University of Wollongong

Research Online
Coal Operators' Conference Faculty of Engineering and Information Sciences

2013

Issues related to stability design of very high spoil

dumps
Leonie Bradfield
University of Newcastle

John Simmons
Sherwood Geotechnical and Research Services

Stephen Fityus
University of Newcastle

Publication Details
L. Bradfield, J. Simmons and S. Fityus, Issues related to stability design of very high spoil dumps, 13th Coal Operators' Conference,
University of Wollongong, The Australasian Institute of Mining and Metallurgy & Mine Managers Association of Australia, 2013,
376-386.

Research Online is the open access institutional repository for the

University of Wollongong. For further information contact the UOW
Library: [email protected]
 2013 Coal Operators’ Conference The University of Wollongong

ISSUES RELATED TO STABILITY DESIGN OF VERY HIGH SPOIL

DUMPS

Leonie Bradfield1, John Simmons2 and Stephen Fityus 3

ABSTRACT: As spoil dumps get higher, particularly in strip mining where most overburden is placed
in-pit, consequences of slope failure become disproportionately greater.

Current understanding of the shearing behaviour of spoil for stability design has involved a combination
of laboratory-scale diagnostic testing and engineering judgment. This is a relatively empirical approach
that provides a linear shear strength envelope for materials known to exhibit non-linear behaviour,
particularly under high confining stresses. A shortcoming to the diagnostic testing is that oversize
particles are usually scalped to accommodate the device capacity. The influence of prototype-size
particles on the geomechanical behaviour of mine spoil is not truly captured.

In response to concerns about overestimating the shear strength and stability of high spoil dumps, and
current plans for coal mine dumps to exceed 400 m in height, there is a need to rationally define the
stress-strain behaviour of more characteristic spoil masses under representative compressive and
shearing loads.

A Large Direct Shear Machine (LDSM) has been designed at The University of Newcastle to generate
reliable stress-strain data on large samples of coal measures spoil (0.72 m x 0.72 m x 0.6 m) subjected to
loads representative of very high dumps (~3.5 MPa). This paper reviews current methods for predicting
shear strength parameters in the context of very high spoil dumps, and presents an overview of the
design considerations of the DSM.

INTRODUCTION

Following the global economic crisis in 2009, the Australian economy sustained growth as a
consequence of continued coal exports. Over the next decade or so, the demand for Australia’s coal is
expected to increase significantly in response to the projected needs of China and India.

The aim of any open pit coal mine design is to provide the steepest possible excavation configuration that
is commensurate with stability, safety and financial requirements. The economic depth for open cut coal
mining, expressed as the ‘strip ratio’, is based on the unit revenue of the coal and the cost to recover it.
The increased demand for Australia’s largest export is seeing the nation’s coal producers secure
significant price rises for the commodity. The flow-on effect is that deeper coal reserves previously
regarded unprofitable due to high stripping ratios are now being developed, with mine owners relying on
experience-based models to establish stability design criteria for pit slopes.

For high-value commodities such as coking-coal, economic depths in excess of 275 m are currently
underway in Australian coal mines, with plans to progress to 350 m or more. This translates to mine spoil
dumps potentially reaching heights of up to 450 m. Current experience-based understanding of mine
spoil stability behaviour falls well short of this, by about 330 m. Furthermore, civil engineering experience
with rockfill dams is that current experience-based models overestimate strength under higher stress
regimes.

Mine owners have relied on geotechnical practitioners for advice on geotechnical risks that could impact
on the success of long-term mine plans. Since current methods are believed to overestimate the stability
of very high spoil dumps to an unknown degree, and the consequences are potentially catastrophic, risks
to mine owners cannot be evaluated and due diligence reviews may find that legal, financial and
operational risks to the business are unacceptable.

1
University of Newcastle and Thiess Pty Ltd, [email protected], M: 04 8776 5555
2
Sherwood Geotechnical and Research Services
3
University of Newcastle

376 14 – 15 February 2013

2013 Coal Operators’ Conference The University of Wollongong

In the context of very high spoil dumps, the uncertainty associated with current strength models can only
be reduced by simulating field-scale conditions in the laboratory.

LIMITATIONS WITH CURRENT DESIGN STRENGTHS

Empirical shear strength envelopes

Linear envelope - BMA Coal strength framework

The Australian coal mining geotechnical community currently uses either totally empirical strength
models or simple models based on extremely small test specimens. The BMA Coal strength framework
was described in 2004 by Simmons and McManus, and represents almost two decades of extensive
in-house research and collaboration with CSIRO on dragline-scale dumps. It is based on laboratory tests
with empirical adjustments using back analyses of several large spoil dump failures.

The framework provides a visual-tactile method for identifying any type of spoil into one of four
categories. Spoil category identification is based on the assessment of five geological attributes,
comprising predominant particle size, consistency, structure, plasticity and age.

Table 1 - Spoil Categories and Attributes (after Simmons and McManus, 2004)

The ‘structure’ attribute is a fabric-related parameter, where larger particles represent the ‘framework’,
and finer particles the ‘matrix’ (Figure 1). Simmons and McManus (2004) describe the framework as the
larger sized fraction that is relatively stiff and transmits most of the forces within the spoil mass when in
particle-to-particle contact. Comparably, they describe the matrix as the finer-sized component that fills
the void spaces between framework particles. Framework-abundant spoils are said to be stronger and
less compressible than matrix-abundant spoils.

Figure 1 - Spoil structure attribute to be used with Table 1 (after Simmons and McManus, 2004)

For each category of spoil, shear strength parameters for three potential strength mobilization modes are
provided, including unsaturated, saturated and remoulded conditions.

14 –15 February 2013 377

2013 Coal Operators’ Conference The University of Wollongong

Table 2 - Shear Strength Parameters for Categories and Mobilisation Modes (after Simmons and
McManus, 2004)

The peak strength parameters cohesion c’ and friction angle ’ are based on a linear fit to data obtained
from laboratory tests simulating stresses experienced in 60-90 m high spoil dumps (Figure 2). The
method has been validated in practice by back-analysis of instability and successful design of spoil
dumps up to 120 m high. The linear Mohr-Coulomb shear strength envelope is expressed as:

(a) (b)

Figure 2 - (a) Three conceptual strength modes for spoil as described by Simmons and McManus
(2004); and (b) modified to explain the linear shear strength approximation adopted in the
framework

It is generally accepted that the frictional strength of mine spoil is strongly dependent on the magnitude of
confining stress. A fundamental shortcoming of the BMA strength framework is that it cannot be
extrapolated to cover the stress states expected in very high dumps. This is because the framework is
based on a Mohr-Coulomb linear fit for data within a defined stress range, and mine spoil shear strength
behaviour is distinctly non-linear. Extrapolation of the framework will overestimate the stability of very
high spoil dumps to an unknown degree, and with plans for coal mine spoil dumps to exceed 400 m there
is a need to investigate the shearing behaviour of representative spoil masses at field stress conditions.

Similarly, extrapolation of the framework to the low stress range will significantly underestimate the
frictional strength for low spoil slopes (<30 m high), with the actual available shear strength considerably
less than that estimated (Figure 3). It is a common occurrence for the lower excavated slope within a
dragline spoil dump to sit at batters steeper than repose angle for short periods of time. Simmons and
McManus (2004) suggest that, in addition to the effects of matric suction, the non-linear shear strength
envelope helps explain this phenomenon.

378 14 – 15 February 2013

2013 Coal Operators’ Conference The University of Wollongong

Figure 3 - Mohr diagram showing framework linear fit with respect to actual strength envelope
(not to scale)

Non-linear failure envelopes

For decades, civil engineers have used non-linear failure envelopes to define the shearing behaviour of
granular materials, particularly in relation to rock fill dam design. A number of studies have used
power-law relationships to explain the dilatant behaviour of compacted rock fills at low effective stress,
and reduced dilation due to particle crushing as stress increases.

Leps (1970) examined rock fill triaxial data obtained from over 20 dam construction projects and
demonstrated a marked increase in friction angle ’ with decreasing confining pressure. Included in the
data were test results from large apparatus capable of handling 1m diameter samples with a maximum
particle size of 200 mm, at confining stresses up to 3.9 MPa. Barton and Kjaernsli (1981) developed a
power-law criterion to model this data, and established a number of simple index tests that could be used
to estimate the peak drained friction angle ’ of rock fill. Recommendations from the Barton and Kjaernsli
study have strong merit for rock-fill projects using the types of rock-fills tested in the study; i.e.
predominantly rocks with high to very high substance strength. However, its applicability to coal
measures spoil of comparatively low substance strength is not well founded.

Charles and Watts (1980) modelled a power-law relationship to describe triaxial data obtained for rock
fills of igneous, sedimentary and metamorphic origin, with varied substance strength and shape. In
contrast to the aforementioned large-sample-high-stress studies, Charles and Watts placed a much
higher importance on the stability of rock fill at comparatively low stresses. Samples with a maximum
particle size of 38mm and specimen diameter of 230 mm were tested over the low stress range
considered critical for lower-slope stability. It was proposed that over a limited stress range (40 kPa to
400 kPa) the shear strength of rock fill is given below, where constants A and b are dependent on rock
type (Figure 4).

= A (σ’)b (1)

The Charles and Watts study is applicable to rock fill embankments up to 50 m high, and covers the
majority of rock types considered suitable for dam embankments. However, a number of limitations exist
in the context of very high mine spoil dump design. The first is the stress range; 400 kPa is equivalent to
a small spoil dump; and the equation cannot be extrapolated out to cover the higher stress range. The
second is that the rock fills tested are heavily compacted well-graded samples; in contrast to mine spoil
dumps that are placed in a loose state, with highly variable gradation as a result of parent rock type and
mining processes. The third is that the samples were scalped to meet apparatus constraints; the
deficiencies of which will be discussed later.

14 –15 February 2013 379

2013 Coal Operators’ Conference The University of Wollongong

Figure 4 - Shear strength for rock fills at low confining pressures, and rock fill constants (Charles
and Watts, 1980)

A CSIRO study (Mallett, et al., 1983) performed direct shear tests on weak pit floor material at South
Blackwater coal mine, and modelled best-fit linear and non-linear relationships to describe its strength
envelope (Figure 5). The material was block-sampled from a clay-filled shear zone, hence not
quantitatively comparable to the shear strength of mine spoil; however it is included in this paper for a
number of reasons. It demonstrates the difference in frictional strength between the models for the low
stress range, as is the case for a typical spoil. In addition it shows agreement for both strength models
within the stress range tested in the laboratory; and infers that accurate shear strength models can only
be achieved by testing representative size samples at the stress range of interest. In the case of very high
spoil dumps, this translates to testing large samples at very high confining stresses.

Figure 5 - Direct shear test results conducted on discontinuity material in a pit floor (Mallett, et
al., 1983)

Scale effects of laboratory testing

The application of laboratory data to spoil dump stability design has been used with caution for decades.
This is partly attributed to the highly variable nature of mine spoil, consisting of large rocks and boulders
and some fines at placement; but more so due to the inability of standard laboratory tests to account for

380 14 – 15 February 2013

2013 Coal Operators’ Conference The University of Wollongong

this variability by testing full-scale samples at as-dumped grading and porosity, and at loads
representative of field conditions.

It is widely accepted that the presence of large rocks and boulders strongly influences the shearing
behaviour of rock fill. A decrease in frictional strength with increasing particle size is well established,
particularly at low confining stresses. Vallerga et al. (1957) found this relationship to be non-existent for
particles less than 5 mm. Barton and Kjaernsli (1981) attribute this to the higher propensity for larger rock
fragments to undergo particle crushing than are smaller particles, as they are more likely to contain
planes of weakness than are sands or fine gravels. However, limitations of standard laboratory apparatus
are such that the shear strength of only the finer fraction of spoil can be measured. Application of such
results to slope design will result in an overestimation of shear strength and stability.

By way of explanation, a standard laboratory direct shear box can test 60 mm x 60 mm samples, of
maximum particle size equivalent to fine gravel, and maximum normal stress up to 2.7 MPa. For testing
coarse materials such as rock-fill, a direct shear box capable of handling larger samples (300 mm x 300
mm) is becoming common-place in geotechnical laboratories, and can test coarse gravels at normal
stresses up to 1 MPa. However, this is still not of sufficient size to handle the cobble and boulder size
particles typical of coal mine spoil.

The overestimation of shear strength when testing only the fine fraction of spoil is demonstrated in Figure
6. Direct shear test results conducted at the University of Newcastle for a typical category 2U spoil
(according to Table 1) are compared with the geotechnical parameters implicit in the BMA Coal strength
framework. A category 2 spoil predominantly consists of fine-grained low-plasticity material (sand) with
larger clasts, is stiff/medium dense and matrix supported. The ‘U’ refers to the unsaturated strength
mobilisation mode (Table 2, Figure 2). The dimension of direct shear test sample was 100 mm x 100 mm
with a maximum particle size less than 4.75 mm. The BMA Coal strength framework is based on a
number of shear strength investigations, including large-sample direct shear tests with size of 1 m x 0.75
m and validated by field slope performance data for dumps up to 120 m high.

Figure 6 - Overestimation of shear strength due to scalping

Groundwater effects

The destabilizing role that water has on the soft rock spoil dumps of coal mines is widely recognized but
difficult to evaluate.

Coal measures rocks are prone to physical and chemical deterioration with changes in moisture content,
resulting in slaking and softening and a reduction in shear strength. Water pressure can have both
positive and negative effects on slope stability; depending on the stress states within the spoil mass, the
moisture content, and time-scale considered. Negative pore water pressure (suction) is usually implicit
where low spoil slopes are observed to sit at batters steeper than repose angle for short periods of time.
Conversely, positive water pressure reduces shear strength by pushing grains apart and reducing their
intergranular friction.

14 –15 February 2013 381

2013 Coal Operators’ Conference The University of Wollongong

It is widely accepted that water pressures must be considered in stability analysis. The civil engineering
field often couples piezometer data with seepage analysis to establish the phreatic surface level and flow
paths through a slope. However, water pressures are not easily ascertained in spoil materials. The
logistical and practical difficulties associated with the installation of piezometers in spoil materials are
many and varied, and often incur significant project costs. Voids are problematic for drilling equipment,
commonly resulting in hole collapse, production disruptions and equipment damage.

Current understanding of water behaviour in strip mine spoils has not advanced far beyond the
conclusions drawn by Gonano (1980, after Boyd, 1976), in that groundwater tends to occupy the basal 3
to 5 m of the spoil foundation, draining through the slope to the level at which water ponds externally. The
BMA Coal strength framework employs a similar model, based on observations of spoil dumps in the
Bowen Basin. It assumes that the phreatic surface is no more than 5 m above the foundation and reduces
to the down-slope toe; with saturated spoil material underlying the phreatic surface.

The inherent uncertainties with groundwater behaviour in spoil dumps characterized by the strength
framework are increased to an unknown degree in the context of very high spoil dumps. In particular, the
extent of saturation and associated reduction in shear strength within the basal layer are significant to
spoil dump stability, as large-scale deep-seated failure mechanisms commonly develop along this plane
of weakness.

There is some conjecture around the potential for mine spoil to achieve effective saturation under the
confining loads characteristic of very high spoil dumps. Preliminary studies at the University of Newcastle
suggest that, although strongly dependent on field moisture content, this could be the case. It was found
that for a typical category 2 spoil subjected to confining loads of 3.5 MPa, compression-induced
saturation could develop at moisture contents greater than 11%. The field moisture content at the time of
sampling ranged between 10 and 11%, following several weeks without rain. The results warrant further
investigation into compression-induced saturation, which will likely be a key stability consideration for
very high spoil dumps.

RESOLVING UNCERTAINTIES ASSOCIATED WITH STRENGTH FRAMEWORKS

Current knowledge of mine spoil shearing behaviour falls well short of that of soil mechanics for several
reasons. The first is due to the significant costs and time associated with the design, construction and
operation of equipment large enough to test characteristic spoil samples at field stress conditions. And
secondly, the BMA Coal strength framework has worked well for nearly a decade, and its limitations are
only starting to emerge with the prospect of deeper mines being developed.

The University of Newcastle is currently undertaking research to update and extend the BMA Coal
strength framework to include very high spoil dumps. Of key importance is the ability to measure the
stress-strain behaviour of characteristic spoil masses under field stress conditions.

Estimation of stress states for very high dumps

The maximum confining stress required to simulate conditions within a 400 m high spoil slope has been
estimated for the two commonly observed large-scale failure mechanisms for spoil dumps; a deep seated
multi-wedge mechanism, and a multi-wedge rill mechanism (Figure 7).

Figure 7 - Large-scale failure mechanisms for spoil dumps (Simmons and McManus 2004)

The slope geometry was simplified to rill angle (37°) for a total slope height of 400 m; the base of which
was assumed saturated to a maximum of 5 m above the foundation floor. A weak remoulded basal layer

382 14 – 15 February 2013

2013 Coal Operators’ Conference The University of Wollongong

occupying the lower 2 m was also assumed. Limit equilibrium analyses explored the effects that
foundation inclination and back-scarp location had on the stress distributions through the spoil mass and
associated Factor of Safety. The Sarma method for non-vertical slices was used to divide the failure
surface into boundaries considered most appropriate for capturing the anisotropic strength state
characteristic of mine spoil (after Simmons and McManus, 2004). This method is based on the principle
of limiting equilibrium and the method of slices, and calculates the critical horizontal acceleration required
to bring the mass of the spoil slope bound by the failure surface to a state of limiting equilibrium (of forces
and moments).

The following observations were made in the analyses:

 The deep-seated multi-wedge failure mechanism developed the highest normal stresses within
the spoil mass, occurring on the base of the passive wedge.
 Enlarging the failure surface geometry by moving the backscarp away from the slope crest
increased the maximum normal stress within the spoil mass. However, this also increased the
Factor of Safety against failure occurring.
 Increasing the foundation inclination had the opposite effect, reducing the Factor of Safety and
the maximum normal stress within the dump.

The analyses suggested that, for a marginally stable 400 m high spoil dump, the maximum normal stress
acting on the base of the passive wedge could range between 3.1 and 3.4 MPa, depending on foundation
inclination. Figure 8 shows an example 2D limit equilibrium analysis for a 400 m spoil dump of simplified
geometry. This analysis was performed using the Galena code software (Clover Technology, 2012) and
the Sarma method for non-vertical slices.

Furthermore, Mohr analysis using the BMA Coal strength framework parameters, and extrapolating to
confining stresses equivalent to 400 m of spoil weight, suggests normal and shear stresses at failure are
unlikely to exceed 3.5 and 2.1 MPa respectively. It is recognized that extrapolation will overestimate the
shear strength to an unknown degree; however is useful in establishing the upper-bound of stress
required for discussion.

Figure 8 - Example 2D limit equilibrium analysis of a simplified geometry for a 400 m spoil dump

Field-scale test apparatus

To the authors’ knowledge, few testing machines exist worldwide that can meet the
large-sample-high-stress requirements for shear strength testing of representative spoil masses at field
stress conditions.

A large direct shear test frame was constructed to test then-representative spoil masses as part of a
CSIRO study into the stability of spoil piles (Mallett, et al., 1983). At the time of the study, spoil dumps
rarely surpassed 90 m in height. The apparatus could handle large samples (1 m x 0.75 m x 0.67 m) and
test at consolidating and shearing loads of 1.3 MPa.

14 –15 February 2013 383

2013 Coal Operators’ Conference The University of Wollongong

It is understood that a large-scale-high-stress triaxial device exists at the IDIEM Laboratory of the
Universidad de Chile in Santiago. The apparatus can test samples 2 m x 1 m in diameter, with maximum
particle size of 203 mm at confining pressures up to 2.5 MPa. The apparatus has been used to
characterize the geotechnical properties of very high waste dumps in the copper mines of Codelco,
Andina Division (CAD) in Chile. The waste rock tested by Linero et al. (2006) is classified as ‘hard-rock’,
consisting of porphyry and granodiorite of high substance strength. To the author’s knowledge, the
apparatus has not been used to perform shear strength studies on the ‘soft-rock’ coal mine spoils.

In comparison, it is evident that the CSIRO apparatus falls short of the required design stress of 3.5 MPa
for very high dumps. Furthermore, logistical limitations prevent the use of the IDIEM Laboratory device for
shear strength investigations into coal measures spoil at the University of Newcastle.

Large direct shear machine

Overview

A large direct shear machine (LDSM) has been designed at the University of Newcastle as part of a
postgraduate research project. Construction of the LDSM commenced in mid-2012, and is expected to be
fully operational by 2013. The design criteria for the LDSM were:

 To be capable of handling large (full-scale) spoil samples;

 To have sufficient hydraulic capacity to simulate field loading conditions estimated for very high
spoil dumps (minimum of 3.5 MPa and 2.1 MPa for normal and shearing loads respectively);
 To maintain constant normal load whilst also minimising tilt of the top plate;
 To apply a constant rate of horizontal displacement;
 To be as rigid as possible to prevent sample rotation and minimise side-wall deflection, whilst
also minimising friction losses;
 To be able to conduct wet (inundated) and dry tests.

Figure 9 shows the schematic of large direct shear machine.

Figure 9 - Large direct shear machine

The dimensions of the internal shear box are as large as possible within budget constraints such that
3.5 MPa confining pressure can be imposed on the shear plane. The initial research proposal indicated

384 14 – 15 February 2013

2013 Coal Operators’ Conference The University of Wollongong

that a 1 m x 1 m shear box would suitably handle the large samples required. However, budget limitations
have restricted the size to 0.72 m x 0.72 m.

The internal shear box consists of a 350 mm-high upper box, and a 250 mm-high lower box. The
sidewalls consist of 120 mm thick solid steel (250 grade); designed to minimise deflection and associated
volume changes under the very-high expected loads. High-tensile strength (Grade 8.8) 30
mm-diameter threaded rods join the sidewalls such that the boxes can be disassembled between tests if
required. Removable water-proof walls (external box) encase the internal box, and enable inundated or
‘wet’ tests to be conducted.

The internal shear box is mounted in a large self-reacting stand-alone frame, fabricated from over 26
lineal metres of 530 mm-deep universal beams. The frame is window-shaped and has the dimensions of
3 m x 1.2 m x 2.5 m (l x w x h), and reinforced with over 100 steel plates up to 16 mm thick. During testing,
the LDSM will weigh more than 8 t.

Some scalping will be necessary to accommodate the sample size. With a maximum sample height of
600 mm, and likely to reduce to 450-500 mm following the application of confining load, a maximum
particle size between 65 mm and 100 mm is anticipated. Note that this is based on the recommendations
from test methods developed for smaller apparatus. A new test method will be devised for the LDSM,
which will set out the nominal maximum particle size, subject to machine trials.

The confining and shearing forces will be applied via a number of 100 t double-acting hydraulic actuators.
Considerations to controlling tilt of the top-plate led to the strategic placement of three actuators such that
corrections for displacement could be individually applied during consolidation, whilst still maintaining a
common constant load between them. The three cylinders can deliver a maximum consolidating stress of
up to 5.7 MPa to the shear plane; however will be restricted to 3.5 MPa in initial stages of testing. Two
cylinders acting in the horizontal direction can transfer up to 3.7 MPa stress to the shear plane, however
it is anticipated that the peak shear stress is unlikely to exceed 2.1 MPa.

Loads will be measured via five load cells; three positioned in series with the actuators applying the
confining loads; and two opposing the horizontal actuators. Horizontal and vertical displacements will be
measured via transducers located as close to the shear plane as possible.

The LDSM allows for 150 mm of horizontal travel; equivalent to approximately 21% horizontal strain.
Small scale direct shear tests on the spoil intended to be tested in the LDSM indicated that a peak shear
stress was generally reached between 10 and 15% horizontal strain. The available 21% allows for any
potential scale effects, and for post-peak stress-strain behaviour to be observed. The nominal rates of
displacement will vary for the category of spoil tested and the saturation condition; and will be determined
from full-scale consolidation tests on the LDSM.

For comparison purposes, Figure 10 shows the internal shear box of large direct shear machine and a
standard size shear box inset.

Figure 10 - Scale comparison: Internal shear box of large direct shear machine, with a standard
size shear box inset

LDSM outputs

It is anticipated that the proposed LDSM will provide peak shear strength data for coal mine spoils
subjected to a wide range of confining loads, particularly those representative of very high dumps. The

14 –15 February 2013 385

2013 Coal Operators’ Conference The University of Wollongong

widest possible range of coal measures spoil types will be tested at various moisture contents in keeping
with the strength mobilisation modes outlined in the BMA Coal strength framework.

The framework is used with confidence to characterize coal measures spoil dumps up to 120 m high. The
LDSM data will be used to generate the full-scale stress-strain data required to extend the BMA Coal
strength framework to reliably assess the stability for spoil dumps up to 400 m high.

Additionally, the LDSM will allow for scale effects to be measured. Although some scalping will be
necessary to accommodate its capacity, it is expected that the stress-strain data can be calibrated by
back-analysis of existing very-high coal mine spoil dumps.

CONCLUSIONS

The uncertainties associated with the application of current shear strength models to very high coal
measures spoil dumps have been presented, and scale-effects inherent to standard laboratory testing of
mine spoils have been discussed. These shortcomings can only be overcome by the construction of a
purpose built machine capable of simulating field-scale conditions; however access limitations prevent
the use of existing large-sample-high-stress machines for testing Australian coal measures spoils. A
large direct shear machine (LDSM) has been designed at the University of Newcastle as part of a
postgraduate research project. It is anticipated that the LDSM will provide reliable stress-strain data for
full-scale samples of spoil subjected to loads representative of very high dumps. The data will be used to
update and extend the BMA Coal strength framework to cover these higher stress ranges.

ACKNOWLEDGEMENTS

This paper describes aspects of the first author’s postgraduate research, funded by Australian Coal
Association Research Program as Project C20019 based at the University of Newcastle. In-kind support
from Thiess Pty Ltd, particularly for provision of the hydraulic components vital to the operation of the
LDSM, is gratefully acknowledged.

REFERENCES

Barton, N and Kjaernsli B, 1981. Shear strength of rockfill, Journal of the Geotechnical Engineering
Division, ASCE, 107(GT7):873-891.
Boyd, G, 1976. Spoil pile stabilization - report on conditions for failure and minimum pit preparation for
stability. Utah Development Company Report.
Charles, J A and Watts, K S, 1980. The influence of confining pressure on the shear strength of
compacted rockfill, Géotechnique, 30(4):353-367.
Clover Technology, 2012. GALENA Slope stability analysis system, version 6.00 [online]. Available from:
www.GalenaSoftware.com.
Gonano, L P, 1980. An integrated report on slope failure mechanisms at Goonyella. Technical Report
Number 114, Division of Applied Mechanics, Institute of Earth Resources, Commonwealth Scientific
and Industrial Research Organisation, Victoria.
Leps, T M, 1970. Review of Shearing Strength of Rockfill, Journal of the Soil Mechanics and Foundations
Division, ASCE, 96(SM4):1159-1170.
Linero, S, Palma, C and Apablaza, R, 2006. Geotechnical Characterisation of Waste Material in Very
High Dumps with Large Scale Triaxial Testing. Proceedings International Symposium on Rock
Slope Stability in Open Pit Mining and Civil Engineering 2007 (ed:Y Potvin). pp 59-75 (Australian
Centre for Geomechanics: Perth).
Mallett, C W, Dunbavan, M, Seedsman, R W and Ross, D J, 1983. Stability of spoil piles and highwalls in
deep surface mines. End of Grant Report Number 150, National Energy Research Development and
Demonstration Program, Australian Government, Canberra.
Simmons, J, and McManus, D, 2004. Shear strength framework for design of dumped spoil slopes for
open pit coal mines, in Proceedings Advances in Geotechnical Engineering: The Skempton
Conference (ed: RJ Jardine, DM Potts & KG Higgins) Vol 2, pp 981-991 (Thomas Telford).
Vallerga, B A, Seed, H B, Monismith, C L and Cooper, R S, 1957. Effect of shape, size, and surface
roughness of aggregate particles on the strength of granular materials, Special Technical Publications,
American Society for Testing and Materials, 212:63-74.

386 14 – 15 February 2013