Search Documents

By A. Andrews

Paper written on project work carried out in partial fulfilment of MSc (Eng) degree Solid state sintered silicon carbide ceramics are used in aggressive environments as sealings. There exist data concerning corrosion, but there are only few data concerning their electrocorrosion behaviour. Therefore the electrochemical corrosion of solid state sintered silicon carbide (SSiC) ceramics has been studied at room temperature in acidic and alkaline environments by using potentiodynamic polarization measurements. The investigation showed a pronounced electrochemical corrosion in bases and acids. In HCl andHNO3pseudo-passivity features due to the formation of a thin layer of SiO2on the surface were observed, whereas in NaOH soluble silicate ions were observed, resulting in more pronounced corrosion. The microstructural observations of initial and corroded samples reveal that the residual carbon found in the microstructure do not dissolve preferentially. Keywords: Corrosion; SEM; SSiC; polarization test.

Jan 1, 2006

By Paulo von Krüger

Nas condições adversas do mercado atual, aliadas à exacerbada competição é uma questão de sobrevivência a otimização de resultados. Para tal é imprescindível que os principais insumos, particularmente os mais gravosos sejam criteriosamente selecionados. Os eletrodos de grafite estão entre os itens principais na operação dos fornos elétricos a arco, tanto do ponto de vista técnico, quanto econômico e a escolha correta deste insumo deve contemplar aqueles dois enfoques. Esta escolha se faz dentre as diversas classes, disponibilizadas no mercado: RP, HP, SHP, UHP. O que se pretende no presente trabalho é sugerir uma metodologia que, contemplando o elenco de parâmetros influentes, permita fazer a escolha correta do eletrodo, dentro daquelas classes, tanto sob o enfoque técnico, quanto econômico.

Aug 17, 2017

By L. S. Hansen

Organized efforts toward the prevention of injuries and occupational diseases have been of comparatively recent origin in the mining industry, dating generally from the beginning of this century. Mining-and particularly underground mining- is more hazardous in many respects than most other major industries, and continual and increasing attention has been given to its health and safety aspects. The importance of the physical well-being of those engaged in mining was recognized by Agricola, who, after describing the hazards encountered in 16th century mining operations, stated that "Some of these evils, as well as certain other things, are the reason why pits are occasionally abandoned." In the past 50 yr particularly, mine managements and labor organizations, separately or collectively, have mounted productive efforts toward the prevention of accidents and industry-associated diseases and illnesses. A statistical study of this period indicates that increasing awareness coupled with the endeavors of all concerned agencies has resulted in decreasing frequency of occurrences of injuries and illnesses. In recent years particularly the mining industry has used the services of safety engineers, industrial hygienists, inspectors, ventilation engineers and other specializing in locating and correcting deficiencies affecting the welfare of employees. Concurrently, the labor movement has incorporated into its organization individuals working in the fields of health and safety, many devoting full time to these responsibilities. Others serve in behalf of their fellow workmen as on-the-job safety representatives. Most authorities agree that accidents are caused by (1) unsafe acts, (2) unsafe conditions and, exceptionally, (3) Acts of God. The "unsafe act" often is cited as the prime cause. but many accidents, and attendant injuries, are the result of a blend of "unsafe acts" and "unsafe conditions," with each chargeable in varying proportions. It therefore is necessary to consider both the physical elements of the mine plant and the training, mental attributes and actions of all personnel involved. Analysis of an accident, if based on hasty reconnaissance and unduly hurried questioning, seldom will disclose more than surface causes and consequently will result in inadequate measures to prevent repetition. Mining involves unavoidable and calculated risks, as in all facets of life. It is most important then that men employed in mining be schooled in the hazards and develop fully the ability to evaluate and minimize risks. The environment, of underground mining often is harsh and strange compared to the more natural surface conditions man has learned to cope with as a part of established life patterns. The miner. therefore, must adapt to unusual situations and conditions and must develop a mental attitude permitting acceptance of change. 3.1-INDUSTRY-AFFILIATED SAFETY AGENCIES L. S. HANSEN There are many organizations primarily interested in health and safety. In addition to industry and labor groups, there are many private and governmental

Jan 1, 1973

By A. Nhasengo

Paper written on project work carried out in partial fulfilment of B. sc. Eng (Mining Engineering) The purpose of this investigation was to assess the mining strategy for steep reefs at Mine A mine in 2005 and 2006. Reefs steeper than 35° were discovered when production commenced at the mine. At that time reefs steeper than 35° were not popular in South African tabular gold mines, therefore new techniques of mining them had to be developed. The underhand mining method was chosen for this purpose. The project was about assessing ventilation, support costs, local support cycles, adherence to the mine?s support standards, mine planning and efficiencies of the underhand mining method. The project work was covered by doing a literature research for underhand mining methods and performing a literature review of the mining methods, support methods, ventilation standards at Mine A. Underground visits were conducted at Mine A mine for data collection, and Mine B was visited to obtain information on support costs and efficiencies. During the investigation it was realized that the underhand mining method was an appropriate method for mining steep reefs; however, there were some challenges related to ventilation and support installation. Efficiencies could be realized with the underhand mining method and planning and scheduling were not different from breast mining.

Jan 1, 2007

The objective in control blasting is to reduce overbreak to control the final pit wall slope, shaft, drift, ditch, bench, etc. to the final planned excavation limit.

Jan 1, 1990

By A. K. Raji, M. O. Ojumu

Robotic subsea dredging crawlers are dynamically and remotely controlled vehicles that are used for deep sea mining and recovery operations. These exploration machines are released from a mother ship and move around on the ocean floor using tracks. Current ocean crawlers such as the MK3 ROST are hydraulically powered. In this paper we develop a scaled-down model for simulating and performing static loading analysis of an electrically powered robotic subsea dredging crawler (EPRSDC). The modeling, simulation, and analysis were carried out using modelling software from Solidworks. The structural frame was assembled using the Tetrix max robotics kit. The kit’s structural components were produced from 1050 aircraft-grade aluminum. The results were used in optimizing and for considering other materials, and the to identify specific areas to be reinforced in future crawler designs.

Oct 3, 2023

By D. E. G. Maschmeyer

The operation of the pilot plant for the S-C Copper Process during 1976 has provided the basic data for the design of a commercial-scale plant. This paper introduces the commercial aspects of the S-C Copper Process through the presentation of a process flow diagram, a schematic flow diagram and basic design parameters for 75,000 tonnes per year (tpy) copper rod production from chalcopyrite concentrate. Basic operating requirements are outlined and a preliminary estimate of operating and capital costs for a Western Canadian location are summarized. The results have been compared with a modern smelter/electrorefinery to highlight the advantages and disadvantages of these two alternatives and to show why the new S-C Copper Process technology should be of interest to the copper industry.

Jan 1, 1978

By using knowledge-based technology to create a Virtual Mining Enterprise and integrating it with existing technical and financial computer systems, a gold mining company has been able to reduce the time taken to produce a mine plan from 14 weeks to less than one week. Typically, the mine planning process requires circulating and managing large amounts of information. Even with the use of computers in todayÆs business, this is often still a manual process involving file transfers and individualised spreadsheets. Mine planning also involves understanding and managing large quantities of complex and interrelated variables. Added to this is the ever changing dynamics and constraints associated with the business of mining. These issues not only hamper the planning process but also question the validity and usefulness of the mine plans being generated. By integrating the information and knowledge of the numerous departments involved in mine planning into one core Knowledge Base - so called æone version of the truthÆ, - mining companies are able to overcome these pitfalls experienced in traditional planning methods. Once set up, the Knowledge Base creates and runs a virtual mining enterprise with which to visualise and test various scenarios of mining, business and logistical activities. The final output is an optimised enterprise-wide plan with clear and detailed instructions for each of the divisions of the company.

Jan 1, 2000

By T. McCallum

Crud is a common concern in SX processes and is formed as a complex solid stabilized emulsion of aqueous and organic.1,3 Crud formation is highly dependent on the amount of solids entering the circuit and is influenced by a number of other operational factors. A common practice for crud processing is interfacial pumping, wherein crud, organic, and aqueous are pumped from the settlers for treatment and utilizes a variety of equipment. These mechanical means of crud processing are effective, however, require significant processing time and require oversight to prevent the return of contaminated organic. Cytec has developed a unique crud treatment process utilizing both chemical and mechanical means to enhance the solid/liquid separation. The use of ACORGA CB? 1000 crud busting reagent allows a rapid separation of solids from the organic phase and provides similar volume recovery to current practices. The process allows operations to return clean organic back to the plant more efficiently and may enable operations to process more crud. In addition, the organic quality of the recovered organic is higher than that recovered by typical means, which can lead to operational cost savings.

Feb 27, 2013

By Regina Krachler, Herbert Ipser

"The principles of the derivation of statistical-thermodynamic models to interpret the compositional variation of thermodynamic properties in non-stoichiometric intermetallic compounds are discussed. Two types of models are distinguished: the Bragg-Williams type, where the total energy of the crystal is taken as the sum of the interaction energies of all nearest-neighbor pairs of atoms, and the Wagner-Schottky type, where the internal energy, the volume, and the vibrational entropy of the crystal are assumed to be linear functions of the numbers of atoms or vacancies on the different sublattices.A Wagner-Schottky type model is used for the description of two examples with different crystal structures:For B'-FeAl (with B2-structure) defect concentrations and their variation with composition are derived from the results of measurements of the aluminum vapor pressure; the resulting values are compared with results of other independent experimental methods.For Rh3~ (with an NiAs-derivative structure) the defect mechanism responsible for nonstoichiometry is worked out by application of a theoretical model to the results of tellurium vapor pressure measurements. In addition it is shown that the shape of the activity curve indicates a certain sequence of superstructures.In principle, there are no limitations to the application of statistical thermodynamics to experimental thermodynamic data as long as these are available with sufficient accuracy, and as long as it is ensured that the distribution of the point defects is truly random, i.e. that there are no aggregates of defects."

Jan 1, 1996

By C. W. JEFFERS

At its annual meeting on March 5, 1928, in Washington, D. C., the Joseph A. Holmes Safety Association gave to the Ford Collieries Co., Curtisville, Pa., a certificate of honor for "having worked 922,628 underground man-shifts from August 14, 1920, to December 31, 1926, producing 5,755,669 short tons of coal without a fatality." This remarkable record certainly deserved recognition. It developed that this company's freedom from fatalities continued until October 2, 1929, hence the full period during which no underground fatalities occurred was from August 14, 1920, to October 2, 1929, or 9 years, 1 month, and 18 days, the total tonnage being over 7,700,000. Table 1, shows the complete safety record of the Ford Collieries Co. since 1920 as given by the officials of the company. The first item in Table 1 shows a remarkable performance. This company since 1920 has mined 15 times more coal per fatal accident than the average pro- duction per fatal accident for the State of Pennsylvania in 1928 (255,257 tons). Excluding the two fatalities (one in 1922 and one in 1929) the severity rate for the same period has been 1.259 per thousand hours worked, a record of which the company may justly be proud. However, including the fatal accidents the rate is 2.407. This compares favorably with an average of 12.4 in 1928 for 66 mines in the National Safety Competition. The Ford Collieries Co. operates three mines located in the vicinity of Curtisville, Allegheny County, Pa. One of these mines, Benjamin No. 1, is idle at the present time. The officials are: A. R. Pollock, general manager, Curtisville, Pa., and C. W. Pollock, assistant general manager in charge of safety and operation.

Sep 1, 1930

"Each and every flotation plant must produce daily/weekly/monthly production reports in which metal balances for payable and penalising elements are presented. Several factors including ore heterogeneity and harsh production environment affect the reproducibility of the reported data. Best practices of metal accounting thus prescribe that production reports be accompanied by measurement quality reports quantifying the reproducibility (a data quality metric often expressed as a relative standard deviation) of each data reported. Ignoring the irreproducibility of measurements may lead to the perception that random errors have been totally eliminated through statistical data reconciliation, which is wrong. This paper aims at explaining the full lifecycle management of reproducibility estimation from the outset at the measurement process to its reporting after having been through the statistical data reconciliation process. Indeed, the reproducibility of reported data can be obtained by performing a sensitivity analysis. Necessary conditions for obtaining a truly representative assessment of measurement reproducibility are stated. Typical levels of measurement reproducibility found in operating flotation plants are also reported. These numbers can be used as benchmarks for reproducibility levels that can be realistically expected.IntroductionEach and every flotation plant must produce daily/weekly/monthly production reports in which metal balances for payable and penalising elements are presented. Such production reports are the most expected output from a metal accounting system (MAS). The duly recurrence of this task combined with an insufficient level of automation has somehow contributed to the misleading perception that it is mainly a clerical work. However, as shown in Figure 1, a MAS encompasses way more than just a reporting system. Acknowledging that the content of production reports indubitably origin from raw measurements has important implications for how both the reports and MAS should be monitored."

Jan 1, 2016

By D. W. Bridges

Solvent extraction has become important in recent years as a unit process in extractive metallurgy. The flexibility and simplicity of the technique have attracted widespread attention, and numerous papers describing actual and projected metallurgical applications have been published. However, differences in terminology, procedure, objectives, and standards, as reflected in reports by different operators and investigators, have sometimes made appraisals and comparisons difficult. Consequently, a need appeared for n series of basic reports on the subject. This Bureau of Mines Information Circular is the first of such a series. It deals with the status of solvent extraction in metallurgy, as well as with the rudiments and terminology of the unit operation. In analytical chemistry, extraction of molecular species from aqueous media by immiscible organic solvents has been used for many years to separate and determine chemical elements (61).4 Complex chlorides of iron and other metals are readily soluble in ethers and ketones, as are the nitrates of uranium and thorium. Most metals are extractable as complex salts of some organic reagent, such as dithizone, 8-hydroxyquinoline, or cupferron.

Jan 1, 1962

By J. -F. Lagueux

Goldex Mine is an underground mine located near the town of Val-d?Or. Multiple innovations were required to work the mine. Due to its low grade, it was important to develop a bulk mining method to economically extract the ore. The method used is a combination between shrinkage stope and long hole stope methods. This modified method is more than adequate; however, all of the ore body must be liberated in a very short period of time. Therefore, to bring down 23 Mt of ore, large blasts up to 1.8 Mt are required. The situation is also complicated by the fact that the mine is located in a residential area. No less that 40 homes are built within a 1 km radius of the mine site. In addition to managing mass blasts, means must also be found to mitigate the vibrations produced by the blast to diminish the disturbance to the neighbourhood. Therefore, the mine used numerous tools to properly plan the blasts and reduce the impacts. A program to control vibrations was one of the first tools put in place. The program?s aim was to draw up an action plan to ensure a follow-up of the homes by inspecting them. The second tool was to develop a methodology to plot a reliable attenuation curve and to adequately determine the maximum charge weight for each blast. Finally, effective communication tools were developed to inform the neighbourhood. There was a huge amount of work to be done in relation to the bench. Reliable and fast loading techniques had to be found to load large blasts in a very short period of time. Adequate logistics had to be developed to send underground more than 20,000 kg of emulsion per day while performing the daily underground operations. Technology had to be developed to adapt the i-kon electronic blasting system to large blasts. Since the control of the charges was the major factor to reduce surface vibrations, it was important to employ loading techniques that would permit decking of the long holes to reduce the charge weight per delays without adversely affecting the loading. All of this work requires an enormous amount of energy but also close collaboration between the different stakeholders (various departments, suppliers and consultants).

Jan 1, 2011

By E. H. Cho

A current environmental concern is the cleanup and restoration of soil contaminated with low-level radionuclides at DOE sites including the Nevada Test Site. At the present, there is no applicable process to economically and effectively deal with the remediation of contaminated soils. This situation requires development of a selective flotation process to separate the radionuclides from contaminated soil efficiently and economically. In this study, a minus 200 mesh size fraction of pure cerium dioxide was used as a surrogate for plutonium oxide which is the major contaminant at the Nevada Test Site. A Hallimond flotation cell was utilized to examine the flotation behavior of cerium dioxide with oleic acid under various conditions of pH and collector concentration Mechanical cell tests were conducted to float the surrogate from a mixture with uncontaminated Nevada Test Site soil in the absence and presence of calcium carbonate as a carrier. Two different sizes of the carrier (minus 150 mesh and 65-150 mesh) were tested. It was found through the Hallimond cell flotation tests that the cerium dioxide could be floated well at pH's 4 and 9 with 10-3 molar oleic acid. It was also found that the addition of the carrier drastically enhanced the flotation recovery of cerium dioxide. The flotation recovery of the surrogate increased from 34 % to 65 % by adding only 0.5 % calcium carbonate (minus 150 mesh) in the feed and also increased to 95 % by adding 10 % of this carrier to the feed. The finer size fraction of calcium carbonate (minus 150 mesh) was more effective in recovering cerium dioxide than the coarser size fraction (65-150 mesh) The flotation method utilizing a carrier of calcium carbonate seems to provide a high potential to be used for the remediation of soil contaminated with radionuclides at various Department of Energy's nuclear weapon sites.

Jan 1, 1996

By A M. Lechner, S Arnold, M J. Gillespie, P D. Erskine, D R. Mulligan, A T. Fletcher, A Gordon

Mine ecological monitoring schemes are often conducted only to address compliance and in some cases do not have the sufficient statistical power or understanding of ecological systems to assess, quantitatively, the meeting of rehabilitation targets or mine impacts. Monitoring schemes for rehabilitation commonly rely on analogue sites or premining baselines (reference sites) to assess whether rehabilitation sites are ecologically similar and thus successfully rehabilitated. In the case of monitoring for mine impacts, if the impacted sites diverge from reference data, the site is considered to be impacted. Fundamental to any monitoring scheme is testing the sampling design to ensure that it can detect these differences if they are present through adequate sample sizes (ie statistical power). However, a good understanding of the ecological system is required to ensure that differences detected are ecologically significant. Modelling ecosystem properties can provide a quantitative understanding of how ecosystems change over time and space, in response to natural environmental variations and thus aid in determining significant ecological differences. Furthermore, modelling methods can be used to predict future ecosystem states. Within the ecological literature, guidelines for robust statistical monitoring designs and modelling methods exist but are not routinely being used in the mining industry. This paper uses simple examples to discuss: problems with current compliance monitoring methods; importance of robust statistical techniques; and how modelling can assist rehabilitation assessment and guide monitoring schemes. The aim of this paper is to describe general monitoring and modelling methods that could be potentially used for mine site monitoring and closure assessment. These methods are often considered by mining companies as an extra expense that provide little practical benefit, and the domain of academia; however, engaging these methods may be more cost-effective in the long-run than conducting untargeted monitoring schemes that give no confidence to regulators on whether rehabilitation criteria have been met. Furthermore, monitoring programs in mining contexts provide a good opportunity to better understand long-term ecological patterns and test the application of more rigorous modelling approaches as the life of many mines tends to be greater than ten years.CITATION:Lechner, A M, Arnold, S, Fletcher, A T, Gordon, A, Erskine, P D, Gillespie, M J and Mulligan, D R, 2012. Embracing modern ecological methods - Monitoring and modelling for mine closure, not compliance, in Proceedings Life-of-Mine 2012 , pp 25-32 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Jul 10, 2012

By Gary B. Glass

Seventeen percent of US coal production now comes from the 19 surface coal mines in the Powder River Basin of northeastern Wyoming. The Powder River Coalfield, which coincides with this topographic and structural basin, is unique to the conterminous US, in that it contains an estimated 934.4 Gt (1.03 trillion st) of in-place coal resources. At least 13.2 Gt (14.6 billion st) is a compliant, strippable reserve base. In 1993, an estimated 162 Mt (179 million st), or 98% of the coal mined in this Wyoming coalfield, came from the thick - average 30.5 m (100 ft) - subbituminous, Wyodak coalbed, or from the Anderson and Canyon coals in the one mine where the Wyodak is split into two separate coalbeds. The other 2.7 Mt (3 million st) of coal mined in the Powder River Basin in 1993 came from the School, Badger, Monarch and Deitz No. 3 coalbeds. In this "world class" coalfield, there are at least 13 other persistent, 3 to 15 m (10 to 50 ft) thick, coalbeds as well as the Big George and Lake De Smet coalbeds, which are each 61 m (200 ft) or more thick in places. None of these latter 13 coals are mined at this time.

Jan 1, 1994

Geological hazards such as faulting, basement ridges, and zones of "thin" (< 6 m) coal have a major impact on mining economics of underground operations in Huntly Coalfield. Experience has shown that drillhole-based investigations do not yield sufficiently detailed models of the coal seam to allow management of the planning risks associated with these hazards. Consequently, operational performance is affected by unplanned costs associated with lower productivity, loss of coal reserves, and expensive strata control remedies in problem ground conditions. Huntly Coalfield is a challenging environment for acquisition of good quality seismic reflection data due to the very thick (10-85m) weathering layer. Since 1994, successful acquisition of 2D and 3D data has been achieved through a combination of careful testing of technical parameters and experimental trials prior to committing to production recording. High resolution seismic reflection (HRSR) is proving to be an investigations technique which generates accurate and reliable models of the coal seam and associated structures, such as normal faults and topography of the surface on which the coal seam rests. The HRSR technique has been applied and developed in the Okowhao Sector where Huntly East Mine is developing and extracting coal reserves. Results of a recent 3D survey demonstrate that this technique is capable of revolutionising risk management in mine planning for underground mines in structurally complicated coal deposits.

Jan 1, 1997

By Jerry D. Pfarr, Gary Loving, David G. McDowall

Introduction The Open Cut is one of the best known landmarks in South Dakota's Black Hills. It is the site of the original discovery of the Homestake ledge, or lead, in 1876. Between that time and 1945, the Open Cut produced about 36 Mt (40 million st) of gold ore, the majority of which was drawn from underground operations. In 1981, a drilling program was undertaken to determine how much ore the old timers left behind. The results of this study indicated an ore body made up of broken and caved ground, and solid blocks in pillars and crowns. Reserves are estimated at about 6.8 Mt (7.5 million st) of gold ore, grading at 4 g/t (0.116 oz per st). Total tonnage of material, including waste, is estimated at more than 60 Mt (66 million st). The mining plan is a combination open pit and underground operations. The bottom 78 m (255 ft) will be mined by the underground panel-caving method. The open pit was designed as a contractor operation up to the expected milling stage or the point of sustained ore production. The pit was designed to avoid disturbance of US Highway 85 near Lead, SD. In 1983, the South Dakota Board of Minerals and Environment approved the mining and reclamation plan. Test mining began in the north of the Open Cut using open pit methods. Upon completion of the test pit, Homestake Mining's board of directors approved an expanded mine plan. They committed an additional $27.1 million to prestrip the ore body that is expected to yield 23 t (750,000 oz) of gold. Project life is estimated at 15 years. Exploration program In early 1980, 10 diamond drill holes were drilled from the surface in the vicinity of the east side of the Open Cut. The reserves delineated by that drilling were too small to justify their development. A continuation of this study attempted to quantify the tonnage remaining in the bottom of the Open Cut in the crushed remnants of old stope-and-crown pillars. This was originally thought to be minable by improved drawhole methods using highly mechanized procedures. By early 1981, however, thinking turned toward surface mining methods. The effort to quantify minable ore tonnage ultimately involved drilling 385 reverse-circulation rotary drill holes. These were drilled from August 1981 to March 1983 in several groups of holes. Each group, or drilling program, had a slightly different immediate goal or purpose. The holes were drilled to various depths using truck-mounted Drill Tech 40 drills with doublewalled 76 mm (3 in.) pipe with 134 mm (5.25 in.) three-cone rotary bits. Initially, wet drilling was employed using a cyclone collection system. Since cutting recovery was not as high as it should have been, all subsequent drilling was dry. The cuttings were collected and deposited by way of a cyclone into the sample bag. The pickup for the cuttings using the three-cone bits was about 50 mm (2 in.) above the bottom of the hole. Overall cutting recovery for the drilling program was estimated at 75%. Samples were taken on each 1.5 m (5 ft) of hole drilled. Each sample was split into three 48-g (1-2/3 assay ton) charges. The accepted value was the straight average of the three charges. Because of the known "nugget effect," assays were nearly impossible to duplicate. Therefore, to validate the regular assays, a series of screen fire assays were run. Based on this method of assays and checks, it was concluded that 10% more ounces would be produced than indicated by mine grade. Test pit The northernmost portion of the mine as planned is known as the Terraville test pit. This portion was mined ahead of the rest and separately. It provided answers to questions on slope stability, reserves correlation, mill recoveries, mining costs, and surface mining conditions in various weather and ground conditions since many boulders and much timber were expected. The mining plan for this portion of the pit was successful. More ore was mined than was expected. Mixing of Open Cut ore with underground ore in the mill feed was successful. Impact on the surrounding communities

Jan 10, 1987

By J. C. Edwards, C. C. Hwang

In ventilated tunnel fires, smoke and hot combustion products may form a layer near the ceiling and flow in the direction opposite to the ventilation stream. The existence of this reverse stratified flow has an important bearing on fire fighting and evacuation of underground mine roadways, tunnels and building corridors. In the present study, conducted by the National Institute for Occupational Safety and Health, a CFD program (fire dynamics simulator) based on large eddy simulations (LES) is used to model floor-level fires in a ventilated tunnel. Specifically, the critical ventilation velocity that is just sufficient to prevent the formation of a reverse stratified layer is simulated for two tunnels of different size. The computer code is verified by checking the computed velocity profile against experimental measurements. The CFD results show the leveling-off of the critical ventilation velocity as the heat release rate surpasses a certain value. At this critical ventilation, the ceiling temperature above the fire reaches a maximum for both tunnels. The velocity leveling-off can be explained from this observation. An extended correlation of Newman (Combust. Flame 57 (1984) 33) is applied to the temperature profiles obtained by CFD. At the critical ventilation, temperature stratification exists downstream from the fire. The computed critical ventilation velocity shows fair agreement with available experimental data taken from both horizontal and inclined fire tunnels. The CFD simulations indicate that the Froude modeling is an approximation for tunnel fires. The Froude-scaling law does not apply to two geometrically similar fire tunnels. The CFD results are compared with two simple theories of critical ventilation by Kennedy et al. (ASHRAE Trans. Res. 102(2) (1996) 40) and Kunsch