Alexandria Engineering Journal (2010) 49, 387–391

Alexandria University

Alexandria Engineering Journal

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

ORIGINAL ARTICLE

Bulging as a pile imperfection

Amr Z. El Wakil *, M. Kassim
Department of Structural Engineering, Faculty of Engineering, Alexandria University, Alexandria, Egypt

Received 17 April 2010; accepted 20 July 2010

Available online 26 January 2011

KEYWORDS Abstract Piled foundation is a method of overcoming difficulties of founding on soft soils to ensure
Bulging; structural safety by transferring structural load through weak soil to a stiff stratum. Due to piles
Shear resistance; importance, it is becoming necessary to avoid pile defects. These defects arouse from inadequate
End bearing embeddement ground investigations, construction phases, pile loading test, and natural unforeseen geological causes.
length; In this paper bulging is studied as a pile imperfection. The effects of bulging width, length, and shape
Ultimate load on the pile ultimate load are investigated. Bulging may increase pile ultimate load though it is still con-
sidered as a pile defect. It was found that bulging can increase the ultimate pile load up to 200%.
ª 2010 Faculty of Engineering, Alexandria University. Production and hosting by Elsevier B.V.
All rights reserved.

1. Introduction
Piles are used in many structures to transfer loads through soft
ground to stiff stratum. The behavior of piles in the ideal state
has been widely investigated, but it is important to study the
behavior of non-ideal conditions on the pile behavior. Imperfect
pile means that there is one or more non-ideal conditions exit.
The pile imperfection may be due to inadequate ground investi-
gation, natural sources, pile loading test, and construction prob-
lems such as necking, voids, weak area, and bulging. Terzaghi
* Corresponding author.
E-mail addresses: amr-elwakil@hotmail.com (A.Z. El
monzirkassim@hotmail.com (M. Kassim).
1110-0168 ª 2010 Faculty of Engineering, Alexandria University.
Production and hosting by Elsevier B.V. All rights reserved.
Peer review under responsibility of Faculty of Engineering, Alexandria
University.
doi:10.1016/j.aej.2010.07.004
[7] mentioned that the earth is never uniform and its proper-
ties are too complicated for theoretical treatment. Wong [8]
conducted a practice survey on quality assurance for bored piles
among the defects reported soft toes, cavities, and short piles not
founded on the desired levels. Poulos [5] divided piles’ defects
into structural and geotechnical defects. Abdrabbo and
Abouseeda [1] observed that only 30% of bored piles founded
on the designed strata due to insufficient subsurface ground
investigation. Zhang and Wang [3] carried out a centrifuge mod-
el on a group of bored piles containing soft toes and short piles
not reaching the designed level, they concluded that the capaci-
ties of the soft toes piles, and short piles were reduced by 15%,
Wakil), and 56%, respectively. Poulos [6] stated that the effect of
construction process on the pile can affect the pile performance
and causes pile defects. He also classified these defects into
natural sources, inadequate ground investigation, construction
causes, pile loading tests, and loading during operation.
2. Testing program and modeling

Bulging is defined as a kind of pile shape imperfection; bulging

Production and hosting by Elsevier
increases the pile cross section in certain areas along the pile
length. To study the effect of bulging on the pile ultimate load
388 A.Z. El Wakil, M. Kassim

a model of 12 m bored concrete pile of 0.5 m diameter embed-

Table 1 Testing program and results.
ded into 8 m of a very soft clay overlying 4 m of dense coarse
sand was investigated. Twenty runs were conducted on the de- Test No. X1/D X2/D d (m) Punum at S = 10%D/PuEC
fined model. Tests were carried out using axisymmetric model 1 0 8 1.03
in Plaxis finite element program where Mohr–Coulomb model 2 0.1 8 1.12
is used. The mesh has radius 10 m around the pile and 8 m be- 3 0.2 8 1.23
neath pile’s toe. The element used is 15-noded byline axisym- 4 0.5 8 1.55
metric triangular element. These conditions were chosen after 5 1.0 8 2.18
performing extensive sensitivity analysis. For comparison pur- 6 0 8 1.03
7 0.1 8 1.11
pose the theoretical ultimate pile load with no bulging zone
8 0.2 8 1.19
was calculated using the Egyptian Code of Soil Mechanics 9 0.5 8 1.44
and Foundation Engineering [2]. Fig. 1 shows the mesh pattern 10 1.0 8 2.05
and its dimension. The sand used has of the following proper- 11 0.2 4 1.25
ties, internal friction 36, cohesion 2 kN/m2, dilation angle of 12 0.5 4 1.55
6, submerged unit weight of 9 kN/m3, Young’s modulus of 13 1.0 4 2.15
65,000 kN/m2, and Poisson ratio of 0.3. The soft clay has 14 0.2 2 1.05
cohesion of 15 kN/m2, submerged unit weight of 7 kN/m3, 15 0.5 2 1.48
Young’s modulus of 9500 kN/m2, and Poisson ratio of 0.45. 16 1.0 2 2.0
Finally the reinforced concrete has a unit weight of 17 0.2 1 0.8
18 0.5 1 1.15
23 kN/m3, Young’s modulus of 2.1 · 107 kN/m2, and Poisson
19 1.0 1 1.85
ratio of 0.18 (Table 1).

3. Testing results and discussion

3.1. Effect of variation of bulging width

The first variable studied in this article is the effect of bulging

width which starts at the pile head and ends at the end of the
clay layer with width X1 as shown in Fig. 2. The bulging width
has values of 0, 0.1D, 0.2D, 0.5D, and D. The relationships be-
tween the ultimate load numerically obtained related to that

Figure 2 Bulging model.

recommended the Egyptian code (Punum/PuEC) and the settle-

ment related to diameter (D/D) for each test was plotted and
gathered in Fig. 3. As observed from the figure, the pile
Figure 1 Mesh design in bulging model. ultimate load increases as the bulging width X1 increases. This
Bulging as a pile imperfection 389

(Pnum/Pu, code)
0 0.5 1 1.5 2 2.5 3 3.5 4
0
Reference case
-0.05 X1=D
X1=0.5 D
-0.1 X1=0.2 D
X1=0.1 D
(Δ/D)

-0.15

-0.2

-0.25

-0.3

Figure 3 (D/D) versus (Punum/PuEC).

might be attributed to the soil profile as the bulging width in-

creases its contribution for pile capacity acts as an enlarged
base for the pile increasing the end bearing resistance of the
pile and hence increasing the pile ultimate load.
The theoretical ultimate load using Egyptian Code was cal-
culated and compared to the numerical ultimate load at settle-
ment equals to 10%D. The relationship between the ratio of
the ultimate load attained at settlement equals to 10%D to that
calculated by the Egyptian Code and X1/D was plotted and
shown in Fig. 4. This relationship shows that the ultimate load
increased up to 2.18 times at X1 = D.

3.2. Effect of bulging height on the pile ultimate load

Figure 5 Bulging model.
Fig. 5 shows that bulging volume starts from the middle of the
clay stratum and ends at the top of the sand stratum with
thickness X2, where X2 equals 0, 0.1D, 0.2D, 0.5D, and D as
Load (Pnum./ Pu, code.)
X1 for comparison reasons. Then the bulging width was kept 0 0.5 1 1.5 2 2.5 3 3.5 4
the same whilst the height is halved. The relationships between 0
Reference case
the ultimate load numerically attained related to that attained -0.05
X2=D
X2=0.5 D
from the Egyptian Code Punum/PuEC and the settlement re- X2=0.2 D
-0.1
lated to diameter D/D for each test was plotted and gathered X2=0.1 D
(Δ / D )

in Fig. 6. As observed from the figure the pile ultimate load in- -0.15

creases as the bulging width X2 increases, confirming the pre- -0.2

vious finding. The theoretical ultimate load using Egyptian
-0.25
Code was calculated and compared to the numerical ultimate
load at settlement equals to 10%D. The relationship between -0.3

the ratio of the ultimate load attained at settlement equals to

Figure 6 Punum/PEC versus D/D.
10%D to that calculated in the Egyptian Code and X1/D
was plotted and shown in Fig. 7. This relationship shows that
the ultimate load increased up to 2.05 times at X2 = D. The in- 2.2

crease in this case is less than before as the lateral surface area 2
Pultnum10%/Pult

1.8

1.6
2.4
1.4
2.2

2 1.2
Pultnum10%/Pult

1.8 1
0 0.2 0.4 0.6 0.8 1 1.2
1.6
X2/D
1.4
Figure 7 X2/D versus Punum10%/PuEC.
1.2

1
0 0.2 0.4 0.6 0.8 1 1.2
X1/D
decreased, and hence the shear resistance occurred in the clay
layer decreased as well, but this decrease is almost negligible as
Figure 4 X1/D versus Punum10%/PuEC. the end bearing effect is more pronounced.
390 A.Z. El Wakil, M. Kassim

Figure 8 Bulging models to compare shear failure surfaces.

Pnum./ Pu, code.

Pnum / P u, code. 0 0.5 1 1.5 2 2.5 3 3.5 4
0 0.5 1 1.5 2 2.5 3 3.5 4 0
0 d =4m
d=4m d =2m
-0.05
-0.05 d=2m d =1m
-0.1
-0.1 d=1m
Δ/D

-0.15
Δ/ D

-0.15
X1= 0.2D
-0.2 -0.2
X1 = D
-0.25 -0.25

-0.3 -0.3

Figure 9a D/D versus Punum/Pu code for bulging width X = D. Figure 9c D/D versus Punum/PuEC for bulging width X = 0.2D.

3.3. Effect of bulging on the shear strength

Pnum / P u, code.
0 0.5 1 1.5 2 2.5 3 3.5 4 To study the effect of bulging on the shear resistance, the bulg-
0
d=4m ing was assumed to be through out the clay stratum with a
-0.05 d=2m width X1 = 0.2D, 0.5D, where D, is the embeddement depth
-0.1 d=1m of the pile into sand, d, has a value of 1 m, 2 m, and 4 m as
shown in Fig. 8. The same runs were conducted on each case
Δ/D

-0.15
and the relationships between Punum at D = 10%D/PuEC
-0.2
X1 = 0.5 D and D/D were plotted and shown in Figs. 9a–c. From the
-0.25 plotted curves it can be shown that as the bulging width X1
-0.3 increases the effect of the embeddement length is less
pronounced, and as the bulging width decreases the effect of
Figure 9b D/D versus Punum/PuEC for bulging width X = 0.5D. embeddement length is more significant.
Bulging as a pile imperfection
4. Conclusions
Bulging may increase the pile ultimate load though it is still
considered as a defect in the pile. From the accomplished sim-
ulation the following conclusions are reached for this particu-
lar simulation case study:
(1) Bulging increases the pile skin friction force.
(2) Bulging acts as an enlarged base in case of end bearing
piles and hence increasing the pile ultimate load.
(3) As bulging width increases the ultimate load attained
increases as well.
(4) Bulging almost doubles the pile compression load
capacity.
(5) As the bulging width increases the effect of embedde-
ment length of pile decreases.
References
[1] F. Abdrabbo, H. Abouseeda, Effect of construction procedures
on the performance of bored piles, in: M.W. O’Neill, F.C.
391
Townsend (Eds.), Proceedings of the Deep Foundations 2002,
Geotech. Spec. Publ. No. 116, ASCE, Reston, VA, 2002, pp.
1438–1454.
[2] Egyptian Code of Practice (ECP 202-2001), Part 4, 2001, pp. 57–
65, 154–157.
[3] L.M. Zhang, Eric Y.W. Wang, Centrifuge modeling of
large-diameter bored pile groups with defects, ASCE J.
Geotech. Geoenviron. Eng. 133 (9) (2007) 1091–1101, ISSN:
1090-0241.
[5] H.G. Poulos, Analysis of pile groups with defect piles, in:
Proceedings of the 14th International Conference on Soil
Mechanics and Foundation Engineering, Balkema, Rotterdam,
The Netherlands, 1997, pp. 871–876.
[6] H.G. Poulos, Pile behaviour––consequences of geological and
construction imperfections, J. Geotech. Geoenviron. Eng. 131 (5)
(2005) 538–563.
[7] K. Terzaghi, Presidential address, in: Proceedings of the First
International Congress on Soil Mechanics and Foundation
Engineering, vol. 3, Cambridge, MA, 1936, pp. 13–18.
[8] E.Y.W. Wong, Behaviour of Large-Diameter Bored Piles Groups
with Defects, M.Phil. Thesis, The Hong Kong University of
Science and Technology, Hong Kong, 2004.