buildings
Article
Effect of Shape, Number, and Location of Openings on
Punching Shear Capacity of Flat Slabs
Ekkachai Yooprasertchai 1 , Yonlada Tiawilai 1 , Theerawee Wittayawanitchai 1 , Jiranuwat Angsumalee 1 ,
Panuwat Joyklad 2, * and Qudeer Hussain 3






Citation: Yooprasertchai, E.; Tiawilai,
Y.; Wittayawanitchai, T.; Angsumalee,
J.; Joyklad, P.; Hussain, Q. Effect of
Shape, Number, and Location of
Openings on Punching Shear
Capacity of Flat Slabs. Buildings 2021,
11, 484. https://doi.org/10.3390/
buildings11100484
Academic Editors: Eva
O. L. Lantsoght and Giuseppina Uva
Received: 10 September 2021
Accepted: 12 October 2021
Published: 17 October 2021
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2021 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Buildings 2021, 11, 484. https://doi.org/10.3390/buildings11100484
1 Construction Innovations and Future Infrastructure Research Center (CIFIR), Department of Civil
Engineering, Faculty of Engineering, King Mongkut’s University of Technology Thonburi,
Bangkok 10140, Thailand; ekkachai.yoo@kmutt.ac.th (E.Y.); yonlada.tiawilai1@gmail.com (Y.T.);
dome84898@gmail.com (T.W.); Jiranuwat27@gmail.com (J.A.)
2 Department of Civil and Environmental Engineering, Srinakharinwirot University,
Nakhon Nayok 26120, Thailand
3 Center of Excellence in Earthquake Engineering and Vibration, Department of Civil Engineering,
Chulalongkorn University, Bangkok 10330, Thailand; ebbadat@hotmail.com
* Correspondence: panuwatj@g.swu.ac.th
Abstract: Experimental evidence have proved that punching shear capacity of flat slabs deteriorate
with the presence of openings located within the critical perimeter around columns. It is understood
that this deterioration varies inversely with the distance of openings from column’s face. However,
effect of the shape of openings on punching shear capacity is not well known. This study presents
experimental results of 14 flat specimens to investigate the effects of the number (2 and 4), shape
(circular, square, and rectangular), and location (1 and 4 times of slab’s thickness from column’s face)
of openings on punching shear strength. It was found that circular openings had least influence
on punching capacity followed by square and rectangular openings, respectively. Further, placing
openings at a distance of four times the slab’s thickness from column’s face had minimal impact on
punching capacity. Further, increasing the number of openings from 2 to 4 substantially reduced
the punching capacity. An effort was made to predict the punching capacities of all specimens
using the descriptive equations of ACI 318-19 and Eurocode 2. Mean of the ratio of experimental
to analytical results and standard deviation of ACI equations were found to be more accurate than
those of Eurocode 2 predictions.
Keywords: punching shear; effect of shape; ACI 318-19; Eurocode 2; flat slabs
1. Introduction
Nowadays, flat slabs are widely used because of their simple reinforcement details,
reduced building heights, rapid construction etc. A flat slab without beams is less rigid
and open to larger vertical displacements than conventional slabs [1,2]. Flat slabs are
prone to punching shear due to the accumulation of high shear stresses in the vicinity of
slab-column junction [3,4]. Punching shear failures are undesirable for their progressive
occurrence without any signs of warnings [5]. Openings in flat slabs are usually provided
to accommodate plumbing, electric wires, air-conditioning systems, etc. Experimental
studies have demonstrated that the presence of such openings within the critical punching
shear perimeter around the periphery of slab-column interface can substantially reduce the
shear capacity of flat slabs.
El-Salakawy et al. [6] reported the results of on the effect of shear studs on the behavior
of slab-column edge connections with openings. They concluded that the existence of
openings substantially reduced the stiffness of slab-column edge connection. They further
recommended that openings as large as the size of the column should not be constructed
as shear studs had minimal effect in this situation. Teng et al. [7] focused on flat plate slabs
https://www.mdpi.com/journal/buildings
Buildings 2021, 11, 484 2 of 24

with openings on rectangular columns. They tested 20 slab specimens under concentrated
loads and found that openings punching strengths of flat plate slabs were substantially
reduced in the presence of openings. They further recommended that if the placement
of an opening becomes inevitable, the best place is to construct along the longer side
of the column. Borges et al. [8] carried out punching shear tests on 13 flat plate slabs
with/without openings or/and shear reinforcement. Openings (1 or 2) were constructed
along the short side of the column. It was shown that for the relatively small openings
studied, the provision of continuous bars adjacent to openings—to replace the areas of
reinforcement—seemed to be an acceptable approach to flexural design. Anil et al. [9]
carried out an experimental investigation on the proximity (adjacent o column and at
300 mm away), location (parallel or diagonal to column), and size (300 mm × 300 mm
and 500 mm × 500 mm) of openings on punching capacities of two-way square slabs
(2000 mm × 2000 mm × 120 mm) subjected to concentrated axial load applied from top of
column (200 mm × 200 mm). It was reported that punching resistance decreased as the
opening size increased. Further, a notable decrease in punching capacity was observed
for the openings placed adjacent to column in comparison to openings placed 300 mm
away from the column. Ha et al. [10] performed experimental investigation on 8 flat slabs
considering the layout and number of openings. It was reported that an L-shaped opening
layout around the corner of a column resulted in further reduction in addition to the
loss of effective critical sections due to the existence of openings. Balomenos et al. [11]
carried out deterministic parametric investigation on the effect of opening distance from
column’s face and found out that irrespective of the size of opening, opening placed at
4 h (where “h” is slab’s thickness”) from column’s face does not have significant effect
on punching capacity of flat slabs. Liberati et al. [12] tested 12 reinforced concrete slabs
without shear reinforcement under symmetrical loading. Slabs were tested in 3 groups
depending upon the number of openings adjacent to column. It was reported that energy
dissipation capacity of slabs decreased as the number of openings increased. Slab with
openings underwent larger displacements than slabs without openings.
Experimental studies have highlighted the adverse effects of openings on shear capac-
ity of slabs. It was reported [13] that openings next to column reduce the concrete area that
sustains punching shear. Shear stresses were increased due to the presence of openings,
load conditions, and unbalanced moments generated from slab geometry. Further, it was
shown that openings located at distances greater than four times of slab effective depth
had negligible effect on slab’s shear capacity and such slabs behaved similar those without
any openings. Moreover, shear strength of slabs with very small openings (50 mm and
70 mm in diameter) located at distance twice the effective depth of slab from the abutment
also remained unchanged. Another study [14] investigated the number and size of circular
openings around column’s periphery. Number of openings were varied from 2 to 4 while
their size was varied as 75 mm, 100 mm, and 150 mm. It was concluded that slabs with
two openings (100 mm and 150 mm) lost stiffness compared to the slab with four 75 mm
openings. Further, size of openings influenced the rotational capacity of slabs.
Size and proximity of an opening from column’s face contribute mainly against
punching shear strength. It has been reported that both the size and distance of an open-
ing from column’s face adversely affect punching capacity [9,15]. Five square flat slabs
(1800 mm × 1800 mm × 130 mm) supported on square columns (150 mm × 150 mm) with
square openings (150 mm × 150 mm) at different distances from column’s face were tested
until failure. Results revealed the presence of very high shear stresses concentrated in
the regions between column and the opening. It was further concluded that any opening
located farther than three times of slab’s effective depth from column’s face had no effect
on the shear capacity of slab [16].
To the author’s knowledge, no experimental work has been conducted till now to
investigate the effect of opening shapes in flat slabs. However, an experimental study [17]
was conducted to study the effect of the shape of opening on shear capacity of conventional
slabs. A total of 15 slabs were tested with rectangular (200 mm × 150 mm and 150 mm ×
Buildings 2021, 11, 484 3 of 24

100 mm) and circular (150 mm and 100 mm) openings placed at the center of slab. It was
found that the size of opening had a direct effect on the shear capacity of slabs. Further, the
shear degradation caused by circular openings was lower than that caused by rectangular
openings. The program involved only the placement of openings at the center of slab and
testing under a Universal Testing Machine (UTM). Such testing methodology is far from
replicating the actual loading scenario in flat slabs. Therefore, a more rational and practical
approach is required to replicate the loading mechanism sustained by flat slabs in actual
structures. Further, ACI 318-19 [18] and Eurocode 2 [19] impose limitations on the distance
of an opening from column’s face. ACI 318-19 and Eurocode 2 consider the maximum
location at which an opening can still affect the punching shear capacity of flat slabs to be
4H and 6d from column’s face, respectively (where “H” and “d” are slab thickness and
effective depth, respectively). It is to be mentioned that ACI 318-19 recently adopted this
distance to be 4H. Previous edition i.e., ACI 318-14 considered this distance to be 10H
from column’s face. However, information regarding the shape of an opening is yet to be
established. Therefore, given the excessive use of flat slabs in modern-day construction
industry along with the inevitable presence of openings in them, there is a significant need
to further extend this domain to various opening shapes and their configurations.
The objectives of this study are to investigate the effect of the shape (circular and
square) on the punching shear capacity of flat slab, to study the effects of the distance of an
opening from the column’s face, and to assess the applicability of existing punching shear
capacity analytical formulations of ACI 318-19 [18] and Eurocode 2 [19] on the results of
this study.

2. Experimental Program
2.1. Test Matrix
A total of 14 flat slabs were tested in this study. Each specimen was sized to 1000 mm,
1000 mm, and 80 mm in length, width, and depth, respectively. A single 100 mm × 100 mm
steel plate was installed concentrically in each slab to simulate a column. Either 2 or 4
openings were provided in each specimen at opposite sides of the column. Size, shape, and
location from column’s face of openings in each specimen are provided in Table 1. Square
(70 mm × 70 mm), rectangular (70 mm × 140 mm), and circular (70 mm) openings were
used to investigate the shape effect of openings on punching shear strength. Effect of the
proximity of an opening from column’s face was studied by placing openings at 1 and 4 h
from the column’s face, where h is the slab’s thickness i.e., 80 mm. Specimens were mainly
divided in 2 groups depending upon the concrete strength. Distance at 4 h was chosen to
cover a wide range of distance from column’s face within which an opening could affect
the punching shear capacity. It basically creates the lower bound as most recently ACI
318-19 has adopted this distance as the safe distance at which an opening could be placed
with minimal effect on the punching capacity of slabs. The other distance (i.e., 1 h) was
chosen for practical reasons and is expected to create the upper bound of the deterioration
in punching capacity caused by the openings. Therefore, both distances were selected to
cover the widest range possible. Each specimen had 1 specimen without any openings
to serve as the reference/control specimen. The remaining 6 specimens were subdivided
into 3 groups depending upon the shape of opening. Each subgroup of 2 specimens had
similar opening shape but its distance varied to the above specified values. Nomenclature
for specimens was chosen to represent the number, shape, and location of openings in the
same order. For example, specimen 2S1H means that this specimen had 2 square openings
located at 1 h from the column’s face on each side of the column. Typical layout of the
openings in each slab is presented in Figures 1 and 2 for group 1 and 2, respectively.
Buildings 2021, 11, x FOR PEER REVIEW 4 of 24
Buildings 2021, 11, 484 4 of 24

Table 1. Test Matrix.

Specimen Specimen Compressive StrengthTable 1. Test Matrix. Opening Opening Size

Number of Opening Location (mm)
No. Name of Concrete (MPa)
Compressive
Shape (mm)
Specimen
1 Specimen
CON1 20.18 Number - of Opening
Control Opening Size
Control Control
Strength of Location (mm)
No. Name Opening Shape (mm)
2 2S1H 20.18 (MPa)
Concrete 2 Square 70 × 70 80 (1 h)
31 2C1H
CON1 20.18
20.18 -2 Circular
Control Diameter
Control 70 80 (1 h)
Control
42 2S1H
2R1H 20.18
20.18 22 Square
Rectangular × ×70140
7070 80
80(1(1h)h)
53 2C1H
2S4H 20.18
20.18 2 2 Circular
Square Diameter
70 × 7070 320 (4h)h)
80 (1
4 2R1H 20.18 2 Rectangular 70 × 140 80 (1 h)
65 2C4H
2S4H
20.18
20.18 2
2 Circular
Square
Diameter
70 × 70
70 320 (4 h)
320 (4 h)
76 2R4H
2C4H 20.18
20.18 22 Rectangular
Circular 70 × 140
Diameter 70 320(4(4h)h)
320
87 CON22R4H 29.71
20.18 2- Control
Rectangular × 140
70 Control Control
320 (4 h)
98 CON2
4S1H 29.71
29.71 -4 Control
Square Control
70 × 70 Control
80 (1 h)
9 4S1H 29.71 4 Square 70 × 70 80 (1 h)
10 4C1H 29.71 4 Circular Diameter 70 80 (1 h)
10 4C1H 29.71 4 Circular Diameter 70 80 (1 h)
1111 4R1H
4R1H 29.71
29.71 4 4 Rectangular
Rectangular 70 ×
70 × 140 140 80 (1h)h)
80 (1
1212 4S4H
4S4H 29.71
29.71 44 Square
Square 7070 × 70
× 70 320(4(4h)h)
320
1313 4C4H
4C4H 29.71
29.71 4 4 Circular
Circular Diameter
Diameter 70 70 320
320(4(4h)h)
70 × 140
1414 4R4H
4R4H 29.71
29.71 4
4 Rectangular
Rectangular 70 × 140 320 (4 h)
320 (4 h)

Figure 1. Layout of specimens with 2 openings (units: mm).

Figure 1. Layout of specimens with 2 openings (units: mm).
Buildings 2021, 11,
Buildings 2021, 11, 484
x FOR PEER REVIEW 55 of
of 24
24

Figure 2. Layout
Figure 2. Layout of
of specimens
specimens with
with 44 openings
openings (units:
(units: mm).
mm).

2.2. Specimen Design

2.2. Specimen Design
All
All specimens
specimens werewere designed
designed to to be
be shear
shear controlled.
controlled. All All specimens
specimens werewere provided
provided
with
with sufficient, and similar flexural reinforcement ratio (0.009) to have higher flexural
sufficient, and similar flexural reinforcement ratio (0.009) to have higher flexural
strength
strength than
than corresponding
corresponding shearshear strengths. Typical layout
strengths. Typical layout of of flexural
flexural reinforcement
reinforcement in in
each specimen is
each specimen isshown
shownininFigure
Figure3.3.ACIACI 318-19
318-19 [18][18] provisions
provisions werewere used
used to calculate
to calculate the
the
shearshear
and and flexural
flexural strengths
strengths of specimens.
of specimens. ShearShear
strengthstrength
provided provided by concrete
by concrete is
is calcu-
calculated using Equation
lated using Equation (1). (1).
 p
  p  
0.33λ
⎡ s λ f 0 , 0.17 + 0.33
0.33
λ s λ f0 , ⎤
vC = min ⎢ 0.33𝜆 𝜆 c𝑓 ′ , 0.17
0.83αs d
 + β p 𝜆 𝜆 c𝑓 ′ ,
𝛽 f0 ⎥ (1)
𝑣 = min ⎢ 0.17 + b0 λ s λ c ⎥ (1)
⎢ 0.83𝛼 𝑑 ⎥
⎢ 0.17 + 𝜆𝜆 𝑓 ′ ⎥
⎣ 𝑏 ⎦
Buildings 2021, 11, x FOR PEER REVIEW 6 of 24

Buildings 2021, 11, 484 6 of 24

where,
• 𝑓 = concrete cylinder compressive strength (MPa)
• where,
𝑑 = effective slab thickness for shear (mm)
• • 𝑏 f c=0 =perimeter
concrete of shear critical
cylinder sectionstrength
compressive 0.5d from loading area periphery (mm)
(MPa)
• • 𝛼 d==40 for interior
effective columns, 30
slab thickness forfor edge
shear columns, and 20 for corner columns.
(mm)
• • 𝛽 b=0 The ratio of of
= perimeter long to short
shear criticalsides of the
section 0.5dcolumns, concentrated
from loading load, or
area periphery reaction
(mm)
• area.
αs = 40 for interior columns, 30 for edge columns, and 20 for corner columns.
• β = The ratio of long to short sides of the columns, concentrated load, or reaction area.
• 𝜆 = q ≤ 1
• λs = .1+0.004d 2
≤1

Figure 3. Typical layout of flexural reinforcement (units: mm).

Figure 3. Typical layout of flexural reinforcement (units: mm).
As per the recommendations of ACI 318-19, critical shear perimeter b0 was reduced to
As per the recommendations of ACI 318-19, critical shear perimeter 𝑏 was reduced
account for the presence of openings located within 4 times of slab’s effective depth. Critical
to account for the presence of openings located within 4 times of slab’s effective depth.
shear perimeter was reduced by an area enclosed by two tangents extended to the outline
Critical shear perimeter was reduced by an area enclosed by two tangents extended to the
of openings from the center of column as shown in Figure 4. ACI 318-19 and Eurocode 2
outline
consideropenings
of the maximum from the centeratofwhich
location column anas shown can
opening in Figure 4. ACI
still affect the318-19
punchingand shear
Eu-
rocode 2 consider the maximum location at which an opening can still affect
capacity of flat slabs to be 4H and 6d from column’s face, respectively (where “H” and “d” the punching
shear capacity
are slab of flatand
thickness slabs to be 4H
effective and respectively).
depth, 6d from column’s face,
It is to respectively
be mentioned (where
that “H”
ACI 318-19
and “d” are slab thickness and effective depth, respectively). It is to be
recently adopted this distance to be 4H. The previous edition, i.e., ACI 318-14 considered mentioned that
ACI 318-19
this recently
distance to be adopted
10H from this distanceface.
column’s to beThis
4H.modification
The previousisedition, i.e., ACI
based upon the 318-14
findings
considered
of Genikomsou and Polak [20], who concluded that slabs with openings situated upon
this distance to be 10H from column’s face. This modification is based farther
thethan
findings
4H fromof Genikomsou
column’s face and Polak [20],
essentially who concluded
exhibited that slabs
same punching with openings
strength sit-
as those slabs
uated farther than 4H from column’s face essentially exhibited same
without openings. Therefore, location of openings were well within the domain specified punching strength
asbythose
ACIslabs without openings. Therefore, location of openings were well within the do-
318-19.
main specified by ACI 318-19.
 Buildings
Buildings 2021,
2021, 11, 11, 484
x FOR PEER REVIEW 7 of 24 7 of 24

FigureFigure
4. Calculation of 𝑏 . of b0 .
4. Calculation

According
According to Eurocode
to Eurocode 2 [19]2the
[19]design
the design punching
punching shearshear resistance
resistance of awithout
of a slab slab without
punching shear reinforcement along the control section
punching shear reinforcement along the control section (𝑣 Rd,c ( v ) may be calculated as shown
, ) may be calculated as
in Equation (2)
shown in Equation (2) 1
vc = CRd,c k(100ρ1 f ck ) 3 (2)
𝑣 = 𝐶 , 𝑘( 100ρ 𝑓 ) (2)
where
where• f ck is in MPa
• 𝑓• iskin
q
= 1MPa
+ 200 d ≤ 2.0 (d in mm)
• k•= 1 + √
ρ1 = ρ≤ly 2.0 ρlz (d
≤in mm)
0.02
• ρ• = ρρly ,ρρlz ≤related
0.02 to the bonded tension steel in y- and z-direction respectively.
• •
ρ , ρ Rd,c
C is 0.18.
related to the bonded tension steel in y- and z-direction respectively.
• 𝐶 , is For0.18
Eurocode 2, design checks are carried out at the critical perimeter and at column’s
face. The
For Eurocode code2,specifies this critical
design checks perimeter
are carried out atto the
be located
critical at a distance
perimeter of at
and 2dcolumn’s
from columns
face (without shear reinforcement). Further, a part of this perimeter
face. The code specifies this critical perimeter to be located at a distance of 2d from needs to be considered
col-
umnsineffective
face (without as long as the
shear shortest distance
reinforcement). between
Further, column’s
a part of thisface and the needs
perimeter opening to does
be not
considered ineffective as long as the shortest distance between column’s face and the the
exceed 6d. Size of ineffective perimeter is enclosed by two tangents ex-tending from
center
opening does ofnot
column
exceed to the
6d. outline of openingperimeter
Size of ineffective as illustrated in Figure
is enclosed by4.two tangents ex-
tending from the center of column to the outline of opening as illustrated as
Similarly, flexural strength of tested specimens was assessed per yield
in Figure 4. line the-
ory [21] descriptive equation as follows in Equation (3) and (4).
Similarly, flexural strength of tested specimens was assessed as per yield line theory
[21] descriptive equation as follows in Equation (3) and (4). !
8
Py = mr 8 L + 2π (3)
𝑃 =𝑚 c −+ 1 2𝜋 (3)
𝐿 − 1
𝑐 
fy

2
mr = ρd f y 1 − 0.59ρ 0 𝑓 (4)
𝑚 = 𝜌𝑑 𝑓 1 − 0.59𝜌 f (4)
𝑓′ c
where
where
• Py = Flexural capacity
• 𝑃• = Flexural
L = Length capacity
of supported slab
• L•= Length of
c = Loading supported slablength
plate side
• c•= Loading plate side length
ρ = Flexural reinforcement ratio
• 𝜌 = Flexural reinforcement ratio
Table 2 lists calculated flexural and shear strengths of all specimens. It can be seen
Table 2 listsstrengths
that shear calculated flexural
of all and shear
specimens strengths oflower
are significantly all specimens.
than their It can bestrengths.
flexural seen
that shear strengths of all specimens are significantly lower than their flexural strengths.
Buildings 2021, 11, 484 8 of 24

Table 2. Nominal flexural and punching shear capacities of tested specimens.

Specimen Specimen Flexural Capacity Shear Capacity (kN)

No. Name (kN) ACI 318-19 Eurocode 2
1 CON1 109.34 56.93 66.54
2 2S1H 109.34 49.26 56.00
3 2C1H 109.34 50.89 58.22
4 2R1H 109.34 41.59 46.04
5 2S4H 109.34 54.24 66.18
6 2C4H 109.34 54.47 66.21
7 2R4H 109.34 51.55 65.81
8 CON2 115.15 69.07 75.70
9 4S1H 115.15 50.47 51.72
10 4C1H 115.15 54.44 50.20
11 4R1H 115.15 31.89 29.07
12 4S4H 115.15 62.94 74.86
13 4C4H 115.15 63.11 74.94
14 4R4H 115.15 56.03 74.02

2.3. Material Properties

Mix design of concrete included Portland type I cement with a maximum aggregate
size of 9.5 mm. To test the compressive strength, 6 concrete cylinders (150 mm in diameter
to 300 mm in height) were prepared. Cylinders were tested by the Compression Testing Ma-
chine at department of civil engineering, faculty of engineering, King Mongkut’s University
of Technology Thonburi, Bangkok to determine the compressive strength ( f c0 ) following
the standard guides of ASTM standards ASTM C39/C39M [22]. The compressive strength
of concrete was 20.18 MPa and 29.71 MPa for group 1 and 2, respectively. Longitudinal
reinforcement comprised of 10 mm deformed steel bars having yield strength of 619 and
545 MPa for group 1 and 2, respectively. This could be due to the reason that steel bars were
ordered from different suppliers at different times. Steel bars were spaced at 175 mm on
center and placed in both orthogonal directions comprising a longitudinal reinforcement
ratio of 0.6% in each direction.

2.4. Load Setup

Experimental setup consisted of a hydraulic jack (with capacity of 60 tons), steel frame,
load cell, wide steel flanges, and concrete block as shown schematically in Figure 5 (see
Figure 6 for actual test configuration). Wide steel flanges were used to support all sides
of specimens. Hydraulic jack was used to apply monotonic load at the center of each
specimen. The hydraulic jack was placed on the top of the load cell to distribute the load
down until it broke through the surface of the specimen. Linear Variable Displacement
Transducers (LVDTs) were used to capture vertical deflection of each specimen as well as
to monitor the rotation in each specimen. Placement of LVDTs is illustrated in Figure 7.
Body of LVDT was affixed to a magnetic stand while its core touch slab specimen.
Buildings 2021,
Buildings 2021, 11,
11, 484 9 9of
of 24
24
Buildings2021,
Buildings 11,xxxFOR
2021,11, FOR PEER
FORPEER REVIEW
PEERREVIEW
REVIEW 9 9ofof2424

Figure
Figure 5.
5.5.Load
Figure5. Load Setup.
Setup.
LoadSetup.
Setup.
Figure Load

Figure 6.6.Actual
Figure6. test
Actualtest configuration.
testconfiguration.
configuration.
Figure
Figure Actual
6.Actual test configuration.

Figure 7.7.Location
Figure7. &&placement
Location& of
placementof LVDTs.
ofLVDTs.
LVDTs.
Figure
Figure 7.Location
Location &placement
placement of LVDTs.

3.3.Experimental Results
3.3. Experimental Results
ExperimentalResults
Experimental Results
3.1.
3.1. Failure
Failure
3.1.Failure Modes
Modes
FailureModes and
and
Modesand Observations
Observations
andObservations
Observations
3.1.
Crack patterns
Crack patternsof ofgroup
group11and and 2are arepresented
presentedininFigures
Figures8–11.8–11. For the specimens
Crackpatterns
Crack patternsofofgroupgroup11and and222arearepresented
presentedin inFigures
Figures8–11. For
8–11.ForForthethe specimens
thespecimens
specimensin in
in
in
group group
1, 1, first
first cracks
cracks initiated
initiated from
from thethe face
face of
of columns.
columns. With
With further
further increase
increase in load,
in load,
group1,1,first
group firstcracks
cracksinitiated
initiatedfrom fromthe thefacefaceofofcolumns.
columns.With Withfurther
furtherincrease
increaseininload,
load,
cracks propagated
cracks propagated towards
towards the the side
side ofof slab
slab specimens.
specimens. ItItcan
canbe beseen
seenin inFigures
Figures 88 and
and 99
crackspropagated
cracks propagatedtowards
towardsthe theside
sideofofslab
slabspecimens.
specimens.ItItcancanbe beseen
seenininFigures
Figures88and and99
that the
that thedistance
distanceofofopenings
openings fromfrom column’s
column’s faceface
had had significant
significant impact
impact on onfailure
the the failure
zone
that
that the
the distance
distance of openings
of openings from
from column’s
column’s face
face had
had significant
significant impact
impact on
on thethe failure
failure zone
zone
zone
of of specimens
specimens (dark (dark
solid solid
lines). lines).
Failure Failure
zone zone
was was narrower
narrower and and
closer closer
to the to the
column column
when
ofof specimens
specimens (dark
(dark solid
solid lines). Failure
lines).atFailure zone
zoneof was
was narrower
narrower and
anddepthcloser
closerfrom to
to thethe column
column when
when
when
openings openings
were were
located located
atataadistance a distance
of 1 times of slab the column face
openingswere
openings werelocated
locatedat a distanceof
distance of111times
times
timesofof slab
ofslab depth
slabdepth from
depthfrom
fromthethe column
thecolumn
columnface face (Figures
face(Figures
(Figures
(Figures
8b–d and 8b–d
9a). and
This 9a). This
signifies signifies
that that
relatively relatively
smaller smaller
slab slab
area area contributed
contributed towards towards
shear
8b–dand
8b–d and9a).
9a).This
Thissignifies
signifiesthat thatrelatively
relativelysmaller
smallerslab
slabarea
areacontributed
contributedtowards towardsshear
shear
Buildings 2021, 11, 484 10 of 24

Buildings 2021, 11, x FOR PEER REVIEW 10 of 24

shear strength when openings were placed nearer the column. Further, for openings at
1H from the column, cracks also initiated from the openings that further deteriorated
strength
the shear when openingsOn
capacity. were theplaced nearerspecimens
contrary, the column.with Further, for openings
openings located at 1H fromfrom col-
farther
theumn’s
column, cracks also initiated from the openings that further deteriorated
face (i.e., at 4H from column’s face) exhibited lesser cracking than their counterpart the shear
capacity. On the contrary, specimens with openings located farther from column’s face
specimens and it is interesting to note that no crack initiated from these openings as well
(i.e., at 4H from column’s face) exhibited lesser cracking than their counterpart specimens
(Figures 8e,f and 9b). Specimens in group showed similar trend as those in group 1 as
and it is interesting to note that no crack initiated from these openings as well (Figures
shown in Figures 10 and 11. Cracks intensity and propagation were more severe when
8e,f and 9b). Specimens in group showed similar trend as those in group 1 as shown in
openings were placed nearer to the column. Analogous to 4H openings in group 1, cracks
Figures 10 and 11. Cracks intensity and propagation were more severe when openings
didplaced
were not initiate
nearerfrom to theopenings at 4H fromtocolumn’s
column. Analogous 4H openings faceiningroup
group1,2cracks
as well.didAnot number of
initiate from openings at 4H from column’s face in group 2 as well. A number of openings2 to 4. For
openings also had an adverse effect on shear resistant zone when increased from
instance,
also specimen
had an adverse 2C1H
effect hadresistant
on shear 2 circular zoneopenings at column’s
when increased from 2opposite side while 4C1H
to 4. For instance,
had four circular openings at each side of the column.
specimen 2C1H had 2 circular openings at column’s opposite side while 4C1H had Comparing their crack patterns at
four
failure,
circular it is interesting
openings at each side to observe that shear
of the column. failuretheir
Comparing zonecrack
extended farther
patterns on sides
at failure, it where
is no openings
interesting were provided
to observe that shear(casefailure ofzone
2C1H). Conversely,
extended farther on shear
sidesfailure
wherezone no open-extended to
ings weredistance
equal providedfrom (caseallof 2C1H).
four sidesConversely,
of column shearinfailure
case of zone extended
4C1H. This to equal explains
clearly dis- the
tance
limit from
thatall four sides
openings of column
impose on the in shear
case ofcapacity
4C1H. This when clearly
placed explains
nearerthe to limit that
the column’s face.
openings
Comparing impose theoncrack
the shear capacity
patterns for when placed
different nearer shapes,
opening to the column’s face.openings
rectilinear Compar- allowed
ingthe
theinitiation
crack patterns for different
of cracks from allopening
of theirshapes,
corners. rectilinear
Further,openings allowed the
crack formation frominiti-
rectilinear
ation of cracks
openings wasfrom all of their
denser corners. Further,
as compared crack
to circular formation
section from rectilinear
(Figures 10g and 11b). openings
was denser ACIas318-19compared to circular section
recommends decreasing(Figures 10g andstrength
punching 11b). by reducing critical shear
ACI 318-19 recommends decreasing punching strength by reducing critical shear pe-
perimeter by an amount that is enclosed by straight lines starting from the centerline of
rimeter by an amount that is enclosed by straight lines starting from the centerline of col-
column to the openings boundary as demonstrated in earlier sections. An important point
umn to the openings boundary as demonstrated in earlier sections. An important point is
is that this reduction should be made for an opening located at a distance smaller than 4H
that this reduction should be made for an opening located at a distance smaller than 4H
from
from column’s
column’s face,face,
wherewhere“H” “H” is thickness.
is slab slab thickness. This highlights
This highlights that an opening
that an opening located located
farther than 4H from column’s face does not affect the
farther than 4H from column’s face does not affect the punching capacity. This explana- punching capacity. This explanation
can be related to the crack patterns obtained in this study.
tion can be related to the crack patterns obtained in this study. For openings located at 1H For openings located at 1H
from column’s face, primary cracks (bold black lines)
from column’s face, primary cracks (bold black lines) were located at the same location were located at the same location
indicating
indicating thatthat
onlyonly a reduced
a reduced slab around
slab area area around
column column
resistedresisted
punching punching shear. Shifting
shear. Shifting
thethe openings
openings awayaway fromfrom column’s
column’s face resulted
face resulted in a wider
in a wider periphery periphery
of primary of primary
cracks cracks
emphasizing
emphasizing on the
on the availability
availability of even
of even largerlarger slabtoarea
slab area resisttopunching
resist punching shear. Another
shear. Another
observation
observation mademade is that
is thatplacing
placing openings
openings on on
onlyonlytwotwo sides
sidesof of
columns
columns resulted
resulted in ain a wider
wider extension
extension of primarycracks
of primary crackson on sides
sides without
withoutopenings
openings (for(for
instance, see crack
instance, pat- pattern
see crack
tern
of of 2S1H).This
2S1H). Thisstates
states that
that thethesides
sideswithout
without openings
openings contribute
contributemoremore
towards resist- resisting
towards
ingpunching
punchingshear shearthanthan thethe sides
sides with
with openings.
openings.This Thistransformation
transformation in the
in thebehavior
behavior of of slabs
slabs from bi-directional to unidirectional action has been reported
from bi-directional to unidirectional action has been reported [16]. In addition, in all slab,[16]. In addition, in all
slab, severe cracks were also observed on the compression side along with the punching
severe cracks were also observed on the compression side along with the punching of the
of the loading plate into the slab as shown in Figure 12.
loading plate into the slab as shown in Figure 12.

(a) (b)

Figure 8. Cont.
Buildings 11,x484
2021,11,
Buildings 2021, FOR PEER REVIEW 11 of 24 11 of 24

Buildings 2021, 11, x FOR PEER REVIEW 11 of 24

(c) (d)

(c) Specimen 2C1H (d) Specimen 2R1H

(e) (f)

(e) Specimen 2S4H (f) Specimen 2C4H

(g)
Figure 8. Crack patterns at failure for group 1 (Line diagrams).
Figure 8. Crack patterns at failure for group 1 (Line diagrams). (a) Specimen CON1; (b) Spec-
imen 2S1H; (c) Specimen (g) Specimen
2C1H;2R4H(d) Specimen 2R1H; (e) Specimen 2S4H; (f) Specimen 2C4H;
Figure 8. Crack patterns
(g) Specimen 2R4H. at failure for group 1 (Line diagrams).

(a) Specimen 2C1H (b) Specimen 2R4H

Figure 9. Crack patterns
(a) at failure for group 1 (Actual photos).
(b)
Figure 9. Crack patterns at failure for group 1 (Actual photos).
Figure 9. Crack patterns at failure for group 1 (Actual photos). (a) Specimen 2C1H; (b) Specimen 2R4H.
Buildings 2021, 11, x FOR PEER REVIEW 12 of 24
Buildings 2021, 11, 484 12 of 24

(a) (b)

(c) (d)

(e) (f)

(g)
Figure 10. Crack
Figure patterns
10. Crack at failure
patterns for group
at failure 2 (Line2diagrams).
for group (Line diagrams). (a) Specimen CON2; (b) Spec-
imen 4S1H; (c) Specimen 4C1H; (d) Specimen 4R1H; (e) Specimen 4S4H; (f) Specimen 4C4H;
(g) Specimen 4R4H.
Buildings 2021, 11, 484 13 of 24
Buildings 2021, 11, x FOR PEER REVIEW 13 of 24
Buildings 2021, 11, x FOR PEER REVIEW 13 of 24

(a) (b)
(a) Specimen 4S1H (b) Specimen 4R4H
Figure
Figure11.11.
Crack patterns
Crack at at
patterns failure forfor
failure group 2 (Actual
group photos).
2 (Actual photos). (a) Specimen 4S1H; (b) Specimen 4R4H.
Figure 11. Crack patterns at failure for group 2 (Actual photos).

Figure 12. Typical cracks and punching of steel plate on compression side.
Figure 12.
Figure 12. Typical
Typical cracks
cracksand
andpunching
punchingofofsteel plate
steel onon
plate compression side.
compression side.
3.2. Ultimate Rotation
3.2.
3.2. Ultimate Rotation
Ultimate Rotation
Rotation of each slab at failure was approximated by using the LVDT values. Figure
13 showsRotation of eachrotation
the ultimate slab
slab atat failure
(Ø ) ofwas
failure was
eachapproximated
specimen or by
approximated byusing
the usingthe
average the LVDT
LVDT
angle invalues.
values. Figure
Figure 13
each speci-
13 shows
shows
men thethe
at failure. ultimate
ultimate
It can be rotation
rotation
seen that(O(Ø ) of
) of
uthe eachspecimen
each
ultimate specimen
rotationsor or the
ofthe average
when angle
average
slabs angle in
openings in each
werespeci-
each specimen
lo-
men
at
cated at 4-time
at failure.
failure. It can It be
canseen
thickness be seen thethat
ofthat the
slab theinultimate
ultimate rotations
both rotations
groups is of of
very slabs
slabs when
when
close openings
openings
to that of the werewere located
controllo-
cated
at at 4-time
4-time
specimen. thickness
However, thickness
of the of
reducing the
slab slab
bothingroups
theindistance both
to 1Hgroups
is isclose
very to
verycolumn’s
from close
thatto
face ofthat
the of the
control
drastically control
specimen.
affected
specimen.
However,
the However,
ultimatereducing reducing
the
rotation capacity ofthe
distance thetodistance
1H from
specimens.tocolumn’s
1H from column’s
Further, face face
drastically
ultimate drastically
rotation of slabsaffected
affected the ultimate
with
the ultimate
circular openingsrotation
were capacity
highest of
(among the specimens.
slab with Further,
openings), ultimate
followed rotation
rotation capacity of the specimens. Further, ultimate rotation of slabs with circular openingsby their of slabs
counterpart with
circular
slabs
were with openings
square
highest (among were
and highest
with(among
rectangular
slab openings,
openings),slabrespectively.
with openings),
followed byThis followed
their trend is by
more
counterpart their counterpart
dominant
slabs withinsquare
and rectangular openings, respectively. This trend is more dominant in the slabsinwith
slabs
the slabswith
withsquare
two and rectangular
openings. openings, respectively. This trend is more dominant
the slabs
two with two openings.
openings.
 Buildings2021,
Buildings 11,x 484
2021,11, FOR PEER REVIEW 14 of 2414 of 24

(a)

(b)
Figure 13.13.
Figure Ultimate rotations
Ultimate of slabs
rotations in (a)
of slabs in2(a)
openings (b) 4(b)
2 openings openings.
4 openings.

3.3. Load
3.3. vs.vs.
Load Deflection
Deflection
Experimental
Experimental results in terms
results of peak
in terms loads
of peak corresponding
loads to punching
corresponding shearshear
to punching for eachfor each
specimen along with their corresponding maximum deflections are listed
specimen along with their corresponding maximum deflections are listed in Table in Table 3. It3.isIt is to
to be
bementioned
mentioned that
that the
the tested
tested 22groups
groups of
ofspecimens
specimens had
had different
different concrete
concrete strengths
strengths since
since
bothboth concrete
concrete groups
groups belonged
belonged to different
to different batches
batches ordered
ordered at different
at different times.
times. To for
To allow
allow for the comparison
the comparison between between
two two groups,
groups, experimental
experimental strength
strength was was normalized
normalized to to
theoretical
0.33 f c0 b0.33 𝑓 𝑏 𝑑.
p
theoretical
strength strength o d.
Buildings 2021,2021,
Buildings 11, x11,
FOR484PEER REVIEW 15 of 15
24of 24

Table 3. Summary of test results.

Table 3. Summary of test results.
Compressive Peak
Specimen Specimen Compressive Peak
Specimen No. Strength
Specimen Name of
Strength of Load Load Deflection FailureFailure
Mode Mode
No. Name Concrete (MPa)
Deflection
Concrete (MPa) (kN) (kN) (mm) (mm)
1 1 CON1 CON1 20.18 20.18 57.73 57.73 7.745 7.745 Punching Punching
ShearShear
2 2S1H 20.18 49.09 5.479 Punching Shear
2 3 2S1H 2C1H 20.18 20.18 49.09 51.10 5.479 5.835 Punching Shear
Punching Shear
4 2R1H 20.18 44.71 4.845 Punching Shear
3 5 2C1H 2S4H 20.18 20.18 51.10 57.04 5.835 7.616 Punching Shear
Punching Shear
4 6 2R1H 2C4H 20.18 20.18 44.71 56.92 4.845 7.802 PunchingPunching
ShearShear
7 2R4H 20.18 55.50 7.236 Punching Shear
5 8 2S4H CON2 20.18 29.71 57.04 70.15 7.616 11.829Punching Shear
Punching Shear
9 4S1H 29.71 45.84 5.677 Punching Shear
6 10 2C4H 4C1H 20.18 29.71 56.92 49.06 7.802 7.244 Punching Shear
Punching Shear
11 4R1H 29.71 38.89 6.029 Punching Shear
7 12
2R4H 4S4H
20.18 29.71
55.50 67.61
7.236 9.299
Punching Shear
Punching Shear
8 13 CON2 4C4H 29.71 29.71 70.15 66.34 11.829 12.195Punching Shear
Punching Shear
14 4R4H 29.71 63.85 9.847 Punching Shear
9 4S1H 29.71 45.84 5.677 Punching Shear
10 4C1H 29.71 49.06 7.244 Punching Shear
3.3.1. Location of Openings
11 4R1H 29.71 38.89 6.029 Punching Shear
12 Figure 14 illustrates the 29.71
4S4H comparison of load 67.61vs. deflection
9.299curves of specimens
Punching Shearwith
openings at 4H and 1H from column’s face, respectively. As evident from Figure 14b,
13 4C4H 29.71 66.34 12.195 Punching Shear
placing openings near the column’s face (at 1H) irrespective of their shape had an adverse
14 4R4H 29.71 63.85 9.847 Punching Shear
effect on the punching shear capacity of the slabs. When the distance of openings was
increased from 1H to 4H (where H is slab thickness), peak loads sustained by each shape
3.3.1. Location of Openings
of openings were found to be very close to the peak load of the control specimen (see
Figure14a).
Figure 14 illustrates the comparison
This emphasizes the risk ofof placing
load vs.openings
deflection curves
nearer thanof specimens
4H from the with
face of
openings
column. at The
4H and
reason1Hcanfrom column’s to
be attributed face,
the respectively. As evident
decreased critical perimeter from
whenFigure 14b,
the openings
placing
wereopenings
placed at 1H nearthatthe column’s face
substantially reduced(at shear
1H) irrespective
capacities of of their shape
respective slabs. had an
Specimens
adverse effect on the punching shear capacity of the slabs.
in group 2 (see Table 3) exhibited similar response as those in group 1. Placing 4 openings
at 1H deteriorated shear capacity more severely than 4 openings at 4H from column’s face.
These results agree well with the previous findings [9,11].

(a)

Figure 14. Cont.

Buildings 2021, 11, x FOR PEER REVIEW 16 of 24

Buildings 2021, 11, 484 16 of 24

(a)

(b)
Figure 14. Normalized
Figure load-displacement
14. Normalized curves
load-displacement (a) Openings
curves at 4Hat(b)
(a) Openings 4HOpenings at 1H.
(b) Openings at 1H.

3.3.2.3.3.2.
Shape Shape of Openings
of Openings
Specimens
Specimens numbered
numbered fromfrom1 to 17 to 7 belong
belong to theto 1st
thegroup
1st group
havinghaving 2 openings
2 openings whilewhile
the rest of the specimens belong to group 2 each having 4 openings.
the rest of the specimens belong to group 2 each having 4 openings. The control specimen The control specimen
in group
in group 1 (CON1)
1 (CON1) was was
able able to sustain
to sustain highest
highest loadload i.e., 57.73
i.e., 57.73 kN. kN. For same
For the the same distance
distance
from column’s face (group 1), specimens with circular openings were
from column’s face (group 1), specimens with circular openings were able to resist highest able to resist highest
loadsloads followed
followed by square
by square and and rectangular
rectangular openings,
openings, respectively.
respectively. For instance,
For instance, peakpeak
loadload
resisted by specimen 2C1H was 51.10 kN whereas corresponding
resisted by specimen 2C1H was 51.10 kN whereas corresponding values for specimens values for specimens
2S1H2S1Hand and
2R1H 2R1H
werewere 49.09
49.09 andand 44.71
44.71 kN,
kN, respectively.AAsimilar
respectively. similar trend
trend was
was observed
observedininpeak
deflection values as circular specimen underwent highest
peak deflection values as circular specimen underwent highest deflections followeddeflections followed by by
square
and rectangular specimens, respectively. As illustrated in Figure
square and rectangular specimens, respectively. As illustrated in Figure 15, slabs with cir-15, slabs with circular
cularopenings
openingsoutperformed
outperformedslabs slabswith
withsquare
squareandandrectangular
rectangularopenings.
openings.TheseThesefindings
findings agree
with the previous findings [17] where circular openings had a
agree with the previous findings [17] where circular openings had a lower deteriorating lower deteriorating effect
effectononpunching
punchingcapacities
capacitiesthan
thanrectangular
rectangularopenings
openingsplacedplacedatatthe
thesame
samelocation
locationininflat
flatslabs.
There can be two reasons behind this. Firstly, rectangular openings had the largest area
slabs. There can be two reasons behind this. Firstly, rectangular openings had the largest
followed by square and circular openings, respectively. Secondly, it has been known that
area followed by square and circular openings, respectively. Secondly, it has been known
stress concentrations at the corners of rectilinear sections adversely affect their performance
that stress concentrations at the corners of rectilinear sections adversely affect their per-
in comparison to circular shape where a uniform stress distribution can be assumed. For
formance in comparison to circular shape where a uniform stress distribution can be as-
the openings located at 4H from the column’s face, peak loads sustained by circular and
sumed. For the openings located at 4H from the column’s face, peak loads sustained by
square openings were comparable. However, rectangular openings still created the lower
circular and square openings were comparable. However, rectangular openings still cre-
bound in terms of peak load. This can be attributed to larger areas of rectangular openings
ated the lower bound in terms of peak load. This can be attributed to larger areas of rec-
in comparison with circular and square shape openings.
tangular openings in comparison with circular and square shape openings.
Buildings 2021, 11, x FOR PEER REVIEW 17 of 24
Buildings 2021, 11,
Buildings2021, 11, x484
FOR PEER REVIEW 17 of 24 17 of 24

Figure 15. Comparison between square, circle, and rectangle shape.

Figure
Figure 15.15. Comparison
Comparison between
between square,
square, circle,circle, and rectangle
and rectangle shape. shape.
3.3.3. Number of Openings
3.3.3.
3.3.3. Number
Number of Openings
of openings
Openings
The number of also had an adverse impact upon the shear capacity. For
The
comparison, The number
number
reduction of openings
ofinopenings
peak loads (atalso
also had had
1H) an an adverse
for adverse
circular, impact impact
square, upon upon
the thecapacity.
shear
and rectangular shear capacity.
openings For For
comparison,
comparison,
in group reduction
1 were reduction in
in peak
11.48, 14.97, peak loads
loads (at
and 22.44% (at 1H) for
1H) for circular,
respectively circular, square,
square,16a).
(see Figure and Peak and rectangular
rectangular openings
loads for the openings
sameinin group
group
specimens 1 but
1 werewere 11.48,
11.48,
with 14.97,
14.97, and and
4 circular, 22.44%
22.44%
square, and respectively
respectively (see Figure
(see Figure
rectangular 16a).(at
openings 16a).
Peak Peak
1H)loads
were forloads
re-the for the
duced same
same specimens
by specimens
30.06, 36.65,butbut
and with
with 4 circular,
4 circular,
44.56%, square,
square,
respectively and
and and rectangular
asrectangular
shown openings
16b.(at (at
openings
in Figure 1H)1H)
were were
re- reduced
by 30.06,
duced 36.65,
by 30.06, andand
36.65, 44.56%,
44.56%,respectively
respectively and
andasasshown
shownin in Figure 16b.
Figure 16b.

(a)

Figure 16. Cont.

Buildings 2021, 11, x FOR PEER REVIEW 18 of 24

Buildings 2021, 11, 484 18 of 24

(a)

(b)
Figure 16. Punching shear capacity of the different the number of openings (a) Openings at 4H (b)
Figure 16. Punching shear capacity of the different the number of openings (a) Openings at 4H (b)
Openings at 1H.
Openings at 1H.
4. Comparison between
4. Comparison Experimental
between ResultsResults
Experimental & Analytical Predictions
& Analytical Predictions
The two approaches used to predict the punching shear capacity of tested specimens
The two approaches used to predict the punching shear capacity of tested specimens
are based upon the recommendations of ACI 318-19 and Eurocode 2. Detailed explana-
are based upon the recommendations of ACI 318-19 and Eurocode 2. Detailed explanations
tions of the calculation procedures are already presented in previous sections. Here, first
of the calculation procedures are already presented in previous sections. Here, first the
the major differences between these two codes are presented. ACI 318-19 states that the
major differences between these two codes are presented. ACI 318-19 states that the critical
critical perimeter for punching shear is located at a distance of d/2 from column’s face
perimeter
(Figure 17) whileforthe
punching
same is shear
locatedis at
located
2d from at column’s
a distanceface
of d/2 from
as per column’s
Eurocode face (Figure 17)
2 recom-
while the same is located at 2d from column’s face as per Eurocode 2
mendations (where d is the effective depth of the slab) as shown in Figure 18. Further, an recommendations
(where
opening candhave
is the effective effect
significant depthon ofpunching
the slab) as shownupintoFigure
capacity 18. Further,
a distance of 4H andan 6d
opening can
have
(where “H”significant
and “d” effect onthickness
are slab punchingand capacity updepth,
effective to a distance of 4H from
respectively) and 6d (where “H” and
column’s
face“d” are slab
defined thickness
by ACI and Eurocode
318-19 and effective depth, respectively)
2, respectively. fromofcolumn’s
Distances openingsface
weredefined
cho- by ACI
318-19
sen to and Eurocode
be within the domain2,specified
respectively.
by theseDistances
codes. of openings were chosen to be within the
domain specified
Table 4 presents by these codes.
a comparison between experimental and analytical predicted punch-
ing shear capacities of all specimens. It was found that the average difference be-tween
experimental and analytical predictions of ACI 318-19 was 4% while the same for Euro-
code 2 was 7%. Figure 19 presents the variation of theoretical predictions from experi-
mental results. Punching strengths of control specimens CON1 and CON2 predicted by
ACI 318-19 were offset only by −1 and −2%, respectively, from their corresponding exper-
imental values. The same had differences of +13 and +7%, respectively, for Eurocode 2
predictions. ACI 318-19 predictions for control specimens (without openings) were not
only closer to the experimental results than Eurocode 2 predictions but they also provided
more conservative results. Among group 1 specimens, the largest percentage difference
obtained in ACI 318-19 predicted strengths was −8%. The counterpart value for Eurocode
2 results was +16%. It was also observed that ACI 318-19 consistently predicted conserva-

Buildings 2021, 11, 484
4R1H. It is to be mentioned that the error obtained in theoretical predictions of both codes
varied with the shape of openings irrespective of their number and location. Therefore,
this greatly emphasizes on the need of further research to account for the shape of open-
ings in determining their deteriorating effect on punch capacities of flat slabs. Statistically,
ACI 318-19 recommendations yielded a slightly lower standard deviation from experi-
mental results than Eurocode 2. Further, the mean of the ratio of experimental to analytical19 of 24
results was 1.04 and 0.93 for ACI 318-19 and Eurocode 2, respectively.

Buildings 2021, 11, x FOR PEER REVIEW 20 of 24

Figure
Figure 17. 17. Definition
Definition of critical
of critical perimeter
perimeter as ACI
as per per ACI 318-19.
318-19.

Figure
Figure 18.18. Definition
Definition of of critical
critical perimeter
perimeter as per
as per Eurocode
Eurocode 2. 2.

Table 4. Comparison between experimental and analytical punching shear strengths.

Analytical Punching Ratio of Experimental to

𝒇𝒄 ′ Experimental Reduction
No. Name Capacity (kN) Analytical
(MPa) (kN) (%)
ACI318 Eurocode 2 ACI318 Eurocode 2
Buildings 2021, 11, 484 20 of 24

Table 4 presents a comparison between experimental and analytical predicted punch-

ing shear capacities of all specimens. It was found that the average difference be-tween
experimental and analytical predictions of ACI 318-19 was 4% while the same for Eurocode
2 was 7%. Figure 19 presents the variation of theoretical predictions from experimental
results. Punching strengths of control specimens CON1 and CON2 predicted by ACI 318-
19 were offset only by −1 and −2%, respectively, from their corresponding experimental
values. The same had differences of +13 and +7%, respectively, for Eurocode 2 predictions.
ACI 318-19 predictions for control specimens (without openings) were not only closer to the
experimental results than Eurocode 2 predictions but they also provided more conservative
results. Among group 1 specimens, the largest percentage difference obtained in ACI
318-19 predicted strengths was −8%. The counterpart value for Eurocode 2 results was
+16%. It was also observed that ACI 318-19 consistently predicted conservative strengths
among group 1 specimens. On the contrary, Eurocode 2 constantly overestimated punching
capacities of all the specimens in group 1. Apart from specimens 4S1H and 4C1H in group
2, ACI 318-19 predictions were conservative for the remaining specimens while Eurocode 2
again overestimated punching strengths of all specimens but 4R1H. It is to be mentioned
that the error obtained in theoretical predictions of both codes varied with the shape of
openings irrespective of their number and location. Therefore, this greatly emphasizes on
the need of further research to account for the shape of openings in determining their dete-
riorating effect on punch capacities of flat slabs. Statistically, ACI 318-19 recommendations
yielded a slightly lower standard deviation from experimental results than Eurocode 2.
Further, the mean of the ratio of experimental to analytical results was 1.04 and 0.93 for
ACI 318-19 and Eurocode 2, respectively.

Table 4. Comparison between experimental and analytical punching shear strengths.

0 Analytical Punching Ratio of Experimental to

fc Experimental Reduction Capacity (kN) Analytical
No. Name
(MPa) (kN) (%)
ACI318 Eurocode 2 ACI318 Eurocode 2
1 CON1 20.18 57.73 - 56.93 66.54 1.01 0.87
2 2S1H 20.18 49.09 14.97 49.26 56.00 1.00 0.88
3 2C1H 20.18 51.10 11.48 50.89 58.22 1.00 0.88
4 2R1H 20.18 44.71 22.55 41.59 46.04 1.08 0.97
5 2S4H 20.18 57.04 1.20 54.24 66.18 1.05 0.86
6 2C4H 20.18 56.92 1.40 54.47 66.21 1.04 0.86
7 2R4H 20.18 55.50 3.86 51.55 65.81 1.08 0.84
8 CON2 29.71 70.15 - 69.07 75.70 1.02 0.93
9 4S1H 29.71 45.84 34.65 50.47 51.72 0.91 0.89
10 4C1H 29.71 49.06 30.06 54.44 50.20 0.90 0.98
11 4R1H 29.71 38.89 44.56 31.86 29.07 1.22 1.34
12 4S4H 29.71 67.61 3.62 62.94 74.86 1.07 0.9
13 4C4H 29.71 66.34 5.43 63.11 74.94 1.05 0.89
14 4R4H 29.71 63.85 8.98 56.03 74.02 1.14 0.86
Average 1.04 0.93
Standard Deviation (SD) 0.08 0.12
Buildings
Buildings2021,
2021,11,
11,x 484
FOR PEER REVIEW 21 21
ofof2424

50
ACI318-19
40 Eurocode 2

Variation (%) from Experimental Results

30

15.7
14.0
13.8
20

13.7
13.2

12.3

12.2

11.5
11.4

9.9.
9.2.

9.7
7.3
10

2.9

2.3
0.3.
0

-0.4
-1.4

-1.6
-4.5

-5.1
-5.2
-10

-7.4
-7.5

-7.7

-14.0
-20

-22.1
-30

-33.8
-40

-50
CON1 2S1H 2C1H 2R1H 2S4H 2C4H 2R4H CON2 4S1H 4C1H 4R1H 4S4H 4C4H 4R4H

Specimen ID

Figure19.
Figure Variation(%)
19.Variation (%)ofoftheoretical
theoreticalpredictions
predictionsfrom
fromexperimental
experimentalresults.
results.

5. Discussion
5. Discussion
This study investigated into the effects of openings on punching shear capacity of flat
This study investigated into the effects of openings on punching shear capacity of flat
slabs. For this purpose, 14 flat slab specimens were tested in 2 groups depending upon
slabs. For this purpose, 14 flat slab specimens were tested in 2 groups depending upon
the concrete strength. The novelty of this research lies in the adoption of different shapes
the
of concrete
openingsstrength. The novelty
for investigating their ofeffects
this research lies in capacities
on punching the adoption of different
of flat shapes
slabs. Although,
ofitopenings for investigating their effects on punching capacities
has been known from literature and current design codes (see earlier sections) that of flat slabs. Although,
itthe
hasdistance
been known of anfrom
openingliterature and current
has inverse relation design
with codes (see earliereffect
its deteriorating sections) that the
on punching
distance of an opening has inverse relation with its deteriorating
capacity. However, this effect has not been related with the shape of the opening in the past. effect on punching ca-
pacity. However,
This study considersthis effect has not
all possible been (including
factors related with the shape
distance from ofcolumn’s
the opening face,inshape,
the past.and
This study considers all possible factors (including distance
their number) associated to the openings around columns in flat slabs to study their effectsfrom column’s face, shape,
and their number)
on punching associated to the openings around columns in flat slabs to study their
capacities.
effectsIton is found thatcapacities.
punching the shape of opening actually affects punching capacity of flat slabs.
Circular,found
It is square, thatandtherectangular
shape of openingopenings actually
were usedaffects punching
in this study. capacity
For the sameof flatdistance
slabs.
Circular, square,face,
from column’s and rectangular
circular openings openings were used
are found to havein this
least study.
effectFor
on the same distance
punching strength,
from column’s
followed face, circular
by square openings are
and rectangular found to
openings, have least effect
respectively. Stressonconcentrations
punching strength, at the
followed
corners of bysquare
square andand rectangular
rectangular openings,
openings respectively.
can be the catalyst Stressfor thisconcentrations
excessive degradationat the
corners of square
of punching and rectangular
strength. openings
Further, circular can have
shape be thesmaller
catalystarea for than
this excessive degrada-
square opening for
tion of punching strength. Further, circular shape have
the same diameter and side dimension. The second factor related to openings was smaller area than square opening
their
for the same
number. Indiameter
this study, and2 side
and dimension.
4 openingsThe were second
used factor
around related to openings
column’s periphery. wasIttheir
was
number.
expectedInfrom this the
study,
slabs2 with
and 4 openings to were
haveused around
smaller column’s
punching periphery.
strength than the It was
slabs
expected from theResults
with 2 openings. slabs with have 4 confirmed
openings to have
this as insmaller
both the punching
groups of strength than increasing
test matrix, the slabs
with 2 openings.
the openings Results affected
adversely have confirmed
the punchingthis asstrength.
in both the Thegroups of test matrix,
last parameter increas-
considered was
ing
thethe openings
distance adverselyfrom
of openings affected the punching
the column’s face.strength.
Openings The last placed
were parameter at 1 considered
and 4 times
the the
was slabdistance
thickness of “H” from from
openings column’s face. Results
the column’s face. suggest
Openings that openings
were placeddo at not
1 and have4
significant
times the slab effect on punching
thickness “H” from strength when face.
column’s placed at 4H from
Results suggest column’s face. However,
that openings do nota
substantial
have significantdecrease
effectin onpunching
punchingcapacity
strengthwas whenobserved
placed for at 4H slabs with
from openings
column’s located
face. How-at
1H from
ever, column’sdecrease
a substantial face. in punching capacity was observed for slabs with openings
locatedCodeat 1H provisions
from column’s of ACIface.
318-19 and Eurocode 2 were deployed to predict the punching
capacities of flat slabs.
Code provisions of ACI 318-19It is found that
and ACI 318-19 slightly
Eurocode edges pasttoEurocode
2 were deployed predict the 2 inpunch-
predict-
ing punching capacities of flat slabs with openings. Nevertheless,
ing capacities of flat slabs. It is found that ACI 318-19 slightly edges past Eurocode 2 in ACI 318-19 predictions
were constantly
predicting punching conservative
capacities whileof flat Eurocode
slabs with2openings.
overestimated punching
Nevertheless, strengths.
ACI 318-19 pre-ACI
318-19 states
dictions were that the punching
constantly strengthwhile
conservative of flat Eurocode
slabs must2beoverestimated
decreased by an amount
punching
that is proportional
strengths. ACI 318-19 to the that
states reduced critical perimeter.
the punching strength of Theflatamount
slabs mustby which the critical
be decreased by
Buildings 2021, 11, 484 22 of 24

perimeter must be reduced has already been stated in earlier sections. An important result
in this regard is that ACI 318-19 updated its clause about the distance from column’s face
within which an opening will have adverse effect on punching strength. This distance was
updated to 4H that previously was 10H in ACI 318-14. Findings of this research agree with
the recent adoption of 4H by ACI 318-19. Openings, irrespective of their shape and number,
had a negligible effect on punching capacity when they were placed at 4H from column’s
face. An important consideration in this regard is the shape of openings. Openings of same
number and at same distance from column’s face but with different shapes resulted in
different punching shear strengths. Though code provisions of ACI 318-19 and Eurocode 2
account for this effect by reducing critical perimeter by an amount that is a function of the
size of an opening, their predictions for different opening shapes resulted in large scatter.
The predictions of both codes for square and circular openings (same number of openings
and at same distance from column’s face) resulted in a close variation from experimental
results. For example, Eurocode 2 predictions for 2S1H and 2C1H (i.e., 2 square and circular
openings at 1H from column’s face, respectively) varied from experimental results by +12%.
But it resulted in a variation of +2.9% for the same number and distance of rectangular
openings. A similar trend could also be observed in ACI 318-19 predictions (see Figure 19).
Therefore, it is much desired to extend this area of research to extend the knowledge about
the effects of openings on punching capacities with special attention to their shapes.

6. Conclusions
An experimental study was conducted to study the effect of different shapes of
openings in flat slabs at different distances from a concentrically loaded column’s face.
A total of 14 specimens were tested in 2 groups each consisting of 7 specimens. Group
1 specimens were provided with 2 openings while 4 openings were provided in group
2 specimens. Effects of shape (circular, square, and rectangular), number (2 and 4), and
distance of openings from column’s face (1H and 4H where H is slab’s thickness) on
punching shear capacity were carefully investigated. Key conclusions drawn from this
study are summarized in this section.
1. In terms of the shape of openings, punching capacity of flat slabs were least affected
by circular openings followed by square and rectangular openings, respectively.
2. Openings provided at 1H from column’s face drastically reduced the punching shear
capacity of flat slabs. On the contrary, openings provided at 4H from column’s face
had little impact on slab’s shear capacity. Rectangular openings had the worst effect
on punching capacity for all opening shapes at 4H from column’s face. Rectangular
openings at 4H reduced punching capacity by 3.86 and 8.98% for 2 and 4 openings,
respectively. Reduction caused by circular and square and circular openings was
comparable at 4H from column’s face and irrespective of their number.
3. Irrespective of the shape of the openings, increasing the number of openings from
2 to 4 substantially reduced the punching capacity of slabs.
4. Ultimate rotation of slabs with openings at 1H from column’s face were significantly
lower than their counterpart slabs with openings at 4H from column’s face.
5. Punching shear capacities were analytically predicted following the descriptive equa-
tions of ACI 318-19 and Eurocode 2. Ratio of experimental to analytical predictions
of ACI 318-19 and Eurocode 2 were 1.04 and 0.93 suggesting that ACI 318-19 recom-
mendations are slightly closer to the real case scenarios. This reason can be attributed
to the different locations of critical punching shear perimeter from column’s face as
stated by the two codes.
6. ACI 318-19 consistently predicted conservative and more closer punching capacities
than those predicted by Eurocode 2. For openings located at same distance and
of same number, the variation of code predictions from experimental results was
marginal for square and circular openings. For instance, ACI 318-19 predicted punch-
ing strengths for specimens 2S4H and 2C4H to be −5 and −4% lower than their
respective experimental values. However, this was not true for rectangular openings
Buildings 2021, 11, 484 23 of 24

as this error increased to −8% for the specimen 2R4H. Similarly, the predictions of
ACI 318-19 for 2S1H and 2C1H resulted in a variation of about −0.4% from their
experimental results. Corresponding prediction for the rectangular openings (i.e., for
specimen 2R1H) resulted in a variation of −7.5%. A similar trend in the predictions
of Eurocode 2 was also observed. Therefore, further research is desired in area to ac-
count for the shape of openings in determining their deteriorating effect on punching
capacities of flat slabs.

Author Contributions: Conceptualization, E.Y., Y.T., T.W. and J.A.; methodology, E.Y., Y.T., T.W. and
J.A.; investigation, E.Y., Y.T., T.W. and J.A.; writing—original draft preparation, E.Y., Y.T., T.W., J.A.,
P.J. and Q.H.; writing—review and editing, E.Y., Y.T., T.W., J.A., P.J. and Q.H. All authors have read
and agreed to the published version of the manuscript.
Funding: This work was financially supported by Office of the Permanent Secretary, Ministry of
Higher Education, Science, Research and Innovation under Grant No. RGNS 63-093.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Acknowledgments: The authors also thank the staff of CIVIL-KMUTT for their technical support in
the laboratory.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Silva, J.A.; Marques, M.G.; Trautwein, L.M.; Gomes, R.B.; Guimarães, G.N. Punching of Reinforced Concrete Flat Slabs with Holes
and Shear Reinforcement. REM Int. Eng. J. 2017, 70, 407–413. [CrossRef]
2. Santos, G.S.; Nicácio, W.G.; Lima, A.W.; Melo, G.S.S.A. Punching Strengthening in Flat Plates of Reinforced Concrete with Carbon
Fiber Reinforced Polymer (CFRP). Rev. Ibracon Estrut. Mater. 2014, 7, 592–625. [CrossRef]
3. Son, K.-S.; Pilakoutas, K.; Neocleous, K. Behaviour of Concrete Columns with Drilled Holes. Mag. Concr. Res. 2015, 58, 411–419.
[CrossRef]
4. King, S.; Delatte, N.J. Collapse of 2000 Commonwealth Avenue: Punching Shear Case Study. J. Perform. Constr. Facil. 2004, 18,
54–61. [CrossRef]
5. Gardner, N.J.; Huh, J.; Chung, L. Lessons from the Sampoong Department Store Collapse. Cem. Concr. Compos. 2002, 24, 523–529.
[CrossRef]
6. El-Salakawy, E.F.; Polak, M.A.; Soliman, M.H. Reinforced Concrete Slab-Column Edge Connections with Shear Studs. Can. J. Civ.
Eng. 2011, 27, 338–348. [CrossRef]
7. Teng, S.; Cheong, H.K.; Kuang, K.L.; Geng, J.Z. Punching Shear Strength of Slabs with Openings and Supported on Rectangular
Columns. Struct. J. 2004, 101, 678–687. [CrossRef]
8. Borges, L.L.J.; Melo, G.S.; Gomes, R.B. Punching Shear of Reinforced Concrete Flat Plates with Openings. ACI Struct. J. 2013, 110,
547–556. [CrossRef]
9. Anil, Ö.; Kina, T.; Salmani, V. Effect of Opening Size and Location on Punching Shear Behaviour of Two-Way RC Slabs. Mag.
Concr. Res. 2015, 66, 955–966. [CrossRef]
10. Ha, T.; Lee, M.-H.; Park, J.; Kim, D.-J. Effects of Openings on the Punching Shear Strength of RC Flat-Plate Slabs without Shear
Reinforcement. Struct. Des. Tall Spec. Build. 2015, 24, 895–911. [CrossRef]
11. Balomenos, G.P.; Genikomsou, A.S.; Polak, M.A. Investigation of the Effect of Openings of Interior Reinforced Concrete Flat Slabs.
Struct. Concr. 2018, 19, 1672–1681. [CrossRef]
12. Liberati, E.A.P.; Marques, M.G.; Leonel, E.D.; Almeida, L.C.; Trautwein, L.M. Failure Analysis of Punching in Reinforced Concrete
Flat Slabs with Openings Adjacent to the Column. Eng. Struct. 2019, 182, 331–343. [CrossRef]
13. Genikomsou, A.S.; Polak, M.A. Effect of Openings on Punching Shear Strength of Reinforced Concrete Slabs—Finite Element
Investigation. Struct. J. 2017, 114, 1249–1261. [CrossRef]
14. Marques, M.G.; Liberati, E.A.P.; Gomes, R.B.; Almeida, L.C.; Trautwein, L.M. Study of Failure Mode of Reinforced Concrete Flat
Slabs with Openings and Studs. Struct. J. 2020, 117, 39–48. [CrossRef]
15. Anıl, Ö.; Ulusoy, B. Nonlinear FEA of Two-Way RC Slabs’ Punching Behavior with Openings. Iran. J. Sci. Technol. Trans. Civ. Eng.
2020, 44, 1109–1124. [CrossRef]
16. da Silva Lourenço, D.; Liberati, E.A.P.; Marques, M.G.; de Almeida, L.C.; Trautwein, L.M. Reinforced Concrete Flat Slabs with
Openings at Different Distances from the Column. Rev. Ibracon Estrut. Mater. 2020, 14, e14111. [CrossRef]
Buildings 2021, 11, 484 24 of 24

17. Bora, A.S.; Gaidhankar, D.G.; Kulkarni, M. Analysis of Flat Slab With And Without Opening. Int. J. Sci. Technol. Res. 2019, 8,
1455–1459.
18. ACI Committee 318 Aci 318-19; American Concrete Institute: Farmington Hills, MI, USA, 2019, ISBN 9781641950565.
19. EN 1992-1-1: Eurocode 2: Design of Concrete Structures-Part 1-1: General Rules and Rules for Buildings; British Standard Institution:
London, UK, 2004.
20. Genikomsou, A.; Polak, M.A. Finite Element Simulation of Concrete Slabs with various Placement and Amount of Shear Bolts.
Procedia Eng. 2017, 193, 313–320. [CrossRef]
21. Johansen, K. Yield-Line Formulae for Slabs; CRC Press: London, UK, 1972.
22. ASTM C39/C39M-14, Standard Test Method for Compressive Strength of Cylindrical Concrete Specimens; ASTM International: West
Conshohocken, PA, USA, 2014.