Alexandria Engineering Journal (2016) xxx, xxx–xxx

H O S T E D 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

KEYWORDS Abstract Superstructure loads are transmitted to the underlying soil strata through a suitably
Footing; designed foundation. Therefore, the foundation of a structure is considered the most crucial struc-
Codes; tural element in a building. The foundation may be classified into two main categories, shallow and
Punching shear; deep foundations. Shallow foundation comprises isolated column footings, combined footings and
Shear span; reinforced concrete mat. The design of isolated column footing is accomplished through the appli-
Correlation; cation of geotechnical and structural analysis concepts. So that, the input research into isolated col-
Contact stress umn footings comes from two 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.
Ó 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 Ó 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/).
Please cite this article in press as: F. Abdrabbo et al., Structural design of isolated column footings, Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.
aej.2016.06.016
2 F. Abdrabbo et al.
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
f0c cylinder compressive strength of concrete
fcu 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
Please cite this article in press as: F. Abdrabbo et al., Structural design of isolated column footings, Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.
aej.2016.06.016
cs material coefficient of steel
cc material coefficient of concrete
Pp predicted failure load due to punching shear
A contact area of footing
Ao 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
Pupc 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 3

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

A0
Pp ¼ Pupc 1 ð1Þ
A
Therefore, the parameter bp is expressed as,
Table 1 Details of test specimens.
bp ¼ Pp =V ð2Þ
Test L=B d C a/d f0c q V test
(mm) (mm) (mm) (MPa) (%) (kN) Assuming the contact stress at footing–soil interface is uni-
formly distributed, and by considering one way shear at the
DF6 1200 395 200 1.27 19 0.87 2836
DF7 1400 395 200 1.52 20.9 0.87 2569
edge of the column, failure load of the footing is calculated as,
DF8 1200 250 200 2.00 22.5 0.88 1203 2Sc
Ps ¼ A ð3Þ
DF10 1200 250 200 2.00 38.1 0.91 1638 BðB  cÞ
DF11 1200 395 200 1.27 21.4 0.87 2813
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 Ps ¼ A ð4Þ
DF19 1200 395 200 1.27 21.8 0.87 2790 BðB  c þ dÞ
DF20 1200 395 200 1.27 35.7 0.87 3037 The parameter bs is expressed as,
DF21 1400 395 200 1.52 36.3 0.87 2860
DF22 1800 395 200 2.03 36.4 0.87 2405
bs ¼ Ps =V ð5Þ
TI 850 175 175 1.92 30.7 0.40 906
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
Fig. 2 and Table 2 present comparison between the measured
TXII 850 125 175 2.70 8 0.40 440
and the predicted failure loads Pus obtained from one-way

Please cite this article in press as: F. Abdrabbo et al., Structural design of isolated column footings, Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.
aej.2016.06.016
4 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 *103

Therefore the proposed adjusted equation for the calcula-
2.5 tion of shear capacity is proposed:
sffiffiffiffiffi
2.0
f
qc ¼ 0:27 cu ð6Þ
1.5 cc
1.0 The failure loads (Ps) were recalculated again using Eq. (6)
0.5 and compared with the measured values, Fig. 3. The figure
indicates better correlation between the measured and calcu-
0.0
0 0.5 1 1.5 2 2.5 3 3.5 lated failure loads. The average value of bs of all tests is
1.03, while the average value of bs of test series DF6 to
Measured failure load kN*103
DF22 is 1.1 after omitting the results of two tests and the aver-
Figure 2 Comparison between predicted and measured failure age value of test series T1 to TXII is 0.83, after omitting the
loads (ECP 203-11) one way shear. results of two tests.
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),
Table 2 indicates that the values of bp obtained from tests
one way shear and punching shear.
DF6 to DF22 vary from 0.68 to 0.98, with an average of
Test Measured Predicted bs Predicted bp 0.80; Tests TI to TXII revealed an average value of 0.47, while
load (kN) failure load Ps failure load Pp values of b ranged from 0.32 to 0.64. The difference in the out-
(kN) (kN)
put results of bp values, between series of tests DF6 to DF22
DF6 2836 1466.2 0.52 1919.7 0.68 and the other series TI to TXII is due to the concentration
DF7 2569 1573.0 0.61 1853.1 0.72 of stresses acting on footing models TI to TXII under column
DF8 1203 814.6 0.68 877.5 0.73 stub. Fig. 4 and Table 2, 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
re-adjusted load, kN*103

DF14 1478 1052.5 0.71 1085.9 0.73

DF15 2750 2103.1 0.76 2688.9 0.98 3.0
DF19 2790 1570.5 0.56 2056.3 0.74 2.5
DF20 3037 2009.8 0.66 2631.4 0.87 2.0
DF21 2860 2073.1 0.72 2442.2 0.85
DF22 2405 2292.5 0.95 2249.7 0.94 1.5
TI 906 501.2 0.55 577.5 0.64 1.0
TII 1050 325.5 0.31 335.4 0.32
0.5
TIX 430 185.4 0.43 179.0 0.42
TX 656 304.5 0.46 332.7 0.51 0.0
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
TXII 440 166.2 0.38 171.2 0.39 Measured failure load kN*103

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 3.5
an average value of 0.70. Footings TI to TXII revealed bs value
3.0
predicted load, KN*103

varies from 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
2.0
the concentration of contact stress, at footing – soil interface,
beneath column stub. It is obvious from Fig. 2 that the code 1.5
provision underestimates the failure load obtained from one- 1.0
way shear at code defined critical section, by an average ratio
of 0.62. 0.5
The following is a proposal to adjust ECP 203-11 provi- 0.0
sions for one way shear to predict values nearly equal to the 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
measured failure load. The multiple factor in the equation Measured failure load KN*103
given by code provisions for the calculations of shear capacity
was calculated for each test by equating the predicted load to Figure 4 Comparison between predicted and measured failure
the measured failure load. By omitting four test results DF13, loads (ECP 203-11) punching shear.

Please cite this article in press as: F. Abdrabbo et al., Structural design of isolated column footings, Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.
aej.2016.06.016
Structural design of isolated column footings 5

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

sffiffiffiffiffi The failure loads of the tested footings were predicted in
qc fcu
qc ¼ 0:43 0:5 þ ð8Þ accordance with the punching shear provisions stated in ACI
bc cc 318-08 code. Table 3 presents the predicted failure loads, the

s ffiffiffiffiffi
/d fcu measured failure loads and the ratio bp. The correlation
qc ¼ 1:09 0:2 þ ð9Þ between experimental and the predicted failure loads was
2ððac þ dÞ þ ðbcþ dÞÞ cc
drawn; Fig. 7, for the sake of comparison. The table indicates
The proposed equations for punching shear calculations that the value of bp for tests DF6 to DF22 varies from 0.63 to
were employed in the prediction of failure loads and compared 0.91, with an average value of 0.75, Tests TI to TXII revealed
with the measured values. Fig. 5 shows a better correlation values of bp vary from 0.30 to 0.60 with an average value of
between predicted and measured failure loads; even though 0.44. The difference in the output results between series of test
scatter in results is noted. It is worth noting that the average DF6 to DF22 and the other series TI to TXII is due to the con-
bp value of tests DF6 to DF22 is 1.09, while the mean value centration of stresses acting on the footing model underneath
of all test results is 1.03. column stub. Generally all test results revealed that code pro-
It is worth noting that the minimum predicted failure loads visions for punching shear underestimate the measured failure
of the tested specimens, in accordance with ECP 203-11 code
provisions are controlled by the loads obtained from one- 4.5
way shear at code-defined critical section, at distance d/2 from 4.0
re- adjusted load, kN*103

the edge of the column.

3.5
y = 1.2023x - 240.14
3.0
6.2. ACI (318-08) provisions [11]
2.5
2.0
The predicted failure loads of the tested footings were obtained
1.5
based on one-way shear at code-defined critical section, at
1.0
0.5
Table 3 Predicted and measured failure loads, (ACI 318-08), 0.0
one way shear and punching shear. 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
Measured failure load kN*103
Test Measured Predicted bs Predicted bp
load (kN) failure load Ps failure load Pp Figure 5 Comparison between re-adjusted and measured failure
(kN) (kN)
loads (ECP 203-11) punching shear.
DF6 2836 4014.2 1.42 1793.1 0.63
DF7 2569 2935.1 1.14 1730.9 0.67
DF8 1203 1161.2 0.97 819.7 0.68
6.0
DF10 1638 1511.0 0.92 1066.6 0.65
DF11 2813 4260.2 1.51 1903.0 0.68
predicted load, kN*103

5.0
DF12 2208 2956.1 1.34 1743.3 0.79
DF13 1839 2467.6 1.34 1599.9 0.87 4.0
DF14 1478 1483.9 1.00 1014.3 0.69
DF15 2750 5611.6 2.04 2511.5 0.91 3.0
DF19 2790 4299.8 1.54 1920.7 0.69
DF20 3037 5502.4 1.81 2457.9 0.81 2.0
DF21 2860 3868.1 1.35 2281.2 0.80
DF22 2405 3241.1 1.35 2101.3 0.87 1.0
TI 906 732.8 0.81 539.4 0.60
TII 1050 400.3 0.38 313.3 0.30 0.0
TIX 430 213.2 0.50 167.2 0.39 0.0 1.0 2.0 3.0 4.0 5.0 6.0
TX 656 405.1 0.62 310.8 0.47 Measured failure load kN*103
TXI 451 283.8 0.63 222.1 0.49
TXII 440 204.4 0.46 159.9 0.36 Figure 6 Comparison between predicted and measured failure
loads (ACI 318-08) one way shear.

Please cite this article in press as: F. Abdrabbo et al., Structural design of isolated column footings, Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.
aej.2016.06.016
6 F. Abdrabbo et al.

loads by 0.65. It is interesting to note that the Egyptian code 5.0

provision for the calculation of failure loads based on punch-
y = 1.2022x - 240.13

re-adjusted load, kN*103

ing shear underestimated the failure load by a factor of 0.70. 4.0
While the Egyptian code provision for the calculation of
one-way shear at distance d/2 from the edge of the column 3.0
underestimate the measured failure load by a factor of 0.62.
The adjustment of code provisions were carried out for 2.0
each test specimen and the multiple factor in the equation giv-
ing the smallest value of shear strength was obtained. The fac- 1.0
tors in the other two equations were adjusted by the same
ratio. 0.0
Three tests which revealed peculiar multiple values, Tests 0.0 1.0 2.0 3.0 4.0 5.0
TII, TIX, TXII, were omitted and the mean values of multiple measured failure load kN*103
factors were calculated. The adjustments revealed the follow-
ing three equations; Figure 8 Comparison between re-adjusted and measured failure
loads (ACI 318-08) punching shear.

 qffiffiffiffi
ac
Vc ¼ 0:25 1 þ 2 k fc0 bo d ð10Þ
bc 6.3. (BS8110-1997) provisions [12]

 qffiffiffiffi
as d
Vc ¼ 0:12 þ 2 k fc0 bo d ð11Þ
bo The failure loads of the tested specimens were predicted by the
qffiffiffiffi implementation of code provisions related to one-way shear at
Vc ¼ 0:48k fc0 bo d ð12Þ distance d from the edge of the column. Table 4 presents the
predicted and the measured failure loads, bs values are also
The three proposed equations for the calculation of punch- presented in the same table. Fig. 9 presents the correlation
ing shear were implemented in the calculation of failure loads between predicted and measured failure loads.
and compared with the measured values, Fig. 8. The figure Test series DF6 to DF22 revealed that the bs values varied
indicates a better correction. The bp value varies between from 0.94 to 1.98 with an average value of 1.33, while test ser-
0.69 and 1.34 with an average value of 1.13. ies TI to TXII indicated bs values varied from 0.36 to 0.71 with
It is worth noting that the minimum predicted failure loads an average value of 0.56. If all test results are considered the
in accordance with ACI 318-08 code provisions is controlled average values of bs becomes 1.09. Again the effect of stress
by punching shear at code-defined critical section, at distance distribution at footing–soil interface on the predicted value
0.5d from the edge of the column. The code provisions under- of the failure load is obvious. Code provisions overestimate
estimated the failure load by an average value of 0.75 for test the failure load of test series DF6 to DF22 by an average ratio
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 Table 4 Predicted and measured failure loads, (BS 8110.1-
with ECP 203-11 and ACI 318-08 code provisions. But ECP 1997), one way shear at distance d and at the edge of the
203-11 code provisions consider one-way shear, contrary to column.
ACI 318-08 code provisions which considered punching shear. Test Measured Predicted bs Predicted failure bs
Both ACI 318-08 code and ECP 203-11 code provisions under- load (kN) failure load load Ps (kN) at
estimate the failure load of isolated column footings by an Ps (kN) at the edge of the
underestimation ratio of 0.65 and 0.62 respectively. 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
3.5 DF10 1638 1537.0 0.94 3600.0 2.20
3.0 DF11 2813 4191.8 1.49 4707.0 1.67
predicted load, kN*103

DF12 2208 2924.3 1.32 5313.9 2.41

2.5 DF13 1839 2433.7 1.32 6572.6 3.57
DF14 1478 1579.0 1.07 3968.6 2.69
2.0
DF15 2750 5449.0 1.98 6397.0 2.33
1.5 DF19 2790 4217.7 1.51 4750.8 1.70
DF20 3037 4971.4 1.64 5688.0 1.87
1.0
DF21 2860 3485.2 1.22 6451.7 2.26
0.5 DF22 2405 2918.8 1.21 7998.8 3.33
TI 906 642.4 0.71 1856.5 2.05
0.0 TII 1050 381.7 0.36 1326.1 1.26
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
TIX 430 237.2 0.55 789.5 1.84
Measured failure load kN*103 TX 656 407.3 0.62 1184.2 1.81
TXI 451 303.5 0.67 940.2 2.08
Figure 7 Comparison between predicted and measured failure TXII 440 184.5 0.42 541.6 1.23
loads (ACI 318-08) punching shear.

Please cite this article in press as: F. Abdrabbo et al., Structural design of isolated column footings, Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.
aej.2016.06.016
Structural design of isolated column footings 7

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

predicted load, kN*103

5.0 1997), punching shear at distance 0.5d from the edge of the
column and at distance d from the edge of the column.
4.0
Test Measured Predicted bp Predicted bp
3.0 load (kN) failure load Pp failure load Pp
(kN) at (kN) at
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
Measured failure load kN*103 DF11 2813 2893.8 1.03 5684.7 2.02
DF12 2208 2665.2 1.21 3634.0 1.65
Figure 9 Comparison between predicted and measured failure DF13 1839 2438.5 1.33 2590.7 1.41
loads (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 DF21 2860 3176.4 1.11 4330.9 1.51
the same provisions are implemented. DF22 2405 2924.6 1.22 3107.2 1.29
The failure loads were predicted using code provisions of TI 906 730.7 0.81 735.8 0.81
TII 1050 461.6 0.44 381.7 0.36
one way shear at the edge of the column. The achieved pre-
TIX 430 287.4 0.67 217.9 0.51
dicted loads along with the measured failure loads are pre- TX 656 482.8 0.74 438.0 0.67
sented in Table 4. Fig. 10 presents correlation between TXI 451 367.0 0.81 303.5 0.67
predicted failure 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 average value of 1.71. From all test results, the code provi- 4.0
sions produce failure loads with an average value of 2.13 times 3.5
predicted load kN*103

the average value of the measured failure load. One can notice 3.0
that the bs values obtained from series DF6 to DF22 are rela-
2.5
tively close to bs values obtained from series TI to TXII com-
pared by values obtained from shear calculations at distance d 2.0
from the face of column. 1.5
By comparing Figs. 9 and 10, one can notice that the diver- 1.0
gence between measured and predicted failure loads increases
0.5
as the control section of one way shear becomes closer to the
edge of 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
obtained implementing punching shear provisions at distance Measured failure load kN*103
0.5d from the edge of the column. The bp values are presented
Figure 11 Comparison between predicted and measured failure
in Table 5. The correlation between the measured and the pre-
loads (BS 8110.1-1997) punching shear at distance 0.5d from the
dicted failure loads is presented in Fig. 11. Table 5 indicates
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.
predicted load kN*103

Tests TI to TXII revealed an average value of 0.66, while the

6 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-
4 tance 0.5d 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
2 difference in output results in bp values between test group
DF6 to DF22 and the other series TI to TXII is due to the con-
0 centration of stresses acting on footing under column stub. If
0 1 2 3 4 5 6 7 8 test results of footings DF6 to DF22 are only considered one
Measured failure load kN*103 may conclude that, the BS 8110.1-1997 code provision for
punching shear at 0.5d from the edge of the column overesti-
Figure 10 Comparison between predicted and measured failure mate the failure load by an average value of 1.13. While ACI
loads (BS 8110.1-1997) one way shear at the edge of the column. 318-08 and ECP 203-11 code provisions underestimate the

Please cite this article in press as: F. Abdrabbo et al., Structural design of isolated column footings, Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.
aej.2016.06.016
8 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,
vc ¼ 0:74f100As =ðBdÞg1=3 ð400=dÞ1=4 =cm
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
re-adjusted load, kN*103
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
ð13Þ of 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;
vc ¼ 0:55f100As =ðBdÞg1=3 ð400=dÞ1=4 =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

3.0 experimental failure loads and the predicted failure loads were
2.5 drawn, Fig. 15, for the sake of comparison.
2.0 The table indicates that the value of bp for tests DF6 to
DF22 varies from 2.06 to 3.6, with an average value of 2.45.
1.5
Tests TI to TXII revealed values of bp varies from 0.38 to
1.0 1.39 with an average value of 0.72. The difference in the output
0.5 results between series of test DF6 to DF22 and the other series
0.0 TI to TXII is due to the concentration of contact stresses act-
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 ing on the footings underneath column stub. Generally, test
Mesured failure load kN*103 results revealed that code provisions for punching shear at dis-
tance 1.5d overestimated the measured failure load, by an aver-
Figure 12 Comparison between re-adjusted and measured fail- age ratio equal to 1.71.
ure loads (BS 8110.1-1997) punching shear at distance 0.5d from The failure loads were predicted by the implementation of
the edge of the column. code provisions of punching shear at the perimeter of the col-

6.0
7.0
re-adjusted load, kN*103

6.0 5.0
predicted load kN*103

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
Measured failure load kN*103 Measured failure load kN*103

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

Please cite this article in press as: F. Abdrabbo et al., Structural design of isolated column footings, Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.
aej.2016.06.016
Structural design of isolated column footings 9

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

predicted load, kN*103

column and at the perimeter of the column. 2.5
Test Measured Predicted bp Predicted failure bp
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 Measured failure load kN*103
DF12 2208 7950.0 3.60 1328.5 0.60
DF13 1839 4131.2 2.25 1314.5 0.71 Figure 16 Comparison between predicted and measured failure
DF14 1478 2670.2 1.81 992.2 0.67 loads (BS 8110.1-1997) punching shear at the perimeter of the
DF15 2750 NA 1599.3 0.58
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
Table 7 presents the average over/underestimation ratio
DF22 2405 4954.7 2.06 1599.8 0.67
(bp) of the tested footing models denoted DF6 to DF22 with
TI 906 1257.1 1.39 633.9 0.70
TII 1050 424.8 0.40 452.8 0.43 contact stress exhibit nearly uniform distribution at footing–
TIX 430 215.4 0.50 269.6 0.63 soil interface and for footings coded TI to TXII which exhibit
TX 656 575.2 0.88 404.4 0.62 contact stress with stress concentration beneath column stub.
TXI 451 337.8 0.75 321.1 0.71 The table revealed that the ratio of the predicted to measured
TXII 440 167.6 0.38 184.9 0.42 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
9.0 is unknown, so a trial was given to predict the load at pre-
sumed different failure planes, and the failure load was consid-
predictedload, kN*10^3

7.5
6.0
4.5
3.0
1.5
0.0
0.0 1.5 3.0 4.5
Measured failure load kN*103
Figure 15 Comparison between predicted and measured failure
loads (BS 8110.1-1997) punching shear provisions, at distance 1.5d
from the edge of the column.
umn. The achieved results are presented in Table 6 along with
the measured failure loads. The ratio bp, of the predicted to
measured 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 varies from 0.41 to 0.71, with an average
value 0.58. Fig. 16 presents correlation between measured and
predicted failure loads. The figure indicates an average under-
estimation ratio of 0.59.
Obviously, the shape of contact stress distribution at foot-
ing–soil interface has no appreciable effects on punching load,
when punching is considered around the perimeter of the
column.
ered the minimum predicted value. Therefore, for each test
specimen, the predicted failure loads were obtained assuming
that punching shear takes place at distance 0.5d, d, 1.5d from
the edge of the column and around column perimeter. Practi-
cally punching around column perimeter sound unrealistic so
that the predicted failure loads corresponding to punching
shear around column perimeter were pulled out from compar-
ison. From Tables 4–6, the minimum predicted failure loads of
specimens coded DF6, DF7, DF11, DF12, DF13, DF15and
6.0 7.5 9.0
DF19 to DF21 were obtained from punching shear provisions
at distance 0.5d from the edge of the column; while the mini-
mum predicted loads of test specimens DF8, DF10, DF14,
DF22, TI and TX were obtained from one-way shear provision
at distance d from 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 1.5d from the edge of the column. The minimum
predicted loads of test specimens TII and TXI were obtained
Table 7 Average over/underestimation ratio corresponding to
punching shear section, BS8110-1997.
Test group Average over/underestimation ratio (b) due to
punching shear provisions at
Edge of the column 0.5d d 1.5d
DF6 to DF22 0.58 1.13 1.65 2.45
TI to TXII 0.58 0.66 0.57 0.72

Please cite this article in press as: F. Abdrabbo et al., Structural design of isolated column footings, Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.
aej.2016.06.016
10 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- Comparison between measured and predicted failure loads
istic high predicted load. Also one way shear provisions at dis- based on code provisions of punching shear at code–defined
tance d from the edge of the column reflect predicted loads sections at distance 0.5d, d, 1.5d, 2d from the edge of the col-
nearly equal to those obtained from punching shear provisions umn and at column perimeter was conducted.
at distance d/2 from the edge of the column, tests specimen The predicted failure loads of the tested specimens were cal-
DF8, DF10, DF13. Therefore, 50% of the tested Specimens culated based on punching shear provisions at distance 0.5d
revealed minimum predicted failure loads based on punching from the edge of the column. The ratios bp of the predicted
shear at distance 0.5d from the edge of the column; the to the measured failure loads are presented in Table 8.
shear-span/depth ratio of the specimens is less than 2.0; Fig. 18 presents correlating between measured and predicted
33.3% of the tested specimen revealed minimum predicted fail- failure loads.
ure loads based on one-way shear at distance d, the shear– Table 8 indicates that bp values varying from 1.38 to 1.87
span/depth ratio of the footing is equal to or bigger than for test specimens DF6 to DF22, and from 0.54 to 1.08 for
2.0, 11.2% of the tested specimens revealed minimum pre- tests TI to TXII. The average values of b of series DF6 to
dicted failure load based on punching shear at distance 1.5d DF22 is 1.59, and the average value of series TI to TXII is
from the edge of the column, the shear span/depth of the foot- 0.86. If all test results are considered, the average values of b
ing ratio is 2.70. 5.5% of the tested specimens revealed mini-
mum 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 Table 8 Predicted and measured failure loads, (EC2-2004),
is related to shear span/depth ratio. Thus, one can conclude punching shear at distance d/2 and at distance d.
that as the value of shear span/depth ratio increases, the loca- Test Measured Predicted bp Predicted bp
tion of critical section of shear failure goes further away from load (kN) failure load Pp failure load Pp
the column edge. Punching shear at distance 0.50d from the (kN) at (kN) at
edge of the column is anticipated in footing having shear span distance 0.5d distance d
to depth ratio less than 2.0. One way shear at distance d from DF6 2836 3913.2 1.38 7687.4 2.71
the face of the column may occur in footings having shear DF7 2569 3718.0 1.45 5069.3 1.97
span/depth ratio bigger than 2.0 and less than 2.70. Footing DF8 1203 1932.2 1.61 1957.7 1.63
having shear span/depth ratio higher than 2.70 exhibit punch- DF10 1638 2329.0 1.42 2359.6 1.44
ing shear either at distance d or 1.5d from the edge of the DF11 2813 4071.5 1.45 7998.3 2.84
column. DF12 2208 3749.9 1.70 5112.9 2.32
Fig. 17 presents the predicted minimum failure loads of DF13 1839 3431.0 1.87 3645.1 1.98
DF14 1478 2324.4 1.57 2381.1 1.61
each footing and the corresponding measured failure loads.
DF15 2750 5135.5 1.87 9997.0 3.64
The figure 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 DF21 2860 4469.2 1.56 6093.6 2.13
low b-values, the ratios of the predicted to measured loads DF22 2405 4114.9 1.71 4371.7 1.82
become equal to 1.0. However if test series TI to TXII are TI 906 980.1 1.08 986.9 1.09
omitted, the average b value becomes 1.11. So that it can be TII 1050 569.2 0.54 470.6 0.45
considered that the BS 8110.1–1997 code provisions of shear TIX 430 335.2 0.78 254.1 0.59
may predict failure loads agree with the measured failure loads TX 656 623.1 0.95 565.3 0.86
within a precision of 10% (90% to 110%), irrespective of the TXI 451 452.6 1.00 374.2 0.83
TXII 440 363.6 0.83 300.6 0.68
shape of contact stress at footing –soil interface.

3.5
5.0
3.0
predicted load, kN*103
predictedload, kN*103

4.0
2.5
2.0 3.0

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
Measured failure load kN*103 Measured failure load kN*103

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.

Please cite this article in press as: F. Abdrabbo et al., Structural design of isolated column footings, Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.
aej.2016.06.016
Structural design of isolated column footings
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
re-adjusted load, kN*103
3.0
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 4.0
mesured failure load kN*103
Figure 19 Comparison between re-adjusted and measured fail-
ure loads (EC2-2004), punching shear at distance d/2.
9.0
7.5
predictedload, kN*103
6.0
4.5
3.0
1.5
0.0
0.0 1.5 3.0 4.5 6.0 7.5
Measured failure load kN*103
Figure 20 Comparison between predicted and measured failure
loads (EC2-2004), punching shear at distance d.
Please cite this article in press as: F. Abdrabbo et al., Structural design of isolated column footings, Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.
aej.2016.06.016
11
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.
The table indicates that the value of bp for tests DF6 to
DF22 varies from 2.3 to 4.66, with an average value of 3.44,
while tests 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 between group of test DF6 to DF22 and the other
group TI to TXII is due to the concentration of contact stres-
ses acting on footing model underneath column stub.
Generally test results revealed that the code provisions for
punching shear at distance 1.5d overestimate the measured fail-
ure load, and the average overestimation ratio is 2.38.
The failure loads of the tested footing were predicted in
accordance with punching shear provisions stated in EC2-
2004 code at distance 2d from the edge of the column. Table 9
6.0
5.0
re-adjusted load, kN*103
4.0
3.0
2.0
1.0
0.0
9.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0
measured failure load kN*103
Figure 21 Comparison between re-adjusted and measured fail-
ure loads (EC2-2004), punching shear at distance d.
12 F. Abdrabbo et al.

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

predicted load,Pss kN*103

punching shear at distances 1.5d and 2d.
10.0
Test Measured Predicted bp Predicted bp
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
DF11 2813 NA NA NA
0.0
0.0 2.0 4.0 6.0 8.0 10.0 12.0
DF12 2208 11185.6 5.07 NA NA
DF13 1839 5812.6 3.16 10342.9 5.62 Measured failure load kN*103
DF14 1478 3720.7 2.52 4996.9 3.38
DF15 2750 NA NA NA Figure 23 Comparison between predicted and measured failure
DF19 2790 NA NA NA loads (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
TI 906 1686.2 1.86 NA NA to 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*103

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-
8.0 ure loads and the corresponding predicted values were drawn;
6.0 Fig. 24, for the sake of comparison. BS (8110) code provisions
under estimate the failure load based on calculations of punch-
4.0
ing shear at the perimeter of the column by a factor of 0.88.
2.0 The predicted minimum failure loads for each test specimen
out of the implementation of punching shear at column
0.0
0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 perimeter, at distances d/2, d and 1.5d in accordance with
Measured failure load kN*103 EC2 provisions were considered. The minimum predicted fail-
ure load of specimens coded DF6 to DF22, TX and TXII was
Figure 22 Comparison between predicted and measured failure obtained from punching shear provisions at the column
loads (EC2-2004), punching shear at distance 1.5d. perimeter. The minimum 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
3.5
the predicted values were drawn; Fig. 23.
Table 9 indicates that the EC2-2004 code provisions of 3.0
predictedload, kN*103

punching shear at distance 2d from the edge of the column

2.5
are not applicable (NA) for most of the footing models in test
series DF6 to DF22. So that it is not easy to achieve a firm 2.0
conclusion from the study. Even though 42% of the tested 1.5
specimens on which the code provisions are implemented,
revealed that the code provisions for punching shear at dis- 1.0
tance 2d overestimate the ultimate loads of the footings. The 0.5
average overestimation ratio 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. Measured failure load kN*103
The 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 Figure 24 Comparison between predicted and measured failure
value of 0.91, tests TI to TXII revealed values of bp varies from loads (EC2-2004) punching shear at the perimeter of the column.

Please cite this article in press as: F. Abdrabbo et al., Structural design of isolated column footings, Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.
aej.2016.06.016
Structural design of isolated column footings 13

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

predicted load, kN*103

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
DF12 2208 1877.7 0.85 0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
DF13 1839 1854.3 1.01
DF14 1478 1402.3 0.95 Measured failure load kN*103
DF15 2750 2281.9 0.83
DF19 2790 1940.3 0.70 Figure 25 Comparison between minimum predicted ultimate
DF20 3037 2984.0 0.98 loads and measured failure loads (EC2-2004), punching shear at
DF21 2860 3002.9 1.05 the 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
The main target of the work presented is to explore the preci-
TXI 451 396.9 0.88
sion of code provisions related to the structural design of iso-
TXII 440 212.3 0.48
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
predicted failure loads of test specimen TII and TXI were the laboratory measured failure loads were conducted. The
obtained from code provisions for punching shear at distance study revealed the following main conclusions.
d from the edge of the column. The minimum predicted failure The predicted failure loads of isolated column footings sub-
load of test specimen TI was obtained from punching shear jected to uniformly distributed contact stress in accordance
provision at distance 0.5d from the edge of the column. with ACI318-08 code provisions are controlled by punching
Table 11 indicates that punching shear at column perimeter shear at code-defined critical section, contrary to ECP (203-
of the tested specimen, prevails predicted failure loads close to 2011) code provisions which the predicted failure loads are
the measured values and the divergence between predicted and controlled by one-way shear at code defined critical section.
measured failure loads increases as the code defined critical The critical section provided by the two codes is at distance
section goes further from the edge of the column. 0.5d from the edge of the column. But ACI 318-08 code provi-
Theoretical calculations in accordance with code provisions sions considered punching shear while ECP 203-11 code provi-
indicated that the minimum predicted failure loads of 79% of sions considered one-way shear.
the 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, the failure loads of isolated column footings subjected to uni-
5.2% of the tested specimen indicates that the minimum pre- formly distributed contact stress. The underestimation ratio is
dicted failure load was achieved by code provisions of punch- 0.75 and 0.70 respectively. If the supporting material of the
ing shear at distance 0.5d, while 10.4% of the tested specimen footing is producing bell shaped contact stress distribution at
revealed minimum predicted failure loads in accordance with footing-soil interface with maximum value beneath column
code provisions of punching shear at distance d, 5.2% of the stub, the underestimation ratio becomes 0.44 in both code
tested specimens indicated a minimum predicted failure load provisions.
by code provisions of punching at distance 1.5d. Observations It is not advisable to consider punching shear at the perime-
of the effect of a/d ratio on the location of punching shear were ter of the column in accordance with BS 8110.1-1997. If section
carried out. of shear failure is considered away from the edge of the col-
These observations indicated that punching shear in foot- umn, at 0.5d, d and 1.5d for punching shear and at d for
ings may occur at column perimeter or at distance 0.5d as long one-way shear, the minimum predicting failure loads of foot-
as the shear span/depth ratio is less than 2.25. Punching shear ings subjected to uniformly distributed contact stress in accor-
at distance d may occur when a/d ratio is bigger than 2.25 and dance with BS 8110-1997 shall equal 1.13 times the measured
less than or equal to 2.7. For greater ratio of a/d punching failure loads. The ratio is 0.56 in case of footings supported
shear at distance 1.5d is anticipated. on geomaterial producing belled shape contact stress
The predicted minimum failure load of each footing was distribution.
assessed out of calculations failure load assuming punching Regarding BS 8110-1997 provisions, punching shear at dis-
takes place at perimeter of column, at 0.5d, 1.5d, and 2d from tance 0.5d from the edge of the column is anticipated in footing
the edge of the column. Fig. 25 presents correlation between having shear span/depth ratio less than 2.0. One way shear at
measured and minimum predicted failure loads. The figure distance d from the edge of the column is anticipated in footing
indicates an average underestimation ratio of 0.85, compared having shear span to depth ratio bigger than 2 and less than
to 0.93 obtained by the implementation of BS (8110-1997). 2.70. Footings having shear span/depth ratio bigger than

Please cite this article in press as: F. Abdrabbo et al., Structural design of isolated column footings, Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.
aej.2016.06.016
14
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.
Please cite this article in press as: F. Abdrabbo et al., Structural design of isolated column footings, Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.
aej.2016.06.016
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.