CI SfB 1976 182 (—1)
rid
foundations
and substructures
This document
itains
104
pages
•
BuildingResearch Estab...
486 downloads
2170 Views
1MB 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
CI SfB 1976 182 (—1)
rid
foundations
and substructures
This document
itains
104
pages
•
BuildingResearch Establishment Report
Bridge foundations and substructures This reportwas prepared for the BuildingResearchEstablishmentby
Dr Edmund C Hambly,Consulting Engineer under the direction of
DrJ B Burland, Head of GeotechnicsDivision, BRS.
Department ofthe Environment Building Research Establishment
London:Her Majesty's Stationery Office
Full details ofall newBREpublications arepublished quarterly in BRE NEWS. Requests for BRE NEWS or for placingon the mailinglist should be addressedto: Distribution Unit Publicationsand Publicity Unit BuildingResearchEstablishment Garston, WatfordWD2 7JR
ISBN 0
11 670761 5
© Crown copyright 1979 First published 1979 Limited extracts from the textmay be reproduced provided thesource is acknowledged.Formore extensivereproduction, please write tothe Publications Officer at the Building ResearchEstablishment.
CONTENTS
INTRODUCTION 1 Survey ofcurrentpractice 2 Objective ofreport 3 4 5
Typesofbridge Textbooksand references Acknowledgements
CHAPTER1 OVERALL PERSPECTWE 1.1 Designprocess and priorities 1.2 Bridge foundations in contextofwhole contract 13 Design for construction and maintenance CHAPTER 2 THESITE 2.1 Site reconnaissance and desk study 2.2 Influences ofsite on construction 23 Subsidence 2.4 Soils investigation 2.5 Groundwater 2.6 Advanced contract pile and plate bearing tests CHAPTER3 CATALOGUEOF SUBSTRUCTURES 3.1 General considerations 3.2 Wallabutments 33 Openabutments 3.4 Strutted, portal and box structures 3.5 Abutments above highly compressiblestrata 3.6 Wingwalls 3.7 Othersubstructures 3.8 Piers CHAPTER4 CHOICE OF FOUNDATION 4.1 Choice offootings or piles 4.2 Differential settlementcriteria 43 Groundimprovement CHAFFER 5 SPREAD FOOTINGS 5.1 Global behaviour 5.2 Loading on substructures and foundations 5.3 Undrained and drained behaviour 5.4 Bearing pressure 5.5 Movements
CHAPTER 6 ABUTMENTEARTH PRESSURESAND STABILITY 6.1 Active and at-rest earthpressures 6.2 Passive pressure 63 Skeleton ('open' or 'spill-through') abutments 6.4 Buried structures 6.5 Stability ofretainingwalls
v V V
Vi Vii
1
2 3
5 6 9 10 13 15
17 18
21 23 26 27 30 31
33 36 37
39 41
42 43 45
47 51 51
52 53
CHAPTER 7 PILEFOUNDATIONS 7.1 Global behaviour 7.2 Selection of pile types 7.3 Contractdocuments 7.4 Pile behaviour 7.5 Raking piles 7.6 Pile groups 7.7 Pile tests CHAPTER 8 DETAILS 8.1 General comments 8.2 Excavation shapeandshear-keys 83 Substructure form and reinforcement 8.4 Construction and movement joints 8.5 Topof abutment 8.6 Drainage and waterproofmg 8.7 Backfill and run-onslabs
55
58 59 60 62 63 65
67 68 69
70 71 72 73
CHAPTER9 PERFORMANCE,MAINTENANCEAND REPAIR 9.1 Monitoring ofperformance 9.2 Inspection and maintenance considerations during design 93 Repairs
75 76 77
CHAPTER 10 ADVICEAND INFORMATIONNEEDS 10.1 Disseminationofinformation 10.2 Advice andinformation needs
79
APPENDIX A SITE INVESTIGATIONINFORMATION
83
APPENDIX B COMMENTS ON PILEINSTALLATION
85
APPENDIXC ARRANGEMENTOF PILES IN GROUP
88
APPENDIXD DEPARTMENTOF TRANSPORTMEMORANDARELATED TO THIS REPORT
89
AUTHORS INDEX
90
INDEX
91
iv
80
INTRODUCTION 1 SURVEYOF CURRENT PRACTICE This report on Bridge Foundations and Substructures summarisesthe findings ofa survey ofthe attitudes of bridge designersin the UnitedKingdom. The objective ofthe survey was to gainsome idea ofthe philosophies adoptedby engineers in their choices of different types of foundations and substructures and of different methods ofanalysis. The reportwas also expectedto highlight those areas where furtherguidance and research are required, and wherefInancial savingsmightbe effected. The survey initiallyinvolved discussions during 1975/76 with about 120 experienced engineers within46 organisationsacknowledgedbelow. During 1977 a considerable amount of additional advice and comment was added as a result ofdetailedcriticism ofapreliminary draftby the Steering Committee. The subsequent Draft Commentary on Current Practice in Design of Bridge Foundations and Substructures was published by the BuildingResearch Establishment to stimulate public comment and it formed the text for a Seminar held at the BuildingResearch Stationin October 1977attendedbyabout 150 engineers. A considerable quantityof further comment, advice and criticism was collected at the Seminar and received in writing from bridge designersin the UK and abroad. This information has also been incorporated in this report.
2 OBJECTWE OF REPORT It became evident earlyon in thesurvey that for themajority ofbridges there are too many independentand individual factors influencing design decisions to enable many useful generalisationsto be madeabout the appropriatedesign or methodofcalculation. However, it was also evidentthat engineers in differentorganisations do have a similar attitudeto many aspects ofdesign, and it was possible to collect a very considerable amountof advice on good practice. As a result the report has concentrated on reproducing this advice and settingout the wide range offactors whichinfluence selection ofdesign, and method of calculation. At the same time anindication is given ofthe varietyof solutions adoptedand their problems. Throughout the reports the statementsof practiceare basedon the commentsofone or more of the engineers interviewed on the survey.
It is hopedthat the reportwill proveto be a usefulreference forboth experienced bridge designers and newcomers. However concernhas beenexpressed that such a compendium ofexperience might be thought a substitutefor theexperience itselfand thatit might be thoughtpossiblefor the inexperienced to producesatisfactory designs using it alone. Thisthey will not be able to do, and the report emphasisesin many places the needto use
judgementbased on experience.
3 TYPES OF BRIDGE This report has concentrated on the foundations and substructures ofsmall and medium size road bridges over roads, riversand railways. This group represents the majority ofbridges built.Much of the discussion is relevant to other bridges (and other works)but in general these require furtherattention.
v
4 TEXTBOOKSANDREFERENCES
The textbookslisted below are recommended by bridge designersinterviewed during the survey. They are
referred to in the report bythe authors'names. A few other specialist references are given in the various sections, and in general these are from published books.No attempt has beenmadeto include comprehensive references ofresearch papers,since it hasnot beenpossibleto studythem critically, and it is probablyeasier for designers to make directuseofcommercial index systems, such as Geodex.
Soilmechanics and foundations Terzaghi K and PeckR B. Soil mechanics in engineering practice. John Wiley,New York, 2nd ed, 1967. Tomlinson M J. Foundationdesign and construction. Pitman, London,3rd ed, 1975. Lambe T Wand Whitman R V. Soilmechanics.John Wiley,New York, 1969. Tschebotarioff G P. Foundations, retaining and earth structures. McGraw-Hill,New York, 1973. Peck R B,HansonW E and ThornbumTH.Foundationengineering.John Wiley,New York,2nd ed, 1974. LittleA L. Foundations. Arnold,London,1961. Bowles, 3 E. Foundationanalysis and design. McGraw-Hill,New York, 1969. NAVFAC DM-7(1971).Design manual —soilmechanics, foundations and earth structures, US Naval Facilities EngineeringCommand, WashingtonDC.
Earthretainingstructures HuntingtonW C. Earthpressures and retaining walls. John Wiley,New York, 1957. (Thisbookis out of print, but it provides muchwise advice and is worth borrowing from a libraryforinspection.) Textbookson concrete with usefuladvice on foundations FaberJ and Mead F. Reinforced concrete.E & FN Spon,London,1967. Reynolds, C E and Steedman J C. Reinforced concrete designershandbook.Cement and Concrete Association, London,8th ed, 1974. ChettoeC S and Adams H C. Reinforced concrete bridge design. Chapman and Hall, London,2nd ed revised 1951.(Thisbook,which was published before the war, is out ofprint and out of date,butit provides some wise advice and isworth borrowing from alibraryfor inspection.) Military EngineeringVol XIV. Concrete, Part II. HMSO,London,1964. Undergraduate textbooks Capper P Laud Cassie W F. The mechanics ofengineeringsoils. E& FN Spon, London,5th ed, 1969. Capper P L, CassieW F and Geddes D. Problems in engineeringsoils. E & FN Spon, London,1971. Scott C R. Soilmechanics and foundations. Applied SciencePublishers, London,2nd ed, 1974. Simons N E andMenziesB K. A short course in foundations engineering.IPC Science and Technology Press, 1975.
i
Geology Blyth F G H andde FreitasM H. A geology for engineers. Arnold, London,6th ed, 1974. Codes andPractice andspecification CivilEngineeringCodeofPractice No 2 (1951).Earth retaining structures. British Standard Code ofPracticeCP2004, 1972. Foundations. British StandardCodeofPracticeCP2001, 1957. Site investigation. Revised 1976 draft reference 76/11937. British StandardBS 153, 1972. Specification for steel girder bridges. British StandardBS 5400, 1978. Steel,concrete and composite bridges. Department ofTransport. Specification for road and bridge works. Her Majesty's StationeryOffice. 1976. Department ofTransport. Notes for guidance on the specification for road and bridge works. Her Majesty's Stationery Office, 1976. Department ofTransport Memoranda relatedto this report are listedin Appendix D
vi
5 ACKNOWLEDGEMENTS
The projectreceived valuableassistance and detailed guidance and critism from an adhoc Steering Committee comprising:
P Effiott
Department ofTransport (BES) Chairman
J B Boden
Transport and RoadResearch Laboratory
L Clements
Scottish Development Department
R W Cooke
BuildingResearch Station
S B Crutchiow
Department ofTransport (E Int)
H G Frost
Department ofTransport (E hit)
H A Hall
DTp,North Western RoadConstruction Unit
J B Holt
G Maunselland Partners
AKnowles
DTp,Midland RoadConstruction Unit
WI J Price
Transport and RoadResearch Laboratory
Dr J E Spindel
British Railways Board
Dr WM C Stevenson
Department ofthe Environment for N.lreland
J P Taberner
Costain Civil Engineeringlimited
V G Towson
Department ofTransport (BES)
In addition many comments and useful criticisms were received during the draftingofthe report from: T A Rochester
Department ofTransport, Bridges
F Shields
Engineering(Design Standards) Division
During the survey ofcurrentpracticemeetings were held with engineers in the organisations listedbelow. Their interest,assistance and valuable advice are gratefully acknowledged. Department ofTransport: Bridges EngineeringDivisions EngineeringhiteffigenceDivision Highway EngineeringComputer Branch North Western Road Construction Unit, Head Quarters North Eastern RoadConstruction Unit, Head Quarters Midland Road Construction Unit, Head Quarters Midland Road Construction Unit, Stafford Sub Unit Eastern RoadConstruction Unit, HeadQuarters SouthEastern RoadConstruction Unit, HeadQuarters SouthWestern RoadConstruction Unit, Somerset Sub Unit SouthWestern RoadConstruction Unit, Devon Sub Unit Building Research Establishment Transport and Road Research Laboratory
Scottish Development Department Department of the Environment for NorthernIreland British Rail (5 Regions) London Transport Devon CountyCouncil Kent CountyCouncil Kent CountyCouncil Highways Laboratory West Yorkshire Metropolitan CountyCouncil Construction Industry Research & Information Association Ove Arup and Partners Blyth and Blyth CostS CivilEngineeringLtd Robert Curtis and Associates R M Douglas Construction Ltd Leonard Fairciough and Co Ltd W A Fairhurstand Partners FoundationEngineeringLtd Frankipile Ltd Freeman, Fox and Partners Kier Ltd Lind Piling Ltd G Maunsell and Partners Sir Alfred McAlpine& Son (Southern)Ltd Scott,Wilson, Kirkpatrick and Partners Dr FW Sherrell Simplex Piling Ltd Soil Mechanics Ltd TB!Group ServicesLtd Tarmac Construction Ltd Taylor Woodrow Construction Ltd Thorburn and Partners M Tornlinson Esq SirOwen Williams and Partners
J
J
vni
CHAPTER 1 OVERALLPERSPECTIVE
1.1 Designprocess and priorities l'his reporthas beenset out to follow as far as possiblethe designprocess. The followingpages ofthis Chapter report the statementsof engineers interviewed during the survey (see Introduction)concerning the design priorities and the relevance ofbridge foundations withinthecontext ofthe whole contract,and general comments relating to economy in construction. Chapters 2, 3 and4 discussthe assessment ofsite and factorsthat influence designers' choices ofappropriate substructure and foundation. Chapters 5, 6 and 7 then reportcomments on the various methods used to design the chosenfoundation. Finally Chapters 8 and 9 complete the design process with comments on details and anticipation ofmaintenance. Although the survey was concerned solely with the design and construction of bridge foundations and substructures, several ofthe senior engineers interviewed prefaced their remarks with a numberofgeneral comments on attitudes,organisation and training ofdesigners, which they considered as prerequisite for gooddesign. Some ofthese recommendations are summarised in thefollowing table.
Comment
Aspect Overallperspective
Maintain a sense of proportionofthe cost of the foundations, and ofthe deck, and consider the influence oftheir construction on the progress and costofthe rest of thecontract.(See Section 1.2).
Construction simplicity
Aim to design works which can be constructedeasily and comparatively quickly and which do not require many specialist trades and processes. (See Section 1.3).
Clarity of concept
A clear appreciation of all the main influences on the interaction of a foundation
with the structure and soil environment is much more importantthan undue refmement of detailed calculations. When predictions cannot be exact make broad judgements of possible extremes of performance based on careful interpretations
ofsoils data. Autonomyof designteam
Retain responsibifity for decisions for all parts ofthe designwithin the design team: ie use specialistsfor advisersbut not formaking choice of designunless fully integrated into the team so that they are aware ofthe effectsoftheir advice on the rest ofthe project.
Training ofbridge designers
Train bridge designersto have a broad understanding of the behaviour of foundations and structureand ofthe construction and costsofboth.
1
1.2 Bridgefoundations in context of whole contract Many engineers on the survey expressed the view that in orderto obtain overalleconomy for a scheme it is essential to maintain a sense ofproportionof the relative importance and cost ofthe various parts withinthe context ofthewhole project. The table below reports comments related to thesignificance ofbridge foundations and substructures withinthe context ofhighway and bridge contracts. Comment Bridges within highway contracts
Most of the bridges built in the last few years have been part of roadworks contracts. The cost ofmaterials in the bridges is usually less than halfofthe cost of the bridges, and these often comprise only about a quarterof the total contract tender. Consequently savings of materials in bridges by a refinement of detail may not be significant in comparison with the contract total. However construction of some bridges can seriously interrupt the progress of other much more expensive parts of the contract, such as muck-shifting. For such bridges simplicity and speed of construction can be importantdesignconsiderations and may affectthe choice of structural form more than reduction ofmaterial content.
Bridge
Several design organisations pointed out that bridge substructures frequently cost more than the superstructures. When piling is required the cost of the pilesalone is often comparable to that of the deck. If overall economy is to be obtainedfor the bridge it is importantto developthe design of all parts simultaneously and not to give one part precedence by designing 'from top down' or 'from bottom up'. Some engineers feel that in the past piles were occasionally adopted unnecessarily, at significant expense, as a result ofthedesignersfirst designingthe decks with unduly tight limits set on differential settlement, and then having to provide piles to obtain sufficiently stifffoundations.
substructure and structure
Comparison of designs
Severaldesignersand contractors pointed out the difficulties ofidentifying the true cost ofany part ofa contract.Engineerswarnedof the dangers ofdrawing too fme conclusions about the relative merits of different designs purely from the basis of rates from contractors'tenders.Tenderers do not always try to make the ratesfor any part reflect the true costofconstruction ofthe part or its complexity or interference with the rest ofthe contract.Tenders haveto be prepared in avery short periodand contractors often fill in all the rates on the basis of pastexperience. Then when this procedure is complete estimators pay attention to the more special features of the works including, the overall programme, the complexity of construction details of the structures, and their effect on the bridge programme and overall programme. At the same time they consider other commercial factors (suchas the design organisation, the locationand competition) and they take accountofall these pointsin the spread ofon costs on the net rates, or in a balancing item, or both. Consequently when a designer wishes to compare estimates for different designshe needs to makesimilar appraisalsofconstruction complexity and interference to programme in addition to comparisons ofmaterial costs. (Although billed rates do not generally indicatethe true distribution of costwithin a contract,global estimates of costs based on themhaveproved sufficiently accurate and very useful for the overall planning of schemes. Several organisations maintain up-to-date costing systems for this purpose.)
2
13 Design for construction and maintenance Several designers and contractors on the survey explained how significant savings can be achieved without detrimental effectson performance when attention is paid to methodsof construction throughoutthe design process. The table below summarisestheir comments and suggestions. Comment
Speedof construction
Simplicity of construction
Speed ofconstructionis usually essential to achieve economy. Consequently the investment of considerable design effort may be justified to make details and procedures as simple and repetitive as possible. Opportunities for saving time can be seized during construction if the programme is flexible, as occurs when structures are designed so that the elements and details can be completed in a variety of sequences,and where restrictive procedures are not necessary. The needforsimplicity of construction, particularly ingroundworks, was emphasised many times during the survey. Simple works not only lend themselves to rapid construction but alsoreducethecontractor's dependence on many specialist trades and processes. Complex details in foundations can cause delays at the start ofa project which have consequences throughoutthe contract.An awkward detail can have as much influence on construction costsas the form ofthe structure. Furthermore complicated details require a management effort out of all proportionto their importance. Even on relatively simple structures the work force needto complete two or threeofa particular designbeforeachievinga rapid turnover. The aspects ofsimple construction most stressed were: 1
Use materials which are readily availableandeasy touse.
2
Construct as much as possiblewith plant at existing groundlevel to easeaccess and use natural crust and drainage.
3
Designexcavations with a level bottom.
4
Use foundations with simple shapes and details (which can, if necessary, be adjusted easily to suiteunforeseen groundconditions).
5
Form all surfaces horizontal or vertical, unless inclined or skew surfaces are essential for structural reasons or appearance.
6
Anticipate re-use offormwork without cuttingor difficult alignment.
7
Where possible design structures
construction.
to be stable withoutproppingat all stages of
8
Fix reinforcement and place concrete in one plane at a time.
9
Use medium size reinforcement, avoid both small bars and alsoheavy large bars which can be difficult to bend and fix accurately.
10 When wallsare likely to be about 450 mm or wider, make themwide enoughfor a man to getinside reinforcement, and anticipate singlepour construction. (Concern has beenexpressed by a few designersthat while an increased adoptionof more straightforward designs might lead to overall economy, the benefit would be felt only by the contractors and not by the owners since billed rateswould not necessarily be adjusted. However other engineers argue that over a period of time contractors'ratesin general alter to reflectchanges in their average costs, while the level ofprofitsis largely influenced by competition.)
3
1.3 (contd) Design for construction and maintenance Comment
4
Site access and restrictions
See Section 2.2. Some designershave on occasions foundit helpful to discusswith a potential contractorearly in design the problems of access of heavy plant, and deployment on a restrictedsite.
Unforeseeable ground conditions and inclement weather
See Section 23 concerning soils investigation. Several designersadvised the need for conservative designsfor foundations because groundconditions can differ significantly from those predicted from the soils investigation. Furthermore, as explained in Section 8.1 and 8.2, construction offoundations can be severely hampered by inclement weather. Foundationswhich can be constructedquickly and which canbe adaptedto suit unexpected ground conditions are not only likely tobe more economic but they could alsoperform betterifthegroundis prevented from softeningfrom long exposure. Once construction is out of the ground progress is much more predictable (and faults can be corrected much more easily).
Temporary works
The costoftemporaryworks can influence the choice ofbridge type. Ifthe false. work requires expensive foundations, such as piles, then it may be that a form of structurewhich does not require such temporary support is more appropriate. Small details on the design can significantlyhelp (or hinder)the fixing oftemporaryworks.
Construction sequence drawings
Although the contractoris usually responsible for choosing the method ofconstruction, several designersand contractors recommended that the designershould provide at least one suggestion for the method of construction when the design is novel or when the sequence is dictatedby the design. Under some circumstances such information may be required to comply with Clause 14 (5) ofthe 5th Edition of theICE Conditions ofContract.
Maintenance
Considerations for future maintenance that can be made during design are discussed in Chapter 9, while drainage details are discussedin Chapter 8.
CHAPTER2 THE SITE
2.1 Site reconnaissanceand desk study Several designersstressed the importanceofmaking the preliminary site reconnaissance and desk study right at thestart ofthe design process before anydecisions have beentakenabout type ofstructureand foundation.Such studies not only provide alarge proportionofthe essential information required for the design, so reducing the risk ofexpensive reappraisalslater, but they also enable effective planning ofthe detailed soils investigation. The Appendices ofCP2001 incorporate several useful lists of the information that can be obtained. Comment Site reconnaissance
The site reconnaissance ofa bridge site is likelyto include: 1
2
Assessmentofproblems of constructing bridge on the site,such as: access, obstructions, traffic, etc.
Study of local ground features which indicate soil characteristics and strata, natural drainage andgroundwater regime: ie exposed soil stratain streams, vegetation etc.
Desk study
3
Survey of existing structures on site and assessment of their influence on the ground andofthe groundon them.
4
Survey ofconditionand type ofneighbouring structures and engineeringworks.
5
Enquiries with the local residents, builders and statutory undertakers.
6
Trial pits with tractor mountedbackhoe orhand auger.
The desk studywhich proceeds simultaneously with the site reconnaissance generally includes: 1
Comparison ofmapsand surveys ofsite, with profiles ofproposed works.
2
Information on previous uses of site (oldmapsuseful).
3
Surveysofexisting and abandoned services.
4
Analysisofgeologicalmapsand memoires ofthe Institute Sciences.
5
Inspection
6
ofGeological
ofaerial photographs including stereopairs. Discussionswith the Institute of GeologicalSciences andlocal organisations on availableborehole records.
7
Discussionswith water authorities, National Coal Board and so on, if appropriate.
Some designers,when unfamiliar with the geology of the region, carry out these studies in consultationwith an engineeringgeologist with local experience.
5
2.2 Influences ofsite on construction The fundamentaldesign decisions ofform ofstructureand substructure and choice of foundations are based on considerations of what can be built and howit can be built most economically, as well as how it wifi perform. The followingtables list some ofthe site characteristics which are found to affectconstruction. Comment
Problem
to be constructedto an isolated bridge site. A contractor that about halfthe bridges on amotorway route have simple access. reported only Use of the route trace is often not practical because of the dominating demands of muck-shiftingin goodweatherand because ofimpassability in bad.
Access
Special access may have
Restricted space
On restrictedsites the choices of foundation and bridge type are often controlled by what it is possible to build in space availableand what plant can be used.
Piles
Piles may be needed to transmitheavy loads down throughrestricted space. Accessfor piling plant is a major problem. Large bored piles require roomfor boring rig, crane with casing and concrete trucks. Clearance of more than 15 m is often required, and in particular directions for rakingpiles.
It is often advantageous to carry out piling from existing ground level to take advantage of access,andthe probable hard crust for working ofplant.Pile trimmingmay be necessary to a greater extent. There are some situations where fill is to follow whenpile caps maybe cast above or at ground level with advantage.
Temporary works for piling over excavations or bank slopes can be very costly.
Poor ground
It is often advantageous to constructas much as possible from existing groundlevel, because the ground usually has a naturalcrust and established drainage. A viaduct over bad ground with poorground level access may be best constructed over-hand. However access along the structurecan then also be difficult.
Headroom
Overhead cables seriously restrict the height and choice of plant.It may be impossible to use piling rigs, cranes, concrete pumps, etc. It has on several occasions beenfound economic for constructionto be sub-dividedinto small parts,sometimes prefabricated. The Factory Inspectorate can serve a prohibitionnotice to limit procedures. Consequently, if it is possible, it maywellbe worth tryingto avoid siting a bridge under 400 KV cables.
6
Traffic interference
Interaction of construction with existing traffic on roads, railways and rivers can cause major disruption to the programme. Ifland is availableit maybe cheaper to spread works rather than suffer expensive delays.
Services
Congested services considerably complicate construction of foundations. Diversions are frequentlylate, and diversion of unexpected services can cause very expensive delays. On exceptionally congested sites it may be economic to prepare altemative or adjustable details which can be selected to suit conditions as found.
Groundwater
Lowering groundwater in excavations or piling can cause settlementof adjacent structures (see Section2.5).
Noise
Designers and contractors pointed out the need to anticipate the effects ofnoise restrictions on piling operations. In special circumstances the restrictions could influence the choice of type offoundation or pile.
2.2 (contd) Influences ofsite onconstruction Problem Bridgesoverrailways General
Possessions
Footings
Comment
The design ofabridge over arailway has to be carriedoutin close cooperation with therailway authorityfrom thestart. The requirements of therailway authoritycanradically affectconstruction procedures, timing and cost. Ifa bridge over arailway is part of alargeprojectit can be advantageous to buildit in anadvancecontractso that the specialconstruction problems do not obstructconstructiontrafficaccess for muckshifting or disruptthe programme of the maincontract. Severaldesignersemphasised theimportance ofconcentrating designeffort into choosing the appropriate type ofsubstructureandstructureto facilitate construction under the particular constraints, rather than concentrating on refmement ofdetail. The influence and restrictions of the railway on construction depend on how busy the line is and what possessions are possible. Spread footings usually have to be placed welloutside 450 line down from edge rail,
to avoid onerousrestrictions and expensive temporaryworks. Restrictions are often alsoplaced on the length ofexcavation that can be placed alongside the track. This can affect the length ofbase pours and reinforcement detailing. Suspect groundmay warrantspecial cautionas additional excavation or overdig maycut 450 line and so require restrictive and expensive temporary works.
Accessto site
Contractors sometimes have difficultyobtaining access to both sides of track, and have only restricted access across.
Clearance
Lackofclearance for working space beside trackcan beveryrestrictive and construction maywellbe simplified ifabutmentsare placed wellbehindminimum clearance allowed. Construction ofabutmentsincuttingscan be particularly difficult, and undercertain circumstances itcanbecheaper to have anincreased span tobankseats atthe top ofslopea
Forordinary constructiona space ofabout 2m is needed betweenthe fence and awall. A small working space is possibleifover-hand or simple constructionis used,such as brickwork acting aspermanentshutter to mass concrete pouredin shallowlifts.Precast concrete, including post-tensioned H-blocks,has beenused for facing abutments. Piles
Methods ofpiling are likely to be severely restricted. Overheadelectrification is likelyto preclude all typesoftall piling plant.Driven piles may not be accepted because of risk ofplant falling on line. Theremaybe insufficient clearance forlargeplant particularly for raking piles. Any risk of ground movement, particularlybeneath controlrods,is opposedby railway authority. Displacementpilescan cause heave oftrack ifdriven in clay,or they can cause settlement if driven in gravel.Yet small cross-section driven piles can cause less disturbance on some soils thanbored piles. Bored piles have advantage that they can often be placed by small tripod rigs. By placing pile caps as high as practical other trackside works are reduced.
Substructure
Ifcranes cannot be used then shuttersand reinforcement may needto be prefabricated intoelements whichmen can handle. Speed and economy ensueifconstructionin small parts can be programmed to provide continuous workload for men involved.
Structure
Decks are usually ofprecastor composite construction with beams placed during possession. (Aesthetic refmement such as edge cantilever, can often be omitted).Decks acting as props to strutted abutements are considered by some designersto be particularly inappropriate, because ofextreme difficulty ofplacing deckbeforebackfill, and virtual impossibility ofreplacing deck.
7
2.2 (contd) Influences ofsite on construction Problem Bridge overrivers General
Comment
The water authorityand other interestedbodies,such as navigation authority and conservancy board,must be consultedfrom the start of design. Their requirements, which may be stringent, can affectthe spans, headroom, positionand shape ofsubstructures, clearance on banks for dredging plant and so on. A bridge whose piers or abutmentsare withinthe river can cause flooding for some distance upstream, as well as affecting scour and/or accretion upstream and downstream. It mayalsoaffecterosionand general stabifity ofthe river. Predictions can be uncertainand difficult to question.
Hydraulic Survey
The Hydraulic Survey needsto be planned by an engineer experienced in the effects of bridge crossings on river behaviour, and it requires as much care as the Soils Investigation. The survey shouldbegin whenthe crossingis first contemplated as it mayaffect the locationofthe crossing, and because the longest possiblerecordis required.
Floods
Floods and flood debris can pose a greaterthreat during construction ofa bridge than during its life. The height and size ofcofferdams maybe restrictedby the river authorityto reduce their influence on floods. Some designers stressed the need to obtain information on likely flooding during construction at the design stage, and later make this availableto tenderers (withoutinterpretation).
Construction
Simplicity of foundation has been found on many occasions to be more important than material economy. Construction over water is much more expensivethan on dry land. It can be economic to divert rivers andbuild bridges in the dry. Otherwise foundations have usually been constructed in cofferdams (carehas to be takenabout position ofcofferdam if raking piles are used). Caissons are unlikely to be used for a small bridge. In recent years large boredpiles, possibly with permanentthin steel skin,have beenplaced in casing from above water level. Pile caps have also been constructed above low water level. Choice oftype of pile maybe governed by what plant can be used and is available.
Access and clearance
Access for plant maybe restrictedto one bank. The benefits of overspanningsmall riversare similarto those for railways.
Scour
Scour is the greatest single cause ofbridge failures. Foundations need to be belowthe estimated lowest level ofscour or bed protection works(gabions, rip rap, etc). Scour can be rapid in times of flood,and even in large slow riversit can go deepand change unpredictably. Scour can be very much worse during flood turbulence than when it can be measured. Piers with streamlined edges cause less eddies and so less scourthan circular columns. However there may be little benefit if the direction of flow alters by a few degrees during floods. A smooth faced footing accelerates side flow and invites scour. Marina mattresses, gabions and rip rap slowdown flow and silting fills their voids; they can be inexpensive. Hard clays and shales softenwith time and erode. Sheet piling used for cofferdams is often left in for scourprotection. When a serious risk ofscouris identifieddesignersshould consult an engineer experienced in river hydraulics at the earliest possiblestage.The Hydraulics Research Station,Wallingford, Oxon,offer an advisory service.
Hidden river beds
River beds change their courses and a flood plane can have several hidden courses, possibly with flowingwater. They can often be located by series oftrial pits. Abutments maybe founded on the sides of a buriedvalley, where positions ofboreholes are critical.
Drawdown
Substructures and embankments subject to flooding are designed to drain rapidlyor
be impermeable, in which case stabilityduring drawdown canbe critical. References _________________
8
C R Neil! (1973). Guide to bridge hydraulics. Published by University of Toronto Press for Roads and Transport Association of Canada. Department of HighwaysOntario(1967).Prediction of scour at bridges, ReportRR1 15.
2.3 Subsidence In regions of potential subsidence the subsidence dominates other design restraints. Particular experience is required beyondscope indicated below. Problem
Coalmining new workings.. old workings...
Comment The majorityofsubsidenceproblems in the UK relate to coal mining. The National Coal Board have experience in prediction and controlofsettlementdue to currentlong-wall mining, but often cannot specify directions ofadvanceofcoalfaces in future. Prediction ofsettlement and collapse due to upward migration ofvoids of old pillarand stall workings is usually not possible. If risk is unacceptable then old workings, and shallow seams that could have beenworked, are grouted. The depth and spacingofgrout injectiondepend on extent of old workings,and depth and soundness of rock above (an example is grouting to 20 m or 30 m at 6 m spacing below roadworks, 3 m spacing below structures, and 1.5 m spacing below special foundations). Grouting can only be done properly by a competentcontractor,and can be veiyexpensive. Water in workings can makegrouting ineffective, and it can be difficultto controlthe flow ofgrout (it is usually progressed up dip of coalseam).It can also be difficult to check that grouthas beensuccessful.
Salt mining
Salt mining is muchless predictable than coal mining, and can also cause verylarge subsidence.
Swallowholes
Swallowholes (generally at surface ofchalkand limestone) can be difficult to identify (even with geophysicalmethods)unless they are visible to the eye or on aerial photographs. When found they are usually filled,grouted or bridged.
Structures
Structures are usually designed to accommodate distortiondue to subsidence by articulating or flexing. Simplicity and predictability ofperformance are essential. It may be easier to predict theperformance of a simple flexible structurethan a three-dimensionallyarticulated structurewith complex mechanism. Many bridges havesimply supportedtorsionally flexible decks. Jacking pockets are generally provided. Considerable care is given to design of joints, rockers and bearings to accommodate thelarge movements. Parts are tied together to prevent separation. (For this reason some designers use strutted abutments for small bridges to ensure compression at joints, but such structures can be subjected to very high earth pressures during the compression phase ofa subsidence wave.)
Substructures
Substructures are designed to tolerate groundstrain and cracks beneath.Refined calculations are not warranted, but carefulchecks are made oflimitingconditions of support,possibly at reducedfactor ofsafety. Spreadfootings are generally used, often with movement permittedin controlled frictionplane (ofsand,or two layersoffelt, or other selected material above a smooth blinding) to ensure that sliding can occur before instability from over-turning or bearing. If there is a risk ofcollapse above a local void the structureis designed to span or cantilever over. Wallsare provided with reinforcement to resistfrictional tensions, or else with plenty ofmovement joints. Culvertshave been articulated. Piles are generally not appropriate unless mineworkingsare very deep, or the piles are constructed (possibly within large sleeves) so that their performance is not affected by subsidence movements. Possibility ofpiles bearing above voids is checked.
Temporary works
Mineworkingscan seriously affect foundations and cost oftemporary works: a bridge designavoiding needfor expensive temporary foundations may be appropriate.
Monitoring
Monitoring ofstructures vulnerable to unpredictable subsidenceis importantand reassuring. InstitutionofCivil Engineers, (1972). Reporton mining subsidence. The Institution, London. Bell,F G (1975).Site investigation in areas ofminingsubsidence. Newnes-Butterworths. Simms, F A and Bridle, R (1966).Bridge design in areas ofmining subsidence. Journal ofthe Institutionof Highway Engineers, November 19—38. NationalCoal Board Mining Department (1975). Subsidence engineers handbook.
References
J
9
2.4 Soilsinvestigation Principles The soils investigation (seeBurland et al reference to Section5.5) is expectedto yield:
A knowledge ofthe soil profileand groundwater conditions across the site, to the depth affecting foundations, set in the context ofthe local geology and tied in with local experience.
1
2 A detailed and systematic description of thesoil in each stratumin terms ofits visualand tactileproperties.
This thouldpreferably be coupledwith routinein-situindicatortests for easy correlation with local experience and practice.
3 Anestimate or determination of themechanical properties of therelevant strata. For the majority ofsmall and medium sized bridges the decisionsas to type and depth offoundations can be made primarily on the basis of (1) and (2) above. In general an investigation cannotbe sufficiently precise to provide accurate quantitative information ofgroundbehaviour and the designerhas to estimatethe limits of behaviour betweenwhich the actual performance islikelyto be. Comment
Subject Planning of investigation
Several designersand geotechnical engineers stressed the importance oftailoring the scope of the investigation to suit the particularrequirements of each bridge and foundationin turn, and notjust following a stereotypedprocedure. By considering the requirements of the particularfoundations the risk of vital information being neglected is reduced,and there is alsoless risk of carrying out a large number of irrelevant tests. A deficient or stereotypedinvestigation could provesufficient for the design ofmany small structures, but for a singificant minorityan inadequate or misleading investigation could result in very expensive changes and delays during construction.
In order to identifythe particularrequirements ofeach foundation, it is often found convenient to carry out the investigation in two stages: Stage 1
To obtain abroad picture ofgroundconditions andidentifyproblems such asgroundwater.It mightincludetrial pits and one deep borehole per bridge with a large number of inexpensivetests such as SPTs to indicatevariability. The conclusions should enable the depth ofthe preferred foundations to be identifiedand indicatethe extent offurtherinvestigations required.
Stage 2 A detailedstudy ofcompetence, variability and construction problems of the strata in which preferred foundations will be placed. This can involve furthertrial pits and two or more boreholes per foundation in a pattern to identify the soil strata in three dimensions along and across the site: but fewer are often found sufficient in regions where strata are predictable. Knowledge of the groundis required to at least bottom of the zone ofsoil influenced by loading (ie 'pressure bulb' shown in sketches below). The investigation may include special tests and installation ofpiezometers. If the preferred foundation cannotbe identifiedwith confidence, information will be needed for the whole range ofstratawhichmight provide support or affect performance.
10
2.4 (contd) Soilsinvestigation Comment
Subject Planning of investigation (contd)
The continuity ofsite investigation between Stages I and 2 depends on the size of contract and planning of the design process. With close co-operation between site investigation supervision and bridge design, the two stages are often consecutive. Itis generally considered advisablefor the designerto be consulted before the first borehole on each site is abandoned. On a large contract it may be practicalto complete Stage 1 boreholes at all bridge sites, while the designer chooses the preferred foundation types,and then to continuewith the Stage 2 investigation. However such preferred procedures are not alwayspractical during the planning stage of a project because ofproblems ofprogramme or access. There can be occasionswhenthe precise position of foundations, and possibly ofa bridge, are not known at the time ofthe mainsoils investigation. Then the designer mightcover the likelyextremepositions of the bridge in plan with the trial pits and boreholes ofthe maininvestigation, and later whenthe designhas progressed he can carry out furtherinvestigation possibly in the form oftrial pits with in-situ testing only. The soils information required for planning constructionprocedures can be quite differentfrom that requiredforpredicting the performance ofthe groundbeneath foundations. Information for construction is requiredby the designer to help him identifythe type ofbridge which can be built most economically, and by the contractor who must work in and on the ground. Ground characteristics which particularly affectconstruction include:
theconsistency ofthematerial to be excavated; thestrength of the ground alongside excavation; groundwater levels andpotential flows; thestrength ofgroundbeneath temporary works; thestrength ofgroundbeneathheavyplant. While the collection of such information as part of routine soils investigations is encouraged, designersemphasise that it isunwise to provide the contractorwith an interpretation ofit in the ContractDocuments. Liaison with bridge designer
Designersandgeotechnical engineers stressed the importanceof the bridge designer being intimately involvedwith the site investigation from its initiation. He knows howcriticalor not criticalground movements are to the structureand he needs to convey the appropriate sense ofproportionto the management ofthe investigation to ensure that their interpretive conclusions are based on correct assumptions. He alsoneeds to check that his understanding ofborehole log descriptions agrees with thecontractor'sby personally carrying out a detailedvisialinspection ofsplitcores earlyon in the investigation.
It can be useful, ifthe designprocess is likely to be protracted,for thedesigner to write a brief report for later reference summarising the types of bridge envisaged during the investigation, together with observations on the shortcomings ofthe investigation.
In contrast,on theoccasionswhena late change is made to the route thedesigner mayneedto anticipate the limitations ofa rushed investigation and adopt a design for bridge and foundations which is conservativeand simple. Supervision
Designersand contractors on the survey emphasised most strongly the importance ofvigilant and competentsupervision in the field. The quality of an investigation cannot be checked later. Small details of great significance,such as waterbearing fissures and silt bands or boulder sizes,caneasilybe overlooked. The engineer charged with making the detailedfoundation recommendations, has the greatest motivation to demandan accurate investigation and so should have a day to day involvement and control over the field drilling and testingprogramme. 11
2.4 (contd) Soils investigation Subject Supervision (contd)
Comment Some design offices check daily logs and preliminary test results immediately and maintain an updatedlog of ground conditions over the site. Thusthey identify inconsistencies in time for checks to be made. Although this requires experienced senior staff it occupies little time ifdone as a routine. Particular attention needsto be paidto the description of soils. Properdescriptions can provide the most valuable information from the investigation. Poorquality descriptions can be misleadingand result in considerable waste ofmoney. The descriptions must be in accordance with the standard of CP200I to ensure correct interpretationby later users.
Reportwriting
Two reportson the investigation are usually produced. The first, written by the contractor, collects together the factual information of borehole logs and tests. The secondcontains the interpretation ofthe facts and recommendations and is written either,by the designorganisation if theyhaveexpertise and manpower, or by their consultant, or most commonly by the contractoron a consultancy basis. Whichever practice is followed itis essential that there is close interactionbetween the person making recommendations and the designer to ensure that the recommendations correctly apply to the particularrequirements ofthe bridge andthe overall scheme. Some designershave criticised reports for containingconservative generalisations, while contractors have complained that they could not obtain adequateinformation about proposed structures. Once a report has beensubmittedit is difficult to relaxrestrictions without further investigation. At the same time care has to be taken that the co-operation is not so close that the objectivity of recommendations is decreased.
The assumptions and calculations (or method) from whichconclusions are drawn needto be includedin an appendix to the report to facilitate verification. Ultimate capacities, load factors and factors ofsafetyneedto be clearly stated andjustified, and the level at which judgementhas beenusedmadeclear.Care needsto be taken when interpreting descriptions of strength since there are significant differences between sometextbooksand CP2001. Also due regard needsto be paidto the dependence ofempirical relationships and factors ofsafety on particular test methodsand sample sizes. Reports should not need to be long. Several designersobserved that the current fashion for encyclopaedic volumes is not found helpful. Ifpiling is involvedit is convenient to contractors if the relevant section is writtenso that it can be separated easily from the rest for forwarding to piling subcontractors.
At present bridge designersusually only issue the factual report to tenderers, because ofthecontractual risk ofinterpretations being proved incorrectduring construction. Contractors complain bitterly at this practice since they feel that a general appraisal ofgroundwater and construction characteristics,by someone with firsthand experience of the region, could be of considerable value to them. Report reading
12
The reportsneedto be readcritically, starting with an assessmentofthe reliability and scope of factual information. It is often forgotten that it is only possible to extract Class 1 samples, suitable for accurate compressibility and strength testing, from a small proportionofreal soils, as is indicated in Appendix A. Some geotechnical engineers feel that some structural engineers place too much reliance on numerical test results while not taking enough notice of qualitative observations.
2.5 Groundwater Thegroundwater regime on a site hasas much influence on construction and performance as the soil. The need for its properinvestigation and assessment cannotbe overstressed.It has in the past beenoverlooked on several occasionsand unexpected groundwater problems have caused someof the most serious construction delays. The subject is discussedin detail by Tomlinson (Ch 11). Problem Assessment
Comment The necessary information can often be obtainedat little extra costsimply by taking more care with observations during normal site investigation. Levels of water in boreholes need to be recorded in detail by dipping at every start and stop ofwork,including breaks and breakdowns, (ifpossible with an intermediate reading toindicatewhetherrise is slowing down), and details given ofcasing levels. It is sometimes appropriate to order specialstandstills for water measurement. While simple borehole records are helpful they are not accurate and can be misleading because ofwater draining into the hole or being addedto assist boring, or iflevel is thoughtstationary whenit is moving.Also, it is never clear at what level wateris entering. Standpipes or piezometers are neededif groundwater is likely to be a serious problem, or the possibility is recognised in preliminary borehole. The appropriate type of piezometer depends on soil type; the Casagrande type open standpipe is simple and often satisfactory. Installation of piezometers in correct strata require very close liaison between installer and designer.
Flowof water through strata can belarge, but can be difficult to estimate : a study oflocal geologyis helpful. Pumping trialsare unlikely to be necessary for foundations of small or medium size bridges. However a trialpit or augered shaft can give a rapid and directindication oflikelyconstruction difficulties. Observationsofgroundwater conditions at a point, like thoseofsoils, can be misleadingabout the regimein the grounden masse, and so shouldalways be treated with caution. In heterogeneous soils conditions can change markedly. Seasonal variations
Seasonal andyearly variations can be very large; so it is importantto recordstandpipelevels regularly and for as long as possible.
Excavations
The flow of water into the majority of excavations for small and medium sized bridges can generally be controlled by pumps withoutmuch difficulty. Butit is not uncommon to fmdone or more foundations on a contractin marginal ground conditions where it is difficult to predictin advance what problems will occur. Then very expensivedelays can occur while the methodofcontrolis selected. The expense is generally much less whenproblems are anticipated.
Ifserious problems arelikely then a differentdesign mightbe appropriate. Permeable ground
The flow can be controlled by dewatering but the expense of excavation is increased enormously, possibly tenfoldcompared to in dry. Lowering of the water-table and! or removal of fmes can affect the stability and settlementofneighbouring structures. In some groundsuch as above gravel stratumit may be impossible, and it may be necessaryto constructunderwater,possibly using massconcrete orhydraulic fill.
Moderate permeability
Pumping from open sumps is most commonly used because of its relativelylow costs. But where the inflow is large there is a risk ofinstability ofsides and base of excavation due to slumping and 'boiling'. Weilpointingcan be used effectively in sands andsandygravel. It has the advantage that water flow can be controlled to improve the stability of excavation. It has the disadvantagesof expense and that equipment is not usually on site. Cofferdams are frequentlypractical where a cutoffin a substratum is achieved. They thenhavetheadvantage that they reducethe uncertaintyofpredicting the efficiency and effectsofpumping. 13
2.5 (contd) Groundwater Comment
Problem Variable ground
Noneof above techniques maybe effective in very variable ground, or fissured weak rocks, or mixturesofbouldersand clay. Drivingofsheeting may not be possible, and it maybe necessary to use otherexpensive techniques, such asgrouting.
Impermeable ground
Impermeable soils, including bedded and laminated rocks, can be vulnerable to heave at thebottom ofexcavations ifsubject to high water-table.
Piles
Construction ofboredpilesis made much more difficultby water ingressto shafts and many expensive delays have occured in past.
Performance
Groundwater affectsthe performance ofbridge substructures by: reducing bearing capacity ofsubstrata; increasinglateralpreures by flooding or perched water-table;
frost heave oflight structures; flooding ofculverts below water-table in embankment; reducing stability ofbankslopes supporting bank seats or abutments. (Horizontal and vertical boreholes for drainage can help stabilise a bankslope in shales, mudstones etc.)
The settlementoffootings is substantially increased ifswellingand uplift has taken placeat the base ofexcavation. The performance of boredpiles isaffectedif soil is loosened or cement washed out during construction. Corrosivewater and soil
Sampling andtestingofgroundwater for corrosive conditions are often inaccurate due to exposure and contamination. If the initial site investigation indicates high acid or sulphate contentsthen much more carefulsampling and rapid testing may be appropriate in thesecond stage investigation. From these results theneed canbe assessedfor sulphate resisting cement, sheathing and so on. CP2004 givesadvice on protective measures.See alsoSection 8.6.
The risk ofsulphate attack depends on the accessibility of the foundation to water. If thefoundation is above thewater-table or thewater is not flowing then the risk is small (however therecan be a significant flow ofwater throughfissures in fissured clay).Sulphate contamination ofclay can be verylocalised due to pocketsofgypsum crystals. These have a low solubility, but sodiumand magnesium sulphates mayalso occur and are highly soluble. Sulphate attack is more severe in acidic conditions. Steelbearing piles are unlikely to lose a significant thickness ofmetal from corrosion where driven into undisturbed soil which is not aerated by flowing water. Uttle information is availableabout the degree ofcorrosion of piles driven into fills and soils which might be aggressive because they containoxygen or anaerobic sulphate producing bacteria.Some piles have beeninspected in such ground after 30 yearsor so,andwhile most have suffered negligibleloss of section a few can be expected to loseseveral mm during life of structure.Piles with protective coatings did not perform significantly better. It is known that parts of piles buried with disturbedsoil are much more vulnerable than piles driven intoundisturbed ground. Fortunately there is a sufficient built-in corrosion allowance in bearing piles for all but the most aggressivesoils, since workingload stresses are limited to only 30 per cent ofyield
to avoid damage during driving.
14
2.6 Advanced contractpile and platebearingtests Comment Advancedcontractpile tests Guidance for design
Designershave found it extremely difficult on many sites to makerealistic predictions ofpile performance, unless they already have experience oftests on similar piles in virtually identical ground conditions. For the majorityofsmall and medium sized bridges on piles this imprecision is generally accepted because it is more economic to adopt a conservative design than carry out an advanced contract pile test. However, where a contractincludes a large numberofpiles a considerable savingmay be achieved by improving the designusing the results ofadvanced tests. It is generally not possible to achieve this from tests in themain contractbecause working piles are only tested to 1.5 times working load, and a delaywaiting for testing and analysing special 'trial' piles is usually unacceptably expensive. See Section 7.7 for detailed comments.
Guidance on construction problems
An advanced contract pile testhas the addedbenefit, which may be of much greatervalue, that particularproblems ininstallation are recognised beforethe main contract and so reducethe risk ofexpensiveunexpected delays in the main contract. On occasions a trialinstallation is done without test. Since the performance and installation problems are very dependent on plant and method it is advanta. geous ifthe advanced pile is installed with conventional equipment, as similar as possible to that likely to be used in the main contract. Then all the records of equipment,rates of work, casing, groundwater, collapses etc could be of considerablebenefit to subsequent tenderers. However it is quite possible that the main contractorwill wish to use a quite differentpile type ofa different subcontractor and then much of this benefit of the advanced contract is unavoidably lost. Unfortunately the nomination ofthe advancedcontract piling contractorfor the main contract can introduce worse contractual problems than the advanced contract avoids.
Platebearing tests Plate bearing tests (see CP2001 Section 5.5) are performed to assessthe bearing capacity andmodulusofthe ground on a greater scale than is possible in conventional borehole investigations. They are useful for checking that piles are not necessary in uncertain ground conditions (a trial pile cannot indicatethis). In heterogeneous soils they often indicate thatmuch greaterbearing pressures can be used than can be predicted from small scale test results while in fissured clays they can indicatethat a lowerbearing capacity shouldbe assumed. Some series ofplate bearing tests have beenperformed rapidly and inexpensivelyin augered holesand trenches using the rig/excavator above for load reaction. They have the limitations that tests are generally impractical in water bearing groundand the region of soil tested is only ofsame scale as the plate,so several tests at differentdepthsmay be required for large foundations. Reference on plate bearing tests: Marsiand, A. (1973).Large in-situ tests to measure the properties ofstiff-fissured clays. BuildingResearch Establishment CurrentPaperCPI /73.
15
CHAPTER3 CATALOGUEOF SUBSTRUCTURES
3.1 General considerations This Chapterforms a catalogue ofthe most commonly used typesofsubstructure,to provide a reference during the conceptual design stage ofbridges. The table below listssome of the general comments relating to philosophy ofdesign reportedduring the survey, while the subsequent tables containthe comments related to various particular types. Sketches are included oftypicalsubstructure shapes, but not ofthe wide varietyof variations and combinations of features which may be adoptedto suit a bridge to its particularconstraints. Details such as shear-keys,sloping footings, ballast walls and so on are discussed in Chapter8.
Subject Interdependence ofsubstructure, deckand approaches
Comment Many designersemphasised the importanceofdeveloping the conceptual design of substructures in parallel with the design of the deck and approaches. Decisions related to span arrangement affect:
thenumber,loading and size offoundations; thezonesofsoil providing support; relative difficulties of construction of, and access to, different foundation positions.
The design loadson a particularfoundation might be adjusted by a change in spans, or articulation,or form ofdeckconstruction. The design bearing pressure ofa bearing stratumdepends on the acceptability of theassociated settlementand differential settlementwhich must be accommodated by thearticulation and flexure of thestructure (see Section 4.2). The approach embankment could settlemore than the bridge (see Section 5.1). Consequently the differential settlement betweenembankment and abutmentneedsto be considered at thesame time as that ofthebridge whenconsideringtheireffects on articulation, drainage, and ride ofthe road. Construction stages
Severalcontractors pointed out the influences ofcomplexity ofstructuralform and details on the cost and time ofconstruction. Simplifications aimed at reducing the number of construction procedures can lead to overall economy and better workmanship.
The construction of works below ground level is often the most susceptible to delays because ofits vulnerability to water from inclement weatherand unforeseen ground conditions. Refinements of geometry and detail of foundations in the bottom ofdeepexcavations are seldom economic (see Section 8.1 .). Structures which are freestanding at all stages of construction are usually easier to
build than structures whichneedpropping. Speed of construction can be assisted if thedesign enables the bridge and earthworks to be completed by more than one sequence so that the programme can have flexibility to accommodate changes of circumstances.
17
3.2 Wall abutments The stabilityofa retaining wall is usually calculated in terms of the forces acting on a vertical plane element of unit length. However economies can sometimes be made by considering the full structureas a single three dimensionalbody. A simple change ofshape can improve stabilityas easily as an increase in weight. However, intersections of component parts of the structure are possibly the most time consuming and expensive portions to constructand consequently are bestkept to a minimum. Reinforced concrete walls and bases are generally more satisfactorywhen designed with more than the theoreticallyeconomic thickness. Flexural stiffness is increased and steel fixing and concreting are much easier.
Type Mass
Comment
Usage
Massconcrete provides a simple form of construction. But the large quantityof concrete is relatively expensive.
Common for walls up to 2 or 3 m. Rare above 6 m
Sometimes when pouredin several lifts the back face is stepped. Placing concrete undera sloping backshutter can be difficult unless men have access. The front face is often raked backto assist resultantload's line of action through central third ofbase. Lackofreinforcement may make it ideal for a small contractwith littleor no other steel fixing. Low strengthconcrete can be used with consequent benefit ofless heat ofhydrationand cracking problems. Semi-mass
Reinforced T
Compromise between simplicity ofmass conceteand low material content ofreinforced concrete.Canbe the most economic form ofconstructionifreinforcement details are very simple
Common
Most common form ofconstruction. Often cheaper than mass concrete but relative merits are balanced. Minimumwidth ofbase is likelyto be achieved with heel larger than toe. However in cutting situationa smaller heelis likely to be economic because of reducedexcavation and working space, though sliding resistance is reduced.
Most common
Complicated reinforcement details make constructionslower than semi-mass.For single lift construction walls needto be wide enoughfor a manto stand between reinforcement to simplify construction and inspection. Counterfort
18
The complicated constructionof counterforts and much formwork makethis uneconomic for walls less than 10 to 12 metresin height. Can be economic for taller walls. Ifwalls are too thin then construction in tall lifts is difficult. Compaction offill between counterforts is difficult and not alwayssatisfactory.
Seldom used exceptfor very tall walls
Sketch
3.2(contd) Wall abutments Type
Comment
Usage
Abutmenton fill
A deepfoundation can often be achieved more Used when economically by placing a small abutmenton appropriate fill than by constructing a large abutment. In difficult ground conditions fill can be selfcompacting, such as mass concrete, and placed without workmen entering the excavations. The abutmentdesign can be standard with depth offill determined on site (but the detail and range ofpossible changes needto beconsidered in designandshown ondrawings since the stabilityofthe combined structure must be checked). Speed and simplicity of construction can more than compensate for apparentcost oflarge quantityofmaterial (See Section 4.1 concerning problems of excavation in difficult groundconditions.)
Sloping abutments
Sloping abutments are sometimes used when lesslateral clearanceis required at the top compared with the bottom and for aesthetic reasons. Construction is much more difficult than for vertical waits, particularly if the bridge has any skew (when re-use of formwork is likely to be minimal). Temporary supportofthe wallduring construction is much more satisfactory ifit can bearon the footing in front.
Oftenused by
A cellular abutmentis an expensivecomplicated form ofconstruction which is usually only used whenvery lowevenbearing pressures are neededand/orwhere piling can be avoided altogether. Cells are provided with positive drainage or filledwith unsaturateable filling such as expanded polystyrene.
Rare, for special situations
Cellular
Sketch
some designers
Cellular abutments havebeenbuilt with open backswith embankment slope inside. Compactionof the spill-throughfill is then difficult to achieve. The design is discouraged by somedesignerswhohave used thembefore. Cellular abutmentscan alsobe filledwith fill to provide a heavy anchorage for resisting horizontal forces. Internalstructurecan then be muchsimpler. Reinforced earth
Reinforced earthis appropriate for situations with embankments behind: but less likely to be suitable in cuttingsor where ties interfere with boundaries and obstructions.
The structurehas a large tolerance for movement and so isideal for siteswith poor ground near the surface (but ifpoorstratum is deep then circular slip is not resisted by ties). Differential settlement betweenabutmentsand embankment shouldbe smooth. A batter to the front face helpsto hide forwardmovement offacingduring construction.
Noabutments have yet been built in UK,
although a few have beenbuilt abroad.
19
3.2 (contd) Wall abutments Type Reinforced earth (contd)
Diaphragm, contiguous bored pile or sheet pile wall
Comment
Usage
Sketch
The wallcan be built overhand with a minimum working space. Damaged facingunits (but not ties) ifappropriately designed can in some situations be replaced during life without affecting stability. Cautionis warranted in assessingthecorrosion life ofties and fittings for abutmentswith long design lives. There is someconcernabout long-term erosion throughgaps in some types offacing. Forspecial These forms ofstructureare convenient for situations. construction from ground level, whentheir high cost is compensated forby the speed of construction and lack of temporaryworks. They usually require some form of facing after excavation. Ifthese abutmentscannot stand as free cantilevers they can be propped by thedeckor by a ground slabor ground beams, or tied backwith anchors. Contiguous bored piles are often the more economic for small orisolated works. The piles can be in-line or staggered to increase the wall thickness and provide a tolerance for variation in pile size: or pilescanbe 'secant' for better water sealing,(reinforcement only in every other pile). Water seeping throughneedsto be collected.
Large diameter piles with wall
in front
20
Diaphragm wallingmaybe economic forlarge or repetitive works. It can be constructedwith fair-faced precast elementswhichdo notneed subsequent covering. The wallmay have to supporta water table behind if adequate drainage cannotbe provided. Steel sheet piling has generallybeenassumedby bridge designersto be unsuitable for permanent works with a long design life. However, recently it hasbeen used to advantage for permanent abutmentswith protection/additional thickness provided against corrosion. In some cases sheet piles have been used forjust retainingthe earth while large piles behindsupport the vertical load.Economy is particularly evident when sheet piles are neededin any case for a temporarycofferdarn. It is importantnot to use too small a section which cannotbe driven into the ground. Abutmentshave beenconstructedwithinrail- For special situations. way embankments by boring two large diameter piles at each abutmentand then sliding in a precast deckto rest on bearings on the piles. Subsequently a retaining wallofcontiguous boredpiles, or conventional construction in cofferdam is placed in front ofthe large diameter piles and then the dumpling is removed. Construction ofthe wallafter placing deckis a difficult process.
'I..
Jww PL&. scctLons
3.3 Openabutments Open abutments, with the end spans ofthe deckbearing on seatings at the tops ofembankments are often preferable andcan look better than retaining walltypes ofabutment. The cost ofthe end span,pier and end supportofanarrowbridge is likelyto be less than that of a closed abutmentwith wing walls and retainedfill. Fora wide bridge thecostofwingwalls is relativelyless significant and theclosed abutmentis then likelyto be more economic. Open abutments are particularly suitable where the ground is not fIrm enoughto support the heavyweight ofa closed abutmentor where horizontal forcesmust be kept to a minimum. Since the bank seat settles with the embankment the problems ofsettlement of backfill behind the abutmentare reduced. Open end spans above embankments can have the disadvantage that the contractor may have considerable difficulty in providingfoundations for falsework for deck. Accessneedsto be provided underend ofspan formaintenance ofbearings and shelf drainage. Comment
Type
Bankseats
Bankseats provide simple economic structuresin semi-massor reinforced concrete.Idealfor bridges over cuttings where the bank seats can be founded on firmundisturbed ground. Some designersuse them on embankments on specially compacted fill without piles, and much prefer themto skeleton abutments(but contractorsdislike theneed to complete earthworks beforestructureand complain at delays waiting for settlement).Settlementof fill is anticipated andbearing pressures are usually keptwithinrange 100 to 200 kN/m2: the lower figure being appropriate to better qualitysuitable material, while the higher figure
Usage
Sketch
Oftenused in cuttings. Less common on embankments.
is likely to require selected granular fill.
The abutmentwith exposed bearing shelfis probably more economical than the buriedtype because ofthe reducedlength ofdeck. The front edge ofthe footing is best kept wellbackfrom the slope because oferosion and frost actionand because compaction ofthe fill on the slopeis slight. Wingwalls are constructed as cantilevers
or on footings. Bankseats on piles
Bankseats are placed on piles beside cuttingsand onembankments when the groundor fill is not strong enough. However settlement of the embankment can subject the pilesto downdrag settlementandloads.They are found more convenientifthe piles canbe placed at the same time as the other pilesin the contract(usually at start). They can be uneconomic ifthe pilesrestrictthe construction sequence or require remobiising plant. Forthe embankment situationthe skeleton abutment(on or without piles) can be more economic and less restrictive on construction sequence, and also more easily designed to resist laterial loads.Installation ofraking piles beside bank slopes can be impractical and expensive depending on the plant used. Bankseats have beenrestrained with deadman anchors while vertical piles supportonly vertical loads.
Used occasionally
21
3.3 (contd) Openabutments Comment
Type Buried skeleton
Usage
Buried skeleton abutments are more expensive Ofted used in and less popular than bank seats, though often embankments preferred in embankments because ofthe lesserrisk ofsettlement. They have advantage that construction can be independentof embankment, particularly ifdesigned so that the bearing shelfcan be placed before or after compaction of fill beneath.
It can be difficult to compact fIll properly
between the legs, and in time the fill behind can settleinto the poorlycompacted zone. Partial skeleton abutmentsare occasionally Rare used for very tall abutmentsor where foundations are very deep. They can be very expensiveifconstruction is not simple. Ifface wall is more than half total depth a buried wall is likely to be more satisfactory. Buried wall
A simple buriedwall can be more economic than a complicated skeleton abutmentbut thecost benefitis unlikely to be reflected in tenders.It has the benefit that propercompactionofthe backfill is more likelytobe achieved.
22
Little used
Sketch
3.4 Strutted, portal, and box structures Comment
Type
Strutted or vertical beam abutments
Many small bridges ofspans up to 15 m (and a few ofgreater span) have beenconstructed with the deckacting as a strut to the top of abutments. Withthe lateralsupport provided by the deckthe substructure and foundations can be narrower and simpler than for free-standingcantilever walls. In special situations such as where installation ofraking piles is difficult or where abutmentsare abnormally tall, the lateralsupportis particularly useful. Usually the walls are monolithic with the footings but somehave been designed with hinges at top and bottom,so forming a mechanism whichis only stabilised by theback till. Various simple details for deck seating have beenused with bearings as simple as sheets of felt or lead. However,care is taken
Usage
Sketch
Oftenused for small bridges,
but popularity decreasing.
Veryseldom used for spans exceeding 20 m.
that thereactionand thrust ofthe deckact at design pointsofcontactand do not cause spalling elsewhere. Some deckshave been designed with details to avoid needfor abutment shelfdrainage. Struttingof abutments is not recommended for skews greater than 200 but effectofskew depends on ratio oflength! width.
Struttedabutmentscan have someserious shortcomingsparticularly for bridges at the larger end ofthe range. Ifthe structure is not stable until complete with backfill expensive restrictions onconstruction maybe necessary. The substructures might have to be strutted and braced, or else the deck might needto be placed prior to backfilling which then has to be carried out simultaneously behind both abutments. Replacement ofthe deckor bearings (whose liveshave been found in past to be less than thesubstructures) could be very difficult. Earth pressures are likely to equal or exceedat-restpressure and since calculations are imprecise sections shouldbe robust.See Section 6.1 for discussion on design pressures.For larger spans earthpressures are likely to be much greaterthan at-rest pressuresdue to contraction/expansion cycles
ofthedeck. Portalframes
Portalframes are used for the same advantages Occasionally as strutted abutments, but for a much larger used. range of spans. It is possible to use a more slender deckthan for asimply supported span, but theportalframes are generally more expensive. It is alsonecessary for the foundations ofportalsto have a greater resistance against sliding since horizontalmovement of the footings can overstress the top corners of
the portal.
23
3.4 (contd) Strutted, portal andbox structures Type
Portal frames (contd)
Comment
Usage
Some portals are designed with hinges at the bottom ofthe legs to improve the predictabilityofthe structure.However the construction sequence is considerably more complicatedwhenthe legs cannotstand as freestanding walls. (It is also necessary to protect thehinges from vehicleimpact.) The effective stiffness of the footings and stiffening by the backfillare difficultto predict and consequently many portalsare designed for the worst conditions ofboth 'free' and 'fixed' feet. Differences in reinforcement are usually small. It is alsogenerally difficultto predictthe intensityofearth pressures,and the design has to be checked for very differentextreme values. Some portals have beendesigned with a soft filler down the backs ofthe legs, and a few have even been provided with separate retaining walls behind, to separate the uncertainties ofearth pressure from predictions ofstructuralaction:it is unlikely that such expedients are economic or significantly improve the accuracy ofcalculations. Some portal frames have reputedlyhad thelegs sloped backto reducetheearth pressures: again it is unlikely that this had achieved economy since the span andbendingmoments have beenincreased. The footings of some portals have beenplaced in front of the legs to move the effective points of supportinwards and so reducespan bending moments.
Box structures
Oftenused for Box structures are often economic for small used fresmall spans. Accordingly they have been of structures quentlyfor culverts and whenthe toes cantilever abutments wouldbe close together. They can also be economic for much larger structures on poor ground, particularly when they avoid the needfor piles. In addition to economy they have the advantage that differential settlement between structureand embankment islikelyto be less than when structure is on piles. The cost of the contractor'sfalsework for the deckis reduced because the lowerslab can be used as a firm foundation. The structurecan be constructed as an open U with simply supported deck. This is particularly appropriate whena precast deck is used.For small structures in particular, it is importantthat the design ofsections and reinforcement should aim at simplicity ofconstructionrather than economy ofmaterials. Box structures are occasionally installed under busyroadsor railwaysby the methodofthrust boring.The high construction cost can be justified by the avoidance ofdisturbanc!to traffic.
24
Sketch
n of tkno,t
J_ it:
3.4 (contd) Strutted, portalandbox structures Comment
Type
Usage
Flexible corrugated metal structures
Flexible corrugated metal structures ofcircular Oftenused or derivativecross-section have beenused for for culverts up to 6 m span many culverts and small underpasses,and a few with spans up to 12 m. Greaterspans are possible. These structureshavethe advantages oflow material cost and speed ofconstruction. (If necessaryone can be pre-assembled alongside an existing embankment and installed during a weekend possession). They also have the advantages that embankment construction is continuousover the top and settlementcan occur without the discontinuities that are common betweenembankment and rigidstructures.
Groundstruts
Sometimes on very poor ground it is convenUsed only for ient to prevent lateralmovement ofabutments exceptional by ground struts or slabs. Raking piles can be reasons avoided. The struts are made flexible ifvertical settlementbeneaththe under-road is anticipated: even rock-fill blankets have been used effectively as struts. The disadvantagesof ground struts are that they interfere with services beneaththe under-road and therecan be a risk oftheir being excavated by mistake in
Sketch
thefuture. Anchors and ties
Groundanchors are often not practical for retaining walls because ofthe problems of
land easements, but they can be practical for abutments. They are treatedwith suspicion by many designersparticularly for cohesivesoils and whereground watermight affectgrouting. When ties are used for deadman anchors or for tying retainingwalls back-to-back provision has to be madefor relative settlement ofwall and fill, and consideration given to constructiondetails which do not interfere with satisfactoryand simple placing of fill. It is also necessary to check that the structure can toleratethe movement necessary to mobilise theanchorresistance.
4;:
In areas ofmining subsidence the foundations ofsome bridges are tied together to prevent
separation. However the effects ofground strain on ties, and anchors needchecking. Ties between foundations are occasionally used on archstructures to prevent separation ifthe ground is not stiffenough: it is sometimes questionable whether the archis then the correct choice ofstructure. Anchors and ties, like ground struts, can be vulnerable to accidental damage during excavation for servicesat a later date.
25
3.5 Abutments above highly compressiblestrata Type
Comment
Abutmentsabove highly compressiblestrata are usually designed to accommodate or prevent large vertical and horizontal movements and possibly a deepseatedslip failure. The problem is minimised whenthe embankinents can be built well in advance ofthe bridge, possibly with surcharge, so that much ofthe movement occurs prior to construction of structures. Ifsubstantial movement cannotbe avoided then the deckand substructures are
Usage
Sketch
Used on exceptionally poor ground.
usually oflightweight design and able to accommodate vertical and horizontal movement.Tilting of substructures is minimised when the bearing pressures are uniformunder footings. Piles are sometimesused only to provide vertical supportwhile being allowed to move laterally with the embankment. Buoyant foundations with or without cellular substructures are very occasionally used iflittle or no additional load can be carried by the ground but the cells mayneed to be verylarge.
Ifmovement of thebridge cannotbe tolerated thesubstructures are designed to be independent ofor resist the movement ofthe ground due to embankment loading. The abutments
are sometimes designed with a smooth back face (possibly with single size gravelfill interface)so that the embankment downdrag is minimised. Ledges, such as underskeleton abutmentshelves,are avoided since voids form underneath into which the fill behindcollapses. Piles raked backwardsbeneatha settling embankment also have tobe avoided since they are likelyto be bent. The lateraldrag effects of ground flowing through piles is sometimesreduced by placing piles within outer cases with a clearance (possibly filled with soft filler). Abutments have been designed with heels extendedto support the embankment behind and to provide alongerflow path for the soil underneath.On somevery soft ground it has beennecessary to design abutments like quays with a sheet pile cut-offwall. Other solutions have involved bulk replacement of alluvium withincofferdams. Embankments have alsoon a few occasions been supported on seperate relieving platforms on piles, or partially supportedby uncapped piles. In a few special situations lateralspreading ofembankments has beenreduced by mattresses and ties.
It is importantto consider theoverallstability of thestructureand embankment in addition
to overturning, bearing and sliding. 26
iii,1TAIATIATAT
I..
3.6 Wing walls
The cost ofwingwalls can be a significant fraction ofthe total cost ofsubstructures, particularly for narrow bridges. They can also have a significant effecton appearances, which can be ruinedby a marginal saving. The choice ofgeometry is usually controlled by the topography of the site, construction restrictions and construction sequence. Ifpossibleit is advantageous ifthe wingwalls can be designed so that the contractorcan construct thembeforeor after the deck. Often they can be structurally independent of the abutment,in which case the joints betweenneedto be designed to permit significant relative movement with generous featuresto hiderelative tilts. Differential settlementwith abutmentcan be large due to differences in bearing pressure. Wing walls cantilevered back from abutmentwalls, or attached to strutted abutmentsor portal frames, are partially restrained against movement andcan be subjected to earthpressures greaterthan active. Severalabutmentshave suffered cracking at thejoints with wingwalls, and consequently many engineers give such abutments robust sections of conservativedesign to transfer the complex loads. Wingwallsare often constructedwith a batter to the front face and the section tapering towardsthe top. When theheight also decreases towards the ends access for construction can be complicated by the front and back shutters impinging unless they are specially cut to fit particularwalls. To achieve economy throughre-use of formwork it may be necessary to build the wallwith constant thickness. The base dimensions can often be designed constantwithoutloss ofeconomy,ifthe wall is designed as a whole and notjust on a unit length basis. Construction is also usually much simpler if the wingwall bases are at the same level as the abutments. Type Wingwalls parallel to abutment
Comment
Usage
Sketch
Wingwalls parallel to the abutmentface are in Common general not as economic in terms ofmaterials as angled walls but are usually the simplest type ofwalls to build.They have advantages that abutmentcan be built quickly with skew only affecting details ofthe abutmentshelf. Forbridges constructedin existing embankmentsthey are likelyto cause least disturbance to embankment. Back-fillcan be properly compacted with roller traversing from side to side. To ensure compaction at sides ofcarriageway it is usually necessaryto have additional horizontallength at top ofwall. They are often designed monolithic with abutment. Sometimes the parapetiscontinued a short distance behind the abutmentsand is supported on short cantilevers off the backof theabutment.These cantileversare usually most conveniently constructedin trenches after completion ofbackfihling.
Wingwallsat angle to abutmentface
Generally the most economic wing walls in terms ofmaterials are on lines bisecting angles between over-road and under-road. The extreme ends can be cantilevers if the walls are ofreinforced or semi-massconcrete construction. But small cantilevers can be more troublethan they are worth. The wing walls can be totallycantilevered offthe abutment wallif they are not too long and the abutment is robust.Horizontal tops to walls and parapet cantilever detail are the same as for walls parallel to abutmentface.
Common
27
3.6 (contd) Wingwalls Type Wingwalls parallel to over-road
Comment
Usage
Sketch
Less economic than angled walls but continue Common line ofdeckand provide support for parapets. Suchwing walls maynot be appropriate when they or their excavations impinge on adjacent structures, servicesor trafficdiversion. They can cause sight line problems where roads merge nearthe end ofa bridge. Also ifroad is curved the continuation of a long bridge chord line can restrict the road curvature. In addition they have disadvantagesthat propercompaction ofbackfill in the corners is seldom achieved and that forming ofexposed edges ofskew abutmentsrequires specialcare. Whenthe wingwalls are structurally attached to theabutmentadvantage can be taken of the stabilityofthe box structure.On narrow bridges orwhere used at same time as counterforts the abutmentwallcan be designed as spanning transversely.If such abox structure is founded on piles, including rakers, care needsto be takenthat the displacement ofthe piles underthe wingwallsis not incompatible with the displacements of the piles underthe abutment.It is sometimes foundbest to have only vertical and forwardraking piles, or to separate structurally wing walls iftheir piles rake sideways.
Cantilever wing walls parallel to over.road
When construction is in a cutting it is sometimeseconomic to split the wing walls in parts founded at differentlevels. However this is not appropriate ifthefounding material cannot standunsupported in steps, and the economy is questionable if each height ofwall requires a new shutter. Occasionallywhen an abutmentis supported on pilesshortlengths of thewing walls are cantilevered offit while theremainder ofthewingwalls are independent and supported on footings.
Uncommon
An abutmentwith large cantilever wing walls looks similar, when complete, to the abutment with wing walls on footings parallel to the over-road. It has the advantage that the whole structureis supported on a single footing which can settleas a singlebody. It has the
Common often with sloping abutments
same problems during construction, but in addition to poor compaction in the corners ithas thedisadvantagethat the compaction is difficult against the cantilevers because offlow ofmaterial underneath.It has on occasions beenfound necessary to construct a mass concrete wall underneath to provide a firm boundary wallfor compaction against.Voids or filler have sometimes beenleft under the cantilevers to permit backward tilting: but 28
H//.
3.6 (contd) Wingwalls Comment
Type Cantilever wing walls parallel to over-road (contd)
they are generally not necessary since such tiltingusually accompanies even greatersetfiement ofthe embankment. The abutmentand cantilever wingwallsneed to be robust and stiff to prevent outward bending.
Cantilever wing walls to box and portal structures
Whencantilever wing walls are attachedto a portal frame or box structurethe stiffness of thewing walls affects thedistribution of moments at the edgesof the abutmentwall and deck slab. Care has to be taken that the wallhas the appopriate reinforcement to transmit the forces attractedby this stiffness.
Usage
Sketch
The wing walls ofbox and portal structures can be thickened up at the top to cantilever horizontally from the deckwhich is avery strong member in-plane. Double wing walls forhigh abutments
A high embankment is sometimes most econo- Rare mically retainedby an abutmentwith two sets ofwing walls. The upper pair are cantilevers parallel to the over-road and retainfill which spills round to the top ofthe lower wing walls, whichare parallel to or angled to the abutment face.
Crib walls
In general wingwalls are ofthesame form of constructionas the abutments,ie mass, semimass or reinforced concrete. The latter is necessary for the more complicated structures with large cantilevers. Reinforced earth or sheet piling can sometimesbe used for wing walls to a concrete abutment.
Crib wallingis alsosuitable for wing wallconstructionand can be economic. (It is not thought suitable for abutments.) Its rustic appearance makes it more acceptable in rural situations than urban situations. Construction requires carefulsupervisionsince the appearance canbe unsatisfactory ifsettingout and elementtolerances are not carefully controlled. Cracking and spalling ofelements can occur if elements are laid with stress concentrations. The backfill needs to be drained and crib filling needscareful selectionsince some granular materials flow out.
29
3.7 Othersubstructures Type
Comment The diagrams illustrate some other substructures which have beenused for particular situations. They are usually not the most economic solution, though they can sometimesbejustified for aesthetic reasons.
Usage
Sketch
Rare, special
cases
Bridgeswith V supports canbe difficult to build and consequently expensive, particularly whenthey have only two footings. In addition
to needing very complicated temporary works
control of ifil at the endshas beena problem. However with simple footings and the bank slope struts pouredagainst the embankment construction neednot be so complicated. (Some designersmakethe strut concave to maintain contactwith the ground).
V supports have been used for cantilever and drop-in span decks. End-fixity has alsobeen obtained by abutmentswith large toes to footings and by massiveanchorblocks. Arches and arching structureswith raking pilesideally require very firmfoundations. Somehave beenbuilt on less suitable ground with large buried substructures to mobilise orbalance horizontal reactions. The decks have to be analysed for the effectsoflateral movements offoundations.
The contract documents forthese complicated structuresbenefit from the inclusion of a constructionsequence drawing, with detailed attention paid to interference of falsework for raking piers and deck, and to striking sequence. Otherwise serious differences ofopinionare likely during the contract.
Footings transmitting inclined loadsto the ground are much easier to construct with horizontal and vertical faces than with bearing face normal to line ofthrust.
30
ttL.
11.
3.8 Piers Type
Comment
Sketch
Usage
The simplest piers to construct,and most Common economic, are usually vertical with uniform rectangular or circular section and fixed at the bottom to the foundations. Inspection and placing of concrete is facilitated ifthe pier is large enoughfor a manto climb down inside.
-
Standardisation of details and reinforcement
inpiers of a bridge or severalbridges can lead to overall economy. Shapingofpiers can have aesthetic advantages.Complicated geometry, such as variation in section, is likelyto increase costunless considerable re-use ofshutters is possiblewithout restrictions on order ofconstruction. Some designersplacelittle confidence on single columns foundedon single large diameter piles because ofthe serious consequences ofsetting out errors.A pile group ofthree or more smaller piles is preferred. In the design of piers on piles or footings the rotationalstiffness of the foundations may have to be calculated to determine the effective height ofthepier.Narrow footings often havea sufficiently low stiffness to behave like pins, at thesame time being wide enoughfor stabilityduring construction. Bearings at the bottoms of the piers are generally not popular unless they can be made rigid during construction. Construction above the bearing is complicated and subsequent temporary propping ofthe piers complicates construction ofthe superstructure. They also have to be protectedagainst vehicle impact. Raking piers on simple foundations require firmground to resistthe arching forces. All stages of construction are considerably more difficultandexpensivethan for vertical piers and subsequent temporarypropping complicates construction ofthe superstructure. Theirhigher costscan often only be justified for high bridges where the rakesignificantly reduces the span.
—
/ f
31
CHAPTER 4 CHOICEOF FOUNDATION
4.1 Choice offootings orpiles Many designerstry to use spread footings wheneverthey can. They try to avoid piles because of the considerable expense and also because ofthe greater risk ofencountering unforeseen and expensive technical and contractual problems during construction. Even in the marginal situationwhere identifiable costs favour a piled foundation, manydesignersgo to considerable troubleto design deckand substructures to work on spread.footings with low bearing pressure and possibly with significant movements. The amountofdifferential settlement that is considered acceptable is discussed in Section 4.2 and 5.5. The following tables list some ofthe groundconditions and construction considerations for which footings or
piles have beenfound to be advantageous.For sites with variable strata the designersoften have to work
through arange of depthsand sizes offoundationto identifythemost economic with acceptable settlement character-
istics.The choice is not always straightforward. Sometimes it is necessary to choose between a foundation which can be constructed to a verifiable qualitywithout disturbing the ground but with a slightly unsatisfactory factor of safety, and one at greater depth with a higher theoreticalfactorofsafetybut with less certaintyabouts its quality.The decision is also affectedby the overallnature and scale of the contract,and it maynot be economic or practical to select what might appearto be themost appropriate foundation for each bridge ifviewed in isolation. None ofthe generalisationsfollowingare correct for every bridge. Ground
Stiff clay, or medium dense dry sand or gravel
Footings preferred Footings are often found to be straightforwardand appropriate for bridges for reasons ofcost, reliability and ease ofconstruction.
Finn clays or loose dry
Some designerspreferto use specialstructures on footings with bearing pressures as low as 100 to 150 kN/m2 possibly with significant settlement(mostusually occurs during construction).
Stiff stratumat
Spread footings at high level supported on mass concrete or granular fill, with formation on firmstratum,have beenfound by several designersto be less expensivethan deep footings.
gravels or sand and gravel
moderate depth with deep water table
High watertable in permeable ground
Stiff ground overlying soft ground eg graveloverclay
Piles preferred Piles mayhave to be used where very heavy concentrated loads have to be transmitted to ground. Piles often adopted.Settlementis less than that of footings, but often not as much less as anticipated (settlement ofa pile group is often as much as 1/2 or 1/3 of that ofa footing and takes place rapidly.)
Many designers and contractors advise against large excavations below the water table. Dewatering can increase the cost of excavation by as much as tenfold, and claims for unforeseen conditions are likely. Driven or cased bored piles are likely to be appropriate, but installation can alsobe difficult. Shallow footings can be designed to makeuse ofload spreadingqualityofthe stiffground. Piles can give rise to drivingor boring problems and by concentrating loadaboveorin clay can cause larger settlements. 33
4.1 (contd) Choice offootings or piles Groundconsiderations Ground
Footings preferred
Piles preferred
Soft silty clays, peat, and uncompactedfills
Spread footings canbe used on fill ifground can be preconsolidated by surcharge. Buoyant/compensated foundations can be more economic than piles, iffirm stratum is very deep, and ifexcavation and construction are not too complicated.
Interbeddedsand and silt layers
Settlement of footings has often beenoverestimated. Excavation has beenhindered by water in layers, but installation of bored piles has also given many problems.
Loose sands increasing in strength with depth
Improving existing ground (Section 4.3) has sometimes beenfound cheaper than piles. But the difficulties of prediction and monitoringthe performance of the special techniques can make themmore expensive and troublesome than a structural solution.
Driven piles compact the soil and can provide high load capacity with minimum settlement. Settlements offootings can be larger than predictedif not artificially compacted. Single size sands can be impossible to compact.
Chalk
Unless chalkis at depth spread footings are generally satisfactory. In the past piles were used more often. But it has been found that even soft chalk consolidates quickly under load (usually during construction) and material which looksin poor condition can take higher loadsthan anticipated.
Where upper surface is at unpredictable depth due to swallow holes, piles may be economic because they can be driven to different depths without delays in contract. Chalk softens during driving or boring and recovers some strengthduring following weeks. It can be worth delaying tests and testingpiles of different lengths to determine the optimum.
Unpredictable and impenetrable ground, such as boulders or rock with clay matrix
Both excavation for footings and installation ofpiles can be very difficult, with predictions of movement unreliable. Shallow footings can be the more economic even with conservative estimates ofbearing capacity and settlement.
Steel H piles have been used and driven either to deep firm stratum or to sufficient depths to mobilise adequate friction.
Compacted fill
Well compacted deposits of suitable or
Steeply dipping rock substratum
Massconcrete fIll can grip stepped interfaces where piles might glance off.
Deep seated instability and mining subsidence
Footings supporting an articulated/flexible structure are likely to be appropriate, since their performance canbe anticipated with greaterconfidence than other foundations. Piles are not effective in preventing deep seated instability. They are alsolikelyto be very expensiveif they have to resistor be protectedfrom loading by moving ground.
Piles are usually adopted. Measures have sometimes had to be taken to prevent damage from lateral loading due to adjacent embankment. (See Section 3.5).
selected fill are generally found to be as good as, ifnot better than, natural deposits ofthe same material.
Ifstratumis at depth then precast concrete piles with rock shoes or steel piles may have to be used.
Furthermoreit can be importantnot to 'dowel'the structureinto groundsubjected
tolarge strains. (See Section 2.3). Overriversand estuaries 34
Piles are usually preferred irrespective ofgroundconditions.
4.1 (contd) Choice of footings orpiles Construction considerations Construction consideration Cost
A large footing can be justified ifit avoids the need to use piles. For the averagemotorway bridge the tender price for piles is about the same as the price ofthe deckor the price of the substructures. Ifthepile depth exceeds 30 m their pricecan be more than thecombined price of deckand substructures.
Delays and consequential costs
Some designers try to avoid piling except whenabsolutely unavoidableprimarily to avoid repeating unfortunateexperience in the pastwhenpiling involved serious contractual problems and unforeseen consequential costs. The consequential costscan be large because thepiling delays, being at the start ofthecontract, can affect much subsequent construction. But it must be remembered that to a certain extent such problems result from the unsatisfactory and unpredictable nature of theground which motivated thechoice of pilesin the firstplace.
Site restrictions
Spread footings can usually be constructed without requiring specialaccess for heavy plant.See Section 2.2 concerning influences ofsite on construction.
Quality control
Modification
Piles preferred
Footings preferred
If piles have to be used savings can sometimes be obtained by raising thepile cap to ground level.
Boundary constraints, such as ad-
jacent buildings, steep slopes, railways and rivers can make con-
struction of spread footingsdifficult.
Quality ofground and foundation can be checked easily for footings. In contrastan engineer cannotinspectsoil in-situ beside small bored piles or piles cast underwater or bentonite. He cannot see whathappens to piles on removalofcasing, and it is probable that a numberofpiles constructed with ground water problems are defective. The qualityof piles is very dependent on quality ofthepiling gang and supervision.
Driven piles have advantages where ground water causes difficulty with
Simple mass and reinforced concrete footings
Pile lengths can be modified to suit uneven depth of substrata.
can be modified to suit excavated ground conditions with minimal delays. In contrast modifications to pile layout or type can have
bored piles.
very expensive contractual consequences. Alternative design
Some designersconsider it a disadvantageof piles that they are sometimes forced to accept alternative designsfor piles for apparent economic benefitwhile riskingworse contractual problems and reliability.
Construction programme
Contractors have much greater flexibility in theorder and timing ofconstructionof footings than ofpiles. They are often restricted as a result oftheir contract with the subcontractorin the numberofrigs than can be used forpiling andby the high costof remobiising piling plant.
35
4.2 Differential settlementcriteria The choice between footings and piles for foundations is sometimes controlled by what differential settlement can be toleratedby the bridge deck. Criteria for acceptable differential settlement differ enormously between designoffices, particularly for foundations for continuous structures. Deck
Criteria
Continuous deck
Some designersaim to design a continuousdeckfor a differential settlementofabout 1 in 800 (25 mm on 20 m span). But othersexpect differential settlementtobe controlled to as little as 1 in 4000 (5 mm on 20 m span). The deckdesigned for 1 in 800 is likely to be slightlymore expensive since differential settlement has to be considered as a primary load,comparable to live load and temperature. However, this cost can be very small in comparison with the cost of providing the very stiff foundations required for 1 in 4000,which are likely to require piles unless on hard rock. In addition to cost, the very stiff foundations have the disadvantagethat the structure does not move with adjacent earthworks to thedetrimentofthe smoothride ofthe road.
Simply supported decks
Foundations for simply supported structures are frequently designed for differential settlements ofthe order of 1 in 800. 1 in 200 (which isjust visible)has been necessaryon occasions. Then careful attention must be given to the effectsof settlement on drainage alignment and headroom. In areas ofmining subsidence much larger movements have to be anticipated. (See Section 2.3). The determination of acceptable bearing pressures to control settlement is discussed in Section 5.4, and the prediction ofsettlementin Section 5.5.
36
4.3 Groundimprovement Iftheupperstrataat a bridge site are weak,then thepossibility of strengthening thegroundmayprove more economic than founding on lower, firmer strata. The betterknown techniques of ground improvement are summarisedbelow. The limitations ofsome ofthe techniques are not clearly understood and the prediction and monitoring oftheir effectivenesscan be difficult. Preparation ofthe specification can also be difficult and an advancetrial maybe required in addition to test loading. Technique
Comment
Surcharge
Surchargingwith embankment has beenused on severaloccasions to speed up settlementand increasethe bearing capacity of soft clays. It is usually ofmost benefitto end supports comprising bank seats whichspread theirload throughthe embankment and do not subject the soft sub-strata to concentrated loading. It has thedisadvantagethat the time required for settlement to occur mayseriously strain thecontractprogramme or even warrant an advanced contract for theembankment, whichis often not pratical. However experience in many British conditions has indicated that settlementoften occurs more quickly than expectedand can be virtually complete by the end ofthe contract.
Sand-drains
Sand-drainshave beenfound effective on occasionsfor accelerating settlement. However they have been used unnecessarily on other occasions when the natural drainage ofthe ground has beenmuch more effective than originally estimated from laboratorytests.
Vibrocompaction
Vibrocompaction has been used satisfactorily for improving the bearing capacity of granular materials, such as loose sand, below concentrated foundations as well as embankments. Bearing pressures ofup to 300 to 400 kNfm2 have proved practical on vibrocompacted gravel. (The technique has been used unnecessarily and with no effectwhen the density ofthe existing ground has beenunderestimated).
Vibro replacement
Loose sand has been consolidated by vibration and further strengthened by stone columns insertedby vibroreplacement. The effectiveness of the technique for soft clays has not yet beenestablished.
Dynamic compaction
Dynamic compaction, using a dropping weight,has beenused for compacting a variety of soils.The range ofapplication and effectivenessis not clearly understood.
Grouting
Grouting has beenused frequentlyfor stabilising fissured rock, particularly in regions of old mineworking. But the techniques can be very costly and unpredictable ifthe dip and fissures are not advantageous.Othersoils are very seldom ideal for grouting and manyengineers are opposed to it because ofthe difficultyof making the grout go where it should and not go where it shouldn't. Much damage can be done by using excessive pressure.
Reference with
InstitutionofCivil Engineers(1976). Ground treatmentby deep compaction. 154 pp.
listsofother references
37
CHAPTER 5 SPREAD FOOTINGS
5.1 Global behaviour Many designersandgeotechnical engineersencourage the practice ofconsidering the global behaviour of the complete bridge and adjacent earthworks before starting the design and detailed calculations for individual foundations. The followingsketches ifiustrate some of the possible global movements that can occur aroundbridges. The truebehaviour can be very differentfrom thesimplifyingassumptions of calculations. Forexample, abutments are considered in calculations to overturn forwards although they frequentlysettlebackwards due to settlement oftheembankment behjnd.The globalmovements due to earthworks can be much greater than thesettlement calculated for an individual foundation. It is often found difficult to make precise calculations and sometimes only broadjudgements are possible. However several designersmentionedthe benefitofmaking separate calculations for favourable and unfavourable assumptions so as to estimate possible limits ofbehaviour betweenwhich theactualperformance should be.
Comment
Surcharge situation
Excavation situation
The 'dish' of settlement inducedby an embankment starts outside the toe and does not reach themaximum until beyond the brow.The lateralmovement ofthe ground is generally muchless than the vertical, unless the upper stratumis so soft that it 'flows'.Because the zoneinfluenced by an embankment is so much greater than that influenced by individual footings, the settlementdue to an embankment on soft groundcan be much greater than that due to footings even though thebearing pressure may be less. The zone ofheave or rebound due to removalof ground in a cuttingis the reflection ofthe settlement due to an embankment. Is is probably ofsmaller magnitude, but even with rock it can be significant for a stiffstructuresuch as an arch.
Ifthebrige is supported on rigidfoundations
whenthe adjacent embankment settles then pile cap and piles can be subjected to heavy loadsfrom embankment arching and downdrag as well as from lateralmovements. Also voids can form beneathledges and later the embankmentcollapses locally into the voids. The globalbehaviour ofpile foundations is discussed in greaterdetail in Section 7.1.
39
5.1 (contd) Global behaviour Comment The construction sequence has serious effects on differential settlementexperienced by the deck. The end supports can settle more than piers during construction due to embankment settle. ment,but later the piers can settle more because most oftheir load results from the deckwhich is constructedlast.
A retaining wall supporting an embankment on stiffground is likely to lean forwards, or slide forwards. The movements behind are predominantly due to slip ofthe active wedge.
In a struttedexcavation therebound due to unloading can exceedother movements. The ground behind the wallcan move down during excavation and later move up over an area extending about 2 times the depth.
A retaining wall supporting an embankment on compressibleground can lean backwards as it settles due to settlement ofembankment
behind,and the ground in front can heave due
to thecircular strain ofground.
The zone of strain round a cantilever wall supporting an excavation can mirrorthat of theretained embankment with theheave accentuated by circular movements. The piles supporting a retaining wallcan also be thought ofas resistingthe circular motion oftheembankment and ground. Iftheground is soft the piles can be subjected to high lateral pressures due to the soil movement as well as vertical loads due to the arching of the settling embankment.
40
Surcharge situation
Excavation situation
5.2 Loading on substructures and foundations The table below liststhe various designloadings for substructures mentioned by designersduring the survey. The loadingsare not considered to act all at the same time, and various combinations are identified during design. Many ofthe loadsare the same as thosespecified for the designofbridge structures in British Standard BS 153 and Department ofTransport Memorandum BE 1/77 (see Appendix D). These wifi eventually be superseded by British Standard BS 5400.Care has to be takennot to confuse the limitstate design procedures of BS 5400 with thetraditionalpermissiblestress procedures currentlyused in foundation design in CP2004 and CP2. The relative importance ofloads ofdifferentduration depends on the time-dependent behaviour of the supporting soil, as discussedin Section 5.3. Sections 5.4 and 5.5 describe the differentmethods designershave to adopt for calculation ofbearing capacity and settlementto take account of the differentbehavioursofnon-cohesive and cohesivesoils under'dead' loads and 'live' loads. Comment
Loading
self weight ofsubstructure; (ii) weight offill supported by foundation; (iii) deadload and superimposed load ofsuperstructure.
Deadload
(i)
Earth pressure
Continuously acting on completed substructure but likely to fluctuate in intensity with substructure movement, vibration, water table and so on. See Section 6.1 concerning relationship betweenearth pressure and movement and restraint of substructure.
Differential settlement
Differential settlement ofembankment relative to substructure transfers loadto substructure and foundations. Differential movements of supports of indeterminate bridge change reactions.
Hydraulic Flood
Piersin riversare subjected to lateralforces from change ofdirectionofwater flow. Forces increasein times offlood. Drawdown condition following flooding ofembankment. Also flooding ofretaining
walldrainage membrane from behind or above. Creepand shrinkage
Creep and shrinkage movements of superstructure can affect reactions and thrusts on substructure.
Temperature
Temperature changesin superstructure can alter reactions and thrustsor apply displacement to supports (which can in turn alter earth pressures).Temperature changes in parts of substructures and ground affect the forces and differential movement between them.Informationon temperatures is in the standards listedabove.
Traffice on bridge
Detailed information is included in standards listed above for loading from: (i) vertical gravitational loadincorporating impactallowance; (ii) centrifugal loadacting radially; (iii) longitudinal loads, which can act towards or awayfrom substructure.
Traffic on abutment
Live load surchargesto represent vertical traffic loading are given in the standards listed above. Horizontal loads due to braking and traction can act on substructure through fill and pavement. The designerswho mentioned this did not fmdCP2 or other references muchhelp and had to rely on firstprinciples.
Wind
Detailed information on wind loads on superstructure and piers is included in the standards listed above.
Impact
Piers are vulnerable to impact from vehicles, trains and river craft. The standards listed above provide guidance for piers alongside highways. Massive abutments and foundations are much less vulnerable and not likely to fail, though some displacement can occur from severe impact. Snow, ice packs, earthquake and so on.
Exceptional Construction
Combinations ofany ofthe above loadscan be critical for incomplete substructures during stages of construction together with loads due to temporaryworks, stored materials, moving loads,and possibly also accompanied by reduction of support. 41
5.3 Undrained and drained behaviour Many engineers are uncertainwhen it is appropriate separately underundrained and drained conditions.
or necessary to consider the stability of a foundation
Whensaturatedsoil is subjected to a sudden increase (or decrease) in loadthe particle packing cannot change suddenly due to the presence ofporewater and the soil strength cannot change from its initial undrained value. Over a period of time, which depends on the ground permeability, the soil consolidates (or swells) and the strength changesto the drained value appropriate to the degree ofcompression. In any construction situation with partially saturated soils and loadsofdiffering durations,the soil condition lies between the extremes of undrained and drained. However for design purposes it is convenient to checkthe stability for the worse of idealised undrained and drained conditions. Foundations are usually designed by considering all loadsas permanent. For permeable materials, such as gravels, this practice is generally realistic because soil test strengths and the dominating loads are virtually all long term in comparison with thetime necessary for thedissipation of excess pore pressures in thedeposit. For materials of lower permeability the practice is usually conservative because under conditions of increased loading the undrained soil strength is less than the drained strength after consolidation and dissipation of excess pore pressure. However there are conditions some of which are listed below under which the practice can be unsafe. and there are also conditions underwhicha fullydrained effective stress analysis can provide a more realistic explanation of the behaviour and stability of the ground which leads to an economy in design. The subject is discussedin detail in most textbooks, cfTerzaghi and Peck,Articles 17 and 18. But unfortunately many naturallyoccuring soil desposits have heterogeneous andcomplex characteristicsand structureand it isoften virtually impossible to predict the response ofthe deposit to change of loador the time for dissipation ofexcess pore pressure from potentialfailure zones. The application of concepts ofundrained and drained behaviour is then far from clear.
The table below indicates some idealised conditions and conditions under which addedcautionis warranted or benefit might be obtainedfrom a drained effective stress analysis. Ground
Undrained/'short-term'
Drained/'long-term'
'Sandgravel'
Impact
Virtually all loads except impact.
'Clay'
Impact, traffic, wind and short-term erection. Temperature, flood and hydraulic fluctuation on slow draining deposits.
Deadload, earth pressure, differential settlement, hydraulic and long-term erection. Temperature, flood and hyd-
raulic fluctuation on quick draining deposits. Permeable bands in many deposits enable much of pore pressure dissipationto occur during construction. Strength increases with consolidation.
Removal of overburden
Removal ofoverburden permits swelling and reduction of effective stresses so that drained strength is less than undrained.
Fissured overconsolidated clays
Slip can occur on polished slickenside with low 'residual' angle of friction. Fissurescan drain quickly so that relevant performance can be effectively drained.
Weathered marls, mudstones and shales
These weakrocks are really heavily consolidated clays and silts and behave as such. Measurement of 'undrained' strengthcan be difficult because of heterogeneity. But short term strength of clay in band or fissure can be critical.
Effective stress parameter from drained tests can be more consistentand reliable than undrained test results.
Chalk
Strong understationary load, but the fabric is destroyed by vibration! mechanical action.
Consolidates quickly. Strength of disturbed chalk increases with time as some fabric reforms.
42
5.4 Bearingpressure There are significant differences in current practice ofoffices and withinoffices in methods ofpredicting loading and bearing capacity. The table below summarisessome of themethods and comments reportedduring the survey. The bearing capacities ofsites are reported to be frequentlyunderestimated, often because borehole information is inadequate. Several designersfelt that foundations are sometimesunnecessarily deep: sometimes 3 or 4 m deep when 1 m would do. Acceptable bearing pressures are usually controlled by settlement. Oftena total settlementof50 mm is acceptable,thoughdesign rules for bearing pressure often relate to only 25 mm.
Ground! foundation Non-cohesive
soil
dry
below or nearwater table Cohesive
soil
Comment
Method/parameter Maximum safe bearing capacity to exceed maximum load.
Assumed by most designers.
Presumed value from CP2004, CP2 or Tomlinson Ch2;alsoother codesand textbooks.
Commonly used. Some designersconsider that values near top of ranges could be used more often.
Terzaghi andPeck empiricalchart of allowablebearing pressures (for 25 mm settlement) vs SF!' N-value.(Terzaghi andPeck,Article 54, andmany textbooks.)
Commonly used,but practicediffers re correcting N-value (see Tomlinson p 99) or not correcting. Several engineers stressed the fact that the SF!' testis onlya crude index testfor comparing differentsoils, and that unduerefinement oftheory canbe misleading.
Sutherland (1974),Peck Hansen and Thomburn,Lambe and Whitman,and others, charts relating N-value bearing pressure and settlement.
Used by a few.
Terzaghi and Peck, and Meyerhof bearing capacity factors. (Terzaghi and Peck,Article33, Tomlinson Ch2).
Not generally appropriate since settlement
Load test.
See Section 2.6.
Halfbearing capacity on dry ground.
Commonly used (see Tomlinson p 98/99).
Factorsofsafetyadoptedto limit settlementand local failure.
Many designersuse F ofS of3 for deadload alone and 2 for combined dead load and live load: but some increase latter figure ifgroundis heterogeneous or if loading is inclined. Lower figures are very occasionally adoptedwhen large settlements can be tolerated and predicted with confidence.
Recentauthorsindicate that bearing pressures up to twice those indicated by Terzaghi and Peck are satisfactory for 25 mm settlement.
governing.
Some use F ofS of 3 for all loadings. (Terzaghi and Peck, Article 53).
An overstress of 25 per cent is sometimes accepted with abnormal live loads if F of S is above 3.
43
5.4 (contd) Bearingpressure
Ground! foundation Eccentric and inclined loads
Allowablepressure
CP2
Footings on slopes or dipping strata Reference
44
Comment
Method/parameter
§ 1.47
Many engineersarescepticalofcommonly accepted pressure distributions and question whether pressures quoted for uniform vertical loads are applicable to peak pressures under eccentric and inclined loads. Textbooksand codes differ. There is little advice on settlement. Site investigation reportsseldom consider it. Commonly used.
CP2004 § 2.3.2.3.7
Inconsistent with CP2 and reported impractical for retaining structures.
Meyerhof's equations (seeLambe and Whitman p211).
Ultimate limit state approach. Used by a few designers. Generally the effectofslopeor dip has to be cal. culated from first principles.
Sutherland, H B (1974).Granular materials. Review Paper Session I. Settlement of structures. Pentech Press. London,473—499.
5.5 Movements Settlement Total settlement does not usually cause problems in highway construction unless bridge headroom becomes insufficient. It is the differential settlementbetween parts offoundations ofa bridge and between bridge and embankment that generally affects performance and use most. However differential settlement can normally only be estimated from total settlementpredictions. Unfortunately predictions of settlement are usually problematical and often inaccurate.
Comment
Aspect
Routinedesign
During the routinedesign offoundations for small bridges on stiffsoils the settlement is generally controlled by the choice ofFactor of Safety for bearing pressure, as discussed in Section 5.4.
Site investigation
Geotechnical engineers stressed that when settlements larger than 50 mm are anticipateddue to soil compression the site investigation needsto pay special attention to them and it is not satisfactory to rely on allowablepressures which are based on the assumption that settlementis related to strength and controlled by FactorofSafety.
Data
A very carefulassessmentofdatais considered muchmore importantthan refinement of calculation. The relevance and reliability ofinformation needs questioning, and it is useful to plotall data against depth to identifyanomalies and variations. Laboratory measurements of the compressibility of many natural deposits are often inaccurate.
assessment
Case records
Engineersin severalparts ofthe countryhave found that caserecords oflarge founda. tions on particulargroundconditions generally providethe only accurate guide for theprediction ofsettlement ofnew foundations on similar ground. It is often found that thestiffness ofundisturbed ground en-massegreatly exceeds predictions basedon tests. Consideringthe importance of settlement and the unnecessary cost in conservative designs,it is surprisingthat few observations are made of bridge settlement, and very few predictions are checked.
Loading
Somedesignerssimply make a conservative calculation ofsettlementundertotal dead load and live load.Others take much care and assess possible limits of performance and consider the effects ofsequence and durationofloads.
A major part of the loadson abutmentfoundations acts earlyin construction, while most ofthe load on pier foundations is only applied when the deckis added. Short-term loadingshave little effect on the settlement ofclay: some designersconsider 10 per cent oflive load as effective with deadload. However settlementof non-cohesivesoils is rapid and it is general practice to consider fulllive loadwith dead load. Differential settlement acceptable magnitude calculation
Section 4.2 reportsthe wide range of differential settlement criteria adoptedby designersofbridge decks. Someexpect the total settlementto be not more than 25 mm. In contrastwhen 1 in 800 is adoptedfor the slopeofdifferential settlementits magnitude can typically be 25 mm on a 20 m span,and the acceptable total settlementcan be as large as 50 mm (note dependence on span). Differential settlements due to differences in stratacan be calculated, but are subject to the combined errors of the separate settlement calculations. Differential settlementsdue to variations in a singleheterogeneous deposit can seldom be predictedand are often arbitrarily assumed to be half the total settlement. (Lainbe and Whitman Ch 25 and reference below report studies). Predictions for critical stages of construction can be calibrated and modified from comparisons ofpredictions and performance
ofearlierstages.
45
5.5 (contd) Movements
Settlement(contd) Comment
Aspect Differential settlement minimisation
Differential settlementof foundations on similargroundis minimised by some designers by designing all foundation to settle uniformly and by same amount. This mayrequire different bearing pressures for footings of different sizes and in different relative position. The accuracy ofthe settlementprediction maybe questionable, but errors shouldbe consistent and differential settlement small. To achieve this reduction or control of differential settlement it is sometimes necessary to make the foundations larger than are strictlyrequiredfor stability. The extra costmaybe welljustified if the differential settlement loads on the deck are reduced, or if piles are avoided.
Heave
See Section 5.1 for mechanisms due to embankment or cutting,Section 2.5 for action ofgroundwater, and Appendix B for effectsofpiling.
A structureacross a deepcuttingcan be subjected to significant long term heave due to removal ofvertical and lateralloadsin ground. Somesoils, such as chalkand mudstones, soften at the surface whenexposed in excavations, but compress again quickly during construction ofsubstructures. Duration of settlement
Settlement of bridge foundations has on several occasionsbeenfound to occur much more quickly than that of adjacent embankments, and tobe largely complete by the end of construction. This is frequently quicker than anticipated because of quick drainage through silt layersand fissures. However homogeneous clays can be very slow.
Predictions ofrate of settlement havebeen found to be more reliablewhenbased on in-situ permeability measurements rather than laboratorytests. Reference
J
Burland, B, Broms, B and De Melo, V F B. (1977).Behaviour offoundations and structures. State of the art report, Session II, Proceedings9th ICSMFE, Tokyo, Vol 2, pp 495—546.
Horizontal movements
Aspect
Prediction
Comment See Section 6.1 concerning relationship between earth pressure and horizontal movement ofwalls. In general horizontal movement is unlikely to be excessiveif the Factor of Safety against sliding exceeds 2. Horizontal movements of abutmentshave generally been small and not noticed, but when problems have arisen they have beendifficult to correct.Predictions are likely to be even more unreliable than settlement predictions unless they are based on field observations (ofwhichfew have been published). The horizontal movement ofground undera foundation half way up an embankment has beenobserved to be greater than the horizontal movement at the top.
46
CHAPTER6 ABUTMENTEARTH PRESSURESAND STABILITY
6.1 Active and at-restearth pressures Themajority of designersuse the concept ofequivalent fluidpressure or nominal earth pressure coefficient when calculatingearth pressures on the abutments ofsmall and medium size bridges. The calculation is usually straightforward since abutments generally supportfill ofone consistency with a horizontal surface. Field measurements (see reference) haveindicated complex non-linear distributions of pressure on abutments. However interpretation and reapplication ofsuch results for routinedesign are generally most conveniently performed by 'fitting'a linear relationship and deducing equivalent fluid pressures or earth pressure coefficient. Simple calculations are generally preferred because ofthe ease with which the designer can take account of the details and three dimensional structureofan abutment. However forlargestructures and where ground conditions are abnormally complexdesignersexpectto make more detailedcalculations with Coulomb wedge and/or Rankine stress analysis and possibly with detailed consideration ofsoil-structure interaction. Although the application ofthe concept ofequivalent fluid pressure is simple the selection of intensityappropriateto a particularstructureis not simple and requires a carefulstudyofthe interactionof the structureand its environment. The following tables report theearth pressure intensities adoptedby some designersfor abutmentswith different restraints. Comment Active pressures on freestanding abutments
Active pressures on drained freestanding abutmentsare usually calculated using an Equivalent fluid pressure 5H kN/m2,where H is in metres ( = 30 H lb/ft2, where H is in feet)
when(i) (II)
(iii)
thewallcan move by tiltingor sliding;
the wallhas positive drainage (seeSection 8.6);
thebackfill is a freedraininggranular material (see Section 8.7).
Because ofthe simplicity and popularityof the rule of '5H' care has to be taken that it is not applied to walls and situations that differ significantly from thosefor which it was developed. The rule was developed over a long period oftime and found satisfactory for simple structures which could move and which were never subjected to theheavy compaction by modernplant.
Otherbackfills
Designers assume higher equivalent pressures than 5H kN/m2 for less satisfactory fills, as indicated in Terzaghi and Peck (Ch 8). It can prove economic to design for suitable fill in regions where selected granular fill is very expensive. But it is usually very expensiveto retain fill ofhigh equivalent fluidpressure and a bank seatis sometimes found to be more economic. Equivalent fluid pressures lower than SH kN/m2 (in accordance with CP2) are occasionallyused for particular fills with properties whichcan be relied upon.
47
6.1 (contd) Active andat-restearth pressures Comment High pressures on restrained abutments
Dependence of pressure on movement
Strutted abutmentsand portal frames
Abutments and parts ofabutmentswhich are fully or partiallyrestrained against lateral movement are generally designed for pressures greater than active. An example ofthe relationship between earth pressure coefficient and displacement of walls supporting sand(as a fraction ofheight)is illustrated in Figure 6.1. The chartis relevant to walls whichmove bodilyor rotate (or flex) about the bottom,but not for rotationabout
the top.
Struttedabutmentsand the legs of portals are propped at the top. Consequently, the wall moves little and earth pressure at-rest is often assumed for their design. However some engineers consider it more appropriate to design for redistribution with the trapezoidal pressure distributionrecommended by CP2 Clause 1.4343 for strutted excavations.
Abutmentson piles
Earth pressure at-rest is assumed by some designers to act on abutments which are supported rigidly on piles. An intermediate coefficient between active and at-rest is appropriate when some lateral movement can occur, as is the case for many piled foundations.
At-rest pressure coefficient
The coefficient ofearth pressure at-rest K0 is generally estimated to be in the range between 0.4 (for well graded granular) to 0.7 (for less suitable backfill). Some designers assume for routine design that K0 is 1.5 times the active pressure coefficient givenin CP2, thoughoccasionally 2 times active is thought appropriate. (These factors are not appropriate for walls designed to retainundisturbed overconsolidated clay without displacement, in which case the horizontal earth pressure can exceed the vertical overburden pressure).
Cantilever T-walls and wing walls
Earthpressure at-rest is assumed by someengineers when designingthe stemofconventional cantilever walls for working conditions, and cantilever wingwalls; while active pressure is assumed appropriate for checking the factor ofsafetyfor ultimate stability ofthe whole structure.Higher pressures than active have beenobserved on instrumented structures but stability failures have been few. It can be argued that when the fill is compacted behind a stem above a heel the lateral pressure beneath the roller approximates to the compaction pressure of the roller. At depthsbelow which K0yH exceeds the compaction pressure the intensityis assumed to be equal to K0'yH. Ifa heavy roller is used it maybe appropriate to consider pressure higher than at-rest and check that these do not cause overall failure during construction.
Effectsof compaction
48
I
6.1 (contd) Active and at-restearth pressures
Comment Vibration
There is little information on the influence of traffic vibration on earth pressures. Many designers assume that abutments will be subject to some vibration and follow therecommendation ofCP2 that active pressure should be considered to act normal to thebackor virtual backofthewall. Heavy vibration is likely to cause earth pressures greater than active, and an argument similarto the above for compaction may be appropriate. There is also little information available about the effects of intermittenthorizontal loadsfrom braking and temperature effects.
Range ofpossible behaviour
Several designersrecommended that for circumstances when prediction of the probable behaviour is not straightforward, it is worth tryingto identifythe limitsof the range of possible behaviour by making alternative calculations based on favourable and unfavourable assumptions.
3-Dimensional form
The design of the 3-dimensional form of an abutment can sometimes havea much greaterinfluence on the stability than variations in earth pressure. Several designers emphasised the importance of choosing the optimum geometry ofstructureand arrangement ofjoints to achieve economy.
Load combination
One designer reported that the adoption of earth pressures greater than active for various degrees ofrestraint,as discussedabove, has not necessarily resultedin heavier structures because the more detailed analysis has also indicated that the worst possible load combinations are less than the cumulative sum of possible loads previously assumed.
Reference
Jones,C J F P and Sims, F A(1975). Earthpressures against the abutments and wing wallsofstandard motorway bridges. Geotechnique Vol 25, No 4, 731—742.
49
6.1 (contd) Active and at-restearth pressures Summary of comments on earthpressures coefficients for abutmentswith freedraining fill Refer to Section 6.1 for explanations.
Wall type 1
Masswall freestanding
Calculation Overall overturning, bearing and sliding
Assumed restraint
Effective earth pressure coefficient
Sketch
Small relieving Ka movements possible
2
Wall design
3 Cantilever T-wall freestanding
Overall overturning, bearing and sliding
Small relieving move- Ka mentspossible
4
Wall stemdesign
Stem restrained against base during
Ka
K0
compaction 5 Cantilever T-wall on raking piles
Pile loads
Wall restrained against horizontal movement
6
Wall stem design
As 4
I(
Pile loads
Wall partially restrained against horizontal movement
Ka
Wall stemdesign
As 4
Wall stemdesign
Retainedfill arches
Redistribution as CP2, Clause 1.4343
Portalframe deck
Relievingeffectsof earth pressure reducedby shrink-
0.75 Ka for low estimate
7 CantileverT-wall on vertical piles
8
9 Strutted abutment, portal frame
10
span sagging
K0
K0
age
11 Skeleton abutment (see Section 63)
Overallover turning, bearing and sliding
12
Column design
13 Cantilever wing walls
Wingwall design
Fill arches on to columns
2 Kaon column width
As 3 forwallpar
2 KaOflcolumn As 4 for wall par Wallsrestrained by abutment
Notes: 1 CoefficientK refers to 'effective' earthpressure. Water pressure needs to be considered as wellifdrainage is poor (seeSection 8.6), and at least to drainage level. 2 Many designersassume active earth pressure acts normal to virtual backsofabutments(and to stem),as recommended by CP2 for walls subject to vibration. 3 The form of an abutmentin 3 dimensions has a major influence on stability. See Section 3.2. 50
6.2 Passive pressure. 6.3 Skeletonabutments 6.2 Passive pressure Comment Risk ofremoval
of passive support
Passive resistance is ignored by most engineers in front oftoes to abutmentsand
retaining walls when they consider sliding and overturning stability, unless the footing is very deep. It is ignored because of the risk of a trench being excavated in front of thewall at some time in thefuture. Furthermore, theresistance ofthe ground for at least a metrebeneaththe surface is likelyto be affected by weathering.
Construction precautions
Ifpassive pressure does have to be assumed then the construction sequence hasto be
Displacement requirement
To develop full passive resistance large displacement must occur,as ifiustrated in Figure 6.1 of earth pressure coefficients against wall displacement. For this reason passiveresistance when it is considered, such as in front ofa shear key, is often attributeda value wellbelowits theoreticalvalue at large displacements. Some engineersattribute to it a value in stabilitycalculationsequalto about 0.7 of the safe
tightly specified and controlled to ensure that the grounddoes not softenduring con struction. Andevenwhen it is ignored many engineers aim to benefit from any resistance that can be obtainedand specify that the backfill ofworking space and overdig in excavation in front of the toe should be in mass concrete.
vertical bearing pressure for the soil.
6.3 Skeleton ('open' or 'spill-through') abutments Comment Variety of methods
A wide varietyofdifferentarbitraryrules are used by designersfor computing pressures on skeleton abutments. (See footnote about survey). The picture is furthercomplicatedby the fact that the design details also differ markedly with very different spacings and widthsto columns. The range ofmethods is illustrated by the four listed below in order ofmagnitude ofassumed earth pressures. 1
2
No lateralearth pressures on columns in stable embankmentswith slope 1 in 2 or shallower. It is argued that a skeleton abutmentis not likely to reducethe stability of theembankment, nor is it likely to have much influence on the movements ofthe embankment. Consequently some movement ofthe abutmentis anticipated, similarto a bank seat on vertical piles, and someprovision is made for the narrowing ofthe deckexpansion joints. Lateral movements ofabutmentshalfway down embankments are reported tobe greaterthan movements at the crest. Chettoe and Adams recommend that columns be designed for an active pressure immediately behindthem plus an arbitraryallowance ofup to 100 per cent addition.
3 Huntington recommends that no reduction from full active pressure over the gross widthbe made if the openings are less than twicethe widthofthe legs, and the fill in front shall not be considered as providing greater resistance than active, with reduction due to the descendingslopetakeninto account. 4 Fullactive pressure over thegross width. Footnote on survey
During the survey 20 statementsofpractice were recorded relating to skeleton abutments,amongst which therewere 12 differentmethods ofcalculation. The extremes ofpractice1 and 4 above were recorded most times, 4 times each. 51
6.4 Buried structures Comment Lateral pressures
During constructionlateral pressures on culverts can build upin a similarmanner to those on strutted abutments. Consequently at-rest pressures are assumed to develop by somedesigners.When culverts are below a perched water-table in an embankment, where drainage maynotbe practical, the possibility ofhigh excess pore-water pressure is usually considered. Lateral forces also act on buriedstructures near the surface as a result ofthe traction and braking loads ('longitudinal loads') along the road above. The amount of load transmittedto the structure depends on the relative stiffness of the structure and covering fill. The effectsare usually ignored ifthe depth of cover exceeds the span.It may be possibleto use the road slab to transferthese loadspastthe buried structure, but the road slab has to be designed structurally for this function.
Vertical pressures
The vertical earth pressure on the roofofa culvert has often beencalculated asequal to the weight of overburden directly above, and some engineers have thought that some reduction is possible due to arching ofthe soil above. But other engineers point out that theearthpressure on top ofa concrete culvert islikelyto be greaterthanthe weight of overburden above since the culvertis stifferthan the fill at either side. The amountofadditional load attractedto the culvert depends on the relative stiffness of the culvert and fill and alsoon the stiffness oftheground beneath.Vertical pressures have beenmeasured on a culvert deepin a rock fill damequalto 2 timesthe weight
of overburden.
52
6.5 Stability of retaining walls There are differences in the Factors of Safety used by differentdesign offices for similar structures. Some engineersare very criticalof the Code of Practice CP2 for being more conservativethan several more recenttextbooks. However,as inthcated below, the lower Factors of Safety ofother texts are usually countered by more conservativeassumptions elsewhere. Comparison with American, Australian and South African practice indicates that the CP2 Factorsare not unduly conservative.It is also argued by some designersthat CP2 is appropriate and needs to be conservative and simple for routineuse by structural engineers for small and medium size bridge abutments. The additionalcostresulting from conservatismis notlikelyto be large. The followingtable summarises themethods reportedduring the survey. (Bearingcapacityis discussed in Section 5.4.)
Factorofsafety Aspect of stability Sliding
Other
CP2
F approx2
Resistance provided by base friction/cohesion and passive earth pressure.
F
Overturning
Comment
1.5
Some designers follow recommendation of Terzaghi and Peck (Article 46) and other authors for F = 1.5. But these generally assume lower angles of base friction than CP2 and ignore passive resistance. 2.0 is used whenpassive resistance is considered.
Resultant in middle third
Gravity walls
F 2
Otherwalls
F >2.25
Used by someengineers when thereis upliftat the heel.
See comment
Some engineers follow recommendation of Huntington that overturning stability should be controlled by keeping resultantthrust on foundations for all types ofwall within the limits: 1
Forwalls on firmsoil, resultant withinmiddlethird.
2 Forwalls on rock, or bottom of vertical stem of solid gravity wall resting on footing,resultantin middle half. 3 Forwalls foundedon very compressible soil, resultantthrust shouldbe at or behind the centre ofthe base to avoid forward tilting. Meyerhof's equation
A few designersare using the theory ofMeyerhof (see Lambe and Whitman p 211) to check the ultimate limit state for combined overturning, sliding and bearing under resultantinclined eccentric load.
It is necessary to check a deep seated slip failure when the soil strataat depth are weaker than thosechecked for sliding stability.
Slip
failure
F
F
'
1.25
Soil strengths based on back analysis of failure of the same type
ofsoil. 1.5
Soilstrengths based on tests. Sucha check maybe necessary for retaining walls on piles as well as on footings, particularly ifthe piles are used to supportthe retaining wall above a layer ofsoft clay. Huntington provides more detailedadvice.
53
6.5 (contd) Stability ofretainingwalls
Factorof safety Aspect of stability Slip failure (contd)
CP2
Other
Comment The slip stabilityofbank seats has to be checked. The value ofthe appropriate factor ofsafetydepends on whetherthe slip is close to thestructureor is a deep seatedembankment slip Slip conditions during construction should not be overlooked;piles to bank seats and legs ofskeleton abutmentshave failed in thepast during construction due to slip ofembankment into excavations for piers.
Reference
In addition to textbookslistedin Introductionsome designersuse: Recommendations ofthe Committee for Waterfront Structures EAU 1970. English Translation 2nd Edition 1971.WilhemErnst & Sohn, Berlin.
54
CHAPTER 7
PILEFOUNDATIONS
7.1 Global behaviour Severaldesignersstressed the benefitofconsidering the global behaviour of pile foundations with their environment beforestarting detail design and calculations. Broad judgements are made to assess primary influences and possiblemodes ofmovement.
Ourunderstanding of the precise interaction ofa pile foundation and its environment is still rudimentary, and a designerhas to rely heavily on his past experience and observation ofthe satisfactoryperformance ofsimilar installations. Fortunatelya considerable bodyofexpertise has beenbuilt up over the years in the practical design ofpile foundations. Nonethelessoccasions do occur when thereis doubt about theprobable performance of a pile foundationand some designersthen find it helpful to make alternative assessments and calculations for favourableand unfavourable assumptions. Designersin general use simple conservativedesign methods (see Section 7.6). Many feel that they are unlikely to obtain a clear understanding ofthepossible overall complexities except for thesimplest ofpile arrangements. Even so it is not always clear whatis conservative and confidence can usually only be based on the apparently satisfactory performance of traditional designs.
The following tables summarise some ofthe possible primaryinfluences and modes of movement considered by designersin the survey. Behaviour Overallconcept
Comment
Sketch
Some designersencourage the drawing ofthe whole of the foundationin its environment. Pile groups are sometimes designed with thought given only to the arrangement ofthe pilesat the cap, with lengths selected separately to provide the necessarybearing capacity. What appears a sensible arrangement of piles at the cap
LI
can look totallydifferentif the piles are drawnto full length (the top sketchwas takenfrom a text book). The piles can be spaced at the cap and appearlikelyto distribute the loadto the substrata; while in fact they unnecessarily concentrate the loadsat depth. A bridge pier may be more efficiently supported by a fan ofpiles than by a groupwith all central piles vertical and edge
uI
_________
ones raked.
Patternof displacement
The pattern ofdisplacements needs to be considered carefully, particularly if the group includes piles of different rakes. A pile groupundera vertical loadis likely to cause a 'dish' ofsettlementin thesub-strata; raking piles at the edgesand corners of a vertical group may bearin groundthat settles less than piles beneaththe centre and so be subjected to much greateraxial and bending loads. Failure ofsuch piles has been known to take place.
55
7.1 (contd) Global behaviour Behaviour
Pattern of displacement (contd)
Comment Relative movements betweenground and a group with differentrakescan inducelarge secondary effectswhich cannotbe predicted. Settlementofan embankment behindan abutment with piles raked backwards can subject these piles to severe bending. For this reason rakingpilesbackunderan embankment is often discouraged. The effects ofrelative settlement between ground and vertical pilesare likely to be less severe and more easily predicted.
Downdrag
Settlement ofan embankment and sub-stratum relative to piles causes downdrag (negative skin friction). Onlyin exceptional circumstances is this likelyto cause failure ofthe pile material but it does increase settlement. If thesettlementoftheupper stratarelative to the bearing stratais small downdrag is not likelyto be a problem since a smaller settlementofthe piles will relieveit. Bitumen slipcoat is sometimes used to reduce downdrag on driven piles. However some engineers question its effectivenessand its ability to survive driving through some soils. It can be cheaper to use larger pileswith strengthto carry the downdrag. Its significance is discussed by Tomlinson (p 40) andinfluence ofpile spacing by Broms (Reference Section 7.4). Ifthestructureis articulated so that thesettlementdoes not impair performance then it may be practical to let the piles settlewith the embankment.
Pile cap load
The contributionof the pile cap to the bearing capacity of thefoundationdepends on muchthe same global interaction of foundation and environment as that whichinfluences negative skin friction. Ifthe groundaroundthe piles is settling relative to the piles, ie the conditions for which downdrag occurs,then the ground will settle awayfrom beneath thepile cap so that the cap only receivessupportfrom the piles. On the other hand ifthe piles settlerelative to the ground around,then the cap will progressivelypickup greater load. Since all the weight ofthe cap is initially applied to the ground whenthe concrete iswet, the bearing load between cap and ground should then increase from thislevel. Some designershave beenable to makesignificant savingsin piles whenthe caps did not have to be supported by them.
Lateral movement
Pile groups are sometimes designed so that they can move laterally with the surrounding soil. A pile group maybe better able to performthe function ofsupporting vertical loadsifit is able to move laterally without having to react with rakers, anchors or a propping deck. Although thereis little guidance availablefor the prediction oflateralmovements it is often found possible to provide more than adequate tolerance in the expansion joints. It is reportedthat a foundation half way down an embankment can move more laterally than points at the crestor toe. Lateral movements have been anticipated for pilesundera pier which could only be supported on a singleline ofpiles between railway tracks.The piles were designed to flex with
the pier during temperature movements ofthe deck. (When large movements are expectedthe change ofeccentricity of the load mayneed to be checked).
56
Sketch
7.1 (contd) Global behaviour Comment
Sketch
Designersdiffer in their demands for the performance ofpiles. Some only require that a pile should be able to supportan axial thrust,with moments resisted by the stiffening effects ofthesoil. To transmit inclined loadsto sub-strata a group ofsuch pileshas tohavethegeometry of a triangular truss. They are not appropriate if the surrounding groundis too soft to provide lateral stability, and it is then necessary for thepilesto havestrengthin bending.
ri
Behaviour
Groupdesign
/11 7'l' t
If the piles are designed to supportmoments then inclined
loadscan be carried by frame action(rather than truss) and it is possible for all thepiles to be vertical. Many engineers concerned primarily with foundation construction consider rakingpiles unsatisfactory and preferto have all their piles vertical, while engineersconcerned with abutmentdesign often preferto use raking pileswith moment capacityand rely on the combined trussand frame action ofthe group. Whateverthe geometry,the horizontal loadon the pile groups may be transmittedto the ground by sub-grade reactionnearthe top of the piles (possibly assisted by the slidingresistance ofthe cap) and/orby the portal action ofthe group transmitting horizontalforces to thelower ends ofthe piles. The precise force system depends on the relative stiffness ofthe piles and the surrounding soil. Methods ofpile group analysis are discussedin Section 7.6. At present no methodofanalysis is able to copewith all thecommon applications ofpiles and themost appropriate methodcan only be selected after a consideration ofthe relevance ofits assumptions to the particularsituation. Lateral loads
Themost serious deficiency of all currentmethods of group ana1ysforabutmentsis that they do not consider theinfluence oftheembankment behindon Lateral loading ofthepiles. It is reasonable to assume that a hard upper stratumcan resisthorizontalforces with subgrade reaction passive pressures on the fronts ofpiles. But ifthat layer ofgroundis soft, then it might apply unfavourable active pressures on the backs ofthe piles. Even whenthe ground is firmthe strain under the edge ofthe embankment may cause the ground to displaceforwards relative to a stiff pile group.There is a wide range ofdifferentforce systems between the extremes ofrelative stiffness. During design it is usually only possibleto makequalitative assessments ofthelimitsofgroundstiffness for which passive resistance oractiveloading are relevant from considerations ofthe global displacements.
Factor of
It may be necessary to study global behaviour when selecting the factors ofsafetyappropriate for pile design. In a large group somevariation ofpile capacity must be expectedand a sub-standard pile is unlikely ofitselfto jeopardise the performance ofthe group. But iffailure of an isolated foundation could lead to progressivefailures ofneighbouring foundations then a higher factoryofsafety
safety
"I
\i
'I't.
may be appropriate. An abutmenton very short piles may need to be checked for overturning stability, as ifon a footing. 57
7.2 Selection of pile type Comment Specialist experience
Consultation with contractors
Responsibility for design
Selection ofpile type and estimation ofperformance is an aspect ofbridge designfor which the majority of designers fmd it necessary to make use of a specialist. Wide experience is requiredof the various typesofpile in differentgroundconditions. The choice is influenced not only by costbut alsoby differences in performance, which dependon soil and groundwater conditions and also on the type of plant used and competence of the gang. The installation process can increase the strength of some soils and decrease the strengthofothers. The choice maynot evenbe straightforward to an expert,ifhe does not have personal experience ofidenticalpiles in similar strata locally,andhe may well request a trialinstallation to confirm his choice. Piling contractors have the greatest fund of experience of installation of their types of pile and many designers seek theiradvice early on in the design process. But it has to be appreciated that commercial considerations can influence contractor's advice. However those offering a wide range ofpile types maybe able to advise without prejudice on the most appropriate one offered. Some designers have found it advantageous to have recourse to an independent expert during thetender period, whenalternatives are likely tobe proposed, to assist in selection and in preparation ofspecial specifications.
The majority of designersexpect to take overall responsibility for the pile foundalionand only make up their mindsafterstudying all sources of information and experience available. The adequacy of the chosen type is ensured by load tests and propersupervision. (Oneorganisation has a break clause in the contractto the effect that if the initialload test is unsatisfactory the piling contractor can be paid for his work and asked to leave the site). However it is sometimes not clear during a contractwhetherdifficult ground conditions were unforeseeable and whois responsible for the additional cost. The difficulties can result from poor quality plantand labour forwhichthe contractoris reponsible. Some designers,when faced with the possibility ofa very expensivedelay, fmd it worth calling in an adviser with experience of installation problems and possibly paying for trials of different methods such as varying the length ofcasing, even when the responsibility is thought not to be theirs.
Vigilant supervision
The importance ofvigilant supervision of installation cannot be stressed too strongly (seeThorbum& Thorbum(1977)).
References
Whitaker T. (1976).The design ofpile foundations. Second Edition,Pergamon Press.
Weltman A J and LittleJ A. (1977). A review of bearing pile types.ReportPG1, CIRIA. London.
J
Thorburn S andThorbum Q. (1977). Review ofproblems associatedwith the construction ofcast-in-placeconcrete piles, ReportPG2, CIRIA, London.
58
7.3 Contractdocuments Many complaints were received during the Survey from piling contractors about the unsatisfactory qualityof contract documents. The table belowsummarisescomments. Problem
Comment
Presentation
All information relating to piling needsto be concentrated in a few pages and drawings so that it can be easily separated and issued to subcontractors. At present subcontractorsoften fmd they are issued with several different sets ofincomplete documentsby the different main contractors. Suchunnecessary duplication is avoided by one design organisation by making the piling information directly available to bona fidepiling subcontractors. It is usually foundimpractical for tenderers to take advantage ofa note that the full Site Investigation Report is available for inspection at the designer's office.Both the office and the site are likely to be at somedistance and senior engineers can neither spare days for travelling nor can they absorb the information at a glance.
Specification differences
There is confusion betweendifferentspecifications (ICE, DTp, GLC, RIBA, Federationof PilingSpecialists, and so on) and piling operatives do not alwaysknow thedifferences.
Clarity
The specification for piles is often scattered over several drawings as well as the Specification. Uncoordinated scattered notes can result in unnecessary small differences in constructionthroughoutcontract. (One contract was reported to involve 8 differentgrades ofconcrete in piles because ofsmallvariations in notes on drawings). To avoid confusion some designers write theircomplete specification on
thepile drawing. Soil
information
The qualityofsoilinformation supplied to piling subcontractors is often poor. Apparentlymany tender enquiries are accompanied by borehole records which do not reach or extend below the bottom of piles. Often little information is available about ground-water, casing depthsandboringproblems.
Unforeseen ground conditions
Contractors at tender stage often tend to be optimistic and not look for problems, but whenproblems are encountered they look for all theinadequacies they can. They try to play on the inadequacy of borehole records and the designer's staff do not always have sufficient knowledgeto rejectquestionable claims. It is in the client's best interestto have a clear idea ofpiling problems before the tender is let.
Measurement
Thebifi of quantities needs to be presentedin a form suitablefor the pilesto be used. It is not possibleto cover with a singlelist ofitemsthe fundamental differences in method ofconstruction ofdifferentpiles or the differentconsequenciesofchanges during construction. Some designers have overcome this problem by providing separate drawings and documents for each typelikelyto be offered (only allowingthe subcontractor to tender with the documents most appropriate). Others design for one likely type and provide separate documents only when an alternative is offered.
Reinforcement
The documents often do not indicatewhether or how reinforcement should be changed ifpile length differs from design, and the method ofmeasurement can confuse the issue by including reinforcement in rate per metre.Changingthelength of a cage at the last minute is usually impractical and often deleterious. The length of reinforcement and details must be indicated.
Length
Designersgenerally consider it advisableto overestimate the lengths ofpilesin tender documents. However possible shortening is often not achieved because tests have not beencompleted or the designer(and checker) is not at hand to approve the change: a delay can quickly cost as much as the extralength.If a subcontractoranticipates shortening he might adjusthis tender to obtain most paymentfor the early stages.
Alternative
Attitudes of designers differ towardsalternative piles offered by contractors. Some designersacceptthem readily, while others consider it is easier to contain claims during construction by sticking to the initial design. If alternative designs are to be encouraged, then design loadsmust be stated on the drawing. Ifthe designloads are dependenton the adoptedmethod of analysisthen thisneedsto be stated also.
designs
59
7.4 Pile behaviour Axial loading Comment
Aspect Design
Highlysophisticated design methodsare generally considered irrelevant because ofthe variability of soil properties. Differences in apparentsoil behaviour resulting from differences in testingprocedures and sample size can influence predictions much more than refinements in design method, and consequently deserve the greater attention. Only experience of similar site conditions provides defmiteguidance to behaviour and only loadtestingprovides confirmation of capacity. In additionto general textbooks in the Introductionthe references below give guidance and lists of other references.
Factor of
Factorsof safetyused in design havenot beenstandardised and practices differ. Many engineersuse a global factor ofsafetyof 3 during designwhenno test results are available,and 2 when test results are availablefor the same type ofpile in the same ground. Otherengineers check pile capacities on site on the basis of tests using a factor ofsafetyof2.5 for shaft frictionand 3 for end bearing. Large diameter piles in clay are frequently designed for the worst ofan overall factory of safetyof 2 or partial factors of 1 for shaft friction and 3 for end bearing.
safety
Settlement group
of
Prediction ofsettlementofpiles and group is considered by many designersto be problematical and liable to error.In clays the frictionat working load can be a very large proportionof the total so that load settlementbehaviour is dominated by the mechanism of friction development. Frictionon the shafts of end-bearing piles may provide a substantial proportionofthe support at working load since it maybe mobiisedby smaller displacements than the end bearing. The quality of installation work can have a marked effect on the load/settlement behaviour ofboth driven and boredpiles. Prediction ofsettlementofpile groups is reviewed by Broms (1976).The shortterm settlementin a loading test ofa singlepile is oflittle helpin predicting the long-term settlementofa group unless it is end bearing on rock or very dense granular soil where elastic compression of the piles is the maincomponent ofsettlement.
The term 'group factor' is used in different ways by different engineers,and applicationis often arbitrary. When applied to the reduction of bearing capacity per pile because ofthe neighbours it is generally less than 1 for groups in clay and also possibly for driven piles in dense sands ifvibration loosens particle packing. It is greater than 1 for loose sand because piling causes densification. 'Settlementratios' areconsidered more useful by someengineers and relate settlementofgroup to that
of single pile (see Whitaker reference below). Tension piles
Designers differ in their assumptions concerning the tensile resistanc obtained from skin frictionwith the ground. Some assume it can be effective in someform for all pilesin tension;othersassume that it cannotbe relied on for long term loadsin clay. A few designersignore it completely and only consider theweight ofthepile. Such differences in assumptions can have significantinfluence on the layout and cost ofa group resistinglarge horizontal forces. Broms (1976) reportson research into differences in skin frictionofpiles in tension and compression.
References
Broms B. (1976). Pile foundations — Pile groups.6th European Conference on soil mechanics and foundation engineering,Vienna, pp 103—132. Tomlinson M J. (1977).Pile design and construction practice. Viewpoint Publication, London.pp 448. Thorburn S and MacVicar R S. (1971). Pile load tests for failure in the Clyde alluvium. Conference on behaviour ofpiles, I.C.E. London,pp 1—7. Burland B and Cooke RW. (1974). The designofbored piles in stiff clays. Ground Engineering,Vol 7, No 4, pp 107—114. Skempton A W. (1959). Cast-in-situ bored piles in London clay, Geotechnique,
J
London,1959,9(4)153—173.
Whitaker T. (1976).The design ofpile foundations. Secondedition. Pergamon Press.
Institutionof CivilEngineers(1977). Piles in weakrock, pp 233. 60
7.4 (contd) Pile behaviour Lateral loading Aspect
Comment
Relevance
Designersgenerally assume that lateralloading ofpiles, whetheractive or passive (subgrade reaction),can be dismissedin the designofthe majority of small and medium size bridges whichare not on exceptionally soft ground.
Active loading
Piles for bridge abutmentsmay need to be designed to resist active lateralloading whenthe substratum is soft and moves forwards relative to the piles due to surcharge effects of the embankment behind.(Huntington provides advice for calculation of this active pressure and slip circle analysis of abutmenton piles). Even if there is a more than adequate factor of safety against slip, piles under an abutmentmay be subjected to some bending due to long-term creepflows of the ground in the potential slip-direction. So there maybe active loadson piles even in stiff ground, but ofsmall magnitude because displacements are small.
Subgrade reaction
In recentyearsseveral methods ofpile groupanalysishave come intousewhich rely
to some extent on subgrade reaction to resist horizontal loads. Subgrade reaction
can usually be relied upon for supportwhen the piles move towards the supporting ground and the ground is stiff. Butsince it can bevery difficultto predict howpiles and soil move relative to each other many engineers are not prepared to place confidence in subgrade reaction for lateral support of foundations subject to long-term loading. Aiso the displacement necessary to develop the required passive resistance can be unacceptable.
There are structures such as pier foundations subject to lateral live loads for which subgrade reaction can provide adequate lateral support. The methods of analysis of Broms (1976) and methods ofbeam on elastic foundation have beenused on many occasions. It is often found that only large diameter piles can develop the required capacity against horizontal loadseconomically. Ifthe top ofthe pile is inweathered ground, then the stiffness of this may have to be ignored, or at least considered as much less than that ofdeeperground. On the other hand ifthe pile is beneatha cap it may be reasonable to assume the full stiffness of the ground, possibly with additional passive resistance and restraining effect of the caps. Sometimes different stiffnesses for the groundhave to be considered forlong-term and short.term loading. Load test
Some engineers consider itjust as importantto load-test piles laterally as wellas vertically ifthe consequences ofunsatisfactory performance jeopardise the structure. In thepast tests have been designed for the particular installation, usually by jackingtwo piles apart. It has not always been clear how the pileshave reacted at depth, or what the maximum moments were.
References
Broms B. (1976).Stability offlexible structures (piles and pile groups). 6th EuropeanConference on soil mechanics and foundation engineering.Vienna, p 230—269. Tomlinson M J. (1977).Pile design and construction practice. Viewpoint Publications, London,pp 448.
61
7.5 Raking piles The installation ofrakingpiles is generally fraught with many more problems thanvertical piles, as discussedin thetable below. As a result engineers concerned primarily with design and construction offoundations generally try to avoid their use. But most designersconcerned with thewhole bridge like to see inclined loadsresisted by a triangle offorces in a truss-like arrangement of piles. Problem
Comment
Cost
The costofraking pilesin poor ground can be as much as two times that of vertical piles. It can be sensible to use vertical piles oftwicethe size if difficult construction problems can be avoided. But in stiffground which can standunsupported the problems ofrake are not so significant.
Maximum rake
The maximum practical rake depends on the ground conditions and methodofinstallation. Many design offices limitthe maximum rakes to the followingvalues: 1 in 8 for large diameter piles 1 in 5 for small diameter bored piles 1 in 4 for small diameter driven piles Piles can sometimesbe installed at greater rakes(1 in 3 is often possible) but not all contractors have the appropriate equipment. In specialsituationssteel piles have been driven atvery much greater rakesto provide an oblique prop or anchor. Tolerances can have a significant effecton the theoreticaldistributionof loadsin a group. Whilevertical piles are generally expectedto be within a tolerance of 75 mm at piling platform level and within 1 in 75, raldng piles are often expectedto be only within 1 in 25 for rakesup to 1 in 6 with greater tolerance at greater rakes. Tighter tolerances maynot be obtainable. If raking piles are installed from a level significantly differentto formation then the 75 mm lateralerror isincreased by theeffect oftheerror in the rake. It is not uncommon for piles to be outside specified tolerances. Ifa group of piles is drawn with the centre lines shown for positions ofextremetolerance, it is found that the ranges ofpositions ofpossible intersections (and load eccentricity) is large. If the lines ofpiles wanderoff course, ie 'banana',the ranges are even larger.
Tolerances
Severalengineers pointedout that during the design of a group thereis no point in tryingto fix positions of piles more accurately than they can be constructed. Rig working space and movement
Most rigs can only rake in particular directions and must have working space on the appropriate side ofeach pile. Ifalternatepiles are raked in different directions, then it can be difficult to move themachine without tracking over piles. It maybe appropriate to stagger lines of differentrake to avoid interference of rig and surplus lengths. Where manoeuvringis difficult it can be worth designingthegroup so that therig can workalong the group.
Bored
The installation of boredpilesis mademore complicated ifnotimpossible by rake because1 Local collapses can occur forming voids above the bore beforea casing can be placed. 2 Drivingthe casing on a rake can be more difficult. 3 It is difficult to get thecage ofreinforcement concentric without bars sagging with inadequatecover beneath. 4 Concrete has to be placed by a tremie to be sure ofquality.This is seldom done and it is difficult to demandit unless it has beenspecified. It is difficult to tremieconcrete deeper
piles
than 20 m. 5 On withdrawing a casing concrete is pulled up more than with vertical casings and therisk of voids forming is greater.
Driven
piles
Load tests —
62
Drivingpiles on a rake does not have such serious problems as boring. But the reduction in driving efficiency can be much greater than acknowledgedby the contractor.A check should be make that piles driven on a rakedo not meet apparentrefusal or adequate set at a higher level than vertical piles in the same strata.
The performance of raking piles shouldonly be checked by tests on raking piles since many of the problems ofinstallation do not occur with vertical piles. A test on a vertical pile maybe useful for proving the load carrying capacity of the ground, but it may well give little indication ofthe quality of installation ofraking piles.
7.6 Pile groups Comment
Lackof knowledge
Problem Many designersexpressed the opinion that current knowledge of the behaviour of pile groups is inadequate. For this reason they felt that a conservative approach is warranted.
Selection of simple arrangement
Severaldesignersstressed that it is much more hnportant to choose a suitable arrangement of piles to resist inclined loads than to select the best method of calculation. For the majority ofsmall and medium size bridges on piles very simple arrangements ofpiles are selected. It is generally considered difficult to estimate the possible performance ofa complicated group with certaintyor to decide what is conservative.
Considerations ofstatics
Many designersdetermine the positionof piles from a simple consideration of statics with piles in only two directions. It is often found that ifpile positions and rakes are selected purely from static considerations then whatever method of calculation is used for the group, the applied loads are primarily resisted by axial forces in piles, with secondary moments and shearforces relatively small.
Non-uniform
load distribution
Appendix C summarisesa simple design method used in several offices for selection ofpile position, using the elastic centre. Many consider this method ofdesign sufficient in itself; somerefme the arrangement using a computeranalysis. It is often assumed that the loadson thepiles in any row are uniformly distributed. Howeverifthe bearing stratumsettles it is likely that the piles at the ends of rows support greaterloads thanpilesin the middle ofrows.To ensure that end piles shed load to others without failure some designerscheck that the compression strength ofthe end piles and shearstrength ofthe pile cap exceedthe ultimate resistance ofthe soil on the piles.
Methods of calculation
Some commonlyusedmethodsofcalculation are reviewed inthe next table.Most designers preferto use the simple traditionalmethods ofanalysis ofCP2 for routinedesign. Many feel that the more complicated methods can seldom be shown to predict loads more reliably or economically. And it is concluded from the satisfactory performance of most pile structures in the past that the traditionalmethods are satisfactory for simple pile groups in average ground conditions. However some form offrame analysis is usually considered necessary for the more complicated arrangementsofpile groups and for simple arrangementsin very soft ground in order to predict the moments in the piles.
Comparisons ofmethods
Comparisonsofpredictions from thedifferentmethodsofanalysis for anarbitrarypile group indicate enormous differences in performance which undermine the confidence ofsome designersin every method.However some designershave found that ifthe group is designed by one methodto supportthe loadsoptimally by axial forces then predictions from different methods indicaterelatively small ranges for axial loadswith only significant rangesin the secondary actions ofpile moments and shears. Even so the similarity ofprediction does not necessarily imply they are correct;none oftheanalyses considers thesignificant influence ofground movements due to embankment surcharge. Comparison between loadscalculated by methods which consider a cross-section of a group (methods1 to 5 of next table) and loadsfrom a 3-D method (7 of next table) is not straightforward since the latter predicts variations in pile loadsalong each row.Suchvariations maycorrespond more closely with realitybut it is not clear howsuch loadpredictions should be used in design. Ductilityofthe ground causes redistribution so that local overstressingdoes not lead to failure. A lowerfactor ofsafetyis appropriate for the piles subjected to the peakloadsin a row than is appropriate for the average capacity of the whole row.Past performance has indicated that current factors ofsafety for averagecapacity are generally satisfactory; so it is unnecessarily conservativeto apply the same factors of safety for the exceptionally loaded piles. Since method 7 is likely to be used more extensively in the future consideration is urgently required as to what factors of safety are appropriate for individual piles. Methods 4 to 7 ofthe following table all make use ofsubgrade reaction to some extent to resistlateralloads. While this added restraint maybe appropriate for a pier foundation,it is not alwaysclear whetherit is appropriate for an abutmentwhere the flow ofground due to embankment surcharge maywell apply active loadsto thebacks of piles (see Section 7.1 and 7.4). However when the upperstratumis firmenough to provide subgrade reaction, raking piles might be reducedor avoided.
Factorsof safety
Subgrade reaction
63
7.6 (contd) Pile groups Summary ofmethodsof calculation ofloadsin pile groups Reference
Method 1 Static method (hand)
2 Rivet group (hand)
3 Elastic centre
(hand or computer)
Details
Comment
CP2 p 154.
Asystemofaxial pile loads is calculated which are simply in equilibrium
with applied loads.Some designersfirstconsider the pile cap as a footing and determine distribution ofbearing pressures,and then placepiles as necessary to carry loads. In the staticmethodand the rivet groupmethod theprimary forces are axial loads,and no account is takenofsecondary moments due to flexure of piles. The methods are thoughtnot appropriate by somedesignerswhen upperstrata are exceptionally soft and pile bending is significant: a frame analysisis then used.
Usage
Commonly used for routine design
CP2 p 151 with simple assumption below for raked piles. Piles are firstconsidered vertical at their positions under the cap and an elastic distributionofaxial load is found in equilibrium with vertical cornponentof applied loadat its positionunder the cap. Then a numberofpiles are raked so that the horizontal components oftheir axial loadsbalance the horizontal component of the applied load. See comment on static method.
Commonly used for routine
CP2 p 155
Commonly used for routine
The piles and cap are considered to form a trussor frame.The cap is assumed to be rigid while the piles are elastic and are supported on rigid supportsat their feet (or equivalent point ofsupport in the length ofthe piles). The cap/pileand pile/support connection can be pinned or fixed.If theconnections are pinned thesystem is then a trusswith only axial loads calculated for the piles. Ifthe connections are fixed bending moments are also calculated. Some engineers design for the worst conditions ofpinned
design
design
and fixed connections.
4 TAMS (computer)
References and example in Appendix of: Dixon H H & Berry D W (1970). Extensions to the Chania-Sasumuawater supply scheme for Nairobi. Proc.
I.C.E,Vol 45,January,pp 35—64.
S Plane frame (computer)
6 Space frame (computer)
7 Elastic half-space (computer)
Footnote on survey
64
Piles are assumed embedded in soil and the stiffnesses of their tops subject to lateraland axial loadsare calculated. Then a stiffness analysis is carried out for thecap supported by springswith lateraland axial stiffnessesof piles. Severalbureaux and 'in house' computer stiffness programmes are used for pile group analysis.Some are ordinary structuralanalysis programmeswhile othershave beenspecially written for pile groups. A wide variety of different assumptions are possible, and are used,relating to: pile/soil axial interaction, lateralinteraction, effective supports,cap stiffness, fixityofconnections.
Used by a few for routine design
Used by a few for routine design. Also used for complex conditions
Three dimensional analysis with facilities of plane frame.Used only for exceptional three dimensional problems.
Exceptional cases only
Department ofTransport Highway EngineeringComputer Branch, HECB/B/7PGROUP. A threedimensional flexibility analysisin which piles are assumed embedded in an elastic halfspace, and pile/soil interaction occurs at a discrete number ofpointson each pile. This methodhas beenavailable for a short periodand has beenfound useful by a few designers.But it is considered by many to be more complicated than is warranted from currentknowledge ofpile behaviour,
Used by a few for routine design or checking. Also used for complex conditions
Of 21 statementsof practice recorded during the Survey: 17 used the simple methods for routine design (static method 4, rivet group 7, elastic centre6), but ofthese 5 used a frame analysis for special conditions; 3 used TAMS or plane frame for routine design and 1 used PGROUP. Some designersused one methodfor design and anotherfor checking.
7.7 Pile tests Comment
Aspect
'Trial' piles to investigate performance
'Trial' piles are generally loadedto failure, ie large displacements,in order to obtain as much information about the groundas possible. (Tests to only 1½ times working load seldom provide information about ground capacity.)The constantrate of penetrationtest is being used increasingly, because it is quicker than incremental tests and for trialpiles often givesa clearly defined failure load.
It is usually foundimpractical to useearlytests of piles in maincontract as investigatory
since the cost ofdelays waiting for results canbe considerable, and often potentialeconomies in design cannotbe achieved because informationis too late. Maximum benefitis most likelyto be achieved ifinstallation and performance can be checked in an advanced contract(seeSection 2.6). One advantage ofan advanced contractis that piles can be left for a longer time beforetesting; so enabling the soil'sstrength to recover. Testing in a rushduring the maincontract can seriously undervalue long term performance. The number of 'trial' piles required depends on the differences of ground conditions over site. If the working pilesbear on a rock substratum ofknown competence 'trial' piles may notbe needed.
'Test' pilesto checkquality of construction
'Test' piles are generally working piles which are loadedto 1½ times working load. Some engineers consider themto be ofquestionable value, except where gross defects are discovered,since they give no indication ofdefects whichcause long term deterioration. Results can be dependenton the position ofthe pile in the group. A pile in the middle of a group can appearmuch stronger than piles at the edge, particularly ifdriving hascompacted ground between.
Most designersretainthe rightto test any pile during the contract.It is not unknown for thelast piles placed to be oftheworst qualitybecause oftherush ofthe contractorto leave site. The number oftests required depends on size ofcontract and difficulties ofinstallation (see Appendix B). In general designers expect to test one pile per abutment or small bridge. On a small contract it maybe cheaper to makethe design more conservative and not test. It canbe very difficult to testlarge diameter piles, but in general their quality can be inspected visually more satisfactorily. Load application
Forsmall piles kentledge hasbeenfound convenient and reasonably representative of future load form. On soft ground attention has to be paid to temporarysupport of kentledge. Forlargepiles kentledge can be uneconomic and unsafe and reaction is often provided by tensionpiles or anchors.Attention has to be given to spacing and interactionin the groundbetweentest pile and anchor piles to ensure that the test pile is transferring load to ground in the same way as the working pile. With kentledge or tension piles the reactionshouldbe applied to the pile top by a jack.
Raking pile tests
See Section 7.5 concerning need to test on rake. Raking piles have been tested with kentledge on two piles ofopposite rake, and also by reaction with raking tension piles.
Dynamic driving records
Dynamic pile drivingformulae or analyses based on the wave equationhave not been found to provide a reliable indication ofpile capacity. However a set empirically calibrated for one pile on a site can be used for comparison ofother similarpiles in other parts ofthe site ifgroundconditions are virtually identicaland driving procedures are the same.
Integritytesting
Methods are reviewedin reference below. In general directmethodsby excavation or coring are themost trustedbut are expensive. The indirect methods relying on sonic, electronic or radiationmeasurements are less reliable and require interpretation.They have the inherent disadvantages of non destructive methods for concrete above ground together with problems ofinterpretation ofunknown shapes in unknownsoil. Consequently many engineers are sceptical of their use at present. However several techniques are progressivelybeing developed and some usefulapplications havebeencarried out.
Reference
Weltman A J. (1977).Integritytesting ofpiles — A Review. ReportPG4,CIRIA, London. 65
CHAPTER8 DETAILS
8.1 General comments The careful design ofsubstructure details can have as much influence on economy ofconstruction as the basic choice oftype ofstructure. Comment Influence on construction programme
Contractors fmd that small details can cause delays in construction with consequential cost out ofall proportionto the importance ofthe detail.This is most likely to happen whenthe design ofthe detail restrictsthe contractor'schoice of construction sequence so that he cannot progressother much larger parts of the contract.
Construction during inclement weather
Construction offoundations is dominated by weatherconditions. Contractors beseech designers to consider during design and preparation of specifications the problems ofexcavating to line and level, blinding, and stabiisingvertical faces during periods of inclement weather. A foundation that can be completed in a weekis not only likely to be very much more economic thanone that requires several processestaking several weeks, but it is also likely to performbetter since the ground will not be softened as much by exposure and long-term release of pressure.
Construction simplicity
Designersand contractors stress the importance of simplicity in design of details as wellas in design ofoverall structural form. Severaldesignershave found that optimisation of designs with respect only to assumed cost-rates and not simplicity of construction tends to result in compact and complexdetails which are not only difficult to build to the desired qualitybut also take an undesirably long time to complete. For these reasons reinforced concrete walls and bases are generally more satisfactory when designed with more than the theoretically economic thickness. Flexural stiffness is increasedand steel fixingand concreting are much easier.
67
8.2 Excavation shape andshear-keys Comment Level
formation
Ground disturbance from shearkey
Shear-key in
front
The simplest abutmentfoundation has a level formation. Sliding must be resisted by friction underthe base, if passiveresistance is ignored in front ofthetoe.Oftena wider base is needed for sliding stability than for bearing or overturning stability. For this reason some designersdimension thebase for thelatter criteria and provide additional resistance to sliding by a shear-key or by inclining formation.However the cost of additional widthofbase or greaterdepth is often no more than that ofthe other remedies and the construction problems and ground softening can be less severe. (To minimise ground softening some designersgive the formation avery small cross fallto improve drainage.) The process of constructing a shear-key frequentlyloosens and softens thevery material on which reliance is placed for passiveresistance. With most ground conthtions a level formation canbe excavated simply and quickly. But if a trench 1.5 m deepis subsequently excavated and is open for a weekor so,then some slippageor loosening usually occurs. Sometimes,the trench acts as a drain for the rest ofthe excavation. Unfortunately shear-keysare most needed in soils which are least able to stand during construction without softening. Ifshear-keyshave to be used then it may be best to design them so that they can be excavated and concreted as a continuousprocess with the blinding and with simple dowels insertedto transferhorizontal loads. The edges can be battered so that theyare stable without movement. Alternatively the shear-key canbe placed in a trench from ahigherlevel with concrete pouredas excavation proceeds. Precise controlofdimensions is seldom important. Shear-keys are generally not placed at the front of footings because ofthe risk of the lateralresistance beinglater removed by excavation for services. Furthermoresoftening ofthe ground underthe toe is much more critical thanelsewheresince it reducestheoverturning stability. For thesame reason the toe needsto be at least 1 m below groundlevel to avoid softening due to frost action, and land drains shouldnot be placed near the toe.
Undermiddle
Shear-keys placed underthe middleoffootings pushduring sliding against soil that is under heavy vertical compression. Excavation can be simpler than at either edge ofthe base since amachine can straddlethe trench. Ifthe reinforcement in theshear-key is continuous with that in thewalls construction is more complicated since the shear-keythen has to be constructedwith the accurate line andlevels ofthe wall.
At back
The shear-key at the backalsoobtains lateralresistance from ground subjected to vertical compression. The possibility of sliding up inclined planes may needtobe checked as wellas sliding horizontally. Excavation at theback of thebase is likely to cause less disturbance than underthe middle, but the depth of excavation on the retained sideis greater.
Inclined formation
A footing with inclined formation can have simpler details than one with
a shear-key. The inclined formation requires deepexcavation of there-
tained side, but the formation shouldnot suffer the same softening that occurs during the construction of a reinforced concrete shear-key. Howeverexcavation and construction on a slope is time consuming and expensive, andit maybe cheaper to constructthe full widthat the lower depth. Increased tolerance and cover are often found desirable. Complex
68
Somedesignersand contractors consider that the time required to construct complexfoundations not only leads to unnecessarysoftening and disturbance of the ground but is also unnecessarily expensive.
•1
ThI
8.3 Substructureform and reinforcement Detail
Comment
Form Shape
Several designersand contractorsconsider that a simple shape is not only likely to be cheaper to build,but alsomore likely to be to the standard ofquality required. It is mucheas&er lo build with surfaces plane and vertical. Ifa wallis curved in planthen cantruction with a batter can be very complicated. Ifa wallhas battersto the faces and a reducing height then formworkextending above the two faces impinges and restrictsaccess for men and materials. Cutting the formworkto fit exactly can greatly increase the costby preventing re-use. Ifan abutmenthas a forward or backward batter to the front face and wingwalls in different planes then it can be very difficult to make the formwork at thecornerand also difficult to supportit adequately. Haunches at the intersection ofretaining wails and bases have been used often in the past, but do not appear to be very common at present. They complicate the constructionat thekickerand compaction of theconcrete maybe poor. It is possible that they improve the performance ofthe wall/basejoint, but detailing is complicated. Haunches are sometimes used at the corners ofportal frames and culverts to reduceshearstresses and so avoid links.
Poursize
It is becoming increasinglyeconomic to construct abutmentsand piers in single large pours from footing to bearing shelf. Some designersthicken up such walls a little, if necessary, to make it easier for men to climb inside the formwork. The standards of workmanship of formwork, steel fixing, cleaning, and concrete pouring and inspection are all likely to be better thanin intricate and slender structures.
Tolerances
Formwork Cost
Bases
Reinforcement
Careful thought needs to be given to the tolerances; as to which matterand whichdo not. Many dimensions ofsubstructures are not critical. Sometolerances are controlled bythe needfor parts oftheconstruction to fit together,someby the ride ofthe road, and some by the needforgoodvisual appearance.
The cost offormwork is a significant proportionof the total costofsubstructuresand considerable savings ensue where multiple re-use is practised. Wastage is reduced if walls are designed so that lengths between construction joints are standardised and if possible relatedto the dimensions ofstandard sheets of piywood. The heights ofwalls, levels offeaturesand stepsin levelsbetween footings shouldbe selected so that as much as possible canbe constructedwith a singleshutterwithout adjustment or unnecessarily restricting the orderofthe construction. It is importantduring the design oftaperedpiers to consider therestrictions on construction sequence that might result from having to modify the formworkto suit different heights. Bases can sometimesbe constructed most economically by casting concrete directlyagainst ground;ie without formwork,and possibly using expendable temporarysupports. Anothermethodis to constructa concrete box in advance with internalforms: the methodis useful where thereis a lot ofrepetitionand the formwork can be struckquickly and re-used with steel fixing following in a clean dry space.
'
.
.
- - —J
L
Contractors findit is much easier to fix reinforcement in vertical and horizontal planes than inclined planes. Fixing ofcorners and joints is much quicker ifone plane can be fixed at a time. One contractorhas found that it takes about twiceas longto fix the reinforcement in abutmentsand piers than in foundations because ofthe problems of knittingheavybars togetherin differentdirections. Fixing is mademuch more difficult ifa crane has to be used to support heavybars in one horizontal direction as they are threadedthrough bars in another plane. Control of dimensions can be difficult with heavy bent bars because of bending tolerances. It is also much easier to fix medium sized bars than numerous small bars. Congestion should be avoided. 69
8.4 Construction and movement joints Detail Construction joints
Comment Contractors fmd that positions ofconstructionjoints and restrictions on lengths of pour have significant influence on the programme. Consequently ifdesignershave stringent requirements they needto be stated clearly in the contract documents and not be left simply to theapproval of theresident engineer. The positions are usually chosen to suit shrinkage and temperature movements of concrete,particularly if crack inducers are not spaced regularly along the pour. However for walls the maximum length of pour is often controlledby the size of shutterthat can be lifted by the availablecranes. Frequently a length ofpour ofabout 10 m is foundmost practical. Making construction joints with stop-ends, and scabblingbeforethe subsequent pour is expensive. Sometimes it is even necessary to incorporate a water bar to prevent penetration through the joint. Removal oflaitenance is not always practical. It maybe more practical and cheaper to form a proper movement joint where strength across the jointis not required. But a movement joint requires exact location with exactreinforcement scheduling and so will not allow minorvariations of construction sequence.
Movement joints
There do not appear to be any defmitive general guidelines concerning the positioning of movement joints. Some designerstry to avoid them ifpossible, and only locate them at changes of direction and sectionto absorb shrinkagemovement, and at changes ofload intensity(as between wingwallsand abutments) and of ground conditionto avoid settlementstresses. Elsewhere they try to controlcracking by reinforcement and crack control joints. Frequently movement joints are designed to accommodate relative movements of25 mm in all directions. The joints in footings and walls often have a shear-key, with walljoints also having a water-bar. In areas ofmining subsidence widerjoints are sometimes provided and the shear-keys PLan vLew of omittedso that they do not pickup loads for which they Feature are not designed. Care has to be takenat joints between — 4 u parts of a structurewhich are likely to move in different directions. At thejointbetween awingwalland an abutAbuin,e.at ment, tiltingofthe wingwallcan be in a differentdirec- movcrntnt tionto that of the abutment,and a shear-key might providea prop against active pressure movements and be subjected to large loads. Where significant movement is anticipatedsome designersprovide agenerous feature
''
Crackcontrol joints
Seals and provisions for leakage
70
(say 75 mm)tohide it. Crackcontroljoints at about 3 m centres have beenused by somedesignersto control the spacing of cracks in very longwallsdue to shrinkageand thermalmovements. Concrete is poured in monolithic form across them and cracks are inducedduring curing by deep grooves on the opposite faces. Cracking maybe assistedby plastic tubes up the middle of thewalland partialdiscontinuity ofreinforcement. A specialwater bar can be attachedto thebackshutter or internally to cover thecrackonceit has occurred. Crack controljoints shouldgive the contractormore flexibility, aswithin reason he may form construction joints where he chooses. TurtonCD. (1974). To crackor not to crack. Concrete, Vol 8, No 11, Nov, pp 32—36. Hughes B P. (1973). Early thermalmovement and cracking of concrete.Current Practice Sheet No 3PC/06/2. Concrete, Vol 7,No 5, May, pp 43—44. Many designersappearto be disifiusionedabout the effectiveness ofpolysuiphide and other plastic seals. Many proprietary seals appearattractive on paperbut are very difficult to install properly under site conditions, and several types deteriorate in a relatively short time. There has in the pastbeena tendencyto follow the lead of manufacturers to use bigger and more complicated typesofjoint. However many designersnowtry to design as simple a joint as possible. Adequate provision is madefor drainage ofwater that does leak through in order to avoid discolouration ofvisible structure.Drainage channels and pipesare large enoughto take the full flow ofwater thatmightenter them and also large enough to be properly cleaned out during constructionand later. A 25 mm diameter hole is likelyto block.
8.5
Topofabutment Detail
Curtain wail form
Comment
Abutmentcurtain walls are frequently designed with complexity of detail out of all proportionto their importance. If the curtain wail is supported on a corbel off the backofthe abutment,then the cost ofmaking the shuttersand fixing intricate reinforcement maybe much greater than constructing a thicker abutmentwall with plane backface.Even quite a large thickening ofthe mainwall can be economic ifit avoidsthe needfor falsework and a soffit shutteron a large over-hang.The saving in time can be of greater significance. In addition compaction of the backfill against a plane face is likely to be more satisfactory than around ledges. Expensive delays can ariseifthe backfill has to be compacted up to the level ofa corbel soffitprior to construction ofcorbeland curtainwall. Contractors have sometimes found that insufficient clearances havebeen provided for properconstruction ofcorbels.
Loads
Curtain walls aresubjected to concentrated vertical andhorizontal loadsfrom the wheels ofvehicles. These loadscause larger local moments thanmight be predicted from considerations of equivalent overburden. Several engineers design curtain walls to resist these moments, having assumed some simple spreading ofthe loadsthroughthe structure,such as at 450•
Reinforcement
On post-tension bridges the curtainwallhas to be detailed so that it can be constructed afterthe deckhas beenpost-tensioned. Ifnot detailedappropriately thereinforcement mayget in the way of thejack.
Longabutment
The reinforcement along the top of an abutmentis often light in comparison to that in thefootings;it is sometimes appropriate in longwalls to increase this reinforcement at the top to provide bending strength or crack control in the event of the ends of the abutmentsettlingmore than the centre.
Wingwalls to bearing shelf
The small wingwalls at the ends ofthe bearing shelfare also commonly designed with unnecessarilycomplicated details. It is not always clear at the designstage whetherit would be more convenient to construct these beforeor after placing the deck. If possiblethey should be designed so that the construction sequence can be determined during the contract.
Drainage of bearing shelf
Some designersconsider that in the past bearing shelves on someabutmentshave not had adequatedrainage, as is demonstrated by thestaining over the outside face.A
considerable quantityofwater can flow down through theexpansion joint,and some designersconsider itwise to provide drainage of sufficient capacity to carry the full flow that might enter in the eventofjointseals being ineffective. They also design the channel along thebearing shelfto be accessible for cleaning. A few designersplace the channel in front ofthe bearingsto make access simple. (Ifaccess is very difficult it may not even be possible to clear out construction debris completely.) The falls along the channel needto be more than adequate since nominal fallscan be obscured by unfavourable tolerances. The effluent pipes need to be roddable. Whilemany engineers slope these pipesdown into the abutmentbackface drainage layer,an increasingnumberare providing independent roddable down-pipes to soakaways.Some designers provide drainage channels along the tops ofpiers if there arejoints in the deckwhich could leak if they deteriorate.
71
8.6 Drainage andwaterproofing Detail Drainage of abutmentback face Need
Capacity
Comment
Many designersstressed the importance of positive drainage down the back face of abutments. Not only does such drainage prevent the building up of high pressures against the wall,but the drainage alsoprevents saturation ofthe fill behind the abutment.During a storm, water flowing throughthe porous sub-baseof the road and service ductsmeets an obstructionat the abutmentand water collects which must be drained away. During construction such flooding ofthe fill can cause ponding and increases subsequent settlement.Drainage needs to be more than adequateto remove thefull flow that could enter during a storm.
The difference in cost between inadequate drainage and more than adequatedrainage is small, and because ofthe importance ofthis item skimping is usually considered not justified. Severaldesignersstressed that the drainage systemneedsto be positive, testableandmaintainable. Currently, a porousblock layerdraining to a pipe capable ofbeing cleared by rodsis generally considered the most satisfactory form of drainage membrane. A450 mm wide dry stone wall may be even more satisfactory in regions where the material and skilled labourare readily available. A layer of sand is difficult to place satisfactorily and even when it accords with a standard specification it maynot be sufficiently permeable. The specification for concrete blocks shouldnot require toohigh a strength otherwise their permeability is likelyto be inadequate.
bLok5
Roado5Le
In addition to providing a positive drainage system, many engineers also incorporate weepholesto provide relief for drainage in the event ofpipe blockage, and visible evidence ofsatisfactory drainage. They are sloped backwardsto prevent weeping and staining whendrainage is satisfactory. Design water pressures
Designersoften assume that there will be no water pressures affecting stability if the wallhas adequatedrainage. However it has to be remembered that the drainage layer is not necessarily at the effective back of the wail in the stability calculation. Sometimes consideration has to be given to water pressures if the backfill is not permeable or if the abutmentis constructedagainst an existing impermeable slope. If adequate drainage cannot be provided behind the abutmentthen some water pressures must be considered. In any case CP2 recommends that it be considered up to the level of
thedrainage outlets.
72
Drainage of bearing shelf
See Section 8.5.
Waterproofmg
The backfaces of most abutmentsare given some form of protective coatings: often two coats of bitumastic paint. Opinions differ as to the value of such coatings; some designersconsider they achieve nothing and are a complete waste of money,while others consider that they provide an inexpensive additional protection to reinforced concrete even if not full waterproofmg. It is thought that coatings reduce seepage of water throughthe concrete,whichin turn reduces the amount of dirt which sticks to the front face. Such protection is not adequatefor structures which are likely to be below the water table and must hold water out, such as subways, and these have to be properly tanked and drained.Clause NG 1601 of Notes for Guidance on the Specification for Roadand Bridge Works givesadvice on protection of concrete in corrosive ground.
8.7 Backfill and run-onslabs Detail
Comment
Abutmentbackfill Material
Designersin most parts ofthe country expectto use afree-draining fill to obtain the benefits of low earth pressure and greaterpredictability of performance. Granular fill also has the advantage that compaction in confinedplacesis easier.
In someregions selected granular fill does not existnaturallyand can be very expensive. In such situationsit can be more economic to design the abutmentsto support the
higher pressures ofthe less satisfactory fill available. But many designersremain cautiousabout using such fill because there is little reliable information on the design pressures. Relatively impermeable materials, such as chalk, have beenused but serious settlementhas occurred on occasions after inundation. In some regions pulverised fuel ash is availableat an economic priceand has beenused successfullyas a standard backfill behind abutments. Where shale and rock fill have beenused it has often been necessary to limit the maximum particle size, particularly if piles have to be placed throughthe fill. Whateverfill is used,somethought should be given to its interaction with the embankment and abutment;an abutmenton piles has failed by backward overturning caused by the settlement ofa rigid mass of cemented fill.
Zone of free-draining fIll
Compaction
Designers'practices differ as to the size of zone specified for the free-draining fill. The sketches indicatesome examples: not all are buildable. The most appropriate shape ofthe zone depends on the construction sequence: so it is unwise to have too rigid requirementsin the design. For walls in a cuttingonly a small zonemay be required. Whatevershapeis designed it is essential that it shouldbe constructable with proper compaction. Some engineers pointedout that it can be impossible to compact fill to the 600 slope that is sometimes indicated to suit a Coulomb wedge. Some engineers design the interface to have a slope of 1 in 1Vz or lessso that the face can be properlycompacted ifone material is placed in advance oftheother.
It can be very difficult to prevent settlement of the backfill behind abutments,and many designers stressed the needforthoroughcompaction. It appears that sometimes even propercompaction does not eliminate it. However contractors are reportedto often underestimate the care that is neededto achieve propercompaction, and supervisionis not always adequate.Compaction in the corners ofa structure is difficult and can only be done by small plant. Some designersspecify an increase in the number ofpasses ofcompaction plant above that generally specified forthe particular material in orderto achievethe same density at least as the approach embankment. (If it is possible to passa large roller along the backofan abutmentthen care shouldbe takenthat compaction pressures do not greatlyexceed the design active pressures). Adequate compaction is particularly difficult to achieve beneathparts ofa structure,such as underwing walls, and where thereis no firmbackground against which to compact. It has beenfound advantageouson occasions to place mass concrete wallsunder parts of skeleton abutmentsand cantilever wingwalls to provide such a firm background. 73
8.7 (contd) Backfill and mn-onslabs Detail Run-on-slabs
Comment Designersand clients generally advise against the use of run-onslabs exceptfor special situations. They used to be more popularbut serious problems of maintenance havearisen. It has been found that making up road levels after embankment settlementhas generally beeneasier and less expensivewhere bridges have not had run-on slabs. It is considered more economic to rely on proper compaction ofbackfill and to makeup the road surfacing as settlementoccurs.
Overdigand working space backfill
74
Backfilling to overdig and working space around foundations is usually done with selected granular fill, where small quantitiesare involved, unless the fill has to transmit bearing pressures or passiveearth pressures in which case concrete fill is generally used.Some engineers consider that if thereis a risk that backfilled excavations might act as reservoirs for water which could soften foundations, then fIllwith a clay content maybe more appropriate.
CHAPTER9 PERFORMANCE,MAINTENANCEAND REPAIR
9.1 Monitoring ofperformance Monitoring ofperformance has in the past only beencarried out in specialcircumstances whenproblems have beenanticipated. However several engineers during the Survey expressed a requirement for more case histories of performance of different types ofbridges in the various regional conditions. The table belowsummarisessome of thecomments on monitoring performance and on possibleprocedures. Comment
Aspect Value
Regular measurements ofline and level of bridge substructures and adjacent ground are ofvalue because: 1
The designer(and others) can develop a muchbetterunderstanding ofthe global behaviour ofdifferenttypes ofbridge and theirenvironments, and so gain an idea ofwhichdesignsperform best and which calculations are relevant.
2 The case histories of full size structures on various ground conditions provide a reliable basis for predictions for future bridges in the region. 3 Observations during construction identify the speed and extent of movements, and so provide a basis for deciding ifprocedures can be speeded up. A significant savingmay be possiblein a contractinvolvinga long pause or considerable repetition.
4 Awarning is obtainedof excessive movement and ofinsufficient support of particular points. Without ahistory ofregular readings it is impossible to tell within a short period after a problemhas been identifiedthe difference between long-term movements and short-term fluctuations and errors.
Procedure
Measurements require careful planning, preferably at the design stage,so that all partiesinvolvedare fullyaware ofwhat is being done.One personshould be responsible for the planning, installation, reading and maintenance of measurement points and particularly datum points. Severalseriesofobservations have beenrendered useless because ofuncertainties about datum positions.
The frequency ofmeasurements needed depends on the ratesofmovement. If measurements are initially taken frequently, such as once a month, then it quickly becomes apparenthow much less (ormore) often the measurements are really needed. Equipment
The instruments and measurement pointsneedto be simple, reliable, stable,inexpensive and easy to install, and above all robust and durable. A few organisations are equipped with precision levels for monitoring settlement. However for the majority of bridges properly calibrated site equipmentshould be satisfactory. Structures often settlemore than is anticipated or appreciated without significant signs ofdistress and measurements with an accuracy ofonly about 3 mm can still be of considerable value.
Contract implications
Monitoring procedures can interfere with construction procedures. Positions ofdatum points and required sight lines need to be indicated on contract drawings otherwise additional costs can arise.
Reference
Symposiumon field instrumentation. Butterworths,London,1973. Burland J B. (1977). Field measurements — Some examples of their influence on foundation design and construction. Ground Engineering, Vol 10, No 7, October. 75
9.2 Inspection and maintenance considerations during design Severalengineers stressed the importance ofconsidering futureinspection needsduring design. Although considerable effort is usually made to minimise the needfor futuremaintenance it is considered prudent to make access provisions for subsequent inspection, cleaningand repair.The table below listscomments on aspects of maintenance which affect substructures. Comment
Subject Inspection facilities
During the design of substructures, consideration needs to be given to what facilities will be required for inspection and maintenance of: 1 Damage from scour. The river authority should be consulted. See Section 2.2.
2 Cleaningand testingof drainage to backface of abutment. See Section 8.6. 3 Cleaningand testingof drainage to abutmentshelf. See Section 8.5.
4
Cleaningofdrainage of otherjoints and crevices.
5 Inspectionof deckbearings and anchorages and provision of pockets for jacking deckfor repairs. Bank seats and buriedabutmentsneedsufficient headroom (900 mm minimum) and level standing area,with steps for access up paved slopes. Tall piers may needinspection galleriesand access thereto.
6
Effects ofhorizontal movements of abutments above soft ground. Expansion joints and bearings can often be designed to accommodate greater movements than anticipated from calculationswithout much increasein cost.
7 Damage from subsidence, particularly mining subsidence. Reductionof headroom can be a problem due to settlementof substructures and resurfacing ofunder-road. Reliance on maintenance
Designers generally try to design their bridges so that the maintenance required is a minimum. Howeverit has on a few occasions beenadvantageous to place a continuousstructure on subsiding foundations and make regular adjustments of levels at supports byjacking. Severaldesignersstressed that this thouldonly be done whenthere is no doubt about the maintenance being carried out.
Feedback of information
Several designersstressed the needfor feedback ofinformation on maintenance problems to the design office,even though the designersresponsibility may terminate at the end of the contractmaintenance period. liaison with the maintenance authoritywas felt to be beneficial because: 1
Feedback of the defects in construction and performance provides guidance for improvements in future design.
2 Observationsofperformance such as settlementof real structures provide the only reliable basisfor the prediction ofthe performance of similarstructures on similargroundconditions. (See Section 9.1). 3 An understanding ofthe performance ofold structures can greatlybroaden a designer's perspective. For example the performance ofslender masonry
retaining walls of traditional constructioncan defy calculations. The difference between safeand unsafe is evident whendrainage becomes defective.
References
Department of Transport. Technical Memorandum BE4/77. Inspection of highway structures. British Railways Board. CE Handbook No 6, Examination of structures. London Transport Executive. Manual ofinspection ofbridges, structuresand buildings.
76
93 Repairs
The followingnotes list someofthe comments madeduring the Survey about repairs to substructures. Problem Improvement to stabffity of abutments and walls
Repair
The stability of abutments and retaining wails has beenimproved by: 1 Anchoring to deadman orgroundanchors.
2 Increasingtoe. 3 Increasingback(possibly with deadman anchors for temporarycondition).
4 Constructing diaphragm wall behind. 5 Underpinningby hit-and-missmethod or small boredpiles or pall radiche piles (defective piles have beeninspected from shafts and tunnels).
6 Treating scourof rockyand granular river beds by refilling with concrete bags spiked together with reinforcement bars.Scour ofsilty river beds has required underpinning.
P
7 Grouting. Care hasto be taken that groutpressures do not exceedexisting ground pressures otherwise more harm can be done than good.
bridge
Whenabridge has to be replaced by another on the same ground advantage can sometimes be taken of the fact that the old bridge has already consolidated the ground beneath.A borehole alongside the old foundations will not necessarily indicate the state of the groundbeneath the foundations and it may be necessary to prove the groundby drilling. Ifa bridge has to be widened then differential settlementbetween new foundations and old foundations can be a problem.
Refurbishing of
Masonry bridges have beenrefurbished by:
Replacement or widening of
masonrybridges 1
Repair ofscourdamage to river bed round foundation.
2 Grouting ofvoids in substructures. Care hasto be taken that grout does not block drainage and ducts.
3 Repointing ofsubstructures. 4 Repointing ofarch. 5 Archis supported while road surface and fIll are removed: fIll is replaced by concrete with light reinforcement.
6 The road is resurfaced and waterproofed with filletat parapet to throw water awayfrom the parapet.
77
CHAPTER 10 ADVICEAND INFORMATIONNEEDS
10.1 Disseminationofinformation Survey of adviceand information needs
Many of the engineers interviewed during the Survey (see Introduction)mentioned theaspects ofdesign ofbridge foundations and substructures for which they required advice or more information. These needs are reportedon the following pagesin four main groups: 1
Guidance on aparticularaspectof a design from someone whohas experience ofsimilar problems on other projects.
2 Advicewhich might be considered part of a general lecturecourse on design of foundations and earth retaining structures. 3 Particular shortcomings in currentknowledge which might be filled by a research project orby a clear presentation and criticalreview ofexisting information.
4 Improvement throughoutthe industryofa common practice or technique which is currentlyimprecise. Finding the adviser
Updating this report
Most of the information needswere of category 1 aboveand were concernedwith obtaining guidance on particularaspects of designs, such as information on the behaviour ofa particularregional soil. In many cases it was thought that the necessary information already existed and that the real problem was in finding the 'expert' whomight be ofmost help.Severalengineers thought there was a need for a national centre for geotechnics which might provide a reservoir of technical expertise but which would alsoredirectenquiries to expertsoutside, particularly for problems ofa regional, practical, or contractual nature. It was felt that often the'expert' was likely to be someone with broaddesign and construction experience who could assessthe importance ofthe problem, and then advise on how to avoid or solve it. This report could help designerskeepup to date with current practice in the design ofbridge foundations. It is hopedthat designerswill report developments in practice and deficienciesof this document to:
The Headofthe Geotechnics Division, BuildingResearch Station, Garston, Watford WD2 7JR,
79
10.2 Advice and information needs
I
Guidance on particularaspects ofdesign andconstruction Some of the followingneedsmight be helpedby general research projects. But in most cases the problems depend significantly on the details ofthe project,and specific advice based on existing experience is likelyto be of more directbenefit. Related pages of Subject
Site investigation
1
Advice on appropriate drilling and sampling methods for various typesofground.
2
Properties and behaviour of soils for different regions, eg Gault clay,Oxford clay,soils of the Midlands, and so on.
3
Appropriate methodsand procedures offield and laboratory testing for assessment of design parameters ofvarious soils, eg plate and pressuremeter testing, interpretation of oedometertests for settlement.
15,83, 84
4
Guidance on the properties ofvarious types of fill with differentdegress ofcompaction.
See Index
5
Advice on hydraulic surveys.
8
6
Advice on survey for, and design ofbridge over old mineworkings.
9
7
Rationalisation ofload combinations.
41
8
Adviceon acceptable levels ofdifferential and total settlement.
36, 45
9
Guidance on appropriate methods ofestimating total and differential settlement.
45,46, 60
and soil properties
Choice of foundation, design and analysis
this report
Need
10,83,84
10 Guidance on the appropriateness ofpiles or spread footings for differentsituations.
33, 34,35
11 Advice on buoyantfoundations and other alternatives to piles on soft ground.
26, 33,34
12 Assessmentofsignificanceofsoft lenses ofsoil at depth.
33,34
13 Selection ofappropriate undrained and drained soil para-
42
14 Advice on groundtreatment,eg when can vibroreplacement rock columns be used underabutmentsand bank seats.
37
15 Advice on the design and construction offoundations on fill.
See Index
16 Choice ofpile for various groundconditions.
58
metersfor design undervarious types ofloading.
Pile
foundations
80
on capability and shortcomings ofpiling plant (someengineers considered this more importantthan advice on choice ofpile).
17 Information
58,62
18 Ranges of skin frictionofpiles in various soils.
60
10.2 (contd) Advice and information needs 1 (contd) Guidance on particularaspects of design andconstruction
Pile foundations (contd)
Relatedpages of this report
Need
Subject
19 Longterm loadtransferalong pile shafts,especially clay soils.
20 Guidance on design and analysis for piles subjected to lateral
61
21 Bendingof piles in ground and effective point offixity.
65
22 Adviceon measurement ofsubgrade reaction.
61
23 Advice on pile group design and optimumarrangement.
63, 64
24 Advice on testingvertical and raking piles.
65
25 Design of cantilever diaphragm walls.
20, 47
26 Design of footings forvertical andlateral load.
43
27 Guidance on choice oflateralpressures in relationto soil
47—50
loads.
Retaining walls and abutments
type, arrangement and stiffness ofsupport system, and methodofconstruction.
28 Guidance on the design of skeleton (spifi-through)abutments. 50 29 Advice on design of flexible corrugated metal structures.
25
Methods of construction
30 Advice onconstructionmethods and costs.
2, 3,4
Details
31 Advice oncorrosion of, and chemicalattack on,
24,72
foundations. Monitoring
32 Advice onwhat to measure, when and how.
75
2 General advice which might be includedina lecturecourse onfoundations andearth retaining structures Soilinvestigation and properties
Soil/structure interaction
Pile foundations
1
Standardisation ofsoil description and logging.
12
2
Advice on validity ofvarious tests for differentsoil types and loading conditions.
83, 84
3
Techniques oflaboratoryand in-situ testingand their interpretation.
83, 84
4
Case records ofthe global behaviour ofstructure,foundations and ground.
39, 40, 55, 56, 57
5
Relationships betweenearthpressure and displacement.
48
6
Occurrence of downdrag.
7
Pile group behaviour and analysis
63, 64,
8
Piles subject to lateralloads andlateral ground movement.
61
81
10.2 (contd) Advice and information needs
3 Particular shortcomings in currentknowledge which might be filledby a research projector by a clear presentationand critical reviewofexistinginformation
Related pagesof this report
Need
Subject
Observations ofthe movements ofmore bridges to determinethe global behaviour ofdifferenttypes ofstructure on thevarious regional ground conditions ofthis country.
39,75
2
Methods ofassessment of Young's modulus, shearmodulus and subgrade reaction ofground.
63,64
3
Properties oftypical soils ofMidlandsand North.
Footings
4
Assessmentofallowable bearing pressure underfootings supporting eccentric and inclined loads.
44
Earth
5
Observationsofearth pressures and associated movements
50
Monitoring
1
Soil investigation and properties
on all typesofretaining walls and substructures, including
pressures
reinforced earth.
Pile foundations
6
Information on earth pressure resistance to fluctuating braking and temperature loads.
7
Development oflimitstate theory for earth retaining structures.
8
Efficiency ofpile drivinghammers, particularly on rake.
58,62
9
The abilityof piles to carry loadin tension.
60
10 Observations ofloadsin full size pile groups. 11
Contribution ofpile cap to pile group performance.
49
63
56, 63, 64
12 Development ofa general methodofpile group analysis whichconsidersembankment/abutment/cap/pile/soil behaviour. Method shouldcover all situations or should comprise a numberofsimple methods with proven ranges of application ofadequatecoverage.
64
13 Factors of safetyappropriate to more rigorous pile group
63,64
design methods. 14 Testing ofpiles to relate to group behaviour.
65
4 Improvement required throughoutindustryofa common practiceor technique which is currentlyimprecise 1
82
Soil and rock description and general site investigation procedures and reporting.
2
In-situ testing,particularly ofgravel.
83
3
Sampling soils and weakrocks, and subsequent testing.
83
4
Predicting ofsettlementand ground movement.
45
5
Intregrity testingofpiles.
65
6
Detection ofservices underground particularly where several lie close together.
6
APPENDIXA SITE INVESTIGATIONINFORMATION Quality
Properties that can be reliably determined
Class 1
Classification,moisture content, density, strength, deformation and consolidation characteristics.
Class 2
Classification,moisture contentand density.
Class 3
Classificationand moisture content.
Class 4
Classification.
Class 5
None; sequence ofstrata only.
Soil type Gravel Cobbles Boulders
Boring and sampling
SF!'gives some indication
Directmeasurement only possible from
unreliable for coarse gravel and cobbles. Dutch cone cannot penetrate dense graveland coarse material.
field plate and field sheartests.
Class 4 samples from trial
Field plate,shearand density in-situ tests are the most reliable but require dry pit.
Class 5 samples from boreholesbecause fines washed
ofrelative density in gravel:
out.
In-situ permeability gives someindication offmes: pumping trial better.
Boring using shell below water: water might be added to assist.
Dutchconebest method for
Sandwith water tendsto 'blow' up borehole so
assessingrelative density. SF!'gives some indication but can be erroneously low due to loosening from 'blow'.
loosening sand below,
Field plate tests etc. bestas
Silt
Strength and compressibility
Trial pit much more satisfactory. Advancingbore and sample recovery difficult, needs shell and casing with chisel for boulders,
pit.
Sand
In-situ testing
Samples generally Class S deficient in fines, Class 4 from splitspoon. Class 3 or 2 from piston sampler ifpossible.
above,
Boring using shell, or clay cutter above water.
Field tests best as above.
Class 5
if fmes washed out,
but Class 3 if recovered intact,
Class 2 possible from piston
Direct measurement only possible from field plate and field sheartests. Approximate values ofstrengthcan be estimatedempirically from SF!' and Dutch cone.
!n-situ permeability and pumping tests for permeability.
As for sand.
Dutchcone bestmethod for assessingrelative density.SPT
givessome indication but can be erroneously low due to loosening from 'blow'.
sampler.
g3
APPENDIXA (contd) SITEINVESTIGATIONINFORMATION Soil type Normally consolidated and lightly over consolidated clays
Over
consolidated clays
Boring and sampling
In-situ testing
Strength and compressibility
Boringusing clay cutter when dry, and possibly shellbelow
Field tests bestas above.
Field tests best.
water.
Vane test useful.
Class 4 for intact lumps from
Dutchcone useful.
Laboratory tests of shear strength and compressibility satisfactory from Class I samples. Large samples needed forgreater reliability
shell and clay cutter. Class 2 from open-drive sampler. Class 1 from pistonsampler.
Boringusing clay cutter when dry, and possibly shell below water. Class 1 offirmto stiff
material from open-drive sampler and pistonsampler: Class 2 or 3 ofstiff and hard material. Class 4 and 5 from shell and clay cutter.
In-situ permeability to test clay with silt bands ifrate of consolidation important. Probing to establish thickness.
Approximate estimates of strength possible from Dutch cone and SPT calibrated empirically from field tests.
Field plate,shear and density Field tests best. in-situ tests needed for accurate assessment,but Laboratory tests of shear require dry pit. strength and compressibility practical ofClass I samples, Dutchcone, SF!'and but results affected by can used. be relative scales of sample and pressuremeter fissure geometry: considerable scatter in undrained test results. Drained tests, even of Class 2 samples, have less scatter. Approximate estimates from Dutchcone and SPT calibrated empirically from field tests may be more reliable than laboratory undrained tests.
Clay with gravelor cobbles
As for gravel, etc, above except that Class 4 samples ofstiff/hard clay can be recovered by rotary core drilling.
Weak rock
Boringusing clay cutter, shell or chisel maybe possible. Class S from shell. Class 4 from clay cutter. Class 3 or 4 from sampler.
Rotarycore drilling of harderrocks. Cores likelyto be deficient of weakbands andweak materials
84
Field tests bestas above. Pressuremeter and SPTgive rough indication ofstrength and compressibility,
Field tests best, and estimates based on visual inspection in shafts callbrated from field tests and foundations elsewhere.
Percussion drilling monitored Approximate estimates of strength possible from SF!' indicate fissure, void and soft calibrated empirically. layer geometry.
by engineering geologistcan
APPENDIX B COMMENTS ON PILE INSTALLATION The following comments ofdesign engineerswere recorded during the Survey. Comments on driven piles Comments/advantages Driven piles are used on more bridges than boredpiles. Segmental driven precast concrete piles have increasinglybeenfound economic. Every pile is 'tested' or at least compared with load-tested piles by the driving process. Steel, precastconcrete,and cased cast-inplace piles have assured cross-section.
Type ofhammerneedstobe selected to suit ground conditions. Forexample, at the limits ofrange of soil types: a slow hammer works well in clays but not sands and gravels,while avibrating hammer workswellin saturated sands and gravelsand not in clays.
Warnings/disadvantages Driving to a set is usually not reliable. The dynamic drivingformulae are found to be unreliable and empirically derived sets are dependenton ground conditions. In some ground, such as chalk, a poor set can be adequateifsoil strength increaseswith time. In contrast in silty soils high pore pressures develop at the base and resistdriving, so giving the appearance of hard ground (on redriving next day drivingmaybe easier), but under long term loading settlementcan be large. The pile's capacity and appropriateset/depth canonly beverified by load test.
In built-upregions noise restrictions may limit choice
of type.
Some types are difficult to extend/cut-back driven differ from those anticipated.
if depths
Closely spaced piles have to be monitored to ensure drivingof one does not lift the adjacentone. Redriving may be necessary. Also shearin soil at side during driving maylater relax and in doing so lift piles.
If pilesmeet refusal at differentdepths then a check
should be made that base bearing loads of the higher ones do not get transferred to deeperones by 450 spread ofload.
Fixedlength precast concrete piles are only appropriate in known uniform conditions since otherwise problems are encountered in extending/shortening. Precast concrete piles and shell piles have been known to break during driving into dense gravel and Keuper Marl.
Itis often worthusinghigh tensile steel, even ifnot needed forload capacity, since driving is easier with less tearing.
SteelH pilescan usually be relied on to drive througha dense layerif required. SteelH and cased piles have advantage oflow mobilisation cost and can be placed by main contractor.(It maybe worth down-grading nominal capacityto compensate for lower standard ofworkmanship). They can readily
be lengthened by welding.
Some very expensive delays have resulted following use of steel H piles to bear on weakrocks and shales, when pileshavebeendesigned on basis ofsteel section capacity rather than rock's capacity. The piles did eithernot meet refusal or reducedcapacity of neighbours as a result ofsplitting rock. Main contractorsdo not generally have the experience of specialistsin avoiding installation problems. Bottom driven cased piles with their steel casings are not suitable if they are too long since casingscan break in tension during driving (lack of movement of top of casing gives appearance of refusal) and/or hard upper strata can grip casing and prevent penetrationof soft stratum at depth.
Vibrationcan cause damage to neighbouring buildings.
85
APPENDIXB (contd) COMMENTS ON PILEINSTALLATION Comments on boredpiles Comments/advantages Bored piles are sometimes chosen for the questionable reason that the designerfeels confident that ifthe pile is installed in accordance with drawings he can be sure ofits performance, and not liable to as many claims as driven piles. However bored piles are not appropriate for all groundconditions and difficulties in installation still lead to claims and expensive delays. Surface of site and adjacent piles are not raised by displacement.
Warnings/disadvantages
Not usually suitable for bearing in granularsoil. Concrete strength is not so important:the needto prevent voids forming is paramount. Borings must be cased or constructedunder bentoniteif ground is unstable. Casingmust be clean to ease extraction. It is usually impossible to drive casing ahead of bore in granular soils and overdig occurs. Voids, necksor 'growers',are likely to occur on lifting of casing, but can usually be avoided by careful contractor. To prevent voids developingnear top an excess head of concrete is maintained, but sometimes this is not practical. Ifthere is a risk ofnecks forming in piles in soft ground the designeris advised to pay for perrnanent casings.Unless the level of the reinforcement has changed it is impossible to tell afterwards from above ground (even the 3-tubemethodofintegrity testingcannot detect slumping ofconcrete outside reinforcement). Drivingcasing is a slow expensiveprocess: the contractor will always try to avoid it and is likelyto try to qualify in thetender that depth ofcasing ofso much hasbeen assumed on basis ofa site investigation. Waterfilled bores in weak rocks and stony ground are likely to have heavy sediment at bottom so that only shaft shouldbe relied on for loadtransfer.Sealingthe casing into rock is likelyto be difficult and silt mayflow in. Some piles have beenconstructedwith a precast base driven out ofthe casing into rock to expelsilt from the bottom. When placing concrete under water the water must have reached equilibrium and a tremie must be used. Segregationand cavities have beenreportedeven whenwater is not under artesian pressure.
Shaft adhesion is affectedby carelessnessofgang. Water added to ease boring reduces adhesion, particularly in short term (ie for test). Reinforcement must notcongestflow ofconcrete.Minimum bar spacing 150 mm. Make reinforcement same length in all piles and not full length: changing lengths to suit borescauses delays and softening of ground; lapping causes congestion. Bobs should be avoided during installation or madetangential: inwards getsin way of concrete,outwards in way of casing. 75 mm cover is usually needed, not 40 mm. Concrete must have high workability. At cutting backcare shouldbe taken not to fracturepile, particularly whenconcrete is green.
86
APPENDIX B (contd) COMMENTS ON PILEINSTALLATION Comments on boredpiles (contd) Comments/advantages
Warnings/disadvantages
Small diameter:
In general groups ofsmall diameter piles are cheaper thanlarge diameter for the same load, particularly for frictionpiles because oftheir greaterperimeter/area ratio. But the pile cap for alarge numberof small piles can be more expensive.
Serious disadvantages that inspection of shaft and base not possible.
Not so likelyto need casing as large diameter piles because the soil is better able to arch around a smallhole. Can be installed with limited headroom.
Actual sizes can be significantly differentfrom nominal sizes specified. If design specifies500 mm a tripod contractor could offer a 19 in, which below level of casing is 17 in. An auger-rigcould only do 18 in which is called 450 mm even thoughit is 20 in around casing. Designersneedto check that what they want is practical and specify it precisely. 500 mm diameteris not a standardfor augered piles in increments of 150 mm from
which have diameters 450 mm.
Proper supervision of large number of tripod rigs can
be impractical. Larger diameter:
Soilcan be inspected and probedin-situ, exceptwhenconstructedwith bentonite. The larger bores are more economic.
A 750 mm diameter hole can be bored as quickly as 450 mm holes. Speed of construction reduces groundrelaxation and so enables greaterbearing pressures and shaft adhesions to be utilised. Capacities of up to and exceeding 2 000 tons can be achieved. Underreams can be madeon pileswith 750 mm shafts and larger ( but not smaller).
Uneconomic for small structures because ofhigh mobilisation costs. Heavy rigs require goodaccess and large clearance. Difficult to install in poor ground because casings are installed full length and not segmentally, and have to be pulled out and replaced by longer lengths every 3m. However use of bentonite is becoming more common. Underreamns must be cleaned and inspected. Defective
equipmentor incompetent driver can result in 7 or 8 skips ofdebris insteadof 2. Pile caps must be thicker to carry high concentrated loads.
If pilot holes are bored prior to main boring to establish base and casing levels these should be beside and not throughmainbore or underream area,so that water encountered does not flood or softenmainboring.
87
APPENDIXC ARRANGEMENTOF PILES IN GROUP The following procedure is followed by some designersto determine the optimumarrangement ofpilesin a group to resistinclined loads.
It is assumed that therows of piles are in two sets with differentrakes (including vertical). At least one set has two or more rows.Consequently theremust be at least three rows in all. 1
Dimensionthe pile cap as ifit were a spread footing.The maximum concentrationofload under a pile cap is often about 400 kN/m2.
2 Draw thelinesofactionofall theloadcaseresultantson a cross-sectionof thepile cap.The point at which they are most nearlycoincident is the optimumpoint for the elastic centre. The elastic centre ofa group with two sets of piles at differentrakesis at the intersection ofthe raking centroid lines ofthe two sets. (Resultantloads through the elastic centre induce only thrusts in the piles in each raking set: the relative magnitude ofthe load in each set depends on the inclination ofthe resultant. The moment on the group due to eccentricity ofa resultantfrom the elastic centreis opposed by variations in axial loadsacross therows ofpiles at each rake: at least one set ofpiles must have two or more rows to provide a lever arm.) 3 By inspection a line is drawn representing the average of the loadcase resultantsand passing through the preferred elastic centre position. Some construction load casesmay not fit the general pattern and these should be ignoredat this stage, though checked in later analysis.
4 A pile group is nowsoughtwhich hasthe above elastic centreand which would have equalloadsin all piles when subjected to the average loadcase resultant. Lines are drawnthroughthe preferred elastic centre at the two
rakesto represent possible centroid lines. The relative numbers ofpiles at each rake are proportional to the relative lengths ofcomponentvectors along the centroid linesthat are required to provide a resultantvector along the averageload case resultantline. The total numberofpiles is found by dividing the total loadby the maximum permissibleload per pile. The piles at each rakeare then arranged about their centroid with an adequatespacing for transferring the loadsto the ground. If it is not possible to designa pile layout so that theelastic centreis as low as the intersection ofload case resultants,thenpiles are inclined as steeply as possible, subject to the maximum practical rake,in orderto bringtheelastic centreas low C.nraLd. of pi.lcs as possible on theaverageload case resultantline.
I
of x
4 Iftheextreme loadcase resultants differ markedly from theaveragethenumberofpilesmight need slight modification. Ifthe load case giving maximum loadscauses a variation in the loadsin parallel
rows ofpiles, then the variation can be effectively removed by moving the whole group laterally by an amountapproximately equalto themomentin theparallel rows (due to theinequality ofpile loads)divided by the resultantapplied load. Fromcomparisons ofmaximum pile loadsfor two positions ofthe group the designercan interpolatethe optimumposition.
5 For abutments it is often convenient to design all thepiles exceptthebackrow with the same slope as the averageload case resultant, and then design the backrow to provide equilibrium for the load case with minimum horizontal component.
88
APPENDIX D DEPARTMENTOF TRANSPORTMEMORANDARELATED TO THIS REPORT Memoranda currentin March 1978: Title
Date
Division
Site investigation procedure (Amendment 7/11/75)
19/10/70
E mt
30/6/72
E mt
Number
Hi1/70
(Amendment 11/8/77) H5/72
Notes on the treatmentofold filledmine shafts and disused shallow coal workings
H3/76
Model contractdocument for site investigation (Amendment 16/8/76)
Feb 1976
E mt
1M4
Pulverisedfuel ash backfiuing to structures
19/12/69
BES
BE1/73
Reinforced concrete for highway structures (1st Revision 9/8/73)
30/1/73
BES
BE7/74
Lnteral loading on piledfoundations
17/12/74
BES
BE5/75
Rules for the design and use offreyssinet concrete hinges in highway structures
7/3/75
BES
BE7/76
Suite ofbridge design and analysisprograms
20/12/76
HECB
Program HECB/B/8(RETWAL) BE1/77
Standardhighway loadings
14/2/77
BES
BE4/77
The inspection ofhighway structures
1/4/77
BET
BE5/77
Suite ofbridge design and analysis programs Program HECB/B/7(PGROUP)
1/4/77
HECB
BE3/78
Reinforced earth retaining walls and bridge abutments for embankments.
27/4/78
BES
89
AUTHOR INDEX Adams, H C, viii,51
Reynolds,C E, viii
BS 5400, viii,41 Bell, F G, 9 Berry, DW, 64 Blyth, F G H, viii Bowles,J E,viii Bridle,R J, 9 BritishRailways Board, 76 Broms, B, 46, 56, 60,61 Burland, J B, 10,46,60
Scott,C R,viii
Simons, N E,viii Sims, F A, 9,49
A
Skempton, W, 60 Steedman, JC, viii Sutherland, B, 43,44
H
Terzaghi, K, viii,42,43,47,53
CP2,viii, 41,43,44,47,48,49,50,53,63, 64,72 CP2001, viii,5, 12,15 CP2004, viii, 14,41, 44
Thorburn,J Q, 58 Thorburn,5, 58,60 Thornburn, H, viii,43 Tomlinson, J, viii, 13, 43, 56, 60, 61 Tschebotarioff, F, viii
T
M
Capper,PL, viii
Cassie,WF,viii Chettoe, C 5, viii,51 Cooke, RW, 60
Weltman,
EAU 1970,54 Faber, J, viii
de Freitas,M H, viii Geddes,J D, viii
Hanson,W E, viii,43 Huntington,WC, viii,51,53,61 Institution of Civil Engineers,
9, 37, 60
Jones, C J F F,49 Lambe,T W, viii, 43, 45, 53
A L, viii
J A, 58 London Transport Little,
Executive,
MacVicar, R 5, 60 Marsland,A, 15
Mead,F, viii deMelo,VFB,46 Menzies,B K, viii Military Engineering,viii
NAVFAC,viii CoalBoard, 9 R,8
National Neill,C
Peck,RB,viii,42, 43,47,53
90
A J, 58,65
T,58, 60 Whitman,R V,vuil,43,45,53 Whitaker,
Department ofHighways,Ontario, 8 Department of Transport, viii, 64, 72, 76, 89 Dixon,H H, 64
Little,
G
76
INDEX Abnormal live loads, 43 Abutments (see also Substructures) earthpressures,47—52
open,21,22 stability, 18, 39, 53, 72, 77 strutted,portal and box, 23, 24, 25, wail, 18, 19,20 Access
Compaction offill(see Fill) ofsand (see Sand) Compensated foundation (see Buoyant foundation) Compressibleground,26,40, 53 Consequential costs, 35 Consolidation,42 Construction, 3,4,6—8, 17,35,67 joints,70 loads, 41
construction, 3,4,5,6, 7, 8, 17 maintenance, 21, 76 Active earthpressure,47—51 onpiles,61 overhand,6, 7,20 Advancedcontract, 7, 37 procedures, 11 pile tests, 15,65 programme, 3, 7, 17, 18, 35, 37, 67 Aerial photographs,5, 9 sequence,4, 27, 30,40,45, 71 Allowable pressure (see Bearing pressure) simplicity(see Simplicity) Alternative design,35,59 speed, 3, 17 Construction problems,5, 17 Anchorage,19 Anchors,20,21, 25, 65, 77 flood, 8 Arch, 30, 39 pile installation, 14, 15, 35, 62, 65 Articulation, 9, 17, 34 portalframe, 24 At-rest earthpressure, 47—52 slip failure, 53 struttedabutment, 23 Augeredshaft,13 Continguousboredpiles, 20 Continuous deck, 36 Back fill(see Fill) Contract Bankseat,7,21, 76 documents, 11,30,59, 75 slipfailure, 53 size, 33,65 Base Corbel, 71 abutment, 18,69 Corrosion, 20 Corrosivesoils, 14, 72 wingwail,27, 28 Base friction, 53,68 Coulomb,47, 73 Batter, 69 Conterfort, 18 Bearingcapacity, 14,15,42 Cover toreinforcement, 67, 68 Bearingpressure, 17, 19, 26, 27,33,37,43 Cracking,18, 29 Crack controljoints, 70 Bearings, 31 Bearingshelf, 71,76 Creep and shrinkage,41 Bitumen slipcoat, 56 Culvert (see Box stmcture) Curtain wall,71 Boiling,13 Boreholes,5, 10, 11,59,77,83 Cyclic loading, 23 Boulders, 14, 34, 83 Box structure, 24, 25, 28,52 Deadload,41,42,43 Brakingloads, 41,49 Deck, 17 (see Structure) Brickwork,7 Building ResearchEstablishment,ii, vii, 79 replacement, 7, 23 Buoyantfoundation, 26, 34 Deepexcavation, 17 Buried skeleton abutment (see Skeleton Delays, 13,35,58,65,71 Designprocess, 1, 17 abutment) Buried structure, 52 Deckstudy,S Buriedwail, 22 Details,67—74 Dewatering,13, 33 Diaphragmwail, 20, 77 Differentialsettiement, 19, 24,25, 27,41, Caissons, 8 Cantileverand drop-in span deck, 30 45,46 CantileverT-wail,18,48, 50 acceptable, 17, 36,45 Cantileverwall,20,40 calculation,45 Cantileverwingwalls, 27, 28,29 minimisation,46 earthpressure, 48, 50 Dip, 34,44 Caserecords, 45, 75 Displacementsof Casing retainingwail, 48,51 borehole, 14 pile group,55,61 pile, 8, 15,62 Drainage,36, 68 Cellularabutment, 19 abutment, 72 Chalk,9,34,42,46, 73 bearingshelf, 71 Circular strain or slip, 40 maintenance,76 Clays,33,34,42,84 membrane, 72 Clearance, 7 piers, 71 Coal mining,9 Drained behaviour,42 Cobbles,83 Dutch cone, 83, 84 Cofferdam,8, 13, 20, 26 Dynamic compaction, 37 Cohesionlesssoil, 43 (see also Sand)
Earth pressure,47—52 abutment, 47—52 active, 47—50 at-rest, 23, 47—51 cantileverwing wall, 27, 50 coefficient, 47—50 compaction, 48,50 loading, 41, 42 movement relationship,48 portalframe, 24,50 struttedabutment, 23, 50 subsidence,9 Earthquake, 41, 74 Eccentric and inclinedloads, 44 Effective stress,42, 50 Elastic centre, 64, 88 Elastic half space, 64 Electric cables,6, 7 Embankment compaction (see Fill) flooding,9,14 movement, 39,46 settiement, 17, 21, 39, 40, 56 Engineeringgeologist, 5 Equivalentfluidpressure, 47 Excavation
belowwatertable, 19, 33,34
beside railway, 7 dewatering, 13 shape, 3,68 site investigationinformation, 11 Expansionjoint,51,56,76 Factor of safety,11,33,43,44,46,53,60, 57,63
Factory Inspectorate, 6 Falsework(see Temporary support) Feature, 70 Fill backfill, 73 compaction, 18, 21, 22, 27, 28, 34, 37,
48, 50, 71, 73
corrosive,14 cribwail, 29 earthpressure,47—52 selectedgranular, 47, 73 specification,73, 74 suitable, 47 under foundations, 18, 33 Fissures,14,42, 84 Flexible corrugated metal structures, 25 Floods, 8,14,41,72 Footings, 33, 34, 35, 39—46
pier, 31 railway, 7
settlement,39 subsidence,9
Form,3, 69 Formwork, 3,27, 31,69 Foundation choice, 33—37 Geologicalmaps, 5 Global behaviour, 39,40,55, 75 Ground anchors (see Anchors) Ground conditions, unforeseen, 4, 17, 59 Ground improvement, 37 Ground level, 3,6 Ground struts,25, 30 Ground ties, 25
91
Groundwater, 13, 14
depth,33,59,72 effects on behaviour, 43,52 effects on construction, 33, 35 effects on grouting, 9 lowering,6 effects on piles, 14, 15, 86 site investigation,10, 11, 13, 14 table 20 Grouting, 9, 14, 37, 77 Gravel,33, 34,42,83 Gravity wails (see Massconcrete) Gypsum, 14 Haunches,69 Headroom, 6 Heave, 14, 39, 40,45, 85 Hinge, 23, 24, 89 Horizontal movement,46, 51, 76 Hydraulicfill, 13 Hydraulicloads, 41,42 HydraulicsResearchStation,8 Hydraulicsurvey, 8
Icepacks, 41 Impact loads, 41,42 Inclined loads, 30, 88 In-situ testing, 11 Instability, 34 Integrity testing, 65 Jacking, 9, 76 Joints construction, 70 crack control,70 expansion, 51,56,70 maintenance, 76 movement, 9, 27, 70 seals, 70 Land drains, 68 Lateral loads onpiles(see Piles) Lateral movements,26,39,51,59 Lateral pressures (see Earth pressure) Liaison between engineers, 11, 12, 76 Limestone, 9 Limits ofbehaviour, 10,39,49,55 Live loads, 41,43 Load combinations,41,49
Loading,41,45,89 Load test, 43 (see also Pile test) Long-term loading,42,45 Maintenance,4, 21,74,76,89 Maps, 5
Mans, 42 Masonrybridges,77 Massconcrete abutment, 18, 50, 53 fill, 13, 19,33,34,51 Measurement,59 Meyerhof,43, 44,53 Miningsubsidence,9, 34, 76, 89 Modifications,4, 35 Monitoring, 9, 75 Movement,19, 26,45,46 global, 39
joints, 70
Muckshifting,6 Mudstone,42,46
92
National Coal Board, 5,9 Noise, 6, 85 Observations,45, 46, 75, 76 Overall stability, 26,57 Overburden,42,52 Overconsolidatedclay, 42,48 Overheadcables, 6 Overstress, 43 Overturningstability, 53 PGROUP,64, 89 Passive pressure, 51,53,68 Passive resistance, 25 Parapet, 27,28 Percussiondrilling,84 Permeabilitytest, 46, 83, 84 Pier, 31,76 Piezometer,10, 13 Pile foundations, 33—35, 55—65 alternativedesign,35, 59 avoidance,19 axial loading, 60
trial, 15,65 types, 15,58,60 vertical, 56,57,62 wingwall, 28 Pile group analysis,63,64 design,55, 57, 62, 63, 88 displacements,55 factorofsafety, 57,63 group factor, 60 overturning stability, 57 portalaction,57 settlement ratio,60 Plane framemethod,64 Plant, 3, 4, 6, 11,62 Plate bearingtests, 15,83,84 Polystyrene, 19
Poorground,6, 7,26
Porewater,42 Portal frame, 23, 24 earthpressures,24, 48, 50 wing walls, 29 Poursize, 69,70 Precast concrete, 7 Pressuremeter,84 bored,6, 7, 8, 20, 33, 35, 60, 61, 62, Propping (see Temporary support) Pulverisedfuelash,73, 89 86,87 caps, 6, 7, 8, 35, 56, 63, 64, 87, 88 Pumps, 13 cased, 85 Pumping trials, 13 casing, 35, 59, 62, 86, 87 coating, 14, 56 construction problems, 15, 35, 62, 65, Railway 85,86,87 auhority, 7 bridge over, 7 continguous,20 corrosion, 14 possessions,7, 25 displacement,7 Raking pier,31 Raking piles (see Pile foundations) driven, 7,33,34,62,85 Rankine theory, 47 drivingformulae, 65, 85 Reinforced concrete abutment, 18, 89 drivinghammer, 85 Reinforced earth, 19, 29 earthpressure,48, 50 end bearing,60 Reinforcement,3, 18,59,62,67,69,71,86 factorof safety, 57,60 Reliability, 17,33,35 Repairs,77 ground waterproblems, 13, 86 Residual angle,42 group (see Pilegroup) Restricted space,6 inspection, 87 Retainingwall (see Abutment) installation, 58, 60, 65, 85, 86, 87 River integrity testing, 65 authority, 8 large bored,6,60,61,62,87 lateral loading, 26, 34, 40, 57, 61, 63, bed,8 89 bridge over, 8, 34 Rivet group method,64 lateralmovement, 26,61 Rock, 14, 34, 39, 53, 73,84 necking, 86 precast concrete, 34, 85 dipping, 34 shoes, 34 raking,6, 7, 21, 23, 26, 28,50, 55—57, weak, 14, 84 62—65, 88 reinforcement,59,62, 86 Rotary core drilling,84 Run-on slab, 74 rigs,6,62, 86, 87 rivers,8, 34 rockshoes, 34 S.P.T (see Standard penetration test) secant, 20 Salt mining,9 segmentaldriven, 85 Sand, 33, 34, 42, 83 settlement, 56, 60 shaftfriction, 60, 86 compaction, 34 Sanddrains, 37 site investigation,12,59 sizes 87, Scour, 8, 76, 77 soft ground, 26,53,57 Seals, 70 Seasonal variations, 13 specification,59 Secant piles, 20 steel bearing, 14,34,85 Semi-mass abutment, 18 subcontract, 35,59 reaction, 57,61 Services,5,6, 25, 28 subgrade Settlement, 43, 45, 46 (see Differential subsidence,9 settlement) tension, 60, 65 abutment, 40,45, 71 test, 15, 58, 61,62,65 'test', 65 acceptable, 17, 36,43 backfill,21, 73, 74 tolerances,31,62
calculation,45,75 Struttedabutment, 23 embankment, 17, 36,56,74 earthpressure,48, 50 fill, 21 Subgradereaction, 57, 61, 63 footings, 33, 34, 36 Subsidence,9, 34, 36,76 ground water, 14 Substructure piers,40, 45 catalogue,17—31 piles, 33, 36, 55, 56 form, 69 ratio,60 on highly compressible ground, 26 sequence,40 open abutment, 21,22 Shale, 42, 73 piers, 31 shear-key,68, 70 railway, 7 Sheetpiles, 13, 20, 26, 29 strutted, portal and box, 7, 23, 24, 25 Short-term loading,42,45 wall abutment, 18, 19,20 Shrinkage,70 wingwall, 27, 28, 29 Shuttering (see Fonnwork) Sulphates, 14 Silt, 34, 83 Supervision,11,58 Simplicity ofconstruction, 1,3,8,9, 17,67 Surcharge,26, 34, 37 Simply supported deck, 36 Survey, vii Singlepour,3, 18, 69 Swallowholes, 9, 34
Site,5—15
groundwater (see Groundwater) influence on construction, 6, 7, 8 reconnaissance,5 soils investigation(see Soilsinvestigation) subsidence(see Subsidence) Skeleton abutment, 21, 22, 50, 51, 53, 73 Skew, 3, 19, 23, 27, 28 Slickenside,42 Slidingresistance,18,53 Slipcoat,56 Slip failure, 19,42,53,61 Sloping abutment, 19
3,51,67,68
Softening, Soilsinvestigation,
5, 10—15
corrosive soil, 14 data assessment,45 informationquality, 83,84
planning,S,10, 11,89 principles,10
report, 11, 12,59
samples(see Soil samples) settlement, 45 supervision,11 Soil profile,10
10, 83, 84
10
strength, 42, 83, 84 Soil samples classification,83, 84 description, 12
(see CantileverT-wall)
Ultimate limitstate,44 Underpinning,77 Undrained behaviour,42 Vane test, 84 Vibration,41,42,49,85 Vibro compaction, 37 Vibro replacement, 37 Visualinspection, 11,84 Void collapse,9, 26, 39,62, 86 V-supports,30
bar, 70
Soil — structure interaction, 47, (see global behaviour) Space framemethod,64 Spanarrangement, 17 Specification,59 Spill-through (see Skeleton abutment) cellularabutment, 19 Spread footings (see Footings) Stability of retaining walls, 3, 53, 72, 77 Standard penetration test, 10, 43, 83, 84 Standpipe (see Piezometer) Static method,64 Statutoryundertaker, 5 Steel bearingpiles(see Piles) Steelfixing (see Reinforcement) Stone coLumns, 37 Structure over railway, 7 struts,
T-wall
authority,5,8
quality, 12, 83, 84
83, 84
Tremie,62, 86 Trial pits,5,8, 10,11,13,83,84
Water (see Groundwater)
inspection, 11
recovery,
Traffic
loads, 41,42,49
Snow,41
stratum,
Temperature effects, 41,42, 70 Temporary support, 3,4, 17, 19, 21, 23, 24, 30, 31, 68, 69 Temporary works, 4,7,9,11, 19 Testload (see Load test) Thrustboring,24 Ties, 9, 20 Tolerances,62,68,69,71 interference, 6
Slope, 21,44
properties,
TAXIS method, 64
pressure, 50, 72 Waterproofing,72 Weather,6, 17,67 Weathering,51,68 Weepholes,72 Wellpointing, 13 Windloads, 41 Wingwails, 21, 27, 28, 29 Workingspace, 7
7 Printed inEngland forHer Maiesty'sStationeryOffice by Harrison & Sons (London)Ltd. 27943 Dd596l5t K24 4/79
93
HER MAJESTY'S STATIONERY OFFICE Government Bookshops
49 High Holborn, London WC1V 6HB 13a Castle Street, Edinburgh EH2 3AR 41 The Hayes, Cardiff CF1 1JW Brazennose Street, Manchester M60 8AS Southey House, Wine Street, Bristol BS1 2BQ 258 BroadStreet, BirminghamB! 2HE 80 Chichester Street, Belfast BT1 4JY Government publicationsare also available
through booksellers
ISBN
0
11 670761 5