Professional Documents
Culture Documents
Steel, Concrete and Composite: Bridges
Steel, Concrete and Composite: Bridges
Part 10 :1980
Incolparting
Amendment No.I
-
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
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
.. . .. . . , . ~ .... . - .. .. . . - __ .. .- . .
. . . . . . . . . . . . . . . . . . . . . ~
. . .- ....
.......
2 . 8ESIO3-1999
P
Issue 2, March 1999 BS5400:Part10:1980
8 BSI 3
- 03-1999
__ .- .-
... ... ._
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
6. Stress calculations
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
__
._ .
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
I
I f ,Welded attachment
-
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
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
1: vehicle
twin tyre
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
4
900 m m
0. tyre +tyre
f 586 7 .. .
- .. .
. -... . .. . . .. . . .. .. ... . .
. .. ... .. -- __ . . -
BS5400:Pan10:1980
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
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. SI.
Lone marking
""i\ - ----
Sl.
I
I
Adj.
I
I
F
I
Adjacent lane Fast lane
.. . .., Hard shoulder or Slow lane
. .._. hard strip
1588 9
_-
... ..- . ..
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
stress range)
fI
!
wheels a
I
vehicle
t -
wheels
Mean paths
o f wheels
Discontinuity
Adjusted stress
- - ,
I
t I
I I
I I
I
I
Impact
allowance
I
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 , . .. - .. ..
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
1591 __ ._- 12 .
BS 5400 :Part 10 : 1980
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
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).
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
...
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
1394
- ...............
..... -. 15
... . . . -
- -- -- - . -- , . - ... m" . . . . .
BS5400:Pan10:1980
..
.... ...
. .
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
..
I595
_... ... . . . .. . . . - .. . . . . .... . . .-... . --
. 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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
.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.
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).
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
I l l 1 1 I 1 I I I
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
I c
c
Number of r e p e t i t i o n s (n1
..
Figure 13. Simplification of a spectrum
in 6.1.3. where
m is the inverse slope of the -Ncurveappropriate to
the detail class and is obtained from table 8.
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
Detail class 0 C 8 s
E
F
FZ
c
W
~~~
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
....
..
.... .....
...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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
, n nu:
-*-=-K2 (g)m
,:7 when Qr 3 uo
class should be between
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
-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
--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
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
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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.
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
Detail class fm -
D t o G. F2 and W
3.01
C
(m - 3.5)
B
( m = 4.0)
S
(m - 8.0)
*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
1607
._.. . . . .. . .
BS5400:Part10:1980
..
..
1 25 50
. ..
. .
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
1608
... . . . . 11 . 29 ~. __ . . . .... . _.._
-.. _
.
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
17.8
32.0
V
- 8.0
v -5.5
w
-3.5
-2.5
10.2
6.3
T
-2.8 -1:9 -1 :5 -0I8
2nd carriageway, adjacent lane 2nd carriageway, slow lane
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
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
(<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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
Lane A
- 12.9 -12.9
Lane B
12.9 12.9
1 - 12.9 -1 2.9
I
I Lane AIB
_.
..... .,... ..
' _...
161 1
- - 32
BS 5400:Pan 10: 1980
Lane A 54
2 cycles per
loading event
0
Lane B "Pt .
0
2 cycles per
loading event
Combined A/B s4
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
l o a d i n g event
Lane I Cycle
number
I
k;;; , Trough
1 Range
0,
I dim+
Class G
I
nc:
I Damage
d*ia
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
*See appendix B.
tSee figure 10.
S e e figure 9.
I 1612
. .33
_
!
BS 5400 : Part 10 : 1980
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
2.7 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
-
2x13 2x11
I .
214.9.5
-1
2x11 2x13
I '
T'
1
35 .
. __ -. ..
,-.,
.. . ..
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
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.
..... .
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
3 total load = 280 t
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
6 total load = 209 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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
L 6 1.8
0;3 5'
-3
6 < L < 15.6
/ oc2
/' /
L >15.6
1617
- . . 38 __. .. .......
x
.( , 4
\
\
........
...........
.. .. .. .. ........
N
- b’
.. .. .. .. ............ . . . .
..........
..........
........................
................
................................
................................
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
................................
................
................
,-
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
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.
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
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
1620
. . -. . 41
BS 5400 :'Part 10 : 1980
..
I
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
r
'6
Length of
s t raight22r
1
Stress
fluctuation
0;
.....
.c....:.... .
, .1
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
.......,..
.... .:,\.
,
..i.:J
(b)
Figure 24. Edge distance
(B BSI 03-1999 43
I - -z- -. .. .. . .
- .. .
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.
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
+t+
/2 min.
3.1 1 3.1 1
3.3
-
3.9
I J
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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 _.
+
..,. . . . .- . ..._. _.
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
0;
Figure 29. Cruciform junction between flange plates
Class
cut out
Class I
Class
A
1627
-. . . - 48 -. . __ . - _. ... .
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
I .
I
i
.. ...
..
.
.. .
--
-.
r
- .-
.
,
. . .._
.-
i!
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
i
.
is blank
Previous page
-
.
.. ....
...
’!
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
.. . .. .... -
~~
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
British Standards Institution . 2 Park Street London W l A 2BS - Telephone 01-629 9000 Telex 266933
Amendment No. 1
I
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
--0
V I -
n-=
NO COPYING WITHOUT BSI PERMISSION EXCEPT AS P E R M I m D BY COPYRIGHT LAW =
*.
.<
- - _-_ -
~... . . . ... . ...
Summary of pages I
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
. . . ..... -
........ . . . . . .
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
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
. . ... ... .- . . - .- . ._ .. . _.
. . -. ... . . ... . . .. .. . .. -.
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
...
........
.. .....
.......
-:..:..
2 0 BSI 03-1999
............. .- . . . . . . ................ . ._
.......... -.
-
issue 2 March 1999 BS 5400 :Part 10 : 1980
@ 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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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.
6. Stress calculations
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
\ . .'.
.
Q BSI 03-1999 4a
... . .. ..- .
-
_- .. . .
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
7- - -'?Welded attachment
Design s t r e s s =(a+%z)
P-P'
IPNe +M 1
IP,e+M)
QN == 6++
QN
A 5++ -
-
qj
4 Vector sum stress
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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 \
' 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
- .-
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.
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
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
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
~ . _ _ _ _ _ _ _ -- - -
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
....
__
.... ._...
.-
...
... . _
.
0
9
s'
E
- .....
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
. .
-
..
...
..
fri tr
-
3
3
is blank
Previous page
. . . .. . .- . ._. ., . . .. -
ii
f
J
i
i
I
L
if!
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
-f
t-
PB
c
0
I 'I
3 A
--
Previous page
...
t . is blank
.-
Licensed copy:Scott Wilson Holdings, 22/06/2006, Uncontrolled Copy, © BSI
-- - _. -- -
. ._. .
.....
.....
....
....
...
....
@
h.. . . . . . . . . . -
- ...... ...