Ground control research continues to be a major part of the portfolio reviewed by SHMRAB. The majority of the work was carried out by Rock Mechanics Technology Ltd (RMT) through ECSC collaborative projects. UK Coal Mining Ltd and the University of Nottingham have also worked on certain aspects. Progress made during 2002 is described under various topic headings below. In some cases, the work programmes described drew on more than one project.
During the year UK research houses undertook work on eight safety-related ECSC research contracts under the following three-year collaborative targeted projects:
HSE contributed to the funding of some of these Projects through 4 contracts entitled:
During the year RMT also participated in an Australian Joint Coal Board research Project entitled, “Acoustic Energy Meter” aimed at evaluation of the instrument for detecting unsafe roof and sides in Australian coal mines and in a South African (SIMRAC) Research Project entitled, “The application of acoustic techniques for identifying rock related hazards in gold and platinum mines”.
The main technical advances during the year have been made under the following subject areas, each of which is discussed in more detail below:
Stability and Support of Mines Sides
This 3-year HSE/ECSC funded Project by RMT has just been completed and the final report is being prepared. The Project was set up partly in response to a fatality due to a rib fall at Maltby colliery in 1997. The Project has involved modifying roof fall hazard risk assessment survey and analysis techniques to incorporate rib fall hazards, developing improved instrumentation for rib monitoring, developing new rib reinforcement consumable testing methods, undertaking laboratory and underground tests on currently available consumables and improving numerical modelling techniques to analyse vertical stress distribution and simulate rib and pillar behaviour. In this part of the report we shall concentrate on the consumable testing aspects of the Project.
There has been considerable innovation in steel and glass reinforced plastic (GRP) reinforcement consumables in the last couple of years and an important part of this Project was to determine appropriate ways of testing these in the laboratory to allow objective comparison of their likely field performance. This work has been undertaken in association with the parallel development of a draft revised British Standard on rockbolting system components for use in coal mines (BS7861:1-1996) by a committee with representatives from HSE, manufacturers, users and research houses.
It was concluded that the Laboratory Short Encapsulation Pull (LSEP) test was a better measure of likely reinforcement system load transfer behaviour in the field than the current Double Embedment Tensile test. For rib consumables it was recommended that the standard LSEP test developed for roof consumables, which involves testing in a confined soft sandstone core, be used rather than a modified version designed for measuring performance in coal. Testing found that large differences between system bond strengths and stiffnesses seen in Double Embedment tests are usually less pronounced in the LSEP test but in some cases behaviour is quite different. This is attributed to the LSEP test taking greater account of the differences in normal forces generated against the borehole wall by varying rockbolt rib designs. Systems producing very high stiffnesses, which might be considered detrimental to reinforcement in low stiffness coal, tend to show much lower stiffness when tested in a relatively soft rock. Also one of the new threaded GRP bar designs produces much better performance than the other when tested using the LSEP, whereas they give similar performance under the double embedment test. This laboratory test result appears to be in agreement with the results of underground pull tests, though a greater database of these is still required.
An important recommendation coming from the Project is that, although there have been considerable improvements in GRP bar technology in recent years, GRP bar is still very weak in bending and torsion, and so steel bar rather than GRP bar should be used to reinforce mine ribsides wherever this is a practical proposition.
Handbook of Mine Support Systems
As part of the deliverables for the HSE/ECSC funded Project on "Safe application of mine roadway support systems" a handbook of mine support systems has been written by RMT. This will be submitted to HSE and the ECSC in 2003. It has been produced through collaboration between RMT and UK Coal in the UK, DSK and DMT of Germany and HBCM and INERIS of France The handbook provides a simple description and photograph of each of the main support systems currently used in Western European coal mines plus a number of other products which have been tested and considered for use. The handbook concentrates on recent innovations, in particular rock reinforcement systems, and excludes steel arches. However it does include a number of packing and cribbing systems in use across Europe.
Each generic system is given one or two pages, which include notes on the "Application History" of the system, "Laboratory Test Results" and "Support System Design Notes". The main aim of the handbook is to provide useful information to mining engineers who are often faced with a wide variety of competing products and confusing claims and counterclaims from manufacturers. It is hoped that it will aid them in making rational decisions about appropriate support systems for different geotechnical environments and operational requirements. However an unintended outcome is that the handbook also illustrates the considerable differences in testing and design philosophies for support systems between the different European countries.
Rock Reinforcement and Crib Testing
The work on laboratory testing of steel and GRP reinforcement has been described in the context of mine sides in section 2 above. RMT's work has also included continued work on rockbolt testing for mine roofs for the revision of Part 1 of BS7861, which is now very close to completion.
Recently, attention has turned to the key issues concerning revision of Part 2 of the Standard, which deals with cable bolting, and a document summarising these has been drawn up and submitted to HSE. It is suggested that a revised standard will need to be broadened to include the many alternative long tendon reinforcement systems that have been introduced into the UK since 1996. This could be achieved to some extent by broadening the definition of a cable bolt but retaining system performance criteria based on the double birdcaged cable bolt. However a complete new section of the Standard is likely to be required for flexible bolts and other cable systems designed for installation through resin capsules. Also it will be necessary to consider how pre-tensionable systems should be treated. Currently it appears that a suitable approach would be to apply the same criteria and tests as for untensioned systems but to add additional safeguards regarding the tensioning equipment and their safety when tensioned. It will also be necessary to consider the safety of these systems when they have been tensioned but prior to full column grouting, a period of indeterminate length.
During the year a number of long tendon systems have been tested in the laboratory including new designs of 7 and 19 strand flexible bolts and a tensionable flexible bolt system. An important aspect of these tests has been to begin to determine a suitable encapsulation length for a Laboratory Short Encapsulation Pull Test for measuring flexible bolt system performance. Current indications are that an encapsulation length of 250mm or more will be necessary to ensure that the loads generated in the systems approach sufficiently close to their tensile strength. This compares with a length of 160mm that has been adopted for rockbolts.
Further to the work reported last year on testing South African rockbolting consumables, further work was undertaken on the behaviour of South African rockbolting resins. This indicated that some resins could be badly damaged when bolts were spun for as little as 5 seconds even though the nominal gel times were much more than this. A similar series of tests was undertaken with UK AT rockbolting consumables to determine whether similar problems might arise. It was reassuring to find that the AT consumables behaved in a much more predictable way in line with the manufacturers' stated expectations.
Work began on assessing a computer program (STOP) produced in the USA by NIOSH for predicting the behaviour of various cribbing systems. This program includes models of load displacement behaviour for conventional timber cribs and some of the new "engineered" cribs, such as Link-n-Lock, which is now used extensively in the UK. It also allows comparison of the costs of employing different cribbing systems to produce the same overall load displacement characteristics. The program allows input of basic timber compressive strength and hardness properties and crib geometry or allows use of its own database of typical American timber properties. Initial testing and comparison of predictions for conventional cribs used at Boulby Potash mine showed good agreement for 4-point cribs but less good agreement for nine point cribs. Comparison of Link-n-Lock predictions with previous test results showed very good agreement. It is now intended to begin an extensive testing programme of typical UK cribbing systems. It is probable that the tests will be undertaken by NIOSH in the USA as they have agreed to do this work at a low cost and suitable test facilities in the UK are no longer available.
Development And Demonstration Of Remote Reading Telltale System
The demonstration of the first full installation of a remote reading telltale system developed by RMT has now been completed successfully in a German coal mine with retreat of the longwall panel. The exercise allowed several improvements to the system to be made; in particular an extensive diagnostics manual has been written which should further reduce cost of ownership. The system has now been recovered from the mine and it is expected that it will be installed in another German experimental rockbolted roadway later this year. As well as operating as an on-line safety monitoring system as intended, the very accurate and detailed data recorded by the system has shown some interesting aspects of strata behaviour in the face end area which are worthy of further study.
The measurement principles employed by the system have also been incorporated into several new transducers during the year. These include a two-point "blast proof" extensometer, which has been successfully tested at the face of a UK tunnel in granite subject to blasting and is now being deployed extensively in platinum mines in South Africa. A prototype multipoint extensometer has also been built which currently awaits field trials. This could be a viable replacement for the sonic extensometer.
In-Situ Integrity Testing Of Rock Reinforcement
Application of the RF system for determining in-situ bolt and cable length in the previously flooded Ollerton 42s area of Thoresby colliery has provided a very useful insight into some of its operational constraints. RMT found that the system worked very well in dry sandstone and siltstone and for newly installed rockbolts in damp rock. The parallel method also worked well between a new bolt and an old (8 years) bolt in damp rock, but it did not work between two old bolts in damp rock. Extensive testing appears to indicate that the resin around newly installed bolts acts as an insulator but this insulation seems to breakdown with time in damp rock, stopping the system working. This is likely to be a basic constraint on the system.
Notwithstanding the above, a new meter has been built with a lower minimum operating frequency and the system has been shown to work well this year in two stone mines and in dry salt at Boulby Potash mine.
Novel Methods For Detecting Broken Rock
In last year's report six physical principles were listed which were being evaluated by RMT for application to the detection of broken rock in mines, particularly loose roof. Of these, work during 2002 concentrated on two principles which showed most promise, these were Thermal and Acoustic response.
Whilst encouraging results were obtained from laboratory experiments and numerical modelling of thermal response, limited success was achieved in field experiments at two mines using an unstimulated thermal response technique. This was mainly attributed to the thermal gradient being too small between the rock and the ventilation air at the sites. A stimulated thermal response experiment is expected to be more successful but prior to setting up such an experiment further laboratory testing will be undertaken.
The best results were achieved using the Acoustic response principle. This was achieved by use of a digital laser vibrometer to remotely measure the vibration response of rock when hit with a physical blow. A similar waveform to that output from the geophone of the Acoustic Energy Meter (AEM) is obtained from the vibrometer. Therefore it is anticipated that similar signal processing to that used in the AEM could provide a suitable measure of rock looseness. These experiments have currently only been undertaken in the laboratory and a detailed underground trial is planned for 2003. This will include recording of the vibrometer output signals such that they can be replayed through various signal conditioning systems to allow their development and improvement. It is still necessary to determine the best means of remote excitation of the rock. Currently a paint ball gun is being utilised with success but alternative means are actively being sought.
Work around the world on assessing the suitability of the Acoustic Energy Meter for measuring the "looseness" of rock has continued with encouraging outcomes in Australian coal mines and in South African gold and platinum mines. Australian IS certification of the system was approved at the end of 2002 and this will allow application of the system in hazardous areas.
Numerical Modelling For Mine Design
During the year, further three-dimensional modelling of rock reinforcement spacing has been successfully undertaken by RMT and it is now believed that this can be applied practically to real design problems. This allows, for example, the behaviour of an 11-bolt pattern at one-metre centres to be compared with a 6 + 5 pattern at 0.5m centres. Until now the conventional two-dimensional modelling has shown no distinction between these two patterns though it is known that their behaviour can be very different in practice.
Modelling has also been successfully applied to simulation of roof failure mechanisms in relatively low stress environments where tensile failure can be induced by a combination of geological slips and low horizontal stresses. Modelling was also used during the year to demonstrate that the roof fall that occurred in the Stanley Main seam at Riccall was likely to be almost wholly attributable to the known change in roof lithology rather than any significant change in the stress environment.
In 2003 it is anticipated that modelling developments will be concentrated on simulation of pre-tensioned cable bolts systems such that their influence on mine roofs can be better predicted.
Improved Rock Stress Measurement Techniques
RMT work on this new Project is currently concentrating on development of stress determination through detection of the Kaiser Effect in acoustic emissions from laboratory compression testing of oriented rock samples. Work to date has shown that this is a viable method of measuring stress when there is already a known indication of principle stress directions. It has also shown that it is not suitable in its current stage of development for measuring stresses below 5MPa and that it appears to be better suited to measurements in rocks with a high quartz content. Work is planned during 2003 to begin to try to achieve successful measurements when there is no known direction of the principle stresses. Work has also begun on undertaking Kaiser effect measurements on core obtained by INERIS from overcoring stress measurements in French coal mines. A draft Suggested Method has recently been published and RMT in collaboration with Cambourne School of Mines (Exeter University) are examining differences between our current techniques and the Suggested Method.
UoN work on this project includes developing Rock Mass Constitutive Models. A testing programme is being developed to investigate the pre and post failure properties of coal measure rock. Particular emphasis is being placed on changes in the stiffness of the strata at various confinements. In general laboratory test values obtained on samples of rock strata need to be reduced to represent in-situ values to account for structural features that are present in-situ. Previous research has been undertaken at the University in the development of a method of characterising in-situ coal measure strata such that their properties can be predicted. The research has led to the development of a dedicated rock mass classification system developed for characterising coal measure strata. Further development and validation of the classification system will be undertaken to assess the optimal relationships for the derivation of constitutive model parameters.
A literature review is being undertaken to obtain published information on the characteristic deformation behaviour of broken caved waste material with a view to developing a Waste Constitutive Model. A testing programme on caved waste material will be developed.
Case study information for a UK coal mine has recently been obtained. Large-scale numerical models that include properties for strata, caved waste material and support elements are being constructed. Once the initial modelling is completed fine tuned optimum constitutive relations will be assessed for simulation of rock strata residual strength caved waste behaviour using newly developed sub-routines.
'Mixed' Support Systems
This UoN project aims to investigate the use of 'mixed' support systems and to provide guidelines on their specification. The term 'mixed support' typically refers to the use of free-standing steel supports, such as steel arches, in conjunction with rock bolts. A comprehensive literature search has been undertaken to allow a fundamental basis for the development of numerical models for mixed support systems.
Modelling has been undertaken using the FLAC 2D-geomechanical finite difference code. The group has recently purchased the FLACD 3D code and it is envisaged that in the future 3D modelling will supplement the 2D models. The initial modelling has been undertaken to develop a structured methodology of simulating mixed support systems, which can then be used to allow the interactions between the different support elements to be appraised.
The three basic components of the mixed support systems considered are:
Development of a methodology of evaluating the interactions between the different support components within the roadway support system and how the relative contributions vary with time was being investigated during the year.
A method that has been developed allows the recording, at each time step of the model, the total of the load transfer of the loads from the rock strata onto the bolting system and onto the free-standing support. This allows appraisal of how the loading of the different support components occurs as a function of time and how the different support system components interact.
A numerical modelling of a case study for a mixed support system used within coal mine gateroads has been undertaken. Although the case study was for a non-European coal mine, it was considered that it would be a worthwhile exercise as detailed instrumentation and monitoring results were available for gateroads in two adjacent panels, one with mixed support systems consisting of free standing supports and bolts, the other with free standing supports only. Monitoring of strain gauges and load cells applied to the steel supports was undertaken as the face retreated towards the monitoring stations. The mining method was similar to the method currently adopted within the United Kingdom consisting of retreat longwall mining using single entry rectangular gateroads with a rib pillar between each panel. The mining however was at a depth of approximately 400 meters, which is typically half the depth of an UK coal mine.
Methodologies are being investigated to assess the optimum manner of using the FLAC 3D code to obtain outputs that can be used to investigate the interactions between the rock strata, rock bolting and free standing steel supports. Detailed appraisal of the nature of the steel support-rock interface properties and the degree of packing of the steel supports are also being undertaken.
Pre-stressing Reinforcing Tendons
The technique of applying pre-stress to reinforcing tendons has been useful and effective in many mining and civil environments. The generation of confinement through pre-stressing reinforcement is claimed to improve the stiffness of the rock mass through closure of fractures and microfractures. This in turn will reduce displacements or closure for a given loading. The techniques application to high stress mining environments however is much more limited partially due to the practicalities of applying pre-stress to the support types used.
The main objective of UoN's contribution is to research the use of pre-tensioned bolts and cables in a coal mining environment and to investigate the beneficial effects of the use of pre-stressing supports on the immediate strata surrounding a coal mine excavation.
The three-year project began in mid-2002 with a literature review in relation to pre-stressed tendon applications, component technology and the theoretical background. Thereafter the programme will include:
The influence of naturally occurring fluids on excavation stability and support system integrity
This UK Coal project is due for completion in September 2003. Early results indicate medium to long term deterioration in support effectiveness can be expected when either crude oil or strata water are present in any part of the near seam geology. The presence of strata water and its likely effects depends on flow rates and the presence of discontinuities. However, crude oils affect support conditions purely by their presence in the surrounding rock mass.
Assessment of non-destructive testing for long term stability of rock bolted roadway
A UK Coal supported project to assess the AMEC GRANIT project was completed at Thoresby Colliery in July 2002. The results did indicate a degree of accuracy. However, the process does rely on a highly developed computing system working with neural networks. This also introduced a major delay in data analysis, which effectively gave no benefits over the existing monitoring systems in operation. The broken bolt detector developed by Rock Mechanics Technology was used in parallel with the GRANIT system. The results indicated that all the bolts tested were intact.