Behavior of external column- wide beam joint with different bar arrangement and existence of

joint shear link under gravity load

AmirFateh
1*
,ArashBehnia
2
,SeyedJamalodinHoseini
3
1
Department Of Civil Engineering, Faculty Of Engineering, University Putra Malaysia
(UPM), Kuala Lumpur, Malaysia
2
Department Of Structure & Materials, Faculty Of Civil Engineering, Universiti Teknologi
Malaysia(UTM), J ohor, Malaysia
3
Department Of Structure & Materials, Faculty Of Civil Engineering, Universiti Teknologi
Malaysia(UTM), J ohor, Malaysia
*
author.

Tel.: +60172965274
E-mail address: gs31295@mutiara.upm.edu.my ,
Amirfateh.saze@yahoo.com





ABSTRACT
An experimental investigation done on the RC exterior wide beam-column joint when
subjected to the gravity load up to failure is reported in this paper. This study was conducted
by applying the concentrated gravity load on full scaled wide beam-column joints with same
area of longitudinal reinforcement to resist for negative moment due to concentrated gravity
load. The joints behavior were considered by effect of different layout of beam longitudinal
bars , existence of shear link in connection zone, spandrel bar and width of the beam in terms
of failure capacity, crack patterns, deflection and rotation. The results shown that the failure
capacity of joints with concentrated longitudinal bars of beam that two-third of bars anchored
in column zone was 24 % higher than even bar distribution. And also the existence of shear
link in connection area and spandrel bar to anchor the longitudinal beam reinforcements that
were outside the connection area is higher than the other specimens without them. Moreover
the wide of beam played important role to enhance the failure capacity.
Keywords: Wide beam-column connection, Gravity loading test, reinforcement layout, Shear
link, Spandrel bar

INTRODUCTION
Wide beams are often used in the medium or high rise building as structural members that
support the slab and transfer the load to the columns or walls. In the usual Reinforced
concrete connections, the width of the beam does not go beyond the width of the column.
However, in many cases, beams width (b
b
) is larger than column width( b
c
) that it is named
wide beam or shallow beam column joint . The main advantages of using the wide beams
as a gravity load resisting system are increasing the width of beam causes the
reduction in both flexural reinforcement and shear links. Furthermore, this kind of
beam has practical advantages such as simplifying the formwork requirement, reduce the
reinforcement detailing and also increasing in the net height floor and result in
reduction of the total height of building. All of these would finally cause a faster
construction and diminish the cost. Gravity load-resisting frames in none seismic prone
normally used reinforcement concrete wide beam frame buildings and composite structural
systems of wide beam framing with other members due to these advantages. Presently, these
benefits have brought about augmented of wide beam systems even in seismically active
areas[1, 2]. Behavior of the wide beam exterior connection under horizontal loading is
influenced by numerous factors such as the ratio of beam to column width ,the amount of
beam longitudinal steel bar anchored in column zone , existence of transverse beam and
etc.[3, 4]. On the other hand, numerous studies were done to find that effect of various factors
on failure capacity of the normal beam-column column joints under gravity loading , such as
shape of anchorage for longitudinal reinforcements of beam in column [5],Axial compression
load on the column[6] .in addition to, the connection failure mode influenced by the
quantity of beam reinforcements that are carry tensile force [5],concrete strength in the
joint zone[7].

In the codes width of the beam ratio and continuation of horizontal link in connection
area[8, 9] were mentioned for ductile design of RC structure when subjected to lateral load. In
the British Standard code (BS) has not provided any design guides for wide beam column
connection.[10] Therefore, the effect of beam longitudinal bars that are anchored in the
connection zone, existence of spandrel bar and joint shear link on connection behavior in
terms of failure capacity, crack patterns, deflection and etc, are investigated in this study .
EXPERIMENTAL PROGRAM
Sample detail
Figure 1&2 illustrate the detail of five exterior wide beam column connection samples
were labeled as C1, C2, C3, C4and C5.this specimens symbolized part of RC building frame
connection and achieved by deleting of the beam and column at their mid point. The points
correspond with bending moment inflection due to vertical loading take place.
figure 1&2 shows the beam longitudinal bars and shear links in columns and beams in
the test specimens. Part of longitudinal reinforcements in the beam that were designed for
negative moment were bent into the joint core with definite length to have sufficient
anchorage length based on BS code [10] as can be seen in figure3.All of five specimens
were casted with a beams length of 1000 mm and the columns height of 2000mm The
columns had a same cross section with 250mm width and 300mm depth in all specimes .
The beams were located at mid-height of columns with 200mm depth. As shown in
Fig.1&2.All beams were cast in the same size with dimension of 600 mm width and200 mm
height with ratio of b
w
/h
c
equal to 3 ,except sample C3 that was cast with 400 mm width and
200 mm height. b
w
and h
c
are beam width and beam height respectively

The differences in specimens are shown in Table1. bar arrangement was even distributed
in samples C1 and C4 whereas in other specimens the arrangements of bars are
concentrated in connection zone. All of specimens included the shear link in joint zone
except C4 and also the spandrel bars were existed in all samples except C5.
The dimension of wide beam-column connection based on the code recommendation
maximum beam width (b
b
) is lesser of b
c
+1.5hc and 3bc (b
c
=column width; h
c
=
column depth), and at least one-third of the wide beam flexural steel should be anchored in or
through the column core to satisfy ductility[8]. So For specimens C2, C3 66.6 % and
C1,C4,C5 33.33 % of top longitudinal bars area anchored in Column core with the length of
at least 40 times of bar diameter that is sufficient anchorage length based on BS code
[10].the anchorage is depicted in figure 3.
Design detail and material properties
Longitudinal bars in beams and columns consisted of high yielded bars with yielded strength
of 460Mpa denoted by T and varied from 10 to 14 mm and mild steel bars denoted by
R were used as a shear links in beams and columns with yielded strength of 250Mpa and
6 mm diameter. The bar are shown as R6, T16, etc,in which T and R denote deformed and
plain bars. All steel reinforcement were free from rust, oil, or other coating that may
destroy or reduce bond. During design of samples the concrete compressive strength for all
specimens was targeted 40 N/mm
2
. Before casting the main samples, several trail mixes
based on BS mix design recommendation, were done to ensure the sufficient strength and
proper workability for placing. Also during the placing of concrete in formwork, every layer
of concrete was vibrated by using vibrator with a head of 25 mm diameter. . All of specimens
were cured for period of seven days in formworks. After seven days the formworks were
removed and store in the laboratory until they were tested, 16 cubic samples with dimension
of 150 x150 x 150 mm were taken and the average of concrete strength was 42.1 N/mm
2
at
the day of test.
Figure 1 &2. Show the top bars in beam include six 16mm diameter high yield reinforcement
(6T16).concrete with 10 mm maximum aggregate size and Portland cement type2 were used
for the samples, with 30 mm concrete cover, the distance between two bars in severe
condition(concentrated bar arrangement into joint zone) is around 53 mm that is greater than
the minimum distance between bars based on BS code [10].shear links were 6 mm
diameter mild steel spaced at 80 mm centers(R6-80).at the free end of beam, the shear link
spacing reduce to 40 mm to avoid the possibility of concrete crushing under
concentrated load. The top bar that were passed in joint area anchored at least 40 times of bar
diameter that is sufficient anchorage length. The column bar consisted of 4 bars with 16
mm diameter (4T16) high yield with 6mm diameter mild steel links, spaced
80mm(R6-100mm).the maximum allowable distance for links in column bars based on
BS code [10] is 12 times diameter of main bar (192 mm).and also at the this spaced was
reduced to 50 mm at the both outer end of column to prevent concrete crushing due to
horizontal load that were produced indirectly from vertical load in beam .
The arrangement of wide beam reinforcement in the specimen C1 was an equal distribution in
the width of wide beam, with existence of shear link in connection zone and also with one
extra spandrel bar that all of the reinforcement out of the connection zone was anchored with
it. Specimen C2 was also cast with same size as specimen C1 with same concrete
compression strength, but the main difference was concentrated bar distributions in
connection zone. The sample C3 was similar as other samples but the beam dimension
was 400X200 mm, 200 mm less than other samples. Two third of bars were
concentrated in connection zone such as sample C2. In this sample shear link and spandrel
beam were existed. Specimen C4 was also cast with same size as specimen C1, the
longitudinal bar distribution in wide beam and concrete compression strength were
as same as specimen C1, but the shear links in connection zone were deleted. And
specimen C5 was cast as same as specimen C2, but in specimen C5 the spandrel bar was
eliminated. The specimens detail is shown in Table.1.
Test arrangement
The test setup is illustrated in Figure 4. The loading apparatus including steel portal frame
was braced to the laboratory floor with aid of bolt and nut system. First the column was
installed in middle frame and kept it from laterally movement at both ends. Lower part of
column was installed in the steel box with size of 300X300X300 with25 mm thickness that is
bolted to the floor. Two incline meters were installed in the middle of column and beam to
measure the rotation between the column and beam. The hydraulic jack with load cell were
placed on top of the column to imposed permanent 130KN downward to the column to
simulate the real position of samples in real RC frame, indeed another hydraulic jack with
load cell was applying load gradually with rate of 5 KN downward at distance of 850mm
from column face as shown in Figure 3 until failure of the beam occurred. Three LVDT were
placed n the middle and side of the beam to measure the deflection.

Nominal design capacities
The theoretical for ultimate load capacity value based on the previous studies and code[5, 10]
V
,
c
=V
c
+0.6(N.V.h
c
/ A
C
. M) [10]
V
,
c
=V
c
+0.6(N.V.h
c
/ A
C
. M) [5]

V=v
,
c
. b
c
.d
c

Where :
v
c
=shear stress
v
,
c
=design shear stress
b
c
=breadth of column cross section
d
c
=effective depth of column
N =column axial load
V =design shear force due to ultimate load
h
c
=depth of column cross section
A
C
=cross section area of column
M=design ultimate moment
The maximum load was calculated 99.98 KN and this value was same and can not be
considered variable factors such as bar arrangement variation and etc.
EXPERIMENTAL RESULTS
Due to the different longitudinal bar arrangement in wide beam, that are anchored in the joint
zone, existence of shear link in connection area, different width of wide beam and also
existence of spandrel bar for anchoring the longitudinal bar of wide beams that were located
outside of wide beam-column joint, different behavior were expected in experimental test in
terms of failure loading capacity, deflection and rotation of wide beam and joint stiffness.
In specimen C1 flexural cracking appear at 30KN (about 35%of ultimate load) in the beam
surface. When the load reached to 35KN, the crack developed in the same pattern, at 45KN
another crack occurred parallel to the first crack at the distance one-third of span from column
face on the beam surface and extend to both sides o beam. At 60 KN, another crack appeared
at two third from column face parallel to another cracks. When the applied load reached at
90%of failure load, the cracks happened in connection area and on the top surface of the wide
beam near connection zone with maximum crack width about 2 mm. The sample failed at
102.5KN due to excessive crack in top surface of wide beam near column. It was passed the
theoretical capacity. The maximum rotation between wide beam and column was 3.27 degree
and the maximum deflection at failure point was 51.58mm.the crack pattern of sample C1
shows in figure4.
In specimen C2, a cracking pattern analogous to the specimen C1.its showed early flexural
cracks. The first crack occurred at 30KN (about24%of ultimate load) at top surface of beam.
When load reached to 40KN, the crack developed in the same pattern. While at 45KN another
crack occurred parallel to the first crack at the distance of one-third of span from column face
on the beam surface and extend to both sides at 55KN. At point 75KN, another crack occurred
at two third from column face parallel to the previous cracks. When the applied load reached
around 80%of failure load, the cracks happened in connection area and on the top of the
surface of the wide beam near connection zone with maximum crack width about 1.5 mm.
The sample failed at 127.6KN due to excessive crack in wide beam near column. The
maximum rotation between wide beam and column was 2.2degree and the maximum
deflection at failure point was 76.58mm.the top view of cracking patterns is illustrated in
figure5.

In the figure6. can be obtained the crack patterns of specimen C3 that were similar to the
other samples, first crack was developed at face of the column-beam intersection at 35KN,
about32%of failure load at beam surface, then next crack happened at around one third of
beam face and extend to both side of beam at45KN. Another crack occurred at two third from
column face parallel to another cracks. When the applied load reached about 45%of failure
load, some cracks also extended in both sides of beam when the load reaches about 90%of
failure load. The maximum crack width about 2 mm occurred in connection area in the top of
the wide beam surface. The sample was failed when the maximum applied load reached at
114 KN, the maximum deflection at failure point was 76.4 mm and rotation was about 3.16
degree.
In sample C4,the cracks patterns look like other specimens were parallel to the wide beam
width and extend vertically to both side of the wide beam, when the load reached at 45KN
two cracks were investigated at distance around one third and two -third of wide beam span
at surface of beam from column face. When the applied load reached about 84% failure, the
cracks were occurred at column in connection zone area. The failure load in this sample was
98 KN and maximum deflection was recorded 48 mm and maximum rotation was 2.17
degree. the crack pattern is depicted in figure7.
As can be found in figure 8, Specimen C 5 The first crack occurred at 40KN (about 37%of
ultimate load) in the column face, parallel to width of the beam and top of the beam surface at
the distance roughly one third from column face. When load reached at 65KN, other cracks
were observed in the same pattern that parallel to first crack at distance of two third from
column side, in surface and extended to both side of wide beam at 90 KN. When the applied
load reached at 84%of failure load, the cracks happened in connection area and on the upper
surface of the wide beam near connection zone with maximum crack width about 2 mm. The
sample failed at107KN due to excessive crack in wide beam near column. The maximum
rotation between wide beam and column was 1.9degree and the maximum deflection at failure
point was 38.89 mm.
DATA ANALYSIS AND CONCLUSION
1. The effect of wide beam bar arrangement in connection zone on the ultimate capacity
of joints can be seen from the results of specimenC1 and C2 in the table 2 and figure
9&10.specimen C2 was concentrated longitudinal wide beam bar arrangement in
connection zone resisted an ultimate load of 127.6KN compared to 102.5 KN for
specimen C1which had even distribution bar arrangement. This result shows the
concentrated reinforcement arrangement in connection area improves the shear
capacity of the joint about 24%.Perivous Tests that were done on normal RC beam-
column connection had also reveals the same trends[11].
2. Diagonal cracking of concrete in connection area can be controlled by two
means. Firstly, provide large column sizes. Secondly, provide closely space close-
loop steel ties around column bar in the connection area. These ties hold the
concrete in the connection together and also resist shear force therefore reduce the
cracking in the zone. based on codes recommendation [8]the continuous transverse
loops around the column bar through the connection zone. Specimens C1 and C4
had same concrete strength and dimension, but in specimen C4, the shear links in joint
zone were eliminated. As can be understood from ultimate load in Table 2 and figure
9, the additional shear links in connection area increase the ultimate shear capacity
around 6%. Specimen C1 which had shear links achieved 102.5KN whereas in
specimen C4, the ultimate load is 98.8 KN. And also, as can be seen from figure 8 the
maximum rotation is belonged to sample C4. existence of shear link in joint zone
reduce the rotation around 42%

3. The effect of spandrel bar in joint area on the ultimate load strength was
investigated from the results of specimens C2 and C5. As can be seen in Table 2, in
specimens C2 with spandrel bar in joint that all beam longitudinal bars outside the
joint area, anchored to it, illustrated more ultimate capacity 127.6 than specimen C5
without spandrel bar. It mean presence of spandrel bar in connection zone enhance
ultimate joint capacity around 18%. A lot of research had been done to find out the
behavior of joint with existence of spandrel beam and the [12] .this study showed
that the existence of slab bars increased the negative moment capacity of beam
4. Specimens C2 and C3 had same concrete strength but different beam dimension.
In sample C1 the width of wide beam was 600 mm where as in specimen C3 was 400
mm. as can be understood from Table 2 specimens C1 resisted ultimate load 127.6KN
while in sample C3,the ultimate load reached114.3.it means that the width of wide
beam increase the capacity of connection about 10%.
ACKNOWLEDGMENT
This study conducted in structure engineering laboratory of universiti teknologi Malaysia
(UTM).the author express his sincere appreciation to Associate Professor Dr. Ramli Abdullah
for his valuable guidance and help



s
p
e
c
i
m
e
n



b
w


mm


h

mm


b
w
/h
reinforcement

B
a
r

d
i
s
t
r
i
b
u
t
i
o
n


S
p
a
n
d
r
e
l

b
a
r


S
h
e
a
r

l
i
n
k



beam
Colum

250X300mm


C1


600


200


3


6T16

4T16 E
*

Yes


Yes


C2


600


200


3


6T16

4T16 C
*

Yes


Yes


C3


400


200


2


6T16

4T16

C
*

Yes


Yes


C4


600


200


3


6T16

4T16 E
*

Yes


No

C5

600

200

3

6T16 4T16 C
*
No

Yes
E
*:
even bars distribution in wide beam
C
*:
concentrated bars distribution in wide beam
Table 1

Table 2

Specimen name

Load(KN)
Max

deflection(mm)
Max

rotation(degree)
C1 102.5 51.58 1.59
C2 127.6 76.4 1.43
C3 114.3 53.15 1.84
C4 98.8 48.47 3.44
C5 107.4 38.89 1.57
Figure1. Samle C1,C2,C3,C5 detail
Figure2. Samle C4 detail

Figure 3.test set up

Figure 4.crack patterns of specimen C 1

Figure5.crack patterns of specimen C 2

Figure 6.crack patterns of specimen C3




Figure 7.crack patterns of specimen C 4

Figure 8.crack patterns of specimen C 5

Figure 9. Load Relative Rotation Curve

0
20
40
60
80
100
120
140
0 0.5 1 1.5 2 2.5 3 3.5 4
L
o
a
d

(
K
N
)
RelativeRotation(Degree)
C1
C2
C3
C4
C5

Figure 10.load-deflection curve

0
20
40
60
80
100
120
140
0 20 40 60 80 100
l
o
a
d
(
K
N
)
Deflection (mm)
C1
C2
C3
C4
C5

REFERENCES
1. Benavent-Climent, A., X. Cahas, and J . Vico, Interior wide beam-column connections in
existing RC frames subjected to lateral earthquake loading. Bulletin of Earthquake
Engineering. 8(2): p. 401-420.
2. Benavent-Climent, A., X. Cahis, and R. Zahran, Exterior wide beam-column connections in
existing RC frames subjected to lateral earthquake loads. Engineering Structures, 2009. 31(7):
p. 1414-1424.
3. Gentry, T.R. and J .K. Wight, Wide Beam Column Connections under Earthquake Type
Loading. Earthquake spectra, 1994. 10(4): p. 675-703.
4. LaFave, J .M. and J .K. Wight, Reinforced concrete exterior wide beam-column-slab
connections subjected to lateral earthquake loading. ACI Structural J ournal, 1999. 96(4).
5. Scott, R., The effects of detailing on RC beam/column connection behaviour. Structural
Engineer, 1992. 70(18).
6. Robert Park, T.P., Reinforced concrete structures. 1975: Willey intersience publication.
7. Marzouk, H., M. Emam, and M.S. Hilal, Effect of high-strength concrete columns on the
behavior of slab-column connections. ACI Structural J ournal, 1996. 93(5).
8. Recommendations for Design ofBeam-Column Connections inMonolithic Reinforced Concrete
Structures, in ACI-ASCE Committee 352. 2002.
9. Murty, C.V.R., Learning earthquake design and construction 11. What are the Indian seismic
codes? Resonance, 2005. 10(1): p. 83-87.
10. Structural use of concrete Part 1: Code of practice for design and construction in BS-8110-1 :
1997. 1997.
11. Park, R. and T. Paulay, Reinforced concrete structures. 1979: World Scientific Publishing
Company.
12. Di Franco, M.A., D. Mitchell, and P. Paultre, Role of spandrel beams on response of slab-
beam-column connections. J ournal of Structural Engineering, 1995. 121(3): p. 408-419.