Civil & Resource Engineering Mark Randolph

Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Offshore Foundation Systems 406

The University of Western Australia

AXIAL PILE RESPONSE

Mark Randolph
Offshore Foundation Systems 406

Axial Pile Response Page 1

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Outline

• Load transfer analysis

• Axial loading
– Load transfer functions
– Strain softening & cyclic loading

Offshore Foundation Systems 406

The previous lectures on piled foundations covered different forms of offshore pile
construction (driven, drilled and grouted, grouted driven), and the main methods of
estimating axial capacity. Offshore design of piles is dominated by design criteria
in terms of capacity, with settlement rarely being critical (apart from estimation of
foundation stiffness, to enable assessment of the dynamic response of the structure).
However, it is still necessary to consider the load-displacement response of pile
foundations, in order to assess the potential for degradation of shaft friction due to
progressive failure or cyclic loading.

Axial Pile Response Page 2

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Structural Model of Piles

Steel jacket structure

Piles, with soil

replaced by axial and
lateral springs

Offshore Foundation Systems 406

Load transfer analysis simulates the pile-soil interaction by means of a series of

springs distributed down the length of the pile (and at the tip). At any location, a
pair of springs would represent the axial and lateral responses. The simplest form of
such springs would be elastic, or elastic, perfectly plastic (i.e. with a limiting
maximum value of local shear stress, or lateral pressure). However, it is customary
to make use of non-linear springs, with gradual reduction of secant stiffness as the
load level increases, and a potential for strain-softening beyond the peak load
transfer.
One of the advantages of the load transfer approach is that the complete jacket and
pile foundations may be analyses in a standard structural analysis package, replacing
the soil with the non-linear load transfer springs. A limitation of the approach is the
loss of direct modelling of interaction between adjacent piles (through the soil
continuum) and a tendency towards empiricism in the choice of load transfer
parameters, divorced from the actual continuum behaviour of the soil.

Axial Pile Response Page 3

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Load Transfer Analysis

d Shear stress, τ
Axial Shaft
spring friction
Elastic range: w ≈ 2 τ
d G

Lateral spring Displacement, w/d

Element of pile
Force per unit length, P
Ultimate
force/unit
length y P / d p net
Elastic range: ≈ =
d 4G 4G

Displacement, y

Offshore Foundation Systems 406

The schematic shows the main features of load transfer springs, and also indicates
the relationships between the initial spring stiffness and the shear modulus, G, of the
soil continuum. It is important that the gradient of the load transfer curve is
expressed in units of modulus (e.g. stress divided by displacement/diameter; or
force per unit length of pile divided by displacement), in order to avoid problems of
different scales (or size of pile).
In particular, for laterally loaded piles, it is common to refer to the horizontal load
transfer as P (load per unit length), rather than an average (net) pressure, p, given by
p = P/d where d is the pile diameter.

Axial Pile Response Page 4

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Governing Equations for Pile

F
w dF dw F d2w πd
dz ≈ − πdτ ; =− ; = τ
τ dz dz (EA )p dz 2 (EA )p

Axial loading

M
H dH dM d2y M
y ≈ −P ; = H; =
dz dz dz 2 (EI )p
dz P
d4y P
Hence =−
dz 4 (EI )p

Lateral loading

Offshore Foundation Systems 406

Actual modelling of the pile response is normally achieved through a computer code
that implements the beam-column equations for axial or lateral loading. The
governing differential equations are standard for structural engineering, and a
variety of codes are available in the industry.

Axial Pile Response Page 5

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Single Pile Stiffness

P (w) Shear Modulus
Gavg Gl Gb
G avg Gl
ρ = ξ=
d Gl Gb

l l/2 db Ep
η = λ=
d Gl
db l For solid cylindrical pile :

Ep =
(EA )steel
Depth πd 2 / 4

2η 2π tanh µl l
++ ρ
P (1 − ν )ξ ζ µl d
=
G l dw 1 8 η tanh µl l
1+
πλ (1 − ν ) ξ µl d
Offshore Foundation Systems 406

The final solution for the elastic response of a compressible pile is given above.
This solution may be incorporated into spreadsheet calculations of pile response
under axial load.
The main load transfer parameter, ζ, is written as:
ζ = ln(rm/ro)
where rm (the maximum radius of influence of the pile) is expressed as
rm = {0.25 + [2.5(1 - ν) ρ - 0.25](Gl/Gb)}l
or, for floating piles (Gl/Gb = 1)
rm = 2.5(1 - ν) ρ l
The numerator of the expression for the pile stiffness is made up of the base
stiffness (first term) and shaft stiffness (second term). The second term in the
denominator is generally small (except for very compressible piles) and may often
be ignored.
The compressibility of the pile is controlled by the tanµl/µl term, where the product
µl is given by
µl = (8/ζλ)0.5 l/d
When µl is greater than about 2, the tanhµl terms becomes close to unity and the
whole expression may be simplified (see next page). At the other extreme, for stiff
piles tanµl/µl becomes unity (as µl approaches zero) and the expression reverts to
that given previously for a rigid pile.

Axial Pile Response Page 6

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Design Chart for Pile Stiffness

40
P 3000
G d
Glwd 30 ξ= l =1 η= b =1
Gb d
1000
G avg
20 ρ= = 0.75
Gl
300
Ep
10 100 λ=
30 Gl
10
0
1 10 100
Pile slenderness ratio, l/d

Offshore Foundation Systems 406

The design chart provides a simple way of estimating the non-dimensional pile
stiffness for a range of stiffness ratios, λ = Ep/Gl, and slenderness ratios, l/d. The
typical range of stiffness ratio, λ, is between 100 and 1000, although lower values
are appropriate for rock sockets.
The division of piles into rigid, intermediate, and compressible, is achieved through
the parameter, µl, which is proportional to the slenderness ratio, l/d, divided by the
square root of the stiffness ratio, λ. Thus, for piles where
l/d < 0.25λ0.5
the pile may be considered as essentially rigid.
On the other hand, for piles where
l/d > 1.5λ0.5
the pile is very compressible and the response becomes independent of the overall
pile length. For this case, the pile head stiffness may be estimated from:
P 1
≈ πρd E p G l ≈ ρd E p G l
w 2ς
For consistency, the shear modulus, Gl, should be replaced by the shear modulus at
a depth corresponding to z = 1.5dλ0.5, which is essentially the depth limit to which
any significant load transfer occurs.

Axial Pile Response Page 7

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Load Transfer for Group Piles

τd
Shear stress, τ w=ς
2G
Single pile: ς ≈ 4
n
Group of n piles: ς* = nς − ∑ ln( 2si / d ) )
Displacement, w 2

1 − ν Pb
wb = ξ
Base load, Pb 2 db Gb
Single pile: ξ = 1
 1 n d 
Group of n piles: ξ* = ξ 1 + ∑ b 
 π 2 s i 
Displacement, wb
Offshore Foundation Systems 406

The same solution for a single pile can be used to estimate the stiffness of a group of
piles, making use of modified load transfer parameters that make allowance for
interaction between neighbouring piles. Interactive effects can lead to a reduction in
effective stiffness of each pile by a factor of 2 or 3.

Axial Pile Response Page 8

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

API Design Guidelines

w/d τ/τs 1

Normalised shear stress

0.8
0.0016 0.30
0.0031 0.50 Clay 0.6
RATZ
0.0057 0.75 0.4 API (0.7)
0.0080 0.90 API (0.9)
0.2
0.010 1.00
0.02 0.70 to 0.90 0
0 2 4 6 8 10
0.1 0.70 to 0.90 Displacement/diameter (%)

w τ/τs

0.025 m 1 Sand

Offshore Foundation Systems 406

The American Petroleum Institute (API) guidelines suggest a default load transfer
curve for axial analysis. This curve gives an approximately parabolic shaped curve,
reaching peak shear stress at a displacement of 1 % of the pile diameter. Beyond
the peak, a degree of strain softening is assumed, with the shear transfer reducing by
up to 30 % by a displacement of 2 % of the pile diameter.
In practice, strain softening for large diameter piles could continue over
significantly larger displacements, with much lower residual values of load transfer
(particularly under cyclic loading).

Axial Pile Response Page 9

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Progressive Failure of Long Piles

Displacement Qactual = RfQideal
τ
A

τ
B

C
w

Offshore Foundation Systems 406

Strain softening effects can lead to progressive failure of long, slender piles, such
that the actual capacity is less than the ideal (rigid pile) capacity. A simple design
chart may be constructed, showing the potential reduction in capacity as a function
of the relative compressibility of the pile. The non-dimensional factor, K, is
essentially the ratio of elastic shortening of the pile under the maximum shaft
capacity, to the displacement needed to strain soften between peak and residual.

Axial Pile Response Page 10

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

One-way vs. two way cycling

Offshore Foundation Systems 406

Cyclic loading can lead to significant degradation of shaft friction. At the element
level (modelling one particular point down the pile), 1-way cyclic loading will tend
to lead to increasing average displacements, followed by failure at a value of shaft
friction that is lower than obtained under monotonic loading.
Under 2-way cyclic loading, the average displacement may not change much, but
the cyclic component will increase, followed by failure at a reduced level of shear
stress.

Axial Pile Response Page 11

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Shaft capacity after 2-way cycling

2-way cycling prior to

static loading to failure

Offshore Foundation Systems 406

Axial Pile Response Page 12

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Cyclic Response of Calcarenite

200
Shear stress Monotonic test
(kPa) 150

100

Cyclic test failed

50 after 6451 cycles

0
0 5 10 15 20
Displacement (mm)
Offshore Foundation Systems 406

A 1-way cyclic loading test on a model pile is shown here, and compared with the
response under monotonic loading. Failure occurs once the cumulative
displacement under cyclic loading reaches the monotonic loading envelope.

Axial Pile Response Page 13

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

RATZ Simulation of Cyclic Loading

100
Monotonic response
75
Relative shear
stress (%) 50

25 2-way cyclic
1-way cyclic loading
loading 0
0 0.01 0.02 0.03 0.04 0.05 0.06
-25

-50
Displacement/Diameter

Offshore Foundation Systems 406

The load transfer code, RATZ, has in-built algorithms to model degradation under
cyclic loading. The plastic displacement undergone within each cycle is treated as
similar to additional monotonic displacement, with resulting degradation of the
available shaft friction.

Axial Pile Response Page 14

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Cyclic stability diagram

Offshore Foundation Systems 406

Axial Pile Response Page 15

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Interaction Diagram

Normalised 1
cyclic shear
stress 0.8
Gerber Monotonic failure
0.6 (parabolic)

0.4

0.2 Goodman
(linear)
0
0 0.2 0.4 0.6 0.8 1
Normalised mean shear stress

Offshore Foundation Systems 406

A stability region may be defined through an interaction diagram, showing regions

where failure would eventually take place under cyclic loading. The diagram may
be made more sophisticated by the inclusion of contours showing numbers of cycles
to cause failure.

Axial Pile Response Page 16

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Rod Shear Test: Typical Response

500

Shear stress 400 Nominal 20 mm diameter grouted rod

(kPa) in 100 mm diameter triaxial sample
300

200

100

0
0 5 10 15 20
-100
Displacement (mm)
Offshore Foundation Systems 406

This shows a typical 'element' response, obtained from a test on a model grouted
pile installed in a sample of calcarenite (so-called 'rod shear test'). The large
amount of strain softening is noteworthy, as is the very low cyclic resistance.

Axial Pile Response Page 17

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Field Tests on Grouted Piles:

Overland Corner
• Woodside Offshore Petroleum GSTs (1987-8)
– Drilled & grouted construction
– Grouted Sections (2 - 5 m long)
– Diameters from 0.4 m to 2 m
– Scaled insert, 0.4 m diameter by 15 m long
• Esso Australia GDPs (1988)
– Grouted driven construction
– Piles 0.4 m to 0.95 m diameter by 10 m long
– Tested before and after grouting
Offshore Foundation Systems 406

Two key sets of field tests were undertaken on grouted piles, in order to establish
design rules for axial response, and to provide data for calibrating load transfer
codes such as RATZ.

Axial Pile Response Page 18

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Grouted Pile Tests at

Overland Corner
30
0.98 x 5.1 m

35 0.44 x 2.4 m

40
Depth 2.08 x 5.1 m 0.95 x 10.0 m
(m) 0.41 x 10.0 m
45
0.44 x 15.6 m
Grouted Driven
Grouted Section Tests Pile Tests
50

Offshore Foundation Systems 406

A total of 6 drilled and grouted pile tests were undertaken, 5 on short sections of
different diameter, and 1 on a scaled 'long' pile representative of a true prototype. In
the separate study by Esso Australia, 3 grouted driven piles were tested.
The zone of testing covered a weak limestone with strength and deformation
characteristics similar to those on the North-West Shelf.

Axial Pile Response Page 19

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Key Geotechnical Parameters

for Overland Corner Limestone
• Cone resistance, qc 10 - 18 MPa
(average: 14 MPa)
• Grouted anchors, τp 526 - 614 kPa
(residual: 25 - 40 % peak
• Hydraulic fracture, pi 1600 - 2300 kPa
(overburden: 500 - 800 kPa)
• Pressuremeter, Gur > 350 MPa (compliance limited)
(su: 600 - 700 kPa)
• CNS tests, τp 160 - 320 kPa
(residual values: 50 - 100 kPa)
• Rod shear tests, τp 300 - 600 kPa
(residual values: 100 - 300 kPa)
Offshore Foundation Systems 406

An extensive site investigation was carried out in order to characterise the site, with
similar types of testing to an offshore site.

Axial Pile Response Page 20

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Diameter Effect: Grouted Section Tests

500
0.44 m diameter
Average 400
shear stress 0.98 m diameter
(kPa) 300
2.08 m diameter
200

100

0
0 0.2 0.4 0.6 0.8 1
Displacement/diameter (%)
Offshore Foundation Systems 406

A consistent trend of decreasing peak shaft friction with increasing diameter of pile
was noted. This arises due to the decreasing effect of dilation at larger diameters, as
the normal stress change around the pile is proportion to the amount of radial
movement as a ratio of the pile diameter. For a given degree of roughness, the
radial movement would be similar for each pile size, and hence the ratio would
decrease as the diameter increased.

Axial Pile Response Page 21

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Modelling 0.44 m Diameter

Grouted Section Test
2

1.5 Field data

Load
(MN) 1
RATZ simulation
0.5

0
0 50 100 150 200
-0.5
Displacement (mm)
Offshore Foundation Systems 406

The monotonic tests on the short elements were used to calibrate the RATZ model
for application to the long (model) pile, which was subjected to cyclic loading.
Note the simulation of the very low cyclic shear stress.

Axial Pile Response Page 22

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Modelling 0.44 m Diameter

Grouted Pile (15 m Long)
7
6
Load
(MN) 5
4
3 RATZ
Field data
2 simulation

1
0
0 10 20 30 40 50 60 70
Displacement (mm)
Offshore Foundation Systems 406

Although the monotonic capacity should have been in excess of 11 MN (scaled

from the shorter elements), cyclic loading led to failure for maximum loads of less
than 7 MN. This test emphasises the potential for cyclic degradation of pile
capacity in calcareous sediments.

Axial Pile Response Page 23

Civil & Resource Engineering Mark Randolph
Offshore Foundation Systems 406 Centre for Offshore Foundation Systems

Measured and Computed Load

Profiles for 15 m Grouted Pile
Load (MN)
0 2 4 6 8
30 Stage 1 Stage 2 Stage 3

35
Depth
(m)
40
RATZ
simulation
45 Field
data
50

Offshore Foundation Systems 406

The profiles of load down the pile shows the gradual transfer of load to the lower
part of the pile, and also a reduction in shear transfer in the upper elements.

Axial Pile Response Page 24