Structural Engineering Documents

Günter Ramberger

Structural Bearings and Expansion Joints

for Bridges

International Association for Bridge and Structural Engineering Association Internationale des Ponts et Charpentes

Internationale Vereinigung für Brückenbau und Hochbau

IABSE AIPC IVBH

About the Author

Günter RAMBERGER

Born 1942 in Aspang, Austria, Günter Ram-berger received his civil engineering degree from the Technical University, Vienna, in 1966. He worked as an assistant to Prof. Dr.-lng. Peter Stein, Institute for Steel Structures, f rom 1967 to 1969 and received his doctor's degree in 1970 with a thesis on orthotropic plates. In 1970 he joined Hein, Leh-mann AG, Düsseldorf, Germany, where he worked in the field of steel bridges, finally, as head of this department. He was involved in design, fabrication and erection of the fol lowing steel bridges: Oberkas-seler Brücke, Düsseldorf, Franklinbrücke, Düssel-dorf, Süderelbebrücke, Hamburg, Hammer-brookbrucke, Hamburg, Hochbrücke, Brunsbüttel, and many others. Since 1981 he has been profes-sor of Steel Structures at the Technical University, Vienna, and was Dean of the Faculty for Civil Engi­neering f rom 1984 to 1987. He has been a member of the Working Commission 2 of IABSE and is a member of sev-eral committees for the standardization of steel structures (CEN TC 250/SC3, ON).

Structural Engineering Documents

6

Günter Ramberger

Structural Bearings and Expansion Joints

for Bridges

International Association for Bridge and Structural Engineering Association Internationale des Ponts et Charpentes

Internationale Vereinigung für Brückenbau und Hochbau

1

IABSE AIPC IVBH

Copyright © 2002 by International Association for Bridge and Structural Engineering

All rights reserved. No part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording, or by any information storage and retrieval system, without permission in writing from the publisher.

ISBN 3-85748-105-6 Printed in Switzerland

Publisher: IABSE-AIPC-IVBH ETH Hönggerberg CH-8093 Zurich, Switzerland

Phone: Int.+ 41-1-633 2647 Fax: Int.+ 41-1-633 1241 E-mail: secretariat@iabse.ethz.ch Web: http://www.iabse.ethz.ch

2

Dedicated to the commemoration of the late Prof. Dr. techn. Ferdinand Tschemmernegg, University of Innsbruck.

Preface

It is my hope that this treatise will serve as a textbook for students and as information for civil engineers involved in bridge construction. My intent was to give a short guideline on bearings and expansion joints for bridge designers and not to mention all the requirements for the manufacturers of such products. These requirements are usually covered by product guidelines, which vary between different countries.

Not all the references are related to the content of this document. They are more or less a collection of relevant papers sometimes dealing with special problems.

I express many thanks to Prof. Dr.-Ing. Ulrike Kuhlmann, University of Stuttgart, chairperson of Working Commission 2 of IABSE, who gave the impetus for this work; to her predecessor of the IABSE Commission, Prof. Dr. David A. Nethercot, Imperial College of Science, Technology and Medicine, London, for reviewing the manuscript, and Prof. Dr. Manfred Hirt, Swiss Federal Institute of Technology, Lausanne, for his contributions and comments.

I wish to thank J. S. Leendertz, Rijkswaterstaat, Zoetermeer; Eugen Brühwiler, Swiss Federal Institute of Technology, Lausanne; Prof. R. J. Dexter, University of Minneso­ta; G. Wolff, Reissner & Wolff, Wels; O. Schimetta †, Amt der OÖ Landesregierung, Linz; Prof. B. Johannsson, Lulea Tekniska Universitet, for amendments, corrections, remarks and comments. I thank also my assistant Dipl.-Ing. Jörgen Robra for his valuable contributions to the paper, especially for the sketches and drawings, and my secretaries Ulla Samm and Barbara Bastian for their expert typing of the manuscript. Finally, I would like to thank the IABSE for the publication of this Structural Engi­neering Document.

Vienna, April 2002 Günter Ramberger

3

Page Blank

In Original

4

Table of Contents

1. Bearings

1.1 Introduction 1.2 The role of bearings 1.3 General types of bearings and their movements 1.4 The layout of bearings 1.5 Calculation of bearing reactions and bearing movements 1.6 Construction of bearings 1.7 Materials for bearings 1.8 Analysis and design of bearings 1.9 Installation of bearings

1.10 Inspection and maintenance 1.11 Replacement of bearings 1.12 Codes and standards 1.13 References

2. Expansion Joints

2.1 Introduction 2.2 The role of expansion joints 2.3 Calculation of movements of expansion joints 2.4 Construction of expansion joints 2.5 Materials for expansion joints 2.6 Analysis and design of expansion joints 2.7 Installation of expansion joints 2.8 Inspection and maintenance 2.9 Replacement of expansion joints

2.10 References

5

7 7 7 9 16 19 29 33 37 38 39 41 42

51 51 51 58 70 72 84 86 87 88

Page Blank

In Original

6

7

1 Bearings

1.1 Introduction

All bridges are subjected to movements due to temperature expansion and elastic strains induced by various forces, especially due to traffic loads. In former times our bridges were built of stones, bricks or timber. Obviously, elongation and shortening occurred in those bridges, but the temperature gradients were small due to the high mass of the stone bridges. Timber bridges were small or had natural joints, so that the full elongation values were subdivided into the elongation of each part. On the other hand, the elongation and shortening of timber bridges due to change of moisture is of­ten higher than that due to thermal actions. With the use of constructional steel and, later on, of reinforced and prestressed concrete, bridge bearings had to be used. The first bearings were rocker and roller bearings made of steel. Numerous rocker and roller bearings have operated effectively for more than a century. With the develop­ment of ageing-, ozone- and UV-radiation-resistant elastomers and plastics, new ma­terials for bearings became available. Various types of bearings were developed with the advantage of an area load transmission in contrast to steel bearings with linear or point load transmission, where elastic analysis leads theoretically to infinite compres­sion stresses. For the bearings the problems of motion in every direction and of load transmission were solved, but the problem of insufficient durability still exists. Whilst it is reasonable to assume the life of steel bearings to be the same as that of the bridge, the life of a bearing with elastomer or plastic parts can be shorter.

1.2 The role of bearings

The role of bearings is to transfer the bearing reaction from the superstructure to the substructure, fulfilling the design requirements concerning forces, displacements and rotations. The bearings should allow the displacements and rotations as required by the structural analysis with very low resistance during the whole lifetime. Thus, the bearings should withstand all external forces, thermal actions, air moisture changes and weather conditions of the region.

1.3 General types of bearings and their movements

Normally, reaction forces and the corresponding movements follow a dual principle -a non zero bearing force corresponds to a zero movement and vice versa. An exception is given only by friction forces which are nearly constant during the movement, and by elastic restraint forces which are generally proportional to the displacement. Usually, the bearing forces are divided into vertical and horizontal components. Bearings for vertical forces normally allow rotations in one direction, some types in all directions. If they also transmit horizontal forces, usually vertical forces are com­bined.

8 1. Bearings

A special type of bearing transmits only horizontal forces, while allowing vertical displacements.

The following table (Table 1.3-1) shows the common types of bearings, including the possible bearing forces and displacements. Friction and elastic restraint forces are not considered.

Nr Symbol Function Construction A Hx H y w ex ey

Mx M y Mz φx

φv φ,

X X X 0 0 0 0 0 0 X X X

Point rocker bearing Pot bearing; Fixed elastomeric bearing; Spherical bearing

All translation fixed Rotation all round

1

2 Horizontal movement in one direction Rotation all around

Constr. point rocker sliding bearing; Constr. pot sliding bearing; Const. elastomeric bearing; Constr. spherical sliding bearing

X X

0

0

X 0

0

X

X

0 0 0 0 X X X

X X X 0 0 0 X X 0 0 0 X

Free point rocker bearing; Free pot sliding bearing; Free elastomeric bearing; Free spherical sliding bearing; Link bearing with universal joints (tension and compression)

Horizontal movement in all directions Rotation all round

3

4

5

All translation fixed Rotation about one axis

Line rocker bearing Leaf bearing (tension and compression)

X X X 0 0 0 X

0

0

X (0)

0

X

X

0 (X)

(X) 0

X

X

0 (0)

X

0

0

X

X

0

0

X 0

0

X

X

0 X

Roller bearing; Link bearing (tension and compression); Constant line rocker sliding bearing

Horizontal movement in one direction Rotation about one axis

6 Horizontal movement in all direction Rotation about one axis

Free rocker sliding bearing; Free roller bearing; Free link bearing

X 0 0 0 X X 0

X

X

0 0

X

0

0

X X

X X X 0 0 0 0 0 X X X 0 Horizontal force bearing

All horizontal transl. fixed Rotation all round

7

8 Horizontal movement in one direction Rotation all round

Guide bearing

0 X

0

0

X X

0

X

X

0 0 0 0 X X X

1.4 The layout of bearings 9

Table 1.3-1

1.4 The layout of bearings

1.4.1 General Bearings can be arranged at abutments and piers (fig.1.4.1-1; fig.1.4.1-2) under the webs of the main girders, under diaphragms (fig.1.4.1-3), and under the nodes of truss bracings. The webs and the diaphragms of concrete bridges have to be properly reinforced against tensile splitting; steel bridges need stiffeners in the direction of the bearing reactions to transfer the concentrated bearing loads to the superstructure and the substructure. Abutments and piers also have to be properly reinforced under the bearings against tensile splitting.

Fig. 1.4.1-1: Bearings at an abutment

Fig. 1.4.1-2: Bearings at a pier

Fig. 1.4.1-3: Bearing at a single pier

10 1. Bearings

The layout of the bearings should correspond to the structural analysis of the whole structure (super- and substructure together!). If the settlement and the deflection of the substructure can be neglected the structural analysis of the superstructure, including the bearings, can be separated from that of the substructure. Sometimes the model for the analysis, especially of the superstructure, will be simplified by assuming the fol­lowing: bearings are situated directly on the neutral axis of the girder (fig.1.4.1-6), the motion of the bearings occurs without restraint, bearings have no clearance, etc. In this case we must consider the correct system (fig.1.4.1-5) at least for the design of the bearings and take into account the influence of the simplifications on the structure.

Fig.1.4.1-4: Reality

Fig. 1.4.1-5: Correct system

Fig.1.4.1-6: Simplified system

On the abutments or separating piers it is normal to use at least two vertical bearings to avoid torsional rotations. At intermediate piers one or more vertical bearings may be used. If more than one bearing is used the rotational displacement at the pier is re­strained. More than three vertical supports of the superstructure lead to statically-in-determinate bearing conditions, but even the simplest bridge has at least four vertical bearings. If the torsional stiffness of the superstructure is low (e.g. open cross sec-tions) it may be neglected and the layout with four bearings becomes isostatic. If the torsional stiffness is not negligible (e.g. box girders) we have to take it into account for the structural analysis, especially for skewed and curved bridges. On a bridge with n > 3 vertical supports, n - 3 bearing reactions can be chosen freely within a reasonable bandwidth. This possibility can be used to prestress the superstructure and to distri­bute the bearing reactions as desired.

If the bearings are situated (nearly) in a plane we need at least one horizontally fixed and one horizontally moveable bearing. The moving direction must not be orthogonal

1.4 The layout of bearings 11

to the polar line from the fixed to the moveable bearing. If more than two bearings in the horizontal direction are necessary, the basic principle should be that an overall uniform extension, caused by temperature or shrinkage, shall be possible without restraint.

In general, there are two possibilities for the arrangement of the bearings: a) arrangement in a horizontal position (fig.1.4.1 -7) b) arrangement in a position parallel to the road or rail surface (fig.1.4.1-8).

Fig. 1.4.1-7: Horizontal arrangement of the bearings (case a)

Fig. 1.4.1-8: Inclined arrangement of the bearings (case b)

Case a) has the advantage that only vertical bearing reactions and no permanent hori­zontal reactions result from vertical loads, but it has the disadvantage that bridges with inclined gradients require a step at the expansion joint due to movements in the super­structure. The greater the elongation or shortening, the greater the step required.

Case b) has the advantage that the slope of the expansion joint is independent of the movement of the bridge. The inclination of the surface of support gives the direction of the normal force. Besides vertical reaction forces, also horizontal reaction forces result from vertical loads. Permanent horizontal actions can lead to a displacement by creep of the concrete and the soil and, thus, to crooked piers.

12 1. Bearings

1.4.2 The layout for different types of bridges For single span girders the layout of the bearings is straightforward. One fixed and one moveable bearing is provided on each abutment, all other bearings are just vertical supports, moveable in any horizontal direction. For wide bridges the horizontally fixed bearings are located in or near the bridge axis.

Formerly, the "classical" arrangement of the bearings for a bridge with two main gird­ers consisted of one fixed and one lengthwise moveable bearing at one abutment and one lengthwise moveable and one free bearing at the other abutment (fig.1.4.2-1). This layout has the advantage that longitudinal horizontal forces (braking and traction forces) can be distributed into the two bearings at the abutment, but it has the disadvantage that horizontal forces in the cross direction (wind) and temperature dif­ferences cause horizontal restraint forces, provided that bearings have no clearance on the abutments.

The author prefers the statically determinate system with only one lengthwise re­strained bearing at the abutment concerned because the actual clearance of a bearing is not determinable in reality (fig.1.4.2-2).

Fig.1.4.2-1: "Classical" layout

Fig.1.4.2-2: Horizontally statically determinate system (better than classical layout)

Fig.1.4.2-3: System with separated vertical and horizontal bearings (statically deter­minate system)

1.4 The layout of bearings 13

For skewed or horizontally curved single span bridges we have to decide whether the horizontal force should be combined with the higher or with the lower vertical reac-tion force. For all bearing constructions it is easier to transfer horizontal forces in com-bination with a high vertical force. In this case the resultant force stays nearer to the centre, its angle to the vertical is smaller and leads to smaller bending moments in sub-and superstructure (fig.1.4.2-4).

Fig.1.4.2-4: Inclination of the resultant force

Thus, the horizontally constrained bearings for skewed bridges should be placed at the obtuse corners of the bridge, for curved bridges at the outer side (fig.1.4.2-5).

Fig. 1.4.2-5: Skewed bridge

Fig. 1.4.2-6: Layout for continuous girders

14 1. Bearings

For straight continuous girders normally two bearings are used at every abutment and pier. If the torsional stiffness is high (box girder) the intermediate piers can be reduced to a round column with one bearing on the axis under the diaphragm. Constrained bearings in the cross direction are the rule at all piers. If the horizontal bending stiff-ness is very high we can transfer the horizontal forces only at the abutments. The same considerations are suitable also for skewed and curved bridges (fig.1.4.2-6).

Bearings for horizontal forces and guide bearings which transfer only horizontal forces may be used in combination with leaf or link bearings which cannot transmit horizontal forces.

The movement of an expansion joint must be linked by a guide like a constraint bear­ing. The main movement of an expansion joint should be in the axis of the traffic way. Generally, this direction does not coincide with the direction of the polar line from the fixed bearing to the moveable bearing at the abutment (fig.1.4.2-7). If all other bearings have the same angle between the polar line and the moving direction there results a layout of the bearings with no restraints on uniform elongation or shortening (e.g. caused by thermal actions or shrinkage), as shown below (fig.1.4.2-8).

Fig.1.4.2-7: Layout for curved bridges

Fig.1.4.2-8: Layout for curved continuous girders (no constraint under overall tem­perature)

Fig.1.4.2-9: Geometrical situation

1.4 The layout of bearings

The elongation is

15

k proportional elongation

The rotation is

One special case of this general rule is well known: the bearings are moveable in the direction of the polar lines with α = 0 (fig.1.4.2-10). However, this layout has the disadvantage that generally the main movement of the joint does not coincide with the movement of the bearing.

Fig. 1.4.2-10: Special case with a = 0

1.4.3 Special bearing conditions, advice etc. It is important to note that the layout of the bearings has a great influence on the struc­tural system. The above mentioned arrangements of bearings are typical for average bridges. The following examples show some special effects which have to be consid­ered for the design of bridges and bearings. These examples do not lay claim to com­pleteness.

a) The already mentioned bearing layout, consisting of one bearing fixed in all sliding directions and one fixed lengthwise at one abutment, leads to high constraint forces not only under horizontal but also under eccentric vertical loading (fig.1.4.3-1). It is interesting that this eccentric loading has no prying effect if the bearings are situated directly on the neutral axis of the girder. This effect results only from the (small) eccentricity of the bearing under the lower flange.

For Ф1 = Ф, the bridge simply rotates as a rigid body without constraint.

16 1. Bearings

Fig.1.4.3-1: Prying effect due to a eccentric loading

b) A similar situation occurs for a continuous girder with chequer pattern loading.

Fig.1.4.3-2: Prying effect due to chequer pattern loading

c) It is not generally known that a skewed bridge with horizontally fixed bearings only in one line exhibits the same effect under vertical loading, as the following figure shows:

Fig. 1.4.3-3: Prying forces for a skewed bridge with vertical loading

Similar effects can occur for curved bridges. For the correct analysis of the bearing reactions it is always necessary to model the bearings at the very point where they are actually situated, and in combination with the substructure. The deflection of the substructure can influence the constraint bearing reactions significantly.

1.5 Calculation of bearing reactions and bearing movements

1.5.1 Actions According to Eurocode 1 (ENV 1991) the actions can be subdivided into:

- permanent actions, - variable actions, - extraordinary actions.

1.5 Calculation of bearing reactions and bearing movements 17

The bridge should take up the desired shape under all permanent loads, at the average temperature (+10°C in most of the European countries) and, if time-dependant displacements occur, at the time t = °°, at which time all moveable bearings should be in the zero adjustment (null position). Variable actions and extraordinary actions lead to deviation from this form.

Variable actions to consider are: - traffic loads, considering the applicable dynamic coefficients - loads due to traffic loads, i.e.

nosing forces centrifugal forces braking forces traction forces

- wind loads wind on construction wind on traffic loads

- settlements of abutments and piers - thermal actions

uniform temperature vertical temperature gradient horizontal temperature gradient temperature differences between individual parts of the bridge (e.g. stay cables, pylon and stiffening girder)

- creep and shrinkage of concrete Extraordinary actions to consider are:

- earthquake actions - vehicle impact - derailment - rupture of the conductor line others

1.5.2 Bearing reactions For permanent actions such as self-weight of the construction, dead load and pre-stressing, the bearing reactions can be calculated as one load case. For the analysis of the bearings it is necessary to consider different combinations of the bearing reactions: - maximum vertical force and the adjacent horizontal force, - minimum vertical force and the adjacent maximum horizontal force, - maximum horizontal force and the adjacent maximum vertical force, - maximum horizontal force and the adjacent minimum vertical force. The simplest way to obtain these combinations is to calculate the variable actions, es­pecially the traffic load, according to the influence line. One should bear in mind that horizontal actions such as centrifugal forces or braking forces are proportional to the vertical traffic load, but other loads, such as wind or traffic or traction forces for rail­ways, are not.

18 1. Bearings

To obtain the extreme bearing reaction it is necessary to consider that all bridges are three-dimensional and not merely plane systems. The influence lines (influence surfaces) of the bearing reactions can be found as the displacement curves (displacement surfaces) of the system, due to unit displacements δ = 1 or (φ = 1, acting at the position and in the direction of the required force. If these analyses are performed on a three dimensional model, the definitive influence area will result directly (fig.1.5.2-1; fig.1.5.2-2). If plane models are used for the analyses, special care is necessary, particularly with continuous girders with open or box sec­tions. The following examples demonstrate the difference:

Fig. 1.5.2-1: Influence area for the vertical bearing reaction A, box section.

Fig.1.5.2-2: Influence area for the vertical bearing reaction A, open section.

1.5.3 Bearing displacements As already mentioned, the zero adjustment (null position) of every bearing has to be defined. The displacements are measured from that position. Thus, for concrete and composite bridges it is usual to consider displacements under time-dependent actions such as creep and shrinkage from the time of installation of the bearing to the time de­fined for the null position (normally t = °°), from which position the displacements due to variable actions are measured.

To obtain the maximum displacements and rotations, again we can use influence lines. The influence line of a displacement can be calculated as the displacement curve due to the corresponding unit force P = 1.

To take into account the imperfections due to installation, the temperature difference for the calculation of bearing displacements should be assumed higher than for the structural analysis of the bridge, or some additional displacement should be consi­dered.

1.6 Construction of bearings

1.6 Construction of bearings

19

Standard type Combinations

Fig. 1.6-1 gives an overview for the most common bearings.

Reinforced elastomeric bearing

Anchored elast. bearing

Elastomeric bearing with fixing device

all translations fixed movement in one dir.

Fixed bearing

Pot bearing

Uni-directional guided

Free pot sliding bearing

Multi-directional non-guided

Constr. pot sliding bearing

Constr. spherical bearing Fixed spherical bearing Free spherical bearing

Point rocker bearing

Line rocker bearing

Free point rocker bearing

Roller bearing

Constr. point rocker bearing

Link bearing with universal (cardan) joints

Link bearing Leaf bearing

Horizontal force bearing Guide bearing

20 1. Bearings

1.6.1 Elastomeric bearings Elastomeric bearings are the simplest types of bearings. In the basic mode they con­sist merely of an elastomeric block (usually rectangular or round). The elastomeric works as a soft part between sub- and superstructure and allows movements in all di­rections by elastic displacements or rotations. Under vertical loads the elastic block bulges, leading to vertical displacements. A solution to this problem was found by re­inforcing the elastic block by thin horizontal steel plates, vulcanized to the elastomer (fig.1.6.1-1). The reinforcing plates prevent the block from bulging, thus leading to very small vertical displacements, but they do not hinder horizontal displacements in every direction and also allow small rotations in all directions. Every displacement and rotation leads to restraining forces and moments which have to be taken into account on the whole structure.

These restraining forces are possible if the friction between bearing and sub- and su­perstructure is sufficient. The friction forces F depend on the compressive force C and the friction coefficient μ, with F = C μ. If displacements take place under a small compressive force, sliding between bearing and sub- or superstructure can occur. To avoid this it is necessary to use elastomeric bearings with resistance to sliding. This can be achieved by applying vulcanized plates on the bottom and on the top of the bearing, which can be connected to the sub- and superstructure by bolts, pins or ap­propriate shapes (fig.1.6.1-2).

Fig.1.6.1-1: Elastomeric bearing (unanchored)

Smaller, short time, horizontal forces can be transmitted by the restraining forces. If these forces are higher or if they are permanent loads a restraining steel construction is required. In these case the elastomeric bearing transmits the vertical force and allows rotations, while horizontal forces in one or two directions are transmitted by the steel construction (fig.1.6.1-3 ; fig.1.6.1-4).

Fig.1.6.1-2: Elastomeric bearing (anchored)

1.6 Construction of bearings 21

Fig.1.6.1-3: Elastomeric bearing constraint

Combination: elastomeric bearing and steel construction fixed in one direction.

Fig.1.6.1-4: Fixed elastomeric bearing

Combination: elastomeric bearing and steel construction fixed in two directions.

1.6.2 Steel bearings Steel bearings are the oldest type of bearings. They have been used for more than 100 years. The principle is simple: a flat plate rolls on another steel plate with a curved sur-face. If this surface is part of a sphere, theoretically we obtain a point tangency. If this surface is part of a cylinder, theoretically we obtain a linear tangency. In the first case we speak of point rocker bearings, in the second case of line rocker bearings. These bearings allow rotations in all or in one direction, but they do not allow displacements (fig.1.6.2-1; fig.1.6.2-4). Under minimal vertical reactions in combination with horizontal loads point rocker bearings and line rocker bearings can exhibit damage of their connections, because of tension. In combination with sliding elements these bearings are very sensitive to this phenomenon, and it causes partial uplift and excessive wear as a result. Linear tangencies can be found also in roller bearings consisting of a roll and a lower and an upper plate (fig.1.6.2-5). These bearings allow rotations in one direction and displacements in one direction. The problem with these bearings is a point or linear concentration of the bearing force, which theoretically leads to infinite stresses. In 1881, the physicist Heinrich Hertz found the solution of this problem: caused by the elastic deformation the theo­retical point of tangency yields to a circle, the theoretical line of tangency yields to a rectangle. The infinite stresses decrease to high but finite stresses, the so called Hertz compression stresses over a very small contact zone. If the radius of the sphere or of the cylinder decreases the Hertz stresses increase. From the local stress concentration the stresses have to be distributed to the contact zones between bearing and sub- and superstructure. Therefore, steel bearings normally need thicker plates for the stress distribution than other types of bearings which transfer the bearing reactions over an area.

22 1. Bearings

Point rocker bearings are used for bearing reactions in the range 500 and 2500 kN, line rocker bearings and roller bearings for loads in the range 200 and 20 000 kN.

Fig.1.6.2-1: Fixed point rocker bearing

Fig.1.6.2-2: Point rocker bearing constraint in one direction

Fig.1.6.2-3: Free point rocker bearing

Fig.1.6.2-4: Line rocker bearing

1.6 Construction of bearings 23

Fig.1.6.2-5: Roller bearing (left side without guide rail; right side with guide rail)

The contact zones of steel bearings cannot be protected against corrosion. Therefore corrosion-resistant layers of high alloyed steel should be used for the contact areas. This can be done by building up a surface by forging or by welding. Between the mild steel and the hardened high alloyed steel of the surface there should be a welded or forged tough buffer zone. The thickness (in mm) of the hardened layer both on the roller (radius R in mm) and of the plate should be t 0,14 R - 2.

1.6.3 Pot bearings These bearings were invented in the 1950s. They combine the two desirable proper­ties: rotation capacity with a very small resistance and transmission of the bearing reaction over a defined area.

The pot bearing consists of a steel pot, filled with an elastomeric disc and a lid or a piston to the top (fig.1.6.3-1). When subjected to high compression forces, the unrein-forced elastomeric disc behaves similarly to a liquid. Rotations can occur due to the nearly constant volume of the elastomer (v = 0,5). Of great importance is the sealing between the elastomeric pad and the lid: if this sealing has a defect the elastomeric pad escapes like a viscous liquid.

The standard type of pot bearing allows only rotation (fig.1.6.3-2). Vertical forces are transmitted to the pad, horizontal forces from the lid to the pot. To release one sliding direction, an additional construction becomes necessary (fig.1.6.3-3 and fig.1.6.3-5). This sliding construction consists of three components: a polytetrafluorethylene (PTFE) disc, a surface of polished stainless steel connected to a sliding plate of struc­tural steel and lubrication grease. PTFE is a plastic with high mechanical and chemi­cal resistance, great toughness and very small friction when combined with polished stainless steel. The PTFE disc is 5 to 6 mm thick, where half a thickness is enclosed by the lid. This disc has small round pockets on the surface for the lubrication grease (normally silicon grease) to reduce friction and wearing.

To constrain the movement in one direction an additional guide is used for the lid. This guiding device allows movements in only one direction (fig.1.6.3-3). Pot bearings are used for vertical bearing forces from 1000 kN up to 100 000 kN. Depending on the standard applied the allowable compression between lid and elas-

24 1. Bearings

tomeric pad should not exceed 4.0 kN/cm2. The allowable compression for the PTFE is 3 kN/cm2 for permanent loads and 4.5 kN/cm2 for short term loads (traffic, wind etc.). Pot bearings have the advantage of a very high vertical stiffness (nearly incompres­sible elastomeric part). It is comparatively independent of the size of bearing and the applied load. This characteristic is important for the bearing of high velocity railway bridges. Bearings with low vertical stiffness can lead to damage of the rails.

Fig.1.6.3-1: Function of a pot bearing

Fig.1.6.3-2: Fixed pot bearing

Fig.1.6.3-3: Pot bearing constraint in one direction

Elastomere disc

Lid

Sealing

Pot - wall

Pot - bottom

Centre of rotation

1.6 Construction of bearings 25

Anchoring plate Sliding plate Polished stainless steel PTFE (Polytetrafiuorethylen) Lid Pot-wall Sealing Elastomere disc Pot - bottom

Fig.1.6.3-4: Members of a pot bearing

Fig. 1.6.3-5: Free pot bearing

1.6.4 Spherical bearings The basic type of spherical bearing consists of three main parts: the pan, the part of a sphere and the upper plate made of constructional steel (fig.1.6.4-1). To allow dis-placements between the parts, sliding surfaces are necessary. The pan has a PTFE plate on the upper surface, the part of the sphere has a chrome-plated polished surface on the underface and a PTFE plate also on the upper surface, and the upper plate has a polished stainless steel plate on the underface. The PTFE plates are chambered over half the thickness and have lubrication pockets with silicon grease, like the sliding plates for pot bearings. The friction resistance of the sliding parts causes reaction moments due to rotations. They must be taken into account to consider additional design stresses of the bearing material.

The vertical bearing reaction is transferred over the compressed areas of the PTFE. The basic model is a moveable bearing (fig.1.6.4-4). To constrain horizontal displace-ments an additional construction to connect the upper plate with the pan becomes necessary (fig.1.6.4-2; fig.1.6.4-3). British and Italian bearings have one sliding plane only and a deeper concave part to take over horizontal forces (fig.1.6.4-5). The construction must be checked for uplift and exceeding the stresses in the contact area. In the bearings with two sliding planes the centre of rotation is between the contact areas of the sliding surfaces, whereas in Italian and British bearings it is somewhere in the bridge structure or in the pier or the abutment.

26 1. Bearings

Like pot bearings, spherical bearings are used for vertical forces in the range of 1000 to 100 000 kN.

Polished stainless steel

Sliding plate

PTFE

Pan

PTFE

Part of sphere

Chrome plated polished surface

Fig.1.6.4-1: Members of a spherical bearing

Fig.1.6.4-2: Fix spherical bearing

Fig.1.6.4-3: Spherical bearing constraint in one direction

Fig.1.6.4-4: Free spherical bearing

1.6 Construction of bearings 27

Fig.1.6.4-5: Italian and British spherical bearing (one sliding surface)

1.6.5 Leaf and link bearings All the above mentioned bearings are able to transfer compression forces. If tensile forces as well as compressive forces must be transferred, leaf and link bearings are used. These bearings can only transmit forces in the direction of the leaf. To transfer forces in the crosswise direction, separate bearings must be used.

A leaf bearing consists of a foot plate, one or two lower leafs with pin holes and two or one upper leaf with foot plate and pin holes, connected by a pin. Leaf bearings al­low free rotation in one direction. Pin and pin holes must have a fit less than 0.3 mm, as in cases of greater slackness and changing forces the pin will punch the hole. Pin plate and pin should be of different types of steel to avoid seizure. Pin plates are made of structural steel, pins often of tempered steel.

For link bearings a pendulum is linked to the foot leaf and to the upper leaf by pins. Link bearings allow rotation and displacement in one direction. For pin holes and pins the same rules apply as given for leaf bearings.

Link bearings with universal (Cardan) joints are used only in special cases. They allow rotation and displacement in all directions.

Displacements 8 of link bearings are always combined with a small displacement δV

in the perpendicular direction. δV = with R equal to the distance between the

axes of the pins. Therefore this distance should not be too small.

1.6.6 Disc bearings Disc bearings were introduced in the late 1960s. The vertical loads are transferred by an elastomeric disc made of polyether-urethane polymer. In contrast to a pot bearing a transverse extension of the elastomeric disc is possible. Bearing capacity and func­tioning is comparable with an elastomeric bearing. Rotations around the horizontal axis are transferred by differential deflection of the disc. The rotations cause a shift of the axis of the load from the centre of bearing, which must be considered in the design. Horizontal forces are transferred by a shear-restriction device which allows vertical deformation and rotation. The basic type is a fixed bearing. Free bearings are con­structed by additional sliding elements and (if necessary) guiding systems.

28 1. Bearings

Fig.1.6.6-1: Fixed bearing

Fig.1.6.6-2: Uni-directional guided

Fig.1.6.6-3: Multi-directional non-guided

top plate

bearing assembly

base plate

top plate

bearing assembly

base plate

top plate

bearing assembly

base plate

1.7 Materials for bearings

1.7 Materials for bearings

29

1.7.1 Steel Structural steel Structural steel is used for all parts of bearings which are not under extraordinary local stress or do not require special properties against corrosion. Structural steel for bearings can be: - Non-alloy structural steels according to EN 10025 - Fine-grained structural steels according to EN 10113 - Quenched and tempered steels according to EN 10082

Eurocode 3 may be used for the design of all bearing components made from struc­tural steel according to EN 10025 and EN 10113 and for all connections (bolts, welds etc.). Quenched and tempered steels are used mostly for non-welded parts under high pressure (parts with Hertz compression, bolts of leaf and link bearings). In contact areas with Hertz compression layers of corrosion-resistant hard steel can be applied by forging or by welding. In the case of hard-surface welding a tough intermediate (puffer) layer must be welded between the steel and the hard-surface.

Stainless steel Stainless steel according to EURONORM 88-2 or ISO 683 can also be used for bear­ings. For design one should use EC 3, part 1-4. Concerning stainless steel for sliding plates see 1.7.3.

1.7.2 Elastomeric parts Elastomeric parts of bearings consist normally of natural or artificial (chloropren) rub-ber (NR or CR, respectively). Artificial rubber has the same good properties as natu-ral rubber, and in addition it has a higher resistance against ozone, ultraviolet radiation and ageing and is more rigid. The characteristic mechanical property is the shear modu-lus G between 0.7 and 1.15 N/mm2 at room temperature, decreasing with increasing temperature. When undergoing stress changes the volume of rubber is nearly constant. So we have a Poisson's ratio v = 0.5 and a Young's modulus of elasticity E = 2 (1+v) G 3 G. The fracture strain of rubber lies between 250 % and 500 %. Rub-ber creeps under stress by up to 50 % of the elastic strain, but creeping ends within some days or weeks. Rubber does not break under compression, it can only break under tensile or shear stresses. Compressing a rubber pad changes its shape. The changing of the shape depends on the possibility of displacement at the compressed areas. If the compressed areas are fixed to a rigid surface, the displacement remains small. Thus we obtain the inequality 1 > 2 > 3 (fig. 1.7.2-1).

Fig. 1.7.2-1: Vertical displacements depending on the lateral expansion

30 1. Bearings

Fig. 1.7.2-2: Stress distribution

If the surface of the rubber is fixed to a rigid body shear stresses develop between the two surfaces under compression (fig.1.7.2-2). Under compression we obtain a virtual modulus of elasticity Ei compr which depends not only on the shear modulus G but also on the thickness of the part between two plates. For rectangular parts a good approxi-mation for Ei compr is given by

The maximum stresses under compression between two rigid bodies are

For bending, the effective modulus of elasticity Ei bending is lower than Ei compr because we obtain a compression in two half waves under a constant rotation angle a. If both halves develop a constant displacement, the virtual modulus of elasticity would be the

1.7 Materials for bearings 31

the maximum σ is not in the middle of one half but nearer the outer side; thus we

This is described very well by the finally obtain:

following approximate formula:

Under the rotation a we obtain a curvature

and a restraining moment

Fig. 1.7.2-3: Rotation - restraining moment

Fig.1.7.2-4: Displacement - restraining forces

1.7.3 Sliding elements For sliding elements in constructional bearings it is normal to use PTFE, also known by the registered trade names Teflon and Hostaflon. PTFE is a so called thermoplast. For bearings it is used in the original (virgin) condition, i. e. not sintered and without fillers. As a counterpart to this rather soft material polished stainless steel plates are normally used, and sometimes acetal resin plates or hardened chromium-plated steel plates. Chromium-plated steel plates are not resistant to fluorine ions and are rather prone to corrosion than stainless steel plates. They are allowed for convex elements only.

The combination of a soft and a hard part has the advantage that there is no danger of cold welding which can occur on polished metal or plastic surfaces under high pres-sure. To minimise the friction silicon grease should be used to provide lubrication. To keep this grease between the two surfaces the PTFE has lubricant pockets on its sur-face, so that a permanent lubrication takes place over several years. The PTFE plates for bearings are normally 5 to 6 mm thick, the depth of lubricant pockets is 2 mm. Un-

32 1. Bearings

der pressure the PTFE yields. To keep the PTFE in the desired shape it is necessary to keep about half the thickness in a «chamber» with sharp edges. Over the sharp edges we obtain a small bulge. It is also possible to glue PTFE to a steel surface. In this case the PTFE is about 2.5 mm thick. The friction coefficient increases with decreasing temperature and with decreasing compression. The static friction coefficient (first movement) is higher than the dy­namic coefficient. After movement has taken place the dynamic friction coefficient re­mains at this value and returns to the static value after a few hours. This might depend on the orientation of the large polymer molecules; during movement they are orientat­ed into the direction of motion within a very thin surface layer. When the motion is stopped, the orientation is lost within a few hours. Fig. 1.7.3-1 shows the design val­ues of the friction coefficient µd between PTFE and stainless steel, depending on the compression force (EN 1337-2).

Fig.1.7.3-1: Friction coefficient depending on the compression force

The design value of the ultimate compression load is

maximum temperature of the bearing.

The wearing of the PTFE depends on a) the product of compression and velocity of the displacement b) the total amount of sliding during the life-time c) the lubrication of the surface (a loss of lubrication leads to extremely high wearing) d) the roughness and the hardness of the stainless steel surface e) the contact pressure near the edge of PTFE (ironing effect)

1.8 Analysis and design of bearings 33

For slow movements caused by thermal actions we obtain long sliding movements but at a low velocity. Quick movements caused by traffic loads have short sliding move­ments but they occur at high velocity. Wearing is mostly caused by the second case.

For the stainless steel plate, austenitic steel X6CrNiMol7122 according to EU-RONORM 88-2, surface n (IIIc), should be used. The stainless steel plate must cover the PTFE plate completely in all situations. The thickness of the plate should be at least of 1.5 mm. The connection to the carrying plate of mild steel can be welded or glued. For 2.5 mm thick plates the connection can be riveted or bolted.

1.8 Analysis and design of bearings

1.8.1 Hertz compression For the design of bearings the following problems should be addressed: compression between two spherical bodies, compression between a spherical and a flat body, com­pression between two cylindrical bodies, compression between a cylindrical and a flat body along a generator line. As already mentioned, Heinrich Hertz obtained the solu­tion under the following assumptions (1881):

1. The two bodies consist of isotropic, homogeneous and infinitely elastic materials. 2. Only normal stresses (no shear stresses) occur at the contact areas. 3. The radius (width) of the contact areas is small compared with the radii of the

involved bodies. Hertz found the following maximum compression stresses max σ and widths b on the contact areas:

Spherical body on spherical body

Cylindrical body on cylindrical body

Fig. 1.8.1-la: A rrangement of the radii

Fig. 1.8.1-lb: A rrangement of the radii

34 1. Bearings

with

F 1 r1,r2

E v max a b

bearing reaction length of the cylinder radii of the bodies in contact Young's modulus Poisson's ratio (v = 0.3 for steel)

maximum normal stress at the contact area half the width of the contact zone

Fig. 1.8.1-2: Stress distribution

For the usual rocker or roller bearings the max σ beneath the vertical bearing reaction greatly exceeds the material yield strength (fig.1.8.1-2). However, at the contact zone we have not only vertical but also horizontal compression stresses. According to the von Mises criterion the comparison stress

the material yield strength fy. In the present three-dimensional compression regime, σv will be less than σ1 and yielding will not begin until σ1 = fy. On the other hand, the maximum strain does not occur at the surface in the middle of the compression zone, so that the hardness of the surface is not the only criterion for the assessment of Hertz compression.

and yielding begins when σv reaches

EN 1337-4 - roller bearings - gives for the design line load pd of a roller bearing

fu tensile strength of the material R radius of the cylinder Ed design value of the modulus of elasticity

1.8 Analysis and design of bearings 35

Compared to Hertz's formula with

max σd =0.418

we find

EN 1337-6 - rocker bearings - gives for the design load Fz .d of a point rocker bearing

(sphere against plane surface)

Compared to Hertz's formula with

maxσ d =0.388

we find

For cylindrical rocker bearings the same formulae as for roller bearings are used.

1.8.2 Pin and pin plate for leaf and link bearings A special problem of all leaf and link bearings concerns the design of the pin and the pin plate. Eurocode 3, part 1-1, gives simple but satisfactory design rules. The design values of the shear force and the bending moment for the pin can be found using the simple model of distributing the force of each pin plate uniformly over the pin.

Fig. 1.8.2-1: Load distribution to the pin

In the case of fig. 1.8.2-1 we obtain the shear force and the bending moment according to fig. 1.8.2-2 and fig. 1.8.2-3.

36 1. Bearings

Fig. 1.8.2-2: Shear force

Fig.1.8.2-3: Bending moment

For normal bridge bearings we have:

The design values for the resistances are

The combination of shear and bending has to fulfil the inequality

In this inequality, the central pin plate is controlling.

The bearing resistance of plate (thickness t and yield strength fy) and pin is:

Fb,Rd = 1.5 t d fy/YMp

fyp field strength of the pin fup tensile strength of the pin YMp = 1.25 according to EC 3-1-1 The bearing capacity of the pin plate at the hole is achieved under one of the following conditions (EC 3-1-1 gives two possibilities):

1.9 Installation of bearings 37

a) Depending on the pin plate thickness t:

t = min (2a, b),

b) Depending on the geometry of the pin plate:

1.9 Installation of bearings

Concerning the installation of bearings, the need for a later simple replacement must be taken into account. So it should be common practice to put every bearing between a lower and an upper steel cover plate. These cover plates are anchored or connected both with the substructure and the superstructure. These cover plates are connected to the bearings during the installation but remain fixed to the structure while the bearings are replaced (fig.1.9-1). Thus, the connection between bearing and cover plates should be constructed in order to allow a simple release. Bolted connections are often used but after many years often the bolts can hardly be unscrewed. According to the author's experience, fastening the bearings with small fillet welds that can be ground off and remade during the replacement process is simpler.

cover plate

bearing

cover plate

mortar bedding

Fig. 1.9-1: Fixing of a bearing

Generally, bearings should not be built directly on the construction beneath. To guar­antee that the area below a bearing is fully sealed a layer of mortar or of a similar prod­uct is used. So the height of the bridge at the abutments or piers can be adapted easily and very exactly. It is useful to fix the bearing to the bridge so that there is no clear­ance at the upper plate and to adjust the bridge by hydraulic jacks. In this situation the

38 1. Bearings

bearings should be adjusted exactly. Thus, the lower plate will get exactly the desired inclination (horizontal or parallel to the gradient, see fig.1.9-1) and all moveable bear­ings will have the desired pre-adjustment, which depends on the temperature of the bridge and the expected shrinkage and creep. The installation of the bearings should be done early in the morning when the bridge has a (nearly) constant temperature. The designer has to provide a table with the pre-adjustment of every bearing depending on the measured bridge temperature. For good functioning, careful handling of the bearings during installation is very im­portant. The bearings must be kept free of dirt, mortar, water and dust, especially from all moving parts. Many bearings, such as pot bearings and spherical bearings, are pro­tected against dust by rubber bulges, but others are not protected at all. These have to be cleaned to remove mortar and sand after the installation. The gap between the lower plate of the bearing and the substructure is normally 3 to 5 cm thick and must be completely filled with a mortar bedding. This can be done in dif­ferent ways: - by a fresh mortar bedding, chambered in the centre where the bearing is set. The

excess of mortar will come out on all sides and must be removed. - by a special joint filling mortar which must be mixed in a pan type concrete mixer

with a precise quantity of water. This mortar is liquid at first and should be poured in a formwork around the bearing only from one side, so that the air can escape on the other side. The special mortar fills the gap without air bubbles, it sets and hard­ens very quickly so that after one day the mortar bedding can be fully loaded and the formwork removed. If the gap is less than 1 cm a two-component epoxy resin should be used instead of mortar. Initially this resin is a lighter fluid than mortar, thus completely filling even very small gaps.

- by boxing up earth-damp mortar in the gap with a wooden stick also from one side to avoid air bubbles. This method will be difficult for the lower plates with a short side larger than half a metre.

All mortars should be non-shrinking.

1.10 Inspection and maintenance

Visual tests of all bearings should be done by qualified personnel at regular intervals. The following properties of the bearings have to be checked: a) sufficient ability to allow movement, taking into account the temperature of the su­

perstructure b) correct positioning of the bearings themselves and of parts of the bearing relative to

each other c) uncontrolled movement of the bearing d) fracture, cracks and deformations of parts of the bearings e) cracks in the bedding or in adjacent parts of sub- and superstructure f) condition of the anchorage g) condition of sliding or rolling surfaces h) condition of the anticorrosive protection, against dust, and of the sealings. For the different types of bearings the following checks are of importance:

1.11 Replacement of bearings 39

Elastomeric bearings: Displacements and rotations, cracks in the elastomer. Roller and rocker bearings: Displacements and rotations, adjustment of the guiding device, no gap in the contact line. Pot bearings: Sufficient mesh of the lid in the pot, tight sealing of the elastomer in the pot (if the sealing has a defect, the elastomer comes out like a pancake!) Sliding devices - PTFE and stainless steel: Thickness of the PTFE, clean surface of the stainless steel.

The result of an inspection should be recorded in a report. EN 1337-10 gives an ex­ample for such a report. For maintenance the bearings should be cleaned, lubricated (if necessary and pos­sible) and coated with paint. Small defects should be repaired as far as possible.

1.11 Replacement of bearings

The replacement of bearings is a normal maintenance operation for bridges. Thus, a bridge designer has to provide measures so that a replacement can be carried out easily. The owner of a bridge has to define in the tender if the replacement of the bear­ings must be carried out under full traffic, restricted traffic or without traffic, depend­ing on the importance of the bridge and the possibility of a traffic ban or a traffic diversion. In case of a replacement under traffic the jacking equipment should allow the same movements as the bearing. To allow rotations the jacks around one bearing should be connected to a single hydraulic circle. That means that the security devices must have a sufficient clearance. Translations are possible by means of additional sliding con­structions.

reinforcement against splitting tension

Fig.1.11-1: Stiffened areas for hydraulic jacks

To replace a bearing, the bridge has to be lifted by one or more hydraulic jacks. For hy­draulic jacks, adequately stiffened areas to transmit the hydraulic jack forces to the sub- and superstructure are required. Concrete parts must be reinforced against split­ting tension, steel parts need stiffeners (fig.1.11-2). Thus, the construction drawings must show in which areas or at which points hydraulic jacks can be set, what are the maximum lifting forces and up to which level the bridge may safely be lifted. This

40 1. Bearings

data is of particular importance if the bridge is supported in a statically indeterminate way at one abutment or pier, in which case the lifting force depends on the height of lift. High stresses can be induced in the cross girder or diaphragm by the lifting device. In such cases it may be necessary to lift the whole cross section uniformly with two or more hydraulic jacks even for exchanging only one bearing. If more than one jack is used the forces can be controlled by hydraulic connection of some or of all jacks: all connected jacks have the same pressure. Hydraulic jacks need some clearance for the installation. For lifting by a few millimetres up to two centimetres flat piston jacks can be used. The following table gives a guide for the required clearances:

Force Required clearance Normal hydraulic jack

Required clearance Flat piston jack

mm mm kN 500 1000 2000 5000

300 360 450 600

150 180 200 250

Table 1.11-1: Required clearance for hydraulic jacks

There are flat jacks with a height of 80 mm and a lifting force up to 5000 kN. But their stroke is only 20 mm and there is no security device. This kind of jack should be ap­plied for special cases only. New bridges should be constructed for normal hydraulic jacks. In all situations, during the replacement of a bearing the hydraulic jack should be se­cured by a mechanical device such as an adjusting nut for the piston or lining plates to avoid dropping in case of pipe rupture or rupture of the piston sealing which some­times can occur (fig.1.11-3 and fig.1.11-2).

Fig.1.11-2: Hydraulic jack with lining plates

1.12 Codes and standards 41

Fig. 1.11-3: Hydraulic jack with thread and nut

If the replacement of a bearing takes a long time so that displacements of moveable bearings will occur, the hydraulic jacks have to be equipped with a sliding device, normally PTFE plus a sliding plate of stainless steel.

Particular care is required when replacing bearings which transmit horizontal forces: if the friction between the jack and the surface of sub- and superstructure is not suffi­cient it is necessary to restrain the movement of the bridge by appropriate devices. If the replacement is done under traffic, in most cases, and especially for railway bridges, these devices have to transmit all horizontal forces due to a possible loss of friction.

1.12 Codes and standards

The first attempts to standardize bearings in national codes were made decades ago. In Europe several codes and national standards are available. The best known national standards in Europe on this topic are Germany: DIN 4141 Lager im Bauwesen (structural bearings),

Teil 1 bis 14. United Kingdom: BS 5400 Steel, Concrete and Composite Bridges.

Section 9.1 Code of Practice for design of bridge bearings Section 9.2 Specification of materials, manufacturing and installa­

tion of bridge bearings

New European Standards about bearings are the following EN 1337 "Structural bearings" with the parts EN 1337-1 General design rules EN 1337-2 Sliding elements EN 1337-3 Elastomeric bearings EN 1337-4 Roller bearings

42 1. Bearings

EN 1337-5 Pot bearings EN 1337-6 Rocker bearings EN 1337-7 Spherical and cylindrical PTFE bearings EN 1337-8 Guided bearings and Restrained bearings EN 1337-9 Protection EN 1337-10 Inspection and maintenance EN 1337-11 Transport, storage and installation

A recommendable American Standards about bearings is the following: AASHO-LRFD: American Association of State Highway Officials (1994).

1.13 References

Books and special chapters about bearings for bridges: Eggert H., J. Grote, W. Kauschke: Lager im Bauwesen. Verlag von Wilhelm Ernst & Sohn, Berlin, München, Düsseldorf 1974. Lee D.J.: Bridge Bearings and Expansion Joints. Second edition by E & FN Spon, London, Glasgow, New York, Tokyo, Melbourne, Madras 1994. Eggert H., W. Kauschke: Lager im Bauwesen. 2. Auflage, Ernst & Sohn, Berlin 1995. Rahlwes K., R. Maurer: Lagerung und Lager von Bauwerken in: Beton-Kalender 1995, Teil 2, Ernst & Sohn, Berlin.

Papers: [1] Albrecht, R.: Zur Anwendung und Berechnung von Gummilagern. Der Deut­

sche Baumeister 1969, Heft 4, Seite 326, und Heft 6, Seite 563. [2] Andrä, Beyer, Wintergerst: Versuche und Erfahrungen mit neuen Kipp- und

Gleitlagern. Der Bauingenieur 5 (1962). [3] Andrä, W. und Leonhardt, F: Neue Entwicklungen für Lager von Bauwerken,

Gummi- und Gummitopflager. Die Bautechnik 39 (1969), Heft 2, Seite 37 bis 50.

[4] Bayer, K.: Auflager und Fahrbahnübergänge für Hoch- und Brückenbauten aus Kunststoff. Verein Deutscher Ingenieure VDI im Bildungswerk BV 1956 (Vor-tragsveröffentlichung).

[5] Beyer, E. und Wintergerst, L.: Neue Brückenlager, neue Pfeilerform. Der Bau­ingenieur 35 (1960), Heft 6, Seite 227 bis 230.

[6] Eggert, H.: Brückenlager. Die Bautechnik 50 (1973), S. 143/144. [7] Bub, H.: Das neue Institut für Bautechnik. Strasse und Autobahn, Band 20

(1969), Seite 189. [8] Burkhardt, E.: Gepanzerte Betonwälzgelenke, Pendel- und Rollenlager. Die

Bautechnik 17 (1939), Seite 230. [9] Cardillo, R. und Kruse, D.: Paper (61/WA-335) ASME (1961). [10] Cichocki, F: Bremsableitung bei Brücken. Der Bauingenieur 36 (1961), Seite

304 bis 305.

1.13 References 43

[11] Clark, E. und Moutrop, K.: Load Deformation Characteristics of Elastomer Bridge Bearing Pads. University of Rhode Island, May 1962.

[12] Desmonsablon, Philippe: Le calcul des piles deformables avec appuis en caoutchouc. Annales des Ponts et Chaussées, Paris 4/1960.

[13] Eggert, H.: Bauwerksicherheit bei Verwendung von Rollen- und Gleitlagern. Strasse Brücke Tunnel 1971, Heft 3, Seite 71.

[14] Eggert, H.: Die baurechtliche Situation bei Lagern für Brücken und Hochbau-ten. Der Stahlbau 39 (1970), Heft 6, Seite 189.

[15] Einsfeld, U.: Erläuterungen zu den Richtlinien von unbewehrten Elastomer-lagern. Mitteilungen Institut für Bautechnik 6/1972.

[16] Franz: Gummilager für Brücken. VDI-Zeitschrift, Bd. 101/1959, Nr. 12, Seite 471 bis 478.

[17] Gent, A.: Rubber Bearings for Bridges. Rubber Journal and International Plas­tics 1959.

[18] Grote, J.: Neoprenelager - einige grundsätzliche Erwägungen. Kunststoffe im Bau 7/1968.

[19] Grote, J.: Unbewehrte Elastomerlager. Der Bauingenieur 44 (1969), Seite 121. [20] Grote, J.: Vermeidung von Rissen und Dehnungsschaden durch gummielasti-

sche Lagerungen. Kunststoffe im Bau 11/1968. [21] Hakenjos, V.: Untersuchungen über die Rollreibung bei Stahl im elastisch-plas-

tischen Zustand. Technisch-wissenschaftliche Berichte der Staatlichen Materi-alprufungsanstalt an der Technischen Hochschule Stuttgart 1967, Heft 67/05.

[22] Heesen: Gepanzerte Betonwälzgelenke, Pendel- und Rollenlager. Die Bau­technik, Jahrgang 25 (1948), Seite 261.

[23] Hütten, P.: Beitrag zur Berechnung der Lagerverschiebungen gekrümmter, durchlaufender Spannbeton-Balkenbrücken. Dissertation TH Aachen 1970.

[24] Jörn, R.: Gummi im Bauwesen. Elastische Lagerung einer Pumpenstation. Der Bauingenieur 36 (1961), Heft 4, Seite 137/138.

[25] Keen: Creep of Neoprene in Shear Under Static Conditions, Ten Years, Trans­actions of the ASME, Juli 1953.

[26] Leonhardt und Andrä: Stützungsprobleme der Hochstrassenbrücken. Beton-und Stahlbetonbau 55 (1960), Heft 6.

[27] Leonhardt, F. und Reimann, H.: Betongelenke, Versuchsbericht, Vorschläge zur Bemessung und konstruktiven Ausbildung. DAfStb, Heft 175. Berlin: Verlag Ernst & Sohn 1966, und Leonhardt, F. und Reimann, H.: Betongelenke. Der Bauingenieur 41 (1966), Seite 49.

[28] Leonhardt, F. und Wintergerst, L.: Über die Brauchbarkeit von Bleigelenken. Beton- und Stahlbetonbau 1961, Heft 5, Seite 123 bis 131.

[29] Maguire, C. und Assoc: Elastomeric Bridge Bearings Pads 1959. [30] Massonnet: Zuschrift zu B. Topaloff, Gummilager für Brücken. Der Bauinge­

nieur 39 (1964), Seite 428. [31] Mönnig, E. und Netzel, D.: Zur Bemessung von Betongelenken. Der Bauinge­

nieur 44 (1969), Seite 433 bis 439. [32] Morton, M.: Rubber Technology. Reinhold Publishing Co. 1959. [33] Mullins, L.: Softening of Rubber by Deformation. Rubber Chemistry and

Technology (Feb. 1969).

44 1. Bearings

[34] Nordlin, E., Stoker, S. and Trinble, R.: Laboratory and Field Performance of Elastomeric Bridge Bearing Pads, Highway Research Board (1968).

[35] Pare u. Keiner: Elastomeric Bridge Bearings. Highway Research Board Bull 242, 1960.

[36] Payne u. Scott: Engineering Design with Rubber [37] Rejcha, C: Design of Elastomer Bearings. Journal of Prestressed Concrete

Institute Oct. 1964, Vol. 9, Nr. 5. [38] Resinger, F.: Längszwängungen - eine Ursache von Brückenlagerschäden. Der

Bauingenieur 46 (1971), Seite 334. [39] Rieckmann, H.-P: Einfluss der Lagerkonstruktion auf die Knicklänge von

Pfeilern. Strasse Brücke Tunnel 1970, Seite 36 bis 42 und Seite 270 bis 272. [40] Sasse, H.-R. und Schorn, H.: Bewehrte Elastomerlager - Stand der Entwick-

lung. Plastik-Konstruktion 1971, Heft 5, Seite 209 bis 227. [41] Schönhofer: Neugestaltungen auf dem Gebiet des Auflagerbaues und auf ver-

wandten Gebieten. Werner-Verlag, Düsseldorf 1952. [42] Sedyter: Über die Wirkungsweise von Bleigelenken. Beton und Eisen 1926,

Seite 29. [43] Shen, M. K.: Über die Lösung des Balkens mit unverschieblichen Auflagern.

Der Bauingenieur 39 (1964), Seite 100. [44] Suess, K. und Grote, J.: Einige Versuche an Neoprenelagern. Der Bauingenieur

38 (1963), Heft 4, Seite 152 bis 157. [45] Thielker, E.: Elastomeric Bearing Pads and Their Application in Structures,

Paper 207 of Leap Conference (1964). [46] Thul, H.: Briickenlager. Der Stahlbau 38 (1969), Seite 353. [47] Topaloff, B.: Gummilager für Brücken - Berechnung und Anwendung. Der

Bauingenieur 39 (1964), Seite 50 bis 64. [48] Topaloff, B.: Gummilager für Brücken. Beton- und Stahlbetonbau 54 (1959),

Heft 9. [49] Uetz, H. und Breckel, H.: Reibungs- und Verschleissversuche mit Teflon.

Sonderheft der Staatl. Materialprüfungsanstalt an der TH Stuttgart, 7.12.1964, Seite 67/76.

[50] Uetz, H. und Hakenjos, V.: Reibungsuntersuchungen mit Polytetrafluoräthylen bei hin- und hergehender Bewegung. Die Bautechnik 44 (1967), Heft 5, Seite 159 bis 166.

[51] Uetz, H. und Hakenjos, V.: Gleitreibungs- und Gleitverschleissversuche an Kunststoffen. Kunststoffe, 59. Jahrgang 1969, Heft 3, Seite 161 bis 168.

[52] Weiprecht, M.: Auflagerung von Brücken. Eisners Taschenbuch für den Bau-technischen Eisenbahndienst, 1967, Seite 231 bis 277, Abschnitt E Brücken-und Ingenieurhochbau.

[53] Zies, K.-W.: Stabilität von Stützen mit Rollenlagern. Beton- und Stahlbetonbau 65 (1970), Seite 297.

[54] AASHO-LRFD: American Association of State Highway Officials (1994). [54] Dupont de Nemours Co.: Design of Neoprene Bridge Bearing Pads, Wilming­

ton (1959). [55] CNR-UNI 10018-68 (Italian Standards for rubber bearings).

1.13 References 45

[56] Ministry of Transport: Provisional Rules for the Use of Rubber Bearings in Highway Bridges, Memo. 802, London (1962).

[57] Mitteilungen, Institut für Bautechnik, 1970, Heft 2 und 4, und 1971, Heft 4 und 6.

[58] Ohne Verfasser. Auflager aus Teflon. Auszüge aus dem Journal of Teflon 1964, 1965 und 1966, Druckschrift der Du Pont de Nemours International S.A. Geneva, Switzerland.

[59] Ohne Verfasser. Brückenlager. Beratungsstelle für Stahlverwendung, Düssel-dorf, Merkblatt 339, 2. Auflage 1968.

[60] ORE Office de Recherches et d'Essais: Verwendung von Gummi für Brücken­lager, Frage D 60, Utrecht (1962, 1964, 1965).

[61] Wiedemann, L.: Zusätzliche Richtlinien für Lager im Brücken- und Hochbau. Mitteilungen Institut für Bautechnik 3/1973, S. 73. Verlag Ernst & Sohn.

[62] Eggert: Vorlesungen über Lager im Bauwesen. Wilhelm Ernst & Sohn 1980/1981.

[63] Kauschke, W.: Entwicklungsstand der Gleitlagertechnik für Brückenbauwerke in der Bundesrepublik Deutschland. Bauingenieur 64 (1989), Seite 109 bis 120.

[64] Battermann/Köhler: Elastomere Federung, Elastische Lagerungen. W. Ernst & Sohn, Berlin, München 1982.

[65] Gerb: Schwingungsisolierungen. Berlin, 9. Auflage 1992, Eigenverlag (gegen Schutzgebühr erhältlich).

[66] Grote, J. und Kreuzinger, H.: Pendelstützen mit Elastomerlagern. Der Bau­ingenieur 53 (1978), Seite 63/64.

[67] Kanning, W.: Elastomer-Lager für Pendelstützen - Einfluss der Lager auf die Beanspruchung der Stützen. Der Bauingenieur 55 (1980), Seite 455.

[68] Maurer/Rahlwes: Lagerung und Lager von Bauwerken. Betonkalender 1995, Ernst & Sohn, Teil II.

[69] Weihermüller, H. und Knöppler, K.: Lagerreibung beim Stabilitätsnachweis von Brückenpfeilern. Bauingenieur 55 (1980), Seite 285 bis 288.

[70] Andrä, W.: Der heutige Entwicklungsstand des Topflagers und seine Weiter-entwicklung zum Hublager. Bautechnik (1984), Seite 222 bis 230.

[71] Eggert, H.: 7 Grundsätze bei der Lagerung von Brücken. 9. IVBH-Kongress Amsterdam 1972, Schlussbericht. Internationale Vereinigung für Brückenbau und Hochbau, Zürich, Schweiz.

[72] Deinhard, J.M., Kordina, K., Mozahn, R., Storkebaum, K.-H.: Der Schadens-fall an der Mainbrücke bei Hochheim. Beton - Stahlbetonbau, 72 (1977), Seite 1 bis 7.

[73] Eggert, H. und Wiedemann, L.: Nutzungsgerechte Lagerung von Stahl- und Verbundbrücken und unterhaltungsgerechte Konstruktion von Brückenlagern. IVBH Symposium Dresden 1975. Vorbericht.

[74] Eggert, H.: Lager für Brücken und Hochbauten. Bauingenieur 53 (1978), Seite 161 bis 168, und Zuschrift 54 (1979), Seite 200.

[75] König, G. et. al.: Spannbeton: Bewährung im Brückenbau. Analyse von Bau-werksdaten, Schäden und Erhaltungskosten. Springer-Verlag Berlin, Heidel­berg, New York, London, Paris, Tokio 1986.

46 1. Bearings

[76] Pfohl, H.: Reaktionskraft am Festpunkt von Brücken aus Bremslast und Bewe-gungswiderständen der Lager. Bauingenieur 58 (1983), Seite 453 bis 457.

[77] Eggert, H. und Hakenjos, V.: Die Wirkungsweise von Kalottenlagern. Der Bau­ingenieur 49 (1974), Heft 3, Seite 93/94.

[78] Lehmann, Dieter: Beiträge zur Berechnung der Elastomerlager. Die Bautech-nik I (1978), Seite 19 bis 22, II (1978), Seite 99 bis 102, III (1978), Seite 190 bis 198, IV (1979), Seite 163 bis 169.

[79] Kordina, K. und Nölting, D.: Zur Auflagerung von Stahlbetonteilen mittels unbewehrter Elastomerlager. Der Bauingenieur 56 (1981), Seite 41 bis 44.

[80] Kordina, K. und Osterath, H.-H.: Zur Auflagerung von Stahlbetonteilen mittels unbewehrter und bewehrter Elastomerlager. Der Bauingenieur 59 (1984), Seite 461 bis 466.

[81] Kessler, E. und Schwerm, D.: Unebenheiten und Schiefwinkligkeiten der Auf-lagerflachen fur Elastomerlager bei Stahlbetonfertigteilen. Fertigteilbau-forum 13/83, Seite 1 bis 5 (Betonwerk + Fertigteil-Technik).

[82] Kessler, E.: Die Anwendung unbewehrter Elastomerlager. Betonwerk + Fertig­teil-Technik, Heft 6 (1987), Seite 419 bis 429.

[83] Bundesminister für Verkehr: Schäden an Brücken und anderen Ingenieurbau-werken. Dokumentation 1982. Verkehrsblatt-Verlag, Dortmund.

[84] Bundesminister für Verkehr: Bericht über Schäden an Bauwerken der Bundes-verkehrswege. Januar 1984. Eigenverlag BMV.

[85] Beyer, E. und Eisermann, G.: Nachstellbare Brückenlager. Erfahrungen beim Bauvorhaben Düsseldorf-Hauptbahnhof. Beton 5/1983.

[86] Dickerhoff, K.J.: Bemessung von Brückenlagern unter Gebrauchslast. Disser­tation Universität Karlsruhe 1985.

[87] Petersen, Chr.: Zur Beanspruchung moderner Brückenlager. Festschrift J. Scheer, März 1987.

[88] Hehn, K.-H.: Prüfeinrichtung zur Untersuchung von Lagern. VDI-Z 118 (1976), Seite 114 bis 118.

[89] N.N., Sanierung der Kölnbreinsperre, Projektierung und Ausführung. 1. Auf-lage Mai 1991. Herausgeber: Österreichische Donaukraftwerke AG.

[90] Hakenjos, V. und Richter, K.: Dauergleitreibungsverhalten der Gleitpaarung PTFE weiss/Austenitischer Stahl für Lager im Brückenbau. Strasse, Briicke, Tunnel 11 (1975), Seite 294 bis 297.

[91] Imbimbo M. und Kelly J.M.: Influence of Material Stiffening on Stability of Elastomeric Bearings at Large Displacements. Journal of Engineering Me­chanics. Sept. 1998.

[92] Zederbaum, J. (1966): The frame action of a bridge deck supported on elastic bearings. Civil Engineering and Public Works Review 61(714), 67-72.

[93] Leonhardt, F. und Andrä, W. (1960): Stiitzprobleme der Hochstrassenbräcken. Beton- und Stahlbetonbau, 55(6), 121-32.

[94] Tanaka, R., Natsukawa, K. and Ohira, T. (1984): Thermal behaviour of multi-span viaduct in frame. In International Association of Bridge and Structural Engineering, 12th Congress, Vancouver, Canada, 3-7 September.

[95] Building Research Establishment (1979) Estimation of thermal and moisture movements and stresses; Part 2, Digest 228, Watford.

1.13 References 47

[96] Emerson M. (1977): Temperature differences in bridges: basis of design re­quirements. TRRL Laboratory Report 765. Transport and Road Research Lab­oratory, Crowthorne.

[97] Emerson M. (1968): Bridge temperatures and movements in the British Isles. RRL Report LR 228, pp.38. Road Research Laboratory, Crowthorne.

[98] Emerson M. (1973): The calculation of the distribution of temperature in bridges. TRRL Report LR 561. Transport and Road Research Laboratory, Crowthorne.

[99] Emerson M. (1976): Bridge temperatures estimated from the shade tempera­ture. TRRL Report LR 696. Transport and Road Research Laboratory, Crow­thorne.

[100] Stephenson, D. A. (1961): Effects of differential temperature on tall slender co­lumns. Concrete and Constructional Engineering, 56(5), 175-8: 56(11), 401-3.

[101] Garrett, R.J. (1985): The distribution of temperature in bridges. The Journal of the Hong Kong Institution of Engineers, May, 35-8.

[102] Comité Euro-International du Béton (1984). Design manual on structural effects of time-dependent behaviour of concrete (Bulletin No. 142). George Publishing Company.

[103] Comité Euro-International du Béton (1985). Manual of Cracking and Defor­mations. Bulletin 158E, Lausanne.

[104] Neville, A.M., Dilger, W.H. and Brooks, J.J. (1983): Creep of Plain and Struc­tural Concrete. Construction Press, London and New York.

[105] Mattock A.H. (1961): Precast-prestressed concrete bridge 5.Creep and shrink­age studies. Journal of the Portland Cement Association Research and Devel­opment Laboratories, May.

[106] Institution of Geological Sciences: National Environmental Research Council (1976), Atlas of Seismic Activity 1909-1968. Seismological Bulletin No.5.

[107] Dollar, A.T.J., Abedi, S.M.H., Lilwall, R.C. und Willmore, R.L. (1975): Earth­quake risk in the UK. Proceedings of the Institution of Civil Engineers, 58, 123-4.

[108] ICE and SECED (1985): Earthquake engineering in Britain. Proceedings of Conference of the Institution of Civil Engineers and the Society of Earthquake and Civil Engineering Dynamics, University of East Anglia, April.

[109] Lee, D.J. (1971): The Theory and Practice of Bearings and Expanison Joints for Bridges, Cement and Concrete Association.

[110] Buchler, W. (1987): Design of Pot Bearings, American Concrete Institute Publication, SP-94, Vol.2, pp. 882-915.

[1ll] Black, W. (1971): Notes on bridge bearings, RRL Report LR 382, Transport and Road Research Laboratory, Crowthorne.

[112] Kauschke, W. and Baignet, M. (1987) Improvements in the Long Term Dura­bility of Bearings in Bridges, American Concrete Institute Publication SP-94, Vol.2, 577-612.

[113] Taylor, M.E. (1970): PTFE in highway bridges. TRRL Report LR 491, Trans­port and Road Research Laboratory, Crowthorne.

[114] Eggert, H., Kauschke, W.: Lager im Bauwesen, Ernst & Sohn, Berlin 1996.

48 1. Bearings

[115] Hakenjos, V.: Lager im Bauwesen mit Komponenten aus Kunststoff verdrän-gen hochbeanspruchbare stählerne Rollenlager. 13th H.F. Mark-Symposium on 19-10-94 in Vienna.

[116] Marioni, A.: Apparecchi di appoggio per ponti e strutture. ITEC, Milano 1983 [117] Campbell, T. I. and Kong, W. L.: TFE Sliding Surfaces In Bridge Bearings. Re­

port ME-87-06, Ontario Ministry of Transportation and Communications, Downsview, Ontario, 1987.

[118] Crozier, W. F, Stoker, J. R., Martin, V. C. and Nordlin, E. F: A Laboratory Evaluation of Full-Size Elastomeric Bridge Bearing Pads. Research Report CA DOT, TL-6574-1-74-26, Highway Research Report, June 1979.

[119] Gent, A. N.: Elastic Stability of Rubber Compression Springs. ASME, Journal of Mech. Engr. Science, Vol. 6, No. 4, 1964.

[120] Jacobsen, F. K. and Taylor R. K.: TFE Expansion Bearings for Highway Bridges. Report No. RDR-31, Illinois DOT, June 1971.

[121] McEwen, E. E. and Spencer, G. D.: Finite Element Analysis and Experimental Results Concerning Distribution of Stress Under Pot Bearings. Proceedings of 1 st World Congress on Bearings and Sealants, ACI Publication SP-70, Niagara Falls, 1981.

[122] Nordlin, E. F, Boss, J. F. and Trimble, R. R.: Tetrafluorethylene (TFE) as a Bridge Bearing Material. Research Report, M & R 64642-2, California DOT, Sacramento, CA, June 1970.

[123] Roark. R. J. and Young, W. C: Formulas for Stress and Strain. 5th Ed., McGraw Hill, New York, 1976.

[124] Roeder, C. W., Stanton, J. F. and Taylor, A. W.: Performance of Elastomeric Bearings. NCHRP Report 298, TRB, National Research Council, Washington, D. C, October 1987.

[125] Roeder, C. W. and Stanton, J. F: State of the Art Elastomeric Bridge Bearing Design. ACI Journal, 1991.

[126] Roeder, C. W., Stanton, J. F and Feller, T.: Low Temperature Performance of Elastomers. ASCE, Journal of Cold Regions, Vol. 4, No. 3, September 1990, pp 113-132.

[127] Roeder, C. W. and Stanton, J. F: Failure Modes of Elastomeric Bearings and Influence of Manufacturing Methods. Proceedings of 2nd World Congress on Bearings and Sealants, ACI Publication SP-94, Vol. 1, San Antonio, Texas, 1986.

[128] Roeder, C. W., Stanton, J. F and Taylor, A. W.: Fatigue of Steel-Reinforced Elastomeric Bearings. ASCE, Journal of Structural Division, Vol. 116, No. 2, February 1990.

[129] Roeder, C. W., and Stanton, J. F: Elastomeric Bearings: A State of the Art. ASCE, Journal of the Structural Division, No. 12, Vol. 109, December 1983.

[130] Saxena, A. and McEwen, E. E.: Behaviour of Masonry Bearing Plates in High­way Bridges. Proceedings of 2nd World Congress on Bearings and Sealants, ACI Publication SP-94, San Antonio, 1986.

[131] Stanton, J. F and Roeder, C. W.: Elastomeric Bearings Design, Construction, and Materials. NCHRP Report 248, TRB, National Research Council, Wash­ington, D. C, August 1982.

1.13 References 49

[132] Stanton, J. F„ Scroggins, G., Taylor, A. W. and Roeder, C. W.: Stability of Laminated Elastomeric Bearings. ASCE, Journal of Engineering Mechanics, Vol. 116, No. 6, June 1990, pp 1351-1371.

[133] Structural Bearing Specification. FHWA Region 3 Structural Committee for Economical Fabrication, Subcommittee for High Load Multi-Rotational Bear­ings (HLMRB), October 1991.

Page Blank

In Original

50

51

2 Expansion Joints

2.1 Introduction

As mentioned in chapter 1.1, movements in old stone and timber bridges were small and no additional devices were necessary to close the gaps between bridges and abut­ments due to bridge movements. The first expansion joints were built for steel railway bridges because their movements were not negligible. With the increase of road traf­fic and of its speed, closing the gaps became necessary for safety reasons, especially at the moveable bearings. Initially, cover plates were used for expansion joints. For longer bridges these cover plates were not sufficient, so that finger joints and sliding plate joints were used. All these types of expansion joints were not watertight and so the water ran down to the bearings and to the abutments. The first watertight expan­sion joints were built using steel rails between rubber tubes to absorb the movements. This principle led to a lot of different multisealed expansion joints which differed in the means of supporting the steel rails, in the rubber profiles and in controlling the gap widths. Another type of watertight expansion joint is the cushion joint, consisting of a rubber cushion with vulcanised steel plates which transfer the traffic loads. In spite of continuous amendments of all constructions for expansion joints, these still remain wearing parts, especially in bridges with high traffic density and high traffic loads. The following chapters give a short survey of expansion joints for different move­ments used in the construction of bridges.

2.2 The role of expansion joints

The role of expansion joints is to carry loads and to provide safety to the traffic over the gap between bridge and abutment or between two bridges in a way that all bridge displacements can take place with very low resistance or with no resistance at all. A further requirement is a low noise level especially in an urban environment. The expansion joints should provide a smooth transition from the bridge to the adjacent areas. The replacement of an expansion joint is always combined with a traffic inter­ruption - at least of the affected lane. Therefore expansion joints should be robust and suitable for all loads and local actions under all weather conditions, moisture and de-icing agents. The replacement of all wearing parts should be possible in a simple way.

2.3 Calculation of movements of expansion joints

Movements of expansion joints depend on the size of the bridge and the arrangement of the bearings. Normally the form of construction depends on the horizontal transla­tion orthogonal to the joint. But it is necessary to consider all translations and rotations to ensure that the displacements will not reach the limits of the joint construction. To describe the movements of an expansion joint in detail we have to consider three translations and three rotations (fig.2.3-1).

52 2. Expansion Joints

Fig.2.3-1: Possible movements

These movements result from temperature, displacements due to external loads, and creep and shrinkage in concrete and composite bridges. We may obtain the move­ments (displacements and rotations) from the structural analysis of the system. Move­ments due to loads depend on the location of the loads. The controlling deformations can be determined with influence lines (fig.2.3-2 and fig.2.3-3). The influence line of a deflection is the bending line due to a unit load acting in the direction of the con­sidered movement.

Fig.2.3-2: Influence line for a translation

Fig.2.3-3: Influence line for a rotation

To obtain the displacement caused by a rotation it is also possible to calculate the rotations; the displacements can be determined from the known rotations.

2.3.1 Horizontal translation in the direction of the bridge axis ux

A change of the environment temperature, creep under normal force and shrinkage lead to a uniform extension or shortening of the bridge (fig.2.3.1-1). The thermal expansion coefficients of steel and concrete have approximately the same value (αT = 1.0...1,2 10-5 /K ). A uniform change of temperature about the cross section causes only a horizontal translation of the joint. This applies to composite bridges, too.

2.3 Calculation of movements of expansion joints

Fig.2.3.1-1: Uniformly extension or shortening

Creep and shrinkage of concrete bridges

Temperature:

Creep:

Shrinkage:

A possible problem is the change of the location of the fixing point or the unknown lo­cation of the fixing point. On arch bridges the superstructure is usually fixed at the crown of the arch. The fixing point is moved by the deformation of the arch due to the asymmetrical load. Buried expansion joints are often used for short bridges (Chapter 2.4). If the fixing point is situated on longer piers, it acts as a horizontal spring bearing. Due to a movement in the joint a plastic deformation of the asphalt layer occurs and the construction has a certain rigidity. A different rigidity of the expansion joints on the right and left abut­ment and a possible longitudinal deformation can lead to the cracking of the asphalt layer at one abutment. As the rigidity of this joint is higher than the rigidity of the piers the new fixing point is situated near the undamaged expansion joint (fig. 2.3.1-2).

Fig.2.3.1-2: Change of the fixing point

Coefficient of creep

Permanent normal force (compression > 0)

Shrinkage coefficient

54 2. Expansion Joints

In the case of an elastic fixing point there are additional movements at expansion joints due to acceleration and braking forces. The actual rigidity of piers can differ from the planned rigidity. Moreover, if the bridge is fixed on more than one pier, the position of the fixing point can differ from the planned position.

Creep and shrinkage in composite bridges (acting in the concrete parts of cross-section only) mainly lead to deflections which result in rotations above the y-axis (fig. 2.3.1-4). Creep can be considered using a reduced section area and a reduced moment of inertia, shrinkage by a substitute tensile force Nsh acting on the free shrinking con­crete. Nsh is a compression force acting on the composite cross-section.

Shrinkage coefficient

Ac Area of concrete

Ec Reduced modulus of elasticity of concrete to consider creep

Fig.2.3.1-3: Equivalent shrinking force

Fig.2.3.1-4: Deflection underload

Horizontal movements of expansion joints can also be caused by vertical movements of the abutments. They are caused by foundation settlements or by replacement of bearings (fig. 2.3.1-5). Statically indeterminate steel and composite bridges can be prestressed by intentional lifting and/or lowering at the bearings.

positive definition:

N s h = εcs · Ac· E c

εsc

2.3 Calculation of movements of expansion joints 55

Fig.2.3.1-5: Displacement of bearings

If a fixing point is located on a high pier the additional movements due to pier defor­mation must be considered in the structural analysis. The movements can result from acceleration, braking forces, uniform and non-uniform temperature actions.

2.3.2 Horizontal translation in direction of the cross-section uy

A horizontal translation in the crosswise direction results if the angle between the joint and the moving direction of the bearing is not 90 ° (e.g. in skew bridges). The magnitude of the movement depends on the magnitude of the movement in the direc­tion of the bridge axis and on this angle (fig.2.3.2-1 and fig.2.3.2-2).

Fig.2.3.2-1: Skewed bridge

56 2. Expansion Joints

Fig.2.3.2-2: Skewed bearing conditions

2.3.3 Vertical translation uz

Vertical translations uz can be caused by the replacement of bearings (fig.2.3.3-3) and the geometrical conditions on the abutment (fig.2.3.3-1 and fig.2.3.3-2).

Fig.2.3.3-1: Sloping bridge with horizontal bearings

Fig.2.3.3-2: Bridge with short cantilever on the abutment

2.3 Calculation of movements of expansion joints 57

Hydraulic jack

Fig.2.3.3-3: Vertical displacement of bearings (due to bearing replacement)

2.3.4 Rotation around the bridge axis φx

In the case of a replacement of one single bearing at one side a rotation (φx occurs (fig. 2.3.4-1). However, it is possible to avoid this movement by uniform lifting over the cross-section.

Hydraulic jack

Fig.2.3.4-1: Lifting on one side

2.3.5 Rotation around the y-axis φy

This deformation is caused by vertical loading and non-uniform temperature. The controlling load positions of the traffic loads can be determined with influence lines.

Fig.2.3.5-1: Rotation due to deflections

2.3.6 Rotation around the z-axis φz

The deformation φz is caused by non-uniform temperature action in the horizontal direction, and by wind loads (fig.2.3.6-1).

58 2. Expansion Joints

Fixed bearing

Fig.2.3.6-1: Non-uniform temperature action

2.4 Construction of expansion joints

2.4.1 General The construction of expansion joints has to fulfil the following requirements: - movement capacity - bearing capacity for static and dynamic loading, - watertightness to save bearings, substructure and possible linkage of expansion

joints from deterioration, - low noise emission, - traffic safety. To fulfil the last two requirements a limitation of gap widths is essential. Additional­ly, it is recommended to avoid slopes exceeding about 3 % and vertical steps between joined surfaces exceeding 8 mm (fig.2.4.1-1).

Fig.2.4.1-1: Recommended safety requirements

Expansion joints are exposed to pollution. The sealing should not be damaged by inclusions of bigger external bodies. If the gap width is reduced due to a movement of the superstructure the joint must be able to expel grit and silt to the carriageway surface.

2.4 Construction of expansion joints 59

In particular, all elastomeric components must be readily accessible and easily re­placeable.

2.4.2 Small movements (up to 25 mm) For movements up to 15 mm it is possible to construct a continuous asphaltic car­riageway pavement with a supporting element covering the gap of the superstructure. This kind of joint is also called a buried expansion joint (fig.2.4.2-1). Up to 10 mm a flat metal plate is sufficient; for movements above 10 mm an elastomeric pad is nec­essary to avoid pavement cracks at the edges of the supporting plate. An additional re­inforcement of the pavement is advisable to provide a uniform strain distribution. The thickness of the pavement should be at least 80 mm and should be equal to the thick­ness of the corresponding parts of the superstructure and the abutment. To fulfil this requirement the cover of the gap is usually extended into a niche. The asphaltic pavement does not provide sufficient watertightness. An additional seal­ing is recommended to protect bearings and substructure from deterioration.

Fig. 2.4.2-1: Buried expansion joint

There are covering elements fulfilling the requirements of support, strain distribution and watertightness without additional sealing, e.g. the following kind of joint con­struction (fig.2.4.2-2 and fig.2.4.2-3).

Fig.2.4.2-2: Buried expansion joint sealed by a rubber profile

60 2. Expansion Joints

Fig.2.4.2-3: Buried expansion joint with continuous sealing and additional rubber profile

For movements between 15 and 25 mm the asphaltic material above the joint can be replaced by a specially modified asphaltic material. Constructions of this kind are called asphaltic plug joints (fig.2.4.2-4 and fig.2.4.2-5). The thickness should be at least 80 mm, while the length should not exceed 700 mm. Though movements exceeding 25 mm could be managed in laboratory tests the influ­ence of temperature and of deformation velocity is not known adequately. Incorrect placement of material results in tearing of the adjacent carriageway pavement. Further problems are yielding of asphaltic material under the wheels of standing vehicles, brake and acceleration forces combined with high environment temperatures, and the development of rutting. Because of their low lifetime (though combined with low relative costs) asphaltic plug joints are recommended for temporary purposes.

Fig.2.4.2-4: Asphaltic plug joint

2.4 Construction of expansion joints 61

Fig.2.4.2-5: Asphaltic plug joint additional sealed by a rubber profile

2.4.3 Medium movements (over 25 mm, up to 80 mm) The absorption of medium movements requires an elastic expansion element or an ex­pansion gap across the carriageway surface. For traffic safety, gaps below 5 mm or over 65 mm are not recommended. Thus, the expansion movement of a simple gap construction is limited to 60 mm. Expansion joints for medium movements consist of a sealing element, edge elements, and fixing elements. The sealing element can be replaced by a cushion element that absorbs movements caused by shear deformation (fig.2.4.3-1).

Fig.2.4.3-1: Construction methods of expansion joints for medium movements

Seals of expansion gaps can be constructed as V-shaped sealing strips (fig.2.4.3-2) or hollow sections (fig.2.4.3-4). Movements are absorbed by the folding of these elements. There are special seals for pavements and cyclist areas to decrease the width of the gap to avoid accidents (fig.2.4.3-3). Traditional cover-plates are prone to rattling and cor­rosion and hinder the accessibility of possible seals, but they provide the best comfort for pedestrians with high heel shoes (fig.2.4.3-6).

62 2. Expansion Joints

Fig.2.4.3-2: V-shaped sealing Fig.2.4.3-3: Special sealing for sidewalks

Fig.2.4.3-4: Hollow section

Fig.2.4.3-5: Expansion joint with V-shaped sealing

Fig.2.4.3-6: Expansion joint with cover plate

Fig.2.4.3-7: Expansion joint for sidewalks

2.4 Construction of expansion joints 63

The use of seals made from cellular neoprene extrusion has the advantage of a closed carriageway surface. In addition to the function as sealing, they are able to transfer traffic loads. Movements up to 80 mm can be accommodated (fig.2.4.3-8).

Fig. 2.4.3-8: Seals made from cellular neoprene extrusion

Elastomeric cushion joints (fig.2.4.3-9) are made from neoprene reinforced with steel plates. Thus, traffic loads can be transferred without significant deflections. The movements are absorbed by increasing and decreasing of the widths of the two gaps on the upper side. The maximum movement is limited by the gap width. The rubber cover of the bearing plate can wear away under traffic or can be damaged (e.g. by snow ploughs) which lowers the skid resistance.

Fig. 2.4.3-9: Elastomeric cushion joint

Especially when using elastomeric cushions and neoprene extrusion seals, the restraining actions can exceed 20 kN/m which in some cases is not negligible.

2.4.4 Large movements (over 80 mm) For large movements, sealing elements and rail elements are coupled. Additionally to the components of a single gap construction, intermediate elements (also called rails), supporting elements and linkage elements are needed (fig.2.4.4-1). Linkage elements cause equal gap widths saving the seals from overextending. They must be able to sustain acceleration and braking forces.

64 2. Expansion Joints

Fig.2.4.4-1: Construction method of expansion joints for large movements

The following figure shows the coupling of cushion elements. In this case a special linkage mechanism is not necessary due to the high deformation resistance of the single elements which actually act as a spring linkage.

Fig.2.4.4-2: Coupled elastomeric cushion joint

A typical construction is the coupling of V-shaped and hollow section sealing ele­ments. It is called multiple seal expansion joint. These expansion joints can be classi­fied by the kind of supporting and linkage. The folding trellis linkages (fig.2.4.4-3) satisfy all supporting and linkage purposes.

2.4 Construction of expansion joints 65

Fig.2.4.4-3: Rails supported by folding trellis linkage

An additional linkage is needed if the rails are supported by parallel beams. One pos­sibility is the spring linkage (fig.2.4.4-4). Springs are made of an elastic material. The portion of the resisting force resulting from friction depends on the number of rails and supporting beams whereas the portion of spring force is independent at the num­ber of springs because of the series connection. A disadvantage of this kind of linkage is that acceleration and braking forces cause non-uniform spring deformations. If the gaps are opened near to the maximum value the seals can be overextended. Another possibility of linkage of parallel supporting beams is the use of horizontal parallel linkages (fig.2.4.4-5).

66 2. Expansion Joints

Fig.2.4.4-4: Rails supported by beams, spacing controlled by springs

Fig.2.4.4-5: Rails supported by beams. Spacing controlled by horizontal parallel linkages

Fig.2.4.4-6: Rails supported by hinged arranged beams (Swivel System)

Fig.2.4.4-7: Linkage

2.4 Construction of expansion joints 67

If the supporting beams are skew (Swivel System, fig.2.4.4-6) they control the gap width by means of the kinematic characteristic of the mechanism (fig.2.4.4-7). The number of supporting beams does not depend on the number of rails. The higher the number of rails the more economical becomes the application of hinged supporting beams. As an alternative to the application of multiple seal expansion joints, special non-watertight constructions like cantilever-toothed joints or rolling leaf joints (also called roller shutter plate expansion joint) are used. Both the cantilever-toothed joint and the rolling leaf joint are as a rule not watertight, so that an additional drainage system is necessary.

The cantilever-toothed joint (fig.2.4.4-8), also called finger joint, is a very robust con­struction but with several disadvantages. The deformation capacity in the crosswise direction is severely limited and vertical deformations of the joint can prejudice traf­fic safety. To accommodate small vertical deformations without hazard the free finger ends should be rounded. Finger joints with supported fingers (fig.2.4.4-9) have proved to be not as good as with cantilever fingers. The rolling leaf joint (fig.2.4.4-10) consists of a tongue plate, a rocker plate, and sliding plates. The acceptable movement depends on the size and number of sliding plates. Rolling leaf joints can exhibit the following disadvantages: - broken hinges (falling shutter plates cause gaps in the motorway), - wear of the bearing surface, - breaking of the restraining spring elements. Some manufactures have carried out important improvements by: - stronger hinges, - use of specially designed bearings for the shutter plates, - stronger restraining elements with elastomeric springs, - rubber seals between the plates (it makes the joint watertight to a great extent).

Fig.2.4.4-8: Cantilever-toothed joint or finger joint

68 2. Expansion Joints

Fig.2.4.4-9: Finger joint with supported fingers

Fig.2.4.4-10: Rolling leaf joint

2.4.5 Expansion joints for railway bridges For the expansion joints for railway bridges it is necessary to consider two elements: - the rails themselves, - the bridge. Nowadays, a continuous track without expansion joints is preferred, due to the com­fort of the passengers. Therefore, many modern railway bridges have no expansion de­vices for the track. Eurocode 1, part 3 (ENV 1991-3), gives rules for the maximum ex­pansion lengths for continuous tracks. The expansion length (i.e. the distance between the "thermal centre" and the opposite end of the deck) should not exceed 60 m for steel structures with a ballast bed and 90 m for concrete and composite structures, again with a ballast bed. If the expansion length exceeds these values expansion devices should be used.

2.4 Construction of expansion joints 69

Two different types of railway expansion joints are in use by the railway authorities. The first type consists of a parallel joint in the rail and works according to fig.2.4.5-1.

Fig. 2.4.5-1: Parallel joint in the rail

The second type is normally used for high speed railways (fig.2.4.5-2). It consists of an ending rail with a slope of 1 : r, with r = 70 to 100, and a tapered rail which is ma­chined in the same slope. This expansion joint has the advantage that there is no gap between the rails and that the wheel load is carried by a full profile section, but the disadvantage that the rail gauge will be widened by the expansion, according to the slope:

E.g., for an expansion of Δl = 200 mm and with r = 70 m the gauge is widened by Δs 6 mm. The expansion joints of the rails should not be located directly over the gap because of the rotation angle φy of the bridge. It is better to adjust the expansion joint on the embankment, at a short distance from the bridge.

Fig.2.4.5-2: Feathered joint

70 2. Expansion Joints

For bridges without a ballast bed the gap between the bridge and the abutment nor­mally stays uncovered. For bridges with a ballast bed there are two possibilities:

a) to enclose the ballast within the bridge and on the embankment, bridging the gap only by the rails. This construction causes problems to the automatic track ballast tamping machine.

b) to build a continuous ballast bed by means of elastic rubber (neoprene) joints or by sliding cover plates. In this case there is no interruption of the ballast bed and no problem for the tamping machine. On the other hand the compactness of the ballast increases and decreases with the expansion in the area of the expansion joint of the rails.

A completely different design philosophy is to install no special expansion devices on bridges with a continuous ballast bed as described above (and, sometimes, with nor­mal bolted rails joints). If high forces caused by temperature-induced expansion or shortening of the bridge occur in the rail, the track will move in the longitudinal di­rection and will become settled by itself, especially under the vibrations of the passing trains.

The forces between the track and the bridge have to be considered especially for the design of the longitudinally fixed bearing(s) of the bridge. ENV 1991-3 gives the lon­gitudinal action per track FTk = ± 8 [kN/m] (LT1-LT2), where LTi are the expansion lengths from the fixed bearing.

2.5 Materials for expansion joints

2.5.1 Steel parts Normally, the supporting members such as edge elements, rails and cross beams are made of mild steel protected by coating or of corrosion-resistant steel. The stirrups of the fixing are curved reinforcing bars. Stainless steel is used for moveable parts like the bolts of a folding trellis linkage and sliding plates connected with PTFE. Members that are difficult to access, e.g. niches for linkage elements, also are made of stainless steel or corrosion-resistant steel. Steel parts embedded in concrete outside of the zone of carbonation, corrosion-resis­tant steel and stainless steel do not need any protection against corrosion. Parts made of mild steel must be protected. Coatings must have a sufficiently high resistance against mechanical stress, temperature actions, oils, and de-icing salt. The coating should be chosen in accordance with the appropriate national standards. However, a coating consisting of a two-component epoxy priming coat with zinc dust and a two-component epoxy final coat with micaceous iron ore is recommended. Steel parts em­bedded within the zone of carbonation need only a priming coat. A protection against corrosion by means of an elastomeric sheathing is possible if the elastomeric material satisfies the requirements of resistance and durability. In the case of protection by gal­vanising, hot-dip galvanising is the normal case. Spray galvanising is expensive but also possible.

2.5 Materials for expansion joints 71

2.5.2 Elastomeric parts Elastomeric parts must be resistant to environmental influences, de-icing salt, alkaline and acidic water. They are classified in two categories (load transferring and non-load transferring). Load transferring elements (e.g. cushion elements or elastic bearings of the rails) are made from polychloroprene or from natural caoutchouc. The material must be age-resisting, despite the presence of de-icing salt. Non-load transferring elements (e.g. sealings) are made from polychloroprene or from ethylene-propylene-caoutchouc with high resistance to tearing and to crack propaga­tion. The thickness should not be below 4 mm. The following table gives the recommended characteristics of applied elastomers.

Characteristic

Hardness Resistance to tearing Tearing strain Resistance to crack propagation Behaviour after a temperature stress (14d;70°C) Change of hardness Change of resistance to tearing Change of tearing strain Resistance against potassium chloride (solution: 4%; 14d;23°C) Change of volume Change of hardness Resistance against hot asphaltic bitumen (30 minutes; 220 °C) Change of resistance to tearing Change of tearing strain Bond with steel

Tab.2.5.2-1: Recommended characteristics of the elastomeric parts

The springs of spring-linked multiple seal expansion joints are made of polyurethane with a high resistance to crack propagation. The material is able to withstand high strains. It can be compressed down to 20 % of the original length. A further advantage is the good damping characteristics.

Non-load transferring elements 55-65 Shore A min. 10 N/mm 2

min. 350 % min. 10 N / m m

max. +7 Shore A max. -20 %

max. -20 %

max. +10% max. -5 Shore A

max. -20 % max.-20%

Load transferring elements 60-70 Shore A min. 15 N/mm 2

min. 400 % min. 15 N / m m

max. +5 Shore A max. -15 %

max. -20 %

max. +10% max. -5 Shore A

max. -20 % max. -20 % Failure within the elastomeric material

72 2. Expansion Joints

Asphaltic plug joints are made of a special modified asphaltic material. This must have a sufficient flexibility to absorb the movements of the gap, combined with a suf­ficient load bearing capacity. The exact composition of the material depends on the producer. However, the binder material usually consists of bitumens modified with plasticizers and polymers. The aggregates, usually, belong to the basalt group.

2.6 Analysis and design of expansion joints

2.6.1 Buried expansion joints and asphaltic plug joints Expansion joints have to satisfy the requirements of ultimate limit state and fatigue strength design. A buried expansion joint or an asphaltic plug joint must only fulfil the construction requirements given in chapter 2.4. The most important rules are: - The thickness of the asphaltic layer should be at least 80 mm. - The asphaltic layer over the supporting construction must have the same thickness

as over the superstructure and over the abutment. - The length of asphaltic plug joints shall not exceed 700 mm. Thin cover plates should be verified by a calculation. The spread of the load can be considered by an angle of 45 ° (fig.2.6.1-1).

Fig.2.6.1-1: Load spread under a wheel

2.6.2 Single seal and multiple seal expansion joints In most cases the ultimate limit state of a single seal and of a multiple seal expansion joint is analysed correctly, while the fatigue was only considered empirically. How­ever, damage is usually caused by fatigue. Therefore a correct analysis is essential [18; 19]. The loading acts for a very short time. The probability that the axles of two vehicles are at the expansion joint at the same time is relatively small and only one axle need be considered. As a rule, standards contain a design load of the following type to analyse single members of a bridge.

2.6 Analysis and design of expansion joints 73

LR Contact length wheel - carriageway surface

R Static load

φ Dynamic factor

Fig.2.6.2-1: Design wheel load

One rail of an expansion joint carries only the portion Fv .k . s ta t of the load, depending on the rail width b, the gap width s and the contact length LR (fig.2.6.2-4).

Fig.2.6.2-2: Factor aa Fig.2.6.2-3: Arrangement of the wheel loads

Fig.2.6.2-4: Load per rail

74 2. Expansion Joints

LN Effective contact length

S i Gap width

n Number of gaps within the contact length

b Rail width

F v . k . s t a t Portion of wheel load

aa Factor of the influence of the angle between expansion joint and driving

direction (fig.2.6.2-2)

If a 9 0 ° the two wheels of the axle do not cause the maximum loading on the rail at

the same time. This fact can be considered by reducing the influence of both wheel

loads by the factor aα.

Horizontal wheel loads result from rolling friction, acceleration and braking forces,

and from the slope of the bridge. Accelerating and braking of a lorry at the expansion

joint cause maximum loads but this is a comparatively rare case and, thus, is consi­

dered only for the ultimate limit state analysis. Horizontal forces due to rolling friction

act at each overrunning and exert an influence on the fatigue of the material.

Ultimate limit state

The ultimate limit state is analysed with the single wheel loads of an axle and consid­

ering the dynamic factors given in the relevant standards.

The acceleration and braking force are determined from the vertical loading. Edge

profiles and their fixing are designed for a horizontal force due to the full wheel load.

Intermediate profile:

Edge profile:

Coefficient of static friction of the standard

Vertical and horizontal dynamic factor

Contrary to the fatigue analysis, for ULS verifications a horizontally and vertically

fixed continuous girder is a suitable model of the rails. Rails and support beams can

be calculated with the E-P or P-P method because actually no yielding occurs due to

the high applicable design loads.

The ultimate limit state is analysed using the semiprobabilistic safety concept as

follows:

2.6 Analysis and design of expansion joints 75

Fatigue design Failure due to fatigue is the main reason for the observed damage. Three types of fatigue fractures have been observed (fig.2.6.2-5):

1) Failure of the welded joint between rail and support beam 2) Failure of the support beam 3) Failure of the rail

Fig.2.6.2-5: Possible cracks due to fatigue

For the fatigue design, the stress range is of interest. At first it is determined by using the loads given in the standards. The horizontal forces due to rolling friction, slope of bridge and acceleration or de­celeration must be considered. However, they are smaller than the horizontal force due to acceleration and braking. The factor ξ consists of three parts:

Factor due to slope Factor due to rolling friction Factor due to locomotive acceleration/deceleration

Fig.2.6.2-6: Determination of the factors

76 2. Expansion Joints

The vertical load acting on an intermediate or edge profile is Fv . k . s t a t. The horizontal loads are determined as follows:

Intermediate profile: Fh.k.stat = ξ . F v . k . s t a t

Edge profile: Fh . k . s t a t = ξ. Rv.k . s ta t

Fig.2.6.2-7: Dynamic loading of a rail

The contact time t1 of the wheel depends on the contact length LR, the velocity v and the width of the profile b.

The impact load is sine-shaped half period). The circular frequency is:

The impact causes a damped sinusoidal vibration (fig.2.6.2-8). For the ultimate limit state analysis the response in the fundamental mode of the system is of interest. It is considered by the dynamic value given in the applicable standards. Fatigue of material is caused by the stress range. Normally, only the first and second amplitude of Fv.k.dyn

exceed the constant amplitude fatigue limit.

2.6 Analysis and design of expansion joints 77

Fig.2.6.2-8: Dynamic loading and response of system

Fig.2.6.2-9: Dynamic model

The static bending moments in the vertical direction can be determined on the sup­ported continuous beam. It depends on the stiffness of the springs if it has to be taken into account or if the springs can be assumed to be rigid. In the horizontal direction the consideration of the elastic fixing is essential (fig. 2.6.2-10).

vertical

horizontal

Fig.2.6.2-10: Vertical and horizontal static system

78 2. Expansion Joints

It is important to use the dynamic stiffness of the springs because it differs from the static value. Both the spring stiffness and the damping coefficient are determined by overrun-tests. The frequency fh and the damping coefficient can be determined from the recorded time-deformation curve. The spring stiffness ch.dyn in the model is varied until the lowest natural frequency according to the experiments is observed. The logarithmic decrement D of the damping coefficient of a spring-linked expansion joint amounts to approximately 10%. Further possibilities to determine the lowest natural frequency are an analysis by FEM or approximate methods. The following method leads to satisfactory solutions.

The fundamental vibration mode shape of the vertical direction can be described by the static bending line of a continuous girder.

A sinusoidal loading causes the following bending deflection curve:

The following formula leads to the stiffness of the spring:

The application of the formulae of the frequency and the rotational frequency leads to the natural frequency of the vertical system:

With known ch.dyn and equal span widths the frequency fh of the horizontal direction can be determined in the same way. But the system is an elastically-supported contin­uous girder. The following figures show some calculated results.

2.6 Analysis and design of expansion joints 79

Fig.2.6.2-11: Lowest natural frequencies of an elastically supported continuous gird­er

m Mass of rail [kg/m]

Ih Moment of inertia [m4]

fh Lowest natural frequency [Hz]

L Single span [m]

Ch.dyn Dynamic stiffness of spring [N/m]

The dynamic values φ1 and φ2 of the first and second modes of the system are added to the value φ. With an assumed logarithmic damping coefficient of 10%, the fol­lowing diagrams give directly the impact factors φ (fig.2.6.2-12). Either the first or second figure can be used. They are suitable for the vertical and horizontal direction.

80 2. Expansion Joints

Fig.2.6.2-12: Dynamic factors

The horizontal axis of the diagram (b) contains the natural frequency of the system. This version shows the frequency of resonance as the maximum of the graph of the de­sign velocity. The values φ of the resonance frequency are comparatively high. Nat­ural system frequencies near the resonance must be avoided at least for the vertical bending. The recommended distance from the resonance frequency is also indicated in the diagram. With a known design velocity a maximum span of the rails can be de­termined. Longer spans cause higher values φ, leading to a higher stress range. An­other disadvantage is an increasing number of stress cycles exceeding the cut-off lim­it, which means that more than.two modes of the system must be considered. With the values φv and φh the dynamic difference moments can be calculated.

The stress range is determined as follows:

The design load of an axle is higher than the actual load. The nominal stresses should be reduced by the factor fred to get the actual design loads. The value of the factor depends on the ratio between design load and loading due to the real traffic situation. The determination of the actual traffic situation requires extensive data for the real loads and their frequency (fig.2.6.2-13). Infrequent high loads exert an advantageous influence on the fatigue behaviour (overloading effect). The maximum load for fatigue design must be determined considering the real fre­quency of the actual traffic loads (e.g. there may be load components occurring only in one of a thousand cases). Instead of the nominal stress also the design load could be reduced.

2.6 Analysis and design of expansion joints 81

Fig.2.6.2-13: Example of a typical loading sequence

The stress ranges up to the chosen limit are used to determine a constant amplitude stress range that causes the same damage (fig.2.6.2-15).

Fig.2.6.2-14: Fatigue strength curve Fig.2.6.2-15: Constant amplitude stress range

This value when compared with the stress range Δσk.max.dyn provides the factor that al­lows the fatigue analysis with design loads given in the standards to be used. For in­stance, [20] recommends the factor fred = 0.75 for the conditions of traffic in Germany, to be applied to the loads of German Standard DIN 1072. A maximum stress deter­mined in this way is exceeded in only one of a thousand cases. The fatigue design has to fulfil the following equation:

Partial safety factor of the fatigue loading (YFf = 1.0)

Partial safety factor of fatigue strength (YMf = 1.15)

Constant amplitude stress range for 100 million cycles

82 2. Expansion Joints

Can be ascertained by the analyses of the real sequence us­ing the Palmgren-Miner summation (α100 0.4).

Fatigue strength for 100 million cycles

The construction members of the expansion joint are three-dimensional and compact. The fatigue strength σL can be taken from the standard used if it contains a suitable detail category, otherwise tests become necessary. The following testing arrangements were recently used with success (fig.2.6.2-16). The required number of tests is nor­mally indicated by the standards.

Fig.2.6.2-16: Recommended arrangement of the tests

The lifetime of a construction can be calculated as a statistical value. It is only appli­cable for the evaluation of that type of construction.

Design life - time in years The number of cycles exceeding the cut-off limit The average of daily lorry traffic in one direction The average number of axles of each lorry The distribution of the DTLV on several lanes p = 1.0 in case of one lane p = 0.85 in case of two lanes p = 0.80 in case of three or more lanes

2.6.3 Elastomeric cushion joint The loads for the ultimate limit state analysis and the reduced loads for the fatigue analysis are determined in the same way as for the seal expansions joints. In the verti­cal direction the analysed element transfers a portion of the wheel load, depending on the zone of influence. Horizontal loads are determined from the vertical loads using the factor ξ,.

Intermediate profile:

Edge profile:

2.6 Analysis and design of expansion joints 83

The horizontal loading of edge profiles and their fixings are analysed considering the complete wheel load. Edge profiles and fixings can be analysed in the same way as for multiple seal joints. A possible intermediate profile can be treated as a single span beam (fig.2.6.3-1).

Fig.2.6.3-1: Calculation of the intermediate profile

The elastomeric parts of elastomeric cushion joints have to withstand stresses and stress ranges due to traffic loads. Their strength can be ascertained by tests. The fol­lowing testing arrangement is recommended.

Fig.2.6.3-2: Recommended arrangement of the test

The specimen is of the same character as the planned construction and has a length of at least 1200 mm. The loads are applied through an elastomeric disk of 50 mm thick­ness which is situated in the middle of the cushion element. LR and BR are the dimen­sions of the load area according to the applicable standard. If the width of sample is smaller than LR, only a reduced load acts on the joint construction. It can be consi­dered by a smaller disk and a force than P. The inclination of P depends on the factor ξ. It considers the sliding friction or the roller friction, the slope of the bridge and the locomotive's acceleration and is different for the ultimate limit and fatigue tests. The applied force P has the following value for the ultimate limit test:

84 2. Expansion Joints

P = Fv.k.stat

Fv.k.stat Wheel load of the standard

For the fatigue test the loads are reduced by the factor fred.

Pred = fred · P

The construction is applicable if experiments prove that the full load P can be sup­ported as a static load, the reduced load Pred for 2 millions of cycles.

2.6.4 Cantilever-toothed joint and rolling leaf joint The Bernoulli-Euler theory of bending gives correct results provided that the height to length ratio of a beam is at least 1/5. Fingers of cantilever-toothed joints are often not within this range. If this requirement is satisfied the ultimate load can be calculated easily. Otherwise tests become essential. The fatigue behaviour must be determined by tests anyway because of the three dimensional character of the connection cantilever / edge element. The testing arrangement and the applied loads are the same as for cushion joints (fig.2.6.4-1). Maximum stresses are caused when the joint expansion is maximum.

Fig. 2.6.4-1: Recommended arrangement of the test

The behaviour of a rolling leaf joint should be checked in the same way. In most cas­es neither the application of the Bernoulli-Euler theory of bending is possible nor do the standards contain suitable detail categories for the fatigue design. The loads must be placed in the most disadvantageous position.

2.7 Installation of expansion joints

The design of an expansion joint is performed by determination of the extreme values of the expected movements and the position of installation. The installation data depends on the planned construction sequence. The expansion joint is adjusted by means of an auxiliary construction. For a spring linkage prestressing is necessary (fig.2.7-1). It is recommended to instal the expan­sion joint in the early morning when the temperature is distributed almost uniformly over the whole bridge.

2.7 Installation of expansion joints 85

Immediately before the installation the actual temperature of the bridge is measured. If it is not within the considered tolerance the adjustment must be corrected. After that the expansion joint is flushed and fixed temporarily. In the case of a steel bridge it is provisionally bolted or tack-welded. The auxiliary construction must be removed im­mediately. After carrying out the final fixing, the protection against corrosion is com­pleted.

In concrete bridges the expansion joints are provisionally fixed by welding together reinforcement and anchoring. The concrete pour should be at least of the same strength as the adjacent material of the superstructure. While pouring the concrete the joint construction should be protected by a cover.

Fig.2.7-1: Possible auxiliary construction for the installation

In the case of a steel bridge the date of installing the expansion joints has no influence on the expected range of movement. In the case of a concrete bridge or a composite bridge, single unidirectional movements (shortening due to creep and shrinkage) oc­cur. These movements begin with the erecting of the construction and stop within some weeks / months / years. Creep is caused by compressive stresses, especially due to prestressing. The movement due to prestessing forces occurs during the prestress-ing work. The joint construction has to accommodate the movements which occur af­ter the installation. Therefore, the dimension and, by this, the costs of a joint con­struction can be reduced by a late installation.

The variation of creep and shrinkage is shown in the following figures by means of the coefficient of creep φ(ºº,t0) and the shrinkage value εSCºº. In various standards, t = 5 years ( 1800 days) to t = 20 years is set equal to t = ºº.

86 2. Expansion Joints

Fig.2.7-3: Variation of creep Fig.2.7-4: Variation of shrinkage

The maximum increments of shrinkage and creep occur immediately after completion or after prestressing. For example after 100 days (about 3 months), about 50 % of the expected creep deformations and 25 % of the shrinkage deformation have taken place.

2.8 Inspection and maintenance

Expansion joints should be checked regularly by means of visual inspection. The fre­quency depends on the sensitivity of the construction. Before the inspection the joint is cleaned, and cover-plates may need to be removed. The check should involve the following items: - Damage of the anticorrosive protection. This should be repaired before advanced

rust formations appear. The new coating must be compatible with the existing one. - Visible cracks due to fatigue in the steel members. - Damages to the seals. The soiled water of the carriageway can lead to the deterio­

ration and corrosion of the bearings, the substructure and possible the linkages. - Workability of the linkage. If it does not fulfil its function, damage of the seals may

result. - Obstruction or damage of the drainage system. The adjacent carriageway pavement should also be checked. A jutting joint construc­tion due to wheelers enhances the impact loading. If it is not possible to repair the entire pavement, asphalt ramps should be erected to protect the joints.

Service-free expansion joints are often demanded by the manufacturers. Nevertheless, it is recommended to clean the gaps from grit and silt to protect seals and linkage. The drainage should also be cleaned regularly.

2.9 Replacement of expansion joints 87

2.9 Replacement of expansion joints

The lifetime of an expansion joint should be the same as the lifetime of the carriage­way pavement. A complete replacement becomes necessary if the steel parts exhibit advanced fatigue damage. On steel bridges only the bolted or welded connections are removed. A replacement on concrete bridges is more expensive. More frequent is the replacement of single members, especially of the elastomer com­ponents. Seals should be replaceable from the carriageway site. Manufacturers offer different systems for easy replacement (fig.2.9-1).

Fig. 2.9-1: Possible fixings to the seal

The gap width must be opened to at least 25 mm. In the case of an elastic linkage, smaller widths are possible because the rails can be displaced. On the other hand the seals must not be stretched fully. Expansion joints for large movements should be ac­cessible from the underside to change members of the linkage like elastomeric springs. In the case of a road with several lanes it is desirable to change the seals of the expan­sion joint in sections. It is possible to join the seals by vulcanization on site. If a replacement of the rails becomes necessary they can also be joined on site. How­ever, the joints should be situated in zones with minimal stress range and must be welded very carefully because of the high fatigue loads.

88 2. Expansion Joints

2.10 References

Books about expansion joints for bridges: Lee D.J.: Bridge Bearings and Expansion Joints. Second edition by E & FN Spon, London, Glasgow, New York, Tokyo, Melbourne, Madras 1994.

Papers: [1] Price, A.R. (1982): The service performance of fifty buried type expansion joints.

TRRL Report SR 740, Transport and Road Research Laboratory, Crowthorne. [2] Price, A.R. (1983): The performance of nosing type bridge deck expansion joints.

TRRL Report LR 1071, Transport and Road Research Laboratory Crowthorne. [3] Price, A.R. (1984): The performance in service of bridge expansion joints. TRRL

Report LR 1104, Transport and Road Research Laboratory, Crowthorne. [4] Department of Transport (1989): Expansion joints for use in highway bridge

decks. Departmental Standard BD 33/88. [5] Department of Transport (1989): Expansion joints for use in highway bridge

decks. Departmental Advice Note BA 26/88. [6] Koster W. (1969): Expansion Joints in Bridges and Concrete Roads. Maclaren

and Sons. [7] Busch, G.A. (1986): A review of design practice and performance of finger joints.

Paper presented to the 2nd World Congress on Joint Sealing and Bearing Systems for Concrete Structures, San Antonio, Texas, September.

[8] Watson, S.C. (1972): A review of past performance and some new considerations in the bridge expansion joint scene. Paper presented to regional meetings of the AASHO Committee on Bridges and Structures, Spring.

[9] Koster W. (1986): The principle of elasticity for expansion joints. Paper present­ed to 2nd World Congress on Joint Sealing and Bearing Systems for Concrete Structures, San Antonio, Texas, September.

[10] Lee, DJ. (1971): The Theory and Practice of Bearings and Expansion Joints for Bridges, Cement and Concrete Association.

[11] Demers, C.E. and Fisher, J.W., Fatigue Cracking of Steel Bridge Structures, Vol­ume 1: A Survey of Localized Cracking in Steel Bridges - 1981 to 1988, FHWA Publication No. FHWA-RD-89-166, McLean, VA, 1990

[12] Standard Specifications For Highway Bridges. 15th edition, American Associa­tion of State Highway and Transportation Officials, Washington, D.C., 1992

[13] Tschemmernegg, F., The Design of Modular Expansion Joints, Proceedings of the 3rd World Congress on Joint Sealing and Bearing Systems for Concrete Structures, Toronto, 1991.

[14] Dexter, R.J., Kaczinski, M.R., and Fisher, J.W; Fatigue Testing of Modular Ex­pansion Joints for Bridges, Proceeding of the 1995 IABSE Symposium, Volume 73/2, San Francisco, CA, 1995.

[15] TL/TP-FÜ 92, Technische Liefer- und Prüfvorschriften für wasserundurchlässi-ge Fahrbahnübergänge von Strassen- und Wegbrücken. Bonn: Bundesministe-rium für Verkehr, Ausg. 1992

2.10 References 89

[16] Richtlinie - RVS 15.45, Brückenausrüstung - Übergangskonstruktion. Wien: Forschungsgesellschaft für das Verkehrs- und Strassenwesen, Arbeitsgruppe «Brückenbau», Arbeitsausschuss «Brückenausrüstung», Ausg. Januar 1995.

[17] Braun, Chr.: Verkehrslastbeanspruchung von Übergangskonstruktionen in Stras-senbrücken. Bauingenieur 67 (1992), P. 229-237.

[18] Tschemmemegg, F. (a.o.): Ermüdungsnachweis von Fahrbahnübergängen nach ENV-1993-1. Stahlbau (1995), P. 202-210.

[19] Pattis, A.: Dynamische Bemessung von wasserdichten Fahrbahnübergängen-Modulsysteme (Dynamic Design of Waterproof Modular Expansion Joints). Ph.D. dissertation. Department of Civil Engineering and Architecture, Universi­ty of Innsbruck, Austria (Dec. 1993).

[20] Herleitung eines Lastmodells für den Betriebsfestigkeitsnachweis von Straßen-brücken. Forschung Strassenbau und Strassenverkehrstechnik Heft 430, 1984.

[21 ] Ramberger, G.: Bearings, expansion joints and hydraulic equipment for bridges, IABSE, 15. Kongress-Bericht Copenhagen, 1996.

[22] Fisher, J.W., Kaczinski, M.R. and Dexter, R.J.. Field and Laboratory Experience with Expansion Joints. IABSE, 15. Kongress-Bericht Copenhagen, 1996.

[23] Braun, C: The Design of Modular Joints for Movements up to 2000 mm. IABSE, 15. Kongress-Bericht Copenhagen, 1996.

[24] Nielsen, H.B.: The Storebaelt West Bridge. Railway Expansion Joints. IABSE, 15. Kongress-Bericht Copenhagen, 1996.

[25] Crocetti, Roberto: Modular Bridge Expansion Joints - Loads, Dynamic Behav­iour and Fatigue Performance. Thesis for the degree of Licentiate of Engineer­ing. Department of Structural Engineering, Division of Steel and Timber Struc­tures. Chalmers University of Technology, 1998.

[26] Barnard, C.P., Cuninghame, J.R.: Practical guide to the use of bridge expansion joints. Application guide 29, Transport research laboratory, UK 1997.

Advertisement

Structural Engineering Documents

Structural Engineering Documents (SED) is IABSE's distinguished series of book-length mono­graphs. So far, six volumes have been published in the SED series, all written by acknowledged ex­perts in their fields. Each SED volume examines a basic structural engineering problem. This fundamental approach, together with high scientific standards, assures the lasting value of SED volumes for practitioners, students and teachers of structural engineering. IABSE maintains a backlist of this series, which to date includes:

SED 1:

Concrete Box-Girder Bridges

Jörg SCHLAICH, Hartmut SCHEEF

German: ISBN-3-85748-032-7,105 pages; 1982 English: ISBN-3-85748-031-9, 108 pages; 1982, out of print

The concrete box-girder, a widely used bridge superstructure system, is comprehensively examined in this informative volume. Emphasis is placed on practical guidelines for the actual design of con­crete box-girder bridges. The monograph follows the sequence of the bridge design process itself: Part 1, Design, covers design principles, and construction methods as they influence design, for substructures, superstructures and complete systems; Part 2, Structural Analysis, examines general analysis procedures and a wide range of structural parameters, with particular attention given to eccentric vehicle loads; Part 3, Dimensioning and Detailing, discusses prestressing, flange and web dimensioning, diaphragms, abutments, construction joints, bearings, piers and bridge finishes.

The attempt of the authors is to provide the design engineer with a compact guide to the essential parameters for the design and analysis of concrete box-girder bridges. Each part has its own list of references, and each proceeds from general principles to specific problems.

SED 2:

Dynamic Response of Reinforced Concrete Buildings

Hajime UMEMURA, Haruo TAKIZAWA

ISBN 3-85748-029-7, 64 pages; 1982, out of print

The performance of reinforced concrete building structures subjected to earthquake-induced mo­tions is the subject of this SED volume. The book emphasises formulations of a basically empirical nature to analyse the complexities inherent in the inelastic and hysteretic response of reinforced concrete structures.

90

Advertisement

SED 3:

Vibrations in Structures

Induced by Man and Machines Hugo BACHMANN, Walter AMMANN

English: ISBN 3-85748-052-X, 176 pages; 1987 German: ISBN 3-85748-051-3, 193 pages; 1987, out of print

This SED volume shows where dynamic problems can occur in structures and presents appropriate countermeasures. Evenly divided between theory and case studies, the monograph first surveys the broad topic of structural dynamics. Various load categories are defined, followed by an account of "source-dependent" effects from human activity, machine operation, wind, water, earthquakes, traf­fic, and explosions. Man-induced vibrations are considered, looking at dynamic loading arising from walking, running and other forms of movement, while machine-induced vibrations are examined according to type (rotating, oscillating, impacting). For both types of vibrations possible counter-measures are proposed.

The second part of the book presents 22 case studies of a wide range of structures: footbridges, sports halls, halls for pop concerts, diving platforms, and eight industrial structures with machine-induced vibration problems. The final section discusses the fundamentals of vibration theory to assure that the reader has an adequate understanding of the problems presented.

SED 4:

Ship Collision with Bridges

The Interaction between Vessel Traffic and Bridge Structures

Ole Damgaard LARSEN

ISBN 3-85748-079-3; 131 pages; 1993

This SED volume is aimed at engineers responsible for the planning, design and assessment of bridges crossing navigation channels. It is a comprehensive source of information on the risks of ship collisions with bridges and presents methods to evaluate the safety of bridges, people and the envi­ronment. It emphasises collision prevention and identifies measures for the protection of structural parts of bridges in the event of ship collision. The monograph also offers advice on the upgrading and retrofitting of existing bridges and navigation channels.

After reviewing some basics of navigation and vessel traffic, and considering collision risk and risk acceptance, the volume focuses on vessel impact forces on bridges and proposes appropriate design criteria. Collision prevention measures such as regulations and navigation channel management systems are also addressed.

91

Advertisement

SED 5:

Introduction to Safety and Reliability of Structures

Jörg SCHNEIDER

ISBN 3-85748-106-4; 138 pages; 1997

Structural engineers devote all their effort to meeting society's expectations efficiently. Engineers and scientists work together to develop methods for solving engineering problems. Given that no­thing is absolutely safe, the discussion of safety can only be in terms of (acceptably small) failure probabilities. Starting from this premise, reliability theory emerged and has become part of the science and practice of engineering today. Its application is not only with respect to the safety of structures, but also in regard to serviceability and other requirements of technical systems that are all subject to some probability of failure.

The present volume takes a broad approach to the safety of structures and related topics. The first chapter introduces the reader to the main concepts and reviews strategies for identifying and mitigat­ing hazards. The second chapter is devoted to processing diverse data into information that can be used in reliability analysis. The third chapter deals with the modelling of structures, and the fourth presents recognised methods of reliability theory. Another chapter focuses on problems related to establishing target reliabilities, assessing existing structures, and the need for effective strategies against human error. The appendix supports the application of the methods proposed and refers read­ers to a number of related computer programs.

This book is aimed at both students and practising engineers. It presents the concepts and procedures of reliability analysis in a straightforward, understandable way, making use of simple examples, rather than extended theoretical discussion. It is hoped that this approach serves to advance the application of safety and reliability analysis in engineering practice.

You may order SED monographs, or get information on other IABSE publications, directly from:

IABSE, ETH Hönggerberg, CH-8093 Zurich, Switzerland Phone: +41-1-633 2647 Fax: +41-1-633 1241 E-mail: secretariat@iabse.ethz.ch Web: http://www.iabse.ethz.ch

92

Structural Engineering Documents

Objective: To provide in-depth infor­mation to practicing struc­tural engineers in the form of reports of high scientific and technical standards on a wide range of structural engineering topics.

Topics: Structural analysis and design, dynamic analysis, construction materials and methods, project manage­ment, structural monitoring, safety assessment, mainte­nance and repair, and com­puter applications.

Readership: Practicing structural engi­neers, teachers, researchers and students at university level, as well as representa­tives of owners, operators and builders.

Publisher: The International Associa­tion for Bridge and Structur­al Engineering (IABSE) was founded as a non-profit sci­entific association in 1929. Today, it has more than 4200 members in over 100 countries. lABSE's mission is to promote the exchange of knowledge and to ad­vance the practice of struc­tural engineering world­wide. IABSE organizes con­ferences and publishes the quarterly journal Structural Engineering International, as well as conference re­ports and other mono­graphs, including the SED series. IABSE also presents annual awards for achieve­ments in structural engi­neering.

For further information: IABSE ETH Hönggerberg CH-8093 Zurich, Switzerland Phone: +41-1-633 2647 Fax: +41-1-633 1241 E-mail: secretariat@iabse.ethz.ch http://www.iabse.ethz.ch

Structural Bearings and Expansion Joints for Bridges

Bridge superstructures have to be designed to permit thermal and live load strains to occur without unintended restraints. Bridge bearings have to transfer forces from the superstructure to the substructure, allowing all movements in directions defined by the designer. The two functions -transfer the loads and allow movements only in the required directions for a long service time with little maintenance - are not so easy to fulfil. Differ­ent bearings for different purposes and requirements have been developed so, that the bridge designer can choose the most suitable bearing.

By the movement of a bridge, gaps are necessary between superstructure and substructure. Expansion joints fill the gaps, allowing traffic loads to be carried and allowing all expected displacements with low resistance. Ex­pansion joints should provide a smooth transition, avoid noise emission as far as possible and withstand all mechanical actions and chemical attacks (de-icing) for a long time. A simple exchange of all wearing parts and of the entire expansion joint should be possible.

The present volume provides a comprehensive survey of arrangement, construction and installation of bearings and expansion joints for bridges including calculation of bearing reactions and movements, analysis and design, inspection and maintenance. A long list of references deals with the subjects but also with aspects in the vicinity of bearings and expansion joints.

This book is aimed at both students and practising engineers, working in the field of bridge design, construction, analysis, inspection, maintenance and repair.