Bridge Engineering Handbook

SECOND EDITION

SUBSTRUCTURE
DESIGN
EDITED BY
Wai-Fah Chen and Lian Duan
Bridge Engineering Handbook
SECOND EDITION

substructure
design
 Bridge Engineering Handbook, Second Edition

Bridge Engineering Handbook, Second Edition: Fundamentals

Bridge Engineering Handbook, Second Edition: Superstructure Design

Bridge Engineering Handbook, Second Edition: Substructure Design

Bridge Engineering Handbook, Second Edition: Seismic Design

Bridge Engineering Handbook, Second Edition: Construction and Maintenance

Bridge Engineering Handbook
SECOND EDITION

substructure
design
Edited by
Wai-Fah Chen and Lian Duan

Boca Raton London New York

CRC Press is an imprint of the

Taylor & Francis Group, an informa business
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
© 2014 by Taylor & Francis Group, LLC
CRC Press is an imprint of Taylor & Francis Group, an Informa business

No claim to original U.S. Government works

Version Date: 20130923

International Standard Book Number-13: 978-1-4398-5230-9 (eBook - PDF)

This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been
made to publish reliable data and information, but the author and publisher cannot assume responsibility for the valid-
ity of all materials or the consequences of their use. The authors and publishers have attempted to trace the copyright
holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this
form has not been obtained. If any copyright material has not been acknowledged please write and let us know so we may
rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or uti-
lized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopy-
ing, microfilming, and recording, or in any information storage or retrieval system, without written permission from the
publishers.

For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://
www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923,
978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For
organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for
identification and explanation without intent to infringe.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com
and the CRC Press Web site at
http://www.crcpress.com
 Contents

Foreword ..................................................................................................................vii
Preface to the Second Edition .................................................................................. ix
Preface to the First Edition ...................................................................................... xi
Editors .................................................................................................................... xiii
Contributors ............................................................................................................ xv

1 Bearings .............................................................................................................................. 1
Ralph J. Dornsife
2 Piers and Columns ......................................................................................................... 35
Jinrong Wang
3 Towers ................................................................................................................................ 63
Charles Seim and Jason Fan
4 Vessel Collision Design of Bridges ............................................................................ 89
Michael Knott and Zolan Prucz
5 Bridge Scour Design and Protection ........................................................................113
Junke Guo
6 Abutments .......................................................................................................................133
Linan Wang
7 Ground Investigation ...................................................................................................155
homas W. McNeilan and Kevin R. Smith
8 Shallow Foundations ....................................................................................................181
Mohammed S. Islam and Amir M. Malek
9 Deep Foundations ........................................................................................................ 239
Youzhi Ma and Nan Deng

v
vi Contents

10 Earth Retaining Structures ........................................................................................ 283

Chao Gong
11 Landslide Risk Assessment and Mitigation ...........................................................315
Mihail E. Popescu and Aurelian C. Trandair
 Foreword

hroughout the history of civilization bridges have been the icons of cities, regions, and countries. All
bridges are useful for transportation, commerce, and war. Bridges are necessary for civilization to exist,
and many bridges are beautiful. A few have become the symbols of the best, noblest, and most beautiful
that mankind has achieved. he secrets of the design and construction of the ancient bridges have been
lost, but how could one not marvel at the magniicence, for example, of the Roman viaducts?
he second edition of the Bridge Engineering Handbook expands and updates the previous edition
by including the new developments of the irst decade of the twenty-irst century. Modern bridge
engineering has its roots in the nineteenth century, when wrought iron, steel, and reinforced concrete
began to compete with timber, stone, and brick bridges. By the beginning of World War II, the
transportation infrastructure of Europe and North America was essentially complete, and it served to
sustain civilization as we know it. he iconic bridge symbols of modern cities were in place: Golden Gate
Bridge of San Francisco, Brooklyn Bridge, London Bridge, Eads Bridge of St. Louis, and the bridges of
Paris, Lisbon, and the bridges on the Rhine and the Danube. Budapest, my birthplace, had seven beauti-
ful bridges across the Danube. Bridge engineering had reached its golden age, and what more and better
could be attained than that which was already achieved?
hen came World War II, and most bridges on the European continent were destroyed. All seven
bridges of Budapest were blown apart by January 1945. Bridge engineers ater the war were suddenly
forced to start to rebuild with scant resources and with open minds. A renaissance of bridge engineering
started in Europe, then spreading to America, Japan, China, and advancing to who knows where in
the world, maybe Siberia, Africa? It just keeps going! he past 60 years of bridge engineering have
brought us many new forms of bridge architecture (plate girder bridges, cable stayed bridges, segmen-
tal prestressed concrete bridges, composite bridges), and longer spans. Meanwhile enormous knowl-
edge and experience have been amassed by the profession, and progress has beneitted greatly by the
availability of the digital computer. he purpose of the Bridge Engineering Handbook is to bring much of
this knowledge and experience to the bridge engineering community of the world. he contents encom-
pass the whole spectrum of the life cycle of the bridge, from conception to demolition.
he editors have convinced 146 experts from many parts of the world to contribute their knowledge
and to share the secrets of their successful and unsuccessful experiences. Despite all that is known, there
are still failures: engineers are human, they make errors; nature is capricious, it brings unexpected sur-
prises! But bridge engineers learn from failures, and even errors help to foster progress.
he Bridge Engineering Handbook, second edition consists of ive books:
Fundamentals
Superstructure Design
Substructure Design
Seismic Design
Construction and Maintenance

vii
viii Foreword

Fundamentals, Superstructure Design, and Substructure Design present the many topics necessary
for planning and designing modern bridges of all types, made of many kinds of materials and systems,
and subject to the typical loads and environmental efects. Seismic Design and Construction and
Maintenance recognize the importance that bridges in parts of the world where there is a chance of
earthquake occurrences must survive such an event, and that they need inspection, maintenance, and
possible repair throughout their intended life span. Seismic events require that a bridge sustain repeated
dynamic load cycles without functional failure because it must be part of the postearthquake lifeline for
the afected area. Construction and Maintenance touches on the many very important aspects of bridge
management that become more and more important as the world’s bridge inventory ages.
he editors of the Bridge Engineering Handbook, Second Edition are to be highly commended for
undertaking this efort for the beneit of the world’s bridge engineers. he enduring result will be a safer
and more cost efective family of bridges and bridge systems. I thank them for their efort, and I also
thank the 146 contributors.

heodore V. Galambos, PE
Emeritus professor of structural engineering
University of Minnesota
 Preface to the
Second Edition

In the approximately 13 years since the original edition of the Bridge Engineering Handbook was
published in 2000, we have received numerous letters, e-mails, and reviews from readers including
educators and practitioners commenting on the handbook and suggesting how it could be improved. We
have also built up a large ile of ideas based on our own experiences. With the aid of all this information,
we have completely revised and updated the handbook. In writing this Preface to the Second Edition,
we assume readers have read the original Preface. Following its tradition, the second edition handbook
stresses professional applications and practical solutions; describes the basic concepts and assumptions
omitting the derivations of formulas and theories; emphasizes seismic design, rehabilitation, retroit
and maintenance; covers traditional and new, innovative practices; provides over 2500 tables, charts,
and illustrations in ready-to-use format and an abundance of worked-out examples giving readers step-
by-step design procedures. he most signiicant changes in this second edition are as follows:
• he handbook of 89 chapters is published in ive books: Fundamentals, Superstructure Design,
Substructure Design, Seismic Design, and Construction and Maintenance.
• Fundamentals, with 22 chapters, combines Section I, Fundamentals, and Section VI, Special Topics,
of the original edition and covers the basic concepts, theory and special topics of bridge engi-
neering. Seven new chapters are Finite Element Method, High-Speed Railway Bridges, Structural
Performance Indicators for Bridges, Concrete Design, Steel Design, High Performance Steel, and
Design and Damage Evaluation Methods for Reinforced Concrete Beams under Impact Loading.
hree chapters including Conceptual Design, Bridge Aesthetics: Achieving Structural Art in
Bridge Design, and Application of Fiber Reinforced Polymers in Bridges, are completely rewritten.
hree special topic chapters, Weigh-In-Motion Measurement of Trucks on Bridges, Impact Efect
of Moving Vehicles, and Active Control on Bridge Engineering, were deleted.
• Superstructure Design, with 19 chapters, provides information on how to design all types of bridges.
Two new chapters are Extradosed Bridges and Stress Ribbon Pedestrian Bridges. he Prestressed
Concrete Girder Bridges chapter is completely rewritten into two chapters: Precast–Pretensioned
Concrete Girder Bridges and Cast-In-Place Posttensioned Prestressed Concrete Girder Bridges.
he Bridge Decks and Approach Slabs chapter is completely rewritten into two chapters: Concrete
Decks and Approach Slabs. Seven chapters, including Segmental Concrete Bridges, Composite
Steel I-Girder Bridges, Composite Steel Box Girder Bridges, Arch Bridges, Cable-Stayed Bridges,
Orthotropic Steel Decks, and Railings, are completely rewritten. he chapter Reinforced Concrete
Girder Bridges was deleted because it is rarely used in modern time.
• Substructure Design has 11 chapters and addresses the various substructure components. A new
chapter, Landslide Risk Assessment and Mitigation, is added. he Geotechnical Consideration
chapter is completely rewritten and retitled as Ground Investigation. he Abutments and

ix
x Preface to the Second Edition

Retaining Structures chapter is divided in two and updated as two chapters: Abutments and Earth
Retaining Structures.
• Seismic Design, with 18 chapters, presents the latest in seismic bridge analysis and design. New
chapters include Seismic Random Response Analysis, Displacement-Based Seismic Design of
Bridges, Seismic Design of hin-Walled Steel and CFT Piers, Seismic Design of Cable-Supported
Bridges, and three chapters covering Seismic Design Practice in California, China, and Italy. Two
chapters of Earthquake Damage to Bridges and Seismic Design of Concrete Bridges have been
rewritten. Two chapters of Seismic Design Philosophies and Performance-Based Design Criteria,
and Seismic Isolation and Supplemental Energy Dissipation, have also been completely rewritten
and retitled as Seismic Bridge Design Speciications for the United States, and Seismic Isolation
Design for Bridges, respectively. Two chapters covering Seismic Retroit Practice and Seismic
Retroit Technology are combined into one chapter called Seismic Retroit Technology.
• Construction and Maintenance has 19 chapters and focuses on the practical issues of bridge
structures. Nine new chapters are Steel Bridge Fabrication, Cable-Supported Bridge Construction,
Accelerated Bridge Construction, Bridge Management Using Pontis and Improved Concepts,
Bridge Maintenance, Bridge Health Monitoring, Nondestructive Evaluation Methods for
Bridge Elements, Life-Cycle Performance Analysis and Optimization, and Bridge Construction
Methods. he Strengthening and Rehabilitation chapter is completely rewritten as two chap-
ters: Rehabilitation and Strengthening of Highway Bridge Superstructures, and Rehabilitation
and Strengthening of Orthotropic Steel Bridge Decks. he Maintenance Inspection and Rating
chapter is completely rewritten as three chapters: Bridge Inspection, Steel Bridge Evaluation and
Rating, and Concrete Bridge Evaluation and Rating.
• he section on Worldwide Practice in the original edition has been deleted, including the chapters
on Design Practice in China, Europe, Japan, Russia, and the United States. An international team
of bridge experts from 26 countries and areas in Africa, Asia, Europe, North America, and South
America, has joined forces to produce the Handbook of International Bridge Engineering, Second
Edition, the irst comprehensive, and up-to-date resource book covering the state-of-the-practice
in bridge engineering around the world. Each of the 26 country chapters presents that country’s
historical sketch; design speciications; and various types of bridges including girder, truss, arch,
cable-stayed, suspension, and so on, in various types of materials—stone, timber, concrete, steel,
advanced composite, and of varying purposes—highway, railway, and pedestrian. Ten bench-
mark highway composite girder designs, the highest bridges, the top 100 longest bridges, and
the top 20 longest bridge spans for various bridge types are presented. More than 1650 beautiful
bridge photos are provided to illustrate great achievements of engineering professions.
he 146 bridge experts contributing to these books have written chapters to cover the latest bridge
engineering practices, as well as research and development from North America, Europe, and Paciic
Rim countries. More than 80% of the contributors are practicing bridge engineers. In general, the hand-
book is aimed toward the needs of practicing engineers, but materials may be re-organized to accom-
modate several bridge courses at the undergraduate and graduate levels.
he authors acknowledge with thanks the comments, suggestions, and recommendations made
during the development of the second edition of the handbook by Dr. Erik Yding Andersen, COWI
A/S, Denmark; Michael J. Abrahams, Parsons Brinckerhof, Inc.; Dr. Xiaohua Cheng, New Jersey
Department of Transportation; Joyce E. Copelan, California Department of Transportation; Prof. Dan
M. Frangopol, Lehigh University; Dr. John M. Kulicki, Modjeski and Masters; Dr. Amir M. Malek,
California Department of Transportation; Teddy S. heryo, Parsons Brinckerhof, Inc.; Prof. Shouji
Toma, Horrai-Gakuen University, Japan; Dr. Larry Wu, California Department of Transportation; Prof.
Eiki Yamaguchi, Kyushu Institute of Technology, Japan; and Dr. Yi Edward Zhou, URS Corp.
We thank all the contributors for their contributions and also acknowledge Joseph Clements, acquiring
editor; Jennifer Ahringer, project coordinator; and Joette Lynch, project editor, at Taylor & Francis/CRC Press.
 Preface to the
First Edition

he Bridge Engineering Handbook is a unique, comprehensive, and state-of-the-art reference work and
resource book covering the major areas of bridge engineering with the theme “bridge to the twenty-irst
century.” It has been written with practicing bridge and structural engineers in mind. he ideal readers
will be MS-level structural and bridge engineers with a need for a single reference source to keep abreast
of new developments and the state-of-the-practice, as well as to review standard practices.
he areas of bridge engineering include planning, analysis and design, construction, maintenance,
and rehabilitation. To provide engineers a well-organized, user-friendly, and easy-to-follow resource,
the handbook is divided into seven sections. Section I, Fundamentals, presents conceptual design,
aesthetics, planning, design philosophies, bridge loads, structural analysis, and modeling. Section II,
Superstructure Design, reviews how to design various bridges made of concrete, steel, steel-concrete
composites, and timbers; horizontally curved, truss, arch, cable-stayed, suspension, loating, movable,
and railroad bridges; and expansion joints, deck systems, and approach slabs. Section III, Substructure
Design, addresses the various substructure components: bearings, piers and columns, towers, abut-
ments and retaining structures, geotechnical considerations, footings, and foundations. Section IV,
Seismic Design, provides earthquake geotechnical and damage considerations, seismic analysis and
design, seismic isolation and energy dissipation, soil–structure–foundation interactions, and seismic
retroit technology and practice. Section V, Construction and Maintenance, includes construction of
steel and concrete bridges, substructures of major overwater bridges, construction inspections, main-
tenance inspection and rating, strengthening, and rehabilitation. Section VI, Special Topics, addresses
in-depth treatments of some important topics and their recent developments in bridge engineering.
Section VII, Worldwide Practice, provides the global picture of bridge engineering history and practice
from China, Europe, Japan, and Russia to the U.S.
he handbook stresses professional applications and practical solutions. Emphasis has been placed
on ready-to-use materials, and special attention is given to rehabilitation, retroit, and maintenance.
he handbook contains many formulas and tables that give immediate answers to questions arising
from practical works. It describes the basic concepts and assumptions, omitting the derivations of
formulas and theories, and covers both traditional and new, innovative practices. An overview of the
structure, organization, and contents of the book can be seen by examining the table of contents pre-
sented at the beginning, while the individual table of contents preceding each chapter provides an
in-depth view of a particular subject. References at the end of each chapter can be consulted for more
detailed studies.
Many internationally known authors have written the chapters from diferent countries covering
bridge engineering practices, research, and development in North America, Europe, and the Paciic
Rim. his handbook may provide a glimpse of a rapidly growing trend in global economy in recent
years toward international outsourcing of practice and competition in all dimensions of engineering.

xi
xii Preface to the First Edition

In general, the handbook is aimed toward the needs of practicing engineers, but materials may be
reorganized to accommodate undergraduate and graduate level bridge courses. he book may also be
used as a survey of the practice of bridge engineering around the world.
he authors acknowledge with thanks the comments, suggestions, and recommendations during the
development of the handbook by Fritz Leonhardt, Professor Emeritus, Stuttgart University, Germany;
Shouji Toma, Professor, Horrai-Gakuen University, Japan; Gerard F. Fox, Consulting Engineer; Jackson
L. Durkee, Consulting Engineer; Michael J. Abrahams, Senior Vice President, Parsons, Brinckerhof,
Quade & Douglas, Inc.; Ben C. Gerwick, Jr., Professor Emeritus, University of California at Berkeley;
Gregory F. Fenves, Professor, University of California at Berkeley; John M. Kulicki, President and Chief
Engineer, Modjeski and Masters; James Chai, Senior Materials and Research Engineer, California
Department of Transportation; Jinrong Wang, Senior Bridge Engineer, URS Greiner; and David W. Liu,
Principal, Imbsen & Associates, Inc.
We thank all the authors for their contributions and also acknowledge at CRC Press Nora Konopka,
acquiring editor, and Carol Whitehead and Sylvia Wood, project editors.
 Editors

Dr. Wai-Fah Chen is a research professor of civil engineering at the

University of Hawaii. He was dean of the College of Engineering at
the University of Hawaii from 1999 to 2007, and a George E. Goodwin
Distinguished Professor of Civil Engineering and head of the Department
of Structural Engineering at Purdue University from 1976 to 1999.
He earned his BS in civil engineering from the National Cheng-Kung
University, Taiwan, in 1959, MS in structural engineering from Lehigh
University in 1963, and PhD in solid mechanics from Brown University
in 1966. He received the Distinguished Alumnus Award from the
National Cheng-Kung University in 1988 and the Distinguished
Engineering Alumnus Medal from Brown University in 1999.
Dr. Chen’s research interests cover several areas, including consti-
tutive modeling of engineering materials, soil and concrete plasticity,
structural connections, and structural stability. He is the recipient of several national engineering
awards, including the Raymond Reese Research Prize and the Shortridge Hardesty Award, both from
the American Society of Civil Engineers, and the T. R. Higgins Lectureship Award in 1985 and the
Lifetime Achievement Award, both from the American Institute of Steel Construction. In 1995, he was
elected to the U.S. National Academy of Engineering. In 1997, he was awarded Honorary Membership
by the American Society of Civil Engineers, and in 1998, he was elected to the Academia Sinica
(National Academy of Science) in Taiwan.
A widely respected author, Dr. Chen has authored and coauthored more than 20 engineering books
and 500 technical papers. His books include several classical works such as Limit Analysis and Soil
Plasticity (Elsevier, 1975), the two-volume heory of Beam-Columns (McGraw-Hill, 1976 and 1977),
Plasticity in Reinforced Concrete (McGraw-Hill, 1982), and the two-volume Constitutive Equations for
Engineering Materials (Elsevier, 1994). He currently serves on the editorial boards of more than 15
technical journals.
Dr. Chen is the editor-in-chief for the popular Civil Engineering Handbook (CRC Press, 1995 and
2003), the Handbook of Structural Engineering (CRC Press, 1997 and 2005), the Earthquake Engineering
Handbook (CRC Press, 2003), the Semi-Rigid Connections Handbook (J. Ross Publishing, 2011), and the
Handbook of International Bridge Engineering (CRC Press, 2014). He currently serves as the consult-
ing editor for the McGraw-Hill Yearbook of Science & Technology for the ield of civil and architectural
engineering.
He was a longtime member of the executive committee of the Structural Stability Research Council
and the speciication committee of the American Institute of Steel Construction. He was a consultant
for Exxon Production Research on ofshore structures, for Skidmore, Owings, and Merrill in Chicago
on tall steel buildings, and for the World Bank on the Chinese University Development Projects, among
many others. Dr. Chen has taught at Lehigh University, Purdue University, and the University of Hawaii.

xiii
xiv Editors

Dr. Lian Duan is a senior bridge engineer and structural steel committee
chair with the California Department of Transportation (Caltrans). He
worked at the North China Power Design Institute from 1975 to 1978 and
taught at Taiyuan University of Technology, China, from 1981 to 1985.
He earned his diploma in civil engineering in 1975, MS in structural
engineering in 1981 from Taiyuan University of Technology, China, and
PhD in structural engineering from Purdue University in 1990.
Dr. Duan’s research interests cover areas including inelastic behavior
of reinforced concrete and steel structures, structural stability, seismic
bridge analysis, and design. With more than 70 authored and coauthored
papers, chapters, and reports, his research focuses on the development of
uniied interaction equations for steel beam-columns, lexural stifness
of reinforced concrete members, efective length factors of compression
members, and design of bridge structures.
Dr. Duan has over 35 years experience in structural and bridge engineering. He was lead engineer for
the development of Caltrans Guide Speciications for Seismic Design of Steel Bridges. He is a registered
professional engineer in California. He served as a member for several National Highway Cooperative
Research Program panels and was a Transportation Research Board Steel Committee member from
2000 to 2006.
He is the coeditor of the Handbook of International Bridge Engineering, (CRC Press, 2014). He received
the prestigious 2001 Arthur M. Wellington Prize from the American Society of Civil Engineers for the
paper, “Section Properties for Latticed Members of San Francisco-Oakland Bay Bridge,” in the Journal
of Bridge Engineering, May 2000. He received the Professional Achievement Award from Professional
Engineers in California Government in 2007 and the Distinguished Engineering Achievement Award
from the Engineers’ Council in 2010.
 Contributors

Nan Deng Michael Knott Charles Seim

Bechtel Corporation Mofatt & Nichol Consulting Engineer
San Francisco, California Richmond, Virginia El Cerrito, California

Ralph J. Dornsife Youzhi Ma Kevin R. Smith

Washington State Department AMEC Environmental and Fugro Atlantic
of Transportation Infrastructure Inc. Norfolk, Virginia
Olympia, Washington Oakland, California
Aurelian C. Trandair
Jason Fan Fugro GeoConsulting Inc.
Amir M. Malek
California Department of Houston, Texas
California Department of
Transportation
Transportation
Sacramento, California Jinrong Wang
Sacramento, California
California Department of
Chao Gong Transportation
homas W. McNeilan
URS Corporation Sacramento, California
Fugro Atlantic
Oakland, California
Norfolk, Virginia
Linan Wang
Junke Guo California Department of
University of Nebraska–Lincoln Mihail E. Popescu
Transportation
Lincoln, Nebraska Illinois Institute of Technology
Sacramento, California
Chicago, Illinois
Mohammed S. Islam
California Department of Zolan Prucz
Transportation Modjeski and Masters Inc.
Sacramento, California New Orleans, Louisiana

xv
 1
Bearings
1.1 Introduction ..........................................................................................1
1.2 Bearing Types ........................................................................................1
Steel Reinforced Elastomeric Bearings • Fabric Pad
Bearings • Elastomeric Sliding Bearings • Pin Bearings • Rocker/
Roller Bearings • Pot Bearings • Disc Bearings • Spherical
Bearings • Seismic Isolation Bearings
1.3 Design Considerations .......................................................................10
Force Considerations • Movement Considerations • Elastomeric
Bearing Design • HLMR Bearing Design
1.4 Ancillary Details .................................................................................14
Masonry Plates • Sole Plates
1.5 Shop Drawings, Calculations, Review, and Approval ..................16
1.6 Bearing Replacement Considerations .............................................16
Ralph J. Dornsife 1.7 Design Examples .................................................................................17
Washington State Design Example 1—Steel Reinforced Elastomeric Bearing • Design
Department of Example 2—Longitudinally Guided Disc Bearing
Transportation References.................................................................................... ....................33

1.1 Introduction
Bridge bearings facilitate the transfer of vehicular and other environmentally imposed loads from the
superstructure down to the substructure, and ultimately, to the ground. In fulilling this function, bear-
ings must accommodate anticipated service movements while also restraining extraordinary move-
ments induced by extreme load cases. Because the movements allowed by an adjacent expansion joint
must be compatible with the movement restrictions imposed by a bearing, bearings and expansion
joints must be designed interdependently and in conjunction with the anticipated behavior of the over-
all structure.

1.2 Bearing Types

Historically, many types of bearings have been used for bridges. Contemporary bearing types include
steel reinforced elastomeric bearings, fabric pad sliding bearings, steel pin bearings, rocker bearings,
roller bearings, steel pin bearings, pot bearings, disc bearings, spherical bearings, and seismic isolation
bearings. Each of these bearings possesses diferent characteristics in regard to vertical and horizon-
tal load carrying capacity, vertical stifness, horizontal stifness, and rotational lexibility. A thorough
understanding of these characteristics is essential for economical bearing selection and design. Pot
bearings, disc bearings, and spherical bearings are sometimes collectively referred to as high-load
multi-rotational (HLMR) bearings.

1
2 Bridge Engineering Handbook, Second Edition: Substructure Design

1.2.1 Steel Reinforced Elastomeric Bearings

Elastomeric bearings are perhaps the simplest and most economical of all modern bridge bearings. hey
are broadly classiied into four types: plain elastomeric pads, iberglass reinforced elastomeric pads,
steel reinforced elastomeric pads, and cotton duck reinforced elastomeric pads. Of these four types, steel
reinforced elastomeric pads are used most extensively for bridge construction applications. Plain elas-
tomeric pads are used occasionally for lightly loaded applications. Cotton duck reinforced elastomeric
pads, generally referred to as fabric pad bearings, are used occasionally. his subsection will address
steel reinforced elastomeric bearings. A subsequent section will address fabric pad bearings.
A steel reinforced elastomeric bearing consists of discrete steel shims vulcanized between adjacent
discrete layers of elastomer. his vulcanization process occurs under conditions of high temperature
and pressure. he constituent elastomer is either natural rubber or synthetic rubber (neoprene). Steel
reinforced elastomeric bearings are commonly used with prestressed concrete girder bridges and may
be used with other bridge types. Because of their relative simplicity and fabrication ease, steel reinforced
elastomeric bearings ofer signiicant economy relative to HLMR bearings.
Prestressed concrete girder bridges use steel reinforced elastomeric bearings almost exclusively.
A concrete bridge application is shown in Figure 1.1. Steel reinforced elastomeric bearings have also
been used in steel plate girder bridge applications. Figure 1.2 depicts one such application in which ser-
vice load transverse movements are accommodated by the shear lexibility of the elastomer while larger
seismically induced transverse force efects are resisted by concrete girder stops.
Steel reinforced elastomeric bearings rely upon the inherent shear lexibility of the elastomeric layers
to accommodate bridge movements in any horizontal direction. he steel shims limit the tendency for
the elastomeric layers to bulge laterally under compressive load, thus limiting vertical deformation of
the bearing. he shear lexibility of the elastomeric layers also allows them to accommodate rotational
demands induced by loading.

1.2.2 Fabric Pad Bearings

Cotton duck, or fabric, pads are preformed elastomeric pads reinforced with very closely spaced layers of
cotton or polyester fabric. Fabric pads are typically manufactured in large sheets under military speci-
ications and with limited guidance from American Association of State Highway and Transportation
Oicials (AASHTO) Speciications (Lehman 2003). he close spacing of the reinforcing ibers, while
allowing fabric pads to support large compressive loads, imposes stringent limits upon their shear

FIGURE 1.1 Steel reinforced elastomeric bearing (concrete bridge) application.

Bearings 3

Centerline of steel plate girder

and centerline of bearing

Elastomeric bearing
reinforced with (10)
steel shims

Transverse girder 1-in. gap (Typ.)

stop (Typ.)

Sole plate

Grout pad

FIGURE 1.2 Steel reinforced elastomeric bearing (steel bridge).

displacement and rotational capacities. Unlike a steel reinforced elastomeric bearing having substantial
shear lexibility, the fabric pad alone cannot accommodate translational movement. Fabric pads can
accommodate very small amounts of rotational movement; substantially less than can be accommo-
dated by more lexible steel reinforced elastomeric bearings.

1.2.3 Elastomeric Sliding Bearings

Both steel reinforced elastomeric bearings and fabric pad bearings can be modiied to incorporate a
PTFE (PolyTetraFluoroEthylene, more commonly known by the DuPont trade name Telon)-stainless
steel sliding interface to accommodate large translational movements. Such modiications extend the
range of use of steel reinforced elastomeric bearings and make fabric pad bearings a viable and economi-
cal solution for applications with minimal rotational demand. A schematic representation of a fabric
pad sliding bearing is depicted in Figure 1.3. A typical fabric pad sliding bearing is shown in Figure 1.4.
PTFE material is available in several forms: unilled, illed, dimpled lubricated, and woven. hese var-
ious forms of PTFE difer substantially in their frictional properties and ability to resist creep (cold low)
under sustained load. Creep resistance is most efectively enhanced by conining the PTFE material in
a recess. Filled PTFE contains glass, carbon, or other chemically inert ibers that enhance its resistance
to creep and wear. Woven PTFE is created by interweaving high strength ibers through PTFE material.
Dimpled PTFE contains dimples machined into its surface. hese dimples act as reservoirs for silicone
grease lubricant. he use of silicone grease in dimpled PTFE reduces the friction coeicient in the early
life of the bearing. However, silicone grease will squeeze out under high pressure and attract dust and
other debris, which may accelerate wear and detrimentally impact a bearing’s durability.
he low-friction characteristics of a PTFE-stainless steel interface are actually facilitated by frag-
mentary PTFE sliding against solid PTFE ater the fragmentary PTFE particles are absorbed into the
asperities of the stainless steel surface. he optimum surface inish is thus associated with an optimum
asperity size and distribution. In order to minimize frictional resistance, a Number 8 (Mirror) inish
is generally speciied for all lat stainless steel surfaces in contact with PTFE. However, recent research
4 Bridge Engineering Handbook, Second Edition: Substructure Design

Sole plate
Fabric pad
PTFE
Recessed bearing plate
Stainless steel sheet

Keeper bars
Masonry plate Anchor rod
FIGURE 1.3 Fabric pad sliding bearing.

FIGURE 1.4 Fabric pad sliding bearing application.

has concluded that stainless steel having a 2B surface inish achieves similarly low-friction properties
with no measurable increase in wear (Stanton 2010). Unlike a Number 8 (Mirror) inish, a 2B inish is
achieved by cold rolling without further polishing. hus it is easier to obtain and more economical. he
research did not investigate the performance characteristics of the 2B inish at very low temperatures.
For a given steel surface inish, friction coeicients for PTFE-stainless steel sliding interfaces vary
signiicantly as a function of PTFE type, magnitude of contact pressure, and ambient temperature. he
AASHTO Load and Resistance Factor Design (LRFD) speciications provide friction coeicients associ-
ated with a Number 8 (Mirror) inish as a function of these variables. Dimpled lubricated PTFE at high
temperature and high contact pressures typically exhibits the lowest friction coeicients, as low as 0.020
(AASHTO 2012). Filled PTFE at very low temperatures and low contact pressures exhibits the highest
friction coeicients, as high as 0.65 (AASHTO 2012).
Resistance against creep of PTFE material is achieved by limiting both average and edge contact
stresses under both permanent and total loads. he AASHTO LRFD speciications limit unconined
unilled PTFE average contact stress to 1500 psi under permanent service load and 2500 psi under total
Bearings 5

service load. hese speciications also limit unconined illed PTFE, conined unilled PTFE, and woven
PTFE iber average contact stress to 3000 psi under permanent service load and 4500 psi under total
service load (AASHTO 2012). he AASHTO LRFD speciications permit slightly higher edge contact
stresses under both permanent and total service load.
In fabric pad sliding bearings, the unilled PTFE material is generally recessed half its thickness into a
steel backing plate. he backing plate is generally bonded to the top of a fabric pad. A stainless steel sheet
is typically seal welded to a steel sole plate attached to the superstructure to provide the low-friction
sliding interface.

1.2.4 Pin Bearings

Steel pin bearings are generally used to support high loads with moderate to high levels of rotation
about a single predetermined axis. his situation generally occurs with long straight steel plate girder
superstructures. Rotational capacity is aforded by rotation of a smoothly machined steel pin against
upper and lower smoothly machined steel bearing blocks. Steel keeper rings are typically designed and
detailed to provide uplit resistance. A schematic representation of the elements constituting a pin bear-
ing is depicted in Figure 1.5. A typical pin bearing of a bridge under construction prior to grout pad
placement is shown in Figure 1.6.

Upper block Sole plate

Lower block Steel pin

Keeper ring
Masonry plate

Anchor rod Nut and washer

FIGURE 1.5 Steel pin bearing.

FIGURE 1.6 Steel pin bearing application.

6 Bridge Engineering Handbook, Second Edition: Substructure Design

1.2.5 Rocker/Roller Bearings

Steel rocker bearings have been used extensively in the past to allow both rotation and longitudinal
movement while supporting moderately high loads. Because of their seismic vulnerability and the more
extensive use of steel reinforced elastomeric bearings, rocker bearings are now rarely speciied for new
bridges. A typical rocker bearing adjacent to a pin (ixed) bearing of an older reinforced concrete bridge
is shown in Figure 1.7.
Steel roller bearings have also been used extensively in the past. Roller bearings permit both rota-
tional and longitudinal movement. Pintles are oten used to efect transverse force transfer by connect-
ing the roller bearing to the superstructure above and to the bearing plate below. Two views of a steel
roller bearing are shown in Figure 1.8. his roller bearing has displaced up against its stop bar and can-
not accommodate any further movement.
Nested roller bearings have also been used in the past. hey are composed of a series of rollers. Without
adequate preventative maintenance, these bearings can experience corrosion and lockup. Figure 1.9 is
a photograph of a nested roller bearing application. Having been supplanted by more economical steel
reinforced elastomeric bearings, roller bearings are infrequently used for new bridges today.

FIGURE 1.7 Steel rocker bearing application.

FIGURE 1.8 Steel roller bearing application.

Bearings 7

FIGURE 1.9 Nested roller bearing application.

Confined elastomer Piston

Sole plate

Flat sealing rings

Circular sealing
ring

Masonry plate

Anchor rod

FIGURE 1.10 Pot bearing.

1.2.6 Pot Bearings

A pot bearing is composed of a plain elastomeric disc that is conined in a vertically oriented steel cylin-
der, or pot, as depicted schematically in Figure 1.10. Vertical loads are transmitted through a steel piston
that sits atop the elastomeric disc within the pot. he pot walls conine the elastomeric disc, enabling
it to sustain much higher compressive loads than could be sustained by more conventional unconined
elastomeric material. Rotational demands are accommodated by the ability of the elastomeric disc to
deform under compressive load and induced rotation. he rotational capacity of pot bearings is gener-
ally limited by the clearances between elements of the pot, piston, sliding surface, guides, and restraints
(Stanton 1999). A pot bearing application detailed to provide uplit resistance is shown in Figure 1.11.
Flat or circular sealing rings prevent the pinching and escape of elastomeric material through the
gap between the piston and pot wall. In spite of these sealing elements, some pot bearings have dem-
onstrated susceptibility to elastomer leakage. hese problems have occurred predominantly on steel
bridges, which tend to be more lightly loaded. Unanticipated rotations during steel erection may con-
tribute to and exacerbate these problems. Excessive elastomeric leakage could result in the bearing expe-
riencing hard metal-to-metal contact between components. Despite these occasional problems, most pot
bearings have performed well in serving as economical alternatives to more expensive HLMR bearings.
8 Bridge Engineering Handbook, Second Edition: Substructure Design

FIGURE 1.11 Pot bearing application.

Urethane disk
Sole plate
Shear-resisting pin

Shear-resisting ring
Masonry plate
Upper and lower bearing plates
Plan view Cross section

FIGURE 1.12 Disc bearing.

A lat PTFE-stainless steel interface can be built into a pot bearing assembly to additionally provide
translation movement capability, either guided or nonguided.

1.2.7 Disc Bearings

A disc bearing relies upon the compressive lexibility of an annular shaped polyether urethane disc to
provide moderate levels of rotational movement capacity while supporting high loads. A steel shear-
resisting pin in the center provides resistance against lateral force. A lat PTFE-stainless steel sliding
interface can be incorporated into a disc bearing to additionally provide translational movement capa-
bility, either guided or nonguided. he primary constituent elements of a disc bearing are identiied in
the schematic representation of a disc bearing in Figure 1.12. Two views of a typical disc bearing applica-
tion are shown in Figure 1.13.

1.2.8 Spherical Bearings

A spherical bearing, sometimes referred to as a curved sliding bearing, relies upon the low-friction
characteristics of a curved PTFE-stainless steel sliding interface to provide a high level of rotational lex-
ibility in multiple directions while supporting high loads. Unlike pot bearings and disc bearings, spheri-
cal bearing rotational capacities are not limited by strains, dimensions, and clearances of deformable
Bearings 9

FIGURE 1.13 Disc bearing application.

Stainless steel sliding surface

Upper sole plate PTFE
Lower sole plate Upper concave block

Guide bar

Lower convex block

Anchor rod
Masonry plate

FIGURE 1.14 Spherical bearing.

elements. Spherical bearings are capable of sustaining very large rotations provided that adequate clear-
ances are provided to avoid hard contact between steel components.
A lat PTFE-stainless steel sliding interface can be incorporated into a spherical bearing to addition-
ally provide either guided or nonguided translational movement capability. he constituent elements of
a guided spherical bearing are depicted in Figure 1.14. his depiction includes a lat PTFE-stainless steel
sliding interface to provide translational movement capability. he steel guide bars limit translational
movement to one direction only. A typical spherical bearing application is shown in Figure 1.15.
Woven PTFE material is generally used on the curved surfaces of spherical bearings. As noted earlier,
woven PTFE exhibits enhanced creep (cold low) resistance and durability relative to unwoven PTFE.
When spherical bearings are detailed to accommodate translational movement, woven PTFE is gener-
ally speciied at the lat sliding interface also.
10 Bridge Engineering Handbook, Second Edition: Substructure Design

FIGURE 1.15 Spherical bearing application.

Both stainless steel sheet and solid stainless steel have been used for the convex sliding surface of
spherical bearings. According to one manufacturer, curved sheet is generally acceptable for contact
surface radii greater than 14 in. to 18 in. For smaller radii, a solid stainless steel convex plate or stain-
less steel inlay is typically used. he inlay is welded to solid standard steel. For taller convex plates, a
stainless steel inlay would likely be more economical.
Most spherical bearings are fabricated with the concave surface oriented downward to minimize dirt
iniltration between the PTFE material and the stainless steel surface. Calculation of translational and
rotational movement demands on the bearing must recognize that the center of rotation of the bearing
is generally not coincident with the neutral axis of the girder being supported.

1.2.9 Seismic Isolation Bearings

Seismic isolation bearings mitigate the potential for seismic damage by utilizing two related phenomena:
dynamic isolation and energy dissipation. Dynamic isolation allows the superstructure to essentially
loat, to some degree, while substructure elements below move with the ground during an earthquake.
he ability of some bearing materials and elements to deform in certain predictable ways allows them to
dissipate seismic energy that might otherwise damage critical structural elements.
Numerous seismic isolation bearings exist, each relying upon varying combinations of dynamic iso-
lation and energy dissipation. hese devices include lead core elastomeric bearings, high damping rub-
ber bearings, friction pendulum bearings, hydraulic dampers, and various hybrid variations.
Efective seismic isolation bearing design requires a thorough understanding of the dynamic charac-
teristics of the overall structure as well as the candidate isolation devices. Isolation devices are diferen-
tiated by maximum compressive load capacity, lateral stifness, lateral displacement range, maximum
lateral load capacity, energy dissipation capacity per cycle, functionality in extreme environments, resis-
tance to aging, fatigue and wear properties, and efects of size.

1.3 Design Considerations

Bearings must be designed both to transfer forces between the superstructure and the substructure and
to accommodate anticipated service movements. Bearings must additionally restrain undesired move-
ments and transmit extraordinary forces associated with extreme loads. his section discusses force
and movement considerations as well as some of the design aspects associated with steel reinforced
elastomeric and HLMR bearings.
Bearings 11

1.3.1 Force Considerations

Bridge bearings must be explicitly designed to transfer all anticipated loads from the superstructure
to the substructure. Sources of these loads include dead load, vehicular live load, wind loads, seismic
loads, and restraint against posttensioning elastic shortening, creep, and shrinkage. hese forces may
be directed vertically, longitudinally, or transversely with respect to the global orientation of the bridge.
In some instances, bearings must be designed to resist uplit. In accordance with the AASHTO LRFD
speciications, most bearing design calculations are based upon service limit state stresses. Impact need
not be applied to live load forces in the design of bearings.

1.3.2 Movement Considerations

Bridge bearings can be detailed to provide translational ixity, to permit free translation in any horizon-
tal direction, or to permit guided translation. he movement restriction thus imposed by a bearing must
be compatible with the movements allowed by any adjacent expansion joint. Additionally, both bearings
and expansion joints must be designed consistent with the anticipated load and displacement behavior
of the overall structure. Sources of anticipated movement include concrete shrinkage and creep, post-
tensioning shortening, thermal luctuations, dead and live loads, and wind or seismic loads. Design
rotations can be calculated as follows:
1. Elastomeric and Fabric Pad Bearings: he AASHTO LRFD speciications stipulate that the maxi-
mum service limit state rotation for bearings that do not have the potential to achieve hard contact
between metal components shall be taken as the sum of unfactored dead and live load rotations
plus an allowance for uncertainties of 0.005 radians. If a bearing is subject to rotation in opposing
directions due to diferent efects, then this allowance applies in each direction.
2. HLMR Bearings: he AASHTO LRFD speciications stipulate that the maximum strength limit
state rotation for bearings that are subject to potential hard contact between metal components
shall be taken as the sum of all applicable factored load rotations plus an allowance of 0.005 radi-
ans for fabrication and installation tolerances and an additional allowance of 0.005 radians for
uncertainties. he rationale for this more stringent requirement is that metal or concrete elements
are susceptible to damage under a single rotation that causes contact between hard elements. Such
bearings include spherical, pot, steel pin, and some types of seismic isolation bearings.
Disc bearings are less likely to experience metal-to-metal contact because they use an uncon-
ined load element. Accordingly, they are designed for a maximum strength limit state rotation
equal to the sum of the applicable strength load rotation plus an allowance of 0.005 radians for
uncertainties. If a bearing is subject to rotation in opposing directions due to diferent efects,
then this allowance applies in each direction.

1.3.3 Elastomeric Bearing Design

Steel reinforced elastomeric bearings and fabric pad sliding bearings are generally designed by the
bridge design engineer. hese relatively simple bearings are easy to depict and fabrication procedures
are relatively uniform and straightforward.
Steel reinforced elastomeric bearings can be designed by either the Method A or Method B procedure
delineated in the AASHTO LRFD speciications. he Method B provisions provide more relief in meet-
ing rotational demands than Method A. he Method A design procedure is a carryover based upon more
conservative interpretation of past theoretical analyses and empirical observations prior to research lead-
ing up to the publication of NCHRP Report 596 Rotation Limits for Elastomeric Bearings (Stanton 2008).
Both Method A and Method B design procedures require determination of the optimal geometric param-
eters to achieve an appropriate balance of compressive, shear, and rotational stifnesses and capacities.
12 Bridge Engineering Handbook, Second Edition: Substructure Design

Fatigue susceptibility is controlled by limiting live load compressive stress. Susceptibility of steel shims
to delamination from adjacent elastomer is controlled by limiting total compressive stress. Assuring ade-
quate shim thickness precludes yield and rupture of the steel shims. Excessive shear deformation is con-
trolled and rotational lexibility is assured by providing adequate total elastomer height. Generally, total
elastomer thickness shall be no less than twice the maximum anticipated lateral deformation. Overall
bearing stability is controlled by limiting total bearing height relative to its plan dimensions.
he most important design parameter for reinforced elastomeric bearings is the shape factor. he
shape factor is deined as the plan area of the bearing divided by the area of the perimeter free to bulge
(plan perimeter multiplied by elastomeric layer thickness). Figure 1.16 illustrates the shape factor con-
cept for a typical steel reinforced elastomeric bearing and for a fabric pad bearing.
Axial, rotational, and shear loading generate shear strain in the constituent layers of a typical elasto-
meric bearing as shown in Figure 1.17. Computationally, Method B imposes a limit on the sum of these
shear strains. It distinguishes between static and cyclic components of shear strain by applying an ampli-
ication factor of 1.75 to cyclic efects to relect cumulative degradation caused by repetitive loading.
Both the Method A and Method B design procedures limit translational movement to one-half the
total height of the constituent elastomeric material composing the bearing. Translational capacity can
be increased by incorporating an additional low-friction sliding interface. In this case, a portion of the
translational movement is accommodated by shear deformation in the elastomeric layers. Movement
exceeding the slip load displacement of the low-friction interface is accommodated by sliding.

Steel reinforced elastomeric Fabric pad

Bearing shown Bearing shown

Area free
to bulge

Plan area of bearing

Shape factor =
Area of perimeter free to bulge

FIGURE 1.16 Shape factor for elastomeric bearings.

Shear strain Axial Shear

Shear strain

Shear strain
Rotation

FIGURE 1.17 Shear strains in elastomeric bearings.

Bearings 13

In essence, elastomeric bearing design reduces to checking several mathematical equations while
varying bearing plan dimensions, number of elastomeric layers and their corresponding thicknesses,
and steel shim thicknesses. Mathematical spreadsheets have been developed to evaluate these tedious
calculations.
Although constituent elastomer has historically been speciied by durometer hardness, shear modu-
lus is the most important physical property of the elastomer for purposes of bearing design. Research
has concluded that shear modulus may vary signiicantly among compounds of the same hardness.
Accordingly, shear modulus shall preferably be speciied without reference to durometer hardness.
Elastomeric bearings shall conform to the requirements contained in AASHTO Speciication M 251
Plain and Laminated Elastomeric Bridge Bearings. Constituent elastomeric layers and steel shims shall be
fabricated in standard thicknesses. For overall bearing heights less than about 5 in., a minimum of ¼ in.
of horizontal cover is recommended over steel shim edges. For overall bearing heights greater than 5 in.,
a minimum of ½ in. of horizontal cover is recommended (WSDOT 2011). AASHTO Speciications M
251 requires elastomeric bearings to be subjected to a series of tests, including a compression test at
150% of total service load. For this reason, compressive service dead and live loads should be speciied
in the project plans or speciications.
As mentioned earlier, the AASHTO LRFD speciications stipulate that a 0.005 radian allowance for
uncertainties be included in the design of steel reinforced elastomeric bearings. his allowance applies
to rotation in each opposing direction. Commentary within the AASHTO LRFD speciications states
that an owner may reduce this allowance if justiied by “a suitable quality control plan.” In the absence
of a very speciic implementable plan, this is inadvisable given that 0.005 radians corresponds to a slope
of only about 1/16 in. in 12 in.
Unlike many HLMR bearing types, elastomeric bearings cannot be easily installed with an imposed of-
set to accommodate actual temperature at installation in addition to any anticipated long-term movements
such as creep and shrinkage. For practical reasons, girders are rarely set atop elastomeric bearings at the
mean of the expected overall temperature range. Rarely are girders subsequently lited to relieve imposed
vertical load to allow the bearings to replumb themselves at the mean temperature. he AASHTO LRFD
speciications statistically reconcile this reality by stipulating a design thermal movement, applicable in
either direction, of 65% of the total thermal movement range. his percentage may be reduced in instances
in which girders are originally set or reset at the average of the design temperature range. For precast
prestressed concrete girder bridges, the maximum design thermal movement shall be added to shrinkage,
long-term creep, and posttensioning movements to determine the total bearing height required.
he material properties of most elastomers vary with temperature. Both natural rubber and neoprene
stifen and become brittle at colder temperatures. herefore, it is important that the type of elastomer
be considered explicitly in specifying the bearing and determining the resulting lateral forces that will
be transferred to substructure elements. he AASHTO LRFD speciications categorize elastomers as
being of Grade 0, 2, 3, 4, or 5. A higher grade number corresponds to greater resistance against stifening
under sustained cold conditions. Special compounding and curing are needed to provide this resistance
and thus increase the cost of the constituent bearing. Determination of the minimum grade required
depends upon the more critical of (1) the 50-year low temperature and (2) the maximum number of con-
secutive days in which the temperature does not rise above 32°F (0°C). he intent of specifying a mini-
mum grade is to limit the forces transferred to the substructure to 1.5 times the service limit state design.
he AASHTO LRFD speciications allow using lower grade elastomers if a low-friction sliding interface
is incorporated and/or if the substructure is designed to resist a multiple of the calculated lateral force.

1.3.4 HLMR Bearing Design

Although design procedures have historically been largely proprietary, the AASHTO LRFD speciications
do provide some guidance for the design of all three primary HLMR bearing types: pot bearings, disc
bearings, and spherical bearings. hus, all three HLMR bearing types may be allowed on most projects.
14 Bridge Engineering Handbook, Second Edition: Substructure Design

Because of their inherent complexity and sensitivity to fabrication methods, HLMR and seismic isola-
tion bearings should generally be designed by their manufacturers (AASHTO/NSBA 2004). Each bear-
ing manufacturer has unique fabricating methods, personnel, and procedures that allow it to fabricate a
bearing most economically. For these reasons, these bearing types are generally depicted schematically
in contract drawings. Depicting the bearings schematically with speciied loads, movements, and rota-
tions provides each manufacturer the lexibility to innovatively achieve optimal economy subject to the
limitations imposed by the contract drawings and speciications.
Contract drawings must show the approximate diameter and height of the HLMR bearing in addition
to all dead, live, and lateral wind/seismic loadings. his generally requires a preliminary design to be
performed by the bridge designer or bearing manufacturer. Diameter of a HLMR bearing is governed
primarily by load magnitude and material properties of the lexible load bearing element. he height
of a pot bearing or disc bearing is governed primarily by the rotational demand and lexibility of the
deformable bearing element. he height of a spherical bearing depends upon the radius of the curved
surface, the diameter of the bearing, and the total rotational capacity required.
Accessory elements of the bearing, such as masonry plates, sole plates, anchor rods, and any appur-
tenance for horizontal force transfer should be designed and detailed on the contract drawings by the
bridge designer. Notes should be included on the plans allowing the bearing manufacturer to make
minor adjustments to the dimensions of sole plates, masonry plates, and anchor rods. he HLMR bear-
ing manufacturer is generally required to submit shop drawings and detailed structural design calcula-
tions for review and approval by the bridge design engineer.
HLMR bearings incorporating sliding interfaces require inspection and long-term maintenance. It
is important that these bearings be designed and detailed to allow future removal and replacement of
sliding interface elements. Such provisions should allow these elements to be removed and replaced with
a maximum vertical jacking height of ¼ in. (6 mm) ater the vertical load is removed from the bearing
assembly. By limiting the jacking height, this work can be performed under live load and without dam-
aging expansion joint components.
HLMR bearings must be designed, detailed, fabricated, and installed to provide a continuous load
path through the bearing from the superstructure to the substructure. he load path must account for
all vertical and horizontal service, strength, and extreme limit state loads. he importance of providing
positive connections as part of a continuous load path cannot be overemphasized. he spherical bear-
ing shown in Figure 1.15 shows both an upper and lower sole plate, with the lower sole plate displaced
longitudinally relative to the upper sole plate. he upper sole plate was embedded in the concrete super-
structure. Because uplit had not been anticipated in the design of this Seattle bridge, the lower sole plate
was designed to it loosely in a recess in the bottom of the upper sole plate. During the 2001 Nisqually
Earthquake, the upper and lower sole plates of this bearing separated, causing the lower sole plate to
dislodge and displace.

1.4 Ancillary Details

HLMR bearings should be detailed and installed in such a way as to allow the bearings to be serviced
and/or replaced during the lifetime of the bridge. A masonry plate connects the bottom of the bearing
to the top of the supporting structural elements below. A sole plate connects the top of the bearing to
the superstructure above.

1.4.1 Masonry Plates

Masonry plates help to more uniformly distribute loads from a bearing to supporting concrete substruc-
ture elements below. Additionally, they provide platforms to facilitate maintenance and repairs of bear-
ings. Analysis shows that a steel masonry plate will deform under concentrated bearing loads (Stanton
1999). his potential deformation, which tends to cause a dishing efect because of the relatively lexible
Bearings 15

nature of the concrete below, must be recognized in the design of the masonry plate. he masonry
plate supporting a HLMR bearing is generally supported either on a thin preformed elastomeric pad or
directly atop a grout pad that is poured ater the superstructure girders have been erected. Each of these
two methods has associated advantages and disadvantages.
A ⅛-in. thick preformed plain elastomeric pad or fabric pad placed atop the concrete bearing sur-
face or grout pad most economically compensates for any minor surface irregularities. Fully threaded
anchor rods can be either cast into the concrete or drilled and grouted into place. An anchor plate can be
either bolted or welded to the bottom of the anchor rod to augment uplit capacity in the concrete. If no
uplit capacity is required, a swedged rod may be substituted for a threaded one. he swedged rod may
be terminated just below the top of the masonry plate and the void illed with a lexible sealant.
A grout pad poured underneath the masonry plate ater girder erection can provide the contrac-
tor more lexibility in leveling and adjusting the horizontal position of the bearing. A variation of this
method incorporating postgrouted hollow steel pipes can be used to substantially increase uplit capac-
ity of the anchor rods and provide some additional anchor rod adjustability. Several methods have been
used successfully to temporarily support the masonry plate until the grout is poured. he two most
commonly used methods are
1. Shim Packs—Multiple stacks of steel shim plates are placed atop the concrete supporting surface to
temporarily support the load on the masonry plate before grouting. Engineering judgment must
be used in selecting the number and plan size of the shims, taking grout lowability, load distri-
bution, and shim pack height adjustability into consideration. To enhance uplit resistance, steel
anchor rods are sometimes installed in hollow steel pipes embedded into the concrete. he steel
pipes have plates welded to their bottoms through which the anchor rods are bolted. Grouting is
accomplished using grout tubes that extend to the bottom of the pipes. Once all pipes are fully
grouted around the anchor rods, the space between the top of the concrete support surface and
the underside of the masonry plate is grouted.
2. Two-step Grouting with Voided Cores/Studs—A two-step grouting procedure with cast-in-place
voided cores can be used for smaller HLMR bearings not generally subjected to signiicant uplit.
Steel studs are welded to the underside of the masonry plate to coincide with voided core loca-
tions. With the girders erected and temporary shims installed between the top of the concrete
surface and the underside of the masonry plate, the voided cores are fully grouted. Once the irst
stage grout has attained strength, the steel shims are removed, the masonry plate is dammed, and
grout is placed between the top of the concrete support surface and the underside of the masonry
plate.
he use of anchor rod leveling nuts, without shim packs, to level a masonry plate prior to grout place-
ment is not recommended. he absence of shim packs results in the application of point loads at anchor
rod locations. his phenomenon is a consequence of the high stifness of the anchor rods relative to the
grout material and can result in warping of the masonry plate (AASHTO/NSBA 2004). Similar consid-
eration must be given to the sizing and number of shim plates as it relates to potential dishing of the
masonry plate under load.

1.4.2 Sole Plates

For concrete bridge superstructures, headed steel studs are typically welded to the top of the sole plate
and embedded into the superstructure. In steel bridge superstructures, sole plates may be bolted or
welded to I-shaped plate girder bottom langes. Sole plate assemblies should be bolted to the bottom
lange of steel box girder bridges because welded connections would require overhead welding, which
may be diicult to perform because of limited access.
Welding of sole plates to steel I-shaped girders allows for greater adjustment during installation
and is generally more economical. Damage associated with removal of the weld as required for future
16 Bridge Engineering Handbook, Second Edition: Substructure Design

maintenance and replacement operations can be reasonably repaired. For these welded connections, it
is recommended that the sole plate extend transversely beyond the edge of the bottom lange by at least
1 in. in order to allow ½ in. of ield adjustment. Welds for sole plate connections should be longitudinal
to the girder axis. he transverse joints should be sealed with an approved caulking material. he longi-
tudinal welds are made in the horizontal position, which is the position most likely to achieve a quality
illet weld. Transverse welds would require overhead welding, which may be diicult to perform because
of limited clearance. Caulking is installed along the transverse seams following longitudinal welding to
prevent corrosion between the sole plate and the bottom lange. he minimum thickness of the welded
sole plate should be ¾ in. to minimize plate distortion during welding (AASHTO/NSBA 2004).
Bolting of sole plates to steel I-shaped girders is also used. Bolting typically requires minimal paint
repair, as opposed to welding, and simpliies removal of a bearing for future maintenance and replace-
ment needs. Oversized holes allow for minor ield adjustments of the bearing during installation.
In some instances, an upper and lower sole plate may be used to simplify the bolted connection to a
steel girder or to account for grade efects. he upper uniform thickness sole plate is bolted to the bottom
lange while the lower tapered sole plate is welded to the upper sole plate. For a concrete bridge, the lower
sole plate may be drilled and the embedded upper sole plate tapped for bolting together. he spherical
bearing depicted in Figure 1.14 includes an upper and lower sole plate to facilitate removal and replace-
ment of bearing elements.
Flatness of the steel mating surfaces may be a concern when bolting a sole plate to a steel girder bot-
tom lange. In lieu of specifying a tighter latness tolerance on the girder bottom lange, epoxy bedding
can be used between the sole plate and the girder bottom lange. Silicone grease is used as a bond breaker
on one of the surfaces in order to allow removal of the sole plate for servicing the bearing during the life
of the bridge.

1.5 Shop Drawings, Calculations, Review, and Approval

As part of the overall process of HLMR and isolation bearing design, the manufacturer generates design
calculations and produces shop drawings for review and approval by the bridge design engineer. he
bridge design engineer is typically responsible for checking and approving these design calculations
and shop drawings. his review shall assure that the calculations conirm the structural adequacy of
all components of the bearing, a continuous load path is provided for all vertically and horizontally
imposed loads, and each bearing is detailed to permit the inspection and replacement of components
subject to wear.
he approved shop drawings should note that all HLMR bearings shall be marked prior to shipping.
hese marks shall be permanent and in a readily visible location on the bearing. hey shall note the posi-
tion of the bearing and the direction ahead on station. Numerous ield problems have occurred when
bearings were not so marked. his is particularly true for minimally beveled sole plates. It is not always
apparent which orientation a bearing must take prior to imposition of the dead load rotation.

1.6 Bearing Replacement Considerations

In some situations, existing bearings or elements thereof must be replaced as a result of excessive wear,
damage, or seismic rehabilitation needs. Bearing replacement operations generally require liting of
superstructure elements using hydraulic jacks. Anticipated liting loads should be stipulated on the
contract drawings. Limitations on lit height should also be speciied. Considerations should be given
to lit height as it relates to adjacent expansion joint components and adjoining sections of safety rail-
ing. As mentioned earlier, new bearings should be detailed to allow replaceable elements to be removed
and replaced with a maximum vertical jacking height of ¼ in. (6 mm). Superstructure stresses induced
by nonuniform liting are limited by imposing restrictions on diferential lit height between adjacent
jacks.
Bearings 17

Experience concludes that actual liting loads nearly always exceed calculated liting loads. Many
factors may contribute to this phenomenon, including friction in the hydraulic jack system and underes-
timation of superstructure dead loads. A typical contract provision is to require that all hydraulic jacks
be sized for 200% of the calculated liting load. In planning a bearing replacement project, the designer
should verify from manufacturers’ literature that appropriate hydraulic jacks are available to operate
within the space limitations imposed by a particular design situation.

1.7 Design Examples

Two design examples are provided to illustrate the bearing design procedure: a steel reinforced elasto-
meric bearing and a longitudinally guided disc bearing.

1.7.1 Design Example 1—Steel Reinforced Elastomeric Bearing

Design of steel reinforced elastomeric bearings, as mentioned earlier, is an iterative process of check-
ing several design requirements while varying bearing plan dimensions, number of elastomeric layers
and corresponding thickness, and steel shim thicknesses. For precast prestressed concrete girders, this
process is somewhat complicated by the need to track camber rotations at various stages under difer-
ent loading conditions. In general, two times are most likely to be critical: (1) ater girders are set but
immediately before the slab is cast, at which time some of the prestressing has been lost and (2) ater the
bridge is constructed and live load is applied, at which time all prestressing losses have occurred. Both
cases should be checked. For each instance, the 0.005 radian tolerance needs to be applied in the most
critical direction, positive or negative.
Excellent examples of elastomeric bearing design for a precast prestressed concrete girders are
included in Chapter 10 Bearings of the Precast Prestressed Concrete Bridge Design Manual (PCI 2011).
A condensed version of one of these examples has been adapted to the following example.

1.7.1.1 Given
A single span precast prestressed concrete girder bridge near Minneapolis, Minnesota, has a total length
of 120 t. (36.6 m) with six equally loaded girders. he abutments are not skewed. Each girder end is
supported on a 22-in. (559 mm) wide by 8-in. (203 mm) long steel reinforced elastomeric bearing. hese
bearings contain four interior elastomeric layers of ½-in. (12.7 mm) thickness and two exterior elasto-
meric layers of ¼-in. (6 mm) thickness. hese layers are reinforced with ive steel plates having a yield
stress of 36 ksi (248 MPa). Assume that one end of the bridge is ixed against movement. he con-
tract documents specify the shear modulus of the elastomer at 73°F (22.8°C) to be 165 psi (1.138 MPa).
Current acceptance criteria allow the actual shear modulus, G, to vary by +/− 15% from the speciied
value. With the exception of checking the bearing against slippage, the critical extreme range value of
140 psi (0.965 MPa) is used in this example.
For the purpose of determining resulting displacements imposed upon each bearing, a sequence of
nine movement phenomena are considered and included in this problem. hese movements are: transfer
of prestressing following girder casting, girder self-weight, creep and shrinkage occurring before each
girder is erected on bearings, creep and shrinkage occurring ater each girder is erected on the bearings,
weight of slab on each girder, diferential shrinkage of the slab ater it is placed, uniform thermal expan-
sion and contraction, lane live load, and truck live load. Because they occur prior to the girders being set
onto the elastomeric bearings, the uniform shortening movements associated with the irst three phe-
nomena do not induce corresponding shear deformations in the bearings. However, because the bottom
of the girder does not have a sloped recess to accommodate anticipated end rotations, all phenomena,
with the exception of uniform thermal expansion and contraction, induce rotation in the bearings.
Nonthermal related longitudinal movements at the top of the bearing at the free end of the bridge
have been calculated as follows, with negative numbers denoting movement toward midspan:
18 Bridge Engineering Handbook, Second Edition: Substructure Design

Δcreep+shrinkage ater girder erection = −0.418 in.

ΔDL slab = 0.333 in.
Δdiferential shrinkage of slab = −0.071 in.
ΔLL lane = 0.109 in.
ΔLL truck = 0.208 in.
It should be noted that the horizontal displacements reported earlier result from a combination of two
efects: (1) change in the length of the concrete girder at its centroid and (2) end rotation of the girder
about its centroid. For instance, creep and shrinkage of the girder following erection causes it to uni-
formly shorten and to delect upward and rotate about its ends. he end rotation causes the bottom of
the girder at the bearing to shit inward (toward midspan), augmenting the shortening efect. Similarly,
diferential shrinkage of the slab causes the girder to uniformly shorten and to delect downward and
rotate about its ends. In this case, the end rotation causes the bottom of the girder at the bearing to shit
outward (away from midspan), reducing the uniform shortening efect. he longitudinal bearing move-
ments listed earlier include both of these efects. For numerical derivation of the individual efects, see
the Precast Prestressed Concrete Bridge Design Manual (PCI 2011).
Rotations imposed upon the bearings have been calculated as follows:
θinitial prestress = −9.260 × 10−3 rads
θDL girder = 3.597 × 10−3 rads
θcreep + shrinkage before girder erection = −2.900 × 10−3 rads
θcreep + shrinkage ater girder erection = −1.450 × 10−3 rads
θDL slab = 4.545 × 10−3 rads
θdiferential shrinkage of slab = 2.370 × 10−3 rads
θuniform thermal = 0.000 rads
θLL lane = 0.997 × 10−3 rads
θLL truck = 1.896 × 10−3 rads
Vertical load efects on each bearing are as follows:
P DL girder = 47.9 kips
P DL slab = 73.3 kips
P LL lane = 33.9 kips
P LL truck = 78.1 kips

1.7.1.2 Requirements
Perform the following design calculations for a steel reinforced elastomeric beating in accordance with
the AASHTO LRFD Bridge Design Speciications, 6th edition. (AASHTO 2012.)
• Determine the design thermal movement
• Check the adequacy of the bearing to accommodate maximum horizontal displacement, using
the AASHTO LRFD Method B design procedure
• Calculate shape factor of the bearing
• Check service load combination
• Check condition immediately before deck placement
• Evaluate stability of the bearing
• Determine required thickness of steel reinforcement
• Determine low temperature requirements for the constituent elastomer
• Calculate approximate instantaneous dead load, the long-term dead load, and the live load
compressive deformation of the bearings
• Consider hydrostatic stress
Bearings 19

• Evaluate the need for providing anchorage against slippage

1.7.1.3 Solution

Step 1: Determine the design thermal movement

AASHTO LRFD Article 3.12.2 includes thermal contour maps for determining uniform tem-
perature efects using the Method B procedure deined therein. hese maps show TMaxDesign
as 110°F (43.3°C) and TMinDesign as −20°F (−6.7°C) for concrete girder bridges with concrete
decks near Minneapolis, Minnesota. hese values are used to calculate the design thermal
movement range, ΔT.

∆ T = αL(TMaxDesign − TMinDesign ) [LRFD Eqn. 3.12.2.3–1]

where L is expansion length (in.); α is coeicient of thermal expansion (in./in./°F).

∆ T = (0.000006)(120)(12)[110 − (−20)] = 1.123 in.

AASHTO LRFD Article 14.7.5.3.2 states that the maximum horizontal displacement of the
bridge superstructure, Δ0, shall be taken as 65% of the design thermal movement range, ΔT,
computed in accordance with Article 3.12.2 combined with the movement caused by creep,
shrinkage, and posttensioning. Note that movement associated with superimposed dead load
is not speciied in this provision.

Design thermal movement = 0.65 ∆ T = 0.65(1.123) = 0.730 in.

his movement can be either expansion or contraction. Uniform temperature change does
not produce girder end rotation augmenting this movement.
Step 2: Check adequacy of the bearing to accommodate maximum horizontal displacement
As noted earlier, for the purpose of calculating the shear deformation in each bearing, the
design thermal movement is added to all creep, shrinkage, and posttensioning efects that
occur ater the girders are set on the bearings.

∆ o = 0.65∆ T + ∆ creep + shrinkage after girder erection + ∆ differential shrinkage of slab

= − 0.730 − 0.418 − 0.071 = −1.219 in.

AASHTO LRFD Article 14.7.5.3.2 requires that the total elastomer thickness, hrt, should
exceed twice the maximum total shear deformation, ΔS. In this example, we take maximum
total shear deformation as Δ0.

hrt = 4(0.5) + 2(0.25) = 2.5 in. > 2(∆ S ) = 2(∆ 0 ) = 2(1.219) = 2.44 in. O.K.

Step 3: Calculate shape factor of the bearing

Shape factor is calculated by the following equation:

LW
Si = [LRFD 14.7.5.1-1]
2hri ( L + W )

where hri is thickness of the ith interior elastomeric layer of the bearing (in.); L is plan dimen-
sion of the bearing generally parallel to the global longitudinal bridge axis (in.); and W is plan
dimension of the bearing generally parallel to the global transverse bridge axis (in.).
20 Bridge Engineering Handbook, Second Edition: Substructure Design

LW ( 8 )( 22 )
Si = = = 5.867
2hri ( L + W ) 2( 0.5 )( 8 + 22 )
Step 4: Check service load combination
In this example, dead loading constitutes static loads while vehicular live loading constitutes
cyclic loads. Vertical bearing force from static loads, Pst and vertical bearing force from cyclic
loads, Pcy are calculated as follows:

Pst = PDL girder + PDL slab = 47.9 + 73.3 = 121.2 kip

Pcy = PLL lane + PLL truck = 33.9 + 78.1 = 112.0 kip

Vertical bearing stresses are calculated as follows:

Pst 121.2
σ a,st = = = 0.689 ksi
LW ( 8 )( 22 )

Pcy 112.0
σ a,cy = = = 0.636 ksi
LW ( 8 )( 22 )

σ s = σ a,st + σ a,cy = 0.689 + 0.636 = 1.325 ksi

Shear strain due to axial static load is taken as

Da σ a,st
γ a,st = [LRFD 14.7.5.3.3-3]
GSi
where Da is a dimensionless coeicient taken as 1.4 for a rectangular bearing and 1.0 for a
circular bearing. Shear strain due to axial cyclic load is taken similarly.
Shear strain due to axial static and cyclic loads are calculated as

Da σ a,st (1.4 )( 0.689 )

γ a,st = = = 1.174
GSi ( 0.140 )(5.867 )

Da σ a,cy (1.4 )( 0.636 )

γ a,cy = = = 1.084
GSi ( 0.140 )(5.867 )

Rotation due to static load is calculated as

θst = θinitial prestress + θDL girder + θcreep+shrinkage before girder erection

+ θcreep+shrinkage after girder erection + θDL slab + θdifferential shrinkage of slab
= ( −9.260 + 3.597 − 2.900 − 1.450 + 4.545 + 2.370 )10−3
= −3.098 (10−3 ) rads

Rotation due to cyclic load is calculated as

Bearings 21

θcy = θLL lane + θLL truck

= ( −0.997 + 1.896 )10−3 = 2.893 (10−3 ) rads
Apply the 0.005 rads tolerance for static rotation as both positive and negative:

θst− = ( −3.098 )10−3 − 0.005 = −8.098 (10−3 ) rads

θst+ = ( −3.098 )10−3 + 0.005 = 1.902 (10−3 ) rads

Shear strain due to rotation is calculated as

2
 L  θ 
γ r = Dr    s  [LRFD 14.7.5.3.3-6]
 hri   n 

where Dr is a dimensionless coeicient taken as 0.5 for a rectangular bearing and 0.375 for a
circular bearing; n is the number of internal elastomeric layers, allowing n to be augmented
by ½ for each exterior layer having a thickness that is equal to or greater than half the thick-
ness of an interior layer. θs is maximum static or cyclic rotation angle. Shear strains due to
static and cyclic rotations are calculated as

2 2
 L  θ   8   1.902 × 10−3 
γ r,st = Dr    st  = ( 0.5 )    = 0.0487
 hri   n   0.5   5

2 2
 L   θcy   8   2.893 × 10−3 
γ r,cy = Dr    = ( 0.5 )     = 0.0741
 hri   n  0.5 5

Longitudinal deformations due to static and cyclic loads are calculated as

∆ st = 0.65 ( ∆ T ) + ∆ creep + shrinkage after girder erection + ∆ DL slab + ∆ differential shrinkage of slab
= 0.730 − 0.418 + 0.333 − 0.071 = 0.574 in.

∆ cy = ∆ LL lane + ∆ LL truck = 0.109 + 0.208 = 0.317 in.

Shear strains due to shear deformation are calculated as

∆s
γs =
hrt [LRFD 14.7.5.3.3-10]

Shear strains due to static and cyclic longitudinal deformations are calculated as

∆ st 0.574
γ s,st = = = 0.230
hrt 2.5

∆ cy 0.317
γ s,cy = = = 0.127
hrt 2.5

Check service limit state requirements (LRFD Article 14.7.5.3.3) for the longitudinal direction:

γ a,st = 1.174 < 3.0 O.K.

22 Bridge Engineering Handbook, Second Edition: Substructure Design

(γ a,st ) (
+ γ r,st + γ s,st + 1.75 γ a,cy + γ r,cy + γ s,cy )
= (1.174 + 0.0487 + 0.230 ) + (1.75 )(1.084 + 0.0741 + 0.127 )
= 3.702 < 5.0 O.K.
Check service limit state requirements (LRFD Article 14.7.5.3.3) for the transverse direction:
γa,st = 1.174 (same as longitudinal direction) < 3.00 O.K.
γa,cy = 1.084 (same as longitudinal direction)
θst = θcy = 0.000
θst+ = 0.005 rads

2 2
W  θ   22   0.005 
γ r,st = Dr    st  = ( 0.5 )   = 0.968
 hri   n   0.5   5 

2 2
 W   θcy   22   0.000 
γ r,cy = Dr    = ( 0.5 )  0.5   5 
  =0
 hri   n 

(γ a,st ) (
+ γ r,st + γ s,st + 1.75 γ a,cy + γ r,cy + γ s,cy )
= (1.174 + 0.968 + 0 ) + (1.75 )(1.084 + 0 + 0 )
= 4.039 < 5.0 O.K.

Step 5: Check condition immediately before deck placement

Pst = 47.9 kip

Pcy = 0.0 kip

Pst 47.9
σ a,st = = = 0.272 ksi
LW ( 8 )( 22 )

Pcy 0
σ a,cy = = = 0 ksi
LW ( 8 )( 22 )

Check the longitudinal direction:

Da σ a,st (1.4 )( 0.272 )

γ a,st = = = 0.464 < 3.0 O.K.
GSi ( 0.140 )(5.867 )

Da σ a,cy (1.4 )( 0 )
γ a,cy = = =0
GSi ( 0.140 )(5.867 )
Rotation due to static load is calculated as

θst = θinitial prestress + θDL girder + θcreep+shrinkage before girder erection + θcreep+shrinkage after girder erection
= ( −9.260 + 3.597 − 2.900 − 1.450 )10−3
= −10.013 (10−3 ) rads

θcy = θLL lane + θLL truck = 0 rads

Bearings 23

Apply the 0.005 rads tolerance as negative:

θst = ( −10.103)(10−3 ) − 0.005 = −15.103 (10−3 ) rads

2 2
 L  θ   8   15.013 × 10−3 
γ r,st = Dr    st  = ( 0.5 )    = 0.384
 hri   n   0.5   5

2 2
 L   θcy   8   0.000 
γ r,cy = Dr    = ( 0.5 )  0.5   5 
  =0
 hri   n 

he only signiicant horizontal displacement imposed upon the bearings immediately prior
to slab placement is creep and shrinkage that occurs ater the girder are erected upon the
bearings. he thermal displacement range during the short interval between when the girders
are erected and the slab is poured is deemed to be negligible.

∆ st = −0.418 in.

∆ cy = 0 in.

∆ st 0.418
γ s,st = = = 0.167
hrt 2.5

∆ cy 0
γ s,cy = = =0
hrt 2.5

(γ a,st ) (
+ γ r,st + γ s,st + 1.75 γ a,cy + γ r,cy + γ s,cy )
= ( 0.464 + 0.384 + 0.167 ) + (1.75 )( 0 + 0 + 0 )
= 1.015 < 5.0 O.K.

Check the transverse direction:

γa,st = 0.464 (same as longitudinal direction) < 3.00 O.K.
γa,cy = 0.000 (same as longitudinal direction)
θst = θcy = 0.000
θst+ = 0.005 rads

2 2
W  θ   22   0.005 
γ r, st = Dr    st  = ( 0.5 )   = 0.968
 hri   n   0.5   5 

2 2
 W   θcy   22   0.000 
γ r,cy = Dr    = ( 0.5 )   =0

 hri   n   0.5   5 

γ s, st = γ s,cy = 0.000
24 Bridge Engineering Handbook, Second Edition: Substructure Design

(γ a, st ) (
+ γ r, st + γ s, st + 1.75 γ a,cy + γ r,cy + γ s,cy )
= ( 0.464 + 0.968 + 0.000 ) + (1.75 )( 0 + 0 + 0 )
= 1.432 < 5.0 O.K.

Step 6: Evaluate stability of the bearing

Per LRFD Article 14.7.5.3.4, bearings shall be investigated for instability at the service limit
state load combination. First, consider stability in the longitudinal direction.

hrt 2.5
1.92(1.92 )
A= L = 8.0 = 0.457 [LRFD 14.7.5.3.4-2]
2L ( 2 )( 8 )
1+ 1+
W 22

2.67 2.67
B= = = 0.311 [LRFD 14.7.5.3.4-3]
( Si + 2.0) 1 +  (5.867 + 2) 1 + 8 
 L 
4W  ( 4 )( 22 ) 

Because 2A = 2(0.457) = 0.914 > B = 0.311, further investigation is required.

he bridge is ixed against horizontal translation in the longitudinal direction, requiring that

σ s = σ a, st + σ a,cy = 0.689 + 0.636 = 1.325 ksi

GSi ( 0.140 )(5.867 )

σ s = 1.325 ksi < = = 5.626 ksi [LRFD 14.7.5.3.4-5] O.K.
A − B 0.457 − 0.311

Next, consider stability in the transverse direction.

hrt 2.5
1.92 (1.92 )
A= W = 22 = 0.0856 [LRFD 14.7.5.3.4-2]
2W ( 2 )( 22 )
1+ 1+
L 8

2.67 2.67
B= = = 0.201 [LRFD 14.7.5.3.4-3]
( Si + 2.0) 1 +  (5.867 + 2) 1 + 22 
W
4L  ( 4 )( 8 ) 

Because 2A = 2(0.0856) = 0.171 < B = 0.201, no further investigation is required.

Step 7: Determine required thickness of steel reinforcement
At the service limit state:

3hri σ s
hs ≥ [LRFD 14.7.5.3.5-1]
Fy
Bearings 25

where Fy is the yield strength of steel reinforcement = 36 ksi; and σs is average compressive
stress due to total load at the service limit state.

3hri σ s ( 3)( 0.5 )(1.325 )

hs ≥ = = 0.055 in.
Fy 36

At the fatigue limit state:

2hri σ L
hs ≥ [LRFD 14.7.5.3.5-2]
∆FTH

where σL is average compressive stress due to live load; ΔFTH is constant amplitude fatigue
threshold for Category A, as speciied in LRFD Article 6.6.

2hri σ L ( 2 )( 0.5 )( 0.636 )

hs ≥ = = 0.032 in.
∆FTH 20

Since the minimum thickness of steel reinforcement hs,min = 0.0625 in. is speciied in the cur-
rent AASHTO M251 speciication, the steel shims shall be at least 1/16 in. thick.
Step 8: Determine low temperature requirements for the constituent elastomer
For the purpose of bearing design, the AASHTO LRFD classiies all bridges in the United
States as being in either Zone A, B, C, D, or E. LRFD Figure 14.7.5.2-1 shows Minneapolis as
being within Zone D. Zone D is associated with a 50-year low temperature of −45°F (−42.8°C).
LRFD Table 14.7.5.2-1 requires a Grade 4 elastomer for bridges located in Zone D unless spe-
cial force provisions are incorporated into the design. When special force provisions are incor-
porated into the design, a Grade 3 elastomer is permissible. In summary, LRFD Article 14.7.5.2
allows three options:
Option 1: Specify a Grade 4 elastomer and determine the shear force transmitted by the
bearing in accordance with LRFD Article 14.6.3.1.
Option 2: Specify a Grade 3 elastomer and provide a low-friction sliding surface, in which
case the shear force transmitted by the bearing shall be assumed as twice that computed in
accordance with LRFD Article 14.6.3.1.
Option 3: Specify a Grade 3 elastomer without providing a low-friction sliding surface,
in which case the shear force transmitted by the bearing shall be assumed as four times that
computed in accordance with LRFD Article 14.6.3.1.
Step 9: Calculate approximate instantaneous dead, long-term dead, and live load compressive
deformation of the bearing
Limiting instantaneous live load delections is important to ensure that deck expansion
joints are not damaged. Steel reinforced elastomeric bearings exhibit nonlinear compres-
sive load-delection behavior. Compressive stifness of an elastomeric layer substantially
increases with increasing shape factor. he total compressive deformation of an elastomeric
bearing is equal to the sum of the compressive deformation of all its constituent elastomeric
layers.
LRFD commentary allows an assumed linear relationship between compressive stress and
compressive strain. Speciically, compressive strain can be estimated as
26 Bridge Engineering Handbook, Second Edition: Substructure Design

σ
ε= [LRFD C14.7.5.3.6-1]
6GSi2

hus, initial dead load deformation can be estimated as follows:

σ a,st 0.689
ε di = = = 0.0238
6GSi2 ( 6 )( 0.140 )(5.867 )2
Note that the smallest acceptable value of shear modulus has been used. his will result in
the largest compressive deformation. Because the bearing is composed of four interior layers
and two exterior layers all having essentially the same shape factor, the total initial dead load
delection can be estimated as

δ d = ∑ ε dihri = ( 0.0238 )(5 )( 0.5 ) = 0.06 in.

Long-term dead load delection includes the efects of creep as follows:

δ lt = δ d + acr δ d [LRFD 14.7.5.3.6-3]

where acr is a factor representing approximate creep deformation divided by initial dead load
deformation.
For an elastomer having a Shore A Hardness of 60 (assumed shear modulus at 73°F between
0.130 ksi and 0.200 ksi), LRFD Table 14.7.6.2-1 estimates acr as being 0.35. Hence,

δ lt = δ d + acr δ d = 0.06 + ( 0.35 )( 0.06 ) = 0.081 in.

Similarly, the instantaneous live load deformation can be estimated as follows:

σ a,cy 0.636
ε Li = = = 0.022
6GSi2 ( 6 )( 0.140 )(5.867 )2

δ L = ∑ ε Lihri = ( 0.022 )(5 )( 0.5 ) = 0.055 in.

Step 10: Consider hydrostatic stress

he bearing has no externally bonded steel plates. herefore, hydrostatic stress is not a
consideration.
Step 11: Evaluate the need for providing anchorage against slippage
he traditional anchorage check contained in previous editions of AASHTO design codes has
been to compare the maximum horizontal force induced in the elastomeric bearing versus
the incipient force required to cause the bearing to slip. his check was generally performed
using service loads and assumed a friction coeicient of 0.20 between the elastomer and the
concrete surface.
he maximum shear displacement of the bearings occurs at the extreme low temperature in
the absence of live loading.

∆ = −0.418 − 0.071 + 0.333 − 0.730 = −0.886 in.

Bearings 27

he maximum longitudinal force induced in the elastomeric bearing, assuming a Grade 4

elastomer, as a result of this shear displacement is

GAbr ∆ s ( 0.190 )( 22 )( 8 )( 0.886 )

Hs = = = 11.85 kip O.K.
hrt 2.5

H sliding = µPDL = µPst = ( 0.2 )(121.2 ) = 24.24 kip > H s = 11.85 kip

he upper range shear modulus of 0.190 ksi (1.31 MPa) is the critical value in this calculation.
It represents 115% of the nominal speciied value. LRFD Article 14.6.3.1 further requires that
the superstructure and substructure be designed to transmit, at the strength and extreme
limit states, the horizontal forces induced by sliding friction or shear deformation of lexible
bearing elements.
Article 14.7.5.4 of the current LRFD speciications requires a check of rotation versus axial strain
for bearings without externally bonded steel plates. A restraint system is required whenever

θs 3ε a
≥ [LRFD 14.7.5.4-1]
n Si

where εa is total of static and cyclic average axial strain taken as positive for compression in
which the cyclic component if multiplied by 1.75 from the applicable service load combination
in AASHTO LRFD Table 3.4.1-1; θs is total of static and cyclic maximum service limit state
design rotations of the elastomer speciied in which the cyclic component is multiplied by 1.75.

θs = θst + 1.75θcy = (1.902 )(10−3 ) + (1.75 )( 2.893)(10−3 ) = 6.965 × 10−3 rads

σ st + 1.75σ cy 0.689 + (1.75 )( 0.636 )

εa = = = 0.0623
6GSi2 ( 6)( 0.140)(5.867 )2

θs 6.965 × 10−3 3ε ( 3)( 0.0623)

= = 0.00139 > a = = 0.0319
n 5 S 5.867
herefore, no restraint system is required.

1.7.2 Design Example 2—Longitudinally Guided Disc Bearing

1.7.2.1 Given
For a steel box girder bridge, the service dead load is 680 kips (3,025 kN). he service live load without impact
is 320 kips (1423 kN). he horizontal strength limit state load is 640 kips (2847 kN). he allowable compres-
sive stress for the polyether urethane material constituting the disc is 5.00 ksi (34.5 MPa). he compressive
stress–strain relationship for the disc may be estimated as σ = E(1+S2)ε, where E is Young’s modulus. For
the polyether urethane used in this bearing, E may be taken as 10 ksi (68.9 MPa). Long-term creep is taken
as 20% of dead load compressive deformation. he disc element is sandwiched by upper and lower bearing
plates having a yield strength of 50 ksi (345 MPa). he 95 ksi (655 MPa) yield strength shear resisting pin is
threaded (12 threads per inch) into the lower bearing plate and bears against a hole in the upper bearing plate.
A longitudinally guided disc bearing difers from the ixed disc bearing depicted in Figure 1.12 in
that it incorporates a horizontal PTFE/stainless steel sliding interface and guide bars similar to those
depicted in the spherical bearing shown in Figure 1.14. he top of the upper bearing plate is recessed
3/32  in. (2.4  mm) for 3/16 in. (4.8 mm) thick unilled PTFE having a diameter of 18½ in. (470 mm).
28 Bridge Engineering Handbook, Second Edition: Substructure Design

he maximum service limit state rotation for the bearing is 0.020 rads. he maximum strength limit
state rotation for the bearing is 0.029 rads. Both these rotations include a 0.005 rads allowance for
uncertainties.

1.7.2.2 Requirements
Perform the following design calculations for a longitudinally guided disc bearing in accordance with
the AASHTO LRFD Bridge Design Speciications, 6th edition (2012).
• Determine the required diameter of the steel shear-resisting pin
• Determine the required diameter of the polyether urethane disc
• Verify the adequacy of a 1–7/8 in. (48 mm) thick disc
• Determine the minimum length of engagement and check the adequacy of the shear-resisting pin
for combined lexure and shear
• Check PTFE average contact stresses and edge contact stress

1.7.2.3 Solution
he design of the disc is governed by AASHTO LRFD Article 14.7.8. he design of the PTFE is gov-
erned by AASHTO LRFD Article 14.7.2. he design of the shear-resisting pin is governed by AASHTO
LRFD Article 6.7.6. Strength limit state resistance factors for shear, bearing, and lexure are taken from
AASHTO LRFD Article 6.5.4.2.
Step 1: Determine required diameter of the shear-resisting pin
he shear force associated with the horizontal strength limit state load determines the mini-
mum diameter of the steel shear-resisting pin.

2
H strength ≤ ϕ v (0.58)( Fy,pin )(π )Dpin,eff /4

0.9743
Dpin,eff = Dpin −
ntpi

where Hstrength is the horizontal strength limit state load (kips); φv is the shear resistance fac-
tor = 1.0; Fy, pin is the yield stress of the steel shear-resisting pin = 95 (ksi); Dpin is the nominal
diameter of the steel shear-resisting pin (in.); Dpin,ef is the minimum efective diameter of the
threaded portion of the steel shear-resisting pin calculated in accordance with ASME B1.1-
1989 (in.); and ntpi is the number of threads per inch.

4( Hstrength )
Dpin,eff ≥
ϕ v (0.58)( Fy,pin )π
4(640)
=
(1.00)(0.58)(95)(3.14)
= 3.85 in.

0.9743 0.9743
Dpin ≥ Dpin,eff + = 3.85 + = 3.93 in.
ntpi 12

To optimize machining operations, Dpin is selected as 5.25 in.

Step 2: Determine the required diameter of the polyether urethane disc
Bearings 29

Pservice (320 + 680)

Areqd ≥ = = 200 in.2
σ disc 5.000

where Pservice is the vertical service load (kips); σdisc is the allowable compressive stress in the
disc (ksi); and Areqd is the required net area of the disc (in.2).
he polyether urethane disc is essentially an annular ring with a steel shear-resisting pin in
the center. A 1/16 in. gap separates the pin from the inside vertical edge of the annular disc.
he outer edge of the disc is V-shaped as depicted in Figure 1.12. he V-shape accommodates
bulging under load. Each leg of the “V” forms a 30° angle with the vertical.

Dinside = 5.25 + 2(0.0625) = 5.375 in.

Alost = π(Dinside )2 /4
= (3.1416)(5.375)2 /4
= 22.69 in.2

where Dinside is the inside diameter of the disc; and Alost is the voided central area of the disc (in.2).

2
πDdisc
Areqd ≤ − Alost
4

where Ddisc is the outside diameter of the disk at its mid-depth.

4( Areqd + Alost )
Ddisc ≤
π
4(200 + 22.69)
=
3.1416
= 16.84 in.

Establish a practical manufacturing diameter of the top and bottom bearing surfaces of the
disk, accounting for the “V”-shaped notch.

Dbase ≥ 16.84 + (1.875)(tan30°) = 17.92 in. → Use 18 in.

Step 3: Verify adequacy of the 1–7/8 in. thickness of the polyether urethane disc

Ddisc = Dbase − (tdisc )(tan θ v ) = 18 − (1.875)(tan30°) = 16.92 in.

where tdisc is thickness of the disc (in.); θv is the angle of the “V”-shaped edge relative to vertical.

2 2
Adisc = π(Ddisc − Dinside )/4

(3.1416)[(16.92)2 − (5.375)2 ]
Adisc = = 202.16 in.2
4

Pservice (320 + 680)

σs = = = 4.947ksi < 5.000 ksi → O.K.
Adisc 202.16
30 Bridge Engineering Handbook, Second Edition: Substructure Design

Calculate the shape factor of the disc.

Plan area of bearing

S = Shape factor =
Bearing area free to bulge

Adisc 202.16
S= = = 2.028
(π )(Ddisc )(tdisc ) (3.1416)(16.92)(1.875)

AASHTO LRFD Article 14.7.8.3 limits instantaneous compressive deformation under total
service load to not more than 10% of the thickness of the unstressed disc. It additionally lim-
its additional deformation due to creep to no more than 8% of the unstressed disc thickness.
his article further proscribes lit of of component elements of the disc bearing, efectively
imposing limits on allowable service limit state rotation.
Calculate the instantaneous compressive deformation of the disk under total service load and
compare with the allowable deformation.

1 1
ε si = 2
σs = σ s = 0.0196σ s
E(1 + S ) (10)(1 + 2.0282 )

ε si = (0.0196)(4.947) = 0.0970 in./in.

δ si = ε sitdisc

δ si = (0.0970)(1.875) = 0.182 in.

where εsi is the instantaneous compressive strain in the disk under full service load (in./in.);
and δsi is the instantaneous compressive deformation of the disk under full service load (in.).

δ si,allowable = (0.10)(1.875) = 0.188 in. > 0.182 in. = δ si O.K.

Calculate the additional creep deformation of the disc under dead load and compare with the
allowable deformation.

Pd 320
σd = = = 1.583 ksi
Adisc 202.16

ε di = 0.0196 σ d = (0.0196)(1.583) = 0.0310 in./in.

δ di = ε ditdisc = (0.0310)(1.875) = 0.058 in.

δ creep = 0.20 δ di = (0.20)(0.058) = 0.012 in.

where Pd is the dead load (kips); σd is the average dead load compressive stress on the disc
(ksi); εdi is the instantaneous compressive strain in the disk under dead load (in./in.); and δdi
is the instantaneous compressive deformation of the disk under dead load (in.).

δ creep,allowable = (0.08)(1.875) = 0.150 in. > 0.012 in. = δ creep O.K.

Bearings 31

Compare rotation at litof to maximum service limit state rotation.

2(δ si ) 2(0.182)
θs,liftoff = = = 0.022 rads ≥ 0.020 rads = θs O.K.
Ddisc 16.92

Step 4: Determine minimum engagement length and check combined lexure and shear on the
shear-resisting pin
he required diameter of the steel shear-resisting pin has already been determined in Step 1.
AASHTO LRFD Article 6.7.6 further stipulates the design of the steel shear-resisting pin as it
relates to bearing and combined lexure and shear. he pin is threaded into the lower bearing
plate and bears against a hole in the upper bearing plate. he minimum engagement length
of the pin against each bearing plate is determined by checking against the allowable bearing
force. he maximum bending moment in the pin is calculated from the required engagement
length and the compressed height of the disc under dead load.

H strength ≤ ϕ b (1.5)( Lengage )(Dpin,eff )( Fy ) [LRFD 6.7.6.2.2-1 and 2]

where Lengage is the engagement length of the pin with each bearing plate (in.); Fy is the lesser of
the yield strengths of the pin and bearing plates (ksi); and φb is the bearing resistance factor = 1.0.

0.9743 0.9743
Dpin,eff = Dpin − = 5.25 − = 5.169 in.
ntpi 12

H strength 640
Lengage ≥ = = 1.651 in.
(ϕ b )(1.5)(Dpin,eff )( Fy ) (1.0)(1.5)(5.169)(50)

Lengage
d = tdisc − δ di +
2
1.651
= 1.875 − 0.058 + = 2.642 in.
2

where d is the distance from the point of maximum bending moment in the pin (top of lower
bearing plate) to the resultant of the bearing force in the upper bearing plate (in.).

M u = H strength (d ) = (640)(2.642) = 1691 in. kips

3
Dpin,eff 5.1693
Z= = = 23.018 in.3
6 6

where Mu is the strength limit state maximum moment in the pin; and Z is the plastic section
modulus of the pin (in.3).

ϕ f M n = Z ( Fy,pin ) = (1.00)(23.018)(95) = 2187 in. kips > 1690 in. kips

32 Bridge Engineering Handbook, Second Edition: Substructure Design

where φf is the lexure resistance factor = 1.0; and Mn is the nominal plastic moment capacity
of the pin.
Check combined lexure and shear.

3
6( Mu )  2.2 (Vu ) 
+  ≤ 0.95 [LRFD 6.7.6.2.1-1]
(
ϕ f Dpin,eff
3
)(
Fy,pin ) ( )(
 ϕ v Dpin,eff
2
Fy,pin ) 

3
6(1691)  2.2(640) 
+ = 0.944 ≤ 0.95 O.K.
1.00(5.169) (95)  1.00(5.169) (95) 
3 2

Step 5: Check PTFE contact stresses

For conined sheet PTFE, AASHTO LRFD Article 14.7.2.4 limits average contact stress for
permanent loads to 3.0 ksi and average contact stress for all loads to 4.5 ksi at the service limit
state. Edge contact stress for all loads at the service limit state is further limited to 5.5 ksi.

π 2 3.1416
Aptfe = Dptfe = (18.5)2 = 268.8 in.2
4 4

where Dptfe is the diameter of the conined PTFE sheet (in.); and Aptfe is the plan area of the
conined PTFE sheet (in.2).

Pd 320
σ d,ptfe = = = 1.190 ksi < 3.0 ksi O.K.
Aptfe 268.8

where σd,ptfe is the average dead load contact stress on the PTFE.

Pservice (320 + 680)

σ s,ptfe = = = 3.720 ksi < 4.5 ksi O.K.
Aptfe 268.8

where σs,ptfe is the average service load contact stress on the PTFE.
Edge contact stress is evaluated by calculating the moment induced in the polyether urethane
disc element due to the maximum service limit state rotation. his moment is transferred
through the PTFE by contact stresses.

π 4 4
π
I disc = (Ddisc − Dinside ) = (16.924 − 5.3754 ) = 3982 in.4
64 64

θs θ
M s = 0.5( Ec )( I ) = 0.5( E )(1 + S 2 ) s
t t

0.020
M s = 0.5(10)(1 + 2.0282 )(3982) = 1086 in. kips
1.875

π 3 3.1416
Sptfe = Dptfe = (18.5)3 = 621.6 in.3
32 32

where Ms is the moment induced in the polyether urethane disc element due to the maximum
service limit state rotation; and Sptfe is the section modulus of the PTFE surface.
Bearings 33

Pservice M s (320 + 680) 1086

σ ptfe,edge = + = + = 5.467 ksi < 5.500 ksi O.K.
Aptfe Sptfe 268.8 621.6

where σ ptfe,edge is the maximum compressive stress on a PTFE edge.

he upper and lower bearing plates, sole plate, masonry plate, and bolted connections need
to be designed to transfer all loads between the superstructure and the substructure. As part
of a continuous load path, the bearing plates need to be designed of suicient thickness to
transfer to the sole and masonry plates the same horizontal loads imposed upon the steel
shear-resisting pins. Stainless steel sliding surfaces need to be detailed to provide suicient
travel distance to accommodate all anticipated movements. Clearances must be adequate to
accommodate unrestrained service limit state movements. Additionally, guide bars need to
be designed and detailed to accommodate the transfer of transverse loads between the sole
plate and the upper bearing block. As noted earlier, it is important that bearings be designed
and detailed to allow for the inspection, maintenance, and future removal and replacement
of all sliding interface elements.

References
AASHTO. 2012. AASHTO LRFD Bridge Design Speciications, 6th Edition. American Association of State
Highway and Transportation Oicials, Washington, DC.
AASHTO/NSBA. 2004. Steel Bridge Bearing Design and Detailing Guidelines, G 9.1 – 2004. American
Association of State Highway and Transportation Oicial/National Steel Bridge Alliance Steel
Bridge Collaboration, Chicago, IL.
Lehman, D.E., C.W. Roeder, R. Larson, K. Curtin. 2003. Cotton duck Bearing Pads: Engineering Evaluation
and Design Recommendations. Prepared for the Washington State Transportation Commission.
Available at: http://www.wsdot.wa.gov/research/reports/fullreports/569.1.pdf
PCI. 2011. Precast Prestressed Concrete Bridge Design Manual. Precast/Prestressed Concrete Institute,
Chicago, IL.
Stanton, J.F., C.W. Roeder, P. Mackenzie-Helnwein, C. White, C. Kuester, B. Craig. 2008. Rotation Limits
for Elastomeric Bearings, NCHRP Report 596, Transportation Research Board, National Research
Council, Washington, DC.
Stanton, J.F., C.W. Roeder, T.I. Campbell. 1999. High Load Multi-Rotational Bridge Bearings, National
Cooperative Highway Research Program Report 432, Transportation Research Board, National
Research Council, Washington, DC.
Stanton, J.F., J.C. Taylor. 2010. Friction Coeicients for Stainless Steel (PTFE) Bearings, Report No. WHRP
10-01, Wisconsin Highway Research Program, Madison, WI.
Washington State Department of Transportation. 2011. Bridge Design Manual, Chapter 9. Washington
State Department of Transportation, Olympia, WA.
 2
Piers and Columns
2.1 Introduction ........................................................................................35
2.2 Structural Types ..................................................................................35
General • Selection Criteria
2.3 Design Loads .......................................................................................40
Live Loads • hermal Forces
2.4 Design Considerations .......................................................................42
Jinrong Wang Overview • Slenderness and Second-Order Efect • Concrete Piers
California Department and Columns • Steel and Composite Columns
of Transportation References........................................................................................................62

2.1 Introduction
Piers provide vertical supports for bridge spans at intermediate points and perform two main functions:
transferring superstructure vertical loads to the foundations and resisting horizontal forces acting on
the bridge. Although piers are traditionally designed to carry vertical loads, these days it is common for
designers to take into account the high lateral loads caused by seismic events. Even in some low-seismic
areas, designers are paying more attention to the ductility aspect of the design.
Piers are predominately constructed with reinforced concrete. Steel, to a lesser degree, is also used
for piers. Steel tubes illed with concrete, known as composite columns, have been used in some recent
projects in China and other countries.
his chapter deals only with piers or columns for conventional highway bridges, such as grade separa-
tions, overcrossings, overheads, underpasses, and simple river crossings. Reinforced concrete columns
will be discussed in detail, whereas steel and composite columns will be discussed briely. Substructures
for arch, suspension, segmental, cable-stayed, and movable bridges are excluded from this chapter.
Chapter 3 discusses the substructures for some of these special types of bridges.

2.2 Structural Types

2.2.1 General
“Pier” is usually used as a general term for any type of intermediate substructures located between hori-
zontal spans and foundations. However, from time to time, it is also used particularly for a solid wall in
order to distinguish it from columns or bents. From a structural point of view, a column is a member
that resists the lateral force mainly by lexure action, whereas a pier is a member that resists the lateral
force mainly by a shear mechanism. A pier consisting of multiple columns is oten called the “bent.”
here are several ways of deining pier types. One is by its structural connectivity to the super-
structure: monolithic or cantilevered. Another is by its sectional shape: solid or hollow; round, octag-
onal, hexagonal, or rectangular. It can also be distinguished by its framing coniguration: single- or
multiple-column bent; hammerhead or pier wall. Figure 2.1 shows a series of columns in a typical urban

35
36 Bridge Engineering Handbook, Second Edition: Substructure Design

FIGURE 2.1 Columns in a typical urban interchange.

FIGURE 2.2 Columns in Skyway structure of San Francisco–Oakland Bay Bridge.

interchange. he smooth monolithic construction not only creates an esthetically appealing structure
but also provides an integral system to resist the seismic forces. Figure 2.2 shows one example of water
crossings, the newly constructed Skyway of San Francisco–Oakland Bay Bridge.

2.2.2 Selection Criteria

Selection of the type of piers for a bridge should be based on functional, structural, and geometric
requirements. Esthetics is also a very important factor of selection because modern highway bridges are
Piers and Columns 37

FIGURE 2.3 Typical cross-section shapes of piers for overcrossings or viaducts on land.

FIGURE 2.4 Typical cross-section shapes of piers for river and waterway crossings.

(a) (b) (c)

FIGURE 2.5 Typical pier types for steel bridges: (a) Solid wall pier (b) Hammerhead pier (c) Rigid frame pier.

oten a part of the landscape of a city. Figure 2.3 shows a collection of typical cross-section shapes for
overcrossings and viaducts on land, and Figure 2.4 shows some typical cross-section shapes for piers of
river and waterway crossings. Oten times, pier types are mandated by government agencies or owners.
Many state Departments of Transportation in the United States have their own standard column shapes.
Solid wall piers, as shown in Figures 2.5a and 2.6, are oten used at water crossings because they can
be constructed to proportions that both are slender and streamlined. hese features lend themselves
well for providing minimal resistance to water lows.
38 Bridge Engineering Handbook, Second Edition: Substructure Design

Hammerhead piers, as shown in Figure 2.5b, are oten found in urban areas where space limitation is
a concern. hey are used to support steel girder or precast prestressed concrete girder superstructures.
hey are esthetically appealing and generally occupy less space, thereby providing more room for the
traic underneath. Standards for the use of hammerhead piers are oten maintained by individual trans-
portation department.
A bent consists of a cap beam and supporting columns forming a frame. Bents, as shown in Figure 2.5c
and Figure 2.7, can be used either to support a steel girder superstructure or as an integral bent where
the cast-in-place construction technique is used. he columns can be either circular or polygonal in
cross section. hey are by far the most popular forms of piers in the modern highway systems.

Monolithic Fixed or expansion

(a)

Monolithic Fixed or expansion

(b)

FIGURE 2.6 Typical pier types and conigurations for river and waterway crossings: (a) Hammerhead (b) Solid wall.

(a) (b)

FIGURE 2.7 Typical pier types for concrete bridges: (a) Bent for precast girders (b) Bent for cast-in place griders.
Piers and Columns 39

A pile extension pier consists of a drilled shat as the foundation and the circular column extended
above the shat to form the substructure. An obvious advantage of this type of pier is that they occupy
a minimal amount of space. Widening an existing bridge in some instances may require pile extensions
because space limitation precludes the use of other types of foundations.
Selection of proper pier type depends on many factors. First, it depends on the type of superstruc-
ture. For example, steel girder superstructures are normally supported by cantilevered piers, whereas
the cast-in-place concrete superstructures are normally supported by monolithic bents. Second, it
depends on the locations of bridges. Pier walls are preferred on river crossings, where debris is a concern
and hydraulics dictates. Column bents are typically used in street crossings and highway separations.
Multiple pile extension bents are commonly used on slab bridges. Last, the height of piers also dictates
the type of selection of piers. he taller piers oten require hollow cross sections in order to reduce the
weight of the substructure. his then reduces the load demands on the costly foundations. Table 2.1
summarizes the general type selection guidelines for diferent types of bridges.

TABLE 2.1 General Guidelines for Selecting Pier Types

Location Tall or Short Piers Applicable Pier Types
Steel Superstructure
Over water Tall piers Pier walls or hammerheads (T-piers) (Figure 2.5a and b),
hollow cross sections for most cases, cantilevered, could use
combined hammerheads with pier wall base and step
tapered shat
Short piers Pier walls or hammerheads (T-piers) (Figure 2.5b and c),
solid cross sections, cantilevered
On land Tall piers Hammerheads (T-piers) and possibly rigid frames (multiple
column bents) (Figure 2.5b and c), hollow cross sections for
single shat and solid cross sections for rigid frames,
cantilevered
Short piers Hammerheads and rigid frames (Figure 2.5b and c), solid
cross sections, cantilevered

Precast prestressed concrete superstructure

Over water Tall piers Pier walls or hammerheads (Figure 2.6), hollow cross sections
for most cases, cantilevered, could use combined
hammerheads with pier wall base and step tapered shat
Short piers Pier walls or hammerheads, solid cross sections, cantilevered
On land Tall piers Hammerheads and possibly rigid frames (multiple column
bents), hollow cross sections for single shats and solid cross
sections for rigid frames, cantilevered
Short piers Hammerheads and rigid frames (multiple column bents)
(Figure 2.7a), solid cross sections, cantilevered

Cast-in-place concrete superstructure

Over water Tall piers Single shat pier (Figure 2.6), superstructure will likely cast by
traveled forms with balanced cantilevered construction
method, hollow cross sections, monolithic, ixed at bottom
Short piers Pier walls (Figure 2.6), solid cross sections, monolithic, ixed
at bottom
On land Tall piers Single or multiple column bents, solid cross sections for most
cases, monolithic, ixed at bottom
Short piers Single or multiple column bents (Figure 2.7b), solid cross
sections, monolithic, pinned at bottom
40 Bridge Engineering Handbook, Second Edition: Substructure Design

2.3 Design Loads

Piers are commonly subjected to forces and loads transmitted from the superstructure and forces acting
directly on the substructure. Some of the loads and forces to be resisted by piers include the following:
• Dead loads
• Live loads and impact from the superstructure
• Wind loads on the structure and the live loads
• Centrifugal force from the live loads
• Longitudinal force from live loads
• Drag forces due to the friction at bearings
• Stream low pressure
• Ice pressure
• Earthquake forces
• hermal and shrinkage forces
• Ship impact forces
• Force due to prestressing of superstructure
• Forces due to diferential settlement of foundations
he efect of temperature changes and shrinkage of the superstructure needs to be considered when
the superstructure is rigidly connected with the supports. Where expansion bearings are used, forces
caused by temperature changes are limited to the frictional resistance of the bearings.
he readers should refer to Chapter 6 of Bridge Engineering Handbook, Second Edition: Fundamentals
for more details about various loads and load combinations and Chapter 7 of Bridge Engineering
Handbook, Second Edition: Seismic Design about earthquake loads. In the following, however, two load
cases, live loads and thermal forces, are discussed in detail because they are two of the most common
loads on the piers but are oten applied incorrectly in the design.

2.3.1 Live Loads

Bridge live loads are the loads speciied or approved by the contracting agencies and owners. hey are
usually speciied in the design codes such as AASHTO LRFD Bridge Design Speciications (AASHTO
2012). here are other special loading conditions peculiar to the type or location of the bridge structure,
which should be speciied in the contracting documents.
Live load reactions obtained from the design of individual member of the superstructure should not be
used directly for substructure design. hese reactions are based on maximum conditions for one beam and
make no allowance for the distribution of live loads across the roadway. Using these maximum loadings
would result in a pier design with an unrealistically severe loading condition and uneconomical sections.
For substructure design, the maximum reaction of design traic lane using either the standard truck
load or standard lane load or a combination of both should be used. In AASHTO LRFD (AASHTO 2012),
Section 3.6 speciies the width of design traic lane as 3.6 m (12 t) and three load combinations:
1. A design tandem combined with the design lane load.
2. A design truck with variable axial spacing combined with the design lane load.
3. Ninety percent of two design trucks spaced a minimum 15.2 m (50 t) between the lead axle of
one truck and the rear axle of the other truck, combined with 90% of the design lane load. he
distance between the 142.3 kN (32 kip) axle should be ixed at 4.3 m (14 t) (Figure 2.8).
Each state transportation agency may add one more load condition that considers its own permit
loads and their combination.
For the calculation of the actual beam reactions on the piers, the maximum lane reaction can be
applied within the design traic lanes as wheel loads and then distributed to the beams, assuming the
Piers and Columns 41

Roadway
** ** *
22ftft 22 ft
ft 22 ftft

66ftft 66ftft 6 ft
W W W
W W
W W
W W W
W

G1
G1 G2
G2 G3
G3 G4
G4 G5
G5 G6
G6

* Design traffic lane = 12 ft Wheel loading W = R2/2

No. of lanes = Roadway width (ft)/12 ft
Reduce to nearest whole number

50 ft

R1 R2 R3 R4 R5

FIGURE 2.8 Wheel load arrangement to produce maximum reaction at R2.

TABLE 2.2 Dynamic Load Allowance, IM

Component IM (%)
Deck joints—all limit states 75
All other components

• Fatigue and fracture limit state 15

• All other limit states 33

slab between the beams to be simply supported (Figure 2.8) if the bent is cantilevered. Wheel loads can
be positioned anywhere within the design traic lane with a minimum distance between lane boundary
and wheel load of 0.61 m (2 t). For integral bent cap, the bent should be modeled as a frame. he calcu-
lated reactions due to the wheel load should be applied to the beam element of this frame. he design
traic lanes and the live load within the lanes should be arranged to produce beam reactions that result
in maximum loads on the piers. hese reactions should be multiplied by a multiple presence factor, m,
as speciied in Section 3.6.1.1.2 of AASHTO LRFD (AASHTO 2012).
Live load reactions shall be increased due to impact efect. AASHTO LRFD (AASHTO 2012) refers to
this as the Dynamic load allowance, IM, and is listed here in Table 2.2.

2.3.2 Thermal Forces

Forces on piers due to thermal movements, shrinkage and prestressing of superstructures can become
signiicant on short, stif bents of prestressed concrete bridges with monolithic bents. Pier design should
be checked against these forces. Design codes or speciications normally specify the design temperature
range. Some codes even specify temperature distribution along the depth of the superstructure member.
42 Bridge Engineering Handbook, Second Edition: Substructure Design

he irst step in determining the thermal forces on the substructures for a bridge with monolithic
bents is to determine the point of no movement. Ater this point is determined, one can calculate the
relative displacement of any point along the superstructure to this point by the distance to this point
times the temperature range and times the coeicient of expansion. With known displacement at the top
and known boundary conditions at the top and bottom, the forces on the pier due to the temperature
change can be calculated by using the displacement times the stifness of the pier.
he determination of point of no movement is best demonstrated by the following example, which is
adapted from Memo to Designers issued by California Department of Transportation (Caltrans 1994).

Example 2.1
A 225.55-m (740-t) long and 23.77-m (78-t) wide concrete box girder superstructure is sup-
ported by 5 two-column bents. he size of the column is 1.52 m (5 t) in diameter, and the
heights vary between 10.67 m (35 t) and 12.80 m (42 t). Other assumptions are listed in the
calculations. he calculation is done through a table. Figure 2.9 demonstrates the calculation for
determining the point of no movement.

2.4 Design Considerations

2.4.1 Overview
Like the design of any structural component, the design of piers or columns is performed to fulill
strength and serviceability requirements. A pier as a structure component is subjected to combined
forces of axial, bending, and shear. For a reinforced concrete pier, the bending strength is axial force
dependent. he shear strength is also afected by bending and axial loads. To consider the actual behav-
ior of a longer column, the bending moment will be magniied by the axial force due to the P-Δ efect.
In current design practice, the bridge designers are paying increasing attention to the adverse efects
of earthquake. herefore, ductility consideration has become a very important factor for bridge design.
Failure due to scouring is also a common cause of failure of bridges. In order to prevent this type of
failure, the bridge designers need to work closely with the hydraulic engineers to determine adequate
depths of cover for the foundations and provide proper protection measures.

2.4.2 Slenderness and Second-Order Effect

he design of compression members must be based on forces and moments determined from an analy-
sis of the structure. Small delection theory is usually adequate for the analysis of beam-type mem-
bers. For compression members, however, the second-order efects must be considered. According to
AASHTO LRFD (AASHTO 2012), the second-order efect is deined as follows: “he presence of com-
pressive axial forces amplify both out-of-straightness of a component and the deformation due to non-
tangential loads acting thereon, therefore increasing the eccentricity of the axial force with respect
to the centerline of the component. he synergistic efect of this interaction is the apparent sotening
of the component, i.e., a loss of stifness.” To accurately assess this efect, a properly formulated large
delection nonlinear analysis can be performed. Discussions on this subject can be found in White
and Hajjar (1994), Galambos (1998), and Chapter 5 of Bridge Engineering Handbook, Second Edition:
Seismic Design. However, it is impractical to expect the practicing engineers to perform this type of
sophisticated analysis on regular bases. he moment magniication procedure given in AASHTO
LRFD (AASHTO 2012) is an approximate process that was selected as a compromise between accuracy
and ease of use. herefore, the AASHTO LRFD (AASHTO 2012) moment magniication procedure is
outlined in the following.
 Piers and Columns
740´

90´ 120´ 160´ 160´ 120´ 90´

X = 64´ X = 80´ X = 56´ 34´

166´ 220´ 100´ 156´

X Conv. PCC CIP P/S X Conv. PCC X

12–
6´–0´ 20´ 7.0´
70 ton 20´
piles Abut
35.0´ 40.0´ Traveled way
A1 Traveled way 2–5´Ø Column (typ) fixed to
40.0´ 42.0´ footing
40.0´ A7
B2
B3 B4 B5 B6

A1 B2 B3 B4 B5 B6 A7
I (Ft)4 1.38 61.36 61.36 61.36 61.36 61.36 102
L (Ft) 5.50 35.0 40.0 Sum 40.0 40.0 Sum 42.0 7.0 Sum
P (kips) @ 1” side sway 1200 + 618 + 415 = 2233 415 + 415 = 830 359 Will slide +600 = 959
D (distance from 1st member of frame) 0 90 210 0 160 0 90
P × D / 100 0 + 556 + 872 = 1428 0 + 664 = 664 0 + 540 = 540
Σ(P × D) / 100 1428 664 540
X= (100) = (100) = 64´ (100) = 80´ (100) = 56´
ΣP 2233 830 959

Notes: Assumptions: Fixed/fixed condition Pinned/fixed condition D.W. Abut 7 = 600 k (assume
linear up to 1” deflection)
Width of structure = 78´ A A
1. Super str. inf. rigid P (Col.) = 12EI P (Col.) = 3EI 3
Diameter of column = 5´–0´ 2. Columns fixed top and bottom L3 L
K/Pile @ 1” deflection = 100 kips 4321 1081 I (abut) = 78 (2.5)3
3. Abutment footing will slide @ @ 1” defl. = @ 1” defl. = 12
Point of no movement = X L 3 L 3
a force equal to D.W.
Refer to properties/piles table 10 10 = 102
4. E (piles) = 4 × 108 psi
E (columns) = 3 × 106 psi

43
FIGURE 2.9 Calculation of points of no movement.
44 Bridge Engineering Handbook, Second Edition: Substructure Design

When a compression member’s cross-sectional dimensions are small in comparison with its length,
the member is said to be slender. Whether a member can be considered slender is dependent on the
magnitude of the member’s slenderness ratio. he slenderness ratio of a compression member is deined
as, KLu/r, where K is the efective length factor for compression members; Lu is the unsupported length
of compression member; r radius of gyration = I/A; I the moment of inertia; A the cross-sectional area.
When a compression member is braced against sidesway, the efective length factor, K = 1.0 can be
used. However, a lower value of K may be used if further analysis demonstrated that a lower value is
warranted. Lu is deined as the clear distance between slabs, girders, or other members that is capable
of providing lateral support for the compression member. If haunches are present, then, the unsup-
ported length is taken from the lower extremity of the haunch in the considered plane (AASHTO LRFD
5.7.4.3). For a detailed discussion of the K-factor, refer to Chapter 18 of Bridge Engineering Handbook,
Second Edition: Fundamentals.
For a concrete column braced against sidesway, the efect of slenderness may be ignored as long as the
following condition is met (AASHTO LRFD 5.7.4.3):

KLu  12M1 
< 34 −  (2.1)
r  M 2 

where M1 and M2 are smaller and larger end moments on a compression member, respectively, the term
(M1/M2) is positive for single-curvature lexure.
For an unbraced concrete column, the efect of slenderness may be ignored as long as the following
condition is met (AASHTO LRFD 5.7.4.3):

KLu
< 22 (2.2)
r

If the slenderness ratio exceeds the above speciied limits, the efects can be approximated by the use
of moment magniication factor. If the slenderness ratio KLu/r exceeds 100, however, a more detailed
second-order nonlinear analysis will be required. Any detailed analysis should consider the inluence of
axial loads and variable moment of inertia on member stifness and forces and the efects of the dura-
tion of the loads.

M c = δ b M 2b + δ s M 2s (2.3)

he factored moments may be increased to relect efects of deformations as follows:

where M2b = moment on compression member due to factored gravity loads that result in no appre-
ciable sidesway calculated by conventional irst-order elastic frame analysis, always positive.
M2s = moment on compression member due to lateral or gravity loads that result in sidesway, Δ,
greater than Lu/1500, calculated by conventional irst-order elastic frame analysis, always positive. Lu is
in same unit as Δ.
he moment magniication factors are deined as follows:

Cm
δb = ≥ 1.0 (2.4)
P
1− u
φ K Pe

1
δs =
1−
∑ Pu (2.5)
φ K ∑ Pe
Piers and Columns 45

(Please note that the above two equations are revised in AASHTO 2012 edition.)
where φ K is stifness reduction factor; 0.75 for concrete, 1.0 for steel and aluminum members; Pu is
factored axial load; and Pe is Euler buckling load that is determined as follows:

π 2 EI
Pe = (2.6)
( KLu )2
Cm, a factor that relates the actual moment diagram to an equivalent uniform moment diagram, is
typically taken as 1.0. However, in the case where the member is braced against sidesway and without
transverse loads between supports, it may be taken by the following expression:

M 
Cm = 0.6 + 0.4  1b  (2.7)
 M 2b 

where M1b and M2b are smaller and larger end moments on a compression member, respectively, the ratio
(M1b/M2b) is positive for single-curvature lexure and negative for double-curvature lexure.
To compute the lexural rigidity EI for concrete columns, AASHTO LRFD (AASHTO 2012) ofers
two possible solutions taken as the greater of

Ec I g
+ Es I s
EI = 5 (2.8)
1 + βd

Ec I g
(2.9)
EI = 2.5
1 + βd

where Ec is the elastic modulus of concrete, Ig the gross moment inertia, Es the elastic modules of rein-
forcement, Is the moment inertia of reinforcement about centroidal axis, and βd is the ratio of maximum
factored permanent load moment to maximum factored total load moment and is always positive. It is
an approximation of the efects of creep, so that when larger moments are induced by loads sustained
over a long period of time, the creep deformation and associated curvature will also be increased.

2.4.3 Concrete Piers and Columns

2.4.3.1 Combined Axial and Flexural Strength
A critical aspect of the design of bridge piers is the design of compression members. We use AASHTO
LRFD Bridge Design Speciications (AASHTO 2012) as the reference code. he following discussion
provides an overview of some of the major criteria governing the design of compression members.
Under the Strength Limit State Design, the factored resistance is determined by the product of nomi-
nal resistance, Pn, and the resistance factor, ϕ. For nonprestressed members, a lower ϕ factor of 0.75 is
used for compression-controlled sections, whereas a higher ϕ factor of 0.9 is used for tension-controlled
sections. he value ϕ is linearly varied from 0.75 to 0.9 depending on the net tensile strain as follows:

d  (2.10)
0.75 ≤ 0.65 + 0.15  t − 1 ≤ 0.9
 c 

where c is distance from the extreme compression iber to the neutral axis, and dt is distance from the
extreme compression iber to the centroid of the extreme tension steel element.
46 Bridge Engineering Handbook, Second Edition: Substructure Design

2.4.3.1.1 Interaction Diagrams

Flexural resistance of a concrete member is dependent on the axial force acting on the member.
Interaction diagrams are usually used as aids for the design of the compression members. Interaction
diagrams for columns are usually created assuming a series of strain distributions and computing the
corresponding values of P and M. Once enough points have been computed, the results are plotted to
produce an interaction diagram.
Figure 2.10 shows a series of strain distributions and the resulting points on the interaction diagram.
In an actual design, though, a few points on the diagrams can be easily obtained and can deine the
diagram rather closely.

Pure Compression he factored axial resistance for pure compression, ϕPn, may be computed by
For nonprestressed members with spiral reinforcement

Pr = φPn = φ0.85 Po = φ0.85 0.85 fc′( Ag − Ast ) + Ast f y  (2.11)

For nonprestressed members with tie reinforcement:

Pr = φPn = φ0.80 Po = φ0.80 0.85 fc′( Ag − Ast ) + Ast f y  (2.12)

For design, pure compression strength is a hypothetical condition since almost always there will be
moments present due to various reasons. For this reason, AASHTO LRFD 5.7.4.4 limits the nominal
axial load resistance of compression members to 85% and 80% of the axial resistance at zero eccentricity,
Po, for spiral and tied columns, respectively.

Pure compression

ε
Axial load, Pn

εcu

Balanced failure

εy

εcu
E
Moment, Mn

εsu > εy

FIGURE 2.10 Strain distributions corresponding to points on interaction diagram.

Piers and Columns 47

Pure Flexure he rectangular section in this case is only subjected to bending moment and without any
axial force. he factored lexural resistance, Mr, may be computed by

  fy  
M r = φM n = φ  As f y d  1 − 0.6ρ  
  fc′  
(2.13)
  a
       = φ  As f y  d −  
  2

where

As f y
a= (2.14)
0.85bfc′

Balanced Strain Condition Balanced strain condition corresponds to the strain distribution where the
extreme concrete strain reaches 0.003 and the strain in reinforcement reaches yield at the same time. At
this condition, the section has the highest moment capacity. For a rectangular section with reinforce-
ment in one face or located in two faces at approximately the same distance from the axis of bending, the
balanced factored axial resistance, Pr, and balanced factored lexural resistance, Mr, may be computed by

Pr = φPb = φ 0.85 fc′bab + As′ f s′− As f y  (2.15)

and

  a  
M r = φM b = φ 0.85 fc′bab  d − d ′′ − b  + As′ f s′(d − d ′ − d ′′) + As f y d ′′ (2.16)
  2 

where

 600 
ab =   β1d
 600 + f y 
(f y in MPa; ab and d in mm ) (2.17)

 87,000 
   ab =   β1d
 87,000 + f y 
(f y in psi; ab and d ininch and)
 a − β1d ′ 
f s′= 0.003 
 a 
Es ≤ f y (f y in the same unit as Es ) (2.18)

2.4.3.1.2 Biaxial Bending

AASHTO LRFD 5.7.4.5 stipulates that the design strength of noncircular members subjected to biaxial
bending may be computed, in lieu of a general section analysis based on stress and strain compatibility,
by one of the following approximate expressions:
When the factored axial load, Pu ≥ 0.10φfc′Ag

1 1 1 1
= + − (2.19)
Prxy Prx Pry Po

when the factored axial load, Pu < 0.10φfc′Ag

48 Bridge Engineering Handbook, Second Edition: Substructure Design

M ux M uy
+ ≤1 (2.20)
M rx M ry

where Prxy is factored axial resistance in biaxial lexure; Prx, Pry are factored axial resistance corresponding
to Mrx and Mry, respectively; Mux, Muy are factored applied moment about the x-axis and y-axis, respec-
tively; Mrx, Mry are uniaxial factored lexural resistance of a section about the x-axis and y-axis, respec-
tively, corresponding to the eccentricity produced by the applied factored axial load and moment and

( )
Po = 0.85 fc′ Ag − As + As f y (2.21)

he above procedure is only used in special circumstances. Generally, designers rely on computer programs
based on equilibrium and strain compatibility to generate a moment–axial interaction diagram. For cases
like noncircular members with biaxial lexure, an interaction surface is required to describe the behavior.
Figure 2.11 shows a typical moment–axial load interaction surface for a concrete section. In a day-to-day prac-
tice, such a surface has little value to designers. Rather, the design program normally gives out a series of lines,
basically slices of the surface, at ixed interval, such as 15°. Figure 2.12 is an example of such plot.
From these lines, one can see that below the balanced condition the moment capacity increases with
the increase of axial load. So, when designing a column, it is not enough to simply take a set of maxi-
mum axial load with maximum bending moments. he following combinations should to be evaluated:
1. Mux max, corresponding Muy and Pu
2. Muy max, corresponding Mux and Pu
3. A set of Mux and Muy that gives largest Mu combined and corresponding Pu
4. Pu max and corresponding Mux and Muy

Pu
Po

θ
My (Mb, Pb)
Mu

M
x

Mu
x

M uy

FIGURE 2.11 he moment–axial load interaction surface for a noncircular section.

Piers and Columns 49

Pn vs. Mn vs. theta angle (0.95% steel)

15,000
90° (transverse)

Nominal axial load, Pn (K) 10,000 75°

60°
5,000 45°
30°
0 15°
0° (longitudinal)
–5,000
0 5,000 10,000 15,000 20,000 25,000
Nominal moment, Mn (K-ft)

FIGURE 2.12 Interaction diagrams generated by a column design program.

2.4.3.2 Shear Strength

Under the normal load conditions, the shear seldom governs the design of the column for conventional
bridges because the lateral loads are usually small compared to the vertical loads. However, in a seismic
design, the shear is very important. In recent years, great efort has been put forth on the evaluation of
shear strength of columns, especially on the interaction between shear and lexure in the plastic hinge
zone. AASHTO LRFD (2012) provides a general shear strength calculation procedure that applies for
both beams and columns. he concrete shear capacity component and the angle of inclination of diago-
nal compressive stresses are functions of the shear stress on the concrete and the strain in the reinforce-
ment on the lexural tension side of the member. It is rather involved and hard to use. ACI Code (2011)
has a set of simpler equations, but they do not address the shear strength in the plastic hinge zones. he
procedure presented by Paulay and Priestley (1992) overcomes both of those shortcomings but does
not include the efect of displacement ductility demand on the shear strength. he procedure adapted
by California Department of Transportation (Caltrans) in its Seismic Design Criteria (Caltrans 2013)
addresses all these factors and is presented here.
he shear strength at a section is given as follows:

Vn = Vc + Vs (2.22)

where Vc is the contribution of the concrete to shear strength and Vs the contribution of shear
reinforcement.

Vc = vc Ae (2.23)

Ae = 0.8 × Ag (2.24)

• Inside the plastic hinge zone

vc = Factor1 × Factor 2 × fc′ ≤ 0.33 fc′  ( MPa ) (2.25)

vc = Factor1 × Factor 2 × fc′ ≤ 4 fc′  ( psi )

50 Bridge Engineering Handbook, Second Edition: Substructure Design

• Outside the plastic hinge zone

vc = 0.25 × Factor 2 × fc′ ≤ 0.33 fc′  ( MPa ) (2.26)

vc = 3 × Factor 2 × fc′ ≤ 4 fc′ ( psi )

where

ρs f yh
0.025 ≤ Factor1 =
12.5
(
+ 0.305 – 0.083µ d ≤ 0.25 f yh inMPa ) (2.27)

ρs f yh
0.3 ≤ Factor1 =
0.150  ksi
(
+ 3.67 – µ d ≤ 3 f yh in ksi )
In Equation 2.27, the value of “ρs f yh ” shall be limited to 0.35 ksi. Figure 2.13 shows value of Factor 1
that varies over the range of displacement ductility demand ratios, µ d.

Pc
Factor 2 = 1 + ≤ 1.5 ( metric units ) (2.28)
13.8 × Ag

Pc
Factor 2 = 1 + ≤ 1.5 ( English units )
2000 × Ag

In Equation 2.28, Pc is in N (lb), and Ag is in mm2 (in2).

For members whose net axial load is in tension, Seismic Design Criteria does not count the concrete
in resisting shear, vc = 0.
For members subjected to minor tension, totally ignoring the shear strength of concrete may be
unnecessarily conservative. ACI Code (2011) uses the following multiplier to account for the reduction
of the strength due to tension, which is equivalent to Factor 2 of above equation:

 Pc 
Factor 2 =  1 +  ( metric units ) (2.29)
 3.45 Ag 

3.5
3
2.5
2 ρs fyh = 0.35 ksi
Factor 1

1.5 ρs fyh = 0.05 ksi

1
0.5
0
1 2 3 4 5 6 7 8 9
Ductility demand ratio, μd

FIGURE 2.13 Factor 1 versus displacement ductility demand ratio, µ d.

Piers and Columns 51

 Pc 
Factor 2 =  1 +  ( English units)
 500  Ag 

In Equation 2.29, Pc is in N (lb), and Ag is in mm2 (in2).

his multiplier should not be less than zero, where Pc is negative for tension,
where Ag is gross section area of the column; Ae is efective section area, can be taken as 0.8Ag; Pc is
axial force applied to the column; and fc′ is compressive strength of concrete.
he nominal shear contribution from reinforcement is given by

Av f yhd (2.30)
Vs =
s

for tied rectangular sections and

π Ah f yh D ′ (2.31)
Vs =  
2 s

for spirally reinforced circular sections. In these equations, Av is the total area of shear reinforcement
parallel to the applied shear force, Ah the area of a single hoop, fyh the yield stress of horizontal reinforce-
ment, D’ the diameter of a circular hoop, and s the spacing of horizontal reinforcement.

2.4.3.3 Ductility of Columns

he AASHTO LRFD (2012) introduces the term of ductility and requires that a structural system of
bridge shall be designed to ensure the development of signiicant and visible inelastic deformations
before failure.
he term ductility deines the ability of a structure and selected structural components to deform
beyond elastic limits without excessive strength or stifness degradation. In mathematic terms, the
ductility μ is deined by the ratio of the total imposed displacement Δ at any instant to that at the
onset of yield Δy. his is a measure of the ability for a structure, or a component of a structure,
to absorb energy. he goal of seismic design is to limit the estimated maximum ductility demand to
the ductility capacity of the structure during a seismic event. For concrete columns, the coninement
of concrete must be provided, and a good detailing practice must be followed to ensure a ductile
column.
According to AASHTO LRFD (2012), for a circular column, the transverse reinforcement for con-
inement inside the plastic hinge zones, ratio of spiral reinforcement to total volume of concrete core,
measured out-to-out of spirals, shall be determined as follows:

fc′
ρs = 0.12 (2.32)
fy

It is recommended that the coninement reinforcement outside the zones should be at least more than
half of that inside the zones but not less than

 Ag  fc′
ρs = 0.45  −1 (2.33)
 Ac  f yh
52 Bridge Engineering Handbook, Second Edition: Substructure Design

For a rectangular column, the total cross-sectional area (Ash) of rectangular hoop (stirrup) reinforce-
ment shall be either

fc′  Ag 
Ash = 0.30ahc −1 (2.34)
f yh  Ac 

or

fc′
Ash ≥ 0.12shc (2.35)
fy

where a is vertical spacing of hoops (stirrups) (in) with a maximum of 4 in, Ac is the area of column core
measured to the outside of the transverse spiral reinforcement (in2), Ag the gross area of column (in2),
Ash the total cross-sectional area (in2) of hoop (stirrup) reinforcement, fc′ the speciied compressive
strength of concrete (ksi), fyh the yield strength of hoop or spiral reinforcement (ksi), hc the core dimen-
sion of tied column in mm in the direction under consideration (in), ρs the ratio of volume of spiral
reinforcement to total volume of concrete core (out-to-out of spiral), and Pu the factored axial load (kip).

Example 2.2: Design of a Two-Column Bent

Problem: Design the columns of a two span overcrossing. he typical section of the structure is
shown in Figure 2.14. he concrete box girder is supported by a two-column bent and is sub-
jected to HL-93 loading. he columns are pinned at the bottom. herefore, only the loads at the
top of columns are given here. Table 2.3 lists all the forces due to HL-93 live load plus impact.
Table 2.4 lists the forces due to seismic loads.
Material data:
fc′ = 4.0 ksi (27.6 MPa) Ec = 3,605 ksi (24,855 MPa) Es = 29,000 ksi (199,946 MPa) fy = 60 ksi
(414 MPa)
Try a column size of 4ʹ (1.22 m) in diameter. Provide 32-#9 (32-#30) longitudinal reinforce-
ment. he reinforcement ratio is 1.78%.
Section properties:
Ag = 12.51 t 2 (1.16 m2) Ast = 26.0 in2 (16,774 mm2)
Ixc = Iyc = 12.46 t4 (0.1075 m4) Ixs = Iys = 0.3338 t4 (0.0029 m4)
he analysis follows the procedure discussed in Section 2.4.3.1. he moment and axial force
interaction diagram is generated and is shown in Figure 2.15.

8.5´ 21.6´ 8.5´

3.0´

+
27.0´

FIGURE 2.14 Example 2.2—typical section.

Piers and Columns 53

TABLE 2.3 Column Group Loads—Service

Live Load + Impact

Case 1 Case 2 Case 3

Dead Trans Long Axial Wind Brake Centrifugal Temperature
Load My max Mx max Nmax Wind on LL Force Force (My) (°C)
HS + IM Lane HS + IM Lane HS + IM Lane
My (k-t) 220 75 35 15 8 32 16 532 153 208 127 180
Mx (k-t) 148 67 26 545 289 50 25 192 86 295 2 0
P (kip) 1108 173 120 131 113 280 212 44 17 12 23 0

IM, dynamic load allowance.

TABLE 2.4 Unreduced Seismic Loads

Unreduced Seismic Forces (ARS)
Case 1 Case 2
Max Transverse Max Longitudinal
My—transverse (k-t) 4855 3286
Mx—longitudinal (k-t) 3126 3334
P—axial (kip) −282 −220

5000

4000
Nominal axial load, P (K)

3000

2000

1000

–1000
0 1000 2000 3000 4000 5000
Nominal moment, Mn (K-ft)

FIGURE 2.15 Example 2.2—interaction diagram.

Following the procedure outlined in Section 2.4.2, the moment magniication factors for
each load group can be calculated, and the results are summarized in Table 2.5.
In which

K y = K x = 2.00

K y × L/R = K x × L/R = 2.00 × 27.0 / ( 0.998 ) = 54.11

I 12.46
r = radio of gyration = = = 0.998 ft
A 12.51
TABLE 2.5 Moment Magniication and Buckling Calculations

54
Moment Magniication Approximate EI for Cracked Section Critical Buckling Axial
Limit
State Case Trans Magy Long Magx Comb Mag EIy (K-t2) EIx (K-t2) Pcy (kip) Pcx (kip) Load Pu (kip)
Str-I 1 1.557 1.578 1.541 2,113,465 2,213,999 7,153 7,494 1,919
Str-I 2 1.549 1.412 1.431 2,036,337 2,476,065 6,892 8,381 1,833
Str-I 3 1.765 1.684 1.728 2,060,541 2,199,323 6,974 7,444 2,267
Str-I la 1.373 1.353 1.365 2,210,094 2,309,333 7,507 7,816 1,531
Str-I 2a 1.361 1.391 1.299 2,145,100 2,528,371 7,260 8,558 1,445

Bridge Engineering Handbook, Second Edition: Substructure Design

Str-I 3a 1.518 1.476 1.498 2,168,298 2,296,280 7,339 7,772 1,879
Str-II 1 1.483 1.683 1.513 1,675,434 1,344,031 5,671 4,549 1,385
Str-II 2 1.483 1.683 1.513 1,675,434 1,344,031 5,671 4,549 1,365
Str-II 3 1.483 1.683 1.513 1,675,434 1,344,031 5,671 4,549 1,365
Str-II 1a 1.286 1.413 1.301 1,764,039 1,344,031 5,971 4,549 997
Str-II 2a 1.286 1.413 1.301 1,764,039 1,344,031 5,971 4,549 997
Str-II 3a 1.286 1.413 1.301 1,764,039 1,344,031 5,971 4,549 997
Str-III 1 1.352 1.425 1.362 2,186,828 1,909,584 7,402 6,463 1,447
Str-III 2 1.483 1.683 1.513 1,675,434 1,344,031 5,671 4,549 1,365
Str-III 3 1.352 1.425 1.362 2,186,828 1,909,584 7,402 6,463 1,447
Str-III la 1.223 1.260 1.228 2,285,095 2,019,060 7,734 6,834 1,059
Str-III 2a 1.286 1.413 1.301 1,764,039 1,344,031 5,971 4,549 997
Str-III 3a 1.223 1.260 1.228 2,285,095 2,019,060 7,734 6,834 1,059
Stx-IV 1 1.670 1.950 1.718 1,632,036 1,344,031 5,524 4,549 1,662
Stx-IV 2 1.670 1.950 1.718 1,632,036 1,344,031 5,524 4,549 1,662
Str-IV 3 1.670 1.950 1.718 1,632,036 1,344,031 5,524 4,549 1,662
Str-IV la 1.286 1.413 1.301 1,764,039 1,344,031 5,971 4,549 997
Str-IV 2a 1.286 1.413 1.301 1,764,039 1,344,031 5,971 4,549 997
Str-IV 3a 1.286 1.413 1.301 1,764,039 1,344,031 5,971 4,549 997
Str-V 1 1.487 1.482 1.485 2,203,359 2217,409 7,458 7,505 1,831
Str-V 2 1.526 1.391 1.414 1,978,537 2,425,521 6,697 8,210 1,731
Str-V 3 1.614 1.600 1.609 2,175,213 2,206,330 7,362 7,463 2,100
 Piers and Columns
Str-V la 1.328 1.326 1.328 2,299,878 2,312,354 7,784 7,826 1,444
Str-V 2a 1.339 1.270 1.280 2,088,594 2,488,095 7,069 3,421 1,343
Str-V 3a 1.422 1.414 1.419 2,274,640 2,302,523 7,699 7,793 1,712
Ext-I 1 1.000 1.000 1.000 2,268,943 2,256,198 7,680 7,636 1,052
Ext-I 2 1.000 1.000 1.000 2,149,076 2,274,357 7,274 7,700 1,064
Ext-I 3 1.000 1.000 1.000 2,149,076 2,274,857 7,274 7,700 1,064
Ext-I la 1.000 1.000 1.000 2,268,943 2,256,198 7,630 7,636 1,052
Ext-I 2a 1.000 1.000 1.000 2,149,076 2,274,857 7,274 7,700 1,064
Ext-I 3a 1.000 1.000 1.000 2,149,076 2,274,857 7,274 7,700 1,064
EKI-II 1 1.412 1.360 1.389 1,700,792 1,376,157 5,757 6,350 1,261
Ext-II 2 1.428 1.280 1.308 1,623,432 2,225,734 5,495 7,533 1,236
Ext-II 3 1.482 1.404 1.445 1,646,910 1,360,966 5,574 6,299 1,360
Ext-II la 1.412 1.360 1.389 1,700,792 1,676,157 5,757 6,350 1,261
Ext-II 2a 1.428 1.280 1.308 1,623,482 2,225,734 5,495 7,533 1,236
Ext-II 3a 1.482 1.404 1.445 1,646,910 1,860,966 5,574 6,299 1,360
Note: Axial load, Pu, is calculated at column top for buckling calculations.

55
56 Bridge Engineering Handbook, Second Edition: Substructure Design

K y luy Kx lux ( 2.0 )( 27.0 )

= = = 54.11
ry rx (0.998)

K y luy K x lux
22 < = = 54.11 < 100 ∴ Second − order effect should be considered .
ry rx

he calculations for limit state, strength I, and case 2 (maximum longitudinal moment) are
demonstrated in the following:
Bending in the longitudinal direction: Mux

Factored load Mux = 1.0[1.25DC + 1.5DW + 1.75( LL + IM + BF + CF ) + 1.2 TU + W + WL ]

βd in Equation 2.8 = max factored dead load moment, MDL/max factored total moment, Mux

M DL = 1.25 × 148 = 185 k-ft(252KN-m)

M ux = 1.25 × 148 + 1.75 (545 + 289 + 295) = 2,160 k-ft(2,929KN-m)

βd = 185 / 2160 = 0.09

Ec I g 3,640 × 144 × 12.46

+ E s Is + 29,000 × 144 × 0.3338
EI x = 5 = 5 = 2,476,065 K/ft 2
1 + βd 1 + 0.09

π 2 × EI x 3.14162 × 2,476,065
Pc = = = 8,381 k   (37,292 KN)
( KL )2 (2.0 × 27)2

Cm = 1.0 for frame not braced against sidesway

1 1
δs = = = 1.412
1,833
1−
∑ Pu 1−
0.75 × 8381
ϕ∑ Pc

he magniied factored moment δs × Mux = 1.412 × 2,160 = 3,050 k-t (4,135 KN-m)
Go through the same procedure, the magniied factored moment in transverse direction =
1,331 k-t. he combined moment is Mu = 3,328 k-t. he nominal moment capacity of the sec-
tion corresponding to the axial force of 1,833 Kip is 4,357 k-t. he factored moment capacity
ϕMn = Mr = 3,428 k-t (ϕ = 0.789).
M r 3,328
herefore, = ≈ 1.0 ∴ Design is OK.
M u 3,428
he analysis results with the comparison of applied moments to capacities are summarized
in Table 2.6.
Column lateral reinforcement is calculated for two cases: (1) for applied shear and (2) for
coninement. Typically, the coninement requirement governs. Apply Equation 2.32 or Equation
2.33 to calculate the coninement reinforcement. For seismic analysis, the unreduced seismic
shear forces shall be compared with the shear forces due to plastic hinging of columns. he
smaller shall be used. he plastic hinging analysis procedure is discussed elsewhere in this
handbook and will not be repeated here.
First, determine the lateral reinforcement by coninement.
Piers and Columns 57

TABLE 2.6 Comparison of Factored Loads to Factored Capacity of the Column

Applied LRFD Factored Capacity (PhMn) Maximum
Limit Tran Long Comb Axial Rebar
State Case Muy Mux Mu Pu = Pr Mr Phi Strain Ratio Mr/Mu
Str-I 1 1575 1312 2050 1919 3412 0.780 0.00262 1.66 OK
Str-I 2 1331 3050 3328 1833 3438 0.789 0.00279 1.03 OK
Str-I 3 1594 1402 2122 2267 3270 0.753 0.00206 1.54 OK
Str-I la 1283 1099 1650 1531 3480 0.821 0.00342 2.06 OK
Str-I 2a 1065 2722 2923 1445 3484 0.831 0.00363 1.19 OK
Str-I 3a 1254 1152 1703 1373 3422 0.784 0.00269 2.01 OK
Str-II 1 675 311 743 1385 3482 0.838 0.00377 4.69
Str-II 2 675 311 743 1385 3482 0.838 0.00377 4.69
Str-II 3 675 311 743 1385 3482 0.838 0.00377 4.69
Str-II 1a 486 188 521 997 3446 0.893 0.00487 6.61
Str-II 2a 466 188 521 997 3446 0.893 0.00487 6.61
Str-II 3a 486 188 521 997 3446 0.893 0.00487 6.61
Str-III 1 1623 647 1747 1447 3484 0.831 0.00363 1.99 OK
Str--III 2 675 311 743 1385 3482 0.838 0.00377 4.69 OK
Str-III 3 1623 647 1747 1447 3484 0.831 0.00363 1.99 OK
Str-III 1a 1374 507 1464 1059 3453 0.883 0.00466 2.36 OK
Str-III 2a 486 188 521 997 3446 0.893 0.00487 6.61 OK
Str-III 3a 1374 507 1464 1059 3453 0.883 0.00466 2.36 OK
Str-IV 1 852 433 955 1662 3464 0.805 0.00311 3.63 OK
Str-IV 2 852 433 955 1662 3464 0.805 0.00311 3.63 OK
Str-IV 3 852 433 955 1662 3464 0.805 0.00311 3.63 OK
Str-IV la 486 188 521 997 3446 0.893 0.00487 6.61 OK
Str-IV 2a 486 188 521 997 3446 0.893 0.00487 6.61 OK
Str-IV 3a 486 188 521 997 3446 0.893 0.00487 6.61 OK
Str-V 1 1859 1292 2264 1831 3438 0.789 0.00279 1.52 OK
Str-V 2 1170 2377 2650 1731 3455 0.799 0.00298 1.30 OK
Str-V 3 1882 1356 2320 2100 3350 0.765 0.00231 1.44 OK
Str-V la 1558 1087 1900 1444 3481 0.831 0.00362 1.83 OK
Str-V 2a 924 2105 2299 1343 3482 0.844 0.00388 1.51 OK
Str-V 3a 1549 1125 1914 1712 3459 0.801 0.00302 1.81 OK
Ext-I 1 1191 773 1420 1052 3818 1.000 0.00502 2.69 OK
Ext-I 2 877 815 1197 1064 3824 1.000 0.00499 3.19 OK
Ext-1 3 877 815 1197 1064 3824 1.000 0.00499 3.19 OK
Ext-I la 1191 773 1420 1052 3818 1.000 0.00502 2.69 OK
Ext-I 2a 877 815 1197 1064 3824 1.000 0.00499 3.19 OK
Ext-I 3a 877 815 1197 1064 3824 1.000 0.00499 3.19 OK
Ext-II 1 535 465 709 1261 3478 0.854 0.00409 4.90 OK
Ext-II 2 479 912 1030 1236 3472 0.857 0.00414 3.37 OK
Ext-II 3 516 468 696 1360 3482 0.841 0.00383 5.00 OK
Ext-II la 535 465 709 1261 3478 0.854 0.00409 4.90 OK
Ext-II 2a 479 912 1030 1236 3472 0.857 0.00414 3.37 OK
Ext-II 3a 516 468 696 1360 3482 0.841 0.00.83 5.00 OK

Note: Permit load was not input; hence, calculation for Str-II limit state is incomplete.
58 Bridge Engineering Handbook, Second Edition: Substructure Design

Inside the plastic hinge zones

fc′ 4.0
ρs = 0.12 = 0.12 = 0.008
fy 60.0
Outside the plastic hinge zones

 Ag  fc′  12.56  4.0

ρs = 0.45  −1 = 0.45  −1 = 0.0057
 Ac  f yh  10.56  60.0

4 Ab
Reinforcement for confinement = ρs =
D ′s
If use #5 bar,
Inside plastic hinge zones: s = 3.4 in say s = 3 in
Outside plastic hinge zones: s = 4.8 in say s = 4.5 in
hen, check the lateral reinforcement for shear.
For let column:
Vu = 167 kip (743 KN) (shear due to plastic hinging governs)
Inside the plastic hinge zones (assume displacement ductility µd = 5):

0.008 × 60
Factor1 = + 3.67 − 5 = 1.87
0.15

(1108 − 508) × 1000

Factor 2 = 1 + = 1.17
2000 × 12.56 × 144

vc = 1.87 × 1.17 × 4000 = 138.4 psi ( 0.95 Mpa )

Vc = 10.56 × 144 × 138.4 = 210.4 kip

φVc = 0.85 × 210.4 = 179 > Vu = 167 kip

∴ No lateral reinforcement is required for shear.

For right column:
he shear force due to plastic hinging is Vu = 199 kip.
However, the axial force in right column is larger than that of the let column, the shear
capacity will be larger. By observation, the shear will not govern.
Final design:
4 t (1.22 m) diameter of column with 32-#9 (32-#30) for main reinforcement and #5@3”
(#16 @ 76.2 mm) for spiral coninement in the top 6 t of column and #[email protected]” (#16 @114.2 mm)
in the rest of the column.

2.4.4 Steel and Composite Columns

Steel columns are not as commonly used as concrete columns. Nevertheless, they are viable solu-
tions for some special occasions, for example, in space-restricted area. Steel pipes or tubes illed
with concrete known as composite columns (Figure 2.16) ofer the most eicient use of the two basic
materials. Steel at the perimeter of the cross section provides stifness and triaxial coninement, and
Piers and Columns 59

FIGURE 2.16 Typical cross sections of composite columns.

the concrete core resists compression and prohibits local elastic buckling of the steel encasement.
he toughness and ductility of composite columns makes them the preferred column type for
earthquake-resistant structures in Japan. In China, the composite columns were irst used in Beijing
subway stations as early as 1963. Over the years, the composite columns have been used extensively
in building structures and bridges (Cai 1987 and 1992; Zhong 1996). he design speciications of
steel and composite columns are given in various codes. (CECS 1990; Galambos and Chapuis 1990;
AISC 2010). In this section, the design provisions of AASHTO LRFD (2012) for steel and composite
columns are summarized as follows.

2.4.4.1 Compressive Resistance

For prismatic members with at least one plane of symmetry and subjected to either axial compression or
combined axial compression and lexure about an axis of symmetry, the factored resistance of compo-
nents in compression, Pr, shall be calculated as

Pr = φc Pn (2.36)

where
Pn = nominal compressive resistance
ϕc = resistance factor for compression = 0.90
he nominal compressive resistance of a steel or composite column shall be determined as

 0.66λ Fe As if λ ≤ 2.25

Pn = Pn =  0.88 Fe As (2.37)
 if λ >  2.25
 λ
in which
For steel columns

2
 KL  Fy
λ= π (2.38)
 rs  Ee
For composite column

2
 KL  Fe
λ= π (2.39)
 rs  Ee
60 Bridge Engineering Handbook, Second Edition: Substructure Design

A  A 
Fe = Fy + C1Fyr  r  + C2 fc  c  (2.40)
 As   As 

 C  A 
Ec = E 1 +  3   c   (2.41)
  n   As  

where As is the cross-sectional area of the steel section (in2), Ac the cross-sectional area of the concrete
(in2), Ar the total cross-sectional area of the longitudinal reinforcement (in2), Fy the speciied minimum
yield strength of steel section (ksi), Fyr the speciied minimum yield strength of the longitudinal rein-
forcement (ksi), fc′ the speciied minimum 28-day compressive strength of the concrete (ksi), E is the
modules of elasticity of the steel (ksi), L the unbraced length of the column (in), K the efective length
factor, n the modular ratio of the steel to concrete, rs the radius of gyration of the steel section in the
plane of bending, but not less than 0.3 times the width of the composite member in the plane of bend-
ing for composite columns (in), and for concrete illed in steel tube, C1 = 1.0, C2 = 0.85, and C3 = 0.40.
In order to use the above equation, the following limiting width/thickness ratios for axial compres-
sion of steel members of any shape shall be satisied:

b E
≤k (2.42)
t Fy
where k is the plate buckling coeicient as speciied in Table 2.7, b the width of plate (in) as speciied in
Table 2.7, and t the plate thickness (in).
Wall thickness of steel or composite tubes shall satisfy the following:
For circular tubes

D E
≤ 2.8 (2.43)
t Fy

TABLE 2.7 Limiting Width–hickness Ratios

k b

Plates supported along one edge

Flanges and projecting leg or plates 0.56 Half-lange width of I-section
Full-lange width of channels
Distance between free edge and irst line of bolts or welds in plates
Full-width of an outstanding leg for pairs of angles on continuous
contact
Stems of rolled tees 0.75 Full-depth of tee
Other projecting elements 0.45 Full-width of outstanding leg for single angle strut or double angle
strut with separator
Full projecting width for others

Plates supported along two edges

Box langes and cover plates 1.40 Clear distance between webs minus inside corner radius on each
side for box langes
Distance between lines of welds or bolts for lange cover plates
Webs and other plates elements 1.49 Clear distance between langes minus illet radii for webs of rolled
beams
Clear distance between edge supports for all others
Perforated cover plates 1.86 Clear distance between edge supports
Piers and Columns 61

For rectangular tubes

b E
≤ 1.7 (2.44)
t Fy

where D is the diameter of tube (in), b the width of face (in), and t the thickness of tube (in).

2.4.4.2 Flexural Resistance

he factored lexural resistance, Mr, shall be determined as

Mr = φ f M n (2.45)

where
Mn = nominal lexural resistance
ϕf = resistance factor for lexure, ϕf = 1.0
he nominal lexural resistance of concrete-illed steel tubes that satisfy the limitation

D E
≤ 2.8 (2.46)
t Fy

may be determined

D E
If < 2.0 , then M n = M ps (2.47)
t Fy

E D E
If 2.0 < ≤ 8.8 , M n = M yc (2.48)
Fy t Fy

where
Mps = plastic moment of the steel section
Myc = yield moment of the composite section

2.4.4.3 Combined Axial Compression and Flexure

he axial compressive load, Pu, and concurrent moments, Mux and Muy, calculated for the factored load-
ings for both steel and composite columns shall satisfy the following relationship:

Pu Pu M M uy 
If < 0.2, then +  ux +  ≤ 1.0 (2.49)
Pr 2.0 Pr  M rx M ry 

Pu P 8.0  M ux M uy 
If ≥ 0.2, then   u +  +  ≤ 1.0 (2.50)
Pr Pr 9.0  M rx M ry 

where Pr is factored compressive resistance (kip); Mrx, Mry are factored lexural resistances about x and
y axis, respectively (kip-t); Mux, Muy factored lexural moments about the x and y axis, respectively
(kip-t).
62 Bridge Engineering Handbook, Second Edition: Substructure Design

References
AASHTO. 2012. AASHTO LRFD Bridge Design Speciications, Customary U.S. Unit, 2012, American
Association of State Highway and Transportation Oicials, Washington, DC.
ACI. 2011. 318–11: Building Code Requirements for Structural Concrete and Commentary, (ACI 318–11),
American Concrete Institute, Farmington Hills, MI.
AISC. 2010. Speciication for Structural Steel Buildings, ANSI/AISC 360–10, American Institute of Steel
Construction, Chicago, IL.
Cai, S. H. 1987. “Ultimate Strength of Concrete-Filled Tube Columns,” in Composite construction in Steel
and Concrete, Proc. of an Engineering Foundation Conference, Dale Buckner, C., and Viest, Ivan M.,
Eds, Henniker, NH. 703.
Cai, S. H. 1992. “Chinese Standard for Concrete-Filled Tube Columns,” in Composite Construction in Steel
and Concrete II, Proc. of an Engineering Foundation Conference, Samuel Easterling, W., and Kim
Roddis, W. M., Eds, Potosi, MO. 143.
Caltrans. 1994. Bridge Memo to Designers (7–10) - Bridge Deck Joints and Deck Joint Seals, California
Department of Transportation, Sacramento, CA.
Caltrans. 2013. Seismic Design Criteria, Version 1.6, California Department of Transportation,
Sacramento, CA.
CECS 28:90. 1990. Speciications for the Design and Construction of Concrete-Filled Steel Tubular Structures,
China Planning Press, Beijing, China. (in Chinese)
Galambos, T. V. and Chapuis, J. 1990. LRFD Criteria for Composite Columns and Beam Columns, Revised
Drat, Washington University, Department of Civil Engineering, St. Louis, MO. December.
Galambos, T. V. 1998. Guide to Stability Design for Metal Structures, 5th Ed., the Structural Stability
Research Council, John Wiley & Sons, New York, NY.
Paulay, T. and Priestley, M. J. N. 1992. Seismic Design of Reinforced Concrete and Masonry Building, John
Wiley & Sons, New York, NY.
White, D. W. and Hajjar, J. F. 1994. “Application of Second-Order Elastic Analysis in LRFD: Research to
Practice,” Engineering Journal, American Institute of Steel Construction, 28(4), 133–148.
Zhong, S. T. 1996. “New Concept and Development of Research on Concrete-Filled Steel Tube (CFST)
Members”, Proc. of he 2nd Int’l Symp. On Civil Infrastructure Systems, December 9-12, 1996, Hong
Kong, China
 3
Towers
3.1 Introduction ........................................................................................63
3.2 Functions..............................................................................................68
3.3 Esthetics ...............................................................................................68
3.4 Towers and Spectacular Bridges .......................................................69
3.5 Conceptual Design ............................................................................. 74
Materials • Forms and Shapes • Erection
Charles Seim 3.6 Final Design .........................................................................................83
Consulting Engineer Design Loads • Other Design Considerations
Jason Fan 3.7 Construction .......................................................................................85
California Department of 3.8 Summary ..............................................................................................87
Transportation References........................................................................................................87

3.1 Introduction
Towers are the most visible structural elements of long-span bridges, because they project above the
superstructure and can be seen from all directions by both viewers and bridge users. Towers give to a
bridge a characteristic identity, a unifying theme, a motif from which people can identify that particular
bridge. Towers project a mnemonic bridge image that people can recall as their lasting impression of
that bridge itself, making towers an important part of the overall esthetics.
As examples of the powerful imagery of towers, contrast the elegant art deco towers of the 1937
Golden Gate Bridge (Figure 3.1) with the utilitarian, but timeless, architecture of the towers of the 1936
San Francisco–Oakland Bay Bridge (Figure 3.2).
hen compare these robust towers to those of the 1964 delicate towers of the Firth of Forth
Suspension Bridge (Figure 3.3); ponder the disproportions between the massive, rugged stone towers
of the 1883 Brooklyn Bridge (Figure 3.4) with the awkward and confusing steel towers of the 1903
Williamsburg Bridge in New York (Figure 3.5).
Alternatively, one may contrast those older, Iconic Bridges, with the new and distinctive San
Francisco–Oakland Bay Bridge East Span with its single-tower suspension bridge (Figure 3.19d, later in
the chapter) and with the quasi-diamond-shaped towers of the 2000 Yeongjong Grand Bridge, Incheon,
South Korea (Figure 3.6). Both of these are self-anchored suspension bridges and have no heavy and
bulky concrete anchorages visible at each end.
hen compare the concrete quasi-diamond-shaped towers of the 1995 Glebe Island Bridge (Figure 3.7)
to the concrete full-diamond-shaped towers of the 2005 Cooper River Bridge (Figure 3.8); the heights of
the roadways dictated the diferences between these tower shapes and not the whims of the designers!
One can easily see that there is great diversity in bridge tower designs; the only requirement that these
towers have in common is that they must resist the loads and forces of nature and be in equilibrium
according to the three equations of statics. Towers surely do impact the appearance of bridges, for good
or for bad.

63
64 Bridge Engineering Handbook, Second Edition: Substructure Design

FIGURE 3.1 Golden Gate Bridge, San Francisco. (Courtesy of Charles Seim.)

FIGURE 3.2 San Francisco–Oakland Bay Bridge. (Courtesy of Charles Seim.)

Towers 65

FIGURE 3.3 Firth of Forth Suspension Bridge. (Courtesy of Charles Seim.)

FIGURE 3.4 Brooklyn Bridge, New York. (Courtesy of Charles Seim.)

66 Bridge Engineering Handbook, Second Edition: Substructure Design

FIGURE 3.5 Williamsburg Bridge, New York. (Courtesy of Charles Seim.)

FIGURE 3.6 Yeongjong Grand Bridge, Incheon, South Korea.

Towers 67

FIGURE 3.7 Glebe Island Bridge, Sydney, Australia. (Courtesy of T. Y. Lin International.)

FIGURE 3.8 Cooper River Bridge, Charleston, South Carolina, under construction. (Courtesy of Charles Seim.)
68 Bridge Engineering Handbook, Second Edition: Substructure Design

he famous bridges noted above are all older than three-quarters of a century. If they are well
maintained, all these bridges could continue to serve for another 100 years.
he service lives of the new self-anchored suspension span of the San Francisco–Oakland Bay Bridge
and the Yeongjong Grand Bridge could be 150 years. hese bridges are excellent examples of enduring
structures; they serve as a reminder to bridge engineers that well-designed and well-maintained structures
can last for 100–150 years, or perhaps longer. Robust designs, durable materials, provisions for inspection
and maintenance access, and a well-executed maintenance program will help to ensure long service lives.
Both suspension and cable-stayed bridges are supported by abutments or piers at the point at
which these structures transition to an approach roadway or to an approach structure. Abutments are
discussed in Chapter 6. Piers and columns that support the superstructure for other types of bridge
structures, such as girders, trusses, or arches, usually do not project above the deck. Piers and columns
are discussed in Chapter 2.

3.2 Functions
“Towers” are usually deined as the vertical steel or concrete structures that project above the bridge
deck to support both themselves and the bridge cables and function to carry the loads and the forces to
which the bridge is subjected to the ground.
hus, by this deinition, towers are used only for suspension bridges, cable-stayed bridges, or hybrid
suspension-cable-stayed structures. he word “pylon” is sometimes used to designate the single-shat
tower of a cable-stayed bridge. In this chapter, the word “tower” is used for structures that are
self-supporting; “pylons” is not used, to avoid confusion.
Recently a new term “spar” has been introduced to describe vertical or near-vertical members that
are not self-supporting and must depend on cables for its support; however, the spar does function as a
tower carrying some bridge loads and forces to the ground. In this chapter, the word “spar” is used to
describe a member that cannot support itself but functions as a tower.
Towers must perform its functions economically, be esthetically pleasing and constructible. Towers
must also be reliable and serviceable for the entire life of the bridge, as unlike other bridge components,
towers cannot be replaced without tearing down the bridge.
Structural serviceability is an important component of good bridge design. his requires that the
bridge and towers be designed to allow for ease of carrying out both inspection and maintenance func-
tions to provide continuous good service to its users. he public demands that bridges and towers be
attractive, esthetic statements having long service lives, so as not to be wasteful of public funds.

3.3 Esthetics
Although the main function of the towers is structural, an important secondary function is visual—
beyond mere esthetics, the towers reveal the true character, or motif, of a bridge. he bridges used as
examples in Section 3.1 are good illustrations of the image of the structure, as revealed by the towers.
Indeed, most are famous because of their towers!
Many people visualize the character of the Brooklyn Bridge by its gothic, arched-masonry towers, the
Golden Gate Bridge by its tall, tapered, red steel towers, and across the Bay, the San Francisco–Oakland
Bay Bridge by its robust-looking cross bracing and shiny aluminum paint. he elegant white, single
tower of the new San Francisco–Oakland Bay Bridge East Span self-anchored suspension bridge will
perhaps leave an even more distinctive impression ater the bridge is opened in 2013.
Seim (1996) measured the aspect ratios of the length, divided by the thickness of the visible compo-
nents of the towers of both the Golden Gate and the San Francisco–Oakland Bay Bridges. He found
important, but subtle, reduction of these ratios with increasing heights above the tower base; the higher
the member, the smaller the aspect ratio. It is these subtle changes in the ratios within the heights of the
towers that produce the much-admired proportions of these world-renowned bridges. he towers for a
Towers 69

long span bridge should be carefully shaped and proportioned so as to give that entire bridge a strong
and sturdy, but graceful, soaring visual image to the eyes of the viewing public.
he two main cable suspension bridges drape in a parabolic curve between towers that many people
instinctively enjoy viewing. he large diameter of the cables makes them stand out as the important
contributors to the overall visual impression of the supporting elements of the roadway. he towers of
these common types of suspension bridges are as wide as the bridge and extend full height, making
them the visual supporting elements, and they project the motif of the bridge design. Just a few suspen-
sion bridges employ a single cable, in which case the towers are usually tapered.
he cables of most cable-stayed bridges are small in diameter and usually do not visually stand out as
do the large cables of a suspension bridge. he cables can be arrayed in a single plane along the centerline
of the bridge, a double plane at the sides of the roadway girder, or a single plane on one side of the tower
and a double plane on the other side. A single plane array is usually used with a single-shat tower and a
double plane array usually used with a two-shat tower. See Chapter 10, Bridge Engineering Handbook,
Second Edition: Superstructure Design, Cable-Stayed Bridges.
However, arrays of the cable stays, such as a fan, radiating fan, or the little-used harp, should be con-
sidered in the context of the form of the tower. he parallel cables of a harp array, for example, usually
will not be as obtrusive to the bridge towers as are other cable arrangements, such as a radiating fan
array that dominates visually over the tower. hus, the cables and the towers together should be consid-
ered as both visual systems and structural systems.
Billington (1983) presents an overview of the importance of the role of esthetics in the history of
the development of modern bridge design. Prof. Billington coined the words “Structural Art” to honor
bridges that are eicient, economical, and elegant structures. Leonhardt (1984) presents many exam-
ples of completed bridges with many tower shapes and cable arrangements for both suspension and
for cable-stayed bridges. Esthetics of bridges is discussed in more detail in Chapters 2 and 3 of Bridge
Engineering Handbook, Second Edition: Fundamentals.

3.4 Towers and Spectacular Bridges

Although eiciency, economy, and elegance are usually the major elements in bridge design, occasion-
ally, since the 1990s, eiciency and economy have oten not been the prime objectives of bridge designers.
his trend started as bridge owners, the public, or both, began demanding “spectacular,” “picturesque,”
or “distinctive” bridges, because bridge engineers could design and construct them!
Such a trend oten calls for coniguring the stay cables in unusual arrays that may dominate the
towers, thus allowing the stay cables to become the principle esthetic statements of these bridges. his
trend also featured curved, inclined, or kinked towers to add “visual impact” to bridges.
hese new spectacular bridge types are designed to attract attention, because eiciency is not an
objective and cost is not restricted. One could also argue that although they may be spectacular, these
bridge types are not elegant. Regardless, such bridges are not considered “Structural art,” as deined by
Billington (1983), because they do not conduct the forces of the bridge to its foundation in the most ei-
cient manner, and they are not economical, because they cost more than a conventional bridge. Instead,
such bridges may be considered “extravagant structural art,” and a form of art, nonetheless.
his extravaganza started in the early 1990s, when proven structural engineering programs became
accessible to most engineers, and high-performance steel and concrete were readily available; thus, it
was inevitable that engineers and architects would begin to exploit these relatively new developments by
designing and constructing spectacular bridges that featured distinctive towers.
One of the irst of these “spectacular” bridges was the Alamillo Bridge (Figure 3.9), constructed for
the 1992 Expo in Seville, Spain. It was designed by Santiago Calatrava, who acted as both architect and
engineer. he bridge features a 142-m (466-t) tall, concrete-illed, steel box tower, angled at 68 degrees;
painted white, it is a visible landmark from the old town of Seville. he concrete box girder roadway is a
200-m (656-t) single span and anchors the single plane, harp-arrayed cables.
70 Bridge Engineering Handbook, Second Edition: Substructure Design

FIGURE 3.9 Alamillo Bridge, Seville, Spain.

FIGURE 3.10 Erasmus Bridge in Rotterdam, the Netherlands. (Courtesy of Charles Seim.)

he very tall tower and the parallel cables create a beautiful, dramatic structure that immediately
attracts the attention of people viewing the bridge. However, this structure is not a genuine cable-stayed
bridge, because the tower is not anchored to the ground with backstay cables (Petroski 1996). he traic
crossing the bridge delects the girder and loads the cables, but the cable loads at the tower are not in
horizontal equilibrium, and the tower simply tilts a little.
he bridge was very costly. However, the people who view the structure see it as a very attractive
bridge and consider it to be well worth the cost. his motif has been successfully used several times
since, most notably on Sun Dial Bridge in California, where the single, pointed tower casts a shadow that
tells the time of day. More importantly, the Alamillo Bridge cleared the way for engineers and architects
to design and construct outstanding bridges, whenever cost is not an important factor to the bridge
owners or to the cities desiring a spectacularly designed bridge as a city icon.
he Erasmus Bridge of Rotterdam, the Netherlands (Figure 3.10), opened in 1996, is another example
of a spectacular bridge that is admired by all who view it. his bridge, designed by architect Ben van
Berkel, features a tapered-steel tower with a “kink” near the midpoint that instantly attracts attention,
Towers 71

because a kink in a tower is highly unusual! Towers are not usually kinked, because they are compres-
sion members; a kink in a compression member introduces a large bending moment, which requires the
engineer to add extra steel to resist that moment, substantially increasing the tower cost.
he modiied, fan-arrayed stay cables in the main span load the upper portion of the tower in com-
pression and bending; they also produce a reaction force at the top of the tower that is resisted by the
two backstay cables that are attached at the top of the tower. he vertical component from the backstay
cables adds large compression forces in the tower. hus, the tower carries the bending moments and the
compression from the main span stay cables, compression forces from the backstays, and the bending
moments from the kink. he sum total of these cable arrangements and the kink added considerable
costs to reinforce the tower to resist the huge bending and compression loads. However, this bridge is a
great success because Rotterdam now has a city icon, and the people can marvel at the bridge’s unique
architecture!
In 2009, the city of Dublin opened the Samuel Beckett Bridge, named ater the famous Irish writer,
and designed by Calavatra. his bridge is certainly a picturesque structure, having a thin, curved tower
described as a forward-leaning, tubular curved spar (Figure 3.11).
he Samuel Beckett Bridge is a short, 120-m long cable-stayed bridge that is balanced as a swing
bridge that pivots on a pier located directly under the base of its 48-m high spar. Each of the two back-
stay cables connects both to the tip of the spar and to the two backstay edge-girders, forming a “V.” he
backstay cables and the forward stay cables combine to create a self-anchored structure that allows the
structure to swing open to provide ship passage. he curved spar acts as a tilted-up arch as it is loaded
transversely with the forward stay cables that support the main span.
his very picturesque bridge is an ingenious assemblage of girders, cables, and a curved spar. Although
costly, it is a true bridge, compared to the Alamillo Bridge, and was supposedly designed to mimic an
Irish Harp laid on its side.
China has built many distinctive bridges, and the Nanjing hird Yangtze Bridge is a good example
of this type (Figure 3.12). his cable-stayed bridge was the longest of this type in China, when it was
opened in 2005 with a central span of 628 m and 215 m tall steel towers. he city fathers wanted each of
the two towers of the bridge to look similar to the curved Eifel Tower in Paris, because one of them had
visited Paris and was impressed with the beauty of Eifel’s masterpiece.
he upper portions of the two steel shats of each tower are straight and braced by three cross-struts;
the lower portions of the two shats do not have cross-struts but are curved to simulate the curvature of

FIGURE 3.11 Samuel Beckett Bridge by Calavatra. (Courtesy of Charles Seim.)

72 Bridge Engineering Handbook, Second Edition: Substructure Design

FIGURE 3.12 hird Nanjing Yangtze River Bridge Towers under construction. (Courtesy of Charles Seim.)

the Eifel Tower. he curvature produced extremely large bending moment in each of the curved lower
portions of the shat, and that required additional steel to reduce the stress from the large bending
moment to an acceptable value. he Eifel Tower has cross-struts spaced along the tower height to reduce
the bending moments in the four corner shats of that Paris icon.
Each tower shat was fabricated in segment approximately 10–12 m high; thus each segment in the
tower shats was fabricated to diferent dimensions and angles. Every segment required geometric
control, which required very accurate ield surveying to ensure that each segment was accurately placed
on the proper curvature. he contractor believed that the dimensions and angles in each segments could
not be controlled accurately enough to use welded connections between the segments and therefore
used bolted connections. hese bolted connections required very thick splice plates and a large number
of high-strength bolts to carry both bending and compression forces. All these items added cost to the
construction of the towers.
In addition, the caisson concrete cap required a massive amount of prestressing steel to contain the
outwardly directed thrust distributed to the caisson cap from each inclined shat at the base of the tower.
The Eiffel Tower emulation added cost to the bridge and tower construction. However, it is a very
successful bridge, because the curved shafts add a dynamic effect to what otherwise could be dull-
looking towers. The City Fathers are delighted, and the bridge users admire the curved appearance
of the towers.
Another example of a distinctive bridge is the Sanhao Bridge in Shenyang City, China (Figure 3.13),
designed by Man-Chung Tang, who also acted as the architect of this bridge. he bridge features two
concrete struts springing from a common support and inclined away from each other and each support-
ing a curved concrete arch spanning across the bridge roadway.
From each inclined tower, cable stays, arrayed in a harp arrangement, support the 100-m roadway
on each side of the piers supporting the towers. Horizontal cables, parallel arrayed, tie the twin towers
together.
Towers 73

FIGURE 3.13 Sanhao Bridge, Shenyang, China. (Courtesy of Man-Chung Tang.)

FIGURE 3.14 Jiayue Bridge, in Chongqing City, China. (Courtesy of Man-Chung Tang.)

he towers and cables added a small cost to this very distinctive bridge. he bridge is a success,
because Shenyang City now has a distinctive icon, and the people who use and view the bridge from city
streets are delighted.
Another distinctive bridge by Man-Chung Tang is the Jiayue Bridge, in Chongqing City, China
(Figure 3.14), which is a conventional cable-stayed bridge with unconventional towers projecting
33 m above the roadway and with a total height of 126 m. The main span is 250 m, but the attrac-
tion of the bridge is not the main span but the portion of the towers that project above the roadway,
acting as out-stretching arms holding up the cable stays. The arms leaning outwardly open up
the bridge to the horizon for drivers compared to the conventional tower types that lean inward,
enclosing the bridge.
74 Bridge Engineering Handbook, Second Edition: Substructure Design

he outward-leaning arms added little cost to the bridge, but they do create a distinctive bridge for all
to enjoy, which is the principle feature of a successful bridge.
From these few examples, one can see that these types of bridges can range from the truly spectacular
to picturesque, to distinctive bridges; all can be considered art, artistic, or even elegant structures; how-
ever, they cannot be considered “Structural Art” according to Prof. Billington’s deinition of eiciency,
economy, and elegance.
hese “bridge” types will continue to be constructed wherever people desire bridge extravaganzas
and have the money to back up such desires. hus, such bridge types as these have entered the repertoire
of the bridges that bridge engineers are required to design, construct, and maintain.
Any future discussion of towers for these spectacular bridges is beyond the scope of this chapter.

3.5 Conceptual Design

he most important step in the design of a new bridge is the structural design concept that will ulti-
mately be developed into a inal design and then be constructed. he cost, the appearance, the reliability
and serviceability of the facility will all be determined, for good or for bad, by the conceptual design of
the structure. Towers act as the bridge framework because the superstructure will hang from the towers;
thus, towers play a signiicant role in the conceptual design process. Once it is constructed, the bridge
will always be there for users to admire or to criticize. he user ultimately pays for the cost of a structure
and also pays for the cost of maintaining that structure.
Gimsing and Georgakis (2012) treat the conceptual design issues of both cable-stayed and suspen-
sion bridges very extensively and present examples to help guide bridge designers. Chapter 1 of Bridge
Engineering Handbook, Second Edition: Fundamentals presents typical practice and general principles
of bridge conceptual design.
A recent trend is to employ an architect to be part of the design team. An architect if employed
should start during the conceptual design phase, as the esthetics of the bridge is set during that phase
of the work.
Generally the role of the engineer is to develop the structure adequacy and ensure the structural
function of the bridge according to the codes of practice. he role of the architect will generally involve
only the esthetics function; however, there are no codes of practice for that.
heir two roles do overlap in achieving an esthetic and functional structural design that is within
budget. As the common objective of both the engineer and the architect is to build an elegant and eco-
nomical bridge, cooperation and respect between them is vital to the success of their joint efort.
However, diferences may occur when the esthetic desires of the architect and the structural calcu-
lations of the engineer conlict. Towers, the most visible components of the bridge, are oten the focal
point for this type of conlict. Each professional should understand if these diferences in viewpoint
occur; they must be resolved so that a successful and fruitful union between their two disciplines will
produce a strong and beautiful bridge.

3.5.1 Materials
Until the 1970s, steel was the predominant material used for towers for both cable-stayed and suspen-
sion bridges. Such towers were oten rectangular in elevation, having cross sections shaped as rectangles,
cruciforms, tees, or other similar shapes that could be easily fabricated in steel.
Two examples of such suspension-bridge steel-tower designs are the typical, rectangular steel towers
of the two Delaware Memorial Bridges: the irst bridge was built in 1951, and the parallel bridge was
built in 1968 (Figure 3.15).
An example of a cable-stayed bridge that is an exception to the rectangular tower form, is the modiied
A frame, weathering steel towers of the Luling Bridge near New Orleans, 1983 (Figure 3.16).
Towers 75

FIGURE 3.15 Delaware Memorial Bridges. (Courtesy of D. Sailors.)

FIGURE 3.16 Luling Bridge, New Orleans, Louisiana. (Courtesy of Charles Seim.)

he cross section of a steel tower is usually designed as a series of adjoining cells, formed by shop-
welding steel plates together in units from 20 to 40 t (6–12 m) long. he steel cellular towers for a
cable-stayed bridge with cables framing into the towers must be designed for the local forces from the
numerous anchorages of the cables. he steel towers for a suspension bridge, and for cable-stayed bridges
with stays passing over the top of the tower in saddles, must be designed for the local, concentrated load
from the saddles.
An excellent example of such a steel tower is the new 525 t (160 m) tower for the Suspension Span of
the San Francisco–Oakland Bay Bridge East Span. his tower is composed of four separated pentagonal,
cross-sectional shaped shats, connected by shear-link beams. he tower shats are separated about 2 m,
allowing light to permeate between the shats that are tapered toward the top to enhance their appear-
ance. he shear-link beams are both attractive esthetic elements, and the structural steel beams yield in
shear and absorb energy when activated by strong earthquakes (Figure 3.17).
76 Bridge Engineering Handbook, Second Edition: Substructure Design

160 m (525 ft)

Shear link
beams
Road decks
Bike
pedestrian path

Tower legs

Tower foundation

Footing box

CIDH piles
Bay mud
Rock sockets
Bedrock

FIGURE 3.17 Tower of new San Francisco–Oakland Bay Bridge self-anchored suspension span.

For suspension bridges and cable-stayed structures, starting about the 1970s, reinforced concrete
began to be extensively used in towers. Concrete towers are usually designed as hollow shats to save
weight and reduce the amount of concrete and reinforcing bars required. As with steel towers, concrete
towers must be designed for the concentrated load from the saddles at the top, if used, or for the local
forces from the numerous anchorages of the cables framing into the tower shats.
Towers designed in steel will be lighter than towers designed in concrete, thus giving potential for
savings in foundation costs. Steel towers will generally be more lexible, more ductile, and can be erected
in less time than concrete towers. Steel towers will require periodic maintenance—painting—although
weathering steel can be used for nonmarine environments as for the Luling Bridge, as noted above.
Costs of steel or concrete towers can vary with a number of factors; hence, market conditions,
contractor’s experience, equipment availability, design details, and site-speciic inluences will likely
determine whether steel or concrete is the most economic material.
During the conceptual design phase of the bridge, approximate construction costs of all the materials
need to be developed and compared. If life-cycle cost is important, then maintenance operations and the
frequencies of those operations need to be evaluated and compared, usually by present worth evaluation.

3.5.2 Forms and Shapes

3.5.2.1 Cable-Stayed Bridge Towers
Towers of cable-stayed bridges can have a wide variety of shapes and forms (Figure 3.18). For conceptual
design, the heights of cable-stayed towers, tower height (TH), above the deck can be assumed to be
approximately 20% of the main-span length, span length (SL). Figure 3.18 lists the ratios of typical bridges.
To this value must be added the structural depth of the girder and the clearance to the foundation to
determine the approximate total tower height. he inal height of the towers will be determined during
the inal design phase. Figure 3.19 lists distinctive towers for cable-stayed and suspension bridges.
Towers 77

(a) Single Tower, I (b) Double vertical shats, (c) Double vertical shats (d) Double cranked shats
H with cross struts above with cross strut above the
the roadway roadway

Stonecutters Bridge Øresund Bridge (Gimsing John James Audubon Talmadge Memorial Bridge
(Morgenthal et al. 2010), 2009; Oresund Bridge Bridge (Fossier and (Tang 1995), carries four
carries dual three-lane 2012), carries four lanes Duggar 2007), carries lanes of US 17 to I-16,
highway, crosses Rambler of European route E20 four lanes of LA 10, crosses Savannah River,
Channel, Hong Kong. and Oresund railway line, crosses Mississippi River, Georgia, USA. Pylon
Pylon height: 298 m crosses Oresund Strait Louisiana, USA. Pylon height: 127 m (418 t),
(978 t), with reinforced between Copenhagen height: 152.4 m (500 t), longest span 335 m
concrete from base up to (Denmark) and Malmö longest span: 482 m (1100 t), clearance below:
175 m level and composite (Sweden). Pylon height: (1583 t), clearance 56 m (185 t), reinforced
top 120 m consisting of 204 m (669 t), reinforced below: 40 m (130 t), concrete, opened:
inner concrete ring with a concrete, longest span: reinforced concrete, November 1990, TH/SL: 0.2
stainless steel skin, longest 490 m (1608 t), clearance opened: May 5, 2011,
span: 1018 m (3340 t), below: 57 m (187 t), TH/SL: 0.23
clearance below: 73.5 m opened: July 1, 2000,
(241 t), opened: December TH/SL: 0.3
2009, TH/SL: 0.22

Roadway direction

Inclined angle
θ = 40° ≈ 60°

(e) Inclined shats, A (f) Inclined shats, (g) Inverted Y (h) Single inclined tower
diamond
Bridge to Russky Island ANZAC Bridge (Moore Yangpu Bridge (Ma and Sundial Bridge (Sundial
(SK-MOST 2011), carries 1996), carries, freeway, Fan 1993), carries Bridge 2013), cantilever
four lanes of roadway, pedestrians and bicycles, six-lane motorway, spar cable-stayed bridge,
crosses Eastern Bosphorus crosses Johnstons Bay, crosses Huangpu River, carries bicycles and
Strait, Vladivostok Sydney, Australia. Pylon China. Pylon height: pedestrians, crosses
(Nazimov peninsula) and height: 120 m (390 t), 223 m (731 t), longest Sacramento River, Redding,
Russian Island longest span: 345 m span: 602 m (1975 t), California, USA, pylon
(Novosiltseva cape). Pylon (1132 t), clearance clearance below: 48 m height: 66 m (217 t),
height: 320.9 m (1052 t), below: 27 m (88 t), (257 t) reinforced clearance below: 8 m
longest span: 1104 m reinforced concrete, concrete, opened: (26 t), opened: July 4, 2004
(3621 t), clearance: 70 m opened: December 2, October 1993,
(230 t), opened: July 2012 1995, TH/SL: 0.27 TH/SL: 0.24
(plans), TH/SL: 0.23

FIGURE 3.18 Generic forms and typical examples of towers for cable-stayed bridges.
78 Bridge Engineering Handbook, Second Edition: Substructure Design

FIGURE 3.18 (Continued) Generic forms and typical examples of towers for cable-stayed bridges.

(a) (b) (c) (d)

Special Pylon with a very Unique Pylon with four hird Nanjing Yangtze Single tower self-anchored
long neck inverted Y-shape, legs River Bridge (Cun et al. suspension bridge, San
instead of straddling the Rio–Antirrio bridge 2009), cable-stayed bridge Francisco-Oakland Bay
roadway, located in (Combault et al. 2005), with two curve Bridge East Span (Nader and
between. ive-span four-pylon ladder-shaped steel Maroney 2007), carries
Shanghai Yangtze River cable-stayed bridge, towers, carries six-lane 10-lane I-80, bike and
Bridge (Zhang and Lu carries six-lane roadway freeway, crosses Yangtze pedestrian way, crosses San
2008), cable stayed bridge, and one pedestrian and River, China. Pylon Francisco Bay, California,
carries six-lane freeway, bicycle lane, crosses Gulf height: 215 m (705 t), USA. Tower height: 160 m
crosses Yangtze River, of Corinth, Greece. Pylon longest span: 648 m (525 t), longest span: 385 m
China. Pylon height: 212 m height: 164 m (538 t), (2125 t), clearance (1263 t), clearance below:
(695 t), longest span: longest span: 560 m below: 24 m (79 t), 30 m (100 t), single steel
730 m (2395 t), clearance (1837 t), clearance opened: October 7, 2005. tower with four legs
below: 54 m (177 t), below: 52 m (170 t), TH/SL: 0.3 connected with shear link
opened: October 31, 2009. opened: August 7, 2004. beams, opened: September
TH/SL: 0.22 TH/SL: 0.2 2013. TH/SL: 0.34.

FIGURE 3.19 Distinctive towers for cable-stayed and suspension bridges.

People driving over a bridge view the towers projecting above the roadway, making this portion of the
towers visually the most important feature of the bridge; thus, the towers should be carefully considered
by the designers of the bridge.
he simplest tower form is a single shat, usually vertical (Figure 3.18a). Stay cables from a single tower
can be arranged in a single plane to align with a longitudinal center girder or can be splayed outwardly
to connect with the longitudinal edge girders. Occasionally, the single shat may be inclined longitudi-
nally, usually away from the main span; rarely toward the main span. Even more infrequently, on short,
Towers 79

curved spans, a single tower is inclined transversely, which adds a dynamic factor to the esthetics of the
bridge. he cables are usually arranged in a star array, radiating from the top of the tower.
Two vertical shats straddling the roadway, with or without cross struts above the roadway, form a
simple tower, which can be used with two planes of cables (Figure 3.18b and 3.18c). he stay cables incline
inward to connect to the edge girders or to the edges of a box girder, introducing a tension component
across the deck support system. he tower shats can also be “cranked” or ofset above the roadway (Figure
3.18d). his allows the cables to be aligned in a vertical plane and attached to the girder that can pass
continuously through the towers. his method was used for the Talmadge Bridge, Georgia (Figure 3.20).
A horizontal strut is always used between the tower shats at the ofset to stabilize the towers.
he two shats of cable-stayed bridges can be inclined inward toward each other to form a modiied
“A” frame, similar to that of the Luling Bridge towers (Figure 3.16) or the two shats inclined to bring
the shats tops together to form a full “A” frame (Figure 3.18e). he two planes of stay cables are inclined
outward, producing a desirable compression component across the deck support system.
Most of the two shats of the H-shaped, A-shaped, and the quasi-diamond- and full-diamond-shaped
towers for cable-stayed bridges are designed as straight members, for ease of construction. A few of the
recently built bridges have curved shats. he hird Nanjing Yangtze Bridge is an excellent example
(Figure 3.19d). As noted in Section 3.5, the form of these towers was copied from the Eifel Tower in Paris
and was the irst cable-stayed bridge in China with curved steel towers.
he form of the towers of a cable-stayed bridge below the roadway is also important for reasons of
both esthetics and costs. People viewing a bridge from a distance will see the towers as part of a complete
structural unit. his total view is important because it displays the motif of the bridge, and it should be
carefully considered by the designers of the bridge.
he shats of the towers for a modiied “A” frame bridge can be carried down to their foundations at
the same slope as was used above the roadway and particularly on sites with low clearances.
However, at high-clearance locations, if the shats of the towers for a full “A” frame or for an inverted
“Y” frame are carried down to the foundations at the same slope as above the roadway, the foundations
may become very wide and costly.

FIGURE 3.20 Talmadge Bridge, Georgia. (Courtesy of T. Y. Lin International.)

80 Bridge Engineering Handbook, Second Edition: Substructure Design

Sometimes the lower shats are inclined inward under the roadway, producing a modiied or “squat”
diamond (Figure 3.18f), similar to the towers of the Glebe Island Bridge, Sydney, Australia (Figure
3.7). For very high roadways, the inward inclination can form a full diamond, as in the Cooper River
Bridge, Charleston, South Carolina (Figure 3.8), or a double diamond as in the Baytown Bridge, Texas
(Figure 3.21). For very long spans requiring tall towers, the “A” frame can be extended by using a single
vertical shat forming an inverted “Y” shape (Figure 3.18g) as in the Yang Pu Bridge (Figure 3.19b) and as
in the Shanghai Yangtze River Bridge, China. his form is very efective for very long spans for which addi-
tional tower height is required, and the inclined legs add stifness and frame action for wind resistance.
he numbers of shats within the towers of cable-stayed bridges can vary from one to four; the Rio-
Antirrio Bridge, Greece, has four shats (Figure 3.19c). hree-shat towers generally are not used for
cable-stayed bridges, except for those with very wide decks. Four-shat towers are best used to support
two separate structures, rather than to support one wide deck. he four shats of a tower may share a
common foundation, or two pairs of shats may have their own foundations, depending on costs.

3.5.2.2 Suspension Bridge Towers

Suspension bridges are designed to be used on much longer spans than are cable-stayed bridges. hus,
the towers of a suspension bridge must be far more robust than are those of a cable-stayed bridge, to
support adequately the large loads and great wind forces a suspension bridge will encounter during its
life span.
Usually the towers of suspension bridges follow a traditional design, using two vertical shats and two
planes of cables, as is illustrated by the steel towers for the Delaware Memorial Bridges (Figure 3.15).
However, concrete towers have recently proven to be economical for some bridges. he very long span
of the 4626 t (1410 m) Humber Bridge, England, 1983, used uniformly spaced multistruts and concrete
towers (Figure 3.22). he crossing of the Great Belt sea way in Denmark (Figure 3.23), opened in 1999,
has concrete towers 833 t (254 m) high with two struts—one near the mid-height and one at the top.
he shats of suspension bridge towers are usually designed for the full height of the towers, from the
foundation to the cable saddles. he tower must accommodate the large aspect ratio for good esthetics.
Only a few single-cable suspension bridges have been designed with an “A” or an inverted “Y” form of
towers. Typical shapes and forms of suspension bridges are shown in Figure 3.24.
For conceptual designs, the heights of suspension bridges towers, above the deck, depend on the sag-
to-span ratio, which can vary from about 1:8 to 1:12. A good preliminary value is approximately 1:10.
To this value, one must add the structural depth of the deck and the clearance to the foundations to

FIGURE 3.21 Baytown Bridge, Texas. (Courtesy of T. Y. Lin International.)

Towers 81

FIGURE 3.22 Humber Bridge, England. (Courtesy of Charles Seim.)

FIGURE 3.23 Great Belt Bridge, Denmark. (Courtesy of Ben C. Gerwick, Inc.)

obtain the approximate total tower height. he shats are usually connected together with several struts,
or cross bracing along the height of the tower, or the shats are connected at the top with a large single
strut. Some form of strut between the towers is usually required for suspension bridges because the large
cables carry lateral wind and seismic loads to the tops of the tower shats, which then need to be braced
against each other with struts or “X” cross bracing to form a tower-frame action.
82 Bridge Engineering Handbook, Second Edition: Substructure Design

(a) (b) (c) (d)

Akashi Kaikyō Bridge Höga Kusten Bridge, Østbroen (East Bridge Golden Gate Bridge,
(HSBA 1998), longest suspension bridge with 2012), suspension bridge, suspension bridge, two
suspension bridge in the two reinforced concrete two reinforced concrete steel towers with moment
world with two steel towers, carries four lanes towers with moment frame, carries six lanes of
X-braced moment frame of European route E4, frame of two vertical US 101, pedestrians and
towers, carries six lanes of crosses the mouth of the shats straddling the bicycles, crosses Golden
roadway, crosses Akashi Ångermanälven river, roadway with two cross Gate, San Francisco, USA.
Strait, Japan. Tower Sweden. Tower height: struts above the roadway, Tower height: 227.4 m
height: 282.8 m (928 ft), 180 m (591 t), longest carries motor vehicles, (746 t), longest span: 1280
longest span: 1991 m span: 1210 m (3970 t), crosses Great Belt, m (4200 t), clearance
(6532 ft), clearance below: clearance below: 40 m Denmark. Tower height: below: 67 m (220 t),
65.7 m (216 ft), opened: (131 t), opened: 254 m (833 t), longest opened: April 19, 1937,
April 5, 1998, TH/SL: 0.11 December 1, 1997, span: 1624 m (5328 t), TH/SL: 0.13
TH/SP: 0.12 clearance below: 65 m
(213 t), opened: June 14,
1998, TH/SL: 0.12

FIGURE 3.24 Generic forms and typical examples of towers for suspension bridges.

3.5.3 Erection
he most crucial stage in the life of a bridge is the erection time of that structure because the risk of
adverse happenings is highest during this phase. Adverse happenings can occur from the high cost of
opening the bridge to service late, to locked-in unanticipated stresses because of faulty erection proce-
dures, to partial or full collapse. Bridge designers have little control of the irst risks; however, unan-
ticipated stresses or partial or full collapse are very troubling because they can be prevented by having
a detailed erection scheme. Ordinary towers can usually be erected without much diiculty; however,
thin, curved, or inclined towers or towers temporarily supporting or resisting erection forces or loads
require a detailed erection plan.
Some bridge designers say that erection is the responsibility of the contractors; however, if something
listed above does happen, everyone will become involved, including the designer, and someone will end
up paying money.
A better solution is to design a detailed erection scheme that will construct the structure to the proper
camber, position, and alignment and with acceptable stresses in all the members. he best person to
design this erection scheme is the bridge designer, because the designer knows the structure intimately,
works on the design for a year, and develops a bridge model for the design of the bridge; that model can
Towers 83

also be used to develop all erection stages for the structure. If this is done, the speciications should
allow the contractor full freedom to modify that scheme or to develop a separate erection scheme. If the
speciications require the contractor to develop the erection scheme, the bridge designer should check
and approve the scheme before erection begins.
During the concept-design phase, many diferent tower forms and cable arrangements may be con-
sidered; each should be evaluated for esthetics, constructability, and cost. Each alternative considered
should have at least one method of erection developed during the concept-design phase to ensure that
the scheme under consideration is constructible. he costs of erecting unusual tower designs such as
inclined towers, or curved spars, can be diicult to estimate and may add signiicant costs to the project.

3.6 Final Design

he AASHTO Standard Speciications for Highway Bridges (AASHTO 2002) and the AASHTO LRFD
Bridge Design Speciications (AASHTO 2012) apply to bridges 150 m (500 t) or less in span. For impor-
tant bridges, and for long-span cable-supported bridge projects, special design criteria may need to be
developed by the owner and/or the designer. he special-design criteria may also need to be developed in
cooperation with the owners of the facility, so as to include their operations, maintenance requirements,
and bridge performance expectations ater large natural events such as earthquakes. See Chapter 9 for
suspension bridge design and Chapter 10 for cable-stayed bridge design in Bridge Engineering Handbook,
Second Edition: Superstructure Design. Troitsky (1988), Podolny and Scalzi (1986), and Walther et al.,
(1999) also present detailed design theory for cable-stayed bridges.
Design methodology for the towers should follow the same practices as does the design methodol-
ogy for the entire bridge. he towers should be part of a global analysis in which the entire structure is
treated as a whole. From the global analysis, the towers may be modeled as substructural units with their
forces and deformations imposed as boundary conditions.
Detailed structural analyses form the basis for the inal designs of the tower, its components, and
its connections. Both cable-stayed and suspension bridges are highly indeterminate and both require
careful analyses by at least one geometric nonlinear program if erections are to be determined. Prudent
design should also include analysis of at least one erection scheme to demonstrate that an experienced
contractor may erect the structure.

3.6.1 Design Loads

he towers are subject to many diferent load cases. he towers, as well as the entire structure, must
be analyzed, designed, and checked for the controlling load case. Chapter 6 of Bridge Engineering
Handbook, Second Edition: Fundamentals presents a detailed discussion of highway bridge loading.
he weight of the superstructure, including the self-weight of the towers, is obtained in the design
process by utilizing the unit weights of the materials used for both the tower and the superstructure.
he forces that are distributed to the tower can be calculated by a structural analysis of the completed
structure. he forces distributed to the tower may be analyzed for a staged erection of the superstruc-
ture, to determine whether the towers will be over-stressed during construction of the superstructure.
Design loads from traic using the bridge, such as trains, transit, trucks, or pedestrians, are usually
prescribed in design codes and speciications or by the owners of the facility. hese loads move across
the bridge, and the forces imparted to the towers from such moving loads must be obtained from a
structural analysis. hese are all gravity efects, acting downward on the superstructure, but can induce
both vertical and horizontal forces on the towers.
A current trend for spanning wide widths of waterways is to design multispan cable-stayed and sus-
pension bridges, linked together to form a long, continuous structure with the towers evenly spaced for
uniform appearance, and having a short span at each end. hese multispan bridge roadways will delect
84 Bridge Engineering Handbook, Second Edition: Substructure Design

excessively unless the towers are specially designed for added stifness. his is because ordinary towers
are not suiciently stif to resist the pull from cables that are supporting the lexible, multispan roadway.
Several methods have been proposed to stifen these towers, such as adding four shats to the towers
as was done to the Rio Antirrio Bridge crossing of the Gulf of Corinth, Greece (Figure 3.19). A second
method would be to use cables arranged in various ways to stifen the towers externally; but this is
beyond the scope of this chapter.
Towers are also subjected to temperature-induced displacements, from the superstructure and the
cables framing into the towers, and the temperature-induced movement of the tower itself. Towers may
expand and contract diferentially along their tower height because of heat from the sun that shines on
them from morning until sunset. Such temperature efects may cause delection and torsional twisting
along the height of the tower.
Wind blowing on the towers as a bluf shape will induce both forces and displacements in the tower.
Force will be induced into the cables by the wind pressure on the superstructure and from the wind
forces on the cables themselves. hese additional forces will be carried to the towers, which must be
designed for them.
For long-span bridges and locations with known high-wind speeds, the wind factor should be treated
as a dynamic loading. his will usually require a wind-tunnel test on a sectional model of a proposed
superstructure in a wind tunnel and for important bridges, a full aeroelastic model test in a large wind
tunnel. See Chapter 22 of Bridge Engineering Handbook, Second Edition: Fundamentals. Under certain
wind lows, the wind may excite the tower itself. In the rare instances where wind-induced excitation
of the tower does occur, appropriate changes in the cross section of the tower may be made, or a faring
added, to change the dynamic characteristics of the towers. If these methods are not efective in chang-
ing the response, installing tuned-mass dampers at various locations within the towers will dampen out
excessive vibrations. hese types of dampers need periodic maintenance, which requires ladders and
elevators for access by maintenance personnel.
Seismic excitations should be treated as dynamic inertia loadings, inducing responses within the
structure by exciting the vibrational modes of the towers. Tuned mass dampers can also be installed to
dampen seismic excitations. Seismic forces and displacement may control tower design in locations with
high seismic activity. For locations with lower seismic activity, the tower design should be checked at
least for code-prescribed seismic loadings. he dynamic analysis of bridges is discussed in Chapter 3 of
Bridge Engineering Handbook, Second Edition: Seismic Design.
A full analysis of the inal design will reveal all the forces, displacements, and other design
requirements for all loading cases for the inal design of the towers.

3.6.2 Other Design Considerations

Suspension bridge cables pass over cable saddles that are usually anchored to the top of the tower.
A cable produces a large vertical force, as well as smaller, but important, transverse and longitudinal
forces from temperature, wind, earthquake, or any unbalanced cable forces between the main and the
side spans. hese forces are transmitted through the cable-saddle anchorage at each cable location, to
the top of the tower. he towers and the permanent saddle anchorages must be designed to resist these
cable forces.
he erection of a suspension bridge must be analyzed and the chosen sequence shown on the con-
struction plans. To induce the correct loading into the cables of the side span, the erection sequence usu-
ally requires that the saddles be displaced toward the side spans. his is usually accomplished for short
spans by displacing the tops of the towers by pulling them with heavy cables. For long spans, the saddles
can be displaced temporarily, on rollers. As the stifening girders or trusses are being erected into posi-
tion and the cable begins to take loads, the towers or saddles are gradually rolled into inal alignment on
the tower. Ater erection of the stifening girders or trusses is completed, the saddles are permanently
fastened into position to take the unbalanced cable loads from the center and the side spans.
Towers 85

At the deck level, other forces may be imposed on the tower, from the box girder or the stifening truss
carrying the roadway. Such forces depend on the structural framing of the connections of the deck and
the tower. Traditional suspension bridge designs usually terminate the stifening truss, or the box girder,
at the towers; that produces transverse and longitudinal forces on the tower at this point. More recent
suspension bridge designs usually provide for the passing of a box girder continuously through the
tower opening; this may produce transverse forces, but not longitudinal forces. For this arrangement,
the longitudinal forces must be carried by the stiing girder or trusses to the abutments.
he most critical area of the tower design is the tower-to-foundation connection. Both shear forces
and moments are at a maximum at this point. Anchor bolts are generally used at the base of steel tow-
ers. Such bolts must be proportioned to transfer overturning loads from the tower to the bolts. he bolts
must be deeply embedded in the concrete footing block in order to transfer their loads to the footing
reinforcement.
Providing good drainage for rainwater running down the tower shats will increase the life of the steel
paint system at the tower base and will provide some protection to the anchor bolts.
Concrete towers must be joined to the foundations with full shear and moment connections.
Lapped reinforcing bar splices are usually avoided as the lapping tends to congest the connections;
the strength of the bars cannot then be developed, and lapped splices cannot be used for high-seismic
areas. Using compact mechanical or welded splices will result in less congestion, with easier place-
ment of concrete around the reinforcement, and a more robust tower-to-footing connection. he
design of the joint of the tower shats to the foundation should produce a constructible, eicient, and
reliable connection.
he cable arrangements for cable-stayed bridges are many and varied. Some arrangements termi-
nate the cables in the tower, whereas other arrangements pass the cable through the tower on cable
saddles. Cables terminating in the tower may pass completely through the tower cross section and then
be anchored on the far side of the tower. his method of anchoring produces compression in the tower
cross section at the anchorage points. Cables can also be terminated at anchors within the walls of the
tower, producing tension in the tower cross section at the anchorage points. hese tension forces require
special designing to provide reliable, long-life support for the cables.
As for suspension bridges, the erection of cable-stayed bridges must be analyzed, and the sequence be
shown on the construction plans. he girders, as they are cantilevered outward from the towers, are very
vulnerable. he most critical erection sequence is just before the closing of the two arms of the girders,
at the center of the span. High winds can displace the arms and torque the towers, and heavy construc-
tion equipment can load the arms that are yet without beneit of the girder continuity to distribute the
loads to towers.

3.7 Construction
he towers and superstructure should be constructed according to an erection plan as noted in
Section 3.5.3.
Towers constructed of structural steel are usually fabricated in a shop by welding together steel plates
and rolled shapes to form cells. Cells must be large enough to allow welders and welding equipment, and
if the steel is to be painted, painters and cleaning and painting equipment inside each cell.
he steel tower components are transported to the bridge site and are erected by cranes and are either
welded or bolted together with high-strength bolts. For bolting, the contractor should use a method
of tensioning the high strength bolts to give consistent results needed to achieve the required tension
such as turn-of-the-nut method. Field welding presents diiculties in holding the component rigidly in
position while the weld is completed. Field welding may be diicult to control when exposed to windy
weather, making ductile welds diicult, particularly the vertical and overhead welds. Field welding
should be made within a protective covering that keeps out water and wind. Full-penetration welds
require backup bars that must be removed carefully if the weld is subject to fatigue loading.
86 Bridge Engineering Handbook, Second Edition: Substructure Design

Towers constructed of reinforced concrete are usually cast in forms that can be removed and reused,
or “jumped,” to the next level. Placing height for concrete is usually restricted to approximately 20–40 t
(6–12 m), to limit pressure from the freshly placed concrete. Reinforcing bar cages are usually preas-
sembled on the ground, or on a work barge, and are lited into position by crane. his requires the rein-
forcing bars to be spliced with each lit. Lapped splices are the easiest to make, but these are not allowed
in seismic areas.
Slip forming is an alternative method that uses forms that are pulled slowly upward, reinforcing bars
positioned and the concrete placed in one continuous operation around the clock until the tower is com-
pleted. Slip forming can be economical, particularly for constant cross-section towers. Some changes
in cross-section geometry can be accommodated. For shorter spans, precast concrete segments can be
stacked together and steel tendons tensioned to form the towers.
Tower designers should consider the method of erection that contractors may use in constructing
the towers. Oten the design can reduce construction costs by incorporating more easily fabricated and
assembled steel components or easily assembled reinforcing bar cages and tower shapes that are easily
formed. Of course, the tower design cannot be compromised just to lower erection costs.
Some engineers and many architects design towers that are angled longitudinally toward or away
from the main span or are curved or kinked. his can be done if such a design can be justiied structur-
ally and esthetically, and the extra cost can be covered within the project budget. hese types of towers
require special erection methods.
Many towers of cable-stayed bridges have legs sloped toward each other to form an “A,” an inverted
“Y,” a diamond, or similar shapes. hese are not as diicult to construct as the longitudinally inclined
tower design. he sloping concrete forms can be supported by vertical temporary supports and cross
struts that tie the concrete forms together for each shat. his arrangement braces the partly cast con-
crete tower legs against each other for support. Some of the concrete form supports for the double-
diamond towers of the Baytown Bridge are visible in Figure 3.19.
As the sloped legs are erected, the inclination may induce bending moments and lateral delection
in the plane of the slope of the legs. Both of these secondary efects must be adjusted by jacking the
legs apart by a calculated amount of force or displacement to release the locked-in bending stresses.
If the amount of secondary stress is small, then cambering the leg to compensate for the delection
and adding material to lower the induced stress can be used. he jacking procedure adds cost but is
an essential step in the tower erection. Neglecting this important construction detail can “lock-in”
stresses and delections that will lower the factor of safety of the tower and, in an extreme case, could
cause a failure.
Tower construction usually requires special equipment to erect steel components or concrete forms
to the full height of the tower. Suspension bridges and some cable-stayed bridges require cable saddles
to be erected on the tower tops. Floating cranes rarely have the capacity to reach to the heights of tow-
ers designed for long spans. Tower cranes, connected to the tower as it is erected, can be employed for
most tower designs and are a good choice for handling steel forms for the erection of concrete towers. A
tower crane used to jump the forms and raise materials can be seen in Figure 3.8. Occasionally, vertical
traveling cranes are used to erect steel towers by pulling themselves up the face of the tower following
the erection of each new tower component.
Because the tower erection must be done in stages, each stage must be checked for stability and for
stresses and delections. he tower construction speciications should require the tower erection be
checked by an engineer, employed by the contractor, for stability and safety at each erection stage. he
construction speciications should also require the tower erection stages to be submitted to the design
engineer for an evaluation and approval. his evaluation should be full enough to determine whether
the proposed tower erection staging will meet the intent of the original design or needs to be modiied to
bring the completed tower into compliance. Chapters 1 and 4 of Bridge Engineering Handbook, Second
Edition: Construction and Maintenance present more detailed construction procedure and techniques
for long-span bridges.
Towers 87

3.8 Summary
Towers provide the structural and visible means of support of the bridge superstructure. Towers project
above the roadway and are the most visible structural elements in a bridge. Towers usually form visible
portals through which people pass as they travel from one point to another. hey give the bridge, for
good or for bad, its character, its motif, and its identifying esthetic statement and form the enduring
impression of the bridge in people’s minds.
Towers are the most critical structural element in the bridge as their function is to carry the weight
of the bridge and the forces imposed on the bridge to the foundations. Unlike most other bridge com-
ponents, they cannot be replaced during the life of the bridge. Towers must fulill their function in a
reliable, serviceable, economical, and esthete manner for the entire life of the bridge. Towers must also
be practicable to erect without extraordinary expense; the exception to this economical requirement is
the owners or the public want a spectacular bridge and are willing to pay for the extra cost.
Practicable tower shapes for cable-stayed bridges are many and varied. hese towers can have one or
several shats arrayed from vertical to inclined, forming various shapes. Practicable tower shapes for a
suspension bridge are usually restricted to two vertical shats connected with one or several cross struts,
although single shats have been used on a few suspension bridges.
In the early 1990s, a trend began where eiciency and low cost were not always an objective because
the owner or the public, or both, desires spectacular, picturesque, or distinctive bridges. his resulted
in coniguring stay cables and a few suspension bridge cables in unusual arrays that can dominate the
towers and act as the principle esthetic statement of the bridge or the opposite of featuring towers that
have unusual shapes, kinks, or inclination to add visual impact. his trend will continue into the fore-
seeable future.
he conceptual design phase is the most important phase in the design of towers for long span bridges.
his phase sets, among other items, the span length, type of deck system, and the materials and shape
of the towers. It also determines the esthetic, economics, and constructability of the bridge. A concep-
tual erection scheme should be developed during this phase to ensure the bridge can be economically
constructed.
he inal design phase sets the speciic shape, dimensions, and materials for the bridge. If a usual
tower design is used, the tower erection should also be shown. It is preferred that the design engineer
follow the project into the construction stages. he design engineer must understand each erection step
that is submitted by the contractor to ensure the construction complies with the design documents. he
owner assured only by this means that the serviceable and reliability that he is paying for is actually
achieved in construction.
he successful design of towers for cable-stayed and suspension bridges involves many factors and
decision that must be made during the conceptual and design phases and the construction phase of the
project. he inal judge of a successful project is always made by the people who use the facility, pay for
its construction and maintenance, and view the results of all the efort to provide a long-life bridge to
service society (Cerver 1992).

References
AASHTO. 2002. Standard Speciications for Highway Bridges, 17th Edition, American Association of State
Highway and Transportation Oicials, Washington, DC.
AASHTO. 2012. AASHTO LRFD Bridge Design Speciications, Customary U.S. Unit, 2012, American
Association of State Highway and Transportation Oicials, Washington, DC.
Billington, D. P. 1983. he Tower and the Bridge, he New Art of Structural Engineering, Basic Books,
New York, NY.
Combault, J., Pecker, A., Teyssandier, J. P. and Tourtois, J. M. 2005. “Rion-Antirion Bridge, Greece-Concept,
Design, and Construction,” Structural Engineering International, 15(1): 22–27.
88 Bridge Engineering Handbook, Second Edition: Substructure Design

Cun, B., Zhao, C. H., Dong, M. and Tang, L. 2009. “Design of Steel-Concrete Segment of Main Tower of
the hrid Naijing Yangtze River Bridge,” Highway, No. 5, Beijing, China. (In Chinese)
Fossier, P. and Duggar, C. 2007. “John James Audubon Bridge Design-Build Project Update,” 2007
Louisiana Transportation Engineering Conference, February 12, Baton Rouge, LA.
Gimsing, N. J. 2009. “From Bridges across Great Belt and Øresund towards a Femern Belt Bridge,” IABSE
Workshop – Recent Major Bridges, May 11–20, Shanghai, China.
Gimsing, N. J. and Georgakis, C. T. 2012. Cable Supported Bridges - Concept and Design, 3rd Edition, John
Wiley & Sons, New York, NY.
HSBA. 1998. he Akashi-Kaikyo Bridge – Design and Construction of the World’s Longest Bridge, Honshu-
Shikoku Bridge Authority, Japan.
Leonhardt, F. 1984. Bridges, Aesthetics and Design, MIT Press, Cambridge, MA.
Ma, X. B. and Fan, Q. G. 1993. “Construction Planning and Management of Yangpu Bridge Main Tower,”
Construction Technology, No. 3. Beijing, China. (In Chinese).
Moore, D. 1996. To build a Bridge, Glebe Island, Sydney, Australia, Chapter & Verse, Sydney, Australia.
Morgenthal, G., Sham, R. and West, B. 2010. “Engineering the Tower and Main Span Construction of
Stonecutters Bridge,” Journal of Bridge Engineering, ASCE, 15(2): 144–152.
Nader, M. and Maroney, B. 2007. “One-of-a-Kind Design, he New San Francisco-Oakland Bay Bridge
Eastern Span”, STRUCTURE magazine, October.
Øresund Bridge. 2013. http://en.wikipedia.org/wiki/Great_Belt_Fixed_Link.
Petroski, H. 1996. Engineers of Dreams, Vintage Books, New York, NY.
Podolny, W. and Scalzi, J. B. 1986. Construction and Design of Cable Stayed Bridges, Second Edition,
John Wiley & Sons, New York, USA.
Seim, C. 1996, “San Francisco Bay’s Jeweled Necklace,” ASCE Civil Engineering, 66(1): 14A, January.
Sundial Bridge. 2013. http://www.turtlebay.org/sundialbridge.
Tang, M. C. 1995. “Talmadge Memorial Bridge, Savannah, Georgia,” Structural Engineering International,
5(1): 15–16.
SK-MOST. 2011. Construction of acable-stayed bridge to the Russky Island across the Eastern Bosphorus
StraitinV ladivostok, http://rusbridge.net/2011/01/.
Troitsky, M. S. 1988. Cable Stayed Bridges, Van Nostand Reinhold Co, New York, NY.
Walther, R., Houriet, B., Isler, W., Moia, P. and Klein, J.F. 1999. Cable Stayed Bridges, 2nd Edition,
homas Telford Ltd. London, UK.
Zhang, C. L. and Lu, Y. C. 2008. “Design of Main Tower of Shanghai Yangtze River Bridge Main Span,”
Shanghai Highway, No. 4, Shanghai, China. (In Chinese).
 4
Vessel Collision
Design of Bridges
Notations ..............................................................................................89
4.1 Introduction ........................................................................................90
Background • Basic Concepts • Application
4.2 Initial Planning ...................................................................................93
Selection of Bridge Site • Selection of Bridge Type, Coniguration,
and Layout • Horizontal and Vertical Clearance • Approach
Spans • Protection Systems
4.3 Waterway Characteristics..................................................................94
Channel Layout and Geometry • Water Depth and
Fluctuations • Current Speed and Direction
4.4 Vessel Traic Characteristics ............................................................95
Physical and Operating Characteristics • Vessel Fleet
Characteristics
4.5 Collision Risk Analysis ......................................................................98
Risk Acceptance Criteria • Collision Risk Models
4.6 Vessel Impact Loads .........................................................................101
Ship Impact • Barge Impact • Application of Impact
Forces • Minimum Impact Requirements • Recent U.S. Barge
Research
4.7 Bridge Analysis and Design ............................................................105
Global Pier Capacity • Local Pier Capacity • Contribution of the
Superstructure • Movable Bridges
4.8 Bridge Protection Measures ............................................................107
Michael Knott Physical Protection Systems • Aids to Navigation
Moffatt & Nichol
Alternatives • Motorist and Vessel Operator Warning Systems
Zolan Prucz 4.9 Summary ............................................................................................109
Modjeski and Masters Inc. References...................................................................................................... 110

Notations
he following symbols are used in this chapter. he section number in parentheses ater deinition of a
symbol refers to the section or igure number where the symbol irst appears or is identiied.
AF = annual frequency of bridge element collapse (Section 4.5.2)
BM = beam (width) of vessel (Figure 4.7)
BP = width of bridge pier (Figure 4.7)
DWT = size of vessel based on deadweight tonnage (1 ton = 2205 lb. = 9.80 kN) (Section 4.4.1)
H = ultimate bridge element strength (Section 4.5.2)
N = number of one-way vessel passages through the bridge (Section 4.5.2)

89
90 Bridge Engineering Handbook, Second Edition: Substructure Design

P = vessel collision impact force (Section 4.5.2)

P BH = ship collision impact force for head-on collision between ship bow and a rigid object
(Section 4.6.1)
PDH = ship collision impact force between ship deckhouse and a rigid superstructure (Section 4.6.1)
PMT = ship collision impact force between ship mast and a rigid superstructure (Section 4.6.1)
PS = ship collision impact force for head-on collision between ship bow and a rigid object (Section 4.6.1)
PA = probability of vessel aberrancy (Section 4.5.2)
PC = probability of bridge collapse (Section 4.5.2)
PG = geometric probability of vessel collision with bridge element (Section 4.5.2)
PF = protection factor for bridge location (Section 4.5.2)
R BH = ratio of exposed superstructure depth to the total ship bow depth (Section 4.6.1)
R DH = reduction factor for ship deckhouse collision force (Section 4.6.1)
V = design impact speed of vessel (Section 4.6.1)
x = distance to bridge element from the centerline of vessel-transit path (Figure 4.7)
φ = angle between channel and bridge centerlines (Figure 4.7)

4.1 Introduction
4.1.1 Background
Vulnerability of critical infrastructures to extreme events have made headlines worldwide in the past
decades due to structural failures, loss of life, and inancial damages due to earthquakes, hurricanes,
storm surge and waves, tsunamis, looding and scour, vessel collisions, and terrorist attacks. For major
bridge structures, the risk and magnitude of such extreme events is oten the controlling load case for
the structure design.
It was only ater a marked increase in the frequency and severity of vessel collisions with bridges that
studies of the vessel collision problem were initiated in the 1980s. In the period from 1960 to 2011, there
have been 36 major bridge collapses worldwide due to ship or barge collision, with a total loss of life
of 342 people. he greatest loss of life occurred in 1983 when a passenger ship collided with a railroad
bridge on the Volga River, Russia. One hundred and seventy six people were killed when the aberrant
vessel attempted to transit through a side span of the massive bridge. Most of the deaths occurred when
a packed movie theatre on the top deck of the passenger ship was sheared of by the low vertical clear-
ance of the bridge superstructure.
Seventeen of the bridge catastrophes mentioned above occurred in the United States, including the
1980 collapse of the Sunshine Skyway Bridge crossing Tampa Bay, Florida, in which 396 m of the main
span collapsed and 35 lives were lost as a result of the collision by an empty 35,000 DWT (deadweight
tonnage) bulk carrier (Figure 4.1). Recent collapse of bridges due to barge collision include the Queen
Isabella Causeway Bridge, Texas, in 2001 that resulted in 8 fatalities, the I-40 Bridge, Oklahoma, in 2002
that resulted in 13 fatalities (Figure 4.2) and the Popps Ferry Bridge, Mississippi in 2009 (Figure 4.3).
A recent collapse due to ship collision was the Eggner’s Ferry Bridge in Kentucky, where a 322-foot
approach span collapsed when hit by 8400 DWT cargo ship on January 26, 2012.
One of the more publicized tragedies in the United States involved the 1993 collapse of a CSX
Railroad Bridge across Bayou Canot near Mobile, Alabama. During dense fog, a barge tow became lost
and entered a side channel of the Mobile River where it struck a railroad bridge causing a large shiting
of the superstructure. he bridge collapsed a few minutes later when a fully loaded Amtrak passenger
train attempted to cross the damaged structure. Forty-seven fatalities occurred as a result of the collapse
and train derailment.
It should be noted that there are numerous vessel collision accidents with bridges, which cause dam-
age that varies from minor to signiicant damage but do not necessarily result in collapse of the structure
or loss of life. A U.S. Coast Guard Study (U.S. Coast Guard 2003) of towing vessels and barge collisions
Vessel Collision Design of Bridges 91

FIGURE 4.1 Sunshine Skyway Bridge ater being struck by the M/V Summit Venture, FL (1980).

FIGURE 4.2 I-40 Bridge over Arkansas River, OK (2002).

FIGURE 4.3 Popps Ferry Bridge, MS (2009).

92 Bridge Engineering Handbook, Second Edition: Substructure Design

with bridges located on the U.S. inland waterway system during the 10-year period from 1992 to 2001
revealed that there were 2692 accidents with bridges. Only 61 of these caused bridge damage in excess
of $500,000, and there were no fatalities during the study period. he study concluded that 90% of the
barge tow accidents were related to human performance (78% to pilot error and 12% to other operational
factors). Only 5% were related to mechanical problems, and for the remaining 5% there was insuicient
information to assign a cause.
In addition to motorist disruption, structural damage and potential loss of life, signiicant environ-
mental damage can also occur in a waterway due to oil and chemical spills as a result of vessel collision.
Examples include the spillage of 170,000 gallons of fuel oil in the Fore River, Maine in 1996 when a col-
lision occurred with a bascule bridge pier of the Million Dollar Bridge that ripped a 9-m hole in a loaded
tanker ship (caused by an underwater protrusion of the concrete support pier footing); and the spillage
of 53,600 gallons of fuel oil into San Francisco Bay in 2007 when a container ship hit one of the main pier
fender systems of the San Francisco–Oakland Bay Bridge during dense fog.
he 1980 collapse of the Sunshine Skyway Bridge was a major turning point in awareness and increased
concern for the safety of bridges crossing navigable waterways in the United States. Investigations and
research subsequent to the Skyway and other major bridge accidents worldwide (National Research
Council 1983; IABSE 1983; Modjeski and Masters 1984; Prucz and Conway 1987) ultimately lead to the
development of the AASHTO Guide Speciication for Vessel Collision Design of Highway Bridges in 1991
(AASHTO 1991). his landmark publication provided the bridge design community (for the irst time)
the ability to evaluate the risk of vessel collision and estimate the magnitudes of impact forces associated
with ship and barge collisions. A second edition of the Guide Speciication was developed by AASHTO in
2009 (AASHTO 2009) to update and incorporate lessons learned from the use of the original 1991 Vessel
Collision Guide Speciication; incorporate current LRFD Bridge Design methodologies; and incorporate
results from barge and ship collision research conducted since the original vessel collision publication.
Current highway bridge design practices in the United States follow the AASHTO speciications
(AASHTO 2009, 2012). he design of railroad bridge protection systems against vessel collision is
addressed in the American Railway Engineering and Maintenance of Way Association (AREMA)
Manual for Railway Engineering (AREMA 2013).
Research and development work in the area of vessel collision with bridges is ongoing, though com-
pared to more mature and established ields such as wind and earthquake engineering, vessel collision
analysis and design is in its infancy stages. Important research needs within the discipline include ship
impact forces, barge impact forces, risk acceptance criteria, physical protection systems, and aids-to-
navigation improvements. As further research results become available, appropriate code changes and
updates could be expected.

4.1.2 Basic Concepts

he vulnerability of a bridge to vessel collision is afected by various factors, including
• Waterway geometry, water stage luctuations, current speeds, and weather conditions.
• Vessel characteristics and navigation conditions, including vessel types and size distributions,
speed and loading conditions, navigation procedures, and hazards to navigation.
• Bridge size, location, horizontal and vertical geometry, resistance to vessel impact, structural
redundancy, and efectiveness of existing bridge protection systems.
• Serious vessel collisions with bridges are extreme events associated with a great amount of uncer-
tainty, especially with respect to the impact loads involved. As designing for the worst case sce-
nario could be overly conservative and economically undesirable, a certain amount of risk must
be considered as acceptable. he commonly accepted design objective is to minimize (in a cost-
efective manner) the risk of catastrophic failure of a bridge component, and at the same time
reduce the risk of vessel damage and environmental pollution.
Vessel Collision Design of Bridges 93

he intent of vessel collision provisions is to provide bridge components with a “reasonable” resistance
capacity against ship and barge collisions. In navigable waterway areas where collision by merchant ves-
sels may be anticipated, bridge structures should be designed to prevent collapse of the superstructure
by considering the size and type of vessel, available water depth, vessel speed, structure response, the
risk of collision, and the importance classiication of the bridge. It should be noted that damage to the
bridge (even failure of secondary structural members) is usually permitted as long as the bridge deck
carrying motorist traic does not collapse (i.e., suicient redundancy and alternate load paths exist in
the remaining structure to prevent collapse of the superstructure).

4.1.3 Application
he vessel collision design recommendations provided in this chapter are consistent with the AASHTO
speciications (AASHTO 2009, 2012), and they apply to all bridge components in navigable waterways
with water depths over 2.0 t (0.6 m). he vessels considered include merchant ships larger than 1000
DWT and typical inland barges.

4.2 Initial Planning

It is very important to consider vessel collision aspects as early as possible in the planning process for a
new bridge, because they can have a signiicant efect on the total cost of the bridge. Decisions related
to the bridge type, location, and layout should take into account the waterway geometry, the navigation
channel layout and the vessel traic characteristics.

4.2.1 Selection of Bridge Site

he location of a bridge structure over a waterway is usually predetermined based on various other
considerations, such as environmental impacts, right-of-way, costs, roadway geometry, and political
considerations. However, to the possible extent, the following vessel collision guidelines should be
followed:
• Bridges should be located away from turns in the channel. he distance to the bridge should be
such that vessels can line-up before passing the bridge, usually at least eight times the length of
the vessel. An even larger distance is preferable when high currents and winds are likely to occur
at the site.
• Bridges should be designed to cross the navigation channel at right angles and should be sym-
metrical with respect to the channel.
• An adequate distance should exist between bridge locations and areas with congested navigation,
port facilities, vessel berthing maneuvers, or other navigation problems.
• Locations where the waterway is shallow or narrow so that bridge piers could be located out of
vessel reach are preferable.

4.2.2 Selection of Bridge Type, Coniguration, and Layout

he selection of the type and coniguration of a bridge crossing should consider the characteristics of the
waterway and the vessel traic, so that the bridge would not be an unnecessary hazard to navigation.
he layout of the bridge should maximize the horizontal and vertical clearances for navigation, and the
bridge piers should be placed away from the reach of vessels. Finding the optimum bridge coniguration
and layout for diferent bridge types and degrees of protection is an iterative process that weighs the
costs involved in risk reduction, including political and social aspects.
94 Bridge Engineering Handbook, Second Edition: Substructure Design

4.2.3 Horizontal and Vertical Clearance

he horizontal clearance of the navigation span can have a signiicant impact on the risk of vessel col-
lision with the main piers. Analysis of past collision accidents has shown that bridges with a main span
less than two to three times the design vessel length or less than two times the channel width are par-
ticularly vulnerable to vessel collision.
he vertical clearance provided in the navigation span is usually based on the highest vessel that
uses the waterway in a ballasted condition and during periods of high water level. he vertical clear-
ance requirements need to consider site-speciic data on actual and projected vessels and must be coor-
dinated with the Coast Guard in the United States. General data on vessel height characteristics are
included in AASHTO (2009) and Larsen (1993).

4.2.4 Approach Spans

he initial planning of the bridge layout should also consider the vulnerability of the approach spans
to vessel collision. Historical vessel collisions have shown that bridge approach spans were damaged in
more than 60% of the total number of accidents. herefore, the number of approach piers exposed to
vessel collision should be minimized, and horizontal and vertical clearance considerations should also
be applied to the approach spans.

4.2.5 Protection Systems

Bridge protection alternatives should be considered during the initial planning phase, because the cost
of bridge protection systems can be a signiicant portion of the total bridge cost. Bridge protection
systems include fender systems, dolphins, protective islands, or other structures designed to redirect,
withstand, or absorb the impact force and energy, as described in Section 4.8.

4.3 Waterway Characteristics

he characteristics of the waterway in the vicinity of the bridge site such as the width and depth of the
navigation channel, the current speed and direction, the channel alignment and cross section, the water
elevation, and the hydraulic conditions have a great inluence on the risk of vessel collision and must be
taken into account.

4.3.1 Channel Layout and Geometry

he channel layout and geometry can afect the navigation conditions, the largest vessel size that can use
the waterway, and the loading condition and speed of vessels.
he presence of bends and intersections with other waterways near the bridge increase the probability
of vessels losing control and become aberrant. he navigation of downstream barge tows through bends
is especially diicult.
he vessel-transit paths in the waterway in relation to the navigation channel and the bridge piers can
afect the risk of aberrant vessels hitting the substructure.

4.3.2 Water Depth and Fluctuations

he design water depth for the channel limits the size and drat of vessels using the waterway. In
addition, the water depth plays a critical role in the accessibility of vessels to piers outside the navi-
gation channel. he vessel collision analysis must include the possibility of ships and barges transit-
ing ballasted or empty in the waterway, as well as the possibility that upstream and downstream
Vessel Collision Design of Bridges 95

water depths may be diferent. For example, a loaded ocean-going barge with a 6-m drat would run
aground before it could strike a pier in 4 m of water, but the same barge empty with a 1-m drat could
potentially strike the pier.
he water level along with the loading condition of vessels inluences the location on the pier where
vessel impact loads are applied, and the susceptibility of the superstructure to vessel hits. he annual
mean high water elevation is usually the minimum water level used in design. In waterways with large
water stage luctuations, the water level used can have a signiicant efect on the structural requirements
for the pier and/or pier protection design. In these cases, a closer review of the water stage statistics at
the bridge site is necessary in order to select an appropriate design water level.

4.3.3 Current Speed and Direction

Water currents at the location of the bridge can have a signiicant efect on navigation and the prob-
ability of vessel aberrancy. he design water currents commonly used represent annual average values
rather than the occasional extreme values that occur only a few times per year and during which vessel
traic restrictions may also apply.

4.4 Vessel Traffic Characteristics

4.4.1 Physical and Operating Characteristics
General knowledge on the operation of vessels and their characteristics is essential for safe bridge
design. he types of commercial vessels encountered in navigable waterways may be divided into ships
and barge tows.

4.4.1.1 Ships
Ships are self-propelled vessels using deep drat waterways. heir size may be determined based on
the DWT. he DWT is the weight in metric tons (1 ton = 2205 lb. = 9.80 kN) of cargo, stores, fuel,
passenger, and crew carried by the ship when fully loaded. here are three main classes of mer-
chant ships: bulk carriers, product carriers/tankers, and freighter/containers. General information
on ship proiles, dimensions, and sizes as a function of the class of ship and its DWT is provided in
AASHTO (2009) and Larsen (1993). he dimensions given in AASHTO (2009) and Larsen (1993) are
typical values, and due to the large variety of existing vessels, they should be regarded as general
approximations.
he steering of ships in coastal waterways is a diicult process. It involves constant communications
among the shipmaster, the helmsman, and the engine room. here is a time delay before a ship starts
responding to an order to change speed or course, and the response of the ship itself is relatively slow.
herefore, the shipmaster has to be familiar with the waterway and be aware of obstructions, navigation,
and weather conditions in advance. Very oten local pilots are used to navigate the ships through a given
portion of a coastal waterway. When the navigation conditions are diicult, tugboats are used to assist
ships in making turns. Ships need speed to be able to steer and maintain rudder control. A minimum
vessel speed of approximately 5 knots (8 km/h) is usually needed to maintain steering. Fully loaded
ships are more maneuverable, and in deep water they are directionally stable and can make turns with
a radius equal to one to two times the length of the ship. However, as the underkeel clearance decreases
to less than half the drat of the ship, many ships tend to become directionally unstable, which means
that they require constant steering to keep them traveling in a straight line. In the coastal waterways of
the United States, the underkeel clearance of many laden ships may be far less than this limit, in some
cases as small as 5% of the drat of the ship. Ships riding in ballast with shallow drat are less maneuver-
able than loaded ships, and, in addition, they can be greatly afected by winds and currents. Historical
accident data indicate that most bridge accidents involve empty or ballasted vessels.
96 Bridge Engineering Handbook, Second Edition: Substructure Design

4.4.1.2 Barge Tows

Barge tows use both deep drat and shallow drat waterways. he majority of the existing bridges cross
shallow drat waterways where the vessel leet is comprised of barge tows only. he size of barges in the
United States are usually deined in terms of the cargo carrying capacity in short tons (1 ton = 2000 lb. =
8.90 kN). he types of inland barges include open and covered hoppers, tank barges, and deck barges.
hey are rectangular in shape, and their dimensions are quite standard so that they can travel in tows.
he number of barges per tow can vary from 1 to over 20, and their coniguration is afected by the con-
ditions of the waterway. A statistical analysis of barge tow types, conigurations and dimensions, which
utilizes barge traic data from the Ohio River, is reported in Whitney, Harik, Griin, and Allen (1996).
In most cases barges are pushed by a towboat. Information on barge dimensions and capacity as well as
on barge tow conigurations is included in AASHTO (2009) and Larsen (1993).
he size of the individual barges afects the collision energy and the shape, size, and strength character-
istics of the bow rake afects the location, extent, and magnitude of the impact loads. he most common
barge type is the hopper barge, which is 10.7 m wide, 59.5 m long, and approximately 4 m deep at its bow
and with a bow rake head log height of 0.6–0.9 m. Because the collision load formulation and the recent
barge tests and studies are all based on a typical hopper barge construction with a 0.6–0.9 m bow rake
head log height (Figure 4.4), it is important for the bridge designer to become knowledgeable of the barge
types transiting the waterway and their orientation in a tow. For example, the use of the hopper barge col-
lision load formulation would have not been appropriate for the much deeper head log of the tanker barge
(Figure 4.5) that hit one of the piers of the I-40 Bridge over the Arkansas River in Oklahoma causing the
collapse of several spans (Figure 4.2). At times, as in this particular case, barges that are usually at the rear
of a tow are turned around with their stronger end, that is normally in contact and pushed by the towboat,
becoming the front of the tow (Figure 4.6) further increasing the likelihood of higher collision loads.
It is very diicult to control and steer barge tows, especially in waterways with high stream veloci-
ties and cross currents. Taking a turn in a fast waterway with high current is a serious undertaking. In
maneuvering a bend, tows experience a sliding efect in a direction opposite to the direction of the turn,
due to inertial forces, which are oten coupled with the current low. Sometimes bridge piers and fenders
are used to line up the tow before the turn. Bridges located in a high velocity waterway near a bend in the
channel will probably be hit by barges numerous times during their lifetime. In general, there is a high
likelihood that any bridge element that can be reached by a barge will be hit during the life of the bridge.

4.4.2 Vessel Fleet Characteristics

he vessel data required for bridge design includes types of vessels and size distributions, transit fre-
quencies, typical vessel speeds, and loading conditions. In order to determine the vessel size distri-
bution at the bridge site, detailed information on both present and projected future vessel traic is
needed. Collecting data on the vessel leet characteristics for the waterway is an important and oten

0.5 m 0.6 m

10.7 m

FIGURE 4.4 Common hopper barge bow rake dimensions.

Vessel Collision Design of Bridges 97

0.9 m
1.9 m

FIGURE 4.5 Bow rake head log height comparison. Typical hopper barge (let) and typical tanker barge (right).
(Note: Barge MM62 was involved in the I-40 Bridge Collapse.)

FIGURE 4.6 Tanker barge approaching a bridge. Note the bow depth of at least 1.8 m and the four push knees.

time-consuming process. he Internet is an important source of navigation data and most U.S. govern-
ment agencies maintain online resources.
Some of the sources in the United States for collecting vessel traic data are as follows:
• U.S. Army Corps of Engineers, District Oices
• Port Authorities and Industries along the Waterway
• Local Pilot Associations and Merchant Marine Organizations
• U.S. Coast Guard, Marine Safety and Bridge Administration Oices
• U.S. Army Corps of Engineers, “Products and Services Available to the Public,” Water Resources
Support Center (WRSC), Navigation Data Center, Fort Belvoir, Virginia, NDC Reports
• U.S. Army Corps of Engineers, “Waterborne Commerce of the United States (WCUS), Parts 1
through 5,” WRSC, Fort Belvoir, Virginia
• U.S. Army Corps of Engineers, “Lock Performance Monitoring (LPM) Reports,” WRSC, Fort
Belvoir, Virginia
• Shipping Registers (American Bureau of Shipping Register, New York; and Lloyd’s Register of
Shipping, London)
• Bridge Tender Reports for movable bridges
Projections for anticipated vessel traic during the service life of the bridge should address both
changes in the volume of traic and in the size of vessels. he following factors need to be considered:
• Changes in region economics
• Plans for deepening or widening the navigation channel
98 Bridge Engineering Handbook, Second Edition: Substructure Design

• Planned changes in alternate waterway routes and in navigation patterns

• Plans for increasing the size and capacity of locks leading to the bridge
• Port development plans
Vessel traic projections that are made by the Maritime Administration of the U.S. Department of
Transportation, Port Authorities, and U.S. Army Corps of Engineers in conjunction with planned chan-
nel deepening projects or lock replacements are also a good source of information for bridge design. As
a very large number of factors can afect the vessel traic in the future, it is important to review and
update the projected traic during the life of the bridge.

4.5 Collision Risk Analysis

4.5.1 Risk Acceptance Criteria
Bridge components exposed to vessel collision could be subjected to a very wide range of impact loads.
Owing to economic and structural constraints bridge design for vessel collision is not based on the
worst case scenario, and a certain amount of risk is considered acceptable.
he risk acceptance criteria consider both the probability of occurrence of a vessel collision and the
consequences of the collision. he probability of occurrence of a vessel collision is afected by factors
related to the waterway, vessel traic, and bridge characteristics. he consequences of a collision depend
on the magnitude of the collision loads and the bridge strength, ductility, and redundancy character-
istics. In addition to the potential for loss of life, the consequences of a collision can include damage to
the bridge, disruption of motorist and marine traic, damage to the vessel and cargo, regional economic
losses, and environmental pollution.
Acceptable risk levels have been established by various codes and for individual bridge projects. he
acceptable annual frequencies of bridge collapse values used generally range from 0.001 to 0.0001. hese
values were usually determined in conjunction with the risk analysis procedure recommended and
should be used accordingly.
he AASHTO provisions (AASHTO 2009, 2012) specify an annual frequency of bridge collapse of
0.0001 for critical bridges and an annual frequency of bridge collapse of 0.001 for regular bridges. hese
annual frequencies correspond to return periods of bridge collapse equal to 1 in 10,000 years, and 1 in
1,000 years, respectively. Critical bridges are deined as those bridges that are expected to continue to
function ater a major impact, because of social/survival or security/defense requirements.

4.5.2 Collision Risk Models

4.5.2.1 General Approach
Various collision risk models have been developed to achieve design acceptance criteria. In general,
the occurrence of a collision is separated into four events: (1) a vessel approaching the bridge becomes
aberrant, (2) the aberrant vessel hits a bridge element, (3) the bridge element that is hit fails, and (4) a
protection factor based on bridge location. Collision risk models consider the efects of the vessel traf-
ic, the navigation conditions, the bridge geometry with respect to the waterway, and the bridge ele-
ment strength with respect to the impact loads. hey are commonly expressed in the following form
(AASHTO 2009, 2012):

AF = ( N )( PA )( PG )( PC )( PF ) (4.1)

where AF is the annual frequency of collapse of a bridge element; N is the annual number of vessel tran-
sits (classiied by type, size, and loading condition) that can strike a bridge element; PA is the probability
Vessel Collision Design of Bridges 99

of vessel aberrancy; PG is the geometric probability of a collision between an aberrant vessel and a
bridge pier or span; PC is the probability of bridge collapse due to a collision with an aberrant vessel; and
PF is an adjustment factor to account for potential protection of the piers.

4.5.2.2 Vessel Traffic Distribution, N

he number of vessels, N, passing the bridge based on size, type, and loading condition and available
water depth has to be developed for each pier and span component to be evaluated. All vessels of a
given type and loading condition have to be divided into discrete groupings of vessel size by DWT to
determine the contribution of each group to the annual frequency of bridge element collapse. Once
the vessels are grouped and their frequency distribution is established, information on typical vessel
characteristics may be obtained from site speciic data or from published general data such as AASHTO
(2009) and Larsen (1993).

4.5.2.3 Probability of Aberrancy, PA

he probability of vessel aberrancy relects the likelihood that a vessel is out of control in the vicinity of a
bridge. Loss of control may occur as a result of pilot error, mechanical failure, or adverse environmental
conditions. he PA is mainly related to the navigation conditions at the bridge site. Vessel traic regula-
tions, vessel traic management systems, and aids to navigation can improve the navigation conditions
and reduce the PA.
he probability of vessel aberrancy may be evaluated based on site-speciic information that includes
historical data on vessel collisions, rammings and groundings in the waterway, vessel traic, navigation
conditions and bridge/waterway geometry. his has been done for various bridge design provisions and
speciic bridge projects worldwide (IABSE 1983; AASHTO 2009; Larsen 1993). he PA values deter-
mined range from 0.5 × 10−4 to over 7.0 × 10−4.
As an alternative, the AASHTO provisions (AASHTO 2009, 2012) recommend base rates for the
probability of vessel aberrancy that are multiplied by correction factors for bridge location relative to
bends in the waterway, currents acting parallel to vessel-transit path, crosscurrents acting perpendicu-
lar to vessel-transit path, and the traic density of vessels using the waterway. he recommended base
rates are 0.6 × 10−4 for ships and 1.2 × 10−4 for barges.

4.5.2.4 Geometric Probability, PG

he geometric probability is the probability that a vessel will hit a particular bridge pier given that it
has lost control (i.e., is aberrant) in the vicinity of the bridge. It is mainly a function of the geometry
of the bridge in relation to the waterway. Other factors that can afect the likelihood that an aberrant
vessel will strike a bridge element include the original vessel-transit path, course, rudder position,
velocity at the time of failure, vessel type, size, drat and maneuvering characteristics, and the hydrau-
lic and environmental conditions at the bridge site. Various geometric probability models, some based
on simulation studies, have been recommended and used on diferent bridge projects (IABSE 1983;
Modjeski and Masters 1984; Larsen 1993). he AASHTO provisions (AASHTO 2009, 2012) use a nor-
mal probability density function about the centerline of the vessel-transit path for estimating the like-
lihood of an aberrant vessel being within a certain impact zone along the bridge axis. Using a normal
distribution accounts for the fact that aberrant vessels are more likely to pass under the bridge closer
to the navigation channel than further away from it. he standard deviation of the distribution equals
the length of vessel associated with each vessel category or grouping. he probability that an aberrant
vessel is located within a certain zone is the area under the normal probability density function within
that zone (see Figure 4.7).
Bridge elements beyond three times the standard deviation from the centerline of vessel-transit path
are designed for speciied minimum impact load requirements, which are usually associated with an
empty vessel driting with the current.
100 Bridge Engineering Handbook, Second Edition: Substructure Design

Bridge pier

rid e
of b terlin
ge
n
BM = Width of ship

Ce
BP = Width of pier

Ship

Ship
Centerline of vessel
ϕ
sailing path
BM/2 BP/2 BP/2 BM/2

Normal Ship/bridge
distribution impact zone

PG

x
Intersection path
to centerline of pier

FIGURE 4.7 Geometric probability of pier collision.

4.5.2.5 Probability of Collapse, PC

he probability of collapse, PC, is a function of many variables, including vessel size, type, forepeak
ballast and shape, speed, direction of impact, and mass. It is also dependent on the ultimate lateral
load strength of the bridge pier (particularly the local portion of the pier impacted by the bow of the
vessel). Based on collision damages observed from numerous ship–ship collision accidents that have
been correlated to the bridge–ship collision situation (IABSE 1983), an empirical relationship has been
developed based on the ratio of the ultimate pier strength, H, to the vessel impact force, P. As shown in
Figure 4.8, for H/P ratios less than 0.1, PC varies linearly from 0.1 at H/P = 0.1, to 1.0 at H/P = 0.0. For
H/P ratios greater than 0.1, PC varies linearly from 0.1 at H/P = 0.1, to 0.0 at H/P = 1.0.

4.5.2.6 Protection Factor, PF

he protection factor is an adjustment to AF to account for full or partial protection of selected bridge
piers against vessel collisions due to protection measures (dolphins, islands, etc.), or due to existing site
conditions such as a parallel bridge protecting a bridge from impacts in one direction, or a feature of
the waterway (such as a peninsula extending out on one side of the bridge) that may block vessels from
hitting bridge piers, or a wharf structure near the bridge that may block vessels from a certain direction.
PF is computed as

PF = 1– ( % Protection Provided /100 ) (4.2)

If no protection of the pier exists, then PF = 1.0. If the pier is 100% protected, then PF = 0.0. As an
example, if dolphin pier protection system provided 70% protection, then PF would be equal to 0.3.
Values for PF may vary from pier to pier and may vary depending on the direction of the vessel traic
(i.e., vessel traic moving inbound versus traic moving outbound) (AASHTO 2009).
Vessel Collision Design of Bridges 101

1.0

Probability of collapse (PC)

0.5

0.1
0.1 0.5 1.0
Ultimate bridge element strength H
Vessel impact force P

FIGURE 4.8 Probability of collapse distribution.

4.6 Vessel Impact Loads

4.6.1 Ship Impact
he estimation of the load on a bridge pier during a ship collision is a very complex problem. he actual
force is time dependent and varies depending on the type, size, and construction of the vessel; its veloc-
ity; the degree of water ballast in the forepeak of the bow; the geometry of the collision; and the geom-
etry and strength characteristics of the bridge. here is a very large scatter among the collision force
values recommended in various vessel collision guidelines or used in various bridge projects.
Ship collision forces are commonly applied as equivalent static loads. Procedures for evaluating
dynamic efects when the vessel force indentation behavior is known are included in IABSE (1983),
Modjeski and Masters (1984), Larsen (1998), Prucz and Conway (1987, 1989), Grob and Hajdin (1996).
he AASHTO provisions (AASHTO 2009, 2012) use the following formula for estimating the static
head-on ship collision force, PS, on a rigid pier:

1
Ps = 0.98( DWT ) 2 (V 16 ) (4.3)

where PS is the equivalent static vessel impact force (MN); DWT is the ship deadweight tonnage in
tons; and V is the vessel impact velocity in knots (see Figure 4.9). his formulation was primarily devel-
oped from research conducted by Woisin in West Germany during 1967–1976 on physical ship models
to generate data for protecting the reactors of nuclear power ships from collisions with other ships.
A  schematic representation of a typical impact force time history is shown in Figure 4.10 based on
Woisin’s test data. he scatter in the results of these tests is of the order of ±50%. he formula recom-
mended (Equation 4.3) uses a 70% fractile of an assumed triangular distribution with zero values at 0%
and 100% and a maximum value at the 50% level (see Figure 4.11).
Formulas for computing design ship collision loads on a bridge superstructure are given in the
AASHTO provisions (AASHTO 2009, 2012) as a function of the design ship impact force, PS, as follows:
• Ship bow impact force, P BH:

PBH = ( RBH ) ( PS ) (4.4)

where R BH is a reduction coeicient equal to the ratio of exposed superstructure depth to the total
bow depth.
102 Bridge Engineering Handbook, Second Edition: Substructure Design

350 80

T
DW
Ship size

00
70

0,0

T
300

DW
16

00
60

Ship impact force (1,000 kips)

0,0
250

10
Ship impact force (MN)
T
50 D W
200 ,0 00
60 T
40 D W
,0 00
150 40
30 T
DW
0 00
20,
100 WT
20 00 D
10,0
50 T
10 2,000 DW

0 0
0 2 4 6 8 10 12 14 16
Impact speed (knots)

FIGURE 4.9 Ship impact force.

Pmax = Maximum impact force

Impact force (P)

P(t) = Average impact force

0 Time (t)

FIGURE 4.10 Typical ship impact force time history by Woisin.

22
At 70% fractile
20
Ps = 0.98 DWT
18 100%
70%
Impact force (MN)

16
14
12
10 50%
8 0% P(t) = 0.88 DWT
6
4 Probability density
2 function
0
0 50 100 150 200 250 300
Ship size (1,000 DWT)

FIGURE 4.11 Probability density function of ship impact force.

Vessel Collision Design of Bridges 103

• Ship deck house impact force, P DH:

PDH = ( RDH ) ( PS ) (4.5)

where R DH is a reduction coeicient equal to 0.10 for ship larger than 100,000 DWT, and,
DWT
0.2 − ( 0.10) for ships under 100,000 DWT.
100,000
• Ship mast impact force, PMT:

PMT = 0.10 PDH

(4.6)
where P DH is the ship deck house impact force.
he magnitude of the impact loads computed for ship bow and deck house collisions are quite high
relative to the strength of most bridge superstructure designs. Also, there is great uncertainty associated
with predicting ship collision loads on superstructures because of the limited data available and the
ship/superstructure load interaction efects. It is therefore suggested that superstructures, and also weak
or slender parts of the substructure, be located out of the reach of a ship’s hull or bow.

4.6.2 Barge Impact

he barge collision loads recommended by AASHTO for the design of piers are shown in Figure 4.12 as
a function of the tow length and the impact speed. Numerical formulations for deriving these relation-
ships may be found in AASHTO (2009, 2012).
he loads in Figure 4.12 were computed using a standard 59.5 × 10.7 m hopper barge. In previous
AASHTO Guide Speciication (AASHTO 2009) and AASHTO LRFD Bridge Design Speciications
(AASHTO 2012), the impact force recommended for barges larger than the standard hopper barge was
determined by increasing the standard barge impact force by a factor related to the ratio of the width of
the wider barge to the width of the standard hopper barge. his approach, although not directly related

5
Tow length
s
ge
ar

20
4b

es
rg
ba
Barge impact force (1,000 kips)

4
3

es
Barge impact force (MN)

a rg
15 2b
3
e
arg
1b
10
2

5
1

0 0
0 2 4 6 8 10 12
Impact speed (knots)

FIGURE 4.12 Barge impact force.

104 Bridge Engineering Handbook, Second Edition: Substructure Design

to the strength of the barge at the point of impact, accounted for the increased likelihood of higher col-
lision loads being associated with larger barges due to other reasons such as deeper bow rake head logs
and stronger structures at push knees and corner locations (Figures 4.5 and 4.6). It is recommended that
the bridge designer evaluate the vessel traic characteristics at the bridge, determine the likelihood of
barges with deeper bows, and use the ratio between the height of the deeper head log and the head log of
the standard hopper barge to increase the standard barge impact force where needed.

4.6.3 Application of Impact Forces

Collision forces on bridge substructures are commonly applied as follows:
• One hundred percent of the design impact force in a direction parallel to the navigation channel
(i.e., head-on).
• Fity percent of the design impact force in the direction normal to the channel (but not simultane-
ous with the head-on force).
• For overall stability, the design impact force is applied as a concentrated force at the mean high
water level.
• For local collision forces, the design impact force is applied as a vertical line load equally distrib-
uted along the ship’s bow depth for ships and along head block depth for barges.
• For superstructure design, the impact forces are applied transversely to the superstructure com-
ponent in a direction parallel to the navigation channel.
When determining the bridge components exposed to physical contact by any portion of the hull or
bow of the vessel considered, the bow overhang, rake, or lair distance of vessels have to be taken into
account. he bow overhang of ships and barges is particularly dangerous for bridge columns and for
movable bridges with relatively small navigation clearances.

4.6.4 Minimum Impact Requirements

AASHTO speciications (AASHTO 2009, 2012) require that all bridge piers located in design water
depths of more than 0.6 m be designed for a minimum impact force associated with an empty hop-
per barge driting at a speed equal to the mean yearly current in the waterway. Owing to the high
frequency of occurrence of barge breakaways resulting in bridge hits during high river stage periods
and the involvement of loaded barges in these incidents, it is recommended that loaded barge scenario
also be considered. Barges can break away from docks and mooring facilities, and they can also break
away from a barge tow in transit that grounded or hit another bridge. A recent study initiated by the
Oklahoma Department of Transportation (Modjeski and Masters 2009) conirmed the need to properly
assess the risks at a given bridge site and account for the likelihood of barge breakaways by type, loading
condition, and current conditions.

4.6.5 Recent U.S. Barge Research

Since the AASHTO Guide Speciication’s adoption in 1991 and its use in analysis and design of bridges
for vessel collision, the speciication has spurred various research projects to better understand the
magnitude of the barge collision loads involved and the bridge response. Of particular importance
are the extensive research eforts conducted by the University of Florida (Whitney, Harik, Griin,
and Allen 1996; Brown and Bollmann 1992; Hoit, McVay and Hayes 1996; Florida Bridge Sotware
Institute 2002, 2007; Consolazio, Hendrix et al. 2004a; Consolazio, Lehr et al. 2004b; Consolazio and
Cowan 2005; Consolazio, Cook, and McVay 2006) and the University of Kentucky (Yuan, Harik, and
Davidson 2008; Yuan and Harik 2008, 2009,2010). he research by these institutions relect the impor-
tance of using dynamic analysis to estimate barge impact forces.
Vessel Collision Design of Bridges 105

A key portion of the research program conducted by the University of Florida involved the use of a full-
scale barge impact testing on several bridge piers of the St. George Island Bridge across Florida’s Apalachicola
Bay (Consolazio, Cook, and McVay 2006). he existing bridge was being replaced by a new bridge; hence,
two of the abandoned bridge piers (a channel pier with a relatively massive mudline foundation and an
approach pier with two waterline footings) were studied in three diferent structural conigurations in a
full-scale test program, which included ramming a small 600-ton barge against the piers at various speeds
(some with the superstructure in place and others with the superstructure removed) and measuring a wide
variety of responses in the structure and soil using extensive measurement and recording systems.
Based on the University of Florida test data from the St. George Bridge program, several general
observations can be made in comparing the measured barge impact forces with those predicted by the
AASHTO equations. For relatively stif piers with below mudline pile supported footings, the mea-
sures impact forces ranged from 50% to 100% of the AASHTO force (with most measurements near the
50% level). For relatively lexible pile-supported piers with the footings at or above the waterline, the
measured impact forces ranged from 100% to 130% of the AASHTO forces (with most measurements
near the 130% level). he test results indicate that the dynamic response of the structure and the stifness
of the underlying soil are key components in the development of the barge impact force transmitted to
the pier. he University of Florida barge test data also indicated that there are diferences in load efects
(e.g., displacements, shears, moments) between the application of the AASHTO static loads versus the
dynamic loads of the test data. Nevertheless, the study indicated that even though there were diferences
in the measured forces versus AASHTO, the static analysis performed using the AASHTO loads appear
to yield foundation design forces that are consistent with results obtained by more reined analysis tech-
niques (e.g., dynamic analysis combined with experimentally measured dynamic loads).
he University of Kentucky conducted analytical studies on multibarge tow impact forces (Yuan, Harik,
and Davidson 2008) and concluded that counting the barges in the length of the tow may yield conserva-
tive impact forces using the AASHTO equations, particularly in those cases where the width of the pier
is smaller (approximately 10%) than the width of the barge. Where the width of the pier is about 50% of
the width of the barge, the barge impact forces are close to the AASHTO values, and where the width of
the pier is about the same or greater than the width of the barge, the AASHTO forces are less than those
computed using inite element models and dynamic analysis. heir research indicates that an “accordion”-
type efect occurs where the barges in the tow length buckle upward/downward, which reduces the impact
energy being transferred to the pier. he study also indicated that the barges in the width of the tow do not
simply break away on impact, but stay connected suiciently to afect the collision energy. Interestingly,
dynamic inite element analysis applied to the entire tow resulted in overall impact forces very similar to
the forces derived using the AASHTO method (i.e., the reduction of the forces by buckling in the length
of the tow is ofset by an increase in force due to the inluence of adjacent barges in the width of the tow).

4.7 Bridge Analysis and Design

Vessel collisions are extreme events with a very low probability of occurrence; therefore, the limit state
considered is usually structural survival. Depending on the importance of the bridge various degrees
of damage are allowed—provided that the structure maintains its integrity, hazards to traic are mini-
mized, and repairs could be made in a relatively short period of time. When the design is based on more
frequent but less severe collisions, structural damage and traic interruptions are not allowed.
Designing for vessel collision is commonly based on equivalent static loads that include global forces
for checking overall capacity of piers and local forces for checking local strength of bridge components.
he contribution of the superstructure to transfer loads to adjacent piers may also be considered. A clear
load path from the location of the vessel impact to the bridge foundation needs to be established, and
the components and connections within the load path must be adequately designed and detailed. he
design of individual bridge components is based on strength and stability criteria. Overall stability,
redundancy, and ductility are important criteria for structural survival.
106 Bridge Engineering Handbook, Second Edition: Substructure Design

4.7.1 Global Pier Capacity

he global pier capacity is determined in terms of a concentrated collision load applied at the design
water elevation, which is commonly the mean high water elevation. It is determined based on the ulti-
mate strength and stability of the pier and its foundation in the Extreme Event II Limit State load com-
bination as deined in AASHTO LRFD (AASHTO 2012). Strength or service limit states may need to be
considered as well, depending on the performance criteria requirements.
he modeling of pile foundations could vary from the simple assumption of a point of ixity to non-
linear soil–structure interaction models, depending on the limit state considered and the sensitivity of
the response to the soil conditions. Lateral load capacity analysis methods for pile groups that include
nonlinear behavior can be found in Kuzmanovic and Sanchez (1992) and Brown and Bollmann (1992)
and the basic features of a inite element analysis computer program developed for bridge piers com-
posed of pier columns and cap supported on a pile cap and nonlinear piles and soil are described in Hoit,
McVay, and Hays (1996). he most recent version of the program, FB-PIER, is commercially available
from the Florida Bridge Sotware Institute. If analysis indicates that piles will be loaded in tension by
vessel impact forces, the design engineer must determine that the piles and their connection to the foot-
ing or cap have adequate pullout resistance.
Transient foundation uplit or rocking involving separation from the subsoil of an end-bearing foun-
dation pile group or the contact area of a foundation footing could be allowed under impact loading
provided suicient consideration is given to the structural stability of the substructure.
Guidelines for the design, detailing, and construction of concrete and steel elements and connections
are included in AASHTO LRFD (AASHTO 2012). Adequate transverse reinforcement, spacing, and
splices must be provided if plastic hinging is allowed to form.

4.7.2 Local Pier Capacity

he local pier capacity is assessed in relation to distributed collision loads in the zone of impact to pre-
vent premature localized failures of the more slender bridge components that can still be reached by ves-
sels hull. Local moment, shear, and shear friction capacity must be checked, and special attention must
be given to detailing in the zone of impact. Concrete will tend to spall upon impact, and it is therefore
important to provide increase concrete cover, closely spaced transverse reinforcement, and to avoid lap
splicing of longitudinal and transverse bars in the zone of impact.

4.7.3 Contribution of the Superstructure

he contribution of the superstructure to the transfer of loads to adjacent substructure units depends
on the capacity of the connection of the superstructure to substructure and the relative stifness of the
substructure at the location of the impact. However, in order to consider partial transfer of lateral forces
to the superstructure, positive steel or concrete connections of superstructure to substructure such as
shear keys must be provided. Similarly, for partial transfer to the superstructure of the longitudinal
component of the impact force, the shear capacity of the bearings must be adequate. When elastomeric
bearings are used their longitudinal lexibility may be added to the longitudinal lexibility of the piers.
If the ultimate capacity of the bearings is exceeded, then the pier must take the total longitudinal force
and be treated as a cantilever.
Simpliied guidelines for determining the distribution of collision loads to adjacent piers are included
in Kuzmanovic and Sanchez (1992). To ind out how much of the transverse impact force is taken by
the pier and how much is transferred to the superstructure, two analytical models may be used. One is
a two-dimensional or a three-dimensional model of the complete pier, and the other is a two-dimen-
sional model of the superstructure projected on a horizontal plane. he projected superstructure may
be modeled as a beam with the moment of inertia referred to a vertical axis through the center of the
Vessel Collision Design of Bridges 107

roadway and with hinges at expansion joint locations. he beam is supported at pier locations by elastic
horizontal springs representing the lexibility of each pier. he lexibility of the piers is obtained from
pier models using virtual forces. he superstructure model is loaded with a transverse virtual force
acting at the place where the pier under consideration is located. he spring in the model at that place
is omitted to obtain a lexibility coeicient of the superstructure at the location of the top of pier under
consideration. hus, the horizontal displacement of the top of pier due to the impact force on the pier
(usually applied at mean high water level) is equal to the true displacement of the superstructure due to
the transmitted part of the impact force. he magnitude of the force transmitted to the superstructure is
obtained by equating the total true displacement of the top of pier from the pier model to the displace-
ment of the superstructure.
he superstructure contribution analysis can also be done modeling the entire bridge within a gen-
eral purpose structural analysis program or more eiciently using a special purpose program such as
FB-MULTIPIER available from the Florida Bridge Sotware Institute that can include soil–structure
interaction and superstructure participation in one model and was developed by the University of
Florida speciically for vessel collision analysis. his program can also perform dynamic analysis of
barge impact.

4.7.4 Movable Bridges

Movable bridges are particularly susceptible to interrupted service as a result of vessel collision because
even minor impact can cause mechanical equipment to jam or fail. Guidelines for the design and pro-
tection of movable bridges are included in the AASHTO Guide Speciication (AASHTO 2009) and the
AASHTO LRFD Movable Highway Bridge Design Speciications (AASHTO 2000).
In addition to the design vessels determined based on the AASHTO LRFD and Guide Speciication
criteria, the AASHTO LRFD Movable Highway Bridge Design Speciications also include an operating
vessel used to minimized damage from routine marine traic and ensure that the bridge remains opera-
tional and to help proportion the fender system so that it is not severely damaged ater minor collisions.
he AASHTO LRFD Movable Highway Bridge Design Speciications also include movable bridge spe-
ciic analysis considerations and design and detailing guidelines.

4.8 Bridge Protection Measures

he cost associated with protecting a bridge from catastrophic vessel collision can be a signiicant por-
tion of the total bridge cost, and must be included as one of the key planning elements in establishing
a bridge’s type, location, and geometry. he following alternatives are usually evaluated in order to
develop a cost-efective solution for a new bridge project:
• Design the bridge piers, foundations, and superstructure to directly withstand the vessel collision
forces and impact energies.
• Design a pier fender system to reduce the impact loads to a level below the capacity of the pier and
foundation.
• Increase span lengths and locate piers in shallow water out-of-reach from large vessels in order to
reduce the impact design loads.
• Protect piers from vessel collision by means of physical protection systems.

4.8.1 Physical Protection Systems

Piers exposed to vessel collision can be protected by special structures designed to absorb the impact
loads (forces or energies) or redirect the aberrant vessel away from the pier. Because of the large forces
and energies involved in a vessel collision, protection structures are usually designed for plastic
108 Bridge Engineering Handbook, Second Edition: Substructure Design

deformation under impact (i.e., they are essentially destroyed during the head-on design collision and
must be replaced). General types of physical protection systems include the following:

Fender systems. hese usually consist of timber, rubber, steel, or concrete elements attached to a
pier to fully, or partially, absorb vessel impact loads. he load and energy absorbing character-
istics of such fenders is relatively low compared to typical vessel impact design loads.
Pile-supported systems. hese usually consist of pile groups connected by either lexible or rigid
caps to absorb vessel impact forces. he piles may be vertical (plumb) or battered depending
on the design approach followed and may incorporate relatively large diameter steel pipe or
concrete pile sizes. he pile supported protection structure may be either free standing away
from the pier or attached to the pier itself. Fender systems may be attached to the pile structure
to help resist a portion of the impact loads.
Dolphin protection systems. hese usually consist of large diameter circular cells constructed of
driven steel sheet piles, illed with rock or sand, and topped by a thick concrete cap. Vessel col-
lision loads are absorbed by rotation and lateral deformation of the cell during impact.
Island protection systems. hese usually consist of protective islands built of a sand or quarry-
run rock core and protected by outer layers of heavy rock rip-rap for wave, current, and ice
protection. he island geometry is developed to stop an aberrant vessel from hitting a pier
by forcing it to run aground. Although extremely efective as protection systems, islands are
oten diicult to use due to adverse environmental impacts on river bottoms (dredge and ill
permits) and river currents (increase due to blockage), as well as impacts due to settlement and
downdrag forces on the bridge piers.
Floating protection systems. hese usually consist of cable net systems suspended across the water-
way to engage and capture the bow of an aberrant vessel, or loating pontoons anchored in
front of the piers. Floating protection systems have a number of serious drawbacks (environ-
mental, efectiveness, maintenance, cost, etc.) and are usually only considered for extremely
deep water situations where other protection options are not practicable.

he AASHTO Guide Speciication (AASHTO 2009) provides examples and contains a relatively
extensive discussion of various types of physical protection systems such as fenders, pile supported
structures, dolphins, protective islands, and loating structures. However, the guide does not include
speciic procedures and recommendations on the actual design of such protection structures. Further
research is needed to establish consistent analysis and design methodologies for protection structures,
particularly because these structures undergo large plastic deformations during the collision.

4.8.2 Aids to Navigation Alternatives

As 60%–85% of all vessel collisions are caused by pilot error, it is important that all aspects of the bridge
design, siting, and aids to navigation with respect to the navigation channel be carefully evaluated with
the purpose of improving or maintaining safe navigation in the waterway near the bridge. Traditional
aids include buoys, range markers, navigation lighting, and radar relectors as well as standard operat-
ing procedures and regulations speciically developed for the waterway by government agencies and
pilot associations. Modern aids include advanced vessel traic control systems (VTS) using shore-based
radar surveillance and radio-telephone communication systems; special electronic transmitters known
as Racon devices mounted to bridge spans for improved radar images indicating the centerline of chan-
nel; and advanced navigation positioning systems based on ship-board global positioning satellite (GPS)
systems and electronic charts. It should be noted that bridge designers are very limited in their ability
to require any modiications that afect operations on a navigable waterway since the responsibility and
authority for implementing such navigation improvements within U.S. waterways belongs to the U.S.
Coast Guard and is protected under Federal Regulations.
Vessel Collision Design of Bridges 109

Following the terrorist attacks upon the United States on September 11, 2001, the Coast Guard has
required that all foreign ships entering the U.S. waterway system to be equipped with various advanced
electronic navigation aids and tracking systems. hese requirements however do not extend to domestic
barge tows on the inland waterway system. It is believed that the use of such advanced electronic navi-
gation systems should also reduce the risk of vessel collision with bridges by providing pilots and vessel
operators with accurate location information. At present, no studies have been performed to analyze
and document the potential reduction in PA due to such electronic aids-to-navigation. If a case can be
made at a particular waterway and bridge site that improved electronic navigation aids would reduce
the PA, then such a factor could be used in the risk analysis—provided it is approved by the owner
(AASHTO 2009).
It should be noted that the traditional isolation of the maritime community must come to an end. In
addition to the bridge costs, motorist inconvenience, and loss of life associated with a catastrophic vessel
collision, signiicant environmental damage can also occur due to spilled hazardous or noxious cargoes
in the waterway. he days when the primary losses associated with an accident rested with the vessel
and her crew are over. he $13 million value of the M/V Summit Venture was far below the $250 million
replacement cost of the Sunshine Skyway Bridge that the vessel destroyed. he losses associated with
the 11 million gallons of crude oil spilled from the M/V Exxon Valdez accident of the coast of Alaska in
1989 are over $3.5 billion. Both of these accidents could probably have been prevented using advanced
electronic navigation systems.

4.8.3 Motorist and Vessel Operator Warning Systems

Motorist warning systems may be used on bridges to minimize the loss of life, which may occur in the
event of a bridge span collapse due to a vessel collision. hese include the following:
• Hazard detection systems, such as ship impact vibration detectors, continuity circuits, and VHF
radio link
• Veriication devices, such as closed circuit television (CCTV), visual delineation devices, and
motorist call boxes
• Traic control and information devices, such as variable message sign, lashing beacons, and
movable gates
Vessel operator warning systems include nonmovement detectors in the vessel operator’s house to
warn if the vessel operator fell asleep or became incapacitated.

4.9 Summary
Experience to date has shown that the use of the vessel impact and bridge protection requirements
such as the AASHTO speciications (AASHTO 2009, 2012) for planning and design of new bridges has
resulted in a signiicant change in proposed structure types over navigable waterways. Incorporation of
the risk of vessel collision and cost of protection in the total bridge cost has almost always resulted in
longer span bridges being more economical than traditional shorter span structures, because the design
goal for developing the bridge pier and span layout is the least cost of the total structure (including the
protection costs). Typical costs for incorporating vessel collision and protection issues in the planning
stages of a new bridge have ranged from 5% to 50% of the basic structure cost without protection.
Experience has also shown that it is less expensive to include the cost of protection in the planning
stages of a proposed bridge than to add it ater the basic span coniguration has been established without
considering vessel collision concerns. Typical costs for adding protection, or for retroitting an existing
bridge for vessel collision, have ranged from 25% to more than 100% of the existing bridge costs.
It is recognized that vessel collision is but one of a multitude of factors involved in the planning pro-
cess for a new bridge. he designer must balance various needs including political, social, and economic
110 Bridge Engineering Handbook, Second Edition: Substructure Design

in arriving at an optimal bridge solution for a proposed highway crossing. Because of the relatively
high bridge costs associated with vessel collision design for most waterway crossings, it is important
that additional research be conducted to improve our understanding of vessel impact mechanics, the
response of the structure, and the development of cost-efective protection systems.

References
AASHTO. 1991. Guide Speciication and Commentary for Vessel Collision Design of Highway Bridges,
American Association of State Highway and Transportation Oicials, Washington, D.C.
AASHTO. 2000. LRFD Movable Highway Bridge Design Speciications, 1st ed., American Association of
State Highway and Transportation Oicials, Washington, D.C.
AASHTO. 2009. Guide Speciication and Commentary for Vessel Collision Design of Highway Bridges,
2nd ed., American Association of State Highway and Transportation Oicials, Washington, D.C.
AASHTO. 2012. AASHTO LRFD Bridge Design Speciication, Customary U.S. Units, 6th ed., American
Association of State Highway and Transportation Oicials, Washington, D.C.
AREMA. 2013. Manual for Railway Engineering, Part 23, American Railway Engineering and Maintenance-
of-Way Association, Lanham, MD.
Brown, D. A. and Bollmann, H. T. 1992. “Pile Supported Bridge Foundations Designed for Impact
Loading”, Transportation Research Record 1331, TRB, National Research Council, Washington, D.C.,
pp. 87–91.
Consolazio, G. R. and Cowan, D. R. 2005. “Numerically Eicient Dynamic Analysis of Barge Collisions
with Bridge Piers”, ASCE Journal of Structural Engineering, ASCE, Vol. 131, No. 8, New York, NY,
pp. 1256–1266.
Consolazio, G. R., Cook, R. A., and McVay, M. C. 2006. “Barge Impact Testing of the St. George Island
Causeway Bridge-Phase III; Physical Testing and Data Interpretation”, Structures Research
Report No. 2006/26868, Engineering and Industrial Experiment Station, University of Florida,
Gainesville, FL.
Consolazio, G. R., Hendix, J. L., McVay, M. C., Williams, M. E., and Bollman, H. T. 2004a. “Prediction
of Pier Response to Barge Impacts Using Design-Oriented Dynamic Finite Element Analysis”,
Transportation Research Record 1868, Transportation Research Board, Washington, D.C., pp.
177–189.
Consolazio, G. R., Lehr, G. B., and McVay, M. C. 2004b. “Dynamic Finite Element Analysis of Vessel-Pier-
Soil Interaction During Barge Impact Events”, Transportation Research Record 1849, Transportation
Research Board, Washington, D.C., pp. 81–90.
Florida Bridge Sotware Institute. 2002. FB-PIER Users’ Manual, University of Florida, Gainesville, FL.
Florida Bridge Sotware Institute. 2007. FB-MULTIPIER Users’ Manual, University of Florida,
Gainesville, FL.
Grob, B. and Hajdin, N. 1996. “Ship Impact on Inland Waterways”, Structural Engineering International,
Vol. 4, IABSE, Zürich, Switzerland, pp. 230–235.
Hoit, M., McVay, M., and Hays, C. 1996. “Florida Pier Computer Program for Bridge Substructure
Analysis: Models and Methods”, Conference Proceedings, Design of Bridges for Extreme Events,
FHWA, Washington, D.C.
IABSE. 1983. Ship Collision with Bridges and Ofshore Structures, International Association for Bridge and
Structural Engineering, Colloquium Proceedings, Copenhagen, Denmark, 3 Volumes (Introductory,
Preliminary, and Final Reports).
Kuzmanovic, B. O., and Sanchez, M. R. 1992. “Design of Bridge Pier Pile Foundations for Ship Impact”,
Journal of Structural Engineering, ASCE, Vol. 118, No. 8, pp. 2151–2167.
Larsen, A. and Esdahl, S., eds. 1998. Proceedings of International Symposium on Advances in Bridge
Aerodynamics, Ship Collision Analysis, and Operation & Maintenance, Copenhagen, Denmark,
May 10–13, 1998, Balkema Publishers, Rotterdam, Netherlands.
Vessel Collision Design of Bridges 111

Larsen, O. D. 1993. Ship Collision with Bridges: he Interaction between Vessel Traic and Bridge Structures,
IABSE Structural Engineering Document 4, IABSE-AIPC-IVBH, Zürich, Switzerland.
Modjeski and Masters. 1984. “Criteria for: he Design of Bridge Piers with Respect to Vessel Collision in
Louisiana Waterways”, Report prepared for Louisiana Department of Transportation and Development
and the Federal Highway Administration, Mechanicsburg, PA.
Modjeski and Masters. 2009. “Vessel Collision Risk Assessment of Bridges Over he McClellan-Kerr
Arkansas River Navigation System With Respect to Loaded Runaway Barges”, Report prepared for
Oklahoma Department of Transportation, Mechanicsburg, PA.
National Research Council. 1983. Ship Collisions with Bridges - he Nature of the Accidents, their Prevention
and Mitigation, National Academy Press, Washington, D.C.
Prucz, Z. and Conway, W. B. 1987. “Design of Bridge Piers Against Ship Collision”, Bridges and Transmission
Line Structures, (Edited by L. Tall), ASCE, New York, NY, pp. 209–223.
Prucz, Z. and Conway, W. B. 1989. “Ship Collision with Bridge Piers-Dynamic Efects”, Transportation
Research Board Paper 890712, Transportation Research Board, Washington, D.C.
U.S. Coast Guard. 2003. American Waterways Operators Bridge Allision Work Group. Report of the U.S.
Coast Guard-American Waterways Operators, Inc. Safety Partnership, Washington, D.C.
Whitney, M. W., Harik, I. E., Griin, J. J., and Allen, D. L. 1996. “Barge Collision Design of Highway
Bridges”, Journal of Bridge Engineering, ASCE, Vol. 1, No. 2, pp. 47–58.
Yuan, P. and Harik, I. E. 2008, “One-Dimensional Model for Multi-Barge Flotillas Impacting Bridge Piers”,
Computer-Aided Civil and Infrastructure Engineering, Vol. 23, pp. 437–447.
Yuan, P., and Harik, I. E. 2008. “Equivalent Barge and Flotilla Impact Forces on Bridge Piers”, Kentucky
Transportation Center research report No. KTC-08-12/SPR261-03-1F, University of Kentucky,
Lexington, KY.
Yuan, P., and Harik, I. E. 2010. “Equivalent Barge and Flotilla Impact Forces on Bridge Piers”, ASCE Journal
of Bridge Engineering, ASCE, Vol. 15, No. 5, New York, NY, pp. 523–532.
Yuan, P., Harik, I. E., and Davidson, M. T. 2008. “Multi-Barge Flotilla Impact Forces on Bridges”, Kentucky
Transportation Center research report No. KTC-08-13/SPR261-03-2F, University of Kentucky,
Lexington, KY.
 5
Bridge Scour Design
and Protection*
5.1 Introduction ......................................................................................113
5.2 Hydrology and Hydraulics ..............................................................113
Hydrology • Bridge Deck Drainage Design • Stage Hydraulics
5.3 Scour Design and Protection ..........................................................121
Scour Analysis • Scour Calculation • Pressure Flow
Scour from Model Tests • Bridge Scour Investigation and
Junke Guo Prevention • Introduction to Bridge Scour Inspection • Real-Time
University of Monitoring • Scour Protection
Nebraska–Lincoln References......................................................................................................132

5.1 Introduction
his chapter presents basic concepts, methods, and procedures in bridge scour design and protection,
including hydrology study, hydraulic analysis, scour evaluation, and scour protection.
Hydrology study is to determine design discharge, either the peak discharge or the lood hydrograph
(in some cases both) at the highway stream crossings. Hydraulic analysis is to convert design discharge
to hydraulic variables such as velocity, low depth, and bed shear stress eroding bed materials around
bridge piers and abutments. Scour design is to evaluate the maximum possible scour depth correspond-
ing to design discharge. Scour protection provides counter measurements resisting scour process. Below
the state-of-the-practice in bridge design and protection is outlined.

5.2 Hydrology and Hydraulics

5.2.1 Hydrology
5.2.1.1 Data Collection
Hydrology data are fundamental in bridge design, obtained from the following sources: as built plans,
site investigations and ield studies, bridge maintenance books, hydraulic iles from experienced report
writers, iles of government agencies such as U.S. Corps of Engineers, USGS, Soil Conservation Service
(SCS) and FEMA, rainfall data from local water agencies, stream gage data, USGS & State water agency
reservoir regulation, aerial photos, and loodways.
Site investigations are conducted except in simple cases. Field studies are important because they
reveal conditions that are not readily apparent from maps, aerial photographs, and previous studies. he
typical data collection during a ield study includes high water (HW) marks, scour potential and stream

* his chapter was updated based on Chapter 61 “Bridge Hydraulics” in the irst edition by Jim Springer and Ke Zhou.

113
114 Bridge Engineering Handbook, Second Edition: Substructure Design

stability nearby drainage structures, changes in land use not indicated on the maps, and debris potential
nearby physical features. See FHWA (1984) for a typical Survey Data Report Form.

5.2.1.2 Drainage Basin

he area of drainage basin above a given point along a stream is a major contributing factor to the quan-
tity of low past that point. For the given conditions, the peak low at the proposed site is approximately
proportional to the drainage area.
he basin shape afects the peak discharge; long narrow basins give lower peak discharges than pear-
shaped basins. he basin slope afects the concentration time; steep slope decreases but latter slope
increases the concentration time. he mean elevation of a drainage basin afects runof; higher eleva-
tion basins receive a signiicant amount of precipitation as snow. A basin orientation with respect to
the direction of storm movement afects peak discharge; storms moving upstream produce lower peaks
than those moving downstream.

5.2.1.3 Discharge
Several methods are used for determining discharge. Most of them are based on statistical analyses of
rainfall and runof records, involving preliminary or trial selections of alternative plans that are judged
to it the site conditions and to accommodate the lood lows selected for analysis.
Flood low frequencies are usually calculated, through the overtopping lood, for discharges of
2.33 years that is considered the “Mean Annual Discharge.” he base lood is the 100-year discharge
(1% frequency). he design discharge is the 50-year discharge (2% frequency) or the greatest of record,
if practical. Many times, the historical lood is so large that a structure to manage the low becomes
uneconomical. In such a case, engineering judgment is needed. he overtopping discharge is calculated
on the site, but may overtop the roadway some distance away from the site.
Changes in land use alter the surface runof so that future land use changes during the bridge life
should be considered in the ield. he surface soil type afects the peak discharge calculation. Rock for-
mations underlying the surface and other geophysical characteristics such as volcanic, glacial, and river
deposits have a signiicant efect on runof. In the United States, the major source of soil information is
the SCS. Detention storage reduces the basin peak discharge by its size and location.
he most commonly used methods to determine discharges are (1) rational method, (2) statistical gage
analysis method, (3) discharge comparison of adjacent basins from gage analysis, (4) regional lood-frequency
equations, and (5) design hydrograph. he results from these methods should be compared, not averaged.

5.2.1.3.1 Rational Method

he rational method was irst employed in Ireland in 1847. his method assumes: (1) drainage area is
smaller than 300 acres; (2) peak low occurs when all of the watershed is contributing; (3) the rainfall
intensity is uniform over a duration equal to or greater than the time of concentration, Tc; and (4) the
frequency of the peak low is equal to the frequency of the rainfall intensity. hese assumptions imply
steady low conditions and the mass conservation law gives
Q = CiA (5.1)
where Q = discharge (t3/s), i = rainfall intensity (in/h) determined from either regional IDF (Intensity
Duration Frequency) maps or individual IDF curves, A = basin area (acres) determined from topo-
graphic map (note: 1 mile2 = 640 acres = 0.386 km2), and C = runof coeicient (%) determined in the
ield and from Tables 5.1 and 5.2 (FHWA 2002) or estimated below for a weighted value if basin is cov-
ered with diferent materials, namely
∑C j Aj
C= (5.2)
∑ Aj
with subscript j for the value in a subarea.
Bridge Scour Design and Protection 115

TABLE 5.1 Runof Coeicients for Developed Areas

Business
Downtown areas 0.70–0.95
Neighborhood areas 0.50–0.70
Residential areas
Single–family areas 0.30–0.50
Multiunits, detached 0.40–0.60
Multiunits, attached 0.60–0.75
Suburban 0.25–0.40
Apartment dwelling areas 0.50–0.70
Industrial
Light areas 0.50–0.80
Heavy areas 0.60–0.90
Parks, cemeteries 0.10–0.25
Playgrounds 0.20–0.40
Railroad yard areas 0.20–0.40
Unimproved areas 0.10–0.30

TABLE 5.2 Runof Coeicients for Undeveloped Area Watershed Types

Soil 0.12–0.16 0.08–0.12 0.06–0.08 0.04–0.06
No efective soil cover, Slow to take up water, Normal, well-drained High, deep sand or other
either rock or thin clay or shallow loam light or medium- soil that takes up water
soil mantle of soils of low textured soils, sandy readily, very light
negligible iniltration iniltration capacity, loams, silt and silt well-drained soils
capacity imperfectly or poorly loams
drained
Vegetal 0.12–0.16 0.08–0.12 0.06–0.08 0.04–0.06
Cover No efective plant Poor to fair; clean Fair to good; about Good to excellent; about
cover, bare or very cultivation crops, or 50% of area in good 90% of drainage area in
sparse cover poor natural cover, grassland or woodland, good grassland,
less than 20% of not more than 50% of woodland, or equivalent
drainage area over area in cultivated crops cover
good cover
Surface 0.10–0.12 0.08–0.10 0.06–0.08 0.04–0.06
Storage Negligible surface Low, well-deined Normal; considerable High; surface storage,
depression few and system of small surface depression high; drainage system
shallow, drainage drainage ways; no storage; lakes and pond not sharply deined; large
ways steep and small, ponds or marshes marshes loodplain storage or
no marshes large number of ponds
or marshes

he concentration time for a pear-shaped drainage basin is determined by the Kirpich equation:

0.77
 L 
TC = 0.0195  0.5  (5.3)
S 

where Tc = concentration time (min), L = horizontally projected length (m) of the watershed, and
S = H/L with H = diference (m) in elevations between the most remote point in the basin and the outlet.
Equation 5.3 combines overland and channel lows.
116 Bridge Engineering Handbook, Second Edition: Substructure Design

5.2.1.3.2 Statistical Gage Analysis Methods

he following two methods are the major statistical analysis methods used with Stream Gage Records:
(1) Log Pearson Type III Method and (2) Gumbel Extreme Value Method.
he use of stream gage records is preferred for estimating discharge/frequencies since they relect
actual climatology and runof. Discharge records are obtained from a state Department of Water
Resources in the United States. A good data set should contain at least 25 years of continuous data.
It is important to review each individual stream gage record carefully to guarantee that the database
is consistent with good statistical analysis practice. For example, a drainage basin with a large storage
facility results in a skewed or inconsistent database since smaller basin discharges are inluenced to a
much greater extent than larger discharges.
he most current published stream gage description page should be reviewed to obtain a complete
idea of the data record. A note should be given to changes in the basin area over time, diversions, revi-
sions, and so on. All reliable historical data outside of the recording period should be included. he
adjacent gage records for supplemental information should be controlled and utilized to extend the
record if possible. Natural runof data should be separated from later controlled data. It is known that
high-altitude basin snow melt discharges are not compatible with rain lood discharges. he zero years
must also be accounted for by adjusting the inal plot positions, not by inclusion as minor lows. he
generalized skew number can be obtained from the chart in USGS (1981).
Quite oten the database requires modiication for use in the Log Pearson III analysis. Occasionally, a
high outlier but most oten low outliers need to be removed from the database to avoid skewing results.
his need is determined for high outliers by using Q H = QH + K.SH, and low outliers by using QL = 
QL + KSL where K is a factor determined by the sample size, QH and QL are the high and low mean loga-
rithm of systematic peaks, Q H and QL are the high and low outlier thresholds in log units, SH and SL are
the high and low standard deviation of the log-distribution. Refer to FHWA (2002) or USGS (1981) for
this method and to ind the values of “K.”
he data plotted are: “PEAK DISCHARGE, Q (CFS)” vs. “PROBABILITY, Pr” as shown in Figure 5.1
that results in a very lat curve with a reasonably straight center portion. An extension of this center
portion gives a line for interpolation of the various needed discharges and frequencies.
he engineer should use an Adjusted Skew, which is calculated from the generalized and station
skews. Generalized skews should be developed from at least 40 stations with each station having at least
25 years of record. he equation for the adjusted skew is

MSEGS (GL ) + MSEGL (GS )

Gw = (5.4)
MSEGS + MSEGL

where Gw = weighted skew coeicient, GS = station skew, GL = generalized skew, MSEGS = mean square
error of station skew, and MSEGL = mean square error of generalized skew.
he entire Log Pearson Type III procedure is found in USGS (1981). he Gumbel Extreme Value
Method is also used to describe the distribution of hydrological variables. For peak discharges, it is
written as

a ( Q −b )
f (Q ) = e − e (5.5)

with a = 1.281/S, b = Q –0.450S, S = standard deviation, and Q = mean annual low.

he characteristics of the Gumbel extreme value distribution are from Equation 5.5. he mean dis-
charge, Q, corresponds to the return period of Tr = 2.33 years and skews toward the high lows or
extreme values as shown in Figure 5.2. Even though it does not account directly for the computed skew
Bridge Scour Design and Protection 117

Probability, Pr
.99 .50 .20 .10 .05 .02 .01

Medina river near San Antonio, Texas

Mean annual flow QL = 3.6392 Computed skew, GL = 0.236

Standard deviation SL = 0.3941

30,000 Probability K KSL QL = QL + KSL Q = 10QL

.01 2.4981 .9845 4.6237 42,042 CFS
20,000 .10 1.3039 .5139 4.1530 14,226 CFS

10,000
Peak discharge, Q (CFS)

5,000

2,000

1,000

6
23
0.
2

=
.25

500 L
,G
–0

Generalized skew, GL = –0.252

ew
=

sk
L
,G

d
te Q = 10QL
ew

Probability K KSL QL = QL + KSL

pu
sk

om .01 2.1411 .8438 4.4830 30,411 CFS

ed

C
.10 1.2514 .4932 4.1323 13,564 CFS
liz
ra

200
ne
Ge

100
1.01 2 5 10 25 50 100
Return period, Tr (YRS)

FIGURE 5.1 Log pearson type III distribution analysis, Medina River, Texas, USA.

of the data, it does predict the high lows reasonably well. Further information about this method is in
FHWA (2002) or USGS (1981). Results from this method should be plotted on a special Gumbel paper,
as shown in Figure 5.2.

5.2.1.3.3 Discharge Comparison of Adjacent Basins

FHWA (1984) contains a list of reports for various states in the United States with discharges at gages
determined for frequencies from 2- to 100-year frequencies. he discharges were determined by the Log
Pearson III method. he discharge–frequency at the gages should be updated by the engineer using Log
Pearson III and the Gumbel Extreme Value method.
118 Bridge Engineering Handbook, Second Edition: Substructure Design

Probability, Pr
.99 .50 .20 .10 .05 .02 .01

Medina river near San Antonio, Texas

Mean annual flow Q = 6,602.5 CFS

Standard deviation S = 7,074.5 CFS

30,000 Probability K KS Q = Q + KS
.01 3.534 25,001 31,604 CFS
Peak discharge, Q (CFS)

.10 1.486 10,512 17,115 CFS

25,000

20,000

15,000

10,000
Q = 6,603 CFS
5,000

0 Tr = 2.33 YRS

1.01 2 5 10 25 50 100
USGS-9-1790 (March 1949)
Return period, Tr(YRS)

FIGURE 5.2 Gumbel extreme value frequency distribution analysis, Medina River, Texas, USA.

he gage data are used directly as equivalent if the drainage areas are about the same (within less than
5%). Otherwise, the discharge is determined by

Qu = Q g ( Au / Ag )b (5.6)

where Qu = discharge at ungaged site, Qg = discharge at gaged site, Au = area of ungaged site, Ag = area
of gaged site, and b = exponent of drainage area.

5.2.1.3.4 Regional Flood-Frequency Equations

If no gaged site is nearby or the record for the gage is short, then the discharge is computed using the
applicable regional lood-frequency equations. Statewide regional regression equations have been estab-
lished in the United States. hese equations permit peak lows to be estimated for return periods varying
between 2 and 100 years, based on the Log Pearson III method (FHWA 2002).

5.2.1.3.5 Design Hydrographs

Design hydrographs (FHWA 2005) give a complete time history of the passage of a lood at a particular
site, including the peak low. A runof hydrograph is a plot of the response of a watershed to a particular
rainfall event. A unit hydrograph is deined as the direct runof hydrograph resulting from a rainfall
event lasting for the unit duration of time. he ordinates of the unit hydrograph are such that the vol-
ume of direct runof represented by the area under the hydrograph is equal to one inch of runof from
the drainage area. Data on low water discharges and dates should be given as it controls methods and
Bridge Scour Design and Protection 119

procedures of pier excavation and construction. he low water discharges and dates are found in the
USGS Water Resources Data Reports published each year, which are determined by reviewing the past
5 or 6 years of records.

5.2.1.4 Remarks
Before arriving at a inal discharge, the existing channel capacity should be checked by the calculated
velocity times the channel waterway area. Note that a portion of the discharge may overlow the banks
and never reaches the site.
he proposed design discharge should also be checked to see if it is reasonable and practicable. As
a rule of thumb, the unit runof should be 300 to 600 t 2/mile2 for small basins (<20 mile2), 100 to
300 t 2/mile2 for median areas (<50 mile2) and 25 to 150 t 2/mile2 for large basins (>50 mile2). he best
results depend on intelligent engineering judgment.

5.2.2 Bridge Deck Drainage Design

5.2.2.1 Runoff and Capacity Analysis
he preferred on-site hydrology method is the rational method, requiring a minimum concentration
time of 10 minutes. Oten the concentration time for the contributing on-site pavement runof is less
than 10 minutes. he initial concentration time is determined by an overland low method until the run-
of is concentrated in a curbed section. Channel low using the roadway-curb cross-section should be
used to determine velocity and low time to the irst inlet. he channel low velocity and looded width
are calculated by Manning’s formula:

1.486 (5.7)
V= AR 2/3 Sf1/2
n

where V = velocity, A = cross-sectional area of low, R = hydraulic radius, Sf = slope of channel, and
n = Manning’s roughness coeicient (FHWA 2012a).
he intercepted low is subtracted from the initial low, and the bypass is combined with runof from
the subsequent drainage area to determine the location of the next inlet. he placement of inlets is deter-
mined by the allowable looded width on the roadway.
Oten, bridges are in sump areas, or the lowest spot on the roadway proile. his necessitates the
interception of most of the low before reaching the bridge deck. Two overland low equations are the
kinematic wave equation

6.92(nL )0.6
to = (5.8)
i 0.4 S 0.3

and the overland equation

3.3(1.1 − C ) L1/2
to = (5.9)
(100 S )1/3

where to = overland low travel time (min), L = length (m) of overland low path, S = slope of overland
low, n = Manning’s roughness coeicient (FHWA 1984), i = design storm rainfall intensity (in/h), and
C = runof coeicient (Tables 5.1 and 5.2).
120 Bridge Engineering Handbook, Second Edition: Substructure Design

5.2.2.2 Select and Size Drainage Facilities

he selection of inlets is based on the allowable looded width that is usually outside the traveled way.
he type of inlet leading up to the bridge deck can vary, depending on the looded width and velocity.
Grate inlets are very common, and curb opening inlets are another alternative in areas with curbs; fur-
ther information is in Brater et al. (1996).

5.2.3 Stage Hydraulics

HW stage is important in bridge design. All available information should be obtained from the ield
and the Bridge Hydrology Report regarding HW marks, HW on upstream and downstream sides of the
existing bridges, high drit proiles, and possible backwater due to existing or proposed construction.
Note that observed high drit and HW marks are not always what they seem. Drit in trees and brush
that could have been bent down by the low of the water is extremely higher than the actual conditions.
Besides, drit may be pushed up on objects or slopes above actual HW elevation by the water velocity
or wave action. Painted HW marks on the bridge should be searched carefully. Some Flood Insurance
Rate Maps and Flood Insurance Study Reports may show stages for various discharges. Backwater stages
caused by other structures in streams should be included.
Duration of high stages should be given, along with the Base Flood Stage and HW for the design
discharge. It should be calculated for existing and proposed conditions that may restrict the channel
producing a higher level. Elevation and season of low water should be given, as this may control design
of tremie seals for foundations and other possible methods of construction. Elevation of overtopping
low and its location should be given. Normally, overtopping occurs at the bridge site but at a low sag in
the roadway away from the bridge site.

5.2.3.1 Waterway Analysis

When specifying the required waterway at the proposed bridge, engineers must consider all adjacent
bridges if these bridges are reasonably close. he waterway section of these bridges should be tied into
the stream proile of the proposed structure. Structures that are upstream or downstream of the pro-
posed bridge may have an impact on the water surface proile. When calculating the efective waterway
area, adjustments must be made for the skew and piers and bents. he required waterway should be
below the 50-year design HW stage.
If stream velocities, scour and erosive forces are high, abutments with wingwall construction may be nec-
essary. Drit will afect the horizontal clearance and the minimum vertical clearance line of the proposed
structure. Field surveys should note the size and type of drit found in the canal. Design based on the 50-year
low requires drit clearance. On major streams and rivers, drit clearance of 2 to 5 m above the 50-year
discharge is needed. On smaller streams, 0.3 to 1 m may be adequate. A formula for calculating freeboard is
Freeboard = 0.1Q 0.3 + 0.008V 2 (5.10)
with Q = discharge and V = velocity.

5.2.3.2 Water Surface Proile Calculation

here are three prominent water surface proile calculation programs available (AASHTO 2005, 2007).
FHWA recommends HEC-RAS (Hydrologic Engineering Centers River Analysis System) that can use GIS
(Geographic Information System) data for input for water surface proile calculations. Besides, WSPRO
(Water Surface Proile) and SMS (Surface-water Modeling System) are two alternatives in practice.

5.2.3.3 Flow Velocity and Distribution

Mean channel, overlow velocities at peak level, and localized velocity at obstructions such as piers should
be estimated for anticipated high levels. Mean velocities may be calculated from known stream discharges
at known channel section areas or known waterway areas of the bridge, using the correct HW stage.
Bridge Scour Design and Protection 121

Surface water velocities are measured roughly by loats during ield surveys. Flow velocities are com-
puted on a uniform channel reach by Manning’s formula (Equation 5.7) if the slope, channel section
(area and wetted perimeter), and roughness coeicient (n) are known.
At least three proiles should be obtained for the channel slope: the channel bottom, the existing
water surface, and the HW surface based on the drit or HW marks. he top of low bank, if overlow is
allowed, should also be obtained. hese proiles are plotted, with existing and proposed bridges or other
obstructions in the channel. he changes of HW slope due to these obstructions and possible backwater
slopes should be estimated.
he channel section used in calculating stream velocities should be more or less uniform. his condi-
tion is usually not always available so that the nearest to uniform conditions should be used with any
necessary modiications made for irregularities.
Velocities may be calculated from PC programs, or calculator programs, if the hydraulic radius,
roughness coeicient, and channel slope are given. he hydraulic radius is the waterway area divided by
the wetted perimeter of an average section of the uniform channel. A section under a bridge whose piers,
abutments, or approach ills obstruct the uniformity of the channel cannot be used because there will
not be uniform low with the structure. If no part of the bridge structure seriously obstructs or restricts
the channel, however, the section at the bridge could be used in the earlier uniform low calculations.
he roughness coeicient “n” for various locations and conditions is found in AASHTO (2005), Brater
et al. (1996), FHWA (1984), and Yen and Chow (1997). At the time of a ield survey, the party chief should
estimate the value of “n” used for the channel section under consideration. Experience is required for ield
determination of a relatively close to actual “n” value. In general, values for natural streams will vary between
0.03 and 0.07. Consider both low and HW “n” value. he water surface slope should be used in this plot and
the slope should be adjusted for obstructions such as bridges, check dams, falls, turbulence, and so on.
he results obtained from this plot may be inaccurate unless considerable thought is given to the
various values of slope, hydraulic radius, and “n.” High velocities between 15 and 20 t/s (4.57–6.10 m/s)
through a bridge opening may be undesirable and require special design considerations. Velocities over
20 t/s (6.10 m/s) should not be used unless special design features are incorporated or if the stream is
mostly conined in rock or an artiicial channel.

5.3 Scour Design and Protection

5.3.1 Scour Analysis
5.3.1.1 Basic Scour Concepts
Scour results from the erosive action of lowing water, excavating and carrying away material from the
bed and banks of streams. Determining the magnitude of scour is complicated by the cyclic nature of
scour process. Designers and inspectors need to carefully study site-speciic subsurface information
in evaluating scour potential at bridges. In this section, the basic bridge scour design procedures and
methods are briely introduced.
Scour should be investigated closely in the ield when designing a bridge. he designer usually places
the top of the footings at or below the total potential scour depth; therefore determining the scour depth
is very important. he total potential scour at a highway crossing is usually comprised of the following
components (FHWA 2012a): aggradation and degradation, stream contraction scour, local scour, and
sometimes with lateral stream migration.

5.3.1.1.1 Long-Term Aggradation and Degradation

When natural or human activities cause streambed elevation changes over a long period of time, aggra-
dation or degradation occurs. Aggradation involves the deposition of material eroded from the chan-
nel or watershed upstream of the bridge, whereas degradation involves the lowering or scouring of the
streambed due to a deicit in sediment supply from upstream.
122 Bridge Engineering Handbook, Second Edition: Substructure Design

Long-term streambed elevation changes result from the changing natural trend of the stream or the
man-made modiication to the stream or river basin. Factors that afect long-term bed elevation changes
are dams and reservoirs up- or downstream of the bridge, changes in watershed land use, channeliza-
tion, cutofs of meander river bends, changes in the downstream channel base level, gravel mining from
the streambed, diversion of water into or out of the stream, natural lowering of the luvial system, move-
ment of a bend, bridge location with respect to stream planform, and stream movement in relation to
the crossing. Tidal ebb and lood may degrade a coastal stream; whereas, littoral drit may cause aggra-
dation. he problem for the bridge engineer is to estimate the long-term bed elevation changes that will
occur during the life time of the bridge.

5.3.1.1.2 Stream Contraction Scour

Contraction scour usually occurs when the low area of a stream at lood stage is reduced, either by a
natural contraction or a man-made contraction (like a bridge). It can also be caused by the overbank
low that is forced back by structural embankments at the approaches to a bridge. here are some other
causes to lead to a contraction scour at a bridge crossing (FHWA 2012a). he decreased low area causes
an increase in average velocity in the stream and bed shear stress through the contraction reach. his in
turn triggers an increase in erosive forces in the contraction. Hence, more bed material is removed from
the contracted reach than from the upstream reach. he natural streambed elevation is lowered by this
contraction phenomenon until relative equilibrium is achieved in the contracted stream reach.
here are two kinds of contraction scour: live-bed and clear-water scours. Live-bed scour occurs
when there is sediment being transported into the contracted reach from upstream. In this case, the
equilibrium state is reached when the transported bed material out of the scour hole is equal to that
transported into the scour hole from upstream. Clear-water scour occurs when the bed sediment trans-
port in the uncontracted approach low is negligible or the material being transported in the upstream
reach is transported through the downstream at less than the capacity of the low. he equilibrium state
of scour is reached when the average bed shear stress is less than that required for the incipient move-
ment of the bed material in this case (Figure 5.3).

5.3.1.1.3 Local Scour

When upstream low is obstructed by obstruction such as piers, abutments, spurs, and embankments, low
vortices are formed at their base, as shown in Figure 5.4 (known as horseshoe vortex). his vortex removes
bed material from around the base of the obstruction. A scour hole eventually develops around the base.
Local scour is either clear-water or live-bed scour. In considering local scour, a bridge engineer needs to
look into the following factors: low velocity, low depth, low attack angle to the obstruction, obstruction
width and shape, projected length of the obstruction, bed material characteristics, bed coniguration of the
stream channel, and also potential ice and debris efects (FHWA 2001, 2012a).

Maximum clear-water scour

Equilibrium scour depth
Pier scour depth, ys

10%
Live-bed scour

Clear-water scour

Time

FIGURE 5.3 Illustrative pier scour depth in a sand-bed stream as a function of time.
Bridge Scour Design and Protection 123

Wake
vortex

Horseshoe vortex

FIGURE 5.4 Schematic representation of local scour at a cylindrical pier.

5.3.1.1.4 Lateral Stream Migration

Streams are dynamic. he lateral migration of the main channel within a loodplain may increase pier
scour, embankment or approach road erosion, or change the total scour depth by altering the low angle
of attach at piers. Lateral stream movements are afected mainly by the geomorphology of the stream,
location of the crossing on the stream, lood characteristics, and the characteristics of the bed and bank
materials (FHWA 2001, 2012a).

5.3.1.2 Designing Bridges to Resist Scour

It is obvious that all scour problems cannot be covered in this special topic section of bridge scour.
A more detailed study is found in FHWA (2012a, b). As described earlier, the three most important com-
ponents of bridge scour are: long-term aggradation or degradation, contraction scour, and local scour.
he total potential scour is a combination of the three components. To design a bridge to resist scour, a
bridge engineer needs to follow the following observation and investigation steps in the design process.
1. Field observation—he purposes of the iled observation are to: (1) observe conditions around
piers, columns, and abutments (Is the hydraulic skew correct), (2) observe scour holes at bends
in the stream, (3) determine streambed material, (4) estimate depth of scour, and (5) complete
Geomorphic Factor Analysis. here is usually no fail-safe method to protect bridges from scour
except possibly keeping piers and abutments out of the HW area; however, proper hydraulic
bridge design can minimize bridge scour and its potential negative impacts.
2. Historic scour investigation—Structures experienced scour in the past are likely to continue dis-
playing scour problems in the future. he bridges that are most concerned with include those
currently experiencing scour problems and exhibiting a history of local scour problems.
3. Problem location investigation—Problem locations include “unsteady stream” locations such as:
near the conluence of two streams, at the crossing of stream bends, and alluvial fan deposits.
4. Problem stream investigation—Problem streams are those that have the following characteristics
of aggressive tendencies: indication of active degradation or aggradation, migration of the stream
or lateral channel movement, streams with a steep lateral slope and/or high velocity, current, past,
or potential in-stream aggregate mining operations, and loss of bank protection in the areas adja-
cent to the structure.
5. Design feature considerations—he following features, which increase the susceptibility to local
scour, should be considered: (1) inadequate waterway opening leads to inadequate clearance to
pass large drit during heavy runof, (2) debris/drit problem: light drit or debris may cause sig-
niicant scour problems, moderate drit or debris may cause signiicant scour but will not create
severe lateral forces on the structure, and heavy drit can cause strong lateral forces or impact
damage as well as severe scour, (3) lack of overtopping relief: water may rise above deck level.
124 Bridge Engineering Handbook, Second Edition: Substructure Design

his may not cause scour problems but does increase vulnerability to severe damage from impact
by heavy drit, and (4) incorrect pier skew: when the bridge pier does not match the channel align-
ment, it may cause scour at bridge piers and abutments.
6. Traic considerations—he amount of traic such as average daily traic (ADT), type of traic,
the length of the detour, the importance of the crossings, and availability of the other crossings
should be taken into consideration.
7. Potential for unacceptable damage—Potential for collapse during lood, safety of traveling pub-
lic and neighbors, efect on regional transportation system, and safety of other facilities (other
bridges, properties) need to be evaluated.
8. Susceptibility of the combined hazard of scour and seismic—he earthquake prioritization list
and the scour critical list are usually combined for bridge design use.

5.3.1.3 Scour Rating

According to California Department of Transportation, structure rating is based upon the following:
1. Letter grading—he letter grade is related to the potential for scour-related problems at this
location.
2. Numerical grading—he numerical rating associated with each structure is a determination of
the severity for the potential scour:
A-1 No problem anticipated
A-2 No problem anticipated/New bridge—No history
A-3 Very remote possibility of problems
B-1 Slight possibility of problems
B-2 Moderate possibility of problems
B-3 Strong possibility of problems
C-1 Some probability of problems
C-2 Moderate probability of problems
C-3 Very strong probability of problems
Scour due to storms is usually greater than from design frequency, say 500-year frequency. FHWA
speciies 500-year frequency is 1.7 times 100-year frequency. Most calculations indicate 500-year fre-
quency is 1.25 to 1.33 times greater than the 100-year frequency (USGS 1981), the 1.7 multiplier should
be a maximum. Consider the amount of scour that would occur at overtopping stages and also pressure
lows. Be aware that storms of lesser frequency may cause larger scour stress on the bridge.

5.3.2 Scour Calculation

All the equations for estimating contraction and local scour are based on laboratory experiments with
limited ield veriication (FHWA 2012a). However, the equations recommended in this section are con-
sidered to be the most applicable for estimating scour depths. Designers also need to give diferent con-
siderations to clear-water scour and live-bed scour at highway crossings and encroachments.
Prior to applying the bridge scour estimating methods, it is necessary to (1) obtain the ixed-bed
channel hydraulics, (2) determine the long-term impact of degradation or aggradation on the bed pro-
ile, (3) adjust the ixed-bed hydraulics to relect either degradation or aggradation impact, and (4) com-
pute the bridge hydraulics accordingly.

5.3.2.1 Speciic Design Approach

he following steps are recommended for determining scour depth at bridges:
Step 1: Analyze long-term bed elevation change
Step 2: Compute the magnitude of contraction scour
Bridge Scour Design and Protection 125

Step 3: Compute the magnitude of local scour at abutments

Step 4: Compute the magnitude of local scour at piers
Step 5: Estimate and evaluate the total potential scour depths

he bridge engineers should evaluate if the individual estimates of contraction and local scour depths
from Step 2 to 4 are reasonable and evaluate the total scour derived from Step 5.

5.3.2.2 Detailed Procedures

Step 1: Analyze long-term Bed Elevation Change
Face of bridge sections showing bed elevation is available in the Maintenance Bridge Books,
old Preliminary Reports, and sometimes in FEMA Studies and USA Corps of Engineers
Studies. Use this information to estimate aggradation or degradation.
Step 2: Compute the magnitude of contraction scour
It is best to keep the bridge out of the normal channel width. However, if any of the following
conditions are present, calculate contraction scour:
Structure over channel in loodplain where the lows are forced through the structure due to
bridge approaches.
Structure over channel where river width becomes narrow.
Relief structure in overbank area with little or no bed material transport.
Relief structure in overbank area with bed material transport.
he general equation for determining contraction scour is:

y s = y 2 − y1 (5.11a)

where y s = depth of scour, y1 = average water depth in the main channel, and y 2 = average water depth
in the contracted section.
FHWA (2012a) provides two methods estimating y2. For live-bed scour, y2 is estimated by

6/7 k1
y 2  Q2   W1 
= (5.11b)
y1  Q1   W 
2

where Q and W are the discharge and channel width, respectively, and subscript “1” is for upstream and
“2” for contraction channel. he exponent k1 is determined as follows:

V*/T K1 Mode of Bed Material Transport

<0.5 0.59 Mostly contacted material discharge
0.5 to 2.0 0.64 Some suspended material discharge
>2.0 0.69 Mostly suspended bed material discharge

Herein, V* is the shear velocity and T the fall velocity of D50. For clear-water scour, FHWA (2012a)
suggests

3/7
 K Q2 
y 2 =  2/3u 2  (5.11c)
 Dm W 
126 Bridge Engineering Handbook, Second Edition: Substructure Design

where Ku = 0.0077 for English units or 0.025 for SI units, Q, Dm, and W are the discharge, median
sediment diameter, and channel width in the contraction reach. Clear-water scour occurs if the average
approach velocity is less than critical velocity Vc for sediment inception described by

Vc = K u y 1/6 D50
1/3 (5.11d)

where Ku = 6.19 for SI units or 11.17 for English units. In general, clear-water scour is about 10% bigger
than corresponding live-bed scour.
Step 3: Compute the magnitude of local scour at the abutments
FHWA (2012a) recommends three methods calculating abutment scour: Froehlich’s equation, HIRE
equation, and NCHRP 24-20 approach. For example, HIRE equation reads as

ys K
= 4 Fr0.33 1 K 2 (5.11e)
y1 0.55

where ys = scour depth, y1 = low depth at the abutment on the overbank or in the main channel, Fr =
Froude number based on the velocity and depth adjacent to and upstream of the abutment, K1 = abut-
ment shape coeicient below, and K2 = coeicient for skew angle of abutment to low.

Description K1
Vertical-wall abutment 1.00
Vertical-wall abutment with wing walls 0.82
Spill-through abutment 0.55

Step 4: Compute the magnitude of local scour at the piers

he pier alignment is the most critical factor in determining scour depth. Piers should align with
stream low. When low direction changes with the stages, cylindrical piers or some variation may be
the best alternative. Be cautious, since large diameter cylindrical piers can cause considerable scour. Pier
width and pier nose are also critical elements in causing excessive scour depth.
FHWA (2012a) provides several methods estimating pier scour depth. For a sand bed channel, an
acceptable method to specify the maximum possible scour depth for both live-bed and clear-water
channel proposed by the Colorado State University (FHWA 2012a) is as follows:
0.65
ys a
= 2.0 K1K 2 K 3   Fr0.43
1 (5.12)
y1  y1 

where y s = scour depth, y1 = low depth just upstream of the pier, K1 = correction for pier shape from
Figure 5.5 and Table 5.3, K2 = correction for angle of attack of low from Table 5.4, K3 = correction for
Bed Condition from Table 5.5, a = pier width, l = pier length, and Fr1 = Froude number = V/(gy)0.5 (Just
upstream from bridge).
Note that Equation 5.12 does not include the efect of sediment mixture. For nonuniform sediment,
Guo’s (2012) equation is recommended as

ys  H 2 /σ 3/2 − H cp
2

= tanh   (5.13a)
ay1  3.75

where Hcp is the critical Hager number for uniform sediment (= 1), determined by

 2  b  1/4 
H cp = 1 −    H c (5.13b)
 3  B 
Bridge Scour Design and Protection 127

L L

a a a

(a) (b) (c)

L L=(# of piers).(a)

a a

(d) (e)

FIGURE 5.5 Common pier shapes: (a) Square nose (b) Round nose (c) Cylinder (d) Sharp nose (e) Group of
cylinders (See multiple coloumns).

TABLE 5.3 Correction Factor, K1, for Pier Nose Shape

Shape of Pier Nose K1
Square nose 1.1
Round nose 1.0
Circular cylinder 1.0
Sharp nose 0.9
Group of cylinders 1.0

TABLE 5.4 Correction Factor, K 2, for Flow Angle of Attack

Angle L/a = 4 L/a = 8 L/a = 12
0 1.0 1.0 1.0
15 1.5 2.0 2.5
30 2.0 2.75 3.5
45 2.3 3.3 4.3
90 2.5 3.9 5

TABLE 5.5 Increase in Equilibrium Pier Scour Depths K 3 for Bed Conditions
Bed Conditions Dune Height H, t K3

Clear-water scour N/A 1.1

Plane bed and antidune low N/A 1.1
Small dunes 10 > H > 2 1.1
Medium dunes 30 > H > 10 1.1–1.2
Large dunes H > 30 1.3

where B = channel width and Hc is the critical Hager number corresponding to

2.33D*−0.25 ( Rh / D50 )1/6 for D* ≤ 10


H c = 1.08D*1/12 ( Rh / D50 )1/6 for 10 < D* < 150 (5.13c)
1.65( R / D )1/6 for D* ≥ 150
 h 50
128 Bridge Engineering Handbook, Second Edition: Substructure Design

where D * = [(ρs/ρ–1)g/v2]1/3D50 is dimensionless sediment size, and R h = hydraulic radius. In practice, the
efect of critical Hager number can be neglected and Equation 5.13a is reduced to

ys  H 2 / σ 3/2 
= tanh  (5.14)
ay1  3.75 

which gives the potential maximum scour depth as ys = (ay1)0.5. Finally, HEC-18 (FHWA 2012a) recom-
mends a revised version of Equation 5.14 as

ys  Η 2 / σ 3/2 
= 1.1K1K 2 K 3 tanh  (5.15)
ay1  1.97 

which is based on both laboratory and ield data and detailed in Guo et al. (2012).

5.3.2.3 Estimate and Evaluate Total Potential Scour Depths

Total potential scour depths is usually the sum of long-term bed elevation change (only degradation is
usually considered in scour computation), contraction scour, and local scour. Historical scour depths
and depths of scourable material are determined by geology. When estimated depths from the above
methods are in conlict with geology, the conlict shall be resolved by the hydraulic engineer and the
geotechnical engineer, based on economics and experience, it is best to provide for maximum antici-
pated problems.

5.3.3 Pressure Flow Scour from Model Tests

Model tests use a small-scale bridge structure to simulate and predict the performance of a full-scale
bridge. he equilibrium scour depth ys is an important parameter, yet requiring a long period to be
attained. In such cases, the following procedures are recommended (Guo 2011).
Suppose a laboratory bridge is scaled by Froude similitude. If the model scour starts with a lat bed
and its depth is η1 at time t1, then the equilibrium scour depth ysm is calculated by

y sm = η1 (1 − e −T1 /T90 )− n (5.16)

where η1 is measured scour depth (m) at time t1 (s), T1 = t1V/hb the dimensionless time with V (m/s) as
approach velocity and hb the bridge opening height before scour, T90 = 1.56 × 105 corresponding to the
dimensionless time at 90% of ys, and n = 0.239. he prototype scour depth is then scaled back according
to Froude similitude as
2
y sp V 
= m (5.17)
y sm  Vp 

where ysp and Vp are the scour depth and approach velocity, respectively, for prototype low.

5.3.4 Bridge Scour Investigation and Prevention

5.3.4.1 Steps to Evaluate Bridge Scour
It is recommended that an interdisciplinary team of hydraulic, geotechnical, and bridge engineers
should conduct the evaluation of bridge scour. he following approach is recommended for assessing
the vulnerability of existing bridges to scour (FHWA 2012a):
Bridge Scour Design and Protection 129

Step 1: Screen all bridges over waterways into ive categories: (1) low risk, (2) scour-susceptible,
(3) scour-critical, (4) unknown foundations, or (5) tidal. Bridges that are particularly vul-
nerable to scour failure should be identiied immediately and the associated scour problem
addressed. hese particularly vulnerable bridges are:
Bridges currently experiencing scour or that have a history of scour problems during past
loods as identiied from maintenance records and experience, bridge inspection records.
Bridges over erodible streambeds with design features that make them vulnerable to scour.
Bridges on aggressive streams and waterways.
Bridges located on stream reaches with adverse low characteristics.
Step 2: Prioritize the scour-susceptible bridges and bridges with unknown foundations by con-
ducting a preliminary oice and ield examination of the list of structure compiled in Step 1
using the following factors as a guide: (1) the potential for bridge collapse or for damage to
the bridge in the event of a major lood; and (2) the functional classiication of the highway
on which the bridge is located, and the efect of a bridge collapse on the safety of the travel-
ing public and on the operation of the overall transportation system for the area or region.
Step 3: Conduct oice and ield scour evaluations of the bridges on the prioritized list in Step 2
using an interdisciplinary team of hydraulic, geotechnical, and bridge engineers:
In the United States, FHWA recommends using 500-year lood or a low 1.7 times the 100-
year lood where the 500-year lood is unknown to estimate scour (Waananen and Crippen
1977), and then analyze the foundations for vertical and lateral stability for this condition of
scour. he maximum scour depths that the existing foundation can withstand are compared
with the total scour depth estimated. An engineering assessment must be then made as to
whether the bridge should be classiied as a scour-critical bridge.
Enter the results of the evaluation study in the inventory in accordance with the instructions
in the FHWA (1995).
Step 4: For bridges identiied as scour critical from the oice and ield review in Steps 2 and 3,
determine a plan of action for correcting the scour problem.

5.3.5 Introduction to Bridge Scour Inspection

he bridge scour inspection is one of the most important parts of preventing bridge scour from endanger-
ing bridges. Two main objectives to be accomplished in inspecting bridges for scour are: (1) To accurately
record the present condition of the bridge and the stream and (2) to identify conditions that are indicative
of potential problems with scour and stream stability for further review and evaluation by other experts.
In this section, the bridge inspection practice recommended by USFHWA (Waananen and Crippen
1977, FHWA 1989) is presented for engineers to follow as guidance.
1. Oice Review
It is highly recommended to make an oice review of bridge plans and previous inspection
reports prior to making the bridge inspection. Information obtained from the oice review pro-
vides a better foundation for inspecting the bridge and the stream. he following questions should
be answered in the oice review:
a. Has an engineering scour evaluation been conducted? If so, is the bridge scour critical?
b. If the bridge is scour critical, has a plan of action been made for monitoring the bridge and/or
installing scour prevention measures?
c. What do comparisons of streambed cross sections taken during successive inspections reveal
about the streambed? Is it stable? Degrading? Aggrading? Moving laterally? Are there scour
holes around piers and abutments?
d. What equipment is needed to obtain streambed cross sections?
e. Are there sketches and aerial photographs to indicate the planform locations of the stream
and whether the main channel is changing direction at the bridge?
130 Bridge Engineering Handbook, Second Edition: Substructure Design

f. What type of bridge foundation was constructed? Do the foundations appear to be vulnerable
to scour?
g. Do special conditions exist requiring particular methods and equipment for underwater
inspections?
h. Are there special items that should be looked at including damaged riprap, stream channel at
adverse angle of low, problems with debris, and so on?
2. Bridge Scour Inspection Guidance
The condition of bridge waterway opening, substructure, channel protection, and scour
prevention measures should be evaluated along with the condition of the stream during
the bridge inspection. The following approaches are presented for inspecting and evaluating
the present condition of the bridge foundation for scour and the overall scour potential at the
bridge.
Substructure is the key item for rating the bridge foundations for vulnerability to scour
damage. Both existing and potential problems with scour should be reported so that an inter-
disciplinary team can make a scour evaluation when a bridge inspection inds that a scour
problem has already occurred. If the bridge is determined to be scour critical, the rating of the
substructures should be evaluated to ensure that existing scour problems have been consid-
ered. he following items should be considered in inspecting the present condition of bridge
foundations:
a. Evidence of movement of piers and abutments such as rotational movement and settlement
b. Damage to scour countermeasures protecting the foundations such as riprap, guide banks,
sheet piling, sills, etc.
c. Changes in streambed elevation at foundations such as undermining of footings, exposure of
piles
d. Changes in streambed cross section at the bridge, including location and depth of scour
holes
In order to evaluate the conditions of the foundations, the inspectors should take cross sections
of the stream and measure scour holes at piers and abutments. If equipment or conditions do not
permit measurement of the stream bottom, it should be noted for further investigation.
To take and plot measurement of stream bottom elevations in relation to the bridge founda-
tions is considered the single most important aspect of inspecting the bridge for actual or poten-
tial damage from scour. When the stream bottom cannot be accurately measured by conventional
means, there other special measures need to be taken to determine the condition of the substruc-
tures or foundations such as using divers and using electronic scour detection equipment. For the
purposes of evaluating resistance to scour of the substructures, the questions remain essentially
the same for foundations in deep water as for foundations in shallow water (FHWA 1995) as
follows:
a. How does the stream cross section look at the bridge?
b. Have there been any changes as compared to previous cross section measurements? If so, does
this indicate that (1) the stream is aggrading or degrading; or (2) local or contraction scour is
occurring around piers and abutments?
c. What are the shapes and depths of scour holes?
d. Is the foundation footing, pile cap, or the piling exposed to the stream low; and if so, what is
the extent and probable consequences of this condition?
e. Has riprap around a pier been moved and removed?
Any condition that a bridge inspector considers to be an emergency or potentially hazardous
nature should be reported immediately. his information as well as other conditions, which do
not pose an immediate hazard, but still warrant further investigation, should be conveyed to the
interdisciplinary team for further review.
Bridge Scour Design and Protection 131

5.3.6 Real-Time Monitoring

Real-time bridge-scour monitoring details the scour developments using technology and communica-
tions systems. It provides timely and quality data of scour developments to bridge managers for deci-
sion making. Since scour is the most common cause of bridge failures and the most expensive kind
of damage to repair bridges, the FHWA actively promotes real-time scour monitoring research and
practice. With two consecutive data (t0,η0) and (t1,η1), the scour ys at the next time step t is found as
follows (Guo 2011).
Assume a quasi-steady state low for time period t0 < t1 < t. he equilibrium scour depth ys is calcu-
lated by

n
 η1/n − η1/0 n exp[−(T1 − T0 ) / T90 ] 
ys =  1  (5.18)
 1 − exp[−(T1 − T0 ) / T90 ] 

where T0 = t0V/hb and T1 = t1V/hb with V and hb the approach velocity and bridge opening height before
scour, respectively, T0 = 1.56 × 105 and n = 0.239. With Equation 5.18, the scour depth η(t) is estimated
by solving the following equation:

y 1/s n − η1/n  T − T0  (5.19)

= exp  −
y s − η0
1/n 1/n
 T90 

with T = tV/hb. According to the value η(t) from Equation 5.19, bridge managers can predict scour
conditions at critical moments and formulate timely corrective strategies. Once a bridge scour is found
critical during loods, the bridge should be closed for public safety and the damages should be imme-
diately repaired ater loods. Note that the above monitoring scheme was developed for pressure bridge
scour, but it can be extended for pier scour by replacing hb with pier diameter.

5.3.7 Scour Protection

Scour prevention measures are generally incorporated ater the initial construction of a bridge to make
it less vulnerable to damage or failure from scour. A plan of preventive action usually has three major
components (FHWA 2012a): (1) timely installation of temporary scour prevention measures; (2) devel-
opment and implementation of a monitoring program; and (3) schedule for timely design and construc-
tion of permanent scour prevention measures.
For new bridges (FHWA 2012a), the following summarizes the best solutions for minimizing scour
damage: (1) locating the bridge to avoid adverse lood low patterns; (2) streamlining bridge elements
to minimize obstructions to the low; (3) designing foundations safe from scour; (4) founding bridge
pier foundations suiciently deep to not require riprap or other prevention measures; and (5) founding
abutment foundations above the estimated local scour depth when the abutment is protected by well-
designed riprap or other suitable measures.
For existing bridges, the following alternatives are used: (1) monitoring scour depths and closing
bridge if excessive bridge scour exists; (2) providing riprap at piers and/or abutments and monitoring
the scour conditions; (3) constructing guide banks or spur dikes; (4) constructing channel improve-
ments; (5) strengthening the bridge foundations; (6) constructing sills or drop structures; and (7) con-
structing relief bridges or lengthening existing bridges. Further scour prevention measures are found
in FHWA (2012a, b).
132 Bridge Engineering Handbook, Second Edition: Substructure Design

References
AASHTO. 2005. Model Drainage Manual, 3rd Edition, American Association of State Highway and
Transportation Oicials, Washington, DC.
AASHTO. 2007. Highway Drainage Guidelines, 4th Edition, American Association of State Highway and
Transportation Oicials, Washington, DC.
Brater, E., King, H., Lindell, J., and Wei, C. 1996. Handbook of Hydraulics, 7th Edition, McGraw-Hill,
New York, NY.
FHWA. 1984. Hydrology, FHWA-TS-84-204, Federal Highway Administration, Washington, DC.
FHWA. 1989. Design of Riprap Revetments, Hydraulic Engineering Circular No 11, FHWA- IP-89-016,
Federal Highway Administration, Washington, DC.
FHWA. 1995. Recording and Coding Guide for the Structure Inventory and Appraisal of the Nation’s Bridges,
FHWA-PD-96-001, Federal Highway Administration, Washington, DC.
FHWA. 2001. River Engineering For Highway Encroachments: Highways In he River Environment, FHWA
NHI 01-004, Federal Highway Administration, Washington, DC.
FHWA. 2002. Highway Hydrology, 2nd Edition, NHI-02-001, Federal Highway Administration,
Washington, DC.
FHWA. 2005. Debris Control Structures Evaluation and Countermeasures, Hydraulic Engineering Circular
No. 9, FHWA-IF-04-016, Federal Highway Administration, Washington, DC.
FHWA. 2012a. Evaluating Scour at Bridges, 5th Edition, HIF-12-003, Federal Highway Administration,
Washington, DC.
FHWA. 2012b. Stream Stability at Highway Structures, 4th Edition, HIF-12-004, Federal Highway
Administration, Washington, DC.
Guo, J. 2011. Time-dependent clear-water scour for submerged bridge lows. J Hydraulic Research, 49(6),
744–749.
Guo, J. 2012. Pier scour in clear water for sediment mixtures. J Hydraulic Research, 50(1), 18–27.
Guo, J., Suaznabar, O., Shan, H., and Shen, J. 2012. Pier Scour in Clear-Water Conditions with Non-Uniform
Bed Materials, FHWA-HRT-12-022, Federal Highway Administration, Washington, DC.
USGS. 1981. Guidelines for Determining Flood Flow Frequency, United States Geological Survey,
Reston, VA.
Waananen, A. O., and Crippen, J. R. 1977. Magnitude and Frequency of Floods in California, Water
Resources Investigation 77-21, United States Geological Survey, Menlo Park, CA.
Yen, B. C., and Chow, V. T. 1997. Feasibility on Research on Local Design Storms, FHWA-RD-78-65, Federal
Highway Administration, Washington, DC.
 6
Abutments
6.1 Introduction ......................................................................................133
6.2 Abutment Types ................................................................................133
Open-End and Closed-End Abutments • Monolithic and Seat-Type
Abutments • Abutment Type Selection
6.3 General Design Considerations......................................................135
6.4 Seismic Design Considerations ......................................................137
6.5 Miscellaneous Design Considerations ..........................................142
Abutment Wingwall • Abutment Drainage • Abutment Slope
Protection • Miscellaneous Details
6.6 Design Example ................................................................................144
Design Data • Abutment Support Width Design • Abutment
Stability Check • Abutment Backwall and Stem Design • Abutment
Linan Wang Backwall Design • Abutment Stem Design • Abutment Footing
California Department Design • Abutment Wingwall Design
of Transportation References......................................................................................................154

6.1 Introduction
As a component of a bridge, the abutment provides the vertical support to the bridge superstructure
at bridge ends, connects the bridge with the approach roadway, and retains the roadway base materi-
als from the bridge spans. Although there are numerous types of abutments and the abutments for the
important bridges may be extremely complicated, the analysis principles and design methods are very
similar. In this chapter, the topics related to the design of conventional highway bridge abutment are
discussed and a design example is illustrated.

6.2 Abutment Types

6.2.1 Open-End and Closed-End Abutments
From the view of the relation between the bridge abutment and roadway or channel that the bridge over-
crosses, the bridge abutments can be divided into two categories: open-end abutment and closed-end
abutment, as shown in Figure 6.1.
For open-end abutment, there are slopes between the bridge abutment face and the edge of the road-
way or the channel. hose slopes provide a widely opened area to the traic lows or water lows under
the bridge. It imposes much less impact on the environment and the traic lows under the bridge than
closed-end abutment. Also it is easier to make future widening on roadway or the channel under the
bridge by adjusting the slope ratios. However, the existing of slopes usually requires longer bridge spans
and some extra earthwork. his may result in the raise of bridge construction cost.
he closed-end abutment is usually constructed close to the edge of the roadways or channels. In the
case of limited right of way, a high abutment wall is usually constructed without front slope to meet

133
134 Bridge Engineering Handbook, Second Edition: Substructure Design

Edge of roadway
or water surface Abutment
backwall Bridge superstructure

H ⩽ 8'–0''
Abutment
stem Edge of roadway
Edge of roadway
or water surface
or water surface

Abutment
footing

(a) (b)

Bridge approach Bridge approach

slab Bridge superstructure slab Bridge superstructure

Abutment
backwall
Abutment
Abutment

H > 8'–0''
stem Edge of roadway
stem Edge of
or water surface
roadway or
water surface

Abutment Abutment
footing footing
(c) (d)

FIGURE 6.1 Typical abutment types. (a) Open end, monolithic type, (b) Open end short stem type, (c) Closed
end, monolithic type, (d) Closed end, short stem type.

the vertical clearance requirements of the traic or the water lows. Since there is no room or only little
room exists between the abutment and the edge of traic or water low, it is very diicult to do the future
widening on the roadways or channels under the bridge. Also the high abutment walls and larger vol-
ume of backill material oten result in higher abutment construction costs and more settlement of road
approaches than for open-end abutment.
Generally the open-end abutments are more economical, adaptable, and attractive than the closed-
end abutments. However, the bridges with closed-end abutments have been widely constructed in the
urban area and for rail transportation system because of the right of way restriction and the large scale
of the live load for trains, which usually results in short bridge spans.

6.2.2 Monolithic and Seat-Type Abutments

Based on the connection type between the abutment stem and bridge superstructure, the abutments
also can be grouped as two categories: the monolithic or end diaphragm abutment and the seat-type
abutment as shown in Figure 6.1.
For monolithic abutment, the abutment stem is monolithically constructed with the bridge super-
structure. here is no relative displacement allowed between the bridge superstructure and abutment.
Abutments 135

All superstructure forces at bridge ends are transferred to the abutment stem and then to the abutment
backill soil and footings. he advantages of this type of abutment are its initial lower construction cost
and its immediate engagement of backill soil that absorbs the energy when the bridge is subjected to
transitional movement. However, the passive soil pressure induced by bridge lateral movement could
result in the diiculty of designing the abutment stem. Also a higher maintenance cost on bridge
approach might be expected for this type of abutment. In the practice, this type of abutment is mainly
constructed for short bridges.
For seat-type abutment, the abutment stem is constructed separately with the bridge superstructure.
he bridge superstructure seats on the abutment stem through bearing pads, rock bearings, or other
devices. his type of abutment allows the bridge designer to control the superstructure forces that are
to be transferred to the abutment stem and its backill soil. By adjusting the devices between the bridge
superstructure and abutment, the bridge displacement could be controlled. his type of abutment may
have short stem or high stem as shown in Figure 6.1. For short-stem abutment, the abutment stif-
ness usually is much larger than the connection devices between the superstructure and the abutment.
herefore, those devices can be treated as boundary conditions in the bridge analysis. Comparatively,
the high-stem abutment may subject signiicant displacement under the relative less forces. he stifness
of high-stem abutment and the response of surrounding soil may have to be considered in the bridge
analysis. he availability of the displacement of connection devices, the allowance of the superstructure
shrinkage, and concrete shortening make this type abutment be widely selected for the long bridge
constructions, especially for prestressed concrete bridges and steel bridges. However, the bridge design
practice shows that the relative weak connection devices between the superstructure and the abutment
usually cause the adjacent columns to be specially designed. Although the seat-type abutment has rela-
tively higher initial construction cost than monolithic abutment, its maintenance cost is relatively low.

6.2.3 Abutment Type Selection

he selection of an abutment type needs to consider all available information and bridge design require-
ments. hose may include bridge geometry, roadway and riverbank requirements, geotechnical con-
dition, right-of-way restrictions, architect requirements, economical considerations, and so on. he
knowledge of advantages and disadvantages for the diferent types of abutments will greatly beneit the
bridge designer to choose a right type of abutment for the bridge structure from the beginning stage of
the bridge design.

6.3 General Design Considerations

Abutment design loads usually include vertical and horizontal loads from bridge superstructure, ver-
tical and lateral soil pressures, abutment gravity load, and the live load surcharge on the abutment
backill materials. An abutment shall be designed as no damage to withstand the earth pressure, the
gravity loads of bridge superstructure and abutment, live load on superstructure or approach ill, wind
loads and the transitional loads transferred through the connections between the superstructure and
the abutment. Any possible combination of those loads, which produce the most severe condition of
loading, shall be investigated in abutment design. Meanwhile, for the integral abutment or monolithic
type abutment, the efects of bridge superstructure deformations, including bridge thermal movements,
to the bridge approach structures must be considered in the abutment design. Nonseismic design loads
at service level and their combinations are shown in Table 6.1 and Figure 6.2. For Load Factor Design
(LFD) (AASHTO 2002) or Load and Resistant Factor Design (LRFD) (AASHTO 2012), the abutment
design loads could be obtained by multiplying the load factors to the loads at service levels. Under the
seismic loading, the abutment may be designed as no support lost to the bridge superstructure while the
abutment may sufer some repairable damages during a major earthquake.
136 Bridge Engineering Handbook, Second Edition: Substructure Design

TABLE 6.1 Abutment Design Loads (Service Load Design)

Case
Abutment Design Loads I II III IV V
Dead load of superstructure X X — X X
Dead load of wall and footing X X X X X
Dead load of earth on heel of wall including surcharge X X X X —
Dead load of earth on toe of wall X X X X —
Earth pressure on rear of wall including surcharge X X X X —
Live load on superstructure X — — X —
Temperature and shrinkage — — — X —
Allowable pile capacity of allowable soil pressure in % or basic 100 100 150 125 150

Truck or equivalent
2' surcharge loading, whichever 2' surcharge 2' surcharge
governs.

Case I Case II Case III

Case II will govern only when live load reaction
falls behind center of gravity of piles.
Truck or equivalent
loading, whichever
2' surcharge
governs.

Temperature
and shrinkage

Case IV Case V
Note : Also consider Case IV with Case V will govern only when dead load reaction
no live load on superstructure. falls ahead of center of gravity of piles.

FIGURE 6.2 Coniguration of abutment design load and load combinations.

he load and load combinations listed in Table 6.1 may cause abutment sliding, overturning, and soil
bearing failures. hose stability characteristics of abutment must be checked to satisfy certain restric-
tions. For the abutment with spread footings in service load design, the factor of safety to resist sliding
should be greater than 1.5; the factor of safety to resist overturning should be greater than 2.0; the factor
of safety against the soil bearing failure should be greater than 3.0. For the abutment with pile support,
the piles have to be designed to resist the forces that cause the abutment sliding, overturning, and bear-
ing failure.
he abutment deep shear failure also needs to be studied in the abutment design. Usually, the poten-
tial of this kind of failure is pointed out in the geotechnical report to the bridge designers. Deep pilings
or relocating the abutment may be used to avoid this kind of failure.
Abutments 137

6.4 Seismic Design Considerations

he investigations of past earthquake damage to bridges reveal that there are commonly two types of
abutment earthquake damage—the stability damage and the component damage.
he abutment stability damage during an earthquake is mainly caused by the foundation failure due
to the excessive ground deformation or the loss of bearing capacities of the foundation soil. hose foun-
dation failures result in the abutment sufering tilting, sliding, settling, and overturning. he founda-
tion soil failure usually occurs at the poor soil conditions such as sot soil and the existence of high water
table. In order to avoid these kinds of soil failures during an earthquake, borrowing backill soil, pile
foundations, high-degree soil compaction, previous materials, and drainage system may be considered
in the design.
he abutment component damage is generally caused by the excessive soil pressure, which is mobi-
lized by the large relative displacement between the abutment and its backilled soil. hose excessive
pressures may cause severe damage to abutment components such as abutment back walls and abutment
wingwalls. However, the abutment component damages usually do not cause the bridge superstructure
lost support at abutment and they are easy to be repaired. his may allow the bridge designer to use the
deformation of abutment backill soil under the seismic forces to dissipate the seismic energy to avoid
the bridge losing support at columns under a major earthquake strike.
he behavior of abutment backill soil deformed under seismic load is very eicient to dissipate the
seismic energy especially for the bridges with total length of less than 300 t (91.5 m) with no hinge,
no skew, or slightly skewed (i.e., <15°). he tests and analysis revealed that if the abutments of a short
bridge are capable to mobilize the backill soil and are well tied into the backill soil, a damping ratio
in the range of 10%–15% is justiied. his will elongate the bridge period and may reduce the ductility
demand on the bridge columns. For short bridges, a damping reduction factor, D, may be applied to the
forces and displacement obtained from the bridge elastic analysis. his factor D is given in Equation 6.1.

1.5
D= + 0.5 (6.1)
40 C + 1

where C = damping ratio.

Based on Equation 6.1, for 10% damping ratio, a factor of D = 0.8 may be applied to the elastic forces
and displacements resulted from the elastic structure analysis; for 15% damping ratio, a factor of D =
0.7 may be applied to such elastic forces and displacements. Generally, the reduction factor D should be
applied to the forces corresponding to the bridge shake mode that shows the abutment being excited.
he earthquake forces that backill soil applied to abutment are very diicult to predict (Goel 1997;
Sorensen 1997). he Mononobe-Okabe method is usually used to quantify the active earth pressure
induced by earthquake for earth-retaining structures with non-top-restrains. For the passive earth
pressure induced by bridge movement at abutment under seismic loading, the study and tests revealed
that the soil resistances mainly depend on the abutment movement direction and magnitude. he “near-
full-scale” abutment tests performed at University of California at Davis (Maroney et al. 1993 and 1994)
shows a nonlinear relationship between the abutment displacement and the backill soil reactions under
certain seismic loading when the abutments move toward its backill soil. his relation was plotted as
shown in Figure 6.3. It is diicult to simulate this nonlinear relationship between the abutment dis-
placement and backill soil reactions while performing the bridge dynamic analysis. However, the tests
concluded an upper limit for the backill soil reaction on the abutment. In design practice, a peak soil
pressure acting on the abutment may be predicted corresponding to certain abutment displacement.
Based on the tests and the investigations to the past earthquake damages, California Transportation
Department guided for the bridge analysis considering abutment damping behavior as follows.
Using the peak abutment force and the efective area of the mobilized soil wedge, the peak soil pres-
sure is compared to a maximum capacity of 7.7 ksf (369 kPa). If the peak soil pressure exceeds the soil
138 Bridge Engineering Handbook, Second Edition: Substructure Design

Displacement (Δs/h) [mm/m]

0.0 16.7 33.3 50.0 64.7 83.3 100.0
6 286.9

5 239.1

Test data
4 151.2

Load (kN/m2)
Idealized trilinear
Load (ksf )

3 143.4
Abutment backwall-soil
load/width/height vs. deflection/height
2 Δs 95.6
Load
h
1 47.8

0 0
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Displacement (Δs/h) [inches/foot]

FIGURE 6.3 Proposed characteristics and experimental envelope for abutment backill load deformation.

capacity, the analysis should be repeated with reduced abutment stifness. It is important to note that
the 7.7 ksf (369 kPa) soil pressure is based on a reliable minimum wall height of 8 t (2.438 m). If the wall
height is less than 8 t (2.438 m), or if the wall is expected to shear of at a depth below the roadway less
than 8 t (2.438 m), the allowable passive soil pressure must be reduced by multiplying 7.7 ksf (369 kPa)
with the ratio of (h/8)2, where “h” is the efective height of abutment wall in feet. Furthermore, the
shear capacity of the abutment wall diaphragm (structural member mobilizing soil wedge) should be
compared to the demand shear forces to ensure the soil mobilizations. Abutment spring displacement
is then evaluated against the acceptable level of displacement 0.2 t (61 mm). For monolithic type abut-
ment, this displacement is equal to the bridge superstructure displacement. For seat-type abutment,
this displacement is usually not equal to the bridge superstructure displacement that may include the
gap between the bridge superstructure and abutment backwall. However, a net displacement of about
0.2 t (61 mm) at abutment should not be exceeded. Field investigations ater the 1971 San Fernando
Earthquake revealed that the abutment, which moved up to 0.2 t (61 mm) in the longitudinal direction
into the backill soil, appeared to survive with little need for repair. he abutments in which the back-
wall breaks of before other abutment damage may also be satisfactory if a reasonable load path can be
provided to adjacent bents and no collapse potential is indicated (Caltrans 1996).
he current seismic design criteria of California Transportation Department (Caltrans 2010) suggests
an efective initial abutment stifness of Ki = 50 t/in/t to be used in seismic analysis. his Ki could gen-
erate a larger backill soil capacity with 0.2 t abutment movement. However, an abutment displacement
coeicient R A is assigned to justify the contribution of the abutment stifness in the analysis.

RA = ∆ D / ∆ eff (6.2)

where:
ΔD = he longitudinal displacement demand at the abutment from elastic analysis
Δef = he efective longitudinal abutment displacement at idealized yield
If R A ≤ 2: It indicates that the bridge stifness is dominated by abutment stifness and Ki used in the
analysis should be realistic.
Abutments 139

Interior
supplemental
shear wall

Increase
wingwall
thickness

EQT

FIGURE 6.4 Abutment transverse enhancement.

If R A ≥ 4: It indicates that the contribution of abutment stifness is not signiicant. he Ki could be

reduced in value or even neglected in the analysis.
If 2 < R A < 4: he abutment stifness may have to be set by try-and-error method following the 7.7 ksf
and 0.2 t rule (Caltrans 2010).
For seismic analysis in bridge transverse direction, since the backill soil is usually slopping away
from abutment wingwall and there is a relatively weak connection between the abutment wingwall and
the stem, the displacement coeicient R A shall not be applied directly in the analysis. Reduced R A or
fully released abutment cases shall be studied. In order to increase the transverse stifness of the abut-
ment, interior supplemental shear walls may be attached to the abutment or the wingwall thickness may
be increased, as shown in Figure 6.4.
Based on the above guidelines, the abutment analysis can be carried out more realistically by try-
and-error method on abutment soil springs. he criterion for abutment seismic resistance design may
be set as follows:
Monolithic abutment or diaphragm abutment (Figure 6.5)

With footing ▸ EQ L ≤ Rsoil + Vdiaphragm

EQ T ≤ Vww + Vkey
Vkeys = 0.75(Vpiles) for pile footing
Vkeys = μ (Dead Load reaction @ bottom of
footing) for spread footing

With footing ▸
EQL ≤ Rsoil + Vdiaphragm

EQ T ≤ Vww + Vpiles
140 Bridge Engineering Handbook, Second Edition: Substructure Design

Soil mobilized
EQL
Shear capacity
of diaphragm

Section

EQT

Shear capacity of
one wingwall (Vww)

Shear keys
Elevation
(a)

Soil mobilized
EQL

Shear capacity
of diaphragm

Section

EQT

Shear capacity of
one wingwall (Vww)
Shear capacity of piles
Elevation
(b)

FIGURE 6.5 Seismic resistance elements for monolithic abutment (a) with footing, (b) without footing.

Seat-Type Abutment (Figure 6.6)

Seat Type Abutment ▸ EQ L ≤ Rsoil

EQ T ≤ Rkeys
Vkeys = Vww + 0.75(Vpiles) for pile footing
Vkeys = Vww + μ (Dead Load reaction @
bottom of footing) for spread footing

where
EQL = Longitudinal earthquake force from an elastic analysis
EQ T = Transverse earthquake force from an elastic analysis
Abutments 141

Soil mobilized

EQL

Section

EQT

Shear capacity of
Shear keys
one wingwall (Vww)

Shear capacity
of piles
Elevation

FIGURE 6.6 Seismic resistance elements for seat-type abutment.

Rsoil = Resistance of soil mobilized behind abutment

Rdiaphragm = ϕ times the nominal shear strength of the diaphragm
Rww = ϕ times the nominal shear strength of the wingwall
Rpiles = ϕ times the nominal shear strength of the piles
R keys = ϕ times the nominal shear strength of the keys in the direction of consideration
ϕ = Strength factor for seismic loading
μ = Coeicient factor between soil and concrete face at abutment bottom
he purpose of applying a factor of 0.75 to the design of abutment shear keys is to reduce the possible
damage to the abutment piles. For all transverse cases, if the design transverse earthquake force exceeds
the sum of the capacities of the wingwalls and piles, the transverse stifness for the analysis should equal
to zero (EQ T = 0). herefore, a released condition that usually results in a larger lateral displacement at
adjacent bents should be studied.
Responding to seismic load, bridge usually accompanies a large displacement. In order to provide
support at abutments for the bridge with large displacement under seismic loading, enough seat width at
abutment must be designed. heoretically, the abutment seat width, as shown in Figure 6.7, has to meet
the requirement in the formula:

NA ≥ ∆ p/s + ∆ cr +sh + ∆ temp + ∆ eq + 4 (6.3)

where
NA = Abutment seat width normal to the center line of bearing (in)
Δp/s = Displacement attributed to prestress shoring
Δcr+sh = Displacement attributed to creep and shrinkage
Δtemp = Displacement attributed to thermal expansion and contraction
Δeq = he maximum relative displacement between superstructure and abutment results in
seismic global or local analysis
142 Bridge Engineering Handbook, Second Edition: Substructure Design

CL Brg

CL Brg
Gap for temperature
movement
Superstructure

Edge distance

½ support width

Support width
Front face of
abutment Support
width
Seat-type abutment Monolithic abutment

FIGURE 6.7 Abutment support width (seismic).

In practice, the minimum abutment support width may be calculated as shown in Equation 6.4:

N A¢ = (12 + 0.03L + 0.12H )(1 + 0.002 S 2 ) (6.4)

where
NA′ = Abutment support width, (in)
L = Length, (t), of the bridge deck to the adjacent expansion joint, or to the end of bridge deck.
For single-span bridges L equals the length of the bridge deck.
S = Angle of skew at abutment in degrees
H = Average height, (t), of columns or piers supporting the bridge deck from the abutment to the
adjacent expansion joint, or to the end of the bridge deck
H = 0 for simple span bridges

6.5 Miscellaneous Design Considerations

6.5.1 Abutment Wingwall
Abutment wingwalls act as a retaining structure to retain the abutment backill soil and roadway soil
to slide transversely. Several types of wingwall for highway bridges are shown in Figure 6.8. he wing-
wall design is similar to the retaining wall design as presented in Chapter 10. However, the live load
surcharge needs to be considered in the wingwall design. Table 6.2 lists the live load surcharge for
diferent loading cases. Figure 6.9 shows the design loads for a conventional cantilever wingwall. For
seismic design, the criteria in transverse direction discussed in Section 6.2.3 should be followed. he
bridge wingwalls may be designed to sustain some damages in a major earthquake as long as the bridge
collapse is not predicted.

6.5.2 Abutment Drainage

A drainage system is usually provided for the abutment construction. he drainage system embedded
in the abutment backill soil is designed to reduce the possible buildup of the hydrostatic pressure, to
control the erosion of the roadway embankment, and to reduce the possibility of soil liquefaction during
Abutments 143

Cantilever wingwall Simple support wingwall

Construction joint

Continuous support wingwall

FIGURE 6.8 Typical wingwalls.

TABLE 6.2 Live Load Surcharges for Wingwall Design

Load Case Equivalent Soil—Height
Highway truck loading 2.0 t (610 mm)
Rail loading E-60 7.5 t (2290 mm)
Rail loading E-70 8.75 t (2670 mm)
Rail loading E-80 10.0 t (3050 mm)

A
S

X
h

S = Surcharge, Ft.
y

h = End height, Ft.

H = Section height, Ft.
H

L = Length, Ft.
P W=
= Equivalent fluid earth pressure

A
2
MAA= WL [3h2 + (H + 4S)(H + 2h)]
24

P = WL [H 2 + (h+ H)(h+ 3S)]

6
MAA
X=
P

FIGURE 6.9 Design loading for cantilever wingwall.

144 Bridge Engineering Handbook, Second Edition: Substructure Design

Pervious backfill
material continous
behind abutment

Weep hole

Pervious backfill material Abutment

in a nonwoven slope
filter fabric

Drainage pipe may be used to instead

weep hole at abutment back

FIGURE 6.10 Typical abutment drainage system.

an earthquake. For the concrete-paved abutment slope, the drainage system also needs to be provided
under the pavement. he drainage system may include the pervious materials, the PSP or PVC pipes,
the wipe holes, and so on. Figure 6.10 shows a typical drainage system for highway bridge construction.

6.5.3 Abutment Slope Protection

he low water scoring may severely damage the bridge structures by washing out the bridge abutment
support soil. To reduce the water scoring damage to the bridge abutment, the pile support, rock slope
protection, concrete slope paving, and gunnit cement slope paving may be used. Figure 6.11 shows the
rock slope protection and the concrete slope paving protection for bridge abutment. he stability of the
rock and concrete slope protection should be considered in the design. An enlarged block is usually
designed at the toe of the protections.

6.5.4 Miscellaneous Details

Some details related to the abutment design are given in Figure 6.12. Although they are only for the
regular bridge construction situations, those details presented valuable references to bridge designers.

6.6 Design Example

6.6.1 Design Data
A prestressed concrete box girder bridge with 5° skew is proposed overcrossing a busy freeway as shown
in Figure 6.13. Based on the roadway requirement, geotechnical information, and the details men-
tioned earlier, an open-end, seat-type abutment is selected. he abutment in transverse direction is 89 t
(27.13 m) wide. Other abutment design information is listed as follows:
Abutment design loads (with load factors)
Factored superstructure dead load = 23.0 kips/t width (γp = 1.25)
Factored normal vehicular load = 8.1 kips/t width (γLL = 1.75)
Special truck vertical load = 11.2 kips/t width (γLL = 1.35)
Maximum bearing pad capacity = 4.6 kips/t width (γ TU,CR,SH = 1.25)
Abutments 145

1.5:1

Berm 2:1
OG.

5'–0''
min ''
2'–6
5'–0''

Rock slope protection

Abutment
6'' min

face
Match roadway
side slope

#4 Continuous
in
m
3''

4'–
0''
#4 @12''
Welded wire fabric
Formed edge
7'' 1'–0''

#4 tot 4

1'–0'' 1'–0''

Concrete slope protection

FIGURE 6.11 Typical abutment slope protections.

915 mm, min

5'– 0'' min

3'– 0'' min

Soil cover
3'– 0'' min
5'– 0'' min

Top of slope

1:1
m
5'– 0'' min ax

FIGURE 6.12 Abutment design miscellaneous details.

146 Bridge Engineering Handbook, Second Edition: Substructure Design

295'–3"
BB EB
147'–7 1/2" 147'–7 1/2"
Mounted sign typ

17'–6" min

Vertical
Full slope paving 1 2 Full slope paving
Abutment 1 HS & FS Abutment 3
Approximate OG along
Bent 2
right edge of bridge

FIGURE 6.13 Bridge elevation (example).

Longitudinal seismic load

Transverse seismic load = 1241 kips
Bridge temperature displacement = 2.0 in
Maximum bridge seismic displacement = 6.5 in (with abutment releases)
Geotechnical information
Live load surcharge = 2 t
Unit weight of backill soil = 120 pcf
Nominal bearing resistance = 6.0 ksf (with resistant factor of 0.5)
Soil lateral pressure coeicient (Ka) = 0.3
Friction coeicient = tan 33°
Soil liquefaction potential = very low
Ground acceleration = 0.3 g
Design criteria
AASHTO LRFD bridge design speciications, customary U.S. units, 2012.
Design assumptions
• he soil passive pressure at abutment toe is neglected
• One feet width of abutment is used in the design
• Reinforcement yield stress, fy = 60,000 psi
• Concrete strength, fc′ = 4000 psi
• Abutment backwall allowed damages in design earthquake

6.6.2 Abutment Support Width Design

Applying Equation 6.4 with L = 295.25 t, H = 21.3 t, and S = 5°.
he support width will be NA′ = 23.6 in. Add 3 in required temperature movement, the total required
support width equals to 26.5 in. he required minimum support width for seismic case equals to the
sum of bridge seismic displacement, bridge temperature displacement, and the reserved edge displace-
ment (usually 4 in). In this example, this requirement equals to 14 in, not in control. Based on the
26.5 in minimum requirement, the design uses 30 in, OK. A preliminary abutment coniguration is
shown in Figure 6.14 based on the given information and calculated support width.

6.6.3 Abutment Stability Check

Figure 6.15 shows the abutment force diagram.
where
qsc = Soil lateral pressure by live load surcharge (γLS = 1.75)
qe = Soil lateral pressure (γEH = 1.5)
qeq = Soil lateral pressure by seismic load (γEQ = 1.0)
Abutments 147

CL Brg = CL Footing

1'– 3"
20'– 0" 3'– 6"

3'– 0" 1'– 0"

3'– 0" 6'– 0"

3"

min
13'– 0"
3'– 9" 4'– 8"

2'– 6"
12'– 0"

FIGURE 6.14 Abutment coniguration (example).

CL Brg = CL Footing

qeq qsc

PDL, PHL – 93, Pp

w1
F, Feq

w4

w2
w5

w3
A
qe

FIGURE 6.15 Abutment applied forces diagram (example).

P DL = Superstructure dead load

P HL–93 = Normal vehicular truck load
P P = Special truck load
F = Maximum bearing pad load with factor of 1.25
Feq = Maximum bearing pad load with factor of 1.0
Pac = Resultant of active seismic soil lateral pressure
hsc = Height of live load surcharge = 2′–0″
ω = Unit weight of soil
Wi = Weight of abutment component and soil block
qsc = ka × ω × hsc = 0.3 × 0.12 × 2 × 1.75 = 0.126 ksf/t width
qe = ka × ω × H = 0.3 × 0.12 × 15.5 × 1.5 = 0.84 ksf/t width
qeq = kae × ω × H = 0.032 × 0.12 × 15.5 × 1.0 = 0.06 ksf/t width
he calculated vertical loads, lateral loads, and moment about point A are listed in Table 6.3.
he maximum and minimum soil pressure at abutment footing are calculated by

p=
∑Vi  1 ± 6e  (6.5)
 
B  B
148 Bridge Engineering Handbook, Second Edition: Substructure Design

TABLE 6.3 Vertical Forces, Lateral Forces, and Moment about Point A (Example)
Load Description Vertical Load (kip) Lateral Load (kip) Arm to A (t) Moment to A (kip-t)
Backwall W1 0.94 — 7.75 7.28
Stem W2 3.54 — 6.00 23.01
Footing W3 4.50 — 6.00 27.00
Backill soil 5.85 — 10.13 59.23
Backill soil — 4.33 5.17 –22.34
Soil surcharge — 1.16 7.75 –8.65
Front soil W4 1.71 — 2.38 4.06
Keys 0.85 — 16.12 13.70
PDL 0.17 — 6.00 1.04
PHS 18.13 — 6.00 27.64
PP 3.15 — 6.00 18.90
F — 2.79 9.25 –25.80
Feq — 3.66 9.25 –33.90
Soil seismic load — 0.47 9.30 –4.37

where p = Soil bearing pressure

Vi = Vertical force
B = Abutment footing width
e = Eccentricity of resultant of forces to center of footing
Mi = Moment to center of base

e=
∑ Mi (6.6)
Vi

Referring to Table 6.3 and Equations 6.5 and 6.6, the maximum and minimum soil pressures under
footing corresponding to diferent load cases are calculated as follows:

pmax pmin Nominal Bearing Resistance

Load case (ksi) (ksi) (ksi) Evaluate
Strength I 5.72 3.46 6.00 OK
Strength II 5.98 2.10 6.00 OK
Extreme Event 4.97 1.35 6.00 OK

Check for the stability resisting the overturning (load case Strength I, III, and Extreme Event):

Eccentricity of resultant from ¼ of the base width from

Load Case center (t) center (t) Evaluation
Strength I 1.02 3.0 OK
Strength III 0.96 3.0 OK
Extreme Event 1.13 3.0 OK

Check for the stability resisting the sliding (load case Strength I, III and Extreme Event):

Load Case Factored Driving Force (kips) Factored Nominal Resistance (kips) Evaluation
Strength I 12.77 20.98 OK
Strength III 12.77 22.41 OK
Extreme Event 12.30 21.55 OK
Abutments 149

Resistance Factor of 0.8 applied for Strength I and III cases. Resistance Factor of 1.0 for Extreme Event
limit state.

6.6.4 Abutment Backwall and Stem Design

Referring to AASHTO load combinations (AASHTO 2012), the maximum factored loads for abutment
backwall and stem design are as follows:

Factored Factored
Location Load Cases Vu (kips) Mu (k-t)
Backwall level Strength I 1.95 4.87
Strength III 1.95 4.87
Extreme Event 2.25 5.89
Bottom of stem Strength I 11.26 63.47
Strength III 11.26 63.47
Extreme Event 11.70 67.09
Footing Bot All cases 19.40 61.16
Footing Top All cases 10.07 22.40

• Footing was modeled as a cantilever supported at stem.

• Maximum bearing combination was applied to design footing bottom reinforcing.

6.6.5 Abutment Backwall Design

Try #5 @ 12 with 2 in clearance on both faces

d = 9.7 in; β1 = 0.85

As f y = ( 0.31)( 60 ) = 18.6 kips ; As′ f y′ = 0

β1 = 0.85; bw = 12.0 in

As f s − As′ f s′ 18.6
c= = = 0.54 in
0.85 fc′β1bw ( 0.85 )( 4.0 )( 0.85 )(12 )

a = β1c = ( 0.85 )( 0.54 ) = 0.46 in

  a  a 
M r = φM n = φ  As f y  ds −  − As′ f s  ds′ −  
  2  2 
  0.46  
0.9 (18.6 )  9.7 −  /12 = 13.2 kip-ft
  2  

Check for shear, simpliied procedure was utilized with β = 2.0, hence

(
Vr = φVn = ( 0.9 ) Vc + Vs + Vp )
Vc = 0.0316β fc′bv dc = ( 0.0316 )( 2.0 ) 4 (12 )( 9.7 − 2.4 ) = 11.07 kip
150 Bridge Engineering Handbook, Second Edition: Substructure Design

0.5φVc = ( 0.5 )( 0.9 )(11.07 ) = 4.98 kips

Since

1.33 M u = (1.33)(5.98 ) = 7.89 kips-ft < M r = 11.07 kips-ft, M cr does not control.

and

0.5φVc = 4.98 kips > 2.25 kips, no shear reinforcement needed.

6.6.6 Abutment Stem Design

Abutment stem could be design based on the applying moment variations along the abutment wall
height. Here only the section at the bottom of stem is designed. Try using #7 @ 12 in at the back face of
the stem and applying the same procedure as for abutment backwall, the results are as follows:

Max. Factored Load Resistance Capacity Evaluation

Mu (k-t) Vu (kips) Mr (k-t) Vr (kips) Moment Shear
67.09 11.7 104 55.0 OK OK

At the front face of stem, using # 5 @ 12 in. in both longitudinal and horizontal direction, it meets the
crack control requirements of the AASHTO speciications.

6.6.7 Abutment Footing Design

Footing will be modeled as a cantilever structure component supported at abutment stem. For the
design of footing bottom reinforcing, the controlling factored maximum and minimum soil bearing
pressures under the abutment footing are shown in Figure 6.16.

Abutment
face
7'–3" 4'–9"

b
2'–2"
2'–6"

qmin

qmax
a b

FIGURE 6.16 Bearing pressure under abutment footing (example).

Abutments 151

he maximum factored demand moment, in all load cases, at Section a-a (design for top lexural
reinforcement):

qa−a = 4.825 ksf/ft wide

M a−a = 61.163 k − ft/ft wide

he maximum factored demand shear, in all load cases, at Section b-b (d = 30 – 3-1 = 26 in from
Section a-a, design for shear reinforcement):

qb− b = 5.23 ksf/ft wide

Vb− b = 14.9 kips/ft wide

For the design of footing top reinforcing, the design model is shown in Figure 6.17.
he maximum factored demand moment, in all load cases, at Section c-c (design for bottom lexural
reinforcement):

qc−c = 7.05 ksf/ft wide

M c−c = 49.57 k − ft/ft wide

he maximum factored demand shear, in all load cases, at Section d-d (d = 30 – 3-1 = 26 in from
Section a-a, design for shear reinforcement):

qd−d = 7.05 ksf/ft wide

Vd−d = 11.14 kips/ft wide

Abutment face

7'–3" 4'–9"

d c

2'–2"

q
2'–6"

d c

FIGURE 6.17 Footing top analysis model (example).

152 Bridge Engineering Handbook, Second Edition: Substructure Design

Try using #8 @ 12, with 3 in clearance at footing bottom, and # 6 @ 12, with 3 in clearance at footing
top. Following the same procedure as for abutment back wall, the factored footing resistance and evalu-
ation results are shown as follows:

Location Max. Factored Load Resistance Capacity Evaluation

Mu (k-t) Vu (kips) Mr (k-t) Vr (kips) Moment Shear
Footing Top 61.16 14.90 90.00 15.97 OK OK
Footing Bot 49.57 11.14 70.52 15.97 OK OK

Here

Vc = 0.0316β fc′bv dv = ( 0.0316 )( 2.0 ) 4 (12 )( 30 − 6.5 ) = 35.64 kips

Since

0.5φVc = 15.97 kips > 14.9 kips No shear reinforcing needed.

and 1.33Mu = 1.33 × 61.16 = 81.34 k-t < Mr = 90.0 k-t

1.33Mu = 1.33 × 49.57 = 65.92 k-t < Mr = 7.052 k-t—Mcr does not control

6.6.8 Abutment Wingwall Design

he geometry of wingwall is
h = 3.0 t; S = 2.0 t;
H = 13.0 t; L = 18.25 t
Referring to Figure 6.15, the factored design load efects at the cantilever support are

Vu = λ EH {
wL 2
6
[ H + (h + H )(h + 3S )] }
0.36 × 18.25 2
= 1.35 ×
6
[13 + (3 + 13)(3 + 3 × 2)]= 46.0 kips
wL2
M u = λ EH
24
[3h2 + ( H + 4S )( H + 2h)]
0.036 × 18.252
= 1.35 × 3( 3)2 + (13 + 4 + 2 )(12 + 2 × 3) = 255.0 k ⋅ ft
24  

Design lexural reinforcing. Try use # 8 @ 12 in at the inside face of the wingwall
Assume fs = fy

As f s = As f y = 13(0.79)( 60 ) = 616.2 kips

As f y 616.2
c= = 2
= 1.37 in < 2.0 in (clearance)
0.85β1 fc′b (0.85) (4.0)(13)(12)

then in the section, using εc = 0.003, the strain in extreme tension steel εt = 0.0178 > 0.005. he section is
tension-controlled, the assumption applied. Also there is no reinforcing in the compression zone. hen
with d = 12–2.0–0.5 = 9.5 in and

a = β1c = ( 0.85 )(1.37 ) = 1.165 in

Abutments 153

  a    1.165  
M r = φM n = φ  As f y  ds −   = 0.9 ( 616.2 )  9.5 −  /12 = 412.12 kip-ft
  2     2  

Since 1.33 × Mu = 1.33 × 255 = 339.15 k-t < Mr OK. No need to check Mcr.
Check for shear. Since

 1.165 
Vc = 0.0316β1 fc′bv dv = 0.0316 ( 2 ) 4 (12 )  9.5 −  = 175 kips
 2 
and
0.5φVc = 0.5 ( 0.9 )(175 ) = 78.75 kips > Vu = 46.0 kips, no shear reinforcing needed.

Since the wingwall allows to be broken of in a major earthquake, the adjacent columns of the bridge
have to be designed to sustain the seismic loading with no wingwall resistant exist.
he abutment section, footing and wingwall reinforcing details are shown in Figure 6.18.

CL Brg = CL Footing

#4 @ 12

#5 @ 12
#6 × 10'–0" @ 12
#7 @ 12
#6 @ 12

12'–0"
#6 tot 5 #8 tot 3 @ abutment ends only
#6 @ 12

#5 tot 12 2" clr

#5 tot 13 typ #6 @ 12
top & bot
#8 @ 12
3" clr

(a)

CL Brg = CL Footing

#5 @ 1'–6" typ #10 @ top

tot 2

3" clr
typ lr
c
2"

#5 tot 2

#8 @ 12 inside face
#5 @ 12 outside face

(b)

FIGURE 6.18 Abutment reinforcement details (example). (a) Abutment-typica section, (b) Wingwall reinforcement.
154 Bridge Engineering Handbook, Second Edition: Substructure Design

References
AASHTO. 2002. Standard Speciications for Highway Bridges, 17th ed., American Association of State
Highway and Transportation Oicials, Washington, DC.
AASHTO. 2012. AASHTO LRFD Bridge Design Speciications, Customary US Units, 2012, American
Association of State Highway and Transportation Oicials, Washington, DC.
Caltrans. 1996. Bridge Memo to Designers 5-1 Abutments, California Department of California
Transportation, Sacramento, CA.
Caltrans. 2010. Seismic Design Criteria, Version 1.6, California Department of Transportation,
Sacramento, CA.
Goel, R. K. 1997. Earthquake behavior of bridge with integral abutment, in Proceeding of the National
Seismic Conference on Bridges and Highways, July, Sacramento, CA.
Maroney, B. H. and Chai, Y. H. 1994. Bridge abutment stifness and strength under earthquake loadings,
in Proceedings of the Second International Workshop of Seismic Design and Retroitting of Reinforced
Concrete Bridges, August,