Search Documents

By Philip J. Daynes

As part of the regeneration of the centre of Dublin, a multi-functional development is being constructed on the site of the historic Smithfield Fruit Market. The site is being developed in three phases, and includes one of the deepest basements (10m) constructed in Ireland to date. Until the recent improved economic climate, there has been little requirement for deep basements in Dublin, and this is the first to be constructed entirely within fluvio-glacial gravels and boulders. The difficult ground conditions led to the adoption of a diaphragm wall solution to ensure a watertight construction. Tie-back anchorages are utilised in the temporary case, and the wall is propped by floor slabs in the permanent case. Where required by high bedrock levels, the diaphragm wall is dowelled into limestone bedrock to ensure toe stability. The recorded lateral movement of the wall is significantly below that calculated by numerical analysis. Temporary bentonite cut-off walls were installed to divide the site into discrete sections. The basement was constructed as a tanked structure, and is subject to uplift water pressure. Passive tension minipiles provides an innovative alternative to the standard stressed anchorage solution. This paper describes:- design and construction aspects of the diaphragm wall, tie-back anchorages, and anti-flotation minipiles, together with recorded lateral wall movements,

Jan 1, 2003

By Peter V. Barr

Despite the advent of modem firing control systems reheating furnace operations frequently allow significant thermal excursions in the discharged slabs or billets after furnace delays. Although HSM furnaces are not immune to this and other related problems, mini-mill operations are most vulnerable due to their more frequent delays and (often) less sophisticated control systems. In order to reduce these problems control systems will commonly include a thermal model, operating either on-line (interacting directly with the controller logic) or in an advisory, off-line mode. Using a new 'fast' thermal model, the paper examines the thermal response of a typical mini-mill furnace during both delays and the subsequent recovery period. A variety of firing strategies are explored and compared on the basis of maintaining thermal excursions within allowable limits for rolling. Other factors, such as the potential for overheating of billets within the furnace during delays and the overall thermal state of the furnace, are also considered in some detail. Although common practice, the use of conventional thermocouple zone temperatures for furnace control during and after delays is shown to be a dubious proposition. Keywords: Reheat Furnace, Modeling, Process Control

Jan 1, 2003

By G. Onal, E. Gock, H. Dincer

"During the conventional chromite dissolution process; chromite ore concentrates were ground to at least <90 µm using the ball mill and after adding Na2CO3 and leaning material (instead of CaCO3 also residues from the leaching process) at a ratio of e.g. 1:0.75:1.2, depending on the specific formulation, follows the digestion e.g. in a rotary kiln, heated in countercurrent. The reaction time is 3 to 8 h. The calcine is comminuted and leached with water or a thin chromate solution. Depending on the type of ore used, the yield is 75-85 %. The disadvantage of the above-described process is that it is impossible, using current technology, to achieve a complete digestion of the chromite ores and that the chromium-containing residues have to be disposed of as special waste, involving a large financial effort.The subject of this study was the investigation of the effect of vibratory milling on chromite dissolution. The impact stress in the vibratory mill causes the mechanical activation of chromite, which is characterised by lattice defects. The degree of mechanical activation can be quantified by measuring the relative X-ray diffraction intensities at the parent lattice planes of chromite. Each relative X-ray diffraction ratio correlates with a certain degree of digestion. As opposed to vibratory milling, a complete digestion of chromite using a ball mill is impossible, even at extremely long grinding times.In this investigation a chromite concentrate from Pulpinar/Kayseri with content of 44 % Cr2O3 and the Eccentric Vibratory Mill was used. In order to verify the effect of the vibrating grinding on chromite dissolution, reaction kinetic investigations on digestion and leaching were carried out."

Jan 1, 2003

"The first 90 years of the last century generated a considerable volume of technical literature on the subject of grinding mill power and its prediction. It is therefore surprising that during this time there were no published models for predicting grinding mill power draw which were validated using a wide range of comprehensive industrial scale data.Notwithstanding this lack of data, the majority of models placed limited emphasis on the internal dynamics of mills and relied on simple assumptions which considered the charge to take up a fixed position and shape.Laboratory based studies (Liddell, 1986), and industrial data (Morrell, 1993), showed that these assumptions did not hold over a wide range of operating conditions, and brought into doubt the ability of the (then) existing models to accurately predict grinding mill power draw. Furthermore no model treated AG,SAG and ball mills as a single class of devices whose power draw could be predicted with the same equations.Through a detailed investigation of the charge dynamics, coupled with a large data base of industrial mill power draws, this situation was rectified by the end of the century, by which time a model had been developed and validated, which accurately predicted the power draw of all wet tumbling mills. This paper describes this model and reviews some of the earlier attempts at mill power modelling.INTRODUCTIONThe early 1900?s saw the beginnings of published papers which tackled the problem of what influenced the power draw of tumbling mills and what equations (models) were suitable for predicting power draw. This interest continued for the rest of the 20th century and the literature contains many papers that attempted to satisfactorily answer these questions. A common and disappointing theme in the majority of them, however, is the distinct lack of suitable experimentally derived power draw data. This deficiency consequently limited the practical application of many of the attempts at modelling mill power draw. Prior to the early 1990?s some limited data on the power draw of grinding mills in the literature can be found, but as Harris et al (1985) remarked they are ?... too frequently unusable simply because one or more essential variables have been omitted?. The absence of published results, from vigorous experimental testing of the various models, has resulted in a general lack of evaluation of the validity of assumptions and hypotheses that such models contain. Harris et al (1985) however, did make some efforts in attempting to evaluate the performance of various power prediction equations and in so doing further developed a semi-empirical one of their own. However, in using manufacturers? published data, which themselves were undoubtedly generated by proprietary equations of unproven validity, they probably succeeded in generating little more than a consensus of the various errors in these models."

Jan 1, 2003

By Jonkion M. Font

An expression for the distribution ratio of arsenic between slag and copper matte or liquid copper was developed by both, an extension of the Reddy-Blander model for deriving a priori the arsenic capacity, and by a relationship between the arsenic capacity and the arsenic distribution ratio. In this study, the a priori arsenic capacities in the FeO-SiO2, CaO-SiO2, MgO-SiO2, FeO-CaO-SiO2, FeO-FeO1.5-CuOo.5-MgO-SiO2, FeO-FeO1.5-CuO0.5-CaO-MgO and FeO-FeO1.5-CuO0.5-CaO-MgO-SiO2 slags were calculated at 1573 K. The a priori arsenic distribution ratio between the FeO-FeO1.5-CuO0.5-CaO-MgO-SiO2 slag and copper matte (L s/m as), and between the FeO-FeO1.5-SiO2 or FeO-FeO1.5-CuO0.5-CaO-MgO slag and liquid copper (L s/Cu as) were calculated. The model calculated data are found to be in good agreement with the reported experimental data. The availability of reliable thermodynamic data for slag components and impurities in the slag-matte-metal system is essential for the accurate prediction of impurity capacities and distribution ratio data. The arsenic distribution ratio model developed here, can be extended for prediction in multi-component slags and mattes or liquid metal, and thereby may lead to the development of new slag systems for the efficient removal of arsenic in the copper smelting, converting and refining processes.

Jan 1, 2003

By R Dunne, G Lane, S Ha, D Dreisinger

The paper outlines the 17 Mt/a Telfer Project plant process design from sample selection to feasibility study design. The 366 Mt Telfer mine reserve consists of a diverse range of copper and gold mineralisation, from massive sulfide reef ores to stockwork ore, oxide copper minerals to chalcopyrite, siltstone to competent quartzite, all with relatively high levels of gravity recoverable gold. The complexity of the orebody demanded a flexible plant that could recover copper and gold values by a combination of gravity recovery, flotation and concentrate leaching. The demands of treating ore from a high grade underground mine and a variable grade open pit (that includes underground old mine workings) necessitated a dual train approach from primary crushing to concentrate production in order to minimise the requirement for campaigning of different ore types. An extensive sampling, testwork and technical evaluation program was undertaken to identify processing characteristics and the optimum process route for each ore type. The design experience included: the development of robust primary stockpiling and pebble crushing concepts; detailed review of scale-up of SAG and ball mill pilot plant data to plant scale; the selection of two 15 MW SAG mills and two 13 MW ball mills for the grinding circuit; and detailed modelling of the pyrite concentrate carbon-in-leach (CIL)/sulfidisation acidification recovery thickening (SART)/acidification volatilisation recovery (AVR) process to optimise design for copper and cyanide recovery from leach liquors. The Telfer plant design also measures the development of process design based on experiences at the Cadia Hill Cu/Au concentrator, commissioned in 1998. The differences between the two plants are compared and contrasted.

Jan 1, 2003

By Richard K. Komarek, Antonios Zavaliangos, Roman T. Dec

"The work in the paper introduces a 2-dimensional finite element model for analyzing particular material processing in a roller press. Two basic roller geometries are considered: (a) smooth rolls (used for milling and compaction) and (b) rolls with cavities, commonly used for briquetting. A model is presented that is suitable to analyze material behavior through all stages of the rolling process: feeding zone, nip region, neutral plane and release section. It can predict pressure, shear stress, slip of the material over the roller surface, material velocity field, failure of the compact during release and the course of other important process variables.MODEL, RESULTS AND CONCLUSIONSAn associated Gurson model was employed as a constitutive model for the material. The calibration of the model was based on experimental results from an instrumented compaction die. Coulomb friction law was used to describe phenomena on the material/roller surface interface. Powder River Basin sub bituminous coal fines with moisture content 9% was used as a model material.Two finite element models have been set up, one for the briquetting roll and one for the “equivalent” flat roll. Approximately 3200 plain strain continuum elements with reduced integration (CPE4R) were used. Lagrangian-Eulerian (ALE) approach in conjunction with judiciously applied remeshing was employed with the commercially-available, dynamic explicit version of ABAQUS finite element code.Comparison of model predictions with experiment shows that there is an agreement in terms of the general trends of experimental values recorded on laboratory roller presses. Gurson model overpredicts stresses at low densities and, as a consequence, overpredicts roll force and roll torque."

Jan 1, 2003

By Justin Brown, Michael Kobler, Dave Ragland, Bob Shufflebotham

PROJECT OVERVIEW As the sun-bronzed San Diego students strolled casually to class, mining crews from Atkinson Construction were working only 9 meters below the surface. The330 meter long, 11.3 meter diameter squat ellipse NATM tunnel is part of the9.3 kilometer Mission Valley East Extension Light Rail Transit (MVE-LRT) being constructed for the San Diego Metropolitan Transit Development Board (MTDB). The SDSU segment of the MVE-LRT was awarded to the Clark Construction Group, the parent company of Atkinson Construction for $82.7 million. URS CORP was the prime designer and Hatch Mott MacDonald performed the NATM tunnel design. The construction managers were Washington Infrastructure Services with Underground Construction Managers, EPC Consultants, and ILF Consultants. A NATM tunnel design was chosen as the most cost-effective way to minimize disruption of the university activities. The tunnel passes close to several buildings and directly below access roads, tennis courts and a baseball field. The NATM method of conventional mining uses shotcrete for ground support often reinforced with steel orplastic fibers, welded wire mesh and lattice girders. The tunnel runs from a cut and cover box up a 4.2% grade in a 217 meter radius left-hand curve with several spirals and ends at the underground station 330 meters away. The depth of cover varies from 16 meters near the west portal to nine meters at the station. The design called for a two-pass system with a primary lining of 300 mm shotcrete with lattice girders at 1 m centers and two layers of welded wire mesh. The lattice girders are used primarily for profile control and are not considered a structural element. The top heading was designed to be excavated in one of three excavation sequences depending on ground conditions and depth of cover. In sequence one excavation, the top heading is excavated in two halves with a one-meter separation. In sequence two excavation, a side gallery is advanced 8 or more meters ahead of the

Jan 1, 2003

After a snap-shot survey of water quality of streams draining various coal measures carried out in 2000 - 2001, West Coast Regional Council expanded its monitoring programme to better establish the condition of streams draining coal mining areas and to characterise the variability of water quality with respect to stream flow. Samples were collected for water quality indicators such as pH, sulphate, turbidity, conductivity, temperature and dissolved oxygen as well as macro-invertebrates, fish and assessments of stream habitat. Water quality sampling was carried out at 27 sites on 21 streams located from Page Stream near Seddonville in the north, to Seven Mile Creek near Dunollie in the South, and to Garvey Creek east of Reefton. There was relatively low variability of pH, conductivity and sulphate concentrations at most of the sites sampled. Conductivity and sulphate concentrations reduced with increasing flow and pH generally increased. Significant effects on macro-invertebrate and fish populations were identified at several of the sites where water quality sampling was undertaken. Of the water quality parameters tested, macro-invertebrates were mostly affected by pH. Although not sampled directly (conductivity was used as a surrogate) heavy metal concentration may explain some of the variation in macro-invertebrate communities. Three creeks affected by mine drainage were sampled for fish and no fish were present. Fish appear to be more sensitive to acid rock drainage that invertebrates. Using data from this survey and other studies conducted in the region, the total length of stream that is actually or highly likely to be adversely affected by acid rock drainage was estimated at 125 km. This figure is useful as a gauge as to the magnitude of the issue in this region and for setting of priorities for remediation. However, as it is only an estimate based on limited sampling sites, it should be used with caution.

Jan 1, 2003

By X. H. Zhang, J. G. Cai

This paper discusses the behaviour of large-span tunnels and caverns in hard rock in a major underground storage project. In-situ stresses were measured using hydraulic fracturing during site investigation and 3-D overcoring during construction, with both showing a horizontal stress ratio of 2-3 times the vertical stress. Rock reinforcement design was based on the Q-system combined with numerical modelling for special design cases. Tunnel design and performance were checked and monitored by in-situ instrumentation during rock excavation, including measurements of deformation, convergence, and load on rock bolts. Deformation measurements using borehole extensometers included one borehole drilled into the virgin rock before rock excavation, which provided a direct and very insightful comparison with the various deformation measurements made during rock excavation. These measurements coupled with numerical calculations have proven that the high horizontal stresses to be very favourable to the stability of the large openings.

Jan 1, 2003

By S. -H. Chang, C. -I. Lee, Y. -K. Lee

Based on results from triaxial compressive tests, an experimental damage model to simulate post-peak behaviours as well as nonlinear behaviours before peak strength was developed. In order to consider pre-peak behaviours, variations of elastic constants at each damage level were regressed by functions of confining pressure. In addition, post-peak behaviours were simulated successfully by using Hoek-Brown constants. As results, the experimental model could predict nonlinear behaviours, peak strengths and stiffness changes observed in experiments. Basé sur des résultats des essais compressifs à trois axes, un modèle expérimental de dommages pour simuler des comportements de poteau-crête aussi bien que des comportements non-linéaires avant la force maximale a été développé. Afin de considérer des comportements de pré-crête, des variations des constantes élastiques à chaque niveau de dommages ont été régressées par des fonctions de pression d&apos;emprisonnement. En outre, des comportements de poteau-crête ont été simulés avec succès en employant des constantes de Hoek-Brunes. Comme résultat, le modèle expérimental a pu prévoir des comportements non-linéaires, des forces maximales et des changements de riidité observés dans les expériences. Auf Basis der Ergebnisse des dreiaxialen Drucktests wurde das experimentelle Schadensmodell, das sowohl das nicht-lineare Verhalten vor der maximalen Stärke als auch das Verhalten nach der maximalen Stärke simulieren kann, entwickelt. Um das Verhalten vor der maximalen Stärke zu berücksichtigen, wurden die Veränderungen der elastischen Konstante bei dem jeweiligen Schadensniveau durch die Funktionen des begrenzten Drucks zurückgegangen. Außerdem wurde das Verhalten nach der maximalen Stärke unter Benutzung von Hoek-Brown Konstanten erfolgreich simuliert. Schließlich konnte das experimentelle Modell das nicht-lineare Verhalten, die maximale Stärke und die Veränderungen bezüglich der Steifigkeit während der Experimente voraussehen.

Jan 1, 2003

PolyMet Mining is developing the NorthMet property near Duluth, MN. Laboratory investigations have led to the development of the Platsol process to dissolve in one step the base metals (copper and nickel) as well as the gold and platinum group metals (PGM). To complete the prefeasibility study, a fully integrated, 10-day pilot plant trial was undertaken. The trial consisted of high-temperature (>220ºC or >428ºF) pressure oxidation in the presence of 5 to 10 g/L chloride (the Platsol process). This was followed by solid/liquid separation, gold plus PGM recovery from solution, conventional copper solvent extractions/electrowinning (SX/EW) and recycling of the copper raffinate to the autoclave. The concentrate tested during the trial period was a NorthMet bulk flotation concentrate. It assayed 14.7% Cu, 3.05% Ni, 0.14% Co, 26.7% S, 1.4 g/t (0.041 oz/st) Au, 2.2 g/t (0.064 oz/st) Pt and 9.9 g/t (0.289 oz/st) Pd Laboratory results were confirmed and, with minor adjustments of the laboratory Platsol conditions, the following leach extractions were obtained during the pilot plant trial: 99.6% Cu, 98.9% Ni, 96% Co, 94.6% Pd, 96 % Pt and 89.4 % Au.

Jan 1, 2003

By Lauri Holappa

Copper converting is mostly performed in Peirce-Smith converters. The process is still very similar to the original one developed in the early 20th century. It is a batch process in a horizontal cylindrical reactor with air blowing via tuyeres along the length of the reactor side. The Peirce-Smith converter has its roots in Bessemer&apos;s inventions in the 1850&apos;s as he tested both horizontal and vertical vessels for steel converting. However, for steel making a vertical pear-shaped converter with air bottom-blowing was established. In the course of time, there was a distinct necessity for oxygen usage instead of air due to harmful effects of nitrogen to steel properties and its cooling effect. In-bath blowing of oxygen did not succeed, however, and top-blown oxygen converter with oxygen lance (LD or BOF) was first developed in he 1950&apos;s and emerged rapidly to a dominating position. Somewhat later oxygen bottom-blowing via special shrouded tuyeres was developed (OBM or Q-BOP) but has attained only fractional share. Bottom-blowing technique had, however, impact on the current combined blown converters which adopted the benefits of both top and bottom blowing. In copper converting Peirce-Smith type converters have kept their position until today although new processes have been introduced and at least few of them seem to be emerging. In one respect copper converting has a clear leading position i.e. in the progress of continuous converting processes (Mitsubishi, Ausmelt). These processes utilise top-blowing lances instead of in-bath blowing tuyeres. A different principle for continuous converting is flash converting by Kennecott Outokumpu, which makes smelting and converting independent of each others. Solid ground matte is used as a feed for flash converting in a reactor shaft similar to Outokumpu matte flash smelting. A comparable process has been presented by Inco too. The paper discusses principal physico-chemical and process technical similarities and differences in copper and steel converting. Although these processes are metallurgically quite different it seems that during the modem history of ferrous and non-ferrous metallurgy, starting from the 1850&apos;s, the exchange of knowledge between these disciplines has been quite intensive. In long-term perspective, many similarities in the development can be found but also divergences due to different ambient circumstances.

Jan 1, 2003

By Pablo González, Rodolfo Espinosa, Jose-Luis Noyola, Sergio Cossio

"Minera Penmont, at La Herradura, is a joint venture between Industrias Peñoles, S.A. de C.V., and Newmont Mining Corporation. The mine is located in the northern Mexican state of Sonora. La Herradura consists of an open pit mine with two-stage crushing and heap leaching facility, the gold is recovered from the pregnant solution using the Merrill-Crowe process; production began in July 1998. La Herradura produces about 130,000 gold ounces per year, and treatment capacity is 22,000 tonnes/day. Ore mineralization occurs in a shear zone with quartz and native gold. Gold distribution is quite heterogeneous due to the coarse visible gold, in some coarser particles ranging from 100 to 600 microns.Sampling and assaying protocols where established during exploration and prior to starting up operations, based on nomographs developed from early heterogeneity tests. But, as the coarse gold problem became noticeable, it was necessary to develop new sampling protocols to have an acceptable relative standard deviation to control the fundamental error within ± 15%.An experimental program was developed and conducted on a selected ore block made of 195 blastholes; regular ore grade control procedures were applied to both sets of data to evaluate the difference in results between the protocol established prior to starting up operations (“existing”) and the new protocol (“leaching”). In the “existing” protocol, subsampling is done through riffle splitters, and coning and quartering. Fire assaying is used to estimate the gold content, on a 50 g sample. In the “leaching” protocol a 12,000 g sample was bottle rolled cyanide leached, and the leach residue was assayed on a 50 g sample using fire assaying, and a weighted average total gold content was calculated. Currently, a new alternative method is being evaluated, where sub-sampling is done using rotary splitters, and fire assay on a 50 g sample; in this “rotary sampling” protocol and for comparison purposes, gold assays were also repeated using the “leaching” protocol.Test results were compared in three groups < 0.30 g/t (below cut-off), 0.30 to 0.8 g/t, and >0.8 g/t. Overall. Results show that the protocol established prior to starting up operations underestimates the gold assays. Detailed technical results, alternative options, and their implications are discussed in the paper."

Jan 1, 2003

By Hans Hein

In its early stage of copper solvent extraction development, SX plants were designed in a series configuration and gradually the copper industry reduced staging and trains for keeping capital investments reasonably low. At a later date of its development, series parallel configuration was applied in some copper SX plants, particularly for increasing copper production at low copper tenor in pregnant leach solution. A gap between series and series parallel configuration had been observed for years. Conventional SX plants in series configuration frequently do not have the flexibility for maintaining designed or expected production level, whenever copper concentrations in pregnant leach solutions are getting low, whereas series parallel flow schemes almost require to double SX feed flows for achieving optimum results. A combined SX configuration breaks the gap and its application potentials are explained.

Jan 1, 2003

By David Humphreys

Assumptions about future commodity prices need to be made with close regard to the purposes for which they are being used. Prices generated for financial planning or trading purposes are unlikely to be suitable for the longer run purposes of project evaluation and reserve estimation. In this latter context, price assumptions need to be realistic, stable and rooted in a rigorous fundamental analysis of long run industry drivers. Amongst the factors to be considered in such analysis are price history, production costs and structural dynamics. It is also important that prices adopted are consistent with other information being used by a company in its decision-making.

Jan 1, 2003

By John S. Pack

The last 25 years have seen rapid expansion in the use of helical screw pile foundations and tiebacks for new and retrofit construction of buildings and earth retention structures. An estimated $100 million was spent in 2002 on helical foundations and tiebacks in the United States alone. While design information is available in the literature, information regarding quality control, inspection and performance monitoring is lacking. Recognition of at least one manufacturer&apos;s helical foundation material by the Uniform Building Code, Standard Building Code, and BOCA (all three codes now combined into the ICC International Building Code) has given added urgency to the dissemination of reliable and practical quality control, inspection and performance monitoring procedures. This paper describes a set of such procedures that have been followed for 17 years by engineers experienced in the industry who have directly installed or been involved in the installation of over 140,000 helical foundations and tiebacks in divers soil conditions for various types of structures, including earth retention structures. Topics covered include design responsibility, building code listings and requirements, manufacturer ISO certification, manufacturing process, material and installation specifications, installation contractor certification, field placement, installation procedures, field inspection, building code inspection and performance monitoring. The paper concludes with a step by step example of practical quality control, inspection and performance monitoring procedures on a typical helical foundation project.

Jan 1, 2003

By Carlos F. Perucca

"Potassium is one of the three basic plant nutrients along with nitrogen and phosphorus. There is no substitute for potassium compounds in agriculture; they are essential to maintain and expand food production.Potash is extracted from buried ancient evaporites by underground or solution mining. This accounts for most of the potassium produced. Another important source is brine from landlocked water bodies, such as the Dead Sea, Salar de Atacama or Great Salt Lake.About 95% of potash produced worldwide is used in agriculture. The rest is found in several other industrial uses, including glass manufacturing, soaps, plastics and pharmaceuticals.Currently, three producers: IMC, PCS and Agrium operate 8 conventional and two solution mines in Saskatchewan, five of which run froth flotation circuits, two heavy media-flotation, two run forced crystallization and one natural crystallization.Potash in SaskatchewanPotash was discovered in Saskatchewan in the early 1940’s while drilling for oil. The deposits lie diagonally across the southern plains of Saskatchewan sloping southerly from a 1,000m depth at Saskatoon to more than 1,600 m depth at Belle Plaine and up to 3,000 m in depth in North Dakota. The known deposits are massive, accounting for more than 40% of the world’s known reserves."

Jan 1, 2003

By Bjarne Ekberg

"The most economical way of dewatering mineral concentrates is to reduce the moisture content in the material mechanically using thickeners and filtration avoiding thermal drying whenever possible. In addition to this, the dewatering process should produce low moisture cake, high quality liquid, consume little energy. As a part of the dewatering process The CERAMEC® capillary filter has been proven to meet these requirements in over 150 installations. The filtration is done by using ceramic filter plates with small pore size. The capillary action in the filter medium secures gas free filtration at good vacuum, and due to the small pore size, filtrate is crystal clean and particle free. The energy consumption is minimal. This paper will discuss results from filters installed at Pyhäsalmi Mine Oy in Finland where the technology was has been used since the late eighties. Results from more recent installations like Los Pelambres in Chile and LKAB in Sweden will also be discussed.INTRODUCTIONThe idea of using capillary action to make dewatering more efficient was invented two decades ago. Ten years later whole filter plants utilizing the new method were delivered to mine concentrators for dewatering of mineral concentrates. Further development of ceramic membrane plates made it possible to meet the requirements in iron ore dewatering. In addition to low cake moisture, high capacity and low energy consumption, there are a lot of environmental advantages in using capillary filtration. One of them is clear filtrate.CAPILLARY FILTRATIONA micro-porous filter material behaves like a large number of narrow capillaries forming a network. When a pressure difference ?P is applied over the wetted, hydrophilic pore with the radius r, the force striving to push the water out of the pore will be ?Ppr2. The force striving to hold the water in the pore will be the vector force 2pr? cosa of the surface tension ? and the cosine contact angle a. Equalling those forces and solving the equation for ?P we get"

Jan 1, 2003

By Takashi Furuyama, Sukeyuki Mori, Takafumi Okuma, Norio Hashiguchi

The Tower mill KD-3 is an improved version of the laboratory mills VD-1, KD-1 and KD-2 in which the classifying column has been modified to ensure a quick discharge of products from the comminution section and, on the other hand, a better separation of coarse and fine particles; the latter being collected in a cyclone with a central spiral ribbon. To level the circling airflow through the classifying column and have a strict control over the maximum size of collected particles, a series of experiments was carried out, modifying the internal structure of the column: insertion of a vertical plate, a cone, or a horizontal net. The relationship established between the maximum size of products collected in the cyclone and the duct airflow rate indicates that, independently of the latter, the insertion of a cone in the classifying column could stabilize the maximum size of products collected in the cyclone around 15.1 - 17.5µm. With a horizontal net, the maximal size could be reduced to 8.4 - 11.3µm. This occurrence might result from the collision of coarse particles blown from the comminution zone with the base of the inserted materials and their settling down the bottom of the classifying column.

Jan 1, 2003