You are on page 1of 12

ACI STRUCTURAL JOURNAL

TECHNICAL PAPER

Title No. 111-S12

Predicting Post-Punching Shear Response of Slab-Column


Connections
by Farshad Habibi, William D. Cook, and Denis Mitchell
The response of slab-column connections after a punching shear
failure is investigated. An analytical model for predicting the postpunching shear response of slab-column connections is presented
that accounts for the individual contributions of each layer of top
reinforcement and each layer of the structural integrity reinforcement. The predictions are compared with experimental results and
the results obtained by the CSA A23.3 Standard design method.
There is a good agreement between the predicted results and the
experimental results.
Keywords: breakout strength; post-punching shear; pullout; slab-column
connections; structural integrity; two-way slabs.

INTRODUCTION
Two-way slabs must be designed to have adequate flexural
and punching shear resistance. There have been many examples of brittle failures due to punching shear. These failures
have occurred during construction, in severe earthquakes, due
to corrosion of the top reinforcement, and due to construction errors.1-3 If the slab-column connection does not have
a secondary defense mechanism, then after punching shear
failure, the loads are transferred to adjacent supports. This
redistribution of moments and shears may result in punching
shear failures at the adjacent slab-column connections,
leading to progressive collapse. After the initial punching
shear failure, the top flexural reinforcement crossing the
punching shear cone and the bottom bars that are adequately
anchored in the column can provide a link between the slab
and the column. This reinforcement can play a significant
role in providing a secondary load-carrying mechanism after
initial failure and can prevent progressive collapse.
Previous research on the behavior of slab-column connections has typically focused on the initial punching shear
failure rather than on the post-punching behavior. Hawkins
and Mitchell4 described the difference in the post-punching
behavior of two-way slabs with and without bottom slab reinforcement properly anchored into the column. Figure 1 illustrates the failure of a slab-column connection that contains
top reinforcement only. After the punching shear failure, the
top bars rip out of the top concrete surface, offering little
resistance and leading to a complete loss of support at the
connection. Figure 2 illustrates the benefit of providing
bottom slab reinforcement that is continuous through the
column. The specially detailed bottom bars provide the
slab an ability to transfer shear to the column and prevent
progressive collapse. Hawkins and Mitchell4 suggested that
the post-punching shear capacity of a slab could be estimated using tensile membrane equations or, alternatively,
the capacity could be taken as not greater than one-half

ACI Structural Journal/January-February 2014

Fig. 1Slab-column connection with top bars only: (a)


ripping out of top bars after punching shear failure; and (b)
loss of support with top bars ineffective.

Fig. 2Slab-column connection containing top bars


and continuous bottom reinforcement: (a) post-punching
resistance provided by both top and bottom bars; and (b)
post-punching resistance provided by continuous bottom
reinforcement.
of the yielding strength of all the bottom reinforcing bars
passing through the column.
Regan et al.5 investigated the role of bottom reinforcement
passing through the column on the post-punching response
of slab-column connections. They compared the response of
a 100 mm (3.94 in.) thick slab with three 8 mm (0.31 in.)
diameter bottom reinforcing bars in each direction with a
slab without bottom reinforcement and concluded that
the presence of bottom reinforcement is fundamental in
achieving post-punching resistance of a slab.
McPeake6 carried out an experimental program on thin
slabs to study the effect of bottom reinforcement on the
post-punching response of slab-column connections. It was
observed that the bottom bars had considerable contribution
in the form of tensile membrane action, rather than dowel
action, at large deflections.

ACI Structural Journal, V. 111, No. 1, January-February 2014.


MS No. S-2012-040.R1, doi:10.14359.51686436, was received June 27, 2013, and
reviewed under Institute publication policies. Copyright 2014, American Concrete
Institute. All rights reserved, including the making of copies unless permission is
obtained from the copyright proprietors. Pertinent discussion including authors
closure, if any, will be published ten months from this journals date if the discussion
is received within four months of the papers print publication.

123

Mitchell and Cook7 reported on tests of 1/4-scale models


of two-way slab structures and recommended a design equation that was adopted in the 1984 CSA Standard8 for the
design of minimum bottom reinforcement based on a tensile
membrane model. The 1984 CSA Standard8 required that
a minimum area of structural integrity reinforcement, Asb,
be placed through columns or support reaction areas and
be made effectively continuous in each span direction. The
minimum area was given by
Asb =

0.5ws l 2 l n
s fy

(1)

where ws is the total specified load per unit area, but not less
than twice the slab self-weight per unit area; ln is the clear
span in the direction moments are being considered measured
face-to-face of supports; l2 is the span length transverse to ln
measured center-to-center of supports; s is the resistance
factor for reinforcing bars (0.85); and fy is the specified yield
strength of reinforcement.
Mitchell9 proposed a simple equation to determine the
amount of structural integrity reinforcement that was
adopted by the 1994 CSA A23.3 Standard10 to design structural integrity reinforcement. This equation gives the total
amount of bottom reinforcement, Asb, connecting the slab
to the column or column capital on all faces of the periphery
of the column or column capital as
Asb

2Vse
fy

(2)

where Vse is the shear transmitted to the column or column


capital due to specified loads, but not less than the shear
corresponding to twice the self-weight of the slab.
The 2011 ACI Code11 requires that the bottom bars in
the column strip be lapped with Class B tension lap splices
and that at least two bottom bars or wires (integrity steel)
in each direction must pass within the region bounded by
the longitudinal reinforcement of the column and shall
be anchored at exterior supports. This typically results in
considerably smaller amounts of integrity reinforcement
passing through the column than that required by the 1994
and 2004 CSA A23.3 Standards.10,12
Melo and Regan13 carried out an experimental program
aimed at predicting the post-punching resistance of slabcolumn connections. They concluded that the post-punching
resistance of the structural integrity reinforcement is limited
by the breakout of the reinforcing bars or by rupture of the
bars rather than by dowel action. They used the ACI Code
for Nuclear Safety Related Concrete Structures14 method
to calculate the breakout resistance of the bottom bars
embedded in the concrete. For the rupture of the bars, they
proposed following equation

124

Vu = 0.44 Asb fu

(3)

where fu is the ultimate strength of the reinforcing steel.


They also indicated that if fu is not known then it can be
assumed that fu = 1.15fy, resulting in the following equation that is identical to the equation used in the CSA A23.3
Standards10,12

Vu = 0.50 Asb f y

(4)

Ruiz et al.15 reported on an experimental program involving


the testing of 125 mm (4.92 in) thick slabs supported on eight
roller bearings with a vertical load applied to a steel plate to
simulate the column reaction. They considered the breakout
resistance of the reinforcing bars and the rupture stress of
the bars in determining the post-punching shear strength.
Ruiz et al.15 used the ACI Code for Nuclear Safety Related
Concrete Structures14 method for the concrete breakout
resistance and accounted for the progressive destruction of
the concrete over the reinforcing bars. To develop a mechanical model capable of predicting the post-punching behavior
of slab-column connections, they assumed that the top mat
of bars in the punching cone region acts together against the
concrete cover. The total breakout resistance of the integrity
reinforcement was calculated for the bars passing through
the column, treating them as having the same depth. The
integrity reinforcement contribution was governed either by
the maximum breakout strength of the concrete above the
bars or by the rupture of the bars.
Habibi et al.16 presented results of experimental studies
on seven two-way slabs to provide additional experimental
verification of the approach taken in the CSA A23.3 Standard12 for the design and detailing of the structural integrity
reinforcement. They also investigated the effects of different
parameters on the post-punching behavior of slab-column
connections, including the thickness of the slab, the length
and distribution of the integrity reinforcement, the influence
of column rectangularity, and the layout of the structural
integrity reinforcement in drop panel regions.
This paper presents an analytical method that has been
developed to predict the post-punching response of slabcolumn connections, taking into account the individual
layers of both the top reinforcement and the structural integrity reinforcement. In this method, the contribution of the
top reinforcing bars is governed by the breakout resistance
of the concrete cover and pullout of the bars after significant deflections. The contribution of the structural integrity
bars is governed by three different failure modes, including
rupture of the bars, concrete breakout of the bars, and pullout
of the bars. The resulting predictions of the complete postpunching responses of different slab-column connections
are compared with experimental results.
RESEARCH SIGNIFICANCE
The analytical method provides a tool for predicting the
complete post-punching shear response of slab-column
connections, including the possible failure modes. The
analysis demonstrates the different roles played by the two
layers of top reinforcement and the structural integrity reinforcement after punching shear failure has occurred. This
ACI Structural Journal/January-February 2014

analysis, together with the experimental results, clearly


demonstrates the importance of properly designed and
detailed structural integrity reinforcement.
MECHANISMS OF POST-PUNCHING RESPONSE
Immediately after an initial punching shear failure, there
is a sudden drop in load, and a different resisting mechanism
is set up involving both the top bars that cross the punching
shear cone region and the bottom structural integrity reinforcement. Figure 3 shows three stages in the response of a
200 mm (7.87 in.) thick slab containing top reinforcement
and structural integrity reinforcement. Figure 3(a) shows
the top surface of the slab soon after punching shear failure
occurred, and Fig. 3(b) shows the slab-column connection
after a number of top bars crossing the punching shear cone
have ripped out of the top surface. At large displacements,
the post-punching shear resistance depends on the bottom
structural integrity reinforcement. The resistance from the
integrity reinforcement is due to the vertical component
of the bars that develop large tensile stresses and become
inclined as the slab deflects (refer to Fig. 3(c)).
Different codes assume different angles for the punching
shear cone, as shown in Fig. 4. The ACI Code11 and the CSA
Standard12 assume a constant angle of 45 degrees (Fig. 4(a)),
resulting in the cone reaching a distance of d from the column
face. The EC2 Code17 assumes that the punching shear cone
extends to 1.5d (34-degree angle) and the CEB-FIP Code18
assumes this distance to be 2d (26.6-degree angle) measured
from the column face (refer to Fig. 4(b) and (c)). It has been
observed in experiments that the failure surface is approximately 45 degrees near the column (refer to Fig. 3(c)), but
becomes flatter near the top surface of the slab, surfacing
approximately 2d from the column face (refer to Fig. 3(a)).
In this study, the failure surface shown in Fig. 4(d) was
assumed to determine the effective number of top bars that
pass through the punching shear failure cone and, hence,
participate in the post-punching response.
After a punching shear failure, both the effective top bars
and the structural integrity reinforcement undergo tension
as the slab deflects. Figure 5 shows the mechanism of shear
transfer after punching shear failure. It is the vertical component of the force in these bars that provides the post-punching
shear resistance. The vertical component is limited by either:
the breakout resistance of the concrete above the bar; pullout
of the reinforcement when the remaining embedded portion
of the bar equals the development length; or rupture of the
bars in tension.
The post-punching shear strength of a slab-column
connection, Vpp, can be calculated as the sum of the contributions of the top bars crossing the punching shear cone,
Vt, and the integrity reinforcement, Vi. It is assumed that
after punching, the dowel action and bending of the bars is
negligible, with tensile stresses in the bars providing the
shear resistance. The exposed length of the bars in the model
includes a concrete damage zone at the locations where the
bar enters the concrete of a length equal to the bar diameter
db (refer to Fig. 5). This debonded length was estimated from
observations of the local concrete damage around the bars
after testing was completed.
ACI Structural Journal/January-February 2014

Fig. 3Stages in the post-punching response of a slabcolumn connection containing structural integrity reinforcement: (a) top surface soon after punching shear; (b) ripping
out of top mat of reinforcement; and (c) inclination of structural integrity reinforcement at large slab displacements.
Based on the equilibrium conditions, the vertical component of the force developed in a reinforcing bar, V, is related
to the tensile stress developed in the bar, fs, and the angle of
inclination of the reinforcement, , by the following equation
V = Asfssin (5)
125

Fig. 4Different assumed angles of punching shear cone: (a) 45 degrees (ACI Code11 and CSA Standard12); (b) 34 degrees
(EC217); (c) 26.6 degrees (CEB-FIP18); and (d) failure plane after punching assumed in this study.
based on an equation proposed by Kunnath et al.19 The relationship is given by

Es s

fs = f y

f + f f u s
y
u
u
u sh

Fig. 5Mechanism of shear resistance after punching shear


failure.
where As is the effective area of the reinforcing bar. The
strain in the bar can be calculated by

s =

1
1
cos

(6)

Applying compatibility of deformations and using Eq. (6),


the deflection of slab can be expressed as

1
= l tan = l tan cos1

1 + s

(7)

where l is the horizontal length of the exposed reinforcing


bar. As shown in Fig. 5, the horizontal exposed length li of
the structural integrity reinforcement is given by

li = d3 + x + 2db (8)

and the horizontal exposed length li of the top bars is given


by

li = 2x + 2db (9)

To determine the relationship between the tensile stress


and strain of the reinforcement, an idealized stress-strain
curve of reinforcing steel has been used. The curve has an
elastic region, a yield plateau, and a strain hardening region
126

if
if
if

y < s sh (10)

sh < s u

s y

where the initial slope of the strain hardening region is Esh


sh
p = Esh u
fu f y
and
CONCRETE BREAKOUT STRENGTH
The ACI Code for Nuclear Safety Related Concrete
Structures14 provides a method for calculating the concrete
breakout strength of an embedded reinforcing bar in concrete
when the bar is loaded in the direction of a free edge. The
breakout strength is taken as the vertical component of the
tensile strength of the concrete acting on the surface of the
concrete breakout cone, and is given by

Vn = Ach fct ,eff =

2
d fct ,eff
2

(11)

where Vn is the nominal concrete breakout strength; Ach is


the horizontal projection of the conical failure surface; fct,eff
is the effective tensile strength of the concrete above the
reinforcing bar; and d is the depth of concrete over the reinforcing bar measured to the center of the bar. It is assumed
that the failure surface is radiating out at 45 degrees from
the location where the bar enters into the concrete (refer to
Fig. 6).
For multiple bars with a center-to-center spacing s greater
than or equal to 2d, the total breakout resistance is equal
to the number of effective bars n times the breakout resistance of a single bar (Eq. (11)). For n bars having a center-toACI Structural Journal/January-February 2014

center spacing of less than 2d, the conical failure surface


is reduced due to the overlapping of the failure cones, as
shown in Fig. 7. The total horizontal projected area of the
conical failure surface for n reinforcing bars with s < 2d is
given by

Ach = n

2
s

d 2 ( n 1) d 2 d sin
2

2
4
(12)

where = cos1(s/2d) is in radians.


To predict the breakout resistance of each layer of the
structural integrity reinforcement, it is necessary to account
for the changing depth of concrete above the bars due to the
influence of the shape of the punching shear failure plane
(refer to Fig. 4) and for the loss of concrete above the integrity bars due to the progressive destruction of the concrete
as the top bars rip out (refer to Fig. 8). Figure 8(a) shows a

section along a single bar with the breakout progressing a


distance of d3 + x from the column face, with the breakout
cone intersecting the punching shear failure plane (that is,
x 2d/3).
As shown in Fig. 8(a), the thickness of the concrete
directly above the bar is x, and it is assumed that in calculating the breakout resistance that the diameter of horizontal
projection of the conical failure surface is 2x.
The concrete breakout resistance can be expressed as the
horizontal projected area, which is a function of x multiplied
by the effective tensile strength of the concrete, as
Vi(x) = Ach(x)fct,eff (13)
To calculate the horizontal projection of the conical failure
surface, Ach(x), for more than one structural integrity bar,
the interaction between the failure surfaces of the breakout
cones must be considered. For n structural integrity bars, the
horizontal projection of the conical failure surface is given
by
2
n x1 A1 ( x1 ) if
2

Ach ( x ) = n x2 2 A1 ( x2 ) if
2
2
n d2 A1 ( d2 ) if
2

2d
2d
2d
<x<
+ d2 cot 2
3
3
3

2d
2d
x
+ d2 cot 2

3
3
(14)

2d
3

where 1 = 45 degrees; 2 = arctan(0.25) = 14 degrees; x1 = x;


2d
2d

x2 = x tan 2 +

3
3
and
Fig. 6ACI Code for Nuclear Safety Related Concrete
Structures14 model for concrete breakout strength above a
single bar assuming a 45-degree failure plane: (a) isometric
view; (b) plan view; and (c) elevation view.

( x) 2 s
A1 ( x ) = 2 ( n 1)
x x sin ( x )
4
2

Fig. 7Concrete breakout failure surface for three reinforcing bars: (a) isometric view; (b) plan view; and (c) elevation view.
ACI Structural Journal/January-February 2014

127

Fig. 8Model to estimate progressive destruction of concrete over structural integrity reinforcing bar.
1 s
cos 2 x
where ( x ) =
0

if
if

where Ach is the horizontal projection of the conical failure


surface, and fct,eff is the effective tensile strength of concrete
in determining the breakout resistance for the top bars,
which can be taken as

s
1

2x

s
>1

2x

fct,eff = 0.55fsp (18)

The term A1(x) provides a correction for the interaction of


the overlapping failure cones when the spacing of the bars s
is less than 2x.
From correlations with the test results,16 where the splitting
tensile strength fsp is known, the effective tensile strength of
the concrete for determining breakout of the structural integrity bars can be taken as

where fsp is the splitting tensile strength of concrete. If the


test results for the splitting tensile strength of concrete, fsp,
is not available, then the effective tensile strength of the
concrete can be estimated by

fct,eff = 0.33 fc (MPa units)


= 4 fc

(psi units)

(19)

fct,eff = 0.85fsp (15)

If fsp is not known, then the effective tensile strength of


concrete can be estimated by15

The reduced values of tensile strength for determining the


breakout resistance of the top bars is due to the fact that the
top concrete cover is in the cracked tension zone of the slab
and is typically subjected to larger shrinkage strains. The
ACI Code for Nuclear Safety Related Concrete Structures14
uses a nominal tensile strength of 4fc psi (0.33fc MPa)
for all design cases, while acknowledging that considerably
higher tensile strengths may be achieved.
It is noted that the use of the 45-degree failure cone,
together with the concrete tensile strengths described previously, gave better predictions than the approach for predicting
concrete breakout given in the 2011 ACI Code.11 The angles
of the breakout failure cones measured on full-scale slabcolumn connections16 are approximately 45 degrees.
For typical top bar spacing and slab concrete covers, the
interaction between the breakout cones need not be considered. Hence the projected area of the breakout cone, Ach(x),
can be calculated by

fct ,eff = 0.50 fc (MPa units)


= 6 fc

(psi units)

(16)

Figure 8(b) shows the situation when the concrete


breakout strength above the structural integrity bars reaches
a maximum. Beyond this stage, the concrete breakout
strength remains constant due to the constant concrete thickness above the bars and, hence, the contribution of these bars
to the shear resistance is also constant.
The same approach used for the structural integrity
reinforcement can be used in determining the progressive breakout of the top bars. The breakout strength of the
concrete cover over a single top bar, with the breakout
progressing to a distance of x from the crack face (refer to
Fig. 5) is equal to
Vt(x)total = Ach(x)fct,eff (17)


128

2 ( x tan 2 )
Ach ( x ) =
d 2
2 1

if
if

x < d1 cot 2

(20)
x d1 cot 2

ACI Structural Journal/January-February 2014

where d1 is the depth of concrete over the top bar measured


to the center of the bar.
PULLOUT OR RUPTURE OF REINFORCING BARS
After punching shear failure occurs, large tensile strains
develop in the reinforcement and the inclinations of the
exposed bars increase with increasing displacement (refer
to Fig. 5). The reinforcement offers resistance over large
displacements until it either ruptures or suffers pullout from
the concrete, at which point the resistance of this reinforcement is assumed to be zero.
At each increment of deflection, the strain in a reinforcing
bar can be determined by the compatibility expression,
Eq.(7). The computed strain in the reinforcing steel is used
to check if rupture of the steel occurs. If rupture occurs,
then the contribution of the steel to the post-punching resistance is zero. If rupture does not occur, then the pullout
resistance of the remaining embedded portion of the bar is
checked. The bar can continue to provide shear resistance
until the remaining embedded length is reduced to the development length of the bar, ld, at which point the bar loses
its anchorage and pullout occurs. Figure 9 illustrates when
pullout is predicted for the top bars and the structural integrity bars (refer to Fig. 9(a) and (b), respectively). In determining the development length, the ACI Code11 design
expression for development length was used, assuming that
the stress in the embedded reinforcement corresponds to the
horizontal component of the force in the exposed inclined
reinforcement. The ACI Code11 design expression for development length is based on the research by Orangun et al.20
and includes an implicit bond strength reduction factor of
0.8. Hence, in checking for pullout failure, corresponding to
the nominal concrete bond strength, the development length
was taken as 0.8 times the development length determined
using the ACI Code11 equation.
SOLUTION TECHNIQUE
To predict the nonlinear response after punching shear has
occurred, the responses of the two top layers of reinforcement and the responses of the two structural integrity layers
of reinforcement are treated separately. For the top steel
contribution, values of x (refer to Eq. (9)) were incremented
and the equilibrium and compatibility relationships were
satisfied until the complete shear-versus-deflection response
was determined. Similarly for the structural integrity layers,
x (refer to Eq. (8)) was incremented to give the load deflection responses of each layer. The complete response of the
combined top and structural integrity reinforcement was
determined by adding the predicted shears for each value of
predicted deflection.
VERIFICATION OF ANALYTICAL MODEL
Figures 10 to 12 show comparisons between the postpunching predictions obtained using the analytical model
and the experimental results for slab-column specimens SS,
RS, and D1 reported by Habibi et al.16 All three specimens
had a mat of 15M top bars (diameter of 16.0 mm [0.63 in.])
and 10M bottom bars (diameter of 11.3 mm [0.44 in.]) and
structural integrity reinforcement as required by the CSA
ACI Structural Journal/January-February 2014

Fig. 9Pullout of reinforcement.


A23.3-04 Standard.12 Flat plate specimens SS and RS had
plan dimensions of 2.3 x 2.3 m (7.55 x 7.55 ft) and a thickness of 200 mm (7.9 in.). Specimen SS had a 225 mm (8.9 in.)
square central column and had two 15M structural integrity
bars passing through the column in each direction protruding
a distance Li of 790 mm (31.1 in.) from the column face.
Specimen RS had a 180 x 270 mm (7.1 x 10.6 in.) rectangular column and had three 15M structural integrity bars in
one direction (Li of 790 mm [31.1 in.]) and two 10M bars (Li
of 600 mm [23.6 in.]) in the other direction. The structural
integrity reinforcement was placed at the same level as the
bottom reinforcement for Specimens SS and RS. Specimen
D1 had a 225 mm (8.9 in.) square column, plan dimensions
of 2.3 x 2.3 m (7.55 x 7.55 ft) and a 160 mm (6.3 in.) thick
slab with a 1.825 x 1.825 m (6.0 x 6.0 ft) drop panel that was
90 mm (3.5 in.) thick. The structural integrity reinforcement
in Specimen D1 was the same as for Specimen SS and was
placed in the bottom of the drop panel with a clear cover of
25 mm (1.0 in.). The top and bottom clear covers were 25
mm (1.0 in.) for all three specimens.
Table 1 gives the average measured reinforcing bar properties and Table 2 gives the average material properties of
the ready mixed concrete at the time of testing used in Specimens SS, RS, and D1. The maximum aggregate size for the
concrete in this series was 20 mm (0.75 in.). The details of
reinforcement for these specimens and those for specimens
tested by others are summarized in Table 3.
Figures 10(a), 11(a), and 12(a) show the predicted contributions of the top bars to the post-punching shear resistance.
The predicted contributions of the structural integrity bars to
the post-punching response are shown in Fig. 10(b), 11(b),
and 12(b). In these figures, the end of the contributions
of the upper layer (Point A) and lower layer (Point B) of
the top bars as well as the upper layer (Point C) and lower
layer (point D) of the structural integrity bars are shown.
It is noted that the top mat of bars becomes ineffective at
a relatively early stage in the post-punching response and
offers little resistance, whereas the structural integrity rein129

Table 1Measured reinforcing steel properties for Specimens SS, RS and D1


Bar size

Area, mm2 (in.2)

fy, MPa (ksi)

y, %

sh, %

Esh, MPa (ksi)

fu, MPa (ksi)

u, %

10M

100 (0.15)

460 (66.7)

0.23

0.75

8400 (1220)

730 (106)

13

15M

200 (0.31)

420 (60.9)

0.21

0.50

8000 (1160)

723 (105)

13

Table 2Measured concrete properties for


Specimens SS, RS, and D1
Slab

fc, MPa (psi)

fsp, MPa (psi)

fr, MPa (psi)

SS

26 (3770)

3.35 (485)

3.46 (502)

RS

30 (4350)

3.16 (458)

3.88 (563)

D1

22 (3190)

2.60 (377)

4.16 (603)

forcement is capable of providing significant post-punching


resistance at very large displacements. The predictions of
the post-punching load-deflection responses of these slabs
are in good agreement with the experimentally determined
responses, including the major behavioral aspects of the
complex progressive damage to the concrete and the highly
nonlinear responses.
Table 3 shows the comparison between the results
obtained by the CSA A23.3 Standard12 design method,
the analytical model, and the test data. It is noted that the
simplified design equation used in the CSA A23.3 Standard12 provides reasonably accurate predictions of the
post-punching shear resistance. There is a good agreement
between the results predicted by the analytical model and the
experimental results. The analytical model described in this
paper is capable of predicting the maximum post-punching
resistance Vpp and the ultimate average displacement u,test,
defined as the deflection reached when the load drops to 80%
of the maximum post-punching shear resistance.

Fig. 10Contributions of top reinforcement and structural


integrity reinforcement in predicting post-punching shear
versus deflection response of Specimen SS. (Note: 1 mm =
0.0394 in.; 1 kN = 0.225 kip.)
130

CONCLUSIONS
The following conclusions can be made based on observations from test results of slab-column connections and from
the behavioral aspects predicted by the analytical model
that was developed to predict the complete response of
slab-column connections after punching shear failures have
occurred:
1. Structural integrity reinforcement can provide significant post-punching shear resistance and deformability.
2. Top slab reinforcing bars, in the vicinity of the column,
provide a limited post-punching shear resistance and become
ineffective at a deflection that is considerably less than that
for the structural integrity reinforcement.
3. An analytical model was developed to predict the postpunching response of slab-column connections. This model
satisfies compatibility of deformation and equilibrium and
includes the nonlinear response of the steel reinforcement.
This model is capable of predicting the complete postpunching response of slab-column connections.
4. The analytical model is capable of predicting the
progressive breakout of each layer of the top reinforcing bars
and the structural integrity bars, the pullout of the top bars
and the structural integrity bars, and the possible rupturing
of the reinforcement.

ACI Structural Journal/January-February 2014

Fig. 11Contributions of top reinforcement and structural


integrity reinforcement in predicting the post-punching
shear versus deflection response of Specimen RS. (Note:
1mm = 0.0394 in.; 1 kN = 0.225 kip.)

Fig. 12Contributions of top reinforcement and structural


integrity reinforcement in predicting post-punching shear
versus deflection response of Specimen D1. (Note: 1 mm =
0.0394 in.; 1 kN = 0.225 kip.)

5. The predicted shear-deflection responses agree reasonably well with the post-punching shear responses of the slabcolumn specimens that were tested. The predicted maximum
post-punching resistance and ultimate deflection agree well
with the large variety of experiments reported by others.

Farshad Habibi is an Engineer at NETRICOM, Mississauga, ON, Canada.


He received his PhD from McGill University, Montreal, QC, Canada. His
research interests include progressive collapse of reinforced concrete
structures and seismic design of structures.

ACI Structural Journal/January-February 2014

AUTHOR BIOS

ACI member William D. Cook is a Research Associate in the Department


of Civil Engineering at McGill University, where he also received his PhD.
His research interests include nonlinear analysis of reinforced concrete
structures and the structural use of high-performance concrete.

131

Table 3Comparison of predicted and experimental results


Integrity
s,
bars mm

L i,
mm

fy,
MPa

Effective
top bars*

Lt ,
mm

fy,
MPa

Vtest,
kN

VCSA,
kN

Vtest
VCSA

Vpp,
kN

Vtest
Vpp

u,test,
mm

u,pred,
mm

u, pred

u,test

Slab

h, mm

fc,
MPa

S1

150

28

2-15M 145
2-15M 145

790
790

457
457

3-15M
1-15M

900
1350

457
457

314

365.6

0.86

290

1.08

183

149

1.22

S2

150

30

2-15M 145
2-15M 145

1000
1000

457
457

3-15M
1-15M

900
1350

457
457

333

365.6

0.91

321

1.04

234

234

1.00

R2

150

33

3-15M
2-10M

77
60

1000
820

457
455

3-15M
1-15M

900
1350

457
457

321

368

0.88

276

1.16

178

192

0.93

SS

200

26

2-15M 120
2-15M 120

790
790

420
420

5-15M
5-15M

1037
1037

420
420

397

336

1.18

381

1.04

188

128

1.46

RS

200

30

3-15M
2-10M

77
60

790
600

420
460

5-15M
5-15M

1010
1055

420
420

360

344

1.04

351

1.02

122

113

1.08

D1

250

22

2-15M 120
2-15M 120

790
790

420
420

8-15M
6-15M

1037
1037

420
420

519

336

1.54

492

1.05

174

160

1.08

S1-B

150

37

3-10M
3-10M

55
55

1037
1037

454
454

6-15M
6-15M

1037
1037

445
445

245

272.4

0.90

270

0.90

75

48

1.56

S1-U

150

37

3-10M
3-10M

55
55

1037
1037

454
454

4-15M
4-15M

1037
1037

445
445

273

272.4

1.00

227

1.20

50

48

1.04

S2-B

150

57

3-10M
3-10M

55
55

1037
1037

454
454

6-15M
6-15M

1037
1037

445
445

298

272.4

1.09

335

0.89

38

51

0.74

S2-U

150

57

3-10M
3-10M

55
55

1037
1037

454
454

4-15M
4-15M

1037
1037

445
445

266

272.4

0.97

282

0.94

54

51

1.05

S3-B

150

61

3-10M
3-10M

55
55

1037
1037

454
454

6-15M
6-15M

1037
1037

445
445

340

272.4

1.25

357

0.95

57

53

1.07

S3-U

150

61

3-10M
3-10M

55
55

1037
1037

454
454

4-15M
4-15M

1037
1037

445
445

281

272.4

1.03

302

0.93

43

53

0.81

PM-9

125

31

2-8
2-8

120
120

616
616

7-8
7-8

601
601

123

121

1.02

135

0.91

42

35

1.20

PM-10

125

31

2-10 120
2-10 120

560
560

7-8
7-8

601
601

159

176

0.90

172

0.92

50

41

1.21

Average

1.04

1.00

1.10

Standard deviation

0.18

0.09

0.22

Coefficient of variation, %

17.3

9.0

20.0

Effective top bars within 2d of column face for each direction.

Specimen reported by Habibi et al.16

Specimen tested by Ghannoum.21

Specimen tested by Ruiz et al.15

Notes: 1 mm = 0.0394 in.; 1 MPa = 0.145 ksi; 1 kN = 0.225 kip.

Denis Mitchell, FACI, is a James McGill Professor in the Department of


Civil Engineering at McGill University. He is a member of Joint ACI-ASCE
Committees 408, Development and Splicing of Deformed Bars, and 445,
Shear and Torsion.

Esh =
fc =
fct,eff =
fct,eff =

ACKNOWLEDGMENTS

The authors acknowledge the financial support provided by the Natural


Sciences and Engineering Council of Canada for funding this research
program. The authors are grateful to those who assisted in the experimental
program, in particular, E. Redl and M. Egberts.

A1

Ach
As
Asb
d
d1
d2

=
=
=
=
=
=

d3
db
Es

=
=
=

132

NOTATION

area correction for interaction of overlapping breakout failure


cones
horizontal projection of conical breakout failure surface
area of reinforcing bar
area of structural integrity reinforcement
effective depth; depth of concrete measured to center of bar
depth of concrete measured to the center of a top bar
depth of concrete measured to the center of a structural integrity
bar
distance measured from bottom face of slab to center of a bar
bar diameter
modulus of elasticity of reinforcement

f s
fsp
fu
fy
h
L i

=
=
=
=
=
=

Lt
l2
ld
li
ln

=
=
=
=
=

lt
n
s
V
VCSA
V i

=
=
=
=
=
=

initial slope of strain hardening stress-strain relationship for


steel
concrete compressive strength
effective tensile strength of concrete used to determine breakout
resistance of structural integrity reinforcement
effective tensile strength of concrete used to determine breakout
resistance of top reinforcement
tensile stress in a bar
splitting tensile strength of concrete
ultimate strength of reinforcing bars
yield strength of reinforcing bars
slab thickness
protruding length of structural integrity reinforcement from
column face
protruding length of top reinforcement from column face
span length transverse to ln, center-to-center of supports
development length of reinforcing bar
horizontal exposed length of structural integrity reinforcement
clear span in direction moments are being considered, face-toface of supports
horizontal exposed length of top reinforcement
number of effective reinforcing bars
center-to-center spacing of bars
vertical component of force developed in a reinforcing bar
CSA Standard prediction for post-punching shear resistance
contribution of structural integrity bars to post-punching shear
resistance

ACI Structural Journal/January-February 2014

Vn
Vpp

= nominal concrete breakout strength


= analytical model prediction for maximum post-punching shear
resistance
Vse = shear transmitted to column or column capital under service
loads
Vt = contribution of effective top bars to post-punching shear
resistance
Vtest = experimental maximum post-punching shear resistance
Vu
= post-punching shear resistance at reinforcement rupture
w s
= total specified service load per unit area
x
= distance from punching shear crack to location where structural
integrity bars enter slab
x
= distance from punching shear crack to location where effective
top bars enter slab

= deflection of slab
u,pred = average deflection when predicted post-punching shear drops to
80% of Vpp
u,test = average deflection when experimental post-punching shear
drops to 80% of Vtest
s
= strain in reinforcing bar
sh
= strain hardening strain of reinforcing bar
u
= ultimate strain of reinforcing bar
y
= yield strain of reinforcing bar
s
= resistance factor for reinforcing bars (equal to 0.85)

= angle of inclination of reinforcement in damaged state

REFERENCES

1. King, S., and Delatte, N. J., Collapse of 2000 Commonwealth Avenue:


Punching Shear Case Study, Journal of Performance of Constructed
Facilities, V. 18, No. 1, 2004, pp. 54-61.
2. Lew, H. S.; Carino, N. J.; and Fattal, S. G., Cause of the Condominium Collapse in Cocoa Beach, Florida, Concrete International, V. 4,
No. 8, Aug. 1982, pp. 64-73.
3. Mitchell, D.; Adams, J.; DeVall, R. H.; Lo, R. C.; and Weichert, D.,
Lessons from the 1985 Mexican Earthquake, Canadian Journal of Civil
Engineering, V. 13, No. 5, 1986, pp. 535-557.
4. Hawkins, N. M., and Mitchell, D., Progressive Collapse of Flat Plate
Structures, ACI Journal, V. 76, No. 7, July 1979, pp. 775-808.
5. Regan, P. E.; Walker, P. R.; and Zakaria, K. A. A., Tests of Reinforced
Concrete Flat Slabs, CIRIA Project RP 220, School of the Environment,
Polytechnic of Central, London, UK, 1979, 217 pp.
6. McPeake, F. A., Post-Punching Resistance of Internal Slab-Column
Connection, BSc honours project, Department of Civil Engineering,
Queens University of Belfast, UK, 1980, 107 pp.

ACI Structural Journal/January-February 2014

7. Mitchell, D., and Cook, W. D., Preventing Progressive Collapse of


Slab Structures, Journal of Structural Engineering, ASCE, V. 110, No. 7,
1984, pp. 1513-1532.
8. CSA A23.3-M84, Design of Concrete Structures, Canadian Standards Association, Rexdale, ON, Canada, 1984, 281 pp.
9. Mitchell, D., Controversial Issues in the Seismic Design of Reinforced Concrete Frames, Recent Developments in Lateral Force Transfer
in Buildings, Thomas Paulay Symposium at La Jolla, CA, Sept. 20-22,
1993, pp.73-93.
10. CSA A23.3-94, Design of Concrete Structures, Canadian Standards Association, Mississauga, ON, Canada, 1994, 199 pp.
11. ACI Committee 318, Building Code Requirements for Structural
Concrete (318-11) and Commentary, American Concrete Institute, Farmington Hills, MI, 2011, 503 pp.
12. CSA A23.3-04, Design of Concrete Structures, Canadian Standards Association, Mississauga, ON, Canada, 2004, 214 pp.
13. Melo, G. S. S. A., and Regan, P. E., Post-Punching Resistance
of Connections between Flat Slabs and Interior Columns, Magazine of
Concrete Research, V. 50, No. 4, 1998, pp. 319-327.
14. ACI Committee 349, Proposed Addition to: Code Requirements for
Nuclear Safety Related Concrete Structures (ACI 349-76), ACI Journal,
V. 75, No. 8, Aug. 1978, pp. 329-335.
15. Ruiz, M. F.; Mirzaei, Y.; and Muttoni, A., Post-Punching Behavior
of Flat Slabs, ACI Structural Journal, V. 110, No. 5, July-Aug. 2013,
pp.801-811.
16. Habibi, F.; Redl, E.; Egberts, M.; Cook, W. D.; and Mitchell,
D., Assessment of CSA A23.3 Structural Integrity Requirements for
Two-Way Slabs, Canadian Journal of Civil Engineering, V. 39, No. 12,
2012, pp. 351-361.
17. Eurocode 2, Design of Concrete Structures Part 1-1: General
Rules and Rules for Buildings, EN 1992-1-1, CEN, Brussels, Belgium,
2004, 225 pp.
18. CEB-FIP MC 90, Design of Concrete Structures, CEB-FIP Model
Code 1990, British Standard Institution, London, UK, 1993, 437 pp.
19. Kunnath, S. K.; Heo, Y.; and Mohle, J. F., Nonlinear Uniaxial Material Model for Reinforcing Steel Bars, Journal of Structural Engineering,
ASCE, V. 135, No. 4, 2009, pp. 335-343.
20. Orangun, C. O.; Jirsa, J. O.; and Breen, J. E., A Reevaluation of Test
Data on Development Length and Splices, ACI Journal, V. 74, No. 3, Mar.
1977, pp. 114-122.
21. Ghannoum, C. M., Effect of High-Strength Concrete on the Performance of Slab-Column Specimens, MEng thesis, Department of Civil
Engineering, McGill University, Montreal, QC, Canada, 1998, 91 pp.

133

Reproduced with permission of the copyright owner. Further reproduction prohibited without
permission.