23/09/2024
Advanced laboratory testing
Material used in this lecture may have been
reproduced and communicated to you in
accordance with section 113P of the
COPYRIGHT Copyright Act 1968.
NOTICE Unless stated otherwise, all teaching and
learning material provided to you are
protected under the Copyright Act and is for
your personal use only. This material must not
be shared or distributed without the
permission of the copyright owner/s
2
23/09/2024
Outline
Consolidation testing
Shear strength testing
Constant head tests as part of
consolidation testing
Various aspects of constant head testing to measure hydraulic
conductivity were discussed previously. When carrying out these
tests in a test such as a slurry consolidometer, an interesting aspect is
the significant change that can occur across the test, as the material
becomes denser, and thus has a lower hydraulic conductivity.
Constant head tests were carried out in a slurry consolidometer.
After equilibrium was achieved in each load, a gradient was created
across the sample using another pump connected to the base porous
stone. This pump was only used at this point, so does not represent a
drainage boundary for the sample during the consolidation stages.
4
23/09/2024
Constant head tests as part of
consolidation testing
Here are the results of a slurry
consolidometer test on a
mixture of laboratory standard
kaolin clay and rock flour. Each
point represents a constant
head test carried out at the end
of a particular consolidation
stage.
Note the order of magnitude
reduction in hydraulic
conductivity from early parts of
the test to the final stage which
is 800 kPa.
Constant head tests as part of
consolidation testing
All of the hydraulic conductivity
values shown are directly measured.
However, this value can also be
estimated by means of cv calculated
for load stages, as described
previously, and the stiffness of the
sample during consolidation with the
follow expression:
ρ
Where
M is confined modulus
ρ is unit weight of water
is gravitational acceleration
6
23/09/2024
Constant head tests as part of
consolidation testing
Solving for k
Directly
measured
Good agreement between the two
means to estimate hydraulic
conductivity is generally seen, and
provides a useful check on the data.
Calculated with
Time dependent processes
8
23/09/2024
Anisotropy
Similar to hydraulic conductivity (kh / kv), coefficient of
consolidation is also generally anisotropic. The value
measured in laboratory tests is related to the vertical
direction (i.e. cv), whereas some in situ tests such as
the CPT dissipation test measure a value more relevant
to the horizontal direction (ch). As with hydraulic
conductivity, this module focusses on the vertical
direction values – kv and cv.
Devices
The following types of devices are typically used to measure the consolidation
behaviour of soils and tailings, each with its own benefits and disadvantages
- Oedometer
- Constant rate of strain (CRS) device
- Control rate of load (CRL) device
- Rowe Cell
- Slurry consolidometer
- Seepage induced consolidation test (SICT)
Some are beyond the scope of this module.
10
23/09/2024
Oedometer
The oedometer is the oldest and simplest form of consolidation test. Loads
are applied to a confined specimen using either dead weights (usually with a
lever arm) or an electro-mechanical load frame. Loads are applied
incrementally – e.g. 10, 20, 40, 80, 160, 320 kPa load. A loading period of 24
hours is often used.
Generally the samples are small – starting the test about 20mm in height,
diameter of usually to 50 – 80 mm. Such sample dimensions are sufficient
for a stiff material, such as:
- A sand, regardless of initial preparation conditions
- Samples obtained from in situ, which have already undergone significant
consolidation, and thus are stiffer than those beginning as a slurry
11
Constant rate of strain (CRS)
A CRS device uses a specimen of similar initial dimensions as an oedometer.
However, in this case, the specimen is back pressure saturated. Rather than
incremental loads such as an oedometer, loading is carried out by moving the
loading ram down into the sample at a set rate. Drainage is only allowed
from the top of the sample, while excess pore pressure occurs at the base
and is measured with a piezometer. The rate of loading must be carefully
selected to ensure excess pore pressure occurs within an optimal range.
With the rate of loading and magnitude of excess pore pressure known,
values of average vertical effective stress, void ratio, stiffness, hydraulic
conductivity, and coefficient of consolidation can all be inferred.
This form of device is the leading practice method to characterize the primary
consolidation behaviour of clayey foundation soils. It also finds use in the
characterization of clayey tailings samples.
12
23/09/2024
Constant rate of Deformation (CRD) test
Sample is compressed at a
constant velocity (rather
than using the load
increment approach of the
oedometer and Rowe cell)
Solutions available that
can provide cv & k value
even if the rate of load
application is not fully
drained
13
Rowe cell
The Rowe cell was discussed previously in the context of hydraulic
conductivity measurement and to introduce the concept of back pressure.
However, this device forms a very popular means to measure the
consolidation behaviour of slurries and other samples. The specimen is
loaded by the water pressure within a diaphragm. Back pressures are used in
most cases to promote saturation. In tailings practice the Rowe cell finds
most common use in the testing of slurries that are first poured into the
device, followed by loading first with weights (to create a sufficient stiff
sample to assemble the diaphragm) followed by diaphgram loading. Single or
double drainage can be used. If single drainage is used, then excess pore
pressure can be monitored at the base of the cell with a piezometer.
After load stages constant head testing can be carried out by using two
pumps, as per discussion previous in this module.
14
23/09/2024
Displacement
transducer
Measurement of volume of water expelled
(against back pressure)
Rigid drainage tube (settles with sample)
Apply pressure to
top of sample
Air pressure chamber
Bellofram seal
Grooved rigid plate
Porous disks (150 mm diameter) Teflon ring Sample
Valve
Base drain for two-way tests, or for
direct permeability measurement
Pore pressure transducer
15
Rowe cell (1D compression)
Features
•Vertical stress applied
hydraulically
•D up to 150mm
•Backpressure can be applied
to ensure Sr=1
•1-way drainage (upwards)
•Pore pressures (at base)
measured during consolidation
process
16
23/09/2024
Rowe cell
Upper part
Lower part
17
Rowe cell
18
23/09/2024
19
20
23/09/2024
21
Can vary both the drainage
condition and/or the boundary
loading condition
• Loading condition (top surface) can be either ‘free
strain’ or ‘equal strain’
• Drainage:
• Vertical: upwards only (with base pore pressure
measurement, or both upwards and downwards (2 way
drainage)
• Horizontal (radial): may be inwards or outwards (estimated
9-fold difference in rate of drainage)
22
23/09/2024
Similarly, permeability tests
can be conducted with flow:
• Upwards
• Downwards
• Inwards
• Outwards
Be careful of ‘rind’ effect when doing
radial (outward) drainage tests
23
24
23/09/2024
One-way
drainage; equal
strain
25
Slurry consolidometer
The slurry consolidometer shares many aspects with a Rowe cell,
in that a slurry can be poured into the device and then vertically
loaded. Slurry consolidometers have undergone a number of
developments over the past few decades, and as such many
versions now exist with upgrades and variations to improve their
testing capabilities.
Some slurry consolidometers make use of back pressure, while
others do not. Arguably, the slurry consolidometer is one of the
least standardized tailings laboratory tests. However, when
applied successfully, they can provide excellent information on
the consolidation characteristics of tailings.
26
23/09/2024
Slurry consolidometer system
Test cylinder
Water
reservoir
Base
Top
cap
Bottom
Initial loads
porous disc
(dead weight)
Top porous
disc
Components before assembly
27
Slurry consolidometer system
Vertical load
applied by shaft
Displacement
measuring device
Sample
Bottom drainage under load
control, to
connect pressure
pump to base
Load frame, to
apply vertical loads
(based moves
upwards)
Test underway
28
23/09/2024
Back pressure slurry consolidometer
Some devices test with back pressure, thus including
additional pumps. Loading is generally applied
hydraulically in such devices as with a Rowe cell.
29
On low-stress testing
A particularly important aspect of testing to measure hydraulic
conductivity of tailings that begin as a slurry is under what
magnitude of vertical effective stress testing can first commence.
For many applications – say in-pit deposition with a high rate of
rise – the hydraulic conductivity under low effective stresses is
important. However, conventional devices such as Rowe Cells
and Slurry Consolidometers cannot measure hydraulic
conductivity of a specimen until it has been consolidated to at
least 10-15 kPa. At lower levels of effective stress, the act of
applying a gradient across the specimen leads it to undergo
consolidation, thus making measurement difficult.
30
23/09/2024
Seepage induced consolidation
test (SICT)
Owing to issues around low stress testing, the use of seepage
forces to consolidate specimens under low effective stresses was
developed. This involves “pulling” water from the base of a
slurry sample, which leads to a negative pore water pressure.
This thereby increases the effective stress of the soil, leading to
consolidation. Very small increments can be applied through
use of different rates of water removal from the base.
This test enables hydraulic conductivity and stiffness to be
inferred at very low levels of effective stress. While a specialized
test, it is finding increased adoption in tailings practice owing to
efforts to improve the characterization of tailings.
In visual appearance and some aspects of testing principles, the
device resembles a slurry consolidometer.
31
Seepage induced consolidation test
32
23/09/2024
Benefits and drawbacks- Hydraulic
conductivity / consolidation devices
The measurement of seepage and consolidation
behaviour was largely focused on the Rowe Cell, Slurry
Consolidometer, and the flexible wall permeater (in a
triaxial).
Each of these devices has their own pros and cons,
which are worth discussing
33
Slurry consolidometer
Advantages
These devices are purpose-built to test material that
begins the test as a slurry. This is the best
representation of tailings deposition we can generally
produce in the laboratory. As such, the Slurry
Consolidometer allows very small loads to be applied to
the tailings slurry after being poured in and/or settling
within the column. Also, given the height of typical
consolidometer columns, a significant thickness of
tailings remains even after applying high loads.
34
23/09/2024
Slurry consolidometer
Disadvantages
Many slurry consolidometer systems do not involve the
application of back pressure. This limits the applicability of
the device to samples that are “poured in” as a slurry (and
are thus saturated by definition). Even then, it could be
argued that the lower quality saturation within the device
(drainage lines, filters stones, the specimen itself) can
reduce the accuracy of results.
There is also a lack of standardization around slurry
consolidometer devices and techniques. This increases the
onus on the laboratory to have quality internal procedures
and equipment
35
Shear strength
Direct simple shear
Advanced triaxial
36
23/09/2024
The simple shear test
The concept: apply a horizontal shear stress while maintaining constant
vertical stress
Types of simple shear devices
Cambridge type NGI type
Wire reinforcement Platen
Hinge
Rubber
End flap membrane
Ideally, complementary shear stresses are
applied on vertical face, but …
37
The NGI type device is very
widely used
Consider stresses within NGI device specimen
´
τv
It is not possible to draw a
unique Mohr’s circle from
NGI simple shear data
Consider the limiting Mohr circle from a
triaxial test (red circle). The simple shear
stress state corresponding to this circle
gives a stress ratio, ρ´, that is less than φ´
38
23/09/2024
Direct simple shear (DSS)
Below is an image of a modern DSS system, with the top loading
platen about to be lowered onto a sample of loose sand. The
sample is confined laterally by a series of Teflon coated rings.
These prevent lateral deformations but enable shear-
deformations to occur later in the test (as they can move across
each other with relatively low friction, being Teflon coated).
39
Direct simple shear (DSS)
Consolidation
Shearing
σ’vc K0 Constant volume condition xy
condition
xy
40
23/09/2024
Direct simple shear (DSS)
Constant volume means that as the sample is sheared, the DSS
device works to prevent the height of the sample from changing.
For example, if a sample “wants” to contract, this would tend to
result in the sample reducing in height. However, the DSS
system detects this slight propensity to decrease in height, and
instead reduces the vertical total stress acting on the specimen
to prevent the decrease in height from occurring. As modern
devices are computer-controlled, this process of working to
prevent height change can occur in a very rapid “loop” between
the vertical actuator, load cell, and computer.
41
Direct simple shear (DSS)
In the constant volume condition, the pore pressure within the
specimen stays at ~0 kPa. Alternatively, the vertical total stress
acting downwards on the sample changes to maintain the
constant height condition. As the area of the specimen does not
change, maintaining constant height means maintaining constant
volume.
While there is no shear-induced pore pressure, it has been
convincingly shown that the change in total stress during
shearing (to maintain constant volume conditions) is equivalent
to the amount of shear-induced excess pore pressure that would
have occurred.
42
23/09/2024
Uniformity (or not) of stresses
within simple shear device
Observed rupture zone in cylindrical apparatus*
*No rupture zone was detected in any of the
tests on loose sand
Distribution of stresses from
an elastic analysis
See: Budhu (1984)
43
Another type of simple shear device
The Berkeley/UWA simple shear device
In this apparatus, the lateral stress is imposed
and thus horizontal stress is also known
44
23/09/2024
Undrained (constant volume)
tests in simple shear apparatus
In virtually all simple shear devices, the sample is not saturated
and the pore water pressure cannot be measured. However, ……
Constant volume conditions can be achieved by ‘locking’ the top
platen, preventing vertical movement while horizontal shearing
takes place.
Work by many researchers (a number of researchers have shown
convincing evidence that the required changes in vertical stress are
equivalent to the excess pore water pressure that would have
developed in a saturated, undrained test.
The beauty of this outcome is that a simple shear test can be
carried out on a dry sand but the resulting behaviour is the same
as would have been obtained from testing a fully saturated sand.
45
Drained residual strength
Soils that comprise a significant proportion of clay-sized particles
that are “true” clays (i.e. not rock flour) can exhibit a
phenomenon referred to as frictional softening. This is where,
after significant shearing of the clay soils, the clay particles
(which are platy in shape) begin to align, forming a slickenslided
surface. Where this occurs, the frictional strength of this
slickenside surface is reduced. This can be an important
condition for stability in some cases – for example when building
embankments on dense, low hydraulic conductivity clays. This
behaviour was implicated in the failure of the Los Frailes TSF in
Spain in 1998.
46
23/09/2024
Drained residual strength
Testing soils to measure their residual friction angle
typically utilizes a device called a ring shear. This device
can shear to as large a deformation as required, owing
to its circular shape. Detailed analysis of drained
residual strength and the ring shear is beyond the
scope of this course, it is mentioned here simply so you
are aware of it and its purpose. In tailings engineering
one finds it primarily applied to characterize certain
clay foundations of TSFs. See for example its use to
characterize the clay foundation of the Cadia TSF in the
investigation panel (Jefferies et al. 2019).
47
Triaxial testing
It is common to keep total horizontal stress constant and
increase vertical stress until peak strength is reached (and
beyond) or until limiting axial strain is reached
qpeak
εa
48
23/09/2024
Tests to determine soil strength
1. The triaxial test
D
σ1
Usually use H/D = 2 or more
Because we impose TOTAL
stresses, the pore pressure
σ3 H must be measured
49
Drained or undrained?
During loading of a triaxial specimen, the pore water will tend to change (may
increase or decrease). If drainage valves are closed, water cannot move to
alleviate the resulting pressure difference, i.e. the test is ‘undrained’. As a
consequence, an excess pore pressure develops. This excess pore water pressure
may be negative or positive.
If drainage valves are kept open during shearing and an appropriate rate of load
application is used such that no excess pore water pressure results, the test is
said to be ‘drained’
+
εv
uw εa Drained
- Undrained
εa
50
23/09/2024
Stress paths
σ −σ
2 uw
σ'3 σ'1 σ3 σ1
The top point of the Mohr circles (for both effective stresses and total stresses)
represent the mean stress and the shear stress for any particular stress condition.
There are two widely used forms of the stress path parameters. We will initially use the
‘MIT’ version, but then switch to the more widely used ‘Cambridge’ option
51
MIT stress path parameters
´ ´ ´ ´
´
−
2 2
Effective stress path for
drained compression test
t Effective stress path for
undrained compression
test with positive excess
Total stress path,
pore pressure inclined at 1:1
Initial conditions, t = 0
s´, s
52
23/09/2024
Cambridge parameters
´
− ´
⅓( 2 ´#
Total stress path for triaxial
q compression test
Effective stress path 3
for dilative material 1
p´, p
53
Undrained shear strength, Su
&
τ% σ& ϕ& This is soil strength expressed in terms of effective
stresses, sometimes termed ‘drained’. It requires
knowledge of the pore water pressure
Soil strength expressed in terms of total stress
% () framework; knowledge of pore water pressure not a
prerequisite
σ −σ
2
σ3 σ1
σ´ σ´
2
54
23/09/2024
Relating drained strength to
undrained strength
φ'
X
c'
σ'3 σ'3 σ'1 σ'1 σ'n
X
Undrained
shear strength
Void ratio
X
Cu
X
Undrained
shear strength
Cu log σ‘v
55
Reconstituting a tailings
specimen for triaxial testing
It is extremely difficult to recover undisturbed samples of most mine tailings from the
field. Although block samples can be taken, they are extremely time consuming to
obtain and can usually only be recovered from shallow depths
It is thus usually a requirement that specimens be ‘reconstituted’ in the laboratory. The
question is, what method is most suitable/representative.
56
23/09/2024
Reconstituting a tailings
specimen for triaxial testing
Air pluviation Water pluviation
Moist tamping
57
Moist tamping
Slurry deposition
58
23/09/2024
Triaxial testing acronyms
CIU: Isotropically consolidated, undrained compression
CAU: Anisotropically consolidated, undrained compression
CK0U: K0 consolidated, undrained compression
Consider the initial condition is a slurry (effective stresses ~ 0)
φ´
σ −σ
2 CIU
´
´
−
Axial
strain, εa
CK0U
Axial
strain, εa
σ σ
2
59
Results from CK0U tests on
Merriespruit tailings
60
23/09/2024
Undrained (constant volume)
tests in triaxial apparatus
In a triaxial apparatus, the sample can
be ‘isolated’ from the water in the
triaxial cell (isolation is by the rubber
membrane, sealed using O-rings to
the top and bottom platens).
By closing drainage valves (e.g. the
‘back pump’ in figure), zero volume
change can be ensured, giving a truly
undrained test.
61
Sample preparation approaches
In some ways, the types of tests, and specific test methods are
the easy part
What can be harder is ensuring that the samples one is testing
are relevant to the actual tailings in situ
62