Articles

A case study of a deep-seated spoil lowwall instability controlled by a classical active-passive wedge mechanism at Anglo American Metallurgical Coal’s Dawson Mine is presented. Following a truck dump extension, widespread displacements were identified during conventional single forward pass de-coaling of Pit 6-8 strip E15. New through-spoil drilling, downhole geophysical logging, and XRD analysis identified a moisture-sensitive tuffaceous claystone unit containing a high proportion of medium-high swelling, mixed-layer illite-smectite clays 11-12m below the lowwall floor. A novel soil mechanics approach was used to determine the mechanical properties of the tuffaceous claystone for which conventional rock mechanics tests could not be applied. Three-dimensional numerical modelling was then undertaken in FLAC3D to: a)validate the characterised mechanical properties; b) determine appropriate buttress slot widths for retreat mining of strip E16; and c) provide a validated base case for the predictive modelling and design of future strips. Operational controls for mining of strip E16 included: surface monitoring (radar, LiDAR); subsurface monitoring (TDR, VWPs); and an adaptive mine plan following the Observational Method. The TDR confirmed that the tuffaceous claystone unit at 11-12m depth was acting as the sole basal horizon controlling the instability and mining of strip E16 was completed safely without coal sterilisation.

Selection of appropriate shear strength parameters for mine spoil slope stability analysis and design is difficult because it requires prohibitively large laboratory equipment to test characteristic spoil samples under meaningful stresses. A convenient alternative to estimate mine spoil shear strength is to adopt published guidelines that have been tried-and-tested in practice. For more than two decades, the Australian coal mining industry has adopted a linear shear strength framework derived from small-scale test data and verified in practice by slope performance of dragline-scale spoil dumps up to 120m in height, and to date this framework has appeared reliable. However, in the field of rockfill dam design there is a broad acceptance of a curvilinear shear strength envelope, and if this is applicable to coal mine spoils, then this industry-accepted framework may overestimate the strength and stability of dumps at higher stress levels. This is particularly relevant in modern times where dump heights (>350m) often exceed the scale (≤120m) for which the framework was developed. This paper explores the applicability of this framework for high-dump situations for a range of coal mine spoils. This is achieved by comparing their framework-assigned strength envelopes with direct measurements of their strength obtained from a custom-built large direct shear machine (LDSM). The machine can test at a much larger scale, in terms of combined specimen size (720mm x 720mm x 600mm) and stress (σ’nup to 4600kPa) than has ever been achieved using a direct shear machine for geotechnical testing of rockfill. A critical outcome is that the LDSM data highlights several non-compliant mine spoils, and stress-dependent shearing behaviour, for which correct application of the published framework will not provide reliable shear strength parameters for design.

Large-scale direct-shear equipment is described and experimental results are presented showing the influence of test boundary conditions on direct shear strength. The equipment was constructed at the University of Newcastle in 2013 for testing square samples up to 720 mm and applying vertical loads up to 4 MPa. Its boundary conditions were modified in 2017 to better comply with the testing philosophy in standards. The main modifications introduced include full compensation for normal load eccentricity induced in the original arrangement and the introduction of a system that would allow the installation of a seamless gap between shear box halves before shearing. Additional modifications considered the restriction of unwanted carriage spinning during lateral displacement and reduction of frictional force between the specimen and the vertical walls of the box. After the modifications were introduced, implementing a careful method for sample construction, repeatability of the friction angle at large displacement in the range of 1° was achieved among single tests. Further equipment improvement strategies were identified

One of the most debated subjects among mine geotechnical practitioners is the degree to which shear strength measurements obtained from conventional geotechnical laboratory equipment can be confidently relied upon to predict the shearing behaviour of actual spoil dumps. The reluctance of practicing engineers to apply shear strength parameters determined from standard laboratory testing equipment relates to uncertainty regarding the significance of scale-effects. This is particularly relevant in current times as spoil dumps approach uncharted heights, and load limitations standard geotechnical testing equipment often results in the extrapolation of Mohr-failure envelopes well beyond their intended stress range. There are two scale-effects of note: The first relates to the degree to which the grading of a spoil sample must be down-scaled to comply with device capacity, such that the influence of prototype-sized particles on shear strength is not anomalous. The second relates to the normal stress limits of the test apparatus; and whether the failure envelope developed from measured strengths can be reliably extrapolated out to the “much higher” stress ranges to simulate field conditions. Scale effects on shear strength have been studied in the literature for soils and rockfills, however, the influence of scale on the shear strength of coal-measures spoil specifically has received little attention. This paper presents the results of an experimental study into the significance of scale effects by comparing shear strength data for a silica sand, considered to be immune to scale effects, with an Australian coal mine spoil tested in standard-sized direct shear machines (100mm, 300mm) and in a large direct shear machine (720mm) which was purpose-built to replicate a large-scale field condition for spoil dumps up to 400m in height. Spoils hear strength is shown to be strongly scale-dependent, both in terms of specimen size and magnitude of normal stress. Furthermore, it is demonstrated that adoption of shear strength measurements obtained from standard laboratory equipment will overestimate the available shear strength of coal mine spoil dumps with height-equivalent stresses ranging between 450kPa-4600kPa.

Coal mine waste rock (spoil) is a weak granular material derived from rocks of sedimentary origin. There are plans for mining operations to construct piles of spoil up to 400m high. Under the consequent burial stresses, the spoil materials undergo substantial compressions, characterised not only by compaction but also by particle breakdown/disintegration. This leads to significant decreases in void ratio with consequent increases in the degree of saturation (Sr) at constant water content. In this study, a series of direct shear tests was undertaken on a typical mine spoil derived from Permian-aged sandstones and mudrocks, compressed under stresses of 1 and 3MPa, with water contents from 0 to 17%. The samples, with different water contents but compressed under the same stress, all attained similar void ratios, but with degrees of saturation ranging from 25 to 100%, and corresponding suctions from 140MPa to 6kPa. When sheared, these differences in suction produced a difference in measured shear strengths of up to 400%, with a consistent trend of decreasing shear strength with decreasing suction. By fitting the measured shear strengths to an extended Mohr-Coulomb shear strength framework, both φ and φb are found to be highly non-linear and suction dependent. Interestingly, when shear strength is plotted against suction on a log scale, the relationship is approximately linear. There is an observed tendency for friction angles to decrease with decreasing suction, and this is shown to be consistent with values back-calculated from spoil pile failures, with the “unsaturated” value of 28° found in the literature being consistent with the values interpreted in this work

This paper presents the results of an experimental study in which a strong Permian mine spoil from the Hunter Valley, Australia was tested under unsaturated and saturated conditions, using a large direct shear machine. The direct shear machine, with a specimen size of 720mm x 720mm x 600mm and a normal stress capacity of 4.5 MPa, was designed and constructed at the University of Newcastle. The spoil material tested is dominated by siltstone with some fine to medium grained sandstone fragments. The large size of the test device allows specimens with particles as large as 100mm to be tested. Tests under unsaturated and saturated conditions recorded considerably different strengths, with peak friction angles of spoil in the unsaturated condition being as much as 6 degrees greater than the same spoil in a saturated state. The reduced strength under saturated conditions is attributed to the effects of reduced matric suction within the rock fragments of the spoil, and the effect this has on the basic rock fragment strength.

Currently, linear Mohr failure strength envelopes are commonly used in the Australian coal industry for shear strength estimation of spoil dumps within certain bounds of normal stress related to historical experiences. Mostly, the linear envelopes appear to be reliable. However, deviation from linearity is fundamentally important for slope design and stability analyses performed for higher values of normal stress, as the true available shear strength may be considerably lower than is estimated from linear models. This is becoming increasingly relevant as spoil dumps are being constructed to unprecedented heights.

This paper presents the results of an experimental study in which a strong Permian coalmine spoil from the Hunter Valley, Australia was tested under both saturated and unsaturated conditions, in a large (720 mm 720 mm 600 mm) direct shear machine over a wide normal stress range, up to that expected for very high spoil dumps. A trilinear Mohr failure envelope was developed from tests performed on 23 specimens, with normal stresses up to 4600 kPa. The secant friction angle values from each test were found to display a corresponding trilinear trend with increasing normal stress. The trilinear behaviour were evident in results from both the unsaturated and saturated samples; however, the linear segment boundaries were lower for the saturated test data. The trilinear behaviour is attributed to differences in the shearing mechanisms that occur under the different stress conditions.

Triaxial testing of mine wastes is commonly undertaken to provide shear strength estimates for stability assessments of mine waste storage facilities. When planning triaxial tests however, the omission of carefully planned testing instructions may produce shear strength parameters which are inaccurate, or not suited to the assessments being undertaken.

This paper details various issues that should be considered when planning triaxial testing of mine wastes. Test results have demonstrated that; test type and initial confining stress are important to define failure envelopes over appropriate stress ranges (which may be particularly important for mine wastes exhibiting curved failure envelopes); laboratory saturation of unsaturated mine wastes may return conservative shear strengths; oversize particles may compromise the reliability of calculated shear strengths, and multistage tests may provide reasonably reliable drained shear strengths, but may overestimate undrained shear strengths.

Civil engineering laboratories use the direct shear test to provide shear strength parameters for soils under drained conditions. Standard shear boxes are of suitable capacity in terms of dimensions and load to provide reliable data for the majority of soils for civil engineering applications. However, the reliability of this data becomes questionable when extended to the field of mine engineering where materials differ from traditional soils, and stresses
may be much greater than those commonly encountered in civil engineering. When performing direct shear tests on mine spoil, geotechnical practitioners and laboratory technicians may be uninformed of the limitations of standard equipment and/or the consequences of applying unrepresentative data to geotechnical models. The purpose of this paper is to outline considerations for using standard laboratory equipment to derive shear strength parameters for coal measures spoil.