Triaxial Testing

Laboratory
Tests
to
Determine
Shear

Strength
of
Soils

Geotechnical
Engineering
II

(ENGI
6723)


Presented
by
Rodney
P.
McAffee,
Ph.D.,
P.Eng.

 Laboratory
Tests
to
Determine
Shear

Strength
of
Soils

•  Lecture
Topics

o Brief
overview
of
direct
shear
test

o Determine
soil
shear
strength
parameters

from
triaxial
tesNng:

•  Unconsolidated
–
Undrained

•  Consolidated
–
Undrained

o Triaxial
test
setup
and
behaviour

o Use
of
results
in
engineering
pracNce

o Examples
of
triaxial
test
results

 Shear
Strength

•  Shear
strength
of
soils
is
required
to
solve
problems

of
stability

o Bearing
capacity,
earth
pressures,
slope
stability,
etc.

•  Shear
strength
is
a
funcNon
of
effecNve
normal
stress

 Laboratory
Tests
to
Determine
Shear

Strength
of
Soils

•  Direct
Shear
Test


•  Triaxial
Tests

o Pore
water
pressure
measurement

o TesNng
under
back
pressure


•  Types
of
Triaxial
Tests

o Unconsolidated
–
Undrained

o Consolidated
–
Undrained

o Consolidated
–
Drained

Direct
Shear
Test

 Direct
Shear
Test

•  Disadvantages

o Failure
plane
is
forced
to
be
horizontal

o Cannot
control
drainage

o Stress
concentraNons
at
the
sample
boundaries

o Uncontrolled
rotaNon
of
principal
planes
and
stresses


•  Advantages

o Test
in
inexpensive

o Fast
and
simple

o Easy
to
prepare
for
cohesionless
samples

 Triaxial
TesNng

•  Casagrande
developed
triaxial
tesNng
in
the
1930s

•  More
complicated
than
direct
shear
tesNng

o But
more
versaNle

•  Drainage
can
be
controlled

•  No
rotaNon
of
principal
stresses

o Some
small
shear
stresses
do
act
on
the
boundaries

•  Stress
concentraNons
are
limited

•  Failure
plane
can
occur
anywhere

•  Stress
paths
to
failure
can
be
controlled

Triaxial
Test
Apparatus
and

assumed
Stress
CondiNons

•  Load
frame
with
stepper
motor
drive

•  Triaxial
cell

•  Data
acquisiton
system

•  Electronic
measurement
transducers

o Load,
displacement,
pressure,
and

volume
change

•  Computer
SoZware
for
triaxial
tesNng

•  Control
panel
and
de‐aired
water

control
system

•  Control
Panel
to
regulate

pressure
and
flows
during

tesNng

 Drainage
Paths
in
Triaxial
TesNng

•  The
3
permissible
drainage
paths
are:


Q-Test
(for “quick” test)

S-test
(for “slow” test)
 Consolidated
–
Undrained
(CU)

Test
Behaviour

•  Sample
is
first
consolidated
under
desired
stresses

•  AZer
consolidaNon
complete,
drainage
valves
closed

•  Typically,
pore
water
pressures
are
measured

o Calculate
total
and
effecNve
stresses

•  Excess
pore
water
pressure
(Δu)
can
either:

o Increase
(+ive):
specimen
contracts
or
consolidates

o Decrease
(‐ive):
specimen
expands
or
swells

•  Axial
stress
increased
incrementally
or
at
constant

rate
of
strain

 Back
Pressure
during
TesNng

•  To
ensure
100%
saturaNon
(necessary
to
give

accurate
pore
water
pressures),
a
back
pressure
is

applied
to
the
pore
water

•  Cell
pressure
also
increased
by
same
amount
to

maintain
the
same
effecNve
consolidaNon
stresses

Typical Stress-Strain
Curves for CU Tests

Principal (effective) stress

ratio is a simple way to
normalize the stress
behaviour with respect
to σ’3 during the test
•  With
pore
water
pressures
measured,
we
can
calculate
both

total
and
effecNve
stresses
at
failure

•  Typically,
a
number
of
tests
over
a
range
of
stresses
is

carried
out

•  NC
clay
specimens
develop
posiNve
pore
water
pressures

(σ’= σ – Δu)

•  OC
clays
tend
to
expand
during
shear
causing
decreased
pore

water
pressures

(σ’= σ –(-Δu))

•  TesNng
on
saturated

sand
with
measurement

of
pore
water
pressure

•  Mohr
envelopes
in
terms
of
both
total
and
effecNve

stresses
for
consolidated
undrained
tests
on
normally

consolidated
clay
samples

•  Mohr
envelopes
in

terms
of
both
total

and
effecNve

stresses
for

consolidated

undrained
tests
on

over‐consolidated

clay
samples

Use
of
CU
Strength
in
Engineering
PracNce


•  Soils
are
fully

consolidated
and

at
equilibrium
with

the
exisNng
stress

system

•  Then,
addiNonal

stresses
are

applied
quickly

•  Example
Test
for
NC
Clay:

o Consolidated
under
a
stress
of
150
kPa

o Then
sheared
undrained
in
axial
compression

o Principal
stress
difference
at
failure
=
100
kPa

o Induced
pore
water
pressure
at
failure
=
88
kPa


c’ and cT are assumed

to be zero
 Unconsolidated
–
Undrained
(UU)

Test
Behaviour

•  Sample
is
placed
in
the
triaxial
cell
with
the
drainage

valves
closed

•  No
consolidaNon
occurs
when
confining
pressure
is

applied

•  Usually,
pore
water
pressures
are
not
measured

•  Sample
is
loaded
to
failure
in
10
to
20
minutes

o Called
Q‐test
(for
“quick)

•  Test
is
a
total
stress
test
and
it
yields
the
strength
in

terms
of
total
stresses

 Typical Stress-Strain
Curves for UU Tests

Note: max. Δσ occurs

at relatively low strains

a)  Remolded and some compacted clays

b)  Medium sensitive undisturbed clay
c)  Highly sensitive undisturbed clay
Typical Mohr failure
envelopes for UU Tests

All samples have same

water content and void
ratio – therefore the
same undrained shear
strength
100% saturated clay

Increased cell pressure

compresses any air voids

Partially saturated clay

•  Change
in
both

strength
and

deformaNon

aZer
remolding

an
undisturbed

sample
of
NC

clay


•  φu
=
0
when

results
are

ploged
with

respect
to
total

stress

Use
of
UU
Strength
in
Engineering
PracNce

•  Engineering

loading
is
assumed

to
take
place
so

rapidly
that
Δu
cannot
dissipate
or

for
consolidaNon

to
occur


•  Change
in
total

stress
does
not

affect
the
in
situ

undrained
shear

strength

•  Scope
 •  Test
Specimens

•  Terminology
 •  Procedure

•  Significance
of
use
 •  CalculaNon

•  Apparatus
 •  Report

ASTM
D
2850:
Unconsolidated
‐
Undrained

•  Terminology

o Failure
is
defined
as
the
maximum
principal
stress

difference
or
that
measured
at
15%
axial
strain

•  Test
Specimens

o Minimum
diameter
of
3.3
cm

o Height
to
diameter
raNo
between
2
and
2.5

o Procedures
to
prepare
undisturbed
and
compacted

samples

•  Procedure

o Axial
strain:
1%/min
for
plasNc
and
0.3%/min
for
brigle

o Test
should
last
approximately
15
to
20
minutes

ASTM
D
2850:
Unconsolidated
‐
Undrained

•  CalculaNons

o Axial
strain,
ε
=
ΔH/H0

o Average
cross‐secNonal
area,
A
=
A0/(1
‐
ε)

o Principal
stress
difference,
σ1
–
σ3
=
P/A

o CorrecNon
equaNons
for:

•  If
all
around
pressure
changes
specimen
length

•  CorrecNon
for
sNffness
of
the
rubber
membrane

•  Report

o Index
properNes
of
material
being
tested

o IniNal
H0,
Diam.,
γd,
void
raNo,
w.c.,
saturaNon,
etc.

o Rate
of
axial
strain,
strain
and
stresses
at
failure

o Stress
–
strain
curve
and
failure
sketch

 ASTM
D
4767:
Consolidated
‐
Undrained

•  Terminology

o Failure
is
defined
as
the
maximum
principal
stress

difference
or
that
measured
at
15%
axial
strain,
or

o Maximum
stress
obliquity,
σ’1/σ’3

•  Test
Specimens

o Same
as
for
UU
test

•  Procedure

o SaturaNon
procedure
to
ensure:

Pore
Pressure
Parameter,
B
>
0.95


(B
=
Δu/Δσ3)

o ConsolidaNon
procedures
to
ensure
specimen
reaches

equilibrium
in
a
drained
state
at
the
effecNve
consolidaNon

stress
required

 ASTM
D
4767:
Consolidated
‐
Undrained

•  Procedure

o Axial
loading
to
produce
equalizaNon
of
pore
water

pressures
throughout
the
sample
at
failure

o Assuming
failure
will
occur
at
4%
axial
strain

•  Rate
of
strain
=
4%
/
(10
x
t50)

o Details
for
measuring
pore
water
pressures

•  CalculaNons

o EquaNons
for
height
and
area
aZer
consolidaNon
(Hc
&
Ac)

o Axial
strain,
ε
=
ΔH/Hc

o Average
cross‐secNonal
area,
A
=
Ac
/(1
‐
ε)

o Principal
stress
difference,
σ1
–
σ3
=
P/A

 ASTM
D
4767:
Consolidated
‐
Undrained

•  CalculaNons

o Calculate
effecNve
stresses
based
on
Δu
measured

o CorrecNon
equaNons
for:

•  CorrecNon
for
filter
paper
strips

•  CorrecNon
for
sNffness
of
the
rubber
membrane

•  Report

o Index
properNes
measured
for
material
being
tested

o EffecNve
consolidaNon
stress,
t50

o Hc
,
Ac
,Diam.,
γd,
void
raNo,
w.c.,
saturaNon
aZer

consolidaNon

o Rate
of
axial
strain,
strain
and
stresses
at
failure

o Stress
–
strain
curve
and
failure
sketch

Interpreting Scatter in Test Results

•  US
Army
Corps
of

Engineers
Procedure:

Draw
strength
envelope

in
a
posiNon
such
that

data
from
two‐thirds
of

the
tests
lie
above
the

failure
envelope

•  Primary
References:

o A.W.
Bishop
and
D.J.
Henkel,
1962.

The
Triaxial
Test

o R.D.
Holtz
and
W.D.
Kovacs,
1981.

Introduc:on
to

Geotechnical
Engineering

o US
Army
Corps
of
Engineers:
Engineering
and
Design

Manual:
Slope
Stability,
Appendix
D
Shear
Strength

Characteris:cs
(EM
1110‐2‐1902,
Oct
31,
2003)

o ASTM
D2850
and
D4767
Standard
Test
Methods