ACI STRUCTURAL JOURNAL TECHNICAL PAPER

Title no. 96-S31

Shear Strength of Normal and High-Strength Fiber

Reinforced Concrete Beams without Stirrups
by Madhusudan Khuntia, Bozidar Stojadinovic, and Subhash C. Goel

This paper presents a rational and unified procedure for pre- part is a summary of previously conducted research and experi-
dicting the shear strength of normal and high-strength fiber ments on normal and high-strength fiber reinforced concrete
reinforced concrete (FRC) beams. A design equation is sug- (FRC) beams. The second part offers a derivation of a rational
gested for evaluating the ultimate shear strength of FRC beams and unified design equation for evaluating the shear strength of
based on the basic shear transfer mechanisms and numerous normal and high-strength FRC beams.
published experimental data on concrete strength up to 100
MPa (14,500 psi). In addition to concrete strength, the influ- RESEARCH SIGNIFICANCE
ence of other variables such as fiber factor, shear Although several semi-empirical relations have been suggest-
span-to-depth ratio, longitudinal steel ratio, and size effect is ed to determine the ultimate shear capacity of steel FRC beams
considered. The modeling approach is similar to that applied (Table 1), few studies are available that give suitable and simple
for conventional reinforced concrete beams, except for some analytical models for the evaluation of shear strength of FRC
modifications suggested in this paper, to account for the effect beams without stirrups. More importantly, there exists no ex-
of the fibers. The comparison between computed values and pression to predict the shear strength of normal concrete and
experimentally observed values is shown to validate the pro- high-strength concrete beams containing fibers in a unified
posed analytical treatment. manner. In this paper, a general method of predicting the ulti-
mate shear strength of FRC beams is proposed. The method is
Keywords: fiber reinforced concretes; high-strength concretes; metal quite versatile. It can be applied across a practical range of con-
fibers; reinforcing steels; shear strength; tensile strength. crete strengths (20 to 100 MPa), fiber factor values (between
0.25 and 2), and shear span-to-depth ratios (more than 0.5).
INTRODUCTION Shear strengths of both normal and high-strength FRC beams
Shear failure of a reinforced concrete beam occurs when the predicted by using the proposed method were compared and
principal tensile stress within the shear span exceeds the tensile found to agree conservatively with the experimental results of
strength of concrete and a diagonal crack propagates through 10 previous investigations considered in the study. The results
the beam web. This failure is usually without any warning due of the study presented in this paper should enable designers and
to the brittle nature of plain concrete behavior in tension. The engineers to use structural steel fiber reinforced concrete with
addition of small discrete steel fibers into the concrete mix helps more confidence.
to improve the postcracking tensile strength of hardened con-
crete, and, hence, significantly enhances the shear strength of
reinforced concrete (RC) beams.
PREDICTION OF SHEAR STRENGTH
A number of semi-empirical equations are available in the lit-
Many reports published over the past 25 years confirm the ef-
erature to predict the ultimate shear capacity of normal and
fectiveness of steel fibers as shear reinforcement.1-16 Fibers are
high-strength FRC beams without stirrups, seven of which are
used to boost the shear capacity of concrete or to replace, in part,
given here (Table 1). The details of specimens used in the test
the vertical stirrups in RC structural members. This relieves re-
program by 10 different investigators are presented in Table 2.
inforcement congestion at critical sections such as beam-col-
umn junctions. Fiber reinforcement may also significantly
reduce construction time and costs, especially in an era of high FACTORS AFFECTING SHEAR STRENGTH OF
labor costs and possibly even labor shortages, since convention- FIBER REINFORCED CONCRETE (FRC) BEAMS
al stirrups require relatively high labor input to bend and fix in As far as shear strength is concerned, the major difference be-
place. Fiber concrete can also be easily placed in thin or irregu- tween the RC beams containing no fibers and those containing
larly shaped sections, such as architectural panels, where it may fibers lies in the significant postcracking tensile strength of
be very difficult to place stirrups. FRC. Therefore, it is rational to infer that the parameters influ-
Another potential area of fiber use is in high-strength con- encing the ultimate shear strength of FRC beams are those that
crete (with compressive strength of 50 MPa or more). The use affect the shear strength of conventional RC beams plus the
of high-strength concrete, which is growing rapidly, is attractive postcracking tensile strength or the split-cylinder strength of
for longer spans and taller structures, as well as for earthquake- FRC. This conclusion has been well established in the experi-
resistant structures of any size where a reduction of mass is of mental observations and analytical models proposed by other
paramount importance. Yet, application of high-strength con- investigators (Table 1). For conventional RC beams, the ACI
crete is hindered by its relative brittleness and lack of ductility.
This drawback can be overcome by inclusion of fibers in the
ACI Structural Journal, V. 96, No. 2, March-April 1999.
high-strength concrete mix. Received September 24, 1997, and reviewed under Institute publication policies.
A unified approach to shear strength evaluation for normal and Copyright  1999, American Concrete Institute. All rights reserved, including the
making of copies unless permission is obtained from the copyright proprietors. Perti-
high-strength fiber reinforced concrete beams without stirrups nent discussion will be published in the January-February 2000 ACI Structural Jour-
will be presented in this paper. This paper has two parts. The first nal if received by September 1, 1999 .

282 ACI Structural Journal/March-April 1999

ACI member Madhusudan Khuntia is a structural engineer with S. K. Ghosh Associ- V c = ( 2.0 f c′ )b w d (U.S. customary)
ates, Inc., Chicago, Ill. He received his BS and MS degrees from Regional Engineer-
ing College, Rourkela, India, and his PhD from the University of Michigan, Ann Eq. (1), while considering the influence of reinforcement ra-
Arbor, Mich. His research interests include application of fiber reinforced concrete tio ρ and equivalent shear span-to-depth ratio a/d, ignores the
and other novel materials in new construction, and behavior of steel-concrete com- size effect, whereas Eq. (2) is regarded as conservative for the
posite structures.
customary range of these parameters. In fact, considering the
ACI member Bozidar Stojadinovic is an assistant professor of civil engineering at practical values of reinforcement ratio (0.25 to 0.75 ρ b ) and
the University of Michigan. He received his PhD from the University of California at shear span-to-depth ratio (2.5 to 6) for a singly reinforced beam,
Berkeley. He is a member of ACI Committee 335, Composite and Hybrid Structures;
an associate member of ACI Committee 341, Earthquake-Resistant Concrete Bridges;
it can be shown that Eq. (1) gives a shear strength in the range
and a member of Joint ACI-ASCE Committee 445, Shear and Joint Torsion. His between 0.16 and 0.20 f c′ bw d, considering both normal and
research interests include earthquake-resistant design of reinforced concrete and high-strength concrete. Thus, it is not unreasonable to use Eq.
composite structures, nonlinear dynamic analysis, and methods for remote testing.
(2) for shear design. Eq. (2) becomes overly conservative, as ex-
Subhash C. Goel is a professor of civil engineering at the University of Michigan, pected, with higher reinforcement ratios ρ and lower shear
where he received his PhD in 1968. His research interests include seismic analysis span-to-depth ratios a/d. However, this should be regarded as a
rational design decision, considering the serious nature of shear
failure. For short beams (a/d < 2.5), it is essential to consider the
Building Code (Code Equation 11-5) presents the following
arch action of concrete compression struts. Using the factor sug-
equations for computing the shear strength of beams without
gested by Zsutty,18,19 which is based on an elaborate statistical
web reinforcement
study, one way to estimate the shear strength of short beams is

V c =  0.157 f c′ + 17.2 ρ V
------- b w d ≤ ( 0.29 f c′ ) b w d
d (1) d
V c = 2.5 -- × ( 0.167 f c′ ) b w d ≤ 0.5 f c′ bw d (SI)
 M (3)
a

in SI units; and
V c = 2.5 d-- × ( 2.0 f c′ ) b w d ≤ 6.0 f c′ b w d (U.S. Customary) (4)
a
V c =  1.9 f c′ + 2500 ρ -------  bw d ≤ ( 3.5 f c′ ) b w d
Vd
 M Thus, the major factor affecting the ultimate shear strength of
in U.S. customary units, or simply RC beams is the tensile strength of concrete or, indirectly, the
compressive strength of concrete, as they are considered to be
interrelated. For short spans, the other significant factor is the
V c = ( 0.167 f c′ ) b w d (SI) (2)
shear span-to-depth ratio ( a/d or M/Vd).
From the above, it can be inferred that the main parameters in-
fluencing the shear strength of FRC beams are compressive

Table 1—Summary of proposed shear strength equations for fiber

reinforced concrete (FRC) beams by different investigators*
Investigator Predictive equation for ultimate shear strength, MPa Application
d d High-strength fiber
Ashour et al. 1 vfrc = ( 0.7 fc′ + 7 F ) -- + 17.2 ρ --
a a reinforced concrete
1⁄ 3
vfrc = 1.25 + 4.68 ( ff f s p )
3⁄ 4 ρd
--  ( d )
–1 ⁄ 3
for a/d > 2.5
 a
Normal strength fiber
Li et al. 4 reinforced concrete
vfrc = 9.16 ( ff ) 2⁄ 3
( ρ) d
1⁄ 3
--  for a/d < 2.5
 a

ρV d
vfrc = 0.41  τ V f ----f  +  0.16 fc′ + 17.2 ----------
l Normal strength fiber
Mansur et al. 6  d f  M  reinforced concrete

d
vfrc = e 0.24 ( fs p ) + 80 ρ -- + 0.41 τ F Normal to medium
Narayanan et al.7 a strength fiber reinforced
concrete
e = 1 for a/d > 2.8; else e = 2.8 d/a

vfrc = kf t  d
-- 
0.25
Recommended by ACI Committee 544 17
 a
10 Normal strength fiber
Sharma k = 1 if ft is obtained by direct tension test; reinforced concrete
k = 2/3 if ft is obtained by indirect tension test;
k = 4/9 if ft is obtained using modulus of rupture.

vfrc = 0.22 f s p + 217 ρ  -- + 0.34τ F for a/d < 3

d
a High-strength fiber
Shin et al. 11 reinforced concrete
= 0.19 f s p + 93ρ  --  + 0.34 τ F for a/d ≥ 3
d
vfrc
 a

l
vfrc = 0.37 τ Vf ----f + vc Normal strength fiber
Swamy et al. 13 df reinforced concrete
factor of 2 d/a to multiply vc for a/d < 2
*In SI units.

ACI Structural Journal/March-April 1999 283

and design of steel and composite structures.

ACI Structural Journal/March-April 1999 284

 Table 2—Summary of beam details and comparison of experimental and
predicted shear stress values
Investigator Fiber Vf lf /d f F1 ρ d fc ′ a/d v frc vo vo / v frc
1.0 60 0.6 2.2 102 22.7 3.0 1.51 3.16 2.09
1.0 60 0.6 1.1 102 22.7 3.0 1.51 2.43 1.61
Li, Ward, and Crimped and 1.0 60 0.6 1.1 102 22.7 1.5 2.04 5.64 2.76
Hamza 4 hooked steel 1.0 100 1.0 2.2 102 26 3.0 2.13 3.55 1.67
1.0 60 0.6 2.2 204 22.7 3.0 1.51 3.05 2.02
1.0 100 1.0 2.2 204 26 3.0 2.13 3.05 1.43
0.5 60 0.3 1.34 197 29.1 2.0 1.53 2.54 1.66
0.5 60 0.3 1.34 197 29.1 2.8 1.31 1.78 1.36
Mansur, Ong, and 0.5 60 0.3 1.34 197 29.1 3.6 1.31 1.52 1.16
Hooked steel
Paramasivam6 0.75 60 0.45 2.0 197 29.9 2.8 1.53 2.2 1.44
0.75 60 0.45 2.0 197 20.6 2.8 1.27 2.03 1.6
0.75 60 0.45 2.0 197 33.4 2.8 1.62 2.91 1.8
0.5 60 0.3 1.1 221 34 2.5 1.41 1.73 1.23
0.5 60 0.3 2.2 221 34 1.5 2.06 4.02 1.95
Lim, Paramasivam, 0.5 60 0.3 2.2 221 34 2.5 1.41 1.9 1.35
Hooked steel
and Lee5 0.5 60 0.3 2.2 221 34 3.5 1.41 1.47 1.04
1.0 60 0.6 2.2 221 34 1.5 2.50 4.39 1.76
1.0 60 0.6 2.2 221 34 2.5 1.85 2.46 1.33
0.25 100 0.25 2.0 130 61 2.0 2.12 2.96 1.40
0.25 100 0.25 2.0 130 61 3.0 1.79 2.77 1.55
0.25 100 0.25 2.0 130 39.2 2.0 1.70 2.71 1.60
0.25 100 0.25 2.0 130 39.2 3.0 1.44 1.94 1.35
0.5 133 0.66 2.0 130 36 3.0 2.00 1.97 0.99
Narayanan and 1.0 100 1.0 2.0 130 36 3.0 2.50 2.97 1.19
Crimped steel
Darwish7 0.5 133 0.66 2.0 130 49 3.5 2.33 2.61 1.12
1.0 133 1.33 2.0 130 57.4 2.0 4.10 5.57 1.36
0.5 133 0.66 3.69 130 36 3.0 2.00 2.24 1.12
0.5 133 0.66 5.72 130 36 3.1 2.00 2.33 1.17
1.0 133 1.33 3.69 130 57.4 3.0 3.78 4.37 1.15
1.0 133 1.33 5.72 130 57.4 3.1 3.78 5.0 1.32
0.5 75 0.37 2.84 215 99 1.0 5.09 9.09 1.79
0.5 75 0.37 2.84 215 99 2.0 3.01 4.82 1.60
1.0 75 0.75 2.84 215 95 1.0 5.90 12.74 2.16
Ashour, Hasanain, 1.0 75 0.75 2.84 215 95 2.0 3.86 6.06 1.57
Hooked steel
and Wafa1 1.5 75 1.12 2.84 215 96 1.0 6.84 13.95 2.04
1.5 75 1.12 2.84 215 96 2.0 4.80 7.21 1.50
1.0 75 0.75 4.58 215 94 2.0 3.84 4.89 1.27
1.0 75 0.75 4.58 215 94 4.0 3.44 3.88 1.13
0.4 100 0.4 4.0 210 44.4 4.5 1.78 2.16 1.21
0.8 100 0.8 4.0 210 46.8 4.5 2.51 3.1 1.23
12 Crimped steel
Swamy and Bahia
0.8 100 0.8 3.05 210 47.7 4.5 2.53 3.22 1.27
1.2 100 1.20 4.0 210 49.8 4.5 3.29 3.13 0.95
0.5 100 0.5 3.55 350 60 0.7 4.85 9.42 1.94
1.0 100 1.0 3.55 350 60 0.46 5.82 13.16 2.26
1.0 100 1.0 3.55 350 60 0.92 5.82 9.97 1.71
Narayanan and
Crimped steel 1.0 100 1.0 3.55 350 67 0.7 6.15 11.48 1.87
Darwish8
1.0 100 1.0 3.55 350 38 0.7 4.63 8.52 1.84
1.0 100 1.0 3.55 350 42 0.7 4.87 9.65 1.98
1.25 100 1.25 3.55 350 68 0.7 6.71 11.39 1.70
0.5 100 0.33 3.59 175 80 2.0 2.62 6.84 2.62
0.5 100 0.33 3.59 175 80 3.0 2.24 3.19 1.43
Shin, Oh, and 0.5 100 0.33 3.59 175 80 4.5 2.24 2.78 1.24
Plain steel
Ghosh 11 1.0 100 0.66 3.59 175 80 2.0 3.36 7.4 2.20
1.0 100 0.66 3.59 175 80 3.0 2.99 4.1 1.37
1.0 100 0.66 3.59 175 80 4.5 2.99 3.44 1.15

285 March-April
 Table 2 (cont.)—Summary of beam details and comparison of experimental
and predicted shear stress values
Investigator Fiber Vf lf / df F1 ρ d fc′ a/d v frc vo v o / vfrc
0.75 60 0.45 2.15 557 54 1.35 2.05 3.3 1.61
1.5 60 0.9 2.15 557 50 1.35 2.77 3.87 1.40
Adebar,* Mindess, 0.4 60 0.24 2.15 557 55 1.35 1.68 2.44 1.45
St. Pierre, Hooked steel
and Olund 20 0.6 60 0.36 2.15 557 56 1.35 1.92 2.77 1.44
0.4 100 0.4 2.15 557 47 1.35 1.83 2.95 1.61
0.6 100 0.6 2.15 557 41 1.35 2.03 2.83 1.39
0.5 50 0.17 1.22 186 28.7 2.0 1.20 1.64 1.22
0.5 100 0.33 1.22 186 32.2 2.0 1.57 1.94 1.17
1.0 50 0.33 1.22 186 29.0 2.0 1.49 2.18 1.39
Murty and
Plain steel 1.0 100 0.66 1.22 186 32.6 2.0 2.18 2.73 1.27
Venkatacharyulu21
1.0 50 0.33 1.22 186 32.1 3.0 1.45 1.58 1.12
1.0 100 0.66 1.22 186 32.3 3.0 1.95 1.98 1.04
1.5 50 0.50 1.22 186 32.8 3.0 1.70 2.42 1.45
Mean† 1.51
Standard deviation 0.37
Median 1.43
*
Arch action is not taken into account because of different testing arrangement.
†Neglecting data for short beams with a/d ≤ 2, mean, standard deviation, and median were computed to be 1.34, 0.27, and 1.30,

respectively.

strength of concrete, shear span-to-depth ratio, and postcracking F 1 = fiber factor expressed as
tensile strength or split-cylinder strength of FRC. The effects of
these factors, as well the influence of the reinforcement ratio lf
and the beam size, are considered in the following sections. F 1 = β V f ---- (6)
df
Postcracking tensile strength of FRC
Before discussing the proposed predictive equation for shear In Eq. (5), it is assumed that the fiber length is smaller than the
strength of FRC beams, it is important to quantify the postcrack- critical fracture length lc given as l c = σ fu × df /2τ , where σ fu = the
ing tensile strength of FRC in a simplified way. Direct tensile fiber fracture stress. This condition is usually satisfied for the
strength of FRC will be used in the following derivations. commonly used steel fibers and concrete strength.
Test data by many investigators (Tables 1 and 2) confirm that As can be seen in Eq. (5), one of the significant parameters in
there is considerable reserve strength in FRC beams failing in the expression for postcracking tensile strength is τ , the fi-
shear after the appearance of first diagonal cracking. The differ- ber-matrix interface bond stress, which has not been satisfacto-
ence is attributed to significant postcracking tensile strength of rily quantified to date. Swamy et al. 12,13 suggested a value of
FRC compared to that of plain concrete. This residual tensile 5.12 MPa, based on the test results for normal FRC containing
stress σ pc depends, among many factors, on the volume fraction, crimped steel fibers. Furthermore, based on their own data and
shape, aspect ratio, and surface characteristics of the fiber, and those of LaFraugh and Mustafa,27 they suggested a value for τ
on the properties of the concrete matrix. Extensive experimental of 4.15 MPa for plain fibers. An interfacial bond stress of 4.15
investigations and analytical studies5,13,22-2 6 have shown that the MPa has been considered by Narayanan and Darwish,7 as well
postcracking tensile strength σ pc can be expressed in the follow- as by Ashour et al., 1 for the computation of shear strength of
ing form FRC beams. Lim et al. 5 used a τ value of 6.8 MPa in their study.
Thus, no explicit relations between the fiber-matrix interfacial
l bond stress τ and the concrete strength fc ′ can be identified. This
σ p c = A β τ V f ----f = A τ F 1 (5) is a major handicap in generalizing the shear strength equations
df
proposed by these investigators for various concrete strengths.
However, a comprehensive study by Naaman and Najm28 on
where bond-slip mechanisms of steel fibers in concrete demonstrated
A = nondimensional constant depending on the orientation that there is a definite increase in interfacial bond stress with the
factor and length efficiency factor of fibers, taken as 0.41.5,13 increase in compressive strength of concrete. In addition, based
τ = fiber-matrix interfacial bond stress on an extensive experimental and analytical study on normal
β = factor for fiber shape and concrete type, taken as 1 for and high-strength FRC, Valle and Buyukozoturk16 pointed out
hooked or crimped steel fibers, 2/3 for plain or round that the interfacial bond stress increases with concrete strength
steel fibers with normal concrete, 3/4 for hooked or because of superior bond characteristics between the fiber and the
crimped steel fibers with lightweight concrete. Factor β matrix for higher-strength concrete. Their results showed that di-
has a similar influence as the bond factor suggested by rect shear strength of FRC is more or less proportional to the
Narayanan and Darwish.7 Though the expression in Eq. square root of concrete strength. Furthermore, in the shear
(5) has been derived for steel fibers, similar equations can strength expression suggested by Sharma 10 (which has been rec-
be contemplated for other types of fibers with modifications ommended by the ACI Committee 544),17 the tensile strength of
in β and τ FRC can be taken proportional to f c′ .
V f = volume percentage Table 3 shows that there exists a reasonable relation between
l f = length the fiber-matrix interfacial bond stress τ and the quantity f c′ .
df = diameter of fiber As seen from Table 3, the minimum ratio τ / f c′ is 0.75 for both

ACI Structural Journal/March-April 1999 286

 Table 3—Relation between fiber-matrix interfacial
bond stress and concrete strength
Concrete Fiber-matrix Ratio
strength fc′, interfacial bond
Investigator Fiber type MPa stress τ, MPa τ/ f c′
33.4 4.31 0.75
28 Hooked
Naaman et al. 51 5.87 0.82
steel
60 7.5 0.97
Crimped
Swamy et al. 12,13 47 5.12 0.75
steel
Hooked 5.3 to 0.91 to
Lim et al. 5,22 34
Fig. 1—Contribution of fibers to shear resistance of FRC beam steel 8.0 1.38
without stirrups.

the normal and high-strength FRC. In another study, Naaman23 V f r = 0.9db w × σ pc = 0.25F 1 f c′ b w d (9)
suggested a typical value of τ as 2 times the tensile strength of
plain concrete for hooked steel fibers. Taking tensile strength of
concrete as 0.33 f c′ (4.0 f c′ , if expressed in psi), the interfa- using Eq. (7) for FRC.
cial bond stress τ amounts to 0.66 f c′ . In this study, the value The concrete contribution, which includes resistance of com-
of the ratio was taken as 0.68 for simplicity, as seen later. This pressed concrete V cc, aggregate interlock Va , and dowel action
value may sometimes underestimate the true tensile strength, of reinforcement V d , can be taken equal to that given by the fol-
especially for high-strength FRC. lowing ACI Building Code equation
The foregoing discussion indicates that the postcracking ten-
sile strength of FRC is a function of the fiber factor F 1 and the V c = 0.167 f c′ b wd
compressive strength of concrete and can be conservatively ex-
pressed as Thus, the ultimate shear resistance of FRC beams can be ex-
pressed as

σ p c = 0.41 F 1 × 0.68 f c′ = 0.28F 1 f c′ , MPa (7)

V f r c = V c + V fr
or
σ p c = 3.35 F 1 f c′ , psi (8)
V f r c = 0.167 f c′ b w d + 0.25F 1 f c′ b w d (10)
Thus, given the fiber content, fiber aspect ratio, and fiber
shape, the postcracking strength of FRC is proportional to f c′ .
As is well known, the other measures of tensile strength for = ( 0.167 + 0.25 F 1 ) f c′ b w d
plain concrete, such as the modulus of rupture and the split cyl- In general
inder strength, are also considered to be proportional to f c′ in
the North American Codes. In the above expressions, the com-
pressive strength of FRC (with fiber content of 2 percent or less) V f r c = ( 0.167 α + 0.25F 1 ) f c′ b w d , (SI) (11)
may be taken equal to that of normal concrete, as there is only a
marginal increase in strength (between 0 and 15 percent) as re-
ported by Fanella and Naaman.29 V f r c = ( 2.0 α + 3.0F 1 ) f c′ b w d , (U.S. customary) (12)

PROPOSED PREDICTIVE EQUATION FOR SHEAR For short beams, the arch action factor α (= 2.5d/a < 3 for a/
STRENGTH OF FRC BEAMS d < 2.5) is to be applied only to the concrete contribution, as the
The development of a general and simple formula to predict fibers contribute little to the arch action in FRC beams.
the shear strength of normal and high-strength FRC beams is It may be noted that the above simplified equations for ulti-
critical for successful application of fibers as shear reinforce- mate shear strength resemble those presented by Swamy13 and
ment in practice. Unfortunately, shear in reinforced concrete is Mansur 6 (Table 1). The major and significant difference lies in
a complex phenomenon that has so far defied purely analytical the fact that their expressions did not consider application to
prediction. Even more rational analyses, such as the modified high-strength FRC. Also, they suggest no explicit relation be-
compression field theory,30 contain important semi-empirical tween concrete strength and the postcracking tensile strength of
expressions, such as expressions for stress-strain relationship of FRC. Thus, the proposed equation is a simplified expression
cracked concrete in tension and aggregate interlock shear. Thus, that can be conveniently and uniformly applied for all practical
there exists a wide divergence of opinions, design approaches, ranges of concrete strength.
and code equations related to the shear strength of reinforced
concrete members. The differences between the proposed shear Comparison of predicted and experimental shear
strength equations are the result of considerable scatter of the strength
experimentally observed shear strengths. A consequence of this A comparison of the predicted shear strength using the pro-
lack of fit to experimental data is the acceptance of conservative posed simplified equation [Eq. (12)] and the experimental mea-
lower limits for code equations. sured failure shear stress as reported in 10 different investigations
Considering the forces acting at a diagonal crack in FRC is shown in Table 2. The experimental data are obtained for a
beam without stirrups (Fig. 1) and assuming a crack inclination wide range of variables: compressive strength of concrete rang-
of 45 deg, it can be shown that the total shear resistance of FRC ing between 20 to 100 MPa, shear span-to-depth ratio ranging
beams is the sum of shear contribution from concrete and shear between 0.5 to 6, fiber factor ranging between 0.25 to 1.33, fiber
contribution from the fibers. As seen in Fig. 1, the fiber contri- volume content ranging between 0.5 and 2 percent, and longitu-
bution can be expressed as dinal reinforcement ratio ranging between 1.1 to 5.7 percent.

287 March-April
 Fig. 3—Comparison of different equations for normal and
high-strength FRC.

Fig. 2—Comparison of experimental and predicted shear

Table 4—Comparison between different shear Fig. 4—Comparison of different equations for different shear
strength equations* span-to-depth ratios.
Standard devi-
Investigator Mean ation Median resents the experimental data better. Table 4 shows the statistical
Swamy et al. 13 2.22 0.63 2.12
properties of the comparison of different shear strength equations
and the 34 experimental data points (for a/d values > 2.5).
Narayanan and 2.44
Darwish7
0.77 2.28 The behavior of the proposed equation and those proposed by
Swamy et al.13 and Narayanan and Darwish7 for typical values
Proposed 1.34 0.27 1.30
of fiber factor, tensile reinforcement ratio, shear span-to-depth
*Based on 34 experimental data points (for a/d values ≥ 2.5).
ratio, and compressive strength of FRC is illustrated in Fig. 5
and 6. Fig. 5 shows that the proposed expression is more conser-
The same comparison between analytical and test results is vative than other two expressions for lower strength FRC and
shown in Fig. 2. less conservative for higher strength. This is mainly due the as-
This comparison demonstrates that the experimental strength sumption of a constant fiber-matrix interfacial bond stress by
is almost always higher than that predicted analytically using the Swamy et al. 13 and Narayanan and Darwish,7 irrespective of the
proposed equation (only two out of 68 strength ratios are margin- compressive strength of the matrix.
ally smaller than 1), more so for lower a/d ratios. Moreover, con-
sidering a shear strength reduction factor of 0.85 (as Effect of longitudinal reinforcement ratio and
recommended by the ACI Building Code for shear),31 the pre- beam size
dicted strength is found always to be conservative. Fig. 2(d) and 2(e) show the variation of ratio of test-to-predicted
There is a scatter of test data for all concrete strengths and all shear strength with change in reinforcement ratio ρ and beam effec-
fiber factors. The influence of concrete strength and fiber factor tive depth d, respectively. In conventional RC beams, there is a
is shown to be adequately represented by the proposed equation. concern in using the ACI simplified equation for shear strength
However, the equation is found to be more conservative for lower [Eq. (2)] when the reinforcement ratio ρ is very small or the
a/d ratio, i.e., for deep beams. The use of a strut-and-tie model beam depth d is very large, as both factors tend to reduce ulti-
may give better correlation for such beams.32 Nevertheless, the mate shear strength. However, such cases in FRC beams, when
results confirm that the proposed equation for computing ulti- treated by using the proposed model, remain on the conservative
mate shear strength of FRC beams can be used with confidence side. This is because a decrease in the safety margin associated
in design practice for all practical ranges of concrete strength, fi- with the concrete contribution (V c) is likely to be compensated
ber factor, and shear span-to-depth ratio. by the extra safety margin associated with the fiber contribution
Fig. 3 and 4 show the comparison of the proposed equation (V fr). The extra safety margin arises from the conservative esti-
with the expressions suggested by Swamy et al.13 and Narayanan mate of postcracking tensile strength of FRC, as discussed in
and Darwish.7 The figures show that the proposed equation rep- preceding sections. Fig. 2(d) and 2(e) substantiate this argument.

ACI Structural Journal/March-April 1999 288

 es of fiber factor (0.25 to 2), concrete strength (3000 to 14500
psi or 20 to 100 MPa) and the shear span-to-effective depth (a/
d) ratio (0.5 to 8) including deep beams.

NOTATION
a = shear span, mm
a/d = shear span/effective depth ratio
A = nondimensional constant in Eq. (5)
As = area of tensile reinforcement, mm2
b,bw = width of web, mm
c = depth of neutral axis, mm
d = effective depth, mm
df = diameter of fiber, mm
ff = flexural strength of fiber concrete beam, MPa
fsp = split cylinder strength of FRC, MPa
ft = tensile strength of FRC, MPa
fc′ = compressive strength of concrete (here FRC), MPa
F = fiber factor suggested by Narayanan et al. equal to bond
factor × Vf lf /df
F1 = fiber factor equal to β V f lf /d f
Fig. 5—Influence of concrete strength on predicted shear h = overall depth, mm
lf = length of fiber, mm
lf /df = fiber aspect ratio
M = bending moment at section, N-mm
v frc = ultimate shear stress in FRC beam (= V frc /bd ), MPa
vc = concrete contribution toward shear strength (= Vc /bd ), MPa
vo = observed ultimate shear stress of FRC beam (= V o/b d), MPa
V = Va + V cc + Vd , shear force at section, N
Va = aggregate interlock shear resistance contribution, N
Vcc = shear contribution in compressed concrete, N
Vc = concrete contribution towards ultimate shear strength of
FRC beams, N
Vcr = first shear crack strength, N
Vd = dowel shear resistance contribution, N
Vfrc = shear strength of FRC beam without stirrups, N
Vf = volume fraction of fiber, percent
Vfr = fiber contribution towards ultimate shear strength of
FRC beams, N
Vo = experimentally observed shear strength at failure, N
α = factor for arch action
β = bond factor considering shape and surface characteristics
of fiber
ρ = tensile reinforcement ratio equal to A s /bd, percent
Fig. 6—Influence of a/d ratio on predicted shear strength. σcc = precracking tensile strength of FRC, MPa
σmu = tensile strength of concrete matrix, MPa
σpc = postcracking tensile strength of FRC, MPa
CONCLUSIONS τ = fiber-matrix interfacial bond stress, MPa
From an extensive literature review and an analytical study
presented in this paper, the following conclusions can be drawn. REFERENCES
1. Inclusion of steel fibers in the concrete mix improves the 1. Ashour, S. A.; Hasanain, G. S.; and Wafa, F. F., “Shear Behavior of
shear strength of RC beams and tends to change the mode of High-Strength Fiber Reinforced Concrete Beams,” ACI Structural Journal,
failure from brittle shear to ductile flexure. V. 89, No. 2, Mar.-Apr. 1992, pp. 176-184.
2. Batson, G.; Jenkins, E.; and Spatney, R., “Steel Fibers as Shear Rein-
2. The ultimate shear strength of FRC beams increases with forcement in Beams,” ACI JOU RNAL , Proceedings V. 69, No. 10, Oct. 1972,
an increase in fiber factor and concrete strength. For short pp. 640-644.
beams (a/d < 2.5), it increases with a decrease in a/d ratio due 3. Jindal, R. L., “Shear and Moment Capacities of Steel Fiber Reinforced
to the presence of arch action. Concrete Beams,” Fiber Reinforced Concrete, SP-81, American Concrete
Institute, Farmington Hills, Mich., 1984, pp. 1-16.
3. Addition of fibers is more beneficial for high-strength con- 4. Li, V.; Ward, R.; and Hamza, A. M., “Steel and Synthetic Fibers as
crete in comparison with normal strength concrete. Shear Reinforcement,” ACI Materials Journal, V. 89, No. 5, Sept.-Oct.
4. The ultimate shear strength of FRC beams without stirrups 1992, pp. 499-508.
can be conservatively computed by using the following simpli- 5. Lim, T. Y.; Paramsivam, P.; and Lee, S. L., “Shear and Moment
Capacity of Reinforced Steel-Fiber-Concrete Beams,” Magazine of Con-
fied equation crete Research , V. 39, No. 140, Sept. 1987, pp. 148-160.
6. Mansur, M. A.; Ong, K. C. G.; and Paramsivam, P., “Shear Strength of
Fibrous Concrete Beams without Stirrups,” Journal of Structural Engineer-
v f r c = ( 0.167 α + 0.25 F 1 ) f c′ , MPa (13) ing , V. 112, No. 9, Sept. 1986, pp. 2066-2079.
7. Narayanan, R., and Darwish, I. Y. S., “Use of Steel Fibers as Shear
Reinforcement,” ACI Structural Journal, V. 84, No. 3, May-June 1987, pp.
v f r c = ( 2.0 α + 3.0 F 1 ) f c′ , psi 216-227.
8. Narayanan, R., and Darwish, I. Y. S., “Fiber Concrete Deep Beams in
Shear,” ACI Structural Journal, V. 85, No. 2, Mar.-Apr. 1988, pp. 141-149.
where the arch action factor α = 1 for a/d ≥ 2.5; 2.5 d/a ≤ 3 for 9. Niyogi, S. C., and Dwarakanath, G. I., “Fiber Reinforced Beams
under Moment and Shear,” Journal of Structural Engineering, V. 111, No.
a/d < 2.5, and the fiber factor F 1 = Vf (lf /d f ) for hooked steel fi- 3, Mar. 1985, pp. 35-40.
bers. 10. Sharma, A. K., “Shear Strength of Steel Fiber Reinforced Concrete
Thus, given fiber factor (or fiber volume percentage and as- Beams,” ACI JO URN AL , Proceedings V. 83, No. 4, July-Aug. 1986, pp.
pect ratio), compressive strength of concrete, and shear 624-628.
11. Shin, S. W.; Oh, J.; and Ghosh, S. K., “Shear Behavior of Labora-
span-to-depth ratio, the shear strength of steel FRC beams can tory-Sized High-Strength Concrete Beams Reinforced with Bars and Steel
be confidently predicted for design purposes using the equation Fibers,” Fiber Reinforced Concrete Developments and Innovations,
proposed herein. The equation can be used for all practical rang- SP-142, American Concrete Institute, Farmington Hills, 1994, pp. 181-200.

289 March-April
 12. Swamy, R. N., and Bahia, H. M., “Effectiveness of Steel Fibers as Concrete Technology Associates, Tacoma, Wash., Jan. 1975, 52 pp.
Shear Reinforcement,” Concrete International: Design and Construction, 28. Naaman, A. E., and Najm, H., “Bond-Slip Mechanisms of Steel
V. 7, No. 3, Mar. 1985, pp. 35-40. Fibers in Concrete,” ACI Materials Journal, V. 88, No. 2, Mar.-Apr. 1991,
13. Swamy, R. N.; Jones, R.; and Chiam, A. T. P., “Influence of Steel pp. 135-145.
Fibers on the Shear Resistance of Lightweight Concrete T-beams,” ACI 29. Fanella, D. A., and Naaman, A. E., “Stress-Strain Properties of Fiber
Structural Journal , V. 90, No. 1, Jan.-Feb. 1993, pp. 103-114. Reinforced Mortar in Compression,” ACI J OURN AL , Proceedings V. 82, No.
14. Tan, K. H.; Murugappan, K.; and Paramasivam, P., “Shear Behavior 4, Jul.-Aug. 1985, pp. 475-483.
of Reinforced Steel Fiber Concrete Beams,” ACI Structural Journal , V. 90, 30. Vecchio, F. J., and Collins, M. P., “The Modified Compression-Field
No. 1, Jan. 1993, pp. 3-11. Theory for Reinforced Concrete Elements Subjected to Shear,” ACI JO UR -
15. Tan, K. H.; Paramasivam, P.; and Murugappan, K., “Steel Fibers as NA L, Proceedings V. 83, No. 2, Mar.-Apr. 1986, pp. 219-231.
Shear Reinforcement in Partially Prestressed Beam,” ACI Structural Jour- 31. ACI Committee 318, “Building Code Requirements for Reinforced
nal , V. 92, No. 6, Nov.-Dec. 1995, pp. 643-652. Concrete (ACI 318-95) and Commentary (318R-95),” American Concrete
16. Valle, M., and Buyukozturk, O., “Behavior of Fiber Reinforced High Institute, Farmington Hills, Mich., 1995, 369 pp.
Strength Concrete under Direct Shear,” Fiber Reinforced Concrete Devel- 32. MacGregor, J. G., “Reinforced Concrete: Mechanics and Design,”
opments and Innovations , SP-142, American Concrete Institute, Farming- Prentice Hall, 3rd ed., 1997, 939 pp.
ton Hills, Mich., 1994, pp. 201-233. 33. ACI-ASCE Committee 426, “Shear Strength of Reinforced Concrete
17. ACI Committee 544, “Design Consideration for Steel Fiber Rein- Members,” Proceedings, V. 99, ST6, June 1973, pp. 1091-1187.
forced Concrete (ACI 544.4R-88),” ACI Structural Journal, V. 85, No. 5, 34. ACI Committee 363, “State of the Art Report on High-Strength Con-
Sept.-Oct. 1988, pp. 563-580. crete (ACI 363R-84),” ACI JOU RNAL , Proceedings V. 81, No. 4, July-Aug.
18. Zsutty, T., “Beam Shear Strength Prediction by Analysis of Existing 1984, pp. 364-411.
Data,” ACI JOU RNAL , Proceedings V. 65, No. 11, Nov. 1968, pp. 943-951. 35. Hsu, T. T. C., “Unified Theory of Reinforced Concrete,” CRC Press,
19. Zsutty, T., “Shear Strength Prediction for Separate Categories of 1st ed., 1993.
Simple Beam Tests,” ACI J OURN AL , Proceedings V. 68, No. 2, Feb. 1971, 36. Naaman, A. E., and Homrich, J. R., “Properties of High-Strength
pp. 138-143. Fiber Reinforced Concrete,” High-Strength Concrete, SP-87, H. Russel,
20. Adebar, P., et al., “Shear Tests of Fiber Concrete Beams without Stir- ed., American Concrete Institute, Farmington Hills, Mich., 1985, pp.
rups,” ACI Structural Journal , V. 94, No. 1, Jan.-Feb.1997, pp. 68-76. 233-249.
21. Murty, D. S. R., and Venkatacharyulu, T., “Fiber Reinforced Concrete 37. Oh, B. H., “Flexural Analysis of Reinforced Concrete Beams Con-
Beams Subjected to Shear Force,” Proceedings of the International Sympo- taining Steel Fibers,” Journal of Structural Engineering, V. 118, No. 10,
sium on Fiber Reinforced Concrete, Dec. 1987, Madras, India, pp. Oct. 1992, pp. 2821-2836.
1.125-1.132. 38. Swamy, R. N., and Al-Taan, S. A., “Deformation and Ultimate
22. Lim, T. Y.; Paramsivam, P.; and Lee, S. L., “Analytical Model for Strength in Flexure of Reinforced Concrete Beams Made with Steel Fiber
Tensile Behavior of Steel-Fiber Concrete,” ACI Materials Journal, V. 84, Concrete,” ACI JOURN AL , Proceedings V. 78, No. 5, Sept.-Oct. 1981, pp.
No. 4, July-Aug. 1987, pp. 286-298. 395-405.
23. Naaman, A. E., “High-Performance Fiber Reinforced Cement Com- 39. Wafa, F. F., and Ashour, S. A., “Mechanical Properties of
posites,” Proceedings of the IABSE Symposium on Concrete Structures for High-Strength Fiber Reinforced Concrete,” ACI Materials Journal, V. 89,
the Future, Paris, France, Sept. 1987, pp. 371-376. No. 5, Sept.-Oct.1992, pp. 449-455.
24. Naaman, A. E.; Moavenzadeh, F.; and McGarry, F. J., “Probabilistic
Analysis of Fiber-Reinforced Concrete,” Journal of Engineering Mechan-
ics Division , V. 100, No. EM2, Apr. 1974, pp. 397-413.
25. Nammur, G. G., and Naaman, A. E., “Strain Rate Effects on Tensile
Properties of Fiber Reinforced Concrete,” Proceedings of MRS Symposium
on “Cement Based Composites: Strain Rate Effects on Fracture,” V. 64, S.
Mind ess, ed., Materials Research Society, Pittsburg, Pa. 1986, pp. 97-118.
26. Swamy, R. N., and Mangat, P. S., “Interfacial Bond Stress in Steel
Fiber Cement Composites,” Cement and Concrete Research, V. 6, 1976,
pp. 641-650.
27. LaFraugh, R. W., and Moustafa, S. E., “Experimental Investigation
of the Use of Steel Fibers for Shear Reinforcement,” Technical Report,

ACI Structural Journal/March-April 1999 290