Contribution of Shear Reinforcements and Concrete To The Shear Capacity of Interfaces Between Concretes Cast at Different Times
Contribution of Shear Reinforcements and Concrete To The Shear Capacity of Interfaces Between Concretes Cast at Different Times
Contribution of Shear Reinforcements and Concrete To The Shear Capacity of Interfaces Between Concretes Cast at Different Times
a
Hubei Key Laboratory of Disaster Prevention and Mitigation, China Three Gorges University, Yichang 443002, China
b
Dept. of Civil Engineering & Architecture, China Three Gorges University, Yichang 443002, China
Received 6 May 2020 Currently, shear transfer models have diverse assumptions about the contribution of shear
Revised 4 November 2020 reinforcements and concrete to the shear capacity of interfaces between concretes cast at
Accepted 17 January 2021 different times (that is, cold joints). Furthermore, the knowledge about the effect of high–
Published Online 26 March 2021 strength shear reinforcements (HSSR) having a nominal yield strength exceeding 414 MPa (60
ksi), is still limited. This study aims to determine the extent to which shear reinforcements and
KEYWORDS concrete contribute to the shear capacity of cold joints and explore the effect of HSSR on
shear transfer behavior. To this end, nine cold joint push–off specimens were tested to
Cold joint investigate the effects of yield strength of shear reinforcement (YSSR) and concrete compressive
Shear capacity strength (CCS) on the interface shear strength and stiffness. In addition, code provisions
Cohesion evaluation was used to identify the degree of concrete contribution, based on a database of
High-strength shear reinforcement 108 cold joint push–off test results established in this study. Test results indicate that the
Push-off test
increase of YSSR (rising from 400 MPa to 600 MPa) has no evident effect on the shear strength
and stiffness of cold joints, whereas the increase of CCS can significantly increase them.
Moreover, when CCS rises from 20 MPa to 50 MPa, the interface shear capacity of push–off
specimens increases by approximately 110%, i.e., from 103.5 kN to 217.5 kN. Hence,
concrete has a considerable contribution, whereas the contribution of shear reinforcements is
not as great as expected. Code provisions evaluation results also indicate that concrete
considerably contributes to the shear capacity in the form of cohesion. On the basis of these
findings and current design equations, a shear capacity equation, which accounts for the effect
of CCS and the limited participation of shear reinforcements, is proposed. This equation is
found to provide reasonable prediction of cold joint shear capacity, especially for the cold
joint composed of high–strength concrete.
1. Introduction the original concept, for the reinforced cold joint, the shear
capacity originates from two aspects: the shear reinforcement
Transfer of shear forces across the interface between concretes and concrete. Birkeland and Birkeland (1966) first proposed the
cast at different times (that is, cold joint) is vital to the strength of “shear friction theory”, based on which the simple shear–friction
many concrete structures. In general, concrete structural applications equation was proposed as follows:
for which shear transfer design is warranted include the interface
V = A f tan ϕ , (1)
between precast concrete girders and cast–in–place concrete n vf y
bridge decks, the interface between the existing girder and the where Avf is area of shear reinforcement; fy is yield strength of
elements newly added in structural rehabilitation, and the interface shear reinforcement; tanφ = 1.7, 1.4, and 0.8 − 1.0 for monolithic
between the precast segments and cast–in–place splice regions of concrete, artificially roughened joints, and ordinary joints,
post–tensioned spliced girder bridges (Barbosa et al., 2017; respectively.
Williams et al., 2017). Nearly all design equations of codes and calculation formulas
This study addresses the shear capacity of cold joints. Back to proposed by researchers contain the parameter Avf fy, thereby
CORRESPONDENCE Qing Wang postwq@163.com Dept. of Civil Engineering & Architecture, China Three Gorges University, Yichang 443002, China
ⓒ 2021 Korean Society of Civil Engineers
2066 J. Liu et al.
implying that the contribution of shear reinforcement to the shear reinforcement contribution. In addition, the push–off tests
interface shear capacity is considered as dominant. Undoubtedly, conducted by Barbosa et al. (2017) and Harries et al. (2012) only
shear reinforcements play a great role in a certain stage of the focused on roughened cold joints, and this study intended to
entire shear resistant process, especially in terms of enhancing carry out push–off tests on smooth cold joints including HSSR,
the shear ductility. However, when the interface reaches the which can serve as useful supplements to the existing tests.
ultimate shear capacity, whether shear reinforcements play a Another aspect that needs consideration is the contribution of
decisive role remains to be discussed. To protect the strain gauges of concrete. Birkeland and Birkeland (1966) first proposed the
shear reinforcement from damage during concrete casting and simple shear–friction equation which did not consider the
from interface slip and cracking, the strain gauges are usually contribution of concrete, based on a limited range of concrete
located at a short distance from the interface itself (Harries et al., strengths (Walraven et al., 1987). In the following decade, nearly
2012; Semendary et al., 2020). Therefore, the actual strain of all the concrete strength of test specimens laid between 3,000
shear reinforcement at the interface is difficult to test. Most and 6,000 psi (21 and 41 N/mm2) (Mattock, 2001). The effect of
literature, as Birkeland and Birkeland (1966) did at first, only concrete strength is relatively small and is explained as the
assumes that the shear reinforcement has yielded when the natural scatter of experimental results; moreover, the contribution of
interface reaches the ultimate shear capacity. Kono and Tanaka concrete was ignored. Due to the simple form and relative high
(2000) and Kono et al. (2001) reported that shear reinforcements accuracy, the shear–friction equation, which maintained its
do not generally yield when the maximum shear is reached and firstly proposed simple format, was codified in the 1977 ACI
the assumption often used in design codes that shear reinforcements Building Code, ACI 318-77 (ACI 318, 1977). With the increase
have yielded at the peak stress should be reconsidered. Harries et of concrete strength in push–off tests, researchers have found
al. (2012) demonstrated that the stress in the interface steel that the increase in concrete strength can increase the shear
reinforcement is remarkably lower than its yield strength when capacity of interface to some extent, and a few researchers proposed
the peak shear load is achieved. Barbosa et al. (2017) showed the calculation formula including a parameter of concrete strength
that for specimens containing #16M (with a nominal diameter of (Walraven et al., 1987). However, as mentioned above, researchers
15.9 mm) reinforcing bars across the interface, the bars reach the still believed that the contribution of shear reinforcement is
nominal yield strain before the peak interface shear load is dominant. Therefore, on one hand, nearly all calculation formulas,
observed. However, the interface reinforcement for the #13M whether including the concrete strength parameter or not, still
(with a nominal diameter of 12.7 mm) specimens does not reach take Avf fy as the main parameter (Walraven et al., 1987; Loov and
the nominal yielding strains. Patnaik, 1994; Mattock, 1994; Mattock, 2001; Kahn and
Given the complex observations of shear reinforcement strain, Mitchell, 2002); on the other hand, instead of rethinking the
another feasible method can be proposed to determine the contribution of concrete, most researchers focused on examining
contribution of shear reinforcement. If the change of shear and modifying the existing calculation models, especially those in
reinforcement strength greatly influences the interface shear codes, to make them suitable for high–strength concrete (Mattock,
capacity, then the shear reinforcement has a great contribution to 2001; Kahn and Mitchell, 2002).
the interface shear capacity, and vice versa. However, the With regard to the degree of concrete contribution, similar to
literature investigation shows that most of the existing cold joint the principle about the shear reinforcement, if the change of
push–off tests focused on normal–strength shear reinforcements, concrete strength has a great influence on the interface shear
and only the push–off tests conducted by Barbosa et al. (2017) capacity, then concrete has a great contribution to the interface
and Harries et al. (2012) addressed the high–strength shear shear capacity, and vice versa. Furthermore, the calculation
reinforcements (HSSR) having a nominal yield strength exceeding accuracy of shear–friction design equations in ACI 318-19 (ACI
414 MPa (60 ksi). Barbosa et al. (2017) concluded that although 318, 2019), AASHTO LRFD 2017 (AASHTO, 2017), and CSA–
the ultimate shear capacity of the specimens reinforced with S6 (2000), can be used to evaluate the degree of concrete
Grade 550 (550 MPa) #16M steel bars is greater than that of the contribution. The reason for this method is as follows. The
Grade 420 (414 MPa) #16M steel bars, a similar increase is not shear–friction design equation of each code includes the friction
observed in the #13M specimens in which negligible differences term, and the value of coefficient of friction is the same, that is,
are observed in the ultimate shear capacity. Harries et al. (2012) the three codes have a consensus on the contribution of shear
reported that the ultimate shear capacity is unaffected by the reinforcement. However, the contribution of concrete is considered
grade of reinforcing steel for push–off cold joint specimens with differently. Regardless of the increasing complexity in terms of
ASTM A615/A615M (with a nominal yield strength of 414 MPa understanding arising from high–strength concrete, the ACI 318–
[60 ksi]) and A1035/A1035M bars (with a nominal yield strength of 19 design equation has maintained the simple format proposed by
690 MPa [100 ksi]). Shear transfer models, which are based on Birkeland and Birkeland (1966), including the friction term only.
“shear friction theory,” are semi–empirical models that have The AASHTO LRFD 2017 and CSA–S6 design equations
been calibrated, mainly using push–off test data (Rahal et al., regard the contribution of concrete as a constant cohesion for a
2016); thus, additional push–off tests containing HSSR are certain interface condition, and the cohesion in the former is
required to explore the effect of HSSR and identify the degree of considerably bigger than the one in the latter. Generally, the
KSCE Journal of Civil Engineering 2067
design equation with a higher calculation accuracy considers the an interface reinforcement ratio of 0.005 (6 × 3.14 × 8 × 8/4/200/
contribution of concrete more reasonably. 300 = 0.005). Three groups of push–off specimens were designed:
This study aims to determine the extent to which shear C2N group, C5N group, and C5T group. In the group names, the
reinforcements and concrete contribute to the shear capacity of first two characters stand for the design CCS, “C2” for 20 MPa
cold joints and explore the effect of HSSR on the shear transfer and “C5” for 50 MPa. The next character denotes the nominal
behavior. Compared with studying the contribution of shear YSSR, “N” for 400 MPa and “T” for 600 MPa. There were three
reinforcement or concrete alone, studying them together can duplicated specimens in each group, and a total of nine specimens
bring a more in–depth and systematic understanding. However, were tested.
research that focuses on them together are rare. In this study, the Wood molds were used to cast all specimens. The old concrete
effects of shear reinforcement strength and concrete strength on portion was cast first, and the new–to–old concrete interface
the interface shear behavior were studied by push–off tests. mold was removed after 24 h and the interface was kept in a
Accordingly, the degrees of shear reinforcement contribution and naturally smooth condition (that is, no interface roughened
concrete contribution to the interface shear capacity were measure was taken), as shown in Fig. 2. After an interval of 28
determined. Furthermore, the calculation accuracy of design days, the new concrete portion was cast. The specimens were
equations in the three codes mentioned above was analyzed on stored in laboratory environmental conditions for curing purposes.
the basis of a database of 108 push–off test results established in The mechanical properties of the shear reinforcement and
this study. Then, the degree of concrete contribution was also concrete on the day of push–off testing (approximately 50 days
identified. Additionally, this study is the only known study of after new concrete casting) can be found in Table 1. The shear
shear transfer behavior of smooth cold joints including HSSR. reinforcement did not have a flat–top yield curve, and the 0.2%
offset method was used to calculate the yield stress (Barbosa et
2. Experimental Program al., 2017; Kahn and Mitchell, 2002).
Tests on push–off specimens were conducted using a
Figure 1 shows the details of the push–off specimen. An L shape servocontrolled actuator with a capacity of 1,000 kN. The load
of old concrete and another of new concrete that are connected was applied through two steel plates with a size of 200 × 150 ×
together at an interfacial zone formed each specimen. The 15 mm (length × width × height) and was concentric with the test
interface zone is rectangular with a dimension of 300 mm long interface, as illustrated in Fig. 3. Before starting the test, a force
by 200 mm wide. The test variables included in this study are of 1.0 kN was imposed on the specimen, and then the load was
CCS (concrete compressive strength) and YSSR (yield strength applied at a rate of 0.3 kN/s until the specimen failed. The
of shear reinforcement). C20 and C50 concrete was selected as vertical relative slip and lateral dilation between the two parts of
the example of low–strength and high–strength concrete, the interface were measured by three LVDTs [see Fig. 3]. The
respectively. HRB400 and HRB600 steel reinforcing bars with a LVDTs were attached to plastic fixers which were glued to one
nominal diameter of 8mm and nominal strengths of 400 MPa side of the specimen through a mortar with epoxy resin. The tips
and 600 MPa, respectively, were used as shear reinforcements
and vertically placed across the interface. All specimens had
three double–legged stirrups crossing the interface, resulting in
rises from 20 MPa to 50 MPa. Therefore, CCS has a remarkable 3.3.2 Effect of CCS on Interface Stiffness
effect on Vu. The greater the concrete strength, the greater the Figure 9 presents the comparison of δult and ωult between groups
ultimate shear capacity of interface. C2N and C5N. As shown in Fig. 9 and Table 2, the displacement
For the shear transfer mechanism of interface, the widely– of each specimen in group C5N is remarkably smaller than that
used shear friction mechanisms include: 1) cohesion (understood of each specimen in group C2N. Furthermore, for groups C2N
as aggregate interlock) between concrete surfaces that form the and C5N, the average values of δult are 1.458 and 0.032 mm,
interface; 2) friction resulting from the normal stress being respectively, and the average values of ωult are 0.231 and 0.038
applied by both normal loads and shear reinforcements that mm, respectively. These results show that when CCS rises from
tension due to the dilation caused by the roughness of the interface 20 MPa to 50 MPa, δult and ωult are reduced markedly. Therefore,
combined with the relative displacements of the concrete parts; CCS has a notable effect on δult and ωult. The higher the concrete
and 3) dowel action from shear reinforcements that cross the strength, the smaller the interface displacements. This is because
interface (Zilch and Reinecke, 2000). When the interface reaches when the concrete strength is low, the concrete cover begins to
the ultimate shear capacity, the dowel action is insignificant, and spall before the interface reaches the ultimate capacity, as
the concrete strength can only explicitly affect the first of these discussed previously, and the behavior of the interface softens
three components (i.e., cohesion). Furthermore, the greater the due to spalling.
strength of concrete, the greater the cohesion of interface (Mattock, According to the analysis above, when CCS rises from 20 MPa
2001; Mansur et al., 2008). Therefore, the experimental result to 50 MPa, Vu increases by approximately 110%, and the
that the shear capacity of interface increases with the increase of corresponding slip δult apparently decreases, indicating that the
concrete strength, indicates that cohesion has a significant increase of CCS can significantly increase the shear strength and
contribution to the ultimate shear capacity of interface. stiffness of interface. Therefore, concrete significantly contributes to
the shear strength and stiffness of interface in the form of
cohesion.
Using the measured shear reinforcement strain, Harries et al.
(2012), Jiang et al. (2016) and Semendary et al. (2020) determined
the shear reinforcement clamping component of shear friction.
Furthermore, the concrete component was calculated by subtracting
the shear reinforcement component from the applied shear load.
They found that the concrete component contributes to the
majority of the shear capacity of the interface. Specifically,
Semendary et al. (2020) stated that the concrete component
contributes to at least 73.5% of the shear capacity. From the
perspective of the effect of CCS, this study shows that concrete
significantly contributes to the shear capacity.
Table 3. Design Provisions in ACI 318–19, AASHTO LRFD 2017 and CSA–S6
Code ACI 318–19 AASHTO LRFD 2017 CSA–S6
Design equation V =A f µ
n vf y Vn =c ⋅ Acv + µ ⋅ ( Avf f y + Pc ) Vn =c ⋅ Acv + µ ⋅ ( Avf f y + Pc )
in the following were adjusted as necessary to maintain consistency data originated from specimens with a smooth and roughened
between different provisions. interface, respectively. The overview and values of experimental–
The nominal capacity in AASHTO LRFD 2017 and CSA–S6 to–nominal shear capacity ratio Vu /VACI, Vu /VAAS, and Vu /VCSA
is based on the cohesion between the interface surfaces and are listed in detail in Tables 4 and 5, wherein Vu is the
friction that results from shear reinforcements crossing the experimental shear capacity, and VACI, VAAS, and VCSA are the
interface. Significantly, the cohesion in AASHTO LRFD 2017 is nominal shear capacities according to the design equations of
assumed to have a greater contribution to the shear capacity in ACI 318–19, AASHTO LRFD 2017, and CSA–S6, respectively.
comparison with that in CSA–S6. The interface shear transfer In addition, Vu /VACI, Vu /VAAS, and Vu /VCSA are plotted against the
model in ACI 318–19 considers shear capacity by friction only. concrete cylinder compressive strength of specimens fc' in Fig.
Thus, the ascending order of the assumed cohesion contribution 10. For cold joint specimens with different concrete strengths on
to interface shear capacity is as follows: ACI 318–19 equation, each side of the interface, the lower cylinder compressive
CSA–S6 equation, and AASHTO LRFD 2017 equation. strength was reported.
The average values of the experimental-to-nominal shear
4.2 Evaluation Results and Analysis capacity ratio shown in Tables 4 and 5 reveal that the ascending
As presented in Table 3, the shear–friction equations in these order of calculation accuracy is as follows: ACI 318–19 equation,
three codes adopt the same coefficient of friction for the interface CSA–S6 equation, and AASHTO LRFD 2017 equation; this
of the same type. However, two major differences exist among order is the same as that of the assumed cohesion contribution to
the provisions. Except for different considerations about cohesion, the interface shear capacity. This analysis result indicates that the
the limitations are also different. ACI 318–19 and AASHTO AASHTO LRFD 2017 equation considers the contribution of
LRFD 2017 limit the design yield strength of any HSSR for concrete more reasonably and cohesion has a significant contribution
interface shear applications to 414 MPa, and CSA–S6 500 MPa. to the shear capacity of interface. In addition, all minimums of
Besides, the upper limits of shear capacity vary. Given the the experimental–to–nominal shear capacity ratio are bigger than
purpose of code evaluation in this study, the limitations in the 0.75, which is the shear strength reduction factor of ACI 318–19.
other two codes are modified to the one in ACI 318–19 in the This finding indicates that even if the ACI 318–19 design equation
following evaluation. In doing so, the consideration about cohesion is modified to the design equation in CSA–S6 or AASHTO
becomes the only difference among the provisions, and the LRFD 2017, retaining the existing limitations in ACI 318–19, it
influence of cohesion on the calculation accuracy of design can still be used to guide engineering designs. The maximums of
equations can be specifically and reasonably considered. On this Vu /VACI are 3.97 and 6.82 for smooth and roughened interfaces,
basis, the code evaluation in this study differs from that in other respectively, indicating that without the cohesion term, ACI 318–
literature. 19 equation is excessively conservative for high–strength concrete.
Based on 108 test data on classical normal–weight cold joint As shown in Fig. 10, Vu /VACI, Vu /VAAS, and Vu /VCSA are positively
push–off specimens (without inclined reinforcement across the correlated with CCS. This is because that the cohesion of
interface or an external force perpendicular to the interface), interface is positively related to the strength of concrete forming
which was composed of 9 test data from this study and 99 available the interface (Mattock, 2001; Mansur et al., 2008), whereas the
published test data collected from the literature (Mattock, 1976; design equations in the three codes do not consider the effect of
Kahn and Mitchell, 2002; Harries et al., 2012; Shaw and Sneed, concrete strength on cohesion. This phenomenon also indicates
2014; Figueira et al., 2015; Barbosa et al., 2017; Liu et al., 2019; that cohesion contributes significantly to the shear capacity of
Semendary et al., 2020), the accuracy of design equations of the interface. In addition, compared with the smooth interfaces, the
three codes was analyzed. Among the test data, 48 and 60 test positive relationship between the experimental–to–nominal shear
KSCE Journal of Civil Engineering 2073
Table 4. Push–Off Tests of Classical Normal Weight Concrete Specimens with Smooth Cold Joint
Reference Specimens fc' (MPa) ρ fy (MPa) Vu (N) Vu / VACI Vu / VAAS Vu /VCSA
Kahn and Mitchell SF–10–1–CJ 98.8 0.0037 572 141,135 3.97 2.53 3.12
(2002) SF–10–2–CJ 83.1 0.0074 572 219,242 3.08 2.40 2.71
Liu et al. (2019) M–1 56.6 0.005 446 252,500 3.39 2.39 2.82
M–2 56.6 0.005 446 216,500 2.91 2.05 2.42
M–3 56.6 0.005 446 211,600 2.84 2.00 2.36
M–4 56.6 0.005 446 207,700 2.79 1.96 2.32
M–5 56.6 0.005 446 212,500 2.85 2.01 2.37
M–6 56.6 0.005 446 254,500 3.42 2.41 2.84
M–7 56.6 0.005 446 214,600 2.88 2.03 2.40
M–8 56.6 0.005 446 213,900 2.87 2.02 2.39
M–9 56.6 0.005 446 201,400 2.70 1.91 2.25
M–10 56.6 0.005 446 197,500 2.65 1.87 2.21
Mattock (1976) C1 40.5 0.0044 351 46,704 1.56 1.00 1.23
C2 40.5 0.0088 351 80,064 1.34 1.05 1.18
C3 41.2 0.0132 348 95,187 1.07 0.90 0.98
C4 41.2 0.0176 356 133,440 1.10 0.97 1.03
C5 42.6 0.022 364 173,472 1.12 1.01 1.07
C6 42.6 0.032 312 196,157 1.10 1.10 1.10
G1 40.5 0.0044 351 35,584 1.19 0.76 0.94
G2 40.5 0.0088 351 58,714 0.98 0.77 0.87
G3 41.2 0.0132 348 85,402 0.96 0.81 0.88
G4 41.2 0.0176 356 111,200 0.92 0.81 0.86
G5 42.6 0.022 364 130,326 0.84 0.76 0.80
G6 42.6 0.032 312 173,027 0.97 0.97 0.97
H1 40.2 0.0044 382 41,811 1.28 0.85 1.03
H2 41.9 0.0088 382 71,613 1.10 0.87 0.98
H3 41.9 0.0132 382 102,304 1.05 0.89 0.97
H4 41.9 0.0176 370 113,424 0.90 0.79 0.85
H5 42.6 0.0248 323 145,450 0.94 0.85 0.89
H6 40.7 0.032 323 169,024 0.95 0.95 0.95
Shaw and Sneed N–5–S–4 33.5 0.013 456 137,221 1.33 1.15 1.24
(2014) N–5–S–5 33.5 0.013 456 154,257 1.50 1.29 1.39
N–5–S–6 33.5 0.013 456 174,139 1.69 1.45 1.57
N–8–S–1 52.0 0.013 456 291,611 2.83 2.44 2.62
N–8–S–2 52.0 0.013 456 237,078 2.30 1.98 2.13
N–8–S–3 52.0 0.013 456 246,108 2.39 2.06 2.21
Figueira et al. (2015) 1 44.2 0.0107 605 202,184 2.03 1.70 1.85
2 44.2 0.0107 605 196,148 1.97 1.65 1.80
3 44.2 0.0107 605 201,321 2.02 1.69 1.85
This study C2N–1 19.8 0.005 450 102,800 1.38 0.97 1.15
C2N–2 19.8 0.005 450 100,800 1.35 0.95 1.13
C2N–3 19.8 0.005 450 107,100 1.44 1.01 1.20
C5N–1 51.0 0.005 450 241,000 3.23 2.28 2.69
C5N–2 51.0 0.005 450 197,200 2.65 1.87 2.20
C5N–3 51.0 0.005 450 214,300 2.88 2.03 2.39
C5T–1 51.0 0.005 645 203,200 2.73 1.92 2.27
C5T–2 51.0 0.005 645 239,400 3.21 2.26 2.67
C5T–3 51.0 0.005 645 221,900 2.98 2.10 2.48
Average 1.99 1.51 1.72
Coefficient of Variation 0.46 0.40 0.42
Maximum 3.97 2.53 3.12
Minimum 0.84 0.76 0.80
Note: The limitations in AASHTO LRFD 2017 and CSA–S6 are modified to the one in ACI 318–19 in the calculation of nominal shear capacity.
2074 J. Liu et al.
Table 5. Push–Off Tests of Classical Normal Weight Concrete Specimens with Roughened Cold Joint
Reference Specimens fc' (MPa) ρ fy (MPa) Vu (N) Vu / VACI Vu / VAAS Vu /VCSA
Barbosa et al. (2017) 4G60–1 30.9 0.0041 473 1,241,180 2.95 1.50 2.28
4G60–2 30.9 0.0041 473 1,141,180 2.71 1.38 2.10
4G60–3 28.9 0.0041 473 1,087,590 2.59 1.31 2.00
4G60–4 28.9 0.0041 473 1,175,370 2.80 1.42 2.16
4G60–5 28.9 0.0041 473 1,200,000 2.85 1.45 2.21
4G80–1 30.1 0.0041 591 1,035,550 2.46 1.25 1.90
4G80–2 28.9 0.0041 591 1,083,240 2.58 1.31 1.99
4G80–3 28.9 0.0041 591 1,123,130 2.67 1.35 2.06
4G80–4 28.9 0.0041 591 1,166,550 2.78 1.41 2.14
4G80–5 28.9 0.0041 591 1,292,740 3.08 1.56 2.38
5G60–1 31.6 0.0064 443 1,215,530 2.47 1.52 2.07
5G60–2 31.6 0.0064 443 1,227,000 2.49 1.53 2.09
5G60–3 28.6 0.0064 443 1,233,780 2.50 1.54 2.11
5G60–4 28.6 0.0064 443 1,210,580 2.46 1.51 2.07
5G60–5 28.6 0.0064 443 1,223,950 2.48 1.53 2.09
5G80–1 31.6 0.0064 589 1,342,060 2.72 1.68 2.29
5G80–2 28.6 0.0064 589 1,359,700 2.76 1.70 2.32
5G80–3 28.6 0.0064 589 1,389,690 2.82 1.74 2.37
5G80–4 28.6 0.0064 589 1,257,330 2.55 1.57 2.15
5G80–5 28.6 0.0064 589 1,322,190 2.68 1.65 2.26
Harries et al. (2012) 615–3A 40.0 0.0041 464 500,400 2.85 1.44 2.20
615–3B 40.0 0.0041 464 429232 2.40 1.22 1.86
615–4A 40.0 0.0073 424 509,296 1.58 1.02 1.36
615–4B 40.0 0.0074 424 573,792 1.79 1.16 1.54
1035–3A 40.0 0.0042 896 400,320 2.27 1.16 1.76
1035–3B 40.0 0.0041 869 467,040 2.65 1.35 2.05
1035–4A 40.0 0.0074 965 603,594 1.88 1.22 1.62
1035–4B 40.0 0.0075 905 504,848 1.57 1.02 1.35
Semendary et al. P–SR–A 63.0 0.0055 414 160,400 2.73 1.58 2.24
(2020) P–SR–B 63.0 0.0055 414 166,900 2.84 1.65 2.33
Kahn and Mitchell SF–7–1–CJ 80.9 0.0037 572 240,192 4.05 1.95 3.05
(2002) SF–7–2–CJ 80.9 0.0074 572 365,181 3.08 2.00 2.65
SF–7–3–CJ 86.0 0.0111 572 490,614 2.76 2.03 2.49
SF–7–4–CJ 86.0 0.0148 572 590,161 2.49 1.96 2.30
SF–10–3–CJ 89.3 0.0111 572 506,672 2.85 2.10 2.57
SF–10–4–CJ 89.3 0.0148 572 560,626 2.36 1.86 2.19
SF–14–1–CJ 101.7 0.0037 572 404,368 6.82 3.28 5.14
SF–14–2–CJ 101.7 0.0074 572 441,197 3.72 2.42 3.20
SF–14–3–CJ 104.9 0.0111 572 599,190 3.37 2.48 3.04
SF–14–4–CJ 104.9 0.0148 572 681,078 2.87 2.26 2.65
Mattock (1976) B1 40.3 0.0044 353 108,309 2.16 1.05 1.63
B2 40.3 0.0088 348 155,680 1.57 1.02 1.35
B3 41.7 0.0132 353 234,410 1.56 1.15 1.41
B4 41.7 0.0176 371 283,782 1.35 1.31 1.31
B5 40.6 0.0248 340 349,168 1.64 1.64 1.64
B6 40.6 0.032 340 378,080 1.77 1.77 1.77
D1 26.0 0.0044 353 131,216 2.62 1.27 1.98
D2 26.0 0.0088 353 204,608 2.04 1.33 1.76
D3 20.3 0.0132 386 224,624 1.72 1.72 1.72
D4 20.3 0.0176 386 222,845 1.70 1.70 1.70
D4A 17.2 0.0176 372 221,066 1.99 1.99 1.99
D5 20.4 0.0248 320 269,104 2.05 2.05 2.05
D5A 19.3 0.0248 319 278,000 2.24 2.24 2.24
B1 40.3 0.0044 353 108,309 2.16 1.05 1.63
KSCE Journal of Civil Engineering 2075
Table 5. (continued)
Reference Specimens fc' (MPa) ρ fy (MPa) Vu (N) Vu / VACI Vu / VAAS Vu /VCSA
Shaw and Sneed N–5–R–4 33.5 0.013 456 262,699 1.53 1.36 1.40
(2014) N–5–R–5 33.5 0.013 456 237612 1.38 1.23 1.26
N–5–R–6 33.5 0.013 456 237,701 1.38 1.23 1.27
N–8–R–1 52.0 0.013 456 329,330 1.92 1.47 1.75
N–8–R–2 52.0 0.013 456 249,488 1.45 1.11 1.33
N–5–R–4 33.5 0.013 456 262,699 1.53 1.36 1.40
Average 2.44 1.59 2.07
Coefficient of Variation 0.34 0.27 0.29
Maximum 6.82 3.28 5.14
Minimum 1.35 1.02 1.26
Note: The limitations in AASHTO LRFD 2017 and CSA–S6 are modified to the one in ACI 318–19 in the calculation of nominal shear capacity.
Table 6. The Statistical Characteristics of the Experimental–to–Nominal YSSR increases from 400 MPa to 600 MPa, the change in
Shear Capacity Ratio Vu /Vpro the ultimate shear capacity of interface Vu and the corresponding
Interface condition Smooth Roughened interface shear slip δult and crack width ωult can be ignored,
Average 1.68 1.63 indicating that YSSR has no significant effect on the shear
Coefficient of variation 0.30 0.21 strength and stiffness of interface. This may be due to the
Maximum 3.07 2.49 fact that when the interface reaches the ultimate shear
Minimum 0.81 1.00 capacity, the shear reinforcement has not yet yielded, and it
is only partially involved in the interface shear resistance.
2. It is observed in the push–off tests that when CCS rises
from 20 MPa to 50 MPa, Vu increases by approximately
110% and δult apparently decreases, showing that the increase
in CCS can substantially increase the shear strength and
stiffness of interface. This indicates that concrete significantly
contributes to the shear capacity and stiffness of interfaces.
In summary, the push–off test results of this study show
that concrete has a significant contribution to the interface
shear capacity, whereas the contribution of shear
reinforcements is not as great as expected.
3. The results of the code provision evaluation show that, the
ascending order of design equation calculation accuracy is
the same as the ascending order of the assumed cohesion
contribution to interface shear capacity (i.e., ACI 318–19
equation, CSA–S6 equation, and AASHTO LRFD 2017
equation). This indicates that concrete has a notable
Fig. 11. Concrete Compressive Strength versus the Ratio of Experimental
Shear Capacity to the Nominal Shear Capacity Computed contribution to the interface shear capacity in the form of
Using the Proposed Equation of This Study cohesion. In addition, the phenomenon in which the
experimental–to–nominal shear capacity ratios Vu /VACI, Vu /
VAAS, and Vu /VCSA are positively correlated with CCS
Table 6 and Fig. 11 reveal that, Vu /Vpro has a reasonable confirms this issue.
average, and compared with Vu /VACI, Vu /VAAS, and Vu /VCSA, it has 4. The proposed shear capacity equation of this study, which
a smaller coefficient of variation and no longer presents a accounts for the effect of CCS and the limited participation
tendency of increasing with the increase of CCS. This indicates of shear reinforcements, can be well used to predict the
that the proposed equation of this study can be well used to cold joint shear capacity, especially for the cold joint
predict the interface shear capacity, and CCS has no effect on its composed of high–strength concrete. In addition, no obvious
calculation accuracy. relationship between the experimental–to–nominal shear
It should be specially pointed out that, for high–strength capacity ratios Vu /Vpro and CCS is noted.
concrete specimens (with CCS larger than 60 MPa), the average 5. Both YSSR and CCS have a significant effect on the
values of Vu /Vpro, Vu /VACI, Vu /VAAS, and Vu /VCSA are 1.69, 3.52, residual strength of interface. This can be explained by the
2.47, and 2.92 for smooth cold joints, respectively, and 1.49, fact that when the residual strength is reached, the shear
3.33, 2.13, and 2.82 for roughened cold joints, respectively. This force is transferred mainly by dowel action, which is
indicates that compared with the three design equations mentioned affected by both shear reinforcements and concrete.
above, the proposed equation of this study can be better applied The results from this study aid in the clarity of the degree to
to high–strength concrete. which the shear reinforcement and concrete contribute to the
shear capacity of cold joints. Variables out of the scope of the
6. Conclusions study (e.g., interface roughness, reinforcing bar sizes, reinforcing
bar spacing, and more reinforcement and concrete strengths)
The contribution of shear reinforcements and concrete to the should be evaluated. It is suggested that more tests should be
shear capacity of cold joints was reconsidered together, and the carried out on this topic.
effect of HSSR was also investigated in this study. Push–off tests
and code provisions evaluation were employed to determine the Acknowledgments
degrees of shear reinforcements contribution and concrete
contribution. Based on the conducted study, main conclusions The authors wish to acknowledge the financial support from
can be drawn as following: National Natural Science Foundation of China (Grant No.
1. The push–off test results demonstrate that when nominal 51641807), Natural Science Foundation of Hubei Province
KSCE Journal of Civil Engineering 2077
(Grant No. 2018CFB664), and Open Research Fund Project of Liu J, Fan JX, Chen JJ, Xu G (2019) Evaluation of design provisions for
Hubei Key Laboratory of Disaster Prevention and Mitigation interface shear transfer between concretes cast at different times.
(China Three Gorges University) (Grant No. 2018KJZ06). The Journal of Bridge Engineering 24(6):06019002, DOI: 10.1061/
(ASCE)BE.1943-5592.0001393
authors also thank Zhongguo John Ma from University of Tennessee,
Loov RE, Patnaik AK (1994) Horizontal shear strength of composite
Knoxville, TN, United States, for his valuable suggestions. concrete beams with a rough interface. PCI Journal 38(1):48-69
Mansur MA, Vinayagam T, Tan KH (2008) Shear transfer across a
ORCID crack in reinforced high–strength concrete. Journal of Materials in
Civil Engineering 20(4):294-302, DOI: 10.1061/(ASCE)0899-1561
Jie Liu https://orcid.org/0000-0003-4730-9497 (2008)20:4(294)
Qing Wang https://orcid.org/0000-0002-8524-1465 Mattock AH (1976) Shear transfer under monotonic loading across an
interface between concretes cast at different times. Rep. SM 76-3,
Department of Civil Engineering, University of Washington, Seattle,
References WA, USA
Mattock AH (1994) Reader comments: Horizontal shear strength of
AASHTO (2017) AASHTO bridge design specifications, 8th edition. composite concrete beams with a rough interface. PCI Journal
American Association of State Highway and Transportation Officials, 39(5):106-108
Washington DC, USA Mattock AH (2001) Shear friction and high–strength concrete. ACI
ACI 318 (1977) ACI 318–77: Building code requirements for structural Structural Journal 98(1):50-59
concrete. ACI 318–77, American Concrete Institute, Farmington Mattock AH (2013) Toward an improved understanding of shear-
Hills, MI, USA friction behavior. ACI Structural Journal 110(5):890-891
ACI 318 (2019) ACI 318–19: Building code requirements for structural Rahal KN, Khaleefi AL, Al–Sanee A (2016) An experimental investigation
concrete. ACI 318–19, American Concrete Institute, Farmington of shear–transfer strength of normal and high strength self compacting
Hills, MI, USA concrete. Engineering Structures 10:16-25, DOI: 10.1016/j.engstruct.
Barbosa AR, Trejo D, Nielson D (2017) Effect of high–strength 2015.11.015
reinforcement steel on shear friction behavior. Journal of Bridge Semendary AA, Hamid WK, Steinberg EP, Khoury I (2020) Shear
Engineering 22(8):04017038, DOI: 10.1061/(ASCE)BE.1943-5592. friction performance between high strength concrete (HSC) and
0001015 ultra high performance concrete (UHPC) for bridge connection
Birkeland PW, Birkeland HW (1966) Connections in precast concrete applications. Engineering Structures 205:110122, DOI: 10.1016/
construction. ACI Journal Proceedings 63(3):345-368 j.engstruct.2019.110122
CAN/CSA–S6–00 (2000) CAN/CSA–S6–00: Canadian highway bridge Shaw D, Sneed LH (2014) Interface shear transfer of lightweight–
design code. Canadian Standards Association, Toronto, ON, Canada aggregate concretes cast at different times. PCI Journal 59(3):130-
Figueira D, Sousa C, Calçada R, Neves AS (2015) Push–off tests in the 144
study of cyclic behavior of interfaces between concretes cast at Walraven J, Frenay J, Pruijssers A (1987) Influence of concrete strength
different times. Journal of Structure Engineering 142(1):0401510, and load history on the shear friction capacity of concrete members.
DOI: 10.1061/(ASCE)ST.1943-541X.0001364 PCI Journal 32(1):66-84
Harries KA, Zeno G, Shahrooz B (2012) Toward an improved understanding Walraven J, Reinhardt H (1981) Theory and experiments on the mechanical
of shear friction behavior. ACI Structural Journal 109(6):835-844 behaviour of cracks in plain and reinforced concrete subject to shear
Jiang H, Fang Z, Ma ZJ, Fang X, Jiang Z (2016) Shear-friction behavior loading. Heron 26(1A):1-68
of groove interface in concrete bridge rehabilitation. Journal of Williams CS, Massey JB, Bayrak O, Jirsa JO (2017) Investigation of
Bridge Engineering 21(11):04016081, DOI: 10.1061/(ASCE)BE. interface shear transfer using push-through tests. ACI Structural
1943-5592.0000953 Journal 114(1):173-185
Kahn LF, Mitchell AD (2002) Shear friction test with high–strength Xiao J, Sun C, Lange DA (2016) Effect of joint interface conditions on
concrete. ACI Structural Journal 99(1):98-103 shear transfer behavior of recycled aggregate concrete. Construction
Kono S, Tanaka H (2000) Interface shear transfer for high strength and Building Materials 105:343-355, DOI: 10.1016/j.conbuildmat.
concrete and high strength reinforcement. 12th World Conference 2015.12.015
on Earthquake Engineering, Upper Hutt, New Zeland, 0641 Zilch K, Reinecke R (2000) Capacity of shear joints between high
Kono S, Tanaka H, Watanabe F (2001) Interface shear transfer for high strength precast elements and normal–strength cast–in–place decks.
strength concrete and high strength shear friction reinforcement. PCI/FHWA/FIB international symposium on high performance
International conference on high performance materials in bridges, concrete, September 25-27, Orlando, FL, USA, 551-560
July 29-August 3, Kona, HI, USA, 319-328