure within the system’s protective shield, cleaning and cooling the camera, lens, and the achromatic relay lenses that carry the images of the flames from the lens to the camera. This self-purging, high-efficiency compressed air filter ensures trouble-free performance and a clear view of the furnace interior. The compressed air finally goes out the front end of the probe. 16.3.2 Fields of View A high-temperature furnace camera can view, through its lens, the exact area that needs to be monitored. Viewing angles can be 105°, 90°, 60°, 45°, and 30°. (See the Figure 16.1 chart of coverage at several viewing angles and viewing distances.) 16.3.3 Direction of View The operator can view directly (straight ahead along the axis of the scope), at a right angle to the axis (basically side viewing at 90°), or forward obliquely from the axis (at 45° halfway between direct and right viewing). For instance, the appropriate lens might look down directly from above, look straight across directly, go through a port and look down at 90°, be rotated to view different areas at 90°, look forward through the roof of the furnace at 45° to view material moving toward the observer, or meet many other observation needs. (See Figure 16.2 illustrations.) 16.3.4 Light Volume Control Light volume control is essential for viewing combustion applications under different conditions, from light-off to maximum output. Advanced furnace cameras utilize
© 2011 by Taylor and Francis Group, LLC
a motorized iris that an operator in the control room remotely utilizes to adjust the amount of light reaching the camera, as needed for the clearest image. Located near the camera and within the camera’s protective housing, this iris avoids problems like “whiteout” (loss of picture due to intense light), eliminates flaring/blooming, can be adjusted to meet varying lighting conditions, and helps provide maximum information for decisions. 16.3.5 Wall Box Mounting The furnace lens must be protected against extremely hostile conditions, from flames and very high temperatures to particles and vapor. A wall box housing, which is the basic mounting and primary coolant device, provides this protection. It also allows air to circulate and helps cool the lens tube components. The wall box may be connected to the forced draft (FD) fan for a boiler or furnace. It also may be an open connection drawing air with negative pressure of the boiler. In such occurrences, the wall box, only used on air-cooled systems, suppresses flames, pushing them away, and preventing them from coming out of the firebox. It functions as a flame-arrestor if a furnace goes to positive pressure, trying to push flames out any opening. The back of the wall box mounts to a flange, providing a laminar flow of air along the lens tube that encapsulates the furnace lens, evenly distributing it down the tube. This prevents the viewing lens from being covered and obscured and the camera not functioning correctly. A high volume of air forces that material away from the viewing port. The wall box also provides a coolant function so that the lens tube doesn’t have to deal with as much heat. Compressed air, very successfully going through the tube, handles the balance of the temperature. If the wall box keeps the temperature to, say, 600°F or 800°F, the coolant passing through the tube is more than enough to keep the tube temperature below 400°F. 16.3.6 Lens Coolant The furnace lens system needs a purge/coolant system to keep the lens clean. The lens must observe clearly, providing crisp, clear images while it is often looking through dusty, dirty environments where there is an enormous amount of particulate flying around. Most furnace camera systems are air-cooled. Twostage coolant systems, using high-volume air and compressed air from separate sources, are very effective. High-volume, low-pressure air travels through a protective wall box, the primary coolant shroud that encloses the lens tube. This air is supplied either by an FD fan or is pulled in by a negative-pressure furnace. The lens tube itself is cooled by compressed air.
358
Fire sight
Pultz
Boiler/furnace camera field of view chart Camera’s viewing angle
Distance between Lenox camera and subject area 5´ (1.5 m) away
10´ (3.0 m) away
15´ (4.6 m) away
25´ (7.6 m) away
35´ (10.7 m) away
50´ (15.2 m) away
H area
V area
H area
V area
H area
V area
H area
V area
H area
V area
H area
V area
95°
8.7´ (2.7 m)
6.6´ (2.0 m)
17.5´ (5.3 m)
13.1´ (4.0 m)
26.2´ (8.0 m)
19.6´ (6.0 m)
43.7´ (13.3 m)
32.7´ (10.0 m)
61.1´ (18.6 m)
45.8´ (14.0 m)
87.3´ (26.6 m)
65.5´ (20.0 m)
90°
8´ (2.4 m)
6´ (1.8 m)
16´ (4.8 m)
12´ (3.6 m)
24´ (7.3 m)
18´ (5.5 m)
40´ (12.2 m)
30´ (9.1 m)
56´ (17.1 m)
42’ (12.8 m)
80´ (24.4 m)
60´ (18.3 m)
72°
5.8´ (1.8 m)
4.4´ (1.3 m)
11.6´ (3.5 m)
8.7´ (2.7 m)
17.4´ (5.3 m)
13.1´ (4.0 m)
29.1´ (8.9 m)
21.8´ (6.6 m)
40.7´ (12.4 m)
30.5´ (9.3 m)
58.1´ (17.7 m)
43.6´ (13.3 m)
Direct
60°
4.6´ (1.4 m)
3.5´ (1.1 m)
9.2´ (2.8 m)
6.9´ (2.1m)
13.8´ (4.2 m)
10.4´ (3.2 m)
23.1´ (7.0 m)
17.3´ (5.3 m)
32.3´ (9.8 m)
24.3´ (7.4 m)
46.2´ (9.8 m)
34.6´ (10.5 m)
Right angle
55°
4.2´ (1.3 m)
3.1´ (0.9 m)
8.3´ (2.5 m)
6.3´ (1.9 m)
12.5´ (3.8 m)
9.4´ (2.9 m)
20.8´ (6.3 m)
15.6´ (4.8 m)
29.2´ (8.9 m)
21.9´ (6.7 m)
41.6´ (12.7 m)
31.2´ (9.5 m)
Direct
45°
3.2´ (0.9 m)
2.5´ (0.7 m)
6.6´ (2.0 m)
4.9´ (1.5 m)
9.9´ (3.0 m)
7.5´ (2.3 m)
16.6´ (5.0 m)
12.4´ (3.8 m)
23.2´ (7.1 m)
17.4´ (5.3 m)
33.1´ (10.1 m)
24.8´ (7.6 m)
Direct
30°
2.1´ (0.6 m)
1.6´ (0.5 m)
4.3´ (1.3 m)
3.2´ (1.0 m)
6.4´ (1.9 m)
4.8´ (1.4 m)
10.7´ (3.2 m)
8.0´ (2.4 m)
15.0´ (4.5 m)
11.2´ (3.4 m)
21.4´ (6.5 m)
16.0´ (4.8 m)
Zoom
72°Z
5.8´ (1.8 m)
4.4´ (1.3 m)
11.6´ (3.5 m)
8.7´ (2.7 m)
17.4´ (5.3 m)
13.1´ (4.0 m)
29.1´ (8.9 m)
21.8´ (6.6 m)
40.7´ (12.4 m)
30.5´ (9.3 m)
58.1´ (17.7 m)
43.6´ (13.3 m)
Zoom
15°Z
1.0´ (0.3 m)
0.8´ (0.2 m)
2.1´ (0.6 m)
1.6´ (0.5 m)
3.2´ (1.0 m)
2.4´ (0.7 m)
5.3´ (1.6 m)
4.0´ (1.2 m)
7.4´ (2.3 m)
5.5´ (1.7 m)
10.5´ (3.2 m)
7.9´ (2.4 m)
Variable zoom
95°V
8.7´ (2.7 m)
6.6´ (2.0 m)
17.5´ (5.3 m)
13.1´ (4.0 m)
26.2´ (8.0 m)
19.6´ (6.0 m)
43.7´ (13.3 m)
32.7´ (10.0 m)
61.1´ (18.6 m)
45.8´ (14.0 m)
87.3´ (26.6 m)
65.5´ (20.0 m)
Variable zoom
43°V
3.2´ (0.9 m)
2.4´ (0.7 m)
6.3´ (1.9 m)
4.7´ (1.4 m)
9.5´ (2.9 m)
7.1´ (2.2 m)
15.8´ (4.8 m)
11.8´ (3.6 m)
22.1´ (6.7 m)
16.5´ (5.0 m)
31.5´ (9.6 m)
23.6´ (7.2 m)
Direct Direct Direct
Direct
Note: Figures are intended as approximate guidelines to help select the appropriate Lenox camera lens and are based on the camera/monitor industry standard 4 × 3 format (H:V). Figure 16.1 Chart shows furnace camera field of view at different angles and distances.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
Diag.°
359
High Temperature Cameras
(a)
(b)
Monitor edges
(c)
Field cone
Field
ld Fie
of vie
w co
ne
v of iew ne co Figure 16.2 Furnace cameras can (a) view directly, (b) at a right angle, or (c) forward obliquely. (“FireSight High-Temperature Remote Viewing Systems.” Lenox Instrument Company brochure with photos and diagrams of system and components.)
Furnace camera systems may also be cooled by water, though this approach is less common. These systems also employ two-stage coolant. A water jacket, which is a triple-wall heat exchanger, takes the place of the wall box. This water, cooling the lens tube, is separate from the compressed air that passes through the tube to purge, clean, and also help cool the lens. Using these two independent sources provides a very reliable system because one will be available at all times. Water pressure and air pressure can fluctuate in all these systems depending on operating conditions in the plant. Water-cooled systems can be used at the highest temperatures—up to 4200°F. Water-cooled systems are generally employed above 3000°F while air-cooled systems are used below 3000°F. Water cooling is frequently employed in steel and glass applications because a highquality water supply is readily available. In many other applications, water may be available, but is not recyclable. In reality, few, if any, applications are as high as 4200°, but products have been tested to that height. They may work well higher than that in operations like vacuum melting or induction melting , but this would be considered experimental, requiring extra caution and employing the proper diagnostic equipment to monitor things like water temperature. 16.3.7 Camera Closed circuit TV (CCTV) cameras (color or black-andwhite) for furnace systems are usually industrial surveillance types (typically 540-line high-resolution). These transmit the images of combustion to normal industrial-type monitors, which may be in a control room or locally at the viewing site. The process can also be digitally recorded for further evaluation.
© 2011 by Taylor and Francis Group, LLC
16.3.8 Camera Housing The protective housing for the camera can be air-cooled or water-cooled. The housing protects both the camera and its light volume control unit. It also cools the compressed air and constantly changes the air in the cavity. (See Figure 16.3 for schematic of complete system.) 16.3.9 Automatic Retraction Systems Retraction systems automatically withdraw the furnace camera several feet back from the firebox if the systems sense loss of either water, air coolant, or high lens tube temperature. This gives the lens additional protection from the heated air that would blow against the lens if fans stopped and negative pressure changed to positive pressure. The retraction system is often air-operated, using an air-reserve tank as a purely pneumatic system component. This makes it totally independent of electricity. On loss of air pressure, water coolant, or high lens tube temperature, a solenoid valve opens and activates a rodless cylinder to withdraw the lens from harms way. The system will not return the camera to its inserted position until the problem is corrected. 16.3.10 Portable Systems Furnace camera systems, including furnace lens, camera, air filtration system and monitor can be supplied in a compact portable case that can be wheeled to the point of usage. The maintenance or operations engineer can evaluate a furnace or boiler right on the spot, with direct or right-angle views. The furnace lens is typically 24-inches or 36-inches long, using an 8-inch wide color monitor. Portable systems allow a maintenance engineer to serve multiple boilers or burners. Typically used for tuning
360
Industrial Combustion Testing
6
1 5
6
4
2
Typical firesight components 1. Wall-box assembly 2. Furnace camera 3. Lens tube assembly 4. Motor drive assembly 5. CCTV camera 6. Air filtration assembly 7. Monitor 3
Figure 16.3 Schematic of furnace camera system shows (1) wallbox, (2) furnace lens, (3) air-cooled CCTV housing, (4) light volume control, (5) compact CCD camera, (6) video monitor, and (7) high-efficiency compressed air filter. (“FireSight High-Temperature Remote Viewing Systems.” Lenox Instrument Company brochure with photos and diagrams of system and components.)
burners, they have been very popular at coal-fired installations that have to tune or adjust such burners, which may be 10 feet apart. The probe is typically placed close to the burner to view it and make the necessary adjustments. These systems can also record what is happening inside the boiler or furnace, allowing engineers in the office to evaluate operations and adjust as necessary (Figure 16.4).
16.4 Applications of Furnace Cameras 16.4.1 Wide Range of Process Applications Furnace cameras are used in a wide variety of applications where steam is generated or where you heat something in order to process it. They provide essential information on critical parameters, verify stable, safe, and efficient performance, and allow precise tuning for complete combustion. 16.4.2 Refineries Crude-oil heaters, also called fired heaters, heat crude oil to begin the process for refining. Open flames in a chamber heat thick oil in boiler tubes to strip off lighter hydrocarbons in the separation process before other processing takes place [5].
© 2011 by Taylor and Francis Group, LLC
Figure 16.4 Portable systems can be wheeled to point of usage. (“FireSight HighTemperature Remote Viewing Systems.” Lenox Instrument Company brochure with photos and diagrams of system and components.)
Operators use furnace cameras to remotely view urners in the numerous fired heaters that are heatb ing this oil to spot such performance defects as flame impingement, flames that are burning too low, clogged burners, irregular flame patterns, flashback, flames that are producing too much pollution, or oil spillage.
361
High Temperature Cameras
More and more refineries now utilize furnace cameras to monitor the burners 24 hours a day much more costeffectively than with manual inspection. Monitoring the quality of combustion, and adjusting it as necessary, they reduce fuel consumption as much as 2%[1]. Detecting flame impingement early via cameras is very important because the flame can damage the tubes if it touches the tube walls directly. Since the tubes are filled with crude oil that is being heated, a leak or other damage caused by flame impingement may be catastrophic to the process if not to the entire refinery. To detect potential impingement or other problems, the furnace camera may be mounted on the wall of the silo process heater (or crude oil heater) at an angle that allows the lens to view the outlet of the burners at the bottom of the silo that are heating the walls of the tubing (see Figure 16.5). An experienced operator, viewing the control room display, can watch clear pictures of each burner and monitor both the flames and the tube walls in the background. He or she can also spot “lazy” flames—too long and too rich in combustibles—without enough air to actually burn everything and thus taking a lot longer to burn. Figure 16.6 shows a variety of flames in an oil refinery boiler, where uniform burning is desired. Operators can also spot a clogged burner, or a flame that may be producing too much pollution (NOx). Since prehistoric times, it has always been true that, when you’re building a fire, you watch it. In the era before furnace cameras, a performance engineer wearing a face shield had to open the boiler door (with the possibility of being burned) to spot problems, and radio the information to the control room. Continuous automatic remote monitoring using a camera is a much preferred procedure to opening a port. A small refinery might install one to four furnace cameras in each crude oil heater. Hydrocarbon, petrochemical, and other industries also employ portable
Figure 16.5 Furnace camera mounted on wall of crude oil heater.
© 2011 by Taylor and Francis Group, LLC
furnace cameras that can be quickly inserted into a port to diagnose and solve burner problems on process heaters or furnaces. Results are viewed on a portable TV monitor instead of a control room. Advanced furnace cameras now installed in oil refineries are explosion proof. 16.4.3 Trash-to-Steam (Waste-to-Energy) Plants Furnace cameras monitor the trash as it goes through the combustion zone on conveyors or grates (Figure 16.7). This allows operators to be sure there is maximum burnout of the waste on the grates, and that the waste stream is moving efficiently. Operators can adjust controls agitating the trash, and optimizing fuel/air ratios, for maximum burning of the many types of waste that passes through. If operators do not achieve even distribution of material on the traveling grate, they need to take corrective action to prevent localized overheating, which can cause severe damage to the waterwall.
Figure 16.6 (See color insert following page 424.) Flames in refinery boiler may have inconsistent flame characteristics, as shown on furnace camera monitor (Section A and Section E). (Baukal, C. E., ed., The John Zink Combustion Handbook, CRC Press, Boca Raton, FL, 2001.)
Figure 16.7 (See color insert following page 424.) Camera monitors trash burning on traveling grate in waste-to-steam boiler.
362
They can make the necessary adjustments so that the trash is evenly distributed and burns uniformly and completely before it leaves the combustion zone. The camera (or cameras) can be installed at grate level to make sure combustion is complete and the fire is out before the ash goes to the ash hopper. Plants don’t want flaming material going down that hopper onto a conveying system, posing the danger of subsequent fires that could disrupt the operation. Operators see the “fireline” in the background and ash, coming toward the camera, in the foreground. If the camera shows flame in that foreground, they need to take corrective action to be sure the fire is extinguished before the material goes to the ash hopper. They would probably stop the grate in an effort to prevent the fire from getting down into the conveyor system. The furnace camera is typically located close to where the ash drops down a funnel to the conveyor. Furnace cameras also help prevent hot spots or overheat situations close to the walls of the plant’s boiler tubes, providing early warning before these burn through the tubes. Some plants view the fire from a spot close to the superheater in order to keep the fire in the center of the furnace. Failure can be very costly. A 2,200ton-per-day plant had to install new boiler tubes after trash piled up and burned a hole in the waterwalls [2]. Typical locations for furnace cameras in waste-to-energy stoker-fired boilers:
1. A port (above the burners) in the nose arch section of the boiler. This can provide a direct 60° or 90° field of view. 2. Installation through the door, viewing the grate upstream at a direct 90° field of view. 3. Installed through port or door above the burners, affording a right angle 55° field of view. 4. Installed through the door, with a direct 60° or 90° field of view, to view down into the ash hopper to look for bridging.
16.4.4 Boilers Furnace cameras are now utilized in thousands of process industry boilers, which make steam to heat the product being processed. The cameras monitor the large fireboxes where air and fuel are mixed for the combustion that heats water in drums, tubes, and headers. Viewing inside the boiler, they help reduce pollution and improve safety. Combined with other techniques, cameras can help reduce pollution as much as 30%. Furnace cameras monitor such boiler conditions as: • Potential flame impingement on the boiler tubes, which can burn holes in water walls • Tube leaks
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
• Confirming ignition (combustion light-off) • Flame stability, allowing operators to optimize fuel/air ratios for maximum burning efficiency • Buildup of slag and ash in the boiler • Smoke and other overfire conditions High temperature cameras can be installed in many locations on the different types of boilers. A typical location for a furnace camera in a front wall-fired boiler would be at centerline height through the opposite wall from the burner grid, with a direct 90° field of view. An opposing wall-fired boiler would be monitored via four camera systems, one in each corner (port or door) viewing the opposite wall. A corner-fired boiler might utilize a camera (direct 60° or 90° field of view) in a port above the burners in the nose-arch section of the boiler, or through the wall above the burners (if elongated tube bends are available). A probe with a right angle lens, as explained earlier, can go through an opening in a waterwall and then look down from above to spot leaks or slag. This is much more cost-effective than installing a tube bend arrangement to allow the probe to penetrate the boiler tubing network. A right angle probe can also be rotated in a portable mode to look for such problems as steam leaks or slag accumulation on the sidewall of the boiler. When burners are located directly below the viewing door, a right angle probe can look down on the burner port to view flames emanating from the burners. Furnace cameras make it possible to quickly adjust combustion in the boiler, minimize the amount of unburned hydrocarbons that downstream equipment needs to clean up and maximize the energy producing steam. Costing $12,000 or less, the cameras are the least-expensive, pollution-control devices for such combustion. Cameras can also complement other equipment to show combustion efficiency in a secondary overfiring air zone, where overfire air is added to achieve secondary combustion, which then burns off residual hydrocarbons, thus allowing adjustments for better fuel economy, where necessary. 16.4.5 Cement Kilns A furnace camera is typically mounted in the front (discharge) end of each rotary kiln—perhaps 520 feet long and 15 feet in diameter and slowly rotating during operation—which heats limestone in temperatures up to 3500°F to produce the powder that is a principal ingredient for cement. Viewing color monitors, operators can adjust the coal feed (fuel), clinker (fused matter), cooling rates, and other variables (see Figure 16.8) [4].
363
High Temperature Cameras
16.4.6 Biomass Furnace cameras ensure that complete combustion is taking place as tons of waste move through combustion zones on a traveling grate. The cameras monitor how uniformly the materials—wood waste, scrap, bark, turkey droppings, walnut shells, almonds, sunflower husks, and other matter—are being burned. They spot large clumps that aren’t dispersed properly, leading to incomplete combustion and smoke. As with trash-to-stream, operators want a nice even distribution of biowaste on the grate so that it burns uniformly. This application area is growing as the U.S. government encourages burning waste streams of natural material since this has the same effect as the material degrading naturally. 16.4.7 Steel Industry
Figure 16.8 Camera retracted in front (discharge) end of cement kiln.
The furnace lens provides a 90° or 60° field of view in the same direction as the flame being blown from the burner. Operators can thus watch the complete flame (including length, shape, color, and direction) on the end of the pipe 12 feet away, the sidewalls, top, and bottom of the kiln, the clinker being formed, and the movement of material down the kiln. The camera’s quartz lens can see 15–25 feet into the kiln, in contrast to the old vidicon cameras, which could only provide a hazy image over a distance of a few feet. Observing their remote screens, operators can adjust for the most efficient combustion—with pulverized coal igniting at the tip of the feed pipe. The furnace lens, flush with the interior wall of the lime kiln, provides a 90° or 60° field of view. The operator can view directly along the axis of the scope. Likewise, he or she can operate a motorized iris and spot filter in the lens to adjust the amount of light reaching the camera. The camera, including the lens, is cooled in the harsh environment by compressed air from a mechanical air conditioner that circulates under positive pressure to cool the camera housing. Compressed air from a separate source goes through the protective lens tube and out its end. An air filtration system removes aerosols, vapor, oil, and particles as small as 0.03 microns. Automatic retractors pull the tube and camera back about 2 feet if the kiln shuts down. Additional cameras can be mounted in the clinker-cooler area to monitor “bridging,” where material could accumulate and block removal of the cooled limestone.
© 2011 by Taylor and Francis Group, LLC
Furnace cameras help attain the primary purpose of watching the flow/alignment of material moving though the furnace on a conveyor. Using the cameras remotely, operators make sure the conveyor is not malfunctioning, causing the material to pile up and become misoriented, with the result that it can’t be removed from the furnace. Operators monitoring the process can make sure pieces don’t hit the sidewall, ceiling of the furnace, or other pieces. Cameras monitor and record material flow processes in reheat furnaces, vacuum degassers, continuous casters, blast furnaces, electric arc furnaces, continuous annealers, galvannealers, tunnel furnaces, and other operations. Though the main interest is material flow, furnace cameras also monitor flames in the steel furnace. The cameras must be rugged and durable to operate in this industries demanding environment, where temperatures can reach over 3000°F. A water-jacketed lens assembly is necessary in place of the air-cooled assembly and CCTV camera housing in the highest temperature ranges (reaching 3800°F). Air purging of the lens system prevents fouling by deposition. Operators at remote stations can also spot bad, irregular pieces of steel, such as blisters with voids in the sheet caused at the reduction mill. 16.4.8 Reheat Furnaces Cameras observe slabs or billets as they are set on “walking beams” to be moved through reheat furnaces. They also monitor the center of the furnaces for proper distance between slabs and to be sure slabs are held properly before they are lifted from the furnace and placed on the rolling or forge line. In these applications, a lens with a forward-oblique 45° direction of view often furnishes a useful perspective as it views the material within the furnace (see Figure 16.9).
364
Camera 2
Air ventilator Camera 1
Camera 3
Figure 16.9 Typical furnace camera coverage of walking beam furnace in steel industry. (“High-Temperature Furnace Camera Systems,” Lenox Application Solutions in the Steel Industry (reference sheets with diagrams for reheat furnaces, vacuum degas-sers, remelt/reverberatory furnaces).)
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
Typical furnace camera coverage in a walking beam furnace
High Temperature Cameras
365
16.4.9 Vacuum Degassers
hours. By reducing the time and costs of bringing the boilers online, cameras also allow greater flexibility for power plants to respond to changes in demand [1]. Cameras are being used to observe burners during combustion tuning in conjunction with such NOx reduction techniques as low excess air (LEA), burners out of service (BOOS), and separated overfire air (SOFA). Balancing the combustion process with the aid of cameras (monitoring the impacts of combustion changes and establishing more uniform combustion) can help reduce emissions as much as 20% [5].
Cameras are mounted through the tank cover or hood of vacuum degassers to observe secondary refining of molten steel to remove oxygen, hydrogen, and nitrogen. They also observe vessel preheating, material being added from the alloy hopper, stirring, refractory condition, surface slag, and the color of the metal. 16.4.10 Remelt/Reverberatory Furnaces Cameras remotely monitor and record, in real time, the push of the batch charge into the hearth and observe melt-surface conditions as flames are blown over the metal during the smelting, refining, or melting process. 16.4.11 U.S. Navy Cameras are monitoring boilers on U.S. Navy ships where safety is the primary concern. The cameras are used to check that unburned fuel is not collecting on the floor of the boiler. Initially employed on the USS Constellation to help preclude explosions, furnace cameras have also allowed greater fuel efficiency by observing results of uneven distribution of air flow to oil burners, potentially reducing fuel consumption by 1–1.5%. The USS Essex and three other LHD-class amphibious assault ships utilize furnace cameras to observe flame patterns and identify potential problems in a boiler combustion monitoring system and also monitor a pushbutton igniter-control system. Previously, a sailor monitored the fires through a small porthole, and an engineer in a flame-resistant jacket lit the boilers with a torch. If unburned fuel was in the boiler, explosions could occur. The camera observes a wide field of view, part of which is the floor on the burner grid. A high-temperature lighting system is often necessary in the dark environment [6]. 16.4.12 Pulp and Paper Industry Furnace cameras monitor flames in the boilers that aid the pulping process. They also monitor the process of burning the black liquor (contaminated stream from the pulp) in a furnace to reclaim, fortify, clean, and reuse it after the hydrocarbon has been burned off. A camera views the tar-like “liquor” as it burns while coming down to a bed. Cameras can also monitor movement of material.
16.4.14 Glass Furnaces Cameras monitor the flat glass or float glass process to aid in conditioning the glass and to ensure that the process moves ahead in an orderly fashion. They monitor both the gas-fired furnaces that formulate the glass and the subsequent process in which the long continuous ribbon of molten glass is flattened and conditioned as it is moved along. Furnace cameras also view ceramic welding repair of refractory linings of glass furnaces, withstanding temperatures as high as 3992°F as they monitor the repair as the furnace is operating [3]. Operators can also remotely view such things as alternating firing of burners on each side of the furnace. They actually want the flames to contact the surface of the material. During the glassmaking process, they can spot “islands” of unmelted silica or other ingredients. In the plant, the operators will put a line across the monitor with a grease pencil to watch that the unmelted material does not move beyond a certain point. If the camera shows that it does move beyond this point, they need to investigate and find out what is happening. 16.4.15 Chemical Incineration Cameras allow operators to make sure that solvents and other liquid chemical streams, as well as biological warfare agents, pharmaceutical waste, and other chemicals achieve maximum uniform combustion as they are incinerated inside a kiln or boiler in an appropriate EPA-approved manner. In some industrial applications, companies pay firms to remove the residue and burn it. Watching the process assures compliance with increasingly stringent regulations.
16.4.13 Electric Utilities Cameras ensure that boilers operate at peak efficiency. They also help reduce NOx emissions. By monitoring burners, they enable operators to quickly confirm proper ignition, and identify flame instability, incomplete combustion, or poor flame distribution. This can shorten troubleshooting and start-up time by several
© 2011 by Taylor and Francis Group, LLC
16.5 Summary: Observing Combustion Problems Just as doctors use medical imaging, like x-rays and MRIs to observe and diagnose conditions, likewise an
366
industrial facility can utilize visual access techniques to correctly diagnose what is happening in a furnace. Using a furnace camera high-temperature visual system, engineers in a control room can observe the situation and make adjustments to optimize operations. To use another analogy, just as an automobile must run correctly to obtain maximum fuel efficiency, so too will an industrial facility obtain fuel economy by optimizing combustion. Furnace cameras provide immediate visual information, confirming the results of adjustments, and crosschecking with other instrument systems in the control room that show such things as too much smoke in the chimney. Typical problems that can be observed before taking corrective action for greater cost-efficiency: 16.5.1 Flame Impingement Camera approach: Visually observe the waterwall or refractory wall directly to be certain flames are not touching the walls, causing potentially catastrophic leaks at sites like crude oil (fired) heaters. Besides spotting flames contacting the external tube surface inside the firebox, the camera can also spot other signs of impingement like tubes with a cherry-red color or bulges in the tube walls [12]. 16.5.2 Boiler or Ash Hopper Slag Camera approach: View waterwall directly to detect slag buildup in early stages. (See Figure 16.10 showing boiler slag.) 16.5.3 Defective Ignition Camera approach: View igniter as fuel is introduced during light-off. Also observe flames emanating from
Industrial Combustion Testing
burners, watching flame shape, color, length, possible detachment, and burner “eyebrows.” 16.5.4 Unbalanced Combustion Camera approach: Using different directions of view and fields of view, observe flame distribution throughout furnace in order to “tune” (properly balance) combustion process, potentially saving thousands of dollars per year and helping comply with stringent emissions standards [12]. 16.5.5 NOx Control Camera approach: Observe burners during tuning to uniformly balance combustion process. High flames (and temperatures) may indicate defective burners, necessitating a change of burners to reduce both the flames and the rate of NOx production. Also observe the color of the flames [7]. One of the largest producers of electricity in the United States has reduced NOx emissions to the second lowest in the country, aided by furnace cameras that monitor five gas-fired boilers. The cameras allow the plant to conduct precise combustion tuning, reducing the 30-day rolling average for NOx emissions by more than 30% [1]. Operators also monitor flame distribution in the boilers and adjust combustion as necessary. This information can be used in conjunction with other NOx reduction equipment. Cameras can monitor and detect the effect of systems that introduce too much oxygen into the combustion process, creating more NOx. Cameras also decrease start-up time by several hours because they confirm burner and igniter flames and allow operators to quickly identify faulty flame-scanner indications. 16.5.6 Steam Leaks Camera approach: Directly observe the waterwall to find steam leaks and qualify them to determine whether or not the boiler can continue to operate until the scheduled outage, avoiding downtime that could cost hundreds of thousands of dollars. 16.5.7 Waste Hot Spots Camera approach: Observe material, such as trash in waste-to-energy plants, as it moves through burn zone on traveling grates or stoker. Camera can detect overheat situations close to boiler tube walls. 16.5.8 Smoking Conditions
Figure 16.10 Camera view of slag in boiler.
© 2011 by Taylor and Francis Group, LLC
Camera approach: View overfire air (as well as stack) to detect smoke, and then properly tune combustion.
367
High Temperature Cameras
to the swirl, recirculation, jet descriptions, and mixing rates for each type [12]. 16.5.11 Flashback Camera approach: Camera can observe flame in the venturi or mixer, a condition that may occur when the pressure is too low and the fire flashes back into the pipes of a boiler or steel furnace [12].
16.6 The Future
Figure 16.11 Smoky flames result from incomplete combustion.
Long smoky flames result from incomplete combustion and can consume far more fuel than is necessary. These “lazy” flames may appear hazy rather than bright and clear. They produce unburned carbon, which is not costeffective. When the operator sees this condition, he or she can provide more air to each burner and then ensure that this air mixes quickly and completely with the fuel to achieve rapid combustion. (See Figure 16.11: flames in smoking condition.) 16.5.9 Flame Lift-Off Camera approach: Monitor flame patterns inside firebox. First indication of lift-off is that the flame is detached from one or two burners [12]. 16.5.10 Pulsating Flame Camera approach: Visual inspection of the flame pattern can see that flames are pulsing and changing in length and volume—a nonsymmetrical pattern from one burner to another. Operators reduce firing rate, supplying sufficient oxygen so that combustion can be completed. Operators can then confirm that these steps have been effective. There are many types of industrial flames, ranging from conventional forward shapes to “headpins,” balls and cones, to long luminous lazy flames. Furnace camera operators learn to recognize these and relate them
© 2011 by Taylor and Francis Group, LLC
Furnace camera technology, while meeting today’s needs, continues to evolve and progress as it incorporates the latest advances. Until recently, the state of the art in these cameras was 480 lines of resolution. Recently, 540-line-resolution cameras became available, with better color rendition and more sophisticated electronics. Light volume control, allowing users to adjust the amount of light to see the different areas and depths within the furnace in a bright application, continues to be very critical and is very advanced. Water-cooled furnace cameras are capable of meeting future high temperature needs reaching 4500°F in some rare applications. Furnace temperatures aren’t expected to exceed this point in the foreseeable future. To surpass 4500°F would require going beyond a lot of material availability in terms of refractories. It might be beyond the point of diminishing returns. Only a very exotic process would require a higher temperature. Thermal imaging, using infrared technology, is currently being developed. It will meet specialized needs. This technology will provide a thermal image as well as temperature acquisition that is accurate and traceable to advanced standards. One of the first throughwall technologies for thermal imaging is currently being developed. It goes through the furnace wall and spreads the fields of view. This is a major advance over a camera that stays outside the furnace and looks through a viewing port, only viewing a small area. One of the hopes of infrared technology is that it will observe better in hostile environments where the air is smoky, or steamy, filled with many small pieces of material. It is possible that at some point, furnace camera systems will be redesigned to be smaller. New significantly smaller units have recently been designed for steel applications, but this is a refinement more than a technological leap. Industry has generally been slow to realize the costsaving potential of furnace cameras. Though these
368
Industrial Combustion Testing
needs for cost-efficiency and quality control through the decades ahead (Figure 16.12).
References Figure 16.12 (See color insert following page 424.) Immediate information on current furnace conditions appears on monitors of furnace camera system. Operators can take corrective action when necessary.
systems are known and utilized in utilities and steel facilities, people running furnaces in other industries often don’t take advantage of them—or aren’t even aware of them. There is definitely an awareness gap despite their huge potential cost efficiencies. As awareness grows with the increased emphasis on fuel economy, process control, and the environment, furnace cameras can be expected to be more widely utilized.
16.7 Conclusion Furnace camera systems provide close vision where incomplete vision, or no vision, hampered operation in the past. These cameras are a very effective tool aiding industries in the use of combustion. They ensure the most cost-efficient combustion, provide advance warning of problems, increase safety, and reduce pollution by monitoring burners, flames, waterwalls, and other locations. Furnace cameras are expected to meet growing
© 2011 by Taylor and Francis Group, LLC
1. McCarty, B. C., and Lang, W. J. “Furnace Cameras Assist in NOx Reduction.” Power Engineering 106, no. 11 (November 2002): 138–40. 2. Lang, W. J. “A Tool for Maintaining Good Combustion.” Solid Waste Technologies, May/June, 1996. 3. Lang, W. J. “Ceramic Welding Used with High Temperature CCTV in Advanced Repair Technique for Glass Furnaces.” Industrial Heating 39 (May 1998): 51–52. 4. Prokopy, S. “Color TV Gives a Clearer Picture from Inside the Kiln.” Rock Products Cement Edition 99, May 1996: 38–39. 5. “Boiler Camera System Reduces Emission by As Much As 20%,” Power Magazine, November 2002. 6. Jean, G. V. “Navy Retrofitting Ships with Fuel-Saving Technologies.” National Defense Magazine, September 2008. 7. Mark McMeilly. “Video System Shows Inside of Turbine Compressor.” Diesel & Gas Turbine Worldwide 32, September 1999. 8. Robert Hoffman. “Monitoring Process Heating at a Refinery (Application Photos and Explanation).” Lenox Insight Publication, October 30, 2006. Lenox Publication. 9. Raymond Feuerhammer “High-Temperature Furnace Camera Systems,” Lenox Application Solutions in the Steel Industry (reference sheets with diagrams for reheat furnaces, vacuum degassers, remelt/reverberatory furnaces). Lenox Publication. 10. Lenox Staff. “FireSight High-Temperature Remote Viewing Systems.” Lenox Instrument Company brochure with photos and diagrams of system and components. Lenox Publication. 11. Raymond Feuerhammer “Potential Savings for Bottomline Energy Efficiency.” Lenox Instrument Company reference table. Huntingdon Valley, PA. 12. Baukal, C. E., ed. The John Zink Combustion Handbook. Boca Raton, FL: CRC Press, 2001.
17 Liquid Fuel Atomization Testing Khaled A. Sallam Contents 17.1 Introduction.................................................................................................................................................................. 369 17.2 Patternation................................................................................................................................................................... 369 17.3 Laser Diffraction.......................................................................................................................................................... 370 17.4 Laser Doppler Technique............................................................................................................................................ 370 17.5 Image Processing of Sprays........................................................................................................................................ 371 17.6 Holography................................................................................................................................................................... 371 17.7 X-Ray Absorption......................................................................................................................................................... 372 References..................................................................................................................................................................................374
17.1 Introduction For oil firing, the burner is designed to first atomize the liquid fuel before it is burned. For atomizing liquid fuels, two kinds of atomizers are generally used: pressure atomizers and air-assisted (twin-fluid) atomizers (e.g., aerated fuel injector shown in Figure 17.1). The breakup of liquid jet or sheet into drops is termed primary breakup and the subsequent breakup of drops into even smaller droplets are termed secondary breakup. The interaction between the turbulence generated inside the injector passage and the aerodynamic forces acting on the liquid upon exiting from the injector controls the outcome of the primary breakup. The formation of a cavitation bubble inside the injector resulting from sharp corners and its subsequent collapse would also influence the atomization of liquid jets. Sauter mean diameter (D32) is reported frequently for fuel injectors because it is representative of the droplet size that has the same volume/surface area ratio as the whole spray. It is given by:
D32 = Σi (N(Di) Di3)/Σi (N(Di) Di2),
(17.1)
where the Di and N(Di) are the droplet size and the number of droplets associated with that size, respectively. Many fuel atomizers generate polydispersed spray and therefore are usually characterized by a probability distribution function (pdf) rather than a simple average diameter. Different distribution functions are popular within the spray community. Among them are Rosin-Rammler distribution, log-normal distribution,
and Simmons’s (1977) universal root normal distribution shown in Figure 17.2. Additionally the intact core length of the jet or sheet, the spray mass flux, and the cone angle of the spray are among the variables used to characterize the effectiveness of the injector design. The droplets size distribution depends on the geometry of the atomizer as well as its operating conditions and how clean its passage is. To test the performance of a fuel atomizer many diagnostics can be used. The next sections summarize a few of these methods focusing on optical methods (Greated, Cosgrove, and Buick 2002), which have the advantages of being nonintrusive.
17.2 Patternation In its earliest form the mechanical patternator consisted of collecting rings used to measure the cone angle of the spray by draining the liquid collected in each ring and weighing it. High resolution mechanical patternators can provide the two-dimensional mass flux of the spray on a plane normal to the injection direction by collecting the liquid into large number of cells, for example, 625 (Hung et al., 2008). The volume of liquid in each cell is used to calculate the mass flux distribution on the plane. The spray angle can then be calculated from the mass flux measurements. However, the drop sizes can not be obtained and the method is inherently intrusive. Optical patternation is a nonintrusive technique based on extinction tomography utilizing diode lasers to form laser sheets and linear photo diode arrays as 369
© 2011 by Taylor and Francis Group, LLC
370
Industrial Combustion Testing
Air
sensors. The system is contained within a ring that allows the spray to be injected in the middle. The extinction measurements yield the local absorptances or simply local surface area densities. An example of such a device employing path-integrated transmittances measurements from six view angles and 256 parallel paths at each view angle was reported by Lim and Sivathanu (2005).
90°
Flow
do
L
100°
25
17.3 Laser Diffraction The laser diffraction method (e.g., Malvern analyzer) is among perhaps the most widely used technique for spray diagnostics. It is a line-of-sight method to measure the droplet size (Li, Nishida, and Hiroyasu 2004) by observing the diffracted light due to the presence of droplets in the laser probe (e.g., 10 mm diameter laser beam).
Liquid
Gas Figure 17.1 Schematic of an aerated fuel injector. (From Sallam, K. A., Aalburg, C., Faeth, G. M., Lin, K.-C., Carter, C. D., and Jackson, T. A., Atomization and Sprays, 16, no. 6, 657–72, 2006.)
2.00
MMD/SMD = 1.2
1.50
d/MMD
1.00 GLR q
0.50
8 8 8 8(a) 8(a) 4(a) 4(a)
0.25 0.10
0.74 0.74 0.74 0.74,4 0.74,4 0.74,4 0.74,4
x/d0 d0 Sym. 0 9 18 25 50 25 50
1 1 1 0.5,1 0.5,1 0.5,1 0.5,1
(a) From Miller et al. (2007) 0.00 0.01
0.2
2
20 50 80 Cumulative volume (%)
98
99.8 99.99
Figure 17.2 Droplet size distribution plotted on a root-normal scale. (Reprinted from Lee, J., Sallam, K. A., Lin, K.-C., and Carter, C. D., Journal of Propulsion and Power, 25, no. 2, 258–66, 2009b. With permission from American Institute of Aeronautics and Astronautics.)
© 2011 by Taylor and Francis Group, LLC
17.4 Laser Doppler Technique The Phase Doppler interferometry (PDI) together with laser diffraction are among the most popular spray diagnostics. Phase Doppler interferometry (also known as PDA, or PDPA) is an extension of laser Doppler velocimetry to measure the droplet velocity and size, number density, and volume flux (Widmann 2002). It uses a continuous laser beam (e.g., Argon-Ion) that is first split into two beams and then the two beams are intersected to form the measuring volume as shown in Figure 17.3. When a droplet crosses that volume the light scattered from it is collected into two sensors. The phase difference between the two collected signals is used to calculate the size of the droplet (assuming a spherical droplet) and the frequency of the signal is used to calculate the velocity of the droplet crossing the measuring volume. Nonspherical droplets typical of those in the near-injector region would result in an erroneous signal (Widmann, Presser, and Leigh 2002). Two dimensional velocities are possible by intersecting four beams, each two are of the same color, with a total of two colors (e.g., λ = 532 nm and 488 nm) as shown in Figure 17.4. Velocity measurements from –25 to 125 m/s and particle sizing up to 145 µm using a 2D PDA device were reported by Dunand, Carreau, and Roger (2005). Three-dimensional velocities are possible by intersecting six laser beams (each two of the same color, with a total of three colors). The measurements are conducted for one droplet at a time (single-particle technique) and the technique is fully automated. The existence of multiple particles in the measurement volume (Roisman and Tropea 2001) and
371
Liquid Fuel Atomization Testing
5 mm
Figure 17.5 Pulsed shadowgraph of aerated liquid jet in subsonic crossflow. (Reprinted from Lee, J., Sallam, K. A., Lin, K.-C., and Carter, C. D., Journal of Propulsion and Power, 25, no. 2, 258–66, 2009b. With permission from American Institute of Aeronautics and Astronautics.) Figure 17.3 Two laser beams used for setting PDA. (Courtesy of Dantec Dynamics.) X
U
Z
ϕ V Y
Scattering plane
V U
Figure 17.4 Sketch of Dual PDA. (Courtesy of Dantec Dynamics.)
burst splitting events, when noisy environment causes over counting of droplets (Widmann, Presser, and Leigh 2001), result in errors in the mass flux measurements.
line-of-sight diffraction methods. Illumination technique is typically based on backlighting (e.g., shadowgraphy; Sallam, Dai, and Faeth 1999). A high-speed digital camera coupled with high repetition laser (e.g., copper vapor) are used to provide time-resolved visualization of the spray (e.g., 10 kHz). This technique is simple and particularly suited to measure nonspherical droplets typical of those in the near-injector region. Special attention should be paid to off-focused droplets in the image that need to be eliminated using robust image processing algorithms with multiple in-focus criteria (Kob, Kim, and Lee 2001). Also heavily overlapped droplets are problematic when measuring droplet sizes. If the overlapped droplets are spherical in shape then the application of Hough transform could be useful (Kim and Lee 2002). The velocity measurements are obtained by double flash illumination as shown in Figure 17.6. To overcome the limited depth of field of the image frame, a scan across the optical axis can be made at fixed intervals. This however would not yield a true size distribution because some of the large droplets would appear in multiple frames due to their large depth of focus. Three-dimensional imaging (i.e., holography) can overcome this shortcoming.
17.5 Image Processing of Sprays Droplet sizes are measured directly from frozen images of the spray, as shown in Figure 17.5, using high magnification long distance lens and a pulsed light source (e.g., Nd-YAG laser or cheap flash lights). Photography is especially beneficial for measuring sizes in the near injector region where large droplets may escape detection by other point-based techniques such as PDPA or
© 2011 by Taylor and Francis Group, LLC
17.6 Holography Holographic diagnostic of sprays eliminates the depth of field correction problem of conventional photographic particle sizing techniques by acquiring
372
Industrial Combustion Testing
three-dimensional images on holographic plates (Li, Nishida, and Hiroyasu 2004). holography is radically different from shadowgraphy. In addition to the wave amplitude, which is the only information provided on direct photographs, holography records the phase of the incident light wave. This recorded phase and amplitude information can then be used to create an image of the objects in the measuring volume during a reconstruction process so that an image of each drop can be focused at its proper position relative to other drops in
Figure 17.6 Double-pulse shadowgraph along the surface of an annular free jet, uo = 20.6 m/s, Δt = 40 µs, x/dh = 8.5, diameter of reference pin = 3.2 mm. (Sallam, K. A., “Properties of Spray Formation by Turbulent Primary Breakup,” PhD thesis, University of Michigan at Ann Arbor, 2002.)
the field. The illuminating beams are produced typically by Nd-YAG laser with short laser pulses (e.g., 7 ns) to essentially stop the motion of all drops in the flow. Off-axis holography can provide a good resolution and sharp drop images in dense spray fields compared with in-line holography where any individual drop image is surrounded by light scattered from out-of-focus images. However, with off-axis holography, the reference beam must be separated from the object beam to avoid overlapping virtual and real images. This is done by splitting the laser pulse into two beams, the reference and object beams. The object beam passes through the spray field. The reference beam is directed through a tube to avoid optical noise from small drops produced in the spray. The reference and object beams are then optically mixed to form a hologram on the photographic plate (e.g., AGFA holographic plates with a 100 mm × 125 mm film format). The optical penetration properties of the holograms can be improved by reducing the diameter of the object beam through the flow. Digital holography eliminates the use of wet developing process of holographic plates and hence makes the technique more popular (Miller et al. 2008; Lee, Miller, and Sallam 2009a; Lee et al. 2009b). The in-line setup is typically used with digital holography because of the relatively low resolution of current CCD sensors compared to holographic plates used in traditional holography. Typical optical setup of digital holography is shown in Figure 17.7. With double pulsing, velocity information can also be obtained as shown in Figure 17.8.
17.7 X-Ray Absorption X-rays can penetrate easily in low atomic numbered materials, which makes it useful for spray diagnostics in the dense near-injector region. Moreover X-rays
Spatial filter (15 µm pinhole) Polarized beam splitter
Plate beam splitter
Collimating Object lens Relay lens
CCD Camera
Nd-YAG Test section Objective lens Nd-YAG
λ/2 plate Mirror
Figure 17.7 Digital double-pulsed in-line holography (DIH) setup. (From Lee, J., Miller, B., and Sallam, K. A., Atomization and Sprays, 19, no. 5, 445–56, 2009a.)
© 2011 by Taylor and Francis Group, LLC
373
Liquid Fuel Atomization Testing
D(µm):
20
52
84
116
148
180
u(m/s):
–20
–10
0
10
20
30
can be used to investigate the inner geometry of the injector and bubbles flow inside the liquid phase due to cavitation as shown in Figure 17.9. X-ray absorption techniques (Yue et al. 2001) have been used successfully to measure the mass distribution of fuel sprays close to the nozzle using a monochromatic, synchrotron X-ray beam as the one available at the
Aerated injector 0
y/d0
10 15
Jet direction
5
Crossflow
20 25 30 0
5
10
15 X/d0
20
0 25 –1
0 Z/d 0
10
Figure 17.8 (See color insert following page 424.) Drop size distribution and three-dimensional velocities near injector region. (Reprinted from Lee, J., Sallam, K. A., Lin, K.-C., and Carter, C. D., Journal of Propulsion and Power, 25, no. 2, 258–66, 2009b. With permission from American Institute of Aeronautics and Astronautics.)
Figure 17.10 X-ray diagnostic experimental setup at Argonne National Lab.
© 2011 by Taylor and Francis Group, LLC
Figure 17.9 X-ray image of liquid jet breakup in still air. The dark circles are droplets whereas the light circles are bubbles within the liquid jet. (From Osta, A., Lee, J., Sallam, K. A., and Fezzaa, K., Proceedings of 11th International Conference on Liquid Atomization and Spray Systems, Vail, CO, July 26–30, 2009.)
374
Advanced Photon Source at Argonne National Lab shown in Figure 17.10. The problem with conventional X-ray sources is the drastic reduction in beam intensity when time-resolved measurements are required. Using tomographic analysis, three-dimensional reconstruction of the dynamics of the spray can be obtained (Cai et al. 2003).
References Cai, W., Powell, C. F., Yue, Y., Narayanan, S., Wang, J., Tate, M. W., Renzi, M. J., Ercan, A., Fontes, E., and Gruner, S. M. “Quantitative Analysis of Highly Transient Fuel Sprays by Time-Resolved X-Radiography.” Applied Physics Letters 83, no. 8 (2003): 1671–73. Dunand, A., Carreau, J. L., and Roger, F. “Liquid Jet Breakup and Atomization by Annular Swirling Gas Jet.” Atomization and Sprays 15, no. 2 (2005): 223–47. Greated, C., Cosgrove, J., and Buick, J. M. Optical Methods and Data Processing in Heat and Fluid Flow. London, UK: Professional Engineering Publishing, 2002. Hung, D. L. S., Harrington, D. L., Gandhi, A. H., Markle, L. E., Parrish, S. E., Shakal, J. S., Sayar, H., Cummings, S. D., and Kramer, J. L. “Gasoline Fuel Injector Spray Measurement and Characterization – A New SAE J2715 Recommended Practice,” SAE Technical Paper 2008-011068. Kim, Y. D., and Lee, S. Y. “Application of Hough Transform to Image Processing of Heavily Overlapped Particles with Spherical Shapes.” Atomization and Sprays 12, no. 4 (2002): 451–61. Kob, K. U., Kim, J. Y., and Lee, S. Y. “Determination of In-Focus Criteria and Depth of Field in Image Processing of Spray Particles.” Atomization and Sprays 11, no. 4 (2001): 317–33. Lee, J., Miller, B., and Sallam, K. A. “Demonstration of Digital Holographic Diagnostics for the Breakup of Liquid Jets Using a Commercial-Grade CCD Sensor.” Atomization and Sprays 19, no. 5 (2009a): 445–56. Lee, J., Sallam, K. A., Lin, K.-C., and Carter, C. D. “Spray Structure in Near-Injector Region of Aerated Jet in Subsonic Crossflow.” Journal of Propulsion and Power 25, no. 2 (2009b): 258–66.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
Li, T., Nishida, K., and Hiroyasu, H. “Characterization of Initial Spray from a D.I. Gasoline Injector by Holography and Laser Diffraction Method.” Atomization and Sprays 14, no. 5 (2004): 477–94. Lim, J., and Sivathanu, Y. “Optical Patternation of a MultiOrifice Spray Nozzle.” Atomization and Sprays 15, no. 6 (2005): 687–98. Miller, B., Sallam, K. A., Bingabr, M., Lin, K.-C., and Carter, C. D. “Breakup of Aerated Liquid Jets in Subsonic Crossflow.” Journal of Propulsion and Power 24, no. 2 (2008): 253–58. Osta, A., Lee, J., Sallam, K. A., and Fezzaa, K. “Investigating the Effect of the Injector Length/Diameter Ratio on the Primary Breakup of Liquid Jets Using X-ray Diagnostics.” Proceedings of 11th International Conference on Liquid Atomization and Spray Systems, Vail, CO, July 26–30, 2009. Roisman, I. V., and Tropea, C. “Flux Measurements in Sprays Using Phase Doppler Techniques.” Atomization and Sprays 11, no. 6 (2001): 667–99. Sallam, K. A. “Properties of Spray Formation by Turbulent Primary Breakup.” PhD thesis, University of Michigan at Ann Arbor, 2002. Sallam, K. A., Aalburg, C., Faeth, G. M., Lin, K.-C., Carter, C. D., and Jackson, T. A. “Primary Breakup of AeratedLiquid Jets in Supersonic Crossflows.” Atomization and Sprays 16, no. 6 (2006): 657–72. Sallam, K. A., Dai, Z., and Faeth, G. M. “Drop Formation at the Surface of Plane Turbulent Liquid Jets in Still Gases,” International Journal of Multiphase Flow 25 (1999): 1161–80. Simmons, H. C. “The Correlation of Drop-Size Distributions in Fuel Nozzle Sprays.” Journal of Engineering for Power 99, no. 3 (1977): 309–19. Widmann, J. F. “Characterization of a Residential Fire Sprinkler Using Phase Doppler Interferometry.” Atomization and Sprays 11, no. 1–3 (2002): 69–90. Widmann, J. F., Presser, C., and Leigh, S. D. “Extending the Dynamic Range of Phase Doppler Interferometry Meas urements.” Atomization and Sprays 12, no. 4 (2002): 513–37. Widmann, J. F., Presser, C., and Leigh, S. D. “Indentifying Burst Splitting Events in Phase Doppler Interferometry Measurements.” Atomization and Sprays 11, no. 6 (2001): 711–33. Yue, Y., Powell, C. F., Poola, R., Wang, J., and Schaller, J. K. “Quantitative Measurements of Diesel Fuel Spray Characteristics in the Near-Nozzle Region Using X-Ray Absorption.” Atomization and Sprays 11, no. 4 (2001): 471–89.
Section III
Burner Testing
© 2011 by Taylor and Francis Group, LLC
18 Process Burners Jeffrey Lewallen, Thomas M. Korb, Jaime A. Erazo, Jr., and Erwin Platvoet Contents 18.1 Introduction.................................................................................................................................................................. 377 18.2 Types of Industrial Burners........................................................................................................................................ 378 18.3 Process Burner Applications...................................................................................................................................... 379 18.3.1 Refining Industry........................................................................................................................................... 380 18.3.2 Reforming Industry....................................................................................................................................... 381 18.3.3 Petrochemical Industry................................................................................................................................. 382 18.4 Process Burner Testing................................................................................................................................................ 383 18.4.1 Objectives and Testing Overview............................................................................................................... 383 18.4.1.1 Testing Parameters and Measurements..................................................................................... 384 18.4.1.2 Process Burner Testing Sequence............................................................................................... 384 18.4.1.3 Burner Testing: Benefits............................................................................................................... 384 18.4.1.4 Burner Testing: Drawbacks......................................................................................................... 384 18.4.1.5 Burner Testing Versus CFD......................................................................................................... 384 18.5 Process Burner Testing Equipment and Methodology.......................................................................................... 385 18.5.1 Test Burners.................................................................................................................................................... 385 18.5.2 Test Furnaces.................................................................................................................................................. 386 18.5.3 Air Delivery Systems..................................................................................................................................... 387 18.5.4 Instrumentation and Control....................................................................................................................... 388 18.5.4.1 Fuel Flow and Composition . ..................................................................................................... 389 18.5.4.2 Flue Gas Analysis......................................................................................................................... 389 18.5.4.3 Flue Gas Temperature and Pressure.......................................................................................... 390 18.5.5 Special Equipment......................................................................................................................................... 390 18.5.5.1 Heat Flux........................................................................................................................................ 390 18.5.5.2 CO Probe........................................................................................................................................ 390 18.5.5.3 Noise............................................................................................................................................... 391 18.5.5.4 Unburned Hydrocarbons, Particulate Matter, and Oxides of Sulfur................................... 391 18.5.6 Test Fuels Selection........................................................................................................................................ 392 18.5.7 Test Procedure................................................................................................................................................ 393 18.6 Conclusion.................................................................................................................................................................... 393 References................................................................................................................................................................................. 393
18.1 Introduction A burner is a device that mixes and burns fuel and an oxidizer in a controlled manner to produce heat used to generate steam or heating air. In most cases, the oxygen present in the air serves as the oxidizer for the combustion process. The combustion process depends on a careful balance of four parameters: heat, oxygen, fuel, and chemical chain reactions. Figure 18.1 illustrates the
parameters required for combustion to occur. A welldesigned burner will maintain this balance over a wide range of operating conditions. This chapter will discuss full-scale testing of industrial process burners. The distinction between a process burner and other types of burners such as boiler burners, oxy-fuel burners, and so on is the application for which the burner is intended. The definition of process burners and the industries where they are used will be described in more detail following a brief discussion of general burner designs.
377 © 2011 by Taylor and Francis Group, LLC
378
Re ac tio
nch ain
Industrial Combustion Testing
He
at
Fuel Air fuel mixture
n
e yg
Ox Air
Air
Figure 18.1 Fire tetrahedron showing the four necessary conditions required to sustain combustion.
18.2 Types of Industrial Burners Burners are typically classified under two general categories, premix and raw gas (also known as nonpremixed or diffusion). A premix burner mixes the fuel and oxidizer together before the constituents enter the combustion zone. A common example of a simple premix burner is the Bunsen burner. A schematic of a premixed Bunsen burner is shown in Figure 18.2. The venturi is designed to entrain air with the fuel from the fuel orifice. An air control register is used to adjust the air flow rate. The fuel air mixture flows through the venturi to an exit tip that is designed to shape the flame to the desired geometry. An advantage of the premix burner is that the burner can be designed such that the air required for combustion can be entrained into the burner with the fuel source. The main disadvantage of premix burners is the limited turndown in firing rate. Turndown is the ratio of the maximum firing rate to the minimum firing rate. Depending on the fuels utilized, when fuel flow rate to a premix burner is reduced the exit velocity of the fuel–air mixture can drop below the flame speed allowing the flame to propagate back through the tip damaging the burner. The term “flashback” is used to describe this situation. Flashback occurs when the flame speed exceeds the velocity of the fuel–air mixture at the tip of the burner. Fuels such as hydrogen and acetylene have high flame speeds relative to other hydrocarbons such as methane and are therefore more prone to flash back. A raw gas burner keeps the air and fuel separate until they are injected into the combustion chamber or furnace.
© 2011 by Taylor and Francis Group, LLC
Fuel
Figure 18.2 Schematic of premix Bunsen burner.
A generic raw gas burner is shown in Figure 18.3. An advantage of a raw gas burner is the high turndown. Depending on the burner design, the turndown of a raw gas burner can be as much as three or four times the turndown of a premix burner. Another advantage of a raw gas burner is that it cannot flashback like a premix burner since the fuel and air are not mixed prior to injection in the combustion zone. A comparative disadvantage of a raw gas burner to a premix burner is the potential for plugging of the fuel injectors. Typically a premix burner utilizes a single fuel orifice port to maximize the air entrainment efficiency. A raw gas burner typically uses multiple ports and injectors to mix the fuel, air, and flue gas within the combustion zone in a specific manner. Since the fuel pressure is specified by the customer and/or the burner design, the need to inject the gas through multiple ports results in ports that are smaller and more prone to plugging from debris either in the fuel or the fuel delivery system (piping). Despite this disadvantage, raw gas burners provide a great deal of flexibility in fuels that can be burned and arrangement of fuel injectors to produce the desired flame geometry and reduce undesirable air pollutants. The purpose of a burner is to complete what is commonly referred to as the 5 M’s. They are Meter, Mix, Maintain, Mold, and Minimize. Metering of fuel and air is necessary to attain the correct ratio of air and fuel to provide efficient combustion. The flow rate of fuel to the combustion zone is controlled by designing the
379
Process Burners
Burner tile
Fuel jets
Air
Fuel Figure 18.3 Schematic of a conventional raw gas burner.
fuel ports on the gas tips such that a known flow rate is achieved at a given pressure for a given fuel composition. The delivery of air to the combustion zone is controlled in a similar manner except the pressure drop between the burner air inlet and the furnace is taken across the burner damper, plenum, and throat (tile section). In order to initiate combustion, the fuel and air must be Mixed and an ignition source provided. Good burner design allows for stable combustion when the ignition source is removed. In order to Maintain ignition, the fuel and air must be mixed sufficiently to allow the fuel and air to burn such that the local temperature is sufficient for self-sustained combustion. The fourth M is Mold. Molding is the mixing of the fuel and air in order to control the flame geometry and achieve a desired flame envelope. The final M is for Minimizing air pollutants like oxides of nitrogen (NOx), carbon monoxide (CO) and unburned hydrocarbons (UHC).
18.3 Process Burner Applications Many different types of burners are used in industrial and commercial applications. Some examples of burners are boiler burners used to produce steam, duct
© 2011 by Taylor and Francis Group, LLC
burners used in power plants and similar applications to heat air or gas turbine exhaust, gas turbine burners used in gas turbine engines, and kiln burners used in kilns. Burners are also used in thermal oxidizer (incinerator) and flaring applications where the primary purpose is to provide heat to thermally destroy unwanted waste materials. Process burners are a specific category of burner utilized in the refining, reforming, and petrochemical processing industries to provide energy to fluids in a specific process within the plant. In all cases, the burners are installed in fired heaters or furnaces, which are, in simple terms, steel enclosures lined with refractory insulation. For purposes of this discussion, the terms fired heater and furnace will be used interchangeably. The fluid being heated in the process flows through tubes that pass through the hot furnace and are heated by radiation and convective heat transfer. These tubes are often referred to as process tubes or tube coils. In some cases, such as crude oil heaters, the fluid is simply heated to increase its temperature either to reduce the fluid viscosity or to preheat the fluid for subsequent processing in a reactor. In other cases, such as hydrogen reformers or ethylene cracking furnaces, a chemical reaction takes place within the process tubes. Many different fired heater and tube coil designs are used in the processing industries depending on the industry segment and the purpose of the heater. The requirements of the process burner are different for each heater design. As a result, process burners are an engineered-to-order product required to meet widely ranging performance criteria. Although specific burner design and performance criteria are unique to each application, there are a number of burner attributes that are universal and apply to all process burners. These attributes can be broadly categorized as safety, performance, reliability, and maintainability [5]. To operate safely, a burner must be stable and burn all the supplied fuel continuously under all operating conditions for which the burner was designed. The burner must be stable with all specified fuels, at off-design conditions such as high and low excess air, reduced firing rate (turndown), and when operating in a cold heater. The burner must also be designed so that limited maintenance can be performed on a burner while other burners in the same heater are still in operation without putting maintenance personnel or equipment at risk. Generally, online maintenance includes the removal and cleaning of fuel (gas or liquid) injection guns and pilots. Universal burner performance criteria include the ability of the burner to achieve the specified maximum and minimum firing rate at the required conditions, flame dimensions, and emissions compliance (NOx, CO, particulate matter, volatile organic compounds (VOC), and noise). Each industry has its own unique design
380
Industrial Combustion Testing
and performance criteria and challenges. Variables that define and segregate the industries are flame shape relative to the furnace, emissions requirements, type of fuel available, location of the burner on the furnace, proximity to the process tubes or coils, radiant heat flux to the process tubes, location of a burner relative to adjacent burners, combustion air delivery system, furnace temperature, and range of operability. Each of the three segments of the processing industry and process burner requirements unique to each segment are discussed briefly in the following sections. 18.3.1 Refining Industry The refining industry or petroleum refining industry converts crude oil into fuels, specialty chemicals and feedstock for use in making more highly purified chemical components. Gasoline, diesel, and jet fuel are the most commonly recognized products of the refining industry. There are many types of furnaces in a refinery and each has a specific operating range depending on the product to be developed or specific purpose of the furnace. Figure 18.4 includes schematics for just three examples of the many different furnace configurations typically used in refining applications. In all cases, the furnace is comprised of a radiant section
where the dominant mode of heat transfer to the process tubes is radiation, and a convection section at the top of the furnace where the hot flue gases enter the stack. The convection section is often used to preheat the process fluid prior to entering the process tubes or to provide heat to another fluid stream such as superheated steam. Vertical cylindrical (VC) heaters have a circular cross section with the process tubes arranged in a circle around the perimeter. The burners are also arranged in a circular pattern, known as a burner circle, closer to the centerline of the heater. Small VC heaters may utilize as few as 3 burners while larger heaters may have 20 burners installed in one or more burner circles. Cabin and box heaters are rectangular in shape and come in many designs, differing in process tube type and location, burner placement, size, and shape. Figure 18.5 is a photograph of a twin cell box heater. This type of heater is basically two separate radiant sections sharing a common convection section and flue gas stack. Some common challenges specific to refinery applications are relatively low furnace operating temperature, significant variation in fuels, and retrofitting existing equipment to utilize modern low or ultra-low emission burners. Relative to reforming and petrochemical applications, most refinery heaters operate at relatively low
Convection tubes
Convection tubes
Convection tubes
Radiant tubes
Radiant tubes
Radiant tubes
Refractory walls
Refractory walls
Refractory walls
Burners
Burners
Burners
Vertical cylindrical
Cabin heater
Central tube wall (Double fired)
Figure 18.4 Examples of three common refinery heater designs.
© 2011 by Taylor and Francis Group, LLC
381
Process Burners
that are taller than the maximum 66% of radiant section height recommended by API 560 [2]. It can also be difficult to fit the new burners in the original mounting locations due to the larger size and spacing requirements of low emissions burners. If the low emissions burners are installed too close together, the flames may interact, resulting in increased emissions and flame length. While full-scale burner testing can be used to measure flame length and diameter, multiburner testing and/or single burner testing used in conjunction with computational fluid dynamics (CFD) modeling may be required to evaluate flame interaction effects. 18.3.2 Reforming Industry Figure 18.5 Photograph of twin cell box heater.
temperatures, 1400–1600°F (760°C–870°C). Depending on the type of fuel, type of burner, and emissions requirements, low furnace temperatures can present a challenge for maintaining low CO and unburned hydrocarbon emissions. Maintaining stable burner operation, especially with ultra-low emissions burners, is also more difficult at lower furnace temperatures. Burners in refinery applications typically burn fuels referred to as refinery fuel gas (RFG), which are a collection of products or by-products of many processes throughout the plant. Due to the large number of processes that exist in a typical refinery, the composition of the available fuel gas varies significantly from plant to plant, from heater to heater, and even over time on a given heater. Variation in fuel properties can significantly impact burner performance parameters such as stability, flame dimensions, and emissions. Testing process burners is an indispensable way to verify the effect of operating temperature and fuel composition on critical burner performance criteria. Due to the age of many refineries, many process burners in refinery applications are being replaced either because they have reached their reliable life span or because regulatory mandates are requiring the installation of modern low emission burners. Replacing burners in an old heater with ultra-low emissions burners can present several unique challenges. Due to the necessary design approach, ultra-low NOx burners generally produce longer and wider flames than conventional raw gas or premix burners. The burners also generally have a larger footprint than conventional burners. The performance of ultra-low NOx burners is also more sensitive to burner spacing and flame-to-flame interaction. Because many refinery heaters were originally equipped with conventional raw gas or premixed burners, replacing them with low emissions burners can result in flames
© 2011 by Taylor and Francis Group, LLC
The reforming segment of the industry produces hydrogen, ammonia, and methanol using specific furnace designs referred to as steam methane reformers. In each case, a natural gas feedstock is passed over a catalyst with steam that produces syngas (from synthesis gas), gas made up primarily of hydrogen and carbon monoxide. Since the reaction is endothermic, process burners are needed to provide heat necessary to drive the reaction. In the methanol process, hydrogen and carbon monoxide are subsequently combined over a catalyst to form methanol. In the hydrogen and ammonia processes the carbon monoxide is further oxidized to carbon dioxide (through the water gas shift reaction) to create additional hydrogen. The hydrogen is then separated and reacted with nitrogen to form ammonia. The most common reformer type is referred to as a down-fired furnace where the process burners are installed on the roof of the reformer and fired vertically downward between the process tubes. Figure 18.6 is a rendering of a down-fired row of burners in a reformer furnace and a photograph of a down-fired burner in operation in a test furnace. The two most significant design challenges associated with down-fired reformer burners are producing short compact flames while still achieving the required NOx emissions and simultaneously firing dual fuels with one of them typically being a low pressure off-gas. Due to the close spacing between burners and between burners and process tubes, a short compact flame is a desirable characteristic of process burners for the reforming industry. A short compact flame serves two purposes; it minimizes the possibility of the flame coming into contact with and damaging the process tubes and it places most of the heat near the top of the furnace. Locating most of the heat input near the top of the furnace is desirable since reforming furnaces are top-fed, resulting in the coolest portion of the process tube being near the peak in radiant heat transfer. Typically, two methods are used to maintain compact
382
Industrial Combustion Testing
Downfired burner
Flue gas
Tunnel
Flue gas Figure 18.6 Schematic of process burners in a down-fired reformer and a photograph of a down-fired burner operating in a test furnace.
flames. The first is to limit the heat release of the individual burners to less than 10 MMBtu/hr (2.93 MW) for natural draft or induced draft cases. The second method is to increase the available air-side pressure drop through the burner and use the energy in the combustion air stream to accelerate the mixing of fuel and air to shorten the flame length. Unfortunately, steps taken to minimize flame length generally tend to increase NOx emissions since low NOx emissions are commonly achieved by expanding the volume of the flame to lower the peak flame temperatures. This results in a trade-off between desired flame geometry and NOx emissions. To improve the overall efficiency of the process and reduce natural gas consumption, burners utilized in reforming applications use low pressure off-gases as a fuel source. During the initial startup phase before the reforming furnace is producing syngas, the burners commonly burn natural gas. When the furnace reaches operating temperature, the off-gas produced in the Pressure Swing Adsorption (PSA) units is burned in the burners, reducing the consumption of natural gas. The PSA offgas is typically low in pressure and also low in energy content due to the high content of carbon dioxide (typically 45–60%). This requires secondary fuel piping and burner fuel distributors to handle the higher volumetric flow rate and lower pressures. Burner testing plays a key role in ensuring the low pressure gas is effectively dealt with in the burner design so that the flame envelope is not compromised.
© 2011 by Taylor and Francis Group, LLC
18.3.3 Petrochemical Industry The petrochemical segment of the processing industries uses cracking furnaces to produce ethylene, which is the base material used to produce most plastics. Hydrocarbons such as ethane, propane, butane, naphtha, or gas oils are heated to very high temperatures in the presence of steam. Depending on the feed and the severity of cracking, typical coil outlet temperatures are in the range of 1435–1615°F (780°C–880°C). As the hydrocarbons reach cracking temperature they decompose through a complex series of radical reactions into a mixture consisting mostly of olefins such as ethylene and propylene, hydrogen and methane. High endothermic heats of cracking combined with extremely short residence times (0.1–0.5 seconds) require high coil heat fluxes and result in very high tube skin temperatures ranging typically between 1995°F and 2057°F (1090–1125°C) at the outlet. An undesired side reaction is the formation of coke (carbon) on the inside process tube walls. Coke production is a strong function of skin temperature. It can exist in various morphologies, depending on feed composition and tube metallurgy, but it always has a very low thermal conductivity and acts as an insulator. When the coke layer becomes too thick, the maximum tube skin temperatures approach the maximum allowable material temperature and the furnace tubes must be decoked using a steam/air mixture. In order to understand the critical role of burners in a steam cracking furnace it is important to keep in mind
383
Process Burners
Figure 18.7 Photograph of a train of seven ethylene cracking furnaces.
that, unlike most process heaters, the steam cracking furnace is the main reactor in the plant. Deviations from an ideal process temperature profile has a direct impact on plant throughput and profitability since it increases coking rates and lowers the coil selectivity. The process temperature profile inside the cracking coils is to a certain degree determined by the layout of the cracking coils, but the radiant heat flux profile generated by the burners remains the key parameter. In early firebox designs the fired duty was mostly provided by radiant wall burners. Since the heat release of a single burner was limited to about 1 MMBtu/hr (0.3 MW), many burners were needed to provide the total firing duty. The number of burners in a naphtha cracking furnace would typically range from 160 in the seventies up to 240 to 300 in the late eighties. The low firing duty meant that a single burner could be considered as a point source of heat. The overall heat flux profile could be tweaked by varying the fuel pressure between fuel headers, by adding more or less secondary air to individual burners, or simply by turning burners off. Figure 18.7 shows a train of ethylene cracking furnaces. The ongoing tendency to reduce investment cost and maintenance cost has significantly changed the character of these furnaces since the early nineties. The firing capacity of a single firebox increased tremendously, which made the use of 100% side wall burners very costly and maintenance-intensive. Instead, the firing is now accomplished mostly by vertically firing floor burners with capacities up to 14 MMBtu/hr (4.1 MW). This means that the heat flux profile from a single floor burner has become one of the most principal parameters for the cracking coil characteristics such as conversion and selectivity. While the firebox duty has increased, the firebox size has not kept equal pace due to efforts to reduce
© 2011 by Taylor and Francis Group, LLC
investment costs. As a result the firing density (expressed in fired duty per firebox volume or floor surface area) has increased faster than the firebox duty and burners have been placed so close together that they have started influencing each other in a negative sense. These burner–burner interactions have been the reason for a number of recent cases with severe rollover of the flames into the cracking coils. The continuous flame impingement resulted in excessive local coking of the process tubes, carburization of the material, and coil cracks. Like all other process furnaces, cracking furnaces have become the subject of tighter NOx and CO emissions limits. Cracking furnaces, however, are especially being scrutinized because of their high firing rates that make them very large producers of NOx in an absolute sense. Moreover, the firebox temperatures are the highest in the petrochemical industry, with bridgewall temperatures up to 2300°F (1260°C) compared to the typical firebox temperatures in refinery furnaces of 1400–1600°F (760–870°C). This means that even on a relative basis the thermal NOx production of a cracking furnace is often two to three times higher than that of a refinery furnace.
18.4 Process Burner Testing As discussed in the preceding sections, each sector of the processing industry has its own unique requirements and challenges to be addressed in burner design and testing. Similarly, there are several different reasons why testing may be conducted and the testing process differs depending on the specific objective of the test. 18.4.1 Objectives and Testing Overview Process burner tests are typically conducted for one of the following reasons: • Research and development purposes • Commercial testing to validate burner performance against customer specifications • Resolution of existing applications issues Each type of test has its own objectives. For example, during a research and development test, a detailed test matrix will be executed to completely characterize the capabilities of a process burner. This information may then be used to validate process burner CFD models and to develop burner-specific performance prediction models used in the production design engineering process. On the other hand, a commercial test focuses on
384
Industrial Combustion Testing
the operating conditions specific to a customer’s needs. A customer may be replacing burners with low-NOx burners to meet more stringent emissions regulations. Refinery fuel gas compositions can change as new processes are introduced; a large change in the fuel gas compositions may require testing of the burner to confirm emissions and flame dimensions. In some cases a customer may require a custom design to meet unique requirements; in these situations a burner test is usually requested to confirm performance. 18.4.1.1 Testing Parameters and Measurements During a process burner test, a wide variety of operating parameters are being controlled, measured, and recorded to quantify the burner performance characteristics [3]. Operating parameters such as pressures and temperatures are measured and recorded to verify that the burner is operating within design specifications. Flue gas analysis is conducted and flame dimensions are recorded to monitor the combustion process in the furnace. The following parameters may be measured and recorded during a burner test.
furnaces. The controlled and isolated environment of a test furnace facilitates the ease and safety of a demonstration. Fuel sources can be quickly isolated and other precautions can be taken to improve the safety of the operation. Common instabilities such as flashback, lift-off, and even acoustic coupling can be demonstrated. The reproduction of these conditions is invaluable to the customer as the flame appearance, fuel pressures, and damper settings can be documented to avoid this behavior in the field. This information can then be relayed to the operators and technicians who work with the burners on a daily basis. Burner research and development are greatly facilitated, since a test furnace will typically accommodate just a single burner. The burner can therefore be studied in greater depth as there is more operating control and better access for instrumentation. Bench-scale testing can provide some insight into the behavior of a full-scale process burner, but there is no substitute for reproducing the flame that will be present in the furnace or heater. Complicated scaling and extrapolations are avoided and real-time data is produced. 18.4.1.4 Burner Testing: Drawbacks
A process burner test can be divided into several general steps. A process burner test starts with the selection of the equipment and the furnace. During the setup of the furnace and burner, instrumentation and control equipment are modified to fit the needs of the test. Insulation patterns within the furnace are modified to produce the required furnace temperature. A customer fuel is simulated using a blend of economically available fuels that approximate the fluid transport and chemical properties of the customer fuel. After the setup and fuel simulation is complete, the burner is ready to be tested for the customer.
In spite of the benefits of a full-scale test there are some real limitations. Actual process furnaces typically utilize many burners to provide the heat needed for the process. A single burner test cannot predict if there will be burner-to-burner interaction in a furnace. Burner-to-burner interaction occurs when burners are tightly spaced together and their flames interact. What typically occurs is that if the flames are close enough to one another, they will coalesce into a single, large flame. This flame may be much larger than each individual flame, which presents the possibility of flame impingement on process tubes and increased pollutant emissions. A quick demonstration of this can be accomplished by holding the flames of two candles close to one another. When brought close enough, the two flames coalesce into one large flame. The process burner test facility will have a limited number of types of furnaces available. It will never have an identical furnace into which the burners will be installed. The flue gas recirculation patterns and heat sinks will be significantly different. It is often the case that the burner will still behave slightly differently in the field than in the test furnace due to these reasons and others. Due to these reasons it can be a challenge to extrapolate the test results to the actual furnace conditions and predict how the burner performance may change as a function of these differences.
18.4.1.3 Burner Testing: Benefits
18.4.1.5 Burner Testing Versus CFD
Having a dedicated test furnace allows for burner experiments that are impossible to do with actual process
The CFD modeling of combustion processes has greatly advanced in recent years. The development of a range
• • • • • • • • •
Fuel flow rate (heat release) and composition Fuel temperature and pressure Furnace draft and burner airside pressure drop Combustion air temperature and humidity Flue gas analysis (CO, O2, NOx, SOx, UHC, particulate matter) Furnace temperatures Radiant heat flux profile Visible flame length and width/diameter Noise measurements
18.4.1.2 Process Burner Testing Sequence
© 2011 by Taylor and Francis Group, LLC
385
Process Burners
of different combustion models for diffusion flames combined with accurate gray or nongray radiation models has lead to good successes predicting burner performance in actual furnaces Reference [1]. Premixed flame modeling remains difficult, however. Pollutant emissions are also very difficult to predict using CFD, due to the large differences in reaction rates between the main combustion reactions and the formation of NOx. Another certain disadvantage of CFD is that the near-burner region needs to be finely resolved, typically leading to very large meshes. For that reason, the convergence times needed to resolve full-scale process burner flames are still substantial compared to the speed of a burner test. This becomes much more evident when transient modeling is attempted to simulate flame instabilities or ignition. So while it has some shortcomings, the great advantage of CFD is that it greatly facilitates the extrapolation of the test results to the actual process furnace. Where the burner test provides insight in ignition, stability, and emissions, a CFD model of the actual furnace can predict whether flame interactions lead to rollover and impingement on the process tubes. So it is the tandem of burner testing and CFD modeling that will yield the most complete burner performance prediction.
18.5 Process Burner Testing Equipment and Methodology A significant investment in specialized equipment is required to properly perform full-scale process burner tests. The following is a list of some of the equipment that will be discussed in detail in subsequent sections. An aerial view of an industrial combustion testing facility is shown in Figure 18.8.
Figure 18.8 Aerial view of industrial combustion testing facility.
© 2011 by Taylor and Francis Group, LLC
• • • • • •
Test burners Test furnaces Fuel measurement and delivery systems Air delivery systems Instrumentation and control Special equipment
18.5.1 Test Burners As previously discussed, burners generally fit into one of two broad categories, premixed and raw gas. Burners are further classified by their firing orientation, flame shape, fuel type, location in the furnace, and several other factors. A basic description of several of the most common burners used in the processing industry is presented in the following paragraphs. A freestanding burner is a burner that is mounted away from the walls of a furnace, and is usually installed in a row or circle of burners where the flame is free to expand without restriction or interference from the furnace walls or other flames. In some instances, freestanding burners can also be horizontally fired when the application requires it, such as in continuous catalyst regeneration (CCR) applications. Figure 18.9 is a rendering of a freestanding burner. Downfired burners are also generally freestanding since they are mounted away from the walls of the furnace. However, downfired burners, which are typically used in reformer applications, are mounted on top of a furnace and fired vertically downward between the process tubes. A wall-fired burner produces a flame that is fired vertically upward along a furnace wall. The flame
Figure 18.9 Rendering of a freestanding process burner.
386
Industrial Combustion Testing
Figure 18.11 Rendering of a radiant wall process burner.
Figure 18.10 Rendering of a floor-mounted, wall-fired process burner.
heats the refractory wall, which in turn radiates heat to process tubes. This type of burner can provide a very uniform radiant heat flux distribution to process tubes. Wall-fired burners are commonly used in thermal cracking of hydrocarbons to form ethylene and similar olefins. As mentioned before, this process requires uniform heat distribution to the process tubes to prevent excessive coke formation and possible plugging of process tubes. Figure 18.10 is a rendering of a wall-fired burner. Premix radiant wall burners are mounted horizontally through the wall of a furnace. The burner tip, which penetrates only a few inches inside the hot face of the furnace wall, fires radially along the wall. These burners are almost exclusively used in ethylene cracking furnaces, either alone or in conjunction with floormounted, wall-fired burners to provide a uniform heat distribution to the process tubes. In some cases several hundred radiant wall burners are installed in a furnace to fine-tune the radiant heat flux to the process tubes. Figure 18.11 is a rendering of a radiant wall burner. Combination burners are typically freestanding burners that can fire both gas and liquid fuels. The liquids that are commonly burned are fuels such as No. 2 fuel oil (diesel) or No. 6 fuel oil. These burners also require connections for the medium used to atomize the liquid fuel, which is typically steam or air. The burner can be used to fire only gas, only liquid, or some combination of the two. Figure 18.12 is a rendering of an up-fired combination burner.
© 2011 by Taylor and Francis Group, LLC
Figure 18.12 Rendering of a combination oil and gas-fired process burner.
18.5.2 Test Furnaces Selecting the correct test furnace for a process burner test is essential to producing valid data that is representative of the actual application for the burner. Due to the number of different process heater designs and the range of burner sizes, a well-equipped process burner testing facility will have a wide variety of test furnaces available to simulate real world operations. Some of the most critical features to be considered when selecting a test furnace are as follows: • Test furnace size and geometry • Burner mounting location and firing direction
387
Process Burners
• Single versus multiburner testing capabilities • Furnace cooling and temperature control Test furnaces are tailored to different process burner designs. The direction a burner will fire in is an important consideration in determining the furnace to be utilized for a test. A poor selection of test furnaces can result in data that does not accurately represent the intended application for the burner. All furnaces need sufficient space to accommodate the flame produced from the burner. Up-fired and horizontally fired, freestanding burners need to be tall/wide enough to prevent flame impingement and/or propagation of the flame into the stack. Combination burners are typically up-fired and also have the same requirements of gas-fired, freestanding burners. Terrace wall firing requires a test furnace with a sloped wall to approximate terrace type furnaces. Multiburner tests involve two or more burners simultaneously firing in a furnace. These types of tests are common in ethylene applications where one or more wall-fired burners on the furnace floor and several radiant wall burners are firing simultaneously. Thus, the furnace needs to be tall enough to approximate the dimensions of a cracking furnace. It is also common to test the radiant wall burner individually prior to the multiburner test. For this reason, small box type furnaces are available to accommodate the small firing rate of a horizontally installed radiant wall burner. Furnace cooling and temperature control are important for a number of reasons. A process burner test can be compromised by operating at temperatures significantly higher or lower than the design temperature specified by the client. A correct firebox outlet temperature is essential for an accurate prediction of the heat flux profile. Temperatures that are too low can negatively impact burner stability or overpredict CO production. Firebox temperatures that are too high could lead to excessive NOx production. The furnace temperature can be regulated through the use of insulation, cooling tubes, and/or a water jacket. Insulation such as refractory and high temperature mineral fiber is used to protect the furnace steel from high temperature gases. Refractory comprised of oxides of aluminum, silicon, and magnesium are used to line the interior of a furnace and to withstand direct flame impingement. Cooling tubes can be used to cool the flue gases and the extent of cooling can be controlled by the flow rate of cooling fluid, such as water, through the tubes or by insulating the tubes themselves. Similarly, water jackets are also used to provide cooling. Photographs of several different test furnaces are shown in Figures 18.13 through 18.16. 18.5.3 Air Delivery Systems Air delivery systems for process burners are typically classified as either natural draft or forced draft. Owing
© 2011 by Taylor and Francis Group, LLC
Figure 18.13 Test furnace used primarily for ethylene applications.
to the buoyant forces of the hot gases inside the furnace, the pressure inside the furnace at the burner level is lower than the pressure outside the furnace. This pressure differential, known as draft, is the driving force used to supply air to the burner in natural draft systems. Natural draft burners have an air inlet that is open to ambient air. Forced draft systems use a fan or blower to provide air to the burner by way of a plenum or air duct. The air being supplied to the burner can be either at ambient temperature or preheated to as high as 900°F depending on the system being simulated. Pitot tubes are used to measure the static pressure just before the inlet to the burner. The difference in static pressure between the duct and the floor of the furnace provide the pressure drop across the burner. Forced draft burners normally operate at an air side delivery pressure that can be in excess of 2 inches of water column (0.5 kPa). They utilize the air pressure to provide a superior degree of mixing between fuel and air. Also, with forced draft systems, air control
388
Industrial Combustion Testing
Figure 18.14 Two test furnaces capable of firing either vertically upward or downward for reforming applications.
Figure 18.16 Test furnace capable of simulating terrace wall-fired heaters.
can be better maintained allowing the operator to realize economic savings, by operating furnaces at lower excess air rates over a wide firing range. Figure 18.17 shows an example of a mobile air preheater used during forced draft testing with or without preheated combustion air. 18.5.4 Instrumentation and Control The instrumentation and control system of the furnace controls and measures important parameters such as fuel flow rate, flue gas composition, and furnace temperature. Below is a list of instrumentation and control equipment functions that will be discussed:
Figure 18.15 Vertical-cylindrical furnace for freestanding upfired burner tests.
© 2011 by Taylor and Francis Group, LLC
• Fuel flow rate and composition • Analysis of flue gases • Flue gas temperature and pressure
389
Process Burners
calculated based on this response. The coriolis meter is a more expensive means of measurement, but this is often offset by its degree of accuracy and its low maintenance requirements. 18.5.4.2 Flue Gas Analysis
Figure 18.17 Self-contained portable combustion air heater and blower used for testing forced draft preheated air burner designs.
18.5.4.1 Fuel Flow and Composition Accurately metering the flow rate of each individual component used to make up a fuel blend is necessary to measure the heat release of the burner. There are many ways to measure flow: differential pressure, magnetic, mass, oscillatory, turbine, and insertion flowmeters, just to name a few. For purposes of burner testing, the differential pressure flowmeter will be discussed. Even limiting this discussion to differential flowmeters, there are still several different methods of measurement available. Measuring the differential pressure across a known orifice plate is the most commonly applied method [6]. The advantage of using orifice plates is that they are versatile and can be changed to match a flow rate and fuel to be metered. Also, there is a significant amount of data concerning measuring fuel flow via an orifice plate. Finally, there are no moving parts to wear out. The drawbacks to orifice plates are that they are precision instruments with an accuracy that is greatly determined by its condition: the flatness of the plate, the smoothness of the plate surface, the cleanliness of the plate surface, the sharpness of the upstream orifice edge, the diameter of the orifice bore, and the thickness of the orifice edge [6]. Another drawback is loss of accuracy when measuring flow rates of dirty fuels. While dirty fuels are a way of life for the refining industry, test fuels are clean (no liquid or solid particles in the gaseous fuels) and this concern is minimized. The coriolis meter is commonly used to measure liquid flow rates. The coriolis meter utilizes the coriolis effect to directly measure the mass flow rate of liquids. The meter is equipped with a specially shaped vibrating tube through with the liquid flows. When a fluid flows through the tube it alters the manner in which the tube vibrates. The mass flow rate of the fluid is then
© 2011 by Taylor and Francis Group, LLC
The measurement of NOx, CO, and O2 concentration in the test furnace flue gas is necessary to evaluate the performance of the process burner. A sample of flue gas is continuously extracted from the furnace using a water-cooled sample probe installed in the furnace stack. The sample flows from the sample probe through heat traced tubing to a sample conditioner inside a data building. The conditioner dries the sample and removes any soot or particulate matter. The sample then flows to a series of emission analyzers where the actual concentration of each species of interest is measured. The methods commonly used to measure each component are briefly reviewed in the following sections. There are a few common methods of measuring oxygen concentration in the gas phase. Electrochemical sensors and paramagnetic sensors are typically used to measure oxygen concentration on a wet and dry basis, respectively. Carbon monoxide (CO) is most commonly measured using a nondispersive infrared technique. A gas sample flows between an infrared radiation source and an infrared detector. Carbon monoxide absorbs infrared radiation, hence the difference in intensity proportional to the concentration of CO in the gas sample. Oxide of nitrogen (NOx) is most commonly measured using chemiluminescence. This method is capable of measuring oxides of nitrogen from subparts per million to 5000 ppm. The principle of operation of these analyzers is based on the reaction of nitric oxide (NO) with ozone. The sample, after it is drawn into the reaction chamber, reacts with ozone generated by the internal ozonator. The above reaction produces a characteristic luminescence with an intensity proportional to the concentration of NO. Specifically, light emission results when electronically excited NO2 molecules decay to lower energy states. The light emission is detected by a photomultiplier tube, which in turn generates a proportional electronic signal. The electronic signal is processed by the microcomputer into an NO concentration reading. To measure the NOx (NO + NO2) concentration, NO2 is transformed to NO before reaching the reaction chamber. This transformation takes place in a converter heated to about 1160°F (625°C). Upon reaching the reaction chamber, the converted molecules along with the original NO molecules react with ozone. The resulting signal represents the total NOx.
390
Industrial Combustion Testing
18.5.4.3 Flue Gas Temperature and Pressure A suction pyrometer (also known as a suction thermocouple or velocity thermocouple) is widely considered as the preferred method for obtaining gas temperature measurements in the harsh environment of an operating furnace. If a bare thermocouple is introduced into a hot furnace environment for the measurement of gas temperature, measurement errors may arise due to the radiative exchange between the thermocouple and its surroundings. A suction pyrometer is typically comprised of a thermocouple recessed inside a radiation shield. An eductor rapidly aspirates the hot gas across the thermocouple. This configuration maximizes the convective heat transfer to the thermocouple while minimizing radiation exchange between the thermocouple and its surroundings, assuring that the temperature measured is nearly that of the true gas temperature. Gas pressure is measured using gauges and pressure transducers. Differential pressures are measured using calibrated manometers in conjunction with differential pressure transducers. 18.5.5 Special Equipment 18.5.5.1 Heat Flux Several techniques have been developed to measure heat flux levels at different locations within a furnace. The instruments designed to successfully obtain heat flux data in the hostile environment of a full-scale furnace
are typically water-cooled probes, which are inserted through a furnace port at the location of interest. The probes may utilize radiometers that measure radiant heat flux levels. The sensing element is typically composed of a thermopile-type sensor that produces a voltage proportional to the temperature difference between the area of the element that is exposed to heat transfer from the furnace and the area that is cooled and kept at a relatively constant temperature per the element design. Common designs utilize a plug-shaped thermopile element with the exposed face at one end and the opposite end cooled by contact with a heat sink (Figure 18.18). The probe utilizes a crystal window, gas screen, or a mirrored ellipsoidal cavity to negate convective heat transfer to the sensor. A radiometer is also often equipped with a gas purge in an effort to keep the crystal clean and free from fouling. Critical parameters to consider when using a heat flux meter include the ruggedness, sensitivity, calibration method, and view angle of the instrument. Since the heat transferred to the furnace tubes is the desired information, the heat flux probe is placed in the same plane as the hot face of the tubes. The heat flux is measured opposite the fired wall via access ports along the vertical axis of the furnace (Figure 18.19). Once the data is collected, it is plotted on a curve and compared to the desired heat flux profile. 18.5.5.2 CO Probe Spatially resolved measurement of local CO concentration (CO probing) can be used to measure the flame
Water jacket Operating length
Water outlet
Thermo couples
Guard rings
Stainless steel plug
Figure 18.18 Schematic of heat flux probe.
© 2011 by Taylor and Francis Group, LLC
Cooling water
Cooling water
Water inlet
391
Process Burners
Process tubes
Access ports
Access ports
Heat flux probe
CO probe
Burner Elevation view Figure 18.19 Schematic of heat flux probe mounted in a test furnace.
envelope much more accurately than visual observation. The CO probing typically defines the flame edge as 2000 ppm of CO. A CO probe is a water-jacketed probe that is inserted into an operating furnace and connected to a sample line that feeds into emissions monitors. Using ports similar to the access ports for heat flux, the CO probe is traversed across the furnace at different elevations to measure flame width and height (Figure 18.20). Near-flame data can be collected and concentrations of oxygen, carbon monoxide, and oxides of nitrogen can be measured. The flames produced from process burners are typically turbulent, therefore CO probe data must be averaged over a period of time. 18.5.5.3 Noise Noise emissions are becoming increasingly important. With some refineries located near populated areas, it is important to keep noise to a minimum. Burner testing is usually conducted on a single burner and noise emissions are usually measured at a distance of approximately three feet from the burner air inlet. Data collected during the test includes an overall A-weighted sound pressure level and the sound level at each octave
© 2011 by Taylor and Francis Group, LLC
Burner Elevation view Figure 18.20 Schematic of CO probe mounted in a test furnace.
band ranging from 31.5 Hz to 8000 Hz. When collecting noise data, it is important to measure it with and without the burner operating, in order to correct the measurements for the background noise.
18.5.5.4 Unburned Hydrocarbons, Particulate Matter, and Oxides of Sulfur Combustion processes can create pollutant emissions other than carbon monoxide and oxides of nitrogen. Unburned hydrocarbons (UHC) is a term describing any fuel or partially oxidized hydrocarbon species that exit the stack of a furnace. The cause for these emissions is typically due to incomplete combustion of the fuel from poor mixing or low furnace temperature. A low temperature environment can be created by operating the furnace at a reduced firing rate or turndown. Particulate matter (commonly called soot) is often produced from fuel rich regions in diffusion flames. Soot becomes smoke if the rate of formation of soot exceeds the rate of oxidation of soot. Oxides of sulfur are formed when sulfur is present in the fuel.
392
Industrial Combustion Testing
18.5.6 Test Fuels Selection The customer fuel can not always be reproduced exactly at a test facility due to the sheer number of components that could be present and the cost associated with supplying and maintaining such an inventory. To circumvent this problem, a test fuel is blended to simulate the customer fuel and reproduce its fluid-transport and chemical properties. Three properties of the simulated test fuels that need to be matched are the isentropic coefficient, molecular weight, and lower heating value (LHV). These properties need to be closely approximated to reproduce the fuel pressure versus heat release relationships. The Wobbe number may also be calculated to determine similarity in fuels. The chemical properties that need to be matched are the adiabatic flame temperature, inert content, olefins and hydrogen content, and the LHV. These properties need to be closely approximated so that the combustion process produces similar flame heights, pollutant emissions, and flue gas temperatures. As an example, Table 18.1 illustrates a refinery gas. The test fuel flow chart shown in Figure 18.21 is a general procedure for composing test fuel blends. By following this flow chart, a test fuel can be composed that
Client fuel specification
will accurately simulate both the fluid-transport and chemical properties of the actual fuel. Based on the fuels available for blending, the hydrogen content is matched, propylene is used to substitute the ethylene content, and a natural gas/propane Table 18.1 Example Refinery Fuel Gas (RFG) Composition Fuel Component Name Methane Ethane Propane Butane Ethylene Propylene Butylene 1-Pentene Benzene Carbon monoxide Hydrogen a b c
Total inert content (N2, CO2, H2O)
Match inert content
Total olefins (unsaturates) (CnH2n)
Match olefin content
Total parafins (saturates) (CnH2n+2)
Sum hydrogen, inert and olefin content
8.13a 19.9a 0.30 0.06 32.0b 0.78 0.66 0.07 0.12 0.22 37.8c
Use natural gas and propane (C3H8) to match lower heating value (LHV) and molecular weight (MW)
No
Are variables within acceptable ranges?
Yes
Test fuel composition
Figure 18.21 Test fuel selection process flow diagram.
© 2011 by Taylor and Francis Group, LLC
CH4 C2H6 C3H8 C4H10 C2H4 C3H6 C4H8 C5H10 C6H6 CO H2
Compare adiabatic flame temperature (AFT), flame speed and air requirement between client and test fuel
Compare LHV, MW between client and test fuel Match hydrogen content
Volume %
Balance of fuel is primarily methane and ethane. Level of olefins in the fuel. Hydrogen content.
Calculate client fuel properties (LHV, MW, AFT. etc.)
Determine hydrogen content
Formula
393
Process Burners
Table 18.2 Comparison of RFG and Simulated Test Fuel LHV (Btu/sch) HHV (Btu/scf) Molecular weight Specific heat ratio @ 60°F Adiabatic flame temperature (°F) Wobbe index
Refinery Fuel
Test Fuel
1031 1124 18.09 1.27 3481 1422
1026 1121 18.38 1.26 3452 1407
mixture is used to simulate the methane and ethane balance as described in Figure 18.21. By holding the hydrogen content fixed at 38%, natural gas and propylene are balanced to obtain a match of the LHV and molecular weight. This balance is then further refined by including the adiabatic flame temperature. A test fuel blend of 34% natural gas, 28% C3H6, and 38% H2 would be acceptable to simulate the refinery fuel gas illustrated in Table 18.1. Table 18.2 gives a side by side comparison of the fuel properties. 18.5.7 Test Procedure Before the burner test starts, all physical dimensions of the burner have to be verified and recorded, in particular the number and arrangement of fuel ports. The test configuration has to be verified as well, such as the burner spacings and the firebox insulation pattern. Each day of testing must begin with a calibration of the flue gas analyzers. A test procedure will vary from burner test to burner test due to difference in application. However, there are several test conditions that are routinely tested to determine the general performance of a burner. The testing of a process burner begins with a cold furnace light off. This is an important test point as it demonstrates burner stability during start-up conditions. Also important during light-off/warm-up period is the production of carbon monoxide. Increasingly stringent emissions regulations are limiting the amount of carbon monoxide that can be emitted during start-up. The lightoff point is usually tested with natural gas or with some other purchased start-up fuel, such as butane. If a pilot is present, its performance may also be verified. Pilot stability can be confirmed by increasing the furnace draft, closing and opening air registers, increasing the airside pressure drop across the burner, or increasing/ decreasing pilot fuel pressure. This phase of the burner test is also used for verification of the flame scanner ability to check the flame of either the pilot or the main burner. Once the flame scanner operation has been confirmed it can be used to verify the burner ignition times and extinction times, whenever these are part of the
© 2011 by Taylor and Francis Group, LLC
burner specification (for example if EN746-2 specifications apply [4]). Design heat release or maximum capacity is demonstrated to confirm that the process burner will provide the heat release required. This test point is often the condition at which the process burner will operate for most of its lifetime. Therefore, documentation of flame dimensions and pollutant emissions at these conditions is important. Carbon monoxide (CO) break is a condition that is achieved by increasing the fuel flow rate until the furnace is almost depleted of excess air. This is conducted to simulate how a burner will respond in a situation where excess air suddenly decreases. Turndown ratio or minimum heat release is demonstrated by reducing the heat release of the burner. These points are repeated if more than one fuel composition is to be tested. Certain fuels such as start-up fuels are only used during light-off and warm-up periods and therefore do not require testing at every condition.
18.6 Conclusion Process burner testing is an invaluable tool in validating client requirements, research and development of new products, and characterizing burner models to develop predictive tools. Process burners are utilized in the refining, reforming, and petrochemical industries to provide heat to furnaces used to produce fuels, base chemicals, specialty chemicals, and hydrogen. Process burner testing is an intensive experimental undertaking that requires careful selection of test furnace and set-up of instrumentation. Testing must be completed in a detailed and disciplined approach in order to collect reliable data. In all cases, process burner testing tailors individual burners to meet the stringent demands of furnaces for any application. By testing process burners, combustion engineers, clients and end-users can be assured that the product to be utilized will be thoroughly developed and meet requirements regarding safety, reliability, operability, and maintenance.
References
1. Baukal, C. E. The John Zink Combustion Handbook. Boca Raton, FL: CRC Press, 2001. 2. API 560. 2007: Fired Heaters for General Refinery Service, ANSI/API Standard 560, 4th ed., August 2007.
394
3. API 535. 2006: Burners for Fired Heaters in General Refinery Services, Recommended Practice, 2nd ed., January 2006. 4. European Standard EN 746-2. Industrial thermoprocessing equipment: Part 2: Safety Requirements for Combustion and Fuel Handling Systems, March 1997.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
5. Gibson, W. 2009: Personal communications with Bill Gibson of Great Southern Technologies, LLC, Tulsa, OK, May 2009. 6. Spitzer, D.W. Flow measurement: practical guides for measurement and control, Research Triangle Park, NC, Instrument Soc. America, 1991.
19 Commercial Boiler Burners Yaroslav Chudnovsky and Mikhail Gotovsky Contents 19.1 Commercial Boilers Population, Applications, and Efficiency Test Methods..................................................... 395 19.2 Burners for Commercial Boilers: Types, Design, Fuel, and Firing Capacity....................................................... 396 19.3 Testing Approach and Methodology........................................................................................................................ 397 19.4 Testing Technique of Gas-Fired Burners.................................................................................................................. 400 19.5 Testing Specifics for Oil-Fired and Dual-Fuel Burners.......................................................................................... 405 References................................................................................................................................................................................. 409
19.1 Commercial Boilers Population, Applications, and Efficiency Test Methods The American Boiler Manufacturers Association (www.abma.com) defined the commercial boilers as boilers having an input capacity between 400,000 (light) and 20,000,000 (heavy) Btu per hour with standard efficiency in the range of 80 to 85% and mostly burning natural gas, propane, and fuel oil. Per Consortium for Energy Efficiency (www.cee1.org) almost 85% of commercial boilers are gas-fired and annually use up to 602 trillion Btu of natural gas. Space heating accounts for approximately 66% of commercial boiler demand. Hot water accounts for most of the rest. More details on commercial boilers characterization can be found in a report by the Energy and Environmental Analysis, Inc [1]. Figure 19.1 illustrates a breakdown of the commercial boilers population and capacity by fuel type. It is clear that natural gas-fired and dual-fuel units dominate the commercial boiler market, so their efficient and reliable performance along with low combustion emissions that comply with the local Air Quality Management District rules and regulations are the main targets for the boiler and burner manufacturers and end users. The commercial boiler market is moving toward high efficiency and condensing technology. Currently, highly efficient condensing boilers are making up 20–25% of the commercial boiler market and exceed the 90% efficiency barrier. It is not quite correct to consider the burner separately from the enclosure (boiler) where the burner to
be installed, so in the majority of cases the fully assembled (packaged) system is tested for the efficiency and performance. However, most of the boilers are very vulnerable to design and operating conditions (such as heat transfer surface fouling, internal geometry, burner turndown, etc.) that may somehow sacrifice an overall boiler efficiency over relatively short or long periods of time. Therefore, the periodic testing of the boiler efficiency is a primary interest and of benefits for the end user. In order to obtain a complete picture of the boiler performance several types of efficiency tests need to be conducted. The main test is a so-called combustion efficiency test. For a better picture, it may be combined with one or more specialized tests such as carbon monoxide testing, hydrogen testing, and others. Combustion efficiency is a characterization of the combustion systems ability to burn fuel. Burners producing a low level of unburned fuel while operating at low excess air levels are considered efficient. Combustion efficiency is different for various fuels and, generally, gaseous and liquid fuels burn more efficiently than solid fuels. Another boiler performance parameter is “thermal efficiency.” Thermal efficiency characterizes the heat exchange effectiveness and measures the ability of the heat exchanger(s) to transfer heat from the combustion product to the water or steam in the boiler. However, thermal efficiency does not account for radiation and convection losses due to boiler body design specifics or other components, so it cannot be considered a true boiler efficiency for economic evaluations. Fuel-to-steam efficiency is a measure of the overall efficiency of the boiler. It accounts for the effectiveness of
395 © 2011 by Taylor and Francis Group, LLC
396
Industrial Combustion Testing
Boiler units
100,000
1,200,00
Boiler units Boiler capacity
1,000,00
80,000
800,000
60,000
600,000
40,000
400,000
20,000
200,000
0
Coal
Oil
Natural gas
Others
Boiler capacity (MMBtu/hr)
120,000
0
Figure 19.1 Commercial boilers population and capacity per type of fuel. (From Energy and Environmental Analysis, Inc., “Characterization of the U.S. Industrial/Commercial Boiler Population,” Report submitted to Oak Ridge National Laboratory, May 2005. http://www.cibo.org/ pubs/industrialboilerpopulationanalysis.pdf)
the heat exchanger as well as the radiation and convection losses. Per the latest revision of ASME/ANSI Power Test Code 4 [2], the fuel-to-steam efficiency of a boiler can be easily determined as: Output Efficiency fuel = 100 × Fuel Input
19.2 Burners for Commercial Boilers: Types, Design, Fuel, and Firing Capacity In 1875 Stephen Wilcox established 12 design rules [7]: • Proper workmanship utilizing the best methods and simplest construction. • The inclusion of a drop-out area (i.e., mud drum) for steam boilers to aid in accumulating and removing waterside impurities. • Steam and water capacities large enough to preclude swings in pressure and/or water level. • Adequate steam release space to prevent carryover and foaming. • Proper circulation to maintain the metals at uniform temperatures. • The division of water space into several sections to prevent a general steam explosion in the event that one section fails. • Design with excess strength. • Provide an adequate furnace to ensure complete combustion within the furnace proper.
or
Losses-Credits × 100. = 100 − Fuel Input
(19.1)
• Provide for cross flow of the gases to the tubes to create turbulence and maximize heat transfer. • Provide easy access for cleaning and repairing. • Maximize capacity and efficiency.
There are plenty of sources available that describe boiler efficiency measurement and supporting calculations (for example, [3–6]); however, the following key factors affecting the boiler efficiency must always be taken into consideration: • • • • • •
Design specifics Fuel specification Excess air Stack temperature Ambient air temperature Radiation and convection losses
High boiler efficiency is always the result of optimal combination of design and operating criteria, including enough heating surface and passes, good burnerboiler compatibility, reliability of combustion controls and safety, as well as cost-effective implementation of advanced waste heat recovery techniques.
© 2011 by Taylor and Francis Group, LLC
• Include the best accessories. These rules can be successfully applied for today’s engineering and development practice, especially in terms of adequacy: for the best performance the burner should perfectly match the boiler! Burners for commercial boilers vary considerably on the worldwide market in design, control scheme, and applications ranging from commercial snow melting and municipal solid waste incineration to water and space heating for hospitals, hotels, and restaurants. Most burners may be classified and described in terms of a few basic features or characteristics. Figure 19.2 illustrates the basic boiler burners’ classification. The detailed discussion of the burner design, operation, and control scheme is beyond this chapter and can be found in comprehensive published sources [8–11] as well as in multiple burner manufacturers brochures and technical publications. Most of the control system and fuel valve train requirements are specified (but not limited to) by the following codes:
397
Commercial Boiler Burners
Using flue gas recirculation (FGR) Low NOx
Boiler burners
Air staged Fuel staged Fuel induced recirculation
Ultra low NOx
Premixed
Conventional
Swirl
Rapid mix
Register
Figure 19.2 Basic boiler burners classification. (From GTI/CEC Report CEC-50005-026, 2006.)
• American National Standards Institute (ASNI Z21.13) • American Society of Mechanical Engineers (CSD-1) • American Society of Mechanical Engineers (Boiler Code: Sections I and Section IV) • National Fire Protection Association (NFPA 85 2001) • Underwriters Laboratories (UL 795)
19.3 Testing Approach and Methodology Fuel combustion in the burner is a complex combination of the flow, chemical, and thermal processes, so the most reliable instrument for the efficient burner development is testing. Testing is defined as experimental evaluation of the qualitative and quantitative characteristics of the combustion system performance while operating at laboratory and/or field conditions. Throughout the world, there are four major types of testing accepted: • • • •
evaluation performance acceptance certification
Evaluation testing is key at the burner development stage. During this testing the burner design is refined and optimal operating parameters are determined in order to meet the development requirements. Each burner developer or manufacturer establishes in-house
© 2011 by Taylor and Francis Group, LLC
evaluation testing procedures and matrices. Evaluation results serve as the basis for the follow-on design engineering and fabrication of the precommercial burner to be further tested for performance characteristics. Upon reaching the targeted performance characteristics, the burner design and operating scheme are revised, if necessary, and the commercial prototype of the burner (or burners) is tested for acceptance. If the burner passed the acceptance testing and satisfied all the technical and environmental requirements set by the targeted specification it is ready for the marketplace. In most cases prior to reaching the marketplace, a burner has to pass certain certification tests. Those certification tests are set by a country, industry, or regulation agency (UL, TUC, CSA, etc.). In the United States most of the commercial boilers are subject to mandatory certification. Underwriters Laboratories (UL) developed a set of standards and procedures for the certification of the gas, oil, and gas–oil burners intended for installation in heating applications such as boilers, furnaces, heaters, ovens, water heaters, incinerators, and so on. Gas, oil, and gas–oil burners are provided with primary safety controls permitting their operation to be without a competent attendant present while the burner is firing. For example, UL 795 requirements [12] apply To factory-built gas appliances having inputs of more than 400,000 Btu per hour, per individual combustion chamber which require flame failure and other precautions and which are intended primarily for commercial and industrial installation. The appliances covered by these requirements are gas burners, comfort heating furnaces, heaters and gas-fired boiler assemblies except watertube boilers having outputs of 10,000 pounds of steam per hour or more. Gas-heating equipment covered by these requirements may be operated without a competent attendant being constantly on duty at the burners while the burners are in operation. Additional installation and operation requirements are available for central-heating gas appliances, domestic conversion burners, floor furnaces, room heaters, unit heaters, and water heaters as defined by the National Fuel Gas Code, NFPA 54, and by the Liquefied Petroleum Gas Code, NFPA 58, as applicable.
UL 726 requirements [13] apply to oil-fired boiler assemblies. The main objective of the certification testing is to prove safety operations and sustained performance of the tested combustion equipment. In the course of the certification testing the following main tests are usually conducted [12]:
398
Industrial Combustion Testing
15
4**
Not permitted
4
4
4
5
1
1
*The pilot flame establishing period shall not exceed four seconds if the pilot input exceeds 0.4 MMBtu/h. **Maximum fuel input at light off not to exceed 2.5 MMBtu/h.
© 2011 by Taylor and Francis Group, LLC
9
8
7
6
5
10 as G
as G
as G
as G
as
as
G
as
G
10
4
10
as
15
3
600
10*
G
> 5.0
10*
as
2.5–5.0
15*
700
2
< 2.5
1
800
Exhaust temperature, F
Pilot flame establishing period, sec Main burner flame establishing period (when ignited by interrupted pilot), sec Main burner flame establishing period (when ignited by electrical igniter), sec Flame failure response time, sec Safety shut-off valve closing time, sec
Maximum Firing Rate per Test Combustion Chamber, MMBtu per hour
G
Rated Timing for Flame Failure Condition
In the case of gas pressure variation by ±50% of the rated pressure downstream of the main burner regulator those switches should cause a safety shakedown of the tested combustion system. All the commercial boiler burners with the firing capacity greater than 5 MMBtu/h have to be equipped with the pilot flame. The electrical ignition system could ignite only a pilot. Burners with firing capacity less than 5 MMBtu/h may be ignited directly by an electrical igniter but the maximum fuel input should not exceed 2.5 MMBtu/h. The corresponding ignition system has to be activated prior to supplying the main fuel input and has to remain active during the flameestablishing period (see Table 19.1). During the ignition test the low-voltage ignition has to be evaluated at 70% of rated voltage for the ignition transformer. Figure 19.3 illustrates the results of the commercial boiler burner ignition testing at different gaseous fuel compositions. The test was performed in the GTI Applied Combustion Research Facility, Des Plaines, Illinois. The burner was ignited and recorded the combustion product temperature exhausting from the test enclosure. The test control system ignited the burner at manufacturer specified conditions and upon reaching a temperature set point (~600°F) shut the burner off,
as
Table 19.1
• Maximum firing rate of the burner exceeds 2.5 MMBtu/h. Or, • Burner is equipped with electrical igniter for main gas ignition (regardless of the firing rate).
G
Prior to certification testing each burner has to be fully assembled at the test facility in the manner to ensure strength, rigidity, and durability. Burner piping components such as the shut-off valves, pressure regulators, and the like should be properly installed to simulate the field operation conditions as close as possible. The pipe fitting torque has to be compliant with the standard requirements and the entire assembly has to be checked for possible leakage by loading with hydrostatic pressure of 1.5 times the maximum operating pressure. The test burner has to be equipped with primary safety controls that de-energize the main burner shutoff valve upon flame failure. Table 19.1 specifies the
required time interval for the de-energization of the safety shut-off valve. A test burner has to be equipped with low and high gas pressure switches if:
G
• Current overload tests, including bonding conductors and connections tests • Torque test: fitting/pipe connections, including gasket tests • Combustion test, including burner, combustion air failure and power interruption tests • Pilot test, if applicable • Ignition test, including all hot wire and delayed ignition tests • Continuous operation temperature test • Blocked inlet and outlet tests • Electrical safety test, including short circuit and combustion safety controls tests • Downdrafts and updrafts tests • Flame, rain, wind, and other applicable tests
500 400 300 200 100 0 0:00:00 0:14:24 0:28:48 0:43:12 0:57:36 1:12:00 1:26:24 1:40:48 1:55:12 Time, hours
Figure 19.3 Results of the ignition test for different gaseous fuels (CEC PIER Contract 500-05-026).
399
Commercial Boiler Burners
purged, and cooled the test enclosure to initial temperature of 200°F followed by reigniting the burner. The ignition test was repeated four times to prove ignition performance consistency. One of the critical ignition tests during the certification is the delayed ignition test when the ignition of the main flame is delayed to make sure that such a delay will not result in flashback of flame to the outside of the test enclosure or any other associated damage. This test is performed by gradually increasing the ignition delay by one second up to the maximum flame establishing period. Upon completion of the ignition test, the burner has to be evaluated for the safe and reliable combustion performance. For example, in order to satisfy UL certification requirements it has to be continuously fired at maximum capacity for 250 hours with daily monitoring of its combustion performance (exhaust temperature and composition, fuel burning rate, controls operation, etc.). All the data are collected and recorded in corresponding data collection sheets that includes the following parameters along with registered visual observation: • Fuel type: natural gas, oil, propane, and so on • Actual firing input capacity, Btu/h • Gas test pressure prior to regulator, in. WC or in Hg • Gas test pressure in the manifold, in. WC or in Hg • Exhaust gas composition (O2%, CO ppm, CO2%, NOx ppm) • Flue gas temperature, °F • Combustion air inlet temperature, °F • Draft over-fire and in-flue, in. WC or in Hg • Stack loss, % • Outlet water temperature, °F • Outlet steam pressure, psig • Other parameters of the secondary equipment at high firing rate
Most of the certification tests are performed when the burner is packaged with the corresponding commercial boiler. Those tests intend to evaluate the maximum temperature that can be reached by the system components while regulating a firing capacity within the test burner turndown ratio, as well as operational limits of the burner when air inlet or vent outlet are partially or completely blocked, and other carefully specified evaluations in the corresponding standard literature [12,13]. During any combustion equipment testing, it is absolutely necessary to follow the set of safety rules established by the testing facility and industry regulations. The major safety items are listed below: • Do not expose yourself to open flame, open viewports, or doors. Combustion pulsations or operating conditions can create a situation of blowout of the hot combustion product through the system openings. • Flame observation has to be performed through the tinted goggles or tinted shield to protect the eyes from harmful light intensity. • Be prepared for hot water in drums and headers during the maintenance or installations. • Always set proper venting conditions in order to evacuate the entire combustion product out of the test space. Never use any toxic or volatile fluids in confined spaces. • Observe at all times the instrumentation reading and flame conditions when making any changes. Remember that extremely low excess oxygen during the boiler operation could end up with catastrophic results. • Do not touch high-voltage parts such as ignition transformers and do not disassemble any units while they are powered to avoid the possibility of electrical shock. • Do not inspect or touch any test facility components until they are cool.
Actual firing input capacity is calculated as
CF = #
[atmospheric pressure (in. Hg) + gas meter pressure (in. Hg)] (520°F)# (30.00 in. Hg) [460°F + gas temperature(°F)]#
⇒ remove from calculation if flow meter is temperature compensated.
sec Btu [gas consumed (ft 3 )] 3600 heating value 3 (CF) ft h Input = [time of consumption (sec)]
© 2011 by Taylor and Francis Group, LLC
(19.2)
400
Industrial Combustion Testing
19.4 Testing Technique of Gas-Fired Burners Putting aside many formalities associated with the certification of commercial boiler burners one can formulate two major testing techniques—so-called cold testing and hot testing. Cold testing intends to obtain flow characteristics of the tested burner including the corresponding flow rates and pressure drop values. During the cold tests, the fuel gas is substituted by the volumetrically equal amount of air while retaining an appropriate stoichiometric air/fuel ratio. In the course of cold testing, the pressure drop coefficients are determined by the separate air blowing through gas and air passages (nonpremixed burners) or joint air/substitute fuel blowing through the burner in the case of premixed burner design. Pressure drop coefficients are calculated as: for fuel gas pass ζ g =
2 Pg ρ g Wg2
for combustion air pass ζ a =
,
2 Pa , ρaWa2
(19.3)
(19.4)
where Pg = substitute gas inlet excessive pressure, kPa, Pa = combustion air inlet excessive pressure, kPa. For forced draft burners flow averaged substitute gas velocity the (m/s) can be calculated as
Wg = (Vg/Fg)⋅(Tg/273)⋅[101.3/(Pg + P0)],
(19.5)
where Vg = substitute gas flow rate, Fg = specific cross section of gas passage, P0 = atmospheric pressure. The flow averaged combustion air velocity (m/s) can be calculated as
Wa = (Va/Fa)⋅(Ta/273)⋅[101.3/(Pa + P0)]
(19.6)
where Va = substitute gas flow rate, Fa = specific cross section of air passage, P0 = atmospheric pressure. For atmospheric burners with a convergent nozzle the following correcting factor should be used for determination of the flow averaged substitute gas velocity
at Pg < 90 kPa: [(Pg + P0)/P0]1/k,
© 2011 by Taylor and Francis Group, LLC
(19.7)
at Pg > 90 kPa: [(k + 1)/2]1/(k−1),
(19.8)
where k = adiabatic exponent for air. Flow density for Equations 19.3 and 19.4 can be estimated as: ρa – density of the combustion air in a specific cross section, kg/m3,
ρa = 1.29⋅(273/Ta)⋅[(Pa + P0)/101.3],
(19.9)
ρg – density of the substitute gas in specific cross section, kg/m3,
ρg = 1.29⋅(273/Tg)⋅[(Pg + P0)/101.3].
(19.10)
For atmospheric burners the corresponding multipliers should be applied to the equation above:
at Pg < 90 kPa: [P0/(Pg + P0)]1/k,
(19.11)
at Pg > 90 kPa: [P0/(Pg + P0)]1/(k−1).
(19.12)
The hot testing acquires burner performance information and can be conducted on the specially designed test rigs or in the field with the qualified application. The burner has to be tested within the entire range of the fuels it was designed for. Prior to the hot testing, the burner has to be visually inspected and properly installed into the test enclosure. If necessary, the burner could be disassembled and reassembled to meet the design specification and blueprints. It would be helpful to check all the movable components and determine their displacement limits. Upon completion of the burner installation, all the piping and flange connections have to be checked for any possible leakage at the working pressure (special liquids or simple soap water can be used). When installing the boiler burner into the laboratory test rig it is very important to simulate (as close as possible) the real-life conditions (for example, keep the same ratio of burner nozzle cross section area to the combustion space cross section area). Necessary measurement and control means having to be properly installed and calibrated prior to the testing. It is extremely desirable to select the measurement sensors as small as possible to minimize their affect on the hydrodynamics inside the burner and test chamber. The test rigs should be equipped with view ports or windows to observe the flame pattern and burner operation. They also have to be equipped with enough mounting capabilities for measurement and control sensors installation. Figure 19.4 illustrates the commercial boiler burners installed at Gas Technology Institute laboratory (Des Plaines, Illinois) prior to the hot testing. All the measurement and control sensors
401
Commercial Boiler Burners
(a)
(b)
Figure 19.4 Commercial boiler burner with input capacity up to 1.5 MMBtu/hr (a), and 3.0 MMBTu/hr (b).
were connected to the specially designed data acquisition system. Figure 19.4a represents a test setup with the water-cooled simulator enclosure while Figure 19.4b represents the test burner packaged with the off-theshelf commercial boiler. Hot testing is usually conducted at stationary conditions by gradually increasing and decreasing the firing rate within the crucial limits (flashback and blow-off). The conditions are considered stationary if exhaust gas temperature change is less than 5° over a 30 minute interval. The critical limits have to be reached at least five times to collect enough reliable data. For better burner performance analysis, it is necessary to have the composition of the fuel, its density, heating value, and Wobbe number (WN). All those characteristics can be obtained from the local analytical laboratory services or measured using gas chromatography techniques followed by the simple calculations:
WN =
high heating value specific gravity with respecct to air
The minimal required excess air is determined on the basis of the combustion products composition and pollutant emissions that comply with the current environmental regulations set for particular burner class, type, and application. The burner flame length is usually considered as a “chemical” length and is determined based on the measurement along the flame until combustion is completed (for example, CO2/CO2max = 0.95 or O2min/O2 = 0.95). However, in some cases (open fire), the visible flame length measurement is acceptable. The flame temperature is measured by a water-cooled suction pyrometer or available optical means. Heat transfer from the
© 2011 by Taylor and Francis Group, LLC
flame to the boiler load is usually determined via calorimeter. Figure 19.5 illustrates the examples of burner flame visual observations at the testing laboratory environment. When performing temperature measurements in the flame or in close proximity of the flame with a nonshielded thermocouple, some errors arise due to the irradiative exchange between the thermocouple and its surroundings. In order to obtain the reliable data during such measurements the suction pyrometers (or aspirated thermocouples) have to be employed. A standard suction pyrometer is a platinum–rhodium thermocouple (R-type), protected from chemical attack by a sintered alumina sheath and surrounded by two concentric radiation shields. The gases are drawn between the shields and over the sheath with high velocity (at least 50 ft/s) so that the equilibrium t hermocouple temperature is nearly that of the gases without the need for correction. The gases are normally sucked in a dynamic position through a hole drilled at the side of the outer shield with the end of the shield closed with a cement plug. However, when low dust concentrations are present in the flame, the hole can be placed in the wake of the shield to minimize blockage. As part of an effort to develop techniques for making more accurate flame temperature measurements, Palmer [15] compared the performance of a self-compensating thermocouple with that of the more complex and expensive aspirated thermocouple. The basic design, implementation, and laboratory installation of the aspirated thermocouple are illustrated in Figure 19.6. Boiler burners are sensitive to the back pressure conditions, so testing the burner performance at various levels of back pressure applied to the burner exit could better assist in the determination of the operating scheme of the burner in the field. The same applies to
402
Industrial Combustion Testing
(a)
(b)
(c)
(d)
(e)
Figure 19.5 (For Figure 19.5a, see color insert following page 424.) Examples of the burner flame visual observations. (a) Primary combustion flame of two-stage forced internal recirculation burner, (b) high velocity burner flame, (c) high temperature oxy-flame, (d) jet impingement flame, and (e) radiant tube flame.
the acoustic impact. Acoustic instabilities generated by the boiler enclosure could propagate downstream of the flame and affect air/fuel ratio. Combustion noise is also one of the captured parameters during the performance testing. The combustion noise is registered by dB meter and has to comply with the established regulations for the specific area and application. Figure 19.7 represents noise test results for the commercial boiler dual-fuel burner with a nominal firing rate of 3.5 MMBtu/h.
© 2011 by Taylor and Francis Group, LLC
For an overall picture of burner performance, the following major parameters to be measured include: • Ambient conditions (temperature, pressure, humidity) • Air and gas flow rates (mass flow meters, orifices) • Air and gas temperature (thermocouple, usually K-type)
403
Commercial Boiler Burners
(a)
120°
1/4˝ GA20
3/8˝ GA18
1/2˝ GA18
To readout
~ 4˝ ~ 10˝
To the pump (providing ~ 50 ft/s flow velocity through the aspirator) (b)
(c)
Figure 19.6 Aspirated thermocouple installed (a) aspirator design sketch, (b) thermocouple/aspirator assembly, (c) at the test chimney. (GTI Applied Combustion Research Facility, Des Plaines, IL)
• Air and gas pressure (pressure sensors, pressure gauges, U-tube manometer) • Outside surface temperature (surface thermocouple, thermal paints, optical pyrometers) • Combustion chamber temperature (thermocouple K-type of R-type) • Exhaust gas temperature (thermocouple or suction pyrometer) • Combustion noise (dB-meter) One of the most critical performance characteristics of the commercial boiler burner is a level of green house emissions such as nitric oxides and carbon monoxide. The combustion product is typically sampling from the exhaust portion of the test setup (laboratory) or exhaust ductwork (field) and then is being prepared (cooled, dried, filtered) and forwarded to the appropriate gas analytical equipment. Figure 19.8 illustrates the set of the analytical equipment typically used for the burner testing at the GTI Applied Combustion Research Facility. The basic set contains the following analytical instruments (detailed description of each analyzer can be found at http://www.emersonprocess.com/raihome/
© 2011 by Taylor and Francis Group, LLC
or corresponding analytical equipment manufacturer documentation): • Chemiluminescence NOx analyzer • • • •
Dispersed infrared carbon monoxide analyzer Dispersed infrared carbon dioxide analyzer Flame ionization total hydrocarbons analyzer Paramagnetic oxygen analyzer
Prior to each testing all of the instruments are calibrated using pure nitrogen to establish the “zero” and an appropriate span gas to set the “gain.” Upon testing completion the calibration check is performed to make any appropriate corrections, if necessary. There are more techniques available on the market for the combustion exhaust composition measurement. For example, the Fourier transform infrared (FTIR) spectroscopy. Continuous emission monitoring system (CEMS) MultiGas™ 2030 provides real-time, simultaneous measurement of the concentrations of flue gas components ranging from water vapor, nitrogen oxides, sulfur oxides, HCl, ammonia, H 2SO4, and many other compounds. Many organic species can
404
Industrial Combustion Testing
120
110 4
Combustion noise, dB
100
90 3 80 2 70 1 60
50
62.5
125
250
500 1000 2000 Frequency, Hz
4000
8000
Figure 19.7 Noise measurements results: (1) background noise, (2) ID fan and purging noise (no combustion), (3) combustion noise at nominal firing capacity, and (4) upper limit noise spectrum.
to establish the sophisticated CEMS on the test site, so there is portable CEMS equipment available that provides accurate and reliable measurement of emissions that is produced by a tested combustion system. For most field tests, GTI uses a portable gas analyzer HORIBA (Model PG-250) that can simultaneously measure up to five separate gas components using the same proven measurement methods used in HORIBAs line of permanent CEMS. The PG-250 is a helpful tool for certification testing as well as for exhaust gas emissions periodical measurements/ monitoring in one or multiple points. The PG-250 uses nondispersive IR detection for CO, SO2, and CO2; chemiluminescence (cross-flow modulation) for NO x; and a galvanic cell or an optional zirconium oxide sensor for O2 measurements. Figure 19.9 illustrates the portable exhaust emission analyzer (overall view (a) and screen sample (b)). Selection of the appropriate emissions measurement means and its calibration is a very important aspect of any commercial boiler burner test due to specified accuracy and reliability of the corresponding measurements. Based on the measured data the following parameters could be calculated: • Burner thermal power, kW
Q = Vfuel LHV,
(19.13)
where Vfuel = fuel flow rate, m3/s LHV = lower heating value, kJ/kg, LHV = 126.4 CO + 108 H2 + 359 CH4 + 644.6 C2H6 + 932.0 C3H8 + 1228 C4H10 + 1461 C5H12 + 595.2 C2H4 + 877.8 C3H6 + 1135 C4H8 + 1414 C5H10 + 1403 C6H6 + 234.4 H2S. (19.14) • Turndown ratio Maximum: Qmax/Qmin Operating: Qnominal/Qmin. • Excess air at boiler exit
Span gases Sample flow controls
Analyzers
Data acquisition
Figure 19.8 Typical analytical instrumentation set.
also be measured. The software includes the calibration data for over 200 gaseous components that may be found in flue gas. Sometimes the burner testing is performed in the field and it is quite complicated
© 2011 by Taylor and Francis Group, LLC
αb =
N2 . N 2 − 3.76O2
(19.15)
• Stack losses (exhaust gas), %
qe =
0 iexh − α b icold air . Ql
(19.16)
405
Commercial Boiler Burners
(a)
(b)
Figure 19.9 Portable gas analyzer (HORIBA PG-250) employed by GTI field testing team.
• Heat losses (chemical incompleteness of combustion), %
qic =
Vg (126.4 CO + 108 H 2 + 358.2 CH 4 ) 100 %, (19.17) Qll
where i exh is the enthalpy of exhaust gas. In all tests it is strongly recommended using the theory and practice of the experimental planning [16]. It will significantly help to minimize the efforts and resources with maximum efficiency of the testing.
19.5 Testing Specifics for Oil-Fired and Dual-Fuel Burners Beyond the above described for gas-fired burner testing, the following characteristics have to be determined when testing the oil and dual-fuel burners: • Quality of the fuel oil atomization (cross uniformity, atomization angle) • Aerodynamic and thermal characteristics of the flame (length, heat flux, emissivity) • Atomization quality • Soot formation During the cold testing, the hydraulic coefficients for the atomization flow passage is determined by blowing out by steam or compressed air, while the oil passage is recommended for the oil or substitute liquid with the same viscosity at the working temperature. Oil passage hydraulics is determined by pumping the oil or substitute liquid with the same viscosity as the fuel oil at the operating temperature. Oil distribution
© 2011 by Taylor and Francis Group, LLC
across the fuel jet is usually determined by using a simulation liquid such as paraffin at 70°C at the special test rig that allows measuring of the droplet size and distribution uniformity across the fuel jet. The following correlations can be obtained as the result of such a test:
Gi = f(dd),
(19.18)
ε = f(Goil),
(19.19)
α0 = f(Goil),
(19.20)
where Gi = Gi ’/G = ratio between the mass of fuel droplet of the certain size dd to the total mass of the sample: Goil = fuel oil flow rate, kg/s, α0 = half of fuel jet spray angle, degree. The reliable and quiet operation of the oil-fired burners very often depends on the quality of the atomizers. It is well known that in any burner—regardless of its air pattern—the atomizing nozzle must assist the burner to do these things: • The spray must ignite smoothly • The nozzle must provide a spray that will maintain a steady, quiet fire • The nozzle must provide a pattern and droplet size that will burn clean • The combination of the air pattern and fuel pattern must provide good efficiency To accomplish these things, nozzles are available in different spray patterns and different spray angles. Figure 19.10 below illustrates the Delavan nozzle and three basic patterns that satisfy a wide range of the oilfired burner requirements.
406
Industrial Combustion Testing
(a)
Steel orifice disc Distributor with tangential slots
Brass body
(b)
Brass screw pin (c) Sintered filter
Figure 19.10 Examples of the spray pattern produced by the Delavan nozzles: (a) hollow, (b) solid cone, and (c) all purpose.
The relationship between the pressure and discharge from a nozzle is a fundamental one [14]. The theoretical discharge from any nozzle is given by the equation
flow rate = CA (2 gH)0.5,
(19.21)
where C = a dimensionless coefficient for the particular nozzle, A = the area of the nozzle orifice, H = the pressure head applied to the nozzle, g = acceleration of gravity. This fundamental equation is modified by various factors encountered in the nozzle design, but from it we arrive at a simple formula, which is of value to anyone using nozzles. 0 .5
P F2 = F1 * 2 , P1
(19.22)
where P1 = pressure at which the nozzle is calibrated, P2 = any pressure at which it is desired to operate a nozzle other than the calibration pressure, F1 = the calibrated flow rate at pressure Pl, F2 = the flow rate at the desired pressure. Figure 19.11 clearly demonstrates the effect of the fuel oil pressure on discharge pattern from the “simplex” pressure-atomizing nozzle [14].
© 2011 by Taylor and Francis Group, LLC
During the hot testing, the following parameters are to be measured: • • • • • • • • •
Ambient pressure and temperature Oil flow rate Oil inlet pressure and temperature Atomization media flow rate (steam or compressed air) Atomization media temperature prior to the burner Combustion air flow rate, pressure, and temperature Combustion product composition Exhaust gas temperature Soot particles concentration
The crucial limits for the oil-fired and dual-fuel burners are similar to ones for the gas-fired burners (flashback and blow-off) with an additional regime for the soot formation that exceeded the limits established by the environmental regulations as well as a reduced atomization regime that is characterized by the occurring liquid droplets at the burner exit. For the gas-fired burners, a pre-purge of the test combustion volume in the case of oil-fired and dual-fuel burners testing plays a very important role in the operating procedure. The purging time has to be sufficient to provide four air volume changes in the combustion
407
Commercial Boiler Burners
(a)
(b)
(c)
(d)
Figure 19.11 Photographs illustrating the atomizing pattern change per pressure increase: (a) 3 psi, (b) 10 psi, (c) 100 psi, and (d) 300 psi.
chamber and associated passages. Table 19.2 provides some of the major requirements for purge and actions required on flame failure at the oil-fired operation [13]. Heat losses due to combustion incompleteness could be determined based on the fuel and exhaust gas compositions: qic = Qc⋅Cp⋅(1 − βfuel/βexh), (19.23) where Qc = carbon heating value (32.8 MJ/kg); Cp = carbon content in fuel, %; βfuel and βexh = parameters of the fuel and exhaust gas accordingly. βfuel = 3⋅(1 – Oox)⋅(0.125⋅OP + 0.039⋅Np)/(Cp – 0.375⋅Sp), (19.24)
© 2011 by Taylor and Francis Group, LLC
βexh = 21 – (CO2 + 0.605⋅CO + O2)/(CO2 + CO),
(19.25)
where Oox = content of the oxygen in oxidizer, % by volume, OP, Np, Sp = content of oxygen, nitrogen, and sulfur in fuel, % by volume, CO2, CO, O2 = content of carbon dioxide, carbon monoxide, and oxygen in combustion product at the test chamber exhaust, % by volume. Heat losses due to incompleteness of combustion could also be determined based on soot particles concentration at the exhaust:
408
Industrial Combustion Testing
Table 19.2 Purge Requirements for Oil-Fired Commercial Boiler Burners Maximum Firing Rate, MMBtu/h < 0.4 Prepurge
Not required
Postpurge Pilot typea and flame establishing period Main burner flame establishing period: Ignition by pilot Direct ignition
Not required N/A
Flame failure reaction time: Ignition by pilot Direct ignition
10 sec maximum (oil 2 and 4) 15 sec maximum (oil 5 and 6)b Not permitted except for low fire start up to 3.0 in which case 15 sec maximum
N/A 90 sec maximum
4 sec maximum 15 sec maximum for < 1.0 4 sec maximum for > 1.0 5 sec maximum
4 sec maximum 4 sec maximum
N/A
Combustion air proving
N/A
Action required on loss of combustion air
N/A
1 sec maximum
One recycle permitted if flame failure response time does not exceed four sec Required if fan is not mounted on burner motor shaft Safety shutdown
Safety shutdown required
Required Safety shutdown
Continuous and intermittent pilots are permitted if only the main burner flame is monitored during the burner operation. If it can be demonstrated by tests that the burner equipped to fire oil 5 and 6 needs more than 15 sec for the main burner flame establishing period in order to avoid nuisance shutdown, the period may be extended to 30 seconds provided not more than 15 seconds of unburned fuel can be discharged during an attempt to establish the main flame.
qic = 3.28 CVexh/LHV,
For stoichiometric conditions (α = 1)
Vfuel = VRO2 + VN2 + VH2O ,
(19.27)
VRO2 = 0.0187 ⋅ K P , m 3 /kg,
(19.28)
VN2 = 0.79 ⋅ Va + 0.008 ⋅ N p , m 3 /kg,
(19.29)
VH2O = 0.11 ⋅ H p − 0.012 W p + 0.016 ⋅ Va 0 , m 3 /kg. (19.30)
© 2011 by Taylor and Francis Group, LLC
For α > 1
(19.26)
where C = soot particles concentration averaged per exhaust cross section, g/m3; Vexh = specific volume of the combustion product per unit of fuel, m3/kg; LHV = low heating value of fuel, MJ/kg.
15 sec minimum Interrupted 10 sec maximum
15 sec maximum 15 sec maximum
One relight attempt permitted only
b
> 3.0 Required four air changes at 60% damper opening and with proven air flow
N/A 90 sec maximum
Fuel valve closing time after de-energization Action required on flame failure
a
0.4–3.0 Not required for < 1.0 Required for > 1.0 (if oil pump operates independently of the burner) Not required Interrupted 10 sec maximum
Vfuel = (1.87 ⋅ K p /RO 2 ′ ) + VH2O .
(19.31)
For α < 1
Vfuel = [1.87 ⋅ K p /(RO 2 + CO + CH 4 )] + VH2O .
(19.32)
Emissivity of the flame produced by oil-fired burner is typically determined by an optical pyrometer per Schmidt’s method [17]:
ε=
E3 + E2 − E1 , E3
(19.33)
where E1 = pyrometer reading when flame radiates to absolute black body, E2 = pyrometer reading flame radiates to absolute black body at hot nonreflecting background, E3 = pyrometer reading when only hot background radiates.
409
Commercial Boiler Burners
Moreover, in the case of oil-fired and dual-fuel burners, the list of measured pollutant emissions is extended by two more components SO2 and H2S.
References
1. Energy and Environmental Analysis, Inc. “Char acterization of the U.S. Industrial/Commercial Boiler Population.” Report submitted to Oak Ridge National Laboratory, May 2005. http://www.cibo.org/pubs/ industrialboilerpopulationanalysis.pdf 2. ASME/ANSI Power Test Code 4, 1998. 3. Wulfinghoff, D. R. Energy Efficiency Manual. (For everyone who uses energy, pays for utilities, designs and builds, is interested in energy conservation and the environment.) Wheaton, MD: Energy Institute Press, 2000. 4. Taplin, H. R. Combustion Efficiency Tables. Lilburn, GA: Fairmont Press, 1991. 5. Payne, F. W., and Thompson, R. E., eds. Efficient Boiler Operations Sourcebook. Englewood Cliffs, NJ: Prentice Hall, 1996.
© 2011 by Taylor and Francis Group, LLC
6. Testing Standard BTS-2000. “Method to Determine Efficiency of Commercial Space Heating Boilers.” Hydronics Institute Division of GAMA, 2001. 7. Packaged Commercial Watertube Boilers. American Boiler Manufacturers Association, 2003. 8. GTI/CEC Report CEC-500-05-026, 2006. 9. Baukal, C. E., Jr. Industrial Burners Handbook. Boca Raton, FL: CRC Press, 2004. 10. Burkhardt, C. H. Domestic and Commercial Oil Burners. New York: McGraw-Hill, 1969. 11. Reed, R. North American Combustion Handbook, Volume II, Gloversville, NY: AIS Engineering Books, 1997. 12. UL Standard 795: Commercial-Industrial Gas Heating Equipment. 13. UL Standard 726: Oil-Fired Boiler Assemblies. 14. Olson, E. Fuel Nozzles for Oil Burners Technical Aspects of Applications. Delavan Technical Publication (www.delavan.com). 15. Palmer, T. “Comparison of Aspirated and RadiationCompensating Thermocouples.” Fire Technology 6, no. 3 (1970): 224–28. 16. Montgomery, D. C. Design and Analysis of Experiments. New York: , John Wiley & Sons, 2005. 17. Schmidt, E. “Messung der Gesamtstrahlung des Wasser dampfes bei Temperaturen bis 1000°C.” Forschung im Ingenieurwesen 3, no. 2 (1937): 57–70.
20 Power Burners Vit Kermes, Petr Beˇlohradský, Petr Stehlík, and Pavel Skryja Contents 20.1 Introduction...................................................................................................................................................................411 20.2 Power Burners.............................................................................................................................................................. 412 20.2.1 Classification Based on a Type of Supply of Combustion Air into Burner........................................... 412 20.2.1.1 Natural Draft Burners.................................................................................................................. 412 20.2.1.2 Induction Supply of Combustion Air........................................................................................ 412 20.2.1.3 Burners With Forced Supply of Combustion Air..................................................................... 412 20.2.2 Classification Based on Fuels....................................................................................................................... 413 20.2.2.1 Gas Fuel Burners........................................................................................................................... 413 20.2.2.2 Liquid Fuel Burners.......................................................................................................................414 20.2.2.3 Combined Burners....................................................................................................................... 415 20.2.3 Classification Based on Methods of NOx Reductions............................................................................... 415 20.3 Fuels............................................................................................................................................................................... 415 20.3.1 Gaseous Fuels................................................................................................................................................. 415 20.3.2 Liquid Fuels.....................................................................................................................................................416 20.3.3 Solid Fuels....................................................................................................................................................... 417 20.4 Oxidizers....................................................................................................................................................................... 417 20.5 Power Burner Testing.................................................................................................................................................. 419 20.5.1 Combustion Chamber Selection.................................................................................................................. 419 20.5.2 Pressure in the Combustion Chamber....................................................................................................... 420 20.5.3 Fuels................................................................................................................................................................. 420 20.5.3.1 Gaseous Fuels................................................................................................................................ 420 20.5.3.2 Liquid Fuels................................................................................................................................... 421 20.5.4 Combustion Air............................................................................................................................................. 422 20.5.4.1 Instrumentation............................................................................................................................ 423 20.5.4.2 Steam and Compressed Air........................................................................................................ 423 20.5.4.3 Instrumentation............................................................................................................................ 423 20.6 Test Planning and Evaluation.................................................................................................................................... 423 References................................................................................................................................................................................. 426
20.1 Introduction A burner may be defined as an entrance gate for fuel and oxidizer into the combustion chamber. The burner as well as the gate have to meet certain requirements. These involve not only compliance with technology requirements, but also with more and more restrictive emission limits for individual groups of technologies. Investors’ demand for low operational costs is also very important and can be achieved by more efficient utilization of energy released during combustion. All these facts contribute to quite a large diversity of burners that
are custom designed for particular types of applications. The following text focuses on presenting testing of power burners combusting mainly gaseous and liquid fuels. These power burners are generally used in the local power supply and various industrial applications, such as chemical and petrochemical industries, incineration of waste coming from several types of industrial productions, and so on. This text does not explore development and testing of burners for industries such as metallurgical and glass that require special burners for direct and indirect heating. The text is divided into two content-related parts. The first part deals with the issue of power burners, fuels, and 411
© 2011 by Taylor and Francis Group, LLC
412
oxidizers. It is basically an introduction for readers who are not familiar with the design and testing of power burners. The second part discusses the requirements of testing facilities, preparation, carrying out, and evaluation of power burner testing. The authors are aware that investments into testing facilities and financial costs of burner testing are obviously restricted by economic means of the burner designers.
Industrial Combustion Testing
with flue gas fan to produce a draft in the combustion chamber. The last method uses forced air supply and needs a combustion air fan. Each of these methods has its pros and cons, which serve as a basis for the selection of burner type. The application of burners is further conditioned by technological and economic restrictions, which means optimization of investment and operating costs. 20.2.1.1 Natural Draft Burners
20.2 Power Burners Trying to find a simple definition for a power burner gives us a clear idea what a power burner is—a weldment of steel components with an incorporated piece of refractory called a burner quarl. This assembly is attached to a wall of the combustion chamber and connected to the fuel and oxidizer. Instrumentation and control usually not part of the power burner and it is usually located in supply pipes near the burner, depending on the requirements and dispositional possibilities. Separation of the fan and instrumentation and control components from the actual burner gives its designers a lot of space for optimizing its geometry (as opposed to block burners that have the fan and control unit integrated into the burner itself). Geometry is a key element in meeting often contradictory requirements on both the technology and emission limits for polluting substances [1], especially nitrogen oxides. Another reason for choosing power burners is that it is possible to install multiple burners in a single furnace. Separation of instrumentation and control components from individual burners enables great simplification of furnace control, burner design, and even distribution of heat in furnace. All the burners do not have to be of equal heat output; boundary burners might have a different heat output than burners in the center of a furnace. Central controlling of multiple burners is much less expensive and very practical. Depending on their design, power burners can be classified according to several parameters [2,3], such as the type of supply of combustion air, type of combusted fuel, methods for NOx reduction, and so forth. It is necessary though to realize that a burner can fall into more than one category. 20.2.1 Classification Based on a Type of Supply of Combustion Air into Burner Basically, there are three methods of supplying a burner with combustion air, first by natural suction using natural draft (this supply may be enhanced by an ejection effect generated by fuel jets). The second type of air supply is induction air suction. This kind of air supply works
© 2011 by Taylor and Francis Group, LLC
Natural draft burners are supplied with combustion air from their surroundings and they do not allow for the preheating of combustion air. They utilize furnace natural draft (stack effect) and often simultaneous ejection effect of fuel streaming from burner jets under a high speed. This is a rather frequent method of combustion air supply and is used for both gas fuel and liquid fuel burners (and possibly for combined burners). The burners are relatively difficult to operate under a low excess of air since the amount of supplied combustion air can only be controlled by a chimney damper or in some cases by flap valves on individual burners, which is rather demanding on the operators. Due to complicated control they are predestined to be used in operations with constant heat consumption without any sudden changes of heat supply, as it would cause product quality deterioration that is, process furnaces in chemical and petrochemical industries [2]. 20.2.1.2 Induction Supply of Combustion Air Combustion air supply for inductive burners is similar to combustion air supply for natural suction burners. However, the draft in a combustion chamber is created by flue gas fan. This method is used whenever there is insufficient natural suction or if a pressure drop disables suction of a sufficient amount of combustion air (e.g., through convective heat exchangers, filters, etc.). Installation of burners with an inductive supply of combustion air as opposed to natural draft burners is less common. 20.2.1.3 Burners With Forced Supply of Combustion Air In contrast to natural draft burners, burners with a forced supply of combustion air can be used in many applications and technologically they can replace natural draft burners. Some of the biggest advantages of burners with a forced air supply include a wide regulation of possibilities and precise controllability. They are often utilized in overpressure combustion chambers. This type of burner also allows for great design variations, motivated by requiring reduction of polluting emissions and fuel savings. Burners with a forced supply of combustion air, which are used in process
413
Power Burners
engineering and power engineering, can be classified into two categories depending on the location of their combustion air fan. The fan and control system of fuel and air supply can be either integrated in the burner body (the so called block burners) or installed outside of the burner body (the so called power burners). Block burners are usually produced in series and they are utilized in the field of heat and steam production and for applications with autonomous and low maintenance operations. Size of the burner heads and flame shape are adapted to a specific target furnace size. Many countries stipulate requirements for control systems of block burners in regulatory rules and thus burners have to undergo several tests during their certification. Compared to block burners, power burners are usually produced separately or in small numbers depending on the needs of individual application. As far as the construction is concerned, it is a weldment of metal pieces with an integrated burner quarl. The design of power burner geometry is usually not strictly restricted by dimensions of the combustion chamber compared to the block burner. This enables a bigger latitude in its design and consequently contributes to a wide range of possibilities of reducing NOx emissions (such as staged fuel and combustion air supply, internal flue gas recirculation, and combination of these) [4,5]. The fact that most of instrumentation and control components as well as the combustion air fan are located separately outside of the burner body allows for savings of fuel costs by using combustion air preheating [6] and for external flue gas recirculation [7] as a technique of NOx emission reductions. Certification and standardization are required only for instrumentation and control components. Operational regime solutions and safety are handled solely by the investor (designer, supplier). 20.2.2 Classification Based on Fuels Distinction between power burners regarding the fuel is independent of the type of their combustion air supply and many various combinations are possible. Currently, the market offers power burners combusting gaseous and liquid fuels separately or combined burners combusting gaseous and liquid fuels alternatively or simultaneously [8]. There are three basic combinations available (e.g., simultaneous combustion of two gaseous fuels, two liquid fuels or a combination of gaseous and liquid fuel). Sometimes it is possible to encounter special requirements for the simultaneous combustion of gaseous and solid fuels, possibly also the combustion of liquid and solid fuels. 20.2.2.1 Gas Fuel Burners Currently, there are many gas fuel burners on the market (see Figures 20.1 and 20.2) that can be classified as power
© 2011 by Taylor and Francis Group, LLC
Flame stabilizer Secondary fuel
Secondary air
Primary fuel Primary air
Air
Gaseous fuel Primary gas throttle Figure 20.1 Section through the model of gas fuel/air-staged burner.
(a)
(b)
Top view
Side view
Figure 20.2 Example of gas fuel/air-staged burner flame.
burners. These burners are usually designed for combustion of well-established fuels, such as natural gas and liquified pertroleum gas (LPG) mixtures that enable fuel substitution by a mere change of jets and adjustment of fuel pressure prior to entering a burner. Adjustments of a burner design for other types of gas mixtures are usually carried out only after specific demands from a client. This mostly concerns combustion of waste gas, such as coking and refinery gases and mining gas that have a sufficient lower heating value or waste gases with a very low heating value that require flame stabilization by noble fuel. Depending on the type of mixing of the fuel with oxidizer power burners, combusting gaseous fuels can be divided into burners with premixed, diffusion, and partially premixed combustion. Considering speed of fuel and oxidizer streaming through the burner head, we can further distinguish between laminar and turbulent combustion; the latter prevails for power burners. Choosing a particular type of combustion determines not only the basic quality of the combustion such as shape, size, and temperature of the flame and formation of polluting substances but also flame stability. Burners with a premixed combustion (i.e., fuel and oxidizer
414
Industrial Combustion Testing
being mixed prior to entering the combustion chamber) demonstrate much higher flame stability compared to diffusion burners. They also do not need a big combustion space for fuel burnout and they are able to transmit much more heat via radiation due to high temperatures in the flame. It is, however, necessary to consider the higher emission of NOx caused by higher flame temperature and also the design of an instrumentation and control system has to prevent the danger of flashback of the flame into the burner body [3]. Burners with diffusion combustion (i.e., fuel and oxidizer being mixed after entering the combustion chamber) demonstrate opposite qualities compared to premixed burners. Flashback of the flame into the burner body is prevented by fuel and oxidizer being mixed in the combustion chamber. It also significantly lowers flame temperature since the mixture of fuel and oxidizer burns out gradually depending on the mixing rate thus reducing NOx emissions. Gradual mixing of fuel and oxidizer and burning of this mixture is however more demanding as to the volume of the combustion chamber compared to premixed burners. This is caused by an increase in diameter and especially in flame length. Further it is necessary to consider flame stability in the burner design, to prevent flame blow-off. In addition to the two above mentioned categories there is their combination (i.e., part of the fuel and oxidizer is premixed prior to entering the combustion space and the rest of the fuel and oxidizer are supplied unmixed into the combustion chamber by separate inlets). One of the biggest advantages of this type of burner is in its high flame stability. 20.2.2.2 Liquid Fuel Burners Liquid fuel burners (see Figure 20.3) as well as gas fuel burners are designed with respect to combusted fuel. In most cases this concerns standardized fuels produced by companies processing oil. It is possible to encounter Atomizing fluid
Ignition spark
Liquid fuel
Air
Figure 20.3 Liquid fuel burner using pressure steam atomization.
© 2011 by Taylor and Francis Group, LLC
several other types of atypical combustion liquids that possess physical and chemical properties quite different from most standardized liquid fuels. A basic requirement of burner combusting liquid fuels is a high-quality fuel atomization [9], necessary for complete evaporation and burnout in the area of the flame. If some fuel drops are not evaporated and combusted in the area of flame, concentrations of carbon monoxide and unburned hydrocarbons (UHCs) in flue gas increase rapidly. For the above mentioned reason most liquid fuel burners are designed as diffusion burners with fuel atomized in the combustion chamber. The fuel atomization system itself is rather dependent on physical and chemical properties of fuel and availability of auxiliary atomizing medium. Thus there are three basic types of atomization [10] (i.e., pressure, pneumatic, and rotary atomization). Besides these, there are other, less frequent types of atomization using vibrational, acoustic, ultrasonic, and electrostatic atomizers or flash liquid atomization. Pressure atomization mainly utilizes transformation of pressure energy of the liquid. Pressure energy of the liquid after release from the atomizer into a combustion chamber changes into kinetic energy and thus causes fuel atomization. Depending on the type of atomized fuel, the required pressure ranges are in MPa. Quality of pressure atomization is highly influenced by physical and chemical properties of the atomized fuel and is not suitable for liquid fuels of high viscosity. One of the biggest advantages of pressure atomization is the fact that it does not require an auxiliary medium and there are no movable components on the burner. Considering the energy intensity, pressure atomization is the least demanding of the three main types of atomization. Pneumatic atomization (i.e., atomization using auxil iary atomizing gas medium) utilizes mostly compressed gas energy that secures sufficient disintegration of streaming liquid. Compressed air and steam are a frequently used media for atomization, depending on the fuel type. Compressed air can be applied for fuels that do not require preheating prior to entering the burner (extra light fuel oil (ELFO)), possibly for fuels with no sharp difference in temperatures between preheated fuel and atomization air (light fuel oil (LFO) of 50°C preheating temperature). Steam is used for atomization of fuels that are necessary to preheat to high temperatures. These include especially of heavy fuel oils (HFOs) and mazut, which require preheating temperature significantly above 100°C. Com pared to pressure atomization, pneumatic atomization is not so demanding on fuel and auxiliary medium pressures that rarely exceed 1.2 MPa. Considering the energy intensity, pneumatic atomization is the most demanding out of the three types of atomization. Rotary atomization utilizes kinetic energy transmitted to the atomized fuel by an element that rotates in high
415
Power Burners
revolutions. Dosing of liquid is done under low pressure; liquid atomization is then caused by centrifugal force affecting fuel particles. One of the biggest disadvantages of rotary atomization is the presence of burner components rotating at high speed. Considering the energy intensity, rotary atomization is more demanding than pressure atomization but does not reach the intensity of pneumatic atomization. Besides the three atomization methods described, there are other, technically more demanding, types such as ultrasonic atomization and atomization based on flash fuel evaporation prior to entering the combustion chamber. Ultrasonic atomization is based on contact of liquid with surface that vibrates with ultrasonic frequency. Some of the advantages of ultrasonic atomization are in its low energy intensity and also the fact that liquid is not supplied under pressure as in the case with pneumatic and pressure atomization. Low jet flow rate and high investment costs are mentioned as some of the disadvantages of ultrasonic atomization, which is caused by high prices of ultrasonic waves generators. 20.2.2.3 Combined Burners Power burners that enable combustion of more than one fuel are designated as combined burners (see Figure 20.4). Their installation is usually conditioned by technical and economic reasons on the investors part. Some of the technical reasons include stabilization of burning secured by noble gaseous or liquid fuel. Stabilization is appropriate in combustion of gaseous or liquid waste products of low heating value and of unstable chemical composition. The choice of two-fuel burners and thus also financially demanding fuel management on the investors part is usually motivated by potential future savings of operational costs and failure in supplies of one of the fuels. A good example of this is heating plant boilers that run on natural gas for the purposes of low NOx and carbon dioxide emissions. Light fuel oil serves
as a back-up fuel that can be stored for long periods of time and functions as a substitute for unexpected natural gas supply failures. 20.2.3 Classification Based on Methods of NOx Reductions One of the other classifying features for the power burners design is the methods employed in primary reduction of NOx formation [11]. Among those are staged fuel supply [12,13], staged combustion air supply [14], and internal/external flue gas recirculation [7]. Sometimes these methods are combined (e.g., see Figure 20.1) or supplemented by injection of water and/or steam into the flue gas.
20.3 Fuels In relation to the previous section that introduced the basic classification of power burners, let us now focus on fuels combusted in these burners. Selection of a proper fuel is based on specific needs of a given technology, legislative restrictions, current as well as future availability of the fuel, and economic assessment of investment and operating costs. Selection also reflects apprehension about payments for greenhouse gases emissions, especially carbon dioxide [15]. This classification of fuels does not aim to be an extensive list. Besides standardized fuels, there are many kinds of fuel such as waste gas and liquid wastes coming from different types of productions. Here we focus on gaseous and liquid fuels; solid fuels are only marginally mentioned since they are rarely combusted in typical power burners. 20.3.1 Gaseous Fuels
Mining gas Natural gas
Figure 20.4 Installation of combined natural gas/mining gas burner.
© 2011 by Taylor and Francis Group, LLC
Gaseous fuels belong to the most widely used fuels due to availability of natural gas and LPG. Popularity of gaseous fuels can be explained by related low investment and operating costs for fuel management (compared to liquid and solid fuels) and flue gas cleaning. Availability of natural gas for its end users in certain areas in the world is grounded in long-term activities related to the building of gas lines. In its liquid form, LPG can be relatively easily transported and to a certain degree can replace natural gas in small power appliances in areas with no gas lines. Utilization of other gaseous fuels is negligible compared to natural gas and LPG. These include mostly local sources from industrial and agricultural productions such as refinery, coke and
416
Industrial Combustion Testing
Table 20.1 Basic Classification of Gaseous Fuels According to CSN 38 5502 (Gaseous Fuels Basic Classification) Classification of Fuel Gases Low-Calorific HHV < 16.8 MJ.m−3
Medium-Calorific HHV = 16.8–20 MJ.m−3
HHV [MJ.m−3]
High-Calorific HHV = 20–50 MJ.m−3
HHV [MJ.m−3]
Very High-Calorific HHV = > 80 MJ.m−3
HHV [MJ.m−3]
HHV [MJ.m−3]
– Stack gas
3.4–6.2
– Town gas
16.8–18.5
– Natural gas
min. 38.1
– Propane
101.0
– Low-pressure ordinary producer gas
3.6–12.0
– Coke oven gas
18.9–19.9
– Refinery gas
41.9
99.0–115.1
– Mining gas
17.6
– Propanebutane – Butane
– Gas from gasification of brown coal
15.9–16.7
– Biogas, sludge gas
23.0
– Hydrogen
12.7
mining gas [16], and biogas [17]. A rough overview of basic gaseous fuels and their classification according to higher heating value (HHV) is stated in Table 20.1. From the point of view of chemical composition of gaseous fuels, it concerns with the exception of the mixtures of gases. Among some of the essential properties of gaseous fuels that depend on their chemical composition and that affect burner design there are LHV, amount of oxidizer necessary for stoichiometric combustion of the mixture, temperature and pressure of gas in gas distribution lines before burner inlet. If gaseous fuels contain shares of liquid carbohydrates or an amount of solid particles, it can cause, in some cases, significant problems. 20.3.2 Liquid Fuels Just like gaseous fuels, liquid fuels can be categorized as standardized and nonstandardized. Some of the standardized liquid fuels made of oil are ELFO, LFO and HFO. Most countries have strict standards on their physical-chemical properties and maximum sulfur content [18]. Mazut (i.e., black oil) is a less frequent fuel with partially standardized physical-chemical properties. All the above mentioned noble fuels are produced and freely distributed on the market. Recently there has been a decrease in combustion of HFO and mazut due to the content of sulfur and chemically bound nitrogen as they both contribute to a higher production of polluting substances [19]. Extra light fuel oil and LFO can be easily atomized by pressure, mechanically (rotary atomization) as well as by pneumatic atomization utilizing compressed air. There is no need for preheating ELFO since its viscosity is sufficiently low. Light fuel oil does not require preheating over 50°C. Since HFO and mazut are highly viscous substances that require preheating over 100°C, pneumatic atomization utilizing steam is more suitable because pressure atomization necessitates high pressures.
© 2011 by Taylor and Francis Group, LLC
133.9
Nonstandardized fuels are the next category of liquid fuels. These fuels come as wastes from various productions or from renewable sources [20]. As to the waste, these are mostly residues from oil and coal processing, polluted liquid carbohydrates, and so forth. As to the liquid fuels from renewables, there are pressed vegetable oils and oils produced as a by-product of biomass gasification. In addition to those, there are wastes coming from biofuels production (such as rape-seed oil methyl ester) and used vegetable oils [21]. Figure 20.5 shows a flame during the combustion of crude glycerin containing methanol and potassium hydroxide in a dual-fuel burner test, in which the stabilization fuel is combusted natural gas. After the test the walls of the combustion chamber are covered by a layer of potassium oxide (see Figure 20.5). The dual-fuel burner is shown pictorially in Figure 20.6. All the above mentioned fuels have certain factors in common (i.e., variable and often unstable chemical composition). This means changing physical-chemical properties and possible presence of large amounts of solid particles. Often there are also dissolved inorganic compounds present. A comparison of selected properties of ELFO, LFO, HFO, rapeseed oil, and its corresponding transesterification products with methanol and ethanol is shown in Table 20.2. Opting for the right type of atomization should be done with certainty, which means that in most cases pneumatic atomization utilizing compressed air or steam is preferred. Pressure atomization can be implemented if the waste has a low viscosity [25]. Combustion of nonstandardized fuels gives rise not only to problems of their atomization. Other problems are related to the combustion of liquid mixtures that are due to the presence of some of the chemical components of a low flash temperature but high viscosity. Possibly it concerns the mixtures with flashpoint that is greatly below the freezing point. Waste incineration with a low heating value might
417
Power Burners
Figure 20.5 (a) An example of flame during the test of the dual-fuel burner. (b) The walls of combustion chamber covered with a layer of Potassium Oxide after the test. Combustion air Flame stabilizer Liquid fuel and atomizing fluid
Gas gun
Primary air
Gaseous fuel Oil gun
Secondary air
Figure 20.6 Experimental staged air dual-fuel (gas and oil) burner.
face unstable burning, incomplete combustion, and high emissions of carbon monoxide and carbohydrates. In this case it is advisable to use standardized gaseous or liquid fuel to secure stable combustion instead.
the usage of typical power burners for combustion of solid fuels is very rare and the burners usually combine solid fuel with gas or oil as a second fuel.
20.3.3 Solid Fuels Combustion of standardized solid fuels is not very common outside the power industry, heating industry, cement industry (coal especially), and metallurgical industry (coke especially). It is dues to their complicated storage, preparation, and dosing into the combustion chamber; furthermore they are affected by restrictions on the maximum concentration of polluting substances, especially sulfur oxides and nitrogen oxides. Dosing of solid and other pseudosolid fuels (paste-like) is performed via several types of equipment that are hard to define as burners. All the above mentioned facts demonstrate that
© 2011 by Taylor and Francis Group, LLC
20.4 Oxidizers Combustion is an exothermic chemical reaction during which reactive substances oxidize. In this respect, the presence of oxygen is an essential requirement for combustion. Atmospheric air is the most widespread and cheapest oxidizer. However, in various sectors it is possible to see that air is enriched with oxygen or that pure oxygen is used during combustion [26]. Both oxygen-enriched air and pure air are used mainly in areas where it is required to achieve high flame temperatures
418
Industrial Combustion Testing
Table 20.2 A Comparison of Some Typical Properties of ELFO, LFO, HFO, Rapeseed Oil and Ester Fuels Rapeseed Oil Property
Extra-Light a Fuel Oil
Specific gravity Kinematic viscosity [mm2/s] Flash point (Pensky and Martens)e [°C] Flash point in open crucible [°C] Freezing point [°C] Ash content [wt %] Sulfur content [wt %]
0.86 at 20°C max. 6
Mechanical impurity content [wt %] Higher heating value [MJ/kg] Lower heating value [MJ/kg]
max. 0.05
a b c d e
b
Light Fuel Oil
a
3.2–18
c
c
Methylester
0.99
0.91
0.880
0.876 at 40°C
0.886
at 80°C at 100°C at 150°C
max. 118 max. 57 max. 10
Ethylester
Distilled d Methylester
Oil
0.91 at 40°C
c
Heavy Fuel Oil
min. 56
min. 66
–
51 –
5.65 179
6.17 124
4.19 184
–
–
min. 110
–
–
–
–
max. −10
summer max. 10 winter max. −5 max. 0.02
max. 40
–21
–15
−10
–
max. 0.14
–
0.002
0.002
–
0.01
0.012
0.014
max. 1·10−6
max. 0.1
lowsulfur max. 1.0 midsulfur max. 2.0 highsulfur max. 3.0 max. 1.0
–
–
–
max. 5·10−6
–
41.5
42–46
40.17
40.54
40.51
–
42.5
44–46
40.9
–
–
–
–
max. 0.01 max. 0.2
lowsulfur midsulfur
max. 1.0 max. 2.0
Adapted from the material safety data sheet of UNIPETROL RPA, s.r.o. (http://www.unipetrolrpa.cz/en/). Adapted from the material safety data sheet of company G7, Inc. (http://www.g7.cz/en/o-spolecnosti.php). Adapted from Klass, D. L., Biomass for Renewable Energy, Fuels, and Chemicals, San Diego, CA: Academic Press, 1998. Adapted from the material safety data sheet of company Agrochem a.s. (http://www.agrochem.cz/o-nas/). According to industry standard ISO 2719: Determination of flash point - Pensky-Martens closed up method.
to increase the heat transfer by radiation (e.g., glass and metallurgical industries for which special burners are designed). As for the installation of power burners, atmospheric air is mostly used to oxidize the fuel. This is mainly due to economic reasons because the production of pure oxygen is rather expensive. Provided that power burners with a forced supply of combustion air has been chosen for the technology (the combustion air fan is not part of the burner), it is possible to increase the energy efficiency of the technology by preheating the combustion air [27]. In case it is required to suppress the formation of NOx, it is further possible to provide primary abatement by external and internal flue gas recirculation burners. For common applications (if preheating is required) it is sufficient to preheat combustion air within the range of 150–300°C. Sometimes a requirement for preheating the air up to 450°C may also occur. The requirement to preheat the air to more than 300°C is mostly based on technological grounds when it is necessary to achieve as high flame temperatures as possible. It is problematic to preheat the air to higher temperatures not only because of the construction of recuperative heat exchangers but
© 2011 by Taylor and Francis Group, LLC
also because the formation of thermal nitrogen oxides increases exponentially. The increased formation of nitrogen oxides is caused by the increased temperature of the flame compared to the temperature when combustion air is not preheated. The life span of the metal parts of the burner is another limitation resulting from the temperature to which combustion air is preheated, and from the increased temperature of the flame relating to it. Using external or internal flue gas recirculation is another construction option provided by power burners; internal flue gas recirculation prevails for power burners. Reduction of NOx emissions is currently the main reason for using flue gas recirculation during which the partial pressure of oxygen is reduced due to mixing flue gas with air, which significantly suppresses the formation of thermal nitrogen oxides. As for external flue gas recirculation, cold flue gas is taken away before it enters the chimney, and mixed with combustion air before it enters the combustion space, or delivered individually to certain places of the combustion chamber in a targeted and focused way. If cold flue gas is mixed with air, the partial pressure of oxygen as well as maximum temperatures of
419
Power Burners
the flame are reduced because a larger amount of mass is heated by the same amount of energy. The formation of NOx may be reduced within the range of 20–75% [7] by external flue gas recirculation, depending on the primary concentration of NOx, the recirculation ratio and the type of combusted fuel. However, it is necessary to be aware that external flue gas recirculation reduces flame stability and that greater mass flows of flue gas affect hydraulic and thermal conditions during convective heat exchange. Recently, producers of process burners have increasingly introduced burners with internal flue gas recirculation to the market. Burners with internal flue gas recirculation use the ejection effect of combustion air to suck in flue gas from within the area of the combustion chamber; the gases are immediately mixed with the air in the body of the burner, which reduces the partial pressure of oxygen in the area of the flame. Significantly lower investment costs of the technology are the main advantage to the use of burners with internal flue gas recirculation. The NOx emissions are substantially reduced, up to cca 40%, in relation to burners with internal flue gas recirculation, which mainly concerns burners using gaseous fuels when preheating combustion air to higher temperatures. In case the air is not preheated, burners with flue gas recirculation do not have good results, and NOx emissions are reduced by 10–20%. When liquid fuels are combusted, the overall impact of internal flue gas recirculation depends on the amount of nitrogen contained in the fuel; in the case of HFO, for example, NOx is reduced by a maximum of 20% when preheated air is used. Recirculation is not really important when liquid fuels are burnt and air is not preheated: NOx emissions are reduced by a few percentage. As in the case of external flue gas recirculation, the length and stability of the flame may be affected. Waste gas from various technologies is used as an oxidizer in extraordinary cases where there is a considerable concentration of either oxygen or air volume. These are mostly facilities disposing of undesirable waste gases containing hydrocarbons, for example. In these cases, it is most important to dispose of undesirable components. Regarding the fact that it is generally not possible to ensure a stable composition of these waste gases, combustion control is more difficult as well. The problem is mostly dealt with by a very sophisticated control system or, more frequently, by setting a higher air equivalence ratio.
20.5 Power Burner Testing Despite the progress made in the area of numerical simulations, development of computers and simulation programs in general, it is not possible to discontinue
© 2011 by Taylor and Francis Group, LLC
physical testing of burners. This is caused by the fact that currently available simulation software capable of providing results within a reasonable time is not sufficiently accurate and reliable, or able to detect limiting states. It is further necessary to realize that the manufacturer may only guarantee values actually measured under conditions corresponding to real conditions, or at least similar to those in a real combustion facility. In many developed countries, tests for certain groups of burners are governed by binding regulations and standards that do not only specify the type of facility but also describe testing of limit states, tests of safety features, and tests of the burner control unit in a very precise manner. Tests have to be carried out in the course of the development, presentation, and delivery of the burner to the customer (or during possible certification if necessary); these tests should unambiguously determine safe maximum and minimum heat output of the burner, flame characteristics (i.e., visible and invisible lengths, diameter, structure, stability), noise emissions, amount of pollutant emissions (nitrogen oxides in particular), for example. As long as the results of measurements performed on the testing facility are to correspond with real values, it is necessary to properly select the testing facility allowing for realistic simulation of all predictable states that may occur during the operation. This is closely connected not only with the selection of the pivotal apparatus—combustion chamber, possibility to prepare a testing fuel (when it is not possible to use the original fuel), oxidizer selection, for example—but also with the economic aspect of the tests carried out because measurements are always very expensive. It is further necessary to realize that the economic profitability of the tests carried out is closely connected with the size and equipment of the testing facility because it is a very expensive investment. The text is further divided into individual areas such as combustion chambers, fuels, instrumentation, and control, for example, because it is very difficult to say which of the areas has a decisive influence on the representative result of the test and which of them is less important. 20.5.1 Combustion Chamber Selection Two basic types of combustion chambers may be seen when power burners are tested. The first group consists of completely insulated adiabatic chambers. The second group is combustion chambers allowing for part of the heat released by combustion to be absorbed in the cooling medium (see Figure 20.7). Water is typically used as the cooling medium, which circulates through the area between the shells of the combustion chamber whereas the internal shell on the side of the flame is not
420
Industrial Combustion Testing
chamber is inappropriate (too big or too small) the results of emission measurements as well as the outlet temperature of flue gas, for example, are distorted. 20.5.2 Pressure in the Combustion Chamber
Figure 20.7 Horizontal combustion chamber with a water-cooled shell.
insulated. Installation of rows of tubes along the walls of the combustion chamber is another possibility. In the case of some designs of combustion chambers allowing for the absorption of heat released by combustion, internal water-cooled shells and rows of tubes are divided into sections [28]. When the number and size of the sections is appropriate for the size of the combustion chamber, the segmentation helps to determine local heat load into the walls of the combustion chamber (e.g., along the length of the flame). The identified values help to optimize the burner so that, for instance, there would be no excessive unbalanced heat load on tubes in heating and reaction furnaces used in chemical and petrochemical industries. The orientation of testing combustion chambers, the installed burner, and the number of burners that can be installed in the testing chambers, are other possible factors according to which testing combustion chambers may be classified. Both horizontal and vertical chambers can be encountered; vertical chambers are predominantly used for testing burners of higher heat output or when such orientation of a burner is expressly required. Testing chambers allowing for installation of several burners are not very common. They are applied for tests that aim to determine how installed burners affect one another during operation (simulation of reaction and heating furnaces). Further, in connection with the selection of the type and orientation of the combustion chamber (insulated or cooled, horizontal or vertical), it is necessary to properly select the size of the chamber with respect to the volumetric and surface heat load of the target application of the power burner. Last but not least, shape of the flame should be taken into account, so that the walls would not be touched by the flame. If the combustion
© 2011 by Taylor and Francis Group, LLC
Negative operating pressure and overpressure relative to ambient atmospheric pressure is another factor affecting the behavior of the burner in the combustion chamber. In the case of standard applications, negative pressure and overpressure values range from minus 200 Pa up to 1000 Pa but requirements for negative pressure of up to minus 3000 Pa and overpressure of dozens of kPa may occasionally occur in the area of using power burners. In order to generate overpressure inside the combustion chamber, the flow of flue gas is throttled behind the outlet from the combustion chamber. The throttling mechanism has to be placed in such a position so as not to affect the flow inside the combustion chamber. Flap valves are most commonly used as throttling mechanisms. In case it is required to generate negative pressure inside the combustion chamber (whereas the negative pressure is higher than the natural stack draft), it is necessary to install a unit in the flue gas exhaust able to generate negative pressure (i.e., an ejector or a flue gas fan). Both the flue gas fan and ejection air fan need to be equipped by suitable control to allow keeping the required negative pressure at a constant value. From the point of view of field instrumentation for measuring pressure inside the combustion chamber, it is possible to use either a glass U-tube manometer filled with fluorescent liquid or an industrial manometer for local measurements and an electronic pressure sensor for remote measurements and collection of data. Reliability and ability to dampen the pressure fluctuation inside the combustion chamber are the advantages of the U-tube and robust industrial manometers. 20.5.3 Fuels Power burners are designed to burn various fuels, either gaseous or liquid ones. This implies requirements for sufficiently extensive fuel management facilities that should not only have sufficient capacity but also a possibility to prepare fuel mixtures replacing original fuels. 20.5.3.1 Gaseous Fuels As already mentioned, in addition to commonly available commercial fuels (natural gas and LPG), other various mixtures of gases are burned. These mixtures of gases include waste products of various productions (refinery gas, stack gas, coke oven gas, etc.) or fuels that
421
Power Burners
are the products of various technological procedures focused directly on the production of these gases for a given local source (biogas, gasification, mining gas, etc.). With respect to the great variety of chemical compositions of real mixtures of gases, it is very difficult to create an identical mixture in the testing facility. Therefore, so called testing fuels are mixed based on the knowledge of the chemical composition of a real fuel; these testing fuels consist of main components of a real fuel. Natural gas, hydrogen, nitrogen, carbon dioxide, carbon monoxide, propane, butane, and other gases (e.g., ethylene) when technically possible, are among the main gases used to prepare testing mixtures. The above mentioned implies the requirements on the gas-mixing equipment to create gas mixtures whose characteristics correspond to a real fuel. With the exception of natural gas that is supplied through a central distribution system, the burner testing facility has to be equipped with other facilities allowing not only to store other gases but also to prepare them to be dosed into a burner or mixed in a mixing station. It is not economic for small testing or research facilities to build fuel management facilities of gaseous fuels allowing storing gases in the liquid state. Therefore, bundles of bottles are used that already have sufficient capacity; the risk of gas leakage is minimized, too. In the case of large testing facilities allowing testing burners of high heat output it is more practical and safe, with respect to the transport of gases, to supply and store gases in the liquid state. Fuel management facilities are then established near the testing site; they consist of liquid gas reservoirs and evaporator stations in which liquid gas is evaporated and from which it is subsequently supplied to the testing facility. 20.5.3.1.1 Testing Fuel Preparation The mixture of gases prepared to substitute a real gaseous fuel needs to be of such quality as to replace it as authentically as possible. The chemical composition, higher and lower heating value, and stoichiometric consumption of the oxidizer per unit amount of fuel are among the basic qualities of gaseous fuels important for combustion. In the case of testing fuels prepared by mixing of the most important components of the original fuel, it is necessary to try to adhere, at least, to the basic qualities mentioned above. The Wobbe Index and adiabatic flame temperature may be further used as other compliance criteria. 20.5.3.1.2 Instrumentation As for the actual test, instrumentation of the testing facility concerning the gas management facility needs to meet requirements for the possibility to dose individual fuels or their components in a precisely controlled manner. This means that it is necessary to ensure
© 2011 by Taylor and Francis Group, LLC
precise measurements of the flow rate, temperature, and pressure in order to be able to convert them to reference conditions. There are no special requirements for the types of measuring instruments used to measure gas temperature and pressure. This is however not case when measuring flow rate. Flowmeters have to be not only accurate but also resistant to hydraulic shocks. From this point of view, using orifice gauges seems to be the most reliable way; these gauges are accurate and sufficiently resistant and can be easily switched according to the required flow rate and the fuel measured. The basic geometry of orifice gauges used to measure flow rate has been standardized (e.g., ISO-5167:2003) and the orifice gauges have to be subsequently calibrated for individual gases or groups of gases. Flowmeters containing rotating parts (such as turbines or rotary piston gas meters) are less suitable; although they guarantee precise flow rate measurements on a large measurement range, they can be easily damaged by pressure shock and impurities contained in the gas. 20.5.3.2 Liquid Fuels Sufficiently low viscosity is a prerequisite for good atomization of liquid fuels and wastes when they enter the jet or the atomizer because if liquid is not atomized into sufficiently small drops, it does not burn out completely. The liquid fuel management facility for testing power burners should therefore store, mix, prepare, and dose liquid fuels while keeping high quality standards in all steps. 20.5.3.2.1 Testing Fuel Preparation In comparison with gaseous fuels, original liquids can be transported to the testing facility, and the fact that they cannot be used during the testing of a burner, is rather a matter of the possibility of long-distance transport because in most cases they are liquids falling under the category of dangerous substances. If the test cannot be performed using an original fuel, it is necessary to try to find a substitute fuel that would be as authentic as possible and that would meet at least the basic qualities of the original fuel (HHV, lower heating value, stoichiometric consumption of air, etc.), which is typically not very difficult for petroleum-based liquids. However, if it is required to burn liquid wastes that include not only various mixtures of hydrocarbons but also mixtures of various residues from industrial productions, it is possible to try to prepare a testing fuel by mixing liquids available on the market (petrol-alcohol, petrol-ether mixtures, etc.). If this cannot be done because it would be difficult to mix the liquids in a tank, it is possible to mix the liquids by means of a static mixer (see Figure 20.8) just before they enter the burner.
422
Industrial Combustion Testing
Manometer Manometer Burner Liquid fuel 1
Liquid fuel 2 Figure 20.8 A static-mixer.
20.5.3.2.2 Dosing Facility The dosing facilities for liquid fuels have to respect the variety of physical and chemical characteristics of the liquids burned (i.e., contrary to common industrial facilities, they have to be sufficiently flexible). The most important characteristics of the liquids in terms of the dosing facility, concern the pumping of viscous liquid at the ambient temperature and temperature at which it is possible to atomize it well. Further, it is the temperature of the liquid in relation to its flash point because a lot of the liquids burned have to be preheated above their flash point. The way of preheating is a closely related issue because it is better, for safety reasons, to choose to heat liquids indirectly by means of a nonflammable, heat-transferring medium (such as steam). It is advisable to use centrifugal pumps to pump liquids with very low viscosity at the ambient temperature as they do not require the liquids to have a capacity to lubricate them. However, if there is a requirement to pump high-viscosity liquids at the ambient temperature, it is advisable to use screw or gear pumps that are, nevertheless, very sensitive to the amount of impurities contained in the liquid, and may wear out when the viscosity of the liquid is too low. Most importantly of all, the selection of control units should allow for sensitive regulation of the flow rates. 20.5.3.2.3 Instrumentation Similarly to gaseous fuels, an important role is played by the instrumentation allowing for precise measurements of process values. However, in comparison with gaseous fuels, there are much higher requirements for the robustness of individual measuring instruments, especially as regards to pressure sensors and flow rate meters that not only have to be resistant to pressure shocks but also have to resist temperatures and physical-chemical properties of liquids. In the case of manometers and pressure sensors it is necessary to avoid that sludge settles down and solidifies inside their bodies. Separators filled with silicon oil are used for that purpose. In order to measure flow rates, it is necessary to choose flowmeters allowing for measurements of flow rates
© 2011 by Taylor and Francis Group, LLC
of various liquids, but it is very difficult to determine objectively what ways are most suitable. Orifice gauges, rotary volumetric flowmeters, and mass flow meters are among the most often used flowmeters. Accurate physical data about the passing liquid is necessary to measure flow rates properly by means of gauges whose construction is simple and nearly maintenance-free; this data is greatly dependent on the temperature of the liquid. Further, sediments often get caught in orifice gauges and the accuracy of measurements is reduced. In comparison with orifice gauges, rotary volumetric flowmeters are not as sensitive to the physical characteristics of the passing liquid. The pressure drop of flowmeters is dependent only on the viscosity of the liquid. Nevertheless, they are much more sensitive to pressure shocks that may damage them. As regards to impurities contained in the liquid measured, they are only able to work as long as the elements do not exceed a certain size level set by the producer. Mass flowmeters working on the principle of Coriolis force are another possibility that is often used. However, in the case of the testing facility it is necessary to know whether it is possible to heat them and, eventually, to clean them from solidified liquids. Contactless flowmeters performing measurements by means of ultrasonic and induction flowmeters need accurate physical and chemical data about the liquids passing through them in order to work properly. 20.5.4 Combustion Air Untreated atmospheric air is mostly used as the oxidizer to test power burners; before it enters the burner, it is possible to enrich it with oxygen or to preheat it. Burners with a forced supply of combustion air are fed with air supplied with a fan. It is necessary to relate the maximum overpressure of the burner and the amount of the air supplied in such a way so that it is possible to test burners even in nonstandard conditions with a high air surplus. Regarding the fact that the combustion air fan serves as a universal source to test burners with different heat outputs, it is necessary to ensure that the flow rate of the combustion air is regulated
423
Power Burners
to a sufficient range (speed regulated by a frequency converter, suction throttling or throttling on the discharge of the fan). Various kinds of preheaters are used to preheat combustion air; these preheaters are mostly stationary ones. If the burners have low heat outputs, it is possible to consider electrical preheating. In the case of burners with outputs of several hundred kilowatt or more, it is useful to choose a preheater equipped with a combustion chamber and a flue gas-air heat exchanger able to change operation modes (i.e., flow rate and preheating temperature), relatively quickly. 20.5.4.1 Instrumentation Natural and induction draft burners suck in combustion air directly from their surroundings and do not have combustion air supply pipes where a flow rate meter could be placed. That is why the total flow rate of the combustion air can only be calculated retrospectively from the chemical composition of flue gas measured just behind the outlet from the combustion chamber. Therefore, when regulating the air supply, it is necessary to take into account the time lag between the change of the setting and the reaction of the flue gas analyzer, whereas the period corresponds to the residence time of flue gas inside the combustion chamber plus the reaction time of the flue gas analyzer. Similarly to other gases, it is necessary to measure three basic values in connection with combustion air: temperature, pressure, and flow rate. In case a combustion air preheater has been installed, temperature and pressure have to be measured not only before air preheater, but also before the preheated air enters the burner. Considering the low pressures, there are no special requirements for temperature and pressure sensors as regards to their robustness, it is only necessary that the signal pipe be long enough when measuring the pressure before it enters the burner, in order to prevent damage to manometers and pressure sensors from high temperature. In the case of burners with a forced supply of combustion air, it is more accurate to measure the flow rate of the combustion air before it is preheated because it is possible to choose more precise methods of measurement. Even in this case however, the flow rate measurements supply only information and it is necessary to carry out accurate flow rate identification based on the required amount of oxygen in flue gas. Turbine flowmeters and hot wire anemometers can be classified as relatively accurate methods of measuring air flow rates in pipes. It is rather inaccurate to use classic Pitot and Prandtl tubes to perform real-time measurements because there are large fluctuations of dynamic pressures, and resulting readings require large damping. The use of orifice gauges at the pipes with large diameter is rather problematic.
© 2011 by Taylor and Francis Group, LLC
20.5.4.2 Steam and Compressed Air Steam and compressed air are additional media often used for pneumatic atomization of liquid fuels. The necessary pressure as well as the consumption per unit of atomized fuel depend on the kind of fuel burned and on the construction of the atomizer. It only pays off to use compressed air to atomize ELFOs that do not need to be preheated, and LFOs that need to be preheated to cca 50°C. If the liquid burned is preheated to a higher temperature (even above 150°C, for instance, combustion of residues from petroleum processing), it is advisable to use steam to atomize the liquid. Steam does not overcool the liquid that is being atomized, and thus eliminates the risk of imperfect atomization. In the case of small units, it is possible to use compressed air heated electrically by a flow-through heater, which is a money and space saving solution as compared to the construction of a steam generator. As regards to the choice of steam sources, we cannot recommend the use of electrical generators because they have high demands on power input. Despite substantially higher investment costs, steam generators burning gas and liquid fuels are a much more suitable option. In order to produce sufficient preheated steam for atomization by reducing the pressure, the difference between the maximum operating pressure used for atomization, and the pressure of the generated saturated steam should be large enough. If the steam is not sufficiently preheated and part of it condenses, the condensate may cause difficulties during the atomization of the fuel. 20.5.4.3 Instrumentation There are no special requirements for the instrumentation used to measure the temperature and pressure of the compressed air. Orifice gauges and hot wire anemometers measure flow rate in a relatively precise manner. As regards to steam, the instrumentation needs to resist not only pressures but also high temperature. In the case of manometers and pressure sensors it is advisable to use cooling separators filled with silicon oil. It is appropriate to use orifice gauges to measure flow rate because high temperature does not affect them.
20.6 Test Planning and Evaluation Gas and liquid fuel burners are characterized by relatively complicated geometry. In order to save money on design optimization and optimization of operating conditions of burners, there has been lately an effort to use various simulations and empirical models obtained
424
on the basis of previous burner tests. However, it is a difficult task to find empirical models that would both involve the influence of the geometry of the burner and the effect of the operating parameters (e.g., on the formation of emissions). When such models are constructed, their predictions are mostly limited to a particular type of a burner and can only be interpreted as qualified estimations. Statistical methods can be used successfully during the actual planning of burner tests and then during the evaluation of tests focusing on the investigation of the influence on individual elements of the burner for the combustion process, the formation of emissions, and the shape and length of the flame. It is ensured by the ability of these methods to analyze large amounts of data measured during experiments. A well-planned experiment is the basis for proper evaluation of experimental data by using statistical methods. Experiment planning is based on three main principles [29]: replication, randomization, and blocking. The principle of replications estimates the experimental error, thus contributing to the accuracy and reliability of the conclusions. The principle of randomization aims to prevent continuous action of a disturbing influence (e.g., operator tiredness) in the course of the experiment. Blocking consists in dividing the whole experiment in blocks so that more homogeneous experimental conditions are achieved within a block in comparison with the conditions within the whole experiment. In the preparation of the experimental plan, it is necessary to consider the objective of the experiment, the choice of the parameter subject to examination (so called response variable), the choice of the number of independent variables (so called factors), the number of levels of the factors, how many times the measurement shall be repeated, and homogeneity and variability of experimental conditions [30]. Last but not least, the plan should take into account available financial resources. We can say, in fact, that the aim of the planning is to minimize the scope of the experiment (i.e., number of measurements) and to maximize the amount and quality of the information obtained. This requirement is widely applied to experiments with a high number of factors. Burner tests are complex, time-consuming, and expensive experiments. In the case of this category of experiments, one should always put a strong emphasis on the planning supported by statistical methods. Should the examined value be influenced by a low number of factors (< 3) and should the dependence between the values be known as linear, it is sufficient to use the so called factorial plan [29] for this type of experiment, or a fractional factorial plan [29]. Nevertheless, most experiments are supposed to have a response variable without linear dependence on some of the factors, and it
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
is thus necessary to have a higher-order model (mostly a second-order one) to describe the relationship between the variables. In order to be able to estimate the regression parameters of the second-order model, it is necessary to include in the experimental plan at least three levels for each of the factors. The so called response surface method [31,32], is a highly efficient industrial tool for the identification of an adequate model. A central composite plan [33] is the most often used kind of plan to construct response surfaces; this plan requires five levels for all factors. The basis of the plan is formed by a factorial plan completed with star and central points [33]. In case there is a large number of factors, a fractional factorial plan is used instead of the factorial one in order to reduce the number of measurements, which also means considerable time and money savings. In practice, it is often difficult to change the five levels of factors; therefore, it is possible to use a modification of the central composite plan (i.e., face-centered central composite plan [34]). This plan only requires three levels for each of the factors as it locates the star points on the centers of the faces of cube [29], as shown in Figure 20.9 for three factors. Prior to testing a burner, a test sheet must be developed. With a well-developed test sheet resembling the one shown in Table 20.3, meaningful data can be collected. Statistical software (e.g., MINITAB [35], STATISTICA [36] etc.) is the most suitable method to design the experiment, evaluate experimental data and construct the empirical model because it provides a tool for experimental planning. The software is also able to graphically display the effects of individual factors on the response variable, as shown in Figure 20.10. The graphs determine the setting of the factors easily to achieve the optimal value (minimum or maximum) of the response variable. On the other hand, these graphs do not provide any information about the course of the combustion process and the shape and stability of the flame. These characteristics then have to be experimentally verified.
X3,max
X3
X2 X1
X2,max X3,min
X2,min X1,min
X1,max
Figure 20.9 A face-centered central composite plan for three factors.
425
Power Burners
Table 20.3 Test Sheets Basic Specifications Date Type of burner Fuel Specifications Fuel Lower heating value Molecular weight Density
Time Air-supply Natural Gas 35,811.92 16.340 0.729
[kJ/m3N] [kg/kmol] [kg/m3N]
Burner Geometrical Setup
Secondary Nozzles
Diameter of swirl generator Pitch angle of swirl generator blades Diameter of primary gas throttle Pitch angle of nozzles Quantity of nozzles Diameter of nozzles Position: Axial position Radial position Tangential orientation
mm ° mm ° – mm mm mm mm
Test Measured Variable
Unit
PID
Fuel Flow rate Temperature Overpressure at burner Heat duty
m3/h °C kPa kW
F60 T12 P39 Q91
Air Flow rate Temperature Overpressure at air fan
m3/h °C Pa
F61 T15 P48
Pa
P47
°C % ppm ppm ppm ppm
T09 Q92 Q93 Q94 Q95 Q96
m m –
– – –
Flue Gas Negative pressure in combustion chamber Temperature O2 CO NO NO2 UHC Flame Characteristics Diameter Length Stability
1
It is important to verify the adequacy of the model in comparison with the measured values, and to analyze residuals after the empirical model is constructed. The adequacy of the model can be assessed by using the so called F-test [37], which compares two types of
© 2011 by Taylor and Francis Group, LLC
Measurement No. 2
3
variances. The first variance indicates how the measured values fluctuate around the average. This variance is only defined when the measurement is performed repeatedly. The second variance defines the distance of the measured values from the calculated ones (i.e., this
426
Industrial Combustion Testing
6,50
45
44 42
6,25
X2
44 6,00 43
40
Y
5,75 40 5,50 1,100
41
41 1,125
1,150
1,175
39
42 40 38
1,200
X1
1,100 1,125 1,150 1,175 1,200 X2
6,50 6,25 6,00 X1 5,75 5,50
Figure 20.10 Information contour and surface plots of the response for two factors.
variance has direct connection with the model). The F-test determines whether these two variances differ from each other in a statistically significant way. Further, it is important to perform tests concerning residuals. These are tests of normal distribution, plot of residuals versus the order of measurements (or time), and plot of residuals versus fitted values.
References
1. England, G. C., McGrath, T. P., Gilmer, L., Seebold, J. G., Lev-On, M., and Hunt T. “Hazardous Air Pollutant Emissions from Gas-Fired Combustion Sources: Emissions and the Effects of Design and Fuel Type.” Chemosphere 42 (2001): 745–64. 2. Baukal, C. E., ed. Industrial Burners Handbook. Boca Raton, FL: CRC Press, 2004. 3. Baukal, C. E., ed. The John Zink Combustion Handbook. Boca Raton, FL: CRC Press, 2001. 4. Reese, J. L., Moilanen, G. L., Bowkowicz, R., Baukal, C., Czerniak, D., and Batten, R. “State-of-the-Art of NOx Emission Control Technology.” ASME paper 94-JPGCEC-15. Proceedings of International Joint Power Generation Conference. Phoenix, AZ, 1994. 5. Ballester, J. M., Dopazo, C., Fueyo, N., Hernandez, M., and Vidal, P. J. “Investigation of Low-NOx Strategies for Natural Gas Combustion.” Fuel 76 (1997): 435–46. 6. Sobiesiak, A., Rahbar, A., and Becker, H. A. “Performance Characteristics of the Novel Low-NOx CGRI Burner for Use with High Air Preheat.” Combustion and Flame 115 (1998): 93–125. 7. Baltazar, J., Carvalho, M. G., Coelho, P., and Costa, M. “Flue Gas Recirculation in a Gas-Fired Laboratory Furnace: Measurements and Modeling.” Fuel 76 (1997): 919–29. 8. Baukal, C. E. Industrial Combustion Pollution and Control. New York: Marcel Dekker, Inc., 2004. 9. Lefebvre, A. H. Atomization and Sprays. Boca Raton, FL: Taylor & Francis/CRC Press, 1989.
© 2011 by Taylor and Francis Group, LLC
10. Bayvel, L., and Orzechowski, Z. Liquid Atomization. Boca Raton, FL: Taylor & Francis/CRC Press, 1993. 11. Beér, J. M. “Combustion Technology Developments in Power Generation in Response to Environmental Challenges.” Progress in Energy and Combustion Science 26 (2000): 301–27. 12. Beér, J. M. “Minimizing NOx Emissions from Stationary Combustion: Reaction Engineering Methodology.” Chemical Engineering Science 49 (1993): 4067–83. 13. Bébar, L., Kermes, V., Stehlik, P., Canek, J., and Oral, J. “Low NOx Burners – Prediction of Emissions Concentration Based on Design, Measurements and Modeling.” Waste Management 22 (2002): 443–51. 14. Ballester, J. M., Sanz, A., and Gonzales, M. A. “Investigation of the Characteristics and Stability of AirStaged Flames.” Experimental Thermal and Fluid Science 32 (2008): 776–90. 15. Klemeš, J., Bulatov, I., and Cockeril, T. “TechnoEconomic Modeling and Cost Functions of CO2 Capture Processes.” Computers & Chemical Engineering 31 (2007): 445–55. 16. Kermes, V., Stasta, P., Šikula, J., Oral, J., Martinak, P., and Stehlík, P. “Substituting Fuel by Mining Gas in Unit for Thermal Treatment of Sludge.” Proceedings of 22nd Annual International Conference on Incineration and Thermal Treatment Technologies. Orlando, FL, 2003 [Proceeding on CD-ROM]. 17. Bhoi, P. R., and Channiwala, S. A. “Optimization of Producer Gas Fired Premixed Burner.” Renewable Energy 33 (2008): 1209–19. 18. Netz, H. Heat and Steam, Industrial Boiler Handbook (Wärme and Dampf, industriekessel handbuch). Germany: Verlag Dr. Ingo Resch GmbH, 1996. 19. Hill, S. C., and Smoot, L. D. “Modeling of Nitrogen Oxides Formation and Destruction in Combustion Systems.” Progress in Energy and Combustion Science 26 (2000): 417–58. 20. Parikka, M. “Global Biomass Fuel Resources.” Biomass Bioenergy 27 (2004): 613–20. 21. Klass, D. L. Biomass for Renewable Energy, Fuels, and Chemicals. San Diego, CA: Academic Press, 1998.
Power Burners
22. UNIPETROL RPA, s.r.o. http://www.unipetrolrpa.cz/ en/ (accessed May 4, 2009). 23. G7, Inc. http://www.g7.cz/en/o-spolecnosti.php (acc essed May 4, 2009). 24. Agrochem a.s. http://www.agrochem.cz/o-nas/ (accessed May 4, 2009). 25. Jedelsky, J., and Jicha, M. “Design of Atomizer for Waste Fuel Combustion with Decreased Exhaust Gas Emissions.” 31st International Symposium on Combustion. Heidelberg, Germany, 34, 2006. 26. Joshi, S. V., Becker, J. S., and Lytle, G. C. “Effects of Oxygen Enrichment on the Performance of Air-Fuel Burners.” In Industrial Combustion Technologies, edited by M. A. Lukasiewicz, 165–70. Materials Park, OH: American Society of Metals, 1986. 27. Tsuji, H., Gupta, A. K., Hasewaga, T., Katsuki, M., Kishimoto, K., and Morita, M. High Temperature Air Combustion: From Energy Conservation to Pollution Reduction. New York: CRC Press, 2003. 28. Kermes, V., Skryja, P., and Stehlik, P. “Up to Date Experimental Facility for Testing Low-NOx Burners.” Proceedings of the 10th Conference on Process Integration, Modeling and Optimization for Energy Saving and Pollution Reduction, 549–54. Ischia Porto, Italy, 2007. 29. Montgomery, D. C. Design and Analysis of Experiments. New York: John Wiley & Sons, 1991.
© 2011 by Taylor and Francis Group, LLC
427
30. Wu, C. F. J., and Hamada, M. Experiments: Planning, Analysis, and Parameter Design Optimization. New York: Wiley, 2000. 31. Chen, X., Du, W., and Liu, D. “Response Surface Optimization of Biocatalytic Biodiesel Production with Acid Oil.” Biomechanical Engineering Journal 40 (2008): 423–29. 32. Yuan, X., Liu, J., Zeng, G., Shi, J., Tong, J., and Juany, G. “Optimization of Conversion of Waste Rapeseed Oil with High FFA to Biodiesel Using Response Surface Methodology.” Renewable Energy 33 (2008): 1678–84. 33. Mason, R. L., Gunst, R. F., and Hess, J. L. Statistical Design and Analysis of Experiments: With Applications to Engineering and Science. New York: John Wiley & Sons, 2003. 34. Beˇlohradský, P., Kermes, V., and Stehlík, P. “Design and Analysis of Experiments for Low-NOx Burners Design for Process Industries.” Proceedings of 8th European Conference on Industrial Furnaces and Boilers. Vilamoura– Algarve, Portugal, 2008 [Proceeding on CD-ROM]. 35. Minitab, Inc. http://www.minitab.com/ (accessed May 4, 2009). 36. StatSoft, Inc. http://www.statsoft.com/ (accessed May 4, 2009). 37. Milton, J. S., and Arnold, J. C. Introduction to Probability and Statistics: Principles and Applications for Engineering and the Computing Sciences. New York: McGraw-Hill, 2003.
21 Regenerative Combustion Using High Temperature Air Combustion Technology (HiTAC) Ashwani K. Gupta, Susumu Mochida, and Tsutomu Yasuda Contents 21.1 Introduction.................................................................................................................................................................. 429 21.2 Preheated Air Combustion......................................................................................................................................... 430 21.3 Air Preheating Method............................................................................................................................................... 430 21.4 Energy Savings............................................................................................................................................................. 431 21.5 Examples of Applications to Combustion System.................................................................................................. 433 21.6 Key Features of HiTAC Flames.................................................................................................................................. 433 21.7 Type of Regenerative Burners: Specific Features, Structure, and Application................................................... 435 21.7.1 History of How Regenerative Burners and Their Dissemination in the Industries............................ 435 21.8 Types of Regenerative Burners.................................................................................................................................. 437 21.8.1 Alternative Combustion-Type Regenerative Burner................................................................................. 437 21.8.2 Radiant-Type Regenerative Burner............................................................................................................. 439 21.8.3 Single-Type Regenerative Burner................................................................................................................ 441 21.9 Key Points for Test Furnace Design (by Nippon Furnace, Japan)......................................................................... 442 21.9.1 Visual Observation, Photo, and Video Recording of the Flame............................................................. 442 21.9.2 Flame Temperature Measurement.............................................................................................................. 442 21.9.3 Measurement of Heat Transfer and Heat Flux of Heated Material........................................................ 443 21.9.4 NOx Measurement......................................................................................................................................... 444 21.9.5 Measurement of Temperature Distribution in the Furnace.................................................................... 444 21.9.6 Measurement of Gas Component in the Furnace..................................................................................... 445 21.9.7 Modeling of Furnace Used for Commercial Application........................................................................ 445 21.10 Summary....................................................................................................................................................................... 446 References................................................................................................................................................................................. 447
21.1 Introduction In industrial furnaces, various methods have been established to improve the thermal efficiency and reduce environmental pollution. Some of these methods for improving efficiency are shown in Figure 21.1 [1]. For the case of continuous heating furnaces, the energy loss associated with exhaust gases, the furnace body, and cooling water can result in a significant decrease of thermal efficiency. A reduction in these losses can improve the furnace thermal efficiency. A reduction of fuel or hydrocarbon species in the exhaust gases results in a simultaneous reduction of pollutants emission and an increase of combustion efficiency. The use of recuperators for extracting energy from the exhaust gases has been one of the most popular approaches
for waste energy recovery for a considerably long time in boiler and ceramic kiln systems. In this approach the exhaust heat is used to preheat the incoming combustion air. The use of a recuperator has been the traditional approach for preheating the combustion air. This is merely a gas-gas heat exchange method. However, for some special furnaces that require high-temperature flames, a regenerator is used. In either case the sensible heat from the exhaust gases is extracted by the heat exchanger. This extracted thermal energy becomes part of the furnace heat input and directly represents the amount of energy that can be saved with the other parameters remaining unchanged. The extracted energy can be used in several ways (e.g., preheating the air or fuel, water heating, steam generation). From the point of energy recovery, the effect is the same as long as the total energy recovery is the same in each case. However, for the case of air preheat the 429
© 2011 by Taylor and Francis Group, LLC
430
Industrial Combustion Testing
Establishment of optimal operation
Decrease in heat loss
Improvement of thermal efficiency
Improvement of heat transfer performance
Improvement of operation capacity Operation with optimal heating capability
Decrease in temperature of furnace end exhaust gas
Improvement in heat pattern
Decrease in cooling water heat loss
Decrease in exhaust gas quantity
Low air ratio combustion
Decrease in furnace wall diffusion heat and radiant flame loss (blowoff and intake loss)
Strengthening the heat insulation of furnace wall
Decrease in exhaust gas heat loss
Improvement of fireplace shape Proper placement of burner Improvement of burner
Exhaust heat recovery
Extension of fireplace length
Utilization of exhaust gas sensible heat Utilization of cooling water sensible heat
Optimization of heating speed
Clogging the opening of furnace wall Operation at proper fireplace pressure
Installation of air preheater Installation of economizer Installation of exhaust heat boiler
Figure 21.1 Methods of improving thermal efficiency in heating furnaces.
adiabatic flame temperature will be increased and result in a subsequent increase of the cycle thermal efficiency. Combustion with elevated temperature air has been shown to provide potential for many engineering realities for the twenty-first century. In an earlier version of high temperature air combustion (HiTAC), excess enthalpy combustion was used. Excess enthalpy combustion is different from highly preheated air combustion. In the former case the enthalpy of the reactants is increased while in the later case the temperature of the incoming combustion air is increased without increasing the temperature in the combustion zone. Thus the combustion conditions in the case of excess enthalpy combustion and HiTAC are much different.
preheated using heat and mass exchange from the hot combustion gases to the incoming air. This new method does not require any catalyst or reaction promoters for the chemical reactions to occur. The temperature is often higher than the auto-ignition temperature of the mixture so that it is not a fixed value but rather dependent on the fuel being used. This is very different than preheated air combustion in which the flame temperature is increased. Air preheats have been used for the combustion of low grade fuels. In contrast no flame stabilizer is necessary since the mixture is incurs auto-ignition of the mixture under high temperature air combustion conditions.
21.3 Air Preheating Method 21.2 Preheated Air Combustion Preheated air combustion can provide energy savings, downsizing of the equipment, and reduction of pollutants emission, all of which occur simultaneously. Note that high temperature air flames use normal air as input to the combustion chamber or furnace. It simply means a new way to introduce the combustion air prior to its mixing and chemical reaction with the fuel. The new introduction method of air requires the air to be preheated with the enthalpy exchange between the exit gases from the furnace section and fresh incoming air. So the air is
© 2011 by Taylor and Francis Group, LLC
In most industrial heating devices, the sensible heat of the flue gas is circulated back in the system with the aid of some heat exchange device. Preheating of the incoming combustion air is an example of heat circulation, see Figure 21.2. The conventional recuperator or regenerator can be incorporated into the system independent of the combustor. In this case, even though a conventional air preheating system is mounted independently of the combustor, it forms an integral part of the combustor. A cross-sectional view of the DEKA recuperative heat exchanger [2] is shown in Figure 21.3. The solid walls in the center act as a barrier. The flue gases flow from
Regenerative Combustion Using High Temperature Air Combustion Technology (HiTAC)
one side (top) and exit at the bottom. The combustion air flows from the left and exits from the right. The heat is mutually exchanged by means of conduction and convection. Ljungström regenerative heat exchanger [3] shown in Figure 21.4 is intended to heat the solid by the fuel gases and then exchange the heat to the combustion
Heat exchanger
431
air via direct contact with the solid surface. A continuous heat exchange takes place in the rotating regenerator moving at a fixed speed so that the heating of solid wall material in the regenerator and air preheating occurs consecutively. Figure 21.5 shows an example of the recuperative heat exchange arrangement with the burner [4]. It is very compact in its appearance as compared to the independent mounting type.
Blower
21.4 Energy Savings
Q Preheated air
P Fuel F
The fuel-saving rate, S, that represents the energy-saving effects, is obtained by preheating the combustion air. The fuel energy savings can be calculated as follows. The available heat HA and HB with no air preheats and with air preheats, respectively, is represented by:
HA = F − Q,
(21.1)
HB = F + P − Q = HA + P,
(21.2)
where F = heating value of fuel (kcal/kg of fuel),
Figure 21.2 Preheated air combustion system.
320 By-cast HT tubes 260
220 140
By-cast LT tubes
Hot water washing 170
Washing system 80 Air 20°C
Glass tubes
120
Figure 21.3 Cross-section of DEKA heat exchanger structure. (From BY-CAST N.V. Company, DEKA Heater Catalog, http://www.by-cast.com/site.htm. With permission.)
© 2011 by Taylor and Francis Group, LLC
432
Industrial Combustion Testing
35
Gas in
Air out
30
25 Drive Fuel saving rate (%)
Rotor (containing heat transfer surface)
Air in
Figure 21.4 Yungstrome regeneraitive heat exchanger. (From Ohana, H., Design Hand book for Heat Exchanger, Kogaku Tosho Edition, 1004, 1990. With permission.)
Ex
11
0°C
90
15 °C
700
Case that the class 1, no.1 heavy oil is burned at excess air ratio of 20%
5
0 100
Exhaust gas
g
st
°C
00
em
t as
u ha
20
10
Gas out
ur
at
r pe
°C
00
3 e1
200 300 Preheated air temperature (°C)
400
Figure 21.6 Percent fuel saving with air preheat temperature. (From The Japan Industrial Furnace Manufacturer Association, “Promotion of EnergySaving Industrial Furnace,” 3–16, 1978.)
Air
The fuel consumption under conditions when no air preheating can be carried out are as follows:
X/HA (kg of fuel)/h,
(21.3)
where X = heat consumption rate in the furnace (kcal/h). Gas
However, under conditions when air preheating is carried out, the fuel consumption will be as follows:
X/HB = X/(HA + P) (kg of fuel)/h.
(21.4)
From the foregoing relations, the fuel-saving rate, S, caused by preheating of the combustion air becomes: Figure 21.5 Cross section of recuperative burner. (From Hosoi, K., “Development of Recuperative Burner,” Periodical of Japan Burner Research Association, no. 53, 1986, 16. With permission.)
Q = heat carried away with the flue gases per unit mass of fuel (kcal/kg of fuel), P = heat carried with the preheated air per unit mass of fuel (kcal/kg of fuel), H = available heat per unit mass of fuel (kcal/kg of fuel).
© 2011 by Taylor and Francis Group, LLC
or
S = (X/HA − X/(HA + P))/(X/HA),
(21.5)
S = P/(HA + P).
(21.6)
Therefore by using the temperature of the exhaust gas as a parameter, the relationship between the preheated air temperature and the fuel-saving rate can be predicted. The results are shown in Figure 21.6. This calculation assumes that combustion occurs at 20% excess air with JIS Class I No.1 heavy oil.
433
Regenerative Combustion Using High Temperature Air Combustion Technology (HiTAC)
21.5 Examples of Applications to Combustion System In the preheated air combustion, severe problems can occur when one increases the temperature of preheated air by increasing the heat recovery from the exhaust gases. Increase of heat recovery increases the adiabatic flame temperature [6], see Figure 21.7. The temperature can increase beyond the permissible level of the furnace material. At higher flame temperatures there is significant deterioration—and hence restriction—of thermal durability of the materials in the metallic heat exchanger. Furthermore at higher temperatures the NOx emission can exceed the regulatory levels due increased thermal NOx formation at high flame temperatures [7]. In addition if the temperature of exhaust gases decreases below the dew point, the corrosion of material in the heat exchanger becomes a problem. The sulfuric acid dew point depends on the S (sulfur) content in the fuel. Only a few parts per million of oxides of sulfur in the exhaust gases cause a significant reduction to the dew point temperature. As an example only 1 ppm of SO2 in the gas can raise the dew point by 62.5°C. Therefore some of the issues in preheated air combustion have been to solve the above problems using different fuels. In the past, excessive increase in flame temperature has been avoided by imposing some restrictions on the rate of exhaust heat recovery. Low NOx burners have been developed by suppressing the peak flame temperature (or hot spot regions) and by eliminating the increase in overall flame temperature. Measures on corrosion prevention of the materials have been taken, for example, by using glass tubes in place of metals in the low-temperature region of the heat exchanger. This is illustrated in the cross-sectional view of a recuperative heat exchanger shown in Figure 21.3. However, the thermal durability and combustion characteristics limit the rate of exhaust heat recovery. This poses a significant barrier for energy saving from the system.
Temperature
Heat recirculating combustion Heat to be recycled
Heat of combustion
Preheating
Heat of combustion Combustion without heat recirculation Process progression
Figure 21.7 Temperature history of heat recirculating combustion of premixed reactants in one-dimensional adiabatic system.
© 2011 by Taylor and Francis Group, LLC
Air
In-furnace
Combustion air Preheated air
Burner
Fuel
Figure 21.8 Example of radiant tube-recuperator combination.
A cross-sectional view of the radiant tube and recuperator combination, shown in Figure 21.8, is an example of the application of preheated air combustion system. A recuperator is installed at the exit section of exhaust gases. An air ejector is used to compensate the pressure drop required for passing the exhaust gases through the recuperator prior to the release of exhaust gases to the atmosphere.
21.6 Key Features of HiTAC Flames The unique features of HiTAC flames are now provided. Specifically we provide information on the flame stability, thermal field uniformity, NOx, and noise emission levels. For other benefits, the reader is referred to the only book available on the subject, High Temperature Air Combustion: From Energy Conservation to Pollution Reduction [8]. The flame stability is much affected by the air preheat and oxygen concentration in the air. The results presented in Figure 21.9 show a stable flame combustion condition with low oxygen concentration air (less than 5%) at air temperatures above 800°C. However for air temperatures below 800°C, the flame stability was much affected at less than 15% oxygen concentration in the air. For example, a stable flame at 700°C could only be achieved at much higher oxygen concentrations of 16%. The physical features of the flame (size, shape, color, and global flame structure) were found to significantly
434
Industrial Combustion Testing
change when the concentration of oxygen was changed from normal 21–3% by volume while maintaining the air temperature constant at about 1000°C. The global features of the flame under conditions of 21% oxygen, 50°C air temperature showed the flame to be short and 1600 Re: air = 1.2 × 103 Re: fuel = 4.2 × 103
Air temperature °C
1200
B: High temp. flame
Stable region
C: H.T.A.C.
800 Unstable region 400
0
Stable boundary Unstable point (CO < 100)
A: Conventional combustion 20
15
10 O2 concentration %
5
0
Figure 21.9 Auto-ignition and blow-off limits of propane in a preheated air or preheated air diluted with nitrogen.
40
60
compact having blue color luminosity. In contrast, the flame observed with air having 3% oxygen concentration and 1000°C air temperature had extremely low luminosity and a very large volume. The color of the flame was found to be bluish green to green depending on input conditions of the air. This suggests that combustion under high temperature and low oxygen concentration air provide lower heat release rate per unit volume as compared to flames obtained with the normal 21% oxygen in the combustion air at normal temperature. The low oxygen concentration and high temperature air flames had extremely wide flame stability limits as compared to the normal combustion air case without the use of any flame stabilizer. Low pressure drop in HiTAC flames reveals further enhancement in performance improvements of HiTAC combustion systems. The temperature profiles from HiTAC flames are significantly different than that obtained with normal temperature air. Temperature and flame fluctuations with HiTAC flames are significantly reduced [8–14]. Measured temperatures in two different kinds of flames obtained with normal temperature air at 21% oxygen and preheated air at 1200°C and 4% oxygen concentration (representing HiTAC conditions) are shown in Figure 21.10. For the normal flame, the peak temperature is located near the vicinity of a fuel nozzle exit. In contrast to the
X mm 80 100 120 140
225
40
200 400
Y mm
Y mm
150
125
75 50
1100°C
175
150
100
X mm 80 100 120 140
225
200°C
200 175
60
600
125 100
800
1200
1300
75
1000
50
1100
25
25
Combustion air (35°C, 21% vol O2)
Fuel LPG
Figure 21.10 Temperature distribution under normal (left) and HiTAC (right) conditions.
© 2011 by Taylor and Francis Group, LLC
Combustion air (1200°C, 4% vol O2)
Fuel LPG
435
Regenerative Combustion Using High Temperature Air Combustion Technology (HiTAC)
Normal room temperature air
1500
1000 Ta = 20°C, O2 = 21%, N2 (x,y) = (100,47) Tmean = 1318°C, Trms = 197°C
500
0
0
50
100 Time, ms
High temperature/low oxygen conc. air
(b) 2000
Temperature, °C
Temperature, °C
(a) 2000
150
200
1500
1000 Ta = 1200°C, O2 = 4%, N2 (x,y) = (80,107) Tmean = 1288°C, Trms = 3.7°C
500
0
0
50
100 Time, ms
150
200
Figure 21.11 Temperature fluctuation in normal air temperature (left) and high temperature and low oxygen concentration air (right).
Noise from normal flame
Case 2
Case 3
Case 4
Noise from CDC flame
Relative sound pressure levels, SPL (dB)
8
the measured flame fluctuations measured with a high speed camera showed low flame fluctuations of HiTAC flames, see Figure 21.13.
7 6 5
21.7 Type of Regenerative Burners: Specific Features, Structure, and Application
4 3
21.7.1 History of How Regenerative Burners and Their Dissemination in the Industries
2 1 0
a
b
c Experimental case
d
e
Figure 21.12 Reduced noise from distributed colorless (HiTAC) combustion.
HiTAC flames having low oxygen concentration and air preheated to very high temperatures, a far more uniform thermal field in the entire combustion chamber can be observed, see Figure 21.10. Temperature fluctuations in normal and HiTAC flames are shown in Figure 21.11. The results show a signifcinat reduction in in the rms value of temperature fluctuations (less than 5°C) with the flame having 4% oxygen concentration in air at 1200°C. The characteristic features of the flame thermal signatures at other locations of the flame showed similar features to those shown in Figure 21.11. These results reveal low combustion noise from HiTAC flames [14]. Indeed the measured sound pressure levels from HiTAC flames are much lower than those obtained from the normal temperature flames, see Figure 21.12. Low temperature fluctuations also reveal flow flame fluctuations. Indeed,
© 2011 by Taylor and Francis Group, LLC
Regenerative combustion has long been used in glass melting furnaces that are operated under high temperature conditions as shown in Figure 21.14. In these furnaces, the combustion gases pass through regenerative sections that possess refractory blocks to store the heat energy in the refractory blocks. The combustion air is introduced into this section and the stored heat energy is transferred into the air, which is heated up to very high temperatures of 1200°C–1300°C. The principle of gaining high temperature preheated air in regenerative burners is similar to that of glass melting furnaces. One switching cycle takes about 20 to 30 minutes to allow the heat to transfer from combustion gases to the regenerator, which is then utilized to preheat the fresh air. Dimensions of the regenerator will accordingly increase because ample heat storage capacity is required for sensible heat exchange to occur from combustion gases during one switching cycle. In contrast the dimensions of the regenerator will decrease if the switching cycle is shortened to say, one to two minutes—from heat storage to heat release—to result in a smaller size compact regenerator. Compact size regenerative combustion burners can be fabricated if such small regenerators are directly
436
Industrial Combustion Testing
Fluctuation (%)
20 cm
10 8 25 cm
6 4
Fuel
2 Air
0
Room temp. 21% O2
1200K, 21% O2
1200K, 4% O2
Figure 21.13 Flame fluctuations in normal room temperature air (left) flame, high temperature (middle), and high temperature and low oxygen concentration flame (right).
Direction of gas flow
Combustion chamber
Waste heat recovered from the gases
Glass furnace
Packed beds
Regenerators Combustion air preheated by the hot regenerator
Direction of gas flow
Combustion air preheated by the hot regenerator
Glass furnace Regenerators Waste heat recovered from the gases Figure 21.14 Glass melting furnace operated with regenerative combustion.
connected or assembled in the burner assembly. Weinberg [6] in 1986 proposed a packed bed-type regenerative burner as an advanced combustion method, see Figure 21.15. This new method is now known worldwide. The use of this type of regenerative burner significantly improves the heat exchange efficiency as
© 2011 by Taylor and Francis Group, LLC
Alternating flow (reactants/products) Figure 21.15 Packed bed combustor. (From Weinburg, F. J., In Advanced Combustion Methods, 200–206, London and New York: Academic Press, 1986.)
compared to the conventional type recuperative heat exchange burner that had been used to recover the waste heat in much of the prior works to improve the system efficiency. Hotwork technology in the United Kingdom [7] has developed a regenerative burner system that has regenerators in the burner assembly as shown in Figure 21.16. These types of regenerative burners were not accepted widely in the market because the air is preheated in the regenerators, which significantly increases the NOx emission levels from the burner. Environmental regulations did not permit such types of regenerative combustion burners to prevail in the market as the NOx emissions from any burners
Regenerative Combustion Using High Temperature Air Combustion Technology (HiTAC)
Figure 21.16 Regenerative burner developed by Hotwork Technology, UK. (Hotwork Combustion Technology Ltd. Regeneretive Burners. Yorkshire: Academic press, 1986.)
must be low and future environmental regulations are only expected to be even lower. Therefore the basic philosophy of the regenerative buners must be such that even with the preheated air, the NOx emissions levels are lower than the best known practices and available with low NOx burners. Nippon Furnace Kogyo (now Nippon Furnace Com pany) and several academic partners (University of Maryland, Osaka University, and other institutions) jointly provided detailed fundamental investigations on heat, flow, and combustion characteristics of new types of regenerative flames. The NFK and NKK Corporations (now JFE Steel Corporation) jointly fostered practical feasibilities incorporating the fundamental elements to develop and demonstrate commercial scale, advanced regenerative combustion burners in 1987. The advanced regenerative combustion system adopted higher speed, air velocity combustion that allowed vigorous recirculation of combustion gasses to provide diluted oxygen for the combustion air. In this manner the NOx emissions decreased significantly as compared to any other method of combustion, including that used for recuperative combustion. In 1993, the Japanese government chose the advanced regenerative combustion as one of the energy saving and low emission technologies and subsidized its further development for commercial application in the industrial furnaces. This government project involved specialists from furnace makers, end users, and major universities
© 2011 by Taylor and Francis Group, LLC
437
in Japan as well as the University of Maryland. The regenerative combustion was thoroughly examined and tested until 1997 by each sector and clarified not only in academic field but also in industrial field. Combustion process of the project was called High Temperature Air Combustion (HiTAC) [8–13]. In the HiTAC project, three major effects were clarified that included energy saving by means of waste heat recovery with the regenerators, ultra low pollution, and uniform temperature combustion (called isothermal combustion reactor), which enabled enhanced performance of the furnace with a smaller size or increased output of the furnace with similar volume. There are two types of regenerators that are often used in furnace applications. One type adopts ceramic honeycombs that have higher efficiency and the other type adopts ceramic balls that incorporate a larger mass of the ceramic material. Presently there are different types of regenerative burners that are in operation around the world for use in a number of reheating furnaces and heat treatment furnaces. The NFK technology is widely accepted and used worldwide with major dissemination in the Pacific Rim countries, Asia, Europe, and North America. Apart from the above Japanese government project, a German manufacturer, WS GmbH developed low NOx regenerative burner technology at almost the same time as Nippon Furnace. The manufacturer in Germany called their regenerative combustion technology as FLOX (Flameless Oxidation). It is reported that a number of FLOX burners are in operation in Europe and North America.
21.8 Types of Regenerative Burners Several different types of renegerative burners exit in the market today. Each type has its own specific features and application. The key features and classification of these burners are described in the following. 21.8.1 Alternative Combustion-Type Regenerative Burner The alternative combustion-type regenerative burners are made of one pair of burners. These burners include Nippon Furnace Co. NFK-HRS-DL-type burner, ChugaiRo Company Diamond-RCB-type burner, Rozai Co RSH-type burner, and Tokyo Gas Company FDItype burners. All of these burners are made by Japanese companies. There are also regenerative burners made by North American Inc., and Bloom Engineering Corp. in the United States. A schematic diagram of typical alternative combustion-type regenerative burners made by Nippon Furnace is shown in Figure 21.17. Certain
438
Industrial Combustion Testing
application of the regenerative burners for commercial furnaces is introduced in Figure 21.18. A major difference in the structure of a regenerative burner is that Nippon Furnace adopts ceramic honeycombs that have a significantly larger surface area for heat exchange, while the other manufacturers adopt alumina balls that offers a significantly low heat exchange area as compared to the honeycomb type so that the efficiency of heat generation with the honeycomb type is significantly high. In contrast, the ball types are robust albeit at a somewhat higher pressure drop across them. Each manufacturer has it own specific design for the nozzle arrangement. Ceramic honeycombs have a larger surface area than ceramic balls and therefore smaller size regenerators can be designed with less pressure drop across the regenerators. Furthermore the honeycombtype regenerator offers shorter switching time to result in higher efficiency.
Burner tile Combustion gas
Fuel Air
Secondary fuel
Four-way change-over valve Combustion gas
Primary fuel Honeycomb regenerative media
Figure 21.17 Schematic diagram for alternative combustion-type of regenerative burner.
Figure 21.19 shows temperature efficiency versus the nondimensional length at different times (nondimensional) for a characteristic of the honeycomb regenerative burner. The temperature efficiency is defined in the figure with the scale showing the temperature efficiency. The time parameter represents a function of the switching cycle. In other words, with a smaller area of heat transfer, the same heat transfer efficiency is maintained with a shorter switching cycle. Assuming that no temperature distribution exists inside the regenerator material, the change of temperature with time of the regenerator is given by:
(21.7)
where θm = temperature of regenerator media, θf = temperature of the fluid, t0 = a time constant ( = cρV/hA). As stated previously, the switching time of air and combustion gases should be shorter to improve efficiency. The change in heat transfer rate is calculated as a function of switching period, τ, using the energy balance equation. The results are shown in Figure 21.20 and point to the importance of a short switching cycle to achieve high heat transfer rates. In the case of a ball-type regenerator, dirt and dust tend to accumulate in the balls but their cleaning method is simple and straightforward as the ceramic balls can be easily removed, washed, or cleaned. With these features, ball-type regenerators can well be employed for dirty gases that contain scale and dust. For normal configuration, a burner nozzle is installed in the longitudinal center axis of the burner and a baffle-type burner is used.
NFK-HRS DL type (Nippon furnace co., ltd.) Figure 21.18 Example of commercial regenerative burners produced by Japanese burner makers.
© 2011 by Taylor and Francis Group, LLC
d θm/dt = 1/τ0(θf − θm),
NFK burner
439
Regenerative Combustion Using High Temperature Air Combustion Technology (HiTAC)
10 20 = Π
=
30
2400
0.4
40 = Π 50 = Π
Λ = Nondimensional length Π = Nondimensional time
0.2 0
0
10
20 30 Nondimensional length, Λ
hm A h A = m (W C)c (W°C)h h AP Π= m M CM Λ=
50
40
A : heat surface area (m2) hm : heat transfer coefficient (kW/m2K) W : flow rate (kg/hr) C : specific heat of fluid (kW/kg K) CM : specific heat of regenerator (kW/kg K) P : switching period (= 1/2 cycle) (hr) suffix h : hot side c : cold side
Figure 21.19 Performance of regenerative heat exchanger (Chugai Ro Co., Ltd.). Q: Thermal energy transferred during 1 cycle of heat storage/release ∆τ = 0.0 τ0 ∆τ = 0.2 τ0 τ0 = 12 sec
Heat transfer rate, Q/2τ (kcal/s)
1000 800 600 400
∆τ = 0.1 τ0
0
30
60 90 Switching period, τ(s)
120
150
Figure 21.20 The effect of switching period on heat transfer rate for regenerative media.
Some manufacturers employ other types of burners on the basis of technical license and specific needs from other manufacturers. The advantage of twin nozzle-type regenerative burners includes: increased combustion capacity of one pair of burners (as much as 8 MW maximum for commercial application). Honeycomb-type regenerative burners of 8 MW capacity are installed at Corus Steel in the United Kingdom as shown in Figure 21.21. Owing to alternative combustion (from the pair of burners), the combustion gases are well circulated in the furnace to provide extensive energy saving in the steel reheating furnaces.
© 2011 by Taylor and Francis Group, LLC
Box shape, walking-beam type
Reheating capacity 230t/h, number of combustion control zones : 10 Slab size 250t × 1950W × 9600L, maximum weight 30t
Burner
200 0
Furnace type
Fuel
∆τ = 0.0 τ0 τ0 = 150 sec
Figure 21.21 Installed HRS DL9 burner on steel reheating furnace in Corus Llanwern Works, UK (manufacturing and installation by ICS Industrial Combustion Systems Sp.zo.o, European Licensee of NFK).
Temperature deviation in a rough bar
1200
Specification of HRS-DL9 Firing rate: 5.8 MW (1 pared burner) Fuel: Natural gas Weight: 3.6 ton/1 burner
2000
0.6
=
(Manufacturing & installation by ICS industrial combustion systems Sp.zo.o (European licensee of NFK))
900
=
Π
0
Π
0.8
Π
Temperature efficiency, E
1.0
By-product-gas of fukuyama works (low heating valve : 11.7 MJ/M3N)
Regenerative burner : 76 units, Regenerator : honeycomb
JFE steel Fukuyama works No. 3 reheating furnace NKK technical review No. 80 (1999) 100 74
No. 3 Furnace (HiTAC)
No. 4,5 Furnace (Conventional)
Figure 21.22 Improvement of skidmark.
For application in steel reheating furnaces, uniform temperature distribution in the entire furnace section is achieved due to dispersed flame in addition to the advantages of energy saving and low NOx. Skid marks of the slab can be diminished with the dispersed flame in a steel reheating furnace. Example of skid mark reduction is shown in Figure 21.22. 21.8.2 Radiant-Type Regenerative Burner The purpose of the radiant-type regenerative burners is to avoid direct contact between hot combustion gases and the material being heated. The regenerative
440
Industrial Combustion Testing
burners for the radiant tube have typical configuration of U type or W type. The heat resistant steel tube is heated by a regenerative burner to generate radiant heat through this tube. Therefore it is an indirect method for heating the materials so as to avoid oxidation. Thus the burner is adapted for the heat treatment furnaces where controlled atmosphere in the furnace is required. In the case of conventional radiant tube burners, the combustion is initiated from one end and combustion gases are fed into the heat exchanger equipped at the other end. However, for regenerative radiant tube, one pair of regenerative burners is installed at both ends of the tube as shown in Figure 21.23. Those burners are
in service alternatively and the tube surface is heated uniformly along the tube length. Extended tube life is obtained due to flat heat flux distribution over the entire tube length as shown in Figure 21.24. Regenerative radiant tube burners are made by Nippon Furnace, Tokyo Gas, and Osaka Gas in Japan and North America. The advantage of regenerative radiant tube burners include: waste heat recovery with higher energy saving rate than that obtained from conventional radiant tube burners. In addition uniform heat flux distribution is available over the entire length to result in longer life. In order to achieve ultra low NOx emissions, a high level of technology is required as the
Combustion air Ignition plug Flame detector Pi
CEM
lot
Flue gas
air
Fuel Burner gun Ceramic honeycomb (regenerator)
Figure 21.23 A schematic diagram of radiant tube-type regenerative burner. 5 inch W-type tube, Fuel: LNG, firing rate: 54 kW, furnace temp.: 850°C, air ratio = 1.2
1000
Conventional recuperative burner
Tube skin temp (°C)
950 900 850
HRS-RT burner
800 REQ-RT HRS-RT
750 700
0
Figure 21.24 Tube skin temperature for radiant tube.
© 2011 by Taylor and Francis Group, LLC
1000
2000
3000 4000 5000 Tube length (mm)
6000
7000
8000
441
Regenerative Combustion Using High Temperature Air Combustion Technology (HiTAC)
combustion gases are recirculated inside the tube. In this regard each burner maker has its own technology for achieving low NOx emission. In the steel industry, comparatively large diameter (six to eight inches) tubes are employed for continuous annealing line (CAL) and continuous galvanizing line (CGL). Each burner maker has its own specific technology for achieving low NOx. For applications other than for the steel industry, electric heaters are mostly employed. These heaters are replaced with regenerative radiant tube burners for energy saving. Regenerative honeycomb-type radiant burners offer more advantages in such replacement than the ball type because of the availability of smaller size burners. 21.8.3 Single-Type Regenerative Burner The single-type regenerative burner is used for both intake of combustion air and for discharge of combustion gases. The fuel is supplied continuously from one nozzle to allow combustion without interruption, see Figure 21.25. This type of regenerative burner does not need to have a pair of burners for alternate combustion that will allow the fabrication of a compact smaller size burner system. Continuous combustion takes advantage of the regenerative burner for
τ sec Combustion air port
τ sec Combustion air port
application to endothermic chemical reaction. A regenerator must handle not only inducing combustion air but also discharging combustion gases and therefore the combustion capacity of single-type regenerative burners is relatively limited as compared to that of alternative-type regenerative burners. In designing this type of single regenerative burners, attention must be paid to prevent the unburned fuel being induced into the outgoing exhaust gas passage. There are several manufacturers of single-type regenerative burners, such as Nippon Furnace, ChugaiRo, Rozai, Tokyo Gas, Osaka Gas, and Yokoi Machinery in Japan and WS in Germany. An example of the single-type regenerative burner of Nippon Furnace (brand name Ux burner) is shown in Figure 21.26. Regemat made by WS is shown in Figure 21.27. These single-type regenerative burners do not have a large capacity and they are typically rated for around 300 kW. The dimensions of such burners can be minimized using a honeycomb for regenerators. The capacity of honeycomb-type regenerative burners can be increased to as much as 2.4 MW and such large capacity single-type regenerative burners are applied particularly for endothermic reactors, such as hydrogen reformers as shown in Figure 21.28. These regenerative burners are well suited and widely utilized in chemical industries.
τ sec
τ sec
Combustion air port
Combustion air port
τ = 3 ~ 8 sec
Fuel Preheated air
Flue gas Switching unit
Flue gas Figure 21.25 Continuous firing with single-type regenerative burner system.
© 2011 by Taylor and Francis Group, LLC
Air
442
Industrial Combustion Testing
994 Switching unit
420.5 Ceramic honeycomb (φ100 × 300L × #100/17)
Pilot air inlet
420 500 Precombustion chamber
Off gas inlet
Exhaust
Fuel gas inlet P’ry air inlet
Main air inlet
Figure 21.26 NFK-UR burner (100d4 type, 170 kW).
21.9.1 Visual Observation, Photo, and Video Recording of the Flame
Flue gases
Gas
Flame
Air
FLOX
Figure 21.27 Single-type regenerative burner Regemat (Weunning, J. G., WS Gmbh. With permission.)
made
by
WS.
21.9 Key Points for Test Furnace Design (by Nippon Furnace, Japan) The purpose of a test should be clarified prior to designing the test furnace. When the test furnace is built for multipurposes, other tests should be well studied and not just the originally intended test purpose. The structure of the test furnace will change according to the requirements and purposes for which the furnace is designed.
© 2011 by Taylor and Francis Group, LLC
For visual observation of the flame conditions, the observation window should be provided with an optical view as large as possible. However, if the window is too large the heat loss will be large, which will then affect the furnace temperature. Attention should be paid to minimize the losses when designing a test furnace. A quartz window should be used to observe, photo, or video record the flame inside the furnace when the furnace temperature is over 1200°C for sustained time duration. When a UV laser is used, a special quality of glass with the desired light passing through should then be adopted, allowing the desired wavelength of UV ray to pass through with negligible light signal attenuation. A UV laser used to measure intermediate combustion generated species from within the flame in a test furnace, using laser induced fluorescence (LIF) diagnostics is shown in Figure 21.29. Optical measurement can be achieved with a shutter that will open or close in order to prevent heat loss during preheating mode of the furnace or during a time when no test is conducted. For the most practical furnace applications, 10 mm thick window glass (200 mm wide and 200 mm long) has provided successful results. 21.9.2 Flame Temperature Measurement It is difficult to measure flame temperature correctly because the temperature values measured with a
Regenerative Combustion Using High Temperature Air Combustion Technology (HiTAC)
443
Installed burner on the test plant in Chiyoda co. Installed burner on the test furnace in NFK Figure 21.28 Large-scale single regenerative burner (2.4 MW).
CCD cameras
Test furnace
Windows for measuring the in-furnace
Nd-YAG laser Burners Dye :
laser
Excimer CCD cameras laser
. . Controller
Pulse generator
Figure 21.29 Photo of test furnace with large size observation/measurement windows.
thermocouple must be corrected for losses. The temperature sensors may have influence from thermal conduction, radiation, and thermal inertia. When the temperature measurement is required for practical application, a thermocouple measuring method is acceptable. For example, an R type thermocouple made of Pt-Pt/13%Rh, even with 50 micrometer diameter wire, an anticatalytic quartz coat should be used for Pt wire. Thermocouple signals for thermal inertia should be corrected with time constant set at 20 ms. If the wire diameter is different then a different time constant must be used. The high frequency noise should be cut off using a low-pass filter, for example cut-off frequency of 5 Hz. An example of the measured temperature profiles in conventional and HiTAC flames is shown in Figure 21.11.
© 2011 by Taylor and Francis Group, LLC
Significant differences in the temporal characteristics of thermal signatures can be observed between the conventional flame and the HiTAC flame. When dealing with high gas velocity or large size test furnace, a thin wire sensor without any protection sheath is not suitable for local measurements. 21.9.3 Measurement of Heat Transfer and Heat Flux of Heated Material Many heat flux meters are available in the market and an appropriate type of instrument can be chosen for such measurement. The decision on sensor installation is very important because it should be easily replaceable when required. An approximation of the heat flux can be
444
Industrial Combustion Testing
easily imagined for the twin-type regenerative burners as the combustion takes place alternatively from the two sections of the burner. In contrast of the singletype regenerative burners, the position of the flame does not change but the combustion air intake flow changes around the nozzle. When the furnace temperature change exceeds a certain range it influences the heating piece. Fast response thermocouples should be installed to measure such quick changes in the thermal field. The diameter of a protective tube should be such that it does not cause much turbulence to the flow. The front end of thermocouples should not be covered so that it can respond quickly to the thermal changes. The operation time of combustion air charging and exhaust gas discharging should be recorded. The temperature change should be measured with a data logger during the operational cycle of the switching valve. Examples of measured temperatures in the test furnace are shown in Figures 21.30 and 21.31. When temperature measurement is required in the center of the furnace (size in excess of two meters in width), the thermocouple should not be supported to hang down in the furnace. The supporting seat should be provided on the furnace wall for installing the protective tube. A water-cooled jacket should be prepared when measuring high temperature distribution in the furnace using R type thermocouples. Thermocouples are often inserted into ceramic tubes. They do not store any heat and therefore a speedy traverse can be achieved. The front end of the thermocouple should be well stretched out from the front end of the watercooled probe in order to avoid the influence of the probe. For example, in the case of a two meter long test furnace, the thermocouple should be stretched out by more than 50 mm from the water-cooled probe as shown in Figure 21.32.
For measurement of NOx emissions, a chemiluminescence analyzer can be used for industrial application. The furnace should be provided with some heat sink mechanism since the furnace temperature must be controlled under constant firing condition. As for heat sink, air cooling is the simplest and most inexpensive way; however, the heat transfer coefficient cannot be large. One of the simple ways is to mix a water mist in the cooling air to increase the heat sink capability. When the large value of heat sink is required for tests, water cooling is the most effective method. For water cooling, the costs are expensive in particular when circulation or a chiller water system is installed. 21.9.5 Measurement of Temperature Distribution in the Furnace In the case of regenerative burners, the position of flame changes with time to cause local changes in the ambient furnace temperature. Change of the flame position is
1300
Exhaust gas
1100 1000
Ordinary combustion
Temperature (°C)
1200
Fuel Air
1300 1200 1100 1000
900 3.80
Fuel Air Fuel
1.90 0.95
Parallel flow type 0 Fuel
Air
Fuel Counter flow type
Exhaust gas
500 1000 0.00 W (mm)
Ordinary combustion Fuel : Natural gas Heat input : 349 kW Air temp. : 950 ~ 1063°C Air ratio : 1.2
1100 1000 900 3.80
2.85 1.90
L/W
0.95 0
500 1000 0.00 W (mm)
Parallel flow type Fuel : Natural gas Heat input : 349 kW Air temp. : 984 ~ 1077°C Air ratio : 1.2
Figure 21.30 Temperature distribution within the test furnace installed with a twin-type regenerative burner.
© 2011 by Taylor and Francis Group, LLC
1200
900 3.80
2.85 Exhaust gas
1300
2.85 1.90
L/W
0.95 0
500 1000 0.00 W (mm)
Counter flow type Fuel : Natural gas Heat input : 349 kW Air temp. : 1012 ~ 1072°C Air ratio : 1.2
L/W
Temperature (°C)
21.9.4 NOx Measurement
Temperature (°C)
made by means of changing the depth of thermocouples. When installing the thermocouple, a sheath diameter of 1 to 3 mm (INCONEL sheath K type thermocouple) should be used. Thermocouples can get easily burnt during measurement. Therefore ample quantities of thermocouples should be prepared in advance prior to testing. The test furnace should be designed in such a manner as to allow easy charge or discharge of the heating piece samples. The sheath wires of thermocouples should be well sealed where it is placed in the furnace. Error protection from thermal conduction of sheath wires should be provided with an insulation cover or built in the junction between refractory and the furnace wall.
445
Regenerative Combustion Using High Temperature Air Combustion Technology (HiTAC)
A
Top section
0
Temperature (°C) C
200
G
C
400
F
1020
A D
E
Y(mm)
B
1060
600 1100
1080
1120
800 1160
1180 1160
1180
1140
1000
1140
1120
Burner 1200
Reaction tube
D
1040
1160 1180
800
600
1140
1180
1120
400
1160
200
0
–200
X(mm) Figure 21.31 (See color insert following page 424.) Temperature distribution within the test furnace installed with a single-type regenerative burner.
Cooling water
φ18
Thermocouple
50
1500
Data acquisition
PC
Figure 21.32 Thermocouple for temperature distribution measurement.
21.9.6 Measurement of Gas Component in the Furnace The gas composition distribution in a test furnace equipped with regenerative burners is also influenced in the same way as the furnace temperature. A gas analyzer should be chosen to provide quick response. For example the gas analyzer should have a response time of less than three seconds for providing full-scale response. When the gases are measured at the sampling point, the timing of correct concentration indicated by the analyzer is determined from the total sum of the gas analyzer response plus the length of sampling line
© 2011 by Taylor and Francis Group, LLC
and the volume of drain removal pot. It is important to measure such factors in advance. The gas sampling shown in Figure 21.33 was used. An example of the CO and O2 concentration in a test furnace (shown in Figure 21.31) is shown in Figure 21.34. 21.9.7 Modeling of Furnace Used for Commercial Application For commercial applications, the location of burners, furnace wall, heating pieces that absorb heat, and combustion load (combustion rate for furnace dimensions)
446
Industrial Combustion Testing
Cooling water
φ 18
2000
O2/NOx/CO analyzer Suction pump
Drain separater
Figure 21.33 Gas sampling probe for local gas concentration measurement.
C
F
200
A
CO concentration (ppm) C
200
A
Bottom section O2 concentration (%) C
E 400 Y(mm)
D
0
400
0 200 500
600
D
1000 800
Burner Reaction tube
1.5
Y(mm)
G B
A
Bottom section
0
600
2.5
800 1500
3000 3000 3000 2500 3000
1000
2000 2500
4.5 4.0 3.5
1000 1200
1200 800
600
400 200 X(mm)
0
–200
D
2.0
4.0 800
4.5
600
4.5 4.0 3.5
3.0 3.5
3.5 4.5
200 400 X(mm)
4.0 0
–200
Figure 21.34 (See color insert following page 424.) Example of measured CO and O2 concentration in the test furnace.
should be determined according to each actual case with as much precision as possible. If the commercial furnace size is quite large, sectional modeling should be used as part of the test furnace but it should be considered the same as a commercial furnace. If the heating material is taken up, it should be clarified whether it is a solid material that can be heated normally like steel, or a heat sink material that absorbs heat with endothermic reaction. Further flame conditions and heat flux distribution should be well studied. It is important to create the same conditions even in the test furnace. An example of sectional model furnace for a steam reformer using high temperature, air combustion technology is shown in Figure 21.35 [15]. Such flames are rather complicated and as such rather difficult to simulate from the regenerative burners with a mathematical model. Therefore, under such situations, it is very important to obtain
© 2011 by Taylor and Francis Group, LLC
actual operation data from tests on the change in temperature and flame characteristics.
21.10 Summary In industrial heating systems, preheated air combustion is exercised in such a manner that the exhaust heat associated with the flue gases heats the material that is then used to preheat the combustion air. The basic principles of HiTAC require high air preheats containing sufficient amounts of diluted air. This air is injected at high velocities into the combustion chamber so as to provide a large entrainment of the combustion gases before ignition of the fuel. The HiTAC exhaust gases
Regenerative Combustion Using High Temperature Air Combustion Technology (HiTAC)
Test reformer for demonstration plant
447
Commercial HICOT reformer
Figure 21.35 Photograph and schematic diagram of sectional model furnace for a steam reformer using high temperature air combustion technology.
are recirculated in the system to provide auto ignition of the fuel immediately after rapid mixing between diluted high temperature air and fuel. The major difference between HiTAC and excess enthalpy combustion lies in the heating levels and the recirculation in relation to the combustion level. The new preheated air combustion method (called here High Temperature Air Combustion method) involves a large amount of heat recirculation. Other names given to this combustion technology include: colorless distributed combustion flames oxidation, or mild combustion, although mild combustion does not seem appropriate. Practical means to provide high temperature air combustion for commercial furnaces requires the use of nonstationary combustion (e.g., via the use of high cycle switching control). They in turn depend on high temperature dilution of the combustion air and/or fuel. Using any of these methods allows one to achieve excellent improvement on heating performance characteristics. In addition, reduction in exhaust heat loss and oxides of nitrogen also occurs. The thermal field uniformity within the heating chamber can be made higher and more uniform with the use of HiTAC technology. The HiTAC flames incorporate enthalpy and mass recovery from the waste gases to provide far more uniform thermal field in the entire combustion chamber. The uniformity is unique and has not been demonstrated in any practical system. The combustion reactor with HiTAC can be called isothermal reactor since it incorporates a highly uniform thermal field. The temperature and flame fluctuations of HiTAC flames are extremely low so that the combustion noise is also very low. The HiTAC flame offers significant fuel flexibility as well as the alleviation of combustion instability that is often present under certain conditions in most combustion systems.
© 2011 by Taylor and Francis Group, LLC
The new preheated air combustion technology with higher heat recirculation rate has the potential for much wider applications than that explored to date for only furnaces and boilers. Some of the potential applications include almost all systems using solid and liquid fuels and wastes. The technology also has the potential for application to stationary gas turbine and internal combustion engines, each application resulting in enhanced benefits. The results obtained to date have shown that considerable performance improvement can be made with HiTAC technology in furnaces and boilers as compared to the conventional heating equipment.
References
1. Nippon Furnace Kogyo, NFK manual, p. 327. Japan: 1981. 2. BY-CAST N.V. Company, DEKA Heater Catalog. http:// www.by-cast.com/site.htm. 3. Ohana, H. Design Handbook for Heat Exchanger, p. 1004. Kogaku Tosho Edition, 1-14-15 kudan-kita, Chiyoda-ku, Tokyo, Japan, 1990. 4. Hosoi, K. “Development of Recuperative Burner.” Periodical of Japan Burner Research Association, no. 53 (1986): 16. 5. The Japan Industrial Furnace Manufacturer Association. Promotion of Energy-Saving Industrial Furnace, 3–16, 1978. 6. Weinburg, F. J. “Combustion in Heat-Recirculating Burners.” In Advanced Combustion Methods, 200–206. London and New York: Academic Press, 1986. 7. Hotwork Combustion Technology Ltd. Regenerative Burners. Yorkshire, UK: Sheffield. 8. Tsuji, H., Gupta, A. K., Hasegawa, T., Katsuki, M., Kishimoto, K., and Morita, M. High Temperature Air Combustion: From Energy Conservation to Pollution Reduction. Boca Raton, FL: CRC Press, 2003.
448
9. Shimada, T., Akiyama, T, Fukushima, S., Mitsui, K., Jinno, M., Kitagawa, K., Arai, N., and Gupta, A. K. “Time Resolved Temperature Profiling of Flames with Highly Preheated/Low Oxygen Concentration Air in an Industrial Size Furnace.” ASME Journal of Engineering for Gas Turbine and Power 127, no. 3 (July/August 2005): 464–71. 10. Gupta, A. K. “Thermal Characteristics of Gaseous Fuel Flames using High Temperature Air.” ASME Journal of Engineering for Gas Turbine and Power 126, no. 1 (January/ February 2004): 9–19. 11. Gupta, A. K., Bolz, S., and Hasegawa, T. “Effect of Air Preheat and Oxygen Concentration on Flame Structure and Emission.” Proceedings of ASME Journal of Energy Resources and Technology 121, (September 1999): 209–16. 12. Ishiguro, T., Tsuge, S., Furuhata, T., Kitagawa, K., Arai, N., Hasegawa, T., Tanaka, R., and Gupta, A. K. “Homogenization and Stabilization during Combustion of Hydrocarbons with Preheated Air.” Proceedings of
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
27th Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 3205–13, 1999. 13. Gupta, A. K., and Li, Z. “Effect of Fuel Property on the Structure of Highly Preheated Air Flames.” Proceedings of International Joint Power Generation Conference (IJPGC-97), Denver, CO, November 2–5, 1997, ASME EC-Vol. 5, 247–58. 14. Mortberg, M., Blasiak, W., and Gupta, A. K. “Combustion of Low Calorific Value Fuels in High Temperature and Oxygen Deficient Environment.” Combustion Science and Technology (CST) 178 (September 2006): 1345–72. 15. Onda, N., Mikuriya, T., Yoshioka, T., Mochida, S., Nakamura, H. and Araake, T., “Demonstration Plant Phase Development on Advanced High-Temperature Air Combustion Technology for Steam Reforming Process”, Proceedings, of American-Japanese Flame Research Committees Joint International Symposium, October 22–24, 2007, Waikoloa, HI.
22 Characterization of Ribbon Burners Colleen Stroud Alexander and Melvyn C. Branch Contents 22.1 Introduction.................................................................................................................................................................. 449 22.1.1 Ribbon Burner Applications........................................................................................................................ 450 22.1.2 Burner Configuration.................................................................................................................................... 450 22.2 Characteristics Defining Burner Stability and Flame Environment Structure.................................................. 451 22.2.1 Burner Stability Parameters......................................................................................................................... 451 22.2.2 Variables Influencing Flame Environment Structure.............................................................................. 452 22.2.2.1 Effect of Burner Configuration on Flame Environment......................................................... 453 22.2.2.2 Parameterized Flow Regime Analysis...................................................................................... 453 22.2.2.3 Heat and Mass Transfer Nonuniformities................................................................................ 456 22.3 Flame Treatment of Polymer Films........................................................................................................................... 458 22.3.1 Overview and Discussion of Wettability................................................................................................... 458 22.3.2 Flame Treatment Parameters....................................................................................................................... 459 22.3.2.1 Effect of Fuel/Air Equivalence Ratio......................................................................................... 459 22.3.2.2 Effect of Film Positioning within the Flame Treatment Environment................................. 460 22.3.2.3 Effect of Temperature and Flow Environments....................................................................... 460 22.4 Roles of Flame and Surface Chemistry in Flame Treating.................................................................................... 463 22.4.1 Flame Chemistry........................................................................................................................................... 463 22.4.2 Surface Chemistry......................................................................................................................................... 465 22.4.3 Effect of Nitrous Oxide Additive................................................................................................................ 466 22.5 Summary and Conclusions........................................................................................................................................ 468 References................................................................................................................................................................................. 468
22.1 Introduction This chapter focuses on recent advances in the understanding of industrial ribbon burners and, by example, the use of these burners in a particularly demanding application [1,2]. Ribbon burners are defined as “a burner having many small closely spaced ports usually made by pressing corrugated metal ribbons into a slot” [3]. There is increasing need for the design of such burners with greater efficiency and stability that enable flames with variable fuel composition and reduced air pollutant emissions. The optimization of the burners for a particular application requires a detailed understanding of the flame structure, chemical composition, chemical kinetics, and flame–surface chemical interactions. The use of ribbon burners, described below, for the surface treatment of polymer films is an excellent example of a very demanding industrial process.
Gas burners have been described as nonpremixed (air mixes with the fuel after the fuel exits the burner port), partially premixed (partial mixing of fuel and air prior to exiting the burner port), or fully premixed (fuel and air are fully mixed prior to exiting the burner port) depending on the degree of premixing of fuel and air prior to combustion [4]. Of these, the most stable and easily controlled is the fully premixed burner with independent control of fuel and oxidizer flow rates. More specifically, the ribbon burner design, which features a series of premixed conical flamelets, has been used in a variety of applications. The ribbon burner offers the advantages of compact design and ease of manufacture making it attractive for such industrial applications as furnaces, baking ovens, plastic and metal forming, and flame treatment of polymer films. Recent advances in the optimization of flame treatment, employing the use of ribbon burners, have resulted from the application of modern experimental and analytical techniques, combustion modeling, and 449
© 2011 by Taylor and Francis Group, LLC
450
an understanding of flame-surface reaction chemistry to this unique process. This progress is also reviewed beginning with a detailed description of the flame structure created by a methane/air mixture and the fluid and thermal environment created by the ribbon burner, and then proceeding to a discussion of the application of burners in the flame treatment process. 22.1.1 Ribbon Burner Applications As described in the previous section, ribbon burners have been employed in a variety of industries due to the positive design characteristics of these types of burners. In the baking and food processing industries, ribbon burners are used extensively. Many types of cereal grain foods and meat food products are baked, cooked, warmed, browned, or toasted in ovens or over grills heated with ribbon burners. In the paper industry, ribbon-burner-supported flames are impinged on paper roll stock during the manufacturing process to remove (via singeing) small fibers protruding from the paper surface. Many molded three-dimensional plastic parts are surface treated with ribbon burners for improved wetting or bonding of inks, paints, coatings, and labels. Ribbon burners are used in this application because they are easily adapted to any firing orientation and can be small enough to mount on robotics or other automated process equipment. For two-dimensional or sheet-stock materials, ribbon burners as wide as eight meters are used in a variety of applications. For metal sheeting, ribbon burners are used to clean or degrease aluminum sheeting prior to coating or other processing. And, as will be discussed in detail in this chapter, ribbon burners are the optimal solution for the surface modification, or flame treatment, of polymer films. Of the applications mentioned above, flame treatment is the most demanding due to the requirement for exceptional uniformity of the flame environment over large physical distances coupled with the need for exceptional control of the flame chemistry and heat-transfer characteristics. These demands require a precise understanding of the combustion environment created by the burner. The goal of flame treatment is the generation of a uniformly modified polymer surface in a high speed industrial process. Flame treating is accomplished by radicals in the post-flame gases that oxidize the surface of the polymer film and increase the wettability of the surface without causing thermal damage to the plastic film. The surface oxidation and consequent increased wettability lead to stronger attachment of adhesives, inks, and other coatings to the surface. The process allows for improved adhesive tape and labeled packaging products, as examples.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
22.1.2 Burner Configuration Many parameters must be considered in the design and manufacture of an effective and efficient burner. A well-designed burner will have a high level of stability, reducing the occurrence of flashback and flame liftoff, which are two types of instabilities that occur due to an imbalance between the fuel/oxidizer mixture flow velocity and the burn velocity of the combustible mixture. The parameters that influence burner stability include port size, shape, number, and proximity within an array. In studying the various burner types, including punched-metal, slot, and ribbon burners, it has been found that the ribbon burner is the most stable under a wide range of conditions while maintaining higher levels of heat release [5]. The formation of smaller momentum-controlled flames allows for more flexibility in the directional mounting of the burner [5]. In addition, the ribbon burner provides a large port area in comparison to other burner types, reducing the physical size of the burner for a given power output [4]. Ribbon burners have a variety of sizes ranging from a single row of ports to multiple-tiered rows, such as the so-called 4-port burner illustrated in Figure 22.1, with additional burner details shown in Figure 22.2. The ribbon is comprised of packed sheets of corrugated stainless steel, which form rows of nearly elliptical outlets, or ports, each having a major diameter of 2.4 mm and a minor diameter of 1.5 mm. The rows are parallel, but offset, so that the centers of the outlets are staggered, as shown in Figure 22.1. In each row, the spacing between the outlets is approximately 2.5 mm center-to-center. With reference to Figure 22.1, ports 1 and 4 are denoted as primer ports, while the central ports 2 and 3 are considered to be the main ports [6]. Alternately, ports 1 and 3 form the upstream pair of burner outlets while ports 2 and 4 form the downstream pair, with the upstream and downstream directions being defined relative to (a)
(b) 1 2 3 4 2.5 mm
Figure 22.1 A photograph coupled with a schematic diagram of the ribbon burner showing port nomenclature. Ports 1 and 4 are primer ports, while ports 2 and 3 are main ports. Ports 1 and 3 form the upstream pair of burner outlets, while ports 2 and 4 form the downstream pair.
451
Characterization of Ribbon Burners
0° or 17° slant
1 cm 31 cm
2 mm
z y, down-web
x, cross-web Burner housing
Figure 22.2 Side view showing burner housing and top view of ribbon placement along with the orientation of experimental axes. Direction x is cross-web, direction y is down-web (-y is up-web).
the motion of the heated, baked, or treated surface (i.e., the surface to be treated is exposed to the flame issuing from the upstream pair of ports first). The surface of the ribbon is generally recessed 3 mm below the surface of the burner housing. The definition of the axes is depicted in Figure 22.2 where the cross-web direction (x) runs along the length of the burner and the downweb direction (y) is defined as crossing the width of the burner. The burner ports can be straight or angled for ease of burner assembly, as shown in Figure 22.2. From a fluid dynamics standpoint, a ribbon burner employed in an industrial process may be treated as an array of multiple jets, often impinging on a nearby surface. Consideration should be given to the multiple jet interaction as well as to the presence and proximity of the impingement surface and any external crossflow generated by the movement of the impingement surface. The multiple jet interactions become complicated with the formation of turbulent eddies that enhance mixing. The overall levels of mixing and entrainment are greater for a multijet array in comparison to that of a single jet, leading to as much as a threefold increase in the total average heat transfer coefficient [7], thus enhancing burner efficiency and performance.
22.2 Characteristics Defining Burner Stability and Flame Environment Structure In this section, the characteristics of the burner design that influence the stability of the flame above a ribbon
© 2011 by Taylor and Francis Group, LLC
burner are described. Then, the structure of the flame cones and subsequent mixing of the combustion products are discussed. In most practical applications, burners must operate safely and stably over a wide range of flow rates, with or without a nearby surface that is being exposed to the high temperature combustion products. Rapid mixing of the combustion products is also desirable for ensuring the spatial uniformity of treatment of a surface exposed to the combustion products and reactive intermediates. Recent studies, summarized below, have shed considerable light on the flow field and mixing characteristics of reactants exiting the burner, reacting in the flame, and leaving the reaction zone, providing insight into the behavior of the fully aerated ribbon burner under a wide range of operating conditions. 22.2.1 Burner Stability Parameters Burner flame stability, which was briefly discussed in Section 22.1.2, will now be addressed in more detail, relating flame instabilities with the specific burner design parameters that help to promote stable burner performance. In general, burner flame stability, which refers to the relatively stationary upright position of the flame, can be achieved by promoting wall attachment [5]. Wall attachment occurs when the flame adheres to a nearby surface because of the presence of a boundary layer along the wall. Near the surface of the wall, the flame is stabilized due to heat losses to the wall. The no-slip condition at the surface must be satisfied and a velocity gradient (from zero to the free stream velocity) forms within the layer. When the stream velocity matches the burning velocity, a stable flame is formed
452
and able to attach. In contrast, flame liftoff occurs when the stream velocity exceeds the burning velocity and combustion is delayed until further downstream. This type of instability is undesirable because it is an indicator of incomplete combustion and can lead to flame extinction. Flashback is another type of instability that can develop in a premixed burner if the burning velocity of the flame exceeds the stream velocity flowing out of the port. The flame front then propagates inside the burner housing and into the reactant delivery system, leading to undesirable and potentially dangerous results. Flashback may occur when burner ports are larger than the reactant gas quenching distance. Under these conditions, the maximum port diameter ensuring necessary thermal losses has been exceeded, thus allowing the flame to propagate back into the mixing chamber. Appropriately sized port diameters enable a flame exhibiting this form of instability to be quenched due to proper heat loss to the port-forming walls, while a balance between the flame-specific burning velocity and the premixed fuel/oxidizer flow velocity is also necessary to ensure proper flame attachment and stability. Jet separation distance, defined as the separation between port openings in the burner and, therefore, also the separation between the flamelets formed, is also crucial in determining flame stability with a ribbon burner. If the jets are too close together, combustion product recirculation near the base of an adjacent flame is restricted. As a result, the local stream velocity from the adjacent jets exceeds the burning velocity and flame liftoff occurs. If the jets are too far apart, the retention of a stable flame is reduced due to excessive heat loss to the burner surface. An optimal flame separation distance for a methane/air mixture was experimentally determined by Pearson, Saunders, and Hargreaves [5] to be 1.25–2.25 mm. Additional studies related to burner configuration and flame stability were performed by Stroud et al. [2] and are described in detail in the next section. The addition of oxygen to the reactant gas has been explored in order to increase the burning velocity of the premixed flames attached to a ribbon burner. The effect of this oxygen enrichment is to increase the rate of energy deposition to an adjacent surface or to a heated environment. Care must be taken in such applications to insure that the ribbon burner ports remain smaller than the quenching distance of the reactant gas mixture as the quenching distance is dependent on the nature of the fuel being used. Since oxygen enrichment inevitably leads to smaller quenching distances, a burner that is safe for operation with air as an oxidizer may not be safe if the oxygen concentration in the combustible mixture is increased [8]. As described in Section 22.1, ribbon burner ports often have an elliptical shape with a major dimension almost
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
twice that of the minor dimension. Berlad and Potter [9] developed a correlation to calculate the quenching distance for an elliptical cross-section tube relative to that for a circular tube. The correlation is described as follows:
2 do2 = d22 , 2 1 + ( d2 d1 )
(22.1)
where the axes of the port are defined as d1 and d2 with d2 > d1 and do is defined as the quenching diameter of the equivalent circular port. There are also extensive experimental data on quenching distances for various fuels in air and oxygen enriched air available in Lewis and von Elbe [10]. The data provided can be used to calculate the effect of oxygen enrichment on quenching distance for ribbon burners and to suggest ribbon burner port diameters for safe operation. The resulting correlation suggests that for stoichiometric methane and air (0.21 oxygen mole fraction in air) a safe port minor dimension is approximately 2.57 mm, while for operation with air enriched to 0.35 oxygen mole fraction, the safe burner port minor dimension is reduced to approximately 1.04 mm. Other important concerns in designing stable and safely operating ribbon burners are the total number of ribbons and ribbon packing. Widening of the ribbon (i.e., increasing the total number of packed corrugated sheets) increases the surface area in contact with the flame. This causes the central portion of the ribbon to be further removed from the cooled burner housing. The conductive and radiant heat transfer mechanisms are thereby diminished, forcing the burner to absorb more energy, potentially leading to failure of the ribbon. Similarly, one must ensure that the ribbon is tightly packed in the burner housing to obtain maximum thermal contact and prevent overheating [5], again highlighting the fact that both stability and safety must be taken into consideration with burner design. 22.2.2 Variables Influencing Flame Environment Structure Similar to the many parameters that influence burner stability, there are many variables that affect the structure of the multiple flame cones being emitted from a ribbon burner as well as the postflame environment created. It is important to understand how these variables interact in order to better predict how the flame environment is influenced. The variables include burner configuration, flame power (in units of W/cm2, calculated based on the fuel flow rate, the heating content of the fuel, and the area of the ribbon face), flame stoichiometry, the presence of an impingement surface, the gap between the
453
Characterization of Ribbon Burners
burner surface and the impingement surface, and the relative motion of the impingement surface. Building on previous studies [4,5], researchers [2] performed qualitative stability and flow regime analyses relating changes in the combustion environment to changes in burner configuration as well as the other additional operating parameters described above. In addition to the flow regime analyses, a heat transfer analysis was also performed and will be detailed in the following paragraphs. The experimental setup mirrors that used in the process of industrial flame treating. The setup consists of a mounted 25 cm diameter roller (the impingement surface) that is driven by a variable-speed drive train allowing for rotational speeds between 0 and 7 Hz, translating to a roller rotational velocity of 0 to 5.5 meters per second (m/s). The roller is chilled to approximately 400 K to prevent any thermal damage that could occur due to the presence of the flame and to prevent the buildup of condensation on the roller surface. The burner is oriented beneath the roller, parallel to the face of the roller. The vertical separation distance or gap between the roller and the burner can be varied along with the horizontal positioning of the burner in relation to the roller. The stability and flow regime analyses utilized a flow visualization process called schlieren imaging. With this technique, the flame thermal gradients are made visible due to the deflection of light resulting from the presence of density gradients within the flame. An inline color filter qualitatively indicates the degree of diffraction that occurs, giving a blue color for smaller thermal gradients, green and yellow for moderate gradients, and orange and red for the largest gradients, providing insight concerning gradient intensity and location. 22.2.2.1 Effect of Burner Configuration on Flame Environment The study done by Stroud et al. [2] relates changes in burner configuration to changes in flame stability and the flame environment created above the burner for laminar, stoichiometric flame conditions incorporating the influence of crossflow resulting from impinging surface motion, which is a condition that is often seen in industry processes. As stated in Section 22.2.1, the optimal flame separation distance has been experimentally determined to be 1.25–2.25 mm [5] under conditions without impinging surface motion. Based on these results, one would expect that three or four typical ribbons between each row of ports would be optimal as each ribbon typically has a width of 0.5 mm. Six burner configurations were considered, including 8-, 4-, and 2-port burner with the row separation distances varying from 1.0 mm to 2.5 mm [2]. The 8-port burner, with a 1.5 mm or 2.0 mm separation distance,
© 2011 by Taylor and Francis Group, LLC
created a stable flame environment. In comparison to the 8-port burner, the 4-port burner was more susceptible to air entrainment at the highest roll rotational speeds (5.50 m/s) due to the reduction in the number of jets providing a cumulative resistance to the rotational velocities. The 4-port burner with a 1.0 mm ribbon separation did not present a stable flame even with a stationary roller. Entrainment began at low roll rotational speeds and the flames lost stability as the rotational speed of the roll increased. The 2-port burner provided a stable flame when the impingement surface was stationary, but because there are no outer flames to shield against the entrained ambient air, the flame was completely extinguished at a high roll rotational speed (5.50 m/s). A 4-port burner with a 2.5 mm row separation showed the same vulnerability to entrainment as the 1.0 mm design mentioned above since the flames are too far apart to provide enough stabilization in the boundary layer formed on the roller at a processing speed of 5.50 m/s. These results [2] suggest that a burner with a 1.5–2.0 mm (three- or four- ribbon) row separation distance and at least four ports is desirable to create a stable and uniform environment above the burner that is resistant to entrainment effects introduced due to impinging surface motion. These results are in agreement with the optimal flame separation distance previously determined by Pearson, Saunders, and Hargreaves [5]. 22.2.2.2 Parameterized Flow Regime Analysis With the optimal row separation distance established, Stroud et al. [2] also performed a parameterized flow regime analysis, where the flame power, burner-to-roll (impinging surface) gap, and roll rotational speed were systematically varied at stoichiometric conditions for a methane/air flame environment created by a 4-port, four-ribbon burner. These changes in the thermal flame structure and postcombustion environment were captured on film utilizing the schlieren imaging technique. The images were visually compared and categorized as laminar, partially mixed, or highly mixed based on the level of small-scale fluctuations that were present in the flame environment. The effects of air entrainment were also considered in order to determine if the free jet velocities were influenced by the roll rotational velocity. Figures 22.3 and 22.4 provide examples of the schlieren images captured and the details of the experimental conditions and categorized results are presented in Table 22.1, with flame powers. In Figures 22.3 and 22.4, the roll is moving in a clockwise direction and is seen as the black circular object at the top center of the images. The ribbon burner is the black object at the bottom center of the images, viewed along its cross-web axis. The colors are an integration
454
Industrial Combustion Testing
Figure 22.3 Schlieren images of nonimpinging conditions (50 mm gap) for a 4-port burner with a clockwise roll rotation of 0.785 m/s; left images: Low flame power (309 W/cm2); right images: High flame power (1570 W/cm2).
Figure 22.4 (See color insert following page 424.) Schlieren images of impinging conditions (7 mm gap) for a 4-port burner with a clockwise roll rotation of 5.50 m/s; left image: Low flame power (309 W/cm2); right image: High flame power (1570 W/cm2).
of the flame diffraction effects along the length of the burner (cross-web or x direction). Figure 22.3 presents schlieren images of the flame structure and the postcombustion environment with a large burner-to-impingement surface gap (H > 50 mm) in which the flame cones generated do not impinge on the roll surface. The left images in Figure 22.3 depict a low flame power condition (309 W/cm2). The flame itself is laminar with a large region of little or no thermal variation at the center of the flame treatment region, as indicated by the dark blue color between the burner and the roll surface. The highest thermal gradients are indicated in red and are found at the outer edges of the flame zone. The right images in Figure 22.3 show a high flame power condition with high reactant flow rates (1570 W/cm2). The presence of taller, more defined flame cones along with
© 2011 by Taylor and Francis Group, LLC
increased small-scale fluctuations are evident and the size of the thermally uniform core (indicated in dark blue) is reduced, indicating a more highly mixed flame environment. Higher-resolution schlieren images were taken to observe at closer range the behavior of the flame cones and the generated postcombustion environment under the same conditions and are shown in the bottom two images in Figure 22.3. As can be seen in both high resolution images, the flame cones, indicated by the areas of cone-shaped yellow and red color located near the burner surface, do not appear to be affected by the presence of the impingement surface. There is a large area of constant high temperature centered above the burner, indicated by the black/dark blue section in the center of the picture in the low flame power
455
Characterization of Ribbon Burners
Table 22.1 Flow Regime Analysis Results Reactant Flow (lpm) 105
160
206
240 305
Roller Rotational Velocity (m/s)
Flow Velocity (m/s)
Free Jet Reynolds No. (Red)
0.785 0.785 2.35 2.35 3.93 3.93 5.50 5.50 0.785 0.785 0.785 2.35 3.93 5.50 5.50 0.785 0.785 0.785 5.50 5.50 5.50 0.785 to 5.50 0.785 to 5.50
4.78
415
7.65
664
9.36
812
10.9 13.9
947 1210
Gap (mm) 10–50 5–7 25–50 5–20 25–50 5–20 20–50 5–15 25–50 15–20 5–10 5–50 5–50 25–50 5–20 40–50 30–35 5–25 45–50 25–40 5–20 5–50 5–50
Flow Regimes* L L L L PM PM PME PME PM PM PM PM PM PM PME ST ST PM to HM PM to HM PM to HM PM to HM HM HM
* Observed flow regimes: laminar (L), laminar with entrainment (LE), partially mixed (PM), partially mixed with entrainment (PME), highly mixed (HM).
configuration. In general, under nonimpinging conditions (H > 50 mm), the low flame powers reflect laminar conditions with little-to-no mixing aside from the entrainment of air near the impingement surface at higher rotational speeds (5.50 m/s). The stability of the flames being emitted from the burner is not affected by surface motion. The high flame power images also reflect extremely stable conditions, but with a high level of mixing. The high-power flames show limited effects due to air entrainment as a result of surface motion. Figure 22.4 captures the conditions within the flame treatment zone as separation distance H is decreased to 7 mm and the roll rotational speed is increased to 5.50 m/s, which is a relatively extreme condition for flame treatment. The left image in Figure 22.4 represents a flame at low flame power (309 W/cm2) impinging on the roll and the right image shows the environment created with high flame powers (1570 W/cm2) at the same gap. At the high flame power, the flame cones are affected by the presence of the impingement surface. Mixing is increased under these conditions as the flame cones deform, causing a redirection of the fluid flow. Under these conditions there is little entrainment of ambient air along the roll surface and the stability of the flame
© 2011 by Taylor and Francis Group, LLC
environment is not affected. In contrast, at the low flame power, the schlieren image shows laminar behavior with few small-scale fluctuations in thermal gradients and minimal mixing. The effects of roll rotation and entrainment are obvious as the flame zone is influenced by the rotational flow, disturbing the upstream environment where ambient air is introduced into the treatment zone. Under these conditions the momentum of the jets is overtaken by the influx of entrained air and the environment above the burner is perturbed. For a flame, or reacting jet, the free jet Reynolds number, Red, is defined as:
Red =
ud , υ
(22.2)
where the characteristic length, d, is the burner port average diameter (1.38 × 10 –3 m), which is calculated based an average port open area of 1.5 mm2, u is the unburned gas velocity (see Table 22.1), and υ is the reactant kinematic viscosity (1.59 × 10 –5 m2/s). In accordance with Shaddix [11], the properties of nitrogen are used for the Reynolds number calculations as they provide a reasonable approximation for the properties of methane–air
456
Industrial Combustion Testing
1. Re < 300 2. 300 < Re < 1000 3. 1000 < Re < 3000 4. 3000 < Re
dissipated laminar jet fully laminar jet semiturbulent jet fully turbulent jet
The experimental observations of Stroud et al. [2] show that the actual flame environment remains laminar below a free jet Reynolds number of 600, starts to transition from laminar to partially mixed at Re > 600, and becomes highly mixed around Re = 900, in comparison to the expected Re = 3000, which is considered to be the fully turbulent condition for a single nonreacting jet. The flame environment created by the array of jets is very highly mixed, resembling a fully turbulent condition even though the Reynolds number is lower than that expected for a single nonreacting fully turbulent jet. This is due to the effects of combustion-induced turbulence and jet interaction, as higher flow velocities are coupled with greater heat transfer and transport properties due to increased mixing. As suggested by Lewis and von Elbe [10], turbulence can be introduced by the chemical kinetics, rapid volume expansion, and general velocity increase occurring within the flame as a result of combustion. The roll rotational velocity can also be influential in the level of thermal gradient fluctuation that is produced, especially with larger gaps between the burner and impingement surface. The increased levels of entrained ambient air can cause a laminar flame environment to appear more partially mixed with increased smallerscale thermal fluctuations as depicted by the top left image in Figure 22.3. In general, however, roll rotation can only induce small changes in the flame environment at low flow velocities and has little to no influence on the flame environment at high flame power, high flow velocity conditions. 22.2.2.3 Heat and Mass Transfer Nonuniformities A significant problem in applications involving multiple impinging jets is the nonuniform heat and mass transfer profiles that can be generated on the impingement surface. To prevent nonuniform heat and mass transfer, the jets need to be well mixed prior to impingement so that a homogenous postflame environment is produced.
© 2011 by Taylor and Francis Group, LLC
For this reason detailed flow velocity and thermal measurements were made by Stroud et al. [2] in order to characterize the flow above the burner, complimenting the flow regime analyses work described in the previous section. Cold-flow velocity field measurements were made, employing hot-wire anemometry to determine the flow characteristics without the influence of combustion. With this diagnostic technique, fluid velocity is measured by sensing the changes in heat transfer from a small, electrically heated sensor exposed to the fluid motion. Under conditions in which fluid temperature, composition, and pressure are constant, the dominant mechanism of heat transfer from the wire is through forced convection, which is directly related to the velocity of the fluid. By measuring the voltage required to maintain the desired hot-wire temperature, the fluid velocity can be accurately determined. The cold-flow velocity field measured at a position z = 2 mm above the ribbon burner operating at laminar conditions (equivalent to 490 W/cm2 for a reacting flow) is shown in Figure 22.5. Under these conditions the highest reactant flow velocity was found to be 6.8 m/s. The velocity field is highly nonuniform over the main combustion zone (10 < y, down-web < 20 mm). Interjet mixing can be observed in the contour plots of the main combustion zone (10 ≤ y ≤ 20 mm) shown in Figure 22.6. The jets formed above the port openings are easily identified at the low separation distance of z = 2 mm, but become increasingly difficult to distinguish as the vertical distance is increased. While velocities are relatively high at z = 10 mm, adjacent jets have combined at this
6 4
5
2
10 Do wn -w
eb
15 pos itio n
(m
m)
20 25
8
6
4
2
0
0
Air velocity (m/sec)
mixtures. All Reynolds numbers were calculated using properties defined at 300 K, assuming that the reactant fuel and air are close to the ambient temperature as they exit the burner ports. The flow regimes established for the environment created by the array of reacting methane-air jets were similar to the published regimes for single nonreacting circular jets, which are defined in the literature as follows [7]:
eb s-w mm) s o Cr ion ( sit po
Figure 22.5 Reactant gas velocity field above 4-port ribbon burner under laminar conditions at vertical position z = 2 mm.
457
Characterization of Ribbon Burners
2 mm
6 mm
(b) 20
14
18 16 14
12 0
1
2 3 4 5 6 7 8 Cross-web position, x (mm)
(c)
12
9
0
10 mm
1
(d)
2 3 4 5 6 7 8 Cross-web position, x (mm)
9
14 mm
20
16 14
20 Down-web position, y (mm)
18
18 16 14
12 0
1
2 3 4 5 6 7 8 Cross-web position, x (mm)
Down-web position, y (mm)
16
Down-web position, y (mm)
18
20
9
Down-web position, y (mm)
(a)
12 0
1
2 3 4 5 6 7 8 Cross-web position, x (mm)
9
Figure 22.6 Reactant gas velocity contours at increasing distance above the ribbon burner surface under laminar conditions. Contour numbers in figure represent velocity in m/s, shown in steps of 1 m/s. (a) 2 mm, (b) 6 mm, (c) 10 mm, and (d) 14 mm.
height, reaching a plane of horizontal uniformity [12] relative to the z = 2 and z = 6 conditions and fluctuations in the flow appear random. At z = 14 mm, in Figure 22.6, the flowfield dissipates significantly and little structure can be identified. Incorporating combustion effects, Figure 22.7 presents temperature data for highly impinging flame environments for laminar (309 W/cm 2) and the highly mixed (923 W/cm 2) flow regimes. The radiation loss corrected thermal characterization of the combustion environment was captured utilizing an Omega R-type thermocouple (platinum/platinum-13% rhodium) positioned above the burner. A traversing system controlled with one automated (y direction) and two manual micrometers (x and z direction) was used for thermocouple positioning. Under the laminar condition (left), similar to the conditions explored in the cold-flow analysis, the combustion environment
© 2011 by Taylor and Francis Group, LLC
temperatures range above 2100 K with the difference between the minimum and maximum local temperatures being less than 230 K. The uniform plane noted with the cold-flow velocity measurements is evident in the thermal measurements as the difference in the average temperature in the x or cross-web direction is less than 100° under laminar flow conditions. In contrast, for the highly mixed flow regime, the difference between the minimum and maximum measured temperatures is approximately 1250 K. This increased temperature difference occurs at these high flow rates, highly impinging conditions because there is little time for mixing of the postcombustion gases that would result in a more uniform thermal environment. For both flame environments, higher temperatures are found between the ports [7,12] and the lowest temperatures are seen above the port openings due to the presence of unreacted fuel and air.
458
Industrial Combustion Testing
14
Temperature (K) 2189 2173 2156 2140 2124 2107 2091 2074 2058 2042 2025 2009 1992 1976 1960
Up-web position (mm)
12 10 8 6 4
Temperature (K) 1971 1882 1792 1703 1613 1524 1435 1345 1256 1166 1077 987 898 809 719
12 10 8 6 4 2
2 0
14
0
1 2 3 4 5 Cross-web position (mm)
0
0
1 2 3 4 5 Cross-web position (mm)
Figure 22.7 (See color insert following page 424.) False color image of temperature above a 6-port ribbon burner. (Left) Laminar, 7 mm gap, 0.785 m/s Roll Rotation, 309 W/cm2; (Right) highly mixed, 4 mm gap, 0.785 m/s Roll Rotation, 923 W/cm2.
22.3.1 Overview and Discussion of Wettability
22.3 Flame Treatment of Polymer Films As mentioned in the introduction to this chapter, the flame treatment of polymer films presents a challenging operating environment for the design of the burner and placement of the treatment surface. The process involves exposing the surface of a thin polymer film to the reactant species in the mixture downstream of the flame in order to produce a film surface that has a range of desired properties, without causing thermal degradation of the film. One of the desired properties of the treated film is an increase in surface wettability so that water-based inks and adhesives will adhere to the surface treated film. For this application, precise control of the flame environment and its spatial uniformity is essential. In this section, the flame treatment process for increased wettability and the flame parameters that need to be optimized to produce the highest degree of wettability are described. These parameters include those that influence burner stability, the combustion flow regimes created, and the uniformity of the heat and mass transfer that results, as discussed in Section 22.2. The surface chemical reactions between the postflame reactants and the polymer film have also been studied extensively and will be discussed in the following section.
© 2011 by Taylor and Francis Group, LLC
Corona discharge treatments have long been the dominant method of treating polymer films, principally because of concerns about the safety of open flames in industrial environments and the observed sensitivity of flame treatments to small changes in process conditions. Recently, however, flame treatment has received renewed industrial interest as a technique for modifying polymer films. Some of the reasons for this renewed interest include: major improvements in the safety, reliability, and ease of operation of flame-treating equipment; energy consumption and equipment space requirements comparable to corona treatment; the ability to achieve high levels of surface oxidation of polymers at extremely short processing times (< 0.5 s); and no generation of toxic, corrosive ozone [13]. Untreated, many polymers, in particular polyolefins, are hydrophobic and therefore tend to repel water and water-based substances. The oxidation of the film through flame surface treatment increases the surface energy of the film, making the material less repellent of water, improving film wettability. Increasing wettability is a necessary component of many manufacturing processes that require water-based inks and adhesive materials to be applied to the polymer film surface.
459
Characterization of Ribbon Burners
F ( x) =
γP bx cos θ ( x ) + , g 2
(22.3)
where F(x) is the measured force on the flame-treated film, γ is the measured surface tension of the water, P is the measured perimeter of the sample, g is gravitational acceleration, θ(x) is the contact angle, b is the buoyancy factor, and x is the position of the sample in the water. The Wilhelmy equation is typically used to calculate a single, average contact angle for the film surface [14–17]. In general, the advancing water contact angle is most sensitive to the low-energy, unmodified components of the surface while the receding angle is more sensitive to the high-energy, oxidized groups introduced by the flame treatment [14]. In addition, in many industrial coating processes, liquids are physically forced to wet a film surface; whether or not the coating continues to uniformly wet the film during the drying or curing of the coating can depend upon the receding contact angle of the coated liquid on the film substrate. Therefore, to best characterize the flame-treated polypropylene, both the advancing and receding contact angles are examined, with the receding angle being the more important
© 2011 by Taylor and Francis Group, LLC
22.3.2 Flame Treatment Parameters 22.3.2.1 Effect of Fuel/Air Equivalence Ratio As previously mentioned, polyolefin films are often exposed to a flame to improve the wetting and adhesion properties of the polymer. Efforts were made by Branch et al. [12] to investigate the relationship between the flame conditions during treatment and the improvement in the wettability of polypropylene film. Figure 22.8 shows a plot of ESCA O/C atomic ratios and the contact angles of water on flame-treated polypropylene as a function of the equivalence ratio, where the equivalence ratio (φ) is defined as: Moles of Fuel Moles of Oxidizer Actual φ= , Moles of Fuel Moles of Oxidizer Stoichiometric (22.4) at a constant burner-to-film separation of 5 mm. The flame-treated polypropylene samples were analyzed by ESCA and contact-angle measurements utilizing the Wilhemy plate method [14,15] previously described in Section 22.3.1. As lower contact angles are indicative of greater wettability, the contact angle axis is plotted from high-tolow values. For comparison, untreated polypropylene 0 20
0.2
Receding angle Advancing angle O/C ratio
0.15
40 60
0.1
80
0.05
100 120 0.68
0.78
0.88 0.98 Equivalence ratio
1.08
ESCA O/C atomic ratio
measurement. Lower contact angles are indicative of higher surface energy, increased wettability, and better adhesion. Specific examples of the Wilhelmy force traces are shown and discussed in detail in Section 22.3.2.3.
Contact angle of water (degrees)
The main method for evaluating wettability is contactangle measurements made utilizing the Wilhelmy plate method [14,15]. With the Wilhelmy plate technique, the advancing and receding contact angles of a probe liquid, such as water, are calculated from the measured force exerted on a film sample as the sample is immersed in or withdrawn from the liquid. The Wilhelmy system comprises an electrobalance, which is used to measure the wetting force on the sample, and a movable stage, which can travel up and down at controlled speeds to advance or recede the liquid front across the film sample [14,15]. The contact-angles measured with this technique are a direct measure of the film wettability with lower angles representing more wettable surfaces. X-ray photoelectron spectroscopy (XPS or electron spectroscopy for chemical analysis (ESCA)) is also utilized as the analytical process provides surface chemical information in the form of an oxygen-to-carbon or O/C atomic ratio, data that are complimentary to contact-angle wettability measurements. Since polypropylene is comprised of purely carbon and hydrogen prior to treatment, the O/C ratios indicate the level of oxidation of the film surface as a result of the treatment process. Again, a higher level of surface oxidation increases the surface energy of the film, therefore improving the wettability. In the measurement of dynamic contact angles utilizing the Wilhelmy plate method with water as the probe liquid, the force on the film is given by the Wilhelmy equation:
0
Figure 22.8 Plot of the ESCA O/C atomic ratio and the contact angles of water on flame-treated polypropylene as a function of the equivalence ratio with a constant burner-to-film separation of 5 mm. Optimum treatment occurs at φ = 0.93 where oxidizing species concentrations are at a maximum.
460
22.3.2.2 Effect of Film Positioning within the Flame Treatment Environment Researchers [16–18] have reported the need to position the film slightly beyond the luminous or visible flame front for improved polymer film surface treatment results. Branch et al. [12] worked to provide a detailed justification of this claim. Although the optimum equivalence ratio is independent of the position of the film relative to the flame, the position of the surface to be treated relative to the flames has a large effect on the level of wettability achieved with flame treatment for a given flame power. Figure 22.9 shows contact-angle measurements as a function of the position of the polymer film relative to the flame at the optimum equivalence ratio of 0.93 and a flame power of 500 W/cm 2. Similar trends are seen at other equivalence ratios. In Figure 22.9, a position of approximately 2 mm represents the location of the tips of the flamelets. Optimal treating at these laminar treatment environment conditions occurs at distances between the tips of the flamelets and the polypropylene film of 0–2 mm. At greater flame-to-film distances, treatment effectiveness declines gradually, although some improvement in wettability is still noted at distances of up to 20 mm.
© 2011 by Taylor and Francis Group, LLC
0 Contact angle of water (degrees)
has an advancing contact angle of 108° and a receding contact angle of 87°. As a function of the equivalence ratio, the contact angles of water on flame-treated polypropylene decrease as φ increases, with a minimum in the contact-angle values occurring at an equivalence ratio of about 0.93, as shown in Figure 22.8. In general, the study performed by Branch et al. [12] indicated that φ = 0.93 for a methane/air flame is optimal for all combinations of flame-to-film distance, flame power, and film speed. As the flames become fuel-rich, there is a steep increase in the contact angles at equivalence ratios from 0.97 to 1.05. Treatment of polypropylene in a fuel-rich flame (φ > 1) yields little improvement in wettability. The surface chemistry of flame-treated polypropylene closely correlates with the wettability of the surface. Figure 22.8 also shows the ESCA O/C atomic ratio of flame-treated polypropylene as a function of the equivalence ratio. The amount of surface oxidation generated by the flame follows the same trend as the wettability. The maximum O/C ratio of 0.18 is obtained at φ = 0.92 – 0.94. Within this range, the gas phase concentrations of surface oxidizers in the flame are high. This surface oxidation is the reason for the improved wettability and adhesion properties of flame-treated polypropylene. A detailed discussion of the flame and surface chemistry involved in surface treating is presented in Section 22.4.
Industrial Combustion Testing
20
Receding angle
40 60 80
Advancing angle
100 120
0
2
4 6 8 10 Distance from burner face (mm)
12
Figure 22.9 Contact-angle measurements as a function of the position of the polymer film relative to the flame at the optimum equivalence ratio of 0.93.
22.3.2.3 Effect of Temperature and Flow Environments Nonuniform treatment of the polypropylene film, leading to irregularities in the wettability of the surface of the material can often occur. Building on the discussion initially presented in Section 22.2, Strobel et al. [16] provide more insight on the formation of nonuniform temperature and flow environments above the ribbon burner by continuing to look at the causes behind the nonuniform surface treatment of polymer films. Depending upon the treatment conditions, polypropylene can be exposed to an inhomogeneous environment because of the conical shape of the flames [12,17]. This nonuniform environment can lead to cross-web variations, or “lanes,” in the wettability of the film [15]. The lower-wettability lanes in the polypropylene are not untreated film; they are merely slightly less wettable than the other portions of the flame-treated polypropylene. However, even this slight variation can have detrimental consequences in the subsequent processing of the flame-treated film. The cross-web lanes have a periodicity that corresponds to the geometrical spacing of the outlets, or ports, along a single cross-web row of the ribbon burner that is used to support the flame [12,17]. For a typical industrial ribbon burner, the spacing of the ports along a single row is approximately 2.5 mm. However, the design of the ribbon burner is such that successive rows of ports are staggered, so that there is a port approximately every 1.25 mm in the cross-web direction. As a result, the film is actually exposed to a flame cone every 1.25 mm in the cross-web direction as it passes by the burner, but the resulting period of variation in film oxidation is 2.5 mm relative to the expected 1.25 mm. This is due to the fact that the upstream pair (Figure 22.1) is more effective in oxidizing the surface of the film
461
Characterization of Ribbon Burners
in comparison to the downstream pair as determined by “registering” the position of the film in relation to a given burner port by marking the film with a scratch [16]. The scratch allows specific features in the Wilhelmy force trace to be associated as a function of position with a given port on the ribbon burner (see Figure 22.10 and 22.11), providing information on the spatial nonuniformities of the flame-treated polypropylene. Two types of registration experiments are presented by Strobel et al. [16] comparing the changes in the Wilhelmy force traces as a function of position when either an upstream or a downstream primer port is registered as well as plugged. Figure 22.10 depicts the
Advancing wilhelmy force (mg)
45 Scratch at location of peak
40
Plugged port
35 30 25
Scratch location
20
15 10 Cross-web position (mm)
5
20
Figure 22.10 Wilhelmy advancing force data for flame-treated polypropylene. To generate this data, a scratch was made in the polypropylene film directly above an upstream primer port while a second upstream primer port located 5 mm away from the scratch was also plugged with a short length of wire. This figure illustrates that laning, or nonuniform treatment has the same spacing as the cross-web port spacing.
Advancing wilhelmy force (mg)
50 45
Peak
40 35
Valley
30
Peak Peak
Valley
Plugged port valley has less effect when downstream Scratch location is in a valley
25 20 15
Peak
5
10
15 20 Cross-web position (mm)
25
Figure 22.11 Wilhelmy advancing force data for flame-treated polypropylene. To generate this data, a scratch was made directly above a downstream primer port while a second downstream primer port located 5 mm away from the scratch was also plugged.
© 2011 by Taylor and Francis Group, LLC
trace when registering and plugging an upstream pair of primer ports. Figure 22.11 depicts the trace when registering and plugging a downstream set of ports. The location of the scratch or registered port is indicated in each figure, along with the locations of the plugged port and the force peaks (lower contact angles) and valleys (higher contact angles). Note the extremely regular 2.5 mm periodicity in the traces. When an upstream primer port is used to make the scratch (Figure 22.10), the scratch discontinuity in the Wilhelmy trace is always located in an expected peak position. Thus, the upstream pair of ports is always associated with higher Wilhelmy forces, lower contact angles, and more effective oxidation of the polypropylene. The plugged upstream primer port that is separated by 5 mm from the scratch also corresponds to a peak position. However, in most instances, the “plugged-port peak” is of lower overall force than the adjacent peaks. This indicates that each burner port comprising the upstream pair contributes to the surface oxidation of the film. When a downstream primer port is used to make the scratch (Figure 22.11), the scratch discontinuity is always located in an expected valley position. The downstream pair of ports is thus associated with lower Wilhelmy forces, higher contact angles, and less effective treatment of the polypropylene film. The presence of the plugged downstream primer port that is separated by 5 mm from the scratch is not easily detected in the Wilhelmy traces. There is little, if any, difference in the forces between the valley formed by the plugged port and the valleys formed by unplugged downstream primer ports. This observation implies that the downstream primer ports have little influence on laning. These experiments [14] show that the upstream pair of burner ports is more effective in oxidizing the polypropylene surface than the downstream pair of ports providing an explanation for the 2.5 mm spacing of the lanes of nonuniform surface treatment. The difference between the upstream and downstream pairs of ports is best illustrated by determining the average temperature of the postcombustion gases in the downstream direction as a function of cross-web position. The term “average temperature” means the average of all temperature measurements made along a downstream line at a particular cross-web position. The average temperature correlates to the number of active oxidizing species that the film is exposed to as the film traverses downstream across the face of the burner. Average temperature data are presented in Figures 22.12 and 22.13 for a flame power of 309 W/cm2. In these figures, the x-origin is located at the centerline of the upstream pair of ports. Figure 22.12 shows the average temperatures near the surface of the backing roll, which is the impingement surface, as a function of cross-web
462
Industrial Combustion Testing
Average temperature (K)
1825 1820 1815 1810 1805 1800 1795 1790
1
0
2 3 Cross-web position (mm)
4
5
Average temperature (K)
Figure 22.12 The average flame temperature (in K) experienced by the polypropylene film during passage through the flame as a function of cross-web position for a 12 mm burner face-to-backing roll gap. The x-origin represents a line passing through the upstream pair of ribbon burner ports. 1800 1780 1760 1740 1720 1700 1680 1660 1640 1620
0
1
2
3
4 5 6 7 Cross-web position (mm)
8
9
10
Figure 22.13 The average flame temperature (in K) experienced by the polypropylene film during passage through the flame as a function of cross-web position with a 4 mm burner face-to-backing roll gap. The x-origin represents a line passing through the up-web pair of ribbon burner ports. The figure shows that the peaks in the average temperature in the cross-web direction have the same spacing as the peaks in surface treatment levels.
position along the burner for a burner face-to-backing roll gap of 12 mm. In both of these cases, the backing roll was positioned such that the thermocouple came as close as physically possible to the surface of the backing roll at an up-web location directly above the centerline of the burner (i.e., at the bottom-dead-center location of the backing roll above the two main ports of the burner). When the burner is located far from the backing roll, the temperature field near the backing roll surface is quite uniform. The range in average temperatures across the measured area in Figure 22.12 is < 25 K and the local temperature profile is similar to that presented for the laminar environment shown in Figure 22.7 (left) with the average temperatures being slightly lower due to the increased gap between the burner and backing roll surface.
© 2011 by Taylor and Francis Group, LLC
At the flame conditions used to generate Figure 22.12, the array of conical flames supported on the ribbon burner is far from the backing roll. The combustion products from each conical flame have ample time to mix prior to impinging upon the polypropylene surface that is held against the backing roll. The relatively homogeneous flame environment at the backing roll under these “wide gap” conditions causes relatively uniform treatment of the polypropylene surface. As determined in another study [15], polypropylene film treated at these conditions had no detectable treatment lanes. A uniform flame temperature field at the backing roll surface translates into a uniform surface treatment of the polypropylene. Unfortunately, increasing the burner-to-roll gap to eliminate laning comes at the cost of a reduction in the overall surface treatment of the polypropylene (see Figure 22.9, for example). As the burner-to-roll gap is decreased at a constant flame power, the products of combustion have less time to mix prior to contact with the backing roll so that the flame environment at the roll surface is much less uniform. Figure 22.13 shows the average temperatures near the surface of the backing roll as a function of cross-web position above the burner for a burner-to-roll gap of only 4 mm. The range in average temperatures across the measured area increases to approximately 120 K at the 4 mm gap in comparison to the wide gap condition and the trends in the local temperature profile are more similar to those presented for the highly mixed environment shown in Figure 22.7 (right), as the port locations, marked by cooler pockets of unreacted fuel and air, are more clearly defined in comparison to the larger gap condition. The small cross-web variation in treatment that is manifest as laning is caused by the nonuniform flame environment above the burner itself. Again, it is presumed that the concentration of active species is also nonuniform above the burner ports. The temperature measurements show a difference between the upstream and downstream pairs of ports. The coolest average temperatures are found along the line of the downstream pair of ports that are associated with lower Wilhelmy forces, higher contact angles, and less effective treatment of the polypropylene film. These “cool-temperature valleys” in Figure 22.13 are spaced 2.5 mm apart. Cooler average temperatures are associated with lower concentrations of active species and lower levels of surface treatment. Conversely, the higher average temperatures in Figure 22.13 correspond to the upstream pair of ports, which are the ports associated with higher Wilhelmy forces, lower contact angles, and more effective treatment. Laning on flame-treated polypropylene is a small, but significant, nonuniformity in the extent of surface oxidation and wettability that is superimposed upon a much larger average increase in oxidation and wettability.
463
Characterization of Ribbon Burners
The wettability and temperature data presented by Strobel et al. [16] show that the polypropylene film that passes above the upstream pair of ribbon-burner ports is exposed to a greater average temperature during passage through the flame. This greater average temperature correlates to an exposure to a greater concentration of the active species that cause the surface oxidation of the polypropylene. It is now believed that some level of laning is inevitable in flame treating because all viable commercial flame-treatment equipment uses an array of conical flames. Conical flames do not quickly coalesce into a uniform flame front, but remain distinct some distance above the burner surface. While other burner designs that may generate a more uniform flame environment than a ribbon burner can be envisioned, such designs are not practical for industrial flame treatment. A large array of conical flames is essential to provide the high flame powers typically needed for treatment at industrial film speeds. Whenever an array of flames are used, there will be an inherent nonuniformity in the flame environment used to modify the substrate. This nonuniformity will be present for all types of flames and for both ribbon and drilled-port burners. The severity of laning can be reduced, but not completely eliminated.
22.4 Roles of Flame and Surface Chemistry in Flame Treating The previous section of this chapter has described the flame parameters that must be controlled and optimized to produce a polymer film of the desired level of wettability. The present section is a review of the current state of knowledge of the chemical kinetic mechanism of the reactions between flame reactants and the surface layer of polymer molecules. The discussion begins with a consideration of the flow, impingement, and quenching of the combustion products and reactive intermediates on the cooled surface of the polymer film. The discussion then proceeds to describe a detailed chemical kinetic mechanism for the surface reaction. The resulting mechanism is able to qualitatively account for the influence of the major flame variables on the wettability of the polymer surface. Finally, the addition of secondary species, specifically nitrous oxide (N2O), to the primary reactants to alter the thermal and/or chemical behavior of the flame is discussed, providing an example of the effects of flame-chemistry modification. 22.4.1 Flame Chemistry The oxidation of the polymer surface as a result of flame treatment occurs through multiple physical processes,
© 2011 by Taylor and Francis Group, LLC
including homogeneous combustion of the premixed reactants, transport of the products of combustion to the polymer surface, and heterogeneous reaction of flame products with the polymer. While the combustion of natural gas is often simplified by the global oxidation reaction:
CH4 + O2 → CO2 + 2 H2O,
(22.5)
the actual combustion chemistry is comprised of hundreds, perhaps thousands, of elementary reaction steps [18]. The short-lived, highly reactive radical species that are present during these chemical interactions are widely believed to be the chemical species that participate most actively in the polymer oxidation process, as compared to the comparatively stable and unreactive final products (CO2 and H2O) [19]. These gas-phase chemical and transport processes were examined numerically by Sullivan et al. [20] through application of the CHEMKIN chemically reacting flow software libraries [21]. In this commercially available modeling package, the species, temperature, and velocity profiles are calculated for numerous chemically reacting flowfields while accounting for finiterate, gas-phase chemical kinetics and multicomponent molecular transport. For these studies [20], the PREMIX and SPIN applications of the CHEMKIN software library were employed to compare the structure and chemistry of a freely propagating methane–air flame, where there is no impingement surface with that of a flame in a stagnation-flow geometry where an impingement surface is present [22,23]. Under the impinging conditions, the flame is strained causing changes in velocity, temperature, and chemistry when compared with the freely propagating flow. By comparing the results of these models, the effects of the stagnation surface on flame structure and homogenous chemistry can be examined. The idealized stagnation flow geometry is shown in Figure 22.14; reactive gases exit a nozzle at velocity Vinlet and temperature Tinlet and impinge upon a surface at temperature Twall positioned at a separation distance L. This chemically reacting flowfield is solved using the SPIN application, while PREMIX is employed to study the freely propagating flame in which the stagnation surface is absent. The GRI 3.0 chemical kinetics mechanism for methane–air combustion is used in both simulations, employing 49 species and 277 elementary reactions steps [18]. In Figure 22.15a, the temperature and velocity profiles are compared between freely propagating flames and flames that impinge upon a cool stagnation surface. In this and the following figures, the gases travel from right-to-left and the stagnation surface is at position x = 0 mm. In this stable, stoichiometric methane–air
464
Industrial Combustion Testing
Nozzle: gas phase reactants at Tinlet and Vinlet
x
L Figure 22.14 Schematic diagram of the impinging flame model. Solution obtained using the SPIN software from Reaction Design.
(b)
Temperature, T(x), (K)
2000
200
V
1500
150
1000
100
500
50
0
0
2
4 6 Distance, x (mm)
Species mole fraction × 10–3
(c)
8
4
10
H2O2*10–2
O2
6
H
O
2
0
0.5
0.5
0
2
4 6 Distance, x (mm)
8
10
0.0
0.10 OH 0.08
O2
HO2*10–2
0
1.0
4
1
0
1.5
OH
0.06 2
2.0
0 < x < 10 mm
0
0 < x < 2 mm
3
8
250
0 < x < 10 mm
T
Species mole fraction × 10–3
2500
Velocity, u(x), (cm/s)
(a)
1 1.5 Distance, x (mm)
H
0.04
O
0.02
2
O2 mole fraction × 10–1
Cool surface at Twall
O2 mole fraction × 10–1
r
flame, the gas velocity reduces from the initial velocity of 100 cm/sec before igniting at the flame speed of 40 cm/sec at x ~ 5 mm. The temperature increases to approximately 2100 K, then decreases significantly at x < 2 mm because of the high heat transfer from the hot gases to the cold wall. These gases finally reach the surface at zero velocity (stagnation) at temperature Tgas = Tsurface = 400 K. Heat transfer to the wall is observed to effect the bulk of the flame by decreasing the maximum temperature below the adiabatic value of 2220 K. High strain is evident as the temperature drops nearly 1700 K over a quench distance of 2 mm. The observed impinging flame structure contains interesting profiles in postflame gas composition. Molecular oxygen (O2), atomic oxygen (O), atomic hydrogen (H), and hydroxyl radical (OH) are the species
0.00
Figure 22.15 Profiles for the CH4–air flame (a) temperature and velocity, (b) O2, OH, H, and O across the computational domain, and (c) O2, OH, H, O, HO2, and H2O2 near the wall. The profiles show recombination of H, O, and OH near the impingement surface.
© 2011 by Taylor and Francis Group, LLC
465
Characterization of Ribbon Burners
normally deemed responsible for the oxidation of the polymer surface [19]. Figure 22.15b shows the mole fractions of these four species in the strained flame across the flame zone to the stagnation surface. In Figure 22.15c, the behavior of these species in the quench zone (x < 2 mm) is revealed; in addition, the mole fractions of hydroperoxy radical (HO2) and hydrogen peroxide (H2O2) are shown. As shown in Figure 22.15b, the lower temperatures encountered at x < 2 mm correlate to a decrease in the concentrations of O2, O, H, and OH by up to an order of magnitude at the stagnation surface. The limited availability of atomic oxygen at the surface (3 ppm) implies that the role of O atoms in the modification of the polymer film is limited. While O2, H, and OH are also consumed in the quench layer, the concentrations remain relatively high at 1700 ppm, 215 ppm, and 112 ppm, respectively. In comparison to all other species, only the stable products of combustion have higher concentration than these three polymer-reactive species. Figure 22.15c shows the sharp increase in the species H2O2 and HO2, whose concentrations increase by an order of magnitude in the quench zone to 33 and 14 ppm, respectively, similar to the results of Aghalayam and Vlachos [24]. While these species have not previously been postulated as significant in the flame treatment of polymer films, they have been found to be reactive with polymers [25,26]. Though these concentrations are moderate, they are significant, implying that HO2 and H2O2 may play a secondary role in oxidizing the surface. 22.4.2 Surface Chemistry Elements of the mechanism for the surface oxidation of polypropylene by flame treatment have been proposed in previous papers [16,17,19,20,27] and the flame chemistry
has been described in detail in the previous section. The defining characteristics of the mechanism include the formation of polymer radicals by hydrogen abstraction, primarily, under typical flame-treating conditions, by the hydroxyl radical (OH). Polymer radical formation is suggested as the rate-limiting step since the oxidation reactions involving the alkyl radical are expected to be fast and nonrate-determining [28–30]. It is these polypropylene alkyl radical reactions with H, O, OH, HO2, H2O2, and O2 that lead to the oxidation of the surface [16,17,19,20,27]. These studies have also shown that there is limited formation of low-molecular-weight oxidized material (LMWOM) on the polymer surface during flame treating [17,31,32]. The formation of LMWOM is associated with reaction of O atoms and O3 molecules [17,31,32], with the alkoxy functional group being the main precursor to LMWOM formation. This, coupled with the chemistry effects of flame impingement discussed in the previous section, suggests that it is unlikely that the alkoxy functional group is a dominant surface oxidized species in the flame treatment process [17,20,31,32]. Dorai and Kushner [33] have reviewed the work of previous investigators and suggested a full reaction mechanism for the surface oxidation of polypropylene through corona treating. It is known that the mechanisms for corona treatment and flame treatment share many of the same elementary reactions. Modeling work described by Stroud and Branch [34], utilized the Dorai and Kushner plasma mechanism [33], and added the hydrogen-containing elementary reactions specific to flame treatment to provide a complete polypropylene surface reaction mechanism. The final reduced mechanism along with the defined reaction rate coefficients are shown in Table 22.2 , with the functional group definitions shown in Table 22.3.
Table 22.2 Surface Oxidation Reduced Reaction Mechanism Reaction Type Initiation Propagation Termination a
Elementary Reactions
Reaction Rate Coefficienta
1. O + PPH = > PP· + OH 2. H + PPH = > PP· + H2 3. OH + PPH = > PP· + H2O 4. PP· + O = > PPO· 5. PP· + O2 = > PPO2· 6. PP· + HO2 = > PPOOH 7. PP· + H2O2 = > PPOH + OH 8. PP· + H2O2 = > PPO2· + H2 9. PPO2· + PPH = > PPOOH + PP· 10. PPO· + PPH = > PPOH + PP· 11. O + PPOH = > PPO· + OH 12. OH + PPOH = > PPO· + H2O 13. H + PP· = > PPH 14. OH + PP· = > PPOH
1.00 × 10–3 1.00 × 10–5 0.0025 0.01 1.00 × 10–2 0.5 0.5 0.5 3.31 × 108 cm2/sec-mole 4.82 × 1010 cm2/sec-mole 7.50 × 10–4 8.20 × 10–3 0.2 0.02
Sticking coefficient unless otherwise indicated.
© 2011 by Taylor and Francis Group, LLC
466
Industrial Combustion Testing
0.008
Mole fraction
Functional Group Definition for Surface Reaction Mechanism
1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0.7
0.007
Term Definition
PPH PP· PPO· or PPO PPO2· PPOOH PPOH PP = O ·PP = O PPH(alt) (OH)PP = O
CHCH3CH2 CCH3CH2 COCH3CH2 CO2CH3CH2 CO2HCH3CH2 COHCH3CH2 COHCH3CH CHCH3CO CHCH3 CHCH3COOH
0.8
0.85 0.9 0.95 1 Equivalence ratio, ϕ
1.05
0.004
1500
O2/10 OH
0.003
0 0.7
1000
O*10
0.001
500
H
0.8
0.9 1 Equivalence ratio
1.1
0
Figure 22.17 Gas phase species mole fraction versus equivalence ratio.
1.1
1.15
Figure 22.16 Functional group production as a function of equivalence ratio: Reduced and modified mechanism from Table 22.2. This figure shows the optimum treatment is near where PPO2, PPO, and PPOH are a maximum.
It is Reaction 3, OH + PPH = > PP· + H2O, and Reaction 5, PP· + O2 = > PPO2·, that have the greatest effect on the formation of surface oxidizing species within this mechanism, which is in agreement with other researchers [31,32]. Reaction 3, where hydrogen is abstracted by the OH radical, is considered the rate limiting step. Reaction 5 is also extremely influential since molecular oxygen is at such high concentrations at lean-to-stoichiometric conditions and small changes in the reaction rate coefficient result in large changes in the oxidation levels of the polypropylene surface and in the production of the peroxy functional group. The effects of the reduced and modified mechanism (Table 22.1) on the functional groups found on the polypropylene surface are given in Figure 22.16. In comparison to the original plasma treatment mechanism [33], the reduced and modified reaction mechanism presented by Stroud and Branch [34] yields surface species functionalities that are more consistent with the data available from the most recent studies [31,32]. In comparing
© 2011 by Taylor and Francis Group, LLC
0.005
0.002
PPH PP· PPOH PPO· PPO2·
0.75
2000
0.006 Mole fraction
Functional Group
2500
Tad
Temperature (K)
Table 22.3
Figure 22.17, which depicts the calculated variation in concentration of hydrogen and several oxidizing gasphase species as the equivalence ratio of the initial gas mixture is varied, with Figure 22.16, one would expect the PPOH concentrations to remain elevated above PPO· mirroring the gas-phase species counterparts, but this is not the case. Throughout all of the variations in the model, if peroxy (PPO2·) surface radical concentrations are driven to increase at stoichiometric conditions, then PPOH remains low in comparison to alkoxy (PPO·) surface radical. Overall, it is the peroxy polymer radical that is the major oxidized surface species under lean and nearstoichiometric conditions, as shown in Figure 22.16. Under stoichiometric conditions, the peroxy, alkoxy, and hydroxyl functional groups contribute almost equally to the oxidation of the surface. Under rich conditions when the molecular oxygen concentration decreases, the alcohol group (PPOH) is the leading surface oxidized species. 22.4.3 Effect of Nitrous Oxide Additive As noted in Section 22.3, flame treatment of polymer films is typically implemented using natural gas (which is largely methane) and air as the reactants. Propane is also commonly utilized as a fuel. It is well known, however, that the characteristics of premixed flames depend greatly on the reactant species. Thus, the addition of secondary species to the primary reactants to alter the thermal and/or chemical behavior of the flame may be envisioned. The addition of an oxidizing additive, nitrous oxide (N2O), is reviewed here as an example of the effects of this type of flame-chemistry modification. One limitation of flame treatment as currently practiced is that it is only capable of oxidation of polymer surfaces. However, there are many instances in which a different type of chemical modification of the surface
467
Characterization of Ribbon Burners
© 2011 by Taylor and Francis Group, LLC
φ = 0.80
Signal strength
(a)
N4
N2 N3
N1
φ = 0.94
Signal strength
(b)
N5
φ = 1.00
Signal strength
(c)
(d)
φ = 1.20
Signal strength
would be useful. In particular, nitrogen-based surface chemistry is often useful for improving the adhesion properties of polymer films [35–38]. Recent work by Strobel and collaborators [39–41] has led to the discovery of a novel flame treatment process that generates nitrogen-containing polymer surfaces. Two sets of experiments were carried out to elucidate the effects of the N2O addition to the standard methane/air flame [41]. These experiments were carried out at the same normalized flame power as previous experiments [12] with standard flames (490 W/cm2), though it should be noted that nitrous oxide-supported flames will have higher flame temperatures than air-supported flames because there are fewer molecules (of nitrogen) per mole of reactant that must be heated to the product temperature. The first set of experiments was performed to understand the effects of a small addition of N2O to a standard methane/air flame at an equivalence ratio of 1.0. Both ESCA atomic ratio and contact angle analysis show that there is a marked enhancement of wettability by a small addition of N2O. Moreover, the data showed that the addition of N2O to the methane/air flame not only adds nitrogen-containing species to the surface of the polymer, but also significantly increases the amount of oxygen affixation. It should be noted that previous studies have shown that no nitrogen, at the level detected by ESCA, is affixed to the polymer surface during flame treatment with standard methane/air flames [39]. The second set of experiments was performed to investigate the effect of equivalence ratio. An addition of N2O totaling 4% of the total oxidizing-species reactant flow was selected. Contact angle measurements showed that (a) the wettability of the polymer surface improved, relative to treatment with methane/air flames, and (b) the optimal equivalence ratio shifted from ca. 0.93 to stoichiometric (φ = 1.0). Correspondingly, ESCA data showed that, while the addition of oxygen to the polymer surface was a maximum at ca. φ = 0.88 – 0.93, the addition of nitrogen was a maximum at φ = 1.0. It was also discovered that the total amount of nitrogen affixation did not vary greatly with equivalence ratio; however, analysis of the ESCA spectral data indicated that the type of nitrogen functionality added to the polymer surface was a strong function of φ (Figure 22.18). In particular, the researchers found that, in lean mixtures (φ < 1.0), there were both highly oxygenated and reduced nitrogen species on the surface, whereas as the reactant mixture became more rich, the amount of oxygen-containing nitrogen species was greatly reduced. Analysis of likely surface reaction paths and known stability of polymer surface functionalities [41] indicated that the main oxygenated forms of nitrogen on the polymer surface were likely nitrate (R–ONO2) and nitro (R–NO2) groups. The main reduced form of nitrogen
412
410
408
406 404 402 Binding energy, (eV)
400
398
Figure 22.18 Nitrogen 1s ESCA spectra (at an 18° electron take-off angle with respect to the surface) of polypropylene treated in flames containing 4% added nitrous oxide at four different flame equivalence ratios (φ). The 0.80 equivalence ratio represents the most fuel-lean flame, while the 1.20 equivalence ratio represents the most fuel-rich flame. The constituent peaks are labeled as follows: N5: nitrate at 408.1 eV; N4: nitro and nitrite at 406.3 eV; N3: nitrosoamine and nitroso at 402.4 eV; N2: oxime, nitrosoamine, hydroxylamine, and amide at 400.9 eV; and N1: amine and nitrile at 399.8 eV. The arbitrary y-axis represents the signal strength.
was determined to be oxime (R = NOH) groups, with nitroso (R–N = O) and nitrosoamine (R–HN–N = O) as additional primary species. Further reduced species such as hydroxylamine (R = NO–R), amine (R–NH2), and nitrile (R ≡ N) were also speculated, resulting from secondary surface reactions.
468
Industrial Combustion Testing
22.5 Summary and Conclusions
References
The results of a series of studies investigating the nature of ribbon burners and their use in the flame treatment process have been presented. These results have shown the power of conventional combustion research tools to provide insight into and understanding of the flow and flame chemistry of the burner, contributing to the optimization of a commercial process relying on flame and surface chemistry. The results can serve as an initial guide to the behavior of an array of flamelets used for flame processing or process heating. The ribbon burner configuration, because of its ability to provide large flame surface and flame stabilization from the interaction of reactants and combustion products from adjacent flames, has a large range of stability as flow rate, equivalence ratio, and reactant gas composition are varied. The flame structure passes from laminar flow conditions at low flow rates to a transitional-mixing regime to a more fully mixed regime as the flow rate is increased. This behavior is similar to that of nonreacting, mixing jets. The flame treatment of polymer films to increase the wettability of a polymer surface has been described as a particularly challenging application of the ribbon burner. Optimum treatment requires a spatially homogeneous postflame reaction zone even with burners up to 3 m in length. The results presented have shown that the conditions for optimum flame treatment correspond to exposure of the polymer film surface to a high concentration of surface oxidizing species. For methane/air flames, the optimum equivalence ratio is near 0.93 where the active oxidizing species concentration near the surface is a maximum. The optimum separation distance from the flame to the polymer surface being treated is the separation distance that again gives the highest local oxidizing radical concentration. Chemical kinetic models of the impinging flame and surface oxidation chemistry of a polymer film were deduced from the existing understanding of the gasphase reactions and the polymer oxidation chemistry in plasma discharge systems. The solution of the reaction mechanisms using software available to the combustion community produced the first realistic models for the reaction chemistry. The model predictions are in good qualitative agreement with the available understanding of the flame variables affecting surface treatment and the expected oxidized species on the polymer surface. While not definitive, it serves as a basis for further understanding of polymer-flame surface chemistry.
© 2011 by Taylor and Francis Group, LLC
1. Stroud Alexander, C., Branch, M. C., Strobel, M., Ulsh, M., and Sullivan, N. “Application of Ribbon Burners to the Flame Treatment of Polypropylene Films.” Progress in Energy and Combustion Science 34 (2008): 696–713. 2. Stroud, C., Branch, M. C., Vian, T., Sullivan, N., Strobel, M., and Ulsh, M. “Characterization of the Thermal and Fluid Flow Behavior of Industrial Ribbon Burners.” Fuel 87 (2008): 2201–10. 3. Reed, R. J. North American Combustion Handbook, 414, 3rd ed., Vol. II. Cleveland, OH: North American Mfg. Co., 1995. 4. Jones, H. R. N. “Fully Aerated Burners.” In The Applications of Combustion Principles to Domestic Gas Burner Design, edited by E. Spon and F. N. Spon, 76–105. London and New York: British Gas, 1989. 5. Pearson, E., Saunders, T., and Hargreaves, K. “Fully Premixed Natural Gas Burners for Use in High Thermal Efficiency Domestic Boilers.” International Gas Research Conference (1983): 683–94. 6. Sullivan, N. P., Branch, M. C., Strobel, M., and Ulsh, M. “Transport Issues when Impinging Laminar Premixed Flames on a Rotating Cylinder.” Proceedings of the Combustion Institute 28 (2001): 1405–11. 7. Viskanta, R. “Heat Transfer to Impinging Isothermal Gas and Flame Jets.” Experimental Thermal and Fluid Science 6 (1993): 111–34. 8. Baukal, C., ed. Oxygen-Enhanced Combustion. Boca Raton, FL: CRC Press, 1998. 9. Berlad, A. L., and Potter, A. E. “Prediction of the Quenching Effect of Various Surface Geometrics.” 5th Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 728, 1955. 10. Lewis, B., and von Elbe, G. Combustion, Flames and Explosions of Gases. 3rd ed. New York: Academic Press, Inc., 1987. 11. Shaddix, C. R. “Correcting Thermocouple Measurements for Radiation Loss: A Critical Review.” Proceedings of the 33rd National Heat Transfer Conference, 1–10, 1999. 12. Branch, M. C., Sullivan, N., Ulsh, M., and Strobel, M. “Surface Modification of Polypropylene Films by Exposure to Laminar, Premixed Methane-Air Flames.” 27th Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 2807–13, 1998. 13. Strobel, M., Branch, M. C., Ulsh, M., Kapaun, R. S., Kirk, S., and Lyons, C. S. “Flame Surface Modification of Polypropylene Film.” Journal of Adhesion Science and Technology 10 (1996): 515–39. 14. Morra, M., Occhiello, E., and Garbassi, F. “Knowledge About Polymer Surfaces from Contact-Angle Measure ments.” Advances in Colloid and Interface Science 9 (1990): 79–116. 15. Chan, C. M. Polymer Surface Modification and Charac terization, 59–62. Cincinnati, OH: Hanser/Gardner, 1994.
Characterization of Ribbon Burners
16. Strobel, M., Ulsh, M., Stroud, C., and Branch, M., “The Causes of Non-Uniform Flame Treatment of Polypropy lene Film Surfaces.” Journal of Adhesion Science and Technology 20 (2006): 1493–1505. 17. Park, J., Lyons, C. S., Strobel, M., Ulsh, M., Kinsinger, M. I., and Prokosch, M. J. “Characterization of Non-Uniform Wettability on Flame-Treated Polypropylene-Film Surfaces.” Journal of Adhesion Science and Technology 17 (2003): 643–53. 18. Bowman, C. T., Hanson, R. K., Davidson, D. F., Gardiner, W. C., Jr, Lissianski, V., Smith, G. P., Golden, D. M., Frenklach, M., and Goldenberg, M. “GRI-Mech 3.0.” http://www.me.berkeley.edu/gri_mech/version30/ text30.html. 19. Koopman, R. N., and Sparrow, E. M. “Local and Average Transfer Coefficients Due to an Impinging Row of Jets.” International Journal of Heat and Mass Transfer 19 (1975): 673–83. 20. Sullivan, N., Branch, M. C., Strobel, M., Park, J., Ulsh, M., and Leys, B. “The Effects of an Impingement Surface and Quenching on the Structure of Laminar Premixed Flames.” Combustion Science and Technology 158 (2000): 115–34. 21. Kee, R. J., Rupley, F. M., Miller, J. A., Coltrin, M. E., Grcar, J. F., Meeks, E., et al. CHEMKIN Release 4.0.2, Reaction Design 2005, San Diego, CA. 22. Coltrin, M. E., Kee, R. J., Evans, G. H., Rupley, F. M., and Meeks, E. “SPIN (Version 3.83): A FORTRAN Program for Modeling One-Dimensional Rotating-Disk/StagnationFlow Chemical Vapor Deposition Reactors.” Sandia National Laboratories Report No. SAND91-8003, 1991. 23. Kee, R. J., Rupley, F. M., and Meeks, E. “CHEMKIN III: A FORTRAN Chemical Kinetics Package for the Analysis of Gas-Phase Chemical and Plasma Kinetics.” Sandia National Laboratories Report No. SAND96-8216, 1996. 24. Aghalayam, P., and Vlachos, D. G. “NOx and Fuel Emission in Combustion of Hydrogen/Air Mixtures Near Inert Surfaces.” 27th Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 1435–42, 1998. 25. Lindén, L. Å., Rabek, J. F., Kaczmarek, H., Kaminska, A., and Scoponi, M. “Photooxidative Degradation of Polymers by HO and HO2 Radicals Generated During the Photolysis of H2O2 FeCl3 and Fenton Reagants.” Coordination Chemistry Reviews 125 (1993): 195–218. 26. Gugumus, F. “Mechanisms of Photooxidation of Polyolefins.” Die Angewandte Makromolekulare Chemie 176/177 (1990): 27–42. 27. Strobel, M., Sullivan, N., Branch, M. C., Jones, V., Park, J., Ulsh, M., Strobel, J. M., and Lyons, C. S. “Gas-Phase Modeling of Impinging Flames Used for the Flame Surface Modification of Polypropylene Film.” Journal of Adhesion Science and Technology 15 (2001): 1–21. 28. Decker, C., and Jenkins, A. D. “Kinetic Approach of O2 Inhibition in Ultraviolet-Induced and Liwer-Induced Polymerizations.” Macromolecules 18 (1985): 1241–44.
© 2011 by Taylor and Francis Group, LLC
469
29. Denisov, E. T. “A Theoretical Approach to the Optimization of Antioxidant Action.” Mechanisms of Polymer Degradation and Stabilization, edited by G. Scott. London: Elsevier Applied Science, 1990. 30. Holländer, A., Klemberg-Sapieha, J. E., and Wertheimer, M. R. “Vacuum-Ultraviolet Induced Oxi dation of the Polymers Polyethylene and Polypropylene.” Journal of Polymer Science Part A-Polymer Chemistry 33 (1995): 2013–35. 31. Pachuta, S. J., and Strobel, M. “Time-of-Flight SIMS Analysis of Polypropylene Films Modified by Flame Treatments Using Isotopically Labeled Methane Fuel.” Journal of Adhesion Science and Technology 21 (2007): 795–818. 32. Strobel, M., Jones, V., Lyons, C. S., Ulsh, M., Kushner, M. J., Dorai, R., and Branch, M. C. “A Comparison of Corona-Treated and Flame-Treated Polypropylene Films.” Plasmas and Polymers 8 (2003): 61–95. 33. Dorai, R., and Kushner, M. J. “A Model for Plasma Modification of Polypropylene Using Atmosphere Pressure Discharges.” Journal of Physics D: Applied Physics 36 (2003): 666–85. 34. Stroud, C., and Branch, M. C. “Modeling of the Surface Oxidation of Flame Treated Polypropylene Film.” Combustion Science and Technology 179 (2007): 2091–2105. 35. Grace, J. M., Chen, J., Gerenser, L. J., and Glocker, D. A. “Use of glow discharge treatment to promote adhesion of aqueous coats to substrate.” US Patent No. 5,425,980, 1995 and “Use of glow discharge treatment to promote adhesion of aqueous coatings to substrate.” 5,582,921, 1996. 36. Gerenser, L. J. “An X-Ray Photoelectron Spectroscopy Study of Chemical Interactions at Silver/PlasmaModified Polyethylene Surfaces: Correlations with Adhesion.” Journal of Vacuum Science & Technology A-Vacuum Surfaces and Films 6 (1988): 2897–903. 37. Liston, E. M., Martinu, L., and Wertheimer, M. R. “Plasma Surface Modification of Polymers for Improved Adhesion: A Critical Review.” In Plasma Surface Modification of Polymers: Relevance to Adhesion, 14–17, edited by M. Strobel, C. S. Lyons, and K. L. Mittal. Utrecht: VSP, 1994. 38. Chan, C. M. Polymer Surface Modification and Characteri zation, 235–53. Cincinnati, OH: Hanser/Gardner, 1994. 39. Strobel, M., Branch, M. C., Kapaun, R. S., Lyons, C. S. “Flame-treating process.” US Patent No. 5,753,754; 1998. 40. Sullivan, N., Branch, M. C., Strobel, M., Park, J., Ulsh, M., and Leys, B. “The Effects of an Impingement Surface and Quenching on the Structure of Laminar Premixed Flames.” Combustion Science and Technology 158 (2000): 115–34. 41. Strobel, M., Sullivan, N., Branch, M. C., Park, J., Ulsh, M., Kapaun, R. S., and Leys, B. “Surface Modification of Polypropylene Film Using N2O-Containing Flames.” Journal of Adhesion Science and Technology 14 (2000): 1243–64.
23 Flameless Burners Joachim G. Wünning and Ambrogio Milani Contents 23.1 Introduction.................................................................................................................................................................. 471 23.2 Flameless Oxidation.................................................................................................................................................... 472 23.3 Principle of Flameless Oxidation............................................................................................................................... 473 23.4 Applying Flameless Oxidation to Burner Design................................................................................................... 475 23.5 Test Furnace Investigations........................................................................................................................................ 476 23.5.1 Test Rig and Measuring Equipment........................................................................................................... 476 23.5.2 CFD Computations........................................................................................................................................ 477 23.5.3 Experimental Results.................................................................................................................................... 478 23.5.4 Confined Flow................................................................................................................................................ 480 23.6 Advanced Diagnostic Investigations........................................................................................................................ 481 23.7 Industrial Validation................................................................................................................................................... 481 23.7.1 The Self-Regenerative Burner....................................................................................................................... 482 23.7.2 Recuperative Glass Smelter.......................................................................................................................... 484 23.7.3 Reformers........................................................................................................................................................ 484 23.8 Control and Safety Standards.................................................................................................................................... 485 23.9 Conclusions................................................................................................................................................................... 485 23.9.1 Application to Heat Processing.................................................................................................................... 485 23.9.2 General Features of Flameless Oxidation.................................................................................................. 486 References................................................................................................................................................................................. 486
23.1 Introduction Not every chemical oxidation reaction is considered to be combustion and not every combustion is accompanied by flames. A classification of different redox-reactions is given in Figure 23.1. Relatively slow redox-reactions with low intensity go on almost everywhere. Examples are rusting of steel, browning of a fresh cut apple, but also various reactions in the earth’s atmosphere. Combustion can happen naturally like a forest fire that could be sparked by a lightning strike. Controlling combustion was one of the fundamental skills of mankind that allowed civilization. Most combustion processes involve flames that enable controlling combustion by visual examination. There are also some combustion processes that are not accompanied by flames. Catalytic surfaces can lower the activation energy for reactions and therefore enable reactions at lower temperatures without formation of flames. A campfire will first burn with blazing flames. Later on the flames will extinguish but reactions will still occur in the fire bed until all char
is burnt to ash. Flameless combustion is therefore combustion in absence of flame [1]. Flameless combustion techniques, in conjunction with regenerative high temperature air preheating, have been applied to steel reheating furnaces, in particular by Japanese and European companies, and are now proposed by equipment manufacturers for heat processing plants in the metal industries (steel, aluminum, etc.) for effective energy saving. Basically, flameless firing allows low NOx (or even very low NOx) with high air preheating, which is possible with regenerative burners [1–5]. As a form of flameless combustion, flameless oxidation or, shortly, FLOX®, is discussed in more detail in the next paragraphs. It has been mainly applied to the design of gas burners in steel heat treatment furnaces; several thousands of such burners have been firing satisfactorily in the last 10–15 years [5–8]. The community of combustion science and technology agrees that flameless combustion has been the most important step forward in the last two decades. In this connection, the present chapter reports the story of developing flameless oxidation technologies from traditional testing until 471
© 2011 by Taylor and Francis Group, LLC
472
Industrial Combustion Testing
more sophisticated, laser-optic diagnostic techniques: the latter are nonintrusive and have time resolution of milliseconds or even microseconds, which closely follows turbulent eddies. This allows much more detailed insight, which is required for combustor applications that are not traditional burners in furnaces [1,9,10]. As a matter of fact, the industrial achievements in steel furnaces have stimulated developments in other domains, including nonconventional power generating equipment, even for high pressure and high power densitylike gas turbines. These are shortly mentioned in the present chapter. Industrial heat processing operating conditions are also briefly covered, because test furnaces cannot reproduce all the boundary conditions influencing the burner performance. In the concluding remarks, features of flameless oxidation and perspectives for widespread use in several applications are summarized. Redox - reaction
Slow reaction
Combustion
Example: - Rusting of steel - Browning of fruit
Explosion - Deflagration - Detonation
Flame
Flameless
Example: - Premixed - Diffusion - Laminar - Turbulent
Example: - Catalytic - Smoldering - Flameless oxidation
Figure 23.1 Redox reactions.
23.2 Flameless Oxidation The flameless oxidation was discovered in 1989 during trials aimed at NO reduction in combustion with highly preheated air [5,6]. No flame could be detected or heard in the combustion chamber and the NOx abatement was above all expectations. Figure 23.2 shows pictures of these trials. On the left-hand side, the burner is shown in flame mode: the stable high velocity flame is pale blue and the high temperature at the burner nozzle causes a local bright luminosity. On the right-hand side, the same burner is shown in flameless oxidation mode. The hot zone at the burner nozzle has disappeared and no flame is visible or any combustion roar is audible. During the following years, the phenomena of the flameless oxidation has been investigated from a basic point of view, but also practical solutions have been developed for transferring the principle into practice. As stated in the introduction, there are many ways of flameless combustion: the flameless oxidation can be defined as “stable combustion without flame with a defined recirculation rate of hot combustion products.” With experiments, but also with computer simulations, it could be demonstrated that in flameless oxidation not only the flame is not visible, but also other typical flame features are not present [1,11]. Flameless oxidation can be used in many circum stances: • • • •
for gaseous, liquid, and solid fuels with or without air preheating with or without fuel preheating for lean, stoichiometric, or rich combustion (e.g., λ ≈ 0.5–3) • for diffusion, partially premixed, or premixed combustion
Flame Fuel
Air
Flameless oxidation (FLOX®) Figure 23.2 Flame and flameless oxidation.
© 2011 by Taylor and Francis Group, LLC
473
Flameless Burners
Some definitions in the literature can be misleading because they are often referred to as hot preheated air, which is in fact the most widely used technical application until now. Therefore it should be noted here that with flameless oxidation: • the combustion air needs not be preheated • the oxygen content in the combustion chamber must not be very low • no catalyst is required • no heat sinks must necessarily be provided in the combustion chamber (adiabatic combustion) • the combustion chamber needs not be large • fuel and air need not be separately injected into the combustion chamber
23.3 Principle of Flameless Oxidation In contrast to the combustion in stabilized flames, flameless oxidation is mixture and temperature controlled and is achieved by specific flow and temperature conditions. A prerequisite for a stable flame front is a balance between flow and flame velocity. This is true in premixed and in diffusion flames and stability depends on species concentrations, flow velocity, flow field, temperature, pressure, and other parameters. Creating flow conditions for flame stabilization is an essential burner design criterion. Swirl or bluff body are most often used to create stagnation points or areas of low velocity for stabilization. The species concentration also plays an important role. Air, with an oxygen content of 21% can create a flammable mixture with
Methane (%)
) (% ne tha Me
0
100
%)
%)
rt (
Figure 23.3 Triangle diagram of combustible mixtures.
20 Oxygen (%)
Oxygen (%)
rt (
Ine
Technical combustion
0 100
Co m m bu ixt st ur ibl es e
Ine
0.5
0
© 2011 by Taylor and Francis Group, LLC
100 0
Kv =
0
λ>1
(23.1)
where M is the mass flow rate and the suffix fg stands for entrained flue gases. A schematic comparison between the combustion rate control modes in flame and in flameless oxidation is reported in Figure 23.4. For flameless oxidation, it is essential that mixing of exhausted recirculated flue gases with fresh reactants takes place before combustion. To the opposite, support recirculation of nondiluted, chemically active species is required for keeping a steady flame front. This latter mechanism is called (burner) internal recirculation, while the former is referred to as external recirculation; it is external to the burner and within the furnace (most common), but it may also be external to the combustion chamber. The basic condition is that external recirculation
80
ible ust s e mb Co ixtur m
10
1
Kv =
λ<1 λ=
kv = Mfg/(Mair + Mgas),
as ust g Exha lation u c recir
en Oxyg ent m h enric
20
many fuels. This is the reason why flames could be controlled as a tool by the human race thousands of years ago. Exhaust gas recirculation increases the content of inerts of a mixture. Flammability limits for combustion of hydrocarbons and air show that it is possible to achieve flammable mixtures for recirculation rates of kv < ~ 0.5 (Figure 23.3) [12]. To provide reliable operating conditions in practical systems, maximum exhaust gas recirculation rates of kv ~ 0.3 can be used as NOx reducing technique. The fact of the matter is that flue gas recirculation is an efficient means to reduce thermal nitric oxide formation: by entraining and mixing inert flue gases into the combustion air and into the fuel, the combustion temperature is reduced because of the large heat capacity involved (inertization). In technical applications, flue gas recirculation up to ≈ 30% are indeed used, but larger values produce unstable flames and quenching. The flue gas recirculation ratio kv is defined as follows:
0
100
474
Industrial Combustion Testing
Flame front stabilization
(a)
Heat to CC
Fuel species Oxygen
Mixing
Flame front
Far field
Near field
Inert
Support recirculation (b) Fuel species Oxygen Inert + exhaust
Exhaust recirculation
Flameless oxidation
Flame front or lifted flames
Mixing of reactants + exhaust recirculation
Heat to CC Near field (distributed reactions)
Far field
recirculation Exhaust recirculation Figure 23.4 Combustion control modes.
FLOX®
Unstable
on Stable combusti
Unstable
Temperature
Stable
FLOX®
No reaction
Internal exhaust gas recirculation Kv
Figure 23.5 Combustion regimes.
entrains inert gases at sufficiently high temperatures to bring the mixture above self-ignition, but at the same time prevents the onset and propagation of a deflagration flame front. It has been found that under defined conditions at high flue gas recirculation rates, a stable combustion regime can be onset. Figure 23.5 shows a scheme of different combustion modes. Burners with burner stabilized flames can be safely operated in the whole temperature range and with flue gas recirculation rates up to circa 30%. With higher recirculation rates the combustion becomes
© 2011 by Taylor and Francis Group, LLC
unstable and is eventually extinguished for low temperatures below self-ignition temperature of the fuel. With lifted and unstable flames the nitric oxide can be reduced, but this regime is not suitable for safe use and for a safety supervision in industrial applications. However, for high temperatures above the self-ignition temperature and for larger recirculation rates, a state of stable combustion can be reached. In such conditions, the combustion takes place without visible or audible flame and the formation of thermal nitrogen oxides is very effectively suppressed. This is called flameless oxidation or, shortly, FLOX®.
475
Flameless Burners
Loop reactor
Plug flow Kv = 0
Loop reactor
Free flow
Well-stirred reactor Kv → ∞
Figure 23.6 Loop reactors.
process; this should clarify the effect of recirculation rate on the maximum combustion temperature. In step I combustion air is mixed with recirculated flue gases. After thorough mixing fuel is injected in step II. The maximum temperature increase, as shown before, depends mainly on the amount of the recirculated flue gases and does not depend in this idealized model on the reaction kinetics. In step III heat is extracted from the process. Eventually part of the flue gas is recirculated and the rest goes out from the process.
Combustion air Process heat III Exhaust
Fuel
I
II
Figure 23.7 Idealized process of flameless oxidation.
The influence of the recirculation rate on the temperature peaks in the combustion can be approximately estimated as follows:
Θmax ≈ Θ0 + Θad/(kv + 1).
(23.2)
In this formula, Θ0 is the temperature of the mixture of the fresh reactants mixed with the recirculated flue gases. For high recirculation rates, Θ0 is close to the flue gases temperature. The adiabatic temperature Θad for rich fuels as natural gas is about 2000°C. This simple estimate makes it clear that the temperature increase ∆Θ = Θmax − Θ0 during combustion is drastically reduced with large recirculation rates. By avoiding temperature peaks, the thermal nitric oxide formation is effectively suppressed, even if the combustion air is preheated: this is the basic trick for low NOx performance. To describe patterns of combustion processes, ideal models are often introduced such as piston flow or a perfectly stirred reactor as well as an ideal stirred boiler. For describing the flameless oxidation the model of the loop reactor is appropriate. Figure 23.6 shows different combinations of loop reactors. Here the piston flow (kv = 0) and the well-stirred reactor (kv = ∞) can be considered as limiting cases of loop reactors. Based on a model of a reactor loop for better understanding, Figure 23.7 shows an ideal flameless oxidation
© 2011 by Taylor and Francis Group, LLC
23.4 Applying Flameless Oxidation to Burner Design Figure 23.8 shows how these ideal process steps can be embodied into technical equipment. The burner can be fired both in flame as well as in flameless oxidationmode (short FLOX® mode). The fuel is introduced via the connection “1” and the burner nozzle “4” into the central combustion chamber “5.” During startup heating time, the burner fires with a stable flame and the air is introduced into the combustion chamber “5” via the air connection “4.” The mixture in the chamber is spark ignited and flows out at high velocity into the furnace chamber. When the furnace chamber has been heated above the threshold required by flameless oxidation, the “start air” can be shut off and the air is now introduced via the air connection “3” and then injected coaxially into the furnace chamber via nozzle “6.” The air jet “A” entrains combustion products or flue gases from the surrounding room “D” and the fuel jet “B” does the same. Only after meeting of the jets in the plane “E” can a reaction fuel/ air take place in the region “C.” The geometrical size and ratios determine the recirculation rate. When the kinetics of the combustion reactions are taken into consideration, there are also other design shapes and operating ranges for the FLOX® burners. Even undesired spontaneous reactions between air and fuel can be suppressed, inasmuch as temperature in the mixing zone is kept low or flow velocities are well above flame
476
Industrial Combustion Testing
2
1
Features: • 3-D fully automatic measurement • Uncooled vertical probe for gas, pressure and temperature (PtRh–Pt 50 µm) • Air cooling for temperature control independant from the capacity • Symmetric conditions • Adjustable air preheat • Vertical arrangement to minimize buoyancy effects • Controlled boundary conditions and • Practical relevance
D 6
A
4
C
B
5 3
6
5
A D
Figure 23.8 FLOX® burner. 1-Nozzle burner
6-Nozzle burner
Figure 23.10 Test furnace.
Flame
1600 ºC 1400
Lifted flame, xNO 60 ppm, 92 dB(A)
Flameless oxidation Figure 23.9 One nozzle and multinozzle FLOX® burner.
speed. From the point of view of mathematical modeling, the assumption “mixed equal burned” must be relaxed and the chemical kinetics of the fuel conversion must be taken into account. Trials have demonstrated that injection via nonseparate nozzles or even injection of a premixed fuel-air in the combustion chamber, under defined conditions, can produce a flameless oxidation. Figure 23.9 shows two further burner designs that can be switched over from flame to flameless firing (flame-FLOX®) and vice versa. For flameless oxidation in one-nozzle burners, fuel and air are injected concentrically into the combustion chamber. In the 6-nozzle burner, six air jets are arranged around the central fuel gas jet. In both cases, gas injection only is switched for flame firing.
23.5 Test Furnace Investigations A small dedicated test furnace with a burner capable of firing in both flame and flameless mode was envisaged and put up in order to verify the concept described above and to validate model predictions. Measurements confirmed the above picture and produced NOx emissions values, which have been subsequently confirmed by feedback from industrial plants. Further experimental devices have been designed and used for confined flows depending on the purpose of the combustion equipment. This is also briefly covered in the following chapter.
© 2011 by Taylor and Francis Group, LLC
Temperature
1200
Flame, xNO 160 ppm, 98 dB(A)
1000
FLOX, xNO 6 ppm, 82 dB(A)
800 600 400 200 0
0
2.5
5 Time
7.5
s
10
Figure 23.11 Temperature recordings.
23.5.1 Test Rig and Measuring Equipment Experimental measurements have been carried out in a test furnace built on purpose (Figure 23.10). A recuperative burner, with a burner integrated heat recovery to bring about an efficient air preheating by means of countercurrent flue gas extraction, was installed at the bottom and fired vertically upward. Air cooling was provided with inserted pipes. The measurements were carried out from above by means of intrusive probes; positioning was computer controlled. For the temperature a very thin thermocouple (0.05 mm) was employed (Figure 23.11), avoiding, however, hot spots above ≈ 1500° in the flame that would destroy the thermocouple. The axial velocity was determined with a Pitot tube. Comparative tests could be carried out as the burner could fire with a burner stabilized flame as well as in flameless mode. In flame mode, the onenozzle burner worked like a usual high velocity burner. In this case the fuel gas was injected radially into the ceramic precombustion chamber to mix with preheated combustion air (Figure 23.9) and the flow pattern was such that a stable high velocity flame could be onset and then supervised with ionization or UV detectors. The flameless operation was carried out by switching the
477
Flameless Burners
1200 800 400
0.2
0
0 Flame 0.15 0.10 m 0.05 ey 0 distanc Radial
2000 °C 1600 1200 800 400 1.0 m 0.8
0.6 Axial dis 0.4 0.2 tance z
0
1800°C
0.0
1600°C ϑmax = 1840°C
ϑfurnace = 1200°C
1800°C 2000°C
1600°C
0 FLOX® m 0.15 0.10 0.05 tance y Radial dis
gas injection in a direction parallel to the high velocity combustion air jet. Typically, the temperature pattern in the burner near field is affected as shown in Figure 23.12: the upper half (flame mode) shows measured temperatures that clearly increase toward the burner nozzle (y → 0, z → 0), while the lower half (flameless mode) shows a profile increasing toward the combustion chamber temperature without forming temperature peaks. This confirms the visual observation of Figure 23.2 and has been checked with mathematical modeling, which confirms the overall outline as described in the next paragraph. 23.5.2 CFD Computations A mathematical modeling has been carried out with a commercial CFD code (Computational Fluid Dynamics), by using the following model assumptions: turbulence: k-ε model combustion: one step with Arrhenius reaction radiation: flux model nitric oxide formation: one-step Arrhenius density: ideal gas heat capacity: depending on temperature and species • viscosity: temperature dependent
© 2011 by Taylor and Francis Group, LLC
1400°C
ϑmax = 2120°C
ϑfurnace = 1200°C
®
FLOX , 200 kW, 950°C air preheat 1600°C
1400°C
ϑmax = 1745°C
Figure 23.12 Measured temperature pattern.
• • • • • •
1400°C
Flame, 245 kW, 600°C air preheat
Temperature
1.0 m 0.8 0.6 Axial dista 0.4 nce z
Flame, 400 kW, no air preheat
Temperature
2000 °C 1600
ϑfurnace = 1200°C
Figure 23.13 Computed temperature field.
A comparison of the combustion temperatures in flame- and in FLOX® -mode should clearly show the reduction of the temperature peaks. Figure 23.13 shows, in the upper diagram, the temperature field of a 400 kW burner in a furnace at 1200°C fired with cold air in flame mode. In the second diagram, the same burner is operated with air at 600°C, but the fuel input is reduced to 245 kW to take into account the preheating and thereby keeping the net heat input constant. The temperature field of a 200 kW FLOX® burner using preheated air at 950°C is represented in the bottom diagram. The cold air burner exhibits the typical pattern of cold air flames with expected peaking temperatures about 2120°C. In the case of the FLOX® burners, the temperature field is substantially modified and the highest temperature is no longer close to the burner nozzle. In spite of the high air preheat the peak temperature is 1745°C, well below the case of the cold air burner. The temperature peaks are the basic reason for the thermal nitrogen oxide formation and this explains why, with FLOX®-burners, the nitrogen oxide emissions can be much lower than with cold air burners, in spite of higher air preheat.
478
°C Temperature
1500
80
1000
2000 °C 1500
/s
m
Temperature
2000
20 m/s 4 60 0 m/s m/ s
Industrial Combustion Testing
/s
100 m
Air flow velocity variation
500 0 0.0
0.5 1.0 Distance from burner
C
0°
80
1000
°C C 400°
600
200°C
Air preheat temperature variation
500
m
1.5
0 0.0
0.2
0.6 0.4 Distance from burner
0.8
m 1.0
2000 °C Temperature
1500
2%
1000
4% 0%
500 0 0.0
Boundary conditions
8%
Fuel: Capacity PA: Furnace temperature ϑOten: Air preheat temperature ϑLuvo: Air ratio λ: Air flow velocity WL:
x∞ = 16% (λ = 4.5) Air ratio variation
0.5 1.0 Distance from burner
m
natural gas 25 kW 1000°C 0.65 ϑOten 1.15 100 m/s
1.5
Figure 23.14 CFD parameter variations.
Figure 23.14 shows the results of a parametric evaluation of the axial temperature profile assuming constant furnace and flue gas temperature = 1000°C, varying: • air inlet velocity: this shows that a high momentum of the input air jet is required in order to entrain a large amount of recirculated flue gases (i.e., a high kv ratio) • air preheating: this affects the axial temperature profile and the onset of combustion reactions above self-ignition • air ratio: its influence on the axial temperature profile is labeled with the residual oxygen content; excess air can either enhance reaction rates or dilute reactants for very lean combustion The measured temperatures are in reasonable agreement with computed values. The measurements showed that the temperatures within the whole furnace chamber are almost uniform apart from the near burner region. As a conclusion, the initial fundamental tests with the experimental furnace described in Section 23.5.1, backed up by the modeling computations, have demonstrated that the fundamental picture of flameless oxidation as a well-defined flameless firing technique is basically correct.
© 2011 by Taylor and Francis Group, LLC
23.5.3 Experimental Results The test furnace (Figure 23.10) has been used not only for the single nozzle burner described above, but also for a six-nozzle burner (refer to Figure 23.9). Measurements with the six-nozzle burner showed very similar characteristics as above. Here the fuel in flame mode was stabilized upstream of the nozzle plane and six small flames were injected into the combustion chamber. In flameless mode the fuel gas is injected axially and the combustion air is injected in axial direction via the six nozzles. By adjustment of defined parameters, this burner could also be operated with lifted unstable flames. Some results of the operating range are schematized in Figure 23.5: • stable flame combustion • unstable (lifted flames) • flameless oxidation (FLOX®) Recording of temperature profiles reported in Figure 23.11 have been measured at a distance of 250 mm from the burner nozzle. This position is downstream of the high velocity flame and inside the lifted flames. The relevant NO values have been measured in the flue gas channel. In flame mode, the temperature signal is only a little affected by time oscillations, whereas the lifted flame clearly showed large fluctuations in the frequency range
479
Flameless Burners
of a few Hertz. The measured amplitude was several hundreds degrees Kelvin. In flameless oxidation, steady temperatures were measured again and found somewhat less than the furnace temperature. Noise measurements are also characteristic for the combustion pattern. Without reaction (i.e., injecting air only), the basic noise was 78 dB (A), while in flame firing almost 100 dB(A) were reached. In the lifted flame region the noise is reduced to 92 dB (A) and is shifted toward lower frequencies. In flameless operation the noise level is about 82 dB(A) just somewhat higher than the background. Drastic differences in the nitric oxide emissions are also quite clear. In the present case, the values of about 160 ppm in flame mode are reduced down to some 10 ppm. The values for lifted flames are between and are subject to strong fluctuations. Figure 23.15 shows the potential of NOx reduction of the flameless oxidation. The values are in a logarithmic scale as the emissions are very much temperature dependent. As reference values, the emissions from air-staged high velocity burners are reported. The NOx values from burners without NOx countermeasures and with preheated air are clearly much higher. The obtainable NOx emissions depend not only on the burner, but are also affected by the particular application. For instance, in a small radiant tube it is clearly more difficult to obtain an efficient recirculation of flue gases as required by flameless oxidation (Figure 23.17). The lowest NOx values down to 1- digit ppm could be reached with burners, for which the combustion chamber had been optimized on purpose for flameless oxidation (combustor). In applications of direct firing in industrial furnaces with air preheating in the range Air preheat ϑL2 (ε = 0.65) 1000
0
200
600 °C
400
800
ppm TA–Luft
A
rner
d bu
age ir st
nt
a adi
R
e
tub
Preheated reactants
fir in
g
10
ire
ct
t = tar
D
NOx (2% O2)
100
600°C of 900°C values, about 50 ppm are typical. With indirect heating with radiant tubes, values in the range of 100 ppm are encountered depending on the particular installation [13,14]. The boundary condition in heating furnaces are such that the air and the fuel jet are injected an a virtually infinite surrounding at a temperature above (or well above) self-ignition at constant pressure. Provided enough momentum is imparted to these jets, turbulent mixing will continue until micromixing at molecular level and reactions will take place until completion (no CO or unburned traces). Avoiding the onset of a flame front is a relatively easy task. The burner design will be straightforward and there is also the possibility of taking into account other requirements beside NOx. The circumstance that combustion stability in flameless oxidation is carried out by entraining recirculated flue gases above self-ignition temperature (or better, above a threshold of 850°C for safety reasons) does not imply that the whole process must be above this threshold [15–17]. The limitation concerns the control volume limits (Figure 23.4) of the combustion chamber, which may be itself inserted into a more comprehensive process scheme. A typical example is postcombustion of gaseous HC solvents in process air, stripped off from drying, painting, printing, and similar processes. The mixture HC-vapors/air is much too lean to produce an adiabatic temperature not even close to 850°C; however, it may be preheated above this threshold and then cooled down again by means of efficient heat exchangers (Figure 23.16). In fact, flameless, volume distributed reactions without flame fronts take place in postcombustion chambers. In these cases, testing is not limited to the burner, but includes the heat recovery equipment (air and lean gases or dispersed vapors). In this way, the difficulty due to low adiabatic temperatures and to very lean mixtures, well below the lean flammability limit, can be overcome without resorting to considerable addition of auxiliary fuel burning in turbulent flames. Catalytic techniques are at lower temperatures, but are expensive and do not accept dirty or contaminated fuels
Com
Flame 0.1
tor
bus
1
0
400
Flameless oxidation 800
1200 °C
Furnace temperature ϑfurnace Figure 23.15 NOx emissions.
© 2011 by Taylor and Francis Group, LLC
Combustion chamber t = tpr 850°C
t = tpr
Countercurrent heat exchanger
1600 Auxiliary fuel
Figure 23.16 Heat recirculation.
Polluted air stream
t = tf
480
Industrial Combustion Testing
(for instance raw biogas). Therefore flameless combustion can also be considered as a valuable technology to burn lean fuels or in general off-specification fuels. 23.5.4 Confined Flow In all preceding examples, the burner mainly consists of separate nozzles issuing into a relatively large, constant pressure combustion chamber that can be schematized as an infinite source of inert hot flue gases above the threshold temperature of 850°C. In such boundary conditions, computational modeling is relatively straightforward (see Section 23.5.2), and a flameless oxidation burner can be designed and tested almost independently of the final use, by just knowing throughput and size. Several thousand high velocity FLOX® burners (especially recuperative natural gas burners with efficient, built-in air preheating) are in fact satisfactorily firing in free flame or in radiant tubes in heat treatment and or heating furnaces [14,18,19]. In other circumstances, however, air and fuel jets from the burner nozzles cannot follow the basic formulae of the free jet and recirculation of flue gases is not unconditional.
Radiant tube Flame tube
mEG
mEG
mA mF mA
Figure 23.17 Single end radiant tube. (After Galletti, C., Parente, A., and Tognotti, L., Combustion and Flame, 151 (4), 649–64, 2007.)
© 2011 by Taylor and Francis Group, LLC
An example concerns a single end radiant tube (Figure 23.17). Radiant tubes are described in a separate chapter of the present book, but this particular case is reported here: the size of the equipment is small and the reduced Reynolds number enhances the effect of flow boundary layers along the walls, the amount of recirculated flue gases entrained along the annular gap by the jet flame is limited and difficult to assess, the heat transfer is complicated [20]. Then flameless oxidation firing is not as clearcut from flame firing; this picture is also supported by mathematical modeling that has been developed for predicting NOx emissions, in particular if the fuel gas contains a large amount of hydrogen. Hydrogen is a special fuel gas because of its very high flame speed that may cause difficulties in preventing a flame front and hence a flameless oxidation regime [20]. This list of designer’s headaches to be taken into account has been quoted to show that a confined flow may be much more difficult to tackle: in such cases, experimental testing must be envisaged in order to include the main features of interest. A further practical example of highly confined combustion is the nozzle field, a rapid heating device based on impinging jets (Figure 23.18): for rapid heating of cold steel strips (to be annealed in a continuous furnace), conventional impinging flames are not admissible because they spoil the surface quality. But FLOX® jets are an attractive solution to avoid hot spots, thanks to the effective suppression of temperature peaks. In this case, recirculation is not well separated from flue gases and the impinging jet is still reacting; there will be competition in reaction rate between chemical kinetic time and convection time depending on the spectrum of turbulent eddies, there will be stability issues and unknown heat transfer between gas phase and solid surfaces. Such a complex system cannot plainly be tested studying individual components or an isolated burner. A full-scale section of the equipment has been built (simulating the upper half only) and tested in the laboratory (Figure 23.18) before building a prototype unit suitable for industrial service [21]. Results have been good; the relevant consideration here is that testing in these cases requires full-scale pilot plant equipment that reproduces at least the basic process steps. In other words, testing cannot be limited to the burner, but must be extended to the whole combustion chamber and simulated in scale 1:1. The most challenging design has concerned a combustor for gas turbines [9,22]. In gas turbines, the adiabatic temperature is moderated by large excess air and the combustion intensity is in order of magnitudes higher than in furnaces, because of size constraints and, of course, because of pressure. Flameless oxidation has been sometimes described as volume combustion in contrast to surface combustion that hints to a turbulent flame front, where volumetric reaction rate is certainly much higher than in the former case. The basic question was then: will a flameless oxidation system cope with the required
481
Flameless Burners
Self-recuperative burners Auxiliary burner
Strip
Water-cooled plate
FLOX– nozzles
Figure 23.18 Test furnace for the nozzle field.
combustion intensity and will such an adiabatic, high residual oxygen content system onset stable combustion? First, simple, room pressure tests of prototype combustor designs have been very encouraging and good luck has accompanied subsequent developments. Further developments of the project have included tests under pressure and quite sophisticated optical inspection techniques, not to mention development of mathematical modeling to be validated by tests and advanced diagnostics [22].
(Figures 23.19 and 23.20) and provide a pictorial overall description [23,24]. Furthermore, laser shots into the reaction zone are so short that even the smallest eddies can be resolved. These instruments supply a large quantity of information that can, or possibly must, be subsequently handled by mathematical modeling (validation).
23.7 Industrial Validation 23.6 Advanced Diagnostic Investigations Experimental testing under pressure of a test combustor, even if simplified and reduced, constitutes a difficult and expensive problem (safety, inspection, size, fluid supply, etc.), and it can be carried out by specialized organizations only. Industrial prototypes of GT combustors are tested in scale 1:1 during short (very expensive) trials; updated, laser-based, diagnostic optical techniques are employed to squeeze as much experimental information as possible to be subsequently supplied to mathematical modelers for quantitative data processing. A detailed insight into the GT sector is out of the scope of the present book and it has been quoted only to remind researchers that testing may become extremely expensive and complicated. Furthermore, Chapters 12 and 13 of this book are devoted to advanced diagnostic techniques that have almost superseded traditional intrusive techniques (water-cooled probes inserted into the combustion chamber for measuring local temperature, gas composition, etc.). Also flameless oxidation is better studied by nonintrusive techniques as the insertion of probes disturbs the local conditions and may cause the onset of a local flame front. Fluorescence (like LIF techniques) as described in other chapters, or even simpler, a UV picture of the reaction zone are very suitable to describe flameless oxidation
© 2011 by Taylor and Francis Group, LLC
Testing an innovative burner design or a flameless oxidation burner on a test furnace only without reproducing the process is not enough to highlight all relevant issues and for predicting the field performance of the combustion system on short- and long-term. Sometimes the flameless oxidation concept is certainly suitable for the process (see the Glasflox® example below), sometimes the flameless oxidation operating range must be adapted to the process rather than to the burner (see the nozzle field example in Section 23.4). The NOx emissions are usually measured in well-controlled test furnaces, but field firing may significantly alter values taken at the stack of operating plants (e.g., because of the load interactions with the flame, in leakages, etc.). Therefore, one has to be careful about guaranteeing NOx emissions as being a property of the burner design only. Demonstration campaigns are a known practice in industrial heat processing and sometimes are a prerequisite for adopting a new technology. In this respect there are no fixed procedures for testing flameless burners because they depend on the process itself, or at least on a significant part of it. Therefore a few examples only, deemed as significant, will be quoted in the following paragraphs. The most widespread burners adopting low NOx techniques based on flame dilution are regenerative burners for reheating and forging steel semiproducts (slabs, billets, ingots, etc.) or melting aluminum; they are large units
482
Industrial Combustion Testing
Highly turbulent premixed methane flame (Φ = 1) 2100
3200 2800
(mm) 6.0
2400
1500
2000
1200
Height =13.0 mm
900
0.0
3.0
6.0 9.0 (mm)
300
12.0
1200
6.0
800
600 0.0
1600
12.0 (mm)
Temperature (K)
1800 12.0
18.0
OH–LIPF (a.U.)
18.0
400 0.0
2.5
0.0
0 5.0 7.5 10.0 12.5 (mm)
0.0
3.0
6.0 9.0 (mm)
12.0
0.0
3.0
6.0 9.0 (mm)
12.0
Flameless oxidation (Φ = 1)
Temperature (K)
1800
(mm)
12.0
6.0
2000 1600
1200
1200
900
300
0.0
3.0
6.0 9.0 (mm)
12.0
600
900 1200 1500 1800 2100 Temperature (K)
300
18.0
2800 2400
1500
12.0
6.0
800
600 0.0
3200
(mm)
Height = 15.13 mm
OH–LIPF (a.U.)
2100
18.0
400 0.0
2.5
0.0
0 5.0 7.5 10.0 12.5 (mm) 0
800 1600 2400 OH–LIPF Intensity (a.U.)
3200
Figure 23.19 Comparison flame and flameless oxidation.
(order of 0.5–5 MWth), connected in couples, and must be robust enough to withstand rough service [1–3]. The effect of regenerative bed clogging is plainly a priority matter and in general the burners are so big that they are an essential part of the overall plant. Chapter 21 of this book is devoted to these regenerative burners and only the particular case of the Regemat® will be described here.
50 kW flame mode
23.7.1 The Self-Regenerative Burner 50 kW FLOX mode 0
200
400 (mm)
600
800
Figure 23.20 Laser induced fluorescence for OH radicals. (After Scherello, A., Konold, U., Schwarz, G., and Koster, B., Tagungsband zur Glastechnischen Tagung, Dresden, 2006.)
© 2011 by Taylor and Francis Group, LLC
The Regemat® is a relatively small capacity burner (200 kWth) and has been designed to be self-regenerative, meaning that both honeycomb cartridges are embedded into the same burner casing, building a compact unit that may replace a conventional burner, instead of using two separate burners for building the couple of heat capacities as usual regenerative burners do. The scheme and the picture in Figures 23.21 and 23.22 show the
483
Flameless Burners
Pneumatic switching valve Exhaust
Orifice
Furnace wall
Regenerator Combustion air
FLOX Flame Fuel
Flame air
Figure 23.21 Scheme of the Regemat® burner.
stations under control effectively and at a low cost. This capability should be taken into serious consideration when planning industrial trials. Clearly, problems like clogging of heat transfer surfaces, trouble free operation of inversion valves, integrity of high temperature components and of onboard electronic devices can only be tackled with demo projects in scale 1:1 and looking at the burner performance within the framework of the industrial process. This has been done as a first step with the first series of Regemat®, by equipping a furnace zone and then revamping the complete furnace as a second step [4]. The first test series demonstrated that clogging was not a problem and that general performance were basically as expected. Subsequent to this feasibility test, the whole annealing furnace for stainless steels strips was revamped and previous feedback was then used for improving design details and maintenance routines. This is a lucky case of burner testing procedure from laboratory scale until industrial acceptance and eventually to market competition. Further conclusions drawn from field testing concern furnace control [1]:
Figure 23.22 Picture inside the furnace.
arrangement of the air/flue gas channels, connected in parallel three by three, to form the heat capacity couple. Switching time to invert flow in the channels is 10 seconds and this switching is carried out with metal valves operated by compressed air. Switching valves are built into the casing and must be reliable, because seizure or leakage might involve destruction of the device in case of unbalanced suction of hot flue gases. For this reason, the Regemat® is equipped with safety devices to monitor regular operation. Remote monitoring is currently made possible by electronic field bus connections (for instance, profibus) that are able to keep many remote
© 2011 by Taylor and Francis Group, LLC
• Burner integrated heat recovery implies that flue gases are basically extracted from the burner; however, a fraction of flue gases, possibly 10–20%, must be left in the combustion chamber and exhausted through the stack. This is in order to manage the furnace pressure avoiding room air in-leakages or excessive out-leakages. • The optimum and by far the most precise, temperature control system in a multiburner furnace is on–off firing (with several options). This requires that burner ignition is prompt and extinction is fast or that transition from a highto low-fire regime is straightforward. Flame performance can be optimized for a well-calibrated burner flow-rate (100% power) and this should accurately match the furnace production rate.
484
• A peculiar feature of flameless oxidation is that it requires recirculation of flue gases above self-ignition threshold (850°C for safety). Below threshold, a burner stabilized flame must be provided (Figure 23.9) and this is carried out by a system capable of selecting between flame mode (the only possible mode below 850°C) and FLOX® mode (above threshold both modes are possible). Below threshold, a flame detector (ionization or UV principle) is required; if FLOX® is selected above threshold, then the flame detector must be neutralized (because there is steady combustion without flame). Dedicated control units have been developed for this purpose [1]; the overall operation of the control system should be carefully checked in operating conditions.
Industrial Combustion Testing
Before revamping: standard burners
After revamping: Glasflox burners – Same gas consumption – NOx emissions abated by 50%
Figure 23.23 CFD comparison of furnace design. (After Scherello, A., Konold, U., Schwarz, G., and Koster, B., Tagungsband zur Glastechnischen Tagung, Dresden, 2006.)
23.7.2 Recuperative Glass Smelter This project for continuous heat recuperative glass melting baths (Figure 23.23) is significant because it has involved a balanced mixture of pilot plant testing with mathematical modeling and with gradual application to industrial revamping [24]. The purpose was to reduce NOx below 500 mg/Nm3 without resorting to end-ofpipe measures (i.e., de-nitrification of flue gases): • Furnace tests at the GWI premises were carried out in order to compare traditional design with a flameless burner, labeled as Glasflox®; this confirmed that the latter was, in principle, substantially better than the conventional lowmomentum design (Figure 23.24). • Partial substitution of the existing burners was possible on the industrial plant: this proved that the new burner could be safely fired in the existing furnace. • Mathematical models have been extensively used in order to interpret observed results and to extrapolate the best solution that was eventually adopted. • Final long-term demo tests confirmed satisfactory performance. This procedure seems to be the virtuous one and it should be followed whenever possible, although boundary conditions may not be as favorable as previously mentioned. It is also worth mentioning that mathematical modeling of the process and combustion chamber was so reliable that it could be employed as a powerful design tool to optimize the combustion system avoiding expensive trials (Figure 23.23). Then modeling could be used in conjunction with experiments in a unified strategy for
© 2011 by Taylor and Francis Group, LLC
Standard burner HWI Pressure drop at 500 kW ~ 2 mbar
Glasflox burner Pressure drop at 500 kW ~ 25 mbar
Figure 23.24 Standard burner and Glasflox®. (After Scherello, A., Konold, U., Schwarz, G., and Koster, B., Tagungsband zur Glastechnischen Tagung, Dresden, 2006.)
better design. The sophisticated measurements quoted in Section 23.6 are carried out in order to develop and to validate advanced mathematical models capable of a more detailed and physically sound description in nonconventional combustion processes. In this case, therefore, process trials are set up to design models and not vice versa (established models are used to design equipment). 23.7.3 Reformers A further example of the nontraditional application of FLOX® burners is small scale steam reformers for the decentralized hydrogen production from natural gas or from other fluid fuel. The concept is based on the Rekumat® burner, which is to make heat available at high temperature and high efficiency at the same time. This is carried out by efficient air preheating and by using combustion heat in the process itself. A detailed process description can be found in the literature [25]. The feature to be outlined here concerns a particular boundary condition: flameless oxidation concerns a part of the equipment and is closely integrated in such a way that the overall size must be compact and that the equipment must be easy to control. In such circumstances, burner testing merges
485
Flameless Burners
additional logic task into account. Even more important, and possibly subject to field testing, is the position of the safety thermocouple(s); it is clearly essential for safe FLOX® firing that temperature is suitably representative of the recirculated flue gases and not too much influenced by local and/or transient conditions. As a last recommendation for industrial operation, flameless oxidation is not visible or audible and the furnace operators, who are often accustomed to control and to check their plants by eyes and ears, must be warned and taught to be confident with the new firing technology.
23.9 Conclusions Figure 23.25 Small reformers: compact and tank station.
with development of an industrial final product and concerns the whole process plant. in this case, this is being done at pilot plant scale and at demo scale in some hydrogen fuel stations in airports (Figure 23.25) or to very small units meant to supply pure hydrogen to Proton Exchange Membrane (PEM) fuel cells.
23.8 Control and Safety Standards A feature that is included in testing flameless oxidation burners concerns combustion supervision: industrial flames can and must be supervised by means of flame detectors that are approved devices specified in accepted standards. These are technically well known and, in fact, they have played an important role in the widespread use of combustion equipment. Flameless and FLOX® firing is, however, not detectable and no similar device has been discovered until now. Therefore, the only way of supervising the process is that the combustion chamber temperature is above a safety threshold. This is ratified in safety standards that define high temperature furnaces, those steadily above 700°C, which is considered explosion free. In fact this means to be above selfignition temperature for most hydrocarbon fluid fuels. For flameless oxidation, the temperature 850°C is usually recommended as a safe threshold, though some hysteresis is also considered when temperature crosses threshold in one or the other direction. This temperature supervision must be error free and therefore two thermocouples are used to minimize default. Below the flameless oxidation threshold, the burner must be automatically switched off or switched to a stable, supervised flame firing. The electronic controls must take this
© 2011 by Taylor and Francis Group, LLC
23.9.1 Application to Heat Processing The flameless oxidation can be adopted where a suitable flue-gas recirculation at high temperature of flue gases can be carried out. These conditions can be fulfilled by numerous heat process plants without modifications of the plant. That is, the existing burners can often be modified and adapted to flameless oxidation. In other processes, especially those at low temperature, the whole plant has been adapted on purpose, which is worth doing in green field construction only. Beside the very low nitric oxide emissions that can be obtained even with high air preheating, the flameless oxidation features other interesting aspects such as: • • • • • •
fuel flexibility temperature uniformity no ignition no flame supervision no overheating close to the burner simple burner geometry
The fuel flexibility derives from the circumstance that with flameless oxidation no stabilization of the flame is required. FLOX® burners work with the largest variety of fuels down to very lean fuels. Gaseous fuels are mainly used, but flameless oxidation is also possible with liquid and solid fuels. For several processes, the temperature uniformity is important in order to obtain a uniform temperature of the stock and to avoid local overheating. The high flow velocities in the absence of temperature peaks contribute to a good thermal uniformity. In high temperature processes, the burner nozzles may be subject to large thermal stresses. In flameless oxidation, the reactions start in the combustion chamber only and, therefore, such stresses are clearly reduced. The nozzle geometries of FLOX® burners are geometrically simpler. Therefore, there is a cost advantage with respect
486
Industrial Combustion Testing
to burners with expensive external recirculation or with multiple air- or fuel-staging. Most FLOX® burners have been utilized until now in the steel industry. Further applications in the fields of glass, ceramic, chemical, power generation, and others are at the moment under development with numerous investigations.
23.9.2 General Features of Flameless Oxidation
The main features of the flameless oxidation can be summarized as follows:
• Thermal NO formation can be efficiently reduced at high combustion air temperatures. • Flameless oxidation is possible when the temperature of the recirculated flue gases is above self-ignition, typically about 850°C. • Flameless oxidation requires high recirculation rates. • In ideal conditions, flameless oxidation takes place without visible or audible appearances. • Flameless oxidation is possible both in adiabatic and nonadiabatic combustion chambers. • Opposite to flame combustion, no high gradients of temperature and composition occur. • Flameless oxidation can be well predicted with commercial CFD software. • With the exception of the burner near-field, the temperature distribution in a furnace is similar to the one onset by high velocity burners.
References
1. Wünning, J. G., and Milani, A. Handbook of Burner Technology for Industrial Furnaces. Essen, Germany: Vulkan Verlag, 2009. 2. Tsuji, H., Gupta, A., Hasegawa, T., Katsuki, M., Kishimoto, K., and Morita M. High Temperature Air Combustion. Boca Raton, FL: CRC Press, 2002. 3. Wünning, J. A., and Wünning, J. G. “Regenerative Burner Using Flameless Oxidation.” International Gas Research Conference, Cannes, France, 1995. 4. Milani, A., Salamone, G. V., and Wünning, J. G. “Abatement of Fuel Consumption with Compact Regenerative Burners in Energy Intensive Furnaces.” EC THERMIE Programme, April 25, 1998. 5. Wünning, J. A. “Flameless Oxidation with Highly Preheated Air.” Chemie Ingenieur Technik 63, no. 12 (1991): 1243–45. 6. Woelk, G., and Wünning J. “Controlled Combustion by Flameless Oxidation.” Joint Meeting of the British and German Sections of the Combustion lnstitute, Cambridge, UK, 1993.
© 2011 by Taylor and Francis Group, LLC
7. Wünning J. A., and Wünning, J. G. “Flameless Oxidation to Reduce Thermal NO-Formation.” Progress in Energy and Combustion Science 23, (1997): 81–94. 8. Wünning, J. A., and Wünning, J. G. “Flameless Oxidation.”Fourth High Air Temperature Combustion & Gasification HTACG, Rome, November 2001. 9. Flamme, M. “New Combustion Systems for Gas Turbines (NGT).” Applied Thermal Engineering 24 (2004): 1551–59. 10. Flamme, M., and Kremer, H. “NOx-Reduction Potential of High Temperature Processes.” International Gas Research Conference, Cannes, France, 1995. 11. IFRF Online Combustion Handbook, Combustion files No. 171–175, International Flame Research Foundation, www.ifrf.net 12. Zabetakis, M. G. “Flammability Characteristics of Combustible Gases and Vapours.” Bureau of Mines, Bulletin 627, Washington, DC, 1965. 13. Wünning J. G. “Flameless Oxidation.” Sixth HTACG Symposium, Essen, Germany, 2005. 14. Milani A., and Wünning J. G. “Design Concepts for Radiant Tubes.” Millennium Steel 2002: 245–49. 15. Cavaliere, A., and de Joannon, M. “Mild Combustion.” Progress in Energy and Combustion Science 30 (2004): 329–66. 16. Chomiak, L., Longwell, J. P., and Sarofim, A. F. “Combustion of Low Calorific Value Gases, Problems and Prospects.” Progress in Energy Combustion Science 15 (1989): 109–29. 17. Berger, R., and Wünning, J. G. “Burners for the Process Integrated Combustion and Utilisation of Biomass or Waste Derived Low Calorific Value Gases.” Eighth European Conference on Industrial Furnaces and Boilers, Vilamoura, Portugal, 2008. 18. Capoferri, G., Mattarini, A., and Ricci, M. “Regenerative Combustion System for Tube Reheating Furnaces.” Millennium Steel 2001: 269–73. 19. Burkhardt, V., Roth, W., Tibbenham, H., and Wünning, J. G. “Annealing and Pickling Lines: Benefits from Advanced Combustion Systems.” Millennium Steel 2006: 284–88. 20. Galletti, C., Parente, A., and Tognotti, L. “Numerical and Experimental Investigation of a Mild Combustion Burner.” Combustion and Flame 151, no. 4 (2007): 649–64. 21. Bonnet, U., Telger, K., and Wünning, J. G. “Direct Fired Strip Preheating.” Sixth HTACG Symposium, Essen, Germany, 2005. 22. Lückerath, R., Meier, W., and Aigner, M. “FLOX® Combustion at High Pressure with Different Fuel Compositions.” Journal of Engineering for Gas Turbines and Power, Trans ASME 130, no. 1 (January 2008). 23. Plessing, T., Peters, N., and Wünning, J. G. “Laseroptical Investigation of Highly Preheated Combustion with Strong Exhaust Gas Recirculation.” Symposium (International) on Combustion 27 (1998): 3197–3204. 24. Scherello, A., Konold, U., Schwarz, G., and Koster, B. “Anwendungen der Flammenlosen Oxidation für Glasschmelzwannen: erste Ergebnisse des Glas- FLOX@ Projektes.” Tagungsband zur Glastechnischen Tagung, Dresden, 2006. 25. Schmid, H. P., and Wünning, J. A. “FLOX® Steam Reforming for PEM Fuel Cell Systems.” Fuel Cells 4, no. 4 (2004): 256–63.
24 Radiant Tube Burners Michael Flamme, Ambrogio Milani, Joachim G. Wünning, Wlodzimierz Blasiak, Weihong Yang, Dariusz Szewczyk, Jun Sudo, and Susumu Mochida Contents 24.1 Introduction................................................................................................................................................................ 487 24.2 Radiant Tube Technologies and its Application Industry................................................................................... 488 24.3 Important Quality Characteristics of Radiant Tubes........................................................................................... 492 24.4 Testing of Radiant Tubes........................................................................................................................................... 492 24.4.1 Testing Facilities.......................................................................................................................................... 493 24.4.2 Measuring Instruments.............................................................................................................................. 494 24.5 Comparative Performance Data Between Regenerative and Recuperative Radiant Tube Burners............... 494 24.5.1 Temperature Profiles................................................................................................................................... 496 24.5.2 Effective Energy from the Radiant Tube.................................................................................................. 497 24.5.3 Energy Balance............................................................................................................................................ 499 24.5.4 Emissions and Pressure Drop................................................................................................................... 500 24.5.5 Conclusions from the Measurements with Different Burners............................................................. 500 24.6 Thermal Performance of the Regenerator.............................................................................................................. 500 24.7 Estimation of Mechanical Strength of the Radiant Tube..................................................................................... 500 24.7.1 Conditions in Testing.................................................................................................................................. 501 27.7.2 Temperature Profile along the Tube Length and the Tube Circumference........................................ 501 24.7.3 Numerical Analysis of the Thermal Stress of the Radiant Tube.......................................................... 501 24.8 Performance of Radiant Tubes at Industrial Applications................................................................................... 502 24.9 Summary and Conclusions...................................................................................................................................... 502 24.10 Nomenclature............................................................................................................................................................. 504 References................................................................................................................................................................................. 504
24.1 Introduction Radiant tubes are used in industry for heat treatment applications in which products are treated under a protective gas atmosphere within heat treatment furnaces. Therefore the heating of such furnaces are performed with indirect fuel-fired systems or electrical heating elements. For indirect fired applications, the flue gas of the combustion process can not enter into the furnace. The combustion takes place within radiant tubes and the heat is transferred—via radiation—from the outer surface of the tube to the process. There are different types of radiant tubes available. For all types the maximum transferred heat is one of the important features of such systems. That means that the maximum radiant tube temperature and the temperature uniformity are important characteristics of radiant tubes. Another issue is the efficiency of the radiant tubes. In
all types of radiant tubes the combustion takes place within the tube and the flue gas leaves the tube at the flue gas outlet. The efficiency of radiant tubes depends on the flue gas outlet temperature. The lower the flue gas temperature the higher is the efficiency of a combustion system. That means as much as possible of the provided energy must be transferred via the radiant tubes to the process and the flue gas outlet temperature must be as low as possible. One way of reducing the flue gas losses is heat recovery from the flue gas for preheating the combustion air. Therefore radiant tubes with internal air preheating technologies like plug-in recuperators, recuperative and regenerative burners have been developed. Another important characteristic of radiant tubes is pollutant emission like NOx and CO. It is important to implement state of the art combustion technology into radiant tubes in order to keep the emissions as low as possible. The lifetime of radiant tubes and maintenance 487
© 2011 by Taylor and Francis Group, LLC
488
costs are also important for the economical application of the technology in heat treatment applications. Experimental investigations of radiant tubes in laboratory furnaces and under practical conditions in field tests are important in order to develop and improve radiant tubes. This chapter will give an overview about the different radiant tube technologies and its application in industry and will describe measurement techniques for experimental investigations of radiant tubes in laboratory furnaces.
24.2 Radiant Tube Technologies and its Application Industry Radiant tubes are used in industry for different heat treatment processes with a protective gas atmosphere in steel and nonferrous metal industry. Radiant tubes are fueled with natural gas, LPG, or light fuel oil. Process heating with radiant tubes leads to lower energy costs compared to electrical heating systems. There are different radiant tubes with different geometries, burners, air preheating technologies, and materials on the market. Radiant tubes can be classified by different criteria. One of the important criteria for the distinction of different systems is if the geometry of the radiant tube allows an internal flue gas recirculation within tube [1]. The internal flue gas recirculation within the radiant tube leads to a better temperature uniformity at the surface of the radiant tube and leads in conjunction with high velocity burners to a recirculation of combustion products into the reaction zone. As a result of the flue gas entrainment the reaction zone will be diluted. Due to the dilution, the flame temperature and the partial oxygen pressure within the reaction zone will be reduced simultaneously. As a consequence the thermal NO formation will be reduced drastically. The dilution of the reaction zone with flue gas leads to a nonvisible flame because the local concentration of OH, CH, and other intermediate products will be reduced and called flameless combustion. The diluted combustion technology is mostly used in conjunction with preheating of the combustion air. Flameless technologies are also described in Chapters 21 and 23 of this book. They are often used in conjunction with high preheating of the combustion air, because they are required to abate NOx emissions and to curb temperature peaks. Figure 24.1 gives an overview about the different types of radiant tubes [4,5]. On the left side of the figure, nonflue gas recirculating radiant tubes are shown. The simplest design is the I-type radiant tube with a burner on one side and the flue gas outlet at the opposite side. This type of radiant tube will be found in Europe only in old
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
installations. The reason for that is the mostly unused combustion air preheating, difficulties with sealing of the moving part of the radiant tube due to thermal distension, the insufficient temperature uniformity at the surface of the radiant tube, and the high flue gas temperatures at the flue gas outlet of the tube. For an enlargement of the surface, I-type radiant tubes can be extended with bends and additional arms in order to build U-type and W-type radiant tubes. With these extensions, the radiation surface of the tubes can be enlarged. For I-type, U-type, and W-type radiant tubes the burners are placed at one end and the flue gas outlet is at the other end. At a one-sided, sealed radiant tube without recirculation, an inner tube will be implemented into an I-type tube and the opposite end will be blocked. The burner will be implemented into the inner tube and the flue gas goes first through the inner tube and flows back in the annulus between the inner tube and radiant tube. Nonrecirculating radiant tubes like the U-type and W-type tubes are mostly used in the American market. In the European market single ended radiant tubes (SER) in combination with auto recuperative burners are the most used types. Single-ended radiant tubes are recirculating types of radiant tubes because there is an annular gap between the burner and the inner tube for flue gas recirculation. Single-ended radiant tubes offer excellent temperature uniformity and they are easy to install into a furnace because they have just one flange and one sealing with the furnace wall at the cold end of the radiant tube. The other end can expand within the furnace due to thermal distension. Horizontally installed steel radiant tubes need an end support at the opposite side of the burner within the furnace. Singleended radiant tubes have, with their inner tubes, one additional part compared to other types of radiant tubes. Other radiant tubes with internal flue gas recirculation are the P-type, double P-type, and A-type radiant tubes. For these tubes the possibility for internal flue gas recirculation is based on the design of the tubes. A uniform temperature distribution is important in order to achieve a uniform heating up of the product within the heat treatment furnace, a maximum heat flux from the tube to the process, and a long life span of the radiant tubes. For nonrecirculating radiant tubes, slowly mixing burners with long flames will be used so that the flame covers the first part of the radiant tube. Overheat ing of the part of the radiant tube close to the burner could be the result of too short flames. Too long flames lead to overheating problems due to turbulences that occur in the first bend of the radiant tube and the completing of the combustion within this part. In conjunction with modulating temperature control, it is difficult to achieve these targets at different loads of the burners. With recirculating radiant tubes, high velocity burners in on/off
489
Radiant Tube Burners
Single-ended radiant tube
SER single-ended radiant tube
U-tube
Recirculating
Non recirculating
Straight tube
P-tube
double-P-tube
A-tube
W-tube
Figure 24.1 Different types of radiant tubes with non flue gas recirculation (left) and flue gas recirculation (right).
operation will be used. The temperature uniformity will be achieved through the intensive flue gas recirculation within the inner part of the radiant tubes. Temperature differences at the surface of the radiant tubes are lower at recirculating tubes compared to nonrecirculating tubes. Figure 24.2 shows the maximum temperature distribution at the surface of different radiant tube technologies [2]. The temperature difference is in the range of 130–150 K for U- and W-shaped radiant tubes and drops down to the range of 20–50 K for single-ended and P-shaped radiant tubes. Single-ended radiant tubes are the ideal geometry in order to achieve a uniform temperature distribution. If the temperature distribution is not uniform, the part of the radiant tube that has a lower temperature has a reduced contribution to the total heat transfer compared to the part that has the maximum surface temperature. This results in a reduction of the total heat, which can be transferred by the radiant tubes. The penalty for this effect leads to an assumed reduction in heat transfer surface of the radiant tube compared to the heat transfer surface at maximum temperature. As a result the investment costs in the heat transfer surface of radiant tubes and in the furnace length could be reduced with radiant tubes designs, which lead to an excellent temperature uniformity. Most of the radiant tubes are made of heat resistant cast alloy or fabricated alloy. The maximum furnace temperature for heat treatment processes is in the range of 900–1100°C depending on the type of material the radiant tubes are made of. Higher temperatures
© 2011 by Taylor and Francis Group, LLC
can be achieved with special alloy or ceramic tubes in conjunction with ceramic auto recuperative burners. Ceramic radiant tubes and parts of ceramic auto recuperative burners are made of reaction bound silicon carbide (SiSiC). Ceramic radiant tubes have been on the market since the mid-1990s. The heat transferred from the radiant tubes to the furnace is a limitation in production capacity of a furnace. With ceramic radiant tubes, the production rate of a furnace can be largely increased due to the higher surface temperature and the higher heat flux of the radiant tubes. The resistance of the silicon carbide material is so high that ceramic radiant tubes can be implemented into a furnace without end support at furnace temperatures of up to 1250°C [3]. The SiSiC is resistant against most of the components of the furnace atmosphere without the alkaline metals potassium and sodium, which can cause corrosion even at low temperatures like 1000°C. Potassium and sodium are often components of washing agents and can be transported into the furnace together with materials that are brought into the furnace for heat treatment if they are not cleaned carefully after washing. Additionally silicon carbide tubes can be affected in dry hydrogen atmospheres [1]. The efficiency of a fuel-fired system is basically the thermal efficiency or also called available heat. The thermal efficiency of a combustion system is depending on the energy that is transported into the system by fuel input minus the thermal energy that is leaving the system at the flue gas outlet. In order to increase
490
Industrial Combustion Testing
W- Rohr 1 burner, 1 plug-in recuperator, efficiency 70%
150°C
870 U- Rohr 2 burner; 2 plug-in recuperator, efficiency 70%
925
1035 1090°C
125°C
870
925
P- Rohr 2 Rekumat SJ2, efficiency 78%
980
1035 1090°C
50°C
870
925
980
1035 1090°C
50°C
Doppel - P - Rohr 1 Rekumat SJ3, efficiency 78%
870 Mantelstrahlrohr 3 Rekumat M150, efficiency 80%
925
980
1035 1090°C
20°C
870
925
Figure 24.2 Effect of inner recirculation on temperature distribution at the surface of the radiant tubes.
© 2011 by Taylor and Francis Group, LLC
980
980
1035 1090°C
491
Radiant Tube Burners
the efficiency, the flue gas outlet temperature must be as low as possible. One effective way is preheating the combustion air by heat recovery from flue gas. In order to increase the efficiency from radiant tubes, plug in recuperators, recuperative, or regenerative burners will be used at radiant tubes to use the energy from flue gas for preheating the combustion air. For U-type and W-type radiant tubes, plug-in recuperators can be used to improve the thermal efficiency. Figure 24.3 shows the application of plug-in recuperators at a U-shaped radiant tube. Another possibility for air preheating at A-, U-, or W-type radiant tubes is the implementation of a regenerative burner pair at such tubes. For this type of radiant tubes one regenerative burner will be installed at each end of the tube. Figure 24.4 shows an installation of regenerative burners at a W-type radiant tube.
In Japan, regenerative burners are applied in U-, W-, and I-type radiant tubes with different kinds of fuels like natural gas, propane, butane, coke oven gas, and process gas [10,11]. In the state of the art regenerative combustion system, the major advantages are ultimate heat recovery rate due to the regenerator, uniformity of temperature profile in the furnaces or inside the radiant tubes, and low NOx capabilities on the basis of High Temperature Air Combustion or HiTAC technology. In U-, W-, and I-type regenerative radiant tube burner systems two identical regenerative burners are installed in each end of the radiant tube. For single-ended, P-shaped, and double P-shaped radiant tubes recuperative burners can be used. Fig ure 24.5 shows the application of a recuperative burner in a SER tube. There are also regenerative burners on Furnace wall
Hot air
Gas
Exhaust
Air Recuperator
Plug in recuperator Figure 24.3 W-type radiant tube with plug in recuperator.
Air Gas
Exhaust
Figure 24.5 Application of a recuperative burner in a single-ended radiant tube.
Gas
Air
Exhaust
Air Gas
Exhaust Figure 24.4 W-type radiant tube with regenerative burners.
© 2011 by Taylor and Francis Group, LLC
492
Industrial Combustion Testing
Regenerators Switching valves Exhaust
Gas
Air
Figure 24.6 Application of a regenerative burner in a double P-shaped radiant tube.
the market that integrate a regenerative burner pair in one unit, for example, the Regemat® burner from the WS Thermal Process Technology, Inc. This burner has been applied at double P-type radiant tubes [5]. Figure 24.6 shows a double P-shaped radiant tube with regenerative burner.
24.3 Important Quality Characteristics of Radiant Tubes One of the most important quality characteristics of radiant tubes is the maximum heat flux from the tubes to the process/product, the temperature uniformity at the surface of the radiant tubes, and the heat flux uniformity versus the length of the radiant tubes. As discussed previously, the temperature uniformity has a strong influence on the maximum heat that can be transferred from the surface of the tube to the product. Additionally the temperature uniformity also has an influence on the life span of the radiant tube. If there are hot spots at the surface of the tube, thermal corrosion and thermal tension of the material will increase at the hot spots and lead to a reduced life span of the system. Additionally the maximum temperature within the hot spot is limited by the material characteristics of the radiant tube and therefore regions with lower temperatures reduce the total heat, which can be transferred from the radiant tubes. Another very important quality characteristic is the combustion efficiency (fuel efficiency) of radiant tubes. The efficiency strongly depends on the flue gas outlet temperature of radiant tube systems. In order to achieve maximum efficiency, air preheating systems use the enthalpy of the flue gas for preheating the combustion air simultaneously reducing the flue gas outlet temperature of such systems. High efficiencies with less flue gas losses result in lower fuel consumptions and lead to a reduction of the CO2 emission per ton of product.
© 2011 by Taylor and Francis Group, LLC
Hazardous emissions like NOx, and unburned or partially burned gases like CO should also be as low as possible. In order to achieve the target and to keep the emission limits of the legislations of the local authority, modern combustion technologies like flameless combustion or HiTAC technologies have to be implemented into the radiant tube system. With standard combustion technologies in combinations with high air preheating, the NOx emissions would increase exponentially. So it is important to use internal flue gas recirculation technologies in order to dilute the combustion zone with flue gas. This is even more important for radiant tubes because the gas temperature inside the radiant tube is higher than the equivalent temperature in free flame (direct heating) conditions. This is inevitable because of the screening effect of the tube wall that hinders direct heat transfer from the flame to the stock. The pressure drop over radiant tube systems including burner and air preheating technology is also an important factor because higher pressure drop costs more electrical power for the air van and has to be taken into account for total efficiency calculations. Other important quality characteristics are required in maintenance and the life span of radiant tube systems. Therefore in experimental investigations, long time experiments will also be necessary in order to get information about these characteristics.
24.4 Testing of Radiant Tubes In order to get information about the quality characteristics of radiant tubes experimental investigations are necessary. With computational fluid dynamic (CFD) simulations, an optimization of combustion systems is possible but finally the combustion systems have to be tested experimentally. With experimental work it is possible to get results from existing radiant tubes or to develop new technologies. The tests can be carried out
Radiant Tube Burners
in lab-scale furnaces with high temperature conditions and also in industrial applications under practical conditions. But in industrial applications it is not easy to carry out experiments without any disturbance of the production. Therefore most of the tests will be carried out at lab-scale furnaces. In lab-scale furnaces, experiments can be carried out with industrial scale radiant tubes because the thermal load of the radiant tubes is not too high for lab-scale conditions. The thermal load of radiant tubes for industrial applications is in general less than 300 kW per tube. In order to get information about the temperature uniformity at the surface of the radiant tube, the maximum heat flux measurements of the temperature distribution along the radiant tube surfaces have to be carried out. Additionally results of the efficiency, the NOx and CO emissions and the maintenance and lifetime of the radiant tubes can be achieved at lab-scale furnaces. It is also possible to test the influence of different burner technologies on the performance characteristics of radiant tubes at lab-scale furnaces. For these tests radiant tubes can be equipped with different burner technologies in order to estimate the influence of the performance of the radiant tubes. Under practical conditions in industrial applications, it is possible to carry out measurements of the emissions and the efficiency (flue gas outlet temperature and excess air ratio) and to get information about the effort for maintenance and the lifetime of radiant tube systems. In special cases it could be possible to change just one radiant tube of an industrial furnace in order to test it under production conditions.
493
23/1/03
Figure 24.7 Front side of the furnace equipped with high-cycle regenerative system.
24.4.1 Testing Facilities In order to test radiant tubes the systems have to be tested in lab-scale furnaces in which one can achieve similar conditions like in industrial applications concerning process temperature, surface temperature of the radiant tubes, and heat flux. Therefore a lab-scale furnace should be equipped with a heat sink in order to control the furnace temperature. For measurements of the efficiency and the emissions like NOx, it is also important to test the radiant tubes at the same conditions like in the industrial applications because the temperature of the radiant tube has an influence on the results. Also tests to get information about the effort for maintenance and life span of the radiant tubes during a longer testing period can be carried out in the same conditions as the industrial application. In the following some lab-scale furnaces for testing radiant tube systems are shown. Figure 24.7 shows the semi-industrial HiTAC test furnace of the Royal Institute of Technology (KTH), Stockholm, Sweden. The outer dimensions of the furnace body is 3.500 × 2.200 × 2.200 m. The furnace body
© 2011 by Taylor and Francis Group, LLC
Figure 24.8 (See color insert following page 424.) Furnace layout for testing of radiant tube performance in pilot-scale industrial furnace.
is insulated with a 0.3 m thick layer of ceramic fiber material. The inner volume of the combustion chamber is 7.4 m3 [7]. Figure 24.8 shows the layout for testing radiant tube performance in a pilot-scale industrial furnace (PSIF) of the Canmet Energy Technology Centre, Ottawa, Ontario, Canada [9]. The furnace is 4.5 m × 3.0 m × 1.0 m (inside dimensions) and it can be modified to simulate any industrial furnace geometry. With a firing rate of 1.2 MW, the temperature, heat transfer, and chemical environment found in most industrial processes can be emulated as well. The furnace is equipped with a calorimeter for total heat flux (34 cooled plates on the floor of furnace).
494
Industrial Combustion Testing
24.4.2 Measuring Instruments For testing the temperature uniformity, thermocouples have to be positioned at the surface of the radiant tubes in order to get information about the surface temperature distribution along the axis of the tubes. For testing the heat flux, total heat flux probes or narrow angle radiation heat flux probes could be used. Modern labscale furnaces are equipped with heat sinks at the bottom of the furnace. These heat sinks are divided into several elements and the heat flux distribution can be measure simultaneously. For the control of the efficiency of radiant tubes, the flue gas outlet temperature has to be measured with a suction pyrometer or a simple thermocouple. For the measurement of the flue gas temperature, the measured temperatures have to be compensated for radiation loses of the thermocouples with the surrounding. Additionally the excess air ratio has to be controlled by O2 measurements within the flue gas of the radiant tube. The emissions of the radiant tubes can be measured with standard flue gas analyzing technologies for NOx and CO. In the case of radiant tube burners without air preheating and with recuperative air preheating technology the measurements should be carried out at steady-state conditions. In the case of the application of regenerative burners for radiant tubes no steady-state conditions will be achieved because of the switching between the regenerators. The unsteady-state conditions strongly depend on the switching time of the regenerators. For measurements in unsteady-state conditions, the time constant of the measuring instruments have to taken into account [12,13]. For more details about measurements at unsteady-state conditions, see Chapter 21 of this book.
24.5 Comparative Performance Data Between Regenerative and Recuperative Radiant Tube Burners Within this chapter, measurements at a radiant tube with recuperative and regenerative burner technologies will be shown as an example of how measurements should be carried out. The measurements have been accomplished at the lab-scale furnace of KTH, Stockholm [6–8], which has been shown (Figure 24.7). A W-shape radiant tube, shown in Figure 24.9, was used in the experiments. The outer diameter of the tube was 0.195 m and its maximum thermal effect was 200 kW. The 74 K-type thermocouples, placed in 41 locations, were mounted along the length of the tube, as shown in Figure 24.10. At least one thermocouple was mounted in each location on the right side of the tube. Points marked
© 2011 by Taylor and Francis Group, LLC
Figure 24.9 W-shape radiant tube used for the experiments.
with 2s mean that there are two thermocouples, one is on the right side and the other is on the left side. Moreover, those locations that are close to burners are marked as 4s, meaning that there are four thermocouples on the left, right, top, and bottom. At elbows, there are three thermocouples, 3s, at the right, left, and at the surface that has the smaller radius. It was not possible to put the fourth thermocouple at elbows because the tube supports are located at that place. Tube temperature reference point (TRP), is located at a distance of 0.995 m from the beginning of the working part of the tube, Figure 24.10. As a conventional system, the recuperative radiant type burning system, manufactured by Chugai Ro Kogyo Co. Ltd., with nominal firing capacity 175 kW, was used in order to compare with the (HRS) high-cycle regenerative system. The Figure 24.11 shows the measurement points in this system. A HiTAC system, manufactured by Nippon Furnace Co., Ltd., (NFK) with nominal firing capacity 158 kW, was used. This system is composed of a pair of burners, where one is working in firing mode and the other is working in regenerative mode at the same time. Modes were changed by a switching valve every 30 seconds. As a regenerator, a ceramic honeycomb was used. Figure 24.12 shows the measurement points in the system. The fuel used for experiments was a typical LPG used by industry, composed of more than 98% propane, 0.9% ethane, and 0.8% butane with lower heating value equal to 93.2 MJ/Nm3. The investigations using both burning systems were made at different operating conditions, such as reference point temperature, firing capacity, and air/fuel ratio. Table 24.1 shows the operating conditions of eight tests set together in order to compare the results
495
Radiant Tube Burners
240
150
230
1650
280
995 Temperature reference point (TRP)
φ 975
4s
2s 4s
2s 4s
2s 4s
2s
41 40 39 38 37 36 35
34
33
21
Bottom regenerative burner
φ 195
31
30
26 16
27 15
28 14
10
11
12
3s
29
3s
13
1200 22 20
3s
25 17
24 18
23 19
4s
280 1
2
4s
2s 4s
50
32
4s
Elbow - 2
φ 1300
Top regenerative burner or recuperative burner
Elbow - 1
200
4
3
5
6
7
2s 4s
2s 4s
200
200
9
8
2s
200
200
200
200
Elbow - 3
200
Figure 24.10 Schema of radiant tube and locations of temperature measurement points on the wall of the radiant tube.
Gas Pilot air
Pilot gas Pilot gas
125 PF
TA TFG
PA
Gas
TA
(b)
PF
PFG PA
TA
PA
XFG
XFG
Pilot air
TFG P FG
Flue gas
Air
XFG
TA Pilot gas
Air
TFG PF Figure 24.11 Recuperative system and location of measurement points.
afterward. Tests numbered from CRS-1 to CRS-4 refer to Conventional Recuperative System, while the others refer to High-cycle Regenerative System (HRS-1 to HRS-4). Comparisons were performed at four sets of reference point temperature (i.e., 840°C, 880°C, 950°C,
© 2011 by Taylor and Francis Group, LLC
XFG
Gas
PF PA
(a) Pilot air
TFG
TFG Flue gas
Figure 24.12 High-cycle regenerative system and location of measurement points.
496
Industrial Combustion Testing
Table 24.1
Table 24.2
Operating Conditions for the W-Type Radiant Tube
Characteristic Temperatures on Longitudinal Profiles
No. − I II III IV
Test No.
Burner System Type
TTRP
V(QF)
O2
−
−
°C
Nm3/h (kW)
%
recuperative regenerative recuperative regenerative recuperative regenerative recuperative regenerative
840
3(78)
3
880
5(129)
3
950
5(129)
3
1000
6(155)
3
CRS-1 HRS-1 CRS-2 HRS-2 CRS-3 HRS-3 CRS-4 HRS-4
No.
II III IV
and 1000°C) and at three levels of firing capacity while the level of oxygen content was kept the same. 24.5.1 Temperature Profiles The longitudinal temperature profiles were obtained using 41 thermocouples located on the right side of the radiant tube. Figure 24.13 shows these profiles for both, recuperative and regenerative, systems at different operating conditions. In the case of a recuperative system, the highest temperature occurred at a location very close to the TRP. At further distances, the temperature decreases along the tube length. The shape of profile is similar to typical shapes obtained when the conventional recuperative system is used. In the case of HRS there is at least one local peak at each end of the tube because of two burners. However, the rest of the temperature profile is almost steady and the temperature is level in about three meters length of the tube. Only at first comparison, the maximum temperature of the wall during tests is lower for HRS compared with CRS due to the difference of reference temperature point equaling 32°C. When the difference of TRP between compared tests is in the order of a few Celsius degrees, the maximum temperature occurred on the profile of HRS. Although the maximum temperature is higher in the HRS case, the temperature difference, between maximum and minimum temperature of the radiant tube wall, is significantly lower, Table 24.2. This is an effect of the HiTAC features, where flame volume is large and temperature gradient in flames is lower. It is certain that uniformity of temperature distribution along the radiant tube is then better, and the longitudinal temperature profile can be treated as symmetrical. These graphs show that the temperature level is much higher along the radiant tube in the case of HRS. This is a preliminary conclusion for a higher level of heat flux. Although the temperature difference increases with the decreased reference point temperature for both systems, this is not significant in the case of HRS.
© 2011 by Taylor and Francis Group, LLC
− I
TTRP
−
°C
°C
°C
°C
847 815 882 876 949 949 1004 1007
842 824 864 878 931 960 987 1021
633 750 699 799 805 895 881 955
209 74 165 79 126 65 106 66
CRS-1 HRS-1 CRS-2 HRS-2 CRS-3 HRS-3 CRS-4 HRS-4
TMAX
TMIN
ΔT
Test No.
It has to be noticed that although the profiles have been compared for the same maximum reference point temperature, the maximum temperature on the profiles, created for the right side of the tube, is not the same. It is due to the location of the TRP on the top of the tube. The real obtained maximum temperatures of the reference point are shown on graphs in Figure 24.13 and in Table 24.2. For all tests, the TRP point is situated on or very close to profile in the case of the HRS, while for the recuperative system this point usually is about 20°C higher than the maximum temperature of the right side profile. This is a result of nonuniform temperature distribution on the circumference, which will be explained further. In the case of the HRS, both peak temperatures are placed closer to the ends of the radiant tube, than the location of TRP (which was equal to 0.995 m). As a result the reference point temperature, which determines the life span of the tube, has to be moved to another place in the case of HRS. This temperature will be a few degrees higher on the top than on the right side, as will be described next. Temperatures around a cross section of the radiant tube at point 17, located in the middle of the tube (Figure 24.10), are shown in Figure 24.14. The graphs also show that for a cross-sectional profile, the temperature differences around the cross section of the tube are lower in the case of the HRS. In the case of the recuperative system, there is a noticeable (around 25°C) temperature difference between the top and bottom sides of the tube. This difference is due to heating by the upper, hotter part and cooling by the lower, colder part of the tube. In the case of HRS, the temperature difference between both ends of the tube is negligible, because of cyclic work of the system and uniform, symmetrical, and longitudinal temperature profile. However, higher temperature at the bottom side, than that on the top, can be observed in contrast with the recuperative system. In this case it is because the considered measurement point is located closer to the lower part of the tube. This difference is negligible and in the result, the cross-sectional temperature profile is also more uniform.
497
Radiant Tube Burners
(b)
850
1st Elbow
800
2nd Elbow
3rd Elbow
Temperature T (°C)
Temperature T (°C)
(a)
750 700 650 600
0
1000
2000
3000
4000
5000
6000
1st Elbow
950
2nd Elbow
3rd Elbow
900 850 800 750
0
1000 2000 3000 4000 5000 6000 Distance from the beginning of the tube (mm) CRS-3 HRS-3
700
0
1000 2000 3000 4000 5000 6000 Distance from the beginning of the tube (mm)
1050
1st Elbow
1000
7000
CRS-2-TRP HRS-2-TRP
2nd Elbow
3rd Elbow
950 900 850 800
7000
3rd Elbow
750
CRS-2 HRS-2 (d)
1000
2nd Elbow
800
CRS-1-TRP HRS-1-TRP
Temperature T (°C)
Temperature T (°C)
(c)
1st Elbow
850
650
7000
Distance from the beginning of the tube (mm) CRS-1 HRS-1
900
0
1000 2000 3000 4000 5000 6000 Distance from the beginning of the tube (mm) CRS-4 HRS-4
CRS-3-TRP HRS-3-TRP
7000
CRS-1-TRP HRS-1-TRP
Figure 24.13 Longitudinal temperature profile of the radiant tube for recuperative and regenerative systems at different reference point temperature TTRP: (a) 840°C, (b) 880°C, (c) 950°C, and (d) 1000°C.
24.5.2 Effective Energy from the Radiant Tube In order to compare both systems, the effective energy from the radiant tube resulting from radiative heat flux, some assumptions were done due to different maximum temperatures on the right side longitudinal temperature profiles. Temperature profiles were displaced to have the same maximum temperature. This temperature was chosen as the average temperature, TAV,MAX, from maximum values, TMAX, in each test. All temperature points were multiplied by factor f, Equation 24.1.
f=
TAV,MAX . TMAX
(24.1)
As a result, the profile in the case of HRS was moved down a few degrees to lower the temperature region and the profile in the case of recuperative system was
© 2011 by Taylor and Francis Group, LLC
moved up a few degrees to the higher temperature region. In the first case it was opposite (Figure 24.15), because of a higher maximum right side temperature for CRS. After these corrections, the radiative heat flux, q, was calculated, Equation 24.2.
q=
σ×ε A
∑ A × (T i
4 AV,i
4 − TAV,FR ).
(24.2)
As 41 temperature measurement points were located with a different length between them, the temperature in every place after displacement TAV,i was assumed to be the same in each surface area Ai. The difference between temperatures on the circumference was neglected, because of low value and because in practice the radiant tube emitted energy mainly from the right and left sides. Calculations were done assuming different
498
Industrial Combustion Testing
HRS-1
710 700
940 930
689
920
680
910 678
682
690
900
CRS-1
670
973
967
HRS-4
914
720
950 935
730
960
911
Temperature T (°C)
740
959
970
Left side Top side Right side Bottom side
923
766
980
705
Temperature T (°C)
750
760
760
990
750
Left side Top side Right side Bottom side
758
770
966
(b) 1000
(a) 780
CRS-4
890
Figure 24.14 Cross-sectional temperature profile of the radiant tube at point 17: (a) for tests CRS-1 and HRS-1, (b) for tests CRS-4 and HRS-4.
850
833
1st Elbow
2nd Elbow
833
Temperature T (°C)
800
3rd Elbow
750 700 650 600
0
1000 2000 3000 4000 5000 6000 Distance from the beginning of the tube (mm) CRS-1
CRS-1-corrected
HRS-1
HRS-1-corrected
7000
Average maximum temperature
Figure 24.15 Longitudinal temperature profiles corrected to the same average maximum temperature.
average temperature inside the furnace (e.g., 250°C, 500°C, 750°C, and minimum temperature, TAV,MIN, of the profiles, which always occurred in the case of a recuperative system). Finally, the heat flux ratio was calculated, Equation 24.3
R=
qHRS . qCRS
(24.3)
The results of calculations (Table 24.3 and Figure 24.16) reveal that the radiative heat flux from the tube is
© 2011 by Taylor and Francis Group, LLC
always higher, when the HRS is used. This difference depends on operating conditions. Obviously it is dependent also on the average temperature in the furnace, TAV,F, Figure 24.17. Graphs in this figure show the area of radiative heat flux at two average furnace temperatures. At lower temperatures, the area is mainly common for two systems and as a result, the heat flux ratio is not significantly high. However it can be clearly seen that at high temperatures of heat-treated material, the area in the case of HRS is about two times bigger than in the case of CRS, resulting in two times higher radiative heat flux. When the temperature inside the furnace is low, the increase of the heat flux is about 6 and 38% for high and low reference point temperatures, respectively. On the other hand, when the temperature inside the furnace is much higher, the increase of the heat flux goes over 100%. In practice, it can be concluded that one of the advantages from the appliance of the HRS is that this technology has particularly huge benefits in the regions of furnace, where the temperature of heating treated material is high. The empty spaces in Table 24.3, first and second comparison, are because the minimum temperatures of the right side longitudinal temperature profile after correction were lower than 750°C and the heat flux calculation would not make sense. From these results, one can conclude that for the same level of heat flux for both systems, the maximum temperature in the case of the HRS will be much lower than for a recuperative system, thereby
499
Radiant Tube Burners
Table 24.3 Radiative Heat Flux and Radiative Heat Flux Ratio TAV,FR (250°C) No.
Test No.
q
−
kW/m
− I II III IV
CRS-1 HRS-1 CRS-2 HRS-2 CRS-3 HRS-3 CRS-4 HRS-4
2
31.8 43.4 42.8 49.7 60.1 67.3 77.6 82.3
TAV,FR (500°C)
R
q
− 1.36
TAV,FR (750°C)
R
q
kW/m
−
21.6 33.2 32.6 39.5 49.8 57.1 67.4 72.0
1.54
1.16 1.12 1.06
2
R
q
kW/m
−
kW/m
−
− − − − 23.0 30.3 40.6 45.2
−
10.8 22.4 12.4 19.3 11.4 18.7 12.4 17.0
2.07
2
1.21 1.15 1.07
TAV,FR (TAV,MIN)*
− 1.32 1.11
R 2
1.56 1.64 1.37
*Values of TAV,MIN are shown in Figure 24.16.
2.2
(a) 60
TAV,MIN = 626°C
50
TAV,MIN = 704°C
1.6
40
TAV,MIN = 818°C
1.2 1 200
20
CRS-1
10
300
400 500 600 700 800 900 Average temperature in the furnace (°C)
I (CRS-1 and HRS-1)
II (CRS-2 and HRS-2)
III (CRS-3 and HRS-3)
IV (CRS-4 and HRS-4)
650 1250 1850 2450 3050 3650 4250 4850 5450 6050 6650 Distance from the beginning of the tube (mm)
50 40 30 20 10 0
24.5.3 Energy Balance In order to do the energy balance for the whole system, the temperature of flue gas was measured as well as the temperature of the surface of burner units and pipes. Compensation of temperature measurement due to radiation to the pipe walls was made in order to find the real exhaust gas temperature. Calculating energy loss with flue gas, QFG; estimating surface losses, Q S; and having chemical energy of the fuel, QF; and thermal energy of the air, QA; the efficiency of the system, η, was calculated, Equation 24.4
© 2011 by Taylor and Francis Group, LLC
0
(b) 60
influencing the life span of the radiant tube in a positive way.
Q + QS η = 1 − FG × 100%. QF + QA
0
1000
Figure 24.16 Radiative heat flux ratio as a function of average furnace temperature.
HRS-1 above CRS-1
30
TAV,MIN = 897°C
1.4
Heat flux (kW/m2)
1.8
Heat flux (kW/m2)
Heat flux ratio (–)
2
(24.4)
CRS-1 0
HRS-1 above CRS-1
650 1250 1850 2450 3050 3650 4250 4850 5450 6050 6650 Distance from the beginning of the tube (mm)
Figure 24.17 Radiative heat flux for CRS-1 and HRS-1 at different average temperatures in the furnace TAV,FR: (a) 250°C, (b) TAV,MIN = 626°C.
The measurement and calculation results (Table 24.4) clearly reveal that efficiency of the whole system is higher in the case of a HRS. The difference is significantly big, up to 25%. It becomes bigger in higher temperatures of the reference point. Higher efficiency for HRS is mainly due to a lower temperature of flue gases. The surface losses are also lower, because of a smaller surface area of burner units and pipes in comparison with a recuperative system.
500
Industrial Combustion Testing
Table 24.4
Flue gas Combustion air
Flue Gas Temperatures, Surface and Flue Gas Losses, and Efficiency of the System No. − I II III IV
Test No.
TFG
QFG
QS
η
Δη
−
°C
kW
kW
%
%
548 261 678 309 730 298 815 325
19.6 8.8 40.4 16.6 44.0 16.4 58.9 21.4
3.3 2.5 4.5 2.9 5.3 3.6 6.1 4.1
71.6 85.1 65.8 83.3 62.9 84.0 58.6 82.8
13.4
CRS-1 HRS-1 CRS-2 HRS-2 CRS-3 HRS-3 CRS-4 HRS-4
17.5
1027°C 59.1 Nm3/hr
1120°C 64.2 Nm3/hr
Heat exchange rate 167 kcal/30 sec = 20,000 kcal/hr
φ 100 × 400L (mm) Flue gas heat recovery rate = 75%
21.1 24.2 280°C
24.5.4 Emissions and Pressure Drop The pollutant emission measurements show that a HRS works with higher efficiency without the cost of higher NO emission. For example NO emission from the conventional recuperative system, calculated for λ = 1, was equal to 179 ppm and 284 ppm in test CRS-1 and CRS-2, respectively, while it was more or less the same using HRS, 167 ppm and 316 ppm for HRS-1 and HRS-2 tests, respectively. No emission of CO was observed during normal operation for both systems. The measurement of pressure difference across the systems revealed that the pressure drop due to HRS is more than 10 times higher than that when the conventional recuperative system is in operation. The pressure drop was 1.7 mbar and 2.8 mbar in cases CRS-3 and CRS-4, respectively, while it was 36 mbar and 30 mbar for HRS-3 and HRS-4, respectively. 24.5.5 Conclusions from the Measurements with Different Burners Temperature profile of the wall of the radiant tube with the use of HRS is much more uniform in comparison with conventional recuperative system. This is applicable to both, the longitudinal and the cross-sectional, temperature profiles. As a result of the uniformity of temperature profile, the heat flux (energy released from the tube) is higher in the case of HRS for the same reference point temperature of the tube. The effective energy from the tube can be doubled in some cases, especially when the temperature of heat-treated material is high. It is much easier to control the temperature of the radiant tube equipped with HRS due to temperature distribution uniformity. The lifetime of the tube will be longer because the required level of heat flux can be achieved for a lower firing rate and with lower maximum temperature. System efficiency in the case of HRS is higher, up to 25% at the same reference point temperature. The above improvement of efficiency and effective energy
© 2011 by Taylor and Francis Group, LLC
50°C Flue gas Combustion air
Figure 24.18 Example of heat balance calculation for a ceramic honeycomb regenerator.
emitted from the tube can be achieved using HRS without the expense of more NOx and CO emission.
24.6 Thermal Performance of the Regenerator An example of heat balance calculation for a ceramic honeycomb regenerator [10,11] is shown in Figure 24.18. Combustion flue gas enters into the regenerator with 1120°C and leaves it with 280°C, where air is preheated from 50°C to 1027°C. In regenerative radiant tube burners the heat exchange rate of the regenerator is 23.3 kW and the flue gas heat recovery rate is 75%. In regenerative radiant tubes, the capability of heat exchange rate per volume of the regenerator is very important in the compactness of the system as well as the low pressure loss nature of the regenerator as 100% of the flue gas leaves the radiant tube through the regenerator and bypassing a part of the flue gas is impossible. This is different from direct fired applications of regenerative burner systems where a part of the flue gas leaves the furnace via the stack without passing through the regenerator.
24.7 Estimation of Mechanical Strength of the Radiant Tube In order to secure the mechanical strength of radiant tubes at a high temperature operation condition, a series
501
of tests were executed by Japan Industrial Furnace Manufacturing Association (JIFMA) in a project organized by The Mechanical Social Systems Foundation (MSSF) under the financial support of Aggregate Corporation JKA in a test furnace measuring the temperature profile of the radiant tube with a regenerative and a conventional recuperative radiant tube burner [10,11]. By using these temperature data, numerical analysis for the thermal stress of the radiant tube for both burner systems were conducted.
Skin temperature (°C)
Radiant Tube Burners
24.7.1 Conditions in Testing
24.7.2 Temperature Profile along the Tube Length and the Tube Circumference Measured tube skin temperature profile along the tube length at furnace temperature of near 900°C with 100% load in a regenerative case is shown in Figure 24.19 and in a recuperative case is shown in Figure 24.20. These data show the regenerative case has a more uniform temperature profile along the tube length compared to the recuperative case. These data show that the temperature difference is larger if the furnace temperature decreases. In the recuperative case the average temperature difference in all measured data is 112°C (standard deviation 24.5°C). In the regenerative case it is 71°C (standard deviation 31.7), so that the value is lower by 41°C but the standard deviation is better by 7°C than the recuperative case. The temperature profile along the tube circumference was measured in both regenerative and recuperative cases. The measured points are two positions along the tube length, one is 1213.5 mm and the other is 2013.5 mm from one end of the tube. In each position, four equal distance points along the tube circumference were measured. The average temperature differences and the standard deviation along the tube circumference for both regenerative and recuperative case are shown in Table 24.5. The results show that there are no significant average temperature differences along the tube circumference between a regenerative and recuperative burner system.
© 2011 by Taylor and Francis Group, LLC
A BNR 0
1000
B BNR 2000
Furnace temp. 900(°C) Excess air ratio 1.22
3000 4000 5000 Tube length (mm)
6000
7000
8000
Figure 24.19 Tube skin temperature profile along the tube length in the regenerative burner. (Sudo, J., and Mochida, S., Internal paper on Radiant Tube Burner Testing, Yokohama, Japan: Nippon Furnace Co., Ltd., 2009. With permission.) 1000 980 960 Skin temperature (°C)
The inner dimensions of the test furnace are 1.0 m (W) × 2.7 m (L) × 2.3(H) and the radiant tube is equipped horizontally. The W-type radiant tube has a total length of 8055 mm and a diameter of 7 inches (outer diameter is 194 mm and inner diameter is 177 mm). The material of the radiant tube is heat resistant cast steel (SCH24). The thermocouple positions is welded on the outside surface of the tube along the length and the tube circumference. The operation parameters in the tests are each 500– 900°C for furnace temperature, 20–100% firing rate and 1.15 to 1.50 for excess air ratio.
1000 980 960 940 920 900 880 860 840 820 800 780 760
940 920 900 880 860 840 820 800
Furnace temp. 899(°C) Excess air ratio 1.18
780 760 0
1000
2000
3000 4000 5000 Tube length (mm)
6000
7000
8000
Figure 24.20 Tube skin temperature profile along the tube length in the recuperative burner. (Sudo, J., and Mochida, S., Internal paper on Radiant Tube Burner Testing, Yokohama, Japan: Nippon Furnace Co., Ltd., 2009. With permission.)
24.7.3 Numerical Analysis of the Thermal Stress of the Radiant Tube With respect to the W-type radiant tube used in the above noted comparative test, stress analysis was carried out in order to investigate the mechanical strength of the radiant tube. The method adopted is a Three Dimensional Elastic Stress Analysis using FEM ANSYS version 9, where the self-weight, measured temperature profile and assumed tube inside pressure at extraordinary combustion are considered. In the thermal stress analysis the measured tube surface temperature at 20 points, which seems to be timely, steady state was used as a smooth temperature curve over the measuring
502
Industrial Combustion Testing
Table 24.5 Average Temperature Differences and the Standard Deviation along the Tube Circumference
Burner
Temp. Difference at RT2 Point °C
Standard Deviation °C
Temp. Difference at RTA4 Point °C
Standard Deviation °C
32.38 34.0
7.59 15.5
22.09 19.75
7.33 9.6
Regenerative Recuperative
Source: Sudo, J., and Mochida, S., Internal paper on Radiant Tube Burner Testing, Yokohama, Japan: Nippon Furnace Co., Ltd., 2009. With permission.
Table 24.6 Calculation Results of the Thermal Stress on the Radiant Tube Maximum Stress on the Tube (MPa) Burner Regenerative Recuperative
Tube Inside Pressure 0 MPa
Tube Inside Pressure 0.6 MPa
14.6 14.5
16.4 16.0
Source: Sudo, J., and Mochida, S., Internal paper on Radiant Tube Burner Testing, Yokohama, Japan: Nippon Furnace Co., Ltd., 2009. With permission.
points. The combustion conditions in the analysis are at 100% load firing and at target furnace temperature of 900°C. This numerical analysis was done for both a regenerative and a recuperative burner. The results are shown in Table 24.6 where the maximum stress on the tube between the regenerative and the recuperative burner is almost the same for both tube inside pressures of 0–0.6 MPa. These values are thought to be low enough than the creep rupture strength at 900°C with a long working time—about 10,000 hours—whereas average value of the creep rupture strength of the radiant tube material (KHR35H equivalent to SCH24) used in the test is 31.0 MPa (Min.26.9 MPa) at 900°C and 10,000 working hours. As a conclusion the possibility of a rising risk in terms of strength of the radiant tube is considered to be very low.
24.8 Performance of Radiant Tubes at Industrial Applications The performance of radiant tube systems under production conditions in industrial applications can be estimated by measuring the flue gas outlet conditions like temperature, oxygen content, and emissions like NOx. Together with measurements of total fuel consumption, conclusions concerning performance of heating systems
© 2011 by Taylor and Francis Group, LLC
can be drawn. In the case of retrofitting an existing furnace with more advanced technology measurements before and after retrofitting gives information about the differences related to efficiency and emissions. As discussed previously the efficiency of radiant tubes depend on the flue gas losses. In order to reduce the losses, systems with high heat recovery from flue gas should be implemented. The following two examples for retrofitting existing furnaces with regenerative radiant tubes will be discussed [10,11]. At a continuous galvanizing and annealing line furnace, 52 pairs of regenerative radiant tube burners were installed in a horizontal second heating zone of 140 tons/hr as shown in Figure 24.21. The 26 U-type radiant tubes, 7 inches in diameter are laid out over and under the strip line. The firing fuel is propane and the burner capacity is 140 kW. The design furnace temperature is 950°C and the outlet material temperature is 750°C. A major results waste heat recovery rate is 80.3% compared to 45.4% for the conventional burner before revamping. This leads to a fuel saving rate of 25.5%. The NOx emission recorded was 118 pm corrected to 11% O2 at dry flue gas with external flue gas recirculation of 20%. Some improvement of the outlet strip temperature uniformity is reported to have been achieved. At a continuous roller hearth type annealing furnace 24 pairs of regenerative radiant tube burners were applied to the heating and soaking zone shown in Figure 24.22. The material to be heat treated is cast iron joints and design net capacity of the furnace is 1.67 ton/hr. The total length of the furnace is 39 m and the maximum furnace temperature is 940°C. The total length of the U-type radiant tube with 6 inches diameter is 4.3 m. The radiant tubes are equipped horizontally in each upper and down side of the roller hearth. The fuel used is natural gas and the design burner firing rate is 81 kW. As a result, the fuel consumption rate was 2,070 MJ/ton, which is equivalent to about 52% reduction against a conventional burner type of the furnace. The NOx recorded was 120 ppm corrected to 11% O2 and dry flue gas.
24.9 Summary and Conclusions Radiant tubes are used in industry for heat treatment application in which products are treated under a protective gas atmosphere within the heat treatment furnaces. There are different radiant tubes with different geometries, burners, air preheating technologies, and materials on the market. In order to increase the efficiency of radiant tubes, plug in recuperators, and recuperative or regenerative burners will be used at radiant tubes to
503
Radiant Tube Burners
2nd heating zone specification Capacity Furnace temp. Material output temp Material Fuel Burner
Strip
Strip
140 ton/hr 950°C 750°C Normal carbon steel coil LPG HRS-RT (Honeycomb), 140 kW 1st heating zone
2nd heating zone (52 pair of HRS-RT) 88 m
Figure 24.21 Schema of continuous galvanizing and annealing line furnace equipped with 52 pairs of radiant tube burners. (Sudo, J., and Mochida, S., Internal paper on Radiant Tube Burner Testing, Yokohama, Japan: Nippon Furnace Co., Ltd., 2009. With permission.) Capacity Heating material Furnace temperature Fuel RT Draft Firing rate per burner
Cooling zone (Conventional RT burner)
:1.67 tons/hr :Joints :940°C (Max.) :Natural gas :6 inch, U type :Pull :81 kW
Heating zone, soaking zone (Regenerative RT burner 24 pairs) Material in
Material out 39 m Width: 2.2 m
Figure 24.22 Schematic drawing of a continuous roller hearth type annealing furnace equipped with 24 pairs of regenerative radiant tube burner. (Sudo, J., and Mochida, S., Internal paper on Radiant Tube Burner Testing, Yokohama, Japan: Nippon Furnace Co., Ltd., 2009. With permission.)
use the energy from flue gas for preheating the combustion air. The most important quality characteristics of radiant tubes are the maximum heat flux from the tubes to the process/product, the temperature uniformity at the surface of the radiant tubes, and the heat flux uniformity versus the length of the radiant tubes. Another very important quality characteristic is the combustion efficiency (fuel efficiency) of radiant tubes. The efficiency strongly depends on the flue gas outlet temperature of
© 2011 by Taylor and Francis Group, LLC
radiant tube systems. High efficiencies with less flue gas losses results in lower fuel consumptions and lead to a reduction of the CO2 emission per ton of product. Hazardous emissions like NOx, and unburned gases like CO should also be as low as possible. In order to test radiant tubes the systems have to be implemented in lab-scale furnaces in which one can achieve similar conditions like in industrial applications concerning process temperature, surface temperature of the radiant
504
Industrial Combustion Testing
tubes, and heat flux. The thermal input of industrial scale radiant tubes is in general less than 300 kW so that it is possible to test industrial scale radiant tubes at labscale furnaces. For the experimental investigations it is important to have a defined condition within the labscale furnace in order to allow a transfer of the achieved result to the industrial furnace. In the case of a comparison of different burners and heat recovery technologies it is important to have the same boundary conditions for the experiments in order to allow an exact comparison of the different tested technologies. With measured temperature profiles a thermal stress analysis of the radiant tube is possible on the basis of numerical simulation. Tests in industrial applications are also possible but not as easy without any disturbances of the production process. On an industrial applications general information about the performance of the radiant tubes, the whole furnace can be collected without disturbance of the process by measuring the flue gas outlet conditions of the radiant tubes and the fuel consumption of the furnace.
24.10 Nomenclature A CRS HRS f Q q R RT T TRP V Δ ε η σ
= = = = = = = = = = = = = = =
surface area, m2 conventional recuperative system high-cycle regenerative system correction factor energy rate, W heat flux, W/m2 ratio radiant tube temperature, °C temperature reference point volumetric flow rate, Nm3/h difference emissivity efficiency Stefan-Boltzmann constant
Subscripts MIN MAX A AV F FG FR S
= = = = = = = =
minimum maximum air average fuel flue gas furnace surface
© 2011 by Taylor and Francis Group, LLC
References
1. Wünning, J. G., and Milani, A. Handbook of Burner Technology for Industriol Furnace. Essen, Germany: Vulkan Verlag, 2009. 2. Milani, A., and Wünning, J. G. “Radiant Tube Technology for Strip Line Furnaces.” IFRF Combustion Journal, Article No 200405, ISSN 1562-479-X, November 2004. 3. Wünning, J. G. “Ceramic Radiant Tubes Extend Performance Limits.” Industrial Heating, March 2002. 4. Wünning, J. G. “Retrofitting Radiant Tube-Heated Furnaces.” Industrial Heating, June 2003. 5. Wünning, J. G. “Self Regenerative Burner for SingleEnded, P and Double-P Radiant Tubes.” Industrial Heating, June 2007. 6. Rafidi, N., and Blasiak, W. “Experimental Study: Thermal Performance of Ceramic Regenerative Heat Exchangers Used in HiTAC Regenerative Burning Systems.” Proceedings of the 25th Topic Oriented Technical Meeting of the IFRF, Stockholm, Sweden, 2003. 7. Szewczyk, D., Blasiak, W., Jewartowski, M., and Rafidi, N. “Increase of the Effective Energy from the Radiant Tube Equipped with High-Cycle Regenerative System (HRS) in Comparison with Conventional Recuperative System.” Proceedings of the 25th Topic Oriented Technical Meeting of the IFRF, Stockholm, Sweden, 2003. 8. Rafidi, N, Blasiak, W., Jewartowski, M., and Szewczyk, D. “Increase of the Effective Energy from the Radiant Tube Equipped with Regenerative System in Comparison with Conventional Recuperative System.” IFRF Combustion Journal, Article No 200503, ISSN 1562479X, 2005. 9. Canmet Energy Technology Centre. “Brochure: Test Bed for New Combustion Technologies.” 2008. http:// canmetenergy.nrcan.gc.ca/fichier.php/codectec/ En/2008-40/Test +Bed+for+Combustion.pdf (accessed November 5, 2008). 10. Sudo, J., and Mochida, S. Internal Paper on Radiant Tube Burner Testing. Yokohama, Japan: Nippon Furnace Co., Ltd., 2009. 11. Japan Industrial Furnace Manufacturing Association (JIFMA)/The Mechanical Social Systems Foundation (MSSF). Investigation Report for Safety Improvement in Heat Treatment Industries, 2007. 12. Szewczyk, D., Mörtberg, M., Rafidi, N., Dobski, T., and Blasiak, W. “Measurements of Temperature and Heat Flux in HiTAC Flame for Unsteady State Conditions.” 5th International Symposium on High Temperature Air Combustion, Yokohama, Japan, 2002. 13. Krishnamurty, N., Szewczyk, D., Blasiak, W., Manuel, S., and Hortawed, T. “Measurements of Temperature Using Fine Wire Compensated Thermocouple in Furnace Equipped with High-Cycle Regenerative System.” 25th Topic Oriented Technical Meeting of the IFRF, Stockholm, Sweden, 2003.
25 Metallic Mat Gas Combustion Giuseppe Toniato, Andrea Zambon, and Andrea D’Anna Contents 25.1 Introduction.................................................................................................................................................................. 505 25.2 Metallic Mat Combustion Technology..................................................................................................................... 505 25.2.1 Manufacture Technologies........................................................................................................................... 506 25.2.2 Internal Distributor....................................................................................................................................... 507 25.2.3 Material........................................................................................................................................................... 507 25.3 Metallic Mat Testing: Combustion, Noise, Pressure Drop, Creep, and Salt Fog Tests...................................... 509 25.3.1 Combustion Tests.......................................................................................................................................... 509 25.3.1.1 Comparison with Diffusive Flames........................................................................................... 509 25.3.1.2 Comparison with Other Premixed Surface Technology......................................................... 510 25.3.2 Noise Tests...................................................................................................................................................... 510 25.3.3 Pressure Drop Test........................................................................................................................................ 510 25.3.4 Creep Test........................................................................................................................................................511 25.3.5 Salt Fog Test.................................................................................................................................................... 512 25.3.6 Particle Emission Tests................................................................................................................................. 513 25.4 Combustion Chamber Design Issues........................................................................................................................ 515 25.5 Conclusions....................................................................................................................................................................516 References..................................................................................................................................................................................516
25.1 Introduction In the last few years gas combustion technology has undergone important innovations especially in residential heating appliances. In particular, requirements on low emissions together with load modulation has led to the use of premixed combustion technologies besides traditional diffusive flames. Constant demand for smaller overall dimensions and cost optimization has led to design combustion chambers with higher combustion intensity and this has led to premixed surface burners, then to radiant burners, and then to metallic mat burners in particular. With the metallic mat combustion flame front stabilizes above a metallic mat and at specific power loads is located inside. It differs from porous matrix combustion where combustion takes place inside a solid. If the initial calculations show that the convection heat exchange is not enough to establish the designed combustion intensity, the use of radiant burners becomes decisive. The radiative contribution of the mat is naturally predominant in the power range when flame develops within the mat itself.
Apart from the radiative contribution, the metallic mat has others functions: it increases the operating margin with regards to flame lift-off and flashback that can happen with the variation in excess air and gas type like the light-back limit gas (as defined in EN676). Metallic mat combustion is widely used in residential heating (at least in Europe) especially in those applications where condensation is sought. It is also used in process applications like bread ovens, paint ovens and where a radiant load is required. In the following, metallic mat combustion technology will be described with special reference to metallic mat production technology and material selection. Next, testing for such technology and especially combustion testing will be presented. At the end some design issues regarding combustion chamber coupled with this technology will be given.
25.2 Metallic Mat Combustion Technology Figure 25.1 compares a metallic mat combustion head (Figure 25.1a) with a standard surface premixed head 505
© 2011 by Taylor and Francis Group, LLC
506
Industrial Combustion Testing
(a)
(b)
Figure 25.1 Combustion head used to develop surface premixed flames: (a) combustion head made up of a metal mat laid on a perforated metal sheet sleeve, (b) combustion head made up of a perforated rolled sheet.
(Figure 25.1b) used for the same power application (35 kW) in a gas heating boiler. The figure on the right is made out of a perforated rolled sheet; within the cylinder that is visible in the figure, a second coaxial cylinder acts as a distributor, a flame trap and silencer. The mat technology showed on the left is made of a metal mat laid on a perforated metal sheet sleeve. In this solution there is no internal coaxial cylinder. The mixture of gas and air enters a cylinder closed at one end and exits through the small holes of the c ylinder lateral surface, traverses the mat and burns above it. Combustion on a metallic mat is not new. A search shows a patent by Osaka Gas Company Limited being filed in France on May 17, 1982. The first claim confirms [1] that it is a metallic mat burner: “burner with a metallic tissue over layed surface.” A more in-depth search shows another patent filed by Shell in 1985. This patent describes an “improved radiant surface combustion burner with a metallic porous element having a high oxidation resistance and thermal shock resistance at high temperature surface combustion conditions, combined with mechanical strength at room temperature” [2]. In particular, an alloy is mentioned whose registered trademark is FeCrAlloy. In the mentioned patent, reference is made to the fact that the alloy contains Yttrium, an element whose possible advantages we will examine later. Although this combustion technology is not new, its application on a wide scale has been hampered by the
© 2011 by Taylor and Francis Group, LLC
cost of the metallic mat. In reality, this cost is strictly connected to the manufacture technology. 25.2.1 Manufacture Technologies Different technologies are available to produce the metallic mat starting from fibers or from yarn. Mats made starting from a fiber are available on the market. This fiber is obtained by stripping a metal wire to about 10 micron diameter: fibers can then be either sintered or gathered in strands. In the case of sintered solution, the manufactured surface could be shaped as needed. The technology patented by Riello [3] and shown in Figure 25.1a provides for the production of a mat on a two-layer flat loom. This technique allows low production costs and a product that, once made, has a cylindrical shape. In this way it can be fitted onto the supporting cylinder and does not need the longitudinal welding many other products on the market have. The knitting loom does not weave a simple yarn but a strand with numerous yarns. In this way the porosity and flexibility needed by the application is achieved. With the same volume, the solutions made from fibers generally have a wider surface. Supposing the corrosive attack is proportional to the exposed surface, it can be assumed that the solution realized by weaving strands of yarns is stronger than the one realized starting from fiber. Moreover the presence of fiber can also be a critical element at high production volumes. New fiber tends to detach from the mat and quality problems in the field
507
Metallic Mat Gas Combustion
came arise from the grounding of the burner ionization probe due to of detached fiber. This problem is not present in the solution starting from yarns.
(a) Airtight aspiration circuit
25.2.2 Internal Distributor The metallic mat, either obtained from fiber or yarn, needs to be supported by an internal frame that could be planar or shaped as a cylinder [1]. This internal frame, besides supporting the mat, has two others important functions: it ensures an homogeneous distribution of the mixture and it functions as a flame trap. The geometry of the holes of the internal frame is a vital element to meet such requirements. Design criteria to avoid flashback is well known and based on the quenching diameter. If a hole has a diameter lower than the quenching diameter the flame will not propagate behind it. In case of a rectangular hole, the same criteria will apply with reference to the smaller dimension [4]. There is a technical limit to the hole dimensions due to the metal sheet thickness from which the internal frame is processed. Holes are obtained by “punching” and they have to be small so the metal sheet can not be very thick. Flashback can happen when the load or gas supply is changed. For example, critical conditions could take place when liquefied petroleum gas (LPG) is used at reduced loads. The LPG–air mixture is characterized by a higher flame speed compared to the methane–air mixture: at reduced loads the flame tends to enter the metal mat and heat the internal frame reducing quenching diameter. Flashback could happen when combustion parameters have been set up in a wrong way. In case the burner in the field is not correctly adjusted, the flame can be stabilized inside the sustaining frame destroying it. In such cases when flashback takes place, application safety is ensured by the use of proportional valves. In fact in case of flashback the counterpressure will close the valve. Mixing of air and gas is possible before or after the fan as shown in Figure 25.2. The second solution is safer because flame front. This is why such a solution is preferred with high-power applications and mixing before the valve is limited to low-power applications. 25.2.3 Material Materials used in metallic mat technology are Fe-Cr-Al alloys. One of the main worries about using metallic materials is the possibility of oxidation or even corrosive phenomena that may affect the functionality and sometimes even the continuity of the component. When components have to work at high temperatures, as in the case of combustion systems, the kinetics of oxidation and/or corrosion undergo considerable accelerations unless surface phenomena stimulate the formation of
© 2011 by Taylor and Francis Group, LLC
Combustion head with holed surface
Flame
Gas
Air (b)
Standard aspiration circuit (no airtight)
Gas
Air Figure 25.2 The two solution for mixing of air and gas: (a) mixing before the fan, (b) mixing after the fan.
long-lasting barriers that hamper further corrosion (formation of permanently stable corrosion-inhibiting barrier films). If corrosion is more specifically hot oxidation, it is obviously necessary to avoid the possible effects of differential expansion that can take place in the component because of two different constraints:
1. “External” constraints caused by the actions preventing the component from expanding (or contracting, when cooling) freely due to its morphology, its position in the equipment, and to the realization of the constraint conditions. 2. “Internal” constraints caused by the material in particular that take place because of the constraint between the surface layer (passive film) and the metal substrate.
The subdivision we propose here shows two ways of dealing with the complex problems resulting from hindered dilation. As for the former condition, it is appropriate to allow the component to adapt to the dimensional variations both during the warming up phase and the cooling phase. To this end it will be necessary to act on the morphology of the component and of the constraint as well as (if and when possible) on the temperature up
508
and down slope, but also (when possible) on a cautious choice of material, with particular attention to the curve of the thermal expansion coefficient according to the temperature. As far as the latter condition is concerned, it must be pointed out that the formation of superficial oxide films goes together with an increase in specific volume, which means an increase of the compression stresses in the superficial film parallel to the surface of the component, partly counterbalanced by a greater thermal expansion coefficient of the substrate to which the oxide film should be permanently and firmly anchored. While the component is cooling, however, the situation that contributed to mitigate the compression stress during the warming up phase and its staying at high temperature often leads to a compression stress overload on the film that, because of its high elastic modulus (if compact), would not be able to contract as much as the relatively high thermal expansion coefficient of the metal substrate requires. This leads to a condition of biaxial stress on the superficial film and, if the connection to the substrate is not good, to its scaling off. The Fe-Cr-Al alloys have been used for some time as heating elements (that is as material for resistors) for kilns, in competition with more expensive alloys with a high Ni content (Ni-Cr, Ni-Cr-Si, Ni-Cr-Fe-Si). The Fe-Cr-Al alloys normally have a chromium content of about 20–22% and an aluminum content of 5–5.5% while the content of yttrium, the most common alloying “auxiliary” element, is normally lower than 0.1%, if present. In comparison with the Ni-Cr and Ni-Cr-Fe alloys, the Fe-Cr-Al alloys have a lower linear thermal expansion coefficient. Mechanical resistance is sufficiently high for the alloys containing 22% of chromium and with an aluminum content of 4.5–5.5% and, therefore, 72.5–73.5 of iron. The above mentioned Fe-Cr-Al alloys have a fairly high melting point (about 1500°C) and a recommended maximum operating temperature of 1280°C and 1375°C, respectively, for the ones with a lower and higher aluminum content. High resistance to oxidation at high temperatures is due to the formation, in the presence of oxygen, of a protective layer of oxide firmly attached to the substrate. At the temperature of about 1200°C this layer of oxide—that at ambient temperature is greywhitish—is formed by about 95% Al2O3, 3% Cr2O3, and 2% Fe2O3. Under normal operating conditions the chemical stability of the oxide layer is satisfactory provided there is no contact with certain refractories at temperatures higher than 980°C. The use of sulfur compounds must be avoided in the refractories and in the cement. In applications with cyclic temperature variations some alloys can undergo progressive lengthening and reduction of the cross section at the same time.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
The Fe-Cr-Al alloys can undergo this phenomenon even without any external forces. A common, even if not totally proven, explanation is that first the alloy undergoes oxidation at a high temperature; while cooling, the high resistance to compression of the superficial layer of oxide would push the metal core (less resistant) to lengthen realizing a plastic deformation; when heated again, because of the differential thermal expansion of the metal core and the oxide superficial layer, the limited resistance to a tensile stress of the latter would cause it to crack when subjected to traction, producing a new localized oxidation process. This cyclic mechanism would cause the metal element to lengthen progressively. It is therefore important for the Fe-Cr-Al alloys to have a suitable support in order to minimize the effects of cyclic growth. The Fe-Cr-Al alloys are normally used in oxidative atmospheres at temperatures between 850°C and 1350°C; reducing atmospheres cause negative effects unless the components have been previously air-oxidized. Their use is not recommended in reducing atmospheres and even less in carburizing atmospheres. High reactivity elements (RE; e.g., cerium, yttrium, zirconium, hafnium) are sometimes added to the Fe-Cr-Al matrix; these help the formation of the alumina protective layer; that is, they speed up the transition from the less to the more stable crystallographic lattices [5,6] and increase its adhesion to the substrate. Secondly this action is assisted by the precipitation of “pegs” made up by the oxides of the reactive elements (RE), partially immersed both in the substrate and in the scale of continuous superficial oxide [6]. However, the same authors state that the formation of the pegs is not vital for the resistance to the scaling off of the layers of superficial oxide. It is important to note that its crystallographic type is α-Al2O3, which is much more effective than the δ, γ, or θ types. It is also vital that a constant supply of aluminum is guaranteed through the diffusion from the substrate in order to ensure the surface is totally covered with alumina. Otherwise Cr2O3 and Fe-Cr spinel would be obtained mainly. The addition of the already mentioned RE in contents of less than 0.1% increase the resistance to oxidation of the Fe-Cr-Al alloys. Moreover the alumina grains are reduced in size. It seems that traces of sulfur and, in part, carbon present in the alloy tend to accumulate on the substrate–protective film interface. The increase in the tendency to form pores is attributed to this and the consequence is a decrease in the adhesion of the oxide. According to some studies, it is possible that the effect of the RE added is their tendency to react with the carbon and the sulfur, binding them in the substrate and lowering the activity of the carbon and
509
Metallic Mat Gas Combustion
for each application while changing power range and excess air. Figures 25.3 and 25.4 show the comparison in CO and NOx emissions measured at the stack when the excess air changes at a specific power. The data refer to a point in the working field (26 kW) but the observations are valid for other working points. It is worth noting that the metallic mat burner is normally set to a value of excess air corresponding to 8.5% CO2, while a diffusive flame burner of the type tested is set to an excess air corresponding to a CO2 of about 9.5%. In its adjustment range the metallic mat head obtains NOx values that cannot be achieved with the diffusive flame. To complete the data, the working point at 10% of CO2 has been added, but it must be considered outside the working field. The low NOx emissions of diffusive heads for such excess air can be correlated to high CO emissions (compare Figure 25.3).
25.3 Metallic Mat Testing: Combustion, Noise, Pressure Drop, Creep, and Salt Fog Tests
80 70 60 50 40 30 20 10 0
First of all it is interesting to compare metallic mat combustion emission to a diffusive flame emission. Tests are performed on the same boiler just changing the burner and comparing the premixed metallic mat head with two diffusive burners: a blue flame burner (or Class 3 according to the EN676 regulation) and a yellow flame burner (or Class 1 according to the same regulation). The CO and NOx emissions are measured
© 2011 by Taylor and Francis Group, LLC
8.1
CO2 (%)
9
Blue diffusive flame
10 Yellow diffusive flame
Figure 25.3 (See color insert following page 424.) CO emissions with three types of gas burners coupled with the same 26 kW cast iron commercial boiler. 200 180 160 140 120 100 80 70 60 40 20 0
Nox emissions (mg/kWh)
25.3.1.1 Comparison with Diffusive Flames
7.4 Premix
25.3.1 Combustion Tests In the heating sector, diffusive burners have dominated the market; even now most liquid fuel burners use this technology. As far as gas is concerned, the demand for lower emissions has led to the development of premixed flame burners. It is interesting to note that this technology was developed in parallel with the spreading of condensing applications. For this reason, it is interesting to compare combustion emissions of the metallic mat technology to diffusive flame emissions and with other premixed combustion technologies.
CO emissions at 26 kW
90 CO emissions (mg/kWh)
sulfur in the alloy. Up to now these interpretations are not considered as sufficiently proven since the interlayers between the scale and the substrate are often extremely thin and therefore not easy to study. There has been experimental confirmation by Moon [5], carrying out comparative tests on Fe–Cr–Al alloys with a different sulfur content—of the oxide superficial layer being more resistant for alloys with an yttrium content higher than 0.2%. It cannot be excluded that in realizing higher deformation percentages, alloying, even with trace elements, might cause difficulties connected with decreases in ductility of the alloys in comparison with the unalloyed homologous, a problem that should be taken into consideration in the manufacturing of products. The assessment on whether to exploit the supposed positive effect of RE on the “endogenous” sulfur-pollution of the alloy or to proceed to a careful selection of the materials in charge for metallurgic processing, concerns financial and technological aspects that go beyond the scope of this chapter. It is thought that problems might arise following the “exogenous sulfur” pollution due to contamination of the environment where the alloy worked at high temperatures.
NOx emissions at 26 kW
Limit class 5 EN 483
7.4 Premix
8.1
CO2 (%)
Blue diffusive flame
9
10 Yellow diffusive flame
Figure 25.4 (See color insert following page 424.) NOx emissions with three types of gas burners coupled with the same 26 kW cast iron commercial boiler.
510
Industrial Combustion Testing
25.3.1.2 Comparison with Other Premixed Surface Technology The mat heads combustion emissions could also be compared to a premixed metal sheet solution. With this target the two combustion heads shown in Figure 25.1 were tested on a wall-hang boiler. Figures 25.5 and 25.6 show the emission values taken at two different heat ratings as the excess of air changes. It is worth noting the margin that mat solutions offer with regards to CO emissions. It is obvious that when the chamber becomes smaller and the combustion intensity increases this becomes increasingly more important. As was mentioned previously, the CO emissions at 10% CO emissions at 8 kW and 35 kW
300 250 200 150 100 50 0
8
8.5
CO2 (%)
9
10
Mat combustion head 35 kW Mat combustion head 8 kW
Metal sheet 35 kW Metal sheet 8 kW
Figure 25.5 CO emissions in a condensing wall-hung boiler with two different types of heads.
A metallic mat is characterized by the specific pressure drop test. For this purpose an experimental device can
NOx emissions (mg/kWh)
120 100 80
110
Class 5 EN 483: 70 mgr/kWh
Noise emissions at the stack CO2 8%
100
60 40 20 0
One of the main benefits of a premixed head in comparison with a diffusive head is the low noise emission at the stack. A premixed burner is generally quieter than a diffusive burner because of the lower combustion turbulence, on the condition there is no instability in the combustion, an instability that is possible with diffusive flames too. Figure 25.7 shows the comparison between the noise at the stack of the configurations previously examined. Like gas emissions, noise measurements are influenced by the combustion chamber type. Regarding sound emissions at the stack, the mat head still exhibits lower emissions in comparison with the metal sheet head as seen in Figure 25.8. It is possible to change the acoustic behavior of the system modifying the internal frame holes distribution. Premixed flames are subjected to more acoustic instabilities compared to diffusive flames: changing dimensions and positions of the holes one can change frequency at which heat is released and avoid unpleasant couplings with combustion chambers and attached chimneys. Transient acoustic phenomena, as could happen at start up, can be managed in the same way, changing hole geometry on the internal frame. Sometimes it is possible to solve acoustic problems fitting specifically designed sectors or diaphragms inside the cylinder. 25.3.3 Pressure Drop Test
NOx emissions at 8 kW and 35 kW
140
25.3.2 Noise Tests
8
8.5
CO2 (%)
9
10
35 kW metal sheet head
35 kW mat head
8 kW metal sheet head
8 kW mat head
Figure 25.6 NOx emissions in a condensing wall-hung boiler with two different types of heads.
© 2011 by Taylor and Francis Group, LLC
Noise pressure dbA
CO emissions (mg/kWh)
350
of CO2 are to be considered outside the working field. In this case NOx values are lower with the metal sheet head due to the higher CO emissions of the same head (see Figure 25.6).
90 80 70 60
Premix
Blue diffusive flame
Yellow diffusive flame
Figure 25.7 Noise emissions at the stack with three types of gas burners coupled with the same 26 kW cast iron commercial boiler.
511
Metallic Mat Gas Combustion
Noise pressure at the stack (CO2 at 9%)
110
Acoustic pressure dbA
100 90 80 70 60
8 kW
35 kW
Metal sheet combustion head
Mat combustion head
Figure 25.8 Noise emissions at the stack in a condensing wall-hung boiler with two different types of heads. Specific pressure drop
1400 1200
(mbar/m2)
1000 800 600 400 200 0
Woven fabric fiber
Sintered fiber
Riello’s mat
Figure 25.9 Specific pressure drops on a ϕ40 mm test sample in a specific test setup.
be realized in order to compare the losses of different technologies. Figure 25.9 shows the comparison between a sample developed in Riello, a sample made in woven fabric, and a sample made in sintered fiber. 25.3.4 Creep Test The multithread mat needs appropriate checks with regards to overall dimensional stability. In fact, it is extremely important that, for an even distribution of the gas, there are no excessive inflexions among the points where the mat is supported so that there are no “pockets” that might cause local unevenness in temperature or the flame passing under the mat causing uneven running. The use of tables and charts describing the creep of the material does not allow an accurate forecast of its
© 2011 by Taylor and Francis Group, LLC
behavior with regards to the configuration of the flame support under study; it has been necessary to carry out comparative tests at the temperature in a muffle furnace. Mats have been tested having the shape of horizontally positioned bands blocked at the ends in a rigid frame suitably constructed to be inserted into the chamber of a muffle furnace. The test was aimed at comparing the performance (in terms of dimensional stability at temperature) of the different threads (the thread used on the Riello heads has a circular cross section while the one used in the comparison, manufactured by a different producer, had a nearly rectangular cross section) and of the type of mat (either cross-ply woven or meshed, respectively). Bands taken along the direction of the warp and of the weft have been used (therefore with a reciprocally orthogonal orientation) in the condition of a band subject only to its own weight, loaded with masses of 0.25 kg (and therefore with an overloading force equal to 2.45 N) on the centerline and with 0.5 kg (overloading force equal to 4.90 N). The deflection of the bands has been noticed and the lengthening evaluated on frames taken by a thermocamera at fixed intervals of 3, 46, and 100 hours stay at temperatures of 750°C and 900°C. Figure 25.10 shows two frames concerning the initial and final situations of the test carried out at 750°C, with the warp oriented according to the length of the strip. In Figure 25.11 two frames are shown concerning the test carried out on the mat manufactured by a different producer, where the load and temperature conditions are the same as for Figure 25.10. As can be noticed, considering the dimensional stability with temperature the design developed by Riello clearly outruns that of the other type of mat. It is quite clear that the dimensional stability is a must for the evenness and stability of the combustion process on the head of the burner. It is worth noting that already during the rising of the temperature of the kiln, the mat from a different producer undergoes remarkable elongations in the 4.9 N load condition, which causes the mat to touch the frame. It is quite impressive the difference in the behaviors of the two kinds of mats: while in the case of the mat produced by Riello, the deformation evaluated both during the test and at the end of it appears to be absolutely acceptable (even at 900°C for 100 h the maximum sagitta was of a few tenths of a millimeter), it appears quite remarkable in the case of the mat from a different producer (at just 750°C for 100 h under the 4.9 N load the sagitta was nearly 6 mm, while at 900°C for 100 h it was 21 mm, despite the further support due to the contact with the frame). As can be noticed in Figure 25.12, the deformation of the mat from a different producer appears to be permanent, even after cooling down to room temperature.
512
Industrial Combustion Testing
(a)
(b)
Room temperature, 0 h
Temperature 750°C, 100 h
Figure 25.10 Arrangement of the Riello samples in woven fabric, on the supporting frame inserted in the muffle kiln for the lengthening tests under temperature stress: (a) initial situation: in the foreground a 4.9 N overloaded fabric, in the background a 2.45 N overload, way in the back an unloaded fabric (own weight), (b) situation after 100 hours at 750°C. (a)
(b)
Room temperature, 0 h
Temperature 750°C, 100 h
Figure 25.11 Arrangement of the mat samples (manufactured by a different producer), on the supporting frame inserted in the muffle kiln for the lengthening tests under temperature stress: (a) initial situation: in the foreground a 4.9 N overloaded fabric, in the background a 2.45 N overload, way in the back an unloaded fabric (own weight), (b) situation after 100 hours at 750°C.
25.3.5 Salt Fog Test As mentioned in relation to hot corrosion resistance, one of the main attack conditions for Fe-Cr-Al alloys with the eventual addition of RE class elements, comes from endogenous contaminations, typically from sulfur, or exogenous, from carbon or sulfur. Generally the main sources of these elements can be found in fuels, as the basic constituent (in the first case in methane or deriving from odorizers in the second case), although we cannot exclude possible contaminations of the metal surface following manipulations. It seems hazardous to realize a test that reproduces the working conditions in relation to the possible variations
© 2011 by Taylor and Francis Group, LLC
in the operative conditions as regards to temperatures, mixing relations, and presumably also in terms of possible variations of the sulfur vectors, considering the possible mixture in the gas fuel supply and the possible variations in the percentage of odorizers. Only a field test carried out in typical working conditions will allow a correct evaluation in this case. That having been said, just in order to evaluate the influence of the morphology of the threads in the possible corrosion kinetics, the mats for the Riello head and for different produced heads, underwent accelerated corrosion tests in a saline mist (in accordance with the ASTM B117), with the warning that, in this ambit, the
513
Metallic Mat Gas Combustion
The examinations carried out with a stereomicroscope, made on the Riello sample (sample 3) and on the different produced sample (sample 4), highlighted a corrosion phenomenon localized by pittings (“corrosion pits”) in sample 4, which brought to the fracture of some threads. None of these phenomena were observed in Riello sample number 3 and, a fortiori, in Riello’s samples 1 and 2. 25.3.6 Particle Emission Tests
Room temperature - after completed cooling down Figure 25.12 Arrangement of the mat samples (manufactured by a different producer): situation after cooling to room temperature, after 100 hours at 750°C.
test can be attributed only a valence, respect to the different configurations, in terms of the morphology of the types of threads utilized (respectively, with a cylindrical and rectangular section), and not of the operative conditions during the working phase. The exposures were protracted, in order to allow differentiating the behavior of the different types of threads, up to 1000 hours. Figures 25.13a, b, and c show the different behaviors of the alloys used and their configuration. From the image sequence, grouped according to the progressively increasing time of permanence in the saline mist (after 100 hrs, 500 hrs, 800 hrs), you can observe the following: • After 100 hours (Figure 25.13a), the samples realized by Riello with circular section threads did not show any corrosive phenomenon, while a widespread corrosion occurred on the different producer mats. • Only after 800 hours (Figure 25.13c) on the Riello mats you can observe a corrosive phenomenon, but only on the mat that was intentionally charged with RE (sample 3). • After 1000 hours, as regards to the Riello samples, you can notice a mild corrosive phenomenon on sample 1, while nothing was noticed on sample 2. The situation in sample 3 did not show any substantial variations in the morphology of the attack. Even the different producer sample (samples 4 and 5) did not show any substantial variations in the morphology of the attack with respect to the situation after 800 hrs.
© 2011 by Taylor and Francis Group, LLC
Metallic mat gas combustion occurs in operating conditions (i.e., premixed combustion with large excess air), which do not allow particle formation. However local conditions, far from the ideal ones, and metal catalytic effects could promote the undesirable formation of particulate matter and its emission in the atmosphere. Measurements performed at the exhausts of commercial residential heating devices have shown that particulate matter emission is negligible below the detection limit of instruments based on mass measurement. However, there is increasing evidence that the concentration of the “number“ of particles (related to their size) rather than their cumulative mass might be responsible for the observed effects of particulate on health and the environment. For these reasons it is important to control the emission of the number of particles of specific sizes, rather than their mass from combustion systems that are widely used, such as residential burners. In particular, the interest should be focused on the ultrafine emitted particles, those with mean sizes below 100 nm that, because of the low sizes, do not contribute massively to mass emission. To this aim, the three configurations of residential heating devices whose combustion test results are presented in previous figures have been analyzed with respect to particle emissions: two combustion heads equipped with premixed burners differentiated by metallic mat and perforated cylindrical heads and one equipped with a blue diffusive flame consisting in a five-tube injector where blue flames of gaseous hydrocarbons can be stabilized. The three flames have been characterized by in-situ optical diagnostics for the identification of the flame structures and the formation of particulate matter, whereas particle emission has been determined by scanning mobility particle sizer (SMPS) and spectroscopic characterization of sampled material. Laser induced emission techniques with sensitivity to mass concentrations down to 0.5 ppb have been used. They are based on the measurement of the laser induced fluorescence and incandescence from the particles. The size distribution functions of the particles in the exhaust pipe have been determined by measuring the differential mobility of charged particles with a nano-differential mobility analyzer TSI Model 3936 SMPS. This apparatus is specifically designed to measure particles in
514
Industrial Combustion Testing
(a)
1
2
3
4
5 100 h
(b)
1
2
1
2
3
4
5 500 h
4
5 800 h
(c)
3
Figure 25.13 Aspect of the various types of mats after the exposure in saline mist for (a) 100 hours, (b) 500 hours, (c) 800 hours; the samples marked with 1 and 2 are made in a Riello mat in a Fe-Cr-Al alloy without an intentional addition of RE. Sample 3 is made in a Riello mat in a Fe-Cr-Al alloy with an intentional addition of RE; samples 4 and 5, respectively, the “longitudinal” and “transversal” orientation of the different producer mats.
the 3–50 nm range. The SMPS system consists of a diffusion charger (Kr-85 Bipolar), a nano-differential mobility analyzer (NDMA, type TSI 3085), and an ultrafine condensation particle counter (UCPC, type TSI 3025A). The SMPS was operated in high-flow mode (aerosol flow set at 1.5 l/min and sheath flow at 15.0 l/min). A water-based sampling technique has been used to collect emitted ultrafine particles. Combustion products, sampled by a probe in the exhaust pipe,
© 2011 by Taylor and Francis Group, LLC
are cooled in order to condense combustion water and drawn through a reservoir containing deionized water, and placed in an ice bath. This sampling procedure allows the collection of very small organic carbon aerosol that have more affinity with water, with respect to larger soot particles. Water samples are then analyzed by spectroscopic measurements. This technique measures ultrafine particle concentrations down to 10 µg/Nm3.
515
Metallic Mat Gas Combustion
1.E+08
dN/dlogDp, cm–3
1.E+07 1.E+06 1.E+05 1.E+04 1.E+03
1
10 DP, nm
100
these species in a water sample has been estimated at 16 kW rated burner power. The three burner configurations form very low concentrations of ultrafine organic particles. The emitted mass concentration of combustion-formed ultrafine particles is about 0.03 mg/Nm3 for the metallic mat burner and increases to about 0.05 and 0.2 mg/Nm3 for the perforated and diffusive burners, respectively. Nano-sized particles are formed in larger concentrations in the flame region but are also strongly oxidized in the secondary oxidation region of the flame, and also in the high temperature flame confinement in a practical condensing boiler.
Figure 25.14 Size distribution function of the particles in the exhaust pipe as measured by SMPS analysis (16 kW and 19% excess air metal premix; drilled premix; diffusive, continuous line ambient air).
On-line standard measurements have also been performed on the exhaust gases for the determination of the concentrations of CO, unburned hydrocarbons, and NOx. The CO2 measurements have been used for the determination of exhaust gas dilution in the exhaust pipe. Figure 25.14 shows the size distribution functions measured at 16 kW rated burner power and 19% excess air in the three configurations (metallic mat, perforted cylinder, and blue diffusive). Emitted particle diameters range from 2 nm, the limit of the detection system, up to 100 nm. In the same figure the size distribution function of the particles collected in the ambient air is also reported. In the 10–100 nm range, typical of soot primary particles, the number concentration of the particles measured at the exhaust of the combustion system is of the same order of magnitude of the number concentration of the particles present in the ambient air and it is reported in the figure as a continuous line. We can, therefore, conclude that the combustion of natural gas in residential heating systems doesn’t produce particles larger than 10 nm. This is in perfect agreement with the optical measurements that show that in all the examined conditions, laser induced incandescence signals due to particles larger than 10 nm are not measured. Below 10 nm, the particle concentration in ambient air is below the instrumental detection limit while a large concentration of very small organic particles are measured at the burner exhausts. Due to their very low sizes, such particles do not contribute to the particulate emission (mass-based concentration). Emissions of particles from a mat premixed burner are lower compared to the emissions from drilled cylinder combustion heads and lower also compared to the emissions from diffusive flame combustion heads. Measurements of organic carbon have been performed in the exhaust pipe by collecting the particles in the condensed combustion water. A concentration of
© 2011 by Taylor and Francis Group, LLC
25.4 Combustion Chamber Design Issues The heat exchange achieved in the combustion chamber by a premixed flame shall be generally different from the one achieved by a diffusive flame; in fact, the thermal gradients achieved will be different. It is therefore necessary to design specific combustion chambers for premixed flames and to proportion the heat exchange downstream accordingly so as to ensure that the flue gases come out at the temperature required. Trying to match premixed burners with old boilers designed for diffusive flame could cause an efficiency loss. Also the higher excess air to which the premixed flame burners are set turns against efficiency. This loss in efficiency is however negligible when flue gases go out at the low temperatures allowed by condensing applications. In resizing the boiler the designer can drastically reduce the dimensions of the combustion chamber due to the higher combustion intensity and the wellidentified flame front not fluctuating in space. If a diffusive flame needs wide volumes as suggested by the EN676 regulation (refers to the dimensions of test tubes), the combustion chamber of a premixed flame can be just a little longer than the combustion head and have a diameter that is about twice the diameter of the head. It seems that the possibility to contain combustion chamber dimensions could be a very attractive argument for boiler producers. To compare the dimension of a premixed mat head with a metal sheet head, tests have been carried out on a combustion chamber of the latest generation of condensing wall-hang boilers. The compared combustion heads are shown in Figure 25.15a. The mat head is 120 mm long and its diameter is 50 mm, while the metal sheet head corresponding dimensions are 120 mm and 70 mm. Figure 25.15b shows a similar comparison between two heads, a metal sheet one and a mat one developed for
516
Industrial Combustion Testing
(a)
(b)
Figure 25.15 Overall dimensions of mat combustion head and metal sheet one: (a) 35 kW combustion heads, (b) 50 kW combustion heads.
a 50 kW application. In this case the combustion area of the mat head is 120 mm long with a 60 mm diameter, while the metal sheet solution is equipped with a combustion area 250 mm long and having a diameter of 70 mm. Therefore the figures highlight the possibility of developing smaller combustion heads using mat heads. It is important to note that the possibility to reduce combustion head dimensions for a certain heating power by exploiting the higher superficial combustion intensity offered by the mat solutions is quite useful in power applications also. With a heat rating higher than 100 kW, the heads undergo such structural stress due to localized thermal gradients that the use of metal sheet heads is prohibitive.
for progressively smaller units and therefore the need to increase the combustion intensity. Apart from the radiative contribution, the metallic mat has others functions: it increases the operating margin with regards to flame lift-off and flashback that can happen with the variation in excess air and gas type. Hot deformation tests on the mat demonstrate that the configuration of the mat starting from yarns allows minimizing permanent deformations that are caused by high temperatures. On the other hand, the circular section of the thread seems to guarantee an improved durability with respect to the rectangular sections.
References
25.5 Conclusions The possibility to contain combustion chamber dimensions, low NOx and CO emissions, and last but not least, achieve low noise intensities in the stack are the main reasons for which the application of premixed surface flames are substituting the solutions with diffusive flames in residential gas heating systems. The different flame front of the premixed flame with respect to the diffusive one allows to design more compact combustion chambers. Tests performed on a wall-hung boiler proved that among the applications of the premixed surface flame, metallic mat heads allow reductions of size, emissions, noise in the stack and improvements of flame stability margins. The element that led the designers to adopt metallic mat combustion head is the need
© 2011 by Taylor and Francis Group, LLC
1. Okabayashi, M., Takeishi, Y., Higuchi, Y., and Taguchi, K. “Bruleur à combustion en surface alimentè totalement en air primaire,” Patent #FR 2526919, Osaka Gas Company, 1984. 2. McCausland, D. A. C., Shirvill, L. C., and Coles, K. F. “Radiant Surface Combustion,” Patent #EP 0157432, Shell Canada, 1989. 3. Lovato, A., and Toniato, G. “Cover Member for Gas Combustion Heads, and Gas Burner Comprising Such a Cover Member,” Patent #EP 1544542, Riello Spa, 2005. 4. Turns, S. An Introduction to Combustion. New York: McGraw-Hill, 1996. 5. Cueff, R., Buscail, H., Caudron, E., Issartel, C., and Riffard, F. “Influence of Yttrium-Aloying Addition on the Oxidation of Alumina Formers at 1173k.” Oxidation of Metals 58, nos. 5/6 (December 2002): 439–55. 6. Moon, D. P. “Role of Reactive elements in Protection.” Materials Science & Technology 5 (August 1989): 754–64.
26 Performance Prediction of Duct Burner Systems via Modeling and Testing Steve Londerville Contents 26.1 Introduction.................................................................................................................................................................. 517 26.2 Applications.................................................................................................................................................................. 518 26.2.1 Cogeneration.................................................................................................................................................. 518 26.2.2 Combined Cycle............................................................................................................................................. 518 26.2.3 Air Heating..................................................................................................................................................... 519 26.2.4 Fume Incineration......................................................................................................................................... 519 26.2.5 Stack Gas Reheat............................................................................................................................................ 519 26.3 General Modeling and Performance Prediction..................................................................................................... 519 26.3.1 1D/2D Mathematical Correlations.............................................................................................................. 519 26.3.2 Physical Nonreacting (PNR) Scale Models................................................................................................ 520 26.3.3 Physical Reacting Test (PRT)........................................................................................................................ 520 26.3.4 Computational Fluid Dynamics (CFD)...................................................................................................... 520 26.3.5 Nonreacting Flow Models............................................................................................................................ 521 26.3.5.1 Reacting Flow Applications........................................................................................................ 522 26.3.5.2 Recommended Modeling and Performance Prediction Methods........................................ 522 26.4 Chemistry and Emission Calculations for Gas-Fired Applications..................................................................... 522 26.5 Emissions...................................................................................................................................................................... 524 26.6 Chemical Kinetics........................................................................................................................................................ 524 26.7 Performance and Prediction....................................................................................................................................... 526 26.7.1 TEG/Combustion Air Flow Distribution................................................................................................... 526 26.7.2 Temperature Profile....................................................................................................................................... 526 26.7.3 Stability........................................................................................................................................................... 526 26.7.4 Emissions........................................................................................................................................................ 527 26.8 Summary....................................................................................................................................................................... 527 References................................................................................................................................................................................. 528
26.1 Introduction A duct burner system can be loosely described as large cross-sectional ducts with high flows that require uniform heat addition for relatively small temperature increases. The flows can be air, fumes, or oxygendepleted streams. The burners used are also called ribbon burners, linear burners, or duct burners designed so that the heat input can be distributed over a relatively large cross section. See Figure 26.1.
Linear and in-duct burners were used for many years to heat air in drying operations before their general use in cogeneration systems. Some of the earliest systems premixed fuel and air in an often complicated configuration that fired into a recirculating process air stream. The first use in high temperature depleted oxygen streams downstream of gas turbines in the early 1960s and were used to provide additional steam for process use in industrial applications and for electrical peaking plants operating steam turbines. As gas turbines have become larger and more efficient,
517 © 2011 by Taylor and Francis Group, LLC
518
Industrial Combustion Testing
Figure 26.1 Typical duct burner assembly. 2. Phase II construction at Dabhol has been under way since October
duct burner supplemental heat input has increased correspondingly [1].
26.2 Applications 26.2.1 Cogeneration Cogeneration implies simultaneous production of two or more forms of energy, most commonly electrical (electric power), thermal (steam, heat transfer fluid, or hot water), and pressure (compressor). The basic process involves combustion of fossil fuel in an engine (reciprocating or turbine) that drives an electric generator coupled with a recovery device, which converts heat from the engine exhaust into a usable energy form. Production of recovered energy can be increased independently of the engine through supplementary firing provided by a special burner type known as a duct burner. Most modern systems will also include flue gas emissions control devices. A typical installation of a duct burner is shown in Figure 26.2. The unique requirements are uniform temperature distribution and low emissions. Reciprocating engines (typically diesel cycle) are used in smaller systems (10 MW and lower) and offer the advantage of lower capital and maintenance costs but produce relatively high levels of pollutants. Consideration of HC’s and soot reduction are complex requiring low temperature oxidation kinetics that must be very accurate. Turbine engines are used in both small and large systems (3 MW and above) and, although more expensive, generally emit lower levels of air pollutants.
© 2011 by Taylor and Francis Group, LLC
Figure 26.2 General Overview of a HRSG. (Courtesy of Coen Company.)
Fossil fuels used in cogeneration systems may consist of almost any liquid or gaseous hydrocarbon, although natural gas and various commercial grades of fuel oil are most commonly used. Mixtures of hydrocarbon gases and hydrogen found in plant fuel systems are often used in refining and petrochemical applications. Duct burners are capable of firing all fuels suitable for the engine/turbine, as well as many that are not, including heavy oils and waste gases. Heat recovery for large systems is accomplished by means of a boiler (commonly referred to as a “heat recovery steam generator” [HRSG]). Smaller systems utilize either a steam or hot water boiler, or alternatively, some type of air-to-air heat exchanger. Supplementary firing is often incorporated into the boiler/HRSG design as it allows increased production of steam as demanded by the process. The device that provides the supplementary firing is a duct burner, so called because it is installed in the duct connecting the engine/turbine exhaust to the heat recovery device, or just downstream of a section of the HRSG superheater. Oxygen required for the combustion process is provided by the turbine exhaust gas (TEG). Depending on the level of O2 concentrations, generation of CO and HC are a significant issue in addition to possible stability problems due to flammability limits. 26.2.2 Combined Cycle Combined cycle systems incorporate all components of the simple cycle configuration with the addition of
Performance Prediction of Duct Burner Systems via Modeling and Testing
a steam turbine/generator set powered by the HRSG. This arrangement is attractive when the plant cannot be located near an economically viable steam user. Also, when used in conjunction with a duct burner, the steam turbine/generator can provide additional power during periods of high or peak demand. 26.2.3 Air Heating Duct burners are suitable for a wide variety of directfired air heating applications where the physical arrangement requires mounting inside a duct, and particularly for processes where the combustion air is at an elevated temperature and/or contains less than 21% oxygen. Examples are: • Fluidized bed boilers installed in combustion air ducts and are only used to provide heat to the bed during startup. At cold conditions, the burner is fired at maximum capacity with fresh ambient air, but as combustion develops in the bed, cross exchange with stack gas increases the air temperature and velocity. Burners are shut off when the desired air preheat is reached and the bed can sustain combustion unaided. • Combustion air blower inlet preheat: Burners are mounted upstream of a blower inlet to protect against thermal shock caused by ambient air in extremely cold climates (–40°F/°C and below). This arrangement is only suitable when the air will be used in a combustion process as it will contain combustion products from the duct burner. • Drying applications where isolation of combustion products from the work material is not required, such as certain paper and wallboard manufacturing operations. 26.2.4 Fume Incineration Burners are mounted inside ducts or stacks carrying exhaust streams primarily composed of air with varying concentrations of organic contaminants. Undesirable components are destroyed both by an increase in the gas stream bulk temperature and through contact with localized high temperatures created in the flame envelope. Particular advantages of the duct burner include higher thermal efficiency as no outside air is used, lower operating cost as no blower is required, and improved destruction efficiency resulting from distribution of the flame across the duct section with (grid-type design). Kinetic destruction rates are relatively slow and residence time is a critical factor in addition to temperature. See Section 26.6 on kinetics.
© 2011 by Taylor and Francis Group, LLC
519
26.2.5 Stack Gas Reheat Mounted at or near the base of a stack, heat added by a duct burner will increase natural draft, possibly eliminating a need for induced draft or eductor fans. In streams containing a large concentration of water vapor, the additional heat can also reduce or eliminate potentially corrosive condensation inside the stack. Often a source of ambient augmenting combustion air is added if the stack gas oxygen concentration is low.
26.3 General Modeling and Performance Prediction Combustion systems are complex involving fluid flow, complex mixing, chemical reactions, and all modes of heat transfer. The most reliable information for performance prediction of such systems is by actual full-scale measurements. Measurements utilizing fullscale tests in many cases can be prohibitively expensive, difficult, and sometimes impossible. Alternatives are small-scale reacting tests, scale physical models, and mathematical models ranging from simple one-dimensional (1D) correlations to threedimensional (3D) computer solutions using computational fluid dynamics (CFD). The selection of the correct simulation or model is driven by the information needed coupled with the models ability to accurately predict the full-scale results from the simulation. For instance, prediction of a piping system pressure loss is usually modeled with a 1D mathematical model using the Darcy pressure loss model and coefficients obtained from the Moody diagram [3]. This is acceptable and appropriate because the model has accurately predicted pressure loss performance in the past and is low cost. For a duct burner system, CFD, physical modeling, scale reacting, full-scale, and other models can be used for various parts of the system where appropriate. 26.3.1 1D/2D Mathematical Correlations Frequently an equipment vendor has access to a large amount of performance data for a particular combustion equipment component such as a duct burner. Using simple to complex correlations the vendor in many cases can correlate predicted performance based on many variables and generate a function or code to predict performance with new and varied applications. For example, for many years Coen Co. (Foster City, CA) uses a NOx correlation to predict NOx emissions for various burner types in an array of furnace arrangement with a variety of fuels with great accuracy 11]. Using a large
520
amount of data and correlating multiple variables, it is possible to include a statistical analysis to add confidence to the prediction. It is common in the industry to successfully use this modeling technique to predict flame shape, NOx, CO, particulate, opacity, furnace heat absorption, and many other performance parameters. 26.3.2 Physical Nonreacting (PNR) Scale Models These types of simulations utilize clear plastic scale geometries of flow sections to simulate flow patterns. The working fluid can be water or air and the scale ranges from 1/5 to 1/20th. Geometric and kinematic similitude is usually maintained. Dynamic similitude is generally ignored. It is common to use this technique to evaluate airflow patterns and improvements to these patterns to obtain uniformity. Smoke and dye streams can be utilized for visualization and are often recorded on video for qualitative analysis. Measurements of the scale model can be preformed for quantitative data but great care may be required due to the scale and low stagnation velocities. In many cases stagnation velocities are less than 0.1 inches W.C. and significant measurement errors are possible. The PNR modeling can be costly, especially if many trials are required to correct flow imbalance. Physical nonreacting models cannot be used for reacting flows, heat transfer, or emission predictions. These models have been used successfully in flow inlet to duct burner systems (see Figure 26.3). 26.3.3 Physical Reacting Test (PRT) Sometimes it is necessary to build an actual prototype unit for testing. An example might be the testing of a burner for stability that cannot now be simulated any
Industrial Combustion Testing
other way. It is common to test scale prototype burner units and scale up the results. This is generally satisfactory since time scales are lower for scale units. In addition, a physical reacting test (PRT) is required for new burner types for emission prediction and verification of correlation parameters or CFD codes. The PRT tests are usually conducted during the development of products by equipment vendors and users rarely need to repeat the testing. The PRT can be costly depending on the size and flow rates. This type of simulation test is common for predicting duct burner emissions. 26.3.4 Computational Fluid Dynamics (CFD) Using CFD, the domain of interest, inlet conditions, flame pattern, temperature distribution (and under some conditions very good correlation with emissions), is formulated on a computer workstation. The volume is then divided into discrete volumes or cells. The number of these cells is dependent on the domain and could be into the millions. Boundary conditions are added for flow injection of air, fuel, or other flows. Physical properties for all flows are inputted. The user must specify certain parameters such as mixing models, kinetic rates, turbulence, and others as required. The CFD model is generally full-scale with complete similitude. The governing differential equations that solve all aspects of mixing, heat transfer, chemistry, turbulence, fluid mechanics, species, and continuity are iterated across the entire model until a converged solution is obtained for all cells and boundary conditions. When converged, the user has complete access to all information such as temperature profiles, pressure drop, velocity profiles, mass and species distribution if the model is appropriate and the model inputs were correct, and the cells are sufficiently small the likelihood of an accurate simulation is very good. The CFD simulation has the capability to provide complete information provided the above is true. The issue of validity has been a hot topic for years. A recent Department of Energy (DoE) report [4] has cited CFD to be capable of:
Figure 26.3 Physical modeling.
© 2011 by Taylor and Francis Group, LLC
1. Predicting catastrophic failure 2. Qualitative trends and parametric analysis 3. Visualization 4. Predicting nonreacting gaseous flows 5. Quantitative analysis of gas velocity and temperature patterns 6. Qualitative analysis of radiation heat transfer 7. Flame dynamics and shape 8. Effects geometry changes
Performance Prediction of Duct Burner Systems via Modeling and Testing
521
Z Y Figure 26.4 (See color insert following page 424.) CFD modeling of Duct burner.
9. Models of temperature and heat release patterns and qualitative trends associated with major species 10. Integration of detailed burner codes with heating process For combustion systems, CFD is the only generalpurpose simulation model capable of modeling reacting flows in order to predict emissions, heat transfer, and other furnace parameters. Users of CFD have done numerous studies for validation to extend the DoE list above to include reacting flows [2,5–9]. Longtime CFD users have the best experience with a combination of experimental data and CFD “calibration” [9]. Coen Co. has been using CFD since 1985 for optimization of burners and combustion systems and has an extensive library of CFD solutions and associated parameters that can be applied to reacting flows [10]. Many commercial CFD codes are available such as Fluent, Star-CD, CFX, and CFDRC. Although these codes are commercial, they are still user specialized and the user must have a high degree of knowledge of the code and limitations in order to apply the code to real problems and utilize the correct code coefficients in addition to experimental validation where applicable. Examples of CFD results for duct burner applications are shown in Figures 26.4 through 26.6. Further information on CFD for industrial applications can be found in Reference [12]. 26.3.5 Nonreacting Flow Models A common modeling application in duct burner systems is inlet flow distribution. This is especially useful for all the ducting up to the duct burner fuel injection point. These nonreacting flow applications can
© 2011 by Taylor and Francis Group, LLC
X
Figure 26.5 (See color insert following page 424.) Temperature profile with maldistribution.
Z Y
X
Figure 26.6 (See color insert following page 424.) CFD of improved temperature distribution.
be adequately modeled with physical nonreacting scale models or CFD. Both models are capable of predicting flow velocities and pressure distributions and optimization. The PNR models must retain geometric and kinetic similitude and great care must be considered when sizing in order to be able to measure very low stagnation pressures. For example see Figure 26.3, where a plastic model is used to develop a distribution grid. The CFD models must be selected with sufficiently small cells for convergence and utilize appropriate coefficients based on experimental data. Also consider vendor mathematical correlations for nonreacting flow.
522
26.3.5.1 Reacting Flow Applications For reacting flow applications the only choices are scale testing, mathematical correlations to full-scale testing, or CFD. If a scale test is possible and the cost is not high, then this is a logical alternative to consider opposed to any mathematical simulation. If a scale test is not feasible or too costly then CFD is the best choice even if coupled with mathematical correlations. 26.3.5.2 Recommended Modeling and Performance Prediction Methods For unreacting flow upstream of a duct burner system by preference:
1. Computational fluid dynamics 2. Physical nonreaction scale models
For duct burner emissions predictions:
1. Physical reacting testing combined with 1D/2D vendor correlations 2. Computational fluid dynamics with kinetic postpossessing
For duct burner downstream temperature distribution:
1. Computational fluid dynamics
For duct burner stability:
Industrial Combustion Testing
Where A = moles carbon/mole fuel in the duct burner fuel B = moles hydrogen/mole of fuel in the duct burner fuel X = moles of water vapor/mole of oxygen in the TEG/air Y = moles of carbon dioxide/mole of oxygen in the TEG/air Z = moles of nitrogen/mole of oxygen in the TEG/air Φ = excess TEG/air ratio where 1 = zero oxygen in the products MWOX = molecular weight of the oxidizer, TEG/air MWF = molecular weight of the duct burner fuel for simple HC fuels = (12A + B) MWPD = molecular weight of the duct burner products, dry MWPW = molecular weight of the duct burner products, wet MW (chemical) = molecular weight of any specified chemical, such as CO, NO PPM = parts per million by volume or molar HHV = higher heating value (Btu/Lbm) for the duct burner fuel LHV = lower heating value (Btu/Lbm) for the duct burner fuel T = temperature Subscripts d, w, m, v = dry, wet, mass, and volume or molar. Subscripts I, p, o = inlet, produced, and outlet. The overall chemistry is defined as:
1. Physical reacting testing combined with 1D/2D vendor correlations
B CA H B + Φ A + [[O 2 ] + X[H 2 O] + Y[CO 2 ] + Z[N 2 ]] → 4
26.4 Chemistry and Emission Calculations for Gas-Fired Applications Since a large majority of duct burner applications utilize clean and simple gas fuels, this section on chemistry and emission calculations is included to familiarize the reader with the simple chemistry, provide insight, and provide the ability to perform some simple calculations. Convert the TEG/air composition so it is molar normalized for the oxygen concentration to = 1.0 and include argon in the nitrogen, for simplicity. All the following equations can apply before the duct burner and after. Convert the fuel composition for simple hydrocarbon fuels so it is molar normalized for subscript A = carbon and subscript B = hydrogen.
© 2011 by Taylor and Francis Group, LLC
A[CO 2 ] + B/2 [H 2 O] + [Φ − 1][A + B/4][O 2 ] + Φ[A + B/4][X[H 2 O] + Y[CO 2 ] + Z[N 2 ]].
(26.1)
The wet TEG/air to fuel molar ratio is:
B TEG/air def Fuel vw = Φ A + 4 [1 + X + Y + Z]. (26.2) The dry TEG/air to fuel molar ratio is:
B TEG/air def Fuel vd = Φ A + 4 [1 + Y + Z].
(26.3)
523
Performance Prediction of Duct Burner Systems via Modeling and Testing
The wet molecular weight of the TEG/air oxidizer is: 32 + 18X + 44Y + 28Z [MWOX] w = . 1+ X + Y + Z
(26.4)
The dry molecular weight of the products dry is: [MWPD] B B A[44] + [Φ − 1] A + 4 [32] + Φ A + 4 [Y[44] + Z[28] = . B B A + [Φ − 1] A + + Φ A + [Y + Z] 4 4
The dry molecular weight of the TEG/air oxidizer is: 32 + 44Y + 28Z [MWOX] d = . 1+ Y + Z
(26.5)
The wet products to fuel mass ratio is:
The wet TEG/air to fuel mass ratio is: TEG/air mw = TEG/air vw [MWOX]w . (26.6) Fuel Fuel MWF
MWOXw Products TEG/air Fuel mw = Fuel vw MWF + 1.
The dry TEG/air to fuel mass ratio is:
TEG/air TEG/air [MWOX]d Fuel md = Fuel vd MWF . (26.7)
[O 2 ] wet
[ Φ − 1] A + B4
. B B B A + + [Φ − 1] A + + Φ A + [X + Y + Z] 2 4 4 (26.8) =
The dry oxygen concentration in the products will be:
chemical fuel mi PPM [ chemical]di ] TEG/air [ MW chemical ] Fuel md 6 10 = . WOX]d [MW
(26.14)
The dry mass of chemical to fuel ratio out is:
The wet molecular weight of the products is:
chemical fuel mo
[MWPW] B B A[44] + [Φ − 1] A + 4 [32] + Φ A + 4 [X[18] + Y[44] + Z[28] = B B B A + + [Φ − 1] A + + Φ A + [X+Y+Z] 4 2 4
(26.10)
© 2011 by Taylor and Francis Group, LLC
(26.13)
Using the above general equations it is easy to solve for duct burner emission contributions such as CO, NO, and HC given the inlet and outlet concentrations. The duct burner contribution is equal to the mass output minus the mass input. The contribution can be either negative or positive. The dry mass of chemical to fuel ratio in is:
[ Φ − 1] A + B4
. B B A + [Φ − 1] A + + Φ A + [Y + Z] 4 4 (26.9) [O 2 ] dry =
Products TEG/air MWOXd Fuel md = Fuel vd MWF + 1.
(26.12)
The dry products to fuel mass ratio is:
The wet oxygen concentration in the products will be:
(26.11)
PPM [ chemical]do] Product [ MW chemical ] Fuel md 6 10 . = [MW WPD]
(26.15)
524
Industrial Combustion Testing
The dry mass of chemical to fuel ratio contribution by the duct burner is:
chemical chemical chemical fuel mp = fuel mo − fuel mi.
(26.16)
The chemical emission based on mass/million BTU is:
[ chemical/(106 BTU)]
mp = [[chemical/fuel] mp/HHV][106 ].
(26.17)
Given the mass flow of the inlet and fuel flow for any specific condition, Φ can be found directly from the mass ratio as: wet inlet mass flow fuel flow
B 32 + 18X + 44Y + 28Z Φ A + A [1 + X + Y + Z] 1+ X + Y + Z . = 12A + B
(26.18)
All of the above duct burner parameters and emissions will then be a direct function of A, B, X, Y, Z (PPM chemical), and Φ. It is convenient to rewrite the dry O2 relations to determine Φ directly as a function of A, B, X, Y, and Z for emission calculations. Frequently it is necessary to convert from actual O2 values to reference 15% O2. The general relationship is: Φ = ([1/[1 + B/4A]] + Y + Z)/(1/((O 2 )dv) − [1 + Y + Z]) + 1.
(26.19)
Given Φ, the use of Equation 26.19 and Equations 26.14 through 26.17 can be used to convert from mass basis to reference 15% O2 for any chemical emission. To compute the product temperature, set the LHV equal to the product enthalpy minus the inlet enthalpy on a wet mass per fuel basis.
26.5 Emissions A duct burner system can either increase or reduce emissions from the generally large volume of mass flow at the input. Generally this flow includes particulate, NOx, CO, and a variety of HC’s including a subset of HC defined as volatile organic compounds (VOC). Where VOC is defined by EPA (40 CFR 51.100, February 3, 1992) as “any compound of carbon, excluding carbon monoxide, carbon dioxide, carbonic acid, metallic carbides, or ammonium carbonate, which participates in atmospheric chemical reaction.” Other compounds are also exempt such as methane, ethane, methylene chloride, methyl chloroform, and other minor chemicals. A subset of emissions is visual opacity from the stack. A direct correlation between opacity, particle loading and size can be computed using simple light scattering computations where sulfur is not a significant contributor. Nitrogen oxides: Combustions systems generate NOx, principally from three mechanisms. 1. Thermal NOx from the dissociation of molecular nitrogen and oxygen at high temperatures. 2. Prompt NOx from the interaction of molecular nitrogen with hydrocarbon radical. 3. Fuel NOx from the reaction of oxygen with elemental nitrogen compounds in the fuel such as NH3, HCN, and CN. Generally, using ordinary fuels, duct burner NOx emissions are dependent on thermal NOx formation with prompt NOx being a minor contributor. The oxidation of carbon monoxide, HC, VOC, and particulate is very dependent on oxygen concentration, temperature, and time. In duct burner systems a large range of temperatures and oxygen gradients exist downstream of the burners. At high temperatures with sufficient oxygen, oxidation of HC is relatively fast. At low temperatures the oxidation is very slow. At medium temperatures the oxidation is defined by mixing and chemical kinetics. Duct burner systems have a complete range of temperatures from low to high and chemical kinetics play a large impact on emission predictions.
[ LHV ] Products TEG/air = f [ H(To)] mw. mw − f [ H(Ti)] Fuel Fuel
© 2011 by Taylor and Francis Group, LLC
(26.20)
26.6 Chemical Kinetics Duct burner systems operate with moderate temperatures and as such exact chemical kinetics is very
525
Performance Prediction of Duct Burner Systems via Modeling and Testing
important for emission prediction, they must be correct. Generally first order reaction rates perform very well. For first order oxidation the general expression is:
(26.21)
−E K = A e , RT
(26.22)
and A = pre-exponential factor/frequency factor in appropriate units R = universal gas constant in appropriate units T = absolute temperature E = activation energy, usually listed in kcal/mole T = time seconds For perfectly stirred reactors well down stream of the duct burner, integration of the first order equation results in a simple equation for constant temperature and O2 mole fraction in a time step as: (1 – (chemical final/chemical initial)) = [1 – e –k(O2)(Δt)]. For utilization, and performance prediction, kinetic data can be utilized from the literature [16]. For instance for CO destruction, several kinetic data are available such as [13]. d[CO] dt 2
−25, 000 P = −1.8107 e − . (CO) (O 2 )0.5 (H 2 O)0.5 RT RT
(26.23)
Most all published CO rates involve H2O because CO destruction requires the (OH)–1 radical to produce the reaction. For HC and VOC incineration, several sources are available such as Edelman [17]. Where in general:
moles 12 , 200 (C H ).5 (O 2 ) Te T a b cm 3 sec
(26.24)
For formation, rather than destruction such as NOx, the equations are similar such as the formation of thermal NOx.
© 2011 by Taylor and Francis Group, LLC
(26.25)
(O)eq =
K0 0.5 . (O 2 )eq (RT)0.5
(26.26)
One generally accepted practice is to assume (O2) in equilibrium with (O) and (O2) concentration using the Westenburg (1971) results for ko [14] for (O2) equilibrium and Zeldovich constants, A and E as measured by Bowman [15]. For particulate oxidation, an equation can be developed from fundamental principals utilizing a combination of diffusion of oxygen and surface reactivity as follows:
dm/dt = (12 Cog Ap )/(1/km + 1/kr ),
(26.27)
where M = mass of particle Cog = molar density Ap = partial surface area Km = diffusion coefficient of oxygen in nitrogen Kr = reaction coefficient of the form A e–E/RT A = frequency factor E = activation energy R = universal gas constant T = temperature The equation can be integrated for constant density particles and using particle tracking in time steps with constant or varying oxygen and temperature. An excellent source of char rate data is available by Smith and Smoot [18]. Utilization of gas kinetic data: Then in all cases, one can postprocess thermal map data in some discrete volume form or insert into a CFD code using the Rayleigh flux theorem as follows: cv
d(C a H b ) = −5.52(108 )P −0.815 dt
d(NO) = 2 A e( − ( E/RT )) (O)eq (N 2 ), dt
and
d(chemical) = K[O 2 ][chemical], dt
where
∂ n ρ dv = ∂t
∫
cs
∫ n ρ(V ⋅ da),
where n = chemical in mass units t = time ρ = density v = volume a = area V = velocity vector
(26.28)
526
Industrial Combustion Testing
Where described in words, the formation of (n) through the volume surface is equal to the integrated rate of f ormation over the control volume. It is a simple extrapolation to extend this concept for even coarse volumes as follows:
∑ dt ρ ∆ v = n ρ(V ⋅ a). dn
(26.29)
This method can be very useful for fully mixed downstream products even with coarse volumes. But one must be careful with coarse volumes to be sure that the temperature and concentrations are uniform.
26.7 Performance and Prediction 26.7.1 TEG/Combustion Air Flow Distribution The turbine exhaust gas/combustion air velocity profile at the duct burner plane must be within certain limits to ensure good combustion efficiency, and in cogeneration applications this is rarely achieved without flow straightening devices. Even in nonfired configurations, it may be necessary to alter the velocity distribution to make efficient use of boiler heat transfer surface. Figure 26.7 shows a comparison of flow variation with and without flow straightening.
26.7.2 Temperature Profile
Relative elevation
Comparison of flow variation 9 8 7 6 5 4 3 2 1
Even with near perfect flow distribution the arrangement of duct burners will produce a nonuniform temperature distribution downstream of the burner location. The degree of this maldistribution is dependent on the duct burner spacing, firing rate, and distance downstream. It is generally impractical to measure in full size units and difficult in scale test units requiring long, high velocity, watercooled probes. The only practical method of defining the downstream temperature is using CFD. See Figures 26.5 and 26.6. Temperature maldistribution is especially important when finned super heater tubes are located at the flame tails.
No flow distribution devices
9 8 7 6 5 4 3 2
Duct burners are commonly mounted in the TEG duct upstream of the first bank of heat transfer tubes, or they may be nested in the boiler superheater between banks of tubes. In the former case, a straightening device would be mounted just upstream of the burner, while in the latter it is mounted either upstream of the first tube bank or between the first tube bank and (upstream of) the burner. Although not very common, some HRSG design configurations utilize two stages of duct burners with heat transfer tube banks in between and a flow straightening device upstream of the first burner. Such an arrangement is, however, problematic as the TEG downstream of the first stage burner may not have the required combination of oxygen and temperature properties required for proper operation of the second stage burner. Perforated plates that extend across the entire duct cross section are most commonly used for flow straightening because experience has shown they are less prone to mechanical failure than vane-type devices, even though they require a relatively high pressure drop. The pattern and size of perforations may be varied to achieve the desired distribution. Vanes can produce comparable results with significantly less pressure loss but require adequate structural reinforcement to withstand the flow-induced vibration inherent in HRSG systems. Regardless of the method used, flow pattern complexity, particularly in TEG applications, usually dictates the use of either physical or CFD modeling for design optimization.
With flow distribution grid
26.7.3 Stability
1 50
75
100 125 Percent flow relative to mean
Figure 26.7 Inlet flow variation.
© 2011 by Taylor and Francis Group, LLC
150
Flame stability is governed by the limits of flammability and ignition temperature of the fired fuel. Due to the complex chemistry it is generally impractical to predict precisely using CFD modeling. Scale tests and actual field data are used to extend simple 1D and 2D models for calibration to predict flame stability. The variables
527
Performance Prediction of Duct Burner Systems via Modeling and Testing
18 Flammability
Ignition
17 16
% O2
15 14 13 12 11
0
200
400
600 800 TEG (F°)
1000
1200
1400
NOx emissions (lb/MMBTU, HHV)
Figure 26.8 Stability limits.
0.08
Duct burner NOx emissions versus stoichiometric adiabatic flame temperature (°F)
0.07 0.06 0.05 0.04 0.03 0.02 3150
3175 Standard tips
3200
3225 3250 Stoich AFT (°F)
Mark I low NOx tips
3275
3300
Mark II low NOx tips
Figure 26.9 NOx testing and correlation.
are upstream composition, temperature, and fuel type. A map of stability is then created for various fuels as shown in Figure 26.8 for methane.
used using actual full-scale data and/or scale testing (see Figure 26.9).
26.7.4 Emissions To a large degree emissions can be predicted using the methods outlined in the section on kinetics. However, the impact of burner design, turbulence, large- and small-scale eddies, and detailed chemistry will require “calibration” of the computational method
© 2011 by Taylor and Francis Group, LLC
26.8 Summary Performance prediction and testing of duct burner systems is highly complex requiring a combination of full-scale and/or subscale data, sophisticated
528
computational models. Significant diligence is required in the selection of the prediction method used and the application of precise kinetics for accuracy.
References
1. Baukal, C. E., Jr., and Schwartz, R. E. eds. The John Zink Combustions Handbook Boca Raton, FL: CRC Press, 2001. 2. Schulze, W. J. “A Systems Approach to Adding Natural Gas Firing to Two 600 MW Units.” Joint ISA/EPRI Conference, June 8, 1993. 3. Moody, L. F. “Friction Factors for Pipe Flow.” Transac tions ASME 66 (November 1944): 671. 4. Department of Energy “Improving Industrial Burner Designs with Computational Fluid Dynamic Tools: Progress. Needs and R&D priorities.” Workshop Report September 2002. 5. Stopford, P. J. “Recent Applications of CFD Modeling in the Power Generation, Metals and Process Industries.” Second International Conference on CFD in the Minerals and Process Industries CSIRO, Melbourne, Australia, December 6–8, 1999. 6. Al-Khalidy, N. “Design Optimization of Industrial Ducts Using Computational Fluid Dynamics.” Third International Conference on CFD in the Mineral and Process Industries CSIRO, Melbourne, Australia, December 10–12, 2003. 7. Bockelie, M., Adams, B., Creamer, M., Davis, K., Eddings, E. Valentine, J., Smith, P., and Heap, M. “Computational Simulations of Industrial Furnaces. Computational Technologies for Fluid/Thermal/Chemi
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
cal Systems with Industrial Applications.” 1998 Joint ASME/JSME Pressure Vessels and Piping Conference, July 1998, San Diego, CA. 8. Mehrotra, V., Diwaker, P., and Vallavanatt, R. “Trouble shooting Furnace Operations Using Computational Fluid Dynamics (CFD).” ASME-PVP’04, San Diego, CA, July 25 –29, 2004. 9. Saripalli, R., Wang, T., and Day, B. “Simulation of Combustion and Thermal Flow in an Industrial Boiler.” 27th Industrial Energy Technology Conference, May 11–12, 2005, New Orleans, Louisiana. 10. Coen Internal Documents Coen Co. “Coefficients for Fluent CFD Applications.” 11. Coen Internal Documents Coen Co. “BurnCalc.” 12. Baukal, C., Gershtein, V., and Li, X. eds. Computational Fluid Dynamics in Industrial Combustion. Boca Raton, FL: CRC Press, 2001. 13. Williams, G. C., Hottel, H. C., and Morgan, A. C. “The Combustion of Methane in a Jet-Mixed Reactor.” 12th Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 1969. 14. Westenberg, A. E. “Turbulence Modeling for CFD.” Combustion Science and Technology 4 (1971): 59–67. 15. Bowman, C. T. “Kinetics of Pollution Formation and Destruction in Combustion.” Progress in Energy and Combustion Science 1 (1975): 33–45. 16. Battelle Columbus Laboratories. “Chemical Aspects of Afterburner Systems,” EPA-600/ 7-79-096, April 1979. 17. Engleman, V. S., Bartok, W., Longwell, J. P., and Edelman, R. B. “Experimental and Theoretical Studies of NOx Formation in a Jet Stirred Combustor.” 14th Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 1973. 18. Douglas Smart, L., and Philip Smith. “Coal Combustion and gasification. pp. 81–88, Plenun press, New York, 1985.
27 Oxy-Fuel and Oxygen-Enhanced Burner Testing Lawrence E. Bool, III, Nicolas Docquier, Chendhil Periasamy, and Lee J. Rosen Contents 27.1 Introduction.................................................................................................................................................................. 529 27.2 Development Tools...................................................................................................................................................... 530 27.3 Experimental Tools...................................................................................................................................................... 530 27.3.1 Laboratory-Scale Furnace Testing............................................................................................................... 532 27.3.1.1 Example of a Purpose Built Laboratory-Scale Furnace: Forehearth Furnace...................... 532 27.3.1.2 Example of a Generic Laboratory-Scale Furnace: High Temperature Furnace................... 533 27.3.2 Pilot-Scale Furnace Testing........................................................................................................................... 534 27.3.2.1 Example of a Pilot-Scale Furnace: L1500................................................................................... 535 27.3.2.2 Example of a Pilot-Scale Furnace: ALICE Test Furnace.......................................................... 536 27.3.3 Open Air Testing........................................................................................................................................... 538 27.3.4 Customer Testing........................................................................................................................................... 539 27.4 Modeling Tools............................................................................................................................................................. 541 27.4.1 Cold Flow Modeling...................................................................................................................................... 541 27.4.2 Dense Phase (PLIF) Modeling..................................................................................................................... 541 27.5 Key Process Variables and Measurement Techniques........................................................................................... 542 27.5.1 Fuel and Oxidant Properties........................................................................................................................ 542 27.5.2 Flame Geometry............................................................................................................................................ 543 27.5.3 Heat Transfer Measurement......................................................................................................................... 543 27.5.4 In-Flame Temperature Measurement......................................................................................................... 544 27.5.5 Temperature Distribution on Furnace Hearth and Crown..................................................................... 544 27.5.6 Stack Gas Temperature Measurement........................................................................................................ 544 27.5.7 Heat Flux Measurement................................................................................................................................ 544 27.5.8 Emission Measurement................................................................................................................................. 545 27.5.9 In-Flame Chemical Composition Measurement....................................................................................... 546 27.5.10 Stack Gas Sampling....................................................................................................................................... 546 27.5.11 Advanced Flame and Aerodynamics Characterization........................................................................... 546 27.6 Oxygen Safety in Burner Testing.............................................................................................................................. 546 27.6.1 Oxygen System Design and Material Compatibility................................................................................ 546 27.6.2 Oxy-Cleaning, Operation, and Maintenance............................................................................................ 547 27.6.3 Training........................................................................................................................................................... 547 27.6.4 Safety in Working With Partners................................................................................................................ 548 Disclaimer................................................................................................................................................................................ 548 References................................................................................................................................................................................. 548
27.1 Introduction Oxy-fuel burners were originally developed for applications that require extremely high flame temperatures that could not be achieved using air/fuel burners. Examples of these applications include cutting torches, glass polishing, and so on. Over the decades, advances in oxy-fuel burner
development have enabled oxy-fuel use in a variety of applications, spanning a range of industries and furnace temperatures. Oxygen addition to air/fuel burners, addition of oxy-fuel burners to air fired zones, and/or conversion of air-fired zones to oxy-fuel-fired zones has been employed in a variety of industries to improve productivity, improve fuel efficiency, and reduce net operating costs as well as reduce pollutant emissions. Industries that have 529
© 2011 by Taylor and Francis Group, LLC
530
utilized oxygen addition in their combustion processes include glass, metals (steel, iron, aluminum, copper), cement, refining, petrochemical processing, incineration, and power generation to name a few. As the list suggests the scale of, and critical requirements for, each process are often very different. This requires a detailed understanding of the burner’s performance and how this performance will integrate with the process. Unlike air-based burner manufacturers that typically focus on market(s) with similar requirements (i.e., burner manufacturers that supply burners for the process industry typically don’t provide burners for the steel or metals industry), oxy-fuel burner manufacturers are often called upon to develop burners for vastly different industries. This need for a portfolio of burners with different, and often contradictory, characteristics requires the capability to test burners under a wide range of process conditions dictated by the wide range of prospective customer’s processes. To accomplish this, a variety of tools are utilized to take a new concept for a burner from idea through to a commercially proven technology. This chapter will summarize the tools utilized by companies that offer oxy-fuel solutions.
27.2 Development Tools One of the most difficult aspects of developing new oxy-fuel technologies is the transition from a novel, and often nebulous, concept to a commercially viable offering. Ideally the development path consists of a number of key stages. The first stage is usually a concept definition step, where the preliminary idea is refined to establish a concrete development path. The second step is the proof of concept stage where the fundamental soundness of the idea is demonstrated. The third stage, which is often the longest, is the scale up and design stage. Finally, the technology is tested and demonstrated for a customer in a commercial furnace. A wide range of tools are available to obtain key information relevant to each stage. These tools include theoretical modeling, physical modeling, and experimental test facilities. Figure 27.1 shows how these tools are commonly used for each stage. Typically, simple theoretical tools or small-scale experiments are used to vet the concept in the first two stages. More complicated theoretical modeling, physical modeling, and larger-scale experimental tools are used in the scale up and design stage. Finally, large-scale experimental tools coupled with modeling can be used to demonstrate the technology in the final stage. Although many commercial technologies have been developed using this strategy, resource and time constraints may limit the scope of the development efforts. Often a compressed development schedule will require
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
Customer testing Outdoor testing Pilot testing Laboratory testing Physical modeling Concept definition
Proof of concept
Scale up and design
Full-scale demonstration
Figure 27.1 Use of development tools.
the concept to go from the definition stage directly to larger scale testing/modeling. In extreme examples a concept may go directly to testing in a customer furnace. In this case a clear understanding of the capabilities and limitations of the various tools is necessary to facilitate rapid commercialization and avoid costly delays or mistakes.
27.3 Experimental Tools A key component to oxy-fuel research and development is careful experimental testing of the concept or technology. A well-designed experimental study can demonstrate proof of concept, enable design and scaling of the technology to commercial size, and minimize the technical risk associated with installing the technology in a customer’s facility. For example, testing early in the development program can demonstrate that the technology meets key performance objectives under ideal conditions. Additional experiments can be used to better understand the underlying mechanisms driving performance, or lack thereof, of the technology. These data can be used for model development/validation and to derive scaling rules for the technology. These fundamental experiments may also identify new technology concepts. Experiments can be designed to specifically evaluate how well the technology works under less ideal conditions typical of normal plant operating conditions. By testing the technology at nearreal world conditions, investigators can identify potential gaps in the technology before investing in large-scale testing. The data can also be used to optimize the technology to best meet the requirements of potential customers. Finally, customer acceptance of the technology typically requires it be demonstrated at the commercial scale. Test rigs used for combustion experiments span a wide range of size and complexity. The test facilities can be
531
Oxy-Fuel and Oxygen-Enhanced Burner Testing
Complexity and realism
Furnace scale
Customer
Outdoor
Pilot
Laboratory 0
20
40 60 Firing rate (MWth)
80
>100
Figure 27.2 Description of furnace scales.
divided into four different types, as shown in Figure 27.2. Since oxy-fuel combustion is used in many industries, there is a substantial overlap in the sizes of the different types. For example, a full-scale burner can be as small as a cutting torch, while a pilot-scale utility boiler can exceed 15 MW. In general, laboratory-scale equipment range from small tube furnaces to ~0.6 MW test furnaces. This type typically provides a very well-controlled environment, but may not be representative of commercial systems. As shown previously in Figure 27.1, laboratoryscale equipment is often used in the proof of concept and scale up/design phase. Pilot-scale facilities range from ~0.9 MW to 15 MW, and tend to be more complex than the laboratory-scale. The added complexity makes it more difficult to derive fundamental mechanisms, but provides a more representative combustion environment. Outdoor (or open air) test facilities are commonly used for burner testing and can evaluate burners up to ~15 MW. Finally, the largest and most complex type is the full-scale. Depending on the customer, the full-scale burner may be fairly small. However, full-scale facilities can exceed 800 MW with multiple burners. Maintaining and operating experimental facilities can be very costly, especially pilot- and full-scale furnaces. If an R&D group focuses exclusively on a single industry, such as the electric power industry, these resources may be well spent. However, for those organizations that support different industries, another option is to combine in-house and external expertise and facilities. Technology developers typically get access to external furnaces through collaborative efforts or by contracting for a specific set of tests. Each strategy has benefits and drawbacks. For example, with in-house R&D the entire process can be well controlled, and the development pace can be defined by the technology developer. Use of in-house resources keeps control of intellectual property within the company. However, key expertise and experimental tools may not be readily available.
© 2011 by Taylor and Francis Group, LLC
Collaborative R&D programs are another way to access a wide range of experimental tools and expertise. These programs can significantly reduce the development risks and time to commercialization for new technologies. Praxair’s oxygen enhanced combustion for NOx control is a classic example of a successful collaborative program [1,2]. In this program Praxair teamed with the U.S. Department of Energy, two universities (University of Arizona and the University of Utah), a commercial computational fluid dynamics (CFD) modeling company (Reaction Engineering International), and a major boiler supplier (Alstom Power). By combining the efforts of academic and industrial groups, Praxair was able to go from laboratory-scale experiments to a commercial installation in approximately two years. Another example of collaborative R&D efforts is the one between Air Liquide and Babcock & Wilcox. This joint R&D program has successfully demonstrated an oxy-coal combustion technology for power generation at a pilot-scale of 30 MWth [3]. The primary downside of a collaborative R&D effort is control of both the pace of the development work and control of the results. Agreements between the participants must be carefully crafted to ensure fair distribution of new intellectual property. Problems at one partner’s laboratory, such as equipment failure on a key furnace, can significantly impact the entire program. Some large experimental facilities will perform specific tests for hire. For example, some universities will perform testing for companies, particularly if they are routine; that is, no new R&D required. Large companies that do in-house research may also do target experiments for hire, provided they are not detrimental to the host company. For these types of tests the intellectual property rights are typically retained by the customer. However, test fees are typically higher than in-house or collaborative programs. In addition, scheduling time in a large pilot-scale facility may have to be done a year or more in advance, which may create scheduling problems. Another key component of oxy-fuel testing is availability of industrial gases. Many of the large industrial gas companies, such as Praxair, Air Liquide, and others have fully integrated in-house furnaces with oxygen and other industrial gases readily available. Until the recent attention on oxy-fuel combustion for carbon capture and sequestration (CCS), most combustion organizations have focused on air combustion. Therefore, before any oxy-fuel experiment can be started an oxygen supply system has to be installed. For small laboratory-scale facilities the oxygen supply may simply be compressed gas cylinders. Larger furnaces will require some kind of bulk liquid oxygen supply, either temporary for shortterm testing or permanent for longer-term testing or for volumes that cannot be accommodated by temporary systems. Figure 27.3 shows an example of a portable oxygen supply skid. The liquid oxygen is typically vaporized
532
Figure 27.3 Portable oxygen supply system capable of supplying 100 tpd. (Courtesy of Praxair.)
with an ambient vaporizer, but could be vaporized with steam or electric heaters if needed. For some development programs a mix of supply methods may be needed. For example, during the previously mentioned OEC for NOx control program, cylinders were used for oxygen supply at the University of Arizona. A small liquid oxygen tank on a trailer with an onboard vaporizer was used for testing at the University of Utah. A larger, portable liquid oxygen tank and vaporizer were used for the testing at Alstom Power. For long-term programs it may be more cost-effective to replace the portable supply systems with a permanent installation. 27.3.1 Laboratory-Scale Furnace Testing Two primary hallmarks of laboratory-scale furnaces are the ability to tightly control and measure key process variables and their relatively low cost of operation. Laboratoryscale furnaces can be as simple as a muffle furnace or as complicated as a self-sustained combustor. A typical furnace has a single burner firing 30 kW–0.6 MW into a well-instrumented combustor. Because of their small size, laboratory furnaces are more flexible than larger facilities, allowing them to be used to test a wide range of combustion conditions. For example, mixing between fuel and oxidant can be varied by modifying the burner or furnace, often in ways that would be impractical at larger scales. Furnace temperature profiles can be controlled by either adding heat through external heaters or by removing heat from the specific areas with water cooled panels or probes. This type of operational flexibility allows investigators to change one variable at a time. This tight control also enables investigators to better understand the fundamental mechanisms for a given technology and to prove the viability of the concept under ideal conditions. Laboratory-scale furnaces are either designed to be extremely generic to allow for a wide range of experiments, or are purpose built to explore specific
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
combustion phenomenon. A generic design offers the flexibility of “quick and dirty” experiments and can be used to evaluate concepts for a wide range of industries and technologies. The generic design, however, may preclude investigators from simulating specific furnaces. For example, it may be difficult to simulate the time-temperature history of a given furnace of interest, even using the previously mentioned strategies. A purpose built furnace can overcome these problems, but at the cost of flexibility. For example, a laboratoryscale furnace that is purpose built to simulate the timetemperature history of a boiler may not be suitable for burner development for the glass industry. A purpose built furnace can also be designed to focus on specific aspects of the combustion process. For example, a furnace that is designed to test near-burner mixing may not fully mimic a full-scale reactor downstream of the near burner zone. Since laboratory-scale furnaces typically fire with a single burner, interactions between multiple burners is missing. A well-designed generic or purpose built laboratoryscale furnaces will include as many access points as possible. These access points, often just penetrations into the furnace, allow both in-situ and extractive measurements to be taken. These ports can be used in conjunction with a gas analysis train to perform detailed composition mapping of the furnace. Other tools, discussed in Section 27.5, can be used to acquire detailed information on the flow field in the furnace, heat flux, particulate loading, temperature, and other variables of interest. Although the small furnace size facilitates these measurements, care must be taken to avoid influencing the combustion process while taking the measurements. For example, inserting a large water-cooled probe into a small combustor can change the temperature profile. Therefore choice of measurement tools needs to take into account their potential impact on the furnace environment. 27.3.1.1 Example of a Purpose Built Laboratory-Scale Furnace: Forehearth Furnace For a number of glass types including insulation or reinforcement fiber as well as container glass, the temperature of the glass flowing out of the melting tank must be finely controlled to guarantee properties adequate to glass forming. When molten glass is flowing in a channel or forehearth, heat is required to maintain and homogenize the temperature and viscosity of the glass. Natural gas burners located inside a refractory block are commonly used for that purpose. Air Liquide has developed and patented a new natural gas oxy-burner, the ALGLASS FH, to increase the energy efficiency of glass forehearths [4]. Using oxygen as an oxidizer usually increases flame temperature and drastically raises radiative heat transfer
533
Oxy-Fuel and Oxygen-Enhanced Burner Testing
Figure 27.5 (See color insert following page 424.) Inside view of the small-scale Air Liquide pilot test furnace in operation. Thermocouple locations (red dots) and view of the burner under testing.
Figure 27.4 View of the small-scale (50 kW) Air Liquide pilot furnace.
of the natural gas flame, leading to improved heat transfer to the glass, but also possibly to burner block overheating or melting if the burner is not carefully designed. This is why detailed characterization of the oxy-burner flame is critical along with its pairing to the refractory block, to ensure both adequate heat transfer to the glass in terms of intensity and location (heat is only really needed along the walls of the forehearth to avoid glass reboiling), and no damage to the burner block refractory. Air Liquide’s 50 kW forehearth pilot furnace shown in Figure 27.4 has been setup for this purpose. This pilot furnace allows detailed characterization of small-scale burners (1–50 kW) in operating conditions matching those of the industrial environment (geometry, temperature, and pressure). For this purpose, pilot oxy-burners are used in one portion of the furnace to warm up the furnace and maintain the wall temperatures equal to those of a glass forehearth (1200°C~1400°C) even at low burner output power, while the burner and burner block to be tested are mounted in another portion. There, thermocouples are installed to monitor the temperature distribution in the burner block and at distances from the flame equivalent to those separating it from the
© 2011 by Taylor and Francis Group, LLC
forehearth elements (glass surface, walls) in the industrial environment. A snapshot of the burner in operation is shown in Figure 27.5, along with the location of additional thermocouples on the side walls. With this pilot furnace, the burner design, the interaction between the burner and its refractory block as well as the interaction between the flame and the forehearth can be monitored and characterized for subsequent burner design adaptation and/or validation. The ALGLASS FH burner design is based on a pipe-inpipe technology that injects the gaseous fuel with a swirl effect to control flame length. Oxygen injection surrounds the fuel injection. Pilot furnace tests facilitated the selection of injection velocities and swirling ratios to fit glass forehearth requirements and burner block geometries. In particular, the swirl effect allows the flame hot spot to be maintained at a constant location (just outside the burner block) over a wide range of burner power inputs. When the power increases, for example, the swirl effect on the fuel injection also increases and increases mixing between the reactants thus avoiding change in the flame length and hence in the location of the hot spot. 27.3.1.2 Example of a Generic Laboratory-Scale Furnace: High Temperature Furnace The High Temperature Furnace at the Praxair Tech nology Center in Tonawanda, NY (Figure 27.6) is a typical laboratory-scale furnace. The furnace consists of nine steel sections; a 25.4 cm long front endcap, seven 45.7 cm long midsection pieces, and a 22.9 cm long rear endcap. The cylindrical-shaped, castable refractorylined furnace has an inner diameter of 0.8 m and an inner length of 3.2 m. The furnace is designed to handle firing rates up to 0.4 MW and wall temperatures
534
Industrial Combustion Testing
Figure 27.6 Praxair’s 0.4 MW laboratory-scale test furnace.
Furnace width, inches
Furnace wall temperature, °F 1865
20 10
1875
0 1800
–10 –20 10
Furnace width, inches
10
20
2015
20
2012
2063
1950 2000 2300 2400 2800 2600 Measured flame shape 2400 Isotherms °F Firing rate = 1 MMBTU/HR “A” burner –design 1
Burner
0
1889
30 40 50 60 70 Furnace length, inches 2039
2025
80
2056
90
100
1961
2150 Burner
0
2300
1700
–10 1580
2150 2000 2050 1800 2300
–20 0
10
20
Measured flame shape Isotherms °F Firing rate = 1 MMBTU/HR “A” burner –design 2
30 40 50 60 70 Furnace length, inches
80
90
100
Figure 27.7 Temperature profiles for different burner configurations in Praxair’s laboratory-scale test furnace.
up to 1560°C. Both the furnace endcaps and the midsection pieces have a number of port openings giving access to the furnace interior to allow for variation in burner or overfire oxidant positioning, extraction of gas samples via water-cooled probes, insertion of water-cooled heat sinks, placement of suction pyrometers and, in general, the insertion of any burner device or diagnostic instrument. Each of the seven furnace midsection pieces is equipped with a thermocouple. The average furnace wall temperature was defined as the average of the seven thermocouples. Figure 27.7
© 2011 by Taylor and Francis Group, LLC
shows the wall temperature profile along the top of each chart as well as temperature isotherms measured by probing the furnace at various locations for a Praxair A-burner configured with two different sets of oxygen nozzles. The two charts show that peak temperature can be reduced while overall temperature uniformity can be improved by modifying the burner from design 1 to design 2. Flue gas exits the furnace through the endcap and passes through a water-cooled pipe to a spray cooling tower. Gas samples are withdrawn with a water-cooled probe as the exhaust leaves the furnace. An oxygen sensor is mounted in the exhaust to measure the exhaust oxygen concentration (volume percentage on a wet basis) as a furnace control parameter. Water is injected into the cooling tower to maintain a stack temperature of less than 120°C. The exhaust gas then passes through an induced draft fan and is exhausted above the roof of the building. As part of a Department of Energy funded program designed to develop new low emission burner technologies, Praxair developed and tested a technology called Dilute Oxygen Combustion (DOC) [5]. These tests included temperature profiles and pollutant concentrations. The goal was to design a burner with a uniform heat flux with much lower NOx emissions. Detailed furnace mapping was done for various gas species (O2, CO, H2, CH4) [6]. An example map is shown in Figure 27.8. These data showed the new burner design was able to distribute the reaction over a wide volume within the furnace thereby resulting in very low NOx emissions and very uniform heat flux. Use of the laboratory-scale furnace allowed the concept to be tested at relatively low cost. 27.3.2 Pilot-Scale Furnace Testing Once lab-scale testing is complete the next step in the typical development pathway is testing the concept at the pilot-scale. A pilot-scale furnace is designed to mimic key operational parameters of specific full-scale combustion systems. For example, a pilot furnace may have the same time-temperature history as the fullscale furnace. The furnace may be designed to have similar near burner and “far field” mixing behavior as the full-scale. Burner-burner interactions, which are difficult to evaluate at small-scale, can be explored in a properly designed pilot furnace. In general, test furnace operating conditions (temperature, pressure, air leakage, thermal load) should match process targets. In particular, furnace heat load should be tuned depending on targeted application and furnace pressure and air-in leakages should be controlled. Pilot-scale test furnaces should provide qualitative (flame size, attachment, stability) and quantitative (emissions, temperature, and
535
Oxy-Fuel and Oxygen-Enhanced Burner Testing
1.0
Heat exchanger
0 0.9 0
Ox
0.8 0 0.7 0 1.0 0 0.9 0 0.8
7.9
0 0 1
0
4.4
0.4
4.
80
0
84
0
1.0 0
0.9 0.8 0 0.7
7
0 0
1
Z/ 0.00 R
27.3.2.1 Example of a Pilot-Scale Furnace: L1500
2.2
2
0 2.0 4.8
2
0
20
4.00 8
l
1.2
0.3
0.
5.4
ue
0.4
Lower furnace plane
0.5
0
8
0
L X/
1.0
0
7
6.00 7.4
4
1.0
24% O2 22% O2 20% O2 18% O2 16% O2 14% O2 12% O2 10% O2 8% O2 6% O2 4% O2 2% O2 0% O2
0
Figure 27.8 Typical furnace O2 mapping from DOC burner tests in Praxair’s laboratory-scale furnace.
heat flux distribution) information and are needed to perfectly understand oxy-burner behaviors and properly operate these burners in industrial furnaces or boilers while minimizing transfer and/or scale up risks. Ideally the pilot-scale is large enough to facilitate the scale up from the pilot results to the full-scale. By testing at the reasonably well-controlled, and measured pilot-scale it is possible to gather key insight that allows a technology developer to transition technology from one field to another. Because pilot-scale furnaces are typically larger and more complex than laboratory-scale combustors, they are often more costly and difficult to operate (but much less costly than full-scale). Although typically better instrumented than full-scale furnaces, the complexity of a pilot-scale system can make obtaining detailed information, such as temperature mapping challenging and time-consuming. In fact, many of the complexities
© 2011 by Taylor and Francis Group, LLC
Figure 27.9 Schematic of L1500. (From Bool, L., Kobayashi, H., Thompson, D., Eddings, E., Okerlund, R., Cremer, M., and Wang, D., 19th Annual International Pittsburgh Coal Conference, Pittsburgh, PA, 2002.)
2.0
0.6
7.3
Staging air injection points
that make the pilot-scale a good test of a technology’s performance at full-scale complicate obtaining detailed measurements.
4.
6.00
7.4
0 Z/ .00 R
0. 0 –1 0 .00 F
9
4.00
0.3
0
10
2.5
0
0.5
L X/
Flue 2.4
0
0.6
0
20
6
0.7
1.0
0.
2.5
0
3
7.4
0.
Coal burner
8.00
7
0 Z/ .00 R
10
.00
12. 00 8.7
7.5
0.
Sampling location
7
Middle furnace plane
0. 0 –1 0 .00
14. 00
0
20
10
nt
.32
10. 00
0
7
25
.00 16
0
0
0.4
4.9
0.3
0. –100 .00
0.
0.
18
0.6
L
X/
0.5
Upper furnace plane
Furnace
ida
A good example of a pilot-scale furnace is the University of Utah’s L1500 facility, shown schematically in Figure 27.9. The University of Utah pilot-scale combustion test furnace referred to as the “L1500” is a nominal 15 MMBtu/hr (4.4 MW) pilot-scale furnace designed to simulate commercial combustion conditions, particularly the thermal history of operating commercial coalfired boilers. The horizontal-fired combustor is 1.1 m × 1.1 m square and nearly 12.5 meters long. The walls have multiplelayered insulation to reduce the temperature from about 1650ºC on the fireside to below 60ºC on the shell side. The combustor is modular in design with numerous access ports and optional cooling panels in each section. This allows the flue gas temperature profile to be adjusted to better simulate commercial equipment. The access ports are used for visual observations, fuel and/ or air injection, and product sampling. The combustion facility includes the air supply system, water supply and cooling system, L1500 combustor, fuel supply systems, a flue-gas-cooling chamber, scrubber, and induced-draft fan and a stack. The dual concentric swirl burner used on the L1500, shown in Figure 27.10, is designed to provide excellent flame stability and offers a wide range of swirl stabilized flames. The burner consists of a central bluff body, a concentric coal pipe, a concentric opening for gas firing (if needed) and two concentric secondary air streams. Each air stream is independently controlled, metered, and can be varied over a wide range of swirl numbers. The L1500 was used to explore the use of oxygen to control NOx from power plants. One of the advantages of using the L1500, is the ability to explore the effect
536
Industrial Combustion Testing
0.6´´ Significant mixing at this interface 0.8´´ Limited mixing at this interface
Tertiary air
7.7´´
Secondary air Stabilizing gas Coal/primary air annulus Bluff body 2´´ Coal/primary air annulus Stabilizing gas
3.2´´
Burner centerline
Limited mixing at this interface
Secondary air
Significant mixing at this interface
Tertiary air
Figure 27.10 Schematic of the test burner in the L1500. (From Thompson, D. R., Bool, L. E., and Chen, J. C., “Oxygen Enhanced Combustion for NOx Control. Final Report,” DOE Report No. 889757, March 2004.) 200.00 180.00 160.00 140.00 NOx (g/GJ)
120.00 100.00 80.00 60.00 40.00 20.00 0.00 0.50
0.60
0.70 0.80 0.90 1.00 First stage stoichiometric ratio Baseline 5% replacement
1.10
1.20
10% replacement 15% replacement
Figure 27.11 NOx data with oxygen injection at L1500. (From Bool, L., Kobayashi, H., Thompson, D., Eddings, E., Okerlund, R., Cremer, M., and Wang, D., 19th Annual International Pittsburgh Coal Conference, Pittsburgh, PA, 2002.)
of a wide range of parameters on the effectiveness of staged combustion. Two different bituminous coals and one sub-bituminous coal from the Powder River Basin (Wyoming and Montana) were used to evaluate the effect of coal characteristics on the effectiveness of oxygen for NOx control. Burner parameters, such as the swirl number, were evaluated, as were the first stage stoichiometric ratio and residence time. Several methods were explored to introduce the oxygen into the first stage of a staged combustion system. As shown in Figure 27.11, experiments with oxygen-enhanced staging showed that the use of oxygen enhances the reactions leading to the conversion of fuel nitrogen to molecular nitrogen. These findings led directly to large-scale pilot testing and full-scale demonstration of the technology [2].
© 2011 by Taylor and Francis Group, LLC
Figure 27.12 Air Liquide 2 MW pilot furnace.
27.3.2.2 Example of a Pilot-Scale Furnace: ALICE Test Furnace Most industrial furnaces rely on burners with a unit power in the range 500 to 3,000 kW, operating in a wide range of temperatures depending on the considered process (crown temperatures usually run from 1100°C for aluminum melting to 1600°C for glass melting) and more or less air-in leakage. Accordingly, dedicated pilot furnaces are needed to accompany the development of new oxy-burners and characterize their performances and relevance for industrial furnaces. An example of such a pilot furnace operated by Air Liquide up to 2 MW with full oxy-burners is shown in Figure 27.12. This ceramic fiber-lined furnace has a low thermal inertia and can operate at a wall temperature up to 1600°C. The combustion chamber is 6.0 m long, and has a rectangular cross section of 1.5 by 2.0 m. These internal dimensions are large enough to minimize flame confinement effects and allow investigating the interaction between the flame aerodynamics and the furnace
Oxy-Fuel and Oxygen-Enhanced Burner Testing
recirculation zones. Heat extraction and furnace outlet temperature can be varied by insulating the sides of the water-cooled furnace floor, and by covering part of the water-cooled panels with a radiative screen consisting of silicon carbide panels. The furnace roof axial temperature profile is measured by 11 type-S thermocouples. The heat extraction profile is obtained by calorimetric measurements over 13 water-cooled floor panels. A first set of analyzers and a suction pyrometer measure the flue gas composition (O2, CO, CO2, and NOx) and temperature at the chimney outlet. A cross section of the furnace is shown in Figure 27.13. In addition, video cameras are positioned on the furnace side, and above the furnace horizontal exhaust section. Image processing is performed on raw video camera data to analyze and quantify flame shape and luminosity. Finally, an electric damper in the chimney duct controls the furnace pressure to simulate various air-in leakage conditions. For the detailed characterization of flame properties, a motorized arm is located on the right-hand side of the furnace as shown in Figure 27.12. This arm allows a sampling probe (connected to a second set of gas analyzers) Thermocouple for flue gas temperature
T1
Flue gas analysis
T11
T5
6.05 m Standard burner
Thermocouple
Chimney outlet
Figure 27.13 Cross-sectional view of 2 MW pilot furnace showing types and locations of measurements in a typical oxy-furnace.
Secondary O2
Fuel + Primary O2
Tertiary O2
Figure 27.14 Air Liquide low NOx ALGLASS SUN burner.
© 2011 by Taylor and Francis Group, LLC
537
and/or a suction pyrometer to be positioned at various locations along the furnace. Probes can be inserted at different depths and angles within the furnace to determine flame chemical composition and temperature (e.g., to determine the chemical size of the flame or to provide data for comparison with detailed numerical modeling). Further details on this furnace and its diagnostics can be found elsewhere [7]. The ALGLASS SUN is an example of an oxy-burner developed by Air Liquide using this pilot facility [8,9]. Dedicated to the oxy-boosting or the full conversion of glass furnaces to oxygen melting, this technology relies on a large separation of the fuel and oxidant streams as well as on the adjustable distribution of the oxidant among three different streams (primary, secondary, and tertiary oxygen) as illustrated in Figure 27.14. As the diameter of the secondary oxygen jet is more important than the tertiary oxygen jet, changing the split between these two injection locations allows the modification of flame momentum and length to achieve the flame characteristics required by the furnace geometry and layout. Indeed when increasing the oxidant repartition through the smaller and more distant injection (tertiary oxygen), flame momentum becomes higher and reactant jets entrain an important flow rate of combustion products: combustion is delayed and takes place over a longer and larger volume as illustrated in Figure 27.15, shot in the 2 MW Air Liquide pilot furnace. In addition, a stream of oxygen (primary oxygen) surrounds the fuel injection to guarantee flame stability, but is small enough to avoid high local temperatures. These characteristics allow the fine tuning of the thermal profile of the flame and help maximize heat transfer to the glass in all circumstances, while avoiding glass volatilization issues and drastically reducing NOx emissions. This is enabled by the large distance between the fuel and oxidant injectors along with the injection of at least 50% of the oxygen in the tertiary stream. Such an approach leads to large recirculation of the combustion products within the flame and accordingly, low peak flame temperatures and low thermal NOx production [8,9].
538
Industrial Combustion Testing
Figure 27.15 Front views of the Air Liquide ALGLASS SUN burner (natural gas version) operating at 2 MW in the Air Liquide pilot furnace. Tertiary oxygen ratio = 50% (left) and 75% (right).
27.3.3 Open Air Testing Open air testing, as the name implies, is the testing of a burner outside a furnace. As was discussed in the previous section, in furnace testing—whether laboratory or pilot-scale—can be used to measure a number of critical burner parameters. Outdoor burner testing is limited in what it can provide in terms of burner testing due to the fact the flame is not in an enclosure. For example, emissions measurements can not be taken on burners tested in this manner. Additionally, newer generation burners that employ in-furnace flue gas recirculation cannot be tested outdoors as the surrounding atmosphere is air and not hot, dilute mixtures of O2, CO2, H2O, and N2. While outdoor testing has its limitations, it can be a valuable tool for studying certain aspects of oxy-fuel burners. Outdoor burner tests can be used to characterize a variety of important burner design features. As oxy-fuel burners have moved away from water cooling to nonwater-cooled designs, refractory burner blocks are required to protect metallic burner components from high temperatures encountered in some furnaces (e.g., glass furnaces). As has been discussed elsewhere [10], oxy-fuel burners typically utilize jets of either oxygen, fuel, or both. Jet theory indicates that turbulent jets will entrain surrounding gases into the jet resulting in expansion of the jet [11,12]. The combination of a turbulent fuel and/or oxygen jet recessed in a refractory port results in a design with the potential to entrain furnace gases into the burner port. If these furnace gases contain condensable species, condensation can occur, leading to fouling of the port and reduced burner performance. Outdoor testing is a tool that can be used to study the entrainment of gases from outside the burner block into the port. This can be done by seeding the area around the port with fine particles (mist, smoke, etc.) and observing the movement of the particles in the near port region. Another parameter that can be evaluated with outdoor testing is flame liftoff. Oxy-fuel burners that are designed to operate at furnace temperatures below 1400°F (760°C) require a mechanism to stabilize the
© 2011 by Taylor and Francis Group, LLC
Figure 27.16 Testing flame stabilization of full-scale oxy-fuel burner over wide range of firing rates.
flame. This stabilization method can be studied in fullscale burners at high firing rates in outdoor testing facilities. Figure 27.16 shows a picture of an oxy-fuel burner operating at 45 MMBtu/hr (13.2 MW) with the flame clearly visible at the exit of the burner. Burners with poor flame stabilization can result in excessive combustion noise. Additionally, significant liftoff can result in loss of flame signal by the UV scanner or burner management system resulting in a burner shutdown. Finally, some oxy-fuel burner designs employ nozzles and mixing strategies that result in coherent jets. These jets do not behave like traditional jets in that they do not entrain surrounding gases for many diameters downstream of the nozzle. These jets can be subsonic or supersonic. Outdoor burner tests can be used to study the efficiency of various nozzle/mixing strategies on coherent jet length. Jet length is typically characterized by dynamic pressure along the jet centerline. This pressure can be obtained by mounting the burner on a stand with
Oxy-Fuel and Oxygen-Enhanced Burner Testing
Figure 27.17 Praxair’s CoJet® lance in outdoor test rig used to measure dynamic pressure along jet centerline.
539
Figure 27.18 Permanent oxygen supply system capable of supplying 250 tpd O2 to a furnace. System shown utilizes atmospheric vaporization.
a water-cooled, Pitot static probe that can be translated in three directions. Figure 27.17 illustrates an example of this test setup on a Praxair CoJet® lance [13]. 27.3.4 Customer Testing The ultimate goal of all development work is to bring solutions to customer’s processes. The final step in this process is demonstrating the benefits of an oxy-fuel burner in a full-scale furnace. Since oxy-fuel combustion benefits are relatively generic and potentially applicable across a wide range of processes and industries the size of the systems can vary over several orders of magnitude. For example, small furnaces used in secondary aluminum processing facilities can use several tons of oxygen per day while larger glass furnaces and steel furnaces can use several hundred tons of oxygen per day. Given the wide range of oxygen requirements the logistics around a demonstration must be evaluated to ensure that meaningful results will be available. Defining the oxygen usage and length of test is a key step in helping to identify the method of supply and the feasibility of various potential sites. There are typically two primary methods of supplying oxygen to a facility for a demonstration. The first is delivery of liquid oxygen via a truck to an onsite storage tank. The onsite storage tank can either be a temporary system or a permanent tank. Liquid oxygen supply systems require vaporizers to convert the liquid to gas for delivery to the furnace. Depending on the proximity and availability of steam these vaporizers can be either atmospheric or steam-assisted. Steam-assisted vaporizers have a significantly smaller footprint for comparable vaporization capacity. A picture of a temporary oxygen supply system capable of supplying up to 100 tpd of oxygen to a process
© 2011 by Taylor and Francis Group, LLC
Figure 27.19 Permanent oxygen supply system capable of supplying 100 tpd O2 to a furnace. System shown utilizes steam-assisted vaporization.
was shown previously in Figure 27.3. Figure 27.18 shows a picture of a permanent liquid supply system capable of delivering 250 tpd of oxygen to a furnace. The system shown in this picture utilizes atmospheric vaporizers. Figure 27.19 shows a picture of a permanent liquid supply system capable of delivering 100 tpd of oxygen to a furnace. This system uses steam-assisted vaporizers. All three systems require some site work that is defined after a site survey by the oxygen supplier. Tank volumes are sized based on a number of factors including proximity to the oxygen supply source and backup volume required. The second method of supply is by gaseous oxygen. This supply can be from a nearby oxygen production facility or a nearby oxygen pipeline. Both would require running a new oxygen pipe to the furnace if one is not already in place. For larger furnaces requiring greater amounts of oxygen, this is often the preferred method if a pipeline or plant is reasonably close. By the time the testing reaches commercial scale the objectives of the test have transitioned to validation of
540
Industrial Combustion Testing
Air8 2650–2700 2600–2650 2550–2600 2500–2550 2450–2500 2400–2450 2350–2400 2300–2350 2250–2300 2200–2250 2150–2200 2100–2150 2050–2100 2000–2050
the temperatures to lower values. These temperatures can be compared to model predictions of furnace conditions. Additionally, water-cooled gas extraction probes connected to portable gas analyzers can be inserted in various locations within the furnace or flue gas ducting to generate a map of gas species at various planes/ locations within the furnace. On furnaces firing fuels with an ash content, ash samples can be collected and evaluated for loss on ignition to measure differences in carbon/fuel burnout between the base case and the oxygen-enhanced case. This information can be used to optimize oxygen injection systems (either individual burners or biasing between different locations/burners within the furnace). Finally, in some furnaces it is possible to insert a grid of thermocouples on a wall (or walls) of the furnace to characterize the wall temperature profile (and thus radiation heat transfer profile) between the air-fired baseline and the oxy-fuel-fired condition.
Back wall
Burner wall
Best with oxygen
Back wall
Best before
Burner wall
predicted benefits rather than measuring specific burner characteristics. Since these furnaces are in commercial operation there is little flexibility to operate the burners outside the requirements set by the process. The sources of information used to evaluate burner performance on commercial scale installations typically falls into one of three categories: (i) combustion performance in the oxy-fuel fired zone, (ii) flue gas emissions from the furnace (including any particulate matter), and (iii) product quality. Combustion performance of the oxy-fuel fired furnace (or zone(s) in a multizone furnace) can be evaluated by a variety of methods. One of the most valuable inputs in evaluating system performance is visual observation of the flames. In many cases, the full-scale test is the first demonstration of an oxy-fuel burner in its true environment. Access to visual observations through viewports, doors, in furnace cameras, and other vantage points often provides very meaningful insight into that specific application. Figure 27.20 shows several images of the near burner region of one of 16 burners on a 125 MW, coal-fired utility boiler with Praxair’s oxygen enhanced combustion low NOx technology installed [2]. The image on the left shows the combustion system operating with air only. The image on the right shows the combustion system operating with a small percentage of the oxygen required for combustion lanced in to the root of the flame. These images show that the root of the flame is much brighter and better anchored to the burner. These characteristics are critical to the system’s performance. While furnaces typically do not have many access ports, a variety of probes can be inserted through ports that are available to measure key process variables that can provide feedback used to fine-tune burner settings. For example, suction pyrometers, such as those described in Section 27.5, can be inserted in the furnace or flue gas duct to measure true gas temperatures as thermocouples are usually placed at locations where significant radiation losses can occur thereby biasing
Figure 27.20 Images of air only and oxygen-enhanced coal flames on one of 16 burners in a 125 MW coal fired utility boiler. (From Bradley, J. L., Bool, L. E., and Kobayashi, H., 29th International Technical Conference on Coal Utilization and Fuel Systems, Clearwater, FL, 2004. With permission.)
DOC8 2650–2700 2600–2650 2550–2600 2500–2550 2450–2500 2400–2450 2350–2400 2300–2350 2250–2300 2200–2250 2150–2200 2100–2150 2050–2100 2000–2050
Figure 27.21 (See color insert following page 424.) Roof temperature profiles as measured by a grid of 20 thermocouples for air-fired operation (left) and oxy-fuel-fired operation (right).
© 2011 by Taylor and Francis Group, LLC
541
Oxy-Fuel and Oxygen-Enhanced Burner Testing
Figure 27.21 shows an example of the roof temperature profile of a furnace fired on both air and Praxair’s DOC firing system. A total of 20 thermocouples were placed on the furnace roof. The roof temperature profiles were used to fine-tune the oxygen and fuel nozzle designs to yield a temperature profile that was more uniform across the furnace (left to right in figure), lower temperature at the inlet to the furnace (bottom of figure) without overheating the wall in any particular area [14]. As the figures indicate this was accomplished. The final and most important source of information that must be monitored is the product quality. This information is product/application specific; however, each process has very tight requirements on the heating process that is taking place within the furnace. Furnaces that process a wide variety of products must be monitored for enough time to observe a representative population of the products that are charged to assure the new combustion system is capable of meeting the heating demands of the process.
27.4 Modeling Tools Modeling tools are used throughout the technology development process. These tools can be as simple as a spreadsheet-based heat and mass balance, or as complicated as a detailed cold flow model. Computerbased tools, such as kinetic or CFD modeling, can be instrumental in evaluating and understanding technology. A detailed discussion of these tools is outside the scope of this chapter. Physical models are used to explore flow through a burner or furnace, or reaction and mixing in a furnace. For this discussion these models are broken into cold flow models and dense phase modeling. A cold flow model is essentially a scaled physical model of a process that is used for flow visualization and evaluating issues such as erosion. Dense flow models are used for flow visualization and reaction modeling. These two models are discussed in more detail in the following sections. 27.4.1 Cold Flow Modeling A cold flow model is essentially any physical model that is used to explore key technology issues without including combustion. These models are typically crafted with transparent materials to allow flow visualization. Key operating characteristics, such as burner momentum, are reflected in the model by carefully scaling the burner flows. Physical differences, such as density differences, can be modeled by using gases with differing densities. A good example of this type of model is a
© 2011 by Taylor and Francis Group, LLC
f urnace being considered for conversion to oxy-fuel firing. By scaling the process flows and using markers, such as smoke or bubbles, it is possible to visualize the gas flow throughout the furnace. This visualization can provide key insight on the impact of a particular technology on furnace operation. For example, if the furnace is designed such that the flow is segregated, then changing the burner momentum may create too much mixing. On the other extreme, if the furnace is normally very well mixed, reducing the flows may negatively impact mixing. These models are normally characterized by their physical simplicity, but may require detailed scaling to ensure the flow fields are representative of the target furnace. Another type of cold flow modeling is designed to test particular equipment components. For example, if there is concern about erosion of a nozzle by entrained solids, it is possible to design a cold flow model to evaluate how fast (and where) the nozzle is likely to erode. Atomization by a burner can be measured by operating the burner under appropriate (scaled) conditions. These tests are typically done as final validation of the design before installation at a customer’s furnace. 27.4.2 Dense Phase (PLIF) Modeling Another technique used to study oxy-fuel burners is Chemically Sensitive–Planar Laser Induced Fluores cence (CS-PLIF) [15]. The CS-PLIF is a noninvasive, quantitative method for visualizing flow, mixing, and chemical reactions in complex geometries typical of commercial furnaces. The technique provides information on burner/burner interactions, burner/wall interactions, and burner interactions with the furnace flowfield/recirculation pattern. This technique utilizes an acid solution, a base solution, a dye tracer that fluoresces over a narrow range of pH and a source of coherent radiation (typically an argon ion laser). The acid and base are mixed in strengths such that when they combine in stoichiometric proportions consistent with the combustion system that is under evaluation, the pH will be within the narrow range that the dye tracer fluoresces. For example, when an oxy/natural gas system is studied, the acid and base are mixed such that when two parts of the oxidizer stream mix with one part of the fuel stream, the pH is within the narrow window in which the dye tracer fluoresces. This technique was used successfully in a project in which Praxair’s DOC system was installed near the charge end of a billet reheat furnace [16]. The objective of the work was to evaluate burner designs that would ensure good flame penetration into the cross flow of waste gases leaving the furnace while preventing unburned
542
Industrial Combustion Testing
fuel from entering the stack. Figure 27.22 below shows an image of the Plexiglas model of the furnace. Figure 27.23 shows several plan view images with the dye introduced through the burner closest to the flue. Figure 27.23a is a schematic of the furnace illustrating the region in view. Figure 27.23b illustrates the fuel jet trajectory with the original firing strategy. The results suggest that fuel is approaching the furnace flue stack with this firing strategy. Figure 27.23c shows the same
Flue
Steel
DOC Burner
Air burner
image with a modified firing strategy to prevent fuel from approaching the flue gas uptake shaft.
27.5 Key Process Variables and Measurement Techniques In oxy-burner testing, generally two types of measurements are involved: (i) measurements inside flames and (ii) global measurements in the furnace. In-flame measurements include chemical composition, flame temperature profile, and optical properties. Global measurements include flame shape, emissions, heat flux, and measurements taken on the load. Figure 27.13 shows typical measurements and their locations in an oxy-pilot furnace. 27.5.1 Fuel and Oxidant Properties
Figure 27.22 Plexiglass model of billet reheat furnace used in CS-PLIF study. (a)
Fuel type and properties are critical in the selection and use of a particular test method. Selected test fuel should closely represent the fuel used at the customer’s site; (b)
Flue
DOC burners
Unburned fuel (c)
Figure 27.23 (a) Schematic of furnace illustrating field of view, (b) fuel jet trajectory for original burner design, and (c) modified firing strategy to avoid fuel near furnace uptake shaft.
© 2011 by Taylor and Francis Group, LLC
543
Oxy-Fuel and Oxygen-Enhanced Burner Testing
otherwise, significant errors could arise. As an example, natural gas, fuel oil (US #2) and heavy fuel oil (US #5) are the most common fuels used in oxy-burners. However, the composition of natural gas varies across the regions within the United States and throughout the world. Other fuels used are propane, liquefied petroleum gas, producer gas, coke oven gas, coal, and petcoke. Fuel properties such as specific gravity, heating value, elemental composition, viscosity, ash, moisture content, and autoignition temperature are important to be made available prior to testing. Wobbe index is an indicator that provides a basis for comparing different fuels. Oxygen purity at the burner will depend on the method of supply. Oxygen from a liquid supply system is typically > 99.5% purity. Oxygen purity from a vacuum pressure swing adsorption supply system is typically between 90% and 94% purity with the balance split almost equally between nitrogen and argon. While these are relatively small quantities of impurities, burner emissions (depending on burner design) may be impacted by the small quantity of nitrogen in the oxygen stream. Tests should be conducted with an oxidizer stream consistent with that which will be employed in the commercial version to ensure performance will meet the requirements. 27.5.2 Flame Geometry Flame shape (for instance, rounded vs. flat) is appropriately designed to meet the heat transfer requirement of the process. Unlike air-fired burners, oxy-burners typically produce bright, intense, and shorter flame. In customer furnace tests, flame must be sufficiently long and wide enough to reach and cover the entire load. Note that in situations where the flames are in direct contact with the load, the flame length may change as the load is being melted. Laboratory and pilot-scale testing must be designed to accommodate these requirements. Open-air testing, as discussed in the previous section, provides a quick and impressive visual observation of the flame. Flame geometrical data such as global length, width, and shape can easily be measured. However, open-air testing is largely influenced by the ambient wind conditions. When testing is conducted inside a test furnace, flames usually become longer and more stable because of the hot environment and represent conditions closer to the commercial furnace. Flame images are taken using endoscope systems, water-cooled chargedcoupled device (CCD) camera with pinhole lens, and special high-temperature cameras mounted at different locations of the furnace (side view and front view). Figure 27.24 presents a side view of a typical oxy-flame from a test furnace taken using a high-temperature camera. Due to higher temperatures associated with standard oxy-combustion, the refractory tiles become much brighter, and hence, emit significant radiation, typically
© 2011 by Taylor and Francis Group, LLC
Figure 27.24 Typical oxy-natural gas flame from a test furnace captured using a water-cooled, high-temperature camera system. (Courtesy of Air Liquide.)
in the visible region making it harder to visualize the flame. This background radiation must be eliminated in order to capture the true image of the flame. Further, the flame usually contains radiation from instantaneous species such as OH, CH, and C2. Hence, appropriate filters such as UV, visible light filters, and intensity attenuator must also be employed. A distinct difference between the test furnace and process furnace is the amount of heat absorbed and emitted by the load. Many industrial processes do not absorb heat at the same wavelength. For instance, in glass furnaces, some clear glasses absorb less heat than that by green/amber glass in the visible range. In an aluminum melting furnace, the heat absorption wavelength largely depends on the dross level. To understand such issues and provide efficient heat transfer to the load in a test furnace, monochrometers can be used to identify and select at which wavelength the load absorbs the heat better and hence, determine the heat transfer efficiency. 27.5.3 Heat Transfer Measurement For a detailed understanding of heat transfer interactions among flame, load, and refractory tiles, it is necessary that in-flame temperature (single-point and 2-D mapping), stack gas temperature, and heat flux measurements are taken. Single-point temperature measurements are commonly taken using thermocouples and suction pyrometers. Noncontact type thermal radiation pyrometers are also widely used to make singlepoint measurements and 2-D mapping. These are ideal for the quick measurement of burner parts, burner tiles, and refractory tile temperature. Due to the high concentrations of CO2 and H2O in oxy-fuel fired systems,
544
Industrial Combustion Testing
it is important to utilize a pyrometer that employs wavelengths outside the strong absorption bands of CO2 and H2O so as not to bias the measurements. 27.5.4 In-Flame Temperature Measurement In-flame temperature measurement provides useful insight into estimation of radiative heat release from flames, kinetics of the reaction, soot formation, and NOx formation. In oxy-burner testing, thermocouple and suction pyrometer are extensively used to take point measurements within the flame. Recently, laser-based advanced diagnostic methods have been developed to map 2-D temperature field of the flame within the furnace [17]. Thermocouples work based on Seebeck effect. When two dissimilar metals are brought in contact at a small region (known as bead), a voltage is produced across the terminals; the voltage is a function of bead temperature. This voltage output is typically converted to a 4–20 mA signal in a transmitter and sent to PLC for visualization and logging. Different types of thermocouples are available based on the metals selected. Usually Type R/S and B thermocouples are employed in oxy-burner testing. Note that thermocouple measurements must be corrected for radiative losses from the bead and conductive losses through the wire. Since the temperature involved in oxy-furnace is high, omission of this correction would result in significant error. Also, the thermocouples are usually enclosed in ceramic sheath to protect the tip deterioration from the process. A suction pyrometer provides a more accurate measurement of the local flame temperature. Here, a thermocouple bead is shielded within a tube and thus the radiant exchanges with the surroundings are minimized. Combustion gases are made to flow over the thermocouple bead at higher velocities (using a suction mechanism) so that only convective effects are dominant. Since radiative corrections are not needed, suction pyrometer measurements usually provide better results than thermocouples. Figure 27.25 shows a schematic of the International Flame Research Foundation (IFRF) suction pyrometer.
Cement
Thermocouple junction
Point measurements are taken along the centerline of the flame at different lengths (for example, 25%, 50%, and 75% of flame length). Thermocouples and/or suction pyrometers are inserted into the furnace to reach the flame centerline. 27.5.5 Temperature Distribution on Furnace Hearth and Crown When oxy-burner testing is conducted in lab-scale or pilot-scale furnace, at several locations on hearth and crown, thermocouples (Type R/S) are typically installed to assess the uniformity of heating. Simulated heat extraction is often employed to represent process load. Hearth temperature profile along with heat extraction data provides an estimation of the efficiency of the burner in heating of the load (i.e., the heat available for process). For instance, Leroux et al. [8] measured the furnace roof temperature using 11 Type-S thermocouples in the test furnace. In oxy-burner fitted customer furnaces, measurement of hearth temperature is tricky since the floor will be filled with load (such as glass, aluminum, and steel). In such furnaces, thermocouples are installed with appropriate shield to measure process temperature. Hence, the laboratory data is useful in estimating a priori the heat transfer to the load. 27.5.6 Stack Gas Temperature Measurement Stack gas temperature measurements are taken to provide a basis for furnace continuous emission monitoring (CEM) and also to setup safety interlocks for emergency shutdown. Hence, accurate measurement of stack gas temperature is crucial to the operation of the furnace. The measurement location must be chosen in such a way that the bulk of the flue gases are sampled and the radiation from refractory tiles is minimal. A suction pyrometer is preferred to a thermocouple for stack gas measurement. Although not recommended, if an unshielded thermocouple is used, it must be corrected for radiative losses from the bead; otherwise, the measurement would result in errors up to the order of 200°F to 400°F. 27.5.7 Heat Flux Measurement
Radiation shields
Cement Operating length Water Water Suction inlet outlet line
Figure 27.25 Schematic of the IFRF suction pyrometer. (From Fricker, N., IFRF Online Combustion Handbook, Reference number: IFRF Doc No/C76/ y/1/17, Combustion File No. CF145. With permission.)
© 2011 by Taylor and Francis Group, LLC
Heat flux measurements are taken to understand the uniformity of heat transfer and hence, efficiency of the oxy-burner. Heat flux probes are commonly used in such measurement. These probes usually have wide angle and measure radiative or total (radiative and convective) heat fluxes. Sensor elements in the probe use a circular foil disk on the front side facing flue gases. The edge of the disk is continuously in contact with a heat sink. The temperature difference between the center of the foil
Oxy-Fuel and Oxygen-Enhanced Burner Testing
and circumference is measured, which is proportional to heat flux. Here, by using a special window, convective heat flux could be eliminated and only radiant heat flux is measured (such types of meters are called radiometers). Thermopile-type probes in which heat conduction is established through a plug are also used to measure heat flux [18]. Usually, a uniform heat flux is desired to produce better yields. Most industrial processes take advantage of flue gas to provide heat flux distribution inside the furnace. Note that in oxy-combustion, the flue gas volume is significantly reduced (by more than three times that of air-combustion). Hence, considerable efforts are needed to properly manage the flue gas recirculation pattern within the furnace and heat flux. Leroux et al. [8] presented a patented burner that provides a better heat flux profile by efficiently managing the injection of reactants. In an oxy-furnace, heat balance calculations are typically done in order to quickly measure global heat transfer properties. In this approach, water jackets are installed on the side or hearth of the furnace to simulate process heat loading. Cooling water is continuously circulated inside the jacket and water temperature measurement at inlet and outlet are taken. Based on the measured data and the knowledge of water flow rate, heat extraction by cooling water can be calculated, and in turn, the global heat balance for the furnace is established. The accuracy of this technique is very good although it does not provide detailed local information.
545
are needed, the entire line must be heat-traced. The insitu method provides wet-based measurements since no extraction is required. As an example, Zirconia probes are widely used for measuring exhaust gas oxygen concentration on a wet-basis. Tunable diode laser absorption spectroscopy (TDLAS) can also be used to take simultaneous in-situ measurement of trace species such as OH, CO, and NO and major species such as O2 and H2O [19]. Concentrations of CO and CO2 are usually measured using a nondispersive infrared (NDIR) analyzer. Many CO2 analyzers are only capable of measuring CO2 up to ~20 vol.% (dry), which is more than adequate for most air-fired systems. Since oxy-fuel fired systems have low concentrations of N2, the resultant concentration of CO2 on dry-based measurements can be > 70 vol.%. Care must be taken in specifying a CO2 analyzer with a range suitable for the measurements that are required. The NOx concentrations can be measured using a chemiluminescence analyzer, but NDIR can also be used when higher emissions are expected. Working principles of these techniques can be found in standard texts. Oxygen concentrations are obtained with a paramagnetic sensor. Figure 27.26 shows a photograph of the typical gas analyzer rack used. In oxy-burner testing, the quantity of nitrogen present in the oxidizer is typically less than 3–5%. Most fuels typically have small quantities of nitrogen
27.5.8 Emission Measurement In an oxy-combustion setup, the components that are of interest are O2, CO, CO2, NOx, SOx, ash, unburned hydrocarbon, H2O, and particulates, depending on the type of the fuel employed. Some of the components are harmful to the environment and such pollutant species data from the furnace must be reported to the local environmental regulatory board. Regulations are often imposed by the U.S. EPA and local agencies. Emission measurements are generally taken inside the flame and near the exit of the stack. There are two methods available for emission measurement: (i) flue gas sampling and (ii) in-situ measurement. In the sampling method, the flue gas stream or a sample from inside the flame is drawn, filtered, conditioned, and sent to a series of analyzers to measure its composition. Here, care must be taken in order not to let the sample gas cool down; otherwise, the measurements could lead to significant errors. With the in-situ method, the sensor is located directly at the point of measurement and the composition is obtained directly. Measurements can be taken on wet or dry basis. In the flue gas sampling method, if wet-based measurements
© 2011 by Taylor and Francis Group, LLC
Figure 27.26 Photograph of a typical gas analyzer rack. (Courtesy of Air Liquide.)
546
as well. Practically, all furnaces will have some air i nfiltration. In many cases, air infiltration is the largest source of nitrogen found in the furnace. Accordingly, it is important to understand the amount of air leakage into the furnace. Helium tracing is one of the best approaches to accomplish this. Here, a known amount of helium is injected into the furnace via oxy-burners. A leak detector such as a mass spectrometer is used to continuously measure the concentration of helium at the exhaust. This provides quantitative data on air infiltration. 27.5.9 In-Flame Chemical Composition Measurement Chemical composition within the flame is done for research purposes. Measurements are usually taken at 25%, 50%, and 75% of the flame length. The concentration data in conjunction with temperature data would help explain the mechanics by which pollutant species are formed and destroyed inside the flame. In addition, one of the most common definitions of flame length is based on iso-CO contours. The CO concentrations below 0.5% usually indicate the end of the chemical flame. The length of the chemical flame is usually longer than the visible flame.
Industrial Combustion Testing
appropriate shut down protocol can be defined. It is important to recognize that oxy-fuel-fired furnaces have much lower N2 concentrations than air-fired furnaces. Consequently, gas species measured on a volume basis (i.e., ppmvd) may be higher than those of the air-fired system. When these measurements are corrected to a lb/MMBtu or kg/MW basis to account for the dilution effect of the N2, the emission rates of oxy-fuel-fired systems are often lower than those of air-fired systems. 27.5.11 Advanced Flame and Aerodynamics Characterization In addition to the above conventional characterization techniques, many advanced methods provide additional insights, mostly based on laser diagnostics and imaging techniques. Flow patterns and flame dynamics can be investigated using laser doppler velocimetry, particle image velocimetry, laser tomography or Schlieren techniques. Further, detailed imaging of the flame with appropriate filters (OH, CH, C2) and/or planar laserinduced fluorescence reveals chemical composition of the flame.
27.5.10 Stack Gas Sampling Stack gas sampling provides the measurement of the level of pollutants leaving the furnace. Most plants are required to follow certain CEMS regulations (such as U.S. EPA) for the instruments and report data to government on a regular interval basis. Sample probe must be located in a place where a well-mixed flue gas stream is available. Figure 27.27 presents an image of a sampling probe with a thermocouple integrated into the probe. Stack gas sampling is also used to monitor the performance of the furnace—denoting oxidizing or reducing conditions. For instance, flue gas oxygen concentration can be tied to safety interlocks and the
27.6 Oxygen Safety in Burner Testing Combustion using pure oxygen or oxygen-enriched air differs significantly from air-based combustion. Although oxygen is by itself nonflammable and chemically stable, it strongly promotes combustion. Several materials that do not generally burn in air can burn in oxygen. In view of potential hazards, oxy-burners are carefully designed by following applicable NFPA, CGA, ASTM codes and internal corporate standards. If the guidelines set forth by the manufacturers are followed and oxy-burner operators are adequately trained, oxygen is safe to handle and use. 27.6.1 Oxygen System Design and Material Compatibility
Figure 27.27 Probe utilized to measure emissions and temperature in stack. (Courtesy of Air Liquide.)
© 2011 by Taylor and Francis Group, LLC
When used for oxygen service, several materials such as pipes, pipe fittings, flanges, gauges, and gaskets could act as fuel and potentially burn if an ignition source is met. Even without the spark, based on the way that the oxygen system is operated, it is possible to create ignition. The goal in oxygen system design is to identify the right materials and minimize chances of undesired ignition. An oxygen system must be designed only by qualified engineers. An oxygen system includes storage tank,
547
Oxy-Fuel and Oxygen-Enhanced Burner Testing
supply system (pipe lines), and distribution station (valve train). Valve trains handling oxygen should be designed according to the codes and standards listed below by including applicable safety interlocks. When testing burner prototype, flexible pipe, fittings, and valves must also be compatible for oxygen service. Standards such as CGA G-4.4, EIGA 13/02, and ASTM G88 describe oxygen system design procedure and offer practical guidelines. The ASTM G94 and ASTM G63 provide guidelines to evaluate metals and nonmetallic materials for oxygen service. The NFPA 53 presents recommended practice on materials and equipment for oxygen service. The NFPA 86, which presents standards for ovens and furnaces, also includes some guidelines for oxygen service. 27.6.2 Oxy-Cleaning, Operation, and Maintenance All parts in contact with oxygen must be free from contaminants, dirt, metal shavings, or any other foreign material. Standard procedure such as CGA G-4.1 and EIGA 33/06 should be employed to clean parts for oxygen use. All cleaned parts must be labeled as “Cleaned for oxygen service.” For testing, standard operating procedures have to be defined and attention must be paid to details. Two examples of careful operation are given below:
1. Block valves in the oxygen line should be opened very slowly first, to avoid adiabatic compression downstream of the valve. 2. Before turning on an oxy-burner in a furnace, the furnace must be purged with air or nitrogen at least four times the volume of the furnace as required by NFPA 86.
In addition to the safety standards followed for air-based combustion (as defined in NFPA 86), flame detection technique employed must be able to monitor oxy-flame. During testing, it is essential to monitor stack O2 and CO concentration. This provides a quick insight into the conditions of the furnace. Safety interlocks should also be tied to these measurements and the furnace must be shut down if undesired conditions do develop. In oxy-burner testing, if the procedures are not followed properly, it is possible to create an oxygen-enriched environment, which could lead to potential fire and explosion hazards. Another important issue with oxy-combustion is the lack of nitrogen and lesser dilution of combustion products, which could create potential CO poisoning. As an example, if the combustion ratio control is off by 10% in air combustion, only 2% of oxygen is needed to reattain the stoichiometry. However, in oxy-combustion, all of 10%
© 2011 by Taylor and Francis Group, LLC
Figure 27.28 Oxygen fire in a valve train due to improper maintenance. (Courtesy of Air Liquide.)
oxygen is needed, and hence, a significant amount of CO can be generated. Note that without nitrogen dilution the CO level would further be higher and lead to more hazardous situations. Hence, the use of proper personal protective equipment such as O2 and CO monitors is strongly recommended to the test operators. Additionally, because the potential for flash fires is higher in the presence of oxygen, it is recommended that anyone working in the area wear flame retardant clothing such as NOMEX® or INDURA®. Further, oxyflames are very bright; hence, proper eye wear must also be chosen to avoid potential eye strain. For instance, in order to directly view the flame, safety glasses with 2% visual light transmittance and more than 95% IR absorption are recommended. All components must be in mechanically good working condition. Otherwise, any accumulated dirt or foreign substance could lead to potential oxygen fire. For instance, conical strainers are typically installed in oxygen valve train at the location specified by the code. Just after start-up and during testing, this strainer should be checked for cleanliness to ensure that oxygen available up to the valve train is free from dirt. Figure 27.28 shows a typical oxygen fire in valve train due to improper maintenance and Figure 27.29 shows an example of an unclean strainer in an oxygen valve train. Such strainers must be removed from the system and cleaned or replaced. 27.6.3 Training All personnel involved in oxygen handling must be trained properly. Training should include identification of potential oxygen hazards, oxy-burner operation, material selection, cleaning for oxygen service, and
548
Industrial Combustion Testing
thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.
References
Figure 27.29 An example of an unclean strainer in oxygen valve train needing maintenance. (Courtesy of Air Liquide.)
xy-burner parts maintenance routine. Engineering o contractors and welding crew are required not only to undergo the training but also to obtain certification for installing oxygen pipeline.
27.6.4 Safety in Working With Partners Many times, oxy-burner testing is done at partners’ site (industry or university) that may not be familiar with oxygen use. In such situations, oxy-burner manufacturer and oxygen supplier should follow safety standards in installing and testing oxy-burners. Proper training should also be given to the partners.
Disclaimer This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency
© 2011 by Taylor and Francis Group, LLC
1. Bool, L., Kobayashi, H., Thompson, D., Eddings, E., Okerlund, R., Cremer, M., and Wang, D. 19th Annual International Pittsburgh Coal Conference, Pittsburgh, PA, 2002. 2. Bradley, J. L., Bool, L. E., and Kobayashi, H. 29th International Technical Conference on Coal Utilization and Fuel Systems, Clearwater, FL, 2004. 3. Farzan, H., McCauley, K. J., Varagani, R., Prabhakar, R., Periasamy, C., and Perrin, N. “Oxy-Coal Combustion Pilot.” IEAGHG International Oxy-Combustion Net work, Yokohama, Japan, March 3–5, 2008. 4. Kalcevic, R., Tsiava, R., Fujinuma, R., Periasamy, C., Prabhakar, R., and Todd, G. “Industrial Results of ALGLASS FH Oxy Fuel Forehearth Burner Operation.” 69th Glass Problems Conference, Columbus OH, November 4–5, 2008. 5. Ryan, H. M., Riley, M. F., and Kobayashi, H. “Dilute Oxygen Combustion-Phase I Report.” Report No. DOE/ ID/13331-1, October 1997. 6. Ryan, H. M., Francis, A. W., Riley, M. F., and Kobayashi, H. “Development of the Dilute Oxygen Combustion Burner for High-Temperature Furnace Applications.” Paper Presented at the 4th International Conference of Technologies and Combustion for a Clean Environment, Portugal, July 7–10, 1997. 7. Dugué, J., Von Drasek, W., Samaniego, J. M., Charon, O., and Oguro, T. “Advanced Combustion Facilities and Diagnostics.” American/Japanese Flame Research Committees 1998, International Symposium Maui, Hawaii, October 11–15, 1998. 8. Leroux, B., Perrin, N., Duperray, P., Recourt, P., Tsiava, R., and Todd, G. “ALGLASS SUN: An Ultra Low NOx Oxy Burner for Glass Furnaces with Adjustable Flame Length and Heat Transfer Profile.” 64th Glass Problems Conference, Urbana–Champaign, IL, October 28–29, 2003.
Oxy-Fuel and Oxygen-Enhanced Burner Testing
9. Leroux, B., Constantin, G., Joumani, Y., Zucchelli, P., Simon, J.-F., and Tsiava, R. “Environmental Performances of Air Liquide ALGLASS Burners: From Modeling to In-Situ Analysis.” 20th ATIV Conference, Parma, Italy, September 14–16, 2005. 10. Kobayashi, H., and Tsiava, R. “Oxyfuel Burners.” In Industrial Burners Handbook, edited by C. E. Baukal, 693– 723. Boca Raton, FL: CRC Press, 2004. 11. Ricou, F. P., and Spalding, D. B. “Measurements of Entrainment by Axisymmetrical Turbulent Jets.” Journal of Fluid Mechanics 11, no. 1 (1961): 21–32. 12. Tacina, K. M., and Dahm, W. J. A. “Effects of Heat Release on Turbulent Shear Flows. Part 1: A General Equivalence Principle for Non-Buoyant Flows and Its Application to Turbulent Jet Flames.” Journal of Fluid Mechanics 415 (July 2000): 23–44. 13. Anderson, J. E., and Snyder, W. J. Coherent Jet Combustion, U.S. Patent 5,100,313, March 31, 1992. 14. Black, E., Erfurth, F., Kitko, G. T., Hernandez, M., Kelly, J., Rosen, L. J., and Tian, K. “Installation of a 145 MMBtu/ hr Oxyfuel Firing System on the #2 Reheat Furnace at ArcelorMittal’s 84” Hot Strip Mill,” 39. AISTech May 4–7, 2009 Proceedings, pp. 39–46, 2009.
© 2011 by Taylor and Francis Group, LLC
549
15. Dahm, W. J. A. “Experiments on Entrainment, Mixing and Chemical Reactions in Turbulent Jets at Large Schmidt Numbers.” PhD diss., California Institute of Technology, 1985. 16. Riley, M. F., and Ryan, H. M. “Dilute Oxygen CombustionPhase 3 Report.” Report No. DOE/ID/13331-3, May 2000. 17. Stricker, W. P. “Measurement of Temperature in Laboratory Flames and Practical Devices.” In Applied Combustion Diagnostics, edited by K. Kohse-Hoinghaus and J. B. Jeffries, New York: Taylor & Francis, p. 155, 2002. 18. Fricker, N. IFRF Online Combustion Handbook, Reference number: IFRF Doc No/C76/y/1/17, Combustion File No. CF138. 19. Allen, M. G., Furlong, E. R., and Hanson, R. K. “Tunable Diode Laser Sensing and Combustion Control.” In Applied Combustion Diagnostics, edited by K. Kohse-Hoinghaus and J. B. Jeffries, New York: Taylor & Francis, p. 479, 2002. 20. Thompson, D. R., Bool, L. E., and Chen, J. C. “Oxygen Enhanced Combustion for NOx Control. Final Report.” DOE Report No. 889757, March 2004. 21. Fricker, N. IFRF Online Combustion Handbook, Reference number: IFRF Doc No/C76/y/1/17, Combustion File No. CF145.
Section IV
Flare Testing
© 2011 by Taylor and Francis Group, LLC
28 Large-Scale Flare Testing Charles E. Baukal, Jr., Jianhui Hong, Roger Poe, and Robert Schwartz Contents 28.1 Introduction.................................................................................................................................................................. 553 28.2 Literature Review........................................................................................................................................................ 555 28.2.1 Large-Scale Flare Test Facilities................................................................................................................... 555 28.2.2 Field Testing of Flares................................................................................................................................... 555 28.3 Large-Scale Flare Test Facility.................................................................................................................................... 555 28.3.1 System Description........................................................................................................................................ 556 28.3.2 Flow Control System..................................................................................................................................... 558 28.3.3 Data Acquisition System............................................................................................................................... 560 28.3.3.1 Thermal Radiation........................................................................................................................ 560 28.3.3.2 Noise............................................................................................................................................... 562 28.3.3.3 Flare Conditions........................................................................................................................... 563 28.4 Flare Pilot Test Facility................................................................................................................................................ 563 28.5 Sample Experimental Results.................................................................................................................................... 563 28.5.1 Hydrostatic Testing........................................................................................................................................ 563 28.5.2 Cold Flow Visualization............................................................................................................................... 563 28.5.3 Ground Flare Burner Interactions............................................................................................................... 564 28.5.4 Un-Assisted Flare........................................................................................................................................... 564 28.5.5 Air-Assisted Flares........................................................................................................................................ 565 28.5.6 Steam-Assisted Flare..................................................................................................................................... 565 28.5.7 Water-Assisted Flare...................................................................................................................................... 565 28.6 Summary....................................................................................................................................................................... 566 References................................................................................................................................................................................. 568
28.1 Introduction The ANSI/API Standard 537 (API 537) defines a flare as a “device or system used to safely dispose of relief fluids in an environmentally compliant manner through the use of combustion” [1]. Flares are used to destroy unwanted gases in a wide variety of applications from petroleum production, to refineries and petrochemical plants, to downstream product handling [2]. These unwanted gases may include off-spec product or vented streams that may be flammable, toxic, or corrosive [3]. In each case the flare system must separate liquids from gases (if necessary), ignite the gases, and provide the stable combustion necessary for destruction while minimizing smoke, thermal radiation, and noise [4]. A flare is an important interface between a plant and its surroundings. Industry, the public, and government regulators are increasingly challenged by the need for factual knowledge of flare performance. Reed (1981) notes that
flares are the most critical element in the safe operation of a process plant [5]. Chaudhuri and Diefenderfer [6] state that flare system design is very important to the economics of plant operation. Flares can be classified by geometry (elevated or ground) or by mixing enhancement (steam-assisted, air-assisted, pressure-assisted, or nonassisted) [7]. Seebold [8] lists flare height, flame stability, purge requirements [9], ignition reliability, materials of construction, and liquid removal as important elements in flare design. Schwartz and Kang [10] list reliable burning, hydraulics, liquid removal, air infiltration, flame radiation, smoke suppression, and noise/visible flame as important design considerations. The API 537 lists the following as high level safety and operating goals for a flare: • to provide safe [11], reliable, and efficient discharge and combustion of hydrocarbons with a high reaction efficiency 553
© 2011 by Taylor and Francis Group, LLC
554
• to ensure the discharged hydrocarbons burn with stable combustion over the entire defined operating range • to ensure a continuity of the flare flame under severe weather conditions • to ensure that ground level concentrations of specified compounds do not exceed environmental limits • to ensure that the back pressure does not exceed the maximum allowable • to ensure that velocity throughout the flare burner does not exceed the maximum specified • to ensure the opacity limit at the smokeless flow rate range does not exceed that allowed • to ensure the flare radiation intensity and noise levels do not exceed the maximum allowable Other important aspects of flare design include determining the plume rise to ensure the plume does not affect people inside or outside the plant [12], ensuring noise levels are acceptable to the plant personnel and surrounding community, and ensuring the flare can handle the range and composition of waste gas flows that will be sent to it. These goals are very difficult to verify in operating flare systems and can generally only be proven with an appropriate experimental facility. Failure to meet these goals can lead to significant problems [13]. Durham, McClain, and Schwartz [14] note that a flare test facility can be used to study variables that are critical to flare performance including: • flame types (momentum vs. buoyancy dominated) • flame length, shape, stability • radiation • smokeless burning rate • nozzle size and number for a given flow, composition, and pressure • nozzle location and spacing • flame stabilization • integration of high and low pressure streams for dual pressure designs • gas discharge velocity • effects of wind and low pressure zones on burner life Other critical issues include the flame stabilization location and acoustic issues that could be high or low frequency. A flare test facility can also be used to validate computational fluid dynamics (CFD; see Chapter 11)
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
predictions [15]. Noise [16] (see Chapter 8) and acoustic instability [17,18] can also be parameters of interest in flare design and operation. Choi notes that flare performance testing may be part of the process for buying flares [19]. Due to the critical safety nature of flares, reliability [20] is an important element that often can only be thoroughly examined through testing. Straitz [21] notes that poor design is one of the major factors leading to poor flare performance [21]. The API 537 provides many details and guidelines for designing flare systems. One of the major challenges for developing these guidelines is collecting operating data from full scale flare systems. In a review article, Brzustowski notes the challenges of scaling many of the important factors related to flare design and operation such as flame length, thermal radiation, and noise [22]. He writes, “full-scale testing is almost out of the question in many aspects of flaring. . . . Full-scale tests will be exceedingly rare.” He is referring to field or in-plant testing that can be very expensive and can interrupt plant production. Cassidy and Massey [23] also note the difficulty of carrying out accurate and dependable tests on in-plant flares [23]. This chapter focuses on testing large-scale flares. There are some unique challenges for collecting test data from large-scale flares, compared to other combustion systems. The first is the extremely large scale. Some of the largest flares are capable of burning over one million lb/hr (0.5 million kg/hr) of a fuel such as propane, which translates to a firing rate of 20 billion Btu/hr (6,000 MW). In many cases, the waste gas composition and flow rate may be unknown because of the nature of various waste streams being vented to the flare and the difficulty in measuring such flows [24]. The lower heating value, flammability limits, and stoichiometric ratio can all vary widely with the waste gas composition. Another problem is the intermittent operation of flares. The highest flow rates normally only occur during some type of emergency upset condition. These are very rare events that generally cannot be predicted in advance. The location of flares is often a problem as the tips may be elevated hundreds of feet above the ground. Add to that the fact that flare flames are usually not enclosed (i.e., there is no combustion chamber). This makes it very difficult, for example, to make emissions measurements [25–27] (see Chapter 6). The weather plays an important role in flare operation and is also very unpredictable. Flares must be capable of operation in both high wind and high rain conditions. High winds can significantly impact flare performance and high rains have the potential to extinguish them. The combination of elevated location, no combustion chamber, and unpredictable weather conditions makes it extremely difficult to instrument large-scale flares.
555
Large-Scale Flare Testing
28.2 Literature Review Relatively few studies have been reported in the literature on large-scale flare testing. Small-scale flare testing is covered in Chapter 29 of this book. The distinction between small- and large-scale testing is somewhat arbitrary. Here, laboratory-size tests, generally conducted indoors, are considered small. Full-scale or near fullscale size flares being tested in an industrial test facility or in the field are considered large-scale. For example, field testing of a relatively small but industrial size flare would be considered large-scale. Large-scale flare testing is categorized in this section into facilities testing and field testing. Test facilities can be used to precisely measure a wide range of parameters under somewhat controlled conditions. Fuel flow rates and compositions can be controlled, but ambient conditions such as wind speed and direction and ambient air temperature, pressure and humidity generally cannot be controlled in large-scale tests. Field testing is where a flare installed in a production site is tested. There is usually much less control over field tests, with fewer measurements being made. Due to the importance of a flare for plant safety, field tests are generally only practical in very limited circumstances. This literature search section is not intended to be exhaustive, but is designed to give the reader examples of various types of larger-scale tests that have been conducted. 28.2.1 Large-Scale Flare Test Facilities Schwartz and coworkers discuss the use of a large-scale flare pilot test facility to develop new flare pilots and monitoring systems [28,29]. Hong et al. [30] showed that large-scale testing was an integral part of designing a new type of steam-assisted flare that has many important improvements including significantly reduced steam consumption [30]. D’Amico and Nazzaro [31] describe full-scale flare testing of air-assisted flares. They studied smokeless performance and measured or calculated combustion efficiency, radiation, temperature, and noise. They also measured a wide range of other parameters such as fuel and air flows, flame length, and ambient conditions such as wind speed and direction and air temperature and humidity. Some of these experiments and others that have been conducted in a large-scale test facilities are discussed later in more detail in Section 28.5. 28.2.2 Field Testing of Flares Oenbring and Sifferman [32] describe full-scale field tests of a 16 inch (41 cm) diameter flare in an operating gas plant and in a refinery. Gas flow rates, flame lengths, and flame angles measured from various
© 2011 by Taylor and Francis Group, LLC
positions around the flares were reported. However, gas composition, wind speed and direction, noise, and radiation were not reported and do not appear to have been measured. Radiation was calculated using several different calculation procedures. McMurray [33] presented flame radiation data collected from field tests conducted over a 10 year period. Flame length, flame shape, position, and radiation were used to correlate flare radiation as a function of a wide range of parameters such as type, size, orientation, waste gas characteristics, and ambient conditions. Blackwood [34] tested the emissions from two very large flares using openpath Fourier transform infrared technology. Different tracers were introduced into each stack to distinguish between the two plumes that were located fairly close together. A range of emissions were measured. Strosher [35] conducted a five year study to investigate the emissions of diffusion flares used to burn solution gases at oil-field battery sites in Canada. The study also included laboratory testing as well. The focus of the study was determining emissions where 199 volatile and semivolatile hydrocarbons were identified. Schwartz et al. [36] discuss successful field testing of a new acoustic flare pilot monitoring system in an olefins plant in the Gulf Coast.
28.3 Large-Scale Flare Test Facility A large-scale flare test facility (see Figure 28.1) will be used as an example to illustrate the concepts discussed in this chapter [37,38]. A significant maximum flaring capacity allows many flares to be tested at full-scale (see Figure 28.2). Today’s industries demand increasingly larger smokeless capacities from flares. Although methods exist for estimating the performance of large flares, fullscale testing is still the most reliable method due to the complexity of the process. Flare design is particularly challenging because of the wide range of operating conditions including fuel composition and flow rates. It is not typically feasible to test new flare designs in the field. This is because operating conditions cannot be modified as desired in the field. One would have to wait for an actual emergency condition to occur with whatever gases happened to be in the system at the time. Discovery of a design flaw in the field during operation would be a big problem as the next time there is a planned shutdown of the flare could be in five or more years when the next turnaround is scheduled. Correcting a problem on an operating flare is often impossible without shutting down the plant, which is very expensive and undesirable.
556
Industrial Combustion Testing
Figure 28.1 World-class flare test facility at John Zink Company LLC in Tulsa, OK. (From Hong, J., Baukal, C., Schwartz, R., and Fleifil, M., Chemical Engineering Progress 102, no. 5, 35–39, 2006.)
Some types of flares do not require any assisting media because they are designed to use the available energy from the waste gas pressures. Figure 28.3 shows a comparison of a Coanda-type flare operating at a normal condition and at a condition where the gas flow is higher than the maximum design flow rate. The Coanda effect is no longer present when the maximum design flow rate is exceeded. The flame becomes bushy and the sound level increases dramatically. This is an example showing the importance of characterizing flare performance to avoid improper application. A flare test facility must be capable of supplying the higher gas pressures typically encountered in these applications. 28.3.1 System Description
Figure 28.2 An air-assisted flare undergoing testing. (From Hong, J., Baukal, C., Schwartz, R., and Fleifil, M., Chemical Engineering Progress 102, no. 5, 35–39, 2006.)
Many flares have some type of assisting media such as steam or air to increase the smokeless capacity [39]. Figure 28.2 shows an example of an air-assisted flare firing propane. The large ducts on the ground, attached to the base of the flare, are connected to two large blowers, one on each side, and provide the assist air. A flare test facility must have the capability of providing large quantities of steam or air for assisted flares.
© 2011 by Taylor and Francis Group, LLC
A block diagram of the facility is shown in Figure 28.4 where the overall flow is from left to right. Starting at the upper left of the figure, a wide range of fuels are available for testing. Standard fuels include propane, propylene, butane, natural gas, or blends of these. The first three fuels are stored as liquids and vaporized to gases for use in testing. Other fuels such as ethylene can be supplied in temporary storage vessels as needed. Inert gases such as nitrogen and carbon dioxide may also be part of the waste gases that need to be simulated. For lower flow rates, the fuels can be sent directly to a given flare. To achieve higher flow rates, a small and/or large storage vessel is/are filled with gases at an elevated pressure to increase the available hydraulic capacity. A compressor can be used to pressurize the tanks and also to circulate the gases to ensure that blends are well-mixed. Sample taps are available on the blend tanks so samples can be taken to measure the blend composition. The fuel then flows through a computer-controlled flow regulation and metering station before going to the flare. Multiple different size metering runs significantly increase the available flow range. A variety of flare testing venues are available, depending upon the flare size and application. Nitrogen is used to purge the lines between tests for safety and to
557
Large-Scale Flare Testing
(a)
(b)
Figure 28.3 Pressure-assisted flare utilizing the Coanda principle: (a) normal operation and (b) “break-away” condition. (From Hong, J., Baukal, C., Schwartz, R., and Fleifil, M., Chemical Engineering Progress 102, no. 5, 35–39, 2006.)
Lower fuel flow rates Pilot test stand
Higher fuel flow rates
Liquified fuels
Vaporizer
Fuel storage tank
Compressor
Fuel flow controls
Fuel blend mixing
Medium-scale air flare
Large-scale air or steam flare
Natural gas supply Enclosed flare
High pressure steam supply
STEAM HOLDER Steam storage tank
Steam flow controls Coanda/ground flares
Figure 28.4 Block diagram of flare test facility. (From Hong, J., Baukal, C., Schwartz, R., and Fleifil, M., Chemical Engineering Progress, 102, no. 5, 35–39, 2006.)
ensure there is no fuel contamination for succeeding tests. Nitrogen is also used to create fuel blends with reduced heat content. The bottom left side of Figure 28.4 depicts the arrangement of the steam supply system used for steam-assisted
© 2011 by Taylor and Francis Group, LLC
flare testing. A high pressure boiler generates steam that can be stored in a holding vessel to increase the available capacity for tests requiring high steam flow rates. A complete metering and control system regulates and measures the steam flow rates.
558
Industrial Combustion Testing
Figure 28.5 Fuel processing system. (Hong, J., Schwartz, R., Baukal, C., and Fleifil, M., Flare Test Facility Ready for the Challenge, presented at the John Zink International Flare Symposium, Tulsa, OK, June 11–12, 2003.)
Figure 28.5 shows the fuel processing system. The large gas storage vessel (V-114) is located in the center of the picture with the gas flow controls on the right. The LPG vaporizer (with the twin exhaust stacks) is located behind the large storage vessel in the upper part of the photo with the small gas storage vessel to its right. Gas blends are produced by sequentially supplying the appropriate gases in the desired quantities to the fuel storage vessels. The mixture is then circulated with an LPG compressor (to the left of the large storage vessel) until the blend is well-mixed. Blends are sampled and tested to verify the proper composition. Appropriate notification systems have been designed to alert appropriate personnel. The test facility is located only a few miles from Tulsa International Airport (Oklahoma). When very large flares are tested, the airport is contacted and permissions must be received from them to run the test, because of the possible impact of a large flame on airplane traffic. The fire department is also contacted in case they receive calls from concerned citizens about a large fire. Before this procedure was put into place, the fire department sometimes came to the test facility during testing of a large flare prepared to put out a fire. Personnel at the test facility are notified of an impending flare test by a series of horn blasts. 28.3.2 Flow Control System The flow controls for the new flare test facility were particularly challenging for several reasons. One was the wide range of fuels and flow rates that it would have to handle. The system is capable of flowing less than 100 lb/hr (45 kg/hr) up to more than 150,000 lb/hr (70,000 kg/hr). Another challenge was the extremely fast response time required. The system must reach the desired flow rate quickly and then maintain that rate as the pressure rapidly declines in the gas storage vessel.
© 2011 by Taylor and Francis Group, LLC
Figure 28.6 A test in progress as viewed from the control room. (From Hong, J., Baukal, C., Schwartz, R., and Fleifil, M., Chemical Engineering Progress, 102, no. 5, 35–39, 2006.)
Depending on the desired flow rate, a test may be completed in less than a minute so the flow needs to be constant during that time, despite the declining fuel tank pressure. Another challenge is synchronously recording data and video images on a single test record. Figure 28.6 shows the inside of a specially constructed control building dedicated to flare testing. The building is reinforced concrete with special impact-resistant glass to protect the operators in the unlikely event of an incident during the testing and development of flare designs. The operator seated to the right is controlling the test of an air-assisted flare firing outside the far left window. The control station and operator are shown on the right in the photo. The operator can manually or automatically control the fuel flow rate using the custom-designed hardware and software systems. Flow rates can be varied during a test as required including ramping the flows up and down. The system was constructed with a wide array of safety devices and instrumentation. There are redundant safety systems to minimize the possibility of an undesired incident. The system can be run with a single operator if desired. An extensive set of procedures was developed based on a comprehensive process hazards analysis (PHA) of the system (see Chapter 2). An environment, health, and safety (EH&S) review is conducted prior to the start of any testing program. Figure 28.7 shows a typical computer screen from the customized software control system. The left side of the screen shows the current wind speed and direction and the ambient temperature, pressure, and humidity. The right side of the screen shows pressures and flow rates (in this case zero as no test was in progress during the
559
Large-Scale Flare Testing
Figure 28.7 Typical flare test control screen. (Hong, J., Schwartz, R., Baukal, C., and Fleifil, M., Flare Test Facility Ready for the Challenge, presented at the John Zink International Flare Symposium, Tulsa, OK, June 11–12, 2003.)
© 2011 by Taylor and Francis Group, LLC
140% Desired mass flow rate
screen capture). Some of the other available functions are shown at the bottom of the screen. The software system is designed to systematically and safely guide an operator through a test. Testing a large flare tip at a very high flow rate can be challenging. For instance, to flow hundreds of thousands of lb/hr (kg/hr) of a blended fuel to a flare tip typically involves compressing fuels in certain proportions into a storage tank and then depressurizing the tank in a relatively short period of time. The reason behind this batch operation is the extreme high flow rate and the physical limitations on equipment such as vaporizer capacity for LPG (propane, propylene, butane, ethylene, etc.). During the batch operation, the pressure of the tank drops rapidly, and the temperature of the fuel blend also drops noticeably due to the Joule-Thomson cooling effect. When properly planned, short tests can yield valuable data and can be cost-effective at the same time. To maintain a constant mass flow rate to the flare tip, flow control valves must continually be opened to compensate for the decreasing tank pressure. Historically, this was done manually by skilled operators who watched a pressure gauge connected to the flare tip fuel plenum and continuously opened a valve in the fuel supply line in an attempt to keep the tip pressure constant. This process had severe shortcomings. It was a manual feedback control dependent on the operator’s skill. During a short duration test, the operator tended to perform an overshoot-
120% 100% 80% 60% 40% 20% 0%
3:11:31 PM 3:11:48 PM 3:12:06 PM 3:12:23 PM 3:12:40 PM 3:12:58 PM 3:13:15 PM
Time
Automatic control
Manual
Figure 28.8 Graph of a typical flow control history for manual and automatic operation. (From Hong, J., Baukal, C., Schwartz, R., and Fleifil, M., Chemical Engineering Progress, 102, no. 5, 35–39, 2006.)
c orrection–undershoot-correction cycle, resulting in oscillation around the target flow rate (see Figure 28.8). For the new system, a sophisticated computer control algorithm was designed and implemented to quickly reach the target flow rate and then maintain that flow rate within a very tight range. This system includes multiple orifice runs, each consisting of an orifice meter and two actuated valves, one upstream of the orifice meter and one downstream. The algorithm that drives these two valves includes a cascade of feedback controls and a
560
Industrial Combustion Testing
f eed-forward control to dramatically improve the response time. This control system is capable of controlling flow rates ranging from 101 to 105 lb/hr (kg/hr) of fuel.
28.3.3 Data Acquisition System At the John Zink Company (Tulsa, OK), various flare designs are tested comprehensively to determine performance parameters such as flame stability, flame length, smokeless capacity, purge rate required, blower horsepower, or steam requirements for assisted flares, tip longevity, radiation, and noise. All relevant data are recorded for each test in a single record. The data acquisition system consists of three computers (see Figure 28.9):
Figure 28.9 Flare test control center. (Hong, J., Schwartz, R., Baukal, C., and Fleifil, M., “Flare Test Facility Ready for the Challenge,” presented at the John Zink International Flare Symposium, Tulsa, OK, June 11–12, 2003.) (a)
1. A computer that provides the human-machine interface (HMI) for the operator. This computer collects general data such as ambient conditions, fuel temperature and flow rates, tip pressure, radiation fluxes and radiant fraction, locations of radiometers and microphones, and so forth. This computer is used to control the actual operation during a flare test. 2. A computer that records the digital videos from multiple cameras positioned to view the flare tests from various directions (see Figure 28.10). These cameras (with pan-tilt-zoom capabilities) are operated from the control room. Videos record multiple frames per second and can be played back simultaneously using a Multiplexer. 3. A computer that records the acoustic data is connected to the high-speed noise signal processor (the so-called front end from Bruel and Kjaer Company). Three different computers are used to ensure that the various software packages (HMI, video recording, and noise recording) have enough computational resources. It is important to have these three computers time synchronized. For example, in order to determine the exact pressures at which the flame becomes detached from the Coanda nozzle (see Figure 28.3), the video was visually inspected to locate the transition from an attached flame to a detached flame, and the time stamp on the video was used to identify the flare tip pressure corresponding to the “break-away” phenomenon from the general data file.
28.3.3.1 Thermal Radiation Thermal radiation is one of the most important considerations in flare designs [40] (see Chapter 30). Figure 28.11 (b)
Figure 28.10 (a) Upwind and (b) crosswind views of a flame during a flare test. (Hong, J., Schwartz, R., Baukal, C., and Fleifil, M., Flare Test Facility Ready for the Challenge, presented at the John Zink International Flare Symposium, Tulsa, OK, June 11–12, 2003.)
© 2011 by Taylor and Francis Group, LLC
Large-Scale Flare Testing
561
Figure 28.12 Radiation measurement system in use during a flare test. (From Hong, J., Baukal, C., Schwartz, R., and Fleifil, M., Chemical Engineering Progress, 102, no. 5, 35–39, 2006.) Figure 28.11 (see color insert following page 424.) Thermogram of a flare flame. (From Schwartz, R., White, J., and Bussman, W., “Flares,” in The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001.)
shows an example of a temperature profile called a thermogram of a flare flame. The stack height of a flare is often designed to be just tall enough to meet certain radiation criteria at specified locations on the ground. Excessive radiation loads can damage equipment and injure personnel [41,42]. The flare tip design can have a large impact on the radiation characteristics of the flare. Effective tip designs reduce the radiation fluxes from the flame, thus making it possible to use a shorter flare stack and reduce the overall cost of the flare system. McMurray [33] argues that the best source of flame radiation data comes from testing. Before a test is started, the coordinates of the flare and the radiometers (see Chapter 6) used to measure radiation are determined by utilizing a laser range finder to measure distances to three fixed objects with known coordinates and a technique called “triangulation.” Multiple radiometers are used to measure various radiant fluxes simultaneously. A photo of the radiation measurement system is shown in Figure 28.12. The measured radiant fluxes, through sophisticated mathematical analysis, are used to determine the coordinates of the effective “epicenter(s)” of the flame, and the radiant fraction, which is defined as the fraction of heat release from combustion that is emitted as thermal radiation [43]. Solar radiation is subtracted from the radiation measurements as appropriate. If the flare test is expected to produce significant levels of radiation, a flare radiation model is run to estimate
© 2011 by Taylor and Francis Group, LLC
the radiation at ground level that could damage equipment and injure personnel. Areas may be blocked off to prevent personnel from entering areas that may have potentially high levels of thermal radiation. In some cases, the grass around the flare may be sprayed with water to prevent it from catching on fire, particularly during the summer when the ambient temperatures are high and there has not been any rain for some time. Numerous calculation methods have been proposed for estimating the radiation from a flare. Schwartz and White [44,45] have shown that a wide range of predictions is possible depending on which model is used and what assumptions are made. Overestimating radiation results in the design of a flare stack that is taller and more costly than necessary. Underestimating radiation means the radiant flux at the ground will be higher than desired and potentially dangerous to personnel in the area during a flaring event. Many designers have relied on the historic models due to their lack of salient data from their flares operating under a variety of conditions. DeFaveri et al. [46] used small-scale experiments to develop a flare radiation model. The problem with such tests is scaling them up several orders of magnitude to actual production flare sizes. A large-scale test facility can be used to compare prediction methods with actual data. The effects of wind on flame radiation can also be studied as flare flames in low wind conditions are more vertical, while flames under high wind conditions may be nearly horizontal. Figure 28.13 shows a plot of constant radiation lines (isoflux lines) at ground level for a high pressure flare such as that shown in Figure 28.3. This plot was generated using measurements from an array of radiometers located at various distances and angles from the flare.
562
Industrial Combustion Testing
–400
Wind speed: 20 m/s Wind direction: 109°
–300 –200
Btu/hr.ft2 1. 500.00 2. 1000.00 3. 1500.00 4. 2000.00
Y axis (ft)
–100 0 100
4 3 2
200
KW/m2 1.58 3.15 4.73 6.31
1
300 400 –400
–200
0 X axis (ft)
200
400
Figure 28.13 Isoflux profiles for typical radiation measurement. (From Hong, J., Baukal, C., Schwartz, R., and Fleifil, M., Chemical Engineering Progress, 102, no. 5, 35–39, 2006.)
Noise (see Chapter 8) is an important emission from a flare that must be adequately controlled to protect personnel in the vicinity of a flare event [47]. To study the effects of noise from flares, a sophisticated sound measurement system was designed for the flare test facility. It consists of state-of-the-art Brüel & Kjær technology and is powered by the latest Pulse Platform. This measurement system (see the photo in Figure 28.14) consists of five stages. The first stage is the transducer where ultra high precision microphones (foreground in the picture) are used. The microphones are equipped with a transducer electronic data sheet (TEDS) that enables the sound system to recognize each microphone for ease of identification and communication. In this stage, the sound pressure from the noise source (flare) is picked up by different microphones that are positioned around the source at predetermined distances, heights, and directions. The use of multiple microphones facilitates resolving the directivity of the noise source (flare) and enables verifying analytical models for flare noise [48,49]. The second stage is the preconditioning circuit in which all signal conditioning such as preamplification and anti-aliasing filtration are carried out for all microphones. The third stage is the data logging system where the analog-to-digital conversion (A/D) is completed and the data for signals from all microphones are simultaneously recorded. The fourth stage is the digital signal processing (DSP) in which all signal processing for spectrum analysis such as fast Fourier transform (FFT) and constant percentage band (CPB) are accomplished simultaneously for all microphones. The fifth stage is the postprocessing and control platform in which all the sound measurement conditions are set through a powerful graphical user interface.
© 2011 by Taylor and Francis Group, LLC
Overall sound pressure level (dBA)
28.3.3.2 Noise
Figure 28.14 Sound measurement system in use during a flare test. (From Hong, J., Baukal, C., Schwartz, R., and Fleifil, M., Chemical Engineering Progress, 102, no. 5, 35–39, 2006.) 90 85 80 75 70 65 60
0
20
40
60
80 100 Time (s) Mic-2
120
140
160
180
Mic-5
Figure 28.15 Typical overall sound level as a function of time. (From Hong, J., Baukal, C., Schwartz, R., and Fleifil, M., Chemical Engineering Progress, 102, no. 5, 35–39, 2006.)
The duration of the measurements, the microphones to be used, the type of data recorded, and the type of spectrum analyzers are among the numerous measurement conditions that can be varied before and after taking the sound measurements. An example of the sound pressure record from two microphones at different locations, for a typical flare test, is shown in Figure 28.15. The spike at 0 seconds and again at about 10 seconds are from a safety horn alerting everyone in the area of an impending flare test. In this particular example, there was a rapid rise in the sound level at the start of the test followed by a steady decline as the fuel flow rate was reduced as dictated by the given test plan. A one-third octave band analysis for the sound signal from microphone No. 5 during the time period from
563
Sound pressure level (dB)
Large-Scale Flare Testing
90 80 70 60 50
20
40
80
160
315 630 1.25k 2.5k Frequency (Hz)
5k
10k 20k
Figure 28.16 Average sound profile for a given time window. (Hong, J., Schwartz, R., Baukal, C., and Fleifil, M., Flare Test Facility Ready for the Challenge, presented at the John Zink International Flare Symposium, Tulsa, OK, June 11–12, 2003.)
20 to 40 seconds is shown in Figure 28.16. Notice that there are three peaks in the spectrum analysis. These peaks correspond to combustion roar and to the noise generated from the assist air and fuel jets. 28.3.3.3 Flare Conditions A number of other data points may be collected during a given test, depending on the type of flare and the intended application. These include, for example, metal temperatures in and around the exit of the flare tip to observe potential overheating problems. Flares are often tested with noncombustible gases such as air to study flow patterns. One example is the use of smoke generators in an air-assisted flare. Another uses handheld Pitot probes (see Chapter 4) to measure velocities. For the enclosed ground flare, a video camera is used to visually observe the combustion process. A gas analysis system can be used to measure the exhaust emissions. Flare performance can be significantly affected by ambient wind speeds. Wind can cause a flame to bend and change the radiation flux on the ground. In addition, high wind speeds can significantly impact flare performance. Data that are continuously recorded include wind speed and direction, and ambient air temperature. Data, such as relative humidity and ambient pressure, which are essentially constant during a test are recorded only once for that test.
28.4 Flare Pilot Test Facility A new test stand (shown in Figure 28.17) was constructed specifically for testing flare pilots [29]. These are critical safety devices that must remain lit under severe weather conditions such as heavy wind and rain. The test stand is
© 2011 by Taylor and Francis Group, LLC
Figure 28.17 Flare pilot test stand. (Hong, J., Schwartz, R., Baukal, C., and Fleifil, M., Flare Test Facility Ready for the Challenge, presented at the John Zink International Flare Symposium, Tulsa, OK, June 11–12, 2003.)
capable of simulating wind speeds over 160 mph (260 km/ hr) blowing against both the pilot and the pilot mixer and simulating rain at over 30 in./hr (76 cm/hr). This test rig was used to develop the WindProofTM [50] and the InstaFireTM pilot (patents pending), which can remain lit when exposed to those extreme weather conditions [29].
28.5 Sample Experimental Results This section gives a sample of some large-scale flare tests. The first few examples are where the flares were not firing. The last few are examples of flares firing. They demonstrate the wide range of test types that may be done to determine the performance of flares. 28.5.1 Hydrostatic Testing In some cases hydrostatic testing is done to evaluate the maximum pressures that can be introduced to a flare tip. Figure 28.18 shows a senior engineer at John Zink carefully taking measurements of a single test element. Pressure was hydrostatically increased until the element began to fail. This scenario was repeated several times to ensure the continuity of the data taken. This element was tested to pressures of more than 20 times the typical working pressure of the unit. 28.5.2 Cold Flow Visualization In some cases, cold flow visualization can be used to inexpensively display patterns in a flare. The researcher
564
Figure 28.18 Hydrostatic flare tip test. (Courtesy of John Zink Co. LLC.)
Industrial Combustion Testing
Figure 28.19 Typical ground flare array. (From Schwartz, R., White, J., and Bussman, W., The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001.)
may be looking for flow uniformity, for any recirculation zones, for mixing lengths, entrainment, or other related fluid dynamics. For example, fluid flow uniformity may be important to ensure the resulting flare flame is uniform and not leaning. Computational fluid dynamics [51,52] (see Chapter 11) can used as well as cold flow visualization (see Chapter 10), where both can be important tools to ensure there are no flow inversions in a given tip design. Smoke can be injected into the air stream of an air-assisted flare to help validate flow patterns predicted by CFD. Cold flow visualization can be a powerful tool to validate CFD predications and to see flow patterns relatively quickly and inexpensively. 28.5.3 Ground Flare Burner Interactions Ground flares (see Figure 28.19), sometimes referred to as multipoint flares, may be used for a variety of reasons. If a very large turndown is required, rows of burners can be staged on as required. If elevated flare flames are not desirable, this type of flare design is an option. Flare radiation can also be shielded with some type of fence. Design tools have been developed to predict the performance of these ground flares, for example, to ensure proper confinement and accurately predict radiation. Single burner testing will suffice to determine turndown capabilities, baseline flame shape, and a range of smokeless performance. Cross-lighting capabilities, total system flame height, and combined radiation confirmation can only be accomplished by more expensive multiburner testing. Figure 28.20 shows a row of ground flare burners being tested to determine flame height. This is compared to previous measurements of single and dual burners. Single and dual burner tests may underestimate the flame height compared to a row of burners. This is important when designing the height of the flare fence.
© 2011 by Taylor and Francis Group, LLC
Figure 28.20 (see color insert following page 424.) Photo of a row of ground flares, with flame heights compared to those previously measured for single and dual burner tests. (Courtesy of John Zink Co. LLC.)
Figure 28.21 shows another example of a test for the type of ground flare burners commonly used in the array shown in Figure 28.19. These are full-scale burners being tested to determine flame heights and centerlineto-centerline distances for cross-lighting. A range of different molecular weight fuels are tested to determine the performance over a range of conditions that might be encountered in the field. 28.5.4 Un-Assisted Flare Another type of flare that uses the Coanda principle (see Figure 28.3) is referred to as an Indair® flare. The Coanda effect occurs within a specific pressure range [53]. Full-scale testing can be used to determine the minimum waste gas flow to initiate the Coanda effect. These flares are spring-loaded, so the minimum pressure is needed to determine the proper spring opening
565
Large-Scale Flare Testing
28.5.5 Air-Assisted Flares
Figure 28.21 Multiple ground flare burners being tested to determine flame heights and cross-lighting distances. (Courtesy of John Zink Co. LLC.)
Air-assisted flares (see Figure 28.2) use one or more blowers to supply combustion air to the flare flame, in addition to the ambient air around the flare that is entrained into the flame. The blowers increase the smokeless capacity of the flare. Figure 28.23 shows a series of photos of an air-assisted flare test where the blower goes from off to full on. The flame goes from heavily smoking to smokeless with the addition of the blower. One safety test that is sometimes done is to simulate a power failure for an air-assisted flare. The blower on an air-assisted flare is designed to increase the smokeless capacity of the flare. A blower failure test is designed to demonstrate the air flare will still safely handle the waste gases going to the flare with the blower off, although smoke will likely be produced with the blower off compared to when the blower is in operation. In particular, the blower failure test is to make sure fuel gas exiting the flare tip does not flow backward in the air duct toward the inlet of the air blower, which could lead to dangerous consequences. This hazardous reverse flow pattern sometimes arises due to the mechanical design of the flare tip. Figure 28.24 shows a high flow rate of propane going to an air-assisted flare with the air blower off. While smoke was generated, the test demonstrated the flare safely operated when the blower was off. 28.5.6 Steam-Assisted Flare
Figure 28.22 Multiple Indair® flare test to determine minimum operating pressure range and tip spacing for cross-lighting. (Courtesy of John Zink Co. LLC.)
pressure. This keeps the tip velocity sufficiently high to negate the smoke typically associated with excessively low turndown pressures. This type of burner is routinely performance tested at application pressures to ensure smoke or sooting is not an issue during normal operation. Figure 28.22 shows multiple Indair® burners being tested to determine centerline spacing for proper cross-lighting. Cross-light timing may also be important as the longer it takes to light flare burners, the more unburned waste gas that could potentially be emitted into the atmosphere.
© 2011 by Taylor and Francis Group, LLC
Steam-assisted flares are similar to air-assisted flares in that additional air is entrained into the flare to increase the smokeless capacity [54]. In steam-assisted flares, steam is used to entrain air into the flare, in addition to the ambient air surrounding the flare that is entrained into the flame. Figure 28.25 shows a series of photos of a steam-assisted flare as the steam flow goes from zero to the required flow rate for smokeless operation. Oversteaming of steam-assisted flares can impact both the stability and the effectiveness of a flare. Figure 28.26 shows the testing of a steam-assisted flare to determine the range of stability when significantly oversteamed. Oversteaming sometimes occurs in a plant when the steam control is manual, so the flare should be stable in case this condition does occur. 28.5.7 Water-Assisted Flare Figure 28.27 shows a flare on an offshore oil production platform. A particular challenge for these flares is the close proximity to personnel, where it is generally not feasible to elevate the flare high enough to minimize thermal radiation loads on the platform. This is due to the structural issues associated with elevating a flare high above a platform that itself is elevated above the
566
Industrial Combustion Testing
(a)
(b)
(c)
(d)
Figure 28.23 Air-assisted flare test: (a) blower off, (b) blower starting up, (c) blower speed increasing, and (d) blower at full speed and flare smokeless. (From Schwartz, R., White, J., and Bussman, W., The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001.)
floor of the ocean. One technique that has proven to significantly reduce both radiation and noise levels from flares is water-injection [55]. Large volumes of water are not always readily available in a land-based plant, but salt water is very plentiful around offshore platforms. Figure 28.28 shows full-scale testing of water-injection in a flare designed for use on an offshore platform [56]. Figure 28.29 shows the results of these tests and how dramatically water injection can reduce radiation from a flare. Figure 28.30 shows how this water injection reduced noise by 13 dBA compared to the same flare with no water injection. Full-scale testing was used to determine the optimum water injector design and injection flow rates.
28.6 Summary
Figure 28.24 Air flare blower failure test. (Courtesy of John Zink Co. LLC.)
© 2011 by Taylor and Francis Group, LLC
A flare test facility can be used for many purposes. One is to develop new designs with enhanced performance. Another is to characterize the performance of existing technologies, particularly for new conditions that may not have been previously encountered. This includes
567
Large-Scale Flare Testing
(a)
(b)
(c)
Figure 28.25 Effectiveness of steam in smoke suppression: (a) no steam, (b) starting steam, and (c) smokeless. (From Schwartz, R., White, J., and Bussman, W., The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001.)
Figure 28.26 Steam-assisted flare test. (Courtesy of John Zink Co. LLC.)
Figure 28.28 Testing a water-assisted flare. (Courtesy of John Zink Co. LLC.)
Radiation level, btu/hr/sq ft
2000
Figure 28.27 Radiation from an offshore flare. (From Baukal, C. E., Industrial Combustion Pollution and Control, Boca Raton, FL: CRC Press, 2004.)
© 2011 by Taylor and Francis Group, LLC
1800 1600 1400 1200
High molecular weight gas
1000 800 600 400 200 0
Low molecular weight gas
0
20 40 60 80 100 Water injected to achieve optimum radiation reduction, %
Figure 28.29 Radiation reduction by water injection. (From Leary, K., Knott, D., and Thompson, R., Oil & Gas Journal, 100, no. 18, 76–83, 2002. With permission.)
568
Industrial Combustion Testing
110 105
Noise level, db
100 95
90 85 80 75 70
10
100
1.000 Frequency, Hz
10.000
100.000
No water injection, 116 db, 114 dba Optimum water injection, 113 db, 101 dba
Figure 28.30 Noise reduction by water injection. (From Leary, K., Knott, D., and Thompson, R., Oil & Gas Journal, 100, no. 18, 76–83, 2002. With permission.)
extending them to new applications such as increasing the operating range of an established design. A wellequipped flare test facility can also be used to validate both proprietary equipment-specific empirical prediction models as well as computational fluid dynamic mathematical models. In the past, flares have been designed using narrowly defined empirical and semianalytical models that sometimes produce disparate results. This was due primarily to the limited amount of credible and comprehensive experimental data from industrial-scale flares tested under precisely measured conditions. The large-scale flare test facility discussed here was designed and built to correct that problem and provide data for greatly improving flare analysis techniques leading to new and improved designs. While field testing may be done under certain limited circumstances, this is not generally practical and typically only provides limited data under a narrow range of operating conditions.
References
1. API Standard 537. Flare Details for General Refinery and Petrochemical Service. 2nd ed. Washington, DC: American Petroleum Institute, December 2008. 2. Schwartz, R., White, J., and Bussman, W. “Flares.” In The John Zink Combustion Handbook, edited by C. E. Baukal. Boca Raton, FL: CRC Press, pp. 588–634, 2001. 3. Banerjee, K., Cheremisinoff, N. P., and Cheremisinoff, P. N. Flare Gas Systems Pocket Handbook. Houston, TX: Gulf Publishing Co., 1985. 4. Schwartz, R. E., and Kang, S. G. “Effective Design of Emergency Flaring Systems.” Hydrocarbon Engineering 3, no. 2 (1998): 57–62.
© 2011 by Taylor and Francis Group, LLC
5. Reed, R. D. Furnace Operations. 3rd ed. Houston, TX: Gulf Publishing, 1981. 6. Chaudhuri, M., and Diefenderfer, J. J. “Achieving Smokeless Flaring: Air or Steam Assist?” Chemical Engineering Progress 91, no. 6 (1995): 40–43. 7. Akeredolu, F. A., and Sonibare, J. A. “A Review of the Usefulness of Gas Flares in Air Pollution Control.” Management of Environmental Quality 15, no. 6 (2004): 574–83. 8. Seebold, J. G. “Practical Flare Design.” Chemical Engineering 91, no. 25 (1984): 69–72. 9. Husa, H. “How to Compute Safe Purge Rates.” Hydrocarbon Processing 43, no. 5 (1964): 179–82. 10. Schwartz, R. E., and Kang, S. G. “Effective Design of Emergency Flaring Systems.” Hydrocarbon Engineering 3, no. 2 (1998): 57–62. 11. Shore, D. “Making the Flare Safe.” Journal of Loss Prevention in the Process Industries 9, no. 6 (1996): 363–81. 12. Fumarola, G., DeFaveri, D. M., Palazzi, E., and Ferraiolo, G. “Determine Plume Rise for Elevated Flares.” Hydrocarbon Processing 61, no. 1 (1982): 165–66. 13. Straitz, J. K. “Flare Case Histories Demonstrating Problems and Solutions.” Process Safety Progress 25, no. 4 (2006): 311–16. 14. Durham, E. B., McClain, D. D., and Schwartz, R. E. “Profitable Flare Selection.” Hydrocarbon Processing 4, no. 5 (1999): 88–90. 15. Castiñeira, D., and Edgar, T. F. “CFD for Simulation of Steam-Assisted and Air-Assisted Flare Combustion Systems.” Energy & Fuels 20 (2006): 1044–56. 16. Banks, D. “Combustion Pulsation and Noise.” Process Heating 14, no. 8 (2007): 30–34. 17. Nathan, G. J., Mullinger, P. J., Bridger, D., and Martin, B. “Investigation of a Combustion Driven Oscillation in a Refinery Flare. Part A: Full Scale Assessment.” Experimental Thermal and Fluid Science 30, no. 4 (2006): 285–95. 18. Riese, M., Kelso, R. M., Nathan, G. J., and Mullinger, P. J. “Investigation of a Combustion Driven Oscillation in a Refinery Flare. Part B: Visualisation of a Periodic Flow Instability in a Bifurcating Duct Following a Contraction.” Experimental Thermal and Fluid Science 31, no. 8 (2007): 1091–1101. 19. Choi, D. W. “How to Buy Flares.” Hydrocarbon Processing 79, no. 12 (2000): 76–78. 20. Straitz, J. F. “Improve Flare Design.” Hydrocarbon Processing 73, no. 10 (1994): 61–66. 21. Straitz, J. F. “Improve Flare Safety to Meet ISO-9000 Standards.” Hydrocarbon Processing 75, no. 6 (1996): 109–14. 22. Brzustowski, T. A. “Flaring in the Energy Industry.” Progress in Energy and Combustion Science 2, no. 3 (1976): 129–41. 23. Cassidy, J., and Massey, L. “Flare System Evolution.” Hydrocarbon Processing 8, no. 9 (2003): 76–78. 24. Gilmer, L., Caico, C. A., Sherrick, J. J., Mueller, G. R., and Loos, K. R. “Flare Waste Gas Flow and Composition Measurement Methodologies Evaluation Document.” Report prepared for the Texas Commission on Environmental Quality, 2006.
Large-Scale Flare Testing
25. Davis, B. C. “Flare Efficiency Studies.” Plant/Operations Progress 2, no. 3 (1983): 191–98. 26. Pohl, J. H., Lee, J., Payne, R., and Tichenor, B. A. “Combustion Efficiency of Flares.” Combustion Science and Technology 50 (1986): 217–31. 27. Baukal, C. E. Industrial Combustion Pollution and Control. Boca Raton, FL: CRC Press, 2004. 28. Schwartz, R. S. “New Products Enhance Flare Stability.” World Refining 12, no. 3 (2001): 20–22. 29. Schwartz, R. E., Hong, J., and Smith, J. D. “The Flare Pilot.” Hydrocarbon Engineering 7, no. 2 (2002): 65–68. 30. Hong, J., Baukal, C., Bastianen, M., Bellovich, J., and Leary, K. “New Steam Assisted Flare Technology.” Hydrocarbon Engineering 12, no. 7 (2007): 63–68. 31. D’Amico, R., and Nazzaro, M. I. “No Smoking.” Hydrocarbon Engineering 13, no. 7 (2008): 47–51. 32. Oenbring, P. R., and Sifferman, T. R. “Flare design … Are Current Methods Too Conservative?” Hydrocarbon Processing 59, no. 5 (1980): 124–29. 33. McMurray, R. “Flare Radiation Estimated.” Hydrocarbon Processing 64, no. 11 (1982): 175–81. 34. Blackwood, T. R. “An Evaluation of Flare Combustion Efficiency Using Open-Path Fourier Transform Infrared Technology.” Journal of Air & Waste Management Association 50, no. 10 (2000): 1714–22. 35. Strosher, M. T. “Characterization of Emissions from Diffusion Flare Systems.” Journal of Air & Waste Management Association 50, no. 10 (2000): 1723–33. 36. Schwartz, R. E., Kodesh, Z., Balcar, M., and Bergeron, B. “Improve Flaring Operations.” Hydrocarbon Processing 81, no. 1 (2002): 59–62. 37. Hong, J., Schwartz, R., Baukal, C., and Fleifil, M. “Flare Test Facility Ready for the Challenge.” Presented at the John Zink International Flare Symposium, Tulsa, OK, June 11–12, 2003. 38. Hong, J., Baukal, C., Schwartz, R., and Fleifil, M. “Industrial-Scale Flare Testing.” Chemical Engineering Progress 102, no. 5 (2006): 35–39. 39. Leite, O. C. “Smokeless, Efficient, Nontoxic Flaring.” Hydrocarbon Processing 70, no. 3 (1991): 77–80. 40. API RP-521. Guide for Pressure-Relieving and Depressuring Systems. 4th ed. Washington, DC: American Petroleum Institute, March 1997. 41. Buettner, K. “Heat Transfer and Safe Exposure Time for Man in Extreme Thermal Environment.” ASME paper 57-SA-20. Proceedings of the Heat Transfer Division ASME Semi-Annual Meeting, San Francisco, CA, June 9–13, 1957.
© 2011 by Taylor and Francis Group, LLC
569
42. Fumarola, G., DeFaveri, D. M., Pastorino, R., and Ferraiola, G. “Determining Safety Zones for Exposure to Flare Radiation.” Institution of Chemical Engineers Symposium, no. 83 (1983): G23–G30. 43. Hong, J., White, J., and Baukal, C. “Accurately Predicting Radiation From Flare Stacks.” Hydrocarbon Processing 85, no. 6 (2006): 79–81. 44. Schwartz, R. E., and White, J. W. “Flare Radiation Prediction: A Critical Review.” Presented at 30th Annual Loss Prevention Symposium of AIChE, Paper 12a, February 29, 1996. 45. Schwartz, R. E., and White, J. W. “Predict Radiation from Flares.” Chemical Engineering Progress 93, no. 7 (1997): 42–49. 46. DeFaveri, D. M., Fumarola, G., Zonato, C., and Ferraiolo, G. “Estimate Flare Radiation Intensity.” Hydrocarbon Processing 64, no. 5 (1985): 89–91. 47. Seebold, J. G. “Flare Noise: Causes and Cures.” Hydrocarbon Processing 51, no. 10 (1972): 143–47. 48. Bussman, W. R., and Knott, D. “Unique Concept for Noise and Radiation Reduction in High-Pressure Flaring.” Offshore Technology Conference, Houston, TX, May 1–4, 2000. 49. Seebold, J. G., and Hersh, A. S. “Control Flare Steam Noise.” Hydrocarbon Processing 51 (1971): 140. 50. Hong, J., Smith, J. D., Poe, R., and Schwartz, R. E. “UltraStable Flare Pilot and Methods.” U.S. Patent 6,702,572, 2004. 51. Baukal, C. E., Gershtein, V. Y., and Li, X. Y. Computational Fluid Dynamics in Industrial Combustion. Boca Raton, FL: CRC Press, 2001. 52. Castiñeira, D., and Edgar, T. F. “Computational Fluid Dynamics for Simulation of Wind Tunnel Experiments on Flare Combustion Systems.” Energy & Fuels 22, no. 3 (2008): 1698–1706. 53. Gregory-Smith, D. G., and Gilchrist, A. R. “The Compressible Coanda Wall Jet: An Experimental Study of Jet Structure and Breakaway.” International Journal of Heat and Fluid Flow 8, no. 2 (1987): 156–64. 54. Selle, G. K. “Steam-Assisted Flare Eliminates Environmental Concerns of Smoke and Noise.” Hydrocarbon Processing 75, no. 11 (1996): 117–18. 55. Leary, K., Knott, D., and Thompson, R. “Water-Injected Flare Tips Reduce Radiated Heat, Noise.” Oil & Gas Journal 100, no. 18 (2002): 76–83. 56. Bussman, W. R., and Knott, D. “Unique Concept for Noise and Radiation Reduction in High-Pressure Flaring.” Proceedings of the Annual Offshore Technology Conference, Vol. 1, 721–32, Houston, TX, May 1–4, 2000.
29 Flare Experimental Modeling Chendhil Periasamy and Subramanyam R. Gollahalli Contents 29.1 Introduction.................................................................................................................................................................. 572 29.2 Parameters Governing Flare Characteristics........................................................................................................... 572 29.3 Experimental Facilities and Instrumentation.......................................................................................................... 572 29.3.1 Wind Tunnel................................................................................................................................................... 572 29.3.2 Model Stack.................................................................................................................................................... 573 29.3.3 Fuel Gas Supply System................................................................................................................................ 573 29.3.4 Instrumentation..............................................................................................................................................574 29.3.4.1 Flame Imaging...............................................................................................................................574 29.3.4.2 Flame Radiation.............................................................................................................................574 29.3.4.3 Combustion Gas Analysis............................................................................................................574 29.3.4.4 Temperature Measurement..........................................................................................................574 29.3.4.5 Velocity Field..................................................................................................................................574 29.3.4.6 Advanced Laser Diagnostics.......................................................................................................574 29.3.5 Data Acquisition System............................................................................................................................... 576 29.4 Regimes of Operation.................................................................................................................................................. 576 29.4.1 Lifted Versus Wake-Stabilized Flames....................................................................................................... 576 29.4.2 Wake-Stabilized Flames................................................................................................................................ 577 29.4.3 Effects of Buoyancy on Flame Configuration............................................................................................ 578 29.5 Global Characteristics................................................................................................................................................. 578 29.5.1 Flame Configuration and Appearance....................................................................................................... 578 29.5.1.1 Low Jet-to-Crossflow Momentum Flux Ratio Flames............................................................. 578 29.5.1.2 High Jet-to-Crossflow Momentum Flux Ratio Flames............................................................ 579 29.5.2 Flame Geometry............................................................................................................................................ 579 29.5.2.1 Flame Size...................................................................................................................................... 580 29.5.2.2 Flame Trajectory........................................................................................................................... 581 29.5.3 Stability Characteristics................................................................................................................................ 581 29.5.4 Flaring Efficiency........................................................................................................................................... 582 29.5.5 Radiation Emission........................................................................................................................................ 583 29.6 Flame Structure Measurements................................................................................................................................. 584 29.6.1 Temperature Field.......................................................................................................................................... 584 29.6.2 Species Concentration Field......................................................................................................................... 585 29.6.3 Velocity Field.................................................................................................................................................. 587 29.7 Miscellaneous Topics................................................................................................................................................... 589 29.7.1 Novel Shapes for Flare Tip........................................................................................................................... 589 29.7.2 Effects of Diluents.......................................................................................................................................... 590 29.7.3 Implications for Scaling of Laboratory Data.............................................................................................. 590 29.8 Concluding Remarks................................................................................................................................................... 591 29.9 Nomenclature............................................................................................................................................................... 591 References................................................................................................................................................................................. 591
571 © 2011 by Taylor and Francis Group, LLC
572
Industrial Combustion Testing
29.1 Introduction Several parameters such as the size and shape of flame tip, input gas composition, flow rate, and wind speed influence the characteristics and behavior of a flare. Prediction of flare behavior in the field is difficult since the speed, direction, and turbulence level of the wind vary widely. In such situations, experimental modeling becomes a useful tool. The purposes of experimental modeling are to understand flare characteristics, delineate the influence of the above parameters, and develop design and operational guidelines. There are two major categories of flares: (i) ground and (ii) elevated flares. Experimental modeling of each type of flare requires specific experimental facility and modeling strategy. In this chapter, we focus on elevated flares and refer to them as flares for simplicity. For modeling study, flare is defined as a large, turbulent diffusion flame in openair crossflow (TDFCF) and the experimental modeling involves appropriately scaled down burners and a wind tunnel to simulate the ambient conditions in the field. Such modeled flares in laboratory settings enjoy the benefits of controlled air environment, fuel flexibility, and burner size and shape variation. All these parameters can be independently varied to study the performance of flares. This chapter deals mostly with model studies conducted at the University of Oklahoma Combustion and Flame Dynamics Laboratory and also presents a critical review of relevant studies available in literature.
29.2 Parameters Governing Flare Characteristics Figure 29.1 shows a schematic of a turbulent diffusion flame in crossflow. Its shape is typically characterized by horizontal length, vertical height, centerline trajectory, and maximum flame width. Horizontal length of a Horizontal projected length
Trajectory U∞ y
Uj
Vertical height
x Model stack Figure 29.1 Schematic of a vertical turbulent diffusion flame in crossflow.
© 2011 by Taylor and Francis Group, LLC
crossflow flame is defined as the horizontal projection of flame, measured from the burner tip to the end of the contiguous flame in the direction of crossflow. Vertical height is defined as the vertical projection of the flame from the burner tip to the end of the contiguous flame in the direction perpendicular to crossflow. Trajectory is usually denoted by the path of the flame centerline. The curvilinear length of the flame centerline is called the trajectory length. The flame width or the thickness is measured perpendicular to the centerline. Besides the trajectory, shape, and size of the flame, thermochemical characteristics such as radiation, smoke liberation, flame stability, combustion efficiency, and pollutant emissions are of interest to the flare designers and operators. In modeled flares, the flame characteristics are primarily influenced by fuel jet to crossflow momentum flux ratio (R), fuel molecular structure, fuel type, fuel mass flow rate and dilution, burner tip size and shape, burner orientation to crossflow, and scaling ratio between the stack and model.
29.3 Experimental Facilities and Instrumentation 29.3.1 Wind Tunnel Low-speed, subsonic-flow, open-loop wind tunnels are usually employed in model flare studies to simulate field ambient conditions. Depending on the desired simulation conditions, the tunnel test section size is chosen. All such tunnels aim at creating a uniform wind flow pattern with a relatively low turbulence intensity (the ratio between RMS of turbulent velocity fluctuations to average velocity at a given location is usually less than 1%) and providing an opportunity to vary the wind speed inside the test section. El Behery, Mohamad, and Kamal [1], for instance, used a maximum wind speed of 10.3 m/s. Alternatively, an open-jet wind tunnel with no side walls or roof has also been used [2]. In their study, the model stack was positioned protruding vertically from the test section floor. Bandaru and Turns [3] made use of the potential core of a large air jet as a simple arrangement to produce a uniform crossflow stream. Figure 29.2 shows the schematic diagram of a lowspeed, suction-type wind tunnel used to simulate the crossflow field at the University of Oklahoma. Overall length of the wind tunnel was 1130 cm (38’), and the test section was 250 cm (8.2’) long, 58 cm (1.9’) wide, and 53 cm (1.7’) high. The inlet section was designed with a contraction ratio of 10. Both side walls of the test section were constructed with 13-mm thick Pyrex glass panels for optical access. The top and bottom panels
573
Flare Experimental Modeling
1130 cm
Tunnel tubing
Screens
Contraction section Flame
530 cm
Exhaust Test section
Fan
53 cm Burner
Pyrex windows 250 cm
Diffuser
130 cm
Figure 29.2 Schematic of the wind tunnel used for model studies.
of the test section walls were made of aluminum sheet metal plates with provisions to introduce probes and burner. The burner protruded into the test section from the bottom panel and was placed perpendicular to the wind stream. The location of the burner tip could be adjusted and was usually set outside the wall boundary layer. The fuel jet was ignited using an electrical lighter from the ceiling of the test section. The wind tunnel was capable of providing a flow velocity ranging from 1 to 6.5 m/s (3.3 to 21.3 ft/s) in the test section. The flow was relatively uniform in the test section except near the wall region due to the formation of the boundary layer. The turbulence intensity in the tunnel test section was about 0.5%. Kalghatgi [4,5] used a large wind tunnel with a test section of 2.65-m-high and 2-m-wide. Bourguignon, Johnson, and Kostiuk [6] and Poudenx et al. [7] also employed large open-loop and closed-loop wind tunnels with 6.1-m high and 3.05-m wide and 1.22-m high and 2.44-m wide test sections, respectively. Both the tunnels were capable of delivering wind speeds up to 35 m/s. On the other hand, Han and Mungal [8,9] employed smaller, induced-draft wind tunnels with a rectangular test section of 30 × 30 cm and a height of 80 cm. Such smaller-size wind tunnels allow the use of advanced laser diagnostic systems for flow field visualization. 29.3.2 Model Stack Flare tips are designed to handle large velocities encountered during emergency situations. Brzustowski [10] describes the factors that influence the choice of tip diameter. Of all the considerations, the turndown ratio is the most important factor. Usually, the stack discharge velocity is over 100 m/s. The combination of discharge velocity and the local wind speed determines the flow pattern near the tip of the stack, which in turn determines
© 2011 by Taylor and Francis Group, LLC
the tip diameter. In TDFCF literature, most studies have employed a circular cross-sectional shape for the burners with a diameter of about 5 mm. However, Kalghatgi [5] used burner tubes of a diameter up to 22 mm. A few studies have also been conducted using elliptic burners [11,12]. Flames from such novel burners are discussed in Section 29.7.1. 29.3.3 Fuel Gas Supply System Flare gas composition depends on the condition at which the flaring is done (i.e., production flaring, process flaring, and emergency flaring [10]. Although it is difficult to reproduce field gas composition on laboratory scale flames, a wide range of pure and composite fuels are used to simulate flare gas. Typically, pure or composite forms of fuel such as natural gas, propane, propylene, ethylene, hydrogen, and commercial butane are employed to model flare gases. In a typical setup at the University of Oklahoma [13–15], the fuel gas supply train consisted of propane, propylene, ethylene and hydrogen gas cylinders, pressure regulators, calibrated rotameters, manifolds, and tubing. There were three different gas supply trains: one for fuels, and one each for hydrogen and nitrogen. Hydrogen was used to attach the flames to the burner and nitrogen was used to suppress smoke liberation. Fuel gases were stored in industrial-size cylinders at a pressure of 17 MPa (2450 psig) and delivered to the system at 135 kPa (5 psig). Each cylinder was mounted with a two-stage pressure regulator. A set of ballrotameters, calibrated with a wet test meter, was used to measure the flow rate of reactants. Before entering the burner, both the fuel and nitrogen lines were combined using a Y connector. Downstream of the Y connection, a 3 m long ( > 150 hydraulic diameter) rubber hose was used to deliver the fuel to the burner.
574
Hydrogen was admitted to the burner as a coflowing stream. A K-type thermocouple was installed in the fuel-nitrogen line just before the burner to monitor the fuel inlet temperature. 29.3.4 Instrumentation 29.3.4.1 Flame Imaging There are several techniques available to image the flames in crossflow. A simple and direct method is the use of 35-mm analog or digital cameras (SLR or non-SLR type). The cameras should have provisions to adjust the shutter speed and exposure time in order to capture the highly fluctuating flames, especially at high jet-to-crossflow momentum flux ratios. Bandaru and Turns [3] employed 35-mm analog cameras with DIN 100 color slide film and an exposure time of eight seconds. Goh, Kusadomi, and Gollahalli [15] used both analog and digital cameras with three different shutter speeds (1, 1/125, 1/250 s) and 1-s exposure time. Huang and Chang [16] employed color photography with short (1/6000 s) and long (2 s) exposure time. Color video camera is also a useful tool to take movie-type images [5,17,18]. Shadowgraphs [2], Schlieren optical techniques, and reactive Mie scattering [16,19] are used to capture the flow pattern. Laser-based planar imaging techniques have also been developed for flow field visualization in a burning environment and are discussed in Section 29.3.4.6. 29.3.4.2 Flame Radiation Radiant fractions can be computed from radiant heat flux measurements. A wide-angle view (150°) radiometer is used for flame radiation measurements in our laboratory. The radiometer is mounted on a tripod at approximately the mid-height of the flame and far enough to satisfy the inverse-square law. For additional details on the mounting of the radiometer, the reader is referred to Bandaru and Turns [3] and Goh and Gollahalli [20]. 29.3.4.3 Combustion Gas Analysis For measuring the concentration profiles of CO, CO2, NO, NOx, O2, and UBHC in a TDFCF, a sample gas stream is drawn from the flame at a rate closer to isokinetic sampling. Sampling probes can be of uncooled [21] or watercooled [3] type. The sample gas is treated to remove moisture and particulate matter and is passed through a series of analyzers and corresponding measurements are taken. For instance, Gollahalli and Nanjundappa [21] used a bent, uncooled quartz probe with an orifice internal diameter of 1 mm to draw the sample gas. The
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
sampling gas flow rate was monitored by a rotameter. A nondispersive infrared analyzer was used for CO and CO2 concentration measurements. Oxygen concentration was obtained by using a Polarographic sensor. The NO/NOx concentration was measured with a chemiluminescence analyzer. Huang and Yang [22] and Bandaru and Turns [3] also used a similar instrumentation setup to measure the concentration of combustion species. 29.3.4.4 Temperature Measurement The temperature field in a TDFCF has been obtained by using thermocouple measurements. Thermocouples of type R and C with a small bead diameter (∼200 µm) have been used to measure the thermal field in the flame [3,20,22]. Note that the temperature measurements obtained with a thermocouple need to be corrected for bead radiation and wire conduction losses [23]. 29.3.4.5 Velocity Field Intrusive measurement techniques such as a Pitot static tube and hot-wire anemometer [24–26], and nonintrusive techniques such as laser Doppler velocimeter and particle image velocimetry (PIV) have been used to study the flow field. Goh, Kusadomi, and Gollahalli [13–15] mapped the velocity field in the flame using a Pitot static tube with a pressure transducer (Barocel). Details of the techniques and selection guidelines are presented in books on experimental aspects of fluid mechanics. Interested readers are referred to Holman [27], Goldstein [28], and Miller [29], to name a few. 29.3.4.6 Advanced Laser Diagnostics 29.3.4.6.1 Laser Doppler Velocimetry Laser Doppler Velocimetry (LDV) is a nonintrusive technique to measure fluid velocity at a focused point in the field. It works based on the Doppler shift of the light scattered from small particles [28,29]. When two laser beams are focused at a location, they form a fringe pattern due to the interference between them. If a particle travels through the fringes, it scatters the light depending on the Doppler frequency. The scattered light is a measure of the particle velocity. Different LDV configurations such as reference beam, dual beam, singlecomponent, and multicomponent have been developed. Detailed information on the hardware setup, data collection, and analysis procedure are presented in several books [28,30,31]. Although the method is nonintrusive, the flow needs to be seeded with tiny particles to act as light scattering media. For accurate measurements, the seeding particles need to be in dynamic equilibrium with the flow.
575
Flare Experimental Modeling
29.3.4.6.2 Planar Imaging Techniques Planar imaging techniques are used to obtain a 2-D, fullfield view of flames [32]. Acquired images provide useful information on the structure of the entire flow field. There are two major planar techniques applicable for the study of TDFCF: (i) planar image velocimetry (PIV) and (ii) planar laser induced fluorescence (PLIF). Hasselbrink and Mungal [33] and Han and Mungal [9] employed the PIV technique to measure the velocity field in a methane diffusion flame in crossflow. In a PIV technique, the flow is seeded with tiny particles like in LDV measurements. A pulsed laser sheet is shone on the field of view and the seeding particles are illuminated. Successive images are acquired using an intensified charge coupled device (ICCD) camera. The displacement of the particles over the known time interval (from the firing rate of laser) is measured from the image. The velocity is then calculated based on the first-order difference scheme. Hydroxyl (OH) and CH radicals are very good markers of the flame reaction zone. Goh and Gollahalli [34] have used the PLIF techniques to measure OH concentrations of propylene flames in a quiescent environment. Han and Mungal [9] and Hasselbrink and Mungal [33] employed the PLIF technique to measure OH and CH concentrations in deflected methane jet flames. In the PLIF technique, a narrow band laser sheet is shone on the flame to selectively excite an atom or molecule to a specific vibrational and electronic level from its original state. This causes a fluorescence emission from such excited species. The fluorescence emission from the 1. Computer with DAQ 2. ICCD camera 3. Camera controller 4. Camera gate and trigger 5. Optical lenses 6. Pulse generator 7. Combustion chamber 8. Flame 9. Bumer
excited species is proportional to the total density of the population of probed species at the ground state. 29.3.4.6.3 Soot Concentration Measurements Volumetric soot concentration in a TDFCF can be nonintrusively measured by laser beam attenuation (LBA) and laser induced incandescence (LII) techniques. The LBA technique was based on a method developed by Yagi and Iino [35] and the basic concepts for LII diagnostics were derived from Melton [36]. In the LBA technique, a laser beam (He–Ne laser at a wavelength of 635 nm, for example) is passed through the flame field. A power detector is placed at the other end of the light source to detect the intensity of light. The beam attenuation due to the presence of soot is obtained by measuring the intensity of light with and without the flame field. The soot volume fraction is obtained by knowing the optical path and the wavelength of laser used [13–15]. Goh and Gollahalli [34] used the LII technique to obtain the soot concentration tomographic plots in propylene flames in crossflow. A typical setup used by them for such measurements is presented in Figure 29.3. In a LII technique, the soot particles are heated up to a temperature above the surrounding gas temperature by a laser beam and the incandescence from soot particles is then measured. The incandescence is a strong function of elevated soot particle temperature. The measured LII signal has a linear relationship with soot concentration at low values of laser fluence. Further information on soot diagnostics with LII is available [see 36–42].
10. To fuel supply train 11. 431 nm filter 12. X-Y traverse mechanism 13. 1064 nm minor 14. Beam splitter 15. Nd-YAG laser 16. Focusing lenses 17. 1064 nm laser
7
12 8
2
1
11
17 5
13
9
16 13
3 15 6
Figure 29.3 Experimental setup for soot concentration measurements with LII.
© 2011 by Taylor and Francis Group, LLC
14 16
10
4
13
576
Industrial Combustion Testing
In addition to DAS, specialized software is also needed for laser diagnostics. Laser signals from the flame are usually captured by a camera, and instrument-specific software is available for controlling the camera. For instance, Han and Mungal [9] employed a MATLAB® based in-house software PIVlab to process their PIV images. Goh and Gollahalli [34] used WinSpec to process their OH-PLIF measurements with ICCD camera.
29.3.5 Data Acquisition System A typical TDFCF grows up to 2–5 m (6.6–16.4 feet), depending on the value of R. As discussed in Section 29.2, there are several parameters that affect the performance of a flare system. In order to quantify the performance, one needs to measure many quantities, and a manual data collection procedure is laborious in such situations. A computer loaded with a data acquisition software (DAS) such as LabVIEW© can be used for this purpose. Such DAS usually accept the analog (differential or single-ended) or digital output from a wide range of instruments. The DAS automatically stores the data into a file, which can be easily processed by any other graphing utility such as Tecplot©. Furthermore, the temperature, velocity, and species concentration profile measurements require that the probe travel a long distance inside the flame. To simplify the operation, the probes could be mounted on a two or three-dimensional traversing mechanism and a DAS-driven stepper motor could control the movement of the traversing unit for accurate and automated positioning of probes. Most DAS’s are capable of controlling the sampling rate and duration, and performing instantaneous statistical analysis. Goh and Gollahalli [20,34] used a sampling rate of 1 kHz and the data were time-averaged over two minutes to yield a most reliable mean value.
(a)
(b)
(e)
29.4 Regimes of Operation 29.4.1 Lifted Versus Wake-Stabilized Flames Primarily, the jet-to-crossflow momentum flux ratio, R, determines the operating mode of crossflow flames. Figures 29.4 and 29.5 show the flame configurations at various operating conditions of model flares fed with propane [2] and natural gas [43], respectively. For low velocities of crossflow (i.e., at high values of R, typically higher than 10), increasing the fuel jet velocity causes the flame to lift, bend over in the direction of crossflow, and stabilize at some distance away above the exit plane of the burner tip. In this configuration, streamwise-oriented, counterrotating vortices are found to influence the mixing and
(c)
(f)
(d)
(g)
(h)
Uj U∞
(i) Blue Orange red Luminous yellow
(j) Turbulent diffusion Flames in cross-wind Fuel: propane
(k)
A B C D E F G H I J K
dj (mm) 5 5 5 5 5 5 5 5 5 5 5
Rej Uj U /U M /M (m/s) × 10–3 j ∞ j ∞ 3.9 5.27 1.16 1.4 5.7 7.07 1.53 3.2 13 16.3 4.92 36 13 16.3 7.11 75 19 23.2 10.1 153 23 28.8 12.8 378 56 69.0 17.6 465 82 100 19.6 574 63 77 21.6 698 72 88 24.6 900 43 52.7 25.4 961
Figure 29.4 Sketches of turbulent propane diffusion flames in crossflow for different jet-to-crossflow momentum flux ratios. (Reprinted from Gollahalli, S. R., Brzustowski, T. A., and Sullivan, H. F., Transactions of the Canadian Society for Mechanical Engineers, 3, No. 4, 205–14, 1975. With permission from Canadian Society of Mechanical Engineering.)
© 2011 by Taylor and Francis Group, LLC
577
Flare Experimental Modeling
(a)
(b)
(c)
(d)
(e)
(f)
(g)
R = 5.29 U∞ = 0.70 m/s Vj = 2.07 m/s texp = 0.001 s R = 1.43 U∞ = 1.33 m/s Vj = 2.08 m/s texp = 0.002 s ηc = 99.4% R = 0.34 U∞ = 2.76 m/s Vj = 2.10 m/s texp = 0.002 s ηc = 99.3% R = 0.15 U∞ = 4.09 m/s Vj = 2.09 m/s texp = 0.001 s ηc = 99.7% R = 0.085 U∞ = 5.49 m/s Vj = 2.11 m/s texp = 0.002 s ηc = 99.2% R = 0.038 U∞ = 8.27 m/s Vj = 2.11 m/s texp = 0.002 s ηc = 96.5% R = 0.021 U∞ = 11.05 m/s Vj = 2.09 m/s texp = 0.002 s ηc = 88.6%
(h) R = 0.0085 U∞ = 17.39 m/s Vj = 2.10 m/s texp = 0.002 s
0.6 m 1.9 m
Figure 29.5 Photographs of natural gas flames in crossflow (Vj —jet exit velocity and texp—exposure time). (Reprinted from Johnson, M. R., and Kostiuk, L. W., Combustion and Flame, 123, 189–200, 2000. With permission from Elsevier.)
combustion characteristics [2,16]. Flares encountered during start-up, shut down, or an emergency situation in petroleum refineries are modeled as turbulent diffusion flames in crossflow with high values of R. For low values of R (lower than 1), the flame is usually attached to the burner tube and stabilizes in the wake of the burner. Increasing the fuel jet velocity, usually, does not cause flame lift-off. In this configuration, the mixing and combustion characteristics are dominated by two-zone flame structures on the lee side of the burner and coherent structures evolving from the upwind side of the jet [16]. Huang and coworkers [16,44,45] further
© 2011 by Taylor and Francis Group, LLC
classified the wake-stabilized flames into several subcategories. Flares encountered in normal operations of refineries typically have low values of R. 29.4.2 Wake-Stabilized Flames Following Huang and Chang [16], Gollahalli and Nanjundappa [21], and Huang and Wang [45], the wake-stabilized flames can be generally categorized as follows: (i) down-wash flame, R ≈ 0.1; (ii) crossflowdominated flame, 0.1 < R < 1; (iii) jet-dominated flame, 1 < R < 10.0, and (iv) strong-jet flame, R > 10.0.
578
In down-wash configuration, the flame is established in the wake of the burner tube. A recirculation vortex in the wake of a burner tube appears as a flame sheet. When R is further reduced, the flame tip is severely deflected by the crossflow. A small recirculation bubble was observed by Huang and Chang [16] at R = 0.04. For a value of R between 0.1 and 1, the impact of a crossflow stream is dominant. An axisymmetric tail flame forms downstream of the recirculation vortex and the flame widens. This structure is characterized by several features such as flickering and bifurcation. In jetdominated mode, the recirculation vortex disappears and only the tail part remains attached to the burner. The transition from crossflow-dominated to jet-dominated conditions occurs from R = 1 to 3. For R > 3, the effect of crossflow becomes negligible; the jet fluid mechanics dictate the flame characteristics. For R > 10, the flame detaches from the burner and stabilizes above the exit plane of the burner tip. Depending upon the jet exit velocity and burner diameter, the flame is either attached to the burner tip or stabilizes as a lifted flame until it blows out. 29.4.3 Effects of Buoyancy on Flame Configuration In quiescent ambient conditions, buoyancy is an important factor in the entrainment of air and must be accounted for flame length and other characteristics. However, the presence of crossflow overwhelms the effects of buoyancy, and hence, it can be neglected. A theoretical investigation of Escudier [46] and experimental observations by Gollahalli, Brzustowski, and Sullivan [2]; Brzustowski [47]; and Kalghatgi [5] confirm the independence of flame configuration on buoyancy. Escudier [46] further noted that the effects of buoyancy become important farther downstream of the flame beyond the flame tip. Nevertheless, the results presented in this chapter are obtained with the assumption of negligible buoyancy.
29.5 Global Characteristics 29.5.1 Flame Configuration and Appearance Flames in the crossflow can exist in numerous shapes depending on the jet-to-crossflow momentum flux ratio, Brzustowski’s research team at the University of Waterloo studied turbulent hydrogen and hydrocarbon flames in crossflow and presented flame shapes, trajectories, and correlations for flame length in a series of papers [2,10,47–51]. As noted before, at low values of R the flame stabilizes in the wake of the burner. At these conditions, due to the intense mixing between the jet
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
fluid and ambient air, the flame appears blue (similar to a premixed flame). The yellow region in the vicinity of the burner signifies the fuel pyrolysis and subsequent soot combustion. At some point downstream, a necking was observed, and the flame assumed a two-zone structure. The point of necking moved closer to the burner as the jet-to-crossflow momentum flux ratio was increased. Kalghatgi [5] described the TDFCF as a frustum of a cone, which was expressed in terms of burner diameter, the burner exit velocity, the crossflow speed, and the density of burner gas. 29.5.1.1 Low Jet-to-Crossflow Momentum Flux Ratio Flames Gollahalli and Nanjundappa [21] observed that at R = 0.27 the flame bent over, stabilized in the burner wake and resided just below the exit plane (Figure 29.6a). A luminous yellow flame and a recirculation zone were seen to attach to the lee side of the burner tube. The attachment length extended up to 20 d, where d is the burner exit diameter, below the burner exit plane. Soot deposit was also observed on the lee side of the burner. A dark zone was observed for about 2 d immediately below the burner exit on the lee side. These two zones corresponded to a planar diffusion flame formed by the fuel entering the recirculation region and the crossflowing wind tunnel stream. A long axisymmetric flame with a luminous yellow-colored tail followed this planar flame in the downstream direction. This region is influenced by the large entrainment into this region and gas-phase reactions. Further, Gollahalli and Nanjundappa [21] observed that as the momentum flux ratio was reduced to 0.17, the planar flame zone became larger and more luminous. The axisymmetric flame became narrower, longer, and less yellowish (Figure 29.6b). When this was further reduced to 0.11, the axisymmetric flame zone became entirely blue (corresponding to a fuel-lean premixed flame). The planar part became whitish and the overall length in the streamwise direction was reduced (Figure 29.6c). With further reduction of R to 0.02, the axisymmetric flame became fully extinct, and the planar zone became more yellow and luminous (Figure 29.6d). These types of flames are characterized by their single-zone structure. Huang and coworkers [16,44,45] have also studied the wake-stabilized jet flames in crossflow. They identified six flow regimes corresponding to different jet-to-crossflow momentum flux ratios: down-wash (R < 0.1), crossflow-dominated (0.1 < R < 1.6), transitional (1.6 < R < 3.0), jet-dominated (3.0 < R < 10.0), and strong jet (R > 10) modes. Figures 29.7a through 29.7f present the flow patterns on the symmetry plane of the jet and wake regions [45]. In the down-wash flow mode, a small
579
Flare Experimental Modeling
(a)
(d)
0
20
–20
0
(b)
(e)
0
20
–20
0
(c)
(f) 40
R = 0.27
(b)
R = 0.17
Z (mm)
(a)
0 (c)
R = 0.11
20
–20
0 0
(d)
R = 0.02
Uj
U∞ Figure 29.6 Photographs of low momentum propane gas jet diffusion flames in crossflow. (Reprinted from Gollahalli, S. R., and Nanjundappa, B., Combustion Science and Technology, 109, 327–46, 1995. With permission from Elsevier.)
recirculation bubble was observed near the burner tip at R = 0.04. At 0.1 < R < 1.6, the jet momentum was not enough to overcome the shearing effects of crossflow stream, and hence, the jet fluid from the burner tip bent through a large deflection angle. The flow structure was characterized by several features such as flickering and bifurcation. In jet dominated and strong-jet modes, the recirculation bubble vanished and only the tail portion remained, as also observed by Gollahalli and Nanjundappa [21]. Experimental results obtained by Gollahalli and Pardiwalla [11] with turbulent flames from circular and elliptic burners in crossflow also confirmed the existence of these flow configurations. 29.5.1.2 High Jet-to-Crossflow Momentum Flux Ratio Flames At higher jet-to-crossflow momentum flux ratios, the recirculation zone or the planar flame zone disappeared
© 2011 by Taylor and Francis Group, LLC
20
40
60
X (mm)
0
20
40
60
Figure 29.7 (see color insert following page 424.) Visualization of flow patterns of the plane of symmetry: (a) R = 0.04 (Uj = 0.8 m/s, Rej = 909), downwash mode; (b) R = 0.16 (Uj = 1.6 m/s, Rej = 1818), crossflow dominated mode; (c) R = 0.70 (Uj = 3.3 m/s, Rej = 3749), cross-flow dominated mode; (d) R = 2.47 (Uj = 6.2 m/s, Rej = 7043), transitional mode; (e) R = 4.32 (Uj = 8.2 m/s, Rej = 9315), jet-dominated mode; and (f) R = 12.60 (Uj = 14.0 m/s, Rej = 15,904), strong jet mode. Flame. (Reprinted from Huang, R. F., and Wang, S. M., Combustion and Flame, 117, 59–77, 1999. With permission from Elsevier.)
and only the tail part remained attached to the burner. Gollahalli, Brzustowski, and Sullivan [2] observed that above a certain value of R, the flame detached from the burner and stabilized above the jet exit plane. Figure 29.4 shows the flame shapes recorded by these authors. At high R, the fuel was stripped from the jet and an intense mixing between the fuel and crossflow stream occurred. This caused the base of the flame to turn blue corresponding to an annular premixed flame regime. The remaining fuel pyrolyzed downstream and burned in diffusion mode. The combustion characteristics in this region were influenced by the entrainment rate and the availability of oxygen. Further increase in R broadened the size of the initial annular blue region. The downstream luminous yellow zone continued to be present. At flame edges, orange–red wrinkles started to form. Also observed was the formation of counter-rotating vortices, which transformed the initial circular shape of the flame to a kidney-shaped section at farther downstream locations. Similar observations were noted by Bandaru and Turns [3] at R = 100, 240, 525, and 900. 29.5.2 Flame Geometry Measurement of the crossflow flame geometry is necessary to the understanding of the extent of flare dispersion, radiative heat transfer to the surroundings, and
580
Industrial Combustion Testing
pollutant emission. A number of experimental investigations have been conducted to quantify the size, shape, and trajectory of such flames [2,5,11,16,21,47–55]. Techniques such as color photography, shadowgraph, SO2 tracing, reactive Mie scattering, and PIV are used to measure the flame geometry. 29.5.2.1 Flame Size Gollahalli, Brzustowski, and Sullivan [2] measured the size and centerline trajectories of TDFCF in high momentum flux ratios. Figure 29.8 presents the normalized flame width and thickness measured normal to the visible centerline at various downstream locations. After the jet was discharged into the crossflow stream, there was an initial sudden raise in the flame thickness due to the initiation of combustion process (see insert Figure 29.8a). The cross section continued to increase as the entrainment of air and combustion of pyrolyzed fuel components occurred downstream. After some point downstream, the rate of combustion decreased, which in turn decreased the cross-sectional dimension. Correlations for normalized flame height as a function of the distance from the fuel orifice along the axis of the curved flame have also been developed. With an increase in crossflow velocity the flame height decreased up to a certain point, beyond which the flame height increased. However, when the blowout was approached, the flame height again decreased. Kalghatgi [3] conducted extensive experiments to study the shape and sizes of four different hydrocarbon flames in crossflow. The angle between the burner
Fuel jet
0.8 0.4 0.0
0.2
2.4
0.4
x
Flame
Frj Rej × 10–3 × 10–3 C 30.5 34.5 34.5 C 30.5 34.5 C 30.5 C 60.8 138.6 C 59.6 134.2 34.5 C 30.5 C 60.8 138.6 C 59.6 134.2 30.4 S 32.4 36.9 S 21.2 93.2 S 50.1 S 62.6 145.5
x
x 0.6
0.8 – S
1.0
1.2
1.4
∆N
2.0 1.6 1.2 0.4 0
2
4
6
8
10 12 14 16 18 20 – S
(29.1)
(b) Mj /M∞ 223 362 527 744 966 1160 1357 1607 316 135 753 1177
C : From color slides S : From shadowgraphs Fuel : Propane dj : 3.8 mm
0.8
Lf 1/2 = K f (ρ jU j ) d j + KUU ∞ . C1f/2
12
Rej × 10–3 Fr × 10–3 Mj /M∞ j
58 39 23 16
10 8 ∆Y
(a)
axis and wind direction was also varied. The author presented the variations of flame width, thickness, and length with velocity ratio. Results obtained with burners located inclined to the wind direction are also presented. Huang and Chang [16] presented the variation of total flame length (blue and orange zone) with jet exit velocity (Uj) for different crossflow velocities. Results indicate that the flame length increased with Uj through all configurations of the flame. However, during the dual-flame mode, a sharp decrease was observed and the flame length continued to increase after that. This decrease was attributed to the shortening of soot-radiating zone. Bandaru and Turns [3] presented the effect of momentum flux ratio on normalized flame length. It was observed that the flame length increased initially, and leveled off or slightly decreased at higher momentum flux ratios (R > 100). Gollahalli and Pardiwalla [11] reported normalized visible flame length as a function of jet-to-crossflow velocity ratio. The flame length was found to increase sharply until the effects of jet started to influence. The visible flame length was approximately 200 times the burner diameter. Majeski, Wilson, and Kostiuk [55] developed a phenomenological model for predicting the length of low-momentum jet flames in crossflow at R ≤ 1. They derived an expression for estimating the flame length in terms of jet and crossflow velocity, stack diameter, and fuel dilution. The expression is presented below:
45.0 20.3 7.1 3.4
1073 242 151 68
Jet fluid : Propane dj : 5 mm
6 4 2 0
2
4
6
8
10 – X
12
14
16
Figure 29.8 Variation of flame cross-sectional dimension of propane diffusion flame in crossflow: (a) along the centerline axis of the flame and (b) along the horizontal coordinate. (From Gollahalli, S. R., Brzustowski, T. A., and Sullivan, H. F., Transactions of the Canadian Society for Mechanical Engineers, 3, No. 4, 205–14, 1975.)
© 2011 by Taylor and Francis Group, LLC
581
Flare Experimental Modeling
The symbols are identified in the Nomenclature s ection. An analytical model for predicting the flame shape and length in crossflow is presented by Karagozian [56]. 29.5.2.2 Flame Trajectory Escudier [46] numerically integrated a set of coupled, nonlinear, first-order, ordinary differential equations using a fourth-order Runge-Kutta method to study the motion of a plume. The author presented plume trajectory in logarithmic coordinates in the near burner and far-field regions. For flames with high momentum flux ratio, Gollahalli, Brzustowski, and Sullivan [2] presented experimental data for visible centerline trajectory. Figure 29.9 shows the results obtained by the authors in nondimensional form. Also shown in the plot are the theoretical solution Z = 2.5 X 0.5 for near nozzle [50] and the experimental correlation Z = 2.05 X 0.28 for the visible centerline of an air jet [57]. Shapes of hydrogen diffusion flames in crossflow are presented in Brzustowski Gollahalli, and Sullivan [49]. Brzustowski [10] described a general calculation procedure for flame shape and length. Rao and Brzustowski [51] presented more experimental results on the trajectory of a TDFCF with SO2 tracing. Results indicate that the trajectory varied as a power law, Z = M X n. With an increase in the flame Froude number, the index of the power law, n, decreased and the constant of proportionality, M, increased. Table 29.1 summarizes the values of M and n used by different researchers for the trajectory calculations of TDFCF. Effects of ambient pressure and buoyancy on the structure and geometry of natural gas and propane gas jet flames in crossflow were studied by Choudhuri and Gollahalli [58] using quantitative rainbow Schlieren deflectometry.
5.0 – Z=
– 2.05 X
29.5.3 Stability Characteristics Under turbulent conditions, a stable flame is established when the local gas velocity is equal to the local turbulent flame speed. The local turbulent flame speed depends on mixture properties and flow conditions. In TDFCF, the crossflow velocity is an important parameter that must be accounted for. Keefer and Baines [59] expressed the local gas velocity, U, in terms of jet and crossflow velocity, as follows: U = C1 (U j − U ∞) + U ∞ ,
where C1 is a constant. In crossflow literature, the jet-tocrossflow momentum flux and velocity ratios are used to explain the stability of the flame. When a crossflow flame is anchored to the burner rim, it is called the burner-attached flame. This occurs when the stabilization point is in the plane of burner rim. If the local gas velocity is increased above a critical condition, denoted as the lift-off limit, the flame detaches from the burner and stabilizes at some distance above the burner tip. The distance between the burner tip and base of the flame is called the lift-off height. If the gas velocity is further increased, the flame blows out (i.e., extinguishes). This condition is called the blowout limit. Gollahalli, Brzustowski, and Sullivan [2] reported that the lift-off distance increases with an increase in jet exit Reynolds number and jet-to-crossflow velocity ratio. Kalghatgi [4] presented a universal nondimensional stability diagram from a set of systematic experiments with different fuels, burner diameters, and jet-to-crossflow velocities. This curve presents the limiting conditions for stable, lifted, bent-over flames. Huang and coworkers [16,44,45] conducted a stability analysis of the propane jet flames in crossflow and
0.28
– Z
2.0 /2
1.0
–1 X 2.5
0.5
– = Z
0.2 0.01
0.03
0.10
0.30 – X
1.0
3.0
(29.2)
10.0
dj mm
Rej × 10–3
5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0
22.2 16.4 25.7 49.7 49.7 39.7 68.3 68.3 115.0 16.4 52.7 42.0
Mj/M∞ Uj/U∞ 63 80 83 160 197 200 289 436 589 645 757 1093
Frj * × 10–3
6.5 6.5 7.4 3.5 7.5 8.7 10.5 33 11.5 33 11.6 21 13.9 64 39.4 64 20.4 184 20.9 3.5 22.9 38 27.2 30
S S S S S S C C C S C C
S-From shadowgraphs C-From color slides
Figure 29.9 Trajectories of propane turbulent diffusion flames in crossflow. (From Gollahalli, S. R., Brzustowski, T. A., and Sullivan, H. F., Transactions of the Canadian Society for Mechanical Engineers, 3, No. 4, 205–14, 1975.)
© 2011 by Taylor and Francis Group, LLC
582
Industrial Combustion Testing
Table 29.1
16
Summary of TDFCF Trajectory Data
14
Pratte and Baines [57] Brzustowski [50] Rao and Brzustowski [51] Huang and Chang [16]
Far-Burner
M
n
M
n
− 2.5 1.99 0.86
− 0.5 0.45 0.37
2.05
0.28
− − 1.07
− − 0.21
12 10
Uj (m/s)
Reference
Near-Burner
Detached
Partially detached
8 6 4
Type II
2 30 Critical regime Mode IV
20
Uj (m/s)
15
Subcritical regime
10
Supercritical regime
Stable
0.2
Mode VI
6
9
12 Uw (m/s)
Uj (high)
Type II flame
5
6
Type I flame
crossflow velocity. Elliptic burners are found to have wider stability range. Potential use of elliptic burners for flares is discussed in Section 29.7.1.
Lower stability limit 15
18
Figure 29.10 Stability diagram for propane jet flames in different operating modes. Mode I: down-washed, Mode II: flashing flames, Mode III: developing flames, Mode IV: dual flames, Mode V: flickering flame, Mode VI: pre-blowout flames. (Reprinted from Huang, R. F., and Chang, J. M., Combustion and Flame, 98, 267–78, 1994. With permission from Elsevier.)
developed a stability diagram with jet and wind velocities as parameters (See Figure 29.10). As described in Section 29.4.2, Huang and coworkers subdivided the burner-attached flames into six categories. Figure 29.10 also identifies the stabilization regimes of those flames. Gollahalli and Nanjundappa [21] reported stability regimes of propane diffusion flames in crossflow. Figure 29.11 presents the variation of jet exit velocity with crossflow velocity at two critical conditions corresponding to burner-attached and detached flames. The lower jet velocities refer to burner-wake stabilized flames, as also studied by Huang and Chang [16]. The upper jet velocity values correspond to the lower stability curve of Kalghatgi [4]. The authors concluded that the overall turndown ratio (highest-to-lowest jet velocity) for a flame in crossflow could be as high as 1000. Gollahalli and Pardiwalla [11] studied the circular and elliptic burner flames in crossflow. For burner-attached flames, the authors presented the variation of jet velocity with
© 2011 by Taylor and Francis Group, LLC
4 U∞ (m/s)
Stable
Mode II
Unstable 3
3
Figure 29.11 Stable range of operation of propane jet diffusion flames in crossflow. (Reprinted from Gollahalli, S. R., and Nanjundappa, B., Combustion Science and Technology, 109, 327–46, 1995. With permission from Elsevier.)
Mode I
0.4
Type I 2 Uj (low)
Mode V
Mode III
5
0.0
Unstable
Upper stability limit
25
0
29.5.4 Flaring Efficiency Efficiency of flaring (ηc) denotes the degree of conversion of flare gas into stable combustion products such as CO2 and water. Generally, the carbon atoms in the flare gas and the combustion products are traced and related to the flare efficiency. The efficiency of a flare based on carbon atom balance is given below:
mass flowrate of carbon available in the form of CO 2 in flame . ηc = mass flowrate of carbon supplied with flare gas
With gas analyzers, the concentrations of carbon containing compounds in the combustion products could be measured, and the combustion efficiency is calculated. Sometimes, flare efficiency is also expressed in terms of combustion inefficiency, defined as (1-ηc), and denoting how much combustible gas is left unburned during a flare operation. Pohl et al. [60] measured the combustion efficiency of propane-nitrogen steam-assisted flares. The results indicated that the size of the flare and the soot did not influence the combustion efficiency. The authors reported an efficiency of over 98% for flares operating with stable flames. Kostiuk and coworkers [6,43,61,62] studied the
583
Flare Experimental Modeling
14 12 Conversion inefficiency, (1–η) (%)
Strsosher [63] measured the emissions of laboratory, pilot, and field-scale flares and calculated the flaring efficiency. Leahey, Preston, and Strosher [64] used empirical relationships to calculate flare efficiency at different conditions such as flare stack velocity, stoichiometric mixing ratio, and wind speed. The calculations showed that the flare efficiency decreased as the crossflow speed increased in conformity with other studies.
Propane Vj = 1 m/s Vj = 2 m/s Natural gas Vj = 1 m/s Vj = 2 m/s Vj = 3 m/s Vj = 4 m/s 60% Propane/40% CO2 Vj = 1 m/s Vj = 2 m/s Vj = 3 m/s
10 8 6
29.5.5 Radiation Emission
4 2 CF1-9-EDNXXPXXCXXVXXUXXLAM-UV13-P2,GRF
0 0
2
4
6 8 U∞/V j 1/3 (m/s)2/3
10
12
14
Figure 29.12 Effect of crossflow and jet exit velocities on the combustion inefficiency of different fuels. (From Johnson, M. R., and Kostiuk, L. W., Combustion and Flame, 123, 189–200, 2000.)
Flare generally appears as a very large turbulent diffusion flame. Radiative emission from such flames could significantly affect the surrounding environment. Radiation from a flame occurs from two sources: (i) infrared emissions of CO2 and H2O, (ii) visible-infrared emissions of soot particles [65]. In order to characterize the overall radiative emission in a global sense, the flame is treated as a point source with radiation emission as a fraction, F, of the total heat release. In TDFCF, the fraction F depends on the type of fuel and aerodynamics of the flame [66]. The radiant heat flux K incident on a unit area of a surface located at a distance D from the point source is estimated as,
combustion efficiency of commercial grade natural gas and propane flares. Figure 29.12 presents results of combustion inefficiency for different fuels. The combustion inefficiency was expressed in terms of U ∞ /U 1j /3 to account for the interaction of buoyancy and momentum. The results suggest that the flare operates at higher efficiencies at low crossflow velocities and/or higher jet exit velocities. A method for computing the combustion efficiency is presented in Bourguignon, Johnson, and Kostiuk [6]. Johnson and Kostiuk [61] further examined the combustion efficiency of flares and at different operating conditions. Based on the experimental results, the following general empirical correlation has been developed for combustion efficiency of flares:
, (29.3) 1/3 ( gU j d j )
(1 − ηc )( LHVmass )3 = A exp B
U∞
where A and B are coefficients that depend on the type of fuel. The correlation is applicable in the following range: Crossflow velocity: 2 m/s ≤ U∞ ≤ 17 m/s Jet exit velocity: 0.5 m/s ≤ Vj ≤ 4 m/s Flare tip diameter: 12.2 mm ≤ do ≤ 49.8 mm Coefficients: A − 156.4 (MJ/kg)3 and B − 0.318 for natural gas
© 2011 by Taylor and Francis Group, LLC
K=
FQ Re l , 4 πD 2
(29.4)
where Q Rel is the rate of heat release by combustion. Brzustowski et al. [66] also delineated the dependence of F on jet and crossflow velocity. Flare radiation can also be estimated by assuming the flame as a surface source [67]. Bandaru and Turns [3] presented experimental results on radiant heat fractions based on the single-point source assumption for flares. They presented radiant heat fraction values between 7% and 12% for ethylene flames and noted that the results did not correlate clearly with momentum flux ratio. For propylene and methane flames, Goh and Gollahalli [20] showed that the effects of piloting caused the flame radiation to decrease. Schmitt [68] measured the radiation and dimensional characteristics of hydrogen flares. Schwartz and White [69] presented a critical review of the methods available for predicting the radiation from flare. The authors compared the measured radiation intensity with the prediction. Table 29.2 lists the test conditions used for the comparison. All the methods overpredict the radiant fraction in the near burner region. However at farther downstream locations, all the models seem to agree with each other and also with experimental data. For a detailed numerical investigation of flare radiative emissions, the reader is referred to Fairweather, Jones, and Lindstedt [70]. The predictions are also compared with laboratory and field-scale experimental data.
584
Industrial Combustion Testing
Table 29.2 Experimental Conditions Used by Schwartz and White Parameter Flare gas composition Molecular weight Lower heating value Wind velocity Stack height Outer diameter
Value Methane − 85%; N2 − 8%; CO2 − 7% 18.96 8600 Kcal/kg 3.7 m/s 46 m 610 mm
Source: From Schwartz, R. E. and White, J. W., Proceedings of the 30th Annual Loss Prevention Symposium of the American Institute of Chemical Engineers; paper #12a, 1996.
29.6 Flame Structure Measurements Measurement of inflame data provides an insight into the global characteristics of flames such as radiation and pollutant emission. Due to the inherent asymmetric nature of TDFCF, inflame measurements are usually taken along horizontal and vertical directions at various locations along the flame trajectory. This section presents a summary of data on inflame temperature, species concentration, and velocity profile measurements in TDFCF. 29.6.1 Temperature Field Inflame temperature measurements are vital to the estimation radiative losses from flares. In a pioneering work, Brzustowski et al. [66] studied the effect of inflame temperature on the radiant heat fraction of propane TDFCF. At high values of R, the flames were characterized by counter-rotating vortex pairs. The flame was blue in the near-burner region followed by a yellow luminous zone. Due to the intense mixing and increased entrainment of air into flames, combustion was more complete in the regions of vortex pair. Temperature measurements of Brzustowski et al. [66] provided an evidence for this observation. The authors also observed off-axis peaks in the horizontal temperature profiles at R = 120. However, at low values of R, combustion, and hence, heat release were found to be more uniform across the flame cross-section. Goh and Gollahalli [20] measured temperature profiles in piloted and nonpiloted propane and propylene flames in crossflow at R = 32–97. The profiles are generally characterized by off-axis single peak structure for all flames. The nonpiloted flames produced higher peak temperatures signifying increased oxidation of O2 under the same flow conditions. Tsue, Kadota, and Kono [71] measured the structure of propane diffusion flame in crossflow, including the flame temperature, velocity and concentration fields. Further, Tsue, Kadota, and
© 2011 by Taylor and Francis Group, LLC
Kono [72] studied the temperature and velocity fluctuations in a hydrogen/methane TDFCF using Rayleigh scattering and LDV, respectively. Largest fluctuations in temperature and velocity were observed in the upper edge of the flame. The temperature field in low R flames was probed by Gollahalli and Nanjundappa [21] and Huang and Yang [22]. Gollahalli and Nanjundappa [21] measured the temperature profiles at R = 0.02 and 0.17. The authors observed that the temperature profile featured a double peak structure with the peak temperature being at the bottom of the vortex eye (See Figure 29.13). The combustion was confined to the vortex edges in the form of thin flame edges. Further, the wake formed on the lee side of the model stack recirculated the combustion gases toward the burner, and thus, offering more time to mix with crossflow stream and burn completely. Due to this reason, the bottom edge of the vortex was usually at a higher temperature than the upper edge. At downstream locations, the flame edges disappeared and the flame burned in premixed or partially premixed mode with single temperature peak. Huang and Yang [22] measured temperature field in detail to understand the thermal structures in downwashed, flashing, developing, dual, flickering, and pre-blowoff configurations of flames in crossflow. The corresponding jet-to-crossflow momentum flux ratios for these six configurations are presented in Section 29.4.2. Results obtained by Huang and Yang [22] were in accordance with the findings of Gollahalli and Nanjundappa [21] and further provided a thorough analysis of the flame structure. Temperature measurements in down-washed flame mode revealed that the flame zone consisted of two “flame sheets.” No flame existed between the sheets. Figure 29.14 presents the temperature field in the downwashed flame mode. Off-axis peak temperatures and a dip at the center shown in Figure 29.14 suggest the existence of these flame sheets. However, at downstream locations, the two flame sheets merged into one sheet with single peak temperature occurring at the flame center. In flashing mode, the flame temperature behaved somewhat similar to the down-washed mode except that the two flame sheets were widely separated near the model stack. In developing flames, a tail flame started to appear and the temperature in the vicinity of the model stack dropped. Also, in the immediate downstream locations, a double peak structure appeared denoting the existence of flame sheets. Farther downstream corresponding to the tail flame, like in the other flame configuration, the flame sheets merged into one flame zone. In dual, flickering, and pre-blowoff flame modes, the dual peak structure of temperature disappeared except in the neck regions of dual flames. The flames had a uniform single peak temperature profile with higher temperature closer to the lee side of
585
Flare Experimental Modeling
Side view (a) 1700
Top view
(b)
(c)
X/d i = 1.64 Z/d i = 0
1275
X/d i = 2.64 Z/d i = 0
X/d i = 4.64 Z/d i = 0
850 425 0 (d) 1700
(e) X/d i = 1.64 Z/d i = 0.6
1275 T (°C)
(f) X/d i = 2.64 Z/d i = 0.6
X/d i = 4.64 Z/d i = 0.6
850 425 0
(g) 1700
(h)
(i)
X/d i = 1.64 Z/d i = 3.4
1275
X/d i = 2.64 Z/d i = 3.4
X/d i = 4.64 Z/d i = 3.4
850 425 0 –2
–1
0
1
2 –2
–1
0 Y/di
1
2
–2
–1
0
1
2
Figure 29.13 Measured inflame transverse temperature profiles in propane flames at R = 0.02. (From Huang, R. F., and Yang, M. J., Combustion and Flame, 105, 211–24, 1996.)
the burner. Further, the experimental data collected just before the blowoff condition indicated that temperature in the regions close to the burner decreased sharply. Temperature profiles of circular and elliptic burner flames in crossflow at R = 0.11 are available in Gollahalli and Pardiwalla [11]. 29.6.2 Species Concentration Field Measurements of inflame species concentration aid in the understanding of global pollutant emission and
© 2011 by Taylor and Francis Group, LLC
smoke release characteristics of flares. Combining with temperature data, the species concentration field can also be used to determine the extent of completeness of combustion, and hence, the efficiency of flaring process. Inflame data can also be used to monitor and control the flare operation. For instance, the flare gas can be injected with diluents such as steam and nitrogen to suppress smoke. This section presents a summary of measurements of CO, CO2, NO, and O2 concentrations in TDFCF under a wide range of jet-to-crossflow momentum flux ratios.
586
Industrial Combustion Testing
2500
X/d = 8.16, Z/d = 0
2000
2000
1500
1500
1000
1000
500
500
0 –40
–30
2500
Temperature (K)
2500
–20
–10
0
X/d = 5.44, Z/d = 0
0 –2
2000
1500
1500
1000
1000
500
500 –30
2500
–20
–10
0
X/d = 2.72, Z/d = 0
0 –2
2000
1500
1500
1000
1000
500
500 –30
–20 Y/d
–10
0
0 –2
0
1
2
X/d = 5.44, Z/d = 7.5
–1
2500
2000
0 –40
–1
2500
2000
0 –40
X/d = 8.16, Z/d = 7.5
0
1
2
X/d = 2.72, Z/d = 7.5
–1
0 Z/d
1
2
Figure 29.14 Inflame temperature profiles in down-wash flame mode (U∞ = 4.86 m/s; Uj = 1.00 m/s). (Reprinted from Gollahalli, S. R., and Nanjundappa, B., Combustion Science and Technology, 109, 327–46, 1995. With permission from Elsevier.)
Figure 29.15 presents the species concentration results obtained by Gollahalli and Nanjundappa [21] at R = 0.02. The authors observed that the CO2 concentration field in the flame followed the temperature field (i.e., a double peak structure in the vicinity of burner wake and the single peak structure downstream). Similar to flame temperature, CO2 concentration also assumed the highest value at the bottom edge of the vortex. Accordingly, the O2 concentration was the highest at the vortex edges and decreased to a minimum value at the vortex eye. Lower O2 concentration at the vortex eye could also be due to the centrifugal effects arising from the differences in molecular weight. This might limit the inward diffusion
© 2011 by Taylor and Francis Group, LLC
of O2 into the vortex core. Hence, the fuel trapped in the vortex core was only partially oxidized to CO. The CO concentration results also supported this behavior. The NO concentration profile exhibited a trend similar to that of temperature—with double peak structures and higher temperature being at the bottom edge of the recirculating vortex. At low jet-to-crossflow momentum flux ratios, detailed measurements of the concentration field were obtained by Huang and Yang [22]. They have presented the variations of CO, CO2, and O2 concentration inside the flame. Figure 29.16 presents the concentration profiles in a flashing flame. The O2 concentration varied
587
Flare Experimental Modeling
(a) 10
X/d = 8.16, Z/d = 0
8
(b) 7
X/d = 8.16, Z/d = 0
20 15
6
10
4
50
4
40
3
30 20
1
10
2
5
0 –30
0 –30
0 –30
10
–10
0
10
X/d = 5.44, Z/d = 0
8
10
CO2 (mole %) –20
–10
5 0
10
X/d = 2.72, Z/d = 0
8
0 –30
4
0
10
0
10
0 –30
40
3
30
2
20
–10
0
10
0 –30
50
10
4
40
3
30
2
20
1
10 0 –30
–20
–10 Y/d
0
10
0 –30
–20
–10 Y/d
0
–20
10
–10
0
10
X/d = 2.72, Z/d = 0
60
5
0 –30
10
NO (ppm v)
70
X/d = 2.72, Z/d = 0
6
0
10
CO (mole %) –20
–10
X/d = 5.44, Z/d = 0
60 50
0 –30
–20
70
4
7
5 –10 Y/d
10
15
2 –20
0
5
1
X/d = 2.72, Z/d = 0
20
6
–10
–10
X/d = 5.44, Z/d = 0
6
O2 (mole %) –20
–20
7
10
10
0 –30
0
15
4
0 –30
–10
X/d = 5.44, Z/d = 0
20
6
2
–20
X/d = 8.16, Z/d = 0
60
5
2
–20
70
X/d = 8.16, Z/d = 0
6
–20
–10 Y/d
0
10
Figure 29.15 Transverse concentration profiles in propane flames in crossflow: (a) CO2 and O2 and (b) CO and NO. (Reprinted from Gollahalli, S. R., and Nanjundappa, B., Combustion Science and Technology, 109, 327–46, 1995. With permission from Elsevier.)
from the free stream value to a minimum value at the center of the flame close to the burner. However, with the formation of “flame edges” in the immediate downstream, the O2 concentration again increased at the core (corresponding to no flame conditions). The locations of low O2 concentration corresponded to the peak locations of CO and CO2 concentration. The results suggest the formation of recirculation vortex on the lee side of the model stack. Gollahalli and Pardiwalla [11] reported the concentration profile measurements of circular and elliptic burner TDFCF at R = 0.11. Fairweather and coworkers [73–76] measured and predicted the behavior including inflame concentration of turbulent reacting jets in crossflow. Some field-scale results are also reported. 29.6.3 Velocity Field Flames in crossflow are largely influenced by complex fluid mechanics and aerodynamic interactions between
© 2011 by Taylor and Francis Group, LLC
the jet fluid and crossflow stream. Velocity measurements have been recorded in the near-burner and far-burner regions of TDFCF. Since many flame characteristics depend on the air entrainment at the flame base, velocity measurements at near-burner locations are very useful. Measurements made at far-flame locations aid the interpretation of radiation and pollutant emission characteristics. Botros and Brzustowski [77] studied the velocity field of propane TDFCF experimentally and numerically. Their study revealed a pair of counter-rotating vortices in the flame. Gollahalli and coworkers [78–80] measured the flow field and turbulent characteristics of gas jet flames in crossflow. At very low values of R, the effects of crossflow stream are more dominant and the jet fluid burns in the wake of a model stack. A recirculation vortex is created and the flame stabilizes on the wall of the recirculation bubble. Figure 29.17a presents the velocity vector field and streamlines obtained by Huang and Wang [45] for down-wash
588
Industrial Combustion Testing
(a) 20
Z (mm)
(a)
0
Top view
Side view (d)
5
CO (%)
4
–20
3
–60
2
–80
0 (b)
(e)
12
6 3
20
40 60 X (mm)
80
100
120
60 ψ (–3.2, 1) ψ (–2.5, 1)
0
0
15 O2 (%)
0
–60
(f )
20
50
100
150 X (mm)
200
250
300
Figure 29.17 Visualized velocity vector field and streamlines in the plane of symmetry: (a) R = 0.04, down-wash flames; and (b) R = 12.6, strong-jet mode. (Reprinted from Huang, R. F., and Wang, S. M., Combustion and Flame, 117, 59–77, 1999. With permission from Elsevier.)
10 5 0 –1.5
–20
–30
0 (c)
ψ (–1.2, 1)
30 Z (mm)
CO2 (%)
9
ψ (–2.2, 1)
ψ (–0.2, 1)
–40
1
(b)
ψ (–3.2, 1)
–1.0
–0.5
0.0 –1.5 Y/di
X/di = 1.64 Z/di = 0 X/di = 1.64 Z/di = 0.6 X/di = 1.64 Z/di = 1.4
–1.0
–0.5
0.0
X/di = 4.64 Z/di = 0 X/di = 4.64 Z/di = 0.6 X/di = 4.64 Z/di = 1.4
Figure 29.16 Measured concentration profiles of CO, CO2, and O2 in flashing flame (U∞ = 4.86 m/s; Uj = 1.00 m/s). (From Huang, R. F., and Yang, M. J., Combustion and Flame, 105, 211–24, 1996.)
flame configuration at R = 0.04. From the figure, one can see that the streamlines in the vicinity of burner tube show an evidence for the presence of recirculation vortex. The boundary of the recirculation bubble is shown as a dotted line in the figure. Streamlines that are emanating from the burner tube are curved downward and continued downstream. In strong jet mode (R = 12.6; Figure 29.17b), all the streamlines are pushed
© 2011 by Taylor and Francis Group, LLC
to downstream locations. The size of the recirculation region continued to decrease from down-wash mode and disappeared in strong-jet mode. Further, Huang and Wang [45] observed a singularity point (shown as a dot in Figure 29.17b) at the immediate downstream location of the burner. A horseshoe-like shear layer was formed due to jet and cross fluid interaction. A reverse flow was also noted in the upstream of the singularity point. Mungal and coworkers [9,33,81] conducted simultaneous measurements of velocity and OH/CH fields in deflected jet flames. The burner was oriented at 0º, 45º and −45º to the crossflow. Measured parameters include the contours of the OH/CH layer, mean velocity profiles and shear strain rate. Figure 29.18 shows the average velocity profile along the direction of crossflow at various axial locations in the flame [9]. The profiles clearly show the asymmetry caused by the crossflow stream. The authors also observed a lower mean strain rate on the lee side of the burner.
589
Flare Experimental Modeling
(a) 80
25
60 40
15
20 10
Uj (m/s)
u (m/s)
20
5
0 1.4
(b) 1.2 1
0
–10
–5 x/d = 3.4
0
y (mm)
5
x/d = 6.8
10
15
x/d = 10.2
Figure 29.18 Mean velocity profiles measured along the direction of crossflow at different axial locations in the flame (r—square root of momentum flux ratio). (Reprined from Han, D., and Mungal, M. G., Combustion and Flame, 133, no. 1–2, 1–17, 2003. With permission from Elsevier.)
29.7 Miscellaneous Topics This section discusses the results of some recent developments in experimental modeling of flares. The following topics are addressed here: i. Novel shapes for flare tip ii. Effects of diluents iii. Scaling of laboratory data 29.7.1 Novel Shapes for Flare Tip The shape of a flare tip influences several parameters including the extent of air entrainment into the flame and subsequent mixing. At low values of R, the shape also affects the stability of operation. Conventionally, flare tips with circular exit cross section have been used. Ho and Gutmark [82] and Gutmark, and Grinstein [83] studied the flow fields of elliptic cold jets in quiescent ambient conditions. Ho and Gutmark [82] studied the entrainment characteristics of a small-aspect-ratio elliptic jet. The entrainment ratio was found to be several times greater than that of a circular jet. Gollahalli, Khanna, and Prabhu [84] experimentally studied the characteristics of burning and nonburning propane jets issued from elliptic jets into a low velocity concentric air stream. Elliptic burner flames had lower soot concentration and radiation emission than those of circular burner flames. Pardiwalla [85] compared the combustion characteristics of propane jet flames from circular and elliptic
© 2011 by Taylor and Francis Group, LLC
0.8 0.6 0.4 0.2 0
0
2
4 Circular
6 8 Uw (m/s) Elliptic
10
12
14
Elliptic
Figure 29.19 Stability regimes of circular and elliptic burner flames in crossflow: (a) upper stability limit and (b) lower stability limit. (From Pardiwalla, D., Elliptic Gas Jet Flames in A Cross-Flow, MS thesis, The University of Oklahoma, Norman, OK, 1996.)
burners at R = 0.11. The aspect ratio (diameter of major axis/diameter of minor axis) of the elliptic burner was 3:1. Two burner orientations were considered: major axis oriented with crossflow, referred to as Emaj and minor axis oriented with crossflow, referred to Emin. The Emin flames were found to be attached to the burner for larger jet velocities than the other two burner flames. This widens the range of stable operation of the burner. Figure 29.19 shows the stability regimes of circular and elliptic burner flames. Visible flame length of Emin configuration was also found to be larger than that of the other two burner configurations. Relative emission indicators show that the elliptic burner flames produced lower CO emission than circular burner flames. However, NO emission in elliptic burner flames increased, which was associated with the higher temperature in elliptic flames. Song, Papanikolaou, and Mohamad [12] studied the stability of natural gas flames issued from circular and elliptic burners with different aspect ratios. When the minor axis was aligned with crossflow (Emin configuration of Pardiwalla [85]), the flare stack was able to withstand larger velocities before it blew out, as also observed by Pardiwalla [85]. In Emin configuration, the crossflow acts on a larger frontal area, which leads to the formation of a longer recirculation vortex. As discussed in Section 29.6, at low values of R, the flame usually resides on the edges of the vortex, and hence, the blowout limits are
590
Industrial Combustion Testing
improved with elliptic burner flames. Stability characteristics are further improved with an increase in discharge area. However, the blowout velocities of elliptic burners with major axis aligned with crossflow are close to or lower than in circular burner flames. In industry practice, other than circular burners, flare tips such as slot burner, Coanda or tulip tips (Kaldair® flare systems of John Zink Company), and multiple jet burners are also employed to meet specific flaring needs. Further investigations on elliptic and other noncircular shapes at different values of R need to be conducted in order to understand the applicability of such burners for flare tips. 29.7.2 Effects of Diluents It is common in flare industry to inject diluents in order to suppress smoke liberation. Laboratory investigations have also been conducted to study the influence of diluents on flare performance. Johnson and Kostiuk [86] studied the effect of CO2 dilution on combustion efficiency. An increase in fuel dilution decreases the input flare gas energy density, and in turn, could decrease combustion efficiency. Similar observations were also noted by Gogolek and Hayden [87] in natural gas-propane flares with nitrogen dilution. Smoke point of a turbulent flame is defined as the critical fuel mass flow rate (CFMFR) beyond which the flame does not smoke. Goh [88] studied the effects of nitrogen dilution on the smoke point characteristics of propylene diffusion flames in crossflow. Figure 29.20 shows the variation of diluent mass flow rate with fuel 0.003
N2 mass flow rate (kg/min)
0.0025 0.002
0.0015 0.001
0.0005 0
0
0.002 0.004 0.006 0.008 Critical fuel mass flow rate (kg/min) 2.0 mps
2.5 mps
3.0 mps
3.5 mps
0.01
4.0 mps
Figure 29.20 Variation of N2 flow rate for different fuel flow rates at turbulent smoke point conditions. (From Goh, S.-F., An Experimental and Numerical Study of Diffusion Flames in Cross-Flow and Quiescent Environment at Smoke Point Condition, PhD dissertation, The University of Oklahoma, Norman, OK, 2003.)
© 2011 by Taylor and Francis Group, LLC
mass flow rate at smoke point over a crossflow velocity range of 2–4 m/s. When the fuel mass flow rate is lowered from CFMFR, additional amount of diluent flow was needed to reattain the smoke point condition (i.e., completely suppress smoke liberation). This region was termed as the momentum-controlled region due to the intense flame turbulence and air-fuel mixing caused by the added diluents. However, at about 20–30% below the CFMFR, the diluent mass flow rate decreased with the continued decrease of fuel mass flow rate. The smoke suppression in this region depends on the reaction rate, and hence this region is termed the chemically controlled region. Goh and Gollahalli [89] also studied the effects of different diluents such as argon, helium, hydrogen, and CO2 on turbulent smoke point characteristics of propane diffusion flames in quiescent environment. 29.7.3 Implications for Scaling of Laboratory Data Although flares have been successfully modeled as turbulent diffusion flames in crossflow, the model results must be carefully interpreted while applying them to full-scale flares. In other words, the differences between full-scale and model flares should be kept in mind while scaling up the laboratory data. It is worth a mention again that in full-scale flares the wind speed and direction are unpredictable and the flare gas composition and flow rate vary depending on the type of flaring. In model flares, several strategies are employed to simulate the field conditions. For instance, Kalghatgi [5] and Han and Mungal [9] set the burner at different incident angles with crossflow stream. Johnson and Kostiuk [61] studied the effect of imposed wind turbulence on combustion efficiency. The velocity profile in the wind tunnel could also be modified to match the field conditions. Further, different flaring conditions could be achieved by adjusting jet-to-crossflow momentum flux ratio. During a model study, wind conditions and stack diameter are appropriately scaled down to ensure dynamic similarity. This suffices only the requirements for cold flow conditions. In a burning environment, however, parameters such as fuel pyrolysis time that depends only on fuel chemistry and temperature [66] are also important to be considered. In addition, buoyancy effects are generally neglected in model flares. For all these reasons, the model results must be compared with field test data to validate the correlations developed and develop scaling laws. Due to the unavailability of such data, quantitative scaling laws are yet to be developed. To date, only a few model test results have been compared with field test data. For instance, Schwartz and White [69] compared predictions of radiative emission from various models with field data. Cook et al. [90,91] conducted field-scale
591
Flare Experimental Modeling
experiments on natural gas flares and measured radiation emission, size, and shape of the flame. Based on the field-scale data, an empirical model has also been developed for calculating the radiative emission. Although the differences between the predictions and field test data have been noted, attempts to fine-tune the models and develop universal scaling laws are rare. There is still a need for generic scaling laws to accurately predict the flaring operation from laboratory scale data.
References
29.8 Concluding Remarks This chapter presents the experimental modeling of flares as turbulent diffusion flames in crossflow. We have reviewed the parameters that affect the flare performance in the field. Experimental facilities and instrumentation employed for model studies are presented. A summary of existing data on flame appearance, geometry, radiation, and stability has been included. Data on inflame temperature, velocity, and species concentration fields have also been discussed. Field test data are to be used in conjunction with laboratory model data to validate the results and derive scaling relations.
29.9 Nomenclature dj jet exit diameter g acceleration due to gravity Cf fuel concentration in the jet Kf fuel parameter (Equation 6 of Reference 55) KU constant of proportionality (Equation 7 of Reference 55) Lf flame length LHVmass lower heating value of fuel M momentum flux R jet-to-crossflow momentum flux ratio U velocity X normalized coordinate along the direction of crossflow, x/d j M j /M∞ Z normalized coordinate along the direction of jet, z/d j M j /M∞
Greek Symbol ρ
density
Subscripts j ∞
jet crossflow
© 2011 by Taylor and Francis Group, LLC
1. El Behery, R. E., Mohamad, A. A., and Kamal, M. M. “Combustion Enhancement of a Gas Flare Using Acoustical Excitation.” Combustion Science and Technology 177, no. 9 (2005): 1627–59. 2. Gollahalli, S. R., Brzustowski, T. A., and Sullivan, H. F. “Characteristic of a Turbulent Propane Diffusion Flame in a Cross-Wind.” Transactions of the Canadian Society for Mechanical Engineers 3, no. 4 (1975): 205–14. 3. Bandaru, R. V., and Turns, S. R. “Turbulent Jet Flames in a Crossflow: Effects of Some Jet, Crossflow, and Pilot-Flame Parameters in Emissions.” Combustion and Flame 121 (2000): 137–51. 4. Kalghatgi, G. T. “Blow-Out Stability of Gaseous Jet Diffusion Flames Part II: Effect of Cross Wind.” Combustion Science and Technology 26 (1981): 241–44. 5. Kalghatgi, G. T. “The Visible Shape and Size of a Turbulent Hydrocarbon Jet Diffusion Flame in a Crosswind.” Combustion and Flame 52 (1983): 91–106. 6. Bourguignon, E., Johnson, M. R., and Kostiuk, L. W. “The Use of a Closed-Loop Wind Tunnel for Measuring the Efficiency of Flames in Cross-Flow.” Combustion and Flame 119 (1999): 319–34. 7. Poudenx, P., Howell, L., Wilson, D. J., and Kostiuk, L. W. “Downstream Similarity of Thermal Structure in Plumes from Jet Diffusion Flames in a Crossflow.” Combustion Science and Technology 176, no. 3 (2004): 409–35. 8. Han, D., and Mungal, M. G. “Stabilization in Turbulent Lifted Deflected-Jet Flames.” Proceedings of the Combustion Institute 29, no. 2 (2002): 1889–95. 9. Han, D., and Mungal, M. G. “Simultaneous Measurements of Velocity and CH Distribution. Part II: Deflected Jet Flames.” Combustion and Flame 133, no. 1–2 (2003): 1–17. 10. Brzustowski, T. A. “Flaring in the Energy Industry.” Progress in Energy and Combustion Science 2 (1976): 129–41. 11. Gollahalli, S. R., and Pardiwalla, D. “Comparison of the Flame Characteristics of Turbulent Circular and Elliptic Jets in a Crossflow.” ASME Journal of Energy Resources Technology 124 (2002): 197–203. 12. Song, G. P., Papanikolaou, N., and Mohamad, A. A. “Flame Stability with Elliptical Nozzles in a Crossflow.” Combustion Science and Technology 176, no. 3 (2004): 359–79. 13. Goh, S. F., Kusadomi, S., and Gollahalli, S. R. “Effects of Cross-Wind on Smoke-Point Flow Rate of NitrogenDiluted Hydrocarbon Fuels.” International Power Generation Conference, paper No. JPGC2001/FACT19097, New Orleans, LA, June 4–7, 2001. 14. Goh, S. F., Kusadomi, S., and Gollahalli, S. R. “Effects of Cross-Wind on the Structure of Nitrogen-Diluted Propylene Flame at the Smoke Point Flow Rate.” Sixth Asia Pacific International Symposium on Combustion and Energy Utilization, Kuala Lumpur, Malaysia, May 20–22, 2002. 15. Goh, S. F., Kusadomi, S., and Gollahalli, S. R. “Effects of Jet Dilution and Co-Flow on Sooting and Emission Characteristics of Hydrocarbon Fuels.” International Journal of Energy for a Clean Environment 4, no. 4 (2003): 285–302.
592
16. Huang, R. F., and Chang, J. M. “The Stability and Visualized Flame and Flow Structures of a Combusting Jet in Cross Flow.” Combustion and Flame 98 (1994): 267–78. 17. Johnson, M. R., Majeski, A. J., Wilson, D. J., and Kostiuk, L. W. “The Combustion Efficiency of a Propane Jet Diffusion Flame in Cross Flow.” Presented at the fall meeting of the Western States Section of the Combustion Institute, paper #98F-38, Seattle, WA, October 26–27, 1998. 18. Majeski, A. J., Wilson, D. J., and Kostiuk, L. W. “Size and Trajectory of a Flare in a Cross-Flow.” Presented at Combustion Canada, Calgary, Alberta, May 26–28, 1999. 19. Yoo, Y. D., and Shin, H. D. “Experimental Study of Jet-Diffusion Flames in Cross Air-Flow.” Journal of the Institute of Energy 67, no. 472 (1994): 121–27. 20. Goh, S. F., and Gollahalli, S. R. “Pilot Flame Effects on Gas Jet Flames in Crossflow.” Journal of Propulsion and Power 18 (2002): 1068–75. 21. Gollahalli, S. R., and Nanjundappa, B. “Burner-Wake Stabilized Gas Jet Flames in a Crossflow.” Combustion Science and Technology 109 (1995): 327–46. 22. Huang, R. F., and Yang, M. J. “Thermal and Concentration Fields of Burner-Attached Jet Flames in Cross Flow.” Combustion and Flame 105 (1996): 211–24. 23. Fristrom, R. M., and Westernberg, A. A. Flame Structure. New York: McGraw-Hill Publications, 1965. 24. Perry, A. E. Hot-Wire Anemometry. New York: Oxford University Press, 1982. 25. Lomas, C. G. Fundamentals of Hot Wire Anemometry. New York: Cambridge University Press, 1986. 26. Brunn, H. H. Hot Wire Anemometry: Principles and Signal Analysis. New York: Oxford University Press, 1995. 27. Holman, J. P. Experimental Methods for Engineers. 6th ed. New York: McGraw-Hill, 1994. 28. Goldstein, R. J. Fluid Mechanics Measurements. 2nd ed. Washington, DC: Taylor & Francis, 1996. 29 Miller, R. W. Flow Measurement Engineering Handbook. 3rd ed. New York: McGraw-Hill, 1996. 30. Albrecht, H.-E., Damaschke, N., Borys, M., and Tropea, C. Laser Doppler and Phase Doppler Measurement Techniques. New York: Springer, 2003. 31. Bernard, P. S., and Wallace, J. M. Turbulent Flow: Analysis, Measurement, and Prediction. Hoboken, NJ: John Wiley & Sons, 2002. 32. Hanson, R. K. “Combustion Diagnostics: Planar Flowfield Imaging.” Proceedings of 21st Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 1677–91, 1986. 33. Hasselbrink, E. F., and Mungal, M. G. “Transverse Jets and Jet Flames. Part 2. Velocity and OH Field Imaging.” Journal of Fluid Mechanics 443 (2001): 27–68. 34. Goh, S. F., and Gollahalli, S. R. “Sooting of PropyleneHydrogen Mixture Diffusion Flames with Diluents Near Smoke Point.” Journal of Propulsion and Power 21, no. 4 (2005): 600–605. 35. Yagi, S., and Iino, H. “Radiation from Soot Particles in Luminous Flames.” 8th Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 288–93, 1960.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
36. Melton, L. A. “Soot Diagnostic Based on Laser Heating.” Applied Optics 23, no. 13 (1984): 2201–8. 37. Quay, B., Lee, T.-W., Ni, T., and Santoro, R. J. “Spatially Resolved Measurement of Soot Volume Fraction Using Laser-Induced Incandescence.” Combustion and Flame 97 (1994): 384–92. 38. Ni, T., Pinson, J. A., Gupta, S., and Santoro, R. J. “TwoDimensional Imaging of Soot Volume Fraction by Use of Laser Induced Incandescence.” Applied Optics 34, no. 30 (1995): 7083–91. 39. Will, S., Schraml, S., and Leipertz, A. “Comprehensive Two-Dimensional Soot Diagnostics Based on LaserInduced Incandescence (LII).” 26th Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 2277–84, 1996. 40. Shaddix, C. R., and Smyth, K. C. “Laser-Induced Incandescence Measurement of Soot Production in Steady and Flickering Methane, Propane, and Ethylene Diffusion Flames.” Combustion and Flame 107 (1996): 418–52. 41. Mewes, B., and Seitzman, J. M. “Soot Volume Fraction and Particle Size Measurement with Laser Induced Incandescence.” Applied Optics 33, no. 3 (1997): 709–17. 42. Bryce, D. J., Ladommatos, N., and Zhao, H. “Quantitative Investigation of Soot Distribution by Laser-Induced Incandescence.” Applied Optics 39, no. 27 (2000): 5012–22. 43. Johnson, M. R., and Kostiuk, L. W. “Efficiencies of Low Momentum Jet Diffusion Flames in Crosswinds.” Combustion and Flame 123 (2000): 189–200. 44. Huang, R. F., and Chang, J. M. “Coherent Structure in a Combusting Jet in Crossflow.” AIAA Journal 32, no. 6 (1994): 1120–25. 45. Huang, R. F., and Wang, S. M. “Characteristic Flow Models of Wake-Stabilized Jet Flames in a Transverse Air Stream.” Combustion and Flame 117 (1999): 59–77. 46. Escudier, M. P. “Aerodynamics of a Burning Turbulent Gas Jet in a Crossflow.” Combustion Science and Technology 4 (1972): 293–301. 47. Brzustowski, T. A. “Hydrocarbon Turbulent Diffusion Flame in Subsonic Crossflow.” 15th AIAA Aerospace Sciences Meeting, paper 77-22, Reston, VA, 1977. 48. Brzustowski, T. A. “A Model for Predicting the Shapes and Lengths of Turbulent Diffusion Flames Over Elevated Industrial Flares.” Paper presented at the 22nd Canadian Chemical Engineering Conference, 1972. 49. Brzustowski, T. A., Gollahalli, S. R., and Sullivan, H. F. “The Turbulent Hydrogen Diffusion Flame in a Cross Wind.” Combustion Science and Technology 11 (1975): 29–33. 50. Brzustowski, T. A. “Turbulent Diffusion Flame Models III: The Buoyant Flame in a Cross-Wind.” Proceedings of 5th Canadian Congress of Applied Mechanics, University of New Brunswick, Fredericton, 1975. 51. Rao, V. K., and Brzustowski, T. A. “Tracer Studies of Jets and Diffusion Flames in Cross-Flow.” Combustion Science and Technology 27 (1982): 229–39. 52. Majeski, A. J., Wilson, D. J., and Kostiuk, L. W. “Local Maximum Flame Length of Flares in a Crosswind.” Presented at The Combustion Institute, Canadian Section, 1999 Spring Technical Meeting, Edmonton, Alberta.
Flare Experimental Modeling
53. Poudenx, P., and Kostiuk, L. W. “An Investigation of the Mean Plume Structures of a Flare in a Crosswind.” Presented at The Combustion Institute, Canadian Section, 1999 Spring Technical Meeting, Edmonton, Alberta. 54. Majeski, A. J., Wilson, D. J., and Kostiuk, L. W. “Measuring and Predicting the Length of a Propane Jet Diffusion Flame.” Presented at the Combustion Institute, Canadian Section, 2000 Spring Technical Meeting, Ottawa, Ontario. 55. Majeski, A. J., Wilson, D. J., and Kostiuk, L. W. “Predicting the Length of Low-Momentum Jet Diffusion Flame in Crossflow.” Combustion Science and Technology 176, no. 12 (2001): 2001–25. 56. Karagozian, A. R. “The Flame Structure and Vorticity Generated by a Chemically Reacting Transverse Jet.” AIAA Journal 24, no. 9 (1986): 1502–7. 57. Pratte, B. D., and Baines, W. D. “Profiles of Round Turbulent Jet in Cross Flow.” Proceedings of American Society of Civil Engineers, Journal of the Hydraulics Division 93, HY6 (1967): 53–64. 58. Choudhuri, A. R., and Gollahalli, S. R. “Pressure Effects on Geometry and Structure of Gas Jet Flames in Crossflow.” Journal of Propulsion and Power 17, no. 1 (2001): 190–96. 59. Keefer, J. F., and Baines, W. D. “Round Turbulent Jet in Cross-Wind.” Journal of Fluid Mechanics 15, no. 4 (1963): 481–96. 60. Pohl, J. H., Lee, J., Payne, R., and Tichenor, B. A. “Combustion Efficiency of Flares.” Combustion Science and Technology 50 (1986): 217–31. 61. Johnson, M. R., and Kostiuk, L. W. “A Parametric Model for the Efficiency of a Flare in Crosswind.” Proceedings of the Combustion Institute 29, no. 2 (2002): 1943–50. 62. Kostiuk, L. W., Majeski, A. J., Poudenx, P., Johnson, M. R., and Wilson, D. J. “Scaling of Wake-Stabilized Jet Diffusion Flames in a Transverse Air Stream.” Proceedings of the Combustion Institute 28 (2000): 553–59. 63. Strosher, M. T. “Characterization of Emissions from Diffusion Flare Systems.” Journal of the Air and Waste Management Association 50, no. 10 (2000): 1723–33. 64. Leahey, D. M., Preston, K., and Strosher, M. T. “Theoretical and Observational Assessments of Flare Efficiencies.” Journal of the Air and Waste Management Association 51, no. 12 (2001): 1610–16. 65. Brzustowski, T. A., and Sommer, E. C., Jr. “Predicting Radiant Heating from Flares.” A.P.I. Preprint, Nos. 64–73, also Proceedings of Division of Refining A.P.I. 53 (1973): 865–93. 66. Brzustowski, T. A., Gollahalli, S. R., Gupta, M. P., Kaptein, M., and Sullivan, H. F. Radiant Heating from Flares ASME paper 75-HT-4, 1975. 67. De Faveri, D. M., Fumarola, G., Zonato, C., and Ferraiolo, G. “Estimate Flare Radiation Intensity.” Hydrocarbon Processing 64, no. 5 (1985): 89–91. 68. Schmitt, D. A. “Radiation from a Hydrogen Flare in a Crosswind,” AIAA Paper, 23rd Aerospace Sciences Meeting, Reno, NV, 14–17 January, 1985. 69. Schwartz, R. E., and White, J. W. “Flare Radiation Prediction: A Critical Review.” Proceedings of the 30th Annual Loss Prevention Symposium of the American Institute of Chemical Engineers. paper #12a, 1996.
© 2011 by Taylor and Francis Group, LLC
593
70. Fairweather, M., Jones, W. P., and Lindstedt, R. P. “Predictions of Radiative Transfer from a Turbulent Reacting Jet in a Cross-Wind.” Combustion Science and Technology 89 (1992): 45–63. 71. Tsue, M., Kadota, T., and Kono, M. “Detailed Measure ments of the Structure of a Jet Diffusion Flame in a Cross Flow.” Symposium (International) on Combustion 28, no. 1 (2000): 295–301. 72. Tsue, M., Kadota, T., and Kono, M. “Temperature and Velocity Fluctuations of a Jet Diffusion Flame in a CrossFlow.” Proceedings of the Combustion Institute 29, no. 2 (2002): 1937–42. 73. Fairweather, M., Jones, W. P., and Marquis, A. J. “Predictions of the Concentration Field of a Turbulent Jet in a Cross-Flow.” Combustion Science and Technology 62 (1988): 61. 74. Birch, A., Brown, D. R., Fairweather, M., and Hargrave, G. K. “An Experimental Study of Turbulent Natural Gas Jet in a Cross-Flow.” Combustion Science and Technology 66 (1989): 217–32. 75. Askari, A., Bullman, S. J., Fairweather, M., and Swaffield, F. “The Concentration Field of a Turbulent Jet in a CrossWind.” Combustion Science and Technology 73 (1990): 463–78. 76. Fairweather, M., Jones, W. P., Linstedt, R. P., and Marquis, A. J. “Predictions of a Turbulent Reaching Jet in a Crossflow.” Combustion and Flame 84, no. 3–4 (1991): 361–75. 77. Botros, P. E., and Brzustowski, T. A. “An Experimental and Theoretical Study of the Turbulent Diffusion Flame in Crossflow.” 17th Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA, 389–97, 1978. 78. Huang, R. F., Savas, O., and Gollahalli, S. R. “Flow Field in the Near-Burner Region of a Partially Lifted Turbulent Gas Jet Flame in a Crossflow.” ASME HTD-223 (1992): 105–10. 79. Huang, R. F., Savas, O., and Gollahalli, S. R. “Turbulence Characteristics in the Flow Field of a Nonpremixed Gas Jet Flame in a Crossflow.” ASME PD-66 (1995): 11–20, 1995. 80. Savas, O., Huang, R. F., and Gollahalli, S. R. “Structure of the Flow Field of a Nonpremixed Gas Jet Flame in a Crossflow.” ASME Journal of Energy Resources Technology 119 (1997): 137–44. 81. Hasselbrink, E. F., Mungal, M. G., and Hanson, R. K. “Simultaneous Planar Velocity Measurements and OH Imaging in a Transverse Jet Flame.” Journal of Visualization 1, no. 1 (1998): 65–77. 82. Ho, C. M., and Gutmark, E. “Vortex Induction and Mass Entrainment in a Small-Aspect-Ratio Elliptic Jet.” Journal of Fluid Mechanics 179 (1987): 665–72. 83. Gutmark, E. J., and Grinstein, F. F. “Flow Control with Noncircular Jets.” Annual Review of Fluid Mechanics 31 (1999): 239–72. 84. Gollahalli, S. R., Khanna, T., and Prabhu, N. “Diffusion Flames of Gas Jets Issued From Circular and Elliptic Nozzles.” Combustion Science and Technology 86 (1992): 267–88. 85. Pardiwalla, D. “Elliptic Gas Jet Flames in A Cross-Flow.” MS thesis, The University of Oklahoma, Norman, OK, 1996.
594
86. Johnson, M. R., and Kostiuk, L. W. “Effects of a Fuel Diluent on the Efficiencies of Jet Diffusion Flames in a Crosswind.” Combustion Institute Canada Section, 1999 Spring Technical Meeting, Edmonton, Alberta, 1999. 87. Gogolek, P. E. G., and Hayden, A. C. S. “Performance of Flare Flames in a Crosswind with Nitrogen Dilution.” Journal of Canadian Petroleum Technology 43, no. 8 (2004): 43–47. 88. Goh, S.-F. “An Experimental and Numerical Study of Diffusion Flames in Cross-Flow and Quiescent Environment at Smoke Point Condition.” PhD diss., The University of Oklahoma, Norman, OK, 2003.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
89. Goh, S.-F., and Gollahalli, S. R. “Mechanism of the Effect of Dilution with Different Inert Gases on Smoke Point of Propylene Diffusion Flames.” Proceedings of ASME Power 2004, Baltimore, MD, March 30–April 1, 2004. 90. Cook, D. K., Fairweather, M., Hammonds, J., and Hughes, D. J. “Size and Radiative Characteristics of Natural Gas Flares. Part 1: Field Scale Experiments.” Chemical Engineering Research and Design 65, no. 4 (1987): 310–17. 91. Cook, D. K., Fairweather, M., Hammonds, J., and Hughes, D. J. “Size and Radiative Characteristics of Natural Gas Flares. Part 1: Empirical Model.” Chemical Engineering Research and Design 65, no. 4 (1987): 318–25.
30 Flare Radiation Wes Bussman and Jianhui Hong Contents 30.1 Introduction.................................................................................................................................................................. 595 30.2 Properties and Characteristics of Radiation............................................................................................................ 596 30.2.1 Electromagnetic Spectrum........................................................................................................................... 596 30.2.2 Units of Flare Radiation................................................................................................................................ 596 30.2.3 Solar Radiation Level.................................................................................................................................... 597 30.2.4 Radiation Level as a Function of Distance from Source.......................................................................... 598 30.2.5 Flare Radiation Spectrum............................................................................................................................. 599 30.2.6 Radiation Transmission Losses through the Atmosphere...................................................................... 599 30.3 Environmental Concerns............................................................................................................................................ 600 30.3.1 Radiation Effects on Equipment.................................................................................................................. 600 30.3.2 Ignition of Flammable Material from Radiant Heating........................................................................... 600 30.3.3 Radiation Effects on Humans...................................................................................................................... 601 30.3.4 API 521 Recommendations........................................................................................................................... 601 30.4 Estimating Flare Radiation......................................................................................................................................... 603 30.5 Measuring Flare Radiation......................................................................................................................................... 606 30.5.1 Description of the Radiometer..................................................................................................................... 606 30.5.2 Useful Tips when Measuring Flare Radiation........................................................................................... 607 30.5.2.1 Radiometer Selection................................................................................................................... 607 30.5.2.2 Sun Interference............................................................................................................................ 608 30.5.2.3 View Angle.................................................................................................................................... 608 30.5.2.4 Convective Cooling Effects......................................................................................................... 608 30.5.2.5 Measuring Technique.................................................................................................................. 609 30.5.3 Radiometer Cube............................................................................................................................................610 References..................................................................................................................................................................................610
30.1 Introduction The hydrocarbon and petrochemical industries use flare systems to dispose of waste gases in a safe, effective manner [1]. During an emergency flaring situation, today’s plants might burn more than a million pounds per hour of gas producing an open flame over several hundred feet long [2]. A portion of the heat produced by the flame will radiate to its immediate surroundings. In many cases, heat radiated from the flare flame during emergency flaring, forms the basis for determining how tall to design the flare stack, where to locate it, and what area surrounding the flare is to have limited access to personnel and equipment. One of the greatest challenges in designing a flare system is predicting the amount of heat radiated from the flame. It is extremely important to accurately predict
thermal radiation. If thermal radiation levels are underestimated, it will result in a flare stack that is too short, which could expose personnel and equipment to potentially dangerous levels of heat. Estimating flare radiation from a pure theoretical approach is currently not feasible due to the complexity of the problem. To overcome this problem, engineers have historically estimated radiation using semiempirical models based on experimental data. Although each flare vendor uses their own proprietary models to estimate flare radiation, each model must rely heavily on experimental data. Therefore, it is extremely important that radiation data gathered from flare testing is accurate. In this chapter we will first discuss the properties and characteristics of thermal radiation. Next, we will discuss the effects of thermal radiation on equipment, structures and personnel. This will be followed by a discussion on 595
© 2011 by Taylor and Francis Group, LLC
596
Industrial Combustion Testing
methods used in the flare industry to estimate radiation from a flare flame. Finally, we will discuss how flare radiation is typically measured in the industry.
30.2 Properties and Characteristics of Radiation 30.2.1 Electromagnetic Spectrum Everyone has experienced the effect of heat transfer by thermal radiation; for example, the warmth of sunshine, or the heat one feels when standing next to a campfire. Unlike conduction or convection, heat transfer by thermal radiation does not require a medium to transfer heat. Thermal radiation is transferred through space by electromagnetic wave phenomena that travel at the speed of light. Radiation consists of a wide range of wavelengths as illustrated in Figure 30.1. Each wavelength corresponds to a frequency and an energy level: the shorter the wavelength of radiation, the higher the frequency and greater the energy. Very short wave lengths fall into a category of gamma rays or x-ray. Radiation with longer wavelengths, such as TV and radio waves, falls on the other end of the spectrum and has much less energy. Thermal radiation falls between these two ends and is defined as the portion of the spectrum between 1 × 10−7 m and 1 × 10−4 m wavelengths [3] that is emitted from solid, liquids, and gases by way of the temperature above absolute zero [4]. The human eye is not able to see all of the energy in the thermal radiation spectrum; the range that the human eye can detect is referred to as the visible spectrum and ranges in wavelength from about 0.4 × 10−6 m to 0.7 × 10−6 m.
30.2.2 Units of Flare Radiation Figure 30.2 shows a flare burning propane at a rate of 60,000 lb/hr corresponding to a total heat release of about 1.2 × 109 Btu/hr. Experimental data collected from this test reveals that approximately 25% of the total heat produced by this flame is radiated to its immediate surroundings; this is referred to as flare radiation. The amount of flare radiation received at a given location (or point of interest) is referred to as the flare radiation level. The radiation level is defined as the amount of heat passing through a given area in a given amount of time. For example, suppose an observer stands at a distance from a flame and holds a 1 ft × 1 ft square frame facing in the direction normal to the radiation incidence as illustrated in Figure 30.3. Assume the observer measures the amount of heat passing through the frame and finds that in a time period of one hour 1000 Btu’s of energy passes through. What the observer has measured is a radiation level of 1000 Btu/hr-ft2. In the flare industry, radiation levels are commonly written in units
Flare flame
Sun Visible - 47% Infrared - 46% UV8% 0.3 to 3 (× 10–6 m) UV
The sun emits thermal radiation with wavelengths primarily in the range from 0.3 × 10−6 m to 3 × 10−6 m. In general, about 47% of the radiation from the sun (at the surface of the earth) falls in the visible spectrum, 46% falls in the infrared spectrum and about 8% in the ultra violet spectrum [5]. The radiation emitted from a flare flame ranges from about 0.3 × 10−6 m to 30 × 10−6 m. Although flare radiation covers a wide spectrum, the human eye cannot see about 99.9% of the radiation emitted because almost all of the energy falls in the infrared region, but we can definitely feel the radiation as heat. Only about 0.1% of the radiation emitted from a flare flame is in the visible spectrum; this small percentage allows us to see the flame.
Propane at 60, 000 lb/hr
Visible - 0.1% Infrared - 99.9% UV 0% 0.3 to 30 (× 10–6 m)
Visible spectrum 0.4 to 0.7 (× 10–6 m)
Infrared
Thermal radiation 1× 10–7 m to 1× 10–4 m
Gamma rays 10–12
X-rays UV
V i s i b l e
Infrared
Microwaves
10–10 10–8 10–6 10–4 10–3 Wavelength, m
Figure 30.1 Electromagnetic spectrum.
© 2011 by Taylor and Francis Group, LLC
TV and radio waves 1
102
Heat radiating from flame to surroundings (referred to as flare radiation)
Figure 30.2 (See color insert following page 424.) Flare firing propane at 60,000 lb/hr corresponding to a heat release rate equal to 1.2 billion Btu/hr.
597
Radiation level - Amount of heat passing through an area in a given amount of time Energy Radiation level = Time–area •
Btu hr–ft2
Joules s–m2
Btu = 3.154 • 1 hr–ft2
Watts
Heat flux Watts m2
m2
Solar radiation (Btu/hr/ft2)
Flare Radiation
350 300 250 200 Normal to solar beam
150 100
Horizontal to the earth surface J
F
M
A
M
J J A S Month of year
O
N
D
J
Object horizontal to the earth surface Object normal to solar beam
Figure 30.3 Illustration defining radiation level.
of Btu/hr-ft2 or Watts/m2. The conversion from one Btu/ hr-ft2 to Watt/m2 is 3.154. For example, a radiation level of 1000 Btu/hr-ft2 corresponds to 3154 Watt/m2.
Sun
Earth
Figure 30.4 Solar radiation level at angles normal to the solar beam and horizontal to the surface of the earth in Tulsa, OK.
Tulsa
30.2.3 Solar Radiation Level When personnel and equipment are exposed to flare radiation, chances are they are also being exposed to the radiation from the sun: referred to as solar radiation. Flare radiation and solar radiation levels can be cumulative. For example, suppose the flare radiation level at a point of interest in 500 Btu/hr-ft2 on a cloudy day. On a clear day, however, when the sun is in full view, the radiation level could increase to over 800 Btu/hr-ft2. Therefore, when estimating flare radiation, it is important that design engineers take into account the contribution from solar radiation because it can impact flare boom lengths and stack heights. Solar radiation should also be taken into account when measuring flare radiation levels. If instrumentation is pointed directly into the sun during data collection, then solar radiation might have a significant contribution to the overall flare radiation level. The purpose of this section is to discuss the solar radiation levels. The amount of thermal radiation we receive from the sun varies with the time of year and depends largely on the path length in the earth’s atmosphere that the sun’s energy must pass through [6]. For example, Figure 30.4 shows radiation levels throughout the year, at high noon, on a clear day in Tulsa, Oklahoma. These levels are based on the solar radiation received at the earth’s surface at angles normal to the solar beam and horizontal to the surface of the earth using a model developed by Bird [7]. In Tulsa, the radiation level received on an object, positioned normal to the solar beam, varies from about
© 2011 by Taylor and Francis Group, LLC
Tulsa
View from sun in January
View from sun in July
Tulsa during winter months θ
Angle of incidence (Solar zenith angle) θ Tulsa during summer months
Earth
Sun
Figure 30.5 View of Tulsa, OK from the sun during the months of January and July [ref].
290 to 305 Btu/hr-ft2 (915 to 946 W/m2). Notice, in the winter and fall, (September through February), the normal radiation is lower than in the spring and summer (March through August). This trend occurs because during the fall and winter months the solar radiation passes through a greater thickness of the earth’s atmospheric air than in the spring and summer months; this can be better explained by referring to Figure 30.5. Figure 30.5 shows the view of Tulsa, relative from the sun, during the months of January and July. Notice in
598
Industrial Combustion Testing
I (θ) = Iθ=0 cos θ,
(30.1)
where θ is the angle of incidence, Iθ = 0 is the radiant flux with an angle of incidence of zero and I(θ) is the radiant flux at a given angle of incidence. 30.2.4 Radiation Level as a Function of Distance from Source As previously discussed, the distance from the earth to the sun can have an effect on the amount of solar energy reaching the earth. Similarly, the length of a flare boom or stack height can have a significant impact on the flare radiation level at a point of interest. To gain better insight into how much the thermal radiation level varies with the distance from a flame, consider an observer located near the base of a flare stack as illustrated in Figure 30.6. To simplify this analysis, assume that all of the heat radiated from the flame is emitted from a single point located near the center of the flame; this point is referred to as the flame epicenter. Let’s further assume that the heat emitted from the epicenter radiates outwardly in all directions and that the heat is evenly distributed over the surface of a sphere. For Case 1, the observer is located at a distance, r, from the epicenter while in Case 2 the observer is located at twice the distance, 2r. The surface area (A) of each sphere can be written as A1 = 4πr2 and A2 = 4π(2r)2 where subscript 1 and 2 represent Case 1 and Case 2, respectively. Taking the ratio of A2 to A1 results
© 2011 by Taylor and Francis Group, LLC
Flame epicenter
Area2 = 4 × Area1
Flame epicenter
Area1 r
2r
Observer
Case 1
Observer
Case 2
Figure 30.6 Effects of doubling the distance of the flame epicenter from an observer. 1.0 Radiation level relative to observer
July, Tulsa faces almost directly toward the sun, however, in January, Tulsa is at a much greater angle relative to the sun; this angle is referred to as the angle of incidence. During the summer months, Tulsa is positioned with an angle of incidence of approximately 15° while in the winter months approximately 75°. A larger angle of incidence requires the solar radiation to pass through a greater thickness of air resulting in more radiation transmission losses. It should be mentioned that the earth’s distance from the sun during its yearly elliptical orbit also affects the amount of radiation the earth receives; this variation in distance changes by about 3.4% resulting in a 6.9% change in the amount of solar energy reaching the earth [8]. Figure 30.4 also shows the radiation level in Tulsa for an object positioned horizontal to the earth’s surface. For this case, the radiation varies much more dramatically: from about 150 to 290 Btu/hr-ft2 (473 to 915 W/m2) throughout the year. Peak radiation levels occur during the summer months mainly because the view allows more solar radiation to strike an area positioned horizontal to the earth. The amount of radiation that strikes an area positioned horizontal to the earth can be described by the Lambert cosine law:
0.9 0.8 0.7
r
0.6
Observer
0.5 0.4
Doubling the distance reduces the radiation level by a factor of four
0.3 0.2 0.1 0.0
1r
1.5r
3r 2.5r 2r Relative distance from observer
3.5r
4r
Figure 30.7 Relative radiation level as a function of distance from the source.
in A2/A1 = 4; this result shows that doubling the distance from the flame epicenter yields a fourfold increase in surface area. Since the heat is evenly distributed over each sphere, one can conclude that the observer will experience a radiation level four times greater for Case 1 as compared to Case 2. Or in other words, doubling the distance from the radiation source reduces the radiation level by a factor of four. Since the surface area of the sphere increases by a factor of r2, then theoretically the radiation level should fall off by a factor of 1/r2, where r is the distance from the radiation source or epicenter. Figure 30.7 is a plot showing how the radiation level falls off relative to an observer located at a distance r from the flame epicenter. Again, notice that as the observer moves out to a distance 2r from the epicenter, the radiation falls off by a factor of four. If the observer further increases his distance from the epicenter by 4r notice that the radiation level falls off by a factor of 16. This example demonstrates
599
Flare Radiation
that sizing the flare boom lengths and stack heights is critical because relatively small changes in distance can have a significant impact on the radiation levels at a point of interest. 30.2.5 Flare Radiation Spectrum
0
IR spectrum
Visible spectrum
UV spectrum
Pulverized coal flame Liquid fuel flame
1
2 3 Wavelength (1 x 10–6 m)
As radiation passes through the earth’s atmosphere, some of the radiation is attenuated. The amount of atmospheric attenuation depends largely on the composition of the air and the spectral characteristics of the radiation. Atmospheric air consists of a mixture of several gases, water vapor, numerous pollutants, and small particles. Table 30.1 gives the approximate composition of dry air by volume fraction [10]. Some of the gases have concentrations that are constant while others are variable. Nitrogen, oxygen, and argon account for about 99.99% (by volume) of the permanent gases in the atmosphere [11]. The concentration of carbon dioxide can be somewhat variable but is typically around 0.03%. Water vapor in the atmosphere is also variable but typically varies between 0% and 4%. Some wavelengths in the infrared, TV, and radio spectrum will pass through the air unabsorbed; the wavelengths that pass through the air unabsorbed are referred to as atmospheric windows. Figure 30.9 shows a generalized diagram of atmospheric radiation transmission through the air in the region of interest for flare radiation. Notice that the atmospheric windows occur at distinct wavelengths. Water vapor and carbon dioxide is largely responsible for the radiation absorption in these bands with water vapor dominant. Water vapor in the atmosphere absorbs thermal radiation at wavelengths similar to where water vapor in the products of combustion emits radiation [12]. For nonluminous flames, where radiation from hot H2O contributes a significant portion of the total emitted energy, absorption by water vapor in
Approximate Composition of Dry Air
4
Constituent
5
Figure 30.8 (See color insert following page 424.) Spectral emission of radiation from luminous and nonluminous flames. (From Baukal, C. E., The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001.)
© 2011 by Taylor and Francis Group, LLC
30.2.6 Radiation Transmission Losses through the Atmosphere
Table 30.1 Non-luminous flame
Violet Yellow Red
Relative radiation intensity
When a flare burns a waste gas, the stable products of combustion typically consist of soot particles and a variety of gases such as carbon dioxide (CO2), water vapor (H2O), carbon monoxide (CO), hydrogen (H2), oxygen (O2), nitrogen (N2), and nitric oxide (NO). The principle contributors of thermal radiation come from two sources: (1) hot CO2 and H2O vapor, and (2) solid particles of soot. Both H2O and CO2 emit radiation in the infrared region and do not contribute to the visible light emitted from the flame; this is referred to as nonluminous radiation. The H2O and CO2 emit radiation at distinct band wavelengths: H2O at 2.7, 6.3, and 20 × 10−6 m and CO2 at 2.7, 4.3, and 15 × 10−6 m. Radiation levels emitted from nonluminous gases depends on the thickness of the gas (optical path) and temperature of the gas; the hotter the gas and thicker the optical path, the higher the radiation. Figure 30.8 shows the spectral emission of radiation from a nonluminous flame of a jet engine combustor. Notice that radiation falls all within the infrared spectrum at distinct band wavelengths. Figure 30.8 also shows the radiation spectrum emitted from flames containing soot particles (coal and liquid flame); these types of flames are referred to as luminous flames. The presence of soot particles in a flame can dominate the thermal radiation. A flame containing a significant amount of soot can radiate three to four times as much radiation as a nonluminous flame [1,9]. The radiation
emitted from soot particles is unlike gaseous radiation with distinct bands; instead, the radiation is continuous over a wide range of wavelengths. Notice in Figure 30.8 that the radiation from the luminous flame falls within the visible spectrum; this is what gives flames a yelloworange color.
Nitrogen Oxygen Argon Carbon dioxide Neon, helium, methane, sulfur dioxide, hydrogen, and other minor gases
Volume Fraction 0.78084 0.20948 0.00934 0.00031 0.00003
Source: From McQuiston, C., and Parker, J. D., Heating, Ventilating, and Air Conditioning, New York: John Wiley & Sons, 1982.
600
Industrial Combustion Testing
Region of flare radiation UV spectrum
50
Infrared spectrum
Visible spectrum
Transparency (%)
100
humidity greater than 10%, and distance from the flame between 100 and 500 feet; however, it can be used to estimate the order of magnitude of τ under a wider range of conditions. Notice for these conditions, it is estimated that approximately 15% to 25% of the radiation can be attenuated over a distance of 500 feet.
0.3 to 30 (×10–6 m)
O2O3
30.3 Environmental Concerns 0 0.2
0.5
1 5 10 20 Wavelength (1 × 10–6 m)
100
Figure 30.9 Generalized diagram showing relative atmospheric radiation transmission of different wavelengths. (From Brzustowski, T. A., and Sommer, E. C., Jr., Predicting Radiant Heating from Flares, Proceedings of the Division of Refining, API 1973, Vol. 53, API Washington, DC, pp. 865–93.)
Fraction of radiation transmitted through atmosphere (τ)
0.95
τ = 0.79(100/RH)(1/16) × (100/D)(1/16)
0.90 0.85
Relative humidity 10% 20% 40% 80%
0.80 0.75
Air temperature = 80°F Flame temperature = 2240°F
0.70 100
150
200 250 300 350 400 450 Distance from flame epicenter, D (feet)
500
Figure 30.10 Estimates of the fraction of flame radiation transmitted through the atmosphere at various distances and percentage of relative humidity. (From Brzustowski, T. A., and Sommer, E. C., Jr., Predicting Radiant Heating from Flares, Proceedings of the Division of Refining, API 1973, Vol. 53, API Washington, DC, pp. 865–93.)
the atmosphere has a significantly greater attenuating effect than for a luminous flame. The amount of atmospheric absorption depends largely on the relative humidity and distance from the point of interest to the flame as demonstrated in Figure 30.10 [12]. Brzustowski and Sommer state that these curves are strictly applicable only under the following conditions: luminous hydrocarbon flame radiating at 2240ºF, 80ºF dry bulb ambient temperature, relative
© 2011 by Taylor and Francis Group, LLC
Excessive thermal radiation from flares can cause equipment and structural damage, ignite flammable material, and pose a safety risk to personnel. The purpose of this section is to discuss the importance of exposure to high levels of flare radiation. 30.3.1 Radiation Effects on Equipment When equipment is exposed to radiant heating, the outer surface of the equipment heats up; the heat can then be transferred to the internal components by convection, conduction, and radiation. If enough heat is transferred, it could melt plastic parts such as valve seals, wiring insulation, and gauges. Many plastics will melt at a radiation level of about 12 kW/m2 (3805 Btu/hr-ft2) while electrical cable insulation will start to degrade at 18–20 kW/m2 (5707–6341 Btu/hr-ft2) [13]. High heat can also dry out lubricated parts such as motors or valves. The amount of damage incurred can vary dramatically from equipment to equipment and depends largely on the design of the equipment, time period of exposure, and local ambient conditions. Due to the multiple parameters involved in the heat transfer of equipment under radiant heating, it is difficult to recommend threshold levels. It is common practice to position equipment where it cannot be in direct view of the flare flame or provide adequate protection to shield it from the radiation. 30.3.2 Ignition of Flammable Material from Radiant Heating Flammable objects near a flare, such as wood objects and vegetation can ignite if exposed to high levels of thermal radiation. For ignition to occur, the material must be heated to a temperature high enough to vaporize the solid fuel within the material. The vaporized gas must then mix with the surrounding air to form a flammable air–fuel mixture. If an ignition source, such as a small flame or a spark is present, the air–fuel mixture can ignite; this is referred to as piloted ignition. If the airfuel mixture is ignited without the presence of a piloted source it is referred to as auto-ignition or spontaneous
601
Flare Radiation
ignition. Piloted ignition requires a much lower temperature than auto-ignition. The surface temperature of the material at which ignition occurs is referred to as the ignition temperature. Babrauskas [14] gives a good summary of the data reported in the literature regarding ignition temperatures of wood for piloted and auto-ignition under radiant heating. Results from numerous experimental studies show that wood will ignite over a wide range of temperatures under radiant heating. Babrauskas states that the large variation in ignition temperature is due to several reasons: definition of ignition, design of test apparatus and operating conditions, specimen conditions (size, moisture, and orientation) and species of wood. Based on a summary of the data reported in the literature, Babrauskas concludes that 250ºC (482ºF) is the best estimate of the ignition temperature of wood regardless if it is piloted or auto-ignition; this value is a suitable limit for design or hazard analysis purposes [15]. In 1965, McGuire [16] suggested that the minimum radiation level for piloted ignition of most wood material is 12.5 kW/m2 (3963 Btu/hr-ft2). At these radiation levels, the time required for ignition is about 10–20 minutes. These test results are based on radiant heat oriented along the grain of the wood (along-grain exposure). Spearpoint [17], however, found that if the wood is exposed to radiation on the end-grain, piloted ignition occurs at lower radiation levels. For end-grain exposure, Spearpoint found piloted ignition for maple occurs at 8 kW/m2 (2536 Btu/hr-ft2) in about 45 minutes; for along-grain exposure, maple will ignite at 12 kW/m2 (3805 Btu/hr-ft2) in about 70 minutes. The time required for piloted ignition of wood depends on several factors such as size and shape of the piece of wood, rate of heat loss from the surface, type of wood, orientation, and radiation level. The auto-ignition of wood has not been studied as extensively as piloted ignition. Test results indicate that wood will auto-ignite at about 4.3 kW/m2 (1363 Btu/hr-ft2) if exposed for hours, rather than minutes. For short-term exposures, a radiation level of 20 kW/ m2 (6341 Btu/hr-ft2) is typical for the auto-ignition of wood [14]. Vegetation in the area of a flare should also be considered such as grass and trees. A radiation level of 25 kW (7926 Btu/hr-ft2) is capable of auto-ignition of trees if the exposure duration is long enough [18].
Living human skin absorbs heat radiated from a fire nearly as a blackbody [19]. When exposed to high levels of thermal radiation on the bare skin, one will first experience pain: a hot, tingling sensation felt when the skin temperature rises to just above 111ºF at a depth of over 0.1 mm. If the temperature of the skin remains above 111ºF pain and injury will occur. The burn injury for human skin is categorized as first-degree, seconddegree, and third-degree burns. If one is exposed to a radiation level of 1600 Btu/hr-ft2 on the bare skin (about five times the level from the sun), a first-degree burn will occur in approximately 20 seconds. The injury will result in damage to the outer layer of skin; this type of burn is extremely painful, leaving the skin reddened that will heal in two to five days. Second-degree injuries burn through the first layer of the skin and damage the second layer of skin. This degree of injury results in intense pain with reddening and blistering of the skin. If treated properly second-degree burns will heal themselves with very little scaring. This degree of injury is considered minor if it involves less than 15% of the body surface area in adults. If the skin is exposed to 1600 Btu/hr-ft2 for more than 50 seconds a third-degree burn will occur. These types of burns may require skin grafting and leave deep scars. In 1958 Stoll and Green [17] conducted an experiment using human subjects to determine the effects of thermal radiation on the skin. A 1000 Watt projection lamp was used to expose a known amount of radiation on an area of the forearm. From the data they were able to establish the time to reach the threshold of pain and the threshold of blistering (second-degree burn) for various levels of radiation. The threshold of pain is defined as when a person first starts to feel pain when exposed to a particular level of radiation. The threshold of blistering is defined as blistering of the skin occurring within a 24 hour period after being exposed to a particular level of radiation. The data from Stoll and Green shows that one can tolerate 1000 Btu/hr-ft2 (about three times the radiation from the sum) directly on the bare skin for about 40 seconds before experiencing pain. At a radiation level of 1500 Btu/hr-ft2 one would feel pain in about 20 seconds. At 3000 Btu/hr-ft2, pain would occur within about six seconds and above 6000 Btu/hr-ft2 one would feel pain almost immediately. The data shows that once pain is felt, blistering will occur fairly quickly if the skin remains exposed to the radiation.
30.3.3 Radiation Effects on Humans
30.3.4 API 521 Recommendations
In this section we will discuss the effects of various levels of radiation on humans. As a benchmark reference, keep in mind that the amount of radiation that reaches sea level at high noon on a clear day is about 317 Btu/ hr-ft2 (1000 W/m2).
The American Petroleum Institute 521 (API 521) provides design radiation levels for personnel working in the vicinity of a flare; these recommendations are listed in Table 30.2 and separated into four radiation levels: 500, 1500, 2000, and 3000 Btu/hr-ft2 [20].
© 2011 by Taylor and Francis Group, LLC
602
Industrial Combustion Testing
Table 30.2 American Petroleum Institute 521 Recommendations for Radiation Exposure to Personnel Permissible Design Level (Btu/h-ft2) 3000
2000
1500
500
a
Conditions Maximum radiant heat intensity at any location where urgent emergency action by personnel is required. When personnel enter or work in an area with the potential for radiant heat intensity greater than 2000 Btu/h-ft2, then radiation shielding and/or special protective apparel (e.g., a fire approach suit) should be considered. SAFETY PRECAUTION: It is important to recognize that personnel with appropriate clothinga cannot tolerate thermal radiation at 2000 Btu/h- ft2 for more than a few seconds. Maximum radiant heat intensity in areas where emergency actions lasting up to 30 s can be required by personnel without shielding but with appropriate clothing.a Maximum radiant heat intensity in areas where emergency actions lasting two to three minutes can be required by personnel without shielding but with appropriate clothing.a Maximum radiant heat intensity at any location where personnel with appropriate clothinga can be continuously exposed.
Appropriate clothing consists of hard hat, long-sleeved shirt with cuffs buttoned, work gloves, long-legged pants, and work shoes. Appropriate clothing minimizes direct skin exposure to thermal radiation.
The API 521 recommends a maximum radiation level of 500 Btu/hr-ft2 at any location where personnel, with appropriate clothing, can be continuously exposed. Appro priate clothing consists of a hard hat, long-sleeved shirt with cuffs buttoned, work gloves, long-legged pants, and work shoes. The API 521 states, however, that three major factors can impact this permissible level: (1) ambient conditions, (2) flare design, and (3) personnel training. Ambient conditions such as wind speed, temperature, and relative humidity can have a significant impact on personnel working in a 500 Btu/hr-ft2 environment. Increasing the ambient temperature and relative humidity increases the heat index and stress on workers. For example, a 500 Btu/hr-ft2 environment might feel comfortable to someone working in Canada on a cold winter day. However, a 500 Btu/hr-ft2 environment might be stressful to someone working in Houston on a hot, humid day during the summer. If a worker is exposed to an additional heat load from a flare flame, his heat stress tolerance will be reduced. In such a case, appropriate adjustments should be made to the design radiant heat intensity. The design of the flare system also impacts personnel working in a 500 Btu/hr-ft2 environment. If personnel are able to temporarily shield themselves from the flare
© 2011 by Taylor and Francis Group, LLC
radiation by positioning themselves under catwalks, scaffolding, or behind the flare stack, they should be able to more effectively tolerate the environment. Finally, training can impact the amount of time personnel can work in a 500 Btu/hr-ft2 environment. If personnel are trained to wear appropriate clothing to minimize direct radiation and can work upwind of the flare to minimize flare radiation levels, they should be more tolerant to the environment. If appropriately clothed, the API 521 recommends that personnel not be exposed to a radiation level greater than 1500 Btu/hr-ft2 for more than two to three minutes. As discussed previously, one would feel pain after about 20 seconds if the skin is exposed to 1500 Btu/ hr-ft2; in about 40 to 50 seconds a second-degree burn would result. It is common for flare manufactures to design flare stack heights so that the maximum radiation level at grade does not exceed 1500 Btu/hr-ft2 during emergency flaring; this allows personnel time to evacuate the area. The API 521 recommends that personnel not be exposed to a radiation level greater than 2000 Btu/hr-ft2 for more than 30 seconds if appropriately clothed. If radiation levels exceed 2000 Btu/hr-ft2 a fire protection suit must be worn. A fire protection suit consists of a silver, nonflammable material containing an air supply bottle. The silver suit helps reflect some of the radiation while the air supply prevents personnel from breathing in the hot ambient air. The suit also consists of a hood designed with a mirrored window that helps reflect some of the incoming radiation but allows personnel to see through. Several other important recommendations API 521 makes include restricted access area to personnel, use of radiation shielding, and location of ladders and platforms; these recommendations are summarized with an illustration in Figure 30.11. The API 521 states that “personnel are commonly protected from high thermal radiation intensity by restricting access to any area where the thermal radiation can exceed 2000 Btu/ hr-ft2.” Some plants locate fences and warning signs around areas where flare radiation levels can exceed 2000 Btu/hr-ft2. The API 521 also states that “it is essential that personnel within the restricted area have immediate access to thermal radiation shielding or protective apparel suitable for escape to a safe location.” Radiation shields are structures sometimes located near the base of the flare stack designed with a galvanized or reflective steel roof. Some plants design a roofed corridor that leads from the base of the stack to a safe location away from high radiation levels. The API 521 also recommends locating ladders and platforms on towers and elevated structures on a side away from the flare. Designing in this manner provides personnel a shield from the radiant heat allowing them more time to evacuate the area.
603
Flare Radiation
“The design of towers or other elevated structures exposed to flare radiation should consider radiation effects ... it is often most effective to accomplish (an escape) by locating ladders and platforms on a side away from the flare.” “Personnel are commonly protected from high thermal radiation intensity by restricting access to any area where the thermal radiation can exceed 2000 Btu/hr–ft2” “It is essential that personnel within the restricted area have immediate access to thermal radiation shielding or protective apparel suitable for escape to a safe location.” Corridor
Radiation shield
Restricted area where radiation can exceed 2000 Btu/hr–ft2 Figure 30.11 API 521 recommendations.
30.4 Estimating Flare Radiation Manufacturers of flares usually provide estimates of flare radiation based on proprietary models. These models are commonly semiempirical in nature in that they use a combination of theory and experimental data. The purpose of this section is to discuss, in general, why and how semiempirical modeling is used to estimate flare radiation and why accurate experimental data is crucial for these models. Every point within a flame radiates heat, however, not every point radiates the same amount of heat. As previously discussed, radiant energy is emitted by the gases and solids present in the flame. The amount of heat radiated from a point within the flame depends largely on the concentration and temperature of H2O, CO2, and soot at that point. Predicting the concentration of gases and solids within the flame, using a theoretical approach, is nontrivial and challenging even with today’s computer power; therefore, it is not a practical method for determining flare radiation for day-to-day engineering design work. As a result, those involved in flare design applications commonly rely on a more practical approach: semiempirical modeling. In general, there are three fundamental approaches for semiempirically modeling flare radiation. The simplest approach is to assume that all of the heat emitted from the flame is concentrated at a single point within the flame (epicenter). All of the heat emitted from the epicenter radiates outward and evenly distributes its energy over the entire surface of a sphere; this type
© 2011 by Taylor and Francis Group, LLC
Single-point model
Multi-point model
Solid-body model
Figure 30.12 Various models commonly used in industry to estimate flare radiation.
of model is referred to as the single-point model and is illustrated in Figure 30.12. Another method commonly used is the multi-point model. This model separates the flame into several segments of equal or varying length and assumes that the heat radiates from the epicenter of each segment and is evenly distributed over the surface of each sphere as illustrated in Figure 30.12; this illustration represents a three-point model. The solid-body model assumes that the flame is a solid-body, typically shaped like a cone, cylinder, or frustum, and that every point on the surface of the solid-body radiates heat. The advantage of the single-point model is that it is the simplest, mathematically, of the three models. The disadvantage, however, is that when the point of interest is located near the flame, the flame cannot be treated as a
604
Industrial Combustion Testing
single-point source. That is, the distance from the point of interest to various locations within the flame differs significantly. On the other hand, when the point of interest is at a significant distance from the flame, the flame can be treated as a single-point source because the distance from the point of interest to every point on the flame is approximately the same. As a rule of thumb, the singlepoint model is typically valid for distance of at least three flame lengths away from the epicenter. Flare manufactures commonly use the multi-point or solid-body model to improve the accuracy at points of interest in close proximity to the flame. To gain better insight into how these models are used, let’s go through a hypothetical example demonstrating the single-point model. Example: Suppose a flare is firing propane at a rate of 50,000 lb/hr and that an observer is located at a distance of 150 feet away from the flame epicenter as illustrated in Figure 30.13. What is the maximum theoretical radiation level at the point of interest? First, let’s calculate the total amount of heat released (or firing rate) from the flame. The total heat release (HR) is calculated by multiplying the fuel flow rate, m ˙ , by the lower heating value (LHV) of fuel. Knowing the LHV of propane is 21,500 Btu/lbm, the heat release can be determined: × LHV HR = m
= 50, 000
Btu (30.2) lbm Btu . × 21, 500 = 1075 × 106 hr hr lbm
Next, let’s calculate the surface area (As) of the sphere with a radius (r) equal to 150 feet:
As = 4 × π × r 2 = 282 , 743 ft 2 .
Propane flame firing 50,000 lb/hr
Flame epicenter
r = 150 ft
Point of interest Figure 30.13 Illustration for example calculation.
© 2011 by Taylor and Francis Group, LLC
(30.3)
Assuming all of the heat is evenly distributed over the surface area of the sphere, the maximum theoretical radiation level (I), at a distance of 150 feet from the flame epicenter, can be determined by dividing the total heat release by the surface area of the sphere:
Btu 1075 HR hr = 3802 Btu . I= = 282 , 743 ft 2 hr − ft 2 As
(30.4)
The actual or measured radiation level would be significantly lower than the maximum theoretical value just calculated because not all of the heat from the flame is transferred to its immediate surroundings by thermal radiation. The fraction of total heat radiated from the flame is referred to as the radiant fraction, F, and is written mathematically as follows:
F=
Actual measured radiation level . (30.5) Maximum theeoretical radiation level
Suppose the actual measured radiation level for this example is 1000 Btu/hr-ft2 at a distance of 150 feet from the epicenter; this would correspond to a radiant fraction of 0.26 (1000/3802). That is, 26% of the total heat released from this flame is transferred to the surroundings by thermal radiation. It should be mentioned that in this example we ignored the transmission loss of the radiation through the air. Recall that as a rule of thumb, 15–25% of the radiation is absorbed by the atmosphere over a distance of 500 feet. The general equation commonly found in the literature for calculating flare radiation levels, I, is written as follows: HR × F × τ I= . (30.6) 4 × π × r2 The term, τ, is referred to as the transmissivity and is the fraction of radiation loss through the atmosphere. Usually the transmissivity is ignored and set equal to a value of one. The flare radiant fraction, F, typically varies from a value of about 0.1–0.4 and is determined from experimental data. The radiant fraction is an overall characteristic of a flame that can be influenced by several variables; three key variables include (1) flare gas composition, (2) mixing, and (3) smoke formation. The composition of the gas being flared can have a tremendous impact on the fraction of total heat radiated from a flame. As mentioned earlier, the presence of soot in a flame usually dominates the radiation emitted over CO2 and H2O. Therefore, hydrocarbon fuels that have a higher carbon-to-hydrogen mass ratio will typically have a higher radiant fraction under similar firing conditions. For example, the carbon-to-hydrogen mass ratio of propane (C3H8) and methane (CH4) is 4.5 and 3,
605
Flare Radiation
(a)
(b)
Figure 30.14 Steam-assisted flare (a) without steam and (b) with steam.
respectively. Since propane has more carbon available to form soot, one would expect it to have a higher radiant fraction than methane under similar firing conditions. The amount of air mixed with the flare gas prior to combustion can also impact flame radiant fractions. If a hydrocarbon is not mixed with an appropriate amount of air prior to combustion, it will have a tendency to create soot within the flame envelope resulting in higher radiant fractions. Propane, for example, chemically “cracks” at about 500ºF forming acetylene. At about 750ºF acetylene combines with radicals to form benzene. Finally, at about 1650ºF benzene agglomerates to form soot particles that are roughly 0.1–10 microns in diameter. This type of reaction is referred to as pyrolysis; a chemical reaction that does not involve oxygen. To reduce the tendency of soot formation, or pyrolysis, an appropriate amount of ambient air must be mixed with the flare gas. Flare manufactures use various flare designs to assist mixing. One type is a steam-assisted flare. Steam-assist flares use high pressure steam to entrain surrounding air and inject it into the core of the flare gas stream. The rapid mixing of the steam and air with the flare gas helps reduce soot formation that tends to lower the flame radiant fraction. Figure 30.14 shows a steam-assisted flare operating under identical flare gas flow conditions with and without steam-assist. Notice without steam-assist, the flame is more luminous and contains more soot; this results in higher radiant fractions. The fraction of heat radiated from a flame can also be greatly increased by the presence of liquid droplets in the gas. Droplets within a hot flame can easily be converted to soot [21].
© 2011 by Taylor and Francis Group, LLC
Smoke acting as a shield to block radiation
Figure 30.15 Smoking flare.
The amount of smoke formation can significantly impact flare radiant fractions. If soot does not burn completely within the flame envelope it will cool, creating black smoke leaving the flame as demonstrated in Figure 30.15. The carbon particles, or smoke, are an
606
excellent absorber of radiation and can act as a shield to block thermal radiation. Notice in the photograph that a majority of the upper part of the flame is shrouded or blocked by the smoke. This can have a dramatic impact on the amount of heat radiated from the flame to its immediate surroundings.
Industrial Combustion Testing
Incoming radiation
1. Sensor surface
2. Thermopiles
30.5 Measuring Flare Radiation As previously discussed, the models used in the flare industry for estimating flare radiation rely heavily on flare radiation data from full-scale test equipment. In this section we will discuss the instrumentation commonly used in the flare industry to measure radiation levels. 30.5.1 Description of the Radiometer Flare radiation is commonly measured using an instrument called a radiometer. There are several brands of radiometers available on the market; Figure 30.16 shows several handheld thermopile types. Radiometers are meters used to detect and measure radiant electromagnetic energy. There are many types of radiometers based on their physics of detection and the intended applications. Due to the spectral range of thermal radiation from flares, it is common to use thermopile-type radiometers to measure flare radiation. A thermopile is a kind of thermometer for measuring heat radiation, consisting of multiple thermocouple junctions in series. A thermocouple is a junction between two different metals that produces a voltage related to a temperature difference relative to a reference temperature.
Figure 30.16 Several brands of handheld radiometers.
© 2011 by Taylor and Francis Group, LLC
3. Heat sink
Figure 30.17 The major components that make up a flare radiometer.
Flare radiometers typically consist of three main components: (1) sensor surface, (2) thermopile, and (3) heat sink as illustrated in Figure 30.17 [22]. The incoming radiation is absorbed on the sensor surface that typically consists of a dark material designed to absorb most of the flare radiation that it receives. The absorbing material is typically very thin, without a lot of thermal mass, so that it will reach temperatures quickly resulting in a fast response time. The heat absorbed on the sensor surface is then transferred through thermopiles located just beneath the absorbing material. Thermopiles consist of very small thermocouples joined together in series. The heat flows through the thermopiles and into a heat sink located just below. The difference in temperature between the sensor surface and heat sink is a function of the heat being transferred and a function of the net absorbed heat flux [22]. The thermopile converts this flow of heat to an electrical signal that is then converted into a radiation level. Radiometers of this type do not require a power supply since the electrical output signal is generated by heat transfer. Some radiometers are designed to give a direct display of the radiation level in units of Btu/hr-ft2 or W/m2. Some require one to measure the electrical output signal and convert the reading to a radiation level. The output signal is usually in units of milli Volts (mV) and can be measured using a handheld voltmeter. Manufacturers of radiometers usually supply the customer with a certificate of calibration that usually consists of a plot showing the response of the output signal at various levels of incident radiation. A Certificate of Calibration for one particular radiometer is shown in Figure 30.18. For this radiometer, the output signal response is linear with the incident radiation level. This particular radiometer has a constant of 293.6 Btu/hr-ft2 per mV. So for example, suppose one aims this radiometer at a flame and the voltmeter reads 7.43 mV. Taking that number and multiplying it by the constant yields the radiation level 293.6 × 7.43 = 2181 Btu/hr-ft2.
607
Flare Radiation
PAGE 2 OF 3
CERTIFICATE OF CALIBRATION 9/5/03 DATE CUSTOMER John Zink Co.
12
TRANSDUCER OUTPUT, millivolts
Tulsa, OK
P.O. NO. 613268 CERTIFICATE NO. 13186–1B MODEL NO. 64–0.5–16/ZnSeW– 1C–150 131861 SERIAL NO. SENSOR TYPE Schmidt–Boelter 0.94 ABSORPTANCE Zinc selenide, 150°VA WINDOW REFERENCE STANDARD 587025A 6 CALIBRATED BY
RESPONSIVITY: 0.003406 mV per Btu/(ft2-Hr), or the inverse: 293.6 Btu/(ft2-Hr) per mV
10
8
CALIBRATION RESULTS SUMMARY: FULL SCALE OUTPUT LEVEL:
6.13 mV at 1800 Btu/(ft2-Hr) RESPONSIVITY: 0.003406 mV per Btu/(ft2-Hr), or the inverse: 293.6 Btu/(ft2-Hr) per mV
6
UNLESS NOTED, CALIBRATION CONDITIONS: Non-condensing Ambiest Air at 23 ± 3°C Relative Humidity Less Than 70% Expanded uncertainty± 3% of responsivity, Coverage factor k-2, –95% confidence level, Test uncertainty ratio (TUR) is less than 4:1.
4
Calibration was performed in compliance with ISO/IEC 17025, ANSI/NCSL Z540–1 and MIL-STD-45662A to MEDTHERM PI-20 with traceability to the National Institute of Standards and Technology.
2
HEAT FLUX ABSORBED Btu/(ft2-Hr) INCIDENT
This certificate applies only to the item described above. It is not to be reproduce, except is its certainly, without writen permission from MEDTHERM corporation.
0
600
1200
1800
POST OFFICE BOX 412/HUNTSVILLE, ALABAMA/TELEPHONE (256) 837–2000/FAX (256) 837–2001 Figure 30.18 A radiometer Certificate of Calibration.
When flare radiometers are designed with optical covers, the manufacturer will usually provide two calibration constants; one with the cover material on and one with the cover material off. The calibration constant with the cover material on should be used for most users. For more advanced users that wish to change the cover material, the calibration with the cover material off is often useful. 30.5.2 Useful Tips when Measuring Flare Radiation Radiation data gathered from full-scale testing is critical information used to design a flare; therefore, accurate measurements are important. Several tips to help eliminate potential error and improve the accuracy of radiation measurements include the following: (1) select the right radiometer, (2) avoid convective cooling effects, (3) measure without sun interference, (4) position the
© 2011 by Taylor and Francis Group, LLC
radiometer so that it views the entire flame, and (5) vary the angle of the radiometer to find the maximum radiation level. 30.5.2.1 Radiometer Selection The reader should be aware that there are several types of radiometers available in the market place. Many of these radiometers are designed to measure solar radiation and, therefore, are not capable of measuring the radiation in the far infrared region where flare radiation is important; recall that approximately 99.9% of the radiation emitted resides in the infrared spectrum. If radiometers specifically designed for solar radiation are used to measure flare radiation, a majority of the radiation in the infrared spectrum will not be captured; this will result in a measured value that will be significantly lower than the actual radiation level. It is therefore important to use a radiometer that will
608
Industrial Combustion Testing
capture radiation over a broad infrared spectrum when measuring flare radiation; a radiometer that will measure in the wavelength range from about 0.3 × 10−6 to 30 × 10−6 meters is usually adequate. It is important to select the correct type of radiometer to ensure the sensor can respond to the range of radiation wavelengths of interest and to select the correct type of optical cover. 30.5.2.2 Sun Interference It is preferred to measure flare radiation with the sun at your back where it will not contribute to the measurement. This is especially important on partly cloudy days because clouds could alter the sun’s contribution during a test. 30.5.2.3 View Angle The radiometer should be position so that the sensor surface views the entire flame. Some radiometers have a narrower view angle depending on how far back the absorbing material is set back into the cavity. If it is set deep into a cavity the view angle might be fairly narrow. Caution should be exercised to ensure the entire flare flame can be covered in the view angle of the radiometer. Figure 30.19 illustrates an example of a radiometer not viewing the entire flare flame. 30.5.2.4 Convective Cooling Effects Wind blowing over a radiometer sensor surface during a measurement could result in an erroneous reading. As wind blows past the radiometer sensor surface, it can cool the sensor surface, resulting in a radiation level reading lower than the actual level. To eliminate the convective cooling effects, radiometer manufacturers commonly use one of the two methods: (1) position the
Flare flame Sensor surface recessed into a cavity
View angle
Cavity
Figure 30.19 Illustration showing a radiometer not viewing the entire flare flame.
© 2011 by Taylor and Francis Group, LLC
Recessed cavity
Optically transparent window
Figure 30.20 Radiometers with different types of windows.
sensor surface into a recessed cavity or (2) cover it with a special window as shown in Figure 30.20. The effectiveness of the cavity method is not well understood at different wind speeds and directions. It is possible that even at quiescent ambient conditions, natural convective cooling can occur and result in lower-than-actual radiation readings. The second method is to use a special optical material placed over the surface of the absorbing material. The cover material is typically in close proximity, but not in direct contact with the sensor surface. The cavity formed between the cover and the sensor surface is sometimes filled with an inert gas to prevent oxidation of the sensor surface. This cover material blocks the wind from cooling the absorbing material on the sensor surface. It is important to use the proper cover material. For example, typical glass that one would find in a window would not work. Although it is transparent to the human eye, a large portion of the infrared radiation emitted from the flare flame will not pass through it resulting in a low radiation reading. One might argue that some sort of lab calibration may correct the problem. However, the percentage of flare radiation that can penetrate the cover material may vary significantly depending on the spectral distribution of the flare flame; this is affected by various factors such as how sooty the flame is, the temperature and concentration fields of the flame, and so forth. Several materials can be used to shield a radiometer from the wind; a common material used for flare radiometers is zinc selenide. Zinc selenide is a light yellow, man-made material rarely found in nature and is highly transparent to the radiation emitted from a flare flame. Figure 30.21 shows the transmissivity curve as a function of wavelength for zinc selenide. The transmissivity of zinc selenide is relatively constant over the wavelength range between 0.7 × 10−6 and 16 × 10−6 m, making it an excellent choice for flare radiation measurement. Figure 30.22 shows the transmissivity curve as a function of wavelength for sapphire. The transmissivity varies with the wavelength between 0.7 × 10−6 and 4.0 × 10−6 m, and more importantly, drops off rapidly at wavelength above 4.0 × 10−6 m. A significant fraction of flare radiation will be clipped and not measured by the sensor if a sapphire cover is used to measure flare radiation.
609
Flare Radiation
100
Transmission (%)
80 60 40 20 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
2.0 3.0 4.0 5.0 Wavelength (µm)
10
20
30
Figure 30.21 Transmissivity curve of zinc selenide.
100
Transmission (%)
80 60 40 20 0 2.0 3.0 4.0 5.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Wavelength (µm)
10
20
Figure 30.22 Transmissivity curve of sapphire. 1.0 0.9 0.8
Incoming radiation
0.7 Reflectivity
radiation of zinc selenide at a wavelength of 1 × 10−6 m (refractive index of 2.49). It can be seen that the reflectivity remains virtually constant when the incidence angle is between 0° and 60°, but starts to deviate at greater angles. Besides the view angle limit set by physical blockage of the radiometer, reflectivity is another important factor affecting the view angle of a radiometer. Dome-shaped optical covers can help reduce reflection; however, constructing a dome-shape cover out of high performance optical material is not cost effective and therefore typically not used for flare radiometers.
Angle of incidence
0.6 0.5 0.4 0.3 0.2 0.1 0.0
0
10
20
30 40 50 60 Incidence angle (degrees)
70
80
90
Figure 30.23 Reflected radiation at a wavelength of 1 × 10−6 m for zinc selenide.
Flare radiometers are often designed with flat optical covers. One shortcoming of a flat optical cover is that the amount of reflected radiation increases dramatically when the angle of incidence is high (Snell’s law) [23]. For example, Figure 30.23 shows the fraction of reflected
© 2011 by Taylor and Francis Group, LLC
30.5.2.5 Measuring Technique Flare gas flow rate and composition, wind speed and direction, the location of the flare flame’s epicenter can vary during a flare test. Due to the Lambert cosine law, the maximum radiation reading will be measured when the radiometer is aimed directly at the flare epicenter. It is often difficult to predict the location of the epicenter and to position the radiometer to aim at the epicenter prior to a flare test.
610
Traditionally, flare radiation is measured manually by trained, properly clothed personnel. During measurement, the operator varies the angle of the radiometer between left and right and between up and down, continuously searching for the maximum reading. This approach is often called manual scanning. Manual scanning suffers two significant shortcomings. First, if the flare epicenter fluctuates in space, this searching method is not effective in tracking the epicenter. Second, if the flare gas flow rate or composition varies in time, this method cannot distinguish the effects from these variations in flow conditions from those caused by the angle perturbation, and therefore fails to track the flame epicenter. As a result, flare radiation measurements using manual scanning are often associated with relatively large errors. The magnitude of these errors depends largely on the specific individuals conducting these tests and flow conditions. 30.5.3 Radiometer Cube A radiometer cube is a state-of-the-art device developed by the John Zink Company to accurately measure flare radiation and is shown in Figure 30.24. This instrument is designed with three radiometers, attached to the three adjacent surfaces of a cube near a corner. The radiometer
Industrial Combustion Testing
cube [24] is placed on a tripod and located in the field at a certain distance from the base of the flare stack. A bubble level, attached to the top surface of the cube, is used to ensure that the top surface is positioned in a horizontal plane. A sighting tube is attached to the top surface of the cube to allow the orientation of the cube to be accurately controlled and determined. The corner of the cube is pointed in the general direction of the flare flame, with the sighting tube pointing at an object with known coordinates. A set of at least two cubes are used to achieve the full capability of the instrument. The relative locations of the radiometer cubes and the flare, as well as the orientations of the cubes are determined prior to the flare test and entered into the computer to allow real-time tracking of key flare radiation parameters. Each radiometer on the cubes is connected to a data acquisition system. The data acquisition system collects the data from each sensor and performs trigonometric calculations using a complex set of equations. The results are real-time parameters of the flare radiation: total radiation levels from both cubes, incidence angles, and the flare epicenter coordinates. This instrument has three main advantages over the handheld instrument:
1. Reduces measurement errors 2. Avoids placing personnel in hazardous situations 3. Locates the flame epicenter
The John Zink proprietary radiometer cubes can be used in conjunction with the API 521 single-point epicenter flare radiation model to obtain objective, experimental values for the radiant fraction and the epicenter locations, which have been often regarded as somewhat subjective and difficult to quantify.
References Figure 30.24 The John Zink radiometer cube.
© 2011 by Taylor and Francis Group, LLC
1. Baukal, C. E. The John Zink Combustion Handbook. Boca Raton, FL: CRC Press, 2001. 2. Baukal, C. E. Industrial Combustion Pollution and Control. New York: Marcel Dekker, Inc., 2004. 3. efunda (Engineering Fundamentals). “Heat Transfer: Radiation,” http://www.efunda.com/formulae/heat_ transfer/radiation/overview_rad.cfm 4. Cornforth, J. R. Combustion Engineering and Gas Utilizations. London: E&FN Spon, 1992. 5. Wikipedia, the free encyclopedia. “Infrared,” http:// wikipedia.org/wiki/Infrared 6. U.S. Department of Energy. “Solar Energy Technologies Program: Light and the PV Cell,” http://www1.eere. energy.gov
Flare Radiation
7. Bird, R. E., and Hulstrom, R. L. A Simplified Clear Sky Model for Direct and Diffuse Insolation on Horizontal Surfaces.” SERI Technical Report SERI/TR-642-761, February 1991, Solar Energy Research Institute, Golden, CO. 8. Wikipedia, the free encyclopedia. “Earth,” http:// en.wikipedia.org/wiki/Earth. 9. Bussman, W. R., and Knott, D. “Unique Concept for Noise and Radiation Reduction in High-Pressure Flaring,” OTC 12160, Offshore Technology Conference, Houston, Texas, May 2000. 10. McQuiston, C., and Parker, J. D. Heating, Ventilating, and Air Conditioning. New York: John Wiley & Sons, 1982. 11. Geog 474. “Energy Interaction with the Atmosphere and at the Surface,” http://www.udel.edu/Geography/ DeLiberty/Geog474/spectrum.jpg 12. Brzustowski, T. A., and Sommer, E. C., Jr. “Predicting Radiant Heating from Flares.” Proceedings of the Division of Refining, API 1973, Vol. 53, API Washington, DC, pp. 865–93. 13. DiNenno, P. J. “Simplified Radiation Heat Transfer Calculations from Large Open Hydrocarbon Fires.” Society of Fire Protection Engineers, SFPE Technology Report 82-9, 1982. 14. Babrauskas, V. “Ignition of Wood: A Review of the State of the Art.” Interflam 2001, Interscience Communications Ltd., London, pp. 71–88, 2001.
© 2011 by Taylor and Francis Group, LLC
611
15. Society of Wood Science and Technology, Teaching unit number 2, http://www.swst.org 16. McGuire, J. H. “Fire and Spatial Separation of Building.” Fire Technology 1 (1965): 278–87. 17. Stoll, A. M., and Green, L.C. 1958, The Production of burns by thermal radiation of medium intensity. ASME Paper No. 58-A-219 (American Society of Mechancial Engineers). 18. Bilo, M., and Kinsman, P. R. “Thermal Radiation Criteria Used in Pipeline Risk Assessment. Pipes and Pipelines International (November–December, 1997): 17–25. 19. Hymes, I., Boydell, W., and Prescott, B. Thermal Radiation: Physiological and Pathological Effects. Rugby, UK: Institute of Chemical Engineers, 1996. 20. American Petroleum Institute. API Standard 521, 5th ed. January 2007. 21. Schwartz, R. E., and White, J. W. “Flare Radiation Prediction: A Critical Review.” Presented at the 30th annual Loss Prevention Symposium of the American Institute of Chemical Engineers, Session 12, February 1996. 22. Hukseflux Thermal Sensors, http://www.hukseflux. com/thermalScience/thermalSensors.html 23. Siegel, R., and Howell, J. R. Thermal Radiation Heat Transfer, 3rd ed. London: Taylor & Francis, 1992. 24. Hong, J., White, J., and Baukal, C. “Accurately Predict Radiation From Flare Stacks.” Hydrocarbon Processing (June 2006): 79–81.
Section V
Testing in Combustors
© 2011 by Taylor and Francis Group, LLC
31 Cement Kilns Eugen Dan Cristea and Givanni Cinti Contents 31.1 Introduction...................................................................................................................................................................616 31.1.1 History..............................................................................................................................................................616 31.1.2 Modern Cement Plant General Layout........................................................................................................617 31.2 Cement Clinker Rotary Kiln Classification...............................................................................................................618 31.2.1 Introduction to Portland Cement Manufacturing and Classification.....................................................618 31.2.2 Pyroprocesses Classification.........................................................................................................................619 31.2.2.1 Wet Process Cement Rotary Kilns...............................................................................................619 31.2.2.2 Long Dry Process Cement Rotary Kilns.................................................................................... 622 31.2.2.3 Semidry Process Cement Rotary Kilns...................................................................................... 622 31.2.2.4 Dry Process Cement Rotary Kilns.............................................................................................. 622 31.2.3 Rotary Kiln Design Parameters................................................................................................................... 624 31.2.3.1 Material Residence Time.............................................................................................................. 624 31.2.3.2 Rotary Kiln Degree of Fill............................................................................................................ 625 31.2.3.3 Rotary Kiln Slope.......................................................................................................................... 625 31.2.3.4 Rotary Kiln Capacity.................................................................................................................... 625 31.3 Conventional and Alternative Fuels.......................................................................................................................... 626 31.3.1 Conventional Fuels........................................................................................................................................ 626 31.3.1.1 Solid Fuels...................................................................................................................................... 626 31.3.1.2 Gaseous Fuels................................................................................................................................ 628 31.3.1.3 Liquid Fuels.................................................................................................................................... 629 31.3.2 Alternative Fuels............................................................................................................................................ 630 31.3.3 PF Firing System Classification.................................................................................................................... 631 31.3.3.1 Direct Firing................................................................................................................................... 631 31.3.3.2 Semidirect Firing........................................................................................................................... 631 31.3.3.3 Indirect Firing................................................................................................................................ 631 31.4 Flame Aerodynamics in Rotary Kiln Section.......................................................................................................... 632 31.4.1 Free Turbulent Jets......................................................................................................................................... 633 31.4.1.1 Isothermal (Constant-Density) Conditions................................................................................ 633 31.4.1.2 Nonisothermal (Combustion) Conditions................................................................................. 634 31.4.2 Confined Turbulent Jets................................................................................................................................ 634 31.4.2.1 Confined Jet with Sufficient Amount of Surrounding Flow................................................... 635 31.4.2.2 Confined Jet with Insufficient Amount of Surrounding Flow............................................... 635 31.4.3 Confined Turbulent Jet with Swirling......................................................................................................... 637 31.4.3.1 Swirl Number................................................................................................................................. 637 31.4.3.2 Modified Swirl Numbers.............................................................................................................. 638 31.4.3.3 Swirl Number Calculation for Cement Rotary Kiln Burners................................................. 639 31.4.4 Combustion System Scaling......................................................................................................................... 639 31.5 Rotary Kiln Burners..................................................................................................................................................... 640 31.5.1 Multichannel Low NOx Burners.................................................................................................................. 641 31.5.1.1 Rotaflam Burner............................................................................................................................ 641 31.5.1.2 FlexiFlame™ Burner..................................................................................................................... 642
615 © 2011 by Taylor and Francis Group, LLC
616
Industrial Combustion Testing
31.5.1.3 M.A.S.® Burner............................................................................................................................... 643 31.5.1.4 Duoflex Burners............................................................................................................................. 644 31.6 Burner Pipe Testing on Cement Rotary Kilns in Industrial Operation............................................................. 645 31.6.1. Semipilot Scale Experimental Tests.......................................................................................................... 645 31.6.2. Burner Pipe Testing on Cement Rotary Kiln........................................................................................... 648 31.7 Process Mathematical and Physical Modeling...................................................................................................... 649 31.7.1. Mathematical Simulation........................................................................................................................... 649 31.7.1.1. PDEs Governing Equations..................................................................................................... 649 31.7.2. Physical Modeling....................................................................................................................................... 652 31.8 Pollutant Formation and Destruction..................................................................................................................... 654 31.8.1. NOx Formation in Clinker Burning Process............................................................................................ 655 31.8.2. SNCR and SCR Technologies..................................................................................................................... 656 31.9 Cement Rotary Kiln Thermal Efficiency................................................................................................................ 656 31.9.1. Introduction.................................................................................................................................................. 656 31.9.2. Goals of H&M Balance................................................................................................................................ 657 31.9.3. Example of H&M Balance of Cement Rotary Kiln................................................................................. 657 31.9.3.1. Materials Chemical Analyses................................................................................................. 657 31.9.3.2. Kiln Feed to Clinker Ratio Calculation................................................................................. 657 31.9.3.3. Specific Heat Consumption Calculation............................................................................... 658 31.9.3.4. Clinker Heat of Formation...................................................................................................... 658 31.9.3.5. Combustion Engineering Calculation................................................................................... 659 31.9.3.6. Gas and Material Heat Flow Rate Calculation..................................................................... 662 31.9.3.7. Shell Heat Losses...................................................................................................................... 663 31.9.3.8. Heat Balance.............................................................................................................................. 663 31.9.3.9. Mass Balance............................................................................................................................. 663 31.9.3.10. Clinker Cooler........................................................................................................................... 663 31.9.3.11. Flow Rates Chart...................................................................................................................... 663 31.9.3.12. Gas Analysis Chart................................................................................................................... 664 31.9.3.13. Electrical Specific Consumption............................................................................................. 664 31.9.3.14. Appendix Formulas of H&M Balance................................................................................... 664 References................................................................................................................................................................................. 668
31.1 Introduction 31.1.1 History The vertical, single shaft kilns were used for burning limestone to produce quicklime 2000 years ago. Romans used such vertical kilns to burn pozzolanic lime. The term pozzolan applies to the incoherent pyroclastic-sialitic rocks occurring in the neighborhood of PozzouliItaly, and it has subsequently been extended to include a wide range of both natural and inorganic materials differing in nature, composition, and structure. In 1824 Joseph Aspdin developed and patented a cement manufacturer process that may be considered a precursor to modern Portland cement. William, his younger son, discovered that if clinkered (overburnt) the material, the strength of the cement is increased. At Northfleet in Kent, William’s kiln produced a cement 2.4 times stronger than the best Roman’s cement. In 1885 an English engineer named Frederick Ransome patented, in England, a slightly tilted horizontal kiln able to be
© 2011 by Taylor and Francis Group, LLC
rotated by a mechanism, to produce cement clinker. This kiln was fired on natural gas. The first rotary kilns have been built having the overall dimensions of an internal diameter up to 2 m and the length up to 25 m able to produce about 50 tpd. At the beginning, the rotary kilns were used for the wet process because of high natural moisture of the raw materials in England. In North America the first rotary kiln developed by Hurry and Seaman of the Atlas Cement Company came on line in 1895. Here the first rotary kilns were operated with a dry powder feed. The Lepol kiln was invented and patented by Dr. Otto Lellep and marketed by Polysius in Germany, the combination of both names leading to Lepol (LellepPolysius). The first kiln was sold in 1928 under the name of “Lepol.” In 1932 Vogel-Jorgensen, an engineer from F.L. Smidth Co., originally submitted the suspension preheater kiln patent to the patent office in Prague. However, in 1951–1953 the Humbolt Company designed and installed the first four-stage suspension preheater system on 300 tpd rotary kiln with 3.2 m internal diameter and 40 m length in Beckum, Germany.
617
Cement Kilns
In the early 1980s the preheater rotary kiln size reached a typical capacity of about 3000 (m)tpd. Further increasing of the outputs were limited by a span life of the brick refractory lining of a rotary kiln shell with a diameter more of 6 m. The precalciner/preheater kilns were developed to reach production up to 10,000 (m)tpd and more, by splitting the total fuel input between the main burner and the auxiliary burners of an additional vertical vessel, called precalciner, located in between the rotary kiln section and the suspension preheater. The more energy efficient preheater and preheater/ precalciner kilns usually route the exhaust gas from the preheater to a raw mill to dry the material in suspension during the grinding process. The flue gas stream exits the raw mill heavily laden with kiln raw material and is exhausted to a baghouse filter or electrostatic precipitator to recover the raw material and any material entrained from the kiln preheater system. The raw material is collected and fed to the blending homogenization silo system to provide the kiln with a homogenous raw feed. 31.1.2 Modern Cement Plant General Layout Currently, the main sections of a modern cement plant are as follows: • • • • •
quarries crushing raw material and raw coal storages raw meal grinding raw meal homogenization silo
Quarry
• burning line (preheater tower, rotary kiln section, and clinker cooler) • clinker storage and dispatching • cement grinding • cement shipment Sometimes in the most advanced burning lines with a rate capacity higher then 4000 tpd, a waste heat recovery and power generation system to enhance the energy efficiency of a cement plant is installed. Figure 31.1 shows the main sections of a modern cement plant flow sheet. The raw materials are drilled and blasted in hard rock quarries that represent the first step in the cement manufacturing process. The following sections are loading, transportation, and crushing to delivery of the raw materials to the cement plant. The raw materials are ground to a very fine powder and then blended in the correct proportions. This blended raw material is called the “raw feed” or “kiln feed” and is burned in the cement rotary kiln, where it reaches a temperature of about 1400°C–1500°C. The raw feed enters the kiln at the back end and gradually passes down to the hot end, then falls out of the kiln and cools down into the clinker cooler. The product formed in the kiln is described as “clinker” and is typically composed of rounded nodules between 1 mm and 25 mm size. After cooling, the clinker may be stored temporarily in a clinker store, or it may pass directly to the cement mill. The cement mill grinds the clinker to a fine powder. A small amount of gypsum (a form of calcium sulfate) is ground up with the clinker. The gypsum controls the setting properties of the cement when water is added.
Raw materials and coal storages Coal grinding
Crushing
Raw meal grinding Hot gas generator
Coal silo
Material Flue gas
Exhaust air Fuel Primary air
Burning line Raw meal homo silo
Clinker silo
Cement silo
Cement grinding Gypsum and additives storage
Figure 31.1 Flow-sheet of a modern cement plant.
© 2011 by Taylor and Francis Group, LLC
Cement packing and shipment
618
The cement is sent to the shipment section and from there is dispatched in bulk or bags.
31.2 Cement Clinker Rotary Kiln Classification 31.2.1 Introduction to Portland Cement Manufacturing and Classification The pyroprocessing stage is generally considered as the heart of the cement-making process. The rotary kilns are typically cylindrical, refractory-lined vessels being slightly inclined from the horizontal and rotated by its driving system. The material is fed in at the higher back end and exits from the lower front end of the kiln. The rotary kiln is fired by a single burner pipe located on either the vessel symmetry axis or in the horizontal position relative to kiln platform, in countercurrent with the material feeding. The cross section of the rotary kiln is characterized by two distinct regions, one is the free board and the other the material bed. The process flue gases flow through the freeboard area and the material occupies the bed. The material is heated from the upper side by thermal radiation from the flame and kiln refractory lining and by forced conduction from the flue gases. The material is also heated below by thermal conduction from the refractory lining. The rotary kiln section is lined with refractory bricks of a type that change along its length in accordance with the longitudinal temperature distribution chart of the flue gas and material. During kiln operation the bricks refractory lining become coated with a layer of clinker, which plays an essential role in the insulation and extension of its life span. Proper flame shape control of the main burner pipe extends the life of the refractory lining in the burning zone of the rotary kiln. Portland cement clinker is formed by a predetermined homogeneous mixture of raw materials comprising of lime, silica, and a small proportion of alumina generally iron oxide, which need to be burned and clinkerized into the cement rotary kilns. The ASTM C 150 standard defines Portland cement as “hydraulic cement, cement that not only hardens by reacting with water but also forms a water-resistant product, produced by grinding clinkers consisting essentially of hydraulic calcium silicates, usually containing one or more of the forms of calcium sulfate as an inter ground addition.” Clinkers are nodules of a clinkerized material that is manufactured when a raw mixture of predetermined composition is burned to a high temperature. The low cost and widespread availability of the limestone,
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
shale, and other naturally occurring materials make Portland cement one of the lowest-cost building materials widely used over the last century throughout the world. Different types of Portland cement are produced to meet different physical and chemical requirements for specific purposes, such as durability and high-early strength. Eight types of cement are covered in ASTM C 150 and AASHTO M 85. These types and brief descriptions of their uses are listed in Table 31.1. Although IA, IIA, and IIIA (air-entraining cements) are available as options, concrete manufacturers prefer to use an air-entraining admixture during concrete manufacture, where they can get better control in obtaining the desired air content. If a given type of cement is not available, comparable results can frequently be obtained by using modifications of available types. High-early strength concrete, for example, can be made by using a higher content of Type I when Type III cement is not available, or by using admixtures such as chemical accelerators or high-range water reducers (HRWR). The production of Portland cements will be affected for years to come by energy and pollution requirements. In fact, the increased attention to pollution abatement and energy conservation has already greatly influenced the cement industry, especially in the production of low-alkali cements. Using high-alkali raw materials in the manufacture of low-alkali cement requires cement rotary kiln bypass systems to avoid concentrating alkali in the clinkers, which consumes more energy. Limiting use of low-alkali cement to cases in which alkali-reactive aggregates are used could lead to significant improvement in energy efficiency. The issue of a European Union directive aimed at removing trade barriers to a variety of products, among them cement, led to the setting up of CEN/TC 51 as well Table 31.1 Portland Cement Types and Their Uses According to ASTM C150 Cement Type I1 II2 III IV3 V4 IA4 IIA4 IIIA4
Use General purpose cement, when there are no extenuating conditions Aids in providing moderate resistance to sulfate attack When high-early strength is required When a low heat of hydration is desired When high sulfate resistance is required A type I cement containing an integral air-entraining agent A type II cement containing an integral air-entraining agent A type III cement containing an integral air-entraining agent
619
Cement Kilns
as the preparation of a series of standards (EN 196) concerning test methods. These were adopted by CEN and later by ISO and are comparable to ASTM standards. The release in 1988 of a further European directive dealing specifically with construction products required a new approach from CEN/TC 51’s work toward cement definition, composition, specifications, and conformity criteria. The EU standard EN-197-1 states five type of cements, 11 groups depending on additive used and 25 subgroups classified function on composition is shown in Table 31.2. The strength requirements of the cements are expressed as early (2–3 days) or late (28 days) ages or both as displayed in Table 31.3. The ENV197-1 is focusing on 28-days strength then introduces three classes 32.5, 42.5, and 52.5; these figures represent the minimum characteristics strength expressed in N/mm2 that the cement is required to achieve after 28 days of aging based on laboratory tests carried out in compliance with the EU standard EN-196-1. The subclasses indicated by the suffix R are introduced in order to provide a measure of classification in respect to early strength. 31.2.2 Pyroprocesses Classification It is customary to classify the cement clinker manufacturing processes according to Table 31.4. The pyroprocessing stage is classified with respect to raw feed water content. There are many different kiln system designs, but they are all in essence performing the following material transformation, starting in order from the rotary kiln feed (back) end: • Evaporating free water, at temperatures up to 100°C. • Removal of adsorbed water in clay materials from 100°C to 300°C. • Removal of chemically bound water in the range 450°C–900°C. • Decarbonation (calcination) of carbonate components from 700°C to 850°C. • Formation of aluminates and ferrites and C2S from 800°C to 1250°C. • Formation of liquid phase melt at temperature higher than 1250°C. • Formation of C3S in temperature range of 1330°C to 1450°C. • Cooling of clinker to solidify liquid phase 1300°C–1240°C. • Final clinker microstructure frozen in clinker < 1200°C. • Clinker cooled in cooler from 1240°C to 100°C.
© 2011 by Taylor and Francis Group, LLC
The chemical reactions below 1300°C are calcination, decomposition of clay minerals as well as the reaction of calcium carbonate (calcite) CaCO3 or calcium oxide (lime) CaO with quartz and clay mineral decomposition products. Calcination of calcite, decomposition of clay minerals are endothermic reactions, while reaction of calcite or lime with quartz and clay mineral decomposition products are exothermic. Calcination of pure calcium carbonate is done according to the reaction:
CaCO 3 → CaO + CO 2 + 1782 kJ/kg CaCO3 . (31.1)
The chemical reactions between calcareous and clay materials constituent oxides are as follows:
2 CaO + SiO 2 → βC2 S − 734 kJ/kg C2 S,
(31.2)
3CaO + αAl 2 O 3 → C3A − 27 kJ/kg C3A, (31.3)
4CaO + αAl 2 O 3 + αFe2 O 3 → C 4AF − 1,782 kJ/kg CaCO3 . (31.4) Heat of reactions are referring to 25°C and 1 atm. After heating up to 1300°C the main phases are belite C2S, aluminate C3A, and ferrite C4AF, and free lime. Burning from 1300°C to 1450°C, a liquid phase representing 20 to 30% of solid phase is formed. The material nodulizes to form clinker, and most of the existing CaO reacts with a large proportion of belite to form alite.
βC2 S + CaO → C3S + 59 kJ/kg C3S.
(31.5)
In addition to the formation of alite, α-belite is transformed to β form and recrystalization and crystal growth proceeds for already formed components. If the clinker is underburned, it will contain an excessive quantity of CaO. The clinker, as a low alite content, is poor quality. If the clinker is overburned, the clinker results in large crystals. Table 31.5 shows the main characteristics and performances of the different classes of cement rotary kilns. 31.2.2.1 Wet Process Cement Rotary Kilns The most predominant technology for clinker manu facturing during the twentieth century was the long wet process kiln, with a length to internal shell diameter ratio (L/D) of as high as 40. The low thermal efficiency of these kilns proved to be their major downfall; they required the evaporation of a large quantity of water from the raw feed, which was introduced as a slurry.
620
Table 31.2 Common Cement Type and Composition According to ENV-197-1 Cement Type I II
Description Portland cement Portland slag cement Portland silica fume cement Portland pozzolana cement
Portland fly ash cement
Portland shale cement Portland limestone cement Portland composite cement Blast furnace cement
IV
Pozzolanic cement
V
Composite cement
© 2011 by Taylor and Francis Group, LLC
I II/A-S II/B-S II/A-D
95–100 80–-94 65–79 90–94
II/A-V II/B-V II/A-W II/B-W II/A-V II/B-V II/A-W II/B-W II/A-T II/B-T II/A-L II/B-L II/A-M II/B-M III/A III/B III/C IV/A IV/B V/A V/B
80–94 65–79 80–94 65–79 80–94 65–79 80–94 65–79 80–94 65–79 80–94 65–79 80–94 65–79 35–64 20–34 5–19 65–89 45–64 40–64 20–39
Pozzolana Natural P % w/w
Industrial Q % w/w
Silica Fume D % w/w
Fly Ash Siliceous V % w/w
Calcareous W % w/w
Limestone L % w/w
Minor Addition % w/w 0–5 0–5 0–5 0–5
6–20 21–35 6–10
6–20 21–35 6–20 21–35
6–20 21–35
0–5 0–5 0–5 0–5 0–5 0–5 0–5 0–5 0–5 0–5 0–5 0–5
6–20 21–35 35–65 66–80 81–95
18–30 31–50
18–30 31–50
18–30 31–50
0–5 0–5 0–5 0–5 0–5 0–5 0–5
Industrial Combustion Testing
III
Notation
Clinker K % w/w
Granulated Blast Furnace Slag S % w/w
621
Cement Kilns
Table 31.3
Table 31.4
Strength Requirements According to ENV-197-1
Classification of Pyropocesses
Compressive Strength (N/mm ) 2
Strength Class
Early Strength
Standard Strength
2 Days
7 Days
− ≥10 ≥10 ≥20 ≥20 ≥30
≥16 − − − − −
32.5 32.5R 42.5 42.5R 52.5 52.5R
Pyroprocess Type
28 Days ≥32.5
≤52.5
≥42.5
≤62.5
≥52.5
−
Wet: Wet long kiln Semiwet: Semiwet long kiln Semidry: Lepol kiln Dry long Dry: SPH/PC kiln
Raw Material
Material Moisture Content, % w/w
Slurry Filter cake
30–40 17–21
Granules Raw meal Raw meal
10–13 0–1 0–1
Table 31.5 Main Characteristics of Cement Rotary Kilns
Maximum rating Kiln length/diameter ratio Production per kiln unit volume Raw material Slurry moisture of kiln feed Kiln inlet feed decarbonation degree Flue gas exit temperature Flue gas exit flow Flue gas for raw material drying Kiln dust (CKD) Heat consumption
Units
Wet Long Kiln
Semiwet Long Kiln
Semidry Lepol Kiln
Dry Long Kiln
Dry SP/PC Kiln
(m)tpd −
2000 32:1–38:1
1200 41:1–44:1
2200 14:1–17:1
2000 32:1–38:1
12,000 11:1–16:1
tpd/m3
0.5–0.6
0.7–0.8
1.9–2.3
0.6–0.9
3.4–4.6
− % w/w
Slurry 30–40
Filter cake 10–13
Granules 10–13
Raw meal 0.5
Raw meal 0.5
%
−
−
~40
−
40–95
°C
160–200
180–200
110–120
300–350
290–340
Nm3/kg clinker −
2.8–3.2
1.8–2.0
1.8–2.2
1.8–2.0
1.3–1.6
No
No
No
No
Yes
% Mcal/t cli
8–10 1200–1500
7–14 850–900
6–9 780–900
0–5 900–1200
0 750–850
Around the time of the energy crisis of the 1970s, these kilns were traded for newer, more efficient technology and only a few have been built since. Another important disadvantage of the long wet kiln technology is the physical size limit and therefore the production limit. Due to problems linked to the sheer load and stress on the mechanically driven parts, stress on the supports, and shell deflection and therefore limited refractory life, a long kiln producing 2000 (m)tpd of clinker is the largest mechanically available. Looking at the process inside the kiln staring from the feed end, the various zones are as follows: • Drying zone: In this zone, the raw slurry is heated and the free water is evaporated. In order to increase the heat transfer between the gas and slurry, chains are hung on the inside of the kiln shell in order to increase the contact
© 2011 by Taylor and Francis Group, LLC
•
•
•
•
surface area. Some processes begin raw slurry drying before entering the kiln using thermal and/or filtration devices. Preheating zone: In this zone, the raw material is heated and prepared for calcinations with the dehydration of clay constituents. Calcining zone: In this zone, the remaining chemical bond water is released by dehydration of clay constituents and then vaporized and the raw meal is decarbonated. The liquid phase begins to form. Burning zone: In this zone, the exothermic clinkerization reaction occurs and the raw meal is transformed into clinker. Cooling zone: In this very short zone, clinker begins cooling as it is introduced into the cooler.
622
31.2.2.2 Long Dry Process Cement Rotary Kilns Popular in North America for over a century, the long dry kiln is similar to the long wet kiln. Having the advantage of dry raw materials, the long dry kiln improved the fuel consumption in clinker production process. Initially requiring water spray cooling to reduce the gas exit temperatures of more than 700°C to reasonable values, the advantage over the wet kiln process was minimal. Over the years, several improvements were made to the process including the installation of chains (as in the wet process), metallic crosses, and ceramic heat exchangers drastically improved the thermal efficiency of these kilns. All of these additions to the preheating zone (first 15–20 m) of the kiln provide an increased contact surface between the gas and the raw meal and therefore improved the heat transfer. By improving the heat recovery between the kiln flue gas and raw meal, the heat consumption was reduced by 30% with gas exit temperatures of 350°C and 400°C. With the reduction of the thermal consumption, the maximum production limit increased by 35–40% with respect to the long wet kiln. Looking to the process inside the kiln starting from the feed end, the various zones are as follows: • Preheating zone: This zone incorporates about 30% of the total length of the kiln and preheats the raw meal from 50°C to 650°C. • Decarbonation zone: This zone incorporates the next 45% of the total length of the kiln and continues the preheating of the raw meal, the chemical drying, and decarbonation with meal temperatures from 650°C to 1100°C. • Burning zone: In this zone, approximately 20% of the kiln length, the exothermic clinkering reactions take place with temperatures in the range from 1100°C and 1450°C. • Cooling zone: In the last 5% of the kiln, clinker begins cooling before it is introduced into the cooler from 1450°C and 1300°C. 31.2.2.3 Semidry Process Cement Rotary Kilns This kiln provided a major improvement in kiln thermal efficiency, reducing by as much as 50% of the heat required in the popular wet kiln process. The new technology was able to reach sizes up to 3000 (m)tpd with specific fuel consumption of 3.3 MJ/kg (800 kcal/kg). The use of a preheating grate, called the Lepol grate, significantly reduces the required residence time in the kiln, reducing therefore the L/D ratio to 12–15. The Lepol grate is covered by a uniform layer of granulated (10–13% water) raw meal (granules), 150–200 mm deep
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
and passes slowly through a single or double pass of the hot kiln exhaust gases. The kiln exhausted gases, in cross-current flow with the raw granules, transfer the majority of their heat to the raw meal by going from 1000°C to just above 100°C. The properties of the raw granules are extremely important to the efficiency of the Lepol grate. Formed in a pan granulator, the granules must be resistant to mechanical and thermal stresses as they are transported to the Lepol grate and heated to 800°C. The decomposition of these granules in the Lepol grate plugging, which increases the pressure drop over the layer and therefore reduces the flow of gas and consequently the productive capacity of the kiln system. 31.2.2.4 Dry Process Cement Rotary Kilns Using cyclones to separate raw meal from hot kiln gasses, the suspension preheater enhances the heat transfer between phases. These cyclone preheaters can have from 1 to 6 stages, depending on several factors, which include raw meal moisture content and required heat consumption; the most common number of stages is 4, which usually guarantees a gas exit temperature of approximately 350°C. The raw meal is introduced into the tower at one location into the inlet duct that splits to feed the two top cyclones. The insertion of the meal is achieved using a simple chute with a splash plate that protrudes inside the duct or a specially designed external splash box. The impact of the incoming raw meal against the plate is intended to disperse the material uniformly across the duct cross section. It is important to note that the flue gas stream exiting the lower cyclone stage travels upward through the top cyclone inlet duct to impact and to transport the dispersed material to the tower first stage. With the separation and heat transfer efficiency available in a suspension preheater, the rotary kiln L/D ratio is relatively low, around 15. Partially calcined raw meal (20%–30%) enters the kiln already heated to 800°C with the first clinkerization reactions already initiated. The suspension preheater also significantly decreases the total system residence time by requiring only 30 seconds residence time to preheat the raw meal and 30 minutes residence time in the kiln with an average kiln rotation speed of 2 rpm. Depending on the technology, the preheater has a pressure drop range from 300 to 600 daPa with optimal gas duct velocities between 15 and 20 m/s. Cristea et al. [28,29] carried out extensive experimental work using a scaled-down laboratory cyclone in order to provide effective validation data to their Computational Fluid Dynamics (CFD) 3D mathematical simulation of cyclone made with Fluent© software. Using an Euler-Lagrangian approach with Reynolds
623
Cement Kilns
Stress Model (RSM) turbulence model was one of the pioneering attempts to simulate a dust-laden swirled flow inside a cyclone. • Suspension Preheater Rotary Kilns with Riser Duct Firing: In order to improve the heat transfer efficiency of a suspension preheater, some of the total heat can be shifted to the riser duct (connection between the kiln and the preheater). This increases the decarbonation degree along with marginally increasing the total production rate, obviously within other process limitations. The CO2 released in the calcination process can cause a temporary increase in suspension and can subject the kiln to material flushing. Increasing the stability of the material suspension by increasing the calcination degree to 50% reduces this problem. Displacing heat from the main burner to the riser duct also helps alleviate thermal stresses on the kiln shell and refractory lining therefore increasing their life span (see Figure 31.2). With this type of kiln system, a higher quantity of excess air is required at the kiln exit in order to burn the additional fuel in the riser duct, but this then fosters the ideal conditions for alternative and waste fuel use. The limit on heat in the riser duct is limited by several factors including duct geometry, system design, and fuel type, and is generally limited to 25% of the total heat. This limit stems from the maximum quantity
of excess air that can be provided in the kiln without destabilizing the burning zone and decreasing the temperatures. In cases where the excess air is critical, the burning zone can be stabilized by oxygen enrichment or oxy-combustion. • Suspension Preheater/Precalciner Rotary Kilns: In order to increase the efficiency of the suspension preheater system, hot air is taken off from the cooler and sent directly to a specially designed combustion vessel, called the calciner. By providing air directly to the calciner and not as excess air in the kiln, 60% of the total heat can be burnt in the calciner, allowing more than 90% calcination of the raw meal before entering the kiln. The relatively low temperatures in the calciner (800°C–900°C) indicate that a conventional flame with a defined shape and geometry is not possible. Several factors affect the calciner efficiency, especially the uniformity of air flow, raw meal, and fuel dispersion. With a preheater/ precalciner kiln system, a nominal production of more than 10,000 tpd clinker can be reached with as little as 3 MJ/kg (700 kcal/kg). The first calciners were designed with a fuel residence time of 2–3 seconds for conventional fuels (coal, HFO, natural gas). With the current trend leaning toward AF, larger calciners with higher residence times provide the flexibility requested for low grade
Flue gas Raw meal 1st stage
Suspension preheater 2nd stage
3rd stage 4th stage
5th stage
Riser duct
Primary air Fuel
Fuel Cooler Cooling air Figure 31.2 Suspension preheater with rise duct firing.
© 2011 by Taylor and Francis Group, LLC
Exhaust air Clinker
624
Industrial Combustion Testing
Flue gas Raw meal 1st stage
Suspension preheater 2nd stage
3rd stage 4th stage
5th stage Precalciner
Tertiary air
Primary Exhaust air air Fuel
Fuel Cooler
Clinker
Cooling air Figure 31.3 ILC: in-line separate calciner.
fuels, requiring five seconds residence time, and also help to solve some of the combustion problems presented in the earlier generations of calciners. As the nominal production capacity increases, preheater/precalciner kilns can be found in many different configurations consisting of one to three preheater tower strings and in-line (IL) (see Figure 31.3) or separate line (SL) calciners (see Figure 31.4). Most recently, a precombustion chamber has been added to enhance the calciner performances. In-line calciners burn fuel in a mix of kiln exit gas and tertiary air, while separate line calciners burn fuel in pure tertiary air—providing an environment generally more adept in burning difficult fuels. Due to the tertiary air requirement of precalciner kiln systems, a grate cooler is a requirement as planetary coolers do not provide tertiary air. Due to the high calcination degree achieved in the calciner, the kiln L/D ratio is reduced to 10–14 with higher kiln rotation speeds, averaging 3.5 rpm. With these parameters, a typical kiln residence time is then reduced to 20–25 minutes. 31.2.3 Rotary Kiln Design Parameters There are typical parameters for the rotary kiln design and operation and they are considered critical in optimizing the performance of a kiln itself. When designing a cement rotary kiln for a certain nominal capacity, or when evaluating an existing one for potential output, there are a number of key parameters that must be evaluated. These
© 2011 by Taylor and Francis Group, LLC
include: burning zone heat loading, secondary/tertiary air velocity, burning zone flue gas velocity, kiln exit gas velocity and temperature, preheater tower gas velocities and pressure drops, preheater tower exit gas temperature, material residence time, clinker cooler grate loading, cooler air supply, kiln dust cycles, and so forth. 31.2.3.1 Material Residence Time The residence time of material in the rotary kiln is determined based on the kiln slope, the speed of rotation, kiln diameter, and by the presence of any internal restrictions either by design or due to ring formation in operation. The residence time, τ expressed in minutes, can be calculated from the formula first given by the U.S. Bureau of Mine:
τ=
1.77 × L × α × F , p×D×n
(31.6)
where L = the kiln length, m p = the kiln slope, degree D = the kiln diameter (inside refractory lining), m n = the kiln rotation speed, rpm α = the angle of repose of material (40° for raw material) F = the constriction factor (usually 1 if no dams, lifters, etc.)
625
Cement Kilns
Flue gas Raw meal 1st stage
Suspension preheater 2nd stage
3rd stage 4th stage
5th stage
Fuel Down draft precalciner
Tertiary air
Primary Exhaust air air Fuel Cooler
Clinker
Cooling air Figure 31.4 SLC: separate line calciner–down draft.
31.2.3.2 Rotary Kiln Degree of Fill Rotary kiln volume load is the percentage of the kiln cross-sectional area filled by the kiln charge, and is usually in the range of 6–16% for most rotary kilns. It should be noted, though, that a fill degree of more than 13% could impair heat transfer in that some of the material in the center of the charge will not be exposed to enough heat. It is sometimes seen that a kiln ring could coincide with high or erratic free lime in the clinker, possibly because the fill degree has exceeded limits for ensuring that all kiln charge material is uniformly heated and burned.
above could be the limitation to a kiln’s output. These limitations will typically manifest themselves as kiln instability and ring or coating build-ups, excessive dust loss, poor refractory lining life, poor clinker quality, or high fuel consumption. Usually the kiln output limitation is dependent upon the same equipment performances such as ID fan capacity, burner thermal power, or grinding capacity of raw materials or coal. The kiln production can be calculated using different empirical formulas, such as:
31.2.3.3 Rotary Kiln Slope Rotary kilns slope from the feed end to the discharge end enabling the material to travel in that direction by gravitational force. The slope is typically 2–4% (1.2° to 2.3°) and is decided in conjunction with the kiln rotational speed. A smaller slope with a higher rotational speed may improve heat transfer because of the greater tumbling of the material. 31.2.3.4 Rotary Kiln Capacity There are design limits for all of the above that may vary between different processes, but any of the
© 2011 by Taylor and Francis Group, LLC
P = 148 × n × D3 × fdeg × ρm × tan β,
(31.7)
with the material particle helical trajectory having the slope given by
tan β = sin p / sin α ,
(31.7a)
where P = the kiln production, kg/h fdeg = the degree of fill of rotary kiln section, percentage ρm = the material density, kg/dm3.
626
31.3 Conventional and Alternative Fuels The cement industry worldwide is very energy-intensive, with fuel cost typically accounting for 30–40% of OPEX (operation expenditures). For this reason the thermal energy efficiency has improved significantly over the last decades and the AF have been used to substitute the conventional fuels in order to further reduce costs. The use of AF in cement rotary kilns is widespread either in Europe or United States with increased usage in other parts of the word such as South-East Asia, Japan, or Australia. The cement rotary kiln can be fired on conventional fuels such as pulverized fossil coal, HFO or natural gas and in combination with AF in co-firing mode. The gasification of the solid waste fuels has been introduced only at a limited number of cement plants. The individual cement rotary kiln characteristics should always be considered when the utilization of AF is foreseen. There is always an impact on kiln production and emissions as well as on the ring and builds up formation, then on the kiln running factor. 31.3.1 Conventional Fuels The fossil fuels (i.e., coal, oil, and gas) are a nonrenewable source of energy. The fuel selection for cement kilns was primarily determined by the cost of exploitation and its availability of the market. The choice of fuel is essential because: • The cost of fuel represents about 30–40% of production cost of the cement. • The fuel must assure trouble-free kiln operation conditions. • Environmental impact (level of NOx, SOx, and volatile organic compounds (VOC) emissions) and flue gas chemical composition. • Its characteristics in terms of: low calorific value (LCV) flammability limits, adiabatic flame temperature, and its flame emissivity. • Volatile matter and ash content as well as ash chemical composition. The operational acceptability of a fuel especially for the pulverized coal or heavy liquid fuels depends on storage, handling, dosing, and combustion behavior, the extent to which the fuel combustion products contaminates the product. 31.3.1.1 Solid Fuels Pulverized coal (PF) firing has many advantages since the coal is generally cheaper then either HFO or natural gas. The pulverized coal firing systems are more
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
hazardous then other fuel systems; the design of these systems is regulated by few codes, the most relevant code is issued by NFPA, namely 8,503–1997: Standard for Pulverized Fuel Systems. In last 20 years the suitability of petroleum coke for the cement rotary kilns has been broadly demonstrated. Petroleum coke is a carbonization product of high-boiling hydrocarbon fractions obtained in petroleum processing (heavy residue) and is extensively used in the rotary kilns. There are two kinds of petroleum coke • green delayed coke • fluid coke Since they contain the impurities from original crude oil, the sulfur is usually high and appreciable vanadium salts may be presented. Some main characteristics of the two types of petroleum coke are shown in Table 31.6. Table 31.7 summarize the main chemical and physical properties of the green delayed petroleum coke in terms of ultimate and proximate analyses, inferior calorific value, free swelling index, HGI, and ash fusion temperature. The Hardgrove Grindability Index (HGI) is an empirical measure of the work required to crush coal based on the generation of #200 mesh coal on air—dried test sample ground in a small ball-and-race mill for 60 rpm. The HGI of 58 is located at the top side of the scale showing the low hardness value of the petroleum coke. The free swelling index FSI, which is a measurement of the increase in volume when coal is heated without restriction, gives an indication of plastic and caking properties. The green delayed petroleum coke does not swell. Under reducing conditions the ash fusion temperatures, mainly the softening temperature of 1316°C, give an indication on fouling and slugging behavior. The matching of coal rank, with the combustion equipment and furnace designs is based primarily on standard analytical techniques on laboratory techniques, which have been developed by the manufacturers and on firing in the test furnace simulating industrial conditions. Table 31.6 Properties of the Typical Petroleum Coke Composition and Properties Volatile matter, %wt. Ash, %wt. Sulfur, %wt. HGI Bulk density, kg/m3
Green Delayed Coke
Fluid Coke
8–18 0.05–1.6
3.7–7.0 0.1–2.8 1.5–10.0 20–30 1500–1600
40–60 (80) 1280–1420
627
Cement Kilns
Table 31.7 Chemical and Physical Characteristics of Petroleum Coke Sample Ultimate Analysis Petroleum Coke
Elemental Analysis
C H N S O (by diff.) Ash Moisture Low calorific value (v = const.)@ 25°C ASTM 2015 DIN 51,700 Free swirling index FSI Ash fusion temperatures Reducing atmosphere, °C
%w/w wet 83.99 2.68 1.43 4.79 0.00 0.29 6.87 38.18 MJ/ kg wet 7687 kcal/ kg wet 0.5 Initial deformation 1293
%w/w dry 90.19 2.68 1.54 5.09 0.00 0.31
The above characteristics are broadly related to coal rank and are experimentally measured, by laboratory analyses based on procedures established by standard associations. The process by which ignition of coal particles into the flame occurs involves a number of steps, which can be extended to the cloud of particles (jet):
1. Initially the PF is heated up by convection of the furnace flue gas and also by radiation from hotter flue gas and material bed. 2. At the temperature above 400°C, the particle starts to decompose and evolves a mixture of volatile matter such as CO, H2, and CnHm as well as in-combustible gas such as CO2 and H2O vapor. 3. At the temperature over 800°C, most of the volatile matter has been evolved and its combustion will sustain the ignition of flame.
© 2011 by Taylor and Francis Group, LLC
90.47 2.89 1.54 5.11 0.00 − − 38.07 MJ/kg d.a.f.
1316
• heat calorific value • grindability and abrasion, which are important in the grinding of coal to the PF • combustibility, which is traditionally characterized by VM (volatile matter) of proximate analysis and the reactivity of the resultant char, and may be related to the petrographic composition of coal • ash properties and chemical composition
%w/w wet
%w/w d.a.f.
− 34.56 MJ/ kg dry 8254 kcal/ 9,104 kcal/ kg dry kg d.a.f. Hardgrove grindability index HGI Softening Spherical
The four most important characteristics in pulverized fuel (PF) combustion are:
Proximate Analysis
1338
VM FC Ash Moisture
10.79 79.17 0.84 9.23
%w/w dry 11.89 87.19 0.93 −
%w/w d.a.f. 12.00 88.00 − −
58 Flow 1449
4. The char residue remaining after devolatilization is than burnt relatively more slowly in the flame.
The char combustion is a much slower process than devolatilization and it therefore determines the time for complete combustion in the furnace. It is generally accepted that the evolved volatile matter then ignites followed by ignition of the residual char at temperatures determined by its reactivity. The burning rate of chars at a pressure of 1 atm for petroleum coke and the different coal rank has been compared by Sergeant and Smith [27]. The collected experimental data show that the char burning rates expressed in (kg m−2 s−1) is slightly lower for the same particle temperature, for petroleum coke against anthracite; however the behavior of both coke residue (petroleum coke) and char (anthracite) is similar. The burning profile technique is a method for predicting the relative combustion characteristics of fuels. • The thermogrammetry is an experimental technique for analysis heat effects (i.e., endoand/or exothermal). The differential thermal analysis (DTA) provides information on heat effect from a qualitative point of view. • The thermogravimetry is an experimental technique based on registration of loss in weight carried out for isothermal or temperature/time programmed conditions and can be displayed in cumulative Thermal Gravimetric Analysis (TGA) or derivative Differential Thermal Gravimetry (DTG) forms.
628
31.3.1.2 Gaseous Fuels The gaseous fuels can be broken down into the following groups based on how they are found in the nature and the type of industrial manufacturing process.
• Fuel gases naturally found in nature. a. Natural gas: Is the term for gases found in oil and coal fields; it mainly consists of methane. The natural gas has a low C/H ratio and burns easily and cleanly. There are different types of natural gas from oil fields (mostly methane): dry or lean (high in methane); sour/sweet: high/low in hydrogen sulfide; wet: high in higher hydrocarbons (C5–C10). b. Methane from coal mines: Natural gas is also found in coal fields (and it is often known as firedamp). • Manufactured gases: Gases may be manufactured from either a liquid or solid feed stocks. • Fuel gases derived from solid fuel. a. Gases derived from coal. b. Gases derived from biomass and waste. c. Gases as by-products from industrial processes (blast furnace gas). • Fuel gases made from crude oil. a. Liquefied petroleum gas (LPG). b. Refinery gases. c. Gases from oil gasification.
Liquefied petroleum gas (LPG) include fuels and refined feed stocks that exist as liquids when stored under pressure, and are vaporized for use as fuels or for synthesis reactions. Examples of common LPG’s include: • Propane–Propylene grades: Commercial (i.e., residential fuel) grade liquid propane (LPG) is predominantly propane and/or propylene. Ethyl mercaptan or thiophane (tetrahydrothiophene) are typical LPG odorant additives. • Butane–Propane (B–P) mixtures: Commercial grade liquid B–P mixtures are predominately used as fuels in climates and applications that do not experience low ambient temperatures. • Butane–Butylene grades: Commercial grade liquid butane is predominantly butane and/ or butylene mixtures that are used as fuels in industrial applications or as residential fuels. Gases from some fermentation process: Gas from the digestion of biomass by bacteria (landfill gas,
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
digester gas). Action of certain bacteria on biomass in the absence of oxygen breaks down the H/C in organic compounds. Synthesis gases: The fuel gases that are chemically made by synthesis process are denoted synthesis gases. The main synthesis is summarized here below: • Producer gas: The gas is produced by blowing an air stream and sometimes a steam stream through an incandescent fuel bed (the process is self-heating). The reaction with air is exothermic, but if insufficient air is added to the synthesis process hence CO is produced. Steam addition results in the formation of hydrogen by the water gas reaction. The gas is a LCV gas mixture produced by the incomplete combustion of coal or coke in an air or air/steam based gas producer unit (inert gases are great than 50%). Efficiency of cracking is of order 75%. The typical composition is: H2 (5–10% v/v), CO (20–35% v/v), N2 (55–65% v/v), and 4% v/v CO2 (equilibrium). Wobbe index is of order 5-6 MJ/ Nm3. • Blue gas or water gas: This is produced in a similar manner to the above but allows the production of a higher LCV fuel by intermittently blasting the incandescent coke bed with air and steam such that the overall heat balance is maintained. The products of the air blast contain the nitrogen that reduces LCV. The products of the steam blast are kept since they have a higher LCV. Efficiency of cracking is of order 62%. Typical composition: CO2 4.7% v/v, CO 41% v/v, H2 49% v/v, CH4 0.8% v/v, N2 4.5% v/v. Wobbe index = 15.64 MJ/Nm3. • Carburetted water gas: Is the result of combining the water gas and oil gas methods. Oil is sprayed into a brick lined chamber during the blow period of the BWG plant with air. This heats the bricks to around 1000°C. The ratio of the two determines the quality. Efficiency of cracking is of order 75%. Wobbe index of the resultant gas of order 24.9 MJ/m3. • Coke and coal oven gas: The coke oven gas is a result of secondary cracking reaction in coke ovens, to which the gases and vapors of carbonization are subjected and hence hydrogen forms its principal components. The typical composition is: H2 (45–54% v/v), CH4 (28–34% v/v), CO (6–7% v/v), N2 (5–6% v/v), CO2 (2–2.5% v/v), and so on. The corresponding gross calorific value (GCV) (saturated) is about 5000 kcal/Nm3.
629
Cement Kilns
31.3.1.2.1 Wobbe Index The flow of gaseous fuel through a control valve is similar to the flow through an orifice plate of the same cross area. The heat release is calculated knowing the GCV of the gas and the volume flow rate. The pressure drop across the orifice plate is calculated from Bernoulli’s equation. The combination of these relations gives the heat release in terms of GCV, pressure drop, density, and area. If a fuel gas is substituted on the same burning installation (i.e., area and pressure drop remain the same) the new heat release rate may be calculated as the ratio of the GCV multiplied by the square root of gas density: Q1/Q2 = (GCV/√ρ1)/(GCV/√ρ2) = (GCV/√SP1)/ (GCV/√SG2) (where SG is the specific gas density relative to air density). Thus, a number may be derived that gives an indication of the interchangeability of the gases, the Wobbe number (Wo = GCV/√SG). In practice, the specific gravity with relation to air is used instead of density. 31.3.1.2.2 Weaver Index A second index is used to define the propensity of the gas to react. This is called the Weaver flame speed factor. It is defined as the ratio between the laminar flame speed of the gas with relation to hydrogen. Thus hydrogen is considered to have a value of 100. The lower the number the lower the flame speed. Weaver speed factor is greatly influenced by the amount of hydrogen in the mixture. In general, the flame speed depends upon the kinetics of combustion and hence varies from gas to gas. The H2 flame is the fastest with the normal velocity having the maximum value of about 2.5 m/s. The CO flame has a lower normal velocity in the range of 0.3 to 0.4 m/s. The flame speed rises if the initial temperature of the combustible mixture is increased or the pressure of the system decreased. If the Wobbe number and Weaver flame speed factor are identical for two gases they are completely interchangeable. Fuel gases are classified according to Wobbe and Weaver numbers. The International Gas Union (IUG) assigns the following gas families [38]: Family 1: Wo = 17.8–35.8 MJ/Nm3/4250–8550 kcal/ Nm3 (COG) Family 2: Wo = 35.8–71.5 MJ/Nm3/8550–12824 kcal/Nm3 (natural gases, town gas) Family 3: Wo = 71.5–87.2 MJ/Nm3/17,072–20,823 kcal/Nm3 (LPG) and gas groups:
© 2011 by Taylor and Francis Group, LLC
High flame speed gases: We = 32 to 45 (LCV = 4000–5000 kcal/Nm3) Intermediate speed: We = 25 to 32 (LCV = 5500– 7500 kcal/Nm3) Low flame speed: We = 13 to 25 (LCV = 7500–10,000 kcal/Nm3). 31.3.1.5 Liquid Fuels The main advantage of liquid fuels over their gaseous fuel alternatives is the extremely high energy density (thermal energy/volume ratio). Typically they would have a GCV of 40 MJ/kg, which in volumetric terms is equivalent to 33,000 MJ/Nm3 compared with natural gas at about 34 MJ/Nm3. Heavy fuel oils are blended products based on the residues from various refinery distillation and cracking processes. The HFO is a general term but other names are used as well to describe this range of products: residual fuel oil, bunker fuel, bunker C, fuel oil No. 6, industrial fuel oil, marine fuel oil, and black oil. The liquid fuels can be classified as heavy, medium, and light fuel oils to give a general indication on the viscosity and density of the products. Heavy fuel oil (HFO) consists primarily of the residue from distillation or cracking crude oil units in the refinery. Historically, the fuel oils are manufactured based on long residues from the atmospheric distillation column; they were known as straight run fuels. Nevertheless the increasing demand of transportation fuels such as gasoline, kerosene, and diesel oil has generated higher atmospheric residue as a feedstock for vacuum distillation and cracking manufacturing processes. Today most HFOs are currently based on short residues and residues from thermal and catalytic cracking manufacturing processes. These liquid fuels are different from the straight run fuels due to the fact that the density and the mean molecular mass are higher. The density of some HFOs can be higher then 1000 kg/m3. The composition of the residual fuels varies widely depending on the refinery process configuration and crude oils characteristics. The residual fuel oils are complex mixtures of high molecular weight compounds with the typical boiling point ranging from 350°C to 650°C. They consist of aromatic, aliphatic, and naphtenic hydrocarbons (carbon numbers from C20 to C50) mixed with asphaltenes and a smaller amount of compounds containing S, N, and O. Asphlatenes are highly polar aromatic compounds with very high molecular mass (2000–5000 kmole/kg); then in the blending of the HFO the asphlatenes must remain in suspension. The HFOs also contain organometallic compounds from their presence in the virgin crude oil. Such trace metal is Va. Other elements that occur are Ni, Fe, K, Na, Al, and Si. Significant concentrations of H2S (hydrogen
630
sulphide) are accumulated in the top space of the storage tanks. Appreciable concentrations of polycyclic aromatic compounds (PAC) can be presented in HFOs depending on the amount of low viscosity diluent used. The physical and chemical properties of liquid fuel oils are provided by different standards: either ASTM D-396 (ASTM-1992) or ISO 8217 (ISO 1996). • Viscosity/pour point: Is a measure of HFO resistance to flow. They have high viscosity (depending on the grade) and often require heating for handling and combustion. The minimum heating temperature is defined by the pour point. The viscosity is not an additive property; therefore oil blends have to be determined experimentally. • Flash point: Is defined as the minimum temperature at which the given fuel oil evolves just sufficient vapor to form an inflammable mixture with air. • Sludge formation: Some oils, particularly cracked oils, do not store well and oxidizable components result in sediment or gum formation that blocks filters. • High C/H ratio: Resulting in sooty, highly luminous flames. This is an advantage for heat transfer in many industrial applications, such as cement kilns. • Sulfur: Residual oils have S contents between 0.2 and 6% w/w. This increases the dew point and corrosion and accelerates gum formation. • Carbon residue: When an oil is heated, both processes of evaporation and creaking are set. The heavier and complex compounds form by decomposition carbonaceous deposit. To determine this residue the Conradson test is used. • Ash: Maximum amount usually present is 0.2%. The content is important. High vanadium ashes have reduced melting temperature and can build up as deposits and attack refractory lining. • Vanadium/Sodium: These compounds in ash cause eutectic compounds with very low melting points (as low as 600°C). This results in a liquid, sticky ash that congeals onto the heat exchanger surfaces resulting in faster build up, corrosion, and problems. 31.3.2 Alternative Fuels Many AF, ranging from hazardous to nonhazardous in solid, liquid, or gaseous phase, are used to be co-fired in the cement kilns as is showing in Table 31.8. The overall
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
Table 31.8 Alternative Fuels Used in Cement Manufacturing Solid Alternative Fuels TDF (tires derived fuel) Meat and bone meal Sewage sludge RDF (refuse derived fuel) Saw dust (waste wood) Rice husks Textiles Bagasse (grinded sugar canes) Coconut, coffee shell MSW (municipal solid waste) Rubber/plastics/ paper waste Hospital waste ASR (Auto Shredded Residue) E&EE (Electric & Electronic Equipment)
Liquid Alternative Fuels Tar Waste oil Paint residue Solvents and waxes Oil sludge Palm kernel oil
Gaseous Alternative Fuels Pyrolosis gases
substation rate of the fossil fuels with AF are very different in EU countries depending mainly on local regulations, local authorities attitude on this topic, and availability of AF on the local market. The utilization of the alternatives fuels through the main burner pipe of rotary kiln section is characterized by some limitation factors: • The LCV of the AF as well as an adiabatic flame temperature influences the heat transfer from the flame to the material. • Affect the emissions of the rotary kiln. • The water content influences the rotary kiln production. • The minor components affect the clinker chemistry and rotary kiln operation. When assessing the Best Available Techniques (BAT) for the usage of different types of fuels, it was found that some waste fuels such as tires are considered a mature technology, but other solid waste have been only recently introduced as AF. Tires are the most geographically and widely used solid waste derived fuel; they can be used in different forms such as entire, chipped, or shredded. The utilization of tires gives complete destruction of the rubber and cotton, which are components of tires. The ultimate and proximate analyses of some AF are displayed in Table 31.9.
631
Cement Kilns
Table 31.9 Ultimate and Proximate Analyses of Some Alternative Fuels UM Carbon Hydrogen Oxygen Nitrogen Chlorine Sulfur Moisture Volatile matters Ash Low calorific value HGI: Hardgrove grinding index P2O5 Hg Cd + Tl Σ(Sb, As, Pb, Cr, Co, Cu, Mn, Ni, V)
%w/w %w/w %w/w %w/w %w/w %w/w %w/w %w/w %w/w kJ/kg
%w/w %w/w ppm ppm
Petroleum Coke
Tires
RDF Fluff
Sewage Sludge
Animal Meal
86.36 3.49 1.85 1.61 0.01 4.68 0.11 10.49 1.28 34,830 50
81.00 6.70 3.00 0.30 0.10 1.70 1.00 61.00 7.50 29,480
53.08 7.26 19.50 0.47 1.20 0.80 17.80 65.00 17.69 14,650
26.60 4.93 16.00 5.73 0.05 0.46 6.60 45.30 46.20 9,849
43.80 5.30 16.90 8.90 0.60 0.50 3.40 68.70 20.50 19,990
<0.01 <10 <3000
<0.01 <20.00 <2000
<1.00 <20.00 <2000
<10.00 <5.00 <10.00 <5000
<5.00 <0.10 <10.00 <1000
31.3.3 PF Firing System Classification Coal
Pulverized coal grinding and firing systems are classified as follows: • Direct firing systems: All systems in which the pulverized coal is fed directly to the rotary kiln downstream of the coal pulverizer (mill) without any intermediate storage. • Indirect firing systems: All systems in which the pulverized coal is stored after the grinding and metered to the rotary kiln from an intermediate surge bin. 31.3.3.1 Direct Firing Figure 31.5 shows the firing system in which the PF is transported from the grinding department using the mill exhaust gas, which acts as burner pipe primary air. 31.3.3.2 Semidirect Firing Figure 31.6 displays the firing system in which a proportion of mill exhaust gas is recirculated back to the mill inlet, thus to increase the transport air in the mill and its dynamic classifier without increasing the primary air to the burner pipe. 31.3.3.3 Indirect Firing Figure 31.7 presents the firing system in which the PF is separated from the mill exhaust gas and it is stored
© 2011 by Taylor and Francis Group, LLC
Coal + gas Primary air fan Ball mill
Hot gas
Figure 31.5 Direct grinding circuit with horizontal ball mill.
in a surge bin. The mill gas is exhausted to the atmosphere through the integral precipitation baghouse filter. The hot gas for the grinding process provided by the suspension preheater (SP) tower of high volatile coal is used or hot exhausted air from the cooler if petcoke or low volatile coal are used. The relative advantages of an indirect firing system against a direct firing system focus on the consistency of PF feeding to the burner, the effect of primary air quantity on the process and emissions, and the effect of water vapor. For safety, the indirect firing system needs more safety measures to be implemented to reduce the risk of fire or explosion to an acceptable level by installation of the explosion doors and inertization system with carbon dioxide or nitrogen. The pulverized coal fed to the burner should be very steady with low fluctuation of set flow rate of the metering system. The variability of raw coal in terms of calorific value is mainly due to the fluctuation in moisture
632
Industrial Combustion Testing
Clean gas Coal + gas Coal
Coal dosing bin Hot gas
Cycle fan
Cement rotary kiln
Primary air fan Recycle Figure 31.6 Semidirect grinding circuit with horizontal ball mill.
Baghouse filter Coal + gas
Waste fan
Coal
Fine coal Hot gas Recycle Primary air fan Figure 31.7 Indirect grinding circuit with vertical meal.
and ash content. The coal moisture removed during the grinding process in the case of indirect firing system is exhausted to the atmosphere from the mill in the grinding department, while in the other two systems the moisture is sent to the rotary kiln.
kinetics; fluid mechanics, and transport phenomena (mass and heat transfer). Most of industrial conventional fuels are hydrocarbons, because their primary elemental components are carbon and hydrogen and sometimes sulfur. The chemistry of these oxidation reactions are involving very complex chain reactions. Nevertheless in industrial application for burning equipment design a so-called simplified chemistry is used; for such a simple chemically reacting system, a simple combustion calculation method is normally applied. Both the chemical kinetics and physical diffusion process control the rate of reactions (combustion process). In general a combustion process is undergoing four different steps: mixing (oxidant and fuel reactants), ignition, chemical reactions, and combustion products release. The rate of combustion is depending on the time scale of the above-mentioned steps; generally speaking in the industrial combustion processes the mixing is the slowest process then the mixing time scale is controlling the industrial diffusion flames. Three main approaches were adopted to predict the industrial diffusion flame characteristics: • Semiempirical correlations • Mathematical modeling techniques by using the CFD • Physical modeling techniques
31.4 Flame Aerodynamics in Rotary Kiln Section Combustion is the study of chemically reacting flows with rapid, highly exothermic reactions, or combustion physics is the science of burning. Combustion is an interdisciplinary science, in nature comprising: thermodynamics, thermochemistry and stoichiometry, chemical
© 2011 by Taylor and Francis Group, LLC
It is not the scope of this chapter to fully present the jet aerodynamics applied to combustion engineering, nevertheless for understanding the cement rotary kiln combustion aerodynamics some specific arguments need to be introduced. The reader can also refer to Stambuleanu [11], who offers an exhaustive presentation of this subject.
633
Cement Kilns
31.4.1 Free Turbulent Jets 31.4.1.1 Isothermal (Constant-Density) Conditions The simplest example of a free turbulent round jet is provided by a nonreacting (isothermal) flow stream expanded at high velocity from a cylindrical nozzle of diameter d. It is assumed that the turbulent round jet flows into a quiescent ambient of the same fluid, as shown in Figure 31.8. The flow is statistically stationary and axisymmetric. Hence, statistics depend on the axial and radial coordinates (x, r), but are independent of the circumferential coordinates (ϑ) and of time. Using the cylindrical coordinates (x, r, ϑ), the axial velocity along the centerline is denoted by um, and the mean velocity leaving the nozzle is u0. The flow is completely defined by u0, d, and ν, cinematic viscosity and hence the only dimensionless parameter is the Reynolds number defined by Re = u0 d/ν. In this case the other velocity components, namely radial v and tangential w doesn’t exist. In the potential core zone the velocity and concentration of the nozzle jet remain unchanged, but outside this zone the entrainment and the mixing occur by turbulent diffusion. The following relationship holds for distances beyond five diameters from the nozzle exit: −1
um x = A , d0 uo
(31.8)
where A is a constant.
Thus, the centerline velocity decay is inversely roportional to the distance downstream of the nozzle p exit and it is assumed that the static pressure remains constant and then the momentum is conserved. The mean radial velocity profile has a Gaussian distribution as follows: r 2 u(r ) = exp − k , um x
(31.9)
where k has values in the range of 82 to 92. The concentration or the temperature decay can also be represented by an expression similar to the formula (2), but with k constant varying from 54 to 57. The axial velocity decrease is due to entrainment from the surrounding fluid. The entrainment is an inherent characteristic of the jet flow. The entrained fluid influences the fluid at the centerline of the jet at a distance equal to five times the jet diameter. The angle of jet is approximately 19°–22° and it is independent of uo. According to Ricou and Spalding [3], the rate of e, in the entrainment of the surrounding environment, m fully developed zone, is given by: e m x = 0.32 − 1, o m do
(31.10)
where e = mass flow rate entrained, kg/s; m o = mass flow rate from nozzle, kg/s; m do = nozzle diameter, m.
Um Uo 1
5
x d0
10
O
d0
Nozzle 0
U0 Potential core
Mixing zone Um
Transition zone Um
Fully developed zone CL
X
α = 19°–22°
r
Figure 31.8 Free jet entrainement. (From Kutz, M., Mechanical Engineer’s Handbook, 3e, Book 4, Energy & Power, New York: John Wiley, 2006. With permission.)
© 2011 by Taylor and Francis Group, LLC
634
Industrial Combustion Testing
This equation is valid for an Re number greater than 2.5 × 104 and the ratio x/d0 > 6. The mass flow rate in each cross section of the jet is written:
x = 2π m
∞
∫
0
ρurdr.
(31.11)
The axial flux of axial momentum of the free jet is conserved, if no external forces act on it, and the momentum can be expressed by:
Gx = 2 π
∞
∫ ρu rdr.
(31.12)
2
0
At the nozzle exit the axial momentum flux is given by:
(31.13)
The relationships developed for the isothermal conditions for the turbulent free jet can be extrapolated to the chemically reacting flows by using the concept of the equivalent nozzle diameter d0ech. This is defined as the diameter of the nozzle through which the entrained fluid of density ρe would expand having the same mass flow rate and momentum as that of the nozzle jet initially considered (Table 31.10). Since u0 = u0eq, the equivalent nozzle diameter may then be written as: ρ d0eq = d0 0 ρe
0.5
.
(31.14)
If the mass flow rate and the momentum flux through the nozzle are available, the formula can be rewritten as follows
d0eq =
0 2m
(G0 πρe )0.5
.
(31.15)
Table 31.10 Mass Flow Rate and Momentum Flux for Nozzle and Equivalent Nozzle Nozzle (Isothermal Conditions) Mass flow . rate mo Momentum flux Go
πd02 ρ0 u0 4 . m0u0
© 2011 by Taylor and Francis Group, LLC
e m ρ = 0.32 e o ρo m
0.5
x − 1, do
(31.16)
where ρe = density of entrained fluid, kg/m3; ρo = nozzle density, kg/m3; e = mass flow rate entrained, kg/s; m o = mass flow rate from nozzle, kg/s; m do = nozzle diameter, m. This relationship holds for all Re > 2.5 × 104 and for x/ do > 6. 31.4.2 Confined Turbulent Jets
πd 2 Gx = G0 = 0 ρ0u0 . 4
31.4.1.2 Nonisothermal (Combustion) Conditions
The more general relationship of the rate of entrain e, for chemically reacting flows is given by the ment, m formula:
Equivalent Nozzle (Nonisothermal Conditions) πd02eq
ρ0 u0 4 . m0u0eq
In the case of the cement rotary kiln, the fluid flow through the kiln freeboard comes from several sources such as secondary hot combustion air, combustion and decarbonation products and inleakage air. In direct firing kilns, the pulverized solid fuel is injected through the burner pipe nozzle with the external diameter in the range of 250 to 600 mm into the kiln with an internal diameter of 2.4 to 6.2 m. In this condition the fuel stream emerges as a jet, and the secondary hot combustion air as well as the inleakage air behave as the surrounding flow. The freeboard flow patterns in the burning zone behave as jets from the entrainment and mode of mixing with the surrounding fluid point of view. The confined jets have been thoroughly studied by Thring and Newby[5], as well as at the International Flame Research Foundation (IFRF) at IJmuiden-Holland. Thring and Newby proposed a simple theoretical treatment of the problem based on the assumption that the rate of the entrainment of the jet is unaffected by the enclosure, but the spread out of the jet is a function of its momentum flux. Craya and Curtet [4] and Barchilion and Curtet [10] have further extended these works to making them more generally valid. The studies initiated by Thring and Newby [5] were focused especially toward carried out experimental works on combustion chamber models and toward the similitude between model and prototype of combustion chamber. When a turbulent jet is confined by an enclosure two different situations are to be considered: A sufficient quantity of surrounding flow is supplied to the enclosure, then the turbulent jet still entrains ambient fluid until it spreads out to impinge the enclosure’s walls.
635
Cement Kilns
The surrounding flow is less than the amount that can be entrained. In this last case, when the turbulent jet issues from the nozzle in the enclosure there will always be a recirculation flow around the jet until impinges the enclosure’s walls.
1. Nonisothermal conditions In the case of chemically reacting flows the concept of the equivalent diameter: ρ d0eq = d0 0 ρe
In all other cases, the existence of external recirculation will depend on the geometry of the enclosure and the supply of surrounding flow. In the case of a sufficient amount of surrounding flow, the confined jet behaves as a free jet.
0.5
,
(31.20)
can be reutilized, then the Thring–Newby [5] parameter becomes: ϑ=
0 +m e d0eq 0 +m e 1 m m or ϑ = . 0 m 2L (G0 πρ f )0.5 L
31.4.2.1 Confined Jet with Sufficient Amount of Surrounding Flow
The Thring–Newby [5] dimensionless parameter ϑ is a similitude parameter, which can be used for nozzle/ enclosure (combustion chamber) scale up purposes. This is a typical situation of the rotary cement kilns where the ratio d0eq/2L < 0.02.
1. Isothermal conditions The Thring–Newby [5] parameter is expressed by the formula: ϑ=
0+m e d0 m . 0 2L m
(31.17)
The mass flow rate of entrained surrounding e up to the point where the jet hits the wall m and completely fills the combustion chamber is obtained from:
e 0.9 m = − 1. 0+m e ϑ m
(31.18)
If that is not the case then for an isothermal confined jet the Thring–Newby [5] parameter acts as a similitude criteria to scale up from model to prototype: +m e d0 m ϑ= 0 0 2 L model m
+m e d0eq m . = 0 0 2 L prototype m
(31.19)
(31.21)
31.4.2.2 Confined Jet with Insufficient Amount of Surrounding Flow The surrounding flow is less than the amount that can be entrained by the confined jet. The following assumptions are taken into considerations: The confined jet expands as a free jet with angle α = 19.4° and thus impinges on a wall at the distance xP. The confined jet behaves as a free jet up to point C, at abscissa xC (i.e., entrainment ceases and disentrainment begins); downstream, midway between the point where entrainment is zero is the point N with abscissa xN, and the point P where the jet impinges the wall has the abscissa xP (see Figure 31.9). e is preIn this case the surrounding flow stream m sumed to be entirely entrained by the nozzle jet before starting the external recirculation process. The mass entrained between the points N and C is the mass of
X Xc
XN
Xp
ERZ L d0
U0 Nozzle
Figure 31.9 Axial confined turbulent jet: External recirculation zone (ERZ). (From Kutz, M., Mechanical Engineer’s Handbook, 3e, Book 4, Energy & Power, New York: John Wiley, 2006. With permission.)
© 2011 by Taylor and Francis Group, LLC
636
Industrial Combustion Testing
external recirculation calculate using the Ricou and Spalding formula [3]: e m m ρ = r = 0.32 e o m o ρo m
0.5
xC − 1. do
θ=
doeq ρo 2 L ρe
• m ≤ 1.5 no recirculation • 1.5 ≤ m ≤ 4 week recirculation • m ≥ 4 significant recirculation
(31.22)
The Thring–Newby parameter [5] is defined by:
The Craya–Curtet parameter [4] is ranging as follows:
m≈
m = −1 . 5 R 2 + R + K
0.5
,
1 − 0 . 5, ϑ2
(31.23)
yields
r 0.72 m = − 1. 0+m e ϑ m
(31.24)
These assumptions lead to the following expressions for the position of the points xN, xC, and xP. The distance from the jet nozzle to the point N is given by
xN = 0.65 ϑL,
(31.25)
xC = 3.125(ϑ + 0.94)L.
(31.26)
(31.28) R2
d0 2 L
2
,
(31.29)
where K = 1 for the round jet
R=
( u0 − u1 )ρ0 d20
2
2
d d u1ρ1 1 − δ + ( u0 − u1 )ρ1 0 2 2
2
,
(31.30)
where δ = the boundary layer thickness. Craya and Curtet [4] have calculated an expression of the mass flow rate of recirculating flow:
For calculations of abscissa of the point P to the origin of jet O:
r m = 0.43 ( m − 1.65 ). 0+m e m
(31.31)
When the jet is considered as a point source, at its origin the relationship between Thring–Newby [5] and xP = 5.85L (for the free jet, α = 19.4º, then the half angle Craya–Curtet [4] parameters are written: α/2 = 9.7°), ϑ = m. (31.32) where 2L is the diameter of the enclosure. The fluid entrained consists of combustion products The points where the ERZ is starting and is ending, that are recirculated back along the walls. The walls respectively: also serve to prevent the jet from spreading out too rapidly. If the walls are hot, this ERZ (external recircula3d f tion zone) serves to recycle hot combustion products xN = if m > 1.5, (31.33) m back into the ignition zone, which is good for flame stability. Craya and Curtet [4] have used averaged integrated 4 . 5d f xP = . (31.34) values of Navier–Stokes equation as well as the conti- 2 nuity equation to predict the confined jet behavior. The dimensionless Craya–Curtet parameter m is given by: The relation rate
G G 1 m= 0 + e − G∞ 2G∞ 2
where G 0 = the nozzle jet momentum, N; Ge = the ambient fluid momentum, N; G∞ = the mixed flow momentum, N.
© 2011 by Taylor and Francis Group, LLC
(31.27)
r m d = 0.43 ( m − 1.65 ) 0 < 0.1, m0 + me df
(31.35)
r m d = 0.60 ( m − 1.40 ) 0.1 < 0 < 0.1. (31.36) 0 +m e m df
637
Cement Kilns
Table 31.11 Main Jet Similitude Parameters for Simple Confined Jets
Thring–Newby parameter Craya–Curtet parameter Curtet number
Symbol
Maximum Value
Minimum Value
Onset of Recirculation
Relationship
θ
0
∞
<0.9
m−0.5
m
∞
0
>1.5
Ct
0
∞
<0.75
Table 31.11 shows the relationships between main jet similarity parameters. 31.4.3 Confined Turbulent Jet with Swirling 31.4.3.1 Swirl Number In swirling free jets, both axial momentum and axial flux of angular momentum are conserved; since both quantities can be considered as characteristics of jet aerodynamic behavior. Chigier and Beèr [45] recommended as criterion of swirl intensity:
S=
Gt = RCGx
rwv ⋅ dA A
∫
RC [ uv + ( p − p∞)] ⋅ dA A
dP w2 =ρ , dr r
where Gx = the axial flux of axial momentum, N; Gt = the axial flux of angular (swirl) momentum, Nm; RC = the characteristic length, m; p = the static pressure, Pa; p∞ = the reference pressure, Pa; v = the velocity vector (u axial, w tangential (swirl) and v radial velocity components), m/s; d A = the elemental area, m2. Swirling jets are used as a means for controlling flame behavior. When rotating motion is imparted to flow upstream of a cylindrical nozzle, the jet expanding has a tangential (swirl) velocity component in addition to the axial and radial velocity components typical for the nonswirling jets. Some recirculation is essential for flame stability. In many cases, the external recirculation originating from confined jets is insufficient to maintain a constant supply of hot gases and/or active species to maintain ignition. In this case, recirculation of hot gas must occur by some other means. The presence of swirl or of tangential velocity component results in the setup of radial and axial pressure gradients. The radial pressure gradient, given by
(31.38)
if radial velocity and viscosity are neglected, but will in all cases be a positive number. This means that the pressure will be lowest in the center of the vortex. As the swirl dissipates in the furnace, the pressure along the centerline increases. This adverse pressure gradient prompts a reverse flow down the center, creating internal recirculation of hot gases as shown in Figure 31.10. The Swirl number, defined as the ratio of the axial flux of angular momentum divided by axial flux of axial momentum times the characteristic radius of the burner nozzle, can be rewritten as:
(31.37)
∫
© 2011 by Taylor and Francis Group, LLC
m−0.5
Gt S= = RCGx
RC
∫
R
0
ρuwr 2 dr
∫0 [ρu2 ( p − p∞)]rdr R
.
(31.39)
The swirl velocity component, w, is a strong function of radius r at the burner exit, and knowledge of this dependence is essential in order to calculate Swirl numbers. This functionality depends on how swirl is generated. The Swirl number, however, is an important parameter describing the flow of swirling flames and can be used for scale up purposes. For S > 0.3, a reverse flow can occur and increasing swirl will always shorten the flame jet and increase its spread. For weakly swirling flows (S < 0.3), the angle of spread is given by
α = 4.8 + 14S.
(31.40)
Mathur and Maccallum [18] have reported results of their experimental work done on hubless and annular vane swirlers for pressure drop across swirler, axial static pressure, and axial and tangential velocity components distribution. It was concluded that the swirling jets experienced a sudden expansion downstream of the swirl generators. An internal recirculation zone (IRZ)
638
Industrial Combustion Testing
Secondary recirculation
Primary recirculation
r x
Swirl
Figure 31.10 Confined swirl jet: internal (primary) and external (secondary) Recirculation Zones. (From Kutz, M., Mechanical Engineer’s Handbook, 3e, Book 4, Energy & Power, New York: John Wiley, 2006. With permission.)
is generated downstream of the vane swirl generator when the vane angle is 45° or higher. These authors argued that tan ϑ was a good measure of the swirler intensity for jets issuing from vane swirler
S=
Gt 2 ( 1 − (R2 / R1 )3 ) tan ϑ , = RCGx 3 ( 1 − (R2 / R1 )2 )
(31.41)
where ϑ = the vane angle relative to swirler longitudinal axis; R1 = the external radius; R 2 = the internal radius. Beltagui and Maccallum [13] based on the proposal of Kerr and Fraser [14] have used an equivalent diameter of vane swirl generator
Gt 1 ( 1 − (R2 / R1 )3 ) tan ϑ. S= = deqGx 3 ( 1 − (R2 / R1 )2 )
(31.42)
The definition of the swirl number provided by Chigier and Béer [44] did not completely characterize the confined swirl jets, because in the case of furnaces downstream of a swirl generator, the swirl flow depends on other parameters such as burner quarl or combustion chamber dimensions. 31.4.3.2 Modified Swirl Numbers Beltagui and Maccallum [13] have proposed a new definition of the swirl number by removing the static term of axial momentum and using as characteristic length the diameter of the combustion chamber (furnace):
© 2011 by Taylor and Francis Group, LLC
S* =
Gt , dFGxd
(31.43)
where Gxd = the dynamic term of axial momentum, N; Gt = the angular momentum, Nm; dF = the characteristic length (the combustion chamber diameter), m. The axial flux of axial momentum has two terms, one is dynamic Gx,d and the other static Gx,s
r
∫ ρu rdr ,
Gx ,d =
Gx , s =
2
0
r
∫ ( p − p ) rdr. 0
∞
(31.44)
(31.45)
They found that the same swirl generator can provide different velocity patterns as the relative furnace size df/ db is changed. At the same time, the size and the shape of the internal or central recirculation zone (CRZ) is primarily dependent on furnace diameter and not vane swirler diameter. Cristea [15,16] in 1984 and 1987, respectively, introduced, relative to a complete system swirler-quarlfurnace, a new modified swirl number in which the characteristic length is the exit diameter of the burner quarl.
S* * =
Gt , rQGxd
(31.46)
where rQ is the burner quarl characteristic radius, with rQ = f(quarlsemiangle), if the quarl length lQ = const. and swirler diameter db = const.
639
Cement Kilns
The analysis of flow pattern models occurring inside the radial distribution of the velocity vector components and the pressure at swirl generator exit are nonuniform; the same burner equipped with an annular vane swirl generator in the same furnace can produce different velocity vector components, when the quarl geometry is changed (dQ/db ≅ 2–3.5); the shape and size of CRZ are primarily a function of quarl geometry and not of vane swirler diameter. The Swirl numbers S and S* are quite different. However, the two swirl numbers S* and S** are correlated with their ratio corresponding to the inverse of the ratio of the respective characteristic length. 31.4.3.3 Swirl Number Calculation for Cement Rotary Kiln Burners For the three-channel burners (or “tuyères” a French term) a special design for cement rotary kilns, there are two ways to calculate the swirl number, both methods being developed in the frame of CETIC (Comité d’Étude Technique des Industries Cimentières) in 1994. S=
Gt . r0Gx
(31.47)
Where r0, the characteristics radius, can be considered either the external radius of the swirler or the characteristics radius of each channel of the three-channel burners. r0 = r2
(31.48)
or
r0 =
( m ax + m tr ,coal + m sw ) , ( πρGx )0.5
(31.49)
where ρ = the air average density The total axial momentum is expressed as the sum of the axial momentum on each channel (axial air stream, swirl air stream, and transport air and pulverized coal streams)
ax uax + m tr ,coal utr + m sw uax . Gx = m
(31.50)
The angular momentum is expressed by:
sw vsw ,tan ) rsw . Gt = ( m
Where the turning radius is given by:
© 2011 by Taylor and Francis Group, LLC
(31.51)
rsw =
2 r23 − r13 . 3 r22 − r12
(31.52)
Second option
ax uax req , ax + m tr ,coal utr req ,tr + m sw uax req , sw ), (31.53) r0Gx = (m ax uax req , ax ) req . Gt = ( m
(31.54)
Mathur and Maccallum [18] have proposed the equivalent radius of each channel (axial, swirl, and transport):
req =
2 r23 − r13 . 3 r22 − r12
(31.55)
31.4.4 Combustion System Scaling As mentioned by Spalding [19], the strict requirements of similarity theory are so numerous that exact scaling of any practical combustion system is impossible because the requirement for one dimensionless group to maintain constant is often incompatible with the another criteria requirements. Two elements must be considered simultaneously when scaling a combustion system (i.e., burners and its quarl) and the furnace forming the so-called complete burning system. Thus a complete combustion should be scaled in accordance with the following main criteria: • Geometric similarity • Kinetics, aerodynamics and mixing similitude between the streams of fuel and oxidant • Thermal similarity The two most frequently applied criteria for the industrial flame scaling are the constant velocity and constant residence time approaches. The cement rotary kiln is effectively an axisymmetrical combustion chamber combined with a moving bed chemical reactor. To maintain consistent heat transfer patterns from the flame to the material bed, the constant velocity scaling criteria is used, but to assure the clinkerization reaction effectiveness the constant residence time in the burning zone is required. It might be argued that the cement clinker rotary kilns should be theoretically scaled up based on the constant velocity criteria for the combustion process and with the constant residence time for the clinker. Unfortunately this method is not practical due to the limits on the tires size, then a compromise should be found such as reducing volume
640
kiln load (tpd/m3). It is more practical to size the kiln based on the constant residence time for the clinker and use the mathematical/physical modeling for the combustion process.
31.5 Rotary Kiln Burners The cement rotary kiln is a typical single burner installation for which the burner is usually acquired as an add-on separate package to an existing or brand new burning line manufactured many times by another supplier. Thus the system combustion aerodynamics design is not in the hand of a burner pipe manufacturer, but is mainly dependent on the kiln designer; therefore the cement rotary kilns operate with poor aerodynamics because the burner and kiln designers are not working together to match secondary air flow pattern with the flame aerodynamics. This can be achieved by using either physical or mathematical modeling to optimize the design of the entire combustion system. Normally through the burner pipe, the cold primary air represents 10–30% of the stoichiometric combustion air and the remaining part including the excess air (10%), namely secondary air, is fed through the kiln hood from the clinker cooler, at the temperature of about 1000°C. The burner must be able to fire conventional fuels such as coal, petroleum coke, fuel oil, and natural gas (or any mixture thereof), ensuring complete combustion, low excess air, and minimum formation of carbon monoxide (CO) and nitrogen oxides (NOx). If relevant, the burner must be able to handle AF without requiring a change of its original design. The requirements of flames were claimed by kiln operators to obtain good quality clinker: • The burner must produce a short, narrow, strongly radiant flame, as this is a condition for good heat transfer from the flame to the material in the sintering zone of the kiln. • Flame formation must be favorable to a dense, stable coating on the brick refractory lining in the burning zone of the kiln as well as to a nodular clinker formation with low dust content and correctly developed clinker phases. • The burner must use as little primary air as possible without compromising stability normally or upset operating conditions. The cement rotary kilns are equipped with burners producing two type of flames (i.e., axial and divergent). The axial flames are generated by simple nonswirling burner pipe connected to a direct coal firing
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
system. The divergent flames are produced by swirling burners that generate a central or internal recirculation zone (CRZ/IRZ) sufficiently developed to expand the PF jet; these swirling burners are normally used in connection with indirect coal firing system. In modern dry-process kilns the multichannel burners have a minimum amount of primary air (calculated as a percentage of stoichiometric combustion air) used to obtain a satisfactory flame shape; however, correct momentum is more important than a low primary air to suppress NOx formation. The main important worldwide burner manufacturers are in strictly alphabetic order as follows: FCTInternational, Australia; Fives Pillard, France; FL Smidth, Denmark; Greco-Enfil, Brazil; KHD, Germany; Polysius, Germany; and Unitherm, Austria. There are three generations of burners specifically developed for the cement clinker rotary kilns. • Monochannel burners generating a long, soft flame, which were originally suitable for the long kilns in the wet and in the dry processes. • Multichannel burners: In the 1970s when the kilns became shorter with the introduction of dry process with suspension preheater and precalciner fired on solid fuels, they were developed in order to reduce the primary air supply and to obtain a short, divergent with strong swirl flame suitable for firing pulverized fossil coal first and petcoke. • Low NOx multichannel burners: To comply with more stringent environmental legislation, at the end of the 1980s a new burner generation was introduced characterized by a slightly longer and less divergent flames. The main advantages are: • High temperature flames with reduced primary air using high velocities at the burner tip (by blowers/high pressure fans utilization) • The environmental pollution control by the reduction of NOx generation and CO production (function of O2 at the kiln feeding end) • Able to burn AF • An improved flame shape matching the furnace geometry • Higher kiln running factor by increasing the brick life • Less preheater plugging in or ring formation in rotary kiln section • Decreased kiln specific heat consumption
641
Cement Kilns
The primary air stream plays two major roles in flame behavior: • controls the rate of mixing between fuels and oxidants • assures the flame stability Primary air is usually defined by a percentage of the stoichiometric air calculated for the total amount of fuel. In the case of PF, it provides a transport air stream that can or can not be considered in the total amount of primary air. Since the very hot secondary air has to be entrained into the fuel-primary air jet, it can have an important impact on the fuel-oxidant macro-mixing. The flow patterns of the secondary stream are mainly determined by the design of cooler uptake and the kiln hood itself. The relationship between primary air jet momentum and secondary air velocity has a significant impact on the flame geometry as well as on the heat transfer to the material and refractory lining. 31.5.1 Multichannel Low NOx Burners Generally speaking there are two design modes regarding the relative positions of the channels of axial air, swirling air, coal and transport air, and core air. The combination of the location of channels starting from the external side to the symmetry axis of burner can be as follows: • axial air channel, swirling air channel, coal and transport air channel and core (central) air (for instance the burners manufactured by FivesPillard, F.L. Smidth, and Greco-Enfil)
• axial air channel, coal and transport air, swirling air channel, and central air (for instance the burners built by KHD and FCT) The Italcementi Group has experienced a certain number of burner brands in its cement works, which are briefly presented here next. 31.5.1.1 Rotaflam Burner The Fives-Pillard has developed its third generation of Rotaflam burner pipe with an enhanced adjustment mechanism, to meet any specific operation condition required by the cement rotary kilns. A new low NOx burner tip has been developed jointly by Italcementi Group and Fives-Pillard with a radial distance between the swirl air channel and pulverized coal channel higher then in the case of the standard tip (Figure 31.11). However the diameter of burner was increased and the central plate diameter was smaller with less room to accommodate the waste fuel jackets. In the case of the petroleum coke, the ignition distance is longer compared with the standard tip. Rotaflam multifuel burners are available for all conventional fuels (i.e., pulverized coal, natural gas, and any kind of HFO). Fuel combustion proportions are achievable within a flexible capacity range up to 100%. In a special design version, the same type of burner uses a wide variety of solid and liquid waste fuels to substitute the conventional fuels (Figure 31.12). The Burner Management System (BMS) allows that three fuels can be burned simultaneously at different capacities. In case of simultaneous co-combustion of LCV waste fuels, and high calorific value fuels, the BMS
Coal
Alternative fuel jacket
Extended annulus
Oil gun Axial air
Swirl air Core air
Ignitor
Figure 31.11 Low NOx burner tip developed jointly by Pillard and Italcementi Group. (Courtesy of Italcementi Group.)
© 2011 by Taylor and Francis Group, LLC
642
must avoid lower flame temperatures by controlling the limitation of LCV waste fuel proportions. To allow good fuels mixing to control the flame shape and avoiding the formation of dual flames, the burner should be adjustable under working conditions. The burner tip adjustments by variation of the cross sections of the axial and swirl air channels are done mechanically or by a hydraulic mechanism. The new design of the Rotaflam burner tip presents a minimum air excess at the burner tip, an axial air outlet coaxial to symmetry axis and only a low divergence at the swirl air outlet. The typical primary air pressure for multifuels firing is 250 mbarG. The primary air amount ranges between 8 and 12% of the total combustion air needed to generate an axial momentum from 8 to 12 N/MWth. The flame holder plate at the end of the central air channel establishes, by the “bluff-body effect,” a flow pattern with a IRZ of negative pressure at the burner tip to help the flame to burn very close at the burner mouth. A number of aspects must be taken into consideration when deciding to burn AF in a rotary kiln. Technical aspects, such as the heat value of different fuel types, are very important. A possible negative impact on both the clinker quality and refractory lining must also be taken into consideration. The refractory lifetime could be reduced. To avoid this, more heat and chemical resistant refractory materials should be installed as refractory lining on the external burner pipe. Economic aspects are also important and should always be kept in mind. The availability of different AF has to be considered, transportation and preparation expenses, the cost of specific equipment together with supplementary additions, plus the cost of emission control devices. Liquid waste fuels are mainly introduced into the flame through specially designed atomizers. In the past solid waste fuels were mainly blown into the flame by separate jackets fitted either on top of the burner or in the kiln head, depending on the kind and preparation
Industrial Combustion Testing
grade of the fuel. Most of the modern designed waste fuel burners are equipped with special jacket tubes to blow the pulverized or shredded waste fuel directly into the center of the flame. This mostly results in increased flame lengths. Particles of the waste material could also leave the main flame boundary and burn in the hot secondary air stream under over stoichiometric conditions. To avoid this situation Fives-Pillard has developed and patented a swirler nozzle; a portion of the primary air is blown through a jacket pipe around the solid waste pipe into a special swirler nozzle at the end of this pipe. This nozzle brings the waste material into rotation that leads to a better outspreading of fuel into the flame. As a result the flame length is similar to the flame length before and the combustion of the alternative solid fuel is more effective. 31.5.1.2 FlexiFlame™ Burner Greco-Enfil proposes the new FlexiFlame™ burner design based on ideas that were seen as antagonists in the past: • The declared goal was to achieve an improved combustion, good flame control, and low NOx emissions. • The use of two swirling air flows with axial and tangential velocity components in a “sandwich” configuration. The burner tip has no moving part, then its adjustment is done by three butterfly valves installed on the cold end of burner. The main burner concept of the burner tip is to have the PF channel located in between two streams of swirling air (Figure 31.13). The multichannels are positioned as follows: • Central cooling air stream with axial, radial, and tangential velocity components flows around several jackets of solid waste fuel injection, External air channel Pulverized fuel channel
Cooling air channel
Figure 31.12 Coal/oil–liquid waste fuel Rotaflam burner installed on the kiln. (From Guse, B., Multi-Burner Flexibility, Fives-Pillard, June 2006. With permission.)
© 2011 by Taylor and Francis Group, LLC
Dispersion air channel
Swirl air channel
Figure 31.13 FlexiFlame™ burner tip. (From Greco Enfil Company, “Greco–Enfil Company,” presentation to Italcementi Group. With permission.)
Cement Kilns
• • • •
guide pipes for liquid waste and fuel oil guns, igniters, flame safeguard sensors, and so on Dispersion air stream with axial and tangential velocity components PF transport air with only axial velocity components Swirling air stream with axial, and tangential velocity components External air stream with only axial velocity component
The central cooling air is used to cool down the central plate of the burner tip and to supply oxygen in a flame core to help the solid/liquid waste fuel burning. The dispersion air stream is used for better pulverization of the PF jet, fast ignition, and quick response of the flame to primary air modification. The swirling air stream keeps the unburnt fuel and devolatized nitrogen inside of IRZ, witch becomes reach in reducing species able to control both fuel bond and thermal NOx. The external air stream is used to increase the hot secondary air entrainment into the flame. The external air flow is expanded through discrete slots to enhance the suction of the hot secondary air and improve mixing of two jets. 31.5.1.3 M.A.S.® Burner Unitherm-Cemcon has developed the M.A.S.® burner pipe in a different manner with others burner pipes existing on the market at that time. Figure 31.14 displays the burner tip with separate jackets for fluff (plastics), animal meal, and for a gas/liquid gun for the ignition
Figure 31.14 M.A.S. burner tip. (From Unitherm Cemcom. “M.A.S. Burner for Traditional and Alternative Fuels,” presented to Italcementi Group, 2005. With permission.)
© 2011 by Taylor and Francis Group, LLC
643
and warm-up. Located in one jacket is a flame safe guard system (process supervision by digital picture processing) enabling a direct view of the flame. The design criteria claims by the manufacturer to be used in concept developing of this burner were: • To simplify the flame length and shape adjustment • To increase the range of the flame shaping • To reduce the primary air pressure drops and to keep the constant air momentum of each flame shape • To reach high continuous stability of the flame independent of its shape • To extend the life time of the burner by better cooling of the external jacket pipe Use multiple steel flexible hoses installed in only one primary air channel for flame shaping to bring the whole primary air flow into a swirl of desired intensity without any reduction of primary air momentum at different flame shapes. Essential benefits of M.A.S.® claims by the manufacturer are as follows: • Very simple flame shape adjustment and improved control of the flame • Total primary air available for cooling of the outer burner pipe • High level of operation, safety, and long life of the burner • Reduced primary air pressure drops due to only one primary air channel. Maximum and constant air momentum independent of flame shape • Efficient mixing of secondary air into the flame because of the single air jets stream effect • Excellent burn-out of the different fuels • Low primary air rate of 6%–12% according to kiln and fuel requirements • Low NOx-emissions, due to the close flame stand-off distance The burner is equipped with a flexible swirl setting device to adjust the flame shape. Quasi whole amounts of primary air flow through the flexible hoses. The swirl intensity of primary air stream is determined by the angle of deflection of the hose nozzles, which are easily adjustable from cold end of the burner. The total primary air is internally divided in the air circuit to the flexible hoses and the central air circuit. It is particularly important that the fuels, the burner primary air, and the kiln secondary air mix properly to ensure
644
Industrial Combustion Testing
the foreseen flame shape and the heat release profile is achieved. The burner is equipped with a flame setting device to adjust the flame geometry. The readings on the scale refer directly to the adjustment of the deflection angle of flexible air hoses. For instance a 10° hose deflection angle is foreseen for a long flame (scale indication 1.5) and 35° for a short flame (scale number 8.5). The burner is controlled by the BMS of Unitherm Cemcon. The BMS enables the adjustment of the fuel flow rates for a set point of the thermal load. The communication between BMS and upper supervision system in the control room is performed by Profibus DP enabling the burner remote control. The refractory lining of the burner has an 8 m length. The last part at the cold end side of the refractory lining is pre drying before installation at 400°C. The ceramic bricks are installed on the burner’s jacket duct and to avoid the thermal stress the dilatation joint are filled with the ceramic fiber. 31.5.1.4 Duoflex Burners The Duoflex burner of F.L. Smith is based on a novel concept featuring a central duct for gaseous and liquid fuels placed inside an annular coal duct. The Duoflex burner fires with all types of fuels and may also be fitted to AF. A high momentum burner, such as the Duoflex G2 burner depicted in Figure 31.15, which provides a vigorous mixing of the particles with the local atmosphere, increases the probability that the char particle will encounter oxygen to facilitate the quick burnout.
F.L. Smith has developed the next generation of the Duoflex burner, known as the Duoflex G2. The Duoflex burner had been introduced to the market in the mid1990s. While the fundamental design and functionality of the burner remains the same as before, various updates with unique design elements, greater alternative fuel flexibility, and simplified maintenance have culminated into the official release of the new burner. The key features of the Duoflex G2 are as follows: • • • •
Replaceable high-pressure central duct nozzle Inclined nozzle for alternative fuel Automatic air nozzle centering device Temperature measurement for early detection of refractory failure
The Duoflex burner fires rotary kilns with pulverized coal or coke, oil, natural gas, or any mixture of these fuels. The burner may be fitted with extra ducts for AF such as plastic chips, wood chips, sewage sludge, and so on. Standard types are available for any fuel combination and a maximum capacity ranging from 20 to 250 MWth., catering for even the largest of rotary cement kilns. The burner is based on a novel concept featuring a central duct for gaseous and liquid fuels placed inside an annular coal duct, which is surrounded by two concentric ducts that form two primary air channels, one for swirling air and one for axial air. The main features of the Duoflex burner are the robust design, the low primary air consumption, the adjustable swirl, the adjustable air nozzle area, and the central fuel injection.
Waste oil jacket
Wood chips jacket Axial & swirl air channels Cool air channel Pulverized coal channel
Fuel oil jacket
Plastic chips jacket Figure 31.15 Duoflex 2G. (From F. L. Smidth Co., “Alternative Fuels,” presented to Italcementi Group. With permission.)
© 2011 by Taylor and Francis Group, LLC
645
Cement Kilns
31.6 Burner Pipe Testing on Cement Rotary Kilns in Industrial Operation The burner pipes for cement kilns have thermal power up to 250 MWth with length of about 10 m and weight of about 15 tons. These burners are tested only on rotary kilns either in the commissioning stage or already in industrial operation after the refurbishing of the combustion equipment. 31.6.1 Semipilot Scale Experimental Tests Due to the lack of understanding for the flame behavior in cement clinker rotary kiln, the Cemflame I Consortium was set up by the most important European cement producers and combustion system manufacturers for financially supporting the IFRF, now located in Pisa and Livorno Italy, to develop an experimental program to evaluate the primary reduction measures of NOx in this specific industry. The NOx reduction techniques applied in the cement clinker rotary kiln are generally classified as primary and secondary measures. The primary methods are applied to the main burner to prevent NOx generation:
• Reduction of amount of the primary air by utilization of low NOx burner pipe and by PF sourced by an indirect firing system. • Utilization of the multichannel burners able to control separately the primary air flow stream on each channel (axial, tangential-swirl, transport, and core). • Utilization of multichannel burners with an important bluff body at the tip to influence the near burner zone hence to obtain the flame ignition point near the burner tip. • Injection of a small amount of water at the flame root to reduce the flame temperature; this it can be done by using liquid AF. The experimental campaigns were carried out on a semi-industrial scale cement kiln simulator facility set up at the former site of IFRF in IJmuinden Holland (see Figure 31.16). The objectives of the research program: • To improve the understanding of cement kiln flames • To determine the influence of burner design, operation, and fuel type
Natural gas (0–40 kg/h) Oxygen (0–300 kg/h)
P
P
Heating up purpose
P
Natural gas
Tertiary air (560°C) Secondary air (500°C) (Total 2000–3000 kg/h)
Precombustor Temperature Experimental kiln
800 1000 °C
Probe holes/view ports
Refractory lining
P
(0–100 kg/h) axial air
(0–100 kg/h) swirling air
(0–30 kg/h) core air
Kiln burner
P
1
2
3
4
5 6 7 8 Watercooled segments
9
Gas composition NOx, CO, CO2, O2 Mono channel air (Coal = 300 kg/h, air=100 kg/h) Coal + transport air Mixer
10
11
12
13
NOx, CO, CO2, O2 and/or temperature Coal
14
15
Chimney
Flue gas measurements NOx, CO, CO2, O2, solids
Open for monochannel (150–800 kg/h)
P
Multichannel air
Open for multichannel
Compressor
Figure 31.16 Experimental cement kiln simulator. (From Van de Kamp, W. L., and Smart, J. P., “Cemflame I: The Effects of Burner Design and Operation and Fuel Type on the Properties of Cement Kiln Flames,” International Flame Research Foundation Doc. No. F 97/y/1, Livorno, Italy, 1992. With permission.)
© 2011 by Taylor and Francis Group, LLC
646
• To create the engineering guide lines for variability, fuel efficiency, and heat transfer characteristics • To determine the influence of flame characteristics, burner type, and operation on NOx formation • To compare two different scaled down burners and to utilize the knowledge to optimize these prototype During the experimental campaign 161 different flames were tested, generated by a generic mono channel burner pipe and two low NOx scaled down industrial burners manufactured by Fives-Pillard (Rotaflam burner) and KHD (Pyrojet burner). An inadequate PF burner/combustion equipment design and operation results in: • reduction conditions in the kiln burning zone • ash heterogeneity occurring in the burning zone • variable burning, then under/over burning (clinker poor quality) • kiln flow patterns generated by the swirling burners depends on the primary air momentum that affects the secondary air rate of entrainment The cement rotary kilns are equipped with burners producing two type of flames, axial and divergent: • The axial flames are generated by simple nonswirling burner pipe and are applied in direct coal firing system. For the axial flames an acceptable micro-mixed controlled flame length of approximately 6 to 7 kiln diameters can be achieved. • The divergent flames are produced by swirling burners that generate a CRZ sufficiently developed to expand the PF jet; these swirling burners are normally used in indirect coal firing system. In the case of the divergent flames, for the same burnout time, would result in flame impingement against kiln refractory lining. Therefore a more rapid mixing/ entrainment of the secondary air stream is required. In modern dry-process kiln the multichannel burners have a minimum amount of primary air used to obtain a satisfactory flame shape; it seems that correct momentum
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
is more important than a low primary air, to suppress NOx formation. Thring and Newby [5] had drawn a simple schematic diagram of flame and aerodynamic flow patterns in the cement kiln burning zone (see Figure 31.17). The IFRF experimental kiln with ID = 2 m and length of 40 m to simulate the burning zone of rotary cement kiln. Prior to ignition, the PF devolatilization is taking place. The critical combustion reactions, as far as rotary kilns are concerned, are as follows:
H + O 2 = OH + O,
(31.56)
CnH m + O = Cn−1H m ,
(31.57)
CO + OH = CO 2 + H,
(31.58)
2 CO + O 2 + M = 2 CO 2 + M,
(31.59)
H 2 O + O = 2 OH,
(31.60)
2 C + O 2 = 2 CO.
(31.61)
Provided that PF fineness and moisture are complying with the foreseen figures, the most critical reactions are Equations 31.56 and 31.58. Hydroxyl radicals are mainly produced by the combustion of the PF volatile matters and their concentration depends on temperature and local stoichiometry. To generate enough OH radicals to oxidize the available CO it is essential that sufficient O2 is available to combust local CO pockets of evolved volatile matters near stoichiometry. It was shown that reactions (Equations 31.56 through 31.60) can be considered to be micro/macro mixing controlled. The flow patterns in the near burner zone are controlled by the burner design. Detailed evaluation, carried out on the IFRF experimental rig, of the effect of momentum on the flame heat transfer has indicated that a minimum 7 N/MWth axial momentum, not allowing for coal and transport air momentum, is required. Since 2001 some of the cement manufacturers, such as Lafarge or Holcim, have used higher axial momentum of about 11 N/MWth and to this day up to 13 N/MWth for their cement plants according to Lowes [26]. The Cemflame 2 Consortium was focused on obtaining more knowledge about NOx formation. The objectives of the work program were the following:
647
Cement Kilns
Axial flame Secondary air 700–900°C
O + OH (Flame front) NO (Post flame front)
H + O2 N2 + O
Primary air (20–35%) + PF at 50–100 m/s
NH3 HCN
+
O2
NO (Preflame front) 3D
1–1.5D
Maximum coal concentration Divergent flame
Primary air 8–15% up to 150 m/s
Macro mixing O2/C necessary by this position
Coal
Secondary air 700–900°C Burner with three concentric annuli
6–7D
H + O2 N2 + O NH3 O2 HCN
Internal reverse flow zone NH2 +NO HCN
3/4 – 1D
OH + O (Flame front) NO (Post flame front) NO (Pre flame front)
N2 (Pre flame front)
1–
1.5D
3–4D
Figure 31.17 Schematic diagram of flame and aerodynamic flow patterns. (From Lowes, T. M., and Evans, L. P., Journal of the Institute of Energy, 220–28, 1989. With permission.)
• To determine the independent effects of PF particle size distribution (PSD) and volatile matter content on NOx, CO, burnout, and flame structure • To determine the effects of burner operation condition on NOx and CO emissions, burnout and flame structure, and the propensity to produce reducing conditions at the flame extremities and in the ERZ • To investigate the effects of burner input on the size, shape, and strength of the IRZ • To determine the effects of fuel type, fineness, and burner input conditions on flame ignition pattern
were varied independently in the respective channels. In total, 189 flames were investigated in different burner input conditions, burner type, and natural gas support. The Cemflame 3 Consortium experiments were concentrated on three different main solid fuels (petcoke, Medium Volatite Bituminous (MVB), and High Volatile Bituminous (HVB) coals) and several alternative fuels (such as sewage sludge, plastics, shredded tires, agglomerates from separation of municipal solid waste, etc.). In total more then 240 flames were measured in a cement kiln simulator and fuel studies were executed in the isothermal plug flow reactor. The objectives for experiments can be summarized as follows:
The generic multichannel burners were of versatile design to allow easy variation of burner input parameters. These included the location of the axial and tangential (swirl) channels, the mean velocities of the respective air streams, and the total primary air amount, expressed as percentage from the combustion stoichiometric air. The mass flow rates and velocities
• to assess different co-firing fuel types and to give recommendations • to maximize the co-firing ratio of AF • to identify flame characteristics when co-firing AF • to evaluate the effect of oxygen boosting
© 2011 by Taylor and Francis Group, LLC
648
Industrial Combustion Testing
31.6.2 Burner Pipe Testing on Cement Rotary Kiln
Table 31.12
In 1995, the Italcementi Group and Pillard Company, in the frame of an EU contract, decided to install a low NOx Rotaflam burner on 700 (m)tpd Lepol kiln at Sarche di Clavino plant located northeast of Italy; the goal was to perform industrial testing of the “wide adjustable” low NOx Rotaflam burner pipe in industrial operation conditions. The adjustment of the primary air flow rate and velocities at burner tip were done in two different modes:
Main Technical Characteristics of Rotaflam Burner at Sarche Plant
• The air pressure on each channel was varied in the range from 220 mbarG to 100 mbarG by the throttle valves installed on the burner. By closing this valve, the air flow rate was decreased due to the reduction of tip exit velocity (from 200 to 100 m/s). • If the pressure is fixed on each channel, the relative position of the axial channel pipe and swirl channel pipe by modifying the exit cross section and hence the air flow rate of streams.
Design Thermal Power Turn down ratio Refractory lining length Trolley type Combustion flow rate Design primary air Transport air flow rate Transport air pressure Transport air temperature
The main technical characteristics of the Rotaflam burner are summarized in Table 31.12. The PF used was the green delayed petroleum coke with the chemical physical characteristics shown in Table 31.13. The primary air was supplied by two high pressure centrifugal fans with the characteristics presented in Table 31.14. Only as an example, the experimental results are given using the tip #7 with the following set up of the Rotaflam burner as it is shown in Table 31.15. The reduction of the transport air didn’t result in a significant reduction on NOx emission, but it had a positive impact on kiln specific heat consumption.
© 2011 by Taylor and Francis Group, LLC
28 1:1.66 5 carriage 29,700 1790 670 600 60
m Nm3/h Nm3/h Nm3/h mbar °C
Table 31.13 Pulverized Fuel Properties LCV Moisture content Volatile matter Ash content Sulfur content Fineness on 90 µm
Since 1996, experimental campaigns in three different stages over 2 years have been carried out: • The preliminary experimental campaign was focused on the establishment of the kiln base line in terms of NOx production with its original monochannel burner; 80% of total heat input was supplied by the main burner and the rest through the auxiliary burner installed on the Lepol grate. • The first experimental campaign with “wide adjustable” low NOx Rotaflam burner to fine tune the operation of the Lepol kiln. • The second and third campaign with “wide adjustable” low NOx Rotaflam burner, a certain number of burner tips as well as different transport air flow rates and petcoke fineness were tested.
MWth
kcal/kg %w/w %w/w %w/w %
8300 0.1 12 0.6 5.5 10
Table 31.14 Primary Air Fan Characteristics Fans Flow rate Static pressure Absorbed power Electrical motor power Rotation speed
Nm3/h mbar kW rpm
#1
#2
2700 220 31 37 3000
1350 200 19.1 37 3000
The ideal low NOx petcoke flame should be ignited and stabilized close to the burner tip with a minimum mixing with the secondary air stream. To minimize NOx generation, the highest flame temperature peak should be located in a zone with low oxygen content. The reduction of NOx emission at the Lepol kiln stack was in the range from 5 to 15% without changing the clinker quality and kiln production as well as without increasing the pollutant emissions. The developing activity of a low NOx burner by testing on industrial cement rotary kiln is very expensive. Currently the burner pipes are designed using CFD or physical modeling, then the fine tuning and optimization is done directly on cement rotary kiln in the phase of the commissioning. Further development is based on industrial data acquirement over long time periods of operation and their elaboration to bring new fresh ideas in design process.
649
Cement Kilns
Table 31.15 The Experimental Settings of the Rotaflam Burner
Channel Axial Swirl Core Transport Total flow rate % Primary air
Settings mm
Primary Air Nm3/h
Pressure mbar
Temperature °C
Fuel Flow Rate kg/h
Kiln Back End NOx ppmv @0% O2
20 23 40 0
1060 410 100 510 2080 11.3
190 190 150 460
50 60 50 60
2040
1000
31.7.1.1 PDEs Governing Equations
31.7 Process Mathematical and Physical Modeling 31.7.1 Mathematical Simulation A detailed presentation of computational fluid dynamics applied to combustion is beyond the scope of this work. The CFD is concerned with the numerical solution of Partial Differential Equations, (PDEs) governing the transport of mass, momentum, energy, chemical species in moving fluids. The Reynolds-averaged Navier–stokes equations (RANS) are obtained by averaging the instantaneous conservation equations for reacting flows. Usually it is talking about turbulent chemically reacting flows of different chemical species. The averaging procedure introduces unclosed quantities that have to be modeled using turbulence models (Wilcox’s k-ω, Wilcox’s modified k-ω, RSM, standard k-ε, RNG k-ε); combustion models (PDF: probability density function; EDF: eddy dissipation model; EBU: eddy breakup model) and radiation models (DOM: discrete ordinate model; DTRM: discrete transfer model, and flux model). Direct numerical simulations (DNS) have changed turbulent combustion modeling. The chemical reacting flows conservation equations are solved without any turbulence model. Large eddy simulations (LES): the objective is to explicitly compute the largest structure of the flow field, typically structures larger then the computational mesh size) whereas the effects of the smallest ones are modeled. Several popular commercial CFD codes are available on the market: • • • • •
Fluent and CFX (ANSYS-USA) Phonics (CHAM-UK) STAR-CD (CD-adapco-UK) FLOW-3D (FLOW Science-USA) MI-CFD (CINAR-UK)
© 2011 by Taylor and Francis Group, LLC
In PF combustion modeling in order to accurately tackle the physical problem, the Euler approach has been employed to numerically investigate the flow characteristics, heat transfer, and species transport by solving the time-averaged global mass, momentum, energy, and species mass fractions conservation equations. In the Lagrangian approach, the elemental control volume is considered to be moving with the fluid as a whole. In the Eulerian approach, in contrast, the control volume is assumed fixed in the space, the fluid is assumed to flow through and pass the control volume. The particle-phase equations are formulated in Lagrangian form, and the coupling between the two phases is introduced through particle sources in the Eulerian gas-phase equations. The standard k-ε turbulence model, finite rate chemistry, and DTRM (discrete transfer radiation model) radiation model are used. Combustion of coal particles included volatile evolution and char combustion to simulate pulverized coal combustion. • Reynolds-averaged Navier–Stokes equations (RANS) 1. Mass conservation equation Conservation of mass in an inertial reference frame, for a viscous, compressible, time dependent flow is given by:
∂ρ ∂ (ρui ) = Sm , + ∂x i ∂t
(31.62)
where ρ = the mixture density; ui = the mixture mass averaged velocity on i direction; Sm is the source term. 2. Chemical species conservation equation The chemical species conservation equation may be expressed in terms of mass fraction
650
Industrial Combustion Testing
of the each chemical species, as dependent variable, ∂ ∂ ∂ ρYk ) + ρuiYk ) = − J ( ( ∂x i ∂x i k , i ∂t
( )
+ R k + Sk ,
(31.63)
where ρ = the mixture density; Yk = the mass fraction of the species k = 1,...,N; J = the diffusion mass flux of the k-chemk ,i ical species in i-direction; Rk = the mass rate of creation or depletion of chemical reaction; Sk = the rate of creation by addition from the dispersed phase. Using Fick’s law and neglecting the diffusion flux due to the thermal gradients (Sorret effect),* it can be written in the form:
∂Y J k ,i = ρDkm k , ∂x i
(31.64)
where Dkm = the diffusion coefficient of k-species in the mixture. For N chemical species, N-1 equations are required. The mass fraction of the last Nth species is calculated by subtracting the mass fraction of N-1 species from 1. The influence of the turbulence on the reaction rate is taken into account by employing the EDM. 3. Momentum conservation equation Conservation of momentum in the i-direction, neglecting body forces, is written as:
∂σ ij ∂ ∂ (ρui ) + (ρui u j ) = + Su , ∂t ∂x j ∂x j
* In compliance with the Onsager reciprocal formula in irreversible processes thermodynamics, the concentration gradients of the chemical species are also able to produce the thermal flux, known as Dufour effect.
∂ (− ρ u 'i u ' j ) + Su , ∂x i (31.66)
+
where (−ρui ' u j ') are the Reynolds stresses to be modeled using, for instance, the standard “two-equation” k-ε turbulence model.
∂u ∂u j 2 ∂uk (− ρ ui ' u j ') = µ t i + − δ ij , (31.67) ∂x j ∂xi 3 ∂xk µ t = Cµρ
k2 , ε
(31.68)
where μt = the turbulent viscosity; k = the turbulence kinetic energy; ε = its rate of dissipation. Under the above mentioned assumptions, the global mass conservation equation does not modify its structure in the regime of turbulent flows. However the chemical species conservation equations and the energy equations, in the framework of k-ε models, make use of the “turbulent” Schmidt and Prandtl numbers. 4. Energy conservation equation The energy conservation equation can be written using the sensible enthalpy as dependent variable: ∂ (ρh) + ∂∂x (ρui h) ∂t i
(31.65)
where σij = the stress tensor, which is usually written in terms of pressure p and viscous stress τij, as σij ≡ τij − pδij being δij the Krönecker delta, and Su the source term representing interactions between the solid and gaseous phase. The Reynolds time average, neglecting density fluctuations, leads to the equation:
© 2011 by Taylor and Francis Group, LLC
∂ ∂ ∂p ∂τij (ρui ) + (ρui u j ) = − + ∂x j ∂x i ∂x j ∂t
=
∂ ∂x i −
k ∂h − c p ∂xi
∂ ∂xi
∑h
k
k
∂Yk ∂xi
Dp ∂ui + Dt + τ ij ∂x + Sh , j
∑h J
k k ,i
k
(31.69)
where h = ΣkYkhk with hk = ∫ TTref c p , k dT ; Yk = the mass fraction of the k-chemical species; Sh = the source/sink terms due to the combustion of volatiles, radiation, and heat transfer between combusting char particles and gas phase.
651
Cement Kilns
In turbulent compressible flows the total (stagnation) enthalpy, defined as h0 = h + ui2 / 2, is often used as dependent variable. Then the energy conservation equation is generally rewritten as:
∂ (ρh0 ) + ∂ (ρui h0 ) = ∂ k ∂h − ∂t ∂x i ∂xi c p ∂xi
∑h
k
k
∂Yk ∂xi
∂p ∂(ui τ ij ) ∂ + Sh . − + hk J k ,i + ∂x j ∂x i k ∂t
∑
(31.70)
• Numerical Simulation procedure 1. Grid generation In the case of Fluent code for instance, the GeoMesh preprocessor for geometry modeling can be used to generate the Computer Aided Design (CAD) geometry of the physical domain, then 2D boundary unstructured grid. The boundary mesh file is imported by T-Grid module, which requires a discretized boundary mesh consisting of either nodes and edges. The T-Grid is generating the complete mesh from the boundary mesh automatically. Smoothing and refining functions were used to improve the mesh quality. 2. Models and solver strategy The turbulence models and the standard wall functions approach are adopted to simulate the effect of turbulence and the influence of bounded wall on the turbulent flow, respectively. The unsteady-state system of partial differential equations is discretized by means of an accurate finite volume cell-centered method. The solution of the algebraic nonlinear equation system arising from the discretization is performed, for instance, by an under-relaxed Point Gauss–Seidel iterative method employing a multigrid technique. Pressure–velocity coupling was realized by the SIMPLE algorithm. The interpolation to determine the face values of the unknown dependent variables was accomplished using the Second-Order Upwind Scheme, to increase the solution accuracy. During the solution procedure the convergence is monitored dynamically by checking the residuals flow pattern in the entire domain or on selected surfaces. It is keeping track of the residuals throughout the iteration process. The residual plots have reached a specified tolerance. The default value of the residuals is 1.e-3 (except for enthalpy 1.e-6, and transport species 1.e-5).
© 2011 by Taylor and Francis Group, LLC
3. Boundary conditions The boundary conditions of the problem are when the PDF model is used by mean mixture fraction and mixture fraction variance; the input of thermal boundary conditions is provided at the flow inlets. In the PDF model the properties of all fluids such as specific heat capacity, molecular mass, and enthalpies of formation are indicated for all flows. Fluent/UNS uses these conditions as the starting point for its time integration of the particle equations of motion (the trajectory calculations). The PSD is assumed to obey the Rosin–Rammler–Benett law. • Coal char combustion: Phenomenological aspect In the regime of “low temperature,” the chemical reaction rate is slow compared with the diffusion through the pores, because the O2 completely penetrates the char matrix. In this case the rate controlling the regime of char combustion is kinetically limited. Therefore the reaction rate is given by:
πd 2 dmc , g/s =−q 4 dt
(31.71)
4 dmc q= − 2 , g/m 2 /s πd dt
(31.72)
q=−
4 d πd 3 ρ , g/m 2 /s. πd 2 dt 6
(31.73)
The combustion then takes place within the porous char, with the changing of its density, rather than of its diameter, then:
q ≈ d.
(31.74)
On the other hand, the rate of chemical reaction may be expressed by the generalized expression of Arrhenius:
q = A exp(−Ea RTp )ps = kc psm , g/m 2 /s (31.75) where A = pre-exponential factor, g/m2/s; Ea = apparent activation energy, kJ/ mol; R = 82.06 = universal gas constant, m3 × atm × mol−1 × K−1 ; Tp = temperature of char particle, K;
652
Industrial Combustion Testing
ps = partial pressure of O2 on external surface area, atm; m = the true reaction order, estimated from the apparent order (n); kc = the chemical rate coefficient based on char external area, g/m2/s/atm O2; then in this “low temperature” char combustion regime,
Ea = E(true) and d ≈ const.
(31.76)
m = n .
(31.77)
If the temperature is increased, the chemical reaction becomes sufficiently rapid for the diffusion of oxygen through the pores to exert a notable rate limiting effect. So that the reaction rate is given by q ≈ d 0 ≈ pns with Ea =
E m+1 and n = . (31.78) 2 2
The apparent reaction rate does not change as rapidly with temperature. Further increase in temperature eventually causes the chemical reaction to become so rapid that the O2 is consumed as it reaches the outer surface of the char particle. In this case the rate controlling the regime of char combustion is entirely, diffusionally limited. q = kd ( p g − ps ), g/m 2 /s.
(31.79)
where kd = the diffusional reaction rate coefficient, g/m2/s/atm O2; pg = partial pressure of O2 in the free stream, atm; q ≈ d − 1 ≈ pg .
(31.80)
The reaction rate calculated based on the internal surface area,
qi = Ri p gm , g/m 2 /s,
(31.81)
Ri = the intrinsic reaction rate of each coal rank.
4. Post processing The display of the quantitative results of CFD simulation is an effective way to study the
© 2011 by Taylor and Francis Group, LLC
physical problem and to interpret the computed solution. Fluent/UNS can generate graphic displays showing grids, contours, profiles, vectors, and path lines. An example of an application of this technique is given by the courtesy of Italcementi Group, which has awarded a contract to CINAR-UK to find the way to reduce CO emission from coal petcoke blend and ASR fired in line precalciner on 4400 (m)tpd kiln in Belgium. CINAR developed and validated its CFD code namely Mineral Industry-Computational Fluid Dynamics (MI-CFD); it was used to simulate the base-case condition followed by a series of optimization simulations. Only as an example, Figure 31.18 displays the mixing of two streams (i.e., kiln flue gases and TA (tertiary air) stream in the precalciner). The Italcementi Group awarded Fives-Pillard a contract to develop a low NOx burner for a white cement kiln in France. In the first stage of development, a mathematical modeling was carried out to simulate the near burner zone of the burner using the Fluent code. Figure 31.19 displays the contour of static temperature of the flame. 31.7.2 Physical Modeling The practice of using the isothermal physical scale models to visualize and study the furnaces and combustion chambers were extensively used in the mid-1960s. The physical modeling requires the construction of a geometrically scaled laboratory model in transparent Perspex using water or air as working fluid. The acid/ alkali flame modeling technique has been successfully used to study the turbulent nonpremixed flames. The technique simulates the mixing in the flame by using dilute solutions of an acid and alkali to represent the air and the fuel, respectively. The alkali contains a colored indicator (phenolphatalein), which becomes clear on neutralization after mixing with the dilute acid. Since 1964 the technique was used by Dr. Rühland to simulate the rotary cement kiln and currently the FCT Company makes intensive use of it. An example of utilization of this technique is given by the courtesy of Italcementi Group, which awarded FCT International-Australia with a contract to reduce CO emissions from their suspension preheater tower of 4000 (m) tpd rotary kiln at Essroc-USA. The high CO at stage #1 exit results mainly from incomplete combustion of coal from the auxiliary (preheater) burners. Another major source of CO is believed to be the incomplete combustion of the organic carbon in the raw meal fed to the rotary kiln.
653
Cement Kilns
d W 40 35 30 25 20 15 10 5 0 –5 –10
c
b a
Figure 31.18 Mixing of streams, kiln gases. and TA stream in precalciner. (From Cinar Ltd. With permission.)
The FCT built a small-scale physical model of the preheater tower, up to cyclones 1A and 1B, having the scale of a laboratory model to an actual plant of 1:35.5. The model was made of transparent acrylic for visibility inside to visualize the flow streams and mixing (Figure 31.20). The existing burners were modeled as slots on the acrylic model, and new FCT design auxiliary model burners were tested as a possible solution. A water-bead modeling reveals the aerodynamics inside of the riser duct; the kiln exhaust was represented by water in the model. For the visualization of flow, neutrally buoyant polystyrene beads follow the water flow accurately and beads allow the flow streams to be visualized: • A thin light sheet is shone into an area of interest and enables the flow in that area to be visualized. • System is operated with the burner flows off and uniform flow in kiln.
© 2011 by Taylor and Francis Group, LLC
• The existing burners have very low momentum, which combined with the aerodynamics in the riser duct, leads to:
i. Poor burner jet penetration into the riser duct ii. Unburnt fuel being carried by the high velocity stream along the east wall iii. Total mixing length approximately halfway between gooseneck and cyclone #1 • This produces poor mixing of the fuel and kiln exhaust air:
i. Even at very high excess air (high exhaust flow) the mixing length is still too long (approximately to gooseneck). ii. At low excess air (low exhaust flow) the mixing length is extremely long (to cyclone 1). • Hence the high CO in the riser duct is due to the poor exhaust and fuel mixing (see Figure 31.21).
654
Industrial Combustion Testing
2.12e+03 2.05e+03 1.99e+03 1.94e+03 1.88e+03 1.83e+03 1.78e+03 1.72e+03 1.67e+03 1.61e+03 1.56e+03 1.50e+03 1.45e+03 1.39e+03 1.34e+03 1.29e+03 1.23e+03 1.18e+03 1.12e+03 1.07e+03 1.01e+03 9.60e+02 9.05e+02 8.51e+02 7.97e+02 7.42e+02 6.88e+02 6.33e+02 5.79e+02 5.25e+02 4.70e+02 4.16e+02 3.61e+02 X 3.07e+02
Y Z
Contours of static temperature (k)
Oct 04, 2007 Fluent 6.2 (3d, segregated, spe, ske)
Figure 31.19 Contours of static temperature of burner pipe flame. (Courtesy of Fives-Pillard.)
Figure 31.21 Poor exhaust and fuel mixing visualization with the actual burners. (From FCT-Combustion Pty Ltd. With permission.)
31.8 Pollutant Formation and Destruction
Figure 31.20 Small-scale physical model of the preheater tower. (From FCTCombustion Pty Ltd. With permission.)
© 2011 by Taylor and Francis Group, LLC
Since September 2007 in the European Union, a revised draft of the BAT Reference Documents (BREF) is available. The BREF-documents are prepared by a technical bureau of the European Union Commission. The European IPPC Bureau in Seville, Spain has prepared five BREF documents that are directly or indirectly related to waste management. The BREF documents
655
Cement Kilns
are prepared by the Technical Working Groups (TWG) consisting of representatives from different interested parties: experts from EU member states, accession countries, EFTA countries, industry, and environmental NGOs. The contents are agreed upon in the international TWG, CEMBUREAU, as well as VDZ-Germany and are all involved in this process. The definition of BAT is given in the Directive as follows: “The most effective and advanced stage in the development of activities and their methods of operation which indicate the practical suitability of particular techniques for providing in principle the basis for emission limit values designed to prevent and, where that is not practicable, generally to reduce emissions and the impact on the environment as a whole.” • “Techniques” shall include both the technology used and the way in which the installation is designed, built, maintained, operated and decommissioned. • “Available” techniques shall mean those developed on a scale which allows implementation in the relevant industrial sector, under economically and technically viable conditions, taking into consideration the costs and advantages, whether or not the techniques are used or produced inside the Member State in question, as long as they are reasonably accessible to the operator. • “Best” shall mean most effective in achieving a high general level of protection of the environment as a whole. Table 31.16 shows the emissions limits according IPCC BAT Directive 2K. Table 31.16 Emissions Limits According to IPCC BAT Directive 2K IPCC BAT Directive 2K
mg/Nm3 @10% O2 dry
NOx (as NO2) SO2 Dust CO CO2, g/Nm3 TOC HF HCl PCDD/PCDF, ng/Nm3
<200–3000 <10–3500 5–200 500–2000 400–500 5–500 <0.4–5 <1–25 <0.1–0.5
<0.4–6 <0.02–7 0.01–0.4 1–4 800–1040 0.01–1 <0.8–10 g/t <2–50 g/t <200–1000 ng/t
400–6000
0.01–0.3
2–600 mg/t
20–600 kg/y
0.001–0.1 0.005–0.3
2–200 mg/t 10–600 mg/t
2–200 kg/y 10–600 kg/y
Metals ∑ (Hg, Cd, Tl) mainly Hg ∑ (As, Co, Ni, Se, Te) ∑(Sb, Pb, Cr, Cu, Mn) ∑ (V, Sn, Zn)
© 2011 by Taylor and Francis Group, LLC
kg/t cli
t/y <20–7000 10–400 1000–4000 0.8–1.04 M 10–1000 <0.8–10 <2–50 <0.2–1 g/y
Environmental legislation typically stipulates the NOx (NO + NO2) emissions limits in units of mg/Nm3 (dry) @ 10% O2 at the stack. Most important pollutants to control are: • NOx (calculated as NO2) < 200–450 mg/Nm3 @10% O2 dry • SO2 < 50 – < −350 mg/Nm3 @10% O2 dry • Dust < 10 – 20 mg/Nm3 @10% O2 dry Minor pollutants are: HCl, HF, CO, and TOC (including PAH: polycyclic aromatic hydrocarbons, PCB: polychlorinated byphenyls, etc.), NH3, heavy metals (in clinker or dust), volatile metals (Cd, Pb, Tl, Hg), Dioxins (PCDDs), and furans (PCDFs). 31.8.1 NOx Formation in Clinker Burning Process NOx is used to refer to NO and NO2. NO is the primary form in combustion products (typically 95% of total NOx). NO is subsequently oxidized to NO2 in the atmosphere. Nitrogen oxide formation occurs through three reaction paths, each having unique characteristics and are responsible for the formation of NOx during combustion processes: • Thermal NOx, which is formed by the combination of atmospheric nitrogen and oxygen at high temperatures according to Zeldovitch’s mechanism. • Fuel NOx, which is formed from the oxidation of fuel-bound nitrogen. • Prompt NOx, which is formed by the reaction of fuel-derived hydrocarbon fragments with atmospheric nitrogen. Prompt NOx hydrocarbon fragments (such as C, CH, CH2) may react with atmospheric nitrogen under fuel-rich conditions to yield fixed nitrogen species such as NH, HCN, H2CN, and CN. These, in turn, can be oxidized to NO in the lean zone of the flame. In most flames, especially those from nitrogen-containing fuels, the prompt mechanism is responsible for only a small fraction of the total NOx. Its control is important only when attempting to reach the lowest possible emissions. In a cement rotary kiln, NOx formation typically takes place by two mechanisms: one is from atmospheric N2 (thermal NOx) and the second from fuel bond N (fuel NOx). The necessary combustion temperature in the burning zone depends on two main parameters: • The raw material burnability, which is a measure of reactivity of the raw meal fed to the
656
Industrial Combustion Testing
cement rotary kiln, indicating the necessary temperature to achieve a good quality of clinker (free CaO less then 1% w/w). • The clinkering process design. If the raw meal burnability is poor, a high clinkering temperature and therefore high flame temperature is required that results in greater NOx emissions. The main techniques to control NOx formation/reduction are displayed in Table 31.17: • Staged combustion or reburning: Staged fuel and/or air to realize a reducing zone in precalciner of the SPH tower. • Selective noncatalytic reduction (SNCR): Injec tion of ammonia or urea solution in the flue gas in a “temperature window” from 900°C to 1050°C. • Selective catalytic reduction (SCR): Ammonia injection in the front of a catalyst ceramic bed at flue gas temperatures around 350°C. The measures to control NOx by staged combustion (reburning) are presented in Table 31.18. Table 31.17 Main Techniques to Control NOx Formation/Reduction Measures Primary measures
Secondary measures
Reported Abatement Rate
Process Low NOx burner Flame cooling Use of mineralizes Stage combustion in precalciner SNCR SCR SNCR and SCR hybrid combination
0–30 0–50 10–15 10–50 10–50 80–85 Emerging technique
Table 31.18 Measures for NOx Control by Staged Combustion (Reburning) Kinetic Parameters Concentration of O2 Concentration of NOx Concentration of NH3 NCN, hydrocarbons CO, catalyst
Process Parameters
Measures for NO Reduction
λ air ratio NOx at kiln inlet Residence time Mixing of gas streams
Air staging Decrease O2 in rotary kiln Calciner design Fine tuning of burners Meal injection (multiple points) Tertiary air Increase fineness Fuel substitution Meal staging
Fuel properties Temperature
© 2011 by Taylor and Francis Group, LLC
31.8.2 SNCR and SCR Technologies The objective of all NOx control technologies is to reach a high NOx reduction with a minimum consumption of the reductant, while the ammonia slip must be kept low. Selective noncatalytic reduction (SNCR) is typically applied in cement rotary kilns where a flue gas optimum temperature window of 900°C–1100°C is identified, which serves to initiate the breakdown of urea solution to form transient species witch lead to effective NOx reduction. The reductant is an aqueous solution (ammonia water, urea) or gaseous phase (ammonia) are sprayed by dedicated lances into hot flue gas stream, then the overall postcombustion reactions are developed, Urea: NH 2 CONH 2 + 2 NO + 1 2 O 2 → 2 N 2 + CO 2 + 2 H 2 O,
(31.82)
Ammonia: NH 3 + 4NO + O 2 → 4N 2 + 6H 2 O. (31.83) At a lower temperature, the reaction rate will slow down, generating an ammonia slip that may result in the formation of ammonia salts. Therefore the ammonia slip should be kept at minimum. Selective catalytic reduction (SCR) is typical performing in much cooler flue gas zones where the oxidation potential of nitrogen species is minimized. The catalyst provides, on its surface, sites that permit the ammonia and NOx to react near perfect utilization. The extent of NOx reduction is often limited by the local ammonia to NOx ratio, the flue gas temperature, the size of catalyst, and the accepted unreacted ammonia slip. The catalyst size is limited by the available space, the resulting gas side pressure drop, the oxidation from SO2 to SO3, and so on. The urea SNCR process is characterized by the following main four steps: • Distribution and mixing of the liquid spray jet in the flue gas stream • Evaporation of water of the reagent solution • Decomposition of the reagent into reactive species • Gas-phase reaction between NH2 and NOx
31.9 Cement Rotary Kiln Thermal Efficiency 31.9.1 Introduction The heat and mass balance problem/exercise is addressed by considering the rotary cement kiln a combustion system surrounded by a control volume across
657
Cement Kilns
which the fuel, oxidizer, and raw material flow in, and the flue gases and final product flow out. The control volume is an imaginary three dimensional computation cell, enclosing the pyroprocessing system; across their boundaries mass and heat flows are to be determined. The heat balance is concerned only with the heat fluxes, which takes place across the boundaries limits of the control volume. The main objective of the rotary kiln section is to transfer the heat flux from the diffusion flames to the material to allow the clinkering reaction and nodulization in the burning zone. As previously described the heat is transferred by conduction, convection, and radiation mechanisms to the material. 31.9.2 Goals of H&M Balance The performing and interpreting of the heat and mass (H&M) balance provide information on various issues necessary to be addressed: • To account for the heat actually supplied to the cement rotary kiln. • Quantification of various heat losses of the cement rotary kiln. • Allows evaluation of various strategies for changing either the raw material, fuels sources, or plant modification to improve efficiency and performance. • Provides information aimed to achieve maximum output of good product quality at lowest practical energy (thermal and electrical) consumption. • Provides a tool to help the plant diagnostics and performance identification. It is mandatory to perform H&M balance in condition of steady-state operation of cement rotary kiln to cover all the possible measuring points in enough time to provide results that are meaningful and accurate using instrumentations well calibrated and certified by a metrological laboratory. The accuracy of measurements is essential and there are some flows that practically can’t be measured directly, for instance, air in leakage. The main burner as well as the auxiliary burners located on the Lepol grate, or in the riser duct or precalciner for drying process kilns need to be fine-tuned and optimized before undertaking the measures campaigns. 31.9.3 Example of H&M Balance of Cement Rotary Kiln An example of H&M balance is provided by the courtesy of Italcementi Group. The burning line design characteristics are summarized in Table 31.19.
© 2011 by Taylor and Francis Group, LLC
Table 31.19 Design Characteristics of the Burning Line Kiln daily production Specific heat consumption • Suspension preheater tower Cyclone #1 stage Cyclone #2 stage Cyclone #3 stage Cyclone #4 stage • Rotary shell total length Burning zone ID • Clinker cooler Total active grate area • Thermal loads in the burning zone • Gas flow velocities Cooler throat Burning zone Preheater #1 stage
2338 (m)tpd 792 Mcal/t cli. KHD Humboldt 2 × Ø4300 2 × Ø6050 2 × Ø6050 2 × Ø6100 67 m 4.6 m Fuller/Combi 820/1050 + IKN KIDS 54.1 m2 4.54 Gcal/m2h 9.0 m/s 10.7 m/s 16.1 m/s
31.9.3.1 Materials Chemical Analyses Tables 31.20, 31.21, and 31.22 provide the chemical analyses of raw meal feed, clinker, and petroleum coke. Based on that are the calculations of the lime saturation factor, silica ratio and alumina ratio of raw meal, and clinker.
1. Raw meal feed: Its chemical composition is given in Table 31.20. 2. Clinker: Its chemical composition is shown in Table 31.21. 3. Green delayed petroleum coke: The ultimate and proximate analyses are displayed in Table 31.22.
31.9.3.2 Kiln Feed to Clinker Ratio Calculation To calculate the raw meal to clinker ratio, a clinker weighing test was carried out over 20 hours, while the kiln feed value was taken from weigh-feeder counters. The mass flow rate of raw meal 157.3 t/h and the clinker 97.42 t/h. In order to evaluate the preheater efficiency, a preheater dust weigh test was carried out by truck. Such a short test is due to the lack of a capable dust bin. The dust weigh test was performed with the raw mill circuit off indicating a figure of 155 t/h. This value shows that the preheater dust represents 3.7% w/w of raw meal feeding counters indication. It is possible to evaluate the real raw meal flow rate by a loss on ignition, L.o.I.-free mass balance, and hence calculate the real raw meal to clinker ratio. The measured preheater dust flow rate (given at a raw meal feed
658
Industrial Combustion Testing
Table 31.20 Raw Meal Feed Chemical Analysis L.o.I. %w/w
CaO %w/w
SiO2 %w/w
34.85 43.27 13.62 LSF-lime saturation factor SR-silica ratio AR-alumina ratio
Al2O3 %w/w
Fe2O3 %w/w
MgO %w/w
SO3 %w/w
K2O %w/w
Na2O %w/w
Cl− %w/w
Moisture %w/w
+90 μm %w/w
+200 μm %w/w
3.44 98.80 2.45 1.63
2.11 %w/w
0.97
0.01
0.90
0.28
0.01
0.54
19.47
1.40
Table 31.21 Clinker Chemical Analysis L.o.I. %w/w
CaO %w/w
SiO2 %w/w
Al2O3 %w/w
Fe2O3 %w/w
MgO %w/w
SO3 %w/w
K2O %w/w
Free CaO %w/w
Cr6+ ppm
5.19 98.78 2.34 1.45
3.59 %w/w
1.39
1.35
1.14
1.26
6.30
0.14 65.21 20.56 LSF-lime saturation factor SR-silica ratio AR-alumina ratio
Table 31.22
Table 31.24
Ultimate and Proximate Analyses
Total Specific Heat Consumption (SHC) Calculation
Units Ultimate Analysis C H S O (by difference) Moisture Net calorific value
82.5 3.64 2.01 4.47 0.69 7904
%w/w %w/w %w/w %w/w %w/w kcal/kg
Proximate Analysis Moisture Ash Volatile matter Fixed carbon
0.69 6.00 12.37 80.94
%w/w %w/w %w/w %w/w
Main burner Auxiliary burners Total specific heat consumption, kcal/kg
Preheater Dust (SPD)
Fuel Ash (FLH)
Raw Meal (RM)
0.14
34.85
0.00
34.85
flow rate of 155 t/h) has to be actualized to the clinker test’s raw meal feed flow rate (157.30 t/h): 157.3 = 5.95 t / h = 0.06 kg/kg cli. 155
Chemical analysis gave as a result the following L.o.I. for each component and is given in Table 31.23 (preheater dust chemical analysis has been assumed equal
© 2011 by Taylor and Francis Group, LLC
Shared Heat, %
7.64 2.11
7904 7904
620 171 791
78.4 21.6
The green delayed petroleum coke has been used both for main and auxiliary burners with the mass flow rates of 7.64 t/h and 2.11 t/h, respectively (see Table 31.24). Total heat consumption is 791 kcal/kg cli and 21.6% of heat is shared at the auxiliary burner.
Clinker (CLI)
SPdust actual = 5.86 ×
SHC kcal/kg
31.9.3.3 Specific Heat Consumption Calculation
L.o.I. Percentage Analysis
NCV kcal/kg
to the kiln feed). The raw meal flow rate measured by the counters was 157.3 t/h and the real raw meal to clinker ratio was 1.568.
Table 31.23
L.o.I (%w/w)
Flow Rate t/h
31.9.3.4 Clinker Heat of Formation The clinker heat of formation only depends on the physical–chemical properties of the raw meal fed to the kiln. For such calculation, it is conventional to use the isothermal analytical method, which consists in quantifying step-by-step the heat required by each chemical and physical transformation of the material during the burning process. The reference temperature to which all heat quantities are referred is 273.16 K. The final heat of reaction is obtained subtracting from the total heat required for endothermic transformations the heat released by the exothermic transformations.
659
Cement Kilns
31.9.3.4.1 Raw Meal and Clinker Chemical Analysis The raw mix mineralogical composition is obtained from the average sample chemical analysis multiplying the present oxides ( a L.o.I. free basis) by the molar mass/ mineral ratio. The calcium and magnesium released CO2 can be calculated multiplying their percentages by the ratio between the molar masses of carbon dioxide and the oxides themselves:
the mineralogical composition of the raw meal (see Table 31.27). 31.9.3.4.3 Clinker Silicates Using Bogue Formula The Bogue formula is used in order to evaluate the clinker silicate constituents starting from the clinker chemical analysis (where C, S, A, F are the relative oxides). The clinker mineralogic composition is displayed in Table 31.28.
Potential Carbonates L.o.I. = 43.27 × 44/56 + 0.97 31.9.3.4.4 Heat of Formation (Isothermal Method) According to Zur-Strassen × 44/40.3 = 35.06%. This method allows an easier calculation of the reacThe raw meal L.o.I. is also due to the vaporization of tion energy since it only considers irreversible reacresidual free water, bound water in kaolin, and small quan- tions and leaves out all sensible heat exchanges. The tities of minor compounds such as sulfides, ammonia, and reference temperature for the reactions is 273.16 K. The so on. In this case the potential carbonates L.o.I. value is formation heat is shown in Table 31.29. The total heat of higher than the dry raw meal L.o.I. value; such situations formation is the difference between endothermic and might be justified by an incomplete carbonation of calcium exothermic reactions heat, which is 418.5 kcal/kg cli. and magnesium oxides. It is reasonable to think that part of CaO is not present as carbonate but as oxide (Table 31.25). 31.9.3.5 Combustion Engineering Calculation Clinker chemical analysis is shown in Table 31.26. The flue gas flow rate calculations are based on the following input data: 31.9.3.4.2 Raw Meal Mineralogical Composition Assuming that all the alumina is present as kaolin • Fuel chemical analysis (2SiO2*Al2O3*2H2O), by means of simple stoichiometric • Fuel flow rates calculations (molar mass ratios) it is possible to identify • Raw and hot meals L.o.I. Table 31.25 • Gas analysis along gas circuit Raw Meal Characteristics • Relative moisture and temperature of ambient Raw Dry Raw Raw Meal Dry Raw air.
Meal, %w/w
Meal, %w/w
L.o.I. Free, %w/w
Meal L.o.I. Free, %w/w
34.85 43.27 13.62 3.44 2.11 0.97 0.01 0.01 0.54
34.31 43.51 13.70 3.45 2.12 0.97 0.01 0.01 −
66.24 20.91 5.27 3.23 1.49 0.02 0.02 0.83
66.60 21.02 5.30 3.25 1.50 0.02 0.02 −
L.o.I. CaO SiO2 Al2O3 Fe2O3 MgO SO3 Cl– Moisture
Clinker Chemical Analysis
65.21
31.9.3.5.1 Combustion Stoichiometry Thermochemistry is concerned with the description of equilibrium states of reacting multicomponent systems. Stoichiometry: the stoichiometric quantity of oxidizer O2 is just that amount needed to completely burn a quantity of fuel: [[1 kg]fuel + [(1/(F/A)stc)kg]air]reactants → [(1 + 1/(F/A)stc) kg]products, (31.84)
Table 31.26 CaO %w/w
The ambient air will be considered at 10°C with 70% relative humidity (see Table 31.30).
SiO2 %w/w
Al2O3 %w/w
Fe2O3 %w/w
MgO %w/w
SO3 %w/w
Free CaO %w/w
20.56
5.19
3.59
1.39
1.35
1.26
where F/A ≡ mass of fuel/mass of air ratio. The stoichiometric reaction can be written on terms of 1 mole of fuel in reaction with air
Table 31.27 Raw Meal Mineralogical Composition Mineralogical Composition kg/kg cli
© 2011 by Taylor and Francis Group, LLC
CaCO3
MgCO3
Kaolin
Bound Water
Quartz
Others
1.183
0.031
0.134
0.019
0.148
0.051
660
Industrial Combustion Testing
Excess of air: e = 100*(1 - Φ)/Φ = 100*(λ – 1), %. (31.91)
Table 31.28 Clinker Mineralogical Composition Clinker Silicates Compositions %w/w
C3S
C2S
C3A
C4AF
60.20
13.60
7.69
10.9
Table 31.29 Heat of Formation Exothermic Reactions
kcal/ kg cli
Kaolin dehydration CaCO3 decarbonation MgCO3 decarbonation Liquid phase formation
19.4 496.8 9.7 5.0
Exothermic Reactions Silicate formation C3S C 2S C3A C4AF Metakaolin crystallization @ 960°C Total exothermic reactions
kcal/ kg cli 77.7 23.7 0.5 2.3 8.3 112.5
The effectiveness of combustion depends greatly on the air to fuel ratio. Insufficient presence of oxygen will cause incomplete combustion, which will result in unburned hydrocarbon products and carbon monoxide in the combustion products. In most combustion calculations it is conventional to approximate dry air as a mixture of 79% v/v N2 and 21% v/v O2 or 3.764 moles of N2 per mole O2. Stoichiometric oxygen:
Table 31.30 Relative humidity Temperature Pressure Density O2 in vol. on dry air Ambient wet air composition H2O N2 O2
70 10 987 1.293 20.95 %v/v 0.86 78.37 20.77
kg O2 32 16 32 C+ H+ S−O . (31.92) 12 2 32 kg fuel
Stoichiometric air or theoretical air to fuel ratio is given by:
Ambient Air Physical Properties
O 2stc =
100 1 Nm 3air A F = O 2stc × 23 × 1.293 kg . (31.93) fuel stc
The obtained has to be referred to the clinker roduced, multiplying it by the so-called index of conp sumption (IC), which is the fuel necessary quantity to produce 1 kg of clinker.
% °C mbar abs kg/Nm3 %v/v %w/w 0.54 76.34 23.12
31.9.3.5.2 Combustion Flue Gas Composition The combustion flue gas products are calculated as follows and expressed in Nm3/kg cli:
CxHy + a(O2 + 3.76 N2) → xCO2 + (y/2)H2O + 3.76 aN2 where a = x + y/4. (31.85)
CO 2 =
18 22.4 × IC, H 2 O = H + H 2 O fuel × 2 18
The stoichiometric air-to-fuel ratio can be defined as: (A/F)stc = (mair/mfuel)stc = (4.76 a/1)(Mair/Mfuel). (31.86) • Air equivalence ratio λ:
λ = (A/F)/(A/F)stc.
(31.87)
• Fuel equivalence ratio Φ Φ = (F/A)/(F/A)stc = (A/F)stc/(A/F) → Φ = 1/ λ. (31.88)
Fuel reach mixtures: ∞ > Φ > 1 or λ < 1 Stoichiometric mixture: Φ = 1 or λ = 1 (31.89) Fuel lean mixture: 0 < Φ < 1 or λ > 1
© 2011 by Taylor and Francis Group, LLC
(31.90)
44 22.4 C× × IC, 12 44
SO 2 =
64 22.4 S× × IC, 32 64
N2 = N ×
22.4 × IC + N stc,air . 28
(31.94)
(31.95) (31.96) (31.97)
Where C, H, S, and O are expressed in percentage w/w and moist is the PF moisture. 31.9.3.5.3 Kiln Feed Dehydration Preheater flue gas contains moisture coming from kiln feed moisture (0.5% w/w) and bound water that is the water obtained with Kaolin dehydration. 31.9.3.5.4 Kiln Feed Decarbonation Calculation The flue gas contains CO2 coming from kiln feed decarbonation (which has to be partially considered for the flue gas
661
Cement Kilns
coming from the kiln and totally for the flue gas coming from the preheater). From the chemical analysis it is possible to calculate such value. For this purpose it is assumed that the preheater dust has the same chemical composition as the kiln feed. Total realized CO2 comes from the effective CaCO3 and MgCO3 present in kiln feed. 31.9.3.5.5 Excess of Air Calculation To ensure the complete combustion, the burning process is usually carried out with an excess of air. The O2 value (% v/v dry) in the flue gas has been monitored by plant instrumentations (kiln back end, 4th stage outlet) and by spot measures carried out during the heat balance (cyclones outlet). The excess of air (referred to a dry flue gas) is calculated as follows:
O2 = 0.21 × excess of air [% dry basis],
To calculate the flue gas flow rate it is necessary to know the decarbonation degree of the hot meal (material coming from the lowermost stage into the kiln): the calculation has been based by assuming that 60% of the residual decarbonation is done in the kiln. The oxygen value at the kiln outlet is the average measure given by field instrumentations. The flue gas at the kiln outlet is composed by (see Table 31.31): • Stoichiometric combustion flue gas • Residual decarbonation CO2 from hot meal • Excess of air
(31.98)
31.9.3.5.7 Flue Gas from 4th Stage of Suspension Preheater O2 × (theoretical dry flue gas + excess The flue gas composition at the 4th stage of the suspension preheater is given in Table 31.32. air) = 0.21 × excess air,
O2 = 3.01% I.C. = 0.078 kg fuel/kg cli (main burner only)
(31.99)
O2 = 0.91% I.C. = 0.100 kg fuel/kg cli (main + auxiliary burners)
O2 % λ= × theoretical dry flue gas, (31.100) 0.21-O2 %
theoretical dry flue gas = theoretical combustion flue gas – fuel moisture – combustion water + decarbonation CO2. 31.9.3.5.6 Flue Gas at Kiln Outlet Using the above mentioned formulas it is possible to calculate flue gas flow rate at the kiln outlet, knowing that:
31.9.3.5.8 Suspension Preheater Outlet Flue Gas O2 = 3.69% I.C. = 0.100 kg fuel/kg cli (main + auxiliary burners) The flue gas composition at the outlet of the suspension preheater is given in Table 31.33.
Table 31.31 Kiln Back End Flue Gas Composition Flue Gas Composition
CO2
CO
H2O
N2
SO2
Fuel Raw meal Excess air Total Dry flue gas Wet flue gas
0.121 0.160*
0.000
0.038
0.521
0.002
0.001 0.039
0.107 0.628 66.83 64.14
0.002 0.26 0.25
Nm3/kg cli Nm3/kg cli Nm3/kg cli Nm3/kg cli % %w/w
0.281 29.90 28.69
0.000 0.00 0.00
4.03
O2
0.028 0.028 3.01 2.89
Total
Flow Rate Nm3/h
0.683 0.160 0.136 0.979
95,372
*The hot meal kiln inlet L.o.I has been considered equal to 24% w/w.
Table 31.32 4th Stage Flue Gas Composition Flue Gas Composition
CO2
CO
H2O
N2
SO2
Fuel Raw meal Excess air Total Dry flue gas Wet flue gas
0.154 0.268
0.000
0.049
0.665
0.003
0.422 37.06 35.53
0.000 0.00 0.00
0.000 0.049
0.039 0.704 61.76 59.20
0.003 0.27 0.26
Nm3/kg cli Nm3/kg cli Nm3/kg cli Nm3/kg cli % %
© 2011 by Taylor and Francis Group, LLC
4.15
O2
0.010 0.010 0.90 0.87
Total
Flow Rate Nm3/h
0.871 0.268 0.050 1.189
115,805
662
Industrial Combustion Testing
Table 31.33 Suspension Preheater Outlet Flue Gas Composition Flue Gas Composition
CO2
CO
H2O
N2
SO2
Fuel Raw meal Excess air Total Dry flue gas Wet flue gas
0.154 0.268
0.000
0.665
0.003
0.422 31.91 29.99
0.000 0.00 0.00
0.049 0.034 0.002 0.085
0.184 0.849 64.16 60.30
0.003 0.24 0.22
Excess air
Nm3/kg cli Nm3/kg cli Nm3/kg cli Nm3/kg cli %v/v % v/v
6.02
Flow Rate Nm3/h
0.871 0.302 0.235 1.409
137,231
Gas Composition at Preheater Outlet and Specific Heat at Constant Pressure @ 327°C
Air inleakage Auxiliary primary air KILN
0.049 0.049 3.69 3.47
Total
Table 31.34
Combustion air
SP
O2
CO2 N2 O2 SO2 H2O
Secondary air Main primary air
Figure 31.22 Sketch for rotary kiln section air circuits.
31.9.3.5.9 Total Air Inleakage The total air inleakage can be calculated by balancing all types of air flow rates involved in the system (Figure 31.22): Total air inleakage = total theoretical combustion air + excess air – total primary air – secondary air = (0.663 + 0.183) + 0.235 – (0.057 + 0.021) – 0.743 = 0.262 Nm3/kg cli = 25,498 Nm3/h.
c–p (kcal/ Nm3 °C)
%v/v Wet Flue Gas
0.451 0.315 0.327 0.483 0.376
29.99 61.37 3.69 0.22 6.02
cp = the gas specific average heat (considering 0°C as reference temperature), kcal/ Nm3 °C; Δt = the gas temperature difference, °C. Dividing then Q by the clinker production, kg cli/h, it is possible to obtain a “specific” enthalpy expressed in kcal/kg cli. The specific average heat has been calculated with a weighed average of each single gas component’s specific heat at gas measured temperature (t = 327°C). cp = cpCO2 × CO 2 + cpN2 × N 2 + cpO2 × O 2 + cpH2O × H 2 O.
Most of the air inleakage is localized between the 4th and the 3rd stage cyclones where the oxygen value suddenly rises up from 0.9 v/v to 2.9% v/v.
(31.102)
The gas composition and specific heat at suspension preheater outlet is shown in Table 31.34. Hence:
31.9.3.6 Gas and Material Heat Flow Rate Calculation 31.9.3.6.1 Preheater Outlet Gas Heat Flow Rate The following formula is used in order to calculate the preheater outlet gas heat flow rate:
.
Q = V × cp × ∆t ,
where Q = the heat flow rate, kcal/h; . V = the gas volume flow rate, Nm3/h;
© 2011 by Taylor and Francis Group, LLC
(31.101)
kcal cp = 0.360 kcal/Nm 3 °C and Q = 165.9 . kgcli
31.9.3.6.2 Inlet and Outlet Air Heat Flow Rate The inlet and outlet air heat flow rates are displayed in Table 31.35. 31.9.3.6.3 Material Heat Flow Rate The material heat flow rate is given in Table 31.36.
663
Cement Kilns
31.9.3.6.4 Raw Meal Residual Moisture Evaporation Heat It is the required heat to evaporate the residual moisture of kiln feed: Raw meal residual moisture heat = 1.568 × 0.54/100 × 597.2 = 5.1 kcal/kg cli. 31.9.3.7 Shell Heat Losses The shell temperatures were monitored on the rotary kiln, preheater, and clinker cooler with an optical pyrometer. Shell thermal losses are given by a radiation and either natural or forced convection heat transfer, which are evaluated as follows (see Table 31.37 and Figure 31.23). 31.9.3.8 Heat Balance The summary of the heat and mass balance are given in Table 31.38. 31.9.3.9 Mass Balance The in-out mass balance is shown in Table 31.39. The air mass balance is shown in Table 31.40. The clinker cooler mass balance is given in Table 31.41. 31.9.3.10 Clinker Cooler The clinker inlet temperature measured at kiln hood discharge with an optical pyrometer is given as average 1430°C. Clinker outlet temperature was measured Table 31.35 Inlet and Outlet Air Heat Flow Rates
Primary air Cooler air Preheater air in leakage Exhaust air Kiln hood dedusting
Flow Rate (Nm3/ kg cli)
Heat (kcal/ kg cli)
Temperature (°C)
cp (kcal/ Nm3 °C)
0.077 1.902 0.262
30 10 10
0.309 0.309 0.309
0.7 5.9 0.8
1.16 0.050
330 156
0.316 0.312
121.0 2.4
Table 31.36 Heat of Material Flow Rates
Fuel Raw meal Clinker Preheater dust
Mass Flow Rate (kg/ kg cli)
Temperature (°C)
cp (kcal/ Nm3 °C)
Heat (kcal/ kg cli)
0.100 1.568 1 0.061
30 50 120 327
0.235 0.201 0.219 0.234
0.7 15.8 26.3 4.7
© 2011 by Taylor and Francis Group, LLC
Table 31.37 Shell Heat Losses Ambient temperature Clinker production Kiln shell inlet diameter Kiln length Emissivity Kiln specific load Kiln shell total heat losses (radiation and convection) Suspension preheater total heat losses (radiation and convection) Clinker cooler total heat losses (radiation and convection)
10°C 2338 tpd 4.6 m 67 m 0.95 2.483 tpd/m3 41.2 kcal/kg cli 13.0 kcal/kg cli
4 kcal/kg cli
taking a sample at the cooler outlet on the pan conveyor as average 120°C. Total clinker cooling air flow rate (185,265 Nm3/h) was measured during the balance. Part of this air is used as secondary air (72,355 Nm3/h), the rest is exhaust going to the downstream dedusting system. During the balance, no hot air from the cooler was used for coal grinding. Since a reliable flow measure is not possible (lack of measuring points), the exhaust air has been calculated by the difference among total cooling air and secondary air. The secondary air flow rate has been calculated by balancing the kiln air flow rates (see Figure 31.24). The primary air flow rate has been measured (all the measured flow rates are resumed in Table 31.40). The exhausted air flow rate (112,910 Nm3/h) is lower than the exhausted air flow rate measured at the cooler stack (130,700 Nm3/h). The difference is given by: • Kiln nose ring dedusting air flow rate (4900 Nm3/h) • Clinker transport dedusting flow rate (7700 Nm3/h) • Air cooled heat exchanger, bag filter, and ducts air in leakages (5190 Nm3/h) Exhausted air temperature was measured at the cooler outlet and resulted to be 330°C. The clinker cooler heat balance is exposed hereafter in Table 31.41. The shown secondary air temperature (955°C) has been calculated to balance the inlets with the outlets since it not possible to get reliable measures in the field. The cooler’s efficiency, expressed as the ratio between the heat recovered as combustion air and the heat from the incoming material (clinker) is equal to 237/376 × 100 = 63%. Actual clinker cooler specific load is 42.98 tpd/m2. 31.9.3.11 Flow Rates Chart Table 31.42 displays the flow rates.
664
Industrial Combustion Testing
350
Kiln shell temperature (°C)
300 250 200 150 100 50 0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67 Kiln length (m)
Figure 31.23 Kiln shell temperatures.
31.9.3.12 Gas Analysis Chart
Specific heat consumption:
Table 31.43 summarizes the measures carried out along the preheater tower.
31.9.3.13 Electrical Specific Consumption The current electrical consumption is shown in Table 31.44. Such values have been taken during the clinker test (19 hours 53 minutes) and hence referred to its respective clinker production. 31.9.3.14 Appendix Formulas of H&M Balance Real raw meal calculation:
FFR × LHV , CLI
(31.105)
where SHC = specific heat consumption, kcal/kg cli; CLI = clinker, kg cli/h; FFR = fuel flow rate, kg/h; LCV = low calorific value, kcal/kg. Mineralogical composition determination: CaCO 3MM 100 = CaO × , kg/kg cli* CaO MM 56 (31.106)
CaCO 3 = CaO ×
RM real × (100 − L.O.I.RM ) + FLH × (100 − L.O.I.FLH ) = CLI × (100 − L.O.I.CLI ) + SPD × (100 − L.O.I.SPD ), (31.103) RM real =
SHC =
CLI × (100 − L.O.I.CLI ) + SPD × (100 − L.O.I.EPD ) − FLH × (100 − L.O.I.FLH ) , (100 − L.O.I.RM )
where RM = raw meal, kg; CLI = clinker, kg; SPD = preheater dust, kg; FLH = fuel ash, kg; L.o.I. = loss on ignition, %.
© 2011 by Taylor and Francis Group, LLC
(31.104)
* It has to be reminded that in our case, since the potential carbonates were higher than the raw meal L.o.I., it has been assumed that not 100% of CaO becomes CaCO3 in order to match the potential carbonates value with the actual raw meal L.o.I.
665
Cement Kilns
Table 31.38 Summary of the Heat and Mass Balance Temperature (°C)
Flow Rate* Inputs Fuels a. Main burner
Enthalpy† (kcal/kg cli)
Heat Load (Gcal/h)
619.9 0.6
60.4 0.1
76.1 0.1
171.2 0.2 15.8 0.5 0.2 5.9 0.8 815.1
16.7 0.0 1.5 0.1 0.0 0.6 0.1 79.4
21.0 0.0 1.9 0.1 0.0 0.7 0.1 100
418.5
40.8
51.9
(%)
− Combustion heat Fuel sensible heat
7.64 7.64
t/h t/h
0.078 0.078
kg/kg cli kg/kg cli
30
t/h t/h t/h Nm3/h Nm3/h Nm3/h Nm3/h
0.022 0.022 1.568 0.057 0.021 1.902 0.262
kg/kg cli kg/kg cli kg/kg cli Nm3/kg cli Nm3/kg cli Nm3/kg cli Nm3/kg cli
− 30 50 30 30 10 10
b. Auxiliary burner Combustion heat Fuel sensible heat Raw meal Sensible heat Primary air to main burner Primary air to auxiliary burners Cooler air Total preheater air in leakage Total inputs Outlets Heat of formation (clinker) Preheater outlets Moisture evaporation Hot gas sensible heat Preheater dust sensible heat Unburned Cooler outlets Clinker sensible heat Exhaust air sensible heat Kiln outlets Kiln hood dedusting sensible heat‡ Shell losses Cooler Kiln Preheater Total outlets Indeterminations
2.11 2.11 157.0 5535 2000 185,265 25,498
19.8 137.231 6.0
t/h Nm3/h t/h
0.008 1.409 0.061
kg/kg cli Nm3/kg cli kg/kg cli
327 327
5.1 165.9 4.7
0.5 16.2 0.5
0.6 20.6 0.6
113.22 97.4 112.910
Nm3/h t/h Nm3/h
0.001 1.000 1.159
Nm3/kg cli kg/kg cli Nm3/kg cli
− 120 330
3.5 26.3 121.0
0.3 2.6 11.8
0.4 3.3 15.0
4,900
Nm3/h
0.050
Nm3/kg cli
156
2.4
0.2
0.3
4.0 41.2 13.0 806.0 9.5
0.4 4.0 1.3 78.5 1.2
0.5 5.1 1.6 %
* The term “flow rate” has to be meant as the ratio between a mass flow rate or a volume flow rate (t/h or Nm /h) and the clinker mass flow rate (kg cli/h), then expressed as either kg/kg cli, or Nm3/kg cli. † The term “enthalpy” (kcal/kg cli) has to be meant as the product given by the previously shown flow rate (kg/kg cli or Nm3/kg cli), the material specific heat (kcal/kg °C) and the temperature (°C). ‡ This term comes from the air used for the nose-ring cooling and sent to the heat exchanger. The reason is this has to be considered as part of heat outcoming from the system.
MgCO 3MM 84.3 = MgO × , kg/kg cli MgO MM 40.3 (31.107)
3
Bound H 2 O = Kaolin ×
MgCO 3 = MgO ×
Kaolin MM 258 = Al 2 O 3 × , kg/kg cli Al 2 O 3MM 102 (31.108) Kaolin = Al 2 O 3 ×
© 2011 by Taylor and Francis Group, LLC
Bound H 2 O MM Kaolin MM
36 , kg/kg cli (31.109) 258 Quartz MM Quartz = SiO 2 − Kaolin × Kaolin MM = Kaolin ×
= SiO 2 − Kaolin ×
120 , kg/kg cli (31.110) 258
666
Industrial Combustion Testing
where, taking as example the first equation, 100 kg/kmol is CaCO3 molar mass, while 56 kg/kmol is CaO molar mass.
Table 31.39 In-Out Mass Balance
Inputs Raw meal Fuels Kiln primary air Auxiliary burners primary air Air inleakage Cooling air Total inputs
Density kg/Nm3
Flow Rate (kg/ kg cli)
%
1.293 1.293
1.568 0.100 0.073 0.027
34.34 2.19 1.61 0.58
0.338 2.459 4.565
7.41 53.86
1.293 1.293
Outputs Clinker Preheater dusts Preheater outlet flue gas Exhaust air Total outputs Indetermination
Kaolin = ( 2 SiO 2 × Al 2 O 3 × 2 H 2 O ).
Hereafter some elements and their molar mass are shown: 1 H 1.008
6 C 12.01
8 O 16.00
12 Mg 24.31
13 Al 26.98
14 Si 28.09
16 S 32.07
17 Cl 35.45
20 Ca 40.0
Clinker composition determination: 1.000 0.061 2.014 1.499 4.574 −0.009
1.430 1.293
21.86 1.34 44.04 32.76
−0.20
Cres = C − CaO free − 0.7 SO 3
(31.111)
C3S = 4.07 Cres − ( 6.72 A + 7.6S + 1.4F ) (31.112)
C2 S = 2.87 S − 0.754C3S
(31.113)
C3A = 2.65A − 1.69F
(31.114)
C 4 AF = 3.04F
(31.115)
Table 31.40 Air Mass Balance Specific Flow Rate (Nm3/ kg cli)
Flow Rate Nm3/h
Rotary Kiln Air Flow Rate Input (Measured) Primary air to main burner 0.057 Total input 0.057
5535 5535
Rotary Kiln Air Flow Rate Output (Calculated) Excess air 0.136 Stoichiometric combustion air 0.663 Total output 0.800
13,263 64,626 77,890
Clinker heat of formation (Zur-Strassen method): Endothermic reactions:
kcal kg Kaolin dehydration: Kaolin × 143 = × kg cli kg
Table 31.41 Clinker Cooler Heat Balance
Mass Flow Rates Cooler Inlets Clinker Cooling air Total Cooler Outlets Clinker Secondary air Exhaust air Shell losses Total
© 2011 by Taylor and Francis Group, LLC
t (°C)
cp (kcal/ kg°C)
Heat Load (kcal/kg cli)
97,420 kg/h 185,265 Nm3/h
1 kg/kg cli 1.90 Nm3/kg cli
1430 15
0.263 0.309
376.44 8.81 385.26
97,420 kg/h 72,355 Nm3/h 112,910 Nm3/h
1.00 kg/kg cli 0.74 Nm3/kg cli 1.16 Nm3/kg cli
120 955 330
0.189 0.335 0.316
22.73 237.51 121.02 4 385.26
(31.116)
667
Cement Kilns
Table 31.43 Excess air
Suspension Preheater Gas Analysis
Primary air
Combustion air
Secondary air
Figure 31.24 Sketch with kiln air flow rates balance.
Point of Measures
O2 %v/v
CO2 %v/v
t (°C)
p daPa
IV stage exit III stage exit II stage exit Preheater exit
0.91 2.81 2.93 3.69
34.33 29.40 30.70 28.70
836 652 522 327
−147 −305 −454 −630
Table 31.42 Table 31.44
Flow Rates Measuring Point Exhaust air stack IKN fan Cooler fan 2 Cooler fan 3 Cooler fan 4 Cooler fan 5 Cooler fan 6 Cooler fan 7 Kiln hood dedusting Cooler dedusting Main burner primary air
t °C
p daPa
Velocity m/s
Flow Rate m3/s
Cross Area m2
Flow Rate Nm3/h
72
−25
19.2
50.1
2.6
136,290
13 15 15 15 15 15 15 156
−30 −18 −17 −17 −18 −36 −253 −32
17.7 15.3 20.7 14.3 12.7 19.7 21.5 11.6
5.5 7.7 10.4 8.7 7.7 12.0 3.9 2.3
0.3 0.5 0.5 0.6 0.6 0.6 0.2 0.2
18,075 25,185 34,105 28,500 25,245 39,155 12,405 4915
60
−85
28.4
2.7
0.1
7660
15
−41
22.5
1.7
0.1
5535
Kiln Specific Electrical Consumption Machines
kWh/t
ID fan Final fan #1 Final fan #2 Kiln hood Kiln drive
9.29 9.91 0.50 1.67
Cooler Fans Conditioning tower Cooling tower compressors ESP (now bag filter) CTP filter CTP final fan CTP heat exchanger Kiln shell fans Total
kcal kg CaCO 3 decarbonation: CaCO 3 × 420 = × kg cli kg (31.117)
3.96 0.99 0.82 0.66 0.83 2.32 0.99 1.49 33.42
kcal kg C 4AF: C 4AF × 21 = × kg cli kg
(31.122)
Metakaolin crystallization at 960°C: kcal kg MgCO 3 decarbonation: MgCO 3 × 310 = × kg cli kg (31.118)
kcal kg C2 S: C2 S × 174 = × kg cli kg
kcal kg C3A: C3A × 7 = × kg cli kg
© 2011 by Taylor and Francis Group, LLC
( Kaolin − Bound H 2 O ) × 72 =
kg kcal (31.123) × kg kg cli
Hot meal residual decarbonation:
Exothermic reactions: Silicate formation: kcal kg C3S: C3S × 129 = × kg cli kg
HM =
CO 2 = HM ×
(31.119)
(31.120)
(31.121)
1 − FLH , HM L.o.I. 1− 100 HM L.o.I. 22.414 × , 100 44
where HM = hot meal, kg/kg cli; FLH = fuel ash, kg/kg cli; HML.o.I. = hot meal lost on ignition, %.
(31.124)
(31.125)
668
Industrial Combustion Testing
Heat losses: Radiation heat losses:
3.6 Q R = S × ε × σ × ( ts4 − ta4 ) × , kcal (31.126) 4.187
Convection heat losses:
3.6 t −t , Q C = S × 11.3 × 3 s a ⋅ ( ts − ta ) × 4.187 ta
kcal for vair ≤ 2 m/s (natural convection)
,
(31.127)
1 3.6 0.7 , Q C = S × 4.6 × ( vair ) × 0.3 × ( ts − ta ) × 4.187 D
kcal for vair > 2 m/s (forced convection)
, (31.128)
where ε = the emissivity factor (0.95); σ = the Boltzman constant (5.67 × 10−8W/m2 K4); ts = the surface temperature, K; ta = the ambient air temperature, K; νair = the speed of ambient air, m/s; S = the considered surface, m2; D = the shell diameter, m.
References
1. Hewelett, P. C., ed. Lea’s Chemistry of Cement and Concrete. Oxford, UK: Elsevier B-H, 1998. 2. Boateng, A. A. Rotary Kilns-Transport Phenomena and Transport Processes. Oxford, UK: Elesvier B-H, 2008. 3. Ricou, F. P., and Spalding, D. B. “Measurements of Entrainment by Axi-Symmetrical Turbulent Jet.” Journal of Fluid Mechanics 11 (1961): 21–32. 4. Craya, A., and Curtet, R. “Sur l’évolution d’un jet en espace confiné” (in French). Comptes-Rendus de l’Académie des Sciences, Paris 241 (1) (1955): 621–22. 5. Thring, M. W., and Newby, M. P. “Combustion Length of Enclosed Turbulent Jet Flames.” 4th Symposium (International) on Combustion, Baltimore, MD, 789–96, 1953. 6. Beér, J. M., and Chigier, N. A. Combustion Aerodynamics London: Applied Science Publishers Ltd., 1972. 7. Gupta, A. K., and Lilley, D. G. Flowfield Modeling and Diagnostics. Cambridge, MA: Abacus Press, 1985.
© 2011 by Taylor and Francis Group, LLC
8. Gupta, A. K., Lilley, D. G., and Syred, N. Swirl Flows. Cambridge, MA: Abacus Press, 1984. 9. Curtet, R. “Confined Jets and Recirculation Phenomena with Cold Air.” Combustion and Flame 2, no. 4 (1959): 383–411. 10. Barchilion, M., and Curtet, R. Some Details of the Structure of an Axisymmetric Confined Jet with Backflow, Journal Basic Engineering (Transaction ASME), 86 (1964): 777–87. 11. Stambuleanu, A. Flame Combustion Processes in Industry. Cambridge, MA: Abacus Press, 1976. 12. Jenkins, G., and Moles, F. D. “Modeling of Heat Transfer from Large Enclosed Flame in a Rotary Kiln.” Transactions of the Institution of Chemical Engineers 59 (1981): 17–25. 13. Beltagui, S. A., and Maccallum, N. R. L. “Aerodynamics of Vane Swirled Flames in Furnaces. The Modeling of Vane Swirled Flames in Furnaces.” Journal of the Institute of Fuel 49, no. 401 (1976): 183–193. 14. Kerr, N. M., and Fraser, D. “Swirl: Part 1: Effect on Axisymmetrical Turbulent Jets, Part 2: Effect of Flame Performance and the Modeling of Swirling Flames.” Journal of the Institute of Fuel 38 (1965): 519–526. 15. Cristea, E. D. “Contribution to Research on Twin Fluid Spray Burners for Industrial Furnaces.” PhD Diss., Polytechnic University of Timisoara-Romania, 1984. 16. Cristea, E. D. “Prediction of the Central Recirculation Zone Size for a Complete Burner-Quarl-Furnace System.” Institute of Aeronautics and Astronautics Journal 25, no. 3 (1987): 457–463. 17. Curtet, R. “Confined Jets and Recirculation Phenomena with Cold Air.” Combustion and Flame 2 no. 4 (1958): 383–411. 18. Mathur, M. L., and Maccallum, N. R. L. “Swirl Effect on Axisymmetrical Turbulent Jets.” Journal of the Institute of Fuel 38 (1965): 519–526. 19. Spalding, D. B. “Colloquim on Modelling Principles, the Art of Partial Modelling.” Ninth Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, 1962. 20. Van de Kamp, W. L., and Smart, J. P. “Cemflame I: The Effects of Burner Design and Operation and Fuel Type on the Properties of Cement Kiln Flames.” International Flame Research Foundation Doc. No. F 97/y/1, Livorno, Italy, 1992. 21. Haas, J., Agostini, A., Martens, C., Carrea, E., and Van de Kamp, W. L. “Cemflame 3: The Combustion of Pulverized Coal and Alternative Fuels-Results on the Cemflame 3 Experiments.” International Flame Research Foundation Doc. No. F97/y/4, Livorno, Italy, 1998. 22. Van de Kamp, W. L., and Daimon, J. “Cemflame 2: Further studies on the Effect of Burner Design Variables and Fuel Properties on the Characteristics of Cement Kiln Flames.” International Flame Research Foundation Doc No. F/97/y/3, Livorno, Italy, 1997. 23. Smart, J. P., Jenkins, B. G., and van de Kamp, W. L. “Studies on NOx Emission from an Experiment Rotary Cement Kiln Fire with Mono and Multi-Channel burners.” ZKG International, N°8, Gütersloh, 1999.
Cement Kilns
24. Lowes, T. M., and Evans, L. P. “Optimization of the Design and Operation of Coal Flames in Cement Kilns.” Journal of the Institute of Energy 62, (1989): 220–28. 25. National Fire Protection Association. “NFPA 8503 Standard for Pulverized Fuel Systems.” Quincy, MA, 1997. 26. Lowes, T. M. “Rotary Kiln Operation Problems and Solutions.” TOTeM-33 of International Flame Research Foundation, February 11, 2009. 27. Sergeant G. D., and Smith, I. W. “Combustion Rate of Bituminous Coal Char in the Temperature Range 800 to 1700 K.” Fuel 52, no. 1 (1973): 52–57. 28. Cristea E. D., Malfa, E., and Coghe, A. ”3-D Simulations and Experiments of Cement Rotary Kiln Preheater Top Cyclone.” Proceedings of Fluent European User’s Group Meeting, Harrogate, UK, 1994. 29. Cristea E. D., Malfa, E., and Coghe, A. ”3D Simulations and Measurements of Strongly Swirling Heavy Laden Flow Inside Cyclone Dust Separator.” 3rd International Symposium on Engineering Turbulence Modeling and Measurements, May 27–28, 1996, Crete, Greece. 30. Cristea, E. D., Malfa, E., and Coghe, A. “3-D Numerical Computation and Validation of High Solid Loading Flow Inside a Gas Cyclone Separator.” ASME Fluids Engineering Division Summer Meeting, June 21–25, 1998, Washington, DC. 31. Meier H. F., Ropelato, K., Forster, H., Less, J. J., and Mori, M. «Computational Fluid Dynamics (CFD) for Cyclone Evaluation and Design, Part1 and 2.” ZKG International 55, no. 4 (2002). 32. Wall, T., Liu, G., Wu, H., and Benfell, K. “The Effect of Pressure on Char Characteristics, Burnout and Ash Formation in Entrained Flow Gasifiers.” International Flame Research Foundation Combustion Journal, May 2001. 33. Bartok, W., and Sarofim, A. F., eds. Fossil Fuel Combustion. New York: Wiley. 34. Lawn, C. J., ed. Principles of Combustion Engineering for Boilers, London: Academic Press, 1987. 35. Van Krevelen, D. W. Coal, Amsterdam: Elsevier, 1993.
© 2011 by Taylor and Francis Group, LLC
669
36. EU Research Contract Programme. Contract N° EV5VCT94-0551. NOx Emissions from Cement Manufacturer and Evaluation of Various Possibilities for NOx Reduction in the Cement Industry, September 15, 1994–September 15, 1997. 37. Maia, J. “Flexiflame™: A Major Step Forward in Burner Design.” ZKG International, no. 8, (2008): 44–53. 38. Sarkar, S. Fuels and Combustion. India: Orient Longman, 1974. 39. ASTM C150-07 Standard Specification for Portland Cement. www.astm.org 40. AASHTO M85 Standard Specification for Portland Cement. www.cement.org 41. EN 196 Method of Testing Cements. www.standardsdirect.org 42. EN 197 Portland Cement. www.iso.org 43. Versteeg, H. K., and Malalasekera, W. An Introduction to Computational Fluid Dynamics-The Finite Volume Method UK: Pearson Education Ltd., 1995. 44. Poinsot, T. M., and Veynante, D. Theoretical and Numerical Combustion. Philadelphia, PA: R.T. Edwards Inc., 2005. 45. Chigier, N. A., and Beèr, J. M. "Velocity and Static–Pressure Distribution in Swirling Air Jets Issuing from Annular and Divergent Nuzzles." Journal Basic Engineering, 4, 1964: 788–796. 46. Kutz, M. Mechanical Engineer’s Handbook, 3e, Book 4, Energy & Power. New York: John Wiley, 2006. 47. Guse, B. Multi-Burner Flexibility, Fives-Pillard, June 2006. 48. Greco Enfil Company. “Greco–Enfil Company.” Presentation to Italcementi, 2008. 49. Unitherm Cemcom. “M.A.S. Burner for Traditional and Alternative Fuels.” Prestentation to Italcementi Group, 2005. 50. F. L. Smidth Co. “Alternative Fuels.” Presentation to Italcementi Group. 51. FCT-Combustion Pty Ltd. Report "Study of High CO Emissions in the Preheater Tower" submitted to Italcementi Group, 2008. 52. Cinar Ltd. Final Report "Mathematical modeling of Gaurain Oreclciner" submitted to Italcementi Group, 2009.
32 Glass Furnaces R. Robert Hayes and Charles E. Baukal, Jr. Contents 32.1 Introduction.................................................................................................................................................................. 671 32.2 Previous Experimental Studies................................................................................................................................. 673 32.3 Instrumentation and Procedure................................................................................................................................ 675 32.3.1 Glass Furnace................................................................................................................................................. 675 32.3.2 Glass Surface Temperature Measurements.................................................................................................676 32.3.2.1 Instrument Description................................................................................................................676 32.3.2.2 Theory and Calibration.................................................................................................................676 32.3.2.3 Measurement Procedure............................................................................................................. 678 32.3.2.4 Data Acquisition and Reduction................................................................................................ 678 32.3.3 Incident Radiant Heat Flux Measurements............................................................................................... 678 32.3.3.1 Hemispherical Ellipsoidal Radiometer..................................................................................... 678 32.3.3.2 Circular Foil Heat Flux Gauge.................................................................................................... 679 32.4 Glass Surface Temperature......................................................................................................................................... 680 32.4.1 Physical Phenomena...................................................................................................................................... 680 32.4.2 Experimental Results.................................................................................................................................... 681 32.4.3 Time Resolved Experimental Results......................................................................................................... 682 32.5 Crown Incident Radiant Heat Flux........................................................................................................................... 683 32.5.1 Firing Measurements.................................................................................................................................... 683 32.5.2 Nonfiring Measurements............................................................................................................................. 685 32.5.3 Time-Resolved Experimental Results......................................................................................................... 686 32.6 Conclusions................................................................................................................................................................... 687 32.7 Nomenclature............................................................................................................................................................... 688 References................................................................................................................................................................................. 688
32.1 Introduction The glass industry consists primarily of four segments: container, flat, pressed and blown, and fiberglass [1]. Glass is formed from raw materials that are fed into a high temperature melting furnace to produce glass that may be clear or colored. The specific furnace and heat transfer issues vary slightly by segment. There are four common furnace designs used to make glass: direct-fired furnaces, regenerative side port, regenerative end port, and all electric furnaces [2]. The type of furnace for melting glass typically depends on the type and quantity of glass being produced, and the local fuel and utility costs. The term “unit melter” is generally given to any fuel fired glass melting furnace that has no heat recovery device. Typically the air/fuel fired unit melters are relatively small in size and
are fired with 2–16 burners. Furnaces range in production from as large as 40 tons (36 metric tons) of glass per day to as small as 500 lb (230 kg) of glass per day. Larger air/fuel unit melters are found in areas where fuel is extremely cheap. Frit, tableware, opthamalic glass, fiberglass, and specialty glasses with highly volatile and corrosive components are produced in unit melters. A recuperative melter is a unit melter equipped with a recuperator. Typically the recuperator is a metallic shell and tube style heat exchanger that preheats the combustion air to 1000–1400°F (540 to 760°C). The furnace is fired with 4–20 individual burners. These furnaces range in size from as large as 280 tons (250 metric tons) per day of glass to as small as 20 tons (18 metric tons) per day of glass. These furnaces are common in fiberglass production but can also be used to produce frit. Some recuperative furnaces are used in the container industry, though this is not common. Furnace life is a function of glass type being 671
© 2011 by Taylor and Francis Group, LLC
672
Industrial Combustion Testing
Primary air
Combustion air from blower Refiner Melter
Secondary air
Recuperator
Flue gas discharge
Batch charger Hot combustion air to burners
Checkers
Furnace Batch charger
Burners
Figure 32.1 Typical recuperative glass melter. (From Eleazer, P. B., and Hoke, B. C., Oxygen-Enhanced Combustion, Boca Raton, FL: CRC Press, 1998.)
produced. For example, a six year furnace life is typical for wool fiberglass. A typical recuperative melter is shown in Figure 32.1. In a regenerative or Siemans furnace, air for combustion is preheated by being passed over hot regenerator bricks, typically called checkers (sometimes spelled chequers). This heated air then enters an inlet port to the furnace. Using one or more burners, fuel is injected at the port opening, mixes with the preheated air and burns over the surface of the glass. Products of combustion exhaust out of the furnace through nonfiring ports and pass through a second set of checkers, thereby heating them. After a period of 15–30 minutes, a reversing valve changes the flow and the combustion air is passed over the hot checkers that were previously on the exhaust side of the process. The fuel injection system also reverses. After reversing, the exhaust gases pass through and heat the checkers that had previously heated the combustion air. The Siemans furnace is the workhorse of the glass industry. Most flat glass and container glass are produced in this furnace type. Regenerative furnaces are also used in the production of TV products, tableware, lighting products, and sodium silicates. There are two common variants of the Siemans furnace: the side-port regenerative melter, and the end-port regenerative melter. End-port regenerative furnaces are typically used for producing less than 250 tons (230 metric tons) of glass per day. In an end-port furnace, the ports are located on the furnace back wall. Batch is charged into the furnace near the back wall on one or both of the side walls. Figure 32.2 shows the layout of a typical end-port furnace. These furnaces are commonly used for producing container glass, but are also used for producing tableware and sodium silicates. For container production, a furnace campaign typically lasts eight years.
© 2011 by Taylor and Francis Group, LLC
Figure 32.2 Typical end-port-fired regenerative glass melting furnace. (From Eleazer, P. B., and Hoke, B. C., Oxygen-Enhanced Combustion, Boca Raton, FL: CRC Press, 1998.) Burner with atomizer Regenerators
Waste gas exit
Regenerators
Air intake Reversal mechanism
Chimney damper Figure 32.3 Typical side-port-fired regenerative glass melting furnace. (From Eleazer, P. B., and Hoke, B. C., Oxygen-Enhanced Combustion, Boca Raton, FL: CRC Press, 1998.)
Side-port regenerative furnaces have ports located on the furnace side walls. Batch is charged into the furnace from the back wall. Figure 32.3 shows the layout of a typical side-port furnace. Side-port regenerative furnaces are typically used for producing greater than 250 tons (230 metric tons) of glass per day. A side-port furnace for float glass commonly produces 500–700 tons (460–630 metric tons) of glass per day. For container glass, side-port furnaces ordinarily produce between 250 and 350 tons (230–320 metric tons) of glass per day. These furnaces are commonly used in container and float glass production, but are also used for the production of tableware and sodium silicates. The design of modern industrial combustion furnaces has become increasingly important from both environmental and economic standpoints [3]. The competing objectives of high efficiency and low emissions have created a complex challenge. A better understanding of the combustion and other transport processes
Glass Furnaces
and the ability to model their behavior more accurately are required to best meet these objectives. These challenges are certainly evident in the area of industrial glass furnaces. Modern design efforts increasingly rely on advanced three-dimensional numerical models of the combustion space and glass melt tank. Significant efforts in this type of modeling are demonstrated by the large number of publications in this field [4–37]. However, experimental data from industrial-scale glass furnaces are relatively scant [29,35–46]. In addition to the important role that these measurements play in increasing the understanding of the glass melting and combustion processes, there is an immediate application for experimental data in the development and evaluation of the sophisticated numerical models created to predict the combustion, batch melting, and melt circulation processes in these industrial furnaces. Important fundamental parameters for model validation and understanding of the combustion and glass melting processes include gas species composition, heat flux, gas temperature, gas velocity, wall temperature, and glass surface temperature. This chapter reports experimental studies in industrial-scale glass furnaces. A specific study will be used to illustrate some of the detailed measurements [44]. This study includes glass surface temperature and crown incident heat flux data in an industrial, gas-fired, flat-glass furnace. The data include time-resolved as well as timeaveraged data for these parameters acquired at six locations within the combustion space of a regenerative, side port, 550 ton/day, gas-fired, flat-glass furnace. The timeresolved information obtained enables the study of transient physical phenomena taking place in the combustion space during both the firing and nonfiring reversal cycles of the regenerative furnace. The data are obtained through six access holes in the crown of a regenerative, side-port, 550 ton/day gas-fired, flat-glass furnace owned and operated by Ford Motor Company Glass Division in Tulsa, Oklahoma [42].
32.2 Previous Experimental Studies There is only limited experimental data from industrial glass furnaces found in the open literature. Exper imental measurements in a large end-port, oil-fired, regenerative, horseshoe-shaped container-glass furnace with a nominal output of 90 tons/day of soda-lime glass and a specific energy consumption of 5.9 GJ/ton are reported by Cassiano, Heitor, and Silva [40]. Two operating conditions were tested to investigate the effect of furnace throughput and air–fuel ratio on the heat transfer characteristics of the furnace. Local gas temperatures, wall incident radiant heat fluxes, and both major
© 2011 by Taylor and Francis Group, LLC
673
and minor gas species concentrations were measured through six existing inspection ports in the furnace. Gas temperatures were measured with a double-shielded, water-cooled suction pyrometer (see chapter 5) and ranged from 1400°C to 1660°C. A total heat flux probe with a blackened receiver was used to measure radiation heat flux with measurements ranging from 230 to 340 kW/m2 (see chapter 6). A water-cooled sample probe was used to extract gas samples that were analyzed using a paramagnetic analyzer for O2 and nondispersive infrared analyzers for CO, CO2, and NOx (see chapter 7). Quantification of cyclic operation of the furnace as well as the three-dimensional nature of the combustion process was reported. No time-resolved information was reported and measurements through the crown of the furnace were not possible because of the lack of access ports in that region. Based on these measurements and modeling of the combustion process, together with an online furnace viewing system, the implementation of a new furnace control scheme is also reported by Victor et al. and Farmer, Heitor, and Vasconcelos [39,38]. Experimental measurements were reported for a similar end-port furnace firing heavy oil with a nominal output of 160 tons/day of amber glass and a specific energy consumption of 4.9 GJ/ton [41]. Local gas temperatures and average gas species concentrations (O2, CO, CO2, and NOx) were measured in both the flames and the combustion space. These measurements were taken through three ports located on the side of the furnace. Gas temperatures were measured with uncoated Pt/Pt13%Rh thermocouples of 300 μm wire diameter. Radiation heat losses from the thermocouple were corrected using a calculation based on diminishing diameter thermocouple measurements using 40, 80, and 300 μm (see chapter 5). The gases were sampled using a water-cooled stainless steel probe and were analyzed using a paramagnetic analyzer for O2 and nondispersive infrared analyzers for CO, CO2, and NOx (see chapter 7). The cyclic operation of the furnace and the three-dimensional nature of the flow and combustion were characterized. In-flame temperature measurements showed considerable uniformity. No radiative fluxes, velocities, or time dependence studies of the regenerative furnace were reported. Profiles of gas velocity, species concentrations (O2, CO, and CO2), average wall incident radiant heat flux, and gas temperature have recently been reported in the combustion space and exhaust ports of a regenerative, side-port, gas-fired, flat-glass furnace with a nominal output of 550 tons/day [29,42]. A straight water-cooled stainless steel pitot tube (see chapter 4) was designed to measure gas velocities because a conventional L-shaped probe could not be inserted through the available small access holes in the furnace. The probe OD was 22.2 mm, had a length of 3 m, and was calibrated in a wind tunnel against a standard pitot tube design. The gases
674
were sampled with a 22.2 mm OD water-cooled stainless steel probe and analyzed with multiple, duplicative gas analyzers using thermochemical and nondispersive infrared techniques. Gas temperatures were measured with a conventional triple-shielded, water-cooled suction pyrometer with a 63 mm OD and 4 m length (see chapter 5). A water-cooled ellipsoidal radiometer consisting of a stainless-steel, water-cooled jacket with a 43 mm OD and 0.65 m length was used to measure radiation heat flux (see chapter 6). A region of fast moving gases near the glass, with axial velocity components exceeding 20 m/s, and a large recirculation zone near the furnace crown have been reported based upon gas velocity measurements. Temperatures as high as 1985 K in the flame and as low as 1750 K in the recirculation zone were reported. A zone of intense reaction near the glass, with large concentration gradients, and incomplete combustion was also observed. The CO2 concentrations were the highest near the batch, where the glass reactions are most intense. Typical concentrations of O2, CO, and CO2 near the vicinity of the flames were 2%, 40,000 ppm, and 10%, respectively. Local average incident radiant fluxes measured at the crown were spatially uniform at 680 kW/m2 over five locations extending laterally across the furnace, along the centerline of the third port from the batch feeder. Significant variations were observed in the exhaust profiles of most measured variables. Large errors in exhaust mass balance suggest a complex, threedimensional flow with recirculation zones along the side walls of the portnecks. A nominal preheat air temperature of 1420 K and a variation in exhaust temperatures from port to port between 1630 K and 1835 K were reported. High O2 concentrations, reaching 8.4%, were measured in the exit, suggesting a bypass of oxygenrich flow around the flame. Limited time-dependent measurements at a few locations were reported for some variables, including species concentration, gas velocity, and gas temperature. These transient measurements were restricted by the time response of the instrumentation, and usually represent data points acquired at 2–3 minute intervals over a 15–20 minute period. One study examined the same furnace after it was rebuilt [45] and compared it to data collected before the rebuild [42]. The natural gas fuel composition and CO2 and H2O generated from the melt were essentially the same before and after the rebuild. Table 32.1 shows a comparison of some other key parameters before and after the rebuild. The same instrumentation (see above) was used for the measurements before and after the rebuild. Post-rebuild profiles of velocity, species concentrations (O2, CO, and CO2), and gas temperature data were reported and compared with the pre-rebuild data measured previously in the same furnace. Additionally, measurements were also taken below one of the regenerators in the tunnel leading to the furnace stack. Fewer
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
Table 32.1 Operating Conditions before and after Furnace Rebuild Parameter Fuel flow rate (kg/hr) Air flow rate (kg/hr) Excess air (based on plant measured values) Fuel flow reduction after rebuild
Before Rebuild
After Rebuild
3496 87,800 70%
2886 55,400 30%
N/A
17%
Source: From McQuay, M. Q., and Webb, B. W., Combustion Science and Technology, 150, no. 1, 77–97, 2000.
variations were observed in the exhaust profiles of most measured variables after the rebuild. Flat inlet velocity profiles were measured with an approximate magnitude of 11 m/s. Exhaust velocities after rebuild are reported to reach maximums of 21 m/s, compared with 26 m/s before rebuild. Exhaust gas temperatures were, generally speaking, higher in the post-rebuild furnace, while inlet preheated air temperatures were observed to be consistently lower after the rebuild. The locations of low O2 concentration in the effluent were shown to be consistent with the high CO concentrations existing both before and after the rebuild. The measurements in the tunnel resulted in very low CO levels, indicating that the combustion reactions continue inside the regenerators resulting in overall complete combustion. Abbassi and Khoshmanesh [37] numerically simulated a 200 ton/day gas-fired, regenerative, side-port melting furnace for making container glass. They also made some experimental measurements to compare against their numerical results. An infrared pyrometer was used to measure furnace wall temperatures at 18 locations during time intervals when there were no flames in the combustion space during firing reversals [47]. The maximum deviation between measured and computed wall temperatures was 126 K (relative error of 7.6%). In the previously mentioned studies, no measurements of the temperature of the glass surface in the combustion space were reported. The temperature of the glass surface inside the combustion space is a critical parameter for the understanding of the combustion and melt circulation processes, as well as validating numerical models. It is crucial to the understanding of the energy transfer to the glass, which is of primary interest in design considerations. Additionally, the glass surface temperature plays an important role in determining the quality of the glass being produced. Furthermore, of the previously reviewed studies, only two works report industrial-scale incident radiant heat flux measurements in the combustion space. The data reported in those instances consisted of average wall radiant heat flux within the combustion space, with one report containing data taken at the furnace wall and the other study presenting limited measurements
675
Glass Furnaces
performed at the crown along one port location. No spatially resolved data along the axial length of the crown were found. No time-resolved heat flux data have been reported, and no measurements were reported during reversal when the flame is off. The incident radiant heat flux within the combustion space is a critical parameter for understanding the combustion and glass melting processes. Quantifying local heat flux values is also essential in validating numerical models. It is crucial to understanding the energy transfer to the glass and refractory, which are of primary interest in design considerations. The levels of incident heat flux upon the surfaces within the combustion space (crown, batch, glass, and walls) play an important role in determining the quality of the glass being produced as well as the life of the refractory. Investigation of the variation in heat flux due to location, as well as differences due to firing and nonfiring periods is fundamental to the understanding of the furnace operation and critical for model validation and future design improvements.
32.3 Instrumentation and Procedure The example study highlighted here [43,44,46] included obtaining glass surface temperature and crown incident radiant heat flux measurements in an operating industrial, regenerative, gas-fired, flat-glass furnace. These data were acquired using three different instruments. Glass temperature measurements were obtained optically using a two-color pyrometer (see chapter 5). Incident radiative heat flux measurements were acquired using both a circular foil heat flux gauge, and an ellipsoidal hemispherical radiometer (see chapter 6). The following discussion describes the furnace in which the data
21.5
were acquired and the instrumentation used, including instrument description, theory and calibration, measurement procedure, data acquisition, and data reduction. 32.3.1 Glass Furnace The data were obtained in a regenerative, side-port, 550 ton/day, gas-fired, flat-glass furnace operated by Ford in Tulsa, Oklahoma. A schematic of the glass furnace can be seen in Figure 32.4. The furnace is approximately 2.5 m high from the glass line to the maximum height of the crown, 11 m wide, and 21.5 m long. A regenerator is located on the north and south sides of the furnace. Connecting each regenerator and the furnace is a set of six portnecks approximately 2.9 m long. At the regenerator opening each portneck is nominally 1.5 m wide and 1 m high. At the furnace inlet the height of each portneck narrows to 0.5 m and the width expands to 1.8 m. Each port has two 4.4-cm-diameter pipe burners located 81.3 cm from the edge of the furnace and 43.2 cm above the glass line. Each burner is angled so that the centerline of the flow will intersect the portneck centerline at the edge of the furnace. The furnace operates on a 15-minute cycle: a 15-minute burn from the north portnecks with exhaust through the south (identified here as N–S configuration or firing direction), a reversal period (approximately 20 seconds) during which no firing occurs, followed by 15 minutes in the opposite direction (i.e., S–N configuration or firing direction). The furnace was constructed with six thermocouple plugs used to measure crown temperature. These plugs, when removed, provided access holes located along the center of the crown apex at axial furnace locations coinciding with the transverse centerlines of each of the six ports (see Figure 32.4). The total combustion fuel flow rates, airflow rates, and firing stoichiometry for each port during furnace operation are shown in Table 32.2. The fuel flow rates
1 2 3 Crown access holes n Crow
m
2.6 m Molten
South
Natural gas burners
Port access holes
11 m (inside) North
tor
nera
glass
Rege
n
Rege
es
s hol
cces
ra erato
2.9 m
Figure 32.4 Schematic (not drawn to scale) of the side-port, regenerative, 550-ton/day, gas-fired, flat-glass furnace where measurements were made. (From Hayes, R. R., Brewster, S., Webb, B. W., McQuay, M. Q., and Huber, A. M., Experimental Thermal & Fluid Science, 24, 35–46, 2001.)
© 2011 by Taylor and Francis Group, LLC
676
Industrial Combustion Testing
Table 32.2 Furnace Firing Conditions in Ports 1 through 6
Port 1 Port 2 Port 3 Port 4 Port 5 Port 6 a b
c
Fuel Flow Ratea (kg/hr)
Air Flow Rateb (kg/hr)
Excess Air c (%)
583 667 666 651 679 243
9680 11,110 10,680 10,240 14,200 6380
12 13 8 6 41 77
43 mm diameter Detector housing 1m Detector detail
Fuel flow rate for each port as reported by the plant. Air flow rates estimated using previously measured exhaust species compositions (From Abbassi, A., and Khoshmanesh, Kh., “Numerical Simulation and Experimental Analysis of an Industrial Melting Furnace,” Applied Thermal Engineering 28, 2008). Values calculated using total measured fuel flow rates and estimated air flow rates.
presented were measured by the plant during operation when the measurements presented herein were performed. The inlet air flow rates shown were not measured, but were calculated for each port. Using the measured fuel flow rates and previously measured exhaust values of O2 and CO2 in each port, the combustion reaction equation was balanced for each port to solve for the inlet air flow rates [42]. As noted in this table, the majority of the fuel is distributed in ports 1 through 5, with a smaller fraction used in the port nearest to the working end of the furnace (port 6). 32.3.2 Glass Surface Temperature Measurements 32.3.2.1 Instrument Description Glass surface temperatures in the combustion space of the glass-melting furnace were measured optically using a two-color pyrometer. Figure 32.5 is a schematic of the instrument. The optical detector housing is mounted on a water-cooled sight tube. The sight tube is approximately 2.8 cm in diameter (I.D.) and 1 m long. Lenses, located in the optical detector housing, image the infrared radiation entering the sight tube onto amplified germanium photodiodes. The radiation is transmitted through interference filters with 30 nm bandwidths centered at 1.27 µm and 1.60 µm. These wavelengths were chosen to avoid the absorption bands of CO2 and water vapor [48]. The signal amplification system for each channel was also constructed with adjustable gain, to permit modification for a range of high temperatures and a variety of data acquisition equipment. 32.3.2.2 Theory and Calibration Through quantum theory, Max Planck described the spectral energy distribution of a blackbody in a vacuum as a function of the absolute temperature
© 2011 by Taylor and Francis Group, LLC
Water Water outlet inlet
Water-cooled sight tube
Photodiode detectors
Photodiode detectors Bandpass optical filters
Figure 32.5 Schematic of the two-color pyrometer instrument used for optical glass surface temperature measurements. (From Hayes, R. R., “Glass Surface Temperature and Incident Heat Flux Measurements in an Industrial, Regenerative, Natural Gas-Fired Flat-Glass Furnace,” MS thesis, Brigham Young University, Department of Mechanical Engineering, 1999.)
of the blackbody and the wavelength of the emitted energy [49]. The corresponding relation is known as Planck’s Law: τ mλ ε λC1 (32.1) Eλ = . ( C /λT ) 5 − 1} λ {e 2 Wien presented a simplified model to describe this phenomenon for low λT products [50]. This relation is known as Wien’s Formula (Equation 32.2) and is accurate to within approximately 1% over the λT range covered by the two-color pyrometer used:
Eλ =
τ mλ ε λC1 . λ 5e(C2 /λT )
(32.2)
Using this formula, assuming that the emissivity of the surface of the emitting body is not spectrally dependent (gray) over the spectral range of the photodiode/filter assembly, and knowing that the voltage potential generated by each detector is proportional to the energy incident on the detector, the following expression (Equation 32.3) for the temperature of the emitting body can be formulated:
Tbody
1 1 C2 − λ 2 λ 1 , = Eλ1 λ2 + 5 ln ln λ1 Eλ 2
where C2 = 1.44 × 107 nm × K λ2 = 1.27 μm λ1 = 1.60 μm.
(32.3)
677
Glass Furnaces
This function relates the absolute temperature of the emitting body to the ratio of the two wavelengths chosen, and the ratio of the emissive power of the radiant energy at those wavelengths. The photodiode detectors are each designed to output a voltage that is directly proportional to the emissive power of the radiant energy incident on that detector. The emissive power at each of the wavelengths can thus be expressed as the product of the voltage output for that wavelength and some proportionality constant P. Eλ1 = P1Vλ1
Eλ 2 = P2Vλ 2 .
The relationship between the emitting body temperature and the voltage output from the detectors at each wavelength can thus be established using an experimentally determined calibration constant, a, which relates the voltages measured at each channel with the spectral emissive power at the given wavelength (Equation 32.4):
Tbody
1 1 C2 − λ2 λ1 = , Vλ1 λ2 + 5 ln + a ln λ1 Vλ 2
(32.4)
where P a = ln 1 . P2 The parameter a is determined from calibration of the instrument that was performed by the detector manufacturer and subsequently checked in the laboratory. The two-color pyrometer was calibrated both with and without the water-cooled sight tube. The detector assembly (without sight tube) was calibrated by the detector manufacturer between 900 K and 2000 K using a small reference blackbody. Later, the detector system with a water-cooled sight tube attached was calibrated in the laboratory using a blackbody cavity designed and built for this purpose. The blackbody cavity was heated using a small electric laboratory furnace (e.g., see Figure 6.20). However, due to limitations of the laboratory furnace, the sight-tube calibration permitted corroboration of the previous detector calibration only to a maximum temperature of 1400 K. To extend the calibration temperature range, the sight tube attached to a torchheated ceramic plate heat source was constructed. Aluminum oxide ceramic material (Coors AD 96; 96 Al 2O3) was selected for its radiatively gray behavior over the range of the two wavelengths used by the pyrometer. A Type S (Pt-Pt10%Rh) thermocouple was
© 2011 by Taylor and Francis Group, LLC
inserted into a blind hole drilled into each side of the 6 mm thick ceramic plate, and an acetylene torch was used to heat the rear face while the optical pyrometer imaged the front face. The ceramic plate was heated to approximately 2000 K. An analysis was developed to estimate the front face temperature from the measured internal temperature. The sight-tube calibration using the blackbody cavity (up to 1400 K) and the ceramic plate heat source (up to 2000 K) enabled the detector assembly to be properly aligned with the sight tube, and confirmed the detector-only calibration performed previously to a maximum temperature of 2000 K. Exploratory measurements with the two-color pyrometer revealed the response of the photodiode/ amplifier components to be somewhat temperaturedependent. Laboratory experiments were performed to quantify this dependence. With a given blackbody source temperature, the two-color pyrometer detector assembly temperature was varied using a specially fabricated water jacket through which water of known temperature was circulated. The water temperature was varied over the anticipated range of the experimental environment. Additional experiments were also performed using the torch-heated ceramic plate to achieve controlled higher source temperatures while varying the two-color detector assembly temperature using a vortex tube cooler with an adjustable air flow rate. In both studies the detector assembly temperature was measured using a Type K thermocouple. The response of the photodiodes at both wavelengths (1.27 and 1.60 µm) was measured over the range of controlled detector temperatures. A mathematical relation was formulated that accurately represents the temperature dependence of the amplifier sensitivity. Interestingly, only the detector centered at 1.60 µm was found to have temperature-dependent response. This temperature dependence is quantified by the following function (Equation 32.5), which enables use of the relation of the difference between a detector body at a standard temperature of 295 K and the actual temperature of the detector body, Tdetector, designated as parameter b shown in Equation 32.6. f (Tdetector ) = 1.2244 − 0.0085361(Tdetector )
+ 7.339116E − 5 (Tdetector )2 ,
b=
f (Tdetector( K ) ) . f ( 295 K )
(32.5)
(32.6)
This correction parameter allows for the expression of the final relationship (Equation 32.7), which indicates
678
Industrial Combustion Testing
the temperature of the emitting body given the voltage from each channel (wavelength) and the detector body temperature (ambient):
Tbody
1 1 C2 − λ 2 λ 1 = . Vλ1 b λ ln + 5 ln 2 + a λ1 Vλ 2
(32.7)
During testing in the glass furnace environment, the detector assembly was cooled using a vortex tube cooler, with an adjustable air flow rate, in order to minimize the temperature excursions. Additionally, the temperature of the detector assembly was measured and documented for later use in data reduction. A statistical analysis of all calibration data including the temperature dependence of the detectors reveals an estimated uncertainty in measured temperature of ±2% of the total measured absolute temperature. 32.3.2.3 Measurement Procedure Surface temperature measurements with the two-color pyrometer were taken in each of the six crown access holes, with location 1 being nearest to the batch feeder and continuing through location 6, nearest to the working end of the furnace. Data were acquired during the reversal of the 15-minute regenerator cycle, when all of the burners are off for approximately 20 seconds. This enabled the unobstructed (no flame) measurement of the molten glass surface below. The measurement procedure included establishing adequate cooling water flow through the probe water jacket and appropriate air flow through the detector assembly vortex tube cooler prior to insertion of the probe. The probe was then inserted through each crown access hole, with the sight tube viewing the glass surface directly below. The estimated optical view of the two-color detector at the glass surface was a circular area approximately 9 cm in diameter. Detailed time-resolved measurements were obtained for all crown access holes, with the exception of location 2, where only time-averaged surface temperature data were taken.
up to 5000 Hz for temperature measurements, enabling a thorough time-resolved investigation of the glass surface temperature variation during the short 20-second reversal period. Insulated, high-temperature, shielded cables were used to transmit the signal directly from the two-color optical assembly on the crown to the data acquisition system located safely on the main floor near the regenerator below. The temperature of the detector assembly, during the acquisition of optical surface temperature data, was also measured and recorded using a type K thermocouple placed inside the detector assembly housing. Data reduction was performed through utilization of the final calibration expression, Equation 32.7, stated previously, including the two experimentally determined calibration constants, a and b, to determine the measured surface temperatures. 32.3.3 Incident Radiant Heat Flux Measurements Incident radiant heat flux measurements were performed using two instruments: a hemispherical ellipsoidal radiometer and a circular foil heat flux gauge radiometer [51]. The former probe having a longer time response (30–60 seconds) and the latter probe capable of higher frequency (up to 250 Hz) measurements. 32.3.3.1 Hemispherical Ellipsoidal Radiometer 32.3.3.1.1 Instrument Description A schematic of the ellipsoidal radiometer is shown in Figure 32.6 The instrument consisted of a stainless-steel, water-cooled jacket, 43 mm in diameter and 0.65 m long, Water jacket 1.7 in (43 mm) Sensing head
For the glass temperature measurements reported, the amplified output from the detector was between 1 and 12 volts. Time-averaged readings were obtained using a voltmeter, while both time-averaged and time-resolved measurements were acquired and recorded using a computer-based, data-acquisition system, with a 12-bit analogto-digital converter. This system provided a sampling rate
© 2011 by Taylor and Francis Group, LLC
Water outlet
Water inlet
Sensing head detail Sensing Nitrogen purge pellet Ellipsoid
32.3.2.4 Data Acquisition and Reduction
2 ft (0.65 mm)
Removable element
Cooling water Constantan wires
Figure 32.6 Schematic of the ellipsoidal hemispherical radiometer used for crown incident radiant heat flux measurements. (From Hayes, R. R., Brewster, S., Webb, B. W., McQuay, M. Q., and Huber, A. M., Experimental Thermal & Fluid Science, 24, 35–46, 2001.)
679
Glass Furnaces
encasing an inner signal wire tube with the detector head mounted in one end, and cooling water and signal lead connections at the opposite end. The radiometer is also equipped with a purge gas tube that allows for a flow of dry nitrogen to exit the ellipsoidal cavity at the detector face to prevent fouling of the ellipsoidal reflective surface by particulate or condensation during use. The hollow ellipsoidal cavity at the sensor face has a 3 mm diameter aperture at one focus and a hemispherical sensing head at the other. The aperture and the sensor head are both centered at the respective foci of the ellipsoidal cavity. 32.3.3.1.2 Theory and Calibration The instrument is designed to utilize the ellipsoidal cavity to focus all incident hemispherical radiant energy onto the sensor surface. The ellipsoidal cavity has an eccentricity of 0.5 and is coated with a 0.05 mm layer of polished gold. The manufacturer states that the reflectivity of the inner ellipsoidal surface is very near unity and is not dependent on the angle of incidence or wavelength of the incident energy. All radiant energy passing through the aperture is reflected onto the detector surface after a maximum of two reflections. The manufacturer suggests a purge system using dry nitrogen to prevent fouling of the reflective inner ellipsoidal surface or detector surface by condensation or particulate. A purge nitrogen flow rate of 45 l/hr is recommended. The ellipsoidal detector head is designed to measure the total (spectrally integrated) hemispherical radiant heat flux incident at the detector face. The instrument converts the heat flux received over a solid angle of 2π steradians into a millivolt potential across a differential thermocouple. The instrument was calibrated accurate to ±5% up to approximately 300 kW/m2 by the manufacturer; above this range the manufacturer indicated that the linear calibration curve could be extrapolated to calculate incident radiative flux. However, this extrapolated value may be somewhat lower than the actual value. The ellipsoidal radiometer requires a nominal response time of 2 to 3 minutes to come up to temperature, after insertion into the furnace environment. After it has reached this level, the detector has a much faster response to changing heat flux, nominally 30–60 seconds. This instrument was therefore most useful for collecting time-averaged data, and only capable of measuring time-resolved trends over longer intervals—lasting several minutes or more. 32.3.3.1.3 Measurement Procedure Time-averaged measurements were acquired through crown access holes at locations 1 through 5. Water cooling flow and purge gas flow were established prior to positioning the instrument detector face inside the combustion space. The probe was inserted through each of the crown access holes, with all measurements acquired with the instrument oriented normal to the glass surface.
© 2011 by Taylor and Francis Group, LLC
The voltage output was measured using a handheld voltmeter. The measurement was then recorded once the output had reached a steady state. The ellipsoidal cavity was inspected between each measurement to ensure that the ellipsoidal mirror was not fouled and determine if the purge gas flow was adequate. The time-averaged data acquired with the ellipsoidal radiometer were only obtained during firing cycles because the radiometer response time exceeds the duration of the nonfiring reversal periods. Time-resolved data were also obtained using the ellipsoidal radiometer over the entirety of several 15-minute firing cycles at location 4, with output recorded every 60 seconds, to examine lower frequency trends over complete firing cycles. 32.3.3.1.4 Data Acquisition and Reduction Radiometer voltage data were recorded from the handheld voltmeter readings after sufficient time had elapsed to ensure a steady-state measurement. The manufacturer of the radiometer supplied calibration data indicating the linear relation between heat flux and the voltage potential from the instrument. The recorded data was thus used with a calibration equation to mathematically determine the incident radiant heat flux. 32.3.3.2 Circular Foil Heat Flux Gauge 32.3.3.2.1 Instrument Description Time-resolved and time-averaged incident heat flux data were obtained with a water-cooled, circular foil, heat flux gauge radiometer. A schematic of the instrument is shown in Figure 32.7. This instrument consisted of a Detector head (45 mm diameter)
Water-cooled brass jacket
19 mm diameter
1.7 m
Water inlet
Water outlet
Detector head – face view
High-emissivity coated circular foil (0.5 mm diameter)
Figure 32.7 Schematic of the circular foil radiometer used for crown incident radiant heat flux measurements. (From Hayes, R. R., Brewster, S., Webb, B. W., McQuay, M. Q., and Huber, A. M., Experimental Thermal & Fluid Science, 24, 35–46, 2001.)
680
brass, water-cooled jacket, 19 mm in diameter and 1.7 m long, encasing an inner signal wire tube with a detector head mounted at one end, and cooling water and signal lead connections at the opposite end. The detector head was approximately 45 mm in diameter with a 0.5 mm diameter circular foil heat flux sensor located at the center of the detector face. 32.3.3.2.2 Theory and Calibration The instrument is designed to receive radiant energy over a full 180° view and measure the total incident hemispherical radiant heat flux up to 850 kW/m2. The sensor is coated with zynolate that results in a sensor emissivity of 0.94. The instrument was calibrated by the manufacturer up to 240 kW/m2, and found to output a voltage that is linear in relation to the total incident hemispherical heat flux, with a relationship of 229.8 W/cm2 per mV output. The transducer accuracy is stated by the manufacturer to be within ±2%. The manufacturer indicated that this calibration line could be extrapolated up to 850 kW/m2 for the instrument with the specified accuracy. The instrument was designed to optimize time-response and view angle for the hemispherical time-resolved measurement application. The circular foil heat flux gauge was tested in the laboratory for time response by subjecting the instrument to a step function of a known incident radiant heat flux and recording the converted response digitally using a high frequency pc-based data acquisition system. This provided opportunity for analysis of the instrument response and calculation of the time-constant, τ, defined as the time required for the signal to reach 63.2% of the step function that was input. The circular foil gauge was found to have a time constant of approximately 4 ms, establishing the capabilities and limitations of the instrument for use in acquiring time-resolved data. 32.3.3.2.3 Measurement Procedure Data from the circular foil heat flux gauge were acquired in each of the six crown access holes, with location 1 being nearest to the batch feed and continuing through location 6, nearest to the working end of the furnace. Measurements were performed using the foil heat flux probe in a manner similar to that previously described for the ellipsoidal radiometer. Adequate cooling water flow was obtained through the use of a booster pump. Time-resolved and time-averaged radiant heat flux data were acquired using the foil gauge during both firing and reversal cycles at each location. The detector face was inspected between measurements and cleaned when necessary to ensure performance reliability and measurement integrity. 32.3.3.2.4 Data Acquisition and Reduction For the data acquired with the foil heat flux gauge, the amplified output from the detector was between 0 and
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
2 volts. These measurements were acquired, digitized, and recorded using a pc-based data-acquisition system, with a 12-bit analog-to-digital converter, capable of obtaining higher frequency signals (above 250 Hz). Insulated, high-temperature, shielded cable was used to transmit the signal directly from the foil heat flux gauge on the crown, to the pc-based data acquisition system located safely on the main floor near the regenerator below. This system provided time-resolved acquisition of heat flux measurements during both firing and nonfiring cycles. Data acquired using the circular foil heat flux gauge was reduced according to the previously stated linear calibration relation provided by the manufacturer, and incident radiant heat flux was calculated directly from the recorded foil gauge voltage output.
32.4 Glass Surface Temperature This section presents and discusses the glass surface temperature experimental results and numerical predictions. Initially, some pertinent complex physical phenomena are briefly discussed, as revealed through independently acquired video data and furnace operating information, to enhance and support the discussion of the experimental results. Next, a three-dimensional numerical model of the batch melting and glass tank, developed by coworkers in conjunction with the present work, is discussed to provide some background on the basis for the predictions to be presented thereafter and to enhance the overall understanding of the physical phenomena involved. The measured time-averaged glass surface temperature data in all six of the crown access holes along the lengthwise centerline of the furnace are then reported and discussed, together with the results of the model of the glass tank and batch melting processes. Lastly, time-resolved glass surface temperature data are presented and discussed. 32.4.1 Physical Phenomena Figure 32.8 shows a schematic (plan view) of the glass surface from the furnace batch feeder wall to the end of port 6 with respect to the general characteristics of the furnace, such as the location of the six portnecks and the batch feeder. This schematic was constructed from video observations made through the crown access holes, and visual observations made through other furnace inspection ports located in the regenerators (along the centerline of the portnecks), and in the working end of the furnace looking toward the batch feeder. The video data obtained by Ford personnel, delineating the brighter
681
Glass Furnaces
Plant estimated location of spring zone
4.27 m
Working end
Port 6
Port 4
Port 3
Port 2
Port 1
Batch blanket
Port 5
Location of bubblers
Batch feeder
3.35 m 3.28 m 3.28 m 3.35 m 3.05 m South side
Figure 32.8 Schematic (plan view) of glass surface in the furnace based on information obtained through specialized video data and visual observations through furnace inspection ports, indicating bubbler location and estimated spring zone position. (From Hayes, R. R., Brewster, S., Webb, B. W., McQuay, M. Q., and Huber, A. M., Experimental Thermal & Fluid Science, 24, 35–46, 2001.)
areas of molten glass and the much darker areas of batch blanket, were acquired using a high-resolution video camera with specialized filtered lenses attached to a water-cooled sight tube inserted through the crown access holes. The video probe apparatus allowed for a field of view of approximately 30°, which provided clear information in the central region of the tank (2–2.5 m wide at the glass surface), but limited observations near the tank edges. Figure 32.8 also indicates additional information about characteristics of the glass tank: (1) the location of bubblers, which are used to enhance mixing of the molten glass, and (2) the approximate location of the “spring zone,” where recirculating molten glass rises from lower regions of the tank to the surface causing the glass surface temperature to reach maximum values in this region of the tank. As illustrated in Figure 32.8, the batch blanket profile shows that the surface boundary between the batch blanket and molten glass along the furnace axial centerline was located in the vicinity of location 1, approximately 4.3 m from the end wall closest to the batch feeder. The schematic also indicates that batch material was nonexistent along the furnace centerline past location 1, but extended along the sides of the furnace in the approximate profile shown, tapering in thickness along the furnace side walls until complete batch meltout, which typically occurs in the vicinity of locations 3 or 4. As expected, the surface boundary between batch and molten glass is not completely steady. The location of the batch blanket varies slightly during furnace
© 2011 by Taylor and Francis Group, LLC
operation with several factors influencing its movement. The batch material is introduced intermittently to the furnace because of the operating characteristics of the batch feeder, causing the batch material to come off the feeder and enter the furnace in discrete clumps or “logs.” In addition, the molten glass recirculation pattern and batch melting contribute to batch movement at the surface near the batch blanket/glass interface. Lateral movement of the batch blanket was also visually observed, with the unmelted batch blanket tending to shift cyclically with firing direction. This cyclic surface motion was observed to be toward the exit port side during firing, returning slightly more to the center during reversal, then shifting to the opposite port side as firing begins in that direction. These factors were identified as contributors to batch blanket movement during furnace operation and to the intermittent existence of “batch islands,” discrete smaller sections of batch material found near the boundary between the batch blanket and molten glass. The transient nature of the surface boundary between batch blanket and molten glass was clearly established, although quantitative details of the contributing phenomena and specifics of their observed transient effects were not thoroughly studied. 32.4.2 Experimental Results Glass surface temperatures measured at locations 1 through 6, averaged over the 20-second reversal period, are shown in Figure 32.9. The lowest average surface temperature measured was 1711 K at location 1, nearest the batch feeder. The average temperature then rises sharply from 1711 K to 1853 K between locations 1 and 2, continues to gradually rise to the peak measured temperatures 1950 Glass surface temperature (K)
North side
1900 1850 1800 1750 1700
0
5
10 15 Axial position (m)
20
25
Figure 32.9 Measured average glass surface temperature profile along the furnace axial centerline in the glass tank. (From Hayes, R. R., Wang, J., McQuay, M. Q., Webb, B. W., and Huber, A. M., Glass Science & Technology, 72, no. 12, 367–77, 1999.)
682
© 2011 by Taylor and Francis Group, LLC
the glass tank, the minimum heat loss from the glass to batch or end walls, and the highest firing rates and lowest excess air values in the flames of ports 4 and 5 (see Table 32.2). Repeated measurements obtained at these locations showed that the average surface temperature magnitudes measured during different reversal periods varied less than 76K at location 4, which is closest to the operating bubblers, and less than 20K at location 5. At location 6 the average measured glass surface temperature is much lower than temperatures measured at locations 4 and 5. Although there is essentially no heat loss to the batch at location 6, there is additional heat loss to the cooler working end of the furnace where the glass exits. Location 6 also experiences considerably less heat transfer from the combustion products than do the other measured locations, due to the reduced firing in this port (see Table 32.2), with significantly higher excess air (77%). Thus the molten glass surface in this region experiences additional heat loss and corresponding surface temperature drop due to convection that results from the unique firing conditions existing in port 6. These factors combine to maintain the lower surface temperature at this final measured location. 32.4.3 Time Resolved Experimental Results Time-resolved glass surface temperature data measured during approximately 20 seconds of the reversal period when the flame is off are shown in Figure 32.10. Typical measurement traces are shown in the figure. The surface temperature at location 1 is seen to vary between 1440 K and 1850 K during one 20-second reversal 2000
Glass surface temperature (K)
of 1912 K at location 4 and 1910 K at location 5, and then falls to approximately 1804 K at location 6, nearest to the furnace working end. The molten glass between the batch feeder and location 1 is shielded from radiant combustion heat by the blanket of unmelted batch on top of the molten glass located in that region of the tank. This, combined with heat loss due to the proximity of the cooler batch and furnace walls, contributes to the lowering of the molten glass temperature in the region. The cooler molten glass is denser and falls toward the bottom of the tank nearest the batch-side tank wall, creating this three-dimensional, recirculation zone. The higher-density glass near the bottom of the tank travels toward the center of the glass tank, rising to the surface at the location of the spring zone. The springzone region is characterized by high combustion firing rates, with minimal heat absorption by batch material. It is in this general area that the highest glass surface temperatures were measured. At the glass surface upstream of the spring zone, glass flows back toward the batch feed area. Downstream of the spring zone, molten glass flows toward the furnace working end. The recirculation and glass flow patterns are important in understanding the analysis of the glass surface temperature data, as well as the overall melting process in the furnace. The low average glass surface temperature measured at location 1 is caused by the fact that this location is at the boundary of the cooler batch and the molten glass in a region where heat loss from the glass to the batch as well as to the tank walls is large. As will be discussed later in conjunction with the time-resolved surface temperature data, both molten glass and batch material were viewed and measured intermittently at location 1. As one moves away from this region where batch and molten glass coexist, a sharp increase in the average measured glass surface temperature is observed along the furnace centerline between locations 1 and 2. In addition to the higher radiant heat flux received from the flame, location 2, being approximately 3.35 meters farther downstream of location 1, experiences less heat loss to the end wall and batch material. As one continues to approach the spring zone, the measured average glass temperature continues to increase gradually between locations 2 and 3. The higher measured surface temperature at location 3 is due to a combination of factors, including: (1) the recirculation flow pattern that feeds hotter molten glass from the spring zone into that region, (2) the increase in firing rate and reduction of the overall excess air in the flame, and (3) lower heat loss to the furnace end wall and to the ever-decreasing batch material along the tank edges. As expected, the highest average glass surface temperatures measured were at locations 4 and 5 in the region of the spring zone. These peak temperatures were due to the combined effects of the recirculation pattern in
Industrial Combustion Testing
1900 1800 1700 1600 1500 1400
0
5
10 Time (s)
15
20
Figure 32.10 Typical time-resolved surface temperature plots acquired during reversal at locations 1, 3, 4, 5, and 6. (From Hayes, R. R., Wang, J., McQuay, M. Q., Webb, B. W., and Huber, A. M., Glass Science & Technology, 72, no. 12, 367–77, 1999.)
683
Glass Furnaces
period. The measured temperature begins relatively low (approximately 1600 K), then rises slowly to a high plateau near 1820 K. It then drops drastically over a period of about 8.5 seconds to below 1500 K before rising slowly again to another high plateau of 1840 K for the remainder of the reversal period. This low frequency, large temperature variation is consistent with what would be expected if portions of the cooler batch blanket traverse the optical view area of the two-color pyrometer for a short time as they shift position in the glass tank in the region of location 1. This batch motion arises from the phenomenon explained previously in the discussion of Figure 32.8. The higher temperature plateaus observed both before and after the lowest temperature interval are both relatively steady and consistent in magnitude with one another, suggesting the measurement of the molten glass surface temperature during these periods. The lower temperature intervals of the data trace are consistent with the measurement of the surface of cooler unmelted or partially melted batch that exists as a continuous blanket along the centerline of the tank from the batch feeder to location 1. It is not known if the intermittently measured batch surface is due to optically viewing the movement of the edge of the continuous batch blanket or moving batch islands that are pulled by molten glass currents. It is known, however, that the lower measured surface temperatures at location 1 are consistent with batch surface temperatures predicted by the model, and the visual observations and physical phenomena discussed previously. The time-resolved temperature signal profile measured in hole 1 is the only one to vary so widely during the reversal period. All other measured locations exhibit a more nearly constant surface temperature during reversal, because the surface temperature measurements at these locations are of molten glass only, with batch material being nonexistent along the axial centerline of the furnace downstream of location 1. The time-resolved surface temperature measurements obtained at access locations 3 through 6 indicate significantly less variation in molten glass surface temperature than at location 1. Locations 3, 4, and 5 all have similar profiles, exhibiting a steady gradual surface temperature decline over the duration of the reversal period while the flame is not present. The data at these locations, shown in Figure 32.10, thus illustrate the cooling of the molten glass surface during the brief nonfiring period. Nominally, the surface temperature is seen to drop 50K–80K over the no-flame reversal period. As would be expected, the slightly steeper decline in temperature takes place when surface temperatures are the highest, such as near the beginning of reversal at locations 4 and 5, while the decline is more gradual when surface temperatures are lower. The time-resolved temperature measurements of the glass surface at location 6, nearest to the glass tank exit,
© 2011 by Taylor and Francis Group, LLC
indicate a steady, nearly constant temperature profile that does not exhibit the same constant gradual decline as the other locations (3–5). It is noted again that location 6 is unique in the combustion cycle process, as reduced air/fuel flow occurs at this port location. The constancy of the measured surface temperature over time during reversal demonstrates that, at this location, the surface temperature of the molten glass is in a state of relative equilibrium, exhibiting no significant cooling during reversal. This is reasonable, as the transient effects of firing and reversal periods are minimized at this port location and the lower surface temperature in this region reduces net radiant heat loss.
32.5 Crown Incident Radiant Heat Flux This section presents and discusses the crown incident radiant heat flux experimental results and numerical predictions. Initially, a three-dimensional numerical model of the combustion also developed by coworkers in conjunction with the present work, is briefly presented, discussing the method and basis for the predictions obtained. Average total incident radiant heat flux data obtained during firing periods, using each instrument, at locations along the lengthwise centerline of the furnace are then presented, and are compared with predictions from the numerical model. Measured radiant heat flux data acquired during nonfiring periods at all six crown locations are also presented and discussed. Comparison between firing and nonfiring measurements is performed to quantify combustion effects on crown incident heat flux. Lastly, time-resolved data, showing heat flux firing cycle variation for both firing directions are presented and discussed. Analysis of the experimental data is supported, where appropriate, with discussion of additional applicable information and pertinent observations, including the glass surface temperature measurements and physical phenomena that have previously been presented and discussed. 32.5.1 Firing Measurements Figure 32.11 shows measured average incident radiant flux to the crown along the axial centerline of the furnace during firing cycles. Time-averaged experimental data measured with both the circular foil and ellipsoidal radiometers are shown in the figure. There is good agreement between the experimental measurements obtained with the two different probes, with differences of less than 12% at all locations. The largest differences were in the regions of highest heat
684
Industrial Combustion Testing
750
Radiant heat flux (kW/m2)
700 650 600 550 Ellipsoidal radiometer Circular foil radiometer
500 450 400
0
5
10 15 Axial position (m)
20
25
Figure 32.11 Measured mean crown incident radiant heat flux along the axial centerline of the furnace during firing cycles. (From Hayes, R. R., Brewster, S., Webb, B. W., McQuay, M. Q., and Huber, A. M., Experimental Thermal & Fluid Science, 24, 35–46, 2001.)
flux, with the ellipsoidal radiometer data, in general, being lower than the foil gauge measurements. This is consistent with the anticipated tendency of the ellipsoidal radiometer to measure somewhat low in the high heat flux range. The lowest average heat flux values measured during firing periods using both the foil gauge and ellipsoidal radiometer were at location 1, with magnitudes of 421 and 429 kW/m2, respectively. This location is the closest to the batch feeder and cooler end wall. The batch blanket and nearby end wall, in addition to being at lower temperatures, have lower surface emissivities than molten glass, resulting in lower radiant energy transfer to the crown in this region from these surfaces. It has been previously shown that this region was found to also have the lowest mean glass/batch surface temperature of all measured locations (See Figure 32.9), which is consistent with lower radiation intensity to the crown from the glass/batch surface. Additionally, fuel firing rates at this location are slightly reduced, being about 12% lower in port 1 than the nominal fuel flow rate in ports 2 through 5 (see Table 32.2). The combination of the slightly reduced firing in this port and the proximity of this region to the cooler entering batch material and end wall results in incident hemispherical radiant heat flux levels at location 1 that are substantially lower than at all other measured locations. As one moves away from the batch feeder, a sharp increase in the average measured crown incident radiant heat flux is observed between locations 1 and 2. Crown incident radiant heat flux levels of approximately 555 kW/m2 were measured at location 2. This location is approximately 3.4 m farther downstream of location 1 and thus experiences less direct influence on incident radiant heat flux due to the previously discussed lower
© 2011 by Taylor and Francis Group, LLC
temperatures and emissivities of the cooler batch blanket and end wall, with a greater proportion and intensity of radiation received from the hotter molten glass surface (see Figure 32.9). Firing rates are also slightly higher in port 2 than in port 1, resulting in higher radiant flux from the flame firing directly beneath measurement location 2. In addition to the significant contribution of heat flux resulting from combustion directly below each measured crown location, it is noted that the hemispherical nature of the measurement also encompasses the additional contribution of incident radiant heat flux on the crown from combustion in adjacent firing ports to the degree to which they are radiatively viewed by the detector. Thus the incident radiant flux on the crown at location 2 also includes smaller radiation contributions from firing in nearby adjacent ports 1 and 3, whereas the heat flux incident at location 1 is affected by radiant energy received from combustion in port 2 and the adjacent cooler end wall. Downstream of location 2, the average crown incident heat flux profile continues to rise to a maximum of approximately 710 kW/m2 at locations 3 and 4, near the center of the combustion space. This rise is expected as locations 3 and 4 are farther downstream from the cooler batch blanket and end wall, while being closer to the spring zone where glass surface temperatures are at a maximum (see Figure 32.9). Further, these port locations experience high fuel firing rates with the lowest percentages of excess air of all the measured locations (see Table 32.2). The resulting higher flame temperatures and increased radiant energy emitted from combustion products in these regions combine with the higher glass surface temperatures to result in the peak crown incident radiant heat flux levels measured. Following the peak at locations 3 and 4, the heat flux profile declines, as one moves away from the spring zone, to near 610 kW/m2 at location 5 and approximately 575 kW/m2 at location 6, near the working end of the furnace. Although port 5 is in a region of high glass surface temperature (see Figure 32.9) and has high fuel flow rates during firing, it also has a high percentage of excess air (44%), resulting in diminished radiant energy transfer from combustion at this location. Location 6 experiences much less heat transfer from combustion than all other measured locations, due to the drastically reduced fuel flow rates during firing in this port, with significantly higher excess air (77%) than any of the other port location (see Table 32.2). Thus, in addition to less energy being released in port 6 due to lower firing flow rates, the flame temperature is also reduced due to higher excess air levels, which further lowers the radiant heat flux from the flame. These combustion effects, along with closer proximity to the cooler working end wall, result in lower incident radiant heat flux levels measured during firing cycles at location 6.
685
Glass Furnaces
32.5.2 Nonfiring Measurements
Crown incident radiant heat flux (kW/m2)
Time-averaged incident radiant heat flux measurements acquired at each location with the circular foil gauge during nonfiring reversal periods are shown in Figure 32.12, with the time-averaged measurements performed during firing cycles also being shown for comparison. During the short, 20–25 second reversal period the only radiant energy incident on the crown is that which is emitted by the molten glass or batch surfaces and interior combustion space walls, with the molten glass surface being the major radiation source. Therefore, these data should also reflect the influence of trends presented earlier in glass surface temperature measurements performed at the same locations (see Figure 32.9). Figure 32.12 indicates that the lowest incident heat flux measured during reversal, when firing does not occur, was also at location 1, with a magnitude of approximately 320 kW/m2. The low heat flux levels measured in this region can again be attributed to the proximity of the cooler batch blanket and end wall to this location. During nonfiring periods the measured heat flux profile downstream of location 1 rises in a similar manner as observed during the firing cycles. A sharp increase is again witnessed between locations 1 and 2, where the measured heat flux increases to approximately 455 kW/m2 as one moves away from the end wall and batch blanket. The trends measured during nonfiring reversal periods are caused solely by changes in radiation intensity from surfaces viewed by the detector at each measured crown location, which are a function of the temperature and emissivity of radiatively viewed
800 700 600 500 400 Firing Nonfiring
300 200
0
5
10 15 Position (m)
20
25
Figure 32.12 Comparison of time-averaged crown incident radiant heat flux measurements acquired during firing and nonfiring periods using the circular foil radiometer. (From Hayes, R. R., Brewster, S., Webb, B. W., McQuay, M. Q., and Huber, A. M., Experimental Thermal & Fluid Science, 24, 35–46, 2001.)
© 2011 by Taylor and Francis Group, LLC
batch, glass, and wall surfaces. The sharp increase in crown incident heat flux between locations 1 and 2 is in harmony with the glass surface temperature profile previously shown in this region (see Figure 32.9). Downstream of location 2, the measured crown incident heat flux without flame increases progressively at locations 3 and 4 to approximately 545 kW/m 2 and 565 kW/m 2, respectively. This rise is expected, as one approaches the hotter spring zone, and the radiative influence of the cooler batch blanket and end wall is reduced. The peak crown incident heat flux measured at location 4 during nonfiring periods is consistent with the estimated location of the spring zone, where glass surface temperatures are at a maximum. Downstream of the estimated position of the spring zone, the final two measured locations continue to exhibit high measured crown incident radiant heat flux levels that are close to those witnessed at locations 3 and 4, with heat flux levels at location 6, nearest to the furnace working end, being equal to those measured at location 4. The nonfiring heat flux measurement acquired at location 5 appears to be slightly lower than expected from the experimental glass surface temperature data (see Figure 32.9). This data point, acquired with the circular foil probe, exhibited higher signal variation than any other location measurement during reversal. Also, the most likely human error to occur is improper positioning of the probe into the access hole in the furnace crown, which would result in an erroneously low measurement. Considering these facts, this data point is suspect to being subject to instrument and/ or human error, resulting in a heat flux measurement that may be up to 10% lower than the actual value at this location. As indicated in Figure 32.12, the crown incident radiant heat flux measured at each location was increased, as expected, during firing cycles as compared with nonfiring cycles. The first four measured locations exhibited increased measured crown incident heat flux due to firing of 25–30%, nominally. At the first two locations, measured radiant heat flux to the crown increased due to combustion by approximately 105 kW/m 2. The radiant heat flux increase from combustion was greatest at the following two locations (3 and 4), with an average increase of over 155 kW/m 2 at these locations, where firing conditions were nearest to stoichiometric (see Table 32.2). At location 5 the increase in measured incident radiant heat flux due to firing is approximately 84 kW/m 2 (15%), where the excess air percentage in the flame was higher than average. Location 6 exhibits the least heat flux increase, with an increase of only 12 kW/m 2 (2%). The previously discussed unique firing conditions in port 6, with lower fuel flow rates, and high excess air, explain the
686
6 5
q"σ /q" (%)
4 3 2 1 0
0
5
10 15 Position (m)
20
25
Figure 32.13 Average firing cycle variation (standard deviation) of measured crown incident radiant heat flux at all locations. (From Hayes, R. R., Brewster, S., Webb, B. W., McQuay, M. Q., and Huber, A. M., Experimental Thermal & Fluid Science, 24, 35–46, 2001.).
minimal difference in firing and nonfiring incident heat flux. 32.5.3 Time-Resolved Experimental Results The higher frequency time-resolved data acquired with the circular foil heat flux radiometer permitted study of the temporal variation in crown incident heat flux during firing cycles. Figure 32.13 shows the standard deviation in measured crown incident heat flux during firing at each of the six measured locations. These data were calculated from measurements acquired continuously over typical 30–60 second periods during firing. The data indicate that the magnitudes in radiant heat flux fluctuations are relatively small, with a standard deviation of less than 6% of the mean heat flux, at each location. Generally speaking, the variation appears to be lowest at port locations with firing nearest to stoichiometric conditions and higher in regions with greater excess air during combustion. Location 6, although having the highest percentage of excess air, exhibited very little variation during firing, due to the previously noted absence of significant radiation from combustion in this port. An examination of the variation in crown incident heat flux over the entirety of typical 15-minute firing cycles is shown in Figure 32.14. Crown incident heat flux measurements obtained at location 4 using the ellipsoidal radiometer for firing cycles in each direction (N–S, and S–N) are shown as a function of time. These data illustrate the crown incident heat flux variation during typical firing cycles in this region. The trend in both cycles shows that heat flux to the crown increases during the first few minutes of firing, due to increased radiant
© 2011 by Taylor and Francis Group, LLC
Crown incident radiant heat flux (kW/m2)
Industrial Combustion Testing
700
650
600
550
N-S firing S-N firing
500
0
5
Time (min)
10
15
Figure 32.14 Cycle variation of measured crown incident radiant heat flux at location 4 for both firing directions over an entire firing cycle period. (From Hayes, R. R., Brewster, S., Webb, B. W., McQuay, M. Q., and Huber, A. M., Experimental Thermal & Fluid Science, 24, 35–46, 2001.)
energy from combustion, and presumably increased radiation from other surfaces (i.e., glass and walls), which are also being heated by the flame. Following this initial increase, the incident flux ceases to increase and begins to gradually decrease for the remainder of the firing cycle. This trend is most likely caused by the effect of the regenerators, which are utilized to preheat combustion air. The regenerators are heated by the previous firing cycle exhaust, and are then used to preheat inlet combustion air when firing reverses direction, recuperating energy, and improving the energy efficiency of the furnace. The heat flux trend observed suggests that the combustion air is preheated to higher temperatures during the earlier stages of a firing cycle, and as would be expected, the regenerators cool as they transfer heat by convection to the cool inlet air. This effect was previously reported [38]. Thus, during later stages of a firing cycle, less energy is available to preheat entering combustion air, resulting in lower inlet air temperatures. As a result, flame temperatures are slightly lowered, resulting in the observed decrease in heat flux radiated to the crown. The difference between the highest and lowest crown incident heat flux levels measured with the ellipsoidal probe over a single firing cycle at location 4, was approximately 10–15 kW/m2 (3–4%). The withinfiring-cycle variation shown in Figure 32.14 is typical of all locations. The nominal magnitude between high and low heat flux levels during a single 15-minute firing cycle at any given location was approximately 10–25 kW/m2 (2–6%). It is also apparent from the data that measured crown incident heat flux levels vary slightly with firing direction, as illustrated in Figure 32.14. A review of all data
687
Glass Furnaces
Crown incident radiant heat flux (kW/m2)
acquired in this study, indicated that the difference in measured heat flux levels due to firing direction did not follow an established pattern (i.e., N–S consistently higher than S–N or vice versa), but rather, the data indicated that variation commonly occurred from cycle to cycle at a given location, independent of direction or other controlled factors. This type of variation is expected, considering the complexity of the furnace physical phenomena, the number of uncontrolled furnace operation variables, and the accuracy of the heat flux instruments. The measured variation between firing cycles at a given location was modest, typically in the range of 20–40 kW/m2 (2–8%). The cycle-to-cycle variation did appear to be greatest in magnitude in regions of highest heat flux. Typical time-resolved data obtained using the circular foil heat flux gauge during nonfiring periods at several measured locations are shown in Figure 32.15. The measurements indicate that during the short 20–25 second reversal period, the crown incident heat flux at most locations measurably decreases. This decrease is most clearly evident in the regions of highest heat flux, with the exception of location 6, where a constant decrease during nonfiring periods was not consistently observed. The decrease in measured heat flux was typically from 20 to 50 kW/m2. This decrease results from the cooling of the molten glass surface and inner furnace walls during the brief time when firing does not occur. These trends are consistent with the previously shown surface temperature data that illustrated and quantified the cooling of the molten glass surface along the furnace axial centerline during typical reversal periods (see Figure 32.10).
600 550 500 450 Location 1 Location 2 Location 3 Location 4 Location 5
400 350 300 250
0
5
10 Time (s)
15
20
Figure 32.15 Typical time-resolved data of measured crown incident radiant heat flux at several locations during nonfiring reversal periods. (From Hayes, R. R., Brewster, S., Webb, B. W., McQuay, M. Q., and Huber, A. M., Experimental Thermal & Fluid Science, 24, 35–46, 2001.)
© 2011 by Taylor and Francis Group, LLC
32.6 Conclusions The objective of this chapter was to report glass surface temperature and crown incident heat flux data in an industrial, gas-fired, flat-glass furnace. Both time-averaged and time-resolved data were acquired in a sideport, 550 ton/day, flat-glass furnace at six crown access hole locations, aligned with the firing portneck centerlines. Glass surface temperatures were measured optically during nonfiring reversal periods, while crown incident heat flux measurements were performed during both reversal periods and firing cycles. Measurements in the combustion space indicate that the average glass surface temperature profile rises sharply from a low near 1700 K closest to the batch feeder to about 1850 K at the next downstream location, then continues to rise steadily down the axial length of the furnace along the centerline to a peak of approximately 1910 K in the region of ports 4 and 5. The molten glass surface temperature then drops to near 1800 K at the measured location closest to the furnace working end. The results of the time-resolved temperature data demonstrate that along the axial centerline of the combustion space, the surface boundary between batch material and molten glass exists in the immediate vicinity of location 1, nearest to the batch feeder. The data at this location reveal measured surface temperatures of both molten glass, near 1830 K, and much cooler batch material, near 1480 K. The time-resolved data also indicate that at downstream locations, a gradual cooling of the molten glass surface occurs during the 20-second nonfiring reversal period. The drop in temperature during this nonfiring interval is nominally in the range of 50K to 80K. The measured glass surface temperature nearest to the furnace working end does not exhibit this same cooling and stays nearly constant over the reversal period. The heat flux measurements indicate that the average incident radiant heat flux profile during firing cycles rises sharply from a low near 425 kW/m2 closest to the batch feeder to approximately 555 kW/m2 at the next downstream location, then continues to rise steadily down the axial length of the furnace along the centerline to a peak of approximately 710 kW/m2 in regions near the center of the combustion space. The measured incident radiant heat flux then declines to near 575 kW/m2 at the measured location closest to the furnace working end. The numerical predictions of crown incident radiant heat flux during firing cycles qualitatively show similar profile trends as the experimental results, however they overpredict the incident heat flux by 5–15% at most locations in the combustion space. During nonfiring reversal cycles, the heat flux measurements indicate that the average incident radiant
688
Industrial Combustion Testing
heat flux profile rises sharply from a low near 320 kW/ m2 closest to the batch feeder to about 455 kW/m2 at the next downstream location, then continues to rise steadily down the axial length of the furnace along the centerline to a peak of approximately 565 kW/m2 in the port 4 region, near the spring zone. A slight decrease in measured radiant crown incident heat flux, typically from 20 to 50 kW/m2, was observed at each location during the brief 20–25 second reversal period. These trends are consistent with the glass surface temperature data presented. The rise in radiant heat flux levels incident at the crown due to combustion during firing cycles is quantified, with a nominal increases of 105 kW/m2 at the two locations closest to the batch feeder and increases of approximately 155 kW/m2 at the two locations in the center of the combustion space. The contribution of radiant heat flux to the crown from combustion is not as appreciable near the working end with increases due to firing of approximately 85 kW/m2, and 12 kW/m2 at the final two measured locations, respectively. The contribution of radiant incident flux on the crown from combustion correlates well with firing conditions in each port location. Variation of heat flux to the crown during 15-minute firing cycles is typically 3–6% of the total incident heat flux, with the peak typically occurring one-third of the way into the cycle 5–6 minutes) and declining during the remainder of the period. The measured variation between cycles due to firing direction, and/or numerous other operating variables, is typically 2–8%, usually being greatest in regions of highest heat flux.
Subscripts
32.7 Nomenclature
a Two-color pyrometer proportionality calibration constant b Two-color pyrometer ambient temperature calibration constant C1 Constant in Planck’s spectral energy (or intensity) distribution C2 Constant in Planck’s spectral energy (or intensity) distribution e Emissive power p Photodiode proportionality constant g Radiative heat flux t Absolute temperature v Voltage Greek ε λ τ
Emissivity Wavelength Transmissivity
© 2011 by Taylor and Francis Group, LLC
b body detector m λ σ
Denotes blackbody property Emitting body Two-color pyrometer detector assembly Mean value Denotes wavelength dependent property Standard deviation
References
1. Eleazer, P. B., and Hoke, B. C. “Glass.” In Oxygen-Enhanced Combustion, edited by C. E. Baukal. Boca Raton, FL: CRC Press, 1998. 2. Drake, R. A. “Combustion Progress, Problems, Needs in the Glass Industry.” In Industrial Combustion Technologies, edited by M. A. Lukasiewicz, pp. 23–25. Warren, PA: American Society of Metals, 1986. 3. Deshmukh, Y. V. Industrial Heating. Boca Raton, FL: CRC Press, 2005. 4. Chen, T., and Goodson, R. E. “Computation of ThreeDimensional Temperature and Convective Flow Profiles for an Electric Glass Furnace.” Glass Technology 13, no. 6 (1972): 161–67. 5. Megahed, I.E.A. “The Prediction of Three-Dimensional Gas-Fired Combustion Chamber Flows.” PhD thesis, University of London, 1978. 6. McConnell, R. R., and Goodson, R. E. “Modeling of Glass Furnace Design for Improved Energy Efficiency.” Glass Technology 20, no. 3 (1979) 100–106. 7. Gosman, A. D., Lockwood, F. C., Megahed, I.E.A., and Shah, N. G. “The Prediction of the Flow, Reaction, and Heat Transfer in the Combustion Chamber of a Glass Furnace.” AIAA 18th Meeting on Aerospace Sciences, Pasadena, California, January 1980. 8. Mase, H., and Oda, K. “Mathematical Model of Glass Tank Furnace With Batch Melting Process.” Journal of Non-Crystalline Solids 38 and 39 (1980): 807–12. 9. Novak, J. D. “Application of Combustion Space Energy Calculations to Commercial Glass Furnaces.” Journal of Non-Crystalline Solids 38 and 39 (1980): 819–24. 10. Gosman, A. D., Lockwood, F. C., Megahead, I.E.A., and Shah, N. G. “Prediction of the Flow, Reaction, and Heat Transfer in a Glass Furnace.” Journal of Energy 6, no. 6 (1982): 353–60. 11. Carvalho, M. G. “Computer Simulation of a Glass Furnace.” PhD thesis, University of London, 1983. 12. Carvalho, M. G., and Lockwood, F. C. “Mathematical Simulation of an End-Port Regenerative Glass Furnace.” Proceedings of the Institution of Mechanical Engineers 199, no.2 (1985): 113–20. 13. Samião, V. S. “Simulação Mumérica de uma Fornalha Industrial.” MSc thesis, University of Lisbon, 1986. 14. Carvalho, M. G., Durão, D.F.G., and Pereira, J.C.F. “Prediction of the Flow, Reaction and Heat Transfer in an Oxy-Fuel Glass Furnace,” Engineering Computations 4, no. 1 (1987): 23–34.
Glass Furnaces
15. Carvalho, M. G., Oliveira, P., and Semião, V. “A Threedimensional Model of an Industrial Glass Furnace.” Journal of the Institute of Energy 143 (1988): 143–56. 16. Post, L. “Modeling of Flow and Combustion in a Glass Melting Furnace.” PhD diss., Delft University of Technology, Holland, 1988. 17. Carvalho, M. G., and Lockwood, F. C. “Thermal Comparison of Glass Furnace Operation with Oil and Natural Gas,” Glastechnische Berichte 63, no. 9 (1990): 233–43. 18. Carvalho, M. G., Semião, V., Lockwood, F. C., and Papadopoulos, C. “Predictions of Nitric Oxide Emissions from a Industrial Glass-Melting Furnace.” Journal of the Institute of Energy Vol. 63, No. 454 (March 1990): 39–47. 19. Boerstoel, G. P., Wieringa, J. A., and van der Meer, Th. H. “Numerical Modeling of Soot Formation in Glass Melting Furnaces.” In Heat Transfer in Radiating and Combusting Systems: 2, Eurotherm Seminar, no. 37, 167–80, 1994. 20. Boerstoel, G. P., van der Meer, Th. H., and Hoogendoorn, C. J. “Numerical Simulation of Soot Formation and Oxidation in High Temperature Furnaces.” Proceedings of ISTF-8, San Francisco, California, 1995. 21. Carvalho, M. G., Nogueira, M., and Wang Jian. “Assessment of Numerical Simulation of Industrial Glass Melter.” Verre 1 no. 5 (1995): 9–13. 22. Carvalho, M. G., Nogueira, M., and Wang Jian. “Mathematical Modeling of the Glass Melting Industrial Process.” Proceedings of 17th International Congress on Glass vol. 6, pp. 69–74, Beijing, China, October 9–14, 1995. 23. Wang Jian, and Khalil, M. “The Mathematical Modeling of Glass Batch Melting in Fired Furnaces.” 1995 SaintGobain Rapport SGR/EM—MK/VL No 4083/95, pp. 1–41. 24. Carvalho, M. G., Nogueira, M., Wang Jian, Ferlin, T., and Malvos, H. “Model-Based Study of a Glass Melting Furnace for Reduced Particle Emission.” Proceedings of International Symposium on Glass Problems, vol. 1, Istanbul, Turkey, September 4–6, 1996. 25. Carvalho, M. G., Speranskaia, N., Wang Jian, and Nogueira, M. “Modeling of Glass Melting Furnaces: Applications to Control, Design and Operation Optimization.” Proceedings of 5th International Conference on Advances in the Fusion and Processing of Glass, Toronto, Canada, July 27–31, 1997. 26. Carvalho, M. G., Wang Jian, and Nogueira, M. “Num erical Simulation of Thermal Phenomena and Particulate Emissions in an Industrial Glass Melting Furnace.” Proceedings of Conference Fundamentals of Glass Science and Technology, pp. 416–21, Växjö, Sweden, June 9–12, 1997. 27. Wang Jian, Carvalho, M. G., and Nogueira, M. “Physically-Based Numerical Tool for the Study of Glass Melt Quality.” Proceedings of 4th International Seminar on Mathematical Simulation in the Glass Melting, pp. 67–76, Vsetin, Czech, June 16–17, 1997. 28. Carvalho, M. G., Wang Jian, and Nogueira, M. “Physically-Based Numerical Tool for the Study of Cleaner Combustion in Glass Melting Furnaces.” Proceedings of 4th International Conference on Technologies and Combustion for a Clean Environment, vol. II, pp. 27.13– 27.20, Lisbon, Portugal, July 7–10, 1997.
© 2011 by Taylor and Francis Group, LLC
689
29. Newbold, J. A. “Combustion Measurements and Modeling of an Industrial, Gas-Fired, Flat-Glass Furnace.” Master’s thesis, Brigham Young University, Provo, UT, 1997. 30. Wang J., Carvalho, M. G., and Nogueira, N. “Investigation of Glass Melting and Fining Processes by Means of Comprehensive Model.” Proceedings of 5th International Conference on Advances in the Fusion and Processing of Glass, Toronto, Canada, July 27–31, 1997. 31. Wang J., Farias, T. L., Carvalho, M. G., and Nogueira, M. “Radiation Modeling Procedures for Numerical Simulation of Glass Melting Furnaces.” Proceedings of 4th International Seminar on Mathematical Simulation in the Glass Melting, pp. 10–20, Vsetin, Czech, June 16–17, 1997. 32. Wang, J. “Three-Dimensional Mathematical Model of Thermal Phenomena Occurring in Industrial Glass Melting Tanks.” PhD thesis, Instituto Superior Técnico, Lisbon, Portugal, 1998. 33. Wang J., and Carvalho, M. G. “Model-Based Study of Sand Grain Dissolution in Industrial Glass Furnaces.” Proceedings of 18th International Congress on Glass, San Francisco, CA, July 5–10, 1998. 34. Wang J., Carvalho, M. G., and Nogueira, M. “An Integrated Methodology for Glass Furnace Design.” Proceedings of 18th International Congress on Glass, San Francisco, CA, July 5–10, 1998. 35. Hoke, B. C., and Schill, P. “CFD Modeling in the Glass Industry.” In Computational Fluid Dynamics in Industrial Combustion, edited by C. Baukal. Boca Raton, FL: CRC Press, pp. 411–453, 2001. 36. Chang, S-L., Golchert, B., and Zhou, C. Q. “Computer Modeling of Glass Furnace Flow and Heat Transfer.” Ceramic Transactions, Advances in Fusion and Processing of Glass III, vol. 141, pp. 399–411, July 27–31, 2003/2004. 37. Abbassi, A., and Khoshmanesh, Kh. “Numerical Simu lation and Experimental Analysis of an Industrial Melting Furnace.” Applied Thermal Engineering 28 (2008): 450–59. 38. Victor, A. S., Costeira, J. P., Tomé, J. A., and Sentieiro, J. “A Computer Vision System for the Characterization of Flame in Glass Furnaces.” Proceedings of the 1991 IEEE Industry Applications Society Annual Meeting, September 28–October 4, 1991, Dearborn, MI. 39. Farmer, D., Heitor, M. V., and Vasconcelos, A. T. “New Methodology for Furnace Monitoring, Analysis and Control.” Glass Industry 10 (1992): 10–16. 40. Cassiano, J., Heitor, M. V., and Silva, T. F. “Combustion Tests on an Industrial Glass-Melting Furnace.” Fuel 73, no. 10 (1994): 1638–42. 41. Costa, M., Mourao, M., Baltasar, J., and Carvalho M. G. “Combustion Measurements in an Industrial Glass Melting Furnace.” Journal of the Institute of Energy 69, No. 479 (1996): 71–81. 42. Newbold, J., Webb, B. W., McQuay, M. Q., and Huber, A. M. “Combustion Measurements in an Industrial GasFired Flat-Glass Furnace.” Journal of the Institute of Energy 70, No. 483 (1997): 71–81. 43. Hayes, R. R., Wang, J., McQuay, M. Q., Webb, B. W., and Huber, A. M. “Predicted and Measured Glass Surface Temperatures in an Industrial, Regeneratively Gas-Fired Flat Glass Furnace.” Glass Science & Technology 72, no. 12 (1999): 367–77.
690
44. Hayes, R. R. “Glass Surface Temperature and Incident Heat Flux Measurements in an Industrial, Regenerative, Natural Gas-Fired Flat-Glass Furnace.” MS thesis, Brigham Young University, Department of Mechanical Engineering, 1999. 45. McQuay, M. Q., and Webb, B. W. “Effect of Rebuild on the Combustion Performance of an Industrial Gas-Fired Flat Glass Furnace.” Combustion Science and Technology 150, no. 1 (2000): 77–97. 46. Hayes, R. R., Brewster, S., Webb, B. W., McQuay, M. Q., and Huber, A. M. “Crown Incident Radiant Heat Flux Measurements in an Industrial Regenerative, Gas-Fired, Flat-Glass Furnace.” Experimental Thermal & Fluid Science 24, nos. 1–2 (2001): 35–46.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
47. Khoshmanesh, Kh. “Mathematical Simulation and Experimental Measurements of a Glass Melting Furnace.” MS thesis, Amirkabir University of Technology, Tehran, Iran, 2000. 48. Siegel, R., and Howell, J. R. Thermal Radiation Heat Transfer. 2nd ed. Washington, DC: Hemisphere, 1981. 49. Planck, M. “Distribution of Energy in the Spectrum.” Annals of Physics 4, no. 3 (1901): 553–63. 50. Wien, W. “Temperatur and Entropie der Strahlung.” Annals of Physics 2, no. 52 (1894): 132–65. 51. Baukal, C. Heat Transfer in Industrial Combustion. Boca Raton, FL: CRC Press, 2000.
33 Thermal Oxidizer Testing Bruce C. Johnson and Nate Petersen Contents 33.1 Introduction.................................................................................................................................................................. 691 33.2 Equipment and Facility Design Objectives.............................................................................................................. 692 33.3 Test Data Accuracy...................................................................................................................................................... 693 33.4 Testing Thermal Oxidizer Burners........................................................................................................................... 693 33.5 Testing Thermal Oxidizer Chambers....................................................................................................................... 694 33.6 Testing Thermal Oxidizer Waste Gas Injection Methods and Configurations.................................................. 695 33.7 Simulating Thermal Oxidizer Input Streams.......................................................................................................... 695 33.7.1 Combustion Air............................................................................................................................................. 695 33.7.2 Quench Medium............................................................................................................................................ 696 33.7.3 Burner Fuel..................................................................................................................................................... 696 33.7.4 Simulating Waste Streams............................................................................................................................ 696 33.7.4.1 Hazardous Vent (Endothermic) Streams.................................................................................. 696 33.7.4.2 Exothermic Waste Streams.......................................................................................................... 697 33.7.4.3 Aqueous Wastes............................................................................................................................ 697 33.7.4.4 Special Components..................................................................................................................... 698 33.8 Instrumentation........................................................................................................................................................... 698 33.8.1 Chemical Species Analysis........................................................................................................................... 698 33.8.1.1 Analytical Methods and Instrumentation................................................................................ 698 33.8.1.2 Chemical Sample Collection....................................................................................................... 700 33.8.2 Flow Measurements...................................................................................................................................... 700 33.8.3 Temperature Measurement.......................................................................................................................... 701 33.8.4 Pressure Measurement.................................................................................................................................. 702 33.9 Conclusion.................................................................................................................................................................... 703 References................................................................................................................................................................................. 703
33.1 Introduction This chapter discusses the advantages in using a thermal oxidation (TO) test facility and the considerations in designing such a facility. With ever increasing demands on TO equipment i.e., to improve destruction removal efficiencies [DREs], to lower NOx emissions, reduce fuel usage, and increase up-time performance, having a facility to test equipment and processes has become increasingly important. Test facilities are used to determine emissions from previously untested wastes or at untried operating conditions. This data can then be used by companies and regulator agencies to help establish operating conditions in permits for new systems. Thermal oxidizer test facilities also provide a place to simulate and work on problems that can be sometimes
encountered in the startup of new one-of-a-kind commercial systems or when the waste stream compositions or flow rates in operating systems have changed. Solving problems in commercial operating systems without having the benefit of test facility data often results in a lengthy and costly method of problem solving. Some of the difficulties that can be encountered in trying to solve commercial system problems result from the inability to make equipment changes since the system is required to stay online, and not being able to significantly alter operating conditions since the desired operating conditions may fall outside the range of the equipments’ capability or operating permit. A test facility can also be used to study conditions that may occur infrequently (i.e., emergency conditions in a plant where performance of the system needs to be proven). This includes testing potentially dangerous 691
© 2011 by Taylor and Francis Group, LLC
692
conditions, where numerous redundant safety systems need to operate correctly to ensure safe operation of the TO system. A test facility is also important for optimizing existing technologies and for developing new technologies [4]. It is generally very difficult to do product development in operating units. A test facility can be used to very carefully control specific parameters to study their effects and test facility equipment is fitted with numerous sample points, pressure taps, viewports, and temperature measuring locations to better understand the effects of equipment and/or operating parameter changes. Testing results can also be used to identify additional areas for research and development. A large scale test facility can be effectively used to train plant operators in a controlled environment, without the time constraints and distractions associated with starting up equipment to meet production demands. Various conditions can be simulated for training purposes in a TO test facility that are not easily obtained in a TO unit operating in a plant. For example, multiple startups and shutdowns can be easily performed in a test facility where this would not be practical in an operating commercial system.
33.2 Equipment and Facility Design Objectives The primary purpose of the test facility is to have equipment capable of creating surrogate waste streams similar to those commonly encountered in industry and destroy them under controlled conditions with adequate instrumentation to quantify parameters that affect the process. Destruction of waste streams is achieved by proper mixing, sufficient temperature, and adequate residence time in the TO. These parameters are often referred to as the three T’s of combustion: time, temperature, and turbulence. In addition to waste destruction, the effects of the three T’s on other parameters such as CO and NOx emissions can also be carefully studied in a test facility. A schematic of a horizontal thermal oxidizer is shown in Figure 33.1. A properly designed test facility TO system should be equipped with numerous sample ports, pressure taps, thermocouples, gas analyzers, or any other instrument that may provide information that can be used to characterize the performance of the equipment. The facility should be designed to be modular to allow components to be used for multiple purposes and also allow for the insertion of additional components that may be needed from test to test. The utilities, control room, and instrumentation also need to be designed to allow for simple additions and change outs that may be required.
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
Flue gas
Waste stream Stack To Air Fuel
Figure 33.1 Schematic of a common configuration of a horizontal thermal oxidizer.
Additionally, the equipment should be designed to have capabilities to operate over a very wide range of conditions such as throughput and temperature. The materials (e.g., refractory) need to be capable of operating over a wide temperature range and have durability to handle significant thermal cycling. Provisions should be in place to allow for a wide variety of fuels and diluents if they are needed. The supply of these components and their metering devices need to be adequate to allow for careful controlling of test conditions so that high quality test data can be obtained. The John Zink test facility has successfully used a horizontal TO system on top of a long metal plate to allow for easy modification of equipment configurations. The horizontal vessel connects directly to the stack and additional vessels and thermal expansion are accommodated by sliding the equipment on top of the metal plate. Flexible piping is used on all of the connections to the system to also allow for thermal expansion and modularity. Stack design should be carefully considered when designing a TO test facility since the stack can be the single most costly equipment component and usually requires the most elaborate and costly foundation. Some of the major considerations that need to be addressed in determining the stack size and design include: Will the stack ever be used to test natural draft burners? What is the maximum firing rare that could ever be expected? And what is the required or desired stack height? Other stack features to consider include providing additional stack nozzles so a single stack could potentially be used with other pieces of equipment and leaving free space around the stack so other equipment could be either temporarily or permanently attached to it. Other system design considerations include providing a concrete test pad that is large enough and strong enough to allow for the future addition of equipment and to permit the maneuvering of a crane and fork truck for installing and changing out system components. The addition of future equipment could possibly include a
693
Thermal Oxidizer Testing
second TO or adding components to an existing TO such as a quench system, venturi scrubber, packed column absorber, or an electrostatic precipitator. With a TO test facility important operating parameters can be eveluated using a multitude of equipment configurations. For example, in order to study the effects of turbulence on DRE different types of burners may be used, the pressure drop of the burner can be changed, the TO vessel length to diameter ratio (L/D) can be changed or other geometric features of the TO could be investigated such as the addition of a choke.
33.3 Test Data Accuracy One of the most important aspects of using a test facility to obtain experimental data is having the ability to check and verify the accuracy of the data being collected. In a test facility with properly calibrated instrumentation mass and energy balance closures within a few percentages are frequently achieved. When field data is obtained from commercial thermal oxidizers the quality of the data is often questionable. This is because there is seldom time to obtain steady-state conditions, means to verify the accuracy of the instrumentation, or even complete data to close a mass and energy balance around the process to see if the measured data appears to be correct. In commercial systems the accuracy of instruments such as fuel and air flow meters is frequently poor due to the type of meter, placement of the meter, initial setup (mol.wt., temperature range, pressure, etc.), or even as the result of erroneous scaling factors. Attempts made to try to determine what instrument or if an instrument is in error by making mass and energy balance calculations are also usually not productive since all of the necessary parameters are usually not available to close a mass and energy balance around the system. For example, waste gas flow volumes are often not measured since they may contain aerosols, solids, or tar-like constituents that can plug or coat flow meters causing them to fail or to have poor accuracy. Even if a waste gas flow rate is accurately measured, the composition of the waste is often constantly varying so closing a mass and energy balance is still not possible as the real-time composition of the gas is still needed and is almost never measured. When collecting data in a TO test facility many things can be done to ensure that the instruments used and the data collected is highly accurate. In the case of fuel gas flow measurements, multiple methods of measurements in series can be used. For example, at the John Zink Thermal Oxidation Test Facility, three different fuel gas flow meters were
© 2011 by Taylor and Francis Group, LLC
installed in series in order to determine that the original, brand new, properly installed and calibrated, gas flow meter was in error by approximately 10%. Once it is determined that a highly accurate fuel flow rate is available, the fuel rate can be used to check the accuracy of the air flow meter if an oxygen analyzer is available. This is done by measuring the flue gas oxygen content while operating the TO at a fixed fuel flow and air flow rate. A mass and energy balance is then calculated using the measured fuel flow and adjusting the air flow rate (if need be) to get the calculated oxygen value to match the measured oxygen value. The difference between the measured and calculated air flow, if there is one, is the error in the air flow meter. This method has been found to be very effective in determining the accuracy of air flow meters because the accuracy of a good extractive-type oxygen meter is very high (within a few hundredths of a percentage) if it is frequently calibrated with certified calibration gases. When performing any testing in a thermal oxidizer test facility it is highly recommended that a mass and energy balance is calculated for each test condition before the test is started. Not only does it help to ensure that the test data is accurate but it allows measurement errors to be quickly recognized so the problem can be immediately fixed rather than finding out later and having to repeat an entire series of tests. Repeating tests can often entail significant expense when it is considered that a TO test system may nominally fire 5 MM Btu/hr of fuel and that it often has to operate up to 24 hours just to ensure that the system is at thermal equilibrium before testing can be started. Even more costly is the labor required to repeat the test and the disruption to other scheduled tests.
33.4 Testing Thermal Oxidizer Burners Testing and developing burners specifically for use in thermal oxidizers is a very important function of a TO test facility since burners used in other applications, such as process burners and boiler burners, are not necessarily well suited in thermal oxidizers. One of the main differences in TO burners is that they fire into a combustion chamber with very hot refractory walls, whereas with many other types of burner applications the combustion chamber walls are relatively cool since heat is being extracted through the chamber walls. Since TO combustion chamber walls are hot, to maximize the destruction of wastes, some NOx reduction technologies that have been developed for burners in processes with cooled walls are not effective with TO burners. The technology employed in cooled wall
694
chambers primarily involves internally recycling flue gases taken from the cooler side walls and inspirating the gases into the burner throat to reduce NOx. However, other techniques can be used to reduce thermal NOx in TO burners. Some of these techniques include fuel gas staging, air staging, partial premix of air and fuel, using numerous small fuel gas tip or ports, and external flue gas recirculation. Another difference between thermal oxidizer burners and other types of burners is that they are frequently required to fire much higher quantities of excess air. The high amounts of excess air are required because wastes downstream of the burner may require significant quantities of air for combustion. Also, high excess air may be required to control the combustion chamber temperature since heat is not withdrawn through the chamber walls. In addition, thermal oxidizers sometimes employ burners that are designed to operate sub-stoichiometrically. Figure 33.2 is a cutaway view of an R&D thermal oxidizer test burner that is equipped with a number of NOx reduction methods so many NOx reduction techniques can be evaluated using a single burner. The test burner NOx reduction methods include primary, secondary, and tertiary fuel gas injection in addition to an internal fuel gas ring in a secondary air annulus around the main burner to allow testing of premixed fuel and air. All of the fuel gas delivery methods contained in the test burner would not typically be used simultaneously, but rather external valving is used to direct fuel gas to various combinations of delivery points and at different percentages. In addition to low NOx emissions, other desirable features of a TO burner include highly turbulent mixing of waste gases or waste liquids, stable operation, a high turndown, quiet operation, reasonable cost, and long life.
Industrial Combustion Testing
Highly turbulent mixing is especially desirable in TO waste gas and waste liquid burners since turbulence ensures that the wastes are rapidly mixed with hot combustion products and that wastes are contacted with the available oxygen to ensure waste burnout. Much of the turbulence in the TO system is generated with the combustion air flow pattern. This is one reason forced draft burners are frequently used on thermal oxidizers. Testing different geometries and pressure drops of burners can be used to develop improved equipment, determine optimum operating parameters for existing equipment, or characterize existing equipment. The relative turbulence created by combustion air entry for a given burner geometry can be correlated to the airside pressure drop of the burner. Therefore, test burners should be equipped with appropriate pressure taps to enable airside pressure drop measurements. When developing new burners, documenting the performance of existing burners, or improving commercial burner designs special attention should be paid to the construction of the test burner. The test burner should be constructed to facilitate modifications, have sufficient spare nozzles and connections, and incorporate sight glasses. Test burners are preferably equipped with separate fuel lines with valves to each set of fuel tips. This allows control of the fuel splits to each set of tips without changing tip drillings. Instrumentation should also be in place to determine the fuel flow to each set of tips. This can be done by placing flow meters in each tip’s fuel line or installing pressure gauges that measure the fuel tip pressure. Even with independent flow control to each set of fuel tips, it is desirable for the fuel tips to be easily removable to enable experimentation with different drillings and port angles. Testing different fuel tip and fuel flow configurations makes it possible to quantify minor species pollutants (NOx, CO) for different configurations that may be used in thermal oxidizer designs. Testing these different configurations at a range of flow rates and fuel splits provides information regarding the operating and stability window for the equipment.
33.5 Testing Thermal Oxidizer Chambers
Figure 33.2 Cutaway view of a thermal oxidizer test burner capable of both fuel and air staging.
© 2011 by Taylor and Francis Group, LLC
Oxidation reactions of the waste occur primarily in the thermal oxidizer chamber and the three T’s of combustion can be evaluated by measurements made in this chamber. Numerous nozzles should be incorporated into the chamber design to enable measurement of gas concentrations and temperature throughout the chamber.
695
Thermal Oxidizer Testing
The effect of residence time on waste destruction is quantified by sampling at various locations along the axial length of the combustion chamber. The effect of turbulence and effectiveness of mixing can be evaluated by measuring gas concentrations and temperature at various radial positions at each axial location. Gas concentration and/or temperature stratification can occur in both the vertical and horizontal directions. Therefore, sampling in both planes is preferable. The gas sampling probe and thermocouple should be designed to have adequate length to have the capability to probe from the vessel centerline to the vessel wall. Care should be taken to seal the sample port after the probe has been inserted to prevent fresh air leaking into the vessel and corrupting the measurements. Care should also be taken when inserting and retracting the probes to prevent injury. The probes are long and get very hot inside the oxidation chamber. Furthermore, conditions may exist where hot gases blow out of the sample port. Adequate personal protection equipment and proper procedures need to be followed when taking samples. Permanent gas sample lines and thermocouples should also be installed at the exit of the thermal oxidizer chamber to allow for reliable operation of the thermal oxidizer.
33.6 Testing Thermal Oxidizer Waste Gas Injection Methods and Configurations There are numerous methods and equipment configurations for injecting waste gases into a thermal oxidizer. This is one of the reasons for designing a modular test unit. Waste gas properties vary widely from each application. Since waste gas properties can vary widely, designing a waste gas injection system also varies widely. For example, a waste stream with a heating value near 900 Btu/scf available at 5 psig could be injected essentially the same as a burner fuel; whereas, a waste stream with a heating value of 1 Btu/scf available at 5″ w.c. might be injected around a manifold downstream of the burner combustion chamber. It is very difficult to ascertain the quantitative effect of a waste stream interacting with the combustion zone without test data. In some cases, waste gases interacting with the burner flame may be beneficial in achieving reduced pollutant emissions. In other cases, the waste gas may act to quench or reduce the rate of oxidation reactions in the flame and inhibit destruction of the waste gases. Testing can be a very valuable tool when evaluating how best to introduce waste gas into a thermal oxidizer.
© 2011 by Taylor and Francis Group, LLC
33.7 Simulating Thermal Oxidizer Input Streams Process streams are unique to the location or plant they are created at and can’t be easily transported to the test facility. The process streams can be simulated by blending each of the pure components together to closely match the process stream, but this is often impractical due to the number of compounds needed, equipment requirements, and economic considerations. Therefore, a priority of physical and chemical properties for the process stream needs to be developed so a simpler, cost-effective surrogate can be used in place of the process stream for the test. The following discussion gives guidance on what properties are important and how they may be prioritized for different process streams. Figure 33.3 shows a metering skid where multiple components can be independently metered and mixed for simulating an input stream. 33.7.1 Combustion Air Thermal oxidizers can be designed to operate with fresh ambient air, preheated air, or even oxygen-containing waste gas steams as the combustion air source. The air source composition and temperature can have a profound effect on the operation of the system. Therefore, it is important to conduct testing with a representative combustion air stream. The physical properties of the test combustion air source should be matched as close as is practical to the actual air source that will be used in the field. The most important properties of the combustion air stream are the temperature and oxygen content. The concentration
Figure 33.3 Test facility metering skid. Multiple components are independently metered in each line and mixed in a header to be delivered into the test equipment.
696
of inerts, especially those with relatively high heat capacities (e.g., CO2, H2O), should also be factored into how a given test should be conducted. 33.7.2 Quench Medium Thermal oxidizers are usually designed as adiabatic chambers. Heat recovery equipment is common but is almost always located downstream of the thermal oxidizer chamber. Therefore, quench fluids are used to control the temperature of the flue gases in the thermal oxidizer chamber by direct cooling. The most common quench fluids are liquid water, steam, or air. The quench fluid affects the composition and volume of the resulting flue gases. Therefore, the tested quench fluid should be the same as the actual quench fluid of interest. 33.7.3 Burner Fuel Thermal oxidizers are generally operated with natural gas as the primary fuel source. However, different types of fuels such as LPG, fuel oil, or refinery gas can be used. Most natural gas blends that are used across the world are similar enough that the performance of the thermal oxidizer will not be significantly affected by conducting a test with any local natural gas supply to simulate another. Surrogates of LPG can be made by blending mixtures of propane, butane, and natural gas. The surrogate LPG fuel should closely match the heating value and molecular weight of the actual fuel. Liquid fuels can be simulated using mixtures of common petroleum oils (such as No. 2 or No. 6). Liquid fuel requires atomization to allow for efficient combustion. The atomization method has a very large effect on the combustion process. Therefore, it is critical to test the liquid under atomization conditions that will be used in the commercial unit. Matching the injection geometry, spray angle, pressure drop, and atomization fluid to liquid fuel ratio is very important in simulating liquid fuels since the oxidation reactions are heterogeneous. The most important fuel characteristics to match for liquid fuels are heating value, density, and viscosity. Refinery gas blends may contain significant quantities of hydrogen, inert gases, methane, and/or heavier hydrocarbons. Refinery gas blends can usually be simulated by matching the H2 concentration, inert concentration (using N2 or CO2) into blends of natural gas to match the heating value of the actual fuel. Refer to the process burner testing recommendations by Baukal et al. [1,2] for additional details on simulating refinery fuels. 33.7.4 Simulating Waste Streams There are multiple combustion properties that need to be considered when attempting to simulate waste
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
streams. Matching all of the combustion properties are usually not possible or practical. Therefore, identifying the most critical concern in the ability to oxidize the waste should be done so that a priority can be given to matching the most important combustion properties. In some situations the TO may be limited by mixing, in other situations it may be limited by temperature, or secondary pollutant formation may be the primary concern. 33.7.4.1 Hazardous Vent (Endothermic) Streams Hazardous vent streams typically contain trace quantities of toxic compounds or pollutants in a diluent such as air, nitrogen, or carbon dioxide. These streams are commonly described as “endothermic” (even though they have a heating value) because a significant amount of energy is required to initiate and sustain oxidation of the species. The primary concerns when treating these waste streams are to maximize DRE and minimize energy consumption. One of the most important physical properties to consider in a hazardous vent stream is the heat capacity of the stream since it will determine how much heat must be supplied to get the mixture to the ignition temperature. Combinations of air, nitrogen, carbon dioxide, and steam should be metered to match the concentrations in the vent stream of interest or at least mimic the heat capacity. Matching the actual diluent components should be done where it is practical since they can participate in the chemistry of oxidation reactions (e.g., O2 content, OH radical formation). Many times the pollutants of interest are toxic to breathe such as H2S or benzene. Testing with hazardous components is strongly discouraged due to safety of personnel. There are many less toxic surrogates that may be used to evaluate whether the target components can be oxidized in the system. Research sponsored by the EPA [3] was done to evaluate the thermal stability of 20 common hazardous waste compounds. Table 33.1 ranks the compounds in the study according to the difficulty of incinerating the compound. Notice that methane (CH4) is more difficult to oxidize compared to most of the compounds studied and, for the most part, is considerably less toxic. Therefore, simulating the hazardous species with an equivalent concentration of methane creates a more difficult stream to oxidize. If a condition can be created that will thermally oxidize trace quantities of methane in an endothermic waste stream, it is reasonable to assume that most other pollutants or fumes can be similarly oxidized. Testing experience at the John Zink Test Facility has indicated that the oxidation of carbon monoxide in endothermic waste streams is also more difficult compared to most other hydrocarbons. Using carbon monoxide as
697
Thermal Oxidizer Testing
TABLE 33.1
TABLE 33.3
Incinerability of Several Common Hazardous Air Pollutants [3]
Simulation of Exothermic Waste Stream
Rank of Difficulty to Incinerate
Exothermic Waste Gas Properties
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Compound Name Acetonitrile Tetrachloroethylene Acrylonitrile Methane Hexachlorobenzene 1,2,3,4-tetrachlorobenzene Pyridine Dichloromethane Carbon tetrachloride Hexachlorobutadiene 1,2,4-trychlorobenzene 1,2-dichlorobenzene Ethane Benzene Aniline Monochlorobenzene Nitrobenzene Hexachloroethane Chloroform 1,1,1-trichloroethane
Compound Formula C2H3N C2Cl4 C3H3N CH4 C6Cl6 C6H2Cl4 C5H5N CH2Cl2 CCl4 C4Cl6 C6H3Cl3 C4H4Cl2 C2H6 C6H6 C6H7N C6H5Cl C6H5NO2 C2Cl6 CHCl3 C2H3Cl3
TABLE 33.2 Simulation of Hazardous Vent Stream Hazardous Vent Stream Composition
Surrogate Vent Stream Composition
79 vol% CO2 4 vol% O2 12 vol% N2 5 vol% H2O 150 ppmv chlorobenzene 50 ppmv propane
81 vol% CO2 19 vol% air (4 vol% O2, 15 vol% N2) 200 ppmv CH4 (natural gas)
a test surrogate is also possible but it is more toxic than methane. Methane (contained in natural gas) and carbon monoxide also make good surrogates because they are relatively easy to obtain, handle, and meter. Furthermore, common CEM analyzers detecting CO and unburned hydrocarbons (UBH) can be used to measure their presence for determining destruction removal efficiency. Table 33.2 illustrates an example of a test surrogate that may be used to simulate a hazardous vent stream. 33.7.4.2 Exothermic Waste Streams An exothermic waste stream contains enough chemical energy to generate temperatures high enough to sustain
© 2011 by Taylor and Francis Group, LLC
88 vol% CO2 6.5 vol% H2O 2.3 vol% CH4 1.6 vol% C3H8 1.2 vol% C5H12 0.4 vol% C8H18 LHV = 126 Btu/scf Air demand = 1.3 mol air/mol waste gas
Surrogate Waste Gas Properties 94.6 vol% CO2 5.4 vol% C3H8
LHV = 125 Btu/scf Air demand = 1.29 mol air/mol waste gas
continuous oxidation without adding additional energy. In some cases, these types of waste streams can present difficulty in achieving oxidation due to mixing limitations, droplet vaporization, and/or complicated fuel chemistry. Testing conducted with exothermic waste streams may be needed to identify optimum operating conditions and/or equipment configurations for systems destroying these wastes. The stream properties having the largest effect on operating conditions are the heating value, oxygen demand, and properties that may effect mixing and dispersion (especially for liquid wastes). Matching these properties as close as practical will ensure that the results from the test equipment are representative of commercial equipment results. To a lesser extent matching the inert components should be attempted when practical with the emphasis on the components with the highest heat capacities such as H2O and CO2. Table 33.3 illustrates how an exothermic waste gas stream may be simulated with a simpler surrogate waste stream. 33.7.4.3 Aqueous Wastes Aqueous wastes contain oxidizable compounds within a liquid water stream and may be endothermic or exothermic depending on the concentration of oxidizable compounds. Proper atomization of the liquid stream is probably the most critical aspect of handling aqueous wastes. This is because the droplet must be small enough to be capable of completely evaporating within the thermal oxidizer chamber and still allow for sufficient residence time to oxidize the combustible compounds. The fluid nozzle, pressure drop, and atomization fluid (e.g., air or steam) demand (if dual fluid atomization is used) should be matched to what is used in the field to obtain a meaningful simulation. The composition of the test fluid should be selected to match the heating value, viscosity, and density in the waste stream of interest.
698
Industrial Combustion Testing
33.7.4.4 Special Components In some cases a species can not be simulated using a simpler surrogate. A particular combustion property of the species may be the limiting case in achieving good combustion. For example, conjugated hydrocarbons (e.g., benzene, butadiene, propylene) have a tendency to form smoke. The double bonds in these species make them prime precursors for forming soot. Simulating the smoking tendency can’t be done with simple hydrocarbon fuels such as methane or propane. The actual compound responsible for the smoke formation may be needed for such a test. At least, a species of similar chemical structure should be used to collect meaningful data. Components with fuel bound nitrogen are another common species that cannot be simulated with simple hydrocarbon fuels. This is not to say that the actual chemical bound species must always be used for the test. Hydrogen cyanide (HCN) is obviously not desirable for testing due to its high toxicity. Ammonia or pyridine could be used to provide the necessary fuel bound nitrogen for the test. The surrogate fuel containing fuel bound nitrogen should be blended in the appropriate amount to match the nitrogen weight composition of the actual fuel.
33.8 Instrumentation Any testing campaign requires measurements to evaluate the thermal oxidizer’s performance. The set of instrumentation used for the test should be carefully chosen. Each instrument will have certain capabilities and limitations that will affect the quality of the test data. The instrument that is used needs to be capable of making measurements over the entire testing range. Furthermore, the instrument needs to have suitable accuracy for the expected measurement range. The durability and maintenance requirements of the instruments will also affect the selection. 33.8.1 Chemical Species Analysis 33.8.1.1 Analytical Methods and Instrumentation There may be many types of chemical species of interest for a given test. The most common types of analyzers associated with combustion equipment are contained in a rack or cabinet as seen in Figure 33.4. The typical analyzers found in one of these cabinets are a paramagnetic oxygen analyzer, infrared absorption carbon monoxide analyzer, UBH analyzer, and a chemiluminescent nitrogen oxide analyzer. Commercial sulfur dioxide analyzers based on infrared or ultraviolet absorption are also readily available to fit into one of these racks. The
© 2011 by Taylor and Francis Group, LLC
Figure 33.4 Commercial test rack with the following components from top to bottom: CO meter, oxygen meter, NOx meter, total hydrocarbons (THC) meter, and gas conditioning system. Smaller, less accurate, portable combustion analyzers are also available. These analyzers are generally equipped with solid state sensors to measure commonly encountered combustion gases.
analyzers found in these cabinets contain sample conditioners to remove water vapor from the stream since the analyzers can not handle condensing water. The water vapor is removed by condensation in chillers and further removed with desiccants. Since water vapor is removed prior to any measurements, the chemical species concentrations are measured on a dry basis rather than the true concentration at the sample point. Mass and energy balances are required to determine how the analyzer dry measurements and the true wet measurements relate to each other. Small portable analyzers equipped with multiple electrochemical sensors are also available for combustion flue gas analysis. Figure 33.5 depicts a common portable combustion gas analyzer. Other species that may be of interest require more specialized equipment that is more difficult to use and/or interpret results. A Fourier transform infrared spectrometer (FTIR), as can be seem in Figure 33.6, can be used to detect a wide array of components. The FTIR requires that the species of interest be “IR-active.” Most polar and polyatomic molecules are IR-active. The FTIR
699
Thermal Oxidizer Testing
Figure 33.5 Portable combustion gas analyzer. A small handheld probe is inserted into the combustion gases and the extracted gases are analyzed using electrochemical cells installed in the instrument.
Figure 33.6 FTIR gas analyzer. Sample gases are continuously flowed through the instrument while it repeatedly scans the absorption spectrum over infrared wavelengths. This unit can be programmed to identify target species and quantify them.
absorption spectrum is useful for characterization of gases since each molecule has its own unique absorption spectra sometimes referred to as a “fingerprint.” The absorption spectrum is also useful for quantification since the molecule concentration is proportional to absorption. Many species may be measured simultaneously with the FTIR but the accuracy and detection limits will vary depending on the absorption spectra. Nonlinear spectral fitting can be used to quantify species with overlapping absorption spectra. Gas chromatography (GC) can also be used to measure component concentrations in a mixture. A GC separates components in a mixture inside a column by differences
© 2011 by Taylor and Francis Group, LLC
Figure 33.7 In situ oxygen analyzer. The oxygen analyzer in this unit is not actually located in the stack. However, the sensing element is very close to the sample location to minimize lag time. Flue gases are rapidly educted across an electrochemical sensor that measures oxygen concentration in real time.
in polarity. Each species in the mixture has a slightly different molecular polarity, which causes the molecules to adsorb and desorb from the column surfaces at different rates, which affects their retention times in the column. As the separated gases exit the column they must be detected. Different types of detectors can be used to detect the presence of target species. Flame ionization (destructive) and thermal conductivity (nondestructive) detectors are the most common detectors used with GC columns. The column type, GC oven temperature, sample quantity, carrier gas flow rate, and other analytical parameters must be optimized to separate and analyze a given mixture. In general, GC is most useful for the characterization and analysis of hydrocarbon mixtures that are easily vaporized. Several other types of analytical instruments are available for chemical characterization and analysis such as electrochemical sensors (such as the in situ O2 analyzer shown in Figure 33.7) or wet chemistry and laboratory methods. Some of the most important factors affecting the selection of chemical analytical equipment are: • • • • • • • •
Measurement accuracy and sensitivity Background (matrix) interference Chemical interference Response time Ease of instrument operation Ease of data analysis Durability of instrumentation Cost of instrumentation and/or analysis
700
33.8.1.2 Chemical Sample Collection Chemical species measurements can be made from extracted samples or nonintrusively. A special concern of nonintrusive measurements is the interference from nontarget species. Electrochemical sensors exhibit cross sensitivities to other gases and can artificially increase the reported value of interest. Optical based measurements may similarly experience interference from other gas molecules and/or particulates. Extractive samples have the advantage of being able to be conditioned to make the sample more amenable to the desired measurement method. Even though measurement interferences can be controlled by extracting and conditioning a sample, the desired measurement may be corrupted by postextractive degradation or reactions. Figure 33.8 shows an extractive sample Probe connected to a flexible sample line that can be used to extract, samples from a vessel at multiple locations and depths. Some of the most important factors affecting the design or selection of a sample collection system are:
Industrial Combustion Testing
• Possible interference or reactions with other species in the sample • Durability of equipment • Extraction (lag) time • Ease of operation 33.8.2 Flow Measurements
• Material compatibility at sampling point (temperature rating, corrosion and fouling resistance, possible reactivity with target compounds) • Instrumentation compatibility with sample (condensation, or chemical attack to sensitive components, temperature limitations) • Time scale required for data acquisition • Sample interactions or reactions with the collection system
Flow measurement is critical during testing since the flow rates need to be metered accurately to carefully control the experimental conditions. The flow measurements are also essential in the analysis and characterization of the test data. There are several methods for measuring flow. Regardless of the method, care needs to be taken in setting up the flow measurement. The flow should be extremely uniform at the point of measurement. Flow straighteners may be needed to eliminate secondary flow patterns in the pipe or duct. The flow is preferably uniform and uninterrupted for 20 pipe diameters upstream of the measurement location and also 10 diameters downstream. Additional instrumentation for density corrections (temperature, pressure) to the flow measurement may also be required. Orifice plate flow meters, as seen in Figure 33.9, are good devices for measuring flows in gases or liquids. Orifice plates are relatively cheap and are very reliable and accurate. The measurement range for a given orifice is limited but different diameter orifice plates can be easily changed out. Orifice plates take higher pressure drops than other flow meters, which may make them unsuitable for fluids only available at low pressure drops.
Figure 33.8 Extractive sample probe. The tube shown in the bottom of the figure is used to draw flue gases out of the test unit. The extracted gases can be conditioned and analyzed. The gas samples usually take 10–30 seconds to be conditioned and analyzed using the equipment shown in the figure.
Figure 33.9 Orifice plate meter. The orifice plate is located between the flanged connection behind the pressure transmitter. The flanges are designed with pressure taps that connect to the pressure transmitter, which monitors the pressure drop across the orifice. The calibrated data acquisition systems continually record the flow.
© 2011 by Taylor and Francis Group, LLC
701
Thermal Oxidizer Testing
Combustion and quench air sources are generally not available at pressures above 1 psig due to increased power requirements and high equipment costs of higher pressure fans and blowers. Therefore, flow meters that do not require much pressure drop, such as thermal mass flow meters or Pitot tubes are often employed. These flow meters are generally less accurate than orifice plate flow meters but they tend to give reproducible results of reasonable accuracy. These flow meters are usually capable of measuring flow over a wider range compared to an orifice plate flow meter. Thermal mass flow meters, as seen in Figure 33.10, have probes with sensing elements on the end of the probe. The sensing element measures heat transferred between two points on the end of the probe and correlates this to mass flow. Since a single probe is used, the measurement is determined locally at the probe position and the overall flow in the duct must be inferred. The duct geometry and gas composition must be known to determine the flow rate using these sensors. Since the overall flow is inferred, irregularities in the flow pattern greatly reduce the accuracy of the measured value since it is assumed to be uniform. There are many other types of flow meters that can be used such as venturi meters, vortex shedder flow meters, and Coriolis flow meters. Any of these may be preferred depending on the application. Fluid flow can also be measured indirectly using chemical analysis and mass and energy balance calculations. Air flow measurements are commonly made this way since an oxygen analyzer is generally more accurate than a thermal mass flow meter. Some considerations to keep in mind in selecting flow meters are:
Figure 33.10 Thermal mass flow meter. The meter’s probe is inserted into a duct of known dimensions. Flowing gases pass through the sensing element at the bottom of the probe. The sensing element consists of two rods. The heat transfer between these rods is correlated to flow. An electrical transmitter and digital display is located in the head of this instrument.
© 2011 by Taylor and Francis Group, LLC
• Range of flow measurement • Possible particle deposition, fouling, or corrosion due to fluid • Accuracy • Reliability • Ease of calibration or modification 33.8.3 Temperature Measurement Thermal oxidizer temperatures are most commonly measured using ceramic sheathed thermocouples as seen in Figure 33.11. These thermocouples give reliable, reproducible results and can be used in most instances when they are placed properly. To minimize measurement errors, the thermocouple should not be positioned near luminous flames or cool surfaces. Suction pyrometers can be used to obtain more accurate measurements of the flue gas temperature. A suction pyrometer recesses the thermocouple in an extractive probe to shield it from radiation exchange with the rest of the vessel. Flue gas is rapidly educted into the extractive probe to maximize the convective heat transfer of the flue gases to the thermocouple (while minimizing radiative exchange with the rest of vessel). There are multiple types of thermocouples that can be used. The most common thermocouples used in combustion applications are type K, type S, and type B. The temperature rating for type S and type B are higher than type K but are significantly more expensive and not quite as accurate. The thermocouple sheathing material and length also needs to be selected to be compatible with a given test. In cases where the environment is too harsh for a thermocouple, the temperature can be measured optically
Figure 33.11 Ceramic sheathed thermocouple. A stainless steel type K thermocouple is sheathed with a ceramic thermowell. This is the most common instrument for measuring temperature in a thermal oxidizer.
702
Industrial Combustion Testing
Figure 33.12 Handheld radiometer. This radiometer is equipped with the HeNe laser used for sighting. The radiometer quantifies the temperature of the sight point. Note that the emissivity of the material at the sight point must be known.
using a pyrometer or radiometer. Additionally, an infrared thermograph can be used to map temperature. Handheld radiometers, as seen in Figure 33.12, are useful for measuring shell temperatures. Optical measurements usually require some knowledge of the optical properties (emissivity, reflectivity) of the medium being measured.
Figure 33.13 Slack tube manometer. The tubes on each side of the manometer are connected to pressure taps (or one is open to atmosphere). The difference in liquid columns is measured with a ruler positioned between them.
33.8.4 Pressure Measurement Pressure and flow measurements are very useful for the characterization and operation of test equipment. Pressures in fuel lines are generally several psi; whereas pressures in the combustion chamber and air duct may be on the order of inches of water column. The expected pressure range will determine what measuring device is best suited for the application. One of the simplest methods to measure pressure is with a water filled manometer when the pressure drop is less than 40″ w.c. Higher pressure drops can be measured but may be impractical because of the length of the manometer. When the pressure drop is more than 1″ w.c. a u-tube manometer, as seen in Figure 33.13, is well suited. When the pressure drop across a u-tube manometer, is less than 1″ w.c. an inclined manometer, as seen in Figure 33.14, is well suited. The only difference between a u-tube manometer and inclined manometer is the angle of the water column. The inclined manometer forces the liquid to travel a longer distance for a given pressure, which makes short water columns easier to measure. Handheld, battery-operated digital manometers can be similarly utilized. These manometers are very durable and very simple to operate but do not permit automatic data acquisition.
© 2011 by Taylor and Francis Group, LLC
Figure 33.14 Inclined manometer. The tubes on each side of the manometer are connected to pressure taps (or one is open to atmosphere). The difference in liquid columns is measured with a ruler calibrated for the inclination angle.
Higher pressures are commonly measured with Bourdon tube gauges as seen in Figure 33.15. Bourdon tube gauges can be designed to measure vacuum pressures or positive pressures. Bourdon tube gauges display the pressure by the displacement of a
703
Thermal Oxidizer Testing
minimize the response time. In most cases ¼″ tubing is adequate.
33.9 Conclusion
Figure 33.15 Bourdon tube pressure gauge. The pressure force exerted on the spring moves the needle on a calibrated dial displaying pressure.
needle. Typically, they do not provide electronic outputs. Bourdon tube gauges should be periodically removed and calibrated on a test stand to ensure their accuracy. Diaphragm pressure transmitters are most commonly used when an electrical signal is desired for data acquisition. These devices are also durable and reliable but require additional calibration and maintenance compared to a fluid manometer or spring gauge. Placement and orientation of the pressure taps need to be considered in the equipment design to allow for pressure measurements. The pressure taps should be positioned perpendicular to the flow of fluid, otherwise the velocity head of the fluid will alter the measurement. Placing ball valves between the pressure taps and measurement device is commonly done (especially when high pressures are involved) so that the pressure measurement device can be safely taken out of service for calibration or replacement. The pressure measurement device is preferably located near the pressure tap but is not essential. The diameter of the tubing connecting the pressure tap to the measurement device should be of adequate diameter to
© 2011 by Taylor and Francis Group, LLC
There are unlimited configurations of thermal oxidizers that can be tested. Each configuration and test objective will have its own requirements for equipment and measurements. Designing the equipment for maximum flexibility and modularity makes it possible to test the vast majority of configurations. Simulating input streams with simpler streams makes the test easier and usually safer to conduct. Equipping the test equipment with the proper instrumentation allows the test to be carefully controlled and provides accurate data for analysis. The thermal oxidizer test facility can be used for many objectives such as research and development, equipment characterization, customer demonstrations, and for training purposes. The information in this chapter is an introduction to the most common approaches to thermal oxidizer testing and hopefully gives insight and considerations for carrying out less common test methods.
References
1. Baukal, C. E., ed. The John Zink Combustion Handbook. Boca Raton, FL: CRC Press, 2001. 2. Baukal, C. E., ed. Industrial Burners Handbook. Boca Raton, FL: CRC Press, 2004. 3. Dellinger, B., Torres, J. L., Rubey, W. A., Hall, D. L., and Graham, J. L. “Determination of the Thermal Decomposition Properties of 20 Selected Hazardous Organic Compounds.” EPA-600/S2-84-138, United States Environmental Protection Agency, October, 1984. 4. Johnson, B., and McQuigg, K. “The Effects of Operating Conditions on Emissions From a Fume Incinerator.” Proceedings of the 1994 International Incineration Conference, Houston, TX, 1994.
34 Utility Boilers Giuseppe Toniato and Silvio Rudi Stella 34.1 Introduction.................................................................................................................................................................. 705 34.2 Boiler Efficiency........................................................................................................................................................... 706 34.2.1 Energy Losses Through the Chimney qsmoke............................................................................................. 707 34.2.1.1 Smoke Temperature and Excess of Air...................................................................................... 707 34.2.1.2 Condensation Index..................................................................................................................... 710 34.2.1.3 Other Parameters Affecting Losses at the Chimney............................................................... 711 34.2.2 Envelope Energy Losses qenvelope.................................................................................................................. 712 34.2.3 Auxiliary Electrical Consumption qaux. ..................................................................................................... 713 34.2.4 Boiler Efficiency in Field: Ambient Conditions and Transient Operations.......................................... 713 34.2.5 Boiler Efficiency and Emissions.................................................................................................................. 714 34.3 Utility Boilers Main Components.............................................................................................................................. 715 34.3.1 Burner.............................................................................................................................................................. 715 34.3.2 Combustion Chamber.................................................................................................................................. 715 34.3.3 Heat Exchangers............................................................................................................................................ 717 34.3.4 Boiler Controls............................................................................................................................................... 717 34.4 Utility Boiler Test Methods........................................................................................................................................ 718 34.4.1 Product Categories........................................................................................................................................ 718 34.4.1.1 Output Function........................................................................................................................... 718 34.4.1.2 Capacity and Configuration Storage.......................................................................................... 719 34.4.1.3 Fuel Configuration....................................................................................................................... 720 34.4.1.4 Boiler Flue Gas/Intake Air System............................................................................................ 720 34.4.1.5 Technology Boiler......................................................................................................................... 721 34.4.1.6 Power Class.................................................................................................................................... 721 34.4.1.7 Boiler/Burner Configuration...................................................................................................... 721 34.4.1.8 Boiler Water Temperature Control............................................................................................. 721 34.4.1.9 Burner Combustion Control....................................................................................................... 722 34.4.1.10 Installation..................................................................................................................................... 722 34.4.2 EN Product Standard.................................................................................................................................... 722 34.4.2.1 Introduction................................................................................................................................... 722 34.4.2.2 Classification of Gas Appliances by Type................................................................................. 722 34.4.3 Eco Design Preparatory Study: Boilers...................................................................................................... 722 34.4.3.1 Assessment Procedure................................................................................................................. 723 34.4.3.2 Test Requirement.......................................................................................................................... 723 34.4.3.3 Other Data Input........................................................................................................................... 724 34.4.3.4 Labeling......................................................................................................................................... 724 34.5 Appendix...................................................................................................................................................................... 724 References................................................................................................................................................................................. 728
34.1 Introduction A big energy saving and CO2 emission reduction could be reached by improving boiler efficiency. Boiler
efficiency is measured by industrial manufacturers in laboratories following standardized test procedures, but values reached in operation could differ a lot. The boiler efficiency situation in Europe is reported in the product studies of Europe Economics [1]: “There is a 705
© 2011 by Taylor and Francis Group, LLC
706
Industrial Combustion Testing
large gap between the theoretical efficiency and the real efficiency of boilers. The majority of boilers are oversized (200% → 400%). The part load of condensing boilers is assumed to be 38.8%, but in tests the real average part load is measured at about 9%. The (theoretical) efficiency of most existing boilers which are older than, e.g. 15 years, is probably 10–15% lower than the newest condensing boilers available.” To fix the boiler efficiency concept we have to image a volume around the boiler, measure all energies in and out, and relate energy output to energy input. This seems quite simple, but going deeper the situation is more and more complex. First of all boiler efficiency is often confused with combustion efficiency. In a heating boiler a fuel is first burnt, then the produced heat is transferred to a fluid (air or water). The first process controlled by the burner takes place in the boiler combustion chamber; here the produced heat starts to be transferred to the intermediate fluid through conduction, convective, and radiative heat process. Downstream the combustion chamber smoke enters a heat exchanger where a prevalent convection mechanism completes the heat transfer to the intermediate fluid. Two efficiency parameters can be easily identified: 1) the combustion efficiency that measures the combustion process or the ability of the burner to burn the fuel and 2) the efficiency of the heat transfer from smoke to the water or air. For evaluating heat transfer efficiency we have to consider that this process takes place both in the combustion chamber and downstream in the heat exchanger. Combustion efficiency depends upon burner and combustion chamber design and is related to unburned fuel and excess air in the exhaust. The combustion chamber design influences the fraction of heat that is transferred here to the intermediate fluid and the smoke temperature at the inlet of the heat exchanger, whose effectiveness is related to the temperature variations of the smoke passing through it. Both combustion efficiency and heat exchanger effectiveness contribute to boiler efficiency. Boiler efficiency definition and parameters affecting it are the argument of the first section of this chapter. The second section describes main components of a
utility boiler while standards relating boiler efficiency are treated in the last section. Several standards are available for different types of boilers defining test conditions under which boiler efficiency is measured in the laboratory. What matters here is to ensure that laboratory tests are representative of real working conditions. As mentioned at the beginning, when a boiler with a laboratorymeasured efficiency is set in operation, real behavior could differ a lot due to the real working conditions. Starting from this analysis current legislation in Europe is promoting new ratings for boiler efficiency taking into account among other things transient conditions, load sizes and heating system characteristics as it will be shown.
34.2 Boiler Efficiency Referring to Figure 34.1, boiler efficiency is defined as the ratio of the useful heat output to the energy input, hence: η=
In the previous formula it is assumed that fuel and reactant air are at the reference temperature of 25° and we neglect for the moment sensible reactants enthalpy. Actual standards in Europe still take as a reference the fuel low heating value and efficiency above 100% could be declared, but this situation is going to change and so
Smoke
Paux
q
Gas Combustion chamber
Heat exchanger
Figure 34.1 Energy flows in a utility boiler. Reactants and water are at a reference temperature of 25°C.
© 2011 by Taylor and Francis Group, LLC
qsmoke Water out
Water in
Air
(34.1)
where: HHV is fuel high heating value, f is fuel mass flow, m Paux is electrical consumption of auxiliaries like the burner fan or boiler pump, q is the heat transfer rate to the intermediate fluid, air, or water.
qenvelope
mf × HHV
q , f ⋅ HHV + Paux m
707
Utility Boilers
we refer to the high heating value. Similar consideration applies to auxiliary electrical consumption: it will be considered in future legislation and motivate the previous formula. The heat transfer rate q to intermediate fluid could be determined directly by measuring the useful heat output and the fuel flow rate, or it can be determined indirectly from the products of combustion. When boiler efficiency is measured in a laboratory, standards provide test schemes to measure useful heat output w is the w (Tfeed − Tret ). See Figure 34.1 where m as: q = m water flow rate and Tfeed and Tret are the feed/return temperatures of the heating system. This direct method is followed in laboratories. Here we refer to the indirect method that is useful to assess where losses arise and thus indicates how the efficiency may be increased. It can also be used in the field giving a good estimation of boiler efficiency. Based on an energy balance, q could be approximate by:
f ⋅ HHV − qsmoke − qenvelope . q=m
(34.3)
It appears that for a specific fuel (HHV) and a specific f ) boiler performances are effected by three fuel flow ( m contributions:
1. Heat transfer rate with the smoke: qsmoke 2. Heat transfer rate to the ambient through the envelope: qenvelope 3. Electrical power of auxiliaries: Paux This is summarized in Figure 34.2.
Heat losses through chimney
Auxiliaries electrical consumption
Boiler efficiency
Figure 34.2 Factors affecting boiler efficiency.
© 2011 by Taylor and Francis Group, LLC
Excess of air
Inlet air humidity
Smoke energy content Unburned hydrocarbons Condensation index
Inlet air and fuel temperatures
Figure 34.3 Variables affecting chimney boiler losses.
34.2.1 Energy Losses through the Chimney qsmoke Main loss affecting boiler efficiency is given by the energy escaping through the chimney: neglecting for the moment qenvelope and auxiliary consumption from the previous equation we get:
(34.2)
Substituting such value of q in the previous equation it results: f ⋅ HHV − qsmoke − qenvelope m , η= f × HHV + Paux m
Smoke temperature
Heat losses through envelope
η = 1−
qsmokes , f ⋅ HHV m
(34.4)
Smoke energy content is affected in turn by several parameters: smoke temperature and excess of air, condensation index, air humidity content, air and fuel inlet temperatures and unburned hydrocarbons. Each of these contributions to chimney boiler losses is shown in Figure 34.3. 34.2.1.1 Smoke Temperature and Excess of Air We can calculate energy escaping from the chimney knowing reactant properties and smoke temperatures. Detailed analysis is outside the scope of this contribution, the reader should refer to specific texts [2–4]. With a simple Excel program and some approximations, we can verify effect of smoke temperature and excess of air on boiler efficiency. Results are reported in Figure 34.4 in case of methane combustion. A 1 m3 of methane requires roughly 11 m3 of air for a satisfactory combustion producing 1 m3 of CO2 and 2 m3 of vapor water apart oxygen and nitrogen (all volumes are referred to standard conditions, which means 25°C and 101 kPa pressure). In terms of mass this means that the combustion of 0.7 kg of methane produces 1.96 kg of CO2 and 1.61 kg of water. Collecting results we can plot boiler efficiency as a function of smoke temperature and excess of air. We get a family of lines like in Figure 34.5. A similar diagram in terms of Btu could be found on the Internet [5].
708
Industrial Combustion Testing
Input data (Nm3/h)
Output data
1.000 0.25 25 25 170
Vmethane methane volume flow Lamda excess of air Tmetano methane temperature (°C) Taria air temperature (°C) Tsmoke smoke temperature (°C)
Air composition %volume %weight Gas N2 79.05 76.76 O2 20.95 23.24 Methane composition Element n° atoms %weight
583
Water heat of vaporization (kcal/kg)
LHV low heating value (kcal/Nm3) HHV high heating value (kcal/Nm3) Qmethane methane sensible heat (kcal/h) Qair air sensible heat (kcal/h) Qsmoke smoke sensible heat (kcal/h) Quseful heat to water (kcal/h) Qinlow LHV*Vmethane+Qmethane+Qair (kcal/h)
8555 9492 0 0 615 7939 8555
C
1
74.9
Qinhigh HHV*Vmethane+Qmethane+Qair (kcal/h)
9492
H
4
25.1
Etaboiler =Quseful/Qinhigh (%)
83,6
CO2
H2O
Complete combustion Units Nm3/h %volume
– kg/h %weight
– kcal/kg kcal/h kcal/h
Reactants O2 N2
CH4
1.00 7.7 – 0.72 4.5 – 1115 798 0
2.50 19.3 – 3.57 22.2 – 0 0 0
Air 11.93 92.3 – 15.36 95.5 – – 0 0
9.43 72.9 – 11.79 73.3 – 0 0 0
Variable Volume flow
Mass flow
Heat of formation Sensible heat
1.00 7.7 9.1 1.96 12.2 13.6 2137 4196 64
2.00 15.5 – 1.61 10.0 – 3208 5157 109
O2
0.50 3.9 4.6 0.71 4.4 4.9 0 0 22
Smoke N2
9.43 72.9 86.3 11.79 73.3 81.5 0 0 419
CO – – – – – – – – –
NO – – – – – – – – –
Total 12.93 100 100 16.07 100 100 – 9353 615
Units Nm3/h %vol. wet %vol. dry
kg/h %weight wet %weight dry kcal/kg kcal/h kcal/h
*Heating values and heat of formation are calculated at 25°C. Sensible heat is calculated from the gas temperature and 25°C. Figure 34.4 Example of methane reaction.
0.9
where hsmoke is the smoke enthalpy and we neglect that mass balance is completed by ash. air − m airstech / If we introduce the excess of air λ = m mairstech we get mair = mairstech × (1 + λ). Also, introducing fuel / m airstech the stoichiometric fuel/air ratio: fstech = m and substituting in the previous expression we have:
Methane
0.88 0.86
Excess of air
0.84 η
Ex = 0
0.82 0.8
Ex = 0.5
1+ λ smoke ⋅ hsmoke = m f ⋅ qsmoke = m + 1 ⋅ hsmoke , (34.6) fstech
Ex = 1
0.78 0.76 350
395
440 485 Tsmokes (K)
530
575
Figure 34.5 Boiler efficiency as a function of exhaust gas temperature and excess of air.
Neglecting any other contributions, the linear correlation between boiler efficiency and smoke tempe rature parameterized on excess of air could be easily demonstrated. In fact energy escaping with the smoke is given by: m smoke ⋅ hsmoke = m f ⋅ air + 1 ⋅ hsmoke , qsmoke = m f m
© 2011 by Taylor and Francis Group, LLC
(34.5)
and for efficiency: η = 1−
1+ λ qsmoke 1 = 1− × + 1 × hsmoke . (34.7) f × HHV m HHV fstech
For a specific fuel HHV and fstech are fixed while smoke enthalpies are function of smoke temperature. The previous equation has the general form:
η = 1 − g(λ) ⋅ f (Tsmoke ).
(34.8)
Inspection of this equation shows that boiler efficiency depends on the boiler capacity to release smoke to the
709
Utility Boilers
12.00
CO2%
11.00
10.00
9.00
8.00 0.00
0.05
0.10
0.15
0.20 Excess of air
0.25
0.30
0.35
0.40
Figure 34.6 Relation between CO2% emissions and excess of air for methane.
atmosphere at low temperatures and work with low excess of air. The highest efficiency would be reached in stechiometric condition, but in boiler practice it is always required to work in excess of air to avoid the risk of CO formation, which is poisonous at very low concentrations. Excess of air ensures that there is always enough air/oxygen to guarantee a complete combustion. The excess of air is actually intended to compensate for several factors: the inhomogeneous mixing of air/fuel (in the case of oil application this is connected to the fuel atomizing process), the fluctuations in atmospheric pressure of the incoming air, in relative humidity of air, and in fuel supply and fuel quality. Wind influence on a chimney has to be taken into account (this is a critical factor especially for atmospheric burners). For instance, an excess of air 0.2 means that 20% of extra air is added with respect of the stechiometrically needed volume. In general in the boiler manual, instead of excess of air the CO2% content the in the smoke is reported. Between excess of air and CO2% for a specific fuel there is a direct correlation as shown in Figure 34.6. Standards require a minimum boiler efficiency measured at a specific excess of air. For example, Figure 34.7 gives excess of air limits requested by EN 303. Boiler manufacturers suggest the correct burner settings for the correct excess of air, but it is not unusual to find boilers in the field that are set up with higher values compared to the suggested ones. In general, installers increase the excess of air to avoid combustion problems and forgetting about possible energy saving. Excess of air (or CO2% setting) suggested by boiler manufacturers depends only upon the type of burner applied on the boiler. Referring to the gas appliance, the most common cases are atmospheric burners as depicted in Figure 34.8.
© 2011 by Taylor and Francis Group, LLC
b)
1.3
1.25
1.2
1.2
1.2
1.15 1.1 101
2
3
4 5 6
8 102
2
3
4 5 6
8 103 7 9 a)
7 9
Figure 34.7 Minimum excess of air required (from EN 303-3): on the x-axis: nominal thermal power in kW, on y-axis air factor (excess of air + 1).
Methane 85% excess of air 6% CO2 NOx < 200 mgr/kWh CO < 100 mgr/kWh Primary air
Secondary air
Figure 34.8 Atmospheric burners.
In these burners gas flow depends on the underpressure created in a venturi system by the air flow. Fuel and air control is not good and a high excess of air is required. The 6% of CO2 contents means that less than double air compared to stechiometric condition is required. Figure 34.8 gives a rough indication of CO and NOx level reached with this technology. Atmospheric burners are premixed burners because air and fuel are mixed before burning, but “premixed burners” for
710
Industrial Combustion Testing
utility boilers are often used to indicate a specific premixed technology where the mixture of gas and air (in general mixed before the fan) is burnt on the surface of dedicated combustion heads as shown in Figure 34.9. In this case it is better to refer to premixed surface burners. Such technology is used for condensing application and is becoming more and more popular in gas utility boilers. Different solutions have been used for gas surface combustion, some of these are presented in this book. Values of CO2% that are reported in Figure 34.9 are typical for the specific technology. It is interesting to compare such performances with a diffusive gas flame burner used in utility boilers as shown in Figure 34.10. It appears immediately that diffusive gas flame burners could work with a lower level of excess of air (higher level of CO2%) compared to a premixed burner and also lower level of CO, but higher NOx emissions (even if the comparison had to be performed at the same level of excess of air). The reasons why surface premixed technology like the one in Figure 34.9 is applied in the condensing application are related to three main aspects: the lower level of NOx emissions that it can reach, the possibility of reducing combustion chamber size compared to a diffusive application, and the opportunity to reduce chimney noise emissions (see Chapter 25). On the
other side diffusive flame burners are easier to stabilize, are more reliable, and so they are preferred in case of high power applications and in all industrial processes. 34.2.1.2 Condensation Index When smoke temperature goes under a specific value related to the fuel used, part of the vapor in the smoke is condensed in water. In such a case, Equation 34.4 is modified in this way:
η = 1−
smoke hsmoke HHV − LHV m + ⋅ α, f ⋅ HHV m HHV
where LHV is the low heating value and α is the condensation index and is given by α = VH2O cond / VH2O theor , which means by the ratio between the measured volume of water condensed (for each cubic meter or kg of burned fuel) and theoretical condensed water for the specific fuel. The difference between high and low heating value (HHV and LHV) of the fuel is the latent condensation heat, which means the heat contained in the water vapor from combustion when condensing. Contribution of the second term in the previous formula
Airtight aspiration circuit Methane 32% excess of air 8.5% CO2 Combustion head with holed surface Air
Flame
NOx < 60 mgr/kWh CO < 80 mgr/kWh
Gas
Figure 34.9 Premixed burners (mixing before the fan).
Mixing Aria Gas Aria
Figure 34.10 Forced draught burners.
© 2011 by Taylor and Francis Group, LLC
(34.9)
Methane 20% excess of air 9.5% CO2 NOx < 80 mgr/kWh CO < 20 mgr/kWh
711
Utility Boilers
1
Dew point temp (°C)
Methane
η
0.95 0.9 0.85
Excess of air = 0 Excess of air = 1
0.8 300
350
400
450
Tsmoke (K)
1
2
3
4
5
6
7 8 9 CO2 (%)
10 11 12 13 14 15
Natural gas (95% CH4) 500
550
600
Figure 34.11 Effect of condensation on boiler efficiency for two different excess of air.
makes the correlation between smoke temperature and boiler efficiency (for a specific excess of air) not more linear. The diagram in this case is modified as indicated in Figure 34.11. In the case of methane combustion around 1.61 kg of water vapor is produced per m³ of methane (see Figure 34.4). The specific latent condensation heat of water is 2.418 MJ/kg at 25°C, so per m³ of methane 4.00 MJ of condensation heat is available. Compared to the HHV of methane of 39.8 MJ/m³, this is 10.04%. Compared to the LHV it adds an extra 11.05%. As natural gas consists mostly of methane, the same numbers apply roughly to natural gas. For other fuels, the stechiometric combustion equations are different and therefore the water vapor and the maximum amount of latent heat are different. For oil-fired boilers this is around 6% and for propane it is 8–9%. Smoke temperature at the chimney distinguishes the noncondensing boilers from the condensing ones. In the past most utility boilers were designed to avoid condensation, while today the efficiency requirements are forcing the market to condensing solutions. In a noncondensing boiler the flue gases leave the boiler at a temperature of somewhere between 100°C and 200°C still in the form of water vapor. All the latent heat of the condensation process is lost for the boiler. The EN standards speak of a dedicated “condensing boiler” at flue gas temperatures lower than 80°C. Condensing relates to the fact that the water vapor in the flue gas comes into contact with a cold surface of the heat exchanger and than turns into liquid, releasing the latent heat of condensation. This condensation of air with a 100% relative humidity takes place at a temperature level that is known as the dew point. This dew point also depends on other parameters, but in general it can be said that condensing starts at a surface temperature of the boiler heat
© 2011 by Taylor and Francis Group, LLC
60 55 50 45 40 35 30 25 20
Gasoil (EL)
Figure 34.12 Dew point for natural gas and oil.
exchanger of just below 57°C for gas and 46°C for oil. At a boiler return temperature of 30°C some 70–80% of all latent heat is recovered. At 35°C, the boiler return temperature around 50% is recovered. At a higher excess of air the dew point will even be lower; at λ = 0.25 the CO2 content decreases and the dew point is approximately 53ºC for gas and 44ºC for oil. From the Figure 34.12, it appears as the dew point for oil is lower compared to gas: condensation with oil is more difficult to reach (and also the energy saving is lower due to the lower water content of oil). Also the sulfur content in fuel impacts on the dew point. Between standard boilers and condensing boilers there are the low temperature boilers (better defined in the third section of this chapter). In a low temperature boiler heat transfer in critical zones is reduced, so allowing lower feed and return water temperatures without the permanent need of a mixing valve. Their main advantage compared to an atmospheric boiler are in part load, where a weather controlled low temperature boiler can modulate with low flue gas temperatures. For this reason the low temperature boiler efficiency is 1–6% better than that of a standard boiler at 30% part load. They do not perform well as condensing boilers, but they represent a good compromise in terms of costs. Noncondensing, condensing, and low temperature boilers require a different design criteria, different materials, different controls, and circuit solutions. 34.2.1.3 Other Parameters Affecting Losses at the Chimney Others parameters affecting losses at the chimney are the air humidity content, the air and fuel inlet temperatures, and the unburned hydrocarbons. Vapor content of air influences chimney losses in two ways: it takes away part of the heat given by the fuel and modifies the dew point temperature as already seen. Not only does the combustion reaction produce water vapor as one of the outputs, but also a—relatively small—fraction of the water vapor
712
in the flue gases comes from the humidity of the combustion air input. Following the example in Figure 34.4 at 70% of relative humidity (as prescribed by EN standard) the water content of air is 14.1 gr/kg at ambient temperature of 25°C.* The density of air at 25°C is 1.2, and 10.98 m³ of air that goes into the combustion resulting in around 0.185 liters of water that needs to be added. When calculating energy balance around a boiler it is important also to consider fuel and air inlet temperatures. In general fuel temperature can be neglected in boiler efficiency analysis, at least it is inside standard conditions. This temperature is more important for the performances of the fuel line and atomization process. The situation is different for inlet air: its temperature can have an important impact on boiler efficiency due to the flow mass involved (compare Figure 34.4). In case the inlet air is preheated by exhaust smoke, the boiler efficiency expression doesn’t change with respect to Equation 34.1. In case the air is preheated by an external source, sensible enthalpy of air has to be considered in both the denominator and the numerator. During combustion, not only CO2 and vapor is produced but also small quantities of pollutants like carbon monoxide (CO) and nitrogen oxide (NOx). Their formation in flame has to be taken into account in the energy balance, but in general (at least until we talk about methane or other typical fuels used in utility boilers) they can be neglected for the boiler efficiency calculation because their formation subtract little energy to the overall balance, but of course they are important from the environmental point of view and this is why they are subjected to specific limits. Concentration of these pollutants depends on several factors like type of fuel (going from the “clean” methane to the “dirty” heavy oil), boiler design, and burner technologies. We have already seen that CO and NOx levels could change switching from a premixed burner to a diffusive burner, but combustion chamber dimensions can also influence such emissions. Boiler installation instructions specify the correct combustion settings, but these adjustments refer to steadystate working conditions. During transient operating conditions, pollutant emissions are higher. Standards like EN 676 for gas burners and EN 267 for oil burners ensure that during burner startup pressure amplitude reduces inside a specific interval after a while, but does not give any emission limits during such transient cycles.† A lot of studies have shown instead that a boiler could emit much more during transient operation. More
Industrial Combustion Testing
information on emissions from oil-and gas fired appliances can be found in [6]. 34.2.2 Envelope Energy Losses qenvelope If the main energy losses are through the chimney, a small percentage of energy is lost through the boiler envelope also. In new boilers such losses are very low. This is due not only to new design solutions and better materials used in insulation, but also in boiler control strategy. In old boilers water temperature was maintained at the fixed value required by the heating system. If the water feed temperature was adjusted to 80°C, temperature in the boiler room could be very high, so high that clothes could be dried there. With the more recent design starting from the low temperature boilers, the water content in the boiler is not fixed but is controlled based on different variables like heat load or external temperature. It is interesting to note that standard EN 303 gives an indication for the envelope losses as in Figure 34.13. Energy losses through the envelope depend of course from the temperature of the room where the boiler is located. This consideration is important not only for the boiler efficiency itself, but it has to be taken into (a)
0.04
0.03
0.02
0.01 0.005 0
101
4
5
6
7
103
8
(b)
9 * EN 303 refers to 20° C as a reference standard temperature but we prefer to refer to 25°C to be coherent with the example and with thermodinamic convention. † For example EN267 imposes that unburned idrocarbons shall not exceed 10 ppm except during the first 20 second after release of the oil.
© 2011 by Taylor and Francis Group, LLC
Figure 34.13 Maximum envelope losses for boiler > 400 kW (from EN 303-2). On the x-axis: nominal thermal power in kW, on y-axis envelope losses in percentage. In case of a combi boiler, values can be improved by 0.005%.
713
Utility Boilers
34.2.3 Auxiliary Electrical Consumption qaux The third contribution to boiler efficiency is given by the electrical consumption of auxiliaries like the burner fan and water pump. It is worth noting that such contribution is neglected in the present regulation in Europe even if such regulation is going to change. Regarding fans in recent years, technology switched from asynchronous and DC motors to brushless motors. In this way electrical consumption has been reduced, bulky transformers are not required anymore, and motor speed can be easily changed, controlling in this way the combustion airflow in place of the traditional butterfly and reducing air circuit losses. Energy saving due to variable speed solution depends a lot from thermal loads: if at maximum load performances are similar to fixed speed solutions, the main benefit appears at reduced loads. The economical threshold between brushless and asynchronous motors is today around 150 watts: under this limit brushless technology is competitive in terms of costs to asynchronous motors; above such limits asynchronous motors are used and the motor speed could be changed by inverters. A similar technology trend took place for the water circulating pump. Also in this case the possibility to regulate the circuit water flow by motor speed instead of a classical 3-way valve, introduces a big opportunity in energy saving. It is impressive how the different electrical consumption that the boiler of the same category could have in different working conditions as shown in Figure 34.14 [7]. In gas boilers electricity used for electrically powered gas valves and gas igniters can be neglected. In oil boilers two other important electrical loads have to be taken into account instead: electrical resistance used to preheat the oil to the right viscosity and the oil pump consumption for pressurizing and atomizing the fuel.
© 2011 by Taylor and Francis Group, LLC
3.0 Electrical use (kWh)
account when the energy balance of a building or a flat is performed. Some standards (prEN 15316-4-1, Table c3) provide a logarithmic formula for the envelope losses: Penv = A + B × log Pn where Pn is the nominal boiler power in kW and A and B are specific coefficients depending on generator insulation type. In case of a combination boiler, the presence of a storage vessel for sanitary hot water introduces some penalties in terms of efficiency. For example for an 80 liter at 65° ( > 100 mm insulation) 55 to 60 watts are lost, which means 500 kWh/year = 5% efficiency loss/year. On the other hand, without a storage vessel boiler, efficiency would be affected by the number of burner cycles per year.
2.5
Boiler 1 Boiler 2
2.0 1.5 1.0 0.5 0
0
20
40
60 80 100 Heat supplied (kWh)
120
140
160
Figure 34.14 Daily electrical usage trends of two similar boiler installations (SEDBUK A-rated systems, with quoted seasonal efficiencies of over 90%). It can be seen that for a given level of heat supplied, boiler 1 uses around two and a half times as much electricity as boiler 2. (From “M-CHP Interim Report,” Carbon Trust: CTC726, available at http://www.carbontrust.co.uk. With permission.)
34.2.4 Boiler Efficiency in Field: Ambient Conditions and Transient Operations When a boiler is set up in field the mentioned parameters like chimney losses, envelope losses, and electrical consumption affect boiler performances in different ways based on ambient conditions. Figure 34.15 shows the components that interact with the boiler: the hydraulic piping, the heating elements, and the control system. A correct matching between boiler and these systems plays an important role in the building’s energy performances and the real efficiency of boilers is closely linked to them. It is reported [1] that “less than 50% of the overall efficiency is due to the quality of the boiler.” In a utility boiler smokes exiting the combustion chamber exchange heat with the heating system water. For a specific thermal power, temperature and flow of the smoke entering the heat exchanger are known. In this case smoke temperature at the chimney depends only from the flow and temperature of the return water. Figures 34.5 and 34.11 can be changed reporting on the x axis water return temperature instead of the exhaust gas temperature. Vapor condensation is possible only if the return water temperature is lower than the dew point temperature. This is the case of a floor-heating system where feed water temperature is 50° and return could be 30° ensuring, in this case, complete condensation of vapor in the smoke. It is interesting to compare performances of a condensing boiler with a different return water temperature. Figure 34.16 compares the efficiency of a wall-hung boiler for different return temperatures. Boiler efficiency can be improved also by using a modulating burner (better control of thermal power and smoke flow) and/or a modulating pump in the water
714
Industrial Combustion Testing
External temperature sensor
Thermal regulator
3-way valve
Boiler
Pump
Heating system
Figure 34.15 Boiler and heating system.
Boiler efficiency at 40% nominal load circuit at two water circuit temperatures 97.0 92.0
%
92.0 87.0 82.0
82.0
86.7
80.2 77.9
77.0
83.1
85.1
77.9
Atmospheric < 1990
New atmospheric
Boiler efficiency at 40% nominal load/feed and return water temperature 80°/60°
Low temperature boiler
Condensing premix
Boiler efficiency at 40% nominal load/feed and return water temperature 60°/40°
Figure 34.16 24 kW wall-hung boiler efficiency for different return temperatures.
flow systems. In Europe new standards based on the EuP Directive (see next paragraph) are proposing a new efficiency-labeling scheme considering transient operation of the boiler and its performances in the heating system. For example, in this new proposal boiler efficiency label changes if the heating system is equipped with a modulating burner or a modulating pump. Another aspect that is taken into account is the transient operation of the boiler. During transient operation, apart from losses due to pollutant emissions already mentioned, a new type of loss has to be added to the energy balance of the boiler. Some of these are the pre-purge losses during startup, the draft losses through the chimney, and after-purge losses during shut off. For safety reasons burners preventilate the combustion chamber to expel preexisting unburned fuel. This phase could take several seconds and this means energy losses (pre-purge losses). If a butterfly doesn’t close the air inlet circuit, during shutoff air enters the combustion chamber and takes away the heat through the chimney (draft losses). It is clear that the more a burner is cycling on and off, the higher are the losses. For example, in the case of an instantaneous boiler that switches at every draw-off, the
© 2011 by Taylor and Francis Group, LLC
number of cycles could be very high (e.g., in the range of 40–50.000 cycles/year). In the past, installers used oversize boilers to avoid customer complaints and this caused a continuous cycling of the boiler. This has been considered in a new label proposal trying to correlate boiler efficiency labeling to the size of the loads. Typical loads are defined based on house dimension, its geographical location, and its age. A boiler labeled best in class for a specific load could be rated in a lower class if it is oversized for that specific load. 34.2.5 Boiler Efficiency and Emissions Boiler efficiency improvements have to be achieved by trying to reduce emissions. Emissions of air-pollutants from combustion in gas and oil utility boilers are nitrogen oxides (NOx), carbon monoxide (CO), and methane (CH4). In oil-fired boilers we also have to consider the presence of sulfur oxides (SOx), volatile organic compounds (CxHy), and soot (particulate matter, PM). Also carbon dioxide (CO2) has to be considered an air pollutant for its global warning potential. While designing a combustion head one tries to reduce temperature flame to avoid formation of thermal NOx, but in this way one runs the risk of increasing CO emissions. Such pollutants go in opposite directions while increasing the excess of air. In recent years huge improvements have been performed in reducing NOx without impacting a high level of CO. What are called low NOx burners are based on specific technologies whose aim is to reduce nitrogen oxides emissions. In the burner design several strategies could be used, the most important are the fuel and air staging and the recirculation of combustion exhausted gases. In gas technologies, premixed surface burners reduce NOx levels to a value not reachable with low NOx diffusive flames, but this is at the expense of higher levels of CO and excess of air. The two main technologies that promise a future reduction in NOx and CO values are the catalytic combustion [8] and the flameless combustion.
715
Utility Boilers
34.3.1 Burner
Table 34.1 Emissions Levels for CO and NOx Class
NOx Emissions in mg/kWh
For gas (EN 676) 1 70 2 120 3 80 For heating oil (EN 267) 1 250 2 185 3 120
CO Emissions in mg/kWh < 100 < 100 < 100 110 110 60
Table 34.1 shows the present limits required for CO and NOx for gas and oil applications today that are valid in Europe (the NOx content of the combustion products is expressed under the reference conditions: ambient temperature: 20°C; relative humidity: 70%). New standards are proposing the limit of 35 mg/kWh for gas products with less than 50% renewable input (otherwise 70 mg/kWh) and 70 mg/kWh for oil (105 in the case of < 50% share). It is interesting to compare such targets to similar targets in automotive fields. Euro 6 (expected in 2014) will require 133 mg/kWh for diesel and 100 mg/kWh for gasoline.* Regarding soot emissions, present standards require a test on smoke numbers for heating oil application [9], but these measurements can roughly filter emissions higher than PM10. For solid fuel applications, EN 303-5 defines limits in terms of weight for volatile organic compounds and dust. Condensing technology is important in utility boilers not only for improving boiler efficiency, but also for reducing soot emissions. It has been demonstrated that in a gas appliance soot emissions can be reduced to half by the use of a condensing boiler [10].
The burner can be clearly identified in oil boilers or in big boilers, while in low power gas appliance the components of the burner are integrated in the boiler itself. For diffusive burners applied in floor standing boilers, technology improved from “yellow” flame burners to “blue” flame burners with a low level of NOx emissions. In some floor standing applications diffusive (blue) flame burners have been replaced by the premixed surface burners. Such technology also replaced the atmospheric burners in wall-hang boilers. Such evolution took place in parallel with the spreading of condensing applications. Premixed surface technology requires smaller combustion chambers as will be shown next and it has lower noise emissions at the stack compared to a diffusive head, due to the lower combustion turbulence on the condition that no combustion instabilities arise. Oil applications are based on pressure atomization technology even if small power application solutions are available based on oil evaporation by heat. Electronic control of the fuel/air ratio previews an actuator to move a butterfly for controlling the air flow and an actuator to move the oil pressure regulator. For gas, two actuators control two butterflies, one for gas and one for air, even if the gas pneumatic valves (whose working principle is based on a pneumatic feedback) are widely used. More advanced burners control air flow changing the fan speed, reducing the pressure drop in the hydraulic pipes and valves. Despite this, a butterfly valve is often necessary to control startup operation, especially for high power applications. The spread of biofuels has increased the applications of the double fuel burners, where a traditional fuel (for example gas) is used as a backup in case of a lack of the renewable resource. Use of bio-oil and biogas increased reliability problems on some specific burner components like pumps and valves. The introduction of a microprocessor in burner control introduced several new functions, starting from diagnosis to fuel consumption indication. 34.3.2 Combustion Chamber
34.3 Utility Boilers Main Components In this section we analyze some boiler components trying to underline main improvements that took place in recent years in heating boilers (at least in Europe). We focus on the burner, combustion chamber, exchangers, and controlling systems.
* Euro 6 = 0.08 g NOx/km for diesel converted at 0.6 kWh/km, based on average fuel consumption of 6.5 ltr./100 km, 36.4 MJ=10 kWh/ltr diesel.
© 2011 by Taylor and Francis Group, LLC
It is important to distinguish combustion chambers designed for diffusive flames (oil or gas appliances) from the chamber designed for premixed flames, in particular surface premixed flame. Sizing the combustion chamber for a diffusive flame burner, one can get some indication from Standards and from some empirical formula. EN 676 for boiler equipped with forced draught gas burners and EN 267 equivalent for oil burners give these indications for the length of the combustion testing tubes:
l[m] = 0.23
Q[ kW ] 10
l = 0.25 × Q[ kg/h].
(34.10)
716
Industrial Combustion Testing
b)
25.000
Nominal thermal power [kW]
3 2 1.5
a1
1 0.8
a2
0.6 b
0.4
2
3
4 5 6
8 102 7 9
2
3
4 5 6
8 103 7 9 a)
Figure 34.17 Combustion chamber minimum dimensions from EN 303-3. On the x-axis nominal thermal power in kW, on the y-axis combustion chamber diameter or length in m. Curve a1 is combustion chamber length in case smoke exits from the back side of the combustion chamber, while curve a2 applies in case smoke inverts the path to the chimney inside the combustion chamber and exits from the front side. Lengths are measured from the burner head where the flame is attached to the bottom of the combustion chamber. Curve b is the combustion chamber diameter (or equivalent diameter in case of a noncircular combustion chamber).
The same standards provide diagrams with standardized diameters ϕ. EN 303-2 (for oil boilers) provides minimum dimensions for the combustion chamber as shown in Figure 34.17. Equations for these curves are:
l[mm] = 101.828+ 0.473 log Q[ kW ] Φ[mm] = 101.923+ 0.291log Q[ kW ].
(34.11)
Considering that: 100 , 473 log Q = 10log Q0 , 473 ≅ Q we find again the exponent 0.5 suggested by EN 676. For a diameter the dependency is based on: 3 Q ≅ 100 ,291 log Q . Other empirical formulas that could be used are: l = 0.25 × Q[ kg/h]
Φ = 0.13 × 3 Q[ kg/h]
l 1 1 = : . φ 4 6
(34.12)
Equations suggested by the EN 676 standard is compared with a database of boilers. Figures 34.18 and 34.19 show the results. From these diagrams it appears that while the diameter equation seems to fit the diameter database, for the length, dependence from a cubic radix seems more appropriate. What is important in designing combustion chambers for a diffusive application is to avoid impingement
© 2011 by Taylor and Francis Group, LLC
y = 6E–08x3,6172
15.000
10.000
5.000
0
0
200
400
600
800
1.000
1.200
1.400
1.600
1.800
Combustion chamber diameter (mm)
Figure 34.18 Combustion chamber diameter versus nominal thermal power for 74 utility boilers for gas and oil application from different suppliers. The curve and the equation refers to EN 676 suggested diameter.
25.000
Nominal thermal power [kW]
0.2 101
20.000
y = 0,0002x1,9894
20.000 15.000 10.000 5.000 0
0
2.000
4.000
6.000
8.000
10.000
12.000
Combustion chamber length (mm)
Figure 34.19 Combustion chamber length versus nominal thermal power for 74 utility boilers for gas and oil application from different suppliers. The red curve and the equation refers to EN 676 suggest length.
of the flame on the back of the chamber or in general make the chamber long enough to avoid a high value of CO. During a burning process, fuel is first transformed in CO: if the flame is suddenly cooled by the wall of the boiler, CO doesn’t complete its transformation to CO2 and goes to the chimney. Regarding the chamber diameter, the design criteria is connected to the smoke pathway to the chimney: the smoke could enter into the heat exchanger from the back or it can invert the flow and enter from the front (inverse combustion chamber). Switching to a premixed burner it should be noted immediately that the heat exchange achieved in the combustion chamber by a surface premixed flame is generally different from the one achieved by a diffusive flame. In resizing the boiler the designer can reduce drastically the dimensions of the combustion chamber as presented in a previous chapter. The
717
Utility Boilers
c ombustion chamber of a surface premixed flame can be just a little longer than the combustion head and have a diameter that is about twice the diameter of the head. The radiative performances of such technologies are limited to a specific power range (generally from 20 to 30% of nominal load). To enlarge such a range it is necessary to use surface ceramic solutions or porous matrix combustion. In general it is very difficult to say the percentage of radiative contribution compared to the convective contribution to the total heat exchange in the combustion chamber. Such contribution depends upon burner technologies and changes with power output and excess of air. In general at lower heat output of a modulating burner (for a specific excess of air) the convection heat transfer decreases, while the radiation heat transfer increases. On the radiation mechanism in the combustion chamber, a classical test is reported [11]. 34.3.3 Heat Exchangers Two important trends took place in the last two decades regarding utility boilers heat exchangers design: • The reevaluating and redesign of the standard cast iron heat exchange used in the past for the floor standing boiler applications. • The integration of the sanitary hot water heat exchanger with the boiler heat exchanger. Making a selection for the heat exchanger material, it is important to compare properties of different materials (see Table 34.2). A simple optimization model shows that the best compromise for the heating application is aluminum. For condensing applications, two technologies are today still competing on the market: the aluminum and the iron steel. Aluminum heat exchangers can be produced by die-casting or can be extruded as tubes that can be shaped in different manners. Iron steel is used in tube shapes; in this case the lower thickness can be used and compensate the lower thermal performances compared to aluminum.
Regarding shape and design three main families can be identified: • heat exchangers based on tubes (finned and un-finned) • heat exchangers based on plates • heat exchanger based on integrated castings (aluminum, cast iron) The traditional type of heat exchanger based on an unfinned tube is the “Shell & Tube” heat exchanger, in which a tube arrangement is placed within a shell and is widely used for high power application. For smaller lightweight appliances other configurations are used, where the tube part consists of one long spiral formed tube, which is placed in a shell. In these applications the tube is finned to increase heat exchange. With die casting it is possible to integrate all necessary connections into the castings, reducing the number of components and assembly times, and the heat exchanger design can be further optimized for radiative heat transfer. Plate heat exchangers are instead considered the most compact and cost-efficient solution for sanitary hot water production. We can identify three types of heat exchangers in a utility boiler: primary, secondary, and tertiary heat exchanger. The first one is immediately after the combustion chamber or sometimes is a part of it. In general it is designed for high temperatures but it doesn’t work in condensation. This takes place in the second heat exchanger. In wall-hang boilers this is little more than an extension of the primary heat exchanger. The condensing heat exchanger has to be designed to allow condense to flow down and evacuate naturally from the combustion chamber. This requirement introduces specific design issues in the secondary (but also on primary if integrated) heat exchanger. The tertiary heat exchange doesn’t exchange heat with the water but with the incoming air to preheat it. This third heat exchanger works with low temperature smoke and that is why it can be designed in some plastic alloy. 34.3.4 Boiler Controls
Table 34.2 Different Thermal Properties for Materials Used in Utility Boilers Heat Exchangers Material Stainless steel Cast iron Aluminum Copper
Thermal Conductivity [W/mK] 27 60 237 390
© 2011 by Taylor and Francis Group, LLC
Thermal Diffusivity [m2/sec] 7,4 × 10−6 1,5 × 10−5 8,1 × 10−5 8,7 × 10−5
Boiler controls have seen the most important improvements in recent years. In the past boiler controls were limited to adjust feed water temperatures and monitor flame operation by the implementation of specific safety sensors like flame, temperature and pressure sensors. Today boiler controls integrate more and more functions: three-way valve, pumps, external heat storage, solar panel controls, and so on. In more advanced utility boilers, thermal power is modulated through fan speed modulation. Air–fuel ratio is controlled by an ionization
718
Industrial Combustion Testing
flame signal even if a more advanced solution with an O2 sensor on the chimney is coming on. For low power oil applications, as used in residential appliances, it is still too expensive to modulate heat power. This is why the technology improved just from a single stage burner to double stage burners, where power is modulated only on two levels. The air-fuel control has no closed loop feedback except for specific applications where O2 or CO sensors on the chimney are used. Regarding the CO sensor, it is interesting to note that there isn’t any univocal correlation between CO emissions and excess of air. In this case, closed loop control reduces excess of air until CO starts to increase and then comes back a few steps, controlling combustion on the borders of the safety limit.
Table 34.3 Utility Boiler Standards • NPR-CEN/TR 1749:2006 Classification of gas appliances by type • EN 297 Atmospheric gas boilers without fan, < 70 kW • EN 303-1 Boilers with forced draught burners: general requirements • EN 303-2 Boilers with forced draught burners: atomizing oil burners • EN 303-3 Gas boilers with forced draught burners: assemblies EN 303-4 Boilers with forced draught burners: oil burners < 70 kW • EN 303-6 Boilers with forced draught burners: combiboilers < 70 kW EN 303-7 Boilers with forced draught burners: gas fired boilers < 1000 kW • EN 656 Gas boilers, type B, 70–300 kW • PrEN 13836 Gas boilers, type B, 300–1000 kW • EN 483 Gas boilers, type C < 70 kW • EN 677 Gas condensing boilers < 70 kW • EN 267 Forced draught oil burners: definitions, etc. • EN 304 Atomizing oil burners: test code • prEN 15034 Condensing oil boilers < 1000 kW • prEN 15035 Room-sealed (type C) oil-fired boilers • EN 676 Forced draught burners for gas • prEN 15456 Electrical power consumption for heat generators
34.4 Utility Boiler Test Methods This third section is dedicated to test methods, with specific reference to the present standard and to the new proposal that is going to be approved by the European Community during 2010. In European Community these are the directives that have promoted boiler efficiency: • 92/42/CE Boiler Efficiency Directive (for the boiler whose power is in the range between 4 and 400 kW) • EPBD Energy Performance of Buildings Directive 2002/91/EC, which involves the boiler as part of the building • EuP: Energy Using Product Directive establishing a framework for the setting of ecodesign requirements for Energy‑Using Products and amending the Council Directive 92/42/EEC. The purpose of Eco-design Directive 2005/32/EC of the European Union is to encourage manufacturers to pay greater attention to the ecological properties of a product even at the development stage. Relevant ecological properties include energy and water consumption, noise levels, and the recyclability of constituent parts. Each of these directives has generated a complex family of product standards that are summarized in Table 34.3. In this paper it is not possible to go into details in all the European Directives and the Product Standards, but it is necessary to explain what the structure is and the main milestones. In the next section, product categories and the technical definitions are presented.
© 2011 by Taylor and Francis Group, LLC
34.4.1 Product Categories The boiler is a product able to supply hot water in a central heating application. The appliance may be developed to provide the sanitary hot water. The performance parameters of the boiler are the nominal heating capacity in kW defined as full load mode and the nominal partial load capacity at 30% of nominal heating capacity. Tests are performed at standard conditions, which means supply plant temperatures: outlet 80°C inlet 60°C and for condensing boilers, supply plant temperatures outlet 50°C and inlet 30°C. The test conditions should be in steady-state operation at fixed low return boiler water temperature or steadystate operation at fixed outlet and inlet boiler water temperatures. Boilers are classified based on 10 criteria: output function, capacity and configuration storage, fuel configuration, boiler flue gas/intake air system technology, power, boiler/burner configuration, boiler water temperature control, burner combustion control, and installation. 34.4.1.1 Output Function
1. Regular boiler: A boiler designed for heating demand only. It may produce the sanitary hot water with a separate appliance (i.e., water storage tank).
719
Utility Boilers
2. Combination boiler (combi): When the boiler is able to satisfy for heat demand and for sanitary hot water.
Hot water
The EN standards add the following qualifications: • Instantaneous combination boiler or instantaneous combi boiler: It is a boiler without an internal hot water store, or with an internal hot water store of a capacity less than 15 liters storage combination boiler, typically there is a plate exchanger. • Storage combination boiler or storage combi boiler: A boiler with an internal hot water tank with an overall volume greater or equal to 15 liters but less than 70 liters. It is a combi boiler with an internal hot water storage of at least 70 liters, where the space heating circuit is not taken directly from the storage. • Combined primary storage unit: It is a boiler where the tank is at least 70 liters and the heated hot water is taken directly from storage water. • Storage combination boilers can be subdivided into: −− primary, where the primary water stored is the water of a space heating circuit. −− secondary, where the water stored is the water used directly for sanitary hot water. Please refer to Figure 34.20. 34.4.1.2 Capacity and Configuration Storage A boiler can have a storage tank for: • primary tank when the water stored is used for central heating water • secondary tank when the water stored is used for sanitary hot water The primary tank combi boiler should be divided into: • No primary store, the overall water volume content is less than 5 liters. • No primary water storage tank, but the unit has a high mass or a high capacity of water. • Integrated thermal store: The boiler is designed to store primary hot water, for space heating and with a secondary heat exchanger to transfer the heat to domestic hot water. • Hot water only thermal store: The product is able to supply only the domestic hot water,
© 2011 by Taylor and Francis Group, LLC
Space heating load
Fuel
Primary storage combi boiler
Hot water Space heating load
Fuel
Secondary storage combi boiler Figure 34.20 Combi boiler configuration.
which is heated by a boiler. There is a secondary heat exchanger between the primary water and the domestic hot water. • Combined primary storage unit (CPSU): The unit is designed to provide both space heating water and the domestic hot water, a burner heats a thermal store that contains the primary water, which is also the space heating circuit water. The store must have a volume of at least 70 liters and the water in the storage is used directly to space heating circuit. Figure 34.21 presents the different configurations. The secondary tank should be divided into: • Instantaneous combi: The sanitary hot water is heated through a heat exchanger (i.e., coil or plate) that is heated directly by the burner. There will be a delay between the request and the answer in terms of waiting time. • Pre-Hot facility: Facility in an instantaneous boiler ( < 15 liters) where the water within the boiler stays hot while there is no demand. The heat could be produced by different heat primary sources (flame, Joule effect, or both). The waiting time, between request and answer, is minimized. • Instantaneous storage combi: It is a product where the temperature sensor is close to the
720
Industrial Combustion Testing
contact directly with ambient air. It is unvented when the storage vessel is sealed.
Integrated thermal tank
34.4.1.3 Fuel Configuration Separate boiler
Hot water
Space heating load
Fuel Thermal store Hot water thermal tank
Separate boiler
Hot water
Fuel
Space heating load
Thermal store Combined primary storage unit
Hot water
Space heating load
Fuel Figure 34.21 Tanks configuration.
domestic hot water inlet. During every tapping, the burner is on until there is domestic hot water demand. • Noninstantaneous storage combi: The storage combination boiler, where the volume capacity may be any size even if it is usually greater than 45 liters. The temperature sensor is inserted in the middle of storage. The right position is optimized under the most efficient condition of operations. The secondary tank could be vented or unvented. The tank is vented when the water of the storage vessel is in
© 2011 by Taylor and Francis Group, LLC
The fuel configuration depends on the primary energy used. Typically we have the following fuels: gas, oil, coal, biomass, electric power, and solar assisted. All boilers indicate the fuels tested on the dataplate. For each different fuel the European Standard requires the specific features and parameters. These parameters depend upon fuel physical propriety and change for the different country or region. Not all fuels have the same diffusion in Europe. There are some fuels that are very popular in Europe along with natural gas and oil; coal and biomass are still a niche in the market, although it is currently increasing on the market. 34.4.1.4 Boiler Flue Gas/Intake Air System This classification for the flue gas/intake air system is applied, at the moment, only to a gas boiler, but it is possible to extend the definitions to other fuel if necessary to explain clearly the type configuration. The letters B and C follow with two digits, generically, these appear as Bxx or Cxx. The letters B and C indicate chimney flue gas kinds used. • Type B takes the combustion air from the indoors in the space around the boiler. • Type C takes the air from the outdoors far from the room where the boiler is installed. The Type B has an open connection between the burner and the inside of the room where the boiler is installed. That is an open system, whereas Type C is defined as room-sealed because the air intake of the burner is taken from outside of the room where the unit is installed. In Type B, the EN and local standards specify health and fire requirements as well as the CO monoxide carbon level, the fuel unburned CxHy, and ventilation system. The system B1x hasn’t got a flue gas damper, whereas the B2x has got a flue gas damper. In Type C units, the first digit indicates the configuration of the air intake and flue gas ducts: • • • •
C1x: Horizontal concentric ducts C3x: Vertical concentric ducts C4x: Side-by-side ducts C5x: Ducts leading to different pressure zones outside the building
721
Utility Boilers
• C6x: The boiler is tested without ducts • C8x: The boilers connected to a negative pressure chimney for the flue gases and to another pressure zone for the air intake A boiler may be suitable for more than one configuration, we note that it is mandatory to describe the different denominations on the dataplate of a boiler. The second digit indicates if present and the position of the combustion air/flue gas fan: • Cx1: There is no fan. • Cx2: There is a fan after the burner but before the flue gas damper. • Cx3: There is a fan positioned before the burner. • Cx4: There is a flue gas fan positioned inside the chimney after the flue gas damper. Burner flue/gas intake kind:
1. Forced draught: It is a forced draught burner when the intake air is fan assisted. 2. Atmospheric: It is an atmospheric burner when the intake air is not fan assisted. Typically the venturi phenomena led the intake air of combustion in the burner.
34.4.1.5 Technology Boiler The main target of a boiler is to capture the overall energy of the flue. This depends on the total heat exchanger surface, the resistance to corrosion, and the resistance to certain temperatures of raw material, for these reasons, the heat exchanger of a gas- or oil-fired water heater or combi can technically be seen as standard, low temperature, or condensing. The boiler efficiency directive (BED, 92/42/EC) defines • Standard boiler: When the average water temperature can be restricted by design. • Low-temperature boiler: When the appliance can work continuously with a return water temperature between 35°C and 40°C, and in certain situations the boiler may produce the condense. • Condensing boiler: When the product is designed to produce condense by combustion gas permanently, which recovers a large part of the water vapor contained in the combustion gases.
© 2011 by Taylor and Francis Group, LLC
There is a proposal by European boiler manufacturers for a different definition of boiler kinds: • A condensing boiler always has a nozzle for condense draining. • A noncondensing boiler never has a nozzle for condense draining and for completing this listing: • If it is a low temperature boiler, the heating system return temperature may be 40°C or lower. • A standard boiler when the heating system inlet temperature must be greater than 40°C. 34.4.1.5.1 Burner Power Control System • On/off boiler: A boiler that only has a single stage of flame burner for space heating. The appliances are included in an on/off boiler even if these have got a burning setup done during the installation within the range approved. • Modulating boiler: A boiler able to move the power burner continuously within the range approved. 34.4.1.6 Power Class The classification approved for gas- or oil-fired boilers (including a combi) are: • The EN standard power classes are: < 70, 70–300, 300–1000 kW, and 1–10 MW. • the Boiler Efficiency Directive power classes are 4–400 kW and above 400 kW. 34.4.1.7 Boiler/Burner Configuration • Burner/boiler assembly, where there are different requirements (and CE-marking tests) for the jet burner and the assembly of the boiler body plus burner (i.e., freestanding boiler and plus burner). • An integrated boiler, where the burner is not sold/tested separately. 34.4.1.8 Boiler Water Temperature Control • Fixed (adjustment at installation) within a range of bandwidth, on/off led by the room thermostat. For rooms without a room thermostat, the temperature control may be implemented by thermostatic radiator valves or separate electronic systems with or without an optimizer.
722
• Modulating PID-type of control on a special modulating room thermostat within a range of bandwidth. For rooms without a room thermostat, the temperature control may be supplemented by thermostatic radiator valves or separate electronic systems with or without optimizer. • Weather controller: An outdoor sensor is setting the operating point for the boiler temperature, along a curve (adjust at installation) that answers to layout and type of heating system. Sometimes the systems are combined with thermostatic radiator valves or separate electronic systems with or without optimizer.
34.4.1.9 Burner Combustion Control The combustion burner control could be to follow several different physics principles, the most actually used are: • Forward control loop: Pneumatic control where the air-fuel ratio is carried out with the venturi principle. Air factor is determined by gas flow rate. • Backward control loop: The air factor is carried out on the basis of the combustion (ionization sensor) and/or the combustion products (e.g., CO sensor). • Forward and backward control: Based on a combination of the above. For each burner combustion control, the ignition could be: • Electronic (through glowing plug or spark plug). • Pilot flame, this technology is going to disappear in the European market.
34.4.1.10 Installation The installation could be: • Indoor installation: The boiler is installed in the building heated. • Outdoor installation: The boiler is installed in open air close to building heated. • The mounting could be: −− Floor standing −− Wall-hung boiler
© 2011 by Taylor and Francis Group, LLC
Industrial Combustion Testing
34.4.2 EN Product Standard 34.4.2.1 Introduction The current EN standards for heating boilers and burners for these heating boilers are split up in a set of categories: • Flue gas system (type C room-sealed, type B without a fan, type B with forced draught burner), by capacity class (up to 70 kW, 70–300 kW, 300–1000 kW) • Fuel (oil, gas) • Configuration (boilers, combi-boilers, boilerburner assemblies, and separate burners) • Technology: Condensing, low temperature, and noncondensing boilers The different Commission mandates lead to different amendments over periods of more than 10 years. These are: • The Gas Appliance Directive 90/368/EC from 1990. • The Boiler Directive 92/42/EC from 1992 that is now incorporated in most relevant standards. Figure 34.22 gives a synthetic overview of all standards available. Details on these standards are reported in the appendix. 34.4.2.2 Classification of Gas Appliances by Type The Sector Forum Gas Utilization Committee and the CEN Technical committee give details of a general scheme for the classification of such appliances according to the flue system. This scheme concerns appliances that could be installed within buildings, it does not apply to outdoor appliances. This form of appliance classification is widely used in the preparation of European Standards for gas appliances and identifies the requirements and methods of tests that are applicable to the various methods of evacuating the products of combustion. The classification introduces the concept of “type.” The type, as well as describing the general scheme, has been used for this purpose. 34.4.3 Eco Design Preparatory Study: Boilers The European commission requested 14 different lots of preparatory Eco-Design implementing studies where the boilers were present in Lot 1. The study was started in 2005 and the final conclusions were printed in February 2008. The scope of this preparatory study was to develop several labels, for an easy way to
723
Utility Boilers
Gas-fired Type B Incl. not fanassisted, atmospheric Burners
EN 676
Fan-assisted (forced draught burner) Boilers
EN 297
Type C
Type B & C
Combi
EN 303–1 (general) New: prEN 15502–1 EN 303-3 (<70 kW) EN 625 EN 656 (70–300 kW) prEN 13826 (300–1000) kW prEN 303-7 (< 1000 kW)
Condensing
Boiler
EN 677
EN 483 (< 70 kW)
EN 15034 (<1000 kW)
EN 15035
Oil-fired
EN 267
(EN 1)
EN 303–1 EN 303–2 EN 303–4 EN 304
En 303-6
Figure 34.22 Classification of appliances and flue gas systems.
communicate the energy efficiency of heating systems to installers and end users. In the following, we will try to summarize the main concept and explain the finale result of this preliminary study on boilers.
34.4.3.1 Assessment Procedure The assessment procedure follows, in general, the test method for CE marking. In some cases current EN test methods are not able to describe correctly the energy efficiency and emissions of a central heating boiler system as a whole. Several test methods for CE marking are carried out at steady-state and so it is not possible to give a real test for efficiency, emissions, and pollution out in the field for heating systems. To describe the behavior of the field, it is necessary to define new methods of assessment that take into account the dynamic behavior of plant and operating cycles. These methods require considerable development time and considerable effort. For those reasons, the implementing study of energy using the product has been proposed to start a transitory regime for the compliance assessment for a central heating boiler system where the strategy is summarized as following: • A set of tests, which take into account the European test standards. • Involvement of the test institutes and notifies bodies for compliance assessment.
© 2011 by Taylor and Francis Group, LLC
• Use a mathematical validation model that takes into account results from European test standards and others. These sets of results are used as inputs to calculate energy efficiency, carbon, and NOx emissions of the central heating system. We think that there will be a transitory period for at least a period of 4–6 years, after which time the European Union, step-by-step, will introduce the necessary dynamic test/emulation methods, as new standards. Those new standards will allow assessing more accurately the following parameters: • • • •
Energy consumption Carbon and NOx impacts CO, CH4, CxHy values Particulate matter emissions
At the moment we are going to describe only the methodology for the transitory period. 34.4.3.2 Test Requirement Below is a list of parameters (some are new, others are taken by present EU standards) previewed by the compliance assessment: • Boiler characteristics: • Power output in kW at 80/60°C
724
• Turndown ratio in percentage • Standby heat loss evaluate at 50°C • Steady-state efficiency at 80/60°C and 50/30°C at nominal load, medium load, and minimum load • Type of fuel • Electricity consumption at nominal minimum load as well as burner off • NOx and CO in steady-state at nominal output power • Noise level • Solar systems: • Types of collectors • Collector loop loss (UL) • Tank heat loss coefficient (UA) • Tank volume capacity • Heat Pump: • Type of heat pump • Nominal output power • Efficiency tests at the Tsink and Tsource • Type of refrigerant (Global Warm Protection)
Industrial Combustion Testing
• Water content (in liter). • Type of air-fuel mixer and/or air factor. • System of control • Room temperature control if included in the package: −− None (fixed BT). −− Weather controlled BT. −− On/off RT. −− Modulating RT. −− Time proportional RT. • System of valve controllers if included in the package offered for CE-marking: • None. • TRV 2 K. • TRV 1 K. • Motor + PID-loop. • Motor + CPU. • For a heat pump used in the central heating system, the refrigerant has to assess by selfdeclaration. 34.4.3.4 Labeling
Fossil fuel is used for the high heating values to assess the input power and the energy efficiency. Electric power inputs are considered conventional as primary energy conversion, the factor 2.5 (1 kWhe = 2.5 kWhprim). Usually for central heating boiler systems (all gas- and oil-fired boilers) third-party testing is current practice, involving specialized test institutes and notified bodies. The study indicates following this good practice for the labeling also. Obviously, for the manufacturer this means an increase of approval costs. A recent round-robin test by Labnet [12] shows tolerances of ±4% on part load efficiency and around ±2% on full load efficiency. The implementing ecodesign measures are very sensitive to the performance of a central heating boiler system and so it is very important to increase the quality level of test institutes in terms of fuel quality, ambient parameters, and measure instruments. 34.4.3.3 Other Data Input Not only the data obtained by harmonized standards are used, several simple data input and measurements will be required. The study indicates: • Boiler • Envelope (in m3). • Weight (in kg). • Room air intake (e.g., from type declaration).
© 2011 by Taylor and Francis Group, LLC
Labeling, in the form of a star rating, was part of the 92/42/ EC, which is now superseded by the Eco-design directive 2005/32/EC. The study has suggested including mandatory labeling as an Eco-design implementing measure.
34.5 Appendix [13] EN 297 Atmospheric gas boiler without fan, < 70 kW Title: EN 297:1994. Gas-fired central heating boilers. Type B11 and B11BS boilers fitted with atmospheric burners of nominal heat input not exceeding 70 kW. Fuel: Gases defined in three gas families. Burners: atmospheric burner without fan. Threshold values: • Nominal input power < 70 kW. • Maximum operating pressure water side not greater than 6 bar. • The maximum temperature of feed less than 95°C. Scope: Test method requirements for construction, efficiency, classification, marking, and design. Directive links: 90/396/EC.
725
Utility Boilers
EN 303 Boiler with forced draught burners EN 303-1 general requirement Title: EN 303-1:1999. Heating boilers. Part 1: Heating boilers with forced draught burners. Terminology, general requirements, testing, and marking. Fuel: Gases defined in three gas families and oil. Burners: forced draught burners. Threshold values: • Nominal output power < 1000 kW. • Combustion chamber with positive and negative pressure. Scope: Terminology, materials used, test of materials. Directive links: 89/106/EC. EN 303-2 Atomizing oil burner Title: EN 303-2:1998 en. Heating boilers. Part 2: Heating boilers with forced draught burners. Special requirements for boilers with atomizing oil burners. Fuel: Oil. Burners: Forced draught burners. Threshold values: • Nominal output power < 1000 kW in accordance with EN 303-1. • Nominal output power < 70 kW in accordance with EN 303-4. • Burners in accordance with EN 267. Scope: Requirements apply to heating boiler testing. Directive links: 89/106/EC. EN 303-3 Assemblies Title: EN 303-3:1998. Heating boilers. Part 3: Gas-fired central heating boilers. Assembly comprising a boiler body and a forced draught burner. Title: EN 303-2:1998 en. Heating boilers. Part 2: Heating boilers with forced draught burners. Special requirements for boilers with atomizing oil burners. Fuel: Gases defined in three gas families. Burners: forced draught burners. Threshold values: • Burners in accordance with EN 676.
© 2011 by Taylor and Francis Group, LLC
Scope: Requirements and test methods for safety, construction, and efficiency. Directive links: 90/396/EC. EN 303-4 Oil burner < 70 kW Title: EN 303-4:1999. Heating boilers: Part 4: Heating boilers with forced draught burners. Special requirements for boilers with forced draught oil burners with outputs up to 70 kW and a maximum operating pressure of 3 bar. Terminology, special requirements, testing and marking. Fuel: Oil. Burners: Forced draught burners. Threshold values: • Nominal output power < 70 kW. • Combustion chamber with positive and negative pressure. • Operating water pressure not greater than 3 bar. Scope: Terminology, materials used, test of materials, and marking requirements. Directive links: 89/106/EC. EN 303-6 combi boiler < 70 kW Title: EN 303-6:2000 en. Heating boilers. Part 6: Heating boilers with forced draught burners. Specific requirements for the domestic hot water operation of combination boilers with atomizing oil burners of nominal heat input not exceeding 70 kW. Fuel: Oil. Burners: Forced draught burners. Threshold values: • Nominal output power < 70 kW • Heating service and domestic hot water service • Domestic hot water production instantaneously or storage feature Scope: Materials used, test of materials, and marking requirements. EN 303-7 gas fired boiler < 1000 kW Title: PrEN 303-7 dec. 2003. Heating boilers. Part 7: Gas-fired central heating boilers equipped with a forced draught burner of nominal heat output not exceeding 1000 kW. Fuel: Gases defined in three gas families. Burners: Forced draught burners.
726
Industrial Combustion Testing
Threshold values: • Maximum output power < 1000 kW. • Burners in accordance with EN 676. • Maximum operating water temperature not greater than 105°C. • Maximum operating water side pressure not greater than 8 bar. • Type boiler B23. Scope: Requirements and test methods for safety, construction, efficiency for standard and low temperature boilers. Directive links: 90/396/EC. EN 656 Gas boilers, type B, 70–300 KW Title: EN 656:1999. Gas-fired central heating boilers. Type B boilers of nominal heat input exceeding 70 kW but not exceeding 300 kW. Fuel: Gases defined in three gas families. Burners: Atmospheric, fan assisted, premix. Threshold values: • Range output power 70 < Pn < 300 kW. • Maximum operating water temperature not greater than 105°C. • Maximum operating water side pressure not greater than 6 bar. • Type boiler B with or without fan: B1, B12, B13, B14, B2, B22, B23. Scope: Requirements and test methods for safety, construction, efficiency, marking, and classification boilers. Directive links: 90/396/EC. PrEN 13836 Gas boiler type B, 300–1000 kW Title: prEN 13836:2006. Gas-fired central heating boilers. Type B boilers of nominal heat input exceeding 300 kW, but not exceeding 1000 kW. Fuel: Gases defined in three gas families. Burners: Atmospheric, fan assisted, premix. Threshold values: • Range output power 300 < Pn < 1000 kW. • Maximum operating water temperature not greater than 105°C. • Maximum operating water side pressure not greater than 6 bar. • Type boiler B with or without fan: B1, B12, B13, B14, B2, B22, B23.
© 2011 by Taylor and Francis Group, LLC
Scope: Requirements and test methods for safety, construction, efficiency, marking, and classification boilers. Directive links: 90/396/EC, 92/42/EC. EN 483 Gas boiler, type C < 70 kW Title: EN 483:1999 en. Gas-fired central heating boilers. Type C boilers of nominal heat input not exceeding 70 kW. Fuel: Gases defined in three gas families. Burners: Atmospheric, fan assisted, premix. Threshold values: • Range output power < 70 kW. • Maximum operating water temperature not greater than 95°C. • Maximum operating water side pressure not greater than 6 bar. • Type boiler C. The requirements for boiler types C41, C51, C61, C71, C81, for external use only, condensing and combi are not completes. Scope: Requirements and test methods for safety, construction, efficiency, marking, and classification boilers. Directive links: 90/396/EC, 97/23/EC. EN 677 gas Condensing boiler < 70 kW Title: EN 677:1998 en. Gas-fired central heating boilers. Specific requirements for condensing boilers with a nominal heat input not exceeding 70 kW. Fuel: Gases defined in three gas families. Burners: Atmospheric, fan assisted, premix. Threshold values: • Range output power < 70 kW. • Maximum operating water temperature not greater than 95°C. • Maximum operating water side pressure not greater than 6 bar. • The standard completes the EN 297, EN 483, EN 625 for condensing boilers. • Type B (only with fan assisted), type C. Scope: Requirements of type testing. Directive links: 90/396/EC, 92/42/EC. EN 267 Forced draught oil burner–definitions Title: prEN 267:2005 en. Forced draught oil burners. Definitions, requirements, testing, marking, will replace EN 267:1999.
727
Utility Boilers
Fuel: Oil. Burners: Forced draught burner. Threshold values: • Fuel having a viscosity range between 1,6 cSt, and 6 cSt. • Single burner with a single combustion chamber. Scope: Requirements of type testing, terminology, general construction, and operation. Directive links: No reference. EN 304 Atomizing oil burners–test code Title: EN 304:1993. Heating boilers. Test code for heating boilers for atomizing. Oil burners. Fuel: Oil. Burners: Forced draught burner. Threshold values: • Oil-fired heating boiler, combined boilers, and water heater. • Single burner with a single combustion chamber. Scope: Requirements of type testing of thermal performance. Directive links: No reference. PrEN 15034 Condensing oil boilers < 1000 kW Title: prEN 15034:2004 en. Heating boilers. Condensing heating boilers for fuel oil. Fuel: Oil. Burners: Forced draught burner. Threshold values: • Maximum output power < 1000 kW. • Condensing boiler declared by manufacturer. • The standard modifies and integrates the following standard: EN 303-1, EN 303-2, EN 304, EN 304/A1, and EN 304/A2. Scope: Supplementary requirements. Directive links: No reference. PrEN 15035 Room-sealed oil fired boiler (type C) Title: prEN 15035:2004. Heating boilers. Room sealed operations for boilers for fuel oil. Fuel: Oil. Burners: Forced draught burner.
© 2011 by Taylor and Francis Group, LLC
Threshold values: • Type boiler Cx3: C13, C33, and C53. • Type C43 with their connection ducts but without the chimney. • Type C83 with their connection ducts but without the chimney. • The output nominal power < 70 kW. • The maximum operating feed water temperature < 100°C. • The maximum operating water side pressure < 8 bar. • Maximum level of NOx and CO. • Air factor value. Scope: Requirements for test methods, safety, efficiency, classification and marking, installation recommendations. Directive links: 92/42/EC. EN 676 Forced draught burners for gas Title: EN 676:2003. Automatic forced draught burners for gaseous fuels. Reference to directive 90/396/EEG PBC 142:2005. Fuel: Gases defined in three gas families. Burners: Forced draught burner, total premix burner, and nozzle mixed burner. Threshold values: • Single burner with single combustion chamber. Scope: Terminology, requirements for test methods, safety, efficiency, device, construction, classification and marking. Directive links: No reference. PrEN 15456 Electrical power consumption for heat generator Title: prEN 15456:2006 Heating boilers. Electrical power consumption for heat. Generators: System boundaries: Measurements. Fuel: Gases defined in three gas families and oil. Burners: Overall burner types. Threshold values: • Boiler with an equipped burner. • Heating boiler sold without burner. • Standard in accordance with EN 267, EN 676, EN 1, EN 303-1, EN 303-2, EN 304, and prEN 15304.
728
• Nominal output power < 10 MW. • Maximum operating feed water tem perature < 110°C in accordance to prEN 14394. Scope: Terminology, definitions, system boundaries, the measurements of determination of electrical consumption, and the waterside resistance in order to carry out the electric auxiliary energy where applicable. Directive links: No reference.
Industrial Combustion Testing
References
1. “Impact Assessment Study on a Possible Extension, Tightening or Simplification of the Framework Directive 92/75 EEC on Energy Labelling of Household Appliances,” Appendix 2: Product Studies Carried Out by Europe Economics, Fraunhofer-ISI, BSR Sustainability and FfE. 2. Borman, G. L., and Rayland, K. W. Combustion Engineering. New York: McGraw-Hill, 1998. 3. Bejan, A. Advanced Engineering Thermodynamics. New York: Wiley-Interscience Publication, 1997. 4. Turns, S. R. An Introduction to Combustion. New York: McGraw-Hill, 1996.
© 2011 by Taylor and Francis Group, LLC
5. U.S. Department of Energy. “Check Burner Air to Fuel Ratios.” Energy Tips May 2002, available at http:// www.eere.energy.gov/industry/bestpractices/pdfs/ et_check_burner.pdf/ 6. Kemna, R., Van Elbury, M., Li, W., and Van Holstijn, R., “Eco-design of CH Boilers-Task 4”, available at http:// www.ecoboiler.org 7. “M-CHP Interim Report,” Carbon Trust: CTC726, available at http://www.carbontrust.co.uk 8. Specchia, S., Toniato, G., and Specchia, V. “Natural Gas Combustion Catalysts for Environmental-Friendly Domestic Burners.” ICOSCAR-3 Third Conference on Structured Catalysts and Reactors Ischia, Gulf of Naples, Italy, September 27–30, 2009. 9. UNI EN30 2: “Heating Boilers with Forced Draught Burners. Special Requirements for Boilers with Forced Draught Oil Burners with Outputs up to 70 kW and a Maximum Operating Pressure of 3 bar: Terminology, Special Requirements, Testing and Marking.” 10. Minutolo, P., D’Anna, A., Commodo, M., Pagliara, R., Toniato, G., and Accordini, C. “Emission of Ultrafine Particles from Natural Gas Domestic Burners.” Environmental Engineering Science December 2008. 11. Hottel, H. C., and Sarofin, A. F. Radiative Transfer. New York: McGraw-Hill, 1967. 12. Schweitzer, J. “Measuring Efficiency of Boilers and Water Heaters. SAVE Conference, Graz, Austria, November 1999. 13. Eco-design of CH boiler: Task 1, available at http:// www.ecoboiler.org
Appendix A: F-Distribution (99%, 95%, & 90% Confidence)
729 © 2011 by Taylor and Francis Group, LLC
730
DF v1 Conf
1
2
3
4
5
6
7
8
9
10
15
20
25
30
50
100
Infinity
1
99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90%
4052 161 39.86 98.50 18.51 8.53 34.12 10.13 5.54 21.20 7.71 4.54 16.26 6.61 4.06 13.75 5.99 3.78 12.25 5.59 3.59 11.26 5.32 3.46 10.56 5.12 3.36 10.04 4.96 3.29 9.65 4.84 3.23 9.33 4.75 3.18 9.07 4.67 3.14
4999 199 49.50 99.00 19.00 9.00 30.82 9.55 5.46 18.00 6.94 4.32 13.27 5.79 3.78 10.92 5.14 3.46 9.55 4.74 3.26 8.65 4.46 3.11 8.02 4.26 3.01 7.56 4.10 2.92 7.21 3.98 2.86 6.93 3.89 2.81 6.70 3.81 2.76
5404 216 53.59 99.16 19.16 9.16 29.46 9.28 5.39 16.69 6.59 4.19 12.06 5.41 3.62 9.78 4.76 3.29 8.45 4.35 3.07 7.59 4.07 2.92 6.99 3.86 2.81 6.55 3.71 2.73 6.22 3.59 2.66 5.95 3.49 2.61 5.74 3.41 2.56
5624 225 55.83 99.25 19.25 9.24 28.71 9.12 5.34 15.98 6.39 4.11 11.39 5.19 3.52 9.15 4.53 3.18 7.85 4.12 2.96 7.01 3.84 2.81 6.42 3.63 2.69 5.99 3.48 2.61 5.67 3.36 2.54 5.41 3.26 2.48 5.21 3.18 2.43
5764 230 57.24 99.30 19.30 9.29 28.24 9.01 5.31 15.52 6.26 4.05 10.97 5.05 3.45 8.75 4.39 3.11 7.46 3.97 2.88 6.33 3.69 2.73 6.06 3.48 2.61 5.64 3.33 2.52 5.32 3.20 2.45 5.06 3.11 2.39 4.86 3.03 2.35
5859 234 58.20 99.33 19.33 9.33 27.91 8.94 5.28 15.21 6.16 4.01 10.67 4.95 3.40 8.47 4.28 3.05 7.19 3.87 2.83 6.37 3.58 2.67 5.80 3.37 2.55 5.39 3.22 2.46 5.07 3.09 2.39 4.82 3.00 2.33 4.62 2.92 2.28
5928 237 58.91 99.36 19.35 9.35 27.67 8.89 5.27 14.98 6.09 3.98 10.46 4.88 3.37 8.26 4.21 3.01 6.99 3.79 2.78 6.18 3.50 2.62 5.61 3.29 2.51 5.20 3.14 2.41 4.89 3.01 2.34 4.64 2.91 2.28 4.44 2.83 2.23
5981 239 59.44 99.38 19.37 9.37 27.49 8.85 5.25 14.80 6.04 3.95 10.29 4.82 3.34 8.10 4.15 2.98 6.84 3.73 2.75 6.03 3.44 2.59 5.47 3.23 2.47 5.06 3.07 2.38 4.74 2.95 2.30 4.50 2.85 2.24 4.30 2.77 2.20
6022 241 59.86 99.39 19.38 9.38 27.34 8.81 5.24 14.66 6.00 3.94 10.16 4.77 3.32 7.98 4.10 2.96 6.72 3.68 2.72 5.91 3.39 2.56 5.35 3.18 2.44 4.94 3.02 2.35 4.63 2.90 2.27 4.39 2.80 2.21 4.19 2.71 2.16
6056 242 60.19 99.40 19.40 9.39 27.23 8.79 5.23 14.55 5.96 3.92 10.05 4.74 3.30 7.87 4.06 2.94 6.62 3.64 2.70 5.81 3.35 2.54 5.26 3.14 2.42 4.85 2.98 2.32 4.54 2.85 2.25 4.30 2.75 2.19 4.10 2.67 2.14
6157 246 61.22 99.43 19.43 9.42 26.87 8.70 5.20 14.20 5.86 3.87 9.72 4.62 3.24 7.56 3.94 2.87 6.31 3.51 2.63 5.52 3.22 2.46 4.96 3.01 2.34 4.56 2.85 2.24 4.25 2.72 2.17 4.01 2.62 2.10 3.82 2.53 2.05
6209 248 61.74 99.45 19.45 9.44 26.69 8.66 5.18 14.02 5.80 3.84 9.55 4.56 3.21 7.40 3.87 2.84 6.16 3.44 2.59 5.36 3.15 2.42 4.81 2.94 2.30 4.41 2.77 2.20 4.10 2.65 2.12 3.86 2.54 2.06 3.66 2.46 2.01
6240 249 62.05 99.46 19.46 9.45 26.58 8.63 5.17 13.91 5.77 3.83 9.45 4.52 3.19 7.30 3.83 2.81 6.06 3.40 2.57 5.26 3.11 2.40 4.71 2.89 2.27 4.31 2.73 2.17 4.01 2.60 2.10 3.76 2.50 2.03 3.57 2.41 1.98
6260 250 62.26 99.47 19.46 9.46 26.50 8.62 5.17 13.84 5.75 3.82 9.38 4.50 3.17 7.23 3.81 2.80 5.99 3.38 2.56 5.20 3.08 2.38 4.65 2.86 2.25 4.25 2.70 2.16 3.94 2.57 2.08 3.70 2.47 2.01 3.51 2.38 1.96
6302 252 62.69 99.48 19.48 9.47 26.35 8.58 5.15 13.69 5.70 3.80 9.24 4.44 3.15 7.09 3.75 2.77 5.86 3.32 2.52 5.07 3.02 2.35 4.52 2.80 2.22 4.12 2.64 2.12 3.81 2.51 2.04 3.57 2.40 1.97 3.38 2.31 1.92
6334 253 63.01 99.49 19.49 9.48 26.24 8.55 5.14 13.58 5.66 3.78 9.13 4.41 3.13 6.99 3.71 2.75 5.75 3.27 2.50 4.96 2.97 2.32 4.41 2.76 2.19 4.01 2.59 2.09 3.71 2.46 2.01 3.47 2.35 1.94 3.27 2.26 1.88
6366 254 63.33 99.50 19.50 9.49 26.13 8.53 5.13 13.46 5.63 3.76 9.02 4.36 3.10 6.88 3.67 2.72 5.65 3.23 2.47 4.86 2.93 2.29 4.31 2.71 2.16 3.91 2.54 2.06 3.60 2.40 1.97 3.36 2.30 1.90 3.17 2.21 1.85
2
3
4
5
6
7
8
9
10
11
12
13
© 2011 by Taylor and Francis Group, LLC
Appendix A: F-Distribution (99%, 95%, & 90% Confidence)
DF v2
15
16
17
18
19
20
25
30
40
50
100
Infinity
99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90% 99% 95% 90%
8.86 4.60 3.10 8.68 4.54 3.07 8.53 4.49 3.05 8.40 4.45 3.03 8.29 4.41 3.01 8.18 4.38 2.99 8.10 4.35 2.97 7.77 4.24 2.92 7.56 4.17 2.88 7.31 4.08 2.84 7.17 4.03 2.81 6.90 3.94 2.76 6.63 3.84 2.71
6.51 3.74 2.73 6.36 3.68 2.70 6.23 3.63 2.67 6.11 3.59 2.64 6.01 3.55 2.62 5.93 3.52 2.61 5.85 3.49 2.59 5.57 3.39 2.53 5.39 3.32 2.49 5.18 3.23 2.44 5.06 3.18 2.41 4.82 3.09 2.36 4.61 3.00 2.30
5.56 3.34 2.52 5.42 3.29 2.49 5.29 3.24 2.46 5.19 3.20 2.44 5.09 3.16 2.42 5.01 3.13 2.40 4.94 3.10 2.38 4.68 2.99 2.32 4.51 2.92 2.28 4.31 2.84 2.23 4.20 2.79 2.20 3.98 2.70 2.14 3.78 2.60 2.08
5.04 3.11 2.39 4.89 3.06 2.36 4.77 3.01 2.33 4.67 2.96 2.31 4.58 2.93 2.29 4.50 2.90 2.27 4.43 2.87 2.25 4.18 2.76 2.18 4.02 2.69 2.14 3.83 2.61 2.09 3.72 2.56 2.06 3.51 2.46 2.00 3.32 2.37 1.94
4.69 2.96 2.31 4.56 2.90 2.27 4.44 2.85 2.24 4.34 2.81 2.22 4.25 2.77 2.20 4.17 2.74 2.18 4.10 2.71 2.16 3.85 2.60 2.09 3.70 2.53 2.05 3.51 2.45 2.00 3.41 2.40 1.97 3.21 2.31 1.91 3.02 2.21 1.85
4.46 2.85 2.24 4.32 2.79 2.21 4.20 2.74 2.18 4.10 2.70 2.15 4.01 2.66 2.13 3.94 2.63 2.11 3.87 2.60 2.09 3.63 2.49 2.02 3.47 2.42 1.98 3.29 2.34 1.93 3.19 2.29 1.90 2.99 2.19 1.83 2.80 2.10 1.77
4.28 2.76 2.19 4.14 2.71 2.16 4.03 2.66 2.13 3.93 2.61 2.10 3.84 2.58 2.08 3.77 2.54 2.06 3.70 2.51 2.04 3.46 2.40 1.97 3.30 2.33 1.93 3.12 2.25 1.87 3.02 2.20 1.84 2.82 2.10 1.78 2.64 2.01 1.72
4.14 2.70 2.15 4.00 2.64 2.12 3.89 2.59 2.09 3.79 2.55 2.06 3.71 2.51 2.04 3.63 2.48 2.02 3.56 2.45 2.00 3.32 2.34 1.93 3.17 2.27 1.88 2.99 2.18 1.83 2.89 2.13 1.80 2.69 2.03 1.73 2.51 1.94 1.67
4.03 2.65 2.12 3.89 2.59 2.09 3.78 2.54 2.06 3.68 2.49 2.03 3.60 2.46 2.00 3.52 2.42 1.98 3.46 2.39 1.96 3.22 2.28 1.89 3.07 2.21 1.85 2.89 2.12 1.79 2.78 2.07 1.76 2.59 1.97 1.69 2.41 1.88 1.63
3.94 2.60 2.10 3.80 2.54 2.06 3.69 2.49 2.03 3.59 2.45 2.00 3.51 2.41 1.98 3.43 2.38 1.96 3.37 2.35 1.94 3.13 2.24 1.87 2.98 2.16 1.82 2.80 2.08 1.76 2.70 2.03 1.73 2.50 1.93 1.66 2.32 1.83 1.60
3.66 2.46 2.01 3.52 2.40 1.97 3.41 2.35 1.94. 3.31 2.31 1.91 3.23 2.27 1.89 3.15 2.23 1.86 3.09 2.20 1.84 2.85 2.09 1.77 2.70 2.01 1.72 2.52 1.92 1.66 2.42 1.87 1.63 2.22 1.77 1.56 2.04 1.67 1.49
3.51 2.39 1.96 3.37 2.33 1.92 3.26 2.28 1.89 3.16 2.23 1.86 3.08 2.19 1.84 3.00 2.16 1.81 2.94 2.12 1.79 2.70 2.01 1.72 2.55 1.93 1.67 2.37 1.84 1.61 2.27 1.78 1.57 2.07 1.68 1.49 1.88 1.57 1.42
3.41 2.34 1.93 3.28 2.28 1.89 3.16 2.23 1.86 3.07 2.18 1.83 2.98 2.14 1.80 2.91 2.11 1.78 2.84 2.07 1.76 2.60 1.96 1.68 2.45 1.88 1.63 2.27 1.78 1.57 2.17 1.73 1.53 1.97 1.62 1.45 1.77 1.51 1.38
3.35 2.31 1.91 3.21 2.25 1.87 3.10 2.19 1.84 3.00 2.15 1.81 2.92 2.11 1.78 2.84 2.07 1.76 2.78 2.04 1.74 2.54 1.92 1.66 2.39 1.84 1.61 2.20 1.74 1.54 2.10 1.69 1.50 1.89 1.57 1.42 1.70 1.46 1.34
3.22 2.24 1.87 3.08 2.18 1.83 2.97 2.12 1.79 2.87 2.08 1.76 2.78 2.04 1.74 2.71 2.00 1.71 2.64 1.97 1.69 2.40 1.84 1.61 2.25 1.76 1.55 2.06 1.66 1.48 1.95 1.60 1.44 1.74 1.48 1.35 1.52 1.35 1.26
3.11 2.19 1.83 2.98 2.12 1.79 2.86 2.07 1.76 2.76 2.02 1.73 2.68 1.98 1.70 2.60 1.94 1.67 2.54 1.91 1.65 2.29 1.78 1.56 2.13 1.70 1.51 1.94 1.59 1.43 1.82 1.52 1.39 1.60 1.39 1.29 1.36 1.24 1.18
3.00 2.13 1.80 2.87 2.07 1.76 2.75 2.01 1.72 2.65 1.96 1.69 2.57 1.92 1.66 2.49 1.88 1.63 2.42 1.84 1.61 2.17 1.71 1.52 2.01 1.62 1.46 1.80 1.51 1.38 1.68 1.44 1.33 1.43 1.28 1.21 1.00 1.00 1.00
Appendix A: F-Distribution (99%, 95%, & 90% Confidence)
14
731
© 2011 by Taylor and Francis Group, LLC
Appendix B: EPA Sample Methods
Method 1 1A 2 2A 2B 2C 2D 2E 2F 2G 2H 3 3A 3B 3C 4 5 5A 5B 5C 5D 5E 5F 5G 5H 5I 6 6A 6B 6C 7 7A 7B 7C 7D 7E 8 9 10
Title Sample and velocity traverses for stationary sources Sample and velocity traverses for stationary sources with small stacks or ducts Determination of stack gas velocity and volumetric flow rate (Type S pitot tube) Direct measurement of gas volume through pipes and small ducts Determination of exhaust gas volume flow rate from gasoline vapor incinerators Determination of gas velocity and volumetric flow rate in small stacks or ducts (standard pitot tube) Measurement of gas volume flow rate in small pipes and ducts Determination of landfill gas production flow rate Determination of stack gas velocity and volumetric flow rate with three-dimensional probes Determination of stack gas velocity and volumetric flow rate with two-dimensional probes Determination of stack gas velocity taking into account velocity decay near the stack wall Gas analysis for the determination of dry molecular weight Determination of oxygen and carbon dioxide concentrations in emissions from stationary sources (instrumental analyzer procedure) Gas analysis for the determination of emission rate correction factor or excess air Determination of carbon dioxide, methane, nitrogen, and oxygen from stationary sources Determination of moisture content in stack gases Determination of particulate matter emissions from stationary sources Determination of particulate matter emissions from asphalt processing and asphalt roofing industry Determination of nonsulfuric acid particulate matter emissions from stationary sources (reserved) Determination of particulate matter emissions from positive pressure fabric filters Determination of particulate matter emissions from the wool fiberglass insulation manufacturing industry Determination of nonsulfate particulate matter emissions from stationary sources Determination of particulate matter emissions from wood heaters (dilution tunnel sampling location) Determination of particulate matter emissions from wood heaters from a stack location Determination of low-level particulate matter emissions from stationary sources Determination of sulfur dioxide emissions from stationary sources Determination of sulfur dioxide, moisture, and carbon dioxide emissions from fossil fuel combustion sources Determination of sulfur dioxide and carbon dioxide daily average emissions from fossil fuel combustion sources Determination of sulfur dioxide emissions from stationary sources (instrumental analyzer procedure) Determination of nitrogen oxide emissions from stationary sources Determination of nitrogen oxide emissions from stationary sources—ion chromatographic method Determination of nitrogen oxide emissions from stationary sources (ultraviolet spectrophotometry) Determination of nitrogen oxide emissions from stationary sources—alkaline-permanganate/colorimetric method Determination of nitrogen oxide emissions from stationary sources—alkaline-permanganate/ion chromatographic method Determination of nitrogen oxide emissions from stationary sources (instrumental analyzer procedure) Determination of sulfuric acid mist and sulfur dioxide emissions from stationary sources Visual determination of the opacity of emissions from stationary sources Alternate method 1: Determination of opacity from stationary sources remotely by lidar Determination of carbon monoxide emissions from stationary sources (Continued)
733 © 2011 by Taylor and Francis Group, LLC
734
Appendix B: EPA Sample Methods
(Continued) Method 10A 10B 11 12 13A 13B 14 14A 15 15A 16 16A 17 18 19 20 21 22 23 24 24A 25 25A 25B 25C 25D 25E 26 26A 27 28 28A 29
Title Determination of carbon monoxide emissions in certifying continuous emission monitoring systems at petroleum refineries Determination of carbon monoxide emissions from stationary sources Determination of hydrogen sulfide content of fuel gas streams in petroleum refineries Determination of inorganic lead emissions from stationary sources Determination of total fluoride emissions from stationary sources—SPADNS zirconium lake method Determination of total fluoride emissions from stationary sources—specific ion electrode method Determination of fluoride emissions from potroom roof monitors for primary aluminum plants Determination of total fluoride emissions from selected sources at primary aluminum production facilities Determination of hydrogen sulfide, carbonyl sulfide, and carbon disulfide emissions from stationary sources Determination of total reduced sulfur emissions from sulfur recovery plants in petroleum refineries Semicontinuous determination of sulfur emissions from stationary sources Determination of total reduced sulfur emissions from stationary sources (impinger technique) Determination of particulate emissions from stationary sources (in-stack filtration method) Measurement of gaseous organic compounds by gas chromatography Determination of sulfur dioxide removal efficiency and particulate, sulfur dioxide, and nitrogen oxides emission rates Determination of nitrogen oxides, sulfur dioxide, and diluent emissions from stationary gas turbines Determination of volatile organic compound leaks Visual determination of fugitive emissions from material sources and smoke emissions from flares Determination of polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans from stationary sources Determination of volatile matter content, water content, density, volume solids, and weight solids of surface coatings Determination of volatile matter content and density of printing inks and related coatings Determination of total gaseous nonmethane organic emissions as carbon Determination of total gaseous organic concentration using a flame ionization analyzer Determination of total gaseous organic concentration using a nondispersive infrared analyzer Determination of nonmethane organic compounds (NMOC) in MSW landfill gases Determination of volatile organic concentration of waste samples Determination of vapor phase organic concentration in waste samples Determination of hydrogen chloride emissions from stationary sources Determination of hydrogen halide and halogen emissions from stationary sources—isokinetic method Determination of vapor tightness of gasoline delivery tank using pressure-vacuum test Certification and auditing of wood heaters Measurement of air-to-fuel ratio and minimum achievable burn rates for wood-fired appliances Determination of metals emissions from stationary sources
Source: Appendix G in C. Baukal, Industrial Combustion Pollution and Control, Marcel Dekker, New York, 2004.
© 2011 by Taylor and Francis Group, LLC
Appendix C: Common Conversions
1 Btu = 1 Btu/ft3 = 1 Btu/hr =
1,000,000 Btu/hr = 1 Btu/hr-ft2 = 1 Btu/hr-ft-°F = 1 Btu/hr-ft2-°F = 1 Btu/lb = 1 Btu/lb-°F = 1 cal = 1 cal/cm2-s = 1 cal/cm-s-°C = 1 cal/g = 1 cal/g-°C = 1 centipoise = 1 cm2/s = 1 ft = 1 ft2/s = 1 g/cm3 =
1 hp =
1 in. = 1J=
1 kcal =
1 kcal/hr = 1 kcal/m3 = 1 kg =
252.0 cal 1055 J 0.00890 cal/cm3 0.0373 MJ/m3 0.0003931 hp 0.2520 kcal/hr 0.2931 W 0.293 MW 0.003153 kW/m2 1.730 W/m-K 5.67 W/m2-K 0.5556 cal/g 2326 J/kg 1 cal/g-°C 4187 J/kg-K 0.003968 Btu 4.187 J 3.687 Btu/ft2-s 41.87 kW/m2 241.9 Btu/ft-hr-°F 418.7 W/m-K 1.80 Btu/lb 4187 J/kg 1 Btu/lb-°F 4187 J/kg-K 2.421 lbm/hr-ft 100 centistokes 3.874 ft2/hr 0.3048 m 0.0929 m2/s 1000 kg/m3 62.43 lb/ft3 0.03613 lb/in3 33,000 ft-lb/min 550 ft-lb/sec 641.4 kcal/hr 745.7 W 2.540 cm 25.40 mm 0.000948 Btu 0.239 cal 1 W/s 3.968 Btu 1000 cal 4187 J 3.968 Btu/hr 1.162 J/s 0.1124 Btu/ft3 4187 J/m3 2.205 lb
1 kg/hr-m = 1 kg/m3 = 1 kW =
1 kW/m2 = 1 kW/m2-°C = 1 lb = 1 lb/ft3 = 1 lbm/hr-ft = 1m= 1 mm = 1 m2/s = 1 m-ton = 1 MW = 1 therm = 1W= 1 W/m-K =
0.00278 g/s-cm 0.672 lb/hr-ft 0.06243 lb/ft3 3413 Btu/hr 1.341 hp 660.6 kcal/hr 317.2 Btu/hr-ft2 176.2 Btu/hr-ft2-°F 0.4536 kg 0.0160 g/cm3 16.02 kg/m3 0.413 centipoise 3.281 ft 0.03937 in. 10.76 ft2/s 1000 kg 2205 lb 3,413,000 Btu/hr 1000 kW 100,000 Btu 1 J/s 0.5778 Btu/ft-hr-°F
Temperature Conversions ºC = 5/9 (ºF – 32) ºF = 9/5 (ºC) + 32 K = ºC + 273.15 ºR = ºF + 459.67
735 © 2011 by Taylor and Francis Group, LLC
Flames into roof tubes
Figure 1.13 Duct burner flame. (From Baukal, C. E., ed., The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001; Courtesy of John Zink Co. LLC.)
Figure 5.1 Flame impingement on the process tubes in the convection section of a process heater.
Outlets
Thermal probe Figure 1.46 Water-cooled bayonets inserted into floor of a test furnace. (Adapted from Bussman, W., Baukal, C., and French, K., “Variable Test Furnace Cooling,” presented at the 2005 Summer Heat Transfer Conference, paper HT2005-72012, San Francisco, CA, July 17–22, 2005.)
View port
Tubeskin
Tubeskin
Figure 5.2 Example of an infrared scan of process tubes in a heater.
© 2011 by Taylor and Francis Group, LLC
*> 900.0°C 895.2 875.4 855.1 834.3 812.5 789.0 762.6 730.8 691.3 648.9 610.4 573.2 530.5 *< 530.1°C Figure 5.17 Infrared thermal image of a flame in a furnace. (From Singh, P., Henneke, M., Jayakaran, J. D., Hayes, R., and Baukal, C. E., The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001.)
Figure 6.13 Example of a luminous flame. (Courtesy of John Zink Co. LLC.)
Figure 6.10 Example of a nonluminous flame. (Courtesy of John Zink Co. LLC.)
© 2011 by Taylor and Francis Group, LLC
Figure 6.14 Surface radiation in a furnace.
In convection section
In convection section
(in
ion
sit
po
8
W est
22 20 18 16 14 Depth 12 10 (in.)
Figure 7.68 NO emissions at locations before and in the middle of the convection section at various depths. (From Bussman, W., Baukal, C., Waibel, R. “Ghost NOx.” Presented at the Air & Waste Management Association’s 98th Annual Conference & Exhibition, Control #28, Minneapolis, MN, 2005.)
51
100 90 80 70 60 50 40 30 6 Before convection section
.)
52
(in
100 90 80 70 60 50 40 30 6 Before convection section
.)
48
53
22 20 18 16 14 Depth 12 10 (in.)
on
50
54
iti
52
55
po s
54
56
8
est
46 48 50 52 54 56 58
W
Ave. NO (ppmdv)
51 52 53 54 55 56 57 58
57
56
46
) (ppmdv) Ave. NOx (NO + NO2
58
58
Figure 7.72 Top surface is in middle of convection section, bottom surface before convection section. (Bussman, W., Baukal, C., Waibel, R. “Ghost NOx.” Presented at the Air & Waste Management Association’s 98th Annual Conference & Exhibition, Control #28, Minneapolis, MN, June, 2005.)
POI@100m
POI@100m
POI@100m
POI@50m
POI@50m
POI@50m
POI@30m
POI@30m
POI@30m
POI@1m
POI@1m
POI@1m
Point source Vert. area source Barrier Receiver Calculation area
> 60.0 dB(A) > 65.0 dB(A) > 70.0 dB(A) > 75.0 dB(A) > 80.0 dB(A) > 85.0 dB(A) > 90.0 dB(A) > 95.0 dB(A) > 100.0 dB(A) > 105.0 dB(A) > 110.0 dB(A) > 115.0 dB(A)
Figure 8.20 Predicted sound pressure field contour plots for a multipoint LRGO flare system. Left: first stage in operation; Middle: stages 1 and 2 in operation; Right: all stages in operation.
© 2011 by Taylor and Francis Group, LLC
Burner exit air velocity distribution burner #1 — after modification
68 F, 90% relative humidity
90 80 70
5000–6000 6000 5000 4000 3000
8000
2000
500
125
60 50 40 30
86 dB 80 dBA 80 dB 74 dBA 79 dB 64 dBA 31.5
Sound power level, dB
Noise spectrum at a distance 1500 ft Noise spectrum at a distance 3000 ft With atmospheric attenuation @ 3000 ft
2000 1000
Frequency, Hz
4000–5000 3000–4000 2000–3000 1000–2000 0–1000
0
Figure 8.40 Effect of distance on flare noise.
Figure 10.9 Exit air velocity distribution of burner #1—after modification.
Burner exit air velocity distribution burner #2 — baseline
Burner exit air velocity distribution of burner #1 — baseline
9000 8000 7000 6000 5000 4000 3000 2000 1000 0 –1000
8000–9000 7000–8000 6000–7000 5000–6000 4000–5000 3000–4000 2000–3000 1000–2000 0–1000 –1000–0
Figure 10.8 Exit air velocity distribution burner #1—before modification.
© 2011 by Taylor and Francis Group, LLC
8000 7000 6000 5000 4000 3000 2000 1000 0 –1000
7000–8000 6000–7000 5000–6000 4000–5000 3000–4000 2000–3000 1000–2000 0–1000 –1000–0
Figure 10.10 Exit air velocity distribution of burner #2—before modification.
Burner exit air velocity distribution burner #2 – after modification
6000 5000 4000 3000 2000
5000–6000 4000–5000 3000–4000 2000–3000 1000–2000 0–1000
1000 0
Figure 10.11 Exit air velocity distribution of burner #2—after modification.
(a)
(b)
(c)
Figure 11.4 Flame characteristics for dry recycle case with burner B: (a) coal particle trajectories, (b) temperature, and (c) oxygen distributions at combustor midsection. (From Chui, E. H., Douglas, M. A., and Tan, Y., Fuel, 82, 1201–10, 2003. With permission.)
© 2011 by Taylor and Francis Group, LLC
Figure 11.5 Isosurface of volatile matter evolution at 10% of the maximum level for dry recycle case with burner B (left) and overlaid with the boundary of fuel-rich zone (shown on right with color red for equivalence ratio = 1). (From Chui, E. H., Douglas, M. A., and Tan, Y., Fuel, 82, 1201–10, 2003. With permission.)
Figure 11.6 Equivalence ratio (left) and NO (right) distributions across combustor midsection for dry recycle case using burner B. (From Chui, E. H., Douglas, M. A., and Tan, Y., Fuel, 82, 1201–10, 2003. With permission.)
© 2011 by Taylor and Francis Group, LLC
0
Axial distance (m)
1
2
3 Down firing
Centerline distribution Burner : Furnace ID
4
B : 0.6 m C1 : 0.6 m B : 0.6 m 0
Burner: Furnace ID Temperature (°C) 800
1000
C1 : 0.6 m
0
8 16 24 32 40
20 O2 (Vol. % dry)
Figure 11.8 Oxygen distribution on furnace cross section and centerline for burners B and C1. (From Chui, E. H., Majeski, A. J., Douglas, M. A., Tan, Y., and Thambimuthu, K. V., Energy, 29, 1285–96, 2004. With permission.) B : 0.6 m
1200
1400
C1 : 0.6 m
1600
0
1800
Figure 11.7 Temperature distribution on furnace cross section for burners B and C1. (From Chui, E. H., Majeski, A. J., Douglas, M. A., Tan, Y., and Thambimuthu, K. V., Energy, 29, 1285–96, 2004. With permission.) Axial distance (m)
1
Peak 3950 ppm
2
3
Centerline distribution Burner : Furnace ID
4
B : 0.6 m C1 : 0.6 m B : 0.6 m
C1 : 0.6 m
50 300 550 800 1050 1300
0
1000 2000 NO (ppm dry)
Figure 11.9 NO distribution on furnace cross section and centerline for burners B and C1. (From Chui, E. H., Majeski, A. J., Douglas, M. A., Tan, Y., and Thambimuthu, K. V., Energy, 29, 1285–96, 2004. With permission.)
© 2011 by Taylor and Francis Group, LLC
0
Axial distance (m)
1
2
3
Centerline distribution burner : furnace ID
4
C1 : 1.0 m C2 : 1.0 m C1 : 1.0 m
C2 : 1.0 m
50 300 550 800 1050 1300 C1 : 1.5 m
C1 : 1.0 m
C2 : 1.0 m
Burner : furnace Temperature (°C): 800
1000 1200 1400 1600 1800
0
1000 2000 NO (ppm dry)
Figure 11.13 NO distribution on stacked furnace cross section and centerline for burners C1 and C2. (From Chui, E. H., Majeski, A. J., Douglas, M. A., Tan, Y., and Thambimuthu, K. V., Energy, 29, 1285–96, 2004. With permission.)
Figure 11.10 Cross-sectional temperature distribution for burners C1 and C2 in 1.0 m and 1.5 m ID furnaces. (From Chui, E. H., Majeski, A. J., Douglas, M. A., Tan, Y., and Thambimuthu, K. V., Energy, 29, 1285–96, 2004. With permission.)
0
1
Axial distance (m)
Temp 2
Hot
Dissimilar flow
3
Centerline distribution burner : furnace ID C1 : 1.0 m C2 : 1.0 m
4
C1 : 1.0 m 0
C2 : 1.0 m
8 16 24 32 40
0
20 O2 (Vol. % dry)
Figure 11.12 O2 distribution on stacked furnace cross section and centerline for burners C1 and C2. (From Chui, E. H., Majeski, A. J., Douglas, M. A., Tan, Y., and Thambimuthu, K. V., Energy, 29, 1285–96, 2004. With permission.) © 2011 by Taylor and Francis Group, LLC
Cool Figure 11.15 Furnace cross section colored by gas temperature, with flow pattern superimposed. (From Chui, E. H., Applications of CFD Modeling in Canadian Industries, Fourth International Conference on CFD in the Oil and Gas, Metallurgical & Process Industries SINTEF/NTNU, Trondheim, Norway, June 2005.)
Figure 11.18 Left: Visualization of burners in cell D based on virtual test results. Right: Photograph of burner in cell D. (From Boisvert, P. G., Majeski, A., Runstedtler, A., and Chui, E. H., Flame Visualization Based on Computational Fluid Dynamics (CFD) Results, The 2008 International Conference on Modelling, Simulation & Visualization Methods, Las Vegas, Nevada, July 2008.)
Figure 11.19 Net heat flux [W/m2] at pipe surfaces. (From Runstedtler, A., Chui, E. H., Majeski, A., Trottier, S., Brunet, D., and Van Doormaal, M. A., Modeling of a Refinery Furnace, American Flame Research Commit tee International Symposium, Livermore, California, October 2003.)
© 2011 by Taylor and Francis Group, LLC
Figure 11.20 Incident radiant heat flux [kW/m 2] at pipe surfaces. (From Runstedtler, A., Chui, E. H., Majeski, A., Trottier, S., Brunet, D., and Van Doormaal, M. A., Modeling of a Refinery Furnace, American Flame Research Committee International Symposium, Livermore, California, October 2003.)
Figure 11.21 Left: Net convective heat flux [W/m2] at pipe surfaces for Cells B and A. Right: Isosurface of temperature (1970°F) in the furnace. (From Runstedtler, A., Chui, E. H., Majeski, A., Trottier, S., Brunet, D., and Van Doormaal, M. A., Modeling of a Refinery Furnace, American Flame Research Committee International Symposium, Livermore, California, October 2003.)
50
y (mm)
40
30
10
20 x (mm)
30
Figure 12.4 Combined measurements of OH-LIF distribution (image) and velocity field by PIV (arrows).
Figure 16.6 Flames in refinery boiler may have inconsistent flame characteristics, as shown on furnace camera monitor (Section A and Section E). (Baukal, C. E., ed., The John Zink Combustion Handbook, CRC Press, Boca Raton, Florida, 2001.)
ωs
ωp
ωas Figure 16.7 Camera monitors trash burning on traveling grate in waste-to-steam boiler.
(ω)
(t)
Figure 13.10 Typical temporal and spectral profiles of the probe beams and the CARS measured signal.
Figure 16.12 Immediate information on current furnace conditions appears on monitors of furnace camera system. Operators can take corrective action when necessary.
© 2011 by Taylor and Francis Group, LLC
D(µm):
20
52
84
116
148
180
u(m/s):
–20
–10
0
10
20
30
(a)
Aerated injector 0
y/d0
10 15
Jet direction
5
Crossflow
20 Figure 19.5 Examples of the burner flame visual observations. (a) Primary combustion flame of two-stage forced internal recirculation burner.
25 30 0
5
10
15 X/d0
20
0 25 –1
0 Z/d 0
10
Figure 17.8 Drop size distribution and three-dimensional velocities near injector region. (Reprinted from Lee, J., Sallam, K. A., Lin, K.-C., and Carter, C. D., Journal of Propulsion and Power, 25, no. 2, 258–66, 2009b. With permission from Americal Institute of Aeronautics and Astronautics.) A
Top section
0
Temperature (°C) C
200
G
C
400
F
1020
A D
E
Y(mm)
B
1060
600
1080
1100 1120
800 1160
1180 1160
1180
1140
1000
1140
1120
Burner Reaction tube
1200
1160 1180
800
600
1140 1120
400
Figure 21.31 Temperature distribution within the test furnace installed with a single-type regenerative burner.
1180
200 X(mm)
© 2011 by Taylor and Francis Group, LLC
D
1040
1160
0
–200
C
F
200
A
CO concentration (ppm) C
200
A
Bottom section O2 concentration (%) C
E 400 Y(mm)
D
0
400
0 200 500
600
D
1000 800
Burner Reaction tube
1.5
Y(mm)
G B
A
Bottom section
0
600
2.5
800 1500
3000 3000 3000 2500 3000
1000
2000 2500
4.5 4.0 3.5
1000 1200
1200 800
600
400 200 X(mm)
0
–200
D
2.0
4.0 800
4.5
600
4.5 4.0 3.5
3.0 3.5
3.5 4.5
200 400 X(mm)
4.0 0
–200
Figure 21.34 Example of measured CO and O2 concentration in the test furnace.
Figure 22.4 Schlieren images of impinging conditions (7 mm gap) for a 4-port burner with a clockwise roll rotation of 5.50 m/s; Left image: Low flame power (309 W/cm2); Right image: High flame power (1570 W/cm2).
© 2011 by Taylor and Francis Group, LLC
14
Temperature (K) 2189 2173 2156 2140 2124 2107 2091 2074 2058 2042 2025 2009 1992 1976 1960
Up-web position (mm)
12 10 8 6 4
Temperature (K) 1971 1882 1792 1703 1613 1524 1435 1345 1256 1166 1077 987 898 809 719
12 10 8 6 4 2
2 0
14
0
1 2 3 4 5 Cross-web position (mm)
0
1 2 3 4 5 Cross-web position (mm)
0
Figure 22.7 False color image of temperature above a 6-port ribbon burner. (Left) Laminar, 7 mm gap, 0.785 m/s Roll Rotation, 309 W/cm2; (Right) Highly mixed, 4 mm gap, 0.785 m/s Roll Rotation, 923 W/cm2.
CO emissions at 26 kW
CO emissions (mg/kWh)
90 80 70 60 50 40 30 20 10 0
7.4 Premix
Figure 24.8 Furnace layout for testing of radiant tube performance in pilot-scale industrial furnace.
© 2011 by Taylor and Francis Group, LLC
8.1
CO2 (%)
Blue diffusive flame
9
10 Yellow diffusive flame
Figure 25.3 CO emissions with three types of gas burners coupled with the same 26 kW cast iron commercial boiler.
NOx emissions at 26 kW
Nox emissions (mg/kWh)
200 180 160 140 120 100 80 70 60 40 20 0
Limit class 5 EN 483
7.4 Premix
8.1
CO2 (%)
Blue diffusive flame
9
10 Yellow diffusive flame
Figure 25.4 NOx emissions with three types of gas burners coupled with the same 26 kW cast iron commercial boiler.
Z Y
X
Figure 26.6 CFD of improved temperature distribution.
Figure 26.4 CFD modeling of Duct burner.
Z Y
X
Figure 26.5 Temperature profile with mal-distribution. © 2011 by Taylor and Francis Group, LLC
Figure 27.5 Inside view of the small-scale Air Liquide pilot test furnace in operation. Thermocouple locations (red dots) and view of the burner under testing.
Air8 2650–2700 2600–2650 2550–2600 2500–2550 2450–2500 2400–2450 2350–2400 2300–2350 2250–2300 2200–2250 2150–2200 2100–2150 2050–2100 2000–2050
DOC8 2650–2700 2600–2650 2550–2600 2500–2550 2450–2500 2400–2450 2350–2400 2300–2350 2250–2300 2200–2250 2150–2200 2100–2150 2050–2100 2000–2050
Figure 27.21 Roof temperature profiles as measured by a grid of 20 thermocouples for air-fired operation (left) and oxy-fuel-fired operation (right).
Figure 28.20 Photo of a row of ground flares, with flame heights compared to those previously measured for single and dual burner tests. (Courtesy of John Zink Co. LLC)
Figure 28.11 Thermogram of a flare flame. (From Schwartz, R., White, J., and Bussman, W., “Flares,” in The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001.)
© 2011 by Taylor and Francis Group, LLC
0
(b)
(e)
0
20
–20
0
Relative radiation intensity
(c)
(f) 40
0
0
20
–20
0 0
20
40
60
X (mm)
0
20
40
60
IR spectrum
–20
Visible spectrum
20
UV spectrum
0
Pulverized coal flame Liquid fuel flame Non-luminous flame
Violet Yellow Red
(d)
Z (mm)
(a)
1
2 3 Wavelength (1 x 10–6 m)
5
Figure 30.8 Spectral emission of radiation from luminous and nonluminous flames. (From Baukal, C. E., The John Zink Combustion Handbook, Boca Raton, FL: CRC Press, 2001.)
Figure 29.7 Visualization of flow patterns of the plane of symmetry: (a) R = 0.04 (Uj = 0.8 m/s, Rej = 909), down-wash mode; (b) R = 0.16 (Uj = 1.6 m/s, Rej = 1818), crossflow dominated mode; (c) R = 0.70 (Uj = 3.3 m/s, Rej = 3749), cross-flow dominated mode; (d) R = 2.47 (Uj = 6.2 m/s, Rej = 7043), transitional mode; (e) R = 4.32 (Uj = 8.2 m/s, Rej = 9315), jetdominated mode; and (f) R = 12.60 (Uj = 14.0 m/s, Rej = 15,904), strong jet mode. Flame. (Reprinted from Huang, R. F., and Wang, S. M., Combustion and Flame, 117, 59–77, 1999. With permission from Elsevier.)
Propane at 60, 000 lb/hr
Heat radiating from flame to surroundings (referred to as flare radiation)
Figure 30.2 Flare firing propane at 60,000 lb/hr corresponding to a heat release rate equal to 1.2 billion Btu/hr.
© 2011 by Taylor and Francis Group, LLC
4
