ACI STRUCTURAL JOURNAL TECHNICAL PAPER

Title No. 112-S68

Bond-Slip Relationship of Beam Flexural Bars in Interior

Beam-Column Joints
by Hyeon-Jong Hwang, Tae-Sung Eom, and Hong-Gun Park
Under cyclic loading, the structural performance of reinforced
concrete (RC) beam-column connections is significantly affected
by the bond slip of beam flexural bars. In the present study, a bond-
slip model was developed to evaluate the bond slip of beam flexural
bars within beam-column joints using simplified bond strength and
bar strain. To address the cyclic loading effect, the bond strength
was determined from the existing test results of beam-column
connections that showed complete bond failure. For verification,
the predictions of the proposed model were compared with the
existing test results of concrete block specimens and beam-column
joint specimens. The results showed that the proposed model
predicted the bond strength degradation and bond-slip relation-
ships with reasonable precision. For performance-based design,
a bond requirement was proposed as the function of the ductility
demand of beam-column joints.

Keywords: beam-column interior joint; bond slip; cyclic loading; simpli-

fied bond strength model.

INTRODUCTION
In reinforced concrete moment-resisting frames subjected
to earthquake load, the structural performance of beam- Fig. 1—Load transfer of interior beam-column joint.
column connections is significantly affected by the bond minimum requirements of hc/db (refer to Appendix A*). In
slip of beam flexural bars in the beam-column joints.1-8 In ACI 352R‑02,11 the minimum hc/db is defined as the func-
the interior beam-column joints subjected to large inelastic tion of the yield strength ratio of reinforcing bars: hc/db ≥
deformation, the bond slip of the beam flexural bars contrib- 20(fy/420) ≥ 20. NZS 3101:200612 and Eurocode 813 use
utes to the total deformation of the beam-column joints by as additional design parameters, including concrete tensile
much as 35%.9 strength, column axial load, and structural performance
Figure 1 shows the load-transfer mechanism in an interior demand. The design requirements were developed on the
beam-column joint. Under reversed cyclic loading, residual basis of existing test results of beam-column connections.
tensile deformations develop in the beam flexural bars after However, the current design codes show significant differ-
flexural yielding. Due to the residual tensile deformation, ences in the requirement of the hc/db.
both the tensile stress and compressive stress of the beam Regarding the existing tests for the bond slip of reinforcing
reinforcing bars can develop even before complete closing bars, Eligehausen et al.14 studied the bond performance of
of the flexural cracks at the joint interface. In this case, to reinforcing bars (in the elastic state) embedded in concrete
transfer the reinforcing bar forces to the joint without contact blocks under monotonic and cyclic loadings, and proposed
of the concrete, the sum of the compressive force and tensile a bond stress-slip relationship. Ciampi et al.15 and Elmorsi
force (T1 + C2 or T2 + C1 in Fig. 1) of the beam reinforcing et al.16 developed a bond model for finite element analysis
bars should be resisted by the bar bond within the joint. of reinforced concrete by modifying the Eligehausen model.
Thus, the bond stress demand in the beam-column joint Viwathanatepa et al.17 developed a bond stress-slip relation-
subjected to cyclic loading can be increased to two times the ship for reinforcing bars after yielding, under cyclic loading.
bond stress under monotonic loading: T1 + C2 or T2 + C1, as
shown in Fig. 1. On the other hand, the development length ACI Structural Journal, V. 112, No. 6, November-December 2015.
of the beam flexural bar is limited by the column depth hc. MS No. S-2014-158.R1, doi: 10.14359/51687708, received October 30, 2014, and
reviewed under Institute publication policies. Copyright © 2015, American Concrete
Thus, in the beam-column joints with small column depth- Institute. All rights reserved, including the making of copies unless permission is
obtained from the copyright proprietors. Pertinent discussion including author’s
to-bar diameter ratio hc/db (db is the diameter of a beam flex- closure, if any, will be published ten months from this journal’s date if the discussion
ural bar), significant bond slip occurs under cyclic loading.3 is received within four months of the paper’s print publication.
To mitigate the bond slip of beam flexural bars, the
*
The Appendix is available at www.concrete.org/publications in PDF format,
current earthquake design codes (ACI 318-11,10 ACI 352R- appended to the online version of the published paper. It is also available in hard copy
02,11 NZS 3101:2006,12 and Eurocode 813) specify the from ACI headquarters for a fee equal to the cost of reproduction plus handling at the
time of the request.

ACI Structural Journal/November-December 2015 827

Fig. 2—Bond resistance mechanism of reinforcing bar.14
Alsiwat and Saatcioglu18 and Marti et al.19 proposed a strain- RESEARCH SIGNIFICANCE
based bond model, separating the post-yield behavior from The proposed bond-slip model was developed based on
the pre-yield behavior. Hong et al.8 and Lowes and Altoon- existing models. However, to address the effects of the actual
tash20 applied the strain-based bond model to interior beam- loading and boundary conditions, the bond strength and
column joints to predict the strain distribution and bond slip stress distribution were redefined by analyzing the existing
of the beam flexural bars. test results of the actual beam-column connections. Further,
While the aforementioned bond models successfully unlike the prescriptive provisions of the current design
described and predicted the bond-slip behavior, the following codes, the bond requirement was defined as the function
studies are required to more accurately describe the bar bond of the ductility demand of beam-column joints, so that the
slip of actual beam-column connections. proposed method can be used for performance-based design.
1. In the existing tests for bar bond slip, concrete blocks
were used to evaluate the bond slip of a reinforcing bar. BAR BOND-SLIP RELATIONSHIP
However, the loading and boundary conditions of the Figure 2 shows the variation of the bond resistance
concrete block specimens differ from the actual condi- according to bond slip in the pullout mechanism of a rein-
tions of the beam-column joints. In the concrete block forcing bar embedded in concrete.14 It is assumed that the
tests, the reinforcing bar is anchored to the concrete inside beam-column joint is well-confined by joint hoops and,
the concrete block, and the stress of the reinforcing bar is thus, splitting failure along the bar development length
determined only by the concrete bond stress. Thus, after the does not occur. Initially, the bond resistance is provided by
peak bond stress (that is, after bond failure), the stress and the bearing of the ribs in a reinforcing bar. When concrete
strain of the reinforcing bar decreases with the bond stress. crushing occurs due to the rib bearing, the bond stress
On the other hand, in actual interior beam-column connec- reaches its peak value. After peak strength, the bearing resis-
tions, even after bond failure occurs in the joint, the beam tance significantly decreases, and residual bond strength is
flexural bars can be anchored to the beam on the opposite provided by the surface friction along the rib tips. Figure 2(c)
side. Thus, even after the post-peak bond stress decreases, shows the bond stress-slip relationship of a reinforcing bar
the yield stress of the beam reinforcing bars is maintained, under load reversals (that is, cyclic loading). After concrete
and the strain can significantly increase. Thus, the degrada- bearing failure occurs (refer to Fig. 2(b)), even under load
tion of the bond stress is expected to be more significant in reversals, the bond strength cannot be recovered due to the
the actual beam-column connections than in the test results existing concrete crushing damage, and only the residual
of concrete block tests. bond strength is developed by the friction mechanism. After
2. In the concrete block tests, the bond stress-slip relation- complete concrete bearing failure, the residual bond strength
ship of a reinforcing bar is only affected by the bond damage is maintained uniformly, regardless of the slip deformation.
between the reinforcing bar and concrete. On the other hand, The concrete bearing failure is closely related to rein-
in beam-column joints subjected to shear, the bond stress- forcing bar yielding. Before reinforcing bar yielding occurs,
slip relationship is affected by joint diagonal cracking as because of the large elastic stiffness of the reinforcing bar,
well as bond damage. The hoops in the beam-column joint the relative deformation between the reinforcing bar and
may affect the bond strength, restraining the concrete diag- concrete is small and, thus, the bond stress can be distributed
onal cracking. along the reinforcing bar length. However, after reinforcing
3. The current design codes prescribe the bond require- bar yielding, large inelastic deformation occurs at the loca-
ment of beam flexural bars, regardless of the ductility demand tion of the bar yielding, and the bond stress is concentrated
of beam-column connections. However, for the perfor- at that location. Thus, concrete cracking and bearing failure
mance-based design, the bond requirement needs to be defined occur at the location of the reinforcing bar yielding.
as the function of the ductility demand of the moment frame. For this reason, the bond strength can be defined sepa-
rately before and after the reinforcing bar yielding that causes

828 ACI Structural Journal/November-December 2015

Fig. 3—Stress and strain distributions of beam reinforcing bars in beam-column joint under cyclic loading.
concrete bearing failure. In the present study, following the d ε 4τ ec
studies of Lowes et al.20 and Hong et al.,8 the bond stresses = for − ε y ≤ ε ≤ 0 (2b)
dx Es db
before and after reinforcing bar yielding were simplified as
the uniform effective stresses of bearing bond strength τe and
friction bond strength τu, respectively (refer to Fig. 2(d)). dε 4τ u
= for ε > ε y (2c)
The values of the bearing bond strength and the friction bond dx Esh db
strength are defined in the “BOND STRENGTH” section.
Using the force-equilibrium in a development length dx,
where εy is yield strain of the reinforcing bar.
the bar stress increment-bond stress relationship can be
Figure 3 shows variations of bond stress distributions, bar
defined as follows.
strain distributions, and bar stress distributions according to
the bond slip of a beam flexural bar in an interior beam-
 d2  d σ 4τ e column joint subjected to cyclic loading. As previously
d σ  π b  = τ e (πdb dx) or =
 4 dx db (1a) mentioned, the bond stress distribution was simplified with
the effective bond strengths before and after reinforcing bar
before reinforcing bar yielding
yielding: bearing bond strength (= τet or τec) and friction bond
strength τu. Figure 3(a) presents the bond behavior before
bond failure. In this case, the bond strength Bj of the joint
 d2  d σ 4τ u is large enough to resist the tensile force of the beam rein-
d σ  π b  = τ u (πdb dx) or =
 4 dx db forcing bar, Bj ≥ Ts. Thus, the partial or full bar compressive
(1b)
after reinforcing bar yielding force at the opposite side, as well as the bar tensile force,
can be resisted by the bond strength of the joint. As shown
where db is bar diameter; τe is bearing bond strength; and τu in Fig. 3(a), after yielding of the reinforcing bar occurs at
is friction bond strength. τe = τet when the reinforcing bar is the joint interface, the bond stresses at the joint interfaces
subjected to tension, and τe = τec when the reinforcing bar is are decreased to the frictional bond stress τu. On the other
subjected to compression. hand, in the elastic region of the reinforcing bar inside the
Using a bilinear bar stress-strain relationship with elastic joint, the bearing bond strengths τet and τec are maintained.
stiffness Es and hardening stiffness Esh (that is, hardening In Fig. 3(a), assuming symmetric cyclic loading, the region
due to repeated cyclic loading effect), the bar stress incre- of the friction bond stress was symmetrically defined in
ment before and after yielding can be defined as dσ = Esdε the joint.
and Eshdε, respectively. Therefore, from Eq. (1a) and (1b), Figure 3(b) shows that, as the inelastic deformation of the
the bar strain rate is defined as beam reinforcing bar increases, yielding of the reinforcing
bar penetrates into the joint, and the bond resistance is
degraded (Bj < Ts). Thus, only part of the bar tensile force
d ε 4τ et
= for 0 ≤ ε ≤ ε y (2a) is resisted by the degraded bond strength and the remaining
dx Es db tensile force is anchored to the compression zone of the
beam at the opposite side. However, the elastic region of the
reinforcing bar remains at the center of the joint. In Fig. 3(c),

ACI Structural Journal/November-December 2015 829

as the bond resistance is further degraded, the bond stress BOND STRENGTH
is decreased to the uniform friction bond strength τu over According to previous studies,14,17 before reinforcing bar
the entire joint length. In this case, the majority of the bar yielding, the bond properties under cyclic loading do not
tension force is anchored to the compression zone of the significantly differ from those under monotonic loading.
beam at the opposite side, rather than to the joint. Thus, the bearing bond strengths τet and τec were determined,
In the case of “Before bond failure” shown in Fig. 3(a), following the study by Lowes and Altoontash,20 which was
in the center region where bar strain εt is less than the yield based on the test results of Eligehausen et al.14
strain εy, bearing bond strengths τet and τec develop. The bond
lengths corresponding to τet and τec are defined as let and lec, τet = 1.8√fc′ for tension bars (in MPa) (4)
respectively. Here, the compressive bearing bond strength
τec is greater than the tensile bearing bond strength τet (refer τec = 2.2√fc′ for compression bars (in MPa) (5)
to the “BOND STRENGTH” section). Therefore, the tensile
bearing bond length let is greater than the compressive bond This is similar to the result of the study by Viwathanatepa
length lec, even when the full compression force of the rein- et al.17 On the other hand, MC201021 specifies a range of
forcing bar is anchored to the joint. In the regions close to bearing bond strength τe according to bond condition.
the joint interface, friction bond strength τu develops, and the
length is defined as lu. Under cyclic loading, once bearing 1.25 f c′ ≤ τ e ≤ 2.5 f c′ (in MPa) (6)
bond failure occurs, the damage cannot be recovered even
under the reversed loading. Thus, considering symmetric To define the friction bond strength τu, which indicates the
cyclic loading, the friction bond lengths were symmetrically residual bond strength after reinforcing bar yielding, Lowes
arranged at the two joint interfaces. and Altoontash20 proposed a range of τu in Eq. (7a), on the
In the case of the “partial bond failure” shown in Fig. 3(b), basis of the concrete block tests performed by Eligehausen
the friction bond length lu is determined as the length of the et al.14 and Shima et al.22 From similar bond test results,
region where bar tensile strain εt exceeds the yield strain εy. Viwathanatepa et al.17 proposed Eq. (7b). Marti et al.19
Considering the effect of symmetric cyclic loading, an iden- proposed Eq. (7c) for monotonic loading. MC201021 spec-
tical length lu is assigned to the two end regions of the joint ifies a range of τu in Eq. (7d) according to reinforcing bar
interfaces. Thus, the length of the bearing bond strength at strain and bond condition.
the center of the joint is determined as let = hc – 2lu.
In the ultimate case of “complete bond failure” shown in 0.05 f c′ ≤ τ u ≤ 0.4 f c′ (in MPa) (7a)
Fig. 3(c), as the yield penetration reaches the center of the
joint, the friction bond length lu is assigned to the total joint
τ u = 0.06 f c′ (in MPa) (7b)
length considering symmetric cyclic loading.
From Fig. 3, assuming a tensile strain of reinforcing bar at
the joint interface, which is the maximum bar strain within τ u ≤ 0.3 3 f c′ 2 (in MPa) (7c)
the joint, the bearing bond lengths (let and lec) and the friction
bond length (lu) can be calculated using the bond strengths 0.075 f c′ ≤ τ u ≤ 1.0 f c′ (in MPa) (7d)
τet, τec, τu, and Eq. (2). The bar strain and stress distributions
can be determined using the bond lengths. The elongation As presented in Eq. (7), the existing friction bond strength
eb of the beam reinforcing bar, which develops at the joint models show large differences. Further, as mentioned, the
interface due to bar-slip, is then calculated by integrating the loading and boundary conditions of the concrete block tests
bar strain over the length of the joint. differ from those of the actual beam-column joints. Thus, in
the present study, the friction bond strength was evaluated

h
eb = ∫0 c εdx (ε ≥ 0) (3) by analyzing the existing test results of beam-column joints.
In the existing beam-column connection specimens that
By integrating the bond stresses, the overall bond strength showed significant bond slip, the lateral load-displacement
of the joint and the forces of the beam reinforcing bars can relationships commonly showed severe pinching after large
be calculated. Detailed sample calculations are presented in inelastic deformation; at the origin of the load-displacement
Appendix C. relationship, the unloading/reloading stiffness decreased to
The proposed bond-slip model was developed based on zero, maintaining a small uniform lateral load P0 (refer to
the existing methods proposed by Hong et al.8 and Lowes Fig. 4(a)). In such specimens, after large inelastic deforma-
and Altoontash.20 However, in the proposed method, the tion under cyclic loading, complete bearing bond failure
effect of cyclic loading was more precisely addressed; the occurs, and the bond stress is decreased to a uniform friction
friction bond strength τu was determined from the test results bond strength (refer to Fig. 3(c)). Further, at the origin of the
of the actual beam-column joint specimens, rather than from load-displacement relationship, due to the residual plastic
the concrete block tests (refer to the “BOND STRENGTH” strains of the beam flexural bars, the internal load transfer
section). Further, as shown in Fig. 3, the symmetric bond between the column and the beams is developed mainly by
damage at the joint interfaces were considered to address the the bond of the beam flexural bars, without contact with
effect of reversed cyclic loading. the concrete at the interfaces between the joint and the two
beams (see the stress-strain relationship of beam reinforcing

830 ACI Structural Journal/November-December 2015

Fig. 4—Evaluation of residual bond stress τu from existing test results of beam-column connections.
bars in Fig. 4(b)). The lateral load-carrying capacity P0 at 2 P0 H
the origin is therefore determined from the friction bond τu = (10)
nb

strength τu. Conversely, in the present study, the friction hs  ∑ πdbi hc  (hc /L1 + 2)
 i =1 
bond strength τu of the specimens was evaluated from the
test results of P0. Figures 5(a) to 5(c) show the lateral loads P0 (at the origin)
As shown in Fig. 3(c), after complete bearing bond failure, of the three representative existing test specimens; the values
the bond stress distribution can be simplified as the uniform of P0 were used to evaluate the friction bond strength. The
friction bond strength τu. Thus, as shown in Fig. 4(b), the P0 values used for the calculation of τu were the average
overall bond strength of the joint is defined as Bj = τu(∑πdbihc). of the positive and negative loadings at the last load cycle
In the existing test results, the overall bond strength can be esti- (that is, the second load cycle or third load cycle; refer to
mated from the sum of the compression and tension forces of Fig. 5(a) to 5(c)). Figures 5(d) to 5(l) summarize the friction
the beam reinforcing bars at the left and right joint inter- bond strengths τu that were calculated from 67 existing test
faces, which are developed by the flexural moments Ml and results.6,B1-B24 All the existing specimens showed significant
Mr of the two beams. Thus, after complete bond failure, the pinching in the load-displacement relationships (Fig. 5(a) to
beam flexural moments at the origin can be expressed with 5(c)), and did not satisfy the requirement of ACI 374.1-0523:
the friction bond strength of the joint (refer to Appendix D the secant stiffness connecting the –0.35% and +0.35% drift
for experimental evidence). ratios was less than 0.05 times the initial stiffness. In the test
specimens, the ranges of design parameters were fc′ = 20.8
 nb  to 84.4 MPa (3.0 to 12.2 ksi), fy = 300 to 858 MPa (43.5 to
M l + M r = (Tb1 + Tb 2 ) hs  (Cb1 + Tb 2 ) hs = τ u  ∑ πdbi hc  hs (8)
 i =1  124.3 ksi), db = 9.5 to 31.8 mm (0.37 to 1.25 in.), and hc =
240 to 550 mm (9.45 to 21.65 in.). The design parameters of
the specimens are presented in Appendix B.
where nb is the number of the bottom flexural bars in a beam
According to Eligehausen et al.,14 the bar bond stress-slip
cross section; dbi is bar diameter; and hs is distance between
relationships are affected by bar diameter, concrete strength,
the top and bottom flexural bars. In beams, the number of
clear spacing between the bars, restraining bars, trans-
top bars is generally greater than that of bottom bars. In this
verse pressure, loading rate, and position of the bars during
case, due to the force-equilibrium in the cross section, the
concrete casting. In actual beam-column joints, the clear bar
bottom bars experience greater inelastic deformation than
spacing corresponds to the spacing of the beam longitudinal
the top bars. Thus, the bottom bars are subjected to severe
bars, the restraining bars correspond to the column longitu-
bond degradation and large bond slip. For this reason, Eq. (8)
dinal bars and joint hoop bars, and the transverse pressure
was defined by the bond strength of the bottom bars.
corresponds to the axial force of the column and the strength
In the test specimens, the lateral load P0 at the origin can
of the joint hoops.
be expressed with the beam flexural moments Ml and Mr by
Considering the bond parameters, as shown in Fig. 5(d) to
using the lateral force-equilibrium for the support reactions
5(l), variations of the friction bond strength were expressed
(refer to Fig. 4(c)).
according to the design parameters that are directly or indi-
rectly related to the bond parameters: concrete compressive
( M l + M r )(hc /L1 + 2)
P0 = (9) strength (√fc′), column flexural strength-to-beam flexural
2H strength ratio (∑Mcol/∑Mbeam), joint shear strength ratio
where L1 is shear span of the beams, and H is net column (Vn /Vu), average axial compressive stress of the column (σc),
height. It is assumed that the beam-column connection has a bar spacing ratio of the beam flexural bars (s/db), tensile
symmetric configuration. strength of the joint hoops parallel to the beam flexural bars
Substituting Eq. (8) into Eq. (9), the friction bond strength (Ahpfyh), tensile strength of the joint hoops orthogonal to the
τu can be evaluated as follows. beam flexural bars (Ahtfyh), and the sum of the two tensile
strengths of the joint hoops (Ahpfyh +Ahtfyh).

ACI Structural Journal/November-December 2015 831

Fig. 5—Variation of friction bond strength according to design parameters. (Note: 1 kN = 0.255 kip; 1 MPa = 0.145 ksi.)
As shown in Fig. 5(d), the friction bond strengths On the basis of the results, the friction bond strength τu
correlated with the concrete compressive strength √fc′, but was defined as the function of the ratio of the hoop tensile
showed large deviations. All the test results belong to the strength to the joint shear force (refer to Fig. 5(k)).
range of Eq. 7(a) proposed by Lowes and Altoontash.20 As
shown in Fig. 5(e) to 5(h), the test results of τu/√fc′ did not (τ u f c′) = 0.16 + 0.11 ( Ah f yh ) Vu (in MPa) (11a)
show good correlations with the design parameters of the
column flexural strength-to-beam flexural strength ratio, joint
shear strength ratio, compressive stress of column, and bar (τ u f c′) = 1.93 + 1.32 ( Ah f yh ) Vu (in psi) (11b)
spacing ratio. On the other hand, as shown in Fig. 5(i) to 5(k),
the friction bond strengths increased in proportion to the ratio where 0 ≤ (Ahfyh)/Vu ≤ 1.5; Ah (= Ahp +Aht) = sum of the area
of the hoop tensile strength to the joint shear force. Particu- of the hoop bars parallel and orthogonal to the beam flexural
larly, as shown in Fig. 5(k), the friction bond strengths showed bars; fyh is yield strength of the hoop bars; Vu is horizontal
good correlation with the ratio of the sum of the hoop tensile shear force applied to the joint (in the case of the test spec-
strength to the joint shear force ((Ahpfyh +Ahtfyh)/Vu). This result imens, Vu =∑Asfy – Pu, where ∑As is the sum of the top and
indicates that the joint shear force decreases the friction bond bottom flexural bars of a beam; fy is the yield strength of
strength, developing diagonal cracking. On the other hand, the the beam flexural bar; and Pu is maximum shear force of
joint hoop strength increases the friction bond strength by the column). In Eq. (11), the range of (Ahfyh)/Vu is limited
restraining diagonal cracking and providing lateral pressure. to 0 ≤ (Ahfyh)/Vu ≤ 1.5, considering the range of the design
As shown in Fig. 5(l), the average friction bond strength τu,avg. parameters of the test specimens. The variable τu in Eq. (11)
  nt nb
  indicates the friction bond strength for the beam-column
( = 4 P0 H  hs  ∑ πdbi hc + ∑ πdbi hc  ( hc L1 + 2) ), using the

 i =1 i =1   connections subjected to two or three repeated cyclic load-
mean bond strengths of the top and bottom bars in Eq. (10), ings at each displacement level.
were compared. The results showed that the relationship
between τu,avg. and (Ahpfyh +Ahtfyh)/Vu did not significantly VERIFICATION OF PROPOSED BOND MODEL
differ from the result for τu based on the bond strength of the For verification, the proposed bond model was applied
bottom bars. to the concrete block specimens tested by Viwathanatepa
et al.17 Although the loading and boundary conditions and

832 ACI Structural Journal/November-December 2015

Fig. 6—Comparison between cyclic test and bond stress model. (Note: 1 mm = 0.0394 in.; 1 kN = 0.255 kip; 1 MPa = 0.145 ksi.)
details of the concrete block specimens differ from the actual
conditions and details of beam-column joints, bond strength
degradation due to cyclic loading was evaluated using the
conditions of the concrete block specimens. Figure 6 shows
the loading and boundary conditions of the test specimens.
In this test, to evaluate the bond behavior between the rein-
forcing bar and concrete under cyclic loading, two identical
loads Fb in the same direction were applied to the reinforcing
bar ends. Thus, the load Fb was equivalent to half of the
overall bond strength Bj: Fb = 0.5Bj. The bar elongation eb
was measured at the bar ends. The design parameters were Fig. 7—Bar strain distribution in No. 14.16 (Note: 1 mm =
bar diameter db (= 19.1 to 31.8 mm [0.75 to 1.25 in]) and 0.0394 in.)
development length hc (= 381 to 635 mm [15.0 to 25.0 in]).
Figures 6(a) to 6(e) show the load-elongation (Fb-eb) rela- cyclic loading, bar bond strength is degraded as the load
tionships of the No.4, No.8, No.11, No.14, and No.17 spec- cycle increases. Thus, for safe design, the proposed fric-
imens, respectively. The Fb-eb relationship was measured at tion bond strength τu in Eq.  (11) was determined from the
one end of the concrete block, and the positive and negative test results corresponding to the second or third load cycle
loads indicate the tension and compression forces Fb applied at each displacement level. To predict the envelope curve
to the reinforcing bar ends, respectively. and bond strength for the first load cycle, the friction bond
Using the proposed bond model, the Fb-eb relationships strength needs to be increased. Figure 6 shows that when the
of the specimens were predicted, and the results were friction bond strength is defined as two times the proposed
compared with the test results shown in Fig. 6(a) to 6(e). bond strength (that is, 2τu), the predictions agree with the
The thick dotted lines indicate the predictions. The proposed envelope curve and the bond strength for the first load cycle
bond strengths τet, τec, and τu in Eq. (4), (5), and (11) were (refer to “prediction of proposed model for the 1st load
used for the predictions. In Eq. (11), τu was calculated as cycle” in Fig. 6).
0.325√fc′. Also, an existing bond model proposed by Lowes On the other hand, the bond model proposed by Lowes
and Altoontash20 was compared in the figures. In the existing and Altoontash20 using τu = 0.23√fc′ overestimated the post-
model,20 τu = 0.23√fc′ was used for the friction bond strength peak bond strengths. Particularly, for Specimens No. 4 and
(that is, residual bond strength). The positive and negative No. 8 with insufficient development lengths as shown in
elongations eb were calculated by integrating the tensile Fig. 6(a) and (b), the predictions overestimated both the peak
and compressive strains of the reinforcing bar, respec- strength and the post-peak strength. The primary reason for
tively (refer to Eq. (3)). The elastic modulus and hardening the overestimation is that the Lowes model did not address
modulus of the steel bars were Es = 200 GPa (28,986 ksi) the symmetric bond damage at the left and right ends of the
and Esh = 0.02Es, respectively. Detailed calculations for the concrete block subjected to cyclic loading (refer to the bond
predictions are given in Appendix C. stress and bond length distributions shown in Fig. 3).
In the bond tests, three cyclic loadings were applied at Figure 7 compares the bar strain distributions for Spec-
each displacement. In Fig. 6, the proposed model underes- imen No. 14. The solid line, dotted line, and dot-and-dash
timated the bond strengths for the first cyclic loading, but line indicate the test result, proposed model, and Lowes’s
agreed well with the envelope curves and bond strengths model,20 respectively. Figure 7(a) and 7(b) show the bar
for the second and third cyclic loadings. Under repeated strain distributions corresponding to initial yielding and a

ACI Structural Journal/November-December 2015 833

Fig. 8—Bar strain distributions of existing beam-column joint specimens after reinforcing bar yielding. (Note: 1 mm = 0.0394 in.;
1 MPa = 0.145 ksi.)

Fig. 9—Elongation of beam reinforcing bars in existing beam-column joint specimens. (Note: 1 mm = 0.0394 in.; 1 MPa = 0.145 ksi.)
large inelastic strain (=0.0105), respectively. In both cases, the beam reinforcing bars, compressive stresses developed
the majority of the bond damage occurred at the right end, even under tensile strain.25 In the proposed model, the effect
subjected to tension yielding in the loading step. Neverthe- of such tensile residual deformation was not considered.
less, the compressive strain at the left end cannot be negli- The bar tensile strains predicted by the Lowes’s model were
gible, and increased as the cyclic damage increased. The greater than those predicted by the proposed model, partic-
proposed model predicted both the tensile and compressive ularly at the right end, which is subjected to tensile strain.
strains at the right and left ends. On the other hand, in the This is because in the Lowes’s model, the bar strains inside
Lowes’s model,20 the compressive strain was zero because the joint increased due to underestimation of the friction
the cyclic damage was not considered. bond strength.
Figure 8 shows the strain distributions of the beam bottom Figure 9 shows the elongation eb of a beam flexural
bars in the beam-column joints tested by Hwang et al.,B1 Lee bar that was measured at the joint interface of specimens
et al.,6 Dai and Park,24 Asou et al.,B22 and Kawai et al.B16 The tested by Kawai et al.B16 and Xian et al.B13 In this figure, the
proposed model was used to predict the strain distributions vertical and horizontal axes indicate the elongation eb and
of the specimens. The predictions were compared with the tensile strain εt of the reinforcing bar at the joint interface,
test results and Lowes’s model.20 The vertical and horizontal respectively. The elongations were predicted by integrating
axes indicate the bar strain and the distance from the joint the bar strain (Eq. (3)). In Fig. 9, the predictions of both
interface, respectively. the proposed model and the Lowes’s model agreed with the
Generally, the bar tensile strains predicted by the proposed test results.
model agreed with those of the test specimens. However, at
the left end of the joints, the proposed bond model showed BOND PERFORMANCE REQUIREMENT
compressive strains, while the test results showed tensile To secure the seismic resistance of beam-column joints,
strains (although the tensile strains were not significant). excessive bar bond slip should be restrained. If complete
This is because, due to the tensile residual deformations of bond failure occurs in a beam-column joint, significant

834 ACI Structural Journal/November-December 2015

pinching occurs in the load-displacement relationship, which
degrades the energy dissipation capacity (refer to Fig. 5(a)
and (b)). Furthermore, as presented in Appendix A, the bond
requirements in the current design codes significantly differ.
Thus, in the present study, the requirements of bond resis-
tance were studied to prevent bond failure.
The following three performance criteria can be used to
derive each bar bond requirement:
1. The bar bond in the joint resists both the incoming and
outgoing bar forces (Bj = Ts + Cs).
2. The bar bond in the joint resists only the tensile bar
force (Bj = Ts).
3) The complete bond failure is permitted (Bj < Ts).
When the first design criterion is used, a very large
column (or joint) depth may be required. Thus, in the
present study, Bj ≥ Ts was used for the bar bond requirement;
in other words, the tensile yield force of the beam flexural
bars should be resisted by the bond strength of the beam-
column joint (refer to Fig. 3(a)). Otherwise, (Bj < Ts); during
unloading/reloading, rigid body slip of beam flexural bars
occurs through the beam-column joint, which significantly
decreases the energy dissipation capacity of the beam-
column connection. Furthermore, the tensile force of the
beam reinforcing bars should be anchored to the other beam
at the opposite side. In this case, the compression zone depth
is increased to satisfy the force-equilibrium for the increased
compression force, which decreases the flexural moment
capacity of the beam by decreasing the effective depth of the
flexural moment. Fig. 10—Comparison of design requirements of hc/db for
To satisfy the criterion of Bj = Ts, as shown in Fig. 3(b), interior beam-column joint. (Note: 1 MPa = 0.145 ksi.)
the yield strength of the reinforcing bar should be resisted
In Eq. (13), the requirement of hc/db is expressed as the func-
by the bearing bond strength (that is, bond strength before
tion of reinforcing bar yield strength fy, concrete strength fc′,
bar yielding), while the strain-hardening strength increase
curvature ductility demand of the beam μϕ = ϕt/ϕy, tension
caused by repeated cyclic loading after bar yielding should
zone depth ratio of the beam section (d – c)/hb, and the fric-
be resisted by the friction bond strength. Thus, the bearing
tion bond strength τu = [0.16 + 0.11(Ahfyh)/Vu]√fc′ (0.16√fc′ ≤
bond length and friction bond length can be defined as let =
τu ≤ 0.33√fc′).
εy(Esdb)/(4τet) and lu = (εt – εy)(Eshdb)/(4τu), respectively. The
In Eq. (13), as the ductility demand of a beam increases,
sum of the bearing bond length let and friction bond length lu
the bond requirement increases. Thus, the proposed equation
should be less than the joint length hc.
can be used for the performance-based design of beam-column
connections subjected to cyclic loading. Basically, when the
Es d b E d
hc ≥ let + 2lu = ε y + 2(ε t − ε y ) sh b (12) areas of the top and bottom flexural bars differ in a beam cross
4τ et 4τ u section, Eq. (13) should be satisfied for both top and bottom
bars. However, generally under cyclic loading, large inelastic
where εt is the maximum tensile strain of the beam flex- deformation occurs in the bottom bars with less area. Thus,
ural bar at the joint interface. Substituting τet in Eq. (4) Eq. (13) is not necessarily required for the top bars.
and τu in Eq.  (11), Esh = 0.02Es (average value of 0.005Es In Fig. 10, the requirements of hc/db specified in Eq. (13),
to 0.04Es),19,26 yield curvature of the beam cross section ACI 318-11,10 ACI 352R-02,11 NZS 3101:2006,12 and Euro-
ϕy = 1.7εy/hb,27 and curvature of the beam cross section code 813 were compared according to concrete strength fc′,
ϕt = εt/(d – c) (hb is beam depth, d is effective beam depth, reinforcing bar yield strength fy, and curvature ductility μϕ.
and c is neutral axis depth) into Eq. (12), the requirement Reinforcing bar yield strength fy = 400 MPa (58.0 ksi) was
of the column depth-to-bar diameter ratio (hc/db) is defined used for Fig. 10(a), and an arbitrary concrete compressive
as follows strength fc′ = 40 MPa (5.8 ksi) was used in Fig. 10(b). In
Fig. 10(c), fy = 400 MPa (58.0 ksi) and fc′ = 40 MPa (5.8 ksi)
 1  were used. The requirements of the current design codes are
hc 0.01
≥ fy  + (ε t /ε y − 1)  presented in Appendix A. For Eq. (13), (d – c)/hb = 0.7 and
db  7.2 f c′ τ u  τu = 0.243√fc′ (which is the mean value of the range of τu =
(13) 0.16√fc′ to 0.33√fc′) were used.
 0.14 1.7µ φ (d − c) / hb − 1 
 fy  + 
 f c′ 100τ u 

ACI Structural Journal/November-December 2015 835

 As shown in Fig. 10(a), ACI 318-1110 and ACI 352R-0211 National University. His research interests include inelastic analysis and
the seismic design of reinforced concrete structures.
have the least hc/db requirements, and the other requirements
decrease as the concrete strength increases. As shown in ACI member Hong-Gun Park is a Professor in the Department of Architec-
Fig. 10(b), the requirements of hc/db increase with the rein- ture & Architectural Engineering at Seoul National University. He received
his BE and MS in architectural engineering from Seoul National University
forcing bar yield strength, except for ACI 318-11.10 Figure and PhD in civil engineering from the University of Texas at Austin, Austin,
10(c) shows that the requirement of Eq. (13) increases with TX. His research interests include inelastic analysis and the seismic design
the curvature ductility demand of beams. The requirements of reinforced concrete structures.
of Eq. (13) are close to those of NZS 3101:200612 and
Eurocode 813 when the ductility demands are μϕ = 6 and 9, NOTATION
Ahp = area of joint hoop bars parallel to beam flexural bars, mm2 (in.2)
respectively. Aht = area of joint hoop bars orthogonal to beam flexural bars, mm2
(in.2)
SUMMARY AND CONCLUSIONS As = area of reinforcing bar, mm2 (in.2)
Bj = bond strength of beam reinforcing bar in joint, kN (kip)
In the present study, a simplified bar bond strength model Cb = compression force of beam reinforcing bars developed by joint
and bond-slip relationship for interior beam-column joints bond strength, kN (kip)
subjected to cyclic loading were studied. The results of this c = neutral axis depth in beam section, mm (in.)
d = effective beam depth, mm (in.)
study are summarized as follows: db = reinforcing bar diameter, mm (in.)
1. To address the bond slip of the beam flexural bars in Es = elastic stiffness modulus of reinforcing bar, MPa (ksi)
beam-column joints, bond stress distribution and bar strain Esh = hardening stiffness modulus of reinforcing bar, MPa (ksi)
eb = reinforcing bar elongation, mm (in.)
distribution were proposed, considering the yield penetra- fc′ = concrete compressive strength, MPa (psi)
tion of the reinforcing bars. Considering the symmetric fy = yield strength of reinforcing bar, MPa (psi)
cyclic loading effect, bond failure zones were used at both fyh = yield strength of hoop bar, MPa (psi)
H = net column height, mm (in.)
ends of the joint. The bond strength in the bond failure zone hb = beam depth, mm (in.)
was defined as the friction bond strength (that is, residual hc = column depth, mm (in.)
bond strength), while in the zone inside the joint, the bond hs = distance between top and bottom flexural bars in beam section,
mm (in.)
strength was defined as the bearing bond strength. L1 = shear span of beams (mm, in.)
2. To evaluate the friction bond strength after complete le = bearing bond length of a tension bar, mm (in.)
bond failure, existing test results of beam-column joints lec = bearing bond length of a compression bar, mm (in.)
let = bearing bond length of a tension bar, mm (in.)
showing complete bond failure were analyzed. The proposed lu = friction bond length, mm (in.)
friction bond strength was defined as the function of joint Mcol = flexural moment of column, kN·m (kip·in.)
hoop bar strength, joint shear demand force, and concrete Ml, Mr = beam flexural moments, kN·m (kip·in.)
nb = number of bottom flexural bars in beam cross section
compressive strength. The joint hoop strength increases nt = number of top flexural bars in beam cross section
the friction bond strength by providing lateral pressure and P0 = load-carrying capacity at origin of lateral load-displacement
restraining diagonal cracking. relationship of beam-column joint, kN (kip)
Pu = maximum column shear force, kN (kip)
3. To verify the proposed model, the bond-slip relation- Tb = tension force of beam reinforcing bars developed by joint bond
ship and bar strain distribution predicted by the proposed strength, kN (kip)
model were compared with existing test results of concrete Ts = tensile force of beam reinforcing bar, kN (kip)
Vn = shear capacity of beam-column joint, kN (kip)
blocks and beam-column joints. The predictions agreed with Vu = shear demand of beam-column joint, kN (kip)
the bond slip, bond stress degradation, strain distribution, s = spacing of beam flexural bars, mm (in.)
and bar elongation of the test specimens. εt = reinforcing bar strain, mm/mm (in./in.)
εy = yield strain of reinforcing bar, mm/mm (in./in.)
4. On the basis of the proposed bond stress model, the ϕt = curvature of beam cross section, 1/mm (1/in.)
requirement of the joint depth-to-diameter of the beam flex- ϕy = yield curvature, 1/mm (1/in.)
ural bar ratio hc/db was proposed to prevent bond failure in μϕ = curvature ductility
σc = average axial compressive stress of column, MPa (ksi)
beam-column joints. Particularly, the proposed model was τec = bearing bond strength for compression bar, MPa (ksi)
defined as the function of the flexural ductility demand of τet = bearing bond strength for tension bar, MPa (ksi)
beams as well as the concrete strength and reinforcing bar τu = friction bond strength, MPa (ksi)
yield strength so that the proposed model can be used for the
performance-based design of beam-column joints subjected ACKNOWLEDGMENTS
This research was financially supported by the Basic Science
to cyclic loading. The proposed requirements are close Research Program through the National Research Foundation of Korea
to those in NZS 3101:200612 and Eurocode 8,13 when the (NRF), funded by the Ministry of Education, Science, and Technology
ductility demands are μϕ = 6 and 9, respectively. (2012R1A1A1003282). It was also supported by the Integrated Research
Institute of Construction and Environmental Engineering, Seoul National
University Research Program funded by the Ministry of Education &
AUTHOR BIOS Human Resources Development. The authors are grateful to the authorities
Hyeon-Jong Hwang is an Assistant Professor in the College of Civil Engi- for their support.
neering at Hunan University, China. He received his BE, MS, and PhD
in architectural engineering from Seoul National University, Seoul, South
Korea. REFERENCES
1. Kitayama, K.; Otani, S.; and Aoyama, H., “Earthquake Resistant
Tae-Sung Eom is an Assistant Professor in the Department of Architec- Design Criteria for Reinforced Concrete Interior Beam-Column Joints,”
tural Engineering at Dankook University, Gyeonggi-do, South Korea. He Pacific Conference on Earthquake Engineering, New Zealand, V. 1, 1987,
received his BE, MS, and PhD in architectural engineering from Seoul pp. 315-326.

836 ACI Structural Journal/November-December 2015

 2. Leon, R. T., “Interior Joints with Variable Anchorage Lengths,” Excitations,” Earthquake Engineering Research Council Report No. 82/23,
Journal of Structural Engineering, ASCE, V. 115, No. 9, 1989, pp. 2261- University of California, Berkeley, Berkeley, CA, 1982, 103 pp.
2275. doi: 10.1061/(ASCE)0733-9445(1989)115:9(2261) 16. Elmorsi, M.; Kianoush, M. R.; and Tso, W. K., “Modeling Bond-
3. Hakuto, S.; Park, R.; and Tanaka, H., “Effect of Deterioration of Bond Slip Deformations in Reinforced Concrete Beam-Column Joints,” Cana-
of Beam Bars Passing through Interior Beam-Column Joints of Flexural dian Journal of Civil Engineering, V. 27, No. 3, 2000, pp. 490-505. doi:
Strength and Ductility,” ACI Structural Journal, V. 96, No. 5, Sept.-Oct. 10.1139/l99-085
1999, pp. 858-864. 17. Viwathanatepa, S.; Popov, E. P.; and Bertero, V. V., “Effects of
4. Meinheit, D. F., and Jirsa, J. O., “The Shear Strength of Reinforced Generalized Loadings on Bond of Reinforcing Bars Embedded in Confined
Concrete Beam-Column Joints,” Report No. 77-1, Department of Civil Concrete Blocks,” Earthquake Engineering Research Council Report
Engineering, Structures Research Laboratory, University of Texas at No. 79/22, University of California, Berkeley, Berkeley, CA, 1979, 304 pp.
Austin, Austin, TX, Jan. 1977, 230 pp. 18. Alsiwat, J. M., and Saatcioglu, M., “Reinforcement Anchorage
5. Bonacci, J., and Pantazopoulou, S., “Parametric Investigation of Joint Slip under Monotonic Loading,” Journal of Structural Engi-
Mechanics,” ACI Structural Journal, V. 90, No. 1, Jan.-Feb. 1993, pp. 61-71. neering, ASCE, V. 118, No. 9, 1992, pp. 2421-2438. doi: 10.1061/
6. Lee, J. Y.; Kim, J. Y.; and Oh, G. J., “Strength Deterioration of Rein- (ASCE)0733-9445(1992)118:9(2421)
forced Concrete Beam-Column Joints Subjected to Cyclic Loading,” 19. Marti, P.; Alvarez, M.; Kaufmann, W.; and Sigrist, V., “Tension
Engineering Structures, V. 31, No. 9, 2009, pp. 2070-2085. doi: 10.1016/j. Chord Model for Structural Concrete,” Structural Engineering Interna-
engstruct.2009.03.009 tional, V. 8, No. 4, 1998, pp. 287-298. doi: 10.2749/101686698780488875
7. Lee, H. J., and Lin, Y. R., “Preliminary Design Recommendations 20. Lowes, L. N., and Altoontash, A., “Modeling Reinforced Concrete
for RC Beam-Column Joints with High-Strength Reinforcement,” 13th Beam-Column Joints Subjected to Cyclic Loading,” Journal of Structural
SEEBUS, Seoul, South Korea, 2011, pp. 1-10. Engineering, ASCE, V. 129, No. 12, 2003, pp. 1686-1697. doi: 10.1061/
8. Hong, S. G.; Lee, S. G.; and Kang, T. H. K., “Deformation-Based Strut- (ASCE)0733-9445(2003)129:12(1686)
and-Tie Model for Interior Joints of Frames Subject to Load Reversal,” ACI 21. fib, “FIB Model Code for Concrete Structures 2010,” fib Journal
Structural Journal, V. 108, No. 4, July-Aug. 2011, pp. 423-433. Structural Concrete, Ernst & Sohn, Berlin, Germany, 2010, 402 pp.
9. Soleimani, D.; Popov, E. P.; and Bertero, V. V., “Hysteretic Behavior 22. Shima, H.; Chou, L. L.; and Okamura, H., “Bond Characteristics in
of Reinforced Concrete Beam-Column Subassemblages,” ACI Journal Post-Yield Range of Deformed Bars,” Japan Society of Civil Engineers,
Proceedings, V. 76, No. 11, Nov. 1979, pp. 1179-1196. V. 10, No. 378, 1987, pp. 213-220.
10. ACI Committee 318, “Building Code Requirements for Structural 23. ACI Committee 374, “Acceptance Criteria for Moment Frames
Concrete (ACI 318-11) and Commentary,” American Concrete Institute, Based on Structural Testing and Commentary (ACI 374.1-05),” American
Farmington Hills, MI, 2011, 503 pp. Concrete Institute, Farmington Hills, MI, 2005, 9 pp.
11. Join ACI-ASCE Committee 352, “Recommendations for Design of 24. Dai, R., and Park, R., “A Comparison of the Behaviour of Rein-
Beam-Column Connections in Monolithic Reinforced Concrete Structures forced Concrete Beam-Column Joints Designed for Ductility and Limited
(ACI 352R-02),” American Concrete Institute, Farmington Hills, MI, 2002, Ductility,” Research Report No. 87-4, Department of Civil Engineering,
38 pp. University of Canterbury, Christchurch, New Zealand, 1987, 65 pp.
12. NZS 3101:2006, “The Design of Concrete Structures,” Standards 25. Eom, T.-S., and Park, H., “Evaluation of Energy Dissipation of
New Zealand, New Zealand, 2006, 698 pp. Slender Reinforced Concrete Members and Its Applications,” Engi-
13. British Standards, “Eurocode 8: Design of Structures for Earthquake neering Structures, V. 32, No. 9, 2010, pp. 2884-2893. doi: 10.1016/j.
Resistance (BS EN 1998-1:2004),” British Standards Institution, UK, 2004, engstruct.2010.05.007
229 pp. 26. Monti, G., and Nuti, C., “Nonlinear Cyclic Behavior of Rein-
14. Eligenhausen, R.; Popov, E. P.; and Bertero, V. V., “Local Bond forcing Bars Including Buckling,” Journal of Structural Engi-
Stress-Slip Relationships of Deformed Bars under Generalized Exci- neering, ASCE, V. 118, No. 12, 1992, pp. 3268-3284. doi: 10.1061/
tations,” Earthquake Engineering Research Council Report No. 83/23, (ASCE)0733-9445(1992)118:12(3268)
University of California, Berkeley, Berkeley, CA, 1983, 162 pp. 27. Priestley, M. J. N., “Performance Based Seismic Design,” Proceed-
15. Ciampi, V.; Eligehausen, R.; Bertero, V. V.; and Popov, E. P., “Analyt- ings of the 12th WCEE, Auckland, New Zealand, 2000, pp. 1-22.
ical Model for Concrete Anchorages of Reinforcing Bars under Generalized

ACI Structural Journal/November-December 2015 837

 NOTES:

838 ACI Structural Journal/November-December 2015

 APPENDIX A: BOND REQUIREMENTS OF CURRENT DESIGN CODES

For interior beam-column joints, ACI 318-11,10 ACI 352R-02,11 NZS 3101:2006,12 and

Eurocode 813 specify the following requirements for column depth-to-bar diameter hc /db.

hc
 20 in ACI 318-11 (A1)
db

hc f
 20 y  20 in ACI 352R-02 (A2)
db 420

hc 1.25 f y
 in NZS 3101:2006 (A3)
db 3.3 f  d  f c

hc  f 1  0.75k D    max
 Rd y in Eurocode 8 (A4)
db 7.5 f ctm 1  0.8vd

In NZS 3101:2006,12 αf = coefficient related to the direction of the beam re-bars; (= 0.85 ~

1.0); αd = coefficient related to the ductility of the plastic hinge of beams (= 1.0 ~ 1.2); γ =

coefficient related to inter-storey drift (= 1.53-0.29δc ≤ 1.0); and δc = inter-storey drift ratio

expressed as a percentage. In Eurocode 8,13 γRd = uncertainty coefficient for the design value of
(2/3)
resistance (= 1.0 ~ 1.2); fctm = tensile strength of concrete (= 0.3fc′ ); kD = coefficient

addressing the ductility (= 0.67 ~ 1.0); ρ′ = compression bar ratio in the beams; ρmax = maximum

tension bar ratio allowed in the beams; and vd = axial load ratio of the column. For comparison,

αf = 1.0, αd = 1.0, γ = 1.0, γRd = 1.2, kD = 1.0, ρ′/ρmax = 1.0, and vd = 0 were used in the cases

shown in Fig. 10.

APPENDIX B: TEST PARAMETERS OF EXISTING BEAM-COLUMN JOINT

Table B1 lists the principal test parameters of the 67 existing beam-column joints6,B1-B24 that
 were used to evaluate friction bond strength τu (Eq. (10)).

Table B1– Test parameters of existing test specimens [1 in = 25.4 mm, 1 ksi = 6.90 MPa]

Concre
Later Bond
Top re-bar Bottom re-bar Joint te Axial
Geometric propertiesa al stren
of beamb of beamb hoopc strengt load
load gth
Specimens h

Nc / τu
fc' P0
L H hb bb hc bc As fy db n As fy db n hs Ah fyh (fc' (MPa
(MPa) (kN)
bchc) )

250 156 40 30 35 35 51 15. 15. 198 51

0 0 0 0 0 0 0 9 9 8 0
119 119
BJ1 250 156 40 30 35 35 51 15. 6 510 15. 6 270 142 51 40.0 0.00 37.4 1.77
4 4
BJ2 0 0 0 0 0 0 0 9 5 510 9 5 270 0 0 40.0 0.00 25.3 1.43
Lee et al.6) 995 995
BJ3 250 156 40 30 35 35 51 15. 4 510 15. 4 270 142 51 40.0 0.00 26.1 1.85
796 796
B1 0 0 0 0 0 0 0 9 3 510 9 3 270 0 0 40.0 0.00 18.2 1.72
597 597
250 156 40 30 35 35 51 15. 15. 142 51

0 0 0 0 0 0 0 9 9 0 0

Hwang et 576 210 50 35 46 50 204 45 25. 102 25. 309 49

S1 4 452 2 394 38.3 0.00 28.8 1.91
al.B1) 0 0 0 0 0 0 0 2 4 0 4 6 6

S1 476 210 50 35 55 50 311 46 25. 6 152 520 22. 4 387 309 49 32.0 0.00 38.8 1.23
Hwang et
S2 0 0 0 0 0 0 0 5 4 5 4 710 2 3 396 6 6 32.0 0.00 36.2 1.48
B2)
al.
S3 476 210 50 35 55 50 193 71 22. 5 116 710 22. 3 396 309 49 32.0 0.00 33.3 1.70
 0 0 0 0 0 0 5 0 2 1 2 6 6

476 210 50 35 45 50 193 71 22. 116 22. 309 49

0 0 0 0 0 0 5 0 2 1 2 6 6

Teraoka et NO4 300 200 40 30 40 40 170 38 19. 170 19. 157 34

6 382 6 310 54.0 0.20 38.6 1.51
al.B3) 7 0 0 0 0 0 0 1 2 1 1 1 0 7

Susanto 400 262 40 20 30 40 51 15. 15. 191 44

S1 995 5 597 510 3 326 33.0 0.00 10.4 1.23
and HuaB4) 0 6 0 0 0 0 0 9 9 7 0

249 224 41 27 36 36 154 33 22. 19. 174 35

113
6 8 9 9 2 2 8 1 2 1 5 2
Durrani X1 4 6 345 4 350 34.3 0.06 25.9 1.65
249 224 41 27 36 36 154 33 22. 19. 261 35
and X2 4 113 345 4 350 33.7 0.06 27.7 1.76
6 8 9 9 2 2 8 1 2 1 8 2
WightB5) X3 3 6 345 3 350 31.0 0.05 26.3 2.23
249 224 41 27 36 36 116 33 22. 19. 174 35
855
6 8 9 9 2 2 3 1 2 1 5 2

NO4 300 200 40 30 40 40 85 19. 19. 142 34

568 568
Hayashi et 6 0 0 0 0 0 0 8 1 2 858 1 2 310 0 7 49.1 0.20 14.5 1.70
170 170
al.B6) NO4 300 200 40 30 40 40 38 19. 6 383 19. 6 280 142 34 49.1 0.20 28.5 1.24
4 4
7 0 0 0 0 0 0 3 1 1 0 7

Sugano et 278 140 40 30 44 44 309 38 22. 309 22. 204 92

J4-0 8 386 8 260 30.4 0.33 90.4 1.67
al.B7) 0 0 0 0 0 0 6 6 2 6 2 8 3

BN1 280 150 47 30 45 45 232 50 22. 6 232 502 22. 6 365 203 81 79.0 0.07 94.1 1.72

Oda et BN2 0 0 5 0 0 0 2 2 2 8 2 502 2 8 350 2 7 79.0 0.07 131.0 1.88

al.B8) BN3 280 150 47 30 45 45 309 50 22. 5 309 535 22. 5 377 203 81 79.0 0.07 110.3 2.04

BN4 0 0 5 0 0 0 6 2 2 5 6 535 2 5 377 2 7 79.0 0.07 117.6 2.17

 BN5 280 150 47 30 45 45 255 53 25. 4 255 535 25. 4 395 203 81 51.3 0.09 84.1 1.85

0 0 5 0 0 0 0 5 4 0 4 2 7

280 150 47 30 45 45 255 53 25. 255 25. 203 81

0 0 5 0 0 0 0 5 4 0 4 2 7

280 150 47 30 45 45 204 53 25. 204 25. 203 81

0 0 5 0 0 0 0 5 4 0 4 2 7

300 175 35 15 30 30 37 12. 12. 30

0 0 0 0 0 0 1 7 7 7

B1 300 175 35 30 30 30 398 37 12. 3 398 371 12. 3 290 340 30 21.3 0.16 5.0 0.72

B2 0 0 0 0 0 0 398 1 7 3 398 371 7 3 290 340 7 20.8 0.16 5.0 0.72

Joh et al.B9)
LH 300 175 35 20 30 30 387 40 12. 3 387 404 12. 3 290 340 37 26.9 0.15 6.8 1.01

MH 0 0 0 0 0 0 387 4 7 3 387 404 7 3 290 678 7 28.1 0.14 8.2 1.21

300 175 35 20 30 30 40 12. 12. 37

0 0 0 0 0 0 4 7 7 7

270 147 30 20 30 30 106 40 13. 13. 36

0 0 0 0 0 0 1 1 0 0 362 8
J1 8 531 401 4 225 25.7 0.08 11.8 1.40
Otani et 270 147 30 20 30 30 106 40 13. 13. 724 36
J2 8 531 401 4 225 24.0 0.08 13.3 1.58
B10)
al. 0 0 0 0 0 0 1 1 0 0 168 8
J3 8 531 401 4 225 24.0 0.08 15.2 1.80
270 147 30 20 30 30 106 40 13. 13. 8 36

0 0 0 0 0 0 1 1 0 0 8

423 257 45 22 40 30 159 30 15. 159 15. 33

Wong et U1 8 300 8 220 896 32.2 0.00 16.9 1.10
8 5 7 9 6 5 2 0 9 2 9 9
al.B11) U2 8 300 8 220 896 28.1 0.00 16.4 1.00
260 257 45 22 40 30 159 30 15. 159 15. 33
 2 5 7 9 6 5 2 0 9 2 9 9

Peer1 365 213 50 40 45 40 157 42 22. 15.

787
4 8 4 8 6 7 6 1 3 2 4 504 9 4 384 0 0 31.8 0.11 17.8 0.96
B12)
Walker 133
Peer2 365 213 50 40 45 40 200 52 20. 6 528 20. 4 360 0 0 38.4 0.09 20.9 0.91
5
2 8 4 8 6 7 6 2 8 6 6

350 247 50 25 45 30 45 12. 12. 216 34

0 0 0 0 0 0 3 0 0 9 8

350 247 50 25 45 30 791 44 15. 15. 173 35

U1 7 791 453 7 383 30.9 0.00 30.3 1.43
0 0 0 0 0 0 804 5 9 9 9 0
U2 4 402 445 2 416 40.8 0.00 17.7 1.34
Xian et 350 247 50 25 45 30 628 49 20. 15. 173 34
U4 2 402 445 2 398 47.2 0.00 17.6 1.86
B13)
al. 0 0 0 0 0 0 942 2 0 9 5 8
U5 3 942 492 3 400 60.7 0.00 39.3 2.49
350 247 50 25 45 30 123 49 20. 20. 300 33
U6 2 942 492 3 420 59.3 0.00 32.7 2.05
0 0 0 0 0 0 2 2 0 0 4 0

350 247 50 25 45 30 46 28. 20. 300 33

0 0 0 0 0 0 3 0 0 4 0

Shiohara et 270 147 30 20 30 30 47 15. 15. 39

S3 995 5 995 470 5 202 512 28.0 0.04 10.1 0.87
al.B14) 0 0 0 0 0 0 0 9 9 0

B04 140 140 24 24 24 24 516 37 12. 4 516 378 12. 4 192 256 39 29.0 0.00 5.2 0.81

B06 0 0 0 0 0 0 645 8 7 5 645 378 7 5 192 256 9 29.0 0.00 6.0 0.74

ShioharaB15 D03 140 140 24 24 24 24 645 37 12. 5 645 378 12. 5 122 256 39 32.4 0.00 6.4 0.80
)
D07 0 0 0 0 0 0 903 8 7 7 903 378 7 7 122 256 9 32.4 0.00 8.5 0.76

D11 140 140 17 24 34 24 903 37 12. 7 387 378 12. 3 182 256 39 32.9 0.00 4.8 1.04

E03 0 0 0 0 0 0 995 8 7 5 995 425 7 5 192 256 9 61.4 0.00 12.6 1.26
 140 140 17 24 34 24 37 12. 12. 39

0 0 0 0 0 0 8 7 7 9

140 140 24 24 24 24 37 12. 12. 39

0 014 0 0 0 0 8 7 7 9

140 00 24 24 24 24 42 15. 15. 39

0 0 0 0 0 5 9 9 9

Kawai et 380 160 45 32 47 47 270 52 22. 270 22. 108 92

I6C 7 522 7 339 66.1 0.20 84.6 1.51
B16)
al. 0 0 0 5 5 5 9 2 2 9 2 0 8

a)
L= beam length (mm); H= column height (mm); hb= beam depth (mm); bb= beam width

(mm); hc= column depth (mm); and bc= column width (mm)

b)
As= area of re-bar (mm2); fy= yield strength of re-bar based on material test result (MPa); and

db= re-bar diameter (mm)

c)
Ah= sum of hoop bar areas (mm2); and fyt= yield strength of hoop bar based on material test

result (MPa).

Table B1– Test parameters of existing test specimens [1 in = 25.4 mm, 1 ksi = 6.90 MPa]

Concre
Later Bond
Top re-bar Bottom re-bar Joint te Axial
Geometric propertiesa al stren
of beamb of beamb hoopc strengt load
Specimens load gth
h

fc' Nc / P0 τu
L H hb bb hc bc As fy db n As fy db n hs Ah fyh
(MPa) (fc' (kN) (MPa
 bchc) )

225 125 35 26 40 30 113 37 19. 113 19. 91

0 0 0 0 0 0 6 8 1 6 1 768 1
J11C 4 378 4 290 57.6 0.24 58.1 2.16
Kaku et 225 125 35 26 40 30 204 37 25. 204 25. 768 91
J31A 4 370 4 290 55.2 0.25 64.4 1.77
al.B17) 0 0 0 0 0 0 0 0 4 0 4 102 1
J32B 4 370 4 290 55.2 0.25 76.9 2.12
225 125 35 26 40 30 204 37 25. 204 25. 4 91

0 0 0 0 0 0 0 0 4 0 4 1

270 147 30 20 30 30 78 12. 12. 32

0 0 0 0 0 0 103 0 7 7 0

A1 270 147 30 20 30 30 2 78 12. 8 516 780 12. 4 213 576 32 30.6 0.06 10.3 1.30

Kitayama A2 0 0 0 0 0 0 103 0 7 8 516 780 7 4 213 576 0 30.6 0.06 9.0 1.14

et al.B18) A3 270 147 30 20 30 30 2 78 12. 6 516 780 12. 4 218 576 32 30.6 0.06 9.7 1.19

A4 0 0 0 0 0 0 774 0 7 6 516 780 7 4 218 576 0 30.6 0.06 10.3 1.26

270 147 30 20 30 30 774 78 12. 12. 32

0 0 0 0 0 0 0 7 7 0

200 140 25 18 25 25 38 12. 12. 42

0 0 0 0 0 0 2 7 7 0
No1 516 4 516 382 4 170 640 28.6 0.16 7.1 1.27
Yoshino et 200 140 25 18 25 25 37 15. 15. 42
No3 603 3 603 379 3 170 640 28.6 0.16 7.4 1.40
al.B19) 0 0 0 0 0 0 9 9 9 0
No4 402 2 402 379 2 170 640 28.6 0.16 5.2 1.48
200 140 25 18 25 25 37 15. 15. 42

0 0 0 0 0 0 9 9 9 0

Kawasazak MKJ 270 147 30 20 30 30 77 19. 19. 67

568 2 568 771 2 220 768 84.4 0.13 13.8 2.29
B20)
i et al. 1 0 0 0 0 0 0 1 1 1 5
 Tateishi 375 180 40 30 35 35 100 33 15. 100 15. 127 36

and AIJ 0 0 0 0 0 0 5 6 9 5 5 336 9 5 314 8 3 23.5 0.04 19.2 1.12

IshibashiB21 HBS 375 180 40 30 35 35 103 38 12. 8 103 388 12. 8 270 127 37 23.5 0.05 25.9 1.38
)
0 0 0 0 0 0 2 8 7 2 7 8 6

Asou et 270 145 40 30 44 44 193 52 22. 193 22. 204 94

No1 5 520 5 298 67.1 0.18 98.6 2.62
al.B22) 0 0 0 0 0 0 5 0 2 5 2 8 9

203 246 30 20 25 25 41 12. 41

2 4 5 3 4 4 4 7 4
BCJ2 516 4 284 414 9.5 4 203 512 30.2 0.00 4.4 1.29
203 246 30 20 30 25 41 12. 41
LeonB23) BCJ3 516 4 284 414 9.5 4 203 512 30.2 0.00 5.0 1.19
2 4 5 3 5 4 4 7 4
BCJ4 516 4 284 414 9.5 4 203 512 30.2 0.00 7.5 1.49
203 246 30 20 35 25 41 12. 41

2 4 5 3 6 4 4 7 4

344 382 45 33 38 38 153 41 25. 19. 203 41

8 3 7 0 1 1 0 4 4 1 2 4
BCJ5
344 382 45 33 38 38 153 41 25. 852 19. 203 41
BCJ8 3 414 3 304 31.1 0.30 10.1 1.41
8 3 7 0 1 1 0 4 4 852 1 2 4
BCJ9 3 414 3 304 31.1 0.00 10.7 1.49
344 382 45 33 38 38 153 41 25. 852 19. 203 41
LeonB24) BCJ1 3 414 3 304 31.1 0.00 10.7 1.49
8 3 7 0 1 1 0 4 4 102 1 2 4
1 2 414 2 304 31.1 0.30 9.3 1.52
344 382 45 33 38 38 153 41 31. 0 25. 203 41
BCJ1 3 414 3 304 31.1 0.30 10.1 1.41
8 3 7 0 1 1 0 4 8 852 4 2 4
2
344 382 45 33 38 38 153 41 25. 19. 203 41

8 3 7 0 1 1 0 4 4 1 2 4
References for Table B1

B1. Hwang, H. J., Eom, T. S., and Park, H. G., “Design Considerations for Beam-Column Joint

with Strengthening Bars”, Engineering Structures, Reviewing.

B2. Hwang, H. J., Park, H. G., Choi, W. S., Chung, L., and Kim, J. K., “Cyclic Loading Test for

Beam-Column Connections with 600 MPa (87 ksi) Beam Flexural Reinforcing Bars”, ACI

Structural Journal, V. 111, No. 4, 2014, pp. 913-924.

B3. Teraoka, M., Kanoh, Y., Taraka, K., and Hayoshi, K., “Shear Strength and Deformation

Behaviour of R.C. Interior Beam-Column Joint Using High Strength Concrete”, Proc. 2nd US-

NZ-Japan-China Multilateral Meeting on structural Performance of High Strength Concrete in

Seismic Resions, Honolulu, 1994.

B4. Susanto, T. and Hua, Z., “Eccentric Reinforced Concrete Beam-Column Joints Subjected to

Cyclic Loading”, ACI Structural Journal, V. 100, No. 2, 2003, pp. 139-148.

B5. Durrani, A. J. and Wight, J. K., “Experimental and Analytical Study of Internal Beam to

Column Connections Subjected to Reversed Cyclic Loading”, Report UmEE82R3, Department

of Civil Engineering, University of Michigan, Ann Arbor, 1982.

B6. Hayashi, K., Teraoka, M., Mollick, A. A., and Kana, Y., “Bond Characteristic of Interior RC

Beam-Column Connections Using High Strength materials (in Japanese)”, Proceeding of the

Japan Concrete Institute, V. 15, No. 2, 1993, pp. 583-588.

B7. Sugano, S., Nagashima, T., Kimura, H., and Ichikawa, A., “Behavior of Beam-Column

Joints Using High-Strength Materials”, ACI SP 123-13, 1991, pp. 359-377.

B8. Oda, M., Kosugi, K., Yamanoka, H., and Tano, K., “Bond Characteristic of Flexural Beam

Reinforcement in Beam Column Connection for High Sgrength Reinforced Concrete Structure

(in Japanese)”, Proceeding of the Japan Concrete Institute, V. 19, No. 2, 1997, pp. 993-998.

B9. Joh, O., Goto, Y., and Shibata, T., “Influence of Transverse Joint and Beam Reinforcement

and Relocation of Plastic Hinge Region on Beam-Column Joint Stiffness Deterioration”, ACI

Special Publication, SP 123-8, 1991, pp. 187-223.

B10. Otani, S., Kobayashi, Y., and Aoyama, H., “Reinforced Concrete Interior Beam-Column

Joints Under Simulated Earthquake Loading”, Proc. 1st US-NZ-Japan Seminar on Design of

R.C. Beam-Column Joints, Monterey, 1984.

B11. Wong, P. K. C., Priestley, M. J. N., and Park, R., “Seismic Resistance of Frames with

Vertically Distributed Flexural Reinforcement in Beams”, ACI Structural Journal, V. 87, No. 4,

1990, pp. 488-498.

B12. Walker, S. G., “Seismic Performance of Existing Reinforced Concrete Beam-Column

Joints”, MS Thesis, Department of Civil and Environmental Engineering, University of

Washington, Seattle, 2001, pp. 308.

B13. Xian, Z. X., Park, R., and Tanaka, H., “Behaviour of Reinforced Concrete Interior Beam-

Column Joints Designed Using High Strength Concrete and Steel”, Research Report, No.92-3,

Department of Civil Engineering, University of Canterbury, 1992.

B14. Shiohara, H. Zaid, S., and Otani, S., “Test of An Innovative Reinforcing Detail for R/C

Interior Beam-Column Connections Subjected to Seismic Action”, Proceedings of the Third

International Conference on Concrete under Severe Conditions, University of British Columbia,

Vancouver, Canada, 2001, pp. 739-746.

B15. Shiohara, H., “Comprehensive Series of Tests on Seismic Performance of Reinforced

Concrete Interior Beam-Column Joints: Part 1”, Seismic performance of concrete joints and

connections, ACI Fall Convention, 2010.

B16. Kawai, T., Kimura, H., Iwata, M., and Watai, T., “Experimental Study of Resistance

Mechanism of RC Beam-Column Connections Using High Strength Materials (in Japanese)”,

Proceeding of the Japan Concrete Institute, V. 19, 1997, pp. 1011-1016.

B17. Kaku, A., Maso, K., Kutoka, T., and Muguruma, T., “Experimental Study about

deformation characteristic of beam column connection in RC structure (in Japanese)”,

Proceedings of the Japan Concrete Institute, V. 15, No. 2, 1993, pp. 559-564.

B18. Kitayama, K., Kojima, C., Otani, S., and Aoyama, H., “Behavior of Reinforced Concrete

Interior Beam-Column Connection Subjected to High Shear (in Japanese)”, Proceedings of the

Japan Concrete Institute, V. 11, No. 2, 1989, pp. 531-536.

B19. Yoshino, M., Takeda, S., and Kamimura, T., “Behavior of Interior RC Beam-Column

Connections after Yielding of Flexural Beam Reinforcement (in Japanese)”, Proceedings of the

Japan Concrete Institute, V. 19, No. 2, 1997, pp. 987-992.

B20. Kawasazaki, T., Kitayama, K., and Noguchi, H., “Bond characteristic of Reinforced

Concrete beam-Column Connections using Ultra High Strength Materials (in Japanese)”,

Proceedings of the Japan Concrete Institute, V. 14, No. 2, 1992, pp. 397-400.

B21. Tateishi, M. and Ishibashi, K., “Experimental Study about Failure Type in Beam Column

Connection after Reaching Yielding of Beam Reinforcement (in Japanese)”, Proceedings of the

Japan Concrete Institute, V. 20, No. 3, 1998, pp. 517-522.

B22. Asou, N., Nagashima, T., and Sugano, S., “Force Characteristic of Beam Column

Connection using High Strength Concrete (Fc600) and Reinforcement (SD490) (in Japanese)”,

Proceedings of the Japan Concrete Institute, V. 15, No. 2, 1993, pp. 553-558.

B23. Leon, R., “Shear Strength and Hysteretic Behavior of Interior Beam-Column Joints”, ACI

Structural Journal, V. 87, No. 1, 1990, pp. 3-11.

B24. Leon, R., “The Effect of Floor Member Size on the Behavior of Reinforced Concrete

Beam-Column Joints”, 8th World Conference on Earthquake Engineering, San Francisco, 1984,

pp. 445-452.

APPENDIX C: SAMPLE CALCULATIONS

In the section “VERIFICATION OF THE PROPOSED BOND MODEL”, the bond slip-

stress relationship of specimen No.14 shown in Fig. 6(d) was determined as follows. The bond

length of the concrete block hc = 635 mm [25.0 in], re-bar diameter db = 25.4 mm [1.0 in],

concrete strength fc′ = 32.7 MPa [4.7 ksi], area of the transverse bars in the joint Ah = 1548 mm2

[2.4 in2], yield strength of hoop bars fyh = 493 MPa [71.4 ksi], area of the beam re-bar As = 510

mm2 [0.8 in2], yield strength of the beam re-bar fy = 469 MPa [68.0 ksi], bond strengths τec (=

2.2√fc′) = 12.6 MPa [1.8 ksi], τet (= 1.8√fc′) = 10.3 MPa [1.5 ksi], τu (= [0.16+0.11(Ahfyh) /Vu]√ fc′)

= 1.86 MPa [0.3 ksi] (Vu = Asfy for the concrete block specimen), yield strain of the beam re-bar

εy = 0.0023, elastic modulus Es = 200000 MPa [28986 ksi], and hardening modulus Esh (= 0.02Es)

= 4000 MPa [580 ksi]. The elongation eb of the re-bar at the joint interface is calculated by

integrating the bar strain ε over the joint depth hc shown in Fig. C1. For example, for a given bar

strain εt = 0.01 and the given bond strengths, the bar strain distribution inside the joint is

determined as follows (see Fig. C1).

 Esh d b
lu    t   y   105.4 mm [4.1 in] (C1)
4 u

Es d b
let   y   hc  2lu   282.3 mm [11.1 in] (C2)
4 et

lec  hc  2lu  let  141.9 mm [5.6 in] (C3)

4 et
 oo   y  let  0.0 (C4)
Es d b

4 ec
 co   lec  0.0014 (C5)
Es d b

4 u
 c   co  lu  0.0016 (C6)
Es d b

Integrating the tensile strains, the bar elongation eb at the joint interface is calculated as

follows.

eb 
 y let

 t   y  lu
 0.97 mm [0.04 in] (C7)
2 2

The bond strength Bj of the joint is calculated by integrating the bond stresses along the joint

depth hc.

B j   let et  2lu u  lec ec  db   405.4 kN [91.1 kips] (C8)

  ec  12 . 6M Pa
 et  10 . 3M Pa
Bond stress 
 u  1 . 86MPa
(MPa)

 co   0 . 0014  t  0 . 010
4 u
 c   0 . 0016  00  0 . 0 E sh d b
1
Strain   y  0 . 0023
(mm/mm)
l u  105 . 4m m l u  105 . 4m m
l ec  141 . 9m m l et  282 . 3m m

h c  635m m

Fig C1– Strain distribution of beam re-bar in specimen No.14 [1 in = 25.4 mm, 1 ksi = 6.90

MPa]

APPENDIX D: EXPERIMENTAL EVIDENCE FOR BAR-SLIP

Eq. (8) was developed assuming that the lateral load P0 at the origin is determined by the

uniform friction bond strength τu within the beam-column joint. Figs. D1 and D2 show

experimental evidence supporting the assumption. Fig. D1 shows the test results reported by

Hwang et al.B2 In the figure, at the origin of the load-displacement relationship, the strains of the

bottom bars in the joint are maintained as a uniform value. Further, the strains remained in

tension in both the positive and negative loadings. Fig. D2 shows the test results reported by

Xian et al.B13 In the figure, at the origin, slip deformation increased without increase of the

lateral load. These results indicate that the uniform lateral load P0 at the origin is determined by

the uniform friction bond strength of the beam flexural bars in the joint. For this reason, in the

proposed model, the bond stress distribution after complete bearing bond failure was simplified

as the uniform friction bond strength τu.

 600 0.004 0.004

Strain (mm/mm)
(a) Hwang  ‐ (b)  (c) Top bars 
300 S1 (2 layers)
Load (kN)

0.002 Gauge 0.002 Gauge

0 Bottom bars
2P 0 0 0 (1 layer)
‐300

‐600 ‐0.002 ‐0.002

‐6 ‐4 ‐2 0 2 4 6 ‐6 ‐4 ‐2 0 2 4 6 ‐6 ‐4 ‐2 0 2 4 6
Story drift ratio (%) Story drift ratio (%) Story drift ratio (%)

Fig D1–Beam re-bar strain-displacement relationshipB2 [1 in = 25.4 mm, 1 kips = 4.45 kN]

200  200 
(a) Xian ‐ U2 (b) Slip 
Load (kN)

100  100  deformation 

0 
(U2)
0 
2P 0 ‐100 
‐100 

‐200  ‐200 
‐6  ‐4  ‐2  0  2  4  6  ‐6  ‐4  ‐2  0  2  4  6 
Story drift ratio (%) Slip deformation (mm)
200  200 
(c) Xian ‐ U4 (d) Slip 
Load (kN)

100  100  deformation 

(U4)
0  0 

‐100  2P 0 ‐100 

‐200  ‐200 
‐6  ‐4  ‐2  0  2  4  6  ‐6  ‐4  ‐2  0  2  4  6 
Story drift ratio (%) Slip deformation (mm)
200 200 
(e) Xian ‐ U6 (f) Slip 
100 100  deformation 
Load (kN)

(U6)
0 0 
2P 0
-100 ‐100 

-200 ‐200 
-6 -4 -2 0 2 4 6 ‐6  ‐4  ‐2  0  2  4  6 
Story drift ratio (%) Slip deformation (mm)

Fig D2–Load-slip deformation relationshipB13 [1 in = 25.4 mm, 1 kips = 4.45 kN]