INFORMAiWN TO USERS

This manuscript ha8 been mpdud f m the mi c d h W. UMI films

the text di - frwn the original or mpy sutmitW. Thus, some thesis and
dissertation ooQies are in typmwW face, whib athem msy be from any type d
amputer pririfer.
The qurlity of this reproduction k epenent upon the qwlity of the
copy submitted. Broken or hkthd print, cokred or par quari illustrations
and pbtographs, Wnt bhdt hwgh, substandard rnargins, and imp-
alignrnent can adverse& #fect iieproduaion.
In the unlikely event that the author did not Send UMI a ampiete manuscript
and there are rnissing ~ages, these will be noted. Alsa, if unauthorireci
copyright material had to be rernoved, a note will indicade the deletion.
Ovene materiab (e.~., maps, dmhngs. cham) are reproduoed by
sectiming the original, beginning at the upper leftharid wmer and continuing
from left to right in equal sections with m l l werlaps.
Photogtaphs induded in the original manuscript have been reprodud
xemgraphically in this -y. Higher quality 6. x W Ma& and white
photographie prints are available for any photographs or illustrations appearing
in this copy for ar~ additmal dmrge. Contad UMI di - to order.
Bell & Howell Infomatbn and Leamim
300 North Zeeb Road, Ann Arbor, MI 481S1346 USA
8G521-0600
ORE DILUTION IN SUBLEVEL STOPING
Oumar Diakit
Department of Miaing and Met.Uurgicril Engineering
McGiII University, Montreal.
August, 1998
A thesis submitted to Faculty of Graduate Sudies and Research in partia! fiilfiUment of the
requirements of the degree of Master of Engineering.
O Oumar Diakit
National Library
l*l ofcanada
BiMithque nationale
du Canada
Acquisitions and Acquisitions et
Bibliographie Services senrices bibliographiques
395 WeHingon S Ut 395. nie Wemgtcm
OttawaON K l A W O(tawaON K l A W
canada
CaMda
The author has granted a non-
exclusive licence allowing the
National Library of Canada to
reproduce, loan, distribute or seU
copies of this thesis in microform,
paper or electronic formats.
The author retains ownership of the
copyright in this thesis. Neither the
thesis nor substantial extracts iom it
may be p ~ t e d or otherwise
reproduced without the author's
permission.
L'auteur a accord une licence non
exclusive permettant la
Bibliothque nationale du Canada de
reproduire, prter, distribuer ou
vendre des copies de cette thse sous
la forme de microfiche/nlm, de
reproduction sur papier ou sur format
lectronique.
L'auteur conserve la proprit du
droit d'auteur qui protge cette thse.
Ni la thse ni des extraits substantiels
de celle-ci ne doivent tre imprims
ou autrement reproduits sans son
autorisation.
The preparation and the reaiization of this thesis have been made with the guidance
of my academic supervisor Professor Hani Mitri. 1 am veiy grateful to his assistance,
patience and advice during my shidy program 1 also appreciate his suppo* help and
encouragement during the whole period of my study.
1 also extend appreciation to all Professon in the mining Department of McGill
University, especially Professon Malcolm Scoble (Presently at the University of British
Columbia), Michel Bilodeau and Fem Hassani for their help and encouragement during
my study period. My appreciation is also addressed to Mr. John Mossop and Mr. Michel
Vachon for their contribution of theu expenences in the rnining field. 1 also extend my
thanks to the mine engineers: Mr. Paul Gauthier, Mr. John Henning and Mr. Marc Ruest
for supplying field data.
I would tike to thank Mrs. Marina Rosati, Mrs. Norma Procyshyn for their kind
help. 1 would also like to express my appreciation to all my coileagues in the Rock
Mechanics program who gave me assistance in gatheting information.
Finally, 1 thank my farnily for their moral support and understanding.
Abstract
The steeply dipping vein orebody of Bousqua 2 mine is extracteci by
sublevel open stope method. The presence of stmctud discontinuities, high induced
stress distribution, and narrow stope widths are major factors which can seriously affect
ore diIution. Theefore, strict ore dilution control is nec- in order to keep the mine
cornpetitive. The thesis describes the selective mining rnethod adapted to this type of
orebody, and subsequent measures taken to minimize ore dilution. For this study, the
locations of stress relaxation are taken as failures zones. Therefore, these zones are more
important since they are subjected to very low stresses which can provoke rock block
sliding or local wall cavhg (tension fdure). Tracing the zone of tende stress in the aope
wali gives reasonable prediction of dilution and pemiits to calibrate the numerical mode1
using the field data obtaiaed from the cavity Monitorllig System (CMS). Stope mining is
made in several cuts accordhg to the width of the ore zone. For an improved recovey,
some stopes are mined using prirnary and ~condary stoping. Blasthole pattern and rnining
sequences are closely linked to the ore zone configuration. The main causes of ore dilution
are sumrnarized. They are a combination of several factors such as ground conditions,
blasting damage, state of stresses around the stope, and stope design. In order to minimize
ore dilution fiom hanging-wall and foot-wali, cable support is instded. Numerical
modelling is carried out for a typical cable bolt pattern. ni e results of modellmg have
demonstrateci that the cable bolts were necessuy ta control hanging wall caving. This is
accompanied by a reduction of ore dilution. Consequently the study has shown the
effectiveness of cables bolt as pre-support of the schistose walls in Bousquet sublevel
stoping environment.
R&um&
Les caractristiques associes au pendage prononc des veines mineraiises de la
mine Bousquet imposent leur extraction par sous-niveaux dont le chantier demeure vide.
La presence de discontinuite srucairales, de contraintes induites eleves, ainsi que
l'ouverture des chantiers d'abattage sont des facteurs majeurs qui peuvent provoquer la
dilution du minerai. Ainsi, un contrle strict de la dilution est necessaire pour maintenir la
mine cornpetitive. Cette thse dcrit la mthode slective adapte pour ce type de
minralisation, et relate les mesures a prendre pour minimiser la dilution. Pour cette tude,
les zones de contraintes de relaxation ont t considdres comme des lieux ayant une
potentielle de xupture, par consquent, elies sont considres puisque ces zones sont
soumises des contraintes de ci dement qui peuvent provoquer des glissements de
blocks ou des effondrements (tension de rupture). En reportant les zones de tension sur les
schemas des chantier il a t possible de predire les zone de dilution; et de calibre le mode1
numerique en utilisant les donnes de terrain obtenues 6 apartir du moniteur de mesure du
chantier (CMS). D'autre part, les zones de concentration de contraintes induites peuvent
tre aEectees par des engrnements, dans le cas majeur par des coups de terrain
(contrainte de rupture). L'extraction du minerai est faite en plusieurs squences
dependenment de la largeur de la zone rnineratise. Pour un meilleur rendement, quelques
chantiers d'abattage sont repartis en chantier primaire et chantier secondaire. Les patrons
de trous de sautage a les squences de sautages sont troitement Lies la configuration de
la zone minralise. Les causes principales de la dilution du minerai se rsumes. Elles
sont une combinaison de plusieurs facteurs tels que les conditions de terrain, les
dommages du sautage, l'tat des contraintes autour du chantier, et le design du chantier.
Pour minimiser la dilution du minerai provenant de l'pontes ~upeneure~des cbles
d'ancrage ont t installes. La modlisation numenque est faite sur le patron de cblage
typique de la mine. Les rsultats ont montr que des cbles taient necessaires pour
stabiliser I'eponte superieure. Cela s'est accompagn par une rduction de la dilution. Par
consequent, l'tude a montr I'fficacit des cables d'anchrage utilises comme pre-
support pour des roches schisteuse des epontes de la mine Bousquet.
Table of Contents
Aeknowledgmcnts
Abstirct
Table of Contents
Liste of Tables
List of Fi gum
In traduction
General
Study problern
Objectives and scope
Bousquet mine geological h g
Thesis outhe
Sublevcl Mining Methods
General
Stability aspects in sublevel mining
Bousquet sublevel stoping
Mine final design
Bousquet 2 stoping
Conclusion
Ore Dilution
Overview
Factors uising ore dilution in sublevel d g
Effkct of ground conditions
Effkct of in-situ stresses
Effkct of stope design
Effect of geology
Effect of blasting
Numerical Modelling
Introduction
GeomechaNd data
Staternent of problern
Analysis of results
Interpretation: stope wail behaviour
Conclusion
Murumement and Control of Ore Dilution
Theoretical calculation
Practicai measurement
Background
Cavity Monitoring System and measurement of open st op
Control against ore dilution
Cable bolting in sublevel mining
General
Cable bolting techniques in Sublevel rnining
Cale bolt design procedure
Factors controllhg cable bolt performance
UncontroliabIe factors
Controllable factors
Cable bolting et Bousquet 2 Mine
Cornpirison Behveenn CMS Murummentr and Numeri d
6-1
ModeUing Ruults
Corn parison 6-1
Discussion 6-1 1
Conclusions and Recommendations 7-1
Conclusions 7-1
Suggestion for funher research 7-2
Refercnccs
Appendix A: Bousquet Mine Plans
Appendis B: Bluthole Pattern dk Blasting Sequenccr
Appendix C: e-z toolr Ditafie for Numerical Modcliing
Appendu D: Ruults of Stope Ligbt Measurement by C MS
List of Tables
Table No. Title
2.1 Sublevel mining methods with their record4 pedonnance.
3.1 Typical joint sets at Bousquet mine.
3 -2 Results of the blasting vibration indicators.
3. 3 Results of overbreak meamirement in Bousquet stope hanging
wall.
4.1 Rock mass quatity modified - Hoek & Brown Criterion.
4.2 Rock mus properties at Bousquet mine, resulting fiom uniaxial
and triaxial tests..
Figure No.
1.1
List of Figurcs
Titie
Actual conditions of minerai reserves at the Complex of
Bousquet Mine.
Geology of the vertical section of Bousquet Mine
Nature of orebody and mining techniques for steeply dipping
orebody.
Influence of rock wall quality on the choice of mining method.
Vertical Crater retreat (VCR).
Transverse blasthole in sublevel vertical retreat.
Longitudinal blasthole in sublevel vertical retreat.
Longitudinal longhole rnining horizontai retreat.
Typical retreat minhg
The mining sequence for Bousquet stope 9- 1- 15.
Design of Bousquet stope blastholes and blasting sequences
lUustntion of diierent types of ore dilution in steeply dipphg
vein rnining.
Page No
Page No
1-4
Eumple of s t o p design showing intentionai and unintentional
ore dilution in sublevel mining.
The causes of additional ore dilution
A)Inmease of principal compressive stresses with depth; B)
Increase of vertical stresses with depth in Abitibi shield rock.
Effect of stress redistribution on ore dilution in open stope.
Stability cwessment at Bousquet 2 mine using the modified
Mathiew stability gr@.
Manner to found input parameters for the design of open stop
by graph method.
Design of smooth blasting pattern.
Assessrnent of controlied blasting design against ore dilution
caused by overbreak in sublevel open stope.
Blast vibration in the hanging wali at Bousquet 2 stopes.
Fi al sketch of zone mode1 for e-z tools modelhg.
Orientation of the principal compressive stresses for Abitibian
area and for Bousquet 2 mine.
Mesh defonned for the purpose of modeKing.
Principal stress distribution around the stope.
Vertical stress distribution around the stope.
Horizontal stress distribution around the stope.
Displacement vectors multipiied by 30
Infiuence of dilution on final grade of ore extracted.
Relationship between planned dilution, additional dilution and
final dilution.
alustration of parameters used for ore dilution estimation.
The choice of an appropriate cable bolt type relative to rock
mass conditions.
Cable bolt performance assessment.
Effe* of rock ma s and mine induced stress change on ultimate
vii
&le bolt bond capacity.
Schematic di- of the modes of rrinforcing loading at a
discontinuity .
Cable bolt layout for 9-1-15 st op in plane.
Hanging wdl cable bolthg in transversal stoping at Bousquet.
Principal stress trajectories around the stope; Primary stope.
Safkty level result using Mohroulomb criterion - Primary
stope.
Safay level redt using Hoek and Brown criterion - Primary
stope.
Superposition of CMS and numerical moelling, tension zone
results - Primary stope.
Superposition of CMS and numerical modelhg, Morh-Coulomb
safty level results - Primary stope.
Superposition of CMS and numerical modelling, Hoek & Brown
safkty level results - Primary stope.
Superposition of CMS and numerical modeliing tension zone
results - Secondary stope - Analysis 1.
Superposition of CMS and numerical modelling tension zone
results - Secondary stope - Analysis 2.
Chapter 1: Introduction
1.1 G t n c d
Ore dilution problems in hard rock mines, epecidy in Abitibi gold vein rnining, are
mainly attributed to induced fdures that invoive in the mining process (stress, blasting,
excavation, haulage, backfiiiing )and to the rock mass quality (discontinuities, rock mass
stifiess, rock nuiss shear strength, etc.). nierefore, the Mures of waste rock within the
stope are the direct consequence of waU imtability. The inelastic and jointed nature of the
rock in archean aras d e that this instab'ility of grwnd the greatest cause of
underground mining problems. Thus, the effkct of unpredicted ground failure due to the
close excavations and rnining operation, is a matter of great economic importance. Waste
dilution increases the cost of production in that money wl be spent in handling the waste
rock causing dilution, and that ore processing fcilities will be engaged for materiai which
contributes very little to final usefiil metd or minera1 production. Because of that, effort
wiil be made to at least minimize this unexpected add of waste rock. The solution of that
requires the understandimg of aU the effects that directly or indirectly provoke wall rock
caving in the stope. This introduces an interdependence between scientist and mine
operator in the selection of principles which are of value in developing a working
knowledge of cause and effe*. Because the acceptabity of the scientSc approach of
ground failure depends upon how accurately it can simulate conditions to be observed in
practice, thus, through a better knowledge of the science of rock mechanics, practical
applications duected towards the study of cause, prediction, and control of rock failure
could be done. However, these p~ci pl es are ofen obscured by local conditions.
Therefore the geomechanical implications are the cause of ore dilution problems in
sublevel open stoping. These are related to the physical properties and structure of rock
and their reaction to stress distribution arising with the geometry of openings. The
relationship can be visualized as a sheu/strength ratio and the aiticai fdure corne
essentidy fiom shear slidiig and ranly firom compressive buckiing. This means to test @y
numencal modeliing) the rock mass instantaneous response to mining. The consequence of
poor performance of the rock mess involves instsntaneous waste block sliding, or even
important waste rock caving; which unfortunately are added to the ore. Consequently, the
factors to be considered in the design of the aope are the date of stress at the excavation
boundary and in the interior of the rock medium, cornparal with the men@ of the rock
mass; and stability of the immediate wds. Good judgement thus depends upon a judicious
blending of the practical with the theoreticai.
1.2 Study problem
Prediction and control of ore dilution is not it means both the analysis of
instantaneous fdure behaviour of immediate wali stope, and the prezise knowledge of the
geology of the orebody and the surrounding rocks. The phenornena accompanying the
rnining process and the vhety of uncertainties on the geologicd knowledge of the deposit
make that ore dilution problems remain difnailt to handle. By understanding and
controing these phenornena there are some means permitting to considerable reduce ore
dilution. Many charactenst ics associated to these deposits make their mining economical
difficult. In effect, the rocks of Canadian Archean Shield had been submitted to several
tectonic brittle forces. As result, while these rocks are strong, they remain difncult to mine
econornically because of the existence of some negative aspects of associated ground
conditions:
The rocks are highly foliated; which foliation cons t i ~es the weakness planes in
open stope conditions;
These rocks are affecteci by several set of joints of which the orientations and
characteristics change with the space;
m The area is localiy affectad by brittle failures, faults and its associated shear
zones; the effect of the most prominent stnictural features of the host rocks (east-west
trending, steeply south-dipping), can not be ignored;
The p ~ c i p a l stress is perpendicular to the weabiess planes (north-south) and
the vein orebody direction,
Presently the mine has ataked a high Ievd of ore amciaion with 17 000 tomes
per day. On the other hanci, poor ground conditions and spatial distribution of rock stress
remain a challenge, especiaiiy for the rccovay of planed stopes. The rnining methods
involve by the parameters of these orebodies much require for these rocks already
weakened. From a rock mechanics perspective, the hanghg wall and footwall rock as weli
as ore zone are competent but highly jointed with evident foliation joint spacing (near the
waIl contacts) and regional schistosity. Thus, the combination of the nature of the rock,
the state of stress and the design of openings could be the main causes of ore dilution
probltms. in sublevel open stoping ore dilution corne from the crWig of wdi rocks.
1.3 Objectives and scope
The ground conditions associated with the above problerns demonstrate the
dzculty to efficiency extract the ore in narrow vein area. Therefore, the problematic of
ore dilution as ground conditions directly corne to defme all the aspects linked to ore
dilution problems. So, the principal objectives of this project are:
1) To i denw and understand the ore dilution problems;
2) To create links between the unexpected loss of grade in ore mucked and both
gr ound conditions (induced stress distributions and rock mass quality) and
blasting effects;
3) To analyse the measurements taken for reducing the dilution.
As the immediate host rocks are offen highly sheared about the hmging wall and footwall
contacts, our attention has been particularly oriented to induced instability of waU rock
within the stope which lead to immediate caving. To firther understand the problem, we
has emphazed Our study on Bousquet 2 mine.
The Complex of Bousquet mine is owned and operated by Barrick Gold
Corporation. It is located between the cities of Val d'Or and Rouyn-Noranda in North-
western Quebec. The original Bousquet mine began in July 1979. In 1986, exploration on
the eastem side of the property led to development of what was to become Bousquet 2
" ..
il 8
-
ACNICO-EAGLE I
Figure 1.1 : Actuai conditions of minaal resewes at the Com~lex of Bousquet mine.
MINE LARONDE
!
A
J
BOUSQUET
RESERVES MIN
Tonnes Oncar
(~000) (~000)
1
IER
Inv. Min.
2,630 536 8
1
RESERVES MINIERES
Toi mes Oncerr Au
(~000) (r0OOi ( q/ l i n)
ProuvCe
e l
Probable
821 308 11,7
Tonnee
(x000)
ProuvCe
et
Probable
3,21
Inv. Yin.
I
L
COMPLEI
RESERVI
Toniies
( ~ 0 0 0 )
ProuvCe
e 1
Probable
4,03
inv. Min. 4,93
mine, which opened in 1989. The mine has 1.1 million ounces of reserves, grading 0.25
ounces of gold per ton. An additional one d i o n ounces of gold mineraiized orebody,
grading 0.186 ounces of gold per ton, have been identifed, in 1996, exploration of a zone
between Bousquet 1 and 2 was folowed by developmmt work to bring this zone into
production in 1999. There is dso fiirther potmtiai to increase reserves at depth and
between the two mines. SUice the beginning of rnining operations (1989). above 2.8
millions tons of ore with average grade of 10.3 g/t Au and 0.95% Cu, had been extracted
The actual conditions of mineral reserves at the Complex of Bousquet is shown in Figure
1.1 (August 1997)
At Bousquet 2, the improvement of mine operation efficiency is a prime
importance for going toward highest profitabity. Therefore, it is not only to focus the
efforts on great tonnage extraction, but aiso to control the factors which maintain the
quality of ore produced, without comprornising the human safety. Actualiy, most of metal
mines in production become deeper and deeper, and the new mine discovered or will be
discovered are located far fiom the surface. Consequently, the potential darnage a d o r
risk provoked from matenal removed are not ignored. This reaiity led to take account al1
the associated ground effects susceptible to a&* millig performance. The ground effects
which must be considered are mainly the factors associated to the stress and those
associated to the ground rock itself The effkcts of stresses are function of depth and their
orientation and distribution consequently are much critical. Therefore the performance of
mining process is assesseci through the following production critena: Rate of production,
Ore recovery, Ore dilution, and Ore losses. The consequences involve each of these
productivity parameters can negatively affect mine life. However, the qualitative controls
of aU these parameters are practically impossible due to the inevitable lack of precision in
deposit knowledge, orebody estimation and mine planhg due to production constraints.
1.4 Bousquet mine geologid setting
The Bousqua mine is located within the southeni margin of Abitibi greenstone
belt. The steeply dipping gold-bearing zones is a deformeci pyritic polymetallic (gold and
copper) archean deposit located within a zone of intense ductile-brittle defonnation. The
orebody consists of massive suiphide veins ocairring between mafic t us (hanging wall)
and rhyolite tuff (footwai). Figure 1.2 shows the geology of the vertical section of the
mine. Tite orebody reprcsents the West depth extension of Dumgami deposit. This district
includes lot of mines (Bousquet 1 and 2, Doyon, and Dumagarni) and approximately
represents about 43 % of Quebec gold production. In this area the rock mass are
competent but affected many orogeny events that provoke ddormations, fiults and other
structural dismntinuities (joints and foliations). In the sal e of the mine, the orebody and
the wall rock show the evidence of four major defonnations (Dl, D2, D3, D4). As result
of these events, the wall rocks ore highly foliated paralletly to the orebody; and the
hanging wall is af kt ed by a major fault. The rock mass quality, characteristics and
properties are described in geomechanical section.
The Bousquet orebody, through its mineral components and its structural
elements, belongs to massive sulphide with a volcaneuous source. The occurrence of
Bousquet district deposit dong and near contacts between rocks of metasedimentary and
met avolcanic-volcaniclastic explains the synvolcanic ongin of gold minerakt ion. The
mode of deposition of gold and the copper in Bousquet area stiii complicated since the
temporal and genetic relationship with the massive sulphide. The principal minerals
associated with the deposit are sulphidic minerais: pyrite, chalcopyrite, pyrrhotite,
sphalerite, galena; hydrothennd minerals siiicate alteration products: quartz, chlorite,
magnesium, carbonates, ac.; and metamorphic mineral: biotite, cordiente, talc, kyanite,
etc. The genetic mode1 for massive sulphide in Abitibian archean greenstone belts
emphases on the combination of hydrothermal activities and the volcmic events. Then,
those mineralisritions are fomed around the discharged vents of submarine hot Springs in
the tectonically active, high heat fiows environment of felsic volcanic centers (Franklin et
al., 1981). According to this hypothesis, the presence of the mineraiisation is linked to
hydrothermal deposition of gold dong fissures and fault systems. In some localities
probably related to emplacement of mafic, intermediate or felsic intmsions. therefore such
mineralisation still controllad loci of a varicty of volcanic and hydrothermal effects ranging
in age fiom pre- to post-mineraiisation. Thus the location of many individuai orebodies is
controiied by faults and fissures. The guides of exploration for those types of deposits
must be firstly on the specific area seledon. For that, wide variety of features which are
meaningfitl must be collected such as volcanic history of the area and all parameters
controliing ore deposition such as fvourable structures (fauits and fissures), fgvourable
rocks (volcanic rocks) and favourable mineralogy bresence of sulphide minerais).
Recently, the application of geophysical methods (indirect method) has frther increased
the possibility to find these types of deposits.
The thesis is divided in six main chapters. In all the chapters the topic begin by the
general definitions andlor general assumptions, and later fiom data provided by Bousquet
2 mine it has been possible to identify important items that can help in the development of
Our conceptual ideas of the mechankm of failure in the hanging wall and the footwall.
Chapter 1 gives the general uifonnation, the problem of narrow steeply dipping vein
mining in Abitibi area and geological conditions. The description of rnining methods is
presented in chapter 2. Chapter 3 is focused on the effects of mining operations on ore
dilution. Chapter 4 is devoted to numerical modeliing of fdure mechanism on typical
stope of Bousquet 2 mine. The following (chapter 5) explains the estimation method of
dilution and the measures taken to reduce it. Chapter 6 discusses the relation between the
failure zones predicted by the numerical modeliing and the stope boundary obseved by the
CMS. The conclusions of the work including some recornmendations are given in chapter
7.
The choice for a specinc mining method result fiom target knowledge on orebody
characteristics as weil as ground conditions. According to Folinsbee and Clvke study
(1981), the procedure of choice should began with the preliminary selection of a rnining
method or methods on which an enginee~g evaluation will be carrieci out. Naturaiy for
steeply dipping vein orebody (where its configuration including s k , shape, grade
distribution is provided through the geologicd data), the preliminiuy choice must k
carried out on some sublevel mlliing methods (cut and fW mining, shrinkage stoping, etc.).
Then conceptual and engineering studies are done for each method , taken account the
technical and economical optimization. The factors considred for the evaluation are
multiple:
Geomechanical data: cornpetence of the ore and host rocks, joints systems,
foliation, faults, groundwater, etc.;
In situ stresses: Orientation of principal stresses, stress levels;
Economical data: grade of the mineable ore, ore grade distribution throughout
the orebody, the value of the ore in the ground;
Geographical data: natural and societal conditions
Although apenence and engineering judgmmt still provide major input into the
selection of rnining method, subtle diierences in the charuneristics of each deposit, which
may affect the method chosen or the mine design, can usually be perceiveci only through
analysis of measured charactenstics. Sublevel mining is a large-scale mine stoping method
and can be divided in five mains types of stoping methods: Sublevel Caving, Shrinkage
Stoping, Block Caving, Cut and Fill, and Sublevel Open Stoping. Among these rnining
methods, the sublevel open stoping is the most cornmon method used in canada. In fact,
the open stoping method provides signifiant advantages in term of operational
performance. As it can be seen in table 2.1, open stoping method with efficient cable bolt
implementation offer relativeiiy lower dilution (5 to 10 %) with better ncovery (rather
97% of recovery). Howevere subkvel open *ope mining is applied for vein orebodies
wit h the following characteristics:
~Steep dip: the inclination of the foot-wail must exceed the angle of repose of
broken rock by some suitable margn. This is required to promote fke flow of fiagmented
rock to the loading cross nits.
~Strong hanging-wall and fmt-wall: since open stoping is unsupporteci from
inside, the strength of orebody and country rock must be sufficient to provide stable
walls, faces and crown for excavation.
Cornpetence of ore and Regularity of orebody boundaries: the orebody
boundaries must be fairly regular, since selective mining is precluded by the requirement
for regular stope outlines, which are associated with the used of long blast holes.
These characteristics define the nature and the quality of orebody and hoa rocks as
illustrated in figure 2.1 and 2.2. Hence sublevel mining method is suited to low grade
orebodies with regular stope outlines. Rock reinforcernent may be used in particular areas,
but is not required as a routine operation as in cut-and-fili rnining.
Mining Method Ore Dilution (Yo) Ore Recovery (%)
Sublevel Stoping 5 - 10 95 - 97
S hrinkage Stoping 10 - 15 93 - 95
Cut and FU 15 - 30 93 - 95
Su blevel Caving 10 - 15 85 - 88
Block Caving 15 - 10 80 - 85
Table 2.1 : Sublevel rnining methods with their recorded performance (Eustace W., 1983).
Under the need to continually improve efficiency in mining operations, more
sefective is opted while a dilution action plan currently targets solutions to dilution and
cavity monitoring measunng system provides surveys of blasted stopes that can be
compared to planned outlines. As a result, much effort is made on cablebolting process to
reduce ore dilution. Consequently this presents in first stage the sublevel stoping
methodology applicable to the steeply dipping orebody of Bousquet mine. Any
particularity which lead to mo d st op goometries. In the second stage, the use of
cablebolt techniques to minimize the dilution coming fiom wall sides.
- -
Regular hegular to discontinous
1 Vertical cross section
--
Delimitation of orebody
-Easy
1 1 oreboby.
-Difficult: density and total length of
prospecthg holes must be increased.
-
1 Distance betwen sublevels
Length and Orientation of blast holes
-ParaIlel long holes
g and length of bIast holes adapted to
the irregularities and discontinuities of the
the distance between
t he sublevels and
application of bulk
rnining method.
Figure 2.1: Nature of orebody and mining techniques (orebody dipping > 459.
Afier Potvin (1988)
-Short, adapted to the irregularities of
w~sMddiscontinuities.
Ore dilution
- Lo w, i f r o c k
quality is adequate.
As it can be noted above, greater detail rock mechanics information at Bousquet
district mua be used to provide realistic estimates of underground opening design, amount
of support, orientation of opening. If ground control or operational problems should be
-High, for sub-level and bulk mining
methods
-Cm be decreased through selective mining
methods
encountered, modiications could be made. Also 0th- parameters that mua be examined
include:
Geornetry and grade distribution of the deposit
Mining costs and upitalization requirements
Environmental wncems and other site -specific considerations.
Figure 2.2: Infiuence of rock wa quaiity on the choice of mining rnethod. Potvin
(1988)
Basically the method entails providing access to the orebody at various
subintervals between the main haulage levels in order to drill and blast the intewening ore.
Stope drillhg is carried out om driiiing drifts on the sublevels and the ore is blasted in
slices towards an open face, wwhich generally is vertical on the downholes and may be
inclined towards the open face for the up holes. The blasted ore gravitated to the bottom
of the aope and is collected through drriwpoints. Dilution with waste rock may occur if
ore boundaries are irregular or if caving ocairs (that currently happens in weak zones);
but 100% of the ore within the stope usually is recovered. While pillar recovery sometimes
is a problem.
Sublevel open aope mining methods can be classifieci into three methodq because
of local rock quality, dependmg of extraction sense of orebody:
8 Vertiul Crater Retreat (VCR) or Vertical retreat (figure 2.3);
- Transverse Blasthole (figure 2.3a)
- Longitudinal Blasthole (figure 2.3b)
0. Horizontal Retreat (figure 2.4);
0' Sublevel Retreat (figure 2.5);
- Longitudinal Longhole (figure 2.6)
This classification is based on foUowing parameters :
O Extraction direction:
= Longitudinal: advancement of blasting moves towards the extension of the orebody;
= Transversal: generally consist to develop mining in way to advance blssting sequence
fiom a wall side towards the other.
O Use of piliar and backfiii
= No pillar non backfill;
= With pillars but no bacldill (medium recovery, for example low grade orebody);
= With pillars and bacW ( use of primary and secondary stopes).
-
Figure 2.3a: Trmsverse blasthole in sublevel vehcal retreat. Pohiin (1988).
Figure 2.3 b: Longitudinal blasthole in sublevel vertical retreat. Potvin (1 988).
Figure 2.4: Longitudinal longhole mining Horizontal Retreat. Potvin (1 988).
. -
Figure 2.5 : Typical sublevel retreat mining. Potvin (1 988).
O blasthole diameters:
Sm& diameters 50 to 64 mm (2 - 2.5 inch.); usually adapted for longhole methods;
Big diameters of 100 to 200 mm (4 - 8 inch.), for blasthole methods.
2.2 Stabity Aspects in Sublevtl Mining
Analysis to determine rock maso response to rnining is apparentiy rarely
undertaken explicitly in d e design. To understand stability considerations in wblevel
mining, it is necessary to emphasiie upon some indispensable geomechuiicel aspects.
Attention may be concentrated on predicting ground perlormance around blocks of
stopes, and on establishing an extraction sequence which will minimize any risks arising
fiom instabity. Increasing depth of mining, or the need for increased extraction ratios
fiom near-surface orebodies, increws the potential for unstable rock mass behaviour.
Sublevel stoping excavation design is distinguished by uncertainties conceming the in-situ
strength of rock, and fiequently by the need to control an a large s d e the perlormance of
overstressed or failed rock. Sometime adverse induced stress redistribution allow
propagation of cracks formed by overstressing, and the formation of tension cracking or
opening up of pre-existing structures adjacent to faults (currently associated to the ore
zones) showing large defonnation. Borton (1997) found that joints at a low angle to the
wall of an excavation result in pat er waii deflections than joints at high angle. High stress
effeas cm be also apparent in the bored cut out raises and in long blast holes. niese blast
hoIes have to be sometime pre-charged immediately after drilling and several months
before blasting. This has several undesirable effects like desensitization of explosive, cut
offs in the line of explosive and wnsequent hang-ups. In addition, the most undesirable
effect is the premature detonation of precharged holes due to rock movements under high
stresses. In sublevel aeeply ore mining W<e Bousquet mine, smd to luge scale
displacements are induced in the rock mass and the energy is dissipated by slip and
cmshing of wall sides. Then the evaluation of orebody ground conditions should start
before opening up the ground for development. Two important parameters in the
economics of a stoping method for which a rock mechanics study can provide estimates
are the with of the stopes and the size of the plar (Nicholas, 1981). In the sublevel
stoping, the width of the a stope is a finction of the immediate and i n t e d i a t e roof
(Alder and Sun, 1968). The immediate roof is characterized with the pressure arch
concept. In the pressure arch concept, the rock is considered to have maximum distance
that it can transfer the load. The ability of the rock to t r ader a vertical stress in a lateral
direction over an underground opening depends on the shear strength of the rock the
horizontal stress, and the strength of the rock pillars. The maximum stope width is twice
the maximum pressure arch. The pillars spaced this distance must be able to cany
tnbutary-area-load. Joint orientation, spacing, and length can be used to defineci the stope
width. The pillars within twice the maximum transfer distance do not have to carry
tnbutary-area-load, but rather the load under the pressure arch, half way to the next
support.
1t can be concluded that, in sublevel d g severai of the i n~t a~l i t i es in crown and
nb pillars are associateci with the development openings which caused high stress
concentrations. More attention appears desirable in mine planning, and also during rnining
operations, to the possible stress concentration around such openings. An numerical
anaiysis 2D or 3D appears adequate to define zone of possible failure (confined and
relaxation zones). Consequently, to remedy unexpected wall caving, ultimate site
investigation is needed for optimum stope design. The contribution of blast stresses to
pillars failure near opening, and to the diation of pillar and wails generally must warrant
investigations.
2.3 Bousquet Sublevel stoping
2.3.1 Mine final design
Mine design is conceived as a process oriented towards technical and econornical
optimization. Every major design and engi nee~g decision is taken as the result of
technical feasible ahematives. The most relevant vertical design plane of bouquet 2 mine
is shown in Appendix A Actually the mine search about 1350 m depth divided in 9 levels
dong the orebody. Each planneci stopes, now being mined or planned to be mined with
sublevel stoping, can be Jan. A vertical sh& (shafi no 2) gives access to difEerent levels.
Originaiiy it is planned with main levels at 30 m intewds. Main levels for personnel
material transport and hadage connect the shaft with the production blocla. Eventually
the main design parameters to think about in Bousquet mine should be situateci in 4
questions; 1) Sublevel distance; 2) sublevel orientation; 4) ore recovery; and ore dilution.
A plan view of level 7 -O of the mine is shown in Appendix A. Access to ore zone and its
transportation has involved several drift and ore-pass. Much data about geology, rock
mass quality and geostnictural data (discussed in last term topic) are available in this
figure.
Figure 2.6: The mining sequence for the stope 9-1-15 stope (Hyett, et al. 1997)
2.3.2 Bousquet 2 stoping
At Bousquet the ore zone is a steeply and dipping (80")ein of disseminated
sulphide. The shape of this orebody , its geometry and geomechanical characteristics of
host rock at Bousquet district, impose its extraction by Sublevel Open Stoping Method.
The open stope rnining referred to here is the rnining of large opening by non-entry mining
methods. This rnethod of mining is initiaiiy development intensive, however, this is
cornpensateci by more fiexibility (combination of other mining methods if necessary) and
0
of appropriated equipment. At the beginning of stoping, an access is developed in the fiil
length of the planned st op (15 m); because to be able to drill production blasthole. The
diameter of those hole is 10 cm; and their pattern is 2 meter (fwdeau) by 2.5 meter
(width); they are drilled parallel or in eventail dependhg to local shape of the orebody.
Also the length and the orientation of each blasthole is adapted to the l o d shape of the
orebody. Each stope have a rise hole with a diameter of 1.2 meter used as initial opening
for the need of effkctive blasting. Generaliy, the explosive used for production blasting is
ANFO for the blastholes situated in the center as well as those situated close to the
footwall; while, blasthole dong the hanging wall are loaded by fairest energy explosive.
This blasting design reoult fiom s M c controlied blasting, and is used to rnuumued
. . .
ore
dilution coming from the hanging wd. For efficient productivity, vertical bu& rnining is
used for single stope. But depending to planned stope sue, mining of the stope may be
made in seved blasting sequences the Figure 2.6 show the mining sequence for stope 9-1-
15. Notice that stope number is established using number of the main level, the number of
the sublevel foilowing by stope number. Sometimes, if geological and geomechanical
permit, largest stope can be planned and extracted. That is the case of stope 7-1-8 where
2 1 3 00 tonnes of ore materid are mined. Mining such a large stope has necessitated five
blasting sequences with much that 2 445 meters of blasthole driiled, as shown in figure
2.7. These blasthole are grouped in 13 ring (R-1 to R-13): characteristics of blasting
sequences and patterns (type of explosive, weight of explosive charge, Basting delay) are
show in Appendk A. Usually the first three sequences are upwards blasted, that create
fiee space necessary for efficient blasting of the next sequences. Next sequences are
venical buik blasting sequences implying larger volume of ore mucked. This process is
made somewhat the removal of the entire stope. The stopes must remain open until the
ore is removed and the opening backfilled. This sublevel vertical retreat leaves the
possibility that any major dilution fiom the walls will fall into mined out area. It also
allows for si11 or rib pillar to be quickly established if ground conditions become
unfavorable. Blast holes are long-holes drilled fiom upper level and its pattern depend of
local ore zone configuration. In regular ore zone blast holes are drilled rather underlying
level, parallel to the waU; therefore inciined at 284' downwards. While in partiailor
irregular or large ore
I
Figure 2.7: Design of Bousquet stop blztsthola and blasting sequmces
zone, blast hole pattern are disposecl in fan of 16*. For this a relatively high powder
factor and drill factors are required in order to minimize wall Mure mechanism and to
ensure good fiapentation. The mining longitudinal sequence is east-west. As at Hemio
Gold mine, the development of cable bolt drifts increases induced instability; hence stress
conditions prevaiiing narrow the stopes coupled to existing development configuration
render the recovery of these stopes hazardous and diicult.
The relative economy of mining in this marner (compared with cut and fi mining)
has become increasingiy attractive at moderate depths with improvements in equipment
for long-hole drilling and ore loading, and increasingly large blasts are being used. The
development of extension steels and special long-hole rockdrills and more recently, the
large-hole blasting techniques, has made sublevel stoping a method of increased
popularity. The complicated and comprehensive development may be seen as a drawback
but is compensateci by efficient ore production. The drling, blasting, and loading
operations can be perfonned independently of each other, offering the potential for high
@
utilization of equipment, and high output with few machine units and operators.
Knowledge of the of the geology, the ore boundaries and a carefiil control of hole
alignment in the long-hole pattern are key factors for the successtiil application.
2.4 Conclusion
The sublevel open stoping remains an improved mining method for Bousquet 2
orebody charactenstics. The presence of several close openings in jointed volfanic rock,
required by the mining laid the challenge of stress and instability distribution dong the
stope walls. Open stope required lot of development works, but because of that many type
of equipment can be used for increasing the production rate. The experiences fiom this
mining method are positive, that involve that the method in the will increase in popularity,
replacing methods as shrinlcage and cut fil1 mining if geomechanical data are not very
unfavorable. However, the use of cable bolts as a pre- reuiforcement method of ground
control must be efficient to minirnize ore dilution fiom st op wall rocks.
Chanter 3: O n Dlution
This chapter presents a study of various Won that may cause ore dilution in
sublevel mining.
3.1 Ovcrview
Ore dilution is the addition of waste rock or non-ore material to the ore during the
mining process. The addition of waste rock decreases ore grade and increases the mined
tonnage for a given geologid reserve. The adverse effects of this remain a major
problem especially in gold vein underground mining- The nature of these types of mines is
such that extraction is always accompanied by the extraction of a certain amount of waste
rock. Sornetirnes the tonnage of waste rock mined is larger than the strict allowable, and
the grade of the run of the mine ore will be lower than the estimateci in-situ grade of the
deposit. For this remion geostatistians refer to dilution problern as ore losses. In fact, if
mine operators expect a grade x and they remver y, there is a loss x-y, a relative loss of
x-Y
'X -
; and it can be said the dilution is (. On tonnage, if they expect x and end
X X
*-Y
up with y, the hannfl gain is =-y, or relatively (. So, ore dilution is an important
X
mining parameter for evduating the efficiency of a rnining method adopted for a particular
orebody. There are four main factors which can result in ore dilution:
Weak ground conditions;
Mining method employed;
Nature of orebody;
Degrae of operation controlled.
There are two types of ore dilution: intentional and unintentional.
Intentionil ore dilution results fiom waste rock within the stope design
bmndaries. Hence this type of ore ddution is planned and the coming ore grade justiq its
extraction; figure 3.1 illustrates ore dilution problern in dope design. The irnprovement of
mine productivity resulted in heavy mechanization and bulk mining methods. The mining
engineer must consider the panmeten aswciated with these choices. The design of
planned dilution is obsemed both in development works (drifng) and in
ORE DiLUTION
='
P l d I \
hm - orebody width
hc - Planned stope width
hr - actual stope opening (width)
Figure 3.1 : Illustration of different types of ore dilution in a production stope in steeply
dipping vein mining (Bourgoin et al., 199 1).
stope mining. In drift development within the orebody, dilution is linked to the size of
drifks imposed by the width of equipment used. Ore dilution resulting from stoping is much
more important, beause stoping remains the main purpose of minkg activity.
Consequently, the design of the stope shrpe, especidiy in narrow vein extraction, may not
inciude more than the strict minimum waste rock (planned ore dilution). While the
incentive for high productivity Jlows for the use of bulk mining and wida mechanical
equipment; it is not simple to arrive at the final stope design in vein orebody rnining.
Figure 3.2 shows the intentional (planned) and the unexpected additional ore dilution in
stope mining. It takes into m u n t the addition of certain qmti ty of waste rock in ore
reserve calculation; consequently this expeaed addition of waste material is due to a
combination of ore deposit discontinuity, its configuration and poor rock mass quality.
Mineralisation
Intentionai Dilution
Unintentionai Dilution
- - -- - - - - - -
Figure 3.2: Example of stope des i p showing intentional and unintentional ore dilution in
sublevel mining.
Unintentional ore dilution is the unexpected addition of waste rock not included
within the stope design. The impact of this unintentional addition of waste rock is that it is
rnixed with stoped material and produced as ore. The unintentional dilution is hence, an
additional ore contamination which could be mainly due to wall rock caving. Figure 3.3
illustrates the sources of additional ore dilution. As can be seen, the geomechanicd aspects
pay a major role. In general, main parameters to consider are:
Wall rock qualities: competency of hanging and foot walls. Tendency of the walls
to cave or slip, depending on stope geometry and its production (blaswig effect);
State of stress: mining process involves induced stresses which directly or
indirectly lead to rock rnass fiacturing and displacements throughout the zone of
influence. This implies locai Uutabity of rock around individual stopes and other
excavations.
Nature of orebody: minhg of narrow vein is the challenge against ore dilution.
Also, orebody local discontinuities and irregularities remain critical. The
irregularities of the orbody make it difncult to mine; while orebody continuity is an
essential element of its quality control. Deposit estimation depends on continuity
appraisais at two levels: 1) definition of the geometry or physical fonn of a
mineral-bearing, geologicai structure, and 2) spatial variabiiity of a value or quality
measured (e.g. grade, thickness) within a rnineralized zone. Another aspect linked
to orebody nature is the facility or not to identi& the ore zones directly in the
mining field.
Other causes are diiectly linked to blast induced damage, bad s t op sequencing
and lack of good knowledge of the orebody. As un be observed, geomechanical aspects
are the main sources of unintentional ore dilution. For this rason the present thesis is
mainly focused on rock mechanics aspects to understand unexpected ore dilution. Stoping
in fair ground conditions implies to process, at al1 mining levels, for additional, yet useless
tonnage. Two other mjor puameters influence ore dilution: choice of stoping rnethod
and mining technique or operational controls. Large deposits with relatively
homogeneous, mineraikation are often mined with stope excavation sized srnaller than the
deposit dimensions. Intentional dilution may be included in the recovered stope; but
unintentional dilution is fiequentiy excluded in this instance.
-- - - -
Geomcchanid AswtS:
- Qu aiity of wall rocks
- State of stress
- Nature of orebody
- Stope design /geometry
- Blasting
- Stoping method
- Haulage of backfill materia
Figure 3.3 Causes of additional ore dilution (Planeta, note course, 1992).
For example, dilution may take place as result of hauling with backfill during the recovery
of pillars adjacent to nIled stopes, stoping adjacent to physical grade boundaries.
Knowing the tonnage of the mucked rock and its appropnate grade extracteci is
not simple. An oversight in orebody grade dculaion is the omission of mineable grade
and Unplicitly the mineable tonnage. In effect orebody estimation is based on geological
basic idonnation. Data provided by geologid investigations give only the average grade
of the whole o r My , and geological tonnage (calculated geometrically). Often the ore
pr odud from rnining process has, its own grade and tonnage values which are different
from geological estimation. Mineable grade and tonnage include intentional dilution and
are estimated by adjusting the reserve outline to reflect the matenal within an economic
cut-off grade. Gracie and tonnage found ore imposed by the mining method and other
parameters linked at them (e. g. rock mas quality, stope design, equipment, etc.). The
effects of these parameters involve both dilution and reduced recovery rate. The ore
recovery is linked to the Iost of ore in the stope:
(~lanned tonnes - Ore lost in stope)
%Recovery = [ 1x100
Phne d tonnes
On the other hand, dilution is linked to the decruvie of ore grade; the amount of matenal
extracted wil be larger than estimated to obtain the sarne equivalent metal content, and
the grade of the run-of mine ore will be lower- This can be expresseci by:
Waste tonnes + Caving tonnes
00 % Dilution =
Planned tonnes 1
For the economic aspects, ore dilution should not exceed 10 % (lower is better); while
rnining recovery mua be ranked between 65 % and 85 %, depending on the type of
deposit and the mining method employed.
In the geologicai context, the grade of the ore deposit is the value of ore taken
inside the orebody; and the configuration of this (orebody) is found by considering the
mineralized zone which has concentrations above the cut off grade of the desired metal.
Hence rock matenal around the orebody can be completely waste or may have some metal
concentration below the cut off grade. By dennition, material below the econornic cut-off
grade wl surround the econornic mineable reserve. Material below ait-off grade wiil al=
be included within this reserve as non-mineralued or lower grade inclusions resulting from
geological or structural effects. If this material cannot be rnined separately, it mua be
included in the calculation as intentional dilution (sa figure 3.2). Consequently final
dilution (intentional dilution + unintentional dilution), with some concentration of desired
metals, is not much wonying (rather some proportion). But for the archean vein orebody
of Bousquet, v o l d c rocks which are in contact with the orebody, do not have gold
concentration; in tiiis condition any unexpected addition of host rocks has considerable
negative consequences. However, using cavity Monitoring measurement combined with a
knowledge of rock propehes, ground conditions, and previous expenence with the
stoping method, an estimate of unintentional stoping dilution can be made. Any grades
attributed to such material would result fiom its anticipated source. Ore dilution due to
waste rock caving fkom the hanging wall or other source may be added as a percentage of
the designed econornic mineable reserves or as a constant related t o the stope width from
footwd t o hanging wd. The estirnated unintentional dilution is added t o the mineable
reserve to give the total tonnage mined from a aope.
3.2 Factors causiag ore dilution in sublevd mining
In sublevel rnining, the nature of the orebody and several processes involved t o the
extraction of ore zones are sometime the sources of ore dilution. Dilution due t o the
nature of the orebody is called structurai ore dilution. This dilution is inherent in the
occurrence of the mineral deposit (Wright, 1982). Two examples are: - the presence of
several barren formations within a deposit in such a way that selective mining is not
possible; - in-situ leaching caused by penetrating underground water (usualiy the case with
some copper and uranium ores). In vein and lense orebodies, structural dilution can be due
to irregular ordwaste contacts, but may dso, be due to the ripping of waste rock t o
provide sufficient working space in the stope (intentional dilution). On the other hand,
production dilution is generally unintentional dilution and has two sources: - in stoping
operation, wall waste rock can cave in the stope and dilute the broken ore; - and in filling
operations, the mixing of the filling matenal with the broken ore. Production dilution is
more unexpected, wnsequently, the understanding of the effas causing related ore
dilution should be usenil. There are rnainly five eEects leading ore dilution in sublevel
stoping: the effects of ground conditions, effects of in-situ stress? effcts of stope
geometry, effects of geology and the effects of blasting.
3.2.1 ERects of ground conditions
In open st op mining the most important aspect of ore dilution control is the
knowledge of ground response to mining. This implies the requirement of some
knowledge of rock mechanic properties in compression, tension, and shear, under the
numerous conditions of loading which may exist and also of the various environmental
factors iduencing the ~tmigth/stress relationship such as confinement, relaxation, and
structural conditions of the rock mas. For their needq mine operators could refr to the
classifications based on these physical properties, rock mus quality, and evaiuate the rock
mass behaviour under ground stress conditions. Data of rnany type of rocks are available
in rock mechanics literature. Particularly, the instability risks in Abitibian jointed rock, is
hence due to poor rock mass quality (see table 4.1) while the wall behaviour is closely
linked to joint systems and joint conditions. Among the rock mass properties which are
most relevant to be considered are: elastic modulus, Poison's ratio, Uniaxial Compressive
Strength, tensile and shear strength, cohesion and density. The rock in this felsic volcanic
area has medium to high strength with Elastic Moduli in the order of 43 to 100 GPa and a
Poisson's ratio ranging fiom 0.16 to 0.30. The local instabiiity of rock around individual
stopes are hence controlled by a combination of the rock properties and constitutive
stress-strain relations.
The effcts of near-field ground performance is controlled by pillar design, while
the stope peripheral rock pefiormance is controiled by the stope design itself (cf Section
3.2.3). The near-field ground wntrol is achieved by the development of load bearing pillar
between the production stopes. The instability inside the open stope is usually provoked
by unfavorable rock properties (low tensile strength, low whesion, etc.) and unfavorable
rock mass nature (structural discontinuities, soft rock, etc.). The rock property affect on
stope wall instability cm be understand using numerical modelling; while discontinuity
approach is based on field obsavations. Hence, poor ground conditions mean inevitably
the poor efficiency of piilar (transverse and longitudinal) left between the dopes. The rock
mass displacement generated under poor performance of pillus can be the sources of rock
instability dong the hanging waU and the footwall. Then, the ore dilution coming nom the
wall caving could be Linked to rock mass self-support capacity. The instantaneous
weakness dong the st op w d have its source in the fdures of piilrus which must sustain
the irnposed States of stress, and may result in the extensive coliapse of the adjacent near
field rock. If the volume of the unnlled mined void is high, the risk is that collapse may
propagate through the piiiar structure (Brady and Brown, 1985). Consequently, the design
of stope and pillars is very significant in open stope mining (cf 3.2.3); its design must take
account of ground conditions in order to rninirnize dilution problems.
3.2.2 Effcts of in-situ tresses
In Abitibian jointed rock mines, the vertical stress approaches the ovehurden load
while the horizontal stresses generdy exceed the vertical ones (Arjang and Herget, 1997).
The majonty of maximum and intermediate principal compressive stresses (61 and 63 are
aligned in a horizontal to sub-horizontal plane. The magnitude of principal compressive
stress increases with depth (Figure 3.4a). The UIcrease of vertical stress component with
depth is shown in Figure 3.4b. While the ratios of the maximum~minimum horizontal
stresses show a large scatter at shallow depth, such scatter decreases with depth exceeding
about 1000 m (Wang and Herget, 1997). The relationship between stress ratios and the
configuration of mine openings suggests that most mining excavations are subjected to
high stress conditions.
The instabity associated with stresses, in open stoping Iinked to post-mining
stress distribution. There are four stress-related factors which may influence ore dilution.
those are:
1. Stress concentrations and relaxations around mine stopes;
2. Stand-up t h e of the rock mass subsequent to a production blast;
3. Magnitude and orientation of in-situ stresses;
4. VeRicai stress due to gravity which is proportional to depth. The rock
elasticity irnplying a resultant horizontal mess.
Figure 3.4: A) increase of principal compressive stresses with depth; B) Increase of
vertical stresses with depth in Abitibi shield rock (Arjang, 1997).
As can be anticipated, ore dilution problems due to stress conditions emanate
rnainly mong other things h m shear sliding failure dong set of inclined joints (associated
with relaxed zone); or rock cmshing in the confined stress areas (stope back for example).
Figure 3.5 illustrates the possible post-rnining with the induced failures provoking both ore
dilution and strainburst. Tensile stress associated with the relaxed zone produces induced
failures as weil as the opening of existing discontinuities (joints, faults). As the relaxation
zones occur dong the contact zones (hanging waii and foot wail), the leading uidability
causes ore dilution by hanging wall caving and foot waii sloughing. On the other hand, the
compression stresses are concentrated a the roof and bottom of the stop as well as in the
si11 plars. Thue confincd zones are submitted to progressive or brittle defonnation
dependmg to the rock mass properties. Generally, in these zones, potential strahburst is
caused by rock cnishing or shear fdure. Therefore, it is important to identify highly
stressed and relaxed areas, so that mining engineers can evaluate the mining cycle and
determine the best approach to stop sequencing, or use destresskg or preconditioning
techniques to reiieve the stress.
I
Mining indoced stmsses
I
Stress concentration
(Compression)
Stress relaxation
(Tensile stress)
I
Progressive or brittle
deformation
I
1- ~ i c k rShg 1
- Shear failure
opening of joints and
Il loughing, unravelling,
Figure 3.5: Effects of rnining induced stress in open stope.
According to data on the joint systems provided by geological investigations, and those
from the laboratory and in-situ tests on the joints d w d rock, the failun mode of shear
sliding should be verified pnor to the commencement of mining operation. to prevent the
shear sliding it is suggested that when the joints have d' i i ent shear proprieties, the
minimum shear strength should be taken as the critical shcar strength.
The process of waste rock cnrshing in the stope is mainly linked to the
compressive failure. But this process occurs, usuaiiy in the crown pillars (e. g. in cut and
fil1 minuig). Thus, high compressive stress directly govems the behaviour of the rock
mas.
3.2.3 Effkct of stope Design
The stability of the rock mass at the boundary of the stope is closely linked t o the
stope design. The design of each stope must be optimized as a fiinction of local rock mass
quality. Larger individual stopes generaiiy have high productivity however, at higher risk
of instability. One of the usefi design tools developed for t he dimensioning of open
stopes is the empirid 'aabiiity graph mahod' originaiiy proposed by Mathews et al.
(198 1); and later by Diederichs and Kaiser (1996) to determine stope critical dimensions
(i.e., maximum), expressed in t ems of the hydraulic radius of the roof and wd, to ensure
stability of the excavation. The hydraulic radius (HR) of a stope face is calculated as the
face surface area divided by the face perirneter. This stability of the open stope is
expressed in stability number N' and illustrated in figure 3.6.
RQD is rock quality designation (Deere and al. 1967);
J. is the joint set number from the rock tunneling quality index, Q (Barton et al.
1974);
J' is the joint roughness number (Barton and al. 1974) for the kinematically critical
joint set;
J. is the joint alteration number (Barton a al. 1974) for the kinematically critical
joint set; and
A, B, and C are respectively parameters which are: rock stress factor, joint
orientation f mo r and Gravity Adjustment Factor (see figure 3.7).
Example Input
Data
Face Hvdraulic Radius. HR (ml
Figure 3.7: Stability assessrnent at Bw~q ua 2 mine ushg the modine Mathiew stability graph.
I
Tnie Angle Be- Face d Joint
(Angle a &twacn Polcs) I
3 O 10 20 30 40 50 60 70 80 90
CL
Dip of Stope Face
O 10 20 30 40 50 60 70 80 90
Dip of Criticai Joint
Figure 3.8: Manncr to found input parameters for thc design of open stope by graph method
(Diederichs and Kaiser, 1996).
Consequently, the stability of an open stope is evaluated taking into account
excavation geometty, rock mass qudity, and induced stress. If the rock quality is too low
or if the stope is too large (high HR), instabity may occur, which leads to dilution of
barren wall rock into the blasted ore. If the instability is severe, it can prevent the stope
from being rnined and prevent d e access to future mining areas. Using the above method
and available data, it is possible to evduate Bousquet's stope stability. At Bousquet mine,
the stopes are excavateci in a fair quality rock mass (RQD = 60) with four dominant set of
joints (Jn = 15) which are smooth planar and unaitered (Ja = 1.0; Jr = 1.0). The sulphide
ore has a compressive strength of 120 MPa., while the maximum nduced wall stress
detennined from modeiiing is approximately 24 MPa in 1100 m deep. From figure 3.7 the
Rock Stress Factor is A = 0.45. The cntical joint set foms an angle of 85" with the stope
face (which is E-W), giving a Joint Orientation factor B = 0.2 and the Gravity Adjustment
Factor C = 7.0. Hence, the resulting stability number N' of Bousquet's stope (at level 7) is
60 1.0
N'= - x x 0.45 x 0.85 x 7.0 = 10.7
15 1.0
The hydraulic radius, of Bousquet s t op can easily be found fiom figure 9. The stope
analyzed is 15 m dong ariked and 38 m down dip. The resulting HR is:
A rea
HR =
-
- 38x l5 z5. 5
Perimeter 76 + 30
The associated point (N', HR) plotted on the stability graph as shown in figure 3.6, is
located at the boundary between the stable and the unstable zones proposed by Potvin
(1988). This indicates that a hydraulic radius of 5.5 represents the maximum prudent
unsupported dimension to avoid instability and ore dilution in this stope.
Expected approach of underground mine opts for large open Hope (desirably
unsupported). But according to Dierderichs and Kaiser (1996) probability studies, a stope
design iike at Bousquet mine (which have hydraulic radius of 5.5, and stabiity number of
10.7) wili have a 35% probability of instability and a 20% chance of major caving. Ore
dilution risk associated with these instability and caving for each stope could be
intolerable. Because of this, significant effort must be made for the irnplementation of an
efficient wall support systern (section 6).
3.2.1 Effects of gdogy
Abitibi archean rock mass is not really intact rock; these are consolidated volcanic
rocks, which are jointed and are affected by shear zones and by both major and minors
faults provoked by tectonic evens. Consequently, the e f f i s of geology on the stability are
strongly linked to the nature and distribution of stmchiral features within the rock mas.
Joints are breaks of geologial origin, when grouped in pardel are called joint set, and
joint sets intersect to fonn a joint system. Joints may be open, filied or healed. Jointed hard
rocks are subjected to wedge fdure. This instab'ity o c w s whm a set of joints strike
obliquely across the stope face. The shear zones represent zones of stress relief in which
tiactured surfaces may be coated with low -fiction materials (produceci by stress relief
process or weathering). Like faults, shear zones have low shear resistance but they rnay be
much more diflticult to identifL visually. The faults are recognised by the relative
displacement of the rock on opposite sides of the fault plane. They rnay be pervasive
features which traverse a mining area or they rnay be of relatively limited local extent on
the scale of meters; they often o m r in echelon or in groups. Their thickness rnay contain
weak materials such as gouge (clay), fault breccia. The ground adjacent to the fault rnay
be disturbed and weakened by associated structures such as drag folds or secondary
faulting. These factors result in faults king zones of low strength on which slip rnay
readily occur. The spatial disposition and orientation (dip azimuths) of these
discontinuities relative to the f'aces of excavations have a dominant effct on the potential
for instability due to falls of blocks of rock or slip on the discontuiuities. At Bousquet
mine the main behaviour of the rock rnass is controlled by weathered foliation. In fact, this
rock is highly foliated and as seen in the table3.1, there are 4 different types of joints
identifiable by their characteristics.
Structure Dip Direction (degree) Dip (degree) Joint spacing (m)
D 1 (foliation) 185 85 0.01 to O. 1
D2 95 45 to 85 O. 5
D3 270 20 to 60 0.5
D4 (sub-horizontal) 345 5 to 25 1
Table 3.1 : Typical joint sets at Bousquet mine ( Henning, 1997)
Instability causing dilution problem is also preocaipying in sob and sedimentary
rocks. Soft and soi1 rocks are very weak and are very heavy jointed or broken, as in a
waste rock dump; hence the dilution is provoked by cailar Mures which are dehed as
single discontinuities but tend to follow ccular failure paths. On the other hand
sedimentary rocks are formed by severai strata separated by bedding planes. These are
generaily immediate hilure planes because on these planes shear resistance would be
purely fictional.
3.2.5 Effects of blwting
High production extraction in hard ore requues (at the present) high energy of
explosive. This blasting mechanism produces vibration energies, a major part of which
fragments surrounding rock material while a residud part travels beyond the far zone
before it finaiiy drops below the background noise level. in hard vein blasting, k e at
Bousquet 2 stopes, the vibration level can be determined by observations (by camera
andlor monitoring systems) of darnage dong the wall faces. Under fair or poor ground
wall conditions, the scale of blast damage could be translateci into induced fdures, and
even sometimes, wall rock overbreak (which means ore diiution). To keep the mine
profitable, blasting proceu needs to be pdormed in a more efficient way to minimize
ground and induced structural vibration. On the other hand, the dilemma is that the power
must remained high enough for maintaining good the agmentation product; because the
size of blasted muck will control ore haulage. Therefore, blast engineer must refer on the
maximum and minimum ground vibration velocity for many combinations of interhole and
interrow delays. The spectrum of the blast sequence is the result of Fourier transfom.
Effective delay times for each blast hole within multihole blast can be calculated by the
formula:
where
XY and Z are coordinates of the vibration monitoring station
V is the propagation velocity of the seisrnic waves,
Tpi is the cumulative nominal (pyrotechnie) delay the, and
Ti is the cumulative effective delay time.
MuiUnllation of blast vibration is restricted by the production requirements, such
as good fragmentation and muck pile size. This means short interhole and interrow delay
times are not suitable. On the other hand, lower blasting energy condition, the rock
fragments remain too large for efficiency haulage; while highest blasting energy can
produce large waste slough in the aope. In this way, it is important therefore to predict
likely size and the risk of oversize sloughing from the wall. An other option in dipper veins
blasting is the stabilization @y pre-instaiiation of cable bolts) of the hangingwall and
footwail, and then increase the blast vibration energy to obtain the desired fiagrnent size.
In archean jointeci, it dmost impossible to achieve a perfect design of the nope (minimum
dilution); the hgrnentation is controlled by the distribution of jointing in the rock mass.
Consequently, the control of ore dilution wiii refen to modify the type of explosive or the
placement of particular explosives such that the damage zone created by the perimeter
holes does not aceed the damage zone produced by the holes within the perirneter. This
process is called smooth wail blasting or cantrolled blasting largely experimented by
1 Change of bluting 1
parameten
- Delay of m g
- Patteni of blastimlts
- Charge wcigtb
- Type of explosive
Explosive
Cbrrrctccristics I
- Wave propsgaicm
- Vibration cncrgy h
- Vibration velo&y
- Effective dcnsity
I I
Rock m m Physical-
Mecbanid Propcrtk
- Elasticity
- plasticity
-stiffirtss
- Density, etc.
1 Peak Particle Vdocity 1
PPV = k-
1 1
1 - On dilution 1
Figure 3.10: Assessmtnt of controlled blasting design against ore dilution
caused by overbrcak in sublcvcl open stope.
CAMNET in drift development, LUotte (1996). The drift for experiment, shown in Figure
3.8, has about 2.4 m by 2.4 m u face dimension with 36 t o 43 holes drilled t o a depth of
2.4m. Even onguial objective is to assess a suitable controlled blasting techniques for
underground development drift, the damage cnteria used could be validly adapted for the
control of dilution. There are direct relationship between actual shape of the openhg and
the level of blast vibration. To understand that, it requin to monitor the blastholes. For the
case of the Canmet experiment, the blastholes were monitored with a multichannel blast
vibration monitor, to which was attached sets of hi@-fiequency geophones and
accelerometen. The blasting darnage level is linked to both the rock rnass physical-
mechanical properties (elasticity plasticity, etc.), and its structural components (Joint,
foliation, etc.) as weU as the power of explosive used. See Figure 3.9. Under fair or poor
ground wai conditions, the scaie of blast damage could be translated into induced failures,
and even sometimes, wall rock overbreak. ControUed blasting is considered a better way
to handle this problem. The blasting of the stope involves two aspects: The explosive
Characteristics (Wave propagation, vibration energy, vibration velocity, ef f dve density)
and the rock mass physical-mechanical properties (elasticity, plasticity, sWhess, density
etc.). The interaction between the explosive material and the geological material creates a
Peak Particle velocity (PPV) which puts in evidence the behaviour of the rock mass under
given blasting parameters, then the blast damage and induced failure are assessed based on
the PPV, and the reflective tensile wave. In most cases, the overbreak causing dilution is
caused by these blast darnage and induced failures. However, it is possible to reduce this
problem by controiled blasting. Controlled blasting also calleci perirneter blasting result s
fiom various combinations of blasting design parameters. It involve the foilowing
components (single or in combination): modification of the firing sequence; the use of
different explosives; modification of the explosives placement procedures; and the
modification of the geometry of the volume where the explosives are placed. Also the
blasting parameters could be changed (fiMg delay, blasthole pattern the effective density
of the explosive, type of explosive. Generally a suitable result is found after various
combinations of these parameters have been tried out. Naturally, inelastic rock material
has brittle behaviour under dynarnic strain of the explosive. Because of the non-
homogenity of geological material, the defonnation is rather plastic and the lirnit of such
behaviour correspond to the zone where the strain level has exceeded the strength of the
6
rock mass. From the classical elasticity Hooke's law (linear defomation = -) , Holmberg
E
et al. (1984) have demonstrated this deformation is proportional to the strain of the
explosive:
where
E is the strain
6 is the stress
E is the young's modulus of elasticity of the rock.
v is the vibration velocity, and
c is the wave propagation velocity.
The problem of controied blasting is considerably mmplicated; most people focusexi their
attention only on the explosive parameter, forgetting the rock mas itselt However, as is
evident fiom the above relationship, the rock mw physical-mechanical properties plays a
great role in controiied blasting. Thus rock ma s can be considerd darnaged when it no
longer reacts elastically and the defonnation is plastic (Lizotte, 1996).
At Bousquet mine, several triaxial geoghones are used to measure the vibration
caused by blasting. The measures have been done for 2 isolated stopes (one is primary
stope while the other is a secondary stope) in level 9. This study was made in 1997 by
Bousquet mine engineers (J. Henning, P. Gauthier, M. Ruest and C. Provenche). The
amplitude of blasting vibration was determineci using the field constants for the mine and
the quantity of explosive loaded in the blastholes of 9-1- 15 and 9-1- 1 1 stopes. The
formula of PPV have been used to detennine the constants "k" and "a" which depend on
the structural and elastic properties of the rock mas . "K' is an indicator of scale of the
charge; and "aS7 is an indicator of vibration attenuation. The amplitude was found to be
lowest in the secondary siope than in the prirnary stope (table 3.2). That is not surprising
because there is highest attenuation of vibration at the relaxation zone in the hanging wall.
Table 3.2: Results of the blasting vibration indicators obtained ( HeMing et al., 1997).
Total database
Data for AMEX only
PNnary stop 9-0- 1 5
K
498
283
Secondary stope 9- 61 1
a
1.19
0.66
K
498
283
a
0.70
1.30
Figure 3.10 Blast vibration in the hanging wsll at Bousqua aopes (J. Hennuig et al.,
1997,).
Table 3.3: Results of overbreak measurement in Bousquet stopes hanging wall (J. Henning
et ai., 1997,).
Primrry st op
Average
Secondrry Stope
Average
Hangingwall
4.0 m
3.1 m
2.0 m
2.2 m
-
Footnill ,
0.5 m .
0.5 m ,
1.5 m ,
1.0 rn
In Figure 3.10, the curves of explosive quantity useci versus maximum blasting vibration
predicted, show the effct of these constants (k and a) calculated for prirnary and
secondary stopes (Henning a al., 1997,). About the h t five meters around the blasthole,
vibrations higher than 1000 d s e c have been measured for the pr i mq stope. Vibrations
calculated for the secondary s t op are relatively lowest. This is certainly one of reasons of
higher ore dilution in p r h ~ stope. To compare the stope overbreak volumes, equivalent
Iinear overbreak factor is used. This is detennined by calculating the area between the
planned stope boundary and the actual geometry of the stope. As can be noticed from
Table 3.3. the primary s t op is affected by more overbregk volume than the secondary
stope.
Therefore, a good approach for reducing ore dilution is to understand the
explosive/rock mass interaction. This means good knowledge of both rock mass and
explosive characteristics, and the design of optimal fragmentation blasthole pattern. The
challenge of successfiil controiled blasting remains still to rninimize damages associated
with the explosive/rock mass interaction while to optimize the fragmentation produced by
explosive energy.
4.1 Introduction
The presence and the distribution of failures which are responsible of ore dilution must
be extensively considend before any masure taken against the problem. Any underground
opening create varieci degrees and magnitudes of stress in both the near and far field area.
Thus, the stoping operation king the essence of mlliing process, put in evidence the rock mass
stability behaviour within the orebody and in the hosted rock This chapter describes the
response of stope boundary conditions under effects of stress redistribution in open stoping at
Bousquet mine. Numerical modehg techniques such as e-2 tools is proven to be effective in
the analysis of these types of geomechanical problem. This software of numerid modelng
(created and developed at McGiil University by. Mitri, 1992) using the finite element method
to divide a representative zone into an assembly of discrete, interacting elements (meshes);
therefore, to dehe problem domains surroundmg an excavation (a st op for Our case study).
The mechanical behaviour of rock mass is evaluated in t em of displacement, deformation,
while, the stresses and the strains in the vicinity of the opening.. As discussed by Brady and
Brown (1985), the assumption in such method is that transmission of internai forces between
the wedges of adjacent dements is represented by interactions at the nodes of the elements.
This geomechanical study is used to analysis the effects that a large open stope has on the
stability of the dope wds and on the local stress field. Therefore, the main objective is
improved understanding of ore dilution involved by wall instabity behaviour under fair ground
conditions.
The stope site selected for the numerical modelling malysis is located about 1 156 m
under surface. Figure 4.1 shows a two-dimensional view of this stope. Material data such as
rock mass characteristics and properties and in situ rock stress are obtained fiom previous
studies carried out by Henning a al., (1996). In order to achieve a realistic analysis of the
behaviour of the hangingwai and footwall, the influence of structural geology is used to
0
provide detailed information.
4- 1
Figure 4.1 : F i sketch of zwc mode1 fa c-z tods modclling
4.2 Geomechanid data
The numerical analysis methods are mainly based, in addition of in situ stress, on rock
4-2
mass quality and properties provided by geomechuUd data. In the others words, any
numerical analysis is not possible without a &a assessrnent of rock maso classincation of the
concemed field. Thus, the rock mass classifications have been assess to evduated its quality
and its behavioural characteristics as a whole under various stress conditions. The strengh and
failure (displacement and deformation) characteristics of the rock is govemed by the
characteristics of the naturaiiy ocairring hcture systems that ocair in the rock mas . The
rock quality is assessed from its containhg in dismntinuity (RQD) and associatecf parameters
such as compressive strength, joint spacing, joint condition and orientation, and ground Hgter
conditions (RMR and "Q System"). Rock properties are assesscd fiom laboratory tests
providing rock strength values and deformation properties. The rock ma s classification studies
at an area like Bousquet involves firstly, investigation at two levels or scales: The identification
and mapping of geological structures that can affect stope wail behaviour. Therefore, the
starting point for the development of an engineering understanding of the rock mass structure
is a study of the generai regional and mine geak~gy as determinecl during exploration. The
information founded (such as bedding, foliation, folds ac.) constitute an essential background
of geomechanical study of the mine. The second level is fbrther studies involving underground
exposures, and logging core of boreholes (RQD and RMR) drilled for this purpose. This gives
precise locale ground quality. The investigation for such study had been made in the level 7-0
at Bousquet 2 mine, the result is located in the Appendix A. The rock mass rating (RMR) and
t he Q-system show a fair wall condition (see table 4.1 below), whereas the orebody is
relatively good quality.
Massive ore Footwdl
Bieniawski
1976
1989
Modified Q'
Table 4.1: Rock mass quality modefied Hoek & Brown critenq H e h g et al., (1996)
RMR
50
50
1 .O0
GSI
50
45
44
RMR
65
66
1.00
GSI
55
50
44
GSI
65
50
44
RMR
55
59
1-00
The knowledge of pre-mining stress nature and their orientation is fndamental to
coherent prediction of the response of the rock ma s t o rnining activity. For dilution problern,
the attentions are focused on the displacement field generated in the boundary of the stope and
eventually in the stope previously bacldilled. At Bousquet the pre-mining ground stresses have
been detemineci at 900 meters depth. This had been done at EUiot Lake Laboratory by B.
Arjang (Canmet 1988). The results have indicated ttiat the maximum and the intermediate
principal compressive stresses are horizontal and are oriented in northeast and southeast
directions. The minimum principd compressive is oriented vertidy. The orientation of the
principal stresses relatively to the existence of the pariicular fabnc element of the rock mass
such as foliation or schistosity, can also help to understand the direction and the mode of
deformation under stress conditions. The orientation of the principal compressive stresses and
the overail trend of the ore zones are illustrated in the Figure 4.2. The average horizontal
compressive gradient amounted to 0.0505 MPa/m, with a ratio of odeHmi n in the range of
1.5. In relation to the ore zones, the a- acts perpendicular whiie aii, is oriented parallel to
the ore zones strike and the main stnictural features. As the in-situ stresses are fbnction of
depth, they can be deterrnined as foliowing:
a- = 0.061 * D
a-= 0.041 D
av = 0.027 D
where ov is the vertical stress; and D is the depth below surface (m). The stresses are
expressed in MPa.
Rock mass properties required for input parameters used in numerical modelling are
determined fiom laboratory and in situ rneasurement. These values are supptied by mine
Engineers. Note that fractures may be dominant influence on the strength of the rock mas.
Table 4.2 gives the elastic and Hoek & Brown parameters of Bousquet rock mas . Uniaxial
compressive strength (UCS), young's rnodulus and poisson's ratio are obtained nom uniaxial
testing of BQ size con. Under uniaxiai testing almost al1 the ore samples undenvent brittle
failure. This is characterized by a sudden release of mergy in an explosive manner. The results
of these tests are adequately supporteci by previous research by Wawersk and Bnce (1971).
Hoek and Brown (1980), Jaeger (1960).
4.3 Statemtnt of Problem
The adequate approach of efficient modelling involves two requirements: 1) A better
choice of numerical mo d e h g techniques which can provide high resolution of ground
behaviour amund the stope and near field rock in accordance with stope design and the
structural elements of the rock m. Sometimes t o avoid a c h o i q several numerical methods
0
could be used in conjunction, dowing for the advantage of each method and providing a
4-5
cornprehensive analysis nom diffrent perspectives. 2) A better choice of a aope which
represents the design and rock nature of any stop in the mine. The respect of t hes
recommendations are fndmental for getting rezrults that could help to define the type of
behavioun which are responsable of failures and caving. Numencal modelhg is d e d on a
typical stope rnined at Bousquet 2 mine. This s t op is located about 1155 m under surface.
Design parameters are shown in the sketch zone mode1 (Figure 4.1). Data concerning rock
mass characteristics as weU as rock mass properties used in modeiiing, are described in
geologicai conditions section. The dip of the orebody measured fiom plan is about 75" West.
The in situ geomechanical data are shown in table 4.2; while the horizontal-to-vertical stress
ratio (K. = 2.26, Kz = 1.52) are obtained fiom previous publication. Datafile pennitting
mnning of the program of numerical modeiiing (e-z tools) is avdable in annexe page.
1 Poisson's ratio 1 0.25 1 0.25
1 28 GPa 1 28G Pa
1 Bulk modulus 1 43 GPa 1 43 GPi
/ Density
/ Compressive strength 80 MPa
I
/ 80 MPI
1 Joint coherion 1 IMPa
HA3 m
1343 S
Joint fiction
1 Joint tensile strength 1 4 MPa I 4 ~ a
Sulphide Ore
I
2.4
0.007
20 degrees 20 degrees
2.4
0.007
20 degrees
Table 4.2: Rock mass propertes at Bousqua mine, resulting fiom ur iMal and triaxial tests.
4.4 Analysis of results
The analysis of the results is
produced fiom such analysis. Then,
the main reason of the modekg. Many uiformation are
data received are used to explore on stress conditions
around the stope, and understand specific ground fdures conditions, therefore to assess
relevant resolution. The procedure used for solving the problem is to divide the concemed
failure zones into srnalier wmponents in orda to get numerid value for whole behaviour. The
parameters pennitting the examination of rock mass behaviours are evaluated in term of stres
redistribution and stress displacements vestor. For stop design purpose, the existence of both
stress concentration induced in rock nuw and the fdures around the stope is closely S i e d to
the stope geometry and size. For Our dilution study, the location of stress relaxation zones are
taken as failures, therefore, are more important since these zones are submitted to shear
stresses which can provoke block sliding or local caving (tension failure). Otheiwise, the zones
of induced stress concentration area can be affected by crushing in major case by bursting
(compressive failure),. The presence of other close openings and the effect of progressive
mining sequence should be taken account for the both cases.
To have good analysis, 64 nodes have been necessary to create 49zones for the need
of modelling. Arnong these 49 zones, 4 are void zones that consist of a longwail stope and 2
drifts of production and 1 cablebolt driff. Mesh can be seen in figure 4.3. This mesh is dense
near the drifis as wel as the hangingwd and the footwd, to have much memrements in these
critical parts. The principal stress (Figure 4.4) is concentrated at both roof and bottom of the
opening. The open stope mining method used at Bouquet 2 implies a redistribution of the in
situ stress field in two mains zones: a relaxation zones in which an increase of arain energy is
stored; and fields where the stress are conmtrated. The maximum principal stress, located at
corners, is about 243 MPa. The figure 4.5 and 4.6 show vertical and horizontal stress
respectively. Like the principal stress, maximum values occur at the roof and the bottom, in the
corners. However, in these figures it possible to idente the tensional zones dong the hanging
wall and the footwall. The maximum vertical stress is about 127 MPa while tensile force is
about 14 MPa. Maximum horizontal stress is about 192 MPa.
Figure 4.3 : Deformed rnesh for the purpose of modelling.
While there are relaxation of stress dong hangingwali and footwdi; then after, as the distance
increases f?om horizontal and vertical axis of the stope, compressive stresses decrease.
Displacement stress vectors multiplied to factor 15 is view in Figure 4.7. Hangingwall and
footwall stresses are onented toward the void. this is due to the greater mass of the wall
materials. As the distance increase from the stope, the displacement stress vector decrease. The
numencal modelling results include also material det y factor (Figure 6.2), that detemiines the
condition of the material under its state of stress and relaxation. It can be seen that near the
opening hangingwal and footwall are considerably unsafe.
4.5 Interprctation: Stope waU bthaviour
Extraction by longhok open stope method like in Bousquet mine, causes Uistantaneous
and continuous stress redistribution in the surrounding strata. The ability to deal with structural
stability problems in bulk mining and associated dilution and risks to mine perso~el is highly
dependent on understanding stress redistribution during dinerent stages of mining. As dilution
is provoked by slip and caving of waste rock, it can be possible to create iinks between
unexpected loss of grade in ore mucked and stress distribution and rock mass quality.
4-8
Figure 4.4: Principal stress distribution around the stope.
Figure 4.5 : Vertical stress distribution around the stope.
Figure 4.6:Horizontal stress distribution around the stope.
m... . a * . . . . . . . . . . . . . a
Figure 4.7: Displacement vecton multiplied by f bor 15.
4-10
a
Independently to the mining method and operation, poor ground conditions and
stresdshear are most likely to induce hstabiity leading irnmedirrte caving. In this manner many
factors directly contribute to unexpected dilution. These are confineci high stresses which are
concentrated in the corners of the stope, high shear stresses which are extended dong the
walis, and geological discontinuities. combination of these three factors are necessary for
understanding of geomechanical source of dilution at Bousquet mine. At the boundary of the
stope, the zone of relaxation along the hangingwd and footwall (Figure 4.5 and 4.6), potential
rockburst can occur during blasting. As seen in table 4.1, the m&c tuff of hangingwall is
relatively weak quaiity, in act, relaxation dong this cm provide much waste material in ore
mucked. the three factors mentioncd above dlowed to say that there are two modes of rock
mass behaviour leading to rock mws instability. One mode involves crushing of rock occurring
close to stope boundaries. The second mode involves slip on naturai or miring-induced planes
of weakness. In fa- instabiiity is provoked both to structural anomalies and induced stress;
near surface excavations are signikantly infiuenced by structural feature, while instability of
a
deep stopes depend more on the response of the rock mass to the induced stress. A current
notion is that unstable discontinuity slip occurs because the dynamic coefficient of fiction for
the discontinuity (joint, hcture) is less than the static value (Brace and Byertee, 1966). Thus,
excess shear stress on a joint or a foliation plan may be used an indicator of the potential for
unstable slip. Having identified state of stress in the host rock mass favourable to unstable
discontinuity slip, the issue then is the measures to be implementd to prevent or reduce the
moment of the event. The response of rock joint to shear loading in situ depend to a large
extent on its surface properties as well as the boundary conditions that are applied across the
joint surfaces. The boundary conditions can be represented by assuming that the the rockburst
of the rock mass surrounding the joint is modelled by a spring with stifniess k = do. I dv where
db. and dv are the changes in joint normal stress and displacement, respectively (Seb and
Amadei). Also fiastures can form fiom slip dong pre-existing joints. As slip increases along
these pre-existing fiactures, secondary fiactures fonn and link the slipping f i aar es to fom
major fiactures. A high density of fracturing develops adjacent to the slip sufiaces and
increases unstable behaviour of the wd .
The mechanism of stope ore dilution cl orl y results fkom the findamental behaviour of wall
rocks. This mechanism of instab'ility provoked by a discontinuity had k e n largely explain by
Brady and Brown in their rock mechanics book (1985). Then, the presence of discontinuities
within the stress relaxation areas (situated dong the rniddle part of the wall) involve shearing
and splitting, and in the worst case the caving of wali waste rock inside the stope. The
mechanism of unstable slip on a single discontinuity plane has been firstly cansidered by Rice
(1983) and re-explain by Brady and Brown (1985). Otherwise, in strong and intact rock the
dilution problem remains relatively minimum. But this situation still rare in Abitiiian area.
However, intact rock under high confineci stress conditions, can be subject to crushing
behaviour. In the both cases, the mechanism governing the behaviour of the stope faces stiUs
the opening and sliding of the existing andlor induced discontinuities dong and close the
boundaries of the stope. Then, the behaviours of stop during mining means to assume the
behaviours of discontinuities in the analysis of large scale opening. In this instance, Yoshida
and Honi (1997) has developed a constitutive mode1 for analysis of discontinuity behaviours in
8 rock mas , called micmmechaniu-based continuum (MBC). nitiay proposed for civil
engineering (tunnelling), this theoretid analysis could be adapted in underground stope
mining, specially in jointed rock area like in Abitibi mining district; because it attempt to asses
relationship equations between average stress and average strain taken account a
representative volume of element which contains a lot of microstructures.
The numerical results give the whole behaviour of rock mass during rnining process,
and are expressed in displacement in rock m s (provoked by the effect of dominant joints),
joint deformation opening displacement. This later could be the ruson of waste sliding in the
stope, because it is initiated by the stress relaxation and stress redistribution due to stope
opening.
Numerical modelhg has dernonstratcd that Bousquet mine stopes are confionted with
instability problems. These are due both to manner of stress redistribution and poor quality of
wail rocks. Redistribution is expressed by confinement of high stresses in roof and bottom
corners of the stope; whe relaxation and shear stresses occur dong the hangingwal and the
footwall. A major consequence of this induced instability is a high risk of unexpected dilution
of ore mucked. From combination of geomechanical data and amaurd anornelies, it has been
identified that more waste material can slide om the hangingwd.
The mechanical behaviour of the stope w d is dfFicult to accurately predict. It is
govemed by induced stress redistribution (relaxation, confinement), rock ma s propetties
(Young's modulus, Poison's ratio) and rock mass quality and nature (rock mass stfiess, joint
sets orientation and distribution, joint effctive length and quality). However, it possible to
8
adopt a numerical method to analysis the effect of some parameters quoted on this behaviour.
An other possibility is to adjusted the rock mass parameters in function of a proposed
numencal analysis fiom the field failure observations; the inverse in dificult to assess but much
relatively tangible for the control of stabiiity.
Cbbi~ter 5: Mcuurement and Conbol of Ore Dution
The estunation and measurernent of ore dilution takes place d e r the stope blasting
operation. Two mahodo of measurement of ore dilution can be used by mine operators:
the theoretical calculation and the dire* estimation by the cavity monitoring systern. The
purpose of this chapta is to dehe an optimai strategy of stope extraction, integrating
both the geological, geometricai, technical and economical aspects of stope operations.
The use of the theoretical approach to estimate ore dilution is based on the data
provided by geochemicai grade analysis of both the ore and the banen rock at the contact
with the orebody, or the knowledge of the parameters of the designed stope and the real
stope. The approach of the desirable result on the performance of the particular stope is
mainly qualitative; that means the evalrution of the grade of ore &er mining operation.
Other criteria being the recovery and the selectivity of the mining method used. There are
many theoretical formulas which can be used for dilution estimation; we propose here
some of these. Laval University (Bourgoin, et al., 1991) have made investigations on
dilution equation formulation in the setting of its search and development project. The use
of one or another formula depends on the avaiiability of parameters and its adaptability to
the mine stope design. In effkct, dilution can be estirnated at diffcrent steps of the mining
process by the following equation:
where: g,, = g, x (1 - 0.0 Id,)
di: dilution ofore at the ih phase (%)
g,,,: grade of the ore at the entry of i* phase (% or g/t)
B: grade of the ore at end of the i" phase (% or gh).
The knowledge of ore dilution at dinetent phases of the mining process is important.
However, the solution of this problem requires knowledge of the perentage of dilution at
each step of mining procas; and t o take the measures to reduce it at this step. The final
ore dilution which considerably iduences the actual profitabity of the mine operation,
can be detennined by two ways:
i) Using orebody reserve grade and the final grade of ore:
where 6: h a 1 dilution of mucked ore at the entry of concentrating process (%);
g: grade of orebody resewe (% or glt);
gr: final gracie of the mucked ore at the eny of concentrating process (% or g/t).
ii) Using the values of "n" different sources of ore dilution (Bourgoin, et d., 199 1):
where di, d2, di, dn are different sources of ore dilution;
n: number of sources of final ore dilution.
There are usually two sources of final ore dilution (n = 2): planneci and unexpected ore
dilution. Therefore, the previaus equation becornes:
d, = dl +d2 - O.Oldld,
(5.4)
When the situation is pedkct i. e. there are the addition ofonly the planned dilution (which
is far from being reaiistic in vein stoping), n = 1, the final ore dilution is equal t o the
planned ore dilution.
dl = d,
The uncertainty of the detailed shape of the orebody and dii3iculty to deiineate the ore in
place (for some mines) make even the planned dilution difficult t o estimate. However, it is
not acceptable t o extract the matenal that you don't need. Al1 the parameters of the
material extracteci must be absolutely known for each individual stope of the mine. These
parameters are the tonnage and the final grade of the ore. This latter must be known and
controlled. There are two gnphs (Figure 5.1 and 5.2) estabshed by Bourgoin, a al.,
(1991) which can help mine operator to not be disagreeably surprised. As it can be note
there is an inverse correlation between final dilution and the grade of ore mucked. A s md
percentage of dilution consideraly reduces the niul grade of the ore. If the grade of the
orebody is, say, 2g/t, at 25% of dilution reduces the final grade t o 1.5g/t. Also it is
possible to calculate the dilution having planned and unexpected dilution by foUowing
equation:
df =d, +d, - O. Ol x d u xdp
( 5 - 5 )
where
df: finai dilution,
dp: pianned dilution,
&: unexpected dilution.
In most cases the final dilution is greater t h the planned dilution (that can be eady
observed by the change of ore qualitative and quantitative parameters). That meam that
the final dilution hides the unexpected ore dilution which is mainly associated to the
stability of the wall rock. Even when the unexpected dilution (additional dilution) can be
estimated beforehand, it is possible to calculate the parameters (tonnage and grade)
associated with the material added to the ore.
where: gr: final grade of ore extracteci;
ta and g. are respectively the tonnage and the grade of material added;
t and g, are respectively the tonnage and the grade of orebody reserve.
Generally, the waste rock has no desuable metal content (g, = O). Thus it become easy to
find the tonnage of materid added.
Sometimes, the rock added to the ore has little quantity of desirable metal (g, r O). In this
case its tonnage is estirnated and its grade is calculated.
Figure 5.1: Infucace of Dilutioa on Finil Gndc of OR extractcd (Bourgoh et al., 1991).
O L/ I 1 I I 1
O 10 i 20 1 301 40' 501
P l d Dilution, d,, (%).
Figure 5.2: Rclationship betwn P l d Dilution (d, ), Additional Dilution (da ) and Final
Dilution (4) (Bourgoh et al., 199 1).
The addition of no waste rock reduces effkct on the final grade of the ore extraaed.
Therefore its addition in the ore can be tolerated rather some proportions ordwaste
established by mine operators. For that, the operational and economic impacts of planned
dilution mua be evduated. From the values such as pl med dilution, as weii as orebody
and waste rock grades Cg, and g.), it becornes possible t o evaluate:
The tonnage of waste rock provoking dilution by mineable orebody tonnes.
This is called the factor of rock provoking dilution (FR).
when the rock provoking dilution is waste for desirable metals (ga = O), this factor
become:
The whole tonnage (ore + waste) by mineabk orebody tonnes. That is known
as factor of planned dilution (FP):
FP= I+FR (5.10)
Consequently, the mineable tonnage must be determined in adjusting the tonnage of
orebody reserve accordiig to both the ore recbvery (OR) and the factor of planned
dilution (FP). Then the adjustecl reserve tonnage (including planned dilution) is:
Ta = O.O1xORxtr (5.1 1)
The tonnage ofwaste rock causing ore dilution is
ta = Ta x FR
The tonnage of ore mucked To(ore + waste rock causing dilution) is:
To = Ta x FP
The grade of ore (gJ afker planned dilution is:
go = g, (1 - 0.0 Li,) (% or gh).
On the other hand, as seen in the Figure 5.2, the final dilution is directly correlated to the
additional ore dilution. That means the unexpected addition of waste rock is completely
independent fkom the planned dilution, since the latter is only associated with the width of
planned stope.
The effects of unplanned addition of waste rock is expressecl in terms of dmease
of ore quality. Using the results of previous investigations and of other mines, it should be
possible to estimate the rate of dilution. However, because of uncertainty knowledge of
al1 the local associated parameten and their implication it becomes impossible to predict
the rate of final dilution. The theoretical approach nwcrtheless, should help in a
prelirninary estimation of ore dilution; and usuay is adopted before the extraction process
begins.
Mine operators opt to manage th& own production systern. In effect, it will mean
the implication of parameters that can be wntrolled in place. As the main source of ore
dilution is linked to the geomechanicai and design characteristics of the stope, it should be
reasonable to use these characteristics in dilution estimation.
5.2.1 Background
The design characteristics of the s t o p can be used to determine the planned ore
dilution. Fust, it is expressed by following quantitative relationship:
where g,, is the grade of ore produced (% or g/t). This grade can be calculated using the
dimensions of the stope Vigure 5.3)
where
g, is the grade in metal of the waste rock causing dilution (% or g/t);
W. is the width of the stope opened (m);
wb is the width of orebody (m).
To express the dilution as function of s t o p design parameters, we can introduce above
equation in the expression of planned ore dilution.
In most case the grade of the wute rock is ni1 or negligible (gw 4). Thm the planned
dilution is:
As can be noticed, the ratio of nope widthlopening of the orebody (wdw.) is signincant in
the assessrnent of planned dilution. The production management reques a st ri a minimum
planned dilution. That means the smallest possible opening. For that, the narrow orebody
should be mined using hand-held air leg drills, scrapers, andlor rd-mounted loaders.
These methods are rnanpower intensive and whe productive enough to support a srnall
operation are not sufficiently flexible or cost effctive t o suppon a larger operation. But if
the ore has high grade or is concmtratted in sub-products (Au, Ag, Cu Zn etc.), it should
be economi d t o use specialised equipment. in effct, now there are small s a l e
equipment for orebodies under 3 meters wide. However, for large scale mining
(mechanized method), the drift sues mua exceed 4.5 m width. But large s a l e mining for
narrow vein (width at least that 3 meters) means the acceptance of high dilution ratios.
Such methods are typically used on even modest deposits while gold price is high. The
ideal performance is found when the vein width is so large that it is not necessary t o
W
include barren rock in the design of the stope: 2 2 1 .
W,
w, - width of the orebody;
w,, - width of plaMed stope;
w, - actuai width of the
ore reserve;
t,, g, - tonnage and grade of
- Cr* L5r
Figure 5.3: iilustration of parameters used for ore diluon esti
5.2.2: The Cavity Monitoring System and mcuurcmcnt of open stopt
The laser light sectionhg meesurement technique was first apped in 1969 to mine
tunnels by Chrzanowski and El Masri. Later it was used in civil engineering tunnels to
detect tunnel deformations over the t he. The technique permits to capture the image on
videotape and then to enhance and evaluate it on a microcornputer. Now the light
measurement technique is improved and thus has become much more suitable for the
measurement of stope overbreak. In Europe, the prototype used is called Rig onginally
constmcted by Scott, now improved by Cambome School of Mines (Wetherelt and Beer,
1996). It consists of a camera fixed to a tube at the end of this section. The rods to rotate
the light source are mounted on the upper surface of the boom. The second section is a 2
m length of boom with a rotation rod mounted on the upper side. In the cavity (or the
stope), the ght source is rotated slowly using the handle at the end of the rotating rods
and a long exposure picture is taken. Then the light section profiles are taken at various
intervals depending primarily on the stope advance or where there is overbreak or
underbreak that requires quantification. In Canada, the Noranda Technology Centre and
Optech System Inc. have created a prototype called Cavity Monitoring System (CMS)
which is efficient in s t op measurement. This new tool provides to operators signifiant
solution against dilution problem in open stope. The CMS consists of a laser scanning
device that can be insertcd into any cavity or st op and rotated in three planes to
accurately survey the entire stope. The e s t laser system ("Optech G150 Laser
Rangefinder") was successfUy tested at Gaspe Mines in 1990. This prototype is actually
used by several mines in Canada which are ~nfronted with ore dilution problem; the
better example shidy of using the CMS is done at Golden Gant Mine by mine operators
(Anderson and Grebenc, 1995). Mer blasting, the openhg area rem* nsky and must be
splitted and wefilly tested against rock caving. However, in sublevel rnining the access
for men to the stope is strictly forbidden; thus measurement device (CMS) is used, after
ore mucking, fiom the upper level of the stope. The Optech laser is rnounted on a set of
aluminium rods that could be inserted inside the stope where it mua be onented
downwards the opening.
The measurements are automatically taken every four or five degrees and the
device itself is equipped with a core processor for storing the measurement data. Me r the
measurement operation, the CMS device is brought outside the mine to be connected to a
computer which has the Optech software. From this computer the output of the CMS is
produced as 3D AutoCad fiies @W or DWG). From the Optech program wire-fime
mesh files are generated which can be directiy overlaid on the planned stope outline in
sirnilar file format. In this way waste overbreak, backfill sloughage and ore left unbroken
can be easily quantifid by reference to the designed aope shape. Appendix D shows the
detailed results of stope 7-2-SB at Bousquet 2 mine. As can be seen, the efficiency of
Optech program is that it becomes possible to obtain automatically on the sarne plan, the
planned design and the actual boundaries of the stope. Also, we can observe several
vertical levels and longitudinal sections of the aope. The final result of the investigation is
very significant since it provides the performance parameters. It includes the real tonnage
of ore broken, the tonnage of wd overbreak, the tonnage of bacldl sloughage; and
finally (when the density and orebody grade is adabl e) the percentages of ore dilution
and the ore recovery. At Bousquet 2 mine al1 the stopes art m e y d in this fshion, and
must have an operational report which contains both the CMS results and a detailed
description of the st op: drawpoint, ehaviour of openings (deformations, sketch of
observation).
As it has been demonstrated above, in sublevel bulk mining conditions the accurate
stope measurement is an important practice for the efficient running of the mine. There are
then three main purposes of use of CMS :
1) To establish the performance of stope extracte&
2) The understanding of the various components of ore dilution in the stope;
3) The understanding of ground movements (instability problem).
In the past it has been difncult or even impossible to establish any of theses parameters as
there was no satisfactory rnethod of acairately detennining the dimensions of the empty
stope. Now the opportunity offered by this technique can be ascertained with high degree
of accuracy and s t o p desi ps can accordingly be modified t o reduce ore dilution and
overbreak while increasing ore recovery.
5.3 Control Agrinst Ore Dilution
The practical approach of dilution evaiuation and measurement involves a tangible
help for the control of ore dilution problem. However, even i fwe can evaluate the planned
dilution, it remains dficult, even impossible to predict the aaud final dilution due to
additionai dilution. This additional dilution is proved @y numerical modelling and CMS)
as the direct product of both blasting effects and structural failures of the stope
boundaries. Consequently, the only way to handle or reduce this unexpected dilution in
sublevel mining methods is the use of cabte bolt.
5.3.1 Cabk Bolting in subkvd mining
Basicdy, cable bolts may be installeci everywhere in target zones in the walls,
backs and piilars of proposed stopes Born laterai access drives as shown in Bousquet 2
stope plane p i p e 5.4). Udorninately, installing cable bolts in radial fans fiom drives
outside of proposed dopes means that reinforcement dmsity is fairfy low, especidy in the
larger spans. An aitemative method for narrow orebodies is to install radial fans of cables
fiom development sub-levels within the planned stope in Bousquet 2 radiai cable bolt
arrays. This creates bands of heavily reinforcement rock that effectively reduce the rock
caving. Careful scheduling of stope extraction and fiiling is required to minMize the span
open dong strike and to allow the cornplete orebody to be mined with minimum ore
dilution.
5.3. 1.2 Cable bolting Techniques in Sublevd mining
The advent of cable insertion equipment and cable bolting machines means that
completely mechanised hole drilling, cable insertion and grouting is now possible. With
these machines stiner grouts are used and the cable bolts are u W y "plunged" into a pre-
grouted hole. The technology of grouting has also improved fkom a simple cernent-water
mix delivered by a single stage pump to high early strength cements delivered by two-
stage, continuous pumps. The practice of pre-tensioning cable bolts is abandoned when it
is realized that tension quickiy develops in untensioned pre-reinforment in response to
rock mass movement induced by blasting or stress redistribution effects during or d e r
excavation. Variation on pre-tensioning now includes plaMg the cable bolt in the hole,
grouting an anchorage at the far end. fitting face restraint, pre-stressing the cable bolt and
then grouting the balance of the hole. Many operations replace this two-stage grouting
process by pre-stressing with a special expansion shell anchor and then grouting that has
been fitted. Some operations simply install a short debonding tube near the collar, grout
the fit11 length of the hole, attach the restraint system and then pre-tension. When a pre-
tension is applied to a fiilly grouted strmd to hold the face restraint only a small force
remains on the system (Thompson 1992).
5.3.1.3 Cable Bolt Design Procedures
The avdabiiity of geomechanicai data is the fht stage of the procedure. Figure
5.5 (C. Windsor, 1992) schematidy illustrates the choice of an appropriate cable bolt
relative to work man condition. When deciding on the arrangement of the cable bolt array
rnay be made. There are four assessment methods that may be usdiil in a cable bolt design
procedure (Wiidsor, 1992):
1. Empincal methods.
2. Analytid methods.
3. Experimental simulation methods.
4. Numerical simulation methods.
The empincal methods and the analytical methods are usefil in helping to propose a
prelirninary design. They are used to determine the overd stability of the s t op and hence
the need of cable. The experimental simulation techniques and numerical simulation
techniques are best suited to checking designs. Empiricai and analytical approaches can be
combined to suggest an initial design. Therefore empiricai and analyticai methods should
be able to assess rock mass stability, predict and identiQ the type of instability, and the
component mechanisms of instability . The numerical simulation backed up by expenment al
investigation is the ideal approach for the design of cable bolts. Ideally, cable
instrumentation is used to help validate th numencal simulation and improve the accuracy
of future modelling studies of subsequent stopes.
5.3.2 Factors coatrolling cable Bolt performince
This section presents some tools which mine operator can use to base operational
decision and to achieve optimal cable bolt design. Some key factors may influence, directly
or indirectly, cable bolt performance. These factors can be divided in two main categories
(see Figure 5.5): uncontrollable and controllable factors.
Rock Mur
D a m p I C d ~ a I K l ,
intermittent anchors
Plain saand
Plain straad and
intermittent anchors
SOA
weathered
jointed
lm !larcss
static laading
Figure 5.4: The choice of aa appropriate cable bolt type relative to rock mass conditions; (Windsor, 1992).
Mining induced effecw Mine development and the extraction of orebody itself allow
the redistribution of stress field, rock rnass quolity and influence cable bolt capacity by
swing the design point on Figure 5.6. This influence can occur in two ways. Fustly
rnining induced stress increase, if high enough, may cause stress fiamring or slabbing
in stope backs which reduces the ef f dve hcture spacing and hence the effective
cable bolt capacity. This tends to shift initial design points such as A vertically down to
lower capacity such as B. Secondly, rnining induced stress decrease (relaxation in
walls), causes a reduction in confinement, and hence in cable capacity. This shiffs
initial design points such as A to the left to some lower capacity such as C. These
effects are not yet weU understood and further reseach in the area of stress control on
cable bolt capacity is needed.
Figure 5.5 : Cable Bolt performance assessment, (Bawden, et al, 1992).
Rock mas effects: Cable bolt capacity and behaviour are We d to rock mass quality
and properties. Three factors strongly influencing these are: 1) natural fiacture
spacing; 2) confinement as related to rock mass quality; and 3) confinement as related
to in-situ stress levels.
Grout: The performance of grouted cable bolt support is greatly Linked to the type of
grout or the w:c ratio. Hence, depending of the test results, mine operator has the
choice betwem the proposed types or make his own w:c mixture. The grout
waterxement ratio factor is largely dependent on the pumping/mixing system used in
the mine. Nature of the cement and how cement is stored and handled on the site may
influence fUul grout characteristics. Type 10 and type 30 are two commonly used
cements; type 30 is less attractive in practical mining than type 10 because of its
in&eased hygroscopicity, its higher cost and its lower uniaxial compressive strength.
Poor Fair Good Excellent W
Increasing Rock Mass Quality (Q' or RMR)
I I I
I w
10 50 100 200
h md n g Intact Rock Modulus (GPa)
Figure 5.6: E f f i of rock mass and mine induced stress change on ultimate cable bolt
bond capacity (Bawden, et al., 1992).
Surfacefixiures: The use of surface fixtures on the ends of cable bolts is cornmonly
used to secure the ground between cable bolts, it inmeases the bearing surace at the
imrnediate collar of the hole; and to act, in effect, to reduce the negative effect of
short embedrnent lengths. The use of such fixtures should: 1) reduce the span between
cable botls, 2) prevent ground from unravehg up the cable due to short embedment
lengths and 3) prevent rotation of the fiee end of the cable under loading conditions.
Plates and straps are used to secure thin slabby gound, effktively creating a higher
embedment lcngths immediately adjacent to the coUar of the hole.
Cabie bo/r configuration: Geometrically there are many types of cable bolts such as
standard, birdcage and nutcase, etc.. For a standard cable the maximum potential
extent for geometric mismatch between the cable and cernent annulus during cable
displacement is srna11 and is equai to the radius of a single strand. For birdcage and
nudcase cables the potential extent for geometric mismatch between the cable and the
cernent annuius, (iicludig the grout trapped within the case structure), is much
greater because the cage structure effectively increases the radius of the cable.
Theoretically, Fable geometries with a higher potentiai for geometric mismatch should
overcome, at least partially, the negative eEect of low radial confinement. That is,
birdcage and nutcase cables should perfon better than standard cables in more highly
fiactured rock masses. in addition they should be less susceptible to problems created
by poor quaiity grouts. The peifonnance of different cables geometries under dxerent
conditions of rock mass confinement and dxerent embedment lengths is an issue
requiring fiirther research.
Cable bolt spncing d orientation: Cabie bolt pattern 2 x 2 m is most cornmon, in
reality there are no method to reliably determine cable bolt spacing. A theoretical
method suggeas the weight of unstable local rock mass divided by the cable capacity
and then adjua this for some factor of safety. The only justification for reducing cable
bolt spacing would be the observation of broken cables in fdures or raveiiing failures
between cables. For cable bolt orientation it is important to assume that if they are
installed such that rock ma s movements induce shear loading the maximum cable
capacity wiii be reduced. Then, cable bolts should be i d e d perpendiculaw to rock
mass diwntinuity planes.
Analysis of reiforcement by cable bolt can show how the cable works in ternis of
its behaviour, load canied, and reduction of local failure. The primary mechanism of load
transfer fiom the rock to the cable is fictional (Stiliborg, 1984). The fictional bond
resistance and therefore the uftimate bond strengh, s, are diuectly related to the interface
pressure as follows:
s = 5 tan (4 + 11
where 5 is the n o d stress
4 is the bolt-grout fiction angle
i the diiation angle.
However, the mechanics of a bonded cable bolt in jointed rock are not so simple. The
cable bolt instaiied in a hole which is already deformed in response to the stresses in the
rock, is subrnitted to load at dimete intervals dong its length at intersections with
discontinuities. Therefore, the cable bolts act as grouted dowels that provide a passive
resistance to any movements in the rock mass. Where these dowels intersect a joint or
fracture, the tendency for the rock ma s to separate due to tende stresses or shearing
accompanied by dilation is resisted in the imrnediate area of the intersection. If a failure of
the grout bond occurs near a joint, the dowel elongation that then occurs is confined to a
short section of its length. If such a joint sepration was initiated across a point anchored
bolt, the bolt would be tensioned and strained uniforrnly over its entue length, resulting in
less resistance to joint separation. As the role of cable boit is to reinforce rock m a s by
holding mainly, near surrounding discontinuities, it is possible for up to six displacements
to occur. Three rotations and three translations (Windsor,l992). These may produce
cornplex interactions involving axial, shear, bending and torsion toading of the reinforcing
element. As shown in Figure 5.7, even when the displacement is simple, the orientation of
the discontinuity in relation to the reinforcernent direction means that only under special
circumstances are loads simplified to pure tension or pure shear.
Figure 5.7: Schematic diagram of the modes of reinforcing loading at a discontinuity (afler
Windsor, 1 992).
5.3.3 Cable bolting rt Bousquet 2 Mine
In order to rninimize the dilution coming fiom the hanging waii and foot wall, an
important measure involved the development of cable bolt driffs for the installation of
cable bolts. From cabling drift, about eight rings of cable bolts are installed toward the
orebody in advance of stoping. Each comprising a fan configuration of about 9 cable
ranging fiom 18 m to 27 m in length. A single bulge cable with a 12 inch bulb spacing was
placed in each 2.5 inches hole, and was grouted fiom toe to coilar using a 0.35 w:c ratio
grout (M. Ruest et al.). Figure 5.8. Show the cable bolt layout for the 9-1-15 stope. While
Figure 5.9 show the overa vertical section view of patterns of cable bolt reinforcement.
The cable bolts used are standard seven-wire cables with steep strap (biidcage) installed
every 2 rneters; the cross sectional area of the single bolt is 138 mm2, with a modulus of
elasticity of 200 000 MPa. As host rock at Bousquet is very foliated and afGected by many
orogeny events that provided fractures, faults and joints, the objective of cable bolting is
based upon supporting some 2 or 3 m thick zone paralleling the wall. In fact the cables are
applied to prevent separation =d slip dong planes of weakness (foliation and joints) in the
rock mass. hduced rock h c t u ~ g due to stress distdmtion is another prevaiiing reason.
Consequently succeu of these meuws against immediate waste caving are observed by
reduction of ore dilution and the distribution of tensional force in cables appied. One
other reason is the pre-reinforcernent of wall rocks while increasing blast energy to get
better fkagmentation of ore produceci. Presently, the main effon of mine engineen is
focused on qualitative and quantitative cable bolting data for foilowing objectives:
The irnprovement of pre-existin8 cable bolt design and pattern;
The understanding of load distribution dong single cable bolt; and
Quality contrai.
For the purpose of efficiency of cable bolting at Bousquet mine, two experiments were
made. The first experiment is made in order to evaluate the use of cable bolts for back
reinforcement; while the second experiment is made in order to evaluate the use of cable
bolts for hanging wall pre-reinforcement, the benefit of changing conventional cable, the
pattern of cabhg. The results of these tests confirm the success of the conventional cable
(against the birdcage cable and the nutcase cable) which work with oniy 35% of its
capacity while for other cables the mpture occurs at 30 mm of displacement. The results
also show that the cable bolts failed by two mechanisms: i) Bond failure invalving pullout
and, in particular, slip at the cable-grout interface; and ii) tensile rupture of the steel cable.
Consequently the decision taken about the choice of cable bolting depends upon the
results from detailed rock mechanics study which includes stabity analysis and numerical
modelling.
~
Stope 9- 1 - 15
Cable Bolts
I I I I I F-' Rings
87654 21
1 1 1 1 1 1 1 1
9- 1-H/W
Access
Figure 5.8: The Cable bolt layout for 9-1- 15 s t o p in plan (Hyett et al., 1997).
5- 19
Cemented NI
or
rock m
1 Figue 5.9: Hmeias\nll cable bolting in tranwasJ
The need that the cable works against stope ounduy Mures, has entoileci the use
in place of tension measuring device. Hyett et al, (1997). have developed a multi-point
extensometer calied Stretch Measurement to Assess Reinforcement Tension or SMART
cable. The advantage of this technology is that it is possible to install sufficient disposable
instrumentation to understand how the patterns of cable bolts are working, not just a
single isolated cable bolt. The result of tests made for the s t op 9-1-15 with this
technology have show that (Hyett et al):
The movement of hanging wail; therefore the big part of load supported dong
the cable is concentratecl near the boundq of the stope (nom between 4 m and 7m);
The increase of cabk load for each blasting Jequence of the stope.
*The cable pich up load imrnediately f i er the blast and continues to pick up load.
Such behaviour of the stop indicate the readjustment of stress redistribution. The rate of
these events decreases with time as stabity is progressively attained.
The andysis of cable bolt pattern behaviour is discussed in the numerical
modelling section (next chapter). Wtthout t hat, the data provided b y Cavity Monitoring
system surveys have proved the performance of cable bolt implementation. ther aspect of
rock support in the mine is the intensive use of rockbolts and mesh for reuiforcement.
Because the rocks in the mine are highly foliated, aU the drifts are systematically bolted at
about 1.5m distance. In poor ground zones, mesh is combined to the rockbolts.
Exceptionally, where drawpoints are hazardous, cable bolts are installecl in the areas.
Cemented rockfill is also an option used at Bousquet 2 to reduce pom miring induced
stress. The primary stope (stope type A) should be backfilled before the recovery of the
relevant secondary stope (type B).
Chapter 6. Comprrison Betwccn CMS Murunments and Numerid Modclling
Results
6.1 Comprrison
The comparison between the actual stope geometry (CMS results) and the
numerical modelling results is attractive since it permits to dbrate/validate the numerical
mode1 using the field data. The prodicted caving zones given by the modelling are assesseci
from: 1) the tensional zones in the stop walls and 2) the s a f i factor ushg rock faiiure
cnteria. Figure 6.1 shows the distribution of principal stresses around the stope. The
caving zones are identifid as the areas of tensile stress where relaxation dominates.
Notice that tensional forces are paralie1 to the st op wall dong the stope wall, while they
are nearly perpendicular at the upper and lower part of the walls.
Figure 6.2 shows the safty fkctor distribution for a primary stope, according to
Mohr-Coulomb criterion (figure 6.2a) and Hoek and Brown criterion (figure 6.2b). It can
be seen that the zones where the det y factors are below 0.5 closely coincide with the
zone of tensional failure zones observed in principal stress trajectones (figure 6.1). The
rock strengh parameters used in the safety level analysis are indicated on the figures.
Figure 6.3 presents a comparison between the CMS measurements and the numerical
modelling results. Figure 6.4 and 6.5 present similar type of cornparisons while using
Morh-Coulomb and Hoek and Brown fgilure criteria, respectively. Referring to the safty
level results, it can be seen that numerical modehg results correspond more closely to the
CMS results when the tensional stress zone is considerd as a basis for evaluating the
caving zones. For the stress analysis of the secondary stope (7-2-SB), it was necessary to
reduce the field horizontal stresses since rnining of the primary stopes causes the ground
to work (or relax). in order to detennine the appropriate level of horizontal stress
reduction, a sensitivity analysis was carrieci out with the foiiowing scenarios:
Analysis 1: & = 75 % of ho
Analysis 2: & = 50 % of ho
where:
ho represents the pro-mining horizontai stress field which was uscd in the analysis
of the primary stop;
9< represents the pre-mining horizontal stress field for the seandary dope.
In the following, only the results of the stress trajectory distribution are presented.
It has already been established fiom the prbmy s t op aiulysis that this method w~el at es
better with the CMS results. Figure 6.6 and 6.7 present results which are superimposed on
the CMS measurements. For the two analyses wnducted, as can be seen, the first anaiysis
with & = 75 %ho correlates better with the CMS results, since at & = 50 Y i 0 there are
not tension failure in the foot wall whereas a part of this zone is caved.
From the above analyses, the foliowing remarks can be made:
1. The use of Morh-Coulomb and Hoek & Brown critena does not appear to give
reasonable estimate of ore dilution due to wall cavhg and sloughing, i.e. in low
or even tende stress zones.
2. A simple method based on tracing the zone of tensile stresses in the stope
walls, and considering such areas to be potentiaiiy caving, appear to give
reasonable prediction of ore dilution as it agms more closely with the CMS
results.
3. Horizontal field stresses acting on the secondary stope are lower than those
acting on the primary stope. At 25% reduction factor of the original @reg
mining) horizontal stress gave reasonable results.
SAFETY FACTOR
Figure 6.2a: Safety level result using Mohr-Coulomb criterion; primary stop.
fhek and Brown Criterion
SIC,
i ; SIG
SIG,
SAFETY FACTOR
Figure 2b: Safety level result using Hoek and Brown criterion - Pnmsry stope.
z
y Stope o -
w
WEl
caving (CMS)
I
4030 El 4030 El
4020 El 4020 El
Potentidly zones
-
Caving (e-z tools)
4010 El
4010 El
I
Optech Orebody Iknsity: 4 3
St op 8 3 - 14 Wd rocks Dtnsity: 2.8
set i on 8644
Grade of Onbody: 17.6g/.
Figure 6.3: Superposition of CMS and numencal modelling, tension zone results - Primary
stope.
- --
Po tentidly zones
cavng (oz tools)
Figure 6.4: Superposition of CMS and numerical modelling, Morh-Coulomb safety level
results - Primary stope.
t
z
Prim ry St op o
ro
I
Optech
Onbody Density: 4.3
8 3 - 14 WiII rocks Density: 2.8
*tien 644
Grade of Orebody: 17.6gIt.
Figure 6.5: Superposition of CMS and numencal modelling, Hoek & Brown safety level
results - Primary stope.
Secion -
Grade of Orebody: 17.6plt.
Figure 6.6: Superposition of CMS and numerical modeling tension zone results -
Secondary stope - Analysis 1.
-4
Optech Orebody Density: 4 3
St oP 7 2 SB
WIU rocks Density: 2.8
section 644
Grade of Orebody: 17.6&
Figure 6.7: Superposition of CMS and numerical modelling tension zone results -
Secondary stope - Analysis 2.
Notice: Dilution is relatively highest in this stope. Usuaily for open stope rnining
acceptable ore dilution rank between 5 to 15 %. The problem of this stope could be due ta
particular local ground conditions (Highiy foated andor jointed); however, cable-bolt
have been efficient in the uppa part of the hanging wdl since this zone is submitted to
much tensional fdure force. Also the level of induced stresses is relatively highest in
primaiy stopes, that can Sct the stability of those nopes.
6.2 Discussion
The expressions "narrow vein" and "ore dilution" are a natural couplet in fair to
poor rock mass quality area. The decision to mine fosta and deeper wch orebody stiii
presents two inevitable options: to appreciate the lower production cost in high dilution
mining, or to opt the higher production cost in reduced dilution rnining. For choice
between these options, the grade or the value of ore is the goveming factor. Highest grade
orebody should be extracted with possible masure against dilution, if necessary by a
relatively high cost method, and total recovery must be required, in the other hand, for
Iowest grade orebody, however, higher dilution risk can be considered for lower con.
Another goveming factor is entireiy geomechanical therefore associated with the applied
stress and quality and the nature of the rock mass itself For the understanding of dilutiorr
problem, we have emphasised on the stope face behavioun; this undergoes the assumption
that the larger part of dilution results fiom the wnsequence of ground instability
conditions. However the data given by CMS shows the stope configuration resulting fiom
the combination of overbreak and rock caving in the stope. Therefore, it remains
complicated to identify the relative proportion of the effkct of each of these unexpecte
components. Absolute better alliance of these two factors (appropriate blasting pattern
and efficient cable bolt support) is the solution for excellent performance in stope rnining.
However, to attain these dextrous combination rernains laborious; and it no rare that some
mines expire their active life without to achieve this objective. The efficient management
0
of rnining operation in such condition infers precarious responsibilities. In fa*, the
continuai need of etneiency performance at lower cost remsins of paramount importance
to al1 concerned in stoping operation. For the correction of ore dilution problem, field
studies and investigations for the assessment of r ecove~g stope extraction as weU as their
working time require much energy and cost. The application of th& recommendations
also impose additional expensive development work; therefore postpone any local active
operations. The cumulative consequences of these decisions rnake timorous mine
operational executives. However, if some measwes are not taken it could evedy destroy
the profitability of the operation. Thus. It is no rational to still without imply the adequate
means to enhance or reduce this operational ailment. It is also important to report that
many operations have ctosed because of uncontroUed ore dilution, and in serious cases
before they even completed their fist year of production. Even if it has been mentioned in
previous sections, it remains signifiant to signai frther that local geologicai occurrence
thoroughly stays the main source of ore dilution. The occurrence of orebody in weak host
rocks (jointed rocks, foated rocks or soii rocks) are usually associatecl with high ore
dilution. Independently of the characteristical parameters of the concerned deposit, the
mining method employed and the delicacy of rnining operations proportionaily induce or
control dilution problem. Without forgetting the ~vucturai nature of the orebody itself
(irregular ore/waste contacts, indistinct ore/waste contacts) which are aiso the potential
sources of dilution.
For narrow vein mining in Abitibi area the effect of magnitude of stress/stability
have critical incidence of the dilution. The significant engineering judgement is to
emphasize that it can bepossible for stability of the stope. This remains crucial big
challenge in jointed and/or foiiated rocks as at Bousquet 2 mine. Because there still two
particular restraints in this types of orebody rnining. One is that the better ore remvery
make uncertain the acceptable stope stability. It is known the stability of the stope as well
as the factor of det y of the whole mine stiU closeiy linked to number and the size of the
siIl pillars. These pillars are in many cases the ore leaving in place. While an ideal situation
should imposes maximum o n recovery. The second rrstnining factor imply the eEea of
0
depth on stope stability. Mining in deeper conditions involve the combination of the stress
magnitude and discontinuitiw. Also the sale of tress literaily increase with the depth; with
crucial grow of stress, the res ubg induce failure will be proportional. In the 0 t h hand
the existence of the wmbination (shear stress-discontinuities) may be the main cause of
diIution in open s t op. The mechanism of stop ore diIution closely results fiom the
fiindamental behaviour of waii rocks. This mechanism of instability provoked by shear
stress and discontinuity is explaineci in stope w d behaviour section.
Chanter 7: Conclusions and recommendations
7.1 Conciusions
Sublevel open stoping remains an improvtd mining method for steeply dipping orebody
in Abitibi archean area. The presence of several close openings, in jointed volcanic rock,
required by this mining method provokes the instability dong the s t op waiis. In such
conditions, the mechanisms causing ore dilution can be surnrnarized by the wmbination of
severai factors such as ground condition, blasting damage, state of stresses iwound the stope,
and stope design. Thus, the evaluation of stope wds behaviour by numerical modeUing and
CMS has perrnitted to understand the various components provoking ore dilution. It hm been
proved at Bousquet 2 mine that dilution wmes fiom both w d rock overbreak (due to
explosive energy) and hangingwall caving and footwali sloughing (due to stress redistribution
and ground conditions). However, it is still complicated to ident* the relative proportion of
overbreak and caving (or sloughing) because of the implication of various parameters such as
0
presence of joints and geological inhomogeneity. From a ground condition point of view,
dilution inside the stope is provoked by univourable rock properties and the occurrence of
discontinuities. Stope instability associated with stresses is iinked to post-mining stress
instantaneous and continuous redistribution in two zones: stress concentration zones with
maximum principal stress about 240 MPa (in the corner of the stope) and stress relaxation
zones; these zones occurring dong the hangingwall and the footwal are subrnitted to tensile
stresses (about 0.65 MPa in horizontal direction, and 14 MPa in vertical direction) provoking
the opening as weli as the sliding of induced fdures. In relaxed field, tende stresses are
proved to be the main cause of dilution; that has pennitted to calibrate the numerical mode!. A
sensitivity analysis has proved that horizontal field stress acting on the stope is reduced by 25%
for the secondary stope; this can be a cause of lower dilution in secondary stope. Othenvise,
stope overbreak caused by blasting effects is due to the inelastic behaviour of the rock mass
under the explosive energy; and the proportion of overbreak is allied to the Peak Particle
Velocity generated by the interaction of explosive material-geological materiai. Controlled
0
blasting has been demonstnted as a good option for reducing dilution. It has been
e
demonstrated by CMS observations and the numetical modelling results, as weii as the results
of blasting effects that the ore of the primary stopes is much more diluted than this of the
secondary stopes. This is due t o the higher level of induced stress around the prirnary stopes
and the higher vibration of wali during blasting. Larger part of dilution wmes fiom caving of
1
the hanging wail (poor quality of this rock and effect of gravity). Thus, primary stopes must be
I
carefiilly prepared against dilution. To control ore dilution, cable bolting of the hanging wall
Yom parallel drift is sti i l the satisfsaory way. The use of cable bolts as a pre- reinforcement
method of ground control has proved effeaive in minunizing ore dilution fiom stope wall
rocks. The result of mode hg has dernonstrated the change in the stress field after cable bolt
installation. This is accompanied by a reduction of ore dilution. Consequently, the study has
shown the eEctiveness of cable bolt pre-support of the f& quality w d in Bousquet sublwel
stoping.
7.2 Suggestions for Further Resurch
The control of open stope wall stability must be done for efficient and econornical ore
recovery. In spite of several studies, ore dilution remains a problem nplying various complex
parameters, difficult to evaluate. These parameters should be cleatly identified and understood.
Even though we have emphasized Our study on some items and data, our ability to quanti@
field stress is probably limited, in part because stress measurements are difficult and expensive
to be carried out, and because the stresses vary t o an unknown degree from point t o point
within the rock mass. Other fture studies which can be considered are:
The use of severai numerical methods in conjunction, allowing for the advantage of
each method and then providing a good analysis fiom dif5erent perspective.
In a particular mine study, it should be interesting to compare dilution values obtained
fiom CMS survey with the correspondkg stability number N. The values N and R (hydraulic
radius) can be plotted in modined stabity graph (proposed by Mathew). The resultant graph
could help to create a relationship between ore dilution and stope stability.
Further studies involving underground borehole drillings to determine the location of
discontinuities as faults joints etc. and their effkct on stope stabity.
Further investigation must be pafonned on the understanding of Mure mechanisms
and criteria to determine rock mass abity to sustain confined and shear stresses before stop
design.
Assessrnent of various cable bolting designs and patterns and then to choose the most
effective one for the stop design.
Anderson B. and Grebenc B.: Controlling dilution at the Golden Gant mine. 1995
C M mine opetators conference.
Ajang B. 1996: " In Situ Ground Stresses in the Abitibi Minhg District". C M
Bulletin; VOL 89. no 996. Pp. 95 -7 1.
Arjang B. 1988' Pre-minhg ground stresses at the Bousquet/Dumagami mines,
Cadillac, Quebec. s CAhIMEz Mining resemch Laboratories. Division reppwf MkU
88 - (lx).
Bawden, W.F. Hyett, AJ and Cortolezzis: "Towards a methodology for performance
assessrnent in cable bolt design." Proceedings of the International Symmum on
rock support, 1992.
Bawden, W.F. and Hyett, AJ: A field cable bolt pull test programme for Lac
Minerals at Bousquet Division - Mineshaft 2. Bousquet mine infernal report, 1994.
Bourchier, F., Dib, E. and O'Flaherty, M.: "Pratical improvements to installation of
cable bolts: Progress at Campbell mine." Proceedings of the Intemutional Symposium
on rock support, 1992.
Bourgoin C., Lafiamme M. and Planeta S. "Techniques d'Exploitation Souterraine des
Gisements MetalMeres Filoniens; Annexes: Etude de la dilution, Rewe documentaire,
Bibliographie*' . Universit Laval, Projet de Recherche et Developpement Aout 1991
Brady B.H.: Rock mechanics and ground control for underground rnining and
construction.- 1989. Prtxeedingofthe 3th US. Symp~sllltl~
Cassidy K. : 'Implementation of a cable bolting progrom at the Con Mine." Mining
Engineering Student; 1989. U' rgonrnd Rock Engineering.
David M. " Dilution and Geostatistics". C M bulletin, June 1988 vol. 8 1 no 9 14; p. 29
- 35.
Diederrichs M. S, and. Kaiser P. K., 1996 "Rock instabiIity and Risk Analyses in Open
Stope Mine Design". Cm. Geotech. J. 33 : p 43 1-439.
Djordjevic N. "MUii.mizing the Environmental Impact of Blast Vibrations" Mining
Engineering April 1997; p 57 - 6 1.
Duk-Won Park & Vojtech Gall: "Supercornputer assisted three-dimensional finite
element analysis of a longwall panel." 1989. Proceeding of the 30rh US.
Symposium
Dune K., and Pakalnis R "Dilution Aspects of a Sublevel Retreat Stope at Detour
Lake Mine". Rock Mechanics, Auberin, H-i & Mri ( eh) 1996 Bakema,
Rotterdan; P. 305 - 3 13.
Eberhard, E.; Stead, D. and Reeves M.I., and Connors, C.: Design of tabular
excavations in foliated rock: an integrated numerical modelling approach.
Geotechnical and GeoiogicaI Engineering, 199 7; 15 4 7-85.
Germain P. and Hadjigeorgiou, 1997. "Influence of Stope Geometry and Blasting
Patems on Recorded Overbreak". Intemational J mmI of Rock A4echanc.s & MN>ing
Sciences. Vol. 34. No % Paper No 1 1 S.
Germain P. Hadjigeorgiou, J. and Lessard, J.F., 1996: On the relationship between
stability prediction and observed stope overbreak. Pruceedings Yd Al orth Amen'can
Rock Mechi cs Symposium on rock support (NARMS*1996), Montreal, pp. 277-
283.
Graham E. P. 'Zow-Cost, High-Dilution Muiing". CM bulletin for JIU&? 1968. P.
847 - 853.
Henning J. et al., 1996 " Elastic and Plastic Material Properties of Core Samples From
Bousquet" . Mine interml report.
Henning J., Gauthier P. and Ruest h& 1997 " Etude sur la Stabiiite des Epontes
Supeneures au Complexe Bousquet". Mine intemol report.
Hyett A J., Bawden W. F., Lausch P., Ruest M., Henning J. and Baillargeon M., 1997
"The S.M.ART. Cable Bolt. An instrument for the detennination of tension in 7-wire
strand cable bolts". Contde de Terrain, 12e Colloque, Val d'Or 1997.
Hyett, AJ. and Hudson J.A.: "In situ stress for underground excavation design in a
naturally fiactured rock mas." 1989. Proceeding of the 30th US. Synipanm.
Hu Y." The Bucldig Failwe Analysis of a Cavem in Jointed Rock". Inter. J. Rock
Mech. &Min. Sci. Vol. 34, no 3-4, no 133 A; 1997.
Jones, P.R and Durston K. J., "Mining Method selection - The Dilution Dilemma".
Presented at the CIMAnmal General Meeting Ca&ary? Mizy 11" 1993.
Kaiser, P.K., mederichs M. and Yazici S.: c Cable bolt peflormance during mining
induced stress change - Three case example. n Proceedings of the Inteniaiional
Symposium on rock support, 1992.
Luotte Y.C. "Controiied Blasting at CANMET Experhental Mine". Mining
Engineering, June 1396; p 74 - 78.
Mitri H. S. 1997: "Finite elements in rnining rock mechanics." focus on underground
mining. Course notesMcGiIl Universi4
Mitri, H. S. , Edrissi R and Henning J. 1997, Sinite Element Modelling of Cable-Bolte
Stopes in Hard Rock Underground Mines." Presontedfor SEM AnmraJ Meeting 1994,
paper nurnber 94- 1 16.
Potvin, Y. 1988 "Empirical open stope design in Canada. PhB. thesis. D e m e n t of
Mining and minera1 Procesihg %e university of British Colmbia, Vmcouver.
Powers B., Rocque P. and Spron D.: "Productivity Improvements and cost savings at
Hemlo Gold Mine's Golden Gant Mine." Hemlo GoldMine ' i nt erd report, 1995
Rocque P. and Bawden, W.F.: Recovery of a double width stope at Golden Gant
Mine: a successful application of numerical modelling and conventional methods for
stability analysis. >> Proceedings of the Intematioml Sp p s i um on rock support,
1992.
Saeb S md Amadei B. ''Effect of boundary condition on the shear behavior of a
dilatant rock joint." 1989. Proceeding of the 30rh US. Symposium.
Toungny G., Doucet D., and Bourget A, 1983 "Geology of Bousquet 2 Mine, an
example of a deformeci, gold-bearing, poIymetallic sulfide deposit.: Economic
Geologv, vol. 88.
Tinucci J. P. "A ground control cornputer program for support analysis of three-
dimensional critical rock blocks. Atasca Consulting Group Inc. Proceedrngs of the
International Symposium on rock support, 1992.
Windsor C., R, 1992 Invited lecture: Cable bolting for underground and surface
excavations. Proceedrngs of t k IntemtiomI S y m p * m on rock support.
Yeung M. R and Leong L.L. " Effects of Joint Attributes on Tunnel Stability". Ihter.
J. Rock Mech. & Min. &i. Vol. 34, no 3 4 , 1997, p. 505 no 348.
Zelanko J. C. Mucho T.P. Iannacchione AT. &Barcd T.M "Stabilisation of a Large,
S haiiow, Potentially Unstable Underground Openhg". Rock Mechmtcs, Aubertn,
Hmscllli & Mitri (eds) 1996 Balkema, Rotterdam; P. 155 - 161.
Appendix A: Bousquet 2 Mine Planes
PRATIQUES STANDARD
Galeries &ns le mi wai
Zone large (4nret +)
Appendix B: Blasthole Pattern & Blasting Sequences
a - , I I q
- - -
ANG.
I I R - /
1 0"; 1 M G O 1 Ltz 1 4 &yE
TROU
334 3.5 Ck,,, 3 -5
No DU
TROU
MET.
v v r i r 1
\
EXPLOSFSTTM (KG)
MATERIEL DE DYNAMITAGE ENDROIT: 3- / CHANTIER # 0
3 EXPLOSIFS
SEQUENCE: t 1 3 1 4 1
# POSTE: 21 999 3/5v 7 b p 999
EXEL D E L
CONNECTEUR DE SURFACE (MS)
1 CODE
POCHES LOMEX 11
POCHES LOMEX III (BLEU)
1 ICAISSES SUPER FRAC jZT30o (BAZOOKA) 1 7
2
3
3
a 5
CAISSES TRIMRITE 2C 32x27 metres
CAISSES APEX ULTRA 65x400
5
6
Appendix C: Datafiles for Numerical Modelling
a
Ditifik for Numericd ModeMing: .dit l e
HW support by cabls-case2
64457710
5 10 3 10 3 10 10
10 101013 3 10 10
1 3 0.0 0.0
2 2 9.1 0.0
3 2 18.1 0.0
4 2 22.3 0.0
5 2 34.85 0.0
6 2 41.85 0.0
7 2 71.85 0.0
8 3 178.55 0.0
9 1 0.0 120.0
10 O 33.3 120.0
11 O 58.8 120.0
12 O 63.0 120.0
13 O 75.55 120.0
14 O 82.55 120.0
15 O 112.55 120.0
16 1 178.55 120.0
17 1 0.0 150.0
18 O 41.0 150.0
19 O 71.0 150.0
20 O 75.2 150.0
21 O 87.75 150.0
22 O 94.75 150.0
23 O 124.75 150.0
24 1 178.55 150.0
25 1 0.0 176.4
26 O 41.0 176.4
27 O 71 .O 176.4
28 O 75.2 176.4
29 O 87.75 153.6
30 O 94.75 153.6
31 0 124.75 153.6
32 1 178.55 153.6
33 1 0.0 180.0
34 O 41.0 180.0
35 O 71.0 180.0
36 O 75.2 180.0
3 7 O 92.4 180.0
38 O 99.4 180.0
39 O 129.4 180.0
0
Datarde for Numericd ModeUing: .cab fde
'1'200000 .O00138 11
'na 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221
's' 20 20 20 20 20 20 20 20 20 20 20
'2' 200000 .O00 13% 1 1
'n'21612162216321642165216621672168216921702171
' s12020202020202020202020
'3' 200000 .O00138 1 1
ln' 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121
' s' 2020202020202020202020
'4' 200000 .O00138 1 1
'n1206120622063 20642065 2066206720682069 20702071
'~'2020 202020202020202020
'5' 200000 .O00138 1 1
'n'20112012201320142015201620172018201920202021
' s' 2020202020202020202020
'6' 200000 .O00 138 1 1
ln' 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971
' s12020202020202020202020
'7' 200000 .O00 138 1 1
'n' 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921
'n' 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871
' s12020202020202020202020
'9'200000 .O00138 1 1
ln' 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821
' s' 2020202020202020202020
'10'200000 .O00138 11
'nt 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771
's' 20 20 20 20 20 20 20 20 20 20 20
'1 1' 200000.000138 11
'n'171117121713 1714 1715 171617171718 1719 17201721
' s' 2020202020202020202020
'12'200000 ,OOC138 1 1
'nt 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671
' s' 2020202020202020202020
'13' 200000 .O00138 1 1
'n'161116121613 16141615 161616171618 161916201621
's' 20 20 20 20 20 20 20 20 20 20 20
END
Appendix D: Results of Stope Light Measurement by CMS
a
COMPLEXE BOUSQUET
e
BARRICK
-
OPTECH
CHANTIER :
TENEUR RSERVE (gltm) :
TENEUR CEDULE (gltm) :
TENEUR MUCK (glbn) :
TENEUR USINE ( g h ) :
O DENSIT REMBLAI :
1 OENSITE ST~RILE :
2 DENSIT~ MINERAI :
Item
6
7
8
9
10
11
12
13
14
15
Type
MINERAI LAISSE DANS LE CHANTIER :
STRILE LAISSE DANS LE CHANTIER :
STRILE TRIE :
MINERAI A L'INTRIEUR DES
LIMITES DE SAUTAGE (14% Dv. enlev) :
MINERAI REEL CASSE :
(-14% Dev. enlev):
MINERAI USINE :
STERILE + REMBLAI EFFONDRE :
12.1 REMBLAI EFFONDR :
12.2 STERILE EFFONDRE HM :
42.3 STERILE EFFONDRE FMI :
STERILE USINE :
MINERAI EFFONDRE COT EST :
MINERAI EFFONDRE COT OUEST :
item 1 Type
16 DILUTION L'USINE :
17 EFFONDREMENT STERILE TOTAL :
18 RECUPERATION L'USINE :
19 MINERAI CAS& LAISSE :
Calculs
Calculs
ILUME A REMBLAYER : 3412 ma
OTE : Chantier vide lors de l'arpentage. Plan lvation 41 01 -6 interprt
17.6
17.6
21.3
N.D.
Tonnes
Pourcentage
IATE ARP. : 27-nov-97
PAR: D.G.. R.C. 8 G. Labbe
Pat: R. Gaulin
Tech. Prod.