Search Documents

By Q. Wu

Neural network methodology is used to predict mean particle size resulting from rock blast fragmentation. A blast data base developed in a previous study is used in the current study. A part of this blast data was used to train a single-hidden layer back propagation neural network model for each of the similarity groups obtained in the same previous study. Levenberg-Marquardt algorithm provided the most stable and efficient training out of the four algorithms evaluated. An extensive analysis was performed using a part of the blast data to estimate the optimum number of units for the hidden layer for each similarity group. The remaining blast data are used to validate the trained neural network models. Capability of the developed neural network models is determined by comparing predictions with measured values and predictions based on one of the most applied fragmentation prediction models appearing in the blasting literature. Prediction capability of the neural network models was found to be strong and better than the existing most applied model. Diversity of the blasts data used is one of the most important aspects of the developed models. The developed neural network models are suitable for practical use at mines.

Jan 1, 2011

By N. O. Lotter

The reliability of a set of flotation results is key to minimizing project risk either at concentrator commissioning, or at the point of implementing an improvement to an existing concentrator operation. High-Confidence Flotation Testing was developed for this purpose, and is part of the list of technologies of various scales used by Xstrata Process Support (XPS) in addressing these projects. The methodology is based on two principles: one, to ensure that the ore sample is representative and has been well-blended and subsampled; two, to perform the flotation tests in sufficient numbers of replicates with appropriate quality controls, so as to improve the reproducibility of the test data. Across ten years of practice and continuous improvement with this methodology at Xstrata Process Support, Sudbury, several improvements to this method and complementary connections into sampling and quantitative mineralogy have been developed. Two case studies are reviewed. These show the capability of this system in both optimizing an existing concentrator operation and in serving a new mine scenario. Further work to improve the efficiency of this procedure has been identified.

Jan 1, 2009

By W. Pariseau, S. Puri, S. Schmelter

Introduction Joints often have profound effects on elastic properties and strength of rock masses and therefore on rock slope stability. In surface mine slopes, joints are usually too numerous to be taken into account individually, so an equivalent properties approach is necessary. Previous work that treats a jointed rock mass as an equivalent composite material of joints and intact rock between resulted in excellent agreement between estimated and true rock mass elastic moduli in case of fully persistent joints (e.g., Pariseau, 1999) Rock mass failure mechanisms on joints and through intact rock were retained obviating the need for an equivalent rock mass strength. Impersistent joints, which are not continuous on a joint plane, can be accommodated (Pariseau, 2003). However, slope stability analysis of a large, deep copper mine indicated a need for computational efficiency even in two-dimensional analysis of vertical sections through pit walls (Puri, 2006). The analyses by the popular finite element method used small elements of bench size near the pit slopes and much larger elements away from the pit walls. The reasons for the graded meshes were numerical quality (small elements) and computational economy (large elements). Small elements contained only a few joints, while the larger elements contained hundreds of joints and led to impractical computation run times (days!). A new modeling procedure that recognizes sufficiently large elements as representative volume elements (REV’s) overcomes this obstacle. A companion improvement recognizes the concept of a representative area element (RAE) in association with joint impersistence and the probability that joint material occupies a considered site. Both are steps towards greater model realism and a practical slope stability analysis procedure. Consider the open pit mine photo shown in Figure 1(a), a companion plan view in Figure 1(b), and the problem of slope stability where high slope angles favor economics but low slope angles favor safety. These competing factors in surface mining generally result in some sections of the mine that have unstable slopes that require monitoring and perhaps mitigating action to improve stability. Although stability of a given section may not be easy to quantify with a single number, useful design guidance can be obtained from a finite element analysis of a proposed mining sequence leading to future pit slopes and profiles. The presence of water from rain and snow melt indicates a need for a coupled analysis that links water pressure, effective stress and total stress (Schmelter, 2001).

Jan 1, 2007

By Hyung-woo Kim, Sup Hong, Sung-Soo Kim, Chang-Ho Lee

A deep-seabed integrated mining system, which consists of four subsystems such as riser pipe, buffer, flexible riser, and tracked vehicles, is presented in Figure 1. In order to collect more manganese nodules, multiple tracked vehicle systems should be used in a deep-seabed integrated mining system.

Sep 1, 2014

By Eduardo Pimenta de Melo, Rambabu Pothina, Rajive Ganguli

Sensors are everywhere in the mining industry, with sensor information being used to monitor and operate processes. Therefore, when sensor information is wrong, economic losses can occur. Unfortunately, sensor errors are usually not detected till they become large. This is problematic as most calibrate their sensors no more than once a year (Beamex, n.d.). Using principles of data mining, where all relevant information is tapped to glean hidden information, Pothina [8] designed an algorithm to detect errors in temperature sensors in gold stripping circuit in Pogo mine, Alaska. This paper continued his work by analyzing the behavior of the algorithm on baseline data and testing the algorithm in new data and under more rigorous conditions. It also made a change to the algorithm. The modified algorithm performed very well in the new data. It also worked well under the new error conditions. Three types of errors were seeded, a fixed additive error (+ 2%), a fixed subtractive error (− 2%), and a noisy, normally distributed error, with a mean value of + 2%. Artificially seeded errors were detected within about 20 gold stripping cycles. Inherent bias in operation impacted algorithm performance by increasing the number of cycles needed to detect errors. This paper reinforced Pothina’s [8] conclusion that when data mining approach is used, sensor errors can be detected even when they are pretty low. Significant economic losses can thus be minimized.

By C. Pritchard

John Marks said in his acceptance of the Hartman Award in 2008 ? ?I guess that without the occasional complaint, your mine is probably over-ventilated? (Marks, 2008). Miners are seldom in this situation and often spend much of the time trying to find enough air to keep the operation running. With examination there are some things that can be done to make the best use of ventilation systems. Should there be surplus air, in which some of these options may result - financial and other benefits can be obtained. Some of the issues mine operators have needed to address recently are increased airflow requirements to dilute diesel particulates in metal and non-metal mines. Also, operations are faced with demands from reserves that are deeper, hotter and further away from fans and shafts, not to mention increased production demands. Mine ventilation air is a costly commodity, especially by the time it has been heated or cooled and moved to the bottom of the shaft and through the mine airways. Most of the options presented in this paper are useful for metal non-metal operations but some may be applicable to coal too. OPTIONS FOR IMPROVEMENT Some of the areas to be addressed are shop ventilation methods, auxiliary equipment areas, dedicated intakes and returns, optimizing development, examination of mine airway utilization and alternative ventilation methods. UNDERGROUND SHOPS First, examine what is being done with the underground shop air. If it is being coursed directly to the returns there may be some potential to better utilize this, in most cases, relatively contaminant free air. Research the legal options for ventilating mine shops. MSHA regulations in 30CFR 57.4761 (1) allow the following options: To confine or prevent the spread of toxic gases from a fire originating in an underground shop where maintenance work is routinely done on mobile equipment, one of the following measures shall be taken: use of control doors or bulkheads, routing of the mine shop air directly to an exhaust system, reversal of mechanical ventilation, or use of an automatic fire suppression system in conjunction with an alternate escape route. The alternative used shall at all times provide at least the same degree of safety as control doors or bulkheads. Discussion This regulation gives some operators flexibility in ventilating underground shop facilities. If shop air is to be coursed through the work area, and then directly to the mine return, the question should be asked ? is the quality of air good enough to be reused in the mine, and if so, what are the risks? Can the risks be accommodated and managed to accomplish the change? If so, a considerable increase in mine level airflow could be attained by reusing shop air. Clearly, regulations are a minimum standard, and can be exceeded. Metal non-metal mine regulations are not as specific as coal due to the multitude of different mining methods and conditions utilized. Risks must be thoroughly examined and addressed to assure a safe environment. Issues Determine what contaminants are being produced in the work area ? welding fumes, diesel exhaust, paint fumes, chemicals and solvents, etc. Survey the area and determine exposures. MSHA requires exposure monitoring in 30 CFR 57.5002 (1) of dust, gasses mists and fumes to assure air quality is being met. Perform surveys and examine existing data to quantify personal and area concentrations of contaminants. Often there is adequate circulation such that exposure levels in shop work areas are low. After sending shop air back into the system any contaminants would be further lowered by dilution with other mine intake air downstream. Look at the shop work schedule. If only on day shift, the other two shifts and possibly weekends have no activity or source of contamination to affect the ventilation system. The worst case planning scenario would be a shop fire. Study and simulate how this would affect the mine and shop area. MSHA regulations require an escape route and a fire suppression system be installed and maintained as follows in 30 CFR 57.4671 (1): (d) Automatic fire suppression system and escape route. If used as an alternative, the automatic fire suppression system and alternate escape route shall meet the following requirements: (1) The suppression system shall be? (d)(1)(i) Located in the shop area; (d)(1)(ii) The appropriate size and type for the particular fire hazards involved; and (d)(1)(iii) Inspected at weekly intervals and properly maintained. (2) The escape route shall bypass the shop area so that the route will not be affected by a fire in the shop area. Case Study In an underground room and pillar mine, active mine workings were advancing further away from the mine shaft with deteriorated returns and leakage having a negative effect on ventilation airflows to the face. Mine level airflows were 236 m3/s (500,000 cfm) and shop airflows were 24 m3/s (50,000 cfm) (see Figure 1A). No additional capacity was available at the main surface fan. A solution was to utilize the shop air then send it to the mine areas instead of directly to the return. Considerable information on shop air quality and exposures from many years of operation was examined and no negative data was discovered. A risk analysis determined that moving forward with a shop air reuse program was desirable. A plan was developed, analyzed and decision to proceed approved by mine management. The change was simulated on the computer with the USBM/MTU MFire program before being implemented. This included utilizing fire scenarios to examine potential exposures to inby mining sections and plan escape routes. A monitoring system was designed and installed in cooperation with the US Bureau of Mines to monitor carbon monoxide

Jan 1, 2009

By Bruce Gellar

For the past six years, the author has made it a priority to attend the Prospectors and Developers Association of Canada’s (PDAC) annual conference. This most recent meeting took place March 10-13, 2002in Toronto, Ontario. In the three years since the author’s last PDAC report, the mining industry has undergone a severe trough. Some the industry’s woes included deeply discounted commodity prices, rampant layoffs, mergers, and overall avoidance by the investment community. However, at each of the PDAC conferences, the author has felt a heightened sense of excitement about the mining industry, more so than at other similar gatherings.

Jan 1, 2002

Arsenic (As) removal behavior from Changgoon mine tailings was investigated using Acidithiobacillus ferrooxidans (A. ferrooxidans) in well-controlled batch reactors by varying pulp density (0.5?4%, pH=1.8, temp=30oC), reaction initial pH (1.8?2.2, pulp density=0.5%, temp=30oC), and temperature (25?35oC, pulp density=0.5%, pH=1.8). The initial concentration of As in the mine tailings used for bioleaching tests was about 30000 ppm, and the size range of tailings used in this study was 63?74 µm. The reaction speed and initial cell concentration were fixed to 150 rpm and 107 cells/mL, respectively. To complement leaching experiments, pH, oxidation/reduction potential (ORP), and amount of ferrous/ferric ions in leachate were analyzed. Scanning electron microscopy and x-ray diffraction analyses were also conducted for the samples before and after leaching to see if any morphological and compositional changes in tailings took place during leaching process. The As leaching efficiency increased with decreasing pulp density, decreasing pH, and increasing temperature in the presence of A. ferrooxidans. Specifically, no significant change in As removal rate was observed when pulp density d 2% while the removal rate dramatically decreased at pulp density of 4%. Regarding pH effect, As removal rate was much more enhanced at lower pHs (i.e., 1.8 and 2.0) compared to pH 2.2. Increasing temperature was observed to tend to enhance As removal rate, showing about 20% enhanced leaching rate under 35oC compared to 25oC and 30oC at the elapsed time of 5 days. The results for As removal behavior at different conditions (i.e., pH, temperature, pulp density) were consistent with the change in pH, ORP, and ferric ion concentration in leachate, showing greater removal rate at the condition with lower pH, and higher ORP and ferric ion concentration in leachate. Furthermore, decreasing removal efficiency was likely in part due to the extent of the formation of by-product (i.e., jarosite) of ferric ions.

Feb 27, 2013

By T. J. Lynch

A new class of polymeric dispersants for industrial minerals has been synthesized and evaluated. Currently available dispersants give initial low viscosities at high mineral loadings, however slurry viscosities can rapidly double or triple on standing over the first several days. The new dispersants give low initial viscosities that are more stable and maintain a lower viscosity for at least one month. The new dispersants are made possible by controlling the polymer molecular weight distributions during their manufacture such that essentially all of the dispersant is adsorbed onto the mineral surface, preventing depletion flocculation.

Jan 1, 2012

By Wilfred P. Schmoe

Introduction An integrated oil company has two options in approaching strategic planning. It has to decide whether it is in business for the long-term. I state this in all seriousness. Those who help manage public companies have the responsibility of good stewardship of other people's money. We must decide whether continued investment in our basic business will generate a reasonable rate of return. With a world surplus of petroleum production capacity, and continuing downward pressure on prices, this question should be addressed by all oil companies. And a positive answer is not necessarily a foregone conclusion in all cases. The decision should be made based on proven competence in finding and developing petroleum reserves on a cost-competitive basis with other producers, and on a frank analysis of future prospects. This should include a realistic assessment of the projected rate of return on specific investment projects. If a company did not pass these tests, its strategy could take one of four paths. Diversification First, it could diversify. But recent history suggests that oil companies are not good at managing other businesses. Second, it could seek a buyer for its business at the optimum return for its owners. Third, it could sell its assets on a piecemeal basis. And, fourth, it could adopt a longer-range plan of self-liquidation by living off current recoverable reserves. In other words, by not investing in exploration and development, it could maximize short-term profit and pass the money back to shareholders in the form of dividends. The Conoco component of DuPont has a below-average cost of finding petroleum, a good track record of efficient production, and a full plate of projects that pass the test of return-on-investment. In short, Conoco is in business for the long-term. That fact, in itself, determines at least one key element in strategy. This is to extend the life of the petroleum reserve base while at the same time maintaining maximum efficient rates of production to provide a strong cash flow. A company that is able to maintain a strong reserve base will ultimately see the value of those reserves escalate as world oil supplies tighten. Our planners expect this tightening to occur within the next decade as demand grows and production in non-OPEC areas of the world begins to decline. An exploration program tailored to meet the requirements of reserve replacement and current production should have two basic parts. The first is relatively low-risk exploration in proven oil provinces. This is augmented by

Jan 6, 1985

By Youngtak Ko, Wonnyon Kim

Korean contract Mn-nodule field in the NE equatorial Pacific is composed of seven sectors showing average water depths of 4,513 - 5,025 m. Of the various factors controlling the property of Mn-nodules, water depth has a potential linkage to chemical composition, shape and abundance of nodules. To clarify such a possibility, we review more than 20 years of research data in one of the northern sectors (hereafter KR1) where there is a marked contrast in water depth. High-resolution seabed mapping clearly separates a northern part from a deeper southern part, cutting across the middle of the KR1. In terms of nodule occurrence, manganese nodules in the deeper north are comparatively larger (2-4 cm) with rough surface (t-type) and discoidal shape (D-type), while those in the south are generally small (<2 cm) with smooth surface (s-type) and irregular shape (Itype). In addition, the nodules in the north have higher metal contents including Cu, Mn and Ni. Such connections of water depths to nodule occurrence and metal contents are commonly observed in the Korea licensed Pacific nodule fields. However, the relation between nodule abundance, size and water depth may be a different story. In KR1, nodule abundance and size are probably controlled by morphological features especially in the north. Seamounts in the north continuously provide rock fragments as new nuclei for the nodules. As a result, the nodules can not grow larger than 2 cm and show the shapes of newly formed nodules (i.e., smooth surface and irregular shapes). Such an interpretation is supported by anomalously high backscattered intensities in the side-scan sonar image map of the north, which possibly reflects a younger generation of magmatic activity.

Sep 1, 2014

By William R. Yernberg

The current state of the copper industry was the topic of discussion at the Copper 99 — Cobre 99conference held Oct. 10 to 13, 1999, in Phoenix, AZ. More than 700 copper-industry engineers, executives, consultants, academics, vendors and other interested parties attended. This successful and well-organized conference was the fourth in a series of international conferences devoted to the mineral processing and extractive metallurgy of copper. The 1999 meeting marked the first time that the conference was held in the United States. The meeting was co-organized by the Minerals, Metals and Materials Society (TMS); the Chilean Institute of Mining Engineers (IIMCH) and the Metallurgical Society of the Canadian Institute of Mining, Metallurgy & Petroleum(CIM). The meeting was also sponsored by The Society for Mining, Metallurgy and Exploration Inc.

Jan 1, 2000

By Charles L. Allie

Since Reserve Mining Co. began commercial operations at Silver Bay, MN, some 25 years ago, major improvements in taconite processing technology have become available. In addition, significant new government standards have been imposed on industrial operations. By implementing a $370-million modernization program, Reserve was able to accomplish two major company objectives: the quality of concentrates produced from the Peter Mitchell mine has improved, and the operation meets both present and foreseeable environmental standards. Process modifications at Reserve were designed to: •Reduce the pellet silica content from 9% to 5%; •Increase the pellet iron content from 62.5% to 65.8%; •Implement a tailings disposal system utilizing "best available technology" water quality controls; and •Reduce air emissions by installing "best available technology" air emission controls. By adding dry cobbing, fine screening, and flotation to the concentrating process, pellet silica content has been reduced; the process reclaims and reuses all process waters; all tailings are stored in a zero discharge tailings basin; and all air emission standards are being met. Initial Beneficiation at Silver Bay Reserve&apos;s Peter Mitchell mine at Babbitt, MN, is on the eastern end of the Biwabik iron formation. The open-pit mine is approximately 16 km long, 0.8 km wide, and presently has a maximum depth of about 50 m. Principal constituents of the ore body are magnetite, quartz, and iron-magnesium silicates. The company&apos;s large commercial plant began operating at Silver Bay, 75 rail km from the mine, in 1955. Facilities have historically had the capacity to process 31.5 Mt/a of taconite, containing an average 24% magnetic iron, into 11 Mt/a of iron ore pellets. Reserve&apos;s initial beneficiation process-which existed up to early 1979-consisted of two crushing stages at the Babbitt mine site, two additional crushing and screening stages at Silver Bay, grinding in a conventional rod-ball¬regrind mill circuit, size classification with cyclones, and concentration with magnetic and hydraulic separators. Concentrate produced by that circuit assayed 65% iron and 8.5% silica, with a screen analysis of 94% -44 µm, and a surface area of 9300 cm2/cm3. Pellets produced from this concentrate assayed 62.5% iron and 9% silica (dry basis). Construction began in 1977 to modify Reserve&apos;s operation. Process modifications were designed from pilot scale operations that had been operated and researched by Reserve for more than 10 years. There were two design restraints imposed on the process: economic constraints that Reserve continue to use much of the existing processes and structures in the new operation; and a substantial number of government-imposed permit constraints. In effect, the tailings process was designed first and then a concentrating process was developed that would "fit" the tailings disposal system. Operation of the modified process began in June 1980. Crushing Remains Unchanged No changes in process flow were made at either the Babbitt mine and crushing plant or at the Silver Bay fine crusher. Gyratory crushers reduce run-of-mine taconite to -90 mm-diam at the mine. Taconite is then transported to Silver Bay by intra-plant railroad. There are 10 fine crushing lines. Each consists of a double deck screen, followed by a heavy duty short head crusher, a second double deck screen, and second short head crusher. All screen undersize material bypasses the crushing lines, resulting in an open circuit crushing process. Fine crusher product is approximately 98% - 20 mm, 80% -12 mm, and 50% -10 mm in diameter. Each fine crusher line processes about 400 t/h of crude taconite. Cobbing Provides Aggregate for Tailings Dams The first major process change was the addition of dry cobbing, a dry magnetic separation step between the fine crusher and concentrator. Dry cobbing serves two major purposes in the new process. Its principal function is to provide large volumes of dry coarse aggregate for use in construction of tailings dams. This process rejects about 20% of the fine crusher product as tailings with a size analysis of 98% -20 mm and 50% - 10 mm. These tailings are free draining and have high frictional shear strengths-two key requirements for dam building aggregates. Fine crusher product is conveyed to a new dry cobbing building. Here crushed ore is discharged into one of two 30-minute capacity surge bins. Crushed ore is withdrawn by 12 constant speed conveyor belts; each discharges 500 t of crushed taconite into a dry cobbing machine. Each machine consists of a 3-m-long riffle splitter and two vari-

Jan 12, 1980

By V. McLemore, N. Dunbar, K. Donahue

There are three principle lines of evidence that indicate the clay minerals found in the Goathill North (GHN) rock pile at the Molycorp mine are predominantly a product of hydrothermal alteration, not weathering. The unweathered drill core and samples from GHN have similar types and abundances of clay-mineral groups based on XRD analysis. Chemical analyses from the electron microprobe reveal similar chemical compositions for the hydrothermal alteration clays in drill-core samples and clays found in the rock pile. Textural evidence gathered from petrographic and microprobe examination also suggests a hydrothermal origin for clay minerals.

Jan 1, 2007

By Russel Delmar

Construction of Ontario Power Generation?s 12.8-m diameter, 10.4-km long water diversion tunnel in Niagara Falls, Ontario, is underway. The tunnel is being excavated by means of a 14.44-m diameter open gripper Tunnel Boring Machine. The tunnel lining is constructed as a two-pass system with the initial lining consisting of rock bolts, mesh, steel channels or ribs, and shotcrete, and the final lining consisting of pre-stressed cast-in-place concrete with a waterproof membrane. This paper describes salient design and construction aspects as follows: ? Regional and Site Geology ? Tunnel Layout and Design ? TBM Tunnelling and Cast-in-place Concrete Lining Construction

Jan 1, 2008

By Miles T. Stephens, Felipe Calizaya, Jacob O. Kingston

Currently, the Bear Canyon Mine is facing the challenge of excessively high coal and rock temperatures in one of its headings. In Mine No.3, the air temperature at the 6th Right panel reached 36.4ºC and the coal 83.9 ºC. Due to these severe work conditions, the mining plan has been altered by changing the mine direction to avoid the high temperatures. To investigate the problem, drill holes were developed from the entries and the rock temperatures monitored using K-type thermocouples. The measurements confirmed the existence of localized “hot spots” in the mine. This paper presents a summary of ventilation surveys that were conducted in this mine. It includes the results of coal and rock temperature measurements from drill holes, interpretation of results using a heat flow simulator and the identification of hot spot areas. It also includes a summary of methods that were developed to reduce the hazard and control the problem.

Jan 1, 2005

By Magno Rodrigues Ribeiro, Alexandre Cabral Stelzer, Cacio José da Silva, Leonardo Nogueira Ferri, Marcelo Ciminelli Mendes, Christophe Bouché, Marco Aurélio Soares Martins, Evenilson Soprani Lopes, José Xavier da Silva Filho, Sérgio Luiz Dalvi

Since 2001, CVRD is carrying out a modernization project in it’s pelletizing plant’s automation systems, that comprehends instrumentation, control and information systems. In addition, the project includes implementation of advanced control systems, using artificial intelligence techniques, including fuzzy logic, expert systems and dynamic physical models, to increase production, reduce fuel and energy consumption and improve quality. This paper reports CVRD’s experiences in carrying out an optimization project.

Jan 1, 2005

By William R. Yernberg

The combined 75th Annual Minnesota Section of SME meeting and 63rd Annual University of Minnesota Mining Symposium was held April 9 and 10, 2002, in Duluth, MN. This year’s theme was “Keys to the Future: Imagination, Innovation and Technology.” The meeting consisted of plenary and technical sessions and a special General Manager’s Session. The technical paper presentations included concurrent mining and mineral processing sessions. In the General Manager’s Session, representatives from the region’s taconite operations presented the current status, concerns and future plans for each of their operations.

Jan 1, 2002

Several of the larger copper projects recently built have included a molybdenum byproduct circuit as part of the overall project due to the favorable molybdenum market conditions and overall economic contribution to the project. This paper provides a summary of moly plant flotation practices across a number of both recent projects and historical ones. Modern Molybdenum flotation flowsheets predominantly incorporate self-aspirated cells and forced air cells with nitrogen addition. The paper provides insight into the process considerations, capital and operating costs associated with both of the cell types. It also reviews metallurgical aspects and important safety aspects. Other aspects including instrumentation and degree of automation are also discussed along with overall plant design.

Feb 27, 2013

Federal regulations require the use of methane monitors on all mining machines. All machine-mounted methane monitors currently used in underground coal mines rely on catalytic heat of combustion sensors. As a comparison, although not currently approved for use on underground mining equipment, instruments with infrared sensors are evaluated in this study. Potential advantages of using infrared sensors include faster response time, long-term stability, and reduced maintenance needs. Tests conducted in the laboratory and at a full-scale model mine compare the performance of catalytic heat of combustion and infrared sensors operating under test conditions designed to simulate airflow conditions underground.

Jan 1, 2009