You are on page 1of 57

BRITISH STANDARD

Steel, concrete and
composite bridges -

Licensed copy:Parson Brinkerhoff, 25/02/2015, Uncontrolled Copy, © BSI

Part 10: Code of practice for fatigue

ICS 93.040

NO COPYING WITHOUT BSI PERMISSION EXCEPT AS PERMITTED BY COPYRIGIW LAW

BS 5400 :

Part 10 : 1980

Issue 1, March 1999

BS 5400 : Part 10 : 1980

Summary of pages
The following table identifies the current issue of each page. lssue 1 indicates that a page has been introduced
for the first time by amendment. Subsequent issue numbers indicate an updated page. Vertical sidelining on
replacement pages indicates the most recent changes (amendment, addition, deletion).

Front cover
h i d e front cover
a
b
C

d
1

Licensed copy:Parson Brinkerhoff, 25/02/2015, Uncontrolled Copy, © BSI

2
3
4
4a
4b
5

Issue

Page

Issue

2
blank
1
blank
blank
1
2
2
2
2
1
blank
2

6

Original
original
2
original

7 to 42
43
44
45 to da
49
50
51
52
53
54
Inside back cover
Back cover

Original

2
blank
2
blank
2
blank
original
2

a

BS 5400 : Part 10 : 1 980

Issue 1, March 1999

Contents

Licensed copy:Parson Brinkerhoff, 25/02/2015, Uncontrolled Copy, © BSI

I

Page
1
Foreword
Cooperating organizations
Back cover
Recommendations
1.
Scope
2
1.1
General
2
1.2
Loading
2
Assessment procedures
1.3
2
1.4
Other sources of fatigue damage
2
Limitations
1.5
2
1.5.1 Steel decks
2
1.5.2 Reinforcement
2
1.5.3 Shear connectors
2
2.
References
2
3.
Definitions and symbols
2
Definitions
3.1
2
3.2
Symbols
3
4.
General guidance
3
Design life
4.1
3
4.2 Classification and workmanship
3
Stresses
4.3
4
4.4
Methods of assessment
4
4.5
Factors influencing fatigue behaviour
3
5.
Classification of details
4
5.1
Classification
4
5.1.1 General
4
5.1.2 Classification of details in table 17
4
5.2
Unclassified details
4
5.2.1 General
4
5.2.2 Post-welding treatments
4
5.3
Workmanship and inspection
4
5.3.1 General
4
5.3.2 Detrimental effects
4
5.4 Steel decks
4
6.
Stress calculations
4
General
6.1
4
6.1.1 Stress range for welded details
4
6.1.2 Stress range for welds
4
6.1.3 Effective stress range for non-welded
details
4
6.1.4 Calculation of stresses
4
6.1.5 Effects to be included
5
6.1.6 Effects to be ignored
5
6.2 Stress in parent metal
5
6.3
Stress in weld throats other than
those attaching shear connectors
5
6.4 Stresses in welds attaching shear
connectors
6
6.4.1 General
6
6.4.2 Stud connectors
6
6.4.3 Channel and bar connectors
6
6.5 Axial stress in bolts
6
7.
Loadings for fatigue assessment
6
7.1
Design loadings
6
7.2 Highway loading
6
7.2.1 General
6
7.2.2 Standard loading
6
7.2.3 Application of loading
6
7.2.4 Allowance for impact
8
7.2.5 Centrifugal forces
8
7.3
Railway loading
8
7.3.1 General
8
7.3.2 Application of loading
8
7.3.3 Standard load spectra
10
8.
Fatigue assessment of highway
bridges
12
8.1
Methods of assessment
12
8.1.1 General
12
8.1.2 Simplified procedures
12
8.2 Assessment without damage
calculation
12
8.2.1 General
12

d

Page

8.2.2 Procedure
8.2.3 Adjustment factors for an, class S
8.3
8.3.1
8.3.2
8.4
8.4.1
8.4.2
8.4.3
8.4.4
9.
9.1
9.1.1
9.1.2
9.2
9.2.1
9.2.2
9.2.3
9.2.4
9.3
9.3.1
9.3.2
9.3.3
9.3.4
9.3.5
10.
11.
11.1
11.2
11.3
11.4
11.5

details only
Damage calculation, single vehicle
method
General
Procedure
Damage calculation, vehicle spectrum
method
General
Design spectrum
Simplification of design spectrum
Calculation of damage
Fatigue assessment of railway bridges
Methods of assessment
General
Simplified procedure
Assessment without damage calculation
General
Procedure
Non-standard design life
Multiple cycles
Damage calculation
General
Design spectrum for standard loading
Design spectrum for non-standard
loading
Simplification of spectrum
Calculation of damage
Fatigue assessment of bridges carrying
highway and railway loading
The Palmgren-Miner rule
General
Design 0 , - N relationship
Treatment of low stress cycles
Procedure
Miner's summation greater than unity

Appendices
A.
Basis of a,-N relationship
B.
Cycle counting by the reservoir method
C.
Derivation of standard highway bridge
fatigue; loading and methods of use
D.
Examples of fatigue assessment of
highway bridges by simplified methods
E.
Derivation of standard railway load
spectra
F.
Examples of stress histories and cycle
counting procedure
G.
Testing of shear connectors
H.
Explanatory notes on detail
classification

12
12
12
12
12
14
14
14
14
14
18
18
18
18
18
18
18
18
20
20
20
20
20
20
20
20
20
20
22
22
22
22
23
25
25
30
34
38
41
41

Tables

1.

2.
3.
4.
5.
6.
7.
8.
9.
10.
11.

12.

Annual flow of commercial vehicles
(nc x 108)
8
Standard load spectra for RU loading
11
Standard load spectra for RL loading
12
Values of k 3 for RU loading of
19
railway bridges
Values of k 4 for railway bridges
19
Values of k 5 for railway bridges
19
Values of k 6 for R L loading of
railway bridges
19
Design 0,-N relationships and constant
amplitude non,propagating stress
22
range values
Mean-line or -N relationships
23
Probability factors
23
Typical commercial vehicle groups
27
Proportional damage from individual
groups of typical commercial vehicles
28
0 BSI 03-1999

Issue 2,March 1999

BS 5400: Part 10: 1980

Page
Typical commercial vehicle gross
weight spectrum
29
14.
Typical commercial vehicle axle
weight spectrum
29
15.
RU loading : annual traffic tonnage for
standard traffic types
34
16.
R L loading : annual traffic tonnage and
composition of standard traffic mix
34
17.
Classification of details
17(a). Non-welded details
49
17(b). Welded details other than at end
51
connections of a member
17(c). Welded details at end connections of
member
53

Licensed copy:Parson Brinkerhoff, 25/02/2015, Uncontrolled Copy, © BSI

13.

-

Figures
Method of indicating minimum dass
requirementson drawings
Reference stress in parent metal
Reference stress in weld throat
Axle arrangement of standard fatigue
3.
vehicle
Plan of standard axle
4.
Designation of lanes for fatigue
5.
purposes
Transverse location of vehicles
6.
Impact allowance at discontinuities
7.
Values of oH for different road
8.
categories
Derivation of ov and &. for damage
9.
calculation
Damage chart for highway bridges
10.
(values of d, 20)
Miner's summation adjustment factor
11.
KF for highway bridges

I ''
I ;;:

12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.

4a
5
5

26.
27.
28.

7
7

29.

9
10
10

30.

13

31.
32.
33.

15

34.

Page
Typical point load influence line
17
Simplification of a spectrum
18
Summary of design or- N curves (mean
minus two standard deviations)
21
Summary of mean-line o r - N curves
23
Typical G-,- N relationship
24
Multiple paths
28
Typical Miner's summation adjustment
curve
29
Trains included in table 2 spectra
35
Trains included in table 3 spectra
37
Typical example of stress concentrations
41
due to geometrical discontinuity
Stress concentration factors
42
Failure modes at weld ends
43
43
Edge distance
Effective width for wide lap
44
connections
45
Type 3 failure modes
Type 3.6 joint
46
Use of continuity plating to reduce
stress concentrations in type 3.7 and
3.8 joints
47
Cruciform junction between flange
47
plates
Example of a 'third' member slotted
through a main member
47
Example of type 3.9 or 3.10 joint
48
Tee junction of two flange plates
48
Alternative method of joining two
flange plates
48
Single fillet corner weld in bending
48

16

17

Foreword

1
1

B S 5400 is a document combining codes of practice to
cover the design and construction of steel, concrete and
composite bridges and specifications for the loads,
materials and workmanship. It comprises the following
Parts and Sections:
General statement
Part 1
Part 2
Specification for loads
Code of practice for design of steel bridges
Part 3
Code of practice for design of concrete bridges
Part 4
Part 5
Code of practice for design of composite
bridges
specification for materials and workmanship,
Part 6
steel
Specification for materials and workmanship,
Part 7
concrete, reinforcement and prestressing
tendons

0 BSI 03-1999

Part 8

Recommendations fot materials and
workmanship, concrete, reinforcement and
prestressing tendons

Part 9

Bridge beatings

bridge bearings
Section 9.2 Specification for materials,
manufacture and installation of
bridgebearings
Part 10 Code of practice for fatigue

1

1

Licensed copy:Parson Brinkerhoff, 25/02/2015, Uncontrolled Copy, © BSI

BS5400:Part10:1980

Issue 2,March 1999

1. s c o p e

3. Definitions and symbols

1 . 1 General. This Part of this British Standard recommends
methods for the fatigue assessment of parts of bridges which
are subject to repeated fluctuations of stress.
1.2 Loading. Standard load spectra are given for both
highway and railway bridges.

3.1 Definitions. For the purposes of this Part of this British
Standard the following definitions apply.
3.1.1 fatigue. The damage, by gradual cracking of a
structural part, caused by repeated applications of a stress
which is insufficient to induce failure by a single application.

1.3 Assessment procedures. The following alternative
methods of fatigue assessment are described for both
highway and railway bridges :
(a) simplified methods that are applicable to partsof
bridges with classified details and which are subjected to
standard loadings;

3.1.2loading event. The approach, passage and
departure of either one train or, for short lengths, a bogie or
axle, over a railway bridge or one vehicle over a highway
bridge.
3.1.3load spectrum. A tabulation showing the relative
frequencies of loading events of different intensities
experienced by the structure.
NOTE. A convenient mode of expressinga load spectrum is to
denote each load intensityas a proportion ( K w ) of a standard load
and the number of occurrences of each load as a proportion (Kn)
of the total number of loading events.

(b) methods using first principles that can be applied in
all circumstances.
1.4 Other sources o f fatigue damage. The following
topics are not specifically covered by this Part of this
British Standard but their effects on the fatigue life of a
structure may need to be considered :
(a) aerodynamically induced oscillations;
(b) fluctuations of stress in parts of a structure immersed
in water, which are due to wave action and/or eddy
induced vibrations ;
(c) reduction of fatigue life in a corrosive atmosphere
(corrosion fatigue).

1.5 Limitations
1 . 5 . 1 Steeldecks. Highway loading is included in this
Part and is applicable to the fatiguedesign of welded
orthotropic steel decks. However, the stress analysis and
classification of details in such a deck is very complex and is
beyond the scope of this Part of this British Standard.
1.5.2 Reinforcement. The fatigue assessment of certain
details associated with reinforcing bars is included in this
Part but interim criteria for unwelded bars are given in
Part 4.
NOTE. These criteria are at present under review and revised
criteria may be issued later as an amendment.
1.5.3 Shearconnectors. The fatigue assessment of shear
connectors between concrete slabs and steel girders acting
compositely in flexure is covered in this Part, but the
assessment of the effects of local wheel loads on shear
connectors between concrete slabs and steel plates is
beyond the scope of this Part of this British Standard.
This effect may, however, be ignored if the concrete slab
alone is designed for the entire local loading.

2. R e f e r e n c e s
The titles of the standards publications referred to in this
standard are listed on the inside back cover.

3.1.4 standardlosdspectrum. The load spectrum that
has been adopted in this Part of this British Standard,
derived from the analysis of actual traffic on typical roads or
rail routes.
3.1.5 stress history. A record showing how the stress at a
point varies during a loading event.
3.1.6 combinedstress history. A stress history resulting
from two consecutive loading events, i.e. a single loading
event in one lane followed by a single loading event in
another lane.
3.1.7 stress cycle (or cycle ofstress). A pattern of
variation of stress at a point which is in the form of two
opposing half-waves, or, if this does not exist, a single
half-wave.
3.1.8stressrange (orrange ofstress)

(U,).

Either

(a) in a plate or element, the greatest algebraic difference
between the principal stresses occurring on principal
planes not more than 45" apart in any one stress cycle : or

(b) in a weld, the algebraic or vector difference between
the greatest and least vector sum of stresses in any one
stress cycle.
3.1.9 stress spectrum. A tabulation of the numbers of
occurrences of a l l the stress ranges of different magnitudes
during a loading event.
3.1.10 design spectrum. A tabulation of the numbers of
occurrences of all the stress ranges caused by all the loading
events in the load spectrum, which is to be used in fatigue
assessment of the structural part.
3.1.1 1 detai/c/ass. A rating given to a detail which
indicates its level of fatigue resistance. It is denoted by the
following :A, B, C, D, E, F. F2,G, S, or W.
NOTE. The maximum permitted dass is the highest recommended
class, that can be achieved with the highest workmanship specified in
Part 6 (seetable 17). The minimum required class to be specified for
fabrication purposes relates to the lowest q N curve in figure 14,
which results in a l i e exceeding the design lie.

-

2

0 BSI 03-1999

etc 'JR max Miner's summation Stress on the core area of a bolt.N cutve in figure 14 that may be safely used with the highest Workmanship standards specified in Part 6 for the detail under consideration.2 Symbols. A cumulative damage summation based on the rule devised by Palmgren and Miner. in a highway bridge) Design stress parameter for bolts Parameter defining the mean line 'Jr -N relationship Parameter defining the ur-N relationship for two standard deviations below the mean line Value of ratio avl B / ~ V ~(highway A bridges) Miner's summation adjustment factor (highway bridges) Proportion factor for occurrences of vehicles of a specified gross weight (320 Kw kN) in any one lane of a highway bridge Fatigue stress concentration factor for re-entrant corners Fatigue stress concentration factor for unreinforced apertures Ratio of actual : standard gross weights of vehicles. The standard design life for the purposes of this Part of this British Standard should be taken as 120 years unless otherwise specified. N 00 UH 3. . . to be taken as 1.1. determined on the basis of the minor diameter Limiting stress range under loading from the standard fatigue vehicle on a highway bridge Stress on net section Constant amplitude non-propagating stress range(oratN= 10') Algebraic value of stress in a stress history Maximum and minimum values of oPfrom all stress histories produced by standard loading Range of stress (stress range) in any one cycle Individual stress ranges (ar) in a design spectrum ((ip max . 3 In 5. flc flC nR P. UN Qo 3..3for hiahwav and railwav.1. Elastic modulus of section Partial safety factor for load(the product Y f . .14 design life. in a design spectrum Number of vehicles (in millions per year) traversing any lane of a highway bridge Effective value of nc Total number of live load cycles (in millions) for each load proportion Kw in a railway bridge Amlied a x i a l forces Basic static strength of the stud .'JR2 . Qy *Y r 4. ..1.4 and 9. Thesymbols in this Part of this British *R .1 are .. etc.Y f-3 .8. QP l.GP min) for a railway bridge Stress ranges (in descending order of magnitude) in a stress history of a railway bridge under unit uniformlydistributed loading Limiting stress range under standard railway loading Nominal ultimate tensile strength. March 1999 - 3.see Part 1) Product of 71.Issue 2. 4. ... etc Standard are as follows. . J'r 3. The Or-Nrelationship adopted in this Part of this British.4 M e t h o d s o f assessment. .3.1 and 9. C r 2 . . This shows the maximum permitted dass for different types of structural detail.2 for railway bridges may be used when the conditions stipulated in 8. A A1 . The basic methods given respectively in 8. repetitions of applied cycles Number of applied repetitions of damaging stress ranges Or. bridaes " mav be used at all times./log Ncurve Number of repetitions to failure of stress range *T OU b V QV etc.1 5 standard design life.3. n 2 . © BSI n di 2 0 KR C KUA Kw ki.. corresponding ton.. ..2).. etc 'JviA uvi B Qx. Stresses should generally be calculated in accordance with Part 1 of this British Standard but clause6 of this Part supplements the information given in Part 1 .1 2 Qr -N relationship or Or -Ncurve.1 is defined the informationto be provided to the fabricator..1 G~ unless otherwise specified Value of Gr under loading from the standard fatigue vehicle (highway bridges) (0. Yf2 Partial safety factor for strength Reciprocal of the antilog of the standard deviation of log N Q BSI 03-1999 max a v l .3 Stresses.1 -13 design ITr-Ncurve. etc. adopted in this Part of this British Standard.5 Factors influencing fatigue behaviour.2.3 for highway bridges and in 9. The class denoted in table 17 determines the design of U.1. The simplified procedures given in 8.1 6 Miner's summation.Standard for design on the basis of 2. etc.max .2and 8. 1.3 % probability of failure.. av2 01 n.GP min) for a highway bridge Values of G" (in descending order of magnitude) in any one stress history for one lane of a highway bridge The largest value of gvlfrom all stress histories (highway bridges) The second largest value of ov. 4.n. G e n e r a l g u i d a n c e 4. The period in which a bridge isrequired to perform safely with an acceptable probability that it will not require repair. U r 2 .1 Design life. Uncontrolled Copy.1. The design life is that period in which a hridge is required to perform safely with an acceptable probability that it will not require repair (see appendix A). from all stress histories (highway bridges) Coexistent orthogonal direct stresses Nominal yield strength Shear stress coexistent with oxand CJ. . trains. Or2 3. All methods of assessment described in this Part of this British Standard are based on the Palmgren-Miner rule for damage calculation (see clause 11). satisfied.ka n L Net area of cross section Effective weld throat area for the particular type of connector Number of standard deviations below the mean line gr-Ncurve Life time damage factor (Miner's summation for 120 million repetitions of a stress range 0. 4. 4. . Each structural steel detail is dassified in accordancewith table 17 (see 5.2 Classification and workmanship.1. 25/02/2015. Yr. P" Z n Yf YfL Ym d . Q r i . The best fatigue behaviour of joints is achieved by ensuring that the structure is so detailed that the elements may deform in their . Number of repetitions to failure of stress ranges O r i .2. 120 years.. The quantitative relationship between Or and N f o r a detail which is derived from test data on a probability basis.P. u p max Qp min j 3. . to ensure that the appropriate quality standards for Part 6 are invoked. bogies or axles in a load spectrum Coefficients in the simplified assessment procedure for a railway bridge Bass length of that portion of the point load influence line which contains the greatest ordinate (see figure 12) measured in the direction of travel Applied bending moments lnverseslope of log *. d Licensed copy:Parson Brinkerhoff.

I . All areas of the structure where welded details classified as dass F or higher are necessary should be shown on the drawings together with the minimum required class and an arrow indicating the direction of stress fluctuation (see figure 1). When this is required the detail should be classified by tests as given in 5. grinding or peening.1. For other details. The effect of residual stresses is taken into account in the classificationtables. unless a superior resistanceto fatigue is proved by special tests.1. Uncontrolled Copy. 25/02/2015.2The classification of each part of a detail depends upon the following : (a) the direction of the fluctuating stress relative to the detail . (2) a weld toe. table 17 (c).) E Fat c ) f ) Fat 5.1 General.3 Workmanship and inspection 5.3. The best joint performance is achieved by avoiding joint eccenbiaty and welds near free edges and by other controls over the quality of the joints. (For important features that change significantly from one type to another see the footnote to table 17. which are not generally specified in Part 6 of this British Standard.2. In the case of members or elements connected at their ends by fillet welds or partial penetration butt welds and flanges with shear connectors. (b) in the throat of the weld. Fat E Figure 1. that may act as stress raisers. Method of indicating minimum class requirements on drawings 4 0 BSI 03-1999 . welded details on surface.I General 5.1.1 General.2. table 17 (b) : (c) type 3. 5.5 The classificationsof table 17 are valid for the qualities of steel products and welds which meet the requirements of Part 6. otherwise the required fatigue life may not be achieved.1 .2. (b) type 2. have a particular class designated in accordance with the criteria given in table 17. Stresses may also be reduced.4 Class A is generally inappropriatefor bridge work and the speaal inspection standards relevant to classes B and C cannot normally be achieved in the vicinrty of welds in bridge work. welded details at end connections of members. If a class higher than F2 is requiredthis has to be speafied on the drawings.2. but not required. In order to dertermine which level of quality and inspection is required in accordance with Part 6. (For these and other classifications that should be used only when speaal workmanship is specified see the footnote to table 17.2.BS 5400: Part 10: 1980 Licensed copy:Parson Brinkerhoff. each part of a constructional detail subject to fluctuating stress should.I .1. Otherwise the detail may be dealt with in accordance with 5. Classification 5. and hence fatigue life increased.1Table 17 isdivided into three parts which correspond to the three basic types into which details may be classified. (1) theendof theweld. the minimum required dass has to be derived. or class W for load carrying weld metal. C l a s s i f i c a t i o n of d e t a i l s 5. except where otherwise noted.2 Classification of detailsin table 17 5. Such tests should be sufficiently extensive to allow the design or-Ncurve to be determined in the manner used for the standard classes (see appendix A).2 Each classified detail is illustrated and given a type number. 5. If a dass higher than F2 is specified. Notes on the potential modes of failure for each detail are given in appendix H.March 1999 intended ways without introducing secondary deformations and stressesdue to local restraints. 5. 5.1. table 17 (a) . Details with a high permitted dass are more seriously affected by such discontinuities because of the restrictions already placed by table 17 on stress raisers inherent in the form of the detail itself.2 Post-welding treatments.2. openings and reentrant corners. Details not fully covered in table 17 should be treated as class G. Table 17 also gives variousassociated criteria and the diagrams illustrate the geometrical features and potential crack locations which determine the class of each detail and are intended to assist with initial selection of the appropriate type number.1 . Such discontinuities can act as fatigue points.3 In welded details there are several locations at which potential fatigue cracks may initiate.1 -2.2 Unclassified details 5. the classifications given in table 17 cover crack initiation at any possible location in the detail. The manufacturingquality determines the degree to which discontinuities.Where the classification of table 17 does not give adequate fatigue resistance. The level of manufacturingquality can affect the fatigue life of all stuctural details. 5.1. (d) the methods and standards of manufacture and inspection. Propagationof cracks takes place in a direction perpendicular to the direction of stress. 5. Where the classification of a detail is dependent upon particular manufacturing or inspection requirements. the necessary standards of workmanship and inspection should be indicated on the relevant drawings.3A detail should only be designated a particular classification if i t complies in every respect with the tabulated criteria appropriate to its type number. NOTE. which may reduce the fatigue life to an unacceptablelevel for the detail under consideration.1. may be introduced during the fabrication process. (3) a change of direction of the weld. © BSI I I Issue 2. the performance of weld details may be improved by post-welding treatments such as controlled machining. 5. Guidance in these aspects is given in table 17 and appendix H.1 . by increased thickness of parent metal or weld metal. 5.2.1 For the purpose of fatigue assessment. where possible.1. (b) the location of possible crack initiation at the detail . For inspection purposes this information should be incorporatedonto the fabricator's shop instructions. NOTE. Performance is adversely affected by concentrationsof stress at holes. (c) the geometrical arrangement and proportions of the detail . Note that a joint may have more than one class requirement if it experiences significant stress fluctuations in two or more directions. 5. These are as follows : (a) type 1.1. an uneconomical fabrication would result. non-welded details. For certain details the maximum permitted class depends on acceptance criteria given in Part 6.) 5. these are as follows : (a) in the parent metal of either part joined adjacent to.2. the crack initiation may occur either in the parent metals or in the weld throat : both possibilities should be checked by taking into account the appropriate classification and stress range.

1 General 6. 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.1. - I 6. 25/02/2015. For non-welded details subject to stress reversals. 6. the effects of fatigue loading may be ignored. (b) accidental arc strikes.l. The classifications given in table 17 should not be applied to welded joints in orthotropic steel decks of highway bridges . should be made. (d) corrosion pitting.4 Steel decks. 6.3. complex stress patterns usually occur in such situations and specialist advice should be sought for identifying the stress range and joint classification. bending and shearing stresses occurring under the design loadingsgiven in clause 7. N O redistribution of loads or stresses. such a s isallowed for checking static strength at ultimate limit state or for plastic design procedures.1 Stress range for welded details.1. Stress calculations 6. Uncontrolled Copy.4.1 Stresses should be calculated in accordance with Part 1 of this British Standard using elastic theory and taking account of a l l axial.3 Effective stress range for non.1. range to be used in the fatigue assessment should be obtained by adding 60 % of the range from zero stress to maximum compressive stress to that part of the range from zero stress to maximum tensile stress.welded details For non-welded details. Licensed copy:Parson Brinkerhoff.1. where the stress range is entirely in the compression zone. © BSI 5. The following occurrences can result in a detail exhibiting a lower performance than its classification would indicate : (a) weld spatter.2 Detrimenta/ effects. The stress range in a plate or element to be used for fatigue assessment is the greatest algebraic difference between principal stresses occurring on principal planes not more than 45" apart in any one stress cycle. For stresses in 0 BSI 03-1999 4a . 6.4 Calculation of stresses 6.1. the stress The effective stress range should be determined as in 6. (c) unauthorized attachments.2 Stress range for welds.5.1 .

the vector difference of the 6. (d) stresses in triangulated skeletal structures due to load applications away from joints. (b) eccentricities necessarilyarising in a standard detail .2 Shear stress may be neglected where it is numerically less than 15 %of a coexistent direct stress. as shown in figure 2a.2. byand r are the coexistent values with appropriate signs of the two orthogonal direct stresses and the shear stresses at the point under consideration. 6. 6. transverse stresses and flange curvature (see Parts 3 and 5 ) . 6.3 The peak and trough values of principal stress should be those on principal planes which are not more than 45" apart. The reference stress for fatigue of a weld throat should be the vector sum of the shear stresses in the weld metal based on an effective throat dimension as defined in Part 3.2). the effects of the following should be included in stress calculations : (a) shear lag.1 The reference stress for fatigue assessment should be the principal stress in the parent metal adjacent to the potential greatest and the least vector sum stress may be used instead aack location. The stressesso calculated should be used with a material factor Y m = 1.March 1999 composite beams the modulus of elasticity of the concrete should be derived from the short term stress/strain relationship (see Part 4). 6.2 Stress in parent m e t a l bearing b e h n parent metals.1. member eccentricities at joints and rigidity of joints (see Part 3). stress concentrations should be taken into account either by special analysis or by the fadors given in figure 22 (see also H.provided that ox22 cq2at both peak and trough. 6. thiseffect should only be taken intoaccount On the evidence of special tests or specialist advice. (d) plate buckling.1.6 Effects to be ignored. Unless otherwise noted in of the algebraic difference ccI 4 -"> I i . where UX. except as required by table 17 . 25/02/2015. In either (a) or (b).3 Stress in weld throats other than those attaching shear connecton. the required stress range will be the algebraic difference between the numerically greater peak principal stress and the numerically greater trough principal stress.5 Effects t o be included.1. e +M1 t = combined size o f Vector sum stress I e f f e c t i v e weld throats (from Part 3) Figure Zb. Where indicated in table 17. This will be achieved if either (a) oX-oyis at least double the corresponding shear stress r a t both peak and trough.1.2. and on the assumption that none of the load is carried in 6.Reference stress in weld throat 0 BSI 03-1999 5 I .Welded attachment Design s t r e s s M' - L M-M' P o t e n t i a l crack Location I / =(/A+ P M/z) \ Stress distribution Figure 2a.4. Where appropriate. © BSI n I BS 5400: Part 10: 1980 table 17. However. 6. the stress should be based on the net section. restrained torsion and distortion.2 The bending stresses in various parts of a Steel orthotropic bridge deck may be significantly reduced as the result of composite action with the road surfacing.2.This is illustrated in figure 2b. (c) stress concentrations. Uncontrolled Copy. (c) cracking of concrete in compositeelements (see Part 5 ) . When calculating the stress range. The effects of the following need not be included in stress calculations : (a) residual stresses. Referencestress in parent metal [ P. Licensed copy:Parson Brinkerhoff. or (b) the signs of oX-oy and r both reverse or both remain the sameat the peak and the trough.Issue 2. (b) effective width of steel plates (see Part3) .

The mean centre line of travel of all vehicles in any lane should be along a path parallel to.2.4.2 and 7.2 Path of vehicles. 7.2. All vehicles less than 30 k N are ignored when considering fatigue.5). the passage of each vehicle forming the load spectrum should be considered to provide a separate loading event.2 Studconnectors. black bolts complying with the requirements ot 6s 41 90 may only be used i f they are faced under the head and turned on shank in accordance with the requirements of B S 41 90. 6 . 1081 mm 6.3 Application of loading 7.1 General. The transverse position of the centre line of the vehicle should be selected so as to cause the maximum stress range in the detail being considered.2.1 Demarcation oflanes. In welded members the dead load stress need not be considered. The design stress for fatigue in bolts complying with the requirements of 6s 4395 and bolts to dimensional tolerances complying with the requirements of 6 s 3692 should be calculated from the following expression : F stress in bolt =-= -/ uB uu where F = 1.for highway bridges this is a single vehicle with a weight of 320 kN. For shear connectors in accordance with the dimensional recommendations of Part 5. when the concrete is of normal density and from 0. The load factors YfL and Y f 3should be taken as equalling 1.2 and 6.3.2 Standard loading 7. See appendix C for the derivationof the standard fatigue vehicle.2.3.2.2. The minimum weight taken for a commercial vehicle is 30 kN.4. 6. The stresses in the weld metal attaching stud shear connectors should be calculated from the following expression : Licensed copy:Parson Brinkerhoff. If for any reason vehicle numbers other than these are adopted.4. Where thedimensions of the shear connectors and/or the concrete haunches are not in accordance with Part 5.1 The stresses in the weld metal attaching channel and bar shear connectors should be calculated from the effective throat area of weld.4 Stresses in w e l d s attaching shear connectors 6.2.2.7kN/mm 2forthreadsof nominaldiarneterupto25mm or F = 2. the fatigue strength should be determined in accordance with appendix G of this Part. 25/02/2015.4.4. transverse to the shear flow. 7.2.1 (e).4.2. which differs i n any way from the standard load spectrum.3. 7. 1018 rnm’ 127 and 102 channel connectors 150 mm long. The standard fatigue vehicle is a device used to represent the effects of the standard load spectrum . The passage of one standard fatigue vehicle along the entire length of one lane should be taken as one loading event. The numbers of commarcial vehicles that are assumed to travel along eachlaneofa bridge per year should be taken from table 1.5 Axial stress in bolts. Loadings for fatigue assessment 7. NOTE. 1697 mm 25 x 25 bar connectors :. suitable adjustments may be made to the fatigue analysis in accordance with 8.2. In determining the maximum range of fluctuating stress.1 kN/mm2forthreadsof nominaldiameterover25 mm b g is the stress range on the core area of the bolt determined on the basis of the minor diameter ou is the nominal ultimate tensile strength of the bolt material in kN/mm2 When subjected to fluctuating stresses. Uncontrolled Copy.3. Highway and railway design loadingsappropriatefor bridges in the UK are given in 7.O (see Part2). the centre line of the lane as shown in figure 6. If a load spectrum is used.2. modified where appropriate to allow for impact as given in 7.4.1 Design loadings. 7.3 Standardloading.3 Channelandbar connectors 6.3.3. They should be designated in accordance with figure 5 and the loading should be applied to the slow and the adjacent lanes only.1 Standardloadspectrum. only the vertical effects of vehicular live load as given in clause 7 should be considered. For the purposes of this Part of this British Standard the lanes should be the actual traffic lanes marked on the carriageway. 1212 mm2 76 channel connectors . and within 300 mm of.2. 7.3 or 8.2 It may assist calculation to note that in normal density concrete.2. 7.4 Non-standardloadspectrum. It consists of four standard axles with the dimensions as shown in figures 3 and 4. For the purposes of this clause the throat area should be based on a weld leg length which is the least of the dimensions tabulated below. 7.85 x throat area when lightweight concrete is used.3). Centrifugal effects need only be considered for substructures (see 7.1. In some instances it may be found that the use of multiple paths results in significantly less calculated damage and guidance on this is given i n C.4. 7.3 respectively.2 H i g h w a y loading 7. generally.3 Number of vehicles. In unwelded members the dead load stress will have to be considered in determining the effective stress range when compression stresses occur (see6.1 General. the design stresses for fatigue in the weld metal should be calculated in accordance with 6. Channel connector Bar connector Half the thickness of beam flange of bar) half the thickness of beam flange 6.3. 200 mrn long.2.3. © BSI stress in weld = longitudinal shear load on stud x 425Nlmm’ appropriate nominalstatic strength (from Part 5) 6.2. The standard load spectrum should beasshown intable 11 whichgivesthe weight intensities and relative frequencies of commercial traffic on typical trunk roads in the UK. Where a crawler lane is provided i t should be treated as an additional slow lane.BS 5400:Part 10: 1980 6. 7.4.2 Standardfatigue vehicle. where the thickness of the beam flange is at least twice the actual weld leg length and the weld dimensions comply with Part 5.2. the effective weld areas are : 50 x 40 bar connectors i : 200 mm long.1.< 150 mm long.

BS 5400 : Part 10 : 1980 80kN - 80 kN 1. 25/02/2015. kN (standard axles 1 -4 Figure 3. Uncontrolled Copy. kN 1. circle may be used Tyre contact area load 20 k N per tyre +tyre +tyre Figure 4.8 m 6. Axle arrangement of standard fatigue vehicle \ vehicle Licensed copy:Parson Brinkerhoff. © BSI twin tyre Alternatively a 225 mn dia.0 m -t - 80. Plan of standard axle 7 .8 m 80.

the static stress at every point affected by a wheel. Where a discontinuity occurs in the road surface. The loads should be epplied to the appropriate lengths of the point load influence lines of not more than two tracks.3.2.0 Not applicable 0. the force should be taken as acting at and parallel to the road surface. The loads to be considered should be the appropriate combination of the nominal live load. x 10 6 ) I Category of road Licensed copy:Parson Brinkerhoff.3. as specified in Part 2 of this British Standard. In welded members the dead load stress need not be considered.5 Not applicable The force assumed for any vehicle should not exceed 30 000 kN 7. so as to produce the algebraic maximum and minimum values of stress at the detail under consideration. The effects of any centrifugal force associated with the fatigue loading defined in 7.3 Railway loading 7.g. The effects of combinations of vehicles are allowed for in clause 8.3).o lanes per carriaaewav Motorway Dual 3 Motorway All purpose All purpose Sfio road Dual Dual Dual Sinale 3 2 All purpose All purpose Slip road Single Single (10 m * ) Single All purpose Single (7. for the individual vehicles of the standard load spectrum shown in table 11 as follows : the centrifugal force per axle = WY2 (kN) 127. where Wis theaxle load of thevehicle (kN) vis 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) 8 . © BSI I I Carriageway layout Number of Number of millions of vehicles per lane. Annual flow of commercial vehicles ( n . as shown in figure 7. should be increased by magnifying the relevant influence line.5 1. 7.5 Methodof loading.3.BS5400:Part10:1980 Table 1. In unwelded membersthedead load stress will have to be considered in determining the effective stress range when compression stresses occur (see 6.2. impact.3 m * ) 2 2 1 I 2 1.4 Allowance for impact.2 need only be considered for substructures.2 Application of loading. Uncontrolled Copy.2. at an expansion joint.5 1. e. 7. lurching and centrifugal force. r-150 7. 7. Only one vehicle should be assumed to be on the structure at any one time and each lane should be traversed separately. per year ( n e ) Each Each 2.1 General.2.1.5 Centrifugal forces. 25/02/2015. at or within 5 m of the discontinuity.0 1. The magnitude of the force should be calculated at the appropriate design speed of the particular road.

F Three lane single carriageway -L -- --ISI. Uncontrolled Copy. Two lane single carriageway Lane marking I c Hard shoulder or hard strip SI. © BSI n Adj.SI. 25/02/2015. Slow l a n e 1 Adj. ' Two lane slip road 1 SI. F Adj. ' SI. Figure S. Designation of lanes for fatigue purposes 9 1 1 F Fast lane 1 SI. F Adj. . Licensed copy:Parson Brinkerhoff. For two lane dual omit fast lane. Two and three lane dual carriageways SI. SI. Adjacent lane NOTE.

the appropriate values of n~ may be obtained by direct proportion. or where a design life other than 120years is specified. Where the volumes of traftic differ from the 27 Y 1O6 tonnes per annum.3. These tables relate proportions of the standard loading Kw to the total number of applied cycles nR x 1 0 6 occurring in a design life of 120 years and for a traffic volume of 27 x 106 tonnes per annum. The load spectrum for a permanent railway bridge subjected to standard loading should be taken from either table 2 for RU loading or table 3 for RL loading. NOTE. For the derivationof load spectra see appendix E. © BSI of mean c e n t r e l i n e of vehicles (selected t o cause stress range) 4 I wheels 4 900 vehicle c maximum wheels -Mean p o t hs of wheels 900 All dimensions are in millimetres Figure& Transverse location o f vehicles Di scontinuit y 4 Adjusted stress / Figure 7. However. 10 . 25/02/2015. Uncontrolled Copy.BS 5400 : Part 10 : 1980 1 t r a f f i c lane Permitted l o c a t i o n ~A:*L3 Mean centre line of t r a v e l I I I I Licensed copy:Parson Brinkerhoff. They also allow for variations in the loading events with influence line length.2 is used. reference to tables 2 and 3 is not necessary when theassessment procedure given in 9. Impact allowance at discontinuities 7.3 Standard/oad spectra. which are assumed in tables 2 and 3.

- U m n 5 11 .Licensed copy:Parson Brinkerhoff. Uncontrolled Copy. © BSI I n P. 25/02/2015.

i f appropriate. The values of OH obtained from figure 8 may be adjusted by multiplying successively by the following factors where appropriate. should be taken as the sum of the flows in those two lanes. 8.7 for bridges designed for 25 units HB 8. NOTE. This method determines the limiting value of the maximum range of stress for a 120 year design life and is generally simpler but more conservative than the more exact methods of 8.2 0. Appendix C gives the derivationof standard highway bridge fatigue loading.2.4.3providesfactorsby which non-standarddesign life.1) . 8.4.128 design life in years (b) Non-standard annual flows : factor = nc (from table 1 ( - ) '. the value of either bp or 0.2 Procedure 8.2.55 0. 8.4.1 ) .1 General.2. (b) The detail may be strengthened in order to reduce the value of ov max or it may be redesigned t o a higher class. 1 (m) 2 3 4 5 7 10 Licensed copy:Parson Brinkerhoff.2.05 0. the simplified procedures of 8. (c) Reduced values of abnormal load capacity (see C. in accordance with 7. © BSI 15 20 30 250 to 3.3.2. (b) the design life is 1 20 years .5 units H B factor = 1. whether resulting from the fatigue vehicle in the same lane or not .3.1 to rotalnumberof liveloadcycles nR Y 10') forvariousloadinggroupsand yp- 9 1 0 0 0 1 1 1 0 1 189 42 0 0 68 10 170 0 29 75 3 74 180 2 75 6 110 0 77 0 38 65 0 10 56 37 0 77 30 15 13 0 49 13 0 0 50 80 0 2 6 5 8 6 0 1 3 0 0 0 80 120 112 NOTE 1. the vector sum stress at the detail being 12 .Where stress reversaldoes not occur.5 0. NOTE. n. and determine the algebraic value of principal stress. (b) Apply the impact allowance of 7.2 0.4.2.2 Procedure 8.2. 8. (a) The detail may be assessed by the alternative procedure given in 8. The values are based on a traffic volume of 27 x 1O6 tonnes per annum. NOTE. In the case where opmsx and op.3 0. Uncontrolled Copy.2 is required or where the standard design life and/or the annual flows given in table 1 are not applicable.2.2 and 8. L is the base length of the point load influence line (see figure 12).1 The following procedure should be used (see appendix D) : (a) apply the standard fatigue vehicle to each slow and each adjacent lane in turn. anddeterminethe maximum and minimumvaluesof principal stress or vector sum stress for weld throat.2.3.2 Assessment w i t h o u t damage calculation 8. Standard load spectra for RL loading Qroup number 1 Lord proportion. o p max'and o p min occurring at the detail being assessed.8s 5400: Part 10: 1980 Table 3.2. It should only be used where all the following conditions are satisfied : (a) the detail class is in accordance with table 17 .2.8.'*' nc (assumed) where nc is the annual flow in the lane loaded to produce Ov max Op max .3.45 0. Kw 3. 8. or for weld throat.4.2.15 0. different traffic flow and design HB loading of less than 45 units may be taken into account. the load spectrum and the assumed annual flow of commercial vehicles.. As an alternative to the more rigorous procedure of 8. the design life. (a) Apply the standard fatigue vehicle to each slow lane and each adjacent lane in turn.1 M e t h o d s of assessment 8.2. 8.4 0. in accordance with 7. (c) the fatigue loading is the standard load spectrum (see 7. If ov may > 1. (a) Non-standard design life : factor = ( 120 . For non-welded details the stress range should be modified asgiven in 6.3Where ov max does not exceed OH the detail may be considered to have a fatigue life in excess of the specified design life. when appropriate.3 for bridges designed for 37. NOTE 2. For class S detail only. (d) obtain the appropriate limiting stress range OH from figure 8.2.2.2) : factor = 1.3. 8. 8.2. The sign convention usedfor b pis immaterial provided it is applied consistently.2.1.1.25 0.2. n~ values apply to one track.if appropriate. 8.3.55 OH for the other classesthis option will not satisfy the recommendations of 8. 25/02/2015.2.4. For intermediatevalues of L.b p min NOTE.1 General.2.4 4 5 6 to to 0.3 and 8. This method determines the fatigue life of the detail in question and may be used where a more precise assessmentthan that provided by the method of 8.1 The following procedure should be used (see appendix D).2 For class S details the values of OH may be adjusted by the factors given in 8. 8. (d) the annual tlows of commercial vehicles are in accordance with table 1 .3if it is not a class S detail.4.OP min.1 General.3 and 8.1.3. (c) determine the maximum range of stress gVmax equal to the numerical value of opmax . or by the procedure given in 8." are produced by loading in two lanes. mln should be taken as zero.-) 0.6 2 to 3 to 0. (b) apply the impact allowance of 7. class S details only.3 may be used provided the conditions stated are satisfied.2. It should only be used where the following conditions are satisfied : (a) the detail class is in accordance with table 1 7 but is not class S. (b) the fatigue loading is the standard load spectrum (see 7.30OH for class S details or > 1.4 Where oVmax exceeds OH either of the following options may be adopted.35 0.3 0.2 Simplified procedures. The selection of the appropriate procedure depends upon the detail classification.4.2.3 Adjustment factors foraH.2.4 if it is a class S detail.5 0. However. permissible stress ranges may be derived from the spectra for the two adjacent lengths shown in the table and the values interpolated. single vehicle m e t h o d 8. Three procedures for the fatigue assessment of details in highway bridges are given in 8.3 and 8. Fatigue assessment of highway bridges 8.3 Damage calculation.1 0 Range Length.

two lane all purpose (1 0 m). C 30 U) U) E c ul 0. dual two lane all purpose 80 70 z2 50 r 40 C 60 E .ategories 13 .=Ov) ul N B \ z - bs d.E 20 c . Uncontrolled Copy.. CI. .80 70 60 50 40 30 Licensed copy:Parson Brinkerhoff. 25/02/2015. Values of OH for different road c. © BSI 20 10 1 2 5 10 20 50 100 200 (b) Uual two lane motorway.-E D E S F F2 -I G W L lm) (c) three lane all purpose. two lane slip road Figure 8. dual three lane all purpose.

more than one cycle results and the individual stress ranges should be determined by the reservoir method given in appendix B.1 General. 20 where the summation includes all the separate lane stress histories as well as the combined stress history. vehicle spectrum m e t h o d 8. (2) where case 2 of figure 9 applies the effective annual flow fiC should be obtained as indicated in figure 9 for case 2. (d) Derive the stress spectrum ovl.1. as shown in figure 9 for lanes C and D.4. result from vehicle positionsin thesamelane (referred toascase 1 in figure 9) the damage should be calculated for the stress historiesfor each lane separately. For nonwelded details the stress range should be modified as in 6. uv2 etc.aeach peak and each trough in the stress history of each lane in turn (see figure 9).3. For a combined stress history from two lanes (see (c) above and case 2 in figure 9) KB should be taken as zero for determining K.2.Ofor the fatigue life of the detail to be acceptable.2. This method involves an explicit calculation of Miner's summation and may be used for any detail for which the a.2.3 In assessing an existing structure. The design spectrum may be divided into any convenient number of intervals. an approximate stress range for the same class of detail can be obtained by multiplying the original value by : (predicted life) "(m+O design life where m is the inverse slope of the appropriate log ar/log N curve given in table 8. where appropriate. 8. The design spectrum should then be determined by combining the stress spectra with the specified numbers of vehicles in the respective lanes. as shown in figure 9 for lanes A and 8 .4 Damage calculation. (h) Determine the predicted fatigue life of the detail from the following. Using the design spectrum.4. allowance for impact should be made in accordance with 7. Licensed copy:Parson Brinkerhoff. 25/02/2015.2 Design spectrum 8.2. with all the stress ranges in any one interval being treated as the maximum range in that interval but low stress ranges should be treated in accordance with 11.1. Where a stress history contains two or more peaks and/or two or more troughs. and the stress range can be determined directly. a design spectrum may be compiled from strain readings or traffic records obtained from continuous monitoring. from each stress history determined from (c).. When c ~ mar p and opmin result from vehicle positions in different lanes (referred to as case 2 in figure 9) an additional combined stress history should be derived.BS 5400 Part 10: 1980 assessed .3. & = nc and may be derived directly from table 1 unless different vehicle flows are adopted . 8. determine the appropriate lifetime damage factor d. This value should not exceed 1. op min.4.4. Tic million. 14 . expression : .3 Simplification of design spectrum.3.4.4 Calculation of d8mege.1 The individual stress spectra for the detail being assessed should be derived by traversing each vehicle i n the load spectrum along the various lanes.2 Where the predicted fatigue life of the detail is less than the specified design life.4. the detail should either be strengthened to reduce the value of uv max or redesigned t o a higher class and then re-checked as in 8. It should be noted that the use of small intervals will reduce the conservatism in fatigue assessment. Account should also be taken of the possibility of higher stressrangesdue to some of the vehicles occurring simultaneously in one or more lanes and/or in alternating sequence in two lanes.2. 8. 8. 2o from the damage chart of figure 10 and multiply each of these values by the appropriate value of ifc. 8. Uncontrolled Copy.2 In the absence of other evidence. appropriate t o each stress spectrum as follows : (1) where case 1 of figure 9 applies.3. Where a stress history contains only one peak and/or only one trough.4. 120 ZKFEcd. fatigue life (in years) = 8. only one cycle results.2. (c) When the maximum and the minimum algebraic values of stress op max.I = should be calculated in accordance with clause 11..4. or viceversa where this is more severe. the value of Miner'ssummation . 8. (g) Determine the value of the adjustment factor K F from figure 11 according to the base length L of the point load influence line (see figure 12) and the stress range ratio KB defined in figure 11.3. (e) Determine the effective annual flow of commercial vehicles. as shown in figure 13. If the detail is to be redesigned to a higher class the procedure given in 11. Inthis case the damage should be calculated for the combined stress history as well as for the separate lane stress histories. It issufficientlyaccuratetocalculateeachpeakor trough value of the direct stress and to obtain the principal stress by combining these with the coincidentshear stress. For the derivation of KF see appendix C.-Nrelationship is known and for any known load or stress spectrum.5( b) may be used as a guide to assess the adequacy of the proposed detail.1. (f) For each stress range ovof each stress spectrum. NOTE. © BSI NOTE. For non-welded details the stress range should be modified as given in 6. which allows for the increased maximum stress range produced by a proportion of the vehicles travelling in alternating sequence in the two lanes. As a guide.

B etc. Uncontrolled Copy. Highest peak and lowest trough withvehicles in different lanes Key 0 Peak stress 0 Trough stress Stress range Q. . A 1 Number of cyclea Lane streas smctrat Lane a t r o u hinow Lane A Lane B Lane C Lane I 1c -p 0 1 Licensed copy:Parson Brinkerhoff.BS 5400 : Part 10: 1980 I I c Lane reference* I - per loedlng *V*nt ffoctlve lane IOW. Imne low. © BSI Case 1. should be obtained by the method given in appendix B. Figure 9. for cycle shown X Datum stress *The lane reference letters A. E. t Values of b. iic Lane A Lane B Lane C k c 1c Lane D I Pv*rre Combined stress history A/B I I 5 min Case 2. f o r damage calculation 15 . CJvi 6 etc. Highest peak and lowest trough with vehicles in same lane ~ Lane straso one n r e u hlnory Lane reference' dumber of cyclea )er loading event apmctrat ffoctlv. 25/02/2015. should be allocated i n descendingorder of magnitude of stress ranges O V l A . Derivation o f gvand is.

Uncontrolled Copy. 25/02/2015.-VI U m em in I 16 . © BSI BS 5400 Part 10: 1980 5 al .Licensed copy:Parson Brinkerhoff.

Uncontrolled Copy. case 2).1. L is the base length of loop containing the largest ordinate measured in direction of travel. Typical p o i n t load influence line 17 . is the ratio - oVl B UVlA where is the largest stress range produced by loading in any one lane ov1B is the nexl largest stress range produced by loading in any other lane ( For non-welded details the stress range should be modified as given in 6. L should be determined from the influence line for the lane producing the largest value of b v i ( = O V l A ) (seefigure9). F2 and W.3. L is the base length of the point load influence line (see figure 12). This figure is applicable only to detail classes B to G. 25/02/2015. Figure 12. QVlA ~ V BC < oV1A). K. ( K e may be taken as zero for the combined history of figure 9. Miner's summation adjustment factor KF f o r highway bridges c - -Largest ordinate ( t v e or -vel NOTE. Figure 11.Licensed copy:Parson Brinkerhoff. For an element of a highway bridge loaded by more than one lane. © BSI BS 5400 : Part 10 :.1980 NOTE 1. NOTE 2.

otherwise it is obtained from 9. 9. In other cases the method may be more conservative than the method given in 9.1 __--1 I I . It should only be used where the following conditions are satisfied : (a) the detail class is in accordance with table 17 . k.2 Procedure 9. 9. (c) Determinethe maximum range of stress OR max equal to the numericalvalue of crp max-bp min.Simplified design spectrum Spectrum as calculated or recorded _--__--- Licensed copy:Parson Brinkerhoff..3. occurring at the detail being assessed.3 Non-st8nd8rd designlife.BS 5400 : Part 0 :1980 Stress ranges 10.1 and 7. © BSI Number of repetitions ( n ) Figure 13.2.3.2 may be used provided the conditions stated therein are satisfied.3. the value of k .2.3. For non-welded details the stress range should be modified as given in 6.1.2 and 9. Where stress reversaldoes not occur under the loading described. asshown in example 4of appendix F.1 Gener81.2.2.3. Slmpliflcstlon of e spectrum (d) Obtain the appropriate limiting stress range b~ from the following expressions: UT = k . (a) Apply the standard railway loading in accordance with 7. The choice of the appropriate method depends upon the detail classification and the nature of the loading. (a) (design life 120in years ) or 120 (b) (design life in years )A where m is the inverse slope of the br -N curve appropriate to the detail class and is obtained from table 8.1.3 Where CYR max is found to exceed GT either of the following options may be adopted : (a) the detail may be assessed by the more precise procedure given in 9. Uncontrolled Copy.3 k 2 = 1. This method determinesthe limiting value of the maximum range of stress for the specified design life.3 the simplified procedure of 9.1 The following procedure should be used.3 when the coefficientsk. (b) the loading is the standard railway bridge loading in accordance with 7. = 1.O if the design life is 120 years.2.O if the loading event produces only one cycle of stress.4 kr is obtained from table 4 k 4 is obtained from table 5 k B is obtained from table 6 k s is obtained from table 7 go is the constant amplitude non-propagating stress range for the appropriate class of detail and is obtained from table 8 NOTE. 'As an alternativeto the more rigorous method of 9.2. by loading the appropriate loops of the point load influence line. (b) Determinethe maximum and minimum values of principal stress or vector sum stress for weld throat u p max and CJ p min.2.2..in should be be taken as zero.1.1 Methods of assessment 9.2. The sign conventionused for bpis immaterial providing it is consistentlyapplied.3.2. The simplified procedure producesthe same results as the method given in 9.3. x k s x d ofor RU loading QT = k l x k 2 x k 4 x k s x ke x dofor RLloading where k . 18 .2. should be taken as the lesser of either : 9. either bpmsx or bp. whether resulting from railway loading on the same track or not.. 9. 9.2 Where U R max does not exceed UT the detail may be considered to have a fatigue life in excess of the specified design life. otherwise it is obtained from 9.2 Assessment w i t h o u t damage calculation 9.2 Simplifiedprocedure. 9. x k.. 25/02/2015.1 Gener81. Two methods for the fatigue assessment of details in railway bridges are given in 9. 9. Where the specified design life is other than 120 years.2.2. k 4 and k 5 (see 9. Fatigue assessment of railway bridges 9.2) are equal to unity. (b) the detail may be strengthenedin order to reduce the value of OR max or it may be redesignedto a higher class. x k 3 x k .

92 2.46 0. Values of ks for RL loading of railway bridges Detail class 0 E F C 8 S F2 G W Length.4 3.0 to 10.BS 5400: Part 10: 1980 Heavy trafflc Detail clem D E C 0 E F F2 0 W S 8 F Licensed copy:Parson Brinkerhoff.9 0.77 1.20 .75 1.0 to 4.0 1.08 1.79 1.10 2.05 1.0 > 20.6 to 0. Table 5.22 1.80 1.09 1.0 4.03 1.0 to 28.30 .6 O 1.0 to 15.0 7.22 1.13 1. 5 to 0.27 1.30 1.37 1.65 1. c L (m) < 3.0 1.46 1.09 1.46 1.57 1.10 2.13 1. Values of k5 for railway bridges 0.14 1.03 1 C B S 1.28 1.0 3.81 2. P .7 to 1.18 1.10 C 8 S E F F2 0 W 1.60 1.91 2.0 10.23 1. Values of k4 for railway bridges 42 to 27 27to18 18to12 12to7 7to5 <5 0.00 1. Trains on two tracks should be considered in the same longitudinalposition.46 1.56 1.28 1.75 0.4 3.03 2.18 2.83 2.0 14.56 2.0 4.71 1.37 1.53 1.71 1.62 1.83 . © BSI F2 0 W - Length.05 2.0 V ~ I J OO S f ha 1.80 1.28 1. P.46 1.36 1. I! is the base length of the point load influence line (see figure 12).o 1.o 0.06 1. 25/02/2015.20 2.71 1.71 1.00 1.9 to 1.83 I NOTE.50 1.79 2.24 2.62 . L (m) Valume of &3 < 3.90 2.89 NOTE.0 to 20.55 1.00 2.o 0.43 1.95 2.34 1.60 1. is the numerical value of stress due to track 2 (where track 1 is the track causing the greater stress at the detail under consideration).87 1.62 -65 1.92 2.71 1.95 1.01 1.19 1.53 1.95 Llght trafflc Mullurntrafflc 1.4 to 4.37 1.20 2. Uncontrolled Copy.45 1.19 2.0 1.27 1.0 10.31 2.31 2.7 0.37 1.42 1.09 1.4 to 4.92 2.46 2.65 1.35 1.23 1.20 2.79 1.68 1.71 1. is the numerical value of stress due to track 1.83 1.125 1.13 1.89 1.19 2.80 2.0 > 28.49 1.42 1.6 4.75 1.62 1.6 Table 6. Table 7.74 3.75 to 0.83 1.0 to 10.50 2.65 2.09 1.56 1.99 2.98 2.09 1.03 2.00 2.0 15.71 1.65 1.0 to 14.36 1.0 to 0.46 .03 .23 1.0 to 3.95 2.83 1.6 to 7.50 2.53 1.

2.3.2 In the case of loading from more than one track.2. I. Such cycles should be counted and the individual strew rangesdetermined by the reservoir method given in appendix 8. These should then be combined with the appropriate total occurrences in the design life of the bridge to compile the overall design spectrum.3.3. For non-welded details the stress range should be modified as given in 6.1 General. coefficientkS which can beobtained from table 6. for a given member after consideration of the fact that where the stress ranges are expressed as proportions of the coincidence of highway traffic on multiple lanes and of maximum stress range. may be appliedtostress histories for railway bridges (see example 4 of appendix F) and will produce the same results as the rainflow methodfor many repetitionsof the loading event.2 the value of OR max should be derived in loadings. passage and departure of a unit uniformly distributed load. As an open track.railway stations. and for non-weldeddetails the stress which should be obtained from 9. the effects of two track loading may be 1.1 ) or OR max/ks(see 9. the sum of the two damage These tables indicate.2.1 Where the loading does not comply with 7. n 2 . which occur spectrum. Where a nonstandard loading is used in accordance with 7.2. in descending order of magnitude..3 Where the approach.3 Design spectrum for non-standard loading N I . .3.5 should be determined from the following the relevant point load influence lines and the resulting expression : stress histories should be analysed by the reservoir method.3. the adjustment factor is not expected to exceed approximation.1 Applying the standard railway loading as given In the case of bridges carrying both highway and railway in 7. for simply supported members.2 Design spectrum for standard loading 10. the or -N relationship is known and for any known load or stress spectra.2. © BSI ( OR2 m OR3 I+(--) +(-) OR1 OR I +..3..2.N.3 or 8.1.3..2. given in appendix B. the effect of the additional in the design life of the structure.2.3.n. The number of repetitionsmay be modified in (see 9.1 repetitions to failure for the same stress ranges. ORZ. a design spectrum may be compiled from strain readings or traff ic records obtained from continuous monitoring. 2) These should then be combined with the appropriate Nn annual occurrences obtained from table 15 or table 16 where proportioned for the required traffic volume and multiplied n . This factor should be determined individual trains forming various standard traffic types. Using the design spectrum. ..2. publishedby the Office for Researchand Experimentsof the InternationalUnion of Railways. . railway traffic on multiple tracks has already been taken into account in assessing the separate damage values. all the cycles should be taken into account.4 and 9. obtained the appropriate train should be traversed across the relevant from 11. 5 'Bending moment spectra and predictedlives of railway bridges'. Thevalue of Miner'ssummation for use trains of figure 19 or figure 20 should be traversed across in 8.3).3. cycles may be obtained by dividing either UR max NOTE.2) by the coefficientkZ accordancewith 11. 120years divided by the accordance with the procedure set out in 9. fl2 . described in appendix 6 for highway bridges.1 General.. The reservoir method of cyclecounting.3. the values should be multiplied by a further adjustment factor equivalent frequency of occurrence of stress ranges of which takes into account the probability of coexistence of varying magnitudes resulting from the passage of the the t w o types of loading. 9.3 Damage calculation the value ot Miner's s u r n m a t i o n Z i should be calculated 9. 9. point load influence lines and the resulting stress histories should be analysed by the rainflow method to derive the xi *The rainflow method is described in ORE D128 Report No.3. the total damage (i.3.4 Multiple cycles. at the individual cycles produced by the approach.1 (a) to (c). rangeshould be modified as given in 6.5 Calculation of damage. are the corresponding numbers of 9. This method involves a calculation of in accordance with clause 11 and should not exceed 1. of either table 2 for RU loading or table 3 for RL loading (amended where appropriate in accordance with 7. are the stress ranges.2. as for instance in multi-span longitudinal or cross members or in continuous deck slabs.3.4 Simplification of spectrum.3. The appropriate standard 11.e. 9. predicted life) should bedetermined for each loading The design spectrum should then be determined by the use condition separately. account should be taken of the possibility of stress Except at very busy. OR 3 etc. 9. 9.2.3.13.1 and 7. 9. where the probability fluctuations arising from the passage of trains on not more of coincidence of rail and road traffic is higher than on the than two tracks.4.4.3.2 In assessing an existing structure.2.3. the design spectrum should be divided into at least 10 equal intervals of stress.3.3. Where the loading event produces more than one cycle of stress the value of k2 should be taken as : Licensed copy:Parson Brinkerhoff. To obtain the total damage. respective stress spectra. It may also be used as a more precise alternative to the simplified procedure of 9.O Miner's summation and may be used for any detail for which for the fatigue life of the detail to be acceptable. where rn is defined in 9.3.1) by the are of the same sign. Uncontrolled Copy. are the specified numbers of repetitions of the by the specified design life to produce the overall design various stress ranges in the design spectrum. both separately and in combination.3. 9.2 where the stressesfrom highway and railway loading obtained by dividing OR max (see 9. 25/02/2015. . All the stress ranges in any one interval should be treated as the mean range in that interval and low stress ranges should be treated in accordance with 11.4 and 9. An illustrationof the multiple cycle stress history is given in example 4 of appendix F.3. 20 . .M i n e r r u l e cycle of stress. Fatigue assessment of b r i d g e s carrying highway and r a i l w a y loading 9. passage and departure of a unit uniformly distributed load produces more than one 11. to derive the respective stress spectra. .B S 5 4 0 0 : P a r t l O : 1980 9.2. or the stress ranges are obtained from strain gauge readings. .1. As an approximation. in accordance with 8.3. 9. The P a l m g r e n .3 UR I . NOTE. . N2.

© BSI BS 5400 : Part 10 : 1980 21 .Licensed copy:Parson Brinkerhoff. 25/02/2015. Uncontrolled Copy.

22 . the value of uofor the new class should be between 12 - Licensed copy:Parson Brinkerhoff.0 10.e. should be reduced in the proportion (Ur/ao) 2 where where m is obtained from table 8 U . the plotted in figure 16 : N following alternative actions should be considered.3%withinthe design life.23 x 1013 78 C (d) Evaluate Miner's summation in accordance with 11. © BSI ( E. if > 1.08 x 1022 I 82 NOTE. The where strengtheneddetail should be satisfactory if the reduced K2 and m have the values given in table 8 for the different values of stresses lie between the limits obtained by detail classes.63 x 10'2 40 accordance with 11.5 14.5 for highwav and railway from either of the following equations. As a guide for It may assist in calculations to note that : upgrading to any class up to 0.e. The F2 13. Uncontrolled Copy. The following procedure should be used Table 8.B S 5 4 0 0 : P a r t l O : 1980 11. Values applicable to non-standardcriteria may be obtained from appendix A.4 and 9.conditions in 8. N = Log10 Kn m Logqour (a) strengthen the detail to reduce the values of Or. 11. ur -N relationships and c o n s t a n t in applying the Palmgren-Miner rule. N x U': = K.3.O.S Miner'ssummation greater than unity.4 Procedure. which have been bridges respectively are not met.) times the value of uoof the original class of the detail.3 Treatment of low stress cycles.leB 14. 25/02/2015.0 10. for 50 %probabilityof failure.43 x 10'2 I 35 number of low stress cycles should be modified in F 13.25 x 10" the design spectrum in accordance with 8. 5 2 ~ 1 0 1 2 53 accordance with 11. If the to failure Nof any one stress range orshould be obtained . (b) redesign the detail to a higher class. a m p l i t u d e n o n . 0 4 ~ 1 0 1 2 I 47 (c) Determinethe number of repetitions to failure Nof each of the stress ranges in the design spectrum in D (3.0 29 0.. W 13.4.3. 13.0 0. The number of repetitions of each stress range Or less than U .01 x l O l s 1100 I I I S 18.0 11.16 x 1012 25 (b) Calculatethe stresses and hence the stress ranges at each detail in accordance with clause 6 and determine G 13. The basic equations and a mean-line 'lmand plot i.2.0 1 1 .2 Design ur -N relationship. i. 11. dividing the original values by the following factors : NOTE. is the stress range given by the equation in 11.2 for or N= 10' and tabulated in table 8. are given in appendix A.0 1 1 . The valuesof K2 correspond to a probablityof failure of 2.p r o p a g a t i n g stress r a n g e values (a) Determinethe class of each detail in accordance with table 17. The number of repetitions 11. Either Log. E 13.3.4 and 9.0 12.

08 x 12. Table 9. The final result is an earlier fatigue failure than could be predicted by assuming that all stress ranges below uoare ineffective.561 10' 1012 10.3 % 2.625 I 0. Summary of mean-linea.2 E D C B S KO 10.13 x Id 10. This gradual enlargement reduces the value of the non-propagating stress range below u0.592 0. an increasing number of stress ranges below uocan themselves contribute to the further enlargement of the defect.862 0.99 x Il. Probability factors A. NOTE. A. Mean-line or-N relationships Detril clard W .Thus. as shown in table 10. © BSI .the larger stress ranges will cause enlargement of the initial defect.2 may be written in basic form as : N x U?' = K O x Ad where N is the predicted number of cycles to failure of a stress range ur KO is the constant term relating to the mean-line of the statistical analysis results m is the inverse slope of the mean-line log Or -log N curve A is the reciprocal of the anti-log of the standard deviation of log N d i s the number of standard deviations below the mean-line.0 18. The equation given in 11.0 13.1 m 13.14% *Mean-line curve.29 x 13.0 Endurance N (cycles) NOTE.57 x 1. 25/02/2015.1. This corresponds to a certain probability of failure. 14. The relevant values of these termsare given in tables 9 and 10 and the mean-line relationships are plotted in figure 15.34 x 12.c Appendix A Basis of a.7 or 8 is considered to have a constant amplitude nonpropagating range ooequal to the value of urobtained from the formula in A. Uncontrolled Copy.0t 0.2.5 Id ::: .313 1012 10" 10l2 101 2 I : . Under fluctuating stress of constant amplitude.BS 5400: Part 10: 1980 - Licensed copy:Parson Brinkerhoff.654 10. so that some of the stress ranges are greater and some less than uo.73 x 13.605 0. This phenomenon has been studied on principles derived from fracture mechanics.0 13.0 13. tThe standard design curve of 11. The value of this'non-propagating stress range' varies both with the environment and with the size of any initial defect in the stressed material.l when N = 1 07.657 10' lOZ3 10.-Ncurves 23 . as time goes on. there is a certain stress range below which an indefinitely large number of cycles can be sustained.i Probmbllity of trlluro 2.617 1014 10.l General. Figurel5. In clean air. When the applied fluctuating stress has varying amplitude.37 x 10.2Treatment of low stress cycles. The use of these curvesfor calculation purposes is not recommended. The or-N relationships 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.23 x 1.-N relationship Table 10. a steel detail which complies with the requirements of Parts 6. less than oocause damage in accordance withtheformula inA. It is found that an adequate approximation to the fatigue performance so predicted can be obtained by assuming that a certain fraction (flr/Oo) of stress ranges Q.0 3.

3 %.06 2.23 1.27 I.9 or 11.23 1.14 1.47 1.23 I.32 5.07 1.18 1. © BSI + Datal1 Numbmrofatmndard davlatlona balowthamun-llna clar W G F2 F E D C B S 1. The probabilitiesof failure associated with various numbers of standard deviations below the mean-line are given in table 10.53 1.28 2.32 1.89 2. which has the inverse slope m 2 where Nisgreaterthanlo’.17 1.20 - 1.6 0.38 3. which are based on two standard deviations below the mean-line. In certain cases. I I \ I I \ \ \ \ I \ I I I \ \ m+2 \ \ \ I \ 1o7 Endurance N (cycles) .15 1.18.1 0. The Qr -Ncurves appropriateto other numbers of standard deviations below the mean-line can be derived from theformula given in A.24 1. where fatigue cracking would not have serious consequences.65 1. 25/02/2015.24 1.56 1.-N relationship 24 -.42 1. a higher probability of failure could be acceptable.5 1.34 1.79 1.0 - 1.27 1.09 1.07 1.63 2.73 2.71 10.40 1.08 1.24 1. or where a crack could be easily locatedand repaired.1 2 2. These points are illustratedin figure 16.07 1.32 1.0 0.20 2.02 2. 2.34 1.30 1.86 2.29 1.52 3. predict the fatigue life (or damage). The standarddesign Qr-Ncurves in figure 14are based on two standard deviations below the mean-line with a probability of failure of 2.88 2.BS 5400:Part 10: 1980 The same result can be obtained by using a notional log ar/log Ncurve.31 1. Only that portion of this figure shown as a full line is based on experimental evidence.log scale NOTE.30 1.62 1.34 1. Figure 16. Uncontrolled Copy.21 1.3 Fatigue l i f e for various failure probabilities.0 1.27 1.78 1.85 2.10 1.\ .05 1.23 1.l.60 1.17 1.19 1.11 1.09 1.6 1.----_-_ Effective curve obtained under variable amplitude loading.22 1.29 1.o 0.21 itatic limitations Constant amp1i tude loading in clean a i r ----.38 1. Typical U. which shows a typical log Or/lOg Ncurve. A.69 1.34 Number of rtandard dOvlatlon8 W G F2 F E D C B S bSlow1 a moan ne .16 1. Where the methodsof clauses 8.15 1.-1.07 1. for example. the life (or damage) appropriate to other numbers of standard deviations below the meanline can be obtained by multiplying the calculated life (or dividing the calculateddamage) by the following factors : Licensed copy:Parson Brinkerhoff.51 1.

2. Mark the highest peak of stress in each occurrence.l. medium and light loading groups (H. in accordance with 8. due to one loading event.1. after many repetitions of the loading event. for the majority of detail classes and influence line lengths.2. Sketch the history due to twosuccessive occurrences of this loading event. Other relatively uncommon vehicle types have been included in the types which are nearest to them on the basis of equivalent damage.BS 5400: Part 10: 1980 Licensed copy:Parson Brinkerhoff. axle arrangements and frequencies of occurrence.6 For non-welded detailsonly. varies between the limits shown in table 12. is suitable when dealing with short stress histories.2.l ~tenderdloedspectrum. 8. and shown in the figure below. It consists of imagining a plot of the graph of each individual stress history as a crosssection of a reservoir. The calculated values of peak and trough stresses may be joined with straight lines if desired.2. The method given in this appendix.1 Derive the peak and trough values of the stress history. mark only tho first such peak in each occurrence. the standard fatigue vehicle with double tyred wheels gives an adequate representation of the wheel damage for all types of vehicles. To allow for variations in the loads being carried by similar vehicles..3successively with each remaining body of water until the whole reservoir is emptied.3 First occurrence I - H i ghes t peak I Appendix C Derivation of standard highway bridge fatigue. 8. If there are two or more equal lowest points the drainage may be from any one of them. compared with the total damage by all vehicles.2. © BSI Appendix B Cycle counting by the reservoir method 6. loading and methods of use C. counting one cycle for each draining operation.1. which is successively drained from each l o w point.2.2. Private cars and light vans below 15 kN unladen weight are not included as their contribution to fatigue damage is negligible. List one cycle having a stress range uvrequal to the vertical height of water drained. which is typical of the full range of commercial traffic on a trunk road in the UK. ovt etc. such as those produced by individual loading events. The proportions of the various types of vehicles of the spectrum have been taken from sample traffic counts.2.3.2 Join the two marked points and consider only that part of the plot which falls belowthis line. 6. 8.1.4 Repeat 8. Uncontrolled Copy.5 Compile the final list which containsall the individual stress ranges in descending order of magnitude G. list them separately. the various types have been divided into heavy.3).2. like the section of a full reservoir.) The proportions of the damage caused by individual vehicle types. will be the same as that obtainable by the rainflow method (see the footnote to9. c. Second occurrence / Shaded oreos a r e thoser p a r t s of t h e reservoir which successively become e m p t y Imaginary reservoir V V 25 .3 Drain the reservoir from the lowest point leaving the water that cannot escape.2. (See 7. Table 11 includes vehicles operating under both the Motor Vehicles (Construction and Use) Regulations and the Motor Vehicle (Authorization of Special Types) General Order.2. 25/02/2015. Standerdfetigue vehicle. Where two or more cycles of equal stress range are recorded. The axle loads have been averaged from weighbridge records ot moving traffic taken between 1971 and 1974. which has been used for some years as the datum axle in the fatigue design of road pavements.(See7.2..2 M e t h o d 8. isgroup 4A-H. listing one cycle at each draining operation. If there are two or more equal highest peaks in one history.1 General.3.) Table 11 shows a 25 band spectrum of commercial vehicle weights.1 (c). The result.2. M and L). 8. As the damage done by the single tyred wheels on vehicles listed in table 11 is normally less than 4 % of the damage done by the double tyred wheels. The axle spacings of the standard fatigue vehicle are the same as thoseforthe short HB vehicle (see Part 2) and the 80 kN axle weight is equivalent to the standard 18 000 Ib axle. The standard fatigue vehicle has been devised to represent the most damaging group which. a horizontal line representing zero stress should be plotted and those parts of thestressrangesin thecompressionzonemodified asin 6. The purpose of cycle counting is to reduce an irregular series of stress fluctuations to a simple list of stress ranges.l Standard loading C. 8. - \ -.2.

as L increases.4.3.2. C.2 and 8.The X factor is the average value of theratio of damage due to the table 11 vehicles to that due to the equivalent standard fatigue vehicles. (See 7. This has the effect of limiting the gross weight of the 18 GT group so that static design stresses are not exceeded. For values of L less than 25 m the design spectrum becomes increasingly influenced by bogie and axle spacings and weights.2). which are labelled X.2.3. Factor X is described in C. The weighbridge records.2 Demege fectofs d .2 and 8. The effects of numbers of vehicles other than the 120 million assumed are allowed for by multiplying the lifetime damage from figure 1 0 by nc. (See figure 8 and 8.The worst assumptions have been made about the stress contributions from each lane and about the allocation of traffic flows between lanes so that the results are always on the safe side. The principle behind the assessment procedures of 8.3 Assessment c h a r t s C. The chance that damaging vehicles will be sufficiently close to each other. where the design is based on reduced H B loading and this is described in Ci4. (See 7. 25/02/2015.1).4 A d j u s t m e n t f a c t o r s C. make up the Miner's summation adjustment factor KF given in figure 1 1. C.3 instead of the complete vehicle spectrum of table 11. adjustment factors are necessary to allow for shorter base lengths and for the effects of combinations of vehicles. and an increase of 10 % (indicated by trial calculations) over the individual values of table 11 has been allowed in deriving the spectrum of axle weights of table 14.)Theannual flows of commercial vehicles nc x 108 given in table 1 are based on the design capacity of the particular road type. The proportion of commercial traffic above 15 kN unladen weight has been taken to be 20 %of all traffic for all-purpose roads and 25 % for motorways. account should be taken of the contribution to damage done by two or more vehicles acting simultaneously. have shown that the increase in damage is not significant and may be neglected. It assumes 106 cycles of stress per year for the 120year design life and the damage. 20.3 Multiple vehicle effects.3. However.3. 26 C. These factors. derived from observations of traftic patterns. (c) vehicles in different lanes in alternating sequence causing stresses of opposite signs and so increasing the stress range. trial calculations.Licensed copy:Parson Brinkerhoff. © BSI BS 5400 : Part 10 : 1980 C. C. either in the same or different lanes.4.3 is that each commercial vehicle in table 11 is represented by one vehicle of the same gross weight but with axle configurations identical to those of the standard fatigue vehicle. each with a range of loop lengths.1.2. . All these effects have been catered for by means of a simple adjustment factor KF (see figure 11) which is described in more detail in C. C. the design spectrum for individual vehicles will be proportional to the gross vehicle weight spectrum.2. which take into account the probability of the occurrence of alternating sequences. as specified by the Department of Transport. instances will occur when more than one vehicle will contribute to the stress in a detail at any particular time and the stress may be increased above that due to either vehicle alone. The adjustment factor K F . has been used in the derivation of the procedures of 8. The resulting load spectrum shown in table 13. The adjustment factorKF has been included in the derivation of figure 8 but should be applied explicitly when figure 10 i s used. assessed by Miner's summation.4. is given in relation to the stress range caused by the passage of a standard fatigue vehicle. have been derived with theaid of the damage chart of figure 1 0 (see C.2Influence line bese length less then 25 m. from figure 11.2 there is an additional adjustment. an alternating succession of vehicles along paths of opposing sign would produce a reduced number of cycles of enhanced stress range. has also been included in the derivation of figure 8.1 Genere/. The standard fatigue vehicle centre line should be traversed along the centre line of each 100 mm strip and a stress spectrum obtained for each strip in accordance with 8. Figure 17 shows a histogram of occurrence of proportion factors for 100 mm wide intervals of carriageway width.1 Limiting stress ranges OH. (See figure 1 0 and 8.2. C. The two groups of vehicles were traversed across 11 different shapes of influence line. from which table 11 is derived (see C.)A significant reduction in assessed damage can be achieved by the consideration of multiple paths when the transverse influence line profile departs rapidly from thevalue of the mean path ordinate.4. During normal conditions of traff ic flow.3.3.show a wider variation in axle loads than in gross vehicle loads.) The graphs of allowable stress ranges in figure 8. The X component of the KF adjustment factor has been obtained by comparing the stress histories for a selection of the most damaging vehicles in table 11 with those derived by representing the vehicles by standard fatigue vehicles with the same gross weights. Because the assessment charts of figures 8 and 10are based on the passage of single vehicles and also on an assumed influence line base length of 25 m. where the various vehicle group weights are expressed as a proportion of the standard fatigue vehicle gross weight. However. The scatter in results between different influence line shapes was found to be acceptable for design purposes. C. When L is greater than 25 m.1.3. In cases where the transverse influence line changessign across the 1300 mm histogram width. These multiple vehicle effects may besub-divided according to the following : (a) more than one vehicle in the same lane simultaneously . Uncontrolled Copy. The adjustment factor K F should be applied to the results from figure 10in order to allow for influence line base lengths of less than 25 m and for the effects of multiple vehicles.4 Multipleperhs. which are based on a 120 year design life and the appropriate traffic flowsfrom table 1. and the Miner's summation for damage calculated using the Or -Nrelationships for all the the detail classes except S. Table 13 has been derived using a datum influence line which is 25 m long and rectangular in shape. For the assessment method of 8. which may be made in the case of class S details.3Stenderdlene flows.4.2.2 and factors Y and 2 in C.1. It should be noted that the values of these factors cannot be determined precisely but that the values of KF given are sufficiently accurate for design purposes.4.3.) The damage chart ot figure 10is based on the cumulative fatigue damage caused by the design spectrum which is obtained from the passage of the vehicles represented by the gross weight spectrum of table 13 over an influence line base length of 25 m. Correction factors have been derived from the comparison between the damage from the combinations of vehicles and the damage due to such vehicles on their own.2.2 Derivation of load spectra based o n t h e standard f a t i g u e vehicle.4. has been assessed on a probability basis. with the annual flow of vehicles in any strip being derived from the appropriate proportion (from figure 17) xnc x 106 (taken from table 1). Y and 2 in figure 18. (b) vehicles in different lanes simultaneously causing stress of the same sign .

z- 0 0 2: 0 0 0 w m m .c" i s c s.l . 25/02/2015.-0 a d 27 m (v .0 3 0 0 bco N 0 e K m m m 0 0 0 N 0 0 0 dQ) N 0 oal d N N E2 h nmQ) 0 0 -03 N 0 0 -03 N 0 0 -00 5K %K 5% ? bar4 N r l - c I x z .l - o m o c o m e ~ dQ) N 0 0 0 F e 0 0 2: c . Uncontrolled Copy.BS 5400 : Part 10 : 1980 5544 U m - 0 0 0 3 0 0 0 0 3 0 0 0 0 3 0 m m m 3 0 3 88 D O 0 0 03 N 3 0 0 T 1-F 0 0 (Do r c 0 0 m m C I C 0 0 Licensed copy:Parson Brinkerhoff. 9 (v - .c n m m 0 0 0 0 m m m 3 m m m m CQCO L 0 e 0 0 0 3 0 0 0 0 0 3 0 0 0 0 0 0 c)mm w m m l- 0 0 .E n mo 3 0 0 0 0 I- (zm .E 0 0 0 bQ) N 0 C c c F -03 N 0 0 7 0 3 N m r m - 0 c m r m .- e m 0 r 0 m 0 0 m 0 I- 0 0 (zco -- 0 0 0 corn Q) me4 l- m - 0 n o o co N - W 1 1 1 1 1 o m o n m o r e m m .I rr! * 3 w m r rz-l g. © BSI c c 0 0 0 22 0 0 dco N 0 0 zz 0 0 coo - i mw .-P (P 0 A Pm p ID 0 c" .d N r n o m b o 0 0 0 # d N ?X8 r .

significant variations occur in the values of X (see C. the ratio between the major stress ranges produced by vehicles travelling separately in the two lanes producing the most severe stress effect.003 % 'Seetablell. The effect given in (c) above is taken account of by the additional combined stress history. Combinations of vehicles in more than two lanesdo not generally increase the total damage significantly. Hence no relaxation is provided in the assessment proceduresof 8. C. Inthis case the effect of vehiclestravelling simultaneously in two lanes may be neglectedand so factor Z can be taken as zero by making K s equal to zero.4. Proportional damage from individual groups of typical commerclal vehicles ~~ Dstril clrw - D to a.e. Hence the adjustment factor X does not apply to class S details and the assessment procedure of 8.4.BS 5400 : Part 10 : 1980 C.which isgiven for severalvalues of factor K g (see figure 11). - i % C 3. 25/02/2015. However.0) 1.4. i.3for these other classes when the bridge is subject to reduced values of HB loading.0) 18 GT-H 9 lT-H 18 GT-M and 9lT-M to 5A. Multiple paths 600 rnm ce 600 mm c .3 cannot be applied.5 1% 1% 5% 4% 3% 4% 57 % 34 % 6% 14% 63 % 8% i 4.5) m 1-25 1.) Table 12 shows that the heaviest abnormal load vehicles of table 11 contribute a very small percentageof the total damage for all detail classes except class S.4.5 L - m 25 8.2 ReducedHB designloeds.2) both with variations in L and with variations in influencelineshapes. F i s n d W 'm 3.1 General. of mean p a t h of vehicles (see figure 6 1 % frequency distribution of vehicles for each 100 mrn interval - I I "I 1 Figure 17.3 (c).2 and 8.4. Uncontrolled Copy. M and L 4 R .5 I L -25 13% 57% 54% 6% 33% 0.3 (c) provides reduction factors where the bridge is designed to carry less than 45 units of HB loading. The effect given in (b) above is allowed for by means of factor 2 (see figure 18). © BSI Table 12. referredto as case 2 in figure 9. I' Licensed copy:Parson Brinkerhoff.2. The effect given in (a) above is allowed for by means of factor Y (see figure 18) which shows that the effect is negligible for L less than 50 m but increases with increasingL due to the greater probability of having two vehicles simultaneously on the same influence line base length. Trial calculations indicate that for class S details.057 % 0.0) -1. these heavy vehicles do contribute the greater proportion of the total damage for class S details and hence 8.2. (See8.H 5A-M and L 4A-H.H to 2 R-L 14% -- m ! Group.4 Class S details C.4.

Uncontrolled Copy.38 5.030 0. This table is based on L < 1.13 2.030 0.000 04 I 0.000 07 0.090 0.61 10. Typical commercial vehicle gross weight spectrum Vehicle Proportion of Proportion of tote1 vehiclea 0. Typical commercial vehicle axle weight spectrum Total axle weight 264 231 176 165 154 143 121 110 99 93 88 77 71 66 61 55 49 44 39 33 22 17 Total I Totalnumber of exlerfor 106 vohicloa 240 120 160 560 100 780 80 90 040 240 280 320 000 59 320 59 350 180 000 59 930 165 000 290 040 150 000 120 000 320 000 380 000 60 000 360 000 2 856 000 axles for 10' vehicles NOTE 1.BS 5400 : Part 10 : 1980 c c 0 1 L Licensed copy:Parson Brinkerhoff. Typical Miner's summation adjustmont C U ~ V O - Table 13.42 0.180 Total 1.00003 0.01 4 50 0.000 01 18GT-M 9lT-H 9TT-M 7GT-H 7GT-M 7A-H 5A-H 5A-M 2.01 5 0.38 10. Thesevaluesinclude the 1 0 %increase referred to in C. © BSI Figure 18.01 5 0.88 0. 25/02/2015.38 0.000 28 0.00003 0.090 0. NOTE 2.M c 4R-H 4R-M 4R-L 3A-H 3A-M 3A-L 3R-H 3R-M 3R-L 2R-H 2R-M 2R-L 0.090 0.5 m.09 2.47 1.170 0.015 0.2.20 10.0 Table 14.75 0. This table is based on L = 25 m (rectangular loop).01 5 0.000 02 0.000 02 0.75 0.1 70 10.01 5 0.13 4A.015 I 0.015 0.09 0.34 14.03 2.67 0.97 1.030 0.28 0. 29 .44 0. I NOTE.

8 -1.3). The U H values given in figure 8 thus provide a simple check which is very suitable for initial design purposes.4 Assessment using 8.8 . while the application of a combined stress history is illustrated in D. since it has the greateststressrange (i.5 N/mm’(seefigure8(b)).under the standard fatigue vehicle. which is adjacent to the transverse weld of UP min = -8.2.5Assessment using8.8 0 0 .l General. slow lane 1st carriageway.2.0 N/mm2 Henceuv max = 17.2 Details of theproblem. D. For all detail classes. © BSI ~ Examples of fatigue assessment o f highway bridges by simplified methods D.1 (a) and (b). provided that the weld end is not within 10 mm of the flange toe.0 N/mm compressiverespectively.8 2nd carriageway.8 .0 1st carriageway. A three span (50 m/75 m/45 m) twin girder highway bridge that carries a dual two lane motorway and is designed to carry standard UK loading.2. the slow lane will be designated lane A.2.2. D.5 2nd carriageway.2. 25/02/2015.0) = 25.9 -2.7% probability of survival) and indicating the extent of required changes in thedetail when the predictedlife is too short.0 The values of uv for the other lanes may be determined in a similar way.(-8.2. 8.2. - D. D.8. with the vehicle positioned in thesame lane.9 of table 17 (b)).The reservoir method (seeappendix 6)may be used to determine the values of oVas shown below : 17. produces principal maximumand minimum stressvalues of 17.3 provides a more precise method for detail classes B to G and F2 and W leading to a life prediction (to 97. Uncontrolled Copy. op may = 17.Appendix D a vertical stiffener. 8.3 Classification. Alternatively. 0.2.8). U H = 20.2 Example using t h e basic assessment procedures f o r a steelwork detail 0. 17.2 and figure 8 providea limiting stress range QH which will always be safe where standard loading conditions are applicable but which may be too conservative in some cases.1 Type of bridge. The basic assessment procedures for a steelwork detail are illustrated in 0.8 ( -8.2. 30 . The potential fatigue crack should be classified as F (from type 2.2. 5 5 ~ H or (b) the detail should be strengthened to reduce UV max or improved to a higher class. in the first carriageway.4.e.8. Fora dual two lane motorway with L = 75 m and a class F detail.8 Nlrnmz tensile and 8. D.0) = 25.2. in accordance with 8.2. Hence UV max exceeds OH and adequate fatigue life is not demonstrated (see 8.2 From D. since uv max < 1 .1 (c)).8 N/mm2 Licensed copy:Parson Brinkerhoff.3. Given thatthe stress histories for each loading event (passage of the standard fatigue vehicle in each lane) are as follows: 17.2.8 N/mm2 (see8. slow lane From figure 9.3.2. adjacent lane -0. adjacent lane d 7 -1. Atypical procedure for shear connectors is given in D.2. given that analysis.4 either (a) the procedure of 8.3 (see8. By reference to 8.2 to D.2.4 give examples of typical calculations for fatigue assessment.3may be used.1 (c)). These assess the fatigue resistanceof the main girder bottom flanges at the mid-span of the main span.

25/02/2015.032 0 Neglect ( c 0.03 1.m 0.2. or the lengthier procedure of 8.e.32 0 Neglect ( < 0.BS 5400: Part 10: 1980 This assessment procedure can be tabulated as follows : numkr 1st Slow 1 l2 *D.1 (h)) which is greater than the specified design life of 120years and so the detail may be regarded as satisfactory. prove a useful 'first stage'as an alternative to the more precise method of 8.843 =i 4zyears(see 8.3.5/25. © BSI Domigo &d. Of uH 31 ."'" . (142 \ o'2si. the procedure of 8.5 I I &: 0.59 x 0.53 = 0.=: 1-1.8 = 0.8 = 0.3 predicts a fatigue life that is in excess of the design life and hence a reduction in cross section could be tolerated. The method of 8. \ ILV/ In contrast.79 times the applied maximum stress range.4.6 1=75m hence KF = 1.59 Total damage = Z K F ECd . 1 10. tSee figure 10.001 ) I 2nd Adj.04. (see figure 11) KB = 15. #See table 1 and figure 9.2.2. Nevertheless.8 Licensed copy:Parson Brinkerhoff.8 0.8 0.2 produces a limiting value equal to 20*5/25. Uncontrolled Copy. thus demonstrating the conservatism of this method. the simplicity of determination of OH is such that the method should.48 1. Adjustment factor K.843 Estimatedlife= 2010. - the stress range could be increased by . By reference to 8.2.2 2 0 C 2nd Slow . 4 10t Puk Cliu F 17. 2o = 1.o 0.001 ) 12.001 ) 7. in many cases. D.02 - t :01 *See appendix B.3.001 0. by 1.o 0.8 Neglect ( < 0.5 -0.6 Comments.3.01 5 - I 1 IlS5 Neglect ( < 0.0 0.001 ) 1.

Given that the analysis predicts.2 does not comply with (b) and (d) of 8. but is deemed to be subject to 1.9 . Given that these are as follows : 12.9 12.3. for a design life of 60 years. D.9 -12. 12. © BSI Lane A .5 Assessment using8. Assume a cross-sectional area of 2200 mmz and determine the stress histories for passage of the standard fatigue vehicle in each lane.1 Typeofbridge. determine an area of cross section for the member.2.3 Cl8ssification.0.3 Example of the application of a combined stress history D.2 x 106 commercialvehicles per year.5. Uncontrolled Copy.1 and therefore 8. A highway bridge that carriesa two lane.weld will beattheedgeofthemember (seealsofigurel).3.3. 0. D. in each lane.4Asssssm8nt using8.9 Licensed copy:Parson Brinkerhoff.3.3.12.2 is not applicable.12. 25/02/2015.3.1 (c) and case 2 of figure 9).2 Details of theproblem. the forces in a transverse bracing member due to the standard fatigue vehicle in alternate lanes and given that the influence line base length is 8 m. single carriageway. as shown in D.3.1 and D.9 Lane A/B 32 -1 2.11 oftable17(b)) sincethe .9 12.3. D.2.9 Lane B Derive the combined stress history (see 8.9 .2. all-purpose road and is designed to carry standard UK loading. The potential fatigue crack should beclassfiedasG (fromtype2. The information given in D. to provide adequate fatigue resistance at the lap-welded bracinglgusset connection.3.3.

Adjustment factor K F (see figure 11) for separate histories KO = 12.:.8 The assessment procedure can be tabulated as follows Lene Range dttot number OV Claaa 0 1 12.03 }I. A 2 12.01 0.9 8.B= 25.01 } . Repeat the assessment with an amended initial assumption of the area.41 Estimated life = 12011. 0.01 12.84 0.08 I I .6 *See appendix E.2.2.47 x 0.2 = 0.3. NOTE.81 for a combined history KB = 0 L =8m hence KF = 1.86 = 0. Iterationprovides a solution of 2027 mm2.6Comments.3. 1 2 :1 .2) . i. tSee figure 10...n The stress ranges uvmay be determined by the reservoir method (see appendix B) and the resulting stress spectra (see figure 9) for the individual and combined lane stress histories will be as shown below : Lane A 54 Z'cycles per loading event 0 "Pt 0 Lane B 2 cycles per loading event 4 Combined A/B Licensed copy:Parson Brinkerhoff. $See figure 9.03 0.O L=8m LI henceKF = 1. fie d i z o 0.81 x 0.055 B 2 -4.055 CVCI.9 0. 33 .9 0.9 8. A.7 0..( h ) ) which is greater than the specified design life of 60 years and so the detail may be regarded as satisfactory.e.9 1 0 14. 2o = 1.6 0.01 0.2 -12. 25/02/2015.01 Damaga "C! = 0.08 -7 1.2 -1 2. The estimated life is in excess of the 60 year design life and hence a reduction in area i s allowable.9 = 1.86 1 :zl ..3. g) 0.9112.6 0.2 = 0.47 Total damage = Z K d .41 = 85 years (see 8.91 7 (see 8.01 Z& dc ao 0. 0. 1. and hence an acceptable area will be 201 7 mm2.21 The reduction factor will be( .7 0. Uncontrolled Copy. © BSI A .

WP. and u ~ = 4 0x 1.2 and illustrated in D. wolght. numbers of these trains.4. but which is been based on the typical trains shown in figure 19. and 30 m for connectors at the shown in table 16. E.2 RL loading.67 2 I253 2. withthe 1. It is not possible to use 8. for connectors at the ends.2 or 6.31 26 986 - 1 1 1 1 - I I I I I I I I 1 1 i!:.070 PUkN at mid-span and 14516 5..10 Puis the nominal strength of the stud from Part 5.1 and Experiments of the International Union of Railways should the weld throat stresses may be calculated according be consulted : to 6. 5 600 4 500 2.4. positioned in accordance Total 127.5 units of HB loading. The load spectra given in table 2 have road and is subject to standard UK loading. assumed for the three broad traffic D. for this example. alternatively. concrete deck slab of flat soff it. the value of L will lie between 15 m. Table 16.3factorallowedfor37. are shown in table 15 together with the make-up of capacity of the attached shear connectors (which are in the total annual tonnages.2.50 5 600 122500 Medium 29.1. the maximum allowable shear load per connector under loading from the I6 I 572 I 2360 1. RL loading: annual t r a f f i c t o n n a g e where a n d composition of standard t r a f f i c m i x A. x lC3kN attheends.5 Comments.5unitsof for standard traffic types HB loading (see8. Report ORE D128/RP7 D.35 standard fatigue vehicle.0 x A.0 Pu/425 = 0. 25/02/2015.4. they also comply with 6.4 Assessment using 8.70 52 N/mmZforconnectorsat theends. Since the connectors are in following reports published by the Office for Research and accordancewith Part 5.2) the maximum 1794 2 257 4. for a dual carriageway all-purpose road and class S Table 15. 2.84 1 I246 111 545 I54 032 I 13. OH 46 x 1.40 52.Licensed copy:Parson Brinkerhoff. R U loading: annual t r a f f i c t o n n a g e (seefigureB(b). x lC3kN a t mid-span and 52. Hence.1 22 PukN at the ends 8.1 Type of bridge.2.9 x A. 1 FcK I I I I +See figure 20. tonnage. for together with the make-up of the total annual tonnages.9 N/mm'for connectorsat mid-span and 7 11120 I 2411 I 2.70 Similarly. Uncontrolled Copy. positioned in accordance with 7.2 Detailsof theproblem. of trains typo.4. the D. D. Hence.4.3. Assuming. 1120 6027 6.8 N/mm2 for connectors at mid-span.2.3. for bar or channel connectors.3 may be expressed as : light 29.74 3 I280 9786 4 I203 6453 1. These spectra will cover most accordancewith Part 5) in a normal density reinforced traffic of this type running on lines in Europe. The numbers under consideration is similar to that of a single simply of each type of train assumed for the standard spectra.2. To investigatethe fatigue types. A 30 m simply supported composite highway bridgethat carries a dual carriageway all-purpose E. Report ORE D128/RP5 The potential fatigue crack should be classified as S (from Report ORE D128/RP6 type3.05 1 allowable shear load per stud under loading from the standardfatigue vehicle. ends of the member.3 = 52N/mmZ Traffic Train Trrln Numbor Total annual tonnage.12oftable17(c). may be expressed as: *See fiaure 19. Train type.4. the limiting value of b p max may be determined as : 13. 29. for Horvy preliminarydesign purposes in this particular type of example it may be noted that: at mid-span up max will equal OP min at the ends bp min will equal zero.00 with 7.4.1 0 where 4 172 (47093 8.2.l RU loading. is the effective weld throat area in mm2for the particulartype of connector. supported girder. The limited to 37.4.3 Clessification. © BSI 0.4.4.4 Exampleof a typical procedure for shear Appendix E connectors D e r i v a t i o n of standard railway load spectra D. obtained from 6. are connectors at mid-span. 'For further information on the derivation of the spectra.3 = 59.3. tonnrs par annum tonnes x 10' The above values may be checked against uvmax as described in 8.2 or. The load spectra given in table 3 have been that the shapeof the shear force influence line of the girder based on the typical trains shown in figure 20.3(c)).3 for the perannum tonneax 10' assessment of shear connectors as the damage chart (see figure 10) does not include factors for class S details.75 8 Thus for stud connectors (see 6. 34 I I .9 Pu/425 = 0.

6 11.5 1.5 1. 3.6 1.1 2.5 2.1Ly-+$.9 l#u 1.6 2.6 2.6 2.6 2.7 3.5 5.5 2.6 $4 t t - 3. © BSI 2.62.2 total load = 344 t Southern Region suburban e 2.5 1.5 2.5 2.9 2.7 k+ 3.5 5.2 t 11.6 3L# 11.+#-+ 3.3 6.7 11.5 2.6 m 3.5 I t 2.4 4 1 - 2.0 11.2 11.6 +++++ +2.5 1.5 V = 145 km/h 2.$3.5 2.6 +t----H 4 2.6 k2.5 2.5 2.2 3.1 21 1.6 +.6 2.3 V = 145 km/h ++ 22 p 3.--IoL m total load = 172 t i1 11. Uncontrolled Copy.52 Q L r eF++F=% 2.6 i m 3.7 11.5 - 11.7 V = 145 km/h total load = 372 t 2 Electric multiple unit V = 160 km/h m .6 2.7+-3c7(L--$+++--.6 2.5 1.5 21 L .2 6 Electric hauled passenger train total load = 572 t Figure 19.5 l0vv b U b n* Y#**H---+++- 41.2 11.2 11.2 6.2 1.2 3 Southern Region suburban 2.5 2.2 11.6 V = 80 km/h total load = 1794 t 1 Steel train 11.O 3. Trains included in table 2 spectra 35 11.2 12 x L n10 .6 ---$-$-+.2 1.6 5 Diesel hauled passenger train 3.6 L-a-a-a-L-a-a-a Licensed copy:Parson Brinkerhoff. 25/02/2015. m total load = 600 t V = 160 km/h 2.BS 5400: Part 10: 1980 225 6.7 -. 3.5 1.

6 4 2. 14 Rue Jean-Ray F.L LIbb 11-7 4.1 4. In deriving the table 2 spectra. I 6.8 U P 1 q m L-B-B-R-B-B-B-B-~~B-B-B-B-B-S-R~B-B-R~B-0-B-B-S 9 Mixed freight total load = 852 t V = 120km/h NOTE.9 5.9 2.1 2.5 ‘ I 3. Uncontrolled Copy.5 6x20 12.0 2.I BS 5400: Part 10: 1980 9 .2 6.11 : : ~ 6.2 2 11.2 3.8 3.8 * v 1 1 v 6x20 .1 4.. t( t 1. impact effects were calculated in accordance with the recommendationsof Leaflet 776-1 R. 25/02/2015.8 YL.1 L.8 Lx20 1:: 12.2 36 22 +++-. (Concluded) 36 .6 21 26 VLI/ 4 4 1 1. 6.5 Licensed copy:Parson Brinkerhoff. © BSI 6x20 1:: ~~ 2.2.2 2 +%+:! 4. 75015 Paris.8 1.2 m 2. published by the International Union of Railways (UIC).L 2.5 m 2.1 V = 72 km/h for medium traffic and 120 km/h tor heavy traffic total load = 1120 t 7 Heavy freight 2.1 V = 72 km/h for medium traffic and 120 km/h for heavy traffic total load = 1120 t 8 Heavy freight 2.5 1.7 ” 1 2.7 &+ 1.8 !2n7:::..6 3. Figure 19.

5 2.4 4 4 3.1 1/1 4 4 A 4 3. 2. In deriving the table 3 spectra.3 2.21 total load = 253 t 2 6 4 4 A 4 total load = 246 t 1 Licensed copy:Parson Brinkerhoff.L z .3 4 A 4 2.4 2.3 2.3 w v 4 4 Lx 10.4 A I 4 4 4 4 8.3 8.3 2.13 7.1 3.3 2.0 2.5 2.0 2.LV 7. © BSI 8.4 2.2 Lx8.13 L 9.4 v v 4 4 A A U 1 4 1 7.9 2.65 2.3 total load = 209 t L x 9.3 2. an impact of 3 0 %was taken for all trains and all spans Figure 20.3 LxlO.4.4 2.2 3.2.2 3.2 A A A A 1 4 4 4 32 3.3 2.3 4 4 4 3.4 2.3 n 4 A 3. 2. Uncontrolled Copy.0 2. 25/02/2015.0 2.65 7.3 2.3 4 4 flfl 3.3U 8.4 2.2 A 3.3 2.0 2.4 8.74 Lx8.74 Lx8.3 2.7L A A / r 4 3.2 NO 9? 7.t 8.3 7.5 2.5 2 3 .4 3.9 2.2 4 /I total load = 231 t NOTE.3 2.0 2.3 8.08 2.3 8.4 2.3 8.08 4 4 4 4 3.9 $4 /In .2 *rn rn- %.4 7.08 7.3 7.4 7.7 ’ 8.L 4 4 4 4 8.L 7.L V A A 4 A 3.3 2.2 A 4 4 3.3 7. 4 8.5 2.2 I31 In. Trains included in table 3 spectra.3 L x 9.3 U A A A 3.5 2.3 2.3 1 A Lx8. 37 v u 3.0 4 4 4 4 4 4 4 4 3. t b4 2.3 2.O N 90 1 3.0 2.3 L x 10.13 8.3 2.3 5 L x8.9 2.4V 8.1v 2.4 2.L 4 4 4 4 3.3 2.9 2 .9 2.

... This shows the midspan bending of simply supported spans loaded by the standard fatigue vehicle and illustratesthe effect of variations in L.. ..8 I 1. .. ... 25/02/2015.6 5 2 / otl ot 1 Stress histories Cycle counting diagrams 38 ... ..84 Licensed copy:Parson Brinkerhoff.8 m Standard fatigue vehicle < 1.' .. 4at80kN 4-i... © BSI Influence line diagram L 6m J.. . . .'. r I ..6 /-\<'<n ..'. Qv2 6 < L < 15. . . .. Highway bridge. e". ...'... .- -w--y--j . . .... ...../...'.'... . .. .. .. Uncontrolled Copy. t.. ... . .... . .Appendix F Examples of stress histories and cycle counting procedure Example 1.8 < L < 3..4I -- I 1. .'.

-U m .-cc C a 0 0 -Q Y U Licensed copy:Parson Brinkerhoff.-C U r U) .-c U) + ai 0 ‘0 L n 39 $ . n U C m c U al B- U) Q 5 > n U Q U m -0C \ \ 5 ‘\ Q \ I n \ U) \ 3 0 3 . © BSI - b’ Em 0) 5 >.-cC m :: C 8C U 0) zn CI m e n E 3 U) - Q) . 25/02/2015.0 .- FI z e X m U) u- E 0 r- WJ m 0 C c C E 9) n m n x? E m L m Q c U) c Q 5 : c U) .( x / \ \ N -b’ c E em CO 23’ . Uncontrolled Copy.

© BSI L = base length of loop containing largest ordinate (measured i n direction of travel) Loading diagram for upmin Loading diagram for uDml. Since the analogy dependssolely on the depth of water retained in each section. . 40 . 4 at 250 k N 80 kNlm 80 k N l m RU Loading to be multiplied by the dynamic factor Influence line diagram Licensed copy:Parson Brinkerhoff. follow the same profile as the stress history.BS 5400: Part 10: 1980 Exemple4. r . 25/02/2015.. / / / . the cycle counting diagrams. for comparative purposes. This shows the mid-span bending of a three-span continuous beam loaded with standard RU loading. Loading diagrams for stress history 2 3 1 Stress history -. Uncontrolled Copy.. Railway bridge. - t pR 3 Cycle counting diagram NOTE. it is immaterial whether the profiles are as illustrated or with successive peaks and troughs joined by straight lines (as proposed in appendix 6 ) . . In examples 1 to 4 given above..

and failure initiates on the surface near the boundary of the compressionring due to ‘fretting’ under repeatedstrain.Licensed copy:Parson Brinkerhoff. non-welded details. 17(b) and 17(c).4.2 Procedure.3for bar and channel connectors. 6. thestress concentrations inherent in the make-up of a welded joint have been taken into account in the classification of the detail.7%probabilityofsurviving 106and 107 repetitions of stress ranges of 146 Nlrnm’ and 82 N/mm2 respectively.-N relationship for the detail (i. 6.1 Notes onpotentialmodes of failure.2to havea97.l General H. This appendix gives backgroundinformation on thedetail classificationsgiven in tables 17(a). © BSI - - H Appendix G Appendix Testing of shear connectors Explanatory notes on detail classification G.S.2 for stud connectors or 6. fatigue cracks normally initiate either at surface irregularities. the resulting stress concentrations should be determined either by special analysis or by the use of the stress concentration factors given in figure 22. Ensurealso that the maximum load on any connector does not exceed 0. in accordancewith clause 6. H.2 Geometricalstress concentration factors. as required by 6. a t the time of testing. Where these conditions are not satisfied.4. failure generally initiates at the edgeof the hole and propagates across the net section. such as a change of crosssection or an aperture (see figure 21 ) and/or where indicated in table 17.3 Class S criteria. 2. but in double covered joints made with H.F. or derived from strain gauge readings. This appendix outlines the procedures which should be followed if the fatigue strength of shear connectors is to be determinedby testing.l Scope. which is holed and connected with rivets or bolts. This includes notes on the potential modes of failure. To enable the weld metal attaching ’ shear connectors to be classified as a class S detail. H.Test the specimens under constant amplitude loading at frequencies not exceeding 250 cycles/min.2Type 1classifications.4. See table 17 (a).2. H. In unwelded steel.e. the . However.at corners of the cross sections.5 times the nominal static strength of the connector (determined in accordancewith Part 5) with the appropriateconcrete strength.4 should be used to assess the fatigue life. 25/02/2015.l. Ensure that the frequency of the applied loading isthe same for each specimen within a particular series of tests. In steel.where the stress ranges are computed in accordance with either 6. Stresses may either be determinedfrom the applied test load. Typical example of stress concentrations due t o geometrical discontinuity 41 . being determined in accordancewith the requirementsof BS 1881. where there is a geometricaldiscontinuity.e n t r a n t corner I The design stress for either location should be taken as the stress on the net section multiplied by the stress concentration factor Figure 21. at holes and re-entrant corners or at the root of the thread for bolts or screwed rods.1.3% probability of failure) should be derived in accordance with appendix A and the method given in 8. the design a.l General.G. H . nl P o t e n t i a l crack locations Welded attachment / A ypical stress distribution The design stress is applied to the appropriate plain material classification At the attachment thedesign stress is applied t o the appropriate joint classification / Manhole or r e . important factors influencing the class of each detail type and some guidance on selectionfor design. bolts this is eliminated by the pretensioning. Unlessotherwise indicated in table 17.providing joint slip is avoided. welds should be shown by tests carried out in accordance withG. Uncontrolled Copy.1.

Uncontrolled Copy. %v (b) Fatigue stress concentration factor for re-entrant corners KRc (based on net stress at X) Figure 22. 25/02/2015. Stress concentration factors 42 . © BSI BS 5400: Part 10: 1980 (a) Fatigue stress concentration factor for unreinforced apertures KUA (based on net stress at X) Length of st raight22r K H RC W Stress fluctuation 0.Licensed copy:Parson Brinkerhoff.

The controlled flame cutting procedure should stressing.1 l).Issue 2. Short a t t a c h m e n t s Long a t t a c h m e n t A f > Weld failure cracksltype. but such componentsshould be subject to special test and inspection procedures. joints). The presence of an aperture or re-entrant from prying action. In welded construction.3 C o m m e n t s onparticuler d e t a i l types The increase in tension will rarely exceed 10 %of an Type 7. stress fluctuations will be governed by the elasticity of the pre-compressed elements. fatigue cracks will normally initiate at the weld ensure that the resulting surface hardness is not suff icient to ends.2 Genera/comments. Where bolts or screwed rods are pre-tensioned to a value in excess of an applied external load. where the load is eccentric.6 to 2. (See figure 24.4. or if these are not present.Awayfrom Type 7. 3 Type 2 classifications. figure 22). where secondary out-of-plane bending of the joint is With the weld reinforcement dressed flush. but when the weld is transverse. for example. w e l d e d details on surface design stress should be the stress on the net section o f member.1 Notes O n p O t 8 n t i 8 l m O d 8 S o f fei/ure. Figure 23. a criterion exists (types 2. In either case the cracks will then Type 7. This covers connectionsdesigned in accordance weld ends.e.3end 7.3.All visible signs of drag lines should be removed external load applied concentrically with the bolt axis. the relativesize of the 'stressed element' and the 'attachment' is not a criterion. For classificationpurposes. Failure modes at weld ends (a) Y Crack t y p e s 2.3. However.1 Edge disfence. Cracks in stressed weld metal of Or.6. This classification applies to failure at the root of H. When the weld is essentially parallel to the direction of Type 1. a t weld surface ripples. 25/02/2015. waisted shanksand thread run-out in components.1 0) to limit the not covered by an appropriate British Standard. failure tends to restrained or does not occur (i.9 or 2. Forthe useof black boltscomplying with the requirements of BS 41 90 and subjected to fluctuating tensile loads.5.2 Generelcomments. 2.11 i f o n e d g e ) Avoid or g r i n d out to a s m o o t h profile.5. Type 7. undercut. A higher unwelded corners as a result of.3.1 to 2. fatigue cracks normally initiate at stop-start with Part 3 for slip resistance a t the ultimate limit state and positions.3. cracks in parent metal may also transference of stress from the main member in the direction initiate from the weld root. double-covered symmetric be associated with weld defects. see 6. Seetable 17(b). cracking will initiate cause cracking.) No edge distance the thread in normal commercial quality threaded criterion exists for continuous or regularly intermittent welds components.2.March 1999 BS 5400: Part 10: 1980 material which has previously been fully heat-treated. © BSI H. at the weld toes.Seefigure23.7 or 2 8 (or 2. Edge distance 0 BSI 03-1999 43 . weld fatigue resistance can be obtained with a rolled thread on Licensed copy:Parson Brinkerhoff. fatigue failure will rarely occur in a region of unwelded material since the fatigue strength of the welded joints will usually be much lower. H. For attachments attaching light bracing members where there is negligible connected by single welds. a further increase will result Types 1. Apart from the particular dimensional requirementsgiven for each type in table 17(b). an 'attachment'shouldbe taken as the adjacent structural element connected by welding to the stressed element under consideration.72. corner implies the existence of a stress concentration and the H . H. Uncontrolled Copy.11 a t edge) NOTE.2. any undercut t o these (to toe of weld) Figure 24. willinitiatefrom the weldroot (seetype3.6. multiplied by the relevant stress concentration factor (see H .11) 1 !\ w Crack types 2.10 ( o r 2. fillets.2. to ensure possibility of local stress concentrations occurring at that they have satisfactory fatigue resistance.3.4.5). but from the flame cut edge by grinding or machining. This type may be deemed to include bolt holes for propagate into the stressed element. Failure initiates by fretting in front of the hole. Attention should be paid to the details of head away from the ends of an attachment (see types 2.

Type 2.2.4). In the case of butt welds made on a permanent backing.3). is not recommended.70.BS5400:PartlO:1980 cause local buckling (see Part 3).7. H. If the reinforcement of a butt weld is dressed flush. J cross brace and a gusset. cover plates or box members (see figure 24 (b) and (c)). It should be noted that transverse butt welds on backing strips may be downgraded by tack welds close to their ends (see H. Type 2.3. fatigue cracks initiate at the weld metal-strip junction and then propagate into the weld metal. hence the criterion given in this part will automatically be met. H. in considering the effects on the stressed element.3. The fillets exclude the possibility of an increase in stress concentration arising at an acute re-entrant angle between the element surface and the toe of the weld. care should be taken to avoid undercut on element corners or to grind it out to a smooth profile should it occur. transfer of a part of the load in the element to and from the attachment will occur through welds adjacent to its ends . need not be considered.2.4 Type 3 classifications.)With the ends of butt welds machined flush with the plate edges.4 Weldforms. or as otherwise given below. Seetable 17(c).3 Stress concentrations. This type includes parent metal adjacent to the ends of flange cover plates regardless of the shape of the ends. Licensed copy:Parson Brinkerhoff. Repair to the standard of a C classification should be the subject of specialist advice and inspection and should not be undertaken in bridgeworks.2Attachment ofpermanent backing strips. The N.3. as in the case of a welded lap type connection between. These welds and those attaching the backing strip should also comply with the relevant class requirements. (c) such load transfer is through joints which are not symmetrical about both axes of cross section of the stressed element. H.3. The significance of defects should be determined with the aid of specialist advice and/or by the use of a fracture mechanicsanalysis.2. The classification will reduce to E or F (type 3. the stress i n the gusset at the end of the cross brace will vary substantially across the section.7 1 . 25/02/2015.3. surface porosity in the dressing area (see H. 44 .3. it is equally important t o ensurethat no accidental undercutting occurs on the unwelded corners of.H. NOTE.. .4. These are increased. H. This type applies regardless of the shape of the end of the attachment. Full or partial penetration butt welded jointsof Tform (such as would connect attachmentsto the surface of a stressed element) should be completed by fillet welds of leg length at least equal to 25 %of the thickness of the attachment. © BSI Part 5 recommends the provision of a minimum edge distance of 25 mm forshear connectors.Where a narrow attachment is transferring the entire load out of a wide member. (See figure 26. H .7and2. where : (a) the weld ends or toes are on. it shouldsubsequently be ground out to a smooth profile.3.3.This applies where any applied shear stress range is (numerically) greater than 50 % of a co-existent applied direct stress range.The classification may be deemed to include stress concentrations arising trom normal eccentricities in the thickness direction. so that the fatigue strength depends largely upon the toe profile of the weld. Long gaps between intermittent fillet welds are not recommended as they increase the risk of corrosion and. Accidental stop-starts are not uncommon in automatic processes.4. and thus. type 3. w e l d e d details a t end connections of member.1 N o t e s o n p o t e n t i a l m o d e s o f failure. Tack welds. For failure in the cross brace at X the cross brace is the 'member' and the gusset is the'attachment'. For assessing the stress in the gusset the effective width should be taken as shown in figure 25.2. (b) the attachment is 'long' in the direction of stressing.The limiting gap ratio m / h applies even though adjacent welds may be on opposite sides of a narrow attachment (as in the case of a longitudinal stiffener with staggered fillet welds).4) at any butt welds in the backing strip or class E at any permanent tack weld (see H.D. for example. fatigue cracks in the as-welded condition normally initiate at the weld toe and propagate into the parent metal. to which they are attached. technique should be selected with a view to ensuring the detection of such significant defects. H.4). and as a result. Type 2. If a permanent backing strip is used in making longitudinal butt weldedjointsitshould becontinuousor made continuous by welding. Effective width for wide lap connections Type 2.3. I Finish machining should be in the direction of Or.5 Tack welds. Types2.4.5 hare required the class should be reduced to F. or near. Whereit does occur. Figure 25. for example. Type 2.2. failure can occur in the weld material if minor weld defects are exposed. it is immaterial whether the attachment is fillet or butt welded to the surface.1) . type 3.3 or 3. e. This type also includes tack welds to the edges of longitudinal backing strips irrespective of spacing.g. Their use in the fabrication process should be strictly controlled. provided that the welds comply in all respects with the workmanship requirements for permanent weldsand that any undercut on the backing strip is ground smooth. type 2. Uncontrolled Copy. If intermediategaps longer than 2. the use of principal stress values may be conservative and accordingly the classification is upgraded. Apart from the width transverse to U. since a similar toe profile results in both cases. in the caseof compression members.3 Comments onparticular detail types Type 2. In all cases. In fillet or partial penetration butt welds. weld returns across a corner should be avoided and the use of cover plates wider than the flange. will provide potential crack locationssimilar to any other weld end. In particular. and hence the fatigue strength is reduced. may NOTE. The effects of tack welds which are subsequently fully ground out or incorporated into the butt weld by fusion. Experimental evidence indicates that where significant shear stress co-exists with direct stress.neither the shape of the end of an attachment nor the orientationor continuity of the weld at itsend affects the class.4.8. This type is only recommended for use in bridgeworks in exceptional circumstances.3. an unwelded corner of theelement (see H.2. Although this criterion can be specified only for the'width' direction of an element. fatigue cracks in weld metal will normally initiate from the weld root. unless carefully ground out or buried in a subsequent run.T. spatter and excessive leg size at stop-start positions or accidental overweave in manual welding.

7 and 3.4. tend to have a better reinforcement shape from the point of view of fatigue than positional. -U or -V forms) are back-gouged to a total width at least equal to half the thickness of the thinner element. H.T.11 3. © BSI ir U e--+ NOTE.9 Licensed copy:Parson Brinkerhoff. This class should not normally be used in bridgeworks (see 5. Type 3.2 3.2. Accordingly.BS 5400 : Part 10 : 1980 - 3. single bevel -J.6. For components tapered in thickness. Bun type welds may also occur within the length of a member or individual plate as. 25/02/2015. Types3. any undercut should be ground out to a smooth profile. H. Wheresuch support is not provided (e. wide plates supported by effectively continuous stiffeners.3. Although such geometries have not been given specific categories in table 17(c).4. i. Welding of reinforcement should comply with Parts 7 and 8.3 Comments on particular detail types Type 3. types 3. their use is not recommended unless subject to special tests and strict procedural control. In these laner cases. the total stress fluctuation should be considered to be transmitted through the welds (e. The N.2 Design stresses. Should it occur. Accordingly.4. For elements where out-of-plane bending is resisted by contiguous construction (e. See also H.6 Partialpenetration butt welds. In allcases.2. provided that the root sides of joints with single sided preparations (i.7 Welding of reinforcing bars for concrete. in the case of : (a) a plug weld to fill a small hole.3 3.Grinding smooth the reinforcement of butt welds until flush with the plate surface on both sides is generally 45 . for fatigue purposes.3 and 3. All butt welds transmitting stress between ends of plates. H.4 may be deemed to cover plug and infill plate welds and types 3. H . Shop welds made entirely in the downhand position.7.2.4.4. the centre of the untapered section should be used.11 3.D.3 3. (c) a hole or slot for a transverse member to slot through a wider member. Unless made on a permanent backing (type 3.g.g. Lap welding of bars is not classified since adequate control cannot be exercised over the profile of the root beads and its use is not recommended under fatigue conditions.failures tend to be associated with plate edgesand careshould be taken to avoid undercut at the weld toes on the corners of the cross section of the stressed element (or on the edge at the toes of any return welds).2. technique should be selected with a view to ensuring the detection of such significant defects.2.2.e. site or submerged arc welds (i.2.4. except. sections or built-up members in bridgeworks should be full penetration. joints made in this manner may be up-graded to class D.e. Type 3. Type 3 failure modes H. Figure 26.1).2.) eccentricities due to axial misalignments in the thickness direction may be neglected.The classifications may be deemed to include for the effects of any accidental misalignments up to the maximum value specified in Part 6. H . column caps and bearing stiffeners).5 Joints weldedfrorn one side only. Uncontrolled Copy.2 or 3.where permitted in types 3.2.4. Fatiguecracks in reinforcingbars will normally initiate in similar locationsto those for structuraljoints.3. See the note to type 3. given similar stress conditionsand joint geometry.8 may be deemed to cover sloned members. larger re-entrant angles at the toes and more uniform profiles). even though the joint may be required to carry wholly compressive stressesand the non-penetrated surfaces may be machined to fit. tension links) the design stress should include an allowance for the bending effects of any intentional misalignment.2 Generalcomments H. the nominal distance between the centres of thicknessof the two abutting components.3Elernentedges. for example.4) welds made entirely from one side are not classified since the root profile will be dependent upon the welding procedures adopted.1 0 (junctions with transverse members).3.4. H -4.4.5). to cater for exceptional circumstances. beam flanges supported by webs. H.2and3.8 and 3.1. The significance of defects should be determined with the aid of specialist advice and/or by the use of a fracture mechanics analysis.g.2. These types do not normally include joints between rolled or built-up sections. either manually or by an automatic process other than submerged arc.e. etc.1 Misalignments.4.4 Part width welds. (b) a weld closing a temporaryaccess hole with an infill plate .Thickness variations and surface misalignments up to the maximum valuesspecified in Part 6 may be deemed to be included (see H.

in the region of the web/flange junction (see f igure 27).9)the detail class will not be reduced below class F unless permanently tacked within 10 mm of the member edge.9. Plane sections may be assumed to remain plane where the main member stress can be continued through the transverse member by additional continuity plating of comparable cross-sectional area. the stress concentration factor may have to be determined by finite element or model analysis.D.2. the stress concentration factor due to the abrupt change of width should be used (see figure 29). Provided that N. a slot in A avoids any risk from lamellar tears. as in the case of flanges at the junction of two girders.8 joint 46 .11). in which case it will be class G (type 2. If B is critical and A is not.4 providedthat the end of the butt weld and the reinforcement within a distanceequal to the radius ( I )are ground flush.The slot or hole dimensions should be in accordance with appendix A of BS 51 35 : 1974. when possible.T.In thistypeof connection it is important that the joint regions of the third member are checked before welding for lameller rolling defects and after welding for lamellar tears. if welded onto permanent backing material.Note thatthis detail should generally be avoided.Butt welds between rolled sections or between built-up sections are prone to weld defects.11 (class G) at point Y. Types3. Uncontrolled Copy. Plug welds should not be used in bridgeworks for transmitting tensile force across two lapping plates. Type 3. which is in line with the main member components (see figure 28). The effect of the stress concentration at thecorner of the joint between two individual plates of different widths in line may be included inthe classification. The welds should be full penetration and should be considered to be equivalent to type 3. Such holes may also be required for stitching laminations or repairing lamellar tears.7and3. the third member should be assumed to transmit the stress which the parent material would have carried before the slot was cut. 25/02/2015. However. which are difficult to detect. Where member B iscalled upon to carry high tensile stress.4). Type 3.Thisgives improvedaccessto the flange butt welds when webs or longitudinalstiffenershave already been attached. weld is made on a permanent backing (type 3. failure through the weld throat should be considered to be class W. see figure 27).BS 5400 : Part 10 : 1980 Licensed copy:Parson Brinkerhoff. a discontinuity in the main memberstress pattern will occur. This joint is frequently made using a semi-circular cope hole. Figure 27. Where the end of one plate is butt welded to the side of another. Where the third member is an open shape. circular cut-outs at the corners of B will improve the class to F (type 2. Type 3. refer to type 3.8. for example.If the backing strip is fillet or tack welded to the plate (type 2.In thiscase.If the weld is a full penetration butt carried out in accordance with all the recommendations for type 3. These types may be used for holes which are either filled with plugs of weld metal or welded infill plates. but where they have to be used. Stress concentrations due to abrupt changes of width can often be avoided by tapering the wider plate (see types 3. Where the third member is a plate it may be assumed that plane sections remain plane in the main members and that axial and bending stress distribution in the or direction are unaffected. These types may be deemed to cover the case where a narrow third member is slotted through a single main member away from an end connection (see figure 30).5. particularly i f different in width. In theabsence of published data on a particular joint configuration. type 3.5 the detail may be classed as F2 without applying a stress concentration factor. In this case the stress parameter should be the peak stress Concentration at the joint.Special preparations. The end of the web butt weld at the cope hole can be consideredto be equivalentto class D with a stress concentration factor of 2. Type 3. 3. with respect to stress fluctuation in member B the detail shown in figure 30 is type 2. this treatment can be assumed to raise the class to D. Types3. Weld metal failure will not govern with full penetration welds.4. based on the minimum throat area projected in the or direction. an I section or a hollow tube. procedures and inspection may be undertaken in exceptional circumstances and type 3.4.6. as slots are difficult to cut accurately and fit-up for welding is often poor. Dressing of the weld reinforcement isadvised to overcome poor reinforcement shape resulting from the greater misalignments which may occur in the jointing of sect ions.3may then be applied unless the Where t w o flat plates intersect in the same plane.4.3 or. Their use for transmitting shear force is not recommended for major structural connections. NOTE.7 (full penetration buttjoints) should be reclassed from F to F2. If the length of the slot is longer than 150 mm in the ar direction. is done after grinding. Cope holes of 45O mitre are not recommended.4.9). type 3. © BSI beneficial.3and3.3and 3.

Useof continuity plating to reducestress concentrationsin type3. 25/02/2015. Example of a'third' member slotted through a main member 47 .7 A Alternative detail if member '8' is critical n NOTE. 'third' rnernber'8' is continuous and is welded all round the slot to'A. Uncontrolled Copy. Figure29. Main rnernber'A issloned. Cruciform junction between flange plates cut out Class F type 3. Figure30.7 and 3. © BSI n 0.8 joints Licensed copy:Parson Brinkerhoff.A Typical stress patterns thickness direct ion Detail X (cruciform joint) \pia t i n g Figure28.

.10 joint Full p e n e t r a t i o n b u t t weld I concentration f a c t o r ? Flgure32.8).8. Type3. are not recommended unlessspecialist advice is sought. The reference to 'end connections' in the title of table 17(c) refers to the end of the member in which failure occurs. Type3.e. diagonals and truss chords. Note that in the case of trusses. or tee or cruciform joints are welded from both sides (asshown in table 17(c)). secondary stresses due to joint fixity should be taken into account. Example o f type3. .5.1 1) will normally govern. Types3. but expert advice should be sought if a higher strength is required. by tearing of the attached flange or in the body of the connector. As far as fatigue failure of the transverse member is concerned. . This can occur in the case of a junction between cross girders and main girders.7 and 3. Whereaxial stress i s also present. 40 Stress distribution \ 0. This type covers embedded shear connectors a t any position along a girder.7 and 3.7 and 3. weld metal failure (see type 3. In this case an axial component in the first member will induce bending moment and hence curvature in the transverse member. Failure from ripples or stop-start positions on the face may give a higher strength than class W. unless the total weld leg length is about twice the element thickness. Members with bolted end connections via transversely welded end plates are particularly susceptible to localincreaseofstress (seefigure31).12.Thesedetail types are distinguished from types 3. in thiscase the welded end of the shear connector. In afillet welded joint. with or without the use of backing material. © BSI I BS 5400:Part 10: 1980 Type3. Alternative method o f joining t w o flange plates Figure=.- if I I 4--b Figure 31. If the transverse member is relatively stiff (i. 25/02/2015.9 or 3. the stress range at the face of the weld may be different from that at the root.tO. the first member is treated as a type 2 attachment and thestress parameter is the stress in the transverse member without the application of a stress concentration factor. single Sided manual metal arc procedures. .11. which if subject to loading that distorts the cross section. Type3. Notethat if the transverse member is an open or hollow section. Unless the latter is very stiff in bending an uneven stress distribution will result.Thestressratioand effective weld sizecriteria of this clause are intended to aid in the exclusion of premature failures by local crushing of the concrete. Where lapped joints are welded on t w o or more sides. Weld metal failure need not be considered.5 times the width of the first member) and a full penetration butt weld is used in accordance with the recommendations ot type 3. This applies particularly to small hollow members with welded corners. Tee junction o f t w o flange plates Figure 33. or in vierendeel frames (see figure 32). the classification in this particular case may be considered to be eftectively F2 with a stress concentration factor of unity.9end3. In hollow or open transverse members this stress is often magnified by local bending of the walls.8).8 by the absence of a similar member in line on the far side of the joint. its width isat least 1. Uncontrolled Copy. Often the load is transmitted from a member to a transverse member primarily via flange plates in the same plane...Licensed copy:Parson Brinkerhoff.73. It wilt also govern in a partial penetration butt welded joint except where reinforced with fillet welds of adequate size. Class W is primarily intended to apply to all fillet or partial penetration butt weld joints where bending action across the throat does not occur. may cause failure of the corner weld in bending (see figure 34). Corner detail Single fillet corner weld in bending . local bending will increase the peakstress further (as in the case of types 3. Otherwise the class shall be F with the appropriate stress concentration factor (see comment on detail types 3. In most cases failure from stress fluctuation in this root will be critical and thisshould always be classified as W. In certain cases difficulty of access may only allow welding to be done on one side of the joint. The fatigue strength of both flange plates may be improved by the insertion of a smoothly radiused gusset plate in the transverse member so that all butt welds are well away from re-entrant corners (see figure 33). such bending action is normally prevented.

Licensed copy:Parson Brinkerhoff. 25/02/2015. P . . © BSI \ \ . . Uncontrolled Copy.

© BSI .Licensed copy:Parson Brinkerhoff. Uncontrolled Copy. 25/02/2015.

Uncontrolled Copy. 25/02/2015.Licensed copy:Parson Brinkerhoff. © BSI .

n Standards publications referred to BS 1881 BS 3643 BS 3692 BS 419O 0s 4395 BS 4604 BS 5135 BS 5400 Licensed copy:Parson Brinkerhoff. Uncontrolled Copy. screws and nuu High strength friction grip bolts and associated nuts and washers for structural engineering The use of high strength friction grip bolts in structural steelwork. © BSI n Methods of testing concrete IS0 metric screw threads I S 0 metric precision hexagon bolts. screws and nuts I S 0 metric black hexagon bolts. 25/02/2015. Metric series Metal-arc welding of carbon and carbon manganese steels Stwl concrete and composite bridges .

in the UK. If permission is granted. No. Ifthese detaik are to be used for any other purpose than implementationthen the prior written permissionof BSI must be obtained. 25/02/2015. © BSI It is the anstant aim of BSI to improve the quality of our products and secv(ces. it is BSI policy to supply the BSI implementationof those that have been published as British Standards. Subscribing members of BSI are kept up to date with standards developments and receive substantial discounts on the purchase price of standards. Committees responsible for this British Standard The Civil Engineering and Building Structures Standards Committee.We would be grateful if anyone finding an inaccuracy or ambiguity while using this British Standard would inform the Seaetary d the responsible technical committee. Uncontrolled Copy. was published under the authority of the Executive Board and comes i n t o effect o n 3 1 January 1980 In response to orders far international standards. stored in a retrieval system or transmitted In any form or by any means electronic. recording or otherwise without prior written permlssion from BSI. Details and advice can be obtained from the Copyright Manager. in the course of implementing the standard. unless othecwise requested. and size. Licensed copy:Parson Brinkerhoff. Buylng standards Orders for all BSI. Fax: 0181 996 7400. Scottish Development Department Welding Institute Amendments issued since publication Amd. Rwloirlonr British Standards are updated by amendment or revision.It presents the UK review on standards in E u r o ~ eand at the htematiial level. British Railways Board British Steel Industry Cement and Concrete Association Concrete Society Limited Constructional Steel Research and Development Organization Department of the Environment (Building Research Establishment) Department o f the Environment (Transport and Road Research Laboratory1 Department of Transport Federation of Civil Engineering Contractors Greater London Council Institution of Civil Engineers Institution of Highway Engineers Institution of Municipal Engineers Institution of Structural Engineers London Transport Executive Ministry of Defence Sand and Gravel Association Lid. international and foreign standards publications should be addressed to Customer Services. ISBN 0 580 10567 9 The following BSI references relate t o rhe work on this standard : Committee reference CSB/30 Draft for comment 74/13197 DC BSI b the independantnationalbody mponsible for preparing Bdtbh Standards. 8 BSI 03-1999 Information on standards BSI provides a wide range of information on national. For details of these and other benefits contact MembershipAdministration. - - This does not preclude the free use. having been prepared under the direction of the Civil Engineering and Building Structures Standards Committee. consists o f representatives from the following Government departments and scientific and industrial organizations: Association of Consulting Engineers Association of County Councils British Constructional Steelwork Association British Precart Concrete Federation Ltd. Fax: 0181 996 7001. Tel: 0181 996 7070. type 01 grade designations. under whose direction this British Standard war prepared.of necessary details such as symbols. Tel: 0181 996 7111: Fax: 0181 996 7048. Contact the Information Centre. BSI offers members an individual updating service called PLUS which ensures that wbsaibers automaticallyreceive the latest ediths of standards. the terms may include royalty payments or a licensing agreement.BS 5400 : Part 10 : 1980 issue 2. Users of 8ritbh Standards should make sure that they possess the latest amendments or editions. Tel: 0181 996 7002. Copyrlght Copyright subsists in all BSI publications. photocopying. Designs and Patents Act 1988 no extract may be reproduced. European and international standards through ita Libmy and its Technical Help to Exporters Service. the identity of which can be found on the inside front cover.Tel: 0181 996 9OOO. BSI also holds the copyright. It is hcorporeted by Royal Charter. of the publications of the international standardizationbodles. Tel: 0181 996 7000: Fax: 0181 996 7001. March 1999 This British Standard. Except as permitted under the Copyright. 9352 I Dateof issue I I March 1999 I Text affected Indicated by a sideline - British Standards InstiMion a 3 8 9 Chiswick High Road London W4 4AL Telephone 0181 996 9OOO Fax 0181 996 7400 .