H Pile Design Guide

P335: H-Pile Design Guide Discuss me ...

Created on 04 July 2011 This material is copyright - all rights reserved. Use of this document is subject to the terms and conditions of the Steelbiz Licence Agreement

SCI PUBLICATION P335

H-Pile Design Guide

A R BIDDLE BSc, CEng, MICE

Published by: The Steel Construction Institute Silwood Park Ascot Berkshire SL5 7QN Tel: 01344 623345 Fax: 01344 622944

P335: H-Pile Design Guide


Discuss me ...

 2005 The Steel Construction Institute Apart from any fair dealing for the purposes of research or private study or criticism or review, as permitted under the Copyright Designs and Patents Act, 1988, this publication may not be reproduced, stored or transmitted, in any form or by any means, without the prior permission in writing of the publishers, or in the case of reprographic reproduction only in accordance with the terms of the licences issued by the UK Copyright Licensing Agency, or in accordance with the terms of licences issued by the appropriate Reproduction Rights Organisation outside the UK. Enquiries concerning reproduction outside the terms stated here should be sent to the publishers, The Steel Construction Institute, at the address given on the title page. Although care has been taken to ensure, to the best of our knowledge, that all data and information contained herein are accurate to the extent that they relate to either matters of fact or accepted practice or matters of opinion at the time of publication, The Steel Construction Institute, the authors and the reviewers assume no responsibility for any errors in or misinterpretations of such data and/or information or any loss or damage arising from or related to their use. Publications supplied to the Members of the Institute at a discount are not for resale by them. Publication Number: SCI P335 ISBN 1 84942 164 4 British Library Cataloguing-in-Publication Data. A catalogue record for this book is available from the British Library.

P:\bcv\bcv500\P335\P335v01d09.doc

ii

Printed 07/09/05


FOREWORD Current practice is to place the responsibility for pile design on the designer (whereas before it was often on the contractor), and there is therefore need for him to become more familiar with the behaviour and advantages of steel piles. It is hoped that this design guide will provide the necessary confidence for practising engineers to use steel H-piling more extensively and become more innovative in the use of steel piling in structural and building foundations. It also covers the use of UC sections for plunge columns. The guide is laid out in Sections which follow the steps involved in a well established design procedure. A new Section 9, Technical and Cost Benefits is based on case studies to demonstrate the practical benefits of using H-piles on various projects.


The SCI database of axial load tests on steel H-piles that was established in 1997 for Steel Bearing Piles Guide, has been used again to validate load capacity prediction methods together with more recent test data and case studies. Partnerships were formed with SCI members who have provided soils data and load test results from steel H-pile tests on their construction sites. Major partners were: Pell Frischmann Group; Volker Stevin Ltd; Stent Foundations Ltd; Testing and Analysis Ltd. Their assistance and time is gratefully acknowledged. Particular thanks is also expressed to the members of the Steel Piling Group who reviewed and contributed to the draft documents or contributed information and photographs for the case studies: Andrew Bond Robin Dawson Marwan Ghannam Simon Griffiths Mike Kightley Steven Lee Norman Mure Ron Mure John Powell Colin Souch David Thompson David Twine John Vincett Mike Webb Cliff Wren

Geocentrix Ltd Dawson Construction Plant Ltd Corus Construction & Industrial Pell Frischmann Group Testing and Analysis Ltd Volker Stevin Ltd Stent Foundations Ltd Stent Foundations Ltd BRE Pell Frischmann Group Dew Group Piling Ltd Ove Arup Geotechnics Tony Gee & Partners Corus Construction & Industrial Stent Foundations Ltd

Grateful thanks is owed to Corus Construction & Industrial who funded the preparation of this publication.


iii

Printed 07/09/05



Discuss me ...


iv

Printed 07/09/05


Contents


Page No. FOREWORD

iii

SUMMARY

vii

1

INTRODUCTION 1.1 Piled foundation choice 1.2 Why choose steel piling? 1.3 Scope of this publication

2

DESIGN BASIS 2.1 General 2.2 Design standards 2.3 Limit state design rules 2.4 Bearing pile structural design 2.5 Design methodology

7 7 7 8 9 10

3

GEOTECHNICAL DESIGN 3.1 Terminology 3.2 Design premise 3.3 Limit State Design 3.4 Geotechnical design methods 3.5 Soil resistance on driven steel piles 3.6 Load / settlement behaviour – friction piles 3.7 Pile-soil load transfer – friction piles 3.8 Load / settlement behaviour – end-bearing piles 3.9 Pile-soil load transfer – end bearing piles 3.10 Site investigation

13 13 13 14 19 21 21 25 26 26 27

4

SELECTION OF SECTION 4.1 Steel piles in bearing only 4.2 Design method examples 4.3 Selection of steel section 4.4 H-Piles 4.5 Plunge Columns

29 29 29 30 30 32

5

AXIAL LOAD RESISTANCE 5.1 Interpretation of soil parameters 5.2 Predictive methods – general 5.3 Pile axial movement models 5.4 Axial resistance in non-cohesive, granular soils 5.5 Axial resistance in cohesive soils 5.6 Axial resistance in rock 5.7 Negative shaft friction 5.8 Measures to increase steel pile axial capacity

36 36 36 38 41 46 48 52 52

6

LATERAL LOAD RESISTANCE 6.1 Introduction 6.2 Methods of analysis 6.3 Assessment of soil properties 6.4 Combined lateral and vertical loading

54 54 55 57 58


v

1 1 2 6

Printed 12/09/05



Discuss me ...

7

PILE GROUP EFFECTS 7.1 Conceptual design axial load resistance 7.2 Methods of lateral load resistance analysis 7.3 Practical pile group design

60 60 61 62

8

THE INSTALLATION AND TESTING OF STEEL BEARING PILES 8.1 Pile driving installation methods 8.2 Offshore experience of pile driving analysis 8.3 Driving formulae and dynamic driving resistance 8.4 Pile load testing 8.5 Steel pile installation tolerances 8.6 Environmental factors with driven piles

65 66 68 69 75 80 81

9

TECHNICAL AND COST BENEFITS 9.1 Steel pile economics 9.2 Soil conditions 9.3 Design configuration 9.4 Case Studies 9.5 Cost comparisons

84 84 84 85 86 92

10

STEEL PILES/STRUCTURE CONNECTIONS

94

11

CORROSION AND PROTECTION OF STEEL PILES 11.1 The need for corrosion protection 11.2 Standard corrosion allowances 11.3 Corrosion in soil 11.4 Corrosion in fills and ‘brownfield’ sites 11.5 Atmospheric corrosion 11.6 Corrosion below water 11.7 Methods of increasing effective life

REFERENCES APPENDIX A


97 97 98 99 100 101 101 101 103

CONTACTS

113

vi

Printed 12/09/05


SUMMARY This publication gives guidance on the selection, design and installation of steel H-piles and UC section plunge columns for foundations to all types of structure. Current practice and experience in this field are presented, discussed and recommendations given.


The characteristics and advantages of steel bearing piles in construction are described in order to assist in the primary process of selection of the correct pile type for any given site and soil conditions. Load transfer mechanisms are described and limit state design methods applied in line with the new Eurocodes. The sections on design include axial and lateral load resistance prediction methods, combined loading effects on retaining walls and pile group analysis. Up-to-date pile driving analysis is presented as a basis for planning efficient installation and as an aid to design. Practical aspects of test loading, installation tolerances and connection details are covered. It is noted that excessive conservatism has been found in current practice and this results in unnecessary overdesign. Currently used specifications for load testing piles only up to 1.5 × working load, are insufficient to reach the ultimate pile resistance and the whole object of the new limit state design (LSD) procedures has been denied. This problem has been compounded by making unrealistically low design assumptions on the soil parameters in pile resistance prediction methods. This publication adopts LSD using the new Eurocodes and suggests more reliance be placed on static and dynamic load test methods to establish ultimate capacity to permit more economic steel pile design.

Guide de dimensionnement des pieux de type H Résumé Cette publication est consacrée au choix, au dimensionnement et à la mise en place de pieux de fondations en acier, de type H et UC, pour tout type de structure. La pratique actuelle est discutée et des recommandations sont données. Les caractéristiques et avantages des pieux en acier sont décrits afin d'aider au choix d'un système correct de pieux pour tout site et toutes conditions de sols. Les mécanismes de transfert des charges sont décrits et les méthodes de dimensionnement aux états limites, selon les nouveaux Eurocodes sont présentées. Les chapitres consacrés au dimensionnement prennent en compte les charges axiales et latérales ainsi que l'effet des murs en retour et des groupes de pieux. Les méthodes les plus modernes de mise en place sont présentées. Les essais de résistance, les tolérances d'installation et les détails d'assemblage sont également abordés. Un conservatisme excessif est constaté dans la pratique courante et dans les spécifications actuelles conduisant à des hypothèses non réalistes dans les méthodes de calcul. Ceci a conduit à des différences considérables entre les calculs et les essais de pieux métalliques in situ ; avec pour conséquence une grande difficulté, pour les praticiens, d'interpréter les résultats d'essais, et ainsi toute la base des nouvelles procédures de dimensionnement à un état limite était niée. Cette publication adopte la méthode des états limites, qui conduit à un dimensionnement plus économique des pieux en acier.


vii

Printed 12/09/05


Leitfaden für Pfähle mit H-Querschnitt Zusammenfassung Dieser Leitfaden gibt eine Anleitung zu Auswahl, Berechnung und Einbau von Stahlpfählen mit H-Querschnitt und „Tauchpfählen“ mit UC-Querschnitt für die Gründungen aller Tragwerksarten. Die gegenwärtige Praxis und Erfahrung auf diesem Gebiet wird vorgestellt, diskutiert und es werden Empfehlungen gegeben.


Die Eigenschaften und Vorteile von Stahlpfählen werden beschrieben, um die Auswahl des richtigen Pfahltyps für jede Baustelle und jeden Baugrund zu erleichtern. Die Mechanismen der Lastübertragung werden beschrieben und der Grenzzustand der Tragfähigkeit gemäß den neuen Eurocodes wird in Relation zu gemessenen Pfahlkopfverschiebungen interpretiert. Die Abschnitte zur Berechnung beinhalten Methoden zur Vorhersage des Widerstands für axiale und horizontale Lasten, Pfahlwände bei kombinierter Belastung und die Berechnung von Pfahlgruppen. Neueste Berechnungen zum Rammen werden vorgestellt als Basis für einen effizienten Einbau und als Berechnungshilfe. Praktische Aspekte aus Versuchsbelastungen, Einbautoleranzen und Verbindungsdetails werden behandelt. Übertriebener Konservatismus wurde in der gegenwärtigen Praxis vorgefunden, was zu unnötiger Überbemessung führt. Gegenwärtige Regelungen für Pfahlversuche mit bis 1,5fachen Gebrauchslasten sind unzureichend um die Grenztragfähigkeit der Pfähle zu erreichen und das Ziel der neuen Berechnungsmethoden der Grenztragfähigkeit wurde bestritten. Dieses Problem wurde bei der Vorhersage des Pfahlwiderstands verbunden mit unrealistisch geringen Berechnungsannahmen hinsichtlich der Bodenparameter. Diese Publikation beinhaltet die Nachweise für den Grenzzustand der Tragfähigkeit nach den neuen Eurocodes und schlägt vor, den statischen und dynamischen Belastungsversuchen zur Ermittlung der Grenztragfähigkeit mehr Vertrauen entgegenzubringen um eine wirtschaftlichere Berechnung von Stahlpfählen zu erlauben.

Guía para pilotes de acero con secciones H Resumen Esta publicación guía la elección, proyecto e instalación de pilotes de acero y compuestos de hormigón y acero con secciones H y UC para cimientos de cualquier tipo de estructura. Se presentan tanto la práctica como la experiencia actuales con su discusión y pertinentes recomendaciones. Se describen las propiedades y ventajas de los pilotes de acero en la construcción con lo que se facilita el anteproyecto del tipo adecuado de pilotes para cualquier tipo de suelo. Se describen también los mecanismos de transferencia de cargas y los métodos de diseño basados en los estados límites de proyecto interpretados en relación a los movimientos medidos en cabeza de los pilotes en línea con los nuevos Eurocódigos. Los apartados relativos al proyecto incluyen métodos de predicción de la resistencia a cargas longitudinales y transversales, efectos de carga combinada en muros de contención y cálculo de grupos de pilotes. Se presentan cálculos de hinca, actualizados, para una planificación efectiva de la instalación y como ayuda de proyecto. También se tratan aspectos prácticos de los ensayos de carga, tolerancias de instalación y detalles de uniones. Se toma nota de que un conservadurismo excesivo ha sido observado en la práctica habitual y en las normas o recomendaciones utilizadas en el ensayo de pilotes, lo que


viii

Printed 12/09/05


además suele venir combinado con hipótesis de proyecto poco realistas sobre los parámetros del suelo que se utilizan en los métodos de predicción de la resistencia, obteniendo como resultado un sobredimensionamiento innecesario. Por todo ello, los proyectistas eran incapaces de interpretar la resistencia última de los pilotes a partir de sus ensayos de carga y ello hizo que el nuevo método de cálculo en estados límites últimos fuese rechazado radicalmente. Esta publicación adopta el cálculo en estados límites últimos que debería permitir proyectos más económicos de pilotes de acero

Guida all'uso di pile portanti in acciaio con sezione trasversale a H Sommario


Questa pubblicazione fornisce una guida per la scelta, la progettazione e l'istallazione di pile portanti con sezione a H e UC in acciaio per fondazioni di differenti tipi di strutture. In particolare,viene presentato lo stato dell’arte sia a livello di prassi progettuale sia considerando le conoscenze acquisite. Sono illustrate le principali caratteristiche e i vantaggi delle pile portanti in acciaio in modo da fornire un importante aiuto nella scelta della corretta forma strutturale della pila in funzione del luogo e del tipo di terreno. Vengono poi descritti in dettaglio i più significativi meccanismi di trasferimento del carico ed è presentato il metodo progettuale agli stati limite applicato secondo i requisiti della versione aggiornata dell’Eurocodice. La parte dedicata alla progettazione propone i metodi per la determinazione della resistenza in presenza di carichi assiali e trasversali, per la valutazione degli effetti combinati sulle paratie e per l’analisi di gruppi di pile. Un aggiornato metodo per l'analisi delle pile e' presentato come base per un conveniente utilizzo e valido aiuto per la fase progettuale. Sono inoltre affrontati gli aspetti pratici delle prove di carico, tolleranze di istallazione e dettagli dei collegamenti. La corrente prassi progettuale e le raccomandazioni attualmente in uso per l'esecuzione di prove di carico risultano eccessivamente penalizzanti e portano ad inutili sovradimensionamenti. Le raccomandazioni attualmente in vigore, che prevedono prove di carico con azioni applicate pari a 1,5 volte quelle di esercizio, risultano inadeguate per valutare la resistenza delle pile e ciò è in disaccordo con la filosofia progettaule legata al metodo semi-probabilistico agli stati limite (LSD). In aggiunta, si hanno ipotesi poco realistiche per quanto riguarda i parametri base del terreno che condizonano la capacità portante delle pile. Questa guida adotta il metodo progettuale degli stati limite in accordo alla recente versione dell’Eurocodice e si basa su indicazioni più appropriate relative alla sperimentazione, statica o dinamica, per valutare la capacità portante e quindi per avere una progettazione economica di pile portanti in acciaio.


ix

Printed 14/09/05



Discuss me ...


x

Printed 12/09/05


1

INTRODUCTION

1.1

Piled foundation choice

The first decision in considering a foundation design is whether piles are required or not. In some cases there may be alternative solutions, for which the costs may be compared with those of a piled foundation. In other cases, the bearing capacity of the soil at the foundation level may be satisfactory but, owing to high loadings, piles are required to keep settlement within acceptable limits. It is important to be clear about the reasons for using bearing piles before weighing the relative merits of using steel or concrete types of driven pile, because there are some essential differences in behaviour that may favour one or the other pile type for a particular project.


Bearing piles are used mostly for supporting vertical loads and for this purpose the main requirements are to: •

Restrict average settlement to a low value.

•

Minimise differential settlement.

•

Achieve an adequate factor of safety or load factor against foundation failure.

Many technical and cost-benefit factors affect the selection of the most appropriate type of pile for a given structure. Very broadly, these factors can be divided into those related to: •

Site location and operating conditions.

•

Type of soil and ground water level during installation.

•

Type and size of the loads to be supported by the foundation.

•

Type of structure, e.g. land or marine.

•

Effect of the pile type on overall construction programme and cost.

In some circumstances there will be additional technical factors that affect the choice of pile, for instance when overturning moments due to wind forces on a tall building have to be resisted, or when severe scouring of a river bed may expose piles supporting a bridge pier. Where piles have to resist tensile loading or absorb energy in bending, as in marine dolphins for ship impact, and in integral bridge piers for vehicle impact, there are special requirements to be considered which favour the selection of steel piles. In particular, the ductility of steel piles creates an elastic ‘compliance’ with the superstructure to absorb the impact energy by deflection. The cost-benefit factors which may favour the choice of steel piles include: •

Total cost of the foundation, where it is important that the comparison between pile types is related to the total construction cost including installation and not just the cost of the pile material.

•

Total construction time, where use of driven steel piling can result in a shorter construction period and an earlier project completion date.

P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

1

Printed 07/09/05


•

Environmental constraints, where the noise and vibration caused during steel pile driving has now been reduced by developing new installation equipment to be within limits stated in UK legislation.

•

Sustainability issues, where steel bearing piles are easy to extract from the ground at the end of structure life and can be reused or recycled so reducing the whole life cost of the building.

Many of the above factors are interrelated, and all require consideration in arriving at the most suitable pile type for a given situation. Broad guidance only is possible in this publication, as each project requires individual examination. For specific technical advice or product information, the organisations listed in Appendix A of this publication should be contacted. There is no single pile type that is both technically and economically appropriate for every structure, site or set of soil conditions. Owing to the many different types of project and construction situations, there will always be a need for a variety of pile types, so selection is an exercise of judgement.


1.2

Why choose steel piling?

Knowledge about the installation and in-service performance of steel bearing piles has progressed over the last 40 years due to increased usage worldwide, particularly in the USA, Japan and in European countries particularly Norway, Finland, Holland, Belgium and Denmark. Research work for the offshore industry has been carried out and reported in the UK[1][2][3] and the transfer of this knowledge was considered beneficial for UK onshore application. The trend towards increased foundation loads is well catered for by steel bearing piles. H-piles are capable of carrying loads of up to 4,400 kN. Steel piles offer many advantages compared to other types including: •

Reduced foundation construction time and site occupation.

•

Reliable section properties without need for onsite pile integrity checking.

•

Ductility also gives high resistance to lateral loads for marine structures and compliance in integral bridge foundations.

•

Larger wall surface area giving better friction capacity than equivalent diameter concrete pile

•

Higher end bearing resistance in granular soils and rocks mobilised by pile driving as compared to boring.

•

Closer spacing possible and therefore smaller pile caps.

•

Pile load capacity can be confirmed during driving by Dynamic Pile Analysis (DPA) on every pile driven.

•

Low displacement of adjacent soil during driving.

•

No arisings and therefore no spoil disposal offsite

•

Easily extracted at end of working life.

•

Reusable or recyclable following extraction to meet Government objectives in sustainable construction


2

Printed 07/09/05


Steel piles have clear-cut advantages on projects such as on river or estuary crossings where soils are typically granular and waterlogged and unsuitable for satisfactory pile boring, or where soft recent low bearing strength alluvium overlies bedrock. On cohesive soil sites, there is a wide selection of acceptable pile types and other construction aspects will govern. Nowadays, steel piling is an attractive and competitive alternative for permanent foundations owing to the research and development in piling technology and changes in the construction industry supply chain. These can be described under three broad headings, durability, performance and economy. Durability


The subject of corrosion and steel protection has received substantial attention both in the UK and abroad over the last 40 years. There is now adequate knowledge on corrosion rates, coatings selection and specifications to permit the designer to make a reasoned judgement on the provision for corrosion prevention. Such information is readily available from Corus publications[94][111]; general guidance is also repeated in Section 11 of this publication. In addition, the corrosion guidance sections of BS 8002[5], BS 6349[6], Eurocode 3: Part 5 (EN 1993-5)[7] and in document BD 42[8] (part of the Design manual for roads and bridges) have embodied earlier research, and further revisions are in progress. Performance Reliable load capacity and driveability predictions are essential for the confident design and installation of driven piling. These topics have been poorly covered in most foundation and piling design textbooks and this publication therefore provides practical advice for the guidance of designers. It was deemed appropriate to examine piling technology used in the offshore construction sector, where there is a body of research and accepted practice, and to transfer relevant practices to the onshore sector. The offshore design methods are simple in concept and the principles involved can be readily understood. They have been used with success in minimising foundation installation costs and the steel tubular piles have performed well for decades on offshore fixed structures. These methods are presented in Sections 4, 5 and 6, and supporting references are given for further detail on usage and applications. For economic pile design, the methods require knowledgeable judgement of soil parameters and this, in turn, requires high quality soils data. Such data is obtainable using routine site investigation techniques, but care must be exercised in the soil sampling and testing specifications, in order to ensure that data collected on soil properties is relevant to driven steel piles as well as to bored concrete piles. In particular, there must be more emphasis on in situ penetration testing (see the advice given in Section 3 on SPT and CPT soil tests). Economy The differential in cost between concrete and steel construction has decreased steadily over recent years; the costs of site labour and concreting materials have increased, whereas the cost of steel has decreased in real terms (see Corus publications[107]). In addition, with the advent of ‘Design and Build’ contracts for civil engineering work, there is more incentive for innovative design to


3

Printed 07/09/05


permit cheaper overall construction by incorporating the piled foundation into the structure concept rather than leaving it separate. Constructing in steel permits prefabrication of larger, but still easily erectable, high quality structural elements that can save construction time; this is an increasingly attractive project consideration. In foundations and basements, steel bearing piles are compatible and easily connectable to the steel frame of a building thereby permitting savings in overall construction costs. Progress has also been made in more effective connection between reinforced concrete superstructures and steel piling using welded-on shear studs or angles, hoop bar connectors and careful detailing in composite connections in bridge engineering. For steel intensive basement construction, cost savings of up to 40% have been reported by designers. Steel foundation piles are ductile and can deflect to absorb energy in marine applications producing a saving in structural section.


Sustainability and environment The world’s available supply of construction materials is becoming more scarce and consequently more difficult and expensive to source and supply. Western governments have agreed to encourage more recycling of construction material in order to reduce the impact of mining more ore and aggregate, and to reduce the volume of waste construction materials from demolition of old buildings. Steel is the world’s most recycled material and is 100% recyclable. In 2003, 965 million tonnes of steel were produced worldwide and approximately 43% of that was from recycled scrap steel. The use of scrap is also essential to the efficient production of the stronger higher grade steels and it therefore has a commercial value that makes recycling economically viable. The supply chain for scrap is well established (see the SCI publication Environmental assessment of steel piling[110]). When assessing the environmental impact of construction, it is important to consider the practicality and cost of removal of the structure at the end of its useful life and the disposal of the demolished materials. The construction industry, in common with many other industries, is now being encouraged to develop new processes that will allow more materials to be recycled or reused, helping to conserve natural resources and reduce waste. Steel piling benefits from being easy to extract from the ground during demolition of previous structures, or after its temporary use as part of the construction process. Extraction equipment includes vibration hammers working under a pullout force from cranes and special high load jacking frames that can pull out the longer bearing piles. This facility creates an additional environmental benefit from being able to easily restore a previous building site to a ‘greenfield’ state without any remaining contamination below ground. The steel piles can either be reused or recycled. Concrete piles on the other hand are difficult to demolish or extract and the process is therefore time consuming and expensive. On many sites the degree of contamination with concrete piles is so expensive to remove that developers have been deterred from using that ‘brownfield’ site and have used a ‘greenfield’ site instead. On some ‘brownfield’ sites, the new piled foundation P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

4

Printed 07/09/05


has been interwoven through the old concrete piles, creating more contamination and rendering the site much worse for any future redevelopment. The large diameter bored concrete under-reamed piles that have often been used on inner city sites such as in London and Manchester, are particularly difficult to remove.


Figure 1.1 shows steel piles extracted from some jetties in Hong Kong harbour and stacked on the quayside. The concrete piles in Figure 1.2 were also part of the same complex of jetties, which took more time to extract and were difficult to break up, illustrating the problems in removing concrete substructures.

Figure 1.1

Recovered steel H-piles from a site in Hong Kong harbour

Figure 1.2

Concrete piling at the same Hong Kong harbour presented considerable demolition and extraction problems


5

Printed 07/09/05


1.3

Scope of this publication

This publication supplements the information given in existing textbooks with up-to-date guidance on those aspects of steel bearing piles that have not been covered elsewhere. Section 2 presents a design basis for H-piles and Section 3 a treatment of Limit State Design (LSD) that is consistent with the new Eurocodes and uses the same notation as those standards. LSD is described in a way that relates pile design to the real performance that is observed in pile load tests, and thereby permits the designer to understand the small pile head settlement that occurs in generating pile load resistance with steel piles. The pile-soil load transfer mechanism is also explained. Section 4 covers the selection of steel section for the intended purpose. Sections 5, 6, and 7 cover the geotechnical aspects of steel pile design in the context of other design references and textbooks.


Section 8 deals with an up-to-date treatment of pile installation and the testing of steel bearing piles, especially the growing use of dynamic analysis of driving as a substitute for expensive static loading procedures. The environmental assessment of noise and vibration during driving is also explained. Section 9 covers Economic Design to illustrate the technical and cost-benefit factors of steel piles by means of case studies. Section 10 presents some typical connection details for the pile to structure interconnection, and Section 11 covers durability aspects, including a discussion of appropriate corrosion allowances.


6

Printed 07/09/05


2

DESIGN BASIS

2.1

General


This publication provides information and guidance to enable an experienced engineer to carry out the design of steel bearing piles. The designer should have sufficient knowledge of the basic physics of soil-pile interaction to be able to design a pile but may need the assistance of geotechnical engineers to interpret the soil parameters from the soils data given in a site investigation report in respect of steel piles. The geotechnical engineer needs detailed references on steel pile load tests in all types of soil to enable characterisation of behaviour that will serve as a basis for the empirical factors to use with generic prediction methods. The information is available from various references including: Clarke et al[1], Biddle and Wyld[22], Jardine et al[112], Harris and Sutherland[91], and Euripides[64]. These are referred to in later Sections. To date, in the UK, design resistance of foundations has been evaluated on an allowable stress basis design (ASD) both for the soils and for the structural components such as piles. However, structural design in the UK has already largely converted to a limit state design (LSD) basis. The structural Eurocodes have all been formulated on an LSD basis whereby partial factors are applied to various elements of the design according to the reliability with which the parameters are known or can be calculated. Eurocode 7 Part 1:Geotechnical Design has already been published in the UK as BS EN 1997-1:2004[9] and presents rules and principles for foundation design on a LSD basis. Limit State Design brings with it a change of emphasis which, when carefully considered, has many benefits for the economic design of piling. The Eurocode approach is particularly rigorous, and this publication adopts the partial factors presented in the Eurocodes. This publication, therefore provides guidance expressed in LSD terminology using the notation given in Eurocode 7[9] where possible, and relates the guidance to previous ASD where it is helpful. However, it has to be recognised that the application of limit state design philosophy to geotechnical design is causing difficulty in a discipline where the Allowable Stress approach and terms such as the ‘allowable bearing pressure’, ‘permissible steel stress’, and ‘allowable pile capacity’ are widely accepted and understood.

2.2

Design standards

British Standards do not cover the geotechnical design of steel piles in any detail, although there is general guidance given in BS 8002[5], BS 6349[6], BS 8004[15]. This publication makes reference to the offshore industry’s recommended practice for steel tubular piles, based on US and UK North Sea experience, which is contained in the American Petroleum Institute Code RP 2A[11] that has been adopted in the ISO Code 13819-2[12]. This has been verified as applicable to steel H-piles by Biddle and Wyld[22] and Jardine et al[112]. Other technical references are used, such as CIRIA Report 103 The design of laterally loaded piles[13], CIRIA Report 104 Design of embedded retaining walls in stiff clay[14], Offshore Technology Conference (OTC) papers and other research papers, and selected textbooks.


7

Printed 07/09/05


BS EN 1997-1:2004[9] was published by BSI in 2004, and the UK National Annex will be available in 2006. It presents a more rigorous treatment of LSD than any of the British Standards relating to foundations so far (BS 8002[5], BS 6349[6], BS 8004[15], BS 8081[16], or BS 8006[17]) and is compatible with the other structural Eurocodes. It is planned that the Eurocodes will co-exist with the British Standards for a period of up to 5 years. The application of LSD methods is only progressed in BS EN 1997-1:2004 to the ‘Rules and Principles’ level but the SCI and Corus have participated in the drafting of Eurocode 3: Part 5 (as EN 1993-5) Design of steel structures piling[7] to ensure there was technical input to that document derived from UK experience and practice. The essence of that work is presented here because it permits adoption of limit state design principles in a rational way for the geotechnical design of steel piles. Allowable Stress Design (ASD) is still permitted in BS 8002[5], BS 8004 and BS 6349[6], to be compatible with the approach taken in BS 449[19]. However, ASD will be phased out as the Eurocodes are adopted between now and 2008. Comprehensive design guidance on all aspects of geotechnical design to Eurocode 7 has recently been published by Thomas Telford[113].


2.3

Limit state design rules

2.3.1 Ultimate limit state axial bearing design of piles Limit state design is a method that achieves a certain level of reliability of structural design against the range of possible adverse variances of presumed loading, strength and behaviour. It assigns partial factors to the presumed values and verifies that, at the ultimate limit state (ULS), the factored design value of resistance (strength) is at least equal to the factored effects (forces, moments, etc.) of the design loads (referred to in the Eurocodes as “actions”). Thus it achieves a level of safety against collapse or failure. A serviceability limit state (SLS) is also considered, at which, typically there is to be no significant permanent deformation or settlement that would affect the use of the structure (or, in the case of a foundation, the structure or other facility that is founded upon it). The SLS is verified using the same presumed loading and strength but smaller partial factors (typically unity). This permits realistic modelling of soil-structure interaction using strains and stiffness to predict pile movements. In BS EN 1997-1 there are three sets of partial factors, one applied to actions, or the effects of actions (denoted by ‘A’), one applied to soil parameters (denoted by ‘M’) and one to resistances (denoted by ‘R’). The values of the partial factors are given in the Eurocode itself but, since the level of safety required is a matter for national choice, the National Annexes are allowed to vary the values of the partial factors. At present there is no UK National Annex and so this publication adopts the partial factors given in the Eurocode (it is likely that the UK NA will also generally adopt those factors). BS EN 1997-1[9] also sets out three possible ‘Design Approaches’ and the National Annex may chose the method to be used. It is likely that the UK will adopt ‘Design Approach 1’ and that approach is described below.


8

Printed 07/09/05


Design Approach 1 For axially loaded piles, adequacy at ULS has to be verified applying two possible combinations of partial factors. These are described as: Combination 1: A1 “+” M1 “+” R1, or Combination 2: A2 “+” M1 or M2 “+” R4. Where A are action factors, M are material factors, and R are resistance factors. See the UK National Annex for the factor values to be used.

2.4

Bearing pile structural design

Currently the structural design of bearing piles is outlined in Section 7 of BS 8004. There it refers to the use of BS 449 for ASD or BS 5950 for buildings or BS 5400 for bridges (the latter two being LSD Codes). Clearly, the designer must choose whether to design his piles to either an ASD or LSD basis. To be consistent with the design basis proposed in this publication, LSD should be used for the structural design but, for information, an overview of both is given below:


ASD design Structural design using ASD principles is referred to in the Corus Piling Handbook[4], in BS 8004 and in BS 8002. These codes invoke the use of procedures that are outlined in BS 449 and in the old CP2: 1951[30]. Neither of these latter codes are in print any longer and only library copies are available. Many temporary works designers in contractors still have an affinity for ASD design because of the increased allowable stresses granted in that code for such temporary works. The ASD basis generally involves use of conservatively assessed safety factors that were appropriate to the then limited knowledge of pile behaviour and the lack of research data using pile instrumentation. Consequently, the degree of utilisation of structural strength of steel piles was lower and the relative cost of the steel option was higher than that used now. This made steel H piles less competitive than concrete piles for many bearing pile applications. The ASD procedures involved using permissible stress values of 0.3fy for axial loading, and 0.5fy for bending moment stresses. For temporary works loads an increase to 0.67fy was permitted. LSD design More economic use of steel piles is now possible as a result of research into pile behaviour using instrumentation to understand the physics of soil-pile interaction under applied static loads and under dynamic forces when driving (see Sections 3,5 and 8). It is now known that fully buried steel piles derive sufficient lateral support from the soil to prevent buckling and no special allowance needs to be made for this in safety factors. (This is covered in the new Eurocode 3: Part 5[7]. The partial factors have values that are related to the degree of confidence that can be attached to each part of the load and resistance equations. For instance, a material factor o = 1.0 is used for steel strength because of the high consistency in high quality steel that is produced in the UK and Western Europe. This factor is applied to the nominal yield strength for each grade of P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

9

Printed 07/09/05


steel. On the other hand, the partial factors on load are 1.35 for dead load and 1.5 for imposed loads, as given in Eurocode 1. The overall factor between nominal load and nominal yield strength is thus between 0.74 and 0.67.

2.5

Design methodology

Obtaining the soil data at the site and the loading data for the project are prerequisites for design. In the first stage of the structural design of a steel bearing pile, the required cross-section is determined based on the design pile resistance needed for the design value of the axial loading that it is required to carry (note that in the structural analysis of the building there is a pinned joint at the connection with the pile and therefore the pile does not need to be designed to carry moment). The pile section shape and steel grade should be selected making an allowance for loss of section due to corrosion according to the required design life (see Section 11).


The second stage is to determine the length of that pile section that is required to provide a design compressive resistance at least equal to the design value of the axial load. Axial load is determined using an appropriate geotechnical prediction method and the soil tests at the site or using measured load resistance from pile load tests (see Section 5). The third stage is to assess the practicality of installing such a pile to the depth required using available driving hammers by using a wave equation programme such as GRLWEAP[27] for a more precise analysis, or using a pile driving formula such as Hiley[28] or Janbu which is comprehensively presented in a paper by Flaate[68] (see Section 8.3) for an approximate check. For friction piles, the desirability of selecting a different pile size or steel grade to adjust the required length can then be judged from sensitivity analyses of various situations to optimise the geometry in relation to driveability, availability from stock, road transport to site, site installation and connections considerations. For end-bearing piles, the provision of a driving shoe might also be evaluated in order to achieve penetration into a sound rock stratum whilst avoiding local buckling of the pile base (see Figure 4.3 for typical examples). Dependent on the pile cross section, this can reduce the available skin friction on the remaining shaft above the tip by over-coring and should be allowed for in design or after pile load tests. The fourth stage is to assess the possible bending stresses that can be induced in the pile, dependent on the type of connection to the structural foundation and the installation tolerances in pile position (see Sections 8 and 10). If bending stresses demand a larger pile size in a group, then a global analysis of the whole foundation (or at least the critical pile group) may be required (see Section 7), in order to apportion the moment between each pile in the group. Each pile will then need a lateral loading analysis, in order to check that the cross section is adequate to take the combined bending and axial force at all levels down the pile according to the pile lateral deflection and the stresses induced, (see Section 6). Fully corroded section properties should be used for the end of design life condition taking account of the corrosion allowance profile with depth. P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

10

Printed 07/09/05


If a different pile size is required for the combined effects, then a new pile length will have to be determined from the geotechnical design and the driveability should be checked anew. The fifth stage is to evaluate the environment and cost benefit factors (see Section 9) for different pile types and configurations and types of connection before selecting a solution and moving onto final detailed design of the connection between pile and structure. As explained in movement. The axial and lateral deflection profile pile length.

Section 3, the generation of soil resistance requires pile new limit state design procedures involve estimation of pile movements in order to satisfy SLS criteria. The lateral will obviously be affected by any change in steel section or


Figure 2.1 shows a flow diagram for the pile design procedure.


11

Printed 07/09/05


Obtain input data: Soil properties Load effects Corrosion allowances

Stage 1

Select initial pile size: Pile type Section size Group configuration

Stage 2

Determine pile length: based on axial resistance

Stage 3

Check pile driveability: Required penetration Driving stresses

Pile driveable?

Change pile type, section or configuration

No


Yes

Stage 4

Design for lateral loads: (if applicable)

Pile suitable?

No

Yes

Evaluate economic & programme aspects: Construction programme, pile types & configurations

Stage 5

Design connections between pile & structure

Figure 2.1


Single pile design procedure flow diagram

12

Printed 07/09/05


3

GEOTECHNICAL DESIGN

3.1

Terminology

Historically, the terminology used by UK geotechnical engineers has been to refer to the ultimate load capacity of a pile as Q, hence Qt=Qs+Qb, where the suffix t is total, s is shaft and b is base. The new Eurocodes have rationalised the symbols used and Eurocode 7 uses R for pile resistance, F for the applied force, and the partial factors involved in the LSD design methods are termed ‘γ’ or ‘ξ’. A comprehensive glossary of definitions is given in BS EN 1997-1[9]. Suffixes to the Eurocode symbols distinguish, inter alia, between characteristic values (R-;k ) and design values (R-;d ). The Eurocode symbols are mainly used in this publication unless indicated otherwise.

3.2

Design premise


The basis of design for any bearing pile is its ultimate axial resistance (‘capacity’) in the particular soil conditions at the site where the structure is to be built. This ultimate resistance can be determined by either: •

load tests on piles at the site, or

•

the use of an empirical formula to predict resistance from soil properties determined by testing.

The design value of the pile resistance is derived from the measured or calculated ultimate resistance by applying appropriate factors and the designer verifies this as adequate to carry the required design loads (actions) from the structure. The ULS procedure is one that is used to ensure that a limit state of failure is avoided. Under ASD (allowable stress design) this used to be achieved by applying ‘Factors of Safety’ but the factors also ensured that settlements were controlled to an acceptably low level. The latter is now ensured by checking the SLS or serviceability limit state separately and it is at this juncture that we must relate to real soil-structure interaction physics to understand what we must achieve in design. Pile movement is needed to generate a soil resistance. The practical design of steel piles therefore involves an appreciation of axial pile strain, shaft wall slip and base movements, and these are obtained from research references and the analysis of pile load tests on site. Reference to a pile head load-displacement (Fc-δ) diagram given in Figure 3.1 permits an understanding of the different pile head deflections that are appropriate to the serviceability limit state (SLS) and ultimate limit state (ULS) specifications for friction pile design. The SLS is generally governed by settlement or deflection, and a working limit of about 10 mm is suggested; and the ULS is generally governed by load to cause failure or near failure, and a practical limit for pile settlement is probably about 40 mm (see Appendix D, page 152 of the ICE Specification for piling and embedded retaining walls)[20].


13

Printed 07/09/05


Reference to Figure 3.8 for end-bearing piles in hard rock shows that even settlements at ULS can be within the elastic compression shortening of a pile, i.e. a few mm (generally <10 mm), where design is governed by the steel material strength and not the rock strength.

3.3

Limit State Design

3.3.1 ULS axial design resistance based on pile load tests The design compressive resistance of a pile, based on load test measurement is given by: R c;d =

R c;k

γt

where: Rc;k

is the characteristic value of compressive resistance

γt

is the partial factor on total resistance for a pile = 1.0 for Set ‘R1’


= 1.3 for Set ‘R4’ (BS EN 1997-1 Table A.6)[9]. The value of the characteristic compression resistance (Rc;k), is a ‘representative minimum’, and depends on the number of pile load tests carried out and the mean and minimum values of the tests. The relationship is expressed in BS EN 1997-1 (§7.6.2.2(8)) as:  ( R c;m ) mean ( R c;m ) min ; R c; k = Min   ξ1 ξ2

  

The factors ξ1 and ξ2 are correlation factors that allow for the confidence by which values can be determined from a small number of tests. The values of the factors are given in Table 3.1, based on Table A.9 of BS EN 1997-1. Table 3.1

Values of correlation factor ξ for static load tests Number of Static Load Tests 1

2

3

4

5

>5

Factor ξ1 on mean Rcm

1.40

1.30

1.20

1.10

1.00

1.00

Factor ξ2 on minimum Rcm

1.40

1.20

1.05

1.00

1.00

1.00

Different correlation factors ξ5 and ξ6 are used for dynamic pile load testing to allow for the confidence by which values can be determined from a small number of tests. The values of these factors are given in Table 3.2, based on Table A.11 of BS EN 1997-1.


14

Printed 07/09/05


Table 3.2

Values of correlation factor ξ for dynamic load tests Number of dynamic Load Tests ≥2

≥5

≥10

≥15

≥20

Factor ξ5 on mean Rcm

1.60

1.50

1.45

1.42

1.40

Factor ξ6 on minimum Rcm

1.50

1.350

1.30

1.25

1.25

Resistance factors The design value of the compressive resistance (Rc;d), is then given by: R cd =

R ck

γt

The value of the partial resistance factor on total resistance, γt is 1.3 for driven piles, as given in Table A.6 of BS EN 1997-1. The partial factors for bored and CFA concrete piles from Tables A.7 and A.8 are also given for comparison purposes in Table 2.2 below.


Table 3.3

Values of the resistance factors, γb, γs and γt in BS EN 1997-1 γb

γs

γt

Driven steel piles

1.3

1.3

1.3

Bored in situ concrete piles

1.6

1.3

1.5

1.45

1.3

1.4

Component factors

CFA (continuous flight auger) in situ concrete piles

For example, comparison with traditional ASD basis for a single driven steel pile test, the relationship between the design value Rc;d and the measured value Rc;m (for Combination 2) is:

R c;d

=

R c;m

γ t ξ1

=

R c;m 1.4 × 1.3

=

R c;m 1.82

Note that the γt resistance factor for driven piles is lower than that for bored concrete piles owing to the greater confidence in driven pile capacity predictions that is due to more consistent behaviour after installation. Traditional practice in allowable stress design procedure (ASD) has been to use a lumped Factor of Safety of 2 for soil resistance on all types of pile, but this has been rationalised due to research that shows that bored concrete piles show more variable behaviour in load tests dependent on the degree of care taken during installation. The limit state design (LSD) procedure taken from BS EN 1997-1, using partial factors on single load tests, gives a total factor of 1.82 between measured and design resistance for steel piles, 2.1 for bored concrete piles and 1.96 for CFA bored piles. This difference is due to the greater reliance of bored concrete piles on end resistance where a lot of disturbance occurs to the soil in the boring process.


15

Printed 07/09/05


It may also be noted that the design value of the load on the pile (against which the resistance has to be verified) is given by: F c;d = γ F × F c;k

where γF is the partial factor on actions and Fc;k is the characteristic value of the load. The value of γF is given in Table A.3 of BS EN1997-1 as 1.35 and 1.5 on permanent and variable actions for Set A1, or 1.0 and 1.3 for Set A2. The overall effect of all the factors is thus approximately to ensure that a factor of 2.5 is maintained between Rc;k and Fc;k.


Axial compressive load

The effect of the application of the separate partial material factors is illustrated in Figure 3.1 using a load-displacement diagram. Measured compressive resistance

R c;m

Characteristic compressive resistance

R c;k

Design value of compressive resistance

R c;d

Characteristic axial load (working load)

F c;k

0

Figure 3.1

10 20 30 40 Pile head displacement (mm)

50 δ

Load-displacement diagram, showing the effect of partial factors

It can be seen that application of the correlation factor ξ, and resistance factor γt, places the design working load on the pile at a level within the elastic range where very little pile head movement is required, thereby satisfying the SLS criterion for allowable settlement, if set at about 10 mm. (By comparison, the generally accepted limit for settlement for structural spread footings is 25 mm and therefore piled foundations give more control over structural movement).

3.3.2 ULS axial design resistance predicted from soil tests The design compressive resistance of a pile, determined from soil tests is given by: R c;d =

R s;k

γs

+

R b;k

γb

which combines equations 7.6 and 7.7 of BS EN 1997-1 (§7.6.2.3) , where: Rs;k

is the characteristic value of shaft resistance of the pile

Rb;k

is the characteristic value of base resistance of the pile


16

Printed 07/09/05


γs

is the partial factor on shaft resistance for driven piles = 1.0 for Set ‘R1’ = 1.3 for Set ‘R4’ (BS EN1997-1 Table A.6).

γb

is the partial factor on base resistance for driven piles = 1.0 for Set ‘R1’ = 1.3 for Set ‘R4’ (BS EN1997-1 Table A.6).

The characteristic values of shaft resistance, Rs;k, and of base resistance, Rb;k, are representative minimum values determined from the relevant geotechnical prediction methods and appropriate soil parameters (see Section 4).


It should be noted that these semi-empirical geotechnical prediction methods are based on load test databases and have to contain conservatively assessed empirical factors, (as stated in the requirement in BS EN 1997-1), to ensure that Rc;k # Rc;m. The overall ‘Factor of Safety’ applied in this limit state procedure therefore comprises a further partial factor, the model factor (to replace the empirical factor in the prediction method), and which allows for the scatter in soil properties and in the load test results. Therefore, where Rs;k, and Rb;k are determined from characteristic values of unit shaft and base resistance, qs;k, and qb;k, the additional model factor should be applied to γs and γb. It is understood that the UK NA (UK National Annexe) will call this factor γ Rd and assign it a value of 1.4. The design compressive resistance of a pile, Rc;d determined from soil tests is then given by: R c;d =

R s;k

γ Rd γ s

+

R b;k

γ Rd γ b

where: Rs;k

is the characteristic value of shaft resistance of the pile

Rb;k

is the characteristic value of base resistance of the pile

γs

is the partial factor on shaft resistance for driven piles = 1.0 for Set ‘R1’ = 1.3 for Set ‘R4’ (BS EN1997-1 Table A.6).

γb

is the partial factor on base resistance for driven piles = 1.0 for Set ‘R1’ = 1.3 for Set ‘R4’ (BS EN1997-1 Table A.6).

γ Rd

is the model factor given in the UK NA that is due in 2005.

3.3.3 ULS axial design resistance for piles end bearing in rock Where the ground conditions at the site include an underlying rock stratum within a driveable depth, the only reliable method for the designer to determine the ultimate load capacity is to carry out a pile load test. The same framework of design rules apply as for soils, except that the endbearing resistance from the rock at the base will dominate. Even if there are


17

Printed 07/09/05


overlying soils, the elastic compression due to applied load on the pile will be so small, due to the high stiffness of the base resistance, that some of the potential frictional resistance on the pile shaft cannot be mobilised. The use of the ‘nominal allowable’ rock bearing pressures that are given, for example, in BS 8004[15], page 11, or in the API Code RP2A[11], such as 10 MPa or 15 MPa, will greatly underestimate most rock resistances and lead to uneconomic designs for steel piles. This is because such ‘limit’ judgements were made for spread footings, and the extension of that bearing capacity theory to deep bored cast in situ concrete piles in clay is not relevant to driven steel piles. In addition, there is the difficulty of drilling a clean rock socket without leaving a layer of soft compressible drill cuttings under the toe of the concrete pile. Further information is given in the CIRIA Guide R181: Piled foundations in weak rock[21].


Due to the high variability of rock types in the UK, site specific load testing is always required to ascertain capacity. The ultimate design resistance of steel piles driven into sound rock is often governed by the allowable stress in the pile section. Steel bearing piles are ideally suited to piling in rock because no excavation is required as with bored concrete piles, and any variations in peak load or in the degree of weathering in the rock can be accommodated by varying the driven length. Their small displacement also ensures penetration to a sound layer (see CIRIA Guide R181). SCI’s database of steel pile load tests, reported in Validation of vertical load capacity prediction methods for steel bearing piles[29], includes tests with end bearing into rock, and those results together with current accepted practice from various sources are given in Section 5.6. The basic recommended procedure is to plan the site investigation to include soil penetration testing (SPTs and CPTs)[26] which will help to differentiate the weathered rock layers from the intact rockhead levels. From offshore experience and European experience with the CPT, it is known from pile driving back-analysis and static load testing that the CPT ‘qc’ value can be assumed as an ultimate unit resistance pressure beneath the steel pile wall tip area. Since the limiting pressure of the load cells within the CPT tool is about 70 MPa to 100 MPa, this should be adequate to cover most of the soft rocks found in the UK, e.g. mudstones, sandstones, chalk and their weathered derivatives. In harder rocks, like granites, metamorphic types, carboniferous limestones and intact unweathered sedimentary types, the unit end bearing resistance is more likely to be of the order of 200 MPa to 400 MPa or more (see Section 5.6). In many cases of end-bearing into rock therefore, the ultimate load capacity of a steel pile is governed by the steel yield stress, and not the rock resistance limit.

3.3.4 SLS axial bearing design SLS axial bearing resistance will be the pile load resistance at a selected pile head settlement that is structurally acceptable to the designer. The design pile load is determined in the LSD procedure by setting all the partial factor values to 1.0.


18

Printed 07/09/05


In practical terms, the designer has two choices, i.e.: •

to specify the SLS criterion to be a pilehead settlement at which there is no permanent ‘set’ of the pile (a ∆s1 of 10 mm is suggested), or

•

to specify a pilehead settlement that is the maximum that the supported structure can sustain without affecting its serviceability (a ∆s2 of 25 mm is suggested for buildings, or perhaps 10 mm for bridge foundations).

Obviously, the proportion of the potential maximum ultimate capacity of the pile achieved at either of these SLS criteria will vary dependent on the pile cross-section, pile length and the soil resistances in friction and end-bearing on a particular site. Also, the major proportion of pilehead settlement will be permanent under the dead load component. However, serviceability criteria, in practice, rarely govern steel pile design because movements are small.

3.3.5 ULS lateral load design resistance The design resistance of a transversely loaded pile is termed Rtr, according to BS EN 1997-1:2004[9], and it must be demonstrated that Ftr;d # Rtr;d


where: Ftr;d

is the design value of transverse load, and

Rtr;d

is the design resistance to transverse loading taking into account any effect of coexisting axial pile loading

Useful guidance is given in Sections 7.3.2.4 and 7.7 of BS EN 1997-1[9] to assist the designer to judge the criteria applicable to the design of piles and pile groups for lateral loading.

3.3.6 SLS lateral load resistance The serviceability limit state for transverse loading of a pile can be defined as the pile head loading and the resulting soil resistance distribution that occurs at the maximum allowable in-service transverse pile head deflection of the supported structure that is permitted or is structurally imposed at the structure/pile connection. The design pile load is determined in the LSD procedures by setting all partial factor values to 1.0. Advice on the geotechnical design and analysis of piles for lateral load resistance is given in Section 6 and where the contribution to lateral loading resistance of a vertical bearing pile is required, the designer is recommended to follow the guidance given in CIRIA Report 103 The design of laterally loaded piles[13] and the textbooks by Poulos and Davis[23], and Tomlinson[24]. Guidance on pile group effects is given in Section 7.

3.4

Geotechnical design methods

Soils are characterised as either ‘clays/cohesive’ or ‘granular/non-cohesive’ types in order to separate their two fundamentally different behavioural responses to applied pile load. The generic formulae used to predict soil resistance to pile load include empirical modifying factors (see Section 5), which can be adjusted according to previous engineering experience of the


19

Printed 07/09/05


influence on the accuracy of predictions by changes in soil type and other factors such as the time delay before load testing. It will be shown in Section 3.7 that the mechanisms of axial load transfer involved in shaft friction Rs and base resistance Rb are completely different. The separate prediction of shaft friction and base resistance therefore forms the basis of all ‘predictive’ calculations of pile load-carrying capacity. The basic equations to be used for this are written as: Rc = Rs + Rb - Wp

(3.1)

and, Rt = Rs + Wp

(3.2)

where: Wp

is the weight of the pile


The weight of the pile (Wp) should be included in the actions acting on the pile foundation and the increased end-bearing resistance due to overburden pressure included in the base resistance. Since these terms often cancel each other out, it is common to ignore them (although strictly they should be included in the calculation). There is a move towards applying reliability criteria to evaluate structural design procedures in construction, but care should be taken in applying these to geotechnical methods. Many geotechnical design methods rely on averaging soil properties over the length of a pile, and practitioners have found that simple formulae can be used with confidence to represent soil response to applied load, provided that expert judgement is applied to the selection of the soil parameters involved. The crucial skill involved is the ‘knowledgeable judgement’, because there is usually such a wide variation in soil strength and properties within a site that it defies use of a precise interpretive formula. Statistical analysis procedures for soil spatial variables are not relevant either, because many of the soil response parameters are also time-dependent. Refinement of geotechnical design methods is difficult to justify because of the considerable scatter in all pile load test databases that compare Rm to Rc, and this indicates that our knowledge of soil-pile interaction and the ways in which we apply it are, as yet, imprecise (see Section 5.1). It is therefore preferable that each formula involves as few variables as possible, to permit designers to appreciate ‘cause’ and ‘effect’ during the analysis of problems and thereby to aid their judgement. The design basis described in the following sections requires the use of either measured pile resistance from load tests or predicted pile resistance using soil tests at the site and empirical generic formulae. The prediction methods and formulae available and the various aspects of soil mechanics are explained in detail in Sections 5, 6, 7 and 8.


20

Printed 07/09/05


3.5

Soil resistance on driven steel piles

Soil resistance is mainly governed by the type of soil, the type of disturbance caused to the soil by the installation of the piles and the nature of the interface that is created between the surface of the pile and the adjacent soil. In the case of steel piles, there is no relaxation or softening of the soils, as there may be in concrete bored or CFA piles, but there is considerable remoulding of the soil surfaces in contact with the steel shaft and base caused by forcing the pile into the soil. To complicate matters further, the soil resistance to any applied load on driven steel piles is time-dependent. During driving, the frictional resistance is lowered in the remoulded soil zones that are immediately adjacent to the pile wall. In fine granular soils, this remoulding is often a liquefaction that is caused by the high local porewater pressures that result from displacement of the soil structure to accommodate the steel pile volume. In clays, this is generally a plastic deformation of the clay structure that is accompanied by porewater pressure changes.


Set-up

The soil frictional resistance to applied pile axial load recovers within a finite time interval, and this time interval is dependent on the permeability of the soil and the structure of the soil fabric (i.e. the presence of discontinuities such as fissures or laminations or lenses within the soil mass can contain more permeable soils and provide local porewater pressure drainage paths). As an indication, full recovery of shaft resistance may take seconds in a coarse granular material; minutes in sand; hours in a silt or clayey silt; days in a sandy clay; and many months in a high plasticity clay. This phenomenon is referred to as ‘set-up’ and appreciation of its time dependency is essential to understanding pile load test results and to planning a trouble-free installation. References Clarke et al[1], Fellenius[113] and Jardine et al[3][112], give data on setup from research measurements on full-scale steel piles. As a result of pile driving, the maximum shaft resistance can only be achieved in pile load tests if sufficient time is allowed between completion of driving and the commencement of loading for full set-up to occur. Where this is not practical, as in heavily overconsolidated very plastic clays, (e.g. London Clay, Oxford Clay, Weald Clay etc.) the effects of set-up must be allowed for in design by applying appropriate empirical factors that have been derived from a load test database for each particular type of soil and pile (see Section 5). Dynamic load testing of piles may also be used to investigate set-up, particularly on test piles (see Jardine et al[112], Fellenius[113], Komurka[115], and Section 8).

3.6

Load / settlement behaviour – friction piles

The settlement of a pile head resulting from progressively increasing compressive load in maintained load stages, i.e. effectively a series of static loadings on the pile, can be represented as a pile ‘load-settlement’ curve, or an Fc - δ diagram, such as that shown in Figure 3.2.


21

Printed 07/09/05


Pile head axial compressive load (Fc )

Discuss me ...

D

Rc

B

Reloading Unloading

A

Initial loading

0 C

Pile head vertical deflection (δ )

Figure 3.2

Axial load-settlement for a friction pile (Fc − δ) curve

Pile head axial compressive load (F c )


The load settlement response is composed of two separate components, the linear elastic shaft friction Rs and the highly non-linear base resistance Rb (see Equation 3.1). These are shown diagrammatically in Figure 3.3.

Rs

Shaft resistance

Rb

Base resistance

0

Figure 3.3

Pile head vertical settlement ( δ )

Resistance Components in (Fc − δ) curve

3.6.2 Linear elastic response of pile Initially, the pile-soil system behaves in a linear-elastic manner up to some point A on the Fc - δ diagram in Figure 3.2. Applying load to the head of the pile produces axial strain in the steel pile shaft wall and a corresponding downward movement with slippage at the pile wall /soil interface. Load transfer occurs in the form of shaft friction that at any level on the pile has an elastic-perfectly plastic load-displacement relationship (see Figure 3.4 for load in pile due to shaft friction resistance).


22

Printed 07/09/05


Hence the upper part of the pile’s shaft compresses and transfers load to the upper soils and if the load is released at any stage up to this point, the pile head will rebound elastically to its original level as the shaft steel relaxes (see Figures 3.5 and 3.6 for examples of pile head load-displacement relationships from load tests that demonstrate the repeatability of this phenomenon). Negligible end-bearing is mobilised up to this point A.

Load transfer (kPa)

300

250

200

150 Penetration m 14 15 16.358 17.6

100

50

0


-50

0

10

20

30

40

50

60

70

80

90

100

110

Movement (mm)

Figure 3.4

Load in steel pile wall at different levels due to shaft friction resistance (as measured in LDP tests, Paper 13 pg 297, Clarke et al [1])

3.6.3 Elastic-plastic response of pile The onset of nonlinear behaviour at point A in Figure 3.2 is associated with the development of base or end bearing resistance Rb as the load strain in the shaft reaches the pile base level and the lower end of the pile starts to move downwards. Further movement will lead to the mobilisation of full shaft friction Rs by some point B. If the load is released at this stage, the pile head will rebound to some point C, the amount of ‘permanent set’ being the distance OC, which is mostly the irrecoverable settlement of the pile base sustained in generating a proportion of the base resistance (∆Rb), the shaft friction movement being, as explained, an elastically recoverable component. The latter phenomenon is illustrated in Figures 3.5 and 3.6. It should be noted that a small residual compression force may remain in the pile wall after unloading, as measured in pile load tests (see Clarke et al[1]), especially for long piles and where the proportion of friction is high. This residual load may cause a corresponding small contribution to the irrecoverable pile head settlement. The pile head settlement required to mobilise the full shaft friction Rs is comparable to the elastic compression of the steel wall, i.e. only of the order of 7 mm to 10 mm for piles of typical length 15 m to 20 m.


23

Printed 07/09/05



Discuss me ...

Figure 3.5

Example pile head load/displacement relationship for repetitive loading in a normally consolidated clay (as measured in LDP tests, Pentre site, Paper 13, pg 283, Clarke et al [1])

Figure 3.6

Example pile head load/displacement relationship for repetitive loading in an overconsolidated clay (as measured in LDP tests, Tilbrook Grange site, Paper 13, pg 283, Clarke et al [1])

The full base resistance of the pile Rb requires a greater settlement for its mobilisation, and the amount of movement is related to the size of the pile base area involved. For unplugged steel piles this will depend on the wall section thickness or in the case of fully plugged piles, on the diameter or full base width of the pile. For H-piles or sheet piles, the movement may be 2 to 3 times the steel pile wall thickness (i.e. 30 to 40 mm) to generate the wall tip bearing resistance (see page 152 of the ICE Specification for piling and embedded retaining walls[20]). For a fully plugged pile on the other hand, δOC on Figure 3.2 may be of the order of 10% of the base diameter or width, depending on the soil type. See Section 5 for further discussion on when to assume plugging. Note that if the pile base is in dense sand or rock, the end bearing may be developed with negligible base settlement and the compression of the pile shaft may be insufficient to mobilise the full potential shaft friction (see Section 3.9).


24

Printed 07/09/05


When the stage of full mobilisation of the base resistance or ultimate base resistance Rb is reached (i.e. at some point D in Figure 3.2), the pile will settle at an increasing rate under only very small further increases of load (near to the ‘ultimate pile resistance’ asymptote). Extended loading periods during pile tests indicate that it is very difficult to achieve the ultimate axial compression resistance, because the curve becomes virtually flat, and to reach the asymptote requires very large settlements, (see Figure 3.1). However, a pile load test in soil should aim to achieve within about 5% of that value and accepted practice for friction piles is to use the load resistance reached at a tip movement of about 30 to 50 mm.

3.7

Pile-soil load transfer – friction piles


The process of driving a steel pile in clays and sands produces a thin layer of completely remoulded soil adjacent to the pile shaft wall that acts as a ‘slip’ and load-transfer layer; its behaviour is now well understood as a result of research on trial piles (Reference Tomlinson[25]; and Clarke et al[1]). If strain gauges are installed at various points along the steel pile shaft, the compressive load remaining in the pile can be measured at each level; the distribution of load in the pile is found to be in the form of that shown in Figure 3.7 (which shows the transfer of load from the pile to the soil at each stage of loading identified in Figure 3.2). Thus when loaded to point A in Figure 3.2, the whole of the load is carried by skin friction on the pile shaft and there is no transfer of load to the base of the pile (Figure 3.7(a)). When the load reaches point B, most of the pile shaft friction is mobilised and the pile base has started to feel load (Figure 3.7(b)). At point D, there has been no further increase in the load transferred in wall friction but the base load will have reached its maximum value (Figure 3.7(c)), i.e. the ‘ultimate pile bearing capacity’ is reached, beyond which the pilehead will move down vertically under nearly constant load. Load on head of pile F c

Full shaft friction resistance Rs Fc

'ULS deflection failure' load on pile Fc =R c Rb

∆Rs

Base of pile (a) F c = ∆ R s

Base reaction ∆ R b
Rs

Full base resistance Rb

(b) F c =R s +∆R b

(c) Fc =R c=R s+ R b

(a) Load in pile at point A on load-settlement curve in Figure 3.2 (b) Load in pile at point B on load-settlement curve in Figure 3.2 (c) Load in pile at point D on load-settlement curve in Figure 3.2

Figure 3.7


Compressive load transfer (pile to soil) from shaft and base for a friction pile

25

Printed 07/09/05


3.8

Load / settlement behaviour – end-bearing piles

Pile head axial compressive load (Fc )


If the pile base is in dense sand or rock, the base or end bearing resistance can be developed with very little base movement and the compression of the pile shaft is often insufficient to mobilise the full potential shaft friction resistance in the soils overlying the rock layer. Often the pile head moves only due to elastic compression of the steel wall up to the predicted ultimate pile resistance since there is negligible ‘set’ at the base, see Figure 3.8. The total pile head movement will obviously be dependent on the pile length required to reach the rock or other dense bearing stratum and the design is governed by steel material strength and not by rock strength.

Rc

C

B

A 0

δ <10 mm Pile head vertical settlement ( δ )

Figure 3.8

Pilehead load-settlement curve for an ‘end-bearing’ pile

In many end-bearing piles, the pile base resistance will be controlled by structural design considerations to limit the stress in the steel wall so as to prevent local yield or ‘buckling’ during driving and not governed by deformation or allowable bearing pressure in the rock at the pile tip. A suggested approach for economic design of such piles is explained in Sections 2.5 and 5.6.

3.9

Pile-soil load transfer – end bearing piles

The pile-soil load transfer diagrams for end bearing piles are very different to those for friction piles and a typical generic diagram is shown in Figure 3.9. It shows the load transfer from the pile to the soil and rock, at points A, B, and C on the pile head load displacement curve shown in Figure 3.8. At point A, proportions of the pile load are taken in both shaft friction and end bearing because the high stiffness at the tip causes reaction at very little pile head movement. By point B, more shaft friction may be developed, but a greater proportion will be carried by the pile base. And by point C, the full pile base resistance has been reached whilst the pile may start to deform plastically due to local yielding near to the pile tip which is the onset of structural failure. In the Figure3.9, ∆Rb denotes a portion of the ultimate base resistance, and ∆Rs denotes the portion of the ultimate shaft friction that is available.


26

Printed 07/09/05


Fc ∆Rs

Fc

Fc

∆ Rb

∆R s

Rb

∆Rs

∆ Rb (a) Fc = ∆ Rc (b) Fc = ∆ R c = R s +∆ R b = R s+ ∆ R b

Figure 3.9

(c) Fc =R c =∆ R s +R b

Compressive load transfer, tip in rock


3.10 Site investigation There are many good publications and references that can be used to decide the scope of soil tests in a site investigation, but very few address the specific requirements in respect of steel bearing piles. It has been found that in situ testing of soils is particularly relevant to all types of driven pile. Soil testing should apply a method of loading to soils that resembles as closely as possible the type of loading that is to be applied by the pile to the soil. In this respect, soil tests are therefore required to provide the properties relevant to predicting the response of soil to the various phases of construction, namely: •

Pile driving.

•

Pile loading during construction.

•

Pile static loading during working life.

•

Pile live loading (transient) during working life.

3.10.1 Soil test data for design Granular soils

In situ soil testing should comprise the use of the following: •

The Standard Penetration Test (SPT) as specified by BS 1377-9: Methods of test for soils for civil engineering purposes: Part 9: In situ tests[26], is a universal test applicable to all types of granular soil for which it has been extensively calibrated for the prediction of pile driving resistance, shaft friction and end-bearing correlations.

•

The Cone Penetration Test (CPT) also specified by BS 1377-9, has been extensively calibrated against steel pile design parameters in fine grained granular soils (sands, silts and clays). An explanation of the interpretation of CPT test results to derive soil design parameters is contained in Cone Penetration testing in geotechnical practice[108].


27

Printed 07/09/05


•

The Dilatometer Marchetti Test (DMT), used to determine the in situ earth pressure coefficients and confined modulus of soils for use in estimates of lateral soil resistance to applied displacement or force.

•

The pressuremeter, used extensively in France and increasingly being applied in the UK to derive in situ soil properties relevant to driven piles.

Laboratory testing should include: •

Saturated and unsaturated bulk densities (unit weight).

•

Shear box tests to determine the angle of internal friction (φ′).

•

Particle size distribution classification tests.

Cohesive soils

For cohesive soils, the geotechnical pile design and resistance prediction methods for axial loading generally rely on correlations of pile behaviour with the undrained cohesive strength cu, but care should be taken to select the soil strength at a consistent strain to failure. This has been addressed in Norwegian and offshore specifications for triaxial soil testing and is taken as the strength at failure or at a strain of 4%, whichever occurs first.


For lateral loading and retaining wall design, the geotechnical methods for limit state design now require the following deformation and stiffness properties: •

Young’s Modulus (E50 and initial tangent modulus).

•

Poisson’s ratio.

•

Coefficients of subgrade reaction and horizontal subgrade reaction.

For earth pressure calculation the geotechnical methods for limit state design require the following properties: •

Coefficient of earth pressure at rest, Ko.

•

Coefficients of active and passive earth pressure (Ka and Kp).

•

Consolidation and permeability characteristics.

3.10.2 Selection of soil parameters As mentioned in Section 3.4, many geotechnical design and prediction methods require the judgement of average soil parameter values for each soil layer. This interpretation requires experience because there are several processes involved in making the judgement, including: •

Classifying and characterising the soils and selecting soil layers.

•

Selecting the soil properties that are best suited to the type of soil and the geotechnical pile design or resistance prediction method that is most appropriate to that type of soil.

•

Collating and interpreting the soils data, including checking the validity of each datapoint, e.g. an cu soil strength value, because some may be too low or too high for various reasons and may therefore not be representative of the soil layer.

The designer should ensure that an engineer with relevant geotechnical experience and knowledge of steel pile design and performance is involved in this judgement process at the concept design stage, otherwise uneconomic or unsafe judgements of soil parameters may result. P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

28

Printed 07/09/05


4

SELECTION OF SECTION

4.1

Steel piles in bearing only


The pile section should be selected in order to try to optimise the shape and pile length according to the soil resistance that is available from the soil layers present at the site and to avoid problems with pile driving. A constraint will be the structural stiffness required to limit any lateral deflection under load; additionally, the pile movements necessary to mobilise the required bearing capacity should be within the serviceability limits specified for the structure. Although pile lateral deflections can be estimated within a tolerable accuracy (see Section 6), the axial load movements cannot be estimated without a pile load test at the site. The inaccuracy can be minimised by using the general guidance contained herein but there will always be an element of over-design until the results of pile load tests determine the actual pile head load-deflection behaviour as explained in Section 3. The same constraints apply to concrete piles, and piled foundation design can only be finalised after pile load testing. If load testing is not carried out at concept design stage, then the piled foundation design cannot be optimised nor can the advantages of steel piling be fully utilised in the construction programme (see Section 9). The traditional approach of passing all risk in piling to the specialist contractor leads to higher structural costs and precludes the selection of the most fit-for-purpose piling solutions that only the designer is able to judge. However, as the designer is ultimately responsible for the foundation design under the CDM Regulations, he must now be involved in the pile design. Understanding the mechanisms of soil resistance to applied pile load is therefore crucial to the acceptance of steel piling for bearing applications. This guide therefore explains the fundamental concepts of steel pile-soil load transfer and presents design rules for all types of pile section that have been soundly derived from pile load testing. Onshore, the commonly used steel pile sections are H piles and tubular piles. To be able to predict load resistance using established offshore methods (such as those for the load resistance of steel tubular piles, as given in the American Petroleum Institute Recommended Practice API RP2A[11]) it was necessary to carry out a validation exercise. The SCI collected, collated and analysed data from load tests on steel piles, other than tubular piles, in a joint industry project in the period 1995 to 1996; selected onshore methods were also compared[22][29]. These methods are explained in Section 5. The methods for granular soils were found to have different reliabilities, depending on the soil type, and for cohesive soils a modified α - method has been derived for H-piles and sheet piles.

4.2

Design method examples

The SCI has produced several publications that provide guidance to assist designers in the application of design methods for steel bearing piles, these are: P163

Integral steel bridges: Design guidance[31]

P180

Integral steel bridges: Design of a single span bridge - Worked Example[32].


29

Printed 07/09/05


P275

4.3

Steel Intensive Basements[109]

Selection of steel section

Steel piles are selected on the basis of overall geometry to suit the geotechnical requirements and wall thickness to suit the load, material strength and the driving stresses to be sustained. H-piles and universal beams (for High Modulus Piles) and other plates and sections are produced to BS EN 10025[34], Grades S275 and S355. Steel tubular piles are produced as linepipe to API 5L[35] Grades X52 up to X80.


4.4

H-Piles

Steel H-piles are very efficient in providing a large surface area for generating shaft friction resistance. For example a 305 mm×305 mm×110 kg/m H-pile section has an external surface equivalent to a concrete pile of diameter 601 mm where its small displacement volume is only 5% of that of the concrete pile, which enables it to be driven with less energy and into more dense soils. The displacement volumes of the Corus range of 305 mm×305 mm piles cover a range of 3% to 8% of that of equivalent concrete piles. In any given foundation plan area therefore, a greater number of steel H-piles can be provided in a group with a standard spacing of 2B (or 2 dia.) than concrete piles and either the load supported can be greater or, if soil conditions permit, the driven steel piles can be shorter to support a given structural load. The stiffness of H-piles is different on each orthogonal axis, allowing designers to select the orientation necessary to achieve the most efficient design. Dimensions and properties of the current Corus range of H-piles are given in Table 4.1. H-piles are also used very effectively to transfer bearing loads into buried rockhead and to get around buried obstructions (see Section 3.3.3). CIRIA R181[21] recommends steel H-piles as ideally suited to piling in weak rock and advises the consideration of special cast steel pile shoes to strengthen the tip and prevent risk of damage or buckling under hard driving conditions.


30

Printed 07/09/05

Universal bearing H-Pile dimensions and properties

31

108.9 346.4

222.9 337.9

186.0 328.3

149.1 318.5

126.1 312.3

110.0 307.9

94.9

88.0

78.9

85.1

71.0

63.0

53.9

44.9

356x368x109

305x305x223

305x305x186

305x305x149

305x305x126

305x305x110

305x305x95

305x305x88

305x305x79

254x254x85

254x254x71

254x254x63

203x203x54

203x203x45

205.9

207.7

256.6

258.0

260.4

306.4

307.8

308.7

310.7

312.9

316.0

320.9

325.7

371.0

9.5

11.3

10.6

12.0

14.4

11.0

12.4

13.3

15.3

17.5

20.6

25.5

30.3

12.8

15.6

17.8

20.3

mm

9.5

11.4

10.7

12.0

14.3

11.1

12.3

13.3

15.4

17.6

20.7

25.6

30.4

12.9

15.7

17.9

20.4

mm

10.2

10.2

12.7

12.7

12.7

15.2

15.2

15.2

15.2

15.2

15.2

15.2

15.2

15.2

15.2

15.2

15.2

mm

Note : Plunge columns use structural UC sections.

200.2

204.0

247.1

249.7

254.3

299.3

301.7

303.7

133.0 352.0 37 3.8

356x368x133

376.0

378.5

152.0 356.4

mm

173.9 361.4

mm

356x368x152

kg/m

160.8

160.8

200.3

200.3

200.3

246.7

246.7

246.7

246.7

246.7

246.7

246.7

246.7

290.2

290.2

290.2

290.2

mm

10.8

9.11

12.0

10.8

9.10

13.8

12.5

11.6

10.1

8.89

7.63

6.27

5.36

14.4

11.9

10.5

9.28

cm4 cm

cm

16.9

14.2

18.9

4100

5027

8860

16.7 10070

13.9 12280

22.4 16440

19.9 18420

18.5 20040

16.1 23560

14.1 27410

1384

1705

3016

3439

4215

5326

5984

6529

7709

9002

8.46 4.92

8.55 4.98

10.5 6.13

10.6 6.17

10.6 6.24

12.8 7.28

12.8 7.31

12.9 7.35

13.0 7.42

13.1 7.49

12.0 33070 10910 13.2 7.58

9.67 42610 14140 13.4 7.73

8.14 52700 17580 13.6 7.87

22.7 30630 10990 14.9 8.90

18.6 37980 13680 15.0 8.99

16.3 43970 15880 15.1 9.05

14.3 51010 18460 15.2 9.13

cm4

Radius Thickness Ratios for Second Moment Mass Depth Width Depth of Area of Gyration of Root between Local Buckling of per of metre Section Section Web Flange Radius fillets Flange Web Axis Axis Axis Axis D B r d t T B/2T d/t x-x y-y x-x y-y

356x368x174

Serial Size

Designation

Universal Bearing Piles Dimensions & Properties to BS4 Part1 1993

Table 4.1


410

493

717

807

966

1099

1221

1320

1531

1755

2076

2596

3119

1769

2158

2468

2823

cm3

Axis x-x

134

164

235

267

324

348

389

423

496

575

691

881

1079

592

732

845

976

cm3

Axis y-y

Elastic Modulus

459

557

799

904

1092

1218

1361

1475

1720

1986

2370

3003

3653

1956

2406

2767

3186

cm3

Axis x-x

T

t

y

x d

206

252

360

409

498

531

595

648

762

885

1065

1366

1680

903

1118

1293

1497

cm3

0.827

0.827

0.827

0.826

0.825

0.832

0.830

0.830

0.830

0.829

0.828

0.827

0.826

0.823

0.822

0.821

0.821

18.6

15.8

20.5

18.4

15.6

23.9

21.6

20.2

17.7

15.7

13.5

11.1

9.55

24.2

20.1

17.8

15.8

0.126

0.158

0.421

0.486

0.607

1.11

1.25

1.38

1.65

1.95

2.42

3.24

4.15

3.05

3.87

4.55

5.37

dm6

19.2

32.7

34.3

48.4

81.8

46.9

64.2

80.0

122

182

295

560

943

84.6

151

223

330

cm4

57.2

68.7

80.2

90.4

108

100

112

121

140

161

190

237

284

139

169

194

221

cm2

Buckling Torsional Warping Torsional Area of Axis Parameter Index Constant Constant H J u Section x y-y

Plastic Modulus

D x

r

B y


Discuss me ...


4.4.1 Special modification to H-piles


Some examples of toe protection shoes for steel H-piles are shown in Figure 3.3 which are taken from standard details provided by specialist suppliers in the USA and Norway. Various shapes are available for different purposes to: •

break through debris, scree, dense gravel, boulders and weathered rock surfaces

•

seat the pile tip into sloping rock with a toothed key to prevent the pile sliding

•

increase the bearing area

•

prevent local buckling damage to the pile tip.

Figure 4.1

Examples of toe protection shoes for steel H-piles

It should be noted that the provision of such shoes to piles can result in a significant loss in skin friction resistance in the soil layers overlying the rock or hard stratum due to ‘overcoring’. Judgement is therefore required as to where the majority of pile resistance to load is to come from before deciding on such section modifications. See also comments in Section 3 on the development of end-bearing and skin friction in terms of the pile head movements required. CIRIA R181[21] also offers advice on the problems in judging the likelihood of an H-pile section ‘plugging’ when driving into weak rock. Research into the relative advantages for using welded-on wing plates and cruciforms to the toe of an H-pile in slatey mudstone has been carried out by Tomlinson as reported in CIRIA Report 066[113]. These tests produced an average end bearing resistance of 41 MPa and a range of 28-58 MPa in the slatey mudstone at Teignmouth in Devon.

4.5

Plunge Columns

4.5.1 Definition Plunge columns are being used with top-down basement construction to form structural columns before casting the ground floor slab. The plunge columns are carefully exposed during the soil excavation one floor at a time, and facilitate support of the concrete floor slabs by means of welded steel platforms. A plunge column generally comprises a steel structural UC section that is plunged into the wet concrete of a bored pile that is formed at ground level before a top-down basement construction starts. It transfers the structural


32

Printed 07/09/05


column load to the concrete pile by means of bond shear stress over a design length or by using welded shear studs.

4.5.2 Available systems There are several plunge column systems available, each proprietary to a construction contractor or foundation contractor. The principal design is similar, though the systems differ in the means by which the steel UC is suspended in the concrete to achieve verticality whilst the concrete sets. The systems generally include a temporary guide or jig on a temporary steel pile casing from which the column-pile is suspended to achieve this verticality. An example is given below. CEMLOCTM System

The CEMLOCTM system, patented by Skanska Cementation Foundations, has been specifically developed to accurately place plunged steel columns into bored concrete piles. Figure 4.2 shows a schematic of the system.


Orientation dowels fixed to casing

Temporary casing length variable

Bore either dry or flooded

Frame of Cemloc overall length adjustable

Column plan position is adjusted before plunging into concrete Pile bore Concrete casting level

Figure 4.2

The CEMLOCTM system (courtesy of Skanska Cementation Foundations)

Standard techniques are used for constructing a rotary bored pile. After the pile bore has been completed, the reinforcement cage is installed. One of two different techniques are used, depending on circumstances. In one technique the concrete is placed by tremie-pipe to a lower casting level. The CEMLOCTM jig is lowered into the temporary pile casing and the correct orientation achieved by means of locating dowels. The unit is then locked onto the inside of the casing using hydraulic rams near the head and near the base of the jig. The column is


33

Printed 07/09/05


lowered through the jig and held with its base above the cast concrete level. The plan location of the column, at ground level and reduced level, is adjusted by the steering system and confirmed by EDM surveying techniques and the verticality by means of laser plumbing. The column is then plunged into the fresh concrete. Alternatively, where protrusions on the flanges (shear studs or welded shear plates) are required along the embedment length, and/or at basement floor levels to ensure load transfer, the column can be accurately positioned first within the reinforcement cage and then the concrete placed by pump hose or tremie-pipe around it. In both techniques, the jig is left in position until the concrete has gained sufficient strength. It is then withdrawn and the annulus around the column is filled with granular material. In certain circumstances, layers of weak concrete fill may be placed to give additional lateral support. Finally, the temporary casing is withdrawn. The installation sequence is shown schematically in Figure 4.3.


1. Temporary casing placed and pile bored to depth (dry or under bentonite)

3. Cemloc unit placed accurately in plan orientation and locked to inside of casing

2. Concrete poured to casing level (above final trim level)

Figure 4.3

5. Column plunged into pile concrete and held until set. Embedment length determinred by column section and load

4. Column placed through Cemloc unit and plan position accurately adjusted at ground and reduced level

7. Backfill placed around column and casing removed

6. Cemloc unit removed and upper spacer fitted

Installation sequence for Cemloc plunge column system

This approach has many advantages, including a faster construction cycle, no requirement for a permanent casing to ground level and no requirement to prepare a pile cap. Also there is no special requirement for safety equipment and procedures since the works are carried out at ground level. Tolerances compatible with fabricated structural steel can be achieved, resulting typically in levels of accuracy of ± 10 mm in plan position, ± 10mm in level and up to 1:600 in verticality. A better verticality is achieved by suspending the column above the wet concrete level and then filling the remainder around it by pump hose or tremie pipe. Where specified tolerances of level are given by the designer, the contractor should always err on the side of being low rather than high, so the tolerance becomes +0, –20 mm, because it is always easier to accommodate a low connection interface by using packs. It is recommended that the design of the steel column section above the concrete pile and the shear connection follow structural design procedures that are laid down in Eurocodes EN 1993-1 and EN 1994-1.


34

Printed 07/09/05


As no definitive design procedures to date have been presented in British Standards for plunge columns and little but no directly relevant information is given in ENV 1994-1, numerous design methods have been postulated by contractors and consultants for the design of the load transfer section between the column and concrete pile (see Section 9 case studies). Skanska Cementation Foundations base their design on bond between steel section surface and the pile concrete over an embedment length. Skanska Cementation Foundations postulate the embedment length is given by: 2W

=

Le

L p × 0.35

f cu

where: W

is the unfactored column load

Lp

is the perimeter length of column section

fcu

is the characteristic strength of concrete.


The maximum length Le is limited to 5 m. If this length does not generate sufficient bond area, then additional load transfer flanges are welded to the steel section.


35

Printed 07/09/05


5

AXIAL LOAD RESISTANCE

5.1

Interpretation of soil parameters

The engineering properties of soils are subject to variation both horizontally and vertically because they are deposited in a variable and random manner. Even within a given geological layer, soils will not be homogeneous. Any attempt to rationalise the behaviour of soil in respect of a physical or mechanical property that is relevant to a particular foundation engineering problem must therefore be accepted as an imperfect approximation.


In addition, there are many difficulties encountered in the reliable measurement of soil properties during site investigation. These include the inevitable disturbance in taking samples of soil out of their in situ state in the ground for laboratory testing, and the lack of precise control of the direction of soil sampling and in situ testing tools within a borehole. Even the basic process of defining the point where natural soil types change is often difficult; geological deposition environments do not usually change abruptly and therefore the change can be transitional with mixed soil types occurring between two clearly different soils in a sequence. Also, samples may not be taken at precisely the right level to permit the judgement of correct depth of soil changes during the exploratory boring, and there may be problems with errors in recording the precise depths of soil core samples and of the reference ground level. In recognition of these difficulties of characterising soils, geotechnical engineers use a process of averaging and taking the mean values of soil properties over depth intervals that are judged appropriate to the soil layering changes encountered. These soil layers are selected to suit the pile resistance prediction methods and may or may not coincide with geological layer boundaries. The best way of ensuring the most appropriate site investigation for a steel pile foundation is for the designer to inform the geotechnical engineer controlling the fieldwork and testing at the outset of his intention to consider steel piles. Appropriate soil testing can then be carried out both in situ and in the laboratory to provide the necessary type of data (see Case Study in Section 9.4.3). For instance, for the driveability predictions involved in steel piling, it is essential to perform in situ testing for reliable soil resistance calculations (see Case Study 9.4.3). Recommendations for soil sampling and testing for steel pile design are given in Section 3 of this publication, and in CIRIA Report 103[13]. Details of the tests are given in BS 1377[26].

5.2

Predictive methods – general

All pile design involves the prediction of soil resistance from soil tests at the initial stage. Major textbooks on foundation engineering and pile design discuss design methods for predicting vertical static pile load resistance for steel piles. However, only generic information is presented and this is often insufficiently detailed to permit economic or confident usage in design. Information on steel pile load test data and its interpretation has been published mainly in offshore pile design references.


36

Printed 07/09/05


It is evident that pile resistance prediction methods only make significant progress and become widely accepted as a result of well organised and well publicised pile load testing, as was the case with large diameter bored concrete piles in the 1960s. Tubular steel piles have always used for supporting offshore tubular steel frame production platforms, and the Offshore Oil and Gas Industry collectively funded full scale loading tests on 762 mm OD tubular piles in the 1960s and 1980s. This has progressed the knowledge of pile behaviour in both sands and clays. The research has included: •

Instrumented pile load tests in both granular and cohesive soils.

•

Improved soil investigation methods and tools for soil strength definition, e.g. Dutch Cone penetration Tests (CPT); Dilatometer Marchetti Test (DMT); push sampling; piston sampling; T-Z probe.

•

Analysis of pile behaviour and calibration of design prediction methods.

•

International pooling of knowledge and best practice.


Data from two major load test programmes on steel piles in UK clays was released into the public domain via a conference at the Institution of Civil Engineers in 1992 (ref. Clarke et al[1]). Load testing on sand sites in Holland, Belgium, France, Denmark, and in the UK has been interpreted and reported in 1996 by Imperial College (ref. Jardine et al[3]). This has resulted in a consensus amongst practising experts in the offshore industry as to the best soil resistance prediction methods to use for tubular steel piles. The methods have a sound theoretical basis, but the limitations of theory are recognised by including empirical adjustment factors. These methods are accepted in the offshore design codes in the USA, namely the American Petroleum Institute (API) Code for the Design and Construction of Offshore Structures (RP2A)[11]; in the UK also by BS 6235[37]; and by the draft ISO Code 13819[12] which is currently being developed. These methods are equally applicable to onshore tubular steel piles and are detailed in Sections 5.4 and 5.5. They have now also been validated by SCI against the results from a load test database[22] of H-piles and sheet piles. Collation and analysis of pile axial load tests on steel H-piles and sheet piles in both granular and cohesive soils in the UK has been carried out by the SCI in 1996 and 1997 in order to clarify the manner in which they can be applied and the level of confidence. It was found that the full area of the pile shaft should be used (i.e. 2D+4B for an H-pile as shown on Figure 5.1) and that ‘plugging’ is very rare and should not be assumed unless it is demonstrated to be developed during driving. The concepts or ‘models’ that can be used with confidence as a basis to calculate the pile load resistance from soil tests are as shown below in Figure 5.1. It should also be noted that the pile resistance prediction methods from soil tests (‘soil model methods’ according to BS EN 1997-1) are applicable to both compressive and tensile pile shaft resistance equally unless otherwise stated. The SCI validation work also included other predictive methods such as the SPT and CPT methods for granular soils, to evaluate their relative reliability. Interpretation of the database[29] forms the basis of the recommendations contained herein. P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

37

Printed 07/09/05


Unplugged H-pile

Plugged H-pile F

Ab

As

f s (i) f s (e)

qb R ult = fs A s + qb A b

R ult = fs A s + qb A b External wall only

Full cross sectional area 9c u in clays

Figure 5.1


5.3

Concepts for axial resistance

Pile axial movement models

As covered in Section 3.7, a designer needs to think in terms of both pile compressive strains and the relative movement of the steel pile wall against the soil which generates frictional soil resistance and thus transfers load from pile to soil. As the pile head load is increased, more of the pile is compressed and the loaded section extends further down the pile length until the pile base starts to feel strain, then the base moves down generating soil end bearing resistance as well. The basic models for axial pile load transfer curves are the ‘t-Z and Q-Z springs’ as used in the offshore design industry, see Figures 5.2 and 5.3. These have been derived for pile wall friction and end bearing by interpreting pile load test results given in the API RP2A Code[11]. Adaptation of the Q-Z curve is required for any steel piles that are not plugged, because of the smaller wall end-bearing area. A typical pile head load/vertical displacement curve is shown in Figure 5.4. There is a marked change in slope or ‘stiffness’ of the response curve at point P that occurs as the ‘stiffer’ load resistance of the wall friction becomes fully mobilised and the ‘softer’ end bearing resistance is added. At point Q, both the wall friction and the end bearing have been fully mobilised and the ultimate compressive load resistance of the pile is reached (the ultimate limit state in Eurocode terminology), after which the pile will continue to move down (‘fail’) under virtually constant load. Onshore full scale pile load testing on steel piles of various cross-sections on many sites has verified that a vertical pile head deflection of 7 to 10 mm is all that is required to generate the full predicted wall friction resistance along the length of most piles (see Clarke et al[1], Jardine et al[3] and Biddle[22]).


38

Printed 07/09/05


t max = f

1.0

t RES = 0.9 fmax

Clay

Range of t RES for clays

0.8

t/t max

t RES = 0.7 fmax

Sand 0.6 Clay: z/D

0.4


0

t/t

max

0.00 0.00 0.0016 0.30 0.0031 0.50 0.0057 0.75 0.0080 0.90 0.0100 1.00 0.0200 0.70 to 0.90 4 0.70 to 0.90

0.2

0 0

Sand:

0.01 0.1

z

t/t max

0.00 0.10

0.00 1.00 1.00

4

0.02

0.03

0.04

0.2

0.3

0.4

z/D

0.05 0.5

Pile head vertical movement,

z (inches)

D

=

pile diameter or pile base width assuming full plugging

t

=

wall friction resistance on pile shaft

Q

=

partial end-bearing resistance on pile cross section

Qp

=

maximum design end bearing resistance on full pile cross section

Z

=

pile head vertical displacement

Figure 5.2

Pile axial load transfer/movement curves, shaft friction (t-Z spring model) (American Petroleum Institute, Code RP2A)

Q/Q p= 1.0

z/D 0.002 0.013 0.042 0.073 0.100

t/t

max

0.25 0.50 0.75 0.90 1.00

z u= 0.10 x pile diameter (D) Pile head vertical movement, ratio z/D

Figure 5.3


Pile axial load transfer/movement curves, end bearing (Q Z spring model) (American Petroleum Institute, Code RP2A)

39

Printed 07/09/05


Pile head movement at the point of ultimate compressive load resistance is typically 30 to 50 mm, as shown on Figure 5.4. Permitted pile head movement in practice is restricted by the application of load and material factors in LSD and lumped factors in ASD (see Figure 3.1), and the settlement of the pile head will therefore be limited to less than 10 mm at its design resistance Rc;d when derived in accordance with Section 3.3. Test load - tonnes

Head settlement in mm

0 A 10

0

50

100

150 Loading

8 mm

200

250

220

233

P

B

20

30 C 40

Q

35 mm

D


Unloading 50

Figure 5.4

Typical pile head load-settlement curve

This magnitude of ‘settlement’ is well within structurally tolerable limits for most structures, especially since the major portion of vertical load for onshore structures is the deadweight of the structure and therefore occurs during the construction period and is confidently predictable if pile load tests are performed at the site. Steel piles that derive the major portion of their load resistance in end bearing are generally those driven into rocks or dense granular soils where the end bearing ‘Q-Z stiffness’ will be high and will dominate. In such piles it will be necessary to limit the design load resistance by consideration of local buckling near the tip. This can be determined from trial pile load tests at the point of departure from elastic behaviour of the Fc-δ curve (see Figure 3.2 and validation testing reported by Biddle[22]). Pile head movements in such ground conditions are small (of the order of 2 to 4 mm and generally less than 10 mm) and therefore well within structurally acceptable levels.


40

Printed 07/09/05


5.4

Axial resistance in non-cohesive, granular soils

5.4.1 Prediction from soil tests – API method Shaft friction

According to the offshore API RP2A Code[11], the ultimate frictional resistance on a tubular steel pile shaft in cohesionless, granular soils for each soil layer can be estimated using the formula: Rs:k

=

Ks poN tan (δ). Αs


where: Ks

=

coefficient of lateral earth pressure against pile shaft wall

po N

=

average effective overburden pressure over soil layer

Αs

=

Exposed area of pile shaft in the soil layer

φ′

=

average angle of internal friction within granular soil layer

δ

=

effective interaction friction angle between the pile wall and the soil in the soil layer, δ, which is equal to (φlaboratory – 5E).

The method has been investigated in an SCI Technical Report[22] and found to be valid for H-piles and sheet piles provided that the whole steel surface area is used and that there are sufficient laboratory soil tests to determine the φ′ values for each layer. Table 5.1 gives typical δ values for different types of granular soil. Table 5.1

Typical δ values for granular soils (taken from API RP2A)

Density

Soil description

Soil - pile friction angle δ (o)

Limiting skin friction values (kN/m2=kPa)

Nq

Limiting unit end bearing values (MN/m2=MPa)

Very loose Loose Medium

Sand Sand-Silt Silt

15

47.8

8

1.9

Loose Medium Dense

Sand Sand-Silt Silt

20

67

12

2.9

Medium Dense

Sand Sand-Silt

25

81.3

20

4.8

Dense Very dense

Sand Sand-Silt

30

95.7

40

9.6

Dense Very dense

Gravel Sand

35

114.8

50

12.0


41

Printed 07/09/05


Base resistance

Similarly, the formula given for the base resistance of tubular steel piles in the API RP2A Code[11], may be used for H-piles. The ultimate pile end bearing resistance in cohesionless soils can be estimated using the formula: Rb:k

poN Nq Ab

=

where: po N

=

effective overburden pressure at the base of the pile

Nq

=

Dimensionless bearing capacity factor from Figure 5.5

Ab

=

as defined in Figure 5.1.

Bearing capacity factors N q , N and Nγ c


1000 800 600 400 200 Nγ 100 80 60 40 20 Nc 10 8 6 4 Nq

2

1

0

10

20

30

Soil friction angle,

Figure 5.5

40

φ'

50

(deg.)

Bearing capacity factors (according to API RP2A[11])

5.4.2 Prediction from soil tests – SPT method Use of the in situ driven SPT test for assessing granular soil properties is common practice within the UK and US piling industries and worldwide. Reference to both SPT and to CPT testing is made in Section 7.5.3. of BS 8004[15] and these should be used as a basis for predicting the load resistance for H-piles.


42

Printed 07/09/05


A detailed explanation of the methods of SPT testing, reporting and calculation methods is available in CIRIA Report 143 The Standard Penetration Test (SPT): Methods and Use[38]. The SPT test is conducted ‘down the hole’ as the borehole is excavated. The test is described in BS 1377-9[26]. An open-ended split barrel sample tube mounted on a string of screwed steel rods is driven into the bottom of the borehole by a 63.5 kg weight falling through a fixed height of 760 mm. The number of blows (N) required to drive the tube a distance of 300 mm is recorded, after its penetration under gravity and a ‘seating’ drive of 150 mm. Usually the blow count is recorded as 6 values, one for each consecutive 75 mm. Where driving is hard and a minimum penetration of 300 mm cannot be achieved without risking severe damage to the rods, an alternative method records the set for 50 blows of the hammer, after the initial seating drive of 25 blows. Since, in the full test, the number of blows required to drive the tube through a set of 300 mm is required, an SPT ‘N’ value has to be extrapolated from the smaller penetration in the alternative dataset. The validation work for an SPT method[22][29] used the formula given in the earlier Steel bearing piles[10] publication to predict load resistance as follows: Created on 04 July 2011 This material is copyright - all rights reserved. Use of this document is subject to the terms and conditions of the Steelbiz Licence Agreement

Ultimate resistance Rc;k = fs As + qb Ab where fs = 2N (kPa), and qb = 400N (kPa) These formulae have been in common use for 40 to 50 years and have proved to be appropriate for steel piling load capacity predictions. The phenomenon of ‘plugging’ did not occur on the 16 pile load test sites for which SPT data was available in the SCI database, and it was therefore necessary to adjust the As and Ab values from those suggested by Cornfield in Steel bearing piles[10] to be the total exposed pile surface area for As and the steel wall end area only for Ab (see Figure 5.1). In accordance with Cornfield’s recommendations, where the soil was submerged below the ground water table, the values of N were factored by 0.67. CIRIA Report 143[38] makes no reference to any factoring required to the N value measured in submerged soils, and given that these generally seem to account for the greater part of the pile length in the UK, this could be a very significant error. However, the average value of the constants given in CIRIA Report 143 is very similar to the constants suggested by Cornfield when the 0.67 factor is applied to the constants rather than the N value. This would suggest that the two methods are essentially analogous in submerged soils.

5.4.3 Prediction from soil tests – CPT method The Cone Penetration Test (CPT) formula for pile capacity prediction uses an in situ test measurement of soil resistance to penetration using a penetrometer with a 60E cone end and a friction sleeve. An instrumented device known as the Static Cone Penetrometer or Dutch Cone, as described in BS 1377-9: In-situ tests[26], is pushed into the soil at a constant rate of 2 cm/sec by a hydraulic jacking system. A load cell mounted just P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

43

Printed 07/09/05


behind the cone continuously monitors the tip load qc being applied to ‘fail’ the soil and thereby to advance the cone. A sleeve section or ‘mantle’ just behind the cone is also strain-gauged and allows direct measurement of the cone ‘shaft’ friction fs. A diagram of a typical 10 cm2 cross-section static cone penetrometer tip is shown in Figure 5.6.

Dirt seal

Seals


1000 mm

Friction sleeve

Watertight seal Dirt seal

60°

35.7 mm

Figure 5.6

A static cone penetrometer tip

Shaft friction

Several empirical formulae have been developed to relate the CPT qc value to foundation design parameters depending on soil type and loading regime (see Cone penetration testing: Methods and interpretation[40]). It has been found that for the more consistent fine sands and silts encountered in Holland, the following formulae can be used in shaft capacity prediction for steel piles: Compressive resistance:

Unit shaft friction =

qc 300

Tensile resistance:

Unit shaft friction =

qc 400

The formulae are not valid in other types of granular soil.


44

Printed 07/09/05


These formulae use the measured cone resistance qc to estimate the shaft friction; however, where cone friction sleeve values fs are also available then the following formula has been derived (see Reference 40): Unit shaft friction = 0.35 fs It should be noted that alternative methods for the prediction of load resistance on offshore steel tubular piles have been developed by Imperial College, London for both cohesive and granular soils. They have been reported in Jardine et al[3][112] and proposed for application to steel H-pile sections using CPT test data. End-bearing


For plugged steel pile sections, the ultimate unit end-bearing is estimated from the cone values after applying Schmertmann’s averaging process for qc values above and below the pile toe, and then adopting limits (see Reference 40). This will produce a conservative prediction of pile capacity that will limit pile settlement. However where plugging is suspected, static pile load testing should always be carried out to confirm the design assumptions. For unplugged steel pile sections, the qc value at any level can be used directly without modification as the ultimate unit endbearing resistance pressure under the steel wall area. This resistance also applies to the tip of the pile during driving.

5.4.4 Conclusions For predicting the shaft friction in granular soils, the SCI validation work[22][29] indicates that either the SPT or CPT method can be used with confidence for fine granular soils and the SPT method for all granular type soils using the total exposed shaft surface area (see Sections 5.4.2 and 5.4.3). To predict the ultimate base resistance on driven steel piles, only the cross-sectional area of the wall should be used together with a unit ultimate bearing pressure equal to the CPT value (as verified by Jardine et al[3]). If CPT values are not available then correlations can be used to deduce an equivalent value from correlations with the SPT. The offshore pseudo-effective stress method for predicting shaft friction in granular soils (see Section 5.4.1) was generally not as reliable as the SPT method in the SCI database because of the absence of measurement of φ′ in routine laboratory soil investigation testing.


45

Printed 07/09/05


5.5

Axial resistance in cohesive soils

5.5.1 Prediction of shaft friction from soil tests According to the offshore API RP2A Code[11], the ultimate frictional resistance on a tubular steel pile shaft in cohesive soils for each soil layer can be estimated using the formula: Rs;k

=

αcu As

where:

α

=

the pile wall adhesion factor (or soil shear strength modification factor) selected for each soil layer

cu

=

average undrained triaxial test (UU) shear strength over the depth interval (layer thickness) being considered

As

=

exposed area of pile shaft in the soil layer (both external and internal pile surface area unless the tube is plugged).

Guidance on the value of the empirical factor α comes from interpretation of a database of over 200 tubular steel pile load tests, ref. Briaud et al[41]. It has been found that reasonable estimates of ultimate wall friction, Rs;k, for steel tubular piles can be obtained by using the values for α of α = 1.0 for cu = 0 to 24 kpa, and α = 0.5 for cu >72 kpa, with a linear relationship in between. These values are interpreted from the API database for steel tubular piles, as shown in the graph in Figure 5.7. Key API test data W. Sole tests (1968 site investigation) W. Sole tests (1978 site investigation) LDPT data

1.25 Friction coefficient, α


(Note: α varies with cu and σv′, see Figure 5.7 for tubular piles)

1.00

0.75 Pentre API RP2A recommended design line

0.50

Tilbrook 0.25

0

0

Figure 5.7

100

200

300 400 500 Undrained shear strength, C u (kPa)

600

API pile load test database for α factor versus cu for clays

The validation work carried out by SCI[22][29] investigated whether the API RP2A method for clays can be used for H-piles and sheet pile sections. For cohesive soils, the shaft friction measured in load tests in the SCI database was found to be adequately predicted by using α =0.25 as a mean value in conjunction with using the average undrained triaxial compression test UU shear strength cu of each clay layer. This is at the lower end of the range for α that is given in BS 8004 and therefore confirms that its adoption as a general rule for design predictions is conservative.


46

Printed 07/09/05


It should be noted that the total surface area of the pile was used in these prediction calculations without assuming any plugging for the H-piles. It was noted that pile load tests are routinely carried out within 2 to 3 weeks of driving the trial piles. In most clays, this would not be sufficient time for dissipation of all the excess porewater pressure set-up during driving of these displacement piles. Therefore the real ultimate capacity had not been measured at that time. The UK instrumented tests on steel piles in overconsolidated clays[1] have established that after a period of weeks or months, a mean α factor of 0.5 is achieved. This confirms the same value of α = 0.5 as is recommended in Section 7.5.3 of BS 8004[15] for bored piles constructed in overconsolidated clays where softening due to exposure is not allowed to occur, i.e. the clay is undisturbed and has no excess pore water pressure. A general confidence level of ± 25% was concluded as being appropriate to cover the inevitable scatter of prediction results around the mean regression line through the SCI dataset[22].

5.5.2 Prediction of base resistance from soil tests The formula given for the base resistance of tubular steel piles in the API RP2A Code[11] may be used for H-piles. The ultimate pile end bearing resistance in cohesive soils can be estimated using the formula:


Rb;k

=

9 cu Αb

where: cu

=

average undrained triaxial test (UU) shear strength of the clay at the pile base (within a depth of 1.5D below the tip level)

Αb

=

full cross sectional area of pile base for plugged steel piles comprising the pile wall and any soil plug; or

=

steel wall cross sectional area for unplugged steel piles.

For steel piles, the calculated potential ultimate pile end bearing across the whole cross-section is compared with the internal soil plug skin friction plus the pile wall end bearing, and the lesser of the two values is taken as the best prediction. Resistance from CPT tests

Compressive and tensile resistance: Unit shaft friction

=

α cu

≡ α

qc NK

where in general for overconsolidated clays, NK = 15 to 20, and cu is the undrained triaxial shear strength of the clay. It has been found in correlation work for clay sites that the conversion factor NK varies considerably with the clay type and its preconsolidation history. Hence the CPT formula is not used on its own for pile capacity predictions in clay. It is used only as an ancillary to check the UU undrained triaxial test cu profile for each clay stratum and thereby to aid judgement of the most representative mean value interpretations in arriving at a design profile.


47

Printed 07/09/05


5.6

Axial resistance in rock

5.6.1 Geotechnical aspects Where the contribution to load resistance in the pile comes almost entirely from pile end bearing in a hard stratum such as a rock layer, the method of geotechnical design will be different to that for soils. CIRIA R181[21] recommends steel H-piles as ideally suited to piling in rock because of their high strength and small displacement that enables them to be driven to virtual refusal and to penetrate well into variable weathered rock surfaces. Unlike bored concrete piles, the presence of groundwater presents no problems and design does not depend on preparing a rock socket.


Prediction of pile ultimate capacity in rock was discussed in Section 3.3.3 with further advice in Section 4.4. Design will depend on the type of rock and its hardness. To generate the maximum potential end bearing per pile in sound rock, generally requires the selection of a cross-section that gives adequate resistance to local buckling because capacity is limited by the steel yield stress. Pile head deflection will not be the overriding criterion because the pile movement required to mobilise base resistance in the rock will only be the axial elastic compression of the complete pile length. Selection of section and any toe strengthening will also be affected by the pile stiffness required to permit any hard driving that is necessary to obtain penetration into the buried rock surface and to confirm development of end bearing. However, this should be achieved without damaging the pile and therefore the pile installation planning must include the prediction of driving stresses. Wave Equation computer models such as GRLWEAP[27] (see Sections 8.2 and 8.3) are available for such analysis. The input for end-bearing resistance should be realistically assessed from in situ penetration tests and/or uniaxial tests on rock cores or reference to driving tests in similar rock in the area. During the site investigation, static CPTs may reach their maximum permissible load without failing the rock or achieving significant penetration into it. This is because the maximum permissible pressure for CPT equipment load cells is only about 70 MPa and yet intact rock strengths are often in excess of 300 MPa (see the Hong Kong Geoguide[42] and the SCI pile load test databases[22]). Typical strengths of different types of hard rock from laboratory compression tests determined by the Norwegian Geotechnical Institute are shown in Figure 5.8. Such sites therefore require a well thought out soil investigation programme with in situ penetration testing for driveability prediction, and good core samples for uniaxial laboratory testing or deep plate bearing tests in the rock within the boreholes for estimation of endbearing resistance, especially if it is significantly weathered and has weak layers. The recommendations given in the CIRIA R181[21] should be followed for good practice.


48

Printed 07/09/05


Rock type

Compressive strength

Hard Weak

100

Leptite Diabase Basalt Granite Syenite Quartz porphyry Diorite, gabbro Quartzite Quartzitic phyllite Metamorphic phyllite Layered phyllite Hornblende Chalkstone Marble Dolerite Oil shale Mica shist Sandstone Lava


Key :

Figure 5.8

200

300

400

Variation about mean Cylinder sample H = D Cylinder sample H = 2D Cube test

Typical rock compressive strengths determined by NGI

Most of the problems of design of piles in UK rocks stem from the inability to obtain good samples in the weathered upper layers, and no amount of applied geological knowledge or accurate description can predict rock resistance design values. Accurate data will only come from trial pile tests at the site as recommended in Section 7.5.1.(1)P of BS EN 1997-1[9] (see also CIRIA R181[21]). Some degree of reliable estimation is needed even to specify the pile load testing procedure to size the piles required, and to design the static load testing frame to the likely magnitude of pile loads involved. Experience shows that this task can require considerable specialist expertise and local knowledge of previous piling in the area. Further guidance is therefore given here for the designer, based on back-analysis of pile load tests on steel piles by SCI and others. Pile testing load reaction frames need to be of a high rating where steel piles bed into hard rock if the ultimate pile capacity is to be achieved. The SCI database[22] records a case at Shaldon Bridge in Devon where a unit wall end bearing of 160 to 200 MPa was measured in a Breccia rock, and ultimate loads of 350 tonnes were recorded on Larssen 25W sheet piles and 356×368×174 kg/m H-piles. Where hard rock is encountered, it is therefore wise to use a load test reaction that permits the yield strength of the steel section to be achieved, and to size the structural reaction frame accordingly. Driven steel piles permit a good estimation of static pile capacity to be obtained from dynamic pile testing during driving. This is a useful first stage that will either be sufficient in itself, or then permit the sizing of a static load test frame to be carried out with confidence. Sufficient time must be allowed in the test pile programme to permit this two-stage process that will ultimately result in


49

Printed 07/09/05


confident data from which the number and length of steel piles can be minimised, and thereby savings to be made in foundation costs. For weak rocks in UK, design estimates of pile load resistance from sample tests have been very unreliable. Many of the empirical design rules have been derived from inappropriate analysis of data from poorly specified pile tests in an era when economy through design competition was not the prime motivating factor. Efficient foundation design can only be achieved through rigorous application of the basic tenets of good site knowledge, thorough research of subsoil in relation to alternative pile construction methods, experienced judgement of risk and cost implications of ‘cutting corners’, and appropriate design to an informed specification. BS 8004: Foundations

The advice given in BS 8004[15] is out of date and is too conservative for economic design of steel piles.


For instance, for prediction of load resistance in rock, end bearing values of 2 to 4 MPa are suggested in BS 8004, whereas end-bearing values between 28 MPa and 58 MPa have been measured on steel test piles in weak weathered mudstones and sandstones[22] in the UK. A geological classification system for chalk materials together with presumed bearing values is given on page 13 of BS 8004, and the text refers to research by Hobbs and Healy[43]. Later research is contained in CIRIA PR 11 Foundations in chalk[44]. The nature of chalk varies significantly across the UK, and the designer should recognise the need to classify chalks in a geological and geographical context, before extrapolating the results of pile load tests to other sites. The hardest chalk is credited in BS 8004 with a bearing value of only 1 MPa to 1.5 MPa. However, SCI research[22] indicates that at Erith in Kent (Location 16) in the Upper Chalk, some 2 to 3 times this value can be obtained by driven steel piles in end bearing on the wall tip area alone. The results at Location 19 in Lincolnshire in the Lower Chalk, indicate that where the chalk is intact and unweathered, even higher values, up to at least 15 MPa, are more realistic. BS 8004 also gives detailed guidance for piles driven into Keuper Marl or the Mercia Mudstone series as it has become known. Data on a series of load tests on H-piles in Keuper Marl[22] indicates that often the shaft friction is best determined by treating all the weathered layers as a ‘granular material’ and using the API RP2A Code[11] formula (see section 5.4) applied to the total surface area of the pile. It is often misleading to treat the Marl as a ‘cohesive’ material for driven piles, due to the presence of numerous fragments and layers of remnant rock in the matrix that break up around the penetrating pile, thus creating a new granular material. The prediction method used must be appropriate to the type of pile and to the soil type and this requires the judgement of an experienced practitioner in each material.

5.6.2 Structural and driving aspects Steel piles to rock should be driven to virtual refusal, with the refusal criterion agreed between the piling specialist subcontractor and the designer such as to limit the driving stresses to not more than 90% of yield strength in accordance with BS EN 12699[70]. The driving stress can be reliably predicted using wave equation programs such as GRLWEAP[27] and checked during driving using


50

Printed 07/09/05


DPA (see below). A guide in this respect is that piles can be driven until the blowcount is at about 10 blows per 20 mm or part thereof, or the top of the pile starts to deform plastically, whichever occurs first. In Norway, where piling into hard rock is an everyday construction problem, they use criteria of 10 blows per 1 to 3 mm to define pile ‘refusal’ provided that the blowcount rate has an increasing tendency.


It is always worthwhile to use dynamic pile driving analysis at the end of driving steel piles in rock, due to the high degree of uncertainty in the theoretical prediction methods for rock resistance. A specialist company can provide a Dynamic Pile Analyser (DPA) that uses measured acceleration and stress wave recordings from pile instrumentation under a blow of the hammer to interpret the distribution of soil resistance down the pile using the CAPWAP[69] analysis. Such measurements have been correlated against static load test records in the SCI validation work for this publication and found to be a reliable estimate of end resistance in rock strata[22]. The use of a DPA also provides direct monitoring of the driving stresses in the pile so that termination of driving can be called when 90% of the yield stress of the steel is approached to comply with BS EN 12699[70]. This advance in technology obviates the use of conservative arbitrary rules, such as are found in BS 6349[6], that were prudent limits in earlier times to provide against the risk of buckling damage during driving of steel piles in the days when neither routine instrumentation nor wave equation analysis were available to the design engineer. Experience indicates that there is generally no risk of overall buckling of steel piles when embedded in soil. Theoretical methods such as ‘Euler Buckling Theory’ are not relevant, because of the considerable lateral restraint offered by the soil to the pile shaft (see BS 6349). The only relevant check is that necessary to prevent local buckling at the pile tip in rock that can be sustained during the driving stage. For example, at the Shaldon Bridge site in Devon[22], four trial piles were extracted and two were found to have toe damage although they were all driven to the same set. If driving stresses in a pile are controlled to within 90% of the steel yield stress using DPA in accordance with BS EN 12699[70] then such damage should be prevented. Driving stresses can be reliably predicted using a modern Wave Equation pile driveability computer analysis programme like GRLWEAP[27], provided that the end bearing resistance force is correctly input by experienced practitioners. The latter can be ensured if CPT qc values are used, as explained earlier. This should be sufficient to ensure selection of a suitable steel pile section for driving. After load testing, test piles can be extracted using a vibratory hammer to check the condition of the pile tip. A change of driving procedure or a change of steel section may be necessary for the working piles if damage has occurred. In addition to the high end-bearing, dynamic pile load testing on steel piles shows that in the lowest section of pile, the wall friction is also very high, particularly in rock and dense granular materials. This cannot yet be predicted reliably using laboratory testing or theoretical models for design and therefore test piles are essential. It is therefore prudent to try to cater for the unknown pile resistance in rock when specifying test procedure with a range of cross-sections or possibly


51

Printed 07/09/05


different types of steel pile to determine the best and most economic solution for each project. For hard driving, it may be advantageous to use a higher steel grade than the minimum required for taking the static design load. The benefit of this provision can be realised by reference to the table in BS EN 1997-1[9] of reducing correlation factors that are allowed for an increased number of pile tests (see Table 3.1 in Section 3.3).

5.7

Negative shaft friction


Where negative wall friction or ‘downdrag’ is likely to occur due to settlement of the adjacent soil relative to the project piles during their working life, it must be allowed for in design. The ‘downdrag’ causes a downward shaft friction on the pile in the settling soil layer and a reduction in the available upward shaft friction that is resisting the structural load on the pile; this results in a longer pile being required. The settlement could originate from ground surface loading adjacent to foundation piles, such as road embankments, or from natural consolidation of recent soils or fills, and is a commonly experienced phenomenon. Methods of design for ‘downdrag’ are explained in Example 7.4 in the publication Designers Guide to EN 1997-1[113]. Essentially, the designer can select one of two methods to find the downdrag force due to the soil settling around the pile. One of these would be to use a soil-structure interaction type analysis to determine the net relative soil movement adjacent to the pile (i.e.
5.8

Measures to increase steel pile axial capacity

The maximum potential axial capacity of steel piles is currently rarely used. To achieve greater economy in the number of piles and the size of steel piles: •

Avoid conservative prediction methods for pile resistance.

•

Improve pile movement predictions.

•

Specify estimated failure loads for test piles not working loads.

•

Consider both steel and concrete pile types that are suitable to the soil profile.

•

Provide for alternative pile types in the preliminary foundation design.

•

Discuss estimated pile design loads between designer and contractors.


52

Printed 07/09/05


•

Be prepared to modify the foundation design and structure/pile connection after pile tests.


As laid down in the guidance section of the ICE Specification for piling and embedded retaining walls[20], economies in the number and size of steel bearing piles required in any foundation can be achieved by using one or other of the following methods for pile testing: •

Carry out trial pile tests to ultimate failure or incipient failure, by specifying a head deflection of a minimum of 40 mm in soils, or to the full allowable steel section load when end-bearing in rock, and then modifying the number of piles to suit.

•

In pile tests, use specialist geotechnical judgement to specify the delay to the start of pile load application to suit the set-up period appropriate to the soil type, so that more of the potential capacity has recovered after driving and before testing.

•

Judge the proportion of resistance attributable to shaft friction and endbearing and then ensure that the steel area in contact with the major contributor is maximised (for example, weld on wing plates to a skin friction pile or use enlarged shoes for an end bearing pile, see Tomlinson et al[106]).

•

Use Maintained increment Load Tests (MLT) not Constant Rate of Penetration Testing (CRP), so that: the load-deflection curve is definitive; the pile reaches an equilibrium between applied load and resistance; and the pile load resistance versus pile head deflection relationship is therefore interpretable.


53

Printed 07/09/05


6

LATERAL LOAD RESISTANCE

6.1

Introduction

Lateral loads on bearing piles range in importance from the major load component in such structures as transmission towers or mooring dolphins, to a relatively insignificant force in the foundations of low-rise buildings. The designer must first judge whether the imposed lateral load on the proposed foundation is significant enough to warrant special analysis. On buildings for instance, the lateral loading is mainly due to wind pressures. For low-rise buildings not exceeding three storeys in height, any foundation shear resistance required is normally accommodated by passive earth pressure acting on the buried pile caps and ground beams of a piled foundation, and on the frictional sliding resistance beneath the ground floor slab and the foundations. Therefore, the lateral loading on the bearing piles would be insignificant.


Advice on the lateral loading of bridge abutment piles from adjacent embankments and appropriate design is covered in the Highways Agency’s design standard BD 74 Foundations[104]. Specific research studies and monitoring of the effects of such lateral loading are referenced there. Lateral load resistance from vertical bearing piles is particularly dependent on soil type, but in most UK soils, it is adequate to resist the braking loads from traffic and to absorb thermal expansion from decks of integral bridges (see SCI publication P250[18] ). In the extreme case of very soft soil, raking piles to an underlying more competent soil or rock would be required to provide significant resistance. Lateral load resistance from vertical bearing piles also requires significant pile displacement, of the order of many centimetres, as evidenced in the lateral pile load tests for the integral bridge at Whaddon Road, Milton Keynes[103]. If such movements are unacceptable then the designer should think of alternative foundation elements to provide it, such as raking piles or embedded sheet pile perimeter walls. Where the contribution to lateral loading resistance of vertical bearing piles is vital and an acceptable solution, the designer is recommended to use CIRIA Report 103 Design of laterally loaded piles[13] and the textbooks by Poulos and Davis[23], and Tomlinson[24]. The design of piles for lateral loading according to LSD rules from BS EN 1997-1[9] is covered by the SLS requirement, because it is governed by pile movement for which there will be limits set by the designer for the structure design. For SLS, as covered in Section 3.3.5, all partial factors are set to 1.0 thus enabling the geotechnical models to use actual unfactored loads from the structure to determine more realistic estimates of pile movement.


54

Printed 07/09/05


6.2

Methods of analysis

The two most extensively validated methods are the P-Y curve method and elastic continuum analysis FE programs. Both are explained in CIRIA Report 103[13].


P-Y curves originate from instrumented lateral load tests carried out on 762 mm OD tubular piles in the USA in the 1960s for offshore design. The models of load resistance were derived from soil resistance distributions required to match the bending stresses measured by the pile shaft strain gauge instrumentation, i.e. curve-fitting to match bending moment diagrams. The P-Y curve method is the only one in which it is possible to allow for significant cyclic loading of piles. This is useful for the structural design of the pile section, but does not give accurate displacements because the single piles had no head restraint. It is explained in detail in the American Petroleum Institute Code RP2A[11] and in computer programs like ‘ALP’ in the OASYS suite[45]. The generic models for P-Y curves for sands and clays are shown in Figures 6.1 and 6.2, where P is the lateral load on the pile and Y is the lateral displacement from vertical. They are lateral resistance springs relating load resistance to deflection for a discrete layer of soil. The curve is dependent on the diameter or width of the pile but not on the shape or stiffness of the pile. Three model types have been derived from pile lateral load tests on 762 mm tubular piles for soft and stiff clay and for sand. The Pu or ultimate lateral resistance force at any level is related to soil strength properties.

Curve defined by p/p u =0.5 ³ √ y/y c

P

p a =0.72p u

Ultimate resistance for static loading

a

p a =p u b

x=

0

x ≥ xr Ultimate resistance for cyclic loading x
r

p = p b(x/x r ) y=3yc y=8yc

Figure 6.1


y=15yc

Y

P-Y curve for clays (according to API RP2A)

55

Printed 07/09/05


P pu =Ap pm=Bp

u

c m

c

1

m

Curve defined by p =Cy1/n

k

y k =(C/n h x)n/n-1 y m=B/60

Figure 6.2

x = x1

y u =3B/80

P-Y curve for sands (according to API RP2A)


The capabilities of the various methods of analysis are summarised in Table 6.1, reproduced from CIRIA Report 103[13]. Table 6.1

Summary of the output of methods of analysis (from CIRIA Report 103)

Model

Limitations

Structural frame

Unrealistic model (the End-bearing pile groups, soil is ignored). with a small lateral load component (say up to 10% of the vertical load).

Axial load on piles is the only reasonable output.

Winkler medium or p-y analysis

A reasonable model for single piles. However, inappropriate for pile groups with s/D<8, because the continuity of the soil is not modelled.

Depth, slope, moment and shear of the pile at any depth.

Elastic continuum

Single piles or pile groups A reasonable model for single piles or pile under working loads. groups at working load. Yield of the soil cannot be included exactly. The limitations depend on the mathematics of the particular computer solution chosen. Available programs are limited to constant or linearly increasing soil modulus with depth.


Application

56

Any laterally-loaded single pile or widely-spaced pile (s/D>3) group. The analysis can provide reasonable predictions for cyclic loading or account for the development of plastic zones if suitable p-y data are selected.

Output

Output depends on the particular program adopted, but typically includes deflection, slope, moment, shear and axial load distribution for each pile in the group, and the overall stiffness and/or flexibility matrix of the pile group.

Printed 07/09/05


For structures like integral bridges, it may be necessary to carry out a soil/structure interaction analysis using appropriate stiffnesses to calculate the movement that occurs at the pile head (see SCI publication P250[18]). This could be a total global frame analysis in 2D for a typical frame of a symmetrical structure, or in 3D if the structure is asymmetric. An iterative procedure will be required to obtain a match of displacement and rotation at the pile head if the structure and foundation are modelled separately. This has been common practice in offshore platform design. A ‘P-Y’ or elastic continuum analysis can be used.

6.3

Assessment of soil properties


There are no clear guidelines for the assessment of soil property values to be adopted for the design of laterally-loaded piles. Many factors influence the actual mobilised values. In particular, the disturbing effect of pile installation is difficult to quantify. Most soils exhibit considerable loss of stiffness under the action of cyclic loading, which is virtually impossible to relate to laboratory test data. Generally, it is desirable to select upper and lower limits to the critical soil properties, and to check the sensitivity of the design to variation within the chosen range. For the realistic design of laterally-loaded piles or pile groups, the upper layers of soil should be thoroughly investigated, because these tend to be particularly variable and strongly affect lateral pile behaviour. In the case of single piles, the stiffness at 50% ultimate strength( E50) of the soil to a depth of only a few pile diameters has a dominant influence on the behaviour of the pile. The most obvious and satisfactory procedure for establishing the response of laterally-loaded piles is by load testing a pile. Many steel pile lateral load tests have been carried out by research institutions and Joint Industry Projects (JIPs), particularly for the offshore oil and gas construction industry. Some of these tests are reported in Offshore Technology Conference papers[46][47][48]. If lateral loads are high, or the structure is subjected to significant cyclic loading, it is strongly recommended that pile tests should be undertaken to confirm the design assumptions (see Reference 103). However for design purposes, the engineer is often faced with the task of selecting suitable parameters from limited site investigation data. The significance of changing the value of a particular soil property will depend on the method of analysis selected for predicting the response of the laterally loaded pile. Most methods of analysis use convenient approximate representations of soil resistance that have been derived from back analysis of load test data. The procedures used to do this were generally iterative curve-fitting to match the measured pile bending stresses and deflections and rotations of the test piles using selected relevant mechanical properties of the adjacent soils. As can be seen in the OTC papers, such curve-fitting was approximate and an exact match was never achieved from the methods used. However, the objectives of the design exercise are to predict the pile deflection and bending moment profile to a sufficient accuracy that permits a confident judgement to be made on the required pile dimensions that are needed to achieve the desired structural response. Experience shows that the two available methods of ‘P-Y’ analysis and ‘elastic continuum’ permit this to be achieved within the framework of applicability stated in Table 5.1. Any attempt to further refine these design methods would involve an extensive programme of


57

Printed 07/09/05



lateral load tests for each type of steel pile to systematically examine: the effects of varying soil type, soil stiffness, elastic-plastic response of soils to lateral loading, soil/pile stiffness ratios, pile slenderness effects, etc. It is doubtful whether such a programme could ever be economically justified when the approximate models that are available do the job acceptably well. Any designer of laterally loaded piles should carry out sensitivity studies of the effects of varying the soil parameters involved in a chosen method of analysis. In general it will be found that the maximum bending stress in a pile will be most dependent on the degree of structural fixity applied to the pile by the connection to the structure. Provided that the correct class of soil is determined to suit the recommendations given in Table 2 of CIRIA Report 103[13], then changing the value of the soil parameter (e.g. Young’s Modulus E; soil modulus G; or coefficient of subgrade reaction K) will not affect the predicted pile effect to a significant degree, unless the change is of an order of magnitude. This permits a wide tolerance in the value of soil properties and their methods of determination. The geotechnical engineers involved in the soil testing should therefore have a good knowledge of this tolerance in order to judge the type and number of soil tests that are necessary to achieve an appropriate characterisation of each soil layer. Structural designers would be wise to specify their design requirements for soils data carefully within a site investigation contract, so that the work includes soil tests appropriate to the type of loading to be applied to the pile. For example, if granular soils are found on site then in situ soil testing, such as CPTs, will be necessary to derive sensible values for lateral loading analysis parameters.

6.4

Combined lateral and vertical loading

Marine structures and integral bridges involve the design of foundations that can resist both vertical and lateral loading. In geotechnical terms this needs to be translated into vertical and lateral structural movement, because soil will only develop resistance to an applied load or displacement from a structure if there is movement at the soil/pile interface. It is well known from pile load testing on driven steel piles that the soil/pile load transfer for shaft friction is governed by a thin remoulded layer of soil immediately adjacent to the pile wall that has already been developed during driving of the steel pile[33]. The full shaft friction is developed after only 7 to 10 mm of local ‘slip’ at this interface and thereafter the friction remains constant. It is therefore reasonable to expect that the magnitude of the total frictional resistance at the wall due to axial loading and due to the development of lateral earth pressure is subject to a finite limit. This limit can be quantified by reference to the SCI pile load test database[22] which illustrates that there are different relationships for clays or sands. Where a foundation moves laterally in a cyclical manner, as in integral bridge abutments, a gap can develop behind the wall to separate the wall from the soil if the soil compacts permanently, this is often called “post-holing”. In view of this possibility, it would be advisable to rely on just the vertical load resistance of the portion of the steel piles that are embedded on both sides, below the cutting or excavation formation level; this has been accepted practice for many years (Figure 6.3). A further and more conservative assumption is to use only one face of the SSP, that is involved in the ‘passive wedge’ portion of the ULS stability calculation, i.e. in the limit equilibrium analysis (Figure 6.4). The P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

58

Printed 07/09/05


latter approach has been used in Integral steel bridges: design of a single-span bridge - Worked Example[32] here the vertical load resistance was found not to be the governing factor. W

Active soil zone moving downwards

Passive soil zone moving upwards

Figure 6.3

Generation of wall-soil friction by pile movement


W

Excavation level

Assumed length of wall providing wall friction resistance

Pile tip

Figure 6.4

Length of sheet pile contributing to wall friction

Steel piling has been used in combined loading situations such as highway bridge abutments for several decades and there has not been any sign of excessive movements or unsatisfactory performance. Therefore, the current design procedures as described herein can be concluded to be satisfactory.


59

Printed 07/09/05


7

PILE GROUP EFFECTS

7.1

Conceptual design axial load resistance

The first stage of a conceptual design is to estimate the number of piles required in the group using the predicted individual axial design resistance from the procedures given in Section 5. These piles will then be laid out in a group within the plan area of the supported structure. Care should be taken to use the characteristic pile resistance because the pile group analysis uses pile deflection and settlement to determine the contribution of each pile to the overall resistance provided by the group to an applied load. However, depending on the relative positions of the piles there may be a group effect that will modify the resistance of each pile.

7.1.1 Assessing pile group effect


Piles in a group may be subject to the following effects: i)

The axial resistance of a group of piles could be less than the sum of the resistances of all the piles in the group acting independently.

ii)

The pile head deflection of the group, or of its pile cap, may be different to that of a single pile.

There are some simple spacing rules that have been derived from experiment and experience, and piles will only interact to cause a ‘group effect’ if they are closer than a particular spacing to each other. This is about 3D (where D = pile diameter) for piles in clay or about 4D if the piles are in sands. When they are closer than those limits, the pile group behaves as a single block with shaft friction around its external periphery and a base resistance over the whole area of the block, because the individual soil resistance ‘envelopes’ overlap. Therefore, in order to achieve best performance and economy of a pile group, the designer should adjust the layout and spacing to comply with these criteria and to ensure that piles act independently wherever possible. If this can be achieved, then no pile group effect will occur and the vertical load resistance of the group is the sum of the individual piles, and the vertical deflection at the pile head is the same as that of an individual pile under the share of the load. It is worth adjusting the size of each pile, removing any conservatism in the assumptions on soil parameters involved in pile load resistance prediction, or waiting for the results of the trial pile load testing, before deciding on a pile group configuration that involves a pile group effect. Where such adjustment of the spacing and arrangement of the piles in a group still infringes the spacing limits, the pile group may have to be designed as the enclosing large block.

7.1.2 Practical considerations As a first approximation, the piles in the group should be arranged to resist the applied loading from a structural point of view (i.e. the centre of action of the pile group should lie near the resultant thrusts of the various load cases). Having outlined a trial pile group, detailed analysis may be carried out to refine the design.


60

Printed 07/09/05


Practical constraints on positioning and aligning piles within a group (particularly raking piles) should also be considered at the design stage. Heavily-raked piles are difficult to control, and they may depart from the intended line by a considerable amount. Misalignment of piles within a group results in higher bending stresses. An analysis of a real pile group by O’Neill, Ghazzaly and Ha[49] indicated a 30% increase in bending moment over that computed for the designed geometry, together with substantial moments on the minor axis of the piles. Raking piles to resist lateral loads should therefore not be ‘battered’ excessively. Practical limits are often considered to be 1 in 3 for driven steel piles using hydraulic hammer rigs and 1 in 2 for hanging leader frames using drop hammers. Although batters up to 45E are theoretically possible, loss of control of angle during driving often results when these steeper batters are attempted.


While it is possible to design vertical piles to resist large lateral forces, high lateral movements are required and it may be more structurally sensible to include some raking piles to absorb part of the lateral load and thereby to appreciably ‘stiffen’ the group. Raking piles also reduce the maximum bending moments to which the vertical piles are subjected. However, if there is significant settlement of the surrounding soil, a heavy batter can introduce problems of unquantifiable secondary stresses and invalidate the initial structural basis of estimation of bending moments.

7.2

Methods of lateral load resistance analysis

Where lateral loading is significant, a pile group analysis is required to determine the interaction of the piles and to predict the load resistance, the group lateral movement, and the design bending moments on individual piles. Traditionally, lateral load resistance from bearing piles has been obtained by using raking piles. This is helpful when the bearing pile section has limited transverse stiffness (e.g. an H-pile), where the upper soils are very soft, or where the bearing pile has a free-standing cantilever height above ground level or mudline, as in marine situations. However, with the progress of research, particularly for offshore platform foundations, the design of vertical steel piling for lateral resistance is now also highly developed. Offshore ‘P-Y’ design methods for single piles are given in the API Code RP2A[11]. However, pile group analysis is not explained in any detail in that Code, a list of references being provided instead. The design of laterally loaded piles and pile groups is covered comprehensively in CIRIA Report 103[13]. Research work by internationally accepted geotechnical experts such as Poulos[50][51]in Australia and Randolph in Cambridge[52] have produced hand calculation methods for pile group analysis under lateral loading. These methods consist essentially of determining ‘pile interaction factors’ that should be applied to modify the calculated single pile resistance and deflection values. Some methods, like that of Poulos[53], are relatively simple because he has produced graphs of interaction values to assist the designer (see CIRIA Report 103, Appendix B, 3.3.1). Computer analysis programs are available to perform this work as well, and these are particularly useful for large pile groups. A number of computer programs using elastic continuum analyses have been published. Two programs which give reasonable results are Repute[55] and PIGLET[56]. P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

61

Printed 07/09/05


The designer should bear in mind that these methods are conservative i.e. the actual resistance of individual piles may not be downgraded as much as the methods suggest, and deflection is probably not going to be increased by as much. This is because the methods are based on model tests in a laboratory and have not been extensively validated in the field. A fuller explanation is available in CIRIA Report 103.

7.3

Practical pile group design

The advice in the section below is largely based on Section 2.2 of CIRIA Report 103 Design of laterally loaded piles[13], which provides a very good description of the design procedure for a pile group, including a flowchart reproduced here as Figure 7.1. It is suggested that the three levels of appraisal detailed in the CIRIA report should be adopted, i.e.: 1.

Consideration of the ultimate failure mechanism of the foundation and incorporation of an overall reserve of strength for safety.

2.

Computation of the lateral translation and rotation of the foundation at working loads, and consideration of the effect of this deformation on the whole structure.

3.

Bending resistance of the piles.


Stability

Inherent uncertainties in assessing the loads and stresses in a laterally-loaded pile group require that a reasonably conservative analysis should be adopted, combined with limits on permissible displacement. The application of lateral loading, or transverse loading as in BS EN 1997-1, to a pile creates plastic failure in the zone of passive soil resistance near the ground surface as the piles are pushed into the soil. This zone is unavoidable in reality and the soil is already at ULS. Tentative guidelines for individual piles are to set limits on lateral pile displacement at the ground surface of not greater than 2% of the pile diameter for sands and stiff clays, and not greater than 5% of the pile diameter for soft clays, subject to the tolerances imposed by the structure itself. Deformation

Three conceptual models are currently in use for the design of pile groups: 1.

Structural frame model - In the static and stiffness methods, the piles are implicitly assumed to be end bearing on a competent layer, and the contribution of overlying soft material to load capacity is entirely discounted. Computation of the forces in the frame is carried out by conventional structural analysis[57][58]. Although this is not a realistic model of actual site conditions, the method has given satisfactory results, and it is still used extensively. The method is now principally useful for a preliminary appraisal of the layout of a pile group and for the design of lightly loaded groups. For a more economic design of pile groups subjected to large lateral loads or moments, other forms of analysis are preferred.


62

Printed 07/09/05


Outline Pile Group Based on engineering assessment of - soils data - structural loading - construction constraints - possible pile types

Assess - reliability of data - sensitivity of structure

Select Method of Detailed Analysis Consider - available SI data - magnitude of lateral loads - complexity of pile data - batter of piles

Static Analysis Refine - pile group - size of piles

Analyse Data Select design parameters

Analysis Single piles 1. Elastic continuum methods 2. Subgrade reaction methods (p-y analysis)


Pile 1. 2. 3.

groups Poulos analysis Randolph analysis Elastic continuum computer models (e.g. PGROUP, LAWPILE)

Output Piles - axial load - shear forces - bending moments Group - vertical deformation - horizontal deformation - rotation - flexibility matrix of group

Minor Structures Appraise performance

Appraise Performance of Foundation - serviceability of structure - consider limit state criteria - global effects

Modify design as necessary

Final design details

Figure 7.1

Pile group design procedure flowchart (from CIRIA Report 103)


63

Printed 07/09/05


2.

Spring idealisation – In this model, the soil is modelled by and infinite number of discrete springs (Winkler medium). Transfer of soil masses within the soil mass is not modelled. The method has been extended (by Matlock and Reese [59]) to include non-linear springs, and it is generally refered to as the ‘P-Y’ or the ‘subgrade reaction’ method. The model is well developed, giving satisfactory predictions of the behaviour of single piles and pile groups. It is felt that the reservations expressed in CIRIA 103 in regard to pile groups are no longer valid because p-y modifiers are now available for groups. Refer to user guidance for PIGLET[56].


3.

Elastic continuum model - An elastic continuum model is useful for the analysis of both single piles and pile groups when the soil can reasonably be assumed to be linearly elastic. In practice, provided an appropriate secant modulus is selected, the method gives satisfactory results for piles at working load in most soil types. Complex elastic-plastic soil models for pile group analysis are not generally available and are not necessary if the limit of 0.02D is used for lateral displacement at ground level.

Continuum analyses (e.g. by Poulos[50][51][53][60] and by Bannerjee and Driscoll[54]) incorporate pile/soil/pile interaction. The method usually involves the use of a computer, and the designer should be fully aware of the limitations of the particular program used. While the method is currently best suited to the final design of the foundation, the publication of parametric studies[61]) makes the method more generally applicable. Several programs are already available commercially, and it is anticipated that other comprehensive programs will be produced, to cater for most common practical problems. One of the most recent programs, Repute[55], allows the study of three-dimensional pile groups with non-linear soil models, thereby overcoming some of the known limitations of simpler programs. Repute’s calculation engine, PGROUPN (Basile)[62], is a direct descendent of the earlier PGROUP program developed by Bannerjee and Driscoll[54]. The idealisation of an elastic continuum allows calculations to be performed which give the designer an insight into the behaviour of the pile group, and which aid understanding of the sensitivity of the group to changing loading conditions and soil parameters. The application of these concepts is summarised in Table 5.1, together with the output of each method. The designer should select the method appropriate to the problem in hand, bearing in mind the complexity of the problem and the resources available. For the design of large piled foundations, an analysis based on the elastic continuum approach is considered to be one of the most satisfactory methods available at present, provided that the limit on displacement is satisfied. Analysis of the foundation at SLS factors set to 1.0 enables the designer to assess the significance of the computer predicted deflections and to include the stiffness matrix of the foundation in the overall design of the structure. Further information is contained in CIRIA Report 103 Design of laterally loaded piles[13].


64

Printed 07/09/05


8

THE INSTALLATION AND TESTING OF STEEL BEARING PILES

The installation of steel piles is a specialist activity, calling for considerable knowledge and experience of handling piles and operating hammers to achieve an acceptable placement within specified tolerances of position and level. Guidance on the practical limits that can be achieved in position and level for driven steel piles is available from the FPS (Federation of Piling Specialists) and in the TESPA (Technical European Sheet Piling Association) publication Installation of steel sheet piles[63] (see also Section 8.5). Guidance is also included at the back of the ICE Specification for piling and embedded retaining walls[20]. The designer should refer to these documents before carrying out a design, because the advice given will often affect the details at connections to the pile cap in the structure.


There is much to gain from matching the stiffness of the pile to the hammer and to the anticipated soil resistance at the site to achieve satisfactory driveability and to ensure achievement of the required design penetration. In addition, following research work over the last 30 years, there is a developing understanding of the benefits of measuring the soil resistance during driving as a check on the designer’s predicted compressive axial static capacity (see Reference 20). Although there are some caveats to be applied to this practice, it is indisputable that both designer and installer gain from using dynamic analysis and that this is of ultimate benefit to their clients in many ways. For instance: •

it is a QA tool to permit the evaluation of piles that have an unexpectedly high resistance at a higher level

•

it can lead to fewer and shorter piles being acceptable

•

it can save construction time by avoiding delays

•

it provides an equitable means of resolving disputes about specifications that turn out to be unrealistic due to inadequate site investigation or inaccuracy in design prediction methods.

The relationship between the static and dynamic load capacity is now better understood and in many soils and rocks there is very little difference between them (see Sections 8.2 and 8.3). However, it is still sound practice to carry out static load testing to check the design penetration because in some soils the pile capacity has been known to decrease after driving. Consequently, the designer should understand pile installation methods in order to benefit from this progress in technology.


65

Printed 07/09/05


8.1

Pile driving installation methods

Pile driving installation methods fall into the following categories: •

Dropping weight.

•

Hydraulically raised impact hammer.

•

Hydraulic double acting hammer.

•

Diesel hammer.

•

Vibratory hammer.

•

Jacking machine.

8.1.1 Drop hammers The dropping weight, or drop hammer, is the traditional method of pile driving and is still employed. Normally, purpose-made rigs hoist the pile into position, support it during driving and incorporate a guide for the drop hammer (Figure 8.1). Rope head wheels

Leader mast head


Pile rope Hammer cage Drop hammer Hammer hydralic ram Pile helmet Flexible hydralic pipe Inner leader mast

Two stage hydralic ram Pile Rear braces Leader foot bottom Upper raking hydralic ram

Leader foot hydralic ram

Two hydralic winches Lower raking hydralic ram Leader mast foot

Figure 8.1

A steel pile driving rig

In guiding the pile, a balance has to be struck between providing suitable directional control, and allowing some freedom for slight pile movement within the guides, particularly at the base of the frame. Some twisting or slight lateral displacement can occur as a pile is driven. Slight out-of-verticality rarely affects the pile performance, and slight deviation from a specified pile location can usually be tolerated (see Section 8.5). When driving in boulder clays, it is


66

Printed 07/09/05


prudent to allow some overall additional tolerance in the foundation width to accommodate the eventuality that a pile may strike a boulder which can cause deflection from its intended position. Drop hammers should have a ram mass of between 0.5 to 2 times the pile weight, with the fall height usually being in the range of 200 mm to 2 m. Variants of the simple winch-raised drop hammer include rams raised by steam or compressed air or hydraulics, free falling from the top of the stroke (single acting hammers), or having pressure applied on the downstroke aswell to assist gravity (double-acting hammers).

8.1.2 Diesel Hammers Diesel hammers are no longer used generally in the UK, nor in Singapore or Hong Kong, and therefore details of their mode of operation is not covered here. However, the hammer library of the WEAP program still contains diesel hammer models as they are still used in the USA.

8.1.3 Hydraulic hammers


This type of hammer is suitable for all sites and soil conditions and are now the most commonly used hammers because they are light and reliable. In a hydraulic hammer, the ram is raised by hydraulic pressure and is then allowed to freefall in single acting hammers but in double-acting hammers the ram has additional acceleration to that from gravity from hydraulic pressure that is switched from lifting through a reversing valve. The stress wave transmitted to the pile has a lower peak and longer duration which is more suited to driving piles in clay where the majority of driving resistance is friction on the pile shaft. Hydraulic hammers are more efficient and quieter in operation than other types, and have replaced drop hammers and diesel hammers for most applications. They are compact and adaptable and may be used on a wide variety of bearing piles, including raking piles, and with only slight modification, can be used underwater too. There is a wide range of hydraulic hammers available worldwide produced by several manufacturers. This competition and their popularity has encouraged their development in both technology and size to tackle the largest sizes of pile and with highly sophisticated instrumentation to measure energy output reliably.

8.1.4 Vibratory methods of pile driving Vibratory hammers are usually electrically powered (but may be hydraulically powered), and consist of contra-rotating eccentric masses within a housing attached to the pile head. The majority of pile vibrators run at low frequencies, typically 20 to 40 Hz. At these frequencies, neither the exposed length of pile nor the soil will be in resonance. Sound propagation is low and in cohesionless soils good rates of progress can be realised. During the driving progress, the granular soil immediately adjacent to the pile is effectively fluidised, and friction on the shaft is considerably reduced. In cohesive soils, fluidization will not occur and vibratory pile driving methods are not generally as effective. Vibratory hammers are often used as the initial drive but the final drive needs an impact hammer. P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

67

Printed 07/09/05


The noise and vibration propagation for vibratory type hammers need to be related to the types of soil at the site in order to minimise the environmental impact.

8.1.5 Resonance pile driving Variable frequency vibrators can be used to good effect in some soils and can be useful in environmentally sensitive areas provided that the equipment is suitable for the soil conditions. If the frequency of vibration is increased up to perhaps 100 Hz, the pile will resonate longitudinally, and penetration rates can approach 20 metres per minute in loose to moderately dense granular soils. At these frequencies, non-cohesive soils are fluidised to the point where the frictional resistance on the pile shaft is reduced to close to zero and more driving energy is delivered to the pile toe. This method of pile installation is potentially very effective but needs thorough investigation by the user and the manufacturer to relate hammer mass and frequency reliably to the type of soil.


8.1.6 Jacking machines Hydraulic jacking machines have been developed in the 1990’s to install steel piles quietly and without vibration in response to the environmental legislation that controls construction noise. Lengths of pile, either in short units or in continuous lengths, are forced into the ground by jacking against a reaction. A good understanding of soil resistance can be obtained during installation by monitoring the pressure in the hydraulic system. The reaction system can be provided by tension piles, adjacent piling or dead loads. The ‘Giken’ system and the ‘Dawson’ system are adapted for use on steel H bearing piles and jack against the tensile reaction provided by gripping adjacent previously driven piles. The jacking method has proved very successful for micro piling (piles less than 300 mm in diameter), since the reaction loads are then provided by the structure being underpinned.

8.2

Offshore experience of pile driving analysis

The offshore construction industry has funded research into the science of pile driveability and driving prediction methods since the 1970’s, because the consequences of premature driving refusal offshore are financially intolerable. As a result of this effort, pile and hammer instrumentation systems were developed that could survive the shocks during driving, and knowledge was gained about the inefficiencies of both steam and diesel hammers and of capblock arrangements (see Figure 8.4). These early types of hammer were also found to give inconsistent behaviour between blows as a result of erratic behaviour in their mode of operation, for example in changes in the steam supply volume and pressure or in the diesel hammer’s explosion force. This was not a reliable basis on which to develop the understanding of soil resistance, because the efficiency and input energy varied, and this did not permit confident application of the blowcount/soil resistance outputs from a wave equation analysis, nor any reliable deductions from the simpler pile driving formulae. The latest forms of hydraulic hammer have none of these problems; they are lighter to handle and more compact; have high efficiency, high reliability, P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

68

Printed 07/09/05


repeatable blow characteristics, and from their fitted instrumentation the input energy to the pile can be continually monitored and controlled during driving. These state-of-the-art installation systems are now used for offshore pile driving and it will be beneficial to transfer this experience to the onshore sector.


Hydraulic hammers have been used in research programs to remove many of the unknowns in hammer input energy and thereby permit back analysis of driving resistance; the results have led to correlations with soil type and strength characteristics (see Clarke et al[1]; Euripides project[64]; Jardine et al[3]; and Akbari and Mure[65]). Although it is not always necessary to use the capability of hydraulic hammers for onshore projects, (i.e. they could be viewed as too ‘high tech’), the advantages that they offer to the designer of having reliable characteristics and variable control to the input energy may be useful on sites where the soil or rock conditions are particularly problematical. Such sites could include those where variable weathered rock or mixed soils are present, and where soil testing has been inconclusive or where the buried rockhead is also at variable depth. The hydraulic hammer will then permit driving stress analysis using pile instrumentation and a PDA to check the static bearing capacity of each pile, once the driving resistance has first been compared to a static result. The control of input energy is useful where driving stress in the pile steel could be the limiting factor for the pile section, for example where driving into intact hard rock is necessary.

8.3

Driving formulae and dynamic driving resistance

8.3.1 Driving formulae Traditional methods of determining the size and type of hammer required, or of estimating the driving resistance that can be overcome, have involved the use of some form of pile driving ‘formula’. This relates the measured permanent displacement (or ‘set’) of the pile per blow of the hammer to the pile ‘dynamic capacity’. Driving formulae are based on an energy balance between the input energy from the hammer and the static work expended to move the pile down permanently a small distance (i.e. to a ‘set’). Various pile driving formulae are in common use, many of which have been reviewed by Whitaker[66], who found that in some situations driving resistances predicted by the different formulae may differ by a factor of 3. This variation is related to the basic unknowns in drop, steam and diesel hammer behaviour, blow to blow inconsistency, capblock and cushion wear, etc, leading to doubt over the efficiency, even the average efficiency over a series of blows. The basic approach to pile driving formulae is outlined and some of the main variables discussed. A simplified picture of the driving process and the assumed relationship between pile resistance and displacement is shown in Figure 8.2.


69

Printed 07/09/05


W

Cushion

Hammer h

Helmet

R

Pile resistance

tc

Wp

Pile

R

c

s Displacement

Figure 8.2

Schematic model of pile driving system and assumed driving resistance/pile displacement relationship


This model of the driving system gives rise to the fundamental pile driving formula: R=

ηW h c  s +  2 

8.1

where: R

is the pile driving resistance

η

is the assumed efficiency of the hammer (allowing for the energy loss on impact)

W

is the weight of the hammer

h

is the drop height

s

is the permanent ‘set’ of the pile

c

is the elastic or recoverable movement of the pile.

Formula 8.1 is widely used in the USA adopting a value of 0.8 to 1 for η and 2.5 mm for c, giving the so-called Engineering News formula [67]. In Europe, both η and c are chosen with regard to the type of hammer; the type of material used in the cushion at the head of the pile, and the physical properties of the pile. Probably the best known variant of the formula is that due to Hiley[28][68], where the overall hammer system efficiency (including hammer, anvil, capblock, cushion) is given by:

η =

k (W + e 2 Wp )

8.2

W + Wp

where: k

is the output efficiency of the hammer (ratio of power delivered at the cushion to the rated hammer power)


70

Printed 07/09/05


Wp

is the weight of the pile

e

is the coefficient of restitution between the hammer and the capblock cushion, or top of pile if there is no cushion.

The recoverable movement c is taken as c = cc + cp + cg , where the cushion compression cc = R tc/(AE)c; tc is the thickness and (AE)c the cross sectional rigidity of the cushion; the elastic shortening of the pile (considered as a column), cp = Rl/(AE)p; and the recoverable movement of the ground cg , which may be taken as 0.5% of the pile diameter. Tables 8.1 and 8.2 show commonly adopted values for the quantities k, e and E, depending on hammer type and cushion material.


The Hiley formula can be presented in the form of a nomogram as shown in Figure 8.3. The Figure shows a chain dotted path of an example to demonstrate its use. The particular example illustrated is a 20 m long, 150 kg/m, steel H bearing pile driven to a final set of 2.5 mm per blow with a 4 tonne ram single acting hammer, the drop height being 1.0 m. The ultimate driving resistance R of 1820 kN can then be read off directly. Thus, applying a factor of safety, in this case 2.0, the maximum working load would be 910 kN. However, this driving resistance is not often equal to the static load resistance of the pile owing to the changes in soil resistance which occur with time in soils. Table 8.1

Values of parameters for hammers

Type of hammer

Energy efficiency k

Hydraulic hammer

1

Drop hammer (trigger fall)

0.9

Steam or compressed air hammer

0.8

Drop hammer (winch operated)

0.7

Diesel hammer

0.5

Table 8.2

Values of parameters for cushion material

Cushion type

Coefficient of restitution e

Young’s modulus Ec (MN/m2)

Micarta plastic

0.8

3x103

Greenheart oak

0.5

3x102

Other timber

0.3

2x102

Another popular formula is that of Janbu, which is comprehensively presented in a paper by Flaate[68] in which he also compares the results of its use with both the Hiley[28] and Engineering News[67] formulae, and with the measured capacities from a database of 116 static load tests on various types of pile.


71

Printed 07/09/05

Figure 8.3


72

75

100

150

175

200

300 250

8

Steel bearing pile

Driving cap

7

5

4

25

3

Pile weight kg/m

30

6

2

1

Plastic or greenheart dolly

Weight of hammer W tonnes

L (m)

Drop h(m)

Hammer W tonnes D

5

S/ A 0.5 0.5 D

A S/ 0 1. 0 D 1.

5 1.

2. 0 2. S/A 0 D S/ A

Stroke or drop h (m)

1. 5

15

10

40

Pile length m

10

15

20

25 25 20

30

30

0

5 2

10

20

30


5

50

Set mm/blow R Factor of saftey

25

500

x78 N/mm²

50

100 Pile weight kg/m

75

100

150

175

200

250

300

1000 1500 2000 2500 3000

Ultimate driving resistance R kN

for a single acting hammer W = Weight of ram

P = Average driving stress = R W

Factor of safety normally = 2.0 D = Drop hammer S/A = Single acting hammer

Working load =


Discuss me ...

Nomogram for Hiley[28] pile driving formula

Printed 07/09/05


It should be emphasised that driving formulae are not sufficiently accurate for assessing, by means of the measured set(s), whether a required driving resistance R has been achieved when using simple drop or diesel hammers. This is because of the uncertainties in the values of the efficiency, k, and the various components of c. The characteristics of both the hammer and of the cushion material can alter significantly over the driving period and from blow to blow, thus altering either the input energy or the efficiency of the system. Both will alter the energy delivered to the pile, and therefore produce different ‘sets’. This is not a basis for reliably deducing the driving resistance and therefore driving formulae should only be used for approximate assessments of the sufficiency of a given hammer to drive a given pile to a required penetration.


Reliable prediction of steel pile driving can be made using a modern wave equation method such as GRLWEAP[27] (see Section 4.6 and 7.3.2) that has been rigorously calibrated against the real performance in the field of all hammers that are available. Such a program enables the selection of an appropriate available hammer to install the designed pile to the required penetration, whilst limiting driving stresses within the strength of the pile material, and with adequate margins to allow for the inaccuracies involved in predicting driving resistance and the overall efficiency of the hammer-pile system. In addition, use of the same wave equation model is utilised in onsite analysis of driving performance in the hammer-pile system employing PDA and the CAPWAP[69] program. Peak driving stresses

The recent Eurocode BS EN 12699: 2001[70] recognises current ability to predict and control driving stresses in steel displacement piles so that their potential strength can be used more efficiently in design and installation. Accordingly, the limit for peak driving stress has been increased to 90% of nominal steel yield strength (see Section 7.7 in BS EN 12699). If it is anticipated that the pile will be driven using relatively inefficient and difficult to control equipment such as diesel hammers, the possible peak driving stresses should be considered. In assessing the suitability of a hammer or choice of pile section, the stresses at the head can be estimated using a wave equation program like GRLWEAP that models the behaviour of a diesel hammer and has all available diesel hammers in its program library. By varying the efficiency over an appropriate range, the risk of the pile receiving an unacceptably high peak driving stress can be evaluated and a basis for site control specified or a case for substitution with a more reliable hydraulic hammer can be made.

8.3.2 Wave equation methods A more rigorous investigation of the driving behaviour of a pile may be achieved by means of ‘wave equation’ dynamic analysis of the pile-soil system. The inertial effects of the soil around the pile and the viscous nature of the soil may both be taken into account by appropriate numerical techniques, such as the finite element method (see Smith and Chow[71]). A common simplification is to treat the soil as a massless medium providing friction resistance alone, while the pile is modelled as a discrete assembly of mass elements, interconnected by springs, see Figure 8.4.


73

Printed 07/09/05


a) Actual system

b) Model Diesel

Air/steam

Ram Anvil Capblock Helmet Cushion

Soil

c) Soil model J

Velocity

Figure 8.4

Static resistance

Ru

Dynamic reistance


Pile

q Displacement

Wave Equation Model for pile driveability prediction

At the design stage, the main objective of dynamic analysis of pile driving is to assess the ‘driveability’ of a pile in given soil conditions and to determine a pile+hammer combination that can satisfactorily achieve the required design pile penetration for the predicted static soil resistance with acceptable driving stresses. This is commonly accomplished using one dimensional wave equation programs: probably the most widely used program is GRLWEAP. The program was originally developed in the USA by Goble and Rausche[72] between 1976 and 1988 and has a comprehensive user manual. The main outcome of a driveability study using GRLWEAP is a curve for each hammer relating the ‘set’ for the hammer blows as a function of the soil resistance on the pile. Figure 8.5 shows a series of such curves for a particular hammer and various efficiencies. A series of such curves may be used to select an appropriate hammer to achieve the desired penetration; to assess the installation time period; and to provide a basis for ‘quality control’ on pile installation (in terms of the required ‘set’ per blow for a given dynamic soil resistance).


74

Printed 07/09/05


Soil resistance to driving

300

ncy cie e ffi 80% cy cien e f fi % 70 y e nc ffici e 60% y ienc effic % 0 5

250

200

276 tons 254 tons 230 tons 205 tons

150

100

50


0

0

2

4

6

8

10

12

14

16

18

20

Hammer blows/inch ('SET')

Figure 8.5

Typical ‘set’ versus soil resistance for a hammer

In addition, a driveability study with the wave equation will include prediction of the maximum compressive and tensile stresses in the pile during driving; and therefore the limiting hammer stroke and ‘set’ that will cause the minimum specified yield stress level in the pile to be exceeded. This can be used to draw up the specification in cases of hard driving to prevent pile damage and for site control. Where the actual operating efficiency of the hammer-pile system is unknown (i.e. with drop and diesel hammers), the size of hammer and the range of hammer stroke that may be needed in order for the hammer to deliver sufficient energy allowing for the worst possible energy losses in the driving system can be deduced. This can be achieved by using a range of assumed efficiency for the whole driving system in the WEAP[72] analysis. If hydraulic hammers are to be used, then the range in hydraulic buffer pressure can be determined for single or double acting hammers.

8.4

Pile load testing

It should be borne in mind that the prediction methods for axial pile resistance from soil tests as described in Sections 5.4 and 5.5, are primarily a means of estimation for the conceptual design stage of the foundation. Reference to the SCI database of results on steel piles (ref. SCI Technical Report 552[22]) and Jardine et al[3][112] shows that although these methods can give an acceptable mean and standard deviation in reliability terms, there is always a scatter about such a mean and the chance of an erratic result on any site due to the variability of soils. This, coupled with the inevitable problem of determining an appropriate measure of ‘soil strength’ from laboratory soil tests, means that P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

75

Printed 07/09/05


wherever possible the opportunity should be taken to conduct a pile load test as the only assured method of load resistance determination. Once a trial pile test has been performed and the pilehead load/resistance curve obtained, the wall friction may be separated from the end-bearing by methods such as CEMSOLVE[73][74][75]. The designer can then calibrate the results against the soil profile and extrapolate across the site using the variations in the soil layering found in the site investigation to predict potential changes in pile resistance.


It is desirable to carry out test loading to failure wherever possible, in order to determine the ultimate pile load capacity and to obtain the full pilehead load/deflection relationship. Generally, driven steel piles can be tested to the ultimate load without affecting subsequent load carrying capacity, because of the manner in which soil resistance is generated (see explanation in Section 3.6, 3.7 and 3.8). A possible exception is where piles are driven to hard rock, and the ultimate load may be governed by local buckling of the steel section as a result of high stresses at the tip, rather than downward movement into the rock. Consequently, pile damage due to overload may reduce subsequent load resistance. In most cases the designer commonly only requires test piles to be ‘proof loaded’, and here the applied load may be specified to be between 1.3 to 2 times the required ‘working’ load; a frequently used factor being 1.5. In ‘proof loading’, it is usually required that the pile shall not have failed at the specified design ‘proof’ load, and that the pile head settlement shall not exceed the specified serviceability limit regardless of how conservative the pile design is. Hence, the client is denied the opportunity to reduce the number or depth of piles if the design is over-conservative. As discussed in Section 3 and shown in Figure 3.1, the designer should specify load testing up to 2 to 3 times the estimated design working load and/or to achieve a pile head displacement of 40 mm (i.e. to ‘failure’) whenever possible, in order to define the ‘ultimate capacity’. The ‘proof’ loading procedure otherwise shuts out any potential cost savings to the project by predetermining the number of piles. Static load testing is now expensive and time consuming as compared to dynamic testing. In addition, if the design estimate of static pile resistance from soil and rock samples is too low, it is rarely possible to increase the load range sufficiently within a predetermined kentledge load or anchor pile test set-up to reach the actual ultimate pile resistance (see Figures 8.6 and 8.7). If dynamic pile testing is used there are no such impediments, and it is therefore potentially a far superior method. Now that the accuracy and reliability of dynamic pile testing has been established by comparing it directly with static testing (see Section 8.4.2), and it has become accepted practice (see the ICE Specification for piling and embedded retaining walls[20]), it has rapidly become the norm. The CAPWAP[69] pile driving analysis also provides another motivation, which is that the force in the pile can be checked during driving, and the driving then terminated for each pile when a certain defined pile resistance has been achieved. Considerable effort is now being applied to analysing pile load test results and comparing static load tests with dynamic pile testing, see Section 8.4.2. Steel piles lend themselves to ready onsite quality control on each pile during driving because they respond reliably to dynamic pile testing analysis (DPA). The dynamic resistance can be directly correlated with static load test results on trial piles and then the capacity of any working pile can be checked during installation to confirm consistency of behaviour over the site. P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

76

Printed 07/09/05


8.4.1 Static pile load testing Two forms of test loading are in general use, the maintained load test and the constant rate of penetration test. Details of both methods of test are given in BS 8004[15] Clauses 7.5.5 and 7.5.6. Figure 5.4 shows a typical result for a pile loaded to failure. The amount of kentledge or tension resistance should always be in excess of the estimated ultimate bearing capacity of the pile. A factor of 1.5 should be a minimum, since loading to twice the ‘working’ or ‘serviceability’ load is the accepted requirement, and the working load is normally about half the ultimate capacity in both LSD and ASD procedures. MLT, the maintained increment load test

In the maintained increment load test (MLT), kentledge or adjacent tension piles or soil anchors are used to provide a reaction for the test load applied by jack(s) placed over the pile being tested. Figures 8.6 and 8.7 show typical arrangements. The load is increased in definite steps, and is held at each level of loading until all settlement has either ceased or does not exceed a specified amount in a stated period of time. Kentledge of cast-iron blocks, concrete etc.


Secondary beams

Main beams

Load cell

Isolated datum beam for measuring settlement

Figure 8.6

Timber crib

Loading jack

Pile under test

Test load arrangement using kentledge

Secondary loading beams Main loading beams

Tie-bars

Load cell

Loading jack

Anchor piles each side Isolated datum beam for measuring settlement

Figure 8.7


Pile under test

Test load arrangement using anchor piles

77

Printed 07/09/05


CRP, the constant rate of penetration test

The second method is the constant rate of penetration (CRP) test. The loading arrangement of the apparatus is the same as for the maintained load test, but the load is increased continuously at a rate such that the settlement of the pile head occurs at a constant rate per minute. The constant rate of penetration test (CRP) has been demonstrated to be unreliable and uninterpretable in respect of deriving and understanding ultimate pile capacity (see Fleming[73]). For static pile load tests, and ULS design procedures, the CRP type of test should be discontinued in favour of a maintained increment load test (MLT) that can give a reliable indication of the ultimate load resistance (see recommendations of Federation of Piling Specialists[105]). In addition, the time allowed for each stage of load application in an MLT should be increased in line with the recommendations of England and Fleming[76] to permit development of the equilibrium resistance to the applied load.


Comment on testing procedures

Excessive conservatism has been found in current practice of proof loading in the currently used specifications for load testing piles, which has been compounded by unrealistically low design assumptions on the soil parameters that are used in pile resistance prediction methods. Consequently, designers are unable to interpret the ultimate pile resistance from their load test results, and the whole basis of the new limit state design procedures to produce greater economy is denied. It is hoped that the guidance given herein will help correct these errors and thereby permit more economic pile design. Testing steel piles nearer to their ultimate capacity instead of to ‘proof loading’ or to over-conservatively assessed design ‘working loads’ will improve knowledge of their behaviour and permit improvements to the currently restrictive design rules.

8.4.2 Dynamic pile load testing For steel piles, the dynamic pile testing method has been proved to be very suitable. The CAPWAP[69] computer program is now used to produce ‘pseudostatic’ pilehead load/displacement diagrams from stress wave measurements with a surprising accuracy of ± 10%. This has been verified by comprehensive comparisons with static load tests on tubular, H-pile and sheet piles (see the Proceedings of the International Conferences on the Application of Stress Wave Theory on Piles[77]). A comparison of interpreted pile dynamic soil resistance versus static load test measured resistance is shown in Figure 8.8 as taken from The analysis of pile driving – A state-of-the-art[116]. Later work by Pile Dynamics Inc. examines the CAPWAP correlations with more rigour for all pile types[117]. The CAPWAP program enables the magnitude and distribution of soil resistance down the pile shaft to be derived by curve matching techniques from stress wave and acceleration recordings in the steel pile taken during driving. This is the basis of the Pile Driving Analyser equipment and of dynamic pile testing. A further subroutine permits the ‘pseudo-static’ pile head load/deflection test graph to be derived from the stress wave taken from a blow at the terminal pile penetration, that can then be compared directly with any static load tests that are available. P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

78

Printed 07/09/05


Case method capacity (tons)

350

300

250

200

150

100

50

0

0

50

100

150

200

250

300

350


Static loading test result (tons)

Figure 8.8

Comparison of dynamic and static interpreted pile capacity

By waiting a period of time and remounting the hammer on the pile, the ‘set-up effect’ or change in soil resistance with time can be determined from restrike blows using the CAPWAP program (see Jardine et al[112], Fellenius[114], and Komurka[115]). Obviously the confidence that the designer will have in the static capacity ‘prediction’ results from dynamic analysis will be greatly enhanced if a correlation can first be carried out against a static capacity test on a trial pile at the site. However, the SCI have obtained good correlation between static and dynamic capacities for granular soils and rocks in their database (refer to SCI Technical Reports 552[22] and 621[29]) and this endorses the use of dynamic test methods for those conditions. On clay sites, dynamic testing has to be carried out with care to understand any set-up effects that may occur after driving the steel piles, and thereby to avoid interpretation of too pessimistic a value for the ultimate load resistance. Equally, there are often cases of a reduction of capacity with time, particularly when driving piles into shales and mudstone bedrocks and when piles are very close together in a group. Hence it is always of value to carry out restrike tests as part of the standard procedure when dynamic testing is available, to avoid reliance on unnecessarily high factors of safety as factors for ‘ignorance’. The ability of CAPWAP to separate shaft friction from endbearing resistance is a very useful tool and is particularly useful in predominantly endbearing piles. Pile stress analysis using a Pile Driving Analyser and pile instrumentation can be an aid to onsite control, particularly when hard driving is required to establish a high end bearing resistance and minimise pile tip damage. On such P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

79

Printed 07/09/05


sites it is often beneficial to extract the test piles to check for such damage. This will permit a more confident basis for the ‘set’ to be used as a control during driving of the working piles. Dynamic load testing of piles may also be used to investigate pile resistance set-up, particularly on test piles (see Jardine et al[112], Fellenius[114], and Komurka[115]).

8.5

Steel pile installation tolerances


It is not possible to install driven bearing piles to such fine tolerances of position or level, or of horizontal or vertical alignment as are possible with plunge columns. The achievable tolerances are strongly related to factors such as: •

the accuracy in setting up the piling equipment

•

the stability of the piling platform

•

the accuracy or ‘repeatability’ of the measurement system

•

the fixity in machine parts

•

any presence of obstructions in the ground and any variations in the properties of the soil especially near the point of entry of the pile into the ground surface

•

the inclination of the strata

•

operator error.

The pile design should allow for bending stresses caused by a specified inaccuracy in the installed position of a pile that has been agreed with the installation contractor. The order of magnitude for tolerances which can normally and reasonably be achieved are quoted in certain specialist publications for example the ICE Specification for piling and embedded retaining walls[20]. Typical tolerances quoted for bearing piles are generally: •

A positional tolerance of 75 mm

•

A maximum of 1 in 75 on axial line for vertical piles.

Much closer tolerances on position can be achieved for plunge columns, (see Section 4.6), because they are lowered into a predrilled hole and not driven. The correct positioning of the machine used to drive the piles is a primary requirement for pile location in plan. In regard to verticality, once a pile has penetrated some short distance into the soil, correction of an initial error becomes more and more difficult as driving proceeds. If the pile has an appreciable upstand above ground level, for example when used as steel column-piles in integral bridges, it is generally possible to correct a deviation in verticality by pulling the piles back using wire strops and tugger winches working against an anchorage or ‘deadman’, or against previously driven adjacent piles.


80

Printed 07/09/05


8.6

Environmental factors with driven piles

Increasing attention has been directed to this subject in recent years with regard to driven piles and this is covered in the SCI Publication P308 Specifiers guide to steel piling [78]. Although the duration of the piling contract may be short in comparison with the whole contract period, noise and vibration perceived by the public may be more acute during the piling phase, especially as they will be fresh to the intrusion of construction noise on the site. Human perception is very intolerant of noise and vibration or shock transmitted through the ground. Prior education of the public is required before the noise of piling becomes acceptable. Efforts made to advise the public and to carefully plan the precise times of driving can reassure those likely to be affected in the vicinity of a pile installation and can result in the necessary cooperation.


In the United Kingdom, the Control of Pollution Act (1974)[79] provides a legislative framework for, amongst other things, the control of construction site noise. The Act defines noise as including vibration and it provides for the publication and approval of Codes of Practice, the approved code being BS 5228[80]. A section of that Code (Part 4) deals specifically with piling noise. This Code was revised in 1992 to include guidance on vibration. Two relevant documents include the TRRL Research Report RR53 Ground vibration caused by civil engineering works[81], and the Corus publication P105A Legislation and Practice on Noise and Vibration Control with particular reference to steel piling[82]. BS 6472[83] deals specifically with Evaluation of Human Exposure to Noise and Vibration in Buildings and provides guidance on the subject.

8.6.1 Noise from piling operations Environmental restrictions on the permitted noise and vibration levels are often imposed in the Conditions of Contract for a project, particularly in urban areas in order to mitigate the effects of construction, especially for pile driving. Pile driving is an inherently noisy operation but has received much attention in recent years to develop quieter plant. Noise levels of 85 decibels within 10 m of the piling plant were quite common compared to 80 decibels now. Typical data on noise levels produced by piling operations were published in CIRIA Report 64 Noise from construction and demolition sites - measured levels and their prediction[84]. These are discussed and interpreted in CIRIA Report PG9 Noise and vibrations from piling operations[85]. However, the data for piling hammers is now out of date because the old types have been replaced by new quieter hydraulic hammers. In particular, unshielded drop and diesel hammers are now never used in the UK and therefore the data should be replaced by that from hydraulic hammers. Investigations into the sources of the noise showed that a large proportion of it arose from secondary effects, such as rebound of the hammer, rope slap, engine noise, etc. Improved design of the components of a piling rig has considerably reduced the high-frequency content of the noise emitted. To dampen the noise sufficiently to be acceptable in urban situations, it has also been necessary to enclose the hammer in an ‘acoustic chamber’ (see Figure 8.9). Hydraulic hammers with a ‘skirt’ have reduced the noise because the point of impact of the ram to the top of the pile is enclosed.


81

Printed 12/09/05



Figure 8.9

Fully enclosed hydraulic hammer driving piles for housing

In areas where severe restrictions are placed on noise levels, pile driving vibrators or the Tosa, Giken or Dawson hydraulic pile jacking machines may be used. Such machines emit a different frequency and acceptably lower level of noise and eliminate the percussive impact, but they involve the use of an auxiliary power unit which could also emit a high level of noise and therefore is also enclosed. Modern diesel engines used in these power units are now much quieter than they used to be owing to developments in mechanical engineering.

8.6.2 Ground vibrations caused by piling It is widely recognised that noise and vibration, although related, are not amenable to similar curative treatment. In the main, noise from site is airborne and consequently the prediction of noise levels is relatively straightforward, given the noise characteristics and mode of use of the equipment. On the other hand, the transmission of vibration is largely determined by site soil conditions and the type of structure involved. General guidance can be derived from a study of case histories of similar situations. Useful references on the subject of ground vibrations are provided by CIRIA Technical Note 142 Ground-borne vibrations arising from piling[86], the publication Dynamic ground movements man-made vibrations in ground movements and their effects on structures[87] and by BRE Digest No. 403 Damage to structures from ground-borne vibration[88] and the references given in Section 8.6.1. Prediction of peak-to-peak acceleration or velocity in real situations is not straightforward. Firstly, the transfer of energy to soil is poorly understood and secondly, attenuation of high-frequency components is rapid. In addition, the response of various forms of construction in adjacent inhabited buildings to ground vibrations is difficult to predict, and some structural details e.g. floor spans which resonate, may lead to a magnification of the effect. The most widely accepted of these criteria are based on the peak particle velocity or the energy intensity of the vibrations induced in the soil adjacent to the foundations of a building. Empirical guidelines have been drawn up using these criteria to P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

82

Printed 07/09/05


define various levels of ‘damage’. One of the most popular of these is included in the German Standard DIN 4150[89], the recommendations of which are listed in Table 8.3. Table 8.3

Maximum allowable peak particle velocity (from DIN 4150)

Class

Description

Max velocity mm/s

1

Ruins and buildings of historical value

2

2

Buildings with existing defects

5

3

Undamaged buildings in technically good condition 10

4

Strong buildings, and industrial buildings

10-40


These recommendations have not been drawn up specifically for ground vibrations induced by piling, and it is considered that they are overly stringent for that purpose. For structures which are not of great prestige or historical value, it is considered that the limits in Table 8.3 could well be doubled without noticeable effect. When considering reasonable limits for ground vibrations, the ambient background level of vibration should be assessed first. In built-up areas, heavy traffic can cause surprisingly high intensities of vibration and peak-to-peak velocities exceeding 3 mm/s have been recorded at a distance of 10 m from a road. The use of empirical limits on velocity or accelerations in specifications and contracts necessitates the use of field instrumentation to observe the actual induced vibrations at the site. Several levels of recording are possible and specialist companies such as Testal or PMC provide a measurement service and can advise on equipment. The simplest is manual recording of peak-to-peak signals and the most complex is a full record of the ground vibrations to enable a frequency analysis to be carried out. In general, human perception of vibrations occurs at levels which are low in comparison with the thresholds of risk for structural damage. BS 6472[83] sets out Tables for vibrations in various different types of accommodation for vibrations in the range 1 Hz to 80 Hz. The vast majority of piling operations currently in use give rise to vibrational energy within this range. The type of pile used greatly affects the intensity of ground vibrations. Steel piles have low displacement and cause much less ground disturbance than full displacement precast concrete piles. Since steel piling takes very little time to install and is an appropriate construction method for most soil types, it has advantages to the contractor. Public irritation and objections to noise and vibration from piling installation can be minimised and their cooperation gained by the contractor and client giving prior notice and careful advice and explanation to nearby residents. Where test piles are used, the opportunity should be taken to also monitor the noise and vibration levels. Data can then be used to agree piling procedures with the local authority’s Environmental Health Officer.


83

Printed 07/09/05


9

TECHNICAL AND COST BENEFITS

9.1

Steel pile economics


Steel H-piles can often be the most economic option for a piled foundation where several of the technical or cost benefit factors affecting pile choice can utilise the advantages listed in Section 1.2. These advantages are repeated below for ease of reference: •

Reduced foundation construction time and site occupation.

•

Reliable section properties without need for onsite pile integrity checking.

•

Ductility also gives high resistance to lateral loads for marine structures and compliance in integral bridge foundations.

•

Larger wall surface area giving better friction capacity than equivalent diameter concrete pile.

•

Higher end bearing resistance in granular soils and rocks mobilised by pile driving as compared to boring.

•

Closer spacing possible and therefore smaller pile caps.

•

Pile load capacity can be confirmed during driving by Dynamic Pile Analysis (DPA) on any pile.

•

Low displacement of adjacent soil during driving.

•

No arisings and therefore no spoil disposal offsite.

•

Easily extracted at end of working life.

•

Reusable or recyclable following extraction to meet Government objectives in sustainable construction.

Different projects will have differing priorities for these technical, environmental and cost benefit factors and this will affect the choice of pile type. The case history examples given later in this Section of the Guide serve to illustrate how these factors can vary according to the site location and the type of project.

9.2

Soil conditions

Steel H-piles can be ideal for situations where buried rock is encountered at shallow depth. H-piles are particularly economic where depth to rockhead is less than 26 m, (which is the maximum length of pile that can be transported by road). Longer piles will need additional lengths welded on site and have this additional cost. As covered in Section 5, the achievable end-bearing pressures are much higher for driven steel piles in rock than for concrete piles. This can give a significantly higher resistance for steel piles that can reduce the number of piles required to support a given vertical load. For other situations, steel H-piles offer a larger wall surface area than concrete piles of similar diameter, which can reduce the required pile length in friction pile design.


84

Printed 07/09/05


As noted in Section 5.2, the risk of pile plugging has been overplayed in the past and this has been demonstrated to be so rare in practice that it is an unreasonable assumption to make in design. Modification to the design can be made if plugging is observed during installation and confirmed in pile tests. Steel piles can sustain high lateral loading by flexing to great advantage on many structures. The design manual for roads and bridges includes rules for design of laterally loaded piles in BD 32/96[90] that recognise this advantage.


There are fundamental differences between driven piling and bored cast-in situ concrete piles that must be understood by designers so that they can select the most suitable and economic type of pile for their project. These differences are as follows: •

Driven piling requires knowledge of the installation driving stresses in either concrete or steel and the soil resistance during driving (see Section 8). Design of steel piles is fairly standard (see Section 4).

•

Design of bored cast-insitu piling requires knowledge of soil disturbance during boring and the effects on soil resistance after concreting in situ. The degree of soil disturbance varies with soil type and must be allowed for in design.

Where the structural designer does not have the necessary experience of piling installation, he should obtain the assistance of a piling design engineer from a geotechnical company to perform the comparative designs for him. For ‘Design and Build’ type contracts, the specifier should ask for comparative steel and concrete pile options in the piling tender.

9.3

Design configuration

Many cost comparisons between steel and concrete piles are performed at design stage by using straight replacement on the same grid or same arrangement of piles. For some soil conditions where buried rock is encountered at shallow depth this may be appropriate, but for most other situations it is necessary to change the grid or spacing of piles in the arrangement to suit the fact that steel bearing piles can withstand more vertical load than concrete sections and therefore fewer piles are needed. To decide on the optimum size of pile to install at a particular site will require a driveability prediction analysis using the GRLWEAP[27] computer program as explained in Section 7. This can be used to determine both the lightest steel section that can be driven and the pile capacity that can be achieved with alternative sections. Using these results, the range of grid spacing and the total number of piles required for each case can be determined.


85

Printed 07/09/05


9.4

Case Studies

Three case studies are included here to illustrate the technical and cost benefit advantages of steel pile applications to foundation construction.

9.4.1 H-piles for Albert Quay, Aberdeen This is an example where the technical benefits of H-piles override important factors.


Discussions with the designer, Arch Henderson & Partners of Aberdeen, who are the retained consultant to the Aberdeen Harbour Authority, revealed that they always specify steel H-piles for the port works in Aberdeen Harbour because from experience they find that H-piles are the most suitable for the soil conditions. The soil layers vary between the various Quays across the harbour but the underlying gravels, hard boulder clays and granite bedrock all have hard driving conditions. H-piles stand up to these conditions better than precast concrete piles enabling the required capacity to be achieved as checked by dynamic load testing. The piling contractor was Volker Stevin, the main contractor Balfour Beatty, and the designer Arch Henderson of Aberdeen. A typical view of the site is shown in Figure 9.1, where the Hitachi piling rig can be seen driving one of the vertical H-piles with a later photograph showing the rig working on the 3 : 1 raking piles. This rig is capable of driving piles of length up to 33 m in one piece without needing welded add-ons.

Figure 9.1

Installing vertical and raking H-piles at Albert Quay, Aberdeen, (Courtesy of Volker Stevin Ltd)

The project has had 2 phases which utilise H-piles to support a relieving platform behind the sheet piled harbour quay walls. There were a total of 380 H-piles with some 50% raking and the remainder vertical piles. The average pile length was 25 m passing through made ground over boulder clay into bedrock. H-piles were specified by the client and were supplied by Corus in


86

Printed 07/09/05


one length delivered by road. Acceptance was by achieving a specified ‘set’ that was agreed with the designer at the outset.

9.4.2 Railtrack Dartford Resignalling project This project is an example where cost-benefit factors of steel H-piles override other technical issues. It was completed in 1998.


Balfour Beatty Rail Ltd as the ‘partner’ for maintenance in the region, agreed with Railtrack their client that the most economic basis for the re-signalling contract in the Dartford area would be to use steel H-pile foundations. This would give overall savings by enabling: •

easy and quick connections between the foundation bases and the gantry superstructures

•

standardisation and predictable installation procedure using a self-propelled truck mounted pile installation rig

•

minimising track occupation time.

The piling train was purpose built by Balfour Beatty Rail Plant and included a diesel train with a crane bogie between two material trucks and an accommodation/messing carriage. The crane handled the piling cage for aligning and supporting the piles which was then fixed to the lead materials truck. The crane handled the pile driving hammer. A typical set up is shown in Figure 9.2.

Figure 9.2


H-pile installation by Stent Foundations Ltd on Gantry base position for Railtrack Dartford Resignalling contract

87

Printed 07/09/05


The piling subcontractor was Stent Ltd who chose to carry out a new soils investigation at each gantry base location and thereby permit site specific pile design. Accordingly, Fugro Ltd were employed to carry out in situ Cone Penetrometer Testing (CPTs), the results of which permitted a direct calculation of the pile driving resistance in the soil profile. The soils contractor calculated the length of pile required at each site and predicted the size of hammer required to drive the H-piles in the most arduous conditions of the 300 pile locations needed and this was standardised throughout to avoid confusion and to minimise pile acceptance time. An in-house Fugro wave equation programme was used that is comparable to the GRLWEAP[27] software produced by Pile Dynamics Inc. in the USA, who are the market leaders.


Steel H-piles were selected because they have good driveability in all soils, are preformed to shop quality thus avoiding integrity testing, and do not use wet concrete so have no delays. They were better able to withstand the predicted hard driving at many locations and connection to the steel gantry superstructure was easy and simple. The H-pile was more suitable than tubular steel for pile add-ons and for connections. An H-pile section was selected to suit the heaviest loadings and driving conditions and standardised for all locations to permit a standardised connection base to be designed for all sites. The pile section lengths were standardised at 8 m to fit the railway wagons and the capacity of the available craneage that was supplied. End plates were welded on and where a design length of greater than 8 m was needed, sections were connected using a bolted splice. Piles were therefore designed for skin friction on the external flanges only. The cross-track gantries required two piles each side and the cantilever signs required three piles, the one at the back working in tension. The tension piles were the longest and were up to 26 m long. For installation, a cantilever piling gate was fixed to the crane wagon to guide the H-piles and hold them vertical whilst driving as shown in Figure 9.2. The pile load transfer plates and base transfer structure are shown in Figure 9.3.

Figure 9.3


Connecting gantry base transfer structure to H-piles. (Courtesy of Stent Foundations Ltd)

88

Printed 07/09/05


9.4.3 Plunge columns for CTRL Ashford Tunnels This is an example of the use of UC plunge columns to give cost and time savings during construction of the cut and cover box tunnels on the CTRL Contract 430 for the 4-track rail tunnel approaching Ashford International Station in 1999/2000. Use of steel plunge columns is becoming very popular as their benefits are increasingly recognised by construction contractors in civil engineering projects. There is already widespread use in basements to permit top-down construction, but this example shows the application to railway tunnels. This Section is a brief summary of a technical paper that was presented in World Tunnelling publication in June 2001 by Howard Roscoe of Skanska Construction UK Ltd and David Twine of Rail Link Engineering[92].


This tunnel is some 570 m long with a rectangular cross-section comprising two external contiguous bored pile retaining walls, and two intermediate ‘walls’ using steel UC section plunge columns set into 900 mm dia. bored concrete piles. The piles support a ground level roof slab that takes heavy loads from roads and utilities that had to be diverted to permit the CTRL construction. There was no choice but to adopt a top-down construction method beneath this slab so that the diverted traffic flow could be maintained without disruption during excavation of the tunnel box below. Steel plunge columns were selected by the designer/project manager Rail Link Engineering and the contractor Skanska working together on the value engineering phase in the initial stages of the work to improve safety and constructability; to reduce disruption to existing roads; and give added surety to the work programme. An open column configuration was required in the intermediate walls to permit spaces for the transverse horizontal propping between the outer retaining walls (Figure 9.4). The steel plunge columns could be installed first to a close positional tolerance before the slab was cast, and provided immediate support during excavation beneath the slab, with no costly construction of pile caps. They are thus permanent props that also give temporary support, similar to their use in basements.

Figure 9.4


Advance box excavation roof supported by plunge columns. (Courtesy of Skanska Construction UK plc)

89

Printed 07/09/05


A high accuracy in position is required with the plunge columns in order to minimise bending moments at the head of the column due to eccentricity, that could cause buckling. This was set at a positional accuracy of ±10 mm in plan and 1 in 200 verticality. This compares to the usual 75 mm in plan position and 1 in 75 verticality normally accepted for piling (see Section 7).

9.4.4 Plunge columns for BMW, Hams Hall basement This is an example of using UC sections as plunge columns to provide permanent structural support within a basement to permit top-down construction, with cost and time saving benefits.


The steel plunge columns were installed first to a close positional tolerance before excavation for the basement commenced and supported the steel framed ground floor and building above (Figure 9.5).

Figure 9.5

UC Plunge columns for intermediate supports at BMW, Hams Hall basement. (Courtesy of Stent Foundations Ltd)

The UC sections were 356×406×235kg/m of grade S355 to take the steel framed building load, and were placed using a jig devised by Stent that was set up onto a temporary steel pile casing (Figure 9.6). The bored pile diameter was increased from 750 mm to 900 mm in order to provide sufficient clearance between the plunge column and the T20 rebar cage within the bore, allowing adequate space for tolerance in position. The required high accuracy in position was achieved using laser levelling and EDM (Electronic Distance Measurement). This achieved a positional accuracy of +/-10 mm in plan and a 1 in 400 verticality was achieved using a laser plumbline system.


90

Printed 07/09/05



Figure 9.6

Lowering plunge column with shear studs into guide casing at BMW, Hams Hall (Courtesy of Stent Foundations Ltd)

Note that shear studs were welded onto the surface of the UC sections to ensure adequate load transfer into the bored concrete pile.

9.4.5 Universal Bearing Piles (H-piles): A cost and performance study The Cost and Performance Study was compiled by consultants Harris and Sutherland in 1977[91] under a commission from the British Steel Corporation (BSC) to investigate the reasons for the selection of steel H-piles on particular projects and to consider when they can be an economic choice compared to concrete piling. Both technical and cost-benefit factors were appraised. The design parameters used in that book were taken from the BSC Guide published as the original Steel Bearing Piles Guide, dated October 1974 by G.M.Cornfield of Constrado[10] (the research division of BSC). The cost data are well presented and carefully analysed. They show that several factors have to be present on the site in order that the steel H-pile alternative is economic in relation to concrete piles. The total cost per metre comparison between H-piles and concrete precast piles is similar to that of today. Costs are generally presented to a unit of ‘per tonne of load carried’. This assumes that all other delivery and installation costs are spread over 100 piles per site. The Harris and Sutherland study presents the technical and cost data from 23 case histories of the use of steel H-piles. The sites were selected from areas all over the UK, mainly from highway bridges where the results of trial and test piles were available. The technical data demonstrate the high load carrying capacity of H-piles and in particular, the high end-bearing obtained by percussion driving into the various rock types encountered in the UK. All but four of the sites were from highway bridges, but of these four, two of them were residential tower blocks of 16-24 storeys in height; one steel works in the USA; and the other a 16 storey office block. High pile loads were a feature of P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

91

Printed 07/09/05


all the sites with design working loads of 70 t to 150 t and test loads from 150te to 450 t. It can be concluded that steel H-piles can be particularly suitable and economic for large construction projects that involve large numbers of highly loaded piles. Most of the jobs also involved hard driving conditions through dense gravels and into bedrock where the small displacement steel H-piles offer the best driveability and high strength necessary for satisfactory installation. The alternative precast concrete piles are high displacement that can refuse prematurely and cannot take the high driving stresses incurred in driving into bedrock or dense gravels.

9.5

Cost comparisons

This Section has been written to assist designers with the cost comparison between steel and concrete piles.


Realistic prices will only be obtained for a steel piling option if main contractors seek bids from competent steel pile installation contractors. Very few specialist piling contractors now provide both concrete and steel piles due to intense competition in the construction market. Most have specialised into either concrete or steel piles and often either preformed piles or bored piles. Ask the main contractor to bid for both concrete and steel pile options and suggest that he asks the FPS and Corus for a list of competent steel pile installation contractors in the UK to put on his bid list. Prior knowledge to the steel manufacturer Corus of impending piling tenders is useful to projects in order that piling supply can be arranged. For preliminary estimates at conceptual design stage, Corus can supply prices to the specialist steel piling installation contractors. A ready reckoner to enable comparison of steel and concrete piling might consist of a generalised equation but this would be out of date as soon as there is any market change in prices. As there are two main elements in costing a piling contract, namely the installation and material supply, there is scope for significant change in either, dependent on the state of the market at any particular time. For preformed displacement piling, the material cost will be affected by the pile length required. This will depend on whether the pile is mainly frictional soil resistance or mainly end-bearing resistance but will also differ between concrete and steel because of the different ratios of cross-section to exposed surface area. The predicted soil resistance to driving or pushing will affect the confidence of achieving a required penetration and so a driveability analysis is needed in order to check that any selected section can be satisfactorily installed with the equipment available from the contractor. This procedure must be carried out before deciding on the piling specification that is put out for bidding. Designers may be experienced at using pile driveability wave equation analysis programs, such as GRLWEAP[27], or they may use specialist geotechnical engineers or pile testing companies to provide this service during the design phase of the project.


92

Printed 07/09/05


The installation cost will be affected by several factors that can vary from site to site, including: •

Degree of reality in the designers specification for the contract bids.

•

Rig delivery cost to site.

•

Rig hire dayrate and predicted productivity (number of piles per day and pile length).

•

Maximum and average pile length (affects rig size required).

•

Pile transport cost to site and onsite handling required.

•

Efficiency and experience of the installation contractor.

•

Cost of spoil removal where required.


In addition, the type of pile (concrete or steel and area ratio of section) will govern the working load that can be achieved and therefore the number of piles required. Concrete piles need larger pile caps whereas steel piles can eliminate need for pile caps altogether. Alternatively for base slabs, the number of piles will affect their spacing and therefore the thickness of slab and the reinforcement required in it.


93

Printed 07/09/05


10

STEEL PILES/STRUCTURE CONNECTIONS

A number of methods for connection of steel piles to the superstructure have been established through construction experience and many of these are given below. Some alternative details are given in BS 6349[6] for marine structures.

Pile cap or slab

Rebar welded to flanges

Pile cap reinforcement

Shear studs; weld on rebar or weld on angle


Compressive loading only

Compressive and tensile loading

Figure 10.1 Steel H-pile connections to r.c. beams and slabs

Figure 10.2 H-pile connection to UB column or stanchion

Where it is required to resist bending moment at the head of a pile, the means of connection becomes more complicated. Reinforced concrete capping beam connections to transfer moment into the head of steel high modulus piles have been designed by the SCI and are described in the publication Integral steel bridges: Design of a single-span bridge - Worked Example[32]. A typical connection is shown in Figure 10.3. Some practical means of providing composite connections to tubular steel piles are shown in the SCI publication Integral steel bridges: Design guidance[31] and these are illustrated in Figures 10.5 and 10.6. Further details are available in the SCI companion publication Integral steel bridges: Design of a multi-span bridge - Worked Example[18].


94

Printed 07/09/05


8 No. T32 at 200 crs. 7 Rows 2 No. Studs/Flange at 200 crs

150 150 7 Rows 2 No. Studs/Flange at 200 crs T40 at 150 crs

Asphaltic plug joint


Concrete deck

Deck beam

Reinforced concrete pile capping beam/wall

Construction joint Note: Similar arrangement for reinforced concrete deck beam

High Modulus Pile Sheet pile

UB

Figure 10.3 High Modulus Pile moment connection to an r.c. capping beam


95

Printed 07/09/05


Asphaltic plug joint

Road construction

Deck beam

Construction joint

Concrete cross head beam /endscreen wall Select granular fill

H-pile


Embankment fill

Figure 10.4 External shear stud composite connections for H-piles to structure

It has been found that often the welding gun does not work well onsite and it is better to carry out most of the welding in the shop prior to delivery to site. Figure 10.5 shows an H-pile connection detail from a bridge where the shear studs were welded vertically in the shop on a backing strip that can then be welded on site to the pile

25 dia. studs 120 long 50 200 750

200 200

Blinding Polystyrene 600 dia. isolation tube 305 x 305 UC Section

Figure 10.5 Detail of H-pile integral bridge abutment P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

96

Printed 07/09/05


11

CORROSION AND PROTECTION OF STEEL PILES

11.1 The need for corrosion protection In the majority of circumstances, steel bearing piles can be used in the unprotected condition. In most applications, such corrosion as will occur is predictable and can be allowed for in design. In more aggressive environments, additional protection may be required dependent on the design life required.


The corrosion of steel that is completely buried in undisturbed natural soil is negligible. This is recognised in Clause 10.3.5 of BS 8004[15], and Clause 4.4.4.4.3 of BS 8002[5]. There are recorded examples where steel piles have been completely buried for up to 40 years and when extracted still had the original millscale with the makers stamp quite clear. Section thicknesses when checked were within the mill rolling tolerance (ref. Irlam Steel Works site, Manchester, Corus STC, 1997[93]).

Figure 11.1 Reusable steel piles recovered after 40 years service

Data has been gathered by several researchers on corrosion losses in various environments and analysed to give average and maximum corrosion rates. Details are given below of the results of accepted references such as those of Romanoff [97] and Ohsaki[98] that have formed the basis for the corrosion rates quoted in BS 8002, BS 8004 and BS 6349[6]; the Highways Agency’s BD 42 Design of embedded retaining walls and bridge abutments[8] and by Corus in their guidance publications on the subject that include: A corrosion protection guide for steel bearing piles in temperate climates[94] and The prevention of corrosion on structural steelwork[95].


97

Printed 07/09/05


11.2 Standard corrosion allowances Table 11.1 advising accepted standard values for corrosion losses in soils for a range of design lives is reproduced of Table 4.1 of EN 1993-5: Piling[7]. Table 11.1 Loss of thickness (mm) due to corrosion of steel piles in soils, with or without groundwater present 5 years

Required design working life

25 years

50 years

75 years

100 years

Undisturbed natural soils (sand, silt clay etc.)

0,00

0,30

0,60

0,90

1,20

Polluted natural soils and industrial sites

0,15

0,75

1,50

2,25

3,00

Aggressive natural soils (peat, marsh, swamps, etc.)

0,20

1,00

1,75

2,50

3,25

Non-compacted and non-aggressive fills (sand, silt, clay, etc.)

0,18

0,70

1,20

1,70

2,20

Non-compacted and aggressive industrial fills (ash, slag, etc..)

0,50

2,00

3,25

4,50

5,75


Notes: 1. Corrosion rates in compacted fills are lower than those in non-compacted ones. In compacted fills, the figures in the table should be divided by two. 2. The values given for 5 and 25 years are based on measurement, whereas those for a greater number of years are extrapolated.

Table 1.2 advising corrosion losses in various waters for a range of design lives is reproduced from Table 4.2 of EN 1993-5: Steel piling. Table 11.2 Loss of thickness (mm) due to corrosion of steel piles in fresh waters or in seawater 5 years

Required design working life

25 years

50 years

75 years

100 years

Common fresh waters (river, ship canals, in zone of high attack (at water line)

0,15

0,55

0,90

1,15

1,40

Very polluted fresh waters (sewage, industrial effluent, etc.) in the zone of highest attack (water line)

0,30

1,30

2,30

3,30

4,30

Seawater in temperate climate in the zone of permanent immersion or intertidal zone

0,25

0,90

1,75

2,60

3,50

Seawater in temperate climate in the zone of highest attack (low water and splash zones)

0,55

1,90

3,75

5,60

7,50

Notes: 1) The highest corrosion rate is usually found in the splash zone or at the low water level in tidal waters. However, in most cases, the highest bending stresses occur within the permanent immersion zone. 2) The values given for 5 and 25 years are based on measurement, whereas those for a greater number of years are extrapolated.

There is no evidence to indicate that the load resistance of steel bearing piles buried in soil is at all affected by corrosion. In consideration of this, it is P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

98

Printed 07/09/05


relevant to know that the research on instrumented driven steel piles shows that the greatest proportion of load resistance is derived from the lowest 25% of pile length in the zone immediately above the pile tip, where the risk of corrosion is lowest. Any corrosion in the upper part of a buried bearing pile therefore only has a significant consequence to its potential load capacity if the loss of section causes the minimum specified yield stress to be exceeded at any time in its design life. To this end BS 6349[6] states that the working stresses should be based on the wall section remaining at the end of the design life and should not exceed the maximum permissible working stress given in BS 8002[5]. The effective life of steel piles will depend on the distribution and magnitude of anticipated stresses in the pile and the distribution and amount of predicted corrosion. The move to LSD procedures in the British Standards and Eurocodes, brings with it a rigour that is useful to the designer in selection of the pile type and its section in respect of corrosion. The limit state design procedures embody a requirement to select an appropriate design life or ‘effective life’ period for each structure. This design life provides the basis for the appraisal of risk of occurrence of the many situations that can affect the serviceability of the structure, including corrosion and accidental loading.


Up-to-date guidance on corrosion is available in the Corus brochure: A corrosion protection guide for steel bearing piles in temperate climates[94].

11.3 Corrosion in soil Corrosion in natural soils is electrochemical in nature and unless the soils are strongly acidic, i.e. pH<4, the corrosion of steel depends on the simultaneous presence of oxygen and water. In undisturbed natural soils, oxygen concentrations are very low at a very short distance below the ground surface. Steel corrosion rates are therefore very low in these circumstances, and are not related to the nature of the soil, its composition or its properties. The best evidence available is that given by Romanoff in his first report on steel piles[96] and supported by his later report[97] on further investigations. In the majority of cases reported by him, the piles were not painted before installation. The following is quoted from Romanoff’s summary in his first report : ‘The data indicate that the type and amount of corrosion observed on the steel piles driven into undisturbed natural soil, regardless of the soil characteristics and properties, is not sufficient to significantly affect the strength or useful life of piles as load-bearing structures.’ Also that: ‘.....Undisturbed soils are so deficient in oxygen at levels a few feet below the ground or below the water table zone, that steel piles are not appreciably affected by corrosion, regardless of the soil types or the soil properties. Properties of soils such as type, drainage, resistivity, pH or chemical composition are of no practical value in determining the corrosiveness of soil on steel piles driven underground.’ Ohsaki[98] evaluated the performance of piles driven into natural soil deposits at ten sites in Japan. The sites were spread widely over the whole country and selected where soil conditions were likely to be corrosive. The report P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

99

Printed 07/09/05


concluded that the average corrosion rate was approximately 0.005 mm/year/side and that the maximum corrosion rate measured was 0.015 mm/year/side. Corus plc has also examined extracted piles from sites in the United Kingdom ranging from beaches, river beds and harbours to inland sites, representing a wide range of soil types and conditions. The results obtained support the findings of Romanoff[97] and Ohsaki[98]. Guidance on corrosion allowances for piles in natural soils is given in Section 4.4.4 of BS 8002[5] where the maximum corrosion rate of 0.015 mm/year/side is advised and no other protection is required. This is within the range quoted in Section 10.3.5 of BS 8004[15].

11.4 Corrosion in fills and ‘brownfield’ sites


Structures are sometimes constructed in ‘brownfield’ sites where piles will be in contact of recent fill or ‘industrial soils’. The nature of fill soils can be variable and a careful assessment is required of how representative the samples are of the fill in situ before selecting samples for test. A full chemical analysis of soil and groundwater will be required to assess the likely corrosion performance of either steel or concrete piles in this environment in relation to the design life required. Soil tests should be in accordance with BS 1377-3[102] or as specified in CIRIA’s series of reports on contaminated land (contact CIRIA for further details). The need for corrosion protection of steel piles in contaminated soil can be assessed by first testing the fill for pH and resistivity, soluble salt content and porosity. The most relevant parameters to assessing the soil’s corrosivity are pH (a measure of the soil’s acidity/alkalinity i.e. the free hydrogen ion content); and resistivity (a measure of the total soluble ion content). In some cases, these tests are supplemented by chemical analysis for chlorides and soluble sulphates and measurements of redox potential, to assess the soil's capability of sustaining bacterial corrosion. Recent work by Corus[111] shows that these parameters can be used to classify a soil as aggressive or non-aggressive so that the appropriate corrosion rate can be selected from the range given in Tables 11.1 and 11.2 taken from EN 1993-5[7]. In a ‘controlled fill’ (i.e. selected granular fill, as referred to in Clause 3.8 of the Design Manual for Roads and Bridges, document BD 42[8]) no special measures are required, and the same corrosion rates as in natural undisturbed soils can be assumed. Further advice on corrosion assessment and protection can be obtained from The Steel Construction Institute or from Corus.


100

Printed 07/09/05


11.5 Atmospheric corrosion The rate of atmospheric corrosion can be taken to be an average of 0.035 mm/year/side, as recommended in BS 8002[5]. In areas of high industrial pollution, higher rates may be encountered, as given in BS 8004[15], but the incidence of these in the British Isles is diminishing as industry complies increasingly with the EC legislation to control pollution emissions to the atmosphere.

11.6 Corrosion below water Design allowances for the corrosion of steel piles wholly immersed in water are covered in BS 6349[6], BS 8002 and BS 8004. A mean corrosion rate of 0.035 mm/year/side, as advised in BS 8002 can be used for permanent immersion in seawater to calculate sacrificial allowances. Corrosion rates in fresh water are generally lower, and guidance is given in BS 8004.


11.6.1 Low water zone corrosion This narrow zone occurs at the bottom of the tidal range where a lack of marine growth is often observed and higher corrosion rates are often experienced, normally about 0.075 mm/year/side. Occasionally, corrosion rates which are significantly higher than this value arise because of specific local conditions, and it is recommended that periodic inspection of this zone on marine piles is undertaken. Investigations into this form of corrosion are reported in Accelerated low water corrosion of steel sheet pile marine walls[99].

11.6.2 Tidal zone corrosion Tidal zones tend to accumulate growths of barnacles and seaweeds, that afford protection to the steel, principally by limiting the supply of oxygen to the surface. Examination of piles in UK harbours indicates a mean corrosion rate of 0.035 mm/year/side, similar to that observed in the fully immersed zone.

11.6.3 Splash and marine atmospheric zones These zones are above the tidal range, the former being subject to wave action and salt spray and the latter mainly to airborne chlorides. Splash zone corrosion depends on the degree of shelter from wave action and thus is variable. Examination of structures in UK harbours produced a mean corrosion rate of 0.075 mm/year/side in the splash zone, though in extreme cases of exposure to full wave and storm conditions values up to 0.125 mm/year/side are possible. Upper range values are quoted in BS 6349 for various exposures. The boundary between the splash and atmospheric zones is not well defined; corrosion rates diminish rapidly with distance above peak wave height and the mean atmospheric corrosion rate of 0.035 mm/year/side can be used for this zone.

11.7 Methods of increasing effective life There are five main options to increase the effective life of a designed steel pile section. These are all covered in BS 8002, BS 6349 and the Corus [94] publication . The action to be taken will depend on individual circumstances


101

Printed 07/09/05


and cost-benefit analysis. The options are: •

Use a heavier section than the design requires.

•

Substitute a pile with a higher yield quality of steel.

•

Apply a protective coating.

•

Use concrete encasement where practicable.

•

Use cathodic protection, usually by an impressed current method on buried or fully immersed parts of the steel pile. (Cathodic protection is not effective in the splash or atmospheric zones).

The above remarks refer to steel of grades S275 and S355 to BS EN 10025-2[34], which are normally used for bearing piles.


In addition to the text in BS 6349 and BS 8002 there are some further points that can be made: •

The distribution and magnitude of anticipated stresses in a piled structure should always be related to the distribution and amount of corrosion because it frequently happens that the most corrosive splash zone is the least stressed.

•

The simplest and most economical way of allowing for corrosion loss is to use high yield quality steel instead of mild steel. A change from S275 to S355 steel gives an increase of about 30% on the useful life for a marginal increase in cost.

11.7.1 Protective coatings Where protection is to be used, the coating should be carefully specified, bearing in mind the comments already made, and it should be applied only to the vulnerable part of the pile. Suggested coatings are given in the Corus publication. Theoretically, maintenance painting is possible between tides, but its practicality will depend on local circumstances. Where protection is considered essential, the choice of a protective system, and its maintenance, will require careful consideration. Coatings are liable to be damaged in transit, in handling and pitching the piles on site, and in rubbing against temporary supports used during driving. Driving in certain types of soil e.g. gravels, can cause removal of some types of coatings, and alternatives should be used instead.

11.7.2 Cathodic protection Cathodic protection is a suitable measure that will be of value to steel which is buried or immersed below water level. It is not likely to be economically justifiable in most normal soil conditions unless the required design life is excessive. In this respect, the design life of 120 years required for highway structures in the UK is an extreme case that justifies special protective measures for both steel and reinforced concrete construction. The Highways Agency is already on record[100] as stating its belief in the merits of cathodic protection to reduce the costs of maintenance and repairs and thereby the whole life costs of its reinforced concrete bridges. BS 7361-1:1991[101] gives guidance on the design and selection of cathodic protection systems for land and marine applications. P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

102

Printed 07/09/05



REFERENCES 1.

CLARKE, J. et al Proceedings of the Conference ‘Recent large scale fully instrumented pile tests in clay’, London, 23-24 June 1992. Institution of Civil Engineers Thomas Telford, London, 1993

2.

BOND, A. J., HEIGHT, D. H. and JARDINE, R. J. Design of piles in sand in the UK sector of the North Sea Offshore Technology Report, OTH 94 457 HSE Books, 1997

3.

JARDINE, R. J. and CHOW, F. C. New Design Methods for Offshore Piles MTD Publication 96/103 Marine Technology Directorate, 1996

4.

Piling Handbook, Seventh Edition British Steel (now Corus), 1997

5.

BS 8002:1994: Code of practice for earth retaining structures British Standards Institution

6.

BS 6349-1:2000 Maritime structures. Code of practice for general criteria BS 6349-2:1988 Maritime structures. Design of quay walls, jetties and dolphins structures British Standards Institution

7.

EN 1993-5: 2004 Design of steel structures. Piling CEN (European Committee for Standardisation)

8.

BD 42/00 Design of embedded retaining walls and bridge abutments (unpropped or propped at the top) In Design manual for roads and bridges: Volume 2 Highway structures design: substructures The Stationery Office, 2000

9.

BS EN 1997-1:2004 Eurocode 7. Geotechnical design. General rules British Standards Institution

10. CORNFIELD, G. M. Steel bearing piles (Fourth Edition) The Steel Construction Institute, 1989 11. API Recommended practice 2A-LRFD (RP2A-LRFD), 1st Edition Recommended practice for planning, designing and constructing fixed offshore platforms - load and resistance factor design American Petroleum Institute, 1993 12. (ISO)13819-2: Petroleum and natural gas industries: offshore structures Part 2: Fixed steel structures International Organisation for Standardisation (ISO), 1995 13. ELSON, W. K. Design of laterally-loaded piles CIRIA Report 103 Construction Industry Research and Information Association, 1984 P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

103

Printed 07/09/05


14. PADFIELD, C. J. and MAIR, R. J. Design of retaining walls embedded in stiff clay CIRIA Report 104 Construction Industry Research and Information Association, 1984 15. BS 8004:1986 Code of practice for foundations British Standards Institution 16. BS 8081:1989 Code of practice for ground anchorages British Standards Institution 17. BS 8006:1995 Code of practice for strengthened/reinforced soils and other fills British Standards Institution 18. WAY, J. A. and BIDDLE, A. R. Integral Steel Bridges: Design of a multi-span bridge – Worked Example (P250) The Steel Construction Institute 1998


19. BS 449: Specification for the use of structural steel in building British Standards Institution 20. INSTITUTION OF CIVIL ENGINEERS Specification for piling and embedded retaining walls (Specification, contract documentation, and guidance notes) ICE; Highways Agency; Federation of Piling Specialists; Ove Arup & Partners Thomas Telford, 1996 21. WALLACE, G. and GANNON, J. CIRIA Report R181: Piled foundations in weak rock Construction Industry Research and Information Association 1997 22. BIDDLE, A. R. and WYLD, M. Technical Report RT 552: Validation of vertical load capacity prediction methods for steel bearing piles The Steel Construction Institute, 1995 23. POULOS, H. J. and DAVIS, E. H. Pile foundation analysis and design John Wiley and Sons, 1980 24. TOMLINSON, M. J. Pile design and construction practice (Fourth Edition) E & FN Spon, 1994 25. TOMLINSON, M. J. Adhesion of piles in stiff clays CIRIA Report No.26 Construction Industry Research and Information Association, 1970 26. BS 1377-9:1990 Methods of test for soils for civil engineering purposes. In-situ tests Section 3.1: Static cone penetration test (CPT) Section 3.3: The standard penetration test (SPT) British Standards Institution 27. GRLWEAP 1997-2 Wave equation software program for pile driveability prediction Goble, Rausche, Likins & Associates, Cleveland, Ohio, 1997


104

Printed 07/09/05


28. HILEY, A. Pile driving calculations with notes on driving forces and ground resistances The Structural Engineer, London, Vol. 8, pp 246-259, 278-288, 1930 29. BIDDLE, A. R. and WYLD, M. Technical report RT 621: Validation of steel axial pile capacity prediction methods: Statistical analysis of results and load tests in rock The Steel Construction Institute, 1995 30. INSTITUTION OF STRUCTURAL ENGINEERS CP2: 1951: Earth retaining structures ISE, 1951 31. BIDDLE, A. R., ILES, D. C. and YANDZIO, E. Integral steel bridges: design guidance (P163) The Steel Construction Institute, 1997 32. WAY, J. A. and YANDZIO, E. Integral steel bridges: design of a single-span bridge - worked example (P180) The Steel Construction Institute, 1997


33. TOMLINSON, M. J. The adhesion of piles driven in clay Proc. 4th Int. Conf. on Soil Mech. and Found. Eng., Vol. 2, pp 66-71 34. BS EN 10025-2:2004 Hot rolled products of structural steels. Technical delivery conditions for non-alloy structural steels British Standards Institution 35. API 5L: Specifications for steel line pipe American Petroleum Institute, 1995 36. TESPA Case Studies Technical European Sheet Piling Association (TESPA), 1991-1993 37. BS 6235: 1982: Code of practice for fixed offshore structures (and draft revision 1985, since withdrawn) British Standards Institution 38. CLAYTON, C. R. I. The Standard Penetration test (SPT): methods and use CIRIA Report 143, 1995 39. BS 1377-9:1990 Methods of test for soils for civil engineering purposes. In-situ tests Section 3.1: Static cone penetration test (CPT) British Standards Institution 40. MEIGH, A.C. Cone Penetration Testing (CPT): Methods and interpretation CIRIA/Butterworths, 1987 41. BRIAUD, J. L., FELIO, G. and TUCKER, L. Influence of cyclic loading on axially loaded piles in clay Research Report for Phase 2, and BRIAUD, J. L., and GARLAND, E. Influence of loading rate on axially loaded piles in clay Research report for Phase 1; both in PRAC 83-42 entitled: Pile Response to Static and Dynamic loads American Petroleum Institute, 1984 P:\Pub\Pub800\Sign_off\P335\P335V01D12.doc

105

Printed 07/09/05


42. GEOTECHNICAL ENGINEERING OFFICE Guide to retaining wall design Civil Engineering Department, Hong Kong 43. HOBBS, N. B. and HEALY, P. R. Piling in chalk CIRIA report PG6, 1979 44. LORD, J. A., TWINE, D. and YEOW, H. Foundations in chalk CIRIA Project Report 11, 1994 45. OASYS Geotechnical suite of programs for analysis and design ALP - Laterally loaded pile analysis Ove Arup and Partners (London), 1991 46. MATLOCK, H. OTC Paper 1204: Correlations for design of laterally loaded piles in soft clay Proc. Second Offshore Technology Conference, Houston, Texas, 1970


47. REESE, L. C., COX, W. R. and KOOP, F.D. OTC Paper 2080: Analysis of laterally loaded piles in sand Proc. Sixth Offshore Technology Conference, Houston, 1974 48. REESE, L. C., COX, W. R. and KOOP, F. D. OTC Paper 2312: Field testing and analysis of laterally loaded piles in stiff clay Proc. Seventh Offshore Technology Conference, Houston, 1975 49. O’NEIL, M. W., GHAZZALY, O. I. and HA, H. B. OTC paper 2838: Analysis of three dimensional pile groups with non-linear soil response and pile-soil interaction Proc. Ninth Offshore Technology Conference, Houston, 1977 50. POULOS, H. G. Behaviour of laterally-loaded piles: pile groups Proc. Am. Soc. Civ. Engrs. - J. Soil. Found. Div. May 1971 97 (SM5), 733 to 751 51. POULOS, H. G. Design of piled foundations Research Report 271, Univ. of Sydney, School of Eng. 1975 52. RANDOLPH, M. F. A theoretical study of the performance of piles PhD Thesis, Cambridge University, 1977 53. POULOS, H. G Lateral load deflection prediction for pile groups Proc. Am. Soc. Civ. Engrs. - J. Geotech. Engng. Div. Jan 1975, 101 (GTI), 19 to 33 54. BANNERJEE, P. K. and DRISCOLL, R. M. Program for the analysis of pile groups of any geometry subjected to horizontal and vertical loads and moments HECB/B/7-PGROUP DoT, Highway Engineering Computer Branch, 1975


106

Printed 07/09/05


55. GEOCENTRIX LTD Repute 1.0 – Pile Group Design Program, available from www.geocentrix.co.uk 2002 56. RANDOLPH, M. F. PIGLET - a computer program for the analysis and design of laterally loaded piles under general loading conditions Cambridge University Research Report CUED/D - Soils TR 91, 1980 57. AWKO, F. Simplified approach to the analysis of piling systems Structural Engineer Mar. 1968 46(3), 83 to 86 58. INSTITUTION OF STRUCTURAL ENGINEERS Earth retaining structures Civil Engineering Code of Practice 2, 1951 59. MATLOCK, H. and REESE, L.C. Generalised solutions for laterally loaded piles Proc. Am. Soc. Civ. Engrs. - J. Soil Mech. Found. Div. Oct. 1960 86(SM5), 63 to 91


60. POULOS, H. G. Behaviour of laterally loaded piles: single piles Proc. Am. Soc. Civ. Engrs. - J. Soil Mech. Found. Div. May 1971 97(SM5), 711 to 731 61. BUTTERFIELD, R. and DOUGLAS, R. A. Flexibility coefficients for the design of piles and pile groups CIRIA Technical Note 108, 1981 62. BASILE, F Non-linear analysis of pile groups Proceedings of the Institution of Civil Engineers, Geotechnical Engineering, Vol. 137, No. 2, April, pp 105-115, 1999 63. Installation of steel sheet piles Technical European Sheet Piling Association (TESPA), 1995 64. Euripides Pile Test Program Tubular steel pile load tests in sand, 1994-1996 Joint Venture Project Reports Fugro-McClelland Engineers B.V. and Geodia S.A. 65. AKBARI, N. A. and MURE, A. Investigation of methods of prediction and measurement of the behaviour of three types of driven piles at Isle of Grain, UK. Paper 21 Piling practice and worldwide trends, edited by M. J. Sands Institution of Civil Engineers, Thomas Telford, 1992 66. WHITAKER, T. The design of piled foundations, 2nd Edition Pergamon Press, Oxford 67. FLEMING, W. G. K., WELTMAN, A. J., RANDOLPH, M. F. and ELSON, W.K. Piling Engineering. 2nd Edition Blackie A and P, 1994


107

Printed 07/09/05


68. FLAATE, K. An investigation of the validity of three pile driving formula in cohesionless material NGI Publication No. 56, pp 11-22 Norwegian Geotechnical Institute, Oslo, 1964 69. CAPWAP. CAse Pile Wave Analysis Program Pile Dynamics Inc., Cleveland, Ohio 70. BS EN 12699:2001 Execution of special geotechnical work. Displacement British Standards Institution 2001 71. SMITH, I. M. and CHOW, Y. K. Three dimensional analysis of pile driveability Proc. 2nd Int. Conf. on Numerical Methods in Offshore Piling, Austin, Texas USA pp 1-10, 1982


72. GOBLE, G. G. and RAUSCHE, F. Wave equation analysis of pile driving - WEAP Program Vol. 1 - Background, Vol. 2 - User’s Manual , Vol. 3 - Program documentation, Report No. FHWA - IP - 76-14.1, 2, 3, National Technical Information Service, Springfield, Virginia, July 1976 73. FLEMING, W. G. K. A new method for single pile settlement prediction and analysis Geotechnique, vol 42, No. 3, 411-425, 1992 74. ENGLAND, M. Pile settlement behaviour: An accurate model Proc. Conference on application of stress-wave theory to piles, A.A. Balkema Publishers, Rotterdam 1992 75. ENGLAND, M. Paper 3-14: The role of driven pile instrumentation Deep Foundations Institute Conference, 1994, Belgium Deep Foundations Institute, New Jersey 07632, USA 76. ENGLAND, M. and FLEMING, W.G.K. Review of foundation testing methods and procedures Proceedings of the Institution of Civil Engineers, Geotechnical Engineeering 1994, 107, July, 135-142 77. Proceedings of the International Conferences on the application of stress wave theory on piles Stockholm 1980, 1984; Ottawa, 1988; The Hague, Holland, 1992; Orlando, Florida, 1996 A.A. Balkema publishers, Rotterdam 78. BIDDLE A. R., and YANDZIO E. Specifiers guide to steel piling (P308) The Steel Construction Institute, 2002 79. HMSO Chapter 40 United Kingdom Control of Pollution Act 1974 80. BS 5228-4:1992 Noise and vibration control on construction and open sites. Code of practice for noise and vibration control applicable to piling operations British Standards Institution


108

Printed 07/09/05


81. TRRL Report RR53 Ground vibration caused by civil engineering works Transport and Road Research Laboratory, 1986 82. Control of vibration and noise during piling. Legislation and practice on noise and vibration control with particular reference to steel piling British Steel Publication 105A, 1994, and later edition British Steel Sections, Plates & Commercial Steels (now Corus C&I), 1997 83. BS 6472:1992 Guide to evaluation of human exposure to vibration in buildings (1 Hz to 80 Hz) British Standards Institution 84. BEAMAN, A. C. and JONES, R. D. Noise from construction and demolition sites - measured levels and their prediction CIRIA Report 64, London, 1977 85. WELTMAN, A. J. Noise and vibrations from piling operations DOE and CIRIA Piling Development Group CIRIA Report No. PG9, Oct. 1980


86. HEAD, J. M. and JARDINE, F. M. Ground-borne vibrations arising from piling CIRIA Technical Note 142, 1992 87. SKIPP, B. O. Dynamic ground movements - man-made vibrations in ground movements and their effects on structures Blackie, Glasgow and London, pp 381-434 88. BRE Digest No 403 Damage to structures from ground-borne vibration Building Research Establishment, 1995 89. DIN 4150: Part 3: Structural vibration in buildings: effects on structures DIN (Deutsches Institut Für Normung) DIN, 1986 (draft 1997 not yet translated into English) 90. BD 32/96 Design of laterally loaded piles In Design manual for roads and bridges The Stationery Office, 1996 91. Universal bearing piles A cost and performance study Harris and Sutherland Consulting Engineers, London, 1977 92. ROSCOE, H. and TWINE, D. Design collaboration speeds Ashford Tunnels World Tunnelling,Vol.14 No.5, June 2001 93. THE BRITISH GEOTECHNICAL SOCIETY ‘Sheet piles stand the test of time’ Ground Engineering Volume 40, Number 4, p4, May 1997 94. A corrosion protection guide for steel bearing piles in temperate climates Corus Construction and Industrial, 2004


109

Printed 07/09/05


95. The prevention of corrosion on structural steelwork British Steel Brochure reference SPCS 501 3 5/96 British Steel Sections, Plates & Commercial Steels (now Corus C&I), May 1996 96. ROMANOFF, M. Underground corrosion National Bureau of Standards, circular no. 579: 1957 US Dept. of Commerce, Washington DC 97. ROMANOFF, M. Corrosion of steel piling in soil National Bureau of Standards, monograph no. 58: 1962 US Dept. of Commerce, Washington DC 98. OHSAKI, Y. Corrosion of steel piles driven in soil deposits Soils and Foundations Journal, Vol. 22, No. 3: Sept. 1982 Japanese Society of Soil Mechanics and Foundation Engineering 99. ROWBOTTOM, D. Accelerated low water corrosion of steel sheet pile marine walls British Steel Piling Technical, Scunthorpe, 1996


100. HIGHWAYS AGENCY Quote from Lawrie Haynes, Director, 1996 101. BS 7361-1:1991 Cathodic protection. Code of practice for land and marine applications British Standards Institution 102. BS 1377-1:1990 Methods of test for soils for civil engineering purposes. General requirements and sample preparation British Standards Institution 103. GRIFFITHS, S., FILIP, R., and BIDDLE, A. R. An innovative integral bridge on stilts at Whaddon Road, Milton Keynes New Civil Engineer November 2003 104. BD 74: Foundations In Design manual for roads and bridges: Volume 2 The Stationery Office, 1996 105. FEDERATION OF PILING SPECIALISTS (FPS) The Essential Guide to the ICE Specification for Piling and Embedded Retaining Walls Thomas Telford Publishing, 1999 106. TOMLINSON M. J., SHERRELL, F.,W. and GEORGE, A.,B. The behaviour of steel H-piles in a slatey mudstone CIRIA Report 066 May 1976, also paper to Geotechnique March 1976, and Proceedings ICE Conference on Piles in Weak Rock, London ,1977 107. CORUS CONSTRUCTION & INDUSTRIAL Supporting the commercial decision – comparing the cost of steel and concrete framing options for commercial buildings; Facts of living – comparing the cost of steel and concrete framing options for multi-storey residential buildings; and Mind the competitive gap – value benefits of steel construction. Corus Construction and Industrial, Scunthorpe, 2004


110

Printed 07/09/05


108. LUNNE T., ROBERTSON, P. K. and POWELL, J. J. M. Cone penetration in geotechnical practice E.& F.N. Spon, London, 1977 109. YANDZIO, E., and BIDDLE, A. R. Steel Intensive basements (P275) The Steel Construction Institute, 2001 110. GORGOLEWSKI, M. Environmental assessment of steel piling (P199) The Steel Construction Institute, 1999 111. CORUS CONSTRUCTION & INDUSTRIAL The durability of steel in fill soils and contaminated land. Report No. STC/CPR OCP/CKR/0964/2004/R. Corus Construction and Industrial, Scunthorpe, 2004 112. JARDINE, R. J., CHOW, F. C., OVERY, R. F, and STANDING,J. R. ICP design methods for driven piles in sands and clays Imperial College , London Thomas Telford 2005


113. FRANK, R., BAUDUIN, C., DRISCOLL, R., KAVVADAS, M., KREBS OVESEN, N., ORR, T., and SCHUPPENER, B. Designers’ guide to EN 1997-1 Thomas Telford, London, 2004 114. FELLENIUS,B. H., RIKER.R.E., O’BRIEN. A. J., and TRACY, G. R. Dynamic and static testing on steel piling in a soil exhibiting set-up Journal of Geotechnical Engineering, Vol 115, No 7, pp 984-1001 American Society of Civil Engineers, 1989 115. KOMURKA, V. E., Incorporating set-up and support cost distribution into driven pile design Current practices and future trends in deep foundations Geotechnical Special Publication No.125, 0-7844-0743-6, pp 16-49 American Society of Civil Engineers 2004 116. GOBLE, G. G., RAUSCHE, F. and LIKINS, G. E. Jr. The analysis of pile driving – A state-of-the-art In Proceedings of the International Seminar on the Application of StressWave Theory on Piles, Stockholm, 4-5 June 1980 A. A. Balkema, Rotterdam, 1981 117. LIKINS, G, RAUSCHE, F., THENDEAN, G. AND SVINKIN, M. CAPWAP correlation studies In 5th International Conference on the Application of Stress Wave Theory of Piles, Orlando, Sept 1996 University of Florida, 1996


111

Printed 07/09/05



Discuss me ...


112

Printed 07/09/05


APPENDIX A

CONTACTS

Corus Group plc Construction & Industrial PO Box1 Brigg Road Scunthorpe North Lincolnshire DN16 1BP. Tel: 01724 404040, Fax: 01724 405564 Federation of Piling Specialists (FPS) 39 Upper Elmers End Road Beckenham Kent BR3 3QY Tel: 0181 663 0947, Fax: 0181 663 0949


Technical European Sheet Piling Association (TESPA) PO Box 8413 D-4000 Dusseldorf 1

and 19 Avenue de la Liberté L-2930 Luxembourg The Steel Construction Institute Silwood Park Ascot SL5 7QN Tel: 01344 623345, 01344 622944

Additional contact information for steel piling contractors and suppliers is given in the Steel Construction Yearbook, available from the SCI. Also, visit www.steelbiz.org


113

Printed 07/09/05

H Pile Design Guide

Recommend Documents