The HKIE Geotechnical Division Annual Seminar 2014

1 INTRODUCTION

Piles supporting Ioundations can be classiIied in accordance with the Iollowing two basic approaches:

Approach (1): To classiIy piles in accordance with construction method, i.e. either displacement piles, or
replacement piles.
Approach (2): To classiIy piles in accordance with the design mode, i.e. either Iriction piles, or end-bearing
piles.

However, the Iollowing points should be noted in the design oI any oI these pile types:

(1) some pile types can be a mixture oI displacement and replacement piles. In the same manner, some oI them
can be a mixture oI Iriction and end-bearing piles; and
(2) a Iriction pile can be installed by either the displacement or the replacement method or by a combination
oI both. The same applies to end-bearing piles.

The design oI end bearing piles in Hong Kong is relatively simple as in most cases these will be bearing on
intact rock. On the other hand, the design oI Iriction piles will be much more involved as these will not only
be aIIected by the soil properties, the ground and groundwater conditions as well as the construction method,
in particular the latter (Brown et al 2007). The purpose oI this paper is thereIore to propose a rational method
oI designing Iriction piles by Iirstly going through the Iundamental soil mechanics theories. This is Iollowed
by processing the data oI a series oI well planned and meticulously executed Iield works carried out in
ArchSD since the mid 1990s. These comprise analyzing/loading tests on 12 nos. oI Iully instrumented piles.

Design Principles Ior Soil Friction Pile Derived Irom Fundamental
Theories & Field Data
H.Y. Wong
Geotech Eng Ltd, Hong Kong
C.Y. Kan, C.T. Wong & M.K. Leung
Architectural Services Department, Hong Kong
ABSTRACT

Traditional replacement non-percussion piles (e.g. pre-bored rock-socketted steel H-piles/ mini-
piles, or large diameter bored piles) usually derive the resistance Irom end-bearing on hard
stratum only. However, on some sites in Hong Kong (e.g. in Mid-Levels) sound bedrock can only
be Iound at very deep level, and a cost eIIective solution is to adopt displacement or replacement
piles relying on the shaIt Iriction component ('Iriction piles). In this paper, the Iundamental
theories governing the Iorces acting on the pile/soil system and the actual soil stress-strain
relationship are Iirstly considered. In order to develop Irom Iundamental theories the general
principles Ior Iriction pile design, Architectural Services Department ('ArchSD) has been
carrying out since the mid 1990s analyzing/loading tests on 12 nos. meticulously instrumented
Iriction piles. With the loading and deIormation at various depths measured Ior each loading
stage, the relative shear displacement and axial load proIiles can be established Ior each
instrumented pile. These together with the SPT-N value (which is a measure oI the shearing
resistance that is available in the soil) can then establish at various depths the ratio oI shearing
strength to SPT-N value thus mobilized and Iinally a more workable Iriction pile design equation.
151
The HKIE Geotechnical Division Annual Seminar 2014

2 FRICTION AT PILE AND SOIL INTERFACE

2.1 Fundamental soil mechanics theories

Consider, as in Figure 1a, the Iorces acting on the pile/soil interIace and the immediately adjacent elementary
volume oI soil. Adopt the Iundamental theories to Iorces operating in this system:

s
c
a

h
tan (1a)

h
K
v
' (1b)

where c
a
adhesion between pile/soil interIace,
s
Irictional shear stress acting along the vertical pile/soil
interIace,
h
horizontal earth stress acting on the vertical pile/soil interIace, Iriction angle at pile/soil
interIace or Iriction angle oI soil immediately adjacent to pile/soil interIace (whichever is smaller), K
horizontal earth pressure coeIIicient, and
v
' eIIective vertical soil stress immediately adjacent to pile/soil
interIace.

c
a
will be ignored, as its value will be considerably reduced during pile installation, and hence its eIIect on
the results oI the back analysis and proposed design equation will be quite minimal. With this conservative
assumption, combining Eqs (1a) & (1b) will yield:

s


K
v
'

tan (2)

Integrating Eq (2) with respect to the depth z along the complete pile/soil interIace will then yield

P
I

H
0
sDdz D

H
0
v
tan ' K dz (3)

where P
I
total soil Irictional resistance acting over the pile/soil interIace, D pile nominal diameter
(assuming to be constant along the whole length), and H total pile length.

It can be seen Irom Eq (3) that P
I
depends on three soil/site parameters, namely, K,
v
'

and , as well as
their variation with depth z, in particular .






(a) Forces acting on pile/soil system


(b) Typical soil stress-strain curves
152
The HKIE Geotechnical Division Annual Seminar 2014


(c) Mobilization oI Iriction vs shear displacement
at pile/soil interIace at various depths

(d) Variation oI shear stress with normal stress at
pile/soil interIace at various depths

Figure 1: Basic stress and stress-strain relationship in soil and at pile/soil interIace


2.2 Effect of hori:ontal earth pressure coefficient (K)

K depends pre-dominantly on construction method and is also the most variable when moving Irom one site to
another with diIIerent pile construction methods (Kulhawy and Mayne 1990; Rollins et al 2005; Kulhawy and
Chen 2007). This can easily change by several times to even more than 10 times. For a normally consolidated
soil beIore pile construction, K can be considered to equal to K
o
so that the Iollowing conventional equation
K
o
(1sinI) can be assumed to be valid, where I eIIective internal soil Iriction angle. K
o
can thereIore vary
Irom 0.577 to 0.357 Ior I varying Irom a loose soil with a value oI 25
o
to a Iairly dense soil with a value oI
40
o
. For an over-consolidated soil K
o
can be in excess oI 1.0. For a vertical pile embedded in soil, K will be
lowest in the replacement piles and highest in displacement pile (O`Neill 2001). The lowest limit must be the
installation oI bored piles, with either bentonite slurry support or just no lateral support at all. Under such
conditions, it is possible Ior K dropping to lower than K
a
. On the other hand, Ior displacement piles, K will
certainly be increased to larger than K
o
, with the corresponding maximum K
p
being the upper bound. As can
be expected, K will increase with the increase in the amount oI outward soil displacement during pile
construction (Kan et al 2013).

2.3 Effect of effective vertical soil stress (
v
')

In general at any depth below the groundwater table
v
'

is given by:

v
' q z
w
(
w
)(z z
w
) (4a)

where q surcharge at ground surIace, soil bulk density,
w
density oI water, and z
w
depth oI
groundwater table.

For depth above ground water table,


v
' q z (4b)

153
The HKIE Geotechnical Division Annual Seminar 2014

Eqts (4a) and (4b) indicate that: (1) the surcharge (q) is normally small, in the order oI 5 to 10 kPa; (2) the
soil bulk density () does not vary too much in moving Irom site to site; (3) the density oI water (
w
) is a
material constant; and (4) the only parameter that can vary more considerably Irom site to site is the depth oI
groundwater table (z
w
). In the extreme case, z
w
can be lowest along sloping ground with z
w
being quite near
to rock head. On the other hand, on reclaimed areas, z
w
will be only a Iew metres below ground surIace. These
two extreme cases can be best illustrated in the Iollowing example, with length oI pile equal to 30m, 19
kN/m
3
, z
w
4 m in Site A and 30 m in Site B. Even in these extreme cases with z
w
Irom 4 to 30m,
v
' in Site A
varies Irom 76 to 310 kPa whilst
v
' in Site B only Irom 76 to 570 kPa.

2.4 Effect of friction angle ()

is a parameter that usually has not been properly deIined or over-simpliIied in most design works. Very
oIten, this is taken as Iriction angle at pile/soil interIace or Iriction angle oI soil immediately adjacent to
pile/soil interIace, whichever is smaller. Figure 1b shows a typical stress strain relationship Ior both a dense
sand (or stiII clay) as well as Ior a loose sand (or soIt clay) based on extensive actual test works (Taylor 1948;
Terzaghi and Peck 1967). It shows that the Irictional shear stress (
s
) at the pile/soil interIace is a Iunction oI
the shear displacement (e
ps
). In general
s
will increase with the increase in e
ps
(Figures 1c & 1d). However,
this is not a linear increase as soil is not a linear elastic material (Frizzi and Meyer 2000; Misra and Chen
2004).

2.5 Summarv

The above paragraphs discuss the individual eIIect oI each oI these three parameters. Nevertheless, what is
more important is the combined eIIect in the Iorm oI K
v
'

tan at various depths in a particular site. For a
given site, K should remain more or less the same at various depths as there is normally no change in the
construction method.
v
'

will increase with depth but the rate oI increase will depend on the depth oI the
groundwater table. Nevertheless, the general trend is Ior K
v
' to increase with depth. Figure 1c shows that
while K
v
' increases with depth, tan decreases with the decrease in shearing displacement, which will in turn
decrease with depth. Thus, the overall picture Ior the term K
v
'

tan can be illustrated in a more quantitative
manner in Figure 1d, with the plotting oI the shearing stress (
s
) versus normal stress (K
v
'

) carried out at the
upper part, middle part and the bottom part as indicated in Figures 1c & d. There is a general trend oI K
v
'

being lowest and tan highest at the upper part and this trend is reversed in moving to the bottom part. The
resulting eIIect that the mobilized shear resistance decreases with depth can be observed in actual pile test
results as indicated in Figure 3.

3 CURRENT METHODS OF DESIGNING FRICTION PILES AND THEIR LIMITATIONS

For some inexperienced designers, Iriction pile design in soil can be based simplest on Eq (3) with taken as
soil Iriction angle determined Irom triaxial or shear box test and K equal to K
o
. Alternatively, there is (total
stress) method Ior over-consolidated clay with
s
c
u
, where c
u
undrained cohesion (Tomlinson 1971). In
Hong Kong, two other methods are commonly adopted, namely, (eIIective stress) method and N-value
method. method is to evaluate at various depths the value which is equal to Ktan. There are again two
approaches to evaluate this value (Brown et al 2007). The Iirst one is the depth-dependent method by
developing Irom Iield load tests its variation with depth. The second one is a more Iundamental approach by
evaluating in terms oI K and . As Ior N-value method, this is just to correlate the SPT-N value directly to the
shaIt Iriction and is more commonly adopted in Hong Kong as the SPT-N values are readily available in
nearly every project. However, though the SPT-N value is a measure oI the shearing resistance available at
various depths, this does not represent the shearing resistance that can be mobilized.
The above review oI the Iundamental theories governing the stress-strain relationship at the pile/soil
interIace at various depths shows that the soil Iriction in pile is controlled by the Iollowing three major
Iactors: (1) pile construction method, (2) shear displacement along pile/soil interIace at various depths, and (3)
soil properties. As a result, there are two major short-comings oI the above two methods. Firstly, as K can
change very drastically with change in construction method, the variation oI with depth by most empirical
equations cannot reIlect this change. Neither can this be accounted Ior even iI K is considered separately.
Secondly, at various depths is not a constant even Ior the same type oI soil because this depends on the
154
The HKIE Geotechnical Division Annual Seminar 2014
relative pile/soil displacement. This change in with depth can be in a way accounted Ior by the depth-
dependent method, but not by the N-value method as the SPT-N value will in general increase with
increasing depth.

4 FIELD MEASUREMENTS

4.1 Instrumentation

In order to develop a simple and workable design equation Ior Iriction piles in soil especially including the
eIIect oI the pile construction method on the K value and the variation oI tan with depth Ior each
construction method, ArchSD has been carrying out loading tests on a number oI meticulously instrumented
Iriction piles (nominal diameter 300-600 mm or more) with diIIerent construction methods since the mid
1990s. Among these piles, at least Iour diIIerent construction methods have been adopted. The typical details
oI these piles are as summarised in Table 1, and the geology and soil strata oI various sites (together with the
SPT-N values versus depths) are summarised in Table 2.

Table 1: Typical details oI instrumented Iriction piles studied by Arch SD
Site
No.
Pile
No
Site
Location
Pile
Type
Construction
method
Nom
D
(mm)
L
(m)
Max
Load
(kN)
Corr
Sett
(mm)
(2)
E
com
(MPa)
Strain
Gauge

(see
Section
4)
1


2


3



4

1/1
1/2

2/1
2/2

3/1
3/2


4/1
4/2
Yuen
Long

Tung
Chung

Ma On
Shan
Area 90

Ma On
Shan
Area 100
(1)
CFA or
PIP
(Pakt-In-
Place)
piles
Boring with a
CFA to the
required
depth, and
cement mortar
to be pumped
through the
auger tip
under pressure
during liIting.
610
25
25

38
38

45
45


45
45
2922
2922

5360
5540

4400
4600


5400
5200
6.28
6.50

6.00
4.78

9.22
12.35


10.13
12.69
27700
27700

34750
34750

30050
30050


31000
31000
3to
4m int.
Irom
2m
depth

13.58
13.58

11.72
11.72

9.70
9.70


14.76
14.76
5


6

5/1


6/1


Kowloon
Bay

Yeung
Long
Bored
piles

Barette type
pile supported
by bentonite,
and RCD
method
supported by
bentonite
2800
800

1200
40


50

5430


5400

9.35


21.90

27500


27500

2 to
4m int.
Irom
2m
depth
2.52


1.89


7 7/1
7/2
Mid-
Levels
Friction
minipiles
with
shaIt
grouted
Steel section
inserted into
pre-bored
hole; cement
grout injected;
tube-a-
manchette
post-grouting
305

49
49
2600
2600
7.43
8.52

50660
50660
5m
int.
Irom
1m
depth
22.76
22.76
Notes : 1. Ordinary PIP piles with reinIorcement cage oI depth oI 24m were used in Site No. 1, and PIP piles
reinIorced by a H-pile (305305149kg/m with length 24 to36m) were used Ior Site Nos. 2 to 4.
2. All Iriction minipiles were inserted with a built-up H-section in Site No. 7 Ior the whole length oI pile.
3. E
com
elastic modulus oI the combined section oI sand/cement mortar (or concrete) and steel
reinIorcement/section.

155
The HKIE Geotechnical Division Annual Seminar 2014

Table 2: Geology and soil strata in each site studied
Site
No.
Site
Location
Geology Soil strata
Average SPT-N value at depth (m)
Rock
head
level
0
-
4
4
-
8
8
-
12
12
-
16
16
-
20
20
-
24
24
-
28
28
-
32
32
-
36
36
-
40
40
-
44
1
Yuen
Long
Marble
area
Fill
Alluvium
Weathered
soil
12 16 27 31 29 26 29 32 45 43 42
Sound
rock at
Iairly
great
depth,
normally
over
50
60m to a
max.
over
100m
2
Tung
Chung
Marble
area
Fill
Alluvium
Weathered
soil
19 17 21 38 40 29 30 29 34 36 40
3
Ma On
Shan
Area 90
Marble
area
Fill
Marine
deposit
Alluvium
Weathered
soil
21 23 16 12 14 24 36 47 32 31 34
4
Ma On
Shan
Area 100
Marble
area
Fill
Marine
deposit
Alluvium
Weathered
soil
13 10 17 16 12 25 34 43 41 39 42
5
Kowloon
Bay
Granitic
area
Fill
Marine
deposit
Alluvium
Weathered
soil
3 14 13 13 5 12 29 35 88 35 38
6

Yeung
Long
Volcanic
area
Fill
Alluvium
Weathered
soil
19 21 8 9 12 14 44 34 28 31 36
7
Mid-
Levels
Granitic
area
Fill
Alluvium
Weathered
soil
6 27 28 28 26 25 35 36 36 42 52

For each oI the two piles in Site Nos 1, 2, 3 and 4, strain gauges were installed along both sides oI the web
oI the steel H-pile at about 3-4 m intervals. For Site Nos 5 & 6, strain gauges were installed on the
reinIorcement cage at about 2-4 m intervals (Wong 2003). The construction method in these two sites were
quite similar as the end condition oI the bored pile wall beIore concreting was the same, both being supported
by bentonite slurry. For Site No 7, vibrating wire strain gauges were installed along the piles at 10 levels with
two gauges at each level. The interval between each level was 5 m and started at 1 m depth (Kan et al 2013).
In each Iully instrumented pile, the axial Iorce and deIormation at various depths were measured at various
load increments during loading tests, and these were then processed with SPT-N values measured beIore pile
installation. With the axial pile Iorce thus measured, the elastic pile shortening and hence the shear
displacement at the pile/soil interIace at each depth can be evaluated. The shearing resistance between
successive depths can also be estimated Irom the corresponding diIIerence in axial Iorce. Accordingly, the
shearing resistance/displacement relationship at various depths can be established.

156
The HKIE Geotechnical Division Annual Seminar 2014
4.2 Results




Figure 2: Axial load proIile in piles with various construction methods


For each instrumented pile, the Iield test results comprise two major parts:

(1) settlement at pile top at various loading stages, which is basically the compliance test Ior a lot oI piling
works, and will yield the load-settlement curve Ior various loading stages; and
(2) readings oI the strain gauges at each load stage oI the pile load test as carried out in (1) above, which will
yield the axial pile load at each instrumentation depth.

From the above results, the distribution oI axial pile load at various depths is as shown in Figures 2a, b &
c. Moreover, with a knowledge oI the elastic modulus value oI the pile materials, the corresponding shear
displacement oI the pile relative to the surrounding soil at this depth can also be deduced.

4.3 Analvsis




Figure 3: Normalized plot oI relative shearing displacement Ior piles with various construction methods
157
The HKIE Geotechnical Division Annual Seminar 2014

Figures 3a, b & c are normalized plots with the ratio oI the shearing displacement relative to the total at the
pile top against z
i
/H together with the trend lines. From Figures 2a, b & c and Figures 3a, b & c, the Iollowing
can be observed:

(1) Both the shearing resistance and shearing displacement decrease with depth. This is in line with the
Iundamental soil mechanics theories as presented in Figure 1b that the lower the shearing displacement, the
lower the shearing resistance.
(2) A very special Ieature oI the normalized plot in Figures 3a, b, & c is that Ior piles constructed by the same
method, all the plotted results will practically converge to the same curve. From these Iigures with
(e
ps
)
ji
j
i
j
ps) (e

, the Iollowing general equation appears to be valid:

(e
ps
)
ji
/(e
ps
)
j0
(1 z
i
/H)
n
(5)

In the above equation, j is an integer varying Irom i to , z

H, and n is a constant Ior a given method oI

pile construction. This means that (e
ps
)
ji
is the total shear displacement between depth z
i
and depth z

.
Accordingly, (e
ps
)
j0
is the total shear displacement between depth z
0
and depth z

(i.e. total shear
displacement at pile top).
(3) For CFA piles in Figure 3a, n 2.5,

i.e. (e
ps
)
ji
/(e
ps
)
j0
(1z
i
/H)
2.5
. (5a)

For bored piles supported by bentonite in Figure 3b, n1.5,

i.e. (e
ps
)
ji
/(e
ps
)
j0
(1z
i
/H)
1.5
.

(5b)

For mini-pile with shaIt grouted in Figure 3c, n 4,

i.e. (e
ps
)
ji
/(e
ps
)
j0
(1z
i
/H)
4.0
.

(5c)

(4) In all cases, the decrease oI (e
ps
)
ji
/(e
ps
)
j0
with z
i
/H is Iairly rapid.

5 PROPOSED DESIGN METHOD

It has been well established that SPT-N value is a measure oI the available shearing resistance in soil.
ReIerring to basic soil stress-strain principles, this corresponds approximately to Point M oI the stress-strain
curve as in Figure 1b. However, in dealing with Iriction pile design in soil, the Iollowing have not been well
established:

(1) the proportion oI the shearing resistance that can be mobilized at various depths: Theses correspond to any
points along the stress-strain curve OLAM in Figure 1b. From Eq (5), the Iollowing general equation
appears to be valid Ior a pile at depth, z
i
with total depth H:

(N
i
)
mobilized
(N
i
)
measured
(e
ps
)
ji
/(e
ps
)
j0
(6a)

where (N
i
)
mobilized
proportion oI shearing resistance that can be mobilized at depth z
i
, and (N
i
)
measured
SPT-
N value measured at depth z
i
.

Substituting Eq (5) into Eq (6a), the Iollowing equation can be obtained:

(N
i
)
moblized
(N
i
)
measured
(1z
i
/H)
n
(6b)

(2) the relationship between shearing strength and the N
mobilized
value: Denote this ratio by the parameter .

158
The HKIE Geotechnical Division Annual Seminar 2014
With such results, Ior a pile with total depth oI H having its SPT-N value measured at various depths z
i
the
maximum load P
max
that can be developed will be given by the Iollowing equation:

P
max

n
0 i
(N
i
)
measured
(1 z
i
/H)
n
(z
i1
- z
i
)D (7)

6 CONCLUSIONS

With the N-value measured at various depths at or near the proposed pile location, the maximum pile capacity
P
max
can be estimated Irom Eq (7), provided that the two newly introduced parameters, namely n and , are
known. Both n and have been Iound in this paper to be essentially the same Ior piles with the same
construction method. To determine n, reIerence can be made to Figures 3a, b & c. The n value that will yield
the best Iitted curve (1z
i
/H)
n
to the test points is the required n. With n thus determined and the axial pile
load at various depths measured, can be evaluated Irom Eq (7). In case that no loading test results are
available, may be obtained Irom Table 1 Ior the corresponding construction methods. In either case in order
to veriIy the design assumptions and parameters, loading test on a trial pile is recommended prior to the
installation works. The purposes oI the loading test on a trial pile are to establish and/or veriIy installation
means and methods oI construction, and to veriIy the design parameters and hence the load carrying capacity
oI the pile. One vital point is to note that the present proposed design method is based on the average SPT-N
values oI each site. It Iollows that its variation Ior site with particular soil type, stress history, pile size and
pile construction method cannot be Iully diIIerentiated. More instrumented pile load tests are also
recommended to be carried out with well deIined drillhole with SPT-N values at the proposed pile location to
veriIy the proposed design method.

ACKNOWLEDGEMENTS

The authors would like to record their thanks to the Director oI Architectural Services, Hong Kong SAR
Government Ior his kind permission in publishing this paper.

REFERENCES

Brown, D.A., Dapp, S.D., Thompson, W.R. and Lazarte, C.A. 2007. Design and Construction of Continuous Flight
Auger (CFA) Piles. Federal Highway Administration, Washington DC.
Frizzi, R.P. and Meyer, M.E. 2000. Augercast Piles: South Florida Experience. In R.Y.S. Pak & J. Yamamura (ed),
Proceedings of Sessions of Geo-Denver 2000, 5-8 August 2000, Denver, Colorado. American Society oI Civil
Engineers, Reston, VA, 382-96.
Kan, C.Y., Wong, C.T. and Leung, M.K. 2013. Relationship oI Iriction and shear displacement along the depth oI shaIt-
grouted mini-piles. HKIE Transactions, 20(4): 206-213.
Kulhawy, F.H. and Chen, J.R. 2007. Discussion oI Drilled ShaIt Side Friction in Gravelly Soils. Journal of Geotechnical
and Geoenvironmental Engineering, 133(10): 1325-1328.
Kulhawy, F.H. and Mayne, P.W. 1990. Manual on Estimating Soil Properties for Foundation Design. Electric Power
Research Institute, Palo Alto, CA.
Misra, A., Chen, C.H., Oberoi, R. and Kleiber, A. 2004. SimpliIied Analysis Method Ior Micropile Pullout Behaviour.
Journal of Geotechnical and Geoenvironmental Engineering, 130(10): 1024-1033.
O`Neill, M.W. 2001. Side Resistance in Piles and Drilled ShaIts. Journal of Geotechnical and Geoenvironmental
Engineering, 127(1): 316.
Rollins, K.M., Clayton, R.J., Mikesell, R.C. and Blaise, B.C. 2005. Drilled ShaIt Side Friction in Gravelly Soils. Journal
of Geotechnical and Geoenvironmental Engineering, 131(8): 9871003.
Taylor, D.W. 1948. Fundamentals of Soil Mechanics. John Wiley and Sons, New York.
Terzaghi, K. and Peck, R. B. 1967. Soil Mechanics in Engineering Practice, 2
nd
edition. John Wiley and Sons, New York.
Tomlinson, M.J. 1971. Some EIIects oI Pile Driving on Skin Friction. In Proceedings of Conference on Behaviour of Piles,
107114.
Wong, H.Y. 2003. Design and construction oI soil Iriction bored piles in Hong Kong, with particular reIerence to marble
areas. Presented at the Hong Kong Institution of Engineers Annual Seminar on Case Histories in Geotechnical
Engineering in Hong Kong, 9 Mav 2003, Hong Kong.
159