St ructuraI Eng i neeri ng Documenjs

Günter Ramberger

Structural Bearings
and Exmnsbn Jobs rn
for Bridges

International Association for Bridge and Structural Engineering IABSE

Association Internationale des Ponts et Charpentec AlPC
lnternationale Vereinigung für Brückenbau und Hochbau IVBH
 Structural Engineering Documents

Günter Ramberger

Structural Bearings
and Expansion Joints I

for Bridges

International Association for Bridge and Structural Engineering IABSE

Association lnternationale des Ponts et Charpentes AlPC
lnternationale Vereinigung für Brückenbau und Hochbau IVBH
Dedicated to the commemoration of the late Prof. Dr. techn. Ferdinand Tschemmernegg,
University of Innsbnick.

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 hridge designers and not to mention all
the requirements for the manufacturers of such products. These requirements are
usually covered by product guidelines. which vary hetween 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 1. S . Leendertz, Rijkswater. t. Zoetermeer; Eugen Brühwiler, Swiss

Federal Institute ofTechnology, Lausanne; Prof. R. J. Dexter, University of Minneso-
ta; G . Wolff, Reissner & Woiff, Weis; O. Schimetta t, Anit der O 0 Landesregierung.
Linz: Prof. B. Johannsson, LuleA Tekniska Clniversitet, for amendments, corrections,
remarks and comments. I thank also my assistant Dipl.-lng. Jorgen 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 ofthe manuscript.
Finally. I would like to thank the IABSE for the puhlicaiion of this Structural Engi-
neering Document.

Vienna, April 2002 Günter Ramherger

Table of Contents J
1. Bearings
I , I Introduction 7
I .2 The role of bearings 7
I .3 General types of bearings and their movements 7
I .4 The layout of bearings 9
I .S Calculation of bearing reactions and bearing movements 16
I .6 Construction of bearings 19
1.7 Materials for bearings 29
I .8 Analysis and design of bearings 33
1.9 Installation of bearings 31
I . 10 Inspection and maintenance 38
I . 1 1 Replacement of bearings 39
I. I2 Codes and standards 41
I. I3 References 42

2. Expansion Joints
2. I Introduction 51
2.2 The role of expansion joints 51
2.3 Calculation of movements of expansion joints 51
2.4 Construction of expansion joints 58
2.5 Materials for expansion joints 70
2.6 Analysis and design of expansion joints 72
2.7 Installation of expansion joints 84
2.8 Inspection and maintenance 86
2.9 Replacement of expansion joints 87
2.10 References 88
1 Bearings J
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 limes 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 of-
ten 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 develop-
ment of ageing-, ozone- and UV-radiation-resistant elastomers and plastics. new ma-
terials 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 compres-
sion 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 ofthe 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 dispkacements 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 niovenient 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 com-
bined.
A special type of bearing transmits only horizontal forces. while allowing vertical
displacements.

The following table (Table 1.3-I ) shows the common types of bearings, including the
possible bearing forces and displacements. Friction and elastic restrain1 forces are not
considered.

Symbol Function C0"SlNCtim

All translation Point rwker bearing
filed Pot bearing: Fixed
Rotation all elastomcric bearing:

-
mnd Spheical bearing
Horizontal Consu. pain1 m k e r
movement in sliding bearing:
me direction Constr. pat sliding
Rotation all bearing: Const.
mud elastomenc bearing;
Constr. rpheical
sliding bearing
Honzmtal Free paint m k e r
mwemtnl in bearing: Free pat
all directions sliding bearing: Free
Roiation all elastameric bearing;
mund Free spherical slidin[
bearing; Linkbcarint
with universal joints
(tension and
compression)
All Uanslauon Line mkwbenring

I fixed
Rotation
Leaf bearing
(tension and

+
about one o l i s eompessim)

HorironLll Roller bearing: Link

movement in bearing (tension and
one direction compression):
Ratation Conslant line mcker
about O M axis sliding M n g
Horimntal Free m k e r sliding

4- movement in
all direction
Rotation
abwt one axis
bearing: Free roller
bearing;Free link
M n g

All honwntal Horizontal farce

Uanrl.Rxed bearing
@ Rotatimall
round
Horizonial Guide bearing
__ movemeni in
__ medirection
Rotation all
mund
1.4 The layout of bearings
1.4.1 General
Bearings can be arranged at abulnicnts and piers (iig.l.4.l-I; fig.1.4.1-2) under the
webs of the miin girders. under diaphragms (fig.1.4.1-3). and under the nodes of
truss bracings. The webs and the diüphragins of concrete bridges have to he properly
reinforced against tensile splitting; steel bridges need stiffeners in the direction of the
bearing reactions to transfer the concentrated hearing loads to the superstructure and
the substructure. Abutments and piers also have to be properly reinforced under the
bearings against tensile splitting.

Fi~.1.4.1-/:BeurinR.7 ut un abiirvtent

Fig. 1.4.1-2: Beoririgs UI u pier

The layout of the bearings should correspond to the structural analysis of the whole
structure (super- and substructure logether!). If the setllement and the detlection of the
substructure can be neglected the structural analysis of the superstructure. including
the hearings, can be separated from that of the substructure. Sometimes the model for
the analysis, especially of the superstructure. will be simplified by assuming the fol.
lowing: hearings are situated directly on the neutral axis of the girder (fig.1.4.1-6)*the
motion ofthe bearings occurs without resiraint, bearings have no clearance, etc. In this
case we must consider the correct system (fig. I .4.1-S)at least for the design of the
hearings and take into account the influence of the simplifications on the structure.

~.

Fig. 1.4.1-4: R e d i -

A
Fig. 1.4.1-5: Crirrccr system

A-.
‘j . ~

‘sPlac@
‘YSICOi

Fig. 1.4.1-6: Simplified.

On the abutments or separating piers it is normal to use at least two vertical hearings
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 re-
strained. More than three vertical supports ofthe superstructure lead to statically in-
determinate bearing conditions. but even the siniplest bridge has at least four vertical
bearings. If the torsional stiffness of the superstructure is low (e.g. open cross sec-
tions) it may he neglected and the layout with four bearings becomes isostatic. If the
torsional stiffness is not negligible (e& box girders) we have to take i t 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 he used to prestress the superstructure and to distri-
bute the hearing reactions as desired.
If the bearings are sitwated (nearly) in a plane we need at least one horizontally fixed
and one horizontally moveable bearing. The moving direction must not be orthogonal
to the polar line from the fixed tu 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.

I n general, there are two possihiliiies for the arrangement of the bearings:
a) arrangement in a horizontal position (fig.I .4.1-7)
b) arrangement in a position parallel to the road or rail surface (fig. I.4. I -R)

,
1d.2 The iayout for different types of bridges
For single span girders the layout ofthe 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 ofthe bearings for a bridge with two main gird-
ers consisted of one fixed and one lengthwise moveable bearing at one abutment and
onelengthwisemoveableand one free bearing at theother 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 hearings at the abutment, but it has the
disadvantage that horizontal forces in the crnss direction (wind) and temperature dif-
ferences cause horizontal restraint forces, provided that bearings have no clearance on
the abutments.

The author prefers the statically detenninate system with only one lengthwise re-
strained bearing at the abutment concerned because the actual clearance of a hearing
is notdeterminable in reality (tig.l.3.2-2).

Fig.1.4.2-2: Horizontullv .staticuIlv determinute stem (better thun rlassicul Iuyoiit)

Fig.1.4.2-3: System with srpurated i,erticul urid horizontal hearings (srutically deter-
minate system)
For skewed or horizontally curved single span bridges we have to decide whether the
horizontal force should be conihined with the higher or with the lower veríical reac-
tion Force, For all bearing construciions 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. I .4.2-4)

fi 11
Fig.l.4.2-
clination ($1 resi inr,furcr

Thus, the horizontally constrained bearings for skewed bridges should be placed at the

r
obtuse corners of the bridge, for curved bridges at the outer side (tig.1.4.2-5).

Fig.1.4.2-5: Skewed bridge

11
Fi& I.4.2-6: Luyout .for cow riniioiis gir(1rr.s
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 he reduced
to a round column with one hearing 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. I .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 bear-
ing. The main movement of an expansion joint should be in the axis of-the traflic 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. I .4.2-8).

Fig.1.4.2-7: L«yoict,fi)rrurvid bridges

Fig.1.4.2-8: íLJ.sout f i r curved continuous girders ( n o constraint under overall tern-

peruture)

Fig.I .4.2-9: Geonirtrii<.ulsinrritioii

The elongation is

Arl = k . rl << r,
A, = k . r << r

k proportional elongation

The rotation is

For Qi = 0,the bridge simply rotates as a rigid hody without constraint

One special case of this general rule is well known: the bearings are moveable in the
direction of the polar lines with a = O (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.

1.4.3 Special bearing conditions, advice etc.

It is important to note that the layouc of the bearings has a great influence on the struc-
tural system. The above mentioned arrangements of bearings are typical for average
bridges. The following examples show some special effects which have to be consid-
ered for the design of bridges and bearings. These examples do not lay claim to com-
pleteness.
a) The already mentioned hearing layout, consisting of one bearing fixed in all sliding
directions and one fixed lengthwise at one abutinent, leads to high constraint forces
not only under horizoiital hut also under eccentric vertical loading (fig. I .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 hearing under the lower tlange.
 IP

k-
7
Fig. 1.4.3-1: Pn9irig efecr due to 11 eccentric h d i n g

a
/
-
-
-~-(&~-j:-z&
b) A similar situation occurs lor a coniiniiouh girder with chequer pattern loading.

f -

7
-. ~ ~~~

Fig. 1.4.3-2: Piyirig effect dire to i.hei/iiPi-l>~itt<’rn

loading

c) It is noigenerally 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: Piyirig.f<’rmsjijr II .s!wvced brirl#e with verticcil louding

Similar effects can occur f«r 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 siiuated, and in conihination with ihe substructure. The defleciion of the
substructure can influence the constraint hearing reactions significantly.

1.5 Calculation of bearing reactions and bearing movements

1.5.1 Actions
According to EuroL.de 1 (ENV 1991)the actions can be suhdivided into:
- permaneni actions,
- variable actions.
-extraordinary actions,
 1.5 C'Aciil;i:ion (11 hearing renctii>nr and hearing movements 17

The bridge should take up the desired shape under all permanent loads, at the average
teiiiperiitiirc i+ 10" C' in most of the European countries) and, if time-dependant
displacciiieiits occur, at the time t = -, at which time all moveable bearings should be
in tlie x r o adjiistinent (null position). Variable actions and extraordinary actions lead
to deviatioii from this form.

Variable iictions to consider are:

traffic Iwads. considering the applicable dynamic coefíicients
~ loads doe tu traffic loads, ;.e.
nosing forces
centrifugal lorces
hinking íorccs
tractioii forces
- wind Iilalls
wind o n constiuction
wind o11traftic loads
- setticinenis 0 1 tihiitnients and piers
-- thermal actions
iiiii form temperature
vcrtic;il ieniperatiire gradient
hwimntiil temperature gradient
ieiriperaiure differences hetween individual parts of the bridge (e.g. stay
cahlcs, pylon and stiffening girder)
- creep arid shrinkage of concrete
E:xtriiordiiiiiry actions to consider are:
- earthqiiukc actions
~~ YClliClC impact
~ dei-ailment
-- rupture of the conductor line
others

1.5.2 Bedring reactions

For pcriii;inent actions such as self-weight of the construction. dead load and pre-
stressing. the he;iriiig reactions cuii he calculated as one load case.
For the analysis of tlie hearings it is necessary to consider d rent combinations of
tlie henririg reactioiis:
- iiiaxiiiiiim venical force and the adjacent horizontal force.
- tiiiniiiiuiii veificül force arid the adjacent niaximuni horizontal force,
~ iiinxiiiitiiii horizont;il force and the adjacent maximum vertical force.
- iiiaxiiiiiiiii horizontal force and the adjacent minimum vertical force.
The sitiiplc.\t w i y to ohtaiii these combinations i s to calculate the variable actions, es-
pecinlly tlie trnfiic I»ad. according to the influence line. One should bear in mind that
h o r i m m l iictioiis such as centrifugal forces or braking forces are proportional to the
vertical iraltic load. hut other load uch as wind or traffic or traction forces for rail-
ways. are not.
To obtain the extreme beanng reaction it i q necessary to consider that all bridges are
three-dimensional and not merely plane systems.
The influence lines (influence surfaces) of the bearing reiictions can be found as the
displacement curves (displacement surfaces) of the system, due to unit displacements
6 = 1 or cp = I, 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 model\ are used for the analysea
special care is necessary, particularly with continuous girders with open or box sec-
tions. The following examples demonstrate the difference:

-....-...

Fig.l.5.2-I: Influence areafor the verfical bearin8 reaction A, box .secIioii.

r
Fig.l.S.2-2: Influence area for the vertical beUriilR reucfionA. open ,section.

1.53 Bearing displacements

As already mentioned, the zero adjustment (null position) of every bearing has to be
defined. The displacements are measured fmm 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 de-
fined 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 = I.
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 consi-
dered.
1.6 Construction of hearings

Combinations
iiastomeric bearing wkh
beanng ixing device

EII3
Anchored elast. bearins 111 iransbtkns h e d movement in one dir.

Jnidirectional guided MuHidiirectional nonguided

.me pol slidinq beanng [Constr pol siding bearing

Yree point rocker beanng IConstr. p ~ nmcker

t beanng

Ll
ILine rocker tmarinq ioiier bearing I

,)$A,
hide bearing
I
1.6.1 Elastomeric hearings
Elastomeric bearings are the simplest types of bearings. In the basic mode they con-
sist merely of an elastomeric block (usuaily rectangular or round). The elastomenc
works as a soft part between sub- and superstructure and allows movements in all di-
rections by elastic displacements or rotations. Under vertical loads the elastic block
bulges, leading to vertical displacements. A solution to this problem was found by re-
inforcing the elastic block by thin horizontal steel plates, vulcanized to the elastomer
(fig.l.6.l-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 possihle if the friction between bearing and sub- and su-
perstructure is sufticient. The fnction forces Fdepend on the compressive force C and
the friction coefficient p, with F = C . b. 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 he achieved by applying VUkdnIZed plates on the bottom and on the top of the
bearing. which can be connected to the sub- and superstructure by bolts. pins or ap-
propriate shapes (fig.I.6.1-2).

I I

I I
Fig.1.6.1-1: Elastonieric hewing (ununchored)

Smaller, short time. horizontal Iorcrs 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. I .6.I-3 ; fig. 1.6.1-4).
 I

Fi,q. 1.6.I-.?: Elo,sroirierichrnrinR consrrainr

I I
Coiiibination:elartomeric bearing and steel construction fixed in one direction.

I
Fig. 1.6.1-4: Fixed elastomeric bearing

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

1.6.2 Steel bearings

Steel bearings are the oldest type of bearings. They have been used for more than I 0 0
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 ihe second case of line rocker bearings. These
bearings allow rotations in all or in one duection, 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
hearings and line rocker bearings can exhihit 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 Heinnch
Hertz found the solution of this problem: caused by the elastic deformation the theo-
retical 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 hearing and sub- and
superstructure. Therefore. steel bearings normally need thicker plates for the stress
distribution than other types of bearings which transfer the hearing reactions over an
area.
Point rocker bearings are used fix hearing reactions in the range SO0 and 2.500 kN,line
rocker hearings and roller hearings for load‘, in the range 200 and 20000 kN.

I “,,,,,U
I
Fig. 1.6.2-2: Point rocker h ~ a r i r (wnsrriiinr
i~ in one direction

Fig 1.6.2-3: Fwe poinl rocker hearing

Fig.1.6.2-4: Line rocker heriring

 I
R#.1.6.2-5: R i i l l ~ i ruil)

The convact zones of steel bearings cannot he protected against corrosion. Therefore
corrosion-resistant layers o f high alloyed steel should he used for the contact areas.
This can be done by building tip 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 rnm) of the hardened layer hoth on the
roller(radiusRinmrn)andoftheplatesh«uldbrt>0,14~R-2.

1.6.3 Pot bearings

These hearings were invented in the 1950s. They combine the two desirable proper-
ties: rotation capacity with a very sm:i11 resisíalice and transmission of the bearing
reaction over a defined area.

The pot hcaring consists of a steel pot. filled with an elastomeric disc and a lid or a
piston to the top (fig. I.6.3-I). When subjected to high compression forces, the unrein-
forced elastorneric disc behaves similarly to a liquid. Rotations can occur due 10 the
nearly constant volume of the clastomer (v = 0,s).Of great imponance is the sealing
between the elastomenc pad and the lid: if this sealing has a defect ihe elastomenc pad
escapes like a viscous liquid.

The standard type of poi hearing all«ws only rotation (fig. I .6.3-7).Vertical forces are
transmitted to the pad. horizontal forces froin 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 polyteirafluorethylene
(PTFE) disc. a surface of polished stainless steel connected to a sliding plate of struc-
tural steel and luhncation grease. PTFE i s a plastic with high mechanical and cherni-
cal resistance, great toughne ind very small friction when comhined with polished
stainless steel. The PTFE disc i s 5 to 6 nini thick, where half a thickness i s 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,m»vement i n one direction an additional guide i s used for the lid. This
guiding device allows movements i n only one direction (fig. I.6.3-3).
Pot hearings are used for vertical bearing forces from lO(x) kN up to 1OOooO kN.
Depending on the standard applied the allowahle compression hetween lid and elas-
tomeric pad should not exceed 4.0 kNlcm'. The allowable compression for the PTFE
is 3 kN/cm' for permanent loads and 4.5 kN/cm' for short term loads (trafic, wind etc.).
Pot bearings have the advantage of a very high vertical stiffness (nearly incompres-
sible 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.

Elastomere disc
Lid
Sealing
-
Pot wall
I Pot - bottom
Fig. 1.6.3-1: Function of u pot heriring

Fig. I . 6.3-2: Fixed pot berirIng

Fig. 1.6.3-3: Pot heuriq construint in one direction

 Anchoring plate
Sliding plate
Polished stainless s l e d
PIi+ (Pol?tetr.lfluornh'l~"~
Lid
Pot - un11
seu1ing
Elastomcrr disc
Pot - hotturn
Fig 1 . 6 3 4 : Members of a pot hetiring

Fig. 1.6.3-5: Free poi 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-I). 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 pans 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.
Like pot bearings. spherical bearings are tised for venical forces i n the i-iii~gcof I O ( H l
to 100000kN.

Polished Sliding plate

stainless steel
I

PrFr: Chrome plated

polished surface
Fig. 1.6.4-1: Mr.niber.s o f u .sphrrird hruriiig

I
Fig. 1.6.4-2: Fix rpheriad heuririg

=I=
Fig. 1.6.4-4: Free sphericui hewing
Fi,y.l.6.4-5: Italiun und British .spherivul h c U r ~ l(one
l ~ slidinx .surf¿¿ce)

1.6.5 Leaf and link bearings

All the ahove mentioned bearings are ahle to transfer compression forces. If tensile
forces as well as compressive forces must he transferred, leaf and link hearings are
used. These hearings can only transmit forces i n the direction of the leaf. To transfer
forces in the crosswise direction. separate bearings must he used.

A leaf hearing 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 hy a pin. Leaf hearings al-
low free rotation in one direction. Pin and pin holes must have a tit less than 0.3 mm.
as in cases of greater slackness aiid changing forces the pin will punch the hole. Pin
plate and pin should he o f differciit types of steel to avoid seizure. Pin plates are made
of structural steel, pins often of tempered steel.

For link bearings a pendulum i h linked to the foot leaf aiid to the upper leaf by pins.
Link bearings ullow rotation and displacement in one direction. For pin holes and pins
the same rules apply as given for I r a l bearings.

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

Displacements 6 o f link bearings are always combined with a small displacement &,
in the perpendicular direction. 6, =
6'
2R
. with R equal to the distance between the
axes of the pins. Therefore this distance should not he too small.

1.6.6 Disc bearings

Disc hearings were introduced in the late 1960s. The vertical loads are transferred by
an elastomeric disc made of polyether-urethane polymer. I n contrast lo a pot hearing a
transverse extension of the elastoiiieric disc i s possihle. Bearing capacity and func-
tioning i s comparable with an elasionieric hearing. Rotations around the horizontal
axis are transferred by differential deflection ofthe disc. The rotations cause a shiíi of
the axis ofthe load from the centre o f hexing, which must be considered in the design.
Horizontal forces are transferred hy a shear-restriction device which allows vertical
deformation and rotation. The hasic type is a fixed bearing. Free bearings are con-
structed by additional sliding elements and ( i f nece r y ) guiding systems.
Fig.1.6.6-1: Fired beurinx

Fig. 1.6.6-2: Uni-directi<inalguided

Fig. 1.6.6-3:Mulii-dirrctionul non-guided

1.7 Materials for bearings
1.7.1 Steel
Shuctural 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 I0025
- Fine-grained structural steels according to EN 101 13
- Quenched and tempered steels according to EN I0082

Eurocode 3 may be used for the design of all bearing components made from struc-
tural steel according to EN 10025 and EN 1 0 1 13 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 conlact
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 IS0 683 can also be used for bear-
ings. For design one should use EC 3, part 1-4. Concerning stainless steel for sliding
plates see I.7.3.

1.7.2 Elastomeric parts

Elastomeric pans of bearings consist normally of natural or artificial (chloropren)rub-
ber (NR or CR, respectively). Artificial ruhher has the same good properties as natu-
ral rubber, and in addition it has a higherresistance against ozone, ultraviolet radiation
and ageing and is more rigid. The characteristic mechanical properly is the shear m d u -
Ius G between 0.7 and 1.15 N/mm’ 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 . ( I + v ) . G = 3 . G . The fracture strain of ruhber lies between 250 T¿ and 500 %#.Rub-
ber creeps under stress by up io SO % 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 slresws. Compressing a rubber pad changes its shape. The
changing of the shape depends on the possihility of displacement at the compressed
areas. If the compressed areas are tined to a rigid surface, the displacement remains
small. Thus we obtain the inequality A , > A 2 > A? (fig, I .7.2-1).

i l r p g
~~

Fi,q.1.7.2-1: Veriical disp1aceiniwr.srlependin,q on rhc Iutrrul expunsion

 i

4
,
Fig. 1.7.2-2:Stress distrihurion

If the surface of the rubber is lixed to a rigid hxly shear stresses develop between the
two surfaces under compression (fig. I .7.?-2).Under compression we obtain a virtual
modulus of elasticity E,,,,,,,, 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 E,,,,,,,, is given by

The maximum stresses under compre.;sioii hetween two rigid bodies are

with 0= -.F
F compression force.
ab
For bending, the effective modulus of elasticiiy E, knd,ny is lower than E,,,,m, 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
same as under conipression. hut with u +a we obtain E, k,id,,ip =
I
E; '<,,"*. Actually,
2 4
the maximum o is not in the middle of one half but nearer the outer side; thus we
a I
finally obtain: a + < E: b r ~ n e= 5 E, I,~,,,~,,.
This is descrihed very well by the
f«llowing approximate formula:

for b 2 a
a
Under the rotation a we obtain a curvature p = = M!...--with 1 = a ' b
~
~ ~

.I
d E,k,Ki*"# 12
and a restraining moment

~ .. .
-
~ . .~. - -..

,%
"K

I
I-: ~. ~~~ I I
Fig. 1. 7.2-4: Displacernenr - resirainingfiJn'es

1.7.3 Sliding elements

For sliding elements in constructional bearings it is nornial to use PTET, also knowii
by the registered trade names Teflon and Hostaflon. PTFE is a so called themioplast.
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 plate\. They are allowed for convex elements
only.
The combination o f a soft and a hard part has the adviuitagr that there is n« 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 pwkets 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 inm. Un-
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 dy-
namic coefficient. After movement has taken place the dynamic friction coefficient re-
mains 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 orientat-
ed into the direction of motinn within a very thin surface layer. When the motion is
stopped, the orientation is lost within IIfew hours. Fig. 1.7.3-1 shows the design val-
ues of the friction coefficient &, between PTFE and stainless steel, depending on the
compression force (EN 1337-2).

0.10
pd t
i I I
0.00 I I I
, I
I
1 c
0.0 0.5 I .o 1.5 2.0 2.5 3.0 p IkN!cm']
Fig. 1.7.3-1: Fricrion coefficient dependin8 on /he compressionjiwce

The design value of the ultimate compression load is

f, = h , 5 ( I - 0.02,(19 - 30.C)) kN/cm' for i9 2 30'C
19 : maximum temperature ofthe bearing.

The wewing of the iTFE 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 siainless steel surface
e) the contact pressure near the edge of PTFE (ironing effect)
For slow movements caused by thermal actions we obtain long sliding movements hut
at a low velocity. Quick movements caused by trafiic loads have short sliding move-
ments hut they occur at high velocity. Wearing ir. mostly caused by the second case.

For the stainless steel plate. austenitic steel X6CrNiM017122 according to EU-
RONORM 88-2. surface n (Illc), should he used. The stainless steel plate must cover
the F'TFE plate completely i n all sitiiations. The thickness of the plate should he at
least of 1.5 min. The connection to the carrying plate of mild steel can he welded or
glued. For 2.5 mm thick plates the connection can be riveted or hoked.

1.8 Analysis and design of bearings

1.8.1 Hertz compression
For the design of bearings the following problems should he addressed: compression
between two spherical bodies. compression between a spherical and a flat body, com-
pression between two cylindrical bodies. compressionbetween a cylindrical and a flat
body along a generator line. As already mentioned. Heinrich Hertz ohrained the solu-
tion under the following assuiriptioiis (1881):
I . The two bodies consist of iwtropic. homogeneous and infinitely elastic materials.
2. Only normal stresses (no shear stresses) occur at the contact areas.
3. The radius (width) o f the contact areas is small compared with the radii of the
involved bodies.
Hertz found the following maximum compression stresses max d and widths b on the
contact areas:

Spherical body on spherical body

Cylindrical body on cylindrical body

with

__ niax o

F beanng reaction
I length of the cylinder
r,. h radii of the bodiei in contact
E Young's modulus Fig.1.8.1-2: Stress distributiori
V Poisson's ratio (v = 0.3 for steel)
max cs maxinium nomal stress at the contact area
b half the width ofthe contact zone

For the usual rocker or roller bearings the max (J 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 hut also horizontal compression stresses. According to the
von Mises criterion the comparison stress
~
.. ~~~

/ 1 3
Ov =v'q,- f O i ' + ( J ? - ~ I ~ ~ - ~ ? ~ ? - ( J Iand
~ I yielding begins when Ov reaches

the material yield strength fy.I n the present three-dimensional compression regime.
(J~,will be less than úi and yielding will not hegin until oi= f,.. On the other hand, the
maxiniuin 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.

EN 1337-4 - roller bearings - gives for the design line load pd of a roller hearing
f,'
(cylindrical body on flat surface): p<(2 I X ' R ' with
~

Ed

f, tensile strength of the material

R radius of the cylinder
E,design value of the modulus of elasticity
Compared to Hertz's formula with

rnaxo, =0.418. íLT

dR
we find
f , 1.77.f,, =o,,,
m a x o , , S 0 . 4 1 8 . ~ 1 8 ~=

EN 1337-6 - rocker bearings -gives for the design load F,,dof a point rocker hearing
f '
(sphere against plane surface) F, ,, S 170. R' . + .
E, -
Compared to Hertz's formula with
, ~~~ ~

we find
maxo,,<0.388.~~170.f,=2.1Sf,=o,,,.
For cylindrical rocker bearings the same formular as for roller hearings art used.

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

Fig.1.8.2-1: Load disíriburioii lo rhe pin

In the case of fig.1.8.2-I we obtain the shear force and the bending moment according
tofig.1.8.2-2andfig.1.8.2-.3.
Fig. 1.8.2-2: Sheur,fiirce

Fig. 1.8.2-3: Bending momenr

h
For normal bridge hearings we have: c =O. a = ~ .
2
The design values for the resistances are
d k
Shear: F,,,, =0.6.A .fup/yM, = 0.6. -.fup/yMp= 0.47 I .d’f,, /y,
4
d’rr
Bending: M,, =O.X.W,,.f,,llr,, =O.X.-.f,/YM, =0.07R5.d’f,,/YM,
32
The cornhination of shear and bending has to fulfil the inequality
[;u):+[;;b) 2

<I

In this inequality, the central pin plate is controlling.

The bearing resistance of plate (thickness t and yield strength f,) and pin is:
F,,,,=1.5.t.d.f,lYMp
f,, field sirength of the pin
f., tensile strength of the pin
y,+ = 1.25 according to EC 3- I - I
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):
a) Depending on the pin plate thickness I:

b) Depending on the geometry of the pin plate:

1.9 Installation of bearings

Concerning the installation of hearings, the need for a later simple replacement must
he taken into account. So it should he conimon 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 hut remain fixed to the structure while the bearings
are replaced (fig. I .9- I ).Thus, the connection hetween 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 he unscrewed. According to the
author's experience, fastening the hearings with small fillet welds that can he ground
off and remade during the replacement process is simpler.

-.--

Fis.I.Y-/: Fixing ( f a bearitil:

Generally. hearings should not be huilt directly on the construction beneath. To guar-
antee that the area below a bearing is fully sealed a layer of moriar or of a similar prod-
uct 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 clear-
ance at the upper plate and to adjust the bridge by hydraulic jacks. In this situation the
bearings should be adjusted exactly. Thus. the lower plate will get exactly the desired
inclination (horizontal or parallel to the gradient, see fig. I .9- I ) and all moveable bear-
ings 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
he 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 im-
portant. The hearings must he kept free of dirt, mortar, water and dust, especially from
all moving parts. Many bearings, such as pot bearings and spherical bearings. are pro-
tected 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 ofthe hearing and the substructure is normally 3 to 5
cm thick and must be completely filled with a mortar bedding. This can be done in dif-
ferent 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 hard-
ens very quickly so that after one day the momr bedding can be fully loaded and
the formwork removed. If the gap is jess 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 hubbies. This method will he 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 hearings should be done hy 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 brarings 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 hedding or in adjacent pans of sub- and superstmcture
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:
 Elasrnnieric hearings: Displacements and rotations, cracks in the elastomer.
Rdíer and rocker bearings: Displacements and rotations, adjustment of the
guiding device. no gap in the contact line.
Por henringr: Sufficient mesh of the lid in the pot, tight sealing of the elastomer
in the pot (if the sealing has a defect, the elasromer comes out like a pancake!)
Sliding devices - PTFEand stainless steel: Thickness of the PTFE, clean surface of
the stainless steel.
The result of an inspection should be recorded in a repon. EN 1337-10 gives an ex-
ample for such a repon.
For maintenance the bearings should be cleaned, lubricated (if necessary and pos-
sible) and coated with paint. Small defects should he repaired as far as possihle.

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 he carried out
easily. The owner of a bridge has to define in the tender if the replacement of the bear-
ings must be carried out under full traffic, restricted traffic or without traffic,depend-
ing on the importance of the bridge and the possibility of a traffic ban or a iraffic
diversion.
i n 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 he
connected to a single hydraulic circle. That means that the security devices must have
a suricienr clearance. Translations are possible by means of additional sliding con-
StNCtionS.

rein;ororcernent against splitting Lnston -'

Fig. I . 11-1: Siiffened areas for hvdraulir j w k s

To replace a bearing. the bridge has to be lifted by one or more hydraulic jacks. For hy-
draulic jacks, adequately stiffened areas to transmit the hydraulic jack forces to the
sub- and superstructure are required. Concrete parts must be reinforced against split-
ting tension. steel parts need stiffeners (fig.I . I 1-2). Thus, the construction drawings
must show in which areas or at which points hydraulic jacks can be ser. what are the
maximum lifting forces and up to which level the bridge may safely be lifted. This
data is ofpUrticular imponance 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 wholecross section uniformly with twoor
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 ihe required clearances:

I Force 1 Requiredclearance I Required clearance I

Normal hydraulic jack Fiat piston jack
kN mm mm
500 3(MI 150
1000 3M) I80
2000 450 200
So00 600 250

Table 1.11-1: Required r:leurunccjor hydraulic juiks

There are ilat jacks with a height of 80 mm and u lifting force up to 5000 kN. But their
stroke is only 20 mm and there is no security device. This kind ofjack should be ap-
plied for special cases only. New bridges should be constmcted for normal hydraulic
jacks.
In all situations, during the replacenient « f a bearing the hydraulic jack should he se-
cured by a mechanical device such us an adjusting nut for the piston or lining plates to
avoid dropping in case of pipe rupture or rupture of the piston sealing which some-
times can occur (fig. I . I 1-3 and fig. I . I 1-2).

Fig. 1.11-2: Hydraulic jack with lining plurer

Fig.I . ¡ ¡ - 3 : Hvdruulic jack wirh thread rind nui

If the replacement of a bear¡ng cakes a long time so that dispkacements of moveable

bearings will occur, the hydraulic jacks have to he equipped with a sliding device,
normally FTFE 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 suffi-
cient 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 hearings in national c d e s 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 hearings).
Teil 1 his 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 hearings

New European Standards about hearings are the following

EN 1337 "Structural M n g s " with the pans
EN 1337-1 General design rules
EN 1337-2 Sjiding elements
EN 1337-3 Elastomeric bearings
EN 1337-4 Roller bearings
EN 1337-5 Pot bearings
EN 1337-6 Rocker hearings
EN 1337-7 Spherical and cylindrical PTFE bearings
EN 1337-8 Guided bedrings and Restrained bearings
EN 1337-9 Protection
EN 1337-10 inspection and maintenance
EN 1337-1 I Transpon. storage and installation

A recommendable American Standards ahout bearings i s [he following:

AASHO-LRFD: American Association of State Highway Officials ( 1994)

1.13 References
Books and special chapters about bearings for bridges:
Eggen H., J. Grote, W. Kauschke: Lager im Bauwesrii. 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.
Eggen H.. W. Kduschke: Lager im Bauwesen. 2. Auflage, Ernst & Sohn, Berlin 1995.
Rahlwes K.. R. Maurer: Lagening und Lager von Bauwerken in: Beton-Kalender
199.5. Teil2. Ernst & Sohn, Berlin.

Papers:
I11 Alhrecht, R.: Zur Anwendung und Berechnung win Gummilagern. Der Deut-
sche Baumeister 1969. Heft 4, Seite 326. und Heft 6, Seite 563.
Andri, Beyer, Wintergerst: Versuche und Erfahningen mit neuen Kipp- und
Gleitlagern. Der Bauingenieur 5 (1962).
Andrd, W. und Leonhardt, F.: Neue Entwicklungen für Lager von Bauwerken,
Gummi- und GUmmitopfldger. Die Bautechnik 3Y (1969). Heft 2. Seite 37 bis
50.
Bayer, K.: Auflager und Fahrbahnükrgange liir Hwh- und Brückenbauten aus
Kunststoff. Verein Deutscher Ingenieure VDI im Bildungswerk BV 1956 (Vor-
tragsveroffentlichung).
Beyer, E. und Wintergerst, L.: Neue Brückenlager, neue Pfeilerform. Der Bau-
ingenieur 35 (1960),Heft 6, Seite 227 bis 230.
Eggen, H.: Brückenlager. Die Bautechnik SO ( 1973). S. 143/144.
Buh, H.: Das neue Institut für Bautechnik. Strdsse und Autobahn, Band 20
(1969), Seite 189.
Burkhardt, E.: Gepanzene Betonwilzgelenke, Pendel- und Rollenlager. Die
Bautechnik 17 (1939). Seite 230.
Cardillo. R.und Kmse. D.: Paoer(61/WA-335)ASME(1961).
Cichocki. F.: Bremsableitung be¡ Brücken. Der Bauingenieur 36 (1961). Seite
304 bis 305.
 Ckirk. E. untl Moutrop, K.: Load Deformation ChWdCteriStics of Elastomer
Bridge Hearing Pads. University of Rhode Island. May 1962.
Desm«iisahl«n. Philippe: Le caicul des piles déformahles avec appuis en
cooutchouc. Annales des Ponts et Chaussées, Paris 4/1960.
Eggcrt. H.: Bauwerksicherheit he¡ Verwendung van Rollen- und Gleitlagern.
Sirasse Briicke Tunriel I97 I , Heft 3. Seite 7 I .
Eggert, H.: D i e haurechtliche Situation he¡ Lagern fur Brücken und Hochhau-
ten. Der Stahlhaii 39 (1970). Heft 6. Seite 189.
Eiiisleld, U.: Erlaiiterungen zu den Richtlinien von unhewehrten Elastomer-
lagcrii. Miiteilungen lnstitut für Bautechnik 6/1972.
Fi-mu: Gumiiiilager für Brücken. VDI-Zeitschrift. Bd. 10111959, Nr. 12. Seite
37 I his 47X.
Gent, A,: Ruhher Bearings f«r Bridges. Ruhher Journal and International Plas-
tics 1959.
Grote. J.: Neoprenelager - einige grundsatzliche Erwagungen. Kunststoffe im
BU 7/1YhX.
Ciri)le. J.: lliibewehne Eiastomerlager. Der Bauingenieur 44 (1969), Seite 121.
Grote. J.: Verineidung von Rissen und Dehnungsschiiden durch gummielasti-
sche Lagerkingen. Kunststoffe im Bau 1 l/l968.
Hnkenjos. V.: liniersuchongen üher die Rollreihung be¡ Stahl im elastisch-plas-
iixheri Zustaiid. Technisch-wissenschaftlicheBerichte der Staatlichen Materi-
alprü1uiigsaiist:ilt an der Technischen Hochschule Stuttgan 1967. Heft 67/05.
Heeseii: Gepanzerte Betonwilzgelenke, Pendel- und Rolleniager. Die Bau-
iechnik, Jiihrgaiig 25 (1948). Seite 261.
Hütteii, P,: Beitrag zur Berechnung der Lagerverschiehungen gekrümmter,
dorchlaufender Spaiinheton-Balkenhrücken. Dissertation TH Aachen 1970.
Jiirn. K.: Curnmi im Bauwesen. Elastische Lagerung einer Pumpenstation. Der
Bauingeiiieur 36 (I961). Heft 4. Seite 137/138.
Keen: Crccp of Neoprene in Shear Under Static Conditions, Ten Years. Trans-
actions ofthe ASME. Juli 1953.
I.eonhardt und Andri: Stützungsprohleme der Hochstrassenhrücken. Beton-
uiid Stahihetonbau 55 (1960).Heft 6.
Leoiihardt. F. und Reimann, H.: Betongelenke, Versuchshericht, Vorschlage
I u r Beriiessiirig und konstruktiven Aushildung. DAfStb, Heft 175. Berlin:
Verliig Ernst &? Sohn 1966, und Leonhardt, E und Reimann. H.: Betongelenke.
I k r RauingeiiieurJI (1966). Seite49.
Lermhiirdt. E und Wintergerst, L.: Üher die Brauchharkeit von Bleigelenken.
Bctoii- iind Stahlhetonhau 1961,HefiS.Seite 123 bis 131.
1291 Maguire, C. und Assoc.: Elastomeric Bridge Bearings Pads 1959.
I301 Massonnet: Zuschrift zu B. Topaloff, Gummilager für Brücken. Der Bauinge-
iiieiir 30 (1964). Seite 428.
MGniiig. E. uiid Netzel. D.: Zur Bemessung von Betongelenken. Der Bauinge-
nieiir 44 (1969). Seitc 433 his 439.
Motion, M.: Ruhher Technology. Reinhold Publishing Co. 1959.
Muiliiis. L.: Softening of Ruhher by Deforniation. Ruhber Chemistry and
Technology (Feh. lY69).
Nordlin, E., Stoker, S . and Trinhle, R.: Laboratory and Field Performance of
Elastomeric Bridge Bearing Pads, Highway Research Board (1968).
Pare u. Keiner: Elastomeric Bridge Bearings. Highway Research Board Bull
242, 1960.
Payne u. Scott: Engineering Design with Rubber
Rejcha, C.: Design of Elastomer Bearings. Journal of Prestressed Concrete
lnstitute Oct. 1964.Vol. 9, Nr. 5 .
Resinger, F.:Langszwangungen - eine Ursache von Brückenlagerschaden. Der
Bauingenieur46 (1971), Seite 334.
Rieckmann. H.-P.: Eintluss der Lagerkonstruktion auf die Knicklange von
Pfeilern. Strasse Brücke Tunnel 1970. Seite 36 bis 42 und Seite 270 bis 272.
Sasse, H.-R. und Schorn, H.: Bewehrte Elastomerlager - Stand der Entwick-
lung. Plastik-Konstmktion 197 I. Heft 5, Seite 209 bis 227.
Schonhofer: Neugestaltungen auf deni Gebiet des Auflagerhaues und auf ver-
wandten Gehieten. Werner-Verlag, Düsseldorf 1952.
Sedyter: Über die Wirkungsweise von Bleigelenken. Beton und Eisen 1926,
Seite 29.
Shen. M. K.: Üher die L6sung des Balkens mit unverschieblichen Auflagern.
Der Bauingenieur 39 (1964). Seite 100.
Suess, K. und Grote. J.: Einige Versuche an Neoprenelagern. Der Bauingenieur
38(1963),Heft 4, Seite 152 his 157.
Thielker. E.: Elastoineric Bearing Pads and Their Application in Stmctures.
Paper 207 of Leap Conference ( 1964).
Brückenlager. Der Stahlbau 38 (1969). Seite 353.
B.: Gummilager für Brücken - Berechnung und Anwendung. Der
Bauingenieur 39 (1964). Seite 50 bis 64.
Topaloff, B.: Gummilager für Brücken. Beton- und Stahlbetonbau 54 (1959).
Heft 9.
Uetz. H. und Breckel. H.: Reibungs- und Verschleissversuche mit Teflon.
Sonderheft der Staatl. Materialprüfungsanstalt an der TH Stuttgart. 7.12.1964.
Seite 67/76.
Uetz, H. und Hakenjos, V.: Reihungsuntersuchungen mit Polytetrafluorathylen
be¡ hin- und hergehender Bewegung. Die Bautechnik 44 (1967), Heft S. Seite
IS9 his 166.
Uetz, H. und Hakenjos. V.: Cleitreibungs- und Gleitverschleissversuche an
Kunststoffen. Kunststoffe. 59. Iahrgang 1969, Heft 3, Seite 161 b
Weiprecht, M.: Autlagerung van Brücken. Elsners Taschenbuch
technischen Eisenbahndienst, 1967. Seite 231 bis 277, Abschnict E Brücken-
und íngenieurhochhdu.
Zies, K.-W.: Stabilitat von Stützen mit Rollenlagern. Beton- und Stahlbetonbau
65 f 1970), Seite 297.
AASHO-LRFD: American Association of State Highway Officials ( 1994).
Dupont de Nemours Co.: Design of Neoprene Bridge Bearing Pads, Wilming-
ton ( 1959).
CNR-UN1 liNl8-68 (Italian Standards for rubber bearings).
Ministry of Transport: Provisional Rules for the Use of Rubber Bearings in
Highway Bridges. Memo. 802, London (1962).
Mitteilungen, Iiistitut für Bautechnik. 1970, Heft 2 und 4. und 1971, Heft 4
und 6.
Ohne Verfasser. Auílager aus Teflon. Ausrüge aus dem Journal of Teflon 1964,
196.5 und 1966. Druckschrift der Du Pimt de Nemoun International S.A.
Geneva. Switzerland.
Ohne Verfasser. Brückenlager. Beratungsstelle für Stahlvenvendung, Düssel-
dorf, Merkhhtt 339. 2. Autlage 1968.
ORE Office de Rccherches et d'Essais: Verwendung von Gummi für Brücken-
lager. Frügc D6O. Utrecht (1962. 1964, 1965).
Wiedemanii. I..: ZiisXtzliche Richtlinien für Lager im Brücken- und Hochbau.
Mitteiluiigen Insiitut lür Bautechnik 3/1973, S.73. Verlag Ernst & Sohn.
Eggen: Vhrlesungen über Lager im Bauwesen. Wilhelm Ernst & Sohn
198o/ I98 I,
Kauschke. W.: Entwicklungsstand der Gleitlagertechnik für Brückenhauwerke
in der Runtlesrepiihlik Deutschlmd. Bauingenieur 64 (1989). Seite I 0 9
bis 120.
Battermaiiii/Kiihler: Elastomere Federung. Elastische Lagerungen. W. Ernst &
Sohn, Berlin, Müiichen 1982.
Gerb: Schwiiiguiigsis«lierungen. Berlin, 9. Auflage 1992, Eigenverlag (gegen
Schutzgebühr erhiiltlich).
Grote, J , und Kreuzinger. H.: Pendelstiitzm mil Elastomerlagem. Der Bau-
ingenieur 5.1 ( 197X). Seite 63/64.
Kanning. W.: Elastomer-Lager für Pendelstützen - Einfluss der Lager auf die
Beanspruchiing der Stützen. Der Bauingenieur 55 (1980). Seite 4.55.
Maurer/R;ihlwe\: Lagerung und Lager vun Bauwerken. Betonkalender 1995,
Emst & Sohn, Teil I I .
Weiheriiiüller, H uiid Knnppler. K.: Lagerreibung heim Stabilitiitsnachweis
von Brückenpfeilerii. Bauingenieur 55 (1980). Seite 285 bis 288.
Andra. W.: Der heuiige Entwicklungsstand des Topflagers und seine Weiter-
entwicklung Zuni Hublager. Bautechnik (1984),Seite 222 bis 230.
Eggen, H.: 7 Grundsatze he¡ der Lagerung von Brücken. 9. IVBH-Kongress
Amsterdani 1977. Schlusshericht. Internationale Vereinigung für Brückenhau
und Hochhau, Zürich. Schweiz.
Deinhard. J.M.. Kordiiia. K.. Mozahn, R., Storkebaum, K.-H.: üer Schadens-
fall an dcr Mainhrücke he¡ Hochheim. Beton - Stahlbetonbau. 72 (1977), Seite
I his 7.
Eggen. H. und Wiedemann, L.: Nurzungsgerechte Lagerung von Stahl- und
Verbundhrücken iiiid unterhaltungsgerechte Konstruktion von Brückenlagern.
IVBH Symposiuni Dresden 1975. Vorberichi.
Eggen, H.: Liigcr ftir Brücke,n und Hochbauten. Bauingenieur 53 (1978). Seite
161 his 168, und Ziischrift54 (1979). Seite2M).
pannbetoii: Bewiihmng im Brückenhau. Analyse van Bau-
werksdüten. Schiiden und Erhditunpskosten. Springer-Verlag Berlin. Heidel-
berg, Ncw York. L(ind»n, Paris, Tokio 1986.
 Pfohl. H.: Reaktionskraft am Festpunkt von Brückeri aus Bremslast und Bewe-
gungswiderstanden der Lager. Bauingenieur 58 ( I 983). Seite 453 bis 457.
Eggert, H. und Hakenjos, V.: Die Wirkungsweise von Kalottenlagern. Der Bau-
ingenieur49 (1974). Heft 3, Seite 93/94,
Lehmann, Dieter: Beitrage zur Berechnung der Elastomerlager. Die Bautech-
nik I(1978). Seite 19 his 22. I1 (1978). Seite Y9 bis 102. III (1978). Seite 190
his 198, IV (1979),Seite 163 bis 169.
1791 Kordina, K. und Nolting, D.: Zur Autlagerung von Stahlhetonteilen mittels
unbewehrter Elastomerlager. Der Bauingenieur56 (1981). Seite 41 bis 44.
1801 Kordina, K. und Osterath, H.-H.: Zur Auflagerung von Stahlbetonteilen mittels
unbewehrter und hewehrter Elastomerlager. Der Bauingenieur 59 (1984).
Seite 461 his 466.
Kessler. E. und Schwerm, D.: Uncbenheiten und Schiefwinkligkeiten der Auf-
lagerflachen fur Elastomerlager be¡ Stahlbetonfertigteilen. Fertigteilbau-
forum 13/83, Seite I bis 5 (Betonwerk t Fertigteil-Technik).
Kessler. E.: Die Anwendung unbewehrter Elastomerlager. Betonwerk + Fertig-
teil-Technik, Heft 6 (1987). Seite 419 bis 429.
Bundesminister für Verkehr: Schaden an Brücken und anderen Ingenieurbau-
werken. Dokumentation 1982. Verkehrsblatt-Verlag. Dortmund.
Bundesminister für Verkehr: Bericht über Schgden an Bauwerken der Bundes-
verkehrswege. Januar 1984. Eigenverlag BMV.
Beyer, E. und Eisermdnn, G . : Nachstellbare Brückenlager. Erfahrungen beim
Bauvorhaben Düsseldorf-Hauptbahnhof. Beton 5/1983.
Dickerhoff, K.J.: Beiiiessung von Brückenlagern unter Cebrauchslast. Disser-
tation Universitat Karlsruhe 1985.
Petersen, Chr.: Zur Beanspruchung moderner Brückenlager. Festschrift J.
Scheer, Miin 1987.
Hehn, K.-H.: Prüfeinrichtung zur Untersuchung von Lagern. VDI-Z 118
(1976), Seite I14 bis 118.
N.N., Sanierung der Kolnbreinsperre, Projektierung und Ausführung. I . Auf-
lage Ma¡ 1991. Herausgeber: Osterreichische Donaukraftwerke AG.
Hakenjos, V. und Richter, K.: Dauergleitreibungsverhalten der Gleitpaaning
FTFE WeisdAustenitischer Stahl für Lager im Brückenbau. Strasse. Brücke.
Tunnel I 1 (1975). Seite 294 bis 297.
1911 linbimbo M. und Kelly J.M.: Influence of Material Stiffening on Stability of
Elastomeric Bearings at Large Displacements. Journal of Engineering Me-
chanics. Sept. 1998.
Zederbaum. J. (1966):The frame action of a bridge deck supported on elastic
bearings. Civil Engineering and Public Works Review 61(714), 67-72.
Leonhardt. F. und Andra, W. ( I 960): Stützprobleme der Hochstrassenbrücken.
Beton- und Stahlbetonbau, 55(6), 12 1-32.
Tanaka, R., NatSUkdwa, 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.
Building Research Establishment (1979) Estimation of thermal and moisture
movements and stresses; Part 2, Digest 228. Watford.
1961 Emerson M. (1977): Temperature differences in bridges: basis of design re-
quirements. TRRL Laboratory Report 765. Transport and Road Research Lab-
oratory, Crowthome.
1971 Emerson M. (1968): Bridge temperatures and movements in the British Isles.
RRL Repon LR 228, pp.38. Road Research Laboratory, Crowthome.
1981 Emerson M. (1973): The calculation of the distribution of temperature in
bridges. TRRL Repon LR 561. Transport and Road Research Laboratory,
Crowthorne.
1991 Emerson M. (1976): Bridge temperatures estimated from the shade tempera-
ture. TRRL Repon LR 696. Transport and Road Research Laboratory, Crow-
thorne.
[ 1001 Stephenson. D.A. ( I 96 I): Effects of differential temperature on tall slender co-
lumns. Concrete and Constructional Engineering, 56(5), 175-8: 56(11),40-3.
[ 1011 Garrett. R.J. (1985):The distribution of temperature in bridges. The Journal of
the Hong Kong Institution of Engineers, May, 35-8.
[I021 Comité Euro-International du Béton (1984). Design manual on smctural
effects of time-dependent behaviour of concrete (Bulletin No. 142). George
Publishing Company.
[ 1031 Comité Euro-international du Béton (1985). Manual of Cracking and Defor-
mations. Bulletin 158E. Lausanne.
I1041 Neville. A.M.. Dilger, W.H. and Brooks, J.J. (1983): Creep of Plain and Struc-
tural Concrete. Construction Press, London and New York.
[ IOSl Mattock A.H. (1961): Precast-prestressed concrete bridge 5.Creep and shrink-
age studies. Journal of the Portland Cement Association Research and Devel-
opment Laboratories, May.
[ 1061 Institution of Geological Sciences: National Environmental Research Council
( 1976).Atlas of Seismic Activity 1909-1968. Seismological Bulletin No.5.
(1071 D«llar,A.T.J..Abedi, S.M.H., Lilwal1,R.C. undWilhore,R.L. (1975):Earth-
quake risk in the UK. Proceedings of the Institution of Civil Engineers, 58,
1234
[ I O81 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.
[ 1091 Lee, D.J. ( 197I ): The Theory and Practice of Bearings and Expanison Joints for
Bridges, Cement and Concrete Association.
[ I 101 Buchler, W. (1987): Design of Pot Bearings, American Concrete Institute
Pubiicdtion, SP-94, vO1.2, pp. 882-915.
[ I I 1 I Black. W. (1971): Notes on bridge bearings, RRL Report LR 382, Transport
and Road Research Laboratory, Crowthome.
( I 121 Kauschke. W. and Baignet, M. (1987) Improvements in the Long Term Dura-
bility of Bearings in Bridges, American Concrete inStiNte Publication SP-94,
V01.2,577-612.
[ 1 I31 Taylor. M.E. (I970): PTFE in highway bridges. TRRL Report LR 491, Trans-
por1 and Road Research Laboratory, Crowthome.
[I141 Eggert, H., Kauschke. W.: Lager im Bauwesen. Ernst & Sohn, Berlin 1996.
11 IS] Hakenjos, V.: Lager im Bauwesen niit Komponenten aus Kunststoff verdran-
gen hochheanspruchbare stiihlerne Rollenlager. 13th H.F. Mark-Symposium
on 19-10-94 in Vienna.
[ I 161 Marion¡, A,: Appmcchi di appoggio per ponti e strutture. ITEC. Milano 1983
Li 17) Camphell,T. I. and Kong, W. I>.:TFE Sliding Surfaces In Bridge Bearings. Re-
port ME-87-06, Ontario Ministry of Transportation and Communications.
Downsview, Ontario, 1987.
[I181 Crozier, W. F., Stoker. J. R.. Martin. V. C. and Nordlin. E. F.: A Laboratory
Evaluation of Full-Size Elastomeric Bridge Bearing Pads. Research Repnrt CA
DOT. TL-6574-1-74-26, Highway Research Report, June 1979.
L I 191 Gent, A. N.: Elastic Siahility of Rubber Compression Springs. ASME, Journal
of Mech. Engr. Science, Vol. 6. Nn. 4. 1964.
11201 Jacobsen. F. K. and Taylor R. K.: TFE Expansion Bearings for Highway
Bridges. Report No. RDR-31. Illinois DOT, June 1971.
I I2 11 McEwen. E. E. and Spencer, G. D.: Finite Element Analysis and Experimental
Results Concerning Distribution of Stress Under Pot Bearings. Proceedings of
1SI World Congress on Bearings and Sedlüiits, ACI Publication SP-70, Niagara
Falls, 1981.
[I221 Nordlin, E. F.. Boss, J. F. and Trimble, K. R.: Tetrafluorethylene (TFE) as a
Bridge Bearing Material. Research Report. M & R 64642-2. California DOT,
Sacramento, CA. June 1Y70.
[I231 Roark. R. J. and Young. W. C.: Formulas for Stress and Strain. 5th Ed..
McGraw Hill, New York. IY76.
[ 1241 Rneder. C. W.. Stanton, J . F. and laylor. A. W.: Performance of Elastomenc
Bearings. NCHRP Report 298, TRB. National Research Council. Washington.
D. C . , October 1987.
[ 1251 Roeder. C. W. and Stantnn, J. F.: Statc of'ihe Art Elastomenc Bridge Bearing
Design. ACI Journal, I 99 I .
11261 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.
[ 1271 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. I , 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.
[ 1291 Roeder, C . W., and Stanton, J. F.: Elüstonieric Bearings: A State of the Art.
ASCE. Journal ofthe Structural Division, No. 12,Vol. 109. December 1983.
1130) Saxena, A. and McEwen, E. E.: Behaviour of Masonry Bearing Plates in High-
way Bridges. Prweedings of 2nd World Congress on Bearings and Sealants.
ACI Publication SP-94. San Antonio. 1986.
11311 Stanton, J . F. and Roeder. C. W.: Elastomeric Bearings Design, Construction.
and Materials. NCHRP Report 248, TRH, National Research Council. Wash-
ington, D.C., Atigust 1982.
I ,I3 ReSerences 49

(1321 Stanton. J. F., Scroggins, G., Taylor, A. W. and Roeder, C. W.: Stability of
Laminated Elastomenc Bearings. ASCE. Journal of Engineering Mechanics,
Vol. 116. No. 6, June 19%. pp 1351-1371.
[1331 Structural Bearing Specification. FHWA Region 3 Structural Committee for
Economical Fabrication, Subcommittee for High Load Multi-Rotational Bear-
ings (HLMRB), October 1991.
2 Expansion Joints
2.1 Introduction
As mentioned in chapter I,I , movements in old stone and timber bridges were small
and no additional devices were necessary to close the gaps between bridges and abut-
ments 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 traf-
fic 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. Ail these types of expansion joints were not watertight and so
the water ran down to the bearings and to the abutments. The first watertight expan-
sion joints were built using steel rails between ruhher 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 rubher profiles and in controlling the gap
widths. Another type of watertight expansion joint is the cushion joint, consisting ofa
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 traftic loads.
The following chapters give a short survey of expansion joints for different move-
ments used in the constniction of bridges.

2.2 The role of expansion joints

The role of expansionjoints is to cany 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 inter-
ruption - 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 weUring 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 consiruction depends on the horizontal transla-
tion 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- I ).
Fig.2.3-I: 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 move-
ment'; (displacementsand rotations) from the structural analysis of the system. Move-
ments due to loads depend on the location of the loads. The controlling deformations
can he 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 con-
sidered movement.

... ...

Fig.2.3-2: Influence linefor a Iratislation

Fig.2.3-3: fnfluence linefor a rotarim

To obtain the displacement caused by a rotation it i s also possible to calculate the

rotations; the displacements can he determined from the known rotations.

2.3.1 Horizontal translation in the direetion of the bridge axis u.

A change of the environment temperature, creep under normal force and shrinkage
lead to a uniform extension or shortening of the hridge (fig. 2.3. I - I ).
The thermal expansion coefficients of steel and concrete have approximately the same
value ( aT= I,0...1,2.10~s/K ), A uniform change of temperature about the cross
section causes only a horizontal translation of the joini. This applies to composite
bridges. tm.
 I

Fig.2.3.1-I: Uniformly enension or shortening

Temperature: % = U, ' XI,AT,

,=I
'

Creep and shrinkage of concrete bridges

<p(-,t,J Coefficient of creep

N,, Permanent normal force

(compression >O)

Ex- Shrinkage coefficient

A possible problem is the change of the location of the fixing point or the unknown lo-
cation of the fixing point. On arch bridges the superstmcture 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 abut-
ment 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 unddmdged expansion joint (fig. 2.3.1-2).

Fig.2.3.1-2: Change of fhejixing point

In the case of an elastic fixing p i n t 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 tixing 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 he considered using a reduced section area and a reduced nioment
of inertia, shrinkage by a substitute tensile force Nlh acting on the free shrinking con-
crete. N,, is a compression force acting on the composite cross-section.

2:; I:- N,, = E, , A , , E,

- ___ - -.-.-._ _ . -
E
-, Shrinkage coefficient

& .-.-. -. -.
A, Area of concrete

Mi E, Reduced modulus of elasticity of

concrete to consider creep
Fig.2.3.1-3:Equivalent shrinking force

Fig.2.3.1-4:Deflection under loud

~ , , . , = @ , . í+e,)
e,
uxd.n=@i'eu+@n'e<>

Horizontal movements of expansion joints can also he 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: cp u,
Fig.2.3.1-5:Displacemeni of bearings
U"d.1 =@,-(%+e,,)
U"&,, = $1 'e" + @,,'e,

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

2.3.2 Horimnial translation in direction of the cm-section u,

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 direc-
tion of the bridge axis and on this angle (fig. 2.3.2-1 and fig. 2.3.2-2).

.J\A
u -:. u,

Fig.2.3.2-I: Skeiued bridge

u, =sin<p.u,,

U" =cos~.ucff
Fig.2.3.2-2: Skewed bearing conditions

2.33 Verücaltransiatioau,
Vertical translations u,can be caused by the replacement of bearings (fig. 2.3.3-3) and
the geometrical conditions on the abutment (fig. 2.3.3- I and fig. 2.3.3-2).

Fig.2.3.3-I: Sloping bridge with horizontal bearings

. ....

..... .. . . . .. .

~ h
Fig.2.3.3-2: Bridge with short cantilever on the abutment
L r f ~~~..~.........

Hydraulic jack
Hydraulic jack

Fig.2.3.3-3: Veriicul displacement of bearings (due to beunng repiacemenr)

2.3.4 Rotation around the bridge axis

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

Hydraulic lack
Fig.Z.3.4-I: Lijiing on one side

2.3.5 Rotation around the y-axis <pr

T h i s 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-I: Rotation due to defleciions

2.3.6 Rotation around the z-axis cp,

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

Fig.2.3.6- I : Non-uniJirm reniperuiim ucrion

2.4 Construction of expansion joints

2.4.1 General
The construction of expansion joints has to fultil 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. Additional-
ly, it is recommended to avoid slopes exceeding about 3 % and vertical steps between
joined surfaces exceeding 8 mm (fig. 2.4. I - I L

Fig.2.4.I - I : Recornniended sufi1.v rryuireinenrs

Expansion joints are exposed to pollution. The sealing should not be ddmdged by
inclusions of bigger external bodies. If the gap width is reduced due lo a movement
of the superstructure the joint must be able to expel grit and silt to the camageway
surface.
In particular, all elastomeric components must be readily accessible and easily re-
placeable.

2.4.2 Small movements (up to 25 mm)

For movements up to 15 mm it i s possible to construct a continuous asphaltic car-
riageway pavement with a supporting element covering the gap of the superstructure.
This kind of joint i s also called a buried expansionjoint (fig. 2.4.2-1). Up to 10 mm a
flat metal plate i s sufficient; for movements above 10 mm an elastomeric pad is nec-
essary to avoid pavement cracks at the edges of the supporting plate. An additional re-
inforcement o f the pavement i s advisable to provide a uniform strain distribution. The
thickness o f the pavement should he at least 80 mm and should be equal to the thick-
ness o f the corresponding parts of the superstructure and the abutment. To fulfil this
requirement the cover of the gap i s usually extended into a niche.
The asphaltic pavement does not provide sufficient watertightness. An additional seal-
ing i s recommendedto protect bearings and substructure from deterioration.

Flexible filler
Reinforcement
, \

/
I
E
E

Fig.2.4.2-I: 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 con-
struction (fig. 2.4.2-2 and fig. 2.4.2-3).

Flexible filler

Fig.2.4.2-2:Buried expansion joint senled by ii rirbher profile

 Flexible filler
// Reinforccmeni

Fig.2.4.2-3. Buried expansion joint with C » ~ ~ I I I U O Usealing

S and udditiunal 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 he managed in laboratory tests the influ-
ence of temperature and of deformation velocity is not known adequately. Incorrect
placement of matenal results in teanng of the adjacent carriageway pavement. Further
problems are yielding of asphaltic matenal 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

 ,Flexible mitenal

Fig.2.4.2-5:Asphalrirplug joint addirional sealed by a rubberpnijile

2.43 Medium movements (over 25 mm, up to 80 mm)

The absorption of medium movements requires an elastic expansion element or an ex-
pansion gap across the caniageway surface. For trafic 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 ).

Cushion element
I "
Fixing element Edge element

1 1

Fig.2.4.3-I: Consrructirin rnerhods of expunsion 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 cor-
rosion and hinder the accessibility of possible seals. hut they provide the k s t comfort
for pedestrians with high heel shoes (fig. 2.4.3-6).
-4N
Fig.2.4.3-2: V-shaped sealing Fig.2.4.3-3: Speciui seuling .for sidmdks

El Fig.2.4.3-4: Hollow section

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

Fig.2.4.3-6: Expunsion joint with cover piate

Fig.2.4.3-7: Expansion joint for .sidei<,alkr

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).

Fi~.2.4.3-X:Seals made f m m cellular neoprene e.rirusion

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-Y: Elasromeric cushionjoinr

Especially when using elastomenc cushions and neoprene extrusion seals, the
restraining actions can exceed 20 kNlm 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- I ). Linkage elements
cause equal gap widths saving the seals from overextending. They must be able to
sustain acceleration and braking forces.
 Iniemediatz slmenl Sealing element

Linkage d e m k S ~ p p welameni
i

Fig.2.4.4-I: Construction method ofexpanrim 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 elustomeric cushirin joint

A typical constmction is the coupling of V-shaped and hollow section sealing ele-
ments. it is called multiple seal expansion joint. These expansion joints can be ciassi-
fied by the kind of supporting and linkage. The folding trellis linkages (fig. 2.4.4-3)
satisfy all supporting and linkage purposes.
Fig.2.4.4-3:Rails supported bv folding trellis linkuge

An additional linkage is needed if the rails are supported by parallel beams. One pos-
sibility 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 num-
ber 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 horizonial
parallel linkages (fig. 2.4.4-5).
Fig.2.4.4-4: Ruils supported by
beams, spaciitg nirtirulled by
springs

e
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 d w s not depend on the number of rails. The higher the
number of rails the more econcimical 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 leafjoints (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-S),also called fingerjoint, is a very robust con-
struction but with several disadvantages The deformation capacity in the crosswise
direction is severely limited and vertical deformations of the joint can prejudice traf-
fic 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 shuíier 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 ,fingerjoint

 Steel fingers with an
elastomeric encasing
’!

* n L. L. A A A n n L. A nln, ’ *,-.Y h n n A * * A A 1 I * A A *
Fixing
Fig.2.4.4-9: FinRerjoint with supportedfingers

Fig.2.4.4-10: Roiling leafjoint

2.45 Expansion joints for railway bridges

For the expansion joints for railway bridges it is necessary to consider two elements:
- the rails themselves,
- thebndge.
Nowadays, a continuous track without expansion joints is preferred, due to the com-
fort of the passengers. Therefore, many modem railway bridges have no expansion de-
vices for the track. Eurocode l. part 3 (ENV 1991-3). gives rules for the maximum ex-
pansion lengths forcontinuous tracks. The expansion length (¡.e. the distance between
the “thermal centre” and the opposite end of the deck) should not exceed 60 rn 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.
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). It consists of
an ending rail with a slope of 1 : r, with r = 70 to 100. and a tapered rail which is ma-
chined 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:

2
As = Al ~

E. g.. for an expansion of Al = 200 mm and with r = 70 m the gauge is widened by

As = 6 rnrn. The expansionjoints of the rails should not he located directly over the gap
because of the rotation angle 'pr 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: Feuthered joint

For bridges without a babasi bed the gap between the bridge and the abutment nor-
mally 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
ramping 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 ofthe 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 nor-
mal 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 di-
rection 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 bearingfs)ofthe bridge. ENV 1991-3gives the lon-
gitudinal action per track FTk= i 8 [kNlm] ( L T 1 - L ) where
, 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 parís 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-resis-
tant steel and stainless steel do not need any protection against corrosion. Parts made
of mild steel must he 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 em-
bedded 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 gal-
vanising, hot-dip galvanising is the normal case. Spray galvanising is expensive hut
also possible.
2.5.2 Elastomenc parís
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& cushion elements or elastic hearings of the rails) are
made from polychloroprene or from natural caoutchouc. The material must he age-
resisting, despite the presence of de-icing salt.
Non-load transfemng elements (e.g. sealings) are made from polychloroprene or from
ethylene-propylene-caoutchoucwith high resistance to tearing and to crack propaga-
tion. The thickness should not he helow 4 mm.
The following table gives the recommended characteristics of applied elastomers.

Characteristic Non-load transferring Load transferring

...-....,
elcmmt<:
... elements
Hardness 55-65 Shore A 60-70 Shore A
Resistance to tearing niin. 10 N I mni2 min. 15 N I mm2
Tearing strain min. 350 o/o min. 400 92
Resistance to crack min. 10 N I mm min. 15 N I mm
propagation
Behaviour after a
temperature stress
(14d:70"C)
Change of hardness max. +7 Shore A max. +5 Shore A
Change of resistance to rnax. -20 % max. -15 %
tearing
Change of tearing strain max. -20 %: inax. -20 %
Resisiance against
potassium chloride
(solution: 4 %: 14 d; 23 "C)
Change o f volume niax.+lO% müx. +lo%,
Change of hardness inax. -5 Shore A niax. -5 Shore A
Resistance against hot
asphaltic bitumen
(30 minutes; 220 "C)
Change of resistance to
tearing niax. -20 ?t max. -20 5%
Change oftearing strain niax -20 #! max. -20 %
Bond with steel Failure within the
elastomeric material

Tuh.2.S.2-I : Rorommended chunic~~ri.stic.s

of /lie elristrimt,rir purrs

The springs of spring-linked multiple seal expansion joints are made of polyurethane
with a high resistance to crack propagarion. The rriatei-id i s able to withstand high
strains. I t can he compressed down to 20 % of the original length. A further advantage
i s the good damping characteristics.
Asphaltic plug joints are made of a special modified asphaltic material. This must
have a suflicient flexibility to absorb the movements of the gap, combined with a suf-
ficient load bearing capacity. The exact coniposition 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 expansionjoints and asphaltic plug joints
Expansion joints have to satisfy the requirements of ultimate limit state and fatigue
strengthdesign.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. I - I ).

Fix.2.6.1-I:Loadspread underu wheel

2.6.2 Single seal and multiple seal expansion joints

In most cases the ultiniate limit state of a single seal and o f a multiple seal expansion
joint is analysed correctly, while the fatigue was only considered empirically. How-
ever, damage is usually caused by fatigue. Therefore a correct analysis is essential
[ 18; 191. The loading acts for a very short tinie. The probability that the axles of two
vehicles are at the expansion joint at the Same tinie is relatively small and only one
axle need be considered.
As a rule, standards contain a design load 01- the following type to analyse single
members of a bridge.
 LR Contact length wheel - carriageway surface

R Static load
cp Dynamic factor

Fi#.2.6.2-/: Design wheel load

One rail of an expansionjoint carries only the portion F,, " ~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 a, Fig.2.6.2-3: Arrangement of the wheel loads

Fig.2.6.2-4: Loadper rail

LN Effective conract length

S, Gap width

n Numher of gaps within the contact length

h Rail width

F,k,,,, Portion of wheel load

a,, Factor of the influence of the angle between expansion joint and driving
direction (fig. 2.6.2-2)

If a # 90 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 a,,.
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 staie

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.

F*.kdY” =<P, ‘F,.,,,

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: F, k,<I~,i = 5 . vh. Fv,ksh,

Edge profile: ‘i~k dyii = 5 vi,

’ ’ R v k.,ini

5 Coeftkient of static friction of the siandard

9”.(mi 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:
Fatigue design
Failure due I« fatigue is the niain reason for the observed damage. Three types of
fatigue fractiires have been observed (fig. 2.6.2-5):

I ) Failure olihe welded joint between rail and suppori beam

2 ) Failure o f t h e support heam
3 ) Failure o f t h e rail

11,y.2.6.2-5: Pr,.s.sihli~
i.rucks dui~fo furipr

For the htigiie design. the stress range is of interest. At first it is determined by using
the lo;ids given in the standards.
The horizorital forces due t o rolling friction, slope of bridge and acceleration or de-
celeration must he considered. However, they are smaller than the honzontal force due
to :iccelerütion and braking. The factor 5 consists of three pürts:

i= 5 s + 5" + 5 k 55 Factor due to slope

ER, Factor due to rolling friction
EE, Factor due to locomotive acceleration/deceleration

I-"15, q&"%
nbnmmilrm

4-
-ip
@m#w
%+k
"*-

-t.l*l

--"cia
%+
.-%
%+

.71

-%,Pi
_Dn

Fig.2.6.2-6:Determination of the factors SR and

 . The horizontal
The vertical load acting on an intermediate or edge profile is Fv.h.s,n,
loads are determined as follows:

intermediate profile: Fhk . ~ , ~ i= 5. Fv L.i,r

J,,-L$..m+ J,

Fig.2.6.2-7: Dynamic kJUding of a rail

The contact time t1of the wheel depends on the contact length L,the velocity v and
the width of the profile h.
b + L,
t , =-
V

The impact load is sine-shaped ( t , = T..,half

. period). The circular frequency is:
2

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 F,.k,arn
exceed the constant amplitude fatigue limit.
The static bending moments in the vertical direction can be determined on the sup-
parted continuous beam. It depends on the stiffness of the springs if it has to be taken
into account or ifthe springs can be assumed to be rigid. In the horizontal direction the
consideration of the elastic fixing is essential (fig. 2.6.2-10).

horizontal y-f- Fh,k.$t3t

Fig.2.6.2-10: Verrical and horizontal static system

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 fhand the damping coefficient can be determined from
the recorded time-deformation curve. The spring stiffness ch.,,?" in the model is vaned
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 I0 %.
Further possibilities to determine the lowest natural frequency are an analysis by E M
or approximate methods. The following method leads to satisfactory solutions.

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

A sinusoidal loading causes the following bending deflection curve:

T"X L' ?T,X
qv(x)=qo.sin ~ w,(x)= ~ qo . sin -
L X" E.1" L
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:
o1= rn
c

With known chdniand equal span widths the frequency fhof the horizontal direction
can be determined in the same way. But the system is an elastically-supported contin-
uous girder. The following figures show some calculated results.
I
 I,
F = j; IHz]

Fig.2.6.2-12: Dyriumicf¿ictor.s

The horizontal axis of the diagram (h) contains the natural frequency of the system.
This version shows the frequency of resonance as the maximum of the graph of the de-
sign velocity. The values Acp of the resonance frequency are comparatively high. Nat-
ural 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 ihe diagram. With a known design velocity a maximum span of the rails can be de-
termined. Longer spans cause higher values Av, leading to a higher stress range. An-
other disadvantage is an increasing number of stress cycles exceeding the cui-off lim-
it, which means that more than two modes of the system must he considered.
With the values Acpband *<PO the dynamic difference moments can be calculated.

Mv.k.dyn ' M %ksial dyn = .Mh.k.sut

The stress range is determined as follows:

The design load of UD axle is higher than the actual load. The nominal stresses should
be reduced by the factor fm,, 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 traffc 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 fre-
quency 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.
 D.2 0.. O.' D.,
AS, ' '%..
I.0

Fig.2.6.2-13:Exumple of u typical loudin8 sequerrt'e

The stress ranges up to the chosen limit are used to determine a constant amplitude
stress range that causes the same damage (fig. 2.6.2-1.5).

W%
4

Fig. 2.6.2-14:Futipue sirengrh curve Fig,2.6.2-15: Canslumamplitude

,siresrange

This value when compared with the stress range Aoknlax.dyn provides the factor that al-
lows the fatigue analysis with design loads given in the standards to he used. For in-
stance, 1201 recommends the factor fe = 0.75 for the conditions of traffic in Germany.
to be applied to the loads of German Standard DIN 1072. A maximum stress deter-
mined in this way i s exceeded in only one of a thousand cases.
The fatigue design has to fulfil the following equation:

YFI Partial safety factor of the fatigue loading (yF, = 1.O)

YMI Partial safety factor of fatigue strength (yMf= I . 15)

AGE.111, = a,,,,.
A ~ J Constant
~ ~ amplitude
~ ~ stress
. ~ range ~ 100 million cycles
~ for
 Can be ascertained by the analyses of the real sequence us-
ing the Palmgren-Miner summation (a,&= 0.4).

ACk. Fatigue strength for LOO million cycles

The construction members of the expansion joint are three-dimensional and compact.
The fatigue strength Ao,.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 nor-
mally indicated by the standards.

Fig.2.6.2-16: Recommended arrurigemenr of rhe rests

The lifetime of a construction can be calculated as a statistical value. It is only appli-

cable for the evaluation of that type of construction.
Y - Ndm.w
-365.DTLV.DAAL.p
Yd Design life - time in years
Ndsmssc The number of cycles exceeding the cut-off limit
DTLV The average of daily lorry traffic in one direction
DAAL = 4 The average number of axles of each lorry
P The distribution of the DTLV on several lanes
p = i .O in ca5e 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 expansionsjoints. In the verti-
cal 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 5.

intermediate profile: Fh.k.w = 5'F v ~ k . , m

Edge profile: Fh.k~w = 5 ' R~.h.w

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 he treated as a single span
beam (fig. 2.6.3-1 ).
I

-6---7l
Fig.2.6.3-1: Calculation of the intermediate profile

The elastomeric parts of elastomenc cushion joints have to withstand stresses and
stress ranges due to traffic loads. Their strength can be ascertained hy tests. The fol-
lowing testing arrangement is recommended.

Fig.2.6.3-2: Recommended arrangement of the rest

The specimen is of the same character as the planned construction and has a length of
at least 12M1 mm. The loads are applied through an elastomeric disk of 50 mm thick-
ness which is situated in the middle of the cushion element. LRand B, are the dimen-
sions of the load area according to the applicable standard. If the width of sample is
smaller than L,. only a reduced load acts on the joint construction. It can be cansi-
dered by a smaller disk and a force than P.The inclination of P depends on the factor
E,. It considers the sliding friction or the roller friction, the slope of the hndge 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:
 P = F, . + 5'
F,,,,,, Wheel load of the standard
For ihe fatigue test the loads are reduced by the factor fmd.
P,, =,,f ' P
The construction is appiicdhíe if experiments prove that the full load P can be sup-
ported as a static load. the reduced load P,, 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 iests 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 I O A ~ S are the same as for cushion joints (fig. 2.6.4-1).
Maximum stresses are caused when the joint expansion is maximum.

D O

Fig.2.6.4-I : Rcronimrnded orrangenienr ofthe test

The behaviour of a rolling leaf joint should be checked in the same way. in most c a -
es 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- I). It is recommended to instal the expan-
sion joint in the early morning when the temperature is distributed almost uniformly
over the whole bridge.
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 ruse of a steel bridge it is
provisionally bolted or tack-welded. The auxiliary construction must be removed im-
mediately. After canying out the final fixing, the protection against corrosion is com-
pleted.
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.

Adjuslable auxiliary ~

Ul”SINCti0”

/
Rsinfovxmeni ”

Fi8.2.7-1: Possible auxiliary constructionfor 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 movemenls (shortening due to creep and shrinkage) oc-
cur. These movements begin with the erecting of the construction and stop within
some weeks I months / years. Creep is caused by compressive stresses, especially due
to prestressing. The movement due to pmtessing forces occurs during the prestress-
ing work, The joint construction has to accommodate the movements which occur af-
ter the installation. Therefore, the dimension and, by this. the costs of a joint con-
stmction can be reduced by a late installation.
The variation of creep and shrinkage is shown in the following figures by means of the
coefticient of creep rp(-,i,) and the shrinkage value E*-. In vaious standards,
t = 5 years (= I800 days) to t = 20 years is set equal to t = -.
::r,,
o.
O2
OD
0
,,

SM
, , ,

rm3
,

Fig 2.7-3: Vanatton of creep

, , , ,

ISM
, , , , ,

ZWO
TIIIW) Id]
~

Fig.2.7-4: Variation of shrinkage

00
Tima ldl

The maximum increments of shrinkage and creep occur immediately after completion
or after prestressing. For example after 100 days (about 3 months), about SO % of the
expected creep deformations and 2.5 To of the shrinkage deformation have taken place.

2.8 Inspection and maintenance

Expansion joints should be checked regularly by means of visual inspection. The fre-
quency 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 construc-
tion 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-freeexpansion 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
The lifetime of an expansion joint should be the same as the lifetime of the camage-
way 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 elastomercom-
ponents. Seals should be replaceable from the camageway site. Manufacturers offer
different systems for easy replacement (fig. 2.9-1 ).

Polychloroprene sealing elemenl

Fig.i.9-1: Possihiejirings to the seal

The gap width must be opened to at least 25 mm. In the case of an elaitic linkage,
smaller widths are possible because the rails can be displaced. On the other hand the
seals must not he stretched fully. Expansion joints for large movements should be ac-
cessible 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 expan-
sion 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. How-
ever. the joints should be situated in zones with minimal stress range and must he
welded very carefully because of the high fatigue loads.
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:
[I] Price, A.R. (1982): The service performance of fifty buried type expansion joints.
TRRL Repon SR 740. Transpori and Road Research Laboratory, Crowthome.
[2] Price. A.R. (1983):The performance of nosing type bridge deck expansion joints.
TRRL Repon LR 107 I , Transport and Road Research Laboratory Crowthome.
131 Price, A.R. (1984):The performance i n service of bridge expansion joints. TRRL
Repon LR 1104, Transpon and Road Research Laboratory. Crowthorne.
141 Department of Transpon (1989): Expansion joints for use in highway bridge
decks. Dep&menid Standard BD 33/88.
[ 5 ] Depanment of Transpon (1989): Expansion joints for use in highway bridge
decks. Departmental Advice Note BA 26/88.
[61 Koster W. (1969): Expansion Joints in Bridges and Concrete Roads. Maclaren
and Sons.
171 Busch, G.A. (1986):A review ofdesign practice and performanceoffingerjoints.
Paper presented to the 2nd World Congress on Joint Sealing and Bearing
Systems for Concrete Structures, San Antonio, Texas, September.
181 Watson, S . C . ( I 972): 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.
[Y] Koster W. (1986): The principle of elasticity for expansion joints. Paper present-
ed to 2nd World Congress on Joint Sealing and Bearing Systems for Concrete
Structures, San Antonio, Texas, September.
[lo] Lee, D.J. (1971 ): The Theory and Practice of Bearings and Expansion Joints for
Bridges, Cement and Concrete Association.
[ I I ] Demers, C.E. and Fisher, J.W., Fatigue Cracking of Steel Bridge Structures. Vol-
ume I: A Survey of Localized Cracking in Steel Bridges - 1981 to 1988. FHWA
Publication No. FHWA-RD-89- 166, McLean, VA, 1990
1121 Standard Specifications For Highway Bridges. 15th edition. AmenCdn Associa-
tion of State Highway and Transpowation Offciais, Washington. D.C., 1992
[ 131 Tschemmemegg. F.. The Design of Modular Expansion Joints. Proceedings of
the 3rd World Congress on Joint Sealing and Bearing Systems for Concrete
Structures, Toronto, 1991.
[ 141 Dexter, R.J., Kaczinski, M.R., and Fisher, J.W.: Fatigue Testing of Modular Ex-
pansion Joints for Bridges, Proceeding of the 1995 IABSE Symposium, Volume
73/2, San Francisco, CA, 1995.
[ 151 T U P - F Ü 92, Technische Liei'er- und Prüfvorschriften für wasserundurchl
ge Fahrbahnübergznge von Strassen- und Wegbrücken. Bonn: Bundesministe-
rium fürverkehr, Ausg. 1992
[ 16) Richtlinie - RVS 15.45. Brückenausrüstung - Übergangskonstruktion. Wien:
Forschungsgesellsch~tfür das Verkehrs- und Strassenwesen, Arbeitsgruppe
.Brückenbau», Arbeitsausschuss *Brlickenausrüstung>>, Ausg. Januar 1995.
[ 171 Braun. Chr.: Verkehrslastbeanspruchung von Ubergangskonstruktionen in Stras-
senbrücken. Bauingenieur 67 (1992). P. 229-237.
[ 181 Tschemmernegg, F. (a.0.): Ermüdungsnachweis van Fahrbahnüberghgen nach
ENV-1993-1. Stahlbau (1995),P.202-210.
[ 191 Pattis, A,: Dynarnische Bemessung von wasserdichten Fahrbahnübergbgen-
Modulsysteme (Dynamic Design of Waterpmf Modular Expansion Joints).
Ph.D. dissertation. Department of Civil Engineering and Architecture, Universi-
ty of Innsbruck, Austria (Dec. 1993).
1201 Herleitung eines Lastmodells für den Betriebsfestigkeitsnachweis van Strakn-
brücken. Forschung Strassenbau und StrassenverkehrstechnikHeft 430, 1984.
12 I I Ramberger, G.: Bearings, expansion joints and hydraulic equipment for bridges.
IABSE, 15. Kongress-Bericht Copenhagen, 1996.
1221 Fisher, J.W., Kaczinski, M.R. and Dexter. R.J.. Field and Laboratory Experience
with Expansion Joints. IABSE, 15. Kongress-Bericht Copenhagen, 1996.
1231 Braun. C.: The Design of Modular Joints for Movements up to 2000 mm.
IABSE, IS. Kongress-Bericht Copenhagen, 1996.
[24] Nielsen, H.B.: The Storebaelt West Bridge. Railway Expansion Joints. IABSE,
15. Kongress-Bericht Copenhagen, 1996.
1251 Crocetti, Roberto: Modular Bridge Expansion Joints - Loads, Dynamic Behav-
¡our and Fatigue Performance. Thesis for the degree of Licentiate of Engineer-
ing. Department of Structural Engineering, Division of Steel and Timber Struc-
tures. Chalrners University of Technology, 1998.
1261 Bamard. C.P.. Cuninghame. J.R.: FYactical 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 serie\ of' hod-length iii<m<>-
graphs. So far. six volumes have k e n published in the SED series. all written hy ;icknowlrdged c x -
pens in their fields. Each SED volume examines a bdsic structural engineering priihlcni. Thih
fundamental approach. together with high scientific standards. a ~ i r e the i lasting value ,it Stm
volumes for praciitionen, students and teachers of structural engineering. IABSE mitinli!ins ü hacklist
of this series. which to date includes:

SED 1:
Concrete Box-Girder Bridges
Jorg SCHLAICH, Hanmut SCHEEF
German: ISBN-3-8574X-032-7. 105 pages; 1Y82
English: ISBN-3-85748-03 1-9. 108 pages; 1982, out of print
The concrete box-girder, a widely used hridge superstructure system. is ciiinprrhrnsivcly examined
in this informative v«Iume. Emph is placed on practical guidelines fbr ihe actuül design 01c o ~ i -
Crete box-girder bridges. The monograph follows the sequence o f t h r hridpe drrigii p r w x iirell
Pan I . Design. covers design principles, and construction niethwia as they influence design. l o r
substructures, supenvuctures and complete systems: Pan 2, Siructural Analysis. examina gener;il
analysis procedures and a wide range of structural parameters. with particular attention b>,\en
' to
eccentric vehicle loads; Part 3, Dimensioning and Detailing, discusses prchtre\sing. Ilaiigr and web
dimensioning, diaphragms. ahutmenis. construction joints. bearings. piers ;ind hridpe tini\he>.
The anempt of the authors is to provide the design engineer with a conip;tct guide ti, the esscnti;d
parameters for the design and analysis of concrete box-girder hridger. Each part has i t s owii list of
references, and each proceeds fmm general principles to specific prohlcm\.

SED 2:
Dynamic Response of Reinforced Concrete Buildings
Hajime UMEMURA, HmoTAKIZAWA
ISBN 3-85748-029-7. -54 pages; 1982, out of print

The performance of reinforced concrete building structures suhjectcd ti, C;irthrlU:Lke-iiiduced mo-
tions i s the subject of this SED volume. The book emphasises foniiulatiiins of il bacically empirical
nature to analyse the complexities inherent in the inelastic and hyrteretic r c r p i i n ~id reinlirrced
concrete structures.
I Adveríisement ,

SED 3:
Vibrations in Structures
I n d u c e d by M a n a n d M a c h i n e s
Hugo BACHMANN, Walter A M M A N N
English: ISBN 3-85748-052-X. 176 pages; 1987
German: ISBN 3-85748-051-3. 193 pages; 1987, out of print
T h i s SED volume shows where dynamic problems can occur i n structure^ and presenis 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. traf-
tic. and explosions. Man-induced vibrations areconsidered. looking at dynamic loading arising from
walking, running and other forms of movement. while machine-induced vibrations are examined
according In type (rotating. oscillating. impacting). For both types of vibrations possible counter-
measures are proposed.
The second pan of the book presents 22 case studies of a wide range o f structures: fwtbridges, sports
halls. halls for pop concern. 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 SlNctUres
Ole Damgnard LARSEN
ISBN 3-85748-079-3 131 pages; 1993
T h i s SED volume i s aimed al engineers responsihle for the planning, design and assessment o f
bridges crnssing navigation channels. I t is a comprehensive source nf information on the risks o f ship
collisions with bridges and presents methods to evaluate the safety of bridges, people and the envi-
ronment. I t eniphasises collision prevention and identifies measures for the protection of structural
partaofbridgesin theeventofshipcollision.Themonographalsooffersadviceonthe upgradingand
retrofitting of existing bridges and navigation channels.
After reviewing some hasics of navigation and vessel trafiic, 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 &a addressed.

. ,
 Advertisement
/-
SED 5:
Introduction to Safety and Reliability of Structures
Jdrg SCHNEIDER
138 pages: 1997
ISBN 3-85748-106-4
Structural engineers devote all their effon t o iineeting society's enp.ctations ellicicntly. Enziiicer?
and scientists work together to develop methods for solving cnginecrinp pmhlenix. Given thiit I M -
thing i s absolutely safe, the discussion of safety can only he in terms <if1:icccpt;ihly snmill fiiiluic
probabilities. Starting from this premia, reliability theory emerged and Inah hccrimr part iif the
science and practice of engineering t d a y . Its applicatiim is no1 iinly with irhpect to tlir \;iiety of
structures, but alsi in regard to serviceiibility and other requirements of technical syrtcnis that ;ire all
subject to some probability of failure.
The present volume takes a broad approach t o the wfety of wucture, and rcI;itcd topics. The tir5t
chapter intriuiuces the reader to the main concepts and reviews *tratepies for identifying ;and iiiiiipiit-
ing hazards. The second chapter is devoted t u prwesiing diverse data into inlo#-ni;iiionthat ciin he
used in reliability analysis. Tite third chapter deals with the nnoiielling < i f structures. and the lourtli
presents recognised methods of reliability theory. Another chapter fiicusc\ <)n priihlrni\ rclatcd ici
establishing taget reliabilities. assessing existing structures. and the need for elfectivc \tiaicgic*
against human error. The appendix suppons the application uf the ineihods prtiporcd and refers read-
ers to a numher of related coniputer programs.

n i i s hook is aimed at both siudcnts and practising rngincers. It preseiiir the concepth ;ind proci.ilure*
of reliability analysis in a straightforward, underitandahle way. iniaking use <if ritiiple exatiiples.
rather than extended theoretical di.wussion. It i s hoped that this appro;icli w v c s to ;idvaicc the
application of safety and reliability analysis in engineering practice.
You may order SED m»nu&raphs. or get information on other IAHSli puhlic;itionx. dircctly I r m i :
IABSE. ETH Honggerberg, CH-8W3 Zurich. Swiwerland
Phone: +4 I- 1-633 2647 I"ax:+jI-I-h331241
E-mail: acretariat~iabse.eth~.ch Weh: hitp://www.ialn~c.ethi.cti