See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/287182523

Effect of fatigue on ultimate load behavior of concrete bridge decks

Article · January 1971

CITATIONS READS

2 63

2 authors, including:

Gouranga Prasad Saha

Construma Consultancy Pvt Limited, Mumbai, India
23 PUBLICATIONS   12 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Strain softening in prestressed concrete beams. View project

All content following this page was uploaded by Gouranga Prasad Saha on 17 October 2016.

The user has requested enhancement of the downloaded file.

 PAPER SP 26-36
9

EFFEX::T OF FATIGUE ON ULT:rnATE LOAD BEHAVIOR OF CONCREI'E BRIDGE D.EX:KS

By F. SAWKO and G. P. SARA

SYNOPSIS

This paper presents the results o~ a series o~

tests on ~atigue o~ model prestressed and rei~orced
concrete bridge slabs o~ different span to width ratios
and di~~erent prestress levels in the transverse direc-
tion. Modern slender prestressed concrete decks, un-
like heavy over-designed rein~orced concrete sections,
· must be investigated more care:full.y than is done in
the classical yield line approach to the ultimate load
analysis o~ bridge decks to investigate their ultimate
load behavior.

Keywords: bridge decks; bridges ( stru.ctures); con-

crete slabs; crack propagation; cyelic loads; ~atigue
tests; models; moving loads; pre&tressed concrete;
rei~orced concrete; research.

ACI member, FELICJAN SAWKO obtained his BSc degree in civil engineering at
the University o~ Leeds, Great Britain in 1958, graduating with a ~irst class
honours degree. He stayed on at the University to study ~or his Master's degree.
In October 1959 he joined a ~irm of consulting engineers to obtain his prac-
tical experience. After 3 years he joined the Department o~ Civil Engineer-
ing, Leeds University as a lecturer in September 1962; in May 1967 he became a
reader in civil engineering and in October 1967 was appointed to the second
chair of civil engineering at the University of Liverpool.

His main research interest is the application of electronic computers to

structural analysis and design with particular re~erence to bridge structures.
He has published extensively in this field.

ACI member, G. P. SARA is a senior bridge engineer with Husband & Co .,

England. He graduated from Calcutta University in 1959. After having three
years experience in industry, mostly on rei~orced concrete bridges, he did
post graduate courses at Leeds University and obtained his MSc in June 1964 and
PhD in July 1966. His main subjects of research are ~atigue of concrete and
analysis o~ bridge decks.

942

•. ;

<ft.
FATIQUE EFFECTS ON BRIDGE DECKS 943

Introduction

The classical yield line approach (l) to the ultimate load analysis of floor
2 6
slabs, has in recent years been applied by several authors( - ) to various

bridge decks to investigate their ultimate load behaviour. This approach has -

not, however, been accepted by everybody as a basis of design and doubts

have been expressed on the validity of analysis under static loading for bridge

decks where vehicular loading is of predominant interest. Re cently Payne and

7
Caldwell ( ) commented, that "colla~se loads and failure mechanisms cannot
be a basis for highway bridge design".

The primary r eason for these doubts i s fatigue. Recent years have seen

considerable improvement in qualit)' control, leading to a lower load factor

employed in design. Whereas in the past heavy over designed reinforced

concrete sections were not affected by fatigue, a modern slender prestres sed

concrete deck subjected to high working stre s ses must be. investigated more

carefully.

Thi s pa per presents the results of a series of tests on fatigue of model

prestressed and reinfo rced concrete bridge s labs of different span to width

ratios , and different prestress l evels in the tran sve r se dire ction. Most of the

·.
 . ~,.:

944 CONCRETE BRIDGE DES IGN

slabs were cast in pairs, one being tested under fatigue loading followed by

static loading to fa ilure, while the other under short time static loading. A

direct comparison could, therefore, be made and conclusions drawn on the

effect of fatigue on the ultimat e load behaviour of bridge decks.

Design and Manufacture of Test Specimens

The dimensions and reinforcing of specimens were chosen with a view to

observipg the failure characteris_tics of both fatigued ·and non -fatigued slabs in

terms of failure patte rn and ultimate load for different span/width ratios and

for different amounts of transverse reinforcements. Full details are given in

Table 1 and Fig 1. These slabs were intended to be 1/10 scale models of

prototype bridge decks.

For the serie s of pre stressed concrete slabs, concrete mix of 1:1:2 with

Rapid-Hardening Portland cement and 3/ 8 inch aggregate was used. ln the

first pair (Slab number s lA and lB), a wat e r / cement ratio of 0.4 was used,

but since s ome honeycombing was noticed the water /cement r atio was increased

to 0.45 for sub sequent prestressed s labs. Plain cold drawn steel wire of 0 . 2

inch diameter having an ultimate strength of 3. 5T was used for pr estressing

a nd the wires were left unbonded . A PSC monowire jack was used for pre -

st r essing the wires . Prestress in t he transverse direction was chosen as

100%, 30% and O% of the longitudinal with a nominal amount of mild steel

reinforcement, as r ecomm ended by C . P . 114 (S), provided in the last case.

2
This con si st ed of 3/16 inch diameter ba r s with a yield s tress of 46000 lb/ in .

l" ... .:... ..

·FATIGUE EFF ECTS ON BRIDGE DECKS 945
Prestressing was provided axially in both directions . ill the longitudinal direction

wires were placed at the centre of the section and in the transverse direction

they were staggered top and bottom at the required intervals (Fig 1). In pre -

stres s ed concrete slabs hdes were left for the prestressing wires with 0. 3 inch

(outside diameter) p.v.c . duct s , a typical arrangement of which being shown in

Fig 2. Wires were tensioned to 2000 lbs each before casting to keep the ducts

st r a ight. Prestressing operation was started from the middle of the slab and

continued outwards in both directions to minimise the loss due to eccentric

prestressing. A prestress of 4250 lbs was applied to all the wires of both

longitudinal and tr~nsverse directions. The level of pres tress was measured

by directly reading the jack dial, measuring the extension of the wires and

recording the concrete strain at the points as shown in Fig 3. As the fricti on

between the slab and the wooden boa rd of the mould did not allow the full elastic

deformation of concrete due to prestress, the s lab was lifted afte r st ressing in

one dire ction to eliminate the friction effect. Due account was ta ken of the

Poisson 's r atio effect in calculating the required level of prestress .

For the r einforced conc r et e slabs , concrete mix of 1:1 !:3 with a wat e r /

cem e nt ratio of 0 . .5.5 was used . Mild ste e l bars of 3/8 inch diameter ha\·ing a
?
yi e ld stress of 46000 lb/ in- were provided in both directions.

Control specimens con si s ting of 9 - -± inch cubes, 3 - 4 inch x 4 inch

x 20 inch pris ms and 6 - 6 inch diamet e r cylinders were ca st for each s lab ,

a nd these were neve r separat ed from the test specimens to ensure identi cal

·.
..
·..: . --=----
 946 CONCRETE BRIDGE DESIGN

curing . Since th~ prestressed concrete slabs were quite thin, one 20 inch x 4

inch x 1 ~ inch specimen was cast for each slab for the use as dummy in the

electrical st rain measurement circuit.

Instrumentation
9
The arrangement of the loading was made t o represent an abnormal vehicle ( )

the contact area of the wheas and the axle spacings being scaled down in the same

ratio as the slabs (Fig 4). Bridges having a span greater than 49 feet accommodate

the whole of the abnormal vehicle for the maximum longitudinal bending moment

and so, for the slabs of span 7 feet 6 inche s and 5 feet all the four axles of

loading were applied. The jack load wp. s applied to the slab through a system

of spreaders an<;! cross spreaders. In the case of s labs of 3 feet 9 in. span,

only two axles of the scal ed abnormal vehicle were placed on the span to cause

maximum l ongitudinal bending moment. A three -point -ball-seating a rrangement

was used to avoid rigidity of the whol e system of the loa ding frame.

Some of the eccentrically loaded slabs wer e prevented f rom uplift at the

far corners and provision was made for fr ee rotation at the supports in the

direction of the span. This was achieved by placing a steel bar at each end on

the top of the s labs exactly above the support bar and by clamping this to the

bearing girder through a joist .

Deflection was measured at five equidi stant points ac r oss the transverse

section of ea ch slab. Dial gauges of 2 iqch travel r eading to 0 . 001 inch were

used to mea sure the static deflection.

·.
 948 CONCRETE BRIDGE DESIGN
.. tension in the case of prestressed concrete slabs and which does not exceed the
~
permissible worldng stresses in the case of r einforced concr ete slabs. In slab

number 38, first a load range of worldng load and the permissible minimum in
6
the machine was applied for 1. 93 x 10 cycles . The m achine was then stopped

and a static cycle was carried out to note the effect. The slab was then loaded

until a crack appear ed and on the cracked section the former load range was

applied. This was done to assimilate the cracldng caused by accidental over-

loading on an actual bridge deck. The repetition was continued for a further
6
0 . 79 x 10 cycles, followed by a static cycle of loading to note the effect. In

the third stage a load range of the cracking loa d a nd half of that value was

applied . The rate of increase of deflection was very high and the machine was

automatically stopped by the knock off switch of the pulsator unit. The slab

• was then tested to destruction by short term loading .

Slab numbers 4 and 5 were first tested statically up to the cracking load

and a dynamic load range with a maximum slightly below that causing cracldng

and the minimum equa l to half of thi s value was then applied . This load range

was maintained for nea rly 3 million cycle s , and was followed by s hort t e rm

loa ding to dest r uction . None of the slabs fail ed in fa tigue and a ll we r e sub -

sequently t ested to desquction under short t e rm loading . Deflection and strain

r eadings were taken at every load increm ent and the propagation of cracks was

not ed.

,.

-
 FATIGUE. EFF~CTS ON BRIDGE DECKS 949
Analysis and Discussion of Results

The results of deflections and propagation of cracks have been fully

discussed e l sewhere (lO). In the follov-ring only the results concerning the

ultimate load behaviour of the slabs a re dealt with.

To understand the effect of fatigue on the ultimate load behaviour of a

st ructure a knowledge of t he fatigue m e chani s m of c oncrete is essential. Although

. (11) .
the latter has been dea lt with by the authors elsewhere 1t was felt that a

pr esentation of some of the conclusions here might prove useful. Accordingly

before a nalysing the r esults of the ultimat e load fatigue mechanism of concr ete

is presented first.

Mechanism of Fatigue of Concrete

When a concret e structu"re is subjected to fatigue loading creep takes place

in the concrete in the direction of applied stress . Although the a ction of this

creep is simila r to that under susmined loading the m e chanism differs. Under

r epeated loading the individua l particles a r e subje cted to an a lte rnating load

which produces a kind of to and fro friction between them on a micros~opic

level. This is probably one of the r easons for the initial g r eat e r r ate of strain

in the concrete than i s obse rved with sustained loading.

With a prolonged cyclic compr essive or tensile loading a definite change

in volu m e of concrete i s noted. No appreciable lateral dilat ion occurs, the

deforma tion mainly occurring in the direction of the applied loading. At any

t ime within a concrete structure s crain is made up of three pa rt s :

•
·.
..
0
.A
'
~
R: 950 CONCRETE BRIDGE DESIGN
it • elastic, delayed e lastic and flow. The last part is not r ecoverable, so a change

of vol um e of concrete accompanied by a perma nent set causes a kind of solid

body compaction of the concrete leading to an increase in strength and stiffne ss

of the section. This inc rea se in strength and stiffness occurs if t he fatigue

loading does not exceed som e cirtica l value . This critical load i s dependent

on both the maximum and minimum values of the fatigue load r a nge . For a

minimum load of S% the maximum load of the fatigue load r ange i s probably

about SO% of the ultimate.

When the applied load exceeds the critical load range micro -cracking of

gel structure of cement past es t akes place a nd thi s together with the micro -

slipping of ihe cement pastes on the aggregat e surface causes a weakening of

..
the section. Th~ equivalent section properties of the fatigued sect ion of a pre.-
•
stressed beam thus l oaded i s shown in Fig 5 .

.
Tensile creep m ovement causes the particles to separate, thereby

reducing stiffness a nd strength all the time .

In the authors ' tests of slabs t he maximum l oad of the fatigue loading

was always below SO% of the ultimate a nd the minimum load was either half

the maximum or the minimum permissible in the machine . Thus it was

always below the critical load r ange. A typical load -deflection relationship

of a fatigued slab together with its non -fatigued companion is shown in Fig 6

which clearly indicates an increase in stiffness due to fatigue particularly in

the post elastic range .

... A
 [.

r 6
FATIGUE EFFECTS ON BRIDGE DECKS

Effect of Fatigue on Ultimate Load and Fracture Patterns

951

Unless a bridge deck is loaded across its entire width the stress is not

constant across any section. Following the principle of creep hardening and

micro -crushing of the compressive zone and creep softening of the tensile zone,

different parts are fatigued according to the stresses developed. Hence the

effect of fatigue on the ultimate load is the combined effect of all parts along

which fallure of the deck takes place.

The ultimate moments per unit length of all the te st slabs are given in
(12)
Table 3. These have been calculated by methods laid down in C. P .115 and
(13) . . (14) .
Janney et ·a l m the case of prestressed sectiOns and Hognestad et al m

the case of reinforced sections. The values specified by the code are invariably
•
conservative and unit moments obtained by Janney ana Hognestad were used for
#

the compari son of the theoretical ultimate loa~s with experimental values. The

loads for zero tension at the bottom of slabs, which have been calculated using

Morice and Little's (lS) distribution analysis a r e a l so presented in Table 2 for

comparison with the ultimate loads. The results of control specimens are shown

in Table 4.

The theoretical ultimate loads for slabs wer e calculated following the
. . 0)
methods described by the authors in another paper . Table 3 give s the

summary of both the experimental and theoretically predicted modes of

failure as well as the transver se moment necessary to obtain the complete

failure across the whole width of the deck.

·.
... a . .-
;.,..T.
t·
~

952 CONCRETE BRIDGE DESIGN

The experimental ultimate load of the prestressed concrete slabs tested

0
with short tenn loading shows a fairly close agreement with the theoretically

calculated values (Table 2). The discrepancies noted in some are partly due

to the small thickness of the section leading in turn to some variability of

concrete strength. In the case of fatigued slab although the applied load was

below the critical, no appreciable increase in ultimate load was noted over

their non -fatigued companions. Tills could be due to the fact that since the

duct s were ungrruted and no restressing was done, the 'gi ve in' of the wedges

at the anchorage might have caused some loss of pre stressing. Even so the

results were quite close to the corresponding non -fatigued s labs. Furthermore

the fracture mode of all th~ fatigued sla bs were s imilar to those as

~ theoretically predicted. Two of the fracture modes obtained in the tests are

• shown in Figs 7 and 8 .

The effect of fatigue on the ultimate load of reinforced concrete s labs

seems to be sm all in spite· of the fact that the maximum loa d of the fatigue load

range was slightly higher than the working load of the slab. The ultimate loads

have been calculated for both shear and flexural patterns, and these a r e almost

identical partly because of the large thickness to span rati o. In the test a

s imultaneou s shear a nd tiexur?l fan failure occurred for both s labs (Figs

9 and 10) . The failure pattern at the top surface definitely indicat es failure

by punching shea r, a nd the very wide top tensile cracks and crushing at the

bottom of the s lab indicate the flexural fa ilure in a fan mode .

&

L.

·.
....., ...
 FATIGUE EFFECTS ON BRIDGE DECKS 953

In the authors' tests all slabs except 18 were s ubjected to eccentric

.. load. This is an extreme case and unlikely to take place in actual bridge

decks subjected to moving loads over the whole widths. In spite of this no

change of fracture mode was observed. The authors' tests covered a wide

range of s labs, and it can therefore be concluded that the fracture modes can

be used as the basis for the calculation of collapse load s of slab bridges.

Conclusions

The following conclusions can be drawn from the authors' inve stigation:

1. The effect of fatigue loading does not a lte r the mode of

collapse of a bridge deck.

• 2. The yield line approach can be successfully used for the

• collapse load of bridge decks s ubjected to moving loads .

Acknowledgement s

The wo rk described in the paper form s a part of the inves tigations carried

. out in the Department of Civil Engineering, Uni versity of Leeds , on various

aspect s of bridge structures . All the test specimens were manufactured with

great care in the Departmental Workshop.

·.
·A
 954 CONCRETE BRIDGE DESIGN

Refe r ences

l. Johansen, K. W. "Yield-line theor y". Transl ated f rom the Danish."

Cement and Cone. Assoc ., London.

2. Mori ce , P . B. "Minimum tra nsverse strength of s la b bridge s" . Mag . of

Cone. Research, Vol. 8, No.23, August 1956 , pp.65-68 .

3. Reynolds , G. C . "The strength of right pre st res sed conc r et e s la b

bridges with edge bea m s" . Cem ent and Cone . Assoc . , London .

Technical Repo rts , TRA/ 237 . Decembe r 1956 .

4. Best, B.C. a nd Rowe , R. E . "Abnormal loading on compos ite s lab

bridges" . Cement and Cone . Assoc. , London . Re s ea r ch Report 7 .

• ;) . Gr a nholm, C . A. a nd Rowe , R. E . "The ultimat e load of simply

s upported skew sla b bridges ". Cem ent a nd Cone . Assoc . , London,

Re search Report 12 .

6. Sawko, F . a nd Saha, G. P. "Ultirrn t e load a na l ysis of b ridg·e decks".

Building Science . Vol. 2, pp. 223 - 23 7 . Perga mmon Pr ess 1967 .

7. Payne , H . L . and Caldwell, L . H . "Ultimat e strength design for

highway bridge s" . journal of the St ructura l Divi s ion, A . S . C . E . ,

Vol. 91 , Oct ober 1965, pp . -! 3 -56 .

8. B. S . Code of Pra cti ce for Reinforced Concr et e No .11 4 , 1965 .

9. Standa rd Highway Loadings, Memorandum No . 77 1 (Amendment t o

Memo randum No . 577). H . M. S. 0 . 1961.

,.

·.
..
c

... -"
 FATIGUE EFFECTS ON BR I DGE DECKS 955

10. Saha, G. P. "Ultimate load analysis of bridge decks including the

effect of fatigue". Ph . D. Thesis, University of Leeds, July 1966.

11. Sawko, F. and Saha, G. P. "Fatigue of concrete and its effect upon

prestres sed concrete beams" . Magazine of Concrete Research, Vol.20,

No.62, March 1968, pp.21-30.

12. B.S. Code of Practice for Prestressed Concrete. No.ll5, 1959.

13. Janney, J. R., Hognestad, E. and McHenry; 0. "Ultimate flexural

strength of prestressed and conventionally reinforced concrete beams".

Proc. Am. Cone . Inst., Vol.52, Januaryl956, pp . 601-620.

14. Hognestad, E., Hanson , N. W . and McHen.ry, 0. "Concret e stres s

distribution in ultimate strength design". Proc. Am. Cone. Inst. ,

•
Vol.52, December 1955, pp.455-479.

15 . Morice, P.B. and Little , G . "Analy sis of right bridge decks subj e cted

t o abnormal loading ". Cement and Cone. A.ssoc., London, July 1956 .

. . . . , .. ·~ -~
 956 CONCRETE BR IDGE DESIGN

S LAB SPAN WIDTH THICKNESS CONCR.ETE

R.EF MIX

Q
"'
V'J
lA , 18 7fl 6111 5 ft I~ ln. 1: 1: 2 wjc 0·4
V'J
"'...~ 2A,28 7ft 6 m 5 ft I~ tn I 1 · 2 W/C 0 ·45
V)

'"'
a:Q ~
V)
3A,38 7ft . 6 In Sft l)l.n 1 1· 2 w/c 0 · 45
....

'"'"' 4 SIC 5 ft IYz In I 1·2 wjc 0 ·45

-"''"'....Q!
~
Q'"'

~
5 Sf t 5 (t IVzin 1: 1: 2 w/c 0 · 4S
'"'(,)
.... ~
6A,68 3ft. 9m Sft . J!im. 11 ·2 wjc 0 ·4 5
.... 0
V)(J
0
Q. 7 A, 78 3ft.91fl. 5 ft. - llf.m 1: 1: 2 w/c 0 · 45

V)
<.) ~ 8A ,88 3ft. 9tn Stt J in 1: /:'t : J W/C 0 ·55
~~

r;l

Table 1 T est data for s labs

• {/) ( 2} (]}
SLAB RANGE OF TOTAL N!!OF CR.ACKING M f)R ICE & LITTLE CALCULATED EX PER./· RATIO OF R.AT/0 OF
REF. LOADS REPETITIONS LOA D LOAD FOR ZERO ULTIMATE MENTA L (2} ( 3)
( Ktps J ( /0 - ) ( Ktp s J TENSION/ WORKING LOAD ULTII-IATE A ND AND
LOAD ( K ips ) ( Ktps ) LOAO(KifJS) (I} ( 2)

/A - - 1- 5 0 876 4. 35
! 5·05 4 · 96 I · 16
18 I 6 - 0· 45 ]·97 ; 6 BF 0 876 4 · 35 4·S 4 · 96 1 · 03

2A - - I ll 0 . 66 4 . 35 4 '1 6 · 59 O · 9t.
28 I 075- 0 45 5 St. I 6 "':: 0 66 4 ·] 5 ]· 45 6 59 0 79

]A
38
-
1075- 0 45
I t. - 0 45
-
2 · 72
2 · 79
I 8
13 A F (Aftu l
1 93 si0 6 Cycl~s) l
0
0 · 66
66 t. 35
4 •35
t. · S
3 . 45
6 . 59
6· 5 9
1· 0 5
0. 7 9 I
i
t. 4 4 - 2 2 2·98 s -2 aF I 7 9· 36 11· 5 5 52 I· 2 2
II
!
I
5 4 ·4 - 2 2 2 · 98 5 ~ E ;- 7 8 30 8. 75 4 88
Ii ; 06 I

6A - -
I
j ·; ; 3 5 ' 89 7. 0 I 4 53 T J9 I

68 30 - I 5 2 · 65 j ==:- 1
- Ii 7 21. 7 ·I I
J
5 . 56 I
0'98
I
i
- I I
I
--
- ; ; I 5 · 21 6 5 I 4 01
I I 24
78 3 0 - I 5 2 9 j ~ a= I s· 21 6·0 4 01 ! I 15

L"
I I
BA
88 _ 6·0 ~ 3·0 1
-
3 · 28
I
t. 0
4 0 3F
i
5 5
5 6
I
22 · 1
22· 1
28 · 5
27·5
I
3 66
3 66
I
1 ·30
I 27
I
I
8 F BfFOR E FA TIGUE A F AF TER FATIGU E

I
Table 2 Test result s of s labs

~
. - ....
_, . . .- "·
 FATIG U E EFFECTS ON BRIDGE DECKS 957

•
SLA B
MOMENTS AS PROVIDED IN SLABS(Kipilji-Js )
END.
(2)
THEORETICAL MODE
EXPER.I-
MEN TAL (3)MO MENT
TRA NS. I
C P 1151114 JANNEY(IJ HOGNESTAD MODE OF NECESSARY
REF. CONDITIONS OF FAILURE FOR. FOR FAILURE
MTS. IN COL. (I J FAILURE
ACROSS( Ki p irYtnS) :
1

LONG TRANS LONG TR ANS

I A
18
0· 982
0 · 98 2
1· 3
1· 3
I· 17
I · 17
7· 45
I ·45
CORNERS
FR EE ..
A CROSS 0 ·234 ( 8 }
0 ·234. (8 }

2A 0 98 2 1 ·3 I 17 I 45
..
"
.. 0 ·234 ( 8 )
28 0 · 98:! I 3 I ·17 I 45 " 0 . 234 {8 )
AS IN

J A 0 982 0 · 605
0 · 6J 5
I · 17
I · 17
i
I
.}· C'24
() · 524
.. .."
COLUMN
{ 2) 0 . 234 (8)
0 · 23 4 (B)
3 B 0 982 "
4 0 ·982 1·3v0 I ·17 I
I 1 · 45 .... ·y·
.' I · 025 ( 8}
5 0 ·982 0 · 605 I ·1 7 0· 5 2 4 TYPE I ·025 (BJ

6A
68
0 -982
0·982
0 · 60 5
0· 605
1 ·17
1·17
I J· 524
u· 5 2 4
..
F ~ R.
· y· TY PE
A CROS S
I -04 0 (8)
0 ·592 (8)
I CORNERS
H£LO F4N
7A
7B
0 ·982
0 · 982
0 · 152
0·152
I · 17
I · 17
0 ·388
0 ·388 .. !
OO iVN
FAN
1· 184 (B)
I · 184 (8)

8A
88
3 ·6
3·6
2 · 99
2· 99
4 · 16
4 ·16
3 · 48
3 · 48 .... I SHEAR.
SHEAR.
12· 9
12· 9
(8)
(8}

THE t-IINI I-I UM TRA NSV E !?.SE MOMENTS NECE SS ARY SHO WN I N COL.f3.J AR.E CA L CULATED FOR. THE
A CTUAL END CONDITIONS AND FOR THE LAYOU T OF REINFORCEM ENTS AS PROVIDED IN THE SL A BS.

T a bl e 3 Fatigue m odes of s labs

SLAB A GE A T C U BE CYLINDER. M OO OF MOO OF

REF TE S T lb/ m 2 lb /,n 2 R UPTURE ELA STICITY
( DAYS) lb/ m 2 lb/in2 ( 10 61

lA 30 9. 9 3 0 5 , 900 652 4 .· 95
IB 56 10.6 5 0 6 ,0 10 7 25 5 · 24

2A 56 9,6 50 5,8 00 662 4 ·9

2B 67 10 ,300 6.400 750 4 . 96

3A 105 9 . 100 6,020 600 4 .5

3B 132 10. 100 6, 33 0 68 7 4 .5

126 10, 200 6.4 10 5

' 6 80 4 .

5 I 42 10. 000 6. 300 600 ' .6 I

6A 132 9 , 600 6,010 590 5 ·0 i
6B 152 10. 100 6 .220 600 5 ·O I
7A 1 43 9.700 5.990 580 4 ·5
7B ·136 10. 200 6, 22 0 570 5· 0

8A 13 ? 8.800 4,700 78 0 l.·t.8

8B I 57 8.80 0 4.800 6 00 l. " 1. 9

T able 4 Strength of control speci men s

·.
'.
...-....
·- -_,
.
c •
!.

958 CONCRETE BRIDGE DESIGN

•
3/41 ~~ ~
0·2 '0/A. TR-AN5VER.5E PR.£5TR.E551NG W112E5

t(I!~~
. •
0 35

w2~ d
+ + + + + + + +
t 0·2'DIA. LONGITUDINAL •
+

PR.E5TR.E551NG WIR.£5 (cil 2Ya c/cs

0 35
'

[SLAB N.~ 1A, lB. 2A.2B & l.~

~
0·2'0/A. TR-AN5VER.SE PR.£5TR.£551NG WIR.E5 •
3.,. J,.../·35
·~ (~1!·:5
7~· 7ft~ 0-~
• • • +
,j. [0 ·2·0/A. LONGITUDI NAL
PR.£5 TR.E551NG WI R.£5 (ci) 2~ clcs

I SLAB N2 3A.3B.5. 6A & 6B. ~

t
r 0 ·2. DfA LONGITUDINAL PR.E5TR.£551NG
3 •

~ ...~+-:~:---- •
WIR.E5 ~ 2!f.• clcs.

t3176• ? M 5 BA R.S
•
C 7$2' c/cs
• • r r~w
~
j_

CLEAR. •
COVER.~
IsLAB N2 7A & 7 B I

•
~~ . . . rj_ I·
-G..s· ~ M.5 BAR.5 CLEAR. COVER. Vi~
In> 2')( c/cs 3 •;.
VB 'f M.5 BAR.S (<i) 2~· c/cs

~N2B A& BB I
Fig 1 Longitudinal section of s labs showing detail s of r einforcement

Fig 2 - Detail of mould and arrangement of prestressing wires before

casting
~

h..
:2

... . . ..
l: ...
'"
a 960 CONCRETE BRIDGE DESIGN

PR. E STR.E S S I N G
S TEEL

1 S E CTION EQU IVA L ENT E l EQUIVA LENT E I A FTE R.

B E FOR.£ FATIGUE FA TIGUE ( LOAD RANG£
ABOVE CR./TICA L)

1
! Fig 5 Equivalent secti on before and afte r fat igue
I

I t
5r---~----.-----r----.----.-----r---~----.----.

I' (, ~~j .. _ ... ~ i·w~~j ~ --= I

~ Jl I• I ~ ..r-:=,t I I I 1-51-\ ----tJ-

"'
Cl
. ~
"'
321 / 1 \,/ ! ~Cl 1vl ' ' I II II ' I

~
<t
Cl
_,
.

POINT I

2 J (, 5 6 0 0·5 1·0 1·5

DEFLECTION (ins } DEFLECTION (ins)
ONE UCLE LOA D - DEFLECTION
CURVES AT VARIOUS STAGES

Fig 6 Typical l oad -deflect i on curves for slabs 2A and 2B

.
'/

·.
..
 View publication stats

FATIGUE EFFECTS ON BRIDGE DECKS 961

Fig 7 Fracture pattern of s lab 7 A

Fig 8 Fracture pattern of s lab 7B Fig 9 Fracture pattern of slab SA (top surface)

Fig 10 Fracture pattern of s lab

SA (bottom su rface)

ill

I
!!illl

_..... .. ,_ --------..n-.