Even though designers often take care to design and detail

elements like beams, columns, footings, and walls, for ductile

behavior they often ignore beam-column joints. As a result, joints
often become the weakest links in the structural system. Joints
are crucial zones for the transfer of forces and moments
effectively between the connecting elements like beams and
columns. When a building is located in non-seismic zone and
designed only for gravity loads, the design check for joints may
not be critical and hence usually not performed. However, the
catastrophic failures reported in the past earthquakes, especially
during the past several earthquakes in India as well as those in
Turkey and Taiwan in 1999, were attributed to beam-column
joints [Saatcioglu et al. (2001) and Rai and Seth (2002), and
Arslan and Korkmaz,( 2007)]. Some of these failures are shown
in Fig. 1.
Hence in this paper, the design and detailing of beam-column
joints for seismic loads are discussed. A comparison of the Indian,
American and the New Zealand provisions are also provided. It
has to be noted that when buildings are subjected to earthquake
loads, as they are often designed only for a fraction of the
earthquake loads (about 10% of load), plastic hinges will be
formed at the ends of the members where there are heavy
bending moments, and subsequently the frames will fail when
there are enough plastic hinges to form a mechanism. Hence, one
has to follow the general rules while planning and designing
buildings in earthquake zones-for example, the buildings should
be regular, soft and weak stories should be avoided, and columns
should be carefully designed (300 mm minimum size is required),
column ties should have closely spaced near plastic hinge zones
and should have 135o hooks with six diameter or greater than 65
mm extensions. Moreover strong column-weak beam principle
should be followed. More details about these aspects may be
found in Subramanian, 2014.

Beam-Column Joints

The performance of framed structures depends not only upon the

individual structural elements but also upon the integrity of the
joints. In most of the cases, joints of framed structures are
subjected to the most critical loading under seismic conditions.
However, despite the significance of the joints in sustaining large
deformations and forces during earthquakes, specific guidelines
are not explicitly included in Indian codes of practice for their
design and detailing until recently (IS 456: 2000 and IS 13920:
1993). Only recently some provisions have been included in the
draft IS 13920 based on ACI 318 and ACI 352 (ACI-ASCE
Committee 352) provisions [Jain et al. (2006)].

While considerable attention is devoted to the design of individual

elements (slabs, beams and columns), no conscious efforts are
made to design joints in the absence of suitable guidelines. It
appears that the integrity and strength of such joints are
assumed to be satisfied by just anchoring the beam
reinforcement in the joints.

One of the basic assumptions of the frame analysis is that the

joints are strong enough to sustain the forces (moments, axial
and shear forces) generated by the loading, and to transfer the
forces from one structural member to another (beams to
columns, in most of the cases). It is also assumed that all the
joints are rigid, and the members meeting at a joint deform
(rotate) by the same angle. Hence, it is clear that unless the
joints are designed to sustain these forces and deformations, the
performance of structures will not be satisfactory under all the
loading conditions, especially under seismic conditions. Since the
mid-1960s, numerous experimental tests and analytical studies
have been conducted to investigate the performance of RC beam-
column connections subjected to lateral earthquake loading
(Hanson and Connor, 1967). Post-earthquake analyses of
structures, accidental loading or laboratory tests show that the
distress in the joint region is the most frequent cause of failure,
rather than the failure of the connected elements [Saatcioglu et
al. (2001) and Rai and Seth (2002), Park and Paulay(1975)].
Analytical models which simulate the response of reinforced
concrete interior beam-column joints have been developed [ See
for example, Mitra and Lowes (2007)]. Even though such models
are currently available and documented in Open Sees website
(www.opensees.berkeley.edu), they are complicated and not
suitable for design office use.

The beam-column joint is defined as the portion of the column

within the depth of the deepest beam that frames into the column
(ACI 352-02). The types of beam-column joints in a moment
resistant frame, can be classified as (a) interior joint, (b) exterior
joint and (c) corner joint (see Fig. 2). When four beams frame
into the vertical faces of a column, the joint is called an interior
joint. When one beam frames into the vertical face of a column
and two more beams frame into the column in the perpendicular
direction it is called an exterior joint. A corner joint is one in
which beams frame into two adjacent vertical faces of a column.
In a roof joint (also called knee joint), the columns will not extent
above the joint, whereas in a floor joint, the columns will extend
above the joint as shown in Fig. 2.
Considerable research and test results are reported on the
strength and behaviour of beam-column joints [Subramanian and
Prakash Rao (2003), Paulay and Priestley (1992), ACI SP
123(1991)]. The revised IS 1893: 2002 has enhanced the lateral
forces on structures considerably compared to the previous
versions, which makes the design of joints imperative.
Recommendations on the design and detailing are available,
which were developed based on the test results [ACI 318-2014,
ACI 352-02, NZS 3101-2006, Draft IS 13920, EN 1998-1:2003].
The design and detailing procedures as per draft IS 13920 [Jain
and Murty(2002)] and ACI 352-02 (which was first published in
1976) are summarized in the following sections. Beam-column
joints in buildings located in earthquake zones and were built
before the development of current design guidelines need to be
repaired and strengthened. A state-of-the art on repair and
rehabilitation of RC beam-column joints is provided by Engindeniz
et al. (2005).

Requirements of beam-column joints

A beam-column joint undergoes serious stiffness and strength

degradation when subjected to earthquake loads. The essential
requirements for the satisfactory performance of a joint in an RC
structure, during earthquakes, can be summarized as follows
[Park and Paulay (1975), Paulay and Priestley (1992)].

(i) A joint should exhibit a service load performance equal to or

greater than that of the members it joins; that is, the failure
should not occur within the joints. Should there be a failure due
to overloading, it should occur in beams through large flexural
cracking and plastic hinge formation, and not in columns
(normally the joint is considered as a part of the column).

(ii) A joint should possess strength not less than the maximum
demand corresponding to the development of the structural
plastic hinge mechanism of the structure. This requirement will
eliminate the need for repair in an inaccessible region in the
structure.

(iii) The joint should respond elastically during moderate
earthquakes.


(iv) The deformation of joints should not increase the storey

drift significantly.


(v) The joint configuration should ensure ease of fabrication

and good access for placing and compacting concrete in the joint
region.

Design and detailing of joints

The problems of detailing and construction of beam-column joints

are often not appreciated by designers. Because of the restricted
space available in the joint block, the detailing of reinforcement
assumes more significance than anywhere else. Indeed, the
conflicting requirement of small size bar for good performance,
and large size bars for ease of placement and concreting is more
obvious at the joints than anywhere else (Subramanian and
Prakash Rao, 2003). This is particularly true for internal joints,
where the beams intersect in both the horizontal directions, and
where large moments are to be sustained by the connections. In
the absence of specifications from the designers, site engineers
often adopt expedient procedures for detailing, which are not
always conducive for satisfactory structural performance.
Some of the incorrect detailing practices adopted by site
engineers in India are (Subramanian and Prakash Rao, 2003): (1)
incorrect bending of beam reinforcement into the beam-column
joint for anchorage. The beam bars (at top) are bent upwards
instead of downwards; such a detailing prevents diagonal strut
formation in the joint and may cause diagonal cracking, leading
to shear failure of the joint. (b) Inadequate anchorage of beam
bars into the beam-column joint, (3) Poor quality concrete at the
critical region of the joint, obviously due to poor quality formwork
coupled with inadequate compaction, and (4) Kinking of column
bars near beam-column joints; Such kinking of bars can damage
concrete at the joint, may cause excessive stresses in the
column, and lead to early distress, especially under lateral forces
induced by earthquakes (Subramanian and Prakash Rao, 2003).
It may also be noted that shear reinforcement are usually not
provided in these joints; even when provided, they are not as per
IS 13920. Further, some site engineers provide the extreme bars
of the beam reinforcement outside the column bars, which is not
a correct practice. Unless we have a wide beam, the beam
reinforcement should be placed within the column cage, without
much kinking (see Fig. 3).
Beam-Column Joints in frames

The beam-column joint in a multi-storey frame, transfers the

loads and moments at the ends of the beams into the columns.
The roof beam-column joint in a frame has the same flow of
forces as the T-joint of Fig. 4(a), and its crack pattern is also
similar to Fig. 4(b). The forces

acting on an interior joint subjected to gravity loading is shown in

Fig. 5(a). As shown in this figure, the tension and compression
from the beam ends and axial loads from the columns are
transmitted directly through the joint. For a four-member
connection as shown in Fig.5(a), if the two beam moments are in
equilibrium with one another then no additional reinforcement is
required.

In the case of lateral loading like seismic loading, the

equilibrating forces from beams and columns, as shown in Fig.
5(b) develop diagonal tensile and compressive stresses within the
joint. Cracks develop perpendicular to the tension diagonal A-B in
the joint and at the faces of the joint where the beams frame into
the joint. Concrete being weak in tension, transverse
reinforcements have to be provided in such a way that they cross
the plane of failure to resist the diagonal tensile forces [Uma and
Meher Prasad (2005)].
Design of Beam-Column Joints

Because the joint block area is small relative to the member

sizes, it is essential to consider localized stress distribution within
the joints. A simplified force system may be adopted in designing
beam to column connections. The quantity of steel required is
calculated on the assumption that steel reaches the design yield
stress and the concrete its design compressive stress. Where
local bearing or bond failure is expected, the lower of the two
capacities for flexural and local failure should be adopted based
on experimental results. It is essential to prevent bond and
anchorage failure within the joints, especially at the external
joints, through proper design and detailing practices.

The principal mechanisms of failure of a beam-column joint are:

– shear failure within the joint
– Anchorage failure of bars, if

anchored within the joint
– Bond failure of beam or column
bars passing through the joint.

As mentioned earlier, the joint has to be designed based on the

fundamental concept that failure should not occur within the
joint; that is, it is strong enough to withstand the yielding of
connecting beams (usually) or columns.

Types of Joints
Typical beam-column joints are grouped as Type 1 and Type 2
joints, as below, as per ACI 352:

Type 1 Joints-these joints have members designed to satisfy

strength requirements, without significant inelastic deformation.
These are non-seismic joints.

Type 2 Joints-these joints have members that are required to

dissipate energy through reversals of deformation into the
inelastic range. These are seismic joints.

Joint Shear and anchorage

Joint shear is a critical check and will govern the size of the
columns of moment resisting frames. To illustrate the procedure,
consider the column bounded by two beams as shown in Fig. 6.
For ductile behavior, it is assumed that the beams framing into
the column will develop plastic hinges at the ends and develop
their probable moment of resistance (Mpr) at the column faces.
This action determines the demands on the column and the beam
column joint.
Hanson and Connor (1967) first suggested a quantitative
definition of RC joint shear, namely that it could be determined
from a free body diagram at mid-height of a joint panel. Fig.7 is
the free body diagram of the joint for calculation of column shear,
Vcol. It is made by cutting through the beam plastic hinges on
both sides of the column and cutting through the column one-half
storey height above and below the joint. In this figure, subscripts
A and B refer to beams A and B on opposite sides of the joint,
and Ve2,A and Ve1,B are the shears in the beams at the joint
face corresponding to development of Mpr at both ends of the
beam. For a typical storey, it is sufficiently accurate to assume
that the point of contraflexure is at the mid-height of column.
Thus, the column shears for sway to right and left (see also
clause 7.3.4 of IS 13920-1993) may be found as
Where hst is the storey height. It has to be noted the probable
(plastic) moment capacity of beams (Mpr,A, and Mpr,B) are
usually calculated by assuming the stress in flexural
reinforcement as 1.25 fy as against 0.87 fy in the moment
capacity calculation. Hence, a factor of 1.4 is used in the
equations given in clause 7.3.4 of IS 13920, which is similar to
Eqn. (1).

Having found the column shear, Vcol, the design horizontal joint
shear Vj can be obtained by considering the equilibrium of
horizontal forces acting on the free body diagram of the joint
shear as shown in Fig. 8. Assuming the beam to have zero axial
load, the flexural compression force in the beam on one side of
the joint may be taken equal to the flexural tension force on the
same side of the joint [Moehle, et al. (2008)].

Thus the joint shear, Vu,j is given by

For an external joint, where the joint has beam on one side of the
joint only, the above equation is written as

The force Tpr is the tension in the reinforcement in the beam at

its probable capacity and is given by

The factor is a is a stress multiplier and = 1 for Type 1 joints,

where only limited ductility is required and = 1.25 for type-2
joints, which require considerable ductility. The value of =1.25 is
intended to account for: (a) the actual yield stress of a typical
reinforcing bar being commonly 10 to 25% higher than the
nominal value; and (b) the effect of strain hardening at higher
strain (ACI 352-02).

Numerous studies have shown the presence of a slab to have a

significant effect on the performance of Type 2 connections [for
example see Durrani and Wight (1987), Durrani and Zerbe
(1987) Ehsani and Wight (1985a), and Wolfgram-French and
Boroojerdi (1989)]. Hence ACI 352-02 recommends including the
longitudinal reinforcement in the slab within the effective width,
in the quantity As to calculate the joint shear force. For the
present, the effective width may be assumed as given in ACI
318:14 or IS 456:2000. For corner and exterior connections
without transverse beams, the effective width is taken as beam
width plus a distance on each side of the beam equal to the
length of the column cross-section measured parallel to the beam
generating the shear (see also Fig. 3.2 of ACI 352-02).

The nominal shear strength of the joint Vn,j should be at least

equal to the required strength Vu,j. Thus,

Where,

Where Aej is the effective shear area of the joint = bjhj, bj is the
effective width of the joint, and hj is the effective depth of the
joint, is the strength reduction factor = 0.85, and is the strength
coefficient, which is dependent on the configuration and
confinement of the joint provided by the beams. The values of
this coefficient for Type 2 connections is provided in Fig. 2 (The
coefficients for Type 1 Connections are about 1.2 to 1.5 times
higher- see Table 1 of ACI 352-02). It has to be noted that Draft
IS 13920 do not define different strengths for roof and typical
floor levels but instead specify using typical values (upper row of
Fig. 2) for all levels. It has to be noted that the NZS 3101 code
suggests a limiting value of irrespective of the confinement
offered by the framing members. Thus, the NZS criterion is based
on the diagonal compression failure of concrete in the joint core,
and hence is assumed to be proportional to the compressive
strength f’c of concrete, whereas the ACI code criterion is based
on the tensile strength of concrete, which is usually expressed as
proportional to vf’c.

It is important to consider the effect of high axial load in the

column on the nominal shear strength of the joint (Hakuto et al.,
2000). Though it is not considered in the ACI and NZS codes,
Eurocode8 (EN 1998-1:2003), includes this important factor
(Uma and Jain, 2006). Recently, Kim and LaFave (2008)
developed joint shear strength models, using Bayesian parameter
estimation method and experimental observations, which are
found to predict joint shear strength more reliably than those
suggested in ACI and NZS codes.

For connections with beams framing in from two perpendicular

directions, the horizontal shear in the joint should be checked
independently in each direction. For the joint to be considered as
fully confined interior joint, the beams on the four faces of the
joint must cover at least three-quarters of the width and depth of
the joint face, where the depth of the joint is taken as the depth
of the deepest beam framing into the joint. If a beam covers less
than three quarters of the column face at the joint, it must be
ignored in determining which coefficient applies as per Fig. 2. For
lightweight concrete frames, ACI 318 code suggests that the
shear capacity from Eqn. 4(b) be multiplied by 0.75. If Eqn. 4(a)
is not satisfied, the size of the column has to be increased.

The area effective in resisting joint shear may not be as large as

the entire cross-sectional area of column since the (web) width of
beam, b, and the column, bc, may differ from each other. The
codes recommend effective joint shear area based on engineering
judgment. Concentric and eccentric joints are shown in Fig.9, and
a comparison of effective joint width, bj, as per different codes is
given in Table 1. ACI 318:2011 suggests that the effective joint
width may be taken as smaller of bc, b + 2x, and b + hc (see Fig.
9 for the definition of these terms). The effective depth of the
joint, hj, may be taken as depth of column, hc, in the considered
direction of shear (see Fig. 9). In ACI 352:2002, the shear
strength of eccentric beam-column connection is reduced by
using a smaller effective width, if the eccentricity of the spandrel
beam with respect to the column centroid exceeds 1/8th of the
column width.
When beams of different width frame into opposite sides of the
column in the direction of loading, b should be taken as the
average of two widths. The average of beam and column widths
usually governs the effective joint shear width, bj, as per ACI
352-02, for RC beam-column connections without joint
eccentricity.
Eccentrically connected beams are found to be the cause of
collapse of a four-storey school building at Hakodate University,
Japan during the 1968 Tokachi-oki earthquake and several
damages to beam-column joints during the 1995 Hyogo-ken
Nanbu earthquake, Japan. However, based on experimental
research, LaFave et al. (2005) and Canbolat and Wight (2008)
determined that the floor slabs significantly reduced the influence
of spandrel beam eccentricity on the behavior of the connection
region under lateral loading. Hence they argue that the effective
joint width suggested in codes is conservative and hence
proposed the following effective width:

After arriving at the design horizontal shear, Vjh, the vertical

shear, Vjv, can be approximated when the columns do not form
plastic hinges, as [Uma and Meher Prasad (2005)]:
Where, hb and hc are the height of beam and column
respectively.

Design of Shear Reinforcement

The role of transverse reinforcement and mechanism of shear

transfer in a beam-column joint for seismic resistance are much
debated and two schools of thoughts prevail. Currently there is
little consensus within the design and research communities as to
declare whether joint hoops serve to confine the core concrete or
to carry joint shear directly. Paulay et al. (1992) proposed shear
transfer mechanisms of a joint as shown in Fig.10, referred to as
diagonal strut mechanism and truss mechanism. They considered
that the strength of the diagonal strut controls the joint strength
before cracking. When the joint shear becomes large, diagonal
cracking occurs in the joint core and the joint reinforcements
come into play; finally the joint fails by crushing of the concrete
in the joint core. Both mechanisms are incorporated in the NZS
3101 code. Thus, NZS 3101 requires a large amount of
transverse reinforcement in a joint to resist a dominant part of
the joint shear by the truss mechanism, relying on the good bond
stress transfer along the longitudinal reinforcement. The use of
larger-diameter and higher-strength bars for beam flexural
reinforcement is limited in NZS 3101 to reduce the bar slippage
within the joint [Kitayama et al., (1991)].

The American codes (ACI 318 as well as ACI 352) assume severe
bond deterioration of the reinforcing bars in the joint and hence
the internal shear forces are resisted only by the diagonal
compressive strut of concrete. Thus, the role of transverse
reinforcement is only to confine the core concrete. These
conflicting concepts about the function of transverse
reinforcement lead to different demand for hoop bars, as well as
disparity in detailing criteria (Hwang et al. 2005). The real
behavior of the structure may be due to the combination of the
diagonal strut and the truss mechanisms with the bond
deterioration of longitudinal reinforcement to a certain degree
during cyclic loading.

Joints Confined by Beams

The behaviour of a beam-column joint is influenced by several

variables which include concrete strength, arrangement of joint
reinforcement, size and quantity of beam/column reinforcement,
bond between concrete and longitudinal bars in beam/column,
and axial load in the column. As per ACI 352-02, for Type 1 joints
the hoop reinforcement can be omitted when the joints are
confined by beams framing into the sides of the column. Such
confinement may be assumed when:
(a) Beams frame into all four sides of the joint and each beam
width is at least 3/4 of the column width and does not leave more
than 100 mm of the column width uncovered on either side of the
beams.
(b) Beams frame into two opposite sides of a joint,
and each beam width is at least three quarters of the column
width, leaving no more than 100 mm of the column width on
either side of the beam. In this case however, horizontal
transverse reinforcement should be provided in the perpendicular
direction.

When such confining beams are not present, ACI 352-02

recommends at least two layers of transverse reinforcement be
provided for Type1 joints, between the top and bottom levels of
longitudinal reinforcement, in the deepest beam framing into the
joint. The primary functions of ties in a tied column are to
restrain the outward buckling of the column longitudinal bars, to
improve bond capacity of column bars, and to provide some
confinement to the joint core.

Confinement Reinforcement

Confinement of the joint core is intended to maintain the integrity

of joint concrete, to improve joint concrete toughness, and to
reduce the rate of stiffness and strength deterioration (ACI 352-
02). Currently, the American as well as Indian (Draft IS 13920)
code provisions emphasize the importance of the confinement of
joint core. The required area of confinement reinforcement in the
joint, when spiral reinforcement is used, as per ACI 352 is given
by Eqn.(7a) and Eqn. (7b) respectively.

(for large columns)

Where ρst = Ag is the gross area of column cross-section and Ak

is the area of concrete core = ,Asp is the area of spiral
reinforcement, Dk is the diameter of core measured to the
outside of the spiral or hoop, f’c is the specified compressive
strength of concrete cylinder, fyt is the yield strength of steel (of
circular hoop or spiral), and s is the pitch of spiral or spacing of
hoops.

The Draft IS 13920 provisions are similar and given by Eqn. (8a)
and Eqn. (8b).
 (for large columns)

Where fck is the characteristic compressive strength of concrete

cube.

When rectangular hoops are used, the ACI provisions are as per
Eqn. (9)

For large columns,

Draft IS 13920 provisions are as per Eqn.(10).

Where, h is the longer dimension of the rectangular confining

hoop measured to its outer face (should not exceed 300 mm),
and gives area of one leg only. For large columns
It has to be noted that both the codes suggest the column
confinement steel to be continued into the joint also.
For type 2
joints, when the joint is confined by beams, transverse
reinforcement equal to at least half the confining reinforcement
required at the end of the column should be provided within the
depth of the shallowest framing member. The spacing of the
hoops should not exceed 150 mm (See Clause 8.1 of IS 13920) in
seismic joints. Hwang et al. (2005) found experimentally that the
ACI requirement of providing hoops to confine the joint is
unnecessary and difficult to construct. Their tests indicated that
hoop reinforcement with wider vertical spacing of up to 300 mm
could be used without significantly affecting the joint
performance. They also developed a softened strut-and-tie (SST)
model to design the hoops.

Spacing requirement for horizontal and vertical transverse

reinforcement as per ACI and NZS codes are compared in Table 2
(see Subramanian (2014) for the definition of the terms used).
The ties within the joint should be provided as closed hoops with
the ends bent as 135o hooks. For single leg cross-tie most of the
codes suggest 135o bend at both ends, though ACI recommends
alternate placement of 90o hook on opposite faces of the column
for easy constructability. A comprehensive comparison of the
provisions of ACI, NZS and Eurocode 8 codes on beam-column
joints is provided by Uma and Jain (2006) and Joshi (2001).

When wide beams are used, confining reinforcement should be

provided through the joint to provide confinement for longitudinal
beam reinforcement outside the column core, if such confinement
is not provided by a beam framing into the joint. In the exterior
and corner joints, all the 135o hooks of the cross-ties should be
along the outer face of the column (see clause 8.1.4 and 8.1.5 of
IS 13920).

For best joint behavior, the longitudinal column bars should be

uniformly distributed around the perimeter of the column core.
Au, et al. (2005) Lu, et al. (2012) and Bindhu et al. (2009),
based on their experimental results found that additional diagonal
bars along the column or beam within the joint region result in
additional strength and ductility of beam-column joint.

Anchorage of Bars at Joints

In interior joints, the flexural reinforcement in beam entering one

face of the joint is usually continued through the joint to become
the flexural steel for the beam entering the opposite side.
However, in exterior or corner joints, one or more number of
beams will not continue beyond the joint, and hence it is difficult
to anchor the beam bars within the column width. For the Type 1
joints, the critical section for development of yield strength of the
beam bars may be taken at the face of the column. However,
during seismic loading, moment reversals take place at beam-
column connections that cause stress reversals in beams, column,
and slab longitudinal reinforcement at the connection. Due to
such stress reversals, spitting cracks along the outer column
cover develops, subsequently separating the cover concrete from
the column core (Paulay and Priestley, 1992). Hence, ACI 352-02
suggests that the critical section for development be taken at the
face of the confined column core. But Paulay and Priestly (1992)
states that this assumption is satisfactory only in elastic joints,
where yielding of beam bars at the face of the column is not
expected.
When plastic hinge develops in the beams adjacent to the joint,
the top bars of beams may go into the strain hardening range
and yielding may penetrate into the joint core with simultaneous
bond deterioration [Paulay and Priestly (1992)]. A splitting crack
may appear along the bar as shown in Fig. 11(a) and the bond
stress distribution around the bar will not be uniform.

Column dimensions seldom permit providing the development

length by straight embedment alone; hence hooks are often
required to anchor negative (top) beam reinforcement, at the far
side of exterior beam-column joints. Usually 90o hooks are used,
with the hook projecting downwards, and extending beyond the
mid-depth of the joint, so that joint diagonal compression strut,
as shown in Fig. 10, can be developed. If the bottom bars are
also required to develop their strength at the face of the joint,
they should also be provided with 90o hooks, which should be
turned upward and extended towards the mid-depth of the joint
[Hakuto et al. 2000]. Hooks should be located within 50 mm of
the confined core, as shown in Fig. 11(c). If the beam has more
than one layer of flexural reinforcement, the tails of subsequent
layers of bars should be located within 3db of the adjacent tail.
A comparison of the different cde provisions for hooks is provided
in Table 3. The development length equation in ACI and NZS
codes consider the beneficial effect of anchoring the bar in the
well confined joint core, and also the adverse effect of the bar
being subjected to load reversals during earthquake.

Use of Headed Reinforcement

The use of hooks in external beam-column joints often result in

steel congestion, difficult fabrication, and construction, and
greater potential for poor concrete placement. Moreover, cyclic
loading tends to degrade the anchorage capacity due to slip.
Anchor plates or heads, either welded or threaded to the
longitudinal bar, can be used as an alternative to the use of
hooked bars in exterior beam-column joints [Wallace et al.
(1998), Chun et al. (2007)]. The use of headed bars offers a
potential solution to the problems posed by hooked bars and may
ease fabrication, construction, and concrete placement. Based on
the works of Bashandy (1996), Wallace et al. (1998), and Wright
and McCabe (1997), ACI 352:2002 was revised to allow the use
of headed bars with a development length Ldh equal to 75
percent of the development length of a standard 90o hooked bar.
The head of bar should be located within 50 mm from the back of
confined core, as per ACI 352:2002, and as shown in Fig. 11(c).

When the side cover at free face of the joint is less than 3 db, ACI
352:2002 suggest that each head should be transversely
restrained by a stirrup that is anchored in the joint. As Type 2
connections may experience significant inelastic deformations,
the hoop leg should be designed for 50 percent of the yield
strength of the bar being developed. In Type 1 joints, it can be
designed for 25 percent of the yield strength of the bar being
developed. If the side cover is greater than 3db, the restraining
force should be determined using the CCD (concrete capacity
design) approach. However, minimum transverse reinforcement
should always be provided.

Head size with a net area of three to four times the bar area was
found to be sufficient to anchor the beam reinforcement
effectively (Chun et al. (2007). A new model that accounts for
head bearing and bond capacity of the anchored bars was
proposed by Chun et al. (2009). Various strut-and-tie models
have also been proposed to consider material strength and also
the structural configuration of the system [Hong et al (2007), and
Thompson et al. (2002)].

Beam and Column Bars Passing Through Interior Joint

The uneven distribution of bond stress around a bar may affect

the top beam bars, the underside of which may be embedded in
inferior quality concrete, due to sedimentation. The following
factors influence the bond response of bars at the beam-column
joint [Paulay and Priestly (1992)]:

1. Confinement, transverse to the direction of embedded bar,

which will significantly improve bond performance under seismic
conditions.


2. The bar diameter, db, has a significant effect on the bond

strength in terms of bond stress.


3. The bar deformations (i.e. the area of ribs of deformed

bars), which improve resistance against slip and hence increase
the bond strength.


4. The clear distance between the bars affect the bond strength
moderately.


5. The compression strength of concrete is not a significant

parameter.
The experimental research has revealed that a displacement
ductility factor of at least µ? = 6 or inter-storey drift of at least
2.5 percent can be achieved if the largest bar diameter in beam,
db, to column depth, hc, ratio at an interior joint was limited to
the following [Paulay et al. (1992):

Paulay and Priestley (1992) also recommend that the average

design bond stress should be around 1.2vfck. Considering several
factors that affect the top bar behavior and assuming that the bar
stress in compression does not exceed fy, Paulay and Priestley
(1992) modified the basic limitation of db/hc as:

Where kj1 is a product of several factors discussed above. Clause

10.4.6.6 of the NZS 3101:2006 code approximates the
product kj kji in the range 2.86 to 3.6.

The purpose of the recommended value of h/db is to limit

slippage of beam and column bars through the joint. ACI
352:2002 has the following limitation:

For Type 2 Joints:

For Type 1 joints no limitation is made in ACI 352:2002.
Interestingly, ACI 352:2002 has a limitation for the height of
beams also as

Clause 7.1.2 of Draft IS 13920 stipulates the following

hc = 300 mm

It is important to note that this length (hc) is not sufficient to

fully anchor the bars in tension, but delay the deterioration of
bond between the bars and concrete in the joint. ACI 352:2002
states that bar slippage is likely to occur with the 20db dimension
of interior column, which will considerably reduce the stiffness
and energy dissipation capacity of the connection zone (As per
ACI 352, a column size of 32db would be necessary to reduce the
slip substantially). Larger development lengths are highly
desirable, especially when the joint is subjected to high shear
stresses and when the column-to-beam flexural strength ratio is
low [ACI 352:2002, Leon (1989), Jirsa (1991), Zhu and Jirsa
(1983)]. Tests on half-scale specimens have shown that
anchorage lengths of 24 to 28 times bar diameter performed
better than those with 16-to 20 times bar diameter. Due to the
smaller size of columns currently being used in India, compared
to those used in USA and New Zealand, an h/db limitation of 15
was chosen in IS 13920 [Jain et al. (2006)].

Larger beam-to-column flexural strength ratios improve the

behavior of connections considerably; to avoid formation of
plastic hinges in joints, the flexural strength ratio should not be
less than 1.4 [Ehsani and Wight (1985b)]. It has to be noted that
NZS 3101 code stipulates beam-to-column flexural strength ratio
of 1.4, whereas ACI 318 and IS 13920 stipulate such ratio of 1.2
and 1.1, respectively.

Summary

The performance of framed structures depends not only upon the

individual structural elements but also upon the integrity of the
joints. However, despite the significance of the joints in
sustaining large deformations and forces during earthquakes,
specific guidelines are not explicitly included in Indian codes of
practice until recently for their design and detailing.

The types of beam-column joints in a moment resistant frame,

can be classified as (a) interior joint, (b) exterior joint,(c) corner
joint, and (d) knee joint. The requirements for the satisfactory
performance of joints are listed. Some of the incorrect detailing
practices adopted in India are highlighted. Various types of joints
and their behaviour under lateral forces are discussed.

When beam-column joints in a frame are subjected to lateral

loading, the equilibrating forces from beams and columns develop
diagonal tensile and compressive stresses within the joint. Cracks
develop perpendicular to the tension diagonal in the joint and
transverse reinforcements have to be provided to resist the
diagonal tensile forces. The principal mechanisms of failure of a
beam-column joint are: (a) Shear failure within the joint, (b)
Anchorage failure of bars, if anchored within the joint, and (c)
Bond failure of beam or column bars passing through the joint.

For design beam-column joints can be categorized as non-seismic
joints (Type 1) and seismic joints (Type 2). The joint shear could
be determined from a free-body diagram at mid-height of a joint
panel. By assuming the point of contraflexure at the mid-height
of column, the column shears could be determined. The tensile
force in reinforcement is taken as Asfy where is taken as 1.25 for
seismic joints. Test data shows that the longitudinal
reinforcement in the slab within the effective width of beam has
to be considered while calculating the joint shear force. The
nominal shear strength is prescribed in ACI 352:2002 or Draft IS
13920 for different joint configurations, based on experimental
results. This nominal shear (with a strength reduction factor)
should be greater than the calculated joint shear force.
Eccentricity of joints is considered by defining effective shear area
of joint.
The role of transverse reinforcement and mechanism of
shear transfer in a beam-column joint are considered differently
in American and New Zealand codes. American and Indian codes
consider that the transverse reinforcement is provided to confine
the joint, and hence suggest the same reinforcement as in
column ends. When joints are confined by beams (with width
greater than or equal to 3/4th of column width) on all the four
sides of the column, no joint reinforcement is required for Type 1
joints and only 50 percent of confining reinforcement need to be
provided for Type 2 joints with a maximum spacing of 150 mm.
The ties within the joint should be provided as closed hoops with
the ends bent as 135o hooks. For best joint behavior, the
longitudinal column bars should be uniformly distributed around
the perimeter of the column core.

In exterior beam-column joints, column dimensions seldom

permit providing the development length by straight embedment
alone; hence hooks are often required to anchor negative (top)
beam reinforcement, at the far side of joint. These 90o hooks
should project inward, to facilitate the formation of diagonal
compressive strut in the joint. Similarly the bottom bars of beams
should be projected upward into the external beam-column joint.
Such hooks should be located within 50 mm of the confined core.
To reduce the congestion of reinforcement in beam columns,
headed bars could be used instead of 90o hooks. To limit slippage
of beam and column bars through the joint, ACI 352 :2002
suggest that hc/db and hb/db should be greater than or equal to
20 (15 as per Draft IS 13920). While detailing beam-column joint
reinforcement constructability issues should be considered
[Murthy(2005) suggests methods for constructability of ties in
beam-column joints].

Acknowledgements

The contents of the paper are extracted from the author’s book
Design of Reinforced Concrete Structures published by Oxford
University Press, New Delhi, which may be consulted for design
examples.

References*

1. ACI 318-2014, Building Code Requirements for Structural

Concrete and Commentary, American Concrete Institute,
Farmington Hills, 473 pp.
2. ACI SP 123 (1991), Jirsa, J.
O.(Editor), Design of Beam-Column Joints for Seismic Resistance,
American Concrete Institute, Detroit, MI, 518 pp.
3. ACI-ASCE
Committee 352. (2002), 352R-02: Recommendation for design of
beam-column joints in monolithic reinforced concrete structures,
American Concrete Institute, Farmington Hills, MI., 37 pp.

4. Arslan, M.H. and H.H. Korkmaz, ‘What is to be learned from
damage and failure of reinforced concrete structures during
recent earthquakes in Turkey?’, Engineering Failure Analysis, Vol.
14, No.1, Jan.2007, pp. 1-22.
5. Au, F.T.K., K. Huang, and H.J.
Pam, (2005), “Diagonally-reinforced beam-column joints
reinforced under cyclic loading”, Proc. Institution of Civil
Engineers- Structures & Buildings, Vol. 158, No.1, pp. 21-40.

6. Bashandy, T. R., (1996), ‘Application of Headed Bars in
Concrete Members’, PhD dissertation, The University of Texas at
Austin, Dec., 303 pp.
7. Bindhu, K.R., K.P. Jaya, K.P. and V.K.
Manicka Selvam, (2009). ‘Behaviour and Strength of Exterior
Joint Sub-assemblage Subjected to Reversal Loadings’, The
Indian Concrete Journal, Vol. 83, No. 11, Nov., pp.9-20.

8. Canbolat, B.B., and Wight, J. K. (2008). ‘Experimental
Investigation on Seismic Behavior of Eccentric Reinforced
Concrete Beam-Column-Slab Connections,’ ACI Structural
Journal, Vol. 105, No. 2, Mar.-April, pp. 154-162.
9. Chun, S.-
C., S.-H. Lee, T. H.-K. Kang, B. Oh, and J.W. Wallace, (2007)
‘Mechanical Anchorage in exterior Beam-Column Joints Subjected
to Cyclic Loading, ACI Structural Journal, Vol.104, No.1, Jan.-
Feb. 2007, pp.102-112.
10. Chun, S.-C., B. Oh, S.-H. Lee, and
C.J. Naito,(2009) ‘Anchorage Strength and Behavior of headed
Bars in exterior Beam-Column Joints, ACI Structural Journal,
Vol.106, No.5, Sept.-Oct. 2009, pp. 579-590.
11. Durrani, A.
J., and Wight, J. K. (1987). ‘Earthquake Resistance of Reinforced
Concrete Interior Connections Including a Floor Slab,’ ACI
Structural Journal, Vol. 84, No. 5, Sept.-Oct., pp. 400-406.

12. Durrani, A. J., and Zerbe, H. E. (1987). ‘Seismic Resistance
of R/C Exterior Connections with Floor Slab,’ Journal of Structural
Engineering, ASCE, Vol. 113, No. 8, Aug., pp.1850-1864.

13. Ehsani, M. R., and Wight, J. K. (1985a). ‘Effect of
Transverse Beam and Slab on the Behavior of Reinforced
Concrete Beam-to-Column Connections’, ACI Journal, Vol. 82, No.
2, Mar.-Apr., pp. 188-195.
14. Ehsani, M. R., and Wight, J. K.
(1985b). ‘Exterior Reinforced Concrete Beam-to-Column
Connections Subject to earthquake-Type Loading ‘, ACI Journal,
Vol. 82, No. 4, July-Aug., pp. 492-499.
15. EN 1998-1:2003,
Eurocode 8: Design of structures for earthquake resistance – Part
1: General rules, seismic actions and rules for buildings,
European Committee for standardization, Brussels, 2003, 215 pp.

16. Engindeniz, M., L.F. Kahn, and A.-H. Zureick, (2005)
‘Repair and Strengthening of Reinforced Concrete beam-Column
Joints: State of the art, ACI Structural Journal, Vol.102, No.2,
Mar.-April 2005, pp.187-197.
17. Hakuto, S., R. park, and H.
Tanaka, ‘Seismic Load Tests on Interior and Exterior Beam-
Column Joints with Substandard Reinforcing Details’, ACI
Structural Journal, Vol. 97, No.1, Jan.-Feb. 2000, pp. 11-25.

18. Hanson, N.W., and H.W. Connor (1967), ‘Seismic
Resistance of Reinforced Concrete Beam-Column Joints”, Journal
of Structural Div., ASCE, Vol. 93, No. ST5, May 1967, pp. 533-
559.
19. Hong, S.-G., S.-C. Chun, S.-H. Lee, and B. Oh, Strut-
and-Tie Model for Development of Headed Bars in Exterior Beam-
column Joints, ACI Structural Journal, Vol.104, No.5, Sept.-Oct.
2007, pp. 590-600.
20. Hwang, S-J., H-J. Lee, T.-F. Liao, K.-C.
Wang, and H.-H. Tsai, (2005), ‘ Role of hoops on Shear strength
of Reinforced Concrete Beam-Column Joints’, ACI Structural
Journal, Vol.102, No.3, May-June 2005, pp.445-453.
21. IS
456:2000, Indian Standard code of Practice- Plain and Reinforced
Concrete, Fourth Revision, Bureau of Indian standards, New
Delhi, July 2000, 100 pp.
22. IS 13920:1993, Indian Standard
code of Practice- Ductile Detailing of Reinforced Concrete
Structures Subjected to Seismic Forces, Edition 1.2, Bureau of
Indian standards, New Delhi, March 2002, 16 pp.
23. IS 13920
Draft (http://www.iitk.ac.in/nicee/IITK-GSDMA/EQ11.pdf

24. IS 1893 (Part 1):2002, Criteria for Earthquake Resistant
Design of Structures, Part 1 General Provisions and Buildings,
Fifth Revision, Bureau of Indian standards, New Delhi, June 2002,
39 pp.
25. Jain, S.K., R.K. Ingle, and G. Mondal, (2006).
‘Proposed Codal provisions for Design and Detailing of Beam-
Column Joints in Seismic Regions, The Indian concrete Journal,
Vol. 80, No. 8, Aug. 2006, pp. 27-35.
26. Jain, S.K., and C.V.
R. Murty, (2002) ‘Proposed Draft Provisions and Commentary on
Ductile Detailing of RC Structures Subjected to Seismic Forces (IS
13920’, Document No. IITK-GSDMA-EQ11-V4.0, IITK-GSDMA
Project on Building Codes, 67 pp.
http://www.iitk.ac.in/nicee/IITK-GSDMA/EQ11.pdf -Accessed on
17th July 2013).
27. Jirsa, J.O. (1991), Editor, ACI SP-123-
Design of Beam-Column Joints for Seismic Resistance, American
Concrete Institute, Michigan, 518 pp.
28. Joshi, D.S. (editor),
Design of Reinforced Concrete Structures for Earthquake
Resistance, Indian Society of Structural Engineers, Mumbai,
2001.
29. Kim, J., and J.M. LaFave, (2008) ‘Probabilistic Joint
Shear Strength Models for design of RC Beam-Column
Connections, ACI Structural Journal, Vol. 105, No. 6, Nov.-Dec.
2008, pp. 770- 780.
30. Kitayama,K., S. Otani, and H.
Aoyama,(1991), Development of Design Criteria for RC Interior
Beam-Column Joints, in ACI SP-123 Design of Beam-Column
Joints for Seismic Resistance, James O. Jirsa, Editor, American
Concrete Institute, Michigan, pp. 97 – 123.
31. LaFave, J.M.,
J.F. Bonacci, B. Burak, and M. Shin, ( 2005), ‘Eccentric Beam-
Column Connections’, Concrete International, ACI, Vol. 27, No. 9,
Sept. 2005, pp. 58-62.
32. Leon, R. T., “Interior Joints with
Variable Anchorage Lengths,” Journal of Structural Engineering,
ASCE, Vol. 115, No. 9, Sept. 1989, pp. 2261-2275.
33. Lu,
X.,T.H. Urukap, S.Li, and F. Lin, (2012) ‘Seismic Behavior of
Interior RC Beam-Column Joints with Additional Bars under Cyclic
Loading’, Earthquakes and Structures, Vol. 3, No. 1, pp. 37-57.
(http://technop.kaist.ac.kr/samplejournal/pdf/eas0301003.pdf –
Accessed 13th June 2013).
34. Mitra, N., and L.N. Lowes,
Evaluation, Calibration, and Verification of a Reinforced concrete
Beam-Column Joint Model, Journal of Structural Engineering,
ASCE, Vol. 133, No. 1, Jan. 2007, pp.105-120.
35. Moehle,
J.P., J.D. Hooper, and C.D. Lubke, (2008)’ Seismic Design of
Reinforced concrete Special Moment Frames: A Guide for
Practicing Engineers’, NEHRP Seismic Design Technical
Brief No.1, NIST GCR 8-917-1, National Institute of Standards
and Technology, Gaithersburg, MD, USA, 27 pp.
36. Murty,
C.V.R., Earthquake Tips-Learning Earthquake Design and
Construction, Dept. of Civil Engg., IIT Kanpur & Building Material
and Technology Promotion Council, Ministry of Urban
Development & Poverty Alleviation, Govt. of India, Sept. 2005, 48
pp.
37. NZS 3101:2006, Concrete Structures Standard, Part
1: The Design of Concrete Structures, Part 2: Commentary on
The Design of Concrete Structures, Standards Council of New
Zealand, Wellington.
38. Park, R., and Paulay, T. (1975),
Reinforced Concrete Structures, John Wiley & Sons, New York,
769 pp.
39. Paulay, T., R. Park, and M. J. N. Priestley (1992).
‘Reinforced Concrete Beam-Column Joints under Seismic Actions,
ACI Journal, Vol. 75, No. 11, Nov. 1978, pp. 585-593.

40. Paulay, T. and M. J. N. Priestley 1992, Seismic Design of
Reinforced Concrete and Masonry Buildings, John Wiley and Sons,
New York, 744 pp.
41. Rai, D.C. and Alpa Sheth, E-conference
on Indian seismic Codes, The Indian Concrete Journal, Vol. 76,
No.6, June 2002, pp.376-378.
42. Saatcioglu, M., N.J. Gardner,
and A. Ghobarah, (2001) “ 1999 Turkey Earthquake/Performance
of RC structures, Concrete International, ACI, Vol. 23, No.3, pp.
47-56
43. Subramanian, N., Design of Reinforced Concrete
Structures, Oxford University Press, New Delhi, 2014, 857 pp.

44. Subramanian, N., and D.S. Prakash Rao, “Design of Joints
in RC Structures with Particular Reference to Seismic Conditions,
The Indian Concrete Journal, Vol. 77, No. 2, Feb. 2003, pp. 883-
892.
45. Thompson et al. (2002)
46. Uma, S.R. and A.
Meher Prasad, (2005) Seismic Behaviour of Beam Column Joints
in Reinforced Concrete Moment Resisting Frames, Document No.
IITK-GSDMA-EQ31-V1.0, IITK-GSDMA Project on Building Codes,
29 pp. http://www.iitk.ac.in/nicee/IITK-
GSDMA/EQ31.pdf Accessed on 17th July 2013.
47. Uma, S.R.
and S. K. Jain, (2006), Seismic Design of Beam Column Joints in
RC Moment Resisting Frames- Review of Codes, Structural
Engineering and Mechanics, Vol. 23, No.5, pp. 579-597 (also see:
http://www.iitk.ac.in/nicee/IITK-GSDMA/EQ32.pdf Accessed on
17th July 2013).
48. Wallace, J. W., S.W. McConnell, P. Gupta,
and P.A. Cote, (1998), ‘Use of Headed Reinforcement in Beam-
Column Joints Subjected to Earthquake Loads’, ACI Structural
Journal, Vol. 95, No. 5, Sept.-Oct., pp. 590-606.
49. Wolfgram-
French, C., and Boroojerdi, A., 1989, ‘Contribution of R/C Floor
Slab in Resisting Lateral Loads,’ Journal of Structural Engineering,
ASCE, Vol. 115, No. 1, Jan., pp. 1-18.
50. Wright, J. L., and
S.L. McCabe,(1997) ‘The Development Length and Anchorage
Behavior of Headed Reinforcing Bars’, SM Report No. 44,
Structural Engineering and Engineering Materials, University of
Kansas Center for Research, Lawrence, Kans., Sept., 147 pp.

51. Zhu, S., and Jirsa, J. O., “Study of Bond Deterioration in
Reinforced Concrete Beam-Column Joints,” PMFSEL Report No.
83-1, Department of Civil Engineering, University of Texas at
Austin, TX, July 1983.