Engineering Structures 171 (2018) 348–361

Contents lists available at ScienceDirect

Engineering Structures
journal homepage: www.elsevier.com/locate/engstruct

An analytical approach to predict shear capacity of steel fiber reinforced T

concrete coupling beams with small span-depth ratio
⁎
Jun Zhaoa, Kou Lia, Fuqiang Shenb, Xiangcheng Zhanga, , Chenzhe Sia
a
School of Mechanics and Engineering Science, Zhengzhou University, Zhengzhou, China
b
School of Civil Engineering, Zhengzhou University, Zhengzhou, China

A R T I C LE I N FO A B S T R A C T

Keywords: Coupling beams with span-depth ratio no more than 2.5 tend to fail in shear-dominant rather than in flexure. The
Steel fiber reinforced concrete use of steel fiber in reinforced concrete coupling beams is considered as an available method to improve the
Coupling beams shear strength and seismic behavior of coupling beams. This paper proposed a theoretical approach based on
Small span-depth ratio strut-and-tie model to predict the shear capacity of steel fiber reinforced concrete (SFRC) coupling beams with
Strut-and-tie model
small span-depth ratio. To develop the model, the shear transfer mechanism was studied based on the test results
Shear capacity
of steel fiber reinforced concrete coupling beams with span-depth ratio no more than 2.5. The proposed model
considered the effect of span-depth ratio, steel fiber with different type and volume fraction and concrete
compressive strength on shear capacity. The shear predictions of the coupling beams were compared with the
experimental values, and results indicated good accuracy of the proposed model.

1. Introduction reinforced concrete coupling beams to simplify the diagonally re-

Coupling beams with small span-depth ratio no more than 2.5 are
usually used in structural design of coupled walls or core walls in order
to meet the requirements of lateral stiffness and seismic performance.
The most common reinforcement forms in coupling beams are con-
ventional reinforcement with longitudinal and transverse bars, or di-
agonal reinforcement. A number of experimental investigations of
coupling beams have been completed to study the shear behavior and
seismic performance of reinforced concrete coupling beams with con-
ventional reinforcement and diagonal reinforcement [1,2]. It is be-
lieved that the coupling beams with diagonally distributed reinforce-
ment can resist higher shear loads and dissipate more earthquake
energy than those using conventional reinforcements. However, the
diagonal reinforced coupling beams are usually configured with a large
amount of steel bars which brings new problems, such as construction
inconvenience, especially used in the coupling beams with small span-
depth ratio. The addition of the reinforcement also causes cast difficulty
of concrete in beam-adjacent wall joint region [3].
Some experimental works focusing on high performance concrete
coupling beams have been performed to solve this problem, such as
using steel fibers to replace partially reinforcement in reinforced con-
crete coupling beams without diagonal reinforcements. Canbolat et al.
[4] demonstrated that the use of high performance fiber reinforced
cement composites could be an alternative design method for
inforcement requirements. In his research study, ultra-high molecular
weight polyethylene (PE) fibers with volume fraction 2.0% was used in
short coupling beams to replace the diagonal reinforcement. It is shown
that the use of HPFRCC allow a simplification of diagonal reinforce-
ment detailing in RC coupling beams while ensuring a stable seismic
behavior. Chaallal et al. [5] have conducted an experiment to study the
seismic behavior of steel fiber reinforced concrete coupling beams
under reversed cyclic loading for the first time. The first specimen was
designed in conventional reinforcement with longitudinal and trans-
verse bars according to standard for design of concrete structures in
Canada. The transverse reinforcements used in the second specimen
were reduced and replaced with steel fibers with volume fraction of
0.76%. The test results showed that the hysteresis curve of the second
specimen was more plump and stable, and its energy dissipation ca-
pacity was 32% more than that of the first specimen, and it also had the
higher shear capacity compared with the first one. Zhang [6] experi-
mentally investigated steel fiber reinforced high-strength concrete
coupling beams with small span-depth ratio to study the effect of span-
depth ratio (from 1.0 to 2.5) and volume fraction of steel fibers (from
0.5% to 1.5%) on ductility and energy dissipation capacity. The con-
crete used in these specimens was C60 with end-hooked steel fiber. He
stated that conventional reinforced high-strength concrete coupling
beams with moderate volume fraction of steel fibers performed better in
ductility and energy dissipate than those without steel fibers. Cai et al.

⁎
Corresponding author at: School of Mechanics and Engineering Science, Zhengzhou University, Science Avenue 100#, Zhengzhou 450001, China.
E-mail address: xczhang@zzu.edu.cn (X. Zhang).

https://doi.org/10.1016/j.engstruct.2018.05.072
Received 17 January 2018; Received in revised form 17 May 2018; Accepted 17 May 2018
Available online 01 June 2018
0141-0296/ © 2018 Elsevier Ltd. All rights reserved.
J. Zhao et al.
[7] stated that the addition of crimped steel fibers could increase the
energy dissipation and deformation capacity of reinforced concrete
coupling beams when the concrete grade was C50. In his study, the
pinched hysteretic response of conventional reinforced concrete cou-
pling beams with small span-depth ratio were improved significantly
when volume fraction of steel fibers increased to more than 2.0%.
Some studies considered the possibility of using steel fibers as shear
reinforcement in concrete elements. The research presented by Dinh
et al. [8] was aimed at experimentally investigating the behavior and
shear strength of SFRC beams. Test results showed that the use of
hooked steel fibers in a volume fraction greater than or equal to 0.75%
could be used in place of the minimum stirrup reinforcement required
by ACI Committee 318. A series of panel tests were performed to assess
the effectiveness of steel fibers as a possible replacement for conven-
tional transverse reinforcement [9]. The test results indicated that the
fiber volume fraction of approximately 1.0% was required to achieve
satisfactory performance in terms of shear strength. An experimental
study on steel fiber reinforced concrete (SFRC) beams was presented to
investigate the effect of steel fibers on shear behavior without con-
ventional shear reinforcement, and results show that a relatively low
amount of fibers could significantly improve the shear strength of
concrete beams [10].
The regions that do not satisfy Bernoulli’s hypothesis are referred to
as Discontinuity regions (D-regions), which commonly occur at mem-
bers with abrupt changes in cross section or the presence of con-
centrated loads. In such regions, the shear resistance is no longer uni-
form over the depth of the regions and cannot be modeled based on the
classical beam theory. Strut-and-tie model could be considered as a
useful method to design and analyze D-region members under com-
plicated flows of loads. The main force flow in D-regions is mainly
through a major compression diagonal and the failure is usually gov-
erned by crushing the diagonal compressive strut [11]. The coupling
beam with small span-depth ratio is a typical example of D-region due
to the complex stress distributions and nonlinear strain inside them.
Such deep beam undertakes higher shear stress than normal beam and
the strength is usually controlled by shear rather than flexure if the
longitudinal reinforcement is used in a normal amount [12,13]. It is
essential to understand the shear behavior of coupling beams with small
span-depth ratio.
Strut-and-tie model design reduces the complex stress and idealizes
D-regions with concrete compressive struts representing the flow of
compressive stress and tensile ties consisting of reinforcing bars and the
tension members, and these struts and ties are joined at nodes to re-
present an idealized truss [12]. Lee et al. [14] applied strut-and-tie
model to coupling beams design of Burj Khalifa based on the STM
procedure in Appendix A of the American Building Code Requirements
for Structural Concrete 318-14 [15]. Khalifa [16] developed strut-and-
tie model to account for the contribution of fiber composite in shear
resistance of short coupling beams based on the experimental results.
Wight et al. [17] used strut-and-tie model to analysis and design deep
beams and proposed detailed design steps. Erwin et al. [18] revealed
that a simple strut-and-tie model that merely satisfies force equilibrium
can give similar accuracy if structural parameters are properly con-
sidered. Margherita et al. [19] proposed a strut-and-tie model to de-
termine the shear strength of exterior reinforced concrete beam-column
joints. Based on the strut-and-tie model, Perera et al. [20] developed a
simple automatic procedure for predicting the shear capacity of RC
beams shear strengthened with FRP, and the optimal configuration of
strut-and-tie mechanism was also resolved. A strut-and-tie model was
proposed by Wael [21] to estimate the shear strength of squat walls,
349
Engineering Structures 171 (2018) 348–361
which was more accurate and reliable than other seven sets of shear
strength predictive formulas that based on the procedures provided in
American code ACI 318-14 [15], the European code EC8 [22], Barda
et al. [23], Hwang et al. [24], Wood [25], Sánchez-Alejandre et al. [26]
and Gulec et al. [27]. The softened strut-and-tie model developed by
Shyh-Jiann Hwang and Hung-Jen Lee was used in estimating the shear
strength of beam column joints, deep beams, corbels and squat wall
[11]. This model was derived to satisfied equilibrium, compatibility
and constitutive laws of cracked reinforced concrete and emphasized
the importance of the compression softening phenomenon [28].
This paper aims at proposing an analytical method for the predic-
tion of shear capacity in steel fiber reinforced concrete (SFRC) coupling
beams with small span-depth ratio no more than 2.5 based on the
softened strut-and-tie model. In order to develop the model, the pre-
vious experimental results of SFRC coupling beams with small span-
depth ratio are reviewed to understand the shear mechanism of such
short coupling beams better. In addition, the proposed model considers
the effect of span-depth ratio, steel fiber with different type and volume
fraction, compressive strength of fiber concrete and horizontal and
vertical reinforcement on shear strength. Then the validity and accu-
racy of the model are assessed by comparing predictions with experi-
mental results.
2. Review of previous experimental results
A number of SFRC coupling beams with small span-depth ratio no
more than 2.5 have been tested under lateral reverse cyclic load [7,29]
according to Chinese test standard. These coupling beams were de-
signed in conventional reinforcement form with longitudinal and
transverse bars. A list of the test used is presented in Table 1. The list
summarizes the main parameters and test results of specimens, and also
includes the references they from. The geometry of the specimen as well
as the reinforcement details are shown in Figs. 1 and 2. The key test
variables were span-depth ratio (from 1.0 to 2.5), compressive strength
of fiber concrete (from 40.5 Mpa to 80.7 MPa) and volume fraction of
steel fibers (from 0.5% to 2.5%).
One feature of these specimens can be concluded from the test re-
sults summarized in Table 1 that most of SFRC coupling beams with
small span-depth ratio tend to behave shear dominant failure as diag-
onal tension mode and diagonal compression mode. The shear diagonal
compression failure was characterized by crushing of the concrete at
diagonal section. This failure mode occurred when the specimen was
provided with sufficient transverse reinforcement to suppress shear
failure along the diagonal crack, otherwise shear diagonal tension
failure will take place if the fiber concrete in diagonal section could
resist more compressive stress.
Regarding specimens CCB2-3, CCB2-4 and CCB2-5, it is notable that
high-strength concrete was used in these specimens. The experimental
results showed that the failure modes of the specimens were changed
from shear failure to flexure-shear failure when the volume fraction of
steel fiber increased from 0.5% to 1.0%. As the volume fraction of steel
fiber continued to increase to 1.5%, the failure mode of the specimen
was changed to flexure failure. The results indicated that the brittle
failure behavior of short coupling beams was significantly improved
with the increase of volume fraction of steel fiber, and steel fibers could
effectively restrain the shear cracking development and made a sig-
nificant contribution in shear resist mechanism of short coupling
beams. In addition, the use of high-strength concrete, on the one hand,
can avoid premature failure of concrete. On the other hand, some re-
search conducted by Naaman [30,31] and Banthia [32] showed that the
 J. Zhao et al.

Table 1
Main specimen parameters and test results.
Investigator Specimens Dimensions Fiber concrete Fiber Longitudinal steel Transverse steel Results

b (mm) h (mm) L (mm) λ (L / h ) f fc f ft fct Vsf (%) Type l f / df f yf (MPa) Tension reinforcement Skin reinforcement ρst (%) f yv (MPa) VExp (kN) Failure
modes
(MPa) ρ (%) f y′ (MPa) ρs (%) f y (MPa)

Cai et al. [7] CCB3-40-1-1F-S 150 400 400 1.0 43.1 3.26 1.71 1.0 Crimped 42 380 1.17 363.4 0.293 295.6 0.056 295.6 295.0 S-T
CCB3-40-1.5-1F-S 600 1.5 43.1 3.26 1.71 1.0 292.0 S-T
CCB3-40-2.5-1F-F/S 1000 2.5 43.1 3.26 1.71 1.0 190.0 F-S
CCB3-30-2-1F-S 800 2.0 40.5 2.62 1.43 1.0 227.0 S-C
CCB3-40-2-1F-S 800 2.0 43.1 3.26 1.71 1.0 238.0 S-T
CCB3-50-2-1F-S 800 2.0 52.9 3.62 1.89 1.0 243.0 S-T
CCB3-60-2-1F-S 800 2.0 66.7 4.27 2.04 1.0 250.0 S-T
CCB3-70-2-1F-S 800 2.0 70.1 4.54 2.14 1.0 253.0 S-T
CCB3-80-2-1F-S 800 2.0 80.7 5.44 2.22 1.0 255.0 S
CCB3-50-2-0.5F-S 800 2.0 54.5 3.82 1.89 0.5 238.0 S
CCB3-50-2-1F-S 800 2.0 54.8 4.64 1.89 1.0 244.0 S

350
CCB3-50-2-1.5F-S 800 2.0 55.9 5.51 1.89 1.5 249.5 S-T
CCB3-50-2-2F-S 800 2.0 55.3 5.96 1.89 2.0 255.5 S-C
CCB3-50-2-2.5F-F/S 800 2.0 54.1 6.66 1.89 2.5 257.0 F-S-C
Zhang et al. [29] CCB1-1 120 500 500 1.0 72.8 8.53 2.04 0.75 End hooked 75 – 0.74 441 0.28 350 0.056 350 341.2 S-C
CCB1-2 400 600 1.5 74.3 8.65 2.04 0.75 0.94 0.35 237.8 S-C
CCB1-3 400 700 1.75 75.3 8.56 2.04 0.75 0.94 0.35 218.9 S-T
CCB1-4 350 700 2.0 73.6 9.26 2.04 0.75 1.08 0.40 194.9 F-S
CCB1-5 280 700 2.5 78.3 9.30 2.04 0.75 1.38 0.51 147.0 F-S
CCB2-3 400 700 1.75 69.8 8.72 2.04 0.5 0.94 0.35 208.3 S-T
CCB2-4 400 700 1.75 66.3 10.0 2.04 1.0 0.94 0.35 229.0 F-S
CCB2-5 400 700 1.75 73.4 9.71 2.04 1.5 0.94 0.35 232 F

Where, b is width of cross section; h is height of specimen; L is length of specimen; λ is span-depth ratio; f fc is the cube compressive strength of fiber concrete; f ft is cube tensile strength of fiber concrete; fct is mean tensile
strength of the unreinforced cementitious matrix; Vsf is volume fraction of steel fibers; lf / df is the aspect ratio of fiber; f yf is yielding strength of steel fibers; ρ, ρs and ρst are ratio of tension reinforcement, skin
reinforcement and stirrups, respectively; f y′ , f y and f yv are yielding strength of tension reinforcement, longitudinal skin reinforcement and stirrups, respectively; VExp is the experimental value of shear capacity. Failure
modes: S-C: shear compression; S-T: shear tension; S: shear dominant; F: flexure dominant; F-S: flexure-shear failure; F-S-C: flexure-shear compression failure.
Engineering Structures 171 (2018) 348–361
J. Zhao et al.
Fig. 1. Details of SFRC specimens in literature [7].
increase in compressive strength of concrete matrix can enhance bond
strength between fibers and concrete matrix, thereby improved the
shear resistance of steel fibers. Therefore, the use of steel fiber re-
inforced concrete in short coupling beam can improve its brittle failure
behavior.
The four graphs shown in Fig. 3(a)–(d) are the typical failure me-
chanisms occurred in the coupling beam specimens with span-depth
ratio of 1.0, 1.5, 2.0 and 2.5 [7], respectively. It could be observed that
the main cracks were developed in the diagonal direction along the
shear span when the span-depth ratio of SFRC coupling beams was 1.0
or 1.5, and an X-type main crack was finally formed under the reversed
cyclic load. But when the span-depth ratio was 2.0 or 2.5, the main
cracks were developed in an inclined direction from mid-span to the
corner of beam-wall joints, and two similar X-type cracks were con-
formed in the specimens. The cracking pattern of the coupling beams
with small span-depth ratio changed with the increase of the span-
depth ratio.
3. Proposed strut-and-tie model
Schlaich et al. [33] recommend that the strut-and-tie model can be
selected based on the principal stress trajectories obtained from linear
elastic finite element analysis. To gain further insight into stress char-
acteristics of coupling beams with small span-depth ratio, three con-
crete coupling beams with different span-depth ratio are simulated by
finite element analysis. The coupling beams were modeled with quad-
rilateral plane-stress elements. The shear force and bending moments
were applied to coupling beam end due to the restraint of the concrete
blocks on each side of the beam. Fig. 4 shows the principal stress
Fig. 2. Details of SFRC specimens in literature [29].
351
Engineering Structures 171 (2018) 348–361
trajectories in the coupling beams with small span-depth ratio sum-
marized from the analysis results. The broken lines in Fig. 4 represent
principal compressive stress trajectories, and the solid lines represent
principal tensile stress trajectories. It can be observed that the principal
compressive stresses trajectories obtained by linear elastic finite ele-
ment analysis correlate closely with the cracking patterns presented in
experiments. When span-depth ratios of coupling beams were 1.0 and
1.5, it can be seen from Fig. 4 that the directional feature of the prin-
cipal stress trajectories were similar, and the inclined compressive
stress trajectories mainly between two compression end zones were
along the shear span or from the compression end zone to the mid-span
section. When the span-depth ratio was increased to 2.0, the principal
compressive stress trajectories mainly inclined from the compression
end zone to the mid-span section.
The key point to idealize such short coupling beams as a truss
structure is to represent the path followed by internal forces within the
structural element. Fig. 5 shows the strut-and-tie modeling of the SFRC
coupling beams with small span-depth ratio under the earthquake-in-
duced forces, which is composed of diagonal mechanism, vertical me-
chanism and horizontal mechanism. The broken lines represent com-
pressive struts, and the solid lines represent tension ties. The
intersection of struts and ties defines the nodes of the model named as
A, B and C, as shown in Fig. 5(a). The diagonal mechanism includes a
direct inclined concrete strut between two compressive end zones. The
vertical and horizontal mechanism account for the contribution of
transverse and horizontal reinforcement to the shear strength of cou-
pling beams with small span-depth ratio, respectively. In addition, the
contribution of steel fiber to shear capacity of SFRC coupling beam
cannot be neglected, and steel fibers can be considered as ties to carry
J. Zhao et al. Engineering Structures 171 (2018) 348–361

(a) L/h=1.0 (b) L/h=1.5 (c) L/h=2.0 (d) L/h=2.5

Fig. 3. Cracking patterns of specimens with different span-depth ratio [7].

tensile stress. 2tanθ−1

γh = , 0 ⩽ γh ⩽ 1
The three resisting mechanisms have a proportional relation and the 3 (4)
ratio changes with the span-depth ratio of coupling beams. Rd , Rh and
2cotθ−1
Rv are the shear ratios resisted by the diagonal, horizontal and vertical γv = , 0 ⩽ γv ⩽ 1
3 (5)
mechanisms, respectively, which can be defined as [34],
If the horizontal tie does not exist or the horizontal tie yields, the
(1−γh)(1−γv )
Rd = coefficient γh = 0 and Rv = γv for the reduced mechanisms, then the
1−γh γv (1) shear forces are resisted by the reduced mechanisms, such as the di-
agonal and vertical mechanism. The same situation occurs between γv
γh (1−γv ) and Rh . In Eqs. (4) and (5), the value of γv and γh are between 0 and 1,
Rh =
1−γh γv (2) thus the equation has the requirement for inclination angle θ of the
diagonal compression strut as 26.56∘ ⩽ θ ⩽ 63.43∘, and the limitation of
γv (1−γh) θ requires the span-depth ratio of coupling beams between 0.6 and 1.6.
Rv = The variation of the span-depth ratio of coupling beam directly af-
1−γh γv (3)
fects the value of inclination angle θ of the single diagonal compression
where, γh is the fraction of diagonal compression transferred by the strut, and influences the proportion of the shear force resisted by the
horizontal tie in the absence of the vertical tie; and γv is the fraction of three kinds of force transmission mechanism sequentially. The angle θ
diagonal compression transferred by the vertical tie in the absence of decreases with the increase of span-depth ratio of coupling beam, when
the horizontal tie. According to Schäfer [35], the values of γh and γv can the span-depth ratio is between 1.6 and 2.5, and the value of θ is less
be defined as, than 26.56°. Thus γh and γv are taken as 0 and 1, Rd , Rh and Rv are taken

M V M
V

V M V
M

(a) L/h=1.0 (b) L/h=1.5

V M

M V

(c) L/h=2.0
Fig. 4. Principal stress trajectories in coupling beams with small span-depth ratio.

352
J. Zhao et al. Engineering Structures 171 (2018) 348–361

A
B
M V C V M M V V M

(a) Strut-and-tie model (b) Diagonal mechanism

M V V M M V V M

(c) Vertical mechanism (d) Horizontal mechanism

Fig. 5. Strut-and-tie model of SFRC coupling beams with small span-depth ratio.

Nodal zone Steep strut

Diagonal strut Vertical tie
kh 0 kh 0 / 3 kh 0 kh 0 /3
A str
h0 V V M
h M h Fv

hw M hw
M V s V

L L

Fig. 6. Diagonal mechanism. Fig. 7. Vertical mechanism.

as 0, 0, and 1, respectively. For this kind of coupling beam with small where, L is the length of coupling beam; h w is the distance of the lever
span-depth ratio 1.6 < L/ h ⩽ 2.5, the shear force is only resisted by the arm from the resultant compressive force to the centroid of the long-
vertical mechanism. itudinal tension reinforcement. The lever arm h w of a doubly reinforced
rectangular section can be estimated by Eq. (7).
3.1. Diagonal mechanism
kh 0
h w = h 0−
3 (7)
Part of the shear forces are transmitted by the concrete compression
zone at the coupling beam end, then a diagonal compression strut is where, h 0 is the effective height of the coupling beam section; kh 0 is the
formed to resist shear forces, as shown in Fig. 6. depth of the compression zone at the end cross-section; and coefficient
The inclination angle θ of the diagonal compression strut is defined k can be defined by Eq. (8) [36].
by Eq. (6).
k= [nρ + (n−1) ρ′]2 + 2[nρ + (n−1) ρ′h 0′/ h 0] −[nρ + (n−1) ρ′] (8)
h
θ = arctan ⎛ w ⎞
⎝ L⎠ (6) where, n is the modulus ratio of elasticity, and n = Es / Ec ; ρ and ρ′ are

353
J. Zhao et al. Engineering Structures 171 (2018) 348–361

the ratio of the longitudinal tension reinforcement and compression width and length of the coupling beam, respectively; θc is an inclination
reinforcement, respectively; and h 0′ is the distance from the edge of the angle of the main crack of the coupling beam. According to the ob-
compression concrete to the centroid of the compression reinforcement. served experimental results, θc is approximately assumed to be equal to
The effective cross section area of the diagonal strut Astr is defined by inclination angle θ of the diagonal compression strut; if the diagonal
Eq. (9) as follows. concrete strut is absent (when the span-depth ratio 1.6 < L/ h ⩽ 2.5), θc
is approximately assumed to equal to inclination angle θs of the steep
kh 0 b
Astr = concrete strut in vertical mechanism.
cosθ (9)
The equivalent cross section area of the vertical tie of steel fibers
Asf , v can be rearranged by Eq. (15).
3.2. Vertical mechanism
bL
Asf , v = 0.41Vsf
cosθc (15)
The vertical mechanism includes one vertical tie and two steep
concrete struts, as shown in Fig. 7. The randomly distributed steel fibers As the crack opens, fibers will be pulled out from the matrix, and the
can be regarded as the micro-reinforcement in concrete which can resist tension behavior of the fiber will depend on the bond stress during the
the shear forces. The tension forces in the vertical tie Fv can be calcu- pullout process of steel fibers. When the vertical tie reaches its limited
lated by Eq. (10) as follows. state, the maximum fsf ,max can be calculated by Eq. (16) as follows.
Fv = Fs, v + Fsf , v (10) lf
fsf ,max = Dsf ⎛⎜ ⎞⎟ τsf ,max ⩽ f yf
where, Fs, v and Fsf , v are the tension force in the vertical tie of stirrups d
⎝ f⎠ (16)
and steel fibers, which can be calculated by Eqs. (11) and (12).
where, Dsf is fiber type impact coefficient, which is taken as 0.5, 0.75
Fs, v = As, v ·fs, v (11) and 1.0 for straight, crimped and end hooked steel fiber [39], respec-
tively. lf is the length of the fiber; df is the fiber diameter. f yf is yielding
where, As, v is the cross section area of the vertical tie of stirrups, it is strength of steel fiber; τsf ,max is the maximum bond shear stress between
roughly assumed that the transverse reinforcements are regarded as fiber and matrix, which is taken as Eq. (17) [40].
70% effective [28]; fs, v is the stress in the vertical tie of stirrups, when
the vertical tie reaches its limit state, fs, v = f yv ; f yv is yielding stress of τsf ,max = 3kb fct (17)
stirrups.
where, fct is mean tensile strength of the unreinforced cementitious
Fsf , v = Asf , v ·fsf (12) matrix. The correlation coefficient kb is taken as 0.4, 0.8 and 1.0 for
straight, end hooked and crimped steel fibers, respectively.
where, Asf , v is the equivalent cross section area of the vertical tie of Thus, the yielding force in the vertical tie Fyv can be calculated by
steel fibers; fsf is the average shear stress for all engaged fibers. Eq. (18).
According to the Ref. [37], the steel fibers distributed along main
cracks can be regarded as vertical or horizontal micro-reinforcement to Fyv = As, v ·f yv + Asf , v ·fsf ,max (18)
simplify the calculation.
The equivalent cross section area of the vertical tie of steel fibers
3.3. Horizontal mechanism
Asf , v can be estimated by Eq. (13).
Asf , v = nsf ·Asf (13) The horizontal mechanism consists of one horizontal tie and two flat
struts, as shown in Fig. 8. The way to estimate the tension force of the
where, Asf is the cross section area of single steel fiber; nsf is the
horizontal tie is similar to the vertical tie, as shown in Eq. (19).
equivalent number of steel fibers, which can be estimated by Eq. (14).
Fh = Fs, h + Fsf , h (19)
bL 1
nsf = η2 Vsf
cosθc Asf (14) where, Fh is the tension force in the horizontal tie; Fs, h and Fsf , h are the
tension force in the horizontal tie of longitudinal skin reinforcement
where, η2 is direction coefficient of steel fibers, which can be approxi- and steel fibers, which can be calculated by Eqs. (20) and (21), re-
mated as 0.41 [38]; Vsf is volume fraction of steel fibers; b and L are spectively.
Fs, h = As, h ·fs, h (20)
Flat strut
where, As, h is the cross section area of the horizontal tie of longitudinal
Horizontal tie skin reinforcement; fs, h is the stress in the horizontal tie of longitudinal
kh 0 kh 0 /3 skin reinforcement, when the horizontal tie reaches its limit state,
fs, h = f y , f y is yielding stress in longitudinal skin reinforcement.
V M Fsf , h = Asf , h ·fsf (21)
h Fh
where, Asf , h is the cross section area of the horizontal tie of steel fibers,
hw which can be calculated by Eq. (22).
M f V
bL
Asf , h = 0.41Vsf
cosθc (22)
L Similarly, the yielding force in the horizontal tie Fyh can be calcu-
lated by Eq. (23).
Fyh = As, h ·f y + Asf , h ·fsf ,max (23)
Fig. 8. Horizontal mechanism.

354
J. Zhao et al. Engineering Structures 171 (2018) 348–361

VVA

VVA B NHB 0.5F h tan A

N HA
NHB N HA Fv
A Fv
B Fh Fv
Fc sin
VVC
0.5F v cot 0.5Fv cot
C
Fc cos
VVC
v Fh
r d 0.5F h tan
C

h Fh

(a) Forces in SFRC coupling beams with small span-depth ratio

Node A
A str
Flat strut

-Fh /cos

Fc
Diagonal strut Steep strut

-Fv /sin
(b) Forces in node A
Fig. 9. Internal forces of strut-and-tie model for SFRC coupling beam with small span-depth ratio.

1 Rd
Fc = − × × VV
sinθ (Rd + Rh + Rv ) (27)
3.4. Shear capacity analysis
1 Rh
Fh = × × VV
3.4.1. Equilibrium conditions tanθ (Rd + Rh + Rv ) (28)
Fig. 9 shows the forces in equilibrium of the coupling beam with
small span-depth ratio calculated through the above strut-and-tie Rv
Fv = × VV
model. The resistances against the horizontal and vertical forces are (Rd + Rh + Rv ) (29)
estimated by Eqs. (24) and (25). When span-depth ratio of SFRC coupling beams is in the range of
VV = VVA + VVC = −Fc sinθ + Fhtanθ + Fv (24) 1.6 < L/ h ⩽ 2.5, the shear force is only transmitted through the vertical
mechanism, then the equilibrium equations are expressed as Eqs. (30)
NH = NHA + NHB = −Fc cosθ + Fh + Fv cotθ (25) and (31).
VV = Fv (30)
where, VV and NH are the vertical and horizontal resultant forces at the
coupling beam end, respectively; VVA and NHA are the vertical and NH = Fv cotθs (31)
horizontal forces in node A, respectively; NHB and VVC are the horizontal
forces in node B and vertical force in node C, respectively; Fc is the tanθs = 2tanθ (32)
compression force in the diagonal strut, Fv and Fh are tension force in where, VV and NH are the vertical and horizontal resultant forces at the
the vertical and horizontal tie, respectively. end section of coupling beam, respectively; θs is the inclination angle of
The three shear resisting mechanisms have a proportional relation. the steep strut.
The ratios assigned among the three mechanisms are assumed by Eq.
(26). 3.4.2. Control conditions
−Fc sinθ : Fhtanθ : Fv = Rd : Rh : Rv (26) The shear strength of the coupling beam can be calculated based on
the assumption that the concrete node zone has reached its maximum
Then, Eq. (26) can be rearranged as, compressive stress. The forces in node A at the end of the beam are

355
J. Zhao et al. Engineering Structures 171 (2018) 348–361

shown in Fig. 9(b). The maximum compressive stresses at node A can be where, f fc is the cube compressive strength of fiber concrete; coefficient
estimated by Eq. (33). η is taken as 0.79 when the concrete strength grade is less than C60. η is
taken as 0.833, 0.857 and 0.875 when the concrete strength grades are
Fh Fv
σd,max = ⎡−Fc + cos(θ−θf ) + cos(θs−θ)⎤/ Astr C60, C70 and C80, respectively.
⎢ cosθ sin θ ⎥
⎣ f s ⎦ (33) By recognizing Eq. (36), when the coupling beam is assumed to
reach ultimate shear capacity, the following Eqs. (40) and (41) can be
where, σd,max is the maximum compressive stress at node A; θf and θs are
obtained.
the inclination angles of the flat strut and steep strut, respectively;
which can be assumed as σd = ζfc′ (40)
2tanθf = tanθ , tanθs = 2tanθ (34) εd = ζε0 (41)
For span-depth ratio between 1.6 and 2.5, the compressive stress in The stress-strain relationships of reinforcement are represented by
the node A is only from steep strut, and the maximum compressive Eq. (42) as follows.
stress of node A is estimated by Eq. (35).
Es ·εs, εs < εy
fs = ⎧
F ⎨ f y , εs ⩾ εy
σd,max = ⎛ v ⎞/ Astr
⎜ ⎟ ⎩ (42)
⎝ sinθs ⎠ (35)
where, Es is the elastic modulus of the reinforcement; fs and εs are stress
and strain in the reinforcement, respectively; f y and εy are yield stress
3.4.3. Constitutive laws and yield strain in the reinforcement.
According to Ref. [41], the softened stress-strain curve of the Based on the simplified constitutive equations for the reinforce-
cracked concrete is represented by Eq. (36) as follows. ment, the stress-strain relationship of the steel fiber is estimated by Eq.
(43).
2

⎪ c ⎣ ζε0
⎪
( )( )
⎧ ζf ′ ⎡2 εd − εd ⎤, |εd | ⩽ |ζε0 |
ζε0
⎦
fsf = Esf εsf (43)
σd = εd 2
where, Esf is the elastic modulus of the steel fiber; fsf and εsf are stress
⎨ ⎡ −
⎛ ε0 ⎞ ζ ⎤
⎪ ζfc′ ⎢1−⎜ 2 − 1 ⎟ ⎥, |εd | > |ζε0 | and strain in steel fiber, respectively.
⎪ ⎢ ζ ⎠ ⎥
⎩ ⎣ ⎝ ⎦ (36) The relationship between forces and strains of the tension ties can
be simplified by Eq. (44).
where, σd and εd are the average principal compressive stress and strain
in concrete, respectively; εr is the average principal tensile strain; and ε0 Fh = Ath Es εh ⩽ Fyh, Fv = Atv Es εv ⩽ Fyv (44)
is the concrete cylinder strain corresponding to the cylinder strength,
where, Ath and Atv are the cross section areas of the horizontal tie and
which can be defined by Eq. (37); ζ is the softening coefficient, which
vertical tie, respectively; εh and ε v are the average strains in the hor-
can be calculated by Eq. (38).
izontal tie and vertical tie, respectively.
f ′−20 ⎞
ε0 = 0.002 + 0.001 ⎜⎛ c ⎟ , 20 MPa ⩽ f ′ ⩽ 100 MPa
c 3.4.4. Compatibility condition
⎝ 80 ⎠ (37)
According to the Ref. [36], two dimensional membrane elements
5.8 1 0.9 should satisfy Mohr circular compatibility condition as shown in
ζ= · ⩽ Fig. 10, and the relationship between εh , ε v and εr , εd can be expressed
fc′ 1 + 400εr 1 + 400εr (38)
by Eq. (45).
where, fc′ is the cylinder compressive strength of fiber concrete, which εh = εr cos2 θ + εd sin2 θ , ε v = εr sin2 θ + εd cos2 θ (45)
can be calculate by Eq. (39).
where, θ is the inclination angle of the principal compressive strain. The
fc′ = η ·f fc (39) basic compatibility condition Eq. (46) is the first strain invariant.

v 2 h

r
d 2
v

h
h
v
d r

(a) Strain state in membrane element (b) Mohr strain cycle

Fig. 10. Compatibility conditions for diagonally cracked concrete.

356
J. Zhao et al. Engineering Structures 171 (2018) 348–361

εr + εd = εh + ε v (46)

3.4.5. Solution procedures

The coupling beam is considered to fail when any one of the strut,
tie or node in the strut-and-tie model reaches its ultimate state. The
solution procedures are divided into four major steps and the flow chart
is shown in Fig. 11. The flow chart starts with a hypothetical shear force
which is assumed to make the concrete node zone at the beam end
reach its maximum compressive stress, and details can be concluded as
follows.

(1) The value of θ , γh, γv , Rd , Rv and Rh can be calculated by Eqs. (1)–(6).

By selecting a value for shear capacity V , the forces Fc , Fh and Fv can
be estimated by Eqs. (27)–(29), and Fyv , Fyh can be calculated by
Eqs. (18) and (23). The result should satisfy the condition Fh ⩽ Fyh
and Fv ⩽ Fyv , otherwise the shear value should be reassumed.
(2) The value of stress σd,max can be calculated by Eq. (33) or Eq. (35),
and the concrete softening coefficient ζ can be calculated by Eq.
(40). Then, the value of strain εd , εh and ε v can be calculated by Eqs.
(41), (44).
(3) The value of strain εr is estimated by Eq. (46), and another concrete
softening coefficient ζ ′ can be calculated by Eq. (38).
(4) If ζ is close to ζ ′, then the assumed shear force can be regarded as
the shear capacity of the coupling beam. Otherwise, if not, then
reassumed shear value and recalculate from step (1).

It is important to note that if Fh calculated from the assumed V is

equal to Fyh or Fv is equal to Fyv , and ζ is calculated less than ζ ′, then the
assumed shear force V which makes any one of the tie reach its yield
strength is the shear capacity of the SFRC coupling beams. It means that
the strut-and-tie mechanism has already failed before the node area
reached its ultimate state due to the yield of the tie.

4. Comparison of predictions and experimental results

The proposed model was used to calculate the shear strength of 22

SFRC coupling beam specimens with small span-depth ratio. The
comparison between the test failure loads and the predictions of the
softened strut-and-tie model are shown in Table 2 and Fig. 12. Table 2
also summarizes the statistical results and main calculating data for
these specimens. For the proposed model, the predictive results were on
average 8.8% larger than experimental results in the case of span-depth
ratios 1.6 < L/ h ⩽ 2.5, while the predictive results were on average
9.2% less than experimental results when the span-depth ratios are in
the range of L/ h ⩽ 1.6 and tend to be more conservative. The shear
mechanism of the SFRC coupling beam changes with different span-
depth ratio, thus affecting their shear capacity. The ratio of shear force
in the diagonal mechanism and horizontal mechanism decreases with
the rise of span-depth ratio, while the ratio of shear force in vertical tie
increases.
The average ratio of the VSTMc / VExp was 1.054 with a coefficient of
variation (COV) of 0.117. The results show that it is reasonable to use
the softened strut-and-tie model for the calculation of shear capacity of
the SFRC coupling beams with small span-depth ratio.

5. Conclusions
Fig. 11. Flow chart for the strut-and-tie model.

This research focused on developing a strut-and-tie model to predict

the shear capacity in steel fiber reinforced concrete coupling beams

357
 Table 2
Experimental verification.
Investigator Specimens b mm h mm L mm λ (L / h ) k Fiber type
J. Zhao et al.

h 0 mm θ Degree θs Degree Astr mm2 fc′ MPa f yv MPa f y MPa l f / df Vsf % fsf MPa Fyv kN Fyh kN

Cai et al. [7] CCB3-40-1-1F-S 150 400 400 357 1.0 0.288 38.90 – 19786.6 34.1 295.6 295.6 Crimped 42 0.01 161.6 140.3 97.5
CCB3-40-1.5-1F-S 150 400 600 357 1.5 0.288 28.28 – 17485.2 34.1 0.01 161.6 186.7 114.1
CCB3-40-2.5-1F-F/S 150 400 1000 357 2.5 0.288 – 38.85 18328.3 34.1 0.01 161.6 296.7 –
CCB3-30-2-1F-S 150 400 800 357 2.0 0.295 – 38.82 20282.3 32.0 0.01 135.1 234.0 –
CCB3-40-2-1F-S 150 400 800 357 2.0 0.288 – 38.90 19786.6 34.1 0.01 161.6 250.9 –
CCB3-50-2-1F-S 150 400 800 357 2.0 0.282 – 38.96 19418.3 41.8 0.01 178.6 261.7 –
CCB3-60-2-1F-S 150 400 800 357 2.0 0.278 – 39.00 19156.8 55.6 0.01 192.8 270.7 –
CCB3-70-2-1F-S 150 400 800 357 2.0 0.276 – 39.03 18989.0 60.1 0.01 202.2 276.8 –
CCB3-80-2-1F-S 150 400 800 357 2.0 0.273 – 39.05 18826.0 70.6 0.01 209.8 281.6 –
CCB3-50-2-0.5F-S 150 400 800 357 2.0 0.282 – 38.96 19418.3 43.1 0.005 178.6 205.2 –
CCB3-50-2-1F-S 150 400 800 357 2.0 0.282 – 38.96 19418.3 43.3 0.01 178.6 261.7 –
CCB3-50-2-1.5F-S 150 400 800 357 2.0 0.282 – 38.96 19418.3 44.2 0.015 178.6 318.2 –
CCB3-50-2-2F-S 150 400 800 357 2.0 0.282 – 38.96 19418.3 43.7 0.02 178.6 374.7 –
CCB3-50-2-2.5F-F/S 150 400 800 357 2.0 0.282 – 38.96 19418.3 42.8 0.025 178.6 431.2 –
Zhang et al. [29] CCB1-1 120 500 500 465 1.0 0.223 40.65 – 16905.6 60.6 350 350 End hooked 75 0.75 367.2 194.9 144.3
CCB1-2 120 400 600 365 1.5 0.253 29.12 – 12704.4 61.9 0.75 367.2 198.7 148.0
CCB1-3 120 400 700 365 1.75 0.253 – 43.67 15345.6 60.6 0.75 367.2 236.8 –
CCB1-4 120 350 700 315 2.0 0.269 – 39.33 13134.1 61.9 0.75 367.2 228.3 –
CCB1-5 120 280 700 245 2.5 0.297 – 32.24 10321.4 62.7 0.75 367.2 217.8 –
CCB2-3 120 400 700 365 1.75 0.253 – 43.67 15345.6 58.1 0.5 367.2 193.1 –
CCB2-4 120 400 700 365 1.75 0.253 – 43.67 15345.6 55.2 1.0 367.2 280.5 –
CCB2-5 120 400 700 365 1.75 0.253 – 43.67 15345.6 61.1 1.5 367.2 367.9 –
Total 22

358
Investigator Specimens Rd Rh Rv Choose V kN Fc kN Fh kN Fv kN σd,max MPa ε0 10−3 ζ εd 10−3 εv 10−3 εh 10−3 εr 10−3 ζ′ |ζ ′−ζ | < 0.001 VSTMc kN VSTMc
Fh ≤ Fyh Fv ≤ Fyv VExp

Cai et al. [7] CCB3-40-1-1F-S 0.449 0.115 0.436 278.6 199.2 39.9 121.5 18.97 2.18 0.5569 1.21 1.92 0.91 4.04 0.5566 0.0003 278.8 0.95
CCB3-40-1.5-1F-S 0.094 0.002 0.904 206.6 40.9 0.94 186.7 16.19 2.18 0.4756 1.04 2.21 0.02 3.27 0.5925 No 206.6 0.71
Fv = Fyv
CCB3-40-2.5-1F-F/S – – 1.0 207.5 – – 207.5 20.87 2.18 0.6131 1.33 1.55 – 2.88 0.6133 0.0002 207.5 1.09
CCB3-30-2-1F-S – – 1.0 234.0 – – 234.0 18.41 2.15 0.5752 1.24 2.08 – 3.32 0.5898 No 234.0 1.03
Fv = Fyv
CCB3-40-2-1F-S – – 1.0 246.7 – – 246.7 19.85 2.18 0.5831 1.27 2.18 – 3.45 0.5835 0.0004 246.7 1.04
CCB3-50-2-1F-S – – 1.0 261.7 – – 261.7 21.43 2.27 0.5129 1.17 2.30 – 3.47 0.5807 No 261.7 1.08
Fv = Fyv
CCB3-60-2-1F-S – – 1.0 270.7 – – 270.7 22.46 2.44 0.4042 0.99 2.37 – 3.36 0.5083 No 270.7 1.09
Fv = Fyv
CCB3-70-2-1F-S – – 1.0 276.8 – – 276.8 23.15 2.50 0.3853 0.96 2.42 – 3.38 0.4878 No 276.8 1.09
Fv = Fyv
CCB3-80-2-1F-S – – 1.0 281.6 – – 281.6 23.74 2.63 0.3362 0.89 2.46 – 3.34 0.4515 No 281.6 1.10
Fv = Fyv
CCB3-50-2-0.5F-S – – 1.0 205.2 – – 205.2 16.81 2.29 0.3903 0.89 2.30 – 3.19 0.5857 No 205.2 0.85
Fv = Fyv
CCB3-50-2-1F-S – – 1.0 261.7 – – 261.7 21.43 2.29 0.4951 1.13 2.30 – 3.43 0.5721 No 261.7 1.05
Fv = Fyv
CCB3-50-2-1.5F-S – – 1.0 276.1 – – 276.1 22.61 2.30 0.5120 1.18 2.00 – 3.18 0.5121 0.0001 276.1 1.07
CCB3-50-2-2F-S – – 1.0 309.0 – – 309.0 25.31 2.30 0.5793 1.33 1.90 – 3.23 0.5798 0.0005 309.0 1.18
CCB3-50-2-2.5F-F/S – – 1.0 311.3 – – 311.3 25.50 2.28 0.5967 1.36 1.66 – 3.02 0.5968 0.0001 311.3 1.17
(continued on next page)
Engineering Structures 171 (2018) 348–361
J. Zhao et al. Engineering Structures 171 (2018) 348–361

0.117
VSTMc
VExp

0.99
0.98

1.09

1.16
1.10
0.93

1.19
1.28
1.05
VSTMc kN

338.2
231.3

236.8

225.8
160.0
193.1

272.0
298.0
|ζ ′−ζ | < 0.001

Average
0.0001

0.0001
0.0001

0.0006
0.0003

COV
No

No

No
0.4356
0.4381

0.4475

0.4384
0.4633
0.4920

0.4654
0.4602
ζ′
εr 10−3

4.81
4.58

4.43

4.57
3.75
4.28

4.53
3.99

Fig. 12. Comparison between experimental results and proposed model.

εh 10−3

1.43
0.05

with small span-depth ratio. Through the analysis of experimental re-

–

–
–
–

–
–

sults, the cracking patterns and failure modes were discussed in order to
εv 10−3

understand the shear mechanism of steel fiber reinforced concrete

2.29
3.50

3.50

3.47
2.57
3.50

3.40
2.84

coupling beams with small span-depth ratio comprehensively. The

conclusions can be made as follows.
εd 10−3

1.09
1.03

0.92

1.11
1.17
0.78

1.13
1.16

(1) An analytical model based on the softened strut-and-tie model was

proposed for the prediction of shear capacity of steel fiber re-
0.4355
0.4055

0.3684

0.4383
0.4632
0.3133

0.4648
0.4599

inforced concrete coupling beams with small span-depth ratio. The

ζ

softened strut-and-tie model used was composed of diagonal me-

chanism, vertical mechanism and horizontal mechanism, and the
ε0 10−3

2.51
2.52

2.51

2.52
2.53
2.48

2.44
2.51

three resisting mechanisms had a proportional relation different

with the change of the span-depth ratio. It also considers the factors
of volume fraction of steel fibers, steel fiber type, compressive
σd,max MPa

strength of concrete.
26.41
25.13

22.34

27.12
29.05
18.22

25.67
28.12

(2) The proposed method based on the softened strut-and-tie model can
be used to calculate the shear capacity of the steel fiber reinforced
Fv ≤ Fyv

Fv = Fyv

Fv = Fyv

Fv = Fyv

concrete coupling beams with small span-depth ratio, and the cal-
Fv kN

127.5
198.7

236.8

225.8
160.0
193.1

272.0
298.0

culated and test results showed good agreement.

(3) The proposed model can be used in shear prediction of SFRC cou-
Fh ≤ Fyh

pling beams with span-depth ratio from 0.6 to 2.5, and the cylinder
Fh kN

58.7

compressive strength of fiber concrete is required between 20 MPa

2.2

–
–
–

–
–

and 100 MPa. The model considers the effect of three kinds of steel
246.1
Fc kN

fibers, including straight, crimped, and end hooked, and the

64.5

available data of volume fraction of steel fibers in this model are

–

–
–
–

–
–

from 0.5% to 2.5%.

Choose V kN

Acknowledgement
338.2
231.3

238.8

225.8
160.0
193.1

272.0
298.0

Financial supports for this research are provided by the National

0.377
0.859

Key R&D Program of China (2016YFE0125600), National Natural

1.0

1.0
1.0
1.0

1.0
1.0
Rv

Science Foundation of China (51078333), Program for Changjiang

0.149
0.005

Scholars and Innovative Research Team in University of Minister of

Rh

Education of China (IRT_16R67), Innovative Research Team (in Science

–

–
–
–

–
–

and Technology) in University of Henan Province (15IRTSTHN026),

0.474
0.136

and Special Research Foundation for Young Teachers of Zhengzhou

Rd

–
–
–

–
–

University (F0000270). These supports are gratefully acknowledged. In

addition, the authors specially thank to Mr. Xin Ma, Bo Jia and Bin Li
for their helps in the tests.
Specimens

CCB1-1
CCB1-2

CCB1-3

CCB1-4
CCB1-5
CCB2-3

CCB2-4
CCB2-5
22
Table 2 (continued)

Zhang et al. [29]

Investigator

Total

359
J. Zhao et al. Engineering Structures 171 (2018) 348–361

Appendix

Ath cross section area of the horizontal tie

Atv cross section area of the vertical tie
Asf cross section area of single steel fiber
Astr effective cross section area of the diagonal strut
Asf , h equivalent cross section area of the horizontal tie of steel fibers
Asf , v equivalent cross section area of the vertical tie of steel fibers
As, h cross section area of the horizontal tie of longitudinal skin reinforcement
As, v cross section area of the vertical tie of stirrups
b width of the coupling beam
df fiber diameter
Dsf fiber type impact coefficient
Es elastic modulus of the reinforcement
Esf elastic modulus of steel fibers
fc′ cylinder compressive strength of fiber concrete
fct mean tensile strength of the unreinforced cementitious matrix
f fc cube compressive strength of fiber concrete
f ft cube tensile strength of fiber concrete
fs stress in the reinforcement
fsf average shear stress in steel fiber
fs, h stress in the horizontal tie of longitudinal skin reinforcement
fs, v stress in the vertical tie of stirrups
fy yielding stress of the longitudinal skin reinforcement
f y′ yielding strength of tension reinforcement
f yf yielding strength of steel fibers
f yv yielding stress in stirrups
Fc compression force in the diagonal strut
Fh tension force in horizontal tie
Fs, h tension force in the horizontal tie of longitudinal skin reinforcement
Fsf , h tension force in the horizontal tie of steel fibers
Fsf , v tension force in the vertical tie of steel fibers
Fs, v tension force in the vertical tie of stirrups
Fv tension force in the vertical tie
Fyh yielding force in the horizontal tie
Fyv yielding force in the vertical tie
h0 effective height of the coupling beam section
h0′ distance from the edge of the compression concrete to the centroid of the compression reinforcement
hw distance of the lever arm from the resultant compressive force to the centroid of the longitudinal tension reinforcement
kb correlation coefficient
kh 0 depth of the compression zone at the end cross-section
L length of coupling beam
lf fiber length
n modulus ratio of elasticity
nsf equivalent number of steel fibers
NHA horizontal force in node A
NHB horizontal force in node B
Rd shear ratio resisted by diagonal mechanism
Rh shear ratio resisted by horizontal mechanism
Rv shear ratio resisted by vertical mechanism
VExp experimental shear strength
VH horizontal resultant force at the coupling beam end
Vsf volume fraction of steel fibers
VSTMc calculated shear strength
VV vertical resultant force at the coupling beam end
VVA vertical force in node A
VVC vertical force in node C
ρ ratio of the longitudinal tension reinforcement
ρ′ ratio of the longitudinal compression reinforcement
ρs ratio of skin reinforcement

360
J. Zhao et al. Engineering Structures 171 (2018) 348–361

ρst ratio of stirrups

γh fraction of diagonal compression transferred by the horizontal tie in the absence of the vertical tie
γv fraction of diagonal compression transferred by the vertical tie in the absence of the horizontal tie
θ inclination angle of the diagonal compression strut
θc inclination angle of the main crack of the coupling beam
θf inclination angle of the flat strut
θs inclination angle of the steep strut
ε0 concrete cylinder strain
εd average principal compressive strain of concrete
εh average strain in the horizontal tie
εr average principal tensile strain
εs strain in the reinforcement
εsf strain in steel fiber
εv average strain in vertical tie
εy yield strain in the reinforcement
σd average principal compressive stress in concrete
σd,max maximum compressive stress at node A
ζ softening coefficient
η2 direction coefficient of steel fibers
τsf ,max maximum bond shear stress between fibers and matrix

References
[1] Paulay T, Binney JR. Diagonally reinforced coupling beams of shear walls. Shear
Reinforced Concr 1974;2(26):579–98. [ACI Special Publication, SP-42].
[2] Paulay T, Santhakumar AR. Ductile behavior of coupled shear walls. J Struct Div
ASCE 1976;102:93–108.
[3] Zhao J, Cai GC, Larbi AS, Zhang Y. Hysteretic behaviour of steel fibre RC coupled
shear walls under cyclic loads: Experimental study and modelling. Eng Struct
2018;156:92–104.
[4] Canbolat BA, Parra-Montesinos GJ, Wight JK. Experimental study on seismic be-
havior of high-performance fibre-reinforced cement composite coupling beams. ACI
Struct J 2005;102(1):159–66.
[5] Chaallal O, Thibodeau S, Lescelleur J, Malenfant P. Steel fibre or conventional re-
inforcement for concrete shear walls. Concr Int 1996;18(6):39–42.
[6] Zhang HZ. Experimental study on shear resistance properties of steel fiber re-
inforced high-strength concrete members. Doctoral Thesis. China: Dalian University
of technology; 2005 [in Chinese].
[7] Cai GC, Zhao J, Degée H, Vandoren B. Shear capacity of steel fibre reinforced
concrete coupling beams using conventional reinforcements. Eng Struct
2016;128:428–40.
[8] Dinh HH. Shear behavior of steel fiber reinforced concrete beams without stirrup
reinforcement. ACI Struct J 2010;107(5):597–606.
[9] Susetyo J, Gauvreau P, Vecchio FJ. Effectiveness of steel fiber as minimum shear
reinforcement. ACI Struct J 2011;108(4):488–96.
[10] Minelli F, Plizzari GA. On the effectiveness of steel fibers as shear reinforcement.
ACI Struct J 2013;110(3):379–90.
[11] Hwang SJ, Lee HJ. Strength prediction for discontinuity regions by softened strut-
and-tie model. J Struct Eng 2002;128(12):1519–26.
[12] Shuraim AB, EI-sayed AK. Experimental verification of strut and tie model for HSC
deep beams without shear reinforcement. Eng Struct 2016;117:71–85.
[13] Oh JK, Shin SW. Shear strength of reinforced high-strength concrete deep beams.
ACI Struct J 2001;98(2):164–73.
[14] Lee OJ, Kuchma DA, Baker W, Novak LC. Design and analysis of heavily loaded
reinforced concrete link beams for Burj Dubai. ACI Struct J 2008;105(4):451–9.
[15] ACI 318-14 building code requirements for structural concrete. An ACI standard,
reported by ACI Committee 318, American Concrete Institute, Detroit; 2014.
[16] Khalifa ES. Analytical model for steel fiber concrete composite short-coupling beam.
Compos B 2014;56(1):318–29.
[17] Wight JK, Parra-Montesinos GM. Strut and tie model for deep beam design. Concr
Int 2003;25:63–70.
[18] Lim E, Hwang SJ. Modeling of the strut-and-tie parameters of deep beams for shear
strength prediction. Eng Struct 2016;108:104–12.
[19] Pauletta M, Luca DD, Russo G. Exterior beam column joints–Shear strength model
and design formula. Eng Struct 2015;94:70–81.
[20] Perera R, Vique J, Arteaga A, Diego AD. Shear capacity of reinforced concrete
members strengthened in shear with FRP by using strut-and-tie models and genetic
algorithms. Compos B 2009;40(8):714–26.
[21] Kassem W. Shear strength of squat walls: A strut-and-tie model and closed-form
design formula. Eng Struct 2015;84:430–8.
[22] European committee for standardization (CEN). European committee for standar-
dization (CEN) (2004)“Design of structures for earthquake resistance, Part 1: gen-
eral rules, seismic actions and rules for buildings Eurocode 8. BSEN 1998–1. BSI
British Standards, London; 2004.
[23] Barda F, Hanson JM, Corley WG. Shear strength of low-rise walls with boundary
elements. Reinf Concr Struct Seismic Zones 1977:149–202.
[24] Hwang SJ, Lee HJ. Strength prediction for discontinuity regions by softened strut-
and-tie model. ASCE J Struct Eng 2002;128(12):1519–26.
[25] Wood SL. Shear strength of low-rise reinforced concrete walls. ACI Struct J
1990;87(1).
[26] Sánchez-Alejandre A, Alcocer SM. Shear strength of squat reinforced concrete walls
subjected to earthquake loading - trends and models. Eng Struct
2010;32(8):2466–76.
[27] Gulec CK, Whittaker AS. Empirical equations for peak shear strength of low aspect
ratio reinforced concrete walls. ACI Struct J 2011;108(1):80–9.
[28] Hwang SJ, Lu WY, Lee HJ. Shear strength prediction for deep beams. ACI Struct J
2000;97(3):367–76.
[29] Zhang HZ, Zhang RJ, Huang CK. Ductility and energy dissipation capacity of steel
fiber reinforced high-strength concrete coupling beams of small span/depth. J
Dalian Univ Technol 2009;49(6):884–90. [in Chinese].
[30] Naaman AE, Najm H. Bond-slip mechanisms of steel fibers in concrete. ACI Mater J
1991;88(2):135–45.
[31] Naaman AE, Namur GG, Alwan JM, Najm HS. Fiber pullout and bond slip. I:
Analytical study. J Struct Eng 1991;117(9):2769–90.
[32] Banthia N, Trottier JF. Concrete reinforced with deformed steel fibres, part I: Bond-
slip mechanisms. ACI Mater J 1994;91(5):435–46.
[33] Schlaich J, Schäfer K, Jennewein M. Toward a consistent design of structural con-
crete. PCI J 1987;32(3):74–150.
[34] Hwang SJ, Lee HJ. Analytical model for predicting shear of exterior reinforced
concrete beam-column joints for seismic resistance. ACI Struct J
1998;96(5):846–57.
[35] Schäfer K. Strut-and-Tie models for the design of structural concrete. National
Cheng Kung University; 1996, Mar:140.
[36] Thomas TCH, Mo YL. Unified theory of concrete structures. Chichester: Wiley Press;
2010.
[37] Gao DY, Shi K, Zhao SB. Calculation method for shear capacity of steel fiber re-
inforced concrete beam-column joints based on softened strut-and-tie model. China
Civ Eng J 2014;47(9):101–9. [in Chinese].
[38] Romualdi JP, Mandel JA. Tensile strength of concrete affected by uniformly dis-
tributed and closely spaced short lengths of wire reinforcement. ACI Mater J
1964;61(6):657–72.
[39] Khuntia M, Stojadinovic B, Goel SC. Shear strength of normal and high-strength
fiber reinforced concrete beams without stirrups. ACI Struc J 1999;96(2):282–90.
[40] Ng TS, Htut TNS, Foster SJ. Fracture of steel fibre-reinforced concrete-the unified
variable engagement model. UNICIV report No.R-460. The University of New South
Wales, Australia, May 2012.
[41] Tan KH, Murugappan K, Paramasivam P. Shear behavior of steel fiber reinforced
concrete beams. ACI Struct J 1993;90(1):3–11.

361