Search Documents

By G. F. Loughlin, C. H. Behre

THE DEVELOPMENT OF CLASSIFICATION WHAT is the use of a classification of mineral deposits? From the days of Agricola, the founder of the science of ore deposits, successive authors on the subject have attempted classifications, none of which have attained unanimous indorsement of miners, engineers, or geologists. The miner and engineer naturally have little regard for a classification that does not help them to find ore. Mining geologists may share this feeling, and deplore the overemphasis that may be placed on classification by their academic brothers. Apparent effects of such overemphasis may be reflected by students or new graduates who have grown to think that each deposit must fit perfectly into some pigeonhole of the more or less arbitrary scheme that has been drilled into them. Diversity in mineral deposits as in other natural phenomena, however, is too great to be fully expressed in the usual tabular classification, and experience is essential for the appreciation of the value or usefulness of a classification. The succession of classifications has closely followed the growth of the science of mineral deposits and reflects that growth, as well perhaps as the limited or comprehensive viewpoint of each proponent. The earliest classifications were based on metal content, and were superseded by those based mainly on form of deposit. Source of materials also became a basis, and was followed by physical and chemical conditions and processes. In spite of this development it is interesting to note that in many of the latest reports on mining regions the authors adopt special classifications suited for their particular districts. Some of these, like the earliest classifications, are based primarily on metal content, and others are most conveniently based on form, and it may be convenient to modify these by reference to source of material or temperature of deposition. Some, including the writers, after noting these different usages, have refused to worry much over classification as an ultimate objective, but,

Jan 1, 1933

By P. Diehl, M. David

"As shortage in the supply of mineral commodities become more probable, there is an increasing need for reliable and comparable reserve/resource data. It is therefore necessary to standardize the traditional classification procedure by quantifying the three basic criteria in resource evaluation: economic feasibility, geologic assurance and recovery. For the portion of fairly well documented deposits and as far as their geologic parameters are concerned, geostatistical methods can considerably improve the quality of ore classification. In respect of economic feasibility and recovery, approved geostatistical methods can be introduced directly into the classification procedure. The concept of geologic assurance, however, requires a new interpretation in terms of geostatistical error spread. As the ""risk"" in estimating an ore parameter is also a function of the respective block size, it is reasonable to define not only upper limits for the risk intervals for each class of uncertainty, but also block dimensions which reflect the typical stage of planning associated with the different ore categories. An algorithm has been developed to block out areas inside a mine or a large panel which satisfies the constraints of an ore category in respect of the risk intervals of a parameter and in respect of their size. The method has been computerized and is demonstrated on a massive copper-nickel deposit.IntroductionFacing an uncertain future with respect to the demand and supply of mineral commodities, both governmental authorities and the mining industries show an increasing interest in a useful framework for the assessment and classification of the remaining mineral resources"". This is reflected by the recently published appraisal and classification scheme of the United Nations as well as by the new or revised classification systems of the USA; Canada or Australia. Moreover there are various classification schemes which are currently discussed in major mining companies."

Jan 1, 1982

Paper written on project work carried out in partial fulfilment of M.Sc. (Chemical Engineering)degree Identification of non-linear systems can be a daunting task and in the process industries the problem is complicated considerably by the presence of noise from various sources, non-stationarity of the data and intermittence, such as observed in particulate flows. Traditional methods, such as frequency analysis and linear model-ling do not handle these systems well. Similar problems arise with the application of non-linear theory developed over the last few decades, where much of the analysis depends on embedding of the data in a phase space or pseudophase space, since these methods were not originally designed to deal with noisy systems. In contrast, singular spectrum analysis is a method which can be used to identify structure and distinguish noise from important information in multivariate data, such as the trajectory matrices of time series measurements. It provides an orthogonal basis onto which the data can be transformed, so that components of the time series can be investigated individually. With Monte Carlo methods, the components can be compared with those from linear surrogate data in order to identify possibly non-linearities in the data. In this paper, this versatile technique is briefly explained and illustrated by means of case studies.

Jan 1, 2003

By S. J. Haynes

"This paper documents the characteristics of the different quartz veins, and their relation to gold, in the turbiditic Meguma domain of eastern Nova Scotia. Eleven different types have been distinguished in the greenschist facies on the basis of form, internal structure, mineralogy, wallrock alteration and crosscutting relationships to each other and to regional structures. Gold is present as rare visible disseminations in host rock slates and in ore-grade, early-formed veins, that pre-date the formation of upright Acadian folds and intrusions of Devonian granitoids. Four vein types pre-date Acadian folding. Five gold-deficient vein types are associated with Acadian folding (syndeformation); whereas two gold-deficient vein types clearly post-date this deformational event: Previously unreported, wall rock alteration is associated with early crosscutting and semiconcordant auriferous veins. In contrast, early stratiform veins are devoid of wallrock alteration. It is suggested that the auriferous vein arrays formed as submarine exhalitive systems over synsedimentary tensional faults and that the Meguma domain may represent the opening of a basin similar to the modernGuaymas rift .IntroductionThe term ""turbidite-hosted gold deposits"" is introduced to distinguish gold deposits contained within a turbiditic sequence of wackes and shales (or their metamorphosed equivalents) from gold deposits in other types of sediments. Usage and justification for the term is similar to ""carbonate-hosted lead-zinc deposits"" or ""volcanogenic massive sulphides"". A range of different characteristics can be present in individual deposits, some of which may be indicative of specific deposit types in the same manner that volcanogenic massive sulphides can be subdivided into ""Cyprus type"", ""Kuroko type"", ""Besshi type"", etc. Historically, gold deposits hosted by turbiditic flysch were an important class of quartz-vein gold mines in the 19th century. They included the goldfields of Victoria, Australia, Otago, New Zealand and Nova Scotia, Canada. Production from most of these fields had ceased by World War II and interest in this type of deposit waned. However, the recent exploration interest in gold has resulted in both re-assessment of the classical mining camps and the discovery that ""turbidite-hosted"" gold deposits are present in many parts of the world.Mapping of the regional geology and gold deposits of the Megumadomain of Nova Scotia was carried out by E.R. Faribault of the Geological Survey of Canada between 1886 and 1939. Faribault produced a comprehensive series of maps and reports for each gold district, a compendium of which was published by Malcolm (1929)."

Jan 1, 1987

By L. A. Readdy

Defining classes of ore using geostatistical methods that match historical classifications has presented problems in the past. A simplistic and practical approach to this problem is presented. An example of the approach is presented using a porphyry type system as a case study. The technique described uses area of drill hole influence and known conditions from similar deposits, or the production history of the deposit being studied. It provides a practical bridging mechanism for defining the class of ore on a block-by-block basis using .the error of estimation value developed during kreiging. The approach provides a filtering method that refocuses attention on the adequacy of drill hole spacing.

Jan 1, 1985

By Nick Barton

The ability to estimate that support is unnecessary for various dimensions of opening in certain rock mass environments is of particular relevance in the mining industry, where most excavations are not expected to remain open indefinitely. As Hoek1 observes: "the basic aim of any underground excavation design should be to utilize the rock itself as the principal structural material, creating as little disturbance as possible during the excavation process and adding as little as possible in the way of concrete or steel support". Approximately two hundred case records have been analysed for the purpose of determining the amount of permanent support normally used when a given span is excavated in known rock mass conditions Some thirty cases concerned unsupported openings. Excavations that can stand unsupported appear to have certain common characteristics as regards rock mass quality. However, the mechanical properties of the rock mass are not the only factors controlling the choice of design. The dimension of the opening (span or height) and the type of excavation are two important variables.

Jan 1, 1975

By Antoni Kidybinski

Mechanical properties of the nether roof rocks play a substantial role in maintenance of mine openings. On longwall faces too low strength is a cause of rock- falls which bring about delays in face advance, and hazard to the miners, while too high strength may result in over hanging of roof rocks in gobs, sometimes over a large area, initiating dynamic periodical pressures. Hence, efforts have been undertaken to develop methods of caveability evaluation for proper selection of roof control system and of longwall-face supports. DESCRIPTIVE CLASSIFICATIONS One of the basic qualitative criteria for classification of roof types with different caveability is the criterion of geological structure. The 26-stages scale worked out for conditions of Donetz Coalfield. USSR, by E.T. Proyavkin et al.(l) may be an example of that type of classification. This scale is presented graphically in Fig. 1. The first types of roofs according to this scale are easily caving immediately after exposure of a small surface, while the last be- long to the strongest roofs which over-hang in waste and require artificial fracturing. Observations and measurements carried out also in Donetz mines lead W.T. Davidyanc (2) to following classification: Class I - weak and densely bedded rocks very easily caving after each advance of supports. The roof subsidence over a waste edge is: [ ] where: h is roof subsidence, m M is thickness of the seam being mined, m S is width of an opening, m Class II - medium caveability and small main roof movements [ ] Class III - hardly breakable rocks, thicker beds with higher strength: caving by large blocks. After exposure of larger roof surfaces the movements of the main roof strata may affect an opening. [ ] Class IV - very hardly breakable rocks - thick, monolithic sandstones: they cause considerable difficulties in working with caving. [ ] Class V - thin-bedded rocks featured by the capacity of a smooth subsidence without disintegration into loose blocks with keeping contact between the particular blocks in a bed. In this classification mining with full roof caving is recommended for class I-III and only exceptionally for class IV. In many coal fields, qualitative and quantitative observations of roof rocks behavior have been carried out: as a result relations and characteristics found in certain mines are given. As an example, Kizielov Coalfield data in the USSR given by B.S. Vinokur (3) may be mentioned. In working thin pitching seams man- less system of undercutting the seam by means of rope saw is used there. The author separated four classes of roofs and gave, for every class, the empirical relation between the extracted area and the time of roof stability. The size of the stable area of exposed roof for particular classes is as follows: 120 to 150 m2 (1291 to 1614 ft2 ), 170 to 200 m2 (1829 to 2152 ft2), 250 to 280 m2 (2690 to 3013 ft2 ) and 350 to 400 m2 (3766 to 4300 ft ). For Kuznetz coalfield in the USSR the following classification of roofs was developed at the WIMI Institute in Leningrad: Class I - completely unstable: impact strength in- dex of rock f = up to 1.5 - the strip of roof 1.4 to 2.0 m (4.59 to 6.56 it) wide, exposed after shearer's cut holds stable during the time not longer than 5 min. Class II - low stability: f = 1.5 to 2.5 - the strip of roof 1.4 to 2.0 m (4.59 to 6.56 ft) wide and up to 4.0 m (13 it) long, exposed after shearer's cut holds stable during 5 to 20 min. Class III - medium stability: f = 2.51 to 3.5 - the strip of roof 1.4 to 2.0 m (4.59 to 6.56 it) wide and 4.0 to 25 m (13 to 82 ft) long, exposed after shearer's cut holds stable from 20 min. to 2 hours. Class IV - stable: f = 3.51 to 4.5 - the strip of roof 1.4 to 2.0 m (4.59 to 6.56 ft) wide and 25 m (82 it) long exposed after shearer's cut is stable during the time longer than 2 hours. Class V - entirely stable: f = over 4.5

Jan 1, 1982

By W. C. Patrick

Introduction The availability of a thorough, simple-to-use roof fall classification system is desirable for several reasons. First, such a system would expedite the reporting of falls for the purposes of estimating material removal costs, analyzing damage levels, and complying with federal regulations to report falls in certain areas of the mine. Second, the classification system would allow the mining engineer to record falls routinely with little time and effort expended. This would encourage him to make observations concerning what type of roof support would work best in a given situation. Third, and probably most important, an adequate roof fall classification system is the first step toward developing a means of predicting the occurrence and extent of future falls. It was with these three reasons in mind that the following classification system was developed. The type of classification chosen reflects the background of the developer and the intended purposes of the system. Hylbert has proposed a geologic-based classification which groups falls as to structural and compositional origin. The need for similar systems is noted by Cox and by Ferm and Melton. Each of these researchers is confronted with a situation requiring a relationship between local geology and roof fall occurrence. Such a relationship would allow the prediction of the relative frequency of falls from a knowledge of local geologic features. The current research is aimed at fulfilling somewhat different requirements, as noted above. Fall geometry has been chosen as the basis of the classification system so fall type can be reported independent of local geological, environmental, and tectonic conditions. This approach parallels that used in the common slope failure classification with groups failures relatively independent of

Jan 1, 1977

By G. Song

"SynopsisAccording to the overlying rock strength and thickness of 10 shallow coal seams in China, the strata are classified into four types: weak and laminated strata (hereinafter referred to as WL), weak and massive strata (WM), strong and laminated strata (SL), and strong and massive strata (SM). The four types of strata, together with longwall panel height (M), panel length (L), and seam depth (H) are factors that influence the loads on hydraulic power supports. Principal component analysis was used to study the seven influencing factors. It turns out that seam depth (H) is multicollinear with WL, WM, SL, and SM. Thus the enumerated seven factors are divided into two models. In Model 1, six factors; panel height, panel length, and layer thickness of the four type strata are involved. In Model 2, three factors; panel height, panel length, and seam depth are included. Using multivariate regression analysis, two equations to calculate loads on supports are obtained for each model. The equations were applied to shallow coal seams. The equation of Model 1 is more accurate than that of Model 2, but Model 2 could be further used in deep coal seams. The equations were coded through VB programming to simplify the calculation.IntroductionTo ensure coal production in the long term, the focus of coal mining has been gradually shifted to the western areas of China in recent years due to rich coal resources in this region. Most of the coal seams in this area are shallow but the support loads are abnormally high. In these circumstances, the chock loads estimating methods used in the past do not work. If the rated working resistance of a support is larger than needed, the shield is unable to fully perform, while a lower working resistance would lead to accidents like shield crushing or even casualties. This problem is frequently found in shallow coal seams such as Shenfu-Dongsheng, Yulin, Tuha, Lingwu etc. (Zhang, 2011). Thus, to ensure safety and efficiency during mining, the estimation of support loads and selection of supports are of great significance."

Jan 1, 2015

By A. J. Rergis

During the past ten years a number of papers have appeared describing zeolite occurrences in the western United States in Mesozoic sedimentary rocks, altered bedded Tertiary pyroclastics, playa lakes, hot springs, and amygdaloidal basalts. With the emergence of natural zeolites as economic industrial minerals it appears, useful to develop a system of classification to delineate those zeolite deposits of economic interest which were not discovered until the middle 1950's. The first attempt at a classification of naturally occurring zeolites was made in 1914 by Lindgren l when he classified the zeolites into eleven groups based on their mode of occurrence. The zoned zeolites in the Mesozoic rocks of New Zealand and elsewhere were classified according to the mineral assemblages by Fyfe, et al.2 in 1958 as a zeolite facies corresponding to low grade regional metamorphism. In 1966, Hay suggested as an alternative to the preceding classification one which would "classify as diagenetic all mineral assemblages produced in sedimentary rocks under conditions in which zeolites are formed." In this paper a classification is proposed with the objective of providing an overall perspective of the zeolite occurrences in the western United States.

Jan 1, 1967

By P. A. Rossouw, G. A. Fourie

Several attempts have been made over the latter half of the twentieth century to devise a system for the classification of underground mining methods. None of these are found to be totally successful. A revised classification is suggested with three main categories being identified by reference to fill requirements, steep stopes and stratified geological environments. Methods always needing lasting mass or structure fill include methods with timber-structured stopes and those with mass-filled (or cut-and-fill) stopes. Methods associated with steep stopes are subdivided into steep-stope-specific caving and steep-stope-specific non-caving (or self-supporting) methods. Methods not in the first two categories, often used in stratified geological situations, are not necessarily restricted to stratified environments and include caving as well as non-caving varieties. This category is further divided into room-and-pillar mining, longwall and shortwall coal mining and the type of open stoping found in gold or platinum reefs comprising overhand, underhand and breast-stoping variations

Dec 1, 1996

By R E. McIvor, K M. Bartholomew

Classification system efficiency (CSE) is a simple, powerful metric that provides a comprehensive measure of classification system performance. CSE directly quantifies the amount of coarse solids in the mill, and as a result, also directly quantifies the percentage of mill power being used effectively. In this manner, it is an effective, practical grinding energy efficiency metric.Three independent variables drive CSE: circulating load ratio, sharpness of classification, and classifier bypass. Due to practical downstream constrains on per cent solids, improvements realised by increasing circulating load reach diminishing returns as sharpness of classification and bypass deteriorate in the most commonly used classification device, the hydrocyclone. This combination of effects leads to the use of CSE, and its relationship to operational economics, as an objective circuit optimisation criterion.Its simplicity in measurement makes CSE a useful tool for the plant metallurgist, and it is not subject to a great amount of experimental error. CSE provides the plant operator a fast, simple, and effective means to evaluate the performance of the plant classification system. Improvements in CSE relate directly to circuit throughput and energy efficiency. This paper, through description and plant case studies demonstrates the power and utility of this emerging grinding system optimisation metric.CITATION:Bartholomew, K M and McIvor, R E, 2013. Classification system efficiency - a comprehensive closed-circuit grinding optimisation metric, in Proceedings World Gold 2013 , pp 389-394 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Sep 26, 2013

By Rob Sloan, Simon Yu, Rob Robison, Mark Schmidt

"In 2000-2009, Highland Valley Copper (HVC) tested new generation 660 mm cyclones supplied by various manufacturers in an aim to replace existing first generation 762 mm cyclones without success. Due to plant infrastructure limitations, new generation larger (838 and 800 mm) cyclones were not feasible. In 2010, an approach of testing hybrid 800 mm diameter cyclones was initiated by HVC. The hybrid cyclone consists of an 800 mm cyclone upper head section and the use of existing 762 mm cyclone conical sections. Three hybrid 800 mm cyclones were engineered and fabricated, targeting a throughput of 850-950 mtph and were installed in the E line cyclone cluster. Since starting operation on September 20, 2011, three hybrid 800 mm cyclones have been running at combined throughput ranges of 750 to 1050 mtph. Based on plant Daily Met Reports, the test hybrid cyclones demonstrated a decrease in misplacement of coarse copper bearing material to the cyclone overflow and better cyclone efficiency when compared to the existing 762 mm cyclones at the same throughput. The hybrid 800 mm cyclones also demonstrated significantly higher capacity per unit when compared to the existing cyclones. Based on the improvement demonstrated in the full plant trials, HVC decided to install the new hybrid 800 mm cyclones in all five grinding lines starting in July, 2012. A more detailed discussion of the full plant trials and initial results of the first complete retrofit of E line will be presented. INTRODUCTIONHighland Valley Copper Partnership (HVC) is Canada’s largest base metals mine and one of the world’s largest producers of Cu-in-concentrate. The mine, located in the central interior of British Columbia, is owned (97.5% interest) and operated by Teck Resources Ltd. In 2011, HVC produced about 210 million lbs. of Cu and 8 million lbs. of Mo. HVC’s Mill Modernization Project (MOP), which was budgeted for $475 M, is currently under intensive construction and is scheduled for commissioning in Q1 2014. The following description was valid at the time of writing this paper.The operation has been expanded a number of times since the startup of the original Lornex mill in 1972. The plant started up with two 9.75 m (32 foot) diameter primary SAG mills. It was expanded in 1981 with the addition of a 10.36 m (34 foot) diameter SAG mill. In 1988, the Highmont concentrator grinding mills and flotation section were moved and attached to the Lornex mill site. This added two10.36 m (34 foot) diameter AG primary mills to the Highland mill. This brought the total number of primary mills to five referred to as Line A, B, C, D and E: three SAG mills (A, B and C) each with two ball mills in parallel and two AG mills (D and E) each with one ball mill. Each AG mill line is equipped with one short head pebble crusher. Ball mill circuits are classified using 8 clusters of ten Linatex 762 mm cyclones to reach a target P80 330 micron (equivalent 43% +100 mesh), before the cyclone overflow product advances to the bulk flotation circuit."

Jan 1, 2013

By Radford B. Langston, John A. Marjerison, Graham C. Howell, Hendrik A. D. Kirsten

Probabilistic design is gaining wider acceptance in the rock engineering community since it allows more rigorous determination of risk relating to ground fall or excavation instability. Risk analysis can be conducted by various means, but the basis is formed by either objectively or subjectively determining probability of occurrence of an event. In the case of rock engineering, this event is either instability or excavation failure. Rock mass classification systems provide objective analysis of data collected on a typically subjective basis that also relate closely to excavation stability. A probabilistic analysis technique is presented that uses statistical distributions of rock mass and material properties, ground support fixture specifications, stress conditions, opening geometry, and ground support installation quality to more rigorously determine probability of failure for an underground opening and the subsequent risk to personnel.

Jan 5, 2007

By Franz Ramstorfer

Impressões de Baumann são utilizadas amplamente nas aciarias para a avaliação e monitoramento da qualidade interna de placas: em particular a segregação central. A análise e monitoramento da segregação central é uma base importante do monitoramento da condição da máquina de lingotamento contínuo, da otimização dos parâmetros do processo de lingotamento contínuo e da qualidade das placas. Em geral, o tipo e a intensidade da segregação central são determinados através da comparação da imagem real com imagens de referência do padrão aplicado. O presente trabalho foi focado no desenvolvimento de uma ferramenta de análise computacional, precisa e fácil de usar pelos operadores do laboratório de Baumann. O objetivo foi estabelecer uma base padronizada e confiável para o processo de monitoramento da segregação central. A metodologia desenvolvida é baseada em processamento de imagem computacional de uma impressão de Baumann em tons de cinza. Observou-se que a classificação de segregação pode ser realizada com uma precisão excelente através da análise computacional dos dados de imagem da segregação central. O procedimento de classificação foi implementado com a utilização de uma rede neural feed forward.

Jul 29, 2018

By A. W. Fahrenwald

THE function of the classifier in modern fine-grinding practice is to remove a finished product from the grinding-mill discharge, leaving material that needs further comminution. The classifier, therefore, makes two products: (1) "finished" product, which overflows the rim of the classifier tank, sometimes referred to as "slimes" or as "classifier overflow," and (2) "unfinished" product, which settles to the sloping bottom of the classifier tank and is removed by dragging or raking. This product is generally referred tows "sands" or "oversize." One of the outstanding features of the closed-circuit classifier is its ability automatically to return the unfinished product of the ball-mill discharge to the feed end or scoop box of the ball mill. This is a distinct advantage because it avoids dilution of the ball-mill feed, and, furthermore, the particular method of removing the settled sand, by raking it up the inclined bottom of the classifier, gives a classifier discharge containing a minimum of water. This leads to maximum classifier efficiency, in this type of classifier, because the maximum volume of water is displaced upward by the settling sands. The closed-circuit classifier is required to deliver a finished pulp of closely specified density requirements, therefore added hydraulic water generally is not permitted.

Jan 1, 1930

By A. Vaillant

Classifiers that are used to separate granular materials into a coarse and a fine stream, are considered. Fundamental Equations are presented which apply to all, such Classifiers and all feed size distributions. The practical, real Classifier is compared quantitatively with the perfect. Classifier. A Formula is introduced for the Real Classifier Selectivity, that defines what portion of each particle size is sent to the coarse outlet duct. This selectivity function is characterized by three Parameters, one of which (b) is a numerical measure of the sharpness of classification. A new dimensionless variable (J) appears in Equations and Graphs for recirculating load, product size distribution and specific surface in the classified product. This variable combines the effects of sharpness of separation, Classifier setting and fineness of a Rosin-Rammler Type feed to the Classifier. Classifier efficiency is defined, based on Specific Surface.

Jan 1, 1967

By P. T. Luckie

Analysis of classifier performance independently of a closed grinding circuit configuration may result in a poor match-up between the grinding device and the classifying device. This paper examines the relationship between classifier and grinding device in dry grinding circuits by employing a semi-empirical methodology.

Jan 1, 1980

By J. Dvorácek

Similar to the case of any other business venture, the activities of mining industries are profit-oriented. Nothing will stop successful businesses in their quest for new localities to be prospected for minerals of their choice, because the useful economic life of any mine is limited. Before starting the process of land acquisition, it is important for any mining company to be able to forecast the profitability of the venture. The focus of this paper has been on the prospects of success, concerning the case of the opencast extraction of aggregates for construction. A set of 70 successful and loss-making quarries was subjected to investigation. An international company owns the quarries, which are located in seven European countries. The method of discriminate analysis provided the criteria for assessing future economic performance. The same criteria served the purpose of classifying other quarries belonging to the company.

Jan 1, 2012

By N. A. Beier, R. Amos, Smith. L., C. Atherton, D. R. Barsi, D. Blowes, D. C. Sego

"Waste rock is one of the most abundant materials at mine sites and must be properly characterized and managed to limit acid rock drainage and other harmful byproducts. An experimental waste rock pile—part of the test piles research program—located at the Diavik diamond mine (Northwest Territories, Canada) was deconstructed and sampled to study the spatial variability of material properties. Bulk samples were classified based on the amount of material finer than 4.75 mm. Hydraulic properties of each sample were estimated and particle size distribution analysis was completed. These datasets and classification methods might be used to improve the understanding of hydraulic and geochemical behaviour of the material and to study waste rock at other sites. ¦ KEYWORDS Acid rock drainage (ARD), Material classification, Particle size distribution (PSD), Waste rock RÉSUMÉ Les stériles font partie des matériaux les plus abondants sur les sites miniers et, en tant que tels, doivent être correctement caractérisés et gérés afin de limiter le drainage rocheux acide (DRA) et la pré¬sence d’autres sous-produits nocifs. Dans le cadre du programme Test Piles Research Program, un amas de roches stériles expérimental situé à la mine de diamant Diavik (Territoires du Nord-Ouest, Canada) a été décomposé et mis à l’essai afin d’étudier la variabilité spatiale des propriétés des matériaux. Les échan¬tillons en vrac ont été classés en fonction de la quantité de matériaux plus fins que 4,75 mm. Les propriétés hydrauliques de chaque échantillon ont été estimées et l’analyse de la composition granulométrique effectuée. Ces ensembles de données et méthodes de classement serviront à mieux comprendre le com¬portement hydraulique et géochimique du matériau et à étudier les stériles sur d’autres sites. ¦ MOTS CLÉS classement des matériaux, composition granulométrique, drainage rocheux acide (DRA), stérilesINTRODUCTION Statement of problem The characterization of mine waste rock remains a challenge in active mining sites. The large volume of material and the large size of particles make it difficult to collect representative samples. Sampling techniques are often dis-ruptive to mining operations because they alter the structure of the waste rock pile or interfere with mine traffic. The convention at most mine sites to collect little to no information on waste rock material means that models and predictions of how waste material will behave and interact with the surrounding environment are based on general information that might not be site specific. This could lead to environmental damage or higher costs if environmental effects are underestimated during the construction of waste rock structures."

Jan 1, 2019