See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/37468420

Application of Critical Shear Crack Theory to Punching of Reinforced Concrete

Slabs with Transverse Reinforcement

Article  in  Aci Structural Journal · July 2009

Source: OAI

CITATIONS READS

146 1,913

2 authors:

Miguel Fernández Ruiz Aurelio Muttoni

École Polytechnique Fédérale de Lausanne École Polytechnique Fédérale de Lausanne
126 PUBLICATIONS   2,102 CITATIONS    219 PUBLICATIONS   3,273 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Projet de recherche AGB 2002/028 Office fédéral des routes (OFROU - Suisse); Bolsa de Doutoramento FCT/BD/13259/2003 - Portugal View project

Seismic behaviour of slab-column connections View project

All content following this page was uploaded by Aurelio Muttoni on 07 May 2014.

The user has requested enhancement of the downloaded file.

 ACI STRUCTURAL JOURNAL TECHNICAL PAPER
Title no. 106-S46

Applications of Critical Shear Crack Theory to Punching of

Reinforced Concrete Slabs with Transverse Reinforcement
by Miguel Fernández Ruiz and Aurelio Muttoni

The traditional approach of codes of practice for estimating the

punching strength of shear-reinforced flat slabs is based on the
assumption that concrete carries a fraction of the applied load at
ultimate while the rest of the load is carried by the shear reinforcement.
Concrete contribution is usually estimated as a fraction of the
punching strength of members without shear reinforcement. The
ratio between the concrete contribution for members with and
without shear reinforcement is usually assumed constant, independent
of the amount of shear reinforcement, flexural reinforcement ratio,
and bond conditions of the shear reinforcement. The limitations of
such an approach are discussed in this paper and a new theoretical
model, based on the critical shear crack theory, is presented to
investigate the strength and ductility of shear-reinforced slabs. The
proposed approach is based on a physical model and overcomes
most limitations of current codes of practice. Its application to
various punching shear reinforcement systems is also detailed in
the paper and its results are compared to available test data.
Fig. 1—Possible failure modes in flat slab with punching
Keywords: critical shear crack theory; flat slabs; punching shear; shear shear reinforcement: (a) crushing of compression strut;
reinforcement; two-way shear. (b) yielding of shear reinforcement with crack localization
within shear-reinforced zone; (c) punching shear outside
INTRODUCTION shear-reinforced zone; and (d) comparison of current design
Punching shear reinforcement is used to improve both the approaches (ACI 318-051 and EC22) and actual behavior.
punching shear strength and the ductility of flat slabs. Many
punching shear reinforcement systems are currently available.
Dimensioning punching shear reinforcement is usually
Such systems can be distributed in the slab near the columns
performed by checking the punching strength within the
(for example, studs, stirrups), placed on top of the columns
shear-reinforced zone (VR,in). Most codes of practice estimate
(that is, steel shearheads or mushrooms), or be a combination
such strength according to the following format
of the previous systems. In this paper, the behavior and
strength of slabs with distributed shear reinforcement under
monotonic and axis-symmetric loading will be investigated. VR,in = ηc · Vc0 + ηs · Vs0 (2)
Considering distributed shear reinforcement, a reinforced
concrete flat slab may develop three different punching where Vc0 is the punching shear strength of the slab without
failure modes (Fig. 1): crushing of the concrete struts near shear reinforcement and Vs0 is the force that can be carried
the column, punching within the shear-reinforced zone, and by the shear reinforcement within the punching cone at
punching outside the shear-reinforced zone. The governing yielding. Thus, the contribution of concrete results in ηc · Vc0,
failure mode can thus be estimated as the one leading to the whereas the contribution of the shear reinforcement is ηs ·
minimum strength of the slab Vs0, where ηc and ηs are factors whose respective values are
lower than or equal to 1.0.
VR = min(VR,crush;VR,in;VR,out) (1) The contribution of concrete (ηc · Vc0) is reduced with
respect to the punching shear strength without shear rein-
forcement (Vc0). This fact is justified because wider shear
In most approaches and codes of practice, checking the
cracks develop in shear-reinforced slabs, thus reducing the
crushing strength VR,crush is usually performed by limiting
ability of concrete to transfer shear.3 The coefficient ηc is
the maximum punching shear strength with respect to the
usually considered constant. For instance, ACI 318-051
punching strength of slabs without shear reinforcement
proposes ηc = 0.50 and EC22 proposes ηc = 0.75. It should be
(ACI 318-051) or by considering a reduced compressive
noted that most codes of practice give empirical formulations
strength of concrete struts near the column (EC22).
for the contribution of concrete (Vc0). Thus, all problems
Checking the punching strength outside the shear-reinforced
zone (VR,out) allows determination of the zone that has to be
shear-reinforced. A similar formulation to that of punching ACI Structural Journal, V. 106, No. 4, July-August 2009.
MS No. S-2007-405.R1 received July 7, 2008, and reviewed under Institute publication
shear in slabs without transverse reinforcement is typically policies. Copyright © 2009, American Concrete Institute. All rights reserved, including the
used, but the control perimeter2 and/or the shear strength1 is making of copies unless permission is obtained from the copyright proprietors. Pertinent
discussion including author’s closure, if any, will be published in the May-June 2010 ACI
modified to suitable values. Structural Journal if the discussion is received by January 1, 2010.

ACI Structural Journal/July-August 2009 485

Miguel Fernández Ruiz is a Lecturer and Research Scientist at the Ecole Polytechnique
Fédérale de Lausanne (EPFL), Lausanne, Switzerland. He received his diploma in
civil engineering and his PhD from the Polytechnical University of Madrid, Madrid,
Spain in 2001 and 2004, respectively. His research interests include the serviceability
behavior of structures, bond, shear and punching shear, and the modeling of structural
concrete using stress fields.

ACI member Aurelio Muttoni is Professor and Head of the Structural Concrete
Laboratory at the EPFL. He received his diploma and PhD in civil engineering from
the Swiss Federal Institute of Technology, Zurich, Switzerland, in 1982 and 1989,
respectively. His research interests include the theoretical basis of the design of reinforced
concrete structures, shear and punching shear, fiber-reinforced high-strength concrete,
soil-structure interaction, and the conceptual design of bridges.

related to such empirical formulations4 are also inherited by

members with shear reinforcement.
With respect to the shear-reinforcement contribution
(ηs ⋅ Vs0), some codes limit the stress in the shear reinforce-
ment at failure2 (ηs < 1.0). Limiting the allowable stress (and
thus the allowable strain) in the shear reinforcement is
logical because concrete can only carry shear forces if crack
widths remain controlled. For instance, in EC2,2 the design
(factored) value of the yield strength in the shear reinforcement
is limited to fywd = (250 + 0.25d [mm]) MPa (36.2 + 0.92d
[in.]) ksi. In ACI 318-05,1 the maximum yield strength in the
shear reinforcement is 414 MPa (60 ksi). The angle of the
punching shear cone (where the shear reinforcement is
considered active) also varies for different codes (for
example, 45 degrees for ACI 318-051 and 34 degrees for EC22).
Figure 1(d) compares ACI 318-051 and EC22 approaches,
where it has been assumed that ηs = 1.0. Also, a theoretical Fig. 2—Punching within shear-reinforced zone: (a) localization
curve describing the actual contributions of a shear-reinforced of strains within shear-reinforced zone; (b) contribution of
concrete slab is drawn. Contrary to the approaches followed concrete and of shear reinforcement; and (c) punching
by codes of practice, the value of ηc varies with the amount strength VR,in , load-rotation relationship of slab and
of shear reinforcement, bond conditions of the shear reinforce- contribution of concrete and of shear reinforcement as
ment, and longitudinal reinforcement ratio. It can be noted that function of rotation in slab.
codes of practice provide rather conservative estimates for low
shear reinforcement ratios. These differences are reduced as
the amount of shear reinforcement increases and the concrete in some cases, leads to very conservative estimates of punching
contribution may even be overestimated (refer to Fig. 1(d)). shear strength of flat slabs, but may also overestimate the
In this paper, a different approach for estimating the contribution of concrete or the shear reinforcement. This
punching strength of shear-reinforced slabs is introduced. paper introduces a physical model based on CSCT that allows
The approach is based on a physical model derived from the one to estimate, on a rational basis, the contributions of
critical shear crack theory (CSCT).3-6 This theory has thus concrete and of shear reinforcement to the punching shear
far been applied to members without transverse reinforcement strength. Because the proposed approach is based on a physical
by adopting the hypothesis that a single shear crack localizes model, it allows to account for the layout of shear reinforcement
the strains in the critical shear region. as well as the diameter, bond conditions, reinforcement ratio,
As previously shown,3,4,6 the width of the critical shear and other mechanical and geometrical parameters. Thus, it is
crack can be correlated to a physical parameter (a reference applicable to a wide number of shear reinforcement systems.
strain for beams in shear and the rotation of a slab for
punching shear in flat slabs), which allows determination of FAILURE CRITERION BASED ON CRITICAL
the deformation capacity of a member at ultimate. Taking SHEAR CRACK THEORY
advantage of the compatibility conditions introduced by Within the shear-reinforced zone, failure develops when a
CSCT (in this case, the rotation of the slab), the contributions critical shear crack opens, localizing the strains as shown in
of concrete and the shear reinforcement are estimated Fig. 2(a). In this case, a fraction of the applied shear can still
depending on the various geometrical and mechanical be carried by concrete (this contribution is limited by the
parameters of the specimen (and not by adopting a constant opening of the critical shear crack and by its roughness),
contribution). Also, because the theory relies on a mechanical while the remaining part of the shear force is carried by the
model, bond conditions can be introduced for the shear shear reinforcement. The punching strength can thus be
reinforcement, allowing one to describe the mechanical written as (refer to Fig. 2(b) and (c))
behavior of various shear reinforcement systems.
VR,in = Vc + Vs (3)
RESEARCH SIGNIFICANCE
Current methods for the design of punching shear reinforce- The concrete contribution Vc can be estimated according
ment rely on empirical formulations to estimate the contributions to CSCT assuming that a single crack develops in the failure
of concrete and shear reinforcement at failure. This approach, zone.5 Based on this theory, Muttoni4 proposed the

486 ACI Structural Journal/July-August 2009

Fig. 3—Punching shear strength in shear-reinforced slabs
based on critical shear crack theory: (a) actual and adopted Fig. 4—Contribution of the shear reinforcement: (a) geomet-
geometry for critical shear crack localizing strains; (b) rical parameters of shear reinforcement and of critical shear
assumed slab kinematics with opening of critical shear crack; crack; (b) opening of critical shear crack and longitudinal
and (c) corresponding crack widths for assumed geometry and transverse relative slips at shear reinforcement; and (c)
and kinematics. contribution of shear reinforcement within the punching cone.

following failure criterion, correlating well with results of w=κ·ψ·d (6)

99 punching shear tests without transverse reinforcement
where κ is a constant whose value is proposed in this paper
as κ = 0.50. Assuming that the critical shear crack is straight
b 0, int ⋅ d f c′
Vc = 3
--- ⋅ --------------------------------------
- SI units (N, mm) (4) (and thus the failure surface is a cone) and that the center of
4 1 + 15 ⋅ ------------------ ψ⋅d- rotation of the crack is located at its tip, the following
d g0 + d g expressions are obtained for the relative displacements of
the crack lips parallel (wbi) and perpendicular (δbi) to the
shear reinforcement (Fig. 4)
b 0, int ⋅ d f c′
V c = 9 ⋅ --------------------------------------
- U.S. customary units (psi, in.)
ψ⋅d
1 + 15 ⋅ ------------------- w bi = κ ⋅ ψ ⋅ h i ⋅ cos ⎛α + β i – π
--- ⎞
d g0 + d g ⎝ 2⎠
(7)
π
where ψ is the maximal rotation of the slab; b0,int is the δ bi = κ ⋅ ψ ⋅ h i ⋅ sin ⎛α + β i – --- ⎞
⎝ 2⎠
control perimeter (defined at d/2 beyond the tip of the crack,
typically at the face of the column [refer to Fig. 2(a)], but
other possible failure locations can be investigated); d is the where hi is the vertical distance between the tip of the crack and
average effective depth of the member; fc is the compressive the point where the shear reinforcement crosses the critical
strength of concrete; dg is the maximum aggregate size; and shear crack, α is the angle of the critical shear crack (proposed
dg0 is a reference aggregate size set to 16 mm (0.63 in.). to be equal to π/4), and βi is the angle between the shear
The contribution of the shear reinforcement can be estimated reinforcing bar and the slab plane (Fig. 4). Thus, the stress in
according to CSCT using its main hypothesis,4,5 which the shear reinforcement (σsi) can be evaluated from wbi (which
establishes that the opening of the shear critical crack is is in turn a function of the rotation of the slab, ψ), depending on
proportional to the product of the rotation of the slab times the the bond properties of the shear reinforcement. If the dowel
effective depth of the member (refer to Fig. 3) action of the shear reinforcement is neglected, the contribution
of the shear reinforcement becomes

w∝ψ⋅d (5) n
Vs = ∑ σsi ( ψ ) ⋅ Asi ⋅ sin ( βi ) (8)
Equation (5) can thus be rewritten as i=1

ACI Structural Journal/July-August 2009 487

 can be taken equal to 0.22l, where l is the span length of the flat
slab), fy is the yield strength of the flexural reinforcing steel, Es
is the modulus of elasticity of the flexural reinforcing steel, and
Vflex is the load necessary to develop the plastic mechanism of
the slab (a detailed solution can be found in References 4 and 7).
It should be noted that cases may exist where the shear
reinforcement is not fully yielded when the load-rotation
relationship intersects the failure criterion. In this case, if the
possible increase in the force carried by the shear reinforcement
is larger than the decrease in the force carried by the
concrete, the strength of the slab may be larger than assumed
Fig. 5—Failure within shear-reinforced zone VR,in: failure in Fig. 5. This potential increase, however, is neglected in the
criterion and load-rotation relationship. present study.

CHECK OF OTHER FAILURE MODES

Punching outside the shear-reinforced zone
Punching outside the shear-reinforced zone develops by
localization of the strains in a critical shear crack (Fig. 1(a)).
According to CSCT, the punching shear strength outside the
shear-reinforced zone (VR,out) can be estimated as the point
where the load-rotation relationship of the slab (Eq. (9))
intersects the failure criterion outside the shear-reinforced
zone (Fig. 6(a))

b 0, out ⋅ d v f c′
V R, out = 3--- ⋅ --------------------------------------
- SI units (N, mm) (10)
4 ψ⋅d
1 + 15 ⋅ -------------------
d g0 + d g

b 0, out ⋅ d v f c′
V R, out = 9 ⋅ --------------------------------------
- U.S. customary units (psi, in.)
ψ⋅d
1 + 15 ⋅ -------------------
d g0 + d g
Fig. 6—Failure outside shear-reinforced zone VR,out: (a) failure
criterion and load-rotation relationship; and (b) reduced
where dv is the reduced effective depth shown in Fig. 6(b) (to
effective depth dv.
account for pullout of shear reinforcement as will be
explained later) and b0,out is the control perimeter (defined at
where Asi is the cross-sectional area of the shear reinforcement. d/2 beyond the outer layer of shear reinforcement and
Appendix 1* details the analytical expressions for σsi(ψ) considering 4d as the maximum effective distance between
for a number of shear reinforcement systems with various two concentric rows of shear reinforcement). This approach
bond conditions. Expressions for other systems can be provides good fitting to test data and is slightly conservative
derived by analogy. because the rotation of the slab is assumed to be concentrated
Finally, once the concrete and the shear reinforcement in the shear-critical crack, although a fraction of the total
contributions are known, the punching shear strength of the rotation develops within the shear-reinforced zone.
member can be obtained by intersecting the failure criterion
(Vc + Vs) with the load-rotation relationship describing the Crushing of concrete struts near support region
behavior of the slab, as shown in Fig. 2(c) and (5). A simplified Crushing of the concrete struts near the column has also
expression describing the rotation of the slab (ψ) as a function been reported as a governing failure mode in some cases.8
of the applied load is given by the following formula The compressive strength of the compression struts is
(introduced in Reference 4 from a number of simplifications strongly influenced by their state of transverse strains, which
of a more general, analytically-derived expression) in turn are functions of the rotation of the slab (because wider
cracks develop for larger rotations). On that basis, the crushing
3 strength of a reinforced concrete slab is estimated as
---
r f
ψ = 1.5 ⋅ ----s ⋅ ----y- ⋅ ⎛ ----------
V ⎞2
(9)
d E s V flex⎠ ⎝
3 b 0, col ⋅ d f c′
V R, crush = λ ⋅ --- ⋅ --------------------------------------
- SI units (N, mm) (11)
4 ψ⋅d-
1 + 15 ⋅ ------------------
where rs is the distance from the edge of the column to the line d g0 + d g
of contraflexure of bending moments (that for regular flat slabs

b 0, col ⋅ d f c′
V R, crush = λ ⋅ 9 ⋅ --------------------------------------
- U.S. cust. units (psi, in.)
*The Appendix is available at www.concrete.org in PDF format as an addendum to ψ⋅d
the published paper. It is also available in hard copy from ACI headquarters for a fee 1 + 15 ⋅ -------------------
equal to the cost of reproduction plus handling at the time of the request. d g0 + d g

488 ACI Structural Journal/July-August 2009

where b0,col is the control perimeter (defined at d/2 of the Also, at the compression face, the pullout strength of the stud
edge of the column or support region). The parameter λ is set is increased by the favorable state of stresses (with compressive
equal to 3.0 for well-anchored shear reinforcement (that is, stresses in two directions) and by the dowel action of the
headed studs). Otherwise, this value is limited to 2.0. The compression reinforcement (if the shear reinforcement is
crushing strength of the slab is obtained again at the point enclosing it).
where the failure criterion (Eq. (11)) intersects the load-rotation Regarding some constructive rules (stud spacing and
relationship of the slab (Eq. (9)), as shown in Fig. 7. diameter, and shear reinforcement enclosing the flexural
reinforcement), the risk of pullout of the shear reinforcement
Pullout of anchorages within the shear-reinforced zone is limited. This is
As shown by other authors,9,10 an anchorage can be pulled confirmed by the results of several tests8,11 where cracks
out from a concrete block if insufficient embedment length developing from the anchorages of the shear reinforcement
is available (refer to Fig. 8(a)). Applications of this failure did not develop a pullout failure surface. This fact can, for
mode to punching in flat slabs has recently been investigated.8,11 instance, be appreciated in the saw-cut of Specimen Z3 by
It should be noted that a direct application of the results of Beutel8 (Fig. 8(b)) in which the width of the cracks originating
pullout tests9,10 to shear reinforcement is not straightforward at the anchorages of the studs are controlled by the next
because the conditions in which the pullout tests are perimeter of studs and, thus, a pullout surface like that of
performed are generally different from the actual conditions Fig. 8(a) does not develop.
of a stud inside a shear-reinforced flat slab. At the tension The width of the cracks developing from the anchorages of
face, the pullout strength of the shear reinforcement is the outer layer of shear reinforcement, however, is not
increased by the dowel action of the flexural reinforcement (if controlled by any other ring of shear reinforcement. Thus,
the shear reinforcement is enclosing the flexural reinforcement). the pullout failure surfaces may join, eventually leading to a
punching shear failure outside the shear-reinforced zone.
Figure 8(c) shows, for instance, the saw-cut of Specimen P1I
by Beutel,8 where the development of a punching cone from
the pullout cracks of the outer layer of shear reinforcement is
clearly visible. This fact, as previously stated, is considered
in the theoretical model by introducing a reduced effective
depth dv in Eq. (10) (refer to Fig. 8(c)), when performing the
check of the strength outside the shear-reinforced zone.
Another failure mode has been reported12 as the development
of horizontal cracks along the upper and lower surface of
anchorages leading to delamination of the member (Fig. 8(d)).
Fig. 7—Failure by crushing of the concrete strut VR,crush : This failure mode may become governing when the shear
failure criterion and load-rotation relationship. reinforcement is not enclosing the longitudinal reinforcement.

Fig. 8—Development of pullout cracks: (a) typical test setup for pullout specimens10;
(b) development of pullout cracks without development of pullout cones (Specimen Z3 by
Beutel8); (c) punching outside shear-reinforced zone by development of pullout cracks
(Specimen P1I by Beutel8); and (d) failure mode by delamination according to Regan and
Samadian.12

ACI Structural Journal/July-August 2009 489

Fig. 9—Application of code-like formulation to design of
shear reinforcement.

CODE-LIKE FORMULATION
Based on the previous approach, a simplified code-like
formulation can also be proposed for design purposes. To
that aim, the same three failure modes (failure within and
outside the shear-reinforced zone, together with crushing of
the compression struts), have to be checked.
The punching shear strength within the shear-reinforced
zone can be calculated on the basis of Eq. (3) as
Fig. 10—Comparison between measured and estimated
punching shear strengths (see details in Table 1): proposed
VRd = Vcd + Vsd (12) model using critical shear crack theory (CSCT) with its
refined and design (simplified) formulation; ACI 318-051
With respect to the shear force carried by concrete (Vcd), a and EC2.2
characteristic formulation of the failure criterion of the
CSCT4 (reaching a target 5% fractile) has to be adopted
Es ⋅ ψ
V sd = ⎛ -------------
- + τb ----- ⎞ ⋅ A sw ≤ f ywd ⋅ A sw
d
b 0, int ⋅ d f ck (15)
1 2 ⎝ 6 db ⎠
V cd = ---- ⋅ --- ⋅ --------------------------------------
- SI units (N, mm) (13)
γc 3 ψ⋅d
1 + 20 ⋅ -------------------
d g0 + d g where τb is the average bond stress, whose value can be
adopted as 5 MPa (725 psi) for ordinary cases.
The use of Eq. (12) for design purposes is rather simple.
b 0, int ⋅ d f c′
V cd = φ ⋅ 8 ⋅ --------------------------------------
- U.S. customary units (psi, in.) As shown in Fig. 9, the rotation at failure (ψd) can be calculated
ψ⋅d for a given value of the shear force (Vd) using the load-
1 + 20 ⋅ -------------------
d g0 + d g rotation relationship defined in Eq. (9). This rotation is then
used to calculate the contribution of concrete (Vcd) using
where fck and fc′ are the characteristic and specified Eq. (13). The remaining part (Vd – Vcd = Vsd) has to be carried
compressive strengths of concrete, respectively (according by the shear reinforcement, where the necessary amount can
to the European and North American practices, a relationship thus be determined directly on the basis of Eq. (14) or (15).
between both strengths is discussed in Reference 13); and Checking the punching strength outside the shear-
where γc is the partial safety factor of concrete (1.5 according reinforced area can be performed on the basis of Eq. (13)
to European practice); and φ is the strength reduction factor for by introducing the suitable control perimeter b 0,out and
punching (0.75 according to North American practice). the reduced effective depth dv as defined previously in Eq. (10).
For slabs with smooth shear reinforcement, and based on the The crushing strength can also be calculated on the basis of
expressions provided in Appendix 1 of this paper, the term Vsd Eq. (13) by multiplying this strength by factor λ as explained
can be easily estimated assuming for all shear reinforcements a in Eq. (11).
mean stress equal to that of a shear reinforcement placed at
0.5d from the border of the support region COMPARISON OF TEST RESULTS
TO CODES OF PRACTICE
Figure 10 compares the punching shear strength of the
Es ⋅ ψ tests detailed in Table 1 (with the shear reinforcement
- ⋅ A sw ≤ f ywd ⋅ A sw
V sd = ------------- (14)
6 systems shown in Fig. 11) to the results of the proposed
model and those of some codes of practice.1,2 The tests are
where Asw is the amount of shear reinforcement within a sorted along the ordinate using the following ratio
perimeter at d from the edge of the support region, and fywd (correlated to V s0 /V c0)
is the design yield strength of the shear reinforcement.
For slabs with deformed shear reinforcement, it can be A sw ⋅ f yw
assumed an increase on the shear reinforcement stress due to --------------------------
- (16)
bond as (refer to Regime 3 in Appendix 1) b 0 ⋅ d ⋅ f c′

490 ACI Structural Journal/July-August 2009

where Asw is the amount of shear reinforcement within a For the proposed model, a good agreement is found (refer
perimeter at d from the edge of the support region, fyw is the to Fig. 10) between the measured punching shear strength in
yield strength of the shear reinforcement, and b0 is a control the tests and the calculated punching shear strength, with an
perimeter at d/2 from the edge of the column. average ratio of 1.06 (values over 1.0 mean conservative

Table 1—Comparison of proposed model (refined and design [simplified] formulations), ACI 318-05,1 and
EC22 with available test data (refer to Fig. 11 for shear reinforcement types)*
Vtest /Vth
Shear reinforcement
Specimen types d, mm d, in. CSCT average CSCT simplified ACI 318-05 EC2
S1 — 124 4.88 1.08 1.26 1.40 1.11
S2 (b) 124 4.88 1.03 1.15 1.27 1.20
S3 (b) 124 4.88 1.01 1.12 1.22 1.12
S4 (b) 124 4.88 1.04 1.21 1.78 1.21
S5 (b) 124 4.88 1.03 1.13 1.21 1.21
S6 (b) 124 4.88 1.04 1.13 1.24 1.23
Birkle11
S7 — 190 7.48 1.02 1.19 1.23 0.96
S8 (b) 190 7.48 0.94 1.03 1.34 1.00
S9 (b) 190 7.48 1.13 1.33 1.62 1.07
S10 — 260 10.2 0.86 1.00 0.97 0.78
S11 (b) 260 10.2 0.99 1.05 1.22 0.99
S12 (b) 260 10.2 1.04 1.19 1.37 0.89
S1 — 159 6.26 0.94 1.09 1.28 0.94
S1A — 159 6.26 0.98 1.14 1.33 0.97
S2 (c) 153 6.02 1.05 1.22 2.16 1.25
S3 (c) 158 6.22 1.06 1.23 2.17 1.20
S4 (c) 159 6.26 1.10 1.31 2.18 1.26
S5 (c) 159 6.26 0.98 1.16 1.78 1.23
Gomes and Regan14
S6 (c) 159 6.26 1.10 1.20 1.65 1.23
S7 (c) 159 6.26 1.20 1.58 2.03 1.37
S8 (c) 159 6.26 1.19 1.41 1.99 1.47
S9 (c) 159 6.26 1.12 1.30 1.88 1.08
S10 (c) 154 6.06 1.19 1.39 2.04 1.27
S11 (c) 154 6.06 1.27 1.43 2.07 1.29
Z1 (a) 250 9.84 0.96 1.24 1.50 1.26
Z2 (a) 250 9.84 1.09 1.32 1.59 1.30
Z3 (a) 250 9.84 1.21 1.54 1.86 1.57
Z4 (a) 250 9.84 1.16 1.41 1.66 1.27
Z5 (a) 250 9.84 1.25 1.51 1.90 1.31
Beutel8 Z6 (a) 250 9.84 1.15 1.40 1.81 1.31
V4 (a) 350 13.7 0.97 1.15 1.09 1.09
P1 — 191 7.52 0.85 0.99 0.94 0.86
P1-I (e) 191 7.52 1.25 1.46 2.03 1.31
P1-II (e) 191 7.52 1.24 1.44 1.92 1.23
P2-I (e) 191 7.52 1.09 1.28 1.28 0.96
Müller et al.15 P22 (d) 153 6.04 1.09 1.28 1.66 0.86
V1 (b) 118 4.65 1.25 1.25 1.29 1.29
Stein et al.16 V2 (b) 118 4.65 0.95 1.11 1.39 1.18
V3 (b) 118 4.65 1.03 1.14 1.17 1.16
20-I (f) 155 6.10 1.03 1.09 1.70 1.16
20-II (f) 152 5.98 0.98 1.10 1.72 1.15
20-III (f) 150 5.90 1.00 1.16 1.68 1.16
Rojek and Keller17,18
30-I (f) 252 9.92 0.95 1.26 1.41 1.05
30-II (f) 260 10.2 0.88 1.24 1.34 0.83
30-III (f) 254 10.0 1.03 1.64 1.72 1.11
Average 1.06 1.25 1.58 1.15
COV 0.10 0.13 0.22 0.15
*
Maximum aggregate size of tests by Stein et al.16 equal to 14 mm (0.55 in.), value given by Dilger in private communication; maximum aggregate size of tests by Rojek and
Keller17,18 equal to 16 mm (0.63 in.), yield strength of flexural steel = 506 MPa (73.3 ksi), values given by Rojek in private communication.

ACI Structural Journal/July-August 2009 491

Fig. 11—Shear reinforcement types: (a) deformed stud; (b)
smooth stud; (c) steel offcut; (d) headed stirrup; (e) stirrup;
and (f) enhanced-bond shear reinforcement.

Fig. 12—Comparison between measured and estimated

rotations at failure.

estimates) and with a small value of the coefficient of variation

(COV = 10%). A comparison of the proposed theory using
the simplified design formulation (setting material safety
factors to 1.0) is also shown in Fig. 10. The average value of
the measured-to-estimated punching shear strength increases
to 1.25, with a value of the COV of 0.13. The ratio for the 5%
fractile is 1.03, thus leading to a satisfactory safety level
(larger than 1.0).
EC22 shows a good agreement with respect to the average Fig. 13—Influence of flexural reinforcement ratio and
value of the previous ratio (1.15). Accounting for the COV, amount of shear reinforcement on contributions of steel and
however, unsafe results may be obtained in actual designs concrete: (a) investigated specimen (d = 260 mm [10.2 in.];
(5% fractile below 1.0). ACI 318-051 leads to larger scatters fc = 25 MPa [3600 psi]; dg = 32 mm [1.25 in.]; fy = fyw =
than EC2;2 however, the rather conservative estimates of the 500 MPa [72 ksi]); (b) failure criteria and estimated failure
measured-to-estimated punching shear strength leads to a loads; and (c) influence of flexural reinforcement ratio on
satisfactory 5% fractile (above 1.0). concrete contribution.
It is interesting to note that with the proposed approach,
not only the punching shear strength but also the rotation of reinforcement ratio ρ. The results, in terms of load-rotation
the slab can be estimated at failure (refer to Fig. 12). A diagrams, are presented in Fig. 13(b) for various cases. For
satisfactory agreement is also found for this parameter with the lowest flexural reinforcement ratio (ρ = 0.33%), the
the proposed model with conservative estimates of the rotation bending strength of the slab is reached even for very small
at failure in general. amounts of shear reinforcement. As the flexural reinforcement
ratio increases, the role of the amount of shear reinforcement
CONTRIBUTIONS OF CONCRETE AND OF SHEAR in the strength of the member becomes more evident. It can
REINFORCEMENT ACCORDING TO CSCT be noted that, for the case where ρ = 1.25%, the bending
Figure 13(a) shows a circular reinforced concrete slab strength of the slab cannot be reached independently of the
where the diameter of the shear reinforcing bars db (deformed amount of shear reinforcement because crushing of the struts
studs in this case) is varied together with the flexural becomes governing. Based on the previous results, the

492 ACI Structural Journal/July-August 2009

 can be easily introduced with the proposed physical model.
Such conditions show a non-negligible influence in the
punching shear strength, although most codes of practice do
not consider them;
6. Very good agreement is found when comparing the
proposed approach to available test data; and
7. A simplified design formulation can be derived on the
basis of CSCT leading to a satisfactory safety level when
compared to available test results.

Fig. 14—Influence of bond conditions on punching shear NOTATION

strength (same specimen as in Fig. 13(a)): (a) punching Asw = cross-sectional area of shear reinforcement
shear strength with smooth and with deformed studs; and Asw1 = sum of cross-sectional areas of shear reinforcements placed at one
(b) punching shear strength for prestressed smooth studs perimeter (reinforcements at same distance from support region)
b0 = perimeter of critical section
(refer to Fig. 15(b) in Appendix 1). b0,col = perimeter of critical section at d/2 of face of column
b0,int = perimeter of critical section (check of punching within shear-
reinforced zone)
contributions of the shear reinforcement and of concrete at b0,out = perimeter of critical section (check of punching shear outside
failure can be estimated and plotted in a format similar to that shear-reinforced zone)
of codes of practice (refer to Eq. (2) and Fig. 1). The results d = distance from extreme compression fiber to centroid of
(Fig. 13(c)) show that codes of practice underestimate the longitudinal tensile reinforcement
db = diameter of reinforcing bar
contribution of concrete for small amounts of shear reinforce- dg = maximum diameter of aggregate
ment (small values of the ratio Vs/Vc0), but may overestimate dg0 = reference aggregate size (16 mm [0.63 in.])
it for large amounts of shear reinforcement. dv = reduced effective depth
The contributions of concrete and shear reinforcement Es = modulus of elasticity of reinforcement
fc = average compressive strength of concrete (cylinder)
also vary depending on the bond conditions of the member. fy = yield strength of flexural reinforcement
Figure 14(a) shows the results obtained for the previous slab fyw = yield strength of shear reinforcement
(Fig. 13(a)) with ρ = 1.25% and db = 14 mm (0.55 in.) when fyw,ef = effective yield strength of shear reinforcement
deformed and smooth studs are used. According to the fywd = design yield strength of shear reinforcement
theoretical model, the punching strength with deformed h = vertical distance between tip of crack and point where shear
reinforcement crosses critical shear crack
studs increases because the strains in the shear reinforcement k = size effect factor in EC2
for the same rotation are larger. A similar effect is also l = span of slab, length
shown in Fig. 14(b), where the smooth studs are prestressed lai = distance between point where shear reinforcement is crossed
at various levels. When prestressing is increased, the by critical shear crack and closest head of shear reinforcement
punching strength of the member also increases according to las = distance between point where shear reinforcement is crossed
by critical shear crack and head of shear reinforcement found
the theoretical model because the stress in the studs is larger farthest from that point
for the same rotation of the slab. Both effects are not ls = stud length
included in current design methods. rs = distance between column of slab and line of contraflexure of
moments
sr = distance between perimeters of shear reinforcement
SUMMARY AND CONCLUSIONS V = shear force
This paper investigates the punching shear strength of Vc = concrete contribution to punching shear strength
reinforced concrete flat slabs with shear reinforcement. Its Vc0 = punching shear strength of slab without shear reinforcement
main conclusions are: Vcd = design concrete contribution to punching shear strength
1. The contribution of concrete to the punching shear Vd = design (factored) action
Vflex = shear force associated with flexural capacity of slab
strength of flat slabs is not constant, contrary to what is Vflex,d = shear force associated with design flexural capacity of slab
assumed in most codes of practice. Codes of practice usually VR = punching shear strength
underestimate this contribution for low amounts of shear VR = design punching shear strength
reinforcement and may overestimate it for large amounts of VR,crush = punching shear strength (governing crushing of concrete
struts)
shear reinforcement;
VR,in = punching shear strength (governing failure within shear-
2. A physical model based on the critical shear crack reinforced zone)
theory can be used to investigate the punching shear strength VR,out = punching shear strength (governing failure outside shear-
in members with transverse reinforcement leading to simple reinforced zone)
analytical expressions; Vs = shear reinforcement contribution to punching shear strength
Vs0 = force that can be carried by shear reinforcement within
3. Based on the critical shear crack theory, it is demonstrated punching cone at yielding
that the contribution of concrete to the punching shear Vsd = design shear reinforcement contribution to punching shear
strength is reduced for large rotations of the slab. This strength
reduction depends on some parameters such as, for Vtest = experimental punching shear strength
instance, the flexural reinforcement ratio or the amount of w = critical shear crack width
wb = relative displacement parallel to shear reinforcement
shear reinforcement; α = angle of critical shear crack with compression face of slab
4. The proposed model can be applied to a number of shear β = angle of shear reinforcement with compression face of slab
reinforcing systems and to investigate the role of the various δb = relative displacement perpendicular to shear reinforcement
geometrical and mechanical parameters; εs = steel strain
φ = strength reduction factor
5. The model accounts for the position, concrete cover, γc = partial safety factor of concrete
and other geometrical parameters of the shear reinforcement. ηc = efficiency factor for contribution of concrete
Bond conditions and prestressing of the shear reinforcement ηs = efficiency factor for contribution of shear reinforcement

ACI Structural Journal/July-August 2009 493

κ = critical shear crack opening factor 9. Fuchs, W.; Eligehausen, R.; and Breen, J. E., “Concrete Capacity
λ = crushing strength parameter Design (CCD) Approach for Fastening to Concrete,” ACI Structural
ρ = flexural reinforcement ratio Journal, V. 92, No. 1, Jan.-Feb. 1995, pp. 73-94.
σs = steel stress 10. Regan, P. E., “Shear Reinforcement of Flat Slabs,” Proceedings of
σsp = prestress the International Workshop on Punching Shear Capacity of RC Slabs,
τb = bond strength TRITA-BKN, Bulletin 57, 2000, pp. 99-107.
ψ = rotation of slab outside column region 11. Birkle, G., “Punching of Flat Slabs: The Influence of Slab Thickness
ψd = rotation of slab outside column region (design value) and Stud Layout,” Department of Civil Engineering, University of Calgary,
Calgary, AB, Canada, 2004, 152 pp.
REFERENCES 12. Regan, P. E., and Samadian, F., “Shear Reinforcement against
1. ACI Committee 318, “Building Code Requirements for Structural Punching in Reinforced Concrete Flat Slabs,” The Structural Engineer,
Concrete (ACI 318-05) and Commentary (318R-05),” American Concrete V. 79, No. 10, 2001, pp. 24-31.
Institute, Farmington Hills, MI, 2005, 430 pp. 13. Reineck, K.-H.; Kuchman, D. A.; Kim, K. S.; and Marx, S., “Shear
2. Eurocode 2, “Design of Concrete Structures—Part 1-1: General Rules and Database for Reinforced Concrete Members without Shear Reinforcement,”
Rules for Buildings,” CEN, EN 1992-1-1, Brussels, Belgium, 2004, 225 pp. ACI Structural Journal, V. 100, No. 2, Mar.-Apr. 2003, pp. 240-249.
3. Muttoni, A., and Fernandez Ruiz, M., “Shear Strength of Members 14. Gomes, R., and Regan, P. E., “Punching Strength of Slabs Reinforced
without Transverse Reinforcement as a Function of Critical Shear Crack for Shear with Offcuts of Rolled Steel I-Section Beams,” Magazine of
Width,” ACI Structural Journal, V. 105, No. 2, Mar.-Apr. 2008, pp. 163-172. Concrete Research, V. 51, No. 2, 1999, pp. 121-129.
4. Muttoni, A., “Punching Shear Strength of Reinforced Concrete Slabs
15. Müller, F.-X.; Muttoni, A.; and Thurlimann, B., “Punching Tests on
without Transverse Reinforcement,” ACI Structural Journal, V. 105, No. 4,
Flat Slabs with Recesses,” Institut für Baustatik und Konstruktion, ETH
July-Aug. 2008, pp. 440-450.
Zürich, Report Nr. 7305-5, Birkhäuser Verlag Basel-Boston-Stuttgart,
5. Muttoni, A., and Schwartz, J., “Behaviour of Beams and Punching in
1984, 118 pp. (in German)
Slabs without Shear Reinforcement,” IABSE Colloquium, V. 62, Zurich,
Switzerland, 1991, pp. 703-708. 16. Stein, T.; Ghali, A.; and Dilger, W., “Distinction between Punching
6. Muttoni, A., “Shear and Punching Strength of Slabs without Shear and Flexural Failure Modes of Flat Plates,” ACI Structural Journal, V. 104,
Reinforcement,” Beton- und Stahlbetonbau, V. 98, No. 2, 2003, pp. 74-84. No. 3, May-June 2007, pp. 357-365.
(in German) 17. Rojek, R., and Keller, T., “Punching Tests using Reinforcement with
7. Nielsen, M. P., Limit Analysis and Concrete Plasticity, second edition, Enhanced Bond, Structural Analysis and Design Approaches,” Beton-und
CRC Press, Boca Raton, FL, 1999, 908 pp. Stahlbetonbau, V. 102, No. 8, 2007, pp. 584-556. (in German)
8. Beutel, R., “Punching of Flat Slabs with Shear Reinforcement at Inner 18. Rojek, R., and Keller, T., “Punching Tests with HFV-Studs,”
Columns,” Rheinisch-Westfälischen Technischen Hochschule Aachen, Fachhochschule Augsburg, Kompetenzzentrum Konstruktiver Ingenieurbau,
Aachen, Germany, 2002, 267 pp. (in German) 2006, 270 pp. (in German)

494 ACI Structural Journal/July-August 2009

View publication stats