Search Documents

By K. G. Lomberg, M. D. Lionnet

The Bafokeng Rasimone Platinum Mine (BRPM) is a joint venture between Rustenburg Platinum Mines Limited and Royal Bafokeng Resources (Pty) Ltd (Bafokeng). The mining lease has been granted over Boschkoppie 104JQ and is located approximately 10km south of the Pilanesberg Alkaline Complex. The Merensky Reef, which constitutes the principal ore body, is geologically complex. A relationship between its geological variability and the associated mineralisation has been established by defining a number of facies types. The Merensky Reef has been sub-divided into seven distinct facies types: the pegmatoidal facies, contact facies, FW1 pothole facies, FW3pothole facies, FW6 pothole facies, and FW7 pothole facies. The Merensky Reef, which does not form part of the potholed reef, has a large component of footwall mineralisation. The pothole facies, in contrast, has mineralisation located in the hangingwall. Statistical analysis of the facies types indicates that they are unique populations and hence need to be evaluated independently. The mining lease has distinct structural domains that have added additional complications to evaluation of the Merensky Reef. The Merensky facies on BRPM exhibit unique geological and statistical characteristics within well-defined structural blocks. Due to the extreme variability of the Merensky Reef on BRPM a unique approach has been developed, within Anglo Platinum, to evaluate the deposit. Mining cuts are defined and composited for the individual facies types, and then modelled individually by applying known geological controls. The resultant grade models are weighted based on statistical and local geological knowledge. The object of weighting the facies is to replicate the variability observed in the ore body. The introduction of stringent geological controls to the evaluation process has resulted in increased confidence in the in situgrades estimated as an input to the mine planning process

Jan 1, 2006

By R. W. O. Kersten

Backfilling is one of the options available to the mining engineer in mine design. The decision as to the type of backfill is dependent on the purpose for which it will be used. For example, in a situation where high-grade, steeply dipping multiple reefs are to be mined, the purpose would be the following: ? to facilitate maximum extraction; a ? to minimize loss of fines; ? to allow closely spaced multiple reefs to be mined; and a ? to create a working platform. The physical properties required which will satisfy the abovementioned requirements can then be defined; the design of the fill plant, as well as the transport of the fill, will be governed by these considerations. The same procedure is also adopted in deciding on backfill in single- reef gold mines. The first question to be asked is the purpose of the fill. It may be to reduce the energy release rate, for environmental reasons, to reduce timber usage, to improve productivity, to reduce stoping costs, to obtain higher mining extraction, and so on. Only after one, or a combination, of these factors has been selected, can the specifications for the fill be decided on, and the plant designed accordingly. The introduction of backfill at Sheba Gold Mine will be used to illustrate the procedure adopted by Anglovaal in determining the type of fill required.

Jan 1, 1988

By A. Refahi

For predicting the energy consumption during the size-reduction process, the Bond approach is often used. In this work, the PFC3D discrete element method (DEM) software was employed to model the crushing behaviour of some rocks with different mechanical properties in a laboratory jaw crusher. FLAC3D software was adopted to analyze the stress distribution in the rocks. The rocks studied were modelled as granular assemblies in the shape of a sphere and/or a cube located between two jaws, and the work done by the jaws in the crusher was determined. Nine different types of rocks were studied and the energies consumed by the crusher were compared to those of the Bond comminution energy estimated from the Bond index. There is considerable difference between Bond crushing energy and work done by the jaw crusher for rocks. It appears that the Bond approach is not a suitable method for predicting single particle fracture energy done by the crusher. To verify the results obtained from DEM models, the fracture behaviour of the crushed rocks was examined and was compared to the PFC3D results. The tensile mode of fracturing is favourably modelled by the PFC3D software while the delamination mode cannot be well modelled by PFC3D software. Keywords: jaw crusher; discrete element method; crushing energy; Bond index

Jan 1, 2009

By R. M. Göktan

A theoretical comparison is made between point-attack and wedge-shaped picks in regard to their cutting efficiencies in coals having different strength characteristics. With specific energy as a criterion of cutting efficiency, it is shown that a mathematical expression can be derived for the assessment of the relative performances of these two groups of picks as a function of coal-strength parameters and pick-tip angle. The ratio of tensile to compressive strength of coal is found to play a significant role in the determination of relative cutting efficiencies, lower ratios favouring the use of point-attack picks.

Jan 1, 1992

By J. P. Coetzee

Dr. K. Gebhard In his excellent and comprehensive survey on the future development of the Iron and Steel Industry in South Africa, Mr. Coetzee has also mentioned the direct reduction processes as a possible method of converting iron ore to iron. Mr. Coetzee has made it quite clear that at this stage one should not consider these processes as a 'magic formula' which herald a 'break through'. I agree entirely with this point of view. Once upon a time I held the opposite view, but in the meantime I have revised my views. To do so, I had 40 years at my disposal. In 1930 I had the privilege of being connected with the first commercial sponge iron plant in Germany1 - a plant which was capable of producing 40 tons of sponge iron a day. We were determined to prove that the blast furnace would become obsolete in the near future. We considered it as an unnecessary detour to convert iron ore first into an iron-carbon alloy, which contained further quite a number of undesirable elements, and to remove this carbon by an endothermic process. Direct iron-making seemed to be a logical short cut. Unfortunately, we were not able to establish this 'short cut,', but the life of our sponge iron plant was cut short. We had solved all technical but not the economic problems. Basically the positon is still the same today - mainly economic and to a lesser degree technological aspects determine the rate at which sponge iron plants are installed. More has been said and written than done about direct reduction plants. I think that there are two main reasons for this stagnation in the introduction of direct reduction processes. The first reason is that modern blast furnaces are just too good and too economic ever to be replaced by direct reduction processes on a large scale. This, of course, already limits the field of application for direct reduction plants. The second reason is that sponge iron production and utilisation is based on technical considerations, which increase not only the production cost of sponge iron but also the conversion cost to steel to such an extent that under special conditions only are direct reduction plants an economic proposition. Are these technical considerations still as forceful and valid today as they were when 'direct steelmaking' was first contemplated? The development of sponge iron production as a method of direct

Jan 3, 1971

By A. Montini

The problems encountered in the design of control systems for milling cir- cuits when the model of the system and the model parameters are uncertain are discussed. Considering these uncertainty factors has led to the development of a technique that gives a quantitative assessment of real system stability, reveals the trade-offs between performance and stability and helps the designer in deciding what level of model complexity to use in the controller design. Furthermore, it allows for on- or off-line optimization of the milling circuit performance as it provides the base for the design of a controller which is stable over the plant operating range.

Jan 1, 1987

High pressure grinding roll (HPGR) technology is used in an increasingly diverse range of applications, predominantly iron ore , gold and diamonds. KHD Humboldt Wedag HPGRs are applied world-wide, and consistently prove to be well designed and reliable operating units, with their performance meeting the pre-set standards. This publication summarizes some of the features and experiences for recent applications in treatment of copper ore, coarse iron ore, and gold ore. Summary data of roll surface wear life and operating data are given, together with operational observations. In addition, effects of feed segregation and truncated feed are discussed, as well as the implications of a product recycle flow sheet.

Jan 1, 2012

By A. J. Reynolds

In recent years, rapid depletion of the higher grade reserves, allied to decreases in paylimits, has led to the exploitation of secondary, erratically mineralized placer horizons on various marginal and low grade gold mines. On Blyvooruitzicht Mine, the rich Carbon Leader Reef has limited reserves, necessitating exploration of the lower grade Middelvlei Reef. This placer has always presented evaluation and exploration problems, to such an extent that it was, at one stage, considered uneconomic. Despite an extensive underground drilling programme no ore reserve targets of reasonable size were apparent. However, when a certain degree of parallelism between sedimentological trends in the Middelvlei Reef and the underlying Carbon Leader was recognized, a meaningful exploration model could be developed. Firstly, all historical sampling records for the Carbon Leader were entered into a computerized system; the database comprised in excess of 340 000 data points. Variables captured consisted of gold value, uranium value, con- glomerate thickness and total reef thickness. The raw data were then averaged into blocks, and plots of different variables were computer-generated for further analysis. This enabled the Carbon Leader to be modelled sedimentologically and, thereby, exploration targets for the Middelvlei Reef were generated, based on the Carbon Leader facies trends. Geostatistical analysis of sampling data in exposed areas of the Middelvlei Reef underground has confirmed the sedimentological breakdown of the placer into distinct facies areas. A facies model relating value distribution patterns to depositional trends, both for the Carbon Leader and the Middelvlei Reef, has thus been incorporated into a mine planning strategy.

Jan 1, 1987

By A. Cichowicz

The objective of this work is to quantify the effect of the rock mass surrounding a stope or a tunnel on the seismic signal so that the rockburst damage on the excavation can be derived. It has been shown that the transfer function of rock mass surrounding an excavation can be modelled using a damped oscillator. The method of estimating the transfer function is extended in this paper. The parameters of a damped oscillator are estimated using inversion techniques and the transfer function is modelled as a time-varying system. The technique is tested on data acquired at three different sites in two deep gold mines. Employment of a modal model of the transfer function was suitable in two cases. Application of this technique is essential to track the time variation of the natural frequency during ground motion caused by a seismic event.

Jan 1, 1999

By K. L. Riemer

The West Wits Line first began yielding gold in the mid-1930s and is among South Africa's leading gold-producing areas. Gold mining operations are carried out amidst the occurrence of seismic events which cause both damage to underground excavations and injury and loss of life to production personnel. The feasibility of establishing a system to minotor the seismicity on all the Gold Fields mines was investigated during the early 1980s. This led to the establishment of a centralized seismic network in 1981. The system was structured around three major objectives. It uses Perkin Elmer (now Concurrent Computer Corporation) 16 and 32 bit mini-computers. The hardware in use at both the central and remote sites is outlined, giving a good insight into the role of computers in the system. The system software reveals the fast flow of data through the network; it can be divided into three categories: (1) data logging and transmission - remote sites, (2) data capture and communications - central site, (3) applications and research - central site. Several examples are given which show how the system is measuring up to its objectives. These case studies also indicate how the computers are effective in representing the seismic data collected. It is concluded that the use of computers enables the pin-pointing of seismic events on the West Wits Line within about 3 -4 minutes of their occurrence. The choice of hardware is largely dependent on the objectives of system, and the Gold Fields Network has chosen mini-computers because of the speed of communication and the large amount of data and storage required for research applications.

Jan 1, 1987

Contribution by F.S.A. de Frey The authors are to be congratulated on a job well done by making, in their paper, such a pool of information available to mining engineers. I would like to comment on the following statement: 'As an alternative to the commonly used deep-level mining methods, consideration should also be given to pillar mining at great depths as a logical extension of the current methods. For example, a 75 per cent extraction Or Ryder's discussion The moderate levels if average pillar stress (APS) (320 MPa) are consistent with pillar stability. Squat 10:1 pillars should never 'fail' in themselves, and foundation strengths are also not exceeded (the stresses are well below the 2 ½ x UCS rule-of-thumb provided the foundations comprise normal strong 200 MPa quartzites). Authors' reply We thank Dr de Frey for his contribution to our paper on shallow hard-rock pillar systems. To considerable extent, we share his enthusiasm the potential role of squat pillars in deep-level mining, although we harbour reservations concerning strata control, mining economics, other issues. We agree wholeheartedly that further research, together with carefully planned field trials, is warranted in pursuit of this potentially highly rewarding concept. .

Jan 1, 1995

By J. E. Nesset

The bubble size, together with the gas rate, determines the bubble surface area flux in flotation machines and, therefore, impacts directly on the kinetics of flotation. Understanding the variables that influence bubble size is one of the keys to understanding flotation recovery. Surprisingly, these relationships are poorly understood, largely due to the difficulties associated with gas dispersion measurements in industrial environments. The McGill gas dispersion sensors (bubble size, gas rate, gas holdup) are capable of reliably making such measurements. This paper reports on a study, using the McGill sensors in a 0.8 m3 Metso RCSTM pilot cell, which investigated the effect on bubble size of frother type and concentration, gas rate, temperature and rotor speed. A Denver laboratory cell was used to investigate the effect of elevation above sea level. Frother concentration proved to have the predominant influence on bubble size, and the effect of frother type can be normalised as a function of the critical coalescence concentration (CCC). Measurements made at many plant sites worldwide are compared to the model developed using the Metso RCSTM unit.

Jan 1, 2007

By D. D. Howat, J. van der Colf

This paper is the first in a series of three dealing with this subject and also with the rate of oxygen removal from the slags in the presence of solid carbon and with the physicochemical properties of the reduced slags. For selected members of the system, the melting temperatures and primary crystallization products were determined. as were the viscosity and the electrical resistivity of the corresponding melts. All the experiments were conducted in a neutral atmosphere. Where determined, the primary phase was found to be either perovskite or magnesium aluminium titanate. The composition variable exerting the most pronounced influence on viscosity or resistivity was found to be the basicity expressed as the mass ratio (CaO + MgO)jSiO,. The ratio CaOjMgO became important only at Iow basi<:ity values. Although generally there is a direct proportionality between log (viscosity) and log (resistivity), slags h&apos;lving either a very high silica content or a very high titania content have anomalously Iow electrical resistance.

Jan 1, 1979

By L. Forsbacka, L. Holappa

Viscosities of quasi-ternary CaO-SiO2-CrOxmelts in contact with metallic chromium were measured by the rotating cylinder method up to 1750ºC. Twelve slag compositions were selected in the composition range 20–55 wt% SiO2, 10–45 wt% CaO and 10–60 wt% CrOx. Estimation of the amounts of CrO and Cr2O3was made by regression analysis based on the earlier data. Professor Iida’s (Osaka University) viscosity model was applied and model parameters (0CrO,CrOand Cr2O3) were evaluated. The Iida model gave good consistency with the measured viscosity values.

Jan 1, 2004

By J. U. Burger

The Elizabeth Bay aeolian deposit in Namibia comprises a sedimentary package of diamondiferous grits that have been deposited, reworked and concentrated to various degrees by dominant southerly winds. Diamond concentration in these deposits is highly variable and often unpredictable. The deposit is marginal from an economic perspective and therefore requires selective mining which in turn requires correct sampling and accurate estimation. Three methods of sampling have been used at Elizabeth Bay for mineral resource estimation: (i) pit samples utilizing an excavator to produce a sample 1 to 3 metres deep with a 10 to 20 m2 footprint, (ii)GB50 grab sampling with a 3.34 m2 footprint and (iii) a BG36 bucket auger drill with a 5 m2 footprint. Pit sampling, being easy to execute, has been the most widely used despite a limited penetration ability and, at times, questionable integrity. The GB50 grab sampling tool was tested at Elizabeth Bay and although it had better penetration ability and sample integrity, it struggled with cemented horizons. No sample optimization was conducted prior to the GB50 grab sampling. The latest phase of sampling was conducted using a BG36 bucket auger drill which is capable of penetrating thick overburden, penetrate cemented horizons and provides an acceptable sample support size. Geostatistical simulation methods were applied to establish the optimum sampling parameters required to estimate a mineral resource at an indicated level of confidence. The results show that a 5 m2 sample support size is adequate and that a 50 m 100 m sample configuration is optimal. Sample data obtained from the three methods confirms that the BG36 bucket auger tool combined with the parameters identified in this exercise produce the best results for diamond resource estimation.

Jan 1, 2009

By T. H. Yang, K. K. Lee, Z. Z. Liang, C. A. Tang, L. C. Li, Y. Tsui, L. G. Tham

Abstract: A series of numerical tests of hydraulic fracturing in heterogeneous rocks were performed to investigate the influence of pore pressure magnitude and gradient on initiation and propagation of tensile fractures. The numerical tests were run with a Rock Failure Process Analysis code, F-RFPA2D, incorporated with a Flow-Stress-Damage coupling model (FSD). To investigate the influence of pore pressure magnitude on fracture behaviour, “pinch-off” breaking tests, originally employed by Bridgeman to investigate the effective stress law for tensile fracture in a uniform pore pressure field, were numerically modelled. In another numerical test, a double-notched sample, with fluid pressure in one notch while keeping the other one open to the atmosphere, is numerically extended to investigate how the fluid flow direction or the pore pressure gradient will influence the fracture behaviour. The simulation results show that fracture is strongly influenced by both pore pressure magnitude and pore pressure gradient. Keywords: seepage mechanics; heterogeneous; pore pressure; numerical simulation; failure process ZUSAMMENFASSUNG: Um den Einfluss der Wasserdruckgroesse der Oeffnung und der Porisitaet auf den Bruchentstanden und Verbreiten der Tension von Gebirgen zu beurteilen, wurde viele nummerische Pruefungen auf den hydraulischen Bruch in dem Gebirgen aufgenommen. Diese Pruefungen wurden mittels der Analyse- Software F-RFPA2D und Flow-stress-Damage Coupling model ( FSD) durchgefuehrt. Um den Einfluess der Wasserdruckgroesse der Oeffnung auf den Bruch zu beurteilen, wurden auch Bruchpruefung und nummerische Analogexperimenten gemacht, und dadurch wurde die ganze Phase der Bruchentstanden und Verbreiten wiedergezeigt, und die Ergebnissen ergeben sich, dass die Wasserdruckgroesse der Oeffnung und der Porisitaet sehr viel auf den Bruchentstanden und Verbreiten der Tension von Gebirgen beeinfluessen. Schluesselwoerter: Durchlaessigkeit , ungeleichmaessig, Wasserdruck der Oeffnung, nummerische Analog, Zerstoerungspahse Résumé : Une série de simulations numériques a été ménée pour analyser l’influence de la grandeur de la pression de pore et du gradient sur la naissance et la propagation des fractures de tensions par le biais de la simulation numérique des fractures hydroliques dans des rochers hétérogènes . Le logiciel F-RFPA (Rock Failure Process Analysis code ) intégrant FSD ( Flow-Stress-Damage coupling model ) y a été employé. Les résultats des simultations montrent tout le processus de la naissance et de la propagation des fractures sous la pression de pore. Ces résultats nous ont permis de constater que la grandeur de la pression de pore et le gradient exercent une influence importante sur la naissance des fractures dans les rochers et sur leur mode de propagation . Mots de clé : le mécanisme de transfusion ; l’hétéogénéité ; la pression de pore ; la simultation numérique ; le processus destructif

Jan 1, 2003

By D. G. Krige

Recent publications have demonstrated the important role that can be played by a priori information of the grade levels, variability, and spatial patterns in adjacent or nearby areas in improving the quality of grade estimates for virgin or sparsely drilled areas. A research project was therefore conducted involving the collection and analysis of extensive sampling and borehole data from 7 reefs on some 30 Witwatersrand mines. This paper shows that the &apos;equivalent&apos; log variance for the two-parameter lognormal model provides a good basis for the study and comparison of the levels and behaviour of relative variances. The trends of these against the corresponding log areas were found to form distinctive patterns, which are useful for these purposes. The patterns for the Basal, Vaal, Carbon Leader, and Main Reef Series form a reasonably close group, whereas those for the V.C.R.(particularly where the facies are mixed), B Reef, and Kimberley-Elsburg Reefs (particularly for narrow channel widths) do not always conform. The patterns shown by underground chip samples are also compared with those obtained from borehole cores. It is shown how this information can be used to improve the borehole valuation of new mines and of virgin areas in existing mines.

Jan 1, 1993

By B. K. Loveday, N. V. Ramlall

A study was carried out to evaluate various batch flotation models for the recovery of minerals in a UG2 platinum ore. The major minerals in a UG2ore can be grouped as platinum group minerals, chromite, and siliceous gangue. This study also examined the entrainment of minerals during flotation. The models were ranked using statistical methods and an analysis of model-fit residuals. Entrainment parameters obtained from model fitting were evaluated for logic using a simple mineral-to-water ratio versus time plot. The foremost conclusion from the study was the importance of entrainment modelling. The measurement of water recovery increased the size of the data-set, and the inclusion of a simple entrainment model was statistically significant. The overall fit to the data was improved, and the entrainment model provided logical information on the recovery of gangue minerals that were not considered to be floatable.

Jan 1, 2015

If an orebody has been drilled to achieve some level of regulatory compliance, then a number of associated geological databases will exist, representing 10 000 to 100 000 m plus of core intervals. The databases usually contain data about assays, lithology, mineralisation and geophysical attributes. They will also include location information such as collar positions and downhole surveys. Obviously this data is collected by geologists for, predominantly, geologists and mining engineers. However, from a process design viewpoint process engineers usually ignore the geology where possible and rely on a small number of metallurgical test samples. Typically these test samples are sourced from convenient and available cores or from a small number of specially drilled metallurgical holes. One of the worst instances known to the author was the design of a SAG mill using only two SAG specific test results (one for each of the major ore types) while, geologically, the orebody was defined by more than 30 000 m of drilling. This paper sets out a philosophy to maximise the use of the geological data, both for process design and in preparation for the commissioning phase of projects. Geological data is used in the identification of preferred locations for metallurgical test samples, for checking the metallurgical validity of the geological ore-type definitions, where necessary developing new ore-type definitions (geometallurgical domains) and developing mathematical relationships for the estimation of important metallurgical factors across the geological database. If the estimation of factors (for example ore competence) is considered reliable enough then there is also the possibility of feeding calculated data back into the block modelling process to allow the generation of metallurgically relevant outputs in the mine plans.

Jan 1, 2007

By R. M. Whyte

The Konkola Deep Mining Project (KDMP) seeks to exploit the deep level potential of the Konkola Deeps orebody, producing high grade copper concentrates for treatment at the Nkana Smelter and Refinery. As an alternative to smelting, Konkola Copper Mines (KCM) has examined a processing route involving pressure leaching of the Konkola concentrates at Nchanga, with the leach liquor being treated by solvent extraction and electrowinning to produce high purity copper cathode. The solvent extraction circuit would be operated to achieve a high ?delta? copper, thereby providing a strong acid raffinate which can be used in the existing Tailings Leach Plant (TLP) to treat Nchanga oxide copper materials. The main focus of the hydrometallurgical option is to capitalize upon the synergy that exists between the treatment of sulphide copper concentrates (acid producing) and oxide copper ores (acid consuming). The paper discusses the key features of a comprehensive piloting campaign carried out at Anglo American Research Laboratories, which involved the co-treatment of Konkola copper concentrates and Chingola refractory ore in a fully integrated pilot plant circuit.

Jan 1, 2001