Magazine of Concrete Research, 2002, 54, No.

3, June, 181–191

Influence of stirrup distribution and support

width on the shear strength of reinforced
concrete wide beams
P. Serna-Ros
, M. A. Fernandez-Prada
, P. Miguel-Sosa
and O. A. R. Debb{

Polytechnic University of Valencia; Mediterranean University of Science and Technology

This study reports on the results obtained in an experimental test programme carried out on 18 wide beams of
750 mm in width, 250 mm in depth and a 3 m free span, designed to fail by shear. From the tests, a clear influence
of arrangement of vertical shear reinforcement, transversal spacing of stirrup legs and support width on the shear
strength is found. The shear strength of the tested beams has been compared with the ones predicted by the codes
ACI 318 and Eurocode 2 and several improvements to both codes are proposed in order to take such effects into
account. Finally, some design recommendations about distribution of stirrups are included in this paper to optimise
the vertical shear reinforcement effectiveness and prevent premature shear failures.

Introduction
secutive stirrup legs less than 20 cm for high shear
In some countries it is common practice to use wide stresses (.1·59 fc1=2 ) and 40 cm otherwise.
3
beams (width . depth) in building work, mainly for Leonhardt and Walther and Regan and Rezai-
4
architectural reasons. This kind of beam has two spe- Jorabi studied the possible decrease in the shear resis-
cial features that may influence the shear strength: tance with the support width and show, from their
support width and transversal spacing of stirrup legs. experimental tests, that the support breath has little
Generally, codes do not consider these effects on the influence.
shear strength, although some authors have studied them
1
before. Hsiung and Frantz tested several beams with
different width=depth ratios, between 0·33 and 1, and
Objectives
considered that there was no influence of transversal
spacing of stirrup legs on shear strength. However, An- This study shows the experimental results about the
2
derson and Ramı́rez found in their experimental work shear strength of wide beams carried out at the Labora-
that shear strength depends on transversal spacing of tory of Concrete Structures of the Technical University
5
stirrup legs and explained the above Hsiung and Frantz’s of Valencia and compares them with those predicted
6 7
results because of the low level of shear stresses in their by the codes ACI 318 and Eurocode 2.
tests. Moreover, they concluded that transversal spacing The scope of this work is focused on shear behaviour
of stirrup legs is important for high level of shear stres- of reinforced concrete wide beams, with constant cross
3
ses, as Leonhardt and Walther reported before. These section, constant longitudinal reinforcement along the
last authors recommended a distance between two con- beam, and transversal reinforcement made up exclu-
sively by closed vertical stirrups (ties).
Our main concern is to analyse the influence on
shear strength of longitudinal spacing between stirrups,

Universidad Politéchnica de Valencia, Deportamento de Ingenieria transversal spacing between vertical legs in each stir-
de la Construcción y Projectos de Ingenieria Civil, 2 Camino de Vera rup, and support width.
s/n, 46071, Valencia, Spain. The analysis will focus on the comparison between
{ Mediterranean University of Science and Technology, Edificio Ga-
the shear strength actually reached in the tests and
lileo Galilei, Av. Los Naranjos s/n, 46022, Valencia, Spain.
those values that would be obtained applying the cal-
(MCR 926) Paper received 29 January 2001; last revised 9 August culation formulae included in the codes mentioned
2001; accepted 13 December 2001 above.
181

0024-9831 # 2002 Thomas Telford Ltd

Downloaded by [ Imperial College London Library] on [28/06/17]. Copyright © ICE Publishing, all rights reserved.
 Serna-Ros et al.

Experimental programme methodology were supported all along the beam width
(750 mm), whereas in the series D the support was
The results to be analysed have been obtained from
limited to the central 300 mm.
the test to failure carried out on 18 wide beams de-
signed to fail by tension in the beam web. The main
features of the experimental programme are In order to characterise the materials used in the
beam manufacture, the compression strength of con-
(a) cross section of beams: 750 3 250 mm2 . crete was determined by means of 150 3 300 mm
(b) support and load system: simple supported beams cylindrical specimens. The specimens were made at the
with a 3 m free span. The load was asymmetrically same time as the beams, were cured like these and
applied at two points (Fig. 1) to make shear failure tested in crushing on the same day as the beams, in
occur in the instrumented span, so that the ratio compliance with rule ASTM C 39-93a. The yield
between shear span and effective depth was equal strength of the stirrups was also determined and the
to 4 (a=d ¼ 4). results are shown in Table 1.
(c) the longitudinal reinforcement was kept constant The load was applied to the beams by a 1000 kN
along the beam. Compression reinforcement was hydraulic jack, and gradually increased at a rate of
seven 16 mm dia. (geometrical ratio r9 ¼ 0:94%) 5 kN=min until the shear failure was reached in the
and tensile reinforcement was seven 20 mm dia. instrumented shear span.
(geometrical ratio r ¼ 2:2%). The anchorage of Due to an unforeseen event, the test control was lost
the bars was secured by welding them to steel during the beam D4 test. Consequently, it was decided
plates placed at either end of the beam. The effec- to unload and to repeat the test. As there was no proof
tive depth was 206 mm for all the tested beams. of the first load registers, the results of this beam
(d ) the reinforcement was designed according to the should be analysed with caution.
code specifications, except for the shear reinforce-
ment spacing, which was considered as one of the
variables in this study.
Analysed theoretical formulae
The 18 beams were made in 4 series. Table 1 shows The analysed theoretical formulae for predicting
the characteristics of each beam tested. In each series, shear strength of beams have been those proposed by
all the beams were made simultaneously with concrete the following codes: Eurocode 2 (art 4, 3, 2); ACI 318
from the same mixture and a beam without shear re- (Cap. 11)
inforcement has been included. Due to the fact that the beams are designed with
The variables analysed were sufficient margins to exclude the failure of compressed
struts in the web, the limits of this last kind of failure
(a) longitudinal spacing between stirrups (s): 85, 127·5
were not considered in the formulae.
and 170 mm, corresponding to stirrup spacing/
effective depth (s=d) ratios of 0·41, 0·62 and 0·83
respectively Formulae: general framework
(b) shear reinforcement arrangement: 2, 4, 6 and 8 The general framework and limitations of the two
legs in each stirrup analysed formulae are shown in Table 2. After studying
(c) diameter of stirrups: 5, 6, 8 and 10 mm. The stir- the formulae, the following aspects could be discerned.
rup diameter was chosen so that two geometrical The shear strength formulae have two terms in both
ratios of shear reinforcement were obtained: codes: the shear strength provided by concrete (Vc ) and
0·089% for R1 and A1 beams and 0·123% for all the shear strength provided by shear reinforcement
other beams (Vs ):
(d ) support width (bs ): beams from series A, R and C Vu ¼ Vc þ Vs (1)
The shear strength of a beam without stirrups is equal
to the concrete contribution (Vc ) of a beam with stir-
415 585
rups and the same characteristics.
P Contrary to Code ACI 318, Eurocode 2 enables us to
0·585 P 0·415 P
compute the shear strength with a variable angle Ł of
compression diagonal in truss analogy. The values of
cot gŁ may be chosen between 0·4 and 2·5, but if the
angle is not 458, the concrete contribution is cancelled
830 1000 out.
3000 ACI Code limits f y to 420 MPa in order to avoid
excessive crack width. Since Eurocode 2 does not limit
Fig. 1. Test arrangement, dimensions in millimetres this value and the actually yield strength of stirrups in
182 Magazine of Concrete Research, 2002, 54, No. 3

Downloaded by [ Imperial College London Library] on [28/06/17]. Copyright © ICE Publishing, all rights reserved.
 Magazine of Concrete Research, 2002, 54, No. 3
Table 1. Tested beams, characteristics of materials and experimental results
Test programme Shear reinforcement Materials Experimental shear strength: kN
Series Beam Reinforcement details Spacing: mm Diameter Geometrical ratio: Concrete f c : Stirrups f y : Total Increase by
% MPa MPa V u exp: stirrups:
Transversal Longitudinal ˜V s,exp
R R0 29·2 244

R1 690 170 26 0·089 29.2 639 266 22

Influence of stirrup distribution and support width on reinforced concrete wide beams
A A0 24·5 187

A1 230 170 16 0·089 24·5 669·6 265 77

A2 690 170 110 0·123 24·5 512·2 234 47

A3 230 170 18 þ 16 0·123 24·5 517:2  669:6 282 94

C C0 25·2 182

C1 690 170 110 0·123 25·2 642·1 258 76

C2 230 170 18 þ 16 0·123 25·2 568:5  575:6 291 109

C3 115 170 15 0·123 25·2 645·8 358 176

C4 138 127·5 15 0·123 25·2 645·8 360 178

C5 230 85 15 0·123 25·2 645·8 343 161

D D0 32·6 218

D1 690 170 110 0·123 32·6 633·5 266 48

D2 230 170 18 þ 16 0·123 32·6 555:2  544:3 296 78

D3 115 170 15 0·123 32·6 622·3 320 102

D4 138 127·5 15 0·123 32·6 622·3 282 64

D5 230 85 15 0·123 32·6 622·3 345 128

183

Downloaded by [ Imperial College London Library] on [28/06/17]. Copyright © ICE Publishing, all rights reserved.
 Serna-Ros et al.

Table 2. Equations used for shear strength calculation

Instruction Formula Limitations
pffiffiffiffiffi
V u ¼ Vc þ Vs Vc . 0:3 f c bd
  
pffiffiffiffiffi Vd d bd
ACI 318 Vc ¼ f c þ 120 r f y < 420 MPa
Md 7

d Vd d
Vs ¼ As f y .1
s Md

V u ¼ Vc þ Vs

Vc ¼  Rd k(1:2 þ 40 r)bd

 Rd ¼ 0:25 f ctk
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
EUROCODE 2 (normalised fctk ¼0·21 3 ( f c  8)2 r < 0:02
method)
d
k ¼ 1:6  ,1
1000
d
Vs ¼ 0:9As f y
s

b ¼ web width (mm)

d ¼ effective depth (mm)
f c ¼ concrete strength in compression (MPa)
f y ¼ steel yield strength (MPa)
Vd , M d ¼ shear and moment in the considered section (N and mm·N)
As ¼ area of shear reinforcement within a distance s (mm2 )
s ¼ longitudinal spacing between stirrups (mm)
r ¼ geometrical ratio of longitudinal reinforcement
f ctk ¼ characteristic tensile strength of concrete (MPa)

the tests amply exceeded 420 MPa, this limit will not where V Rd2 is the shear resistance for compression of
be considered. concrete in the web:
 
1 : fc fc :
Limits in longitudinal spacing of shear reinforcement V Rd2 ¼ 07 0 9bd (4)
2 200 1:5
Code ACI 318 limits the stirrup spacing (s) to:
where f c , 40 MPa.
s . d=2 According to these criteria, only beams C5 and D5
(2)
s . 600 mm fulfil ACI Code limitations, as for the others the long-
itudinal stirrup spacing is higher than half of the effec-
When shear stresses are high (higher than those tive depth.
reached in this experimental programme), these limit- In order to specify the Eurocode 2 limits, V Rd2 has
ations are reduced to half. to be calculated taking into account the actual concrete
Eurocode 2 proposes different limitations depending strength of each series. Accordingly, the maximum
on the design shear value (VSd ): longitudinal spacing of shear reinforcement for all
( beams should be 0:6d ¼ 124 mm. Hence, only beams
s . 0:8d
For VSd < 1=5V Rd2 . . . . . . . . . . . . . . . C5 and D5 could be acceptable for both ACI 318 and
s . 300 mm Eurocode 2.
(
s . 0:6d
For 1=5V Rd2 , VSd < 2=3V Rd2 . . . : (3)
s . 300 mm Transversal spacing limitations in stirrup legs
(
s . 0:3d Only Eurocode 2 sets out limits for transversal spa-
For VSd . 2=3V Rd2 . . . . . . . . . . . . . . . cing of stirrup legs (st ). These limitations might be
s . 200 mm summarised as follows:
184 Magazine of Concrete Research, 2002, 54, No. 3

Downloaded by [ Imperial College London Library] on [28/06/17]. Copyright © ICE Publishing, all rights reserved.
 Influence of stirrup distribution and support width on reinforced concrete wide beams
(
st . d calculated by the different theoretical formulae based
For VSd < 1=5V Rd2 . . . . . . . . . . . . . . . on the f y and f c values shown in Table 1 for each
s t . 800 mm beam.
( In order to compare the test results with those pre-
s t . 0:6d
For 1=5V Rd2 , VSd < 2=3V Rd2 . . . : (5) dicted by the theoretical equations, an experimental/
s t . 300 mm theoretical shear strength ratio () has been calculated
( (Table 4) for every beam according to the following
s t . 0:3d expression:
For VSd . 2=3V Rd2 . . . . . . . . . . . . . . .
s t . 200 mm V u,exp
¼ (7)
V u,cal
In order to analyse the influence of the transversal
spacing of stirrup legs on the shear strength, the tested Table 4 also shows the ratio between the experimen-
beams have to cover a range of st values, which do not tal and theoretical values of increase in shear strength
fulfil this limitation, in the cases of beams with stirrups provided by shear reinforcement:
of only two and four legs.
˜V s,exp
s ¼ (8)
V s,cal

Test results and theoretical values

Table 1 shows the maximum shear reached for each Analysis of results
beam (Vu,exp ), which includes the effect of the self-
weight of the beam and the weight of the test equip- General considerations
ment. This table also includes the increase in shear From Tables 3 and 4 the following general considera-
strength (˜Vs,exp ) provided by stirrups, quoted as tions emerge.
‘increase by stirrups’. This value is obtained for each
beam by subtracting the shear strength (V u0,exp ) of the (a) All the theoretical shear strength values calculated
reference beam (the beam without stirrups) from the are lower than the mechanical capacity reached in
achieved shear strength: the tests, except for beams with only two stirrup
˜V s,exp ¼ V u,exp  V u0,exp (6) legs, where EC-2 was followed. However, it should
be noted that these beams exceed the limits for
Table 3 shows the shear strength (V u,cal ) and the transversal spacing of stirrup legs.
increase in shear strength (Vs,cal ) provided by stirrups, (b) The results obtained by ACI Code are more con-

Table 3. Theoretical shear strength (kN) predicted by Codes

Series Beam ACI 318 Eurocode 2

Total Increase by stirrups Total Increase by stirrups

(V u,cal ) (V s,cal ) (V u,cal ) (V s,cal )
R R0 133·7 173·2
R1 221·3 87·6 252·0 78·8
A A0 123·7 146·6
A1 215·5 91·8 229·2 82·6
A2 221·2 97·5 234·3 87·7
A3 232·6 108·9 244·6 98·0
C C0 125·3 150·7
C1 247·5 122·2 260·7 110·0
C2 234·0 108·7 248·5 97·8
C3 248·2 122·9 261·3 110·6
C4 248·2 122·9 261·3 110·6
C5 248·2 122·9 261·3 110·6
D D0 140·5 191·3
D1 261·1 120·6 299·8 108·5
D2 245·4 104·9 285·7 94·4
D3 258·9 118·5 297·9 106·6
D4 258·9 118·5 297·9 106·6
D5 258·9 118·5 297·9 106·6

Magazine of Concrete Research, 2002, 54, No. 3 185

Downloaded by [ Imperial College London Library] on [28/06/17]. Copyright © ICE Publishing, all rights reserved.
 Serna-Ros et al.

Table 4. Experimental/theoretical shear strength ratios

Series Beam ACI 318 Eurocode 2

Total  Increase by stirrups  s Total  Increase by stirrups  s

R R0 1·82 1·41
R1 1·20 0·25 1·06 0·28
A A0 1·51 1·28
A1 1·23 0·85 1·16 0·94
A2 1·06 0·48 1·00 0·54
A3 1·21 0·87 1·15 0·97
C C0 1·45 1·21
C1 1·04 0·62 0·99 0·69
C2 1·24 1·00 1·17 1·11
C3 1·44 1·43 1·37 1·59
C4 1·45 1·45 1·38 1·61
C5 1·38 1·31 1·31 1·46
D D0 1·55 1·14
D1 1·02 0·40 0·89 0·44
D2 1·21 0·74 1·04 0·83
D3 1·24 0·86 1·07 0·96
D4 1·09 0·54 0·95 0·60
D5 1·33 1·07 1·16 1·19

servative than those provided by Eurocode 2, espe-

with the same geometrical shear reinforcement ratio
cially for beams without stirrups.
(0·123%) and the same longitudinal stirrup spacing
(c) Neither ACI 318 nor EC-2 takes into account
(170 mm); and Figs 3(b) and 3(d) represent the  s ratio
transversal spacing of stirrup legs (st ), longitudinal
versus s for beams with the same geometrical shear
spacing of shear reinforcement (s) or support width
reinforcement ratio (0·123%) and the same transversal
(bs ). Therefore, V u,cal keeps constant when they
spacing of stirrup legs (230 mm).
change. However, V u,exp is significantly influenced
In series A and C, totally supported along the their
by these parameters.
whole width, the  s ratio reduces to ,1 for high
transversal spacing, which corresponds to beams with
Influence of the parameters analysed (s t , s and b s ) only two stirrup legs (see Figs 3(a) and 3(c)). It means
Figure 2 shows the reduction of  ratio with increas- that the shear reinforcement does not contribute to
ing transversal spacing of stirrup legs (st ) and long- improve shear strength as would be expected by the
itudinal stirrup spacing (s). Figs 2(a) and 2(c) represent Codes. This deficit is compensated by the major con-
the  ratio versus st for beams with the same geome- crete contribution actually achieved. However, for low
trical shear reinforcement ratio (0·123%) and the same transversal spacing the shear strength provided by shear
longitudinal stirrup spacing (170 mm). Figs 2(b) and reinforcement reaches values much higher than code
2(d) represent the  ratio versus s for beams with the predictions.  s ratio reaches up to 1·6 (EC-2 Code) or
same geometrical shear reinforcement ratio (0·123%) 1·4 (ACI Code) when transversal spacing reduces to
and the same transversal spacing of stirrup legs 115 mm. This increase is not so high if the beams are
(230 mm). partially supported.
It can be seen from Figs 2(a) and 2(c) that the  Figures 3(b) and 3(d) show that the  s ratio is higher
ratio increases by 38% when the transversal spacing of than 1 if the beams are fully supported across their
stirrup legs is reduced from 690 mm to 115 mm, for whole breadth, whereas this ratio is lower than 1 for
beams supported along the whole beam breadth. Never- beams partially supported with longitudinal stirrup spa-
theless, this increase is reduced to 20% if the beam is cing of 170 mm. The  s ratio clearly increases as the
only partially supported (series D, bs ¼ 300 mm). value of s decreases.
Figures 2(b) and 2(d) show an increase of the  ratio
up to 12% in series C and up to 11% in series D when
the longitudinal stirrup spacing changes from 170 to
85 mm. Shear strength calculation: proposed
Following Fig. 2, Fig. 3 shows the reduction of  s
formula
ratio with increased transversal spacing of stirrup legs
and longitudinal stirrup spacing. In the same way, Figs On the basis of the results obtained, shear strength
3(a) and 3(c) illustrate the  s ratio versus st for beams might be re-expressed as follows:
186 Magazine of Concrete Research, 2002, 54, No. 3

Downloaded by [ Imperial College London Library] on [28/06/17]. Copyright © ICE Publishing, all rights reserved.
 Influence of stirrup distribution and support width on reinforced concrete wide beams
Series A (0·123%)

Series C (0·123%)
ACI 318 EC-2
Series D (0·123%)

1·6
1·4
1·2
1·0
0·8
Φ

0·6
0·4
0·2
0
0 200 400 600 800 0 200 400 600 800
Transversal spacing of stirrup legs, st: mm Transversal spacing of stirrup legs, st: mm
(a) (c)

1·6
1·4
1·2
1·0
0·8
Φ

0·6
0·4
0·2
0
0 50 100 150 200 0 50 100 150 200
Longitudinal spacing of stirrups, s: mm Longitudinal spacing of stirrups, s: mm
(b) (d)

Fig. 2. Variation of -ratio versus shear reinforcement spacing

Series A (0·123%)

Series C (0·123%)
ACI 318 EC-2
Series D (0·123%)

1·6
1·4
1·2
1·0
Φs

0·8
0·6
0·4
0·2
0
0 200 400 600 800 0 200 400 600 800
Transversal spacing of stirrup legs, st: mm Transversal spacing of stirrup legs, st: mm
(a) (c)
1·6
1·4
1·2
1·0
Φs

0·8
0·6
0·4
0·2
0
0 50 100 150 200 0 50 100 150 200
Longitudinal spacing of stirrups, s: mm Longitudinal spacing of stirrups, s: mm
(b) (d)

Fig. 3. Variation of s -ratio versus shear reinforcement spacing

Magazine of Concrete Research, 2002, 54, No. 3 187

Downloaded by [ Imperial College London Library] on [28/06/17]. Copyright © ICE Publishing, all rights reserved.
 Serna-Ros et al.

V u ¼ V c þ s V s (9) Fig. 5 shows the calculated  ratio from proposed

formulation applied to both codes, ACI 318 and EC-2.
where  s is an expression that should take into account When the proposed formula applies to ACI Code, 
the influence of the analysed variables (s, st and bs ). ratio ranges from 1·12 to 1·43, if beam D4 is excluded.
The experimental values of  s obtained in this work This range narrows to 1·07–1·21, when EC-2 is used.
are shown in Table 4 and Fig. 3. Considering the limitations of the test programme,
In Fig. 4, the experimental values of  s have been and in order to generalise the proposed expressions,
plotted against examination should be broadened to include other vari-
d ables more wide ranging than those analysed here.
k ¼ pffiffiffiffiffiffi , The experimental results obtained in this work can-
ss t
not be explained from a plane truss model, because
distinguishing between those obtained from the beams they concern variables not included in this kind of
partially supported and those totally supported. It is modelling. A space truss model should be used for this
observed that there is a linear relationship between purpose and should be considered in future research.
these two variables. Adjusting by minimum squares,
two straight lines can be drawn with a very good
regression coefficient. Since  s must be equal to 0 if
k ¼ 0, as it can be observed in Fig. 4,  s function may
be expressed as: New trends in Eurocode 2
 s ¼ k (10) At present, Eurocode 2 is being revised. The second
9
draft of January 2001 incorporates significant changes
where  depends on bs .
in shear strength treatment. The main changes that
From the experimental results, we have proved that
concern this article are summarised as follows.
¼e 1 for beams fully supported across their breadth
(bs ¼ b) and is given by the following expression, for
beams partially supported (bs , b): (a) Transversal spacing of stirrup legs and longitudinal
stirrup spacing are limited to 0·75d in both cases
 Æ
bs and, hence, are independent of the shear force
¼ (11) level.
b
(b) A variable angle truss model is considered for
where Æ should be taken 0·41 in accordance with the beams with shear reinforcement. The value for
obtained experimental values. shear resistance is given by:
Regarding beams fully supported along their breath,
the proposed formula (equation (9)) is similar to the A sw
8 Vu ¼ f y z cot gŁ 1 < cot gŁ < 2:5 (12)
one introduced by Mphonde for slender beams, where- s
in  s ranges from 1 to 16, though it does not consider
the effect of the transversal spacing of stirrup legs in where Asw is the cross sectional area of shear
the assessment. reinforcement.
The use of the proposed formulation yields more (c) For beams without shear reinforcement, the shear
uniform  and  s ratios in the tested beams (Table 5). resistance is given by:

Total support
Partial support
ACI 318 EC-2
1·8
1·6 y = 1·0566x – 0·0523
y = 0·951x – 0·047 R2 = 0·9604
1·4
R2 = 0·9604
1·2
1·0
Φs

0·8
0·6
y = 0·7251x + 0·0275
0·4 y = 0·6526x + 0·0247 R2 = 0·9863
0·2 R2 = 0·9863

0
0 0·5 1·0 1·5 2·0 0 0·5 1·0 1·5 2·0
d(sst)–½ d(sst)–½

d
Fig. 4. Experimental values of s versus k ¼ pffiffiffiffiffi
sst
188 Magazine of Concrete Research, 2002, 54, No. 3

Downloaded by [ Imperial College London Library] on [28/06/17]. Copyright © ICE Publishing, all rights reserved.
 Magazine of Concrete Research, 2002, 54, No. 3

Table 5. Shear strength (kN) and  ratios from proposed formulation applied to ACI-318 and EC-2

Influence of stirrup distribution and support width on reinforced concrete wide beams
Series Beam Proposed parameters Shear strength from proposed formulation  ratios

 k ACI 318 EC-2 ACI 318 EC-2

Total Increase by Total Increase by Total Increase by Total Increase by

V u, pro stirrups V s, pro V u, pro stirrups V s, pro  stirrups  s  stirrups  s
R R0 1·00 133·73 173·22 1·82 1·41
R1 1·00 0·60 186·40 52·7 220·63 47·4 1·43 0·42 1·21 0·46
A A0 1·00 123·71 146·57 1·51 1·28
A1 1·00 1·04 219·31 95·6 232·61 108·9 1·21 0·82 1·14 0·91
A2 1·00 0·60 182·35 58·6 199·34 75·6 1·28 0·80 1·17 0·89
A3 1·00 1·04 237·15 113·4 248·66 125·0 1·19 0·84 1·13 0·93
C C0 1·00 125·26 150·68 1·45 1·21
C1 1·00 0·60 198·77 73·5 216·84 91·6 1·30 1·03 1·19 1·15
C2 1·00 1·04 238·50 113·2 252·60 127·3 1·22 0·96 1·15 1·07
C3 1·00 1·47 306·36 181·4 313·68 188·4 1·17 0·97 1·14 1·08
C4 1·00 1·55 316·16 190·9 322·49 197·2 1·14 0·93 1·12 1·04
C5 1·00 1·47 306·36 181·1 313·68 188·4 1·12 0·89 1·09 0·99
D D0 0·69 140·48 191·28 1·55 1·14
D1 0·69 0·60 190·30 49·8 236·11 95·6 1·40 0·96 1·13 1·07
D2 0·69 1·04 215·56 75·1 258·85 118·4 1·37 1·04 1·14 1·15
D3 0·69 1·47 260·34 119·9 299·15 158·7 1·23 0·85 1·07 0·95
D4 0·69 1·55 266·83 126·3 304·99 164·5 1·06 0·51 0·92 0·56
D5 0·69 1·47 260·34 119·9 299·15 158·7 1·33 1·06 1·15 1·18
189

Downloaded by [ Imperial College London Library] on [28/06/17]. Copyright © ICE Publishing, all rights reserved.
 Serna-Ros et al.
Series R
Series A
Series C
ACI 318 EC-2
Series D
1·6
1·4
1·2
1·0
0·8
Φ

0·6
0·4
0·2
0
0 200 400 600 800 0 200 400 600 800
Transversal spacing of stirrup legs, st: mm Transversal spacing of stirrup legs, st: mm
(a) (c)
1·6
1·4
1·2
1·0
0·8
Φ

0·6
0·4
0·2
0
0 50 100 150 200 0 50 100 150 200
Longitudinal spacing of stirrups, s: mm Longitudinal spacing of stirrups, s: mm
(b) (d)

Fig. 5. Calculated -ratio from proposed formulation

0:18 vative, even if beams with stirrup spacing higher

Vu ¼ k(100r f c )1=3 bd
ªc than the code limits are taken into consideration.
rffiffiffiffiffiffiffiffi (b) The analysed formulae undervalue the shear
200 strength of the beams supported across their whole
with: k ¼ 1 þ (d in mm) (13)
d width and reinforced in their cross section with
Asl multiple leg stirrups (C3, C4 and C5).
r¼ . 0:02 (c) The use of concentrated supports clearly reduces
bd
the shear strength of wide beams.
where Asl is the cross sectional area of the tensile (d ) The shear strength provided by stirrups remains
reinforcement and ª c is the partial safety factor for clearly overvalued when the transversal reinforce-
concrete. ment is made up of stirrups with very few legs,
otherwise this contribution is underestimated,
If this model is used, the proposed formula in equa-
whenever the beams are supported across their
tion (9) cannot be applied. In this case, the improve-
breath.
ment of shear resistance provided by the reduction of
(e) In order to improve the shear strength predicted by
stirrup spacing or spacing of stirrup legs could be taken
codes ACI 318 and EC-2, we propose to include
into account by increasing the upper limit of cot gŁ as
two factors: one which takes into account the influ-
s and st reduce. In the same way cot gŁ should be
ence of the shear reinforcement distribution; and
reduced if beams are partially supported across their
the other which considers the support width in the
breath.
calculation formulae. Therefore, the proposed for-
mulation would be:
Vu ¼ Vcu þ k Vsu (14)
Conclusions
where
Despite the limitations of our research, the following
 0:41
conclusions can be drawn. d bs
k ¼ pffiffiffiffiffiffi and  ¼
(a) The two analysed standards are, in general, conser- ss t b

190 Magazine of Concrete Research, 2002, 54, No. 3

Downloaded by [ Imperial College London Library] on [28/06/17]. Copyright © ICE Publishing, all rights reserved.
 Influence of stirrup distribution and support width on reinforced concrete wide beams

( f ) If a variable angle truss model is adopted for shear sense, this proposal is in accordance with EC-2
resistance as proposed by EC-2 (draft), the above (draft).
proposed formulation would not be applicable. In
such a case, upper limit of cot gŁ should be for-
mulated depending on the parameters studied in References
this article. 1. HSIUNG W. W. and FRANTZ G. C. An exploratory study of the
shear strenth of wide reinforced concrete beams with web reinfor-
cement. Research Report CE 83-151, Department of Civil Engi-
neering, University of Connecticut, 1983.
Design recommendations 2. ANDERSON N. S. and RAMÍREZ J. A. Effect of the detailling of
stirrup reinforcement of the ultimate strength and behavior of
From the results obtained, the following recommen- reinforced concrete members failing un shear, Structural Engineer-
dations could be made in order to optimise the shear ing Research Report, CE-STR-87-2, 1987, 367 pp.
design of wide beams. 3. LEONHARDT F. and WALTER R. The Stuttgard Shear Test, 1961,
Translation n8 111, Cement and Concrete Association, London
(a) The maximum spacing among vertical legs in a 1964.
stirrup could be limited to values close to the beam 4. REGAN P. E. and REZAI-JORABI H. Shear resistance of one way
depth, but it does not seem necessary to reduce slabs under concentrated loads. ACI Structural Journal, 1988, 85,
150–157.
that limit. Accordingly, the use of two stirrup legs
5. DEEB OSAMA A. R. Comportamiento en servicio y rotura de vigas
in wide beams should be banned. planas de hormigón armado para forjados de edificación’. PhD
(b) The longitudinal stirrup spacing limit (s ¼ 0:50d) thesis, Universidad politéctina de Valencia, 1996.
stipulated by Code ACI 318 seems too conserva- 6. ACI 318-99. Building Code Requirements for Structural Concrete.
tive. Although a better transversal reinforcement American Concrete Institute, Michigan, 1999.
7. ENV 1992-1-1. Eurocode 2: Design of concrete structures. Part 1-
performance may be obtained applying this limit, it
1: General rules and rules for building (Spanish version AENOR.
is unreasonable unless this favourable effect is UNE-ENV 1992-1-1), AENOR, Madrid, 1998.
included in the formulae. In addition, in wide 8. MPHONDE A. G. Use of stirrup effectiveness in shear design of
beams, this limit forces the stirrups to be closely concrete beams. ACI Structural Journal, 1989, 86, 541–545.
spaced and, therefore, it would complicate the pro- 9. prEN 1992-1. Eurocode 2: Design of concrete structures–Part 1:
General rules and rules for buildings. CEN, Brussels, 2001.
blems that arise during construction. Within the
scope of this study it would be enough to limit the
maximum longitudinal spacing among stirrups to a
value of 0·8 times beam depth, keeping it constant Discussion contributions on this paper should reach the editor by
even if VSd values are higher than 1=5V Rd2 . In this 13 November 2002

Magazine of Concrete Research, 2002, 54, No. 3 191

Downloaded by [ Imperial College London Library] on [28/06/17]. Copyright © ICE Publishing, all rights reserved.