INDIAN GEOTECHNICAL SOCIETY CHENNAI CHAPTER

ABSTRACT: A large scale experimental program using vertical and batter model piles in sand subjected to pull out loads has been
carried out in a model tank of size 1m x 1m x 1m. Mild steel pipes of varying diameters, lengths, and shapes are used as model piles. A
poorly graded river sand having specific gravity G=2.67, Uniformity coefficient=3.53, emax =0.81, emin = 0.54 has been used as
foundation medium. The influence of pile inclination, pile length, diameter, surface characteristics and shape were investigated. It is
inferred that net ultimate pullout capacity increases significantly with increase in length to diameter ratio. Pullout capacity also increases
with increase in diameter. Net ultimate pullout capacity increases with increase in batter angle attains a maximum value and then
decreases. Sand coated piles are found to be resisting more pullout forces than smooth piles. It is also found that for piles of varying
shape but of constant volume, circular pile will resist more uplift force compared to square or rectangular piles. The experimental values
of net ultimate pullout capacities have been compared with predictions made by available theories.
Pullout Capacity of Model Piles in Sand
K. Kimi Bose
1
and A. Krishnan
2

1
M.E Student, Department of Civil Engineering, Government College of Technology, Coimbatore kimikichu@yahoo.co.in
2
Asst. Professor, Department of Civil Engineering, Government College of Technology, Coimbatore.
Introduction
When structures are constructed below the
ground water table or if they are constructed under water
uplift forces are to be applied on the basement of
structures. Also in the case of transmission line towers,
mooring systems for ocean surface or submerged
platforms, tall chimneys etc are usually subjected to
overturning moments due to wind, wave pressure or ship
impact etc. These overturning moments are transferred
to structures foundation in the form of compression on
some elements and pullout on others. The type of
foundation usually recommended is a combination of
vertical and batter piles. In this paper the behaviour of
vertical and batter piles under pull out loads has been
investigated.
To study the effect of pile inclination, pile length,
diameter, shape, surface characteristics and pile tip
properties on uplift capacity of piles, laboratory
experimental investigation is carried out.
Meyerhof(1973) presented an analysis to
determine the axial pullout resistance of batter piles. For
pile of inclination, with vertical axis , and vertical depth
of embedment, D, ignoring pile weight, the pullout
resistance, Pu is given as,
( )
u u s
P 'K tan A

= (1)
where,
As =Embedded pile surface area

o =Average effective overburden pressure= D/2

Ku =Uplift coefficient
=Pilesoil friction angle
Awad and Ayoub (1976) studied about the
ultimate uplift capacity of vertical and inclined anchors in
cohesion less soil. An empirical equation was developed
for determining the ultimate uplift capacity of inclined
piles.
( )
cos
p p
cos +tan


=

(2)
Where Po =Net ultimate uplift capacity of vertical
pile
For vertical piles,
o u
P =P -W
Hanna and Afram(1986) suggested an analytical
procedure to evaluate ultimate pullout resistance P as
( ) P =Pcos /2 (3)
Chattopadhyay and Pise (1986)
proposed a theoretical analysis for predicting the axial
uplift capacity of inclined piles, embedded in sand. They
concluded that for equal length piles, the ultimate uplift
capacity of inclined pile increases with increase in
inclination of pile and decreases after reaching maximum
value of L at =(15to 22.5). Pise and Sharma (1994)
carried out extensive work on uplift behaviour of anchor
piles in sand under axial pulling loads. A comparison was
made between between the experimental and theoretical
and experimental values. It was concluded that uplift
capacity increases with increase in pile friction angle,
depth of embedment and B/d ratio.
Different theories regarding behaviour of piles
under different loading conditions have been developed
over the years. The reliability of the theories can be
demonstrated only by comparison of experimental results
on model or field piles with the theoretical predictions.
Full-scale field tests, though highly desirable, are
generally expensive and difficult to perform. In the
absence of resources and scope of testing prototype
small scale laboratory model test conducted on piles in
foundation medium prepared under controlled condition
may serve the purpose to some extent. Properly
conducted laboratory tests, with known parameters
KEYWORDS: Pullout capacity, Model piles, Sand bed, Batter Piles, Pile roughness

STUDENTS PAPER COMPETITION 2009

50
affecting the soil-pile response under pulling loads would
provide information on qualitative and quantitative
contributions of such parameters on ultimate resistance
of piles in the absence of field test results.
Compared to previous studies in this area, this
investigation proposes to consider wider range of
parameters and their effects on the uplift capacity of
piles.
Experimental Set Up and Model Tests
Testing Programme
Tests under axial pullout have been carried out on
tubular mild steel model piles having outer diameter
27.5mm, 33mm, 47mm at different angles with vertical
axis as 0, 10, 20 and 30.The model piles have been
Test Arrangement
Axial pullout loads were applied to the piles
through double pulley arrangement The steel loading
frame, movable along the length of chamber with an
inverted pulley was used to align the axis of batter pile
and wire rope. The non-extensible steel wire rope was
attached to the pile top by bolting. The wire rope was
taken first through an inverted pulley and then over the
second pulley. Loading pan where dead weights were put
for loading was fixed at the other end of wire as shown in
Figure1. The position of first pulley was fixed according to
the alignment of the wire rope and pile axis as per the
inclination of the pile. A long steel flat plate was placed
along the width of the chamber to mount magnetic base
of two dial gauges. Two dial gauges were fixed
equidistant from pile axis. The loads were applied by
dead weights in the loading pan starting the smallest,
with gradual increase in stages. Dial gauge readings
were observed for both dial gauges for each increment of
loading when it becomes stable. Average value of
displacement as recorded from both the dial gauges have
been taken as axial displacement of the pile
corresponding to the pullout load applied.

tested for L/d ratios 8, 16, 24 and 32 and three different
surface roughnesses of piles. All tests have been
conducted in model chamber of size 1000mmx 1000mmx
1000mm deep.
The tests were performed in dry local
river sand of unit weight 16.68 kN/m
3
; angle of shearing
resistance of 38; relative density,Dr 80%.
The test programme consisted of the following
arrangement.
> Vertical and inclined piles subjected to axial pull.
The angle of inclination being 10,20and30 with
respect to vertical direction.
> Vertical piles of constant length but different
diameters subjected to vertical pull.
> Vertical and inclined piles of constant length but
different surface characteristics subjected to
vertical pull.
> Vertical piles of constant volume but different
shapes such as round, square and rectangle
subjected to vertical pull.
Properties of Soil used in the Test
A poorly graded river sand having specific gravity
G=2.67 was used for the tests. The D10, D 30, D60 of the
soil are 0.17mm,0.25mm and 0.6mm. The coefficient of
curvature (Cc) and uniformity coefficient (Cu) of the soil
are 0.61 and 3.53. According to Indian Standard
Classification System, this soil can be classified as poorly
graded sand with a letter symbol SP.
To determine the density and void ratio of sand,
a number of trials have been carried out for varying
heights of fall. It was understood that the height of fall of
sand goes on increasing, the density of sand increases.
To verify this, a steel rectangular box of size
335mmx235mmx35mm depth was used to pour the sand
with help of hopper. The box was filled with sand for
different heights of fall ie, 0cm, 5cm, 10cm, 15cm, 20cm,
25cm, 30cm, 35cm, 40cm, 50cm, 60cm.
For every height of fall, corresponding unit
weight, void ratio, relative densities were calculated and
are shown in Table 1
Height of
fall (cm)
Unit Weight
(kN/m
3
)
Void Ratio
Relative
Density in %
0 14.73 0.813 0
5 15.4 0.734 30
10 15.7 0.700 42
15 15.99 0.670 53
20 16.23 0.645 62
25 16.45 0.623 70
30 16.68 0.601 80
35 16.88 0.582 85
40 17.06 0.565 92
50 17.25 0.548 98
60 17.32 0.542 100
Table 1 Height of Fall and Relative Density
Fig 1. Model Pile and Test Set-up

PULLOUT CAPACITY OF MODEL PILES IN SAND

51
Test Procedure
The technique of sand placement plays an
important role in the process of getting reproducible
densities in a reasonable amount of time. The reliability
of the results depends much on the density of foundation
material and therefore the required density of sand was
predetermined. According to that, the sand was poured in
the chamber continuously through the slit of hopper
keeping the height of the fall of sand of about 30cm. This
technique was used by Pise, (1969); Pal,(1983); and
Chattopadhyay,(1986).

After placing the pile in position the sand was
poured uniformly around the pile by moving hopper by
hand. The dial gauge fixing arrangement was attached to
the pile top by tightening bolt and screw arrangement.
This method of sand pouring gave the placement unit
weight of 16.68 kN/m
3
corresponding to relative density
Dr80%.
Test Results
Pullout Load-Axial Displacement Diagrams
Pullout Load versus Axi al Displacement
Diagrams for Variation in Length and Batter
Angle
In general the load displacement responses for all
the piles are similar. It is observed that load displacement
diagrams are practically linear at initial stages of loading
for different batter angles and then they are non linear.
The Figures 2 to 4 show the load displacement diagram
for L/d=16 for different surface finishes. Figure 5 shows
typical load-displacement curves for piles having L/d=16
for different surface finishes at =0. It can be clearly
seen that for a particular displacement the pile resistance
to pullout load increases with surface finishes smooth to
rough.
Figure 6 show typical load-displacement curves
for piles having different L/d ratios for batter angle of
=20. It can be clearly seen that for a particular
displacement the pile resistance to pullout load increases
with increase in L/d ratio of the pile. In general the
responses are found to be similar. As the L/d ratio of pile
increases, the resistance offered by the pile at any axial
displacement increases significantly.
0
0.02
0.04
0.06
0.08
0.1
0.12
-0.5 0 0.5 1 1.5 2 2.5 3
AXIAL DISPLACEMENT (mm)
P
U
L
L
O
U
T

L
O
A
D

(
k
N
)
=0
=10
=20
=30
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
-0.5 0 0.5 1 1.5 2 2.5 3 3.5
AXIAL DISPLACEMENT (mm)
P
U
L
L
O
U
T

L
O
A
D

(
k
N
)
=0
=10
=20
=30
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 1 2 3 4
Axial Displacement(mm)
P
u
l
l
o
u
t

L
o
a
d
(
k
N
)
=0
=10
=20
=30

Fig 2. Pull out Load Versus Axial Displacement in mm for
L/d=16 for Smooth Piles
Fig 3. Pull out Load Versus Axial Displacement in mm for
L/d=16 for Medium Rough Piles
Fig 4. Pull out Load Versus Axial Displacement in mm for
L/d=16 for Rough Piles
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
-1 0 1 2 3
AXIAL DISPLACEMENT (mm)
P
U
L
L
O
U
T

L
O
A
D

(
k
N
)
smooth
medium rough
rough
0
0.05
0.1
0.15
0.2
0.25
0.3
-0.5 0 0.5 1 1.5 2 2.5 3
AXIAL DISPLACEMENT IN mm
P
U
L
L
O
U
T

L
O
A
D

I
N

k
N
for L/d=8
for L/d=16
for L/d=24
for L/d=32
Fig 5. Influence of Surface Finish on Pullout
Load - Displacement Response
Fig 6. Influence of Length on Pullout
load - Displacement Response =20
STUDENTS PAPER COMPETITION 2009

52
Pullout Load versus Axi al Displacement
Diagrams for Variation in Diameter
Figure 7 shows the variation of uplift capacity with
diameter of pile. It is found that as diameter increases the
uplift capacity also increases.
Net Ultimate Pullout Capacity Diagrams
The experimental value of net ultimate pullout
capacity, Pu has been calculated by deducting the
component of weight of pile i.e, Wcos from the ultimate
pullout capacity, Po. Where W is the weight of model pile.
Variation of net uplift capacity with batter angle
The variation of net ultimate pullout capacity, Pu
with respect to the batter angle is shown in Figure10.
From this figure it is seen that net ultimate pullout
capacity increases with increase in , attains maximum
value and then decreases. The maximum value of Pu
corresponds to batter angle 20.
Pullout Load versus Axi al Displacement
Diagrams for Variation in Shape (Figure 8)

0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
-0.5 0 0.5 1 1.5 2
AXIAL DISPLACEMENT(mm)
P
U
L
L
O
U
T

L
O
A
D
(
k
N
)
DIA=27.5mm
DIA=33mm
DIA=37mm
Fig 7. Variation of Uplift Capacity with the Diameter of Pile
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
-0.5 0 0.5 1 1.5 2 2.5 3
AXIAL DISPLACEMENT(mm)
P
U
L
L
O
U
T

L
O
A
D

(
k
N
)
square
rectangle
round
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
-0.5 0 0.5 1 1.5 2 2.5 3 3.5
AXIAL DISPLACEMENT (mm)
P
U
L
L
O
U
T

L
O
A
D

(
k
N
)
=0
=10
=20
=30
Pullout Load versus Axi al Displacement
Diagrams for Variation in Surface Characteristi cs
Fig 9. Variation of Uplift Capacity for Medium Rough Piles

Fig 8. Variation of Uplift Capacity with the Shape of Pile.
0
0.05
0.1
0.15
0.2
0.25
0.3
0 5 10 15 20 25 30 35
BATTER ANGLE
n
E
T

U
L
T
I
M
A
T
E

P
U
L
L
O
U
T

C
A
P
A
C
I
T
Y
(
k
N
)
L/d=8
L/d=16
L/d=24
L/d=32
Fig 10. Net Uplift Capacity Versus Batter Angle
Variation of net ultimate pullout capacity, Pu
with diameter of pil e (Figure 11)
Variation of net ultimate pullout capacity, Pu
with shape of pile (Figure 12)
0.128
0.13
0.132
0.134
0.136
0.138
0.14
0.142
0 5 10 15 20 25 30 35 40
DIAMETER OF PILE (mm)
N
E
T

U
L
T
I
M
A
T
E

P
U
L
L
O
U
T

C
A
P
A
C
I
T
Y
(
k
N
)
L/d=24
Fig 11. Net Uplift Capacity Versus Diameter of Pile
0
0.02
0.04
0.06
0.08
0.1
0 0.2 0.4 0.6 0.8
LENGTH OF PILE (m)
N
E
T

U
L
T
I
M
A
T
E

P
U
L
L
O
U
T

C
A
P
A
C
I
T
Y

(
k
N
)
VOLUME=0.003
m3
Fig 12. Net Uplift capacity versus length of pile
PULLOUT CAPACITY OF MODEL PILES IN SAND

53
Comparison of Results
The validity of any experimental work can be
proved only by comparing those results with established
results. The experimental results are used to check the
validity by comparing them with those values calculated
from theories proposed by Meyerhof (1973), Awad and
Ayoub(1976), Chattopadhyay and Pise(1986) and Hanna
and Afram(1986). These comparisons are presented in
Figures 14 to 17. Meyerhofs theory, Hanna and Aframs
theory and Awad and Ayoub gives general decreasing
trend for Pu values with respect to increasing values
which is found to be conflicting with observed
experimental variation. The results from Chattopadhyay
and Pises theory show general trend of initial increase
and then decrease in Pu values with respect to increase
in -value having maximum value of Pu at 20. Upto
60% plus or minus variation has been found from the
experimental Pu values.
Conclusion
> From the laboratory investigations that have been
carried out, the following conclusions are drawn.
> Axial pullout load versus axial displacement
diagrams for batter piles are practically linear at
initial stages of loading and non-linear at later
stages.
> The resistances offered by the pile at any axial
displacement increases significantly with increase
roughness of pile i.e, sand coated piles are found
to be resisting more pullout forces than smooth
piles.
> The resistances offered by pile at any axial
displacement increases significantly with increase
in L/d ratio.
> Axial displacement of about 3% to 10% of pile
diameter is required to attain the ultimate pullout
capacity.
> The net ultimate pullout capacity of a pile
increases significantly with increase in L/d ratio.
> Net ultimate pullout capacity of a pile increases
with increase in batter angle,, attains maximum
Variation of net ultimate pullout capacity, Pu
with L/d Ratio (Figure 13)
0
0.05
0.1
0.15
0.2
0.25
0.3
0 5 10 15 20 25 30 35
L/d
N
E
T

U
L
T
I
M
A
T
E

P
U
L
L
O
U
T

C
A
P
A
C
I
T
Y
(
k
N
)
=0
=10
=20
=30
Fig 13. Net Uplift capacity versus L/d Ratio
0
0.01
0.02
0.03
0.04
0.05
0.06
0 10 20 30 40
BATTER ANGLE
N
E
T

U
L
T
I
M
A
T
E

P
U
L
L
O
U
T

C
A
P
A
C
I
T
Y
MEYERHOF
HANNA and AFRAM
AWAD and AYOUB
CHATTOPADHYAY and
PISE
PRESENT
EXPERINENTAL
INVESTIGATION
BATTER ANGLE
0
0.02
0.04
0.06
0.08
0.1
0.12
0 10 20 30 40
BATTER ANGLE
N
E
T

U
L
T
I
M
A
T
E

P
U
L
L
O
U
T

C
A
P
A
C
I
T
Y
(
k
N
)
MEYERHOF
HANNA and AFRAM
AWAD and AYOUB
CHATTOPADHYAY
and PISE
PRESENT
EXPERIMENTAL
INVESTIGATION
0
0.05
0.1
0.15
0.2
0.25
0 10 20 30 40
BATTER ANGLE
N
E
T

U
L
T
I
M
A
T
E

P
U
L
L
O
U
T
C
A
P
A
C
I
T
Y
(
k
N
)
MEYERHOF
HANNA and AFRAM
AWAD and AYOUB
CHATTOPADHYAY and
PISE
PRESENT EXPERIMENTAL
INVESTIGATION
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0 10 20 30 40
BATTER ANGLE
N
E
T

U
L
T
I
M
A
T
E

P
U
L
L
O
U
T

C
A
P
A
C
I
T
Y
(
k
N
)
MEYERHOF
HANNA and AFRAM
AWAD and AYOUB
CHATTOPADHYAY and
PISE
PRESENT EXPERIMENTAL
INVESTIGATION
Fig 17. Net ultimate pullout capacity - Batter angle for
L/d=32
Fig 16. Net ultimate pullout capacity - Batter angle for
L/d=24
Fig 15. Net ultimate pullout capacity - Batter angle for
L/d=16
Fig 14. Net ultimate pullout capacity- Batter angle for L/d=8

STUDENTS PAPER COMPETITION 2009

54
value and then decreases. The maximum value of
Pu occurs at 20, and it is about 10% to 20%
more than the vertical pile capacity.
> Pullout capacity also increases with increase in
diameter.
> It is also found that for piles of varying shape but
of constant volume, circular pile will resist more
uplift force compared to square or rectangular
piles.
> The variation of net ultimate pullout capacity with
batter angle, by Chattopadhyay and Pise (1986)
analysis is qualitatively similar to observed
experimental variation. The theory also predicts
maximum capacity at 20.
> Method of Meyerhof (1973) and Hanna and Afram
(1986) Awad and Ayoub(1976) predict decreasing
trend of Pu values with increase in -value.
References
A.Awad And Ayoub(1976): Ultimate Uplift Capacity of
Vertical and Inclined Piles in Cohesion less Soils.,
Proc.5
th
conference on Soil Mechanics and Foundation
Engineering,Budafest, pp. 221-226.
Chattopadhyay, B.C. And Pise, P.J .(1986): Axial Uplift
Capacity of Inclined Piles, Indian Geotechnical Journal
Vol.16, No. 3, pp.198-213.
Das, B.M. And Seeley, G.R.(1975): Uplift Capacity of
Buried Model Piles in Sand, Journal of GTE Div.,ASCE,
Vol.10, pp.1091-1094.
Hanna, A.M. And Afram, A.(1986): Pullout Capacity of
Single Batter Piles in Sand, Canadian Geotechnical
Journal, Vol.23, No.3, pp.387-392.
Ismael, N.F. And Klym, T.W.(1979): Uplift and Bearing
Capacity of short piers in sand, Journal of GTE Div.,
ASCE, Vol.105, No.5, pp.579-594.
Meyerhof, G.G. And Adams, J .I.(1968): The ultimate
uplift capacity of foundation, Canadian Geotechnical
Journal, Vol.5, No.4, pp.225-244.
K.Rajagopal And V.Srihari(1998): Experimental
Investigations on Pullout Capacity of Vertical Anchors,
Indian Geotechnical J ournal, Vol.28(2), pp147-166.
Shanker, K., Basudhar, P. K. And Patra N. R.(2006):
Uplift Capacity of Single Piles Embedded in Sand, IGJ.,
Vol. 36. No. 4, pp.334-347.
B.V.R Sharma And P.J Pise.(1994): Uplift Capacity of
Anchor Piles in Sand Under Axial Pulling Loads, Indian
Geotechnical Journal, Vol.24 (3), pp181-202.