UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Geotechnical studies of retreat pillar coal mining at shallow depth Cullen, Michael 2002

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2002-731464.pdf [ 13.02MB ]
Metadata
JSON: 831-1.0081103.json
JSON-LD: 831-1.0081103-ld.json
RDF/XML (Pretty): 831-1.0081103-rdf.xml
RDF/JSON: 831-1.0081103-rdf.json
Turtle: 831-1.0081103-turtle.txt
N-Triples: 831-1.0081103-rdf-ntriples.txt
Original Record: 831-1.0081103-source.json
Full Text
831-1.0081103-fulltext.txt
Citation
831-1.0081103.ris

Full Text

GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH By Michael Cullen B.Eng., McGill University, 1986 M.Eng., McGill University, 1988 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY In THE FACULTY OF GRADUATE STUDIES (Department of Mining and Mineral Process Engineering) We accept this thejsi^ as confornjiflg to thejequired standard THE UNIVERSJTY/OF BRITISH COLUMBIA March 2002 © Michael Cullen 2002 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of A*^ p M,toft£>i Pfc^ v &i£ g-u^/o WUL\ o ^_ The University of British Columbia Vancouver, Canada Date GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH ABSTRACT This thesis presents the results of research into geotechnical aspects of retreat pillar coal mining at shallow depth (less than 100m). The fieldwork component was carried out over a four-year period at the Quinsam Coal Mine. The geotechnical aspects investigated were excavation stability, excavation support, pillar design, gob cave prediction, and subsidence. These were all considered critical aspects of safe and cost effective mining. Geotechnical design at the Quinsam Coal Mine was initially carried out using existing design tools. For the most part these tools were developed at moderate to deep depth mines (greater than 100m). The suitability of these tools at shallow depth mines was previously not known; as such the primary objectives of this research were: 1) to determine if the existing geotechnical design tools were applicable at shallow depth, 2) to develop new design methods where the existing tools were found to not be suitable, 3) to improve safety, productivity and costs at shallow depth retreat pillar coal mines. Suitable existing tools where found for design of ground support, pillar size, excavation size, and gob cave prediction. Existing tools where not found to be suitable for prediction of caving and surface subsidence. The cave height at shallow depth was determined to extend much higher than predicted by existing methods. The extent of the ground surface affected by subsidence was determined to be much less than that predicted by existing methods. New tools for predicting caving and subsidence have been developed based on measurements at the Quinsam Coal Mine supplemented with numerical modelling. The distinct response of the rock mass to mining at shallow depth is attributed to the magnitude and orientation of the induced stresses. At shallow depth low compressive and tensile stresses are more likely to occur, and more likely to extend for a greater distance into the rock mass above the excavation. Although this work is based on studies carried out at the Quinsam Coal Mine it is believed that the findings can be applied to other shallow depth coal mines in similar geological environments. Since adopting many of the recommendations and design procedures developed through this work, the Quinsam Coal Mine has achieved significant improvements in safety, productivity, and costs. i i TABLE OF CONTENTS ABSTRACT ii LIST OF TABLES vi LIST OF FIGURES vii LIST OF PHOTOS ix ACKNOWLEDGMENTS x CHAPTER 1 INTRODUCTION 1 CHAPTER 2 THE QUINSAM COAL MINE 8 2.0 INTRODUCTION 8 2.1 G E N E R A L G E O L O G Y OF THE Q U I N S A M C O A L A R E A 8 2.2 MINING A T THE Q U I N S A M C O A L MINE 11 CHAPTER 3 SITE CHARACTERIZATION 21 3.0 INTRODUCTION 21 3.1 GEOLOGIC DISCONTINUITY D A T A 21 3.2 R O C K M E C H A N I C A L PROPERTIES 24 3.3 R O C K M A S S CLASSIFICATION 27 3.4 IN SITU STRESS 28 3.5 IMPLICATIONS FOR MINING 31 3.5.1 Geologic Structure 31 3.5.2 Mechanical Properties and Rock Mass Classification 33 3.5.3 Mining Induced Stresses 33 3.6 CONCLUSIONS 40 CHAPTER 4 EXCAVATION STABILITY 43 4.0 INTRODUCTION 43 4.1 L I T E R A T U R E R E V I E W 43 4.2 E X C A V A T I O N C L O S U R E MONITORING P R O G R A M 46 4.2.1 Mechanical Borehole Extensometers 46 4.2.2 Rod Extensometer 48 4.2.3 Magnetic Multipoint Borehole Extensometer 50 4.3 E X C A V A T I O N STABILITY VERSUS R O C K M A S S CLASSIFICATION 53 4.4 CONCLUSIONS 55 CHAPTER 5 GROUND SUPPORT 58 5.0 INTRODUCTION 58 5.1 R O C K BOLTS 59 5.1.1 Visual Observations of Performance 61 5.1.2 Rock Bolt Pull Tests 61 5.1.3 Dri l l Hole Size 65 5.1.4 Rock Bolt Load 67 5.1.5 Full Column Resin Rock Bolts 68 5.2 EXISTING SUPPORT DESIGN METHODS 68 5.2.1 Rock Mass Support Design Methods 69 5.2.1.1 US Corps of Engineers 69 i i i GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 5.2.1.2 Farmer and Shelton ! 70 5.2.1.3 French Matrix System 72 5.2:1.4 "Q" Classification System 72 5.2.1.5 " R M R " Classification System 75 5.2.1.6 C M R S Classification System 76 5.2.2 Analytical Support Design Methods 76 5.2.2.1 Beam Building and Arching 76 5.2.2.2 Dead Weight Load 80 5.3 C O M P A R I S O N OF THE EXISTING SUPPORT DESIGN METHODS TO EXPERIENCE A T T H E Q U I N S A M C O A L MINE 84 5.4 CONCLUSIONS 85 CHAPTER 6 PILLAR DESIGN 88 6.0 INTRODUCTION 88 6.1 L I T E R A T U R E REVIEW 89 6.1.1 Pillar Stress 89 6.1.2 Pillar Strength 94 6.1.3 Factor of Safety 97 6.1.4 Pillar Failure Modes 98 6.2 P I L L A R P E R F O R M A N C E A T THE Q U I N S A M C O A L MINE 99 6.2.1 Coal Strength 99 6.2.2 Visual Pillar Performance 100 6.3 P I L L A R DESIGN M E T H O D 105 6.3.1 Method Verification 106 6.4 R E C O M M E N D A T I O N S FOR P I L L A R DESIGN A T S H A L L O W D E P T H R E T R E A T P I L L A R C O A L MINES 108 6.4.1 Barrier Pillars '. 108 6.4.2 Roadway or Mains Pillars 109 6.4.3 Panel Pillars 109 6.4.4 Remnant Pillars 110 6.5 CONCLUSIONS 110 CHAPTER 7 GOB C A V E PREDICTION 113 7.0 INTRODUCTION 113 7.1 L I T E R A T U R E R E V I E W 113 7.2 GOB C A V E STUDIES A T THE Q U I N S A M C O A L MINE 115 7.2 GOB C A V E M O D E L 121 7.4 CONCLUSIONS 124 CHAPTER 8 SUBSIDENCE 126 8.0 INTRODUCTION 126 8.1 SUBSIDENCE DEFINITIONS 128 8.2 L I T E R A T U R E R E V I E W 133 8.2.1 Sub-Surface Ground Movements 135 8.2.2 Angle of Draw 137 8.2.2.1 Angle of Draw Prediction 139 8.2.3 Cave and Fractured Zone Height 141 8.2.3.1 Predicting the Caved Zone Height 143 8.3 C A V I N G A N D SUBSIDENCE A T THE Q U I N S A M C O A L M I N E 146 8.4 C O M P A R I S O N OF THE EXISTING SUBSIDENCE PREDICTION METHODS TO E X P E R I E N C E A T THE Q U I N S A M C O A L MINE 155 8.5 N U M E R I C A L M O D E L I N G OF SUBSIDENCE 158 8.6 INVESTIGATION OF STRESSES A R O U N D THE C A V I N G FRONT 171 8.7 CONCLUSIONS 174 iv GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH CHAPTER 9 CONCLUSIONS AND RECOMMENDATIONS 179 9.0 INTRODUCTION 179 9.1 MINING INDUCED STRESSES 180 9.2 E X C A V A T I O N STABILITY 181 9.3 E X C A V A T I O N SUPPORT DESIGN 183 9.4 G R O U N D SUPPORT WITH R O C K BOLTS 184 9.5 P I L L A R DESIGN 185 9.6 GOB C A V E PREDICTION 186 9.7 C A V I N G A N D SUBSIDENCE 187 9.8 CONTRIBUTIONS TO THE A D V A N C E M E N T OF K N O W L E D G E 189 9.9 R E C O M M E N D A T I O N S FOR FURTHER W O R K 192 REFERENCES 195 APPENDIX 1 QUINSAM C O A L MINE SITE CHARACTERIZATION 207 F A U L T A N D JOINT M E A S U R E M E N T S F R O M THE 2N MINE 208 U N I A X I A L COMPRESSIVE STRENGTH TESTS 215 POINT L O A D I N D E X TESTING 238 DIRECT S H E A R S T R E N G T H TESTS 242 S L A K E D U R A B I L I T Y TESTS 245 MOISTURE C O N T E N T TESTS 246 DENSITY TESTS 246 Q R O C K M A S S CLASSIFICATION P A R A M E T E R S 247 R M R R O C K M A S S CLASSIFICATION P A R A M E T E R S 247 C M R R R O C K M A S S CLASSIFICATION P A R A M E T E R S 247 APPENDIX 2 ROOF CONVERGENCE MEASUREMENT DATA 248 ROD E X T E N S O M E T E R D A T A 249 APPENDIX 3 R O C K BOLT PULL TEST DATA 250 RESULTS OF P U L L TESTS O N ROOFBOLTS 251 APPENDIX 4 PILLAR CLASSIFICATION RECORDS AND FACTOR OF SAFETY VALUES 253 P I L L A R CLASSIFICATION RECORDS 254 APPENDIX 5 SURFACE SUBSIDENCE 260 S U R F A C E SUBSIDENCE STATION L O C A T I O N , INITIAL COORDINATES, A N D F I N A L E L E V A T I O N S 261 S U R F A C E SUBSIDENCE L E V E L S U R V E Y D A T A : STATION N U M B E R , D A T E OF S U R V E Y , A N D E L E V A T I O N S 265 S U R F A C E SUBSIDENCE: S U M M A R Y OF N U M E R I C A L M O D E L I N G RESULTS \ 271 v GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH LIST OF TABLES Table 3.1: Mechanical Properties of Rocks 26 Table 3.2: Point Load Strength and Estimate of UCS 26 Table 3.3: Typical Rock Mass Classification Values 27 Table 3.4: Estimated In Situ Stress For k = 1.0 29 Table 3.5: Mining Induced Stresses Determined from Computer Simulations 37 Table 4.1: Roof Convergence Data from the Literature 45 Table 5.1: Average Finished Drill Hole Diameter 66 Table 5.2: Pull Tests Results For Different Dril l Hole Size and Resin Cartridge Size 66 Table 5.3: Support Design Recommendations of the US Corps of Engineers 70 Table 5.4: Support Design Recommendations of Farmer and Shelton 71 Table 5.5: Support Design Recommendations of French Matrix System 73 Table 5.6: Support Design for 6m Wide Excavation in Competent Siltstone 85 Table 6.1: Empirical Values Used in Pillar Strength Formulas 95 Table 7.1: Gob Stability A N D Convergence Data from the Literature 115 Table 7.2: Roof Convergence at Time of Gob Cave 117 Table 7.3: Gob Cave Data, Quinsam Coal Mine 119 Table 8.1: Subsidence Data From the Quinsam Coal Mine 150 Table 8.2: Surface Fractures Over 2S Mine Sections 101 153 Table 8.3: Prediction of the Height of the Caved Zone and the Fractured Zone 155 Table 8.4: Prediction of the Angle of Draw 156 Table 8.5: Material Properties Used in Numerical ModelS 161 Table 8.6: Summary of Numerical Model Results 164 Table 8.7: Induced Stresses Above the Caved Zone Determined from Computer Simulations 171 vi GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH LIST OF FIGURES Figure 2.1: General Location Map For The Quinsam Coal Mine 9 Figure 2.2: Typical Section At The Quinsam Coal Mine 10 Figure 2.3: Regional Geology of the Middle Quinsam Coal Basin 12 Figure 2.4: Typical Mine Layout for Room and Pillar Mining 13 Figure 2.5: Process of Pillar Extraction "Depillaring" 18 Figure 2.6: Map of 2N Mine at Quinsam Coal, March 1996 20 Figure 3.1: Stereonet Projection Showing Contours of Pole Concentrations and Four Major Joint/Fault Orientations 23 Figure 3.2: Histogram of Fault Displacements 23 Figure 3.3: Stereonet Projections Showing Poles of Fault Orientations As Well as Fault Displacement 24 Figure 3.4: Horizontal Stress Measurements from Coal Mines in the Western U.S., 29 Figure 3.5: Models of Stress Induced Failure at the Quinsam Coal Mine 30 Figure 3.6: Potential Wedges Formed in N-S and E-W Trending Roadways 32 Figure 3.7: Orientation and Concentration of Induced Stresses around a 10m Excavation 36 Figure 3.8: Extent of Stress Induced Failure around an Excavation at 50m and 400m Depth 39 Figure 3.9: Idealised Rock Mass Response to Stress 41 Figure 4.1: Example of Monitoring Roof Convergence to Predict Instability 44 Figure 4.2: Mechanical Extensometer 47 Figure 4.3: Mechanical Extensometer Installed in Potentially Unstable Wedge 48 Figure 4.4: Telescoping Rod Extensometer 49 Figure 4.5: Data from the Rod Extensometer 49 Figure 4.6: Sonic Probe M P B X 51 Figure 4.7: Results From M P B X Installed in 2N Mine, #2 Mains 52 Figure 4.8: Deformational Model of Bed Separation 52 Figure 4.9: Relationship Between Stability, C M R R , and Depth (Stress) for Unsupported Excavations Greater Than 6m Long 54 Figure 4.10: Relationship Between Stability, C M R R and Width for Supported Roadways 54 Figure 4.11: Relationship Between Stability, C M R R and Width for Supported Roadway Intersections 55 Figure 5.1: Load Versus Displacement 64 Figure 5.2: Rock/Resin Bond Stress Versus Displacement 64 Figure 5.3: Load History on Rock Bolt Installed In Active Retreat Pillar Area 67 Figure 5.4: Support Design Guidelines Using the "Q" Classification 74 Figure 5.5: Problem Geometry for Voussoir Arch Stability Analysis 78 Figure 5.6: Formation OF a Reinforced Rock Arch 79 Figure 5.7: Anchoring Weak Strata to Strong Strata 81 Figure 5.8: Example of a Wedge Defined by Three Joints above a Mine Excavation 83 Figure 6.1: Mining Induced Stress Distribution 89 Figure 6.2: Redistribution of Stresses to Pillars 91 Figure 6.3: Abutment Load on Pillars at the Edge of a Gob 93 Figure 6.4: Distribution of Abutment Stresses at the Edge of the Gob 94 Figure 6.5: Typical Modes of Pillar Failure 99 Figure 6.6: Example of Pillar Classification Map 103 Figure 6.7: Visual Pillar Classification versus Distance from the Gob 105 Figure 6.8: Different Loading Conditions Considered in A R M P S 106 Figure 6.9: Pillar Factor of Safety versus Visual Classification 107 vii GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 7.1: Electronic Single Point Borehole Extensometer 116 Figure 7.2: Roof Convergence in Section 101 2N Mine 117 Figure 7.3: Example of Detailed Structural Mapping and Controlled Pillar Extraction 121 Figure 7.4: Proposed Gob Failure Mechanisms 122 Figure 8.1: Definition of Subsidence Angles 130 Figure 8.2: Zones of Ground Movement 130 Figure 8.3: Schematic Diagram of Discontinuous Surface Subsidence Profiles 131 Figure 8.4: Common Shapes of Caving 144 Figure 8.5: Development of Chimney Type Cave in a Four-Way Intersection 145 Figure 8.6: Location of Subsidence Monitoring Stations Over 2N Mine 148 Figure 8.7: Location of Subsidence Monitoring Stations Over 2S Mine 149 Figure 8.8: Map of Surface Fractures Over Panel 101, 2N Mine, August 9, 1995 152 Figure 8.9: Angle of Draw versus Depth of Cover 157 Figure 8.10: Subsidence Factor versus Depth of Cover 159 Figure 8.11: Flow Chart of F L A C Modelling Method 161 Figure 8.12: Example of Graphical Output from F L A C 165 Figure 8.13: Example of Graphical Output from F L A C 165 Figure 8.14: Example of Graphical Output from F L A C 166 Figure 8.15: Angle of Draw Versus Depth of Cover 168 Figure 8.16: Angle of Fracture Versus Depth of Cover 168 Figure 8.17: Contours of Tensile Stress, and Principal Stress Orientation and Magnitude around a Rectangular Shaped Cave 175 Figure 8.18: Contours of Tensile Stress, and Principal Stress Orientation and Magnitude around a Wedge Shaped Cave 175 Figure 8.19: Concept of Yielded and non yielded Zones 180 vii i GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH LIST OF PHOTOS Photo 2.1: Continuous Miner Cutting a 6m Wide Roadway 14 Photo 2.2: Roof Support Being Installed By a "Fletcher Roof Bolter" 14 Photo 2.3: Stump Pillars within a Gob 17 Photo 2.4: Initial Gob Cave 17 Photo 2.5: Sinkhole Subsidence 19 Photo 5.1: Ground Support Elements in General Use at The Quinsam Coal Mine 60 Photo 6.1: Pillar Class 0 101 Photo 6.2: Pillar Class 2 101 Photo 6.3: Pillar Class 3 102 Photo 6.4: Pillar Class 4 102 Photo 6.5: Pillar Class 5 103 Photo 7.1: Open Gob with Crushed Wooden Post in Background 114 Photo 7.2: Gob Immediately after the Cave 114 Photo 8.1: Open Surface Fracture at the Quinsam Coal Mine 127 Photo 8.2: Stepped Fracture at the Quinsam Coal Mine - Partially Obscured by Vegetation 127 Photo 8.3: Open Fracture At the Quinsam Coal Mine 154 ix GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH ACKNOWLEDGMENTS The fieldwork for this thesis was carried out at the Quinsam Coal Mine that is owned and operated by Hillsborough Resources Ltd. I would like to thank the management and staff at the mine for their technical assistance, financial support and opportunity to carry out this research. Special thanks are extended to Mr. M . Hoffman P.Eng. and Mr. K. Galovich P.Eng. Financial and technical assistance for this research was also provided by CANMET under DSS Contract 23440-3-9105/01-SQ. Special thanks go to Mr. D. Forrester, P.Eng. ITASCA Consulting Group Inc. kindly provided technical assistance and comments on the numerical modelling of caving and subsidence. Special thanks go to Mr. M . Board, P.Eng. Dr. Rimas Pakalnis of the University of British Columbia acted as my Research Supervisor. My sincere thanks go to Dr. Pakalnis for his continued support and encouragement of this work. GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH CHAPTER 1 INTRODUCTION This thesis presents the results of research into geotechnical aspects of retreat pillar coal mining at shallow depth. For this research, shallow depth is defined as being less than 100m; however, in the case of subsidence prediction it is defined as being less than 30 times the height of extraction. A value of 30 times the height of extraction was determined to be the upper limit for discontinuous surface subsidence to occur at the Quinsam Mine. Retreat pillar mining is considered by many to be the most dangerous method of coal mining. In the United States room and pillar mining accounts for approximately half of the coal production but has twice the ground fall incident rate to that of longwall mining (Papas et al 2000). Pillar retreat accounts for only 10% of the U.S. coal production yet 25% of all ground fall fatalities (Mark et al 1997). The fieldwork component of this study took place between 1993 and 1997 at the Quinsam Coal Mine, located near Campbell River on Vancouver Island. Underground mining at the Quinsam Coal Mine was carried out by the retreat pillar extraction method, at depths of cover ranging between 30m and 100m. These depths are relatively shallow when compared to most coal mines operating throughout the world. The Quinsam Coal Mine Management recognised that geotechnical aspects of mine design would play a significant role in achieving the desired underground production rates in a safe and cost effective manner. Underground mining commenced at the Quinsam Mine in 1988. Geotechnical design was initially carried out using established industry practices. However, most of the established design methods were developed at mines operating at depths greater than 100m. The suitability of the existing design tools to shallow depth coal mines (less than 100m) was previously unknown. The purpose of this research was as follows: 1. To verify the suitability of existing geotechnical design procedures to retreat pillar mining at shallow depths. 1 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 2. Where necessary, improve upon the existing geotechnical design procedures for retreat pillar mining at depths less than 100m. 3.- Improve safety and productivity at retreat pillar mines operating at depths less than 100m. 4. Investigate the causes of the differences between mining at shallow and deeper depths. Five areas of geotechnical design were investigated as part of this research, namely: 1. Roadway excavation stability. 2. Roadway support. 3. Pillar design. 4. Gob stability 5. Subsidence and caving. The Quinsam Mine Management considered these aspects to be geotechnical components to safe and cost effective retreat pillar coal mining. The methodology used for this research involved a step by step literature search, field and laboratory measurements, data analysis, computer simulation, and model development. The literature review was carried out to determine the state-of-the - art design methods for the 5 areas of research noted above. This was followed by the field and laboratory measurements. The fieldwork consisted of instrument measurements and detailed observational measurements of the rock mass response to mining. The laboratory measurements consisted of determination of the mechanical properties of the rock. The collected data was subsequently analysed and numerical modelling of stresses and rock mass response to mining were carried out. The analysis attempted to analysis for differences attributed to stress, structure, and rock mass. The step by step investigation is considered the most appropriate method for these types of research were several different activities are being investigated. The methodology is considered applicable to research carried out at hardrock mines. Site visits were made to three coal mines and two research facilities. The mines visited were Phalen Mine (Nova Scotia), Spruce Creek Mine (West Virginia), and Cumberland Mine (Pennsylvania). The research facilities visited were the Cape Breton Coal Research Laboratory (Nova Scotia), and the United States Bureau of Mine Research Facility 2 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH (Pennsylvania). The purpose of the site visits was to gain first hand experience of operating conditions and practices at other coal mines, and to determine the state of the art in geotechnical aspects of mine design. This thesis is divided into 9 Chapters. Chapter 2 presents general details on the Quinsam Coal Mine including regional geology, local geology, and details of the retreat pillar mining method. The first phase of the research was site characterisation. This work included collecting data on the geotechnical properties of the intact rock, rock mass, and geologic discontinuities at the Quinsam Coal Mine. An evaluation of the in situ stresses and mining induced stresses was carried out using visual observations and two dimensional numerical modelling. The purpose of this phase of the work was as follows: 1. Obtain geotechnical and geological information to facilitate meaningful comparison to other mines. 2. Establish the values for geotechnical properties that were considered to affect rock mass response to mining. 3. Investigate the stress regime at depths less than 100m and compare it to depths greater than 100m. The geotechnical properties of the intact rock, rock mass, and discontinuities at the Quinsam Coal Mine were found to be similar to those of other mines operating in similar geologic environments. The magnitude and orientation of the mining induced stresses at shallow depth mines were determined to be different to those at deeper mines. The detailed results of this phase of the research are presented in Chapter 3. The second phase of the study involved measurements and observations of the rock mass response to mining, and comparison of the observed response to the predicted response using existing design methods for each of the 5 aspects listed above. This work is presented in Chapters 4 to 8. Each of these chapters includes a literature review that presents background 3 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH information and the state-of-the-art. Analysis and discussions of the rock mass response at shallow depth is also included. For the most part roadway excavation design at the Quinsam Coal Mine was dictated by the size of the mining equipment. The excavations at the mine were typically 6m wide and 2.2m high. The 6m wide excavations were found to be stable in most circumstances. However, unstable excavations did occasionally occur. It was found that roadways that were stable when first excavated generally remained stable, while unstable roadways were usually unstable right from the start. Additional support was installed to control unstable areas; however, there was no means of verifying the effectiveness of the additional support. The purpose of the excavation stability study was to provide scientific verification of the observed stability conditions as well as to determine a means for distinguishing stable from unstable mine excavations at shallow depth. Excavation convergence, which is a well established indicator of stability, was used to complete this portion of the study. Ground support at the Quinsam Coal Mine was primarily achieved with rock bolts. Rock bolts represented a large portion of the mining costs as well as being a key element in ensuring the safety of the underground excavations. Over-support results in unnecessary expenses while under-support may result in potential safety hazards. The purpose of the ground support study was to first evaluate the effectiveness of the rock bolt support being used, and secondly to determine appropriate support design methods for use at shallow depth retreat pillar coal mines. Pillar support is an essential element in ensuring the safety of underground excavations. Over-designed pillars results in unnecessary sterilisation of resources while under-designed pillars may result in ground falls that lead to injuries and lost production. Where pillars are being used to prevent caving and/or subsidence, under-designed pillars may result in damage to surface resources or significant water ingress into the mine. The purpose of the pillar support study was to first evaluate the effectiveness of the pillar designs being used at the Quinsam Coal Mine, and secondly to determine appropriate pillar design methods for use at shallow depth retreat pillar coal mines. 4 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Gob caves are an integral part of the retreat pillar mining method; as pillars are mined out the roof collapses. At the Quinsam Coal Mine unpredicted caves have occurred that have caused damage to mining equipment, lost production, and injury. The first phase of the investigation was to study gob cave behaviour at the Quinsam Coal Mine. The second phase was to establish a more accurate method of gob cave prediction. The third phase was to determine methods to reduce the hazard of unpredicted gob caves at shallow depth retreat pillar coal mines. Surface subsidence is an unavoidable consequence of the retreat pillar mining method. Prediction of subsidence is critical for protection of surface resources, as well as for protection of the underground environment. Surface ground movements may result in damage to surface facilities such as buildings, roads, and power lines. At shallow depths fractures may extend to surface causing short-circuiting of ventilation or water ingress into the mine. Subsidence protection is usually achieved with proper excavation span and pillar design. The first phase of the study was to investigate subsidence at the Quinsam Coal Mine. The second phase was to determine appropriate caving and subsidence design methods for use at shallow depth retreat pillar coal mines. Chapter 9 presents the conclusions and recommendations from this study along with recommendations for further work. A discussion of the scientific significance and contribution of this work is also presented in this chapter. Numerical simulations showed that there is a significant difference between the induced stresses at shallow and deep mines. These differences result in significantly different rock mass response to mining. Since most of the existing geotechnical design tools were developed at mines operating at depths greater than 100m it was unknown if the existing tools would be applicable at depths less than 100m. It was determined that there are existing design tools for excavation stability analysis, support design, pillar design, and gob cave prediction that are applicable to retreat pillar coal mines operating at shallow depths. 5 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Existing tools where not found to be suitable for prediction of caving and surface subsidence. The cave height at shallow depth was determined to extend much higher than predicted by the existing methods. The extent of the ground surface affected by subsidence was determined to be much less than that predicted by the existing methods. New tools for predicting caving and subsidence have been developed based on measurements at the Quinsam Coal Mine coupled with numerical modelling. This study is focussed on work carried out at the Quinsam Coal Mine. However, the properties of the rock and rock mass at the Quinsam Coal Mine were determined to be similar to other mines working in similar geological environments. As such, it is believed that the findings from this research can be applied successfully to other shallow depth mines operating in similar geological environments. Since adopting many of the recommendations and design procedures developed by this work the Quinsam Coal Mine has achieved significant improvements in both safety and production. This research has contributed to the advancement of knowledge in the following ways: 1. Geomechanical properties of the rock mass at the Quinsam Coal Mine have been determined. 2. Models to explain the differences between rock mass response at shallow depth and deep depth have been developed. 3. Existing tools were found that are suitable for support design, excavation stability analysis, pillar design, and gob cave prediction at retreat pillar coal mines operating at depths less than 100m. 4. The critical convergence rates for excavation stability and gob cave prediction at the Quinsam Coal Mine were determined. The critical convergence rates were found to not be a function of stress or depth at shallow mines. 5. Existing tools for subsidence and cave height prediction are not applicable for shallow depth mines. Tools for predicting cave height and the extent of ground affected by subsidence have been developed. 6 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH The most significant findings of this work relate to the cave height and extent of subsidence over shallow depth mines. The height of caving was determined to be as much as 14 times the excavation height. Al l previous research has indicated that the upper bound for caving is 10 times the excavation height; this value is commonly used for engineering design purposes. A potential 40% increase in cave height over the previously accepted criteria is considered an extremely important finding that may significantly impact upon engineering evaluations of new and abandoned mines. Only the area immediately above shallow depth retreat pillar mine excavations is affected by significant subsidence. The angle of critical deformation, which defines the extent of damaging ground movements, was determined to be less than 0° for excavations at depths less than 30 times the extraction height. Existing subsidence prediction tools suggest that the angle of critical deformation ranges from between 4° and 26.5°. Knowing that critical ground deformations are restricted to the area above the excavation will significantly reduce the amount of resources sterilised in protective pillars. The findings from this research are believed to be applicable to other shallow depth mines; this hypothesis needs to be confirmed by measurements at other mines. The recommendations for future work also include refining the design methods through collection and statistical analysis of additional field data; this should include data from other mines. Data collection should include information on both the hazard and the consequence of failures such that a risk assessment method might be developed. Verification of some of the assumptions made, such as the in situ stress regime, is recommended to validate the differences between induced stresses at shallow and deep mines. Also, there is much scope for improvement to numerical modelling capabilities and techniques for the investigation of caving and subsidence over shallow depth excavations. 7 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH CHAPTER 2 THE QUINSAM COAL MINE 2.0 INTRODUCTION The Quinsam Coal Mine is located approximately 20km south-west of Campbell River, on Vancouver Island, Canada. Figure 2.1 shows a general location map for the Mine. The Mine has been in operation since 1986. Underground operations commenced in 1989. Underground mining is being carried out by the room and pillar method with retreat pillar extraction. Most of the coal is being mined from the No. 1 Seams of the Middle Quinsam Lake Sub-Basin of the Comox Basin Formation. Several underground mines are in operation at the property. Most of the research work for this project was carried out at the 2N Mine. 2.1 GENERAL GEOLOGY OF THE QUINSAM COAL AREA The Comox Basin Formation stretches from Campbell River to Parksville, a distance of approximately 120 km. The basin is seldom wider than 40 km; its average width is 13 km. The sedimentary deposits occur in the lowlands along the paleo coastline. The sediments are members of the Nanaimo Group. The Quinsam Coal Mine is located in a small sub-basin known as the Middle Quinsam Lake area. This sub-basin is separated from the main basin by a ridge of volcanic rock from the Karmutsen Formation. The Middle Quinsam Lake area consists of up to a 200m thickness of the lowermost part of the Comox Formation. Two distinct members are present within this area; a lower member comprising of medium greenish grey siltstone and brownish grey mudstone, and an upper member comprising of white to grey medium to coarse grained sandstone with minor mudstone. The lower member hosts the No. 1 seam which is the largest and most ecomomically important seam. Most of the coal extraction at the Quinsam Coal Mines has come from this seam. Figure 2.2 presents a typical stratigraphic section for the Quinsam Coal Mine. 8 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 9 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 2.2: Typical Section At The Quinsam Coal Mine, depth of cover 100m a CU •. •*. It Depth (m) . Rock Type 0 to5 Basal till. Brown, dense, silty sand with trace clay, cobbles. 5 to 60 Sandstone with silty interbeds. Grey, medium to fine grained sandstone. Dark grey-black siltstone and mudstone interbeds. #3 Coal Seam (where present). 60 to 70 Shaly siltstone and coal with mudstone interbeds. #2 Coal Seam 70 to 100 Siltstone. Dark grey, fine to coarse grained siltstone. Becomes muddy siltstone with depth. 100 to 105 #1 Coal seam. >100m Carbonaceous mudstone, dirty coal seams, siltstone, volcanic rocks. Gardner (1992) reports that the change in lithology between the upper and lower members is due to a change in the depositional environment. The lower member was formed in a quiescent coastal swamp or lagoonal environment, while the upper member was formed in a higher energy fluvial or deltaic environment. This difference is reflected in the character of the coal seams. The No. 1 Seam, located in the lower member, consists of a relativly uniform thickness of clean coal, 2.0m to 6.0m thick. The coal is of bituminous rank with an average ash content of 10%, and a sulpher content of 0.6%. The coal seam varries from dull to bright. Minor coatings of calcite and pyrite crystals are common on the cleat surfaces. A carbonaceous mudstone "dirt band" parting is common within the seam. The No. 2 and No. 3 Seams, located in the upper member, are of variable thickness. They have sandstone partings and a sandstone roof and floor. The No. 2 Seam is found approximately 20 to 30m above the No. 1 Seam. This seam averages just over lm in thickness. The No. 3 Seam is found approximately 35 to 45m above the No. 2 Seam. This seam averages 5m in thickness, of which about 2.5m is clean coal, the remainder being numerous partings of sandstone and mudstone. The No. 3 seam is not always present. Thin "rider" coal seams occur above the No. 1 and No.2 seams. The rider seams are located between several centimetres and several meteres above the main coal seams. In localized situations they are incorporated into the main seams. 10 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH In general terms, the structure within the Middle Quinsam Lake area is that of a gentle syncline plunging to the northwest. The coal seams outcrop along the southwest limb of the syncline on the north side of Middle Quinsam Lake. On the eastern limb of the syncline the sedimentry formation butts up against the volcanic ridge that separates the sub-basin from the coastal area. The synclinal structure is approximately 4km wide and 10km long. Regional bedrock geology and structure is shown on Figure 2.3. The structure of the area is complicated by numerous faults and folds. Displacements on major faults is reported by Gardner (1992) to be as much as 18m. Locally the dip of the coal seam on the limbs of the folds can reach 20 degrees. 2.2 MINING AT THE QUINSAM COAL MINE The mining method used at the Quinsam Coal Mine was room and pillar with retreat pillar extraction. This is one of the most common methods of coal extraction in North America. In the United States, room and pillar mining accounts for approximately half of the coal production; most of the remainder comes from longwall mining. Mining commenced by driving a set of three tunnels, called "Mains", into the coal seam. In most cases one tunnel was used for access and ventilation intake, one tunnel was used for the conveyor that transports the coal from the mine, and the third tunnel was used for ventilation exhaust. The tunnels were spaced at 36m centres and connected by perpendicular tunnels called "cross cuts" which were located every 36m. This arrangement resulted in pillars that were 30m wide and 30m long as indicated on Figure 2.4. Most of the tunnels at the Quinsam Coal Mine were 6m wide. This was the optimum size for the mining equipment used, which was 3m wide Continuous Miners. Photo 2.1 shows a Continuous Miner cutting coal. Mine tunnels that were more than 6m wide required a third pass of the mining equipment; tunnels that were less than 6m wide required a partial pass that made less efficient use of the equipment. 11 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 2.3: Regional Geology of the Middle Quinsam Coal Basin, after BRINCO Mining Ltd. u QCWt-- a CM f • to. 1 2 E >- u. o 2 CL 1 t 1 3 . 7i «•> 3 E Q 12 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 2.4: Typical Mine Layout for Room and Pillar Mining MAINS DEVELOPEMENT J ~ _ SLAB CUT 7 PANEL ROADWAY EXCAVATION BARRIER PILLAR CROSSCUT BARRIER PILLAR MAINS P I L L A R 30 o 6 CO t p l cot COl CO SLAB CUT BARRIER PILLAR SLAB CUT 7 CM CO CM CD C\J PANEL PILLAR "30" o -o CN* SLAB CUT SCALE: 1:2000 50 100 13 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Photo 2.1: Continuous Miner Cutting a 6m Wide Roadway 14 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Where unfavourable ground conditions were encountered, the width of the tunnels was often reduced; however, reducing the width had the disadvantage of slowing down shuttle cars (coal transporters) which worked most efficiently with a full 6m width. The tunnels were typically advanced in 6m wide and 6m deep increments called "cuts". The 6m depth was selected as it is the distance between the cutting head on the Continuous Miner and the front of the operators' controls. When the Continuous Miner was advanced 6m the operator would still be beneath ground that had been supported in the previous cut. Where unfavourable ground conditions were encountered the cut size was often reduced to as little as 3m wide and 1.5m deep. After each cut the Continuous Miner was pulled out and the ground support was installed. The primary ground support consisted of resin anchored roof bolts and wire screen; wood posts and wood cribs were installed as needed. Photo 2.2 shows a bolt being installed with a "Fletcher Roof Bolter". "Panels" were driven off the mains. The panels were first developed in a "room and pillar" fashion. There were usually 5 tunnels or roadways per panel. The tunnels were typically 6m wide and spaced at 18m centres. Cross cuts, spaced at 36m intervals, connected the tunnels together to form a regular pattern of pillars 12m wide and 30m long as indicated on Figure 2.4. The length of the panels ranged from 100m to 500m. Barrier pillars were left between adjacent panels as well as between the panels and the mains. The barrier pillars between panels typically started at 30m wide but were then reduced to 18m wide by taking a "slab cut" as shown on Figure 2.4. In some instances these pillars were reduced to as little as 3m. The sole purpose of the barrier pillars between panels was to prevent connectivity of air, and in some cases water; they were seldom required to be load bearing. In most cases it was of little consequence if these pillars yielded. The barrier pillars between the panels and the mains were typically 30m wide. These pillars were designed to prevent connectivity of gases and water, as well as protect the mains from damage induced by stresses transferred from the mined out panels. 15 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH After the room and pillar layout was completed, "depillaring" commenced. The pillars were extracted starting from the far end of the panel and working back towards the mains. A "Christmas Tree" cut pattern was used to recover the pillars. Angled cuts, 6m deep, were made with the Continuous Miner, as shown on Figure 2.5. In this way, a significant amount of the 12m wide pillars was extracted. The mined out area is referred to as the "gob". Stump pillars and fender pillars (see Figure 2.5) were often left in the gob for ground control and safety reasons. These pillars were used to provide local support to unfavourable geologic structure as well as to warn of the impending gob cave. Photo 2.3 shows two stump pillars in the mined out gob. Point pillars are the triangular shaped pillars left at the roadway intersections. Mark (1997) reports that, in the US, 50% of the fatalities associated with retreat pillar mines occurred when mining the point pillars. Initially the practice at the Quinsam Coal Mine was to recover the point pillars if at all possible. This practice was changed to leave all point pillars in place. When the area of the gob reached a sufficient size the roof collapsed; this is called a "gob cave". The initial cave height was usually only a few metres, as shown in Photo 2.4. As the overall size of the gob increased, the height to which the caved zone extended also increased. Once a critical gob size was exceeded the effects of mining were seen on the ground surface above the mine. At shallow depth the effects are often quite significant and include open fractures and sinkholes such as shown in Photo 2.5. The mine development and extraction described above resulted in a somewhat regular underground arrangement of roadways, pillars and panels, as illustrated in Figure 2.6. This Figure shows a map of the 2N Mine at Quinsam Coal in 1996. The map shows panels in the process of development, panels that have been fully developed, panels in which depillaring has started, and panels that have been fully depillared (now caved). Most of the fieldwork conducted for this research project was carried out at the 2N Mine in the areas shown on Figure 2.6. 16 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Photo 2.3: Stump Pillars within a Gob Photo 2.4: Initial Gob Cave 17 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 2.5: Retreat Pillar Mining o - POSTS A - ROADWAY B - STUMP PILLARS C - FENDER PILLAR, Other Mined Out or Left Depending on Roof Conditions D - REMAINING COAL PILLAR E - INTERSECTION POINT PILLARS, not to be mined 1st PASS , 45deg. to ROADWAY 2nd PASS . 60deg. to ROADWAY 18 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MPWING AT SHALLOW DEPTH Photo 2.5: Sinkhole Subsidence; Depth Of Cover Is 40m 19 20 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH CHAPTER 3 SITE CHARACTERIZATION 3.0 INTRODUCTION The first phase of this research project was to establish a geomechanics database to characterise conditions at the Quinsam Coal Mine. The purpose of this database was to facilitate meaningful comparison of the rock mass to other mines, as well as to establish the values for properties that were considered to affect rock mass response to mining at shallow depth. The literature review suggested that there are many properties which influence rock mass response. These can be divided into the following three groups: 1. Geologic discontinuities (faults, joints, bedding planes, slickensides, cleats). 2. Intact rock and rock mass mechanical properties (strength, deformation, durability, etc.) 3. In situ stress and mining induced stress changes. 3.1 GEOLOGIC DISCONTINUITY DATA Geologic discontinuities at the Quinsam Coal Mine consisted of faults, joints, slickensides, bedding planes, and cleats. The characteristics of the various features are discussed below. The implications to mining are discussed later in this chapter. Bedding planes were found throughout the property, in every rock unit. The bedding planes were less well developed in the sandstone. Bedding plane orientation was typically dip to the north-east at between six to ten degrees. Cleats are similar to joints, but typically they only occur in the coal seams. Cleats are created by stress changes; however, there is still some debate over the precise mechanism of formation (Jeremic 1985). There were two cleat directions at the Quinsam Coal Mine. Both were near vertical; one strikes south-west, the other north-east. The intensity of cleating was observed to vary considerably throughout the property. 21 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Slickensides are discontinuities that formed prior to consolidation of the sediments. They are caused by such factors as differential settlement and small slump failures along the sides of river channels. Slickensides have smooth surfaces with little or no cohesive or tensile strength. At the Quinsam Coal Mine slickensides predominately occurred in the siltstone above the coal seam. The orientation of the slickensides was found to be highly variable. Most of the faults identified at the Quinsam Coal Mine were normal faults. These are the most common types of faults found in coal measures rocks (Whittaker and Reddish 1989). Normal faults are considered to occur under a state of high vertical stress and low horizontal stress. Fault displacements at the Quinsam Coal Mine varied from several millimetres to many meters. When the offset was very small it was often difficult to distinguish between faults and joints; the two are commonly associated and have similar orientations. The distinction between faults and slickensides was also difficult to make at times. Faults usually crossed through all strata while slickensides only occurred in the siltstone. During the course of the field research the major faults and joints were mapped in the 2N Mine. A complete listing of the characteristics of these features (dip direction, dip, off set, infilling) is given in Appendix 1. Al l orientations used in this report are based on the Quinsam Coal Mine co-ordinate system that is rotated 52 degrees counter-clockwise from the true north co-ordinates. Figure 3.1 shows a contour plot of the poles to the major faults and joints. The mean values of the 4 major planes, as determined from the contour analysis, are also shown. It should be noted that there were many joints and faults with random orientations. Figure 3.2 shows a histogram of fault displacements for the faults mapped in the 2N Mine. The mean fault displacement was found to be 0.85m with a standard deviation of 0.82m. More than 25 % of the faults had a displacement greater than 1.0m, while approximately 42% had no displacement. Figure 3.3 plots the faults as poles on a stereonet with the displacements indicated. The dominant fault orientation was found to be strike north-south and dip either to the east or west. The north-south striking faults typically had the greatest persistence and were the ones most likely to have displacements greater than 1.0m. 22 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 3.1: Stereonet Projection Showing Contours of Pole Concentrations and Four Major Joint/Fault Orientations 2N Mine: Faults and Major Joints, 289 observations MAJOR PLANES M ORIENTATIONS # DIP/DIR. 1 M 46/Q95 2 n 44/176 3 n 42/267 4 M 49/338 EQUAL ANGLE LUR.HEMISPHERE Figure 3.2: Histogram of Fault Displacements 2N Mine: Histosrran of Fault DisplaceKents ( H e t n - s ) Use caution when u t i l i z i n g GLOBAL MEAN UECIORS trend/plunm : Unweighted : 328 / 84 : Weighted 328 / 84 ct Arithiwtio 3 3 St.Deu.= Fault x - ^ ^ V B.8211 Offset A g 26 (m> 2 ' i8<'Number "•^ 13 of faults 2 5 BJB1 Bar length • . - represents _ 1 5 V. of total J poles I 3 . 5 ' 1 1 1 1 1 1 1 1 i 20 40 60 80 100 Percentage of faults in each category 23 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 3.3: Stereonet Projections Showing Poles of Fault Orientations As Well as Fault Displacement 2N Mine: F a u l t Poles With Displacement <metres) POLE PLOT H * 7X, y ° V / + • ° + « \ -j + + + + * + Y / 8 *" * \ ~i * ° o 4+ r~ 1 • e * + *+° **+° L \ • L \ ° + / \ + * * L \ ° / D1SP. + < 0 • < 0 5 o < 1 o < i 5 • < 2 . < 2 5 • < 3 < 3.5 EQUAL ANGLE LMR. HEMISPHERE 173 POLES 173 ENTRIES Faults were the single biggest cause of unstable ground at the Quinsam Coal Mine. As well, faults with significant displacements disrupted the continuity of the reserves, making mine planing and coal recovery difficult. Further discussion on the implications of faults is given in Section 3.5.1. 3.2 ROCK MECHANICAL PROPERTIES The mechanical properties of intact rock from the Quinsam Coal Mine was determined by laboratory tests. Al l testing was carried out at the University of British Columbia, in general conformance with ASTM and/or ISRM recommended procedures. 24 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH The mean values for the mechanical properties determined from the laboratory tests are given in Table 3.1. Insufficient testing was carried out to perform a statistical analysis or comment on the distribution of the results. Typical values from the literature (Lama and Vutukuri 1978, Bell 2000) are also given on Table 3.1. Complete records of the testing are given in Appendix 1. CANMET (1996) also determined the uniaxial compressive strength (UCS) and Young's Modulus for coal, siltstone and sandstone samples from the Quinsam Coal Mine. The values reported by CANMET were 20 to 30% higher than those determined by the Author. For the uniaxial compressive strength tests (UCS) cylindrical samples and cubes were cored and saw cut from large blocks of rock. Due to the very weak bedding plane cohesion in the coal and siltstone it was only possible to obtain test samples perpendicular to bedding. Attempts to core and cut parallel to bedding resulted in sample breakage. This condition made sample preparation extremely difficult as well as resulted in testing in one direction only. Attempts were made to grind and polish the ends of the test specimens, as per ISRM standards, however, due to the weak nature of the rock this was not always possible. Anomalous results, such as failure during seating of the load platens on the UCS test rig, where excluded. Hanna et al (1984) report similar problems; they concluded that laboratory test results from weak rocks may be biased as the weaker samples may not survive the necessary preparation for testing. The UCS was also estimated from point load strength index values (Is(so)). Point load tests were carried out on approximately 200 samples obtained from drill cores. The results are summarised in Table 3.2. Full details of the point load tests are presented in Appendix 1. The Is(50) values determined at the Quinsam Coal Mine are consistent with the IS(5o) values determined for similar rock types by others, i.e. Molinda and Mark (1996), Bell (2000). The estimated UCS utilised a conversion factor of 12.5 for siltstone, and 17.4 for sandstone and coal, as per the recommendations of Vallejo (1989, reported in Molinda and Mark, 1996). The UCS values, estimated from the point load strength are seen to be almost double the test frame values. This is attributed to the problems with test frame sample preparation and the 25 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH use of larger samples in the test frame (greater potential for weakness planes to affect results). The anisotropy index represents the ratio between the (IS(5o)) perpendicular to bedding (axial) and the strength parallel to bedding (diametral). As was expected, the strength parallel to bedding is significantly less than that perpendicular to bedding. In general, the anisotropy ratio was found to increase with decreasing rock strength and increasing prominence of bedding planes. The tensile strength was determined by the Brazilian test. Shear strength was determined using a direct shear box. The shear tests were conducted to determine shear strength parallel to bedding as this was considered to be the weakest discontinuity. Based on limited testing it appears that the angle of friction on bedding planes is small and that cohesion accounts for most of the shear strength. Table 3.1: Mechanical Properties of Rocks. Mean Values from Quinsam Coal and Typical Values from the Literature (Lit.). Values in brackets are the number of tests carried out. UCS (MPa) Quinsam Lit. Young's Modulus (GPa) Quinsam Lit. tensile strength (MPa) Quinsam Lit. peak shear strength at a n =0.6 MPa Quinsam density (g/cc) Quinsam Lit. slake durability index Quinsam coal 9.6 6 to 2.8 1 to 1.0 0.5 0.6 1.8 1.8 97 (10) 69 (5) 29 (11) to 3 (3) (2) (4) weak <1 0.5 1.8 siltstone to 2.3 to 2.5 (2) siltstone 8 1 to 8.9 0.5 0.2 2.7 2 to 91 (6) 100 (3) to 96 to 36 (4) 2.8 (3) sandstone 40.5 4 to 20.7 1 to Ito 2.6 2 to 97 (8) 168 (2) 100 19 (2) -2 .8 (1) Table 3.2: Point Load Strength and Estimate of UCS Is (50) Diametral (MPa) IS(50) Axial (MPa) UCS Diametral (MPa) UCS Axial (MPa) ANISOTROPY INDEX coal 0.1 (6) 1.0 (6) 1.7 17.4 10 siltstone 0.2 (32) 1.5 (32) 2.5 18.7 7.5 sandstone 2.4 (31) 3.7 (29) 41.8 64.4 1.5 26 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 3.3 ROCK MASS CLASSIFICATION Rock mass classification systems combine together the environmental conditions and the mechanical properties considered to govern rock mass behaviour. Rock mass classification is an important tool in geotechnical engineering. Systematic classification of a rock mass provides a semi-quantitative description that can be used for comparison and prediction of rock mass behaviour. Many classification systems are currently in use at mines, the most common being: the rock quality system "Q" (Barton 1974), the rock mass rating system "RMR" (Bieniawski 1974), and the coal mine roof rating system "CMRR" (Molinda and Mark 1994). Al l of these systems emphasise the influence of geologic structure on the competence of the rock mass. Only the CMRR was developed specifically to account for the bedded nature of coal measures rocks. Full details of these classification systems can be found in the cited references. Rock mass classification was carried out throughout the Quinsam Coal Mine. The average values that were obtained are presented in Table 3.3. Additional details of the collected data are given in Appendix 1. The siltstone was broadly divided into two categories: weak siltstone and strong siltstone. Weak siltstone commonly occurred at the coal contacts and between the #1 Seam and the rider seam where they were in close proximity. The quality of the siltstone was noted to improve as the separation distance between the #1 seam and Rider seam increased. The mudstone occurred as a "parting" layer within the No. 1 Seam. The thickness of the mudstone was seldom more than 0.3m. Table 3.3: Typical Rock Mass Classification Values Type R M R Q C M R R General Description coal #1 seam 45 16 fair coal #1 rider 45 8 40 poor to fair mudstone 24 0.1 very poor to poor siltstone -weak 29 0.3 25 to 35 poor siltstone -competent 40 1.3 35 to 45 poor to fair sandstone 82 45 80 good 27 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 3.4 IN SITU STRESS A state of stress exists prior to any underground mining occurring; this is referred to as the in situ or pre-mining stress condition. The ability to predict mining induced stress changes first requires knowledge of the in situ stress. The pre-mining state of stress at the Quinsam Coal Mine was not directly measured; it was estimated using local knowledge, visual observations and theoretical methods. As has been discussed previously, the in situ stresses at the Quinsam Coal Mine were considered to be low when compared to most underground coal mines. The vertical component of the pre-mining state of stress (crv) at the Quinsam Coal Mine was estimated as being equal to the load imposed by the overlying rock and determined as: o-v = y g H (3.1) where: H = depth y = unit weight of overlying rock This estimate is based on proven experience throughout the mining industry, Herget (1988), Hoek.et al(1995). The horizontal components of the pre-mining stresses (ah) are a result of the sedimentary depositional environment, paleo tectonic stresses, active tectonic stresses, and a component of the vertical stress due to the Poisson effect. Although much of Vancouver Island is known to be subject to high horizontal stresses as a result of past and present tectonic activity, the pre-mining horizontal stresses at the Quinsam Coal Mine were estimated to be quite low. The k value (cm/ov) w a s estimated to be 1.0 based on visual observations, theoretical estimates, and measurements of stress at coal mines in the Western United States. Table 3.4 presents the estimated in situ stress values for several depths. Based on an elasto-static thermal stress model, Sheorey (1994, reported in Hoek et al, 1995), developed the following equation for estimating k: k = 0.25 + 7E h (0.001 + 1/H) (3.2) where: Eh= horizontal modulus of deformation (in GPa) 28 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Hoek et al (1995) report that this model predicts results that are consistent with actual stress measurements throughout the world. Using an average value of 8 GPa for Eh, k equals 1.02. Recent stress measurements at coal mines in the United States show that at depths less than 100m, the k ratio varies between approximately 0.8 and 2.1, with most values close to 1.0. Figure 3.4 plots the summarised results from these tests. The data scatter is consistent with results from hard rock environments; Hoek and Brown (1980) report considerable scatter in measurements of horizontal stress at shallow depth in hard rock. Figure 3.4: Horizontal Stress Measurements from Coal Mines in the Western U.S., after Mark and Mucho (1994) MAXIMUM HORIZONTAL STRESS. MDa Table 3.4: Estimated In Situ Stress For k = 1.0 Depth of Cover (m) Vertical Stress o v (MPa) Horizontal Stress o h (MPa) 50 1.3 1.3 100 2.6 2.6 200 5.2 5.2 500 13.0 13.0 29 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH There was no evidence of unusually high horizontal or vertical stresses at the Quinsam Coal Mine, with the exception of very occasional floor heave and squeezing of the mudstone parting. These occurrences were attributed to the presence of locally weak material, and not to high stress per se. Figures 3.5a and 3.5b illustrate these types of stress related failures. In the case of the very weak mudstone parting (Figure 3.5a), the mudstone squeezed out much like toothpaste. The flowing material would sometimes drag or undercut the skin of the pillars, which in turn would result in sloughing off the sides of pillars. Where floor heave occurred it was attributed to the presence of a very thin beam of competent coal over a weak carbanaceous mudstone, see Figure 3.5b. Figure 3.5: Models of Stress Induced Failure at the Quinsam Coal Mine f r$ mudstone Squeezing mudstone drags skin off pillars with it. A: Weak mudstone parting is squeezed out under relatively low vertical stress. Squeezing action B: Weak carbonaceous mudstone is squeezed out under relatively low vertical stress. A thin beam of competent coal is pushed into the excavation. Horizontal stresses may contribute to failure due to buckling—mechanism in the thin coal beams. 30 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH The strength of geologic discontinuities is typically much lower than that of intact rock. Stresses that are too low to cause intact rock to fail may still cause tensile or shear failures on weak discontinuities. High stress is a relative term and must be reported in relation to the strength of the material being considered. 3.5 IMPLICATIONS FOR MINING 3.5.1 Geologic Structure Based on the site visits to other mines, personal communication with Mark (1998), and from the literature review, it was determined that the frequency of the joints and slickensides at the Quinsam Coal Mine was similar to that of other North American coal mines; the frequency and displacement of faults was dissimilar to most other mines. It was determined that the effects of joints and slickensides on mining were similar to other mines operating in similar rock types. Faults at the Quinsam Coal Mine had a greater frequency and greater displacements than most other mines in North America. Hence, faults also had a greater impact on mining. Faults with displacements greater than 1 .Om significantly affected the mining method used as well as the economics of mining at the Quinsam property. On account of these faults, a very flexible mining method was required; longwall mining would not have been economically viable, as this method requires relatively large continuous blocks of coal with uniform strike and dip. Analysis of the fault and joint data shows that there is significant potential for block and wedge type failures. Also, as the frequency of discontinuities increases, the rock mass typically becomes weaker and more susceptible to progressive unravelling. Instability related to geologic discontinuities is referred to as structurally related failures; this is discussed in detail in Chapter 6. 31 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH As discussed in Section 3.1, the majority of the faults and joints had a north-south orientation. Underground excavations that trended north south (parallel to the major faults) were more likely to have stability problems than excavations that trended east-west. Wedges with the kinematic potential to fall were more likely to form in the north-south excavations. The wedges in the north-south excavations would also be much larger. Figure 3.6 illustrates these difference using the mean dip and dip direction from Figure 3.1. As can be seen the potential wedge in the east west excavations is only 1.6m high and 0.8 tonne in weight. This contrasts sharply with the north-south excavations where the potential wedge is up to 2.7m high and 307 tonne in weight. Efforts were made at the mine to minimise north-south drives, and to cross all known faults at right angles whenever possible. Figure 3.6: Potential Wedges Formed in N-S and E-W Trending Roadways 32 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Faults and joints are also known to affect caving and subsidence. The gob will cave at a smaller size where faults and major joints are present. Unexpected gob caves may also occur; this is discussed further in Chapter 7. Depending upon the orientation, faults may either result in an increase or decrease in the area affected by surface subsidence. Faults may also result in unusual surface displacements and connectivity for air and water between surface and the underground workings. The effects of faults on subsidence are discussed in Chapter 8. 3.5.2 Mechanical Properties and Rock Mass Classification The mechanical properties and rock mass classification values determined at the Quinsam Coal Mine were within the range of typical values reported in the literature. As such, the effects of the rock properties on mining are expected to be similar to those at other mines. 3.5.3 Mining Induced Stresses Mining alters the in situ stress conditions creating areas of both stress concentration (higher stress) and stress relief (low compressive and tensile stresses). Understanding the relationship between stress, geologic structure, rock strength, and excavation stability is critical to mine design. Stress induced failures are a function of both the rock strength and the applied stresses; i.e. moderate strength rocks may be stable in a low stress environment, but may fail in a high stress environment. Geologic discontinuities are also an important factor in excavation stability. The stability of an underground excavation tends to decrease with increasing span; this is the result of several factors including: • Potential to expose more geologic structure. • Longer beam lengths across the span. • Increase in induced stresses at the edges of the excavations. • Formation of a stress relief zone over the centre of the excavation span. 33 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH At shallow depth, the horizontal and the vertical in situ stresses are low, as are the mining induced stresses. As the depth of cover increases so do both the in situ and induced stresses. The two dimensional numerical model FLAC (Fast Lagrangian Analysis of Continua -ITASCA 1999) was used to investigate the induced stresses around mine excavations. FLAC is an explicit finite difference program designed to simulate the behaviour of structures built in or on rock or soil. The program is designed to simulated plastic flow of materials without numerical distress. Modelling is carried out by creating a grid of quadrilateral elements. The FLAC program further divides each element into two sets of constant strain elements. The grid is adjusted by the user to the desired shape of the problem. Each element within the grid is assigned a constitutive model type; this provides the stress/strain law that governs the element response to the applied forces and boundary constraints. The stresses and strains for each element can be determined at any time during the solution process. The program uses an explicit Lagrangian calculation scheme, mixed discretization, and the full dynamic equations of motion to model plastic collapse and flow accurately, and without numerical instability. Once the user defined yield criteria is exceeded the model material will yield and flow, the grid will deform and move with the material. The ability to model large strains and grid deformation was one of the reasons for selecting FLAC. Unfortunately it was found that the program was not able to handle the very large block deformations and rotations associated with caving ground. This is discussed further in Chapter 8. The finite difference method operates by translating the differential equations that govern the stress/strain behaviour into matrix equations; however a global stiffness matrix is not formed. The first step in the solution is to apply the equations of motion to derive new velocities and displacements from the applied stresses. The second step is to derive new stresses from the strain rates obtained from the velocities in the first step. This operation is carried out on an element by element basis, the newly calculated stresses are assumed to be frozen and do not affect the velocities. Although it is known that a stress change in one element will affect velocities in adjoining elements, the time step used by FLAC is so small that information 34 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH cannot physically pass from one element to another. After several computational cycles the disturbance will propagate outwards to other elements. Coetzee et al 1998 consider that "The central concept of the explicit method is that the 'calculation wave speed' always keeps ahead of the physical wave speed so that the equations always operate on known values that are fixed for the duration of the calculation. " Lagrangian analysis refers to the update of grid co-ordinate locations at each time step. The incremental displacements determined after each step are added to the existing co-ordinates so that the grid moves and deforms with the material it represents. More detailed technical discussions of the operation of FLAC can be found in Coetzee et al (1998) and ITASCA (1999). Numerical simulations with FLAC were carried out for four different depths and five different excavation span widths using a k value of 1.0. A purely elastic constitutive model was used for this analysis. The results from the analysis are summarised in Table 3.5. Both the minimum (o"mjn or CT3) and maximum (CTmax or cn ) induced stresses are reported along with the horizontal stresses that occur in the roof at the centre of the excavations (cxhor). The centre of the excavation is the location where the horizontal stresses are expected to be lowest. In all cases it was found that the maximum stresses occurred near the corners of the excavation with near vertical orientation. The minimum stresses occurred above the centre of the excavation. Typical graphical output from the FLAC program showing the orientation and relative magnitude of the principal stress, along with contours of the minimum and maximum stresses, are presented in Figure 3.7. The in situ stresses increase linearly with depth as indicated in Table 3.5. The relationship between the mining induced stresses a m a x and depth, and o-hor and depth, is also approximately linear for a given excavation span. Where amj n is compressive it also increases linearly with depth; however, where it is tensile (negative values) no relationship to depth is apparent. The magnitude and height to which the tensile stresses extend above the excavation increase with decreasing depth of cover and with increasing excavation span. 35 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 3.7: Orientation and Concentration of Induced Stresses around a 10m Excavation JOB TITLE : 50m Depth. 10m Span FLAC (Version 3.40) LEGEND 29-Sep- 1 20:21 step 12375 1.906E+01 <x< 5.140E+01 -7.026E+01 <y< -3.792E+01 Principal stresses Max. Value = 3.537E+06 Maximum principal stress Contour interval 5.00E+05 -1.500E+06 -1.000E+06 inirrmm_princip_a[ stress Contour interval 2.50E+05 A: -1.500E+06 G: 0.000E+00 Boundary plot I I I I I I 0 5E 0 Michael Cullen A X X * * x x x n ^ A X ^ x * * * + X * x X + + A A S X . i 1 < * + + + + xV_x x + + + + + - / - V / H- + k k ^k kk* k k k k k k XX XX XX X X X X + + + -V + k k + k k k X k k k k X X kkkXX X AT. k X X X X k^X-X XXX X X X~\X X k XX X X - X X X X X X x~ X X X X X X X X X X X X x x x x x x X x x x x-x. £ X ^ X\~*- -«-^.-t-.E: x x , x V X x v^-^ ~*T-+r-H--S»»-X*X X x ^ T A - A - I - : ^ X X x v. - A ^ - ^ F H 1 X X V>X, -A - A - A - t -X X X -X "A-Act-. + Bh X X - A - A + + X X X - A - A - A + -)—h X-X. -A A- -A - f -++ X x""-X--A-*-rA -fc - W : X. -A.-A--A-X ^ - A ^ T * --X X a X X X^X' X X - t -V X X X, X X.X ^.Jr-X'X X 4--A--V X X k-k-k-kx -t--Y- -V-Y-Y '-k-kk-XX + 4->r-VJf -V-V X X; A X X; X X X X X/X" X X X X X X X X X X xxx-_X-X X X X X J<X X Jf X X J c X - X A A + + + + A A A A A A x * A x x * x >*x • X X X x x x x x x x x x x x x x x x • < A A - A x A -A -A X A A A x x A X -X X x % X X X •* X X X X X xxx a) 50m Depth of Cover JOB TfTLE : 400m Depth. 10m Span FLAC (Version 3.40) LEGEND 2-Oct- 1 7:04 step 6418 1.951E+01 <x< 5.139E+01 -4.182E+02 <y< -3.863E+02 Principal stresses Max. Value = 2.712E+07 I 0 1E 8 Maximum principal stress Contour interval^ 2.50E+06 G: -1.000E+07 H: -7.500E+06 Minimum principal stress Contour interval 2.00E+06 B: -8.000E+06 H: 4.000E+06 Boundary plot I I I I I I 0 5E 0 Michael Cullen -A X V- X X A -A -A -A A A A -A •f A A + + A ..+ + + + + +' + + + + + 4-+ + + + + + + + + + + -f-+ + + + + + + + + + + +, + + k + k'k + -if k k k k kk k k'k X k X X X X-.X x v \ x XXX x x >< x x X X •A X •A X A A * - A A A - A A A + + A + + + +/+ + + + + yw kk k k k k k k X x x X X x x x X X V x x :U x x,'x A XjX A Xj. +A\ xxx X X X x x x ,k x x x k x x x x x X + _+-H~BK.-t-. + - A X X X ,*"•+• + + + 4- A% m x^ x x Jr + + + + + + -f. » % X J T + + + + + + X I ' . A X X *r A 1 f - -^-Y-X X V X X -+-..-IT & - . . - + - " A ' A X X V>-, X^ A? H 1—f~ ^ - K X X X >• -\——k^>>x X X x x x x • ~i h- -r- -v-y X X ^ -A"-f^.^t7£|T-+^-v- - V X X yU X x V x ' A - A - A - ( - - * - - t - ^ - A - A ' ^ X X X X V X " A - * - - + - - t - - t - - t - - * - - * ' j f ' ' x X X X x y-N~.-A •+• +d- -+- +>V> x x X X x - A - A 4- ^ F ^ F + -v- A - A - X X X X X X - A - A - A - t - + + + J i - J r X X X % H i - ) - + + + + + + + ^ A - X X X X X - A - A - / - + -t- + + + - V - - > r A - X X X X - A X . - A + - f - - r - + + + J r A - - > r X X X X X X X X X X X X X k k k k X k k k k + k + + + + + \ + + + + '+ + -f + . '+ + A -f>+ A A A A A A A A X A A X X X X X X X X X x x x XXX \ k X \ k k k k k k k k k + + + + + + + + '+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + ;+ + + A+ + AVA + A A A A A, -A A A; A X X; A X * X X .X X X / X X A X X 3.250 3.750 (•10-1) b) 400m Depth of Cover 36 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Table 3.5: Mining Induced Stresses Determined from Computer Simulations. The numbers in brackets are the distance, in metres, to which tensile stresses extend above the roof. Depth (m) In Situ Stress (MPa) 5m Span Stress (MPa) 0~max G"min ^hor 10m Span Stress (MPa) G"max ^rnin ^hor 20m Span Stress (MPa) 0~max ^min ^hor 40m Span Stress (MPa) 0"max Omin ^hor 60m Span Stress (MPa) 0~max ^min ^hor 50 1.3 2.6 0.1 1.2 3.5 -0.1 (1) 0.7 4.9 -0.1 (4) 0.3 7.5 -0.1 (15) 0.1 9.7 -0.2 (30) -0.1 (1) 100 2.6 5.3 0.1 2.7 6.9 -0.0 (1) 1.4 9.5 -0.0 (2) 0.8 13. 7 -0.1 (9) 0.4 18. 6 -0.1 (17) 0.3 200 5.1 10. 5 0.3 4.7 13. 7 0.1 2.8 18. 6 -0.0 (1) 1.5 27. 1 -0.1 (5) 0.8 37. 2 -0.1 (10) 0.6 400 10.2 20. 9 0.6 9.4 27. 3 0.1 5.5 37. 3 0.0 3.1 54. 3 -0.0 (4) 1.7 74. 5 -0.1 (7) 1.2 The o"min stresses above the centre of the excavations were found to be sub-vertical in orientation except at the excavation boundary where only tangential stresses can occur. The potential for bed separation, beam bending, and gravity failures is greatest where a m m is tensile. The horizontal compressive stress in the roof decreases as the excavation span increases and the depth of cover decreases. Eventually a r 0 0f becomes tensile; in the elastic computer simulations this occurred at a 60m span and 50m depth of cover. The simulations indicated that a r o of is very low (less than 1.0 MPa) for all depths of cover less than 100m and spans greater than 10m. The above findings are contradictory when compared to published elastic stress distributions such as those found in Hoek and Brown (1980). The elastic analysis in these publications show that the extent of the stress relief zone is independent of depth for a given excavation size. Only the magnitude of the stress changes within the stress relief zone. This discrepancy is due to the presence of the ground surface and gravity. The published stress distributions typically assume that the depth is more than 10 times the span of the excavation, and that the applied stresses are uniform over the model area (i.e. they do not consider gravity). The extent of mining induced stresses is in the range of 2 to3 times the excavation size; any boundaries, such as the ground surface, within this distance will affect the resulting stress field. Gravitational stresses will also affect the stress field; Hoek and Brown (1980) consider that gravity must be considered in the stress analysis of large excavations at shallow 37 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH depth. Using FLAC it was determined that gravity and the ground surface have a significant influence on the mining induced stresses around a 40m wide by 4m high excavation up to a depth of approximately 150m. At deeper depths the mining induced compressive stresses may exceed the strength of the rock. At shallow depth the stresses are less likely to exceed the strength of the rock. For example, the FLAC simulations indicated that the maximum induced stress for a 10m wide excavation at 100m depth is only 6.9 MPa. Numerical simulations were carried out to assess the potential for stress induced failures. The program FLAC was used to model the extent of failure around excavations at depths of 50 to 400m. A Mohr-Coulomb constitutive model was used for this analysis. The following material properties were used as the input parameters: • Bulk modulus 2.0 GPa • Shear modulus 1.0 GPa • Cohesion: 4.0 MPa • Tensile cut off: 0.4 MPa • Angle of friction: 30 degrees • Density: 2600 kg/m3 Results from the analysis of a 10m wide excavation at depths of 50m and 400m are shown in Figure 3.8. The extent of the failed zone around the excavation is seen to double in size between a depth of 50m and 400m. Although both excavations are the same size and have the same rock properties, the response of the excavations and the support requirements would be quite different. 38 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 3.8: Extent of Stress Induced Failure around an Excavation at 50m and 400m Depth 16-OcH 18:42 step 11212 2.197E+01 <x< 4.931 E+01 -6.496E+01 <y< -3.762E+01 stale ~ j Elastic H Elastic. Yield in Past Maximum principal stress Contour interval 5.00E+05 E: -1.500E+06 F: -1.000E+06 Boundary plot I I I I I I 0 5E 0 Michael Cullen JOB TITLE : 400m Depth, 10m Span, Mohr-Coulomb Model FLAC (Version 3.40) LEGEND 16-Oct- 1 23:09 step 73621 2.043E+01 <x< 4.966E+01 -4.154E+02 <y< -3.862E+02 state • Elastic H Elastic. Yield in Past | At Yield in Tension Boundary plot I I I I I I 0 5E 0 Maximum principal stress Contour interval^ 2.50E+06 G: -1.000E+07 H: -7.500E+06 Michael Cullen 39 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 3.6 CONCLUSIONS The mechanical properties of the Quinsam Coal Mine rock mass were found to be similar to other coal mines operating in similar geologic environments. The occurrence of bedding planes, cleats, joints, and slickensides was also found to be similar to other coal mines operating in North America. The frequency and magnitude of faults was determined to be anomalous for coal mines in North America. Faulting had a significant impact on both operational and safety considerations at the Quinsam Coal Mine. Faults, and in some instances joints, created potentially unstable wedges in the excavation roofs. The frequency and size of the wedges was found to be greatest in excavations driven in a north-south orientation (mine grid). The Quinsam Coal Mine operates at relatively shallow depth of cover where in situ stresses are low compared to most mines. The vertical stress was estimated to be equal to the overburden pressure. The ratio between horizontal and vertical stresses (k) was estimated to be 1.0. Numerical simulations show that there are significant differences between the mining induced stresses at shallow and deep mines. Most notable is the finding that as the depth of cover decreases the size of the stress relief zone increases. The stress relief zone consists of low magnitude compressive and tensile stresses. The magnitude and height to which the tensile stresses extend above the excavation increase with decreasing depth and increasing excavation span. Rock masses are weak in tension. Geologic discontinuities often have little or no tensile strength. Where low compressive or tensile stresses exist along with geologic structure gravity related failures such as block falls, wedge falls, bed separation, beam bending and progressive unravelling may occur. Based on the stress results from the numerical simulations the potential for gravity related failures is highest at shallow depth mines. 40 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Gravity related block and wedge failures may also occur in high stress environments. However, where inclined joint and fault planes are present, higher stresses may reduce the potential for gravity failures. It can be shown that under even moderate stresses (greater than 1.0 MPa) joint planes with inclinations greater than 45° to the horizontal will be stable provided the effective angle of friction is greater than 45°, see Chapter 5. If the normal stresses across the joint planes are very low, (i.e. at shallow depth) wedges may fall regardless of the effective angle of friction. Most failures in a coal mine rock mass are associated with geologic structure. There are three common scenarios for structurally controlled failures at coal mines: 1) bedding plane dominated, 2) joint dominated, 3) both bedding plane and joint dominated. Figure 3.9 provides an idealised picture of the rock mass response of these three scenarios to a low and high stress environment. At the Quinsam Coal Mine bedding planes and joints (including faults and slickensides) were observed to control instability and ground falls in most situations. In situ stress levels at shallow depth mines, such as the Quinsam Coal Mine, are lower than at most other underground coal mines operating in North America. As such, the response of the rock mass to mining induced stresses is expected to be different. For example, the design of ground support in a high stress and low stress environment is quite different. In a low stress environment gravity induced failures dominate; the support requirements can usually be determined based on dead weight loading (gravity). Assessing the support requirements in a high stress environment is typically more difficult as both gravity and stress induced failures must be considered. Most of the existing geotechnical design tools were developed at mines operating at depths significantly greater than 100m. It therefore stands to reason that the existing design tools may not be directly applicable to shallow depth mines. 41 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 3.9: Idealised Rock Mass Response A: Failures under low stress conditions. CONTINUOUS JOINT SETS DOMINATE Gravity induced wedge failure. BEDDING PLANES DOMINATE Gravity induced beam failure. BEDDING PLANES AND JOINT SETS DOMINATE Gravity induced unraveling of blocks defined by joints and bedding planes. Stress B: Failures under high stress conditions. CONTINUOUS JOINT SETS DOMINATE Stress induced failures at corners of excavation. Horizontal stresses clamp wedges. BEDDING PLANES DOMINATE Stress induced buckling type failure parallel to bedding. BEDDING PLANES AND JOINT SETS DOMINATE Stress induced buckling and crushing failures at corners couped with gravity failures of failed blocks and wedges defined by joints and bedding planes. 42 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH CHAPTER 4 EXCAVATION STABILITY 4.0 INTRODUCTION For the most part, excavation design at coal mines is dictated by the size of the mining equipment. At the Quinsam Coal Mine the roadway excavations were typically 6m wide; this was the optimal size for the equipment used (3m wide continuous mining machines). This size is typical for coal mines using similar types of mining equipment. It is possible to create smaller excavations; however, this is not as cost efficient, as the full capacity of the continuous miner would not be used. As well, coal hauling is slower on account of reduced manoeuvrability of the equipment in the narrower roadways. Photo 2.1 shows a continuous miner starting a new excavation from a 6m wide roadway. Experience at the Quinsam Coal Mine showed that 6m wide excavations in siltstone, at depths of cover less than 100m, were generally stable once they were supported. Excavations that were stable when they were first developed usually remained stable. Most unstable excavations were unstable as soon as they were developed. The main exceptions to these observations were areas that deteriorated by weathering, and areas with adverse geologic structure. The purpose of this portion of the study was to provide scientific verification of the observed stability conditions at shallow depth, and to determine a means to identify potentially unstable excavations. Attempts were also made to evaluate if depth of cover and stress were factors affecting the methods used to determine instability. 4.1 LITERATURE REVIEW All underground excavations are subject to convergence or closure: primarily as roof sag. In a stable environment the rate of convergence is small and usually remains steady or decreases with time (Serata 1989). In an unstable environment the rate and amount of convergence is 43 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH found to increase with time. As such, rock convergence is often used to predict stability in underground mines. Figure 4.1 illustrates the use of roof convergence to predict stability. Figure 4.1: Example of Roof Convergence to Predict Instability; after Maleki (1988) Caved Rate = .585 in7rrn Cut ft Lunch 6 7 *- Tramming out One problem is that the amount of convergence, rate of convergence, and length of time convergence occurs prior to failure is variable. For example, in hard competent rock the amount of movement (convergence) may be very small and occur just prior to failure. In evaporite strata the amount of movement may be extremely large, and commence immediately after an excavation is created. In coal measures rocks the rate and amount of movement typically falls between the two extremes noted above. Table 4.1 summarises the results of coal mine convergencevstudies found in the literature. Several examples of convergence rates in hard rock mines are also given. Attempts have been made to present the data in a standardised format; however, this was not always possible. Definitions of stable and unstable ground varied from author to author, as did the methods of measuring convergence; in some instances it included the roof and floor, in 44 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH others only roof convergence was measured. For the purposes of this study, critical convergence is defined as the rate beyond which failure is imminent. Table 4.1: Roof Convergence Data from the Literature MINE DEPTH (m) R O C K S P A N (m) C O N V E R G E N C E Total Rate Condition mm rnm/min C O M M E N T S R E F E R E N C E Mary Lee 200 Shale 6.7 <25. 4 stable Total in 1 yr. Parketal 1992 Plateau mudstone 6 12.2 0.02 cave Maleki 1988 Plateau mudstone 6 .0004 critical Failure within 2 to 10 days Maleki 1988 Plateau mudstone 6 <,0002 stable Longwall Maleki 1988 Sufco 6 .0006 critical Maleki 1988 Island Creek 600 Shale/ sandstone 12.5 >25 >.0003 potential unstable Intersection >30m from gob Pothini, Schonfeldt 1978 shale 4.2 13 critical Fair quality rock mass Ghosh, Ghose 1992 Panelec 6 >1.2 unstable Failure is imminent Serata 1989 Panelec 6 .004 stable Serata 1989 Justus 6 >0.4 unstable Serata 1989 Justus 6 <.004 stable Serata 1989 Sufco gob 5.1 critical Pillar extraction Maleki 1988 Umgala sandstone gob 45 .01 critical Pillar extraction Naismith, Pakalnis 1982 Plateau mudstone gob 199 14.9 cave Pillar extraction Maleki 1988 Plateau mudstone gob 7.6 critical Cave within 20 minutes Maleki 1988 Plateau mudstone gob <5.1 stable Pillar extraction Maleki 1988 Helvetia gob 1.27 cave Adjacent to gob Serata 1989 Helvetia gob >0.42 unstable Adjacent to gob Serata 1989 Lucerne gob 2.1 unstable Cave-in imminent Serata 1989 Lucerne gob <0.42 stable Serata 1989 Ron court gob 60 >0.66 critical Deniauetal 1982 Mairy gob 30 >0.27 critical Deniauetal 1982 Serrouville gob 8 >0.25 critical Deniau et al 1982 Campbell 800 Norite 4 30 rockburst Hardrock Cullen 1988 Detour Lake Mafic >.0007 unstable Hardrock CANMET 1995 White Pine 6 .0003 critical Hardrock Maleki 1988 Several observations can be made from Table 4.1. • The critical rate of movement in coal mine roadways varies between 0.0004 and 1.2mm/min. • The critical total movement in coal mine roadways varies between 12.2 and 25.4mm. • The critical rate of movement for gob caves is greater than that for roadways. • The critical convergence rate in hard rock mines is typically less than that in coal mines. 45 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Many factors have been identified which influence roof instability and roof convergence in coal mines. Kane and Karmis (1986) statistically determined that discontinuity type, discontinuities orientation, discontinuity density, and presence of fossils were the most important factors to roof stability. Maleki (1988), considers roof span to govern instability, Ghosh and Ghose (1992) statistically determined that rock mass quality and span are the critical factors in roof stability. Pothini and Schonfeldt (1978) found that lateral stress and proximity to faulting had no appreciable affect on stability, while slickensides and proximity to the gob where very significant. Thus it appears that different factors may be significant in different geotechnical environments. Serata (1989) proposed that stability criteria can be established for any individual mine by analysis of the room closure rate versus the age of opening. 4.2 EXCAVATION CLOSURE MONITORING PROGRAM A monitoring program was set up at the Quinsam Coal Mine to evaluate excavation closure rates and stability. Several different instruments were used including simple mechanical borehole extensometers, rod extensometers, and magnetic borehole extensometers. Details of the instrumentation and results from the monitoring program are given below. Due to safety considerations the instruments could only be installed within supported ground; no measurements of stability in unsupported ground were carried out. It was also necessary that the instruments be intrinsically safe (safe for use in a potentially explosive environment). 4.2.1 Mechanical Borehole Extensometers The mechanical borehole extensometers used were single point instruments installed into drill holes. Instruments were installed in holes up to 5m long; this was the limit for the drilling equipment used (Fletcher Roof Bolters). The anchor point used was a simple spring clip. The measurement point was a scale inserted into a reference sleeve at the collar of the hole. The measurement point is connected to the anchor by either wire or rods (fibreglass chimney cleaning rods were found to work very well). The instruments recorded total 46 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH movement in the roof between the anchor point and the drill hole collar. Figure 4.2 shows a typical measuring point. Readings were taken visually; no enhancement of measurements was possible. Resolution was found to be +/- 1mm at close range, decreasing to +/- 10mm at approximately 10m. Figure 4.2: Mechanical Extensometer Green Yellow Red A n c h o r Cable Reference Tube Roofline Cable Crimp A total of 25 mechanical extensometers were installed throughout the Quinsam Coal Mine; five were installed to monitor the stability of a weak siltstone roof, seven were installed to monitor potentially unstable wedges formed by joints and faults, and 12 were installed in active retreat pillar mining panels. The instruments were installed to depths ranging between 3 and 5m. Installation depth was selected based on the estimated maximum height of instability. In all but one instance no movement was detected by the instruments. The one exception was an instrument installed to monitor the stability of a wedge of rock defined by three joints. The wedge was identified by detailed structural mapping after another wedge fell from the roof nearby. An analysis of the joints indicated that the maximum height of the wedge was 2m; a 4m extensometer was installed as indicated on Figure 4.3. Additional rockbolts were also installed into the wedge. Shortly after installation, the extensometer registered 3mm of movement. For the next few weeks the instrument was monitored constantly. No further movement occurred. In summary, it was found that the mechanical extensometers could be used to detect movement where it exceeded 1mm. However, the instruments cannot be read to a useful 47 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH resolution from a distance of more than a few meters. This makes them of only limited use except to detect large scale movements (more than 10mm). Figure 4.3: Mechanical Extensometer Installed in Potentially Unstable Wedge 4.2.2 Rod Extensometer A high-resolution telescoping rod extensometer called the "SERATA Ground Safety Meter" was used to monitor convergence throughout the Quinsam Coal Mine. This instrument consisted of a telescoping rod with dial readout. The instrument measured the total convergence of the excavation between the roof and floor, see Figure 4.4. Excavation closure rate was measured by installing the instrument between the roof and the floor of the excavation. The resolution of the instrument was +/- 0.0005 mm. Remote monitoring was not possible. Full details of the instrument construction and use are given in Serata (1989). Measurements were taken in roadways throughout the Quinsam Coal Mine. Locations included roadways just excavated (and supported), roadways which had been open for several years, and roadways at the edge of the gob. Safety considerations prevented readings from being taken in known unstable ground. The collected data is plotted in Figure 4.5 as excavation closure rate versus the age of the opening. A complete listing of the data is provided in Appendix 3. Mechanical Extensometer Joints 48 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 4.4: Telescoping Rod Extensometer, after Serata (1989) Figure 4.5: Data from the Rod Extensometer E X C A V A T I O N C O N V E R G E N C E R A T E V E R S U S E X C A V A T I O N A G E 0.0070 0.0060 0.0010 • S T A B L E R O A D W A Y S • S T A B L E E D G E O F G O B A A U N S T A B L E E D G E O F G O B A • • • • • • t • • • • -1 II • • A • - • 0.01 0.1 1 10 100 1000 Age of Excavat ion (days) 49 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH In the case of newly excavated and supported roadways, the convergence rate was found to be in the range of 0.001 to 0.005mrn/min. The convergence rate quickly diminished with time. In older stable roadways the convergence rate was found to be less than O.OOOlmm/min. At the edge of the gob, the convergence rate was found to range between 0.0001 and 0.003mm/min with lower rates being the most common. There were no visible signs of instability in the immediate vicinity of any of these measurements. Four measurements were taken in areas where there were visible signs of decreasing stability. In two instances anomalously high convergence rates were recorded; in the other two instances the convergence rates were not elevated. 4.2.3 Magnetic Multipoint Borehole Extensometer The magnetic borehole extensometer used was a G E O K O N 7000 Series Sonic Probe multipoint borehole extensometer (MPBX). The measurement points were magnets that were anchored within a drill hole. The magnet at the end of the hole served as the reference point. Measurements were taken by inserting a flexible probe into the hole; electrical pulses were used to determine the distance between each magnet. Figure 4.6 shows a schematic diagram of an instrument installation. The resolution of the instrument was +/- 0.025mm. Remote measurement was not possible as the flexible probe must be inserted into the borehole by hand. 50 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 4.6: Sonic Probe MPBX T o p m a g n e t is r e f e r e n c e p o i n t . M a g n e t s a n c h o r e d t o r o c k G u i d e t u b e T o r e a d o u t S o n i c e x t e n s o m e t e r p r o b e i n s e r t e d in g u i d e t u b e . Three MPBX were installed in the roof of roadways within retreat pillar extraction panels. The instruments were installed in 7.6m deep drill holes. Magnets were placed at 0.5m intervals. Monitoring was carried out as the gob approached. The data from two instruments indicated that no movement occurred prior to the instruments passing into the gob. The data from the third instrument indicated some movement occurred in the roof as the gob approached. The data from this instrument is presented graphically in Figure 4.7. The relative movement of 6 of the anchor points is plotted against date and distance to the gob. The anomalous spike that occurred when the instrument was 40m from the gob cannot be explained at this time. The maximum movement between the top anchor and the excavation roof was found to be only 1.4mm. The rate of movement was measured to be 0.0002mm/min (0.26mm/24hour). Most of the movement was found to be taking place in the intervals between 0.1 and 0.9m, and between 1.9 and 2.4m above the roofline. It is believed that the movement was related to bed separation at the contact between the rider coal and siltstone (located at 0.5m) and at a bedding plane in the siltstone located in the interval between 1.9 and 2.4m. This is illustrated in Figure 4.8. 51 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 4.7: Results From MPBX Installed in 2N Mine, #2 Mains -0.5 6-May 11-May 16-May 21-May Date 26-May 31-May Figure 4.8: Deformational Model of Bed Separation Beds separate Stress relief zone with bed sagging Rider coal Coal 52 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 4.3 E X C A V A T I O N S T A B I L I T Y V E R S U S R O C K M A S S C L A S S I F I C A T I O N It is a well proven fact that excavation stability decreases as the rock mass quality decreases (Bieniawski 1989). Several researchers have recently completed extensive studies relating the span of coal mine excavations to the CMRR (Coal Mine Roof Rating; see Chapter 3). Mark (1999) has studied the relationship between the stability of unsupported excavations, CMRR, and depth of cover. The results of this investigation are reproduced in Figure 4.9. This Figure clearly shows the significance of induced stress (which is directly related to depth of cover) on the stability of underground excavations; as the depth increases the stability of the excavation decreases for a given excavation size and CMRR value. Mark and Chase (1994) investigated the relationship between the stability of excavations, CMRR, and the width of excavations in longwall mines. The results of this investigation are reproduced in Figure 4.10. Molinda et al (2000) investigated the relationship between the stability of excavations, CMRR, and width of roadway intersections. The results of this investigation are reproduced in Figure 4.11. In Figure 4.11 the intersection span is determined as the sum of the diagonals across the intersection. The ground fall rates are defined as follows: • Low: 0 to 0.001 falls per 100 intersections. • Moderate: 0.001 to 0.05 falls per 100 intersections. • High: greater than 0.05 falls per 100 intersections. Spans at roadway intersections are typically much greater than elsewhere. Mark and Barczak (2000) report that although intersections account for less than 25% of the underground roadway system, they account for approximately 70% of all roof falls. This work clearly shows the relationships between stability and the CMRR. 53 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 4.9: Relationship Between Stability, CMRR, and Depth (Stress) for Unsupported Excavations Greater Than 6m Long; after Mark (1999) Depth of Cover (m) 0 100 200 300 400 500 600 100 j 1 1 -+- -H H Y O 90 J-80 J . 70 60 J. 50 4-40 30 _L 20 . . 10 J . 0 I I O A i A O Quinsam Coal Mine average values + A O Never Stable A Sometimes Stable • Always Stable Extended Cut Equation H 1 1 1 - i 1 1 H 200 400 600 800 1000 1200 1400 1600 1800 2000 Depth of Cover (ft) Figure 4.10: Relationship Between Stability, CMRR and Width for Supported Roadways; after Mark and Chase (1994) 12 100 90 O 80 < cc 70 O O cr LU < o o 60 50 40 30 20 14 ENTRY WIDTH, ft 16 18 Strong . roof Moderate roof m Weak roof 20 $ • y • n 5 ob B - o — A -i • 3.7 4.3 4.9 5.5 ENTRY WIDTH, m m a -a-22 Entry width=5.5m J r » ( 1 8 f t ) KEY • Satisfactory cases • Unsatisfactory cases Quinsam Coal Mine average values 6.2 6.8 54 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 4.11: Relationship Between Stability, CMRR and Width for Supported Roadway Intersections; after Molinda et al (2000) Data obtained from the Quinsam Coal Mine have been added to Figures 4.9, 4.10 and 4.11. Average CMRR for competentsiltstone (CMRR of 40) and weak siltstone (CMRR of 30) are shown. The weak siltstone is seen to plot below the recommended design lines with other data identified as "sometimes stable" and "not satisfactory", this is consistent with experience at the Quinsam Coal Mine when this rock forms the immediate roof. The competent siltstone also plots slightly below the recommended design lines; however, experience at the Quinsam Coal Mine is that this material is usually stable with a low rate of ground falls. The Quinsam data should not be considered anomalous as there are other data points (both successful and unsuccessful) that plot on the wrong side of the design lines. It is believed that the design lines were drawn to provide a high degree of confidence. 4.4 CONCLUSIONS Convergence measurements have been used to verify that 6m wide roadway excavations in competent siltstone (CMRR value of 40), at depths of cover less than 100m are generally stable. Excavation convergence measurements have been shown to be an effective way of 55 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH assessing the stability of excavations where the potential instability is caused by wedge failures or bedding plane separation. Where failure is due to progressive unravelling, the effectiveness of excavation closure as a means to evaluate stability is questionable. Although no measurements were obtained, the author considers that the incremental movements associated with progressive unravelling are likely too small, too irregular, and occur over too short a time period, to be amenable to stability evaluation using convergence measurements. When roadways are first excavated and supported, the convergence rate may be as high as 0.005mm/min (7.2mm/24hour). The convergence rate for stable excavations is expected to quickly diminish. In older, stable, excavations the closure rate is expected to be less than O.OOOlmm/min (0.14mm/24hour). Based on limited measurements of unstable roadways it is suggested that, at the Quinsam Mine, where the rate of convergence is increasing, the critical rate of convergence for roadway excavations should be approximately 0.0035mm/min. Convergence rates greater than 0.0035mm/min (5mm/24hour) may indicate the onset of instability. This rate is consistent with values reported in the literature, including values from mines operating at significantly deeper depths. The critical convergence rate at coal mines is seen to be significantly greater than that at hard rock mines, which is typically around 0.0007mm/min (1 mm/hour). Insufficient data was collected to determine a critical value for total closure in roadway excavations. At this time all that can be said is that it is greater than 3mm. The Quinsam Coal Mine continues to install single point mechanical borehole extensometers in potentially unstable areas of the mine. These extensometers have been arbitrarily set up to indicate "caution" after 10mm of movement, and "danger" after 20mm of movement. The mine reports that no movement has ever been seen in any of these instruments (Galovich, 2001). The stable roadway excavations at the Quinsam Coal Mine were found to remain stable right up to the edge of the gob. Minor movement associated with bed separation may occur in the roof as the gob approaches; however, this does not necessarily indicate instability. No 56 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH indication of impending gob cave could be drawn from the convergence measurements taken at the edge of the gob. The affect of stress and depth of cover on critical convergence rates are not well defined. Within the literature some authors maintain stress and depth are factors (i.e. Ghosh and Ghose 1992) while others maintain they are not factors (i.e. Pothini and Schonfeldt 1978). Results from the Quinsam Coal Mine are inconclusive suggesting that depth and horizontal stress are not important factors in determining the critical convergence rate for excavation stability where the depth of cover is less than 100m. 57 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH CHAPTER 5 GROUND SUPPORT 5.0 INTRODUCTION The purpose of ground support is to provide a safe working environment while allowing economic extraction of the mineral resource. Hoek and Wood (1988) consider that support systems work best when they knit the rock together to enhance and mobilise the inherent strength of the rock. In terms of coal measure rocks this means the following: 1. Provide confinement. Rock is much stronger when it is confined. 2. Increase the normal forces across geologic discontinuities. The shear strength of bedding planes and joints is directly proportional to the normal stresses across these structures. 3. Prevent deformation. Rock masses that have been allowed to dilate are weaker than when in an undisturbed state. 4. Prevent progressive unravelling. Maintaining the integrity of the original excavation boundaries helps to provide confinement and prevent deformation of the rock beyond the excavation boundary. 5. Anchor weaker rock to stronger rock. The stability of weaker rocks around an excavation boundary may be enhanced by connecting them to stronger rocks located beyond the boundary. There are many factors that influence the support requirements and design at a mine. The following list presents some of these factors: • State of stresses (directly related to depth of cover). • Geological discontinuities (orientation and properties). • Mechanical properties of the intact rock and the rock mass. • Ground water conditions and susceptibility of the rock mass to weathering processes. • Size of the excavation. • Direction of the excavation. • Purpose of the excavation. • Equipment available and skill level of work force. 58 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Many methods of support design have been developed. However, most of the existing support design methods do not consider the depth of cover or stress regime. The induced stresses are known to have considerable affect on rock mass response to mining, hence support requirements. Since most of the existing support design tools were developed at mines where the depth of cover is significantly greater than 100m, it stands to reason that they may not be applicable in shallow depth mines. Three general types of ground failures are recognised to occur at mines: failures related to stress, failures related to geologic structure, and failures related to the rock mass. As discussed in Chapter 3, the in situ and induced stresses at mining depths less than 100m are relatively low such that stress induced failures are not expected to be a significant problem. Gravity induced failures are expected to cause most of the stability problems in shallow depth mines This chapter starts off with the results from a detailed evaluation of the rock bolt support used at the Quinsam Coal Mine. Visits by the Author to several US and Canadian coal mines as well as the literature review have indicated that the trend in artificial ground support in coal mines is towards rockbolts. Rock bolts are faster to install than most other methods, they can provide active support, and do not inhibit ventilation. 5.1 ROCKBOLTS The standard rock bolts used at the Quinsam Coal Mine were 1.8m or 2.4m long rebar, 20mm diameter, grade 60, with a yield strength of 13 tonne, and an ultimate strength of 21 tonne. Steel bearing plates were used on all rock bolts. In most cases rock bolts and wire screen provided adequate roof support. Where the ground conditions were very favourable rock bolts alone were used. Where unfavourable ground conditions were encountered wooden cribs, wooden posts, and occasionally metal straps and steal arches were used. The most common support elements are shown in Photo 5.1. 59 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SrIALLOW DEPTH Photo 5.1: Ground Support Elements in General Use at The Quinsam Coal Mine The rebar rock bolts used were point anchored with approximately 0.76m of epoxy resin placed at the toe of the hole. The bolts were tensioned with an applied torque of approximately 200Nm. This torque resulted in an applied tension of between 2.5 and 4.5 tonne (based on the torque-tension relationship of 0.013 to 0.023 tonne per Nm established by the Author). The Fletcher Roof Bolters used to install the bolts (see Photo 2.2) only had the capability to drill vertical holes; as such, rock bolts were only installed in the roof. No rib bolts were installed. The anchoring horizon for the rock bolts included coal, weak siltstone, competent siltstone and occasionally sandstone. The performance of the rock bolts was studied using visual observations of failure, rock bolt load measurements, and rock bolt pull tests. 60 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 5.1.1 Visual Observations of Performance Over the course of the field study (1993 to 1997) very few rock bolts were observed to have failed within roadway excavations. The few observed failures where related to one of the following: • Rock bolts installed in fault zones (unravelling of rock around the bolt or failure above the bolt). • Rock bolts installed in wedges defined by geologic structure (bolt capacity exceeded). • Rock bolts installed in weak siltstone (unravelling of rock around the bolt). • Rock bolts installed in ground subject to weathering (unravelling of rock around the bolt). 5.1.2 Rock Bolt Pull Tests In order to ensure that the resin anchorage capacity matched the bolt strength, rock bolt pull tests were carried out throughout the mine. Random pull tests were carried out on production bolts as well as on bolts specifically installed for the purpose of pull tests. Several mechanical anchor and Split Set bolts were tested in addition to the rebar bolts. A total of 104 pull tests were carried out. Details of the test are presented in Appendix 3. Several standards are in common use to evaluate if a rock bolt pull test is acceptable or not; these standards are as follows: • Acceptable if less than 3.18mm of anchor displacement occurs at a 7.25 tonne load (Mark 1995). • Acceptable if the bond stress at failure exceeds 5 MPa. Failure is considered to have occurred when the slope of the bond stress versus anchor displacement curve drops below 0.75 MPa. per mm (British Coal, 1992). • Acceptable if the bolt can be loaded to the yield strength without sustaining unrecoverable deformation (Cullen, 1989). 61 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Random pull tests were carried out on 60 standard production bolts. To minimise the need to replace tested bolts most tests were carried out to a maximum load of 13 tonne, which was the yield load of the rebar. The results from these tests are summarised below: • The average failure load was 18 tonne (excluding failures at the nut). • The minimum failure load was 8.5 tonne (failure at the rock/resin interface). • The average load at which nut failure occurred was 15 tonne. • The minimum load at which a nut failure occurred was 9.8 tonne. In 20% of the random pull tests, failure occurred by the nut stripping, which clearly indicated a problem with the bolt manufacture. This problem was brought to the attention of the bolt supplier and the problem was rectified by changing the way that the threads were formed. Specific pull tests were carried out to analyse anchor capacity in different rock types and for different size drill holes. Several of these tests were carried out in conjunction with CANMET (Payne 1994). The tests were carried out in general conformance with the ASTM and ISRM standards. The tests monitored the deformation as well as the load applied to the bolt. Deformation due to stretching of the rebar was backed out of the calculations using the elastic modulus of the steel. Beyond the rebar yield load (13 tonne) this calculation could no longer be performed, as stretching became non elastic. As the primary interest of these tests was to establish a bond capacity, it was desirable to have the bond fail below the steel yield load. To accomplish this, the resin encapsulation length was limited to between 0.22 and 0.35m; these types of tests are often referred to as "short encapsulation" pull tests. The first short encapsulation tests were carried out to evaluate the anchor capacity in competent siltstone. The length of resin encapsulation was 0.305m. In all but one test the yield point of the steel was reached (13 tonne). This indicated that 0.305m of resin encapsulation provided sufficient anchorage to reach the roof bolt yield load. One test failed at a 3 tonne load due to improper resin mixing. 62 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Of particular interest was the long-term roof bolt performance in the weak siltstone. Twelve short encapsulation bolts were installed in an area of dry, weak siltstone, with numerous slickensides. The rock bolt installation details were as follows: • 0.508m long bolts were installed in 0.406m long holes. • The holes were drilled with a 26mm diameter cross bit. The resulting hole diameter was 27.5mm. • The resin cartridge length was 0.190m. The resin cartridge diameter was 23mm. This resulted in an encapsulation length of 0.224m. • To prevent spin out of the resin during mixing, electrical tape was wrapped around the bolt at 0.224m length. Pull tests were carried out at the following intervals: 1 day, 32 day, 65 day, and 380 day. Results for the tests conducted at 65 days are presented graphically in Figures 5.1 and 5.2. The results indicated that there was no time dependent loss of anchor capacity in weak siltstone that was dry. The results also indicated that the average resin anchor capacity is 24 tonne/m resin. The optimum encapsulation length, which is the minimum resin length required to reach the ultimate strength of the rebar, is therefore 0.877m. The pull test acceptance criteria discussed previously were met for most of these tests. Pull tests of production bolts installed in weak siltstone that was wet were also carried out. Yield failure, related to resin-rock anchor slippage, began to occur at approximately 4 tonne. The average failure load was only 10.9 tonne. These tests did not meet any of the pull test acceptance criteria. These results indicated that rock bolts installed in weak siltstone might not be acceptable if moisture is present. The following recommendations were made to deal with this problem: • Install longer bolts that anchor in competent rock beyond the weak siltstone. • Design the support based on a peak load of approximately 4 tonne, which would require that bolt spacing to be reduced. Despite its shallow depth the Quinsam Coal Mine was relatively dry. Within mine roadways water was only a problems in the vicinity of major faults. 63 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 5.2: Rock/Resin Bond Stress Versus Displacement. Short Encapsulation Pull Test Results for Bolts Installed in Weak Siltstone after 65 Days: ROCK/RESIN BOND STRESS (MPa) 3 4 ANCHOR DISP. (mm) +— TEST 7 - a - TEST 8 — T E S T 9 64 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Bolt tension measurements were carried out at the same time as many of the pull tests. Very little tension bleed off was found to occur with properly installed bolts in competent siltstone. Significant tension bleed off (up to 50%) occurred on bolts installed in weak, wet, siltstone. It was concluded that bolt tension could be used to assess the quality of the rock bolt installations. Where tension bleed off greater than 20% occurred, the bolt may be improperly installed or the anchor at the end of the hole, or at the collar, may be yielding. An anchor problem would be suspected if tension bleed off continued to occur after the bolt was re-torqued. A total of five Split Set Bolts and three mechanical anchor bolts were tested. Al l tests were carried out in competent siltstone. From these tests it was determined that the average yield load was 2.9 tonnes per metre length for the Split Set bolts, the average anchor failure load for mechanical bolts was 8.3 tonnes. The results of the tests are summarised in Appendix 3. Compared to the point anchor resin bolt the capacity of the Split Set bolts and mechanical anchor bolts is low. The use of Split Set or mechanical anchor bolts was not recommended where the roof rock consisted of siltstone. 5.1.3 Drill Hole Size The size of the drill hole, relative to the size of the bolt, is a critical factor in the quality of a resin anchor rock bolt installation. If the annulus between the drill hole and bolt is too large the resin may not be mixed properly and may be "spun away" from the anchor zone. The standard drill bit used at the Quinsam Coal Mine was a 26.2mm cross bit. Borehole callipers were used to determine the finished dimension of the drill hole. Measurements were taken in different rock types throughout the mine. The average drill hole diameters for the different rock types are given in Table 5.1. 65 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Table 5.1: Average Finished Drill Hole Diameter coal 28.0mm weak siltstone: 28.5mm hard siltstone: 27.5mm mudstone: 29.0mm Pull tests were carried out in a siltstone using both 23mm and 28mm diameter resin cartridges, and 33 and 35mm diameter drill holes. The results from the pull tests are presented in Table 5.2. The data for the 23mm cartridge in the 28mm diameter hole was found to have a normal distribution. Insufficient tests were carried out in the other catagories to comment on the distribution. The reader is referred to Appendix 3 for a complete listing of the data. For comparative purposes the results have been normalised to tonnes per metre of resin length. The results show a reduction in anchor capacity with increasing hole size. Table 5.2: Pull Tests Results For Different Drill Hole Size and Resin Cartridge Size (numbers in brackets indicate number of tests) Drill Hole Diameter (mm) Average Capacity of Rock Bolt Anchor (tonne/m of resin) 23mm cartridge 28mm cartridge 28 >37 (50) 33 37 (3) 35 4.8 (3) 25 (6) The primary cause of the decrease in anchor capacity with increasing hole size is improper mixing of the resin and/or resin being spun away from the anchor point. These problems increase as the size difference between the bolt and the drill hole increases. "Glove fingering" of the resin occurs when the resin cartridge is punctured but not properly shredded and mixed. This phenomena usually occurs where the size of hole is more than 10mm larger than the size of bolt being installed, or when the rebar is pushed either to quickly or too slowly through the resin (Ulrich et al 1991). 66 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 5.1.4 Rock Bolt Load A single load cell was placed on a rock bolt to study the load history. The bolt was installed with the load cell approximately 40m back from an advancing gob. The load cell was left in place until just before it passed into the gob, at which time it was recovered. The load history is shown in Figure 5.3. As can be seen, the load on the rock bolt did not change from the time of installation to the time of removal (just prior to passing into the gob). Since there was no increase in load on the bolt, it can be concluded that there was no deformation in the roof rock even as the gob approached. This also infers that there was no change in the induced stresses in the immediate roof. Figure 5.3: Load History on Rock Bolt Installed In Active Retreat Pillar Area 3 5 0 0 -f 3 0 0 0 - = 2 5 0 0 9 2 0 0 0 -f o "* 1 5 0 0 - -1 0 0 0 -5 0 0 installed load 3600 kg load call recovered just prior to bolt passing into gob I N S T A L L E D -+-1 0 / 2 3 / 9 4 1 0 / 2 4 / 9 4 1 0 / 2 5 / 9 4 1 0 / 2 6 / 9 4 1 0 / 2 7 / 9 4 DATE 1 0 / 2 8 / 9 4 1 1 0 / 2 9 / 9 4 67 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 5.1.5 Full Column Resin Rock Bolts Full column resin bolts have several advantages over point anchor bolts: • Filling the hole with resin reduces moisture ingress and deterioration of the rock. • The resin provides anchorage along the full length of the bolt including the collar rock, this helps prevent unravelling around the collar. • If the rock around the collar deteriorates a full column rock bolt remains effective. With point anchor bolts once the rock around the collar is lost the bolt is essentially ineffective. • Full column resin increases the bolt stiffness and improves shear resistance. At the Quinsam Coal Mine, deterioration of rock around the collar of the bolts was found to be a significant problem in the weak siltstone. Installation trials of full column resin bolts were carried out. During the trials, 30% of the bolt installations failed due to bending of the rebar as it was being inserted into the hole. It is common industry practice to install 1.8m long, 20mm diameter bolts into 27mm diameter holes, with full resin encapsulation. Resin manufacturers and rock bolt manufacturers were brought in to examine the installation problem at the Quinsam Coal Mine. They concluded that the Fletcher Roof Bolters were not capable of drilling a straight enough hole or pushing the rebar through the resin in a straight enough line to assure trouble free full column resin installation. 5.2 EXISTING SUPPORT DESIGN METHODS An extensive literature review revealed nine support design methods commonly used in mining and civil applications. The methods can be broadly divided into rock mass support design methods, rock structure support design methods, and stress support methods. The rock mass support methods include the simple rules of thumb through to rock mass classification methods. All are essentially empirical in nature. The design recommendations 68 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH from these methods are usually somewhat conservative to account for unknown conditions and anomalous behaviour. This approach is analogous to the application of a factor of safety in the analytical methods. The rock mass classification methods are the most sophisticated of the empirical design methods. These methods seek to match the support requirements to the quality of the rock mass and site conditions. Rock mass classification quantifies the parameters considered to affect support design thus making it possible to use the methods in a range of geological and environmental conditions. Most of the rock mass classification systems recognise the significance of geologic structure to rock mass behaviour; the input parameters typically include discontinuity properties such as spacing and strength. These methods do not consider discrete blocks or wedges formed by geologic structure. Most of the methods only consider stress in general terms. The analytical methods match the support requirements to the expected mode of failure. Typical modes of failure include gravity driven rock structure failures and stress driven failures. Prior to selecting a design method, it is necessary to determine the expected mode of failure. Numerical methods are an extension of the analytical methods where numerical simulations are used to analyse mining induced stresses and deformations. Rock failure criteria are then applied to the results of these analyses. Ground support requirements are typically based on the depth and extent of the predicted failure zones around the excavation. Numerical methods are primarily used where failure is being driven by stresses. As failure in shallow mines is seldom due to high stresses, numerical analysis was not considered a useful tool for support design in this study. 5.2.1 Rock Mass Support Design Methods 5.2.1.1 US Corps of Engineers The US Corps of Engineers (1980) developed empirical rules for support design based on analysis of support in underground chambers, tunnels and shafts, no mine excavations were 69 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH included. The rock quality is assumed to be "average", as such, the recommended support must be adjusted if the rock mass is of good or poor quality. The rules are presented in Table 5.3. Table 5.3: Support Design Recommendations of the US Corps of Engineers (1980) Parameter Empirical rule Minimum length and maximum spacing Minimum length Greatest of: (a) 2 x bolt spacing (b) 3 x thickness of critical and potentially unstable rock blocks (Note 1) (c) For elements above the springline; spans <6m: 0.5 x span spans between 18 and 30m: 0.25 x span spans between 6 and 18m: interpolate between 3 and 4.5m (d) For elements below the springline: height <18m: as (c) above height >18m: 0.2 x height Maximum spacing Least of: (a) 0.5 x bolt length (b) 1.5 x width of critical and potentially unstable rock blocks (Note 1) (c) 2.0m (Note 2) Minimum spacing 0.9 to 1.2m Minimum average confining pressure Minimum average confining pressure at yield point of elements (Note 3) Greatest of: (a) Above springline: either pressure = vertical rock load of 0.2 x opening width or 40 k N / M 2 (b) Below springline: either pressure = vertical rock load of 0.1 x opening height of 40 kN/m 2 (c) At intersections: 2 x confining pressure determined above (Note 4) 1. Where joint spacing is close and span relatively large, the superposition of two reinforcement patterns may be appropriate (e.g. long heavy elements on wide centres to support the span, and shorter, lighter bolts on closer centres to stabilise the surface against ravelling). 2. Greater spacing than 2.0m makes attachment of surface support elements (e.g. weldmesh or chain link mesh) difficult. 3. Assuming the elements behave in a ductile manner. 4. This reinforcement should be installed from the first opening excavated prior to forming the intersection. Stress concentrations are generally higher at intersections and rock blocks are free to move toward both openings. 5.2.1.2 Farmer and Shelton Farmer and Shelton (1980) developed rules for support in rock masses having a maximum of three discontinuity sets with clean tight interfaces. The rules are based on the work of several others. The rules are summarised in Table 5.4. 70 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Table 5.4: Support Design Recommendations of Farmer and Shelton; after Farmer and Shelton(1980) t/5 H Z u S 2 o u SO '_ c " - V w — rz ta) I u "3 * -1. z « o O — -J J J a "5 5 S c £ .2 u ,= « IT - r i = •a - s •= a .= u. = « J~ w . . G u ' J .= <A ^  C ~ C o - w (A 3 u ~ c .= « o - .a - « 5 C — t l .2 -.A <u f- £ -a = > Q. . u ~ *> S: >> > T3 2 3 C O 2 -1 £ j2 u c ;s £ 3 SO -s .52 •- c « O . " ao en : 2 "c — <u BO . c O0 = — an _ * c 2 o i j 3 C O O ..: "3 <u - SJ 73 73 5 »> «J u <i s = « * \G c a " - N ~ ey 30 JS = .a ™ a =* C so ~ c 73 -.2 "5 73 _ ;r = = ™ C3 2 ' *> * = = 2?T3 ^ * U ' = 3 5 0 U xi 5 c 73 (A ^ . 'sr. 73 5 £ -C „ t« - — e» <— *i — 2 £ £ «= O C w O « OQ C C u -= i t , «3 C 2 •z c > u 5 w X C U r-— cn - c j . 2 u i 5 « .2 m ^  ^ I"* s § . ! . z •- 4> U a. v; — G ^-- O t> u 2- w .= « «• 5 = u J . 2 « ^ «! 3 § S m u «. r C ^ C J£ C x; — 1A o •= .2, o . Tj • , V / / ; < y < v V A 5 ' . / . ' / A W v ' W v Z a U a H - J O CQ si . c 3 U c o p.-.2 a. 5 "O (A ~ C ' ca II II o a = * tn — C S (9 "(A o o X T I II «3 so so (A • vei y hoi lary I- <u c primai secoiul V c <c J" o c >^ re G u c ao c ° d = « II 2 OQ _j ^ <^ T « d o <-H II II II 'I 7 - <NJ ~ _ ty) V . — U. 3 O Z 06 H U Z CQ O S c/j H ^ — [J z a 35 o < E > ~~ < Z U < T I o •G = s 5 — J = vi 2 c - s •a 'C *> o .= s o O r-) o VI <^  5 % •a VI A 71 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 5.2.1.3 French Matrix System Historical experience at coal mines in France led to the development of rules which relate in situ conditions to support requirements. This method is one of the few specifically developed for use at coal mines. It is also the only method that specifically considers the stress induced failures. To derive the required support pattern it is necessary to ascertain the existing conditions and work through a matrix developed by Newson (1987). The matrix and accompanying notes are presented in Table 5.5. 5.2.1.4 "Q" Classification System The Norwegian Geotechnical Institute "Q" rock mass classification system was developed by Barton et al (1974). Since its inception it has gained wide acceptance in quantifying rock mass quality, and relating quality to support requirements. The Q value is determined as follows: Q = (RQD/Jn) x (Jr/Ja) x (Jw/SRF) where: RQD = rock quality designation (Deere 1964) Jn = parameter related to number of joint sets Jr = parameter related to roughness, and of joints Ja = parameter related to alteration, and openness of joint sets Jw = parameter related to water SRF= parameter related to stress and rock strength Details of the determination of the individual parameters can be found in most rock mechanics texts such as Hoek et al (1995), and Bieniawski (1989). This classification system is one of the few that considers the relative strength of the rock compared to the induced stress. 72 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Table 5.5: Support Design Recommendations of French Matrix System; after Newson (1987) Type of rock and Results of Effects Bolting Parameters strata stress field on time Anchorage Length Diameter Density Lagging Deep and superficial stability Stabilised deformation Support not necessary Delayed Light Short Small Low Light deformation point shotcrete Homogeneous or Deep Stabilised Any Short Small Medium Light slightly fractured stability and superficial instability deformation Delayed deformation Light or strong point Short Small Medium Light continuous Deep and Stabilised deformation Strong point or full column Medium to long Medium High Heavy superficial instability Delayed deformation Strong point Medium Medium High Heavy Deep and superficial stability Stabilised deformation Delayed deformation Light point Short Support Small not necessary Low Light shotcrete Stratified Deep Stabilised Any Medium Small Medium Light and hardly vertically stability and superficial instability deformation Delayed deformation Light or strong point Medium Medium Medium Light continuous Deep and Stabilised deformation Strong point or full column Long Medium High Heavy superficial instability Delayed deformation Strong point Long Large High Heavy Stabilised Point or full Short Small Medium Light Deep and deformation column continuous superficial stability Delayed deformation Point or full column Short Small Medium Light continuous Irregular Deep Stabilised Full column Medium Medium Medium Light lenticular stability deformation or fractured and in several directions superficial instability Delayed deformation Strong point or full column Medium Medium Medium Heavy Stabilised Full column Long Large High Heavy Deep and deformation superficial instability Delayed deformation Strong point or full column Long Large High Heavy 1. Rock Type : This is a simplistic classification designed to characterise the strata immediately surrounding the roadway. Essentially, a judgement is made on whether the rock is solid or liable to slab. 2. Stress Field in the surrounding Strata: This information should be obtained from detailed field investigations. Two distinct categories have been defined: superficial and deep. The superficial category concerns the strata behaviour to a depth of 1 m around the roadway. The deep category concerns the overall deep-seated rock mass behaviour surrounding the roadway. 3. Time Effects: Two simple parameters have been categorised: stable deformation and delayed deformation. These parameters refer to the rheological properties of the strata and mining factors such as adjacent workings that might cause time dependent deformation. Detailed field investigations should be done to determine these properties. 4. Anchor Type: This column is self-explanatory, although in France there are two typed of point anchors: 1) where the bolt fails before the bond (strong) and 2) where the bond normally fails first (weak). 5. Length: short bolts are about 1 m long. Medium bolts are about one-third the width of the structure, normally 1.8 to 2.4 m. Long Bolts are one-third to one-half of the structure width. 6. Diameter: Small diameters are less than 20 mm. Medium is between 20 and 25 mm, while anything greater than 25 mm is considered large. 7. Density: Low density is less than 0.75 bolts/m2, minimum is 0.75 to 1.25 bolts/m2, and high is over 1.25 bolts/m2. 8. Lagging Light discontinuous lagging consists of steel strips or beams. Light continuous lagging consists of welded wire mesh. Heavy lagging consists of heavy duty wire mesh either with or without shotcrete. 73 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH The Q value has been related to support by both graphical methods and by empirical relationships. Both these methods make use of a parameter called the Excavation Support Ratio (ESR), which is essentially a risk factor related to the type of excavation being developed; for permanent mine openings the ESR is taken as 1.6, for temporary mine openings it is taken as 3 to 5. A graphical method of support design is shown in Figure 5.5. The empirical equations, proposed by Barton (1984) for estimating the length and spacing of rock bolts are given below. Bolt length L (m) = 2 + 0.15 B/ESR (5.1) Bolt spacing S (m) = V(C x lO"3) / P (5.2) where: C = bolt capacity (usually taken as yield strength) P = support pressure = 2 V(JnxQ" 0 3 3 3)/(3 Jr) B = span of excavation (m) Figure 5.4: Support Design Guidelines Using the "Q" Classification; After Grimstad and Barton (1993) REINFORCEMENT CATEGORIES 1) Unsupported 2) Spot bolting 3) Systematic bolting 4) Systematic bolting with 40-100 mm unreinforced shotcrete 5) Fibre reinforced shotcrete, 50 - 90 mm, and bolting 6) Fibre reinforced shotcrete, 90- 120 mm, and bolting 7) Fibre reinforced shotcrete, 120- 150 mm, and bolting 8) Fibre reinforced shotcrete, > 150 mm, with reinforced ribs of shotcrete and bolting 9) Cast concrete lining 74 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 5.2.1.5 it RMR" Classification System The rock mass rating (RMR) classification system was developed by the South African Council for Scientific and Industrial Research in 1974 (Bieniawski 1974). Since this time, it has gained wide acceptance in quantifying rock mass quality, and relating quality to support requirements. The RMR system assigns values to five basic rock properties as listed below: • rock strength • rock mass designation (RQD) • fracture spacing • discontinuity condition • groundwater condition The values assigned to each property are then added to arrive at a preliminary RMR value. Adjustments may then be made for discontinuity orientation. Details for determination of the input values can be found in many texts such as Hoek et al (1995), and Bieniawski (1989). Several researchers have developed support guidelines based on the RMR classification including Merritt (1972), Laubscher and Taylor (1976), and Unal (1983). Only Unal's work will be discussed further as it was carried out exclusively for coal mining. Merritt's work focused on tunnels while Laubscher and Taylor's was carried out for hard rock mines. Unal (1983) carried out an extensive study of roof support at coal mines in the US. From this work he developed the following equations for support design. These equations have also been used to generate design charts. The design charts indicate that as rock quality deteriorates and/or span increases supplemental support in the form of posts and/or planks must be used. For point anchor bolts where: Bolt spacing: S (ft) = V (C ) / (fos x y x ht) Bolt length: L (ft) = ht/2 y = unit weight of rock (lbs/ft) C = bolt capacity -usually taken as yield strength (lb) (5.3) (5.4) 75 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH h t = rock load height (ft) = (100-RMR)xB/100 B = span (ft) fos = factor of safety (recommended to be 1.5) For full column resin bolts Bolt Spacing: S = V ( C ) / (fos x y x ht) (5.5) Bolt Length: L = V ( B 2 x ht) / 300 (5.6) 5.2.1.6 CMRS Classification System Dhar et al (1992) developed the coal measures roof support (CMRS) geomechanical classification system. The CMRS is essentially a modification of the RMR system to account for specific conditions found in coal mines that were believed to be controlling factors in roof stability. The parameters analysed included: layer thickness, structural features, weatherability, rock strength, water seepage. Statistical analysis was used to establish the relative weighting for each parameter in determining the CMRS value as it relates to support design. 5.2.2 Analytical Support Design Methods 5.2.2.1 Beam Building and Arching The creation of stable beams and arches are similar concepts. The objective is to create a reinforced zone that is capable of supporting itself as well as the weight of the overlying material. The extent of the rock material that must be supported is a function of many variables. In horizontally bedded deposits with no cross joints Fayol (1885, reported in Brady and Brown, 1985), demonstrated that at a certain height above an excavation, the gravitational load of the upper beams is transferred to the abutments rather than adding to the load on the lower 76 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH beams. The load transfer, described as arching, is the result of the mobilisation of frictional resistance between beds. Bedding planes typically have a low, or no, tensile strength and a low shear strength. Under gravity loading the bedding planes may separate and sag. The different layers will slide against one another, similar to a deck of cards that is bent. The capacity to share load with upper units is lost once sagging and slipping occurs. If no cross cutting structures are present the stability of the immediate roof can be analysed by beam theory or plate theory; however, cross cutting structures are present in most mines such that the standard beam and plate theories cannot be applied. In order to account for the presence of cross cutting joints, Evans (1941, reported in Brady and Brown,1985) applied the concept of the Voussoir arch from masonry construction. The method developed by Evans has been modified by several researchers, including Brady and Brown (1985) and Hutchinson and Diederichs (1996). There are three possible modes of failure predicted by this theory: • Shear at the abutments. • Crushing at the central hinge points. • Buckling or "snap through" at the central hinge points. The Voussoir arch problem geometry is shown in Figure 5.5. Equations for solving the problem, as well as the many limitations and assumptions required, are presented in the above referenced texts. The equations are statically indeterminate; an iterative process is used to determine an approximate solution. The solutions may be used to assess the initial stability, as well as to design bolting patterns. The installation of rock bolts will reduce the effective beam span as well as increase the beam thickness. The equation solutions are very sensitive to rock modulus rock strength values. As such, a factor of safety of between 1.5 and 3 is recommended (Hutchinson and Diederichs 1996). 77 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 5.5: Problem Geometry for Voussoir Arch Stability Analysis; after Hutchinson and Diederichs (1996) If tensioned rock bolts are installed a normal stress (o"n) will be created that increase the shear strength across the bedding planes. The shear strength (xs) of the bedding planes can be described by the Mohr-Coulomb strength criteria: xs = c + o-nTan(<j) + i) (5.7) where: c = cohesive strength of the bedding plane or joint surface <) = angle of friction of the bedding plane or joint surface i = dilation angle due to roughness of the bedding plane or joint surface (j) + i = effective friction angle a n = normal stress across the bedding plane or joint surface The Mohr-Coulomb strength criterion can be extended to the entire rock mass. Fictitious shear planes are assumed to exist within the rock mass. A rock failure envelope can then be defined in terms of the principal stresses cr\ and 0 -3 . Arching theory can be applied in a general form to rock masses.. The installed rock bolts are considered to create a zone of compressed reinforced rock around the excavation. The reinforced zone acts as a structural member capable of supporting itself as well as the weight of the loose rock above. This concept is illustrated in Figure 5.6. 78 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 5.6: Formation of a Reinforced Rock Arch The Norwegian Institute for Rock Blasting (Stillborg 1986) has proposed the following empirical formulas for support in the central section of such an arch: Bolt length L (m) = 1.4 + 0.184 x B (5.8) Bolt spacing S (m) = 0.5 x L. (maximum 3 x the spacing between joints) (5.9) In order for a stable arch to be created it is important that all "key" blocks be supported. The key blocks are those that control the stability of the other blocks surrounding them, much like the key blocks in a masonry arch. Figure 5.6 illustrates this concept. If the key blocks are not supported progressive unravelling and subsequent loss of the reinforced arch may occur. Support for key blocks should be sufficient to hold the dead weight load of the blocks. Where pattern bolting is used the pattern may need to be adjusted or additional "spot" bolts installed to support the key blocks. 79 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH t Daws (1987) has shown that, using a rock mass confinement approach, the load to be supported by the bolting system (Pb) can be determined from the following equation: P b (tonnes/m) = (q x C x t) / S2 (5.10) where: q = tan (45 + cp/2): this term considers the increase in strength of the reinforced arch due to the confinement provided by the rockbolts. d> = angle internal friction of rock t = thickness of reinforced roof beam = the smaller of: 0.66 x L or tensioned length of bolt S = bolt spacing (m) L = bolt length (m) C = bolt capacity -usually taken as yield strength (tonnes) The load on the support system (P ) can be estimated from the following expression developed by Unal (1983): P = h t x B x y (5.11) where: h t = rock load height = (lOO-RMR)xB/ 100 The initial bolting pattern is then determined by equating equations 5.10 to 5.11. Daws recommends using a factor of safety of 1.6 with the calculation. 5.2.2.2 Dead Weight Load Dead weight load design is applicable to situations where large blocks or wedges may fall under gravity load into the excavation, or where a layer of weak rock is to be anchored to more competent rock. Support is designed with sufficient capacity to hold up the "dead weight" of the unstable blocks, wedges, or layers. Where a weak strata is overlain by a more competent strata, the "dead weight" of the weaker strata can be supported on rock bolts that anchor into the overlying competent strata, as 80 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH illustrated in Figure 5.7. For horizontal strata the bolt spacing S in metres can be determined from the following equation: C / (fos x y x h) (5.12) where: C Y h = bolt capacity -usually taken as yield strength (tonnes) = unit weight of the rock (tonne/m3) = thickness of weak or unstable layer to be supported (m) The rock bolts must be of sufficient length such that the anchor point is fully within the competent strata and the fos should not be less than 1.5. Figure 5.7: Anchoring Weak Strata to Strong Strata Competent strata. Geologic discontinuities may form discrete blocks or wedges that can fall into an excavation. The number of bolts (N) required to support a wedge under gravity may be determined from the following general equation: N = W x f o s / C (5.13) where: C = bolt capacity - usually taken as yield strength (tonnes) W = weight of wedge or block (tonnes) 81 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH The rock bolts must be of sufficient length such that the anchor point is fully within the competent strata above the unstable wedge and the fos should not be less than 1.5. For the simple 2 dimensional case, i.e. wedges defined by just two joint sets, the number of bolts per metre length of the wedge can be determined from the following equation: N = V2 h w x B w x fos / C (5.14) where: h w = maximum height of the wedge (m) B w = width of the wedge (m) y = unit weight of the rock (tonne/m ) In some cases it is desirable to include the shear strength of a discontinuity in the analysis of stability. This is usually done using the Mohr-Coulomb strength criteria given in equation 5.7. The factor of safety of a wedge then becomes: fos = (N x C) + (c + o-^ x Tan (ty + i) (5.15) W + T where: x = the shear stress, or stress parallel to the joint surface. In most cases, the vertical stress in the vicinity of an excavation roof is negligible such that the normal stress a n and the shear stress x may be determined from the horizontal stress (ah) with the following equations: a n= c>hcos0 (5.16) x = o~hsin0 (5.17) where: 6 = the angle of the joint surface to the horizontal As the angle of the joint plane increases, the shear strength increases and the shear stress decreases. It can be shown that, provided there is a moderate horizontal stress, a wedge will be stable without any support if (<|> + i) + 0 >90 degrees. In this case the discontinuity shear strength will be greater than the shear stress. The excess strength is used to support the 82 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH gravitational load of the wedge. As demonstrated in Chapter 3 the horizontal stresses over the centre of an excavation decrease with increasing span and decreasing depth of cover. A-decrease in the horizontal stress will result in a decrease in the shear strength of the joint surface. Where wedges are defined by three or more joint sets the visualisation and determination of the wedge weight becomes more difficult. An example of a three dimensional wedge is shown in Figure 5.8. The size and weight of a wedge can be determined by graphical methods such as those described by Hoek and Brown (1980) or with a computer program such as UNWEDGE (Carvalho et al 1994). The UNWEDGE program will evaluate the shear strength of the geologic structures by considering a user defined cohesion and angle of friction. The program also allows the user to simulate rock bolt installation and determine a supported factor of safety against failure. Figure 5.8: Example of a Wedge Defined by Three Joints above a Mine Excavation 83 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 5.3 COMPARISON OF THE EXISTING SUPPORT DESIGN METHODS TO EXPERIENCE AT THE QUINSAM COAL MINE Nine of the rock mass support design methods discussed above were used to determine the recommended support for a 6m wide excavation in siltstone at a depth of 100m. The rock properties used in the design were those for competent siltstone as found at the Quinsam Coal Mine. The results are tabulated in Table 5.6. Table 5.6: Support Design for 6m Wide Excavation in Competent Siltstone DESIGN M E T H O D B O L T T Y P E B O L T L E N G T H (m) B O L T SPACING (m) B O L T C A P A C I T Y (tonne) OTHER SUPPORT C O M M E N T S "Q", chart method (Grimstad and Barton 1993) 1.8 1.5 shotcrete "Q", equation method (Barton etal 1977) 2.3 0.7 Farmer and Shelton (1980) 2.3 1.1 - 1.5 F R E N C H M A T R I X , (Newson 1987) 1.8-2.4 0.8-1.3 screen and strap US CORP. E N G . ( 1980) 3 0.9-1.5 Unal(1983) resin - point anchor 1.8 1.5 7 posts at 2.2m spacing fos=1.0 Rock confinement, (Daws 1987) resin - point anchor 1.8 0.8 13 fos = 1.0 C M R S , (Dhar et al 1992) rope dowel 1.5 1.0 lagging and trusses Arch, (NGI - Stillborg 1986) 2.5 1.3 support used at the Quinsam Coal Mine resin - point anchor 1.8 1.2 13 screen and straps as required Almost all the existing support design methods overestimated the support levels that were found to work successfully at depths less than 100m at the Quinsam Coal Mine. In terms of providing a safe excavation it appears that any of the methods would be acceptable. However, using a method that overestimates ground support will unnecessarily increase the production costs. The French Matrix method was the only method that recommended ground support levels that were consistent with the levels used at the Quinsam Coal Mine. 84 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH The main shortcomings of most of the existing rock mass support design tools are that they do not consider the stress regime and/or they were not developed for use in coal mines. For example: • Unal's method was developed exclusively for coal mines; however, the data set used to develop the method was obtained primarily from mines operating at depths greater than 100m, and no account is taken of mining induced stresses. • The Q, US Corp of Engineers, and Farmer and Shelton methods were developed for use in civil applications. 5.4 CONCLUSIONS Excavation stability at shallow depth retreat pillar coal mines should primarily consider geologic structure and the rock mass. Support design should be based on both an empirical rock mass method as well as an analytical geologic structure method. This research has shown that most of the rock mass support design methods overestimate the support requirements at shallow depth; the one exception is the French Matrix method. The UNWEDGE computer program was found to be suitable for determination of support for geologic structure. Where the depth of cover is less than 100m the induced stresses are expected to be low, and stress induced failures are not expected to be a significant problem. However, it must be recognised that the affects of stress are proportional to the strength of the rock mass; the potential for stress related instability is greater in weaker rock, i.e. mudstone, weak siltstone, and rocks weakened by weathering processes. As the depth of cover decreases the stresses over the top of an excavation will decrease and may become tensile. This situation favours structurally controlled gravitational type failures such as bed separation, beam bending, unravelling, and block type failures. Gravity induced failures are expected to be the most prevalent type of failure at shallow depth coal mines where well defined geologic structure is present. The UNWEDGE computer program 85 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH developed by Carvalho et al (1994) was determined to be an excellent tool for visualisation of wedges and determination of support requirements to secure the wedges. This research has shown that most of the existing rock mass support design methods overestimate support requirements at shallow depth coal mines. There are three reasons for this: • The methods do not generally consider depth of cover or stress. • Most of the methods were developed for uses other than at coal mines. • Most of the methods were developed using data from excavations greater than 100m. Only the French Matrix Method (Newson 1987) was found to match the support levels in use at the Quinsam Coal Mine where the depth was less than 100m. This is the only rock mass support design method that was developed specifically for use in coal mines that also considers the mining induced stresses. Support design based on most of the other rock mass methods evaluated should provide a safe excavation but not a cost effective excavation. Rock bolts consisting of point anchor rebar were found to be a suitable means of ground control in competent siltstone. In siltstone that is dry the average resin bond capacity was determined to be 24 tonne/m. If 20mm diameter grade 60 rebar is used the optimum resin encapsulation length is 0.88m; this is the resin length required to hold the ultimate strength of the rebar (21 tonne). In weak siltstone that is wet the resin bond capacity was found to be less than 12 tonne/m. Yield at the resin-rock interface began at loads as little as 4 tonne. The ability of resin anchored rebar to provide adequate anchorage in weak wet siltstone is questionable. Where weak siltstone is present the bolt length should be increased to anchor into competent rock; if this is not possible and there is a potential for the rock to become wet the support design should utilise a design load of 4 tonnes per bolt. Consideration must also be given to the problem of unravelling and weathering type failures that commonly occur in weaker rocks. 86 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH This research has shown that rock bolt tension can be used for quality assurance testing at the Quinsam Coal Mine. Where tension bleed off exceeds 20% either there is anchor slippage or the bolt was not installed properly. Bolt tension and load were found to remain essentially unchanged throughout the life of stable excavations. This finding is consistent with the results of the convergence study reported in Chapter 4. 87 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH CHAPTER 6 PILLAR DESIGN 6.0 INTRODUCTION Pillars are primarily used at coal mines to maintain stability of the underground excavations, and to protect surface facilities from subsidence damage. Secondary uses of pillars include control of the mine environment such as ventilation and water. The proper design of pillars is essential to underground coal mining. Undersized pillars may result in pillars being unable to perform their intended duty; this may have both economic and safety implications. Oversized pillars may unnecessarily sterilise mineral reserves. Pillar design is generally carried out by matching the pillar strength to the expected load, as defined by the following equation: forces resisting pillar failure (strength) = forces driving pillar failure (load) A factor of safety (fos) of between 1.2 and 2.0 is usually applied to this equation. A literature survey indicated that the forces resisting pillar failure (strength) are dependent on many factors including: • geologic structure • time • in situ material properties • environment (temperature, moisture) • pillar geometry (size and shape) • pillar/roof and pillar/floor country rock interaction The forces driving pillar failure (load) are dependent on: • in situ stress (directly related to depth) • mining induced stress • pillar geometry and orientation • seam orientation 88 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH The fos is used to compensate for uncertainty in the input parameters. In theory a fos greater than 1 represents a stable condition. A fos less than 1 represents an unstable condition. In practice, the fos is increased as the level of confidence in the input parameters is reduced, or as the importance of the need for stability is increased. Stress and depth are not variables that control pillar strength; however, they are the primary factor determining pillar load. Stress is well accounted for in the existing pillar design tools. As such it is expected that the existing design tools should be suitable for use in shallow depth mines. 6.1 LITERATURE REVIEW 6.1.1 Pillar Stress As mining occurs, the pre-mining stresses are re-distributed. This redistribution results in an increase in pillar load. The stress distribution is typically greatest at the edge of a pillar, decreasing towards the centre; this is schematically illustrated in Figure 6.1 Figure 6.1: Mining Induced Stress Distribution; after Brady and Brown (1985) post-mining post-mining pillar abutment stress stress distribution distribution 89 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH There are two common methods for determining the load on pillars: numerical simulations, and the tributary area theory. The tributary area theory assumes that each pillar will support its share of the load that must be redistributed after an excavation is created. The numerical simulations use computational methods to determine the stress distribution in the pillars. Both methods require prior knowledge of the in situ stresses. Numerical methods are capable of accounting for such factors as irregular pillar shapes, irregular pillar layouts, and inclined seams. Many types of numerical models are now available for the prediction of the stress distribution in pillars. Brady and Brown (1985) consider that the tributary area theory is only applicable to coal deposits with a regular pillar arrangement that extends over an area greater than the depth of cover. Where the mining geometry or pillar layout is irregular or the area being analysed is small, the tributary area theory should not be used. The tributary area theory assumes that mining induced stresses are distributed equally to the surrounding pillars as illustrated in Figure 6.2. The tributary area equation for horizontal seams, with uniaxial loading (vertical loading only) takes the form: The vertical pre-mining stress is usually considered to be equivalent to the weight of the rock above the excavation, as determined by equation 3.1. a p = rjv (w + B)(l +B) / (w x 1) (6.1) where: o v B w vertical pillar stress (average) pre-mining vertical stress roadway width length of pillar side width of pillar side ( J v = y g H (3.1) 90 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 6.2: Tributary Area Theory; Redistribution of Stresses to Pillars Section A-A' Insltu (pre-nlnlng) s t r e s s distribution. Ver t ica l s t r e s s e s are evenly distributed. 1 1 1 1 1 1 1 1 ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ Tributary area f o r pillar A A ' t V7, Section A-A ' Mining s t r e s s distribution. All vert ical s t r e s s e s are within tr lbuary area f o r pillar A are now carr ied by pillar A. Trlbuary a rea f o r pillar A The distribution and magnitude of induced stresses can change significantly in inclined seams; high compressive and shear stresses will occur in the roof on the up dip side, and in the floor on the down dip side. Pariseau (1982), and Hedley and Grant (1972) have modified the basic tributary area formulae to account for the effects of horizontal stresses on pillars in a dipping seam; their respective equation for a p are: 91 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH o-p = q v (d + k) + (\ - k) cos (2 a)V2 (6.2) (1-R) o~p = CTv cos2 oc + CTh sin2 a (6.3) (1-R) where: k = ratio of horizontal to vertical stress (normal to pillar surface) a = seam dip R = extraction ratio = fw + B¥l + B ) - w l (w + B)(l + B) Pillars adjacent to an active gob are subject to additional "abutment stresses" that are much higher than predicted by the tributary area theory. Several researchers have proposed methods to calculate this additional stress. Szwilski (1982) proposed the following formula to predict the additional stress in chain pillars. The formulae considers the theoretical cantilevering of the immediate roof over the area being mined: CTr = CTv (1 + S) CEFW + 2w + 3S) (6.4) 2 w l where: S = spacing of chain pillars (m) EFW = extraction front width (usually the width of the mining panel) Whittaker and Singh (1979), and Mark (1990) proposed that the additional load is due to a triangular block defined by the abutment angle ((3), as illustrated in Figure 6.3. The additional load is referred to as the abutment load (LA). Chase and Mark (1993) developed equations 6.5 and 6.6 to predict the abutment load. The equations are based on the work by Whittaker and Singh (1979) and Mark (1990). 1. When the extent of the gob (GL) is greater than 2(H x tan(3) LA = 0 . 5 x y x H 2 (tanp)xEFW (6.5) 2. When GL is less than 2 (H x tan p) L A - y (0.5 x H x GL - GL 2 / 8 tan p) x EFW (6.6) 92 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 6.3: Abutment Load on Pillars at the Edge of a Gob; after Mark et al (1994) KEY AMZ Active mining zone B Abutment angle EFW Extraction front width GL Mined out area H Depth of cover LA Abutment load LD Development load Research by Mark (1990) indicated that pillars within a distance of 9.3 VH from the edge of the gob carry all the additional abutment load; 90% of the load is carried by pillars within a distance of 5VH. The distance 5VH has been defined by Chase and Mark (1993) as the active mining zone (AMZ) as shown in figure 6.3. The vertical stress distribution within the AMZ is illustrated in Figure 6.4. The vertical stress (0"V>) at any point within the AMZ is determined by the following equation: OV = (3LA / 9.3VH) x (9.3VH - X) 2 (6.7) where: X = distance from the edge of gob to point of analysis 93 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 6.4: Distribution of Abutment Stresses at the Edge of the Gob; after Mark et al (1994) Abutment stress Pillar Pillar Pillar Pillar 1 2 3 4 5.0 VrT——«-| 9.3 VrT J 6.1.2 Pillar Strength Methods to determine pillar strength are divided into two broad categories: empirical and analytical. Empirical methods were developed from experience that relates the pillar strength to the rock strength and pillar geometry (height, length, width). Empirical methods are widely accepted in the mining industry. The analytical methods employ failure criteria, such as the Mohr-Coulomb or Hoek Brown. The majority of the empirical formulas take the form: Sp = K x (A + B x (w7hp b)) (6.8) where: Sp = pillar strength K = coal strength parameter that accounts for size effect w = pillar width hp = pillar height A, B, a, b = empirical constants related to geomechanical properties. A summary of the empirical constants (A, B, a, b) used in commonly applied pillar strength formulas are given in Table 6.1. 94 G E O T E C H N I C A L S T U D I E S O F R E T R E A T P I L L A R C O A L M I N I N G A T S H A L L O W D E P T H Table 6.1: Empirical Values Used in Pillar Strength Formulas Method or Author A B a b w/h D range Bunting (1911) 0.7 0.3 1.0 1.0 0.5 to 1.0 Obert and Duvall (1967) 0.778 0.222 1.0 1.0 0.5 to 2 van Heerden (1974) 0.704 0.296 1.0 1.0 1.14to3.4 Bieniawski(1981) 0.64 0.36 1.0 1.0 1.0 to 3.1 Sorenson, Pariseau (1978) 0.963 0.307 1.0 1.0 0.5 to 2.0 Steart(1954) 0 1.0 0.5 1.0 Holland (1964) 0 1.0 0.5 0.5 Greenwald et al. (1939) 0 1.0 0.5 0.833 Salamon and Monro (1967) 0 1.0 0.46 0.66 Sheorey et al. (1987) 0 1.0 0.5 0.86 less than 4 Griffith (1912) variable 1 0.5 Gaddy (1956) 0 variable 1 0.5 Evans and Pomeroy (1966) variable -0.17 0 Skelly(1977) .78 .22 1.0 1.0 Hustrulid (1976) 0 variable 1 0.5 Zern(1928) 0 1.0 0.5 0.5 Equation 6.8 is only applicable to square pillars; however, rectangular pillars are known to be stronger than square pillars due to increased confinement in the length wise dimension. Several methods have been proposed to account for this increase in strength. Sheorey and Singh (1974) proposed that the width term should be based on the average dimension of the two sides. Wagner (1980) proposed an effective width term as follows: w e = 4 Ap/r (6.9) where: w e = effective pillar width (m) to a maximum of 2w A p = cross section area of pillar (m2) r = pillar circumference (m) Salamon and Oravecz (1976) proposed the following effective width term: w e = ( w x l ) 0 5 (6.10) 95 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH The empirical pillar strength formulas do not explicitly consider the stress distribution; however, they do imply a stress gradient with the w/h term (Mark and Iannacchione 1992). These authors mathematically derived an implied stress gradient for Bieniawski's square pillar formula as follows: where: CTp' = pillar stress at analysis point Si = in situ coal strength d = distance from pillar edge to point of analysis. This equation was then modified to account for rectangular pillars by integrating over the area of a rectangular pillar and dividing by the area. The resulting rectangular pillar strength formula is as follows: No upper limit is indicated. The formula predicts that a pillar that is twice as long as wide will be 10% to 20% stronger than a square pillar, depending on the w:hp ratio. The strength of coal has been shown to decrease as the size of the specimen tested increases (Bieniawski 1968, Evans 1961, Gaddy 1956). This phenomenon is called the size effect and is attributed to the fact that larger specimens will contain more geological flaws than smaller specimens. This accounts for the significant difference between the uniaxial compressive strength (UCS) determined in the laboratory and the in situ strength of the rock. The in situ strength has been found to be as low as 20% of the laboratory UCS value. The strength of a pillar will decrease with increasing size up to a critical size where the mean density of geologic discontinuities is reached. Bieniawski (1968) found that the critical cubic specimen size in South African coal mines was 1.5m. Bieniawski (1981) notes that the size effect is more pronounced in the case of soft fissured rock than in hard intact rock, and tends to be more pronounced in tension than in compression. The parameter Si (in situ coal strength) in CTp. = Si (0.64 +(2.16 xd/hp) (6.11) Sp =S,(0.64 + (0.54 w/hp) - (0.18 w 2/lh p )) (6.12) 96 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH equations 6.11 and 6.12 is routinely determined using the following formula developed by Gaddy (1956) and Hustrulid (1976): Si (psi) = UCS VD / (V36) for h p greater than 3 feet (6.13) where: D = diameter or side dimension of the UCS test specimen (inches) UCS = uniaxial compressive strength (psi) Mark and Barton (1996) have shown that the size effect is related to coal structure. For friable coal, the size effect may be negligible or non-existent. These findings are echoed by Khair (1996), who found that the effect of coal specimen size on strength is limited to the relative size of inhomogeneity in the material with respect to the size of the specimen. Mark and Barton (1996) have statistically analysed many case histories of pillar failures. Their findings indicate that there is a poor correlation between sample strength, size effect, and pillar stability. The best statistical correlation of pillar stability was arrived at when the in situ coal strength was given a constant value of 6.2 MPa. Several other researchers have also suggested that despite wide variations in laboratory strength the actual in situ strength may fall within a narrow range (Galvin 1995, Madden 1991). 6.1.3 Factor of Safety The factor of safety (fos) for a pillar is equal to (pillar strength)/(pillar stress). The fos is used to account for variations in input parameters, and parameters that are not considered in the pillar strength equations (i.e. time and environmental conditions). Based on an analysis of US industry practices Bieniawski (1981) proposed that a fos of between 1.5 and 2.0 be used for long term stability; the average fos using Bieniawski's PSU pillar equation was found to be 1.73. Mark (1994) reports that case studies using the ARMPS program (see Section 6.4) have indicated that, for short term pillars, 94% of designs are satisfactory when the fos exceeds 1.5, whereas 92% of designs fail when the fos is less than 0.75. Sheorey et al (1987) proposed that the factor of safety used should vary between 0.6 and 2, depending upon such factors as depth, pillar type and packing (backfilling). 97 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Barron et al (1982) approached safety in terms of acceptable probability of failure. They proposed that the tolerable probability of failure is dependent upon the type of pillar. The following tentative values were proposed: 6.1.4 Pillar Failure Modes Based on a review of field observations and laboratory tests, Brady and Brown (1985) have divided pillar failure modes into those related to rock mass and those related to geologic structure. Where failure is related to the rock mass three further divisions are made. 1. In massive rock and in pillars with high width to height ratios, failure initiates with spalling at the edges and progressively moves inwards to create an "apple core", Figure 6.5a. 2. In regularly jointed rocks with low width to height ratios, failure may take the form of an inclined shear plane which transects the pillar, Figure 6.5b. 3. Where the pillar - roof/floor interface is weak, transverse tractions over the ends will occur. These may result in internal axial splitting within the pillar. Failure will manifest itself as bulging or barrelling of the pillar surface, Figure 6.5c. This is confirmed by Coates (1981), who reports that soft partings will induce horizontal tension. Where failure is related to specific geologic structure, several possible modes of failure occur. 1. Where a geologic structure cuts the pillar, failure can be expected if the shear stresses exceed the shear strength of the structure, Figure 6.5d. The amount of movement required for yield and relaxation of the elastic stress state may be very small. 2. Where geologic structures are parallel to the pillar walls, failure will be by buckling type failure, Figure 6.5e. Pillar Type barrier mains panel Acceptable Probability of Failure 0.1 % 1.0% 5.0 % 98 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Biswas et al (1995) found that where a weak horizontal parting is present within the pillar, or at the roof/floor contact, squeezing of the parting material and slippage at the interfaces of the weak parting and coal may occur. Figure 6.5: Typical Modes of Pillar Failure, after Brady and Brown (1985) 6.2 PILLAR PERFORMANCE AT THE QUINSAM COAL MINE 6.2.1 Coal Strength Prior to evaluating pillar strength, it was necessary to ascertain the in situ strength of the coal at the Quinsam Coal Mine. Numerous laboratory strength measurements were carried out by the author and others, see Chapter 2. The results from these tests were found to be quite variable. This variation is not unreasonable considering the variable nature of the coal and associated geologic structure. An average UCS value of 20 MPa was selected after considering the results from all sources. An in situ coal strength of 6.2 MPa was determined using equation 6.13. This value is consistent with the recommendations of Mark and Barton (1996). 99 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH A weak parting is often found in pillars at the Quinsam Coal Mine. This material is a mudstone with a UCS of less than 1 MPa, and a slake durability value of 5. The mudstone tends to squeeze and flow causing failure at the skin of the pillar as illustrated in Figure 3.5. Overall pillar stability was observed to not be adversely affected by the pillar skin failures around the mudstone parting. 6.2.2 Visual Pillar Performance A visual pillar classification system was developed to assess the deterioration of pillars adjacent to the gob. Based on past pillar behaviour, an 8 Class system was developed where Class 0 represents no pillar deterioration and Class 7 represents complete pillar failure. Photos 6.1 to 6.5 show pillar Classes 0, 2, 3,4, and 5. Descriptions of the extent of failure are given with each Photo. Descriptions for Class 1, 6 and 7 are as follows: • Pillar Class 1. Less than 0.1m of sloughing from the pillar. Cracks penetrate less than 0.25m into the pillar. The crack height is less than half the pillar height. Deterioration is greatest at the pillar corners. • Pillar Class 6. Sloughing exceeds 2m. The failed material entirely obscures the pillar. A slope of failed material, at angle of repose, extends from the top of the pillar out. • Pillar Class 7. Total failure of the pillar. Over 230 observations of pillar deterioration were made in the 2N Mine during depillaring operations. The records.include the following information on the pillars: location, size, depth of cover, thickness of mudstone parting, age, proximity to gob, visual classification, and any unusual factors. The collected data is summarised in Appendix 4. Figure 6.6 shows an example of the pillar classification maps generated during the study. No Class 6 or Class 7 pillars were observed during the course of study with the exception of stump pillars left in the gob. 100 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Photo 6.1 Pillar Class 0. No detectable pillar deterioration Photo 6.2: Pillar Class 2. Less than 0.25m of sloughing from the pillar. Cracks penetrate less than 0.5m into the pillar. The cracks extend for more than half the pillar height. Deterioration is greatest at the pillar corners. Note weak mudstone parting in photo. 101 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Photo 6.3: Pillar Class 3. Less than 0.5m of sloughing from the pillar. Cracks penetrate less than 0.75m into the pillar and extend almost to the full pillar height. Hourglass shape is developing. Note weak mudstone parting in photo. Photo 6.4: Pillar Class 4. Less than 1.0m of sloughing from the pillar. Cracks penetrate less than 1 5m into the pillar rib and extend over the full pillar height. Hourglass shape is developing. Note weak mustone parting in photo. 102 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Photo 6.5: Pillar Class 5. Up to 2m of sloughing from the pillar. Cracks penetrate more than 2.5m into the pillar and extend full pillar height. Hourglass shape is well developed. Figure 6.6: Example of Pillar Classification Map 103 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Several conclusions were drawn from the visual pillar classification study. 1. Pillar deterioration occurs with time; smaller pillars are more susceptible to deterioration than larger pillars. 2. Pillar deterioration increases with proximity to the gob. Figure 6.7 shows a plot of pillar classification versus distance to the gob. Although a trend of increasing deterioration with proximity to the gob is apparent, a statistically significant relationship could not be determined. Pillars with classifications of 0 and 1 often occur at the edge of the gob. 3. Pillar deterioration, which is the result of stress transfer, was generally found to occur only within 40m of the gob. This result is consistent with the findings of Mark (1996), who reports that 90% of the abutment load at the edge of the gob is carried by the pillars within a distance of 5 V H . At 50m depth, 5 V H =35m. 4. Proximity to a larger barrier pillar does not significantly affect pillar deterioration. 5. Pillars with visual classifications less than 4 posed no extraction difficulties. Pillars with visual classifications of 4 or 5 posed only minor extraction difficulties. 6. Where unusual deterioration was observed, it was always associated with discrete geologic structures such as major faults or very weak parting. The noses of pillars were the areas most susceptible to anomalous deterioration. Figure 6.7: Visual Pillar Classification versus Distance from the Gob 5 a m O 4.5 4 3 5 II 2 5 2 * 1.5 1 0.5 am 10 3 0 4 0 5 0 2 0 5 0 Distance From Gob (m) 104 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 6.3 PILLAR DESIGN METHOD Stress is well accounted for in the existing pillar design methods; as such, it is expected that existing design tools should be suitable for use in shallow depth mines. An evaluation of the existing pillar design methods indicated that the Analysis of Retreat Mining Pillar Stability (ARMPS) method developed by the USBM is the most suitable method for use in shallow depth retreat pillar coal mines. There are several reasons for this: 1. The method was designed specifically for use with the retreat pillar mining method. 2. The method uses the Mark-Bieniawski pillar strength formula, a modified version of Bieniawski's (1967) pillar strength formula. Bieniawski's formula is well accepted in the North American coal industry. The modifications take into account the increase in strength of rectangular pillars over square pillars. 3. The method accounts for the increased abutment stresses at the edge of a gob. 4. The method considers the stability of a group of pillars, not just individual pillars. 5. The method can be used to analyse four common situations in retreat pillar coal mining, namely; 1) development loads only, 2) development load and front abutment load, 3) development load, front abutment load, and one side abutment load, 4) development load, front abutment load, and two side abutment loads. The four scenarios are illustrated in Figure 6.8. 6. The method has been compiled as an easy to use computer program. The ARMPS program calculates a fos for a group of pillars within the active mining zone (AMZ). The AMZ includes all pillars within a distance of 5VH of the gob. As discussed in Section 6.1.4, this is the distance over which the majority of the abutment stresses are re-distributed. If an individual pillar within the AMZ is determined to have failed, the excess load on the failed pillar is transferred to the adjacent pillars. The reader is referred to Chase and Mark (1993) and Mark et al (1995) for further discussion of the ARMPS program. 105 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 6.8: Different Loading Conditions Considered in ARMPS, after Mark et al (1995). 0 0 O D D a a I • a o 0 0 0 0 ! Q O O • • o a • g o a o o a G O O o Q o a o o o a o a o o o a a o • • 0 0 0 0 0 o a a o o 0 0 0 0 a D O o a o a o a a 0 0 D O 0 0 0 0 0 a o 0 0 0 0 a a o 0 0 0 o a a a o a a 0 0 0 0 a a o 0 0 0 0 • a o 0 0 a a a o o o o 0 0 • 0 0 0 0 0 0 • 0 0 • • a a a o o 0 0 a o a a o o a a a 0 0 0 a a 0 0 a o o o a o a o o a o a o l a a a o D o a 1 • • • • • • o O Q Q O D O O a a a g o o a 1 o o a o n o o • • • o a o o o o a a o o a Q O C O Q O O O Q O O O Q O • • • • • o a o o o o o o o a o o a o o o o o o o o o o o a a o a a o o o o o o o o • 0 0 0 0 0 0 o o o o o o o • o a a o o a o o o o o o o • • a a o 0 0 o o o a a o a o o D o a a o o a o o a o o o o a a o o a —I a a o 1 — • o a a o o o a o o o o o o o o o o o o o o o o o a a o a o o i o a a o a o o o o o o o o o o o a a o o a • a o a o a a a o o a o o o o o o o o o o a 0 0 0 0 a o o o o o o o o o o o o o o o a a o o a o o a o o a o o o a o o a o o o o a a o a a a • • a o a o o a a o o a o o o o o o o o o 0 0 0 o a a a o o a a o o a o o a a a a o o a a a a a a o a o • a a L E G E N D . f Active mining zone Front gob First side gob Second side gob • O D O O O O O O O a D O O D O O O O O O O O O 0 0 0 0 0 0 • o a o o a o o o o o a o o o o o a o 0 0 0 0 0 a o o o a o O O D O O O O Q O O O O O • • • • • • o a o o o o a o o o o o o o o o a o o o ' o o o o o o o a o o a o a o a o a a a DO a a o o a o o o o a a o o o a a o o a o o a o o a o o a o o a o o a o o o 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 o o o o o o o a o o o a a a o a a a a a o 0 0 0 a o a o o o a a a o o a o o o o a a 0 0 a o o o o 0 0 0 o o 0 0 o a o o 0 0 0 0 0 0 0 0 0 0 0 0 0 o • 0 0 0 • • a o o o a o a o 0 0 0 0 • a o a a 0 0 gassBss 0 0 0 0 0 o a a o o o o D O 0 0 0 0 0 o a 0 0 0 0 0 0 0 O D D 0 0 o a • o a o a 0 0 • 0 0 o a o o • o a o o • • 0 0 0 o a a a o a o o a 0 0 o a a o a 0 0 • o a o o o a 0 0 0 0 0 0 0 a o o o o a o • • o a o o a o a o o a o o • o a o o o a o a a 0 0 o a 0 0 0 o a a a a o o o 0 0 a a o o a o o o o o o o o o o o a a o o a o o o o o a o o o o o o o o o a o o a a o • • • D O O O o a o \ — • a o OOOOOOOi a o o a o o o . o o o o o o o a g o b o a I o a a l o o o o o o o l o o o o o o o b o o o o o o L o o o o o o o l o o o o o o o l o o o o o o o o o o o o o o o o o o o a o o o o o o o o • o o a a o a o o o o o o o a o o a o o o o o o o o a o o o o o o o o UOODOOO l a a a a a a o a o o a o o o J o 0 0 0 0 0 0 . —1 a 0 0 r~—^ — l a o a I — G Q Q Q O Q O Q O D O D O • • O • 0 • O O 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 o o a o a o o o D o a o o a o o o a o o a 0 0 0 o a o a o o o o a a a o o o o a o o o o a o o o o o a o o o o o o 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 o o a o o o o o o o 0 0 a a a 0 o • 0 a a 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 o a o o a o o a a o • • D O O 0 0 o o o o 0 0 0 a o o o o a 0 0 0 0 o o a a o development load only development load and front abutment load o o o o o o o o o D a o o o o o o o o o o a o o o o O D O o O Q o a o o o o o Q Q O o o o o a o a o o D o o o o a o o o 0 0 o o • o a r 0 0 0 0 0 o a 0 0 0 0 0 a a 88888 0 0 0 a a • o a o o 0 0 o o o o o a o o a a a o 0 0 o o a a o o a a o o o o o o 0 0 0 o o o a 0 0 0 0 a o a a a o a a o a 0 0 0 a o o a o a o o a 0 0 a a o a a a o a o o a o o a a a a 0 0 o a a o o a o o o 0 0 0 a o a a a o o o o o a o o o o o o o a o o o o a a a a a • • • ia o o 0 • o a a a o a o a o o a o a o o a o o o a o o o a o o a • a o 0 0 0 0 o o o o o o o 0 0 0 o o o o a a o o a 0 0 a o o a o o o o o o o o a o o a a a a a a o o o a o o 0 o o o o o o o o o o o a o a a o o a o o o — 1 0 0 a a a a o o o o o o o o o o o o o o o a o a o o a o a a 0 0 a O o o o o o o o • a a a a a a a a o o o o o • 0 0 0 0 o o a a o o a a o 0 0 0 • • • • o a a o o o a o o a a a o o o a a o o o o o o a o o a o o a a o a o o o o o a o o o o a o o o o a a o o a a o o o • • D O D O O D O O O O Q • 0 • O • O O O O O O O O O 0 O O O • O O Q • O O O • o o o o a o a a o o a a o a o o a 0 0 a o o o a o o a a a a a o o a o a o a a a o o o D O O O D o o o O D O O D O O O O Q O D O Q O Q O O O O O O D O Q Q Da a o r o a o L development load, front abutment load and one side abutment load • o o o o o o o o o a o o o o o o o o o o • o a o o a o a o o o o o o o o o o a o a o o a o o o o 0 0 a a o a a o o o o o o a o o o o o a o a o o o a a o a o a o o o a a o a o a a a 0 o D O o a o o a a o a o o a o o a o o a o o o o • o o o a o a a o o o o a o o a a a o o a o o a o o o o o o o a a o o o o a o a • • o a a o o • • • o o o o o 0 0 o a 0 0 • a o o o a o o o o o o o o Q a o Q development load, front abutment load and two side abutment loads 6.3.1 Method Verification To verify the suitability of the ARMPS program in shallow depth mines the program was used to calculate the fos for all of the pillars within the Quinsam Coal Mine visual classification database. The calculated fos was then compared to the observations of pillar deterioration. The individual pillar fos was also calculated, using the Mark-Bieniawski formula, and compared to the ARMPS fos and visual classification value. The data for these comparisons is presented in Appendix 4. 106 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 6.9 plots the ARMPS and individual pillar fos versus the visual pillar classification. As expected, there is a general trend of increasing pillar deterioration with decreasing fos. However, this trend does not become apparent until after a visual classification of 2.0. This indicates that pillar skin failures less than 0.5m deep have little or no bearing on overall pillar stability. However, such skin failures may still pose a safety concern. For the Quinsam Coal Mine pillars a fos less than 1.0 was calculated for many individual pillars at the edge of the gob. The ARMPS group fos was always greater than 1.0. The ARMPS program redistributes the load from any individual pillars determined to have a fos less than 1.0. Provided that the combined strength of the pillars within the AMZ exceeds the total applied load, the ARMPS fos will be greater than 1.0. The ARMPS method of determining a fos is consistent with observations of pillar performance at the Quinsam Coal Mine; i.e. no pillars where observed to have failed during the course of this study. This finding verifies the appropriateness of the ARMPS method, and shows that evaluation of pillar stability on an individual pillar basis is not appropriate in shallow depth retreat pillar coal mines. Figure 6.9: Pillar Factor of Safety versus Visual Classification 10 *• b re (/) "5 5 i -o ' Individual Pi l lars l A R M P S 2 3 Visual Classification 107 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 6.4 RECOMMENDATIONS FOR PILLAR DESIGN AT SHALLOW DEPTH RETREAT PILLAR COAL MINES Several types of pillars are used in retreat pillar coal mines including: roadway or mains pillars, barrier pillars, panel pillars, fender or stump pillars, and point pillars. The different pillar types are illustrated in Figures 2.4 and 2.5. Although each pillar type is required to perform a slightly different function, the design method is essentially the same. The ARMPS program was developed to evaluate the short term stability of pillars during retreat mining. Mark et al (1995) reports that a suitable ARMPS fos for long term pillar stability is not known. However, the calculations in the ARMPS program are essentially the same as those proposed by Bieniawski (1981) who recommends a fos of 1.5 to 2.0 with an average value of 1.73. As such, it is considered appropriate to use an ARMPS fos of 1.73 for the design of long term pillars. 6.4.1 Barrier Pillars Barrier pillars are used to separate areas within the mine and provide regional stability to main access corridors. Barrier pillars are also used to separate mining blocks when water or ventilation control is required. In most circumstances barrier pillars must be indestructible. Wagner and Madden (1984), and Mark et al. (1997) have reported that sudden pillar failures will not occur beyond a width:height ratio of 4 and 3 respectively. Bieniawski (1981) suggests that no pillar failure will occur beyond a width:height ratio of 10. The recommended design criteria for barrier pillars in retreat pillar mines at shallow depth are as follows: 1. Utilise a minimum width to height ratio of 10 if pillars are required to be indestructible. 2. Utilise a minimum width to height ratio of 4 if yielding is acceptable. 3. Utilise a minimum ARMPS factor of safety of 1.73. 4. Increase the pillar size to account for significant faulting that transects the pillar, and weak roof or floor rock. 108 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 6.4.2 Roadway or Mains Pillars Mains pillars are used to provide stability and to separate the mine access roadways. These pillars are typically required to remain stable until the access is no longer required, at which time the pillars are recovered. Pillar design should include dimensioning to maximise recovery and minimise roof support during pillar recovery. The recommended design criteria for mains pillars in retreat pillar mines at shallow depth are as follows: 1. Utilise a minimum width to height ratio of 4. At shallow depth, the barrier pillars adjacent to the mains are expected to take most of the abutment stresses. 2. Utilise a minimum ARMPS factor of safety of 1.73. 3. Increase the pillar size to account for significant faulting that transects the pillar, and weak roof or floor rock. 6.4.3 Panel Pillars Panel pillars are used to ensure short-term stability within retreat pillar panels. As the pillars are recovered, a gob is created and abutment loads occur. The abutment loads will change as pillar extraction progresses and as gob caves occur. The recommended design criteria for panel pillars in retreat pillar mines at shallow depth are as follows: 1. Utilise a minimum width:height ratio of 4. 2. Utilise a minimum ARMPS factor of safety of 1.5. 3. Review pillar stability on a case by case basis if the following conditions occur: • Competent roof rock that does not readily cave. • Presence of faulting that transects the pillar. • Presence of weak roof or floor rock. 109 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 6.4.4 Remnant Pillars Remnant pillars are left in the gob; they include fender, point and stump pillars. Fender pillars are left between adjacent passes of the Continuous Miner; hence, they are long and slender. Point pillars are left at the noses of pillars and are usually triangular in shape. Stump pillars are located down the pillar centre line at the limit of reach of the Continuous Miner; they are typically irregular in shape. Stump pillars may also be left at other locations to provide local support to hazardous geologic structure. Remnant pillars are often left to improve safety and increase productivity. Safety is increased as a result of the additional support provided to the roof and the reduced excavation spans in the gob. Productivity is increased as the time associated with waiting for an unstable gob to cave may be reduced. These pillars are expected to remain stable or to fail in a stable manner such that unexpected roof caves do not occur. It is very important that the remnant pillars be properly sized; however, the irregular shape and pattern of these pillars means that they cannot be analysed with conventional design tools. If remnant pillars are too small, the roof will continue to cave with little change in mining conditions. If the pillars are too large, there is a possibility of a massive pillar collapse or run occurring. There are no published design recommendations for remnant pillars. At this time, there is insufficient information to provide design recommendations for remnant pillars. 6.5 CONCLUSIONS The pillar designs used at the Quinsam Coal Mine were determined to be adequate. Based on observations at the Mine the following conclusions on pillar performance at shallow depth retreat pillar mines have been drawn: 110 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 1. Pillars with a minimum dimension of 12m will be stable, even at the edge of the gob. 2. Pillar deterioration increases with proximity to the gob. Deterioration may cause local safety concerns; however, it seldom posses any difficulty to pillar extraction. Pillar failures to a depth of 0.5m have no significant impact on overall pillar performance. 3. Pillar deterioration is generally limited to the zone within 40m of the gob. This result is consistent with the findings of Mark (1990) who reports that 90% of the abutment load at the edge of the gob is carried by the pillars within a distance of 5 VH. 4. Where adverse geologic structures are present the extent of pillar deterioration may be greater and conclusions 1, 2 and 3 above may not apply. 5. The presence of a sub-horizontal mudstone parting in the pillar results in pillar skin failure and local safety concerns; however, it does not have a significant impact on overall pillar performance. This research has shown that in retreat pillar mines pillar stability should be determined for pillar groups, not for individual pillars. Use of individual pillar design methods will tend to result in a very conservative pillar layout. A review of the existing pillar design tools determined that the ARMPS program developed by the USBM (Mark et al 1995) is the most suitable method for shallow depth retreat pillar coal mines. The ARMPS program utilises a modification to the tributary area theory to account for the additional stresses that occur adjacent to a gob. As such it is suitable for use where the pillar area to be analysed is not large compared to the depth of cover. As well, the program uses a modified pillar strength formula that accounts for the increase in strength of rectangular pillars over square pillars. The ARMPS program calculates a group fos based on the total load and total strength of all pillars within the active mining zone. The program first analyses the fos for each pillar; if a pillar is determined to have a fos less than 1.0, the excess load is transferred to an adjacent pillar. The ARMPS program may be used to design barrier pillars and mains pillars. A minimum ARMPS fos of 1.73 should be used for all permanent pillars. Panel pillars that are to be extracted on the retreat should be designed with a minimum ARMPS fos of 1.5. I l l GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH During the course of this study it was determined that stump, fender, and point pillars left in the gob play a very important role in safety and productivity. At this time, there are no suitable design tools for these pillars. Developing design methods is considered a very important consideration for future work. While this research was being undertaken recommendations were made to the Quinsam Coal Mine to leave additional pillars in the gob, especially where adverse geologic structure was present. Since adopting this recommendation, the mine has not experienced any injuries or equipment damage due to gob caves. This is discussed further in Chapter 7. 112 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH CHAPTER 7 GOB CAVE PREDICTION 7.0 INTRODUCTION Caving of the roof is an integral part of the retreat pillar mining method. As pillars are mined out, the roof in the gob is expected to cave. Occasionally the mining equipment gets caught in these caves; this may cause significant damage to the equipment, result in lost production, and is potentially dangerous to the underground workers. As such, it is extremely important to be able to evaluate roof stability and predict when the mined out gob areas will cave. Photo 7.1 and Photo 2.3 show open gobs prior to the cave. Photos 7.2 and 2.4 show the gob just after the cave has occurred. These photos illustrate the magnitude and potential hazards of gob caves. At most retreat pillar mines the underground operators assess the stability of the gob using criteria such as pillars crushing, rock bolt heads popping, roof slaking, and rock noises. These criteria are all responses to roof convergence. The purpose of this portion of the study was to extend the work carried out on roadway stability to gob stability. As determined in Chapter 4, stability can be accurately evaluated using excavation convergence measurements. 7.1 LITERATURE REVIEW Instrumentation measurements have been used by many researchers to evaluate gob stability. Details of these studies are presented in Chapter 4, and summarised in Table 7.1. In general, it has been found that gob caves will occur when the rate of movement and/or the total movement exceeds a threshold value. The rate of movement is considered to be the more useful tool as it is independent of previous movement. Total movement analysis requires that instrumentation measurements commence immediately after an excavation has been developed (Maleki 1988). 113 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Photo 7.1: Open Gob with Crushed Wooden Post in Background Photo 7.2: Gob Immediately after the Cave 114 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Table 7.1 Gob Stability and Convergence Data from the Literature Mine Depth (m) Rock Type Convergence Total Rate Condition mm mm/min Comments Reference Sufco 5.1 critical Pillar extraction Maleki 1988 Umgala sandstone 45 .01 critical Pillar extraction Naismith, Pakalnis 1982 Plateau' mudstone 199 14.9 cave Pillar extraction Maleki 1988 Plateau mudstone 7.6 critical Cave within 20 minutes Maleki 1988 Plateau mudstone <5.1 stable Pillar extraction Maleki 1988 Helvetia 1.27 cave Adjacent to gob Serata 1989 Helvetia >0.42 unstable Adjacent to gob Serata 1989 Lucerne 2.1 unstable Cave-in imminent Serata 1989 Lucerne <0.42 stable Serata 1989 Roncourt 60 >0.66 critical Deniauetal. 1982 Mairy 30 >0.27 critical Deniau et al. 1982 Serrouville 8 >0.25 critical Deniau etal. 1982 7.2 GOB CAVE STUDIES AT THE QUINSAM COAL MINE 7.2.1 Mine Operators Stability Assessment The underground operators at the Quinsam Coal Mine monitored several variables to assess stability and predict gob caves; these included: rock cracking sounds, crushing and slaking of pillars, crushing of wood posts, slaking of roof rock, and rock bolts "popping" (breaking). General experience and intuition were also reported to be very important in predicting failure. Photos 7.1 and 2.3 show examples of the visible effects of ground movements in the gobs prior to caving. Photo 7.1 shows a broken wooden post while Photo 2.3 shows a failed stump pillar. 7.2.2 Instrumentation Stability Assessment The excavation stability study reported in Chapter 4 determined that mine roadways were generally stable right up to the edge of the gob, and that convergence measurements at the edge of the gob could not be used to predict stability within the gob. In order to measure roof convergence and assess stability within the gob it was necessary to install instruments within the gob that could be remotely monitored. 115 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Single point electronic borehole extensometers were selected for this purpose. These instruments utilise an LVDT (linear variable displacement transducer) to determine movement between the anchor point and hole collar. The resolution of these instruments is +/- 0.01mm. The instruments were installed in boreholes much like the mechanical extensometers described in Section 4.3. Figure 7.1 shows a schematic diagram of the instruments used. Both the instrument and the readout unit (an ohmmeter) had to be certified as intrinsically safe for use in coal mines. Remote monitoring was achieved by extending the instrument electrical cable to a location outside the gob. Figure 7.1: Electronic Single Point Borehole Extensometer Eight instruments were installed in retreat pillar extraction panels. Useful data was collected from only three of the instruments. Of the remaining instruments, two were destroyed by mining equipment and three were not being monitored at the time that the cave occurred; the cave occurred unexpectedly. The data obtained from the three useful instruments is summarised in Table 7.2. Figure 7.2 shows the results from the two instruments installed in Panel 101. One instrument was installed to a depth of 1.73m, the other to a depth of 6.30m. It is interesting to note that significantly greater movement occurred on the deeper anchor. This indicates that the initial Cable Crimp Roof Platen Anchor Cable 116 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH failure and dilation occurred between the two extensometers. This was confirmed by visual observations that revealed the cave height to be between 3 and 4m. Table 7.2: Roof Convergence at Time of Gob Cave Location Depth (m) Rock type immediate roof Anchor depth (m) Rock mass classification RMR CMRR Radius factor Convergence at time of gob cave Total Rate (mm) (mm/min) #3 Mains 90 siltstone 4.7 28 34 8.0 17.5 0.66 2N, 101, 60 siltstone 1.73 35 42 3.5 2.7 0.03 2N, 101, 60 siltstone 6.30 35 42 3.5 15 1.98 Figure 7.2: Roof Convergence in Section 101 2N Mine, after Cullen et al (1996) 16.00 14.00 12.00 g - 10.00 O < 8.00 tn u. O § 6.00 4.00 2.00 ANCHOR HEIGHT —•—6.30 m •••••1.73m first movement, ft«.EA D P I N J T £ . « . £ S T primary cave 3-4m_ 177T gob heavy continuous miner pulled out I , i 1/ /t ! no mining, gob quiet w J no mining, minor pillar sloughing and roof falls in gob mining cutX mining point Y 0.00 4#«—I • ! I 'i ' 1 1 1 7/4/95 7/4/95 7/4/95 7/4/95 7/4/95 7/4/95 7/4/95 7/4/95 7/4/9S 7/4/BS 7/4/95 7/4/B5 7/4«5 7/4/95 7/4/95 7/4/95 7/5/95 7/S/9S 7/5/95 7/5/95 14:24 15:00 15:36 16:12 16:48 17:24 18:00 18:36 19:12 19:48 20:24 21:00 21:36 22:12 22:48 23:24 0:00 0:36 1:12 1:48 DATE AND TIME The significant difference in total convergence between the two instruments highlights the need to ensure that the reference anchor is above the height of failure. The critical convergence rates recorded at the Quinsam Coal Mine were similar to those reported in the literature, see Table 4.1. In order to provide adequate warning of the 117 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH impending cave, a convergence rate less than the critical must be selected. Based on the limited field data, a rate of approximately 0.3 mm/min was recommended for use at the Quinsam Coal Mine. This value is expected to provide adequate warning without an excessive number of false alarms. Many factors affect stability such that additional studies are recommended to better determine a range for the critical convergence rate. The instruments used in this study were installed well ahead of the approaching gob. None of the instruments recorded any roof convergence until they were at the edge of the gob, or had passed into the gob. This was consistent with the findings of roadway stability, reported in Chapter 4, where roadway excavations at the Quinsam Coal Mine were found to be stable right up to the edge of the gob. 7.2.3 Gob Size Assessment Table 7.2 includes information on rock mass classification and radius factor. The radius factor, developed by Milne et al (1996), is a measure of the excavation size. It is a refinement of the hydraulic radius term, commonly used in hard rock mining for stability assessments. The radius factor accounts for the presence of pillars and irregular excavation geometry. It is considered a much better measure of excavation size than the span, area, or hydraulic radius. Excavation size and rock mass quality are well documented factors in excavation stability; see Chapter 4. As more data becomes available, it is anticipated that better cave prediction models may be developed by statistically incorporating size and rock quality data with convergence rate. Attempts were made to use the data from the Quinsam Coal Mine to relate gob caves to the size of the gob and rock mass quality. Data was collected on the first gob cave that occurred in eight different panels. By limiting the data to the very first cave it was possible to analyse the data without consideration of the effects from adjoining gobs. The data collected included excavation size at time of cave (recorded as the effective radius), depth, CMRR classification value, number of faults, time to cave, and extraction ratio. The collected data is presented in Table 7.3. 118 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH At this time, there is insufficient data to determine any statistically meaningful relationships. It appears that there is a general trend of increasing stable gob size with increasing CMRR value; however, further data is required to verify this. Table 7.3: Gob Cave Data, Quinsam Coal Mine Location Depth Faults Extraction Time to CMRR Radius Comments (m) per m 2 Ratio Cave Rock Mass Factor (hour) Classification #2 Mains 83 0.0008 90 24 34 8.4 2N, 208 60 0.0001 85 24 40 11.7 2N, 302 50 90 48 35 8.1 initial cave was less than 2m thick 2N, 204 53 85 48 40 7.8 initial cave was less than 2m thick 2N, 207 82 0.0010 90 40 11.5 2N 101 55 0.0007 100 42 10.8 2S,101 35 0.0047 75 18 40 12.7 fault controlled 2S,102 40 0.0057 80 45 11.8 Based on the data presented in Table 7.3, depth or stress do not appear to be factors in predicting gob stability at shallow depth. Similar results were found with the roadway stability study discussed in Chapter 4. This finding is somewhat unexpected as the numerical simulations indicated that the extent of the stress relief zone increases as the depth of cover decreases. Within the stress relief zone bed separation and beam bending are expected to be more pronounced. Bed separation and sagging within the stress relief zone is postulated to be one of the mechanisms of gob failure, see Section 7.3 below. Table 7.3 indicates that it often takes more than 24 hours for the gob to cave once the operators have determined it is unstable. This may represent lost production time if there is no other areas available to be working in. Convergence instrumentation can be used to minimise this time by better predicting when a gob cave will occur. 7.2.4 Control of Gob Caves The mining equipment occasionally gets caught in unpredicted gob caves. During the course of this fieldwork, there were eight instances where a Continuous Miner was buried in a gob 119 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH cave. A review of the circumstances around these events indicates that in every case existing geologic structure was a contributing factor to the failure. In half of the cases, the unexpected cave occurred when the point pillars were being removed. Mark et al (1997) reports that nearly 50% of the fatalities that occurred in retreat pillar mines occurred when the point pillars were being extracted. The following two recommendations were made to the mine: 1. Leave all point pillars. 2. Carry out detailed mapping of geologic structure in retreat panels. Where geologic structures that form wedges are identified, additional stump and fender pillars should be left for support. The effectiveness of these strategies was demonstrated in the #2 Mains at the 2N Mine. This was one of the locations where a Continuous Miner was buried. The cause of the burial was determined to be a structurally controlled wedge failure that occurred as a pillar was being extracted. After the burial incident, the geologic structure in the remainder of the panel was mapped in detail. Support pillars were left in the gob wherever geologic structure with the potential to form wedges was identified. Figure 7.3 illustrates the extent of the mapping, along with details of the controlled pillar extraction that followed. Despite the large amount of adverse geologic structure, the remainder of the panel was mined successfully with no stability concerns. 120 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 7.3: Example of Detailed Structural Mapping and Controlled Pillar Extraction Detailed structural mapping, controlled pillar extraction, and leaving all point pillars, is now standard practice at the Quinsam Coal Mine. Since adopting these practices in 1998 there have been no injuries or equipment damage from unexpected gob caves. 7.2 GOB CAVE MODEL Based on field observations and the limited instrumentation measurements it was concluded that there are three principal mechanisms of gob caves at the Quinsam Coal Mine. These are, 1) failure related to blocks and wedges formed by geologic structure, 2) failure related to span between unmined pillars, 3) failure related to span of cantilevered beams created as pillars are removed. These mechanisms are illustrated in Figure 7.4. 121 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 7.4: Proposed Gob Failure Mechanisms , Pil lar to be recovered. Dilation of joint surface prior to failure. Wedge defined by geologic structure. Fails when supporting pillars are recovered. a) Gob Cave caused by Geologic Structure. Stress relief zone Sagging of beams and dilation of bedding planes prior to failure. Failure occurs when sufficient sag occurs for snap through or shear failure at abutments. b) Gob Cave caused by exceeding crit ical span. Bending of beams and dilation of bedding planes prior to failure. >. Pillar to be recovered. _ i ^ ~—t-. . . ^ ^ ^ ^ ^ ^ ^ ^ / / / / / / /V/ / / / / J r i >' t i L ' * / / / / / / / / / / / / . / / / / / / / / / / / / / . O caved O v. Failure occurs when stresses at abutment of canti levered beams exceed strength. c) Gob Cave caused by exceeding critical cantilever span. 122 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Wedges defined by geologic structure are often supported by pillars. As the pillars are removed the support is lost and failure of the blocks becomes kinematically feasible. This type of gob failure is illustrated in Figure 7.4a. Depending upon the arrangement of the wedge and supporting pillar it may only be necessary to remove support from one side of the wedge to induce failure. It is expected that there will be dilation of the discontinuity surfaces that define the wedge prior to failure. However, the magnitude and length of time over which dilation occurs prior to failure is expected to be much less than that for the other two mechanisms. The magnitude of roof bed sagging and separation is a function of the span between supporting pillars. As the span increases, the amount of bed sagging (beam bending) will increase. As bending occurs, shear and compressive stresses are generated at the abutments while tensile stresses are generated in the middle of the beam. When the strength of the rock is exceeded, failure will occur. This failure mechanism is illustrated in Figure 7.4b. In practice, failure usually occurs on pre-existing geologic structure or planes of weakness, not through the intact rock. Where a caved gob is present on one side of the pillars being extracted the mechanism may be cantilever type failure of the roof beams as illustrated in Figure 7.4c. Removing the supporting pillar results in the sudden formation of a long cantilevered beam. As the beams bend, shear stresses will be generated at the abutments. Once the strength of the rock is exceeded, failure will occur. In practice, failure usually occurs on pre-existing geologic structure or planes of weakness, not through the intact rock. Al l three of the proposed failure mechanisms have an element of deformation and dilation that will cause excavation convergence. This convergence can be used to predict failure. To successfully predict a gob cave it is necessary that the monitoring instruments are located within the area about to fail, borehole instruments must be anchored into rock beyond the failure surface. 123 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH The numerical simulations have shown that as the depth of cover decreases, the magnitude of the induced stresses decreases and the extent of the stress relief zone increases. Where failure is caused by exceeding a critical span, it is expected that there will be a difference in gob caves at shallow and deeper depths. However, as discussed in Section 7.2 there is no evidence to suggest that stress or depth of cover are important factors in gob cave stability. A possible explanation for this finding is that gob caves at the Quinsam Coal Mine are primarily driven by geologic structure. 7.4 CONCLUSIONS Experience at the Quinsam Coal Mine has shown that operators can predict most gob caves with reasonable accuracy; however, unpredicted gob caves may occasionally occur. Accurate and reliable gob cave predictions can be made using roof convergence instrumentation. To be effective the instrumentation must be continuously monitored, and must be located in the section of the gob that is about to cave. Additionally, if borehole instruments are used they must have the reference anchors located beyond the height of the cave. Based on results from the Quinsam Coal Mine, and the literature review, the critical rate of movement for retreat pillar coal mines is expected to be in the range of 0.2 to 1.0 mm/min (432 mm/24 hour). In order to provide adequate warning of a cave, a convergence rate less than the critical must be selected. Based on the limited field data, a rate of approximately 0.3 mm/min was recommended for initial use at the Quinsam Coal Mine. This value is expected to provide adequate warning without an excessive number of false alarms. Many factors affect stability such that additional studies are recommended to better determine a range for the critical convergence rate. Depth and stress do not appear to be significant factors in the critical rate of convergence for gob stability. Similarly, depth and stress do not appear to be factors in determining the ultimate size that the gob reaches before it caves. Rock mass quality and the size of the excavation appear to be factors that influence the stability of the gob in shallow depth mines; however, additional data is required to confirm this. Geologic structure is the factor that has the most significant impact on gob stability. Identifying the geologic structures and, where 124 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH necessary, leaving pillars to support blocks and wedges formed by structure, is considered the best method of maintaining control of gob caves. The Quinsam Coal Mine has adopted the recommended practices of leaving all point pillars and conducting detailed structural mapping in all retreat panels. Where adverse geologic structure is identified the mine now leaves additional stump and fender pillars for support. Since these practices were adopted, there has been no injuries or equipment damage caused by unexpected gob caves. 125 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH CHAPTER 8 SUBSIDENCE 8.0 INTRODUCTION Ground movement over mines results from the collapse of rock into the underground excavation. When the ground movement reaches surface, it is termed subsidence. The potential for ground movement exists over all underground excavations with the greatest probability, and greatest magnitudes, occurring over shallow mining operations using caving methods such as retreat pillar coal mining. The ability to predict ground movements is necessary to protect surface facilitates from subsidence damage, and to protect the underground mine from problems such as water inundation. Buildings, roads, and power lines that are situated within the zone of subsidence may sustain irreparable damage. Subsidence fractures may provide a pathway through which water from overlying bodies of water may enter the mine. Subsidence prediction is necessary to establish protective surface set back distances and underground coal reserve areas. Subsidence has been extensively studied over high extraction coal mines operating at moderate to deep depths (greater than 100m). Reasonable capabilities now exist for predicting surface ground movements in these situations (Gray 1990). An understanding of, and an ability to predict, ground movements over shallow depth mines is less well developed. The existing subsidence prediction methods have been found by many researchers to provide inaccurate predictions at shallow depth mines (i.e. Salamon 1976, Peng 1992). The developers of the existing predictive methods do not, for the most part, advocate their use at shallow depth (NCB 1975). However, on account of the lack of suitable tools, the existing subsidence prediction methods are sometimes applied to shallow mines. Fortunately, the results of this erroneous application are generally conservative in nature. Where the depth of cover is moderate to deep, the ensuing ground movements result in a well defined, smooth, surface subsidence profile. At shallow depth the surface ground movements are irregular and of greater magnitude; they often include open fractures, stepped fractures, and sink holes as shown in Photos 2.5, 8.1 and, 8.2. 126 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Photo 8.1: Open Surface Fracture at the Quinsam Coal Mine Photo 8.2: Stepped Fracture at the Quinsam Coal Mine - Partially Obscured by Vegetation 127 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Due to the magnitude and irregularity of the surface ground movements over shallow depth mines, damage to surface facilities is expected to occur. The most important consideration is the extent to which damaging ground deformations extend out beyond the edge of the underground excavations; this is defined by the angle of draw or the angle of critical deformation. The actual magnitude of the ground movements is seldom an issue at shallow depth as all movements may potentially cause damage to surface facilities. The term shallow depth, as it relates to subsidence, is better defined as the limit depth for the occurrence of damaging discontinuous ground movements. As discussed below, discontinuous ground movements are not expected to occur beyond a depth of cover of 30 times the excavation height. This chapter provides a review of the existing tools for predicting the extent of caving and subsidence. Possible reasons for the poor performance of existing design tools to predict subsidence at shallow depth are explored. New guidelines for predicting the extent of surface subsidence over shallow depth mines are developed based on data collected at the Quinsam Coal Mine and through numerical modelling. An extensive literature review was carried out that included subsidence and caving over both soft rock and hard rock mines. The behaviour of softer (sedimentary) rocks has generally been considered to differ significantly from that of harder (crystalline) rocks. However, it was believed that useful insights might be gained from a review of behaviour in both rock types. 8.1 SUBSIDENCE DEFINITIONS There are several terms that are unique to subsidence engineering. Within the literature, the definition of these terms was found to vary. The definitions used in this report are defined below; these definitions were found to be those most commonly used in the North American coal mining industry (i.e. Peng 1992, Brauner 1973, Abel and Lee 1980). Al l angles used in the definitions are measured from the vertical. 128 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Angle of Draw (8): the angle between a vertical line at the edge of the excavation and a line drawn between the edge of the excavation and the maximum extent of surface subsidence; see Figure 8.1. The maximum extent of subsidence is typically taken as the location where the vertical displacement is less than 6mm. Angle of Critical Deformation (8'): the angle between a vertical line at the edge of the excavation and a line drawn between the edge of the excavation and the extent of "critical" ground movement; see Figure 8.1. Critical ground movement is defined as that necessary to cause non-repairable damage to the structure under consideration. The value is dependent upon the type of structure being acted upon. The type of ground movement causing critical deformation will also be different for different structures; for example, concrete footings are most affected by strain deformation while power poles are most affected by ground tilt or curvature. Angle of Fracture (co): the angle between a vertical line at the edge of the excavation and a line drawn between the edge of the excavation and the extent of surface fracturing; see Figure 8.1. At shallow depth this angle is equal to the angle of critical deformation. Angle of Cave (i|i): the angle between a vertical line at the edge of the excavation and the line along which the rock is caving; see Figure 8.1. This angle is often measured from the horizontal. Caved Zone: zone of rock immediately above the excavation, that has fallen into the void; see Figure 8.2. The rock falling from the roof rotates. This results in a very high bulking factor. The original structure within the rock strata is lost. In deeper mines, the caved zone extends to where either the caving is choked off, or to where a competent rock unit bridges the cave. In shallow depth mines, the caved zone may extend to surface where it results in a highly irregular surface profile with open fractures, stepped fractures, and sinkholes; see Figure 8.3. 129 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 8.1: Definition of Subsidence Angles UNAFFECTED Fractured Zone: zone of rock that overlies the caved zone; see Figure 8.2. The strata in this zone are fractured; however, blocks do not rotate, and the basic structure of the rock mass is preserved. Hydraulic connectivity between surface and underground may still occur. In deeper mines, the fractured zone extends to where the stresses no longer exceed the strength of the rock mass. In shallow depth mines, the fractured zone may extend to surface where it results in open fractures with occasional steps, see Figure 8.3. Figure 8.2: Zones of Ground Movement, after Peng and Chiang (1984) 130 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 8.3: Schematic Diagram of Discontinuous Surface Subsidence Profiles OPEN FRACTURE STEP SINK FRACTURE SEAM CAVED ZONE • Rock rotates and dilates. • Stratified nature is lost. A: Caved Zone extends to surface. STEP FRACTURE OPEN FRACTURE FRACTURED ZONE • Rock fractures but does not rotate. Stratified structure is maintained. CAVED ZONE SEAM -7, B: Fractured Zone extends to surface. Continuous Deformation Zone: zone of rock that overlies the fractured zone; see Figure 8.2. The strata deflect downwards, but seldom fracture except over high strain areas. At the ground surface continuous subsidence results in a smooth profile with no stepped cracks. Discontinuous Subsidence Profile: an irregular surface subsidence profile with open fractures, stepped fractures, and sinkholes. There is hydraulic connectivity between the surface and underground. This is the type of subsidence that typically occurs over shallow depth retreat pillar coal mines where the caved or fractured zones extend to surface; see Figure 8.3. 131 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Continuous Subsidence Profile: a smooth surface subsidence profile with no steps. Open fractures are only found in the vicinity of the edge of the excavation where tensile strains are typically greatest. This type of subsidence occurs when the ground surface is located within the continuous deformation zone. The ground movements associated with continuous subsidence are characterised by seven parameters as follows—Xand Fare the horizontal and vertical Cartesian axes through the cross section of interest: Vertical displacement (s): movement in the vertical plane, usually downwards; however, small upward movement may occur over the excavation abutments. Horizontal Displacement («): movement in the horizontal plan, usually towards the centre of the subsidence basin. Slope (/): the change in vertical displacement over a given horizontal distance (X), i.e.: i = ds dX Curvature (A): the difference in surface slope between two adjoining line sections, i.e.: dX2 k = d2s Horizontal Strain (e): the change in distance between two points in the horizontal plane. If the distance is lengthening, the strain is tensile; if the distance is becoming shorter, the strain is compressive, i.e.: e = du dX Twisting (T): the difference in slope between two parallel line sections divided by the distance between the line sections, i.e.: T = d2s dXdY 132 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Shear Strain (7): the changes in internal angles of a square on any cross section, i.e.: 7 = du dY 8.2 LITERATURE REVIEW Ground subsidence due to mining likely occurred with the advent of underground excavations to recover chert for stone tools, pigment, and gems (Gray 1990). Primitive mining was carried out at shallow depth such that the propensity for discontinuous subsidence was high. There are records of severe surface damage from the 15th century (Young and Stoek, 1916). The first known theories and predictive methods for surface subsidence were developed in the 19th century by Belgian, French and German engineers (Shadbolt 1987). In recent years, much research has been carried out and many theories and methods for predicting surface subsidence and ground movements have been developed. Most of the previous subsidence research has focussed on the prediction of surface behaviour over moderate to deep depth mines with very high extraction ratios, and extraction areas that are large in relation to the depth. The resulting surface subsidence for this type of mining is continuous in nature. In most cases, a large percentage of the overburden rock falls within the zone of continuous deformation. Surface subsidence for these scenarios can now be predicted within acceptable limits (Gray 1990). It is generally accepted that the existing predictive tools are not applicable for shallow depth mines (Salomon 1976, Peng, 1992). Nonetheless, a brief review of the existing tools is presented below. The Author is aware of several instances where the tools for deeper depth mines have been used to predict shallow depth subsidence, presumably due to the fact that there are no existing tools for predicting shallow depth subsidence. The existing predictive methods can be categorised into the following three types: 1) empirically and semi-empirically derived relationships, 2) analytical/numerical methods, and 3) physical models. A brief discussion of these methods is presented below. Further details 133 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH can be found in most texts on subsidence such as Peng (1992), Whittaker and Reddish (1989), Kratzsch (1983). Empirically derived relationships include simple "rules of thumb", statistical methods, profile or curve fitting methods, and influence function methods. The simple rules tend to be site specific, while the influence and profile function methods are more general but require calibration at each new site. The empirical methods are the most widely used for surface subsidence prediction. This is a reflection of the relative ease of use and ability to provide predictions with reasonable accuracy. The newer empirical methods attempt to provide improved predictions by incorporating more site specific information such as geology and gob treatment (Cleaver et al. 1995, Shadbolt 1987, Tandanand and Powell 1982). Most of the empirical subsidence prediction methods have been developed for longwall mining; the application of these methods to room and pillar mining is subject to much debate. The NCB Subsidence Engineers Handbook (1975) sets the following limiting conditions for its use: 1. The working panels should extend for a distance of about 0.7 times the depth in front of, and beyond the surface point where subsidence is to be predicted; i.e. the total length of the panel must be 1.4 x (depth of cover). 2. The working panels should have no centre gates or other zones of special packing apart from those at the main and tailgates. 3. Where the sides of a panel are not parallel (owing to faulting, etc.), the average panel width must be developed. Moebs (1982) reports that attempts by the USBM to apply longwall subsidence prediction methods to room and pillar sites met with only limited success owing to the erratic and unpredictable subsidence upon removal of pillars. Kohli et al (1982) reports that room and pillar mining with a 80 % extraction results in subsidence profiles similar to longwall. Peng (1986) suggests that longwall subsidence prediction techniques can be applied to room and pillar mines provided that total extraction occurs (100% pillar recovery). Even where high extraction is achieved there appears to be subtle differences in subsidence over the two 134 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH mining methods: for example, Darmody et al (1989) reported significant differences in the effects of high extraction room and pillar mining versus longwall mining on crop yields. In practice, total extraction seldom occurs in retreat pillar mining, even where total pillar recovery is planned. At the Quinsam Coal Mine the extraction ratio seldom exceeded 95% even under ideal conditions. The average extraction ratio was found to be 85% during the course of this study. Numerical models can be used to theoretically determine both surface and subsurface rock movements in any situation. Provided that an accurate constitutive model is used, and the material properties and in situ conditions are known, it is theoretically possible to model any scenario. In practice, determination of an accurate constitutive model, subsurface conditions, and input parameters is very difficult. Starfield and Cundall (1988) consider that all numerical modelling in the field of rock mechanics falls into the class of "data limited problems"; one seldom knows enough about the rock mass to model it unambiguously. Numerical modelling is a useful tool for parametric studies and to gain additional insights into subsidence but is unlikely to provide definitive answers. This said, many researchers have used numerical modelling to predict subsidence (i.e. Choi and Dahl 1981, Su 1992, Najjar et al 1993, Szostak-Chrzanowski and Chrzanowski 1991, O'Connor and Dowding 1992). These researchers have generally reported good results once the models were properly calibrated to the conditions being considered. Physical modelling requires construction of a scale model of the conditions to be evaluated. Materials used in the models must be selected such that the properties can be scaled to match the in situ conditions; careful attention to the problem of similitude between lab and field material properties is required. Although not as common as it once was researchers continue to use physical modelling to gain insights into the subsidence process (i.e. Whittaker et al 1990, Sutherland et al 1984). 135 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 8.2.1 Sub-Surface Ground Movements Surface subsidence is merely the manifestation of sub-surface ground movements; therefore, to fully understand surface subsidence it is considered necessary to understand what is happening in the rock between the excavation and surface. Sub-surface rock movements have been studied by numerical models (i.e. Sutherland and Schuler 1981, Su 1992, Peng 1994), physical models (i.e. Sutherland and Schuler 1981, Kratzsch 1983, Whittaker et al 1990), geophysical investigations (i.e. White 1992) and a few in situ monitoring programs (i.e. Conroy and Gyarmaty 1983, Kolebaeva 1968, Holla and Buizen 1990, Howell et al 1976, Liu 1981, Peng and Chiang 1984, Wade and Conroy 1980). The information available on sub-surface behaviour indicates that ground movements can be divided into caved, fractured, and continuous deformation zones as indicated in Figure 8.2. Rock failure in the caved and fractured zones is attributed to various mechanisms including shear and tensional failure of beams, de-lamination of bedded strata, and gravitational type failures such as progressive unravelling and block falls. Most of the conventional theories of subsidence suggest that rock failure in the continuous deformation zone involves beam bending and de-lamination only; these ground movements result in the smooth continuous surface subsidence profiles that occur over deeper mines. The influence of geologic structure on sub-surface ground movements and ultimately on surface subsidence has been recognised by many researchers (i.e. Crane 1929,1931, Heslop 1974, Boyum 1961, Kantner 1964, Fletcher 1960, Parker 1978, Kotz 1986, Mahtab 1976, North 1980, Hoek 1974, Peng 1992, Nelson 1981, Holla and Buizen 1990, Lee 1966, Shadbolt 1987, Hellewell 1988, Whittaker et al 1990, O'Conner et al). Many observations of the influence of geologic structure have been made; however, only a modest amount of research work has been carried out. Hellewell (1988) reports that understanding the effects of geologic structure is complicated by the fact that the results of scientific investigations are in some instances contradictory. 136 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Crane (1929) carried out extensive measurements of caving at iron ore mines in Michigan. He concluded that failure in rocks over mining excavations is largely controlled by existing planes of weakness. Failure and movement will take place initially on the steepest dipping failure plane, and continue progressively to the most shallow failure plane along which failure is kinematically possible. The most pronounced movement will be on the most prominent failure plane. Parker (1973) noted that in rocks with no significant geologic structure the angle of cave is usually consistent and can be predicted with reasonable confidence, otherwise the angle of cave is determined by geologic structure. Parker further found that a well defined steeply dipping fault plane, which is parallel to a mining face, will result in fast caves defined on surface by the trace of the fault plane. If the predominant joints and faults are near perpendicular to the mining front, caving will be inhibited and negative angles of draw may occur. Physical modelling of the influence of faults carried out by Whittaker et al (1990) indicated the following: 1. Foot wall workings are most likely to activate movement along a fault. 2. Surface steps occur only when the workings are very close to the fault. 3. Faults within a horizontal distance of 0.5 x depth of the excavation will affect subsidence. 4. Foot wall workings may result in additional surface fracturing due to large tensional strains. Whittaker and Reddish (1989) have summarised the observed responses to faults as follows: 1. A stepped surface fracture will result if the fault outcrops at surface. 2. If mining is only on the hanging wall side of a fault, the fault will usually limit surface movement to that side of the fault. If the fault dip is steeper than the angle of draw, the extent of surface subsidence will be reduced. If the fault dip is less than the angle of draw, the extent of surface subsidence will be increased. 3. If mining is only on the foot wall side or on both sides of a fault, the ultimate angle of draw will be unchanged by the fault. 137 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 8.2.2 Angle of Draw The focus of the literature review was shallow depth mines. The review turned up hundreds of references to the angle of draw over soft rock mines and hard rock mines; however, the amount of published information on shallow depth mining was found to be very limited. The collected information is summarised in Appendix 5. Some examples of angle of draw from deeper mines are also included in Appendix 5. A possible explanation for the lack of published information from shallow mines is that significant ground movements and damage to structures within the subsidence zone are a foregone conclusion such that little research has been considered necessary. A review of the collected data shows that the angle of draw for shallow coal mines varies between 0° and 29°. The reported angle of draw for shallow hard rock mines varies between 0° and 16°. For both hard rock and coal mines there is a general trend of increasing angle of draw with increasing depth; however, exceptions are common, especially where geologic structure is sited as a factor controlling the angle. Abel and Lee (1980) carried out an extensive analysis of angle of draw data from literature sources. Their statistical analysis indicated the following: 1. The general shape of the subsidence trough and the angle of draw are similar for both longwall and room and pillar mining; only the magnitude of subsidence is different. 2. The angle of draw decreases with an increasing percentage of limestone in the overburden; however, the percentage of shale, limestone and sandstone is by itself a relatively poor predictor of angle of draw. Peng and Geng (1982) statistically analysed data from over 100 US and Chinese longwall coal mines. Their statistical analysis indicated the following: 1. The angle of draw and angle of critical deformation are affected by section dimensions; the angles will be greatest (and constant) when the critical dimensions are exceeded. The angles decrease as the dimensions become sub-critical. 2. The angle of draw is approximately 18.2° larger than the angle of critical deformation. 138 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 3. The angle of draw and the angle of critical deformation decrease with increasing mining depth. 4. The angle of draw and angle of critical deformation increase with increasing strength of the overburden rock. 5. The time taken to achieve full subsidence is significantly less when the overburden is comprised of soft rock. Singh and Singh (1968) consider the nature of the strata to be the most important parameter in determining the angle of draw at Indian coal mines. Dip of the seam, packing and depth of cover are considered secondary factors. Negative angles of draw often occur in the Giridih coal fields of India; this phenomenon is attributed to shallow depth and strong overburden. The angle of draw and angle of cave are commonly represented as straight lines. However, field observations have often shown the lines to be parabolic. Royce (1941) determined the line to be concave towards surface, while Heslop (1974) reports that the line is convex towards surface and may deflect to a shallow angle close to surface where confinement is reduced. Crane (1931) developed a system for predicting the angle of draw over hard rock mines based on joint measurements. Crane's observations led him to conclude that in the absence of faults and dykes, joint dip determines the angle of draw. The angle of draw for a mine will be equal to the dip of the most prominent joint; prominence is a measure of the number of observations, persistence, continuity, and character of the joint surface. 8.2.2.1 Angle of Draw Prediction Several methods have been developed for prediction of the angle of draw. Simplifying assumptions such as homogeneous rock masses are required in most cases. None of the methods have gained universal acceptance. However, all the methods are reported to correlate well with field observations under the conditions that the methods were developed. 139 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH The NCB Subsidence Engineers Handbook (1975) recommends an angle of draw of 35 and an angle of critical deformation of 26.5° for horizontal deposits in the UK coal fields. O'Donahues's rule of 1907 (Morrison 1976) for UK coal fields proposes an angle of draw of 8° for horizontal strata. This angle was used to establish set back distances to protect against subsidence damage; therefore, it is in reality the angle of critical deformation. Stratham's rule (Morrison 1976) for coal seams in South Africa, is as follows: For horizontal seams, Stratham's rule is equivalent to O'Donahues's rule. This angle was designed to establish set back distances to protect against subsidence damage; therefore, it is in reality the angle of critical deformation. Khair and Begley (1992) attempted to relate the angle of draw to the rock quality designation (RQD) and the percentage of strong strata in the overburden. Strong strata was defined as being either sandstone or limestone. The following equation was developed based on a back analysis of measured and published data. Although the concept appears reasonable, attempts to apply the equation by this author met with unusual results. Peng et al (1995) developed the following empirical formulas based on statistical analysis of 110 case histories from US coal mines: 5 = 8U + a (24 - 8u)/24 where: a = angle of seam inclination (8.1) 8 = 71 / 2 (RQD P 0 U + 0.06) (57.2958 degree/radian) where P 0 = percent of strong strata in overburden. (8.2) 5 = 6.87 - 0.0072H + 8.872 x 10"6H2 (8.3) 6' = 27.96 - 0.02426H + 6.9 x 10"6H2 (8.4) where H = depth of cover in feet. 140 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH The maximum angle of draw in the database was 24°, in 95% of the cases the angle of draw was less than 20°. Negative values for the angle of draw were reported to occur in "thin" overburden. The maximum angle of critical deformation was 23°; the minimum angle of critical deformation was -14°. No definition for "critical deformation" was given. It is interesting to note that the formulas predict an increase in angle of draw with increasing depth, but a decrease in the angle of critical deformation with increasing depth. No explanation is given as to why the general trends of these two similar angles are opposite. At shallow depth the angle of critical deformation is predicted to be greater than the angle of draw. No explanation is given for this discrepancy. 8.2.3 Cave and Fractured Zone Height Many researchers including Peng (1992), Kratzsch (1983), Whittaker and Reddish (1989), and Garrard and Taylor (1988) have reported on the factors affecting the height to which the caved and fractured zones extend. The following factors are considered to be the most important: extraction height, extraction width, size and location of pillars, rock mass properties (including bulking properties), geologic discontinuity properties and orientation, and ground water. The actual mechanics of caving are not well understood. Based on an analysis of stresses, Coates (1981) proposed that caving is initiated by either tensile failure in the centre of the excavation (where tensile stresses are predicted to be greatest) or by compressive failures at the edge of the excavation (were compressive stresses are predicted to be greatest). It is possible for both types of failure to occur simultaneously. The compressive and tensile stresses around an excavation are discussed in detail in Section 3.4 of this report. Abel and Lee (1980) carried out an extensive literature review and presented the following summary of the development of caving in hard rock mines: 1. "Collapse of rock progresses upward from the mining horizon (undercut level) as ore is withdrawn from below. The resulting column of caved and broken rock is confined above the area of extraction. 141 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 2. The ground surface does not begin to measurably subside until the collapse has so thinned the overlying intact rock that it cannot transfer the load of the overlying rock to the adjacent solid rock ribs. The overlying solid rock will then begin to deflect downward toward the collapsed rock below. Lateral movement of adjacent rock into the collapsed rubble column is resisted by the active (and possibly passive) pressure of the caved rock. 3. Further extraction of caved ore from below results in increased subsidence of the ground surface above and adjacent to the area of extraction. The overlying intact rock is progressively thinned by the further upward migration of the broken rock, which causes intact rock to deflect into the caved rock. The initial trough subsidence is similar in shape to the trough subsidence observed above coal mines. 4. Continued extraction of ore will result in breaching of the surface. The initial breach is typically in the form of a circular pit, commonly referred to as a chimney. The chimney is roughly centred over the mining area. Offsets may occur if geologic weaknesses are present. 5. If ore extraction continues, the surface breach will grow laterally near the surface. The rock adjacent to the subsided chimney either slides along geologic weaknesses, such as joints or faults, or topples into the open crater. 6. The final or ultimate angle of draw is determined by the lesser angle produced by either geologic weakness, or angle of repose of the broken rock mass. " Crane (1929, 1931) presents an extensive study of rock failure over hard rock mines. He reports that caving occurs in an orderly and systematic manner, and proceeds from the top of the workings towards the surface where it gradually spreads laterally until the ultimate angle of draw is reached. Where faults, dykes and other "abnormal" features do not exist failure is controlled by the joints, which he maintains are present to some degree in all rock types. Laubscher (1981) reports that two forms of caving commonly occur in block caving operations: 1. Stress caving occurs in virgin cave blocks when stresses in the cave area exceed the rock mass strength. The mechanism of failure is localised shear resulting in sloughing from the roof. In these situations caving will stop when a stable arch develops in the roof; 142 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 2. Subsidence caving occurs when: a) adjacent mining has removed the lateral restraint and the failed zone is undercut, resulting in rapid propagation of the cave, often with limited bulking; b) the rate of undercutting exceeds the rate of stress caving of the roof and then it fails en mass. The mechanism of failure is shear, usually along pre-existing structure. Results from numerical model studies (Singh et al 1993) suggest that in the absence of significant geologic structure, tensile failure is the main cause of progressive caving. Gravity loads result in tensile stresses in the rock mass. When the strain limit of the rock mass is exceeded, tensile fractures develop and caving occurs. In coal mines, the cave zone height is reported to extend to between 2 and 10 times the excavation height. The fracture zone is reported to extend to between 20 and 90 times the excavation height (Peng 1992, Whittaker and Reddish 1989, Piggot and Eynon 1978, Bai 1995, Wade and Conroy 1980, Conroy and Gyarmaty 1983, O'Connor and Dowding 1992). Caving is terminated when it is choked off by bulking of the caved material, or a competent rock unit capable of bridging the caved zone is encountered, or the ground surface is reached. In deeper mines with large spans, caving is usually terminated by being choked off. The fractured zone is terminated when the stresses are no longer sufficient to cause rock rupture, or the ground surface is reached. In hard rock mines, CANMET (1995) reports that the caving height is typically 0.5 times the excavation span. A failure height of 0.5 times the excavation span is consistent with the theoretical Mohr-Coulomb failure height, using an effective friction angle of 45 degrees; see Chapter 5 of this report. Hunt (1980) observed that in Illinois retreat pillar coal mines, caving will extend to surface where the depth of rock cover is less than 23m or where the total depth is less than 46m and the overburden soil to rock ratio is less than 1.2. 143 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 8.2.3.1 Predicting the Caved Zone Height Where bulking of the fallen rock chokes off the caving, the height of the caved zone (z) can be estimated using simple geometric relationships and the bulking factor (K). Three common caving shapes are illustrated in Figure 8.4. The formulas for calculating the maximum caving height for these three shapes are given below the figures. With equations 8.5 to 8.7, only the excavation height is a factor in the cave height; the length and width of the caved area are not considered. Piggot and Eynon (1978) report that K is typically in the range of 0.3 to 0.5 for coal measures rocks. These authors also report that the maximum cave height observed in the field is 10 times the excavation height. The predicted cave heights for K=0.3 and 0.5 are as follows: conical shape caves: 6 to 10 times the excavation height wedge shape caves: 4 to 6.7 times the excavation height rectangular shape caves: 2 to 3.3 times the excavation height Figure 8.4: Common Shapes of Caving R E C T A N G U L A R C A V E W E D G E C A V E i i C O N I C A L C A V E rectangular shape caving z = M/K (8.5) wedge shape caving z = 2M/K (8.6) for conical shape caving z = 3M/K (8.7) 144 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH where M = excavation height K = bulking factor = ( V c - V 0 ) / V 0 V 0 = in place volume of rock V c = disturbed (caved or loose) volume of rock Whittaker and Reddish (1989) have extended the above calculations to include the material that spills into the adjoining excavation as shown on Figure 8.5. The equation to calculate the height of the cave is as follows: z = 4(2BM2cotd>' + M B 2 ) / ((K) B2TT) (8.8) where: B = span of cave (assumed to equal excavation span) d/ = angle of repose of caved rock within adjoining excavations Figure 8.5: Development of Chimney Type Cave in a Four-Way Intersection, after Whittaker and Reddish (1989) 1. Initial collapse condition of mine junction. 2. Progressive development of roof collapse with caved rock spilling into adjoining rooms to roof level. 3. Caved rock fills remainder of chimney during self choking process. 145 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Using equation 8.8 the cave height in four-way intersections can theoretically be as much as 10 times the seam height. As the excavation width increases, the effects of the material that spill out into the adjoining roadways decreases, and the predicted cave heights from equation 8.8 and equation 8.5 will converge. Garrard and Taylor (1988) determined that bulking theory is not always applicable for prediction of the caving height over shallow room and pillar coal mines. In a study of over 150 collapsed workings they found that in 70% of the cases caving was arrested by arching or bridging, not by being choked off. Based on the collected data, these authors could not determine a direct relationship between the excavation height and the cave height; however, 99% of the data fit within the bounds of the maximum cave height being less than 10 times the excavation height. For excavations less than 75m deep, and less than 18m wide, the excavation span was found to provide the best estimate of cave height. The following relationship was developed: z = 2.68xB (8.9) Garrard and Taylor report that atypical caving heights greater than those predicted by the above equations may occur where the following occurs: 1. Well developed sub-vertical geologic discontinuities. 2. The span of the excavation is such that arching cannot develop. 3. Caved material is washed away from the caved zone into the surrounding mine voids. For estimating the extent of surface fractures, Salamon (1976) developed the following rules of thumb for shallow dipping South African coal mines: 1. If depth to seam thickness ratio is less than approximately 20, a discontinuous subsidence profile will occur, i.e. the fracture zone will extend to surface. 2. If the depth of cover is less than 50m, sinkholes may occur. 3. If depth to seam thickness ratio is between 30 and 50m, surface cracks may open up parallel to face, then close up as the face advances. Cracks may remain open over the abutments. 146 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 8.3 CAVING AND SUBSIDENCE AT THE QUINSAM COAL MINE To evaluate the extent of surface subsidence at the Quinsam Coal Mine, more than 230 surface subsidence stations were installed over 14 panels at the 2N and 2S Mines. The locations of these stations are shown on Figures 8.6 and 8.7. The northing, easting, and elevation co-ordinates for all the points are given in Appendix 5. The stations consisted of rebar or steel spikes driven into the ground to refusal. The surficial soils into which the spikes were driven consisted of a dense glacial till. The subsidence stations were installed and surveyed prior to mining. Level surveys, using a transit and stadia rod, were carried out to determine vertical movement as mining took place and again upon completion of mining. Additional surveys were carried out up to two years after mining had been completed. Where the depth of cover was less than 50m, surveys were not carried out during mining on account of the potential for the sudden formation of large fractures and sinkholes. The collected data from these surveys is presented in Appendix 5. Many of the survey lines were located in moderately dense forestland on undulating terrain. Conducting precise surveys proved to be very difficult in some places. In some areas the accuracy of the surveys was found to be as low as +/- 0.1m. The data from the level surveys was used to determine the angle of draw, angle of fracture, maximum vertical subsidence, subsidence factor, and type of subsidence occurring over the panels within the study area. The results are summarised in Table 8.1. This table also includes information on the panel geometry and lithology of the overlying rock. 147 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 8.7: Location of Subsidence Monitoring Stations over 2S Mine 149 X O H W Q O - J <: x H < a < o o < •J J E H < § O U 5 => H 00 J < U UJ f -O w O O U 1 C a ^ 3 O ei td Q <u o PI u JO 't/i X> O C/3 00 £ 00 (-1 60 O 00 & 2 00 fc 00 4= 2 CD o • c X o 3 > o o oo ^ H 2 o J 3 , Q w *3 OI o OI "^ 3 ^ bfj 3 rt O us w 6 S •a a o 4^  (A O 3 8 £ o GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH The results from the subsidence investigation are summarised in point form below: • The angle of draw has a mean value of 0.6° and a range of 0° to 5° (where not affected by faults). • Where faults are present the angle of draw may be greater. • The angle of draw generally increases with increasing depth of cover. • The angle of fracture has a mean value of -1.9° and a range of 0° to -10°. • The subsidence factor has a mean value of 0.31 and a range of 0.05 to 0.64. • The subsidence factor generally decreases with increasing depth. • Sinkholes may develop where the depth:height ratio is less than 14. • Discontinuous subsidence will occur where the depth to height ratio is less than 20 and may occur to depth to height ratios up to 30. Detailed mapping of surface fractures was carried out over Panel 101 at the 2N Mine and Panel 101 at the 2S Mine. Most of the vegetation and some of the topsoil had been stripped off from above these panels prior to mining. This provided the opportunity to collect detailed information on subsidence fracture formation. Experience had shown that fractures with displacement of less than about 0.1m may be obscured by vegetation. Photo 8.2 shows an example of a surface subsidence fracture that is partially obscured by vegetation. The surface fractures were mapped approximately every 7 days as the panels were mined. Figure 8.8 shows an example of the data collected over Panel 101 at the 2N Mine on August 9, 1995. Table 8.2 provides a summary of the fractures mapped over Panel 101 at the 2S Mine at the completion of mining. These fractures are shown on Figure 8.7 along with the faults mapped underground. Analysis of the surface fracture data showed that the strike orientation was similar to that of the faults mapped underground. However, the dip of the surface fractures was found to be dissimilar. Most of the surface fractures had dip values near 90 degrees whereas the major faults typically dip at 40 to 50 degrees (see Chapter 3). 151 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 8.8: Map of Surface Fractures Over Panel 101, 2N Mine, August 9, 1995 152 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Table 8.2: Surface Fractures over 2S Mine Sections 101 No. Width Depth Length Material Dip C O M M E N T S max max (m) (deg.) (m) (m) 1 .01 .1 6 sandstone 90 2 0.01 0.05 4 sandstone 3 0.01 0.1 3 sandstone 90 4a 0.02 0.15 18 sandstone 90 two parallel cracks 0.5m apart 4b 0.1 1.5 18 sandstone 90 terminates under rock fill 5 0.05 0.5 5 sandstone 90 terminates under rock fill 6 0.02 0.1 3 sandstone 90 several parallel and perpendicular hairline cracks 7a 0.01 0.05 5 sandstone 90 7b 0.05 0.5 4 sandstone 90 8 0.1 1.0 8 sandstone 90 terminates under rock fill 9a 0.1 1.0 30 till 90 saw tooth 10 0.1 1.5 6 t i l l , sandstone 90 60 terminates under rock fill 11 0.1 2.0 16 till sandstone 90 80 straight in till saw tooth in rock 12 0.1 0.8 3 till sandstone 90 80 straight in till saw tooth in rock 13 0.1 0.8 13 till sandstone 80 60 straight in till saw tooth in rock 14 0.05 0.5 11 till sandstone 15 0.8 3.0 25 sandstone 80-90 16a 0.02 0.5 33 sandstone 80 17 0.15 1.0 7 sandstone 90 18 0.1 0.5 9 till fill 90 90 19 0.05 0.5 20 sandstone terminates in forest vegetation 20 0.05 0.5 6 sandstone 90 21 0.05 0.5 7 sandstone 90 22 0.05 0.5 10 sandstone 90 23a 0.02 0.2 rock fill 90 23b 0.2 3.0 rock fill tension cracks due to movement of rock fill 24 3.5 6.0 12 sandstone 90 sinkhole (see Photo 8.4) 25 0.1 0.5 7 till terminates in forest vegetation 26 1.0 2.0 7 sandstone terminates in forest vegetation 27 0.1 0.5 6 sandstone terminates in forest vegetation 153 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH The size of the surface fractures was found to increase with decreasing depth of cover. Several large fractures over lm wide, 3 m deep, and 10m long occurred over Section 101; see Photo 8.3. No surface fractures were found to extend beyond the edges of the underground excavations. Photo 8.3: Open Fracture at the Quinsam Coal Mine 154 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 8.4 COMPARISON OF THE EXISTING SUBSIDENCE PREDICTION METHODS TO EXPERIENCE AT THE QUINSAM COAL MINE Several of the methods discussed in Section 8.2 were used to predict the height of the caved and fractured zones, and the angle of draw at the Quinsam Coal Mine. The results are summarised in Tables 8.3 and 8.4. Table 8.3: Prediction of the Height of the Caved Zone and the Fractured Zone Method Maximum Height of Fractured Zone or Discontinuous Subsidence (m) Maximum Height of Cave (m) C O M M E N T S Salamon(1976) 56 50 20 x excavation height Bai(1995) 33.6 12 x extraction height Whittaker and Reddish (1989) 28 based on bulking theory Garrard and Taylor (1988) 48 using a value of 18m for span Liu (1981) 33.6 Peng(1992) 55 to 110 28 Piggot and Eynon (1978) 28 10 x extraction height Note: Predictions used an excavation height of 2.8m and an excavation span of 90m. From Table 8.1 it is seen that at the Quinsam Coal Mine, the fractured zone typically extended to a height of less than 60m (20 x excavation height) but may extend to as high as 82m (30 x excavation height). The caved zone extended to a height of up to 40m (14 x excavation height). The criteria developed by Salamon (1976) and Garrard and Taylor (1988) provide the best matches to measurements at the Quinsam Coal Mine. Both of these methods are based on data from shallow depth mines. Garrard and Taylor limited their method to excavations with spans less than 18m, as this was the upper limit of the span size within their database. Most previous researchers, including Garrard and Taylor, have suggested that the upper bound for caving is 10 x excavation height. The data from the Quinsam Coal Mine clearly shows this to be an erroneous assumption. Caving at the Quinsam Coal Mine occurred to 14 x excavation height. 155 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH The methods of cave and fractured zone height prediction developed by Peng (1992), Liu (1981), and Bai (1995) are based on observations from deeper mines. As such, it is not unexpected that the results from these methods do not match observations at the Quinsam Coal Mine. It is apparent that at shallow depth mines, caving can extend much higher than at deeper mines. Based on the data collected from the Quinsam Coal Mine it also appears that the use of bulking theory to predict maximum caving height may be misleading at shallow depth. Table 8.4: Prediction of the Angle of Draw Method Depth of Cover (m) Angle of Draw (deg.) Angle of Critical Deformation (deg.) National Coal Board not specified 36 26.5 Stratham's Rule of Thumb not specified 8 US coal data base 50 6.0 - upper bound 3.5 -average 23.0 - upper bound 8.0 - average US coal data base 100 5.5 - upper bound 4.0 - average 20 - upper bound 6.0 - average The different methods predict a large range of values for the angle of draw and angle of critical deformation for depths less than 100m. Al l the methods predict an angle of draw that is greater than the average value determined at the Quinsam Coal Mine. The US Coal Data Base method values were closest to those measured at the Quinsam Coal Mine. The NCB predicted value is seen to be significantly greater than the values obtained at the Quinsam Coal Mine. The NCB method was not intended for shallow depth room and pillar retreat mining; see Section 8.2. The US Coal Data Base method contains data from 38 mines with a depth of cover less than 100m. No limitations with respect to mining method or geologic environment were given with this method. The Quinsam Coal Mine angle of draw measurements are plotted against the depth of cover on Figure 8.9. A best fit curve for the data is plotted along with the angle of draw prediction curves for the US Coal Database and NCB methods. The trend line for the Quinsam data suggests that the angle of draw is essentially constant at 0.6°. 156 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 8.9: Angle of Draw versus Depth of Cover o> , J o o o> 10 -NCB Angle of Critical Deformation • Quinsam Data —Linear (Quinsam Data) US Coal Data Base Angle of Draw • • y = -0.0018x +0.8363 60 60 70 Depth of Cover (m) None of the existing methods predicted angle of critical deformations consistent with observations at the Quinsam Coal Mine. Although the angle of critical deformation was not of its own measured, the angle must be less than the angle of draw. The angle of fracture is proposed as a reasonable approximation for the angle of critical deformation over shallow depth coal mines. The angle of fracture is a new term defined as the angle between a vertical line at the edge of the excavation and a line drawn between the edge of the excavation and the extent of surface fracturing. From Table 8.1, it is seen that the average angle of fracture at the Quinsam Coal Mine was -1.9°. The existing prediction methods significantly overestimated the angle of critical deformation, as determined from the angle of fracture. Several of the existing methods for the prediction of continuous subsidence imply a relationship between the width to depth ratio and surface ground movements. These methods also report a "critical" width to depth ratio. If the ratio is less than critical, the maximum possible vertical ground movements will not be reached. In addition, if the ratio is less than critical, the inflection point for the subsidence profile will be located outside the under mined area (over solid coal). The inflection point is the location where the horizontal strain (e) changes from compressive to tensile. Peng (1992) reports that the critical width to depth ratio is in the range 157 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH of 0.9 to 2.2. The subsidence data from the Quinsam Coal Mine indicates that there is no relationship between the width to depth ratio and any of the subsidence investigated. No critical span could be determined that governed the amount of vertical movement or the location of the inflection point. It was determined that below a critical span the excavation will remain stable and no collapse or subsidence will occur. At the Quinsam Coal Mine, when not influenced by geologic structure, the minimum effective radius at which caving occurred was 7.8m. The minimum effective radius at which surface ground movements occurred was determined to be 10.8m; this is equivalent to an excavation that is 43m by 43m. Figure 8.10 plots the subsidence factor "a" (maximum subsidence/extraction height) versus depth for the data obtained from the Quinsam Coal Mine, as well as for data obtained from the literature. Best fit curves have been plotted for the two data sets. It is apparent from this figure that there is a significant difference between maximum vertical subsidence observations at the Quinsam Coal Mine and those elsewhere. Vertical ground movements in zones of discontinuous subsidence are very irregular. As such, the measurement of vertical movement will also be irregular and dependent upon the location of the survey points. In the case of the Quinsam Coal Mine none of the survey points fell within open cracks or sinkholes; had such measurements been included the subsidence factor would have been higher. It is proposed that the magnitude of vertical subsidence can not be predicted with acceptable accuracy where discontinuous subsidence occurs. 8.5 NUMERICAL MODELING OF SUBSIDENCE Shallow depth ground movements and subsidence were investigated using the two dimensional finite difference program FLAC (ITASCA 1999). A brief discussion of the technical basis of FLAC is given in Chapter 3. Several researchers (i.e. Kay et al 1991, Noor et al 1997) have used FLAC to model subsidence over mines where the resulting surface subsidence was continuous in nature. There are no reported cases of using numerical codes of any kind to model ground movements over shallow depth retreat pillar coal mining operations where the surface subsidence is discontinuous in nature. 158 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 8.10: Subsidence Factor versus Depth of Cover ra 0.60 k. o t) S. 0.50 • • • • • • Literature data m • y = -0.1296Ln(x)+ 1.1701 —, * • • — • • • • Quinsam Mine Data 1 • • y = 4.7216x* 6 8 8 3 • 30 40 50 60 70 80 90 100 Depth of Cover (m) Very little is known about the actual processes of block movements associated with caving; this makes verification of model results very difficult. As well, most of the existing numerical models lack the ability to handle the large deformations and rotations associated with caving. Modelling of caving and subsidence is very much a "data limited problem" as described by Starfield and Cundall (1988); field data such as in situ stresses, material properties and the presence of discontinuities will never be completely known. ITASCA (1999) suggests that in these situations it is futile to expect that modelling will be accurate enough for detailed design work. They suggest that numerical modelling of data limited problems is an effective tool to carry out parametric studies and to provide a picture of the mechanisms that may occur in the physical system. The traditional way of creating excavations within finite difference models is to create a null or void zone. The initial FLAC model runs performed as part of this study were done in this way. However, it was found that once caving commenced the model would become 159 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH numerically unstable as large displacements occurred. To prevent numerical instability it was necessary to "trick" the model. Excavations were created by replacing the coal seam with an "undercut material". The stresses in the newly created undercut material were re-initiated to zero. The undercut material properties are given in Table 8.5. The modulus values assigned to the undercut were found to significantly affect the height of the yielded rock and the magnitude of the vertical movements within the overlying rock mass. O'Connor and Dowding (1992), and Szostak-Chrzanowski (1988) similarly found that the material modulus values were inversely proportional to the magnitude of the vertical ground movements. The process of caving was simulated by replacing yielded elements with a "caved material". After each model iteration, the state of every element within the model was checked. Based on the Mohr-Coulomb failure criteria, any element determined to have yielded was flagged. The flagged elements were then replaced with the caved material, and the stresses in the newly caved material were re-initiated to zero. The caved material properties are given in Table 8.5. The modulus values used were 1/1 Oth of the rock mass values. Szostak-Chrzanowski and Chrzanowski (1991) suggest a modulus reduction of 1/3 while O'Connor and Dowding (1992) suggest a reduction of between 1/10 and 1/100 for rock that has failed. The modelling process used in the simulations is illustrated in the flow chart shown in Figure 8.11. A shortcoming of the caving simulation method used is that all elements that yield have their material properties changed to that of caved rock. Not all elements that yield will actually cave. At a height of 10 X the excavation height interaction between the caved rock and overlying rocks is expected to occur. Elements that yield above this height should have their material properties changed to some value other than caved material. 160 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Table 8.5: Material Properties Used in Numerical Models Property Coal Seam Undercut Material Caved Rock Property Set#l Property Set #2 Property Set #3 Property Set #4 Elastic bulk modulus, K (Pa) 2.0e9 2.0e6 2.0e8 2.0e9 2.0e9 2.0e9 3.0e9 Elastic shear modulus, G (Pa) 1.0e9 1.0e6 1.0e8 1.0e9 1.0e9 1.0e9 2.0e9 Angle of friction, (|> (degree) 35 20 25 30 30 35 40 Mass density, p (kg/m3) 2600 1800 2000 2600 2600 2600 2600 Cohesion, c (Pa) 6.0e6 0 0 2.0e6 4.0e6 8.0e6 20e6 Tension Limit a ' (Pa) 6.0e6 0 0 2.0e6 4.0e6 8.0e6 20e6 Joint angle of friction, <j)j (deg) 20 25 25 25 30 Joint cohesion, Cj (Pa) 0 0.2e6 0.4e6 0.8e6 1.0e6 Joint tension Limit a\ (Pa) 0 0.2e6 0.4e6 0.8e6 1.0e6 Figure 8.11: Flow Chart of FLAC Modelling Method f s T A R T Set up simulation. Solve F L A C model to ensure initial equilibrium. Create undercut in 12m increments. Coal seam is replaced with "undercut" material. Re-initiate stresses to zero. Replace flagged elements with "caved" material Re-initiate stresses to zero Y I E L D Solve F L A C model Check each element for yield Flag all elements that have yielded since last run. NO Y I E L D r Save information from simulation Continue simulation with additional mining 161 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH The simulations were all run using the Mohr-Coulomb ubiquitous joint constitutive model. This model applies the Mohr-Coulomb shear failure criteria to both the solid mass and to planes of weakness imbedded in the solid. The weak planes are assigned a direction as well as material properties, and are assumed to be present at every location throughout the solid. Yield may occur in the solid and/or on a plane of weakness, depending upon the stress state and assigned material properties. A Mohr-Coulomb failure surface (fs) is defined in FLAC by equation 8.10. Yield occurs if the value offs is determined to be less than 0 within any element in the model. fs = o-i - 0-3(1 + sin4>)/(l - sin<|) ) + 2c V((l + sin(j))/(l - sindj )) (8.10) where: a 1= major principal stress (within FLAC compressive stresses are negative) o"3= minor principal stress c= cohesive strength of the bedding plane or joint surface 4>= angle of friction Tensile yield may also occur within the model if 0-3 is equal to or greater than the tensile strength assigned to the solid or to the planes of weakness. Four different rock mass material property sets were used as input parameters for the numerical simulations. The values used are listed in Table 8.5. The four property sets were selected to approximate the following rock types: Property Set # 1: very weak Property Set #2: weak Property Set #3: moderate Property Set #4: strong The weak rock is considered to approximate a siltstone from the Quinsam Coal Mine. The values were determined by calibrating the model to the observed response of underground excavations at the mine. The input parameters were adjusted until the model predicted 5mm of roof closure, with no failure, in an 8m wide excavation. 162 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Over 40 numerical simulations were carried out to assess the effects of depth of cover, excavation span, rock mass properties, and in situ stress. The resulting subsidence parameters investigated with the models were the angle of draw, angle of fracture and angle of critical deformation. The angle of draw was determined as the point on the surface where vertical deformation reduced to 0.01m. The angle of fracture was determined as the point on the ground surface where no further yielded elements occurred. The angle of critical deformation was determined as the point on the ground surface where the horizontal strain became less than 0.002 m/m. Table 8.6 summarises all the results from the study. Figures 8.12 to 8.14 present examples of the graphical output from the FLAC simulations. Figures 8.12 and 8.13 show the simulations for a 24m wide and a 36m wide excavation using property set #3. The figures show the extent and type of element yield as well as contours of vertical displacement. In Figure 8.12, the height of yielded elements is 15m; this is analogous to the cave zone height. No ground movements occurred at surface. When the excavation span was increased to 36m (Figure 8.13), the yield zone extended to surface, and surface subsidence occurred. The surface ground movements were limited to the area directly above the excavation. The angle of draw, angle of fracture and angle of critical deformation were all less than -1.5 degrees. Both Figures 8.12 and 8.13 show the formation of sub-vertical shear failure boundaries at the edge of the excavations. These boundaries define the extent of the yield zone as well as the extent of significant ground movements. "Orphaned zones" of non yielded elements occur within the yield zone. Verification runs to evaluate the impact of the orphaned zones were carried out. The orphaned zones were converted to caved material. In some cases, it was found that the orphaned zones had no effect on the final subsidence; in other cases, converting the orphaned non-yielded elements to caved material resulted in significant changes to the surface ground movements. The extent of orphaned zones was found to decrease as the strength values of the material were decreased; see Figure 8.14. 163 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Table 8.6: Summary of Numerical Model Results Run Depth (m) Width (m) Stress Ratio Property Set Angle 1 Draw Angle 2 Fracture Angle 3 Critical Comments qr22 40 36 1:0.5 l -5.0 -5.0 -5.0 qr23 40 72 1:0.5 l -5.0 -5.0 -5.0 qr24 40 108 1:0.5 1 • -5.0 -5.0 -5.0 qrl3 80 36 1:0.5 l -1.4 -1.4 -1.4 qrl4 80 72 1:0.5 l -3.6 -3.6 -3.6 qrl5 80 108 1:0.5 1 -3.6 -3.6 -3.6 qr2 80 72 1:1 l 0 0 0 qr3 80 108 1:1 l -2.0 -2.0 -2.0 qtl-72 120 72 1:1 l 7 6 6.0 qtl-108 120 108 1:1 l -2 -3 -1 qs2-36 40 36 1:0.5 2 -14 -14 -14 qs2-72 40 72 1:0.5 2 -12 -12 -12 qs2-108 40 108 1:0.5 2 -12 -11 -10 qrl3a 80 36 1:0.5 2 -5.3 -5.3 -5.3 qrl4a 80 72 1:0.5 2 -1.4 -1.4 -1.4 qrl5a 80 108 1:0.5 2 -2.9 -2.9 -2.9 qrl6a-l 160 72 1:0.5 2 35 -4 0 qrl6a-2 160 108 1:0.5 2 33 -2 -1 qrl7a 160 144 1:0.5 2 36.8 -3.6 -3.9 qrl8a 160 216 1:0.5 2 33.0 -2.9 -7.8 qs3-72 80 72 1:1 2 -2 -2 -0.5 qs3-108 80 108 1:1 2 -1 -0.5 -1 qt2-36 120 108 1:1 2 6 3 4.3 qv3-36 40 36 i;i 3 -11 - -No breakthrough. No tensile strain greater than 0.0005. Cave height 31 m. qv3-72 40 72 i;i 3 -11 -26 -26 qv3108 40 108 i;i 3 -14 -12 -12 qu3-36 80 36 1:1 3 -3.0 -3.0 -3.0 No tensile strain greater than .0005 qu3-72 80 72 1:1 3 12.2 -2.5 -1.0 qu3108 80 108 1:1 3 37 -15 -12.0 qp8-36 95 36 1:1 3 11 11 No tensile strain greater the .002 qp8-72 95 72 1:1 3 1.5 1 1.2 qp8108 95 108 1:1 3 -9.5 -9.5 -9.5 qv4-36 40 36 1:1 4 No subsidence greater than 0.0001. No break through to surface. No tensile strain greater than 0.002. No cave. qv4-72 40 72 1:1 4 0 No break through to surface. No tensile strain greater than 0.0002. No cave. qv4108 40 108 1:1 4 10 No break through to surface. No tensile strain greater than 0.0005. No cave. qu4-36 80 36 1:1 4 No subsidence greater than 0.005. No break through to surface. No tensile strain greater than 0.0001. No cave. qu4-72 80 72 1:1 4 9.2 - -No break through to surface. No tensile strain greater than 0.0005. No cave. qu4108 80 108 1:1 4 14 - -No break through to surface. No tensile strain greater than 0.0005. No cave. qt4-36 120 36 1:1 4 - -No break through at surface. No subsidence greater than 0.005m. No tensile strain greater than 0.0001. Cave height is 0m. qt4-72 120 72 1:1 4 14 - -No break through at surface. No tensile strain greater than 0.0001. No cave. qt4-108 120 108 1:1 4 22.5 - -No break through at surface. No tensile strain greater than 0.0001. No cave. 1 The angle of draw was determined as the location where the vertical subsidence was less than 0.01m. 2 The angle of fracture was the determined as the location where failure reached the ground surface. 3 The angle of critical deformation was determined as the location on the ground surface where the horizontal ground strain became less than 0.002 m/m 164 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 8.12: Example of Graphical Output from FLAC. JOB TITLE: 40m depth, 24m undercut, property set #3 FLAC (Version 3.40) LEGEND 11-Nov-1 10:54 step 47021 -7.778E+00 <x< 4.140E+01 -4.442E+01 <y< 4.761E+00 state Elastic iHl At Yield in Shear or Vol. •Elastic, Yield in Past Slip Along Ubiq. Joints Grid plot i 0 1E 1 Y-displacement contours Contour interval 5.00E-02 Minimum: -9.50E-O1 Maximum: 0.00E+00 Michael Cullen & m & % m m~: mt. ma mt mi i - i 1.300 POM) Figure 8.13: Example of Graphical Output from FLAC. JOB TITLE: 40m depth, 36m undercut, property set #3 FLAC (Version 3.40) LEGEND 11-Nov-l 12:12 step 146554 -8.317E+00 <x< 4.086E+01 •4.498E+01 <y< 4.195E+00 state 8Elastic At Yield in Shear or Vol. Elastic, Yield in Past Slip Along Ubiq. Joints Grid plot ' 0 1E 1 Y-displacement contours Contour interval 5.00E-02 Minimum: -9.50E-01 Maximum: 0.00E+00 Michael Cullen -0.900 0.000 0.SX 1.000 1.300 POM) 2.000 2.300 3 000 3.SO0 <0O0 165 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 8.14: Example of Graphical Output from FLAC. JOB TITLE : 40m depth, 4Bm undercut, property set #1 F L A C (Version 3.40) LEGEND 13-Nov-l 10:10 step 176988 -9.152E+00 <x< 4.116E+01 -4 498E+01 <y< 5.326E+00 state Elastic At Yield in Shear or Vol. J Elastic. Yield in Past Slip Along Ubiq. Joints Displacement vectors Max Vector = 3.823E-01 i i 1E 0 Boundary plot 0 1E 1 Michael Cullen Figure 8.14 shows the extent of the yield zone and elements displacement vectors for a 48m wide undercut, at 40m depth, with property set #1 (very weak rock). The surface ground deformations were limited to the area directly above the excavation. The angle of draw, angle of fracture, and angle of critical deformation were all less than -3.0 degrees. From the displacement vectors it was seen that there is essentially no ground movement beyond the shear yield boundaries. The numerical simulations indicated that failure initiates by shear yield at the edge of the excavations. As discussed in Chapter 3, stress concentrations are greatest at the edge of the excavations. When an element yields, the stresses are transferred. The element directly above the yielded element will now experience an increase in stress concentration and may also fail. This process results in the formation of the sub-vertical shear boundary. The formation of a boundary of maximum shear was first proposed by Kratzsch (1983) who suggested that this boundary controls the extent of tensile stresses over an excavation. 166 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Analysis of the FLAC simulations revealed that there is no relationship between the excavation span and the angle of draw or angle of fracture. Below a critical span the model excavation remained stable and no collapse or subsidence occurred. The critical span was dependent upon the rock mass properties and the induced stresses. These results are consistent with measurements from the Quinsam Coal Mine, as well as records from the literature. As the ratio of horizontal to vertical stress was increased, the angle of draw, angle of fracture, and angle of critical deformation all increased. At the same time the maximum subsidence determined by the model was seen to decrease. The FLAC results clearly showed a relationship between the depth of cover and the angle of draw, angle of fracture, and angle of critical deformation. There was also a relationship between the rock strength and the angle of draw. The results from Table 8.6 are plotted in Figures 8.15 and 8.16 as the angle of draw versus the depth of cover, and the angle of fracture versus the depth of cover. Best fit curves have been plotted for each of the different rock property sets used in the models. Figure 8.15 and 8.16 clearly show a trend of increasing angle of draw and increasing angle of fracture with increasing depth of cover. These results are not consistent with measurements from the Quinsam Coal Mine but they do agree with general trends reported in the literature. At the Quinsam Coal Mine, the angle of draw and angle of fracture do not appear to be dependent upon the depth of cover where depth is less than 30 times the excavation height. The angle of draw had an average value of approximately 0.6° while the angle of fracture had an average value of-1.9° at these depths. Figure 8.15 shows that the angle of draw is dependent upon the strength of the rock. As the strength of the rock increases so does the angle of draw. Within the literature there are conflicting reports of the influence of rock strength on the angle of draw. The numerical modelling results suggest that the angle of fracture is not dependent upon rock strength; see Figure 8.16. 167 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 8.16: Angle of Fracture Versus Depth of Cover, k=1.0 15 10 5 0 i "5 ? -10 -15 -20 -25 -30 30 Rock Property A • Set#1 • Set #2 A Set #3 * . g Set#1 .tT-"5**] — • ,A Set #2 A /"set #3 / A • 50 70 90 Depth (m) 110 130 168 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH The modelling showed that at depths of cover less than 80m, the angle of critical deformation was equal to the angle of fracture. At depths of cover less than 100m, the angle of critical deformation may be greater than the angle of fracture; however, the difference is less than two degrees. These findings verify that, at shallow depth, the angle of critical deformation can be approximated by the angle of fracture. The numerical simulation data in Table 8.6 was analysed using the artificial neural network (ANN) program "Braincel" (Promised Land Technologies, 1996). ANN analysis is analogous to a crude low level biological neural system. The analysis method involves training the model with a historical set of input and output data. During training, or learning, the weight of each input parameter is iteratively adjusted until a statistical best fit model is achieved. ANN are ideal for pattern recognition in complex problems, such as ground movements, where the exact nature of the relationship between variables is not known. An ANN predictive model for the angle of draw was developed using depth, width, and rock properties as input variables. The model was used to back analyse the same data that was used as the input. The coefficient of determination was found to be only 0.69; the value should be close to 1.0 for such a back analysis. Similar results were obtained for models developed for the angle of critical deformation and the angle of fracture. The main reason for the poor results is believed to be insufficient input data. The minimum number of truthed historical data sets to train an ANN on is generally considered to be 100. As the complexity of the problem increases, or the number of input variables increase, so does the number of historical data sets required to produce a good model. Anomalous results from the numerical models are also considered to be a contributing factor in the inability of the ANN to develop a good model. They also account for the data spread and low correlation of determination in the linear analysis. Several of the anomalous results encountered while carrying out the numerical modelling are listed below: 169 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 1. Large angles of draw and angles of fracture occasionally occurred when the depth of cover was less than 100m. This phenomenon was more common with smaller excavation spans. The anomalous angles would usually disappear as the excavations were enlarged. The large angles are not consistent with the results of the majority of the model runs nor field measurements. 2. Small changes in the input parameters occasionally resulted in large and unexpected model results. 3. Inconsistent results from the analysis of orphaned zones. Several simulations were re-run with orphaned zones located and converted to caved material. In some instances, there was no change in the model results; in other cases, there were significant changes in the results. Over 60 FLAC verification and calibration runs were carried out to investigate the affects of material properties, orphaned zones, constitutive model, and model set up. The results from these runs are summarised in Appendix 5. The verification runs were unable to account for all the unusual results. Several possible causes for the unexplained model inconsistencies and unusual phenomena are suggested below: 1. The method used to simulate the caving process is not appropriate. 2. The caving problem is three dimensional in nature. 3. Caving and discontinuous subsidence are chaotic and sensitive. The model is simply a reflection of this reality. ITASCA (1999) reports that in some models, particularly those involving discontinuous materials, the results can be extremely sensitive to very small changes in the initial conditions or trivial changes in loading conditions. Such systems are termed chaotic. The observed sensitivity of the computer model is simply a reflection of similar sensitivity in the real world. 170 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 8.6 INVESTIGATION OF STRESSES AROUND THE CAVING FRONT The numerical model FLAC was used to investigate the stresses that occur around the caving front. The analysis considered a 60m wide excavation at depths of 50m and 200m in an elastic medium. In Chapter 3 it was shown that the compressive horizontal stresses in the roof decrease as the depth of cover decreases. At a span of 60m and depth of 50m, the horizontal stresses in the roof become tensile. The caved zone was modelled by creating a 20m high void. This approach is valid up to a height of approximately 10 x the excavation height.' This is the height that interaction between the caved rock and uncaved rock is expected to occur (choking off). Simulations were carried out for both wedge (triangular) and rectangular shaped caves. The results are summarised in Table 8.7. Both the minimum (<Jmm) and maximum (amax) induced stresses are reported along with the horizontal stress that occurs in the roof at the centre of the excavation (o-hor). Graphical output showing the orientation and relative magnitude of the principal stress, along with contours of the tensile stresses around rectangular and wedge shaped caves, is shown on Figures 8.17 and 8.18. As discussed in Chapter 3 the centre of the excavation is the location where the horizontal stresses are expected to be tensile. This was found to be true for the rectangular shaped caves but not for the triangular shaped caves. Table 8.7: Induced Stresses above the Caved Zone Determined from Computer Simulations. (The numbers in brackets are the distance, in metres, to which tensile stresses extend above the roof) Depth (m) In Situ Stress (MPa) 60m Span 2m High Excavation Stress (MPa) 60m Span 20m High Rectangular Cave Stress (MPa) 60m Span 20m High Triangular Cave Stress (MPa) ^rnax Chor ^max ^min C>hor ^max ^min <?hor 50 1.3 9.7 -0.2 (30) -0.1 (1) 6.8 -0.5 (30) -0.5 (6) 7.7 -0.1 1 2.5 200 5.1 37.2 -0.1 (10) 0.6 32.4 -0.07 (9) 1.2 37.4 -0.011 • 2.8 1 The minimum stress values were tensile; however, they were located in pockets on the sides of the triangle as opposed to over the center of the excavation as occurs with a rectangular shape. At the center of the excavation, a m i n was found to be compressive. 171 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH In the case of the triangular shaped caves, <jmm was found to be tensile in pockets on the sides of the triangle and compressive at the apex as shown on Figure 8.18. In all cases, it was found that the maximum compressive stresses occurred near the edge of the excavations. The orientation of these stresses was near vertical. From Table 8.7 it is seen that for rectangular caves at 50m depth the magnitude of the tensile stresses (o-hor and o~mjn) increase as the cave height increases. This condition would contribute to greater instability and caving. At a depth of 200m the magnitude of the tensile stresses (ami n) decrease in magnitude as a cave develops. The ahor stresses, which are strictly compressive, increase in magnitude as a cave develops. These conditions would contribute to increasing stability. As such, from a stress perspective, caves at shallow depth are more likely to propagate upwards than caves at deeper depth. At both 50m depth and 200m depth it was seen that a wedge or triangular shape is more stable than a rectangular shape. The peak compressive stresses (amzx) at the edge of the excavations are similar for both shapes. Both o-hor and Gmm at the centre of the excavation are less with a rectangular shape. Figure 8.17 and 8.18 show contours of tensile stress around a rectangular and wedge shaped void at a depth of 50m. It is evident that the tensile stresses are much greater, therefore stability is less, around the rectangular shaped cave. Tensile stresses are greatest in the centre of the rectangular shaped caves. As such, failure is most likely to occur in the centre. If failure were restricted to the centre of the excavation the shape of the cave may change from rectangular to triangular. The stress analysis showed that low magnitude tensile stresses do occur over the sides of a triangular shaped cave. If these stresses are sufficient to cause failure, then the shape of the cave would revert back to rectangular. The potential for these tensile stresses to cause failure would be greatest where sub-horizontal planes of weakness, such as bedding planes, exist. As such, it is expected that in horizontally bedded strata, the cave zone will maintain a rectangular shape. 172 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Figure 8.17: Contours of Tensile Stress, and Principal Stress Orientation and Magnitude around a Rectangular Shaped Cave JOB TITLE : SOm Depth, 60m Span, 20m High Rectangular Cave FLAC (Version 3.40) LEGEND 25-Nov- 1 15:20 step 40062 9 574E+00<x< 8.882E+01 -6.598E*01 <y< 1.326E+01 Boundary plot 0 2E 1 Principal stresses Max Value = 6.847E+06 linuiuiWmuud 0 2E 7 Minimum principal stress Contour interval 2.50E*05 Minimum: 0.00E+00 Maximum: 1.00E+06 Michael Cullen i fflmm Figure 8.18: Contours of Tensile Stress, and Principal Stress Orientation and Magnitude around a Wedge Shaped Cave JOB TITLE: 50M DEPTH, 60M SPAN, 20M TRIANGULAR CAVE 173 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 8.7 CONCLUSIONS This investigation into caving and subsidence has clearly shown that the ground movement response of a rock mass at shallow depth is quite different to that at deeper depths. For subsidence investigations shallow is defined as the height to which discontinuous surface subsidence will occur. At the Quinsam Coal Mine it was determined that, where not influenced by faults, discontinuous subsidence will occur up to a height of 30 times the excavation height. Most of the existing methods for predicting subsidence were developed at deeper mines where continuous surface subsidence occurs; these methods were found not to be applicable at shallow depth. Discontinuous subsidence occurs where the caved or fractured rock extends to surface. It is next to impossible to protect surface facilities from damage, or to prevent surface water inundation, where discontinuous subsidence occurs. As such, at shallow depth the most important consideration is to what extent discontinuous subsidence extends out beyond the limits of the underground excavation. The term "angle of fracture" was coined to define the extent of discontinuous subsidence over shallow mines. It is defined as the angle between a vertical line at the edge of the excavation and a line drawn between the edge of the excavation and the extent of surface fracturing. This research has shown that, at shallow depth, the angle of fracture is analogous to the angle of critical deformation, this is the angle used to define the extent of damaging ground movements in deeper mines. Any ground movements that occur beyond the angle of fracture are expected to be very small in comparison to those occurring within the angle of fracture. The angle of fracture at shallow depth mines is expected to be very small or even negative. At the Quinsam Coal Mine the angle of fracture was determined to have an average value of -1.9°. This implies that significant ground movements only occur directly above the mine excavation. Ground movements over shallow mines are seen to be extremely limited when compared to the large trough subsidence basins that occur over deeper coal mines. It must be 174 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH noted that large ground movements or fracturing may still occur beyond the edge of the excavations where adverse faulting is present, or where the overburden is comprised of a significant thickness of soil or very weak rock. Analysis of these situations must be carried out on a case specific basis. At very shallow depth caving may propagate through to surface. When this happens sinkholes and large open fractures may appear. Many researchers have suggested that caving will not occur where the ratio of depth of cover to excavation height exceeds 10. Engineers commonly use this value for design purposes. This research has shown that an upper bound of 10 times the excavation height is incorrect. At shallow depth caving may extend to a height of up to 14 times the excavation height. Where faulting is present the height to which caving occurs may be greater yet. Existing subsidence prediction tools relate the magnitude of the vertical subsidence (S) to the depth of cover and extraction height. Determination of S is needed for the prediction of other ground deformation parameters such as slope and curvature. This research has shown that S cannot be predicted with meaningful accuracy where discontinuous subsidence occurs. At deeper mines it has been found that there is a strong relationship between the excavation width to depth ratio and ground subsidence. The concept of a critical width to depth ratio does not appear to be valid over shallow depth mines where discontinuous subsidence occurs. The results from the numerical modelling, as well as from the data collected from the Quinsam Coal Mine, indicate that there is no relationship between the width to depth ratio and any of the subsidence parameters investigated. Induced stresses are the main reason for the significant differences between shallow and deeper depth caving and subsidence. Numerical modelling has shown that the magnitude and extent of the tensile stresses increases with decreasing depth of cover and increasing excavation width. It was also found that as the cave height increases, the magnitude of the tensile stresses will increase over shallow excavations, but they will decrease over deeper excavations. 175 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH If the shape of the cave changes from rectangular to triangular the stresses over the centre of the excavation may change from tensile to compressive. This may result in stabilisation of the cave. Where sub-horizontal planes of weakness (such as bedding planes) are present the cave is expected to remain essentially rectangular; tensile stresses that occur on the sides of the triangle are expected to induce structurally controlled gravity failures thus restoring the rectangular shape. The numerical simulations carried out indicated that shear failures will occur at the edges of an excavation where stresses are concentrated. If the concentrated stresses cause a model element to fail (yield in shear), the stresses will be transferred to the next element up, which in turn may fail. This process creates a line or boundary of shear failures similar to the boundary of maximum shear stress proposed by Kratzsch (1983). Within the boundary of shear failure, the models show that the rock yields. Outside the boundary of shear failure there is little or no yield or ground deformation. This concept of yielded and non yielded zones is illustrated in Figure 8.19. The zone of yield is equivalent to the caved and fractured zones in the field. Discontinuous subsidence will occur if the caved and fractured zones extend to surface. Based on the data from the Quinsam Coal Mine this will occur where the depth to height ratio is less than 30. At deeper mines the yield zone does not extend to surface. The yield zone is overlain by a zone of continuous deformation with limited yield, as illustrated in Figure 8.19B. Ground subsidence within the zone of continuous deformation is continuous in nature. At shallow depth it is suggested that the shear failure boundary, defining the caved and fractured zones, can be approximated by the angle of fracture. The extent of the zone of continuous deformation above the caved and fractured zone is defined by the angle of draw (8). The results from the numerical modelling were found to be in general agreement with the field measurements from the Quinsam Coal Mine. In particular, the numerical modelling results support the finding that significant surface deformations do not occur beyond the edge of the underground excavation over shallow depth retreat pillar mines. The hypothesis that 176 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH the angle of critical deformation can be approximated by the angle of fracture at shallow depth mines was also confirmed by the numerical modelling. Numerical modelling proved to be a useful tool for studying the process of shallow depth caving and subsidence. However, at this time it cannot be used on its own as a design tool. The work carried out indicated that there are several outstanding problems with numerical simulation of shallow depth caving and subsidence. The model must be "tricked" in order to handle the large displacements associated with caving. The validity of this method to simulate the physical process of caving and subsidence could not be validated. The numerical model occasionally returned anomalous results. The best explanation for the anomalous results is that caving and subsidence are sensitive and chaotic processes; as such, model results may also be sensitive and chaotic. There remains much scope for improvement of numerical models and modelling techniques for shallow depth caving and subsidence. The most important improvement required would be the ability to realistically model the large deformations and block rotations associated with caving. If this were achieved the need to "trick" the models to simulate caving by changing material properties could be eliminated. 177 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH ;ure 8.19: Concept of Yielded and Non Yielded Zones ZONE OF INTACT ROCK WITH ' MINIMAL DEFORMATION DISCONTINUOUS SURFACE SUBSIDENCE > - ^ T '//////////, Angle of fracture lines Boundaries defined by shear yeild CAVED & FRACTURED ZONES (ZONE OF YIELOED ROCK) UP TO 30M V///////. M -A: Shallow Depth (relative to M) 178 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH CHAPTER 9 CONCLUSIONS AND RECOMMENDATIONS 9.0 INTRODUCTION This thesis has presented the results of a study into geotechnical aspects of retreat pillar coal mining at shallow depth. The geotechnical aspects investigated were excavation stability, excavation support, pillar design, gob cave prediction, caving, and subsidence. Shallow depth is generally defined as being less than 100m; in terms of subsidence prediction it is defined as being less than 30 times the height of the excavation. A value of 30 times the height of extraction was determined to be the upper limit for discontinuous surface subsidence to occur. The field work component of this research was carried out at the Quinsam Coal Mine located on Vancouver Island, Canada. The properties of the rock and rock mass at the Quinsam Coal Mine were determined to be similar to other mines working in similar geological environments. As such, it is believed that the findings from this research can be applied successfully to other shallow depth mines operating in similar geological environments. Most of the existing geotechnical design tools for retreat pillar coal mines were developed at mines operating at depths greater than 100m; the suitability of these tools to shallow depth mines was previously not known. The primary objectives of this research were therefore as follows: 1. To determine if the existing geotechnical design tools were applicable at shallow depth. 2. To develop new design methods where the existing tools were found to not be suitable. 3. To improve safety, productivity, and costs, at shallow depth retreat pillar coal mines. Sections 9.2 to 9.7 of this chapter summarises the results from the investigation into excavation stability, excavation support, pillar design, gob cave prediction, caving and subsidence. Section 9.1 presents a summary of the differences in the mining induced stresses at shallow and deep mines; it is these differences that account for the distinct response of the rock mass to mining at depths less than 100m. Contributions to the advancement of 179 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH knowledge are summarised in Section 9.8, and recommendations for further work are given in Section 9.9. 9.1 MINING INDUCED STRESSES In shallow depth coal mines the magnitude of both the vertical and the horizontal in situ stresses are generally low in relation to the rock strength. The vertical in situ stress is typically derived solely from the overburden load, while the horizontal in situ stress is expected to be approximately equal to the vertical stress. Numerical simulations using the above noted in situ stresses showed that there are significant differences between the mining induced stresses at shallow and deep mines. Most notable is the finding that as the depth of cover decreases the size of the stress relief zone above the excavation increases. The stress relief zone consists of low magnitude compressive and tensile stresses. Using an elastic model, the horizontal stresses over an excavation were found to become tensile at a depth of 50m and excavation span of 60m. Sub-vertical tensile stresses occur over most excavations greater than 10m wide (except at the excavation perimeter where only tangential stresses exist). The numerical simulations showed that the maximum induced stresses occurred near the corners of the excavation. The minimum induced stresses occurred above the centre of the excavation. The maximum stress was found to be directly related to the depth of cover. The minimum stress was also directly related to the depth of cover where it was compressive; however, where it was tensile, no relationship to depth was apparent. Where low compressive or tensile stresses exist along with geologic structure, gravity induced structurally controlled failures such as block falls, wedge falls, bed separation, beam bending, and progressive unravelling are most likely to occur. These types of failures are illustrated in Figure 3.9. The numerical modelling carried out clearly demonstrated that the mining induced stresses at shallow depth are low or even tensile, such that the potential for 180 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH gravity induced, structurally related failures will be greater at shallow mines than at deep mines. Gravity related block and wedge failures may also occur in high stress environments. However, where inclined joint and fault planes are present, higher stresses may reduce the potential for gravity failures. It can be shown that under even moderate stresses (greater than 1.0 MPa) joint planes with inclinations greater than 45° to the horizontal will be stable provided that the effective angle of friction is greater than 45° . If the normal stresses across the joint planes are very low, (i.e. at shallow depth) wedges may fall regardless of the effective angle of friction. At deeper depths, the mining induced compressive stresses may exceed the strength of the rock. Numerical modelling showed that the maximum induced stress around a 10m wide excavation at 50m depth is only 3.5 MPa.; at 400m the maximum induced stress is 27.3 MPa. Numerical simulations using a Mohr-Coulomb constitutive model showed that the failed zone around a 10m wide excavation doubles in size between a depth of 50m and 400m. Although both excavations are the same size with the same rock properties, the response of the excavations and the support requirements would be quite different. Figure 3.9 illustrates the differences in rock mass response under a low stress and high stress environment. 9.2 EXCAVATION STABILITY The low stresses found in shallow depth mines tend to favour structurally controlled, gravity induced failures, such as wedge falls, block falls, bed separation, beam bending and progressive unravelling. As such, geologic structure is the most important consideration to stability in shallow depth mines. Stress related instability is seldom a problem except where very weak rocks are present. The orientation of the mine excavations in relation to geologic structure will have significant stability implications; excavations that run parallel to dominant geologic structure will have 181 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH more stability problems than those that cross cut the structures. It is critical that geologic structure be evaluated in the planning and pre-development stage of a mine such that mine designs can be developed to minimise the number of excavations that run parallel to geologic structure. The orientation of the mine excavations to the principal horizontal stress direction is unlikely to be a significant factor in stability due to the low magnitude of the stresses. Excavation convergence measurements are an effective way of assessing the stability of excavations where the potential instability is caused by wedge falls, block falls, bedding plane separation and beam bending. Where failure is due to progressive unravelling, the effectiveness of excavation closure as a means to evaluate stability is questionable; the incremental movements associated with progressive unravelling are likely too small and too irregular for stability evaluation using convergence measurements. At the Quinsam Coal Mine it was determined that when roadways are first excavated and supported the convergence rate may be as high as 0.005mm/min (7.2mm/24hour). The convergence rate for stable excavations diminishes quickly. Within 24 hours of being supported the closure rate in stable excavations is expected to be less than O.OOOlmm/min (0.14mrn/24hour). Based on limited measurements of unstable roadways it is suggested that, at the Quinsam Mine, where the rate of convergence is increasing, the critical rate of convergence for roadway excavations should be approximately 0.0035mm/min. Convergence rates greater than 0.0035mm/min (5mm/24hour) may indicate the onset of instability. The critical convergence rate at coal mines is seen to be significantly greater than that at hard rock mines, which is typically around 0.0007mm/min (lmm/24hour). Roadway excavations at the Quinsam Coal Mine were found to remain stable right up to the edge of the gob. Minor movement associated with bed separation may occur above the excavation as the gob approaches; however, this does not necessarily indicate the onset of instability. The effect of stress and depth of cover on convergence rates and excavation stability are not well defined. Within the literature, some authors maintain stress and depth are factors (i.e. 182 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Ghosh and Ghose 1992) while others maintain they are not factors (i.e. Pothini and Schonfeldt 1978). Results from the Quinsam Coal Mine are inconclusive, suggesting that depth and horizontal stress are not important factors in determining the critical convergence rate for excavation stability at shallow depth. 9.3 EXCAVATION SUPPORT DESIGN Support design at shallow depth mines should be based on both a rock mass and a rock structure method. As with excavation stability, geologic structure is the most important consideration for support design in shallow depth mines. Support based on the rock mass is also important, especially if systematic pattern bolting is to be used. Where the depth of cover is less than 100m the induced stresses will be low and stress induced failures are not likely to be a significant issue. However, it must be recognised that the effects of stress are proportional to the strength of the rock mass; the potential for stress related instability is greater in weaker rock, i.e. mudstone, weak siltstone, and rocks weakened by weathering processes. As the depth of cover decreases, the stresses over the excavation will decrease and may become tensile. This situation favours gravitational type failures such as bed separation, beam bending, unravelling, and block type failures. Gravity induced failures are expected to be the most prevalent type of failure at shallow depth coal mines where well defined geologic structure is present. Detailed analysis is needed where geologic structure forms well defined blocks and wedges. The UNWEDGE computer program developed by Carvalho et al (1994) was determined to be an excellent tool for visualisation of wedges and determination of support requirements to secure the wedges. This research has shown that most of the existing rock mass support design methods overestimate support requirements at shallow depth coal mines. There are three reasons for this: • The methods do not generally consider depth of cover or stress. 183 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH • Most of the methods were developed for uses other than at coal mines. • Most of the methods were developed using data from moderate to deep excavations. Only the French Matrix Method (Newson 1987) predicted the support levels found to be effective at the Quinsam Coal Mine where the depth was less than 100m. The French Matrix Method is the only rock mass support design method that was developed specifically for use in coal mines that also considers mining induced stresses. Support designs based on most other methods may provide a safe excavation, but not a cost effective excavation. 9.4 GROUND SUPPORT WITH ROCK BOLTS Rock bolts have become the primary means of ground support at most underground coal mines in North America. A study of the effectiveness of point anchor rock bolts was carried out at the Quinsam Coal Mine. Point anchor rebar rock bolts were found to be a suitable means of ground control at the Quinsam Coal Mine. In siltstone that is dry, the average resin bond capacity was determined to be 24 tonne/m. If 20mm diameter grade 60 rebar is used in a 28mm diameter hole the optimum resin encapsulation length is 0.88m; this is the resin length required to hold the ultimate strength of the rebar (21 tonne). In weak siltstone that is wet, the resin bond capacity was found to be less than 12 tonne/m. Yield at the resin-rock interface began at loads as little as 4 tonne. The ability of resin anchored rebar to provide adequate anchorage in weak siltstone is questionable. Where weak siltstone is present it is recommended that the bolt length be increased to anchor into competent rock. If this is not possible and there is potential for the rock to become wet, the support design should utilise a design load of only 4 tonnes per bolt. In weak rocks, consideration must also be given to the problem of unravelling and weathering around the rock bolts. 184 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH This research has shown that bolt tension can be used for quality assurance testing. At the Quinsam Coal Mine, rock bolt tension and load were found to remain essentially unchanged throughout the life of all stable excavations. It was determined that if tension bleed off exceeds 20% either there is anchor slippage or the bolt was improperly installed. 9.5 PILLAR DESIGN At shallow depth mines, the stress on pillars will not generally be high in relation to pillar strength. The exceptions are pillars adjacent to the gob and small isolated pillars. Based on observations and measurements at the Quinsam Coal Mine, the following conclusions on pillar performance at shallow depth retreat pillar mines have been drawn: 1. Pillars with a minimum dimension of 12m will be stable, even at the edge of the gob. 2. Pillar deterioration increases with proximity to the gob. Deterioration may cause local safety concerns; however, it seldom poses any difficulty to pillar extraction. Pillar failures to a depth of 0.5m have no significant impact on overall pillar performance. 3. Pillar deterioration generally only occurs within 40m of the gob. This result is consistent with the findings of Mark (1990) who reports that 90% of the abutment stresses are carried by the pillars within a distance of 5 V(depth of cover) of the edge of the gob. 4. Where adverse geologic structures are present, the extent of pillar deterioration may be greater and conclusions 1, 2 and 3 above may not apply. 5. The presence of a sub-horizontal mudstone parting in the pillar may result in pillar skin failure and local safety concerns; however, it does not have a significant impact on overall pillar performance. 6. Load sharing occurs between adjacent pillars. Pillar stability calculations should be based on the strength of pillar groups, not on individual pillars. A review of existing pillar design tools determined that the ARMPS program developed by the USBM (Mark et al 1995) is the most suitable method for shallow depth retreat pillar coal mines. The following pillar design recommendations were developed for use at these mines. 185 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Barrier Pillars 5. Utilise a minimum width to height ratio of 10 if pillars are required to be indestructible. 6. Utilise a minimum width to height ratio of 4 if yielding is acceptable. 7. Utilise a minimum ARMPS factor of safety of 1.73. 8. Increase the pillar size to account for significant faulting that transects the pillar. 9. Increase the pillar size to account for weak roof or floor rock. Mains Pillars 4. Utilise a minimum width to height ratio of 4. 5. Utilise a minimum ARMPS factor of safety of 1.73. 6. Increase the pillar size to account for significant faulting that transects the pillar. 7. Increase the pillar size to account for weak roof or floor rock. Panel Pillars 4. Utilise a minimum width to height ratio of 4. 5. Utilise a minimum ARMPS factor of safety of 1.5. 6. Review pillar stability on a case by case basis if the following conditions occur: • Competent roof rock that does not readily cave. • Presence of faulting that transects the pillar. • Presence of weak roof or floor rock. 9.6 GOB CAVE PREDICTION There are three principal mechanisms of gob caves at shallow depth coal mines: 1) failure related to blocks and wedges formed by geologic structure, 2) failure related to span between unmined pillars, and 3) failure related to span of cantilevered beams created as pillars are removed. These failures are illustrated in Figure 7.4. Where geologic structure is present, it will usually control the failure mode. Al l three failure mechanisms have an element of deformation and dilation that will cause excavation convergence. This convergence can be measured to predict gob caves. 186 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Excavation convergence measurements are an effective way of accurately predicting gob caves. The critical rate of movement for retreat pillar coal mines is expected to be in the range of 0.2 to 1.0 mm/min. In order to provide adequate warning of an impending cave, a convergence rate less than the critical must be selected. Based on the limited field data, a rate of approximately 0.3 mm/min was recommended for initial use at the Quinsam Coal Mine. This value is expected to provide adequate warning without an excessive number of false alarms. Many factors affect stability such that additional studies are recommended to better determine a range for the critical convergence rate. To be effective the convergence instrumentation must be continuously monitored, and must be located in the section of the gob that is about to cave. Additionally, if borehole instruments are to be used they must be anchored beyond the projected height of the cave. In shallow depth coal mines the depth of cover and stress were not found to be significant factors in determining the critical rate of movement. Geologic structure is the factor that has the most significant impact on gob stability. Identifying the geologic structures and, where necessary, leaving pillars to support blocks and wedges formed by structure is considered the best method of maintaining control of gob caves. The Quinsam Coal Mine adopted the recommended practice of carrying out detailed structural mapping in all retreat panels. Where adverse geologic structure is identified the mine now leaves additional stump and fender pillars for support. Since this practice was adopted, along with the practice of leaving all point pillars, there have been no injuries or equipment damage caused by unexpected gob caves. 9.7 CAVING AND SUBSIDENCE Caving and subsidence over shallow depth mines is distinct from that of deeper mines. Induced stresses are the principal reason for the significant differences between the two. Numerical modelling has shown that the size of the stress relief zone, and the magnitude and extent of the tensile stresses, increases with decreasing depth of cover. It was also shown 187 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH that as the cave height increases the magnitude and extent of the tensile stresses will increase over shallow depth excavations, but will decrease over deeper excavations. Where low compressive or tensile stresses exist along with geologic structure, gravity induced structurally controlled failures such as block falls, wedge falls, beam bending and strata separation are most likely to occur. Shear failures are most likely to occur at the edges of an excavation where mining induced stresses are concentrated. These failures define a boundary line that is similar to the maximum shear stress boundary proposed by Kratzsch (1983), see Figure 8.19. Within the shear failure boundary the rock yields to create the zones of caving and fracturing. Outside the boundary there is little or no yield or deformation. At shallow depth the shear boundary is approximately defined by the angle of fracture. The term "angle of fracture" was coined to define the extent of discontinuous subsidence. It is defined as the angle between a vertical line at the edge of the excavation and a line drawn between the edge of the excavation and the extent of surface fracturing. The angle of fracture is analogous to the angle of critical deformation. Ground movements that occur beyond the angle of fracture are expected to be very small in comparison to those occurring within the angle of fracture. The angle of fracture at shallow depth mines is expected to be very small or even negative. At the Quinsam Coal Mine the angle of fracture was determined to have an average value of -1.9°. This implies that significant ground movements only occur directly above the mined out excavation. Ground movements over shallow mines are seen to be extremely limited when compared to the large trough subsidence basins that occur over deeper coal mines. Large ground movements or fracturing may still occur beyond the edge of the excavations where adverse faulting is present, or where the overburden is comprised of a significant thickness of unconsolidated material. Analysis of these situations must be carried put on a case specific basis. 188 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH The yield zone defines the caved and fractured zones. It is proposed that discontinuous surface subsidence will only occur when the yield zone extends to surface as shown in Figure 8.19A. At deeper mines the yield zone does not extend to surface, a zone of continuous deformation occurs between the yield zone and the ground surface as shown on Figure 8.19B. This research has shown that the upper bound limit for discontinuous surface subsidence is 30 times the excavation height. This is the maximum height to which the fractured zone was observed to extend at the Quinsam Coal Mine. This upper bound limit for discontinuous surface subsidence is used as a refined definition of shallow depth. Knowledge of the depth at which subsidence changes from discontinuous to continuous is critical for protection of surface structures as well as to protect against mine flooding when working beneath water sources. At very shallow depth caving may propagate through to surface. When this happens, sinkholes and large open fractures may occur. Many researchers have suggested that the upper bound limit for caving is 10 times the extraction height. This research has shown that caving may extend to a height of 14 times the excavation height. Where faulting is present, the height to which caving occurs may be greater yet. 9.8 CONTRIBUTIONS TO THE ADVANCEMENT OF KNOWLEDGE This research has contributed to the advancement of knowledge in several ways. The contributions have included determination of the properties of the rock and rock mass at the Quinsam Coal Mine, determination of the factors that cause the distinct response of the rock mass to mining at shallow depth, and determination of geotechnical design methods for retreat pillar coal mining at shallow depth. The most significant contributions are discussed below. Mining induced stresses over shallow depth mines are significantly different to those over deeper mines. The size of the stress relief zone and the magnitude of the tensile stresses 189 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH increase with decreasing depth of cover. At shallow depth the magnitude and extent of the tensile stresses above the excavation will increase as the caved zone propagates upwards. At deeper mines the magnitude and extent of the tensile stresses will decrease as the caved zone propagates upwards. Rock masses are typically weak in tension and shear. Where low compressive or tensile stresses exist along with geologic structure, gravity induced structurally controlled failures are most likely to occur. At shallow depth mines, geologic structure is the most significant factor in overall stability and the rock mass response to mining. It was determined that there are existing design tools for excavation stability analysis, support design, pillar design, and gob cave prediction, that are applicable to retreat pillar coal mines operating at depths of cover less than 100m. Convergence measurements were found to be an appropriate method for determination for excavation stability as well as for prediction of gob caves. Stress and depth of cover were found to not be significant factors in determining the critical convergence rates for either excavation stability or gob caves at shallow depth. The ARMPS computer program (Mark et al 1995) was determined to be the most appropriate method for pillar design at shallow depth. The program considers overburden and abutment stresses, pillar shape, and pillar group interaction. Al l these factors have been shown to be important for pillar design in shallow depth retreat pillar mines. Ground support design in shallow depth mines must consider both the rock mass and geologic structure. The French Matrix method (Newson 1987) was the only existing rock mass support design method that did not overestimate the support requirements. Any of the evaluated methods could be used to design a safe excavation; however, only the French Matrix method results in a safe and cost effective excavation. The UNWEDGE computer program (Carvalho et al 1994) was determined to be an excellent tool for visualisation of wedges and for determination of support requirements to secure the wedges. 190 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Non of the existing design tools were found to be suitable for prediction of caving and subsidence over shallow depth retreat pillar coal mines where discontinuous subsidence occurs. Discontinuous subsidence occurs where the caved or fractured rock extends to surface. It is next to impossible to protect surface facilities from damage, or to prevent surface water inundation where discontinuous subsidence occurs. As such, at shallow depth, the most important consideration is to what extent discontinuous subsidence extends out beyond the limits of the underground excavation. This research has shown that the commonly used ground deformation parameters such as vertical displacement, strain, and curvature have little or no application and can not be determined in a meaningful way in areas of discontinuous surface subsidence. The most significant findings of this work are considered to relate to the cave height and to the extent of subsidence over shallow depth mines. The height of caving was determined to be as much as 14 times the excavation height. Most previous research has indicated that the upper bound for caving is 10 times the excavation height; this value is commonly used for engineering design purposes. A potential 40% increase in the cave height over the previously accepted criteria is considered an extremely important finding that will significantly impact upon engineering evaluations of new and abandoned mines. Only the area immediately above shallow depth retreat pillar mine excavations is affected by significant subsidence. The angle of critical deformation or angle of fracture, which define the extent of damaging ground movements, was determined to be less than 0° for excavations at depths less than 30 times the extraction height. Existing subsidence prediction tools suggest that the angle of critical deformation ranges from between 4° and 26.5°. Knowing that critical ground deformations are restricted to the area above the excavation will significantly reduce the amount of coal resources sterilised in protective pillars. Large ground movements or fracturing may still occur beyond the edge of the excavations where adverse faulting is present, or where the overburden is comprised of a significant thickness of unconsolidated material. 191 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH 9.9 RECOMMENDATIONS FOR FURTHER WORK The recommendations for future work include refining the design methods through collection and statistical analysis of additional field data, verification of the assumptions made such as the in situ stress regime, improvements to numerical simulation methods for investigation of caving, and most importantly, verification of the applicability of the findings to other mines working in similar geologic settings. The areas considered the most important for further work are discussed below. The in situ stress values used for the numerical simulations in this research were derived from theoretical considerations, visual observations, and comparison to measurements carried out at other mines. Verification of in situ stresses is considered important since mining induced stresses are considered to be the main cause of the distinct behaviour of the rock mass at shallow depth mines. Field data was collected on several geotechnical aspects of retreat pillar coal mining at shallow depths. When the field data was analysed it was apparent that relationships existed between the various parameters and the rock mass response to mining; however, there was insufficient data to derive statistically meaningful relationships. Examples of this are as follows: • Rock mass quality and the size of the excavation were recognized to be related to the stability of the excavations; they may also be factors in determining the critical convergence rate for roadway and gob stability. • Pillar size, pillar age, and pillar proximity to the gob were recognized to be factors that influence pillar stability. • Depth, rock mass quality, and rock strength were recognized to be factors that influence the angle of draw and angle of fracture. By collecting and analysing additional field data on the above parameters it is anticipated that statistically meaningful relationships may be recognised, and that refined geotechnical design tools may be developed. The field data used in this study was primarily obtained from the Quinsam Coal Mine supplemented with information from literature, where available. In 192 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH order to ensure the general applicability of the design methods future work should include data collection and analysis from other shallow depth retreat pillar coal mines. Data collection should include information on both the hazard and the consequence of failures such risk assessment might be carried out along with the geotechnical design. During the course of this study, it was recognized that stump, fender, and point pillars left in the gob play a very important role in safety and productivity. It is very important that the remnant pillars be properly sized; however, there are no published design recommendations for remnant pillars. The irregular shape and location of these pillars makes it difficult, or impossible, to carry out stability analyses with conventional design tools. If remnant pillars are too small, the roof will continue to cave with little change in mining conditions. If the pillars are too large, there is a possibility of a massive pillar collapse or pillar run occurring. At the present time the design is usually left up to the underground operators. Developing design methods for these pillars may result in improved safety and productivity as well as improved recovery. As such, developing pillar design methods for point, stump, and fender pillars left in the gob is recommended for future work. Numerical modelling proved to be a useful tool for studying the process of shallow depth caving and subsidence. However, at this time it cannot be used on its own as a design tool. The work carried out indicated that there are several outstanding problems with numerical simulation of shallow depth caving and subsidence. The model must be "tricked" in order to handle the large displacements associated with caving. The validity of this method to simulate the physical process of caving and subsidence could not be validated. The numerical model occasionally returned anomalous results. There remains much scope for improvement of numerical models and modelling techniques for shallow depth caving and subsidence. The most important improvement required would be the ability to realistically model the large deformations and block rotations associated with caving. If this were achieved the need to "trick" the models to simulate caving by changing material properties could be eliminated. 193 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Sub-surface ground movements that occur between the underground excavation and the ground surface remain an aspect of coal mining that has not been extensively investigated. Surface subsidence is merely the manifestation of sub-surface ground movements. By gaining a better understanding of the sub-surface ground movements it is likely that significant improvements in the methods of predicting caving and subsidence will follow. As such, it is strongly recommended that future work be carried out to gain knowledge on the rock movements that occur between underground and surface. The findings of this research are based primarily on fieldwork carried out at the Quinsam Coal Mine. It was determined that the properties of the rock and rock mass at the Quinsam Mine are similar to those of other coal mines working in similar geologic environments. As such, it is believed that the findings from this research can be applied at other mines operating in similar geologic settings. However, this has not been tested. Future work should include verification of the applicability of these results to other shallow depth retreat pillar coal mines. 194 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH REFERENCES Abel, J.F., Lee, F.T., 1980, "Subsidence Potential in Shale and Crystalline Rocks", USGS, OFR 80-1072. Bai, M. , Kendorski, F.S., Roosendaal, D.J., 1995, "Chinese and North American High Extraction Underground Coal Mining Behaviour and Water Protection Experience and Guidelines", Proceedings, 14th International Conference on Ground Control in Mining, ed. S. Peng. Barron, K., Wright, P.L., Smales, T., 1982, "Guidelines for the Design of Coal Pillars in the #4 Seam at Mclntyre Mines Ltd., Grand Cache, Alberta", Proceedings, Conference on Stability in Underground Mining, ed. C. Brawner. Barton, N. , Lien, R., Lunde, J., 1977, "Estimation of Support Requirements for Underground Excavations", Proceedings, 16th US Rock Mechanics Symposium. Barton, N., Lien, R., Lunde, J., 1974, "Engineering Classification of Rock Masses for the Design of Tunnel Support", Rock Mechanics, Vol. 6, No. 4, 1974, pp 189 - 236. Bell, F.G., 2000, "Engineering Properties of Soils and Rocks", 4th Edition, Blackwell Science Ltd., Oxford. Bieniawski, Z.T., 1968, "The Effects of Specimen Size on the Compressive Strength of Coal", Int. J. Rock Mech. Min. Sci., Vol. 5, pp 325 - 335. Bieniawski, Z.T., 1974, "Geomechanics Classification of Rock Masses and its Application in Tunnelling", Proceedings, 3rd International Congress on Rock Mechanics, ISRM, Vol. 11 A. Bieniawski, Z.T., 1981, "Improved Design of Coal Pillars for US Mining Conditions", Proceedings, International Conference on Ground Control in Mining, ed. S. Peng. Bieniawski, Z.T., 1987, "Strata Control in Mineral Engineering", Wiley and Sons, New York. Bieniawski, Z.T., 1989, "Engineering Rock Mass Classification—a Complete Manual for Engineers and Geologists in Mining Civil and Petroleum Engineering", Wiley and Sons, New York. Biswas, K., Peng, S.S., Tsang P., 1995, "Influence of Partings in a Coal Seam on Long Term Pillar Stability", Proceedings, 14th International Conference on Ground Control in Mining, ed. S. Peng. Boyum, B.H., 1961, "Subsidence Case Histories in Michigan Mines", 4th US Symposium on Rock Mechanics, Perm. State University. 195 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Brady, B.H.G., Brown, E.T., 1985, "Rock Mechanics for Underground Mining", George Allen Unwin, London. Brauner, G., 1973, "Subsidence Due to Underground Mining, Part 2: Ground Movements and Mining Damage", USBMIC 8572. Brauner, G., 1973, "Subsidence Due to Underground Mining, Part 1: Theory and Practices in Predicting Surface Deformation", USBM IC 8571. British Coal, 1992, Code of Practice 01/30, "The Support of Mine Roadways by Rock Bolts." Bunting, D., 1911, "Chamber Pillars in Deep Anthracite Mines", Transactions ATME, Vol. 42, pp 236-245. CANMET, 1995, Design Guidelines for Sublevel Mining Method, DSS File 15SQ.23440-2-9142.CANMET. CANMET, 1996, Quinsam Coal Mine Unconfined Compressive Strength Testing, Cape Breton Coal Research Laboratory, unpublished results. Carvalho, J.L., Hoek, E., Bin L i , 1992-94, Unwedge, Version 2.2, Software, Rock Engineering Group, Department of Civil Engineering, University of Toronto. Chase, F.E., Mark, C , 1993, "Ground Control Design for Pillar Extraction", Soc. Min. Eng., AIME Preprint No. 93-282, Reno. Choi, D.S., Dahl, H.D., 1981, "Measurement and Prediction of Mine Subsidence over Room and Pillar in Three Dimensions", Proceedings, Workshop on Surface Subsidence Due to Mining, ed. S.S. Peng. Cleaver, D.N., Reddish, D.J., Shadbolt, C.H., Dunham, R.K., 1995, "Statistical Regional Calibration of Subsidence Prediction Models", 14th International Conference on Ground Control in Mining, ed. S. Peng. Coates, D.F., 1981, "Rock Mechanics Principles", CANMET, Monograph 874. Coetzee, M.J., Hart, R.D., Varona, P.M., Cundall, P.A., 1998, FLAC Basics: An introduction to FLAC and a guide to its practical application in geotechnical engineering, 2nd Revised Edition, Itasca Consulting Group, Inc., Minneapolis, Minnesota. Conroy, P.J., Gyarmaty, J.H., 1983, "The Mid Continent Field: Result of a Subsidence Monitoring Program", Surface Mining Environmental Monitoring and Reclamation Handbook, Elsevier, New York. Crane, W.R., 1929, "Subsidence and Ground Movement in the Copper and Iron Mines of the Upper Peninsula, Michigan", USBM Bulletin 285. 196 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Crane, W.R., 1931, "Essential Factors Influencing Subsidence and Ground Movement", USBMIC 6501. Cullen, M. , 1988, "Studies of Destress Blasting at Campbell Red Lake Mine", M.Eng. Thesis, McGill University, Montreal. Cullen, M. , 1989, "Rock Bolt Quality Assurance at the Myra Falls Mine" Westmin Resources Ltd. internal notes for short course. Cullen, M. , 1993, "Report on Site Visits to US and Canadian Coal Operations", internal report, Quinsam Coal Corporation. Cullen, M. , Pakalnis, R., Payne, D., 1996, "Predicting Gob Cave by Rock Movement at the Quinsam Coal Mine", Proceedings, North American Rock Mechanics Symposium, A A Balkema, New York. Darmody, R., Jansen, S., Carmer S., Steiner, J., 1989, "Agricultural Impacts of Coal Mine Subsidence: Effects on corn Yields", J. Environ. Qual., vol 18, pp 265 - 267. Daws, G., 1987, "Coal Mine Roof Bolting", Mining Engineer, pp 147 - 154. Deniau, R., Leonet, O., Barou, S., 1982, "Detection and Warning System for Roof Falls Underground", Proceedings, 7th International Strata Control Conference, Liege, Belgium. Deere, D.U., 1964, "Technical Description of Rock Cores for Engineering Purposes", Rock Mechanics and Engineering Geology, Vol. 1, No. 1, pp 17-22. Dhar, B.B., Saxena, N.C., Singh, U.K., 1992, "Rockmass Characterization for Estimation of Support Load in Underground Galleries and for Prediction of Surface Subsidence in Indian Coal Fields", Proceedings, Eurock '92, Thomas Telford, London. Evans, I., Pomeroy, D. C , Berenbaum, R., 1961, "The Compressive Strength of Coal", Colliery Engineering, Vol. 2, pp 75 - 80. Evans, I., Pomeroy, CD. , 1966, The Strength, Fracture, and Workability of Coal, Pergamon, New York. Evans, W.H., 1941, "The strength of undermined strata, Trans Inst. Min. Metall. No., pp 50, 475 - 532. Farmer, I.W., Shelton, P.D., 1980, "Factors that Affect Underground Rockbolt Reinforcement Systems Design", Trans. Inst. Min. Metall, No.89 pp A68 - A83. Fayol, M. , 1885, "Sur les movements de terrain provoques par l'exploitation des mines", Bull. Soc. l'lndustrie Minerale, 2 n d series, 14, 818. (reported in Brady and Brown 1985) 197 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Fletcher, J.B., 1960, "Ground Movement and Subsidence from Block Caving at Miami Mine", AIME Transactions, Vol. 217. Gaddy, F.L, 1956, "A study of the Ultimate Strength of Coal as Related to the Absolute Size of Cubical Specimens Tested", Bulletin No. 112, Virginia Polytechnic. Galovich, K., 2001, personal communication. Galvin, J.M., 1995, "Roadway and Pillar Mechanics Workshop: Stage 2—Design Principals and Practice", Course Notes, University of Sydney. Gardner S. 1992, "Geology of the Quinsam Coal Mine" Quinsam Coal Corporation internal report. Garrard, G.F.G., Taylor, R.K., 1988, "Collapse Mechanisms of Shallow Coal Mine Workings From Field Measurements", Proceedings, Engineering Geology of Underground Excavations, eds:Bell, Cueshaw, Crippps, Lowel.. Ghosh, C.N., Ghose, A.K., 1992, "Estimation of Critical Convergence and Rock Load in Coal Mine Roadways—An approach Based on Rock Mass Rating", Geotechnical and Geological Engineering, No. 10, pp 185 - 202. Gray, R.E., 1990, "Mining Subsidence - Past, Present, Future", Int. J. Mine. & Geol. Eng. No. 8, pp 400 - 408. Greenwald, H.P., Howarth, H.C., Hartman, I., 1939, "Experiments on Strength of Small Pillars of Coal in the Pittsburgh Bed", USBM Technical Paper 605. Griffith, W., Conner, E.T., 1912, "Mining Conditions Under the City of Scranton, PA", USBM B25. Grimstad, E., Barton, N., 1993, "Updating the Q-System for NMT", Proceedings, international symposium on sprayed concrete—modern use of wet mix sprayed concrete for underground support, eds: Kompen, Opsahl and Berg, Norwegian Concrete Association. Hanna, K., Conover, D., Haramy, K., 1984, "Coal Mine Entry Intersection Behaviour Study", USBM, RI 9337. Hedley, D.G.F., Grant, F., 1972, "Stope and Pillar Design for the Elliot Lake Uranium Mines", Canadian Institute of Mining and Metallurgy, Bulletin No. 65. Hellewell, F.G., 1988, "The Influence of Faulting on Ground Movement due to Coal Mining: The U.K. and European Experience", Mining Engineer, 147, pp 334 - 337. Herget, G., 1988, "Stresses in Rock", A.A. Balkema, Rotterdam. 198 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Heslop, T.G., 1974, "Failure by Overturning in Ground Adjacent to Cave Mining at Havelock Mine", 3rd ISRM Congress, Denver. Hoek, E., 1974, "Progressive Caving Induced by Mining an Inclined Orebody", Trans. Inst. Min. Metall., No. 83, pp A113 - A119. Hoek, E, Brown, E.T., 1980, "Underground Excavations in Rock", BVLM, London. Hoek, E., Wood, D.R., 1988, "Rock support", Mining Magazine, Oct., pp. 282 - 287. Hoek, E., Kaiser, P.K., Bawden, W.F., 1995, "Support of Underground Excavations in Hard Rock", A.A.Balkema, Rotterdam. Holla, L., Buizen, M. , 1990, "Strata Movement Due to Shallow Longwall Mining and the Effect on Ground Permeability", Proceedings, Aus. IMM, No. 1, pp 11 - 18. Holland, C.T. 1964, "The Strength of Coal in Mine Pillars", Proceedings, 6th U.S. Symposium on Rock Mechanics, University of Rolla. Howell, R.C., Wright, F.D., Dearinger, LA., 1976, "Ground Movement and Pressure Changes Associated with Short Wall Mining", Proceedings, 17th US Symposium on Rock Mechanics. Hunt, S.R.,1980, "Surface Subsidence Due to Coal Mining in Illinois", Ph.D. Thesis, University of Illinois at Urbana-Champaign. Hustrulid, W.A., 1976, "A Review of Coal Pillar Strength Formula", Rock Mechanics, pp. 115-145. Hutchinson, Jean D., Diederichs, Mark S., 1996, Cablebolting in Underground Mines, BiTech Publishers Ltd., Richmond, British Columbia. Itasca Consulting Group, Inc., 1999, Fast Lagrangian Analysis of Continua (FLAC) version 4, Proram and User's Guide, Minneapolis, Minnesota. Jeremic, M.L., 1985, "Strata Mechanics in Coal Mining", A.A.Balkema, Boston. Kane, W.F., Karmis, M. , 1986, "Geologic and Geotechnical Controls on the Stability of Coal Mine Entries", Rock Characterization Techniques, ISRM. Kantner, W.H., 1934, "Surface Subsidence over Porphyry Caving Blocks, Phelps Dodge Corporation, Copper Queen Branch", AIME Transactions, Vol. 109. Kay, D.R., McNabb, K.E., Carter, J.P., 1991, "Numerical Modelling of Mine Subsidence at Angus Place Colliery", Computer Methods and Advances in Geomechanics, Proceedings, 7th International Conference, Cairns, Australia, Vol. 2, A.A. Balkema, Rotterdam. 199 G E O T E C H N I C A L S T U D I E S O F R E T R E A T P I L L A R C O A L M I N I N G A T S H A L L O W D E P T H Khair, W.A., Begley, R.D., 1992, "Mechanistic Subsidence Prediction Model (MSPM)," Proceedings, Third Workshop on Surface Subsidence Due to Underground Mining, ed S. Peng. Khair, W.A., 1996, "Integrated Coal Specimen Strength Formula", Proceedings, 2nd North American Rock Mechanics Symposium, eds: Aubertin, Hassani, Mitri, A A Balkema, Rotterdam. Kohli, K.K., Peng, S.S., Thill, R.E., 1982, "Subsidence Experience in the Room and Pillar Mines of the Northern Appellation Coalfield", Ground Control in Room and Pillar Mining, ed. Y.P. Chugh, SME-AJME. Kolebaeva, J.M., 1968, Theory of Ground Movement Rock Mass, Nedra, Moscow (in Russian, summarised in Peng, 1992). Kotz, T.J., 1986, "The Nature and Magnitude of Surface Subsidence Resulting From Mining at Relatively Shallow Depths on Platinum Mines", SANGORM Symposium, The Effect of Underground Mining on Surface, South Africa Institute of Civil Engineers, Yeonville, S.A. Kratzsch, H., 1983, "Mining Subsidence Engineering", Springer Verlag, Berlin. Lama, R.D., Vutukuri, V.S., 1978, "Handbook on Mechanical Properties of Rocks - Volume II", Trans Tech Pub., Clausthal. Laubscher, D.H., 1981, "Planning Mass Mining Operations", Design and Operation of Caving and Sublevel Stoping Mines, ed. R.D. Stewart, AIME, New York. Laubscher, D.H, Taylor, H.W., 1976, "The Importance of Geomechanics Classification of Jointed Rock Masses in Mining Operations", Proceedings, Symposium on Exploration for Rock Engineering, Vol. 1, Johannesburg. Lee, A.J., 1966, "The Effect of Faulting on Mine Subsidence", Mining Engineer, 125, pp 417 -427. Liu, T.Q., ed. 1981, Surface Movements, Overburden Failure and Its Applications, Coal Industry Publishing Co., China (in Chinese, summarised in Peng, 1992). Madden, B.J., 1991, "A Re-Assessment of Coal-Pillar Design", Journal South African Inst. Min. Metall., Vol. 91, No. 1, pp 27 - 37. Mahtab, M. , 1976, "Influence of Rock Fractures on Caving", Transactions, AIME, Vol. 260, 1976. Maleki, H., 1988, "Detecting Stability Problems by Monitoring Rate of Roof Movement", Coal, December. 200 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Mark, C , 1990, "Pillar Design Methods for Longwall Mining", USBMIC 9247. Mark, C , Chase, F., 1994, "Design of Longwall Gate Entry Systems Using Roof Classification", Proceedings, New Technology for Longwall Ground Control, USBM SP 94-01. Mark, C , Mucho, T.P., 1994, "Longwall Mine Design For Control of Horizontal Stress", Proceedings, New Technology for Longwall Ground Control, USBM SP 94-01. Mark, C , Chase, F., Zipf, K., 1997, "Preventing Massive Pillar Collapses in Coal Mines", Proc. New Technology for Ground Control in Retreat Mining, NIOSH, IC 9466. Mark, C , Chase, F., Campoli, A.A., 1995, "Analysis of Retreat Mining Pillar Stability", Proceedings, 14th International Conference on Ground Control in Mining", ed. S.Peng. Mark, C , 1995, personal communication. Mark, C , Barton, T.M, 1996, "A New Look at the Uniaxial Compressive Strength of Coal", Proceedings, 2nd North American Rock Mechanics Symposium, eds: Aubertin, Hassani, Mitri, A A Balkema, Rotterdam. Mark, C , McCall, F., Pappas, D., 1997, "A Statistical Overview of Retreat Mining of Coal Pillars in the United States", Proceedings, New Technology for Ground Control in Retreat Mining, NIOSH IC 9446. Mark, C , 1998, personal communication. Mark, C , 1999, "Application of Coal Mine Roof Rating (CMRR) to extended cuts", Min. Eng. 51(4), pp 52 - 56. Mark, C , Barczak, T.M., 2000, "Fundamentals of Coal Mine Roof Support", Proceedings, New Technology for Coal Mine Roof Support, NIOSH IC 9453. Merritt, A.H., 1972, "Geologic Prediction for Underground Excavations", Proceedings, Conference on Rapid Excavation and Tunnelling, AIME, New York. Milne, D., Pakalnis, R., Felderer, M. , 1996, "Surface Geometry Assessment for Open Stope Design", Proceedings, North American Rock Mechanics Symposium, A A Balkema, Rotterdam. Moebs, N.N., 1982, "Subsidence over Four Room and Pillar Sections in South Western Pennsylvania", USBM RI 8645. Molinda, G.M., Mark, C , 1994, "Coal Mine Roof Rating (CMRR): A Practical Rock Mass Classification for Coal Mines", USBM IC 9387. 201 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Molinda, G.M., Mark, C , 1996, "Rating the Strength of Coal Mine Roof Rocks", USBM IC 9444. Molinda, G.M., Mark, C , Dolinar, D., 2000, "Assessing Coal Mine Roof Stability through Roof Fall Analysis", Proceedings, New Technology for Coal Mine Roof Support, NIOSH IC 9453. Morrison, R.G.K., 1976, "A Philosophy of Ground Control", Ontario Department of Mines. Naismith, W.A., Pakalnis, R.T., 1982, "Monitoring a Coal Pillar Extraction Operation", Proceedings, 1st International Conference on Stability in Underground Mining, ed. CO. Brawner, ATME, New York,. Najjar, Y., Chen, D., Zaman, M. , 1993, "Application of Non-linear Finite Element Method in Prediction of Ground Subsidence Due to Underground Mining", Canadian Symposium on Numerical Applications in Mining and Geomechanics, ed. H. Mitre, Montreal. NCB, 1975, "Subsidence Engineers Handbook II", National Coal Board Mining Department. Nelson, J.W., 1981, "Faults and their Effects on Coal Mining in Illinois", Illinois State Geological Survey Circular 523. Newson, S.R., 1987, "A Re-assessment of Strata Bolting", Mining Eng., Vol. 146, No. 306, pp 585 - 597. Noor, M. , Lloyd, P.W., Reddish, D.J., 1997, "Longwall Surface Subsidence Prediction through Numerical Modelling", Proceedings, 16th International Conference on Ground Control in Mining, ed S. Peng. North, P.G., Callaghan, R.P., 1980, "Subsidence Associated with Mining at Mt. Lyell", The Aus. IMM Conference, New Zealand. Obert, L., Duvall, W.I., 1967, "Rock Mechanics and the Design of Structures in Rock", Wiley and sons, New York. O'Connor, K .M. , O'Rourke, J., Carr, J., 1983 "Influence of rock Discontinuities on Coal Mine Subsidence", USBM OFR 30-84 O'Connor, K.M. , Dowding, C. H., 1992, "Distinct Element Modelling and Analysis of -Mining-Induced Subsidence," Rock Mechanics and Rock Engineering, No 25, Springer-Verlag, Austria. Parker, J., 1973, "What Can Be Learned from Surface Subsidence", E&MJ, July. 202 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Park, D.W., Jiang, L.A., Morley, L.A., Keeton, W., 1992, "Stability Analysis of a Room and Pillar Mine with Thinly Laminated Roof, Strong Pillars and Weak Floor", Mining Engineer, November, pp 1355 - 1360. Pariseau, W.G., 1982, "Shear Stability of Mine Pillars in Dipping Seams", Proceedings, 23rd U.S. Symposium on Rock Mechanics. Payne, D., 1994, "Geotechnical Assistance at the Quinsam Coal Mine, Campbell River, B.C.", CANMET Client Report MRL 93-075(CL). Papas, D.M., Bauer, E.R., Mark, C , 2000, "Roof and Rib Fall Incidents and Statistics: A Recent Profile", Proceedings,: New Technology for Coal Mine Roof Support, NIOSH IC 9453. Peng, S.S., Geng, D.Y., 1982, "Methods of Predicting the Subsidence Factors, Angle of Draw and Angle of Critical Deformation", Proceedings, State-of-the-Art of Ground Control in Longwall Mining and Mining Subsidence, SME-AIME, Littleton, CO. Peng, S.S., Chiang, H.S., 1984, "Longwall Mining", Wiley and Sons, New York. Peng, S.S., 1986, "Coal Mining Ground Control", Wiley and Sons, New York. Peng, S.S., 1992, "Surface Subsidence Engineering", Society For Mining, Metallurgy and Exploration Inc., Littleton, Colorado. Peng, S.S., 1994, "Numerical Prediction of Subsidence", 13th International Conference on Ground Control in Mining, ed. S. Peng. Peng, S.S., Luo, Y., Zhang, Z.M., 1995, "Development of a Subsidence Database and Determination of Subsidence Parameters", 14th International Conference on Ground Control in Mining, ed. S. Peng. Piggot, R.J., Eynon, P., 1978, "Ground Movements Arising from the Presence of Shallow Abandoned Mine Workings", Proceedings, Conference on Large Ground Movements and Structures, Pentech Press, Plymouth. Pothini, B.R., Schonfeldt, H.V., 1978, "Roof Fall Prediction at Island Creek Coal Company", Proceedings, 1st International Symposium on Stability in Coal Mining, Vancouver, ed: C. Brauner, I Dorling, Miller Freeman Pub., San Francisco. Promised Land Technologies Inc., 1996, Brancel Users Manual Version 3.0. Royce, S., 1941, "Mining Engineers Handbook", R. Peele, J. Church, eds., Vol. 1, Wiley and Sons, New York. Salamon, M.D.G., Monro, A.H.A., 1967, "A Study of the Strength of Coal Pillars", Journal South African Inst. Min. Met., No. 68, pp 55-57. 203 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Salamon, M.D.G., Oravecz, K.I., 1976, "Rock Mechanics in Coal Mining", Chamber of Mines of South Africa, Johannesburg. Serata S., 1989, "Quick Measurement of Roof Deterioration and Timing of Roof Fall", Serata Geomechanics Inc. Shadbolt, C.H., 1977, "Mining Subsidence—Historical Review and State of the Art", Proceedings ,Conference on Large Ground Movements and Structures, ed. J.D. Geddes, Pantech Press, London. Shadbolt, C.H., 1987, "A Study of the Effects of Geology on Mining Subsidence in the East Pennine Coalfield", Ph.D. Thesis, University of Nottingham. Sheorey, P.R., Singh, B., 1974,"Strength of Rectangular Pillars in Partial Extraction", Int. J. Rock mech. Min. Sci. & Geomech. Abstr. Vol 11 No 41. Sheorey, P.R., Das M.N., Barat, D., Prasad, R.K., Singh, B., 1987, "Coal Pillar Strength Estimation from Failed and Stable Cases", Int. J. Rock Mech. Min. Sci. and Geomech. Abst, Vol. 24, pp 347 - 355. Sheorey, P.R., 1994, "A Theory for In Situ Stresses in Isotropic and Transversely Isotropic Rock", Int. J. Rock Mech. Min. Sci. Geomech. Abstr., Vol. 31, pp 23-34 (reported in Hoek et al 1995). Singh, U.K., Stephansson, O.J., Herdocia, A., 1993, "Simulation of Progressive Failure in Hanging Wall and Foot Wall for Mining with Sub Level Caving", Transactions, Inst. Min. Metall., Section. A, No. 102. Singh, T.N., Singh, B., 1968, "Angle of Draw in Mine Subsidence", J. Mines, Metals and Fuels, pp 253 - 258. Skelly, W.A., Wolgamott, J., Wang, F., 1977, "Coal Mine Pillar Strength and Deformation Prediction through Laboratory Sample Testing", Proceedings, 18th US Symposium on Rock Mechanics. Sorenson, W.K., Pariseau, W.G., 1978, "Statistical Analysis of Laboratory Compressive Strength and Young's Modulus Data for Design of Production Pillars in Coal Mines", Proceedings, 19th U.S. Symposium on Rock Mechanics. Stankus, J., Peng, S.S., 1996, "A New Concept for Roof Control", Coal Age, Sept., pp 67 -72. Starfield, A.M. , Cundall, P.A., 1988, "Towards a Methodology for Rock Mechanics Modelling", Int. J. Rock Mech. Min. Sci. & Geomech. Abstr., Vol. 25, No. 3, pp 99 - 106. 204 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Steart, F.A., 1954, "Strength and Stability of Pillars in Coal Mines", J. Chem. Met. Min. Soc., South Africa, Vol. 54, pp 309 - 315. Stillborg, B., 1986, "Professional Users Handbook for Rock Bolting", Trans. Tech Pub. Su, D.W.H., 1992, "Finite Element Modelling of Surface Subsidence Induced by Underground Coal Mining", Proceedings, Third Workshop on surface Subsidence Due to Underground Mining, ed. S. Peng. Sutherland, H.J., Schuler, K.W., 1981, "A Review of Subsidence Prediction Research Conducted at Sandia National Laboratories", Proceedings, Workshop on Surface Subsidence Due to Underground Mining, eds: S. Peng, M Harthill. Sutherland, H.J., Heckes, A. A., Taylor, L.M., 1984, "Physical and Numerical Simulations of Subsidence above High Extraction Coal Mines", Sandia Report: SAND83-1191C. Szostak-Chrzanowski, Anna, 1988, "An Iterative Modelling of Ground Subsidence Using Non-Linear elastic Finite Element Analysis," Proceedings, 5th Int. Sym.on Deformation Measurements, eds: A.Chrzanowski, W.Wells. Szostak-Chrzanowski, A, Chrzanowski, A., 1991, "Use of Software 'Femma' in 2-D and 3-D Modelling of Ground Subsidence," Proceedings, 2nd Canadian Conference on Computer Applications in the Mineral Industry. Szwilski, A.B., 1982, "Sizing of Chain Pillars Around Longwall Panels", Proceedings, 1st International Conference on Stability in Underground Mining, ed. Brawner, AIME, New York. Tandanand, S., Powell, L.R., 1982, "Assessment of Subsidence Data from the Northern Appalachian Basin for Subsidence Prediction", USBM RI 8630. Thomas, L.A., "Subsidence and Related Caving Phenomena at the San Manuel Mine, San Manuel, Arizona", Magma Copper Company, San Manuel Division, unpublished report. Ulrich, B.F., Wuest, W.J., Stateham, R.M., 1991, "Relationships between Annulus Thickness and Integrity of Resin-Grouted Roof Bolts", USBM RI 9253. Unal, E., 1983, "Development of Design Guidelines and Roof Control Standards for Coal Mine Roofs", Ph.D. Thesis, Pennsylvania State University. U.S. Corps of Engineers, 1980, "Engineering and Design: Rock Reinforcement", Engineer Manual Eml 110-1-2907, Washington D.C. Vallejo, L.E., 1989, "Correlation Between Unconfmed Compressive and Point Load Strengths for Appalachian Rocks", Rock Mechanics as a Guide for Efficient use of Natural Resources, ed W. Khair. 205 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH van Heerden, W.L., 1974, "In Situ Determination of Complete Stress Strain Characteristics of 1.4m Square Specimens with Width to Height Ratios up to 3.4", CSLR Research Report ME1265. VPI&SU, 1994, "Surface Deformation Prediction System—Version 4 User's Manual", Department of Mining and Minerals Engineering, Virginia Polytechnic Institute and State University. Wade, L.V., Conroy, PJ., 1980, "Rock Mechanics Study of a Longwall", Mining Engineer, pp 1728- 1734. Wagner, H., 1980, "Pillar Design in Coal Mines", J. S. Afr. Inst. Min. Metall., vol 80. Wagner, H., Madden, BJ . , 1984, "Fifteen Years' Experience with the Design of Coal Pillars in Shallow South African Collieries: an Evaluation of the Performance of the Design Procedures and Recent Improvements", Proceedings, Design and Performance of Underground Excavations, ISRM. White, H., 1992, "Accurate Delineation of Shallow Subsurface Structure Using Ground Penetrating Radar", Symposium on Construction over Abandoned Mines, SANGORM, Yeoville, South Africa. Whittaker, B.N., Singh, R.N., 1979, "Design and Stability of Pillars in Long Wall Mining", Min. Eng., No. 138, pp 59 - 73. Whittaker, B.N., Reddish, D.J., 1989, "Subsidence: Occurrence, Prediction and Control", Elsevier Science Publishers, Amsterdam. Whittaker, B.N., Gaskell, P . , Reddish, D.J., 1990, "Subsurface Ground Strain and Fracture Development Associated with Longwall Mining", Mining Sci. and Tech. No. 10, Elsevier Science Publishers, Amsterdam. Young, L.E., Stoek, H.H., 1916, "Subsidence Resulting from Mining", University of Illinois Bulletin #91,Urbana. Zern, E.N., 1928, "Coal Miners Pocketbook", McGraw Hill, New York. 206 G E O T E C H N I C A L S T U D I E S O F R E T R E A T P I L L A R C O A L M I N I N G A T S H A L L O W D E P T H APPENDIX 1 QUINSAM COAL MINE SITE CHARACTERIZATION 1) FAULT AND JOINT MEASUREMENT DATA 2) MECHANICAL PROPERTIES TEST DATA 3) ROCK MASS CLASSIFICATION DATA 207 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH FAULT AND JOINT MEASUREMENTS FROM THE 2N MINE F = FAULT J = JOINT F E A T U R E # DIP (deg.) DIP DIR (deg.) T Y P E D I S P L A C E M E N T (m) l 60 340 f 1.8 2 45 100 f 1.5 3 75 95 J 0 4 40 270 J 0 5 55 65 f 0.3 6 62 235 J 0 7 25 90 J 0 8 80 60 J 0 9 30 100 J 0 10 25 100 J 0 11 45 235 J 0 12 60 280 J 0 13 65 100 J 0 14 65 90 J 0 15 50 105 J 0 16 46 95 J 0 17 39 90 J 0 18 40 90 f 0.5 19 40 100 f 2 20 75 115 f 0.1 21 80 270 J 0 22 40 255 J 0 23 30 90 J 0 24 30 90 J 0 25 35 275 J 0 26 35 275 J 0 27 60 275 J 0 28 65 95 J 0 29 45 225 J 0 30 80 65 f 0.4 31 40 245 J 0 32 50 100 J 0 33 30 105 J 0 34 30 105 J 0 35 50 260 f 0.2 36 65 85 f 0.2 37 50 285 f 1.2 38 60 5 J 0 39 60 185 J 0 40 70 5 J 0 41 45 185 f 0.3 42 50 260 J 0 208 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH F E A T U R E # DIP (deg.) DIP DIR (deg.) T Y P E D I S P L A C E M E N T (m) 43 50 195 J 0 44 60 5 J 0 45 40 85 J 0 46 40 105 J 0 47 30 20 J 0 48 45 135 J 0 49 45 115 J 0 50 55 110 J 0 51 60 115 J 0 52 70 120 J 0 53 50 90 J 0 54 45 90 J 0 55 50 85 J 0 56 55 190 J 0 57 55 195 J 0 58 40 95 J 0 59 30 100 J 0 60 40 240 J 0 61 30 270 J 0 62 45 250 . J 0 63 45 260 J 0 64 45 235 J 0 65 30 255 J 0 66 45 90 J 0 67 50 90 J 0 68 60 85 J 0 69 60 85 J 0 70 35 280 J 0 71 35 265 J 0 72 35 85 J 0 73 40 90 J 0 74 50 90 J 0 75 80 260 J 0 76 35 280 J 0 77 35 85 f 3.2 78 45 340 f 0.3 79 30 80 f 0.4 80 30 100 f 0.8 81 30 100 J 0 82 30 100 J 0 83 35 115 f 1.5 84 40 95 f 0.5 85 45 95 J 0 86 40 245 J 0 87 40 280 f 1.5 88 45 100 f 1.5 209 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH F E A T U R E # DIP (deg.) DIP DIR (deg.) T Y P E D I S P L A C E M E N T (m) 89 60 325 J 0 90 60 340 J 0 91 50 350 . J 0 92 65 350 J 0 93 50 85 J 0 94 55 90 J 0 95 65 90 J 0 96 45 90 J 0 97 60 335 J 0 98 45 90 J 0 99 30 0 J 0 100 42 98 f 0 101 45 104 f 1.5 102 51 71 f 0.5 103 45 344 f 0 104 70 154 f 0.6 105 55 142 f 0.4 106 36 339 f 0 107 70 78 f 0.5 108 50 82 f 1.7 109 85 28 f 3 110 52 88 f 2.5 111 63 81 f 1 112 63 68 f 0.8 113 70 269 f 2.6 114 42 72 f 0 115 45 80 f 0 116 45 322 f 0.5 117 54 342 f 0.5 118 25 232 f 1 119 30 244 f 0.6 120 45 196 f 0.6 121 40 195 f 0.2 122 40 242 f 0.5 123 75 210 f 2.5 124 75 210 f 3 125 60 83 f 2 126 30 270 f 0 127 60 274 f 2 128 55 265 f 1 129 40 258 f 0.2 130 40 258 f 0.2 131 45 260 f 1.4 132 45 265 f 2 133 50 255 f 0.1 134 60 346 f 1 210 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH F E A T U R E # DIP (deg.) DIP DIR (deg.) T Y P E D I S P L A C E M E N T (m) 135 30 248 f 0.6 136 40 245 f 0.1 137 40 254 f 1.6 138 75 85 f 1.5 139 35 257 f 0.5 140 22 307 f 0.5 141 28 307 J 0 142 20 305 J 0 143 35 305 f 0.2 144 35 325 f 0.1 145 25 305 J 0 146 25 305 f 0.1 147 45 250 J 0 148 30 290 f 0.3 149 20 300 J 0 150 35 300 J 0 151 40 300 J 0 152 36 300 J 0 153 35 310 f 0.2 154 80 290 J 0 155 5 275 J 0 156 60 270 J 0 157 42 117 J 0 158 50 292 J 0 159 35 280 f 0.4 160 35 320 f 0.4 161 70 265 J 0 162 82 320 J 0 163 85 320 J 0 164 72 270 J 0 165 80 280 J 0 166 90 360 J 0 167 85 340 J 0 168 85 135 J 0 145 90 265 J 0 146 87 270 J 0 146 45 90 J 0 146 35 108 J 0 146 55 154 J 0 146 75 242 J 0 146 45 230 J 0 146 30 90 J 0 146 47 258 J 0 146 42 345 J 0 146 46 118 J 0 146 47 163 J 0 211 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH F E A T U R E # D I P (deg.) DIP D I R (deg.) T Y P E D I S P L A C E M E N T (m) 146 47 258 J 0 146 43 43 J 0 146 57 82 J 0 147 38 340 f l 148 45 330 f 0.5 149 45 325 f 0.2 150 40 325 f 0.7 151 40 322 f 0.3 152 40 340 f 0.5 153 25 336 f 0 154 40 295 f 1.6 155 50 240 f 0 156 50 236 f 0.1 157 50 93 f 1.5 158 35 154 0 159 35 156 f 0.7 160 48 171 f 0.4 161 45 325 f 1 162 35 122 f 0.6 163 40 45 f 0.4 164 90 122 J 0 165 90 141 J 0 166 40 158 f 0.3 167 35 153 f 0.6 168 30 91 f 1.5 169 65 346 f 0.5 170 62 108 f 0.6 171 40 187 f 0.7 172 30 10 f 0 173 70 101 f 1.5 174 35 165 f 0 175 45 135 f 0.3 176 40 270 f 0.5 177 70 94 f 0.6 178 60 263 f 2.8 179 70 108 f 1 180 30 102 f 1 181 50 93 f 0 182 50 265 f 0.1 183 55 103 f 1 184 62 270 f 3 185 75 180 f 0 186 50 182 f 2 187 45 178 f 1 188 40 260 f 0.2 189 45 125 f 0.3 212 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH F E A T U R E # DIP (deg.) DIP DIR (deg.) T Y P E D I S P L A C E M E N T (m) 190 50 263 f l 191 70 110 f l 192 40 173 f 0 193 57 272 f 0.5 194 70 130 f 0.3 195 40 180 f 0.4 196 40 173 f 0.3 197 45 107 f 2 198 45 95 f 0.8 199 50 112 f 1 200 20 120 f 0 201 45 245 f 0.3 202 50 252 f 0.2 203 58 268 f 0.3 204 50 89 f 1.6 205 35 284 f 0.8 206 45 279 f 1 207 50 290 f 0.2 208 40 102 f 3.2 209 75 55 f 1.3 210 50 345 f 0 211 60 159 f 1 212 60 75 f 1.5 213 75 23 f 1 214 45 82 f 1 215 50 190 f 1.5 216 50 157 f 2.2 217 58 334 f 0.6 218 40 260 f 2 219 55 266 f 0.5 220 45 92 f 1.2 221 50 183 f 0.2 222 70 273 f 1.2 223 55 91 f 2 224 60 59 f 0 225 45 345 f 0 226 50 55 f 1.2 227 40 252 f 1 228 30 165 f 1.5 229 50 238 f 0.1 230 65 82 f 1.8 231 50 80 f 2.5 232 65 36 f 0.2 233 40 105 f 0.4 234 30 290 f 2 235 50 304 f 2.3 213 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH F E A T U R E # DIP (deg.) DIP DIR (deg.) T Y P E D I S P L A C E M E N T (m) 236 40 172 f 0 237 40 294 f 2 238 30 296 f 1 239 40 103 f 1 240 30 104 f 0.2 241 40 284 f 0.1 242 50 170 f 0.2 243 60 85 f 0.3 244 50 263 f 0.6 245 45 288 f 3 246 22 290 f 2 247 30 156 f 0 248 65 195 f 3 249 55 195 f 0.2 250 50 180 f 0.5 251 50 176 f 0 252 40 180 f 0 214 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TESTS UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: #1 coal seam - upper seam TEST NUMBER: QC 1 SAMPLE LOCATION: #1 Mains B Road. TEST DATE: 4 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 25 November, 1993 TEST PREPARATION: cored block DIAMETER (inch): 1.821 HEIGHT (inch): 3.953 AREA (sq. inch): 2.604 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (psi) NOTES 0 .0000 .0000 0 0 1.0 .0100 .0025 200 76.8 .0125 .0032 300 115.2 .0143 .0036 400 153.6 1.5 .0158 .0040 500 192 .0167 .0042 600 230.4 .0194 .0049 700 268.8 .0231 .0058 800 307.2 2.5 .0250 .0063 900 345.6 .0264 .0067 1000 384.0 1100 Bedding is perpendicular to axis .0279 .0071 1200 460.8 Longitudinal fractures on cleat (axial) .029 .0073 1300 499. .0300 .0076 1400 537.6 3.0 .0309 .0078 1500 576.0 Sample failed in brittle fashion, still intact 1600 614.4 3.2 .0362 .0092 1200 460.8 .0416 .0105 1100 422.4 .0479 .0121 100 384.0 UNIAXIAL COMPRESSIVE STRENGTH (psi): 614 215 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: #1 rider seam TEST NUMBER: QC 2 SAMPLE LOCATION: #1 Mains B Road face. TEST DATE: 4 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 25 November, 1993 TEST PREPARATION: cored block DIAMETER (inch): 1.821 HEIGHT (inch): 4.736 AREA (sq. inch): 2.604 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (psi) NOTES 0 .0000 .0000 0 0 .0014 .0008 700 268.8 .0019 .0010 1000 384 .0023 .0013 1000 384.0 0.5 .0038 .0021 400 153.6 UNIAXIAL COMPRESSIVE STRENGTH (psi): 384 216 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: #1 coal seam upper seam TEST NUMBER: QC 3 SAMPLE LOCATION: #1 Mains B Road face TEST DATE: 4 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 25 November 1993 TEST PREPARATION: cored block DIAMETER (inch): 1.824 HEIGHT (inch): 3.068 AREA (sq. inch): 2.611 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (psi) NOTES 0 .0000 .0000 0 0 .0030 .0010 400 153.2 .0053 .0017 600 229.8 Fast loading rate 0.5 .0084 .0027 700 268.1 .0094 .0031 800 306.4 .0100 .0033 900 344.7 .0104 .0034 1000 383 .0106 .0035 1100 421.3 .0111 .0036 1300 497.9 .75 .0118 .0039 1500 574.5 .0124 .0040 1700 654.9 .0131 .0043 1900 727.7 .0137 .0045 2100 804.3 1.0 .0171 .0056 2900 1110.7 .0173 .0056 3000 1149 .0194 .0063 3500 1340.5 - Brittle failure .0224 .0073 3400 1302.2 - Specimen collapsed into numerous pieces at 1.4 .0234 .0076 0 0 failure - Axial fractures on cleat UNIAXIAL COMPRESSIVE STRENGTH (psi): 1340 217 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: #1 coal seam upper seam TEST NUMBER: QC 4 SAMPLE LOCATION: #1 Mains B Road face TEST DATE: 4 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 25 November 1993 TEST PREPARATION: cored block DIAMETER (inch): 1.828 HEIGHT (inch): 3.186 AREA (sq. inch): 2.624 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (psi) NOTES 0 .0000 .0000 0 0 Bedding perpendicular to axis 100 38.1 .0008 .0003 200 76.2 .0032 .0010 300 114.3 400 152.4 500 190.5 .0073 .0023 600 228.7 1.0 .0109 .0034 700 266.8 .0129 .0040 800 304.9 .0142 .0045 900 343 .0149 .0047 1000 381.1 1100 419.2 .0163 .0051 1200 457.3 1300 495.4 1.75 .0172 .0054 1400 533.5 .0235 .0074 1500 571.6 1600 609.8 .0292 .0092 1700 647.9 1800 686 - Brittle failure .0336 .0105 1900 724.1 - Sample Broke into 3 pieces with much crushed 2.5 .0413 .0130 2000 762.2 coal UNIAXIAL COMPRESSIVE STRENGTH (psi): 762 218 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: #1 coal seam upper seam TEST NUMBER: QC 5 SAMPLE LOCATION: #1 Mains B Road face TEST DATE: 21 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 15 December 1993 TEST PREPARATION: cored block DIAMETER (inch): 1.825 HEIGHT (inch): 3.016 AREA (sq. inch): 2.615 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (psi) NOTES .0294 .0097 2500 956 .0310 .0103 2600 994.3 3.0 .0344 .0114 2700 1032.5 - Brittle failure - Sample broke into 2 main pieces with several smaller ones and some crush UNIAXIAL COMPRESSIVE STRENGTH (psi): 1032 219 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: #1 coal seam upper seam TEST NUMBER: QC 6 SAMPLE LOCATION: #1 Mains B Road face TEST DATE: 21 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 15 December 1993 TEST PREPARATION: cored block DIAMETER (inch): HEIGHT (inch): AREA (sq. inch): TIME D E F O R M A T I O N STRAIN L O A D STRESS NOTES (minute) (inch) (inch/inch) (pounds) (psi) 0 .0000 .0000 0 0 .0007 .0002 200 76.5 .0016 .0005 300 114.7 .0025 .0008 400 153 .0030 .0010 500 191.2 .0047 .0016 600 229.4 .0076 .0025 700 267.7 .0087 .0029 800 305.9 1.0 .0103 .0034 900 344.2 .0110 .0037 1000 382.4 .0117 .0039 1100 420.7 .0121 .0040 1200 458.9 .0126 .0042 1300 497.1 .0130 .0043 1400 535.4 .0139 .0046 1500 573.6 1.5 .0145 .0048 1600 611.9 .0155 .0051 1700 650.1 .0183 .0061 1800 688.3 .0199 .0066 1900 726.6 2.0 .0208 .0069 2000 764.8 .0220 .0073 2100 803.1 .0234 .0078 2200 841.3 .0239 .0079 2300 879.5 2.5 .0250 .0083 2400 917.8 UNIAXIAL COMPRESSIVE STRENGTH (psi): 917 220 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: #1 coal seam upper seam TEST NUMBER: QC 7 SAMPLE LOCATION:#l Mains B Road face TEST DATE: 21 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 15 December 1993 TEST PREPARATION: cored block DIAMETER (inch): 1.825 HEIGHT (inch): 3.016 AREA (sq. inch): 2.615 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (psi) NOTES .0143 .0047 2200 841.3 .0147 .0049 2300 879.5 .0154 .0051 2500 956 .0159 .0053 2400 917.8 5.5 .0168 .0056 1600 611.9 - Fractures formed along cleat parallel to bedding. UNIAXIAL COMPRESSIVE STRENGTH (psi): 917 221 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: #1 coal seam upper seam TEST NUMBER: QC 8 SAMPLE LOCATION:#l Mains B Road face TEST DATE: 21 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 15 December 1993 TEST PREPARATION: cored block DIAMETER (inch): 1.825 HEIGHT (inch): 3.016 AREA (sq. inch): 2.615 TIME D E F O R M A T I O N STRAIN L O A D STRESS NOTES (minute) (inch) (inch/inch) (pounds) (psi) 0 .0000 .0000 0 0 100 38.2 .0013 .0004 200 76.5 .0027 .0009 300 114.7 .0035 .0012 400 153 .0046 .0015 500 191.2 600 229.4 3 .0083 .0028 700 267.7 .0094 .0031 800 306 .0106 .0035 900 344.2 .0107 .0036 1000 382.4 .011 .0037 1100 420.7 .0112 .0037 1200 458.9 .0116 .0039 1300 497.1 .0118 .0039 1400 535.4 .0121 .0040 1500 573.6 .0124 .0041 1600 611.9 .0127 .0042 1700 650.1 .0129 .0043 1800 688.3 .0133 .0044 1900 726.6 .0136 .0045 1800 688.3 .0138 .0046 2000 764.8 .0142 .0047 2100 803.1 UNIAXIAL COMPRESSIVE STRENGTH (psi): 803 222 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: #1 coal seam upper seam TEST NUMBER: QC 9 SAMPLE LOCATIONS 1 Mains B Road face TEST DATE: 21 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 15 December 1993 TEST PREPARATION: cored block DIAMETER (inch): 2.690 HEIGHT (inch): 6.600 AREA (sq. inch): 5.683 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS NOTES - Brittle failure - Fractures saw tooth along cleat 40 sec 3800 680 Failure load UNIAXIAL COMPRESSIVE STRENGTH (psi): 680 223 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: #1 coal seam - upper seam TEST NUMBER: QC 10 SAMPLE LOCATION:* 1 Mains B Road face TEST DATE: 21 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 15 December 1993 TEST PREPARATION: saw cut block DIAMETER (inch): HEIGHT (inch): 2.09 AREA (sq. inch): 3.738 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (psi) NOTES Brittle failure Numerous axial fractures parallel to cleat 35sec 6000 1604 UNIAXIAL COMPRESSIVE STRENGTH (psi): 1604 224 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: #1 coal seam - upper seam TEST NUMBER: QC 11 SAMPLE LOCATION:#l Mains B Road face TEST DATE: 21 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 15 December 1993 TEST PREPARATION: saw cut block DIAMETER (inch): HEIGHT (inch): 2.00 AREA (sq. inch): 4.2328 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (P s i) NOTES - Brittle failures - Numerous axial fractures parallel to cleat - Failed sample remained intact 31 sec 6300 1488 UNIAXIAL COMPRESSIVE STRENGTH (psi): 1488 225 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: #1 coal seam - upper seam TEST NUMBER: QC 12 SAMPLE LOCATION:#l Mains B Road face TEST DATE: 21 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 15 December 1993 TEST PREPARATION: saw cut block DIAMETER (inch): HEIGHT (inch): 2.435 AREA (sq. inch): 4.264 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (psi) NOTES - Brittle failure - Corners broke off - Center intact with axial fractures parallel to cleat 27sec 9860 2313 UNIAXIAL COMPRESSIVE STRENGTH (psi): 2313 226 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: #1 coal seam - upper seam TEST NUMBER: QC 13 SAMPLE LOCATION: #1 Mains B Road face TEST DATE: 21 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 15 December 1993 TEST PREPARATION: saw cut block DIAMETER (inch): HEIGHT (inch): 2.885 AREA (sq. inch): 4.101 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (P s i ) NOTES - Brittle failure - Axial fractures parallel to cleat - Concentrated failure at corners - Center of sample intact 30sec 7200 1755.7 UNIAXIAL COMPRESSIVE STRENGTH (psi): 1755 227 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: Weak siltstone TEST NUMBER: QC 14 SAMPLE LOCATIONS Mains B Road TEST DATE: 04 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 25 November 1993 TEST PREPARATION: cored block DIAMETER (inch): 1.812 HEIGHT (inch): 3.376 AREA (sq. inch): 2.578 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (P s i ) NOTES 0 1'25" 0 .0022 .0034 .0039 .0050 .0054 .0093 0 500 700 900 1300 1500 1000 0 193.9 271.5 349.1 504.2 581.8 387.9 Saw tooth @ 65° tea Brittle - some noise at failure sample remained in intact UNIAXIAL COMPRESSIVE STRENGTH (psi): 581 228 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: Parting siltstone TEST NUMBER: QC 15 SAMPLE LOCATIONS 1 Mains B Road TEST DATE: 04 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 25 November 1993 TEST PREPARATION: cored block DIAMETER (inch): 1.815 HEIGHT (inch): 3.378 AREA (sq. inch): 2.587 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (psi) NOTES 0 0 0 0 .0016 600 231.9 .0034 900 347.9 .0044 1400 541.2 .0062 2300 889.1 .0064 2900 1121 .0124 4600 1778.1 .0194 2600 1005 T02" 1100 425.2 Non violent - no noise at failure Cored perpendicular to bedding UNIAXIAL COMPRESSIVE STRENGTH (psi): 1778 229 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: Parting siltstone TEST NUMBER: QC 16 SAMPLE LOCATION:Sec 102 GOB TEST DATE: 23 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 15 December 1993 TEST PREPARATION: cored block DIAMETER (inch): 1.820 HEIGHT (inch): 3.018 AREA (sq. inch): 2.601 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (psi) NOTES 5 3 sec 1200 461.4 Non Violent UNIAXIAL COMPRESSIVE STRENGTH (psi): 461 230 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: Competent Siltstone TEST NUMBER: QC 17 SAMPLE LOCATION:#l Mains Portal TEST DATE: 04 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 21 November 1993 TEST PREPARATION: cored block DIAMETER (inch): 1.826 HEIGHT (inch): 3.153 AREA (sq. inch): 2.618 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (psi) NOTES 0 0 0 0 .0004 100 38.2 .0006 500 191 .0014 800 305.6 .0018 1900 725.7 .0019 2000 763.9 .0020 2200 840.3 .0021 2600 993.1 .0023 2800 1069.5 .0024 3000 1145.9 .0029 2600 993.1 Brittle failure 0'43" .0037 2400 916.7 Rapid loss of strength UNIAXIAL COMPRESSIVE STRENGTH (psi): 1146 231 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: Competent Siltstone TEST NUMBER: QC 18 SAMPLE LOCATION:#l Mains Portal TEST DATE: 04 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 21 November 1993 TEST PREPARATION: cored block DIAMETER (inch): 1.823 HEIGHT (inch): 3.395 AREA (sq. inch): 2.609 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (psi) NOTES 0 0 0 0 .0002 200 76.7 .0005 300 115 .0008 400 153.3 .0010 500 191.6 .0022 600 230 .0026 700 268 .0027 800 306.6 .0028 900 345 .0029 1000 383.3 .0030 1100 421.6 Brittle failure .0031 1300 498.3 Rapid loss of strength .0032 1500 574.9 .0033 1700 651.6 .0035 1900 728.2 .0037 2100 804.9 .0038 2300 881.6 .0039 2500 958.2 .0041 2700 1034.9 .0043 2900 1111.5 .0045 3100 1188.2 1 '51" .0095 1300 498.3 UNIAXIAL COMPRESSIVE STRENGTH (psi): 1188 232 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: Competent Siltstone TEST NUMBER: QC 19 SAMPLE LOCATION: Sec 102 GOB TEST DATE: 23 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 17 December 1993 TEST PREPARATION: cored block DIAMETER (inch): 1.800 HEIGHT (inch): 3.355 AREA (sq. inch): 2.544 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (psi) NOTES 45sec 4600 1808.2 Brittle failure Rapid loss of strength UNIAXIAL COMPRESSIVE STRENGTH (psi): 1808 233 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: Sandstone TEST NUMBER: QC 20 SAMPLE LOCATIONS S Pit - Blasted Rock TEST DATE: 23 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 18 December 1993 TEST PREPARATION: cored block DIAMETER (inch): 1.800 HEIGHT (inch): 3.660 AREA (sq. inch): 2.545 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (psi) NOTES 40sec 10300 4047.2 Brittle failure UNIAXIAL COMPRESSIVE STRENGTH (psi): 4047 234 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: Sandstone TEST NUMBER: QC 21 SAMPLE LOCATIONS S Pit - Blasted Rock TEST DATE: 23 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 18 December 1993 TEST PREPARATION: cored block DIAMETER (inch): 1.800 HEIGHT (inch): 4.220 AREA (sq. inch): 2.545 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (psi) NOTES 3 5 sec 14000 5501 Brittle failure Rapid loss of strength UNIAXIAL COMPRESSIVE STRENGTH (psi): 5501 235 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: Sandstone TEST NUMBER: QC 22 SAMPLE LOCATIONS S Pit - Blasted Rock TEST DATE: 23 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 18 December 1993 TEST PREPARATION: cored block DIAMETER (inch): 1.800 HEIGHT (inch): 4.325 AREA (sq. inch): 2.545 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (psi) NOTES 28sec 13000 5108 UNIAXIAL COMPRESSIVE STRENGTH (psi): 5108 236 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH UNIAXIAL COMPRESSIVE STRENGTH TEST ROCK TYPE: Sandstone TEST NUMBER: QC 23 SAMPLE LOCATIONS S Pit - Blasted Rock TEST DATE: 23 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 18 December 1993 TEST PREPARATION: cored block DIAMETER (inch): 1.800 HEIGHT (inch): 3.470 AREA (sq. inch): 2.545 TIME (minute) D E F O R M A T I O N (inch) STRAIN (inch/inch) L O A D (pounds) STRESS (psi) NOTES 20sec 11000 4322.2 Brittle failure Rapid loss of strength UNIAXIAL COMPRESSIVE STRENGTH (psi): 4322 237 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH POINT LOAD INDEX TESTING POINT LOAD INDEX PARALLEL TO BEDDING (Diametral Test) ROCK DDH DEPTH DIAMETER D (cm) LENGTH W (cm) GAUGE (psi * 100) LOAD (kN) Is (MPa) Is5o (MPa) sandstone med. -course 94-05 54.05 7.4 14 12 14.8 2.7 3.2 sandstone med. 94-05 56.7 7.4 15 13.5 16.7 3.0 3.6 sandstone med. + silt bands 94-05 59 7.4 9.5 0.1 0.1 0.0 0.0 siltstone 94-05 60 7.4 11 0.3 0.4 0.1 0.1 siltstone 94-05 62 7.4 12 0.5 0.6 0.1 0.1 sandstone very fine 94-05 63.5 (2.8m above coal) 7.4 8.5 0 0.0 0.0 0.0 silty sandstone very fine 94-05 64.8 (1.5m above coal) 7.4 14 0.2 0.2 0.0 0.1 silty sandstone very fine 94-05 65.8 (0.5m above coal) 7.4 8.0 0 0.0 0.0 0.0 siltstone 94-05 71.8 (under coal seam) 7.4 7.0 0.2 0.2 0.0 0.1 sandstone 94-04 64.3 7.4 10 10.5 13.0 2.4 2.8 sandstone 94-04 65.5 7.4 9.0 16.0 19.7 3.6 4.3 sandstone 94-04 67.0 7.4 9.5 9.0 11.1 2.0 2.4 sandstone 94-04 68.5 7.4 12.0 11.0 13.6 2.5 2.9 siltstone 94-04 69.5 7.4 14.5 0.3 0.4 0.1 0.1 siltstone 94-04 71.0 7.4 8.0 0.0 0.0 0.0 0.0 siltstone 94-04 72.5 (1.0m above coal) 7.4 9.5 0.0 0.0 0.0 0.0 coal (#1 seam) 94-04 77.0 7.4 10.5 0.1 0.1 0.0 0.0 siltstone 94.04 81.3 7.4 10.0 0.2 0.2 0.0 0.1 siltstone 94-20 117.5 (lm above #2 ) 7.4 12 0 0.0 0.0 0.0 siltstone 94-20 144.5 (2.0m above coal) 7.4 9.5 0.0 0.0 0.0 0.0 sandstone course 94-06 71.2 (2m above coal) 7.4 12 9.5 11.7 2.1 2.5 siltstone 94-06 72.8 (0.5m above coal) 7.4 8.5 0.1 0.1 0.0 0.0 coal 94-06 74 7.4 15 0.2 0.2 0.0 0.1 sandstone 94-07 38.5 7.4 10 24.0 29.6 5.4 6.4 sandstone course 94-07 39.5 7.4 15.5 1.0 1.2 0.2 0.3 siltstone 94-07 41.2 7.4 9.5 2.0 2.5 0.5 0.5 sandstone fine grain 94-07 42.5 7.4 12.5 0.5 0.6 0.1 0.1 sandstone med. -course 94-07 44.2 7.4 10.0 13.5 16.7 3.0 3.6 sandstone med. 94-07 45.5 7.4 9.5 19.0 23.4 4.3 5.1 sandy siltstone 94-07 47.0 7.4 9.5 0.1 0.1 0.0 0.0 siltstone 94-07 48.0 (0.2m above coal) 7.4 8.0 0.0 0.0 0.0 0.0 coal - #2 seam 94-07 48.8 7.4 9.0 0.1 0.1 0.0 0.0 siltstone + carb. banding 94-07 50.5 (1.0m below coal) 7.4 8.0 0.0 0.0 0.0 0.0 238 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH ROCK DDH DEPTH DIAMETER D (cm) LENGTH W (cm) GAUGE (psi * 100) LOAD (kN) Is (MPa) Is5o (MPa) siltstone 94-07 52.0 7.4 11.5 0 0.0 0.0 0.0 siltstone 94-07 53 7.4 10.0 0.3 0.4 0.1 0.1 siltstone 94-07 54.5 7.4 9.0 0.5 0.6 0.1 0.1 siltstone 94-07 56.4 (1.0m above coal) 7.4 8.5 0.1 0.1 0.0 0.0 siltstone + carb. banding 94-07 57.0 (0.4m above coal) 7.4 10 0.1 0.1 0.0 0.0 coal 94-07 58.0 7.4 12 0.2 0.2 0.0 0.1 coal 94-19 116.4 7.4 10 7.3 9.0 1.6 2.0 sandstone 94-19 109.0 7.4 10 15.0 18.5 3.4 4.0 sandstone 94-19 106 7.4 9.4 8.5 10.5 1.9 2.3 coal 94-17 151 7.4 9.0 0.3 0.4 0.1 0.1 siltstone 94-17 148 7.4 11.0 1.0 1.2 0.2 0.3 siltstone 94-17 149 7.4 6.0 0.5 0.6 0.1 0.1 sandstone 94-18 94.5 7.4 9.0 12 14.8 2.7 3.2 sandstone 94-18 96 7.4 9.0 10 12.3 2.3 2.7 sandstone 94-18 97.5 7.4 8.0 18 22.2 4.1 4.8 sandstone 94-18 98.5 7.4 12.0 16.0 19.7 3.6 4.3 sandstone 94-18 100.5 7.4 10.0 7.5 9.3 1.7 2.0 sandstone 94-18 101.5 (above #2 seam) 7.4 10.0 5.5 6.8 1.2 1.5 carbonaceous siltstone 94-18 103.0 (base of #2 seam) 7.4 0 0.0 0.0 0.0 siltstone 94-18 104.5 7.4 11.0 0.2 0.2 0.0 0.1 siltstone 94-18 108 7.4 11.5 0.2 0.2 0.0 0.1 siltstone+carb stringers 94-18 107 7.4 10.0 0.1 0.1 0.0 0.0 siltstone 94-18 109.5 7.4 10.5 0.5 0.6 0.1 0.1 siltstone 94-18 111.0 7.4 10.8 0.0 0.0 0.0 0.0 siltstone 94-18 112.5 7.4 10.5 0.5 0.6 0.1 0.1 siltstone 94-18 114.0 7.4 9.5 5.2 6.4 1.2 1.4 silty sandstone 94-18 115.5 7.4 8.0 1.5 1.9 0.3 0.4 sandstone 94-18 117.0 7.4 9.5 12.5 15.4 2.8 3.4 sandstone 94-18 118.5 7.4 11.0 13 16.0 2.9 3.5 sandstone 94-18 120.0 7.4 7.5 19.0 23.4 4.3 5.1 sandstone 94-18 120.0 7.4 7.5 21.0 25.9 4.7 5.6 silty sandstone 94-18 121.5 7.4 8.0 0.5 0.6 0.1 0.1 silty sandstone 94-18 123.0 7.4 9.0 13.0 16.0 2.9 3.5 siltstone 94-18 124 7.4 10.0 0.5 0.6 0.1 0.1 siltstone 94-18 124.5 (2.5m above coal) 7.4 9.0 12.5 15.4 2.8 3.4 siltstone 94-18 126.0 (lm above coal) 7.4 8.5 0 0.0 0.0 0.0 coal 94-18 127.5 7.4 9.5 0.5 0.6 0.1 0.1 239 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH POINT LOAD INDEX PERPENDICULAR TO BEDDING (Axial Test) ROCK DDH DEPTH DIAMETER D (cm) LENGTH W (cm) GAUGE (psi * 100) LOAD (kN) Is (MPa) Is5o (MPa) sandstone - course 94-05 54.05 7.4 6.0 17.5 21.6 3.9 4.8 sandstone med. 94-05 56.7 7.4 5.5 11.0 13.6 2.7 3.2 sandstone + silt 94-05 59 7.4 5.0 11.0 13.6 2.9 3.5 siltstone 94-05 60 7.4 4.0 6.0 7.4 2.0 2.2 siltstone 94-05 62 7.4 4.5 12.0 14.8 3.6 4.1 sandstone very fine 94-05 63.5 (2.8m above coal) 7.4 5.0 9.5 11.7 2.5 3.0 silty sandstone fine 94-05 64.8 (1.5m above coal) 7.4 6.5 8.5 10.5 1.8 2.2 silty sandstone fine 94-05 65.8 (0.5m above coal) 7.4 5.5 0.8 1.0 0.2 0.2 siltstone 94-05 71.8 (under coal seam) 7.4 7.0 1.0 1.2 0.2 0.2 sandstone 94-04 64.3 7.4 4.5 22.0 27.1 6.6 7.5 sandstone 94-04 65.5 7.4 4.5 19.0 23.4 5.7 6.5 sandstone 94-04 67.0 7.4 4.0 11.0 13.6 3.7 4.1 sandstone 94-04 68.5 7.4 4.5 14.0 17.3 4.2 4.8 siltstone 94-04 69.5 7.4 6.5 8.0 9.9 1.6 2.1 siltstone 94-04 71.0 7.4 3.5 4.8 5.9 1.8 2.0 siltstone 94-04 72.5 (1.0m above coal) 7.4 5.0 0.2 0.2 0.1 0.1 coal (#1 seam) 94-04 77.0 7.4 4.5 0.1 0.1 0.0 0.0 siltstone 94.04 81.3 7.4 5.0 0 0.0 0.0 0.0 siltstone 94-20 117.5 (lm above #2 ) 7.4 12 0 0.0 0.0 0.0 siltstone 94-20 144.5 (2.0m above coal) 7.4 9.5 0.0 0.0 0.0 0.0 sandstone course 94-06 71.2 (2m above coal) 7.4 12 9.5 11.7 1.1 1.5 siltstone 94-06 72.8 (0.5m above coal) 7.4 8.5 0.1 0.1 0.0 0.0 coal 94-06 74 7.4 15 0.2 0.2 0.0 0.0 sandstone 94-07 38.5 7.4 7.5 27.5 33.9 4.9 6.4 sandstone course 94-07 39.5 7.4 7.0 8.0 9.9 1.5 2.0 siltstone 94-07 41.2 7.4 7.0 7.0 8.6 1.3 1.7 sandstone fine grain 94-07 42.5 7.4 5.5 10.0 12.3 2.4 2.9 sandstone course 94-07 44.2 7.4 4.5 11.0 13.6 3.3 3.7 sandstone med. 94-07 45.5 7.4 5.0 13.0 16.0 3.5 4.1 sandy siltstone 94-07 47.0 7.4 6.0 12.5 15.4 2.8 3.4 siltstone 94-07 48.0 (0.2m above coal) 7.4 4.5 0.1 0.1 0.0 0.0 coal - #2 seam 94-07 48.8 7.4 4.5 0.2 0.2 0.1 0.1 siltstone + carb. 94-07 50.5 (1.0m below coal) 7.4 4.0 0.8 1.0 0.3 0.3 siltstone 94-07 52.0 7.4 5.0 0.5 0.6 0.1 0.2 siltstone 94-07 53 7.4 5.0 6.0 7.4 1.6 1.9 siltstone 94-07 54.5 7.4 3.5 10.0 12.3 3.8 4.1 siltstone 94-07 56.4 (1.0m above coal) 7.4 4.0 1.0 1.2 0.3 0.4 siltstone +carb. 94-07 57.0 (0.4m above coal) 7.4 4.5 2.5 3.1 0.7 0.8 coal 94-07 58.0 7.4 5.0 0.3 0.4 0.1 0.1 240 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH ROCK DDH DEPTH DIAMETER LENGTH GAUGE LOAD Is Is50 D (cm) W (cm) (psi * 100) (kN) (MPa) (MPa) coal 94-19 116.4 7.4 5.0 0 0.0 0.0 0.0 sandstone 94-19 109.0 7.4 6.0 17.0 21.0 3.8 4.7 sandstone 94-19 106 7.4 7.6 20.0 24.7 3.5 4.6 coal 94-17 151 7.4 5.0 0.8 1.0 0.2 0.3 siltstone 94-17 148 7.4 4.0 7.0 8.6 2.3 2.6 siltstone 94-17 149 7.4 5.0 10.0 12.3 2.7 3.1 sandstone 94-18 94.5 7.4 5.0 2.0 2.5 0.5 0.6 sandstone 94-18 96 7.4 4.5 8.0 9.9 2.4 2.7 sandstone 94-18 97.5 7.4 4.0 17.5 21.6 5.9 6.5 sandstone 94-18 98.5 7.4 5.0 15.5 19.1 4.2 4.9 sandstone 94-18 100.5 7.4 4.0 5.5 6.8 1.8 2.0 sandstone 94-18 101.5 (above #2 seam) 7.4 5.5 7.8 9.6 1.9 2.3 carbonaceous silt. 94-18 103.0 (base of #2 seam) 7.4 4.0 2.8 3.5 0.9 1.0 siltstone 94-18 104.5 7.4 4.5 6.5 8.0 1.9 2.2 siltstone 94-18 108 7.4 5.5 0.1 0.1 0.0 0.0 siltstone + carb 94-18 107 7.4 5.0 0.5 0.6 0.1 0.2 siltstone 94-18 109.5 7.4 5.0 1-5 1.9 0.4 0.5 siltstone 94-18 111.0 7.4 5.0 1.0 1.2 0.3 0.3 siltstone 94-18 112.5 7.4 5.0 3.5 4.3 0.9 1.1 siltstone 94-18 114.0 7.4 5.5 8.1 10.0 2.0 2.4 silty sandstone 94-18 115.5 7.4 3.5 7.8 9.6 3.0 3.2 sandstone 94-18 117.0 7.4 3.5 11.0 13.6 4.2 4.5 sandstone 94-18 120.0 7.4 3.0 15 18.5 6.7 6.9 silty sandstone 94-18 121.5 7.4 4.5 10.0 12.3 3.0 3.4 silty sandstone 94-18 123.0 7.4 13.0 4.0 4.9 0.4 0.6 siltstone 94-18 124 7.4 6.5 3.5 4.3 0.7 0.9 siltstone 94-18 124.5 (2.5m above coal) 7.4 4.5 8.5 10.5 2.5 2.9 siltstone 94-18 126.0 (lm above coal) 7.4 5.5 .5 0.6 0.1 0.1 coal 94-18 127.5 7.4 4.0 6.0 7.4 2.0 2.2 241 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH DIRECT SHEAR STRENGTH TESTS DIRECT SHEAR STRENGTH TEST ROCK TYPE: coal TEST NUMBER: QC DS 3 SAMPLE LOCATION: 2N Mine #1 Mains B TEST DATE: 21 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 15 December 1993 TEST PREPARATION: cored block DIAMETER (inch): 2.700 HEIGHT (inch): AREA (sq. inch): 5.72 NORMAL LOAD (pounds): 200 PISTON DIAMETER (inch): 1.65 NORMAL STRESS (psi): 32 PISTON AREA (sq. inch): 2138 TIME (minute) D I S P L A C E M E N T (inch) PRESSURE (pounds) STRESS (psi) NOTES 0 .0000 0 0 Tested Parallel to bedding. Unstable displacement measurements omitted. 650 113.6 peak Saw tooth failure across bedding planes. Concrete cast failed PEAK SHEAR STRENGTH (psi): 113.6 RESIDUAL SHEAR STRENGTH (psi): 242 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH DIRECT SHEAR STRENGTH TEST ROCK TYPE: coal TEST NUMBER: QC DS 4 SAMPLE LOCATION: 2N Mine #1 Mains B TEST DATE: 21 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 15 December 1993 TEST PREPARATION: cored block DIAMETER (inch): 2.700 HEIGHT (inch): AREA (sq. inch): 5.72 NORMAL LOAD (pounds): 500 PISTON DIAMETER (inch): 1.65 NORMAL STRESS (psi): 87 PISTON AREA (sq. inch): 2138 TIME (minute) D I S P L A C E M E N T (inch) PRESSURE (pounds) STRESS (psi) NOTES 0 .0000 0 0 Tested Parallel to bedding. Unstable displacement measurements omitted. 680 119 Peak strength Saw tooth failure across bedding planes. 0 0 Residual strength. PEAK SHEAR STRENGTH (psi): 119 RESIDUAL SHEAR STRENGTH (psi): 0 243 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH DIRECT SHEAR STRENGTH TEST ROCK TYPE: coal TEST NUMBER: QC DS 5 SAMPLE LOCATION: 2N Mine #1 Mains B TEST DATE: 21 December 1993 SAMPLE TYPE: block TESTED BY: M . Cullen SAMPLE DATE: 15 December 1993 TEST PREPARATION: cored block DIAMETER (inch): 2.700 HEIGHT (inch): AREA (sq. inch): 5.72 NORMAL LOAD (pounds): 500 PISTON DIAMETER (inch): 1.65 NORMAL STRESS (psi): 87 PISTON AREA (sq. inch): 2138 TIME (minute) D I S P L A C E M E N T (inch) PRESSURE (pounds) STRESS (psi) NOTES 0 .0000 0 0 Tested Parallel to bedding. Unstable displacement measurements omitted. 600 104 Peak strength Saw tooth failure across bedding planes. PEAK SHEAR STRENGTH (psi): 104 RESIDUAL SHEAR STRENGTH (psi): 244 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH SLAKE DURABILITY TESTS slaking medium was tap water at 18° C in all cases. Sample description: #1 seam coal, upper section of seam Sample location: #1 Mains, C Road face, 15, Dec. 1993 Number of tests: 1 Slake Durability index: 97.0 notes: coal edges and corners rounded, several small chips created, fine material in tank with some suspended solids. Sample description: #1 seam coal, top of seam at parting contact Sample location: #1 Mains, B Road face, 25, Nov. 1993 Number of tests: 1 Slake Durability index: 97.1 notes: coal edges and corners rounded, 5% 2-10mm chips created, Minus 2mm material in tank was rectangular chips with some suspended solids. Sample description: siltstone parting, competent rock. Sample location: #1 Mains, B Road face, 25 Nov. 1993, 1-02 panel, 14 Dec. 1993. Number of tests: 2 Slake Durability index: 94.2, 79.2 notes: material broken into many platy chips: 60% 2-10mm, 40% 10-30mm. Minus 2mm material 90% 0.5-2mm 10% very fine, some suspended solids. Sample description: siltstone parting, soft mud-like material. Sample location: 1-02 panel, 15 Dec. 1993. Number of tests: 2 Slake Durability index: 5.9, 4.2 notes: a few rounded platy pieces remained in basket. Minus 2 mm material platy with some very fine suspended solids. Sample description: Rider seam. Sample location: 1-02 panel gob, 14, Dec. 1993. Number of tests: 2 Slake Durability index: 97.7, 97.5 notes: Corners and sharp edges rounded, a few smaller pieces of rock created. Minus 2 mm material is fine with some suspended solids. Sample description: siltstone above Rider seam. Sample location: #1-02 panel gob 14 Dec. 1993 Number of tests: 1 Slake Durability index: 98.8 notes: Edges rounded, a few small chips created. Minus 2mm material all very fine. 245 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Sample description: sandstone Sample location: 2S pit Number of tests: 1 Slake Durability index: 96.8 notes: Edges rounded. Minus 2mm material very fine. MOISTURE CONTENT TESTS #1 coal seam #1 Mains face: 2.2 % siltstone parting competent, #1 Mains face: 3.2% siltstone parting soft, #1 Mains face: 7.6% DENSITY TESTS The test method utilised water displacement followed by pulverisation and treatment with a surfactant. Due to unexpected surfactant reaction it was not possible to determine porosity except with coal material. R O C K L O C A T I O N DENSITY POROSITY DENSITY GROUND (%) R O C K R O C K (g/cc) (g/cc) coal 2N Mine #1 Mains 1.2, 1.3 - 29 1.8, 1.7 mudstone Panel 102 2.4, 2.5 rider coal Panel 102 1.3, 1.4 28 1.8 siltstone Panel 102 2.7,2.8 sandstone 2S Pit 2.6, 2.6 246 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Q ROCK MASS CLASSIFICATION PARAMETERS R O C K L O C A T I O N RQD Jn Jr Ja Jw SRF Q weak siltstone 2N Mine #2 Mains 30 9 .75 3 1 2.5 0.3 siltstone 2N Mine portal 40 6 1 1.5 1 2.5 1.8 coal Panel 202 60 6 2 1.0 1 2.5 8 coal Panel 203 60 6 3 .75 1 2.5 16 siltstone 2N Mine #1 Mains 30 6 .75 2 1 2.5 0.8 sandstone 2S 90 6 3 1.0 1 1 45 mudstone 2N Mine #1 Mains 20 12 .75 4 1 5 0.1 RMR ROCK MASS CLASSIFICATION PARAMETERS ROCK LOCATION RQD STRENGTH JOINT SPACING JOINT CONDITION GROUND WATER RMR coal 2N Mine #1 Mains 8 2 5 25 10 45 siltstone 2N Mine #1 Mains 8 2 5 20 10 40 sandstone 2S 18 4 25 25 10 82 mudstone 2N Mine #1 Mains 3 1 5 5 10 24 weak siltstone 2N Mine #1 Mains 3 1 5 10 10 29 CMRR ROCK MASS CLASSIFICATION PARAMETERS (bolted horizon) R O C K Strength MPa Moisture Sensitivity Discontinuities cohesion roughness spacing persistence number Ground Water C M R R siltstone competent 10 to 30 slight mod. to slick wavy to planar >1.8 to <0.05 >9 to <1.0 2 to 3 damp 35 to 45 sandstone 40 to 60 not 2 2 2 damp 80 coal #1 rider 20 not mod wavy to planar >1.8 to <0.05 >9to <1.0 2to3 damp 35 to 40 weak siltstone 5 to 10 moderate weak to slick wavy to planar <1.8to O.05 >9to <2.0 3 damp 25 to 35 247 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH APPENDIX 2 ROOF CONVERGENCE MEASUREMENT DATA 2 4 8 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH ROD EXTENSOMETER DATA L O C A T I O N C O N V E R G E N C E R A T E (mrn/min) E X C A V A T I O N A G E (days) TIME TO C A V E (days) C O M M E N T S #2 M A I N S 0.003175 360.0 5 edge of gob #2 M A I N S 0 10.0 2 edge of gob #2 M A I N S 0.000169333 1.0 2 inside gob #2 M A I N S 0.001778 360.0 5 inside gob #2 M A I N S 0 360.0 16 gob edge #2 M A I N S 0.003175 360.0 16 gob edge #2 M A I N S 0.001989667 400.0 16 gob edge #2 M A I N S 0.002159 400.0 15 inside gob #2 M A I N S 0.001989667 400.0 15 inside gob #2 MAINS 0.001905 500.0 15 inside gob #2 M A I N S 0 2.0 15 inside gob #2 M A I N S 0.000211667 500.0 15 gob edge #2 M A I N S 0.001566333 2.0 15 inside gob #2 M A I N S 0.003471333 2.0 15 inside gob #2 M A I N S 0.001651 2.0 15 inside gob #2 M A I N S 0.001905 2.0 15 inside gob #2 M A I N S 0 730.0 15 gob edge #2 M A I N S 0.004826 7.0 0.5 gob edge - ribs working #2 M A I N S 0 5.0 11 gob edge - ribs working #2 M A I N S 0.000846667 0.1 0.5 in gob #2 MAINS 0.005926667 7.0 0.5 gob edge - ribs working 2S-102 0 14.0 2 edge of gob 2S-102 0 14.0 2 edge of gob 2S-102 0 21.0 3 edge of gob 2S-102 0 21.0 3 edge of gob 2S-102 0 28.0 2 edge of gob 2S-102 0 0.0 2 inside gob 2S-102 0.001693333 14.0 2 gob edge #2 M A I N S 0.001947333 0.1 5 just cut #2 M A I N S 0.001820333 0.1 5 just cut #2 MAINS 0.001778 0.1 5 just cut #2 MAINS 0.001905 0.1 5 just cut #2 MAINS 0.002667 0.1 5 just cut 2S-102 0.00381 0.2 14 roadway just cut 2S-102 0.00381 0.2 14 roadway just cut 2S-102 0.005503333 0.2 14 roadway just cut 2S-102 0 0.2 14 roadway just cut 2S-102 0 0.1 14 roadway just cut #2 M A I N S 0.000973667 0.1 5 just cut #2 M A I N S 0.002328333 0.1 5 just cut just cut #2 M A I N S 0.002116667 0.1 5 just cut #2 M A I N S 0.003259667 0.1 5 just cut #2 MAINS 0 0.1 7 just cut 249 GEOTECHNICAL STUDIES OF R E T R E A T PILLAR C O A L MINING A T S H A L L O W DEPTH APPENDIX 3 ROCK BOLT PULL TEST DATA 25o GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH RESULTS OF PULL TESTS ON ROOFBOLTS DATE LOCATION ROCK AT LOAD COMMENTS MINE, SECTION ANCHOR (tonne) 4/8/94 2N, 205 B Rd siltstone 11.3 nut stripped 4/8/94 2N, 205 B Rd siltstone 12.3 holding 4/8/94 2N, 205 A Rd siltstone 0 4/8/94 2N, 205 A Rd siltstone 9.3 holding 4/8/94 2N, #2 Mains siltstone 12.3 holding 4/8/94 2N, #2 Mains siltstone 12.3 holding 4/8/94 2N, #2 Mains siltstone 13.8 4/8/94 2N, #2 Mains siltstone 12.3 holding 4/8/94 2N, #3 Mains siltstone 12.3 holding 4/8/94 2N, #2 Mains siltstone 12.3 holding 4/8/94 2N, 302 siltstone 12 nut stripped 4/8/94 2N, 302 siltstone 11.6 4/8/94 2N, 302 siltstone 11.3 holding 4/8/94 2N, 302 siltstone 12.3 holding 8/12/93 2N, #1 Mains parting 19.7 8/12/93 2N, #1 Mains parting 20.4 8/12/93 2N, #1 Mains parting 16.8 8/12/93 2N, #1 Mains parting 14.7 nut stripped 10/11/93 2N,#1 Mains siltstone 16.4 10/11/93 2N,#1 Mains siltstone 19.7 10/11/93 2N,#1 Mains siltstone 16.4 14/12/93 2N,#1 Mains parting 18.5 14/12/93 2N,#1 Mains parting 19.7 14/12/93 2N,#1 Mains parting 18.5 15/03/93 2S, 101 siltstone 14 nut stripped - pinch thread bolt 15/03/93 2S, 101 siltstone 19 holding - pinch thread bolt 15/03/93 2S, 101 siltstone 21 nut stripped - pinch thread bolt 15/03/93 2S, 101 siltstone 19.7 nut stripped - pinch thread bolt 18/03/93 2S, 103 siltstone 8 resin/rock failure - 250mm resin encapsulation 18/03/93 2S, 103 siltstone 9.8 resin/rock failure - 250mm resin encapsulation 18/03/93 2S, 103 siltstone 14 nut stripped - 250mm resin encapsulation 6/4/94 2N,103 rider seam 11 6/4/94 2N,103 rider seam 11.0 holding - torque = 225ftlb 6/4/94 2N,103 rider seam 11.0 holding - torque = 100 ftlb 6/4/94 2N.103 rider seam 11.0 holding - torque = 100 ftlb 6/4/94 2N,103 rider seam 10.0 resin/rock failure - torque = 100 ftlb 6/4/94 2N,103 rider seam 9.5 resin/rock failure - torque = 100 ftlb 21/04/94 2N,103 rider seam 11.3 holding - torque= 100 ftlb 21/04/94 2N,103 rider seam 8.5 resin/rock failure - torque = 100 ftlb 21/04/94 2N,103 parting 7.7 resin rock failure - 254mm resin encapsulation 21/04/94 2N.103 parting 10.2 resin rock failure - 254mm resin encapsulation 21/04/94 2N,103 parting 9.0 resin rock failure - 254mm resin encapsulation 14/06/94 2S,#1 Mains siltstone 17.0 holding - pinch thread bolt 14/06/94 2S, #1 Mains siltstone 17.0 holding - pinch thread bolt 14/06/94 2S, #1 Mains siltstone 17.0 holding - pinch thread bolt 14/06/94 2S,#1 Mains siltstone 17.0 holding - pinch thread bolt 09/06/94 2S,#1 Mains siltstone 14.5 nut stripped 09/06/94 2S,#1 Mains siltstone 16.0 nut stripped 09/06/94 2S, #1 Mains siltstone 17.0 nut stripped 10/8/94 2N, 103 parting 11.0 resin/rock failure - 254mm resin encapsulation 10/8/94 2N, 103 parting 7.8 resin/rock failure - 254mm resin encapsulation 10/8/94 2N, 103 parting 9.5 resin/rock failure - 254mm resin encapsulation 251 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH DATE LOCATION ROCK AT LOAD COMMENTS MINE, SECTION ANCHOR (tonne) 10/8/94 2N, #4 Mains parting 13.1 holding - torque = 100 ftlb 10/8/94 2N, #4 Mains parting 13.0 holding - torque = 100 ftlb 17/8/93 2N, 204 siltstone 17.2 holding - pinch thread bolt 21/2/94 2S, #1 Mains siltstone 16.4 holding - pinch thread bolt 21/2/94 2S, #1 Mains siltstone 16.4 holding - pinch thread bolt 24/2/94 2S, portal siltstone 8.8 resin/rock failure - 35mm drill hole, 23mm resin 24/2/94 2S, portal siltstone 1.0 resin/rock failure - 35mm drill hole, 23mm resin 24/2/94 2S, portal siltstone 1.0 resin/rock failure - 35mm drill hole, 23mm resin 9/2/94 2S, 101 siltstone 12.0 holding - 254mm resin encapsulation 9/2/94 2S, 101 siltstone 12.0 holding - 254mm resin encapsulation 9/2/94 2S, 101 siltstone 12.0 holding - 254mm resin encapsulation 9/2/94 2S, 101 siltstone 2.7 resin failure - poor mixing - 254mm resin encapsulation 9/2/94 2S, 101 siltstone 13.0 holding - 254mm resin encapsulation 9/2/94 2N, #2 Mains siltstone 13.0 holding - 254mm resin encapsulation 9/2/94 2N, #2 Mains siltstone 13.0 holding - 254mm resin encapsulation 9/2/94 2N, #2 Mains siltstone 13.0 holding - 254mm resin encapsulation 15/07/92 siltstone 9.1 resin failure - 33mm hole, 355mm encapsulation 15/07/92 siltstone 13.1 resin failure - 33mm hole, 355mm encapsulation 15/07/92 siltstone 20 holding - 33mm hole, 355mm encapsulation 6/01/93 2N,205 siltstone 16.4 resin/rock failure - 35mm hole 6/01/93 2N,205 siltstone 9.8 nut stripped 6/01/93 2N,205 siltstone 16.4 resin/rock failure 6/01/93 2N.205 siltstone 15.5 resin/rock failure 6/01/93 2N,205 siltstone 15.5 resin/rock failure 9/9/92 2N, #3 Mains rider seam 11.8 resin failure - 35mm hole, 400mm encapsulation 9/9/92 2N, #3 Mains rider seam 11.8 resin failure - 35mm hole, 400mm encapsulation 9/9/92 2N, #3 Mains siltstone 10.0 resin failure - 35mm hole, 400mm encapsulation 9/9/92 2N, #3 Mains siltstone 11.3 resin failure - 35mm hole, 400mm encapsulation 9/9/92 2N, #3 Mains siltstone 9.7 resin failure - 35mm hole, 400mm encapsulation 9/9/92 2N, #3 Mains siltstone 9.3 resin failure - 35mm hole, 400mm encapsulation 9/9/92 2N, #3 Mains rider seam 14.7 resin failure - 35mm hole, 400mm encapsulation 9/9/92 2N, #3 Mains rider seam 13.6 resin failure - 35mm hole, 400mm encapsulation 9/9/92 2N, #3 Mains rider seam 15.3 resin failure - 35mm hole, 400mm encapsulation 9/9/92 2N, #3 Mains rider seam 14.0 resin failure - 35mm hole, 400mm encapsulation 9/9/92 2N, #3 Mains siltstone 11.3 resin failure - 35mm hole, 400mm encapsulation 9/9/92 2N, #3 Mains siltstone 9.8 resin failure - 35mm hole, 400mm encapsulation 5/04/93 2N, 301 siltstone 15.2 nut stripped 5/04/93 2N, 301 siltstone 17.9 resin failure 5/04/93 2N, 301 siltstone 17.5 nut stripped 5/04/93 2N, 301 siltstone 14.3 nut stripped 5/04/93 2N, 301 siltstone 16.4 resin failure 14/02/94 2N, #1 Mains siltstone 14.2 holding 14/02/94 2N, #1 Mains siltstone 14.2 holding 15/07/92 5.9 6 ft. Split Set, 33mm hole 15/07/92 5.9 6 ft. Split Set, 33mm hole 15/07/92 siltstone 7.9 mechanical bolt, B12 short bail soft rock anchor 15/07/92 siltstone 7.9 mechanical bolt, Dl long bail soft rock anchor 15/07/92 siltstone 9.3 mechanical bolt, Dl long bail soft rock anchor 15/07/92 2.9 5 ft. Split Set 15/07/92 3.9 5 ft. Split Set 15/07/92 4.5 5 ft. Split Set 252 G E O T E C H N I C A L S T U D I E S O F R E T R E A T P I L L A R C O A L M I N I N G A T S H A L L O W D E P T H APPENDIX 4 VISUAL PILLAR CLASSIFICATION RECORDS AND FACTOR OF SAFETY VALUES 253 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH PILLAR CLASSIFICATION RECORDS LOCATION No Pillar Dimensions (m) Depth (m) Age (days) Dist. to Gob (m) Parting Thick (m) Visual Rating F O S Ind.1 FOS2 A R M P S Length Width Height 2N #2Mains 1 12 15 2.5 82 3 15 0.5 2.5 3.4 1.9 2N #2Mains 2 12 15 2.5 82 3 0 0.5 2.5 2 1.2 2N #2Mains 19 12 15 2.5 82 35 20 0.5 3 3.4 1.9 2N #2Mains 19 10 24 2.5 82 3 12 0.5 1 3.3 2.6 2N #2Mains 20 10 24 2.5 82 54 0 0.5 3 2.3 2.6 2N #2Mains 21 8 11 2.5 82 7 0 0.5 4 1 1.2 2N #2Mains 22 6 17 2.5 82 4 0 0.5 4 0.9 2N #2Mains 23 8 12 2.5 82 2 0 0.5 4 1 1.2 2N #2Mains 24 11 12 2.5 82 1 0 0.5 4 1.4 1.2 2N #2Mains 25 6 29 2.5 82 1 0 0.5 4 1.4 2N #2Mains 26 12 51 2.5 82 3 20 0.5 1 4.8 3.3 2N #2Mains 27 12 51 2.5 82 3 20 0.5 1 4.8 3.3 2N #2Mains 28 10 12 2.5 82 11 25 0.5 1 2.5 2.2 2N #2Mains 3 12 40 2.5 82 3 15 0.5 I 4.7 3.3 2N #2Mains 3 12 40 2.5 82 12 15 0.5 2 4.7 3.3 2N #2Mains 4 8 26 2.5 82 300 18 0.5 1 2.7 1.8 2N #2Mains 4 8 26 2.5 82 335 10 0.5 2 2.7 1.8 2N #2Mains 5 11 11 2.5 85 5 0 0.5 2 1.3 1.1 2N #2Mains 5 11 11 2.5 85 9 0 0.5 2.5 1.3 1.1 2N #2Mains 29 12 38 2.5 82 5 0 0.5 1 4.6 2.9 2N #2Mains 30 12 38 2.5 84 20 10 0.5 1.5 4.6 3.3 2N #2Mains 6 12 20 2.5 85 5 0 0.5 1.5 2.5 2.3 2N #2Mains 6 12 20 2.5 85 9 0 0.5 2 2.5 2.3 2N #2Mains 31 6 13 2.5 84 20 0 0.5 2 0.9 2N #2Mains 7 12 26 2.5 85 6 12 0.5 0.5 4.2 3.3 2N #2Mains 7 12 26 2.5 85 9 0 0.5 2 2.9 2.9 2N #2Mains 8 12 26 2.5 85 6 30 0.5 0 4.2 3.9 2N #2Mains 8 12 26 2.5 85 9 10 0.5 2 4.2 3.3 2N #2Mains 9 6 38 2.5 84 20 0 0.5 2 1.7 2N #2Mains 9 6 38 2.5 84 23 0 0.5 2 1.7 2N #2Mains 32 11 11 2.5 85 3 60 0.5 0 2.5 2.2 2N #2Mains 10 10 41 2.5 85 3 25 0.5 0 3.9 3.5 2N #2Mains 10 10 41 2.5 85 7 20 0.5 0.5 3.9 3.5 2N #2Mains 10 10 41 2.5 85 15 0 0.5 1 2.9 2.2 2N #2Mains 11 12 32 2.5 85 3 25 0.5 0 4.4 3.6 2N #2Mains 11 12 32 2.5 85 7 20 0.5 0.5 4.4 3.6 2N #2Mains 11 12 32 2.5 85 15 0 0.5 1.5 2.9 2.9 2N #2Mains 12 11 30 2.5 85 3 40 0.5 0 4 3.8 2N #2Mains 12 11 30 2.5 85 7 20 0.5 0.5 4 3.2 1 The factor of safety for individual pillars was calculated using the Mark-Bieniawski pillar strength formula. 2 The factor of safety for A R M P S considers a group of pillars. Both fos determinations account for abutment stresses. 254 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH LOCATION No Pillar Dimensions (m) Depth Age Dist. Parting Visual FOS FOS2 Length Width Height (m) (days) to Gob (m) Thick (m) Rating Ind.' A R M P S 2N #2Mains 12 11 30 2.5 85 25 0 0.5 1 2.8 2.8 2N #2Mains 13 12 30 2.5 85 3 40 0.5 0 4.4 3.8 2N #2Mains 13 12 30 2.5 85 7 25 0.5 0.5 4.4 3.2 2N #2Mains 13 12 30 2.5 85 25 0 0.5 1.5 3.2 2.9 2N #2Mains 33 11 30 2.5 85 300 40 0.5 0 4 3.8 2N #2Mains 34 6 30 2.5 85 304 0 0.5 1 1.5 2N #2Mains 35 6 30 2.5 85 30 0 0.5 2 1.5 2N #2Mains 36 5.5 14 2.5 85 30 0 0.5 2.5 0.9 2N #2Mains 14 29 35 2.5 85 300 35 0.5 0 10 4 2N #2Mains 14 29 35 2.5 85 302 10 0.5 0.5 10 3.6 2N #2Mains 14 29 35 2.5 85 303 10 0.5 1.5 10 3.6 2N #2Mains 15 11 26 2.5 82 1 90 0.5 0 3.8 3.8 2N #2Mains 15 11 26 2.5 82 20 24 0.5 1 3.8 3.8 2N #2Mains 16 11 26 2.5 82 1 90 0.5 0 3.8 3.8 2N #2Mains 16 11 26 2.5 82 20 24 0.5 0.5 3.8 3.8 2N #2Mains 17 11 26 2.5 82 1 90 0.5 0 3.8 3.8 2N #2Mains 17 11 26 2.5 82 15 24 0.5 1.5 3.8 3.8 2N #2Mains 37 12 42 2.5 82 10 50 0.5 0 4.7 3.9 2N #2Mains 38 12 42 2.5 82 10 50 0.5 0 4.7 3.9 2N #2Mains 39 12 42 2.5 82 10 50 0.5 0 4.7 3.9 2NS101 1 8 10 2.5 50 1 0 0.5 2 1.9 2NS101 2 5 12 2.5 55 2 0 0.5 2 1.3 2NS101 3 5 16 2.5 50 1 0 0.5 1 1.6 2N S101 4 12 36 2.5 55 100 10 0.5 2.5 7.2 5.1 2N S101 5 15 16 2.5 52 100 10 0.5 2 6.4 4.6 2NS101 6 12 30 2.5 50 100 0 0.5 0 5.5 4.9 2N S101 7 6 30 2.5 50 1 0 0.5 0 2.7 2NS101 8 15 18 2.5 55 100 10 0.5 0 6.8 5.2 2N S101 8 15 18 2.5 55 102 5 0.5 1 6.8 5 2NS101 9 5 12 2.5 55 2 0 0.5 2 1.4 2N S101 10 6 10 2.5 55 2 0 0.5 2 1.5 2NS101 11 6 16 2.5 52 2 0 0.5 2 2 2NS101 12 6 36 2.5 55 1 0 0.5 0 2.8 2NS101 13 6 38 2.5 50 5 0 0.5 0 2.9 2NS101 14 6 32 2.5 46 4 0 0.5 1 2.7 2N S101 15 6 6 2.5 55 2 0 0.5 3 0.9 2NS101 16 6 13 2.5 55 2 0 0.5 3 1.7 2NS101 17 12 19 2.5 55 100 10 0.5 0.5 5.8 4.3 2N S101 18 15 19 2.5 57 100 5 0.5 2.5 7 5 2NS101 19 6 8 2.5 57 100 0 0.5 5 1.2 2NS101 20 8 12 2.5 57 100 0 0.5 2.5 1.7 2NS101 21 12 30 2.5 57 100 15 0.5 1.5 6.9 5.1 2NS101 22 6 30 2.5 48 2 0 0.5 1.5 3.3 2NS101 23 12 34 2.5 50 100 10 0.5 1 7.1 5.1 2NS101 24 12 34 2.5 50 100 10 0.5 1 7.1 5.1 2NS101 24 12 34 2.5 50 110 0 0.5 1 5.8 5.1 255 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH LOCATION No Pillar Dimensions (m) Depth (m) Age (days) Dist. to Gob (m) Parting Thick (m) Visual Rating F O S Ind.1 FOS2 A R M P S Length Width Height 2NS101 25 12 34 2.5 52 100 10 0.5 0.5 7.1 5.1 2NS101 25 12 34 2.5 52 110 0 0.5 1 5.8 4.9 2NS101 26 12 34 2.5 55 100 10 0.5 0 7.1 5.1 2NS101 27 6 14 2.5 55 2 0 0.5 2 1.8 2NS101 28 6 12 2.5 55 1 0 0.5 5 1.7 2NS101 29 12 32 2.5 55 100 5 0.5 0.5 7 4.9 2NS101 30 12 20 2.5 55 100 5 0.5 0.5 5.9 4.9 2N S101 31 12 34 2.5 48 100 0 0.5 2 5.8 4.9 2N S101 32 12 34 2.5 54 100 0 0.5 0.5 5.8 4.9 2NS101 33 12 34 2.5 56 100 0 0.5 0 5.8 4.9 2NS101 34 12 34 2.5 58 100 10 0.5 2 7.1 5.1 2NS101 35 6 28 2.5 56 1 0 0.5 3.5 2.6 2NS101 36 7 12 2.5 58 1 0 0.5 3.5 1.8 2N S101 37 6 10 2.5 58 1 0 0.5 5 1.3 2N Sec401 1 11 12 2.5 93 110 0 1 1.4 2.3 2N Sec401 2 6 10 2.5 93 110 0 1 0.6 2N Sec401 3 5 10 2.5 93 110 0 1 0.5 2N Sec401 4 11 11 2.5 93 110 12 1.5 2.1 2 2N Sec401 5 11 11 2.5 93 110 20 1 2.1 2 2N Sec401 5 11 11 2.5 93 150 20 1 2.1 2 2N Sec401 6 11 11 2.5 93 110 10 1 2.1 2 2N Sec401 6 11 11 2.5 93 150 10 1 2.1 2 2N Sec401 7 11 47 2.5 75 90 18 0.5 4.5 3.8 2N Sec401 8 11 47 2.5 75 90 18 0 4.5 3.8 2N Sec401 9 11 52 2.5 75 90 18 0 4.6 3.8 2N Sec401 10 11 26 2.5 60 90 18 0 4.1 3.8 2N Sec401 11 11 11 2.5 75 90 0 1.5 1.4 2.6 2N Sec401 12 11 11 2.5 75 90 0 1 1.4 2.6 2N Sec401 13 12 46 2.5 75 90 0 1 4.1 4.2 2N Sec401 14 12 13 2.5 70 70 0 2 2.2 3.4 2N Sec401 15 12 13 2.5 70 70 0 1.5 2.2 3.4 2N Sec401 16 12 13 2.5 70 70 0 1 2.2 3.4 2N Sec401 17 12 12 2.5 70 70 0 1.5 1.9 3.3 2N Sec401 18 12 12 2.5 70 70 0 2 1.9 3.3 2N Sec401 19 18 24 2.5 65 60 0 1 6.1 4.1 2N Sec401 20 12 18 2.5 60 50 0 0.5 3.3 3.8 2N Sec401 21 12 18 2.5 55 50 0 1 3.4 3.8 2N #5Deeps 1 9 12 2.4 120 30 0 0.5 0.6 1.6 2N #5Deeps 1 9 12 2.4 120 36 0 0.5 0.6 1.6 2N #5Deeps 1 9 12 2.4 120 42 0 0.5 0.6 1.6 2N #5Deeps 1 9 12 2.4 120 52 0 1 0.6 1.6 2N #5 Deeps 1 9 12 2.4 120 82 0 1 0.6 1.6 2N #5Deeps 1 9 12 2.4 120 96 0 1 0.6 1.6 2N #5Deeps 1 9 12 2.4 120 136 0 1 0.6 1.6 2N #5Deeps 2 9 12 2.4 120 30 0 0.5 0.6 1.6 2N #5Deeps 2 9 12 2.4 120 36 0 0.5 0.6 1.6 256 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH LOCATION No Pillar Dimensions (m) Depth (m) Age (days) Dist. to Gob (m) Parting Thick (m) Visual Rating F O S Ind.1 FOS2 A R M P S Length Width Height 2N #5Deeps 2 9 12 2.4 120 42 0 0.5 0.6 1.6 2N #5Deeps 2 9 12 2.4 120. 52 0 1 0.6 1.6 2N #5Deeps 2 9 12 2.4 120 82 0 1.5 0.6 1.6 2N #5Deeps 2 9 12 2.4 120 96 0 1.5 0.6 1.6 2N #5Deeps 2 9 12 2.4 120 136 0 1 0.6 1.6 2N #5Deeps 3 9 12 2.4 120 30 0 1.5 0.6 1.6 2N #5Deeps 3 9 12 2.4 120 36 0 1.5 0.6 1.6 2N #5Deeps 3 9 12 2.4 120 42 0 2 0.6 1.6 2N #5Deeps 3 9 12 2.4 120 52 0 2 0.6 1.6 2N #5Deeps 3 9 12 2.4 120 82 0 2.5 0.6 1.6 2N #5Deeps 3 9 12 2.4 120 96 0 2 0.6 1.6 2N #5Deeps 3 9 12 2.4 120 136 0 2.5 0.6 1.6 2N #5Deeps 4 9 12 2.4 120 20 0 2.5 0.6 1.6 2N #5Deeps 4 9 12 2.4 120 26 0 2.5 0.6 1.6 2N #5Deeps 4 9 12 2.4 120 32 0 2.5 0.6 1.6 2N #5Deeps 4 9 12 2.4 120 42 0 2.5 0.6 1.6 2N #5Deeps 4 9 12 2.4 120 72 0 3 0.6 1.6 2N #5Deeps 4 9 12 2.4 120 86 0 3 0.6 1.6 2N #5Deeps 4 9 12 2.4 120 126 0 3 0.6 1.6 2N #5Deeps 5 11 12 2.4 120 10 0 1.5 0.7 1.6 2N #5Deeps 5 11 12 2.4 120 16 0 1.5 . 0.7 1.6 2N #5Deeps 5 11 12 2.4 120 22 0 1.5 0.7 1.6 2N #5Deeps 5 11 12 2.4 120 32 0 2.5 0.7 1.6 2N #5Deeps 5 11 12 2.4 120 62 0 2.5 0.7 1.6 2N #5Deeps 5 11 12 2.4 120 76 0 3 0.7 1.6 2N #5Deeps 5 11 12 2.4 120 116 0 3 0.7 1.6 2N #5Deeps 6 12 12 2.4 125 1 0 1 0.8 1.7 2N #5Deeps 6 12 12 2.4 125 7 0 1 0.8 1.7 2N #5Deeps 6 12 12 2.4 125 11 0 1 0.8 1.7 2N #5Deeps 6 12 12 2.4 125 21 0 1.5 0.8 1.7 2N #5Deeps 6 12 12 2.4 125 53 0 2 0.8 1.7 2N #5Deeps 6 12 12 2.4 125 67 0 2 0.8 1.7 2N #5Deeps 6 12 12 2.4 125 107 0 2 0.8 1.7 2N #5Deeps 7 12 12 2.4 125 1 0 0 0.8 1.7 2N #5Deeps 7 12 12 2.4 125 11 0 1 0.8 1.7 2N #5Deeps 7 12 12 2.4 125 43 0 1.5 0.8 1.7 2N #5Deeps 7 12 12 2.4 125 57 0 1.5 0.8 1.7 2N #5Deeps 7 12 12 2.4 125 97 0 1.5 0.8 1.7 2N #5Deeps 8 12 12 2.4 125 1 0 0 0.8 1.7 2N #5Deeps 8 12 12 2.4 125 11 0 0.5 0.8 1.7 2N #5Deeps 8 12 12 2.4 125 43 0 1 0.8 1.7 2N #5Deeps 8 12 12 2.4 125 57 0 1 0.8 1.7 2N #5Deeps 8 12 12 2.4 125 97 0 2 0.8 1.7 2N #5 Deeps 10 30 30 2.4 120 40 15 0 6 5.6 2N #5Deeps 10 30 30 2.4 120 46 15 0 6 5.6 2N #5 Deeps 10 30 30 2.4 120 52 15 0 6 5.6 257 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH LOCATION No Pillar Dimensions (m) Depth (m) Age (days) Dist. to Gob (m) Parting Thick (m) Visual Rating FOS Ind.1 FOS2 A R M P S Length Width Height 2N #5Deeps 10 30 30 2.4 120 62 15 0 6 5.6 2N #5Deeps 10 30 30 2.4 120 92 15 0 6 5.6 2N #5Deeps 10 30 30 2.4 120 106 15 0 6 5.6 2N #5Deeps 10 30 30 2.4 120 146 15 0 6 5.6 2N #5Deeps 11 12 30 2.4 120 40 15 0 2.9 2.4 2N #5Deeps 11 12 30 2.4 120 46 15 0 2.9 2.4 2N #5Deeps 11 12 30 2.4 120 52 15 0 2.9 2.4 2N #5 Deeps 11 12 30 2.4 120 62 15 0 2.9 2.4 2N #5 Deeps 11 12 30 2.4 120 92 15 0.5 2.9 2.4 2N #5Deeps 11 12 30 2.4 120 106 15 1 2.9 2.4 2N #5Deeps 11 12 30 2.4 120 146 15 1 2.9 2.4 2N #5Deeps 12 30 30 2.4 125 30 15 0 6.2 5.8 2N #5Deeps 12 30 30 2.4 125 36 15 0 6.2 5.8 2N #5Deeps 12 30 30 2.4 125 42 15 0.5 6.2 5.8 2N #5Deeps 12 30 30 2.4 125 52 15 1 6.2 5.8 2N #5Deeps 12 30 30 2.4 125 82 15 1.5 6.2 5.8 2N #5Deeps 12 30 30 2.4 125 96 15 1.5 6.2 5.8 2N #5Deeps 12 30 30 2.4 125 136 15 1.5 6.2 5.8 2N #5Deeps 13 30 30 2.4 125 30 15 0 6.2 5.8 2N #5Deeps 13 30 30 2.4 125 36 15 0 6.2 5.8 2N #5Deeps 13 30 30 2.4 125 42 15 0 6.2 5.8 2N #5Deeps 13 30 30 2.4 125 52 15 0 6.2 5.8 2N #5 Deeps 13 30 30 2.4 125 82 15 0.5 6.2 5.8 2N #5Deeps 13 30 30 2.4 125 96 15 1 6.2 5.8 2N #5Deeps 13 30 30 2.4 125 136 15 1 6.2 5.8 2N #5Deeps 14 30 30 2.4 130 12 15 0 6.2 5.8 2N #5Deeps 14 30 30 2.4 130 42 15 1 6.2 5.8 2N #5Deeps 14 30 30 2.4 130 56 15 1.5 6.2 5.8 2N #5Deeps 14 30 30 2.4 130 96 15 1.5 6.2 5.8 2N #5Deeps 15 30 30 2.4 130 32 15 0 6.2 5.8 2N #5Deeps 15 30 30 2.4 130 46 15 1 6.2 5.8 2N #5Deeps 15 30 30 2.4 130 86 15 1.5 6.2 5.8 2N #5Deeps 9 12 12 2.4 130 1 5 0 1.2 1.6 2N #5Deeps 9 12 12 2.4 130 13 0 1 0.8 1.6 2N #5Deeps 9 12 12 2.4 130 47 0 1.5 0.8 1.6 2N #5Deeps 9 12 12 2.4 130 87 0 1 0.8 1.6 2N #5 Deeps 16 30 30 2.4 120 20 0 0 6 2N #5Deeps 16 30 30 2.4 120 34 0 1 6 2N #5Deeps 16 30 30 2.4 120 74 0 na 6 2N #5Deeps 17 12 30 2.4 125 15 0 0 2.2 2N #5Deeps 17 12 30 2.4 125 29 0 1 2.2 2N #5Deeps 17 12 30 2.4 125 69 0 na 2.2 2N #5Deeps 18 12 30 2.4 125 15 0 0 2.2 2N #5Deeps 18 12 30 2.4 125 29 0 2 2.2 2N #5Deeps 18 12 30 2.4 125 69 0 na 2.2 2N #5Deeps 19 12 30 2.4 125 5 0 0 2.2 258 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH LOCATION No Pillar Dimensions (m) Depth (m) Age (days) Dist. to Gob (m) Parting Thick (m) Visual Rating FOS Ind.' FOS2 ARMPS Length Width Height 2N #5 Deeps 19 12 30 2.4 125 19 0 1.5 2.2 2N #5Deeps 19 12 30 2.4 125 59 0 1.5 2.2 2N #5Deeps 20 12 30 2.4 130 1 15 0 2.6 2 2N #5Deeps 20 12 30 2.4 130 14 0 1 2 2N #5 Deeps 20 12 30 2.4 130 54 0 1.5 2 2N #5 Deeps 21 12 30 2.4 130 6 0 0 2 2N #5Deeps 21 12 30 2.4 130 46 0 0 2 2N #5 Deeps 22 22 30 2.4 135 4 15 0 2N #5Deeps 22 22 30 2.4 135 44 15 0 2N #5Deeps 23 6 12 2.4 135 20 0 2.5 2N #5Deeps 24 6 12 2.4 135 10 0 1.5 2N #5 Deeps 259 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH APPENDIX 5 SURFACE SUBSIDENCE 1) SURFACE SUBSIDENCE STATION LOCATION AND COORDINATES 2) SURFACE SUBSIDENCE LEVEL SURVEY DATA 3) SMMARY OF NUMERICAL MODELING OF SURFACE SUBSIDENCE 260 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH SURFACE SUBSIDENCE STATION LOCATION, INITIAL COORDINATES, AND FINAL ELEVATIONS Location Station # Northing Easting Initial Elev. Final Elev. 101 CS2 102741.770 99408.980 345.386 C S 7 103256.440 99431.380 326.484 101 1A 102741.097 99452.073 329.908 302 1B 102741.176 99467.181 329.423 302 1C 102741.714 99784.383 329.138 302 1D 102741.046 99507.250 329.416 302 1 102741.075 99410.956 335.024 302 2 102741.309 99537.269 328.749 302 3 102740.950 99564.979 328.233 302 4 102741.071 99590.356 327.346 302 5 102741.065 99604.830 326.980 302 6 102741.019 99618.708 326.240 302 7 102741.111 99629.943 325.850 325.643 302 8 102720.024 99633.024 326.655 326.230 302 9 102706.262 99635.130 326.750 326.926 302 10 102679.225 99639.146 326.793 326.256 302 11 102764.385 99624.071 325.436 325.369 303 12 102780.069 99619.893 325.297 325.238 302 13 102734.235 99665.662 324.611 324.429 302 14 102729.705 99689.133 323.793 323.645 302 15 102726.740 99717.391 323.102 322.988 302/#3 16 102721.829 99748.870 320.660 320.568 #3 mains 17 102710.400 99764.496 319.301 319.210 #3 mains 18 102700.331 99781.511 319.449 319.365 #3 mains 19 102697.044 99802.172 319.651 319.565 #1 mains 23 103100.453 99980.049 311.895 308.870 #1 mains 23B 309.005 103 24 103069.903 99979.929 310.942 310.925 103 25 103040.402 99979.057 309.365 309.143 103 26 103009.706 99979.697 309.285 308.811 103 27 102980.614 99979.582 311.103 310.197 103 28 102948.857 99979.476 311.259 310.860 103 29 102945.919 99962.161 315.725 315.645 103 30 102944.100 99953.122 317.145 316.930 103 31 102939.541 99920.483 316.580 316.612 103 32 102951.044 99999.600 309.750 308.989 103 33 102953.284 100019.449 308.116 307.595 103 34 102954.976 100038.871 306.715 306.492 103 35 102918.237 99979.386 309.082 308.135 103 36 102888.969 99979.419 310.694 310.241 103 36A 313.049 312.800 103 36B 312.431 312.267 103/206 37 102858.110 99978.815 312.645 312.512 103/206 38 102828.937 99977.997 314.200 314.186 206 39 102798.075 99978.165 314.958 314.906 261 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Location Station # Northing Easting Initial Elev. Final Elev. 206 40 102770.174 99976.153 314.239 313.431 206 41 102737.017 99979.086 313.532 313.051 206 42 102736.528 99959.289 315.897 314.898 206 43 102736.900 99939.681 317.559 316.302 206 44 102737.358 99919.725 318.547 317.640 206 45 102736.728 99998.116 312.083 312.123 206 46 102738.620 100019.034 310.354 310.493 206 47 102739.721 100039.498 308.502 308.693 206 48 102710.519 99977.404 312.767 312.605 206 49 102679.802 99977.430 313.620 313.311 206 50 102650.301 99976.547 313.772 313.506 206 51 102620.279 99977.375 314.307 314.071 206 52 102589.527 99977.128 313.873 313.650 #2 mains 53 102557.573 99977.986 311.082 310.704 #2 mains 54 102535.734 99976.412 304.538 303.988 #2 mains 55 102504.806 99979.062 290.264 289.565 #2 mains 56 102470.363 99984.194 286.697 286.075 208 57 102442.735 99985.937 286.635 286.054 208 58 102429.907 99975.732 286.788 286.112 208 59 102405.174 99975.661 280.146 279.102 208 60 102403.512 99956.177 285.844 285.260 208 61 102401.759 99941.510 288.063 287.839 205/208 62 102399.524 99926.400 289.734 289.510 208 63 102399.178 99994.329 272.909 272.253 208 64 102411.027 100007.680 272.430 271.774 208 65 102420.562 100018.807 271.209 270.553 208 66 102371.018 99976.189 276.149 275.540 101 70 102814.719 99428.740 328.669 328.643 101 71 102814.023 99494.714 328.645 328.685 101/303 72 102818.890 99520.933 327.703 327.682 303 73 102819.284 99555.751 326.109 326.106 303 74 102817.168 99530.891 324.908 325.024 303 75 102815.048 99620.118 323.830 323.525 303 76 102814.682 99652.082 323.514 323.286 303 77 102813.622 99686.574 321.396 321.141 303 78 102814.171 99712.306 320.243 320.086 303 79 102795.908 99685.361 322.376 322.661 303 80 102836.394 99682.093 321.519 321.804 101 85 103070.390 99584.492 336.441 336.957 101 86 103068.498 99624.308 335.566 335.854 102 87 103067.480 99730.713 329.814 329.841 101 88 102969.466 99519.214 339.675 339.078 101 89 102970.111 99540.569 339.657 338.979 101 90 102970.411 99561.246 338.458 338.378 101 91 102969.558 99581.163 338.047 337.747 101 92 102968.961 99691.118 337.674 337.424 101 93 102969.093 99621.054 336.822 336.668 101/102 94 102969.182 99640.845 335.286 335.181 101/102 96 102969.438 99660.987 334.648 334.631 102 97 102968.887 99682.064 334.297 334.161 102 98 102968.822 99703.503 333.124 332.661 262 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Location Station # Northing Easting Initial Elev. Final Elev. 102 99 102969.270 99724.133 331.558 331.423 101 102 102892.776 99519.630 340.208 339.325 101 102 A 102893.723 99537.005 339.026 338.427 101 102B 102892.948 99556.356 338.889 338.351 101 102C 102893.360 99575.594 338.109 337.580 101 103 102893.344 99591.683 337.800 337.216 102 104 332.283 331.979 101/201 105 102667.733 99398.120 343.400 343.877 101 106 102690.294 99400.234 344.213 344.673 101 107 102717.212 99403.575 345.464 344.887 101/201 108 102741.877 99428.426 343.551 343.376 101 109 102742.260 99389.951 343.183 342.625 101 110 102742.892 99369.192 342.805 343.117 101 111 102743.293 99349.909 342.804 343.290 101 111A 102743.666 99339.599 344.322 344.829 101 112 102765.250 99419.230 344.282 343.872 101 113 102792.112 99421.623 343.608 343.234 101 114 102815.183 99425.800 343.025 342.430 101 115 102821.620 99446.1 343.010 342.26 101 116 102809.717 99405.308 341.556 340.998 101 117 102816.020 99386.555 341.760 341.538 101 118 102835.542 99430.886 343.364 342.701 101 119 102857.449 99435.859 343.928 343.384 101 120 102876.543 99438.196 343.825 343.331 101 121 102893.282 99444.883 343.329 342.850 101 122 102909.733 99453.975 342.443 342.096 101 123 102893.039 99469.174 340.178 339.59 101 124 102894.826 99486.005 340.033 339.329 101 125 102893.937 99502.555 340.022 339.278 101 126 102893.175 99421.298 340.914 340.971 101 127 102893.277 99406.610 341.379 341.575 101 128 102924.960 99469.490 341.114 340.129 101 129 102940.006 99488.933 339.640 339.153 101 130 102953.925 99502.663 338.690 338.192 101 131 102969.470 99498.920 338.636 338.491 101 132 102970.670 99479.430 337.781 337.962 101 135 102971.960 99453.080 336.765 337.013 101 136 102972.610 99434.150 336.845 337.100 101 137 103012.210 99460.050 335.765 335.98 101 138 103001.380 99476.690 337.013 337.227 101 139 102989.639 99537.707 339.512 338.810 101 140 103006.810 99553.810 338.703 338.088 101 141 103033.410 99565.540 338.577 337.896 101 142 103055.970 99575.280 337.425 337.445 101 143 102813.893 99441.158 342.407 341.751 101 144 102816.852 99521.211 341.210 341.236 #1 mains 146 103126.600 99590.100 332.722 332.591 pillar 148 103396.100 99762.840 311.036 311.185 101 149 103124.860 99638.920 333.005 332.795 101 150 103098.800 99631.800 334.259 333.898 101 151 103078.650 99625.120 335.415 334.735 263 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Location Station # Northing Easting Initial Elev. Final Elev. 101 152 103110.970 99634.720 333.540 333.317 101 153 103103.290 99633.450 334.007 333.761 101 154 103091.720 99629.080 334.888 334.450 101 155 103086.380 99627.630 335.109 334.527 101 156 103072.620 99623.700 335.319 334.594 101 157 103137.830 99636.730 333.209 333.026 101 158 103144.940 99635.330 332.889 332.702 #1 mains 159 103167.650 99616.770 329.785 329.635 #1 mains 160 103199.920 99630.010 327.535 327.403 #1 mains 161 103227.430 99652.340 324.922 324.760 101A 162 103276.860 99672.190 320.634 320.478 101A 163 103323.940 99689.590 316.425 316.241 101A 164 103363.070 99720.930 312.778 312.567 101A 165 103347.240 99739.790 312.878 312.588 pillar 166 103443.350 99737.890 309.188 308.922 #2 mains 200 102639.105 99395.771 343.575 343.503 #2 mains 201 102513.355 99397.415 343.670 343.416 #2 mains 202 102432.295 99396.201 345.255 345.031 #2 mains 203 102467.760 99393.511 345.114 344.814 #2 mains 204 102587.519 99396.339 342.223 342.099 #2 mains 205 102513.226 99412.557 343.375 313.126 #2 mains 206 102513.751 99431.187 345.221 344.984 #2 mains 207 102513.578 99448.051 346.706 346.463 #2 mains 208 102514.131 99470.552 347.237 346.972 #2 mains 209 102514.508 99491.215 347.339 347.048 #2 mains 210 102514.515 99510.514 347.198 346.916 #2 mains 211 102514.570 99531.607 346.729 346.428 #2 mains 212 102514.221 99556.272 345.253 344.914 #2 mains 213 102515.061 99576.195 345.413 345.031 #2 mains 214 102515.589 99600.296 344.505 344.088 #2 mains 215 102516.197 99624.519 343.642 343.148 #2 mains 216 102517.131 99652.961 341.992 341.436 #2 mains 217 102517.576 99680.497 338.242 337.651 #2 mains 218 102518.099 99716.458 334.716 334.049 #2 mains 219 102508.227 99738.693 333.773 333.042 #2 mains 220 102518.689 99765.838 333.235 332.426 #2 mains 221 102519.822 99784.640 332.453 331.532 #2 mains 222 102437.642 99421.102 347.762 347.538 #2 mains 223 102437.899 99449.503 347.613 347.391 #2 mains 224 102437.913 99475.640 347.757 347.503 #2 mains 225 102437.741 99507.348 347.569 347.293 #2 mains 226 102436.097 99541.742 345.937 345.599 #2 mains 227 102436.723 99565.773 343.869 343.489 #2 mains 228 102438.264 99596.927 342.046 341.613 #2 mains 229 102438.844 99622.240 339.289 338.813 #2 mains 230 102439.208 99656.139 336.947 336.438 #2 mains 231 102439.456 99686.880 334.609 334.050 264 z o H < > - J W Q P O H < Q pf W ca P o H H r/5 < Q P -> -U z w Q r/3 ca p w u p CZ5 30/7/96 | 342.922 | 326.848 | 343.262 | 342.822 | 342.559 | [ 340.787 | 342.065 | [ 341.127 | 340.140 | 339.730 | [ 338.993 | 326.484 | 26/9/95 | | 6/9/95 | [326.4841 30/8/95 | 344.910 | 343.248 | [342.816 | [342.576 | [342.8381 [342.1251 [341.213] 9/8/95 | 344.910| 343.0501 342.6081 [342.3731 [343.6281 [341.9281 340.9261 [339.9451 126/04/951 308.870 | 311.895 | [310.925 | [ 309.143 | [308.811 | | 310.197 [ | 310.860 j | 315.645) | 317.343 | | 317.160] | 309.449 | 308.059 | | 307.053 | 308.673 | 310.655 12/04/951 311.895 | 308.870 | 310.949 | 309.293 | | 309.285 | [310.227 | [310.494 | [315.276 | | 316.930 | | 316.612 | [ 308.989 [ | 307.595 j | 306.492 j | 308.135 ] | 310.241 | |17/03/951 | 311.895 | | 310.945 | | 309.292 | | 309.285 | | 311.015 | | 311.285 | | 315.754 | | 317.200 | | 316.625 | | 309.817 | | 308.140 | | 306.720 | | 308.983 | | 310.523 j 29/08/94 | 22/07/94 | 2/06/94 | 26/04/94 | | 310.942 | j 309.365 | | 309.431 | | 311.103 j [ 311.259 ] | 315.725 [317.145] | 316.580 | 309.750 | 308.116 | | 306.715 | 309.082 | 310.694 23/03/94 | [16/01/94 | | 310.941 | | 309.231 | | 309.146 | | 310.870 | | 311.176 | | 315.644 ) | 317.086 | | 316.475 | 309.684 | | 308.034 | | 306.609 | 308.971 | 310.446 | 18/8/93 | [ 329.908 | 1329.423 | [329.1621 1329.420 | 1335.024 | 1328.728 | 1328.187 | 1327.244 | 1326.848 | 1326.116 | 1325.760 | 1326.624 | 1326.351 | 1326.374 | [325.4881 [325.3701 1324.571 | 1323.7951 1323.1451 1320.741 | 1319.390 | 1319.558 j 1319.770 | 1311.288 | 1309.129 18/6/93 | [329.8661 [329.701 | 1329.115 | 329.369 | 335.024 | 335.669 | 328.145 | 327.171 | 326.769 | 326.008 | 325.643 | 326.230 | 326.926 | 326.2561 325.369 | [325.2381 [324.4291 1323.6451 [322.9881 1320.568 [ |319.210[ 1319.365 j 1319.5651 24/03/931 329.855 | 329.393 | 329.138 | 329.416 | 335.024 | 328.749 | 328.233 | 327.346 | 326.980 | 326.240 | 325.850 | 326.655 | 326.750 | 326.793 | 325.436 | 325.297 | 324.611 | 323.793 | 323.102 | 320.660 | 319.301 | 319.449 | [319.651 | [ 311.895 | | 310.941 | | 309.289 | | 309.277 j | 311.018 j | 311.288 | | 315.756 | 317.198 | 316.587 | 309.796 | 308.146 | 306.721 | 308.987 | 310.524 21/8/92 | | 8/07/92 | 25/04/92 | 20/01/92 | | Station | 1 CS2 I I ISO I < CQ O Q CM CO in CD CO o> o CN CO in CD CO 01 CO c\j 1 23B [ •s-CM tn CM CO CM CM CO CM cn CM o CO CO CM CO CO CO S in CO CO CO O l 30/7/96 | | 341.391 | [ 341.679 | 26/9/95 | 6/9/95 | [336.957 30/8/95 | 1336.630 9/8/95 | 341.261 [341.651 : 26/04/951 313.267 | 312.790 | [313.199] 313.311 | 313.506 | 314.071 | 313.650 | 310.704 | 303.988 | 289.565 | [ 286.075 | 286.054 | [286.112 | [279.102 | [ 285.260 | | 271.826 | | 275.251 | 12/04/951 312.800 | [312.267 | [312.512 | [314.177 | 314.877 | [313.362 | 17/03/951 313.049 | 312.431 | 312.547 | 314.162] 314.868 | 29/08/94 | 22/07/94 | 2/06/94 | 26/04/94 | 312.645 | 314.238] 315.005 | 313.623 | 313.123 | 314.898 | 316.302 | 317.640 | 312.123 | 310.493 | 308.693 | 312.605 | 313.628 | 313.802 | 314.311 | 313.976 | 311.211 | 304.704 | 290.419 | [ 286.874 | | 286.753 | [ 286.760 | | 279.633 | 23/03/94 | | 328.643 | | 328.685 j | 327.682 ] | 326.106 j [ 325.024 [ | 323.525 | 323.286 | 321.141 | 320.086 16/01/94 | 312.431 | 314.001 | [314.621 | 313.186 | 312.836 | [314.621 | 316.031 | 317.374 | 311.833 | 310.198 | 308.343 | 312.323 | 313.353 | 313.668 | [314.158 | [313.768 | [311.068 | [ 304.576 | [ 290.335 | | 286.807 | | 286.705 | | 286.657 | | 279.490 | | 285.620 | | 287.839 | | 289.510 | | 272.253 | | 271.774 | | 270.553 | | 275.540 j 18/8/93 | 317.590 | 318.820 | 316.490 | 316.054 | [314.9151 316.376 | 1317.7281 [313.000 | [310.314 | j 308.454 | 1312.490 | 1313.4431 1313.678 | 1314.236 | 1313.832 | 1310.700 | 1328.669 [328.6501 1327.703 1326.097 1324.898 1323.837 1323.514 | 1321.396 1320.243 1322.661 1321.804 | 336.441 18/6/93 | 1328.669 1328.645 1327.703 1326.109 1324.908 1323.830 1323.255 1321.117 1319.945 1322.376 1321.519 1336.441 24/03/931 312.552 | 314.200 | 314.958 | 314.239 | 313.532 | [315.897 | [317.559 | | 318.547 | | 312.083 | | 310.354 | | 308.502 | | 312.767 | | 313.620 j | 313.772 ) | 314.307 | 313.873 | 311.082 | 304.538 | 290.264 | 286.697 | 286.635 | 286.788 | 280.146 | 285.844 | 288.063 | 289.734 | 272.909 | 272.430 | 271.209 | 276.149 21/8/92 | 8/07/92 | 25/04/92 | 20/01/92 | | Station | 1 36A I 1 36B I CO CO CO Oi CO o 5 CM CO * r in CO - J CO O) OS CM W CO IT) 3 in m CO m r~ m CO m o> m o CO CO CM CO CO CO S in CO CO CO o CM 1^  CO I-- in CO CO o> h- o CO m CO MO MO CN 30/7/96 | 339.078 | 339.325 | 338.427 | 338.351 | 337.580 | 337.216 | | 344.887 | j 343.376 | | 343.117 | | 343.290 | | 344.829 | | 343.872 | | 343.234 | | 341.538 | | 342.701 | | 343.384 | | 343.331 | | 342.850 | | 342.096 | [ 339.59 | | 339.329 | | 338.022 | | 340.129 | 26/9/95 | 6/9/95 | 339.1831 338.9791 337.7471 337.4241 336.6681 335.181 | [342.908 1342.156 1341.138 1341.728 1340.790 30/8/95 | 335.854 | 339.183 | 339.021 | 338.192 | [337.908] 1337.074 | 1335.579 | 1339.3451 1338.491 | 1338.460 | 1337.722 | 1337.4251 1343.877 | 1344.673 | 1344.917 | 1343.3851 1342.6251 1343.105 | 1343.290 | 1344.823 | 1343.6831 1343.190 | [342.430] 1340.9981 [341.455] 1342.6251 1343.285] 1343.238 1342.755 1342.002 1339.519] 1339.282 1339.278 1340.971 1341.575 1340.651 9/8/95 | 1339.8491 1339.0261 1338.9451 1338.1971 1337.7971 1334.671 j 1334.458 j 1334.7411 1343.1951 1342.5831 1342.899 1343.077 1344.6011 1343.6601 1343.010 1342.257 1340.820 1341.277 1342.471 1343.156 1343.139 1342.729 [342.080 1339.625 1339.546 1339.659 1340.811 1341.392 1340.934 26/04/951 12/04/951 17/03/951 29/08/94 | 22/07/94 | 2/06/94 | 26/04/94 | 23/03/94 | 329.841 | 339.579 | 338.378 | 337.999 | 337.733 | 336.849 | 335.381 | 334.631 | 334.161 | 332.661 | 331.423 | 340.208 | | 337.800 | | 331.979 | 16/01/94 | 18/8/93 | 335.566 | 329.814 | 339.675 | 339.650 | 338.458 | 338.047 | 337.674 | 336.822 | 335.2861 334.6481 334.297 | 333.124 | 331.558 | 18/6/93 | 335.566 | 329.814 | 339.675 | 339.657 | 338.458 | 338.047 | 337.674 | 336.822 | 335.2861 334.6481 334.297 | 333.124 | 331.423 | 340.028 | j 337.7081 1332.2831 24/03/931 21/8/92 | 8/07/92 | 25/04/92 | 20/01/921 | Station | CD CO oo CO 00 01 oo o 01 CM CD CO O) CO Ol CO Ol O) Ol CM O 1 102A I 102B 102C CO o o m Cl CD O O CO o 01 o Cl < CN CO IO CO CO Ol o CM CN CM CM CO CM CM ID CM CD CN CM CO CM CN 30/7/96 | 338.192 j 338.810 338.088 | 337.896 | 339.812 | 332.591 | 311.185 | 332.795 | 333.898 | 334.735 | 333.317 I 333.761 | 334.450 | 334.527 | 334.594 [ 333.026 j j 332.702 | | 327.403 | 320.478 1 | 316.241 | | 312.567 | | 312.588 | | 308.922 | 332.818 | 334.015 | 334.801 | 333.345 | 333.785 | 334.480 | 334.575 | 334.674 333.042 | [332.7131 332.617 | 311.200 | 332.818 | 329.635 | 1327.385 | [324.760 | 1320.479 | 1316.231 | 1312.570) | 312.596 26/9/96 | 332.722 | 333.005 | 334.203 | 335.229 | 333.523 | 333.974 | 334.816 | 334.991 | 335.114 | 333.208] 332.865 | 6/9/95 | 339.2701 338.2721 338.5481 338.122 | 337.1741 337.2631 337.3681 339.0331 338.7761 338.9151 337.7411 30/8/95 | 339.1531 338.1881 338.491 | 337.9621 337.0131 337.100 | 335.979 | 337.2271 339.197 | 338.7581 338.742 | [337.4451 341.751 | 332.7221 9/8/95 | 339.6351 338.6901 j 339.5121 1341.5861 [341.2361 26/04/951 | 343.503 | 343.416 | 345.031 | 344.814 | 342.099 | 313.126 I 344.984 | 346.463 | 346.972 | 347.048 12/04/951 17/03/951 29/08/94 | | 343.433 | 344.999 | 344.873 | 343.145 | 344.984 | 346.445 | 346.965 | 347.061 22/07/94 | | 343.575 | 343.556 | 345.073 | 344.873 | 342.088 | 343.222 | 345.021 | 346.449 | 347.007 | 347.143 2/06/94 | 1343.575 1343.670 1345.255 1345.114 1342.223 1343.375 1345.221 1346.706 1347.237 1347.339 26/04/94 | 23/03/94 | 16/01/94 | 18/8/93 | 18/6/93 | 24/03/931 21/8/92 | 8/07/92 | 25/04/92] 20/01/921 1 Station 1 Oi CM o CO CO CM CO LO CO CO CO CO CO CO O) CO o CM CO •q-CD CO O) ^-o LO CM m CO CO 3 LO LO CD LO r~ LO CO w 0) m o to CO CM CD CO CD S LO CD CO CO o o CM o CM CM O CM CO O CM •q-o CM m o CM CD O CM o CM CO O CM o> o CM 0 0 M 3 30/7/96 | 26/9/95 | 6/9/95 | 30/8/95 | 9/8/95 | 26/04/951 346.916 | ! 346.428 | 344.914 | 345.031 | [ 344.088 | 343.148 | 341.436 | [ 337.651 | 334.049 | 333.042 | 332.426 | [331.532 | [ 347.538 | [ 347.391 | | 347.503 | | 347.293 | | 345.599 | | 343.489 | | 341.613 | | 338.813 | | 336.438 | | 334.050 | 12/04/951 17/03/95 | 29/08/94 | 346.935 | 346.466 | 344.984 | [345.117 | 344.201 | [ 343.299 | [341.634 | 337.880 | [334.343 | [ 333.384 | [ 332.838 | | 332.004 | | 347.486 | | 347.333 | | 347.476 | | 347.304 | | 345.655 | | 343.576 | | 341.738 | | 338.958 | | 336.603 j | 334.248 ) 22/07/94 | 347.042 | 346.515 | 345.004 | 345.187 | 344.130 | 343.166 | 341.536 | 337.846 | [ 334.324 | [333.419 | 332.938 | 992.169 | 347.515 | 347.351 | 347.546 | j 347.273 | | 345.658 j | 343.631 j | 341.411 | | 338.694 | | 336.430 | 334.100 2/06/94 | 347.1981 346.7291 345.253 | 345.413 | 344.505 | 343.642 | (341.992 | 1338.242 | 1334.716 | [333.7731 1333.2351 1332.4531 1347.762 | 1347.613 | 1347.757 | 1347.569 | 1345.937 | 1343.869 | 1342.046 | 1339.289 | 1336.947 | 1334.609 | 26/04/941 23/03/94 | 16/01/941 18/8/93 | 18/6/93 | 24/03/93 | 21/8/92 | 8/07/92 | 25/04/92 | 20/01/92 | 1 Station! o eg CM CM CM CO CM •<r CM m CN CD CM CM CO CM O) CM o CM CM CN CN CN CN CN CO CM CN <3-CM CN in CM CM CO CM CM CM CM CO CM CM o> CM CM o CO CN CO CM NO CN GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH SURFACE SUBSIDENCE: SUMMARY OF NUMERICAL MODELING RESULTS Run Depth (m) Panel Width (m) Stress Ratio H : V Model Property Set Angle 1 Draw (deg.) Angle" Fracture (deg.) Angle1" Critical (deg.) Max Sub H Comments qc 80 28 1:1 initial Run to evaluate affect of moduli and transient stress effects. Moduli have no effect on failure. Transient stresses induced by excavation do not significantly affect results. qd 80 20 1:1 initial Run to calibrate material model and properties. Significant failure of coal rib: need to increase strength of coal, use , "caved" rock instead of null to provide confinement. qe 80 8 1:1 initial Run to calibrate material model and properties. Significant failure of coal rib: need to increase strength of coal, use "caved" rock instead of null to provide confinement qf 80 20 1:0.5 initial Run to calibrate material model and properties, and stress. Minor failure of coal rib: need to increase strength of coal, use "caved" rock instead of null qhl 80 24 1:0.5 1 Mohr-Coulom b Run to calibrate material model and properties. Very odd results, prompted change to ubiquitous joint model. qn 80 24 1:1 1 Run to calibrate material model and properties. Extensive rib failure. Excessive deformations resulting in "bad geometry" need to increase caved rock moduli. qn-a 80 18 i ; i 2 Run to calibrate material model and properties. qn-c 80 36 1:1 1 + Run to calibrate material model and properties. Odd results; ladder type failure-many orphaned non-failed zones. qn-d 80 36 1:1 1+ 14.7 1.2 Run to calibrate material model and properties. Reduced properties of 271 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Run Depth (m) Panel Width (m) Stress Ratio H : V Model Property Set Angle' Draw (deg.) Angle" Fracture (deg.) Angle'" Critical (deg.) Max Sub (m) Comments failed rock results in less orphaned zones qslc -8 80 8 1:1 1 Run to calibrate material model and properties. Maximum deflection is 0.0125m. Significant failure in roof. qslc -20 80 20 1:1 1 Run to calibrate material model and properties. Maximum deflection is 2.27m. Significant failure in roof, mostly at shoulder. qs3c -8 80 8 1:1 3 Run to calibrate material model and properties. Maximum deflection is 0.005m. qs3c -20 80 20 1:1 3 Run to calibrate material model and properties. Maximum deflection is 0.0125m. Minor failure in roof, mostly at shoulder. qs3b -20 80 20 1:0.5 2 Run to calibrate material model and properties and stresses. Maximum deflection is 0.225m. Significant failure in roof, mostly at shoulder. qs3a -8 80 8 1:1 2 Run to calibrate material model and properties and stresses. Maximum deflection is 0.009m. Moderate failure in roof, mostly at shoulder. qs3a -20 80 20 1:1 2 Run to calibrate material model and properties and stresses. Maximum deflection is 0.15m. Significant failure in roof, mostly at shoulder. qr22 40 36 1:0.5 1 -5.0 -5.0 -5.0 0.42 qr23 40 72 1:0.5 1 -5.0 -5.0 -5.0 0.59 qr24 40 108 1:0.5 1 -5.0 -5.0 -5.0 0.65 qrl3 80 36 1:0.5 1 -1.4 -1.4 -1.4 0.49 qrl4 80 72 1:0.5 1 -3.6 -3.6 -3.6 1.30 qrl5 80 108 1:0.5 1 -3.6 -3.6 -3.6 1.75 qrl6 160 72 1:0.5 1 Not realistic results qrl9 320 40 1:0.5 1 Rib failed to 240m. Not realistic results qrl 80 36 1:1 1 12.7 12.7 0.46 Anomalous results. qsl 80 36 1:1 1 25.5 25.5 Run to evaluate anomalous result of qrl and evaluate 272 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Run Depth (m) Panel Width (m) Stress Ratio H : V Model Property Set Angle 1 Draw (deg.) Angle" Fracture (deg.) Angle"1 Critical (deg.) Max Sub (m) Comments effect of mirror lower boundary. Results are more anomalous qr2 80 72 1:1 1 0 0 0 1.02 qr3 80 108 1:1 1 -2.0 -2.0 -2.0 1.4 qt l -36 120 36 1:1 1 17.5 15 15.0 0.4 Anomalous results. qt l -72 120 72 1:1 1 7 6 6.0 0.72 qtl-108 120 108 1:1 1 -2 -3 -1 1.1 qr7 80 36 1:1.5 1 Siltstone above coal failed to 60m. Banding. qr8 80 72 1:1.5 1 Siltstone above coal failed to 73m. Banding. qr9 80 108 1:1.5 1 Siltstone above coal failed to 78m. Banding. qr4 80 36 1:2 Coal Rib failed to 4m Siltstone above coal failed to 90m. Banding. Non realistic. Highlights importance of stress vs strength parameters. qr4a a 80 36 1:2 1 1.15 Run to evaluate affect of Szz. There is only minor difference in subsidence results when Szz = Sxx and when Szz = Syy. This is expected, as Mohr-Coulomb failure does not consider sig2. qr22 a 40 36 1:0.5 2 -8.5 nob/t .01 many orphaned non failed zones qr22 aa 40 36 1:0.5 2 -8.5 nob/t .01 Run to evaluate affect of orphaned non failed zones. No significant differences in subsidence when orphaned zones changed to "caved" material. qr23 a 40 72 1:0.5 2 4.3 -9.9 0.33 Anomalous result. qs2-36 40 36 1:0.5 2 -14 -14 -14 0.01 Run to evaluate use of mirror on lower boundary for qr22a.The extent of Orphaned zones remained, however, the cave broke through to surface and the angle of draw decreased by 5.5 deg. qs2-72 40 72 1:0.5 2 -12 -12 -12 0.13 Run to evaluate use of rnirror on lower boundary 273 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Run Depth (m) Panel Width (m) Stress Ratio H : V Model Property Set Angle 1 Draw (deg.) Angle" Fracture (deg.) Angle1" Critical (deg.) Max Sub (m) Comments for qr23a. The "anomalous" failure zone and angle draw of qr23a disappeared. Similar angle of fracture results. qs2-108 40 108 1:0.5 2 -12 -11 -10.0 0.14 qrl3 a 80 36 1:0.5 2 -5.3 -5.3 -5.3 0.18 Many orphaned non failed zones qrl3 aa 80 36 1:0.5 2 1.4 -7.2 0.1 Run to evaluate distance to lower boundary; distance increased from 20 to 120m. 5.4deg. change in angle of draw with 10mm, 3 deg change with 20mm. qrl3 aaa 80 36 1:0.5 2 -7 -7 00 Run to evaluate use of mirror on lower boundary. 8.6 deg difference in draw. qrl4 a 80 72 1:0.5 2 -1.4 -1.4 -1.4 0.69 Many orphaned non failed zones qrl4 aa 80 72 1:0.5 2 0 0 0.6 Run to evaluate distance to lower boundary; distance increased from 20 to 120m. Change not considered significant. qrl4 aaa 80 72 1:0.5 2 -5.5 -5.5 -5.5 0.34 Run to evaluate use of mirror on lower boundary. 5.5 deg difference. qrl5 a 80 108 1:0.5 2 -2.9 -2.9 -2.9 0.75 Many orphaned non failed zones qrl5 aaa 80 108 1:0.5 2 -2.5 -3.5 0 0.87 Run to evaluate use of mirror on lower boundary. Very similar results. qrl6 a 160 36 1:0.5 2 29.9 1.4 1.4 0.16 Anomalous results. qrl6 a-1 160 72 1:0.5 2 35 -4 0 2.5 qrl6 a-2 160 108 1:0.5 2 33 -2 -1 3.4 qrl7 a 160 144 1:0.5 2 36.8 -3.6 -3.9 4 qrl8 a 160 216 1:0.5 2 33.0 -2.9 -7.8 4.9 qrl9 a 320 40 1:0.5 2 45 9.2 9.9 0.4 Anomalous results. qrl9 aa 320 40 1:0.5 2 Run to evaluate distance to lower boundary; distance increased from 80 to 480m. Significant difference -coal rib fails out to a distance of 400m! 274 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Run Depth (m) Panel Width (m) Stress Ratio H:V Model Property Set Angle 1 Draw (deg.) Angle" Fracture (deg.) Angle"1 Critical (deg.) Max Sub (m) Comments qr20 a 320 144 1:0.5 2 Rib failed to 360m. Not realistic results. qs3-36 80 36 1:1 2 5 5 5.7 0.08 Anomalous results. qs3-72 80 72 1:1 2 -2 -2 -0.5 0.48 qs3-108 80 108 1:1 2 -1 -0.5 -1 0.65 qt2-36 120 36 1:1 2 22 20 22 0.50 Used 0.02 as cut-off for angle of draw. Anomalous results. qt2-108 120 108 1:1 2 6 3 4.3 1.10 qv3-36 40 36 l ; l 3 -11 0.01 No breakthrough. No tensile strain greater than 0.0005. qv3-72 40 72 i ; i 3 -11 -26 -26 0.14 qv31 08 40 108 i ; i 3 -14 -12 -12 0.18 qv3-36A 40 36 i ; i 3 -14 0.01 Run to evaluate orphaned zones. No significant change to qv3-36. No breakthrough. No tensile strain greater than 0.0005. qv3-72A 40 72 i ; i 3 -5.8 -4.2 -4.2 0.18 0 Run to evaluate orphaned zones. Significant difference in all angles as compared to qv3-72. Similar subsidence. Results appear more likely. qv3-108 A 40 108 i ; i 3 5 -27 -27 .35 Run to evaluate orphaned zones. Significant difference in all angles as compared to qv3-108. qu3-36 80 36 1:1 3 -3 -3 -3.0 0.04 No tensile strain greater than 0.0005 qu3-72 80 72 1:1 3 12.2 -2.5 -1 0.39 qu31 08 80 108 1:1 3 37 -15 -12 0.54 qp8-36 95 36 1:1 2/3 11 11 .12 no tensile strain greater the 0.002 qp8-72 95 72 1:1 2/3 1.5 1 1.2 .62 qp81 08 95 108 1:1 2/3 -9.5 -9.5 -9.5 0.86 qt3-36 120 36 1:1 3 0 -1 0.05 No tensile strain greater than 0.0002. Anomalous results. qf3- 120 72 1:1 3 -8 -9 0.6 No tensile strain greater 275 GEOTECHNICAL STUDIES OF RETREAT PILLAR COAL MINING AT SHALLOW DEPTH Run Depth (m) Panel Width (m) Stress Ratio H : V Model Property Set Angle 1 Draw (deg.) Angle" Fracture (deg.) Angle1" Critical (deg.) Max Sub (m) Comments 72 than 0.0002. Angle of draw uses a 0.017m cut off. Anomalous results. qt3-108 120 108 1:1 3 -8 -8 -7.1 0.99 Anomalous results. qt3-36a 120 36 1:1 3 2.5 -1.5 -1.5 0.23 Run to evaluate orphaned zones. No significant change to qt3-36. qt3-72a 120 72 3 -7.5 -9 -9 0.56 Run to evaluate orphaned zones. No significant change to qt.3-72. Angle draw uses a 0.02m cut off. qt31 08a 120 108 1:1 3 qv4-36 40 36 4 .000 1 No subsidence greater than 0.0001. No break through to surface. No tensile strain greater than 0.0001 qv4-72 40 72 1:1 4 0 0.03 No break through to surface. No tensile strain greater than 0.0002 qv41 08 40 108 1:1 4 10 0.06 No break through to surface. No tensile strain greater than 0.0005 qu4-36 80 36 4 0.01 no subsidence greater than 0.005. No break through to surface. No tensile strain greater than 0.0001 qu4-72 80 72 1:1 4 9.2 .034 No break through to surface. No tensile strain greater than 0.0005 qu41 08 80 106 1:1 4 14 0.05 No break through to surface. No tensile strain greater than 0.0005 qt4-36 120 36 4 0.01 No break through at surface. No subsidence greater than 0.005m. No tensile strain greater than 0.0001 qt4-72 120 72 1:1 4 14 0.02 No break through at surface. No tensile strain greater than 0.0001 qt4-108 120 108 1:1 4 22.5 0.05 No break through at surface. No tensile strain greater than 0.0001 I The angle of draw was determined as the location where the vertical subsidence was less than 0.01m. I I The angle of fracture was the determined as the location where failure reached the ground surface. I I I The angle of critical deformation was determined as the greater of the following: 1. the location where the horizontal ground strain was less than 0.002 m/m or, 2. the angle of fracture. 276 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0081103/manifest

Comment

Related Items