Steel, Concrete and Composite: Bridges

BRITISH STANDARD BS 5400 :
Part 10 :1980
Incolparting
Amendment No.I
Steel, concrete and

composite bridges -
Part 10: Code of practice for fatigue
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
NO COPYING WITHOUT BSI PERMISSION EXCEPT AS PERMITTED BY COPYRIGm L A W

nn.
-
U - -
n-=
U -
b e 1, March 1999 BS 5400 :Part 10 : 1980
Summary of pages
The following table identifies the current issue of each page. Issue 1 indicates that a page has been introduced
for the 5rst time by amendment !hbsquent issue numben indicate an updated page. Vertical sidelining on
replacement pages indicates the most recent changes (amendment, addition, deletion).
Issue
Front cover 2 6 Original

Inside h n t cover blank ?to42 Original
a 1 43 2
b blank 44 Original
C blank 45ma Original
d 1 49 2
1 2 50 blank
51 2
2 2
3 2 62 blank
4 2 63 2
4a 1 64 blank
4b blank Inside back cover Original
5 2 Back cover 2
Q BSI 03-1999 - a
. .. __
.. .. ... . - . ., . -
BS5400:Pan10:1980
Contents _..
Pago Page
Foreword 1 8.2.2 Procedure 12
Cooperating organizations Back cover 8.2.3 Adjustment factors for OH, class S
Recommendations details only 12
1. scow :! 8.3 Damage calculation, single vehicle
1.1 General :! method 12
1.2 Loading i! 8.3.1 General 12
1.3 Assessment procedures Z! 8.3.2 Procedure 12
1.4 Other sources of fatigue damage 2! 8.4 Damage calculation, vehicle spectrum
1.5 Limitations i! method 14
1.5.1 Steel decks i! 8.4.1 General 14
1.5.2 Reinforcement 2 8.4.2 Design spectrum 14
1.5.3 Shear connectors 2 8.4.3 Simplification of design spectrum 14
2. References 2 8.4.4 Calculation of damage 14
3. Definitions and symbols 2 9. Fatigue assessment of railway bridges 18
3.1 Definitions 2 9.1 Methods of assessment 18
3.2 Symbols 3 9.1.1 General 18
4. General guidance 3 9.1.2 Simplified procedure 18
4.1 Design life 3 9.2 Assessment without damage calculation 18
9.2.1 General 18
Classificationandwodmanship
11s
4.4 stresses
Methods of assessment
3
4.
4
9.2.2
9.2.3
Procedure
Non-standard design life
18
18
4.5 Factors influencing fatigue behaviour 3 9.2.4 Multiple cycles 20
5. Classification of details 4. 9.3 Damage calculation 20
5.1 Classification 4 9.3.1 General 20
5.1.1 General 4 9.3.2 Design spectrum for standard loading 20
5.1.2 Classification of details in table 17 4 9.3.3 Design spectrum for non-standard
5.2 Unclassified details 4 loading 20
5.2.1 General 4 9.3.4 Simplification of spectrum 20
5.2.2 Post-welding treatments 4 9.3.5 Calculation of damage 20
5.3 Workmanship and inspection 4 10. Fatigue assessment of bridges carrying
5.3.1 General 4 highway and railway loading 20
5.3.2 Detrimental effects 4 11. The Palmgren-Miner rule 20
5.4 Steel decks 4 11.1 General 20
6. Stress calculations 4 11.2 Design ur-N relationship 22
6.1 General 4 11.3 Treatment of low stress cycles 22
6.1.1 Stress range for welded details 4 11.4 Procedure 22
6.1.2 Stress range for welds 4 11.5 Miner’s summation greater than unity 22
6.1.3 Effective stress range for non-welded Appendices
details 4 A. Basis of U,-N relationship 23
6.1.4 Calculation of stresses 4 B. Cycle counting by the reservoir method 25
6.1.5 Effects to be included 5 C. Derivation of standard highway bridge
6.1.6 Effects to be ignored 5 fatigue; loading and methods of use 25
6.2 Stress in parent metal 5 D. Examples of fatigue assessment of
6.3 Stress in weld throats other than highway bridges by simplified methods 30
those attaching shear connectors 5 E. Derivation of standard railway load
6.4 Stresses in welds attaching shear spectra 34
connectors 6 F. Examples of stress histories and cycle
6.4.1 General 6 counting procedure 38
6.4.2 Stud connectors 6 G. Testing of shear connectors 41
6.4.3 Channel and bar connectors 6 H. Explanatory notes on detail
6.5 Axial stress in bolts 6 classification 41
7. Loadings for fatigue assessment 6
7.1 Design loadings 6 Tables
7.2 Highway loading 6 1. Annual flow of commercial vehicles
7.2.1 General 6 (nc x 109 8
7.2.2 Standard loading 6 2. Standard load spectra for RU loading 11
7.2.3 Application of loading 6 3. Standard load spectra for RL loading 12
7.2.4 Allowance for impact 8 4. Values of k, for RU loading of
7.2.5 Centrifugal forces 8 railway bridges 19
7.3 Railway loading . 8 5. Values of k. for railway bridges 19
7.3.1 General 8 6. Values of ks for railway bridges 19
7.3.2 Application of loading 8 7. Values of k,, for RL loading of
7.3.3 Standard load spectra 10 railway bridges 19
8. Fatigue assessment of highway 8. Design U,-N relationships and constant
bridges 12 amplitude nontpropagating stress
8.1 Methods of assessment 12 range values 22
8.1.1 GeneraI 12 9. Mean-line Ur-N relationships 23
8.1.2 Simplified procedures 12 10. Probability factors 23
8.2 Assessment without damage 11. Typical commercial vehicle groups 27
calculation 12 12. Proportional damage from individual
8.2.1 General 12 groups of typical commercial vehicles 28
.. .
d. mIxxKMm9 ...
BS 5400:Part 10: 1980
Page Page
13. Typical commercial vehicle gross 12. Typical point load influence line 17
weight spectrum 29 13. Simplification of a spectrum 18
14. Typical commercial vehicle axle 14. Summary of design U,-N curves (mean
weight spectrum . 29 minus two standarddeviations) 21
15. RU loading: annual traffic tonnage for 15. Summary of mean-line U,-N curves 23
standard traffic types 34 16. Typical or-N relationship 24
16. RL loading :annual traffic tonnage and 17. Multiple paths 28
composition of standard waffic mix 34 18. Typical Miner's summation adjustment
17. Classification of details curve 29
17(a). Non-welded details 49 19. Trains included in table 2 spectra 35
17@). v\FeldeddetaitSotherthanatend 20. Trains included in table 3 spectra 37
I ooMlectiwofamember
17(c). Welded detaii at end connections of
51 21. Typical example of stress concentrations
due to geometrical discontinuity 41
22. Stress concentration factors 42
member 53 23. Failure modes at weld ends 43
Figures 24. Edge distance 43
I Method of indicating minimum dass

requirementsondrawings 4a
25.
26.
Effective width for wide lap
connections
Type 3 failure modes
44
45
I 2a.
2b.
Referencestress in parent metal
Referenceskssinweldthroat
5
5
27. Type 3.6 joint 46
28. Use of continuity plating to reduce

3. Axle arrangement of standard fatigue stress concentrations in type 3.7 and
vehicle 7 3.8 joints 47
4. Plan of standard axle 7 29. Cruciform junction between flange
5. Designation of lanes for fatigue plates 47
purposes 9 30. Example of a 'third' member slotted
6. Transverse location of vehicles 10 through a main member 47
7. Impact allowance at discontinuities 10 31. Example of type 3.9 or 3.10 joint 48
8. Values of Un for different road 32. Tee junction of two flange plates 48
categories 13 33. Alternative method of joining two
9. Derivation of uv and zc for damage flange plates 48
calculation 15 34. Single fillet corner weld in bending 48
10. Damage chart for highway bridges
(valuesof dtl0) 16
11. Miner's summation adjustment factor
KF for highway bridges 17
Foreword
8s 5400 is a document combining codes of practiceto Part 8 Recommendations for materialsand
cover the design and construction of steel, concrete and workmanship, concrete, reinforcementand
composite bridges and specificationsfor the loads, prestressing tendons
materialsand workmanship. It comprisesthe following Part 9 Bridge bearings
I PartsandSections: W o n 9.1 Code of pradice for design of
Part 1 Generalstatement bridge bearings
Part 2 Specification for loads Secbion 9.2 Specification for materials,
I Part 3 Code of practice for design of steel bridges manufacture and installation of
Part 4 Code of practice for design of concrete bridges bridgebearings
Part 5 Code of practice for design of composite Part 10 Codeofpracticeforfatigue
bridges
I
Part 6 Specitication for materialsand workmanship, A British Standard does not purport to indude all the necessary
steel prwisions of a contact Users of British Standards are
Part 7 Specification for materialsand workmanship,
responsible for their cocrecf applimh.
concrete, reinforcementand prestressing
tendons
Complianke with a British Standard does not of itself
confer immunity from legal obligations. I
Q BSI 03-1999 1
.. . .. . . , . ~ .... . - .. .. . . - __ .. .- . .
. . . . . . . . . . . . . . . . . . . . . ~
. . .- ....
.......
BS 5400: Part 10 :1980 issue 2 March 1999
1. Scope 3. Definitions and symbols

1.1 General. This Part of this British Standard recommends 3.1 Definitions. For the purposesof this Part of this British
methodsforthe fatigue assessmentof partsof Wdgeswhich Standardthe followingdefinitions apply.
aresubject t o repeatedfluctuations of stress. 3.1.1 fatigue. The damage, by gradualcracking of a
1.2 Loading. Standardload spectra are given for both structural part, caused by repeated applicationsof a stress
highway and railway bridges. which is insuff icient to induce failure by a single application.
1.3 A-ssment procedures, The following alttmative 3.1.2 loading event. The approach, passageand
methods of fatigue assessment are described for both departure of either one train or, for short lengths, a bogie or
highway and railway bridges: axle, over a railway bridge or one vehicle over a highway
(a) simplified methodsthatareapplicabletopai'tsof bridge,
bridges with classified details and which are suhjectedto 3.1.31oad spectrum. A tabulation showing the relative
standard loadings; frequencies of loading events of different intensities
(b) methods using first principles that can be applied in experienced by the structure.
all circumstances. N0TE.A convenient mode of expressinga load spectrum is to

denote each load intensily aa a proportion (Kr)of a standardload
1.4 Other sources of fatigue damage. The following and the number of occurrences of each load as a proportion (Kn)
topics are not specifically covered by this Part of this of the total number of loading events.
British Standard but their effects on the fatigue life of a
structure may need to be considered : 3.1 -4standardloadspectrum. The load spectrum that
has been adopted in this Part of this British Standard,
(a) aerodynamically induced oscillations: derived from the analysis of actual traffic on typical roads or
(b) fluctuations of stress in parts of a structureimmersed rail routes.
inwater, which are due to wave action and/or eddy 3.1.5 stress history. Arecord showing how the stress at a
induced vibrations; point varies during a loading event.
(c) reduction of fatigue life in a corrosive atmosphere
3.1.6 combinedstress history. A stress historyresulting
(corrosion fatigue).
from two consecutive loading events, i.e. a single loading
1.5 Limitations event in one lane followed by a single loading event in
1.5.1 Steel decks. Highway loading is includedin this another lane.
Part and is applicable to the fatigue design of welded 3.1.7stress cycle (orcycle ofsrress). A pattern of
orthotropic steel decks. However. the stressanalysisand variation of stressat a point which is in the form of two
classification of details in such a deck is very complex and is opposing half-waves, or, if this does not exist, a single
beyond the scope of this Part of this British Standard. half-wave.
1.5.2 Reinforcement. The fatigue assessment of certain 3.1.8stressrange (orrange ofsrress) (U,). Either
details associated with reinforcing bars is included in this
Part but interimcriteria for unwelded barsaregiveriin (a) in a plate or element the greatest algebraic difference
Part 4. between the principal stresses occurring on principal
NOTE. These criteria are at present under review and revised planes not more than 45'apartin any one stress cycle :or
criteria may be issuedlater as an amendment. (b) in a weld, the algebraic or vector difference between
1.5.3 Shear connectors. The fatigue assessment of shear the greatest and least vector sum of stresses in any one
connectors betweenconcrete slabs and steel girders acting stresscycle.
compositely inflexure is covered in this Part, but the 3.1.9stressspectrum. A tabulation of the numbersof
assessment of theeffects of local wheel loadsonshear occurrencesof all the stress ranges of different magnitudes
connectors between concrete slabs and steel plates is during a loading event.
beyond the scope of this Part of this British Standard.
This effect may, however, be ignoredif the concrete slab 3.1.10 designspectrum. A tabulation of the numbers of
alone is designed for the entire local loading. occurrencesof all thestress ranges caused by all the loading
events in the load spectrum, which is to be used in fatigue
assessment of the structural part.
2. References
3.1.11 det8iIc/ass. Arating given toa detail which
The titles of the standards publicationsreferred to in this indicatesits level of fatigueresistance. I t isdenoted by the
standardare listed on the inside back cover. following :A. B. C, 0, E, F, F2, G, S, or W.
NOlE.TfN3marhwrmpermitteddassblhe highest recommended
dass,thatcanbeedilavedvdhthehighestworkmanshipspedtiedin
Pan6 (seetable 17).The fnhimum requireddass to be spedfied for
faklcatbnpurposes relates tothe lowest q-Narrve in @urn 14,
which resubh a 6fe exceeding the desgn life.
2 . 8ESIO3-1999
P
Issue 2, March 1999 BS5400:Part10:1980
3.1.1 2 Or -N relationship or ar -N curve. The Miner's summation

quantitative relationshipbetween and Nfor a detail ffl3 Stress on thecore area of a bolt, determinedon
which is derivedfrom test data on a probability basis. the basis of the minor diameter
3.1.13 design ffr-NCunw. The ut-Nrelationship ffH Limiting stress range under loading from the
adopted in this Part of this British Standardfor design on standardfatigue vehicle on a highway bridge
the basis of 2.3 %probability of failure. UN Stress on net section
ff0 Constant amplitude non-propagating stress
3.1.1 4 design life. The period in which a bridge is required range (Cr a t N= 10')
to performsafely with an acceptable probability that it will Algebraic value of stress in a stress history
not require repair. Maximumand minimumvaluesof u p from all
ffp max '
3.1.1 6standarddesignlife. 120years, adopted in this ffp min stress histories produced by standardloading
Part of this British Standard. ffr Rangeof stress (stress range) in any one cycle
3.1.1 6 Miner's summation. A cumulative damage ffrir an Individualstressranges (Or) in a design
summation based on the rule devisedby PalmgrenandMiner. ...etc spectrum
ffR mex (upmsx-up min) fora railway bridge
3.2 Symbols. Thesymbols inthis Part of this British ~ Stress ranges (in descendingorder of
Standard are as follows. ...etc
QRI, f f ,
magnitude) in a stress history of a railway bridge
under unit uniformlydistributed loading
Net area of crosssection Limiting stress range under standardrailway
Effectiveweld throat area for the particular type =T
Ioading
of connector Nominal ultimate tensile strength, to be taken as
Number of standard deviations below the mean ffu
1.1uy unless otherwisespecified

line ff,-Ncuwe Value of Or under loading from the standard
ffv
Life time damage factor (Miner's summation for fatiguevehicle (highway bridges)
120 million repetitions of a stress range ovin a (up max-up min) fora highway bridge
Uv max
highway bridge)
Values of u v (in descendingorder of
...etc
OVl. ffVl
Designstress parameter for bolts
magnitude) in anyone stress history for one
Parameterdefining the mean line a,-N
lane of a highway bridge
relationship
UvtA The largestvalue of ffvI from all stress histories
Parameterdefining t h e q -Nrelationshipfw
(highway bridges)
two standard deviations below the mean line
ffvc B The second largest value of uv1from all stress
Value of ratio uwB/ffvcA (highway bridges)
histories (highway bridges)
Miner'ssummation adjustmentfactor (highway
(7x0 ffy Coexistent orthogonaldirect stresses
bridges)
ffY Nominal yield strength
Proportionfactor for Occurrences of vehicles of
T Shear stress coexistentwith ux and UY
a specified gross weight (320 Kw kN) in any one
4. General guidance
Fatiguestress concentration factor for
re-entrant corners 4.1 Design life. The design life is that period in which a
Fatiguestressconcentration factor for bridge is requiredto performsafely with an acceptable
probabilitythat it will not requirerepair (seeappendixA).
unreinforced apertures
Ratio of actual :standard gross weights of The standard design life forthe purposes of this Part of this
vehicles, trains, bogies or axles in a load British Standard should be taken as 120 years unless
spectrum otherwise specified.
Coefficientsin the simplified assessment
procedurefor a railway bridge 4.2 Classification and workmanship. Each shdural steel
Base length of that portion of the point load detail is dassified in accOcdalIce with table 17 (see5.1 2).This
influence line which contains the greatest shows the maximum pennitleddass for different types of
ordinate (see figure 12) measured in the sbuduraldetail. The dass denoted in table 17 determines the
direction of travel designof 4-Ncurve in figure 14 that may be safely usedwith
Applied bending moments the highestworkmanship standards specified in Part 6 for the
Inverseslope of log ffr/lOg Ncurve detaii under consideration.
Number of repetitionsto failure of stress range
ffr In 5.3.1 is deiined the informationto be pmvided to the
Number of repetitions to failure of stress ranges fabricator, to ensure that the appropriate quality standards for
.
ffrs, u r I . .etc., correspondingtonl,nl.. . etc.. Part 6 are invoked.
repetitions of applied cycles
Number of applied repetitions of damagingstress 4.3 Stresses. Stressesshouldgenerally be calculated in
rangesa,,. .etc., in a design spectrum accordance with Part 1 of this British Standard but clause 6
Number of vehicles (in millions per year) of this Part supplementsthe information given in Part 1.
traversing any lane of a highway bridge 4.4 Methods o f assessment. All methods of assessment
Effectivevalue of nc described in this Part of this British Standard are based on
Total number of live load cycles (in millions) for the Palmgran-Miner rule for damage calculation (see
each load proportionKw in a railway bridge clause 11).The basic methods given respectively in 8.4
Applied axial forces and 9.3for highway and railway bridges may be usedat all
Basic static strength of the stud times. The simplified procedures given in 8.2 and 8.3 for
Elastic modulus of section highway bridges and in 9.2 for railway bridges may be used
Partialsafetyfactor forload(the product YfI . when the conditions stipulated in 8.2.1,8.3.1 and 9.2.1 are.
nn.Yfr. see Part 1) satisfied.
.
Product of Yft Y h 4.5 Factors influencing fatigue behaviour.The best
Partialsafety factor for strength fatigue behaviour of joints is achieved by ensuring that the
Reciprocalof the antilog of the standard structure is so detailed that the elements may defortn ;n*hair
deviation of log N
8 BSI 3
- 03-1999
__ .- .-
... ... ._
BS 5400: Part 10: 1980 issue 2,March 1999
intended ways without introducing secondary 5.1.2.4 Class A is generally 'happropriatefor bridge work and
deformationsand stresses due to local restraints. Stresses thespecial inspection standards relevant todasses B and C
may also be reduced, and hencefatigue life increased, by cannot nomraly be achieved in the vidnity of welds in kidge
increasedthicknessof parent metal or weld metal. work (Forthese and otherdassificationsthatshould be used
The best pint pertormanceis achieved by avoiding jctint only when special wodunanship is specifiedsee the tbolnote to
ea&kily and welds near free edgesand by othermnbols table 17.)
overtheqwfiofthejDints.Performanceisadverselyaffeded
by concentrationsof stress at hdes, openingsand m-entmt !i1.2.5Theda&idhs oftable 17 are valid forthe q u a l i i
ofsteelproducQandwektswhidrmeettherequbemenEsof
I
axners.Guidance inttreseaspectsisgiven in table 17and
a m i H.The effect of residual stresses is taken into
account in thedassificationtables.
5. Classification of details
Part6.sccept~otherunsenoted. For ceRain details the
m*nllnnpemrittedctassdependson acceptancecriteria
giveninPart6.
5.2 Unclasdfled details
I
5.1. Classification 6.2.1 Gener81. Details not fully covered in tible 17 should
5.1.1 08n8?81 betreated as class G, or class W for load carrying weld
5.1.1.1 For the purposeof fatigue assessment, each part of metal, unless a superior resistanceto fatigue is proved by
a constructional detail subject to fluctuating stressshould, special tests. Such testsshould be sufficiently extensiveto
where possible, havea particular class designatedin allow the design ur-Ncurve to be determinedin the
accordance with the criteria given in table 17. Othewise manner usedfor the standardclasses (see appendix A).
the detail may be dealt with in accordance with 6.2. 5.2.2 Post- welding treatments. Where the
5.1.1 2 T h e classification of each part of a detail depends classification of table 17 does not give adequatefatigue
upon the following : resistance, the performanceof weld details may be
(a) the direction of the fluctuating stressrelativeto the improvedby post-welding treatments such as controlled
detail: machining, grinding or peening. When this is requiredthe
detail should be classified by tests as given in 5.2.1.
NOTE. Propagationof crackstakes place in a direction
perpendicularto the ditectionof mess. 5.3 Workmanship and inspection
(b) the location of possiblecrack initiation at the detail ; 5.3.1 General. Where the classification of a detail is dependent
(c) the geometricalarrangement and proportions of the upon particularmanufacbwingor inspection requirements,
detail ; w h i i are not generally specified in Part 6 ofthis
I (d) the methods and standards of manufactureand British Standard, the necessary standards of workmanship and
inspection. inspection should be indiited on the relevant drawings.
5.1.1.3 In welded details there are several locations at which Au areas of the structure where welded details dassified as
potential fatigue cracks may initiate: these are as follows : dass For higherare neoessiily should be shown on the
(a) in the parent metal of either part joined adjacent to : drawings together with the minimum required dass and an
(1) theendoftheweld, amw indiiting the direction of stress fluctuation (seeNure 1).
For inspection purposesthis inkmukm * shouldbe
(2) a weld toe,
incorporatedonto the fabricator%shop insbuctions.
(3) a change of direction of the weld,
(b) in the throat of the weld. Note that a joint may have more than one dass requirement if it
experiences signiticantstress fluctwtions in two or more
Inthe case of members or elements connected at their ends
by fillet welds or partial penetration butt welds and flanges diredions.
with shearconnectors, the crack initiation may occur either NOTE me level of manufsQlfingq w l i i can affecth e fatigue Gfe of all
in the parent metals or in the weld throat: both possibilities s~raldetaihThemanutachrringq~determinesthedegreet0
should be checked by taking into account the appropriate whichdiscontinuities,that may adasstressrdm,may be htroduced
classificationand stress range. For other details, the duhg thekbricatiaclproass. Suchdiscontinuities can ad as fatigue
points.whk)lmayreQlesthefatiguerlfetoanuMccepQblelaveldathe
classificationsgiven in table 17 cover crack initiation at any detail underfms&mth Detab.~ a high pennitEed&as am m o ~ )
possiblelocation in the detail. Notes on the potentialmodes seriousry-bY-- bacallseofths~
of failure for each detail are given in appendix H. akeadypbcedbytaMe17on~raisenhherentinthefamofthe
detail itself.
5.1.2 Ciassificetion of deteiisin tebie 77
In order to dettermbw w h i i lavelof quality and inspectionis requiredin
5.1 2.1 Table 17 is divided into three parts which acadanw wilh Pa1t6.the minLnum required dass hastobederkred.
correspondto the three basic types into which details may ifa dass higher than F2 is rqhdthis has to be spedlied on the
be classified. These are as follows : drawings, o(hemiseuW requbedfatigue life may not be achieved. ifa
(a) type 1, non-welded details, table 17 (a) ; dass higher than E!is specified. but not required. an uneananical
fabricationwouldresuk
(b) type 2, welded details on surface, table 17 (b) ;
(c) type 3,welded details at end connections of
members, table 17 (c).
5.1.2.2 Each classified detail is illustrated and given a type
number. Table 17 also gives various associatedcriteria and
the diagramsillustrate the geometricalfeatures and
potential crack locations which determine the class of each
detail and are intended to assist with initial selection of the
appropriate type number. (For important features that
change significantly from onetype to another see the
C C -
footnote to table 17.) FstE FatE
5.1.2.3 A detail should only be designateda particular
classification if it complies inevery respectwith the figure 1. Method of indlcatlng mlnlmumclass requirements on
tabulated criteria appropriateto its type number. drawings
4 0 BSI 03-1999
Issue 1, March 1999 BS5400:Part 10: 1980
5.3.2 Detrimental effects. The following occurrences

can result in a detail exhibiting a lower performancethan its
classification would indicate :
(a) weld spatter ;
(b) accidental arc strikes;
(c) unauthorized attachments;
(d) corrosion pitting.
5.4 Steel decks. The classificationsgiven intable 17
should not be applied to welded joints in orthotropic steel
decks of highway bridges; complex stress patternsusually
occur in such situations and specialistadvice should be
sought for identifying the stress range and joint
clasjfication.
6. Stress calculations
6.1 General
6.1.1 Stress range for welded details. The stress range
in a plateorelement to be used for fatigueassessmentis the
greatestalgebraic difference between principal stresses
occurring on principal planes not more than 45'epart in
any one stress cycle.
6.1.2 Stressrange for welds. The stress range in a weld
is the algebraic or vector difference between the greatest
and least vector sum of stresses in any one stress cycle.
6.1.3 Effective stress range for non- welded details
For non-welded details, where the stress range is entirely
in the compression zone, theeffects of fatigue loading may
be ignored.
For non-welded details subject to stress reversals, the stress
range should be determinedas in 6.1 .l. The effective stress
range to be used in the fatigue assessment should be
obtained by adding 60% of the range from zero stress to
maximumcompressivestress to that part of the range from
zero stress to maximum tensile stress.
6.1.4 Calculation o f stresses
6.1 -4.1 Stresses should be calculated in accordance with
Part 1 of this British Standard using elastic theoryand
taking account of all axial, bending and shearing stresses
occurring under the design loadings given in clause 7. NO
redistribution of loads or stresses. such as is allowed for
checking static strength at ultimate limit state or for plastic
design procedures, should be made. For stresses in
0 BSI 03-1999 4a
__
._ .
Issue 2, March 1999 BS5400: Part 10: 1980
compositebeams the modulusof elasticity of the concrete table 17, the sb-ess should be basedonthe netsection.Whem
should be derived from the short term stress/strain indicatedin table 17. stress conentratiow should be taken hto
relationship (see Part4).ThestressessocalcuIated should account eitherby spedal analyskorbythe factors given in
be used with a materialfactorym = 1. @re 22 (seealso H.1.2).
6.1.4.2The bending stressesinvariousparts of a steel
‘orthotropic bridge deck maybe sbnificantlyreduced as the 6.2.2 Shearstress may be neglectedwhere it is
result of composite action with the road surfacing. numericaltyless than 15 %of e coexistentdirect stress.
However, thiseffect should only betaken intoaccount On 6.2.3 The peak and trough values of principal stress
the evidenceof special tests or specialistedvice. should be those on principal planes which are not more
6.1.5 Effects to be included. Were appropriate, the than 45’ apart. This will be achieved if either
effects of the following should be included in stress (a) 0- is at least double the corresponding shear
calculations: stress r a t both peak and trough, or
(a) shear lag, restrainedtorsionand distortion, transverse (b) thesignsofcruyandrbqthreverseor bothremain
stresses and flange curvature (see Parts3 and 5) ; thesameatthe peakand thetrough,
(b) effective width of steel plates (see Part3) ; where
(c) cracking of concrete in composite elements (see U X ,uyand rare the coexistentvalues with
Part 5) ; appropriate signs of the two orthogonal direct stresses
(d) stresses in triangulatedskeletal structures due to load and the shear stressesat the point under consideration.
applications away from joints, member eccentricities81 In either (a) or (b),provided that&> uy’at both peakand
joints and rigidity of joints (see Part 3). trough, the required stress range will be the algebraic
difference between the numericallygreater peak principal

6.1.6 Effects to beignored. The effects of the following stress and the numericallygreater trough principal stress.
neednot be included in stress calculations:
(a) residual stresses; 6.3 Stress Inweld throats other than those attaching shear
(b) eccentricitiesnecessarily arising in a standard detail ; connectors. The reference stressfor fabgue of a weld throat
(c) stress concentrations, except as required by tabls 17 ; shwld be the vector sum of the shear sbsses in the weld
(d) plate buckling. metal based on an effedive throatdimension as defined in
Part 3, and on the assumptionthat none of the load is canied in
6.2 Stress in parent metal bearing between parent rnetatslhis is illustrated in figure a.
When calwlating the stress range, the vector differenceof the
I
6.21 The reference stress for ktigue assessment should be
I
the principal stress in the parent metal adjacent to the potential greatest and the least vedor sum stress may be used instead I
I aack tocation, as shown in tigure 2a. unless otherwise noted in of the algebraic difference.
I
I f ,Welded attachment
fq- Design stress =(/A+

P M/z)
-
M’ I /
M-M’
Potential crack Stress
location di st ri but ion
I Figure ?a. Referencestress In parent metal
t = combined size of
1 Vector sum stress effective weld throats
(from Part 3)
P
= $t
1 Figure Zb. Reference stress in weld throat
Q BSI 03-1999 5
__ . .- . . - - .. .-
BS 5400: Part 10: 1980
6.4 Stresses in welds attaching shear connectors 7. Loadings for fatigue assessment
6.4.1 General. For shear connectors in accordance with 7.1 Design loadings. Highway and railway design
the dimensionalrecommendationsof Part 5, the design loadingsappropriateforbridgesin the UK are given in 7.2
stressesforfatigue in the weld metal should be calculated and 7.3 respectively.
in accordance with 6.4.2 and 6.4.3. Where the dimensions The load factors YfL and yf, should be taken as equalling 1.O
of the shear connectorsand/or the concrete haunches are (see Part 2).
not in accordancewith Parts. the fatiguestrength should be
determinedin accordancewith appendix G of this Part. 7.2 Highway loading
6.4.2 Studconnectors The stresses in the weld metal 7.2.1 General. In determining the maximumrange of
attachingstud shear connectonshould be calculated from fluctuating stress, generally, only the vertical effects of
the following expression : vehicular live load as given in clause 7 should be
considered. modified where appropriateto allow for impact
stress in weld = as given in 7.2.4. Inwelded membersthe dead load stress
longitudinal shear load on stud need not be considered. In unweldedmembers the dead
x 425 N/mm2 load stress will have to be consideredin determiningthe
appropriatenominalstaticstrength(fromPart5)
effective stress range when compression stressesoccur
6.4.3 Channelendber connectors (see 6.1.3).
6.4.3.1 The stresses in the weld metal attaching channel Centrifugaleffects need only be consideredfor
and bar shear connectorsshould be calculated from the substructures (see 7.2.5).
effectivethroat area of weld, transverse to the shear flow,
7.2.2 Standardloading
when the concrete is of normaldensity and from 0.85 X

throat area when lightweight concrete is used. For the 7.2.2.1 Standardloadspectrum. The standardload spectrum
purposes of this clausethe throat area should be based on a should be as shown intable 11which gives the weight
weld leg length which is theleast of the dimensions intensitiesand relative frequencies of commercial traffic on
tabulated below. typical trunk roads i n the UK.The minimum weight taken
for a commercial vehicle is30 kN. All vehicles less than
Channal connoetor Bar connector 30kN are ignored when consideringfatigue.
7.2.2.2 Standardfatigue vehicle. The standardfatigue
vehicle is a device used torepresent the effects of the
of bar) standard load spectrum ;for highway bridgesthis is a single
Half the thickness of half the thickness of vehicle with a weight of 320 kN. It consistsof four standard
axles with the dimensionsasshown in figures 3 and 4.
NOTE. See appendix C.for thederivationof thestandard fatigue
vehicle.
6.4.3.2 It may assist calculation to note that in normal
density concrete, where the thickness of the beam flange is 7.2.2.3 Number olvehicles.The numbers of commercial
at least twice the actual weld leg length and the weld vehicles that areassumedtotravel alongeachlaneofa bridge
dimensions comply with Part 5, the effective weld areas are : per year should be taken from table 1. If for any reason
vehicle numbers other than these are adopted, suitable
50 :-:40 bar connectors x 200 mm long, 1697 mrrtl
adjustments may be made to the fatigue analysis in
25 :i 25 bar connectors 2 200 mm long. 1018 rnm’
accordance with 8.2.3or8.3.2.1 (e).
127 and 102 channel connectors :. 150 mm long,
1272mm’ 7.2.3 Application of loading
76 channel connectors <; 150 mm long, 1081 mm:’ 7.2.3.1 Demarcation of lanes. For the purposes of lhis Part
6.5 Axial stress in bolts. The design stress for fatigue in of this British Standardthelanesshould betheactual traffic
bolts complying with the requirementsof BS 4395 arid lanes marked on the carriageway. They should be
bolts to dimensional tolerances complying with the designated in accordance with figure Sand the loading
requirements of BS 3692 should be calculated from the should be applied to the slow and the adjacent lanes only.
following expression : Where a crawler lane is provided it should be treated as an
F additional slow lane.
stress in bolt = - x us 7.2.3.2 Path of vehicles.The mean centre line of travel of
Qu
where allvehicles in any laneshould be along a path parallelto,
and within 300 mm of,the centre line of the lane asshown
F = 1.7kN/mmzforthreadsof nominaldiameteruptO25mm in figure 6. The transverse position of the centre line of the
or vehicle should be selectedso as to cause the maximum
F = 2.1 kN/mmzforthreadsofnominaldiameterover25mm stress range in the detail being considered. In some
GB is the stress range on the core area of the bolt
instances it may be found that the use of multiple paths
determinedon the basis of the minor diameter results in significantly less calculateddamage and guidance
uuis the nominal ultimate tensile strength of the bolt on this is given in C.1.4.
material in kN/mmz
7.2.3.3 Standardloading.The passage of one standard
When subjected to fluctuating stresses, black bolts fatigue vehicle along the entire length of one lane should be
complying with the requirementsot BS 41 90 may only be taken as one loading event.
used ifthey are faced under the head and turned on shank in
accordance with the requirements of BS 41 90. 7.2.3.4 Non-standardloadspectrum. If a load spectrum is
used. which differsin any way from thestandard load
spectrum. the passage of each vehicle forming the load
spectrum should be consideredto provide a separate
loading event.
0
m
BS 5400 : Part 10 : 1980
b.'-',
. . 80kN 80 kN 80.kN 80, kN (standard axles)
I I
- 1.0 m
-- -- 6.0 m
Figuref. Axle arrangement of standard fatiguevehicle
1: vehicle
twin tyre
4
900 m m
0. tyre +tyre
Figure 4. Plan of standard axle
f 586 7 .. .
- .. .
. -... . .. . . .. . . .. .. ... . .
. .. ... .. -- __ . . -
BS5400:Pan10:1980
Table 1. Annual flow of commercial vehicles (n, x 1 06)
Type
Motorway
Motorway
All purpose
All purpose
i 1[
Carriageway
layout
Dual
Dual
Dual
Dual
Number of
lanes per
carriaoewav
1.5
.O i adjacent
1.5
1.o
Slip road Single

3
2 [,.o
Slip road Single 1 applicable
All purpose Single (7.3 m * ) 2
*The number of vehicles in each lane of a single carriageway between7.3 rn and 10 rn

wide should be obtained by linear interpolation.
7.2.3.5 Method of loading. Only one vehicle should be The force assumed for any vehicle should not exceed
assumed to be on the structure at any one time and each 30000 kN
lane should be traversed separately. The effects of 1 x 0
combinations of vehicles are allowed for in clause 8.
7.3 Railway loading
7.2.4 Allowance for impact. Where a discontinuicy
7.3.1 General. The loads to be considered should be the
occurs in the road surface. e.g. at an expansion joint. the
appropriate combination of the nominal live load, impact,
static stress at every point affected by a wheel, at or within
lurching and centrifugal force. asspecified in Part 2 of this
5 m of thediscontinuity, should be increased by British Standard.
magnifying the relevant influence line, as shown in figure 7.
In welded members the dead load stress need not be
7.2.5 Centrifugalforces. The effects of any centrifugal
considered. In unwelded members the dead load stress will
force associated with the fatigue loading defined in 7.2.2
have to be considered in determining the effective stress
need only be considered for substructures ;the force should
range when compression stresses occur (see 6.1.3).
betaken as actifig at and paralle! tc the road surface. The
magnitude of the force should be calculated at the 7.3.2 Application of loading. The loads should be
appropriate design speed of the particular road, for the applied to the appropriate lengths of the point load
individual vchicles of the standard load spectrum shown in influence lines of not morethan two tracks. so as to produce
table 11 J S fo!lows : thealgebraic maximum and minimumvalues of stressat the
wv2
detail under consideration.
thecentrifugal force peraxle .: (kN)
127,
where
Wistheaxleloadof thevehicle (kN)
Y is the design speed of the road (km/h)
r is the radius of curvature at the particular lane on which
the vehicles are assumed to travel (m)
1587
BS 5400: Part 10: 1980
SI. Adj. F F Adj. SI.
Two and three lane dual carriageways

Two lane slip road
SI. ' F SI.
Three lane single carriageway
SI. SI.
Two lane single carriageway
Lone marking
""i\ - ----
Sl.
I
I
Adj.
I
I
F
I
Adjacent lane Fast lane
.. . .., Hard shoulder or Slow lane
. .._. hard strip
NOTE. For two lane dual omit fast lane.

F i g u r e 5 Designation of lanes for fatigue purposes
1588 9
_-
... ..- . ..
BS5400:Part 10: 1980
c
1
I
I
lane marking lane marking
Permitted location - -
of mean centre line
--+.--. --.-Mean centre line of travel
of vehicles (selected to cause maximum
stress range)
fI
!
wheels a
I
vehicle
t -
wheels
Mean paths
o f wheels
All dimensions are in millimetres

Figure6. Transverse location of vehicles
Discontinuity
Adjusted stress
I Influence line for static stress
- - ,
I
t I
I I
I I
I
I
Impact
allowance
I
Figure 7. Impact allr,wance at discontitiuities
7.3.3 Ster,rl.7rd/oads~ecrra.Tlic loiid spcctruni frir i t leiigth. Howevcr, rcfcrcnce to Iiiblcs 2 illid 3 is not tieccssary
pcriiiitiii:til r:iifw;iy l)rklgi: : ~ i i l i i ~ ~ : ~to
( ! stmiditrd
rl loitdit ig w h c n thciisscsstiicnt proccdurc givcti in 9.2 is used. Where
should Iic! Iiikcii Iroiii i:illii!r ~ i i l i l c2 for RU Ioiirliiig or tiililc 3 thcvoluiiicsol Iriiflit:diffcr frotii thc 27 . 1 0 . tontics per
for RL loitding. Tlii:st: liiftl(:s rol;ilc proporliotis o l thc ntiiiutii. which arc itssutiicd iii tables 2 and 3.or wherea .. ...
.. . .
stii!idiird loiiditi!j Kw to 11ii: 1(ttiiI tiittiil)i:r of itpplicd c:yc:lcs desigti lilc olhcr than 120yciirs is spccified.theappropriate '._..)
/)R i i o f 120 yciirs iiiid l o r ;I
10'.oi:curriiig iii ;I 1 1 i ; ~ i ~ jlilc valucs of r i H ni;iy bc obtained by dircct proportion.
traffic volitmc 01 27 1 0 . loiitics ljor ; i t i t i u t i i . Tlicy itlso NOTE. For ilw di!riv:itioti of loiid spcclr;t see appendix E.
iillow I o r viiriii~i(iiisiii l l t c loitdiiig cvctits with iiifluciicc titic
1589 10 , . .. - .. ..
ncu
\\I
-cNmIn
Vm=rmCIOIn000
x)moooooooo
-000000000
Ea
BS5400:Part 10: 1980
..,. ...
Table 3. S t a n d a r d l o a d spectra for R L l o a d i n g (c) determine the maximumrange of stressbv max equal ..
-
to the numericalvalue of UP max UP min. For non-welded
Group number 1 1 2 3 4 5 6 detailsthe stress range should be modified as given
in 6.1.3;
Load propoition. XW 0.55 0.45 0.35 0.25 0.15 0.05 (d) obtain the appropriate li,mitingstress range UH from
0.5 0.4 0.3 0.2 0.1 0 figure 8.
NOTE. The sign convention used for upis immaterial providedit i s
Range 0.6 0.5 0.4 0.3 0.2 0.1 applied consistently. Where stress reversaldoes not occur, the
Totelnumberof live loadcycles
value of either bpmer orb, mlnshould be taken as zero.
Length. L (m)
(nR 2 10') forvariousloading groupsand 8.2.2.2 For class S details the values of OH may be adjusted
by the factors given in 8.2.3, when appropriate.
2 9 120 189 42 0 0 8.2.2.3 Where avmax does not exceed OH the detail may be
3 1 112 68 10 170 0 consideredto have a fatigue life in excess of the specified
4 0 29 75 3 74 180 design life.
5 0 6 110 0 i! 75
7 0 38 65 0 0 77 8.2.2.4 Where aVmax exceeds U H either of the following
10 1 10 56 37 0 77 options may be adopted.
15 1 13 0 49 30 15 (a) Thedetailmay beassessed bythealternative
20 1 13 0 0 50 80 proceduregiven in 8.3 if it i s not a class S detail, or by the
30 0 8 6 0 0 2 6 5 proceduregiven in 8.4 if it is a class S detail. However, If
3 50 1 1 3 0 0 0 80 UV m x > 1.30 OH for class S details or > 1 5 5 UH for the
other classes thisoption will not satisfy the recommen-

dations of 8.3 and 8.4.
(b) The detail may be strengthenedin order to reducethe
value of crVmax or it may be redesigned to a higher class.
8.2.3 Adjustment factors for UH. class S details only.
The values of gH obtained from figure 8 may be adjusted by
multiplying successively by the following factors where
appropriate.
8. F a t i g u e assessment of h i g h w a y bridges
(a) Non-standard design life :
8.1 Methods o f assessment 0.12s
factor = .. .- ..
8.1.1 General. Three proceduresfor the fatigue assessment design life in years
of details in highway bridgesare given in 8.2,8.3 and 8.4.
The selection of theappropriate procedure depends upon (b) Non-standard annual flows:
the detail classification. the design life, the load spectrum
and the assumed annual flow of commercialvehicles.
8.1 .ZSimplified procedures. Asan alternative to the
factor =,
where
nc (fromtablel
('&(assumed) 1 0.12'
more rigorous procedureof 8.4. the simplified procedures

n c is the annual flow in the lane loaded to produce
of 8.2 and 8.3 may be used providedthe conditions stated
are satisfied. Uv max Up max - Up min
NOTE. In the case where a, max and U,, m,n are produced by
NOTE. Appendix C gives thederivation of standard highway
loedingintwolanes,n.should betakenasthesumofthe flows
bridge fatigue loading.
in those two lanes.
8.2 Assessment w i t h o u t damage calculation (c) Reduced values of abnormal load capacity
8.2.1 General. This method determines the limiting value (see C.4.4.2) :
of the maximum range of stress for a 120 year design life .
and is generally simpler but more conservativethan the
--
factor 1.3 for bridgesdesignedfor 37.5 units HB
factor 1.7 for bridgesdesignedfor 25 units HB
more exact methodsof 8.3 and 8.4. It should only be used
where all the following conditions are satisfied : 8.3 Damage calculation, single vehicle method
(a) thedetail class is in accordance with table 17 ; 8.3.1 General. This method determines the fatigue life of
(b) the design life is 120 years ; the detail in question and may be used where a more precise
(c) the fatigue loading is the standard load spectrum assessment then that provided by the method of 8.2 is
(see7.2.2.1) ; required or where the standard design life and/or the annual
flows given in table 1 are not applicable. It should only be
(d) theannual tlows of commercial vehiclesarein used where the following conditions aresatisfied :
accordance with table 1.
(a) the detail class is in accordancewith table 17 but is
NOTE. For classS detail only. 8.2.3 provides factors by which
non-standarddesign life, different traffic flow and design HB not class S.
loadingof less than 45 units may be taken into account. (b) the fatigue loading is thestandard load spectrum
8.2.2 Procedure (see 7.2.2.1).
8.2.2.1 The following procedureshould be used (see 8.3.2 Procedure
appendix D) :
8.3.2.1 The following procedureshould be used (see
(a) apply the standard fatiguevehicle to each slow and appendix D).
each adjacent lane in turn, in accordance with7.2.3 ;
(a) Apply the standard fatigue vehicle to each slow lane
(b) apply the impact allowance of 7.2.4. if appro- and each adjacent lane in turn, in accordance with 7.2.3.
priate, anddeterminethe maximum and minimumvaluesof ... ....,_. ... .
(b) Apply the impact allowance of 7.2.4, i f appropriate. .. .
principal stress or vector sum stress for weld throat,
mp max'and mp min occurring at the detail being assessed,
and determine the algebraicvalue of principal stress, or
whether resulting from the fatigue vehicle in the same lane for weld throat, the vector sum stress a t the detail being
or not;
1591 __ ._- 12 .
BS 5400 :Part 10 : 1980
I"
1 2 5 10 20 50 100 200
L
(a) Dual three lane motorway (b) Uual two lane motorway, dual three lane all
purpose. dual two lane all purpose
80
70
.=OVIU)
%2?
60 o u
t
N
E
-
-
z
50 B
bx 40 C
Q
m
C
30
v) D
-
VI
L
E
U) S
-
.c
.-
cn
20 F
F2
.-E
A
G
10
1 2 5 10 20 50 100 200
L [ml
(c) three lane all purpose. two lane all purpose (1 0 m), (d) two lane all purpose (7.3 m) ;single lane slip road
two lane s l i ~
road
figure8. Valuesof OH f o r d i f f e r e n t r o a d categories
1592 4. :
'Z
-- . .. ..- .. _. . 13 -- . . .. .
BS 5400: Part 10: 1980
assessedforeachpeakandeachtrough in thestress 8.3.2.2 Where the predicted fatigue life of the detail is less ..
history of each lane in turn (see figure 9). than the specified design life, the detail should either be
NOTE. It is sufficientlyaccuratetOCalCUlateeaCh peak or strengthened toreduce the value of UV max or redesigned to
trough value of the direct stress and to obtain the principal stress a higher class and then re-checked as in 8.3.2.1.
by combining these with the coincidentshear stress. orvice- As a guide, an approximate stress range for the same class
versa where this is more severe.
of detail can be obtained by multiplying the original value
(c) When the maximum and the minimum algebraic by :
values of stress l i p max, n p min. result from vehicle
predicted life "(m+''
positions in the same lane (referred to as case 1 in figure
9) the damage should be calculated for the stress
histories for each lane separately.
(
where
design life)
When c p max and n p nlin result from vehicle positions in m is the inverse slope of the appropriate log a,/log N
different lanes (referred 10 as case 2 in figure 9) an curve given in table 8.
additional combined stress history should be derived, If the detail is to be redesigned to a higher class the
which allows for the increased maximum stress range procedure given in 11.5(b) may be used as a guide to
produced by a proportion of the vehicles travelling in assessthe adequacy of the proposed detail.
alternating sequence in the two lanes. In this case the 8.4 Damage calculation, vehicle spectrum m e t h o d
damage should be calculated for the combined stress
8.4.1 General. This method involves an explicit calculation
history as well as for the separate lane stress histories.
of Miner's summation and may be used for any detail for
(d) Derive the stressspectrum nvl,(zv2etc., from each which the u+"elationship is known and for any known
stress history determined from (c).
load or stress spectrum.

Wherea stress history contains only one peak and/or only 8.4.2 Design spectrum
one trough, only one cycle results, as shown in figure 9
8.4.2.1 The individual stress spectra for the detail being
for lanes C and D. and the stress range can be determined
assessed should be derived by traversing each vehicle in the
directly.
load spectrum along the various lanes. Account should also
Where a stress history contains two or more peaksand/or betaken of the possibilityof higherstressrangesdue to
twoormoretroughs.asshowninfigure9forlanesAand some of the vehicles occurring simultaneously in one or
B. more than one cycle results and the individual stress more lanes and/or in alternating sequence in two lanes.
ranges should be determined by the reservoir method For non-welded details the stress range should be modified
given in appendix B. as given in 6.1.3.
(e) Determine the effective annual flow of commercial 8.4.2.2 In the absence of other evidence, allowance for
vehicles. nc million, appropriate to each stressspectrum impactshould be made in accordance with 7.2.4. The
as follows : design spectrum should then be determined by combining
(1) where case 1 of figure 9 applies, ifc = nc and may be the stress spectra with the specified numbers of vehicles in
derived directly from table 1 unless different vehicle the respective lanes.
flows are adopted ; 8.4.2.3 In assessing an existing structure, a design spectrum
(2) where case 2 of figure 9 applies the effective annual may be compiled from strain readings or traffic records
flow fiC should be obtained as indicated in figure 9 for obtained from continuous monitoring.
case 2. 8.4.3 Simplification of design spectrum. The design
( 1 ) For each stress range uvof each stress spectrum, spectrum may be divided into any convenient number of
determine the appropriate lifetime damage factor d , zo intervals. as shown in figure 13, with all the stress ranges in
from the damage chart of figure 10 and multiply each of anyone interval being treated as the maximum range in that
these values by the appropriatevalue of Fc.For non- interval but low stress ranges should be treated in
welded details the stress range should be modified as accordance with 11.3. It should be noted that the use of
in 6.1.3. small intervals will reduce the conservatism in fatigue
(9) Determine the value of the adjustment factor KF from assessment.
figure 11 according to the base length L of the point load 8.4.4 Calculation of damage. Using the design spectrum,
influence line (see figure 12) and the stress range ratio
Ke defined in figure 11. the value of Miner's summationx!should be calculated in
N
For a combined stress history from two lanes (see (c) accordance with clause 11.This value should not exceed
above and case 2 i n figure 9)KB should be taken aszero 1.O for the fatigue life of the detail t o be acceptable.
for determining KF.
NOTE. For !he derivation of KF see appendix C.
(h) Determine the predicted fatigue life of the detail from
the following - expression
. :
120
fatigue life (in years) =
I K F i c d i 20
where the summation includes all the separate lane stress
histories as well as the combined stress history, where
appropriate.
...
y; .C. ..
..........
.. *_
..._ ..
1593
BS 5400:Part 10: 1980
..
Lane reference' Lane stress history Lane stresa Number o f cycles Effective lane
apectrs t mer loedlna event low. ii,
Lane A
Lane 0
5
Lane C IC
Lane D
Case 1. Highest peak and lowest trough with vehicles in same lane
Lane reference* end stress hlsiory Lane atress Number of cycles tfective lens
spectret per loading event low. iic
Lane A %A
.. ....
10
Lane B
. I
Lane C I t ncc
Lane D nco
ZAIO
.........
Combined stress history A/B pp min
I '
I
Case 2. Highest peak and lowest trough with vehicles in different lanes
Key
0 Peak stress Stress range U,
0 Trough stress for cycle shown
X Datum stress
. . *The lane reference letters A, B etc. should be allocated in

\: ','
descending order of magnitude of stress ranges U W A . Uvie etc.
t Values of U,should be obtained by the method given i n
appendix B.
Figure 9. Derivation of ovand ii, for damage calculation
1394
- ...............
..... -. 15
... . . . -
- -- -- - . -- , . - ... m" . . . . .
BS5400:Pan10:1980
..
.... ...
. .
..
I595
_... ... . . . .. . . . - .. . . . . .... . . .-... . --
BS 5400 :Part 10 :.1980 I
. 3.0 K, = 1.0
2.8
2.6
K, = 0.5
2 .I
2.2
2.0 K, = 0.0
1.8
KF
1.6
1.5
1.2
1.3
1.2
.I
1.1
1.0
1 2 3 4 5 6 810 20 30 LO 50 60 00 100 200
L (m)
NOTE 1. f is the base length of the point load influence line (see figure 12).
K B is the ratio Q V l B -
OVlA
where
O V l A is the largest stress range produced by loading in any one lane
CVIBisthenextlargeststressrangeproduced byloading inanyotherlane ((TVCBe Q V l A ) .
For non-welded details the stress range should be modified as given in 6.1.3.
( K Bmaybetakenaszeroforthecombined historyoffigure 9.case2).
NOTE 2. This figure is applicable only to detail classes B to G. F2 and W.
Figure 11. Miner's summation adjustment factor KF for highway bridges
NOTE. f is the base length of loop containing the largest ordinate measured in direction of travel. For an element of a highway
bridge loaded by more than one lane. L should bedetermined from the influence line for the lane producing the largest value of
Qvi (= C V l A ) (seefigure9).
Figure 12. Typical point load influence line
1596
__ - . __ - - ...--
.. ... .. 17
i
BS 5400: Pan 10: 1980
,
Stress
I ranges
1
(0;
Ill
f i lm
piS,L ied design s p e c t r u m
Spectrum 'as calculated

/ -or recorded
I l l 1 1 I 1 I I I
I c
c
Number of r e p e t i t i o n s (n1
..
Figure 13. Simplification of a spectrum
(d) Obtain the appropriate limiting stress range UT

9. F a t i g u e assessment of r a i l w a y b r i d g e s from the following expressions :
= k , x k z x k 3 x k g x k S x aofor RU loading
9.1 M e t h o d s of assessment
UT = k , x k , x k . x k s x k e x uoforRLloading
9.1.1 General. Two meihodsfor the fatigue assessmentof
where
details in railway bridges are given in 9.2 and 9.3. The
choice of the appropriate method depends upon the detail k , = 1.O if the design life is 120years. otherwise it is
classification and the nature of the loading. obtained from 9.2.3
9.1.2 Sitnplifiedprocedure. As an alternative to the k, = 1.O if the loading event produces only one cycle of
stress, otherwise it is obtained from 9.2.4
more rigorous method of 9.3 the simplified procedure
of 9.2 may be used provided the conditions stated therein
k j is obtained from table4
are satisfied.
k 4 isobtainedfromtable5
k S isobtainedfromtable6
9.2 Assessment w i t h o u t damage calculation k, is obtained from table 7
9.2.1 General. This method determines the limiting value a. is the constant amplitude non-propagating stress
of the maximum range of stress for the specified design life. range for the appropriate class of detail and is
It should only be used where the following conditions are obtained from table 8
satisfied : NOTE. The sign convention used for upis immaterialproviding
(a) thedetailclassisin accordancewith table 17: it is consistently applied. Where stress reversal does not occur
under the loading described, either upm,r or upmm should be
(b) the loading is the standard railway bridge loading in be taken as zero.
accordance with 7.3.
9.2.2.2 Where OR max does not exceed UT the detail may be
The simplified procedure produces the same results as the considered to have a fatigue life in excess of the specified
method given in 9.3 when the coefficients k , . kz. k and design life.
k s (see 9.2.2) are equal to unity. In other cases the method
9.2.2.3 Where U R max is found to exceed q e i t h e r of the
may be more conservative than the method given in 9.3.
following options may be adopted :
9.2.2 Procedure (a) the detail may be assessed by the more precise
9.2.2.1 The following procedure should be used. procedure given in 9.3;
(a) Apply the standard railway loading in accordance (b) the detail may be strengthened i n order to reduce the
with 7.3.1 and 7.3.2. value of OR max or it may be redesigned to a higher class.
(b) Determine the maximum and minimum values of 9.2.3 Non-standard design life. Where the specified
principal stress or vector sum stress for weld throat design life is other than 120years. the value of k , should be
up max and n p min. occurring at the detail being assessed, taken as the lesser of either :
by loading the appropriate loops of the point load
influence line, as shown in example 4 of appendix F,
whether resulting from railway loading on thesame track
(a) (design life
lZO in years )A
or not. or
(c) Determine the maximum range of stress U R max equal
to the numerical value of up max-up min. For non-welded
.......
....
details the stress range should be modified as given .. _.._....
<.g
..I
in 6.1.3. where
m is the inverse slope of the -Ncurveappropriate to
the detail class and is obtained from table 8.
1597 '. __ .. . . . . .- . __.... . .... . . ... . . - - .-

-.
,-- Table 4. Values of k3 for R U loading of railway bridges
H a a w traffic Medium traffic - trafflo

Uaht
Datal1class D C B S D C B s D C B . s
E E E
F I F
FZ
G
W
122 W
Length. f (m)
<3.4 1.00 1.00 1.01 1.14 1.09 1.09 1.13 1.28 1.37 1.60 1.60 1.71
3.4 to 4.0 1.09 1.09 1.13 1.28 1.23 1.22 1.30 1.46 1.53 1.79 1.80 1.71
4.0 to 4.6 1.23 1.22 1.30 1.46 1.37 1.36 1.46 1.46 1.71 1.79 1.80 1.71
4.6 to 7.0 1.37 1.36 1.46 1.65 1.53 1.56 1.62 1.65 1.92 2.05 2.00 1.95
7.0 to 10.0 1.53 1.56 1.62 1.65 1.71 1.75 1.81 1.83 2.19 2.31 2.24 2.20
10.0 to 14.0 1.71 1.75 1.62 1.65 1.92 1.95 2.03 1.83 2.46 2.31 2.50 2.20
14.0 to 28.0 1.92 1.95 2.03 1.83 2.19 2.18 2.03 1.83 2.74 2.56 2.50 2.20
> 28.0 2.19 1.95 2.03 1.83 2.46 2.18 2.03 1.83 3.06 2.87 2.50 2.20
Table 5. Values of k4 for railway bridges
42 to 27 27to18 18to12 12t07 7to5 <5
0.89 1.o 1.13 1.27 1.42 1.6
Table 6. Values of k, for railway bridges
pz
Ratio 2-
PI
P
- P2
' Ratio 5
Pl
p2
0. 5 to 0.6 0.6 t o 0.75 0.75 to 0.9 0.9 t o 1.0 0.0 to 0.7 0.7 t o 1.0
1.42 1.27 1.1 25 1.o 1.o 0.89
NOTE. P, is the numerical value of stress due to track 1.

'. . P, is the numerical value of stress due to track 2 (where track 1 is the track
causing the greaterstressat thedetail under consideration).
Trains on two tracks should be considered in the same longitudinal position.
Table 7. Values of k6 for R L loading of

railway bridges
Detail class 0 C 8 s
E
F
FZ
c
W
~~~
Length, f (m) Values of k.
< 3.0 1.23 1.28 1.35 1.65

3.0 to 3.4 1.34 1.37 1.45 1.71
3.4 to 4.0 1.43 1.49 1.55 1.80
4.0 to 10.0 1.57 1.62 1.68 1.91
10.0 to 15.0 1.77 1.79 1.90 2.10
15.0 to 20.0 1.98 1.99 2.00 2.10
> 20.0 2.08 2.05 2.09 2.10
NOTE. f is the base length of the point load influence line (see
figure 12).
1590
-----
BS 5400 : Part 10 : 1980
9.2.4 Multiple cycles. Where the loading event produces respective stress spectra. These should then be combined .,. ...
r;....
more than one cycle of stress the value of k2 should be with the appropriate total occurrences in the design life of
taken as : the bridge to compile the overall design spectrum. For
m non-welded details the stress range should be modified as
(I+(%)
URa
-I-(-)
U R9
+.... given in 6.1.3.
9.3.3.2 In assessing an existing structure, a design spectrum
where may be compiled from strain readings or traffic records
m is defined in 9.2.3 obtained from continuous monitoring.
,.
U R U R ~ OR,
, etc. are the stress ranges, in descending 9.3.4 Simplification of spectrum. Where a non-
order of magnitude, at the individual cycles produced by standard loading is used i n accordance with 7.1, or the
the approach, passage and departure of a unit uniformly stress ranges are obtained from strain gauge readings, the
distributed load. design spectrum should be divided into at least 10equal
NOTE. Such cycles shouldbe counted and the individualstress intervals of stress. All the stress ranges in any one interval
ranges determinedby the reservoir method given in appendix B. should be treated as the mean range in that interval and low
An illustrationof the multiple cycle stress history i s given in stress ranges should be treated in accordance with 11.3.
example 4 of appendix F.
9.3.5 Calculation of damage. Using the design spectrum
9.3 Damage calculation n
the value ot Miner'ssurnmation& shouldbe calculated
9.3.1 General.This method involvesa calculation of
Miner'ssummation and may be used for any detail for which in accordance with clause 11and should not exceed 1.O
the ur-Nrelationship is known and for any known load or forthefatigue lifeofthedetail to beacceptable.
stress spectra. It may also be used as a more precise
alternative to the simplified procedure of 9.2.

9.3.2 Design spectrum for standardloading 10. Fatigue assessment of b r i d g e s carrying
9.3.2.1 Applying the standard railway loading as given highway a n d railway loading
in 7.3.1 and 7.3.2 thevalueof OR max should bederived i n In the case of bridgescarrying both highway and railway
accordance with the procedure set out in 9.2.2.1 (a) to (c). loadings, the total damage (i.e. 120years divided by the
The design spectrum should then bedetermined by the use predicted life) should be determined for each loading
of either table 2 for RU loading or table 3 for RL loading condition separately, in accordance with 8.3 or8.4and 9.3.
(amended where appropriate in accordance with 7.3.3). To obtain the total damage, the sum of the two damage
These tables indicate, for simply supported members, the values should be multiplied by a further adjustment factor
equivalent frequency of occurrence of stress ranges of which takes into account the probability of coexistence of
varying magnitudes resulting from the passage of the the two types of loading. This factor should be determined
individual trains forming various standard traffic types, for a given member after consideration of the fact that
where the stress ranges are expressed as proportions of the coincidence of highway traffic on multiple lanes and of
maximum stress range. railway traffic on multiple tracks has already been taken
9.3.2.2 Inthecaseof loading from more than one track, into account in assessing the separate damage values.
account should be taken of the possibility of stress Except at very busy railwaystations, where the probability
fluctuations arising from the passage of trains on not more of coincidence of rail and road traffic is higher than on the
than two tracks. both separately and in combination. As an open track, the adjustment factor is not expected to exceed
approximation. the effects of two track loading may be 1.2 where thestressesfrom highway and railway loading
obtained by dividing OR max (see 9.3.2.1) by the are of the same sign.
coefficient k , which can be obtained from table 6.
9.3.2.3 Where the approach, passageand departure of a
unit uniformly distributed load produces more than one
cycle of stress, as for instance in multi-span longitudinal or 11. The P a l m g r e n - M i n e r r u l e
cross members or in continuous deck slabs, all the cycles
should be taken into account. The appropriate standard 11.1 General. The value of Miner'ssummationz; for use
trains of figure 19 orfigure 20should be traversed across
the relevant point load influence lines and the resulting in 8.4.4 and 9.3.5 should be determined from the following
stress histories should be analysed by the reservoir method, expression :
given in appendix B. to derive the respective stress spectra.
Theseshould then be combined with the appropriate
E;=($+$+ ...... +-
Nn
annual occurrences obtained from table 15 or table 16
proportioned for the required traffic volume and multiplied where
by the specified design life to produce the overall design ..
n , , n 2 . nn are the specified numbers of repetitions of the
spectrum. As an approximation. the'effect of the additional various stress ranges in the design spectrum, which occur
cycles may be obtained by dividing either U R max in the design life of the structure.
(see 9.3.2.1 ) or OR max/ks(see 9.3.2.2) by the coefficientk2 NOTE. The number of repetitionsmay be modifiedin
which should be obtained from 9.2.4. accordancewith 11-3, and for non-weldeddetails the stress
range should be modifiedasgiven in 6.1.3.
9.3.3 Design spectrum for non-standard loading
NI, N, ...Nnare thecorresponding numbers of .
9.3.3.1 Where the loading does not comply with 7.3.1
repetitions to failure forthesamestress ranges, obtained
theappropriate train should be traversed across the relevant
from 11.2.
point load influence linesand the resulting stress histories
should be analysed by the rainflow method' to derive the
....
..
.... .....
...I.
7wd
+
. ..
*The rainflow method is described in ORE D128 Report No. 5 'Bending moment spectra and predictedlives of railway bridges',
published by the Office for Research and Experiments of the International Union of Railways. The reservoir methodof cycle counting.
described in appendix B for highway bridges. may be applied to stress histories for railway bridges (see example4 of appendix F) and
will producethe same results as the rainflow method for many repetitions of the loadingevent.
1599 __ - __ .. - 70 ._ __-
. .-- - ._. .
~
I
I
I
l i
'
".
..
"
1600
BS 5400: Part 10: 1980
BS 5400 : Part 10 : 1980
11.2 Design ur--Nrelationship. The number of repetitions 11.5 Miner's summation greater than unity. If the
to failure N of any onestress range n r Should be obtained conditions in 8.4.4 and 9.3.5 for highwav and railway
from eitherof the following equations. which have been
I bridgesrespectivelyare not met, i.e. if > 1.O, the
plotted in figure 16 :
N x U? = Kz following alternative actions should be considered. ' I
Either
Log,, N= Log,, K 2 - m Log,, br
(a) strengthen the detail toreduce the values of Qr. The
where
strengthened detail should be satisfactory ifthe reduced
K2and rn have the values given in table 8 for the different values of stresses lie betweenthe limits obtained by
detail classes. dividing the original values by the following factors :
I
NOTE. The values of K2 correspond to a probablityof failure of
I
2.3% within the design life. The basic equations and a mean-line
plot. i.e. for 50 %probability of failure, are given in appendix A.
(1;)'lrnand
11.3Treatment of l o w stress cycles. The number of
repetitions of each stress range Or less than bo should be
reducedin the proportion (or/u,)2
where
where m is obtained from table 8
isth the stress range given bytheequation in11.2for or
N= 1O7andtabulatad intable8. (b) redesignthe detail to a higher class. As a guide for
It may assist in calculationsto note that: upgrading to any class up to D, the value of uofor the new
, n nu:
-*-=-K2 (g)m
,:7 when Qr 3 uo
class should be between
times the value of uoof the original class of the detail.

. 11.4 Procedure. The following procedureshould be used
' Table 8. ur - N relationships a n d c o n s t a n t
in applying the Palmgren-Minerrule.
I (a) Determine the class of each detail in accordance with
Dstailclasa m I' a,, (Nlrnm')
table 17.
(b) Calculate the stresses and hence the stress ranges at W 13.0 0.16 x 10" I 25
each detail in accordance with clause 6 and determine
the design spectrum in accordance with 8.4 and 9.3. The
G 13.0 1 0 . 2 5 ~
10" I 29
number of low stresscycles should be modified in
F2 13.0 10:43 x 10" I 35
accordance with 11.3. F 13.0 (0.63 i : 10'' I 40
(c) Determine the number of repetitions to failure N o f E
~ 13.0 11.04 .a: 1 0 ' 2 I 47
I
each of the stress ranges in the design spectrum in D 13.0 (1.52 ~ 1 0 , ~53 I
accordance with 11-2. C 13.5 14.23 .r. 1013 I 78
(d) Evaluate Miner'ssummation in accordancew i t h l l . l . 8 14.0 11.01 :, 1015 lioo
S 18.0 12.08 x 1 0 2 2 1 82
NOTE. Valuesapplicable to non-standardcriteria may be obtained
from appendix A.
.._.
......, ..,.
..........
. , .s.
'.&J'
1401 -
. .-. . Appendix.A Table 10. Probability factors
..
Basis of a,-N relationship Probabllity of
falluro
A.7 General. Thecr-Nrelationships have been
established from statistical analyses of available
experimental data (using linear regression analysis of log or
and log N) with minor empirical adjustments to ensure
compatability of results between the various classes. 16%
The equation given in 11.2 may be written in basic form as : 2.3 % 2.0t
N x o?‘= K O x Ad 0.14% 3.0
where *Mean-line curve.
tThe standard design curve of 11.2.
Nis the predicted number of cycles to failure of a stress
range or A.2Treatment of low stress cycles. Under fluctuating
KOis the constant term relating to the mean-line of the stress of constant amplitude, there is a certain stress range
statistical analysis results below which an indefinitely large number of cycles can be
sustained. The value of this ‘non-propagating stress range’
m is the inverse slope of the mean-line log ar -log N c u w e varies both with the environment and with the size of any
d is the reciprocal of the anti-log of the standard initial defect in the stressed material. Inclean air, a steel
deviation of log N detail which complies with the requirements of Parts6.7 or
dis the number of standard deviations below the 8 isconsidered t o have a constant amplitude non-
propagating range goequal to the value of orobtained from
mean-line.
the formula in A.l when N = 10’.
-. NOTE. This corresponds to a certain probabilityof failure. as
shown in table 10. When the applied fluctuating stress has varying amplitude,
The relevant values of these terms are given in tables 9 so that some of the stress ranges are greater and some less
and 10 and the mean-line relationships are plotted in than 00, the larger stress ranges will cause enlargement of
figure 15. the initial defect. This gradual enlargement reduces the value
of the non-propagating stress range below b0.Thus, as
Table 9. Mean-line or-N relationships time goes on, an increasing number of stress ranges below
I uocan themselves contribute to the further enlargement of
Datal1 clasr KO A ! m thedefect. The finalresult isan earlier fatigue failurethan
could be predicted by assuming that a l l stress ranges below
uoare ineffective.
F2 11.23 :.. 10’2 10.592 13.0 This phenomenon has been studied on principles derived
from fracture mechanics. It’isfound that an adequate
F 11.73 :c 1012 10.605
~ ~~~ 13.0 approximation to the fatigue performance so predicted can
E 3.29 >: 10” 0.561 3.0 be obtained by assuming that a certain fraction (flr/Gc) of
D 3.99 .*. 10” 0.617 3.0 stress ranges Cr less than bocause damage i n accordance
with the formula in A . l .
C 11.08 ... 10’ I0.625 13.5
B 12.34 .: 10’ 5 10.657 14.0
S 12.13 :.: 1023 10.313 18.0
Endurance N (cycles)
NOTE. The use of these curvestor calculationpurposes i s not recommended.
Figure15 Summaryof mean-lineor-Ncurves
1602 . _- _- 23
The same result can be obtained by using a notional The simplified assessment methodof 8.2 gives limiting
.
log o,/log Ncurve, which has the inverse slopem -I- 2 stress ranges. For numbers of standard deviations below
where Nis greater than 10'. the mean-lineother than two (as used in the methods),
These points are illustratedin figure 16, which shows a these limiting stress ranges can be multiplied by the
typical log nr/logNcurve. following factors :
A.3 Fatigue l i f e f o r various failure probabilities. The Number of rtendard devlatlonr
below tha mean-lino
standard design Or -Ncurves in figure 14 are based on two
standard deviations below the mean-line with a probability 1.6 1.0 0.6 0.0
of failure of 2.3 %. In certain cases, a higher probability of W 1.07 1.15 1.24 1.32
failure could be acceptable, for example, where fatigue
G 1.07 1.15 1.23 1.32
cracking would not have serious consequences, or where a
crack could be easily located and repaired. F2 1.09 1.19 1.30 1.42
The probabilitiesof failureassociated withvarious numbers F 1.09 1.18 1.29 1.40
of standard deviations below the mean-line are given in E 1.10 1.21 1.34 1.47
table 10. The of -Ncurves appropriate to other numbersof D 1.08 1.17 1.27 1.38
standard deviations below the mean-line can be derived
from the formula given in A.1. C 1.07 1.14 1.22 1.31 ,
Where the methodsof clauses 8,9 or 11, which are based B 1.05 1.11 1.17 1.23
on two standard deviations below the mean-line, predict S 1.07 1.16 1.24 1.34 ..
the fatiguelife (or damage), thelife (or damage) appropriate
to other numbers of standarddeviations below the mean-

line can be obtained by multiplying the calculated life (or . ......._ .
... <._......
. .
dividing the calculateddamage) by the following factors : . .
Detail Number of rtanderddeviatloni
cl.. belowthemean-line
-1.5
-
0.0
W 1.24 2.34
G 1.23 2.28
F2 1.30 2.85
F 1.29 1.65 2.73
E 1.34 1.78 3.1 8
D 1.27 1.62 2.06 2.63
C 1.27 2.02 2.56
B 1.23 1.52 1.88 2.32
S 1.79 3.20 10.21
I--
Static limitations
Constant am pl i t u.de Ioa d ing

in clean a i r
al
CO-- -- - - - - - - - _- - ------------
U
v) Effective curve obtained under
variable amplitude loading,
-0,
0
equivalent to changing slope o f
--b'
1 i
I
\
\
-
N curve above N = 10'
\
0, I \
\
0,
C I
2 I \
ul
v) I
E I \
z 1 I \\
NOTE. Only that portion of this figure shown asa full line is based on experimental evidence. ..........
.. .
. ..
.....
Figure 16. TypicaIu,-Nrelationship
'._.'
1603 74
rn
BS 5400: Part 10: 1980
./-.. Appendix B Appendix C

C y c l e counting by the r e s e r v o i r m e t h o d D e r i v a t i o n of standard highway b r i d g e fatigue;'
loading and m e t h o d s of use
8.1 General. The purposeof cycle counting is toreduce an
irregular series of stress fluctuationsto a simple list of stress C.l Standard loading
ranges.The method given in this appendix, and shown in the C.l.l Stendardloedspectrum. (See 7.2.2.1 .) Table 11
figure below, is suitable when dealing with short stress showsa 25 band spectrum of commercial vehicle weights,
histories, such as those produced by individual loading axle arrangementsand frequencies of occurrence, which is
events. It consists of imagining a plot of the graph of each typical of the full range of commercialtraffic on a trunk road
individual stress history as a cross section of a reservoir, in the UK. Other relatively uncommon vehicle types have
which is successivelydrained from each low point, counting been included in the types which are nearest to them on the
one cycle for each draining operation. The result, after many basis of equivalent damage. Privatecars and light vans
repetitions of the loading event, will be the same as that below 15 kN unladen weight are not included as their
obtainable by the rainflow method (see the footnote contribution to fatigue damage is negligible.Table 11
to 9.3.3). includes vehicles operating under both the Motor Vehicles
8.2 M e t h o d (Construction and Use) Regulationsand the Motor
8.2.1 Derive the peak and trough values of the stress Vehicle (Authorization of Special Types) General Order.
history, due to one loading event, in accordance The proportions of the various types of vehicles of the
with 8.3.2.1 (c). Sketchthe history dueto twosuccessive spectrum have been taken from sample traffic counts. To
occurrences of this loading event. The calculated values of allow for variations in the loads being carried by similar
peak and trough stresses may bejoined with straight lines vehicles, the various types have been divided into heavy,
if desired. Mark the highest peak of stress in each mediumand I.ightloading groups (H, M and L). The axle
i
.... occurrence. If thereare two or more equal highest peaks in loads have been averaged from weighbridge records of
.. .'.,' one history, mark only thofirst such peak in each moving traffic taken between 1971 and 1974.
occurrence.
C.1.2. Standard fatigue vehicle. (See 7.2.2.2.) The
8.2.2 Join the two marked points and consider only that proportions of the damage caused by individual vehicle
part of the plot which falls below this line, like the section types, compared with the total damage by all vehicles,
of a full reservoir. varies between the limits shown in table 12. The standard
8.2.3 Drain the reservoirfrom the lowest point leaving the fatigue vehicle has been devised torepresent the most
water that cannot escape. If there are two or more equal damaging group which, for the majority of detail classes and
lowest points the drainage may be from any one of them. influence line lengths, is group 4A-H.
List one cycle having a stress range owl equal to the vertical The axle spacings of the standard fatigue vehicle are the
height of water drained. sameasthosefortheshort HBvehicle(see Part2) andthe
8.2.4 Repeat 8.2.3 successively with each remaining body 80 kN axle weight is equivalent to the standard 18 000 Ib
of water until the whole reservoir is emptied, listing one axle, which has been used for some years as the datum axle
cycle at each draining operation. in the fatigue design of road pavements. As the damage
8.2.5 Compilethe final list which contains all the done by the single tyred wheels on vehicles listed in table 11
individual stress rangesin descending order of magnitude is normally less than 4 %of the damage done by the double
ovl,ov2 etc. Where two or more cycles of equal stress range tyred wheels, the standard fatigue vehicle with double tyred
are recorded, list them separately. wheels gives an adequate representation of the wheel
8.2.6 For non-welded details only, a horizontal line damage for all types of vehicles.
representing zero stress should be plotted and those parts of
thestressrangesinthecompressionzonemodified as in 6.1.3
First occurrence Second occurrence

I- - -~ .. -m
Highest peak, ,Imaginary r e s e r v o i r
I
i
5;
V'
Shaded areas are thoser Time
p a r t s of the reservoir
which successively (NTS))
become empty
1604
25
{
BS5400:Part10:1980
C.1.3 Standardlane flows. (See 7.2.2.3.) The annual C.3.2 Damagefactorsd,20.(See figure 10and8.3.) The
flowsofcommercialvehiclesnc/106givenin table 1 are damage chart ot figure 10is based on the cumulative fatigue
based on the design capacity of the particular road type, as damage caused by the design spectrum which is obtained
specified by the Department of Transport. The proportion of from the passage of the vehicles represented by the gross
commercial traffic above 15 kN unladen weight has been weight spectrum of table 13 over an influence line base
taken to be 20 %of all traffic for all-purpose roads and 25 % length of 25 m. It assumes 1O Bcycles of stress per year for
for motorways. the 120 year design life and the damage, assessed by
C.1.4 Multiplepaths (See 7.2.3.2.) A significant Miner's summation, is given in relation to the stress range
reduction in assessed damage can be achieved by the caused by the passage of a standard fatigue vehicle.
consideration of multiple paths when the transverse The adjustment factor KF should be applied to the results
influence line profile departs rapidly from the value of the from figure 10 in order to allow for influence line base
mean path ordinate. Figure 17 shows a histogram of lengths of less than 25 m and for the effectsof multiple
occurrence of proportion factors for 100 mm wide intervals vehicles.The effects of numbers of vehicles other than the
of carriageway width, derived from observations of traftic 120 million assumed are allowed for by multiplying the
patterns. lifetime damage from figure 10bync.
The standard fatigue vehicle centre line should be traversed C.4 Adjustment factors
along the centre line of each 100 mm strip and a stress C.4.1 General. Becausethe assessmentcharts of figures 8
spectrum obtained for each strip in accordance with 8.3.2, and 10 are based on the passage of single vehicles and also
with the annual flow of vehicles in any strip being derived on an assumed influence line base length of 25 m,
from the appropriate proportion (from figure 17) x nc x 106 adjustment factors are necessary to allow for shorter base
(taken from table 1). In cases where the transverse influence
lengths and for the effects of combinations of vehicles.

line changes sign across the 1300 mm histogram width, These factors, which are labelled X, Y and 2 in figure 18, ..
. ..
an alternating succession of vehicles along paths of make up the Miner's summation adjustment factor KF ...:. . ..
'
opposing sign would produce a reduced number of cycles of given in figure 1 1. Factor X is described in C.4.2 and
enhanced stress range. However, trial calculations, which factors Y and 2 in (24.3. It should be noted that the values
take intoaccountthe probabilityofthe occurrenceof of these factors cannot be determined precisely but that the
alternating sequences. have shown that the increase in values of KF given are sufficiently accurate for design
damage is not significant and may be neglected. purposes.
C.2 Derivation of load spectra based o n t h e standard The adjustment factor KF has been included in the
fatigue vehicle. The principle behind the assessment derivation of figure 8 but should be applied explicitly when
procedures of 8.2 and 8.3 is that each commercial vehicle figure 10 is used. For the assessment method of 8.2 there is
in table 1 1 isrepresented by one vehicleof thesamegross a n additional adjustment. which may be made in the case
weight but with axle configurations identical to those of the of class S details, where the design is based on reduced HB
standard fatigue vehicle. The resulting load spectrum shown loading and this is described in C.4.4.2.
in table 13, where the various vehicle group weights are
C.4.2 Influence line base length less than 25 m. The X
expressedas a proportion of the standard fatiguevehicle
component of the& adjustment factor has been obtained
gross weight, has been used in the derivation of the
by comparing the stress histories for a selection of the most
procedures of 8.2 and 8.3 instead of the complete vehicle
damaging vehicles in table 11 with those derived by
spectrum of table 11. Table 13 has been derived using a
representing the vehicles by standard fatigue vehicles with
datum influence line which is 25 m long and rectangular in
the same gross weights. The two groups of vehicles were
shape. This has theeffect of limiting thegross weight of the
traversed across 11 different shapes of influence line, each
18 GTgroup so that static design stresses are not exceeded. with a range of loop lengths, and the Miner's summation for
For values of L less than 25 m the design spectrum becomes damage calculated using the a,-Nrelationships for all the
increasingly influenced by bogie and axle spacings and the detail classes except S.The X factor is the average value
weights. The weighbridge records, from which table 11 is of the ratio of damage due to the table 11 vehicles to that
derived (see C.1.1). show a widervariation in axle loads due to the equivalent standard fatigue vehicles. The scatter
than in gross vehicle loads, and an increase of 10% in results between different influence line shapes was found
(indicated by trial calculations) over the individual values to be acceptable for design purposes.
of table 1 1 has been allowed in deriving the spectrum of
C.4.3 Multiple vehicle effects. During normal conditions
axle weights of table 14. When L is greater than 25 m, the
of traffic flow, instances will occur when more than one
design spectrum for individual vehicles will be proportional
vehicle will contribute to the stress in a detail at any
to the gross vehicle weight spectrum. However, as L
particular time and the stress may be increased above that
increases. account should be taken of thecontribution to
due to either vehicle alone. These multiple vehicle effects
damage done by two or more vehicles acting simultaneously.
may be sub-divided according to the following :
All theseeffects have been catered for by means of a simple
adjustment factor KF (see figure 11) whichisdescribed in (a) more than one vehicle in the same lane
more detail in C.4. simultaneously :
C.3 Assessment charts (b) vehicles in different lanes simultaneously causing
stress of the same sign :
C.3.1 LimitingstressrangesnH. (See figure8 and 8.2.)
The graphs of allowable stress ranges in figure 8, which are (c) vehicles in different lanes i n alternating sequence
based on a 120year design life and the appropriate traffic causing stresses of opposite signs and so increasing the
flows from table 1, have been derived with the aid of the stress range.
damage chart of figure 10 (see C.3.2). The worst The chance that damaging vehicles will be sufficiently close
to each other, either in the same or different lanes, has been .
assumptions have been made about the stress
contributions from each lane and about theallocation of assessed on a probability basis. Correction factors have been
traffic flows between lanes so that the results are always o n derived from the comparison between the damage from the
the safe side. The adjustment factor KF. from figure 11, has combinations of vehicles and the damage due to such .
.... ...
, ..
.. .-.
vehicles on their own.
,:.::..
also been included in the derivation of figure 8. XY
1505 . ..
BS5400:Pat-t 10: 1980
o a
N d
o c
mr-
3 0 0 0 0 0 0 0 0 0.00
3 0 0 0 0 0 0 0 0 0 0 0
3 0 0 0 0 0 0 0 0 0 0 0
3 0 0 lnlnlnooo
n m m r c e 0 c ) c )
sg
hl
C
o
w
a 0 b
a 0 -0
?
N
o m o
(Dln-3
0 0
2 E
-
0
a .FJ m
a m ID- r-
ml-
.. . $:
E‘
rt
. . ~.:
1 .-
U
.-Km
t-
F
w m
7 >r
0
Y
I 1606
BS5400:Part 10: 1980
The effect given in (a) above i s allowed for by means of C.4.4 Class S details
factor Y (see figure 18) which shows that the effect is (24.4.1 Genere/. Trial calculations indicatethat for class S
negligible for L less than 50 m but increaseswith details, significant variations occur in the values of X
increasingL due to the greater probability of having two (see C.4.2) both with variations in L and with variations
vehicles simultaneously on the same influence line base in influence lineshapes. Hencethe adjustmentfactor Xdoes
length. not apply to class S details and the assessment procedure
The effect given i n (b) above is allowed for by meansof of 8.3 cannot be applied.
factor 2 (see figure 1 8 ) which is given for several valuesof C.4.4.2 ReducedHB designloeds. (See8.2.3 (c).) Table
factor KB (see figure 1 1 ) . i.e. the ratio between the major 12shows that the heaviest abnormal loadvehiclesof table
stress ranges produced by vehicles travelling separately in 11 contribute a very small percentageof the total damage for
the two lanes producing the most severe stress effect. all detail classes except class S. Henceno relaxation is
Combinations of vehicles in more than two lanesdo not provided in the assessment proceduresof 8.2 and 8.3for
generally increase the total damage significantly. these other classeswhen the bridge is subject to reduced
The effect given in (c) above is taken account of by the values of H B loading.
additional combined stress history, referredto as case 2 in However, these heavy vehicles do contribute the greater
figure 9. Inthiscase the effect of vehicles travelling proportion of the total damage for class S details and
simultaneouslyin two lanes may be neglected and so factor hence8.2.3 (c) provides reduction factors where the bridge
2 can be taken as zero by making K B equal to zero. is designedto carry less than 45 unitsof HB loading.
Table 12. P r o p o r t i o n a l damag vehicles
Detail class fm -
D t o G. F2 and W
3.01
C
(m - 3.5)
B
( m = 4.0)
S
(m - 8.0)
Group. IL - 1 . 5 (L-25 L-1.5 I L-25 L-1.5 lL-25 L-1.5 IL-25
1 8 GT-H 1% 2% 1% 5% 2% 11% 27% 80 %

9 lT-H 1% 2% 1% 4% 1% 7% 9% 15.2 %
1 8 GT-M and
9 l T - M to 5A-H 2% 4% 3% 6% 4% 9% 32% 4.72 %
5A-M and L 4% 14% 4% 14% 3% 13% 1% 0.02 %
I
4A-H, M and L 57% I 67% 57% 63% 57% 54% 22% 0.057 %
4 R-H to 2 R-L 35% 111% 34% 8% 33% 6% 9% 0.003 %
*See tablell.
f of mean p o t h of vehicles
(see f i g u r e 6 1
-
-
I
1
7
I
-
% frequency
distribution o f
vehicles for each
100 mm interval & 7 11 16 18 16
I I
I
I I i i 1 i 1
Figure 17. Multiple paths

600 mm 600 mm
' 4
1607
._.. . . . .. . .
BS5400:Part10:1980
..
..
1 25 50
Figure 18. Typical Miner's summation'edjustmentcurve

. ..
. .
Table 13. Typical commercial vehicle gross Table 14. Typical commercial vehicle axle
weight spectrum weight spectrum
Vehicle
designation
I
Proportion of
standard fatigue
vehicle aross waiaht I Proportion
of total
vehicles
Total sale
weight
264
Total numberof ariesfor
106 vehicles
240
18GT-H 6.75
18GT-M 12.38 I 0.000 01
0.000 03
0.000 02
231
176
165
120
160
560
0.000 04 154 . 100
7GT-H 0.000 03 143 780
7GT-M 0.000 07 121 80
110 90 040
7A-H 12.47 I 0.00002 99 240 280
5A-H 11.97 I 0.00028 93 320 000
0.01 4 50 88 59 320
0.01 5 77 59 350
4A- H 0.090 71 180 000
4A- M 0.090 66 59 930
0.090 61 165 000
55 290 040
49 150 000
44 120000
4R-L 10.38 I 0.015 39 320 000
3A-H 10.67 I 0.030 33 380 000
22 60 000
17 360 000
3R-H 0.01 5 Total 2 856 000 axles for 1O6
3R-M 0.01 5 vehicles
0.01 5
2R-L
li%p .
0.1 70
10.170
0.1 80
NOTE 1. This table is based on L .< 1.5 m.
NOTE 2. These values include the 10%increase referred to in C.2.
Total 1.0
NOTE. This table is based on 1 =: 25 m (rectangular loop).
1608
... . . . . 11 . 29 ~. __ . . . .... . _.._
-.. _
.
BS 5400 : Part 10 : 1980
Appendix D a vertical stiffener, given that analysis, in accordance .E.-..,

...
-.
with 8.2.2.1 (a) and (b), produces principal maximum and
Exarnplesof fatigueassessrnent of h i g h w a y minimum stress values of 17.8 N/mm2 tensile and
bridges b y simplified methods 8.0 N/mmz compressive respectively, under the standard
0.1 General. 0.2 to 0.4 give examples of typical fatigue vehicle, with thevehicle positioned inthesame lane.
calculations for fatigue assessment. D.2.3 Classification. The potential fatigue crack should
For all detail classes, 8.2 and figure 8 provide a limiting be classified as F (from type 2.9 of table 17 (b)), provided
stress range OH which will always besafe where standard that the weld end is not within 10 mm of the flange toe.
loading conditions are applicable but which may be too D.2.4 Assessment using8.2
conservative i n some cases. The U H valuesgiven in figure 8 From 0.2.2, UP max = 17.8 N/mm'
thus provide a simple check which isvery suitable for initial
U P min = -8.0 N/mmz
design purposes.
Alternatively, 8.3 provides a more precise method for detail
classes 8 to G and F2 and W leading to a life prediction (to
Henceov msx = 17.8 (-8.0)
(see8.2.2.1 (c)).
- - 25.8 N/mmz
97.7% probability of survival) and indicating the extent of For a dual two lane motorway with L = 75 m and a class F
required changes in the detail when the predicted life is too detail,
short. OH = 20.5 N/mm2 (seefigure8(b)). ,_
The basic assessment procedures for a steelwork detail are Hence uVmax exceeds OH and adequate fatigue life is not
illustrated in D.2, while the application of a combined stress demonstrated (see 8.2.2.3). By reference to 8.2.2.4
history is illustrated in D.3 (see 8.3.2.1 (c)). A typical either
procedure for shear connectors is given in 0.4.

(a) the procedure of 8.3 may be used, since
0.2 Example using t h e basic assessment procedures uv max < 1 . 5 5 0 ~
f o r a steelwork detail or
D.2.1 Type o f bridge. A three span (50 m/75 m/45 m) (b) the detail should be strengthened t o reduce uv max
twin girder highway bridge that carries a dual two lane or improved to a higher class.
motorway and is designed to carry standard UK loading.
0.2.5 Assessment using8.3. Given that the stress
0.2.2 Details o f theproblem. These assess the fatigue histories for each loading event (passage of the standard
resistanceof the main girder bottom flanges at the mid-span fatiguevehicle in each lane) are asfollows:
of the main span, which is adjacent to the transverse weld of
17.8
32.0
V
- 8.0
v -5.5
w
-3.5
-2.5
1SI carriageway. slow lane 1st carriageway, adjacent lane
10.2
6.3
T
-2.8 -1:9 -1 :5 -0I8
2nd carriageway, adjacent lane 2nd carriageway, slow lane
From figure 9. in the first carriageway, the slow lane will be

designated lane A, since it has the greatest stress range (i.e.
17.8 - (-8.0) 25.8).Thereservoirmethod (seeappendix
8 ) may be used to determine the values of ovas shown
below :
17.8
0 0
....
...,
. .
. -.,.
...
-......_.... .
- 8.0
The values of bvfor the other lanes may be determined in a
similar way.
1609
30 -....
.._ .... -- -.
_. - -
. -..
BS5400:Pan10:1980
.......
... This assessment procedure can be tabulated as fo
Rderencm Cych Rang* dad ZCt
uv CIau F
-8.0 25.8 0.32

1st Slow ( O V l A) 1.5
5.5 Neglect
12 10 1-5-5 ( c 0.001 )
IB
1 15.6 0.032 0.03
1st Adj. (Ovvla) 1 .o
2.5 Neglect
I 2
(<0.001)
1 10.2 -2.8 13.0 0.015 0.02
2nd Adj. C 1.o
2 0 -1.9 1.9 Neglect
(<0.001)
1 6.3 -1.5 7.8 0.001 0.00
2nd Slow 0 1.5
2 0 -0.8 0.8 Neglect
(<0.001)
I .../". *.:. .
,
*.
*See appendix B.
tSee figure 10.
$See table 1and figure 9.
KB 15.5125.8 0.6 -
Adjustment factor KF (see figure 11) 0.2.6 Comments. The method of 8.3 predicts a fatigue life
that is in excess of the design life and hence a reduction in
cross section could be tolerated. BY reference to 8.3.2.2,
L-75m
-
hence KF 1.59
Total damage = XKF ZCd, 1o = 1.59 x 0.53 = 0.843
the stress range could be increasedby 142 "'"i.e. by 1.04.
( 1 20)
Estimated life = 120/0.843 = 142 years (see 8.3.2.1 (h)) In contrast, the procedure of 8.2 produces a limiting value
which is greater than the specified design life of 120 years of OH equal to 20.5/25.8 = 0.79 times the applied
and so the detail may be regardedas satisfactory. maximum stress range, thus demonstrating the
conservatism of this method. Nevertheless, the simplicity of
determination of OH is such that the method should, in many
cases, prove a useful 'first stage'asan alternative to the
more precise method of 8.3. or the lengthier procedure
of 8.4.
. .... . - . .__
. 31 .. ....
BS5400:Part10:1980
D.3 Example of the application of a combined stress provide adequatefatigue resistance at the lap-welded ..
.....
history bracingfguuet connection.
0.3.1 Type of bridge. A highway bridgethat carries a two D.3.3 Classification. The potential fatigue crackshould
lane, single carriageway. all-purpose road and is designed be classfied as G (from type 2.1 1 of table 17(b)) since the
to carry standard UK loading, but is deemed to be subject to weld will beat theedgeofthemember (seealsofigurel).
1.2 L 1OEcommercialvehicles per year, in each lane, for a D.3.4Ass~~smsntusingS.2. The information given
design lifeof 60years. in 0.3.1 and D.3.2 does not comply with (b) and (d)
0.3.2 Details of theproblem. Given that the analysis of 8.2.1 and therefore82 is not applicable.
predicts, as shown in D.3.5. the forces in a transverse D.3.5 Assessment using 8.3. Assume a cross-sectional
bracing member due to the standard fatigue vehicle in area of 2200 mm 2 and determinethe stress historiesfor
alternate lanes and given that the influence line base length passage of the standardfatigue vehicle in each lane.
is 8 m. determine an area of cross section for the member, to
Given that theseareasfollows:
12.9 12.9
Lane A
- 12.9 -12.9
Lane B
Derivethe combined stress history (see 8.3.2.1 (c) and

case 2 of figure 9).
12.9 12.9
1 - 12.9 -1 2.9
I
I Lane AIB
_.
..... .,... ..
' _...
161 1
- - 32
BS 5400:Pan 10: 1980
The stress ranges ovmay be determined by the reservoir

method (see appendix B) and the resulting stressspectra
(see figure 9) forthe individual and combined lane stress
histories will be as shown below:
Lane A 54
2 cycles per
loading event
0
Lane B "Pt .
0
2 cycles per
loading event
Combined A/B s4
l o a d i n g event
The assessment procedure can be tabulated as follows :
Lane I Cycle
number
I
k;;; , Trough
1 Range
0,
I dim+
Class G
I
nc:
I Damage
d*ia
1 12.9 0 12.9 0.055 0.03

A
2 12.9 4.2 8.7 0.01 }1.:%= 0.6 0.01
B
1
2
0
4 . 2
-12.9
-12.9
12.9
8.7
0.055
0.01 J
.:.2zl
= 0.6
.2
0.03
0.01
112.9 I 4.2 I 8.7 I 0.01 I J = 0.6 10.01
*See appendix B.
tSee figure 10.
S e e figure 9.
D.3.6 Comments. The estimated life i s in excess of the 60

Adjustment factor K F (see figure 1 1 ) year design life and hence a reduction in area is allowable.
for separate histories
K , = 12.9/12.9 = 1.0
The reduction factor will be( $)0.25
i.e. 0.917 (see8.3.2.2)
L=8m and hence an acceptable area will be 2017 mm 2 . Repeat the

henceKF = 1.81 assessment with an amended initial assumption of the area.
for a combined history NOTE. Iteration provides a solution of 2027 mm'.
KB = 0
L =8m
hence KF = 1.47
Total damage = Z K F& d, 20
. .
... ;.
..._.-.
...
- 1.81 x 0.08 + 1.47 x 0.86
. .-
I
= 1.41
Estimated life = 120/1.41 = 85years (see8.3.2. (h))
which is greater than the specified design life of 60 years
and so the detail may be regardedas satisfactory.
I 1612
. .33
_
!
BS 5400 : Part 10 : 1980
D.4 Exampleof a typical procedure for shear Appendix E -_.

.
. . ...
_.._
. .
connectors
Derivation of standard railway load spectra
D.4.1 Type of bridge. A30 m simply supported composite
highway bridgethat carries a dual carriageway all-purpose
E.l R U loading. The load spectra given in table 2 have
road and issubject to standard UK loading, but which is been based on thetypical trainsshown in figure 19. The
limited to 37.5 units of HB loading. numbers of these trains, assumedfor the three broad traffic
0.4.2 Details o f theproblem. To investigate the fatigue types, are shown in table 15together with the make-up of
capacity of the attached shear connectors (which are in the total annual tonnages. These spectra will cover most
accordance with Part 5) in a normal density reinforced traffic of this type running on lines in Europe.
concrete deck slab of flat soffit. For further information on the derivation of the spectra, the
0.4.3 Classification. Since the connectorsare in following reportspublished by the Office for Researchand
accordance with Part 5, they also comply with 6.4.1 and Experiments of the International Union of Railwaysshould
the weld throat stresses may be calculated according be consulted :
to 6.4.2 or 6.4.3. Report ORE D128/RP5
The potential fatigue crack should be classified as S (from Report ORE D128/RP6 '
type3.12of table17(c).
Report ORE D128/RP7
0.4.4Assessmentusing8.2. Assuming, for this example,
that the shape of the shear force influence line of the girder E.2 RL loading. The load spectra given in table 3 have been
under considerationis similar to that of a single simply based on the typical trains shown in figure 20. The numbers
supported girder, the value of L will lie between 15 m, for of each type of train assumed for the standardspectra,
connectors a t mid-span, and 30 m for connectorsat the together with the make-up of the total annual tonnages, are
ends of the member. shown in table 16.
Hence, for a dual carriageway all-purpose road and class S Table 15. R U loading: annual t r a f f i c tonnage
(seefigure8(b). with the 1.3factorallowedfor37.5unitsof for standard t r a f f i c t y p e s
HB loading (see8.2.3(c)),o~ = 46 x 1.3 = 59.8 N/mmz
for connectors a t mid-span, and 0 ~ = 4 0x 1.3 = 52N/mmz
for connectorsat the ends.
Traffic
type
I Train
two*
I Train
weight,
tonnes
I Number
of train.
per annum
I Total
annual tonnage.
tonnea x 10'
The above values may be checked against uvmax as I
describedin 8.2.2 and illustrated in 0.2 or, alternatively, for Heavy 7 1120 4821 5.40
preliminarydesign purposes in this particular type of
8 1120 7232 8.1 0
example it may be noted that :
-
at mid-span n p mex will equal U Pmin
9 852 15845 13.50
at the ends ~p min will equal zero.
r I
Total I 27.00
I
Hence. the limiting value of up max may be determinedas: Medium 5 I 600 122500 113.50
29.9 N/mmzfor connectors at mid-span and 7 11120 I 2'411 ' I 2.70
52 N/rnmz for connectorsat the ends.
Thus for stud connectors (see 6.4.2) the maximum
8 11120 1 6027 I 6.75
allowable shear load per stud under loading from the
1 11794 I 2 257 I 4.05
standard fatigue vehicle, positioned in accordance
with 7.2.3 may be expressed as:
Light
29.9 Pu/425 = 0.070 Pu kN at mid-span and
52.0 Pu/425 = 0.122 PukN at theends 5.40
where 8.1 0
PUis the nominalstrength of the stud from Part5. 4 I 172 147093 I 8.10
Similarly, for bar or channel connectors, the maximum 5 I 600 I 4 500 I 2.70
allowable shear load per connector under loading from the
standard fatigue vehicle, positioned in accordance
16 I 572 I 2360 I 1.35
with7.2.3, may be expressedas: Total 127.00
29.9 s A \. 1W3 kN a t mid-span and *See fiwre 19.
52.0 S A , \ 10JkNattheends.
where Table 16. RL loading: annual t r a f f i c tonnage
Al istheeffectiveweld throatareainmm2forthe and c o m p o s i t i o n of standard t r a f f i c m i x
particular type of connector, obtained from 6.4.3.1.
Train Train Number Tots1
0.4.5 Comments. It i s not possible to use 8.3 for the t w o weight. of trains annual tonnage.
assessment of shear connectors as the damage chart (see tonnes parannum tonnesx 10'
figure 10) does not include factors for class S details. 1 246 111 545 1 2.84
2 253 I54 032 I 13.67
26 986 5.64
3 463 0.80
...
. ..
......
.--....
*See ligura 20.
1613 34
.- ..... /..
...
- 15 x 6 w 18.5 -
+--- -4
225 6.1 23 19 1.5 1.5 5.2 1.5 1.5 1.5 1.5 5 2 1.5 1.5
+--++-.+H ##'x xx'x xw- m
2.5 23 4 .O 3.52
1 Steel train total load = 1794 t V= 80 km/h
2.6 11.5 2.6 2.6 11.5 2.6

j m
3.6
L-a-a-a-L-a-a-a
total load= 372 t V= 145 km/h
2 Electric multiple unit
2.7 11.5 2.6 2.6 11.5 2.6 2.6 11.5 2.7 2.7 11.5 2.6 2.6 11.5
d-t ivt I $ Xt'tY 4 %%a t* ++- I2.6t %2.6k 11.5
4+2.7 m
3.2 3.2 3.2 3.2 3.2
3 Southern Region suburban total load = 344 t V = 145 km/h
-
2x13 2x11
I .
214.9.5
-1
2x11 2x13
I '
T'
1
2.7 11.5 2.6 2.6 11.5 2.6 2.6 11.5 2.7

+# 4,) % k kC % k kk- m
3.2 . 3.2
4 Southern Region suburban total load= 172 t V = 145 km/h
1.L 2.2 6.9 2.2 3.62.6 11.5 2.6

4+"*-
2.2 22 1.8
5 Diesel hauled passenger train total load= 600 t V = 160 km/h
3.3 6.7 3.3 2.6 11.5 2.6

p m
3.2
6 Electric hauled passenger train total load= 572 t V - 160km/h
Figure 19. Trains included in table 2 spectra
35 .
. __ -. ..
BS 5400 :Part 10 : 1980
,-.,
.. . ..
7 Heavy freight total load = 1120 t V = 72 km/h for medium traffic and 120 km/h tor heavy traffic
t
, ~ I 6x20 ~ i
: ;:e+tS
2.2 6.9 2.2 5.5
-YYX YFY #.e-+ m
22 2.2 3.5
8 Heavy freight total load = 1120 t V = 72 km/h for medium traffic and 120 km/h for heavy traffic
-5116x20~ ~:+%, ~ !2xt7:;:/:

: : 4x20 ii:!+s+
111::11 : :6x20~ ~ ~
4x20 11, t
2.1
-kw7wbF--.##---t
2.1
L.4 21
21 4.6
6.5
3.8
6.5 1.8
3.7
12.8 1.8
3.7
6.5 2.1
4.6 21
4.4
/* +
23 1.8
2.1 4.5
12.8 1.8
m
L-B-8-R-8-8-B-B-8-B-B-B-B-B-S-R-B-B-R-B-B-B-B-S
9 Mixed freight total load = 852 t V = 12Okm/h
NOTE. Inderiving the table 2 spectra, impact effects were calculated in accordance with the recommendations of Leaflet 776-1 R.
published by the International Union of Railways (UIC),14 Rue Jean- Ray F. 75015 Paris.
Figure 19. (Concluded)
..... .
I,
..,....
1615
_.. 36
-. .. _. _. . -. - ... . .. .... ..- .. ... . .... . . . ... . ..- -
. . . .
BS5400:Part10:1980
e-,-.. Wpc
.,..
..
.. ....
2.4 7.9 2.4 2.4 a0 2.4 2 . ~ 8.0 2 . ~2.4 8.0 2 . ~2 . ~7.9 2 . ~2.4 8.0 2.4 2.4 7.9 2.4
v u
A 4 4 A 4 4 A A 4 4 4 4 4 . l 4 4 4 A 4 A A 4 4 4 4 4 A 4
3.2 3.1 3.1 3.2 3.2 3.2
1 total load = 246 t
lnh) bID om (orr9y
2.6 7.9 2.4 2.4 8.0 2.4 2.4 8.0 2.L 2.4 8.0 2.4 2.L 7.9 2.L 2.L 8.0 2.4 2.L 7.9 2.L
Y v v VV v v v v v v v v v v
4 4 4 4 4 4 n n /I'l 4 4 4 A A 4 A 4 4 4
3.2 3.1 3.1 3.2 3.2 3.2
4 total load =I 203 t
I I ;x8.6SI I:; L ~ 8 . 6 5 ~ [r8.7LI

~ ~,1 I:;I ' ~ 8 . 7 4 ~ li;l ;x8-7Li ~ ;f18.65:
: ~ tf
2.3 Y
8.5 2 3 2 3 7.3 2.3 2.3 8.5 2 3 2.3 73 2.3 2.3 7.3 2.3 2.3 8.5 2.3
A 4 A 4 4 4 A 4 4.9 E 4 n A A A 4 1 4 4 A n 4 4
3.3 3.2 33 32 3.3
,I 4 x 9.08 ~ !:;!Lx9.0a !::!

~ ~ ~ Lx 10.13
~~~~ j.1063 Lx10.13 ~ :: L :~ 9~
~ .08~ ! !,
2.3 8.5 2.3 2 3 7.3 2.3 2.3 8.5 2.3 2.3 7.3 2.3 2.3 7.3 2.3 2.3 8.5 2.3
w v
4 4 4 4 4.l A 0 A A 4 4 n n n 4 4 4 4
33 3.2 3.3 3.2 "3.3
6 total load= 231 t
NOTE. In deriving the table 3 spectra, an impact of 30 %was taken for all trains and all spans
I
Figure 20. Trains included in table 3 spectra.
~
8s 5400: Part 10: 1980
Appendix F
Examples of stress histories and cycle counting procedure
Example 1. H i g h w a y bridge. This shows the midspan bending of simply supported spans loaded by the standard
fatiguevehicleand illustratesthe effect of variations in L.
4 at 80 k N
d- 4- - - 4-
I c:
Influence line diagram Standard fatigue vehicle
L 6 1.8
0;3 5'
1.8 < f < 3.6

f uv3
n<L<n ///,f r-- - . .w .......

...................... -y--j
....... .... .... ................
.... ... ... ... ...
....
..
@Vl
5 2
-3
6 < L < 15.6
/ oc2
/' /
L >15.6
Stress histories Cycle counting diagrams
1617
- . . 38 __. .. .......
x
.( , 4
\
\
........
...........
.. .. .. .. ........
N
- b’
.. .. .. .. ............ . . . .
..........
..........
........................
................
................................
................................
................................
................
................
,-
m
b’ 5 c
> \
n \
\
U \
a,
U \
\
\
-
9
0 \
\
\
\ E
a, I
s2
\
\
-
.-C
C
0
0
C
m
2
E
Q)
-
N
m
0
0
.-C
U
C
al
n
L
.-C
0
4
C
9
n
U)
9Q)
1618
:
BS5400:Part10:1980
Example4. R a i l w a y bridge. This shows the mid-span bending of a three-span continuous beam loaded with
. ..
. ..
standard RU loading.
4 at 250 kN
Influence line diagram RU Loading to be multiplied by the dynamic factor

L- base length of loop containing largest ordinate (measured in
direction of travel)
Loading diagram for U,,, Loading diagram for up
Loading diagrams for stress history
2
3
Stress history
0-
R2 Dkr
Cycle counting diagram
NOTE. Inexamples 1 to 4 given above. the cycle counting diagrams, for comparative purposes, follow the same profile as the stress
history. Since the analogy dependssolelyon thedepth of water retained in each section, it is immaterial whether the’profiles are as
illustrated or with successive peaks and troughs joined by straight lines (as proposed in appendix 6 ) .
........
......
..........
..I
...._.
1619
.. _ _ ..
Appendix G Appendix H
I T e s t i n g of shear c o n n e c t o r s Explanatory notes on d e t a i l classification
G.l General. This appendix outlines the procedureswhich
H.l General
should be followed if the fatiguestrength of shear
connectors is to be determined by testing, as required H.l.l Scope. This appendix gives backgroundinformation
by 6.4.1. on thedetailclassificationsgivenin tables17(a), 17(b) and
17(c). This includes notes on the potential modes of failure,
6.2 Procedure. Test thespecimens under constant
important factors influencing the class of each detail type
amplitude loading at frequencies not exceeding 250
and some guidance on selection for design.
cycles/min. Ensure that the frequency of the applied
loading isthe same for each specimen within a particular H.1.2 Geometricaf stress concentration factors.
series of tests. Ensure also that the maximum load on any Unlessothewise indicated in table 17, the stress
connector does not exceed 0.5 times the nominal static concentrations inherent in the make-up of a welded joint
strength of the connector (determined in accordance with have been taken intoaccount in theclassification of the
Part5) with the appropriateconcretestrength, at the time of detail. However, where there is a geometricaldiscontinuity,
testing, being determinedin accordancewith the such as a change of cross section or an apeiture (see figure
requirementsof BS 1881, 21) and/or where indicated in table 17, the resulting stress
Stresses may either be determined from the applied test load, concentrations should be determinedeither by special
in accordancewith clause 6, or derivedfrom strain gauge analysis or by the use of the stress concentration factors
readings. given in figure 22.
G.3 Class S criteria. To enablethe weld metal attaching H.2.Type 1 classifications, non-welded details.
shear connectors to be classifiedas a class S detail, the See table 17 (a).
welds should be shown by tests carried out in accordance H.2.1 Notes onpotentialmodes of failure. In
with 6.2 to have a 97.7 96 probability of surviving 105 and unweldedstee1,fatigue cracks normallyinitiateeitherat
1O7 repetitionsof stress ranges of 146 N/mmz and surface irregularities, at corners of the cross sections, at
82 N/mm’ respectively, where the stress ranges are holes and re-entrant corners or at the root of the thread for
computed in accordance with either 6.4.2 for stud bolts or screwed rods. In steel, which is holed and connected
connectors or 6.4.3 for bar and channel connectors. with rivets or bolts, failure generally initiates at the edge of
Where these conditions are not satisfied, the design a -N the hole and propagates across the net section. but in double
relationship for the detail (Le. 2.3% probability of failure) covered joints made with H.S.F.G. bolts this is eliminated by
should be derived in accordance with appendix A and the the pretensioning. providing joint slip is avoided, and failure
method given in 8.4 should be used to assess the fatigue initiates on the surface near the boundary of the
life. compressionring due to’fretting’ under repeatedstrain.
P o t e n t i a l crack locations
elded attachment
ypical stress distribution
1
*r
7-
The design stress is At t h e attachment thedesign
applied to the appropriate stress is applied tothe
plain material classification appropriate joint clossificat ion
Manhole or r e -entrant corner he design stress for either location should be

aken as the stress on the net section
multiplied by thestress concentration factor
Figure 21. Typical example of stress concentrations due t o geometrical discontinuity
1620
. . -. . 41
BS 5400 :'Part 10 : 1980
..
I
r
'6
(a) Fatigue stress concentration factor for unreinforced apertures Kun

(based on net stress at X)
Length of
s t raight22r
1
Stress
fluctuation
0;
(b) Fatigue stress concentration factor for re-entrantcornersKRc

(based on net stress at X)
Figure 22. Stress concentration factors
.....
.c....:.... .
, .1
1621 __.. . .......

.- ... - .... . .. 42 .. - . .. . _.
Issue ,2,March 1999 BS 5400: Part 10: 1980
H.2.2 Generdcommenrs. In welded construction, materialwhich has previously been fully heat-treated. but
., fatigue failure will rarety occur in a region of unwelded such eomponentsshouldbe subject to special test and
materialsincethefatiguestrength of the welded joints will inspection procedures.
usually be much lower. For the use of black boltscomplying with the
requirementsof BS 4190 and subjected to fluctuating
tensile loads, see 65.
Where boltsor screwed rodsare pre-tensionedto a value in
excess of an applied external load, stress fluctuations will be
H.2.3 Comments onparticular detailtypes governed by the elasticity of the pre-compressedelements.
The increase in tension will rarelyexceed 10% of an
Type 1.3. All visible signs of drag lines should be removed externalload applibd concentrically with the bolt axis, but
from the flame cut edge by grinding or machining. where the load is eccentric, a further increasewill result
7ypes 7.3.9nd 1.4.The presence of an aperture or re-entrant from prying action.
corner implies the existence of a stress concentration and the H.3 Type 2 classifications, welded details on surface
design stressshould bethe stress on the net section of member. See table 17(b).
multiplied by the relevant stress concentration factor (see
figure 22). H.3.1 Notes onpotenti8lmodes o f failure.Seefigure23.
Type 1.4.The controlled flame cutting procedureshould When the weld is essentially parallelto the direction of
ensure that the resulting surface hardness is not sufficient to stressing, fatigue cracks will normallyinitiate at the weld
cause cracking. ends, but when the weld is transverse, cracking will initiate
at the weld toes. Ineithercasethe cracks will then
Type 1.5. This type maybe deemed to include bolt holes for propagate into the stressed element. For attachments
attaching light bracing members where there is negligible connectedby single welds, cracks in parent metal may also
transference of stress from the main member in the direction initiate from the weld root. Cracks in stressed weld metal
of Qr. will initiate from the weld root (see type 3.1 1).Away from
Type 1.6.This covers connections designed in accordance weld ends, fatiguecracks normally initiate at stop-start
with Part 3 forslip resistanceat the ultimate limit state and positions. or ifthese are not present, at weld surfaceripples.
where secondary out-of-plane bending of the joint is With the weld reinforcementdressed flush, failure tends to
restrainedor does not occur (i.e. double-covered symmetric be associatedwith weld defects.
joints). Failure initiates by fretting in front of the hole.
H-3.2 General comments.
Type 1.12. Thisclassification appliestofailureattherootof H.3.2.1 Edge distance. (See figure 24.) No edgedistance
the thread in normal commercialquality threaded criterion exists for continuous or regularly intermittent welds
components. Attention should be paid t o the detailsof head awayfromtheendsofanattachment (seetypes2.1to2.5).
fillets, waistedshanksand thread run-out in components, However, a criterion exists (types 2.6 to 2.1 0) to limit the
not covered by an appropriate British Standard, to ensure possibility of local stress concentrationsoccurring at
that they havesatisfactory fatigue resistance. A higher unwelded cornersas a result of, for example, undercut, weld
fatigue resistancecan be obtained with a rolled thread on
Long attachment Short attachments
A A
f \ r
-
-
Weld failure crocks(type.3.111
Crack types Crack type; 2.9 or 2.10

2.6.2.7 or 2 8 ( o r 2.11 i f o n edge)
(or 2.11 a t edge)
NOTE. For classificationpurposes, an 'attachment'should be
taken as the adjacent structuralelement connectedby welding 10
the stressedelement under consideration. Apart from the
particulardimensional requirementsgivenfor each type in table
17(b). therelativesizeofthe'stressed element'and the
'attachment' is not a criterion.
Figure 23. Failure modes a t weld ends
Avoid or g r i n d out to a smooth profile,
any undercut to thes
.......,..
.... .:,\.
,
..i.:J
(b)
Figure 24. Edge distance
(B BSI 03-1999 43
I - -z- -. .. .. . .
- .. .
BS5400:Part 10: 1980
spatter and excessive leg size at stop-start positionsor cause local buckling (see Part3). If intermediategaps
accidentalovemeavein manual welding. Although this longer than 2.5 I, are requiredthe classshould be reducedto .-,.. . .
criterioncan bespecifiedonlyforthe'width'ditectionofan F. This type also includestack welds to the edges of .. .
element, itisequally important to ensure that noaccidental longitudinal badting stripsirrespectiveof spacing. provided
undercuttingoccurs on the unwelded corners of, for that the welds comply in all respectswith the workmanship
example, cover platesor box members (see figure 24 (b) requirementsfor permanentwelds and that any undercuton
and (c)).Whereitdoesaccur, itshould subsequentlybe the backing strip is ground smooth. The effects of tack welds
groundout toa smooth profile. which are subsequentlyfully ground out or incorporated
into the butt weld by fusion, need not be considered.
Part 5 recommendsthe provision of a minimum edge
distanceof 25 mm for shear connectors, hence the Type 2.7. The classificationmay be deemed to include stress
criterion given in this part willautomatically be met. concentrationsarising trom normaleccentricities in the
thickness direction.
H .3.22 Attuchment of permanent backing strips. If a
permanent backingstrip is used in making longitudinal butt This type includes parent metal adjacent to the ends of
welded joints it should be continuous or made continuous flange cover platesregardless of the shape of the ends.
by welding. These welds and those attaching the backing Types2.7and2.8. Where a narrow attachmentis
stripshouldalsocomply with therelevant class transferringthe entire load out of a wide member, as in the
requirements.Theclassificationwill reduceto Eor F (type case of a welded lap type connection between, for example,
3.3 or3.4) at any butt welds in the backing strip or class E a cross braceand a gusset, the stress in the gusset at the end
atanypermanenttackweld(see H.3.3,type2.4). It should of thecross brace will varysubstantlally across thesection.
be notedthat transverse butt welds on backing strips may be For assessingthe stress in the gusset the effective width
downgraded by tack welds close to their ends (see H.4.3, should be taken as shown in figure 25.
type 3.4).
H A2.3 Stressconcentr8tions. These are increased, and Gusset
hence the fatigue strength is reduced, where :
(a) the weld endsor toes are on, or near, an unwelded Slope = 1 :2
corner of theelement (see H.3.2.1) ;
(b) the attachmentis'long'in the direction of stressing,
and as a result. transfer of a part of the load inthe element Effective
to andfromtheanachment willoccurthroughwelds width X '0;
adjacent to its ends ;
(c) suchload transferisthrough joints whicharenot
Crack I
Y
0 0 0
symmetrical about both axes of cross section of the

stressed element.
H.3.2.4 Weldforms. Full or partial penetrationbutt welded
jointsof Tform (such as would connect attachmentsto the NOTE. Forfailurein thecross braceatxthecros braceisthe
surface of a stressed element) should be completed by fillet 'rnember'and thegusset is the 'attachment'.
weldsof leg length at least equal to 25 %of the thickness of Figure 25. Effective width forwide lap connections
theattachment.Thefil1etsexcludethe possibilityof an
increase in stress concentration arising at an acute re-entrant Type 2.1O.This applies whereany applied shear stressrange
angle between the element surface and the toe of the weld. is (numerically) greater than 50 %of a co-existent applied
and thus, in consideringthe effects on the stressed element. direct stress range.
it is immaterialwhether the attachment is fillet or butt
Experimental evidence indicatesthat where significant
welded to the surface, since a similar toe profile results in shear stress co-exists with direct stress, the use of
both cases. principalstress values may be conservativeand
H.3.2.5 Tuck welds. Tack welds, unless carefully ground out accordingly the classificationis upgraded.
or buried in a subsequentrun, will provide potential crack Type 2.1 1. This type applies regardlessof the shape of the
locationssimilartoany other weld end. Their use in the end of the attachment. In all cases, care should be taken to
fabrication process should be strictly controlled. avoid undercut on element corners or to grind it out to a
NOTE. Apart fromthe width transverse to U,,neither the shape of smooth profile should it occur. In particular, weld returns
theendolanattachment northe orientationorcontinuityofthe across a corner should be avoided and the use of cover
weld at itsendaffects the class.
plateswider than the flange, to which they are attached, is
H.3.3 Comments onparticular detail types not recommended.
Type 2.1 Finish machining should be in the direction of H.4 Type3 classifications, welded details a t end
cr. Thesignificanceof defects should be determinedwith connectionsof member. Seetable 17(c).
the aid of specialist advice and/or by the use of a fracture H.4.1 Notes onpotentia/modes o f failure. (See figure
mechanicsanalysis. The N.D.T. technique should be 26.) With the ends of butt welds machinedflush with the
selectedwith a view to ensuring the detection of such plate edges, or as otherwise given below, fatigue cracks in
significant defects. This type is only recommendedfor use in the as-welded condition normally initiate at the weld toe
bridgeworksin exceptional circumstances. and propagate into the parent metal, so that the fatigue
Type2.2.Accidental stop-starts are not uncommon in strengthdepends largely upon the toe profile of theweld. If
automatic processes. Repair to the standard of a C the reinforcementof a butt weld is dressed flush, failure can
classificationshould be the subject of specialist advice and occur in the weld material i f minor weld defects are exposed.
inspectionand should not be undertaken in bridgeworks. e.g. surface porosity in the dressingarea (see H.4.3. type
Type2.4.The limiting gap ratiomlh applies even though 3.3).
adjacent welds may be on opposite sides of a narrow In the case of butt welds made on a permanent backing, . '. I
attachment (as in the case of a longitudinal stiffenerwith fatigue cracks initiate atthe weld metal-strip junction and LJ
staggered fillet welds). Long gaps between intermittent then propagate into the weld metal.
fillet welds are not recommendedas they increasethe risk In fillet or partial penetrationbutt welds, fatigue cracks in
of corrosion and, in the case of compression members, may weld metal will normally initiate from the weld root.
BS 5 4 0 0 : P a r l l O : 1980
+t+
/2 min.
3.1 1 3.1 1
3.3
-
3.9
I J
NOTE. Fatiguecracks in reinforcing bars will normally initiate in

similar locations to those for structuraljoints, given similar stress
conditionsand joint geometry.
Figure 26. Type 3 failure modes
H.4.2 Generalcomments dependent upon the welding proceduresadopted.

H.4.2.1 Misalignments.The classifications may be deemed Accordingly, their use is not recommendedunless subject
to include for the effects of any accidental misalignmentsup to special tests and strict proceduralcontrol.
to the maximum value specified in Part 6,provided that the H.4.2.6Partialpenetration butt welds. All butt welds
root sides of joints with single sided preparations(i.e. single transmitting stress between ends of plates, sections or
bevel -J, -U or -V forms) are back-gouged to a total width built-up members in bridgeworksshould be full penetration,
at least equal to half the thickness of the thinner element. except where permitted in types 3.8 and 3.10 (junctions
H.4.2.2 Design stresses. For elements where out-of-plane with transverse members). In these latter cases, even
bending is resisted by contiguous construction (e.g. beam though the joint may be required to carry wholly
flanges supported by webs, wide plates supported by compressivestressesand the non-penetratedsurfacesmay
effectively continuous stiffeners, etc.) eccentricities due to be machined to fit. for fatigue purposes, the total stress
axial misalignmentsin the thickness direction may be fluctuation should be consideredto be transmitted through
neglected. Where such support i s not provided (e.g. tension the welds (e.g. column caps and bearing stiffeners).
links) the design stress should include an allowance for the H.4.2.7 Welding of reinforcing bars for concrete. Welding
bending effects of any intentional misalignment, i.e. the of reinforcementshould complywith Parts7 and 8. Lap
'-. nominaldistance between the centres of thicknessof the welding of bars is no! classified since adequate control
two abutting components. For componentstapered in cannot be exercisedover the profile of the root beads and its
thickness, the centre of the untaperedsection should be use is not recommended under fatigue conditions.
used. H .4.3 Comments on particular detail types
H.4.2.3 Elementedges. In all cases,failurestend to be Type3.1. The significance of defects should be determined
associatedwith plate edges and care should be taken to with the aid of specialist advice and/or by the use of a
avoid undercut at the weld toes on the corners of the cross fracture mechanicsanalysis. The N.D.T. technique should
section of the stressed element (or on the edge at the toes be selected with a view to ensuring the detection of such
of any return welds). Should it occur, any undercut should significant defects. This class should not normally be used
be ground out to a smooth profile. in bridgeworks (see 5.1.2.5):
H.4.2.4 Pan width welds. Butt type welds may also occur Type3.2. Shop welds made entirely in the downhand
within the length of a member or individual plate as, for position, either manually or by an automatic process other
example, in the case of: than submergedarc, tend to have a better reinforcement
(a) aplugweldtofillasmallhole; shape from the point of view of fatigue than positional, site
(b) a weld closing a temporary access hole with an infill or submerged arc welds (i.e. larger re-entrant angles at the
plate; toes and more uniform profiles). Accordingly, to cater for
exceptional circumstances, joints made in this manner may
(c) a hole or slot for a transverse member to slot through be up-graded to class D.
a wider member.
Types3.2and3.3. Thicknessvariations and surface
Although such geometrieshave not been given specific
misalignments uptothe maximumvaluesspecifiedin Part 6
categories in table 17(c), types 3.3 and 3.4 may be deemed
to cover plug and infill plate welds and types 3.7 and 3.8 may bedeemed to beincluded (see H.4.2.1).
L2 may be deemed to cover slotted members. See also H.4.3. Thesetyf.ms do not normally include joints between rolled
or built-up sections. See the note to type 3.6.
H.4.2.5 Joints weldedfrom oneside only. Unless made on
a permanent backing (type 3.4) welds made entirely from Type 3.3. Grinding smooth the reinforcement of butt welds
one side are not classified since the root profile will be until flush with the plate surface on both sides is Qenerally
1624 45
-
. . - .. .
> i
BS5400:Pan10:1980
beneficial. Providedthat N.D.T. is done after grinding, this weld is made on a permanent backing (type 3.4, see figure
treatment can be assumed to raisethe class to D. 27). Dressingof the weld reinforcement is advised to
Types3.3and3.4. These types may be used for holes which overcome poor reinforcementshape resultingfrom the
areeitherfilled with plugs of weld metal orwelded infill greater misalignments which may occur in thejointing of
plates. Such holes may also be requiredfor stitching sections.
laminationsor repairing lamellar tears. The welds should be NOTE. This joint is frequently made using a semi-circular cope
full penetrationand should be consideredto be equivalent hole. This gives improved access to the flange butt welds when
webs or longitudinalstiffeners have already beenattached. The
to type 3.3 or. if welded onto permanent backing material. end of the web butt weld at the cope holecan beconsideredto be
type 3.4.The slot or hole dimensionsshould be in equivalent to class D with a stress concentration factor of 2.4
accordancewith appendix A of 8s 5135 :1974. provided that the end of the butt weld and thereinforcement
Plug weldsshould not be used in bridgeworksfor within a distanceequalto the radius ( I )areground flush. Cope
transmitting tensile force across two lapping plates. Their holes of 45" mitrearenot recommended.
use fortransmitting shear force is not recommendedfor Types 3.7end3.8. Weld metalfailure will not govern with
major structuralconnections, but where they have to be full penetrationwelds.
used, failure through the weld throat should be considered Where the third member is a plate it may be assumed that
to beclass W, based on the minimum throat area projected plane sections remain plane in the main membersand that
i n the gr direction. axial and bending stressdistribution in the a,direction are
T y p e 3 k I f the backing strip is fillet or tack welded to the unaffected.Where the third member is an open shape, for
plate (type 2.9)thedetail class will not be reduced below example, an I section ora hollow tube, particularlyi f
class Funless permanentlytacked within 10mm of the different in width, a discontinuity in the main memberstress
memberedge, in which case it will beclass G (type 2.1 1). pattern will occur. In this case the stress parameter should
Type 3.5. The effect of the stress concentration at the corner be the peak stress concentration at the joint. In the absence
of the joint betweentwo individual plates of different of publisheddata on a particularjoint configuration, the
widths in line may be included in the classification.Where stress concentration factor may have to be determinedby
the end of one plate is butt welded to the side of another, finite element or model analysis.
refer to type 3.9. Planesections may be assumed to remain plane where the
Stress concentrationsdue to abrupt changes of width can main member stress can be continued through the transverse
often be avoided by tapering the wider plate (see types 3.2, member by additional continuity plating of comparable
3.3and 3.4). cross-sectionalarea, which is in line with the main member
components (see figure 28). Inthistypeof connection it is
Type3.6. Butt welds between rolled sections or between important that the joint regionsof the third member are
built-up sectionsare prone to weld defects. which are checked before welding for lameller rolling defects and after
difficult todetect, in the region of the web/flange junction welding for lamellar tears.
(see figure 27). Special preparations, procedures and
inspection may be undertaken in exceptional Where two flat plates intersect in thesame plane, as in the
circumstancesand type 3.3may then be applied unless the case of flangesatthe junction of two girders, the stress
concentrationfactor dueto the abrupt change of width
should be used (see figure 29). If the weld is a full
penetration butt carriedout in accordance with all the
recommendationsfor type 3.5the detail may be classed as
F2 without applying a stressconcentrationfactor.
These types may be deemed to cover the case where a
narrow third member i s slotted through a single main
memberawayfroman endconnection (see figure 30).In
thiscase, the third membershould beassumedto transmit
the stress which the parent material would have carried
beforetheslotwascut. lfthelength oftheslotislonger
than 150 m m inthe or direction, type3.7 (full penetration
buttjoints) should be reclassedfrom F to F2.Notethat this
detail should generally be avoided, when possible, as slots
are difficult to cutaccurately welding isoften
poor. Where member B is called upon to carw high tensile
stress, a slot i n A avoids any risk from lamellartears.
However, with respect to stress fluctuation in member B the .
Radius r"i detailshowninfigure30istype2.11 (classG) atpointy.
If B iscriticaland Ais not, circularcut-outsatthecornersof
Figure 27. Type 3.6 joint B will improve the class to F (type2.9).
1625 46 _.
+
..,. . . . .- . ..._. _.
BS 5400 : Part 10 : 1980
f -x Typical stress patterns

!. ' .I
without with
< 0.5t3
Through
thickness
direction
Detail X (cruciform joint)
Figure28. Use o f continuity plating to reduce stress concentrations in type3.7 and 3.8 joints
0;
Figure 29. Cruciform junction between flange plates
Class
cut out
Class I
Class
A
Alternative detail if member 'B' is critical

. .
NOTE. Main member'A isslotted. Ihird'member'B' is
..
continuous and is welded all round the slot to 'A'.
Figure 30. Example o f a 'third' member slotted through a main member

Type3.8. In afillet weldedjoint weld metal failure (see
type 3.1 1) will normally govern, unless the total weld Stress distribution
leg length isabout twice the element thickness. It will also
govern in a partial penetration butt welded joint except
where reinforcedwith fillet welds of adequatesize.
\
Types3.9and3.10.Thesedetail typesare distinguished
from types 3.7 and 3.8 by the absence of a similar member in
line on the far side of the joint. In this case an axial
component in the first member will induce bending moment
and hence curvaturein the transverse member. Unless the
latter isvery stiff in bending an uneven stress distribution
will result. Memberswith bolted end connectionsvia
transversely welded end platesareparticularlysusceptible
to local increase of stress (see figure 31). Note that if the
transversemember is an open or hollow section, local
bending will increase the peakstressfurther (as in the case
of types 3.7 and 3.8). ' :... .,
As faras fatigue failure of thitransverse member is Figure31. Exampleof type3.9or3.10Joint
concerned, the first member is treated as a type 2
attachmentand the stress parameter is the stress in the
transverse member without the application of a stress F u l l penetration
concentrationfactor. In hollow or open transverse members

this stress is otten magnified by local bending of the walls.
butt weld
I
Often the load is transmitted from a member to a transverse
member primarily via flange plates in the same plane. This
can occur in the case of a junction between cross girders and
main girders, diagonals and truss chords, or in vierendeel ronsverse
frames (see figure 32). If the transverse member i s relatively
stiff (i.e. its width is at least 1.5 times the width of the first
member) and a full penetration butt weld is used in
accordance with the recommendationsof type 3.5, the
classificationin this particular case may be consideredto be
eftectively F2 with a stress concentrationfactor of unity.
Otherwise theclass shall be F with the appropriatestress
concentration factor (see comment on detail types 3.7 and
3.8).Note that in the case of trusses, secondary stressesdue
to joint fixity should be taken into account. The fatigue concentration factor
strength of both flange plates may be improved by the
insertionof a smoothly radiusedgusset plate in the Figure32.Teejunctionof twoflange plates
transverse member so that all butt welds are well away from
re-entrantcorners (see figure 33).
Type3.7 1 . Class W is primarily intended to apply toall
fillet or partial penetration butt weld joints where bending
action across the throat does not occur. Where lappedjoints
are welded on two or more sides, or tee or cruciformjoints
areweldedfrom bothsides (asshownin tablel7(c)),such ypes 1.3 or 1.4 + stress
bending action is normally prevented. I n certain cases
difficulty of access may onlyallow welding to be done on
one side of the joint. This applies particularlyto small
hollow members with welded corners, which if subject to
loading that distorts the cross section, may cause failure of
thecornerweld in bending (see figure 34). Where axial
stress is also present, the stress rangeat the face of the weld
may be different from that at the root. Failurefrom ripples or
stop-start positions on the face may give a higher strength
than class W, but expert advice should be sought if a higher
strength is required. In most casesfailurefrom stress
fluctuation in this root will be critical and this should always Figure 33. Alternative method o f joining two flange pleteo
beclassified as W.
Type 3.12. The stress ratio and effective weld size criteria of
this clause are intended to aid in the exclusion of premature
failures by local crushing of the concrete, by tearing of the
attached flange or in the body of the connector.
This type covers embedded shear connectorsat any
position along a girder. The reference to 'end connections' in
the title of table 17(c) refers to the end of the member in
which failure occurs; in this case the welded end of the .. . .
....
:. .:.::
shear connector. -
....,
. ...
Type3.13. Single sided manual metal arc procedures, with
or without the use of backing materia!, are not recommended Corner detail
unlessspecialistadvice is sought. Weld metal failure need
not be considered. Figure 34. Single fillet corner weld In bending
1627
-. . . - 48 -. . __ . - _. ... .
I .
I
i
.. ...
..
.
.. .
--
-.
r
- .-
.
,
. . .._
.-
i!
i
.
is blank
Previous page
-
.
.. ....
...
’!
4
-
1
1
f
f
e
/i
it
c
c
is blank
Previous page
jl
Standards publications referred to
BS 1881 Methods of testing umcrete
BS 3643 Is0 metric screw threads
BS 3692 IS0 metric precision hexagon bolts, screws and nuts
BS 4190 IS0 metric black hexagon bolts, screws and nuts
BS 4395 High strength friction grip bolts and associated nuts and washers for structural engineering
BS 4604 The use of high strength friction grip bolts in structural steelwork. Metric series
BS 5135 Metal-arc welding of carbon and carbon manganese steels
BS 5400 Steel concrete and composite bridges
.. . .. .... -
BS 5400 : Part 10.:1980

This British Standard. having been prepared under the direction o f implementing the standard, of nemsary details such as symbols and <--.
the Civil Engineering a n d Building Structures Standards Committee, size, type or grade designations. Enquiries should be addressed t o
was published under the authority of the Executive Board and the Publications Manager, BSI. Linford Wood. Milton Keynes
comes into effect on 31 January 1 9 8 0 . MK14 6LE. The number for telephone enquiries is 0908 220022
(JBritish Standards Institution. 1980 and for telex 825777.
Contract requirements. A British Standard does not purport to
ISBN 0 580 10567 9
include all the necessary provisions of a contract. Users of British
The following BSI references relate t o the work o n this standard: Stand8rdS are responsible for their correct application.
Committee reference C S B / 3 0 Draft for comment 74/13197 D C ~~
~~
Revision of British Standards. British Standards are revised, when

British Standards Institution. Incorporated b y Royal Charter. BSI i s necessary. b y the issue either o f amendments or o f revised editions.
the independent national body for the preparation of British It is important that users of British Standards should ascertain that
Standards. It i s the UK member of the International Organization they ara in possession of the latest amendments or editions.
for Standardization and U K sponsor of the British National Automatic updating senicr. BSI provides an economic, individual
Committee o f the International Electrotechnical Commission.
and automatic standards updating service called PLUS. Details are
In addition to the preparation and promulgation of standards, BSI available from BSI Enquiry Section at M i l t o n Keynes, telephone
of fers specialist services including the provision of information 0908 221166, telex 825777.
through the BSI Library and Standardline Database; Technical Help
to Exporters; and other services. Advice can be obtained from the Information on all BSI publications is in the BSI Camlogus.
Enquiry Section. BSI, Milton Keynes M K 1 4 6LE. telephone supplemented each month b y BSI News which i s available t o
0908 221 166. telex 825777. subscribing members o f BSI and gives details of new publications.
Copyright. Users o f British Standards are reminded that copyright revisions. amendments and withdrawn standards. A n y parson who,
subsists in all BSI publications. N o part o f this publication may be when making use of a British Standard. encounters an inaccuracy or
reproduced in any form without the prior permission in writing of
SI. This does n o t preclude the free use#in the course of
ambiguity, i s requested to n o t i f y BSI without delay in order that
the matter may be investigated and appropriate action taken.
.$. -
C o o prating organizations
The Civil Engineering and Building Structures Standards Committee, Department o f the Environment (Transport and Road Research
under whose direction this British Standard was prepared. consists Laboratory)
of representatives l r o m the following Government departments and Department o f Transpon
scientific and industrial organizations: Federation o f Civil Engineering Contractors
Association of Consulting Engineers Greater London Council
Association 01 County Councils Institution of Civil Engineers
British Constructional Steelwork Association Institution of Highway Engineers
British Precast Concrete Federation Ltd. Institution of Municipal Engineers
Brilish Railways Board Institution of Structura! Engineers
British Steel Industry London Transport Executive
Cemenr and Concrete Association Ministry of Defence
Concrete Society Limited Sand and Gravel Association Ltd.
Constructional Steel Research and Development Organization Scottish Oevelopment Department
Department 01 the Environment IBuilding Research Establishment) Welding lnstirute
Amendments issued since publication
Amd. No. Date of issue Text affected
British Standards Institution . 2 Park Street London W l A 2BS - Telephone 01-629 9000 Telex 266933
8903 -9-0 6 k - B CSBl30
-__. I632 .. ...... .. ._:__._ . , . _.,,. ,..

... .- . . . . . . . . . .- ......... . . . . . . . - . . .- .... ._ -___-
Amendment No. 1
I
published and effective from 16 March 1999

to BS 6400 :Part 10 : 1980
Steel, concrete and composite bridges -

Part 10. Code of practice for fatigue
~ ~~~ ~
Instructions for replacement of pages

The following pages contain new or revised text. Please remove any superseded pages and insert the new or
revised pages in the position given in the sumnlary of pages (see page a). Where only one of the two pages on
each sheet has been updated,the other page has been reprinted.
F'ront cover and inside front cover

aandb
candd
land2
3and4
4a and 4b
5
43
49
51
53
Back cover
You may wish to retain the superseded pages, e.g. for reference purposes. If so please nurk them 'Superseded by
issue X",where 'X" is the appropriate issue number. If you do not wish to retain the superseded pages, please
destroy them.
BRITISH STANDARD BS 5400 :
Part 10 : 1980
Steel, concrete and

composite bridges -
Part 10: Code of practice for fatigue
--0
V I -
n-=
NO COPYING WITHOUT BSI PERMISSION EXCEPT AS P E R M I m D BY COPYRIGHT LAW =
*.
.<
- - _-_ -
~... . . . ... . ...
M e 1, March 1999 BS5400:Part10:1980
Summary of pages I
Prge XS8Ue p43e Isam

Front cover 2 6 Original
Inside h n t cover blank 7to42 original
a 1 43 2
b blank 44 Original
C blank &to48 original
d 1 49 2
1 2 50 blank
2 2 51 2
3 2 62 blank
4 2 53 2
4a 1 64 blank
4b blank Inside back cover Original
5 2 Back cover 2 I
0 BSI a 1 9 9 9
a
. . . ..... -
........ . . . . . .
BS 5400:Part 10: 1980 Issue 1, March 1999
Contents .-.
PaBe Page
Foreword 1 8.2.2 Procedure 12
Cooperating organizations Beck cover 8.2.3 Adjustment factors for OH, class S
Recommendations details only 12
1. scope 2 8.3 Damage calculation, single vehicle
1 .l General 2 method 12
1.2 Loading 2 8.3.1 General 12
1.3 Assessment procedures 2 8.3.2 Procedure 12
1.4 Other sources of fatigue damage 2 8.4 Damage calculation, vehicle spectrum
1.5 Limitations 2 method 14
1.5.1 Steel decks 2 8.4.1 General 14
1.5.2 Reinforcement 2 8.4.2 Design spectrum 14
1.5.3 Shear connectors 2 8.4.3 Simplification of design spectrum 14
2. References 2 8.4.4 Calculation of damage 14
3. Definitions and symbols 2 9. Fatigue assessment of railway bridges 18
3.1 Definitions 2 9.1 Methods of assessment 18
3.2 Symbols 3 9.1.1 General 18
4. General guidance 3 9.1.2 Simplified procedure 18
I 4.1 Design life 3 9.2 Assessment without damage calculation 18

42 Classificationand workmanship 9.2.1 General 18
3
4.3 StleSSeS 4 9.2.2 Procedure 18
4.4 Methods of assessment 4 9.2.3 Non-standard design life 18
4.5 Factors influencing fatigue behaviour 3 9.2.4 Multiple cycles 20
5. Classification of details 4 9.3 Damage calculation 20
5.1 Classification 4 9.3.1 General 20
5.1 .I General 4 9.3.2 Design spectrum for standard loading 20
5.1.2 Classification of details in table 17 4 9.3.3 Design spectrum for non-standard
5.2 Unclassified details 4 loading 20
5.2.1 General 4 9.3.4 Simplification of spectrum 20
5.2.2 Post-welding treatments 4 9.3.5 Calculation of damage 20
5.3 Workmanship and inspection 4 10. Fatigue assessment of bridges carrying
5.3.1 General 4 highway and railway loading 20
5.3.2 Detrimental effects 4 11. The Palmgren-Miner rule 20
5.4 Steel decks 4 11.1 General 20
6. Stress calculations 4 11.2 Design ur-N relationship 22
6.1 General 4 11.3 Treatment of low stress cycles 22
6.1.1 Stress range for welded details 4 11.4 Procedure 22
6.1.2 Stress range for welds 4 11.5 Miner's summation greater than unity 22
6.1.3 Effective stress range for non-welded Appendices
details 4 A. Basis of U,-N relationship 23
6.1.4 Calculation of stresses 4 B. Cycle counting by the reservoir method 25
6.1.5 Effects to be included 5 C. Derivation of standard highway bridge
6.1.6 Effects to be ignored 5 fatigue: loading and methods of use 25
6.2 Stress in parent metal 5 D. Examples of fatigue assessment of
6.3 Stress in weld throats other than highway bridges by simplified methods 30
those attaching shear connectors 5 E. Derivation of standard railway load
6.4 Stresses in welds attaching shear spectra 34
connectors 6 F. Examples of stress histories and cycle
6.4.1 General 6 counting procedure 38
6.4.2 Stud connectors 6 G. Testing of shear connectors 41
6.4.3 Channel and bar connectors 6 H. Explanatory notes on detail
6.5 Axial stress in bolts 6 classification 41
7. Loadings for fatigue assessment 6
7.1 Design loadings 6 Tables
7.2 Highway loading 6 1. Annual flow of commercial vehicles
7.2.1 General 6 (nc x 106) 8
7.2.2 Standard loading 6 2. Standard load spectra for RU loading 11
7.2.3 Application of loading 6 3. Standard load spectra for RL loading 12
7.2.4 Allowance for impact 8 4. Values of k, for RU loading of
7.2.5 Centrifugal forces 8 railway bridges 19
7.3 Railway loading . 8 5. Values of k
. for railway bridges 19
7.3.1 General 8 6. Values of ks for railway bridges 19
7.3.2 Application of loading 8 7. Values of ks for RL loading of
7.3.3 Standard load spectra 10 railway bridges 19
8. Fatigue assessment of highway ' 8. Design or-N relationships and constant
bridges 12 amplitude nonlpropagating stress
8.1 Methods of assessment 12 range values 22 . ._
.i::::,:,
8.1.1 General 12 9. Mean-line Cr-N relationships 23 .......
.......
........
.....
8.1.2 Simplified procedures 12 10. Probability factors 23
8.2 Assessment without damage 11. Typical commercial vehicle groups 27
calculation 12 12. Proportional damage from individual
8.2.1 General 12 groups of typical commercial vehicles 28
d BSI 03-1999
1 -. . . . . . . . - .- - - - - Q
. . ... ... .- . . - .- . ._ .. . _.
. . -. ... . . ... . . .. .. . .. -.
lssue 2, March 1999
Page Page
13. Typical commercial vehicle gross 12. Typical point load influence line 17
weight spectrum 29 13. Simplification of a spectrum 18
14. Typical commercial vehicle axle 14. -
Summary of design 'U, Ncurves (mean
weight spectrum . 29 minus two standarddeviations) 21
15. RU loading: annual traffic tonnage for 15. Summary of mean-line ur-N curves 23
standard traffic types 34 16. Typical or-N relationship 24
10. R L loading :annual traffic tonnage and 17. Multiple paths 28
composition of standard traffic mix 34 18. Typical Miner's summation adjustment
17. Classification of details CUNB 29
17(e). Non-welded details 49 19. Trains included in table 2 spectra 35
20. Trains included in table 3 spectra 37
I 17@). Weldeddetarlsotherthanatend
oonnedionsdamember
17(c). Welded detaii at end ##\necbonsof
51 21. Typical example of stress concentrations
due to geometrical discontinuity 41
22. Stress concentration factors 42
member 53 23. Failure modes at weld ends 43
Figures 24. Edge distance 43
1. Method of indicating m i n i m dass 25. Effective width for wide lap
4a connections 44
requirements~df=ms
2a. Referencestress in parent metal 5 26. Type 3 failure modes 45
27. Type 3.6 joint 46
2b. Referencestressinweldthroat 5
28. Use of continuity plating to reduce

3. Axle arrangement of standard fatigue stress concentrations in type 3.7 and
vehicle 7 3.8 joints 47
4. Plan of standard axle 7 29. Cruciform junction between flange
5. Designation of lanes for fatigue plates 47
purposes 9 30. Example of a 'third' member slotted
6. Transverse location of vehicles 10 through a main member 47
7. Impact allowance at discontinuities 10 31. Example of type 3.9 or 3.10 joint 48
8. Values of UM for different road 32. Tee junction of two flange plates 48
categories 13 33. Alternative method of joining two
9. Derivation of uv and for damage flange plates 48
calculation 15 34. Single fillet corner weld in bending 48
10. Damage chart for highway bridges
(values of A l o ) 16
11. Miner's summation adjustment factor
KF for highway bridges 17
Foreword
BS 5400 is a document combining codes of practice to Part 8 Recommendationsfor materialsand
cover the design and construction of steel, concrete and workmanship, concrete, reinforcement and
composite bridges and specificationsfor the loads, prestressing tendons
materialsandworkmanship. It comprises the following Part 9 Bridge bearings I
I PaRsandSections: Section 9.1 Code of practice for design of
Part 1 General statement bridge bearings
Part 2 Specification for loads Section 9.2 Speciticabjon for materials.
1 Part 3 Code of practicefor design of steel bridges manufacture and installation of
Part 4 Code of practice for design of concrete bridges bridge bearings
Part 5 Code of practice for design of composite Part 10 Code of practiceforfatigue
bridges
Part 6 Specification for materialsand workmanship,
I
A BritishStandard does not purport to indude all the necessary
steel provisions of a contract Users of British Standards are
Part 7 Specification for materialsandworkmanship,
responsible for their comet3application.
concrete, reinforcementand prestressing
tendons
Complianke with a BritishStandard does not of itself
confer fmmunity from legal obligations. I
Q BSI 03-1999 1
...
.
_ _. . ... ....- . .. ._
....... .
BS 5400: Part 10: 1980 h2March1999
1. Scope 3. Definitions and symbols

1.1 General. This Part of this British Standardrecommends 3.1 Definitions. For the purposesof this Part of this British
methods far the fatigue assessment of parts of bridges which Standardthe following definitions apply.
are subject to repeated fluctuations of stress. 3.1.1 f8figue. The damage, by gradualcrackingof a
1.2 Loading. Standard load spectra are given for both structural part, caused by repeatedapplications of a stress
highway and railway bridges. which is insufficient to induce failure by a single application.
1.3 Assessment procedures. The following alternative 3.1 .2108ding event. The approach, passageand
methods of fatigue assessment are describedfor both departure of either one train or, for short lengths, a bogie or
highway and railway bridges: axle, over a railway bridge or one vehicle over a highway
(a) simplified methodsthat are applicableto partsof bridge.
bridges with classified details and which are subjected to 3.1.3losd spectrum. A tabulation showing the relative
standard loadings : frequencies of loading events of different intensities
(b) methods using first principlesthat can be applied in experienced by the structure.
all circumstances. NOTLA convenientmode of expressing a load spectrumis to

denote each loadintensityas a propordon (Kw) of a standard load
1.4 Other sources of fatigue damage. The following and the number of occurrencesof each load as a proportion (&)
topics are not specifically covered by this Part of this of the total number of loadingevents.
British Standard buttheir effects on the fatigue life of a
structure may need to be considered: 3.1.4standardloadspectrum. The load spectrum that
has been adopted in this Part of this British Standard,
(a) aerodynamically induced oscillations-: derivedfzom the analysis of actual traff ic on typical roads or
(b) fluctuations of stress in parts of a structureimmersed rail routes.
in water, which are due to wave action andlor eddy
3.1.6stresshistory. Arecord showing how the stress at a
induced vibrations;
point varies'during a loading event.
(c) reduction of fatigue life in a corrosiveatmosphere
3.1.6 combinedstress history. A stress history resulting
(corrosion fatigue).
from two consecutive loading events, i.e. a single loading
1.5 Limitations event in one lane followed by a single loading event in
1.6.1 Steeldecks. Highway loading is ihcludedin this another lane.
Part and is applicable to the fatigue design of welded 3.1.7stresscycle(orcycleofstress). Apatternof
orthotropic steel decks. However, the stress analysisand variation of stress a t a point which is in the form of two
classification of details in such a deck i s very complex and is opposing half-waves, or, if thisdoes not exist, a single
beyond the scope of this Part of this British Standard. half-wave.
1.5.2Reinforcement. The fatigue assessment of certain 3.1.8stressrange (orrange ofstress) (0,). Either
details associated with reinforcing bars is included in this
Part but interim criteria for unwelded barsaregiven in (a) in a plate or element, the greatest algebraic difference
Part 4. between the principal stresses occurring on principal
NOTE. These criteria are at present under review and revised planes not more than 45"apart in any one stress cycle; or
criteria may be issued later as an amendment. (b) in a weld, the algebraic or vector difference between
1.5.3 Shear connectors. The fatigue assessment of shear thegreatest and least vector sum of stresses in any one
connectors between concreteslabs and steel girders acting stress cycle.
compositely in flexure is coveredin this Part, but the 3.1.9stressspectrum. Atabulation of the numbers of
assessment of the effects of local wheel loads on shear occunences of all the stress rangesof different magnitudes
connectors between concrete slabs and steel plates is during e loading event.
beyond the scope of this Part of this British Standard.
This effect may, however, be ignored if the concreteslab 3.1.10 design spectrum. A tabulation of the numbers of
alone is designed for the entire local loading. occunencesof all the stress rangescaused by all the loading
events in the load spectrum, which is to be used in fatigue
assessment of the structural part.
2. References
3.1.11 detailclass. Aratinggiven toa detail which
The titles of the standards publicationsreferred to in this indicates its level of fatigue resistance. I t is denoted by the
standardare listed on the inside back cover. following :A, B, C. D, E, F, F2, G, S, or W.
~ 7 h s m e x m u l l l ~ c l a bthehighestreammended
Ss
dass,that can be a c h i i w w t h e tlghedworknanshp spedfied in
Pan6 (seetable 17). The minmumrequired dass to be specifiad for
fakkatknpurposes f e h b $the bvvest 4 - N a m e In Csgw 14.
which results in a life excaeding thedasign L.
...
........
.. .....
.......
-:..:..
2 0 BSI 03-1999
............. .- . . . . . . ................ . ._
.......... -.
~~ ~~~~~~ ~~~~~~ ~~ ~~~

. . .. . .
.. . . . .. . , . .... . .... . - . . . . . __ - __
-.
-
issue 2 March 1999 BS 5400 :Part 10 : 1980
3.1.12 ur-Nnvlationship oru,-Ncurve. The % Miner's summation

quantitative relationship between ur and Nfor a detail Q0 Stress on the corearea of a bolt determined on
which is derivedfrom test data on a ptobability basis. the basis of the minor diameter
3.1.13 design ffr-NCume. The ur-Nrelationship on Limiting stress range under loading from the
adopted in this Part of this British Standard for design on standardfatigue vehicleon a highway bridge
the basisof 2.3 %probability of failure. UN Stress on net section
QO Constantamplitude non-propagating stress
3.1.14 design life. The period in which a bridge is required range (uratN= 107)
to performsafelywith an acceptableprobabilitythatit will Algebraic value of stress in a stress history
not requirerepair. Maximumand minimumvalues of up from all
ffp max
3.1.1 5 stsndsrddesign life. 120years, adopted in this
Pan of this British Standard.
Up min
4
} stress histories produced by standard loading
Range of stress (stress range) in any one cycle
3.1.1 8 Miner's summstion. A cumulativedamage Url, Un Individual stress ranges (Or) in a design
... etc spectrum
summation basedon the rule devisedby Palmgrenand Miner.
OR max -
(ffp max ffp min) for a railway bridge .
3.2 Symbols. The symbols in this Part of this British f f ~ ~ , f Stress
f ~ ~ranges (indescending orderof
Standardare as follows. ... etc magnitude) in astress history of a railway bridge
under unit uniformly distributed loading
A Net area of crosssection Limiting stress range under standardrailway
Effectiveweld throat area for the particular type UT
Ai loading
of connector Nominal ultimatetensilestrength,to be taken as
d ffu
Numberof standarddeviations below the mean
1.10~unless otherwise specified

line Or -Ncurve Value of Or under loading from the standard
ffv
di 2 0 Life time damagefactor (Miner's summation for fatigue vehicle (highway bridges)
120 million repetitionsof a stress rangeU, in a
highway bridge)
UV max (Up max- min) for a highway bridge
uvt. uvr Valuesof uv (indescendingorderof
F Designstress parameterforbolts
KO Parameter defining the mean line Ur-N
... etc magnitude) in any one stress history for one
relationship lane of a highway bridge
Parameter defining the ur-Nrelationship for uv1A The largest value of uvlfrom all stress histories
K2 (highway bridges)
two standarddeviatiohs below the mean line
Ovl B The second largest value of uv1 from all stress
K0 Value of ratio uvqB/ffy,A (highway bridges)
Miner'ssummation adjustmentfactor (highway histories (highway bridges)
KF OX, U, Coexistentorthogonal direct stresses
bridges)
UY ' Nominal yield strength
Kn Proportionfactor for occurrencesof vehicles of
r Shear stress coexistent with uxand 0,
a specifiedgross weight (320Kw kN) in any one
4. General guidance
KRC Fatiguestress concentration factor for
re-entrant corners 4.1 Design life. The design lifeis that period in which a
Fatiguestress concentration factor for bridge is requiredto performsafely with an acceptable
KUA probabilitythat it will not requirerepair (see appendix A).
unreinforcedapertures
Kw Ratio of actual :standardgross weights of The standarddesign life for the purposes of this Part of this
vehicles, trains, bogies or axles in a load British Standard should be taken as 120years unless
spectrum otherwise specified.
kq- kr Coefficientsin the simplified assessment
procedure for a railway bridge 4.2 Classification and workmanship. Each strudural steel
L Base length of that portion of the point load detail is dassified in accordam wilh table 17 (see5.1.2). Ru's
influence line which contains the greatest shows the maximumpermitted dass for different types of
ordinate (see figure 12) measured in the struGbd detail The dass denoted in table 17 determines the
direction of travel design of q- N anre in figure 14 that may be safely used with
M. M I Applied bendingmoments the hlghestworkmanship standards specitied in Part 6 for the
m Inverseslope of log Ur/lOg Ncurve detail under oonsideration.
N Number of repetitionsto failure of stress range
ffr
In 5.3.1 is defined the informationto be plovidedto the
Ni.Nz Number of repetitions to failure of stress ranges fabricator, to ensure that the appropriate quality standards for
or,, Vr2 ...etc., correspondingtonl. n 2 . . etc.. . Part 6 are invoked.
repetitions of applied cycles
nl.nz .. Number of applied repetitions of damagingstress 4.3 Stresses. Stresses should generally be calculated in
etc. ranges Uti. Ur2.. .etc., in a design spectrum accordancewith Pan 1 of this British Standard but clause 6
nc Number of vehicles (in millions per year) of this Part supplements the information given in Part 1.
traversing any lane of a highway bridge 4.4 Methods o f assessment. All methods of assessment
ZC EHectivevalue of ne described in this Part of this British Standardare based on
nR Total number of live load cycles (in millions) for the Palmgren-Minerrule for damagecalculation (see
each load proportion Kw in a railway bridge clause 11).The basic methods given respectively in 8.4
P.Pl Applied axial forces and 9.3for highway and railway bridges may be used at all
Pu Basic static strength of the stud times. The simplified procedures given in 8.2 and 8.3for
z Elastic modulus of section highway bridges and in 9.2 for railway bridgesmay be used
Yf Partialsafety factor for load(the product Yft . when the conditionsstipulated in 8.2.1.8.3.1 and 9.2.1 are.
.
Y h Yt,, see Part 1) satisfied.
YfL Product of Yfl Yf2 . 4.5 Factors influencing fatigue behaviour. The best
Ym Partial safety factor for strength fatigue behaviour of joints is achieved by ensuring that the
A Reciprocalof the antilog of the standard structure i s so detailed that the elements may deform in
deviation of log N
@ BSI 03-1999
. -. ,. . , , - _.. . . .. ..
3 . ....
BS 5400:Part 10: I980 Issue 2,March 1999
intended ways without introducing secondary 5.1.2.4 Class A is genetally inappropriatefor bidge vmk and
defmtionsandmessesdue to local restraints. Stresses the spedd inspection standads releMnttodassesBand C .. ..- .
-
may also bemduced, and hen* fatigue life increased, by caMlot~beacMevedin~v#nityofweldshkidge
increasedthickness of parent metal or weld metal. w o r k ( F O r t h e s e a n d 0 t h e r ~nlatshwldbeused
Thebestjointperfonnanceisachievedbyavoidingjoint onlvhspedal- *pisspecified see the footnote to
eaenbidtyandwktsnearfmedgesandbyolhercontrots table 17.)
OV~the~iOftbjdnts.PerfwmanceiSadverselyaffected
bY sdsbessathdes,openingsandleemlnt & 1 2 5 T h 0 oftable17afevalidforthequalities
~
comen.GuidaKE in these aspectF is given in table 17and ofsaeelpoduetsandwe#swhidr~the~of
Pad8.accept~atherwlsenoted.ForcertaIndetaaSthe
I appendix H.The effect of residualstmsses is taken into
accuunththedassificationtables. maximum~dassdependsonacceptanceaitefia
giveninPart8.
5. Classification of details 6.2 Unclassified details
6.1. Classiftcation 6.21 Genere/. Details not fulfy coveredintible 17 should
6.1.1 Generel be treatedas class G, or class W for load canying weld
5.1.1.1 For the purposeof fatigue assessment, each part of metal, unlessa superior resistanceto fatigue is proved by
a constructionaldetail subject to fluctuating stress should, special tests. Such testsshould be sufficientty extensive to
where posu'ble, havea particular class designated in allowthedesign u,-Ncuwe to be determinedin the
accordance with the criteria given intable 17. Otherwise manner used for the standardclasses (see appendix A).
the detail may be dealt with in accordance with 6.2. 6.2.2 Post- welding tnatments. Where the
I 5.1.1.2lhe classification of each part of a detail depends classification of table 17 does not give adequatefatigue
1 upon the following : resistance, the performance ot weld details may be
(a) the direction of thefluctuatingstressrelativetothe improved by post-welding treatments such as conuolled
detail; machining. grinding or peening. When this is requiredthe
j NOTE. Propagationof cracks takes place in a direction
perpendicularto the directionof stress.
detail should be classified by tests as given in6.2.1.
5.3 Workmanship and inspection
I
(b) the location of possiblecrack initiation at the detail ; 5.3.1 G m d . Were the dassificaton of a detaii is dependent
(c) the geometricalarrangementand proportions of the uponparticuhrman~ngor~requirements,
detail; which are notgeneraDyspecifiedin Pad6 ofthii
I (d) the methods and standardsofmanufadmand British Standard, the necessary standardsof workmanship and
inspectron inspedionshould be in&cated on the relevant drawings.
5.1.1.3 Inwelded details there are several locations at which All areas of the &re w b m welded details dassiiied as
potential fatigue cracks may initiate; these are as follows : dass F or higherare n e k w y should be shown on the
(a) in the parent metal of either part joined adjacent to: drawingstogetherwith the minimum required dass and an
(1) theendoftheweld, amMind~thedirectionofstressflucbation(seefigurel),
For inspecb'on purposesthis information should be
(2) a weld toe,
incorporatedonto the fabricatds shop insbuctions.
(3) a change of direction of the weld,
(b) in the throat of the weld. Note thata joint may have m t h a n one dass requirement if it
experiencessignificant stress fluduations in hvo or more
In the case of membersor elementsconnected at their ends
by fillet welds or partial penetration butt welds and flanges diredions.
with shear connectors, the crack initiation may occur either NOIEThs level of manufaduringqwlity can affectthefa!igue life of all
in the parent metalsor inthe weld throat: both possibilities studuradetadetashThemanufachrrirgguatitydetermkwsuledegreel0
should be checked by taking into account the appropriate wtliidii-thatmayaas-9qbt3-
during the~hefikicationpl#rrss.SuchdtocontinuWescan a c t a s f a t i ~ ~
classificationandstressrenge. Forotherdetails, the ~~mayreduarthefafigrsSAsb8nunacoepQble)ewlforthe
classifications given intable 17 cover crack initiation at any detaaunderawrddeafknbetaib~ahighparmitteddatsaremont
possiblelocation in the detail. Notes on the potentialmodes seriously~bysochdix#lthrdbesbecausedtherasbidions
of failure foreach detail are given in appendix H. aLeadyptasdbytable17un~rakersinherenthlheformofthe
detai ibey.
5.1.2 Ci8Ssificationo f details in tebie 17
InordertodertermhewhkhleveldquaWyandinspectionisraquiredIn
5.1.2.1 Table 17 is divided into three parts which acmrdana,with Pate. the mininumreqULsd dass has to bedefived.
correspondto the three basic types into which details may Ha dau higherthanF2 is requLedthishastobespecified on the
be classified. These are as follows : drawinqs.o(hemketherequiredfatigueGbmaynotbeachieved.Ifa
(a) type 1, non-welded details, table 17 (a) ; dass higher than F2 k specif&, but not requid, an unearmmical
faMcathwouldresuL
(b) type 2, welded details on surface, table 17 (b) ;
(c) type 3,welded details at end connections of
members. table 17 (c).
5.1.2.2 Eachclassifieddetail is illustrated and given a type
number. Table 17 also gives varlous associated criteria and
the diagrams illustrate the geometricalfeatures and
potential crack locations which determine the class of each
detail end are intended to assist with initial selection of the
appropriatetype number. (For important features that
changesignificantly from one type to another see the
cccc
footnote totable 17.) FmtE WE
5.1.2.3A detail should only be designated a particular
classification ifit complies in every respect with the mum 1. Methodd Indicatingmtnbnum class requlllementp on
tabulated criteria appropriateto its type number. dram'ngs
4 (P BSI 03-1999
... .- . . . . . . . '' ~ _ _ _.. -_
.............. ...
h u e 1, March 1999 BS5400:Pat-t 10: 1980
....-
. . 2.
5.3.2 Detrimental effects. The following occurrences

can result ina detail exhibiting a lower performancethan its
classificationwould indicate :
(a) weld spatter;
(b) accidental arc strikes:
(c) unauthorized attachments;
(d) conosion pitting.
5.4 Steel decks. The classificationsgiven in table 17
should not be applied to welded joints in orthotropic steel
decks of highway bridges: complex stress patterns usually
occur insuch situations and specialistadvice should be
sought for identifying the stress rangeand joint
claqification.
6. Stress calculations
8.1 General
6.1.1 Stress range for welded det811s. The stress range
in a plate or element to be used for fatigueassessmentis the
greatest algebraic difference between principal stresses
occurring on principal planes not more than 45'apart in
any onestress cycle.
6.1.2 Stressrange for welds. The stress range in a weld
is the algebraic or vector difference between the greatest
and least vector sum of stresses in any one stress cycle.
6.1.3 Effective stress range for non-weldeddetails
For non-welded details, where the stress range is entirely
in the compression zone, theeffects of fatigue loading may
be ignored.
For non-welded details subject to stress reversals, the stress
range should be determined as in 6.1 .lThe. effective stress
range to be used in the fatigue assessment should be
obtained by adding 60% of the range from zero stress to
maximumcompressive stress to that part of the range from
zero stress to maximum tensile stress.
6.1.4 Calculation of stresses
6.1.4.1 Stresses should be calculated in accordance with
. Part 1 of this British Standard using elastic theoryand
I
\ . .'.
.
taking account of all axial, bending and shearing stresses

occurring under the design loadings given in clause 7. N O
redistribution of loads or stresses, such as is allowed for
checking static strength at ultimate limit state or for plastic
design procedures, should be made. For stresses in
Q BSI 03-1999 4a
... . .. ..- .
-
_- .. . .
Issue 2, March 1999 BS 5400: Part 10 :1980
composite beamsthemodulusof elasticity of the concrete table 17, the sbess should be based on thenet section. Where
should be derived from the short termstress/strain indicated in table 17, stress concentrationSshould be taken into
relationship (seePart4).~hestressessocalculatedshould account either by special analysis a by the faders gimn in
be usedwith a materialfactor Ym = 1. figure 22 (seealso H.12).
6.1.4.2 The bendingstresses invarious partsof a steel
'orthotropic bridgedeck may be significandyreduced as the 6.2.2 Shear stress may be neglectedwhere it is
result of compositeaction with the road surfacing. numerically less than 15 % of a coexistent direct stress,
However, thiseffectshould ontybeteken intoaccount On 6.2.3The peak and trough values of principal stress
the evidence of special tests or specialist advice. should be those on principal planes which are not more
6.1.5 Effects to be included. Where appropriate, the than 45" apart. This will be achieved if either
effects of the following should be included in stress (a) a d is at least double the correspondingshear
calculations: stress g at both peak and trough. or
(a) shear lag, restrainedt,orsion and distortion, transverse (b) the signs of Q- and r both reverse or both remain
stresses and flange curvature (see Parts3 and 5) ; the same at the peak and the trough,
(b) effectivewidth of steel plates (see Part3) ; where
(c) cracking of concrete in composite elements (see OX.ayand r are the coexistentvalues with
Part 5) ; appropriatesigns of the two orthogonal direct stresses
and the shear stressesat the point under consideration.
(d) stressesin triangulated skeletal structures due to load
In'either (a) or (b), providedthat a 2 3 q 2 a tboth peakand'
applicationsaway from joints, member eccentricities at
trough, the requiredstressrange will be the algebraic
joints and rigldityof joints (see Part3).

difference betweenthe numerically greater peak principal
6.1.6 Effects to beignored. The effects of the following stress and the numericallygreater trough principal stress.
need not be included in stress calculations :
(a) residualstresses ; 6.3 Stress in weld throats other than thase attaching shear
(b) eccentricities necessarily arising in a standard detail ;connectors.The reference stress for fatigue of a weld throat
(c) stress concentrations. except as required by table 17 : should be the vedor sum ofthe shear sbeses in the weld
(d) plate buckling. metal based on an effective throat dimension as defined in
Part 3, and on the assumptionthat none of the load is carried in
6.2 Stress in parent m e t a l bearing between parent meta~s.~h~ is illustrated in~ u r ea. I
6.2.1 The reference stress for fatigue assessment should be When calculatingthe sb-ess range, the vector diffemmof the
the prindpalstress in the parent metal adjacent to the potential greatest and the least vedor sum slress may be used instead
I cradc location. as shown in figure 2a. Unless otherwise noted in of the algebraic difference.
7- - -'?Welded attachment
Design s t r e s s =(a+%z)
P-P'
Potent ial crock Stress

location distribution
I Figum Za. Referencestress In parent metal
IPNe +M 1
IP,e+M)
QN == 6++
QN
A 5++ -
-
qj
4 Vector sum stress
1 Flgum 2b. Referernestress in weld throat
Q BSI 03-1999 5
. .- .. . __ __
. ....
6.4 Stresses in welds attaching shear connectors . .....
7. Loadings for f a t i g u e assessment
6.4.1 General. Forshear connectors in accordancewith 7.1 Design loadings. Highway and railway design
the dimensionalrecommendationsof Part 5. the design loadings appropriatefor bridges in the UK are given in 7.2
stresses for fatigue in the weld metal should be calculated and 7.3respectiveb.
in accordancewith 6.4.2 and 6.4.3. Where thedimensions The load factors YfLand Yf, should be taken as equalling 1.O
of the shear connectorsand/or the concrete haunchesare (see Part2).
not in accordance with Part 5. the fatigue strength should be
determined in accordancewith appendix G of this Part. 7.2 Highway loading
6.4.2 Studconnectors The stresses in the weld metal 7.2.1 General. I n determiningthe maximum range of
attaching stud shear connectorsshould be calculated from fluctuating stress, generally, only the verticaleffects of
the following expression : vehicular live load asgiven in clause 7 should be
considered, modified where appropriateto allow for impact
stress in weld = as given in7.2.4. Inwelded membersthe dead load stress
longitudinal shear load on stud need not be considered. In unweldedmembersthe dead
x 425 N/m+ load stress will haveto be consideredin determiningthe
appropriatenominalstaticstrength(fromPart 5)
effective stress range when compression stresses occur
6.4.3 Channelandbar connectors (see 6.1.3).
I 6.4.3.1 The stresses inthe weld metal attaching channel Centrifugal effects need only be consideredfor
and bar shear connectors should be calculated from the substructures(see 7.2.5).
effectivethroat area of weld, transversetothe shear flow, 7.2.2 Standadloading
when the concreteis of normaldensity and from 0.85 x
throat area when lightweight concreteis used. For the’ 7.2.2.1 Stund8fdloadspectrum.The standardload spectrum
purposesof this clausethe throat area should be based on a should beasshownin table11 whichgivesthe weight
weld leg length which is the least of the dimensions intensities and relativefrequencies of commercialtraffic on
typical trunk roads in the UK.The minimumweight taken
t
tabulated below.
fora commercial vehicle is30 kN. Allvehiclesless than
Channel connector Bar conneator 30 kN are ignored when consideringfatigue.
Actual leg length 0.6 x actual leg length 7.2.2.2 Standardfatigue vehicle. The standard fatigue
Thicknessof channelweb +
0.125 x (height breadth vehicle is a device usedtotepresent the effects of the
of bar) standard load spectrum;for highway bridgesthisis a single
Half the thicknessof half the thickness of vehicle with a weight of 320kN. It consistsof four standard
beam flange beam flange axles with the dimensionsasshown in figures3 and 4.
NOTE. See appendix Cforthederivationofthe standardfatigue
vehicle.
6.4.3.2 It may assist calculation to note that in normal
density concrete. where the thickness of the beam flange is 7.2.2.3 Number ofvehicles. The numbers of commercial
at least twice the actual weld leg length and the weld vehicles that are assumed to travel along eachlaneofa bridge
dimensionscomply with Part 5, the effective weld areas are : per year should be taken from table 1. If for any reason
vehicle numbersother than these are adopted, suitable
50 x 40 bar connectors x 200 mm long, 1697 mmz
adjustmentsmay be made to the fatigue analysisin
25 x 25 bar connectors x 200 mm long, 1018 mm’
accordance with 8.2.3 or8.3.2.1 (e).
127 and 102 channel connectors .. 150 mm long,
1272 mm 2 7.2.3 Applicetion of loading
76 channel connectors x 150 mm long, 1081 mm2 7.2.3.1 Demarcationoflanes. For the purposes of this Part
6.5 Axial stress in bolts. The design stress for fatigue in of this British Standard the lanes should be the actual traffic
bolts complying with the requirements of BS 4395 and lanes marked on the carriageway. They should be
bolts to dimensionaltolerances complying with the designated in accordance with figure Sand theloading
requirements of 8s 3692 should be calculated from the should be applied to theslow and theadjacent lanesonly.
following expression : Where a crawler lane is providedit should be treated as an
,C additional slow lane.
stress in bolt = - X oB 7.2.3.2 Path ofvehicles. The mean centre line of travel of
@U
where allvehicles in any lane should be along a path parallelto.
and within 300 mm of, the centre line of the lane as shown
F = 1.7kN/rnm’forthreadsof nominaldiameterupto25mm in figure 6. The transverse position of the centre line of the
or vehicle should be selected so as to cause the maximum
F = 2.1 kN/mm’forthreadsof nominaldiameterover25mm stress range in the detail being considered. In some
4 8 is the stress range on the core area of the bolt instances it may be found that the use of multiple paths
determinedon the basisof the minor diameter ’ results in significantly lesscalculateddamageand guidance
uuis the nominalultimate tensilestrength of the bolt on this is given in C.1.4.
material in kN/mm2
7.2.3.3 Standardloading. The passage of one standard
When subjected to fluctuating stresses, black bolts fatigue vehicle along the entire length of one lane should be
complying with the requirementsof BS 4190 may only be taken as one loading event.
used ifthey are faced under the head and turned on shank in
accordancewith the requirementsof BS 4190. 7.2.3.4 Non-standardloadspectrum. If a load spectrum is
used, which differs in any way from the standard load
spectrum. the passage of each vehicle forming the load
spectrum should be consideredto provide a sqparate
loading event.
6
lssue 2, March 1999 BS5400:Part 10: 1980
H.2.2 Generelcommenrs.In welded construction, materialwhich has previously been fully heat-treated, but .
fatigue failurewillrarely occur in a region of unwelded such componentsshould be subject to special test and
materialsincethe fatiguestrength of the welded joints will inspection procedures.
usually be much lower. For the use of black boltscomplying with the
requirementsof BS4190and subjectedto fluctuating
tensile loads, see 6.6.
Where bolts or screwed rodsare pre-tensionedtoa value in
excess of an applied externalload, stress fluctuations will be
H.2.3 Comments Onp8rtiCUl8rdet8i/~peS governed by the elasticity of the pre-compressedelements.
The increase in tension will rarely exceed 10% of an
Type 1.3.All visible signs of drag lines should be removed externalload applied concentrically with the bolt axis, but
from the flame cut edge by grinding or machining. where the load is eccentric, a further inueasewill result
Types 1.3and 1.4.The presence of an aperture or re-entrant from prying action.
corner impliesthe existenceof a stress concentration and the H.3 Type 2 classiflcatlons, welded details on surface
designstress should be the stress on the net section of member. Seetable 17(b).
multiplied by the relevantstress concentration factor (see
figure22). H.3.1 Notes on potenrielmodes of f8ilufe. Seefigure 23.
Type 1.4. The controlled flame cutting procedureshould When the weld is essentiallyparallelto thedirection of
ensure that the resulting surface hardness is not suff icient to stressing, fatigue crackswill normally initiate at the weld
cause cracking. ends, but when the weld is transverse, cracking will initiate
at the weld toes. Ineithercase the cracks will then
Type 1.5.Thistype may be deemed to include bolt holesfor propagateinto thestressed element For attachments
attaching light bracing members where there is negligible connected by single welds, cracks in parent metal may also
transferenceof stress from the main member in the direction initiate from the weld root Cracks in stressed weld metal
Of or. will initiate from theweldroot (seetype3.11).Awayfrom
Type 1.6.This covers connectionsdesigned in accordance weld ends, fatiguecracks normallyinitiate at stop-start
with Part3forslipresistanceatthe ultimate limitstateand positions, or if these are not present a t weld surface ripples.
where secondary out-of-plane bending of the joint is With the weld reinforcement dressed flush, failure tends to
restrainedor does not occur (i.e. double-covered symmetric beassociatedwith weld defects.
joints). Failure initiates by fretting in front of the hole.
H.3.2 Gener8lcomments.
Type 1.12. Thisclassification applies tofailureat theroot of H.3.2.1 Edgedisfance. (See figure 24.) No edgedistance
the thread in normal commercialquality threaded criterion exists for continuous or regularly intermittent welds
components. Attention should be paid to the details of head awayfromtheendsofanattachment (seetypes2.1 to2.5).
fillets, waistedshanksand thread run-out in components, However.'a criterion exists (types 2.6 to 2.1 0) to limit the
not covered by an appropriate British Standard, to ensure possibility of local stress concentrationsoccurring at
that they havesatisfactoryfatigue resistance./\ higher unwelded corners as a result of, for example, undercut, weld
fatigue resistance can be obtained with a rolled thread on
Long attachment Short at to ch m ent s
A A
f \
Weld failure crocks (type.3.111
' I
4
I I
+
Crack types
Y
Crack types 2.9 or 2.10
2.6.2.7 o r 28 ( o r 2.11 i f on edge)
[or 2.11 a t edge)
NOTE. For classificationpurposes, an 'attachment' should be
taken as the adjacent structuralelement connectedby welding to
the stressedelement under consideration. Apart from the
particular dimensional requirementsgivenfor each type in table
17(b). therelativesizeofthe'stressed element'and the
'attachment'is not a criterion.
Figure 23. Failure modes at weld enda
Avoid or grind out to a smooth profile,
any undercut to thes
. I
(a)
(to Ge of weld)
Figure 24. Edge distance
0 BSI 03-1999 i,. 43

.. . .
. __ .. - .. .- .
... ....
- .-
BS 5400:Part 10: 1980
spatter and excessive leg size at stop-start positionsor cause local buckling (see Pan 3). If intermediategaps
accidentalownweaveInmanual welding. Although this longer than 2.5hare requiredtheclassshould be reducedto
criterioncanbespecifiedonlyforthe'width'directionofan F. This type also includes tack welds to the edgesof
element it isequally Imponant to ensure that noaccidental longitudinal backing strips irrespectiveof spacing, provided
u n d e r m h goccurs on the unwelded corners of, for that the weldscomply inall respects with the workmanship
example,~platesorboxmembers(seefigure24 (b) requirements for permanent welds and that any undercut on
and (c)).Where it does occur, it should subsequentlybe the backingstrip isground smooth. The effects of tack welds
groundouttoasmooth profile. which are subsequentlyfully ground out or incorporated
into the butt weld by fusion. need not be considered.
Part 5recommendsthe provision of a minimum edge
distanceof 25 mm forshear connectors, hencethe Type 2.7. The classification may be deemed to include stress
criteriongiven in this part willautomatically be met. concentrationsarising from normaleccentricitiesin the
thickness direction.
H .3.2.2Attachment ofpermanent backing strips If a
permanentbackingstrip is used in making longitudinal butt This type includes parent metaladjacent to the ends of
welded joints itshould be continuous or made continuous flange cover platesregardlessof theshape of the ends.
by welding. These welds and those attaching the backing Types2.7and2.8. Wherea narrow attachment is
strip shouldalso comply with the relevant class transferringthe entire load out of a wide member, as in the
requirements. The classification wilt reduce to E or F (type case of a welded lap type connection between, for example,
3.3 or 3.4) at any butt welds in the backing strip or class E a cross brace and a gusset, the stress inthe gusset at the end
at any permanent tack weld (see H.3.3. type 2.4). It should of thecross brace willvary substantiallyacross the section.
be notedthat transverse butt welds on backingstrips may be For assessingthe stress in the gusset the effective width
downgraded by tack welds close to their ends (see H.4.3, should be taken asshown in figure 25.
type 3.4).
Gusset
i
H.3.2.3Stress concentrations. These are Increased, and
hencethe fatigue strength is reduced, where :
(a) the weld ends or toes are on, or near, an unwelded
corner of theelement (see H.3.2.1) ;
y ,Slope=1:2
(b) the attachmentis 'long' in the direction of stressing,

and asa result, transfer of a part of the load in the element
to and from the attachment will occur through welds
adjacent to its ends ;
(c) suchload transfer is through joints whicharenot
symmetrical about both axes of cross section of the
stressed element.
H.3.2.4 Weldforms. Full or partial penetrationbutt welded
jointsofTform (such as would connect attachmentstothe NOTE. For failure in the cross braceat X the cross bracei s the
'
surfaceof astressedelement) should be completedby fillet 'rnember'and the gusset is the 'attachment'.
weldsof leg length at least equal to 25 %of thethickness of
Figure 25. Effective width for wide lap connections
the attachment.Thefilletsexclude the possibilityof an
increaseinstressconcentrationarising at an acutere-entrant Type2.1O.Thisapplieswhereanyapplied shear stressrange
angle between the element surface and the toe of the weld. is (numerically) greaterthan 50%of a co-existent applied
and thus. in considering the effects on the stressed element. direct stress range.
it is immaterial whetherthe attachment is fillet or butt
Experimentalevidence indicatesthat where significant
welded to the surface, since a similar toe profile results in
both cases. shear stress co-exists with direct stress, the use of
principal stress values may be conservativeand
H.3.2.5 Tack welds. Tack welds, unless carefully ground out accordingly the classificationis upgraded.
or buried in a subsequent run, will provide potential crack
Type 2.11.This t;iri'e appliesregardless of the shape of the
locationssimilar to any other weld end. Their use in the
end of the attachment. Inall cases, care should be taken to
fabrication process should be strictly controlled.
avoid undercut on element cornersor to grind it out to a
NOTE. Apartfromthewidthtransversetoa, neithertheshapeof smooth profile should it occur. In particular, weld returns
the end of an attachment nor the orientationor continuity of the across a corner should be avoidedand the use of cover
weld at itsendaffects the class.
plates widerthan the flange, to which they are attached, is
H.3.3 Comments onparticular d e t a i l types not recommended.
Type2. Finish machining should be in the direction of H.4Type3 classifications, welded detailsat end
cr. The significance of defects should be determinedwith connections of member. See table 17(c).
the aid of specialistadvice and/or by the use of a fracture H.4.1 Notes onpotentialmodesof failure. (See figure
mechaniuanalysis. The N.D.T. technique should be 26.) With theends of butt welds machined flush with the
selected with a view to ensuring the detection of such plate edges, or as otherwise given below. fatigue cracks in
significantdefects. This type is only recommended for use in the as-welded condition normally initiate a t the weld toe
bridgeworksin exceptional circumstances. and propagateinto the parent metal, so that the fatigue
Type2.2.Accidental stop-starts are not uncommon in strength depends largety upon the toe profile of the weld. If
automatic processes. Repair to the standard of a C thereinforcementof a butt weld isdressedflush, failurecan
classificationshould be the subject of specialistadvice and occur in the weld material if minor weld defects are exposed.
inspectionandshould not be undertaken in bridgeworks. e.g. surface porosity in the dressingarea (see H.4.3, type
Type 2.4.The limiting gap ratio m/h applies even though 3.3).
adjacentwelds may be on opposite sides of a narrow In the case of butt welds made on a permanent backing,
attachment (as in the case of a longitudinal stiffener with
staggeredfillet welds). Long gaps between intermittent
fatiguecracksinitiateat the weld metal-strip junction and
then propagateinto the weld metal.
'd
fillet welds are not recommendedas they increasethe risk In fillet or partial penetrationbutt welds, fatigue cracks in
of corrosionand, in the case of compression members, may weld metal will normally initiate from the weld root.
44
~ . _ _ _ _ _ _ _ -- - -
....
__
.... ._...
.-
...
... . _
.
0
9
s'
E
- .....
. .
-
..
...
..
fri tr
-
3
3
is blank
Previous page
. . . .. . .- . ._. ., . . .. -
ii
f
J
i
i
I
L
if!
-f
t-
PB
c
0
I 'I
3 A
--
Previous page
...
t . is blank
.-
-- - _. -- -
Issue 2, ~ a r c h1YYY US5400 :Part10 :1980

Thia British Standard, hwhg kenpnperd under the direction of
the civil Ewineering nd Building Stnnxurea Standerdr Committee.
was publishad under Ih.ruthorlty of tho Exrcut'ba Ward and
coma into e f f m 00 31 January 1980
F*,
Q BSI 03-1999 [',<.*::
.....
..
...
ISEN 0 580 tossf 9
...
The following BSI rdwmces relate to the work on this standard:

Committee referenceCSB/30 Draft for comment 74/13197 DC
. .
. ._. .
.....
.....
....
....
...
Committees responsible for this British Standard

The Civil Engincaring a d Building Structures Standards Committee. Department of the Environment (Transport a n d Road Research
under whose direction this British Standard waa prepared. consins Laboratory)
of representatives from the lollowing Gowrnmmt departments a n d Dapartmant of Transport
acientilic and industrialorganizations: Federation of Civil Engineering Contractors
Association of Consulting Engineers Greater London Council
Association 01 County Councilr Institution of Civil Engineerr
British Constructional Steelwork Association Institutionof Highway Engineers
Brirish Precast Concrete Federation Ltd. lnatitution of Municipal Engineers
British Railways Board Institution of Structural Engineers . .
British Steel Industry London franrport Executive ..
Cement and Concrete Association Ministry of Defence
Concrete Society Limited Sand and Gravel Association l i d .
Constructional Steel Research and Dmlopkent Organization Scottish Development Deparrment
Department of 1.50 Environment (Building.ResearchEstablishment) Welding Institute
Amendments issued since publication

Amd. No. I Date of issue 1 Text affected
9352 March 1999 Indicated by a sideline
....
@
h.. . . . . . . . . . -
- ...... ...

Steel, Concrete and Composite: Bridges

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Steel, Concrete and Composite: Bridges

Uploaded by

Copyright:

Available Formats

BRITISH STANDARD BS 5400 :

Steel, concrete and

NO COPYING WITHOUT BSI PERMISSION EXCEPT AS PERMITTED BY COPYRIGm L A W

Front cover 2 6 Original

I Method of indicating minimum dass

28. Use of continuity plating to reduce

BS 5400: Part 10 :1980 issue 2 March 1999

1. Scope 3. Definitions and symbols

all circumstances. N0TE.A convenient mode of expressinga load spectrum is to

3.1.1 2 Or -N relationship or ar -N curve. The Miner's summation

1.1uy unless otherwisespecified

BS 5400: Part 10: 1980 issue 2,March 1999

5.3.2 Detrimental effects. The following occurrences

Issue 2, March 1999 BS5400: Part 10: 1980

difference between the numericallygreater peak principal

fq- Design stress =(/A+

1 Figure Zb. Reference stress in weld throat

when the concrete is of normaldensity and from 0.85 X

Figuref. Axle arrangement of standard fatiguevehicle

Figure 4. Plan of standard axle

Table 1. Annual flow of commercial vehicles (n, x 1 06)

Slip road Single

*The number of vehicles in each lane of a single carriageway between7.3 rn and 10 rn

wide should be obtained by linear interpolation.

SI. Adj. F F Adj. SI.

Two and three lane dual carriageways

Two lane slip road

SI. ' F SI.

Three lane single carriageway

Two lane single carriageway

NOTE. For two lane dual omit fast lane.

BS5400:Part 10: 1980

All dimensions are in millimetres

I Influence line for static stress

Figure 7. Impact allr,wance at discontitiuities

other classes thisoption will not satisfy the recommen-

more rigorous procedureof 8.4. the simplified procedures

figure8. Valuesof OH f o r d i f f e r e n t r o a d categories

load or stress spectrum.

. . *The lane reference letters A, B etc. should be allocated in

Figure 9. Derivation of ovand ii, for damage calculation

BS 5400 :Part 10 :.1980 I

Figure 11. Miner's summation adjustment factor KF for highway bridges

Figure 12. Typical point load influence line

Spectrum 'as calculated

(d) Obtain the appropriate limiting stress range UT

1597 '. __ .. . . . . .- . __.... . .... . . ... . . - - .-

H a a w traffic Medium traffic - trafflo

Table 5. Values of k4 for railway bridges

42 to 27 27to18 18to12 12t07 7to5 <5

0.89 1.o 1.13 1.27 1.42 1.6

Table 6. Values of k, for railway bridges

1.42 1.27 1.1 25 1.o 1.o 0.89

NOTE. P, is the numerical value of stress due to track 1.

Table 7. Values of k6 for R L loading of

Length, f (m) Values of k.

< 3.0 1.23 1.28 1.35 1.65

alternative to the simplified procedure of 9.2.

times the value of uoof the original class of the detail.

to other numbers of standarddeviations below the mean-

Constant am pl i t u.de Ioa d ing

./-.. Appendix B Appendix C

First occurrence Second occurrence

1597 '. .. . . . . .- . .... . .... . . ... . . - - .-