Search Documents

By Christopher Mark

Coal bursts are typically associated with highly stressed coal. Most bursts occur during retreat mining (longwall mining or pillar recovery) in highly stressed locations like the tailgate corner of the longwall panel. Others are associated with multiple seam interactions. However, a small but significant percentage of coal bursts have occurred during development or in outby locations unaffected by active mining. Most development bursts have been relatively small, but some have been highly destructive. No theory of coal bursts can be complete if it does not account for this type of event. This paper focusses on the development mining coal burst experience in the US, putting it into the context of the entire US coal burst database. The first documented development coal burst occurred almost exactly 100 years ago during slope drivage at the Sunnyside Mine in Utah. Sunnyside subsequently had a long history of bursts, mainly during retreat mining but also during development. Several Colorado mines have also experienced multiple development bursts. Some, but by no means all, of the development bursts in these western US coalfields have been associated with faults. In the Central Appalachian coalfields, most development bursts have occurred in multiple seam situations. In some of these cases, however, there was no retreat mining in either seam. The paper closes with some lessons from this history, with implications for preventing such events in the future.

Jan 1, 2017

By Stephen C. Tadolini, John P. McDonnell

"High-strength cable bolts have been used for ground support in underground coal mines for many years. Cable bolts are routinely used as secondary ground support in numerous longwall gateroad support applications, areas that possibly experience the most severe conditions in current underground coal mining. Additionally, when mining ground conditions deteriorate, engineers oftentimes recommend cable supports as a principal solution because of the established credibility for successful ground control. Nevertheless, there is still a reluctance to utilize non-tensioned cable supports as an effective primary roof support alternative.This paper provides historical and recently documented examples of non-tensioned cable supports utilized as an effective primary ground support method in a western U.S. underground coal mine, and it will continue the technical discussion between the differences of various degrees of tension in cable supports with respect to ground behavior and reactions.INTRODUCTIONResin-grouted cable supports are widely used to support difficult ground conditions throughout the world. Through the use of innovative applications and configurations they have proven to be a major paradigm shift in roof support and ground control technology in the U.S. (Tadolini and Koch, 1993; McDonnell et al., 1995; Tadolini and Gallagher, 1994; Tadolini and Trackemas, 1994)Today, the basic cable bolt support consists of a high-strength steel cable installed in bore holes that range in diameter from 25 to 50 mm (1 to 2 in) depending on the strand size and drilling equipment. Cables are offered in both bright and galvanized, for corrosion protection, and range in capacity from about 30 to 60 tons. An example of a resin-grouted cable bolt and the key components is shown in Figure 1. The steel cables are flexible and the support length is not limited by the height of the opening. Lengths of up to 10 m (32 ft) have been supported with high pattern densities when standing supports were not operationally feasible. The most common lengths are 3 to 5 m (10 to 16 ft). An important physical characteristic is that PC-strand has an ASTM minimum elongation requirement of 4%. For comparison purposes, the ASTM 432-08 Standard Specifications for Roof and Rock Bolts and Accessories specifies elongation requirements for traditional roof bolts are a minimum of 12% for grade 40, 9% for grade 60, and 8% for grade 75 materials, respectively. Traditional roof bolts can stretch and elongate twice the length that cable bolts can, which makes design considerations challenging."

Jan 1, 2010

By Bruce Hebblewhite, Jim Galvin

"A coal burst occurred on 15 April, 2014 at the Austar Coal Mine, located west of Newcastle, NSW, Australia. The burst resulted in fatal injuries to two men working as part of the mining crew at the development face. At the time, a continuous miner was being used to mine a longwall development gate road through heavily structured coal, at a depth of approximately 550m. A number of pre-cursor bumps had occurred on previous shifts, emanating from the coal ribs of the roadway, in proximity to the coal face.This paper reviews the geological, geotechnical and mining conditions and circumstances leading up to the coal burst event; and presents and discusses the available evidence and possible interpretations relating to the geomechanical behaviour mechanisms that may have been critical factors in this incident. The paper also discusses some key technical and operational considerations of ground support systems and mining practices and strategies needed for operating in such conditions in the future.INTRODUCTIONAustar Coal Mine is an underground longwall coal mine located near Cessnock in the Hunter Valley of New South Wales (NSW), Australia and is the only underground mine still extracting the Greta Seam in this region. The mine was the first in Australia to adopt the Chinese-developed Longwall Top Coal Caving (LTCC) method for thick seam extraction. Typically seam thickness ranges from 4m to 7m and depth of mining from 480m to 560m, with future mining planned down to depths of up to 700m, making it one of the deepest operating coal mines in Australia.On 15 April 2014, a pressure burst occurred in the left hand rib at the active mining face of B Heading, 2 to 3 cut-through, Maingate A9 panel, during development of the gateroads for the ninth longwall top coal caving panel. Strata in the general vicinity was affected by disturbed geology and multiple geological structures. Figure 1 shows a section of the mine plan as at October 2014 (six months after the accident), indicating the current longwall extraction panel (AS) and the development panel A9 where the accident occurred. (Note: Neither development nor longwall face positions changed significantly between the time of the accident and the date of this plan). At the time of the accident the current longwall face was in excess of 1,000m away from the Maingate A9 development panel face position."

Jan 1, 2016

By Raj R. Kallu

Underground gold mines in Nevada are typically hosted in geologic zones with a wide range of geotechnical properties ranging from blocky competent rock to faulted and highly altered soil-like material. Geotechnical engineering design in weak rock utilizes rock mass classification systems as inputs to both empirical and numerical methods. The most common rock mass classification system utilized in the gold mines of Nevada is the easy-to-use Rock Mass Rating (RMR) system; however, in practice it is difficult to apply in weak to very weak rock masses. This paper attempts to improve the sensitivity of RMR in weak rock masses by utilizing W-RMR?a modified procedure for calculating RMR?and evaluates its performance against other methods of determining RMR for predicting in situ moduli. In-situ rock mass moduli were determined using a portable plate loading device (PPLD), and these data were used to quantify the improved sensitivity of the modified RMR versus existing classification systems. The PPLD was designed for performing rapid and economical in-situ deformability tests in a mining environment. Tests were performed at eight sites yielding thirteen rock mass moduli measurements. Statistical analyses indicate Detailed-Discontinuity RMR-89 and W-RMR provide the best correlations to the in-situ moduli compared to other methods of calculating RMR.

Jan 1, 2014

By Bret E. Leisemann

The geomechanical environment in which highwall mining must operate represents a hybrid between surface and underground operations. Mine design must take into consideration the complexities associated with unsupported spans, pillars with extreme aspect ratios and the instability related to the interaction between coal extraction and the highwall stability. The rock mass adjacent to the highwall, and for a significant distance in, is affected by lateral stress relief and blast damage. This paper describes a rigorous site investigation program aimed at providing some of the input parameters for the design of highwall mining layouts at the Moura Mine and focuses upon defining the nature and extent of the relaxed zone behind the highwall. The combination of borehole imaging and C factor analysis allows delineation of blast damage and relaxation of the rock mass immediately adjacent to the highwall excavation, as well as the more subtle relaxation effects deeper into the rock mass. This has led to the formulation of a proposed geomechanical model for the relaxed zone for direct input into highwall instability analysis and mine design.

Jan 1, 1993

By William Monaghan

Over the period 2000 to 2003, roof falls have accounted for 4 to 14% of the fatalities in underground mining operations. The National Institute for Occupational Safety and Health (NIOSH) is conducting research to reduce the frequency, exposure, and risk of these events through an ongoing program of field and laboratory studies. One area of research involves the evaluation of polyurethane grouting technology that is commonly used to stabilize fractured mine roof strata. Concurrently, NIOSH is conducting work to evaluate the application of ground penetrating radar for advanced delineation of problematic mining areas. In this study, NIOSH partnered with Sub-Technical, Inc. who injected a polyurethane grout (also known as glue) into a roof area of NIOSH's Safety Research Coal Mine. The mine roof area was scanned using ground penetrating radar technology before and after grout injection in an attempt to determine the extent of grout infiltration. A comparison of the pre-and post-grouting radar records showed a significant change at a mine roof depth of about four to five feet. The interpreted radar records were then compared with drill core information, borescope evaluation of roof-bolt holes in the study area, and underground observations. At this site, the interpretations of the radar records correlated with data obtained from the core holes, borescope evaluations and underground observations. This study showed that ground penetrating radar technology can be a useful tool for detecting changes in mine roof due to the injection of the grout.

Jan 1, 2004

By Marc T. Filigenzi

Two of the major ground control safety issues confronting underground mine operations today are shaft pillar stability and the failure of rock around active mine openings. Failure of a mine shaft can lead to the entrapment of workers. Failure of rock around active underground mine openings can lead to roof falls, which in turn can result in worker injuries and fatalities. Finite-element analysis has proven to he a reliable method for predicting stresses and displacements around underground mine openings. However, this is a complex and time consuming technique and is not used as often as it could be in mine planning. The purpose of this paper is to demonstrate one technique developed at the Spokane Research Center that allows the user to create a finite-element model of a two-dimensional section of an underground mine in a relatively straightforward manner. This model is then used to calculate stresses, displacements, and safety factors around mine openings. With this information, mine planners can evaluate the stability of mine openings as well as the stability of pillars and mine shafts. This analysis will help develop mining sequences and layouts that minimize stresses in that section of the mine. This, in turn, should minimize the occurrence of shaft failure, roof falls, and other hazards associated with underground mining.

Jan 1, 1997

By D. G. Lynch, M. R. Martin, R. J. Coutts

"Broadmeadow Mine is an underground longwall mine located in Central Queensland, Australia. The mine introduced a top coal caving (TCC) longwall face in 2013 but since then has experienced severe convergence events at the start of each panel after 60–70m of retreat, resulting in equipment damage and the longwall almost becoming iron bound. A significant geotechnical monitoring effort during LW10 and LW11 to characterise the caving mechanics and operational changes during LW11 were successful in mitigating the risk posed by these initial convergence events. This paper describes the geotechnical monitoring programme undertaken during LW11, the operational methods used to mitigate these initial convergence events, and the contribution to these events of the TCC technique.A thick sandstone unit close to the extraction roof was likely the cause of the convergence events at the start of each longwall panel. However, stress change monitoring cells, surface extensometers, an inclinometer, and the blockage of holes designed for pre-conditioning using hydraulic fracturing indicated that the convergence events were not caused by the sudden collapse of sandstone bridging across the goaf. Rather, they were caused by the propping action of strong units within the overburden strata. Equipment design modifications were made to allow the longwall to remain operational during the convergence events and an operational strategy was implemented to mitigate convergence in LW11. The initial convergence event was successfully negotiated, and Broadmeadow went on to be the highest producing longwall for CY16 in Australia.INTRODUCTIONBroadmeadow Mine is an underground longwall mine located in Central Queensland, Australia, owned and operated by BHP Mitsubishi Alliance (BMA). The mine is a punch longwall mine using open cut pit access provided by Goonyella/Riverside Mine. Broadmeadow commenced longwall operations in 2005 using a conventional high reach longwall, but later changed to the top coal caving (TCC) method of extraction. Broadmeadow extracts the easterly dipping 5.5–7.5m-thick Goonyella Middle Seam in ranging in depth from 90m to 330m. The coal is washed and blended at the adjacent Goonyella/Riverside open cut mine to produce a premium coking coal product for export."

Jan 1, 2018

By Y. P. Chugh

The use of wooden supports, supplementary to the primary method of roof control using roof bolts in mines, consumes over 3 million cubic meters of hardwood in the U.S. annually. These supports arc primarily used in the form of cribs and posts. Along with a rapid depletion of high quality timber resources, supplies fluctuate seasonally, costs are escalating rapidly, and product quality is highly variable. Environmental considerations of global climate change put additional pressure on deforestation for utilization of timber for artificial supports. On the other hand, management of coal combustion byproducts (GCBs) from power plants has become a major issue to maintaining a healthy coal industry. The USA currently produces over 100 million tons of CCBs annually and the average cost for their management in the US is about $20/ton. This cost is expected to escalate rapidly with tougher environmental regulations. The principal author has been directing research, development, and demonstration studies for CCBs-based artificial supports and lightweight structural materials in mines for the past seven years and has made significant progress toward commercially viable products. More specifically, first and second generation crib elements and post elements have been developed, characterized analytically, and demonstrated in the field. It is estimated that about 2.5 x 10" tons of fly ash could be utilized for artificial supports in U.S coal mines and this would double if these materials were utilized in non-coal mines as well. The author believes that CCBs-based artificial supports have the potential to replace wooden supports and blocks with better performance and cost competitiveness while managing CCBs in an environmentally sound manner. This paper focuses on the development and field demonstration of CCBs-based artificial supports for use in underground mines, and its pilot scale fabrication facility. Recent analysis results on engineering performance evaluation of the developed crib under various loading conditions using finite element method are presented. The following comments highlight the developments: I ). Engineered artificial supports, crib and post elements, using CCBs as a major constituent, have significant potential for application in mines because their structural performance is significantly better than wooden cribs counterparts. 21. Engineered crib elements provide about 50% larger area for air-flow than a conventional wooden crib. 3). Engineered crib elements are not flammable. 4). Each engineered crib element utilizes about 10 kg of fly ash, about 40 % of the weight of crib element- Therefore, a typical crib 2.0 m high utilizes about 200 kg of fly ash. 5). Each 16.5 cm diameter engineered post utilizes 50 kg of fly ash. 6). Assuming 3.0 million cu. meter of wood consumed for post and crib support systems, it is estimated that about 4 - 5 million tons of fly ash may be utilized for supports fabrication in the USA. 7). An engineered post has excellent post-failure characteristics similar to wood. In addition, the engineered post also has excellent loading and unloading characteristics close to the yield point of about 75% of the ultimate load carrying capacity.

Jan 1, 2004

By S. M. Matthews

The magnitude and orientation of the in situ stressfield has been determined as a key factor in controlling the stability of openings for both coal mine development and extraction. Monitoring of roadway and pillar behaviour in conjunction with numerical analysis has identified specific stress control measures that have been successfully applied to improve productivity in underground coal mining operations. The influence of the major horizontal stressfield on the stability of openings has been identified at a wide range of sites including mines in Australia, Britain, the USA, New Zealand and Japan. Stress mapping, stress measurement and stress monitoring techniques have been applied to define the in situ stress regime and, together with the correlation of measurements of rock deformation and rock reinforcement behaviour, has enabled appropriate stress control measures to be developed. Stress control measures for development headings have required the selection of roadway orientation, roadway sequencing, reinforcement density and pillar size. The monitoring of behaviour at a range of trial sites has confirmed significant improvements in development stability. Stress control measures for longwall extraction have required rational selection of longwall orientation and extraction direction. Where a mine layout design incorporating the optimum longwall layout design has not been possible, the use of sacrificial roadways has been successfully applied to protect key roadways from stress concentration effects. Comprehensive in situ monitoring of behaviour has been carried out to confirm the mechanisms involved.

Jan 1, 1992

By Douglas Morrison

Hardrock mines in Ontario generally operate at between 3,000 and 7,500 ft below surface and generally experience significant rockburst activity below 5,000 ft. Since the introduction of bulk mining techniques around 1980, the safety record of Canadian hardrock mines has improved dramatically. However, the process of improvement was not a smooth one and, in the early years, the number of accidents and fatalities actually increased. One incident in 1984 that resulted in 4 fatalities in Ontario initiated a complete re-examination of mining practice and by the mid 1980s many changes had been introduced, driven by new non-prescriptive mining legislation. These changes affected not only the technical discipline of mining engineering but also the management of operational procedures and, together, they resulted in the significant improvements that have continued to this day. The analysis of the rock-fall related fatalities showed 5 major causes (loose, scaling, installing support, collapse and rockburst) and by the mid 1990s fatalities in all of these categories, except rockbursting, had been eliminated and even then, all these rockburst fatalities occurred in one mine which was closed shortly afterward. This paper discusses both the technical and social reasons for the initial failure and the ultimate success in improving the safety of these mines, with implications for mines in other major mining jurisdictions.

Jan 1, 2003

By Rajendra Singh

In India, pillar mining is contributing more than 90% of underground production which involves extraction of a part of coal from bords with pillars around as natural support in the development stage. #Liquidation of these pillars with splitting and stooking, broken nature of face during mining under complex geomining factors like massive roof, varying treatment of goaf and over or underlying workings result in high stresses and deformation in and around the neighbouring piilars, causing operational problem and hazards to mine safety. This paper suminarises the results of three case studies of depillaring under shallow depth with or without stowing. Assessment of nature and amount of mining induced stresses and deformation was done under these conditions, by insitu instrumentation and observation at three depillaring panels of Girmint. Samla and Lachhipur collieries of ECC at 54, 90 and 110 m depth cover respectively. The value of induced stresses and associated deformations was comparatively low in well caved and stowed waves, whereas for massive roof the roof fall followed overriding of pillars and stooks, with very high value of induced stresses and deformaton in absence of stowing. The stress in the centre and outer edges of the pillars changed with respect to depillaring face advance or splitting of pillars. The caving panel of Girmint colliery under weak coal measure formation, showed maximum induced stress in the pillar at 4 m distance from the face which was 1.2 times (16.6 kg/sq.cm) the preminlng stress of the area. The main roof fall over partially stowed goaf of Lachhipur colliery under massive roof strata caused overriding of pillar after crushing 39 stooks of 7.5 x 7.5 sq.m in rythmic sequence with development of 133lsq.cm maximum induced stress. Well stowed Waf of the same panel reduced this value of abutment stress to 2.3 times (65 kg/sq.cm.) body stress after 40 m further advance of the face from the crushed zone. The depillaring at Samla colliery even under massive roof strata induced low stress, equal to 0.8 times field stress only under the same condition with stowed goaf. The studies revealed the effectiveness of stowing in the goaf over loading of pillars and stooks.

Jan 1, 1990

By David Bigby

The paper describes research undertaken on an ECSC/K HSE project which investigated several potential physical principles which might be harnessed to provide a non-intrusive means of determining the extent of failure in the near roof of mine roadways and tested the most promising in the laboratory and in the field. It was concluded that the principles which warranted further development were thermal response and laser vibration measurement. In the latter case there is a need to develop an efficient means of imparting energy to the seam to induce sufficient vibration. A new, SIMRAC funded Project will progress work in the field of thermal response.

Jan 1, 2004

By W. G. Pariseau

A new approach to the direct inclusion of the effects of joints and geologic structure in rock mechanics analyses is applied to an elastic analysis of stability of a Lucky Friday undercut and 4111 longwall stope (WL) using the finite element method (FW). The approach is base+ on the concept of equivalent properties of a heterogeneous test volume, but does not depend on the assumption that a test volume is a representative volume element. Comparison with conventional FW results show greater variability in fields of stress, strain, displacement, safety factor and energy densities. Stability requires a knowledge of rock mass strengths as well as elastic moduli.

Jan 1, 1989

By Hideki Shimada

A low cost methodology that allows the determination of in- situ stress using oriented core specimens has been investigated. The technique can be used to determine the stresses during the early stages of a project, even in undeveloped areas of a mine. The research aim was to compare the experimental results estimated by the AE and DRA methods to those estimated by conventional ANZI (Australia and New Zealand Inflatable) cell measurements. The studies have been focused on fill stress tensor determination from a single oriented cored rock. In all cases the rock core specimens recollected the in-situ stress reasonably well compared with the standard overcoring method.

Jan 1, 2001

By Rossen A. Halatchev

This study introduces a new probabilistic approach to seismic stability analysis of embankments and slopes. The approach is based on Sarma's solution, which is modified through presentation of the seismic force by its horizontal and vertical components; i.e. the force may have an arbitrary inclination. The probabilistic modelling includes Monte Carlo simulation of the shear strength parameters of the soil mass, which are assumed to have a normal distribution. The probability of failure is determined based on the seismic co-efficient, which is treated as a random quantity. The probabilistic approach is illustrated through a solution of a numerical example from an opencast coal mine.

Jan 1, 1992

By Gift Makusha

Pre-feasibility work at Goedehoop Colliery indicated that stooping of the undersized pillars could be undertaken safely and economically. In light of the findings a decision was taken in may 2003 to undertake the project in the 600 area where three panels (628, 630 and 632) were identified for pilot test sites. The findings of the pre-feasibility studies were presented in a paper to the 22nd International Conference on Ground Control in Mining held at Morgantown in 2003 (1). This paper is a progress report for the project 12 months down the line. Work commenced to prepare the sites for mining including, acquisition of equipment, equipping the panels, selecting and training people to undertake the project. Three monitoring sites were set up distributed evenly along panel 628. Each site comprised of a surface extensometer station in the overburden strata adjacent to a pillar underground containing hard inclusion vibrating wire extensometer cells. This combination would be used to record pillar loading ahead of stooping as well as the depth of caving and strata deformation in the goal zone behind the stooping face. In July 2003, support work began to secure the roof and rib sides. The support was required to be at least three-pillar center distances ahead of the mining front to ensure that the effective stooping zone is secured at all times. The NEVID method of pillar extraction developed at Sasol and described in detail by Ismet and Moolman (2) in a Coaltech 20-20 report was adopted for the pilot project. As part of training the selected stooping team was send for a week to Sasol's Bosmanskrans Colliery to familiarize with the NEVID method of stooping. Further training sessions were carried out on site especially on the use of Mobile Roof supports. The first cut was taken on 20 August 2003 by implementing a form of chequerbord mining taking every second row of pillars. Three rows were mined this way which period was considered training to get workers familiar with the new equipment and mining method. On the fourth row the full design span equivalent to mining 4 pillars in a row was implemented taking out all the pillars in successive rows. Production figures gradually increased from 300 to more than 1,000 tons run of mine coal per shift. Panel 628 was mined out at the end of December 2003 and the second of three panels (630) is being mined. Substantial data from the monitoring set up has been recorded since the start of the project. Monitoring will continue to the end of the pilot project and the data will be analysed and fed back into the design tools to improve the design in future undertakings. The pilot project has been a success so far and other old workings are being evaluated for stooping.

Jan 1, 2004

By Lorraine Kent

Comparison of in-situ stress overcore measurement results with those determined from laboratory evaluation of Acoustic Emission (AE) testing on oriented sub-core samples has been carried out for a variety of mining environments. These include UK and German Coal mines at moderate depth, shallow-level South African Coal mines and deep-level South African gold mines. Results show that the `Kaiser Effect' or 'Stress Memory-Damage is readily determinable throughout the brittle deformation process, although particular care is required in interpretation of results during the 'bedding-in' phase of testing and the region associated with unstable crack growth prior to failure. The results suggest that AE provides a useful complimentary method for comparison with existing methods of stress determination. The success of the method can, however, be dependent on both rock type and the likely magnitude of in-situ stress in comparison to the strength of the rock. Keywords: in-situ stress, acoustic emission, Kaiser Effect

Jan 1, 2002

By C. J. H. Brest van Kempen

The techniques developed should provide a useful tool, not only during the initial formulation of a suitable bolting plan for a new section, but also for periodic check¬ing on the utilization factor of hardware installed in the roof. Indications are that with present practice, utilization seldom exceeds 507.. Retro-fittable bolter equipment now available (FCBS?) allows doubling of the utilization factor in many cases. We are also introducing a computer pro¬gram based on the material outlined in this paper. This program can save a great deal of time in examining the real effect of implementing alternate bolting plans or using different installation techniques. It allows fine-tuning of bolt lengths, spacing and installation tension to mini¬mize roof support cost, while maximizing safe life of the mine opening. We have developed a procedure which allows easy, site-specific quantification of the need for reinforcement of a mine roof. The concept is based on the forma¬tion of a direct link between theory and empirical data to custom-fit each mine section. In its simplest form, the proce¬dure requires only recording of roof bolt torques in a sample working place (16 to 25 bolts) several times during the first week or so after initial bolting. The way in which the torque pattern changes during this time can reveal whether or not the bolts carry enough tension to maintain stability in that specific roof. From the same data, the nature of the anchorage horizon is derived. Specifically, shale or limestone can be distinguished from sandstone. If the procedure is repeated a few times (within the same mine section) with different bolt installation tensions, we can numerically characterize the roof in terms of one single value: an "effective angle of internal friction". This number can then serve to calculate trade-offs in bolting plan design, such as bolt length, bolt spacing and optimum bolt tension. Additional refinement is possible if a sagmeter is installed in the sample work¬ing place being monitored and if bolt anchorage capacity is determined. The lat¬ter can be done automatically on each bolt as it is being installed, using a system we developed and patented. Alternately, a number of pull tests could be performed. Using the sagmeter, bolt torque and anchor¬age data, it is possible to project the safe life of the supported opening. The details of the procedures described apply specifically to tensioned roof bolting, but a brief discussion paints the way for a similar analysis of full-column resin bolting also.

Jan 1, 1986

By Karl Jr. Zipf

The famous Homestake gold mine in Lead, SD. closed recently alter 125 years of operation. However, the mine may receive a new lease on life as a National Underground Science Laboratory (NUSL) supported by the National Science Foundation (NSF) and dedicated to high-energy physics experiments involving neutrinos. These experiments require the construction of many large underground chambers tells of meters wide at depths of over 21 18 m. Construction of such large openings at such extreme depths has few precedents m civil-engineering-type construction. The Spokane Research Laboratory has a long history of ground control research at the Homestake Mine and has supported NSF and the physics community in their efforts to develop a NUSL at I Homestake or other suitable location. This paper will summarize underground science and describe proposed underground science laboratories ill the United States, in particular tile proposed NUSL at Homestake. Next, the paper describes an empirical design approach using the Q system of rock mass classification for these large Openings. .M initial site investigation that produced three good case histories is described. The actual support system used at the three sites agrees well with support recommendations from current support design charts, giving credibility to estimates of the classification parameters and Q. Feasibility of, and support requirements for, 25-, 50-, and 100-m spans arc estimated in the anticipated rock mass. This analysis indicates that 25- and 50-m spans are probably feasible, but a stable 100-nn span may be unachievable. The empirical predictions are then compared to detailed numerical analyses. The paper concludes with preliminary support recommendations and necessary site investigations needed to make a proposed NUSL a successful reality. These studies have been important input to NSF for their decisions on the technical feasibility of developing the proposed NUSL. If NUSL is developed, it could enable scientific experiments in areas beyond high-energy physics. 'the laboratory could also enable research in rock mechanics. ground water hydrology. mining engineering, nine safety and health, and ground control.

Jan 1, 2002