Alexandria Engineering Journal (2016) 55, 2665–2678

HOSTED BY
Alexandria University

Alexandria Engineering Journal

www.elsevier.com/locate/aej
www.sciencedirect.com

ORIGINAL ARTICLE

Structural design of isolated column footings

Fathi Abdrabbo, Zaki I. Mahmoud *, Mariana Ebrahim

Structural Eng. Dept., Faculty of Eng., Alexandria Univ., Egypt

Received 21 November 2015; revised 31 May 2016; accepted 9 June 2016

Available online 9 August 2016

KEYWORDS
Abstract Superstructure loads are transmitted to the underlying soil strata through a suitably designed
Footing; foundation. Therefore, the foundation of a structure is considered the most crucial struc-tural element in
Codes; a building. The foundation may be classified into two main categories, shallow and deep foundations.
Punching shear; Shallow foundation comprises isolated column footings, combined footings and reinforced concrete mat.
Shear span; The design of isolated column footing is accomplished through the appli-cation of geotechnical and
Correlation; structural analysis concepts. So that, the input research into isolated col-umn footings comes from two
Contact stress different disciplines, geotechnical and structural. This may be one of the main causes that attributed to
the limited research input to the subject. Therefore, the structural design of isolated column footings is
based on empirical rules and the calculations of bending moments (BM) and shearing forces (SF)
induced in a footing are based on the rules of beam theory, which is questionable. On the other hand,
punching theory was developed on relatively thin floor slab, even though the theory is implemented for
the calculation of punching shear in relatively thick footings. Also experimental research on isolated
column footings is scarce, due to the difficulties involved in the setup of the laboratory models and the
cost of experiments. The work presented in this article deals with the correlation between failure loads
predicted by different code provisions, ECP203-11, ACI318-08, BS 8110.1-1997 and EC2-2004, of
isolated column footings, and the cor-responding measured values.

The study showed that shear span to depth ratio of a footing and distributions of contact stress at
footing–soil interface are key factors in the structural design of the footing. ECP203-11, ACI318-08, and
EC2-2004 code provisions, underestimate the structural failure loads of isolated column foot-ings, while
BS 8110.1-1997 overpredicts the failure loads of isolated column footings, if punching provisions at
perimeter of column are pulled out from the code.
2016 Production and hosting by Elsevier B.V. on behalf of Faculty of Engineering, Alexandria
University. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/
licenses/by-nc-nd/4.0/).

1. Literature review [1,2]. The most important parameters that influence punching
shear are the effective or total footing depth, the flexural rein-
It was established that the failure mechanism of floor slab and forcement ratio, and compressive strength of concrete [1]. The
foundation plate depends on the shear slenderness ratio (a/d) angle of shear cracks of foundation plate is between 50 and 60

*
Corresponding author.
Peer review under responsibility of Faculty of Engineering, Alexandria University.
http://dx.doi.org/10.1016/j.aej.2016.06.016
1110-0168 2016 Production and hosting by Elsevier B.V. on behalf of Faculty of Engineering, Alexandria University.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
2666
Nomenclature
L length of footing
B width of footing
d depth of footing
c width/length of column stub
a shear span, distance from edge of column to edge
of footing
0
fc cylinder compressive strength of concrete
f
cu cube compressive strength of concrete
q reinforcement ratio
V experimental failure load
(1 vertical to 1.19–1.73 horizontal) which is significantly higher
than the angle for slender slabs, which varies between 30 and 40
(1 vertical to 0.57–0.83) [2].
The difficulties associated with laboratory modeling and
testing of isolated column footings, lead to that the experimen-tal
research input to the subject is quite scarce [3]. It is worth noting
that, the majority of technical regulations do not distin-guish
between punching through floor slabs and punching through
foundation slabs. A comparative analysis [4] indicated that
foundation slab failure mechanism is different when com-pared to
slender slabs. Theoretical explanation of the plate punching
phenomenon, based on the critical shear crack for the reinforced
concrete slab without and with transverse rein-forcement was
emphasized [5,6]. The theory is referred as crit-ical shear crack
theory (CSCT). The theory was recognized by new fib model
code 2010, Draft Bulletins 55 and 56. The differ-ence between the
punching mechanism of foundation plates and floor slabs has
generally been neglected in technical regu-lations [7]. This can be
attributed to that experimental research related to foundations has
so far been quite scarce, because of the complicated arrangement
of such experiments. Further-more, there is noticeable difference
in the calculations of punching loads given by different codes [7].
Experimental study on 17-column footing revealed that the shear
span/depth of footing, which is called shear slenderness ratio,
significantly affects the bearing capacity to punching-shear [7].
The punching failure through a footing is brittle, and the use of
shear reinforcement increases the punching capacity sig-
nificantly, and increases the ductility and the possibility of
redistribution of forces [8,14].
A review of the theoretical and experimental research work
including Codes/Regulations for punching calculation of col-umn
footings leads to that cracks pattern because punching depends
upon a/d ratio, in which cracks are inclined in case of column
footings with greater a/d ratio, than in case of col-umn footings
with a smaller a/d ratio [9]. In Switzerland, the shear
reinforcement in footing is calculated on the basis of the theory of
plasticity, according to SIA 262, and the contribu-tion of concrete
to punching capacity is neglected, which leads to conservative
calculation results for shear reinforcement [9].
2. Objectives of the research
The main objective of the presented work was to correlate
between the predicted structural failure loads of isolated column
footings, through the implementation of code
F. Abdrabbo et al.
cs material coefficient of steel
cc material coefficient of concrete
Pp predicted failure load due to punching shear
A contact area of footing
A
o punching contact area of footing at the level of
flexural steel enclosed by line at distance d from
the edge of the column
Sc one way shear capacity of footing
Ps predicted failure load based on one way shear
P
upc punching shear capacity of footing
provisions, and the measured failure loads documented in the
literature. A trial was given to adjust some code provisions during
the correlation process to obtain a better correlation results. ECP
203-11, ACI 318-08, BS 8110.1-1997 and EC2-2004 code
provisions are considered for the prediction of fail-ure loads. The
structural design of isolated column footing, most often is
controlled by punching shear induced in the foot-ing. So that, the
most attention is given to code provisions dealing with punching
shear. No attention was paid to the behavior of footings included
shear reinforcement due to the very limited experimental work on
such footings.
3. Procedure of the study
The work presented was accomplished through the following
steps:
1. The ECP 203-2011, ACI 318-08, BS 8110.1-1997 and EC2-
2004 provisions related to the structural design of isolated
column footings were used through spread sheets for the
calculation of ultimate failure loads.
2. The available previous work documented in literature was
reviewed, for experimental work on isolated column footings.
3. The laboratory work completed with enough data on class of
concrete, footing dimensions, failure load, reinforcement and
grade of steel was tabulated, as data base.
4. The predicted loads using code provisions were obtained using
experimental data. One way and punching shear according to
ECP 203-11, ACI 318-08, BS 8110.1-1997 provisions were
implemented in the prediction. Punching shear only of EC2-
2004 provision is considered in the study.
4. One way and punching shear code provisions
Code provisions consider two types of shear in the design of
reinforced concrete isolated column footings subjected to axial
loads, One-way shear and punching shear. The Egyptian code
provisions ECP 203-2011 defined the critical section of one-way
shear and punching shear at distance d/2 from the edge of the
column as shown in Fig. 1. ACI (318-08) provisions con-sidered
critical section for one-way shear at distance d from the edge of
the column and punching shear at distance d/2. BS (8110-1997)
provisions considered the control section of one-way shear at
distance d from the edge of the column, and to
Structural design of isolated column footings 2667

a/d varies between 1.28 and 3.37. The concrete compressive

Figure 1 ECP code – critical section.
calculate punching shear at distances 0.5d, d, and 1.5d from the
edge of the column, and at the perimeter of the column. Euro code
(2-2004) stated that the control section of punching shear is at
distance 2.0d from the edge of column and the lowest value of
resistance at different sections controls the design.
5. Data base
strength, fc0, varies from 8.0 MPa to 38.10 MPa, and steel rein-
forcement ratio from 0.4% to 0.91%. The measured failure load
(v) varies from 430 kN to 3037 kN. The yield strength of steel
reinforcement is 552 MPa, and the tensile strength 634 MPa. Tests
TI to TXII after Bonic and Folic [9], while tests DF6 to DF22
after Heggar et al. [7].
It is worth noting that in Table 1, tests denoted DF6 to DF10
were carried out while the footings are supported on sand. Tests
marked TI to TXII were carried out on supporting sand gravel;
25% by weight sand content, maximum nominal size of gravel 50
mm. In tests coded DF11 to DF22 the soil was simulated by small
hydraulic jacks transferred its load via steel beams to two
polytetrafluoroethylene (PTFE) coated sliding bearings. Therefore
scale effects may influence the test results of tests denoted TI to
TXII. It is interesting to note that the ratio of maximum nominal
size of gravel to footing width in tests TI to TXII is 1/170 which is
unrealistic ratio compared by practical applications. While, in
footings DF6 to DF10 where the footings are supported on sand,
the maximum sand size to footing width varies from 1/600 to
1/700 assuming max-imum nominal size of sand size is 2 mm. It
is documented that the tested specimens failed when the induced
punching shear stress in specimens attained the punching shear
strength pro-vided by concrete section.

Nineteen experimental test results on reinforced concrete foot- 6. Procedure of comparison

ings were collected from the available literature, to carry out
comparison between predicated and experimental failure loads.
The data of the tested footing models such as; amount of
reinforcement, dimensions of footings, and failure loads were
tabulated. Table 1 presents the dimensions of the tested footings
L, B and d. The length L equals to the width B. The side
dimension of the footing varies between 850 mm and 1800 mm,
while the depth (d) varies between 100 mm to 470 mm. The
dimensions of column stubs (c) are varying from 175 mm to 200
mm. Th ratio of shear span to footing thickness
The comparison between predicted and measured failure loads
was conducted taking into consideration that the cube strength of
concrete was assessed by dividing cylinder concrete strength fc0
by 0.80, to implement Egyptian and British code provisions. The
material coefficients cs and cc of steel reinforcement and concrete
were taken equal to unity, in the prediction of failure load.
The total failure load Pp was predicted as,
Pp ¼ Pupc1 A0 ð1Þ
A

Therefore, the parameter bp is expressed as,

Table 1 Details of test specimens.
fc0 bp ¼ Pp=V ð2Þ
Test L=B d C a/d q V test
(mm) (mm) (mm) (MPa) (%) (kN) Assuming the contact stress at footing–soil interface is uni-
DF6 1200 395 200 1.27 19 0.87 2836
formly distributed, and by considering one way shear at the edge
DF7 1400 395 200 1.52 20.9 0.87 2569 of the column, failure load of the footing is calculated as,
DF8 1200 250 200 2.00 22.5 0.88 1203 2Sc
DF10 1200 250 200 2.00 38.1 0.91 1638 A
DF11 1200 395 200 1.27 21.4 0.87 2813 Ps ¼ BðB cÞ ð3Þ
DF12 1400 395 200 1.52 21.2 0.88 2208 And for one way shear at distance d/2 from the edge of the
DF13 1800 395 200 2.03 21.1 0.87 1839 column,
DF14 1400 295 200 2.00 21.2 0.88 1478
2Sc
DF15 1400 470 200 1.28 21.7 0.85 2750
DF19 1200 395 200 1.27 21.8 0.87 2790
Ps ¼ BðB c þ dÞ A ð4Þ
DF20 1200 395 200 1.27 35.7 0.87 3037
DF21 1400 395 200 1.52 36.3 0.87 2860 The parameter bs is expressed as,
DF22 1800 395 200 2.03 36.4 0.87 2405
TI 850 175 175 1.92 30.7 0.40 906 bs ¼ Ps=V ð5Þ
TII 850 125 175 2.70 30.7 0.40 1050
TIX 850 100 175 3.37 17 0.40 430 6.1. ECP (203-2011) provisions [10]
TX 850 150 175 2.25 17 0.40 656
TXI 850 125 175 2.70 15.4 0.40 451
TXII 850 125 175 2.70 8 0.40 440
Fig. 2 and Table 2 present comparison between the measured and
the predicted failure loads Pus obtained from one-way
2668 F. Abdrabbo et al.

3.5 DF22, TII and TXII, an average value of the multiple factor
becomes 0.27 instead of 0.16.
3.0

predicted load, kN *10 3

Therefore the proposed adjusted equation for the calcula-tion
2.5 of shear capacity is proposed:
2.0 sffiffiffiffiffic
f
cu
1.5 qc ¼ 0:27 c ð6Þ
1.0
The failure loads (Ps) were recalculated again using Eq. (6)
0.5 and compared with the measured values, Fig. 3. The figure
0.0 indicates better correlation between the measured and calcu-lated
0 0.5 1 1.5 2 2.5 3 3.5 failure loads. The average value of bs of all tests is 1.03, while the
3 average value of bs of test series DF6 to DF22 is 1.1 after
Measured failure load kN*10
omitting the results of two tests and the aver-age value of test
Figure 2 Comparison between predicted and measured failure loads series T1 to TXII is 0.83, after omitting the results of two tests.
(ECP 203-11) one way shear.
Employing ECP code punching shear provisions, the failure
loads were calculated and compared with the measured failure
loads and the ratio bp was drawn, Fig. 4.
Table 2 Predicted and measured failure loads, (ECP 203-11), one
way shear and punching shear. Table 2 indicates that the values of bp obtained from tests DF6
b b to DF22 vary from 0.68 to 0.98, with an average of 0.80; Tests TI
Test Measured Predicted s Predicted p to TXII revealed an average value of 0.47, while values of b
load (kN) failure load Ps failure load Pp ranged from 0.32 to 0.64. The difference in the out-put results of
(kN) (kN)
bp values, between series of tests DF6 to DF22 and the other
DF6 2836 1466.2 0.52 1919.7 0.68 series TI to TXII is due to the concentration of stresses acting on
DF7 2569 1573.0 0.61 1853.1 0.72 footing models TI to TXII under column stub. Fig. 4 and Table 2,
DF8 1203 814.6 0.68 877.5 0.73 indicate that the code provision for
DF10 1638 1060.0 0.65 1141.9 0.70
DF11 2813 1556.0 0.55 2037.4 0.72
DF12 2208 1584.3 0.72 1866.4 0.85 4.0
DF13 1839 1839.2 1.00 1712.8 0.93 3.5
10 3

DF14 1478 1052.5 0.71 1085.9 0.73

kN*

3.0
DF15 2750 2103.1 0.76 2688.9 0.98
DF19 2790 1570.5 0.56 2056.3 0.74 2.5
load,

DF20 3037 2009.8 0.66 2631.4 0.87 2.0

DF21 2860 2073.1 0.72 2442.2 0.85
adjuste

DF22 2405 2292.5 0.95 2249.7 0.94 1.5

d
-

0.5

TI 906 501.2 0.55 577.5 0.64 1.0

TII 1050 325.5 0.31 335.4 0.32
TIX 430 185.4 0.43 179.0 0.42
re

0.0
TX 656 304.5 0.46 332.7 0.51
TXI 451 230.8 0.51 237.8 0.53 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
3
TXII 440 166.2 0.38 171.2 0.39 Measured failure load kN*10
Figure 3 Comparison between re-adjusted and measured failure loads
(ECP 203-11) one way shear.
shear at code defined critical section. The predicted failure loads
were drawn against measured failure loads. Table 2 pre-sents the
ratio bs of the predicted to the measured failure loads. The ratio
bs varies from 0.52 to 1.0 for tests DF6 to DF22 with an average 3.5
value of 0.70. Footings TI to TXII revealed bs value varies from 3.0
predicted load, KN*10 3

0.31 to 0.55 with an average value of 0.44.

The bs –values of tests TI to TXII are less than the values 2.5
obtained from tests DF6 to DF22, This may be explained by the 2.0
concentration of contact stress, at footing – soil interface, beneath
column stub. It is obvious from Fig. 2 that the code provision 1.5
underestimates the failure load obtained from one-way shear at 1.0
code defined critical section, by an average ratio of 0.62.
0.5
The following is a proposal to adjust ECP 203-11 provi-sions 0.0
for one way shear to predict values nearly equal to the measured 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
failure load. The multiple factor in the equation given by code 3
Measured failure load KN*10
provisions for the calculations of shear capacity was calculated
for each test by equating the predicted load to the measured Figure 4 Comparison between predicted and measured failure loads
failure load. By omitting four test results DF13, (ECP 203-11) punching shear.
Structural design of isolated column footings 2669

punching shear underestimates the failure load by an average
value of 0.7. If test results of series TI to TXII are omitted, the
average value of bp becomes 0.80.
The code provisions were adjusted by the same procedure
outline in one-way shear. Table 3 indicates unexpected small
values of bp corresponding to tests TII, TIX and TXII. The test
results of these specimens were omitted and the mean values of
multiple factors in the equations given by code provisions for
calculating the shear capacity were assessed. The achieved mul-
tiple factors and the proposed equations are
sffiffiffiffiffi
f
cu
qc ¼ 0:43 cc
sffiffiffiffiffi
distance d from the edge of the column and compared with the
measured failure loads. The achieved results are presented in
Table 3 and Fig. 6.
Comparison between the predicted and the measured fail-ure
loads indicates that the bs values, vary from 0.92 to 2.04 for tests
coded DF6 to DF22 with an average value of 1.36 and from 0.38
to 0.81 for tests TI to TXII with an average value of 0.57. The
achieved results revealed that the distribu-tion of contact stress at
footing-soil interface has a great influ-ence on the shear strength
induced in the footing. Table 3 indicates that the distribution of
contact stress at footing-soil
ð7Þ
f
qc cu

b ð8Þ
qc ¼ 0:43 0:5 þ
c ¼ þ 2ðða
c
c þ dÞ þ ð bcþdÞÞ
cc
sffiffiffiffiffi
/d
q 1:09 0:2
f
cu ð9Þ

c
c
interface is a major factor in the calculations of
one-way shear induced in the footing.
The failure loads of the tested footings were
predicted in accordance with the punching shear
provisions stated in ACI 318-08 code. Table 3
presents the predicted failure loads, the measured
failure loads and the ratio bp. The correlation
between experimental and the predicted failure
loads was drawn; Fig. 7, for the sake of
comparison. The table indicates

The proposed equations for punching shear calculations were Table 3 Predicted and measured failure loads, (ACI
employed in the prediction of failure loads and compared with the 318-08), one way shear and punching shear.
measured values. Fig. 5 shows a better correlation between Test Measured Predicted bs Predicted bp
predicted and measured failure loads; even though scatter in
load (kN) failure load Ps failure load Pp
results is noted. It is worth noting that the average bp value of (kN) (kN)
tests DF6 to DF22 is 1.09, while the mean value of all test results DF6 2836 4014.2 1.42 1793.1 0.63
is 1.03. DF7 2569 2935.1 1.14 1730.9 0.67
It is worth noting that the minimum predicted failure loads of DF8 1203 1161.2 0.97 819.7 0.68
the tested specimens, in accordance with ECP 203-11 code DF10 1638 1511.0 0.92 1066.6 0.65
provisions are controlled by the loads obtained from one-way DF11 2813 4260.2 1.51 1903.0 0.68
shear at code-defined critical section, at distance d/2 from the DF12 2208 2956.1 1.34 1743.3 0.79
edge of the column. DF13 1839 2467.6 1.34 1599.9 0.87
DF14 1478 1483.9 1.00 1014.3 0.69
6.2. ACI (318-08) provisions [11] DF15 2750 5611.6 2.04 2511.5 0.91
DF19 2790 4299.8 1.54 1920.7 0.69
DF20 3037 5502.4 1.81 2457.9 0.81
The predicted failure loads of the tested footings were obtained DF21 2860 3868.1 1.35 2281.2 0.80
based on one-way shear at code-defined critical section, at DF22 2405 3241.1 1.35 2101.3 0.87
TI 906 732.8 0.81 539.4 0.60
TII 1050 400.3 0.38 313.3 0.30
TIX 430 213.2 0.50 167.2 0.39
TX 656 405.1 0.62 310.8 0.47
TXI 451 283.8 0.63 222.1 0.49 that the value of bp for tests DF6 to DF22 varies from 0.63 to
TXII 440 204.4 0.46 159.9 0.36 0.91, with an average value of 0.75, Tests TI to TXII revealed
values of bp vary from 0.30 to 0.60 with an average value of 0.44.
The difference in the output results between series of test DF6 to
DF22 and the other series TI to TXII is due to the con-centration
of stresses acting on the footing model underneath column stub.
Generally all test results revealed that code pro-visions for
punching shear underestimate the measured failure

4.5
4.0 y = 1.2023x - 240.14

1
0
k

*
3
3.0

3.5

load,
2.5

adjus
2.0

ted
-
0.5

1.5
1.0

re
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
3
Measured failure load kN*10

Figure 5 Comparison between re-adjusted and measured failure loads

(ECP 203-11) punching shear.

6.0
3
5.0
kN*1

4.0
load, 0

3.0
predict
ed

2.0

1.0

0.0
0.0 1.0 2.0 3.0 4.0 5.0 6.0
3
Measured failure load kN*10
Figure 6 Comparison between predicted and measured failure loads
(ACI 318-08) one way shear.
2670

loads by 0.65. It is interesting to note that the Egyptian code ac

provision for the calculation of failure loads based on punch-ing qffiffiffiffi
shear underestimated the failure load by a factor of 0.70. While Vc ¼ 0:25 1 þ 2bc k fc0bod ð10
Vc ¼ 0:12 as d þ 2 k fc0bod ð1
qffiffiffiffi
bo

the Egyptian code provision for the calculation of one-way shear

at distance d/2 from the edge of the column underestimate the
measured failure load by a factor of 0.62.
The adjustment of code provisions were carried out for each
test specimen and the multiple factor in the equation giv-ing the
smallest value of shear strength was obtained. The fac-tors in the
other two equations were adjusted by the same ratio.
Three tests which revealed peculiar multiple values, Tests TII,
TIX, TXII, were omitted and the mean values of multiple factors
were calculated. The adjustments revealed the follow-ing three
equations;
qffiffiffiffi
Vc ¼ 0:48k f c0 bo d ð12
The three proposed equations for the
calculation of punch-ing shear were implemented
in the calculation of failure loads and compared
with the measured values, Fig. 8. The figure
indicates a better correction. The bp value varies
between 0.69 and 1.34 with an average value of
1.13.
It is worth noting that the minimum predicted
failure loads in accordance with ACI 318-08 code
provisions is controlled by punching shear at code-
defined critical section, at distance 0.5d from the
edge of the column. The code provisions under-
estimated the failure load by an average value of
0.75 for test series DF6 to DF22 and by an average
value of 0.44 for series TI to TXI, and by 0.65 for
all the tested specimens. The ulti-mate loads of
isolated column footings are controlled by shear at
distance 0.5d from the edge of the column in
accordance with ECP 203-11 and ACI 318-08
code provisions. But ECP 203-11 code provisions
consider one-way shear, contrary to ACI 318-08
code provisions which considered punching shear.
Both ACI 318-08 code and ECP 203-11 code
provisions under-estimate the failure load of
isolated column footings by an underestimation
ratio of 0.65 and 0.62 respectively.

3.5
3.0
predicted load, kN*10 3

2.5
2.0
1.5

1.0
0.5
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Measured failure load
3
kN*10
Figure 7 Comparison between predicted and
measured failure loads (ACI 318-08) punching shear.
 F. Abdrabbo et al.

5.0
4.0 y = 1.2022x - 240.13
3
kN*1
re-adjustedload, 0

3.0

2.0

1.0

0.0
0.0 1.0 2.0 3.0 4.0 5.0
3
measured failure load kN*10

Figure 8 Comparison between re-adjusted and measured failure loads (ACI 318-08)
punching shear.

6.3. (BS8110-1997) provisions [12]

The failure loads of the tested specimens were predicted by the implementation
of code provisions related to one-way shear at distance d from the edge of the
column. Table 4 presents the predicted and the measured failure loads, bs values
are also presented in the same table. Fig. 9 presents the correlation between
predicted and measured failure loads.
Test series DF6 to DF22 revealed that the bs values varied from 0.94 to 1.98
with an average value of 1.33, while test ser-ies TI to TXII indicated bs values
varied from 0.36 to 0.71 with an average value of 0.56. If all test results are
considered the average values of bs becomes 1.09. Again the effect of stress
distribution at footing–soil interface on the predicted value of the failure load is
obvious. Code provisions overestimate the failure load of test series DF6 to
DF22 by an average ratio

Table 4 Predicted and measured failure loads, (BS 8110.1-1997), one way shear at
distance d and at the edge of the column.

Test Measured Predicted bs Predicted failure bs

load (kN) failure load load Ps (kN) at
Ps (kN) at the edge of the
distance d column
DF6 2836 4028.8 1.42 4435.2 1.56
DF7 2569 2899.4 1.13 5276.2 2.05
DF8 1203 1275.2 1.06 3054.7 2.54
DF10 1638 1537.0 0.94 3600.0 2.20
DF11 2813 4191.8 1.49 4707.0 1.67
DF12 2208 2924.3 1.32 5313.9 2.41
DF13 1839 2433.7 1.32 6572.6 3.57
DF14 1478 1579.0 1.07 3968.6 2.69
DF15 2750 5449.0 1.98 6397.0 2.33
DF19 2790 4217.7 1.51 4750.8 1.70
DF20 3037 4971.4 1.64 5688.0 1.87
DF21 2860 3485.2 1.22 6451.7 2.26
DF22 2405 2918.8 1.21 7998.8 3.33
TI 906 642.4 0.71 1856.5 2.05
TII 1050 381.7 0.36 1326.1 1.26
TIX 430 237.2 0.55 789.5 1.84
TX 656 407.3 0.62 1184.2 1.81
TXI 451 303.5 0.67 940.2 2.08
TXII 440 184.5 0.42 541.6 1.23
Structural design of isolated column footings 2671

6.0
Table 5 Predicted and measured failure loads, (BS 8110.1-1997),

3
kN*1 5.0 punching shear at distance 0.5d from the edge of the column and
at distance d from the edge of the column.
load,0 4.0
Test Measured Predicted bp Predicted bp
3.0 load (kN) failure load Pp failure load Pp
predict

(kN) at (kN) at
ed

2.0 distance 0.5d distance d

1.0 DF6 2836 2781.3 0.98 5463.7 1.93
DF7 2569 2642.5 1.03 3603.0 1.40
0.0 DF8 1203 1391.1 1.16 1409.4 1.17
0.0 1.0 2.0 3.0 4.0 5.0 6.0 DF10 1638 1676.7 1.02 1698.8 1.04
3 DF11 2813 2893.8 1.03 5684.7 2.02
Measured failure load kN*10
DF12 2208 2665.2 1.21 3634.0 1.65
Figure 9 Comparison between predicted and measured failure loads DF13 1839 2438.5 1.33 2590.7 1.41
(BS 8110.1-1997) one way shear at distance d. DF14 1478 1668.1 1.13 1708.8 1.16
DF15 2750 3769.0 1.37 7336.9 2.67
DF19 2790 2911.7 1.04 5719.9 2.05
of 1.33, while code provisions underestimate the failure load of DF20 3037 3432.0 1.13 6742.1 2.22
test series TI to TXII by an average ratio of 0.55, even though the DF21 2860 3176.4 1.11 4330.9 1.51
same provisions are implemented. DF22 2405 2924.6 1.22 3107.2 1.29
TI 906 730.7 0.81 735.8 0.81
The failure loads were predicted using code provisions of one
TII 1050 461.6 0.44 381.7 0.36
way shear at the edge of the column. The achieved pre-dicted TIX 430 287.4 0.67 217.9 0.51
loads along with the measured failure loads are pre-sented in TX 656 482.8 0.74 438.0 0.67
Table 4. Fig. 10 presents correlation between predicted failure TXI 451 367.0 0.81 303.5 0.67
loads and the measured value. TXII 440 223.6 0.51 169.4 0.39
Test series DF6 to DF22 revealed that bs values varied from
1.56 to 3.57 with an average value of 2.32, while test series TI to
TXII indicated that bs values varied from 1.23 to 2.08 with an
4.0
average value of 1.71. From all test results, the code provi-sions
produce failure loads with an average value of 2.13 times the
N

3.5
k

*
1
0
3

average value of the measured failure load. One can notice that 2.5

the bs values obtained from series DF6 to DF22 are rela-tively 3.0
load

close to bs values obtained from series TI to TXII com-pared by

values obtained from shear calculations at distance d from the 2.0
face of column.
predicte

1.5
By comparing Figs. 9 and 10, one can notice that the diver- 1.0
d

gence between measured and predicted failure loads increases as

the control section of one way shear becomes closer to the edge of 0.5
the column. 0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
The predicted failure loads of the tested specimens were
3
obtained implementing punching shear provisions at distance 0.5d Measured failure load kN*10
from the edge of the column. The bp values are presented in Table
5. The correlation between the measured and the pre-dicted Figure 11 Comparison between predicted and measured failure loads
failure loads is presented in Fig. 11. Table 5 indicates (BS 8110.1-1997) punching shear at distance 0.5d from the edge of
the column.

8 that the values of bp obtained from test results DF6 to DF22 is

varying from 0.98 to 1.37, with an average value of 1.13. Tests TI
N
o

1
0
a

*
k
l

4
3

6
to TXII revealed an average value of 0.66, while the values of bp
ranged from 0.44 to 0.81. The test results revealed that the
predicted failure loads based on punching shear at dis-tance 0.5d
predicted

from the edge of the column of tests TI to TXII are less than the
measured values while, the predicted failure loads of tests DF6 to
DF22 are higher than the measured values. The difference in
2
output results in bp values between test group DF6 to DF22 and
the other series TI to TXII is due to the con-centration of stresses
0 acting on footing under column stub. If test results of footings
0 1 2 3 4 5 6 7 8 DF6 to DF22 are only considered one may conclude that, the BS
3 8110.1-1997 code provision for punching shear at 0.5d from the
Measured failure load kN*10
edge of the column overesti-mate the failure load by an average
Figure 10 Comparison between predicted and measured failure loads value of 1.13. While ACI 318-08 and ECP 203-11 code provisions
(BS 8110.1-1997) one way shear at the edge of the column. underestimate the
2672 F. Abdrabbo et al.

failure load of the same footings by an average value of 0.75 and
0.80 respectively.
The adjustment of code provisions was carried out, for each
test specimen, by assessing the multiple factor in the code shear
equation in order to equalize the predicted and the measured
failure loads. Four peculiar test results TII, TXI, TX and TXII
were omitted from the calculation of the mean of the multiple
factor in shear equation. Therefore the proposed equation for
calculating punching shear capacity of isolated column footing is,
1 3
vc ¼ 0:74f100As=ðBdÞg = ð400=dÞ = =cm ð13Þ
Fig. 12 indicates a better correlation between the measured
and the calculated failure loads, implementing the proposed
equation, the bp value varies from 0.75 to 1.28 with an average
value of 1.02.
The predicted failure loads of the tested footing models were
calculated by implementing code provisions of the punch-ing
shear at distance d from the edge of the column and the ratio bp of
the predicted to measured loads were presented in Table 5. The
collected experimental and the predicted failure loads were
drawn, Fig. 13, for the sake of comparison.
Table 5 indicates that the value of bp for tests DF6 to DF22
varies from 1.04 to 2.67, with an average value of 1.65. Tests
4.0
3.5
3
TI to TXII revealed an average value of 0.57, while the values of
bp ranged from 0.36 to 0.81.
The difference in output results in bp values between series of
tests DF6 to DF22 and the other series TI to TXII is due to the
concentration of contact stresses acting on footing under column
stub. Generally, test results revealed that the code pro-visions for
punching shear at distance d from the edge of the column
overestimate the failure load by an average ratio equal to 1.65, if
the results of tests series TI to TXII are omitted. If all test results
are considered the over predicted ratio is 1.30.
The code provisions were adjusted to achieve the equality of
1 4
the measured and the predicted failure loads. The multiple factor
in the code equation was adapted to achieve that requirement. The
factor for each test was calculated. Test results TII to TXII were
omitted and the mean value of the multiple factors was obtained
as 0.55, therefore the equation may be written as;
1 3 1 4
vc ¼ 0:55f100As=ðBdÞg = ð400=dÞ = =cm ð14Þ
The above equation was implemented in the calculation of the
predicted failure loads of test series DF6 to TI and com-pared
with the measured values, Fig. 14. A better agreement was
achieved. The value of bp varies from 0.56 to 1.85 with an
average value of 1.10.
The failure loads of the tested footings were predicted in
accordance with punching shear provisions stated in BS 8110.1-
1997 code at distance 1.5d from the edge of the column, Table 6.
The values of the measured failure loads, the pre-dicted loads and
the ratio bp are presented. The collected experimental failure

3.0
kN*1

loads and the predicted failure loads were drawn, Fig. 15, for the
2.5 sake of comparison.
re- adjusted load, 0

2.0 The table indicates that the value of bp for tests DF6 to DF22
1.5 varies from 2.06 to 3.6, with an average value of 2.45. Tests TI to
TXII revealed values of bp varies from 0.38 to 1.39 with an
1.0
average value of 0.72. The difference in the output results
0.5 between series of test DF6 to DF22 and the other series TI to
0.0 TXII is due to the concentration of contact stresses act-ing on the
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 footings underneath column stub. Generally, test results revealed
3 that code provisions for punching shear at dis-tance 1.5d
Mesured failure load kN*10
overestimated the measured failure load, by an aver-age ratio
Figure 12 Comparison between re-adjusted and measured fail-ure equal to 1.71.
loads (BS 8110.1-1997) punching shear at distance 0.5d from the The failure loads were predicted by the implementation of
edge of the column. code provisions of punching shear at the perimeter of the col-

7.0
6.0
re-adjusted load, kN*103

6.0
predicted load kN*103

5.0

5.0

4.0

4.0

3.0
3.0

2.0 2.0

1.0

1.0
0.0
0.0
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0
3 3
Measured failure load kN*10 Measured failure load kN*10

Figure 13 Comparison between predicted and measured failure loads Figure 14 Comparison between re-adjusted and measured fail-ure
(BS 8110.1-1997) punching shear at distance d from the edge of the loads (BS 8110.1-1997) punching shear at distance d from the edge of
column. the column.
Structural design of isolated column footings 2673

3.5
Table 6 Predicted and measured failure loads, (BS 8110.1-1997),
punching shear at distance 1.5d from the edge of the column and 3.0

3
at the perimeter of the column.

kN*1
2.5
Test Measured Predicted bp Predicted failure bp

predicted load, 0
2.0
load (kN) failure load load Pp (kN) at
Pp (kN) at perimeter of the 1.5
distance 1.5d column
1.0
DF6 2836 NA 1267.2 0.45
DF7 2569 7882.3 3.07 1319.0 0.51 0.5
DF8 1203 2253.8 1.87 872.8 0.73 0.0
DF10 1638 2716.5 1.66 1028.6 0.63 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
DF11 2813 NA 1344.8 0.48 3
DF12 2208 7950.0 3.60 1328.5 0.60 Measured failure load kN*10
DF13 1839 4131.2 2.25 1314.5 0.71
DF14 1478 2670.2 1.81 992.2 0.67
Figure 16 Comparison between predicted and measured failure loads
DF15 2750 NA 1599.3 0.58 (BS 8110.1-1997) punching shear at the perimeter of the column.
DF19 2790 NA 1357.4 0.49
DF20 3037 NA 1625.1 0.54
DF21 2860 9474.8 3.31 1612.9 0.56
DF22 2405 4954.7 2.06 1599.8 0.67 Table 7 presents the average over/underestimation ratio (bp) of
TI 906 1257.1 1.39 633.9 0.70 the tested footing models denoted DF6 to DF22 with contact stress
TII 1050 424.8 0.40 452.8 0.43
exhibit nearly uniform distribution at footing– soil interface and
TIX 430 215.4 0.50 269.6 0.63
for footings coded TI to TXII which exhibit contact stress with
TX 656 575.2 0.88 404.4 0.62
stress concentration beneath column stub. The table revealed that
TXI 451 337.8 0.75 321.1 0.71
TXII 440 167.6 0.38 184.9 0.42
the ratio of the predicted to measured failure loads depends upon
the contact stress distribution at column footing soil interface, and
on the location of punching surface.

Practically the footing fails by shear at a certain critical plane

of failure, where the imposed shear stress is equal to the shear
strength of concrete. But, in theory, the failure plane is unknown,
9.0 so a trial was given to predict the load at pre-sumed different
failure planes, and the failure load was consid-ered the minimum
kN*1
0^3

7.5 predicted value. Therefore, for each test specimen, the predicted
predictedload

6.0 failure loads were obtained assuming that punching shear takes
place at distance 0.5d, d, 1.5d from the edge of the column and
4.5 around column perimeter. Practi-cally punching around column
,

3.0 perimeter sound unrealistic so that the predicted failure loads

corresponding to punching shear around column perimeter were
1.5 pulled out from compar-ison. From Tables 4–6, the minimum
0.0
predicted failure loads of specimens coded DF6, DF7, DF11,
0.0 1.5 3.0 4.5 6.0 7.5 9.0 DF12, DF13, DF15and DF19 to DF21 were obtained from
3 punching shear provisions at distance 0.5d from the edge of the
Measured failure load kN*10 column; while the mini-mum predicted loads of test specimens
Figure 15 Comparison between predicted and measured failure loads DF8, DF10, DF14, DF22, TI and TX were obtained from one-way
(BS 8110.1-1997) punching shear provisions, at distance 1.5d from shear provision at distance d from the edge of the column.
the edge of the column.
The minimum predicted load of test specimens TIX and TXII
were obtained from code provision of punching shear at distance
umn. The achieved results are presented in Table 6 along with the 1.5d from the edge of the column. The minimum predicted loads
measured failure loads. The ratio bp, of the predicted to measured of test specimens TII and TXI were obtained
failure loads are presented in the same table. The values of bp for
test series DF6 to DF22 varies from 0.45 to 0.73, with an average
value of 0.58, while test series TI to TXII revealed values of bp Table 7 Average over/underestimation ratio corresponding to
varies from 0.41 to 0.71, with an average value 0.58. Fig. 16 punching shear section, BS8110-1997.
presents correlation between measured and predicted failure Test group Average over/underestimation ratio (b) due to
loads. The figure indicates an average under-estimation ratio of punching shear provisions at
0.59.
Edge of the column 0.5d d 1.5d
Obviously, the shape of contact stress distribution at foot-ing–
soil interface has no appreciable effects on punching load, when DF6 to DF22 0.58 1.13 1.65 2.45
punching is considered around the perimeter of the column. TI to TXII 0.58 0.66 0.57 0.72
2674 F. Abdrabbo et al.

from code provision for punching shear at distance d from the 6.4. (EC2-2004) provisions [13]
edge of the column. It was observed from the table that one way
shear provision at the edge of the column produce unreal-istic Comparison between measured and predicted failure loads based
high predicted load. Also one way shear provisions at dis-tance d on code provisions of punching shear at code–defined sections at
from the edge of the column reflect predicted loads nearly equal distance 0.5d, d, 1.5d, 2d from the edge of the col-umn and at
to those obtained from punching shear provisions at distance d/2 column perimeter was conducted.
from the edge of the column, tests specimen DF8, DF10, DF13. The predicted failure loads of the tested specimens were cal-
Therefore, 50% of the tested Specimens revealed minimum culated based on punching shear provisions at distance 0.5d from
predicted failure loads based on punching shear at distance 0.5d
the edge of the column. The ratios bp of the predicted to the
from the edge of the column; the shear-span/depth ratio of the
measured failure loads are presented in Table 8. Fig. 18 presents
specimens is less than 2.0; 33.3% of the tested specimen revealed
correlating between measured and predicted failure loads.
minimum predicted fail-ure loads based on one-way shear at
distance d, the shear– span/depth ratio of the footing is equal to or Table 8 indicates that bp values varying from 1.38 to 1.87 for
bigger than 2.0, 11.2% of the tested specimens revealed minimum
test specimens DF6 to DF22, and from 0.54 to 1.08 for tests TI to
pre-dicted failure load based on punching shear at distance 1.5d
TXII. The average values of b of series DF6 to DF22 is 1.59, and
from the edge of the column, the shear span/depth of the foot-ing the average value of series TI to TXII is 0.86. If all test results are
ratio is 2.70. 5.5% of the tested specimens revealed mini-mum considered, the average values of b
predicted failure load based on punching shear at distance d from
the edge of the column, where a/b is bigger than 2.7. Therefore
the location of the critical shear section is related to shear
span/depth ratio. Thus, one can conclude that as the value of shear Table 8 Predicted and measured failure loads, (EC2-2004),
span/depth ratio increases, the loca-tion of critical section of shear punching shear at distance d/2 and at distance d.
failure goes further away from the column edge. Punching shear Test Measured Predicted bp Predicted bp
at distance 0.50d from the edge of the column is anticipated in load (kN) failure load Pp failure load Pp
footing having shear span to depth ratio less than 2.0. One way (kN) at (kN) at
shear at distance d from the face of the column may occur in distance 0.5d distance d
footings having shear span/depth ratio bigger than 2.0 and less DF6 2836 3913.2 1.38 7687.4 2.71
than 2.70. Footing having shear span/depth ratio higher than 2.70 DF7 2569 3718.0 1.45 5069.3 1.97
exhibit punch-ing shear either at distance d or 1.5d from the edge DF8 1203 1932.2 1.61 1957.7 1.63
of the column. DF10 1638 2329.0 1.42 2359.6 1.44
DF11 2813 4071.5 1.45 7998.3 2.84
DF12 2208 3749.9 1.70 5112.9 2.32
DF13 1839 3431.0 1.87 3645.1 1.98
Fig. 17 presents the predicted minimum failure loads of each
DF14 1478 2324.4 1.57 2381.1 1.61
footing and the corresponding measured failure loads. The figure DF15 2750 5135.5 1.87 9997.0 3.64
indicates an average underestimation ratio of 93%. DF19 2790 4096.7 1.47 8047.9 2.88
If, Two test results out of nineteen test results, which repre- DF20 3037 4828.8 1.59 9486.0 3.12
sent 10% of the tested specimen, are omitted due to the very low DF21 2860 4469.2 1.56 6093.6 2.13
b-values, the ratios of the predicted to measured loads become DF22 2405 4114.9 1.71 4371.7 1.82
equal to 1.0. However if test series TI to TXII are omitted, the TI 906 980.1 1.08 986.9 1.09
average b value becomes 1.11. So that it can be considered that TII 1050 569.2 0.54 470.6 0.45
3

the BS 8110.1–1997 code provisions of shear may predict failure TIX 430 335.2 0.78 254.1 0.59
N

1
0
*
k

loads agree with the measured failure loads within a precision of TX 656 623.1 0.95 565.3 0.86
10% (90% to 110%), irrespective of the shape of contact stress at TXI 451 452.6 1.00 374.2 0.83
TXII 440 363.6 0.83 300.6 0.68
footing –soil interface.
p

d
e

a
c
r

,
i

3.5
5.0
3.0
0 3
kN*1

4.0
2.5
predictedloa

2.0 3.0
d,

1.5 2.0
1.0
1.0
0.5
0.0 0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 0.0 1.0 2.0 3.0 4.0 5.0
3 3
Measured failure load kN*10 Measured failure load kN*10
Figure 17 Comparison between minimum predicted and Figure 18 Comparison between predicted and measured failure
measured failure loads (BS 8110.1-1997). loads (EC2-2004), punching shear at distance d/2.
Structural design of isolated column footings 2675

becomes 1.36. But there is doubt about the achieved results of test
series TI to TXII, due to unrealistic simulation of the sup-porting
soil. If these test results are omitted, the average value of Bp
becomes 1.5. On the other hand, BS 8110.1-1997, ACI 318-08
and ECP 203-2011 revealed values of 1.13, 0.75 and 0.8
respectively, for the same test series.
The code provisions may need to be adjusted. However one
should realize that test specimens failed by punching shear, but
the location of failure plane is not defined. This may affect the
procedure of justification of code provisions.
The adjustment was carried out for each test specimen by
adjusting the multiple factor in code punching shear equation, to
fulfill the requirement that the predicted failure load becomes
equal to measured failure load. Four tests results TII, TIX, TX,
TXII were omitted from the calculation of the mean value of the
multiple factor in shear equation, hence the mean value was
obtained as; 0.12 instead of 0.18 of in the code. The proposed
equation was implemented in the pre-diction of failure loads of
the tested specimens.
Fig. 19 indicates a better correlation between the measured
and the predicted failure loads; the mean value of b is 1.02.
The predicted failure loads were assessed from punching shear
provisions at code-defined section at distance d from the edge of
the column and the ratio bp was presented in Table 8. The
collected experimental failure loads and the corresponding
predicted failure loads were drawn, Fig. 20, for the sake of
comparison.
4.0
3.5
N
Table 8 indicates that the value of bp for tests DF6 to DF22
varies from 1.44 to 3.64, with an average value of 2.32. Tests TI
to TXII revealed an average value of 0.75, while the values of bp
ranged from 0.45 to 1.09.
The difference in output results in b values between series of
tests DF6 to DF22 and the other series TI to TXII is due to the
concentration of contact stresses acting on footing under col-umn
stub. Generally test results revealed that the code provi-sions for
punching shear at distance d from the edge of the column are not
realistic for most of footing models in the case study.
The code provisions were adjusted to insure the equality of the
predicted failure load and the measured value. The multi-ply
factor in punching shear equation was adapted to achieve this
requirement. Five tests out of nineteen test results, which
represent 26% of tests, were omitted. These tests are TII, TIX,
TX, TXII, because of the resulted low values of the mul-tiple
factor. The mean value of the multiple factor of all tests except the
omitted tests is 0.09.
The failure loads were re-predicted using the above multiple
factor in the equation of punching shear and compared with the
measured failure loads, Fig. 21. A better agreement was achieved;
the mean value of b is 1.11. It is interesting to note that the five
omitted test results are in series TII to TXII.
The failure loads of the tested specimens were predicted in
accordance with punching shear provisions stated in EC2 code
provisions at distance 1.5d from the edge of the column. Table 9
presents the predicted failure loads, the measured fail-ure loads
and the ratio bp. The collected experimental failure loads and the
predicted failure loads were drawn; Fig. 22, for the sake of
comparison.

310
*
k

3.0 The table indicates that the value of bp for tests DF6 to DF22
2.5 varies from 2.3 to 4.66, with an average value of 3.44, while tests
adjusted load,

2.0 TI to TXII revealed values of bp varies from 0.50 to 1.86 with an

average value of 0.90. The difference in the out-put results
1.5
0.5 between group of test DF6 to DF22 and the other group TI to
-

1.0 TXII is due to the concentration of contact stres-ses acting on

footing model underneath column stub.
0.0 Generally test results revealed that the code provisions for
re

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 punching shear at distance 1.5d overestimate the measured fail-
3 ure load, and the average overestimation ratio is 2.38.
mesured failure load kN*10
The failure loads of the tested footing were predicted in
Figure 19 Comparison between re-adjusted and measured fail-ure accordance with punching shear provisions stated in EC2-2004
loads (EC2-2004), punching shear at distance d/2. code at distance 2d from the edge of the column. Table 9

9.0 6.0
7.5 5.0
3

load,kN*10 3
kN*1

6.0 4.0
0 edload,
predict

4.5 3.0
adjusted

3.0 2.0
0.0
re-

1.5 1.0

0.0
0.0 1.5 3.0 4.5 6.0 7.5 9.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0
3 3
Measured failure load kN*10 measured failure load kN*10

Figure 20 Comparison between predicted and measured failure loads Figure 21 Comparison between re-adjusted and measured fail-ure
(EC2-2004), punching shear at distance d. loads (EC2-2004), punching shear at distance d.
2676 F. Abdrabbo et al.

Table 9 Predicted and measured failure loads (EC2-2004) 12.0

punching shear at distances 1.5d and 2d.

0 3
kN
*1
10.0
Test Measured Predicted bp Predicted bp

predicted load,Pss
load (kN) failure load Pp failure load Pp 8.0
(kN) at (kN) at
distance 1.5d distanced 2d 6.0

DF6 2836 NA NA NA 4.0

DF7 2569 11090.2 4.32 NA NA
DF8 1203 3130.5 2.60 NA NA 2.0
DF10 1638 3773.3 2.30 NA NA
0.0
DF11 2813 NA NA NA
0.0 2.0 4.0 6.0 8.0 10.0 12.0
DF12 2208 11185.6 5.07 NA NA
3
DF13 1839 5812.6 3.16 10342.9 5.62 Measured failure load kN*10
DF14 1478 3720.7 2.52 4996.9 3.38
DF15 2750 NA NA NA Figure 23 Comparison between predicted and measured failure loads
DF19 2790 NA NA NA (EC2-2004) punching shear at distance 2d.
DF20 3037 NA NA NA
DF21 2860 13330.9 4.66 NA NA
DF22 2405 6971.2 2.90 12404.6 5.16 Table 10 Average over/underestimation ratio corresponding to
TI 906 1686.2 1.86 NA NA punching shear section, (EC2-2004).
TII 1050 523.9 0.50 758.8 0.72
TIX 430 251.2 0.58 289.2 0.67 Test group Over/under prediction ratio (b) due to punching
TX 656 742.4 1.13 1880.1 2.87 shear at
TXI 451 416.5 0.92 603.4 1.34 Edge of the column 0.5d d 1.5d 2d
TXII 440 334.6 0.76 484.7 1.10
DF6 to DF 0.91 1.59 2.32 3.44
TI to TXI 0.8 0.85 0.75 0.90 2.70

14.0
0.48 to 1.14 with mean value of 0.8. All test results revealed a
predicted load,1.5dkN*10 3

12.0 mean value of 0.88. Tests results revealed that the code provi-
10.0 sions for punching shear at the perimeter of the column are
underestimating the measured failure loads. The collected fail-ure
8.0 loads and the corresponding predicted values were drawn; Fig.
6.0 24, for the sake of comparison. BS (8110) code provisions under
estimate the failure load based on calculations of punch-ing shear
4.0 at the perimeter of the column by a factor of 0.88.
2.0 The predicted minimum failure loads for each test specimen
0.0 out of the implementation of punching shear at column perimeter,
0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 at distances d/2, d and 1.5d in accordance with EC2 provisions
3 were considered. The minimum predicted fail-ure load of
Measured failure load kN*10 specimens coded DF6 to DF22, TX and TXII was obtained from
Figure 22 Comparison between predicted and measured failure loads punching shear provisions at the column perimeter. The minimum
(EC2-2004), punching shear at distance 1.5d. predicted failure loads of test specimen TIX were obtained from
punching shear provision at distance 1.5d from the edge of the
column. The minimum
presents the predicted failure load, the measured failure loads and
the ratio bp. The collected experimental failure loads and the
3.5
predicted values were drawn; Fig. 23.
Table 9 indicates that the EC2-2004 code provisions of 3.0
3
kN*1

punching shear at distance 2d from the edge of the column are not 2.5
applicable (NA) for most of the footing models in test series DF6
predictedload, 0

to DF22. So that it is not easy to achieve a firm conclusion from 2.0

the study. Even though 42% of the tested specimens on which the 1.5
code provisions are implemented, revealed that the code
provisions for punching shear at dis-tance 2d overestimate the 1.0
ultimate loads of the footings. The average overestimation ratio 0.5
bp is 2.6. 0.0
Table 10 presents the predicted failure loads based on 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
punching shear provisions at the perimeter of the columns. The 3
Measured failure load kN*10
measured failure loads and the ratio bp. The value of bp for tests
DF6 to DF22 varies from 0.6 to 1.24, with an average value of Figure 24 Comparison between predicted and measured failure loads
0.91, tests TI to TXII revealed values of bp varies from (EC2-2004) punching shear at the perimeter of the column.
Structural design of isolated column footings 2677

3.5
Table 11 Predicted and measured failure loads (EC2-2004)
punching shear at the perimeter of the column. 3.0

predicted load, kN*10 3

Test Measured Predicted failure bp 2.5
load (kN) load Pp (kN) at
perimeter of column 2.0

DF6 2836 1711.9 0.60 1.5

DF7 2569 1853.5 0.72
1.0
DF8 1203 1263.6 1.05
DF10 1638 1993.0 1.22 0.5
DF11 2813 1908.1 0.68
0.0
DF12 2208 1877.7 0.85
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
DF13 1839 1854.3 1.01
3
DF14 1478 1402.3 0.95 Measured failure load kN*10
DF15 2750 2281.9 0.83
DF19 2790 1940.3 0.70 Figure 25 Comparison between minimum predicted ultimate loads
DF20 3037 2984.0 0.98 and measured failure loads (EC2-2004), punching shear at the
DF21 2860 3002.9 1.05 perimeter of the column.
DF22 2405 2985.1 1.24
TI 906 1033.4 1.14
TII 1050 738.1 0.70 7. Conclusions
TIX 430 347.5 0.81
TX 656 521.2 0.79
TXI 451 396.9 0.88 The main target of the work presented is to explore the preci-sion
TXII 440 212.3 0.48 of code provisions related to the structural design of iso-lated
column footings, and to illustrate their relations from test results.
To achieve this target, comparisons between predicted failure
loads of column footings based on code provisions and the
predicted failure loads of test specimen TII and TXI were laboratory measured failure loads were conducted. The study
obtained from code provisions for punching shear at distance d revealed the following main conclusions.
from the edge of the column. The minimum predicted failure load The predicted failure loads of isolated column footings sub-
of test specimen TI was obtained from punching shear provision jected to uniformly distributed contact stress in accordance with
at distance 0.5d from the edge of the column. ACI318-08 code provisions are controlled by punching shear at
Table 11 indicates that punching shear at column perimeter of code-defined critical section, contrary to ECP (203-2011) code
the tested specimen, prevails predicted failure loads close to the provisions which the predicted failure loads are controlled by one-
measured values and the divergence between predicted and way shear at code defined critical section. The critical section
measured failure loads increases as the code defined critical provided by the two codes is at distance 0.5d from the edge of the
section goes further from the edge of the column. column. But ACI 318-08 code provi-sions considered punching
Theoretical calculations in accordance with code provisions shear while ECP 203-11 code provi-sions considered one-way
indicated that the minimum predicted failure loads of 79% of the shear.
tested specimens were achieved in accordance with code Both ACI 318-08 and ECP 203-11 provisions underestimate
provisions of punching shear at the perimeter of the column, 5.2% the failure loads of isolated column footings subjected to uni-
of the tested specimen indicates that the minimum pre-dicted formly distributed contact stress. The underestimation ratio is 0.75
failure load was achieved by code provisions of punch-ing shear and 0.70 respectively. If the supporting material of the footing is
at distance 0.5d, while 10.4% of the tested specimen revealed producing bell shaped contact stress distribution at footing-soil
minimum predicted failure loads in accordance with code interface with maximum value beneath column stub, the
provisions of punching shear at distance d, 5.2% of the tested underestimation ratio becomes 0.44 in both code provisions.
specimens indicated a minimum predicted failure load by code
provisions of punching at distance 1.5d. Observations of the It is not advisable to consider punching shear at the perime-ter
effect of a/d ratio on the location of punching shear were carried of the column in accordance with BS 8110.1-1997. If section of
out. shear failure is considered away from the edge of the col-umn, at
These observations indicated that punching shear in foot-ings 0.5d, d and 1.5d for punching shear and at d for one-way shear,
may occur at column perimeter or at distance 0.5d as long as the the minimum predicting failure loads of foot-ings subjected to
shear span/depth ratio is less than 2.25. Punching shear at distance uniformly distributed contact stress in accor-dance with BS 8110-
d may occur when a/d ratio is bigger than 2.25 and less than or 1997 shall equal 1.13 times the measured failure loads. The ratio
equal to 2.7. For greater ratio of a/d punching shear at distance is 0.56 in case of footings supported on geomaterial producing
1.5d is anticipated. belled shape contact stress distribution.
The predicted minimum failure load of each footing was
assessed out of calculations failure load assuming punching takes Regarding BS 8110-1997 provisions, punching shear at dis-
place at perimeter of column, at 0.5d, 1.5d, and 2d from the edge tance 0.5d from the edge of the column is anticipated in footing
of the column. Fig. 25 presents correlation between measured and having shear span/depth ratio less than 2.0. One way shear at
minimum predicted failure loads. The figure indicates an average distance d from the edge of the column is anticipated in footing
underestimation ratio of 0.85, compared to 0.93 obtained by the having shear span to depth ratio bigger than 2 and less than 2.70.
implementation of BS (8110-1997). Footings having shear span/depth ratio bigger than
2678
2.70 may exhibit punching shear either at distance d or 1.5d from
the edge of the column.
EC2-2004 provisions underestimate the ultimate load of
isolated column footings; the average underestimation is 0.91 in
case of footings supported on geometrical producing uni-formly
distributed contact stress and 0.8 for footings on geo-metrical
producing belled shape contact stress.
In accordance with (EC2-2004) provisions, punching shear in
footings at column perimeter or at distance 0.5d from the edge of
the column may occur, if shear span/depth ratio of footings is less
than 2.25. For shear span/depth ratio bigger than 2.25 and less
than or equal to 2.70, punching shear is anticipated to be at
distance d from the edge of the column. For bigger ratio of shear
span/depth ratio, punching shear may be occurred at distance
1.5d.
Shear span to depth ratio of footing and distribution of contact
stress at footing soil interface are key factors, among others, in
the structural design of isolated column footings, even though
these two factors are not addressed in various code provisions.
References
[1] M.A. Staller, Analysis studies and numerical analysis of punching
shear failure in reinforced concrete slabs, INF. Workshop on
punching shear capacity of R.C slabs, Stockholm, 2000, p. 8.
[2] M. Hallgren, M. Bjerke, Nonlinear finite element analyses of
punching shear failure of column footings, Cement Concr. Compos.
24 (2002) 491–496.
F. Abdrabbo et al.
[3] J. Hegger, A.G. Sherif, M. Ricker, Experimental investigations on
punching behavior of reinforced concrete footings, ACI Struct. J.
(July–August) (2006) 604–613.
[4] Fib model code 2010, first complete draft, Published by the
International Federation for Structural Concrete, Switzerland, 2001.
[5] A. Muttoni, Punching shear strength of reinforced concrete slabs
without transverse reinforcement, ACI Struct. J. 105 (4) (2008) 440–
450.
[6] A. Muttoni, Application of the critical shear crack theory to punching
of R/C slabs with transverse reinforcement, ACI Struct. J. 106 (4)
(2008) 485–494.
[7] J. Hegger, M. Ricker, G. Sherif, Punching strength of reinforced
concrete footings, ACI Struct. J. (2009) 706–716.
[8] L. Tassinari, Punching of concrete slabs with shear reinforcement,
EPFL ENAC IIC, IBETON2010, <http://
ibeton.epfl.ch/resherche/arrmpoinconnement/default_e.asp>
(pristupljeno januara 2013).
[9] Z. Bonic, R. Folic, Punching of column footings, comparison of
experimental and calculation results, Gradevinar 65 (10) (2013) 887–
899.
[10] Egyptian code of practice for design and construction of reinforced
concrete structures, code no. ECP 203-2011, HBNRC, Cairo.
[11] ACI Committee 318, Building Code Requirement for Structural
Concrete (ACI 318M-08) and Commentary (318R-08), American
Concrete Institute, Farmington Hills, MI, 2008.
[12] BS 8110-1:1997, Code of practice for design and construction.
[13] European Committee for Standardization (CEN), Euro code 2:
Design of Concrete Structures – Part 1.1: General Rules and Rules
for Buildings, Brussels, Belgium, 2004.
[14] Hamed S. Askar, Repair of RC plat plates failing in punching shear
by vertical studs, Alex. Eng. J. (2015) 541–550.