Geotechnical Aspects of Open Stope Design at BHP Cannington

G C Streeton1

ABSTRACT GENERAL STOPING CONDITIONS AT

Open stoping dimensions are typically controlled by a number of factors CANNINGTON
including production drilling capacity, scheduling, fill constraints,
orebody dimensions and host wall rock conditions. Frequently host wall Geology and rock mass conditions
rock conditions are the key control on stope dimensions. As a new mine
is brought into production, there is a steep geotechnical learning curve The geology of the Cannington deposit has been described
for those associated with the development of the mine. Initially, the elsewhere, and the interested reader is referred to Bailey (1999)
experience and observations gained at other mines is used to guide the and Streeton (1997). In general the stoping conditions at
stope design. As experience is gained at the new operation, site specific Cannington are characterised by a higher quality ore material
techniques can be validated for use. This has been the case at BHP
Cannington, where a heavy emphasis has been placed on the validation of
forming the stope side walls along the strike of the orebody, and
empirical stope stability techniques for site specific conditions and a poor quality sillimanite muscovite schist and quartzites forming
experience. the stope hangingwall and in places the stope crowns. The two
A range of empirical and numerical techniques can be applied during main open stoping orebodies at Cannington are the Glenholme
the design of open stopes. The empirical techniques are an experience Breccia orebody and the Hangingwall Lead orebody.
based application of known performance levels during the design of open The Glenholme Breccia is a quartzite hosted silver lead zinc
stopes. The numerical modelling and structural tools can be used to breccia. Typically the quartzite has a UCS of around 125 MPa to
compliment the empirical techniques so that an understanding can be 160 MPa. The Hangingwall Lead is a mafic pyrobole tending to
gained of the likely stress and structurally controlled failure mechanisms.
banded mafic quartzite hosted massive lead-silver orebody. The
Hangingwall Lead ore has a higher UCS than the Glenholme
INTRODUCTION Breccia, trending from 180 MPa to 250 MPa. Additionally, the
Hangingwall Lead ore tends to be more massive. In both cases
Production stoping has been in progress at the Cannington Mine
the ore forms competent walls. There is however a key structural
since 1997. The stoping activities are undertaken in a series of
control over the stability of stope sidewalls and that is a series of
geologically and geotechnically distinct orebodies, which display
strong persistent joints and faults. These structures can control
a range of rocktypes and geotechnical environments. localised instability leading to overbreak from sidewalls and blast
Geotechnical considerations have become one of the key design hole rings. Oversize material reporting to the drawpoints from
criteria during stope and mining block design the fall-off can have a detrimental impact on production.
This paper describes the techniques which are used to assess The sillimanite muscovite schist wraps around the hangingwall
and improve stope stability at Cannington, and also details the and crown of the Glenholme Breccia and Hangingwall Lead
key areas of knowledge gained from three years of stoping. The orebodies. This material is a poor to very poor schist which
analysis techniques employed include; empirical stability typically displays very low RQD’s in the range of 0 to 30, and
methods, three-dimensional numerical modelling, and typically has an in situ UCS of 10 MPa to 30 MPa. When
trigonometric wedge analysis. Examples are also given to intersected in development drives this material can display a
demonstrate a modified open stoping system used in an adverse tendency to cave if surface support in the form of fibrecrete is not
geotechnical environment. applied very soon after firing. In some areas the schist abuts
directly against the cut-off ore boundary, and at other times there
THE CANNINGTON SILVER LEAD ZINC MINE may be a buffer of several metres of poor quality quartzite
between the ore boundary and the schist. A stress path technique
The Cannington Silver Lead Zinc Mine is wholly owned and (Board, 2000), has proven to be very useful in helping to
operated by BHP Minerals Pty Ltd, a wholly owned subsidiary of generate an understanding of the failure mechanisms that occur
the Broken Hill Proprietary Company Limited. The orebody was in this material.
discovered by BHP Minerals Pty Ltd in 1989. Prefeasibility and
feasibility studies were undertaken during the early to Stress conditions
mid-1990s. Full-scale construction commenced in 1996, and
stoping and milling operations were under way by late-1997. The Cannington Mine is operating in a relatively low stress
Currently the mine produces around 24 Moz of silver in 290 Kt environment when compared to those found in other deeper
of silver rich lead concentrate, and 130 Kt of zinc concentrate operations. The mine does not operate stopes at depths greater
annually. Cannington is the world’s single largest mine producer than 575 metres. Typically, the virgin principal stress level prior
of silver, and is a dominant supplier in the lead concentrates to mining at the vertical centre of the orebody, 450 metres below
market (Harvey, 2000). surface, is around 25 MPa. The maximum principal stress
direction is approximately east-west and is subhorizontal. It has
The underground mine currently produces at a nominal rate of
proven to be difficult to obtain a comprehensive understanding of
1.5 Mt of ore annually. The bulk of the mine production is
the stress field at Cannington. Stress measurements have been
sourced from sublevel open stopes. Bench stoping is also used. undertaken at five separate locations through out the mine.
Localised variations in stress direction have been found.
Difficulties in undertaking the measurements have been
commonly experienced due to the structured nature of the
1. MAusIMM, Geotechnical Engineer, BHP Cannington, PO Box 5874 rockmass.
TMC, Townsville Qld 4810. E-mail: [Link]@[Link] Mining induced stress changes are occurring as would be
expected. The effects of localised stress concentration has been
observed in pillars prior to their extraction. In other areas, stress
reduction and even relaxation is occurring, leading to stability

MassMin 2000 Brisbane, Qld, 29 October - 2 November 2000 747

G C STREETON

issues related to the removal of confining stresses from adopted mine grid. The first stoping blocks were designed with
structurally formed sidewall blocks. Numerical modelling the sidewalls oriented E-W and the hanging and footwalls
techniques are employed to predict the likely rockmass response oriented N-S.
to mining, and will be discussed elsewhere in this paper. The significance of the orientation of a regularly occurring
dominant fault series was not appreciated at the time of the stope
Extraction sequence design. This fault set has a dip/dip direction of around 80°/330°,
and is now recognised throughout the mine. The unfavourable
The stoping strategy at Cannington is driven by maximising intersection angle of this fault set with the grid aligned stope
grade and extraction of the resource (Harvey, 2000). Open
sidewalls resulted in the immediate observation of large-scale
stoping commenced in the TU89 mining block, targeting the
block fall-off from the sidewalls. The fall-off of large blocks of
Glenholme Breccia and Hangingwall Lead orebodies. Extraction
is currently taking place in the ST67 and TU89 blocks, and is several thousand tonnes each resulted in poor mucking
scheduled to move to the southern end of the deposit over the conditions, and undercut fill masses for the adjacent secondary
next few years. Extraction in the southern section of the mine has stopes.
not yet commenced, but a retreat method of open stopes A decision was made to alter the orientation of the stope
retreating back to the northern end of the mine will be employed. sidewalls by 30 degrees to improve the intersection angle of the
dominant fault set with the stope sidewalls. This has resulted in
• The only pillars left are long term isolated access pillars. improved sidewall performance. Structurally controlled fall-off
Stress has not been an overriding factor in determining the
from competent wall rock still occurs, but at much reduced scale
extraction sequence. Stopes typically range in size from one
than observed in the first stopes. Fall-off of material from unfired
sublevel to three sublevels. The sublevel interval is 25
ring faces remains an issue, which is being dealt with by
metres, and stope plan dimensions range from 20 x 10 m to
changing blast practices and COS orientation wherever possible.
30 x 30 metres. Whenever ground conditions permit,
multiple sublevel stopes are used due to their superior
productivity over single sublevel stopes. APPLICATION OF THE MODIFIED STABILITY
Paste fill is used to assist with achieving maximum extraction. GRAPH METHOD
The high value of the ore means that full extraction of the The empirical stability graph method has been employed as the
orebody is preferred and remains viable when the cost of cement main stope stability assessment tool at Cannington. The
for the paste fill is considered. As the bulk of the resource is technique has been well documented elsewhere and is in use in a
contained within a 300 metre cube, there is a need for fast wide range of operations. The technique has been subjected to
turnover of stopes, as production cannot be spread out over a significant review since the original method was first proposed
long strike length. The rapid cure time of the paste fill allows fast by Mathews (Mathews et al, 1980). In particular, the
stope turnover and provides an environmental benefit by modifications proposed by Potvin (1988) and Stewart and
reducing the amount of process plant residue requiring surface Forsyth (1995) have been adopted at Cannington. The Stewart
storage. Cannington has exposed significant fill masses of up to
and Forsyth approach, known as the Modified Mathew’s Method
50 metres by 25 metres wide six weeks after placement with
now forms the basis of the technique employed at Cannington.
cement contents of around 3.5 to 4.5 per cent.
A combination of transverse and longitudinal stoping is The first level of assessment is undertaken during the
employed. The orientation is dictated by the orebody geometry, preliminary block design for a new stoping block. At this stage
development access and to an increasing degree, the ground of the assessment, individual stope shapes have not yet been
conditions. generated. Geological and geotechnical data from the limited
access development and the diamond drilling database is
collected and collated on a rock type basis. Predictions of the
GEOTECHNICAL DATA COLLECTION likely wall forming rocktypes are made in conjunction with
Geotechnical data is routinely used during the stope design geological staff, and a statistical analysis is undertaken of the
process, and is sourced from a number of areas. The primary data RQD, joint set, and joint alteration data for each rock type.
source is diamond drilling, from which a range of structural Using the geotechnical data, a prediction can then be made of
features and alteration data is collected. Fortunately Cannington the likely stable spans for each wall forming rock type in the
has collected RQD, fracture frequency, joint set and alteration mining block. An important part of this process is comparing the
data from the commencement of diamond drilling before the statistical populations of RQD and joint sets for a new mining
feasibility study. This has resulted in a large database which can block with the data from an active mining block where the wall
be easily interrogated using Datamine and Surpac. performance is known. The likely stable span predictions can
Routine geological mapping of all mine development is then be used during the stope block design process. For a given
undertaken by the geological staff. The mapping provides stability number (N’), there will be a maximum permissible
information on the persistence, orientation and nature of key hydraulic radius as defined by the stability curve. A maximum
structures that intercept the development in the vicinity of the hydraulic radius is identified for each rocktype forming stope
stope. walls and crown. The various wall dimensions can then be
Stress measurements have been undertaken, and have been iterated to produce the most efficient stope shape within the
discussed earlier in the paper. confines of the maximum allowable hydraulic radii. For example
a 20 metre high by 20 metre stope wall has the same hydraulic
Rock property testing has been undertaken on all the radius as a 40 metre high by 13.3 metre wide stope wall. The
Cannington rocktypes. Limited triaxial testing has been higher stope wall may allow a greater level of production
complemented by extensive point load testing of drill core. efficiency to be achieved.
The first step in the design of an individual stope is the first
REORIENTATION OF STOPES IN RESPONSE TO pass generation of a stope shape by the planning engineers. This
SIDEWALL FALL-OFF shape will be based on the geological wireframe, and the mine
operating constraints such as maximum blast hole length,
The first application of back analysis and stope design
sublevel interval, etc. Owing to the complex orebody geometry,
modification occurred soon after the commencement of open
every stope design is assessed on an individual basis, and if
stoping in the Glenholme Breccia orebody. In the northern part of
necessary modified as further information comes to light. The
the mine, the N-S strike of the orebodies is parallel to the
stope shape is assessed for wall stability using the Modified

748 Brisbane, Qld, 29 October - 2 November 2000 MassMin 2000

 GEOTECHNICAL ASPECTS OF OPEN STOPE DESIGN AT BHP CANNINGTON

Mathews method. The diamond drilling data for each stope wall or CMS. These devices have become widely used in open
is statistically analysed, and the median value of RQD and Jn is stoping operations throughout the world. The key benefit they
used as input into the stope design process. The use of the offer is to enable the performance of stope sidewalls to be far
median value of RQD and Jn has been found to provide a good more accurately examined (Miller, Potvin and Jacob, 1992). In
correlation between predicted and actual stope performance. The the case of Cannington they helped significantly in developing
joint alteration and roughness values are obtained from an an understanding of the various modes of failure that can occur
inspection of any development present in the stope walls and in the different geotechnical domains present in the stope wall
from inspecting the relevant diamond drillcore. rocks.
The stress factor (A) is evaluated using the intact rockmass The ‘Cannington Stability Curve’ was generated by plotting
strength and an estimate of the induced compressive stress. the stope performance data and fitting a curve to the data. As can
Where the stope wall stresses cannot be easily predicted, as in be seen from the accompanying charts in Figures 1 and 2, there is
the case of a secondary stope pillar, numerical modelling is not a significant difference between the Modified Mathew’s
undertaken. curves and the Cannington curve. Despite the differing
Geological mapping of surrounding development is reviewed geotechnical domains, the use of the one curve appears to be
to obtain the dip/dip direction of any key structures, for valid for both orebodies.
ascertaining the value of the joint orientation factor (B), and for
assessing the likely failure mode for the gravity adjustment factor NUMERICAL MODELLING
(C).
Numerical modelling is undertaken at Cannington in two key
An automated spreadsheet has been developed, which
areas. The first is the examination of mine wide issues such as
calculates many of the input values based on the data provided.
the long-term performance of key infrastructure and pillars, and
The end result of this process is a unique stability assessment will not be discussed in this paper. Localised numerical
for each wall of the proposed stope. The design is then reviewed modelling is undertaken at a smaller stope scale, to aid in the
in light of the findings of the stability analysis. The need for prediction of stope wall performance, in the development of
cablebolt support is evaluated, or if the wall stability appears to extraction and firing sequences, and cut-off-slot (COS)
be unsustainable the dimensions can be changed. At times there orientation. Typically, the stopes targeted for numerical
may be scope to increase the stope dimensions due to favourable modelling are secondary or tertiary stopes, particularly isolated
conditions. pillars prior to their extraction. An important role for numerical
modelling is to aid in the understanding of stress related failures.
BACK ANALYSIS AND VALIDATION One mode of failure that has impacted on mine production
When the Modified Mathew’s stability graph technique was first from tertiary pillars has been large blocks falling off the face of
applied at Cannington, a record of stope performance case rings into the COS, when the COS has been oriented parallel to
history did not yet exist. Initially, the stability curves applied the principal stress. A suggested mechanism for the block fall-off
during the stope design process were those suggested by the is that the pillar is subject to an increased principal stress, but
developer’s of the Modified Mathew’s stability chart (Stewart that the confining stress is removed when the COS is opened.
and Forsyth, 1995). Gradually as a record of stope extractions This then allows small-scale reactivation of the cross-cutting
has been developed at Cannington, a program of back analysis faults leading to block loosening and fall-off.
and comparison of actual versus predicted stope wall When it is possible to select between orienting the COS
performance has been undertaken. The key of this process, has normal or parallel to the principal stress during pillar extraction,
been to develop a site specific stability curve that best reflects the a model can be run using Map3DTM to aid in the
record of stope performance at Cannington. It has eventuated, decision-making process. If the COS is put in normal to the
that the Cannington stable line is not dissimilar to the lower principal stress there is the potential for significant relaxation or
bound of the Modified Mathews Method Transitional zone line. even tension forming which can also lead to significant block
It could be argued that by adjusting the input parameters for fall-off during extraction. The use of numerical models to assist
any given stope wall, it would be possible to manipulate the decision-making during the stope design process has been
stability number to fit the Mathews curves. For example, much of invaluable. An important area of investigation over the coming
the published literature mentions the use of RQD, but does not 12 months will be the instrumentation and monitoring of a pillar
discuss which RQD value to use if the rockmass under during formation and extraction, as a means of confirming the
investigation is variable. In order to provide a uniform hypothesis on pillar behaviour at Cannington.
benchmark for purposes of comparison, a decision was made to
use the the 50th percentile value of RQD, which results in a CABLEBOLT SUPPORT DESIGN
stability line slightly lower than the Modified Mathews
As in the case of the stability graph stope design process, during
transitional line. If for example a 60 percentile value of RQD was
the extraction of the first stopes at Cannington heavy reliance
used, it may have eventuated that the resulting stability numbers
was made on empirical cablebolt design tools developed from
fitted the Modified Mathews curves exactly. An estimated value
case histories at other operations (Hutchinson and Deiderichs,
for the other Q components of Jn, Jr and Ja is also generated,
1996). The design charts were used to assess the potential
using fracture frequency data from drillcore, and visual
effectiveness, and the length and spacing of the cablebolts
observations of joint conditions in drillcore and mapping.
required.
The stability number calculated using the 50th percentile RQD
Typically, there are two key wall rock failure mechanisms that
value is regarded as a ‘most likely scenario’ for N’. A most likely
commonly occur at Cannington, and the approach to cablebolting
‘best case’ and ‘worst case’ stability number for each wall be
for each of these mechanisms is different. Sidewall failures in
using the 75th and 25th percentile’s respectively. This is useful,
competent rock are typically structurally controlled, and the
as it can give an indication of the potential variability in stability
cablebolt design tends to focus on attempting to prevent the
of the rockmass forming a stope wall.
mobilisation of large blocks of material. In this case, the
The currently used Cannington specific curves are based on cablebolts installed are typically twin strand Garford Bulb
the extraction records from around 20 stopes. The stope sidewall cables. The length and orientation of the bolts are driven by the
and crown performance has been assessed using the data from a shape of the structurally controlled block and the location of
laser stope survey system known as a Cavity Monitoring System access points to install cablebolts from.

MassMin 2000 Brisbane, Qld, 29 October - 2 November 2000 749

G C STREETON

HL Stability back analysis

Updated Mathew's Stability Graph Date: 18/2/00
(Stewart and Forsyth 1995)
Stability Number N'

FIG 1 - Hangingwall lead stope back analysis with ‘Cannington Stability Line’.

Glenholme Breccia Back Analysis

Updated Mathew's Stability Graph Date: 13/9/99
(Stewart and Forsyth 1995)
Stability Number N'

FIG 2 - Glenholme Breccia stope back analysis with ‘Cannington Stability Line’.

750 Brisbane, Qld, 29 October - 2 November 2000 MassMin 2000

 GEOTECHNICAL ASPECTS OF OPEN STOPE DESIGN AT BHP CANNINGTON

<---- Poor Ground - Large Stopes Good Ground - Small Stopes ----->
<------ Increasing bolt density

Bolt pattern spacing

``

FIG 3 - Cablebolt performance design chart.

The other key type of failure mechanism present in stopes at BLASTING AND FRAGMENTATION
Cannington is an unravelling or caving mechanism formed in
stope hangingwalls and crowns where the wall rock is the Primary breakage of ore during blasting can have a significant
sillimanite muscovite schist mentioned previously. Two impact on stope and mine productivity. An excessive level of
uncontrolled failures in this material have occurred resulting in oversized material reporting to drawpoints can result in costly
the abandonment of stopes. The aim of cablebolting in this secondary breakage and lower mucking rates. Alternatively, if
material, is not to prevent the occurrence of failure, but to control not enough coarse material is fed into an AG mill, there will be
the propagation of failures between sublevels. Cablebolting in insufficient grinding media available (Scott, Segui and
this material aims for density and depth, and is designed to create Kanchibotla, 2000). Ideally, the blasted material needs to strike a
a zone where unravelling can be halted. Typically, surface balance between the two, so that the issues associated with too
support is also used in these instances. much oversize and insufficient coarse grinding material can be
Cablebolting can be a cost-effective method of dilution control avoided.
in open stope mining, particularly if the mine is adequately Cannington has experienced problems at both ends of the
equipped for the installation of cablebolts. It has been estimated fragmentation spectrum. Production has at times been hampered
in Canadian open stoping operations that the cost of cablebolting by too much oversize presenting to drawpoints. At other times
was approximately equal to the average cost per tonne of dilution orepass blockages have been experienced due to cohesive arching
(Nickson, 1992). Cost models developed by the author have also of fines in the ore stream. The mine to mill concept recognises
indicated this ratio of cablebolt cost versus dilution cost is valid that the energy requirements and costs are lower if greater
at Cannington. There comes a point however, where regardless of breakage is achieved from blasting, and less energy is required
the amount of cablebolts installed, a significant failure is for crushing and grinding (Scott, Segui and Kanchibotla, 2000).
unavoidable and the stope wall is not viable at its current Prior to undertaking a mine to mill program, most blasting at
dimensions. Cannington was undertaken by applying relatively uniform blast
To aid in the identification of non-viable stope walls, a design parameters to all stopes. As the material properties of
cablebolt effectiveness chart has been developed specifically for both ore types can vary widely, this led to significant variation in
Cannington conditions (Figure 3.). The chart compares the fragmentation of ore reporting to the drawpoints. It was
stability number divided by the hydraulic radius (N’/HR) against recognised that if a standard blast design was being applied, then
the design cablebolt density. The use of N’/HR is useful, as it the factors that were resulting in differing fragmentation levels
allows the comparison of different combinations of wall would include the in situ strength and degree of jointing of the
dimensions and rockmass stability number. It can be seen below rockmass being blasted. An intensive investigation revealed that
that when the value of N’/HR is below a threshold value of 0.7, both of these parameters can vary widely even within the same
that a failure is likely, regardless of the density of cablebolts orebody.
installed. This chart is used to identify these walls so that As a substantial body of rockmass joint data exists within the
fundamental changes to the stope design can be made. The diamond drilling database, it has been possible using blasting
design chart for cablebolt length provided by Hutchinson and models, to develop a simple predictive tool that assists with the
Deiderichs (1996) is used for assessing cablebolt length identification of an appropriate powder factor. All of the fracture
requirements. frequency data within a stope wireframe is extracted from the

MassMin 2000 Brisbane, Qld, 29 October - 2 November 2000 751

G C STREETON

0.30
0.28
0.26
0.24 FF=1
0.22 FF=3
0.20 FF=5
0.18 FF=10
0.16 FF=15
0.14
0.12
0.10
75 100 125 150 175 200

Powder Factor - kg/t UCS - MPa

FIG 4 - Powder factor selection chart.

database. The median fracture frequency is calculated, and

graphed against the insitu rockmass compressive strength. Using
Figure 4, an appropriate powder factor can then be identified.
The use of this prediction tool has resulted in an increase in
powder factor in the harder more massive hangingwall lead ores,
and a reduction in powder factor for some of the weaker more
highly structured material found in the Glenholme Breccia.
As part of the routine stope geotechnical assessment process
undertaken for all stopes during the design phase, a
recommended main rings powder factor is calculated.

AN EXAMPLE OF GEOTECHNICAL ISSUES

DRIVING STOPE DESIGN
Multiple lift stopes can offer productivity advantages such as
increased conventional mucking, reduced number of flat
bottomed stopes, and increased drilling efficiencies. These
benefits are lost however, if the stope walls display high levels of
instability. A solution to the instability is to reduce the height of
the stopes, leading to the loss of some of the productivity
advantages. The excessive and unacceptable dilution resulting
from the poor performance of the sillimanite muscovite schist
indicated that it was not sustainable to continue with multiple lift
stopes with schist hangingwalls.
Simultaneously with the realisation that the stoping geometries
would need to be altered to reduce mine dilution levels, was the
observation of the good performance of the paste fill placed in
the first primary stopes upon their exposure. It became apparent
that a paste fill exposure of up to 20 metres wide and 50 metres
high offered substantially lower dilution levels than a schist
exposure of similar dimensions.
The geometry of the Glenholme orebody is such that it has
proven possible to design a stacked stoping sequence that takes
advantage of the benefits of multiple sublevel stopes whilst
minimising the height of the schist exposure in any one stope. To
achieve this, part of the schist exposure is replaced by paste fill FIG 5 - Example of a stacked open stope layout in the
from an adjacent preceding stope. Glenholme Breccia orebody.

752 Brisbane, Qld, 29 October - 2 November 2000 MassMin 2000

 GEOTECHNICAL ASPECTS OF OPEN STOPE DESIGN AT BHP CANNINGTON

This stacked open stope technique (Figure 5) does run a risk of described of past and present staff at BHP Cannington is also
allowing some ore dilution due to scour of paste fill from the acknowledged, in particular Gavin Power of the JKMRC, and
adjacent stope, but has been successful at reducing the high Clint Scott of Coffey Geosciences Pty Ltd.
dilution levels experienced from the schist. To-date, paste
dilution from scouring has not occurred. In addition to the REFERENCES
benefits of reduced stope dilution, the technique has allowed the
production efficiency benefits of multiple level stopes to be Bailey, A, 1999. Cannington Silver-Lead-Zinc Deposit, in Geology of
retained. An added advantage is that the resultant geometry of the Mineral Deposits of Australia and Papua New Guinea, pp
387-394 (The Australasian Institute of Mining and Metallurgy:
the stopes results in smaller crowns, narrower sidewalls, and the
Melbourne).
elimination of flat bottomed stopes. The fall-off from the crowns
Board, M, 2000. Mining and Modelling of Orebodies. Underground
of the previous stope into the mucking horizon of the next lift of Mining Methods Seminar, Australian Centre for Geomechanics.
a flat bottomed stope has been an issue at Cannington and Perth, May 2000.
elsewhere (Potvin and Hudyma, 2000). Harvey, P, 2000. Open Stope Mining - Design of the Cannington Silver
Mine. Underground Mining Methods Seminar, Australian Centre for
SUMMARY Geomechanics. Perth, May 2000.
Hutchinson, D J and Deiderichs, M S, 1996. Cablebolting in
There has been a significant period of learning and back analysis Underground Mines (BiTech Publishers Ltd: British Columbia)
associated with the first few years of stoping at BHP’s Mathews, K E, Hoek, E, Wyllie, D C and Stewart, S B V, 1980.
Cannington mine. When the mine commenced stoping, stability Prediction of Stable Excavation Spans for Mining at Depths Below
analysis techniques were sought from other operations and 1000 Metres in Hard Rock. CANMET Report 802-1571.
published material to provide a template to apply at Cannington. Miller, F, Potvin, Y and Jacob, D, 1992. Laser Measurement of Open
Over time these techniques have been validated and customised Stope Dilution, Proceedings of the 93rd CIM Annual General
to suit the local conditions. Meeting, (Can Inst of Min, Metall and Petroleum: Montreal).
The value of a good data set of structural data is immense Nickson, S D, 1992. Cablebolt Support Guidelines for Underground Hard
when it comes to the individual analysis of stope walls. Rock Mine Operations. MASc Thesis (unpublished), University of
British Columbia, Vancouver.
Significant variations in wall rock quality and geometry between
stopes mean that individual stope assessments are required rather Potvin, Y and Hudyma, M, 2000. Small and Large Scale Open Stoping in
Canada. Underground Mining Methods Seminar, Australian Centre
than applying a uniform set of design dimensions to all stopes. for Geomechanics. Perth, May 2000.
The data required to undertake the individual analyses can be Potvin, Y, 1988. Empirical Open Stope Design in Canada. PhD Thesis,
readily obtained as part of a resource diamond drilling program, University of British Columbia, Vancouver.
and the additional cost of collecting RQD, alteration and Scott, A, 1999. Blasting and Fragmentation Report. Internal Report to
structural data is minimal. BHP Minerals Pty Ltd.
The complex and variable ground conditions found at Scott, A, Segui, J and Kanchibotla, S, 2000. Ore Characterisation for
Cannington have necessitated the use of flexible stoping layouts Mine to Mill Fragmentation, in Proceedings Fourth International
which take advantage of the engineered fill mass as a means of Mining Geology Conference, pp 247-253 (The Australasian Institute
limiting the exposure of some of the very poor quality schists. of Mining and Metallurgy: Melbourne).
Stewart, S B V and Forsyth, W W, 1995. The Mathew’s Method For
Open Stope Design, Bull Can Inst Min Metall Petrol, July.
ACKNOWLEDGEMENTS Streeton, G, 1997. Geological Modelling of a Stoping Block at BHP’s
The author would like to thank the management of BHP Cannington Ag-Pb-Zn Mine, in Proceedings Third International
Cannington for their permission to publish this paper. The Mining Geology Conference, pp 123-126 (The Australasian Institute
technical contribution to the development of the systems of Mining and Metallurgy: Melbourne).

MassMin 2000 Brisbane, Qld, 29 October - 2 November 2000 753