Understanding Shaft Resistance

in Rock Socketed Piles

Dr Chris Haberfield
Lecture 2 Outline

• Origin of the Scatter (Load test results)

• Back to Basics
• Some Research Findings
• Laboratory Testing of Interfaces
• Roughness
• From Laboratory to Field
• Explaining the Scatter
• Summary
Acknowledgements

• Dr Julian Seidel
• Foundation QA for use of Rocket
• Researchers at Monash University
Shaft Resistance
1
aAdhesion Factor ,

2 3
3
2
2
0.1 2

LEGEND:
Mudstone,shale
Sandstone
What is the origin of the scatter ?
Indicates test not to failure
0.01
0.1 1 10 100
Unconfined compressive strength, qu (MPa)
Parameters affecting Shaft Resistance

Rock
• type, structure, weathering
• strength
• stiffness
Construction
• socket diameter
• socket roughness
• socket cleanliness
• concrete pour
• contractor experience and expertise
 Origin of the Scatter

• socket roughness 1000

• (drilling
pile diameter
Annotations denote mean absolute
2000
1400 asperity angle in degrees
1400
tools, artificial roughening) Annotations denote bentonite thickness
800

(kPa)
Increasing smear thickness

(kPa)
Increasing clean socket

resistance (kPa)
1200
1200
15.0 17.5
1200 Increasing
(structural considerations) normal stress
roughness clean socket

(kPa)
1500 2 mm
600 Increasing diameter
1000

resistance
Shaft resistance
12.5
1000
800 900 Increasing filter cake
10.0 thickness
4 mm
Are normal
• initial their models
stresswe can use to800
predict these effects
0.35 ?

resistance
7.5
800 400 •62mm
mm
1000 600

Shaft resistance (kPa)

5.0
600
600
600
(concrete placement) 4 mm

What engineering properties are required ?

Shaft
0.6

Shaft
200 300 8 mm

• rock smear 400 2.5

400
500 0.96 mm

Shaft
100
400 10 mm
200
2000 30 1.58 mm
(socket cleaning & bonding) Annotationsdenote
Annotations denotesmear
initialthickness
normal stress
(mm) in
2.0kPa
10 mm
0200
00 0 5 10 15

• residual drilling fluids 000 55

Displacement
Annotations
10 10
(mm)
denote diameter
10 15
of shaft in (m)
1515
(socket cleaning & bonding) 0 Shaft displacement
Displacement
Displacement (mm)
(mm)
(mm)
0 5 10 15
Increasing roughness Displacement (mm)
Shaft Resistance - Back to basics
• Shaft resistance is developed
through friction (t) from intimate
contact between the concrete of
the pile and the rock
Pressure friction
• The wet concrete applies a
from fluid
pressure (sn) to the socket wall concrete
which is locked in when the
concrete hardens
• frictional resistance (+ adhesion)
t  c  sn tan 
must be overcome before slip at
the pile/rock interface can occur
Shaft Resistance - Back to basics

• The rock socket is not

smooth, but has
undulations referred to
as roughness
• For slip to occur at the
interface, the socket
must dilate
• Thus increasing the
normal stress on the
interface and the
frictional resistance
Socket Dilation
Vertical Socket dia.
displacement D + DD
Pile and of pile
Pile dia. D
socket dia.D
. .
. . .
. . .
.. . .. . .
..
.
. .. . . . . . .
... Rough wall . . .. .
.
rock socket
. Increased
Normal Shaft section .
.. . . . .
force of concrete Normal ... Normal
pile . . force force
. .
.. . .. .
. Shear
. . Shear .
force .
.
. . force
. . .. .
.

(a) Pile before displacement (b) Pile after displacement

 Constant Normal Stiffness (CNS)

Increase in normal stress

Em Dr Dr
Dr
Dsn = r r
(1+ nm ) r
Em = rock mass Young’s modulus
nm = rock mass Poisson’s ratio
r = D/2 = radius of socket
Dr = dilation of socket Dsn Em
Dsn = change in normal stress K= =
Dr (1+ nm).r
K = normal stiffness
Modelling socket shear in the laboratory
CNS Direct Shear Tests
Spring stiffness, K Increased
Normal force Normal force
Dilation
.
. . .
.
. .
. . . .
. . Shear force
.
Concrete
. . .
.
Concrete . .
. .
. .
. . . . .
. .
.
.

Shear force Rock

Rock

Normal force Increased

Shear Displacement Normal force

(c) Equivalent 2-D model (d) Equivalent 2-D model

for before displacement for after displacement
Laboratory Testing : CNS Rig

• Computer feedback control

• 250 kN hydraulic actuators

for shear and normal stress

• Monotonic and cyclic loading

• Stress or strain control

• Automatic data logging

 CNS Direct Shear Testing Rig

Actuator Shaft
Load Cell
Transfer Plate
Bearing Plate
Needle Rollers Vertical Piston

Rigid Frame

Rock-concrete interface
Split Shear Box

Horizontal Carriage Cam Followers

Rough interface sample prior testing

Concrete
S

Mudstone
CNS test results : Impact of roughness, normal stress and stiffness
Class A
Class C

1000
Class A (300/300)
Class A (300/900)
800 Class A (900/300)
Class C (300/300)
Shear stress, t (kPa)

Class C (600/600)
Class C (150/150)
600

400
Increasing roughness increases strength and stiffness of the interface response
Increasing stiffness
200 increases the strength and stiffness of the interface response
Increasing initial normal stress increases the strength and stiffness of the
0
interface response
0 10 20 30 40 50
Shear displacement, x (mm)
CNS Test Samples - Triangular Asperities

5 deg. x 3.75 mm high regular triangles

10 deg. x 7.5 mm high regular triangles

12.5 deg. x 6 mm high regular triangles

15 deg. x 7.5 mm high regular triangles

17.5 deg. x 9.5 mm high regular triangles

22.5 deg. x 9.5 mm high regular triangles

27.5 deg. x 11.5 mm high regular triangles

 Typical Test Results - Triangular Asperities
4.0

Shear stress ( kPa )

1200 Asperity failure 1200
Shear stress ( kPa )

Dilation ( mm )
3.0
Asperity sliding
800 800 2.0

1.0
400 400
0
Post peak sliding
-1.0
0 10 20 30 40 0 400 800 1200 0 10 20 30 40
Shear displacement (mm) Normal stress (kPa) Shear displacement (mm)

TEST No. M1b 16/4/1992

Normal stress ( kPa )
Normal stress ( kPa )

1200 1200 Johnstone sample 6C

12.5 + 0
Bandsaw cut / gladwrap o'nite
800 800

400 400
Initial nor. stress = 330 kPa
Normal stiffness = 321kPa/mm
Shear rate = 0.5 mm/min
0 10 20 30 40 -1.0 0 1.0 2.0 3.0 4.0
Shear displacement (mm) Dilation (mm)
 Summary ‘A’ for Triangular Asperities
304 kPa + 300 kPa/mm 330 kPa + 321 kPa/mm 292 kPa + 300 kPa/mm
Shear stress ( kPa )

1200 1200 1200

Shear stress ( kPa )

Shear stress ( kPa )
10o 12.5o 15o
800 800 800

400 400 400

+i

0 0 0
400 800 1200 400 800 1200 400 800 1200
Normal stress (kPa) Normal stress (kPa) Normal stress (kPa)

288 kPa + 300 kPa/mm 334 kPa + 285 kPa/mm 288 kPa + 300 kPa/mm
Shear stress ( kPa )

Shear stress ( kPa )

Shear stress ( kPa )

1200 1200 1200
17.5o 22.5o 27.5o
800 800 800

400 400 400

0 0 0
400 800 1200 400 800 1200 400 800 1200
Normal stress (kPa) Normal stress (kPa) Normal stress (kPa)
 Summary ‘B’ for Triangular Asperities
304 kPa + 300 kPa/mm 330 kPa + 321 kPa/mm 292 kPa + 300 kPa/mm

Shear stress ( kPa )

Shear stress ( kPa )

Shear stress ( kPa )

1200 1200 1200

1200

10o 12.5o 15o

800 800 800
800

400 400 400

400

0 0 0
10 20 30 40 10 20 30 40 10 20 30 40
Shear displacement (mm) Shear displacement (mm) Shear displacement (mm)

288 kPa + 300 kPa/mm 334 kPa + 285 kPa/mm 288 kPa + 300 kPa/mm
Shear stress ( kPa )
Shear stress ( kPa )

Shear stress ( kPa )

1200 1200 1200

27.5o
800 800 800
22.5o
17.5o
400 400 400

0 0 0
10 20 30 40 10 20 30 40 10 20 30 40
Shear displacement (mm) Shear displacement (mm) Shear displacement (mm)
 Simple Sliding Model

Shear Stress, t ( kPa )

1200
t  s n tan(  sl + i )

800
sl + i

sn 400

0 400 800 1200

i Normal Stress, sn (kPa)
Video image of regular 12.5o siltstone
asperity at failure

Concrete

toe of failure Rock rotational

surface failure

At a critical shear displacement, the asperity can

no longer support the applied load and fails.
Effective Roughness

If asperities are too steep, there will be no sliding and no

dilation. As a result, the interface may have lower
shear strength and will behave in a more brittle manner.
There is an optimal level of roughness, beyond which
no improvement to shear performance will occur.

“Grooving”
 Deformation and dilation
Asperities deform under load and reduces dilation
(to less than the asperity angle)

N
de
S dy
.
i n
dx
 Behaviour after Failure
• After failure of the asperity, there is a wedge of compressed
rubble which effectively acts as a door-stopper

.
.
. Concrete.
.
.
dx (1- e).dx
.
. a wedge of compressed
rock

e dx
Unfailed rock b

Relative movement occurs between both the concrete and the wedge
and the wedge and unfailed material. This results in a residual
strength greater than the residual strength of the rock.
5o asperity profile - measured vs predicted
Shear and Normal Stress ( kPa )
2000
2.00
Normal
1500

Dilation ( mm )
1.00

1000

Shear
0.00
500
Test MJR_5_1 : 5 x 3.8mm high :
s no = 588 kPa; K = 600 kPa/mm
0
0 10 20 30 40 50

Shear displacement (mm)

10o asperity profile - measured vs predicted

1500 4.00
Shear and Normal Stress ( kPa )

1200 Normal 3.00

Dilation ( mm )
900 2.00

Shear
600 1.00

300 0.00
Test MJR_10_1 : 10 x 7.5mm high :
s no = 304 kPa; K = 300 kPa/mm
0 -1.00
0 10 20 30 40
Shear displacement (mm)
15o asperity - measured vs predicted

Shear and Normal Stress ( kPa ) 1500 4.00

Normal
1200 3.00

Dilation ( mm )
900 2.00

Shear
600 1.00

300 0.00
Test MJR_15_1 : 15 x 7.5mm high :
s no = 292 kPa; K = 300 kPa/mm
0 -1.00
0 10 20 30 40
Shear displacement (mm)
22.5o asperity - measured vs predicted

Shear and Normal Stress ( kPa ) 1200 7.00

900 5.00

Dilation ( mm )
Normal
600 3.00

Shear
300 1.00
Test MJR_22_3 : 22.5 x 9.5mm high :
s no = 150 kPa; K = 150 kPa/mm
0 -1.00
0 10 20 30 40

Shear displacement (mm)

Extension to rough profiles

Class A

Class B

Class C

Class D

Class E

0 100 200 300 400 500

Horizontal Dimension (mm)
 Same basic behaviour but more complex !

Regular triangular asperity profiles

The conditions at every asperity are the same

Rough asperity profiles

The conditions at every asperity are different

Asperity Deformation and Load Sharing

si

chord j chord i
D
Elastic Modulus, E
Laboratory Validation : fractal profiles
Class A Profile - measured vs predicted
Shear and Normal Stress ( kPa )
1200 1.00
Test MJF_A1C2 : Profile A1 coarse :
s no = 300 kPa ; K = 900 kPa/mm
900 0.67

Dilation ( mm )
Normal 0.33
600

300 0.00
Shear

0 -0.33
0 5 10 15 20
Shear displacement (mm)
Class C Profile - measured vs predicted

Shear and Normal Stress ( kPa ) 1200 7.00

Test MJF_C1C5 : Profile C1 coarse :
s no = 150 kPa ; K = 150 kPa/mm
900 5.00

Dilation ( mm )
600 3.00

Normal

300 1.00
Shear

0 -1.00
0 5 10 15 20
Shear displacement (mm)
Class C Profile - measured vs predicted

Shear and Normal Stress ( kPa ) 1200 1.00

Normal
900 0.67

Dilation ( mm )
600 0.33
Shear

300 0.00
Test MJF_C1C2 : Profile C1 coarse :
s no = 300 kPa ; K = 900 kPa/mm
0 -0.33
0 5 10 15 20
Shear displacement (mm)
Understanding Roughness
2000
1200 10 degree
12.5 degree
Class15
A degree
1200 Triangular profile Class17.5
B degree
1500 Class C
22.5 degree
Class D

• Impact is complex 900 Class27.5

E degree

Shear stress (kPa)

stress (kPa)
900
• Shape dependent 1000
600

(kPa)
• Optimal roughness500 is not to causeIrregular profile
600

tShear
interlock but to generate maximum 300

300
dilation 0
0 10 Increasing
20 30 40 50
roughness
• Scale dependent 0 Shear displacement (mm)
0
0 10 20 30 40

As asperity
0 angle
10increases
Shear- displacement
20 strength
30 and
40 stiffness
(mm) 50 increases
x (mm)
As roughness increases - strength and stiffness increase
Scale effects

Standard deviation of angle, Sl,q

l = 40 mm, S40,q = 3o

l = 10 mm, S10,q = 5.5o

 Roughness Parameter vs Scale
If we want to represent real roughness as a set of statistics - e.g.
standard deviation of asperity angle (or height), what length scale
(chord length) is appropriate?
Standard deviation of angle (deg.)

15

10

0
0 10 20 30 40 50 60

Chord Length, l (mm)

Answer : All scales, but …

in practical terms : it depends on the scale at which performance
(often displacement) is being considered.
Some CNS direct shear test results
800

Shear Stress ( kPa )

Class C - fine (512 x 1mm)
Class C - medium (128 x 4mm)
Class C - coarse (32 x 16mm)
600

Coarse, medium and fine

400
approximations of the same profile

200
Fine Approximation
0
0.0 2.5 5.0 7.5 10.0
Medium Approximation Shear Displacement (mm)

• Stiffness is systematically
Coarse Approximation higher for finer profiles
• Peak shear strength does
vary systematically
0 100 200 300 400 500
Horizontal Dimension (mm)
Some more CNS direct shear test results
800

Shear Stress ( kPa )

Class D
Profiles with different geometry but 600
similar statistics also perform in an
essentially similar manner. Class B
400

200 Profile 1
Profile 1 Profile 2
Profile 3
0
0.0 2.5 5.0 7.5 10.0
Profile 2 Shear Displacement (mm)

Profile 3

0 100 200 300 400 500

Horizontal Dimension (mm)
Impact of Roughness - Summary

• All scales of roughness important

• Small scale roughness impacts on initial
stiffness
• Large scale roughness impacts on peak
shear strength
• “Grooving” may not be advantageous
 From laboratory test to rock socket
What are differences in roughness?
Patch contact
and loading
Strip contact
and loading

3D socket roughness

Rock socket roughness is

3D not the 2D approximation
used in CNS laboratory
testing 2D CNS roughness
.
 From laboratory test to rock socket

Displacements and load sharing between

asperities in laboratory sample and field
socket are different
Take care when extrapolating CNS laboratory test
results directly to field socket behaviour

Plane-strain : extensive but moderate Patch-loading : intensive but localized

deformation effect about loaded strip effect about loaded patch both along
and along axis .of sample. Thickness pile axis and around circumference.
of sample is finite. Thickness is infinite.
 ROCKET input parameters
• Shear strength parameters - c’ and ’
- drained triaxial tests
- UCS and Hoek Brown
• Sliding friction angle - direct shear tests.
• Rock mass modulus and Poisson’s ratio
- pressuremeter tests.
- triaxial tests, correction for jointing ?
- moisture content correlations, correction for jointing ?
• Socket diameter (structural strength requirements)
• Socket roughness
- direct measurements
- back calculated from load tests
• Initial normal stress - estimated
 Field Validation
800

600
Pile head stress ( kPa )

400

Middleborough Rd. Pile M2

200

Pile Load Test

"Rocket" prediction

0 10 20 30
Pile head movement (mm)
 Field validation

800 1200
Pile Load Test Westgate Freeway Pile WG303/2
Pile head stress ( kPa )

Pile head stress ( kPa )

"Rocket" prediction
600 900

400 600

200 300
Pile Load Test
Stanley Avenue Pile S5
"Rocket" prediction

0 10 20 30 0 10 20 30
Pile head movement (mm) Pile head movement (mm)
Field validation

16 12

14
10 Bahrain
12
9 m sockets in
Pile Head Load (MN)

Pile Head Load (MN)

8
10
1 to 2 MPa
8 6 calcareous
6
Measured Measured
siltstone
4
ROCKET ROCKET
4

2
2

0 0
0 10 20 30 40 0 10 20 30 40
Pile Head Displacement (mm) Pile Head Displacement (mm)

(a) TP1 (b) TP2B

Field validation

20 30

18

16
25 Bahrain
14
12 m and 15 m
Pile Head Load (MN)

Pile Head Load (MN)

20
12
sockets in 1 to 2
10 15 MPa calcareous
8 Measured
Test 1
Measured siltstone, toe of
10 Test 1
6 Measured Measured 15 m socket in
Test 2 Test 2
4 ROCKET
5
ROCKET strong limestone
2

0 0
0 10 20 30 40 0 10 20 30 40
Pile Head Displacement (mm) Pile Head Displacement (mm)

(e) TP5 (f) TP6

Field measurement of roughness
 Example Profiles

0 Siltstone: Elm Towers Project (Melbourne, Wagstaff Piling)

MW- HW,
50 Augered

100 MW- HW,

Grooved
y (mm)

150
Basalt: Federation Square Project (Melbourne, Wagstaff Piling)

SW, Cored
200

MW- HW,
250 Cored

0 100 200 300 400 500 600 700 800 900 1000

x (mm)
Some results in Siltstone
18.0
Williams (1980) Baycan (1996) - R40 Piles, D = 600mm
16.0 Elm Towers - Effective Roughening Baycan (1996) - R20 Piles, D = 600mm
Mean Roughness Height (mm)

Elm Towers - Ineffective Roughening UB & LB for Effective Roughening

14.0

12.0

10.0

8.0

6.0

4.0

2.0

0.0
0 50 100 150 200 250
Chord Length (mm)
Back-calculated from pile load tests, l = 50 mm
Williams & Ervin (1981), Extremely Jointed Rock.
50

PILES - Shaft resistance fully mobilised

PILES - Full shaft resistance not mobilised
Effective Roughness Height (mm)

40 ANCHORS (D < 450mm)

Sockets known to contain sidewall smear
Seidel et al. (1996) - Proposed roughness bounds
Proposed upper & lower bound roughness guidelines

30

20

10

2 3 4 5 1 2 3 4 5 10 2 3 4 5 100 2 3

Unconfined Compressive Strength (MPa)

Does this explain the empirical load test data ?

1
Effective
upper limit
4
3
Adhesion Factor

0.1 Effective lower limit

4
3

0.01
Piles in Clay (after Kulhawy & Phoon, 1993)
Piles in Rock (after Kulhawy & Phoon, 1993)
4
3
2 3 4 2 3 4 2 3 4 2 3 4 2
0.01 0.1 1 10 100

Unconfined Compressive Strength (MPa)

Revisit : Parameters affecting Shaft Resistance
Affects normal stiffness
and increase in normal Affect strength
and stiffness
Rock stress with dilation

• type, structure, weathering Major impact on interface

• strength behaviour wrt stiffness
and strength of response
• stiffness Affects shear strength
of asperities, load
sharing between
Construction asperities
• socket diameter May impact on
• socket roughness soundness and
integrity
Affects socket dilation, normal of pile
• socket cleanliness Affects
stiffness and increase
interface
ininteraction
normal between
friction
and
• concrete pour stress, asperity deformation, asperities
load sharing
• contractor experience andbetween asperities
expertise
 Summary

• Understanding shaft resistance is of prime importance to

predicting rock socketed pile performance
• Shaft resistance is highly dependent on rock properties, socket
roughness and construction effects. Socket diameter also has
an impact.
• Be aware of differences between CNS laboratory testing
performance and field socket performance
• Sockets should be roughened – “grooving” may not be
advantageous”