ACI MATERIALS JOURNAL TECHNICAL PAPER

Title no. 93-M24

Prediction of Shear Strength of Reinforced Concrete

Beams without Web Reinforcement

by Jin-Keun Kim and Yon-Dong Park

A rational and mechanics-based equation is proposed for the prediction of Test results have shown that shear strength of reinforced
shear strength of reinforced concrete beams without web reinforcement. concrete beams without web reinforcement depends mainly
This prediction is based on basic shear transfer mechanisms, a modified
on concrete strength, longitudinal steel ratio, shear span-to-
Bazant's size effect law, and numerous published experimental data,
including high-strength concrete beams with compressive strengths of con-
depth ratio, and effective depth. Of course, factors such as
crete up to 100 MPa (14,500 psi). Comparisons with experimental data maximum aggregate size, diameter of the bars, and spacing
indicate that the proposed equation estimates properly the effects of pri- of the flexural cracks show some minor contribution. A part
mary factors, such as concrete strength, longitudinal steel ratio, shear or all of these primary factors are included in the existing
span-to-depth ratio, and effective depth. It is shown that the proposed shear strength prediction models, but the effects of these fac-
equation is considered to be better than the other equations compared in tors are estimated differently according to the models.
this study with respect to accuracy and estimation of primary factors. A
simplified design equation is also derived within the limited range of effec-
Recently, high-strength concrete has been increasingly
tive depth for practical purposes. used in practice. With the development of concrete technol-
ogy and the introduction of superplasticizers and silica fume,
Keywords: aggregate interlock; beams; dowel action; failure mode; shear the compressive strength of concrete in the field of ready-
strength; size effect. mixed concrete reached 100 MPa (14,300 psi) and higher.
Since the mechanical properties of concrete are changed in
According to the shear span-to-depth ratio (a/d), shear high-strength concrete, a reevaluation of the prediction model
failure of a reinforced concrete beam without web reinforce-
ment is divided into two modes, as shown in Fig. 1. For a/d
greater than 2.0 ~ 3.0, the inclined cracking load exceeds the
shear compression failure load. With the formation of the in-
clined crack, a beam without web reinforcement becomes
unstable and fails. This type of failure is usually called “di-
agonal tension failure.” For a/d less than 2.0 ~ 3.0, however,
the failure load exceeds the inclined cracking load. If suffi-
cient anchorage length is provided, after the inclined crack
develops, failure may occur by concrete crushing in the
upper end, and this type of failure is called “shear com-
pression failure.”
Shear force in reinforced concrete member is transferred
in various ways. For slender beams where a/d is greater than
2.0 ~ 3.0, shear force is carried by the shear resistance of un-
cracked concrete in the compression zone, the interlocking Fig. 1—Effect of a/d on shear failure mode
action of aggregates along the rough concrete surfaces on
each side of the crack, and the dowel action of the longitudinal ACI Materials Journal, V. 93, No. 3, May-June 1996.
reinforcement. For relatively short beams, however, after the Received Feb. 8, 1994, and reviewed under Institute publication policies. Copyright
 1996, American Concrete Institute. All rights reserved, including the making of
breakdown of beam action, shear force is resisted mainly by copies unless permission is obtained from the copyright proprietors. Pertinent discus-
sion will be published in the March-April 1997 ACI Materials Journal if received by
arch action. Dec. 1, 1996.

ACI Materials Journal / May-June 1996 213

 RESEARCH SIGNIFICANCE
Jin-Keun Kim , MACI, is a professor of civil engineering at the Korea Advanced Insti-
tute of Science and Technology (KAIST), Taejon, Korea. He obtained his BS and MS
Research reported in this paper provides a rational and ac-
degrees from Seoul National University, and his PhD from Northwestern University. curate equation for the prediction of shear strength of reinforced
His research interests include the application of fracture mechanics to concrete struc- concrete beams without web reinforcement. The results
tures and inelastic analysis of concrete structures.
show that the proposed equation predicts the existing exper-
Yon-Dong Park is a senior research engineer at Tong Yang Central Laboratories, imental data more accurately than the other equations in this
Korea. He obtained his PhD from KAIST. His research interests include production study. A simplified design equation is also proposed for
and mechanical properties of high-strength and high-performance concrete and shear
behavior of reinforced concrete members.
practical purposes without a significant accuracy reduction
compared with the original equation.

is necessary to reliably estimate the shear strength of beams BASIC SHEAR TRANSFER MECHANISMS
made with high-strength concrete. Moreover, because of the For slender beams where a/d is greater than 2.0 ~ 3.0, the
shear force in a cracked section of a reinforced concrete
wider range of concrete strength used, more accurate predic-
beam is mainly resisted by the shear resistance of com-
tions of shear strength of reinforced concrete members are
pression zone, interlocking action of aggregates, and dowel
required. action, as shown in Fig. 2. For rectangular beams, after an in-
The shear failure of reinforced concrete beams without clined crack has formed, the proportion of the shear force
web reinforcement has been known to be a typical case of transferred by the various mechanisms is as follows: 8 20 to
brittle failure and indicates significant size effect. In 1981, 40 percent by the uncracked concrete of compression zone;
Reinhardt1 introduced fracture mechanics in the prediction 33 to 50 percent by interlocking action of aggregates; and 15
of shear strength. He analyzed limited test data for shear fail- to 25 percent by dowel action. Meanwhile, in a relatively
ure based on linear elastic fracture mechanics. Subsequently, short beam, the load is transferred directly from the loading
points to supports owing to arch action.
it was established that the size effect implied by linear elastic
fracture mechanics is too strong in the case of concrete, and
Shear resistance of uncracked concrete
that brittle failures of concrete structures are better described In a reinforced concrete beam, after the development of
by nonlinear fracture mechanics. Meanwhile, a simple and flexural cracks, a certain amount of shear is carried by the
approximate size effect law on the basis of nonlinear fracture concrete in the compression zone. It is clear that the shear
mechanics was proposed by Bazant.2 Several studies3-6 have failure in the uncracked concrete is recognized as the failure
shown that Bazant’s size effect law is in good agreement under combined compression and shear9 and the area of un-
with test results. However, there is some discrepancy be- cracked zone. The position of the neutral axis in a beam after
tween the prediction by Bazant’s law and the test data, par- flexural cracking is mainly dependent on the elastic modulus
of concrete and the longitudinal steel ratio because elastic
ticularly for large-sized specimens. Recently, Kim and Eo 7
modulus of steel is nearly constant. Therefore, the shear
proposed a modified Bazant’s size effect law to reduce the
force carried by the uncracked concrete in the compression
discrepancy. zone can be represented by the compressive strength of con-
In the present study, a simple and accurate equation pre- crete and the longitudinal steel ratio, since the concrete
dicting the shear strength of reinforced concrete beams with- strength in the biaxial state and elastic modulus of concrete
out web reinforcement is proposed based on basic is the function of the compressive strength of concrete.
mechanisms of shear transfer and a modified Bazant’s size
effect law deduced by Kim and Eo, and it was verified by the Interlocking action of aggregates
Previous experimental studies8-10 have shown that a large
published test data. In addition, a simplified equation is also
portion of the total shear force on the beam without web re-
proposed for practical design purposes. These equations that
inforcement is carried across the cracks by aggregate inter-
include the effects of all the factors previously mentioned are locking. Among many variables, the width of the crack and
supported by test results and are compared with other pre- the concrete strength are likely to be the most important factors.
diction equations for the shear strength of beams without Since the flexural crack width is approximately propor-
web reinforcement. tional to the strain of tension reinforcement, the crack width
at failure becomes smaller as the longitudinal steel ratio is in-
creased. Also, with increasing a/d, the strain of tension rein-
forcement at failure is increased. Meanwhile, it is naturally
expected that the interlocking force will be increased when
the strength of concrete is high.

Dowel action
When shear displacement occurs along the cracks, a cer-
tain amount of shear force is transferred by means of dowel
action of the longitudinal bars. Although there is some con-
tribution in dowel action by the number and arrangement of
Fig. 2—Shear transfer mechanisms of slender beams longitudinal bars, spacing of flexural cracks, and the amount

214 ACI Materials Journal / May-June 1996

of concrete cover, etc., the main factors influencing this According to Reineck,11 the maximum value of shear
action are flexural rigidity of longitudinal bars and the stress along the crack surface due to aggregate interlock de-
strength of the surrounding concrete. pends on the tensile strength of concrete and is decreased lin-
early with increasing crack width ∆n as follows
Arch action
In relatively short beams, applied loads are transferred di- ∆n
τfu = 0.45ft′  1 – --------- with ∆n u = 0.9mm (6)
rectly to the supports by arch action. The main factors influ-  ∆n u
encing this action are the span-to-height ratio of the
analogous arch and the strength of the compression strut.
The span-to-height ratio of the analogous arch is approxi- Crack width is resolved as ∆n = 0.71εs scr, in which εs is the
mately equal to the shear span-to-depth ratio. The strength of strain of the longitudinal reinforcement and scr is the spacing
the compression strut is closely related to the compressive of the primary cracks. From theoretical considerations and
strength of concrete and the area of tension reinforcement. experimental comparisons, the spacing of the primary cracks
was also derived by Reineck as s cr = 0.7(d - kd). Substituting
DEVELOPMENT OF EQUATION the values of ∆n and s cr into Eq. (6) and assuming that k = 0.3
As previously discussed, shear resistance of uncracked for the usual simplification and d = 500 mm as a reference size
concrete in the compression zone can be expressed as
τfu = 0.45f ′t ( 1 – 193.3εs ) (7)
l
V c = c1 ( f ′c ) ( b k d) (1)
Eq. (7) can be replaced approximately by the following sim-
in which fc′ is the compressive strength of concrete, b is the pler expression within the practical range of εs , i.e., 0.0008 ≤
width of the beam, kd is the neutral axis depth, and c1 and l εs ≤ 0.003.
are empirical constants. The constant l depends on the ratio
of shear stress to compressive stress in an uncracked zone,
and it is in the range of 0.5 to 1.0. According to the classical
bending theory of reinforced concrete beams with only ten-
sile reinforcement and a negligible tensile capacity of con-
crete, we would have

2 2 1⁄2
k = ( n ρ + 2 nρ ) – nρ (2)

in which n = Es /Ec = ratio of elastic moduli of steel and con-

crete. Eq. (2) is, however, unnecessarily complicated and
may be replaced by the following simpler expression

m
k = c2 ( nρ ) (3)

in which c2 and m are certain constants. Within the practical Fig. 3—Comparison of two equations on neutral axis
range, i.e., 5 ≤ n ≤ 10 and 0.005 ≤ ρ ≤ 0.05; consequently,
0.025 ≤ nρ ≤ 0.5. These constants can be chosen so that the
values given by Eq. (2) and (3) are almost undistinguishable.
Fig. 3 shows the optimum values of c2 and m.
The elastic modulus of normal-weight concrete can be rep-
resented as

0.5
E c = c3 ( f c′ ) (4)

in which f c′ is the compressive strength of concrete and c3 is

a certain constant. Therefore, substituting with Eq. (3),
where c2 and m are the values shown in Fig. 3, and Eq. (4)
into Eq. (1), we obtain

( l – 0.18) 0.36
V c = c5 ( fc′ ) ρ bd (5)

Fig. 4—Comparison of two equations on aggregate inter-

in which c5 is a constant. locking stress

ACI Materials Journal / May-June 1996 215

 τ f u = 0.012f t′ ⁄ εs (8) As previously discussed, the shear capacity of slender
beams may be represented as Eq. (13), although the primary
shear transfer mechanisms are coupled
The comparison of Eq. (7) and Eq. (8) is shown in Fig. 4. In
reinforced concrete beams, as long as the tension reinforcement
Vu = V c + V a + Vd (13)
has not yielded, the strain of the longitudinal reinforcement
at the ultimate state can be approximately expressed as
Substituting Eq. (5), (10), and (12) into Eq. (13), we obtain
V u a eff
εs = -------------------- (9) (l – 0.18 ) 0.36
E sA sj d V u = [ c5 ( f ′c ) ρ
(14)
+ c8 ( f ′c ρd ⁄ a)
0.33 0.5 r
+ c10 ( f ′c ) ρ ]b d
in which Vu is ultimate shear force and aeff is effective length
of shear span. Substituting Eq. (9) into Eq. (8) and tenta-
tively assuming that shear strength of reinforced concrete It is noticed in Eq. (14) that the powers of f c′ have similar
beams is approximately proportional to 1 /3 power of (f c′ × values. In practice, uncracked parts of reinforced concrete
ρ × d /a) suggested by Zsutty, the shear force carried by the beams are stresed in two directions (biaxal stress condition).
aggregate interlock [= τfu b(d - kd)] is expressed as The power of f c′ in the first term of Eq. (14), therefore, may
be varied in some range. Considering this fact and the minor
contribution of dowel action (approximately 20 percent of
Va = c 8 ( f c′ ρ d ⁄ a )
1⁄3
bd (10) total shear resistance), it can be assumed for simplicity that
the power of f c′ is identical for all three terms in Eq. (14). In
in which c8 is some constant. addition, the values of power of ρ in all three terms of Eq. (14)
According to Vintzeleou and Tassios,12 the dowel force is are very similar to each other. Consequently, Eq. (14) is then
represented as expressed as

1⁄3
V d = c 9 b c t d b ft′ (11) s t
V u = c11 ( fc′ ) ρ [ c 12 + ( d ⁄ a ) ]bd (15)

in which c9 is a certain constant, d b is diameter of the bar, and in which c11 , c12 , s, and t are certain constants. Because the
b ct is net width of the section (= b - qdb , q being the number term [c12 + (d/a )1/3] can be replaced by [c13 + c14( d/a )] with-
of bars). Eq. (11) can only be used when the number and di- out a large difference (in which c13 and c14 are certain con-
ameter of the bar are known. Assuming the number of the bar stants), as shown in Fig. 5, the shear capacity of reinforced
is not changed, the dowel force can be expressed approxi- concrete beams with constant effective depth can be ex-
mately by more general terms ρ as follows pressed as

s t
0.5
V d = c 10 ( f ′c ) ρ b d
r
(12) Vu = c 15 ( fc′ ) ρ ( c 16 + d ⁄ a )bd (16)

in which c10 is some constant and r, varying from 0.3 to 0.5 in which c15 and c16 are certain constants.
in a practical range, is a parameter that is dependent on spac- To reduce the discrepancy between the prediction by
ing of the reinforcement. Bazant’s size effect law and the experimental data, especial-
ly for very large-sized specimens, Kim and Eo7 recently sug-
gested a modified Bazant's size effect law based on the
concept of dissimilar initial cracks. The formula is as follows

k 1 σr
σN = ------------------------------ + k 2 σr (17)
1 ⁄2
( 1 + k3 β )

in which σN is the nominal strength of a specimen at failure,

σr is the nominal strength of the specimen with reference
size, and k1 , k2 , and k3 are empirical constants. β is brittleness
number, expressed as

2
( ft′ ) D
β = ---------------- (18)
E c Gf

in which f ′t is tensile strength, Ec is modulus of elasticity, Gf

Fig. 5—Comparison of two equations on ξ is fracture energy, and D is the characteristic size of the

216 ACI Materials Journal / May-June 1996

structure. f ′t and Ec are mainly dependent on compressive in which c17 = c15 k1 and k5 = k2/k1 . Eq. (20) has been com-
strength, and G f is dependent on maximum aggregate size pared to all other important experimental data, with regard to
(d a ) and compressive strength. G f is increased with da and not only the effects of concrete strength, longitudinal steel
ratio, and shear span-to-depth ratio, but also to the effect of
compressive strength, although the increase is minimal in the
effective depth.
higher strength range. For reinforced concrete beams, D is
From the statistical analysis of existing experimental
effective depth. Consequently, β can be expressed as a data,13-44 the values of all six parameters in Eq. (20) have
function of f c' , d a , and d. If stirrups are used, Gf is increased been determined. The experimental data used in the analysis
because crack localization is prevented, to some extent, by are only three or four point test results of simple beams with
stirrups. Thus, the effect of stirrups in the size effect of rein- rectangular sections except for Shioya’s test results,37 which
forced concrete beams in shear can be considered. Mean- applied uniform load. Moreover, test results of beams with
while, since the effects of factors influencing G f are not clear small d [less than 100 mm (3.94 in.)] or small maximum ag-
gregate size [less than 5 mm (0.2 in.)] were excluded for
and the ranges of f c' and d a are limited in the construction
practical purposes. Cubic compressive strength of concrete
field, experimenal data have shown that the capacity of reported in some tests was converted to the cylindrical com-
beams is not significantly influenced by d a; therefore, it is ra- pressive strength according to Neville’s empirical relation.45
tional to assume that (f′t )2 /Ec Gf is a constant. Therefore, Eq. (17) Finally, the following equation is proposed for the mean
can be expressed as nominal shear strength of reinforced concrete beams with
a/d equal to or greater than 3.0
k 1 σr
σ N = ----------------------------- + k 2σ r (19) 1⁄3 3 ⁄8
( 1 + k4 d )
1⁄2 ν u = 3.5fc′ ρ ( 0.4 + d ⁄ a) λ ( d ) (21)

in which
in which k4 is an empirical constant. Eq. (16) does not take
into account the size effect. According to Eq. (19), we fi-
1
nally obtain λ ( d ) = ------------------------------ + 0.18
1 + 0.008d

ν u = c17 ( fc′ ) ρ ( d ⁄ a + c 16 )  ----------------------------- + k 5

s t 1
1⁄2
(20) fc′ and vu are in MPa and d is in mm. Fig. 6 shows the ratio
( 1 + k4 d ) of shear strength tested to predicted with a/d ≥ 3.0 for each

Fig. 6—Ratio of shear strength for each primary factor

ACI Materials Journal / May-June 1996 217

factor by the proposed equation. As shown in the figure, it The ratios of shear strength (tested to predicted) with
can be concluded that the proposed equation estimates prop- the bulk of existing experimental data determined by three
erly the effects of primary factors. other equations and the proposed equation are shown in
For relatively short beams, the failure mode is changed Fig. 7. For the ACI Code equation, the scatter is much
gradually from diagonal tension to extreme shear compression. larger than those of the other equations. It is shown that the
Consequently, the effect of concrete strength on the shear ratio of shear strength obtained by Zsutty’s equation increases
strength is also gradually increased with decreasing a/d. Previ- gradually with increasing shear strength. This phenome-
ous experimental work 32 has shown that the effect of concrete non is mainly due to the fact that Zsutty’s equation ignores
strength on the shear strength of reinforced concrete beams be- size effect and slightly underestimates the effect of a/d when
comes more significant with decreasing a/d. If we roughly as- a/d is less than 2.5. Bazant’s equation predicts the shear
sume that the effects of ρ and d are not changed with a change of strength of experimental data comparatively well, although
failure mode, the shear strength of relatively short beams can be the scatter is somewhat larger than that of the proposed equa-
expressed as tion. As shown in the figure, the proposed equation, which has
a standard deviation of 0.162 with respect to average ratio of
α⁄3 3⁄8 shear strength and a correlation coefficient of 0.926, predicts
ν u = 3.5fc′ ρ ( 0.4 + d ⁄ a )λ ( d ) (22)
more accurately than the other equations.
Fig. 8 shows the ratios of tested-to-calculated shear
where α is the failure mode index that is dependent on the a/d ra- strength of existing test data failed by diagonal tension (a/d ≥
tio. From the compatibility condition, which is α = 1 at a/d = 3.0, 3.0) with variation of d. The ratios of shear strength obtained
and the statistical analysis of existing experimental data, α is de- from the ACI Code equation and Zsutty’s equation are de-
termined as creased with increasing d due to the disregard of size effect,
whereas Bazant’s equation and the proposed equation have
a⁄ d nearly constant value. Although Bazant’s equation reflects
α = 2 – ---------- for 1 ≤ a ⁄ d < 3 (23)
3 not only the effect of d but also the effect of maximum ag-
gregate size, the scatter is somewhat larger than that of the
EVALUATION OF PROPOSED EQUATION proposed equation.
Many equations have already been proposed to estimate For relatively short beams failing in shear compression,
the shear strength of reinforced concrete beams. To eval- the scatter of test data is larger than that of slender beams
uate the proposed model, three well-known equations are failing in diagonal tension. As shown in Fig. 9, the standard
selected for comparison: 1) Zsutty’s equation 46 deduced deviation of the ratio of shear strength for the ACI Code
by multiple regression analysis; 2) Bazant's equation 5 derived equation is more than twice that of the proposed equation.
based on Bazant’s size effect law; and 3) the ACI Code Although scatters of the ratio of shear strength for Zsutty’s
equation. 47 These equations are as follows: equation and Bazant’s equation are lower than that of the
ACI Code equation, they are larger than that of the proposed
ACI Code equation equation, to some extent. Similar to the case of slender
beams, the proposed equation with a standard deviation of
0.230 and a correlation coefficient of 0.901 is considered to
V ud
νc = 0.1578 fc′ + 17.25ρ w --------- ( MPa ) ( a ⁄ d ≥ 2.5) (24) be better than the other equations in terms of accuracy and
Mu estimation of primary factors.

PROPOSED SIMPLE DESIGN EQUATION

ν c =  3.5 – 2.5---------u  × [ Eq. ( 24) ]
M
( a ⁄ d < 2.5) (25)
 V ud From a practical point of view, design equations should be
as simple as possible within a range of acceptable accuracy.
Although the proposed equation is relatively simple and
Zsutty's equation
accurate, and estimates the effects of the primary factors
well, it is somewhat complicated to use in practical design,
1⁄3
ν u = 2.1746  fc′ ρ --
d
(MPa) ( a ⁄ d ≥ 2.5) (26) especially regarding the size factor λ(d). The effective depth
 a of reinforced concrete beams in construction is usually lim-
ited within the range of d ≥ 250 mm (9.84 in.). Since small-
sized structures that conform to the strength criterion are not
νu =  2.5--  × [ Eq. ( 26)]
d
( a ⁄ d < 2.5) (27) presented, the term λ(d), which indicates size effect, can be
a
approximately represented using a simpler expression as follows
Bazant's equation
k6
λ ( d ) = ------- + k7 (29)
1 + 5.08 ⁄ d a
ρ  × ------------------------------------- d
νu = 0.54 3 ρ  f′ c + 249 ------------
 5
1 + d ⁄ ( 25 d a )
a⁄d
(28) in which k6 and k7 are some constants. From the regression
(MPa) analysis of existing experimental data, the values of k6 and k7
have been determined. As a result, the simplified size factor

218 ACI Materials Journal / May-June 1996

 Fig. 7—Comparison of various equations with bulk of existing data

Fig. 8—Comparison of various equations with increasing d (a/d ≥ 3.0)

ACI Materials Journal / May-June 1996 219

is nearly identical with the one in the CEB-FIP model code.48 identical accuracy compared with the original equation,
Consequently, the following simplified equation is derived although some approximation has been made.
for the mean nominal shear strength Assuming a normal distribution of the ratios of shear
strength, the design equation for the shear strength of reinforced
( 0.4 + d ⁄ a ) ------- + 0.07
α⁄3 3⁄8 1
ν u = 19.4fc′ ρ concrete beams with 90 percent reliability is represented as
 d 
(30)
for d ≥ 250 mm (9.84 in.) ν u, d = 15.5f ′c
α ⁄3
ρ
3⁄ 8
( 0.4 + d ⁄ a )  -------
1 + 0.07 (31)
 d 

in which α = 1 for a/d ≥ 3.0 and 2 - (a/d)/3 for 1.0 ≤ a/d < 3.0.
Fig. 10 shows the comparison of results obtained from This equation is shown by the straight line in Fig. 11. As
the original equation and the simplified equation. As shown, only a few data points lie below the equation, and
shown in the figure, the simplified equation has nearly no data points fall significantly below this equation.

Fig. 9—Comparison of various equations with increasing a/d (a/d < 3.0)

Fig. 10—Comparison between original equation and simplified equation

220 ACI Materials Journal / May-June 1996

 CONCLUSIONS Vc = shear resistance of compression zone
On the basis of the results obtained in this study, the fol- Vd = dowel force
Vu = shear capacity of beams
lowing conclusions may be drawn.
vu = shear strength of beams
1. Based on shear transfer mechanisms, modified size ef- α = failure mode index
fect law, and numerous published experimental data, the β = brittleness number
following equation is derived for the nominal shear strength ∆n = crack width
∆nu = ultimate crack width
of reinforced concrete beams without web reinforcement
εs = strain of tension reinforcement
λ(d) = size factor
ρ, ρ w
( 0.4 + d ⁄ a )  ------------------------------ + 0.18
α⁄ 3 3⁄8 1 = longitudinal steel ratio
ν u = 3.5f c′ ρ σN = nominal strength of specimen at failure
1 + 0.008d σr = nominal strength of specimen with reference size
τfu = maximum shear stress along crack surface
in which α = 1 for a/d ≥ 3, 2 - (a/d)/3 for 1 ≤ a/d < 3.
2. Comparison with published experimental data indicates REFERENCES
1. Reinhardt, H. W., “Masstabseinfluss bei Schubversuchen im Licht der
that the proposed equation estimates properly the effects of
Bruchmechanik,” Beton-und Stahlbetonbau (Berlin), No. 1, 1981, pp. 19-21.
primary factors, such as concrete strength, longitudinal steel 2. Bazant, Z. P., “Size Effect in Blunt Fracture: Concrete, Rock, Metal,”
ratio, shear span-to-depth ratio, and effective depth. Journal of Engineering Mechanics, ASCE, V. 110, EM4, 1984, pp. 518-535.
3. As a result of the comparison with the ACI Code equation, 3. Bazant, Z. P., and Kim, J. K., “Size Effect in Shear Failure of Longitu-
dinally Reinforced Beams,”ACI JO URNAL, Proceedings V. 81, No. 5, Sept.-
Zsutty’s equation, and Bazant’s equation, the accuracy of the
Oct. 1984, pp. 456-468.
proposed equation is better than that of any other equation. 4. Bazant, Z. P., and Cao, Z., “Size Effect of Shear Failure in Prestressed
4. For practical design, a simple and accurate equation Concrete Beams,” ACI JOU RNA L, Proceedings V. 83, No. 2, Mar.-Apr.
deduced from the simplification of the originally pro- 1986, pp. 260-268.
5. Bazant, Z. P., and Sun, H. H., “Size Effect in Diagonal Shear Failure:
posed equation is derived as
Influence of Aggregate Size and Stirrups,” ACI Materials Journal, V. 84,
No. 4, July-Aug. 1987, pp. 259-272.

( 0.4 + d ⁄ a )  ------- + 0.07

α⁄ 3 3 ⁄8 1 6. Bazant, Z. P., and Kezmi, M. T., “Size Effect on Diagonal Shear Fail-
νu , d = 15.5f ′c ρ
 d  ure of Beams without Stirrups,” ACI Structural Journal, V. 88, No. 3, May-
June 1991, pp. 268-276.
for d ≥ 250 mm (9.84 in.) 7. Kim, J. K., and Eo, S. H., “Size Effect in Concrete Specimens with
Dissimilar Initial Cracks,” Magazine of Concrete Research (London),
V. 42, No. 153, Dec. 1990, pp. 233-238.
8. Taylor, H. P. J., “Investigation of Forces Carried Across Cracks in
The accuracy of the simplified equation is nearly identical Reinforced Concrete Beams in Shear by Interlock of Aggregate,” Tech-
compared with the original equation and is considered to be nical Report No. 42.447, Cement and Concrete Association, London, Nov.
better than those of the other equations compared in this study. 1970, 22 pp.
9. Guralnick, S. A., “Shear Strength of Reinforced Concrete Beams,”
5. The effect of concrete strength on the shear strength of Proceedings, ASCE, V. 85, No. ST1, Jan. 1959, pp. 1-42.
beams with a/d < 3.0 is estimated well by introducing failure 10. Fenwick, R. C., and Paulay, T., “Mechanisms of Shear Resistance of
mode index α. Concrete Beams,” Proceedings, ASCE, V. 94, No. ST10, Oct. 1968, pp.
2325-2350.
11. Reinock, K. H., “Ultimate Shear Force of Structural Concrete Mem-
ACKNOWLEDGMENTS bers without Transverse Reinforcemt Derived from a Mechanical Model,”
Support for this research was partially provided by the Engineering Re-
search Center (ERC) on Advanced Structure Research (STRESS) of KOSEF.
The STRESS is gratefully acknowledged.

NOTATION
aeff = effective length of shear span
As = area of tension reinforcement
a/d = shear span-to-depth ratio
b = width of beam
bct = net width of beam section
c1 , c2, ..., c1 7 = constants
D = characteristic size of structure
d = effective depth of beam
da = maximum aggregate size
db = diameter of reinforcing bar
Ec = elastic modulus of concrete
Es = elastic modulus of steel
f ′c = compressive strength of concrete
f t′ = tensile strength of concrete
Gf = fracture energy
jd = internal lever arm
k 1 , k 2, ..., k 7 = constants
kd = neutral axis depth
l, m, p, s, t = constants
Mu = ultimate moment
n = ratio of elastic moduli of steel and concrete
scr = spacing of primary cracks Fig. 11—Comparison between experimental data and calcu-
Va = shear force carried by aggregate interlock lated values by using design equation

ACI Materials Journal / May-June 1996 221

ACI Structural Journal , V. 88, No. 5, Sept.-Oct. 1991, pp. 592-602. Strength of Reinforced Concrete Beams, Part 1—Tests of Simple Beams,”
12. Vintzeieou, E. N., and Tassios, T. P., “Mathematical Models for ACI J OURNAL, Proceedings V. 51, No. 3, Nov. 1954, pp. 317-332.
Dowel Action under Monotic and Cyclic Conditions,” Magazine of Con- 32. Mphonde, A. G., and Frantz, G. C., “Shear Tests of High- and
crete Research (London), V. 38, No. 134, Mar. 1986, pp. 13-22. Low-Strength Concrete Beams without Stirrups,” ACI J O URNA L, Pro-
13. Alimad, S. H.; Khaloo, A. R.; and Poveda, R., “Shear Capacity of ceedings V. 81, No. 4, July-Aug. 1984, pp. 350-357.
Reinforced High-Strength Concrete Beams,” ACI J O URN AL, Proceedings
V. 83, No. 2, Mar.-Apr. 1986, pp. 297-305. 33. Niwa, J.; Yamada, K.; Yokozawa, K.; and Okamura, H., “Re-Eval-
14. Akiyama et al., “FIexural and Shear Behavior of Beams with High- uation of the Equation for Shear Strength of Reinforced Concrete Beams
Strength Concrete and High-Strength Reinforcement,” Research Report, without Web Reinforcement,” Proceedings, JSCE, No. 372, Aug. 1986,
Institute of Kajima Consruction Technology, Japan, No. 38, Oct. 1990, pp. 167-176.
pp. 75-84. 34. Placas, A., and Regan, P. E., “Shear Failure of Reinforced Concrete
15. Bial, N. S., “Uber den Einfluss der Balkenhohe auf Schubtragfahigkeit Beams,” ACI JOURNAL, Proceedings V. 68, No. 10, Oct. 1971, pp. 763-773.
von einfledrigen Stahlbetonbalken mit und ohne Schubberwehrung,” dis- 35. Rajagopalan, K. S., and Ferguson, P. M., “Exploratory Shear Tests
sertation, Universitat Stuttgart, 1968, 124 pp.
Emphasizing Percentage of Longitudinal Steel,” ACI J OURNA L, Proceed-
16. Bresler, B., and Sedelis, A. C., “Shear Strength of Reinforced Con-
ings V. 65, No. 8, Aug. 1968, pp. 634-638.
crete Beams,” ACI J OURNAL, Proceedings V. 60, No. 1, Jan. 1963, pp. 51-74.
17. Chana, P. S., “Some Aspects of Modeling the Behavior of Rein- 36. Rusch, H.; Haugli, F. R.; and Mayer, H., “Schubversuche an
forced Concrete under Shear Loading,” Technical Report No. 543, Cement Stahlbeton-Rechteckbalken mit gleichmassing verteilter Belastung,” Bul-
and Concrete Association, July 1981, 22 pp. letin No. 145, Deutcher Ausschuss fur Stahlbeton, Berlin, 1962, pp. 4-30.
18. Diaz de Cossio, R., and Siess, C. P., “Behavior and Strength in Shear 37. Shioya, T., “Shear Properties of Large Reinforced Concrete Mem-
of Beams and Frames without Web Reinforcement,” ACI J O URN AL, Pro- bers,” Report No. 25, Institute of Technology of Shimizu Corporation,
ceedings V. 56, No. 8, Feb. 1960, pp. 695-735. Japan, 1989, 198 pp.
19. Elzanty, A. H.; Nilson, A. H.; and Slate, F. O., “Shear Capacity of 38. Taylor, H. P. J., “Shear Strength of Large Beams,” Proceedings,
Reinforced Concrete Beams Using High-Strength Concrete,” ACI
ASCE, V. 98, No. ST1 1, Nov. 1972, pp. 2473-2490.
J OU RN AL, Proceedings V. 83, No. 2, Mar-Apr. 1986, pp. 290-296.
20. Gaston, J. R.; Siess, C. P.; and Newmark, N. M., “An Investigation of 39. Taylor, R., “Some Shear Tests on Reinforced Concrete Beams
Load Determination Characteristics of Reinforced Concrete Beams Up to without Shear Reinforcement,” Magazine of Concrete Research (Lon-
the Point of Failure,” Civil Engineering Studies, Structural Research Series don), V. 12, No. 36, Nov. 1960, pp. 145-154.
No. 40, University of Illinois, Urbana, Dec. 1952. 40. Taylor, R., and Brewer, R. S., “Effect of the Type of Aggregate on
21. Ikeda, S., and Uji, K., “Studies on the Effect of Bond on the Shear Diagonal Cracking of Reinforced Concrete Beams,” Magazine of Con-
Behavior of Reinforced Concrete Beams,” Proceedings, JSCE, No. 293, crete Research (London), V. 15, No. 44, July 1963, pp. 87-92.
Jan. 1980, pp. 101-109. 41. Thorenfeldt, E., and Drangsholt, G., “Shear Capacity of Reinforced
22. Johnson, M. K., and Ramirez, J. A., “Minimum Shear Reinforcement
High-Strength Concrete Beams,” High-Strength Concrete—Second Inter-
in Beams with Higher Strength Concrete,” ACI Structural Journal , V. 86,
national Symposium, SP-121, American Concrete Institute, Detroit,
No. 4, July-Aug. 1989, pp. 376-382.
1990, pp. 129-154.
23. Kani, G. N. J., “Basic Facts Concerning Shear Failure,” ACI J OUR -
N AL, Proceedings V. 63, No. 6, June 1966, pp. 675-692. 42. van den Berg, F. J., “Shear Strength of Reinforced Concrete Beams
24. Kani, G. N. J., “How Safe Are Our Large Reinforced Concrete without Web Reinforcement, Part 2—Factors Affecting Load at Diagonal
Beams?” ACI J OURNAL, Proceedings V. 64, No. 3, Mar. 1967, pp. 128-141. Cracking,” ACI J OU RNAL, Proceedings V. 59, No. 11, Nov. 1962, pp.
25. Kim, J. K., and Park, Y. D., “Shear Strength of Reinforced High- 1587-1600.
Strength Concrete Beams without Web Reinforcement,” Magazine of Con- 43. Walraven, J. C., “Influence of Depth on Shear Strength of Light-
crete Research (London), V. 46, No. 166, Mar. 1994, pp. 7-16. weight Concrete Beams without Shear Reinforcement,” Stevin Labora-
26. Ko, K. I., and Kim, W., “Shear Strength of High- Strength Rein- tory Report No. 5-78-4, Delft University of Technology, 1972, 36 pp.
forced Concrete Beams,” Proceedings, KSCE, V. 9, No. 4, Dec. 1989, pp.
44. Walraven, J. C., “Size Effect: Their Nature and Their Recognition
63-71.
in Building Codes,”Proceedings of JCI International Workshop on Size
27. Krefeld, W. J., and Thurston, C. W., “Studies of Shear and Diagonal
Tension Strength of Simply Supported Reinforced Concrete Beams,” ACI Effect in Concrete Structures, Japan, 1993, pp. 295-314.
J OURNAL, Proceedings V. 63, No. 4, Apr. 1966, pp. 451-476. 45. Neville, A. M., “General Relation for Strengths of Concrete Speci-
28. Leonhardt, F., and Walther, R., “Beitrage zur Behandlung der Schub- mens of Different Shapes and Sizes,” ACI J O URNAL , Proceedings V. 63,
probleme in Stahlebtonbau,” Beton-und Stahlbetonbau (Berlin), Mar. No. 10, Oct. 1966, pp. 1095-1109.
1962, pp. 32-44, 54-64. 46. Zsutty, T. C., “Shear Strength Prediction for Separate Categories of
29. Mathey, R. G., and Watstein, D., “Shear Strength of Beams without Simple Beam Tests,” ACI J OURNA L, Proceedings V. 68, No. 2, Feb. 1971,
Web Reinforcement,” ACI JOURNAL , Proceedings V. 60, No. 2, Feb. 1963, pp. 138-143.
pp. 183-208.
30. Mattock, A. H., “Diagonal Tension Cracking in Concrete Beams 47. ACI Committee 318, “Building Code Requirements for Reinforced
with Axial Forces,” Proceedings, ASCE, V. 95, No. ST9, Sept. 1969, pp. Concrete (ACI 318-89),” American Concrete Institute, Detroit, 1989, 353 pp.
1887-1900. 48. Comite Euro-International du Beton (CEB), “CEB-FIP Model
31. Moody, K. G.; Viest, I. M.; Elstner, R. C.; and Hognestad, E., “Shear Code 1990,” Switzerland, July 1991.

222 ACI Materials Journal / May-June 1996