UNITED STATES DEPARTMENT OF LABOR MINE SAFETY AND HEALTH ADMINISTRATION COAL MINE SAFETY AND HEALTH REPORT OF INVESTIGATION Fatal Underground Coal Mine Explosion January 2, 2006 Sago Mine, Wolf Run Mining Company Tallmansville, Upshur County, West Virginia ID No. 46-08791 By Richard A. Gates District Manager, Coal Mine Safety and Health, District 11, Birmingham, AL Robert L. Phillips Coal Mine Safety and Health Specialist, Coal Mine Safety and Health, Arlington, VA John E. Urosek Chief, Ventilation Division, Technical Support, Pittsburgh, PA Clete R. Stephan General Engineer, Ventilation Division, Technical Support, Pittsburgh, PA Richard T. Stoltz Supervisor, Ventilation Division, Technical Support, Pittsburgh, PA Dennis J. Swentosky Supervisor, Coal Mine Safety and Health, District 2, New Stanton, PA Gary W. Harris Supervisor, Coal Mine Safety and Health Special Investigator, District 7, Barbourville, KY Joseph R. O’Donnell Jr. Supervisor, Coal Mine Safety and Health, District 11, Bessemer, AL Russell A. Dresch Electrical Engineer, Coal Mine Safety and Health, District 5, Norton, VA Originating Office Mine Safety and Health Administration Office of the Administrator Coal Mine Safety and Health 1100 Wilson Boulevard Arlington, Virginia, 22209 Kevin G. Stricklin, Acting Administrator May 9, 2007 TABLE OF CONTENTS OVERVIEW...................................................................................................................... 1 GENERAL INFORMATION ........................................................................................ 4 EVENTS LEADING TO THE ACCIDENT................................................................ 9 DESCRIPTION OF THE ACCIDENT ...................................................................... 13 The 2nd Left Parallel Miners .............................................................................. 29 NOTIFICATION AND SAMPLING......................................................................... 34 RESCUE AND RECOVERY OPERATIONS ........................................................... 44 Mine Rescue Protocol........................................................................................... 44 Mine Gases............................................................................................................. 45 Mine Exploration .................................................................................................. 50 2nd Left Parallel Exploration .............................................................................. 51 Rescue Borehole Chronology ............................................................................. 56 MINE RECOVERY........................................................................................................ 60 INVESTIGATION OF THE ACCIDENT................................................................. 62 Mine Emergency Evacuation and Firefighting Program of Instruction..... 63 Notification ........................................................................................................ 63 Evacuation of the Mine .................................................................................... 64 SCSRs .................................................................................................................. 65 Belt Fire Detection System .............................................................................. 65 Barricading Instructions .................................................................................. 68 Barricading ............................................................................................................. 69 Carbon Monoxide Poisoning.............................................................................. 71 Self-Contained Self-Rescuers............................................................................. 75 Daily Inspection ................................................................................................ 77 90 Day Inspection.............................................................................................. 78 Training............................................................................................................... 79 Recordkeeping................................................................................................... 79 Evaluation........................................................................................................... 79 Miners Working Outby 1st Left ..................................................................... 80 Miners on the 1st Left Mantrip ...................................................................... 82 Miner Working Near the Mouth of 2nd Left Parallel................................ 87 Miners on 2nd Left Parallel............................................................................. 88 Miners Attempting Rescue Effort.................................................................. 94 Other SCSRs Recovered and Evaluated ....................................................... 96 Mine Ventilation Plan........................................................................................ 100 Mine Ventilation................................................................................................. 102 Development Sections ................................................................................... 103 Ventilation of Seals ........................................................................................ 103 Methane Ignitions........................................................................................... 103 Methane Liberation ........................................................................................ 103 Methane in the Sealed Area.......................................................................... 104 Ventilation Survey and Computer Simulations ....................................... 105 Barometric Pressure ........................................................................................ 107 ii Roof Control Plan ............................................................................................... 108 Geology ................................................................................................................. 110 Evaluation of Two Linear Features near Survey Station 4010................ 110 Cleanup Program and Rock Dusting .............................................................. 111 Mine Dust Survey ............................................................................................... 111 2 North Mains - Survey No. 1(b) .................................................................. 113 1st Left - Survey No. 2 .................................................................................... 113 2nd Left Parallel - Survey No. 3.................................................................... 113 2nd Left Mains - Survey No. 4 .......................................................................... 114 MSHA Mine Dust Sampling Prior to Accident ........................................ 114 Examinations........................................................................................................ 114 Training................................................................................................................. 116 Communications ................................................................................................. 117 Equipment ........................................................................................................ 117 Equipment Status............................................................................................ 119 Mine Rescue Communications......................................................................... 120 Underground Mine Rescue Communications........................................... 120 Seismic Location System ................................................................................... 121 Introduction ..................................................................................................... 121 System Deployment ....................................................................................... 122 Mine Emergency Evacuation and Firefighting Program of Instruction123 2nd Left Parallel Crew.................................................................................... 123 System Response............................................................................................. 123 Seals ....................................................................................................................... 124 Manufacturing and Testing of Omega Block ............................................ 124 Seal History and Construction ..................................................................... 128 Seal Testing ...................................................................................................... 139 Electrical Power and Equipment...................................................................... 146 Electrical Power System................................................................................. 146 Grounding Systems ........................................................................................ 147 Potential Ignition Sources ................................................................................. 150 Other Sources................................................................................................... 150 Roof Falls ........................................................................................................... 151 Lightning Overview ....................................................................................... 153 Lightning as an Ignition Source................................................................... 158 Origin .................................................................................................................... 172 Flame ..................................................................................................................... 176 Force....................................................................................................................... 180 Deflagration ..................................................................................................... 180 Pressure Piling................................................................................................. 181 Detonation........................................................................................................ 182 The Sago Mine Explosion.............................................................................. 184 ROOT CAUSE ANALYSIS ....................................................................................... 187 CONCLUSION............................................................................................................ 188 ENFORCEMENT ACTIONS .................................................................................... 189 iii TABLES Table 1 - Accident Incidence Rates.............................................................................. 7 Table 2 - Enforcement Actions in 2005........................................................................ 7 Table 3 - Air Quality Measurements......................................................................... 38 Table 4 - Air Quality Measurement by BCMR........................................................ 41 Table 5 - Summary of Toxic Effects Following Acute Exposure to Carbon Monoxide................................................................................................................ 73 Table 6 - Summary of Toxic Effects Following Acute Exposure to Carbon Monoxide................................................................................................................ 74 Table 7 – Summary of Information on the SCSRs at the Sago Mine.................. 99 Table 8 - Air Sample Results .................................................................................... 104 Table 9 - Dimensions of the 2 North Mains Seals ................................................ 132 Table 10 - Mine Explosions in Sealed Areas with Lightning as a Possible Ignition Source .................................................................................................... 156 Table 11 - Results of Explosion Test #501 and #502 at Lake Lynn ................... 183 FIGURES Figure 1 - Sketch of Sago Mine..................................................................................... 5 Figure 2 - Location of 2 North Main Seals.................................................................. 9 Figure 3 - CO Measurements at the No. 1 Drift Opening ..................................... 43 Figure 4 - CO Measurements from a Mine in Virginia.......................................... 49 Figure 5 - Borehole No. 1 Carbon Monoxide Results............................................. 59 Figure 6 - Damaged Stopping at 59 Crosscut, No. 4 Belt ....................................... 60 Figure 7 - Damaged Overcast at 58 Crosscut, No. 4 Belt........................................ 61 Figure 8 -Drawing of Barricade .................................................................................. 70 Figure 9 - Location of Miners and Their Carboxyhemoglobin Levels................ 75 Figure 10 - CSE SR-100 ................................................................................................. 76 Figure 11 - Components of the SR-100 SCSR .......................................................... 76 Figure 12 - Fan Chart................................................................................................... 102 Figure 13 - Barometric Pressure for Buckhannon, WV ........................................ 108 Figure 14 - Square and Round Plates ...................................................................... 108 Figure 15 - Wire Mesh ................................................................................................ 109 Figure 16 - Anomaly ................................................................................................... 110 Figure 17 - Picture of an Omega Block.................................................................... 125 Figure 18 - Sketch of the Lake Lynn Mine ............................................................. 129 Figure 19 - Mortar in Vertical Joint.......................................................................... 132 Figure 20 - Post-Explosion Location of Seal No. 1 ................................................ 133 Figure 21 - Post-Explosion Location of Seal No. 2 ................................................ 134 Figure 22 - Post-Explosion Location of Seal No. 3 ................................................ 134 Figure 23 - Post-Explosion Location of Seal No. 4 ................................................ 135 Figure 24 - Post-Explosion Location of Seal No. 5 ................................................ 135 iv Figure 25 - Post-Explosion Location of Seal No. 6 ................................................ 136 Figure 26 - Post Explosion Location of Seal No. 7................................................. 136 Figure 27 - Post-Explosion Location of Seal No. 8 ................................................ 137 Figure 28 - Post-Explosion Location of Seal No. 9 ................................................ 137 Figure 29 - Post-Explosion Location of Seal No. 10 .............................................. 138 Figure 30 - Test No. 1 Lake Lynn Mine Layout ..................................................... 140 Figure 31 – Test No. 2 Lake Lynn Mine Layout..................................................... 141 Figure 32 - Test No. 3 Lake Lynn Mine Layout ..................................................... 142 Figure 33 - Test No. 4 Lake Lynn Mine Layout ..................................................... 143 Figure 34 - Test No. 5 Lake Lynn Mine Layout ..................................................... 144 Figure 35 - Test No. 6 Lake Lynn Mine Layout ..................................................... 145 Figure 36 - Cloud to Ground Lightning.................................................................. 155 Figure 37 - Intra-cloud Lightning............................................................................. 155 Figure 38 - Cloud to Cloud Lightning ..................................................................... 155 Figure 39 - Upward Lightning .................................................................................. 156 Figure 40 - Damaged Tree.......................................................................................... 157 Figure 41 - Damaged Insulator ................................................................................. 161 Figure 42 - Damaged Lightning Arrester................................................................ 161 Figure 43 - Cable Coupler.......................................................................................... 170 Figure 44 - Coupler with Pins and Conductors ..................................................... 171 Figure 45 - Two Pieces of the Cable......................................................................... 171 Figure 46 - Picture taken Near Survey Station 4010 ............................................. 174 Figure 47 - Picture Taken Near Survey Station 4011 ............................................ 174 Figure 48 - Picture Taken Near Seal No. 8.............................................................. 175 Figure 49 - Picture of Debris Outby Seal No. 2 ..................................................... 178 Figure 50 - Contours Near Seal No. 10 in 2 North Mains, No. 9 Entry.............. 182 v APPENDICES Appendix A - List of Deceased and Injured Miners Appendix B - Detailed Map of Mine Appendix C - Mine Rescue Personnel and Teams Responding Appendix D - BCMR Air Quality Measurements Taken On January 2 and 3, 2006 Appendix E - Gas Chromatograph Analysis Results for the No. 1 Drift Opening and Borehole No. 1 Appendix F - Accident Investigation Data – Victim Information Appendix G - Lists of Individuals Who Assisted with the Investigation Appendix H-1 through H-9 - Mapping of the Entire Mine Appendix I - Executive Summary of “Investigation of Pyott-Boone Electronics MineBoss Monitoring and Control System” Appendix J - Bottom Mining Supplements to the Ventilation Plan Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals Appendix L - Pre-Explosion Simulation of the Mine Ventilation System Appendix M - Post-Explosion Simulation of the Mine Ventilation System with the Damaged Ventilation Controls Appendix N - Post-Explosion Simulation of the Mine Ventilation System with the Initial Repairs made to the Damaged Ventilation Controls Appendix O - Evaluation of Potential for a Roof Fall to Ignite a Methane-Air Mixture Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 Appendix Q - Results of the Mine Dust Survey Appendix R - Map Showing the Location of all Intended Mine Dust Sample Locations and Results Appendix S - Executive Summary of Inspection of Sago Mine Voice Communications Equipment Appendix T - Executive Summary of the Trolleyphone Repeater Report Appendix U - An Executive Summary of Investigation of the Motorola Two-way Radios Appendix V - Executive Summary of the Evaluation of the Uniaxial Compressive Strength of Burrell “Omega” Blocks vi Appendix W - Sampling and Testing of Mortar Bed Cores Taken from Failed Ventilation Seals Appendix X - Experimental Study of the Effect of LLEM Explosions on Various Seals and Other Structures and Objects Appendix Y-1 and Y-2 - Map of the Electrical System, Equipment, and Associated Items Appendix Z - Executive Summary of Portable Gas Detector Testing Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports Appendix BB - Map Showing Sago Mine in Relation to Recorded Location of Lightning Strikes, a Lightning Damaged Poplar Tree and the Mine’s Phone and Power Lines Appendix CC - Results from Analysis of Seismic Data Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Appendix EE - Report on the Investigation of the Well Heads and Gas Pipeline System Appendix FF – Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix GG - Map Showing Sago Mine in Relation to Recorded Locations of Lightning Strikes, Gas Wells and Gas Lines Appendix HH - Observation and Sampling Collection Methodology Appendix II - Executive Summary of Submersible Pump Parts Recovered from Sago Mine Appendix JJ - Sago Mine Pump Cable Test Appendix KK - Map Showing Earth Resistance Measurement Values Appendix LL - Mine Map Detailing the Extent of Flame and the Direction of the Primary Explosion Forces vii SAGO MINE DRIFT OPENINGS Explosion Origin 2nd Left Parallel Section 1st Left Section Recently Sealed Area 2 North Mains Seals Previously Sealed Areas Drift Openings OVERVIEW On January 2, 2006, an explosion occurred at approximately 6:26 a.m. in Wolf Run Mining Company’s Sago Mine. At the time of the explosion, 29 miners were underground. Twelve miners lost their lives, and one was seriously injured. The explosion occurred inby the 2 North Mains seals, and destroyed all ten of the seals used to separate the area from the active portion of the mine. The weather conditions at the mine were unseasonably warm with the temperature near 45 degrees Fahrenheit (F). A storm, accompanied by heavy rain, thunder and lightning, was in the area. Before entering the mine, some Sago miners saw lightning strikes near the property. A preshift examination of the mine had been conducted. One mine examiner remained underground. The 2nd Left Parallel crew and another miner entered the mine at about 6:00 a.m. The 1st Left crew and three other miners entered the mine shortly thereafter. The 2nd Left Parallel crew arrived on their working section, and the 1st Left mantrip arrived at the 1st Left switch. Shortly thereafter, an explosion occurred. One miner died of carbon monoxide (CO) poisoning shortly after the explosion. The 2nd Left Parallel miners’ attempt to evacuate was unsuccessful, and they barricaded themselves on the 2nd Left Parallel section. All other miners eventually evacuated the mine. Mine management officials entered the mine in an attempt to assess the situation. The 1st Left Foreman remained underground and eventually joined this group. They found that the explosion damaged ventilation controls. In an effort to reach the missing miners, they attempted to restore ventilation, using temporary ventilation controls. They were unable to clear the smoke and gases, and eventually ended their rescue attempt and evacuated the mine. Federal and state agencies responded to the accident. Mine rescue teams were organized, a command center was established, and a rescue effort was initiated. Entry into the mine was delayed because of elevated levels of CO and methane. Preparations were started to drill a borehole into the 2nd Left Parallel section for sampling and communications purposes. Rescue teams entered the mine after the concentration of gases stabilized. They found the first victim on January 3, near the 2nd Left Parallel track switch. Later that evening, rescue teams advanced into the 2nd Left Parallel section where twelve miners were found behind a barricade. One miner was found alive. He was rescued and transported to a hospital. On January 4, the 12 victims were 1 recovered from the mine. A list of the deceased and injured miners is contained in Appendix A. Working with the West Virginia Office of Miners’ Health, Safety and Training (WVMHS&T), the mine operator, and miners’ representatives, the Mine Safety and Health Administration (MSHA) launched an investigation into the events surrounding the fatal accident. The investigative team interviewed people with knowledge of the mine or the accident. Investigators mapped the mine, reviewed mine records and gathered relevant physical evidence from underground. The evidence was evaluated. Investigators determined that methane began to accumulate within an area which had previously been mined and then sealed with 40 inch thick Omega block seals. The explosion occurred within the sealed area and destroyed the seals. This caused portions of the mine to fill with toxic levels of CO. At MSHA’s request, the National Institute for Occupational Safety and Health (NIOSH) conducted a full-scale testing program designed to determine the strength of the Omega block seals and to gather information about explosions in sealed areas. The mine operator failed to build the seals in accordance with the approved plan. However, the testing showed that the seals, as built at the mine, would likely have withstood pressures of 20 pounds per square inch (psi), as required by regulation. The explosion in the mine is believed to have generated pressures in excess of 93 psi. The discrepancies between the actual seal construction and the approved plan, as well as all other non-contributory conditions observed during the investigation, were cited under a separate inspection activity. MSHA collected Self-Contained Self Rescuer units (SCSRs) used by the miners, and tested them. The mine operator did not keep adequate records on all of the units, and one unit was out-of-date. Some of the miners had trouble donning their SCSRs and breathing through them. However, testing indicated that the units produced oxygen as intended. Investigators determined that coal dust did not play a major role in the explosion. Potential ignition sources were investigated. There was no evidence that cutting, welding, mining operations, smoking, or spontaneous combustion were involved in the ignition. Electrical systems and equipment were also ruled out as possible ignition sources. Although a roof fall cannot be definitively excluded as a potential ignition source, it is a highly unlikely ignition source. Lightning strikes were recorded near the mine at approximately the same time as a seismic event occurring in the area and the initial alarm from the mine’s atmospheric monitoring system (AMS). MSHA contracted with Sandia Corporation, the operator of the Sandia National Laboratories (Sandia), to 2 perform modeling and testing to ascertain if it was possible for lightning to cause electrical energy to enter the mine and cause an explosion. Sandia determined that a lightning strike could create enough energy in the sealed area to initiate an arc. Lightning has been determined to be the most likely ignition source. 3 GENERAL INFORMATION The Sago Mine is located near Tallmansville, Upshur County, West Virginia. The mine opened in 1999 as the Spruce No. 2 Mine operated by BJM Coal Company. The mine changed ownership in 2002. Anker West Virginia Mining Company, Inc., a subsidiary of Anker Group, Inc., acquired the company and renamed the mine as Sago Mine in 2003. International Coal Group, Inc. (ICG) acquired Anker Group, Inc. in 2005. The Anker West Virginia Mining Company, Inc. was renamed the Wolf Run Mining Company in late 2005. Principal Officers of ICG were Bennett K. Hatfield, President and Chief Executive Officer; Oren E. (Gene) Kitts, Senior Vice President, Mining Services; Samuel R. (Sam) Kitts, Senior Vice President of Operations; and Timothy Martin, Corporate Director of Health and Safety. ICG owns and operates a number of mining properties throughout the United States. The management structure at the mine was similar to that traditionally found at coal mines throughout the United States. The direct line of supervision consisted of mine superintendent, mine foreman and foremen. Mine Superintendent Jeffrey Toler was head of the on-site mine management organization at the mine and was responsible for the overall operation of the mine. Mine Foreman Carl Crumrine was responsible for all underground activities including countersigning various mine record books. Safety Director James Schoonover was responsible for mine safety issues and training, as well as accompanying state and federal inspectors while on mine property. Maintenance Superintendent Denver Wilfong was in charge of all electrical and equipment related issues. A number of shift foremen, section foremen and outby foremen were responsible for coal production and general support operations. The mine opened into the Middle Kittanning coal seam through five drift openings. The drift openings were located in a box cut where the overburden material was removed down to the coal seam level. Drift openings were numbered with the No. 1 Drift Opening on the extreme left side. The mine fan was located in the No. 5 Drift Opening on the extreme right side of the highwall. The developing entries were numbered separately from the drift openings, from left to right, with the No. 1 entry on the extreme left side. Mine personnel identified locations in the mine by the numbered crosscut along with the corresponding main belt, for example, 34 Crosscut, No. 2 Belt. Coal was produced from the 1st Left and 2nd Left Parallel sections. The majority of the 1st NE Mains was sealed. The 2nd Left Mains were also sealed. A map of the mine 4 is illustrated in Figure 1 to provide an overview of the mine. A detailed map of the mine is shown in Appendix B. Figure 1 - Sketch of Sago Mine The mine work schedule consisted of two overlapping 10 hour shifts beginning at 6:00 a.m. and 3:00 p.m. and one maintenance shift beginning at 12 midnight, Monday through Thursday. The weekend schedule (Friday, Saturday and Sunday) was composed of two overlapping 13 ½ hour production shifts, and one 8 hour maintenance shift. Coal was produced from two sections. Two remotecontrolled, continuous mining machines and two twin boom roof-bolting machines operated in each of the 1st Left and 2nd Left Parallel sections. There were three electrically powered shuttle cars located in 1st Left and four in 2nd Left Parallel. The two continuous mining machines in each section were not operated simultaneously. One mining machine completed a cut sequence and was idled. The other mining machine proceeded to cut another sequence. Sections were developed by advancing eight entries. The approved roof control plan allowed for main entries, sub-main entries and rooms to be developed 20 feet wide, on centers from 48 feet to 110 feet. Crosscuts could range from 54 feet to 140 feet centers in the mains, 48 feet to 140 feet centers in the sub-mains and 40 feet to 140 feet centers in rooms. The average mining height was approximately 7 feet. In addition to the normal mining development, bottom mining was conducted in some areas of the mine. The bottom mining was the removal of the lower bench of the Middle Kittanning coal seam. When mining was completed in an area, or adverse conditions were encountered that ended development, this method was used to maximize coal yield. Because of the extreme heights that would have been created during initial development, the bottom portion of the coal seam was not mined at that time. During and after removal, no miner was permitted in the 5 mined out area. This precaution eliminated exposure to high, unsupported coal ribs. Rock dust was applied during initial development as required. Additional rock dust was not applied in areas that had been bottom mined. A portion of the sealed 2nd North Mains and 2nd Left Mains area had been bottom mined. The A-1 and A-2 Panels off of 1st Left were also bottom mined. Verizon provided telephone service to the surface buildings. The underground mine communication system consisted of pager phones, trolleyphones and wireless handheld two-way radios. Battery-powered track mounted personnel carriers (mantrips) and locomotives were used to move men and materials throughout the mine. The mine dispatcher was located in an office on the surface. The dispatcher directed and monitored all traffic entering, traveling throughout, and exiting the mine. He also monitored an AMS that consisted of sensors placed throughout the mine that relayed information to a central computer. This system displayed a continuous readout of CO levels at each sensor located along each belt conveyor entry, belt startup, belt shutdown and mine power status. The Mine Emergency Evacuation and Firefighting Program of Instruction designated the dispatcher as the responsible person in the event of an emergency. A large portion of the mine was wet, and pumps controlled the water accumulations. Coal was transported from the working sections to the surface by a series of conveyor belts, and was then loaded onto trucks, transported to a nearby cleaning plant, and processed. In 2005, the mine produced approximately 1,700,000 tons of raw materials, which resulted in 507,775 tons of clean coal. This resulted in a recovery ratio of approximately 30 percent. Reportedly, this ratio did not change significantly during bottom mining. A blowing fan, located on the surface, ventilated the mine. The mine fan produced approximately 146,000 cubic feet per minute (cfm) of air. Mine inspection records in October 2005 indicated that the mine liberated approximately 90,500 cubic feet per day (cfd) of methane. A single split ventilation system was used in each of the two sections. Intake air was typically directed through the Nos. 7 and 8 entries and returned out the Nos. 1 and 2 entries. The Nos. 3 through 6 entries were ventilated with intake air generally traveling in the outby direction. Air lock doors were installed in the track entry, one door was located between 8 and 9 Crosscuts, No. 1 Belt and another door was located between 12 and 13 Crosscuts, No. 1 Belt. These doors allowed for the passage of men and materials without disrupting the air current. To accomplish this, only one door was opened at a time. The mine employed approximately 135 underground miners and six surface miners. At the time of the accident, the miners were not represented by a labor 6 union for collective bargaining purposes. During the investigation, two separate miners’ representative groups were selected to represent the miners. One group of miners selected the United Mine Workers of America (UMWA) and the other selected a group of Sago miners. Both groups participated in portions of the onsite investigation. Table 1 shows the Fatal and Non-Fatal Days Lost (NFDL) accident incidence rates for the mine along with the comparable national rates for all underground coal mines, for years 2004 and 2005. Table 1 - Accident Incidence Rates Calendar Year Incidence Rate Sago Mine Incidence Rate National National All Incident Rate 2004 NFDL/Fatal 15.90/0.00 NFDL/Fatal 10.22/0.00 NFDL/Fatal 5.98/0.04 NFDL/Fatal 5.42/0.03 National/Sago 8.42/19.88 National/Sago 7.71/12.41 2005 MSHA completed its last regular health and safety inspection of Sago on September 30, 2005. MSHA started a new inspection on October 3, 2005. The inspection was ongoing at the time of the accident. Table 2 summarizes MSHA enforcement actions at the mine in 2005 prior to the accident, and references the number of citations issued to the operator under provisions of the Federal Mine Safety and Health Act of 1977. Table 2 - Enforcement Actions in 2005 Number Initiated - 208 (2 vacated actions excluded) Type Enforcement Action 104(a) non-S&S citation 85 104(a) S&S citation 96 104(b) order 1 104(d)(1) citation 1 104(d)(1) order 2 104(d)(2) order 14 107(a) order 1 314 (b) safeguard 5 103(k) order 3 7 At the time of the accident, eight citations had not been terminated. They were not associated with the accident. Three of the citations involved tunnel liners, two were in the primary escapeway, two were electrical and one was for guarding. These violations occurred in outby areas not related to or directly affected by the explosion. Based on enforcement action taken during previous inspections, the operator was subjected to a higher level of enforcement pursuant to section 104 (d) of the Federal Mine Safety & Health Act of 1977. 8 EVENTS LEADING TO THE ACCIDENT Development of the 2 North Mains was stopped in June of 2005 due to excessive water and adverse roof conditions. The 2nd Left Mains were subsequently mined until August of 2005 when adverse roof conditions and water inflow again caused development to stop. On September 28, 2005, the operator submitted a plan to bottom mine the 2nd Left Mains. The plan was approved on September 28, and bottom mining was started shortly thereafter. On October 3, 2005, the operator submitted a plan to extend bottom mining in the inby portions of 2 North Mains. The plan was approved on October 4, and bottom mining of the 2 North Mains was conducted. Upon completion of the bottom mining, the equipment was moved to the 2nd Left Parallel. On October 12, the mine operator submitted a plan to MSHA to seal the 2 North Mains inby the 2nd Left Parallel. The mine operator also submitted a plan to use Omega Blocks to construct 40 inch thick Omega Block seals. On October 24, the mine operator’s requests were approved. Seal construction began on October 24, 2005. By November 9, seven seals had been completed. The operator subsequently completed the next seal in the 63 Crosscut, No. 4 Belt between entry Nos. 2 and 3. The locations of the seals are shown in Figure 2. Figure 2 - Location of 2 North Main Seals Ventilation controls, including stoppings and overcasts, also had to be modified to accommodate the air change associated with sealing. By December 11, 2005, 9 the operator had completed the last two seals in the Nos. 1 and 9 entries, and made those ventilation changes. On Friday, December 30, 2005, coal was produced. Miners did not produce coal on Saturday, December 31, 2005, but two shifts performed equipment maintenance, roof bolting, rock dusting, relocating equipment outby from the faces, and other duties. Miners did not produce coal on Sunday, January 1, 2006, but four miners worked the day shift, hauling and installing track ballast, performing maintenance on water pumps in 2nd Left Parallel section and at 46 Crosscut, No. 4 Belt and repairing the trolleyphone communication system. After performing maintenance on the pump in 2nd Left Parallel, they pumped the standing water in that area, and turned off the power to the pump. They repaired the trolleyphone communication system by resetting an electrical breaker located at 9 Crosscut, No. 4 Belt. According to the miners, the trolley system worked fine for the rest of the shift. After the completion of the day shift, the mine was idled. Mine Examiners Fred Jamison and Terry Helms arrived at the mine around 2:15 a.m. on January 2, 2006, to conduct preshift examinations prior to the oncoming day shift. Dispatcher William Chisolm said he arrived about 3:30 a.m. to monitor communications and the AMS. Helms and Jamison indicated that they entered the mine at approximately 3:00 a.m., although Chisolm believed it was 4:00 a.m.1 Helms traveled into the mine by mantrip through the track entry. Jamison walked into the mine through the belt entry, and examined that entry to the 11 Crosscut, No. 1 Belt area where he walked into the track entry and met Helms. Jamison boarded the mantrip with Helms and they traveled to the No. 3 Belt drive. Jamison exited the mantrip at the No. 3 Belt drive and walked the belt entry to No. 4 Belt drive. Helms continued to the No. 4 Belt drive where he left the mantrip, traveling the No. 4 Belt to the mouth of the 1st Left Section and examined the 1st Left section. Jamison boarded the mantrip at No. 4 Belt drive and traveled the track entry to 2nd Left Parallel switch. He then examined the belt entry into the 2nd Left Parallel section. Jamison started his examination in the No. 1 entry at approximately 4:00 a.m. He determined the air quantity in the last open crosscut between the intake and return aircourses, which measured 11,241 cfm. Jamison continued across the section from left to right conducting his examination of the working places and the remainder of the section. He detected no methane during his examination of the section, which he completed at approximately 4:25 a.m. He examined the track entry to the 2nd Left Parallel 1 Jamison and Chisolm confirmed that they had a conversation prior to Jamison and Helms going underground, so their recollections regarding times may not be completely accurate, which they both acknowledged. 10 switch where he boarded the mantrip and traveled to 1st Left switch. He called the dispatcher and informed him that he would leave Helms’ dinner bucket and coat at the 1st Left switch. Jamison continued on the mantrip toward the mine opening. He stopped at a power center at 17 Crosscut, No. 3 Belt. He walked to the water pump in the return entry at 22 Crosscut, No. 3 Belt either before or after making an unsuccessful attempt to start the pump by resetting the breaker. Jamison boarded his mantrip and rode to No. 3 Belt drive. He exited the mantrip and examined the belt drive. Jamison walked to No. 2 Belt drive, examined it, and returned to No. 3 Belt drive. Jamison drove to No. 1 Belt drive. From there, he walked to No. 2 Belt drive and checked the pump across from No. 2 Belt drive. He then returned to No. 1 Belt drive, boarded the mantrip and proceeded to the surface, arriving at approximately 5:40 a.m. While Helms and Jamison were conducting the preshift examination, other miners were arriving on the surface and preparing to start their 6:00 a.m. shift. Jamison exited the mine and told Pumper John N. Boni about the malfunctioning water pump at 22 Crosscut, No. 3 Belt. He also informed 2nd Left Parallel Section Foreman Martin Toler Jr. what he found during his preshift examination. Jamison entered his preshift examination results in the preshift examination record book, noting no hazards and no methane. Jamison walked back into the mine at approximately 6:00 a.m. and went to the No. 2 Belt drive to shovel coal spillage. After Helms left the mantrip at No. 4 Belt drive, he walked to the 1st Left section and conducted a preshift examination between 4:20 and 4:50 a.m. Helms decided not to come out of the mine, so he called outside and told 1st Left Section Foreman Owen Jones what he found during his preshift examination. Owen Jones did not know Helms’ location when he called. The preshift examination of the section revealed no methane, 14,510 cfm in the last open crosscut between the intake and return air courses and no hazards. Chisolm called Helms underground to inform him where Jamison had left his lunch bucket and coat. Chisolm did not know Helms’ location when he spoke with him. Helms eventually proceeded to the 2nd Left Parallel switch area. The 2nd Left Parallel section crew consisted of 12 miners: Thomas P. Anderson, Alva M. Bennett, James Bennett, Jerry Groves, George Hamner Jr., Jesse Jones, David Lewis, Randal McCloy Jr., Martin Toler, Jr., Fred Ware, Jackie Weaver, and Marshall Winans. They boarded a mantrip operated by Jesse Jones and entered the mine through the track entry at approximately 6:00 a.m. Twelve miners were on the 1st Left section crew: Denver D. Anderson, Paul Avington, Gary B. Carpenter, Randall Helmick, Eric M. Hess, Owen Jones, Hoy 11 S. Keith, Jr., Arnett R. Perry, Gary Rowan, Harley J. Ryan, Christopher Tenney, and Anton R. Wamsley. The crew started to board a mantrip with John Boni, Belt Cleaner John P. (Pat) Boni, and Mine Examiner Ronald Grall. The crew realized that the mantrip they had was too small, and exchanged it for a larger one. The 15 miners boarded the mantrip operated by Owen Jones and entered the mine at approximately 6:05 a.m. The 1st Left mantrip traveled to the 1 Right switch where John Boni exited the mantrip at approximately 6:14 a.m. He walked to the power center at 17 Crosscut, No. 3 Belt. John Boni moved the electrical plug from one receptacle to a receptacle protected by a larger breaker. He then walked to the pump at 22 Crosscut, No. 3 Belt in the return air course and confirmed it was operating. He started walking back toward the belt entry. At approximately 6:19 a.m., Pat Boni exited the mantrip in the track entry near No. 4 Belt drive and walked to the belt drive. Pat Boni walked to 39 Crosscut, No. 3 Belt, refilled the trickle duster with rock dust and turned it on. The mantrip stopped at the 1st Left switch. Perry exited the mantrip and threw the switch. He picked up a ladder near the switch and placed it on the mantrip. 12 DESCRIPTION OF THE ACCIDENT At approximately 6:26 a.m., Perry re-entered the mantrip at the 1st Left switch and was sitting down when a violent blast of air, smoke, dust and debris struck the mantrip and the miners. Owen Jones, the operator of the mantrip, tried to get into position to move the mantrip, but the force knocked him off of it, causing him to lose his hard hat. Owen Jones did not hear an explosion but estimated that the force lasted about 6 to 8 seconds. The other crewmembers’ estimates ranged from 4 to 15 seconds. Owen Jones also stated that the dust in the air was so thick that he was unable to see, but that he did not smell any smoke at that time. Owen Jones’ handheld detector alarmed, but he was unable to read the display showing the concentrations of methane, oxygen and CO in the air because of the dust. The force knocked off the hats, lights and glasses of some of the other 12 miners, and forced dust into some of their eyes and faces. The crew stated that they did not hear an explosion, or see any type of flash or flame. The records of the AMS indicated that it alarmed at 6:31:31 a.m. but the clock in the AMS was four minutes and 56 seconds fast, so the time of the alarm was actually 6:26:35 a.m. At that time, the CO sensor at 57 Crosscut, No. 4 Belt alarmed, showing 51 parts per million (ppm). This was the first indication on the surface of something unusual occurring underground. John Boni was in the No. 3 return entry at 22 Crosscut, No. 3 Belt next to the mandoor leading into the belt entry when he felt a rush of air. The power to the pump went off. Boni felt that the air was not very forceful, and was similar to a small pillar fall. He did not see any dust. Pat Boni checked No. 4 Belt drive and started walking inby in the belt entry to check the belt take-up unit, when he a felt a rush of air and dust from inby hitting him in the face. He grabbed his hat to keep from losing it, but estimated that the rush of air lasted only a second. The visibility after the rush of air was about 14 to 15 feet. His initial thought was that a roof fall had occurred nearby. As Jamison was shoveling at the No. 2 Belt drive, he felt pressure in his ears. He thought there might have been a roof fall. Once the rush of air subsided, the 1st Left crew began to exit the mantrip. Some of them felt heat. Rowan stated that the mantrip could not be used to evacuate the mine due to debris on the track. Rowan said that Perry shouted that the mine had blown up. Wamsley described the air as a yellowish brown. Owen Jones immediately instructed his crew to stay together and begin their evacuation outby on foot by walking in the track entry. However, the group did not stay together. Grall, Hess and Wamsley went ahead of the others working their way outby along the left rib of the track entry. Anderson stated that he thought two 13 or three people said the crew should put their SCSR on, but he did not remember who they were. Wamsley stated that he thought Owen Jones shouted to the crew to put their shirts over their mouths until they donned their SCSRs. Visibility was very poor due to dust and smoke, with some miners describing it as no more than 8 to 10 inches. Hess stated that initially some miners tried to stay together by grabbing another miner’s shirt, belt, belt loop or anything else they could. Their attempts to stay together were made even more difficult as they stumbled over debris from damaged ventilation controls. Grall, Hess and Wamsley arrived at the first mandoor at 48 Crosscut, No. 4 Belt where they checked the No. 7 entry (the primary intake escapeway) and found the atmosphere to contain heat, dust and smoke, causing poor visibility. Hess and Wamsley decided that the situation was not good, and donned their SCSRs while in the No. 7 entry before going back to the track entry. Neither person had any difficulty donning their SCSR. Grall did not don his SCSR. He traveled back to the track entry and met Avington. Grall and Avington continued outby in the track entry looking for another mandoor, in an attempt to re-enter the primary intake escapeway. Somewhere between the mantrip and three to four crosscuts outby, Avington asked Grall if they should don their SCSRs. Grall said no, that they should just keep moving outby. Hess and Wamsley re-entered the track entry through 48 Crosscut, No. 4 Belt. When they arrived in the track entry, they met Ryan and Anderson. Wamsley suggested that Ryan and Anderson don their SCSRs, and assisted Ryan with his. Ryan told investigators that he had difficulty grasping the tab to open the unit, and difficulty removing the bottom portion of the unit. They had to jerk the bottom of the unit two or three times to remove it. In addition, Ryan had difficulty breathing with the unit and, as a result of not having teeth, had difficulty keeping the mouthpiece in his mouth. Hess assisted Anderson in donning his SCSR. Anderson stated that he had no trouble donning his SCSR. Hess said that he had trouble helping Anderson remove the SCSR from its pouch, since it had sealant on it, and because there was a pair of channel locks in the pouch. Once he removed the channel locks, Hess was able to pull the unit from the pouch, remove the metal straps from the top and bottom, and hand it to Anderson. Anderson was then able to complete the donning process and activate the unit. Anderson felt that his SCSR performed well, and the only problem was that the unit became warm. Keith stated he was a little disoriented, and Wamsley assisted him in donning his SCSR. Wamsley stated that Keith’s SCSR would not activate. He pulled the activation cord but it did not work. Keith thought the SCSR did function as intended, but that it did not make it easier to breath, because of the dust in his 14 mouth. Hess remembered Keith stating that his SCSR was not working the way it should. Wamsley stated that when they re-entered the track entry he heard someone outby shouting to come his way. After donning their SCSRs, Ryan, Anderson and Keith continued outby in the track entry. Helmick, Tenney and Carpenter stated that they did not don their SCSRs because they did not have trouble breathing, and thought that they may need them later. Owen Jones did not don his SCSR but acknowledged that he should have. As Helmick, Tenney and Carpenter made their way outby in the track entry; they assisted other crew members who were having some difficulty walking. Tenney also noticed that air was hitting him in the face. This would indicate that the air was flowing from the outby to the inby direction, meaning that the air had reversed. As Owen Jones made his way outby he came upon Perry, who had fallen down. Helmick arrived, and Owen Jones asked him to help Perry continue his evacuation. Helmick told Owen Jones that the other crew members were coming behind him, but he could not see them due to the thick dust. Perry said that he had lost his hat and the lens was broken on his cap lamp. During his evacuation in the track entry he decided to drop his damaged light so that he would have less weight to carry. Grall and Avington continued outby ahead of the others, feeling their way along the dusty and debris-filled track entry, looking for another mandoor between the track and primary intake escapeway entry. Eventually they arrived at 37 Crosscut, No. 4 Belt where they entered the primary intake escapeway in the No. 7 entry. While there was still dust in the air there, the visibility was much better than in the track entry, with Grall reporting it being about 10 feet. Grall’s detector was alarming, and the methane reading was one percent and falling. Grall did not state what the reading had been in the track entry. Grall continued to monitor his detector; with the last reading he recalled being 0.8%. He also recalled that the CO level at that time was 66 ppm and dropping. Avington remained in the intake while Grall returned to the track entry to check for the remaining crew members. The track entry was still dusty, and he could not see anyone. He shouted but did not receive an answer. He traveled outby to 30 Crosscut, No. 4 Belt and then returned inby to 38 Crosscut, No. 4 Belt without seeing anyone. He continued to shout in an attempt to make contact with the remaining crew members. He finally saw lights through the dusty atmosphere as the crew made their way in his direction. Grall estimated it took about another 5 minutes for everyone to reach the intake entry at 37 Crosscut, No. 4 Belt. 15 Ryan stated that once he reached the intake at 37 Crosscut, No. 4 Belt the “bottom part” of his SCSR air bag started to collapse. As he escorted Perry it became difficult to breath, and the unit was getting warm. Ryan felt that he needed more air than was being produced. Therefore, he would slow down when the unit would become warm. He said that the unit never actually stopped producing oxygen, and that the fresh air produced by the SCSR was better than the atmosphere in the mine. After the rush of air, Jamison walked along No. 2 Belt toward No. 3 Belt drive to see if a stopping had been damaged. He did not find any damaged stoppings. Pat Boni went to the mine phone located near the No. 4 Belt drive at approximately 6:32 a.m., called Chisolm on the surface and asked what had happened. Chisolm replied that lightning had knocked out some of the underground power. Pat Boni replied he did not think that was what had happened, since dust was moving inby rather that outby, the opposite direction in which air normally flowed. Chisolm confirmed after talking with Yardman Gary Marsh that the fan was still running. Pat Boni reiterated that the air was flowing inby, indicating something was wrong, and asked about the belts. Chisolm responded that Nos. 1, 2 and 3 Belts were operating, but Pat Boni could see that Nos. 3 and 4 Belts were not running, and he told Chisolm so. Pat Boni also said that the rock duster which he had started earlier was not operating, and the power center at No. 4 Belt drive was not energized. John Boni went through the mandoor into the belt entry and continued to the track entry. There he saw a large amount of white dust that appeared to be rock dust. However, he did not see smoke, feel heat or hear anything. He noticed that the dust was just hanging in the air and not moving. John Boni immediately went to the mine phone at 1 Right switch to call Chisolm. He heard Pat Boni’s conversation with Chisolm. Pat Boni told John Boni what he observed at No. 4 Belt drive and that he thought there may have been a roof fall. Pat Boni said he would walk inby on the track entry to look for one. Pat Boni stated he then went to the track entry and walked inby to the maintenance shanty, which he thought was at 8 Crosscut, No. 4 Belt.2 He found it to still be dusty with the air still moving inby at 9 Crosscut, No. 4 Belt. He called the dispatcher from the phone at 9 Crosscut, No. 4 Belt to find out what happened. Chisolm responded that he did not know. Pat Boni told him that the air was going inby, and that he thought a fire or explosion had occurred. 2 The maintenance shanty was actually at 9 Crosscut, No. 4 Belt. 16 As the crew members were arriving at 37 Crosscut, No. 4 Belt and making their way into the No. 7 entry, Owen Jones stopped to use the mine phone near 37 Crosscut, No. 4 Belt in the track entry. He estimated that the call was made about 5 minutes after the explosion. At approximately the same time the explosion occurred, Chisolm was speaking on the mine phone with Mine Superintendent Jeffrey Toler. Jeffrey Toler was in the building next to the dispatcher’s office when a flash of lightning and loud thunder occurred. Chisolm heard a loud popping/ringing noise in the phone that caused pain in his ear, and made him drop the phone. After picking the phone back up, he told Jeffrey Toler that he had lost the AMS and that the belts were down. Jeffrey Toler could hear the AMS alarms over the mine phone. Jeffrey Toler told Chisolm to radio the 1st Left and 2nd Left Parallel Crews and ask them to check all the CO sensors which were alarming to determine the problem. Chisolm also spoke on the phone to Wilfong, who was in his office, and told him that he had lost communications on the AMS. Wilfong thought that fuses must have blown, so he gave Maintenance Foreman Vernon Hofer, who was in his office at the time, a handful of fuses. He instructed Hofer to check the AMS and replace any blown fuses. Hofer proceeded to the dispatcher’s office, checked in and obtained his cap lamp. He looked at the CO monitor screen to see which belts were affected and went to the pit to prepare to go underground. The Nos. 1 and 2 Belts were operating but the Nos. 3 and 4 Belts had lost power. The Nos. 5 and 6 Belts had not been operating when the explosion occurred. Owen Jones called outside about “five minutes maybe” after they felt a rush of air. Owen Jones spoke over the mine phone to Chisolm, Jeffrey Toler and Wilfong on the surface, while John and Pat Boni listened in. Owen Jones said “I called out and I said, we’ve had a mine explosion in here. I said, get mine rescue team here now.” He also indicated that there was a rush of air from the direction of 2nd Left Parallel, and that there was smoke. He was directing his men to the primary intake escapeway. After completing his phone conversation Owen Jones made his way to the No. 7 entry where he joined his crew members. While at No. 3 Belt drive, Jamison overheard Owen Jones on the mine phone. Owen Jones was relaying his belief that there had been an explosion and that he was going to have his men evacuate the mine. Jamison decided to start walking toward the surface. While evacuating the mine he noticed that most of the mandoors along No. 1 Belt were open, and he shut them as he walked out. He did not indicate on which side of the track the doors were located. Jamison did not don his SCSR but said that he had it in his hand ready to don if needed. 17 Wilfong told Pat Boni over the phone to get in the intake and evacuate the mine. Pat Boni walked back to No. 4 Belt drive and picked up his lunch bucket and coat. He entered the primary intake escapeway through a mandoor across from the No. 4 Belt drive where the air was clear. Pat Boni walked the primary intake escapeway out to 4 Crosscut, No. 1 Belt. There he opened a mandoor between the intake and track entry. Seeing no smoke, he exited into the track entry and walked four crosscuts to the surface, arriving at about 7:25 a.m. During his evacuation, he did not see any other miners. Pat Boni did not don his SCSR. He felt that he was in good air since he did not see or smell smoke. When he arrived, there was no one in the pit area. Pat Boni immediately called the dispatcher from the phone in the pit to notify him that he was out of the mine. John Boni asked Chisolm what the situation was, and he replied that there was a storm and that the power to No. 3 or No. 4 Belt had been lost, but that Nos. 1 and 2 Belts were still operating. John Boni stated that Wilfong and one of the other mechanics were on the phone, and that one of them said that they were coming in to reenergize No. 3 Belt. John Boni told Chisolm to have them wait until he checked for a possible roof fall. He walked inby on the track entry about eight to ten crosscuts, but did not find a roof fall and returned to 1 Right switch. However, he did notice that dust was hanging in the air. John Boni called the surface again and spoke to Marsh. John Boni was thinking that there may have been an explosion and asked Marsh if there were any AMS sensors showing readings of CO. Marsh replied: “the ones on Two Left, the Two Left belt line, were showing CO. He told me what it was, 107 and 170 or something like that.” John Boni finished his conversation with Marsh when Owen Jones and Jeffrey Toler spoke on the phone. Jeffrey Toler asked John Boni where he was located. He then told John Boni to stay there so that he could pick him up. Owen Jones also informed John Boni that he was sending his crew out the primary intake escapeway and asked him to watch for them. After the conversation with Owen Jones, Jeffrey Toler was concerned that the 2nd Left Parallel crew had not responded. He shouted to Safety Director James Schoonover, who was across the hall, to prepare to go underground. Jeffrey Toler, Schoonover and Wilfong then prepared to go underground. Wilfong called Hofer, who was in the pit area, and told him to wait so they could go underground together. Hofer moved a mantrip to the drift opening and waited. As Jeffrey Toler, Schoonover and Wilfong were leaving for the pit area, Wilfong told Chisolm to continue trying to contact the 2nd Left Parallel crew. Once in the pit, Wilfong went to the main mine fan to check the fan pressure recording gauge. He told investigators that he noticed nothing unusual at that time, but 18 that later in the day he recognized a fine line on the recording chart indicating an instantaneous spike in pressure. Wilfong also later noticed that the chart had not been replaced after one revolution, and had run over, so he replaced the chart. Wilfong, Jeffrey Toler, Schoonover and Hofer boarded the mantrip operated by Schoonover and proceeded underground. They did not take any gas detection instruments with them. No one could explain this oversight. Jeffrey Toler estimated that 10 to 15 minutes had elapsed from the time the explosion occurred until the time they started underground. The underground mine power system was not de-energized prior to them going underground. While waiting for Jeffrey Toler, John Boni walked back and forth between the Nos. 6 and 7 entries watching for both the 1st Left crew to approach in the primary intake escapeway and for Jeffrey Toler to come in on a mantrip in the track entry. Realizing that there may have been an explosion, John Boni made a third call to either Chisolm or Marsh requesting that Jeffrey Toler bring in gas detectors, because he did not have one. However, Jeffrey Toler and the others had already started underground. Once Ryan reached the intake entry at 37 Crosscut, No. 4 Belt he assisted Perry in donning his SCSR. Ryan stated that they did not have any trouble during the donning process. However, Ryan stated that Perry’s SCSR air bag collapsed after walking about one crosscut (80 to 90 feet). He removed the mouthpiece and continued walking outby. Perry reported that the bag did not inflate at first. Perry also stated that he pulled the mouthpiece plug out, but did not pull on the activation cord on the bottom of the SCSR. Since Perry was short-winded and breathing hard, the breathing bag on his SCSR began collapsing. At that point, he exhaled into the bag to inflate it, but it was uncomfortable. He kept removing and reinserting the mouthpiece because he felt that he was not getting enough air. He also stated that the goggles were uncomfortable and were pushing on his eyes, so he turned them down away from his eyes. Rowan stated that he did not don his SCSR until he reached the intake entry at 37 Crosscut, No. 4 Belt because they were in a panic, they were hoping to get to fresh air, and they needed to communicate with each other, which was difficult when wearing the unit. After reaching the intake at 37 Crosscut, No. 4 Belt it was still very dusty. Rowan decided to don his SCSR there. He did not experience any problems while donning the unit or while breathing with it. He acknowledged that he should have donned it immediately after the explosion occurred. After Owen Jones’ conversations with surface personnel, he walked to the intake entry where he joined his crew at 37 Crosscut, No. 4 Belt. Owen Jones instructed his crew members to immediately continue outby in the primary intake 19 escapeway. He stated that he was going to stay, but some of his crew pleaded with him to evacuate. He said that his brother, who was a miner on the 2nd Left Parallel crew, was inby and that he was going to see if he could do anything. Grall insisted that Owen Jones evacuate with them, saying he needed to think of himself, but Owen Jones refused. The crew members then proceeded outby without Owen Jones. Grall estimated that the crew was at 37 Crosscut, No. 4 Belt for about 2-3 minutes. The crew made its way outby 37 Crosscut, No. 4 Belt in the No. 7 primary intake escapeway entry. Grall and Avington advanced ahead of the others and, when Grall looked back, he could no longer see anyone behind them. As Rowan traveled outby, he assisted Keith, who was having difficulty breathing. It appeared that Keith’s SCSR was working because the bag was inflated. On several occasions, Rowan removed the mouthpiece on his own SCSR and had Keith take a few breaths from it, in case Keith’s was not functioning properly, but that did not seem to help. Perry had lost his hardhat and had removed his cap lamp and battery earlier due to a broken cap lamp lens, so Ryan helped him as they made their way outby. Ryan would move ahead a crosscut and wait for Keith and the others who were helping him. Ryan would then move forward another crosscut and wait. Ryan stated that as the crew traveled outby, the visibility improved. Hess stated that Avington and Tenney had handheld radios and made an unsuccessful attempt to contact the 2nd Left Parallel crew. Avington stated that he used his handheld radio while in the primary intake escapeway to tell Tenney to hasten their evacuation from the mine, and Tenney acknowledged him but was not sure what was said. Tenney stated that he had turned his handheld radio on while outside to check the battery, and then turned it off. He was planning to turn it back on when he arrived on the section, but never did. Tenney and Avington stated that they made no attempt to contact the 2nd Left Parallel crew. Hess and Tenney stated that although the crew was spread out to some extent during their travel out of the mine through the primary intake escapeway, they did stay within sight of each other. Once the crew members left to continue their evacuation from 37 Crosscut, No. 4 Belt, Owen Jones traveled back and forth from the primary intake escapeway to the track to check the conditions. The dust was starting to settle in the track 20 entry, but he could breathe better in the primary intake escapeway. Owen Jones’ detector was alarming. He cleaned the display and discovered it was in the failure mode3, but he did not turn it off. Owen Jones decided to travel inby in the No. 7 entry in an attempt to find the 2nd Left Parallel crew. However, after traveling about half a crosscut, he thought about his detector alarming and realized he could be overcome by CO. He retreated to the phone in the track entry that he had used earlier. He said that the air smelled like oil or coal burning. His detector read 0.2% methane. As the 1st Left crew was making their way out through the primary intake escapeway, Jeffrey Toler, Schoonover, Wilfong and Hofer entered the mine and traveled about two to three crosscuts, where they met Jamison, who was making his way out along the track entry. They asked if he was all right. When he replied that he was, Jeffrey Toler instructed him to continue to evacuate the mine. Jeffrey Toler, Wilfong, Hofer, and Schoonover continued their travel into the mine until they arrived at 1 Right switch where they met John Boni. John Boni stated that he had been waiting for about 10 minutes, and had not seen the 1st Left crew. Jeffrey Toler asked John Boni for a detector but John Boni did not have one. John Boni boarded the mantrip and continued into the mine with the others. John Boni told Jeffrey Toler that he thought there was an explosion. Jeffrey Toler said that there could not have been an explosion, and questioned how it could have happened. John Boni responded that he did not know how, but that he thought it had occurred. They continued into the mine and stopped near 25 Crosscut, No. 4 Belt. Wilfong used the phone to call the dispatcher at approximately 7:10 a.m. to see if he had heard from the 2nd Left Parallel crew. Chisolm responded that he had not heard from anyone. Owen Jones spoke to Wilfong and Chisolm on a phone near 37 Crosscut, No. 4 Belt. Wilfong thought that Owen Jones was attempting to make his way inby in an attempt to get to his brother. Wilfong told Owen Jones to get out of there before he was overcome by CO, and travel outby to their location. 3 Jones was probably carrying one of the following two types of detectors: Industrial Scientific Model LTX310 or Model ISC Model iTX. Neither of the two instruments will show the words “failure mode” on the display. The Model LTX310 instrument will show BATTERY FAIL” on the display when the instrument has insufficient charge to operate. The Industrial Scientific Model ISC Model iTX instrument will show “FAIL” on the display. The instrument was examined by MSHA. The manufacturer was contacted and stated that the most likely reason for “FAIL” showing on the display would be that an attempt was made to calibrate the instrument in high concentrations of CO. 21 The 1st Left crew, with the exception of Grall and Avington who were some distance ahead, were continuing their way out through the primary intake escapeway, and came together at 27 Crosscut, No. 4 Belt. As the miners assembled at 27 Crosscut, No. 4 Belt they discussed using a scoop that was parked near their location to evacuate the mine. As they formulated their plans, they heard a mantrip in the track entry. After Wilfong spoke to Chisolm on the surface, he returned to the mantrip and rode inby. Wamsley asked Ryan to go through the mandoor and flag down the mantrip. Ryan crawled halfway through the door at 27 Crosscut, No. 4 Belt and waved his light and shouted. Wilfong stopped and asked Ryan who was with him. Ryan responded that the whole crew was with him except for Avington and Grall. Ryan also indicated that Keith was not breathing well and had trouble walking, and that Perry had lost his hat and cap lamp and had trouble walking. Wilfong instructed Ryan to get everyone out to the track entry where there was fresh air, and said that he would take them outside. Ryan then went back through the door and told the others that a mantrip was there, and that everyone should travel to the track entry. As the crew boarded the mantrip some of the crew members were relating to Wilfong, Jeffrey Toler, Schoonover, Hofer and John Boni what had happened, and that a stopping was out at 32 Crosscut, No. 4 Belt. Grall said that when he and Avington reached 25 Crosscut, No. 4 Belt the air was clear, and they could see for a distance of about 500 to 600 feet. Grall and Avington continued their evacuation in the primary intake escapeway and approached 9 Crosscut, No. 4 Belt, where the maintenance shanty was located. There they heard a mantrip vehicle on the track entry, and traveled toward it. Grall noticed that the two large metal doors were open on the front of the maintenance shanty. He told Avington to check the track entry while he used the mine phone at that location. During his travel into the mine, Wilfong had not observed any signs of an explosion. After seeing the condition of the 1st Left crew, and hearing their description of what they had experienced, Wilfong realized that the situation was more serious than he had first thought. As the crew continued to board the mantrip, Wilfong asked John Boni and the others to take a head count. Wilfong then ran back to the phone, and made another call to the surface and spoke with Chisolm, who at that time was talking with Assistant Director of Safety and Employee Development John B. Stemple, Jr. Wilfong told Chisolm to alert both the federal and state agencies, and stated that mine rescue teams were needed immediately. 22 As Wilfong was talking on the phone with Chisolm, Grall spoke on the phone, and informed Wilfong that he and Avington were at the maintenance shanty at 9 Crosscut, No. 4 Belt. Wilfong told Grall that the crew was boarding the mantrip near 24 or 25 Crosscut, No. 4 Belt and would be evacuating, and that they would pick him and Avington up. Grall first told Wilfong that he would walk, but that Avington preferred to ride, but then informed Wilfong that they would both wait for the ride. Grall estimated the time to be about 7:15 a.m. Once Chisolm finished his conversation with Wilfong, he patched the land line phone into the mine phone, enabling Stemple to speak directly with Jeffrey Toler underground. It was approximately 7:15 a.m., and Jeffrey Toler advised Stemple that he was not sure what had happened. He said that they had found the 1st Left crew, and they were bringing them to the surface. Jeffrey Toler related that the 1st Left Crew stated that there were several intake stoppings out, and that there was smoke and dust in the air as they traveled along the primary intake escapeway. Stemple also learned from Jeffrey Toler that there had been no contact with the 2nd Left Parallel crew. He told Jeffrey Toler that he needed to re-establish ventilation as deep into the mine as he could in an attempt to prevent a short circuit of air to the 2nd Left Parallel section. Jeffrey Toler stated that he told Stemple to contact mine rescue teams. Jeffrey Toler told Wilfong to take the 1st Left crew outside while he, Schoonover and Owen Jones remained underground. Wilfong then asked Schoonover and Jeffrey Toler to get Owen Jones, and to assess the damage and determine how far they could advance, while Wilfong, Hofer and John Boni were taking the 1st Left crew to the surface. He also then mentioned that the stopping at 32 Crosscut, No. 4 Belt just inby their location was out. Hofer then operated the mantrip carrying the 1st Left Crew, John Boni and Wilfong toward the surface. They traveled outby to 9 Crosscut, No. 4 Belt where they picked up Grall and Avington. They continued toward the surface and arrived at the electric air lock doors along No. 1 Belt. Hofer asked Wilfong if they should switch from the electric doors to the manual doors. Wilfong said yes, and Hofer closed the manual doors but left the electric doors open. They continued out the track entry and arrived on the surface at approximately 7:30 a.m. Owen Jones traveled outby to meet Jeffrey Toler and Schoonover at the mine phone at 25 Crosscut, No. 4 Belt. Jeffrey Toler noticed that Owen Jones did not have a hard hat, and instructed Owen Jones to stay at the phone while he and Schoonover traveled inby to assess the damage. As Jeffrey Toler and Schoonover traveled inby on the track entry, they noticed the first stopping damage at 32 Crosscut, No. 4 Belt where the stopping was blown out from the intake toward the track entry. They continued to travel inby to about 42 or 43 Crosscut, No. 4 23 Belt and noticed that other stoppings were blown out toward the track entry as well. Jeffrey Toler and Schoonover decided to withdraw because they did not have any detectors with them, there was more than one stopping out and they did not know what conditions they would encounter. They traveled back to 41 Crosscut, No. 4 Belt where Jeffrey Toler stated a mine phone was located. Jeffrey Toler called outside and spoke to Marsh. Jeffrey Toler told Marsh to have Wilfong and Hofer bring in curtain, nails, boards, saws, all available detectors and a hardhat for Owen Jones. Jeffrey Toler and Schoonover then walked back to where Owen Jones was, and waited for Wilfong and Hofer to return with supplies. Marsh and miners Casey Short and George Brooks gathered the supplies, including two detectors, loaded them on a forklift, and took them into the pit area. Wilfong and Hofer arrived on the surface with the 1st Left crew. The 1st Left crew exited the mantrip and went to the bathhouse. Hofer informed Brooks that the batteries on the mantrip were low, and instructed him to get a fully charged mantrip for their return trip into the mine. Brooks obtained another mantrip and Marsh, Brooks and Short began loading the supplies. Wilfong told Hofer to stay in the pit while he went to the surface substation to de-energize the remaining power to the underground portion of the mine, including the AMS. Wilfong then signaled Hofer to pull the visual disconnect at the pit mouth, lock and tag out the underground mine power. Hofer then disengaged the knife blades on the pole in the pit and locked them out. The AMS system was equipped with a battery backup that maintained power to the system when there was a loss of mine power. The system would remain energized until it was manually disconnected. That was not done at this time, and the AMS remained energized until discovered by mine rescue teams during exploration. Hofer then went to the mine office and obtained handheld gas detectors and SCSRs. He returned to the pit area for the return trip underground with Wilfong. From the substation, Wilfong went to his office where he encountered Perry, who had dirt and debris in his eyes. Wilfong provided brief assistance to Perry. From there, he obtained telephones, hammers and other materials and returned to the pit area to assist Marsh, Hofer, Short, and Brooks load the remaining 24 supplies ordered by Jeffrey Toler. Once the supplies were loaded and Hofer and Wilfong were prepared to enter the mine, Hofer mentioned that they would stop at the maintenance shanty at 9 Crosscut, No. 4 Belt to obtain additional hand tools. Wilfong and Hofer entered the mine. Hofer turned one of the detectors on so that he could monitor the atmosphere as they traveled into the mine. The detector did not show any contaminants. They stopped at the maintenance shanty at 9 Crosscut, No. 4 Belt where Hofer obtained a sledgehammer, a slate bar and a pole axe. They proceeded inby and met Jeffrey Toler, Schoonover and Owen Jones. All five men went to 32 Crosscut, No. 4 Belt and installed a check curtain across the damaged stopping between the Nos. 6 and 7 entries on the intake side. Hofer stated that there was light air pressure toward the track entry. Hofer also stated that they noticed the return stoppings at 33 and 34 Crosscut, No. 4 Belt were damaged. They did not repair those controls. They boarded the mantrip and rode inby to 42 Crosscut, No. 4 Belt. Hofer walked inby on the track entry. About half way between 42 and 43 Crosscut, No. 4 Belt he heard his detector alarming. He looked down at the detector and saw that the alarm light was also flashing. He retreated to 42 Crosscut, No. 4 Belt and moved to the intake entry. He checked the detector and it was showing 40 to 50 ppm CO. He also indicated that the CO was dropping on the detector at that time. The CO would have been higher in the track entry. Jeffrey Toler’s detector also alarmed, but he could not recall any readings. Concerned about causing another explosion, the men decided to de-energize the mantrip by disconnecting the batteries, and to leave it at 42 Crosscut, No. 4 Belt. At 42 Crosscut, No. 4 Belt, Schoonover noticed a small amount of dust and smoke moving in the outby direction in the track entry. Wilfong gathered curtain, nails, spads, an axe and a detector and then proceeded into the intake entry. He installed a check curtain between the Nos. 6 and 7 entries at 42 Crosscut, No. 4 Belt. They unloaded the remaining supplies at 42 Crosscut, No. 4 Belt. Jeffrey Toler, Schoonover, Wilfong, Hofer and Owen Jones then started to repair stoppings between the Nos. 6 and 7 entries as they moved inby. Some were damaged while others were not. They were not sure how many damaged stoppings they repaired as they moved inby on foot. Jeffrey Toler stated that the stoppings were blown out from the intake to the track entry, and the amount of damage ranged from partial to complete. During the investigation, it was determined that they installed check curtains at damaged or completely blown out stoppings at the following locations: 32, 42, 43, 45, 46, 47, 49, 54, 56 and 57 crosscuts along No. 4 25 Belt. They also installed a check curtain at the damaged overcast at 51 crosscut along No. 4 Belt. Wilfong stated that they installed the check curtains starting from the outby end of the crosscut, working their way toward the inby end. This was done to remain in fresh air and force the CO inby and away from their work area. Wilfong noted that their detectors would alarm as they advanced, but as they installed a check curtain the air would clear and the alarms on the detectors would drop from high to low. They did not recall any actual readings. He also believed that the detectors would malfunction at times. When the check curtains had been installed up to 49 Crosscut, No. 4 Belt, visibility improved and Hofer noticed more damage in that area. Jeffrey Toler asked Hofer where the closest phone was. Hofer responded that there was a phone in the track entry hanging from a roof bolt. Jeffrey Toler stated he wanted to have a phone in the fresh air, so he went to the track entry to obtain the phone. Jeffrey Toler stated that when he was in the track entry he observed in excess of 700 ppm CO on the detector he had with him. He did not want to cut the phone line in the track entry leading to the 2nd Left Parallel section, so he went to the 1st Left Belt drive at 49 Crosscut, No. 4 Belt and cut the phone line there. Jeffrey Toler then worked the phone line over to the No. 7 intake entry. Jeffrey Toler moved two Emergency Medical Technician boxes and a stretcher located in the crosscut between the track and intake entry to the No. 7 intake entry. Crumrine stated that shortly after arriving at the mine he spoke on the mine pager phone to Jeffrey Toler who was underground. Crumrine recalled that Jeffrey Toler was either at 1st Left switch or near 42 or 43 Crosscut, No. 4 Belt. Jeffrey Toler told him that there had been an explosion or fire. He said that they did not have as much air volume or velocity as they should have. He said that there may be a stopping blown out behind him, meaning outby. He asked Crumrine to walk the intake entry into the mine and check ventilation as well as the ventilation into 2 Right. However, after completing the conversation with Jeffrey Toler, West Virginia Mine Inspector John Collins informed Crumrine that he was not permitted to enter the mine. It was between 8:15 a.m. and 8:30 a.m. Hofer connected the phone line to the phone, and called outside to see if the phone was working. Marsh answered the phone and spoke to Hofer. Marsh told Hofer “I notified him at that time that we had a (k) order4 and they were to evacuate the mines and not to proceed any further.” Hofer stated that Marsh 4 The 103(k) order was issued verbally over the telephone by Satterfield to Stemple at 8:32 a.m. According to Stemple, Satterfield said “No one is to enter the mine or do any work at the mine from 8:32 on.” 26 told him “he told me that there was a (k) order on the mines.” Hofer relayed the information to Owen Jones, who was standing beside him. Hofer asked Jeffrey Toler if he needed more curtain material. When Jeffrey Toler replied yes, Hofer went back to the mantrip at 42 Crosscut, No. 4 Belt to obtain more curtain material. Jeffrey Toler stated that they placed a check curtain in 51 Crosscut, No. 4 Belt where an overcast over the track entry was damaged. The group continued inby in the No. 7 entry, installing check curtains at the damaged stoppings between the Nos. 6 and 7 entries. At 42 Crosscut, No. 4 Belt, Hofer obtained a roll of curtain material, spads and nails and delivered the material to the area near 56 Crosscut, No. 4 Belt in the No. 7 entry where Jeffrey Toler, Schoonover, Wilfong and Owen Jones were waiting. He left the material there and returned through the intake entry back to 42 Crosscut, No. 4 Belt. He noticed that the visibility in the track entry was becoming poor due to smoke. He then began moving extra SCSRs, an extra detector, and two rolls of curtain into the No. 7 intake entry. One detector had apparently failed, so there was only one left. Wilfong stated that at some point Hofer had brought a couple of SCSRs to where Jeffrey Toler, Schoonover, Wilfong and Owen Jones were installing check curtains. However, they were not used. As Jeffrey Toler and the others moved inby installing check curtains, they noticed that the air velocity was not as strong as it should have been. In the area between 55 to 57 Crosscuts, No. 4 Belt, Jeffrey Toler started thinking that they may have missed one or more damaged controls outby. Jeffrey Toler and Wilfong told Owen Jones to take a roll of curtain and go outby with Hofer in the primary intake escapeway to check ventilation controls between the primary intake escapeway and track entry, and to install curtain wherever there was a damaged stopping without a curtain. Owen Jones proceeded to 42 Crosscut, No. 4 Belt where Hofer was moving the supplies from the mantrip area to the intake. Owen Jones and Hofer proceeded out the primary intake escapeway following the reflectors, and checked for any short circuit of air. As Hofer and Owen Jones walked out the primary intake escapeway, Jeffrey Toler, Wilfong and Schoonover continue to install check curtains. After installing a curtain in 57 Crosscut, No. 4 Belt they advanced inby between 57 and 58 Crosscut, No. 4 Belt where they observed the conditions and listened. The smoke was extremely dense, hanging down about three feet from the roof and swirling. Visibility was very poor and getting worse. The smoke was too dense to permit them to hang a curtain in 58 Crosscut, No. 4 Belt. Jeffrey Toler wanted 27 to go inby. Wilfong told him they should not do that and that the best thing they could do was to go outside and report what the conditions were and where they had been. They could hear noises inby, which sounded like something falling. They repeatedly shouted in the general direction of the noise in an attempt to make contact with the 2nd Left Parallel crew, but did not receive a response. There is no testimony or other information to indicate that this group had with them or used any of the non-permissible handheld radios in an attempt to contact the 2nd Left Parallel crew. Wilfong estimated that they waited about 15 to 20 minutes, and then discussed the situation. They believed that they had diverted all of the airflow toward the 2nd Left Parallel section, but thought that they should have more air at their location than they had. They thought a damaged stopping must have been missed. Wilfong stated that the three of them did not think that the 2 North Mains seals could have been blown out. Schoonover had taken some mine rescue training, and had some concerns. They decided that there was a potential of another explosion resulting from their actions, since they were forcing fresh air into areas where explosive gases might be present. Jeffrey Toler stated that the detector they had was still beeping, but he did not know what the readings were. Wilfong stated that the detector had reached its maximum reading and was in malfunction mode. Jeffrey Toler suggested that they should evacuate the mine and “let the professionals come in,” because they are “trained in this.” The others agreed. Jeffrey Toler, Wilfong and Schoonover then started outby in the primary intake escapeway. When they arrived at 49 Crosscut, No. 4 Belt Jeffrey Toler called outside on the phone that was moved into the intake entry earlier and spoke with WVMHS&T mine inspector Collins. At approximately 9:30 a.m., he told Collins that they had made it to 58 Crosscut, No. 4 Belt, that their detectors were burned up and that they had run out of air, and that the soot and smoke were so bad that they could not go into the track entry. When Owen Jones and Hofer arrived at 2 Right they found that the overcast over the No. 7 intake entry at 12 Crosscut, No. 4 Belt was damaged. Owen Jones stated that he noticed that a large amount of the intake air was short circuiting over the overcast to the main return. Owen Jones and Hofer picked up a piece of curtain material lying under concrete blocks from the damaged overcast, carried it to the overcast across the track entry and used it to install a check across the overcast over the track entry at 12 Crosscut, No. 4 Belt. This reduced the short circuit of air to the return entry at this location, thereby forcing more air inby. Jeffrey Toler, Wilfong and Schoonover continued outby to 42 Crosscut, No. 4 Belt where the mantrip that Wilfong and Hofer had used on their return trip underground was parked. Wilfong stated that he thought that they might use 28 the mantrip to evacuate the mine, but when he looked through the check curtain that had been installed earlier and saw more smoke and dust there than when they had parked the mantrip, he decided that they should walk out the primary intake escapeway to the surface. Jeffrey Toler had Wilfong and Schoonover travel the No. 9 entry, and Jeffrey Toler stayed in the No. 7 entry to check the stoppings. When they arrived around 12 Crosscut, No. 4 Belt at 2 Right, they saw Hofer and Owen Jones. Jeffrey Toler stated that he noticed some damage to a couple of overcasts in the 2 Right area. The walls of the overcasts were blown out. Owen Jones and Hofer told him they had already installed a check curtain on top of the overcast on the track entry at 12 Crosscut, No. 3 Belt. Owen Jones stated that Wilfong told him and Hofer that they should get out of there, since all the fresh air was now flowing inby, which could force methane over any fire that might exist and cause another explosion. Jeffrey Toler, Wilfong, Schoonover, Hofer and Owen Jones then walked out of the mine in the primary intake escapeway and arrived on the surface at about 10:35 a.m. At no time did Jeffrey Toler, Schoonover, Wilfong, Hofer or Owen Jones don an SCSR. Schoonover stated that no one donned an SCSR because he felt that there was no need to do so. Wilfong stated that he was saving his until he needed it. The 2nd Left Parallel Miners Twelve miners were on the 2nd Left mantrip, which was operated by Jesse Jones. Thomas P. Anderson, Alva M. Bennett, James Bennett, Jerry Groves, George Hamner Jr., Jones, David Lewis, Randal McCloy Jr., Martin Toler Jr., Fred Ware, Jackie Weaver and Marshall Winans entered the mine through the track entry about 6:00 a.m. Many of the following details concerning the events of the 2nd Left Parallel miners were obtained from physical evidence gathered during the investigation and from interviews of various mine rescue team members. Other details were provided by McCloy. He provided investigators with valuable information that only he would know. However, McCloy was still recovering from the effects of the accident at the time of his interview. As the crew made their way to the 2nd Left Parallel section, McCloy did not recall speaking to or seeing Helms. The crew arrived on the section and exited the mantrip. The crew was walking toward the face when the explosion occurred. The initial effects of the explosion were noise, pressure, wind and a haze. McCloy stated he was not knocked over. There was pressure but his ears 29 did not pop. McCloy stated that Martin Toler took charge and gathered everyone together after the explosion. McCloy indicated that no one tried to call out because all of the communication devices were damaged. He did not know if anyone tried to use the handheld radio communication system but he did not think it would have worked. McCloy stated that they boarded the mantrip operated by Martin Toler and started outby on the track entry in an attempt to escape. During their travel outby, they encountered an atmosphere filled with smoke. They continued outby until the mantrip hit debris on the track at 10 Crosscut, No. 6 Belt. They exited the mantrip.5 A mine rescue team later indicated that the mantrip appeared to have encountered an Omega block that had been blown into the center of the track between the rails. The mantrip appeared to have come in contact with the block and moved it in the outby direction. As the block moved forward, the soot deposited on the gravel between the track rails was disturbed. It also appeared that the mantrip was then moved inby away from the block about two to three feet. The crew donned their SCSRs, but McCloy could not remember exactly where or when. The top and bottom covers from twelve SCSRs were found at 11 Crosscut, No. 6 Belt in the No. 7 entry. According to McCloy, Martin Toler suggested that they don their SCSRs because they were in a small amount of smoke. McCloy stated that his SCSR worked fine, but that the SCSRs used by Groves, Anderson, Jesse Jones and Martin Toler did not work. McCloy indicated that he thought the other miners seemed to know how they worked, and indicated that they had been trained in their use numerous times. McCloy indicated that when they discovered that the SCSRs did not work, there was some yelling and there was a lot of controversy. When asked how he knew that the SCSRs did not work he stated that it was a “no-brainer,” since the miners had been trained extensively. He also indicated that the crew had to remove the mouthpieces from their SCSRs in order to communicate. At some point, Groves gave his SCSR to McCloy because Groves could not get it started. McCloy worked with the unit in an unsuccessful attempt to get the unit to work. 5 During their initial exploration, the mine rescue teams found the empty 2nd Left Parallel mantrip at 10 Crosscut. 30 McCloy stated the 2nd Left Parallel crew attempted to evacuate, and Martin Toler encouraged everyone to stay together. They tried several places to get out but everywhere they went it was smoky. However, McCloy said the visibility was never so poor that it was necessary to place their hands on each other or attach themselves in some manner like mine rescue teams. A mine rescue team found footprints in the soot on the mine floor indicating that the 2nd Left Parallel crew traveled to 11 Crosscut, No. 6 Belt in the No. 7 entry where they apparently donned their SCSRs. The team continued to follow the footprints outby in the No. 7 entry a crosscut or two until they could no longer see the footprints. Due to the smoke filled atmosphere limiting visibility, toxic gases, destroyed stoppings, and the debris on the track, the crew may have felt that all their options were exhausted, and there was no way out. They may have theorized that to try to travel on foot as a group in an attempt to escape would be extremely difficult. Although all of the information that was available to the 2nd Left Parallel crew as they were considering their options is not known, it is possible to consider what information they may have had. They knew that the 1st Left crew had entered the mine after them. They knew that the mine had been idle the previous shift. They knew the mine was not very gassy. Although they knew the results of the preshift examination for the 2nd Left Parallel section, they may not have known the results for the preshift examination for the 1st Left section. History has indicated that most explosions are the results of the actions of men or machinery. Based on these considerations, it is possible they believed that an explosion occurred in the 1st Left section as the crew entered the section or just shortly thereafter. It would not have been likely that they would have considered an explosion originating from behind the sealed area. Although explosions had occurred in the past damaging seals, there was no history of an explosion of this magnitude or level of destruction. There was no obvious ignition source present, such as spontaneous combustion or an active fire. If they considered that the explosion had originated in the 1st Left section, then the conditions observed on the 2nd Left Parallel section would not be as destructive as what they may have expected to encounter in the mains as they attempted to escape. They may have considered the distance that they would have to travel and speculated that it would be impossible for them to accomplish it safely. Martin Toler suggested that they go back to the section. Everyone agreed to go back to the section. As they traveled back toward the section in the belt line, they initially could not see very well. 31 They decided to build a barricade. McCloy recalled Martin Toler directing the installation of the barricade curtains. Toler, Anderson and McCloy assisted in the installation of the curtains. He thought that there could have been additional miners helping but could not recall who. They tried to make them “leak-free.” They decided to use curtain material from the face area since some of the crew indicated their SCSRs were not working. Although there was concrete block nearby, they felt that using block would take more work and “it would just not work.” McCloy recalled that visibility was good during installation of the curtains. He said that he removed his SCSR during the installation process. Once behind the barricade it took several hours before the miners calmed down. They turned all their cap lamps off except for one, as Martin Toler suggested. There was conversation between them. The area they were in was large, and they would have to shout to each other at times. McCloy indicated that the crew thought they would be rescued. They took turns using a sledgehammer to bang on a roof bolt. McCloy said that as each miner took his turn, he would take off his SCSR because he would get exhausted. McCloy said that this was the only time he removed his SCSR. McCloy thought that rescuers would bring the machine that locates people to the mine. According to McCloy, the crew thought that they would hear shots on the surface, rescuers would drill a hole in the right spot, and they would be taken out. They thought that they would be rescued, and discussed how long it would take. However, as time passed it did not look good. They were waiting for the borehole but felt that the rescuers must not have had the right equipment. McCloy indicated that about an hour and a half after entering the barricade, Martin Toler and Anderson exited it. They walked to the power center across from the tailpiece. He thought that they did not have SCSRs with them. He believed that they were looking to see if the air was clearing and to see how far they could get. They made it to the power center but then returned. When they re-entered the barricade they were coughing and gagging, and were exhausted. McCloy said that Toler and Anderson said that there was too much smoke and that it was hard to breathe. While in the barricade, McCloy removed his goggles. McCloy shared his SCSR with Groves while in the barricade. He was aggravated that Groves’ SCSR would not work, so he again made an unsuccessful attempt to get it to function. McCloy said that his and other miners’ SCSRs were depleted, but he could not recall whose. During the time the miners spent in the barricade, some of them wrote personal notes to their family members. According to a note written by James Bennett at 11:40 a.m., they had air but the smoke was bad. At 2:45 p.m., Weaver wrote that 32 the fumes were getting terrible, but everyone was still partially ok. James Bennett wrote at 3:07 p.m. that the air was bad and that he did not know how much longer they could last. At 4:22 p.m., he wrote that time was running out and at 4:25 p.m., he wrote “we not heard anything from the outside.” The quality of the writing in each segment of the notes deteriorated with time. McCloy did not see Martin Toler make any gas checks, did not hear any alarm from a gas detector, and did not think that Toler had a detector. McCloy indicated that it was a long time before any of the miners went to sleep, or appeared to be sleeping. However, they did not all succumb at one time. McCloy did not know if all the others fell asleep before him because they were not all together. Some of his fellow miners were some distance away and it was difficult to see them. 33 NOTIFICATION AND SAMPLING Chisolm telephoned Stemple at home by 7:00 a.m., and patched him through to Wilfong, and then to Jeffrey Toler in the mine at 7:15 a.m. Around 7:20 a.m., Stemple contacted the General Manager of ICG’s Buckhannon Division, Charles Dunbar, and told him that something had occurred at the mine but that he did not know exactly what. Stemple said he would call back once he got more information. At approximately 7:30 a.m., ICG Purchasing Director Jerry Waters told ICG’s Manager of Safety for West Virginia and Maryland, Harrison Tyrone (Ty) Coleman, that something had happened at the mine, and that Coleman should contact Stemple to obtain more details. Ty Coleman was Stemple’s supervisor. Ty Coleman left home between 7:35 a.m. and 7:40 a.m. and drove to the mine. While driving, he called Stemple and told Stemple to activate the mine rescue teams and put them on standby. Stemple replied that he had already contacted them. Ty Coleman also contacted Dunbar, but Dunbar was already aware of the event and was either at the mine or on his way. Ty Coleman called ICG’s Production Coordinator for the Buckhannon Division, Raymond Coleman, to inform him of the event at the mine. Ty Coleman estimated it took approximately 20 minutes to drive to the mine. At approximately 7:45 a.m., Chisolm telephoned Crumrine to notify him of the events at the mine, but was unsuccessful and left a message. When Crumrine returned his call, Chisolm told Crumrine that there had been an explosion in the mine. Crumrine left his home and drove to the mine. Stemple first attempted to contact personnel at the WVMHS&T Fairmont, West Virginia office around 7:40 a.m. He was unsuccessful, and left a message on the answering machine. At around 7:46 a.m., Stemple attempted to contact Collins at home. He was unsuccessful, and left a message on his answering machine. Shortly thereafter, Collins returned Stemple’s phone call, learned of the event, notified his supervisors and drove to the mine. Stemple called MSHA’s Bridgeport, West Virginia Field Office Supervisor Kenneth Tenney at home around 7:50 a.m. He was unsuccessful, and left a message on his answering machine. Tenney was not at home and did not learn of the accident until later. Dunbar and Crumrine arrived at the mine at approximately 8:00 a.m. Dunbar went to the dispatcher’s office to talk to Chisholm. Crumrine went to his office and soon received a briefing from Chisolm. Crumrine assembled his mine gear and overheard Chisolm in a nearby office talking to Jeffrey Toler on the mine phone. Crumrine interrupted that conversation and talked to Toler. Jeffrey 34 Toler told Crumrine that they had some trouble in the mine and that an explosion, a fire, or something else had happened. Jeffrey Toler said that they did not have the quantity of air they should have, and asked Crumrine to walk the intake into the mine to check the ventilation system. Ty Coleman arrived at the mine sometime after 8:00 a.m. and traveled to the superintendent’s office to be briefed, and to be near a mine phone. At 8:04 a.m., Stemple tried to contact Jeffery Rice, a member of the Barbour County Mine Rescue Team. Stemple then tried to contact personnel at the MSHA District 3 office in Morgantown, West Virginia at 8:05 a.m., but the office was closed due to the federal holiday. The telephone answering machine at the MSHA District office provided Stemple with a list of names and telephone numbers to contact in case of a mine accident or emergency. He proceeded to call them. Stemple unsuccessfully tried to reach MSHA District 3 Assistant District Managers Carlos Mosley and William Ponceroff, and District Manager Kevin Stricklin, and left messages on their cell phones concerning the accident. Stricklin’s cell phone registered Stemple’s message at 8:13 a.m. Collins arrived at the mine at approximately 8:15 a.m. He was the first of many WVMHS&T representatives who would arrive at the site throughout the day. Collins went into the mine office and was briefed by Dunbar. Collins then saw Crumrine, who was exiting his office with the intent to enter the mine to check the ventilation system as Toler had instructed. Collins asked Crumrine to wait until more information could be obtained, and asked if anyone was monitoring the mine’s return air. At about 8:28 a.m., Stemple called MSHA’s Bridgeport, West Virginia Field Office Supervisor James Satterfield and informed him of the events at the mine. After being briefed, Satterfield notified Stemple that he was issuing an order under section 103(k) of the Federal Mine Safety and Health Act. According to Stemple, Satterfield told him that nobody was to enter the mine or do any work at the mine after 8:32 a.m. Satterfield then attempted to notify MSHA District 3 Staff Assistant Ron Wyatt, and left a message about the accident. At 8:30 a.m., Collins issued an order to the mine operator to preserve the scene of the accident. He explained the order and its requirements to Crumrine. Dunbar notified ICG’s Senior Vice-President Sam Kitts that there may have been an explosion in the mine, and that 18 miners were unaccounted for. He also reported to Sam Kitts that the 1st Left crew had managed to get out, but that others had gone inside to investigate. Sam Kitts then telephoned and left a message for ICG President and CEO Ben Hatfield. Sam Kitts also called ICG Vice-President of Mining Services Gene Kitts, and asked him to notify other senior management officials at ICG. 35 At 8:35 a.m., Stemple contacted the mine, talked to someone whom he believed was either Chisolm or Marsh, and told him that he had notified the appropriate federal and state agencies. Stemple also notified him about MSHA’s issuance of a 103(k) order. At some point Stemple also contacted the pastor of the Sago Baptist Church and obtained his permission to use the church as an assembly area for families, news media and mine rescue teams. Stemple made contact with a mine rescue team at approximately 8:37 a.m. by speaking with Chris Height of the Barbour County Mine Rescue (BCMR) Team. Stemple asked Height to have the team assembled. The Barbour County team members assembled at their Volga, West Virginia station, prepared their equipment and headed for the mine at approximately 10:30 a.m. They were then to reassemble at the Sago Baptist Church across the road from the mine and wait for further instructions. As the day progressed, various mine rescue teams and personnel responded to the mine. A list of mine rescue personnel and teams responding is contained in Appendix C. Around 8:40 a.m. Satterfield contacted Mosley on his cell phone and informed him of events at the mine. Satterfield informed Mosley that he had contacted MSHA inspectors Ron Postalwait and Argil P. Vanover and was meeting them at the Bridgeport field office to travel to the mine. Collins asked contract Foreman James Scott and another foreman to monitor the mine’s return air. At 8:40 a.m., Scott and the foreman acquired air quality and quantity measurements in the No. 1 Drift Opening. The air quality was 47 ppm CO, 0.0% methane and 20.4% oxygen. They determined the air quantity to be 93,204 cfm. At 9:10 a.m., Scott and WVMHS&T mine inspector Jeff Bennett obtained more air quality measurements in the No. 1 Drift Opening. Each took their own readings with their own instrument. Bennett’s readings were 23 ppm CO, 0.0% methane, and 20.3% oxygen and Scott’s instrument indicated 50 ppm CO, 0.0% methane, and 20.6% oxygen. Around 8:43 a.m., Wyatt telephoned Satterfield concerning the message left earlier on his answering machine. Satterfield then briefed Wyatt on the events at the mine. Wyatt then contacted Ponceroff at his residence and informed him. At around 9:03 a.m., Wyatt contacted Stemple and obtained a briefing on the situation at the mine. During this conversation Wyatt asked Stemple if mine rescue teams had been contacted and Stemple informed him that they had been notified. At 9:05 a.m., Wyatt contacted Ponceroff and Mosley to provide them with an update on the situation. Ponceroff then asked Wyatt to meet him at the District office so they could then travel to the mine. Gene Kitts telephoned ICG Corporate Director of Health and Safety Timothy Martin at home, told him the information he had about what had occurred at the 36 mine, and informed him that two crews were unaccounted for. At approximately 9:10 a.m., Martin contacted Bob Gardner, Vice-President and General Manager of Viper Coal in Williamsville, Illinois, to request that he activate the Viper mine rescue team. Martin then made several telephone calls to arrange transportation for the Viper mine rescue team from Illinois to the mine. He estimated that the team would arrive in Charleston at about 1:30 p.m. Prior to Sam Kitts leaving his residence at approximately 9:15 a.m., he received a call from Stemple informing him of the event at the mine. Sam Kitts informed Stemple that he had already been notified by Dunbar, and that he was on his way to the mine. At about 10:00 a.m., WVMHS&T mine inspector Brian Mills contacted Joe Prevola of the Tri-State mine rescue team and asked that he have his team members respond to the mine accident. Prevola telephoned the team members and instructed them to gather at their office in Kingwood, West Virginia, to assemble their equipment. Once the team was gathered and assembled, they proceeded to the mine. Shortly after 10:00 a.m., Mills contacted Spike Bane of CONSOL Energy, Inc. to inform CONSOL personnel of the accident at the mine and the potential need for CONSOL’s mine rescue teams to assist in a mine rescue. Mills later had further telephone conversations with Bane on the events unfolding at the mine. Mills stated that he provided Bane’s contact information to someone at ICG with instructions to contact Bane to get CONSOL’s mine rescue team personnel on site. Wyatt contacted MSHA Headquarters personnel to notify them of the accident at the mine around 10:00 a.m. Personnel there contacted the Chief, Mine Emergency Operations (MEO), Dr. Jeffrey Kravitz, about the accident around 10:15 a.m.; so that he could mobilize the agency’s other resources, including MSHA’s Mine Emergency Unit (MEU) members. Shortly before 10:30 a.m., Satterfield and Vanover arrived at the mine and met with Jeffrey Toler. Postalwait arrived at the mine at about the same time. Satterfield instructed Postalwait and Vanover to monitor the pit area, including the return air exiting the No. 1 Drift Opening. Postalwait and Vanover made their first air quality measurement at 10:47 a.m., which indicated 500 ppm CO, 0.8% methane, and 19.8% oxygen.6 6 Air quality measurements were made using MSA Solaris multi-gas handheld detectors, which can detect a maximum carbon monoxide level of 500 ppm. At carbon monoxide concentrations exceeding 500 ppm, the Solaris instrument display screen will display 500 ppm. 37 Kravitz notified the Chiefs of the Ventilation and Physical and Toxic Agents Divisions of MSHA’s Technical Support about the mine accident between 10:45 and 10:50 a.m. They proceeded to notify their respective personnel in order to mobilize each Division’s available mine emergency capabilities. Those groups’ ability to respond was restricted since personnel and materials from both groups were at the West Elk Mine in Colorado. Most of their mine emergency equipment and manpower had been sent there to respond to a mine fire. In addition, some MEU equipment had also been deployed to the West Elk Mine. Kravitz contacted Stricklin at 10:59 a.m. at his residence to request permission to use District 3 mine rescue personnel. This was how Stricklin first became aware of the accident. Stricklin made several telephone calls and traveled to the mine. Bennett, Postalwait and Vanover made additional measurements of the return air exiting the No. 1 Drift Opening until about noon, the results of which are shown in Table 3. Date Time 1-02-06 1-02-06 1-02-06 1-02-06 1-02-06 1-02-06 1-02-06 1-02-06 1-02-06 1-02-06 11:02 a.m. 11:02 a.m. 11:15 a.m. 11:28 a.m. 11:28 a.m. 11:30 a.m. 11:37 a.m. 11:37 a.m. 11:45 a.m. 12:00 p.m. Table 3 - Air Quality Measurements Collector Instrument Carbon Methane Monoxide (CH4) (CO) (%) (ppm) 7 Bennett Explorer 4 472 1.0 Vanover Solaris 500 1.1 Vanover Solaris 500 1.1 Bennett Explorer 4 472 0.8 Vanover Solaris 500 0.9 Vanover Solaris 500 0.7 Bennett Explorer 4 472 0.7 Vanover Solaris 500 0.8 Vanover Solaris 500 0.7 Vanover Solaris 500 0.6 Oxygen (O2) (%) 19.0 19.4 19.4 19.3 19.7 19.8 19.5 19.7 19.8 19.8 At about 11:30 a.m., Barbour County mine rescue team members assembled at the Sago Baptist Church to wait for further instructions. However, since the miners’ family members were using the church to wait for news, the team 7 Bennett’s air quality measurements were made using a CSE Corporation Explorer 4 handheld detector. The maximum CO which it is able to detect is 500 ppm. At CO concentrations exceeding 500 ppm, the Explorer 4 instrument display screen will display 500 ppm. However, if the CO sensor is weak or the instrument is out of calibration, a lower value may be displayed. 38 relocated to the mine. Team members set up their equipment at the mine and were ready to don their apparatuses by about 12:30 p.m. Prior to Sam Kitts arriving at the mine, Hatfield telephoned him and received an update on the mine accident. Sam Kitts arrived at the mine site around 11:45 a.m., and met with mine management personnel to assess the situation. At 12:00 p.m., Ponceroff and Wyatt arrived at the mine site. Mine management briefed them, and they traveled into the mine pit and met with Postalwait and Vanover, who were taking air quality measurements. Finding a CO reading of 500 ppm, Ponceroff and Wyatt decided to withdraw everyone from the pit area. Sampling in the No. 1 Drift Opening was conducted every 15 minutes and personnel entering the pit area continuously monitored the air quality. At 12:17 p.m., Bennett used an Industrial Scientific 270 Multi-gas detector to measure the air quality in the No. 1 Drift Opening. The measurement indicated CO in excess of 1,999 ppm.8 Martin arrived at the mine at approximately 12:15 p.m. and was briefed by mine personnel. He then assisted with ongoing activities until assuming the responsibility of ensuring that mine rescue teams were available and properly staged. Carbon monoxide continued to be a concern, not only in the pit area, but in the surface buildings. At 12:20 p.m., Vanover issued an imminent danger order under section 107(a) of the Mine Act because of the extremely high CO levels detected in the No. 1 Drift Opening. The order required the withdrawal of all non-essential personnel from the pit and the surface buildings. Barbour County mine rescue team members were mobilized to conduct the sampling of the No. 1 Drift Opening. At 12:30 p.m., Kravitz notified MSHA Mine Emergency Operations Group personnel to prepare the seismic system for possible deployment. Two technicians arrived in Pittsburgh, Pennsylvania at 2:45 p.m. to prepare the unit. At 1:00 p.m., Sam Kitts went to the Sago Baptist Church and provided a briefing to the miners’ families. At about the same time, elevated CO concentrations were measured outside and inside surface buildings. Those CO levels were 330 ppm and 130 ppm, respectively. MSHA personnel directed that all office and nonessential personnel leave the mine site. Shortly before the evacuation, Stemple arrived at the mine site. Stemple obtained a briefing from Crumrine and other 8 The Industrial Scientific 270 multi-gas handheld detector detects maximum carbon monoxide levels of 1999 ppm. 39 mine personnel and was informed that levels of CO were greater than 2,300 ppm in the pit mouth. Stemple noticed that Crumrine’s handheld gas detector was in alarm, measuring 61 ppm CO. Stemple met with Ponceroff, Satterfield, and other MSHA and WVMHS&T officials, and assisted with evacuating nonessential personnel from the mine site, to either the Sago Baptist Church, or to the training room at the ICG cleaning plant located about a mile from the mine site. Around this time, Ty Coleman established the command center in Toler’s office and started to assign personnel to set up the room for a command center, monitor the mine entrance, guard the mine site, and provide a workspace for engineering. Toler’s office would serve as the command center for the rest of the rescue and recovery effort. Shortly after 1:00 p.m., a formal plan was developed by the mine operator and approved by MSHA and WVMHS&T personnel to monitor mine gases in the pit mouth. The plan required two mine rescue team members to approach the mine entrances wearing full apparatus, and to monitor the gases exiting the mine. The plan required the results to be reported to the command center. The plan also required that two mine rescue team members wearing full apparatus stand at the edge of the pit to serve as backup to the personnel in the pit. At 1:05 p.m., BCMR personnel started to take air quality measurements in the mine pit. Two rescue team members entered the pit and two watched from the top of the pit as emergency backup each time an air quality measurement was made. These measurements were made in Nos. 1– 4 Drift Openings. The results from the log are shown in Table 4. 40 Table 4 - Air Quality Measurement by BCMR Date Time 1-02-06 1:05 p.m. 1-02-06 1:07 p.m. 1-02-06 1:09 p.m. 1-02-06 1:11 p.m. Collector Instrument9 BCMR BCMR BCMR BCMR iTX iTX iTX iTX Drift No. 1 2 3 4 CO10 (ppm) +2000 +2000 4400 1700 CH4 (%) 0.0 0.0 0.0 0.0 O2 (%) 20.9 20.7 20.7 20.7 BCMR personnel continued to obtain air quality measurements in the mine pit. BCMR air quality measurements taken between 1:25 p.m. on January 2 and 11:00 a.m. on January 3 are shown in Appendix D. The Tri-State mine rescue team arrived at the mine site at approximately 1:30 p.m. At the same time, non-essential mine personnel were being allowed back on the mine site after being evacuated because of high CO levels. The CO levels in the mine office decreased. Tri-State member Chris Lilly noticed that some CONSOL teams were already on the property. The Command Center told TriState team members that CO levels measured at the No. 1 Drift Opening were too high for them to enter the mine. The WVMHS&T mine rescue trailer containing their mine rescue gear arrived. CONSOL team members were arriving at the mine, and CONSOL sent safety department personnel to assist with coordinating and directing their teams’ activities. CONSOL also sent a gas chromatograph and personnel to operate it to help in monitoring the mine’s atmosphere for gases. 9 These air quality measurements were made using an Industrial Scientific iTX multi-gas handheld detector. The iTX can be equipped with either of 2 types of CO sensors. The 4 series sensor has a range of 0 - 999 ppm CO. The maximum indicated concentration on the instrument display is 999 ppm CO. The display indicates “OR” if the maximum range is exceeded. This is the standard sensor. The 7 series sensor, used at Sago, has a range of 0 - 9,999 ppm CO. The maximum indicated concentration on the instrument display is 9,999 ppm CO. The display indicates “OR” if the maximum range is exceeded. 10 BCMR air quality measurements were documented by MSHA. The 1:05 p.m., 1:07 p.m., 1:09 p.m. and 1:11 p.m. Carbon Monoxide measurements were found to be inaccurate after they were entered into the log. The CO peak readings stored in the iTX multi-gas handheld detector memory were checked, and actually indicated a maximum CO value of 1,386 ppm, not the CO values of +2,000 ppm, +2,000 ppm and 4,400 and 1,700 ppm documented in the log. The log was not corrected. Also, the maximum CO value for the sensor installed in this instrument might have been exceeded. Following this series of readings, MSHA personnel provided training to BCMR personnel on use of the iTX multi-gas handheld detector. 41 Stricklin arrived at the mine site between 1:45 p.m. and 2:00 p.m. He obtained a briefing from Ponceroff, Wyatt and Satterfield on the status of the missing miners, the miners who had escaped, mine management’s rescue attempt and the condition of the mine. Personnel from MSHA’s Ventilation and Physical and Toxic Agents Divisions organized and readied for transport a set of infrared and electrochemical gas analyzers, several thousand feet of 3/8 inch PVC tubing, vacuum pumps, four handheld permissible radios, a gas chromatograph, and the associated computers needed to operate the gas chromatograph and analyze the gas results. They left Pittsburgh, Pennsylvania with this equipment at around 2:00 p.m. CONSOL’s gas chromatograph was placed in the WVMHS&T’s mine rescue trailer and readied for operation. CONSOL technicians calibrated the instrument and had it operational by 3:00 p.m. to analyze air samples collected in the mine drift openings. The gas chromatograph provided the capability to monitor additional gases and allowed a means to verify the readings for the CO, methane and oxygen being obtained from the handheld instruments. The Viper mine rescue team arrived at the Charleston, West Virginia airport around 1:40 p.m. and was escorted to the mine by West Virginia state police, arriving around 3:30 p.m. At about this time, personnel designated by the Command Center briefed the mine rescue team captains concerning the accident. At about 3:30 p.m., construction of a road to provide access to the 2nd Left Parallel borehole drill site was begun. The construction and site preparation took about 3 hours to complete. Air quality measurements from the drift openings indicated a downward trend in the levels of dangerous gases. It was after 4:00 p.m. when the mine operator submitted requests to send mine rescue personnel into the mine. However, MSHA and WVMHS&T denied these requests because the levels of CO exiting the mine were still too high, reflecting a substantial risk of fire and the possibility of another explosion. The mine rescue teams were briefed at 4:15 p.m. 42 The air quality readings continued trending downward. Figure 3 illustrates the results of CO measurements obtained in the No. 1 Drift Opening. While they were still at dangerous levels, it was determined that they were low enough to allow rescue efforts to commence. At 4:55 p.m., the mine operator submitted a plan for the start of exploration which was approved by MSHA and WVMHS&T. The plan called for Tri-State Team A to enter the No. 5 intake entry and to explore the first 1,000 feet. Tri-State Team B would serve as their backup in the event Team A personnel experienced any type of difficulty. At 5:12 p.m., the mine operator submitted a new plan switching the Tri-State teams to the CONSOL teams, since CONSOL’s teams had more experience in mine rescue than any other team present. Sago Mine Main Return Carbon Monoxide Explosion Occurs at 6:26 AM on Day 1 2500 2000 CO, ppm Mine Rescue Teams Enter the Mine at 5:25 PM on Day 1 3000 1500 1000 500 CO From Explosion Stabilization Begins to Occur 12:00 AM 12:00 PM 12:00 AM 12:00 PM 12:00 AM 12:00 PM 12:00 AM 0 Day 1 Day 1 Day 2 Day 2 Day 3 Day 3 Day 4 Date and Time Figure 3 - CO Measurements at the No. 1 Drift Opening MSHA’s Ventilation and Physical and Toxic Agents personnel arrived at the mine site at approximately 5:15 p.m. and were briefed by Stricklin. They began to set up atmospheric sampling equipment, consisting of infrared and electrochemical instantaneous monitoring equipment, a gas chromatograph, and all associated equipment. During the set up process, electrical power had to be provided. A sampling line had to be extended to the No. 1 Drift Opening since a previously installed line was plugged. Four handheld permissible radios were distributed to MSHA’s MEU personnel. At 5:25 p.m., the CONSOL Robinson Run A mine rescue team entered the mine through the fan house and proceeded inby exploring the mine. 43 RESCUE AND RECOVERY OPERATIONS Mine Rescue Protocol A basic mine rescue protocol has evolved over the years based on rescue efforts made during previous mine disasters. However, each mine disaster is unique, presenting a number of situations requiring difficult decisions. Most operations begin with establishment of a command center, which is headed by the mine operator. State and federal officials and sometimes miners’ representatives are generally part of the command center. MSHA issues a section 103(k) order which requires a written plan to be proposed by the mine operator. It must be approved by MSHA and agreed to by the parties in the command center before it can be implemented. All of the decisions concerning the rescue operation, including mine rescue team movement, the areas of exploration and all related work are made in this manner. All teams are briefed before entering the mine and debriefed upon exiting. A mine rescue team establishes a fresh air base (FAB) which includes a hardwired communications system running to the surface command center. The FAB is the communication hub between the exploring teams and the command center. Exploration begins with one rescue team generally composed of five members. Each member is equipped with a breathing apparatus weighing approximately thirty-five pounds and consisting of a full-face mask and a supply of oxygen. A back up team is stationed at the FAB. They are ready to assist the exploring team if needed. A third team is on the surface, ready to provide support to the teams underground. Communication from the exploring team to the FAB is made by handheld permissible radios or by using a hard-wired communication system connected directly to the FAB. Rescue teams can typically explore about 1,000 feet from the FAB. After an area has been explored, ventilated and made safe for travel, the FAB is advanced. This continues until the operation is completed. It is critical that ventilation not be changed in an area that has not been explored. This may allow explosive gases to come in contact with an ignition source, such as a fire, causing a subsequent explosion. If ventilation has been severely disrupted such that it is no longer possible to establish a FAB that is in fresh air, it may become necessary for the mine rescue team to begin to airlock as they advance. The mine rescue team builds temporary ventilation controls across all entries just inby the existing FAB. They then completely explore the next 1,000 feet in each entry. They build another set of temporary ventilation controls at that location. They repair ventilation controls between the two sets of airlocks. They remove the first set of airlocks and re-ventilate the area, relocate the FAB and start the process again. Airlocking efforts are labor and time intensive. 44 MSHA deploys MEU personnel to mine disasters. The purpose of the MEU is to provide technical and expert assistance during emergency operations. MEU members have extensive experience in mine rescue and recovery operations throughout the nation. MEU is self-supported and provides an assortment of specialized equipment such as permissible radios and handheld air quality detectors. During any mine related exploration, MEU personnel are assigned to the exploration team and to the back-up teams. Their presence has proven invaluable to mine rescue operations. Mine Gases Methane and coal dust explosions have occurred in underground coal mines. These explosions can develop overpressures of 20 psi or more. MSHA investigated numerous methane and/or dust explosions and, with 2 possible exceptions,11 had not observed evidence of explosion overpressures exceeding 20 psi. The pressures generated by an explosion are well in excess of the 2 to 4 psi that ventilation controls, such as stoppings, are able to withstand. Investigators have found that damage to ventilation controls after an explosion is quite common. This damage usually causes a short circuit in the ventilation system which may allow methane to accumulate. The ignition temperatures of coal, wood, and other combustible materials found in a coal mine are less than the temperature of the explosion flame. However, the speed of the flame from an explosion can propagate in excess of 1,000 feet per second.12 At this speed, the explosion flame contacts each point in the explosion zone for only a brief period of time, typically less than 100 milliseconds. This 11 An explosion occurred in the Production Shaft of Consolidation Coal Company’s Blacksville No. 1 Mine on March 19, 1992. A cap was placed on the Production Shaft. An explosive methane/air mixture began to accumulate in the shaft. It was ignited by arc welding operations occurring on top of the cap. The subsequent explosion generated overpressures at the top of the shaft of approximately 1000 psig. The unusual circumstances resulted in a detonation of the fuel. An explosion occurred in a sealed area of U.S. Steel Mining Company’s Oak Grove Mine on July 9, 1997. Although a lightning strike of +145,200 amperes was determined to be the cause of the ignition, the path of that lightning strike into the sealed area was not defined. The sealed area had a total of 38 seals. Access into the sealed area after the explosion was not possible. Seven cementitious seals may have been damaged by the forces. Four seals had minor damage which may have affected their strength. Three seals were partially or completely displaced. The compressive strength of two of these three seals were found to be below the minimum acceptable limit of 200 psi. The information that the third seal had a compressive strength in excess of 200 psi led the investigators to indicate that the explosions forces required to damage the seal was in excess of 20 psi. Subsequent opinions by MSHA determined that the number of samples subjected to compressive strength testing was inadequate to fully support this conclusion. 12 The Explosion Hazard in Mining, U.S. Department of Labor, Mine Safety and Health Administration, Informational Report 1119 (1981), John Nagy, Page 61. 45 length of time is generally too short to directly involve all these materials in a massive fire. The flame of the explosion also includes suspended coal dust that is heated to above the ignition temperature of various combustible materials. As the flame of the explosion slows and terminates, this coal dust drops out of suspension and accumulates on available surfaces. These surfaces may be the mine entry, crib blocks, roof support posts, or other combustible materials. It is possible that materials with a low ignition temperature may begin to smolder and eventually ignite under these conditions. When an explosion occurs, large volumes of toxic and flammable gases are produced due to the incomplete combustion of these fuels. These gases include carbon monoxide, carbon dioxide, hydrogen, acetylene, and ethylene. The concentration of these gases can vary depending on the concentration of the fuel involved in the explosion. For example, in a coal dust explosion, CO concentrations can be 1,000 ppm when the initial coal dust concentration is 0.1 ounce per cubic feet. 13 The data also indicated that if coal dust was the sole fuel source involved in the explosion at a concentration of one ounce per cubic foot, CO could be formed to as high as 46,000 ppm.14 In the case of methane explosions, post-explosion CO concentrations are approximately 500 ppm when a 9% methane/air mixture is ignited. CO levels can reach 80,000 ppm when the initial concentration of ignited methane increases to 12%.15 Other gases are produced during explosions which are asphyxiates. These gases may not be toxic or flammable but can displace the oxygen necessary to sustain life. When fires first begin, they produce barely detectable levels of CO. As the fire begins to grow and intensify, the level of CO production also begins to grow. The rate of growth of the fire depends on a number of factors, including the fuel and the amount of available oxygen. Carbon monoxide levels can reach well in excess of 10,000 ppm during a fire. 16 The temperature of the flame of a fire exceeds 1,500 degrees F. The ignition temperature of methane is 1,000 degrees F.17 13 The Explosion Hazard in Mining, U.S. Department of Labor, Mine Safety and Health Administration, Informational Report 1119 (1981), John Nagy, Page 63. 14 Id. 15 Id. 16 Mine Fires Prevention, Detection, Fighting, Donald W. Mitchell, P.E., (1996), 3d Ed., pp 69-70. 17 The Explosion Hazard in Mining, U.S. Department of Labor, Mine Safety and Health Administration, Informational Report 1119 (1981), John Nagy, Page 52. 46 The most important consideration after an explosion is the safety of the mine rescue persons and that of any missing miners. Before sending any person, including mine rescue teams underground, the atmosphere in the mine must be assessed. The atmosphere should be monitored as close to the area of the explosion as possible. This can be accomplished with a borehole. However, a borehole is generally not present and it may take a considerable amount of time for one to be drilled. The only monitoring location available may be where the return air exits the mine. The air at the monitoring location may be diluted and may not give an accurate representation of the conditions in the area of the mine where the explosion occurred. Monitoring of the mine atmosphere should begin as soon as possible. After a mine fire or explosion, the mine atmosphere can be monitored with handheld instruments, infra-red equipment, or gas chromatographs. The handheld instruments are the most readily available and are usually the first equipment on site. They can detect methane, CO, and oxygen. The detection levels vary for each instrument. Generally, the detection level for methane is 0 to 5%. The detection level for CO varies but generally ranges from 0 to 500 ppm for some instruments or from 0 to 999 ppm for others. It is important to know and understand the detection levels of the instrument being used. Infra-red equipment is generally used to measure methane in ranges from 0 to 100%, CO in ranges from 100 to 20,000 ppm, and carbon dioxide in ranges from 0 to 4.0%. Gas chromatographs are generally not very portable but are highly accurate and able to monitor most ranges of gases including methane, oxygen, carbon monoxide, carbon dioxide, and fire gases such as hydrogen, ethylene, and acetylene. Determining trends of the mine gases may be accomplished with any of the described detection equipment, but the gas chromatograph is generally used for this purpose as it is the most accurate. It is also important to determine what the fuel was for the explosion. This information is very difficult to determine initially. Again, since a borehole in the area where the explosion is thought to have occurred is generally not available, monitoring of the return air where it exits the mine may be the only available option. Coal mines liberate methane. It is important to know the normal concentration of methane and the air volume in the monitored air. If it is different than normal, the cause for the difference must be determined before allowing personnel to proceed underground. For example, if the concentration of methane is lower than normal and the volume of air is the same, this may indicate a major short circuit in the ventilation system and methane may be accumulating in the area where the explosion occurred. This is also the area where a fire is most likely to be occurring. If the methane concentration is higher than normal and the volume of air is the same, this may indicate that there was an accumulation of methane somewhere in the mine that was not consumed by 47 the explosion. This could indicate an accumulation of methane is still in existence in the mine. Unfortunately, there is recent history of fires starting after explosions. Most recently: • An ignition/explosion occurred at a mine in Virginia and the miners were evacuated. The air exiting the mine was continually monitored for CO and other fire gases. Before the CO trend had stabilized, it began to trend upward and the mine was subsequently sealed at the surface. When the mine was reopened, evidence of two separate fires was discovered. One was relatively close to the reported location of the ignition/explosion origin. Evidence of a second fire was found thousands of feet away. • A fire occurred after a series of explosions occurred in a mine in Alabama. Although mine rescue teams re-entered the mine, they were subsequently withdrawn after elevated concentrations of methane and a fire was discovered. The area was subsequently sealed. • An explosion occurred at a mine in Illinois. It appeared to have originated inby the longwall face. After the atmosphere in the mine went through a stabilization period, mine rescue teams were permitted in the mine. During their exploration, they found a crib block still burning only a few feet away from an accumulation of explosive methane. The forces of the explosion disrupted the mine ventilation system. It took a period of time for the CO generated from the explosion to reach the main return, No. 1 Drift Opening. As previously shown in Figure 3, the CO began to increase dramatically, peaked and then began to decrease. The CO trend eventually stabilized. Persons generally should not re-enter a mine until the atmosphere has stabilized. The generally recognized stabilization time period is 72 hours. This minimizes the risk to persons from a secondary explosion. 48 Figure 4 illustrates the results of CO measurements obtained in one of several Virginia Mine Main Return return shafts in a mine in Virginia after Carbon Monoxide an ignition/explosion and subsequent fire. The ignition/explosion occurred in the mine at 4:20 p.m. on Day 1. Although there were no samples collected immediately after the ignition/explosion, the gases from the event reported to multiple return shafts. 3000 12:00 PM Day 1 Day 1 Day 2 12:00 AM 12:00 AM 0 12:00 PM 500 12:00 AM 1000 CO From Fire 12:00 PM 1500 12:00 AM CO, ppm O 2000 Ignition/Explosion Occurs at 4:20 pm on Day 1 2500 Day 2 Day 3 Day 3 Day 4 Day and Time Figure 4 - CO Measurements from a Mine in Virginia It took time for the fire to become large enough to be readily detected at a return shaft but eventually began to increase dramatically. The mine was subsequently sealed at the surface because of the increasing trends. It can be seen how CO being produced from a developing fire could be masked by the CO that had been produced by an explosion. If a fire would have started in the Sago Mine after the explosion, as it did in this Virginia mine, it would have taken a significant period of time until the CO produced from the fire exceeded the CO produced from the explosion. At the Sago Mine, a borehole into the area where the explosion occurred was not initially available. Monitoring of the main return was initiated with handheld instruments after the explosion occurred. The initial information indicated the volume of air exiting the mine had not changed significantly. It also indicated relative low CO and methane levels. About 10:30 a.m., the CO levels exceeded the detection limits of the handheld equipment and the methane levels were greatly in excess of the normal levels. The levels of CO remained above the detection level for handheld equipment. A gas chromatograph became available at about 3:00 p.m. Gas chromatograph analysis results for No. 1 Drift Opening are shown in Appendix E. The mine air analysis from the gas chromatograph confirmed the elevated CO and methane levels. These levels continued in a downward trend throughout the afternoon. The CO and methane levels were still trending downward but had not yet stabilized. It was not possible to know with any certainty if the explosion had started a fire in the mine. The elevated methane levels confirmed the possibility of methane accumulations in the inby areas of the mine. Even though these conditions existed, at 4:55 p.m., the command center made the decision to permit the mine rescue teams to begin to explore underground. There was a high degree of risk associated with this decision and it was discussed with all parties including the mine rescue teams before they started underground. 49 Mine Exploration At 5:15 p.m., the CO at the No. 1 Drift Opening was still dangerously high at 1,740 ppm, but the downward trend had been continuing for several hours. At 5:25 p.m., the CONSOL Robinson Run A mine rescue team entered the mine through the fan house and proceeded inby exploring the mine. The CONSOL Blacksville No. 2 mine rescue team was assigned as their backup team. By 5:57 p.m., exploration had reached 9 Crosscut, No. 3 Belt. Exploration then paused to allow team members to check air quality, air quantity and water depths in the explored area. By 7:20 p.m., MSHA’s instantaneous sampling equipment was set-up and monitoring the mine atmosphere exiting No. 1 Drift Opening. Initial readings indicated 1,200 ppm CO, 0.2% methane, and 20.6% oxygen. Gas measurements were recorded about every 15 minutes during the rescue operation. A trend analysis of these measurements was maintained. The mine operator raised a concern about the need to start dewatering the return entries at the inby end of No. 1 Belt to prevent water from blocking the return air course. To address this issue, the mine operator submitted a plan, which was approved by MSHA and the WVMHS&T. This plan permitted the mine rescue team to energize power to a 150 kilovolt-ampere (kva) transformer, located at 23 Crosscut, No. 1 Belt, to power a dewatering pump located in the adjacent return entries. The water pump was energized at approximately 7:55 p.m. At about 8:05 p.m., rescue teams continued their search. The rescue teams continued pushing into the mine until 2:13 a.m. on January 3, when they reached 32 Crosscut, No. 4 Belt. The team saw a red light glowing at approximately 36 Crosscut, No. 4 Belt in the belt entry. Rescue team members identified the light as coming from the AMS system, and believed the system to be energized by electrical power. Due to the risk of an explosion which such an energized component could cause, team members were ordered to retreat out of the mine at 2:40 a.m. The command center ordered the rescue teams to maintain their positions out of the mine until the AMS power was de-energized, which was completed at 3:57 a.m. The teams were then to re-enter the mine. However, an effort to drill a borehole into the mine was occurring at the same time as the rescue effort. Personnel at the drill site notified the command center that the borehole into the 2nd Left Parallel section would be completed in about one hour. At that time, all mine rescue personnel would need to be withdrawn from the mine due to the explosion hazard which drilling through the roof could create. The command center ordered the rescue teams to hold their positions until the borehole was completed. The 2nd Left Parallel borehole penetrated the mine at 5:35 a.m. at a 50 depth of 258 feet. The borehole intersected the section at 23 Crosscut, No. 6 Belt in the No. 4 entry. An air quality sample taken from the borehole at 5:53 a.m. indicated 1,052 ppm CO and 20.4% oxygen. The drillers tapped on the drill steel and listened, hoping to hear a response from the trapped miners. No response was heard. Rescue teams re-entered the mine at approximately 6:57 a.m. In addition, MSHA’s robot was transported into the mine with the teams to 27 Crosscut, No. 4 Belt. The robot was to be used as an additional rescue tool to travel the track entry into 2nd Left Parallel. The rescue teams arrived at 27 Crosscut, No. 4 Belt at approximately 7:34 a.m. Team members unloaded the robot and sent it inby toward 2nd Left Parallel. The teams then began to explore inby 27 Crosscut, No. 4 Belt, independent of the robot. At about 8:48 a.m., the robot became disabled at 32 Crosscut, No. 4 Belt. At 10:45 a.m., as teams continued to explore, the mine operator submitted a plan to have the teams explore to 48 Crosscut, No. 4 Belt, then proceed to explore the return entries of 1st Left for a distance of six crosscuts. Teams were also to examine the overcasts at 49 Crosscut and 51 Crosscut, No. 4 Belt. Once these examinations were completed, exploration was to proceed inby toward the 2nd Left Parallel section by exploring and using the Nos. 7, 8 and 9 intake entries until reaching and examining the seals inby 62 Crosscut, No. 4 Belt. Once the seals were examined, exploration was to continue toward the 2nd Left Parallel section. The plan was approved and the mine rescue teams continued their exploration. At 2:13 p.m., they found the 1st Left crew’s abandoned mantrip between 49 Crosscut, No. 4 Belt and 50 Crosscut, No. 4 Belt. Rescue team personnel disconnected its power, and continued their exploration. At 5:20 p.m., the rescue teams located the first victim, Terry Helms, in the track entry between 57 and 58 Crosscut, No. 4 Belt. By 5:50 p.m., the rescue teams had explored the previous seal locations inby 62 Crosscut, No. 4 Belt. At 6:18 p.m., rescue team members found that seal No. 10 in the No. 9 entry was destroyed. They continued to explore across the seal line and by about 6:47 p.m. had found that the other 9 seals were destroyed as well. They appeared to have been blown in an outby direction. The mine rescue teams finished exploring the seal area and then turned toward the 2nd Left Parallel section. 2nd Left Parallel Exploration At approximately 7:48 p.m., team members found the 2nd Left Parallel crew’s abandoned mantrip at 10 Crosscut, No. 6 Belt. At 8:10 p.m., rescue team members found evidence of 12 SCSRs opened at 11 Crosscut, No. 6 Belt in the No. 7 entry. They also saw footprints heading in an outby direction. Rescue team members traveled outby in an attempt to follow the tracks, and to search 51 for any additional signs of the 2nd Left Parallel crew, but they turned up no further evidence. After 9:40 p.m., a rescue team explored to 17 Crosscut, No. 6 Belt and then retreated back to the FAB by traveling the belt entry. At 11:12 p.m., the command center implemented a plan to extend the search distance beyond the normal 1,000 feet. The plan was to explore to the 2nd Left Parallel faces with a mine rescue team by extending communication using permissible handheld radios. The command center believed that the atmosphere in the mine, including in the 2nd Left Parallel, had stabilized to a point where it would not be life threatening. In an effort to locate the missing miners as quick as possible, a plan was developed that did not adhere to standard mine rescue procedure. Adhering to the standard procedure of advancing the FAB incrementally or airlocking would have added several hours to the search and rescue effort. The teams had to stretch communications as far as possible. The teams were taking a risk in order to try to find the miners as soon as they could. By doing so, communications could be compromised by overextending the handheld radios’ capabilities. Three of the four permissible radios were available. A fourth radio had become non-operational at some point during the rescue. McElroy rescue team members were contacted at the FAB and asked by the command center if they would go beyond normal rescue protocol. They agreed. At 11:17 p.m., McElroy mine rescue team was authorized to search the entries toward the faces of the 2nd Left Parallel section. Two Tri-State team members were stationed on the track at 59 Crosscut, No. 4 Belt. One of those members had a permissible radio to relay communications back and forth to the McElroy team as they advanced into 2nd Left Parallel. The second Tri-State team member had a voice activated mine rescue hard line communication system to communicate to a person at the FAB. As the McElroy mine rescue team explored the 2nd Left Parallel section, they encountered water in the track entry that was approximately knee deep near 8 Crosscut, No. 6 Belt. At this point, the communication on the handheld radio that was used to talk with the Tri-State team member stationed at 59 Crosscut, No. 4 Belt began to break up. Therefore, a member of the McElroy rescue team was positioned near 8 Crosscut, No. 6 Belt to maintain communications with the Tri-State team member stationed at 59 Crosscut, No. 4 Belt. However, as the McElroy team continued to explore inby, the McElroy team member at 8 Crosscut, No. 6 Belt had to walk inby to 10 Crosscut, No. 6 Belt to maintain communication with the inby team members, and walk back outby to maintain communications with the FAB. The distance from the track at 59 Crosscut, No. 4 Belt to 9 Crosscut, No. 6 Belt in 2nd Left Parallel was approximately 620 feet. The handheld radios become less reliable as the distance between users is increased or when the users are not in 52 direct line of sight of each other. The track in this area was not straight and it dipped near 8 Crosscut, No. 6 Belt, resulting in a lack of a direct line of sight. As the McElroy team continued to advance toward the face, they would contact the McElroy team member at 10 Crosscut, No. 6 Belt with the information they wanted relayed to the surface. He would then travel through the water near 8 Crosscut, No. 6 Belt to communicate with the Tri-State mine rescue team member stationed at 59 Crosscut, No. 4 Belt. The Tri-State team member would relay the message to a team member standing next to him, who was manning the hard line device. This team member would then relay it to the FAB located in the No. 7 entry at 57 ½ Crosscut, No. 4 Belt. The team member at the FAB would communicate the information to the command center on the surface. As the McElroy team approached the faces of 2nd Left Parallel they had to leave the track entry, losing sight and radio contact with the McElroy team member near 8 Crosscut, No. 6 Belt. This caused the outby team member to repeatedly wade through the water while trying to maintain communication with the inby rescuers and outby rescuers. He went back and forth numerous times during the rescue and recovery operation, but was not always able to maintain communication with both groups. The distance in the track entry from 10 to 23 Crosscut, No. 6 Belt was approximately 920 feet. This caused messages to break up and be difficult to understand. The McElroy mine rescue team began searching the faces. They found a check curtain constructed across the No. 3 entry and heard a moan coming from behind it. The McElroy team member stationed between 8 and 10 Crosscut, No. 6 Belt heard someone say in an excited voice “there’s noises, there’s guys behind it … we’ve got to go around another break.” He lost contact with them once they went around the crosscut. The team went through the curtain and found the miners. They began to administer first aid to the miner who was making the noise. Other rescue team members immediately went to each of the other eleven miners to make an assessment of their condition and to provide assistance if needed. It soon became apparent that McCloy was the only miner alive and they prepared him for transport to the FAB. A MEU team member left the rest of the team in the barricade and traveled to the power center in the track entry at 23 Crosscut, No. 6 Belt to get a stretcher and to report their findings to the McElroy team member stationed between 8 and 10 Crosscut, No. 6 Belt. Using his handheld radio, he told the McElroy team member near 8 Crosscut, No. 6 Belt that they had “all 12 guys” accounted for and that “we have one alive.” He also asked for immediate help. The entry in which the MEU team member stood had several obstacles in the entry such as supply cars, which weakened the signals of the radios. In addition, the radios’ batteries were weak, and were scheduled to be changed in less than an hour. 53 The McElroy team member stationed near 8 Crosscut, No. 6 Belt stated that someone hollered over the radio “we need help, we’ve found them, we found all the men, we need help.” He recalled that the MEU team member told him “we need medical help. We have two people we’ve got down, we’ve got to have stretchers, we need help.” The McElroy team member shouted into his radio “they found them, they need help, there’s men down.” The McElroy team member stationed near 8 Crosscut, No. 6 Belt was frustrated by the poor radio communications. He ran back and forth trying to improve reception. During this hectic time, the mine rescuers were quickly relaying information. The information communicated from the sender was not being repeated to verify the accuracy of what the recipient had heard. The McElroy rescue team member stationed near 8 Crosscut, No. 6 Belt had to run back and forth several hundred feet to maintain communications, and could not take the time to have people verify all the communications sent and received by him. The mine rescuers at the FAB in 57 ½ Crosscut, No. 4 Belt stated that they received a message of “12 alive” over the headset from 59 Crosscut, No. 4 Belt and that they immediately called outside to the Command Center and repeated “12 alive.” The information communicated to the FAB from the team members inby was not confirmed by the FAB before it was relayed to the Command Center. After the Command Center received this information, they requested and received a confirmation from the FAB. At 11:46 p.m. on January 3, it was recorded in the Command Center log that the message “12 people alive” was received. The mine rescue members at 59 Crosscut, No. 4 Belt stated that they heard the McElroy rescue team member near 8 Crosscut, No. 6 Belt say over the radio that “we found them alive” and “we need help now.” One Tri-State mine rescue member at the FAB and the two at 59 Crosscut, No. 4 Belt traveled to the face to help the rescuers. In addition, a MEU member and a WVMHS&T team member traveled to the face. This resulted in a further breakdown of the communication system. However, the McElroy team member continued to move between 8 and 10 Crosscut, No. 6 Belt trying to maintain communications. Upon reaching the barricade, those five rescue team members assisted in assessing the victims and in transporting McCloy. The team members were all wearing heavy apparatus as they carried McCloy to the FAB. Team members took turns carrying the stretcher through knee-deep water and over concrete block rubble from destroyed stoppings. Some team members were running low on oxygen. As they were approaching 9 Crosscut, No. 6 Belt, one of them stated, “we’ve only got one alive … we think we’ve only got one alive.” The McElroy team member stationed near 8 Crosscut, No. 6 Belt ran a couple of crosscuts outby and relayed back to 59 Crosscut, No. 4 Belt that 54 “they’ve only got one person alive.” He did not wait for a response and did not know if this information was received. He ran up and met the team members near 10 Crosscut, No. 6 Belt and helped carry McCloy to the FAB. By approximately 12:30 a.m. on January 4, the rescuers reached the FAB. According to one rescuer, there were “a bunch of men ready to help, thinking there’s still 12” men alive. Rescue team members placed McCloy on a mantrip and transported him outside. Upon learning of the communication error, the McElroy team captain contacted the command center, and informed the command center that only one person was alive, and eleven were deceased. The command center then ordered all mine rescuers to exit the mine. By about 1:00 a.m., McCloy had been transported to the surface and placed in an ambulance. By about 1:20 a.m., all rescuers had exited the mine. The command center debriefed the rescue teams. After the debriefing, the command center decided to send the Viper mine rescue team to the barricade to verify the initial findings made by the McElroy and Tri-State mine rescue personnel. The Viper Mine rescue team members, who were emergency medical technicians (EMTs), were provided stethoscopes to confirm the status of the miners in the barricade. The Viper rescue team and their back up, the Robinson Run rescue team, entered the mine around 1:38 a.m. The Viper team also experienced communication problems. They were unaware of how the McElroy team had dealt with the gaps in communication, and planned to post a member of their team at 9 Crosscut, No. 6 Belt in 2nd Left Parallel Section to relay communications back to the FAB. However, they were unable to maintain communication with this member, and decided to take him to the barricade with them. As a result, the Viper team did not have communications with the FAB for a period of time. After the EMTs on the Viper team confirmed that the other miners had perished, they could not report this information back to the FAB because they did not have a rescue team member with a radio near 10 Crosscut, No. 6 Belt to relay messages. Their confirmation did not reach the FAB until after the Robinson Run team came up to re-establish communications at about 4:14 a.m. The command center and the rescue teams discussed the recovery of the deceased miners. Normal procedure would be for the area to be re-ventilated prior to any recovery, to limit the exposure of rescue team personnel to any hazards. However, rescue team members volunteered to re-enter the mine and, under apparatus, recover the deceased miners. By around 9:22 a.m. the victims had been recovered and transported to the FAB. Shortly thereafter, the mine rescue teams and the victims were transported to the surface. Appendix F contains the victim information data sheets. 55 Rescue Borehole Chronology Joseph Myers had been the Chief Engineer for the mine operator since July 29, 2005. He reported to Dunbar. He was responsible for the development of the coordinates for the drilling program, and in charge of planning the drilling of the boreholes during the rescue efforts at the mine. Myers was notified of the accident at around 10:30 a.m. on January 2. Myers left his home and drove to the mine. While on the way, he made calls to Alpha Engineering (Alpha), an engineering group contracted to perform the mapping at the mine. Myers did not know exactly what would be needed. He asked Alpha to send a mapping grade handheld global positioning system (GPS)18 and a survey grade handheld GPS, as well as a conventional survey. Myers had in his possession a GPS unit that was accurate to plus or minus 30 to 50 feet. He tried to use the GPS unit numerous times by placing it in front of his car’s windshield during his trip to the mine. Each effort was unsuccessful due to low signal strength from the satellites. Gary Hartsog, President of Alpha, was notified by an Alpha employee of the accident around 10:45 a.m. At around 11:00 a. m., Hartsog tried to telephone Myers but was unsuccessful. Hartsog then telephoned Dunbar to obtain more details of the accident and determine how Alpha’s resources should be used. Hartsog then contacted David Prelaz, an Alpha employee, and instructed him to work on completing an updated mine map for Myers. Myers estimated that he arrived at the mine at around 11:30 a.m. He again checked the GPS but had virtually no signal. He entered the mine building and was briefed by Dunbar. At about 12:07 p.m., Myers again contacted Alpha, to determine when the mapping grade GPS would arrive, and to obtain an updated map of the mine. Alpha personnel downloaded a file copy of the updated mine map to a file transfer protocol site to which both companies had access. Myers then downloaded the file to a computer, which allowed him to start looking at potential areas for drilling. 18 A GPS uses a worldwide radio navigation system formed from 24 satellites and their ground stations. The system was designed for and is operated by the U.S. military. GPS provides specially coded satellite signals that can be processed in a GPS receiver, enabling the receiver to compute position, velocity, and time. Four GPS satellite signals are used to compute positions in three dimensions and the time offset in the receiver clock. Their accuracy varies depending on the receiver system deployed. 56 In addition, Myers used a planametrics map to assist in identifying surface structures, terrain and other surface features, to find potential drilling sites. At about 1:00 p.m., Myers contacted Hartsog and again requested a survey grade GPS unit. During this conversation, Hartsog informed him that both the mapping grade and survey grade GPS equipment would be sent. Hartsog stated that the mapping grade GPS equipment was located in Sutton, West Virginia, and a person was ready to transport it to the mine. The survey grade GPS equipment required preparation time. Also due to the holiday, time was needed to organize people to operate and transport the equipment from Danville, West Virginia. Based on the mapping information, Myers selected a drill site and forwarded it to mine management for approval. At about 1:35 p.m., Ty Coleman and Myers discussed with Satterfield and WVMHS&T personnel a proposal to drill a borehole into the 2nd Left Parallel section. Myers explained that the area above the 2nd Left Parallel conveyor belt feeder had the gentlest grade on which to develop a road and drill pad site. The hillside was much steeper at other potential drill sites further inby in the mine. At approximately 2:07 p.m., Alpha employee Matt Ashley arrived on site with a map grade GPS unit. Ashley informed Myers that the map grade GPS had poor signal strength. Myers provided Ashley with a specific coordinate derived from the maps. Myers said that he would verify that coordinate when it was approved. He asked Ashley to locate that potential borehole site. At approximately 3:00 p.m., the mine operator obtained permission from the landowner for development of an access road and a borehole drill site. Myers estimated that the 2nd Left Parallel drill site was generated with the map grade GPS unit around 3:30 p.m. However, the initial drill site coordinates were not satisfactory due to poor signal strength. The software used by the GPS units requires that a certain number of satellites be in communication in order to complete a survey. The unit displays the Positional Dilution of Precision (PDOP) reflecting the signal strength and the number of satellites used. Weather, trees and structures may affect the PDOP. When the PDOP number is high, the results are less reliable. Alpha encourages its surveyors to use a PDOP that is in the range of 5 to 7. On the day of the accident, the surveyors were receiving a PDOP in the 16 to 18 range. Although the coordinates obtained were not deemed accurate enough to drill, construction of the pad and road began immediately. It was finished at about 6:30 p.m. The mine operator’s project engineer, Kermitt Melvin, arrived between 5:00 p.m. and 6:00 p.m. and began to assist Alpha personnel. At around 5:30 p.m., the command center discussed the proposed location of the borehole and gave 57 approval to drill the hole at the specified location, to within 20 feet of the coal seam. An Alpha survey crew arrived on site about 6:00 p.m. They started a survey, working from permanent monuments at the mine and at the Spruce Fork No. 1 Mine, a nearby ICG mine. A survey grade GPS unit arrived on site at approximately 9:30 p.m. The surveyors attempted to use a real-time GPS19, which involved radio communication between GPS units. Adequate radio communication could not be established between the units to perform at an acceptable level. The surveyors informed Myers of the problem with the survey grade unit. Therefore, the surveyors resorted to using observations of GPS receiver units on permanent monument points at the Sago Mine and Spruce Fork No. 1 Mine to provide a baseline. Once those observations were made, the results were downloaded into a computer and processed. Calculations were made using a mathematical model to provide coordinates for two points in relatively close proximity to the drill site. That process was performed in the field on a laptop computer and was completed at about 11:00 p.m. The conventional surveying method20 was employed, using two points to locate the exact site for drilling. This was completed at 11:30 p.m. The original site of the drill hole determined by the mapping grade GPS was off by about 30 feet. By approximately 2:00 a.m. on January 3, the drill site had been resurveyed and the drill rig mast had been plumbed by a survey crew. Drilling of Borehole No. 1 commenced at 2:45 a.m. Because the rescue teams had withdrawn from the mine, the borehole penetrated into the 2nd Left Parallel section at 5:35 a.m.at a depth of 258 feet. The borehole intersected the section at 23 Crosscut, No. 6 Belt in the No. 4 entry, over the conveyor belt feeder. The crew repeatedly struck the drill steel trying to get a response from the missing miners, but no response was heard. An air quality sample taken from the borehole at 5:53 a.m. indicated 1,052 19 Real time GPS surveying techniques can provide measurements to the accuracy of a centimeter, over 10 kilometer baselines, by tracking five or more satellites and using real-time radio links between the reference and remote receivers. 20 Conventional surveying consists of an instrument such as a transit or a total station being placed over a point, and being used to accurately determine points and lines of direction (bearings) on the earth's surface. Maps or plans are prepared from the data generated. 58 ppm CO and 20.4% oxygen. Figure 5 illustrates the results of CO measurements obtained in Borehole No. 1. Appendix E contains the gas chromatograph analysis results for samples collected at the Borehole No. 1. Sago Mine Borehole No. 1 Carbon Monoxide 1400 Explosion Occurs at 6:26 AM 1200 CO (ppm) 1000 800 600 400 200 01/07/06 12:00 AM 01/06/06 12:00 PM 01/06/06 12:00 AM 01/05/06 12:00 PM 01/05/06 12:00 AM 01/04/06 12:00 PM 01/04/06 12:00 AM 01/03/06 12:00 PM 01/03/06 12:00 AM 01/02/06 12:00 PM 01/02/06 12:00 AM 0 Date and Time Figure 5 - Borehole No. 1 Carbon Monoxide Results At 6:30 a.m., the crew lowered a camera into Borehole No. 1. The images displayed indicated that the area surrounding the conveyor belt feeder was relatively undisturbed. There was no sighting of the missing miners. Two additional borehole sites were located to enable drilling to penetrate into 1st Left section and the outby end of 2nd Left Parallel section. The crew began drilling Borehole No. 2 at approximately 6:50 a.m., after the drill rig had been relocated from Borehole No. 1. Borehole No. 2 reached the hold depth of 360 feet, approximately 30 feet above the mine, at 2:24 p.m. The personnel in the command center decided not to complete the hole since rescue teams had advanced in the mains inby 1st Left. Borehole No. 2 was finished at a later date to aid in the recovery of the mine. The drilling of Borehole No. 3 was started at about 2:35 p.m. but was stopped short of penetrating the mine because the hole was generating 60 to 80 gallons of water per minute, which it was feared could cause additional problems in the mine. This hole was never completed. 59 MINE RECOVERY The 1st Left and the area inby the 2 North Main seals had not been explored. The explosion caused extensive damage to the ventilation controls. Air quality monitoring at the No. 1 Drift Opening, Borehole No. 1, and eventually, through a series of additional boreholes was initiated. Air quality monitoring continued until the mine atmosphere was stable and safe for miners to re-enter the mine and restore ventilation. On January 5th, Borehole No. 2 was completed into the track entry at 31 Crosscut, No. 5 Belt. It was used to monitor air quality in 1st Left. Boreholes No. 4 – 7 were drilled into the 2nd Left Mains. Borehole No. 4 was started on January 6th and completed on January 8th. Borehole No. 7, the last borehole drilled was started on January 17th and completed on January 19th. Boreholes No. 4 - 7 were used for air quality monitoring, dewatering, and/or ventilation. Dewatering of the 2nd Left Mains was started on January 12th and was not satisfactorily completed until January 20th. The air quality analysis of the mine atmosphere remained favorable throughout the mine recovery. On January 21, 2006, with ventilation established to the boreholes at the inby end of the 2nd Left Mains, mine rescue/recovery teams entered the mine to examine and re-establish ventilation following the approved plan developed by the operator. In addition to establishing ventilation, some areas of the mine had to be dewatered. Dewatering required the restoration of portions of the underground mine electrical system. After the mine rescue/recovery teams examined and established ventilation throughout the mine, the underground portion of the investigation commenced on January 26, 2006. Figure 6 is a photograph of a damaged ventilation control between the Nos. 6 and 7 entries at 59 Crosscut, No. 4 Belt found during recovery of the mine. Figure 6 - Damaged Stopping at 59 Crosscut, No. 4 Belt 60 Figure 7 is a photograph of a damaged overcast in the No. 2 entry 58 Crosscut, No. 4 Belt found during recovery of the mine. Figure 7 - Damaged Overcast at 58 Crosscut, No. 4 Belt 61 INVESTIGATION OF THE ACCIDENT MSHA’s Administrator of Coal Mine Safety and Health appointed a team to investigate the accident at the mine. The team consisted of personnel from MSHA Coal Districts 2, 5, 7, and 11 and from Technical Support. The team utilized numerous resources, including personnel from MSHA Headquarters, Educational Field Services, Small Mines, and Technical Support. The Administrator appointed Richard A. Gates, District Manager of Coal District 11, as accident investigation team leader. A portion of the investigative team arrived at the mine on January 2, and the full team arrived at MSHA’s Bridgeport, West Virginia field office by January 8. The investigation was conducted jointly with WVMHS&T. The mine operator and two groups appointed by Sago miners, the UMWA and an employee group, also participated in the investigation. Appendix G lists the individuals who assisted with the investigation. Preliminary information and records were obtained from MSHA’s Coal District 3 and from the mine operator. The investigation consisted of both in-mine and out-of-mine activities. At the mine, the investigative procedures included mapping the entire mine, photographing the affected areas, and collecting physical evidence. The mapping of the entire mine is included in Appendices H-1 through H-9. The physical evidence was examined or tested on-site and/or later in an appropriate facility. The underground investigation could not begin until the rehabilitation work of drilling boreholes, dewatering, and restoring ventilation was completed. This delayed the investigation team from entering the underground mine until the work was completed. The entire underground mine was then examined and deemed safe for entry. The underground portion of the investigation began on January 26. The investigative team identified numerous people who had knowledge relevant to the accident and conducted 80 interviews. These included officials of ICG, miners, a past employee, contractors, MSHA inspectors, WVMHS&T inspectors, mine rescue team members, and medical professionals. The interviews were conducted at the U.S. Bankruptcy Court and the U.S. District Court in Clarksburg, West Virginia; the Wingate Hotel in Bridgeport, West Virginia; the Renaissance Hotel in Morgantown, West Virginia; and MSHA offices in Bridgeport, Summersville, and Morgantown, West Virginia. Investigators conducted follow-up interviews of four previously interviewed witnesses. Additional information was obtained from contractors, and state and local authorities. Pertinent records were obtained and reviewed during the course of the investigation. The findings in this report are based on the information obtained during the investigation. 62 Mine Emergency Evacuation and Firefighting Program of Instruction MSHA approved the Mine Emergency Evacuation and Firefighting Program of Instruction on February 3, 2004. The mine operator later submitted two requests to revise page 3, which MSHA approved on May 3, 2005, and on November 8, 2005. These supplements changed the Emergency Alert Chart containing the persons to be notified in the event of an emergency. The program identified the dispatcher on duty as the “Responsible Person” in the event of a mine emergency involving a fire, explosion or gas or water inundation. The program stated in part: “The responsible person shall have current knowledge of the assigned location and expected movements of miners underground, the operation of the mine ventilation system, the location of the mine escapeways, the mine communications systems, any mine monitoring system if used, and the mine emergency evacuation and firefighting program of instruction. The responsible person shall initiate and conduct an immediate mine evacuation when there is a mine emergency which presents an imminent danger to miners due to fire or explosion or gas or water inundation. Only properly trained and equipped persons essential to respond to the mine emergency may remain underground.” In addition to being designated as the responsible person, the dispatcher had other duties, including controlling the mine traffic underground and monitoring the AMS. Notification The program included a list of persons that the dispatcher on duty was to notify immediately in the event of an emergency involving a fire, explosion or gas or water inundation. The list contained mine management, MSHA and WVMHS&T personnel. Chisolm was on duty at the time of the emergency. He called Stemple at 7:00 a.m. and spoke to him for about 15 minutes regarding the events taking place at the mine. At about 7:15 a.m., Stemple was patched through to Jeffrey Toler who was underground assessing what had happened. Jeffrey Toler advised Stemple that he was not sure what had happened. He said that they had found the 1st Left crew, and they were bringing them to the surface. Jeffrey Toler related that the 1st Left Crew stated that there were several intake stoppings out, and that there was smoke and dust in the air as they traveled along the primary intake escapeway. When Stemple learned from Jeffrey Toler that there was dust and smoke in the air and that there had been no contact with the 2nd Left Parallel crew, he told Jeffrey Toler to re-establish ventilation as deep into the mine as he 63 could in an attempt to prevent a short circuit of air to the 2nd Left Parallel section. Stemple made other calls before attempting to notify MSHA’s Bridgeport, West Virginia Field Office Supervisor Kenneth Tenney at his residence. He left a message on Tenney’s answering machine at 7:50 a.m. At 8:28 a.m., Stemple reached Bridgeport Office Supervisor James Satterfield at home. Satterfield issued a verbal 103(k) order at 8:32 a.m. Stemple notified the mine of the order at 8:35 a.m. Evacuation of the Mine The Mine Emergency Evacuation and Firefighting Program of Instruction states as follows: “In the event that you are notified of or discover a mine fire, evacuation and fire fighting procedures shall begin immediately for those in the mine. Only those necessary to fight the fire shall remain in the mine. Those in outby areas or away from the mine phones will be notified by sending a messenger to their work area. From any area of any section the primary escapeway should be used first, and the alternate used only if the primary cannot (due to smoke, fire, water, roof fall, bad top, etc). The Dispatcher should be notified of your intention to evacuate by using the mine phone or the trolleyphone communication system." Immediately following the explosion, Owen Jones directed his crew and the other miners present to evacuate the mine. The miners traveled the track entry to 37 Crosscut, No. 4 Belt and entered the primary intake escapeway. Owen Jones had a phone conversation with Chisolm and told him that something had happened. He said that he felt a force of air coming from the direction of 2nd Left Parallel section, and he thought there must have been an explosion. Wilfong was listening, and instructed Jones to take his crew to the primary intake escapeway and evacuate the mine. The miners were already evacuating and continued to do so. Wilfong, Jeffrey Toler, Schoonover and Hofer entered the mine on a battery powered track mantrip. They did not take any gas detection instruments with them. They picked up John Boni along the track entry as they traveled underground. They found the miners that were evacuating near 27 Crosscut, No. 4 Belt. They all evacuated the mine with the exception of Jeffrey Toler, Schoonover and Owen Jones, who remained underground. After taking the miners from the 1st Left crew to the surface, Wilfong and Hofer re-entered the mine with curtain, nails, boards, saws, detectors and a hard hat for Owen Jones and rejoined Jeffrey Toler, Schoonover, and Owen Jones. They then traveled inby to 32 Crosscut, No. 4 Belt. Even though they realized there had 64 been an explosion, they began making ventilation repairs. They installed check curtains in the crosscuts where stoppings had been damaged between the track entry and the intake entries up to and including 57 Crosscut, No. 4 Belt. Once the curtain was installed at 57 Crosscut, No. 4 Belt Jeffrey Toler, Schoonover and Wilfong observed the area inby 58 Crosscut, No. 4 Belt where the air velocity had diminished. Smoke was very thick and was not dissipating, hindering visibility. After making unsuccessful verbal attempts to contact the missing miners, they evacuated the mine. SCSRs The Mine Emergency Evacuation and Firefighting Program of Instruction states, “Where emergency evacuation is required, personnel should immediately don their Person Wearable Self Contained Self Rescuer (PWSCSR).” The mine operator provided the miners with CSE-SR 100 Self-Contained Self-Rescuers. Three miners working at outby locations did not don their SCSR units during their evacuation. Only seven of the 13 miners who were at the 1st Left track switch donned their SCSRs during the evacuation. The 12 miners in 2nd Left Parallel section donned their SCSR units while trying to evacuate. The miner found near 2nd Left Parallel switch had not donned his SCSR. Belt Fire Detection System The mine used an AMS to detect gases that might result from a fire in the mine. The AMS was a Pyott-Boone Mineboss system that included a computer located on the surface in the dispatcher’s office, which had multiple surface and underground sensors. The AMS required only one computer for the system to function. The system monitored the CO levels at all of its sensors, and showed belt operations and power status. The program required that the system initiate fire alarm signals at a surface location where a responsible person was always on duty when persons were underground. The responsible person was to be trained in the operation of the AMS and the proper procedures to follow in the event of an emergency or malfunction. A map or schematic identifying each belt flight and the details of the monitoring system was displayed on the monitor. Carbon monoxide sensors were required to be spaced along the conveyor belt at 1,000 foot intervals, and a sensor for the section tailpiece had to be between 50 and 100 feet inby or outby the section tailpiece depending on the direction of airflow. A CO sensor was required for each belt drive and tailpiece. However, where a belt drive discharges coal onto another belt tailpiece as a continuation of a belt conveyor system, without a change in direction and on the same split of 65 air, only one sensor was required. An air velocity of 50 feet per minute (fpm) or greater and a definite and distinct movement in the designated direction was required by the program. The system was required to have both visual and audible alarm signals. A visual or audible alert signal was required to activate when a sensor detected 10 ppm CO above the ambient level established for the mine. An audible alarm signal distinguishable from the alert signal was required when a sensor detected 15 ppm CO above the ambient established for the mine. The established ambient for the mine was 0 ppm. When the system gave an alert signal (10 ppm CO), the program requires all persons to be withdrawn to a safe location outby the working places and action taken to determine the cause of the alert. When the system gave an audible alarm (15 ppm CO), all persons in the same split(s) of air were to be immediately withdrawn to a safe location outby the sensor(s) activating the alarm unless the cause was known not to be a hazard to the miners. If an alarm signal (15 ppm CO) was given at shift change, no one was permitted to enter the mine except those qualified persons designated to investigate the source of the alarm. On January 2, 2006, at 6:04:54 a.m.,21 a CO monitoring sensor at the 1st Left section tailpiece, identified as “station 1.99 1 left section” was taken off scan (manually turned off). At 6:05:05 a.m. the sensor initialized and was placed on scan (manually turned on). At 6:05:10 a.m. the system alarmed and indicated a reading of 26 ppm CO. According to the CO monitoring log, problems with this sensor had been occurring off and on since December 9, 2005 when the sensor was calibrated. Several hours after calibration, the device reported an event that generated an alarm indicating a reading of -1 ppm, and the alarm reset a few seconds later. Between December 10, 2005 and December 31, 2005, several events were recorded showing that the sensor lost communication with the Master Control Station on the surface for periods of 5 seconds to 1 hour and 19 minutes. Between December 15, 2005 and December 31, 2005, there were numerous entries in the CO event log indicating that the alarms and latch resets were increasing. The maximum “Alarm Latch Set” value was 26 ppm on December 31, 2005 and the maximum recorded CO value was 38 ppm on December 30, 2005. Additionally, several entries in the log during this time period indicated that the device had lost and regained communications. 21 The record of the AMS times was 4 minutes and 56 seconds fast. The times shown are corrected times. 66 Two methods are used to reset the warning and alarms after they have activated or latched. When the CO level recedes to a point below where the alarms are set, the sensor will reset automatically if the system is in the auto reset mode. If it is not in the auto mode it can be reset from the surface or it can be reset by personnel at that sensor’s location. This CO sensor was in the automatic reset mode when inspected during the investigation. The value reported by the sensor at Station 1.99 at the 1st Left section tailpiece was not correct. With clean air applied, the unit reported a value of 26 ppm CO. When CO calibration gas containing 50 ppm was applied, the sensor reported 74 ppm. The difference between the ‘zero’ and ‘span’ points was 48 ppm. When coupled with the event log readings showing: (a) steadily increasing alarm readings, and (b) the calibration adjustment attempted on December 15, 2005, it appears that the sensor had a zero drift. The connections between the CO sensor and the remote alarm on the section were incorrectly wired. The remote alarm could not be activated by the attached sensor or from the surface. Additionally, when properly calibrated and the wiring to the alarm unit on the section was corrected, this CO sensor would cause the attached alarm to give audible and visual warnings continuously in clean air. In conclusion, the CO sensor with address station 1.99 1 left section was communicating with the system on February 1, 2006 when inspected during the investigation. However, the unit was not measuring CO within acceptable limits. The remote alarm would give an audible and visual warning when the test buttons were pressed on the device, but it would not provide warning signals at the section loading point when actuated by the CO sensor or the surface master control station. It was determined that the system was malfunctioning because the wiring between the CO sensor and the alarm was incorrect. The data suggests that the sensor and remote alarm did not function properly at the time of the explosion. Furthermore, the data suggests that some corrective action had been attempted in the early morning hours of December 31, 2005, and that the system operator had attempted to reset the device at approximately 6:05 a.m. on January 2, 2006. Appendix I contains an executive summary of “Investigation of Pyott-Boone Electronics MineBoss Monitoring and Control System.” The Ventilation Plan and the Mine Emergency Evacuation and Firefighting Program of Instruction contained guidelines for the installation, use and maintenance of the system, and outlined the appropriate responses to the signals provided by the system. The plan and program also outlined procedures to follow if the system was partially or completely inoperative. The program states “When the carbon monitoring warning system gives an audible alarm at 15 ppm above the established ambient level at shift change, no 67 one shall be permitted to enter the mine except qualified persons designated to investigate the source of the alarm.” At 6:05:01 a. m. the sensor at Station 1.99 at the 1st Left section tailpiece indicated a reading of 26 ppm CO. The 2nd Left Parallel crew had entered the mine about 6:00 a. m. The program required that, “If miners are enroute into the mine, they shall be held at, or be withdrawn to, a safe location outby the sensor(s) activating the alarm.” The 2nd Left Parallel crew’s route of travel and eventual work location was not affected by this alarm, and they continued into the mine. There was no indication that they were contacted. The 1st Left crew entered the mine about 6:05 a. m. There was no indication that any efforts were made to investigate the alarm. The Program further states “When a determination is made as to the source of the alarm, and that the mine is safe to enter, the miners shall be permitted underground.” A responsible person was required to be on duty at all times when miners were underground. The person was to be situated so that he could see or hear the alert and alarm signals. As noted above, the responsible person for the system was to be trained in the operation of the AMS and in the proper procedures to follow in the event of an emergency or malfunction and, in that event, was to take appropriate action immediately. However, some dispatchers at the mine were unaware of the correct alert and alarm levels, or of the proper procedures to follow when those alert and alarm levels were reached. In addition, dispatchers were improperly using the AMS to signal miners on the working sections to answer the mine phone. Barricading Instructions The Mine Emergency Evacuation and Firefighting Program of Instruction provided guidelines for barricading when miners are entrapped by toxic gases from fires or explosions. The program stated that the miners should collect tools, timbers, boards, brattice cloth, water, dinner buckets, self-contained self rescuers and whatever else may be useful. Barricade construction should begin as soon as possible. A place of several hundred feet of entries or rooms should be chosen to provide as much oxygen as possible and the area should be made air tight in an attempt to shutout toxic gases thereby creating a toxic gas-free atmosphere. Theoretically, an average size person breathes approximately one (1) cubic yard of air per hour. A rule of thumb is that about 8 feet of entry length should provide air to sustain one person for one day. The ventilation current outby the barricade should be shut off or short-circuited as soon as possible by opening personnel doors or knocking out permanent stoppings or overcasts. If a series of controls are built (air lock) to ensure an air tight seal, a sign should be placed outby the first control indicating persons are behind the barricade. To conserve oxygen, persons should remain as quiet as possible, near the floor and separated by several feet. However, one person should walk around occasionally to mix the air. Flame safety lamps should be extinguished and cap lamps should be 68 turned off after the barricade is completed. Persons should listen for 3 shots from the surface. They should return a signal by pounding on the mine roof 10 times. The persons should repeat pounding on the mine roof about every 15 minutes. This should be repeated until they hear 5 shots which would indicate that they have been located. Oxygen cylinders, such as those used for oxygen/acetylene cutting torches, could provide an additional source of oxygen in a barricade. Barricading The erection of a barricade by miners who cannot escape after an explosion or fire can be a life-saving measure as a last resort. Miners who have been physically blocked by an explosion may seal themselves promptly behind welllocated and well-constructed barricades, bulkheads, or stoppings. Since the first records were maintained in 1909, the United States Bureau of Mines (USBM) has recorded that lives have been saved by barricading.22 Explosions change the mine atmosphere and create high concentrations of CO, low levels of oxygen, and other gases in a short period of time. A wellconstructed barricade should be practically airtight to prevent ingress or egress of air. The miners who go behind the barricade are dependent upon the air within this enclosed area. Barricading was an option of last resort after all avenues of escape to the outside were believed to be cut off. After the 2nd Left Parallel crew encountered smoke and gases during efforts to exit the mine on the mantrip, they attempted to find other possible exits. When these attempts failed, they retreated to the section and tried to isolate themselves from poisonous gases by building a barricade. Records indicated the 2nd Left Parallel crew had been trained in the methods of barricading and location of barricading materials during annual refresher training. The miners knew that ventilation controls had been blown out. McCloy recalled Martin Toler instructing the miners to construct a barricade from curtains. McCloy stated they decided to use curtain from the face area since some miners did not have SCSRs, and because it would take more time and effort to use concrete blocks. McCloy indicated that they attempted to construct the barricade to keep the smoke out. He further described that initially there was some smoke inside the barricade, but that the smoke faded and the air cleared a little bit. 22 Saving Life By Barricading In Mines And Tunnels At Times Of Disaster, United States Department of Commerce, Information Circular 6701, Harrington, D. et al. 69 However, James Bennett wrote at 11:40 a.m. that “we have air right now but the smoke is bad.” Experiments by the USBM show that a man in a confined space needs about a cubic yard of normal air each hour. The barricade location selected by the miners was inby and included the last open crosscut of the No. 3 entry. The width of the crosscut between the Nos. 3 and 4 entries was about 17 to 20 feet. The width of the No. 3 entry outby the last open crosscut was about 18 to 20 feet. Curtains were installed across these locations. A diagonal curtain was installed from the right rib of the inby corner, to the left rib of the outby corner, in the No. 3 entry. The diagonal curtain was approximately 29 feet in length from rib to rib and balled up on the outby end, according to the captain of the McElroy mine rescue team. The volume of the larger area, (curtain in the crosscut and the curtain in the entry to the face) was about 23,800 cubic feet. The volume of the smaller area, (diagonal curtain to the face) was 15,350 cubic feet. The larger area calculated into cubic yards for 12 miners would be 73 cubic yards per miner. The smaller area for 12 miners would be only 47 cubic yards per miner. This shows that the miners had enough air to sustain them for at least 47 hours if they remained in the smaller area within the barricade and if normal air was in the barricade. This is about 6 hours longer than it took for mine rescue teams to reach the barricade. Figure 8 is a drawing of the barricade. Figure 8 -Drawing of Barricade The CO concentration at the time the barricade was constructed is unknown. A borehole was drilled into the No. 4 entry in the area of the conveyor belt feeder at 5:35 a.m. on January 3, 2006. The first bottle sample analysis taken from the borehole showed 1,052 ppm of CO. The captain of the McElroy mine rescue team was the first person to enter the barricade sometime after 11:30 p.m. on January 3, some 41 hours after the explosion. He and an MSHA mine rescue 70 team member stated that the CO level was 300-400 ppm around and in the barricade area, at the time they found the barricade. The CO level at the borehole at 9:30 p.m. was 205 ppm.23 According to these witnesses, the curtain across the crosscut between the Nos. 3 and 4 entries was loosely hung and open about one foot at the inby side. The diagonal curtain was open about one foot at both ends when the barricade was entered. They did not notice what was used to hang the curtains (nails, wire, etc.). They also said no coal or other sealing material was on the bottom of the curtain to weigh it down for a tight fit. Other possible barricading material was present on the supply car in the track entry at 17 Crosscut, No. 6 Belt, including four pallets of 6-inch concrete blocks, mortar, wedges, headers and cap boards. Fifty 6-foot and sixteen 8-foot posts, and spray sealant were located at 5 Crosscut, No. 6 Belt. Carbon Monoxide Poisoning Carbon monoxide is a colorless, odorless, and highly toxic gas. It is formed as a by-product of burning organic compounds. The composition of the mine atmosphere after the explosion would have been dictated by the methane concentration and quantity, the uniformity of the methane mixture, the amount of coal dust ultimately involved in the explosion, and to a lesser extent other variables such as humidity, turbulence and other materials which were in the explosion zone. NIOSH, formerly the USBM, has provided research data on the composition of an atmosphere after a methane or coal dust explosion in the laboratory. The data indicated that if a methane concentration of 12% was present and if methane was the sole fuel source prior to the explosion, CO could be formed to a concentration as high as 80,000 ppm (8.0%). The data also indicated that if coal dust was the sole fuel source involved in the explosion at a concentration of one ounce per cubic foot, CO could be formed to a concentration as high as 46,000 ppm.24 Also, NIOSH has conducted numerous explosion tests at its experimental mine and collected mine atmosphere samples after the tests. This data indicated that the CO concentration could reach as high as 117,000 ppm (11.7%).25 When CO is inhaled, it is diffused into the bloodstream and displaces oxygen from the hemoglobin that is found in red blood cells. It combines with hemoglobin about 200 to 250 times faster than oxygen. Inhalation of even small 23 The CO reading of 205 ppm was determined using a gas chromatograph. The hydrogen level was 136 ppm. Hydrogen is an interference gas that often causes the handheld detectors to read high. 24 The Explosion Hazard in Mining , United States Department of Labor, MSHA Informational Report 1119, (1981), John Nagy, page 63. 25 Id., page 64. 71 amounts of CO can cause oxygen deficiency, known as hypoxia. Hypoxia may cause headaches, nausea, dizziness, fatigue, confusion, drowsiness, rapid breathing, increased pulse rate, vision problems, chest pains, convulsions, seizures, loss of consciousness, and may eventually cause death. A person with elevated levels of carboxyhemoglobin or CO poisoning is often described as anemic, due to a low hemoglobin level available to bind to oxygen. Carbon monoxide increases the release of nitrous oxide in the system. Nitrous oxide causes a drop in blood pressure by interfering with cellular respiration, resulting in a decrease in the amount of blood flow to the brain, as well as a reduction in the amount of oxygen in the blood that is flowing. Reduced oxygen in the blood also alters the hemoglobin molecule so that it will not release oxygen as readily to the cell. Carboxyhemoglobin is the amount of hemoglobin attached to CO, and is measured in blood to detect CO toxicity. At 20% and above, a person starts to have trouble with motor skills. They may be conscious but nauseated, and begin suffering a headache. Thinking skills and even emotions may start to deteriorate beyond a 20% carboxyhemoglobin level. At 30% to 40%, the person may be quite confused, experience difficulty performing tasks and, depending on their risk factors, suffer unconsciousness. Carboxyhemoglobin that is greater than 80% is immediately fatal. Tables 5 and 6 summarize the effects of carbon monoxide. Not all individuals will respond similarly to the effects of CO inhalation. Certain risk factors must be considered, for example, a person with moderate cardiac or pulmonary disease, emphysema or anemia, or a long-term smoker, may respond more severely to a lower level of CO than someone without those conditions exposed to a higher level. The deprivation of oxygen caused by CO poisoning causes a variety of physical ailments. There are also neuropsychological problems associated with the poisoning. 72 Table 5 - Summary of Toxic Effects Following Acute Exposure to Carbon Monoxide26 Carboxyhemoglobin Signs and Symptoms In Blood (%) <2% No significant health effects 2.5%-4.0% Decreased short-term maximal exercise duration in young healthy men 2.7%-5.2% Decreased exercise duration due to increased chest pain (angina) in patients with ischaemic heart disease 2.0% - 20.0% Equivocal effects on visual perception, audition, motor and sensor motor performance, vigilance and other measures of neurobehavioral performance 4.0%-33.0% Decreased maximal oxygen consumption with short-term strenuous exercise in young healthy men 20%-30% Throbbing headache 30%-50% Dyspnea, dizziness, nausea, weakness, collapse, coma > 50% Convulsions, unconsciousness, respiratory arrest, death Individuals who have high to low levels of carboxyhemoglobin in their body may seem fine initially, but may experience memory loss a few days later. This delayed reaction is neurologic syndrome and is associated with CO poisoning. These delayed symptoms, including psychological disability, may occur anywhere from forty-eight hours to months and even years afterwards. The hippocampus (memory), the basal ganglia (motor function) and the cerebellum (balance) are referred to as watershed areas because they are located deep within the brain and at the end of the blood circuit. Collateral circulation is the process of providing blood flow through an intricate network of vessels from healthy areas of the brain to areas that have been damaged. This network of blood vessels branches deep into the brain, becoming smaller and smaller until they reach the end of the blood circuit. A person with a significant degree of CO poisoning will be affected in these three areas of the brain. 26 www.camr.org.uk/chemicals/compendium/carbon_monoxide/acute.htm 73 Table 6 - Summary of Toxic Effects Following Acute Exposure to Carbon Monoxide27 PPM CO Percent Symptoms Experienced by Healthy in Air CO in Adults Air Less than 0.0035% No effect in healthy adults 35 ppm 100 ppm 0.01 % Slight headache, fatigue, shortness of breath, errors in judgment 200 ppm 0.02% Headache, fatigue, nausea, dizziness 400 ppm 0.04% Severe headache, fatigue, nausea, dizziness, confusion, can be lifethreatening after 3 hours of exposure 800 ppm 0.08% Headache, confusion, collapse, death if exposure is prolonged 1500 ppm 0.15% Headache, dizziness, nausea, convulsions, collapse, death within 1 hour 3000 ppm 0.3% Death within 30 minutes 6000 ppm 0.6% Death within 10-15 minutes 12,000 ppm 1.2% Nearly instant death Injury to the hippocampus causes varying degrees and types of memory loss or memory impairment. The most common is anterograde amnesia (memory dysfunction). Anterograde amnesia usually begins at the time of the exposure. There is difficulty forming new memories. A person can learn and recall how to do simple tasks. A person with severe CO poisoning that has a hippocampus injury will have difficulty remembering the contents of a conversation ten minutes later. Retrograde amnesia causes loss of memory of events from a fixed period in the past. Some affected individuals may suffer the loss of three years worth of memories, while others may be unable to remember a 15-year period. 27 Washington State Department of Labor ,www.Ini.wa.gov/Safety/Topics/AToZ/carbon 74 According to Dr. Raymond Roberge, M.D., depending on the degree of exposure, most victims will have some memory of events that occurred before the onset of amnesia. The cause of death for all of the victims was carbon monoxide intoxication/poisoning. Helms was found near the mouth of the 2nd Left Parallel section and had a carboxyhemoglobin saturation of 78%. The deceased miners found in the barricade had carboxyhemoglobin saturation levels ranging from 64% to 78%. The levels do not appear to be age or size dependent but indicate a trend relative to their distance from the barricade curtains with McCloy, the surviving miner, being the furthest inby. Figure 9 shows the location of the miners in the barricade and their carboxyhemoglobin saturation levels. Figure 9 - Location of Miners and Their Carboxyhemoglobin Levels Self-Contained Self-Rescuers Introduction Section 75.1714 requires the mine operator to make available to each miner an approved self-rescue device, which is adequate to protect the miner for one hour or longer. The operator must provide for training, proper inspection, testing, maintenance and repair of the units. 75 The mine operator supplied a CSE SR100 person-wearable self-contained selfrescuer (SCSR) to each miner. These units were manufactured by the CSE Corporation in Monroeville, Pennsylvania. The MSHA/NIOSH approval number is TC-13F-239. Figure 10 shows the CSE SR-100. Figure 10 - CSE SR-100 The SR-100 provides about 100 liters of usable oxygen for a rated duration of 60 minutes. The unit uses a bi-directional rebreathing system in which the exhaled gas makes multiple passes through a carbon dioxide/oxygen generation canister where carbon dioxide is absorbed and oxygen is generated before the gas can be returned to the user.28 Potassium superoxide (KO2) is used to produce oxygen, as well as absorb carbon dioxide. It is yellow solid but turns a dark grey as it is reacted. Lithium hydroxide (LiOH), which is a white solid, is also used to scrub the carbon dioxide. Figure 11 - Components of the SR-100 SCSR The unit is certified for one hour of operation based on 42 CFR Part 84 and the maximum service life is 10 years. The SR-100 is designed to quickly isolate a miner’s respiratory system from a potentially dangerous atmosphere. It is approved as an escape-only self-contained breathing apparatus and should not be used for rescue, firefighting or underwater breathing. Figure 11 shows the components of the system. Initially, the unit should be removed from the carrying pouch. The tab on the security band is pulled, thereby releasing the band and the top and bottom covers of the unit to open the unit. The manufacturer indicates that after the unit 28 Donning Procedures for Person-Wearable Self Contained Self Rescuer, CSE Corporation. 76 is opened, a properly trained user should be able to activate the oxygen, insert the mouthpiece, and put on the nosepiece in approximately 10 seconds. The oxygen is released from the oxygen cylinder by pulling on the oxygen actuator tag. The miner will hear the faint hiss of the oxygen being released from the cylinder for a few seconds. He should also notice the breathing bag fill. It is important that the mouthpiece plug remains in the mouthpiece during this operation as the oxygen can escape into the atmosphere through the mouthpiece rather than fill the breathing bag. If the breathing bag does not fill for any reason, such as the failure of the compressed oxygen cylinder or the oxygen vents from the unit as stated above, the SCSR can be manually started. This procedure requires that the miner inhale ambient air and exhale into the unit three to six times. The miner then puts on the nose clips to completely close the nostrils, puts on the goggles, adjusts the unit’s straps, replaces his hard hat and evacuates. An SCSR is a closed-circuit breathing apparatus which provides safe, breathable air, independent of the ambient atmosphere. It is designed to be used only for escape from an un-breathable atmosphere. Once donned, an SCSR must not be removed, even to talk, until safety is reached, or its oxygen supply is exhausted. Attempting to conserve, save, or share the oxygen supply in an SCSR, by removing the mouthpiece, may expose a miner to the risk of breathing toxic atmosphere. Re-inserting the mouthpiece, or trying to restart the SCSR provides no means to filter, absorb, or otherwise protect the wearer from what they have already inhaled. Repeated donning, re-inserting the mouthpiece, or trying to restart the SCSR could also interfere with its proper function. The SCSRs performance may be compromised. The SCSR may not restart, or provide protection for its rated duration. Daily Inspection The SCSR must be inspected for damage and for the integrity of its seal each time it is worn or carried by a miner. The unit should be checked daily to insure that it is less than 10 years old, the security band is secure, the top and bottom moisture indicators are blue, the temperature indicator (if applicable) is white, the top and bottom covers are properly aligned, and that there are no signs of significant damage. Any unit that does not meet these criteria must be removed from service. There is no requirement to document the results of the daily inspection, unless the unit is removed from service. 77 90 Day Inspection On November 23, 1998, MSHA and NIOSH notified the mining industry of a potential problem on some CSE SR-100 SCSR devices. MSHA and NIOSH had tested a large number of SCSRs and found that some units produced a higher than normal level of carbon dioxide (CO2). In order to identify the SR-100s most likely to exhibit a higher than normal level of CO2 and remove them from service, an Acoustic Solids Movement Detector (ASMD) test was incorporated into the required 90 day examination of SR-100s. Any unit that failed this test was to be taken out of service.29 The percentage of carbon dioxide in the air affects a miner’s breathing. In air, it is about 0.03%. Miners exposed to about 0.5% carbon dioxide in air may breathe a little deeper and faster, that is, their lung ventilation would increase. Their lung ventilation may double when 3.0% carbon dioxide in air is present. When the carbon dioxide levels in air reaches 10%, a miner may only be able to tolerate it for a few minutes even if he is at rest. A mixture of 10% carbon dioxide in air contains about 18.9% oxygen. 30 SCSRs shall be tested in accordance with approved instructions. The person making the test shall certify by signature and date that the tests were done. The manufacturer states the SR-100 must be tested with the ASMD at least once every 90 days. Training materials are provided by the manufacturer. The test is conducted by attaching the ASMD to the front center of the SR-100. The test can be performed by shaking the unit up and down, several times. A LED indicator on the ASMD will inform the user if the unit passes the test. Any unit which does not pass the test must be removed from service. 31 The ASMD test is a Pass/Fail test. It does not distinguish the precise condition of damage, nor predict to what degree the performance will be degraded. Nor does the ASMD test apply to the functioning of the oxygen startup cylinder. SR-100s that fail the test may not provide an acceptable level of life support. Since the 29 This information is described in more detail in Program Information Bulletin (PIB) No. P99-5 dated April 5, 1999, which is outdated. The PIB states that “although the affected devices will continue to provide protection and miners should not suffer any long term health effects while wearing the device, self-rescue devices that exhibit a higher-thannormal level of CO2 do not conform to the approval requirements and must be removed from service.” The current version of the ASMD Instruction Manual describes the testing methodology for the Daily Visual and 90 day inspection criteria but does not discuss the specific reasons for the test. 30 Mine Gases, Mine Enforcement and Safety Administration, p 11. 31 Daily Visual and 90 day Inspection Criteria, CSE Corporation. 78 extent of damage cannot be known, all SR-100s failing the ASMD test must be removed from service. Training Miners are required to be trained on all types of SCSRs used at the mine, and a record kept of that training. The training must include instruction and demonstration in the use, care and maintenance of an SCSR. The training must also include complete donning procedures. This training must be provided during the Training of New Miners, Experienced Miners Training, or Annual Refresher Training. A review of these records was completed for each miner that was underground on January 2. Recordkeeping Various records were reviewed to determine which SCSR was assigned to each miner. These include the 90 day inspection record of the SCSRs maintained by the mine operator for the Sago Mine and the Spruce Mine, records that were obtained by MSHA inspectors during inspections in 2004 and 2005, purchase orders, and information obtained from the mine operator. It was not always possible to determine which SCSR was assigned to which miner underground on January 2. Additionally, because it was possible for miners to switch SCSRs, it is not possible to conclusively state that a particular SCSR was assigned to, carried by or used by that miner. Evaluation The SCSRs that were recovered were sent to NIOSH’s National Personal Protective Technology Laboratory (NPPTL) in Pittsburgh, Pennsylvania for evaluation. These evaluations were conducted blind, that is, during the evaluation, NPPTL and MSHA personnel had no prior knowledge of the circumstances surrounding the use or deployment of any particular unit, other than the only unopened SCSR belonging to Terry Helms. This evaluation included a visual inspection for any irregularities, such as significant damage. It was not possible to state conclusively that the units evaluated by NIOSH would have passed or failed visual inspection prior to the explosion. For example, NIOSH could not evaluate the condition of the seals and the top and bottom covers, the moisture or temperature indicator, and the security band. The condition of the breathing hose and bag, as well as the condition of the other components was evaluated. An observation of the activation of the start-up oxygen was made. This observation can determine whether the oxygen cylinder was activated, but it cannot determine if the oxygen bottle was full. The evaluation cannot determine if the oxygen exited the unit through an open mouthpiece during the donning process. A visual estimate of the amount of 79 chemical used was completed, and a determination of whether the unit produced oxygen was made on the opened SCSRs. The visual estimate of the amount of chemical used is related to the amount of oxygen the unit produced. This estimate is based on color change of the KO2 in the chemical bed. A performance test on the Breathing and Metabolic Simulator was conducted on the unopened SCSR. In order to supplement the visual estimates of oxygen utilization, chemical analyses were performed at the laboratories of CSE and an independent laboratory, Alternative Testing Laboratories, Inc (ATL). The investigation team requested that CSE utilize a modification of their quality control procedures for analyzing pure KO2 samples to obtain estimates of oxygen utilization. The modification dealt with the handling and preparation of the sample, and did not alter the analytical method used. The procedure was replicated at ATL and witnessed by MSHA, for verification purposes. The results from the three analyses differ due to the fact that they are estimates. Visual estimates took into account the characteristics of the whole chemical bed. This type of analysis is especially useful when most of the chemicals have been utilized and the bed material is “fused” or stuck together. The chemical analyses use a representative sample from the un-fused chemicals. The material was mixed in an attempt to achieve a homogeneous mixture. The results obtained from this procedure should be regarded only as an estimate of the utilization of the chemical bed by the wearer of the SCSR. Exposure of the chemical to moisture or carbon dioxide will be detected as bed usage by this procedure. Exposure can occur between the time the units were worn and the time that the canisters were opened and evaluated by the laboratory. A laboratory test was conducted in 2006 to determine how much of the chemical bed was used when a CSE SR-100 SCSR was activated and left exposed to the atmosphere. During the test, three SR-100s were opened, activated, and left with the mouthpiece plug removed for a 48 hour period. After 48 hours, the units were placed in plastic bags. These units were about 1 year old. The relative humidity in the atmosphere was 40% to 53% and the temperature was 70 to 75 degrees Fahrenheit. The units were then tested by CSE. The results of the test indicated that approximately 3.3%, 3.1% and 1.5% of the chemicals in the unit had been depleted. Miners Working Outby 1st Left At the time of the explosion, John Boni, Pat Boni and Jamison were working in outby areas. After the explosion, they evacuated the mine without donning their SCSRs. Records indicated which SCSR was assigned to each of the three miners. It was not possible to state conclusively that each miner was carrying the SCSR that had been assigned to him at the time of the accident. The mine operator’s 80 records indicated that the 90 day inspection was completed for two of the three SCSRs. One of the three miners had received training on the SCSR within one year. John Boni stated that he had signed a training form without taking the training. Pat Boni indicated that he did not don the unit as part of the training, and his training was given by a person who is not listed as an approved instructor for Experienced Miner Training. Records indicated the last Experienced Miner Training Pat Boni received from an approved trainer occurred more than one year prior to the accident. John Boni - The mine operator’s inspection records indicated SCSR 106186 was assigned to John Boni. The last 90 day inspection occurred on November 14, 2005. This SCSR was manufactured in July of 2004. During the evacuation, John Boni did not don his SCSR. He was near the pump at 22 Crosscut, No. 3 Belt when the explosion occurred. He indicated “there was no smoke or --- there was, like I said, mainly rock dust in the area that I was in.” He stated that his last training on the unit “would have been probably a year and a half ago.” He indicated that he missed a scheduled training class, but signed a form stating that he had received the training. The records show that he had Experienced Miner Training at the Sago Mine on October 11, 2004. The training form indicated, “Hands on SCSR/Tour.” The records showed he had Annual Refresher Training at the Sago Mine on October 7, 2005. Pat Boni - The mine operator’s inspection records indicated SCSR 100991 was assigned to Pat Boni and the last 90 day inspection occurred on November 16, 2005. Although the unit was manufactured in December of 2003, the mine operator’s records show it was manufactured in November of 2004. Pat Boni was in the belt entry near No. 4 Belt drive when the explosion occurred. During the evacuation, he did not don his SCSR. He stated that he “knew I was in good air,” and that there was never a time that he smelled or saw smoke. He stated that he had training on the unit. However, he did not don the SCSR as part of the training exercise. He stated, “he showed us how to do it.” The records show that Pat Boni had Experienced Miner Training at the Sago Mine on December 29, 2004. The training form indicated, “Hands on SCSR/Tour.” The records show that he had Experienced Miner Training at the Sago Mine on July 5, 2005. The training form was signed by a person who was not listed as an approved instructor for Experienced Miner Training. Jamison - The Spruce Mine inspection records indicated SCSR 91947 was assigned to James F. Jamison. The 90 day inspection occurred on February 14, 2005. It was manufactured in March of 2002. There were no records from the Sago Mine indicating which SCSR was assigned to Jamison, or that the 90 day inspection was ever conducted. He was near the No. 2 Belt drive when the explosion occurred. During the evacuation, Jamison did not don his SCSR. He indicated that he did not observe any smoke or feel any heat. He stated, “I had it 81 in my hand. I just was making sure. You know, I had it ready to go.” The records indicate that he had Annual Refresher Training at the Spruce Fork Mine No. 1 on January 28, 2005. The training form indicated, “Hands on CSE.” The records indicate that he had Experienced Miner Training at the Sago Mine on June 27, 2005. Miners on the 1st Left Mantrip At the time of the explosion, Denver Anderson, Avington, Carpenter, Grall, Helmick, Hess, Owen Jones, Keith, Perry, Rowan, Ryan, Tenney, and Wamsley were on the 1st Left mantrip at the 1st Left switch. After the explosion, they encountered dust, smoke and other contaminants. Seven of the thirteen miners eventually donned their SCSRs as they evacuated the mine. Records indicated which SCSRs were assigned to twelve of the thirteen miners, and the records for the remaining miner indicated that he was assigned a different SCSR than was recovered. It was not possible to state conclusively which SCSR was carried or used by the miners at the time of the accident. The mine operator’s records indicated that the 90 day inspection was completed for eight of the thirteen SCSRs, including four of the seven miners who had donned their SCSRs. Although records indicated that twelve of the thirteen miners had received training on the SCSR within one year, the remaining miner had received training in December of 2005. The type of training was not indicated on the training form. His training was given the same day the other miners at the Sago Mine received Experienced Miner Training. One miner indicated he had difficulty removing his SCSR from his pouch. One miner indicated he had difficulty opening his unit. All of the seven miners who donned SCSRs reported that the units worked, but three indicated they had some type of difficulty. NIOSH evaluated three of the seven units and indicated the SCSRs activated and produced oxygen. Denver Anderson – The Spruce Mine inspection records indicated SCSR 83566 was assigned to Denver Anderson. The 90 day inspection occurred on February 14, 2005. Although the unit was manufactured in May of 2001, the Spruce Mine records show it was manufactured in May of 2004. There were no records from the Sago mine indicating which SCSR was assigned to Anderson, or that the 90 day inspection was completed. Anderson donned his SCSR shortly after he exited the 1st Left mantrip, and had no difficulty doing so. Hess assisted Anderson and stated “his rescuer had BlocBond on it and he was having trouble with where it was on his belt, getting it up out of the pouch. So he had the channel locks down in his pouch, too, so I pulled those out and of course, you know, I'm beside him so I kept my hands under it and got it pushed up out.” Anderson put his rescuer on, “because of all the smoke and that.” The unit worked as expected. He continued to use the unit until he got in the mantrip to evacuate the mine. As SCSR 83566 was not recovered, there are no NIOSH 82 evaluation results available. The records show he had Experienced Miner Training at the Sago Mine on December 8, 2005. Paul Avington – The Spruce Mine inspection records indicated SCSR 63277 was assigned to Paul Avington. The 90 day inspection occurred on February 14, 2005. It was manufactured in July of 1998. There were no records from the Sago Mine indicating which SCSR was assigned to Avington, or when the 90 day inspection was completed. During the evacuation, Avington did not don his SCSR. He indicated, “I just didn’t think I needed it.” The records show he had Annual Refresher Training at the Spruce Fork Mine on August 27, 2004. The training form indicated “CSE 100.” The records show he had Experienced Miner Training at the Sago Mine on December 8, 2005. Gary Carpenter - The Spruce Mine inspection records indicated SCSR 75648 was assigned to Gary Carpenter. The 90 day inspection occurred on February 14, 2005. It was manufactured in March of 2000. There were no records from the Sago Mine indicating which SCSR was assigned to Carpenter or whether the 90 day inspection was performed. During the evacuation, Carpenter did not don his SCSR. He stated, “We never really discussed, you know, but there was an explosion because of the air and the debris. It was just kind of obvious.” The records show he had Annual Refresher Training at the Spruce Fork Mine on August 20, 2004, and Experienced Miner Training at the Sago Mine on December 8, 2005. Ron Grall - The mine operator’s inspection records indicated that SCSR 92943 was assigned to Grall and that the most recent 90 day inspection was performed on November 17, 2005. The unit was manufactured in May of 2002. During the evacuation, Grall did not don his SCSR. He indicated, “[t]he reason I didn't put mine on is because I didn't smell any smoke. I could smell --- the taste of dust, sulfur taste, but you couldn't see --- couldn't taste no --- smell no smoke or anything so I figured as long as I could breathe, I wasn't putting mine on.” He said that training should be held more often, “the self-rescuer, they need to do that more frequently. I mean, because once a year, you kind of forget that stuff.” The records show he had Annual Refresher Training at the Spruce Fork Mine on August 21, 2004, and Experienced Miner Training at the Sago Mine on September 16, 2005. Randall Helmick - The mine operator’s inspection records indicated that SCSR 56505 was assigned to Helmick, and that the last 90 day inspection took place on November 15, 2005. The unit was manufactured in September of 1997. During the evacuation, Helmick did not don his SCSR because he was saving it. He stated, “I didn't put mine on because I was still breathing. You know, I didn't feel like I was having any difficulty of breathing. And we didn't know if, you know, we was going to have a second explosion or what.” The records show he 83 had Annual Refresher Training at the Spruce Fork Mine No 1 on April 2, 2004. The training form indicated, “Hands on SCSR.” The records show he had Experienced Miner Training at the Sago Mine on December 8, 2005. Eric Hess - The mine operator’s inspection records indicated that SCSR 88170 was assigned to Hess and the 90 day inspection was completed on November 14, 2005. Although the unit was manufactured in December of 2001, the mine operator’s records show it was manufactured in December of 2002. After the explosion, he exited the 1st Left mantrip and traveled outby where he checked and found that he did not have fresh air in the primary intake escapeway. He then donned his SCSR. He had no difficulty in donning it, and it worked as expected. He continued to wear the unit until he encountered clear air at approximately 26 Crosscut, No. 4 Belt. SCSR 88170 was not one of the units recovered for evaluation. Therefore, there were no evaluation results available. The records show that Hess had Annual Refresher Training at the Spruce Fork Mine on April 16, 2004. The training form indicated, “Hands on SCSR.” The records show he had Experienced Miner Training at the Sago Mine on October 6, 2004, but this training was received more than a year prior to the date of the accident. While the records indicate that Hess had some type of training at the Sago Mine on December 8, 2005, the type of training was not marked on the 5000-23 form required by MSHA. However, as many other miners received Experienced Miner Training on December 8, 2005, it is likely that this is the type of training that Hess received. Owen Jones - The mine operator’s inspection records indicated that SCSR 92933 was assigned to Jones and that the 90 day inspection was done on November 14, 2005. The unit was manufactured in June of 2002. Jones was the section foreman for the 1st Left crew. During the evacuation, Jones did not don his SCSR. He stated, “I should have, but I didn't.” He stated, “my carbon monoxide detector went off immediately after the explosion.” Jones went to the doctor the following week due to an odd feeling in his chest. A blood test revealed that he had a “high level of carbon monoxide.…” He did not evacuate to the surface with his crew but rather stayed near the phone near 37 Crosscut, No. 4 Belt. He met with Jeffrey Toler, Wilfong, and Schoonover, and they installed ventilation controls up to 57 Crosscut, No. 4 Belt. The records show he had Annual Refresher Training at the Sago Mine on March 18, 2005. The training form indicated, “Hands on SCSR.” The records indicate that he had Experienced Miner Training at the Sago Mine on December 8, 2005. Hoy Keith - The Spruce Mine inspection records indicated SCSR 60035 was assigned to Hoy S. Keith, Jr. The 90 day inspection occurred on February 14, 2005. It was manufactured in January of 1998. There were no records from the Sago Mine indicating which SCSR was assigned to Keith or if the 90 day inspection was completed. Keith donned his SCSR shortly after he exited the 1st 84 Left mantrip. He indicated “I was just a little bit disoriented whenever it happened” and other miners that already had donned their SCSR assisted him, including Wamsley. Wamsley stated,” I helped him get it on, around his neck, nose clips on, everything. I pulled the thing and it didn't activate.” When Wamsley was asked if Keith tried to start the unit manually, he stated, “No. I don't even know if he had enough wind to do that.” However, Keith indicated that it worked as expected. Rowan indicated that Keith was having difficulties breathing and he stayed with him, sharing his SCSR with Keith as they evacuated the mine. Rowan said that Keith’s SCSR appeared to be working but he could not tell for sure, and that Keith was upset with the situation. Rowan said, “I'm not sure that he actually even had any trouble with his. Like I said, he just kind of --- I know that the bag was out on his and everything like that. I mean, it looked like it was working.” Keith indicated he continued to wear the unit until he got to fresh air and entered the mantrip to evacuate the mine. SCSR 60035 was not recovered, therefore there were no NIOSH evaluation results available. The records show that he had Annual Refresher Training at the Spruce Fork Mine No 1 on August 20, 2004, and Experienced Miner Training at the Sago Mine on December 8, 2005. Arnett Perry – The mine operator’s inspection records indicated that SCSR 102138 was assigned to Perry and the unit had a 90 day inspection on November 15, 2005. The manufacture date was January of 2004. He exited the mantrip and did not don it immediately “Because that's all I could remember, one hour. And I thought; well, now I've been told it takes two hours to walk out of here.” He traveled outby and got into the intake entry. He then donned his SCSR. Ryan assisted him. Perry said, “I suppose it worked all right. Other than I was trying to breathe too hard and it sucked the bag together.” He did not believe that he pulled the oxygen activator tag. Instead, he manually started the unit. “Every little bit, I was taking it (the mouthpiece) out because I wasn’t getting enough air it seemed like.” Ryan stated, “Got it open, got the bag and everything out, he (Perry) got it in his mouth, put his nose clips on, I activated it, the bag blew wide open. Within a block, the bag collapsed. He couldn't breathe. He had to take it out of his mouth. And I tried to get him to leave it in his mouth and just breathe with what he could get, but he said he couldn't breathe, so he took it off.” Perry removed the mouthpiece from his mouth when he arrived at the mantrip. The records show he had Experienced Miner Training at the Sago Mine on December 8, 2005. SCSR 102138 was recovered and was evaluated by NIOSH. The dust shield had some cracks, but the canister was not dented. The damage was not significant. NIOSH established that the start-up oxygen was activated, that the unit produced oxygen, and that approximately 20% of the chemicals in the unit 85 were used. The goggles were attached to the unit.32 CSE and ATL conducted chemical analyses of the unit. They reported that approximately 29% and 31% of the chemicals in the unit had been depleted, respectively. Gary Rowan - The mine operator’s inspection records indicated that SCSR 59965 was assigned to Rowan and the last 90 day inspection was performed on November 14, 2005. The manufacture date was February of 1998. However, SCSR 59965 was not recovered. MSHA inspection records for the Sago Mine from June of 2004 indicated that SCSR 86537 was assigned to Rowan. The record indicated the unit was manufactured in September of 2003. The manufacture date was September of 2001. There are no records from the Sago Mine indicating that the 90 day inspection was completed for that unit. He exited the mantrip, traveled outby into the intake entry and donned his SCSR. He stated, “I should have put it on as soon as it happened.” He did not have any difficulty in donning his SCSR, and it worked as expected. During the evacuation, he assisted Keith and stated, “ I took my mouthpiece out and let him take, you know, deep breaths so he could take some air and stuff out of mine and stuff because he said he wasn’t sure that his was working or not.” He indicated he left his SCSR on until he got outside. The records show that Rowan had Annual Refresher Training at the Sago Mine on March 18, 2005. The training form indicated, “Hands on SCSR.” The records show he had Experienced Miner Training at the Sago Mine on December 8, 2005. SCSR 86537 was recovered, and was evaluated by NIOSH. “Gary Rowe” was written on the unit. The dust shield had some cracks but the canister was not dented. The damage was not significant. NIOSH reported that the start-up oxygen was activated, the unit produced oxygen and approximately 10% of the chemicals in the unit were used. CSE and ATL conducted chemical analyses of the unit. They reported that approximately 19% and 28% of the chemicals in the unit had been depleted, respectively. Harley Joe Ryan - The mine operator’s inspection records indicated that SCSR 97144 was assigned to Ryan and the last 90 day inspection was done on November 16, 2005. The manufacture date was December of 2003. He exited the mantrip and walked outby where he donned his SCSR. Wamsley assisted him. Ryan stated, “You just couldn't get the tab off. You couldn't get ahold of it for one thing. ....The bottom part of mine, we had to jerk on it two or three times to get it to unseal.” He had difficulty with the mouthpiece, “You can't hold something in your mouth if you don’t have teeth that's designed for something to hold with your teeth. What they're going to do about that, I don't know. I kept it in my mouth. I had trouble keeping it in, but I kept it in.” He further 32 The goggles may have been placed there by the evidence teams as they were recovered. 86 stated, “I just know I was with him, walking with him when mine started getting to the point I couldn't breathe real good with it...But I was going slow enough with Roger that I wasn't asking this thing for more than what I was getting out of it.” and “I knowed it was overriding what was left in it. And I would rather breathe what it was giving me than the air that was out there.” He indicated he did not remove the unit until he was outside the mine. The records show that Ryan had Annual Refresher Training at the Sago Mine on January 27, 2005. The training form indicated, “Hands on SCSR.” The records show that he had Experienced Miner Training at the Sago Mine on December 8, 2005. SCSR 97144 was recovered and evaluated by NIOSH. The dust shield had some cracks but the canister was not dented. The damage was not significant. There was foreign matter in the mouthpiece that appeared to be snuff, but the breathing tube was not obstructed. According to NIOSH, the start-up oxygen was activated, the unit produced oxygen, and approximately 40% to 50% of the chemicals in the unit had been used. CSE and ATL conducted chemical analyses of the unit. They reported that approximately 42% and 48% of the chemicals in the unit had been depleted, respectively. Christopher Tenney - The mine operator’s inspection records indicated that SCSR 52409 was assigned to Tenney and the 90 day inspection was completed on November 14, 2005. The manufacture date was June of 1997. During the evacuation, Tenney did not don his SCSR. He stated, “well, actually I wasn't having any trouble breathing and I didn't know what we were going to encounter further out and I don't know what I was thinking, I guess maybe save it in case I needed it at a later point.“ The records show that Tenney had Annual Refresher Training at the Sago Mine on March 18, 2005. The training form indicated, “Hands on SCSR.” The records show he had Experienced Miner Training at the Sago Mine on December 8, 2005. Anton Wamsley - The mine operator’s inspection records indicated that SCSR 88981 was assigned to Wamsley, with the last 90 day inspection occurring on November 17, 2005. The unit was manufactured in November of 2001. He exited the mantrip, traveled outby and found that he did not have clear air in the primary intake escapeway. He donned his SCSR with no difficulty, it worked as expected, and he indicated that he kept it on until he got almost outside. SCSR 88981 was not one of the units recovered for evaluation. Therefore, there were no evaluation results available. The records show that Wamsley had Annual Refresher Training at the Sago Mine on February 25, 2005, and Experienced Miner Training at the Sago Mine on December 8, 2005. Miner Working Near the Mouth of 2nd Left Parallel Terry Helms – SCSR 90223 was found with Helms. The Spruce Mine inspection records indicated SCSR 90223 was assigned to Terry Helms and the 90 day 87 inspection occurred on February 12, 2005. There were no records from the Sago Mine indicating that the 90 day inspection was completed for an SCSR belonging to Terry Helms, but there were records for SCSR 90223. The Sago Mine inspection records indicated that this unit was not assigned to any miner, and that the 90 day inspection was completed in a timely manner, with the last inspection occurring on November 14, 2005. The manufacturing date was December of 2001. Helms did not don his SCSR. The records show he had Annual Refresher Training at the Spruce Fork Mine #1 on April 21, 2005. The training form indicated, “Hands on SCSR.” The records show he had Experienced Miner Training at the Sago Mine on June 27, 2005. SCSR 90223 was recovered. It was evaluated by NIOSH. The dust shield had cracks but the canister was not dented. Some pieces of the dust shield were missing. The damage was significant. “Terry Helms” was written on the unit. It did not pass the ASMD test. The mine operator’s inspection records indicated the unit passed the ASMD test on November 14, 2005. It was not possible to state conclusively whether this unit would have passed or failed these required tests prior to the explosion. The SCSR was tested on a breathing simulator. It did not have sufficient start-up oxygen to fill the breathing bag. However, the unit did pass the breathing simulator test when started manually. It operated for 64 minutes before being fully consumed. Miners on 2nd Left Parallel At the time of the explosion, Thomas Anderson, Alva Bennett, James Bennett, Groves, Hamner, Jesse Jones, Lewis, McCloy, Martin Toler, Ware, Weaver, and Winans were on the 2nd Left Parallel section. They tried to evacuate and eventually donned their SCSRs. McCloy indicated that Jones, Anderson, Toler, and Groves felt that their units were not functioning. Records from the Sago Mine and the Spruce Mine indicated which SCSRs were assigned to ten of the twelve miners, and the mine operator provided information on the assignment of SCSRs for the remaining two miners. However, it was not possible to state conclusively that the SCSRs were carried and used by the miners to whom they were assigned at the time of the accident. The mine operator’s records indicated that the 90 day inspection was completed for six of the twelve SCSRs. The mine records indicate that only one of the four SCSRs that McCloy stated did not function as intended had been inspected within the previous 90 days. The SCSR assigned to one miner was more than 10 years old. Records indicated that all of the twelve miners had received SCSR training within the previous year. All of the twelve SCSRs recovered were evaluated by NIOSH. These evaluations indicated that they all were activated and produced oxygen. The visual observations indicated that eight of the twelve units had depleted approximately 30% or less of the chemicals. 88 Thomas Anderson – The Spruce Mine inspection records indicated SCSR 92652 was assigned to Tom Anderson and the 90 day inspection occurred on February 14, 2005. There were no records from the Sago Mine indicating any SCSR for Anderson or that the 90 day inspection was completed for SCSR 92652. Although the unit was manufactured in March of 2002, the Spruce Mine inspection records show it was manufactured in March of 2003. This unit was found opened in the barricade, about two feet from the victim. McCloy indicated that Anderson had problems with his SCSR. The records show he had Annual Refresher Training at the Spruce Fork Mine on August 13, 2004, and Experienced Miner Training at the Sago Mine on December 8, 2005. SCSR 92652 was recovered. NIOSH evaluations indicated that the goggles were attached to the unit33, the dust and heat shield were off of the unit, the dust shield was cracked with pieces missing, the pads were missing on the nose clips, there was a tear on the breathing tube close to the saliva trap, the top bushing was displaced threequarters of the way down the unit and the bottom bushing was cut and dislodged but in place. The bottom corner of the canister was damaged, there was dirt on the breathing bag and there were possible signs of an inward leak of dirt onto the bag. The damage was significant. NIOSH reported that the start-up oxygen was activated, the unit produced oxygen and approximately 40% of the chemicals in the unit were used. CSE and ATL conducted chemical analyses of the unit. They reported that approximately 39% and 48% of the chemicals in the unit had been depleted, respectively. Alva Bennett – The Spruce Mine inspection records indicated SCSR 89765 was assigned to Alva M. Bennett and the 90 day inspection occurred on February 12, 2005. There were no records from the Sago Mine for any SCSR for Alva Bennett or that the 90 day inspection was completed for SCSR 89765. The unit was manufactured in December of 2001. This unit was found opened in the barricade, about 17 feet from the victim. The records show he had Annual Refresher Training at the Spruce Fork Mine No. 1 on April 21, 2005. The training form indicated, “Hands on SCSR.” The records show he had Experienced Miner Training at the Sago Mine on December 8, 2005. SCSR 89765 was recovered. It was evaluated by NIOSH. The dust shield had cracks but the canister was not dented. The damage was significant. The laboratory also indicated that the mouthpiece plug was tied to the bottom of unit, the neck strap and part of the waist strap were missing, the dust shield was broken and the heat shield was damaged, and the bottom filter showed evidence of mineralization. Mineralization occurs when some of the chemical contained in the breathing bag is dissolved in water and is re-deposited in a fine layer on the bottom of the filter. NIOSH reported that the start-up oxygen was activated, the unit produced 33 The goggles may have been placed there by the evidence teams as they were recovered. 89 oxygen and approximately 25% of the chemicals in the unit were used. CSE and ATL conducted chemical analyses of the unit. They reported that approximately 38% and 54% of the chemicals in the unit had been depleted, respectively. James Bennett – The mine operator indicated that SCSR 56495 was assigned to Bennett. The manufacture date was November of 1997. The Spruce Mine inspection records indicated SCSR 89203 was assigned to Jim Bennett and the 90 day inspection occurred on February 14, 2005. Those records also indicated that SCSR 56495 belongs to another miner. There were no records from the Sago Mine for SCSR 56495 or for any SCSR for Bennett in the 90 day inspection record. SCSR 56495 was found opened in the barricade about 17 feet from the victim. The records show that he had Annual Refresher Training at the Spruce Fork Mine #1 on April 22, 2005, and Experienced Miner Training at the Sago Mine on July 27, 2005 and on December 8, 2005. SCSR 56495 was recovered. NIOSH evaluations indicated the dust shield had some cracks and the canister had some dents, but they were not severe. The damage was not significant. The breathing tube was set but was pliable and open. “Smargo” was written on the unit. There was green paint on the unit. NIOSH’s report indicated that the start-up oxygen was activated, the unit produced oxygen and approximately 25% to 35% of the chemicals in the unit were used. Jerry Groves – The Spruce Mine inspection records indicated SCSR 57878 was assigned to Jerry Groves and the 90 day inspection occurred on February 14, 2005. There were no records from the Sago Mine for SCSR 57878 or for any SCSR for Jerry Groves in the 90 day inspection period. Although the unit was manufactured in December of 1997, the Spruce Mine inspection records show it was manufactured in February of 1997. SCSR 57878 was found opened in the barricade about 10 feet from the victim. McCloy indicated that Grove’s unit was not functional. He stated that the breathing bag did not inflate when the unit was opened, or when attempts were made to start the unit manually. The records show he had Annual Refresher Training at the Spruce Fork Mine on August 27, 2004. The training form indicated “CSE.” The records show he had Experienced Miner Training at the Sago Mine on December 8, 2005. SCSR 57878 was recovered. NIOSH evaluations indicated that the dust shield had some cracks and the canister had some dents, but they were not significant. The damage was not significant. The laboratory also indicated that the breathing tube was set but was pliable and open, there was a cut in the top canister bushing and a blister on the neck of the breathing bag, the bottom of the canister showed mineralization, and there was paint on the dust shield. NIOSH reported that the start-up oxygen was activated and that the unit produced oxygen, and estimated that approximately 40% to 50% of the chemicals in the unit were used. CSE and ATL conducted chemical analyses of the unit. They reported that approximately 48% and 77% of the chemicals in the unit had been depleted, respectively. 90 George Hamner – The mine operator indicated that SCSR 101868 was assigned to Hamner, but MSHA inspection records from June of 2004 indicated that SCSR 101868 was assigned to another miner. The manufacture date was January of 2004. There were no records from the Sago Mine indicating that the 90 day inspection was completed for SCSR 101868. The mine operator’s inspection records indicated that a different unit, SCSR 101838, was assigned to Hamner and that the 90 day inspection was completed in a timely manner, with the last inspection occurring on November 18, 2005. SCSR 101838 was not recovered. SCSR 101868 was found opened in the barricade about 49 feet from the victim. The records show he had Annual Refresher Training at the Sago Mine on June 24, 2005. The training form indicated, “Hands on SCSR.” The records show he had Experienced Miner Training at the Sago Mine on December 8, 2005. SCSR 101868 was recovered. NIOSH evaluations indicated the dust shield had some cracks and the canister was dented. The damage was not significant. “Walker” was written on the unit. NIOSH indicated that the start-up oxygen was activated, the unit produced oxygen and approximately 25% of the chemicals in the unit were used. CSE conducted a chemical analysis of the unit. It reported that approximately 31% of the chemicals in the unit had been depleted. Jesse Jones – The mine operator’s inspection records indicated that SCSR 46433 was assigned to Jones and that the 90 day inspection was completed in a timely manner, with the last inspection occurring on November 17, 2005. Although the unit was manufactured in August of 1995, the mine operator’s inspection records show it was manufactured in August of 1996. The unit should have been taken out of service on its 10 year anniversary, almost five months prior to the accident. This unit was found opened in the barricade about 27 feet away from the victim. McCloy indicated that Jones had problems with the SCSR. The mine records show that Jones had Annual Refresher Training at the Sago Mine on March 18, 2005. The training form indicated, “Hands on SCSR.” The records indicate that he had Experienced Miner Training at the Sago Mine on March 22, 2004. The training form indicated, “Hands on SCSR.” SCSR 46433 was recovered. NIOSH evaluations indicated the dust shield had cracks but the canister was not dented. The damage was significant. The unit would not pass the visual exam because it was past its service date. The breathing tube was set but pliable and open, the heat shield was loaded with dirt and there was a stain on the breathing bag at the lanyard tie point. NIOSH reported that the start-up oxygen was activated, the unit produced oxygen and approximately 10% to 20% of the chemicals in the unit were used. CSE and ATL conducted chemical analyses of the unit. They reported that approximately 41% and 50% of the chemicals in the unit had been depleted, respectively. David Lewis – The mine operator’s inspection records indicated that SCSR 101831 was assigned to Lewis and the 90 day inspection was completed in a timely manner, with the last inspection occurring on November 18, 2005. The 91 manufacture date was January of 2004. SCSR 101831 was found opened in the barricade, about 10 feet from the victim. The records show he had Annual Refresher Training at the Sago Mine on April 22, 2005. The training form indicated “Hands on SCSR.” The mine records show he had training at the Sago Mine on December 8, 2005, but the type of training is not marked on the 5000-23 form. However, as many other miners received Experienced Miner Training on December 8, 2005, it is likely that this is the type of training that Lewis received. The mine records show he had Experienced Miner Training at the Sago Mine on December 15, 2005. SCSR 101831 was recovered. NIOSH evaluations indicated the dust shield was cracked and the canister was dented. The damage was not significant. NIOSH indicated that the start-up oxygen was activated, that the unit produced oxygen and that approximately 10% to 20% of the chemicals in the unit were used. CSE conducted a chemical analysis of the unit, and reported that approximately 25% of the chemicals in the unit had been depleted. Randal L. McCloy Jr. – The mine operator’s inspection records indicated that SCSR 106154 was assigned to McCloy and the 90 day inspection was completed in a timely manner, with the last inspection occurring on November 18, 2005. The manufacture date was July of 2004. SCSR 106154 was found opened in the barricade, about 21 feet from where he was found. McCloy stated that his unit “worked fine.” The records show he had Annual Refresher Training at the Sago Mine on August 19, 2005, and Experienced Miner Training at the Sago Mine on December 8, 2005. SCSR 106154 was recovered. NIOSH evaluations indicated the dust shield had no cracks and the canister was not dented. The damage was not significant. There was green paint on the breathing tube. The laboratory also indicated that the lenses on the goggles were displaced and the relief valve was sticking closed. Although the sticking relief valve could have eventually affected the performance, it was not likely to affect the initial performance. It may have occurred after the unit was used and may not conclusively reflect the condition of the unit prior to the explosion. NIOSH reported that the start-up oxygen was activated, the unit produced oxygen, and approximately 20 to 25% of the chemicals in the unit were used. CSE and ATL conducted chemical analyses of the unit, and reported that approximately 28% and 29% of the chemicals in the unit had been depleted, respectively. Martin Toler Jr. – The Spruce Mine inspection records indicated SCSR 57604 was assigned to Martin Toler and the 90 day inspection occurred on February 14, 2005. There were no records from the Sago Mine indicating any SCSR for Martin Toler or that the 90 day inspection was completed for SCSR 57604. There were also other records from the Sago Mine indicating that SCSR 106022 was assigned to “Toler JR” and that the 90 day inspection was completed in a timely manner, with the last inspection occurring on November 17, 2005. The manufacture date was December of 1997. SCSR 57604 was found opened in the barricade, about 32 feet from the victim. McCloy indicated that Martin Toler had problems with the 92 SCSR. Martin Toler may have been confronted with a situation in which the miners felt they did not have enough working SCSRs to escape through the heavy smoke. McCloy stated that Toler said “this ain’t safe like this. Let’s go back to the section.” SCSR 106022 was not recovered. The records show he had Annual Refresher Training at the Spruce Fork Mine on August 27, 2005 and Experienced Miner Training at the Sago Mine on December 8, 2005. SCSR 57604 was recovered. NIOSH evaluations indicated the dust shield had some cracks but the canister was not dented. The damage was marginal. Some pieces of the dust shield were missing, the breathing tube was set but was pliable and open, and there was a stain on the breathing bag at the lanyard tie point. NIOSH indicated that the start-up oxygen was activated, the unit produced oxygen, and approximately 10% to 15% of the chemicals in the unit were used. CSE and ATL conducted chemical analyses of the unit. They reported that approximately 21% and 27% of the chemicals in the unit had been depleted, respectively. Fred Ware – The mine operator’s inspection records indicated that SCSR 56880 was assigned to Ware and the 90 day inspection was completed in a timely manner, with the last inspection occurring on November 17, 2005. The manufacture date was October of 1997. SCSR 56880 was found opened in the barricade near the victim. The records show that Ware had Annual Refresher Training at the Sago Mine on March 25, 2005. The training form indicated, “Hands on SCSR.” The records show he had other training at the Sago Mine on December 8, 2005. The type of training was not marked on the 5000-23 form. However, as many other miners received Experienced Miner Training on December 8, 2005, it is likely that this is the type of training that Ware received. SCSR 56880 was recovered. NIOSH evaluations indicated the dust shield had some cracks but the canister was not dented. The damage was marginal. There was tape around the relief valve, and the breathing tube was set but was pliable and open. “Fred Ware Jr.” was written on the unit. NIOSH evaluations showed that the start-up oxygen was activated, the unit produced oxygen and approximately 10% to 20% of the chemicals in the unit were used. CSE and ATL conducted chemical analyses of the unit, and reported that approximately 39% and 41% of the chemicals in the unit had been depleted, respectively. Jackie Weaver – The mine operator’s inspection records indicated that he was assigned SCSR 57334. There were records indicating the 90 day inspection was completed in a timely manner, with the last inspection occurring on November 16, 2005. The manufacture date was December of 1997. SCSR 57334 was found opened in the barricade, about 19 feet from the victim. The records show that Weaver had Annual Refresher Training at the Sago Mine on October 14, 2005. The training form indicated, “Hands on SCSR.” The records show he had Experienced Miner Training at the Sago Mine on December 8, 2005. SCSR 57334 was recovered. NIOSH evaluations indicated the dust shield had cracks but the canister had no dents. The damage was significant. The breathing tube was set 93 but was pliable and open, the tag was missing on the lanyard for the firing lever, the dust shield was broken and cracked and the heat shield was loaded with dirt. There was rust at the relief valve lanyard attachment point to the breathing bag but the lanyard attachment was still solid. NIOSH reported that the start-up oxygen was activated, the unit produced oxygen and approximately 30% of the chemicals in the unit were used. CSE and ATL conducted chemical analyses of the unit. They reported that approximately 39% and 34% of the chemicals in the unit had been depleted, respectively. Marshall Winans – The mine operator’s inspection records indicated that he was assigned SCSR 52478. There were records indicating the 90 day inspection was completed in a timely manner and that the last inspection occurred on November 14, 2005. The manufacture date was June of 1997. SCSR 52478 was found opened in the barricade about 27 feet from the victim. The records show that he had Annual Refresher Training at the Spruce Fork Mine on August 13, 2004 and Experienced Miner Training at the Sago Mine on December 8, 2005. SCSR 52478 was recovered. NIOSH evaluations indicated the dust and heat shields were missing, and the canister had dents. The damage was significant. The breathing tube was set but was pliable and open, the upper bushing was missing, there was staining on the breathing bag at the lanyard tie point, the nose clips were stuck together, and there was evidence of moisture in the bottom filter. NIOSH reported the start-up oxygen was activated, the unit produced oxygen and 50% to 60% of the chemicals in the unit were used. CSE and ATL conducted chemical analyses of the unit. They reported that approximately 72% and 87% of the chemicals in the unit had been depleted, respectively. Miners Attempting Rescue Effort After the explosion, Hofer, Schoonover, Jeffrey Toler, and Wilfong entered the mine. They did not use their SCSRs. Records indicated which SCSRs were assigned to each of the four miners. However, it was not possible to state conclusively that an SCSR was carried by the miner to whom it was assigned at the time of the accident. The mine operator’s records indicated that the 90 day inspection was completed for three of the four SCSRs. The records indicated three of the four miners had received training on the SCSR within the past year. Vernon Hofer – The mine operator’s inspection records indicated he was assigned SCSR 63274. There were records indicating that the 90 day inspection was completed in a timely manner, with the last inspection occurring on November 17, 2005. The manufacture date was October of 1998. Hofer also entered the mine after the explosion. He did not don his SCSR. He stated,” I wasn’t having trouble breathing. I wasn’t --- didn't notice any adverse effects from the conditions that we were working in….” The records show he had Annual Refresher Training at the Sago Mine on February 28, 2005. The training 94 form indicated, “Hands on SCSR.” The records indicate that he had Experienced Miner Training at the Sago Mine on February 4, 2004. James Allen Schoonover– The mine operator’s inspection records indicated he was assigned SCSR 104889. There were records indicating the 90 day inspection was completed in a timely manner, with the last inspection occurring on November 17, 2005. The manufacture date was June of 2004. Schoonover entered the mine with Toler after the explosion. He stated, “We would repair whatever, whatever stopping it was and the detector, of course, it would go down, it wouldn't have any alarm. It would advance. You know, you could --your detector would go off again, get a piece of curtain hung, and we would bring the fresh air behind us.” He did not don his SCSR. He stated, “Because I felt there was no need to at that time.” Schoonover was responsible for the training at the mine. The records show he had Annual Refresher Training at the Spruce Fork Mine on August 9, 2002 and Experienced Miner Training at the Sago Mine on January 12, 2004. The records indicated that it had been over a year since Schoonover received SCSR Training. Jeffrey Toler – The mine operator’s inspection records indicated he was assigned SCSR 104831. There are records indicating that the 90 day inspection was completed in a timely manner, with the last inspection occurring on November 14, 2005. The manufacture date was June of 2004. Jeffrey Toler entered the mine after the explosion. Along with the others, he repaired damaged ventilation controls to re-establish airflow. He did not don his SCSR. “With us keeping fresh air with us, I never felt like we were in a concentration of CO that I felt that I needed it,” he stated. When he was in the track entry at 49 Crosscut, No. 4 Belt, he stated, “I think it (the concentration of carbon monoxide) was in excess of 700 parts per million at that point.” The records show he had Experienced Miner Training at the Sago Mine on August 2, 2005. Denver Wilfong – The Spruce Mine inspection records indicated SCSR 55656 was assigned to Denver Wilfong and was manufactured in September of 1997. There were no records from the Sago Mine indicating SCSR 55656 was assigned to Wilfong or if the 90 day inspection was completed for that unit. Wilfong entered the mine after the explosion as well, but did not don his SCSR. He stated,” I was saving it ‘til I needed it, I guess.” The records show he had Annual Refresher Training at the Spruce Fork Mine on August 22, 2003. The training form indicated, “Hands on SCSR.” The records show he had Experienced Miner Training at the Sago Mine on December 6, 2005. 95 Other SCSRs Recovered and Evaluated SCSRs 106603 and 107966 were recovered and believed to belong to miners on the 1st Left mantrip. There were no records from the Sago Mine for SCSR 106603 SCSRs 109419, 57517, 106615, 109482 and 109455 were found in the barricade. SCSR 101106 was found on the 2nd Left Parallel Section. They were opened and activated on January 3 - 4, 2006. These units were believed to be opened during the rescue of McCloy. According to testimony, they were only used for a brief period of time. They were recovered by the investigation team and stored in plastic bags until they were evaluated on March 27 - 31, 2006. The visual observations indicated that between 5% and 10% of the chemicals in the units were used. This shows that any change that might have occurred in the chemical beds of the units as a result of either the units’ exposure to the mine atmosphere until they were recovered, the storage procedure used or the length of time that elapsed between recovery and evaluation, was minimal. This conclusion is further supported by the results of the laboratory test conducted in 2006. SCSR 106603 was recovered and is believed to belong to one of the miners on the 1st Left mantrip. However, the Spruce Mine inspection records indicated SCSR 106603 was assigned to another miner and the 90 day inspection occurred on February 12, 2005. There were no records from the Sago Mine for SCSR 106603. The manufacture date was in August of 2004. SCSR 106603 was evaluated by NIOSH. The dust shield had no cracks and the canister was not dented. Any damage was not significant. NIOSH evaluations established that the start-up oxygen was activated, that the unit produced oxygen, and that approximately 20% to 25% of the chemicals in the unit were used. CSE and ATL conducted chemical analyses of the unit. They reported that approximately 23% and 46% of the chemicals in the unit had been depleted, respectively. SCSR 107966 was recovered and is believed to belong to one of the miners on the 1st Left mantrip. The mine operator’s inspection records indicated that SCSR 107966 was not assigned to any miner and the 90 day inspection was completed in a timely manner, with the last inspection occurring on November 16, 2005. The manufacture date was November of 2004. SCSR 107966 was evaluated by NIOSH. The dust shield had no cracks and the canister was not dented. Any damage was not significant. “Walker” was written on the unit. There was evidence of moisture on the bottom filter. NIOSH evaluations established that the start-up oxygen was activated, that the unit produced oxygen, and that approximately 80% to 90% of the chemicals in the unit were used. CSE and ATL conducted chemical analyses of the unit. They reported that approximately 48% and 63% of the chemicals in the unit had been depleted, respectively. 96 SCSR 109419 was recovered in the barricade and is believed to have been opened during the rescue effort. There were no records from the Sago Mine for SCSR 109419. The manufacture date was in October of 2004. SCSR 109419 was evaluated by NIOSH. The dust shield had no cracks and the canister was not dented. Any damage was not significant. NIOSH evaluations established that the start-up oxygen was activated, that the unit produced oxygen, and that approximately 5% to 10% of the chemicals in the unit were used. CSE and ATL conducted chemical analyses of the unit. They reported that approximately 11% and 14% of the chemicals in the unit had been depleted, respectively. SCSR 57517 was recovered in the barricade and is believed to have been opened during the rescue effort. The Spruce Mine inspection records indicated SCSR 57517 was assigned to another miner not on the 2nd Left Parallel crew and the 90 day inspection occurred on February 14, 2005. There were no records from the Sago Mine for SCSR 57517. The manufacture date was December of 1997. SCSR 57517 was evaluated by NIOSH. The dust shield had no cracks and the canister was not dented. Any damage was not significant. NIOSH indicated the breathing tube was set but was pliable and open. The cap was missing from the relief valve. The breathing bag had an impression from the goggles or a stain on the bag. NIOSH evaluations established that the start-up oxygen was activated, that the unit produced oxygen, and that approximately 5% to 10% of the chemicals in the unit were used. CSE and ATL conducted chemical analyses of the unit. They reported that approximately 20% and 34% of the chemicals in the unit had been depleted, respectively. SCSR 106615 was recovered in the barricade and is believed to have been opened during the rescue effort. The mine operator’s inspection records indicated SCSR 106615 was assigned to another miner not on the 2nd Left Parallel crew and the 90 day inspection occurred on November 16, 2005. The manufacture date was August of 2004. SCSR 106615 was evaluated by NIOSH. The dust shield had no cracks and the canister was not dented. Any damage was not significant. NIOSH indicated that there was mineralization on the bottom filter. NIOSH evaluations established that the start-up oxygen was activated, that the unit produced oxygen, and that approximately 10% of the chemicals in the unit were used. SCSR 109482 was recovered in the barricade and is believed to have been opened during the rescue effort. There were no records from the Sago Mine for SCSR 109482. The manufacture date was in October of 2004. SCSR 109482 was evaluated by NIOSH. The dust shield had no cracks and the canister was not dented. Any damage was not significant. NIOSH evaluations established that the start-up oxygen was activated, that the unit produced oxygen, and that approximately 5% to 10% of the chemicals in the unit were used. 97 SCSR 109455 was recovered in the barricade and is believed to have been opened during the rescue effort. The mine operator’s inspection records indicated SCSR 109455 was not assigned to any miner and there was no record of the 90 day inspection. The manufacture date was in October of 2004. SCSR 109482 was evaluated by NIOSH. The dust shield was dented and the canister was dented. Any damage was marginal. NIOSH evaluations established that the start-up oxygen was activated, that the unit produced oxygen, and that approximately 10% of the chemicals in the unit were used. CSE conducted chemical analyses of the unit. They reported that approximately 21% of the chemicals in the unit had been depleted. SCSR 101106 was recovered on the 2nd Left Parallel section and is believed to have been opened during the rescue effort. The Spruce Mine inspection records indicated SCSR 101106 was assigned to a miner not on the 2nd Left Parallel crew, and that the 90 day inspection occurred on February 14, 2005. There were no records from the Sago Mine for SCSR 101106. The manufacture date was May of 2004. SCSR 101106 was evaluated by NIOSH. The dust shield had no cracks and the canister was not dented. Any damage was not significant. NIOSH evaluations established that the start-up oxygen was activated, that the unit produced oxygen, and that approximately 10% of the chemicals in the unit were used. Table 7 is a summary of the SCSRs that were assigned to the miners who were underground at the time of the explosion, were assigned to the miners who traveled underground during the rescue attempt, and that were used by the mine rescue team assisting McCloy. 98 Table 7 – Summary of Information on the SCSRs at the Sago Mine Miner Location Serial No. Date of SCSR Training Was Unit Donned 90 Day Inspection Record at Sago Evidence of Oxygen Production NIOSH Visual % Used CSE % Used John N. Boni John P. Boni James Jamison Outby Outby Outby 106186 100991 91947 10/07/051 12/29/042 06/27/05 No No No 11/14/05 11/16/05 None - - - Denver Anderson Paul Avington Gary Carpenter Ronald Grall Randall Helmick Eric Hess Owen Jones Hoy Keith Arnett Perry Gary Rowan Harley Ryan Christopher Tenney Anton Wamsley 1st Left 1st Left 1st Left 1st Left 1st Left 1st Left 1st Left 1st Left 1st Left 1st Left 1st Left 1st Left 1st Left 83566 63227 75648 92943 56505 88170 92933 60035 102138 86537 97144 52409 88981 12/08/05 12/08/05 12/08/05 09/16/05 12/08/05 12/08/053 12/08/05 12/08/05 12/08/05 12/08/05 12/08/05 12/08/05 12/08/05 Yes No No No No Yes No Yes Yes Yes Yes No Yes None None None 11/17/05 11/15/05 11/14/05 11/14/05 None 11/15/05 None 11/16/05 11/14/05 11/17/05 na na na Yes Yes Yes na na na na 20 10 40 - 50 na na na na 29 19 42 na Terry Helms 2nd Left 90223 06/27/05 No 11/14/05 - - - - Thomas Anderson* Alva Bennett James Bennett Jerry Groves* George Hamner Jesse Jones* David Lewis Randal McCloy Jr. Martin Toler Jr.* Fred Ware Jackie Weaver Marshall Winans Barricade Barricade Barricade Barricade Barricade Barricade Barricade Barricade Barricade Barricade Barricade Barricade 92652 89765 56495 57878 101868 46433 101831 106154 57604 56880 57334 52478 12/08/05 12/08/05 12/08/05 12/08/05 12/08/05 03/18/05 04/22/05 12/08/05 12/08/05 3/25/05 12/08/05 12/08/05 Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes None None None None None 11/17/05 11/18/05 11/18/05 None 11/17/05 11/16/05 11/14/05 Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes 40 25 25-35 40-50 25 10-20 10-20 20-25 10-15 10-20 30 50-60 39 38 48 31 41 25 28 21 39 39 72 48 54 77 50 29 27 41 34 87 Vernon Hofer James Schoonover Jeffrey Toler Denver Wilfong Rescue Rescue Rescue Rescue 63274 104889 104831 55656 02/28/05 01/12/04 08/02/05 12/06/05 No No No No 11/17/05 11/17/05 11/14/05 None - - - - 1st Left 1st Left Barricade Barricade Barricade Barricade Barricade 2nd Left 106603 107966 109419 57517 106615 109482 109455 101106 - Yes Yes Yes Yes Yes Yes Yes Yes None 11/16/05 None None 11/16/05 None None None Yes Yes Yes Yes Yes Yes Yes Yes 20-25 80-90 5-10 5-10 10 5-10 10 10 23 48 11 20 21 - 46 63 14 34 - Other Other Other Other Other Other Other Other na - SCSR was not available or unable to determine the user * - Miners identified by McCloy as having difficulties with their SCSRs 1 - Did not participate in training but filled out the form 2 - Did not don the unit during training, training on 07/05/05 by a person not listed as an approved trainer 3 - Box on training form not marked 99 ATL % Used na na na 31 28 48 na Mine Ventilation Plan The Ventilation Plan in effect at the time of the explosion addressed specific ventilation requirements. MSHA approved the plan, which included a number of addendums, on May 5, 2005. Six-month reviews were conducted as required. MSHA completed the last six-month review prior to the accident on October 25, 2005. The plan required that when a Joy 14CM15 continuous mining machine was used and coal was cut, mined or loaded, and the scrubber was on, a minimum of 6,000 cfm and a maximum of 9,000 cfm of air was required at the inby end of the line curtain. When the scrubber was not running, a minimum of 6,000 cfm was required at the inby end of the line curtain. The line curtain was required to be within 40 feet of the point of deepest penetration with blowing face ventilation. When the continuous mining machine was equipped with a scrubber but it was not used, the line curtain was required to be maintained to within 20 feet of the area of deepest penetration with exhaust ventilation. When an Eimco 2810-2 continuous mining machine was used with the scrubber on, a minimum of 6,000 cfm and a maximum of 8,000 cfm of air were required at the inby end of the line curtain. When the continuous mining machine was in the working place with the scrubber off, a minimum of 6,000 cfm was required. When the Joy 14CM15 or the Eimco 2810-2 was not equipped with a scrubber, the plan required a minimum of 5,880 cfm of air at the inby end of the line curtain, or a minimum of 60 fpm mean entry air velocity, whichever was greater. The line curtain was to be maintained within 20 feet of the area of deepest penetration with exhaust ventilation when the machines were not equipped with a scrubber. During roof bolting, the line curtain was required to be maintained to the second full row of roof bolts outby the face and was advanced until the curtain was within 10 feet of the face. A minimum of 3,500 cfm of air was required while the roof bolter was in operation. A minimum of 9,000 cfm was required in the last open crosscut of each split. The mine used an AMS as the automatic fire warning system required by Section 75.1103. The plan addressed the type of system, capabilities of the system, air velocity along the belt entry, type of activation signals, inspections, examinations, testing and procedures to follow when the system or a portion of the system became inoperative. On September 28, 2005, MSHA approved a supplement to the plan for a test area of about 300 feet in length, which detailed the ventilation and evaluation of the 2nd Left Mains during and after mining the lower bench of the Middle Kittanning Coal Seam (bottom mining). This bottom mining was to be completed while retreating out of the 2nd Left Mains area. The mine operator was to install seals after completion of the bottom mining. 100 On October 4, 2005, MSHA approved a supplement to the plan to extend the test area in 2nd Left Mains for mining the lower bench of the Middle Kittanning Coal Seam and for the ventilation and evaluation of the area. The area covered by the supplement was the remainder of 2nd Left Mains and a portion of 2 North Mains extending from the face to a point about one crosscut outby the entrance to 2nd Left Mains. This plan also contained additional safety precautions to further protect persons while bottom mining. Bottom mining was conducted in some areas of the mine to recover additional coal reserves in the lower bench of the Middle Kittanning Coal Seam that was separated from the upper bench by a layer of rock. Normal mining height during initial development averaged seven feet in the 2nd Left Mains and 2 North Mains. After bottom mining was conducted, the mining height ranged from about 10 to 20 feet. Development was completed in a section before any bottom mining could begin. The belt loading point and equipment were moved outby. A new belt loading point was established and bottom mining commenced at an outby point and moved inby. The continuous mining machine operator commenced mining by cutting a ramp down to the desired depth and continued inby to a pre-determined stopping point. Bottom mining was only conducted in entries. Crosscuts were not bottom mined. Roof support installation was not necessary since the roof had been supported during initial development. To provide some protection against overhanging ribs, the mining plan did not permit bottom mining wider than the development mining. Once the mining was completed in all designated entries for that setup, the belt loading point and equipment were once again moved outby and the process repeated. Once an area was completed, no person was permitted in the area. This would eliminate exposure of persons to the heightened coal ribs. This process continued progressively outby until the designated area to be bottom mined was completed. Two additional supplements were submitted, and then approved on October 21, 2005 and December 19, 2005 for bottom mining in the A-1 and A-2 Panels off 1st Left. These approved supplements were similar to the approved plan for the 2nd Left Mains and 2 North Mains areas. Appendix J contains the four bottom mining supplements to the ventilation plan. The ventilation plan contained a set of guidelines for the installation of preloaded solid concrete block (Packsetter) seals. MSHA also approved supplements to the plan providing for non-hitched Omega Block seals. These supplements outlined the location and the procedures for installation and ventilation of the seals during and after construction. MSHA approved two supplements on October 24, 2005. The first supplement contained procedures for installation of a 40 inch thick, up to 8 foot high and up to 20 foot wide Omega Block seal. The second supplement described the sequence of constructing seals 101 for the 2 North Mains area and making air changes. The first change was to show ventilation during construction of the seals and the second was to show the final ventilation after completion of the seals. MSHA approved the third supplement on December 8, 2005. This supplement contained procedures for the installation of the three different configurations of non-hitched Omega Block seals. The first was again a 40 inch thick, up to 8 foot high and up to 20 foot wide seal. The second was a 40 inch thick, up to 10 foot high and up to 20 foot wide seal. The third was a 40 inch thick, up to 12 foot high and up to 20 foot wide seal. The three configurations were submitted and approved in preparation for sealing the A1 and A2 Panels off 1st Left, where the entry exceeded 8 feet in height. Appendix K contains the three supplements to the ventilation plan concerning Omega Seals. Mine Ventilation The mine was ventilated with a blowing ventilation system. The drift openings were numbered from left to right facing inby. Airflow entered the mine through the No. 5 Drift Opening and exited through No. 1 and the three other drift openings, which consisted of a track, conveyor belt and one other common opening. According to the mine record books, the total quantity of intake air entering the mine through the blowing fan at the No. 5 Drift Opening was 146,566 cfm on December 28, 2005. The total quantity of return air exiting the mine through the No. 1 Drift Opening was 101,088 cfm. The remaining 45,478 cfm would have exited the mine through the Nos. 2, 3 and 4 Drift Openings. The blowing fan was an 8 foot diameter, Joy Model Number M96-50 fan, with a blade setting of 8 degrees and operating at about 1.9 inches of water gauge. Figure 12 is a copy of the chart which was on the fan pressure recorder when the explosion occurred. Although the fan chart shows a pressure spike about 6:00 a.m., the explosion occurred about 6:26 a.m. This indicates that the fan chart was not correctly aligned on the pressure recorder to correspond with actual time. Figure 12 - Fan Chart 102 Development Sections The 1st Left and 2nd Left Parallel were developed with eight entries. The sections were ventilated with a single split of air. The entries were numbered from left to right facing inby. The Nos. 7 and 8 entries on the right side of the section served as intake air courses. The Nos. 1 and 2 entries on the left side of the section served as return air courses. The No. 5 entry was the track entry, the No. 4 entry was the conveyor belt entry and the Nos. 3 and 6 entries were common with the belt and track. The sections did not use belt air at the face and the airflow in the Nos. 3, 4, 5 and 6 entries was in an outby direction. The preshift examination record books for the 1st Left and 2nd Left Parallel sections on the day of the accident indicated that the quantity of air measured in the last open crosscut was 14,510 and 11,241 cfm, respectively. Ventilation of Seals Two sets of mine ventilation seals were installed to separate worked-out portions of the mine from the active areas. The seals were located across 1 NE Mains and 2 North Mains. The 1 NE Mains seals were ventilated with return air. The 2 North Mains seals were ventilated with intake air that was directed across the seals and over a set of overcasts to the return air course at the mouth of the active 2nd Left Parallel section. Neither the 1st Left nor the 2nd Left Parallel sections were being ventilated with air that passed these seals. Methane Ignitions There had been one methane ignition reported at the mine since its opening. The ignition occurred on February 8, 2001. At that time the mine was known as Spruce No. 2 and owned by BJM Coal Company. A section foreman and four miners were preparing to install a tunnel liner into the face area of the No. 5 entry of the 1 NE Mains. In order to install a tunnel liner, a crossbar suspended by three roof bolts about 10 feet outby the face of the No. 5 entry had to be removed. The miners used an oxygen/acetylene cutting torch to cut off the roof bolts holding the crossbar. After the section foreman had completed cutting two of the roof bolts, he raised the cutting torch toward the third roof bolt and ignited methane. The flame extinguished itself, but not before causing first and second degree burns to the four miners. The mine operator later sealed off the 1 NE Mains using preloaded solid concrete block “Packsetter” seals. Methane Liberation During each MSHA quarterly inspection of the mine, inspectors collected an air sample in the No. 1 Drift Opening (return air course) to determine the daily methane liberation. To assist in that determination, the air quantity at the 103 sampled location was determined. The air sample was sent to MSHA’s Mount Hope, West Virginia Laboratory to be analyzed. The lab determined the amount of methane in the sample and calculated the quantity of methane liberated from the mine in cubic feet per day. The analysis of the last four quarterly collected air samples is shown in Table 8. Table 8 - Air Sample Results Date January 10, 2005 April 29, 2005 July 18, 2005 October 5, 2005 Methane (%) 0.09 0.07 0.09 0.10 Quantity (cfm) 53,074 94,446 83,136 62,901 Liberation (cfd) 68,784 95,202 107,744 90,577 A ventilation survey was conducted on March 1-2, 2006 as part of the accident investigation. Air samples were collected at the drift openings which were outgassing mine air during the study. The analysis of those samples indicated that the mine liberated 92,460 cfd of methane. Two methane studies were conducted in the area previously sealed inby the 2 North Mains seals, on February 7–9, 2006 and March 2–3, 2006. Both studies were conducted by collecting information in the mine and from the Nos. 5 and 7 boreholes located in the previously sealed area. Information collected included air samples, air velocities, air temperatures, air pressures, borehole diameters and regulator opening dimensions. This information was used to calculate the methane liberation from the area. The results from the February and March studies indicated that the previously sealed area liberated about 13,220 cfd and 12,090 cfd of methane, respectively. Methane in the Sealed Area Methane has a specific gravity of 0.55 and is lighter than air. It is only explosive in methane-air mixtures that range from 5% to 15%. Methane-air mixtures above or below these concentrations will not burn or become involved in an explosion. Generally, methane enters the mine in concentrations in excess of 80% and is diluted by the ventilation current. After the seals were completed, the atmosphere behind the seals was stagnant. Methane entering the area would have the tendency to form layers, with higher concentrations near the mine roof. Additionally, it cannot be assumed that methane would have been liberated equally in each entry or crosscut or from the roof, ribs, or floor. It is likely that the average concentration in the entries would differ throughout the sealed area based on the liberation in each entry. 104 The results of the methane liberation studies were used to evaluate the volume of methane in the sealed area prior to the explosion. The rate of decay of the methane liberation was considered to be linear. Based on the studies and the assumptions, the methane liberation on December 11, 2005, the day the area was sealed, was approximately 16,350 cfd and on January 2, the day of the explosion, was approximately 15,220 cfd. The average liberation rate for the 22 days that the area was sealed was approximately 15,790 cfd. Therefore, the volume of methane in the sealed area just prior to the explosion was approximately 347,300 cubic feet. The mine operator provided information to MSHA indicating that the total open volume behind the seals just prior to the explosion was about 2,938,156 cubic feet. Based on this calculation, the volume of methane in the sealed area of 347,300 cubic feet indicates an average homogeneous methane concentration of over 11%. However, it is not likely that a homogenous mixture of 11% was present throughout the sealed area at the time of the explosion. After the explosion, the volume of air and the concentration of gases exiting the mine through the return drift opening and at the boreholes were monitored. Initially, the methane concentrations at these locations were elevated. Concentrations eventually declined and stabilized to a background level. The volume of methane in the air that exited the mine through the return drift opening and the boreholes that was greater than the background level was calculated. This volume was considered to be excess methane that was in the sealed area prior to the explosion and which was not consumed by the explosion. This excess volume of methane was determined to be approximately 205,500 cubic feet. Since this volume was not involved in the explosion, it was likely in concentrations less than 5% or greater than 15%. Calculations indicate that 141,800 cubic feet of methane (347,300 – 205,500) was consumed by the explosion or ventilated from the mine through another means. This information further supports the conclusion that a homogeneous explosive methane/air mixture did not exist in the sealed area prior to the explosion. Ventilation Survey and Computer Simulations Investigators obtained information pertaining to the mine ventilation system from a variety of sources, including mine records, fan charts, mine rescue team maps, mine recovery team maps, underground investigation findings, and interviews and discussions. In addition, they conducted a ventilation survey on March 1-2, 2006, after the mine operator reconstructed the ventilation system to a production-ready configuration. The ventilation survey consisted of collecting and recording measurements of air velocities, mine entry heights and widths, and air pressures at predetermined 105 locations throughout the mine. Those locations included but were not limited to air splits, regulators, and the fan. The ventilation survey determined that very small air pressure differentials induced airflow. These pressures differentials were so small, they were often beyond the ability of the instruments being used to accurately measure, thereby making aircourse resistance calculation extremely difficult. The airflow measurements were balanced so that a computer program could use the data. The entry resistance data was normalized for any given aircourse characteristic. The typical ventilation program data would include airway resistance, airway area, airway length, pressure drop and air quantities. This data would allow the program to make calculations of the ventilation network using the Hardy Cross method. The program may be used to generate tabulated reports, graphs of fan curves and fan operating points, and network distribution diagrams showing pressure drops, resistance, and airflow distribution. However, the program was not written to permit calculations to the degree needed when dealing with air course pressure differentials as small as onethousandth of an inch of water. When the ventilation system was reconstructed, it did not replicate the system as it existed on the morning of the accident, prior to the explosion. Differences in the ventilation system included the ventilation of the previously sealed area, the addition of a second bank of overcasts at 1st Left, and the elimination of a set of overcasts at the mouth of 2nd Left Parallel. In order to replicate the ventilation system, MSHA developed computer simulations to depict the pre- and post-explosion ventilation. However, due to very small pressure differentials of various aircourses throughout the mine, the pre- and post-simulations depicting aircourse patterns and air quantities should be used for demonstrative purposes only. The simulation depicts the mine ventilation system prior to the explosion, as shown on a map in Appendix L. This illustrates airflow direction and quantities, and ventilating pressures. The results were compared to the information obtained from the sources listed above to verify their accuracy. The simulation indicates that the air quantity delivered on the 1st Left section was 43,300 cfm and on the 2nd Left Parallel section was 46,900 cfm. The simulation indicates the fan would be blowing approximately 172,300 cfm of air into the mine at a fan pressure of 1.85 inches water gauge. MSHA also developed two post-explosion computer simulations of the ventilation system of the mine after the explosion. The simulation depicting the mine‘s ventilation system after the explosion with the ventilation controls damaged is shown on a map in Appendix M. This simulation shows that the 106 outby damage to the overcast and stoppings at 2 Right and the stopping at 32 Crosscut, No. 4 Belt, decreased the available quantity of intake air moving inby 32 Crosscut, No. 4 Belt from a pre-explosion quantity of 118,400 cfm to 53,000 cfm. The model shows that the damage to the ventilation controls inby 42 Crosscut, No. 4 Belt created major ventilation short circuits, thereby limiting any mechanically induced ventilation inby 49 Crosscut, No. 4 Belt. The model also shows there was no mechanically induced airflow to the mouth of the 2nd Left Parallel. The second post-explosion simulation is shown on a map in Appendix N and depicts the placement of curtains hung by mine management during their attempt to reach the 2nd Left Parallel crew. Controls include the curtain installed between the intake and track to 57 Crosscut, No. 4 Belt and the curtain hung at the 2 Right overcast. The model indicates repairs to the ventilation controls outby 57 Crosscut, No. 4 Belt created mechanically induced ventilation to 57 Crosscut, No. 4 Belt. The model also indicated that there was mechanically induced airflow to the mouth of the 2nd Left Parallel. Although the simulations indicate that the repairs to the ventilation controls may have had an impact on the atmosphere in the 2nd Left Parallel, the extent of that impact or its affect on the 2nd Left Parallel miners could not be determined. Barometric Pressure Changes in barometric pressure can cause the expansion and contraction of accumulated gases within unventilated (sealed) and poorly ventilated areas of mines. Generally, changes in the barometric pressure have little impact on the atmosphere in the sealed area in a mine, except for the areas just inby and outby the seals. During a period of falling barometric pressure, the atmosphere tends to leak from the sealed area into the active workings of a mine. When the barometric pressure is rising, the atmosphere tends to leak from the active area into the sealed area. The barometric pressure for Buckhannon, West Virginia, at 12:00 a.m. on January 2 was approximately 30.01 inches of mercury. The barometric pressure was falling from 1:00 a.m. to 4:00 a.m. At 4:00 a.m., the barometric pressure was 29.92 inches of mercury. From 4:00 a.m. to 6:30 a.m., the pressure varied between 29.90 and 29.94 inches of mercury. The pressure was about 29.93 inches of mercury at 6:30 a.m. 107 Figure 13 is a graph of the barometric pressure for Buckhannon, West Virginia from 12:00 a.m. on January 1 through 12:00 a.m. on January 4, 2006. These changes in barometric Barometer Pressure pressure did not appear to significantly influence the conditions within the sealed area just prior to the explosion, since the point of origin for this explosion was more than 300 feet from the seals. Buckhannon, WV 30.50 . 30.40 1/2/06 6:30 AM Barometric Pressure (Inches of Mercury) 30.30 30.20 30.10 30.00 29.90 29.80 29.70 29.60 29.50 1/1/06 12:00 AM 1/1/06 8:00 AM 1/1/06 4:00 PM 1/2/06 12:00 AM 1/2/06 8:00 AM 1/2/06 4:00 PM 1/3/06 12:00 AM 1/3/06 8:00 AM 1/3/06 4:00 PM 1/4/06 12:00 AM Time Figure 13 - Barometric Pressure for Buckhannon, WV Roof Control Plan MSHA approved a Roof Control Plan for the mine on October 16, 2003. Sixmonth reviews were conducted as required. MSHA completed the last sixmonth review prior to the accident on June 29, 2005. MSHA approved four foot and six foot fully grouted resin bolts and five foot fully grouted resin tension bolts as the primary roof supports. Ten and fourteen foot resin cable bolts, prop setters, square and round plates, metal straps, wire mesh, brow tenders and other approved devices were used as supplemental support throughout the mine. Figure 14 shows pictures of square and round plates. Figure 14 - Square and Round Plates 108 The approved plan required a four foot by four and a half foot roof bolt installation pattern in the main and sub main entries of the mine. The plan required the operator to bolt wire mesh to the roof of the track and belt conveyor entries during the roof bolting cycle, to within two bolt rows of the face. The wire mesh was required to be at least 8 gauge, with openings no greater than four inches square, and measuring at least five feet by thirteen feet overall. The wire mesh is shown in Figure 15. Figure 15 - Wire Mesh The plan required at least one of the following in the primary intake escapeway and one return air course entry maintained evenly with the section tailpiece: • • • • A roof sealant applied to the mine roof; A 17 inch square or larger plate (roof cap) installed with each roof bolt; Wire mesh bolted to the mine roof as described above; or Two rows of posts or equivalent supports installed to create a six foot wide walkway on not more than five foot advancing centers. Of these four options, the mine operator installed the wire mesh in the primary intake escapeway and a return entry. The mine conveyor belt structure was suspended from the mine roof. Belt support brackets were anchored to the mine roof with roof bolts. These brackets and bolts were installed against the wire mesh for belt installation and not as roof support. 109 Geology The mine was developed in the Middle Kittanning Coal Seam. The overburden in the 2nd Left Parallel, measured from the base of the seam to the surface, ranged from 230 feet to 320 feet. The immediate roof consisted of gray shale grading upward into sandy shale and sandstone with shale bedding. A description of the mine geology and roof falls is contained in a report titled “Evaluation of Potential for a Roof Fall to Ignite a Methane-Air Mixture” in Appendix O. Evaluation of Two Linear Features near Survey Station 4010 Two linear geologic features were observed during the investigation. These two prominent features were located in the roof near survey station 4010, within the formerly sealed area of 2nd Left Mains. The features generated interest because they were located in the area where the explosion originated. Because the features seemed uncommon, they were referred to as “anomalies.” Due to their location in the area where the explosion originated, some parties speculated that the linear “anomalies” might represent the effects of lightning. A picture of the anomaly is shown in Figure 16. Anomaly Anomaly Figure 16 - Anomaly Light brown linear streaks along the trend of the parallel linear ridges represent knife scratch marks from an attempt to collect fossil material. Location is the vicinity just inby survey station 4010 intersection. Twin parallel ridges pass beneath the embossed, square skin control plate. 110 An analysis of the features concluded that the linear features represent the remnants of a pair of fossilized trees, with each linear feature representing the top, tangential edge of a single tree. The rough texture of the linear feature represents the trace fossil impression of the tree bark as preserved against the bottom layer of the overlying muscovite-rich gray shale, and the pair of parallel ridges represents compaction of the muscovite-rich gray shale downward around the formerly circular boundary of the tree trunk. Although the fossil tree was removed by mining, the linear features represent the expression of the top edge of the tree where it tangentially contacted the bottom of the bedding plane exposed in the shale roof. An analysis and description of the linears near survey station 4010 is contained in Appendix P in two reports titled “Evaluation of Features” and “Description of Features Observed in the Roof Inby Spad 4010.” Cleanup Program and Rock Dusting The mine operator established a program for regular cleanup and removal of accumulations of coal and float coal dusts, loose coal, and other combustible materials at the mine. The program included an examination of active haulage ways prior to the end of each shift. Any loose coal accumulations were to be removed from the mine. Miners were to examine mining equipment used at the face and to remove accumulations of loose coal, coal dust, oil and grease before the end of each shift. They were also to remove any accumulations of loose coal, coal dust, oil and grease from the section tailpiece by the end of each shift. Rock dust was to be applied and maintained to within 40 feet of each working face. Accumulations of loose coal, coal dust or other combustibles along belt and track travel ways were to be removed or reported to the mine foreman each shift. Rock dust was applied in the 2 North Mains and in 2 Left Mains during initial development. Additional rock dust was not applied in areas after they had been bottom mined. Miners stated that 36 one-ton bags and several pallets of 50pound bags of rock dust were delivered to the track switch at the mouth of 2nd Left Parallel before the 2 North Mains seals were completed. Miners applied rock dust by hand and with rock dusting machines around the sealed area and outby the seals for a distance of approximately four crosscuts. According to miners, the depth of the rock dust in the area varied between one-half and threefourths of an inch. Mine Dust Survey Investigators conducted a post-explosion mine dust survey. The mine dust samples were analyzed at MSHA’s Laboratory in Mount Hope, West Virginia. Each sample was subjected to an Alcohol Coke Test and an incombustible 111 analysis. The incombustible analysis identified the percentage of incombustible material in each sample. The Alcohol Coke Test identified the portion of coke in each sample. The results of the mine dust survey are contained in Appendix Q. The locations of all intended mine dust samples are shown on the mine map in Appendix R. Samples were collected by band or perimeter method from entries. Material was gathered from an area on the floor up to one inch deep and six inches wide and combined with dust from the roof and ribs to make up a one band or perimeter sample. The material was collected with a small flat scoop and brush, placed in a collection pan, and sifted through a 10 mesh screen. The sifted material was placed on a clean rubber sheet. If the amount collected was too large for the collection bag, then the sample was thoroughly mixed and quartered, reducing the desired amount to a half bag. If an insufficient amount was gathered, an additional, adjacent band sample would be taken. Where it was impractical or unsafe to collect full perimeter samples because of excessive height, a floor sample and a sample from the ribs was collected to the maximum height that could be done safely. Each bag was long enough to allow tying a knot in the open end of the bag. An identifying tag was secured to each bag by the tag string and secured within the formed knot of the bag. As a sample was collected, the location was marked on the identifying tag that corresponded to the predetermined location on the mine dust survey map. The incombustible content of the combined coal dust, rock dust and other dust must be maintained to at least 65% in the intake air courses and at least 80% in the return air courses, in the absence of methane, to meet regulatory requirements. The area evaluated was extensive; therefore, the survey was divided into five separate survey areas, as follows: Survey No. 1(a) - 2 North Mains (outby the location of the 2 North Mains seals) Survey No. 1(b) - 2 North Mains (inby the location of the 2 North Mains seals) Survey No. 2 - 1st Left Survey No. 3 - 2nd Left Parallel Survey No. 4 - 2nd Left Mains (inby the location of the 2 North Mains seals) MSHA intended to collect mine dust samples at 685 designated locations. However, 458 locations could not be sampled because of wetness, inaccessibility, or because the area was unsafe to travel. A total of 227 locations were successfully sampled. 112 2 North Mains - Survey No. 1(a) The starting point for this survey was 50 feet inby survey station 3483 of the 2 North Mains track entry, and extended inby for approximately 5,700 feet to the location of the 2 North Mains seals. There were 247 designated locations identified for sampling in this survey. A total of 141 mine dust samples were collected. The other 106 locations could not be sampled because of wetness, inaccessibility, or because the area was unsafe to travel. The results of the 141 samples collected indicate that 39 of the samples, or 28%, were substandard. However, due to the area where the explosive force propagated, it cannot be determined if these samples were contaminated by dust and other materials. Therefore, the incombustible content of the samples taken could not be used to determine compliance with the regulatory requirements. 2 North Mains - Survey No. 1(b) The starting point for this survey was inby the 2 North Mains seals and extending toward the faces of 2 North Mains. There were 64 locations identified for sampling. Mine dust samples were collected at 29 locations. The other 35 locations could not be sampled because of wetness, inaccessibility, or because the area was unsafe to travel. The results of the 29 samples collected indicate that 26 of the samples, or 90%, were substandard. The explosion occurred inby the seals and the incombustible content of the samples taken could not be used to determine compliance with the regulatory requirements. 1st Left - Survey No. 2 The starting point for this survey was at the mouth of 1st Left. There were 43 locations identified for sampling in 1st Left. Mine dust samples were collected at four locations. The other 39 locations could not be sampled because of wetness, inaccessibility, or because the area was unsafe to travel. The incombustible content results of the four samples indicated that two of the four samples, or 50%, were substandard. 2nd Left Parallel - Survey No. 3 The starting point for this survey was at the mouth of 2nd Left Parallel. There were 222 designated locations for sampling in 2nd Left Parallel. Mine dust samples were collected at 42 mine locations, the other 180 locations could not be sampled because of wetness, inaccessibility, or because the area was unsafe to travel. Of the 42 samples analyzed, 14 of the samples, or 33%, were substandard. 113 2nd Left Mains - Survey No. 4 The starting point for this survey was at the mouth of 2nd Left Mains. There were 109 locations identified for sampling in the 2nd Left Mains. Mine dust samples were collected at 11 locations, the other 98 locations could not be sampled because of wetness, inaccessibility, or because the area was unsafe to travel. However, due to the area where the explosive force propagated, it cannot be determined if these samples were contaminated by dust and other materials. The results of the 11 samples collected indicate that 4 of the samples, or 36%, were substandard. Therefore, the incombustible content of the samples taken could not be used to determine compliance with the regulatory requirements. MSHA Mine Dust Sampling Prior to Accident MSHA conducted mine dust surveys during regular health and safety inspections prior to the accident. The areas that were evaluated for incombustible content as required by Section 75.403 included areas beginning approximately 600 feet outby the 2 North Mains seals and extending through the sealed area and into 2nd Left Mains. Based on the inspectors’ observations and evaluation, this entire area could not be sampled because of excessive water. Additionally, mining had stopped because of increased water inflow and deteriorating roof conditions. As discussed previously, a large area was evaluated before the accident and was very wet. Similar conditions were found after the accident. Therefore, it appears that the area may have also been wet at the time of the explosion. Examinations Sections 75.360 and 75.362 require that examinations of the mine be conducted by certified mine examiners. Section foremen were normally assigned to conduct preshift and onshift examinations during production shifts. Hourly mine examiners were normally assigned to conduct preshift examinations on nonproducing shifts. Other mine examiners were normally assigned to conduct onshift and preshift examinations along the belt and track entries. Section 75.360 requires an examination by a certified person within 3 hours preceding the beginning of any 8-hour interval during which any person is scheduled to work or travel underground. The certified examiner is required to examine for hazardous conditions, test for methane and oxygen deficiency, and determine if air is moving in the proper direction at specific locations, such as travelways, working sections, and seals along intake air courses where intake air passes by a seal to ventilate working sections. The 2 North Mains seals were not required to be examined during preshift examinations unless miners were 114 scheduled to work in the area. Preshift examinations were performed based upon three 8-hour time periods. The 8-hour intervals scheduled for starting preshift examinations were 6:00 a.m., 2:00 p.m. and 10:00 p.m. On Sunday, January 1, 2006, the day shift mine foreman and two other miners worked on the track and on a pump on the 2nd Left Parallel section. One of the hourly employees was a motorman who was also certified to conduct mine examinations. Preshift or supplemental examinations were not conducted prior to these employees entering the working area. On January 2, 2006, two mine examiners, Helms and Jamison, conducted a preshift examination of the underground areas of the mine before the crews entered the mine. This break in routine was due to the holiday weekend. Jamison examined the 2nd Left Parallel section and exited the mine to complete his report. Helms examined the 1st Left section, remained underground and called his report to the surface. No unsafe conditions or dangers were noted or reported. Section 75.364 requires a weekly examination of worked-out areas and the bleeder system. It also requires an examination for hazardous conditions at specific locations that include at least one entry of the intake and return air courses in their entirety and at each seal along a return or bleeder entry. Measurement of air volume and tests for methane at specific locations are also required. Hourly employees who were also examiners were assigned to conduct the majority of the weekly examinations. Interviews with mine personnel and a review of the weekly examination records conducted during the last quarter of 2005 indicated deficiencies. The records indicated that a weekly examination of the mine was not conducted during the week of December 14, 2005. The examiner conducting the weekly examination on November 23, 2005, failed to make the required air reading where air leaves the main return at the mouth of 1st Left. The mine examiner conducting the weekly examination for hazardous conditions found and recorded 0.2% methane in the air course at the 2 North Mains seals on December 28, 2005. He also stated that he found 1.2% methane exiting the sample pipe at the No. 10 seal. This was the only time he had found methane during his examinations. The mine examiner reported the incident to the mine foreman. On December 30, 2005, the mine foreman found 0.2% methane in the split of air ventilating the seals. Section 75.312 requires a daily main mine fan examination to assure electrical and mechanical reliability of each main mine fan and its associated components. This includes the devices for measuring or recording mine ventilation pressure. A trained person designated by the operator shall examine the fan for proper 115 operation at least once each day unless a fan monitoring system is used. Hourly and management employees are trained by the operator to conduct these examinations. Interviews with mine personnel and a review of the daily fan pressure recording charts indicated the operator failed to change the main mine fan pressure recording chart before the beginning of a second revolution on four occasions during the last quarter of 2005. The required test of the automatic fan signal device was not performed by stopping the fan every 31 days. MSHA’s underground and surface standards require the operator to examine and test electrical equipment at specific intervals. Section 75.512 requires all underground electrical equipment to be examined and tested at least weekly by a qualified person to assure safe operating conditions. Five pieces of equipment were not tested or examined consistently on a weekly basis. Section 75.900-3 requires all low- and medium-voltage circuit breakers and their auxiliary devices to be tested and examined by a qualified person on a monthly basis. The circuit breaker that protected the 58 horsepower (hp) pump was not tested or examined at least once each month. Section 75.900-4 further states that each breaker test, examination, repair, or adjustment will be noted in a written record. The record of the tests of all circuit breakers did not list each breaker individually. Section 75.800-3 requires the testing and examination of high-voltage circuit breakers and their auxiliary devices protecting underground circuits by a qualified person on a monthly basis. Section 77.502 requires all surface electrical equipment to be examined, tested and properly maintained by a qualified person at least monthly, to assure that it is in safe operating condition. Training The approved Part 48 Training Plan for underground and surface areas of the mine was evaluated to assure that the plan met the requirements of Section 48.3 and Section 48.23. Course material, course outlines, evaluation methods, visual aids, and equipment available for use by the instructor(s) as required by Section 48.3(e) and Section 48.23(e) were reviewed. A review of all Task Outlines was conducted for each position at the mine, as required by Section 48.3(b) (8) and Section 48.23(b) (8). Evaluations were done of the mine operator’s MSHA Form 5000-23, Certificates of Training, with emphasis on the 1st Left and 2nd Left Parallel crews. The course material, course outlines, evaluation methods, visual aids, and equipment used for training were reviewed to assure that all items listed in the Approved Part 48 Training Plan were available for use by the instructor(s) as required by Section 48.3(e) and Section 48.23(e). The approved Part 75 and Part 77 training plans for certified and qualified persons were reviewed. Course materials and course outlines for Part 75 and Part 77 including but not limited to, Principles of Mine Rescue, Provisions of Part 75 and Part 77, and task training as required by Section 75.161 and Section 77.107-1 were evaluated. 116 A list of miners that carry a methane/oxygen detector was requested from the mine operator. These miners were checked on the MSHA Standardized Information Systems (MSIS) to ensure that they had been tested as required by Section 75.151 and Section 77.102. The electrical retraining plan for underground and surface as required by Section 75.153(g) and Section 77.103(g) was reviewed. A list of electricians at the mine was checked on MSIS for up-to-date certifications. All instructors that conducted training for the mine, which included Part 48 Approved Instructors and Electrical Instructors, were checked for up-to-date qualification on the MSHA MSIS program. The Mine Emergency Evacuation and Firefighting Program of Instruction was reviewed and compared to the course outlines in Sections 48.25, 48.6, and 48.8, to assure that the outline addressed the needs of the miners. The following is a list of deficiencies that were found: • Ten miners whose job duties required testing for methane had not demonstrated to the satisfaction of an authorized representative of MSHA that they were qualified to test for methane; • The annual refresher training was not adequate. A miner was not provided with hands-on SCSR training; • Underground electrical qualification retraining was conducted at the mine without an approved underground electrical retraining program; • Surface electrical qualification retraining was conducted at the mine without an approved surface electrical retraining program; • A form 5000-23 was signed by a miner and a qualified instructor verifying that annual refresher training had been completed when in fact no training had been given; and • Six miners did not receive any annual retraining as required. Communications Equipment The mine used several communication systems. The dispatcher’s office was the communications hub. Verizon supplied telephone service to the surface office buildings and the dispatcher’s office. The dispatcher had the capability to route the Verizon service into the mine through the mine phone system. The underground mine phone system was comprised of pager phones, which were located throughout the mine and in the working sections, as well as in the pit area, dispatcher’s office and other mine offices. Any pager phone on this system could page to all of the other pager phones. A conversation between any two people using these phones could be heard from any of the other pager 117 phones. This system allowed a number of miners to communicate with each other simultaneously. Mine pager phones were connected together by two wires. Each phone had a battery. If the battery was disconnected or depleted, then that phone would not operate. If the wiring became severed, then there would no longer be two-way communications between the phones inby the damage and the phones outby the damage. However, the phones outby could communicate with each other and the phones inby could communicate with each other. The mine also employed a Gai-Tronics Corporation trolleyphone communication system. These phones were located in the dispatcher’s office and on the battery powered rail mantrips and locomotives. Although the system was referred to as a trolleyphone system, there was not an electrically-powered trolley system at the mine. The trolleyphone system used an antenna wire, a carrier repeater and an electrical connection to the track at the drift opening and at the carrier repeater. The antenna wire was installed on the mine roof above the track. The carrier repeater was used to amplify the signal to maintain trolleyphone communication throughout the mine. It was installed in the crosscut between the No. 4 Belt entry and the No. 5 track entry, 9 Crosscut, No. 4 Belt. The carrier repeater was powered by 120 volts received from the No. 4 Belt power center installed at the same location. The trolleyphone system allowed communication between miners on mantrips and locomotives, and the dispatcher. This system could receive communications from the pager phone system, but could not transmit to it. When needed, the dispatcher would relay communications between the two phone systems. The trolleyphone system would not operate if the carrier repeater was deenergized. For example, if the power center for the No. 4 Belt drive was deenergized, the repeater would be de-energized and the trolleyphone system would not operate. If the antenna wire to the system was damaged, trolleyphones inby the damage would not function, but those outby the damage might. Motorola two-way handheld radios were used on both sections. The two-way radios would not interact with any other communication system. MSHA personnel indicated the radios may have a maximum range of 1,500 feet within the same entry, with severely limited range around corners. This range is highly dependent on coal seam height, entry geometry, and infrastructure within the entry. Battery strength also affects the range of the radios. One miner stated that the units had a range of about 1,000 feet when in direct line of sight and less than that distance when not in direct line of sight. 118 The dispatcher and the yardman each had a handheld radio. These radios could transmit and receive communications with each other and the pager phones via the Interlink 3000 unit located within the dispatcher’s office. The radios could also receive alerts and alarms from the AMS. The dispatcher used the radio when his assignments required him to leave the dispatcher’s office. Equipment Status The pager phone system was operational prior to the accident. The explosion damaged wiring and several pager phones. The most outby damage to the wiring occurred approximately 50 feet inby the 1st Left track switch, near survey station 3869. Pager phone communication inby this point to 2 North Mains and 2nd Left Parallel was no longer possible. Information on the mine pager phone is included in Appendix S, which is an executive summary of a report entitled “Executive Summary of Inspection of Sago Mine Voice Communications Equipment.” At the start of the day shift on January 1, 2006, the trolleyphone system did not function. The carrier repeater for the system lost power. At about 8:00 a.m., a maintenance foreman reset the circuit breaker and the trolleyphone system worked. The trolleyphone system was working at the end of this shift. The dispatcher indicated that the trolleyphone system again failed to function on January 2, 2006. Before the accident, he only heard static on the system. After the explosion, the carrier repeater lost power. The most outby damage of the antenna wire was approximately 20 feet inby survey station 3854, located near 50 Crosscut, No. 4 Belt. The trolleyphone system could not be used to communicate with the mantrip used by the 2nd Left Parallel crew. During the investigation, the carrier repeater was removed from the mine and tested, and was found to be functional. The executive summary of the reports for the trolleyphone system are contained in Appendices S and T. The 1st Left crew was located at the track switch when the explosion occurred. It is over 1,400 feet from the 1st Left track switch to the location of the 2nd Left Parallel mantrip. These two locations were not in a direct line of sight. Therefore, it is not likely that the 1st Left and 2nd Left Parallel crews could have communicated with each other with the radios. An “Executive Summary of Investigation of the Motorola Two-way Radios” is contained in Appendix U. 119 Mine Rescue Communications The following surface locations at the mine had pager phones during the mine rescue operations: • • • • • • • • • Command center (mine superintendent’s office) Maintenance superintendent’s office Small office behind the maintenance superintendent’s office Mine foreman’s office Foremen’s office Dispatcher’s office MSHA’s mine rescue vehicle (phone was connected between 6:00 p.m. and 12:00 midnight on January 2, 2006) WVMHS&T’s mine emergency vehicle (phone connected during rescue efforts) Building in mine pit (phone disconnected at approximately 6:58 a.m. on January 3) A command center was established at about 1:00 p.m. on January 2, 2006 in the mine superintendent’s office. Underground Mine Rescue Communications Mine rescue teams used Motorola two-way handheld, MSHA approved permissible radios. During this rescue operation, MSHA provided four units, but one malfunctioned. Interviews conducted with each MEU team member and their surface support personnel determined that the handheld permissible radio communication system performed as expected. Communications are difficult in mine rescue scenarios where rescuers are wearing full face masks. Literature provided by the radio manufacturer discusses the range of the radios in general terms. The literature states that more power will increase the range. As the batteries discharge power, the range of the radios will decrease. In addition, proper tuning will increase the range of the radios. The range is shorter in a building than it is when used outside in an area with no obstructions. The range for the permissible radios is similar to that of the non-permissible radios discussed previously when used underground. 120 Seismic Location System Introduction In 1970, the National Academy of Engineering (Academy) reported that a seismic system might be able to detect and locate trapped miners. The Academy proposed that a miner could strike part of the mine with a heavy object and the resulting vibrations could then be detected on the surface by using seismic transducers or geophones. The vibrations would be converted into electrical signals by the geophones and then amplified, filtered, and recorded. By comparing the arrival times of the signal at several different geophone locations, the trapped miner could be located. In 1971, the Westinghouse Electric Company built and tested a truck-mounted system. From 1972 until 1981, Westinghouse, MSHA and the USBM modified and tested the system in a variety of mines. There were 15 field tests conducted to define a signal model, background, noise levels, and geophone location performance. Since 1981, MSHA has conducted intermittent field tests to check and maintain operational familiarity with the system. Tests indicated that, under certain conditions, the truck-mounted system can be an effective means of detecting and locating trapped miners. Signals from miners pounding on the roof of a mine can be of sufficient strength to enable detection over an area of the mine. The signals are affected by ground conditions, the depth of the mine, and seismic noise sources. Estimations of the location of the trapped miner can be of sufficient accuracy to aid the rescue team or aid in the positioning of the rescue drill.34 However, a significant amount of time is required to set up the system and conduct an accurate survey. MSHA’s truck-mounted seismic location system is maintained by personnel from the Pittsburgh Safety and Health Technology Center of MSHA’s Technical Support. The seismic equipment, as well as the other related mine emergency equipment and personnel, is not automatically deployed when a mine emergency, such as a fire or explosion, occurs. The deployment of the equipment is based on the preliminary information received about the nature of the mine emergency, and is often made based on consultation with Technical Support personnel. 34 Evaluation of the Seismic System for Locating Trapped Miners , Bureau of Mines Report of Investigations, RI 8567, John Durkin and Roy J. Greenfield, (1981). 121 A minimum of six people are required to prepare and operate the system in a timely fashion. The Chief, Mine Emergency Operations (MEO), directs the setup and operation of the system, assisted by two Technical Support personnel. Several MSHA MEU team members have also been trained to assist in the setup and operation of the system. However, the use of the MEU at a mine emergency for this purpose could reduce the resources available for mine rescue exploration. In March 1977, during the rescue efforts at the Porter Tunnel Mine Inundation near Tower City, Pennsylvania, the MSHA truck-mounted seismic system was deployed and was not able to detect signals from a trapped miner, due to seismic noise and overburden conditions, using geophones installed on the surface over the mine. MSHA installed cables and geophones from the surface into the mine, attempting to receive signals from miners. This also was not successful. This event prompted MSHA to develop a mini-seismic system in the 1980’s. This system was designed to be quickly deployed. It is portable and designed to be taken underground and used by mine rescue teams. The system can be carried by two people and will easily fit in a small truck. However, the mini system has very limited capabilities, employing only 4 geophones. It cannot pinpoint the specific location of miners, but may detect their presence in some situations. It was not designed to be used from the surface of a mine. However, when used in this configuration, it can detect signals at a very limited depth, reportedly less than 200 feet. System Deployment Following a mine disaster in which it has been determined that use of the seismic location system would be helpful and is requested to be deployed, the system is transported to the mine site. A geophone array is positioned over the suspected area of entrapment. Each of the seven geophone sub-arrays must be accurately surveyed and tied to the mine survey. A refraction survey must also be performed to determine the ground velocities. In order to improve the possibilities of detecting and locating a trapped miner, the geophones should be placed around the miner’s most likely location. If the trapped miner is not within the area covered by the geophones, he may still be detected, but determining his location accurately may be more difficult. The system does not give an exact location for the trapped miners. Information from the system, along with the underground mine maps, helps determine where miners may be located. The accuracy of the system is limited to 50-100 feet. Telemetry is used to connect the system base station with the geophone arrays. It is important to locate the geophones away from any vehicle or personnel activity during attempted reception of seismic signals, because they interfere 122 with signal reception. Other natural and man-made seismic noise sources hinder the system’s ability to detect signals from trapped miners Mine Emergency Evacuation and Firefighting Program of Instruction The Program provided that when miners are trapped by toxic gases from fires or explosions and are able to take refuge where the air is comparatively good, they should make every effort to protect themselves from deadly, poisonous gases by erecting a barricade or bulkhead. The miners behind the barricade should do the following: 1) 2) 3) 4) Listen for three shots, then Signal by pounding hard on the roof 10 times. Rest for 15 minutes, and Repeat . . . . until 5 shots are heard which would indicate that you have been located. 2nd Left Parallel Crew After attempting to evacuate, the 2nd Left Parallel crew built a barricade in the face area of 2nd Left Parallel. The miners used a sledgehammer to pound on a roof bolt. Investigators found the sledgehammer and an obviously beaten roof bolt in the barricade. McCloy indicated that they took turns pounding but he was unable to provide a time as to when they started or stopped. It is likely that they started pounding in the morning of January 2, and stopped in the afternoon or evening of that same day. The exact timeframes are unknown. System Response MSHA headquarters personnel contacted the Chief, MEO, Dr. Jeffery Kravitz at about 10:15 a.m. Only limited information was available at the time, including the fact that an explosion may have occurred at the mine, that a number of miners underground had not been accounted for, and that miners had gone underground after the event. Based on this information, headquarters personnel requested Kravitz to dispatch MSHA’s mine rescue and gas analysis equipment and personnel to the mine. Kravitz’s first priority was to notify MSHA district managers to request that their mine rescue team members respond to the mine. He then started contacting the required MEU members at their homes. At about 12:30 p.m., Kravitz instructed his staff members to prepare the truck-mounted seismic system for possible deployment to the mine. He called the trucking company that hauls the supply trailer, which is an integral part of the system, and put them on alert. At 2:00 p.m., Kravitz traveled to Pittsburgh and then went to the Technical Support 123 offices. The mini-seismic system was readied for deployment in the event it was needed. He departed with it at 5:15 p.m., arriving at the mine at 8:30 p.m. MSHA officials at the mine had gathered information about the accident throughout the day on January 2 and updated headquarters staff on the situation. They learned that the 1st Left crew and the other miners who were outby 1st Left at the time of the explosion evacuated the mine safely. They concluded that an explosion had occurred and that the 2nd Left Parallel crew did not evacuate. Miners entered the mine, found damaged ventilation controls that had short circuited the ventilation system, and made temporary repairs to those controls to advance the ventilation in the mine to the mouth of the 2nd Left Parallel. At this location, they encountered smoke, elevated CO concentrations and insufficient ventilation current to continue, and evacuated the mine. The early information indicated that the explosion occurred somewhere on 2nd Left Parallel and that the miners were still located there. Mine rescue teams arrived at the mine throughout the day. The mine operator had started work on surveying the area on the surface over 2nd Left Parallel to drill a borehole, but the survey effort was hampered by conditions and the lack of appropriate survey equipment on site. Although there was an initial upward trend in the gas concentrations at the monitoring locations, the trend eventually went downward, making it likely that mine rescue teams could enter the mine. Based on this information, MSHA officials decided that the approximate location of the miners was known and that mine rescue teams would be able to enter the mine if the downward trend continued. The truck mounted seismic system would take over eight hours to set up once a surveyed location was determined, and all rescue operations, including drilling, would have to cease during the test. Therefore, the truck-mounted seismic system was not deployed to the mine site. The terrain and depth of cover over the 2nd Left Parallel made use of the miniseismic system from the surface impractical, so it was not used. Preparations for a borehole into 2nd Left Parallel were ongoing. The mine rescue teams were progressing steadily underground and did not need it. Seals Manufacturing and Testing of Omega Block Seals are constructed in underground coal mines to separate the worked-out areas from the active workings. Stoppings and other ventilation controls are also constructed to direct ventilation through the mine. Seals, stoppings, and other controls can be constructed from a variety of materials, provided that these materials and the methods of construction have been deemed suitable by MSHA. In order for MSHA to determine that seal materials and the methods of 124 construction were suitable, full scale seals were constructed and tested underground in NIOSH’s Lake Lynn Experimental Mine (Lake Lynn). Lake Lynn is an underground limestone mine that was converted into a federal research facility. Prior to the accident, seals had been built from various materials and tested at Lake Lynn by utilizing different methods of construction and subjecting the seals to explosions generating a static pressure of 20 psi or more. This 20 psi testing pressure was required by federal regulation and was based on USBM research. MSHA accepts materials for use as seals provided they are constructed in the same manner as tested. Materials such as solid concrete blocks, wood, pumpable cementitious materials, and lightweight blocks, such as Omega blocks, had passed this explosion testing prior to December 31, 2005 and been accepted for use as seals. If an explosion occurs in direct line with any seal, the total pressure from the explosion is exerted on the seal. The total pressure is the sum of the static pressure and dynamic pressure. The static pressure is pressure exerted in all directions. The dynamic pressure is the pressure exerted by the movement of gases, or wind pressure. For example, an explosion in an entry exerts a static pressure only on seals destroyed in crosscuts and exerts the total pressure on seals destroyed in the same entry. During explosions, seals are exposed to either, 1) the static pressure only if the seal is not in the direct line of the explosion, or 2) the total pressure, including both static and dynamic pressure, if the seal is in the direct line of the explosion and is destroyed. Omega blocks are lightweight, polyester fiber-reinforced blocks manufactured by Burrell Mining Products International, Inc. (Burrell). The nominal size of a single block is 8 inches by 16 inches by 24 inches, weighing between 40 to 50 pounds. Laboratory testing has shown that Omega blocks are noncombustible. Figure 17 shows a picture of an Omega block. Figure 17 - Picture of an Omega Block 125 MSHA initially approved Omega blocks as a construction material for stoppings. However, full-scale testing at Lake Lynn revealed that Omega blocks could be used to build seals that withstand a static horizontal pressure of 20 psi. Since 1990, various configurations of Omega block seals had successfully passed testing. The initial Omega seal which passed explosion testing was 24 inches thick and included a center pilaster and hitching. In 2001, a 40 inch thick Omega seal without a pilaster or hitching passed explosion testing. The proper construction of Omega seals will be detailed in a subsequent section of this report. Burrell manufactures Omega blocks at plants located in Bluefield, West Virginia; Garards Fort, Pennsylvania; and Price, Utah. Burrell produces other products at these plants as well. At each plant, the manufacturing occurs in a facility adjacent to an enclosed storage area. After manufacturing, the Omega blocks are protected from the environment in an enclosed storage area. Suppliers provide the necessary ingredients to each plant for the manufacture of Omega blocks. The ingredients include Portland cement, water, foaming agent, polyester fiber, and Type F fly ash. The cement and fly ash are very fine powders. A computer-controlled system combines the ingredients into a batch mix. After appropriate quantities are entered, a mixing process occurs, which results in a uniform distribution of ingredients throughout the mix. The batch is discharged from the mixer, and a Burrell employee pumps it into forms. The employee must maintain the discharge hose in continual motion to properly fill the forms. As individual forms are filled, they are moved from the filling area to a holding area for approximately 24 hours. This period allows the product to harden to the point where the forms can be removed. A full curing period is 28 days due to the cement in the mix. After 24 hours, pallets of filled forms are individually positioned at a large, electronically-controlled band saw. According to Burrell, allowing the material to cure longer than 24 hours prior to sawing would cause excessive wear on the saw. When the forms are removed, the material is cut into 8 inch by 16 inch by 24-inch sections. As individual pallets of Omega blocks complete the sawing phase, they are subjected to a quality control check. Initially, visual observations are made of each pallet load of cut Omega blocks. Any Omega blocks with defects or obvious differences in dimensions are removed from the pallet and discarded. A single block from each pallet is examined for size and weight. During this phase, Omega blocks generally weigh between 45 and 47 pounds. Blocks must weigh between 40 pounds and 50 pounds to be acceptable. Up to five pounds of water loss may occur in individual blocks during the curing phase. 126 After the quality control check is completed, a shrink wrap is fitted to each pallet load of Omega blocks. This wrap protects the Omega block from atmospheric conditions, such as precipitation, and serves to maintain the integrity of the block during the curing and shipping process. An identification tag is affixed to each pallet with a date stamp marked on it to show the manufacture date. Each pallet is moved to a storage area where it is kept for at least two weeks before shipping. This two week period allows for continued curing of the Omega blocks. Omega blocks are shipped directly to underground coal mines or to mine supply distributors. Uniaxial compressive strength tests were conducted on Omega blocks from a variety of sources. The purpose of the testing was to establish whether any strength differences existed between dry and wet block, between new block from each of the three plants, between blocks from lots used previously at Lake Lynn, or between blocks cored from different sides. Preparation and testing was conducted by MSHA’s Roof Control Division at their Bruceton facility. The complete results of all uniaxial compressive strength tests of Omega blocks are contained in a Report of Laboratory Testing dated July 11, 2006. Appendix V is a copy of the executive summary of that report. The Omega blocks were received from ten (10) separate locations as follows: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Burrell’s Bluefield, West Virginia plant An underground coal mine in Utah NIOSH’s Lake Lynn – 2002 NIOSH’s Lake Lynn – 2006 Sago Mine –2 North Mains seal remnants Sago Mine – Supply yard blocks dated 2004 Sago Mine – Supply yard blocks dated 2005 Sago Mine – Supply yard loose blocks undated Burrell’s Price, Utah plant Burrell’s Garards Fort, Pennsylvania plant Burrell does not conduct compressive strength testing on any Omega blocks manufactured at any of their three plants. Therefore, no direct comparisons could be made between the Omega blocks tested as a part of this investigation and results of past testing during the manufacturing phase. A range of compressive strengths between 45 psi and 120 psi is typical for Omega blocks. Of the 109 samples tested, 108 (99.1%) samples fell within or exceeded the expected range. Only one (0.9%) sample fell below expectations. The results indicate that there are no differences in the average compressive strengths between wet and dry specimens or between cores removed horizontally or vertically with a drill. Core orientation had little influence on compressive strength since the Omega material is a mixed product poured into a 127 mold. Sample degradation (i.e. surface cracking) was observed as samples dried. However, moisture content did not influence the compressive strength. Seal History and Construction Federal regulations require that areas of underground coal mines be ventilated or sealed. Sealing eliminates exposure to hazardous conditions, such as adverse roof conditions, and allows for areas to be abandoned where mining has ceased. Sealing eliminates the need to ventilate and examine sealed areas. Many underground coal mines choose to construct seals. Seals are to be constructed according to the federal regulations contained in Section 75.335. In addition, Section 75.335 (a) (2) permits seals to be constructed using alternative methods or materials if they can withstand a static horizontal pressure of 20 psi.35 The method of installation and the material used are approved in the ventilation plan. Prior to 1992, federal regulations stated that pending the development of specifications for explosion-proof seals or bulkheads, seals or bulkheads could be constructed of solid, substantial, and incombustible materials sufficient to prevent an explosion that may occur on one side of the seal from propagating to the other side. There were no performance standards prior to 1992 that defined seal construction. However, in 1992, MSHA promulgated revised safety standards for underground coal mine ventilation. The standards included a 20 psi static horizontal pressure requirement on seals constructed of alternative methods or materials. The 20 psi requirement was based on USBM Report of Investigations (RI) 7581 entitled “Explosion-Proof Bulkheads.” According to RI 7581, a seal or bulkhead may be considered explosion proof when its construction is adequate to withstand a static load of 20 psi, if there is sufficient incombustible material on both sides of the seal to abate the explosion hazard. With adequate incombustible material and minimum coal dust accumulations, USBM considered it doubtful that pressures exceeding 20 psi could occur very far from the origin of the explosion. The coal mining industry and the general public were afforded the opportunity to comment on the proposed ventilation regulations before those regulations became effective. The regulations were intended to prevent explosions on either side of a seal from propagating to the other side. MSHA partnered with NIOSH to develop a full-scale seal-testing program at Lake Lynn. Figure 18 is a sketch of Lake Lynn. Alternative seal designs have 35 MSHA has since issued an interim requirement that newly constructed seals must withstand a 50 psi overpressure. MSHA PIB No. P06-16. 128 been tested and determined to meet the requirements of 75.335(a)(2). Seals that MSHA has determined to be suitable for construction in underground coal mines included seals constructed of Omega blocks. All seals that are deemed suitable for construction in underground coal mines must be constructed in the same manner as those that passed explosion testing at Lake Lynn. The size limitations for all seals are not to exceed 8 feet in height or 20 feet in width. Seals can be constructed in larger openings but they must be evaluated by MSHA on a caseby-case basis prior to installation. Deviations in the method of construction or the materials used result in an untested seal with strength characteristics that may not be appropriate. Consequently, such seals are not suitable for construction in underground coal mines until they successfully pass testing. Figure 18 - Sketch of the Lake Lynn Mine When the full-scale testing program was initiated, manufacturers submitted their intended designs to MSHA and the USBM. Seal designs were evaluated to determine whether their intended purpose could be met. Seals were constructed underground at Lake Lynn and provided time to cure. The USBM documented the steps necessary for construction. Air leakage guidelines were developed regarding the amount of air leakage that would be acceptable at various air pressure differentials. Visual observations and air leakage tests were used to determine if the seal met the regulatory requirements. The guidelines show that an air leakage of up to 100 cfm is acceptable at an air pressure differential of one 129 inch water gauge and up to 250 cfm is acceptable at an air pressure differential of four inches of water gauge. A pre-explosion air leakage test was conducted. Seals were required to meet or exceed the guidelines. Afterwards, an explosion was initiated which generated a pressure of about 20 psi static horizontal pressure on each seal. Subsequent to testing, seals were again required to meet the air leakage guidelines. Manufacturers shared their successfully tested designs with mine operators. Mine operators submitted some of these designs for inclusion in their ventilation plan to MSHA for approval. The MSHA district office could contact MSHA Technical Support for technical information and guidance on any specific seal design prior to approval. MSHA Technical Support provided training, distributed technical information, and responded to specific inquiries regarding seal construction. The manufacture and testing of individual Omega blocks has been described in a previous section of this report. Omega blocks had been found suitable for seal construction, when mortared together with BlocBond in the proper configuration. The first Omega block seals which passed explosion testing were 24 inches thick, including a 48 inch square center pilaster and hitched six inches deep along both ribs and the floor. A pilaster is an additional center column built of Omega blocks from floor to roof as an integral part of any individual seal. Hitching is accomplished by cutting a trench along the floor from rib to rib and cutting a trench in each rib from roof to floor. The seal is to be set into the hitch as a means to prevent perimeter failures. Attaching angle iron to both ribs and the floor on both sides of the seals is an acceptable method for providing artificial hitching. The 24 inch thick Omega block seals successfully passed explosion testing and were deemed suitable for construction in underground coal mines. In 2001, the 40 inch thick Omega block seal design without a pilaster or hitching passed explosion testing at 20 psi. As with other seals, there was no attempt to test these seals to their maximum strength. Consequently, the maximum explosive force which a 40 inch thick Omega block seal could withstand was not determined at that time. The construction of the 40 inch thick Omega seal is documented in a NIOSH publication titled, “Designs for Rapid In-Situ Seals” and includes adequate site preparation, roof support, and the following necessary factors: 1. No hitching was used. 2. Joints were staggered. 3. Final seal thickness was 40 inches plus the thickness of face coatings. 130 4. BlocBond, a high-strength mortar, was applied ¼-inch thick as a mortar for all vertical and horizontal joints and as a face coating on both sides of the seal. 5. No pilaster was used. 6. The gap between the top of the seal and the roof was about 2.5 inches. 7. Three rows of 1 inch thick by 8 inch wide by 10 feet long wood planks were run lengthwise from rib to rib across the top of the seal. One row was placed in the middle of the seal and two rows were placed symmetrically on each side with their respective edges flush with the inby and outby side of the seal. Each row was wedged on about 1 foot centers and the gaps between wedges and between wood rows were filled with BlocBond. At the Sago Mine, the mine operator planned to construct seals across the nine entries of the 2 North Mains, which would effectively seal the inby areas of the 2 North Mains and all of the 2nd Left Mains. As a result, the mine operator submitted a plan detailing the construction of 40 inch thick Omega block seals. The plan, which was approved by MSHA, provided details on the method of construction of the 40 inch thick Omega seal. The applicable addendums to the plan are included in Appendix K. The plan included the following: 1. No hitching was to be used. 2. Total thickness of the completed seal shall be 40 inches. 3. Joints were to be staggered. 4. All joints shall be a minimum ¼-inch thick and be mortared using BlocBond. 5. Three rows of wood planks running the entire length of the seal shall be installed across the top of the seal. 6. Wedges will be placed on one foot centers or less with BlocBond used to fill the gaps. 7. BlocBond shall be used as full face coating on both sides of the seal. 8. The opening where the seal is to be constructed was limited to 8 feet in height and 20 feet in width. 9. Seals shall be at least 10 feet from the corner of the pillar. Subsequently, the mine operator submitted plans for the construction of Omega Block seals in locations where the opening is up to 10 feet high and 20 feet wide and also where the opening is up to 12 feet high and 20 feet wide. However, these plans were intended for future seal locations and not for the seals constructed in 2 North Mains. The method of construction for these larger designs was never utilized by the mine operator. The dimensions of the locations in 2 North Mains where the ten seals had been constructed were measured and are listed in Table 9. 131 Table 9 - Dimensions of the 2 North Mains Seals Seal No. 1 2 3 4 5 6 7 8 9 10 Maximum Width (feet) 21.7 20.4 19.7 18.9 18.8 19.5 19.2 19.6 19.1 18.3 Maximum Height (feet) 8.9 8.7 7.4 7.3 7.2 7.4 7.5 6.3 6.7 6.3 Testimony indicated that: • Mine management knew prior to seal construction that the location of the No. 1 seal exceeded 20 feet in width. • Up to three inches of dry BlocBond was spread on the floor prior to seal construction. • Each course was laid dry and mixed mortar was spread across the top of each course and an attempt was made to force mortar into the vertical joints by hand. This is shown in Figure 19. The darker material is the BlocBond, which only slightly filled the vertical joint. • Three wood planks were not always used on top of seals and wedges were not always installed properly. Figure 19 - Mortar in Vertical Joint Similar seals were constructed at Lake Lynn and withstood a 21 psi explosion. The pressures created by the explosion at Sago Mine significantly exceeded 20 psi. The differences in seal construction, listed above, did not affect their ability to withstand the explosion. 132 Statements indicated that dry BlocBond was spread across the mine floor at each seal location as the initial step in seal construction. A dry powder such as BlocBond must be properly mixed with quantities of water designated by the manufacturer to form mortar. The quality of the BlocBond observed after the explosion varied. The BlocBond which remained on the ribs appeared to be properly mixed. It remained attached to the ribs after the explosion. It was dark gray to black and was extremely difficult to remove. The BlocBond observed on the floor, after Omega blocks were removed, was light gray and easily removed. Core samples were removed from the floor at each of the ten seal locations. These samples were submitted to an independent laboratory to establish the quality and composition of the mortar in the setting beds. The laboratory studies included petrographic examinations, visual examinations, and compressive strength testing. A memo and executive summary of the report on the “Sampling and Testing of Mortar Bed Cores Taken from Failed Ventilation Seals” is included in Appendix W. The average compressive strength of the mortar cast in the laboratory exceeded 8000 psi. Only one mine core sample had a comparable compressive strength, however, the remaining mine core samples only had strengths from 830 to 2810 psi. Strength discrepancies in the mine core samples occurred because of inadequate mixing, incorrect water contents, inclusion of extraneous materials, or from fissures or tears that occurred after the mortar stiffened. The ten 2 North Main seal locations were evaluated during the investigation. The post-explosion location of Seal No. 1 is shown in Figure 20. Several whole and partial Omega blocks remained at the location of Seal No. 1 after the explosion. An exposed horizontal layer of BlocBond was easily removed from the remaining Omega blocks, indicating the lack of good bonding. No BlocBond was observed in some vertical joints. This seal had been constructed on a diagonal and was not perpendicular to either rib. Figure 20 - Post-Explosion Location of Seal No. 1 133 The post-explosion location of Seal No. 2 is shown in Figure 21. Several Omega blocks remained at the location of Seal No. 2 after the explosion. Unburned paper material was observed imbedded within the mortar along one rib and also in a remaining joint between Omega blocks. No significant thickness of BlocBond was found between remaining Omega blocks. The vertical joint between two Omega blocks included BlocBond for only approximately 25% of the joint. Figure 21 - Post-Explosion Location of Seal No. 2 The post-explosion location of Seal No. 3 is shown Figure 22. BlocBond on the rib was difficult to remove, indicating proper mixing prior to application. BlocBond on the floor had very little strength, indicating improper or no mixing with water prior to application. Figure 22 - Post-Explosion Location of Seal No. 3 134 The post-explosion location of Seal No. 4 is shown in Figure 23. BlocBond was observed on the floor as a smooth surface, indicating a lack of adherence to the Omega block. Figure 23 - Post-Explosion Location of Seal No. 4 The post-explosion location of Seal No. 5 is shown in Figure 24. BlocBond and Omega blocks were set on loose floor material at this location. Figure 24 - Post-Explosion Location of Seal No. 5 135 The post-explosion location of Seal No. 6 is shown in Figure 25. It appeared that pieces of Omega block were used, along with dry BlocBond, to level the floor prior to construction of the seal. Figure 25 - Post-Explosion Location of Seal No. 6 The post-explosion location of Seal No. 7 is shown in Figure 26. Very little BlocBond was observed on the ribs. The BlocBond was difficult to remove, indicating good strength characteristics. Figure 26 - Post Explosion Location of Seal No. 7 136 The post-explosion location of Seal No. 8 is shown in Figure 27. Very little BlocBond was observed on the ribs. The BlocBond was difficult to remove, indicating good strength characteristics. Roof conditions deteriorated during the investigation at this location. Figure 27 - Post-Explosion Location of Seal No. 8 The post-explosion location of Seal No. 9 is shown in Figure 28. Several Omega blocks remained at the location of Seal No. 9 after the explosion. Unburned plastic and paper material was observed imbedded within the mortar along one rib. A coating of dry, unmixed BlocBond was observed on the floor. Figure 28 - Post-Explosion Location of Seal No. 9 137 The post-explosion location of Seal No. 10 is shown in Figure 29. No Omega blocks remained on the floor. There were some indications of mortar on the ribs. Figure 29 - Post-Explosion Location of Seal No. 10 The actual construction of the ten seals was different from the requirements of the MSHA approved plan and from the initial NIOSH testing of 40 inch thick Omega block seals. Unburned plastic was imbedded inside the cured mortar along the rib, indicating that it was used as filler between the seal and the rib. Unburned paper material was also found imbedded in the cured mortar between the seal and the rib. Paper was found in some joints between Omega blocks. The dimensions of two of the ten seal locations exceeded the maximum approved dimensions of 8 feet high and 20 feet wide. One seal was not set back at least 10 feet from the corner of the pillar. After removal of remaining portions of Omega block from the floor, a layer of dry BlocBond material was evident. It appeared to be BlocBond that was spread on the floor dry and not BlocBond that had been mixed with water. Vertical joints were not coated with at least a ¼-inch thick application of BlocBond. Mortar was applied to the top horizontal surface and was spread by hand. Most of the mortar reaching the vertical joints was forced in by hand. During the underground investigation and, subsequently, during laboratory examination and testing, the limited extent of vertical joint mortar was noted as shown above in Figure 19. A center plank was not always incorporated into the top of the seal. The planks were not wedged properly, as required by the ventilation plan. In some cases, these planks did not extend from rib to rib. The space between planks and the space between wedges was not completely filled with BlocBond. In addition, wedges were sometimes driven between the Omega block and the wood plank. This forced the wedge into the Omega block rather than allowing the load to be more evenly distributed across the top of the seal. Wedges were also driven parallel to the wood planks rather than perpendicular, as shown in the approved plan. Although each wood plank would be completely wedged, 138 the wedges were placed skin-to-skin, which caused wedges to replace mortar. This procedure may have affected the strength of the seal. Seal Testing Due to these differences in the method of construction, MSHA requested NIOSH assistance in evaluating the explosion resistance of various Omega block seal designs. As a part of this investigation, all seals were constructed at Lake Lynn. The purpose of this testing was to establish whether any detrimental effects resulted from the differences in construction and to establish the magnitude of pressures that may have occurred during the explosion at Sago Mine. This was the first time that seals had been subjected to full-scale test explosions generating total pressures on the seals, and the first time tests were conducted within a completely sealed area. An executive summary titled “Experimental Study of the Effect of LLEM Explosions on Various Seals and Other Structures and Objects” is contained in Appendix X. Prior to the first test explosion, two 40 inch thick Omega block seals were constructed underground. A typical solid concrete block seal was constructed in No. 1 Crosscut. The Omega block seal in No. 2 Crosscut was constructed in the same manner as the one which successfully passed explosion testing in 2001. The Omega block seal constructed in No. 3 Crosscut incorporated several changes in the method of construction. These changes include applying unmixed mortar on the mine floor, not applying mortar directly to the vertical joints of the first course of blocks, and modifying the installation of wood planks and wedges between the last course of the Omega blocks and the mine roof. This second Omega seal, referred to as a hybrid seal, was not intended to accurately represent the seals that were destroyed at the Sago Mine. The seals were cured for 22 days. This represented the shortest curing period for any portion of the ten 2 North Mains seals. Two cribs were constructed in the entry just outby the No. 3 Crosscut. Belt hangers, roof plates, and roof bolt bearing plates were installed along the entry. A battery charger, removed from the Sago Mine, was located in the entry in which the explosion occurred. 139 The first test explosion was conducted on April 15, 2006 and generated a static pressure pulse of about 23 psi on the seal in No. 2 Crosscut, and 25 psi on the seal in No. 3 Crosscut. Each of the three seals successfully withstood the pressure pulse. The battery charger was moved outby by a distance of at least 21 feet as a result of this explosion. The crib blocks were blown a maximum distance of about 883 feet from their initial location as a result of this explosion. Figure 30 is a sketch of the Lake Lynn Mine layout for Test No. 1. Test No.1 - Sketch of NIOSH’s Lake Lynn Laboratory BD + + + CB X-1 + + + + + + OB + + + + + + + + + + + + + + + Drift C HB X-2 X-3 X-4 X-5 X-6 X-7 Drift B Drift A CB – Concrete Block Seal OB – 40-inch Omega Block Seal HB – 40-inch Hybrid Omega Block Seal - Cribs +++ - Roof Bolt Plates with Pie Pans and Belt Hangers Arranged Alternatively BD - - Bulkhead Door - Battery Charger Not to Scale Figure 30 - Test No. 1 Lake Lynn Mine Layout For the second test explosion, the solid concrete block seal and both of the Omega block seals from the previous test remained in place. In addition, a 40 inch thick Omega block seal was constructed in the Drift C outby No. 3 Crosscut. This third Omega block seal was constructed in the same manner as the one which successfully passed explosion testing in 2001. The construction of this third seal across the drift effectively sealed off the inby area. The seals cured for 28 days. No cribs were constructed as part of this test. All damaged roof plates, and roof bolt bearing plates were replaced. The explosion was initiated in the sealed area. The purpose of this test was to impact the seals in the crosscuts with a static pressure pulse and the seal in the drift with a total pressure pulse from the explosion. 140 The explosion generated a pressure of 22 psi on the seal in No. 2 Crosscut, 39 psi on the seal in No. 3 Crosscut, and 51 psi on the seal constructed in Drift C. The test was conducted on June 15, 2006. The solid concrete block seal and the Omega block seal in No. 2 Crosscut successfully withstood the pressure pulse. The Omega block seals in No. 3 Crosscut and in Drift C were destroyed. The battery charger was moved outby by a distance of at least 79 feet as a result of this explosion. The greatest distance that seal debris was thrown as a result of this explosion was about 822 feet. Figure 31 is a sketch of the Lake Lynn Mine layout for Test No. 2. Test No.2 - Sketch of NIOSH’s Lake Lynn Laboratory BD + + + CB + + + + + + X-2 + + + + + + + + + + + + Drift C OB HB OB X-1 + + + X-3 X-4 X-5 X-6 X-7 Drift B Drift A CB – Concrete Block Seal OB – 40-inch Omega Block Seal HB – 40-inch Hybrid Omega Block Seal - Cribs +++ - Roof Bolt Plates with Pie Pans and Belt Hangers Arranged Alternatively BD - Bulkhead Door - Battery Charger Not to Scale Figure 31 – Test No. 2 Lake Lynn Mine Layout For the third test explosion, the solid concrete block seal and the Omega block seal in No. 2 Crosscut remained in place. Omega block seals that were similar to those constructed at Sago Mine were constructed in No. 3 Crosscut and in Drift C just outby No. 3 Crosscut. These Omega block seals incorporated several changes in the method of construction. These changes include; applying unmixed mortar on the mine floor, not applying mortar directly to any vertical 141 joints, and modifying the installation of wood planks and wedges between the last course of the Omega blocks and the mine roof. The construction of this third seal across the drift effectively sealed off the inby area. One dry-stacked stopping was constructed just outby the seal in the drift and one dry-stacked stopping was constructed in No. 3 Crosscut, behind the seal. Two cribs were built on both the inby and outby side of the seal in the drift. All damaged roof plates, and roof bolt bearing plates were replaced. The seals cured for 28 days. The explosion was initiated in the sealed area. The purpose of this test was to impact the seals in the crosscuts with a static pressure pulse and the seal in the drift with a total pressure pulse from the explosion. The explosion generated a pressure of 13 psi on the seal in No. 2 Crosscut, 16 psi on the seal in No. 3 Crosscut, and 17 psi on the seal constructed in Drift C. The test was conducted on August 4, 2006. Each of the four seals successfully withstood the pressure pulse. Figure 32 is a sketch of the Lake Lynn Mine layout for Test No. 3. Test No.3 - Sketch of NIOSH’s Lake Lynn Laboratory BD CB X-1 + + + + + + + + + OB + + + # # + + + # # HB SB SB X-3 X-2 + + + Drift C + + + + + + B X-4 X-5 X-6 X-7 Drift B B Drift A CB OB SB – Concrete Block Seal - Battery Charger – 40-inch Omega Block Seal – 40-inch Sago Omega Block Seal +++ - Roof Bolt Plates with Pie Pans and Belt Hangers Arranged Alternatively B - Hollow Concrete Block Stopping - Cribs BD - Bulkhead Door Not to Scale Figure 32 - Test No. 3 Lake Lynn Mine Layout 142 For the fourth test explosion, the four seals from the previous test remained in place. The fourth test was designed to increase the static and total pressures. The explosion was initiated in the sealed area. The explosion generated a pressure of 15 psi on the seal in No. 2 Crosscut, 18 psi on the seal in No. 3 Crosscut, and 21 psi on the seal constructed in Drift C. The test was conducted on August 16, 2006. Each of the four seals successfully withstood the pressure pulse. Figure 33 is a sketch of the Lake Lynn Mine layout for Test No. 4. Test No.4 - Sketch of NIOSH’s Lake Lynn Laboratory BD CB X-1 + + + + + + + + + OB + + + # # + + + # # HB SB SB X-2 + + + X-3 + + + + + + Drift C B X-4 X-5 X-6 X-7 Drift B B Drift A CB OB SB – Concrete Block Seal - Battery Charger – 40-inch Omega Block Seal – 40-inch Sago Omega Block Seal +++ - Roof Bolt Plates with Pie Pans and Belt Hangers Arranged Alternatively B - Hollow Concrete Block Stopping - Cribs BD - Bulkhead Door Not to Scale Figure 33 - Test No. 4 Lake Lynn Mine Layout For the fifth test explosion, the four seals from the previous test remained in place. The fifth test was designed to significantly increase the static and total pressures. The explosion was initiated in the sealed area. The explosion generated a pressure of 26 psi on the seal in No. 2 Crosscut, 35 psi on the seal in No. 3 Crosscut, and 57 psi on the seal constructed in Drift C. The test was conducted on August 23, 2006. The solid concrete block seal and the Omega block seal constructed in No. 2 Crosscut successfully withstood the pressure pulse. Both of the Omega block seals that were similar to those constructed at Sago Mine were destroyed by the pressure pulse. The battery charger was 143 moved outby by a distance of about 30 feet as a result of this explosion. The crib blocks were blown a maximum distance of about 438 feet as a result of this explosion. Figure 34 is a sketch of the Lake Lynn Mine layout for Test No. 5. Test No. 5 - Sketch of NIOSH’s Lake Lynn Laboratory BD CB X-1 + + + + + + + + + OB + + + # # + + + # # HB SB SB X-2 + + + X-3 + + + + + + Drift C B X-4 X-5 X-6 X-7 Drift B B Drift A CB OB SB – Concrete Block Seal - Battery Charger – 40-inch Omega Block Seal – 40-inch Sago Omega Block Seal +++ - Roof Bolt Plates with Pie Pans and Belt Hangers Arranged Alternatively B - Hollow Concrete Block Stopping - Cribs BD - Bulkhead Door Not to Scale Figure 34 - Test No. 5 Lake Lynn Mine Layout For the sixth test explosion, the solid concrete block seal and the Omega block seal in No. 2 Crosscut remained in place. A solid concrete block seal was constructed in No. 3 Crosscut and a seal constructed of Omega blocks from the Sago Mine was constructed in Drift C. The sixth test was designed to significantly increase the static and total pressures. The seals cured for 28 days. Two cribs were built on both the inby and outby side of the seal in the drift. One dry-stacked stopping was constructed in No. 3 Crosscut, behind the seal. The explosion was initiated in the sealed area. The explosion generated a pressure of 51 psi on the seal in No. 2 Crosscut, 49 psi on the seal in No. 3 Crosscut, and 93 psi on the seal constructed in Drift C. The test was conducted on October 19, 2006. Both of the solid block seals successfully withstood the pressure pulse. The Omega block seal in No. 2 Crosscut withstood the pressure pulse. The Omega block seal constructed in Drift C was destroyed by the explosion. 144 The greatest distance that seal debris was thrown as a result of this explosion was about 918 feet. The battery charger was moved outby by a distance of about 356 feet as a result of this explosion. The greatest distance stopping debris was thrown, as a result of this explosion, was about 748 feet. Figure 35 is a sketch of the Lake Lynn Mine layout for Test No. 6. Test No. 6 - Sketch of NIOSH’s Lake Lynn Laboratory BD CB X-1 + + + + + + + + + OB + + + # # + + + # # HB SB CB X-2 + + + X-3 Drift C + + + + + + B X-4 X-5 X-6 X-7 Drift B Drift A CB OB SB – Concrete Block Seal - Battery Charger – 40-inch Omega Block Seal – 40-inch Sago Omega Block Seal +++ - Roof Bolt Plates with Pie Pans and Belt Hangers Arranged Alternatively - Hollow Concrete Block Stopping B - Cribs BD - - Bulkhead Door Not to Scale Figure 35 - Test No. 6 Lake Lynn Mine Layout The belt hangers were intended to simulate those belt hangers that were installed in the Sago Mine at the time of the explosion. When any belt hanger displayed damage during a test, it was replaced with a new belt hanger prior to the next explosion test. Maximum explosion pressures ranged from 17 psi to 93 psi. One belt hanger, located 403 feet outby from the explosion origin, was significantly bent in test No. 5 and No. 6. The damage was most likely caused from projectiles, such as crib blocks, striking it. Inby belt hangers were not damaged even though they were exposed to higher pressures. From these tests, it does not appear that significant damage to belt hangers can occur at pressures less than 93 psi. The test explosions have shown that solid concrete block seals can successfully withstand static explosion pressures of at least 49 psi. The Omega block seal that was constructed in the same manner as the one which successfully passed explosion testing in 2001 can successfully withstand static explosion pressures of at least 50 psi. Omega block seals constructed in a manner similar to those that were built prior to the January 2, 2006 explosion at Sago Mine can successfully withstand total explosion pressures of at least 21 psi. The Lake Lynn testing did not result in the same level of damage to the seals as observed at the Sago Mine. 145 Electrical Power and Equipment Electrical Power System The Allegheny Power Company supplied 138,000 volt alternating current (vac) electric power to the French Creek Substation, located approximately two miles from the mine, where it was reduced to a 12,470 vac solidly grounded system.36 The power circuit supplied two high-voltage circuit breakers installed in a fenced area adjacent to the French Creek Substation. The 12,470 vac was transmitted through surface transmission lines to a branch transmission line. The branch line had visible disconnects at its first pole and lightning arresters were installed at the next pole. This branch line extended to the Sago Mine substation. This circuit was protected by a high-voltage circuit breaker, visible disconnects, and lightning arresters. A spare circuit breaker was installed in the same fenced area and was not in use. Each circuit breaker contained relays designed to provide overcurrent, short circuit and grounded phase protection. Three 1,250 kva transformers located in the mine substation reduced the 12,470 vac to a 7,200 vac resistance grounded system37 for underground distribution. A high-voltage circuit breaker, visible disconnects, and lightning arresters located in the surface substation provided circuit protection for the underground distribution system. The circuit breaker contained relays designed to provide overcurrent, short circuit, grounded phase, under-voltage, and ground monitor protection. The power circuit was provided with visible disconnects and lightning arresters at the mine openings where it entered the underground area of the mine via a 4/0 American Wire Gauge (AWG) high-voltage cable. All connections between underground power centers were made with 4/0 AWG mine power, ground, ground check, 8 kilovolt (kv) rated high-voltage cable. The power circuit was further reduced by underground transformers to 995 vac, 575 vac, 480 vac, and 240/120 vac for use by underground electric equipment. The 12,470 vac was also reduced to 4,160 vac by three 167 kva pole-mounted transformers for the surface fan. The power circuit was provided with visible disconnects, fuses and lightning arresters. Three 100 kva pole-mounted transformers reduced the 12,470 vac to 480 vac and 240/120 vac for the surface electric equipment and mine facilities. 36 A solidly grounded system is one that has the neutral of the transformer electrically connected to the grounding medium without any intentional impedance. The grounding medium is usually earth or something serving as earth. 37 A resistance grounded system is one that has the neutral of the transformer electrically connected to the grounding medium through a resistor. The purpose of the resistor is to limit the amount of current and voltage during a fault condition. 146 Power centers were located throughout the mine to reduce the voltage for use by the conveyor belt system, water pumps, battery chargers, trickle rock dusters, AMS, trolley communication system, underground workshop, outby work area lighting, the 1st Left section and 2nd Left Parallel section equipment and other miscellaneous equipment. The mine incorporated three splitters, or switchhouses, into its power distribution system. A splitter contains a disconnect switch and a circuit breaker. It can contain more than one set of protective devices. It is used to establish branch circuits that may be de-energized independently of the main circuit. Maps of the electrical system, equipment, and associated items are shown in Appendices Y-1 and Y-2. The high-voltage cable was damaged by the explosion near the mouth of 1st Left and 2nd Left Parallel. This caused the circuit breaker in the single splitter, located at 21 Crosscut, No. 1 Belt to de-energize the high-voltage circuit. Only the circuits outby the splitter remained energized such as the Nos. 1 and 2 Belt drives. The surface power remained energized as well. Grounding Systems The Allegheny Power Company established a safety ground system38 for the French Creek Substation. Two grounded neutral conductors were installed above the power conductors from French Creek to the branch circuit leading to the mine substation. One neutral conductor was continued from the branch circuit to the mine substation. This conductor was installed below the power conductors and connected to the safety ground system for the mine substation and surface electric equipment. A second safety ground system was installed at the mine site. This safety ground system was separated from other safety ground systems by more than 25 feet. Its purpose was to establish a resistance grounding system for the underground power system and equipment. The lightning arresters at this mine were also connected to ground fields. The lightning arresters located within the substation were connected to the surface safety ground system. The lightning arresters at the mine drift opening used for the underground power were connected to a separate ground field. A separate ground field was established for the lightning arresters protecting the AMS. 38 A safety ground system is designed to limit step and touch potentials between grounded components during a fault condition. Part of the safety ground system is the grounding medium (ground field). 147 A lightning arrester is a device that limits the overvoltage of lightning or other electrical surges by providing an electrical path between an ungrounded conductor and earth which is used as the grounding medium. A simple lightning arrester consists of two contacts that are separated by an air gap. One contact is connected to the transmission line and the other is connected to earth. The normal voltage of the circuit cannot bridge the gap. When an overvoltage occurs it sparks over the gap between the contacts. This creates an electrical path for the excess energy to discharge to earth. Abandoned Pump in 2nd Left Mains Sealed Area The mine operator abandoned a submersible pump, its controller and a No. 6 AWG, 2,000 Volt cable with a male cable coupler in the 2nd Left Mains area. The pump and its components are shown in Appendix Y-2, “Electrical Map, 2nd Left Mains, 2 North Mains Inby Crosscut 57.” The label on the pump indicated it was 8.1 horsepower with 10.2 full load amperage and it requires 575 vac, three phase power. The pump, controller, and several hundred feet of cable, located in the No. 6 entry, were under water at the time of the explosion. The majority of the cable was along the No. 5 entry with the coupler near survey station 4028. Portions of the cable were found hung on the ribs and roof near the controller and pump but the majority of it was found on the mine floor. The cable was approximately 1,300 feet long and was found in four sections. Personal Equipment Twelve miner cap lamps were recovered from the barricade on 2nd Left Parallel section and submitted to A&CC for evaluation and testing. The report from A&CC concluded that there were no signs of a short circuit in any of the cap lamp assemblies which would be the source of a spark ignition in a methane-air atmosphere. Further, based on the results of previous testing during the approval process of the Koehler 5000 Series battery, the batteries were incapable of igniting a methane-air atmosphere due to arcing caused by a short circuit of the battery voltage. There were no signs of overheating in any of the cap lamp assemblies which would be the source of a thermal ignition in a methane-air atmosphere. All of the cap lamp bulb envelopes were intact with no exposed filaments. Therefore, no thermal ignition in a methane-air atmosphere could have been initiated by a hot filament. All of the bulbs were labeled with part numbers which were previously accepted and tested. Further, based on the results of previous testing during the approval process, the bulbs were incapable of igniting coal dust on the lens surface or a methane-air atmosphere inside the headpiece. All but one of the cap lamp assemblies illuminated correctly. Exhibit KLH-8 illuminated intermittently. Several discrepancies were identified, but none were considered to be an ignition hazard. The complete report is titled 148 Laboratory Inspection of Twelve Cap Lamps Recovered from a Mine Explosion at Wolf Run Mining Company’s Sago Mine, I.D. No. 46-08791, PAR 92104. Three Motorola non-permissible handheld radios were recovered from the barricade on 2nd Left Parallel section and two Motorola non-permissible handheld radios were recovered from the 1st Left crew. These handheld radios were submitted to A&CC for evaluation and testing. The Motorola PR400 radio is not MSHA approved for use in permissible areas of underground coal mines, but is approved by Factory Mutual (FM) as Intrinsically Safe for use in above ground explosive atmospheres, including methane-air mixtures. MSHA does not accept the FM approval in lieu of an MSHA approval. The functionality of the radios were compared with two new Motorola PR400 radios and functioned as well above ground as the new units did. None of the radios exhibited visual signs that the radio produced a spark or thermal ignition source for the ignition of coal dust or methane-air mixture. Information obtained through the A&CC’s Emergency Communications and Tracking System Committee indicates that radios operating in the UHF band communicate an approximate maximum distance of 1500 feet within the same entry, with severely limited propagation around corners. This is highly dependent on coal seam height, entry geometry, and infrastructure within the entry. See “Executive Summary of Investigation of the Motorola Two-way Radios” in Appendix U. The methane detectors carried by the miners on the 2nd Left Parallel section were recovered and tested. Jesse Jones, Ware, and Winans had CSE Model 102LD portable methane detectors. They were capable of measuring methane concentrations from 0% to 5% and did not have datalogging capability. Two of the three CSE Model 102LD portable methane detectors did not respond to methane within acceptable limits before calibration. After calibration, all three methane detectors responded to methane within acceptable limits. Helms and Martin Toler had Industrial Scientific Model LTX310 portable multigas detectors. They were capable of measuring methane from 0% to 5%, CO from 0 to 999 ppm, and oxygen from 0% to 30% and had datalogging capability. The ISC Model LTX310 that belonged to Helms did not respond to methane and CO within acceptable limits before or after calibration. A bump test of the instrument indicated that the concentrations displayed on the LTX310 for methane was 12% too high while the CO display indicated a 400% higher level. The response to oxygen was not available after it was initially turned on because the oxygen reading went blank and remained blank for the duration of the tests. The memory captures and records the peak methane and CO values and minimum oxygen levels to which the instrument was exposed. It does not 149 indicate when that exposure occurred. The peak values reported before calibration for methane was 1.7% and for CO was 59 ppm. These values appear to be very similar to a calibration gas mixture. These recorded values indicate that the instrument was probably not on at the time of the explosion. The ISC Model LTX310 assigned to Martin Toler did not respond to methane and CO within acceptable limits before calibration. It did respond to oxygen within acceptable limits before calibration. A bump test of the instrument indicated that the concentrations displayed on the LTX310 for methane was 45% too low while the CO display indicated a 200% higher level. After calibration, it responded to all three gases within acceptable limits. The memory captured and recorded the peak methane and CO values and minimum oxygen levels to which the instrument was exposed. It did not indicate when that exposure occurred. The peak values reported before calibration for methane was OR (Over Range), for CO was OR, and for oxygen was 14.6%. These recorded values indicate the instrument probably was on after the explosion and may indicate the concentration of the gases to which the miners may have been exposed, methane and CO greater then 5% and 999 ppm, respectively. The executive summary of the portable gas detector testing is contained in Appendix Z. Potential Ignition Sources An atmosphere containing between 5% and 15% methane and over 12% oxygen can be an explosive mixture. The temperature required to ignite an explosive methane-air mixture is approximately 1,000 degrees F. An explosive mixture is easily ignited by an electrical arc, frictional spark, heated surface or open flame. The amount of energy necessary for ignition will vary with gas concentration, however, as little as 0.3 millijoule of electrical energy is required. This is equivalent to about 1/50 of the static electricity accumulated by an average sized man walking on a carpeted floor on a dry day. The average lightning strike has well over one billion millijoules of energy. Potential ignition sources for the explosion in the sealed area were evaluated, including lightning and roof falls. Other sources, including cutting and welding, mining operations, smoking and spontaneous combustion were considered but were eliminated as potential ignition sources for this explosion. This is discussed below. Other Sources Electric circuits, cables and equipment were examined for evidence that they may have provided the ignition source for the explosion. Physical evidence and testimony indicated that some circuits and equipment were not energized prior to the explosion. 150 There was no evidence that the ignition source originated from the mine’s underground electrical circuits, cables or equipment in the active portion of the mine. This includes the power system, conveyor belt system, water pumps, battery chargers, welders, mantrips, locomotives, rock dusters, AMS, the pager phones, trolleyphone system, radios, gas detectors, cap lamps, electric equipment contained in the underground workshop, outby work area lighting, electric doors and the 1st Left and 2nd Left Parallel section equipment. Several additional ignition sources were considered as potential ignition sources for the explosion. These ignition sources include: the operation of cutting and welding torches, mining operations, smoking, and spontaneous combustion. Each of these ignition sources were initially considered but were eventually dismissed. There were no cutting and welding operations on-going in or near the sealed area at the time of the explosion. Mining operations were not occurring within close proximity to the 2 North Main seals. There was no person near the sealed area at the time of the explosion. Additionally, there were no smoking articles found during the investigation. The mine had no history of spontaneous combustion and there was no evidence of spontaneous combustion found during the investigation. Roof Falls Roof falls can ignite explosive methane-air mixtures either by generating frictional heat or by releasing piezoelectric energy. During a roof fall, rocks forming the strata comprising the immediate and the main roof rub against one another as the roof breaks and falls. In rare cases, the resulting friction from rubbing or from impact can cause temperatures above the ignition temperature of methane. The USBM has conducted rubbing friction and impact friction experiments. Under carefully controlled laboratory experiments, the USBM was only able to ignite methane-air mixtures in a small percentage of tests, even when the methane concentration was optimum for ignition. An ignition can also be generated by piezoelectric discharges during certain roof falls. This type of event is typically associated with rock containing crystalline structures such as tourmaline, quartz, topaz and Rochelle salt. These crystals produce electric charges on parts of their surface when they are compressed in particular directions. In coal mining, the most notable crystal formation found is the quartz content of sandstone. See “Evaluation of Potential for a Roof Fall to Ignite a Methane-Air Mixture” contained in Appendix O. Although a roof fall cannot be definitively excluded as a potential ignition source, it is a highly unlikely source for the following reasons: 151 • • • • Seven roof falls were located within the sealed area. Prior to the explosion, three pre-sealing roof falls had been identified on the mine map. During the investigation, it was observed that these three preexisting falls had extended. Also four additional roof falls were observed that were not shown on the mine map prior to seal completion. See drawing in Appendix O. It is not known exactly when these four roof falls occurred. These four additional roof falls were located approximately 200 to 600 feet from the center of the origin of the explosion. Of these seven falls, the rubble and exposed fall cavity of the five closest roof falls within 440 feet were inspected. Access to the two roof falls beyond 450 feet was obstructed by deep water in the bottom mined areas. Shale is the predominant rock type visible in the roof fall rubble. Specifically, the material referred to as shale was classified as “laminated siltstone” with low quartz content in a soft matrix that inhibits quartz grain-to-grain contact. This low quartz rock type was not as conducive to frictional heating or piezoelectric sparking as sandstones that have been suspected as ignition sources in roof falls. The roof falls extended 7 to 12 feet above the mining horizon. Three roof fall cavities (see Appendix O) had sandstone beds exposed at the top of the fall rubble roughly 8 to 12 feet into the immediate roof above the underlying shale. The samples collected from the roof fall rubble were a variety of sandstone that was micaceous, and characterized by thin, alternating laminations of fine sand, silt, and mica partings. In contrast, the sandstones associated with piezoelectric sparking and rock-on-rock frictional heating are commonly considered to be dominated by quartz, exhibit stronger cementing or even quartz grain fusing (i.e. the metamorphic rock “quartzite”), and occur in more massive beds. Furthermore, the roof falls observed were outside the area where the explosion originated. Thus, rock-on-rock or piezoelectric ignitions are unlikely ignition sources. The only metal roof supports noted in the fall rubble were fully grouted bolts and the wire mesh noted under the rubble of one fall. These steel roof support materials have not been associated with ignitions in experiments or in documented observations of gob ignitions. It was not possible to determine whether cable bolts noted near the roof falls were in the fall rubble. However, previous laboratory testing of the sparks from cable bolt failure did not ignite methane-air explosive mixtures. Since there were no roof falls in the proximity of the origin of the explosion, wicking of methane from the roof falls to the origin was considered. Methane is lighter than air and is released into the mine atmosphere in concentrations generally in excess of 80%. Layering of methane can occur in a mine atmosphere where the velocity of the airflow is minimal and not sufficient to generate turbulence in the airflow. Upon ignition, the layer may burn without the generation of forces and without 152 • generating turbulence in the mine atmosphere, commonly known as wicking. For wicking to occur, a methane layer must be continuous, within its explosive range of 5% to 15%, and would generally be located near the roof. The burning methane layer may eventually contact a larger accumulation, resulting in an explosion. However, a roof fall generates turbulence in the mine atmosphere mixing layers that may have been present. Additionally, the distance, elevation, and uneven roof conditions from the observed falls to the origin of this explosion make this highly unlikely. Computer simulations have predicted that air temperature could increase rapidly to the point of igniting methane or coal dust during a roof fall. The area was sealed, wet, and without air movement, so that any existing coal dust could not have been suspended. The roof falls observed in the 2 North Mains seal area that were not noted on the mine map prior to sealing were too small to ignite methane by compression. Lightning Overview Lightning is an electrostatic discharge (the same kind of electricity that can deliver a shock when touching a doorknob) between a cloud and the ground, between clouds, or within a cloud. Lightning is mostly associated with thunderstorms but is also created during volcanic eruptions, dust storms, forest fires and tornados.39 Nearly 1,800 thunderstorms occur at any moment around the world and lightning strikes the earth 100 times per second.40 Lightning occurs less frequently in the winter because there is not as much instability and moisture in the atmosphere as in the summer.41 West Virginia experiences thunderstorm activity approximately 30-50 days per year.42 Thunderstorms have very turbulent environments. These environments include strong updrafts and downdrafts that occur often and close together. The updrafts carry small liquid water droplets from the lower regions of the storm to heights between 35,000 and 70,000 feet. At the same time, downdrafts are transporting hail and ice from the frozen upper parts of the storm. When these particles collide, the water droplets freeze and release heat. This heat keeps the 39 www.nssl.noaa.gov/primer/lightning/ltg_basics.html. 40 www.moncooem.org/thunderstorms.htm. 41 www.nssl.noaa.gov/primer/lightning/ltg_faq.shtml. 42 www.moncooem.org/thunderstorms.htm. 153 surface of the hail and ice slightly warmer than its surrounding environment, and a soft hail, or graupel forms. When graupel collides with additional water droplets and ice particles, a key process occurs involving electrical charge. Negatively charged electrons shear off of the rising particles and collect on the falling particles. The result is a storm cloud that is negatively charged at its base, and positively charged at the top. Opposite charges attract one another. As the positive and negative areas grow more distinct within the cloud, an electric field is created between the oppositely charged thunderstorm base and its top. The farther apart these regions are, the stronger the field and the stronger the attraction between the charges. The atmosphere is a very good insulator that inhibits electric flow. A huge amount of charge has to build up before the strength of the electric field overpowers the atmosphere's insulating properties. A current of electricity forces a path through the air until it encounters something that makes a good connection. The current is discharged as a strike of lightning. While all this is happening inside the storm, a positive charge begins to pool on the surface of the earth beneath the storm. This positive charge will shadow the storm wherever it goes, and is responsible for cloud to ground lightning.43 Most of these flashes originate near the lower-negative charge center of the storm and deliver a negatively charged lightning strike to Earth.44 However, the electric field45 within the storm is much stronger than the one between the storm base and the earth’s surface, so about 75 to 80% of lightning occurs within the storm cloud.46 The voltage of lightning discharges can range from 100 million to one billion volts.47 43 www.nssl.noaa.gov/primer/lightning/ltg_basics.html. 44 thunder.msfc.nasa.gov/primer/primer2.html. 45 An electric field is a field or force that exists in the space between two different potentials, such as between negatively and positively charged regions of a thunderstorm. 46 www.srh.noaa.gov/mlb/ltgcenter/whatis.html. 47 www.nssl.noaa.gov/primer/lightning/ltg_faq.shtml. 154 Cloud to ground lightning is defined as lightning that discharges to earth. This is shown in Figure 36.48 These are negative discharges most of the time. Positive discharges account for less than 10% of all cloud to ground strikes, and most often occur on the periphery of a thunderstorm. The peak current of a positive discharge is often much larger than a negative one, resulting in greater potential for damage.49 Figure 36 - Cloud to Ground Lightning Intra-cloud lightning occurs within separate charge centers of a cloud. This illuminates portions of the cloud without any visual evidence of the lightning strike that is occurring within the cloud. 50 This is shown in Figure 37.51 The lightning discharge may be positive or negative depending on the charge center. Figure 37 - Intra-cloud Lightning Sometimes an intra-cloud discharge occurs between charge centers of different clouds. This results in a cloud to cloud discharge. 52 This is shown in Figure 38.53 Figure 38 - Cloud to Cloud Lightning 48 thunder.msfc.nasa.gov/primer/primer2.html. 49 www.srh.noaa.gov/mlb/ltgcenter/whatis.html. 50 thunder.msfc.nasa.gov/primer/primer2.html. 51 www.nssl.noaa.gov/primer/lightning/ltg_basics.html. 52 thunder.msfc.nasa.gov/primer/primer2.html. 53 www.nssl.noaa.gov/primer/lightning/ltg_basics.html. 155 Also, upward lightning has been known to occur. It is a discharge from a tall structure to a cloud. It develops from the pool of positive charge shadowing the storm.54 This is shown in Figure 39.55 Figure 39 - Upward Lightning In the past, lightning has been identified as a possible ignition source for explosions in sealed areas of underground coal mines. Prior to the Sago accident, MSHA had not conducted full underground investigations of post-explosion sealed areas, because hazardous conditions did not permit full exploration or investigation of these areas. Table 10 is a list of some of these occurrences. Table 10 - Mine Explosions in Sealed Areas with Lightning as a Possible Ignition Source Mine Name JWR No. 3 Mary Lee No. 1 Oak Grove No. 1 Beatrice Gary 50 Oak Grove Oasis No. 1 Oasis No. 1 Oak Grove Soldier Canyon Pinnacle Big Ridge Year Explosion Occurred 1986 1993 1994 1994 1995 1996 May, 1996 June, 1996 1997 2000 2001 2002 Metal Conduit Present Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Number of Seals Destroyed Shaft cap 2 plus shaft cap 5 Shaft cap None 1 4 4 3 Shaft cap Shaft cap 1 54 Upward Lightning Flashes, Wada, A., Miki, M., Asakawa, A., Central Research Institute of Electric Power Industry, Nagasaka, Japan (2004) 55www.rf-web.tamu.edu/about/ 156 Lightning detection networks track lightning throughout the United States using sensors located at various locations. MSHA obtained reports from two lightning detection companies regarding lightning strikes near the mine on the morning of the explosion. See “Vaisala Group and AWS Convergence Technologies, Inc. Reports” contained in Appendix AA. The Vaisala Group tracked the lightning through the National Lightning Detection Network (NLDN). AWS Convergence Technologies, Inc. used the United States Precision Lightning Network (USPLN). USPLN is owned and operated by TOA Systems and Weather Decision Technologies. As reported by the lightning detection companies, both of these lightning detection systems have limitations. The NLDN has an accuracy of 1,640 feet on average while the USPLN has an accuracy of 820 feet. Both systems require that at least three sensors detect the discharge before it is recorded. If less than three sensors detect a discharge, it will not be recorded as a strike. Also, upward lightning initiated by tall structures cannot be detected by these systems. The USPLN has a detection probability of 95% for cloud to ground lightning and 60% for intra-cloud lightning in the West Virginia region. The NLDN does not record cloud to cloud or intra-cloud discharges. It also has a detection probability between 80-90 percent. Therefore, unrecorded lightning discharges can occur during a storm along with the recorded discharges. The NLDN recorded two lightning strikes near the mine area at the time of the accident. The first strike occurred at 6:26:35.522 a.m. and reportedly occurred more than one mile south of the mine drift openings. This was a positively charged strike with a magnitude of 38,800 amps. Several unsuccessful attempts were made to locate evidence of a strike in this area. The other lightning strike occurred at 6:26:35.680 a.m. and was about one mile north of the mine drift openings. It was also a positive lightning strike with a magnitude exceeding 100,000 amps. Evidence of this strike hitting a tree was found. The tree had freshly splintered pieces scattered around it. This is shown in Figure 40. Figure 40 - Damaged Tree 157 USPLN recorded one lightning strike near the mine area at the time of the accident. The strike occurred at 6:26:35.522 a.m. and reportedly was about a half of a mile south of the drift openings. This was a positively charged strike with a magnitude of 35,000 amps. Several unsuccessful attempts were made to locate evidence of a strike in this area. A map contained in Appendix BB titled “Sago Mine in relation to recorded location of lightning strikes, a lightning - damaged poplar tree and the mine’s phone and power lines” shows the three recorded lightning strikes. Lightning as an Ignition Source The Virginia Polytechnic Institute and State University’s Department of Geosciences concluded that a seismic event most likely occurred at or near the Sago Mine within a four-second interval centered at 06:26:38 a.m. on January 2, 2006. A copy of that report titled “Results from Analysis of Seismic Data…”is contained in Appendix CC. In addition, the atmospheric monitoring system recorded the first presence of CO at 06:26:35 a.m. The nearby lightning strikes recorded by NLDN and USPLN occurred at approximately the same time as the seismic event and the initial alarm for the AMS. To determine if lightning energy may have entered the mine, MSHA contracted with Sandia Corporation, Sandia National Laboratories (Sandia). They performed modeling and testing to simulate whether lightning energy could enter the mine by direct contact or indirect inductive coupling. Sandia has unique capabilities to characterize and mitigate lightning effects on high value assets with the Department of Energy and other agencies as part of a national security mission in nuclear weapons stockpile stewardship. From November 5 through November 9, 2006, personnel from Sandia conducted direct and indirect tests at the mine site. They compared the energy levels recorded from these tests with the levels required to initiate an arc. Sandia also analyzed the raw data provided by two lightning detection databases for other lightning discharges that failed to meet detection standards. They failed to find evidence of another cloud to ground strike in the correct timeframe. The Sandia report concluded “that lightning-induced electrical arcing was not only plausible, but highly likely.” See report titled “Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine” contained in Appendix DD. 158 Based on this information, MSHA concluded that lightning is the most likely ignition source for this explosion. Several plausible lightning strike scenarios illustrate how significant energy could ignite methane in the sealed area of the Sago Mine. These were evaluated to determine the most likely possibility. Three scenarios for energy from lightning to enter the sealed area were evaluated and are listed below as A, B and C. A. A recorded strike occurred in the proximity of the mine, hitting a tree. Two apparent paths for energy from this recorded lightning strike to reach the portal are through 1) the telephone grounding system or 2) the high-voltage power system. Further evaluations were undertaken to determine if the energy from lightning could be transported from the portal to the sealed area. B. A lightning strike delivered from the surface area directly through a conductor over the sealed area, such as gas wells and their interconnected piping system or water in the strata overlying the sealed area. C. A lightning strike over the sealed area indirectly energizing metallic objects within the sealed area. Scenario A - A recorded strike occurred in the proximity of the mine, hitting a tree. Two apparent paths for energy from this recorded lightning strike to reach the portal are through 1) the telephone grounding system or 2) the high-voltage power system. Further evaluations were undertaken to determine if the energy from lightning could be transported from the portal to the sealed area. 1) Surface Telephone Grounding Conductor A resident living near the damaged tree stated that his telephone service was interrupted for two days after the lightning strike. An investigation of this area revealed that the strike hit the tree and left a hole in the ground at the base of the tree. Investigators analyzed the area around the tree and found that an underground telephone communication cable was located approximately 40 feet from the tree. A telephone junction box was located approximately 100 feet from the tree. The communication cable was routed through junction boxes between the tree that was struck by lightning and the mine. Each junction box was connected to the earth by a ground electrode and was connected to the metallic shielding on the communication cable. The location of the tree struck by lightning along with the identified locations of each telephone pole and telephone junction box leading to the mine are shown in Appendix BB. 159 Earth resistance measurements were conducted at the tree near the mine, which was struck by lightning, and at a Verizon telephone junction box located approximately 40 feet from the tree. These measurements were taken to determine the soil resistivity. The earth resistance test from the tree to the Verizon junction box revealed an earth resistance value of 4.92 ohms when taken by the three pole method with the Lem Unilap NGI tester. Four additional earth resistance measurements were conducted at right angles from the tree with the Lem Unilap NGI tester. These four tests revealed an average low earth resistance of 4.82 ohms within a 60 feet diameter of the tree. Resistance measurements were taken with a Fluke MegOhmMeter from the Verizon junction box to a power pole located at the supply trailer located on mine property. All grounds were connected common and to earth at the base of that pole. These measurements revealed the total resistance of the ground circuit for the telephone system from the junction box to the power pole was 204.8 ohms. The telephone system and earth provided a low resistance path that extended from the junction box located near the tree struck by lightning to the mine. This path was also connected to the surface telephone system ground, the safety ground system for the surface equipment and the safety ground system for the underground mine electric power system. All of the installed grounding conductors, with the exception of the bathhouse and foremen’s offices, were connected to the surface metal belt structure, mine track system and trolleyphone system. 2) High-Voltage Mine Power Electric System A lightning strike occurred in the proximity of high-voltage transmission lines near the mine, hitting the tree. The lines which extended from the French Creek Substation to the preparation plant and to the mine were examined. The purpose of this examination was to determine if lightning may have struck the main transmission or branch lines and entered the mine on one or more of the branch lines. The examination of the transmission line from the French Creek Substation and branch line that extends to the mine revealed damage to a phase insulator and a lightning arrester. The phase insulator was damaged on the main 160 transmission line. The lightning arrester was damaged at the second pole of the branch lines leading to the mine. A map showing the high-voltage transmission and branch lines is shown in Appendix BB. Figures 41 and 42 are photographs of the damaged insulator and lightning arrester below. Figure 41 - Damaged Insulator Figure 42 - Damaged Lightning Arrester It was not possible to determine if the lightning storm that occurred on the day of the accident caused the damage to the lightning arrester and insulator. A previous storm or other event may have caused this damage. 161 Electric circuits and equipment were examined and tested on the surface and underground to determine if lightning entered the mine through one or more of the high-voltage power conductors. These examinations and tests revealed that none of the lightning arresters on mine property or the surge suppressors installed in the mine high-voltage power centers were damaged by lightning. There was no flash-over arcing or tracking identified that could be caused by lightning in the electric equipment installed on the surface at the mine or underground. MSHA took electrical earth resistance measurements of the safety ground systems for the mine substation and the underground power system and equipment. These measurements were taken to determine if the grounding systems were of a low resistance value. The measurements were taken with a Lem Unilap NGI ground resistance tester. This device used the three-pole method to check resistance with the poles spaced at 20 foot intervals. The test of the safety ground system for the surface mine substation revealed that the system had a resistance value of 9.38 ohms. The test of the safety ground system for the underground power system and equipment revealed the system had a resistance value of 7.4 ohms. These measurements showed that both grounding systems had low resistance values. These systems provided an adequate means for dissipating electrical surges from system operation or other sources such as lightning. Because a lightning strike occurred in proximity to the high-voltage transmission lines, an induced voltage could have occurred. When voltage is induced in transmission lines, it can affect the entire system. Induced voltages would have the potential to damage the transformers at either the French Creek or Sago substations due to direct current applied on an iron core transformer. If this had occurred, the French Creek substation would have detected a significant voltage increase and would have de-energized the entire system. Lightning arresters at the French Creek substation and at the mine should have dissipated this voltage. There was no interruption of electrical power at the mine, so any induced voltage on the high-voltage transmission lines was not significant. However, the two grounded neutral lines above the transmission lines could have transmitted an induced voltage. These grounded neutral lines were connected to the surface safety ground. Therefore, an induced voltage could have been produced on the grounded neutral lines and transmitted to the surface safety ground which was also connected to the surface telephone system ground and the safety ground system for the underground mine electric power system. All of the installed grounding conductors, with the exception of the bathhouse and foremen’s offices, were connected to the surface metal belt structure, mine track system and trolleyphone system. 162 Lightning Energy from the Portals to the Sealed Area The interconnection of the surface telephone system grounding conductor, surface lightning arrester grounding conductors, the continuous metal structure of the belt line, the mine track, and the surface and underground mine power system grounding conductors were common. Since they were common, they created a low resistance path for the energy from a lightning strike to possibly enter the mine, and extend to the No. 6 belt drive, which is approximately 400 feet from the 2 North Mains seals. The grounding conductors for the belt starters and motors on all conveyor belts were connected to the frames of the starters, the belt drive motors, and the metal frames of each conveyor belt. The No. 6 belt drive for the 2nd Left Parallel was at 58 Crosscut, No. 4 Belt along with the No. 4 belt tailpiece. The conveyor belts were suspended from the mine roof by metal chains attached to brackets bolted to the roof. The belt drive and belt tailpiece were provided with metal guarding materials to minimize the likelihood of persons contacting moving parts. The metal guarding and the metal supports for the guarding were anchored firmly against the mine roof and in contact with the wire mesh installed on the roof. The area where the ten seals were located prior to the accident had metal wire mesh installed against the roof in the Nos. 5, 6 and 7 entries of 2 North Mains, and in the Nos. 4, 5 and 7 entries in the 2nd Left Mains. The wire mesh assisted in controlling the effects of roof spalling and was installed rib to rib, overlapped with roof bolts and spider plates to hold it against the mine roof. The metallic bolts and spider plates attached the wire mesh against the roof firmly, and provided a good electrical connection throughout the wire mesh. The wire mesh was removed from the area of the roof immediately above the location of the seals. The mesh in the No. 5 entry (seal No. 6), the No. 6 entry (seal No. 7) and the No. 7 entry (seal No. 8) had gaps 10, 11 and 4 feet wide, respectively. There were gaps in the wire mesh in several other locations as well. Appendix Y-2, “Electrical Map, 2nd Left Mains, 2 North Mains Inby Crosscut 57” shows the wire mesh installed outby and inby the sealed area. A 2 inch diameter, 40 foot long galvanized steel pipe was installed to provide a conduit for an air-sampling pipe through seal No. 10 in the No. 9 entry. The pipe extended into the sealed area and was supported on cribs. About three feet of the metal pipe extended from the seal on the active side, which was reduced to ½-inch diameter. A ½-inch diameter copper pipe was installed inside the steel pipe with a ball valve connected to the end of the copper pipe. The sampling pipe was located on the left side of the seal approximately 12 inches from the roof. Wire mesh was not installed in the No. 9 entry. This sampling pipe was not connected to a low resistance path. Therefore, the sampling pipe was not a likely path for lightning or any energy to reach the origin of the explosion. 163 Thirty-one earth resistance measurements of the mine roof and floor were taken. Measurements were taken from the end of the 2 North Mains track (survey station 3923) in the No. 6 entry to the origin of the explosion. The distance from the mine track to the origin is approximately 1,100 feet. A Lem Unilap NGI tester using the four-pole method was used. These earth resistance measurement values as well as the location of each previously installed seal, and the areas of the mine roof and floor where tests were conducted, are shown on the map titled “Earth Resistance Measurement Values” in Appendix KK. Measurements between these points revealed a low resistance path, so energy could flow easily between them. Electrical earth resistance measurements were taken underground using the Lem Unilap NGI tester with the four-pole method. The results of the measurements are shown in Appendix KK. One series of tests was conducted at each location of the ten 2 North Mains Seals. Two other series of tests involved acquiring data in the belt and track entries to inby the 2 North Mains seals location. These measurements were taken to determine the resistivity of the path between these points. Three measurements were taken at each location. The four poles were attached to roof bolt plates during one test. Measurements were taken from metal pins that were installed in the mine roof. For the third test, the poles were installed in the mine floor. Measurements between these points revealed a low resistance path. Measurements were also taken from the No. 6 Belt starter to various sites near survey station 4010 and No. 1, No. 2 and No. 9 seal areas. These measurements were taken to determine the resistivity between the points. A Megger resistance tester and a Beckman HD 110 multi-meter were used in conjunction with 2-12 AWG wires connected in parallel to obtain resistance readings. Measurements between these points revealed a low resistance path. The results of the resistivity measurements, including the locations of the measurements and the values, are shown in Appendix KK. The underground mine power system grounding conductors, belt support structure, metallic guards, roof support and wire mesh were connected together. Measurements indicate that even with gaps in the wire mesh, a low resistance path continued from the No. 6 Belt drive into the sealed area. Sandia conducted a direct drive test to determine if lightning energy could enter the mine through the low resistance path. They applied a test signal at the portal to the belt conveyor structure, trolley communication antenna, high-voltage cable grounding medium, and the track rail. They monitored the signal with current and voltage probes at several locations in the mine including where the mine belt structure, trolley communication line, and the track rail were closest to the 2 North Mains seals. 164 Resistivity measurements indicated that electrical energy could travel from the surface of the mine to the point of origin in the sealed area. However, Sandia concluded the energy is divided sufficiently by earth grounding so that only a relatively small amount of energy is directed into the mine near the sealed area. It is unlikely this method could provide an adequate amount of energy at the point of origin. The ground wires in the high-voltage cable exhibited a relatively high current at the closest point to the 2nd Mains Seals. The high-voltage cable does not end at the 2nd Left Parallel switch, but continues to the 2nd Left Parallel section power center. Sandia analyzed the possibility that lightning energy on this portion of the high-voltage cable may have induced energy onto the abandoned pump cable in the sealed area. Sandia concluded that any energy induced on the pump cable would be too low to ignite methane. Energy induced onto the pump cable is discussed further in Scenario C. Sandia further concluded that it is highly unlikely a 100,000 amperes lightning strike attached at the mine portal to the belt conveyor structure, trolley communication antenna, high-voltage cable grounding medium, and the track rail could generate sufficient voltage on the pump cable within the sealed area to initiate electrical arcing. Therefore, it is not likely that methods discussed in this scenario could ignite methane in the sealed area. Scenario B – A lightning strike delivered from the surface area directly through a conductor over the sealed area, such as gas wells and their interconnected piping system or water in the strata overlying the sealed area. Direct Strike Over the Sealed Area Both lightning detection systems have limitations and do not record all lightning strikes. An unrecorded cloud to ground or an upward discharge may have occurred over the sealed area. Upward lightning may have been initiated from a nearby communications tower. Four towers are within about one mile of the sealed area, the closest one being about a half mile away. The lightning energy could have been delivered from the surface area directly through a conductor over the sealed area, such as gas wells and their interconnected piping system or water in the strata overlying the sealed area. 165 Gas Wells and Interconnected Metal Piping System Several gas wells and interconnecting metallic pipelines were installed in the vicinity of the mine. One gas well was located about a half mile northwest of the tree that was struck by lightning. A pipeline connected this well with other gas wells in the area, and extended approximately 2.4 miles to an active gas well (API# 47-097-01251) located near the previously mined sealed area in 2 North Mains. The well casing was not located in the sealed area itself. The location at the surface of the well indicates it may have been as close as 107 feet from the mine workings. The well extended from the surface and penetrated the coal seam at approximately 377 feet, extending approximately 4,755 feet deep to the natural gas reservoir. The well was encased with 10 feet of 13 inch diameter metal casing, 713 feet of 8 ¼ inch metal casing, and 4,032 feet of 4 ½ inch metal casing. The metallic gas pipeline over the sealed portion of the mine where the explosion occurred was not tested due to liability concerns of all participants. If the pipeline is viewed as a conduit of energy and if it was installed as a continuous metallic structure, then it could conduct lightning energy. The cell towers that are in close proximity to the pipeline could have experienced a lightning strike or an upward-going positive strike that was not recorded by the lightning detection networks. The cell towers are well grounded to earth and therefore would conduct the lightning energy into earth and on to the pipeline. This could generate two results. Result 1: The pipeline is not in direct contact with the surrounding earth and exhibits a high resistance contact with earth. In this case the pipeline would be looked upon as an insulated conductor that would conduct lightning energy. The lightning energy would be conducted to the gas well that the pipeline was connected to and then the lightning energy would be dissipated into the earth through the gas well casing. While the lightning current is flowing through the pipeline it will generate a similar electromagnetic pulse that a lightning strike would generate. This could then induce a voltage in the cable that was in the sealed area that could result in an electrical arc at the cable. Result 2: The pipeline is in direct contact with the earth which is of low resistivity and therefore well connected to earth. Resistivity tests above the sealed area show that the immediate earth exhibits low resistance. In this case the lightning charge would dissipate very quickly into the surrounding earth as was shown by Sandia in their direct drive tests conducted on the track that entered the mine. The track, like the pipeline, was in direct contact with earth. Therefore, it is unlikely sufficient current to ignite methane would be able to enter the sealed area of the mine. 166 The first result does provide a possible explanation as to how lightning energy could have traveled on the surface to induce energy and a resultant arc in the cable left in the sealed area of the mine. Neither of these results would have by themselves provided a direct drive but could have enhanced the indirect drive to initiate an arc in the sealed area of the mine. The investigators evaluated the possibility that energy from a direct lightning strike penetrated into the sealed area through the metal gas well casing and provided the energy to ignite methane. No visible evidence of the lightning strike to the metal well heads or gas lines was observed. Additionally, no damage to the ground was observed in the vicinity of the wells or along the length of buried metal lines. The only damage observed in the area was to a tree located near a buried metal gas line on a hilltop approximately 4,600 feet north of a recorded strike. When the tree was damaged is unknown, but neither of the lightning detection networks recorded a strike at this location. There was no evidence underground of the explosion originating in the entries nearest the gas well. Although wicking of a methane layer is possible, the distance, elevation, and uneven roof conditions from these entries to the origin make this unlikely. See Appendix GG for a map titled “Sago Mine in Relation to Recorded Locations of Lightning Strikes, Gas Wells and Gas Lines” and Appendix EE for a report titled “Investigation of the Well Heads and Gas Pipeline System.” A measurement was taken to determine the earth resistance of Gas Well API# 47097-01251, on the surface near the sealed area. A Lem Unilap NGI tester and the four-pole method were used. The test revealed the surface of the earth around the gas well had a resistance value of 2.49 ohms. Measurements between these points revealed a low resistance path. This indicates that lightning energy may readily dissipate in the earth near the well rather than travel into the mine. Conductors to the Sealed Area HydroGeophysics, Inc. conducted two other tests during the weeks of June 12 and July 17, 2006. The company conducted a geophysical survey to map the subsurface electrical properties of the region above the sealed area. This survey relied on induced fields to map all magnetic and electrical properties of this area and any metallic features such as well casings that might be present but unknown. The company also performed a surface to mine resistivity test, which mapped zones of gradient electrical resistance to establish the conductivity of the earth above the sealed area. These tests indicated that a direct, vertical low resistance metallic path or zone of reduced resistivity for lightning energy to travel from the surface to the sealed area did not exist. See Appendix FF an executive summary report titled “Geophysical Survey of the Old 2 Left Section of the Sago Mine.” 167 Water in the Strata Over the Sealed Area Water samples were collected from surface streams, the right rib of the No. 8 entry on 2nd Left Mains and from the track entry between 32– 33 Crosscut, No. 3 Belt of the 2 North Mains below Trubie Run stream to assess the pH and electrical conductivity of water in the mine. This study was done to determine if electrical energy was capable of passing from the surface into the sealed area through water. The sample obtained from the No. 8 entry on 2nd Left Mains exhibited a high conductive property. A sample was not collected at or near the origination of the explosion because there was no water present. Based on the area of origination of the explosion in relation to the samples collected, it is highly unlikely that water entering the mine from the surface created a path for electrical energy to enter the sealed area and ignite an explosive mixture of methane gas. See Appendix HH, a report titled “Observation and Sampling Collection Methodology.” Based on observations and testing, Scenario B is unlikely. Sandia stated that it is unlikely that the vertical pipes would induce a significant amount of voltage onto the pump cable in the sealed area because the cable is perpendicular to the pipes.56 Similar to the wire mesh, the gas lines and the well are grounded at regular intervals and would not support a large voltage potential. There was no evidence underground of the explosion originating in the entries nearest the gas well. Wicking of a methane layer from the area closest to the gas well is unlikely based on the distance, elevation, and uneven roof conditions from these entries to the origin of the explosion. HydroGeophysics, Inc. did not find a low resistance vertical path for lightning to travel into the mine. Also, conductive water was eliminated based on the origin of the explosion. Scenario C - A lightning strike over the sealed area indirectly energizing metallic objects within the sealed area. Indirect Energy Transfer To Sealed Area The current in a lightning strike has an associated magnetic field. Due to the relatively low frequency content of lightning (<100 kHz), electromagnetic energy can readily propagate through hundreds of feet of earth and induce a voltage onto an antenna or receiver. This process is referred to as indirect coupling. An electromagnetic field propagates through the earth as a result of a cloud to ground lightning strike or a long, low-altitude horizontal A conductor that is perpendicular to another is severely limited in its ability to induce a voltage on the other, as compared with a conductor that is parallel to another. 56 168 current channel from a cloud to ground strike. Unlike direct coupling, indirect coupling does not require the presence of metallic conductors in a continuous path from the surface to areas inside the mine. This scenario involves lightning occurring over the sealed area. There are several ways in which this could have occurred. The horizontal portion of a recorded lightning strike may have traveled over the sealed area. As discussed previously, both lightning detection systems have limitations and do not record all lightning strikes. An unrecorded strike may have occurred over the sealed area. This strike could have been a cloud to cloud, intra-cloud, cloud to ground or an upward discharge undetected by either system. A lightning strike in this area would induce a voltage on all nearby metallic objects, on the surface and underground. Sandia conducted indirect drive tests. A test signal was generated on the surface over the sealed area where the explosion initiated and then was measured underground with instrumentation, a computer and an antenna. The objective was to identify the mechanism that would allow electromagnetic coupling of lightning energy into the sealed area of the mine. The soil and rock resistivities play a major role in determining the amplitude and frequency dependency of indirect coupling. The electric fields measured in the sealed area of the mine had amplitude and frequency characteristics which confirmed that they were caused by diffusion coupling from currents above the sealed area through the soil and rock overburden. The soil and rock resistivities used to model the coupling were comparable to those determined by HydroGeophysics, Inc. The abandoned submersible pump in the 2nd Left Mains had a cable approximately 1,300 feet long which was found in four sections. It appeared to have been damaged by being pulled apart, rather than being severed. Mine management indicated that the cable was pulled into one or more sections as they tried to retrieve the pump prior to sealing the area. It could not be determined if some or all of the damage was caused by the explosion. Also, there was no evidence of arcing or sparking on any of these cable ends. See “Executive Summary of Submersible Pump Parts Recovered from Sago Mine” contained in Appendix II. Approximately 196 feet of the pump cable abandoned in the sealed area was retrieved for testing by MSHA. The retrieved portion of the pump cable extended from cable break 1 to where the cable coupler was removed. That is the portion of cable nearest the origin of the explosion. The location of the cable is shown in Appendix Y-2, “Electrical Map, 2nd Left Mains, 2 North Mains Inby Crosscut 57.” This portion of the cable had three permanent splices, one temporary splice and numerous damaged places in the outer jacket of the cable. 169 Testing of the three insulated power conductors (red, black, white), two ground conductors, and an insulated ground check conductor within the pump cable was performed. The test was conducted to determine if there had been a failure in the conductor’s insulation and the cable’s outer jacket. The ground wires in this cable are not provided with insulation. The insulation on the power conductors is rated for 2,000 volts. The red conductor failed the test at about 700 volts. The black conductor failed when 1,600 volts was applied to it. The white conductor passed the test. Further testing on the white conductor was conducted to determine at what level failure would occur. It failed when 5,500 volts was applied. Each of the two ground conductors failed when 24 volts were applied to them. The ground check conductor passed the test with minimal leakage current. See Appendix JJ titled, “Sago Mine Pump Cable Test.” The Sandia test data revealed that during a lightning strike the insulated conductors of the abandoned pump cable could receive voltages as high as 20,500 volts. This voltage would be of a short duration but the energy generated would be adequate to cause an arc and ignite methane. A cable with insulated conductors in an underground mine can act as an antenna or receiver. If a lightning strike occurs on the surface there could be a voltage induced onto the insulated conductors of the underground cable which may result in component failure. The component failure will be in the form of an insulation breakdown or arcing. If the conductors in the cable were frayed, they would be of such a small size that they could not carry the induced energy upon them by the lightning strike. The frayed portion of the conductors would act like fuses and burn apart causing an arc. When a cable is connected to a coupler, the four insulated conductors are in close proximity to grounded conductors and the grounded shell of the coupler. A cable coupler contains exposed bare copper pins that connect to the red, white, black, ground and ground check conductor. Figure 43 is a photograph of the cable coupler. Figure 43 - Cable Coupler 170 Figure 44 is a photograph of the end of the cable coupler containing the connecting pins and the conductors. These pins become energized to the same level as their connected conductors. The coupler was lying on the damp mine floor, the coupler pins were exposed to moist dirt that could provide a path to the grounded metal shell resulting in an arc. Figure 44 - Coupler with Pins and Conductors Damaged areas of a cable or its coupler are potential locations where arcs can occur. This cable had numerous damaged areas. Figure 45 is a photograph of two pieces of the cable. Therefore any cuts, nicks, pinholes, or other damage to the cable are potential points where an arc between the red, black, or white conductor and the grounding conductors could occur. Figure 45 - Two Pieces of the Cable Any of these arcs are potential ignition sources for the methane. Although there was no observed sign of arcing on the conductors, this does not rule out the possibility that an arc occurred, initiating the explosion. The pump cable was not connected to a power source, and power sources were not located in the sealed area. No other equipment was found in the sealed area. Other metallic objects near the origin of the explosion include roof bolts, spider plates and wire mesh. These objects were not considered as plausible receivers or antennas of the electromagnetic energy that propagated underground because measurements indicated they were well grounded at regular intervals to the roof of the sealed area, and therefore would not support a large voltage potential. Corona discharge can occur to any power conductor that is energized. When energized, it produces an electric field around the power conductor, unless the power conductor is shielded with a grounding shield. The strength of the electric field developed around the conductor is proportional to the conductor wire size, the shape of the conductor and the amount of voltage applied to the conductor. When the electric field strength reaches a specific value, the air molecules surrounding the conductor become ionized. Higher voltage levels produce a cloud of ionized gas surrounding the conductor. This process is called corona discharge and is a precursor to an arc. A corona discharge may ignite an 171 explosive gas mixture.57 Sandia concluded it is unlikely that a corona discharge would develop before an electrical arc occurs due to the short duration of lightning. Sandia’s field measurements and analysis indicate that significant electromagnetic energy can be coupled into the sealed area of the mine. A lightning source, as stated above, would create an electromagnetic field similar to a magnetic field that is produced between the north and south poles of a magnet. The electromagnetic energy created by the lightning discharge would have then radiated through earth onto the pump cable, which could act as a receiver or antenna. The electromagnetic energy could induce a voltage onto the pump cable which generates an arc near the explosive methane mixture in the sealed area. Eyewitness accounts of simultaneous lightning and thunder above the sealed area at the time of the explosion lend further credence to this possibility. Measurements and analyses indicate that the pump cable is the most likely receiver of electromagnetic energy in the sealed area. This is the most likely ignition source for this explosion. Origin The origin of an underground coal mine explosion is the location where the explosion begins. It is identified as the location from where primary explosion forces propagate in all directions. Primary explosion forces are the initial forces that occur at each location. In addition, the origin must be a location that includes a suspended accumulation of fuel, sufficient oxygen to support combustion of the fuel, and the ignition source. In some cases, the ignition source can occur a short distance from the origin of an explosion. In these cases, the ignition source is located within an explosive concentration of layered fuel. Methane is a gas which normally forms layers in underground coal mines under certain conditions. The methane layer must be continuous, within its explosive range of 5% to 15%, and would generally be located near the roof. Upon ignition, the layer burns without the generation of forces and without generating turbulence in the mine atmosphere. This is commonly called “wicking.” The burning layer eventually contacts a larger accumulation, resulting in an explosion. The explosion immediately generates primary forces propagating in all directions away from the origin. 57 Sacks, H. K., and Novak, Thomas, Corona Discharge Initiated Mine Explosion, IEEE Transactions on Industrial Applications, Vol. 41, Sept/Oct 2005. 172 Layering can occur in a mine atmosphere where the velocity of the airflow is not sufficient to generate turbulence in the airflow. The lack of turbulence prevents the mixing of gases throughout the mine opening. Sufficient ventilation can disperse a layer of mine gases. On December 28, 2005, an examination was conducted at the 2 North Main seals and 0.2% methane was detected. The quantity of air measured in the split of air ventilating the seals was 4,392 cfm. The examiner detected 1.2% methane exiting the sample pipe in Seal No. 10. On December 30, 2005, the mine foreman visited the area and also found 0.2% methane in the split of air ventilating the seals. It is unlikely that wicking through the seal could occur due to the fully-mortared face on the active side of each seal. The gas sampling line in Seal No. 10 included a valve. The valve was reported to be closed and the line was not installed against the roof. The water trap in Seal No. 1 was reported to be full of water and was installed near the floor. There were no ignition sources near the seals. Although the burning of a layer leading to an explosion can be possible, no evidence was found to support wicking from outby the sealed area through the seals. The distance through which a burning layer can pass is dependent on the conditions of the roof, such as undulations, and the ability of the layer of methane to remain within its explosive range. Within the sealed area, there were locations where accumulations of methane were possible. If the burning layer contacted any accumulation, an explosion would result. It is unlikely that a layer in the sealed area would have the ability to burn for more than a short distance. Physical evidence was observed throughout the underground areas affected by the explosion. Physical evidence includes the deformation of structural materials, including belt hangers, roof support plates, and wire mesh. Also, the deposition of mine dust on mine surfaces and on equipment and roof bolts is considered to establish the direction of primary forces. This evidence was evaluated and was used to establish the point of origin, the extent of flame, and the direction of the primary explosion forces. Evidence indicated that the explosion was initiated within the sealed area near survey stations 4010 and 4011 in the 2nd Left Mains. Primary forces propagated away from this location in all directions, thus identifying this location as the origin of the explosion. A typical underground coal mine explosion begins as methane is ignited. The ensuing fireball rapidly enlarges and eventually begins to propagate through the fuel. When propagation through the unburned fuel remains at a velocity below the speed of sound, the explosion is termed a deflagration. The forces developed during a deflagration are dependent on the speed of the flame. The faster the flame propagates, the higher the forces become. Deflagrations show limited, if any, pressure damage near the origin due to the fact that the flame is developing. 173 Figure 46 was taken near survey station 4010, facing outby in an area that had been second mined. All but two of the roof plates appear to be unaffected by the forces of the explosion, indicating that forces had not reached their peak magnitude. Figure 46 - Picture taken Near Survey Station 4010 The picture shown in Figure 47 was taken from a location near survey station 4011. There is no damage to roof support members, including wire mesh or roof plates. The origin of this deflagration-type explosion was located within the sealed area near survey stations 4010 and 4011 in the 2nd Left Mains. This particular location did not appear to be exposed to the same magnitude of pressures as the surrounding areas. Figure 47 - Picture Taken Near Survey Station 4011 174 Figure 48 was taken near Seal No. 8. It shows extensive damage to the wire mesh. Pressures from a deflagration increase as the flame travels away from the origin in all directions. Higher pressures are achieved as the flame speed continues to increase. Figure 48 - Picture Taken Near Seal No. 8 Bottom mining was not conducted in the areas where the seals were constructed. Therefore, as the flame approached the location of the seals, the size of the entries decreased. These restrictions caused the mine atmosphere to become pressurized prior to the arrival of the flame front. When the flame entered the pressurized mine atmosphere, the pressures increase. This is commonly known as pressure piling. Pressures associated with a deflagration followed by pressure piling are different from pressures associated with detonations of fuel. Deflagrations begin with low pressures and low flame speeds. When restrictions are encountered during a deflagration, pressure piling effects can result in excessive pressures. Pressure piling during a deflagration can result in a deflagration to detonation transition (DDT). A detonation occurs when the reaction moves through the unburned fuel at a speed that exceeds the speed of sound. Explosions that begin as detonations result in excessive pressures at the origin of an explosion. The same direction of 175 pressure can occur as in a deflagration but variations in the magnitude can be used to identify the type of explosion and the origin. A methane accumulation was ignited within the sealed area near survey stations 4010 and 4011 in the 2nd Left Mains. A deflagration began as the flame propagated in all directions from the origin. During the outby propagation of the explosion, flame speeds and pressures increased. As the flame approached the location of the ten Omega block seals, it propagated through an area with a pressurized mine atmosphere caused by the presence of the seals. In addition, the mine openings became restricted as the flame passed out of the area that had been bottom mined. The pressurized mine atmosphere, along with the increased pressure due to height restrictions, caused pressure piling to occur. This condition resulted in excessive pressures which completely destroyed the ten seals. Appendix LL contains a mine map that details the extent of flame and the direction of the primary explosion forces. Flame Extent of Flame A Mine Dust Survey was conducted throughout underground areas after the explosion. The mine dust samples were sent to MSHA’s Laboratory in Mount Hope, West Virginia for analysis. Each of the mine dust samples was subjected to an Alcohol Coke Test. The Alcohol Coke Test identified the portion of coke in each of these samples. Coke occurs as coal is de-volatilized during a heating process, allowing mainly carbon to remain. The results of the Alcohol Coke Test indicated the quantity of coke in each sample as either none, trace, small, large, or extra large. Large and extra large quantities of coke in the post-explosion mine dusts are indicative of underground areas exposed to explosion flame. However, it is possible for mine dust samples within the flame zone to show none, trace, or small quantities of coke. For example, the explosion flame can travel at a velocity that is too fast to allow sufficient time for coal to coke or if coal dust is not dispersed into the explosion flame. In the 2nd Left Mains, located entirely within the sealed area, coke indicating flame was found in every sample. In the 2 North Main entries inby the location of the ten seals, coke indicating flame was found in most samples. The flame from the explosion did propagate throughout an extensive area of the 2nd Left Main entries and the inby portions of the 2 North Main entries. However, due to the large number of mine dust samples that could not be collected in these entries, it was not possible to accurately determine the location of the inby edge of the flame. 176 In the 2 North Main entries outby the location of the ten seals, coke indicating flame was only found in two samples. The flame from the explosion did not propagate significantly outby the location of any of the ten seals, with the exception that flame extended for a short distance outby the seal locations in the Nos. 7 and 8 entries. In both 1st Left and 2nd Left Parallel, coke indicating flame was not found in any samples. The flame from the explosion did not propagate into either 1st Left or 2nd Left Parallel. The flame of an explosion is extinguished due to a lack of fuel, suspension, heat, oxygen, confinement, or a combination of these five factors. The extent of flame is shown on the mine map in Appendix LL. Fuel and Suspension Methane is naturally suspended as it enters the mine workings. Prior to the explosion, coal dust would not have been suspended in the mine atmosphere on either side of the ten seals. When the minimum explosive concentration of coal dust is suspended, the cloud is so dense that you cannot see through it nor can you breathe in it. Research has shown that ignition of as little as 13 cubic feet of methane, diluted to within the explosive range, would be sufficient to suspend and ignite a coal dust cloud.58 Coal dust may have been involved to a limited degree throughout the sealed area as the flame propagated. Methane provided the primary fuel for the explosion. After ignition, the explosion propagated away from the origin in all directions. Explosive quantities of methane, in the range of 5% to 15%, were initially available for the explosion. It is possible that these explosive accumulations existed throughout the entire cross-sectional area of some entries and crosscuts. Concentrations in excess of the explosive range of methane were probably present in the sealed area prior to the explosion. A portion of these methane layers may have been diluted into the explosive range due to the turbulence of the propagating explosion. As the methane explosion propagated, a shock wave occurred with its resultant overpressure. This overpressure may have resulted in the suspension of mine dust, including coal dust, from the mine roof, ribs, and floor. Methane remained suspended for the duration of the explosion. The flame was not extinguished due to a lack of fuel or a lack of suspension. 58 The Explosion Hazard in Mining, U.S. Department of Labor, Mine Safety and Health Administration, Informational Report 1119 (1981), John Nagy, Page 56. 177 Heat Explosion flames exceed the ignition temperatures of both methane and coal dust. Rock dust and other incombustible dusts in suspension reduce the heat available for continued flame propagation. When an area is wet, coal dust will not become readily suspended during an explosion and therefore will not become involved in its propagation. Rock dust or other incombustible dusts in sufficient suspended quantities can extinguish or prevent a coal dust explosion. In addition to the available rock dust, Omega blocks are manufactured of incombustible material. A significant quantity of Omega blocks were used to construct the ten seals. Broken and unused Omega blocks were sometimes placed along rutted entries and crushed by the operation of mining equipment. This deliberate action filled ruts and unintentionally provided additional incombustible material in the area. In addition, the force of the explosion, coupled with a high degree of impact damage from collision with ribs and wood crib blocks, resulted in the pulverization of a significant portion of many Omega blocks and the suspension of the resulting dust. Figure 49 shows debris along the rib outby Seal No. 2. This debris included a significant amount of pulverized Omega blocks. Figure 49 - Picture of Debris Outby Seal No. 2 Research has shown that explosion flame cannot successfully penetrate ten feet into a cloud of coal dust suspended at a concentration of 5.0 ounces per cubic 178 feet.59 This shows that dense dust clouds, regardless of their composition, will not allow the propagation of explosion flame to penetrate. Such a sufficiently dense cloud of suspended dust may have existed at the location of the ten seals as the explosion flame approached. This suspended dust may have acted as a heat sink, which prevented the continued propagation of the explosion flame into the active workings. Therefore, the loss of sufficient heat may have been a factor responsible for extinguishing the explosion flame at the location of the ten seals. Oxygen Methane requires at least 12% oxygen to become or to remain involved in any combustion process.60 Where flame evidence existed throughout the sealed area, it is certain that oxygen concentrations above this minimum occurred. The active workings would have contained an oxygen concentration of about 20.9% before and during the explosion. The flame of the explosion consumed most of the oxygen. The flame of an explosion would generally not be able to burn back through the same area because of the lack of oxygen immediately after the explosion. Oxygen concentrations after a methane explosion could be less than 4%, depending on the initial methane concentration.61 The explosion flame propagated outward in all directions from the point of origin near survey stations 4010 and 4011 in the 2nd Left Mains. The lack of oxygen prevented the flame from burning through the same area twice but did not extinguish the propagating flame front traveling in all directions. Confinement Confinement is related to the cross-sectional area of the opening where a propagating explosion travels. It allows pressures to continue and, if the explosion is fueled by dust, it keeps the fuel particles in close proximity to one another. If the opening size increases or if additional entries become available for the flame front, confinement could be lost. The loss of confinement would cause a decrease in the speed of the explosion and a resulting reduction in pressure. The lack of confinement did not occur and was not responsible for extinguishing the explosion flame. 59 Id, page 36. 60 Limits of Flammability of Gases and Vapors, U.S. Bureau of Mines, Bulletin 503, (1952), Page 131. 61 The Explosion Hazard in Mining, U.S. Department of Labor, Mine Safety and Health Administration, Informational Report 1119, (1981), John Nagy, Page 63. 179 Sealed Areas Past explosion research was concentrated in unrestricted entries. Prior to 2006, neither the USBM nor NIOSH conducted full-scale explosions in sealed areas. No specific mining publications were available detailing the effects of the flames and forces of a propagating explosion in a sealed area. Also, MSHA has not traveled into the sealed area during the investigation of any previous explosions which occurred in sealed areas. Little research has been conducted to quantify the effects of pressure piling in coal mines. Force An explosion can propagate as a deflagration or a detonation. A deflagration occurs when the reaction moves through the unburned fuel at a speed that remains below the speed of sound, which is about 1,129 feet per second (fps) at 70°F. A detonation occurs when the reaction moves through the unburned fuel at a speed that exceeds the speed of sound. In the underground coal mine environment, deflagrations are typical. Both deflagrations and detonations can produce excessive pressures. A factor which can significantly increase the pressures at any particular location is known as pressure piling. Pressure piling occurs when the mine atmosphere becomes pressurized prior to the arrival of the flame front. One physical factor that can lead to pressure piling occurs when the total dimensions of the opening through which the explosion is propagating become increasingly restricted, thus the flow of gases for pressure equalization is inhibited. Deflagration After ignition of methane, the flame of a deflagration heats the mine atmosphere. The heated atmosphere expands as a result. This expansion exerts a pressure on mine surfaces, equipment, ventilation controls, and miners. This pressure is sometimes referred to as a shock wave. The faster the flame propagates, the higher the pressures become. Regardless of the speed of the flame, it can not overtake the shock wave. Research has indicated that flame speeds of approximately 400 feet per second (fps) may result in pressures of about 7 psi. When the flame increases in speed to near 1,000 fps, research indicates that the expected force may be on the order of 17 psi.62 Many conditions underground can affect the magnitude of explosion pressure including, but not limited to; change in height or width of opening, the presence of large equipment, change in the number of entries or crosscuts, the concentration of fuel being consumed, the 62 Id, page 62. 180 strength of the ignition source, the percentage of suspended dust that is incombustible, and the amount of oxygen available for combustion. The calculated theoretical value of maximum static pressure for coal dust or methane explosions in closed vessels is about 140 psi.63 The maximum pressure that can be expected from an explosion fueled by either coal dust or methane traveling at deflagration speeds in an underground coal mine would be less than 100 psi.64 In an underground explosion, complete combustion does not occur and heat is lost to the mine surfaces, which accounts for the lower pressure. As the speed of the flame decreases and the flame eventually terminates, pressures reduce and are eliminated. According to the U. S. Bureau of Mines (USBM) Report of Investigations (RI) 7581 entitled “Explosion-Proof Bulkheads,” with adequate incombustible material and minimum coal dust accumulations, it is doubtful that pressures exceeding 20 psi could occur very far from the origin of the explosion. Several oscillations of pressure can occur before ambient conditions are reached. This allows for pressure in both directions to occur at each underground location within the explosion zone. The mine map shown in Appendix LL shows the direction of the primary forces. The primary force is the first, or initial, force at each location shown. Pressure Piling The explosion pressures achieved during pressure piling are contingent upon the compression of the fuel ahead of the flame. This compression of the fuel increases as the speed of the flame increases and as the opening becomes more restricted. The explosion pressure in a pre-compressed fuel/air mixture is proportional to the absolute pressure. If the mine atmosphere is pre-compressed to about 45 psi, the instantaneous explosion pressure could be as high as 300 psi. As an example of pressure piling, computations after one coal dust explosion experiment in a dead end entry indicated that a peak static pressure of not less than 595 psi was reached.65 63 Id, page 69. 64 Id, Page 58. 65 Coal Dust Explosions and Their Suppression, National Center for Scientific, Technical and Economic Information, Warsaw, Poland, (1975), p. 284. 181 Figure 50 shows a typical side view of the entry in which Seal No. 10 was constructed. The seal is shown on the left. The line across the top of the figure shows the elevation of the roof. The line across the bottom of the figure details changes in the mine floor due to bottom mining. Those areas to the right of the seal location shown in the figure are part of the sealed area. The height of the entry at the bottom of the ramp, where bottom mining commenced, is approximately 2.5 times the height at the location of the seal. Figure 50 - Contours Near Seal No. 10 in 2 North Mains, No. 9 Entry As the explosion propagated toward the location of the ten seals, the mine atmosphere immediately behind the seals would have experienced increased pressure due to the shock wave of the approaching flame front. Pressures would have increased dramatically as the explosion propagated up the ramp into more restricted entry heights. The pressurized mine atmosphere immediately behind the seals would have caused pressure piling effects. Detonation In a detonation, shock waves may develop at the flame front. These shock waves advance ahead of the flame and reinforce each other in the unburned fuel/air mixture. When the energy in these shock waves is sufficient, self-ignition of the mixture occurs and new, multiple flame fronts develop. The instantaneous static pressure from the detonation may be several times higher than 100 psi. The static pressure in a mine explosion can be as little as a fraction of a pound per square inch or more than 600 psi.66 66 The Explosion Hazard in Mining, U.S. Department of Labor, Mine Safety and Health Administration, Informational Report 1119, (1981), John Nagy, Page 60. 182 NIOSH Assistance NIOSH, at MSHA’s request, initiated two test explosions in the face area of Drift C of Lake Lynn to compare the effects of an explosion in a sealed area and an explosion in an open entry. The same amount and concentration of methane and the same ignition source were used for both explosions. In the first explosion, labeled as #501, seals were constructed in the first three crosscuts between Drift C and Drift B at distances of 59 feet, 156 feet, and 256 feet from the face. Drift C remained open. Also, two cribs were constructed in the entry about 312 feet from the face. In the second explosion, labeled as #502, the seals in the crosscuts remained in place. A fourth seal was constructed across Drift C at a distance of about 320 feet from the face, effectively sealing the inby area of Drift C. The seal in No. 1 Crosscut was the same for both tests and was constructed of solid concrete blocks, according to 30 CFR 75.335. The seal in No. 2 Crosscut was the same for both tests and was an Omega block seal constructed in the same manner as the Omega block seal that passed explosion testing in 2001. The seal in No. 3 Crosscut was the same for both tests and was the hybrid Omega block seal. This seal was built with the following conditions: applying dry mortar on the mine floor, not applying mortar directly to the vertical joints of the first course of blocks, and modifying the installation of wood planks and wedges between the last course of the Omega blocks and the mine roof. For test #502, an Omega seal constructed in the same manner as the Omega block seal that passed explosion testing in 2001 was built across Drift C. Pressure readings were recorded at locations where transducers were mounted in panels along Drift C. Table 11 contains the results for both explosion tests #501 and #502. Table 11 - Results of Explosion Test #501 and #502 at Lake Lynn Location No. 2 Crosscut No. 3 Crosscut Crib Test Seal Outby Outby Outby Outby Distance from Face (feet) 156 256 312 320 403 501 598 757 Test #501 Pressure (psi) 22 25 28 -14 9 6 4 Test #502 Pressure (psi) 22 38 -50 5 4 3.5 3 The results shown above indicate that seals do affect the magnitude of the pressure achieved at locations both inby and outby the seals. The test seal in Drift C was about 320 feet from the face. In Test #501, the maximum pressure achieved was 30 psi and occurred at the face. The pressure on the seal in No. 2 183 Crosscut was 22 psi and on the seal in No. 3 Crosscut was 25 psi. This slight increase in pressure was possibly attributed to the distance it took to involve the suspended coal dust. Subsequently, this pressure began to deteriorate as it traveled outby, however the two cribs constructed in the entry caused the pressure to increase to 28 psi. This increase was recorded at a distance of 312 feet from the face. Within 53 feet, the pressure had dropped to 14 psi and after traveling another 195 feet, which corresponds to the location approximately 598 feet from the face, the pressure had dropped to 6 psi. At 757 feet from the face, the pressure had dropped to 4 psi. In Test #502, the seal in No. 2 Crosscut was 156 feet from the face and the maximum pressure on this seal was 22 psi. The seal in No. 3 Crosscut was 256 feet from the face and the maximum pressure on this seal was 38 psi. This significant increase in pressure can most likely be attributed to pressures rebounding or reflecting after impacting the seal in Drift C, prior to its destruction. The maximum pressure increased to 50 psi at the location of the seal in Drift C, which was 320 feet from the face. When the pressure wave reached 403 feet from the face, the pressure had dropped to only 5 psi. This drop is very significant in that pressures decreased 90%, from 50 psi to 5 psi, in a distance of only 80 feet. The flame of the explosion had most likely consumed all the available fuel and propagation of the explosion was not continuing. However, the pressures that developed impacted the seal across Drift C and rebounded toward the face. The primary thrust of the pressure wave did not head outby after rebounding. At 757 feet from the face, the pressure had dropped to 3 psi. The Sago Mine Explosion A methane explosion initially occurred in 2nd Left Mains in the general area of survey stations 4010 in the No. 6 entry and 4011 in the No. 7 entry. These survey stations are located in the No. 2 Crosscut. As the flame from this explosion expanded, it began to propagate through explosive concentrations of methane in all directions. Some of the sealed area may have included no methane or methane at concentrations below 5%. Additionally, concentrations of methane above the explosive limit were probably present in some locations throughout the sealed area prior to the explosion. A portion of this methane may have been diluted into the explosive range due to the turbulence of the propagating explosion. Methane that remained in concentrations outside the explosive range did not become involved in the explosion. It appeared that the flame and the associated forces initially traveled inby and outby through the Nos. 6 and 7 entries. The length of flame and the generation of forces in any direction are dependent on the amount of explosive methane accumulations in that direction. It is typical in underground coal mine explosions that limited forces occur at the origin of the explosion. 184 The explosion propagated inby in the Nos. 6 and 7 entries of 2nd Left Mains. The flame and forces traveled in both directions from these entries through each crosscut. Subsequently, the flame and forces involved each entry of 2nd Left Mains and continued propagating inby. Although the inby extent of the flame is unknown due to the lack of mine dust samples, the forces would have affected all of 2nd Left Mains to varying degrees. The most inby mine dust sample was taken in the No. 5 Crosscut. The magnitude of the forces would have varied greatly, especially as the explosion was directly affected by change in crosssectional dimensions of the entries and crosscuts. The mine map in Appendix H4, “2 Left Mains” shows the extent of bottom mining throughout the 2nd Left Mains. As the flame and forces traveled through the crosscuts toward the No. 1 entry, stoppings were destroyed. Stoppings and overcasts can be destroyed by explosion forces of between 2 and 4 psi. Ventilation controls were damaged throughout the area affected by the explosion. This and other damage throughout the 2nd Left Mains is indicated on the mine map in Appendices H-1 through H-9. The explosion propagated outby in the Nos. 6 and 7 entries of 2nd Left Mains. The explosion flame and forces entered the 2 North Mains throughout Nos. 65 and 66 crosscut. The flame and forces headed in both directions through each entry in 2 North Mains. Subsequently, the flame and forces involved each entry of 2 North Mains and continued propagating both inby and outby. Although the inby extent of the flame is unknown due to the lack of mine dust samples, the forces would have affected all of 2 North Mains inby the seals to varying degrees. The most inby mine dust sample was taken in the No. 3 entry, just inby No. 66 crosscut. The magnitude of the forces would have varied greatly, especially as the explosion was directly affected by change in cross-sectional dimensions of the entries and crosscuts. The mine map in Appendix H-4, “2 Left Mains” shows the extent of bottom mining throughout the 2 North Mains. Bottom mining occurred inby the location of the seals in the entries but not in the crosscuts of 2 North Mains. Bottom mining occurred as close as about 60 feet inby the location of the seals. Entry heights increased from about 6 feet at the location of the seals to about 20 feet at some locations in the areas that had been bottom mined. As the explosion propagated outby in the entries of 2 North Mains, the total height of the opening through which the explosion was propagating became increasingly restricted as the explosion approached the seals. This restriction resulted in pressure piling, as explained earlier. A resulting and drastic increase in the explosion pressures occurred at the location of the restriction and at each of the seals. The increase in pressures was not on the order of magnitude necessary to cause a deflagration to detonation transition (DDT). 185 Visual observations were made of the remnants of the ten seals at the Sago Mine. Visual observations were also made of the post-explosion condition of each of the seals constructed at NIOSH’s Lake Lynn Experimental Mine. Conditions at these two mines are not identical, but comparisons were made concerning the destruction of the Omega blocks at different pressures. The damage to the seals at the Sago Mine was more extensive. This comparison would indicate that the pressures exceeded 93 psi at the location of the seals at the Sago Mine. One victim who suffered fatal injuries was found about 556 feet outby the seal in the No. 6 entry of 2 North Mains. The exact location of this miner at the time of the explosion could not be established. His death was attributed to carbon monoxide intoxication. No traumatic physiological injuries were present, indicating pressures of less than 5 psi at his location. The 2nd Left Parallel crew survived the flame and forces of the explosion without experiencing any known traumatic physiological injuries. McCloy indicated that he felt wind and heard noise. He felt pressure in his ears but they did not pop. Although the miners may have been on the section, the exact location at the time of the explosion could not be established. A significant reduction in explosion pressures occurred in 2 North Mains just outby the seals. These reduced pressures would have propagated into the crosscuts of 2 North Mains and traveled hundreds of feet before having any impact on the 2nd Left Parallel crew. This indicates they may have been affected by a pressure wave of less than 2 psi. The 1st Left crew was located in a mantrip at the 1st Left switch when the explosion occurred. They were in direct line with the seals and the explosion forces. They were impacted by flying debris and a rush of air, which reportedly lasted for about 8 seconds. The mantrip operator was knocked down by the force of the explosion. They did not hear any noise. They did not smell any smoke initially. They did not see any flash or flame. After the rush of air, the atmosphere was very dusty. They did not experience any traumatic injuries resulting in physiological damage such as lung damage from pressure, ruptured ear drums or broken bones. Tests of the mine dust, as well as their testimony, indicated the flame from the explosion did not extend to their location. This indicates they may have been affected by a pressure wave of about 2 psi. 186 ROOT CAUSE ANALYSIS A root cause analysis was conducted. Root causes were identified that could have mitigated the severity of the accident or prevented the loss of life. Listed below are root causes identified during the analysis and their corresponding corrective actions to prevent a recurrence of the accident. Root Cause: The 2 North Main seals were not capable of withstanding the forces generated by the explosion. Corrective Action: Seals should be designed and installed to prevent an explosion from propagating to the opposite side. Root Cause: The atmosphere within the sealed area was not monitored and it contained explosive methane/air mixtures. Corrective Action: The atmosphere in existing sealed areas should be monitored and maintained inert when the seals are not capable of withstanding the forces of an explosion. Root Cause: Lightning was the most likely ignition source for this explosion with the energy transferring onto an abandoned pump cable in the sealed area and providing an ignition source for the explosion. Corrective Action: Insulated cables and conductors should be removed from the area to be sealed prior to seal completion. 187 CONCLUSION On January 2, 2006, an explosion occurred at approximately 6:26 a.m. in the mined-out area known as 2 North Mains and 2nd Left Mains of the Sago Mine. Lightning was the most likely ignition source for this explosion with the energy transferring onto an abandoned pump cable in the sealed area and igniting the methane that had accumulated within the sealed area. The ensuing explosion generated forces in excess of 93 psi and destroyed the seals, filling portions of the mine with toxic levels of carbon monoxide. One miner died of carbon monoxide poisoning shortly after the explosion. The 2nd Left Parallel miners’ attempt to evacuate was unsuccessful and they barricaded themselves on the 2nd Left Parallel section. Unfortunately, the barricade was not able to prevent high levels of carbon monoxide from reaching the miners before they could be rescued. As a result, 11 additional miners perished and one survived. 188 ENFORCEMENT ACTIONS A 103 (k) order was issued to Wolf Run Mining Company, Sago Mine, on the morning of the accident to insure the safety of all persons at the mine. The order was modified numerous times to allow for the rescue and recovery operations, and then the accident investigation, to proceed. This order remained in place for the extent of the investigation. A contributory citation is one issued for a condition that leads to the causes and effects or the severity of the accident. No contributory citations were issued to the mine operator as a result of the accident investigation. As indicated in the report, safety standard violations were identified with respect to seal construction, SCSR training, emergency notification to MSHA and mine rescue teams, lightning arresters and various other violations. In addition to the 103 (k) order, 149 non-contributory citations/orders were issued as a result of this investigation. One hundred seventeen were issued previously and 32 have been issued with the release of this report. Some of the more significant enforcement actions are addressed below along with an explanation of why they were not deemed contributory. • Although the 2 North Main seals were not built in accordance with the approved ventilation plan requirements, the forces generated by the explosion would have completely destroyed the seals even if they had been built in compliance with the plan. • Several miners did not don their SCSRs immediately after the explosion as required and some apparently removed the units to communicate or to perform physical work. However, those who did not don their SCSRs successfully evacuated the mine. The miners on the 2nd Left Parallel section donned their SCSRs but were exposed to high levels of CO far beyond the one-hour time capacity for each SCSR. • MSHA and mine rescue teams were not immediately notified of the accident. This is an important requirement in order to maximize professional assistance. However, had agency officials and the mine rescue teams arrived earlier, the teams would not have been permitted underground any earlier than actually occurred due to the high levels of toxic gases and the possibility of another explosion. Even if the mine rescue teams had been on site and entered the mine immediately after the accident, they would have been withdrawn when they encountered the high carbon monoxide levels. 189 • Five electrical circuits entering/exiting the mine did not have lightning arresters. Testing indicated that a direct lightning strike onto these circuits could not have traveled far enough into the mine to initiate the explosion. Even though these violations did not directly lead to the cause and effect or the severity of the accident, they are important matters that miners and the mining industry should be aware of and attentive to in order to prevent and minimize coal mine accidents. 190 Appendix A - List of Deceased and Injured Miners Deceased Miners Name Thomas P. Anderson Alva M. Bennett James Bennett Jerry L. Groves George J. Hamner Terry Helms Jesse L. Jones David W. Lewis Martin Toler Jr. Fred G. Ware Jackie L. Weaver Marshall Winans Occupation 2nd Left Parallel Shuttle Car Operator 2nd Left Parallel Mining Machine Operator 2nd Left Parallel Shuttle Car Operator 2nd Left Parallel Roof Bolter Operator 2nd Left Parallel Shuttle Car Operator Mine Examiner 2nd Left Parallel Roof Bolter Operator 2nd Left Parallel Roof Bolter Operator 2nd Left Parallel Section Foreman 2nd Left Parallel Mining Machine Operator 2nd Left Parallel Electrician 2nd Left Parallel Scoop Operator Injured Miner Name Randal McCloy Jr. Appendix A - Page 1 of 1 Occupation 2nd Left Parallel Roof Bolter Operator Injury Carbon Monoxide Poisoning Appendix Sago Mine, MSHA ID 46-08791 Wolf Run Mining Company Mine Map 1000? 500? [j GU A JIQDDEDEDD El 0' DD 1/ [1 D: ENDED-D CID-D DEEDS mgr: - EDD-DEBS DCH: :3 13:11:] DEED - 9; ul? 7W EEDEEDEDE EEK: DI .1 EH: .- -1- SEEDS - - :1 QUIZZES 0?3 Emma UDUU 0 DOD DOD DO CID 0 at) BUD EIDQIUDE 3 alumgmu I??g?g?g DE Appendix C - Mine Rescue Personnel and Teams Responding Consol Energy Corporation Mine Rescue Personnel Spike Bane Elizabeth Chamberlain Rick Marlow Kevin Whetzel Blacksville No. 2 James Ponceroff Richard Tolka Lonny Meyers David Rush Robert Wade Tony Casini McElroy Mine Rescue Danny Beyser Dennis Crow James Klug Kelvin Jolly Robert Rohoe Michael Clark James Smith Randy Clark Jack Price William Blackwell Shoemaker Mine Rescue Cliff Ward Charles Fisher Ted Hunt Silas Stavischeck Glenn McWhorten Jim Jack Okey Rine Robert Hines Shan Michener Robinson Run Mine Rescue Alfred Bell Craig Carpenter Sherman Goodwin Phillip Morgan Larry Tenney Jerry Bienkoski Gary Given Mark Koor William Reed Loveridge Mine Rescue Leslie R. Cosner Nick A. Tippi Donald A. Jack Robert Hovatter Appendix C - Page 1 of 3 James Clendenen Richard Shockley Charles P. Layman Gary Hayhurst Appendix C - Mine Rescue Personnel and Teams Responding (cont’ d) Eighty-Four Mine Rescue Don Klek Dale Tiberie Richard Gindlesperger Kenneth Clark Robert Volpe Adrian Gordon John Stowinsky Dan Puckey Brad DeBush Mickey Miskiewiez Bailey Mine Rescue Dennis Vicinelly Mike Spears Dave Cass Gene Menozzi Larry Cuddy George Joseph Kevin Williamson Steve Edgehouse Bob Calhoun Barbour County Mine Rescue Mark W. Chewning Fred Radabaugh Brian Curtis Roger Hedrick Teddy Hickman Ryan Jeran Clyde M. Tenney Doug Andrews Paul Maxson Jeff Byard John Cottrill James Paugh Viper Mine Rescue Brad Kaufman Ty Hunt Paul Perrine Brandon Sanson Clifford Bryant Jr. Allen Setzer Bret Bushong Tri-State Christopher C. Lilly Mike Grimm Kerry Lilly Mark Thorn Dan Bismark Appendix C - Page 2 of 3 Andrew Lilly Ben Wilson Don Firn Chris Sisler Gary Bolyand Appendix C - Mine Rescue Personnel and Teams Responding (cont’ d) State Mine Rescue Region 1 Barry Fletcher Jeffery Bennett John Scott John Hall Region 4 Clarence Dishman Mike Rutledge Eugene White William Tucker Randy Smith Region 3 James Hodges MSHA Mine Emergency Unit Virgil Brown Jerry Cook Charles Pogue Mike Hicks Ronald Hixson Greg Ison James Langley Jan Lyall Scott Mandeville Fred Martin Frank Thomas Stanley Sampsel Clayton Sparks Mike Shumate Tony Sturgill Pittsburgh Safety and Health Technology Center John Urosek William Francart Gary Shemon Mike Valoski Appendix C - Page 3 of 3 Richard Stoltz George Aul George Durkt Tony Argirakis Appendix D - Barbour County Mine Rescue Team Air Quality Measurements Date 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 Time 1:25 PM 1:27 PM 1:29 PM 1:30 PM 1:46 PM 1:47 PM 1:48 PM 1:50 PM 2:04 PM 2:04 PM 2:06 PM 2:06 PM 2:08 PM 2:08 PM 2:09 PM 2:09 PM 2:27 PM 2:27 PM 2:33 PM 2:33 PM 2:38 PM 2:38 PM 2:42 PM 2:42 PM 3:14 PM 3:16 PM Drift Opening No. 1 Drift Opening No. 2 Drift Opening No. 3 Drift Opening No. 4 CO Methane Oxygen CO Methane Oxygen CO Methane Oxygen CO Methane Oxygen (ppm) (%) (%) (ppm) (%) (%) (ppm) (%) (%) (ppm) (%) (%) 1372 0 20.6 77 0 20.9 24 0 20.7 14 0 20.5 1860 0 155 0 20.7 54 0 20.7 32 0 20.6 2430 0.6 19.7 2700 0.5 19.6 417 0 20.6 544 0 20.5 178 0 20.6 179 0 20.4 135 0 20.7 105 0 20.5 1880 .0.6 19.7 2280 .0.5 19.6 292 0 20.7 212 0 20.5 127 0 20.6 92 0 20.5 72 0 20.5 69 0 20.4 1800 0.5 19.5 507 0 20.3 Appendix D - Page 1 of 3 Appendix D - Barbour County Mine Rescue Team Air Quality Measurements Date 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/3/2006 1/3/2006 1/3/2006 1/3/2006 Time 3:35 PM 3:37 PM 3:52 PM 4:02 PM 4:04 PM 4:07 PM 4:09 PM 4:35 PM 5:19 PM 5:21 PM 5:40 PM 5:42 PM 5:55 PM 6:30 PM 6:35 PM 7:45 PM 8:15 PM 8:50 PM 9:37 PM 10:17 PM 10:58 PM 11:55 PM 12:30 AM 1:00 AM 1:30 AM 2:00 AM Drift Opening No. 1 Drift Opening No. 2 Drift Opening No. 3 Drift Opening No. 4 CO Methane Oxygen CO Methane Oxygen CO Methane Oxygen CO Methane Oxygen (ppm) (%) (%) (ppm) (%) (%) (ppm) (%) (%) (ppm) (%) (%) 2136 0.4 19.6 280 0 20.4 1985 0.4 19.9 2251 0.4 20.2 410 0 20.9 120 0 20.9 82 0 20.9 2064 0.5 20.1 1626 0.3 19.9 369 0 20.4 1375 0.3 20.2 255 0 20.4 1358 0.3 19.9 1198 0.3 19.9 275 0 20.4 16 0 20.7 17 0 20.5 0 0 20.7 0 0 20.9 0 0 20.9 0 0 20.0 0.6 0 20.9 0 0 20.9 0 0 20.9 0 0 20.9 0 0 21.1 Appendix D - Page 2 of 3 Appendix D - Barbour County Mine Rescue Team Air Quality Measurements Date 1/3/2006 1/3/2006 1/3/2006 1/3/2006 1/3/2006 1/3/2006 1/3/2006 1/3/2006 1/3/2006 1/3/2006 1/3/2006 1/3/2006 1/3/2006 1/3/2006 1/3/2006 1/3/2006 1/3/2006 1/3/2006 1/3/2006 Time 2:30 AM 3:00 AM 3:30 AM 4:00 AM 4:26 AM 4:55 AM 6:26 AM 6:55 AM 7:25 AM 7:55 AM 8:25 AM 8:55 AM 9:25 AM 9:55 AM 10:00 AM 10:32 AM 10:35 AM 10:54 AM 10:55 AM Drift Opening No. 1 Drift Opening No. 2 Drift Opening No. 3 Drift Opening No. 4 CO Methane Oxygen CO Methane Oxygen CO Methane Oxygen CO Methane Oxygen (ppm) (%) (%) (ppm) (%) (%) (ppm) (%) (%) (ppm) (%) (%) 0 0 20.9 0 0 20.9 0 0 20.9 0 0 20.9 0 0 20.9 0 0 20.9 0 0 20.9 0 0 20.9 0 0 20.9 0 0 20.9 0 0 20.9 0 0 20.9 0 0 21.1 0 0 20.9 322 0.2 20.7 0 0 20.9 311 0.2 20.6 0 0 20.9 0 0 20.9 Note: Team continued to take handheld measurements throughout rescue effort. Appendix D - Page 3 of 3 Appendix E - Gas Chromatograph Analysis Results No. 1 Drift Opening ICG, Inc. Sago Mine 46-08791 No. 1 Drift Opening Date and Time 1/2/2006 14:45 1/2/2006 15:00 1/2/2006 15:10 1/2/2006 15:30 1/2/2006 15:45 1/2/2006 16:00 1/2/2006 16:30 1/2/2006 17:15 1/2/2006 17:45 1/2/2006 17:55 1/2/2006 18:30 1/2/2006 19:20 1/2/2006 19:20 1/2/2006 19:55 1/2/2006 20:00 1/2/2006 20:30 1/2/2006 21:00 1/2/2006 21:00 1/2/2006 21:30 1/2/2006 22:00 1/2/2006 22:00 1/2/2006 22:30 1/2/2006 23:00 1/2/2006 23:30 1/3/2006 0:00 1/3/2006 0:30 1/3/2006 1:00 1/3/2006 1:30 1/3/2006 2:00 H2 ppm 755 890 893 567 792 752 671 599 657 726 639 O2 % 19.26 19.71 20.26 19.68 19.91 19.15 20.41 19.84 20.46 20.43 20.46 20.26 20.44 20.12 20.48 20.06 20.49 20.46 20.52 20.51 20.36 20.42 20.52 20.54 20.60 20.55 20.57 20.57 20.56 Appendix E - Page 1 of 10 GAS CONCENTRATIONS CH4 CO CO2 C2H2 C2H4 C2H6 % ppm % ppm ppm ppm N2 % 78.15 78.10 78.10 78.09 78.10 78.10 78.06 78.12 78.11 78.10 78.13 0.29 0.27 0.27 0.25 0.24 0.22 0.22 0.22 0.20 0.19 0.18 0.18 0.14 0.18 0.14 0.17 0.14 0.17 0.18 0.17 0.17 0.17 0.15 0.14 0.14 0.14 0.13 0.13 0.13 2600 2570 2340 2130 1970 1870 1750 1740 1510 1420 1290 1130 1101 1060 1025 1000 940 960 950 893 920 900 845 806 779 751 725 699 654 Page 1 of 6 0.28 0.27 0.29 0.24 0.23 0.23 0.22 0.22 0.20 0.18 0.17 0.16 0.15 0.15 0.15 0.14 0.14 0.14 0.13 0.15 0.13 0.13 0.13 0.13 0.12 0.12 0.12 0.12 0.11 12 0 21 32 22 31 26 28 19 6 8 15 5 10 10 30 21 25 19 14 21 21 0 0 0 0 0 0 0 10 10 10 10 10 6 10 6 10 11 10 10 28 10 10 16 4 0 0 0 0 0 Ar % 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 Appendix E - Gas Chromatograph Analysis Results No. 1 Drift Opening ICG, Inc. Sago Mine 46-08791 No. 1 Drift Opening Date and Time 1/3/2006 2:30 1/3/2006 3:00 1/3/2006 3:30 1/3/2006 4:00 1/3/2006 4:30 1/3/2006 5:00 1/3/2006 5:30 1/3/2006 6:00 1/3/2006 6:30 1/3/2006 7:00 1/3/2006 7:30 1/3/2006 8:00 1/3/2006 8:30 1/3/2006 9:00 1/3/2006 9:30 1/3/2006 10:00 1/3/2006 10:30 1/3/2006 11:00 1/3/2006 11:30 1/3/2006 12:00 1/3/2006 12:30 1/3/2006 13:00 1/3/2006 13:30 1/3/2006 14:30 1/3/2006 15:00 1/3/2006 15:30 1/3/2006 16:30 1/3/2006 17:00 1/3/2006 17:30 1/3/2006 18:30 1/3/2006 19:00 1/3/2006 19:30 1/3/2006 20:00 H2 ppm 632 632 359 491 435 543 410 374 379 358 404 301 363 267 250 230 207 205 184 170 190 188 164 115 99 83 70 61 56 55 52 50 50 GAS CONCENTRATIONS CH4 CO CO2 C2H2 C2H4 C2H6 % ppm % ppm ppm ppm O2 % N2 % 20.64 20.61 20.61 20.60 20.61 20.65 20.65 20.67 20.71 20.68 20.68 20.71 20.71 20.71 20.71 20.70 20.73 20.72 20.78 20.80 20.79 20.79 20.81 20.82 20.81 20.83 20.84 20.83 20.85 20.85 20.86 20.88 20.86 78.06 78.10 78.14 78.14 78.14 78.10 78.13 78.12 78.08 78.11 78.11 78.18 78.10 78.11 78.11 78.13 78.11 78.11 78.06 78.13 78.07 78.06 78.12 78.07 78.06 78.07 78.06 78.07 78.06 78.06 78.06 78.03 78.06 Appendix E - Page 2 of 10 0.13 0.13 0.12 0.12 0.12 0.12 0.12 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.10 0.10 0.11 0.10 0.10 0.11 0.10 0.09 0.09 0.09 0.09 0.09 0.09 0.09 0.08 0.09 0.08 Page 2 of 6 635 604 570 550 531 507 478 445 432 414 389 375 355 340 334 314 295 282 271 216 213 263 212 167 170 164 148 141 136 119 119 118 107 0.11 0.11 0.10 0.09 0.10 0.10 0.09 0.09 0.08 0.09 0.08 0.08 0.08 0.08 0.08 0.08 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.06 0.06 0.06 0.06 0.06 0.06 0.05 0.06 0.05 0.05 8 4 0 4 0 0 0 0 0 0 0 9 0 0 7 0 6 0 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 16 18 0 0 0 0 0 0 0 11 0 10 0 0 0 8 8 0 8 0 0 0 0 4 5 0 0 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Ar % 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 Appendix E - Gas Chromatograph Analysis Results No. 1 Drift Opening ICG, Inc. Sago Mine 46-08791 No. 1 Drift Opening Date H2 and Time ppm 1/3/2006 20:30 45 1/3/2006 21:00 42 1/3/2006 21:30 40 1/3/2006 22:00 41 1/3/2006 23:00 33 1/3/2006 23:30 31 1/4/2006 0:01 30 1/4/2006 1:15 29 1/4/2006 1:30 27 1/4/2006 2:00 26 1/4/2006 2:30 26 1/4/2006 3:00 24 1/4/2006 3:30 24 1/4/2006 4:00 23 1/4/2006 4:30 22 1/4/2006 5:00 21 1/4/2006 5:30 20 1/4/2006 6:00 20 1/4/2006 6:30 20 1/4/2006 7:00 19 1/4/2006 7:30 17 1/4/2006 8:00 18 1/4/2006 9:00 15 1/4/2006 10:00 15 1/4/2006 11:00 23 1/4/2006 12:00 24 1/4/2006 13:00 19 1/4/2006 14:00 18 1/4/2006 15:00 18 1/4/2006 16:00 17 1/4/2006 18:00 15 1/4/2006 20:00 14 1/4/2006 22:00 12 1/5/2006 0:00 10 O2 % 20.86 20.86 20.87 20.86 20.87 20.87 20.87 20.86 20.87 20.86 20.81 20.86 20.87 20.87 20.86 20.87 20.83 20.82 20.87 20.87 20.87 20.87 20.87 20.85 20.85 20.83 20.82 20.82 20.83 20.81 20.84 20.85 20.85 20.84 Appendix E - Page 3 of 10 N2 % 78.06 78.06 78.06 78.05 78.06 78.06 78.06 78.07 78.06 78.07 77.84 78.07 78.07 78.07 78.07 78.07 77.91 77.89 78.07 78.07 78.07 78.07 78.07 78.08 78.09 78.10 78.10 78.11 78.10 78.13 78.09 78.08 78.08 78.10 GAS CONCENTRATIONS CH4 CO CO2 C2H2 C2H4 C2H6 Ar % ppm % ppm ppm ppm % 0.08 102 0.05 0 0 0 0.93 0.08 128 0.05 0 0 0 0.93 0.08 120 0.05 0 0 0 0.93 0.08 109 0.06 0 0 0 0.93 0.08 97 0.05 0 0 0 0.93 0.08 95 0.05 0 0 0 0.93 0.08 95 0.05 0 0 0 0.93 0.08 80 0.05 0 0 0 0.93 0.08 76 0.05 0 0 0 0.93 0.08 72 0.05 0 0 0 0.93 0.08 69 0.03 0 0 0 0.93 0.08 66 0.05 0 0 0 0.93 0.08 61 0.05 0 0 0 0.93 0.08 58 0.05 0 0 0 0.93 0.08 54 0.05 0 0 0 0.93 0.08 51 0.05 0 0 0 0.93 0.08 49 0.05 0 0 0 0.93 0.08 49 0.05 0 0 0 0.93 0.08 46 0.05 0 0 0 0.93 0.08 43 0.05 0 0 0 0.93 0.08 43 0.05 0 0 0 0.93 0.08 39 0.05 0 0 0 0.93 0.08 31 0.05 0 0 0 0.93 0.08 26 0.05 0 0 0 0.93 0.08 30 0.05 0 0 0 0.93 0.09 27 0.05 0 0 0 0.93 0.09 24 0.05 0 0 0 0.93 0.09 24 0.05 0 0 0 0.93 0.09 25 0.05 0 0 0 0.93 0.09 22 0.05 0 0 0 0.93 0.09 20 0.04 0 0 0 0.93 0.09 18 0.05 0 0 0 0.93 0.09 15 0.05 0 0 0 0.93 0.09 13 0.04 0 0 0 0.93 Page 3 of 6 Appendix E - Gas Chromatograph Analysis Results No. 1 Drift Opening ICG, Inc. Sago Mine 46-08791 No. 1 Drift Opening Date and Time 1/5/2006 2:00 1/5/2006 4:00 1/5/2006 6:00 1/5/2006 8:00 1/5/2006 10:00 1/5/2006 12:00 1/5/2006 14:00 1/5/2006 16:00 1/5/2006 18:00 1/5/2006 20:00 1/5/2006 22:00 1/6/2006 0:00 1/6/2006 2:00 1/6/2006 4:00 1/6/2006 6:00 1/6/2006 8:00 1/6/2006 10:00 1/6/2006 12:00 1/6/2006 13:30 1/6/2006 15:30 1/6/2006 17:30 1/6/2006 19:30 1/7/2006 10:00 1/7/2006 10:30 1/7/2006 11:30 1/7/2006 12:30 1/7/2006 13:30 1/7/2006 14:30 1/7/2006 15:30 1/7/2006 17:30 1/7/2006 19:30 1/7/2006 21:30 1/7/2006 23:30 H2 ppm 9 8 7 7 6 6 5 5 4 3 4 3 3 3 3 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 GAS CONCENTRATIONS CH4 CO CO2 C2H2 C2H4 C2H6 % ppm % ppm ppm ppm O2 % N2 % 20.84 20.85 20.84 20.85 20.85 20.85 20.85 20.85 20.85 20.85 20.85 20.85 20.83 20.85 20.85 20.86 20.86 20.86 20.86 20.85 20.87 20.86 20.85 20.87 20.87 20.86 20.87 20.85 20.85 20.85 20.86 20.85 20.82 78.10 78.09 78.10 78.09 78.09 78.09 78.09 78.10 78.10 78.10 78.09 78.10 78.11 78.09 78.10 78.09 78.09 78.08 78.09 78.02 78.08 78.09 78.10 78.09 78.09 78.10 78.11 78.10 78.09 78.10 78.09 78.10 78.00 Appendix E - Page 4 of 10 0.09 0.08 0.08 0.08 0.09 0.09 0.08 0.08 0.08 0.08 0.08 0.08 0.09 0.09 0.08 0.08 0.08 0.08 0.08 0.08 0.07 0.08 0.08 0.08 0.07 0.07 0.05 0.08 0.08 0.08 0.08 0.08 0.08 Page 4 of 6 10 10 9 8 7 6 6 5 4 4 3 4 3 3 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.05 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.05 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Ar % 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93 Appendix E - Gas Chromatograph Analysis Results No. 1 Drift Opening ICG, Inc. Sago Mine 46-08791 No. 1 Drift Opening Date H2 and Time ppm 1/8/2006 1:30 1 1/8/2006 7:00 1 1/8/2006 9:00 1 1/8/2006 12:00 1 1/8/2006 14:30 1 1/8/2006 15:30 1 1/8/2006 17:30 1 1/8/2006 19:30 1 1/8/2006 21:50 1 1/9/2006 2:30 1 1/9/2006 4:30 0 1/9/2006 6:30 0 1/9/2006 8:30 1 1/9/2006 10:30 1 1/9/2006 12:30 1 1/9/2006 14:30 1 1/9/2006 16:30 1 1/9/2006 18:30 1 1/9/2006 20:30 NDA 1/9/2006 22:30 1 1/10/2006 2:30 0 1/10/2006 4:30 1 1/10/2006 6:30 0 1/10/2006 9:15 1 1/10/2006 11:15 1 1/10/2006 13:15 1 1/10/2006 15:15 1 1/10/2006 17:15 NDA 1/10/2006 19:15 1 1/10/2006 21:15 1 1/10/2006 23:30 1 1/11/2006 4:30 1 1/11/2006 6:30 0 1/11/2006 8:30 1 O2 % 20.85 20.83 20.86 20.84 20.74 20.84 20.85 20.84 20.85 20.75 20.69 20.82 20.85 20.89 20.88 20.87 20.88 20.88 20.88 20.89 20.83 20.87 20.85 20.84 20.87 20.84 20.85 20.86 20.86 20.85 20.83 20.84 20.84 20.84 Appendix E - Page 5 of 10 N2 % 78.10 78.00 78.10 78.11 77.74 78.10 78.09 78.05 78.08 77.80 77.60 78.08 78.09 78.07 78.08 78.08 78.08 78.08 78.07 78.07 77.89 78.11 78.04 78.12 78.09 78.12 78.11 78.10 78.09 78.06 78.12 78.11 78.10 78.11 GAS CONCENTRATIONS CH4 CO CO2 C2H2 C2H4 C2H6 Ar % ppm % ppm ppm ppm % 0.08 1 0.04 0 0 0 0.93 0.13 1 0.04 0 0 0 0.93 0.08 1 0.04 0 0 0 0.93 0.08 1 0.05 0 0 0 0.93 0.08 1 0.04 0 0 0 0.93 0.09 1 0.04 0 0 0 0.93 0.09 1 0.04 0 0 0 0.93 0.13 1 0.04 0 0 0 0.93 0.10 0 0.04 0 0 0 0.93 0.17 1 0.04 0 0 0 0.93 0.16 0 0.03 0 0 0 0.93 0.13 1 0.04 0 0 4 0.93 0.09 0 0.04 0 0 0 0.93 0.08 1 0.04 0 0 0 0.93 0.08 2 0.03 0 0 0 0.93 0.08 1 0.03 0 0 0 0.93 0.08 1 0.03 0 0 0 0.93 0.03 0 0 0 0.93 0.07 NDA 0.09 1 0.04 0 0 0 0.93 0.07 1 0.03 0 0 0 0.93 0.07 2 0.04 0 0 0 0.93 0.07 2 0.02 0 0 0 0.93 0.14 2 0.04 0 0 0 0.93 0.06 1 0.04 0 0 0 0.93 0.07 1 0.04 0 0 0 0.93 0.07 1 0.04 0 0 0 0.93 0.07 1 0.04 0 0 0 0.93 0.07 1 0.04 0 0 0 0.93 0.08 1 0.04 0 0 0 0.93 0.07 1 0.09 0 0 0 0.93 0.08 1 0.04 0 0 0 0.93 0.08 1 0.04 0 0 0 0.93 0.08 1 0.04 0 0 0 0.93 0.08 1 0.04 0 0 0 0.93 Page 5 of 6 Appendix E - Gas Chromatograph Analysis Results No. 1 Drift Opening ICG, Inc. Sago Mine 46-08791 No. 1 Drift Opening Date H2 and Time ppm 1/11/2006 10:30 0 1/11/2006 12:30 1 1/11/2006 14:30 1 1/11/2006 16:30 1 1/11/2006 18:30 1 1/11/2006 20:30 1 1/11/2006 22:30 2 1/12/2006 2:15 1 1/12/2006 9:30 1 1/12/2006 11:30 1 1/12/2006 13:30 1 1/12/2006 15:30 1 1/12/2006 17:30 1 1/12/2006 19:30 1 1/12/2006 21:30 1 1/12/2006 23:30 1 1/13/2006 5:30 0 1/13/2006 9:30 1 1/13/2006 11:30 0 1/13/2006 13:30 0 1/13/2006 15:30 1 1/13/2006 17:30 0 1/13/2006 21:30 0 1/13/2006 23:30 1 1/14/2006 4:00 0 1/14/2006 7:40 0 1/14/2006 8:15 1 1/14/2006 10:30 1 1/14/2006 12:30 1 1/14/2006 14:30 1 1/14/2006 16:30 1 1/14/2006 18:30 1 1/14/2006 20:30 0 1/15/2006 9:30 1 O2 % 20.90 20.86 20.86 20.86 20.88 20.87 20.86 20.86 20.86 20.86 20.85 20.86 20.86 20.86 20.86 20.86 20.84 20.86 20.85 20.85 20.86 20.87 20.85 20.86 20.85 20.80 20.86 20.88 20.89 20.90 20.91 20.91 20.91 20.86 Appendix E - Page 6 of 10 N2 % 78.05 78.09 78.09 78.09 78.08 78.08 78.09 78.10 78.10 78.14 78.10 78.09 78.09 78.09 78.09 78.09 78.09 78.08 78.10 78.09 78.08 78.08 78.08 78.09 78.09 77.90 78.08 78.07 78.06 78.05 78.06 78.05 78.05 78.12 GAS CONCENTRATIONS CH4 CO CO2 C2H2 C2H4 C2H6 Ar % ppm % ppm ppm ppm % 0.07 1 0.04 0 0 0 0.93 0.08 1 0.04 0 0 0 0.93 0.08 1 0.04 0 0 0 0.93 0.08 0 0.04 0 0 0 0.93 0.07 1 0.04 0 0 0 0.93 0.08 0 0.04 0 0 0 0.93 0.07 1 0.04 0 0 0 0.93 0.06 1 0.04 0 0 0 0.93 0.07 1 0.04 0 0 0 0.93 0.07 2 0.05 0 0 0 0.93 0.07 0 0.05 0 0 0 0.93 0.07 1 0.04 0 0 0 0.93 0.08 1 0.04 0 0 0 0.93 0.08 1 0.04 0 0 0 0.93 0.07 0 0.04 0 0 0 0.93 0.08 1 0.04 0 0 0 0.93 0.09 1 0.05 0 0 0 0.93 0.08 1 0.05 0 0 0 0.93 0.08 1 0.05 0 0 0 0.93 0.08 0 0.04 0 0 0 0.93 0.09 1 0.04 0 0 0 0.93 0.09 1 0.04 0 0 0 0.93 0.09 0 0.05 0 0 0 0.93 0.09 0 0.03 0 0 0 0.93 0.09 1 0.04 0 0 0 0.93 0.37 0 0.01 0 0 0 0.93 0.09 0 0.04 0 0 0 0.93 0.08 1 0.04 0 0 0 0.93 0.08 0 0.04 0 0 0 0.93 0.07 1 0.04 0 0 0 0.93 0.07 0 0.03 0 0 0 0.93 0.07 1 0.04 0 0 0 0.93 0.07 0 0.04 0 0 0 0.93 0.07 1 0.03 0 0 0 0.93 Page 6 of 6 Appendix E - Gas Chromatograph Analysis Results Borehole No. 1 ICG, Inc. Sago Mine 46-08791 Borehole No.1 Date and Time H2 ppm O2 % N2 % GAS CONCENTRATIONS CH4 CO CO2 C2H2 % ppm % ppm C2H4 ppm C2H6 ppm Ar % 1/3/2006 5:53 1045 20.45 78.04 0.23 1052 0.14 16 0 0 0.93 1/3/2006 6:55 963 20.45 78.08 0.21 914 0.14 20 30 7 0.93 1/3/2006 10:45 713 20.66 78.04 0.16 508 0.09 10 14 4 0.93 1/3/2006 11:15 450 20.68 78.04 0.16 491 0.09 9 15 5 0.93 1/3/2006 12:45 377 20.70 78.05 0.15 411 0.08 11 11 4 0.93 1/3/2006 13:15 369 20.72 78.05 0.15 394 0.08 5 0 0 0.93 1/3/2006 14:30 339 20.73 78.05 0.15 341 0.07 11 13 5 0.93 1/3/2006 15:30 279 20.74 78.04 0.16 337 0.07 10 10 5 0.93 1/3/2006 16:33 274 20.73 78.07 0.15 305 0.07 8 0 0 0.93 1/3/2006 17:30 244 20.77 78.03 0.15 289 0.07 0 0 0 0.93 1/3/2006 19:40 334 20.77 78.02 0.15 254 0.06 14 0 0 0.93 1/3/2006 21:30 136 20.80 78.04 0.14 205 0.06 0 0 0 0.93 1/4/2006 8:30 39 20.82 78.02 0.16 125 0.05 8 4 0 0.93 1/4/2006 11:30 55 20.78 78.07 0.15 63 0.05 0 0 0 0.93 1/4/2006 15:30 46 20.79 78.06 0.17 55 0.05 0 0 0 0.93 1/5/2006 8:50 14 20.82 78.04 0.16 17 0.04 0 0 0 0.93 1/5/2006 8:55 14 20.80 78.07 0.15 18 0.04 0 0 0 0.93 1/5/2006 13:10 10 20.83 78.05 0.15 11 0.04 0 0 0 0.93 1/5/2006 13:50 10 20.83 78.05 0.15 11 0.04 0 0 0 0.93 1/5/2006 16:10 8 20.82 78.05 0.15 10 0.04 0 0 0 0.93 1/5/2006 16:15 9 20.79 77.88 0.15 9 0.04 0 0 0 0.93 1/5/2006 19:00 7 20.83 78.04 0.16 8 0.04 0 0 0 0.93 1/5/2006 22:00 6 20.83 78.05 0.15 6 0.04 0 0 0 0.93 1/6/2006 2:00 5 20.82 78.05 0.16 5 0.04 0 0 0 0.93 1/6/2006 5:00 4 20.82 78.05 0.16 4 0.04 0 0 0 0.93 1/6/2006 8:00 3 20.84 78.04 0.15 4 0.04 0 0 0 0.93 1/6/2006 11:00 3 20.84 78.04 0.15 3 0.04 0 0 0 0.93 1/6/2006 14:00 2 20.84 78.04 0.14 2 0.04 0 0 0 0.93 1/6/2006 17:00 1/6/2006 20:00 1/6/2006 23:00 2 2 2 20.84 20.84 20.84 78.04 78.04 78.05 0.15 0.15 0.15 2 1 2 0.04 0.04 0.04 0 0 0 0 0 0 0 0 0 0.93 0.93 0.93 1/7/2006 3:00 2 20.82 78.07 0.15 2 0.04 0 0 0 0.93 1/7/2006 8:00 1 20.84 78.05 0.15 1 0.04 0 0 0 0.93 1/7/2006 11:00 1 20.84 78.04 0.15 1 0.04 0 0 0 0.93 Appendix E - Page 7 of 10 Page 1 of 4 Appendix E - Gas Chromatograph Analysis Results Borehole No. 1 ICG, Inc. Sago Mine 46-08791 Borehole No.1 Date and Time H2 ppm O2 % N2 % GAS CONCENTRATIONS CH4 CO CO2 C2H2 % ppm % ppm C2H4 ppm C2H6 ppm Ar % 1/7/2006 14:00 2 20.83 78.05 0.15 1 0.04 0 0 0 0.93 1/7/2006 18:00 1 20.83 78.05 0.15 2 0.04 0 0 0 0.93 1/7/2006 22:00 1 20.83 78.04 0.15 1 0.04 0 0 0 0.93 1/8/2006 1:00 1 20.83 78.06 0.14 1 0.04 0 0 0 0.93 1/8/2006 8:30 1 20.84 78.05 0.14 1 0.04 0 0 0 0.93 1/8/2006 11:00 1 20.81 78.04 0.17 1 0.04 0 0 0 0.93 1/9/2006 8:00 1 20.82 78.02 0.18 1 0.04 0 0 0 0.93 1/9/2006 11:00 1 20.85 78.03 0.15 2 0.03 0 0 0 0.93 1/9/2006 14:00 1 20.85 78.04 0.15 2 0.03 0 0 0 0.93 1/9/2006 17:30 1 20.86 78.04 0.14 2 0.03 0 0 0 0.93 1/9/2006 20:30 1 20.86 78.04 0.14 1 0.03 0 0 0 0.93 1/10/2006 4:00 1 20.86 78.06 0.12 1 0.02 0 0 0 0.93 1/10/2006 8:17 1 20.81 78.09 0.12 1 0.04 0 0 0 0.93 1/10/2006 11:00 1 20.85 78.10 0.12 2 0.04 0 0 0 0.93 1/10/2006 14:15 1 20.82 78.08 0.13 1 0.04 0 0 0 0.93 1/10/2006 19:00 1 20.84 78.06 0.13 2 0.04 0 0 0 0.93 1/11/2006 4:00 1 20.82 78.07 0.14 2 0.04 0 0 0 0.93 1/11/2006 13:00 1 20.85 78.04 0.14 1 0.04 0 0 0 0.93 1/11/2006 16:15 1 20.84 78.05 0.14 1 0.04 0 0 0 0.93 1/11/2006 18:35 1 20.85 78.05 0.13 1 0.03 0 0 0 0.93 1/11/2006 20:50 1 20.86 78.08 0.13 1 0.00 0 0 0 0.93 1/11/2006 22:35 1 20.85 78.04 0.14 2 0.04 0 0 0 0.93 1/12/2006 0:35 1 20.82 78.06 0.15 1 0.04 0 0 0 0.93 1/12/2006 2:35 1 20.82 78.06 0.15 2 0.04 0 0 0 0.93 1/12/2006 4:30 1 20.83 78.05 0.16 2 0.04 0 0 0 0.93 1/12/2006 6:30 1 20.83 78.05 0.15 1 0.04 0 0 0 0.93 1/12/2006 8:32 1 20.83 78.05 0.15 1 0.04 0 0 0 0.93 1/12/2006 10:45 1 20.85 78.03 0.15 3 0.04 0 0 0 0.93 1/12/2006 12:29 1 20.81 78.06 0.15 1 0.04 0 0 0 0.93 1/12/2006 14:32 1 20.82 78.05 0.16 3 0.04 0 0 0 0.93 1/12/2006 16:25 1 20.82 78.05 0.16 1 0.04 0 0 0 0.93 1/12/2006 18:20 1 20.83 78.02 0.18 1 0.04 0 0 0 0.93 1/12/2006 20:20 1 20.83 78.03 0.17 1 0.04 0 0 0 0.93 1/12/2006 22:20 1 20.84 78.02 0.17 1 0.04 0 0 0 0.93 Appendix E - Page 8 of 10 Page 2 of 4 Appendix E - Gas Chromatograph Analysis Results Borehole No. 1 ICG, Inc. Sago Mine 46-08791 Borehole No.1 Date and Time H2 ppm O2 % N2 % GAS CONCENTRATIONS CH4 CO CO2 C2H2 % ppm % ppm C2H4 ppm C2H6 ppm Ar % 1/13/2006 0:35 1 20.81 78.05 0.16 1 0.04 0 0 0 0.93 1/13/2006 2:40 1 20.77 77.90 0.26 1 0.04 0 0 0 0.93 1/13/2006 4:34 1 20.82 78.04 0.16 1 0.05 0 0 0 0.93 1/13/2006 6:40 1 20.82 78.05 0.16 1 0.04 0 0 0 0.93 1/13/2006 8:38 1 20.85 78.01 0.17 1 0.04 0 0 0 0.93 1/13/2006 10:35 1 20.82 78.04 0.18 2 0.04 0 0 0 0.93 1/13/2006 12:32 1 20.79 78.04 0.20 1 0.04 0 0 0 0.93 1/13/2006 14:30 1 20.83 78.02 0.18 1 0.04 0 0 4 0.93 1/13/2006 16:30 1 20.82 78.02 0.19 2 0.04 0 0 0 0.93 1/13/2006 18:30 1 20.83 78.01 0.20 1 0.04 0 0 0 0.93 1/13/2006 20:30 1 20.86 78.09 0.08 1 0.04 0 0 0 0.93 1/13/2006 22:30 0 20.88 78.09 0.06 4 0.04 0 0 0 0.93 1/14/2006 0:30 0 20.85 78.12 0.05 0 0.04 0 0 0 0.93 1/14/2006 2:30 0 20.86 78.11 0.06 0 0.04 0 0 0 0.93 1/14/2006 4:30 0 20.86 78.11 0.06 1 0.04 0 0 0 0.93 1/14/2006 6:30 1 20.82 78.03 0.18 1 0.04 0 0 0 0.93 1/14/2006 8:30 1 20.82 78.02 0.19 1 0.04 0 0 0 0.93 1/14/2006 10:27 1 20.85 78.02 0.16 1 0.04 0 0 0 0.93 1/14/2006 12:31 1 20.87 77.99 0.17 2 0.04 0 0 0 0.93 1/14/2006 14:37 1 20.87 78.00 0.19 1 0.04 0 0 0 0.93 1/14/2006 16:29 1 20.88 77.98 0.17 1 0.04 0 0 0 0.93 1/14/2006 18:35 0 20.93 78.07 0.04 0 0.04 0 0 0 0.93 1/14/2006 20:28 0 20.93 78.07 0.03 0 0.04 0 0 0 0.93 1/14/2006 22:28 0 20.91 78.07 0.05 0 0.04 0 0 0 0.93 1/15/2006 0:31 1 20.95 78.04 0.04 0 0.04 0 0 0 0.93 1/15/2006 2:33 0 20.88 78.02 0.14 1 0.04 0 0 0 0.93 1/15/2006 4:37 1 20.91 78.09 0.04 0 0.04 0 0 0 0.93 1/15/2006 6:34 1 20.87 77.96 0.20 0 0.04 0 0 0 0.93 1/15/2006 8:31 1 20.88 78.00 0.15 1 0.04 0 0 0 0.93 1/15/2006 10:34 1 20.83 78.05 0.15 1 0.04 0 0 0 0.93 1/15/2006 12:33 1 20.84 78.03 0.16 1 0.04 0 0 0 0.93 1/15/2006 14:49 1 20.85 78.06 0.16 1 0.04 0 0 0 0.93 1/15/2006 16:34 1 20.84 78.03 0.16 1 0.04 0 0 0 0.93 1/15/2006 18:40 0 20.88 78.08 0.07 1 0.04 0 0 0 0.93 Appendix E - Page 9 of 10 Page 3 of 4 Appendix E - Gas Chromatograph Analysis Results Borehole No. 1 ICG, Inc. Sago Mine 46-08791 Borehole No.1 Date and Time H2 ppm O2 % N2 % GAS CONCENTRATIONS CH4 CO CO2 C2H2 % ppm % ppm C2H4 ppm C2H6 ppm Ar % 1/15/2006 20:33 1 20.90 78.09 0.05 0 0.04 0 0 0 0.93 1/15/2006 22:37 1 20.87 78.08 0.08 1 0.04 0 0 0 0.93 1/16/2006 0:40 0 20.88 78.09 0.05 1 0.04 0 0 0 0.93 1/16/2006 2:42 0 20.86 78.08 0.09 1 0.04 0 0 0 0.93 1/16/2006 4:35 1 20.77 77.94 0.31 1 0.04 0 0 4 0.93 1/16/2006 6:40 1 20.82 78.03 0.16 1 0.04 0 0 0 0.93 1/16/2006 8:42 1 20.83 78.03 0.17 1 0.04 0 0 0 0.93 1/16/2006 10:38 1 20.83 78.01 0.18 2 0.04 0 0 4 0.93 1/16/2006 12:40 1 20.81 78.05 0.17 2 0.04 0 0 3 0.93 1/16/2006 14:31 1 20.82 78.05 0.17 1 0.04 0 0 0 0.93 1/16/2006 16:40 1 20.81 78.04 0.17 1 0.04 0 0 0 0.93 1/16/2006 18:30 1 20.82 78.04 0.17 1 0.04 0 0 0 0.93 1/16/2006 20:30 1 20.83 78.04 0.16 1 0.04 0 0 0 0.93 1/16/2006 22:28 1 20.79 77.96 0.28 1 0.04 0 0 4 0.93 1/17/2006 0:35 1 20.81 78.04 0.18 1 0.04 0 0 0 0.93 1/17/2006 2:33 1 20.80 78.05 0.18 1 0.04 0 0 0 0.93 1/17/2006 4:39 1 20.78 78.02 0.23 2 0.04 0 0 0 0.93 1/17/2006 6:35 2 20.80 78.05 0.17 1 0.04 0 0 0 0.93 1/17/2006 8:40 1 20.84 78.01 0.18 1 0.04 0 0 0 0.93 1/17/2006 10:45 1 20.85 78.00 0.18 1 0.04 0 0 0 0.93 1/17/2006 12:40 2 20.84 78.00 0.19 1 0.04 0 0 0 0.93 1/17/2006 14:40 1 20.84 78.00 0.20 2 0.04 0 0 0 0.93 1/17/2006 16:35 1 20.84 77.99 0.19 1 0.04 0 0 0 0.93 1/17/2006 18:40 1 20.84 77.99 0.19 1 0.04 0 0 0 0.93 1/17/2006 20:37 1 20.84 78.00 0.19 1 0.04 0 0 0 0.93 1/17/2006 22:23 1 20.83 78.00 0.19 1 0.04 0 0 0 0.93 1/18/2006 0:35 1 20.82 78.01 0.19 1 0.04 0 0 0 0.93 1/18/2006 2:35 1 20.82 78.01 0.20 1 0.04 0 0 0 0.93 1/18/2006 4:35 1 20.83 78.01 0.18 1 0.04 0 0 0 0.93 1/18/2006 6:35 1 20.84 77.99 0.19 1 0.04 0 0 0 0.93 1/18/2006 8:42 1 20.83 78.01 0.19 1 0.04 0 0 0 0.93 1/18/2006 10:26 1 20.82 78.01 0.20 0 0.04 0 0 0 0.93 1/18/2006 12:25 1 20.82 78.03 0.18 1 0.04 0 0 0 0.93 Appendix E - Page 10 of 10 Page 4 of 4 Appendix - Accident Investigation Data - Victim Information Accident Investigation Data - Victim Information U.S. Department of Labor Event NumberMine Safety and Health Administration Victim Information: 1 1. Name of Employee: 2. Sex 3. Victim's Age 4. Last Four Digits of SSN: 5. Degree of Injury: Teny Helms 50 I 01' Fatal 6. DatetluttuliDDt?l'Y) and Time-{24 Hr.) or Dearth: r. Date and Time Started: a. Date: armada ta. Time' 11:00 a. Date: 01mm b. Time: 0:00 a. Regular Job Title: 0. Work Activity men Injured: 10. Was this work adivity part of regular job? 1195 Preehiit?er 092 walking yes I IMO I I 1 ?rm? Years Weeks Days bl Regular Years Weeks Days a This Years Weeks Days d. Tom Years Weeks Days Wont mutiny; ti 0 0 Job Title: 0 26 0 Mine: 0 25 0 Mining: 23 a 13. Nature ol' unmet Iilness: 110 carbon monoxide intoxication 12. Directly Inllicted Injury or Illness? 023 carbon monoxide gas from an explosion 14. Training De?ciencies: Hazard: I I Nefviijeu?y-Employed Experienced ?ner: I I Annual:I I Task: I I 1 5. Company of Employmenttif diltererlt from production operator] Operator Independent Contractor ID: (if applicable) 1e. Dn-site Emergency MediceiTreetrnent: I I First-Aid: I CPR: I I EMT: I I Medical Professional: I I None: I 17. Part 50 Document CONN Number: (form 18. Union Al?liation of Vlctim: 999-9 None {No Union Af?liation) Victim Information: 2 1. Name or Injurediill Employee: 2. Sex 3. Victim's Age 4. Last Four Digits of SSN: 5. Degree of Injury: Jackie L. Weaver 51 01? Fatal 3. DatetltIIVDDn?Yj and Time{24 Hr.) Of Death: 7. Date and Time Started: a. Date: almond a Tim: 11:00 3- 93?: an?? 55?? 0. Regular Job Title: 0. Work Activity Mien Injured: 10. Was this work activity part of regular job? on: Eiectrican are revenue we assignment I I "a I I 1; "?r?mw Years Weeks Days b. Regular Years Weeks Days I: This Years Week Days '1 Total Years Week: Day: Work Activity: 26 0 0 Job T1153: 25 0 0 Mine: 2 t} Mining: 25 a 13.Nature ct Injury or Illness: 110 carbon monoxide intcixication 12. What Directly lnilicted injuny or Illness??I 02.? carbon monoxide expiosion 14. Training De?dendes: Hazard; I I NewINewty-Employed Experienced Miner. I I 15. Company of Employment: (it different from production operator) Operator 1e. Dn-site Emergency Medical Treatment: Annual: I I Task: Independent Contractor ID: (If applicable) Not Applicable:_I First-Aid: I I CPR: EMT: I I Medical Professional: I None: I I 1T.Part 5U Document Control Numben (form TOGO-1) 15? Union Af?liation ufwc?m: 9999 ?one (No Union A?m??am Victim Infon'nation: 3 1. Name of Employee: 2. Salt 3. Victim's Age 4. Last Four Dig its of SSN: 5. Degree of Injury: James A. Bennett 51? I or Fatal 6. and Time(24 Hr.) Of Death: Date and Time Started: e. Date: 01102;?005 tr. Time: 17:00 a. Date: 0mm b.1'ime: 0:00 a. Regular Job Title: 9. Work when injured: 10. Was this week activity part of regular job? use Shuf?e Ceererator 07" TEN-imam Yes I Ian I I Ewen?: Years Weeks De Years Weeks .Years Week Years Weeks Do a. nus b. Regular c: This an d. Total Work Activity: 23 0 Job Title: 23 0 0 Mine: 0 20' 0 Mining: 25 a 12. What Directly Intlicted Injury or Illness? 13. Nature of Injury or Illness: 023 carbon monoxide from expiosbn 100 carbon monoxide minimisation 14. Training De?clellcl' "es: Hmm; I I vaiNewty?Employed Experienced Liner. I I Annual: I I Task: I I 15.Comparly of different from production operator) om Independent Contractor ID: {if applicath to. Dn-slte Emergency Medical Treatment: Not Applicable: I First-Aid: I I CPR: I I EMT: I I Medical Professional: I I None: I I Part 50 Document Control Number. {form 7000-1) 16. Union Af?liation of Victim: 9999 None (No Union Af?tiatrbn) MEI-IA Form Dec 1994 Printed UZi14i200? 10:15:18 AM Appendix - Page 1 of 5 APP Accrdent Investigation Data - Victim Information EventNumber:I4IiI3I4I4I1I4I Victim Information: ll endix - Accident Investigation Data - Victim Information U.S. Department of Labor Mine Safety and Health Administration 1. Name of Employee: 2. Sex Atva M. Bennett 3. Victim's Age 51 4. Last Four Digits of SSN: 5. Degree oi Injury: 01? Faint 7. Date and Time Started: a. Date: 000M006 a. Time: 6:00 6. and Time{24 Hr.) Of Dealn: a. Date: 01/022006 bfnme: 1?:00 3. Regular Job Title: 9. Work Activity when Injured: 10. Was this work eulvity part of regular job? ?Continuous Miner Operator 975 TMWIINU Ia Yes I I I Exhi?moe Years Weeks Days h. Regular Years Weeks Days c: This Years Weeks Days 6- Total Years Weeks Days Work Activity: 25 0 0 Job Title: 25 a 0 Mine: 2 26 0 Mining: 29 a 12. What Directly Intlicted Injury or Illness? 13. Nature of Injury or Illness: 023 carbon monoxide [cm explosion 110 carbon monoxide intoxication 14. Training De?ciencies: Hmm; I I Newaewly?Employad Experienced miner: I I Annual: I I Task: I I 15. Company at different from production operator) 0mm, Independent Centractor ID: (if applicable} 16. On-slte Emergency Medical Treatment: 5 I Firsthid; I can: I I EMT: I I Medical Professional: I I lid-1n: I I 17. Part 50 Document Control Number: Item 1000-1} 13. Union Af?liation of Victim: 9999 None (No Union Af?liation) Victim Information: 5 1. Name of Employee: 2. Sex Thomas p. Andersen 3. Victim?s Age 30 5. Degree of Injury: 01? Fatal. I 4. Last Four Digits Date and Time Started: a. Date: 011020006 and TimeI2-?I Hr.) or Death: b. Time: 17:00 b. Time: 6:00 a. Date: OTMMDOG 3. Regular Job Title: 9. Work Activity Wit!" Iniumdi 1 D. Was this werk activity part of regular job? can Shuttle oar Operator 07? '0 ?mnm I I "0 I I 11. Experience: a. This Years Weeks Days Raul? Years Weeks Days c7111? Years Week Days {1 TON Years Weeks Days Work Activity: 2 a Job Title: 2 Mine: 0 to 0 Mining: 10 a a 13.Nature ot Injury or Illness: 110 carbon monoxide intoxication 12. What Directly ln?icted Injury or Illness? 023 carbon monoxide li'om explication 14. Training De?ciencies: Hazard: I NewiNeMy-Employad Experienced Miner: I I 15. Company of Employment: {If different from production operaer Annual:I I Task: I I Independent ID: (if applicable) Operator to. Dn?slte Emergency Medical Treatment: Not Applicable: I I First-Aid: I can: I I EMT: I I I hiedical Professional: I None: I 11PM 5'3 '3':me Control Humbert {form la. Union Al?llation of Victim: 9999 None {No Union Victim Information: 5 1. Name of Employee: 2. Sex 3. Victim's Age it. Last Four Digits of 38 N: 5. Degree of Injury. Martin Toier Jr. ill 51' 01 Fetal 6. DatetMMiDDN?t) and Time{24 Hr.) Of Death: 7. Date and Time Started: a. Date:01r02r2006 D.Time: 11:00 a. Date: b. Time: 0:00 8. Regular Job Title: 9. Work Activity when Injured: 10. Was this work activity part of regular job? 949 m" 016 Traveling to workth Yes I IMO I I 11? Experience Year: Weeks Days Years Weeks Years Week Days Years Weeks I This ti. Regular c: This d. Total Work Activity: 25 0 0 Job Title: 25 0 0 Mine: 0 14 0 Mining: 32 0 0 12. What Directly Innkited Injury or Illness? 13_ "mum of ?jury ?mass; 023 carbon monoxide from explosicln to Carbon monoxide intoxication 14. Training Deficiencies: Hazard; I I NelHiNeMy-Employed Experienced Minen I I Arvlualz I I Task: I I 15.DDmpany of Employmenqu different from production operator] Operator 16. Dn-e'lte Emergency Medical Treatment: Not Applicable: I First-Aid: I 17. Part 50 Dowment Control Number: {form round) Independent Contractor ID: (If applicable) EMT: I I Medical Professional: I I None: I 18. Union Attitiation None {No Union Af?iction) 9999 MSHA Form Dec. 1994 Printed 0314i200? 10:15:19 AM Appendix - Page 2 of 5 r1 Appendix - Accident Investigation Data - Victim Information Accident Investigation Data - Victim Information U.S. Department of Labor Event NumberMine Safety and Health Administration Victim Information: 1" 1. Name of Employee: 2. Sex 3. Victim's Age 4. Last Four Digits of SSN: 5. Degree of injury: Fred G. Ware 58 01 Fate! 5. and 'Iimei24 Death: Date and Time Started: a. Date: 011022006 am: 17.00 a. Date: 011022006 0. Tints: 6:00 a. Regular Job Title: 9. negativity When Injured: 10. Watt this activity part or regular job? A Cpntinuous Miner Operator 975 f0 Yes I It?, I I 1 :f?em Years Weeks Days ?1 Regular Years Weeks Days c: This Years Weeks Days {1 Total Years Weeks Days Work Activity: 15 0 0 Job Title: 15 0 0 Mine: 1 36 0 Mining: 37 a 12. What Directly Initiated Injury or Illness? 13. Nature of lottery or Illness: 023 Carbon monoxide non: explosion 1'10 Carbon monoxide intoxication 14. Training De?ciencies: Hazard; I I NewiNawa-Empioyed Experienced Miner: I I Amuai: I I Task: I I 15. Company of Employment:{tf different from production operator} 0mm, Independent Connector ID: {if applicable) i?GTD?anite-Emergency Medical Treatment: Not Applicable: I First-Aid: I CPR: I I Eittli?: I I Medical Professional: I I None: I I 17. Part 50 Document Control Number. [form moo-1} 1 8. Union Af?liation of Victim: 9999 None (No Union Armando) Victim Information: 8 1. Name of Injurediill Employee: 2. Sex 3. Victim's Age 4. Last Four Digits of SSN: 5. Degree oi Injury: Jesse Jones 44 I or Fatai e. DateiMIliliDDiYY} and Hr.) or Deali'l: Date and Time Started: a Data. 01mm hm: 13.90 a. Date: 01mm b. Time: 6:00 a. Regular Job Title: 9. Work Activity wt'tert Injured: 10. Was this we; actiy-ity?part of regular job? .945 075 Travelian Messianmeni Yea I I No I I 2'er Years Weeks Days b. new,? Years Weeks Days I: Tim Years Week Days EL Tm. Years Weeks Days Work Activity: 14 a Job Title: 14 0 Mine: 0 as 0 Mining: 15 a a Directly Indicted Injury or Illness? 13.Natura of Injury or Illness: 523 Carbon monoxide from extrication 110 Carbon monoxrde intoxication 14. Training Deficiencies: Hazard; I I NewiNewiy?Employed Experienced Miner: I I Annual: I I Task: I I 15. Company of Employment: {It dttl'erent tron-t production operator) 0mm Independent Contractor ID: applicable} to. Dn-aite Emergency Medical Treatment: N?applicable: I I First-Nd: I I CPR: I I EMT: I I Medical Professional: I I None: I I 11PM 50 Document Control Number: lion? 7000-1} 13. Union Affiliation orvtotim: 9999 None (No Union Af?iction] Victim lnfonnation: 9 1. Name of Employee: 2. Sex 3. Victim's Age 4. Last Four Digits of SSN: 5. Degree of Injury: Marsheii Winona 5'0 01' Forget a. Determwoorm and Time{24 Hr.) or Death: 7. Date and Time Started: a. Date: OTMMOOE b.Tr'me: 1?:00 a. Date: 0110242000 tt.'i'ime: 0:00 3. Regular Job True: 9. Work Activity when Injured: 10. Was this wotk'ectiuity part of regular Ion? 028 Scoop Operator 93" Toronto ?0 Yes I Inc I I u-?xizenmm- Years Weeks Days b. Regular Years Weeks Days c: Thu Years Week Days 6- TOM Years Weeks Days Work Activity: 5 0 0 Job Title: 5 t} Mine: 1 8 0 Mining: 23 a a 12. What Directty In?icted Injury or Illness? 1a, Nam", of Iriqu or "mm; 023__ __Ca_;rbog_ monoxide from expiosion 1'10 Carbon monoxide intoxication 14. Training De?ciencies: Hazard; I I NBWiNBMy-Empinyad Experienced Miner. I I Annual: I I T3511: I I 15.Company of Employment?t onetime production operator) 0mm. Independent Contractor ID: {it applicable] 15. Dn?site Emergency Medical Treatment: Not Appw; I I First-Aid: I I CPR: I I EMT: I I Medical Professional: I I None: I I 17. Part 50 Document Control Number: (form Toad-1} 15. Union Af?liation of Victim: 9999 None {No Union intonation) MSHA Form Dec 1994 Printed 02i?i4t?200? 10:15:20 AM Appendix - Page 3 of 5 Appendix - Accident Investigation Data - Victim Information Accident Investigation Data - Victim Information U.S. Department of Labor Event NumberMine Safety and Health Administration Victim Information: 10 1. Name of Employee: 2. Sex 3. Victim?s Age 4. Last Four Digits of SSN: 5. Degree of Injury: David W. Lewis 28 or Fatal Date and Tlme Started: a. Date: 01.02.0006 h.Trrne: 6:00 6. DateIMilrIr'DDN?r?j and Tlmetzit Hr.) 01 Death: a. Date: 011022006 b. Time: 1?:00 9. Work Aolivity when Injured: 10. Was this and: activity part of regular Iob? 6. Regular Job Title: 94y- Roof?Buffer am?, on Yes I Iran I I x?mw Years Weeks Days b. Rag-"m Years Weeks Days a This Years Weeks Days 0.1.0131 Years Weeks Days Work Activity: 1 0 Job True: 1 0 Mine: 1' 32 Mining: 1 32 a What?Directly Inflicted Injury or Illness? 13. Nature of Injury or Illness: 023 Carbon menolldde [ringleer 119 Carbon monoxide intoxication 14. Training De?ciencies: Hmm; I I Nethewty-Employed Experienced Miner. I I Annual: I I Task: I I 15. Company of different from production operator) Independent Contractor ID: [If applicable] 16. Dn-site Emergency Medical Treatment: Not Applicable: I First-Nd: I can: I I EMT: I Medical Professional: None: I I 17. Part 50 Document Control Number: (form 7000-1} 18. Union Af?liation of 1liltinlin'l: 9999 None (No Union A?ltta?onj Victim Information: 11 1. Name of Employee: 2. Sex 3. Victim's Age 4. Last Four Digits oi SSH: 5. Degree of Injury: L. Grates 56 or Feral 6. and Time{24 Hr.) Of Death: T. Date and Tlme Started: a. Date: moments is. T'Inle: tron 9- Dare: 010M005 5- TM: 5500 3. Regular Job 9. Work Activity Vitth injured: ?1 Was this my? part of mama'- iob? 04.? Rear sans:- 075 Trawlingto work assignment Yes I IMEI I I 11. Experience: a. This Years Weeks Days bl Regular Years Weeks Days c: This Years Week Days d, Total Years Weeks Days Work Activity: so a 0 Job Title: 20 0 Mine: 1 0 Mining: 25 a a 13.Nature of Injury or Illness: 110 Carbon monoxide intoxication 12. What Directly lnllictecl Injury or Illness? 023 Carbon monoxide from explosion 14. Training De?ciencies: Hmm; I I "w-hl?ewrFNewa-Employed Experienced Miner: I I 15. Company of Employment: {If different from produdion operator) Operator 13. (in-site Emergency Medical Treatment Annual: I I Task: Independent Contractor ID: (If applicable} Not Applicable: I First-Aid: I CPR: EMT: I I Medical Professional: I I None: I I 5? Dimmer? 30W Number: 15. Union Af?liation or Victim: 9999 None {No union antisean Victim Information: 12 1. Name of Employee: 2. Sex 3. Victim's Age 4. Last Four Digits of SSN: 5. Degree of Injury. Genres J. Hamner 54 or Fetal, . 6. and Hr.) Of Death: Date and Time Started: a. Date: 01.022006 b. Time: 1100 a. Date: b. Tune: 6:00 8. Regular Job Title: a. Work Activity when Injured: 10. Was this work activity port of regular job? 550 stem Car 0mm.- OYE Trsyetingtotvorksestmment Yes I INO I I Years Weeks Days Regular Years Weeks Days a Tm Years Week Days d, Total Years Weeks Days Work Activity: 13 CI IJ Job Title: 13 0 Mine: 1 26 0 Mining: 26 0 12. What Directly Injury or lHness'r 023 Carbon wide from egpi?osion 13. Nature oI Injury or Illness: 10 Carbon monoxide intoxication Annual: I I 14. Training De?ciencies: Hazard: I I NewMeMy?Employed Experienced Miner. I I t5.Compeny of Employment-{If diIferent Irorn production operator} Operator Task: I I Independent Contractor ID: (if applicable) ls. on-site meanness Treatment: Not Applicable: I I Firawtid: I I CPR: I I EMT: I I Medical Professional: I I None: I I 17. Part 50 Document CO?tl'Ci Number: {term 18. Union Affiliation of Victim: 9999 None (No Union An'r?ir'etfoni MSHA Form D301994 Printed 10:15:21 AM . 421 a - Appendix Page 4 of 5 Appendix - Accident Investigation Data - Victim Information Accident Investigation Data - Victim Information U.S. Department of Labor Event NumberMine Safety and Health Administration Victim information: 13 1. Name 2. Sex 3.Victim'aAge 4. Last Four DigitsofSSN: 5. Degreeoflnjury: 25 I oz Permanent total disability 7. Date and Time Started: a. Date: 012022098 b.T!me: 6:00 B. and TimeIZ-d Hr.) Of Death: a. Regular Job Title: n. Walk Activity when Injured: 10. Was this work activity part oi regular job? 045 900mm open,? 076' Trevol'r'ngtowork assignment I )5 IN, I I 11' Experience Years Weeks Years Weeks De Years Weeks De Years Weeks Do a. This 11. Regular c: This Total V3 Work Activity. 6' 24 0 Job Title: ti 24 0 Mine: 1 16 0 Mining: 4 12 a 12. What Directly Intlicted Injury or Illness? 13. Nature of Injury or Illness: 023 Carbon monoxide from explosion 110 Carbon monoxide poisoning 14. Training De?ciencies: Hazard; I I Matrimony-Employed Experienced Miner. I I Annual: I I Task: I I 15. Company of different from production operator) open?, Independent Contractor ID: (if applicable] 16. Dn-eite Emergency Medical Treatment: Not Applicable: Firet?Aid.? one: I I EMT: litedical Professional: I None: I I 1 Part 50 Document Control Number: (form 13. Union Attiltati'on of Victim: 9999 None (No Union Af?liation} 'v?Ictim Information: 1. Name of Injurediill Employee: 2. Sex 3. Victim's Age at. Last Four Digits ot SEN: 5. Degree of Injury: 5. and Time-(24 Hr.) Of Death: T. Date and Time Started: 5- Revular Job Title: 9- Work Activity when Injured: 10. Was this work activity part of regular job? in I [No I 11. Experience: a, This Years Weeks Days by Regular Years Week: Days 6: This Year: Week Days ?1 Total Years Weeks Days Work Activity: Job Title: Mine: Mining: 12. What Directly Indicted Injury or Illness? 13.Neture of Injury or Illness: 14. Training De?dencies: Hazard: I "WNW-Employed Experienced Miner: I I Annual: I I Task: I I 15. Company of Employment: {if diII'erent from production operator) Independent Contractor ID: {if applicable} 13. Dn?site Emergency Medical Treatment: Not?ipplicable: I First-Aid: I CPR: I I EMT: I I Professional: I I None: I I 1T.F'art 50 Document Number {531111 7000-1} 15_ union A?ilia?on ?Wm: Victim Information: 1. Name of Employee: 2. Sex 3. Victim's Age 4. Last Four Digits ot SSH: 5. Degree of Injury: 8. and Time{24 Hr.) Of Doom: 7. Date and Time Started: a. Regular Job Title: 9. Work Activity when Injured: 10. Was this work 51:21va part of regular job? Yes I 11- Experience: veers Weeks om Years Weeks Days veers Week on Years Weeks ea ,1 This In. Regular c: This ya d. Total Work Activity: Job Title: Mine: Mining: 1:2. What Directly In?icted Injury or 13. Nature 01? Injury or Illness: 14. Training De?ciencies: Hazard; I I Hewl'NewavEmphyed Experienced Miner: I I Annual: I I Took: I I 15.Company of Employmenth different from production operator} I Independent Contractor ID: (if applicable) 15. Dn-eite Emergency Medical Treatment: Not AEEncabIe; I I First-Aid: I I CPR: I I EMT: I Medical Professional: I I None: I 17. Part 50 Document Control Number. (form rode-1) 18. Union Af?liation MSHA Form Dec 1994 Printed 10:15:22 AM Appendix - Page 5 of 5 Appendix G - Lists of Individuals Who Assisted with the Investigation International Coal Group, Inc. Samuel R. Kitts John B. Stemple Charles C. Dunbar Timothy A. Martin Senior VP of Operations WV & Maryland Region Assistant Director of Safety and Employee Development General Manager, Buckhannon Division Corporate Director of Health and Safety Wolf Run Mining Company Carl L. Crumrine Jeffery Toler Burlin Wright Bradley L. Hamrick James A. Schoonover Roger D. Hendrick Vaughn Miller Kermitt Melvin Gary D. Carpenter Richard Bragg Joseph Ryan Joseph Myers Ron Helmic William Saltis Ralph Tanner John Travise Jr. Philip R. Clevenger Sago Miners Jeremy R.Toler Brian E. Curtis Chester Runyon Teddy J. Hickman Basil J. Chidester Chris Chisolm Ronnald E. Grall Joseph Runyon Roy L. Williams William L. Chisolm Charles R. Wilson Edmund B. Payne Appendix G - Page 1 of 3 Travis J. Anderson Craig D. Newson Mike W. Butcher Harold Baisden Jr. Kenneth Anderson Gary L. Marsh Roger L. Shiflet Francis Johnson Denver D. Anderson Thomas L. Everson Nathan H. Eye Darrel Lucas Appendix G - Lists of Individuals Who Assisted with the Investigation (Cont’d) United Mine Workers of America Ron Bowersox Max Kennedy Ted Hapney Dennis Bailey Gary Trout Butch Oldham Mark Cochran State of West Virginia Brian Mills Jeff Bennett Mike Rutledge Jim Hodges J.D. Higginbotham Monte Hieb John Hall Barry Fletcher Phil Atkins John Collins John Cruse Doug Conaway MSHA - Educational Field Service Preston T. White MSHA - District 4 James D. Honaker MSHA - National Mine Health and Safety Academy Donald C. Starr Theodore G. Farrish Arthur D. Wooten David S. Mandeville Harold E. Newcomb MSHA - Pittsburgh Safety and Health Technology Center Thomas A. Morley James D. Baca Kim S. Diederich Scott K. Johnson Terence M. Taylor Michael Gauna Mark A. Pompei William J. Francart Dean Skorski Appendix G - Page 2 of 3 Dennis A. Beiter Gary J. Shemon Mark E. Schroeder C.W. Moore Richard Allwes John R. Cook George N. Aul Donald A. Sulkowski William Helfich Appendix G - Lists of Individuals Who Assisted with the Investigation (Cont’d) MSHA - Approval and Certification Center Kevin L. Hedrick Robert J. Holubeck Department of Labor, Office of the Solicitor James B. Crawford Robert S. Wilson Appendix G - Page 3 of 3 Timothy S. Williams Last row of installed roof bolts Water across place (shore line) SCSR top SCSR bottom and top SCSR goggles SCSR goggles Cluster of 3 used and 2 SCSR bottoms: handwritten identification 134173, handwritten identification 129841, handwritten identification 134124, one additional glove in cluster, potential brought in by rescue teams all 3 SCSR goggles SCSR bottom SCSR bottom 1 glove SCSR goggles SCSR pouch SCSR pouch and goggles SCSR bottom CSE calibration kit (brought in by rescue team) SCSR goggles SCSR top Used SCSR 21306, handwritten identification "63022" Water line Empty crunchy bar wrapper SCSR top 8 02. water bottle, Damaged 2 claw hammers and a ball point pen SCSR top 91 . i. Ht Hard Hat Black Skinny yellow tape Randall McCloy or a Used SCSR against rib 23964, handwritten identification "64552", green, white red reflective tape on SCSR Channel locks Red handkerchief SCSR top SCSR pouch Methane detector CSE 102 large digital readout 4588 Center of pointed area on rib Curtain laying on bottom with green paint on it (small red pool in center of curtain) water jug Used SCSR A15162, handwritten identification "134632 Bucket No. 8\ Pick hammer Curtain with green paint on it, laying on mine bottom Rock hammer Thermos - No' Used SCSR 22526, handwritten identification 62974, written on front is SMAGR Hydraulic oil can AW68 Bucket No. 7 U1 Thermos (Green jacket under thermos) Bucket No, 10 ,Jr, Homner 1 LO 1 11/11 1 Used SCSR A06285, handwritten identification "123670" Bucket No. 6 Center of canvas carry bag containing 3 (brought in by rescue team) Latex rubber glove Curtain laying on mine bottom Bucket No. 5 Used A10031, handwritten identification "131939" Bucket 9 Bucket 4 Used SCSR 19109, handwritten identification "60084", empty water bottle(8oz.) underneath SCSR 60084 Cluster of tools consisting of channel locks, claw hammer, pick hammer, Used SCSR 23104. handwritten identification "54353" 2 gloves with keys attached to clip, Note: 2 keys, utility knife Empty crushed water bottle Lense from SCSR goggles Curtain laying on bottom. Green paint outlining curtain rib Used SCSR 64586, handwritten identification "2763" SCSR goggles Safety glasses, plastic Used SCSR 108166, handwritten identification "55558" 1 pair of SCSR goggles with left lense missing (laying on mine bottom) 4 1 Explorer 4 CSE 2950, 1 case of brought in by rescue team SCSR goggles to (D (.11 ill Hooded sweatshirt Bulldog Athletics (pockets empty) SCSR dust cover, broken Ball point pen Drinking water box, 8 02. bottles; 1 empty, 1 Full, 72 of a 1 liter bottle of water from GO MART. A neck strap and empty cracker paper under the box Used SCSR 105202, handwritten identification "52210" State of West Virginia emblem Bucket No. 2, Possible Fred Ware matimated curtain attachment Empty snuff can 7 Barricade curtain lnby corner of line point of curtain along rib attachment obtained from Used SCSR evidence 47986 . on ?b handwritten identification "7845" Empty sandwich bag Rubber gloves on Two SCSR lids and straps Box with 8 oz. bottled water 7 full bottles, 2 gloves Empty 8 02. water bottle Strap for carrying off a stretcher Center of head of 8 lb. sledge hammer with 1.6 ft. handle 1 curtain on floor Used SCSR 124040 handwritten identification Lancaster premium chewing tobacco on curtain on floor Barricade curtain Center of curtain laying in pile of coal in face 4?x4' looks as if it had been set on in two locations Outby corner of line curtain along rib . Cluster of brown?orange SCSR dust cover debris Roof bolt head, damaged and deformed Cable bolt head, damaged and deformed Square of curtain folded 24"x24" and 60"x96" burn sheet from first aid kit . Red paint spray can Estimated curtain attachment Estimated curtain attachment Empty water bottle Nail bucket Left comer of curtain Estimated curtain attachment Bucket No. 1, Possibly Jesse Jones Empty plastic water bottle Silica gel pack from SCSR Curtain attached to roof bolt Barricade curtain SCSR goggles (found intact) Appendix H-1 Sago Mine, MSHA ID 46-08791 Wolf Run Mining Company Mine Map No. 3 Face of 2nd Left Parallel Barricade 0? 5? 10? BE mots: Gas bubbling in water at face area Lost row of permanent ast row of bolts with 2 reflectors supports With 2 reflectors hanging from We curtain hung to face Roof Control Plan laying around in pieces face curtain Fire boss slip with notes hung; outby FJ 42?? am?) end of curtain hang on bolt ine curtain hung 7' from face Fire boss slip 4:37?) 1 flypad Note: 2 packets Of silica gel Face Curtain SCSR Top Bottom Last row of Ser. NO. 106167 permanent Check curtain (rolled up) with Goggles supports, 2 reflectors on *Notej?I last row of Curtain appeared bo ts five Gob in Face Last row of permanent supports Curtain Down Rock Dust End Of curtain Face Curtain I 33231313? power Cor Danger Sign "High Voltage? End of line curtain Last row of permanent supports Shore of Preshift 1 2 06 Fire boss tag with notes water Last row Of bolts F.) 4:33am?) Fire boss tag with notes curtain hung Water (211?2 om?) 0n End of line curtain\ _/Last row of permanent supports on ground 12-5, fr?m rib comer _/Last row of permanent supports Items on top of Power Center through CheCk End of hung curtain un throu check\_ Electrical Su lies Curtain see Append'x 9 0n roof bolter pie pan fire pp hung from boss initials 8c dates time roof bolt ~20? Piece of *Joy #1 Miner Remote on FJ 4:37 am") brattice cloth?1 with scrubber right Side #3 Fletcher Model DDO 15 - 1. Spad Gun 2. 2 FE Roof bolt, curtain hung off of it 3. Unopened SCSR #106829 on ground 82105/2004320 4. Steel :30 P'ece of fl ad on round 5. High VOItoge Gloves .1, 6. Stretcher Dual boom ?etCher boner s/N 7 (heck curtain Piece of run thru Curtain 7. Microwave 85004/2002328 Roof ranger 2?15 /Curtain hung 3/4 entry 8. Electric meter Preshift chalk 1?2?06 FJ 4:Probes (for methane Checks) empy Curtain 2O Laying . . Back-up th on cans Run thru 10 First Ald box unopened . . check Jpg?ia?ug ?mam checks 11 Escapeway maps (1 rolled, 1 folded) Continuous miner #2 with 12 Large First Aid Box (green, opened) Check curtain 6? on 13. CSE Trama Kit (silver, opened) . . laying on back Deep Water line . Pieces of curtain Used SCSR A11397 128856 14 D11. on top Piece of curtain~35 to 40 ?oor~2o? this Area Curtain attached . . . . . . to run thru, 1 . 15. Anemonmeter With case 7 7 7 la in on round piece 0 7 7 7 9 fly check Ear Plug 23527) FE Hi Volta Sled CheCk curtain Joy shuttle Joy water Pump 9 hollow Empty oil can 40' of curtain ~20? curtain Scoop Charger $1696"th 5 Joints Plastic cm pa cm ung Pipe approx' 20 5 Bags rock dust 4 Empty Ladder in debris Oil Cans 3?B?bond bags Pressure Pump scoop \1 Belt Sled 4.2 X8 Battery oc us ag bond bags \Check curtain Piece of 8'x6? partially hung /curtain on ground 0 0 Tool box 3 Tanks laying on car model 10 SC Bundle Of wire bottom 1OSC32-64BKXE s/N 2007\chw. SCSR mesh 5' 13' pile of El'16527 laying on b'OCks Stamler bottom feeder 13717 Joy shuttle car-g Curtain/l s/N 2008 laying on Bore hole bottom 1 Belt ~20' of tailpiece curtain laying EDEJEDCW $1 on floor CO sensor (mod 1700N. and alarm pyott?Boone . . Appears stopping buckled Fire Protection and fell In on itself Silver Bullets, Be? FE Ecko System Pieces curtain curtain \Wire mesh . laying on . P'ece, of floor in a pile Standmg curtaln 20?x8? Date board with laying on 1?2 FJ 4:16 am Fairchild Battery 20' Piece Belt Structure in bottom mantrip of curtain\B . 5T Track Jack/ rubber tires from Of mantr'p Belt sled 5 bags rock dust on rib corner 2 tool boxes/ 8? step Looseq ladder 12 extra block T?p 8x20 10 lb_ FE /Piece of 8? 20' curtain on ground 1 Piece of fly pod Scoop T339 336 ags Model 35 CH Fairchild Curtain . . Quickcrete 10 pound FE. ~20 Curtains Fire valve CSE s/N 496 2 large tool boxes I 0 .d 4 small tool boxes aylng ou SI Date board 1 piece run deck 37 Extra thru curtain concrete 4115 AM Pallet of resin blocks dl f6? blts/ un resm Bucket end Pie pans \Toolbox 5 bags rock Stock of roof dust, empty P'e I30n bolt pie pans Hung ioxe?s? in rib 12 from if 55:: ~16 extra Cable o'd dr'" b'ts bottom Fly curtain concrete block Mechanics supply sled?List of items on sled: 2 pallets of E39 Sfcrew kbfigst - crib block 09 r00 us p'le Cat supply parts 2 Pieces of wire mesh 5' 13? Curtain approx 20? long\ Curtain/ 4 bags rock dust Rockust bag Scoop ca?lght in 2 Partial pallets of rock dust . battery '0 er 39 bags of Dry chem fire . 1?2 FJ rock dust Battery Date, Lincoln electric #1 welder, with 4 tanks tied to the back side, Charger written on top (2 acetylene tanks, 2 oxygen tanks with tape on caps) 4n valve of rail on car Supply car outermost rail of armoured cross ties Armor ties\ and 19 rails 2?pallets of oil\ 6 Bags /40 mss?es? \10 lb. FE rock 7 headers 26 Supply Empty dust car?1 bUCket /oil can used roof 10 extra bolt bitsDoor handle concrete empty pallet Roof piece 0 Wll'e mes agalns ri broke inward blOCk bolter toward #3 su lies Partially hung o" can 2.7, between supp y car, pp CPI-tam cars (coupled) Aw65 on cans, 4 8 20 piece Open direction skids 6" concrete 1 blocks, quickrete, wedges, cap boards 10 Rock . . gust Belt structure/ Empty Center Of 8 20 curtain 09$ 3 Posts Run thru fly Bolts Plates 5313\45? long /2 Run thru Cable bolts 8 8 piece Of Curtain?1 piece Rib Roll 3'x2.4x1. CSE 102 gm checkwiieongmg Pie pans Bottom of SCSR of Wire mesh 5 13 on ground coverlng control 8c 4843 in cab mesh 52478 mf 6?97 C0 lines 7? and pie Fairchild scoop pans SN T339-327 Plastered Stopping with a 4"x6? hole patched Bottom of SCSR with black not plastered. Wooden wedges along 57878, 12_97 the top not plastered 10 7 Rock Drill steels, methane 0\ 8'x50' Curtain laying on ground 6 dUSt resin empty, 5 deteCtor o d check 2 - 5'x13' wire mesh on the ground bags armor ties, 0? 0" t0P of scoop curtain Rolled up curtain 8?x50' py OI cans, 1 . . piece of Wire mesh 0" grease . can Curtain on ground 2_1o, cable belts of SCSR Bottom of SCSR 56880, 10-97 6? tie 11_97 Largest piece Of broken SCSR case, 5 B?'te? Bits 4 bags of other smaller pieces at the left of largest 5 Flyboards ton rock dust 2 buckets ,9 Section Map Dated 11/11/05 \miner bits\ Top of SCSR 41 concrete) 0f SCSR Rag $339311: \Bu?ld "1?12 b'??ks Top Of SCSR CO (unused) Center of cribs 0f SCSR 8'x25' 0? St?pping curtain line concrete Belt tools mingrxesit: Shells 1 5x13 Wire mesh on ground blocks plastered It b~t Bottom of SCSR 0 er on both sides 5 Oil can bucket I 1 5'x13' Bent wire mesh on ground 57604, 12?5x13 Wire mesh against rib /concrete blocks tyrofoam Stopping bowed LIOocigncre . 17 bags rock and cracked stocked Empty 0" 57334? 12?97 Pull tab to SCSR dust\~ 4 bundles 8?xs?xs" post 6? bolts /15 8?x5?xe? post Bottom of SCSR Belt StrUCtUl'e\ . 25 Unused Kennedy Panels TOP of SCSR 92652, 03?02 1?4? . 14 Belt Structures/ N3 . 22 8 P?sts Glove blue pipe 8 BOttom mllers 1 ton rock can 3 boxe.s I res'n dust bag 1 Piece wire mesh, aessq.piepaes - - smashed Oil can 2 Pallets 8" Plates . 1 P'ece Of scrap curtain 101831, 01?04 CO sensor 3 Unused coupler bk?j olf tcable 1 G?b Pi'e paes Goggles from SCSR 5 4 PVC pipe to Top Of SCSR 8'x0.5' Post Bottom of SCSR 2 Holster for SCSR s/N 45433, 08?95 Cribs dustags 32 pieces Of 0,1 belt structure 1 Run thru 15 8'xO.5' posts on the ground Bottom of SCSR I can 4 osts; 2 new belt - Top of SCSR 4, structure 1 5ix13. ire curtain 101868, 01-04 x' x? 4" valve mesh 1 5?x13' Wire mesh with a piece of curtain attached 50 splice ottom of SCSR Debris 32"x32" I bars 0" field ?f mandoor Ire VG ve 20 track TOP Of SCSR 106154? 07?04 0" 8x6x16 3 bolts I can concrete blocks in Bottom of SCSR Top of SCSR Oil con\ 89765, 12?01 Ru" thrud 1 piece of wire mesh 5? 13' on Bags of 13 concrete blocks. on groun round rock dust dgoppfd 0? b?th S'des 0f cribs 4 Cribs Top of SCSR 8'x20? Curtain Ell:st Note: Found 24 silicon gel packs (unmapped) blocks 5?x13' wire sh 10 20 Curtain laying on bottom crib :5 Bags of/ i . 'b rock dust/ I. rl pie pans Crib Scoop battery Old Date board stopping 7' bolts and . line Ple pan pie pans Piece Of . run thru . Partial skid curtain of B?Bond 2-4"x20' PVC Pipe\ Posts 1 Side plastered Debris extra struCtUre/Belt structureV field of concrete stockpile Empty rock dust 8x6x16 block pallet concrete Battery 9 3?x4?x6"/ blocks charger Safet 36? m'l\ 9 Pieces of Pie concrete block Shovel no handle Curtain along "b 8?x36" Curtain on ground 25 Oil can\ 2 bags . . . 2.5 0.2 plece unused rook of wooden pallet concrete 15 flyboards 16 11,: blocks Broken piece 5'x5' area piece of curtain . Pipe 3 long (not in use) .. . PLASTERED CONCRETE BLOCK STOPPING OMEGA BLOCK 4 i?headers Flypad roned .up/ x.20 Hi_Line under water 4" PVC 3,611 2, It PLASTERED OMEGA BLOCK STOPPING pipe Al 4832 le 0 PARTIAL OMEGA BLOCK with Belt structure 12 Fly boards CSE SCSR STOPPING WITH MANDOOR some rock dust not 2.. Bolts Unused DAMAGED VENTILATION CONTROLS CONCRETE BLOCK moved concrete 1?Ton rock dust bag block REGULATOR X15 PI PARTIAL CONCRETE BLOCK Curtain as ere ??ncre 6 0? PERMANENT SEAL LOCATION Water DAMAGED SEAL LOCATION HALF HEADER 2 6 X20, Debris contai 3 PVC PlPe misc. bolts, ates, CHECK CURTAIN grease bucket oof bolter tire OVERCAST DECKING 4 full/ Pie Pan cans oil Shuttle car tire OVERCAST Bottom of tunnel liner 1 WEDGE Water ll Fly pod 0.8 wideg steel Stack of DAMAGED OVERCAST 1 Oil can unused CRIB BLOCK concrete fogfgiist Cable bolts - 302 block /-10 Pieces of wire mesh FAN HOUSE Old fire boss tag\ 7 7 i. "use 8' 8' piece of water concrete blocks 4. curtain on ground EQUIPMENT DOOR ELONGATED OBJECT Hi volt couplers Water STOPPING DEBRIS . . 5 em Concrete Block Debris come hung mm, Empty Skid bulk ?028 3 cribs, 1 displaced FLYBOARD 2n suction i all. board line gettingver608t\? A. Man Door ~17 crib blocks FALL FIRE EXTINGUISHER Gob pile peaked to Post 8f glue laying on ground within 3? of top Na? .91. 1o Ome a were n. stackg'dega 8?x75? piece of 0mm? Styrofoam knockedgover Mud bucket "0m sump WATER KENNEDY STOPPING PANEL and broken 4L 8 Unused ?J/T?ll? CRIBS PACER LOCATION Coil 2? PVC '9 el? blocks \9 31%? 7 4* CONVEYOR BELT /3 Omega/ 2 bags Rock dust 1 Ton TRACK PACER LINE CONNECTION 6 PVC stacked 0" ca 0 Color change aego?gwn 11 Beam Safety glasses blocks 1 CO SENSOR LOCATION from white to Bent rail 3 bundle wedges stacked I block Glove, Tool box 11 BOTTOM MINING AREA 4 Mandoor, unused ?2:ng spl'ce 2 water Styrofoam o4:I \4 Bags 1 ?ne I can 3 Styrofoam Trolley wire down/ B?ntp /Quiokrete on belt Pile scraped up mud and concrete block with 1ft dia lid concrete block and more mud blown on top Pieces trolley Wire 0 Water so we- smog ~weme 636 art sealant Damaged Pump 1.: mandoor original .0 0" 20 pieces Oil can location 3 lg left over3x10 2 Skids 0" 31 omega 6 m' 'l concree 5,6x7? ai iac Styrofoam blocks 15 omega Prescri tion Pile of . 4 wedges\ /safety glasses debris Flyboard not skinned but plastered /-Wire mesh a 4L I See Appendx rt?B I I MAP LOCATION DIAGRAM I . a I Appendix H-2 Sago Mine, MSHA ID 46-08791 I 3 Wolf Run Mining Company See -Append'>< Mine Map 2nd Left Parallel 50? 100? BE in Damaged stopping: 12 omega block out Damaged omega stopping, Top half One block moved, One block pushed out /3 Omega blocks stacked I I NE PLASTERED CONCRETE BLOCK STOPPING DAMAGED VENTILATION CONTROLS FALL DAMAGED SEAL LOCATION Line Curtain Intact Continuous Mining Machine Curtain Leibritw\ 3 bolts\ \l Continuous Mining Machine Twin Boom Roof Bolter Line Twin Boom Roof Bolter T7 ICurtO'n Run?thru nstalled Llne Curtain Installed Curtains Curtain (on rib) 5 Curtains up Shuttle car CheCk Run?thru curtains curtain Shuttle car Scoops . . Shuttle cor Check curtain lnstalled Feeder Curtain I 5, :heck curtairll ddown - mer enc se Run?thru check curtain intact Tallp'ece 2 acetylene tanks 7 ower Line Center . Kennedy stopping supplies Sled pump 0 on belt Toolbox #7 Belt drive . Be t Starter box Bet tool sled Mechanics tool sled Old tailpiece-? . Eneocr?iig Welders sled End of Track . Charger cart Plle Of Power deluge Center Light on belt\ Fork Lift Check curtain down Pile of 16' Fly boards Wire mesh 1\ Pie?pans scattered in this area Damaged stopping: 33 omega block out \i Scoop batteries Battery charger Outby tail\ ?31 was.lef.t out for A . CO sensor/ Ventllotlon Purposes 1. Kennedy i outx on belt Kennedy stopping supplies I . /Wire mesh One block outj? curtain installed an 2 Pumps Scoop batteries Belt structure Roof Bolt Resin . Battery charger Rollers frame Damaged stopping: 27 Scoop omega ?Ut One block out 0 Not all blocks here, some on other side WireB/ Shuwe cor of temporary seal. mesh Belt StrUCture Partial Run?thru curtain /Water Line Sling Duster I .i ?L/Hi?Line Sled Omega stopping out, appears to be knocked/f Bantam Duster/ /Wire Mesh out not blown. Stopping fell straight down I Skid of Omega Blocks Supply Hole with: -Skld of ?2 Tires ?3 Skids Omega Blocks I Roll of Cudoin/ -Pile of 4? Bolts . Cracked block probably 5 Skids of concrete hit by equpment I Supply Hole with: A -Skid of Hydraulic on . - - - - One block, bottom 5 Sk'd 0 -Skid of Pie Pans Note. Omega block stoppings bunlt in this area row fell over 0 mega _?kigk.Omega blacks out Lift of Rock Dust One block missing\T \Pieces RAE A A ?n ?a a Oil Can Mandoor\ Gob?l- I Oil Can Damaged Omega om?gg Stogplilng . pus or ino b OCk StOppmg bottom cut area Bl /Oil Can Chain fence Oil Can Sump \Power switch for pump 2" hard plastic line/ Omega block pieces/ '3 . I i Water I Oil Can, Empty Oil Can, Empty /Plastic from bag, rock .l dust bag, possible Pump cable through hole in the Sl'ght heat top of stopping above mandoor Oil Can Empty Damaged Small hole upper right corner, chipped Oil Can, Empty Small hole upper right corner, chipped out /Smo hole, made by equipment Oil Can, Empty, Small hole upper right corner, chipped out Top 7 corner concrete blocks, not plastered I Small pieces of I /styrofoam in entry Slurry duster Pie Pan Mortar bucket Piece of pallet next to track 8mg? hole? mock Plastic mortar bucket pushed out prob. equip Sump A ??Plastic mortar bucket A Bucket on top .0. of roof bolt - l- Ti See Append?x LEGEND EQUIPMENT DOOR OMEGA BLOCK PLASTERED OMEGA BLOCK STOPPING STOPPING DEBRIS PARTIAL OMEGA BLOCK ELONGATED OBJECT STOPPING WITH MANDOOR CONCRETE BLOCK FLYBOARD REGULATOR PARTIAL CONCRETE BLOCK FIRE EXTINGUISHER PERMANENT SEAL LOCATION WATER HALF HEADER KENNEDY STOPPING PANEL CRIBS CHECK CURTAIN CONVEYOR BELT OVERCAST DECKING PAGER LOCATION OVERCAST TRACK WEDGE PAGER LINE CONNECTION DAMAGED OVERCAST BOTTOM MINING AREA CRIB BLOCK CO SENSOR LOCATION FAN HOUSE MAP LOCATION DIAGRAM Appendix H-3 Sago Mine, MSHA ID 46-08791 Wolf Run Mining Company Mine Map 1st Left 0? 20? 40? BE /No. 13 Pump 13 Pump Control Box /?Edge of water 12:10 PM 0 . Fall Crib blocks scattered throughout area, some pushed up to face\ go Post Dee water Proba ation of fl in fully in 7 A Light plastic tied to roof bolt, no heat/ ater 2? pipe attached to roof No 13 mp Cable 81.4r Feet Total Leng Prob pping locat Small pieces of plastic in wire mesh throughout xcut, bags etc. . . end om mInIng Wire mesh at stopping outby damaged . Stopping location, probably mostly underwater at time .0 Piece of muddy curtain, no heat . . . Wire from wire mesh in water Forces . . . 2 Pieces of pipe floating in water Mandoor, no damage Pieces of omega blocks and 6 wooden wedges in wire mesh// Stop ation, coatin oof ri Split Generally brow tenders appear to be pointing inby?? Wire own and to 0'2' Lead . 96 on brow . :0 . .0 Piece of curtain laying across entry, Oil 0 edded in wi Mandoor frame, damaged Bubbles in sump U) 3 'o a: 3 (-0- 2" pl Pieces of concrete block scattered throughout area Wire mesh damaged with rubble from concrete block inside Heavier concentration of broken block Damaged wire mesh and brow tenders diagonally into x?cut Damaged red paint can in wire mesh paint discharged Roll of 3? plastic pipe, no heat, approx. 10.5? did. 5? piece of wire mesh hanging down Original stopping location, coating on roof rib Small pieces of concrete block in area Square Post Damaged wire mesh hanging Damaged wire mesh hanging Small pieces of concrete block in xcut into entry and against for rib Unus rete block Wire amaged i toward Brow pulled to o. 2 entry, of B?bo 1.5 blocks bonded together Piece rtain, no Heav ntration of pieces of te block Post Dam re mesh, . Dam re mesh on Dam re mesh, Dam re mesh on Wires from mesh on floor Wire ent down, 3 bent tow ut Brow xcut Damaged Wire meSh on floor Force ared to co this direc Run through curtain on floor, no heat 0 Wire . intersectio 0 ed if from xcut Possible initial shoreline of water Damaged oil can Force ire mesh i Damaged oil can Belt stand Debris field of broken block Piece of wire mesh on floor, damaged Crib blocks scattered throughout area Old stopping location coating on ribs and roof Piece of damaged wire mesh on floor Pieces of concrete blocks and cribs along the rib Brow 3 damaged ll Piece of wire mesh on bottom Piece of curtain, heat Piece of damaged wire mesh Break Old stopping location coating along roof and rib Cable Wire mesh damaged and missing Piece of damaged wire mesh Damaged wire mesh Pieces of concrete block imbedded in wire mesh in intersection . Crib . . . Blocks Sc blown thro ghout are 3 Jacks . . 0 Wire . mesh ans, force damaged Cribs force 333+ blo n1 0 set here Pu ble: 93 8' of wire mesh hanging down, appears to be blown - - - Fe I Length . Wi damaged ha appears to be Cri re set here Stopping location coating on ribs and roof both sides\ Gob Pi 3h curtain St roof w're meSh damaged? appears to be pUShed Inby ~40?5 bo ging from Wire mesh intact, some pie pans and belt hangers along rib looking inby appear to be bent inby 5' of wire mesh hanging down, appears to be blown Wire mesh hanging down, damaged, appears to be blown .7 Super Post Pieces of concrete block scattered throughout Crib block so alon inby rib thro ghou crosscut location 9f efore Line of pie pans across entry, outby end appears to be bent In Brad toward the inby L. .o.7 Unused Small pieces of concrete block scattered along rib Ho ow . . . . Piece of curtain, heat Block ib Super Post - - - - bl Small piece of brattice cloth, not heat Ottere A. Header . Gob . . . . pans bent, appears bothways, some roof bolt plates bent IndIcatIng Edge of roof fall looks clean force from outb a th Origi 0 p0 ible ro en oar rom op\ crib I cot; Pump 188 Feet ength Super Posts (Crib ma have been ged 5 mS on he jacks Header durin min gen tact throu rosscut. Small pieces Of concrete block scattered throughout xcut\ . Crib scattered ut .I 7'.7 Super Posts 2 Half headers nailed together header block Big John Post 4 half headers 9 Half headers nailed together omega Break 1 4 Half headers nailed together Debris Pie pans damaged Damaged flyboard Looks Cable Reel Omega under it Clean 22?6? Resin bolts . Crib bloc rea 7.7 [7.7 7.7 7.7 Pie pans damaged .0. Plastic thermos water jug Flyboard attached this side\ Some roof fall debris fresh, some old Lit le da age *3 to plates U) (-Inby Double "ne AZ 39's ?00? 49 utby Double ?lnby Dou le line line tb bl AZ 253?25? 00? (a me 0 Out gle line I Crib blocks The "Anomaly" throughout Curtain roun Cable Coup ("Cath crib, at Crib Standing 9 . 2 partial C?bs Stopping location Piece of wire mesh on floor Stopping appears to be out 0.7 o..7 0 ob ile . Crushed oil can\\ . . 3 Pile of appr x. 4 Pie pans damaged . 0 . LEGEND Gob Pile . . Basketball sized Omegas 00 49 3O crib bloc plastered on both sides . A PLASTERED CONCRETE BLOCK STOPPING OMEGA BLOCK /Trace of B?bond . . PLASTERED OMEGA BLOCK STOPPING Piece of curtain wedged in gob, heat 00b Plle . STOPPING WITH MANDOOR PARTIAL OMEGA BLOCK Pie pans damaged\\ /Broken, pointin outby . DAMAGED VENTILATION CONTROLS CONCRETE BLOCK 4. h. h\ Stopping rib, 1 bl ck 6 REGULATOR . Ie lg 6? ng rand wire . PARTIAL CONCRETE BLOCK Pie pans damagedPERMANENT SEAL LOCATION ce of plasti Piece of plasti 0 fro ?oor (heat) . . A .. DAMAGED SEAL LOCATION HALF HEADER Piece of plastic 0 0 CHECK CURTAIN 21 crib blocks in pile Damaged meta Ic handle Show hea OVERCAST DECKING Damaged 8 area bent mostl . . . Pie pans bent in 0? directions Some aged ates In xcut. mag nt manly from .2 entry 0 No.1 t. OVERCAST WEDGE 6? header block on cribs p?zie?'sfut?b A 0 DAMAGED OVERCAST CRIB BLOCK 1 Sea 3 - 0 PP'ng . . FAN HOUSE . . 0.. 9 . EQUIPMENT DOOR I . . 5 ELONGATED OBJECT 0 a STOPPING DEBRIS FLYBOARD FALL . A FIRE EXTINGUISHER WATER KENNEDY STOPPING PANEL 0. 0 0 . CRIBS PAGER LOCATION . CONVEYOR BELT PAGER LINE CONNECTION . TRACK . . . . '0 . 0 CO SENSOR LOCATION Se . . . MINING AREA 0 . No. Seal . gee pp NO 860 (Inby direction) Few damaged pie pans bent inby between xcuts in #8 entry Roof fall Outby fall Shown on map, roof fall visible location inby xcut appears . dark dust covered outby end of fall just inby bottom out . Approx. inby end of bottom out MAP LOCATION DIAGRAM No. Se Appendix H-4 No.1 0 eal Sago Mine, MSHA ID 46-08791 Wolf Run Mining Company Mine Map 2 Left Mains 50? 100? 55:: I I Heated plastic band Piece of heated plastic Piece of hea plastic wrapped on roof bolt on floor Piece of heated plastic I Plastic band Crushed can Damaged mandoor Crushed o?l an Wilrred frame Ni: Piece of plastic Piece Of curtain I Crushed C:Pie rpadn cllestroyghe Pie pan pieces piece of heated curtain 3iece of shredded, heated curtain bmgga?d Ciigide . rus OI con Damaged Plastic 'u inside Piece of shredded, heated plastic - Belt splice Heated bUCket ple pan crushed metJalg bucket If lt Plece Of . 0 on ere mes /W?e Pie pan 3? PVC Hm? p?lece Of?wood ?pipe Pie . Wire Piece of heated plastic 'ece Wire as mesh AA 0 Wire ponP'\e pan"; ?eSh-\ w_n_ Piece of plastic madriver n. 7? I mesh 3 A (16Pieces of metal can\ A handle shows heat Piece of curtain . piece of heated curtain Piece of heated cudain?f2-?g-r2wggeocr: 0f pallet Heated plastic band Piece of conveyor belt in debris Piece of A Pieces of mesh wire throughout Post plastic Yellow reflector 55L Piece of plastic . A (2 8) A i Entry full of gob :59w??d 6 PVC 3 plastic Wire w. A J). pipe pipe mesh H20 pushed 1 left A Angle fork undepe?ail \Ple pan Track tie AA Plastic band Shirt Pallet board /.oL (4o) allepgeggogg Damaged metal Pieces of allet board Small ieces of allet board 3 bUCket (4551 ?ll-(20; I3?Rubber glove '99: what) ?32 .91. ?32 3L\Plastic band (no heat) Curtain on Concrete block Omega can . Beams Stocked no eomss alIets A Wood skid heat Board 14p 2 Booms Piece of heated curtain (3) 3 mandoors 3 Beams Damaged metal bucket (26) Piece 01' pallet boord 1 Rail Beams . . 4" Curtain A ?In. 0 8 41 n- A 4?99.- Beam Board A RUbber glove/#L 0? (35) Aft-A 9- as: ?is - Shredded wood in pile .oi. ., A 4 A A 111%: can '0 MA (A) A (7) In, A1541Overcast 5 high Pall?:ClilS Bea ndogr A A i Curtain Pallet eoms Beams on it .I eoms . Broken prop setters Wire Boom Beam Beam 51,5, Wire ground m' Fresh water line 8mg Prop setters Board Post pi . 561) . Dama ed to belt stucture Oil can Pallet board Oil can . Pleoe 0f powEa ord 6 water line 9 Piece of pallet Piece of pallet board Debris with 10 crib blocks small pieces of 4L Bucket top? Wt wrap Damaged top belt StUCture f. omega, crib blocks, concrete pieces, wire mesh 1% A Piece of curtain?453?? ~Boards Top roller uarding, belt structure, Hi?Line guard, wood, Piece of pallet Piece of curtain 9 4L Damaged top 4 large omega pieces A Be Oil can 1* Pallet Belt I'Ol'el' IEe'ft SgUCture . Top roller .oL Q) n- Bottom roller hanger Piece of pallet Top belt structure anger ere A Bottom belt hanger Damaged pie pan n. P. . 6" piece of water line P. rt . Belt hanger me . mesh Top roller 4L 'ece 0 mm rcpe Piece of curtain - - Piece of heated curtain n- E'w fece 0 CU wooden Sk'd mastic bucket Piece of plastic Pieces of wood Piece of heated rock dust bag A i of pallet A A 4A A 4L Heated plastic .Stee' bond . Piece of plastic I t. piece of heated curtain a of curtain n- A Pleoe 0f Piece of heated curtain pas '0 Piece of plastic piece of wood Muck pile . m- M. A Bottom roller plpe Bottom roller Pallet board Piece Of heated 473?3: .91. 4L A :31: . 6 CI I Wire 8" _beom Curtain Damaged pie pan 4L a . Belt stucture Ski sh Guarding Wire mesh Top roller 2 Discharge line Bottom rouer Piece of 6? lastic pipe A and belt Tog obelt stucture. bottom roller Piece of plastic 6 structure All guards down along belt take?up Do ?3 IS UP 090m?" bent UP guard Steel Channel piece of heated pie pan Piece Of plostic Piece of heated curtain . . Piece Of p' n. Debris: 1O cribs concrete blocks w'fte meSh l?dged 'n the take?up Debris (stacked up to drive motor) contains: \Cable bolt Pieces of wood . A . Plooe 0f heated CUFtom heated plastic . '30? 0 e0 3 p03. '0 omega pieces, bottom belt pushed \ggit toke_up 30 crib blocks, numerous cables (small), Piece of 6? plastic pipe Piece of heated curtain Piece Free of wood P'ece of heated curtom . . 0 Buck 3 Pieces of plastic belt drive guarding We" to Cr'b? t0P belt StrUCtUre 'ntC'Ct Fresh water line Elece of heated plastic 332:3 Crossover board Piece of heated plastic woo'den piece of Piece of heated curtain on roof saw .91 Drive motor coupler Belt toke_up power unit /Speed reducer 'e pan piece skirt rubber, 3 pie pans, loose coal, A Piece of heated Piece of plastic pipe Lock?N?Load Phone head May be signs of heat on spad ribbon on top of debris . HV plug A pieces of heated plastic, plastic guarding curtain'\ piece of wood Metal latch from piece Belt drive motor guard? bent P'e pan Piece of pie pan . A Piece of heated curtain Lock?N-Load 2 ?le?063 We mesh I Steel p0rt 0? 0? armored t'e Piece of heated A Wedge driven into roof bolt piece of heated cumin 4% 4L 4L 4L curtain Piece Of Damaged oil can A $41 A a A heated piece of A 4.. 5" Metal clamp a Lock?N?Load {Curto'n plyboard .91.91. A A A A M- Damaged . I . Pie pm A ow Piece of heated plastic Mine water hose 1: A .041. Ila-4L w/ A A Piece of wire mesh HV guard 0' Belt mils/ (g3) Omega block in place 2 Drive rollers Pie pan Damaged pie pan . . le an Mondoor (39.) 2 skids of rock dust Wire mesh Motor coupler sitting onr??gr Ogegg'r?g Pagerpphone. no headset \3 it . . Pieces of wire mesh Speed reducer, coupler and coupler Deluge box 6 ml Pallet vonous cople P'eces 9f bINOOde?nt wedg: Piece of heated curtain 3 . . . . bolts, banding curtain In 0 '00 . . Small piece of curtain Debris piled against discharge chute and thoughout Greg bolt pan from Small piece of curtain inby Metal bucket papers 't 0 CO monitor on the ground guarding missmg . . Belt Chain missing Starter box guarding on 2Left discharge. Debris Fire valve missin from fresh water line Xmas tree coupling on fresh water contains: 42 crib blocks, omega block Reflective marker hanging from roof 9 Fire suppresion pipes hanging from roof pieces, concrete block pieces, wire meshChunks of omega blocks, "Bowling Ball Size?, Damaged Piece of CO 6? waterline broken 8? 8c 6? waterline pieces, top belt monitor Damaged fire valve under belt Guarding still in place this side\ structure, roof jack, date board, 6 9 Cable 1 bne't Chonfle' Lock-N?Load post A /thoughout crosscut . Trickle duster O'Umlnum plpe Piece of 6 water line Free of Ape Wire /Wire mesh i $43- AL Piece of 6" waterline D9m?ged meSh Part of Lock?N-Load post W're meSh EZEOSS . A A a: A An- (af- p'ece Afr: A entry 3- FE n- on . . crib blocks Ph End of 6? waterline roof 2 pieces of 8 waterline . this area honZRE/ 6" Victaulic tee with two valves Roll of 2? fire hose P'e [3'6 pan 6 this area Date board w/ light bu'b frame 6 It 03, 2 Pipes Iggy Belt 5 Rock dust bags /Mostly intact?Zr blocks out Deployed SCSR Bottom belt roller w're meSh Board . 13%: Jacket Hydraulic saw with hose. Front wall missmg SN. 131718 Bottom belt roller 2 pager batteries Spool of Dr pepper can Curtain 820m Written AO8739 \5 Post ire mesh I 00? . . Wire Seat made of Lunch Pa? cable Rubber glovgv Curtain Top of Heintzmon Jack . Jocks with top A Piece of curtain Jo??pe?g_ Board Eel; channel 3 Cribs Glove Track tie on rib Montrip charger lid 47L Belt structure stacked against rib Filler Fl rock dust bags Rubber Belt splice Drill steel Angle iron (2) Rails Piece of pallet plank Line curtain a glove CmBCO'jfd gigs/gm /-Roi tie Paper bags throughout area (2) Armored ties w, Pal'?t Plank A 7/ . 5 Rail Pallet Hanks /10 coupleShovel handle 39?er P'Ote Damaged pallet Skid a Line curtain Pallet. plank Brattice End of wire mesh FE Crumpled Oil can Wire mesh bent A A A Lock-N?Load back 7 2 Pie Pan \Piece Plev?i?Belt ac Jack - splice Cople 6 Rock A . Glue box Ho .nem zmon 0? Cable Kennedy angle $0 .3 2 bundles Belt DamaggeFCDI switch throw a Belt mils \Armored tie 0's Damaged 4X43 Piope ?30 'e . Kenned elt rail 4" Llne CUVtoln Door damaged Curtain? g1?ennedy Jack for 5m?: ?re Damaged Steel ?03 . A Bottom belt rollers w" . A n- Wire mesh bent toward Hondkereh'ef' ?llers G'ngles motor 1'1? wwman Equ'pment 49 Be? StrUCture Angle iron 100mb rook dUSt bog Line curtain'\ . bent back Fire suppresion 0" beam i.i p'e"es meei?l) Steel plates . doors ILA Equipment door Cable bolt\ P'ece Of pallet Wooden wedges in wire mesh on roof . Lock and load bracket canister bracket leg 5? (2) Jack tops He'ntzman 100k PU'VeriZed omega b'00k\ 2.. Vg'ret r268: C?Chonne' Recove Beitin eoms pipes ?2 Partial roll of cutain Air lock door . . en 0? Pallet Piece of post Suit ry I Metal piece I. undamaged 6? Jack AJI- Piece Of metal 4 Rows Of Traces of omega block in floor Belt hanger/ \l boom Cboardd4 GOB Ce Moto' Strip from oxygen Transformer njf? "he Crushed oil can - . FOSS Ie- . rus allet oar Piece of post Strip of belt_ equ'ppng?gf 88:; acetylene tanks not in use Smashed Muck pile_\ tied to rib . Plank Smashed oil can Belt hanger Piece of wood 0? can Piece of Wedges in wire mesh on Oil Can\ GOB Wire mesh El A Board A A - amage A .n.w 97' J3, Pallet board Pallet r??f a i. soil, i [Ice . _n_w 4L AW i A 11Wire meshed 5:8? Of Wire meSh roofing pulled Post . 8 6 CI I . Piece of Pal'et Post in 0?1th Pallet board . metal 300,500? post Pallet .direCtion C?channel Oil can piece Board boards hangmg down Pallet board Line curtain .PL Batteries Part Of trOSh can Pallet board Piece Of pallet Line curtain Board detached . Debris in piles pushed by scoop, iirgleri?ees hOf 7 omega blocks, 5 crib blocks, 5 Pallet small wood blocks board on A A Pallet plank Crushed Oil 9 Board Pallet can 3 . . Cons 42%. Wooden "e board Post Oil Pie pan 4 It .I can Mangled pie pan on floor CrUShed 0" w/?E?ecie [Sf Load . rais . . Pallet A can 0" 00? . A Half of Lock?N?Load post a Small P'ece curto'n (heat) Fly board rotated outby ww A A .2.- A a 39! 00? Board Pallet around bolt 5 p? m: Curtain n- poni board .91. i IL A vvNo.9 Seal Post Board Pallet board n- A .91. Pallet laying against rib A Pallet board Board 4L Pallet board Guard Gob Pile pushed by scoop Small piece of wire Post Pallet . 2) Damaged Lock?N?Loads a; small piece curtain (heat) 15 long 2 alum. deluge pipe Part of fly board bent on roof bolt A 0 blip: 15 bags of rock dust Gob ?f w. mes ire tsio in My; sevem' Pieces Wire mesh, 4L 1,921 Plastic line curtain (little heat) pp 9 10 pieces of belt structure meSh W'thN 50 A Gob pile/(g /Bucket xe .oi. Wire_/a I Mangled pie plate piece on floor .oL . Wooden Pallet mesh :1 core block 2 pallets Several 2 Post pieces at top 4L .91. Pieces of damaged pie plates n- A . 1 4L Small pieces brottice Bent pie plate piece on floor 4L 4th cloth (plastic) shredded Wire /meSh Spl? tered wood JAM In HE debris 53' Plastic bucket no heat apparent and apparent heat edges Broken ?y board pieces I p, Damaged . n? 1L 4i. 2 steel seal sample pipe HF a 4L Board . w're Brattice mesh Cable bolt . . . \Piece of heated plastic curtain 2 Steel p'pe (360' mon'tor'ng p'pe) Piece Of tie wire on rib spad plastic coating melted NOIO Seal 1/2? copper sample tube (sticking out) Pieces of shredded curtain Piece of crib block on inby side of roof bolt plate Metatarsal leather glove-no apparent heat LEGEND 12:: PLASTERED CONCRETE BLOCK STOPPING EQUIPMENT DOOR OMEGA BLOCK MAP LOCATION DIAGRAM PLASTERED OMEGA BLOCK STOPPING STOPPING DEBRIS PARTIAL OMEGA BLOCK ELONGATED OBJECT STOPPING WITH MANDOOR CONCRETE BLOCK FLYBOARD 5 DAMAGED VENTILATION CONTROLS FALL REGULATOR PARTIAL CONCRETE BLOCK FIRE EXTINGUISHER ago me, - PERMANENT SEAL LOCATION WATER DAMAGED SEAL LOCATION HALF HEADER KENNEDY STOPPING PANEL W0?c Run Mining Company CHECK CURTAIN CRIBS CONVEYOR BELT OVERCAST DECKING PACER LOCATION 2 North Mains 56 to 64 OVERCAST TRACK WEDGE PAGER CONNECTION 2 O, 40, DAMAGED OVERCAST BOTTOM MINING AREA CRIB BLOCK 0 CO SENSOR LOCATION FAN HOUSE Bottom pushed inby by equipment Concrete block debris Bucket . Entry rt en CU 0m 7 Rows of proposed Empty o? can full of overcast wall gob Door open . . . . Spray Sealant . 24 blocks, plastered together 7 Rows Of proposed overCOSt Wu? Omega b'OCk debrlS 3.22222222 Stopping With bottom Spray Sealant knOCked SCOOP Metal o/c decking extending through All blocks on wall top of overcast gone Old Stopping the hole in the bottom of the stopping 2 Blocks out of wing wall on top of overcast 7? Spray Sealant Piece Of tin from top 3333333533 Empty wood cable spool 2 Decking sheets popped up JOCket Spray Sealant Spray Sealant\ _n_w n- A 5 'Qh. 4L A n- A No damage to side wall . Omega Block Debris A H. L. ine uar All blocks on wall top of overcast gone 9 #5 take up power pack Door removed Gobbed Wire w, I Piece of wire mesh re rn es Wire mesh Plastic bucket n? n- 13nun-'0' n- ?an. n. n. An. Wynn?belt motors A n- Pieces of A .0. Piece of curtain Wire mesh 3 ome a n- . Red nan-A Piece of Wire mesh 11.0. reflector\ #5 belt starter box . . A A Omega Block Debris 43- Belt guard Wire mesh 2 pieces 19 omega blocks . . S.d Omega Block Debris 3 pieces of 6 water line 9 W0 0 A {g git-q 38 omega blocks 429. A 1m- .111). . Belt owered duster on round Don er 3L Empty on can Piece of belt 9 Li ht bulb uord Shut off valve on Signg omega block on belt 24 omega piled in xcut 01? 9 9 fresh water line . overcas amage . . 1L 0 Trickle duster P'ece 0f guard Part of light fixture Curtain blown against chain Oil can . . . on ground/ Cross 2x12 wedged in structure . '3 4L Fire under Slde WOII 0k BrefaITIthruglicij a on groun Board Suppl Curtain wra ed around a chain 64 unused Omega guard 1?5 Plece . - 9 Posts Set pleceS 0f Cl'lb set pp 51 Ome a ieces Sludge pump of waterline b'??ks beltin and in offset 9 5 pieces of 0/0 prior to event If 2?6? decking 2 rails 0 Piece of i float'n 'n . I I we er\ The omega 0? the top 0 3? plastic 6? Pipe P'pes Ouse of the overcast are missing. T0 of overcast and outb - . . (Omega block) 9 on. line Belting 6 Pipe (5) 6? Pipes SU lies; sides) were flush with brow ~50?6" Pipe can bUCket 5 4? line coupler 6? waterline Crin: caps water Top 2 courses A D'schorge ?he tie Wire 0 0n . End cap for 72OOV HVL Plug handle bugket Stockp'le line couplers out 0? Side 511.0081}: Date Boar 9% Belting Curtain\ Belting bucket Belt r9"\ trnesh '3 Lrackh du ba 2.4" ?eces . BUCket me Of r??k Bent overcast rail . . i . Pallet r00 '6 re.nc BL 00 9 of plgigloss Oil can dust bogs A Only 2 rails remain Monfdoori 5 Timber . on bucket spray sealant kit . Wire a a Wire A 0 st M- a 4L-0L . 3L n- an. 3L0??llAIL Baa-(3T8 walkway .53- APE A wa er COP light, RP stance pump line cou [er Operating pager hanging Jar of soup, flagged 8'x4'x piece Old belt-n A Rose gun (bmken lef?sev End cop forp Post/ . BTool box of metal bent Top course out .39- in walkaag I I, 3:33? yondoor/ w're mesh bl Brattice cloth Igloo BUCket brOken sw'tCh) 7200V HVL Fire eXtingUiSher hanging on side wall ma?a: 03:; a Wire mesh Bogg?ne Part of Electrical Box \0 . on can Concrete Block Debris Fire extin uisher hon in BL/Original regulator. Tail Roller doors open Bmushed Hfletnear top 9 9 9 More block blown out than Was cr'b knoc ed over Belting Not in service P. . . . In iece Stopping was a combination pp 9 3239969 Damaged power center 2 Skid of Sgginglaa?eTgygisi enin 16 Rock Dust bags Bucket angle and omega Door Congfenteer 2 EMT kits b'OckS Mine Rescue Team 9 16?8? Pipe stockpile post Angle pieces panel 0.08 metal panel Plle band Str?p\ i? a . Curtain torn down . block debris Block Pile Mandoor Bottom row in by fall COB 15 Unused Whole Omega block El 42L El Mine Rescue Team phone Flat 0il line under fall I Beam Shovel blade Ball of Wire Mesh Wire 1" 008 mesh Lagging board\ . 5 allon oil can P'mesh 47L 41. WNW ?n?n'xL 4" 4L 47?- n? Bucket ?gLn. 413- . . COB meSh debrlS Floating omega debris 5? 13' Wire Bucket A A a mesh/ 2h. M- 9 Omega <36 1 bUCket Ha Am. CB. headers, Unused concrete blocks under soot covered wedges one Concrete curtain No heat A block out of top Muck Pi'e 4 headers, . in middle Omega from A 3 wedges as" stopping wall 22 Omega blocks and Crushed 5 Gallon 1 block out Balled .up wire 3 Concrete blocks partial Omega blocks empty oil can 72 block removed mes? 9mm? stacked against rib ATBucket Muck Piles Mud? Pile P'l as a \6 '6 Several pieces between omega rib El 4L El 5 mm 9 All pie pans on roof Bucket Crushed bucket Crib debris bolts are in good shape Concrete blocks Crib debris no damage on roof stacked against rib LEGEND PLASTERED CONCRETE BLOCK STOPPING EQUIPMENT DOOR OMEGA BLOCK I PLASTERED OMEGA BLOCK STOPPING STOPPING DEBRIS PARTIAL OMEGA BLOCK ELONGATED OBJECT STOPPING WITH MANDOOR MAP LOCATION DIAGRAM DAMAGED VENTILATION CONTROLS FALL CONCRETE BLOCK FLYBOARD Appendix H-6 PARTIAL CON PERMANENT SEAL LOCATION WATER CRETE BLOCK FE FIRE EXTINGUISHER Sago M. MSHA ID 46 08791 Ine, - DAMAGED SEAL LOCATION CRIBS HALF HEADER KENNEDY STOPPING PANEL Wolf CHECK CURTAIN un ompany CONVEYOR BELT OVERCAST DECKING i PAGER LOCATION Mine Map OVERCAST TRACK I WEDGE PAGER LINE CONNECTION 2 North Mains crosscuts 46 to 55 a DAMAGED OVERCAST 20, 4O, BOTTOM MINING AREA CRIB BLOCK 0 CO SENSOR LOCATION Gob two feet from roof ESQ Omega Block DebrIs . MAJ Tunnel IIner . Post (7) Metal Gob two feet Stopping from roof Panels, Intact A Omega Block Debris .0. Mortar bucket RUbble from 5 I .I clean?up of Curtain 90 0' can a roof fall installed Bundle of 2? 47L MRT a CrIbs set w/s Of belt . 4 steel sets Fall area 6 steel sets 3 MT oil cans 4 Whole blocks resin bolts Man Door Debris field with 2 New blocks, No Plos?ter crossties 6 long COB A can "d GOB CM head Iight/ 4 Crushed setting on 3 Blocks around the mandoor are empty 0" Cons missing on the track side. There is no sign of debris, This may be pre?explosion damage. - (:03 Plastic Bucket wad of curtain in wter Pile of wire mesh Stopping G93 and . Mon Door WIre mesh In 13/ Removed '3 One block out 5 Gallon empty oil can 5 Gallon empty Oil can 5 Gallon empty Oil can Man Door removed against empty Full oil can on rib 5 Gallon empty Oil can Gob SStOCked omega block Damaged metal stopping panels Omega Block Debris IL Omega Block Deb?s Omega Block Debris Wm 4L gallon oil can Reflective oxygen activator tag of SCSR TMX 412 pump I L. unne Iner Omega Block Debris RBOttom 0f SCSR Bad Top and Gob 5 Gallon empty oil can Smashed empty Oil can oil can GOB 3 of entry high Pile of fly boards Crushed 5 Gallon empty oil can Empty ProSeaI bucket 5 foot wide steel wire mesh (twisted) Empty Pro?seal Bucket LEGEND PLASTERED CONCRETE BLOCK STOPPING EQUIPMENT DOOR OMEGA BLOCK PLASTERED OMEGA BLOCK STOPPING STOPPING DEBRIS PAR STOPPING WITH MANDOOR OMEGA BLOCK ELONGATED OBJECT DAMAGED VENTILATION CONTROLS FALL CONCRETE BLOCK FLYBOARD MAP LOCATION DIAGRAM REGULATOR App H-7 PERMANENT SEAL LOCATION WATER CONCRETE BLOCK DAMAGED SEAL LOCATION CRIBS HALF HEADER KENNEDY STOPPING PANEL 8390 Mine: MSHA ID 46'08791 CONVEYOR BELT OVERCAST DECKING PAGER LOCATION Wolf Run MInIng Company OVERCAST TRACK CHECK CURTAIN WEDGE PAGER LINE CONNECTION 2 North Map 36 45 DAMAGED OVERCAST I BOTTOM MINING AREA CRIB BLOCK 0 CO SENSOR LOCATION 20 40 BE FAN HOUSE 7.8 {If'e 6?x8? hole for dischar I- II D- I. mug Opp'ng Omega Stoppings Removed 9 I. Discharge line tr'lsc Gige .me 2? Dia. hole where I Uzve4Miigrs thru Stopping noruhosleOPP'ng? discharge line is a NO. 4 Belt No? 4 Be? take?up Cable thru wall: No hole thru stoppmg Discharge2X3 h0 e Belt Tailpiece 2? 2, It Pum hous hOIe . .0. Power hole hole CO monitor hole hole hole fl? {1f ?m 2? 4 B/Center TB 1 2hx 4 L'ne Spl'tter . 10 steel trusses TB TB oe oe TB Stacks of Power \Rock on top I) f) omega center\EI of hung Hi?Line TB New roll of wure\ \Half block \Bowed x4 ?3 x4 ?3 Jocket/ . Location Of new two" of new track TB StoPpmg 0e 06 TB TB D. TB hanging c?gfger 3 door (not framed) door (not framedcrib in bOttom of wedges - ISC ar me ru TB - -B door?n2 hole . Stopping Concret hf]: ien Ho'e In bOtt?Zm (ff Pile of concrete b ock I I I 53:) 1 18 XIO water block from old wall debris Door Omega runnmg toward track Only joints plastered this side opened stopping l3T TB removed Wire mesh/_Roof bolter stockpile ?9oB Loose check I Loose Loose check check\6' Run Run Loose/ Run Loose Loose Check check check 0? 100? 200? Belt control line severed Concrete block overcast wall intact 3 Concrete blocks,and 1 crib block Overcast Damaged OX 0 ec 3 Concrete blocks on belt . CO monitor Overcast wall has blocks Concrete block overcast wall intact M'ssmg: '3 7'9 9&3, hOIe 6 Concrete blocks, 1 crib Power center block and 1 wedge on ground TB New foam applied on top of overcast Ports Shonty\ Overcast Damaged. Appears that decking has shifted. Gaps visible between decking panels. Pump Joints Only Foamed Return Side Check curtain installed on top of overcast Omega block from wall on top of overcast laying on the ground 9.8?x1.8? Hole on top of overcast wall 14 Concrete blocks and 5 wedges overcast Damaged bl . Concrete block overcast wall intact more 6 0? overcos WC 00 All omega blocks on top of overcast were plastered both sides Overcast Damaged Concrete block overcast wall intact Concrete block wall built on top of the overcast. Holes are patched with omega blocks. Concrete block overcast wall damaged. . Cribs set under brow on Concrete blocks on mine floor 8.3?x2.0? Hole on top of overcast wall top of overcast mil Heigth on top of overcast: 6.4' 14 Concrete blocks and 1 wedge Omega block wall on top Concrete block overcast wall intact of overcast still intact Hole, 2 {x6? Hole, top of stopping: 0.8? 3.7' 9 Omega ~18 mostly full omega blocks block debris /5.9?x8.0? Hole Shuttle Car Feeder/breaker This stopping was built with concrete blocks. A hole (approx 6'x10') was patched with omega blocks. The remaining concrete blocks are undamaged. The stopping was plastered on both sides Plastered on the outby side. Hole 6?x5.4?. The remaining blocks are bowed toward the outby. ET Bowed toward return side LEGEND PLASTERED CONCRETE BLOCK STOPPING PLASTERED OMEGA BLOCK STOPPING STOPPING WITH MANDOOR DAMAGED VENTILATION CONTROLS REGULATOR PERMANENT SEAL LOCATION DAMAGED SEAL LOCATION CHECK CURTAIN OVERCAST DAMAGED OVERCAST FAN HOUSE EQUIPMENT DOOR STOPPING DEBRIS FALL WATER CRIBS CONVEYOR BELT TRACK BOTTOM MINING AREA Remaining Mostly whole block in wall omega blocks Joints Only Foamed Return Side bowed to the return side CO Monitor Spray%' Sealant 1 omega block, 2 cribs, 1O wedges OMEGA BLOCK PARTIAL OMEGA BLOCK CONCRETE BLOCK PARTIAL CONCRETE BLOCK HALF HEADER OVERCAST DECKING WEDGE CRIB BLOCK ELONGATED OBJECT FLYBOARD FIRE EXTINGUISHER KENNEDY STOPPING PANEL PAGER LOCATION PAGER LINE CONNECTION CO SENSOR LOCATION SEALED MAP LOCATION DIAGRAM Appendix H-8 Sago Mine, MSHA ID 46-08791 Wolf Run Mining Company Mine Map 2 North Mains crosscuts 3 Belt 36 to 4 Belt 36 how .02 .0 mm :wm 9 $2950 mgsommoho 8:22 5.62 >cmano cam 550.8 9 .83 comm mi 5282 QZmon :52 9.258% EDD EDD BUD QB UDDUQEDDSD UDQUDDQD WEED SEED DD DDUQUUQQUU mega? ID I m_ 3w oo Appendix I - Executive Summary of "Investigation of Pyott-Boone Electronics MineBoss Monitoring and Control System" u.s. Department of Labor Mine Safety and Healt h Ad minist rat ion Industri al Park Road RR1, Box 251 Triadelphia, Wes t Virginia 26059 April 19, 2007 MEMO RAN DUM FOR RICH ARD A. GATES District Manager, Coal Mine Safety and Health District 11 FROM: JOH N P. FAINI c::;y:}­ Chief, Approval ~cf Certification Center SUBJECT: Executive Summary of Investigation of Pyott-Boone Electr onics MineBoss Monitoring and Control System A computerized monitoring system manufactured by Pyott-Boone Electronics was in use at Wolf Ru n Mining Company's Sago Mine at the time of an explosion on January 2, 2006. Portions of the hardware and software associated with this system, called ' MineBoss Mon itoring and Contr ol System,' were evaluated to de termine operational status. Additionally, d ata associated with recordable events stored in th e com puter was extracted and a copy of the computer's hard disk drive was made. On January 11 and 30, 2006 and February 1 and 2, 2006, the Pyott-Boone Electro nics MineBoss Monitoring and Con trol System was inspected, tested, and evaluated to determine its operational status. The system was used to measure the carbon monoxide (CO) level in the conveyor belt haulage entries and near a battery charging station, in the mine and report those levels to a surface location. Certain events, such as CO concen trations above pre-defined alarm levels, were recorded by the system via a printer and stored on magnetic media. Visual and audible alarms were loca ted undergr ou nd at the 1 Left Section and 2 Left Section conveyor belt tailpieces, and mounted to an outside wall of the dispatcher's office tr ailer located on the surface. The system was also used to monitor and control the operation of underground conveyor belts. Again, certain events associated with the operation of the conveyor belts were recorded and stored by the system. The stored data, or 'event log,' was used in this evaluation. Additionally, the operation of the system was observed, and the CO monitors were inspected and tested by ap plication of a known concentration of CO in air. All dates and times were those recorded in the event log; they were not revised to reflect the difference between actual time and the computer's real-time clock. H owever, it was reported by Marshall W . Ro binson of Allegheny Surveys, Inc., that the computer's real-time clock , and th erefore Appendix I - Page 1 of 5 Appendix I - Executive Summary of "Investigation of Pyott-Boone Electronics MineBoss Monitoring and Control System" the time recorded on the event log, was 4 min u tes and 56 seconds ahead of the actual time. The follo w ing are the significant findings of the investigation. Following these items is an approximate repro d uction of the map of the underground com ponents of the CO monitoring system, wi th graphical reproduction of each device. • The Pyott-Boone Model805C remote alarm located at the tailpiece of the 1 Left Section belt was not operational when tested. It was wired incorrectly, such that it would not provide visual or audible signals when manually operated by th e dispatcher or automatically operated by the adjacent CO monitor. Based on a review of the eve nt log, and assuming that the wiring had not been m odified sinc e the time of the accident, the alarm would not have pro vided aud ible or visual warnings at the time of the accident. • The Pyott-Boone Model 1700 CO monitor located adjacent to the remote alarm at the tailpiece of the 1 Left Section belt did not operate properly when tested. It read '26' in clean air and '74' with 50 parts per m illion (ppm ) CO applied to the sensor head. Ad d itionally, it was improperly w ired to the aforementioned Model 805C remote alarm, so that the alarm unit would no t in itiate. When w ired properly, this CO monitor would cause the Model805C re m ote ala rm to give aud ible and visual w ar nings continuously, regardless of the CO reading. The data in the event log suggests that this condition existed at the time of the explosion. Furthermore, the data suggests that the response of this monitor was drifting, or changing without a corresponding change in the carbon monoxide content of the mine atmosphere. It appears that some corrective action was attempted on several occasions, most notably during the early morning hours of December 31, 2005. Also, it appears that the system operator ha d attem pted to reset the device, by taking it 'off scan' and placing it back'on scan,' at approximately 6:09 am on January 2, 2006. • The CO monitor with address 1.34, located beside the #2 Belt near crosscu t 7, w as measu ring CO properly on January 30, 2006/ but was not reporting the value to the surface. Tw o fuses located in the 'Data +' and ' Data -' circuits were open­ circuited. Review of the eve nt log indicates that communications w ith this CO monitor w ere lost on Jan uary 2, 2006, at an indicated time of 6:32 am; this is most likely due to open-circu iting of the fuses. The event that caused the fu ses to operate in the data communications circuitry is unknown. • Nineteen (19) of the twenty-five (25) CO moni tors inspected underground gave readings within 10% of the intended value w hen a test gas containing 50 ppm CO w as applied to the sensor heads with the Pyott-Boone calibra tion adapter 2 Appendix I - Page 2 of 5 Appendix I - Executive Summary of "Investigation of Pyott-Boone Electronics MineBoss Monitoring and Control System" and flow regulator. Ad d itionally, one (1) of the CO monitors inspected underground was damaged, not connected to the system, and could not be tested underground . • The mon itors that did not respond properly to the test gas, or were non­ functional, were as follows: Address Location Motor Barn Spur Zero Readino 0 Span Readinc 40 1.29 1.39 #3 Belt near Crosscut 38 109 109 1.40 #4 Belt near Crosscut 8 0 75 1.46 #4 Belt near Crosscut 57 0 19 1.47 Tail #4 Belt (intended location) - - 1.80 #5 Belt near Crosscut 15 110 110 1.99 5 Belt tailpiece just outby the section feeder 26 74 Comments Device failed on January 30, 2006 Found face down on mine floor , covered in soot Fragment found on mine floor beside # 4 Belt between crosscuts 44 and 45, Damaged , could not test in mine Device failed between Jan 2 and Jan 30, 2006 . • The event log indicates that, at the time of the explosion, conveyor belts identified as #1, #2, #3, and #4 were most likely running. It is not possible to d etermi ne the status of the #6 belt, because of damage in the area of the belt drive, but the event log does not include an entry that indicates that it was running at the time of the explosion. It's likely that the #5 belt was not running at the time of the explosion. The event log includes entries for Belt #7 before the ti me of the explosion and the last entry in the even t log for this belt was on December 29, 2006. The physical evidence indicates that the equipment associated with this belt was in the process of being dismantled. • The fragm en t of a Pyott-Boone CO monitor recovered from the mine was determ ined to be the unit with address 1.47. It is the subject of a separate investigation to d etermine if it contributed to the explosion. • With the exception of the unit with address 1.47, Exhibit Number 114P, there was no evid ence that any of the CO monitors produced cond iti ons that would have provided enough energy to ignite a flammable methane-air mixture. The explosion risk of Exhibit Number 114P is the subject of a separate report. 3 Appendix I - Page 3 of 5 Appendix I - Executive Summary of "Investigation of Pyott-Boone Electronics MineBoss Monitoring and Control System" • The entries in the eve nt log that were recorded on the morning of Jan uary 2, 2006, were evaluated. Definitions of each entry were provided an d the actions that could have caused those entries were described . Com prehensive inspection and test results can be obtained from the Chief of the A&CC, RR 1, Box 251, Industrial Park Road, Triadelphia, West Virginia 26059. 4 Appendix I - Page 4 of 5 Appendix I - Executive Summary of "Investigation of Pyott-Boone Electronics MineBoss Monitoring and Control System" 1.47 1.46 1.45 1.44 1.43 1.42 1.41 1.40 4 1.39 1.38 1.37 1.36 1.35 3 2 1.80 1.48 6 UNDERGROUND CARBON MONOXIDE MONITORING SYSTEM COMPONENTS SAGO MINE 1.81 5 1.100 KEY: 1.XX (Green Rectangle) Most Likely Functional at Time of Explosion Responded within Acceptable Limits 1.XX (Yellow Triangle) Most Likely Functional at Time of Explosion Did Not Respond within Acceptable Limits 1.82 1.29 1.33 1.99 1.XX 1.XX 1.XX (Yellow Oval) Most Likely Functional at Time of Explosion Accuracy Unknown (Yellow Explosion) Status at Time of Explosion Unknown, Damaged Accuracy Unknown (Red Stop Sign) Not Functioning Properly at Time of Explosion Did Not Respond within Acceptable Limits (Speaker Symbol) Red - Not providing audible or visual warnings at time of explosion Green - Providing audible and visual warnings at time of explosion Appendix I - Page 5 of 5 1.34 1.32 1.31 1 1.30 Appendix] - Bottom Mining Supplements to the Ventilation Plan U.S. Department of Labor Mine Safety and Health Administration 504 Cheat Road Morgantowa, West Virginia 25508 WITH FIELD DATE [17: SEP 2 2095 on.) Mr. Jeffrej,r K. Toler Superintendent Anker WV Mining Company, Inc. Route 9, Box 50? Buckhannon, West Virginia 26201 Dear Mr. Toler: The request filed September 23, 2005, for a test area as shown on the accompanying map for the ventilation and evaluation of the worked-out area as a result of mining the lower bench of the Middle Kittanning seam of the 2nd Left Mains at the Sago Mine, LD. No. has been reviewed and is approved. This information will be included in your currently approved mine ventilation plan. You are reminded that all changes or revisions to the mine ventilation plan, as speci?ed in 30 CFR 75.370fd), must be submitted to and approved in writing by this of?ce before they are implemented. If you have any questions, please feel free to contact this of?ce. Sincerely, Kevin G. Stricklin Kevin G. Stricklin District Manager EParrish:aew bcc: Bridgeport W. Ponceroff E. Parrish Health Section Map File ,Mai'n File Appendix] - Page 1 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan 1. . sea-'5 MerWest Virginina sesame E3 Compmy i September 28, 2005 Kevin Stricklin, District Manager Department of Labor, Mine Health and Safety Administration 604 Cheat Road Morgantown. WV 26508 Attn: Torn Hlavsa. Submittal 2a?2ventlFinal. Dear Mr. Stricklin: The following correspondence is concerning amending our Sago Mines, identi?cation number 46-08?91} approved ventilation control plan. These proposed amendments will allow recovery of additional resources, in that the lower bench of the Middle Kittanning seam that is being proposed to be mined. This mining application will apply to the lower coal seam of the :2"d Left Mains at the Sago Mine, l.D.No.46- 08791. Please refer the attached drawing (Number 1 Proposed Typical Ventilation Plan), which depicts the proposed ventilation plans for ventilating the area to be mined during the bottom split advancement. We also this time wish to utilize an evaluation point so as not to expose examiners to undue hazards of raised areas and heightened coal ribs. Be advised that we wish to respectfully submit for your review and subsequent approval a bleeder system for a non-pillared worked out area "Please refer to Evaluation Point Designation Plan ?so as not to expose examiners to undue hazards of raised areas and heightened coal ribs. In addition this amendment will include the ?Inactive Bleeder Systems and NonePillared Worked Out Areas? of the current approved ventilation control plan .The examiner will place his initials and date at the evaluation point and record the results in a book loted outside for that purpose. Appendix] - Page 2 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan 0 Page 2 September 28. 2005 It should noted that the proposed evaluation system is to be used only for a brief period of time as we plan to seal this area following the completion of the bottom split mining. Please refer to the attached list of ?Safety Provisions? that will address in detail safe work procedures for this mining process. in closing, your prompt review and approval of this proposed amendment will be greatly appreciated. If you have any questions oonceming this correspondence please feel free to contact me at 1-304- 471-3400. Al Schoonover Safety Director Appendix] - Page 3 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan DRAWING NO. 1 PROPOSED TYPICAL VENTILATION PLAN FOR BOTTOM SPLIT MINING SAGO MINE MSHA ID 46?08791 U-ZDI 6-98A Note: I .This may be repeated, as well as ul?rered due ?ro condl?rlons. Appendix] - Page 4 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan DRAWING NO. 3 TYPICAL CUT SEQUENCE FOR SECOND DEVELOPMENT WORK SAGO MINE MSHA ID 46?08791 RIB STABILIZATION PLAN STRATA PRODUCTS - (TOP VIEW) i i . 5.00 a: smg? )Esno 10.00, 10.00, $1 10.00, 10.00, Lu}? 10.005 10.00. a5? 10.00, 10.00? a< 3% 10.005 10.00? 5.001%( M.Sme Noie 1: P0515 will be -. Nofe 2: Wooden Headers Ins?fdlled 0" FI- 8 and foo?rers be ulilized In the Immediate work area. a] on The Lock?N?Loads Ln Loco?rion of Lock?N?Load Supporf SIDE VIEW OF SIDES AacC CURRENT MINED AREA - RE BOTTOM SPLIT TO BE SIDE VIEW OF SIDES CURRENT MINED AREA BOTTOM SPLIT TO BE MINED 5? .5 No?re 3: Rib Supporls will be lef?r in place. Appendix] - Page 5 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan EVALUATION POINT DESIGNATION PLAN SAGO MINE FTJR 2ND LEFT MAINS AREA MSHA LD. 46?08791 Ml EP (Worked Ou?r Monitoring Poin?r) Stopping L_Ine Wqu BE TEST AREA I In Of Lost Open egresa FOR 2ND LEFT MAINS --, I I m3 PS Well Location (6/04/04) 4?-09??o1251 (Prod. Gas) EP J. @?Jlntake Evaluation Point) Appendix] - Page 6 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan Anker Mining Company, Inc. Page 3 Sago Mine Bleeder System A description of the Future bleeder system to be used is shown on the mine ventilation map submitted in accordance with 30 CFR 75.372. The description includes; the bleeder system design, the location of the evaluation points for measurement of methane and oxygen concentrations and for test air quantity' and direction, and the location of ventilation devices such as regulators, stoppings, and bleeder connectors used to control air movement through worked out areas. Active Bleeder Systems: Certi?ed personnel designated by the operator will travel to the location of evaluation points and measuring points. Bleedcr entries will be examined by traveling to the point of furthest penetration from the B.E.P. to check the quality of air. These travels will be made at least every seven days to determine the effectiveness of the bleeder system. The examinations will consist of measurements for methane, oxygen de?ciency, air quality and a determination whether the air is ?owing in the proper direction. At each underground monitoring point location the name of the monitoring point as well as the direction of the air?ow will be identi?ed. The examiner will place his initials and date at the evaluation point and record the results in a book located outside for the purpose. The examiner will notify the Shi? or General Foreman inunediately of signi?cant changes (reversal of air ?ow direction, changes of more than 25% in the quantity of air, or more than 1% change in the content of methane or oxygen). If warranted, the Shift or General Mine Foreman will make an investigation into the cause of the changes and take action to correct any hazardous conditions found. This action will be recorded in the appropriate book on the Surface. Bleeder entries will be maintained free of obstructions through the use of: posts and cribs, to control the roof; and through ditches andfor dewatering pumps, to control water. Prior to intersecting accessible areas such as bleeder entries or other splits of air, precautions will be taken to avoid adversely affecting the mine ventilation such as building stoppings, hanging check curtains, building andfor adjusting regulators. Inactive Bleeder Systems and Non-Pillared Worked Out Areas: Certi?ed personnel designated by the operator will travel the perimeter of non-pillared worked out areas at least every seven days, examining for methane, oxygen de?ciency, air quantity, air flowing in the proper direction, and hazardous conditions. These measurements shall be made at approved evaluation points andtor measurement point locations. The examiner will place his initials and date at the evaluation point and record the results in a book located outside for the purpose. All approved evaluation point andtor measurement point locations, shall, at all times, be maintained in a safe condition. Any hazardous condition will be recorded in a hook located outside for that purpose. For the purpose of ventilation of structures, area or installations that are required to be ventilated to return air courses, and for ventilation of seals, other air courses designated as return air courses are shown on the mine ventilation map submitted in accordance with SDCF 75.3?2. The location, if different from that submitted on the mine ventilation map, and sequence of construction of proposed seals will be submitted to the District Manager and approved prior to the construction of seals. Appendix] - Page 7 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan Sago Mine LD. Number 46-08791 Safety Provisions: Note: The safety provisions listed below will be reviewed with all persons working in the affected area prior to commencing work and record there of made. 1. No person will be allowed inby the second mining area so as to eliminate exposure of persons to heightened coal ribs. 2. The Shuttle car operator will remain under the protective canopy at all times while inby the second mining area. 3. The Shuttle cars will be equipped with ?Back Boards? so as to protect the operator from lateral material falls. (Refer to the Attached Equipment Schematic) Seetwt?ewel?w 4. All access points to raised areas created by second mining will be dangered off with yellow ribbon or equivalent marterial.The ribbon will be affixed from rib to rib. and noted in the pre?shift Ion-shift examination book. 5. Tests for methane gas will be conducted prior to the cutting and loading of coal and every 20 minutes there after by remote means. This may be accomplished by utilizing a remote probe or by traveling inby on the upper level parallel and above the area to be mined. 6. In the event mining equipment becomes disabled the ribs will be supported prior to commencing repairs to said piece of equipment. All work will be conducted under the direct supervision of a W.V. certi?ed underground mine foreman. 7. Cable handling will be accomplished via remote means utilizing pull ropes and additional personnel if needed. At no time will persons go inby to accomplish this task unless the coal ribs are supported. 8. The lower level mining entries will not be wider than the upper level. Appendix] - Page 8 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan U.5. Department of Labor Mine Safety and Health Administration 604 Cheat Road Mergamown, West Virginia 26508 uuoesosouso HINE FILE moron Dismissal: WITH new ornce: 2035 m, I SM 1 I Mr. Ieffrey K. Toler 1/ a: I Superintendent Anker West Virginia Mining Company, lnc. WU Route 9, Box 50? Buckhannon, West Virginia 26201 Dear Mr. Toler: The request filed October 4, 2005, to extend the test area as shown on the accompanying map for the ventilation and evaluation of the worked-out area as a result of additional mining of the lower bench of the Middle Kittanriing seam of the 2"d Left Mains at the Sago Mine,r ID. No. 46-08791, has been reviewed and is approved. This information will be included in your currently approved mine ventilation plan. You are reminded that all changes or revisions to the mine ventilation plan, as specified in 30 CFR 75.3.70 must be submitted to and approved in writing by this office before they are implemented If you have any questions, please feel free to Contact this office. Sincerely, Kevin G. Strickiin Kevin G. Stricklin District Manager ]Hayes:si bcc: Bridgeport Field Office (2) Wi Ponceroff I. Hayes Map File g/r'Main File Appendix] - Page 9 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan MWWesW-mhia MINE MiningGompany OCT 4 2005 0mm, 3' 2005 Kevin Stricklin. Distn'ct Manager 1 CID Department of Labor, Mine Health and Safety Administration 604 Cheat Road Morgantown, 26508 Attn: Tom Hlaysa Submittal 3. Dear Mr. Stricklin: Anker West Virginia Mining Company wishes to amend our September 27, 2005 submittal which allowed our Sago Mine, (MSHA ID 46-08791), and more speci?cally our Left Mains unit. to mine the lower bench of the Middle Kittanning Seam. We wish to modify this plan to allow for additional minin in this area. This additional area is shown on the attached map, and displayed and denoted with hatching. It should be noted that we will comply with all details and information complied in the September 27, 2005 submittal. it should also be noted that we have moved both the intake. as well as the return monitoring points, and EP-ZMIS outby so as to cover the additional area we plan to add. if you have any questions concerning this correspondence please feel free to contact me at 1-304-471-3300. inoerely, . AI Schoonover Salety Director Appendix] - Page 10 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan Sago Mine MSHA ID. Number 46-0878]; ID No. U?201ti-98A Safety Provisions: Note: The safety provisions listed below will be reviewed with all persons working in the affected area prior to commencing work and record there of made. 2. it. No person will be allowed inbv the second mining area so as to eliminate exposure of persons to heightened coal ribs. The shuttle car operator will be remain under the protective canopy at all times while inblv the second mining area. The Shuttle Car will be equipped with ?Back Boards" so as to protect the operator from lateral material falls. (refer to the Attached Equipment Schematic). All access points to raised areas created by second mining will be dangered off with yellow ribbon or equivalent material. The ribbtm will be af?xed from rib to rib, and noted in the pre-shiftlon-shift examination book. Tests methane gas will be conducted prior to cutting and loading of coal and every 20 minutes there after by remote means. This will be accomplished by utilizing a remote probe or by traveling inby on the upper level parallel and above the area to be mined. In the event mining equipment becomes disabled the ribs will be supported prior to commencing repairs to said piece of equipment. All work will be conducted under the direct supervisions of a WM. certified underground mine foreman. Cable handling will be accomplished via remote means utilizing pull ropes and additional personnel ifneeded. At no time will persons go inhy to accomplish this task unless the coal ribs are supported. The lower level mining entries will not be wider that the upper level. Persons will be withdrawn from the immediate area during second advance mining in the event of loose and or overhanging ribs are encountered. Dutbv the line depicted as on the attached map, additional ribi?roof support wilI be added so as to provide additional roof support for the miner operator. This will be accomplished utilizing one of the methods shown below: We will position one ofour twin-head roof bolter in a crosscut to a point where the ATRS support is set at the junction ofthe crosscut and entry. Once the ATRS is set the roof bolters operator's canopy. nearest the corner in which the miner operator is going to position himself to operate1 will be swung towards the inbju,r comer and rib area. In doing such. this will create a protected area where?hy the miner operator can operate the continuous miner from. This support will remain in place until the miner operator has completed the cut and has safely positioned himself in the main entry away, outby from the intersection. Either 2. (two). Prop-setter supports or 2. {two} Lock-N-Load Supports will be installed on 5. (five) foot centers. with screen meshing being attached on the irtbg,r side. These supports will be installed with wedges being driven from the outby portion of the support towards the inbv corner or rib line. By installing these supports in this fashion in conjunction with a removal rope. these supports can be remoter removed by using a scoop to safely remove these devices. Once removed. the rope. which had been previouslyr attached to the sccop can be pulled taught in order to remove these supports to the middle of the intersection where they can be safely recovered. Appendix] - Page 11 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan Either the top will be screened to cover an area approximately 4? 12?, and installed utilizing 4, (four) renf'bolts. Appendix] - Page 12 of 26 Appendix] - Page 13 of 26 see sofefy precaution number 10. i Gutby the fine Iabeled A provisions, ER 2mg ?Whit: gill 1 B11 a! U. litmuiixlii. :3 z. 1 Proj Projected Test Area .t h. if GEM, in..th hi, MINE MSHA LD. 46*0879?! To Be Added 3. U- ?i?lultl?l=li 3 WVOMHS8CT LD. ec?red Test Area 2ND LEFT R. i-vi A Appendix] - Bottom Mining Supplements to the Ventilation Plan AREA Appendix] - Bottom Mining Supplements to the Ventilation Plan Department of Labor Mine Safety and Heaith Administration 604 Cheat Road Morgantown, West Virginia 26503 SENT TO ANWOR DISCUSSED WITH FIELD OFFICE: Humans A DATE glimmer: sir: 2 1 bait? m3. farm?s THE Mr. Ieftrey K. Toler Superintendent Anker WV Mining Company, lnc. . Route 9, Box 50? Buckhannon, West Virginia 26201 Dear Mr. Toler: The request ?led October 2005, for a test area as shown on the accompanying map for the ventilation and evaluation of the worked~out area as a result of mining the lower bench of the Middle Kittanning coal seam in the A?Panel at the Sago Mine, ID. No. 46?08?91, has been reviewed and is approved. This information will be included in your currently approved mine ventilation plan. You are reminded that all changes or revisions to the mine ventilation plan, as specified in 30 CFR must be submitted to and approved in writing by this office before they are implemented. If you have any questions, please feel free to contact this of?ce. Sincerely, Kevin 13. Strickiin Kevin G. Stricklin District Manager JHayes:aew bcc: Bridgeport W. Ponceroff E. Parrish I. Hayes Health Section Map File .M'ain File Appendix] - Page 14 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan EALTH Nicerme MINE - - 3 OCT I 7 FEES ADMINISTRATION October 16, 2005 Kevin Stricldin, District Manager CIO Department of Labor, Mine Health and Safety Administration 604 Cheat Road WV 26508 Attn: Nelson Blake, Tom Hiavsa. Submittal Dear Mr. Stricklin: The following correspondence is concerning the second mining of our Sago Mine, identi?cation number 46-08791 8; State ID. U- We wish to respectfully request that a Test Area be approved for the A?Panei area of the Sago Mine for second mining of the lower bench of the Middle Kittanning Seam for both the entries and cross-cuts alike Refer to attachment labeled {Projected Test Area} which shows proposed ventilation circuits and evaluation points. For your information I have attached a detailed cut sequence map that will eliminate exposure of persons to heightened areas. A list of the safety precautions that have been successfully utilized in previously mined areas has been included that will be in effect during this application. All previously approved submittals concerning this mining application will still be in effect for this mining application. in closing, your prompt review and approval of this request will be greatly appreciated by this department. If you have any questions concerning this correspondence please feel free to contact me at 1604? 47'1-3442. Appendix] - Page 15 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan Projec?red Tes?r Area <74TEST AREA FOR AREA SAGO NHNE MSHA 46?08791 1.0. Appendix] - Page 16 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan Sago Mine MSIIA ID. Number 46-08?91; ID No. Safety Provisions: Note: The safety provisions listed below will be reviewed with all persons working in the affected area prior to commencing work and record there of made. 1. 2. No person will he allowed inby the second mining area so as to eliminate exposure of persons to heightened coal ribs. The shuttle car operator will be remain under the protective canOpy at all times while inby the second mining area. The Shuttle Car will be equipped with ?Back Boards? so as to protect the operator from lateral material falls. [refer to the Attached Equipment Schematic}. All access points to raised areas created by second will be dangered off with yellow ribbon d: or equivalent material. The ribbon will be af?xed from rib to rib, and noted in the pre-shifh?on?shi? examination book. . Tests for methane gas will be conducted prior to cutting and loading of coal and every 20 minutes there alter by remote means. This will be accomplished by utilizing a remote probe or by traveling inby on the upper level parallel and above the area to be mined. In the event mining equipment becomes disabled the ribs will be supported prior to commencing repairs to said piece of enuipment. All work will be conducted under the direct supervisions of a W.V. certi?ed underground mine foreman. Cable handling will be accomplished via remote means utilizing pull ropes and additional personnel if needed. At no time will persons go inby to accomplish this task unless the coal ribs are supported. The lower level mining entries will not be wider that the upper level. Persons will be withdrawn ?'om the immediate area dining second advance mining in the event of loose and or overhanging ribs are encountered. Ill. Outby the line depicted as on the attached map, additional riblroof support will be added so as to provide additional roof support for the miner operator. This will be accomplished utilizing one of the methods shown below: We will position one of our twin-head roof bolter in a crosscut to a point where the ATRS support is Set at the junction of the crosscut and entry. Once the ATRS is Set the roof bolters operator?s canopy, nearest the corner in which the miner operator is going to position himself to operate, will be swung towards the inhy corner and rib area. In doing such, this will create a protected area whereby the miner operator can operate the continuous miner item. This support will remain in place until the miner operator has completed the cut and has safely positioned himself in the main entry away, outlay ??om the intersection. Either 2, (two), Prop-setter supports or 2, (two) Lock-N-Load Supports will be installed on no more than 5. foot centers, with screen meshing being attached on the inby side. These supports will be installed with wedges being driven From the outlay portion of the support towards the inby comer or rib line. By installing these supports in this fashion in conjunction with a removal rope, these supports can be remotely removed by using a scoop to safely remove these devices. Once removed, the rope, which had been previously attached to the scoop can be pulled taught in order to remove these supports to the middle of the intersection where they can be safely recovered. Appendix] - Page 17 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan Either the top will be screened to cover an area approximately 4? 12?, and installed utilizing a minimum of 4, (four ?.)roof bolts. 11. During the ?rst cuts of Sequence (See Diagram the continuous miner operator can be positioned inlay the corner of Sequence provided the following measures have taken place: 0 Prior to starting the ?rst cuts a screen must be attached to at least two roof bolts along the row of roof bolts located closest to the right hand rib. Attachment can be by means of running a cable hanger through the screen and connect it to the hanger loop in the roof bolt plate. I Once this is completed, either two Prop-Setter Supports or two Lock-N- Load supports will be installed as close as possible to the rib and underneath the screen. By installing these supports in this fashion the screen will be forced to the top, as well as towards the rib line. - A?er the above actions have been completed the continuous miner operator can be taking the ?rst cuts ?'om Sequence 0 Removal of the screen and posts will occur as follows I First the cable hooks will be unhooked from the roof bolt plates; then, I We will follow the removal action described in Item #10 above, with the exception that continuous miner may also be used to remotely remove the temporary supports. Appendix] - Page 18 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan DIAGRAM #1 PRCIPUSED SCREEN AND POST LOCATION SCREEN (2) PROF SFTTER DR (2) sman LINE Huxinun FL Between Rattan 01" Screen and Floor FLOQR 1 PROWSED MINER OPERATGR Lem TION (mp. FIRST CUTS OF SEQUENCE #1 MINER OPERATOR HST CUTS 01" SEQUENE #1 SECOND CUTS 0F SEQUENCE #1 LOCATION MINER OPERATOR ECOND CUTS 0F SEQUENCE #1 Appendix] - Page 19 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan vpun-o - Ii:- II - ?!1Snr?r I - . . I Gin-erreegm - 5'15? 559'] . i .. I I drama mun-Insist: 5' L?ylezo-??Imam :l L1 I 'I'l Ihz-n m'lgl?'Ielile?-r.M "a - -1 hunk?u: - . - ?Iran-21mm'lJ?u-rm-onume - - . I. T11 3- - ?u??amPEI non-21.30109 rcml-pl'mumn I m" lgi?'? - -?L0'm?ilbat?: . ?5 magma-Inuit:35 25-3}? - inn-II I-?lanIBA-r?H1551 Flu-l: rem ml - manual-In?u- mn??muwpa?wuamln ,g mmnr-m-amm-mww I . tuba-Iain .. . . 11.mulai?-? "lJ .. . - In 1:4 FIND HI . nu ma: ?fn inynacniwirlaunru 1-01-1140!- - I . I ilu'lrlIt): cup in 1. - . Irv-Him! umuw- '1 mar: um. i 3 m?n A In?: H1: wins Di!? (I) 5131515?5 mug33?s?. at" shit-"IDAppendix] - Page 20 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan DRAWING NO. 3 I TYPICAL CUT SEQUENCE FOR SECOND DEVELOPMENT WORK SAGO MINE MSHA ID 46?03791 wvowsm RIB STABILIZATION PLAN STRATA PRODUCTS - (TOP VIEW) Shoo, 10.00? 10.00? H) 10.00' 10.00? :5 . 10.00 10.00 a I 10.00? 10.00 a 10.00' 10.00? 5.00"a 5.00? . Note 1: Posts wilt be - Note 2: Wooden Headers and footers be on the Lock-N?Loads In E: in ihe Immediate work area. Li?) - Loca?rion of Lazku?N?Load Support i insIaIled on 4 Fi. Centers SIDE VIEW OF SIDES CURRENT MINED AREA I . . Viki?: \r?x?f <9 BOTTOM SPLIT TO BE MINED, SIDE VIEW OF SIDES 3&0 CURRENT MINED AREA BOTTOM SPLIT TO BE MINED {yr :11 I- 01', a! Noi?e 3: Rib Supperfs be 12H in place. Appendix] - Page 21 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan LIST or mantis Allin 55:51:13 mum Iii: . The LDCk~N~Loadm can be removed and reused by . I. -: Lanna-Loan amusmm SIJZPPOJIT Adjustable without any cutting of the timber and is available to ?t mining haughts from 3 fast to 15 feet (1m to Sm). Available in different and capacities of Either 5. or 20 tons. Can be installed by one person and may be preloaded using the simple wrench that can be purchas: 2500 ibsl?llkl?d) of preiaad can be app-lied. rereasmg the ciamps. The Lock-N~Loacl can be packaged with conventional cap blocks and header boards. In addition. 1.rai-ioius hu? Lulu: i ~14mm 'ii??ahl - I steel fittings are available to tie into steei or wooden beams. - The Lock-N-Load can be applied in place of steel jacks, water props, or posts as either a temporal?)? ar permanent supp-art. It can also he used as formwork for stoppings, seals. barricades and uentilatinn curtains. For use ii: inn; min a UN ill Inga. More: The nonayi'eiding Lack-N-and is not dassi?ea? as a roof support ands. 30. LDEK-N-LDAD SPECIFICATIONS 5 TON SUPPORT CAPACITY Part Lock 513-5 Lock SM-E Closed Haight 3 ft. 4 Ft. Loci: 5,354 5 ft. 51?6-8 Appendix] - Page 22 of 26 5ft. Open Height Weight 6ft. 3? 8ft. 19 Ibs. 24 lbs. 33 lbs. Appendix] - Bottom Mining Supplements to the Ventilation Plan LDCK-N-LOAD 8 TON SUPPORT CAPACITY Part Ciosed Height Open Height Weight 3 . 5 ft. Lack Fl: 26 lbs. Lock SM-E 4 Ft. 6 ft. 32 tbs. Loch 5 Ft. 7 ft. 39 lbs. Lack 3/6-3 6 Ft. 3 Pt. 45 lbs. Loco: SKI-9 Ft. 9 ft. 52 Ebs. LUCK SIB-10 3 ft. 10 ft. 53 Has. Lock SKID-12 10 Pt. 12 ft. F1 lbs. 20 TON SUPPORT CAPACITY Part Closed Height Open Height Weight Lock 203-5 3 ft. 5 ft. 54 tbs. LOCK ZOE-E- 4 ft. 6 ft. 67 lbs. Lack 205-? 5 ft. 7 ft. 30 lbs. Lock EDIE-8 6 fr. 5 93 ibs. Lock 3" Pt. 9 ft. 106 lbs. Lock 3 R. 10 ft. 118 lbs. Lem: 9 ft. 11 Ft. 131 lbs. Lock 20310-12 10 ft. 12 ft. 144 lbs.fINCHES -- 2.5 13.5 minus 2.5 i 3.5 in Lemon: h.5 fr. high 15 n. height D's-aran Adana 33f fiie a? Lack-N-Load product sheet. 51:33 dome Strata Mme 5'!me Request l-Ir?relnformatrun 1 New; .5 Appendix] - Page 23 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan U.S. Department of Labor Mine Safety and Health Administration 604 Cheat Road Morgsmown, West Virginia 26505 unease ac $451513; 1 SENT TO ANDFOFI DISCUSSED FIELD OF FIG DEC I 9 2305 12:3 1 ME Mr. Jeffrey K. Toler Superintendent Anker West Virginia Mining Company, Inc. Route 9, Box 50? Buckhannon, West Virginia 26201 Dear Mr. Toler: The request filed December 2005, for a test area as shown in red on the accompanying map for the ventilation, evaluation to mine the lower bench of the Middle Kittanning seam and future seal locations of the A-Z Panel at the Sago Mine, ID. No. 46-08?91, has been reviewed and is approved. This information will be included in your currently approved mine ventilation plan. You are reminded that all changes or revisions to the mine ventilation plan, as Specified in 38 CFR 175.370 must he submitted to and approved in writing by this office before the);r are implemented. If you have any questions, please feel free to contact this office. Sincerely, cinema Kevin G. Stricklin District Manager EParrish:si bec: Bridgeport Field Office W- Ponceroff E. Parrish Health Group Map File g?M-a-in File Appendix] - Page 24 of 26 Appendix] - Bottom Mining Supplements to the Ventilation Plan ANKER WEST VIRGINIA was I We eucxHANrionNmzoi in i' i9 November 30. 2005 -- Mr. Kevin Stricklin MSHA 604 Cheat Road Morgantowo. WV 26508 Dear Mr. Stricklin: The following correspondence is concerning the second mining of our Sago Mine. (MSHA D. No. 46-08791 State l. D. No. U-2016-QBB). We wish to respectfully submit an amendment to our current approved ventilation and roof control plans for the AZ-F'anel area of the Sago Mine for second mining of the lower bench of the Middle Kittanning Seam for both the entries and cross-cuts alike. Refer to attachment labeled (Projected Area) which shows proposed ventilation circuits and evaluation points and future seal locations once the panel is abandoned. Note: In the set of seals labeled 1 through 5, seals 1 and 5 will be built last, and in the set of seals labeled 6 through 1D. seals 6 and 10 will be built last. For your information I have attached a detailed out sequence map that will eliminate exposure of persons to heightened areas. A list of the safety precautions that have been successfully utilized in previously mined areas has been included that will be in effect during this application. All previously approved submittals concerning this mining appiition will still be in effect for this mining application. in closing, your prompt review and approval of this request will be greatly appreciated by this department. if you have any questions conceming this correspondence picase feel free to contact me at 1404-4171-3303. John B. Ste-mole Jr. ,2 will. in . We. Assistant Director of Safety And Employee Development Appendix] - Page 25 of 26 Appendix] - Page 26 of 26 PROPOSED SEALS SAGO MINE PROJECTED AREA FOR A-Z PANEL MSHA LD. 46?0379'1' u?zms?ga Projected Area EP - Appendix] - Bottom Mining Supplements to the Ventilation Plan Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals Appendix K - Page 1 of 18 Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals Appendix K - Page 2 of 18 Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals Appendix K - Page 3 of 18 Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals Appendix K - Page 4 of 18 Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals ` Appendix K - Page 5 of 18 Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals Appendix K - Page 6 of 18 Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals Appendix K - Page 7 of 18 Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals Appendix K - Page 8 of 18 Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals Appendix K - Page 9 of 18 Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals Appendix ____ Appendix K - Page 10 of 18 Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals Appendix K - Page 11 of 18 Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals Appendix K - Page 12 of 18 Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals Appendix K - Page 13 of 18 Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals Appendix K - Page 14 of 18 Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals Appendix K - Page 15 of 18 Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals Appendix K - Page 16 of 18 Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals Appendix K - Page 17 of 18 Appendix K - Three Supplements to the Ventilation Plan Concerning Omega Block Seals Appendix K - Page 18 of 18 Appendix L Sago Mine, MSHA ID 46-08791 Wolf Run Mining Company Pre-Explosion Simulation of the Mine Ventilation System Appendix M Sago Mine, MSHA ID 46-08791 Wolf Run Mining Company Post-Explosion Simulation of the Mine Ventilation System with the Damaged Ventilation Controls Appendix N Sago Mine, MSHA ID 46-08791 Wolf Run Mining Company Post-Explosion Simulation of the Mine Ventilation System with the Initial Repairs made to the Damaged Ventilation Controls Appendix O - Evaluation of Potential for a Roof Fall to Ignite a Methane-Air Mixture u .S. Department of Labor Mine Safety and Health Administration Pittsburgh Safety & Health Technology Center P.O. Box 18233 r') Pittsburgh, PA 15236 Roof Control Division September 7, 2006 MEMORANDUM FOR RICHARD A. GATES District Manager, CMS&H District 11 ~Z~ . THROUGH: KELVIN K. WU ~?~~4' Acting Chief, Pittsburgh Safety and Health Technology Center M. TERRYH9e:~rv~- . Chief, Roof Control Division ~7yi¡~-:b~~~~ FROM: MICHAEL GAUNÁ I,/. ,,7. Mining Engieer, Roof Control Division " íLI. .fl U/~' ~;~OOK -: ( Mining Engineer, Roof Control Division SUBJECT: Evaluation of the Potential for a Roof Fall to Ignte a Methane- Air Mixtue at the Wolf Run Mining Company, Sago Mine, Upshur County, West Virginia, MSHA 1. D. No. 46-08791 An explosion initiated in the sealed 2 Left area in the northern portion of the Sago Mine on January 2, 2006. Maps indicate that three roof falls occurred in this area prior to seal construction. Examinations after the explosion ctetermined that additional roof falls had occurred that were not shown on the mine maps. The precise timing of these falls relative to the mine explosion is not known. The Sago Accident Investigation Team requested that Roof Control Division (RCD) personnel assess the likelihood that these roof falls ignited explosive concentrations of methane at the Sago Mine. The RCD evaluated the possibilty of a roof fall initiation through background literature searches and in-mine investigations. (~, Appendix O - Page 1 of 10 Appendix O - Evaluation of Potential for a Roof Fall to Ignite a Methane-Air Mixture 2 Background Roof Control - The primary roof support consisted of %-in. x 6-ft., fully grouted bolts on approximately 4-ft. centers. The bolts were installed with 8- x 8-in. bearing plates which were typically supplemented with larger "Spider" or "Pizza Pan" plates for additional surface control. In some areas, welded wire mesh was installed with the 8- x 8-in. plates for improved roof surface control. Cable bolts also were noted in the sealed 2 Left area. In the areas investigated by RCD, the cable bolts were only used cribs and wood Propsetter standing supports also had been used on an infrequent basis. Explosive forces had warped and folded the "Spider" and "Pizza Pan" plates, torn welded wire mesh from the roof in places, and dislodged wood supports. occasionally and there was evidence that wood Pilar stabilty was evaluated using Analysis of Retreat Mining Pilar Stabilty (ARMPS) 55- x 80-ft. center pilar, 18-ft. mining width, 15-ft. bench software. For the typical mining height, and 320-ft. overburden, the pilar stabilty factor (SF) is 2.2. The effective pilar stabilty is actually higher because the 15-ft. mined height only applies to the panel entries and not the crosscuts. The crosscut mining height of only 7 to 8 ft. serves to reinforce and improve the pilar stabilty. In the areas traveled, no evidence of abnormal pilar stress or pilar dilation was encountered. This observation is consistent with the satisfactory SF value. The pilar rib conditions in the entries and crosscuts appeared to be stable. 2 Left Roof Falls Mining was completed in 2 Left in late October and the seals were completed on December 11, 2005. Prior to the January 2nd explosion, three pre-sealing roof falls had been identified on the mine map. Roof Control Division personnel visited the mine on January 30, 2006, and observed that these three pre-sealing roof falls had extended (see Drawing 1). Also, four additional roof falls were observed that were not shown on the mine map prior to seal completion (see Drawing 1 green shaded falls labeled "Before 1/27/06"). It is not known exactly when these four newer roof falls occurred. The roof fall areas observed were consistent with roof fall information collected by other investigators during initial exploration on January 27, 2006. Roof Control Division personnel again observed the 2 Left area on May 11, 2006, and found additional roof falls that were not present on January 27 or 30 (see Drawing 1 purple shaded falls labeled" After 1/27/06"). Other investigators have determined that the explosive forces propagated in every direction from the area near surveying spads 4010, 4011, 4047, and 4048 (see Drawing 1). The seven roof falls that were observed during the January 30 investigation range in distance from approximately 150 ft. to 470 ft. from this area. The Appendix O - Page 2 of 10 Appendix O - Evaluation of Potential for a Roof Fall to Ignite a Methane-Air Mixture 3 rubble and exposed fall cavity of the five closest roof falls (within 440 ft.) were inspected. Access to the two roof falls beyond 450 ft., was obstructed by deep water in bench mined entries. The roof falls extended 7 to 12 ft. above the mining horizon. Gray shale was the predominant rock type visible in the fall rubble and in the exposed cavity of the roof falls. However, thinly bedded sandstone beds, interspersed with shale layers were exposed at the top of the fall rubble, roughly 8 to 12 ft. into the immediate roof in three locations (see Drawing 1). The fall rubble consisted of rock slabs of varying thickness and geometry. The falls encompassed the entire entry width and primarily affected the entries and adjoining intersection(s) as opposed to crosscuts. Thus, there appears to be a general tendency for north-south migration of the roof fall areas (see Drawing 1). Roof support in the vicinity of the roof falls consisted of %-in.-diameter, 6-ft.-Iong, fully grouted resin bolts installed with 8- x 8-in. roof bearing plates and "Spider" or "Pizza Pan" plates. Cable bolts were installed near some of these roof falls, wire mesh had been installed near the perimeter of two of the roof fall cavities, and wire mesh was noted under the fall rubble of a third roof fall. The fully grouted bolts were the only roof support that could be observed within the roof fall rubble. Geology The Sago Mine is developed in the Middle Kittanning coal seam. The overburden, measured from the base of the seam to the surface, ranges from 230 to 320 ft. in 2 Left and the immediate roof consists of gray shale grading upward into sandy shale and sandstone with shale bedding. Exploratory Dril Hole SF17-97 is situated immediately adjacent to the sealed area (Drawing 1). Dril core from this hole was used to assess the stratigraphy above the Middle Kittanning coal seam (see Table 1). The roof falls noted in the course of the investigations are within an 800-ft. radius of this hole. It is reasonably likely that the same sequence of units is present above the coal seam in the vicinity of the roof falls in the sealed area. Coal measure geology is known to change substantially over short distances (e.g. due to depositional features such as sand channels). However, the rubble observed in the falls appeared to be gray shale overlain by bedded sandstone (i.e. generally consistent with Table 1). Table 1 provides an example of the thickness of lithologic units, the distance from the top of the Middle Kittanning seam, and the distance from the top of the typical mining horizon to the lithologic units based on information from Dril Hole SF17-97. In much of 2 Left, 3 to 5 ft. of shale roof (3.6 ft. individual average) typically was mined with the coaL. Appendix O - Page 3 of 10 ~ Appendix O - Evaluation of Potential for a Roof Fall to Ignite a Methane-Air Mixture 4 Table 1 Dril Hole SF17-97 Lithology I fImmediate40f t.fR 0 E xamp. e 0 Lithologic Description f ba ove M'ddl 00 1 eKi ttannng Coais eam Thickness, Distance to Distance to ft. Lithologic Lithologic Unit Unit from Top from Top of of Coal Seam, Mining Horizon(l), ft. ft. Dark Gray Shale Dark Gray Sandy Shale Shale Dark Gray Shale Sandstone with Shale Streaks Dark Gray Sandy Shale Dark Gray Shale 15.70 9.30 5.30 5.40 3.30 7.20 8.30 Shale 41.39 32.09 26.79 21.39 18.09 10.89 2.59 37.8 28.5 23.2 17.8 14.5 7.3 Top of Mining Typically Within this Unit Typically Mined 2.59 0 Bone - top unit of coal seam 0.30 Note (1) = Top of mining at 3.6 ft. average depth into overlying shale Shale Description - Shale samples from the immediate roof in the vicinity of spad 4010 were studied microscopically for the Sago Mine explosion investigation. The samples were classified based on grain size and bedding spacing as "laminated siltstone" according to Potter's 1980 textural classification of shales. They are characterized by very similar textures having a matrix composed of very fine-grained (0.005-0.2 mm) muscovite lathes, which are randomly oriented, but arranged in thin bedding layers. Contacts between adjacent bedding layers are gradational, defined by different grain sizes or mineral contents. The very fine-grained, muscovite-dominated layers host approximately 8-12% angular quartz grains, which are approximately 0.01 mm in diameter and isolated by the surrounding matrix. Coarser-grained layers are dominated by angular quartz grains, which are approximately 0.1 mm in diameter and touch along tangential contacts to leave angular interstices that are filed with finergrained muscovite. The very finest-grained layers host very fine-grained, clay sized (0:0.003 mm) muscovite with no quartz, and represent planes of preferential weakness along which delamination preferentially occurs. Sandstone Description - Three sandstone samples (RCD-SSA, RCD-SSC, and RCD-SSD) collected from the fringe of the roof fall rubble are described below. The sample locations are depicted in Drawing 1. Appendix O - Page 4 of 10 Appendix O - Evaluation of Potential for a Roof Fall to Ignite a Methane-Air Mixture 5 Sample RCD-SSA is characterized by Ij16-in. to 1j8-in. crossbedded laminations of light-colored, fine-grained quartz sandstone that form beds % in. to 1/2 in. thick, and are bounded by 1j64-in. dark-colored laminations that host abundant muscovite and biotite flakes. The sandstone laminations are well indurated, although scratch marks from a knife blade are visible. Sandstone laminations commonly pinch down from % in. to 1j16 in. over a distance of 3 in., to be bounded by dark-colored micaceous laminations. Sample RCD-SSC is characterized by Ij16-in. to 1j32-in.laminations of alternating light-colored, fine-grained quartz sandstone and dark-colored siltstone. The light- colored quartz sandstone laminations are well indurated, and alternate with moderately indurated dark-colored siltstone laminations, which host very fine-grained flakes of biotite mica. Fine-grained flakes of muscovite mica are commonly distributed within the light-colored quartz laminations, which may also host microcline or orthoclase grains, due to a faint pink tint. Very thin (lj64-in.) carbonaceous bedding partings are distributed at approximate 11/2-in. intervals. Laminations of all compositions can be easily scratched with a knife blade, indicating that quartz grains are not sutured. Sample RCD-SSD is characterized by 1jI6-in. to 1j8-in. crossbedded laminations of light-colored, fine-grained quartz sandstone that alternate with Ij32-in. dark-colored laminations of very fine-grained siltstone, which hosts abundant 1j16-in. flakes of muscovite mica. The sample also hosts a %-in.-thick bed of fine-grained, dark-colored, well indurated siltstone that hosts fine-grained biotite and muscovite mica, and contains Ij 64-in. stringers of light-colored quartz siltstone. The entire sample is approximately 2 in. thick, and is bounded by muscovite-rich bedding partings. Historical Research on Roof Falls and Ignitions The majority of methane-air ignitions can be attibuted to frictional ignitions by some form of machine or mechanical action. (d,f) However, within the time frame from 1960 to present, four instances were found where the most likely source for the ignition was a roof fall (c, d, k, 1). One instance involved a roof fall on a mining section and three instances referred to falls of ground within the extracted area of a longwall paneL. The precise ignition mechanisms could not be determined conclusively. However, the most likely scenario from these cases was determined to be ignition through rock-on-rock frictional forces. The factors involving ignition from roof falls have been studied with laboratory testing where the ignition capabilty (incendivity) of both mine roof rock and steel roof support materials were investigated. In addition, the igntion potential from compression of methane-air-coal dust mixtures has been studied in the laboratory. Appendix O - Page 5 of 10 Appendix O - Evaluation of Potential for a Roof Fall to Ignite a Methane-Air Mixture 6 Steel Roof Support Incendivity - Tests have been performed in which roof bolts and cable bolts were broken in tension and roof bolt heads were pulled through plates in an explosive methane-air mixture. Tests on roof bolts and plates produced no sparks or ignitions (c). However, sparking was observed in tests on cable bolts. In fact, sparking had been observed from breaking cable bolts in underground coal mines in the U. S. in the early 1990's. In response to these observations, laboratory testing was conducted to assess cable bolt failure incendivity. The test results indicated that although sparks are produced by breaking cable bolts these sparks are not hot enough, not large enough and are not of sufficient duration to ignite an explosive methane-air mixture (a, b). Tests have also been performed to try to determine the possibilty of igniting an explosive methane-air mixture by impact friction. These tests evaluated the incendivity of various combinations of materials when impacted together (i.e. by dropping one from a fixed height onto another). Samples included sandstone, shale, roof bolt steel and aluminum. Several combinations produced sparks, but the only ignitions were initiated by dropping aluminum on a rusty steel plate (c). Despite these findings, however, the researchers determined that sparks from failng steel roof supports cannot be conclusively ruled-out as an ignition source because of the limitations of laboratory testing simulating the actual underground environment. Rock-on-Rock Frictional Incendivity - Laboratory work indicates that specific rock types (e.g. sandstones) do have an incendivity potential (c,d,f). Studies have attempted to determine whether or not an ignition could occur due to heat andj or sparks produced by the friction of rocks rubbing together during a roof fall. In laboratory settings, two rock specimens have been rubbed together by pressing a rock against another rotating rock wheeL. Ignitions have been produced in these experiments with varying rock types under varying test conditions. Video records of these experiments indicate that the ignitions appeared to be from the heat trail behind the hot spot on the rocks and not the sparks that are produced (d). Rocks high in quartz content appear to be most susceptible to producing the friction required for heating but, rock composition is also a large factor (d). The study indicated that the rock composition, (ie. the overall proportion of quartz, feldspar and rock fragments in the grain framework) was a better indicator than quartz content alone of the incendivity of a particular rock (d). It has been noted that quartz-rich rock types (sandstones and quartzites) can produce a voltage when minutely deformed by applied mechanical stress. The mechanism known as piezoelectricity was discovered in 1880 by Pierre and Paul-Jacques Curie. They found that when certain types of crystals including quartz, tourmaline, and Rochelle salt, were compressed along certain axes, a voltage was produced on the surface of the crystal. In a piezoelectric crystal, the positive and negative electrical charges are separated, but symmetrically distributed, so that the crystal overall is electrically neutral. When a mechanical stress is applied, this symmetry is disturbed and the crystals are polarized, and the charge asymmetry generates a voltage across the Appendix O - Page 6 of 10 Appendix O - Evaluation of Potential for a Roof Fall to Ignite a Methane-Air Mixture 7 materiaL. The charge separation may be described as a resultant electric field and may be detected by a voltmeter as a voltage between the opposite crystal faces. The phenomenon of piezoelectricity is widely used in a variety of electronic devices, including igniters. Currently, synthetic material such as carefully prepared ceramics are used as igniters since they exhibit the most efficient piezoelectric properties (g,i). As a geologic phenomenon, piezoelectricity has been invoked to explain certain effects associated with earthquakes, such as "earthquake lights", the lightnng or fireballs that have been reported in the vicinity of earthquake epicenters. Piezoelectricity also has received some attention in the field of earthquake prediction, where some suggest that the mechanical stress imparted by shifting tectonic plates my induce voltages in rocks, which might be recorded as a precursor to earthquakes. Although it is thought that in rock types where crystals are randomly oriented the piezoelectric effect is self canceling, it may be that in rock types with preferentially oriented quartz crystals (such as gneiss or quartzite), such voltages may be generated (j. Methane-Air and Coal Dust Compression - Computer simulations have predicted that air temperature could increase rapidly to the point of igniting methane or coal dust laboratory tests simulated air compression from a confined fallng object and verified that ignitions could occur with certain methane and coal dust mixtures. The laboratory tests had no ignitions with any methane-air mixture in the absence of coal dust. Also, the numerical simulation for a full-scale mine scenario indicated that ignition could only be achieved with a fallng block of at least 65- x 65-ft. planar area fallng simultaneously (e). during a roof falL. Subsequently, Summary It is difficult to definitively exclude a roof fall as a potential ignition source for the explosion at Sago Mine. However, it appears to be an unlikely source for the following reasons: . Shale is the predominant rock type visible in the roof fall rubble. Specifically, the material referred to as shale is classified as "laminated siltstone" with low quartz content in a soft matrix that inhibits quartz grain-to-grain contact. This rock type is not as conducive to frictional heating or piezoelectric sparking as sandstones that have been suspected as ignition sources in roof falls (d). An exploration dril hole in the vicinity indicates that rock classified by core logging as sandstone exists above the mining horizon. Three roof fall cavities had sandstone beds exposed at the top of the fall rubble roughly 8 to 12 ft. into the immediate roof above the underlying shale. The samples collected from the roof fall rubble are a variety of sandstone that is micaceous, and characterized by thin, alternating laminations of fine sand, silt, and mica partings. In contrast, the sandstones associated with piezoelectric sparking and rock-on-rock frictional heating are Appendix O - Page 7 of 10 Appendix O - Evaluation of Potential for a Roof Fall to Ignite a Methane-Air Mixture 8 commonly considered to be dominated by quartz, exhibit stronger cementing or even quartz grain fusing (i.e. the metamorphic rock" quartzite"), and occur in more massive beds. Furthermore, the roof falls observed are outside the area where the explosion is inferred to have originated. Thus, rock-on-rock or piezoelectric ignitions are unlikely ignition sources. . The only metal roof supports noted in the fall rubble were fully grouted bolts and the wire mesh noted under the rubble of one fall. These steel roof support materials have not been associated with ignitions in experiments or in documented observations of gob ignitions. It was not possible to determine whether cable bolts noted near the roof falls could be hidden in the fall rubble. However, previous laboratory testing of the sparks from cable bolt failure did not ignite methane-air explosive mixtures. . All of the roof falls observed in the 2 Left seal area that were not noted on the mine maps prior to sealing, encompassed a much smaller area than the 65- x 65-ft. highly confined area required in computer simulations to ignite methane by compression. Attachment Appendix O - Page 8 of 10 Appendix O - Evaluation of Potential for a Roof Fall to Ignite a Methane-Air Mixture 9 References a. Mazzoni, RA., Brown, W.J., Carpetta, J.E., Spark Temperatures from 7-Strand Cable Bolts. Technical Support Roof Control Memorandum, December 19, 1994. b. Mazzoni, RA., Laboratory tests to evaluate cable bolt sparks as a possible methane ignition source. Technical Support Roof Control Memorandum, September 9, 1996. c. Nagy, J., Kawenski, E.M., Frictional Igntion of Gas During a Roof Fall. U.s. Bureau of Mines, RI 5548, 1960. d. Ward, C.R, Crouch, A., Cohen, D.R, Identification of potential for methane ignition by rock friction in Australian coal mines. International Journal of Coal Geology, 2001, pp. 91-103. e. Lin, W., The Ignition of Methane and Coal Dust by Air Compression - The Experimental Proof. Masters Thesis, Virginia Polytechnic Institute and State University, 1997. f. Powell, F., Bilinge, K, The Frictional Ignition Hazard associated with Collery Rocks. The Mining Engineer, 1975, pp. 527-533. g. http:j j en.wikipedia.orgjwikijPiezoelectricity h. http://webphysics.davidson.edu/ alumni! MiLeelTLab / Crystallography WW/piezo .htm 1. http:j j ww.britannica.comj eb j article-9059986 j. http:j j professionalmasters.science.orst.eduj Studentwebs j Mellonj ThesisOl J un04Final. pdf k. McKinney, R, Crocco, W., Tortorea, J. S., Wirth, G. J., Weaver, C. A., Beiter, D. A., Stephan, C. R, Report of Investigation, Underground Coal Mine Explosions, July 31 August 1,2000, Wilow Creek Mine - MSHA ID No. 42-02113, Plateau Mining Corporation, Helper, Carbon County, UT i. Carico, A.D., Methane IgnitionjExplosionjMine Fire Accident, February 14, 2005 at Buchanan Mine #1, Consolidation Coal Co., Mavisdale, Buchanan County, V A, ID No. 44-04856. Appendix O - Page 9 of 10 Appendix O - Evaluation of Potential for a Roof Fall to Ignite a Methane-Air Mixture Appendix O - Page 10 of 10 Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 U.S. Department of labor Mine Safety and Health Administration Pittsburgh Safety & Health Technology Center P.O. Box 1 B233 ( Pittsburgh. PA 15236 Roof Control Division August 31, 2006 MEMORANDUM FOR RICHARD A. GATES District Manager, CMS&H District 11 d" /:¿J . THROUGH: Kf~ K. WU /7~Æ ~ Acting Chief, Pittsburg~ty/ and Health Technology Center M. TERR~ A r:' 0i" l. Chief, Roof Control Division. / . C ~l. . í'~~l ,~, ~ /! l't1?it1..L4L. . I' _. '/L-jr'"- /IYU!..f FROM: SANDIN E. PHILLIPSON; i J Geologist, Roof Control Division ( SUBJECT: Evaluation of Features at Wolf Run Coal Company, Sago Mine, MSHA 1. D. No. 46-08791 Observations As requested by the MSHA Accident Investigation Team (Sago), observations of geologic features were performed in the formerly sealed 2nd Left Mains, in the vicinity of spad 4010 on February 21, 2006. The purpose of the observations was to evaluate and document two linear features in the mine roof in the vicinity of spad 4010. Observations were restricted to the #5, #6, and #7 Entries, between the 1st and 3rd Crosscut from the #1 Entry of the Main. The 2nd Left Mains are developed at an approximate 60° angle from the left side of the Mains, such that the first crosscut in the 2nd Left Main in the #6 Entry is actually the third crosscut in the #1 Entry. Observations began just inby spad 4010, in the #6 Entry, and proceeded down-grade into the next, benched intersection at spad 4047. The observation traverse proceeded east from spad 4010 into the #7 Entry through the intersection with spad 4011, and then inby along the benched #7 Entry for two crosscuts to the spad 4063 intersection. Observations continued in the unbenched crosscut between spads 4045 and 4047. l( Appendix P - Page 1 of 28 Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 2 Detailed observations concluded just inby the spad 4010 intersection, where the two linear roof features were scrutinize. A similar feature was briefly examined in the neighboring #5 Entr, just inby the spad 4028 intersection. The observation area is characterized by a variety of abundant structural geologic features and stress-related features. Abundant, very well developed joints were observed in the roof (Figure 1). The dominant joint set is oriented with a strke of N 85°E, and is characterize by nearly vertical joints that are spaced approximately 12-20 inches apart. Joints of ths set were preent across the entire observation area, from the spad 4010 intersection to the spad 4063 intersection, a distance of two crosscuts. Two minor, irregularly spaced sets of joints, oriented respectively at N 57°W and N 300E, are aligned parallel to the trend of slickenside planes. A prominent slickenside plane that controlled a zone of buckled roof strata was oriente N 300E, with a dip of 35° toward the southeast: and is located in the southeast corner of the spad 4047 intersection. A pair of slickenside planes, oriented N 67°W and dipping 50° NE, formed a linear, coffn-shaped roof cavity that trended through the spad 4045 intersection, crosscuttng a wide, deep horizontal stress pot-out. - r .. t l or ~ ll.. -:: g i I / .~ Fi~ 1. Very well developed joint set, characteri by N 85°E-strg joints spaced 12-20 inches apar. Photo taken in the croscut between spads 4010 and 401. Appendix P - Page 2 of 28 p Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 3 Horizontal stress pot-outs were common in the observed area, and were consistently r r oriented with a long axis aligned along a bearig of approximately N 5-7"E (Figure 2). Long-running cuttrs, localized at the intersection between the roof and rib, were consistently located along the west rib of the observed entres. In the #7 Entr, a longrunning cutter left the rib and crossed though the spad 4063 intersection along a bearing of N 100E. ...,l , '.,', .,.,'., i . . . ~ ' pi '", '&'; ~.' .J', ~;V. "~~~~';~:~:~ .1,' . Fip,e 2. Downward-bucked zone of thinly laminted shale represnts a stress pot-out tht follows a trnd of approxiately N 5-7"E. Other liear buckled zones of shale are aligned a1onr. the same bearing throur.hout the observed area. Ground conditions were partcularly degraded in the observed portion of the #7 Entr, with abundant stress pot-outs and cutters developed at the projected intersection of the mutually perpendicular slickenside planes. Detailed observations concluded just inby the spad 4010 intersection, where a small scaffold was constructed to reach the roof and observe two linear features that were present (Figure 3). Each linear feature was characterized by a pair of parallel ridges that trended across the exposed flat plane of the roof. One pair of parallel ridges was oriented along a bearing of N 43°E, while the other pair of parallel ridges was oriented along a bearing of N 700E. The parallel ridges were spaced approximately 2-3 inches apart, and protrded approximately % inch below the flat roof horizon. The roof horizon is characterize by thinly laminated, muscovite-rich gray shale that in the immediate vicinity of the area hosts oval-shaped, downward-buckled stress pot-outs. Appendix P - Page 3 of 28 ~ i Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 4 The parallel ridges are characterid by an irrgular, rough texture, but are bounded by immediately adjacent patchy areas of approxiately 5-10 cm2 that represent a flat: smooth, slickenside plane that follows the base of the muscoviterich gray shale (Figure 4). No part of the linear ridges appeared to extend upward into the thn shale layers of the roof, as indicated by a thin brow that intersected the edge of the linear features along the trnd of a prominent strs cuttr. The collection of a piece of the protrding ridge was attempte with a knfe blade, but the ridge repreents only a very thin (-01 mm) coating of slickensided shale, and scratching with the knfe blade immediately expose the overlying muscoviterich gray shale above the thn coating. This resulte in the whitish straks shown in Figure 5. Fi~ 3. Two pair of pael ridr.es expod on the underside of the shae roof, and disrupted wher a shallow strs pot has broken out of th roof. No eviden of the liear feature was foun in th thin brow of the str pot along the trnd of the liear feature, indicatir. tht it doe not extend upward into the ro Appendix P - Page 4 of 28 . r r t i . Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 5 --, - .. . ..l Fir,re 4. Two pair of linear, parallel ridges exposed on the bottom surface of the way shale. Center left which is oriented feature lies alonr. a trnd of N 70oE, forinr. an acute anr.le with the featu at alonr. a trnd of N 43°E. There is no indication of the linear feature extendinr. above the thinly lamated imediate layer of the roof, as exposed in the th brow formed by the str pot-Qul. Appendix P - Page 5 of 28 . Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 . 6 r ~ .. ~ .. OJ , 4'ft.~ I) ~~',J' li. ~ .. .... ~"- .. . .JFI . )0 ..... :: :: - " i;,.' ~/l-, , ,,-".'" ~ " ~ ...~~ 'P '!.l ~~.~ ia ~ .. liear ridges represent kne scatch marks from an aUemptto collect fossil materiL. Location is the viánty just inby the spad 4010 intersection. Twin parallel ridges pass beneath the embossed, square skin control plate. Fir,re 5. Ught brown linear straks along the trnd of the parallel Discussion The purpose of the February 21" mine visit was to observe and identify two pairs of linear features located in the vicinity of spad 4010, in the 2nd Left Mains. Although there are abundant structural geologic discontinuities in the surrounding area, including joints and slickensided faults, the pair of linear features in question is not structural geologic features. Instead, the linear features observed just inby spad 4010 in the #6 Entr, and portayed in Figures 3-5, represent the remnants of a pair of fossilized trees, with each linear feature representing the top, tangential edge of a single tr. The rough texture of the linear feature represents the trace fossil impression of the tree bark as preserved against the bottom layer of the overlying muscoviterich gray shale, and the pair of parallel ridges represents compaction of the muscoviterich gray shale downward around the formerly circular boundary of the tree trunk. Although the fossil tree was removed by mining the immediate shale roof, the linear features represent the expression of the top edge of the tree where it tangentially contacted the bottom of the bedding plane exposed in the shale roof. If you should have any questions regarding this report, or if we can be of furter assistance, please contact Sandin Philipson at 304-547-2015. Appendix P - Page 6 of 28 , Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 Mine Safety and Health Administration Pittsburgh Safety & Health Technology Center P.O. Box 1 B233 U.S. Department of Labor r) Pittsburgh, PA 15236 Roof Control Division September 1, 2006 MEMORANDUM FOR RICHARD A. GATES Distr:ct Manager, CMS&H i¿trict 11 . THROUGH: d. riAf/ KELVIN K. WU ~7~ Acting Chief, P~~afety and Health Technology Center M. ~RRY Horn Chief, Roof Control Division If f". E jJli J .11" pa'n eU/'v . r::-"~1¡J. _?A jrLVI FROM: 0\ SANDIN E: PHILLIPSON f2 . 'J Geologist, Roof Control Division SUBJECT: Description of Features Observed in the Roof Inby Spad 4010, 2 Left Mains, in Wolf Run Mining Company, Sago Mine, MSHA _j.D.NoA6~0879L_ _. ___ _ ______ __________h.___________ _ n _ BackgroundAs requested by the Sago Accident Investigation Team, the author witnessed the extraction of a mine roof sample on March 1,2006 by personnel from R. J. Lee consultants. The sample extraction area is located just inby spad 4010 (Figure 1) where two prominent features are located in the roof (Figure 2). The features generated interest because they are located in the area where the explosion in 2nd Left Mains is believed to have originated. Because the features were not recognized as being widespread, they were quickly referred to as "anomalies." Due to their location in the area interpreted as the explosion site, some parties speculated that the linear "anomalies" might represent the effects of lightning arcing across the mine roof. Although initial observations conducted by Roof Control Division (RCD) personnel on February 21, 2006 (RCD February 27, 2006 Draft Memo) indicate that the linear features represent compaction along the length of a tree fossil, consultants retained by the mine collected samples of the features in order to document any possible effects of lightning. l)) \ Appendix P - Page 7 of 28 Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 2 r I a 50 100 feet ~i '.~ ~ L I _ Fiinre 1. Map of ¡ieologic feature in a portion of the 2n Left Mai, showi¡i reults of mappin¡i from Februar 21 and March 20, 200. Sample collecon ara is centered on dark jnn featur just inby spad 4010. Dashed purle line indicates Februar 21, 200 observation trvers. Appendix P - Page 8 of 28 Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 3 -.. , I Fiinre 2. Two sets of paired, liear rid¡ies define an acute an¡ile in the roof horion just inby spad 400. R. l. Le sample collecon effort on Mar 1, 20 extracted saples of this feature. In th photo, the linear feature is trcated by a shallow str pot-out. The effects of lightnng have ben documented in unconsolidated soil, loose sand, and solid rock. The preserved effects of lightning on rock and soil can form silica glass known as "fulgurite". Fulgurite has ben found in soil and sand dunes, forming a small tunnel with walls of silca glass, presumably formed by high temperature melting and fusing of quartz sand grains (Figure 3). Other experiments documented on various websites indicate that fulgurite can be formed in any rock composition with suffcient voltage. The longest fulgurite tunnel was reportedly approximately 20 feet long, and a search of available literature suggests that the fulgurite tunnels are 2-3 inches in diameter. Photos available on websites indicate that the cylindrical, glass-walled tunnels undulate, twst, and turn, commonly branching or bifurcating though the unconsolidated soil materiaL. Although lightnng can afect solid rock, available observations indicate that fulgurites in rock are restrcted to the top several feet of mountain peaks, and seldom penetrate more than a few inches into the rock. Lightning can magnetize iron minerals in rock outcrop, as observed by the author at a location in the Colorado Rocky Mountains. Appendix P - Page 9 of 28 Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 4 ----- --Set-6 I i I L _' _ - -L Fiinre 3. Sample of fu¡nrite for sale on internet website, showin¡i branchin¡i texture of bubbled silca ¡ilass. Thus, visible effects of lightnng on a rock would be expected to include the formation of silica glass or quart grains that showed sign of partial melting or fusing. Glass, of which volcanic obsidian is an example, is very distinctive in the geological environment Most geologically formed glass is associated with volcanism, in which high-temperature molten rock is frozen before crystals can nucleate and grow. Methodology The mine's consultants obtained four rectangular samples from the roof and retained thee for testing. Samples were obtained using a battery operated "ripsaw" to define a rectangular cut sequence to delineate the sample. Aftr a notch was cut to provide working room, a wide, flat chisel was used to force separation along a delamination plane along bedding to remove the sample from the roof (Figure 4). Appendix P - Page 10 of 28 Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 5 p r ~ t i Fi¡iur 4. Show reinlar box remain where samples of th linear" anomaly' were retreved on Apri 6, 20 (renter). Saples wer collecd frm the sae "anomaly" on Mar 1, 20 by R l. Le as indicated by shallow box locted at ri¡iht of photo. Samples 305477 and 3075 were retreved from the box on the ri¡iht side of the field of view. Splits of the thre samples obtained by R J. Le were passe to MSHA on March 13, 2006, and obtained by the author on March 16, 2006. Two of the samples were cut with a water-cooled, diamond blade rock saw at the Approval and Certfication Center (A&Cq to obtain a cross section though the area where the linear feature appears (Figure 4). The cross setion slice was annotated with five retangular blocks to be prepared for thn setions. Loations of the rectangular blocks were marke on the mating surface of the original sample split (Figure 5). Each outlned block was then sawed from the cross setion slice to define an individual sample (Figure 6). The chips were then sent via FedEx to Spectrm Petrographics in Vancouver, Washington, to prepare thin sections of the samples. Thin setions are slices of the rock that are ground so thin that light can pass through the sample, while glued to a microscope slide so that microscopic textures and details can be documented. The complete thn sections were received on April 7, 2006 (Figure 7). Appendix P - Page 11 of 28 Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 6 '" -..j . ~ p .. .4'. .# .,;0 .~-'-1~ L - .. .. J -i ~ ;.- - I .. " ~"~ ~~ ~ / r '" ~ .~ " -' I ~~ .~~ '~ ,. :;_,\\_.,"., -L.~.". r _ 1 """'i'''~.Î ~_..~~ L:'. J I 11 \ . I ~=,,- i-""-' _ \ - - - i Fiinre 5. Split of Saple 305477 obtaed by R l. Le on Mar 1, 20 at A&C, showin¡i cros seon across the liear feature obsered inby spad 400, 2nd Left Mai. The work shown was peformed at the Approval and Cercation Center. Appendix P - Page 12 of 28 Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 7 ~.. ~~ ~ , .. -- ~ ./,,' 4;':'' if' :'-'\ \-",- -' 1-... 'C' r- ~\-.. ~ T... 7 'c, ~~tr:;-'_ I. "\ ~¡ lr, l.~ - I ,l ,-- ~-i ~ 1- 'i ~1 1 ~l . .,-- " ~ .r'.;~/.(f . c'. " '~.-"~ ì . ..l?:/' ~ ~ -- :"""~ .. ."" "';Cl 'E .~.QJ oJ ~ ~ i _ -,. -" --..:.i ~ )0 . -~ Fi¡ie 6. Cross section slice from Saple 3045477 (top pair) and Sample 305475 (bottom pair) furter separated into individual chps ready to be made into thn sections for detailed study. The work shown was peformed at the Approval and Cercation Center. Appendix P - Page 13 of 28 Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 8 -~ I --~ ,- Qí .. I'.~.. .'j, II -- '-- --..' . 1- .1 Fi¡iure 7. Completed th sectons (¡ilass microscope slides) and ori¡iinal sample chips prepard at A&CC retured by Spe Petro¡iaphics laboratory on Apri 7, 200. Summar of Rock Textue Observations Subsequent to sample preparation at A&CC, the chip from Sample 3045475 was observed to exhibit a striking texture. The sample hosts a very thin layer of black, coal- like material that appears to represent carbonized (coalified) plant bark, as indicated by a series of parallel lines that are similar to the cellulose of plant fibers (Figure 8). The carbonized, fossilze plant material is located at the core of the twin, parallel linear ridges that trend across the roof of the area inby spad 4010 in 2nd Left Mains. The thin sections of Samples 3045477 and 3045475 were studied with a Meiji 9400 Series polarizing light microscope at viewing scales of 40X to 100X. The samples of shale are classified based on grains size and bedding spacing as "laminated siltstone" according to Pottr's 1980 textural classification of shales. Because all six samples were collected from the same sedimentary horizon, within approximately 2 inches from the mine roof, they are characterized by very similar textures. Each of the six samples is characterid by a matrx composed of very finegrained (0.005-0.2 mm) muscovite lathes, which are randomly oriented but arranged in thin bedding layers. Contacts between adjacent bedding layers are gradational, defined by different grain sizes or mineral contents. The very fine-grained, muscoviteAppendix P - Page 14 of 28 Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 9 dominated layers host approximately 8-12% angular quartz grains, which are approximately 0.01 mm in diameter and isolated by the surrounding matrix. Coarsergrained layers are dominated by angular quartz grains, which are approximately 0.1 mm in diameter and touch along tangential contacts to leave angular interstices that are filed with finer-grained muscovite. The very finest-grained layers host very finegrained, clay sized (':0.003 mm) muscovite with no quartz, and represent planes of preferential weakness along which delamination preferentially occurs. Textures in all samples are very similar, characterized by muscovite-dominated layers corresponding to alternating grain sizes of "fine silt" and "medium silt". This material represents approximately 80% of the layers in each small, rectangular thin section. The remaining approximately 20% of layers are represented by "very fine quartz sand". Bedding layers are generally of uniform thickness, remaining parallel in relation to the bedding parting that represented the mine roof horizon. One notable exception to this is represented by Sample 3045477-4, which hosts a series of thin, discontinuous iron hydroxide stringers that suddenly ramp up away from the mine roof horizon, such that the stringers become closer together as they rise into the roof. This texture is characteristic of compaction of unconsolidated sediments around obdurate objects, and is referred to as draping. The parallel bands of "very fine sand" quartz, located approximately 5 mm higher in the section, exhibit the same rising at the same point on the traverse. The area defined by the compaction texture is at the margin of one of the two protruding ridges, which define the "linear anomaly" observed in the mine roof just inby spad 4010. The presence of the compaction texture, combined with the thin layer of carbonized plant materiaL, suggest that the twin linear ridges observed in the mine roof represent local compaction of the muscovite-rich laminated siltstone immediate roof around a linear tree trunk. No silica glass or magnetite was observed in any of the thin sections, and no textural evidence was observed to indicate that grains. have been fused together. Appendix P - Page 15 of 28 I Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 10 Fi¡iure 8. Enlrp,ed view of a small reangular sample chip prior to bein¡i sent for thin secon preparation. Th piece of Sample 3045475 exhbits a black area tht repreents carbonized fossil plant bark. Paralel lies are interreted to reprent cellulose plant fiber. The pair of linear features obseed inby spad 4010 in 200 Left Mai is cored by th carbonid fossil materiaL. The sample is approxiately 7/8 inch wide x 13J inches lon¡i. Appendix P - Page 16 of 28 Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 II Appendix of Thin Section Descriptions Sample 3045477-1 (Figures 9 and 10) The sample is composed of fine laminations of randomly oriented, fine-grained, ragged muscovite lathes. Although muscovite lathes appear randomly oriented in detaiL, partings between some laminations are sharp and distinct. Most micaceous bedding layers host isolated grains of angular quartz that are diffusely scattered parallel to bedding laminations. Individual quartz grains are commonly surrounded by a thin, diffuse halo of very fine-grained muscovite that may represent diagenetic sericitization. Locally, angular quartz grains occur in suffcient quantity to define quartz-dominated interbeds that are parallel to bedding laminations. Quartz grains in the discontinuous interbeds touch along tangential contacts, and individual grains remain partially surrounded by a matrix of fine-grained muscovite lathes that are randomly oriented. Laminations defined by very fine-grained muscovite commonly represent preferential delamination horizons. The sample contains approximately 15% quartz, which ranges in size from 0.01 mm ("fine silt") to 0.1 mm ("very fine sand"). The remaining approximately 85% of rock volume is represented by muscovite, which ranges in size from 0.005 mm ("fine silt") to 0.04 mm ("medium silt"). Based on the size of grains, thickness and nature of bedding layers, and content of clay-sized materiaL, the shale sample is classified as a muscoviterich laminated siltstone. Textures suggest a low degree of compaction because individual mineral grain long axes are not strongly aligned with bedding planes. Long axes of angular quartz grains commonly form an obtuse angle with bedding laminations, indicating that grains were not forced to rotate. Although bedding textures are commonly diffuse, thin, discontinuous stringers of iron hydroxide are aligned parallel to bedding and highlight laminations. Despite the presence of iron hydroxide, no magnetism is present, as tested with a small, powerful magnet that is weakly attracted to samples with as little as ':: % magnetite. Appendix P - Page 17 of 28 p Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 12 r .A:-.~,.".;~...(.;.. '.:;J',.,~,..,~.~ 1¡'~. ¡~*'~"'.;.'~.'~~~: "~~.': '.,l. . ~~; .j .. ~'" ,~"'.i" .~' ..~ ~ ~JI. '4.:""~'''''''' ,'..~~:.!R"if'~ "~.. . ~,:.." ...~.-.~~". ,.._.:!,....,~~ "'Ji.'~,~,,, '" .i..;.~\J."r.... f ,",,, .~'lI ..#. '..,~"t:).:;!;...;~r...l~'r :.~ .-:ir:i~~.~r.~ ";t'Ol;:H~. \"~:Â' '(, - i: '':...,.. , '..1 .:-'1i.-:..~. ....,.. ,ol 'i~""~_"''' . '\ ~ ~\ .," .......rfl,..,.'"(C.. '-~"'."'~ ......... .- 'l;':4"'\.~.. '~-""""'" ".".:r:. ," -.:.', ......:. .' .. .t?.. "-'),,;l.... .~, .,' 'i-..'~'?",.;, ..-A."..~.'..W:l:.;ir..~...'! ~.-!.~ - ,....~,,'i:../:o(tl.'~ .."l'.._'~~~'" ".h.".~.~~;~. .l"., ."..--. .ji'-~lc!'.~'.-".. 'l-~~~-:'r'.:-. .. .. ~at-l ""."4It..~~.. l "iO ~..;J'~~",g h' . ..~,:i....f--... "'\. '~\.,~"'-.:." .1-i- , ....~'_...,"';~~"'..\~...~-..' ..~'" \.; V_'4.t"""~~\:,~.ll:.1'-._ ._...y.~',¡I~'a.~"tr,~.:v . ,-.~. ,.................'. ,:'..~. ,,~..; ,&1",''" 9-.l-~.."~"J.¡"',," .r-l " ",-.- ¡'~... 'to .. '''f\.. ~'s'iT'''.'Ñ:'.~~~.J''..' -i..:. ~";r'~' ..~ ~~. .... .,.... ,'. .. ':,;..-~~~ _~.¡\~~,.\~tt~_.1i-.~l.~i.; '~, ..,,lt."A: .i'.!'~A....~.C"~~'j~,~ ...-:.,. . .;. ~:i;';.~~ ...:.... t~~,..~r~.._ l:'~ .~~~l~t~~~~(:t!!\.".t.i..." ~.~;r-"".~I'~~:'I~T~_,!.;.: ',. . . '~. .J ";,.I..':'!.. "_'Ij~ -...... .~.-r."ti'i)-- -~.......~I~..,-i ..~-.,..ri.J~. ._..~1'.. ",,!.~.,.....,..,.'.,.t..~f'¡....,..~..~4::"'. ':'.' \." .:'",',li 'toO ....., ....:-~.~....~".:~1.-~!-~~~:r., ~.r.-'..".......,,"l¡~..( :~_.. i~~- '\ -... "i:~~."':"..-c.'l..t. "'".l" ..~f _:-'J--~ø:~::~....U.~.',.:.:.,._.V.r.._ J ~~.t, .'" -.~",' ~. 'I. ¡,~",..., .'...'...~.... ' ..~"';;.."'. .,~.-.."......r."i'!..~~.~...:J.~.,. -r,;.~~¡:~ ;"'.t.........~. ..... .l.~...'l.~'l1f._,."'~~-~llJl..~.::~.,!'~l_."'..ß-!....-o..-~..t. ..~~ . ':' ~.. ....,... :.. ~~~.:~t,t.¡-.. ..~;" .1~~'..~..~.,.I.-_--r-~~l;l~~ir:l''',l.i-,::~r\~.)I'~'''.~~. '~._~;: _ . ~-...l.'~....~~N 't'7"-~.~/".\~.';'.,-N..;~.~~:..~ ....of" -. . ~'! ..t(:. It,,~,\ ;;.......-,..; .:-~..I:~..:.~...Ci:~~~ .l .~~,k..~¡~:ttì~"\~i-".~.,--l-":"~.:. l-:¿ .~.... .'i.c .~!,,/....1.1.~ :!~1 .#¡"..~.~J\.~.......-,~~.:;..;,.i.~...;t.i:~.'.~."t.....' ...Al'~V: ";:.i. :_~. ~/ \:~. lo ,,t,':.h '" ..g)~", .ic7i ~";"'i+..-R.., ,It¡.,...~:i'l.:. --.~..oitl-:".f~"i.. "...k',...~....~llt~."'..\ '.. -l"¡ - :. l.:,..¡ i..1:'\..~,.._....;'_.,'!.1.., ;i'\\,-'( 1"." "'~,~, - ,-,~'.""..~:,:"~".". ~ . / I. ...,.l. .~', .:~/""il'\l ¿""."~"''¿ ;'~ "';~#~l:\'\-jt'..S:i..\.. ~;_~-.-;."t:-:. ""~.~''' .~. ....,':.. . .-~'. . ._....'. ..:-(,.!....- ..;;L;_.......~~~.~-..~.. .~. ,l~"..' J-";' . .... "'.-iIf :'....... ... . .. .\.~ .. ,'" '.. .-i" .1,- ..\ '. .., .. ... ~ ,-K .,~'" .,,, .--1... . . _ ~ JI ..~--. _- "-- '-';' '..."! ", l. . .. ... ~ .. , ;... ..''' .", ~.._ i~./i ~'~. ..,.r...-' ,~o. _, ;'.' J . '. . .. ~¡... .. '. Fiinre 9. Lowest layer of shale imediate roof expos at mie roof horion, showi¡i an¡ilar quart ¡irain (bri¡iht white) scattered in a matr of very fine-waied lathes of muscovite (recn¡iular, bri¡ihUy colore yellow/pink/blue). Brown repreents patchy iron stainin¡i. Field of view 2.4 mm at 40X, taken under crossed polars. ~ .~ i --;r ~" ..,.. ~,.'~ . . ,". --.....~k.. ..¡i4l;....,.. .¡; J ".'~'...J': '.. ;. ,.",__ .r...... ..~ ..'. I' ",__ ....... ~.;? -~:';' ¡e.l'""",~" r......~',;- ,:"~ ".'\ ,'. ......: ~.t.~..;. '. f~.:,..,..,.r;:;.'..: ','~....~.'I"'; ',~..". .~z, .'. ..,. "..~ ., .......,., _?j." ..,.'" ,~;. . .."".... "-~1 .,.J....~'\~ ~... .-f ,'~ 'l'.c..,'~" .,:...)....-:.:0'.'\...'~... ,.. ~.. ....~...~.. ~.. .~.'i.'. . -4 ..'~~ '4~:'f~.r .~.,..--.....-l ~ .,_r'.. .. .ë;:-.. .. ~'- .. /~.' .,."'...1' .. .v- ..-, .....;1' '-. ",. .' -' . "'.. 't ,'. ':..."..."!... ~il"i.'\..""..~..... .....S_w:. ~~...-l..::..~-...' . .~ \ ;.""'..',,,,;' 'I' .~.T-..'''~ .,. ',.-. ... ..: _ '," ,..:(..' _~... ...... "-' ,,'O c~ ;-.."_ . ..." ..'".." .' ." ", ." .. ~"J . .... .' ....l. í'",'...._~.r A:-'1....~ ....-.-'",.;,.';,/y..,........ ....~ .'. . ..~. a',.',:'~ ~,~-,~lrl.' .'f"Q. .~~.-:.;.,', .l~~' ~:~..l~. ~~. .:~ :..: !.... ,. ~..'" t;---.~",...v:". - ~.~,.." ~ ...ó. ".'_. ':~.:.~.. .,. ~.,~-oc";." ~' , '-' J'-" ,,"_.-r;i.' ,,".,.. ..., .j~~)' .;....\ .."..'O .. .'. a., ~ ,,: .'.W' . ~.;.; . .~..~.. ...1- .,.' "~. . j '.' 'j .._.,....4.'.~.¡.. ,-!. .,\ll..,,,.' ... -.,.. ., - r::..l . ...'.."...,,. .'. '.~..--..,'.-' - ..-.. .;.,rt "., .,' ..:. '''t.'' ,.,,!i .. _..,..' ...,.... -l' (::.~.'" "....-.7 .. p.1'..... to -:_. .~,-' -t' i~.. ... 11........ _~.,..__...I .:l....~ --.11.... . ....... .'. ,. '.' -.~¡,~"-" .",. ., ..".'" "': -. ........ .. .,' ..,.... t,l.__l...~,".,'-l"'_ - -í. :..&. r..-.... ..J",f" ". '.","!...J'! '" "..~. __ .....,.... :"..."'J'" .'. ''i''... "'~ ",' ."','" ~, . ~ '. ".'"t, .....". :.." .~'.'..~ .-· "-# ~'."Ot~', . .:.; .., ..",' '-''' .', ;,:., :".' ..' "?,,' -',3..' ';," .~. c".''' '- .,,. . ... ~."., '); .'..,:. . r:.., ..,J. ~..... - _' ;' ..' ~'" -'."" ."" ..c.,.., r .' ~ '" ,:~" ," : .., :'. .~~~ ;.'.,' í' ~~ ~.,c--. .:~~' ".', '. , '. -' .: ~ _~,' '. ".A, ...~ : : ,è .. 'l-' ..' .'. ~ L" _ .~,.~. -l'~:;*" .,.' ;.-..J.; -.,....... -;.' .. ..'l..... ' '",;" ~ . .. _ .." .,..... ,.'. ....,"-'.,4. ~'I.. . ll. ... . ,..__,__._.,.....i... ". ,.. "w.: _.' ~.... ..l"',".' ;~. . . I. ... -l..l" ~ ~.;' ~ .......... , .fI. ._.:. ~.. '... '". . ,.40 ". '. ....- -, ~ ......l.. ~.. "'. ,'. . . '.1.. . ..~r;- \." 'f.' r .. .... \,.''' ..,....-',J.R.,\... ~ . Wi."" ,.. -.... .:r (..-'II :Jr-.'. i-' _ __ "'." . . .' ..... .. ,L '''',' .".. .' ,,~.~. r, ... ,,~.,.,. .. ",'" .' l .,l ...:;. ...~ ... .'.. _~", ; ~:.;.: ~ '...t..... . ~ - ..... ~/. .. _... .._.. .._'. :.. lo'. .' -.1'... ....~ Fi¡nre 10. Laation of aninlar quart ¡irai of "ver fine sand" size. An¡iuI ¡irai touch alon¡i tan¡iential contact. Lon¡i axes of quarz ¡i and recn¡iular muscovite lathes are not stron¡ily oriented parallel to beddil~ indicatin¡i that burial compaction was not intense enou¡ih to foræ ¡irain rotation. Field of view 2.4 mm at 40X, taken under crosd polar. Appendix P - Page 18 of 28 I Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 13 Sample 3045477-2 (Figures 11 and 12) This sample is characterized by a matrix of fine-grained, randomly oriented muscovite lathes that are arranged in diffuse bedding laminations. Two beds are dominated by angular quartz grains that are sporadically distributed within a very fine sand-sized band. In the muscovite-dominated portion of the rock, angular quartz grains are sporadically distributed, with individual grains isolated by the muscovite-dominated matrix. In the quartz-dominated bed, angular quartz grains touch along tangential boundaries, and are intermixed with coarser-grained, randomly oriented, thin muscovite lathes, The finest-grained portions of the muscovite-dominated matrix host bedding-parallel delamination zones that are planes of preferential weakness, Discontinuous stringers of iron hydroxide, which may represent alteration of original biotite, are aligned parallel along diffuse bedding laminations, Despite the abundance of the discontinuous, bedding-parallel iron hydroxide stringers, a powerful magnet is not attracted to the sample. The sample contains approximately 19% quartz, which ranges in size from 0.01 mm ("fine silt") to 0.1 mm ("very fine sand"). The remaining approximately 81 % of the rock volume is dominated by muscovite, which ranges in size from 0.005 mm ("fine silt") to 0.04 mm ("medium silt"). Based on the size of grains, thickness and nature of bedding layers, and the content of clay-sized materiaL, the shale sample is classified as muscovite-rich laminated siltstone. In coarser-grained interbeds, the long axes of quartz grains are not strongly aligned with bedding laminations, forming obtuse angles, which indicates a low degree of compaction, In the fine-grained matrix, muscovite lathes are randomly oriented. Appendix P - Page 19 of 28 Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 14 I -."" Ìi..~..~.... ~'"-''' . ",1:-..'i(..~.t'..~;~. .! '-'r~' .~ ~-"",,-.-...,,.,...~. :¡:,,,,.....~'J"..i,.):.~,' :':oiq,~.¡o .i.~""~ .~..~f.~~..,4i. r ~..:~~~.i.l:~-tJ);.~... .~~~:~:1-r:~::~':.~.._~'(.. ..:.~.....:.~- ~~;:l"~; ,~ .,~E,k~..it:.:~~~.. ;..~t(¡::¡.:.!l.2't:::;:~~1,'1 ,'t~:,~;~:~ tç~~..1~,,':. , ~',,'~ ~r5~~ ..'.í:. -r,,:,,~-;,,~o~"~-..ì.~.\' ','7' .~)/...._\. ,t~,. .~;,j,,;-~-',;. ~"''''""'~'zi~,,.~.~j. ';..i-..:"-r.\...:..;.'Í",.~....f~~. ~ ~. 'lf,.¡i... -,~'l.!.~~~:i~ ;iT...~:.y.."l....~..,~___..~,.~-".......;....: ~,.'..t ~,~..;$t...:r;l;.l~2.;~~;"!:::¡;¡.;.~'t.;~-; ~y~..iI !!$~~! ilt-i, ._~ ~ ~ ~."\~.. .i.;.t;¡'''i;..;I.;::;;,;l'l;~l.~_t..,~H';~~'~P.'~~rl",;,,~ ~~.,,~,.. JI...\ .' ~....:f 'l.-'j.. \.., ."r,M i ~~...."'~\. ''' .a..-~., '':''lc ". ;J.r,...-... '1f"1'~~il:~\':",,.''f:i'':-i''~.~''''-; -..':'_r""~... ,'.~:;t~!.'f~ . ,.:, ""1;l', . l....'\....Y~,;_. ....:l.."~ 'I\ "1.. 'lt~~'';''l'~.I....-l.~.:~.._..~... f ~_-'t':'.' ,;,,&.,.~,:', _..¿:~..-..~....~;, "~'_~.. .."9~'" _t"~;"'.~#'l '_. . °t.. _.:~~."'.' ...~. .....":: .:....~~.,'....i::. 6-""". ,:;¡#~ '.:q .,..-'It,~;;;.:_~~.. :t~"~'\.";'~;"rn. ..,~. "?:".,;4~v....' ""'~:~~";s'~~.J_,~.~t ri~!i.~..\.~oi; ~.".~""~~~i:...1t."I-l/..~:~:''' "y'"- ;~",-"':..;~::v.'-,,._y..!\.',,_~~~.~/..~t:~,,.....(..~~.., \-~ Z';-"'-.~,."l-i.r~.ít.. ~ .~~. .'.. o#-;....,~.~ -t' ø.l::"'..l-.l.:-~~:_~.lt..:lJ'\~'(--.. 4 ...it~-r~~f::..~ll.£.. ¡;f..;! ,,'\ oí't..~~' .. _. X~C' " ,'" .._. ....~~.,~ ~'\. ......-,.~ "'.. .,... . 4i...... :~'t~.. ~:'~."c-~J': ¡:""-:-'. ...~~ .....-il.,.: ....~,... ¡.'l ..~~... ...::~..,....:~:\.1 :å,:.._".,,-...:- 11. ~"'l" ~ r..:" , ...... .' -._ ....... .... .~.:.~~~ ,r... ~ ..~.-.: .'..':;'.:lJ'_ "/'; --';' .:oo.t ~ (: ..' ",::.' .~;.. r ..... ~'l ...:\ ,~; .l:~::~~~,:;:1..J -'~!t;"l.'$""~f;:-.~.i~l"~"':;/~¡"'::'-~A~~F..~..r;:~~.~-.~~~' ~;" "; ~ .. ~1! "'-'.~" ,'I- "('')-~'''''-ll -;........~.'. .,..).... ,'" --. __ ' ',' .11_..'" ,. ..,...' II ..~... ,.....~..."....i.-.. ":'".":'.;:"'~'¡" ¡li ì""' ~.,i~..' .. ."''-i"~.',.'''I'.~l'',.... W".' '-~-r'" .... '"--'~"I'.IÒ'1,1. '.-..."'.-,' ....._ .s-l~j...J.'" ~'. ~~...,,, "_:l .~ ".~ ~ . ." ':" \ AK!" "... ....... "'r,' ...'" . . .;-~ ~~ '--r'¡L .. c.. :.~ "'"'f'!'--. .....;~'.;f.r. -'~ ". ...~ .~..Il. '..~.,~..,:,,~'*,......... '.,.1~t;..~l'~. .... -Ôl' ~ l....-'., .'\_~1"'\ .,,'t-....,;..:"'4.._'-~:t.: '.:r."-:"'~';i.""---::-r,.",,~~,J;.t",:'~JJ' ..;,...... :.....:l""~. ~ r" .' ,...' " ' " . ~ .... i. ."" "'(~=i' ..I.. -.... .'(,...C'.~tl'~ .....,,~..- c. .... .~ ,. -- . ",' ~.- ',. .... ,., ',... . ~.". .J,. .. .., :.. !" . .. t.. . '~,"l .... .i ~ Fi¡nre 11. Lowest .. .,. layer of shale imediate roof exposd at mine roof horion, showin¡i an¡iul quar ¡irains (bri¡iht white) scattered thou¡ihout a matr of very fie-waied muscovite (bri¡ihUy colored pink/yellow). The muscovite-domiated matrx hosts patchy iron stain¡i (brown) that is oriented alon¡i beddin¡i laations, and may repreent leached original biotite flakes. Field of view2.4 mm at 4OX, taken under croso; polars. . VI .-. ,.d.1L 4~' l"" .. ~..#.... _ Ñ'~' -. ~. /t _. .-,""'l ' " S".l c' .., ....~.... :.)....:!~.:¡;., ....,,, '\1.~....~~~it .':,l..'t;r' :...., .J. ~..¿ ~.,.. . ......"'--.. ~,.. .',.. '~.. ~ "~-"' ~.. ..,# . ~~ .J ~"''' l.~ . ....~~~'.'~ ....~,~~, ~ ~~ ""':~"_.'." ...\.~ ..'. ...~..._~ ,. ;'11'1 -; .~.!~ .... i.¡ /,.-'1. ": ,-N l...... ..~''' '........¡a ,...J,.; f~o( ~"-.. ~ .. ", 'l." ..... , 'I ~ J ..,- -"T:~. ~- ''''?''.~7 t~..:..~wl" ~"~~'''''''''/~~J' ':'"--:."-/,"" - ... p .' '..',i", "i JiA ' ,,;; , ., ", ~. ~....1.':,. 1 l _'.... õf~...l.. J".A,' ~ t-..,... ,¿.'C.,1~ :-.. #;~o!~.¡" "'/" :~iM. ",.' ~ '~'~-'.~~'" -..., .:,6'...~,.~.-. ' . "'.~ _#"). ,.~"#,, ~ ..".. ii ,.,. ':" 5: ~. ...'.._- . '. ,';' ;1('" ,~;.,.~¡,~, ';). ,..:. ~~? ~ -.-:,..'.... ..... _£..-". .\'~ " ¡~..7. .I'J... '! Jl'1",. -0'" .",~ '.-.. .~~'. .... i-~.. ." '.'... ',¡W : . ,''''''lE, "" "_ . ." .-,. .". ".a ,".~ ~..-.. ":1 -... "'. ~ .¿- '.'! ~ .~~. /__ ....-rJ..,- ~__,'..r !.'" i"~4'i..;.....l,+ - I"'...';...."...;. " .ø",. -li -"-c"'- ft'....,.-...:J" ,._" ~,..:; .~., --: -rl..3,J .,.~'!~'J.", #4'~~..._..t;/.'" 4tr.. ...~.. .., t ..,... .. L,'ll...' '" "~S¿, '.., 'l .. ...-r~h...'..;..";.. .J .. f.. ~ .'-f.:. "'.:l"#... T "_. .~- -~ . i'"r~"'f: _ -4'l-. 'l- 1'/0' "* J--'" ... *. ;1," ,~''.'"''4~~' .#fit.... L- '"... l ;../,'~ -#",#....; - "......,.c;..i'jl'~.../"-l~l-.,~ ""..... ~ 7:.... -- ~ ', .~'i"-''' ~. -- -,,,,,";"--r;'. .I "'r..' ,z., .. . ~",-, ;.,',.. '.."..~:..1' ,: "".. ;.. "V; ¿~ ~~ ....¡....:;a.# --,,~ "."r"L-f"l' ..:i..'.â".Y 4'" ...::. = .:. '!$"',~i)" '..,;.~'~.'.-r ,/,J l,;~~#.'~"" ...'~,';.,... '. ...,."4~. _6':,... ........_,T l.~;. ;.. "'.'" ....;.... ,,--: ~# ~. ~,-~'''''.v-f;~v# ... .;: ~.;' ~:"'..~.t,; . r.d.. clt.'.. .~ ..l.. li ,. "f( T'. ..'.., ., ; ,.. t~.1W' .. #:"_. r' -.. ~..~. ;" ç.., ..~ ". .'" '1. ~" \.. ;,... .~. .,~. '/:-:." "';I':/1 ''l:.:' ,,/ l ...ll.~ -'iei... '~ .' l"~%""., :.. L.~ . ~"i~ - ;;..... ,:--" .~r -!~ '.. .. ,.,;;; l~ .\,,~.. 'v -~ ' .~...... .i¿... .'IIl#" 'A"~ .. ~.. '" ~ 'J' ,,.' ..ál'''--i # /,..,~'...~" .. ..''._ ",..\1." , -' -. ..#'l( ~r. _...~., "f"" -. ..,:1--.' ...~'~1:-;i -T-lI''_.~ .....1./!~.'l!- "."" "~;;";"".lr 't'17#?",;" .~.."/ ,~lt..:~:.:l . ~:l ~'.~~~",~l.~i /;::., .L";f.".~ .....#."I~--,. ~ ':'.. ~ .. Fi¡nre 12. Same area as previous photo, showi¡i individual, an¡nlar quart wain (white and p,y) isolated by suroundinll randomly oriented ra¡i¡ied fles of muscovite (yellow/pin). Brown patcy areas repn.'5nt iron stainin¡i. Field of view 1 mm at 100X, taken under crossed pola. Appendix P - Page 20 of 28 t i Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 15 Sample 3045477-3 (Figures 13 and 14) This sample is characterized by a matrix of fine-grained, randomly oriented muscovite lathes that are arranged in diffuse bedding laminations. Contacts between laminations are generally gradational, characterized by a changing grain size or mineral content. Several thin laminations are dominated by grains of angular quartz that are coarsergrained than those found in the muscovite-dominated portions of the rock. In the finegrained, muscovite-dominated portion, angular quartz grains are sporadically scattered, with individual grains isolated by the surrounding muscovite matrix. In coarse-grained layers, quartz grains touch along angular, tangential boundaries or are more commonly slightly separated by a rim of very fine-grained muscovite. This sample exhibits more quartz-dominated laminations that are more sharply defined with respect to alternating muscovite layers, compared to the other samples. Thin, discontinuous stringers of iron hydroxide are abundantly distributed, aligned parallel to the bedding laminations that are defined by grains size and mineral content. The stringers may represent diagenetically altered biotite flakes, Despite the abundance of the stringers, a powerful magnet is not attracted to the sample. Very fine-grained laminations represent delamination horizons that are planes of preferential weakness. The sample contains approximately 23% quartz, which ranges in size from 0.02 mm ("medium silt") to 0.2 mm ("fine sand"). The remaining 73% of the rock is dominated by muscovite, which ranges in size from 0.005 mm ("fine silt") to 0,2 mm ("fine sand"), The matrix of randomly oriented muscovite lathes, and the poorly aligned long axes of individual quartz grains in coarser-grained laminations indicates that the sample was not strongly compacted enough to force grain rotation. Appendix P - Page 21 of 28 Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 p 16 r ~ t i !0. Fi¡n 13. Lowest layer of shae imediate roof exposd at mie roof horiolL showi¡i wadationa contact between ver fine-wained, musovite-domiated layer and overlyi¡i coarsr layer that hosts ¡ireater quart content and lar¡ier wain siz. Lower, very fie-wained layer loc delaination zones (parallel black lies reprent ¡ilass of microsope slide where roc separated). -i....'_ " ~.4.~ I.j-". l " !'Y.g ~ , '..' II ø · j ~ .. , ........ ',':'d ~ ,'*~' .': ~ _ .~, ~~, . ~ "\~'~' ii l .. '.. ..:,"~.~' ':~. t l- -. ' .. ~ ,.~. ,.,' ...,.. _..~~, ", 1J _'".'.;.~,Ì' ~ ......4I. '.~;".~. .., ~. I: ,. .' .. '~,. ..... . (, .' r --~ .J l~,""... à '..) '.t~~" '. '" '_~'li.~~' .i,' " "',.,i. .- ",' 1-.,.... .~ .. \. ~ ," ~"'. .' Jí' ',,,0" .1. " ~~ . l-, .J. ";-'f.... 'f , ...'~-.-,;~./ if.. .~" \ ,...~:.y. It ;. .. t r~. ' ,,'~. ., r:.":.-1';'" .~'AI-- , ". "r, .T, k ,.' . .-l ~ ,.. -:. i-¡' .. .. ,..f'. ,.Itle' . :, .i/,i:;' ,d~1Ítt"\.~,fl. ~::~, "~'l;~' .'~ " . '. . -t . l- '-: ,,'..4' .- . . . . _ .:: y ", "), . ~. .~':-d ~ ; ~. .~ 'i "0: .... .. ~ '''., ~ Fi¡iure 14. View of a coarer-¡iraied, quartz.rich lamination, showi¡i an¡iar quar wain (white and ¡iray) islated by the sUIOundi¡i matr of fine-¡irained muscovite lathes (pin/blue/ yellow / ¡iren). Brown areas represent patchy iron stainin¡i. Field of view 1 mm at 100X, taken under crosd polar, Appendix P - Page 22 of 28 Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 , 17 Sample 3045477-4 (Figures 15 and 16) This sample is characterized by a matrix composed of very fine-grained, randomly oriented muscovite lathes that are arranged in diffuse bedding laminations. Contacts between laminations are generally diffuse, characterized by a gradational change in grain size and mineral content. In general, the very finest laminations host only muscovite, with increasing grain size associated with increasing quartz content, until some laminations are dominated by quartz. In fine-grained layers, angular quartz grains are sporadically distributed, with individual grains isolated by the surrounding matrix of fine-grained, randomly oriented muscovite. In coarser-grained layers, angular quartz grains dominate and touch along angular, tangential boundaries, or may be slightly separated by a rim of very fine-grained muscovite. The very finest layers host bedding-parallel delamination horizons that are planes of preferential weakness. Thin, discontinuous stringers of iron hydroxide are abundantly distributed, aligned parallel to bedding laminations. The stringers may represent diagenetically altered biotite flakes. Despite the presence of abundant stringers, a powerful magnet is not attracted to the sample, At the mine roof horizon, several of the thin stringers abruptly change their distance from each other along traverse, defining a compaction zone. This sample was collected from a portion of the R, J, Lee sample along which one of the pair of linear ridges ("anomalies") is located. A quartz-dominated lamination located 5 mm higher than the mine roof horizon also mirrors the iron hydroxide stringer-defined compaction zone. Although these textures suggest draping around an obdurate object, the matrix of randomly oriented muscovite lathes and layers of moderately aligned quartz grains indicate that the rock was not subjected to burial compaction significant enough to force grains to rotate into parallelism, The sample hosts approximately 13% quartz, which ranges in size from 0,01 mm ("fine silt") to 0.09 mm ("very fine sand"). The remaining 87% of the rock volume is dominated by muscovite, which ranges in size from 0.005 mm ("fine silt") to 0.2 mm ("ine sand"). Appendix P - Page 23 of 28 I , Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 18 r l r i Fiinre 15. Lon¡i str¡iers of iron hydroxide (black very dark brown) defie beddin¡i laminations in a compaction zone locted near the margi of one of the liear rid¡ies in Sample 30477. Anii quartz wains (bri¡iht white) are scttered throu¡ihout the matr of fine-wained musovite (spekled pin/ yellow with brown iron stairJ Field of view 24 mm at 40X, taen under crse pola. ,. ,. u ....., '. . , _ ,'lfl-..,. ~. l-...". _.... " ........ . , ''':' v. ~~,, -. ~-" ,~i:; ,\...,;.:i . "':'" ::;'.'. .-.'--II.. . ..', ,.', '""'~l~r..,~:,.~., .¥." ~L: .:'r.:; ",.' '," ',. .~~. - ~_ -," -..,.. ." .,"'." .~. ,,' .. .. y"".. ..... ''\. ", -~" .,,~", .....-,..'" '''.,r ,('.. .' ,'.;".\X ../.,~, ~"f .... ','-:...~, ..'"~t.. ~ '.'.. '....'t.. .,..J ~'" . """,.' '"_ ..,.cr..i--'.~,.'-'."!''V""", ..t -, '.' ,:,,....,r... lt-t., .""'''..,,~ ......".. "." . .."'; '"r .. . ""~"". .-.. ",y ." "~,_ .. _ . . . ...., - , -... -, ...."., '" '... '. . - . ~ .oil -" ~ ... ... ~ ...', .. ..._ ;è.,... _ . . , ' , ~-.. · ...\¡. ~...0'1 'i""'~n~ .,""....., ~ .. .l": ", _ , \ -'.. --. ''to'.,.... .~ _~.. ~ _ . '" ~ ¥ . ... ". .,', ',". .' .'~r"""':'è!' "...-, "'"'','.. --i'i'.~ '...,;~ '. '. '.,.. ,'. .'0.. ',.... ._'~ . '., ........; ' '..to.. :...~. . ~..~- I .~" ,,' .:-~ ~~.. ',' '-è'..,. . '. .' _ ..~. ~ . . . . ,'.. '. --,..... ..... -,. . ".... . -.t!"To', ,I, -.. ~.;.,.'''','.~'",~ ",,:l ',;: '1. ..'.:.~: ;i::~.,- ~-,i ¡t",., "jT. ,.': .,~ ..... .... .....,,,. , "~'" .~(,~ '.". .'. "''-;' ~'.. ...- .-,,'-1.. " "" '.' -.7, V 'A;I,"Y~ ~ .~, b... 4l'. ~.. .' '"',.-"',~ ,"'.' '~~. '... " -~" ~. ~'...., ", ",,. ...... "', .' '~'?"'. -' '-,""., """.-( A-' "'_"",,~,,;.., ." ,._~'¿' '. ... "!1 '... "~'""V.'~~\-#'-'-ß""""-'.__'! ,...i;._,.:....,. ..y...... 'I"'~'.. ).. .,~.. -\, "".._ ',' __,. J,'o. ."". '..... ~'" ~ ~'~.¡l~:¡ ". '¡~'-" ~,:lI i- '.. ... .::"~ .' . ,.' :. A .':" ~" 0. '.~-,..,. . .. .'".. "1. ...~;"~ .. . . ; \ .. 'i ',,".'..', or' '1....... .. '.....'L..'.., .' _ . ,. ,..., -.:,'.' :' ..;~. .. -., ~"'. ", "', ..-.,;. :..., ., # . 'p,/,.. -~''¡ C'C'." .lt,~,._ _, ".' "'. , '.' . ......, . ,.,' ,. .~ . ~'..- -i'"~ ~;'"~,- ',. .' :t' .,. ", .. '." ''¡ ',~,.l ;: ;;:..._...~..,;-'" .. ':, '~".._:~:~ _,.. ...._ "'..: .,.,. ",. ~ ',... ."'.',_~ ,l'.. .." :.-' ~"0.. 'r' ""'l"~' ;...." ,".. , ',. '...."... .. '"'.", . .;', '..C . "'. ~ ... " . ll, ,..~ ", ...,""._, "",' ....~ -.".', '"... ".. .. ".of 'l- .,~., .1. . " " ". .~~:, · "l ...~. ~ .. . . ~ " " :" .1 Q.- M" . . ~' . Fi¡iure 16. Field of view approxiately 5 mm above the area in Fi¡nre 15, showin¡i interbeds of quar that ¡ienUy rie from ri¡iht to left above the compacton zone. Althou¡ih loclly a compacton zone, the lon¡i axes of quart wains and muscovite lathes ar not stron¡ily oriented parallel to beddin¡i indicati¡i that buri compaction was not suffcient to force ¡¡rain rotation. Field of view 2.4 mm at 4OX, taen under crssed polar. Appendix P - Page 24 of 28 p Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 19 r r Sample 3045477-5 (Figures 17 and 18) very fine-grained, randomly oriented muscovite lathes that are arranged in diffse beding laminations that exhbit gradational contacts based on changes in grain size and mineral content The finegrained laminations host scattered, fine-grained, angular quart grains, with individual grains isolated by the surrounding muscovite matrix. Coarser-grained layers are dominate by angular quartz grains that touch along angular, tangential boundaries that are parallel to bedding contacts. Abundant, thin stringers of iron hydroxide are aligned parallel to beding laminations and my represent diagenetic alteration of original biotite flakes. Despite the presence of abundant iron hydroxide, a powerful This sample is characterized by a matrx of magnet is not attacted to the sample. In this sample, beding contacts are particularly continuous and paralleL. The finest-grained layers host delamination horizons that are planes of preferential weakness. Although beding layers maintain constant thickness, the randomly oriented muscovite lathes and moderately aligned long axes of quartz grains indicate that the rock was not subjected to signifcant burial compaction. The sample contains approximately 16% quart, which ranges in size from 0.01 mm ("fine silt') to 0.2 mm ("very fine sand"). The remaining approximately 84 % of the rock volume is dominated by muscovite, which ranges in size from 0.005 mm ("fine silt') to 0.2 mm ("fine sand"). , . ",.. , .'...,~. ..'tt--.-. . ':I '(.r'~-l;""""'J.~" - '~~.,. '..,. ,. ... -":'''""'~~~'i~;'..~~_~ll'''!' ~""_*'- ." . _.. ~ ,i;. ',~..~. _iJ"' ~î¡¡"";'~'~" '~,'~~,"-" ll.t..¡:,." 'i; ,~s~,';ç~.i,~~~ì'~~.:.~I.,~~:t¡f.~.. ,'~ ~'''n:li~''''' """.~,l,.,..~~...~4.r..'t";;';:'Ik'ü.. .;.,.,,,,.....',~~.','c..¡"~ "~~ . .' 9_,,~~ï.).,...i ~;f iY't~ ~._'r.':.~.\.,c,t~~... ~..~..fLil~fi;'i:"'~?J"t¡....fi~v'"~,1-~."'._"'~.~!¡ ~" ~,?!.-..." ~:lr.!...~'.......;. '. "'~~~..~:.~ ~~~..'" 'f-~-";:' ~,¿;~-.-..1I,..~ j~. ..~:\~":~~ ..~.. K..v. :l.ø",~r.'.., .',q,"'~;rt~, 1',- ni:,~' 't,ft~;i",,,,,..,,,,,~,,,,,,, '~"""'''~'~, .'. :0" .'P1'', ,..,~".~"'..",,!'(. _"" - ""'/"";.~~:l ~'l "-q,' .,' ..;r.",." ,.. ','-~ ..,.!ì.. -;~:i~~.;'t.'.""-"''''' ..F'.:...'..'. ..i-~ _~. ..~~.... "no",~ ~"...._"'..._g.r..,~.._;i.1...r..;,:. . .Ua. _'C' ....'¡_._~:l__..-,.~. ~ ~ ;. .~'",., .... ",. ;.;....,;~~,,:.,iii;¥.f.Clt.r-.;~,. ~~..,~~~i,. ~. .'r', '.","; ~ '~A-.i,.:i' .~.... -.i'~,,,4-l-ti~..'''''. '.... .~, ~~"::..~.,,ii, ;_ ø,~," ..;:~"i",..O"..i..-i:-.~~.: ~ -" '.. \' ,'I~.. :i:..~.",.¿ i,~ ~ ¡"'I..,.:~~~-.\! ..'...~-:::.-i.:~;~........l-~::~:;~;" '~rx~""-":..~~.;.~:j'~" ~~..:).~ ~~--:;,~~:~~-r.;' "-:rw~~~~~.., :; ;:.....~1"~l\~~,~"ii.~..'1.;..,,:~,lt.',.,.~ l.~rw.~-~~. .~. ~.~ :i~--~~\" 41;l7:.~'"", ....l,..y....I\I""_.._.I. . =-"'-i,.--~...J ") ~_ r:l~ r;..;.'~.r;~...:;l",:~n:"'..':'~~-i!vi~~.:-~..~~.. -;.~ ~!,l.:.:_it;l~."'.i ~. :'~ ..;;:';.~.,,~~ ii~~~ ~ ,t..-..#..;i~'~t...'y~..,.. ..,...~ ..~...,.i-i..¡,-.:'~......~~l._~.....~_.,.. ,¡;"I~. .,~,,,',_;"',;.;l'~....,,~ ~~~..~ 't' "~"'..-"~9":'~lF,;~'\;r~~~',:l.~I1..;-"i.p.~:'.~p i''',i.i~-K~".~~A,;l~~~'~~_i .. ~:o~'l..~, ...,~...J~ .tR' it" .~,!~....~..,..i.. :': i....,.....~,.,.. i.i.~,;~ :ylt .._~~.¡..,,r-.. ;o.....,,'l..-"__ ,. ..';.....~~.t. ....~~ .' ;;r.i":ii"'l-""d,._.tr~~ _:..,..!.,,,t-1-~,:.,,--;.~..-~i-.....:.;,.,¡:s~ ;., .,~.~ t_,!t"\~ '''....,\- :1...;cr:~..;l..~!",-t. '.,"1o.!lL~.~ ,¡..,~1r~:...-Irl..;:..~""l-..~.l- '¡-~'.. ,'~1-~~ .. L"'\'4-.lt~ '~''''7 "rt~t_"I-., ....,.-..,.....;:;-.:;1."......j ....,...-...n.,,.1'L... .' -..¿i'C-.-\-: :-.f. ... ..~..,.. .iI-;~',I:_ ,_.' ..'t)l- .,--¡.,i...,)"-~..';,"~....~t; . "''Ii ;,._...."'..'_........ or. ..f,,":~....~ _....\."~.'''..,. ,. . ¡ """""''-.-''. .r ~."''-'.'' ~'-. ..':, ':--1 \(...oii~.,".'''; !"'-lji. '....,,~.. .. ....ñ....:. :.'!:.! ."...' ~ ...,;... , ,';". ~~.::. "tf'-'" ",-''l~-i~l''C-~:'~ ":'';'_~''_'' .. :..--,r¿tp.1. ...,"-. . ,__ fo_-.):..~~." '.., -t:',':'..rJ .."... ..-, -~. ..r.:'... .".-4'" ~-. ~r -,f'.1° ...;..;"'.....~;-.~.!.,.. -.."' -.. ;',p ':.,~-' ...L...".,--, '.''1', :-: ~,...... "9..,... "~.-': ...,' .~".- ,,-:..~ì'(~.:a4t *,~,:~ .1I-". -.:-.;./....:'$.' "i ; e:~...-¡;...i~.. 'I.""~ ~i.-b ~. ..'F; ~-",~~,,":l..1. ...,"-; .'r.'. ..~~.;, .,~"'t-.,:.1...';..':., 4",...1" "' ,'-. ...J ~:r. ";;"'~~''''l...! \" ..J~-l~: -\t¿:..,-1 ...~..:. '.l~i-~ ;: ~"ìl~.'" ~+~.t ',,'~...~ .-".:,;~. t- .J..'~;r.. '.\01" .......-"-,... ....., ....0....,...., ..~'~-,¡. ..l...,.........l..I..,.'~.,:....,...,.,... ..~-. ...1 "".-. ...'.... '!'......l"7". '., '.l"- --..' ..-' ,'.", .-,' .. ó-. 'Jl..... il ... ..~ ,.~~¡. ~?:..:-..~)'..~; ....~;, _- ~;.. ~,.. "~;.-i'l',,. .,.:;:~..t"?::~:.1.,.~. ,..". ~~. i..;,...'.~,.~.'v\t;v~~~c ~. ri..:..;..J o;-,-~'~~ l: ~"_ "';?- '.: t,.., J;':::,-.ti-..~~""..~.;~,.')l,i-~'! .4- '" t-!'i l "~:"'"i .-....;''...... j~~ ~¡.'-. :..~I~ ~ ... ..r-...."..'.~.. "J ,i.'-i'z,,-,r'l'- ""'Il-4 . "p..'_ ,~:-...i~' "'-'" ..~.' ," ~'..._,.--..~i~' 'l "_' .. ~ '". ,;"" ",~;.-#~ ~.. '. -..' l. .,.. ...,;...7".:t.r:~ .!o:-.,;..i-::'..;:;: 'J .r'-"1-r' _. ...~.J.. ~,'~.."' , 1"''':~ _~_...-..:.,. ~ -._.:.;.-..:.,...~~..,-...;. J:'¡,i......._....."~)..-;-~~' ..."" ;:,.... ';.',. ".:......._ o"'~ .~....i..",1l t~"J". .. . .....,~ ,~....._.,-:..\ .,... ..."...';. c. -'....'..~ ..tt 'I --..-~\.t ~ - ....(. ..l-t--~'...li.l,,'.. ' ;,..i.t~: ;a:;J.,~.....l...~........~:!'-. ......... '.. ... .,_; ~;_... _.~ J:-....-::"~~. .... .~t ~~:'l:.;v1. ..~:~.:.l~it~._i.-;l........'~.~..l-~ ~..'~.:..';:~_.~ .~..':. .,.,,':..... \ ...... '!, '. . 'l -.,." ....... ...., ~ .'".~. ... ,_.......i,..~... .- t'-i. .1',.-,......"0-( ...~".--. .'- ....... ~ .."'', i .,.-' .., ., t .,r.. '''''4,''''~''''.o " .~..-;....' -. t,l.,o-: . '--..l- .." .....,.y ..... Fi¡iure 17. Lowest level of shale imediate roof expos at mie roof horion, showi¡i an¡ir quartz grais (white) scattered and isolated in the very fine-wained, muscovite-dominated matr. Field of view 2.4 mm at 40X, taken under crosed polar. Appendix P - Page 25 of 28 ~ i Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 20 ~~,., ,~t~'..'.., '., ',,', ,..~, '-t . 'lL~ ~'... ~ ,-' '. ,"'~.; ... '" ~~.~.Pc::.\,~~... ...,..."...a-: .'"'JI'''';''- l...;. -"",.. :1'; ~;';'~""~~'~;fí.,~:¡"$~.Q~.A-l.~..~:;~i~~~~~;~;-$jf~i~~ ",-l;-,ti",'.",v~"""~~-;'.4;'1i :f",~~;.,::,,-?;'tyt.PZ".~:i~:.'.'CJ ~"-N~"': d'~.i',,"""~"'~"" ,,~...'f;~,' '#t""_ '-' ". . :1' ~.'~ll '-. r..,~~.~ '" '.:.i¡,.,..'l,." .;;-"::'; 'h, ~::~~¡r.~ ,~;I:,.-~.z;¡!"l~:4".f""", ..,,.......01 .. ,.~..~~' -"", ',.i~,' ',." ~ ,- . -. ... ....,,;, '.., ,..~~r.:;.."'."l foi4. ~,:,'~:'._'.J~"'~"'" '¡"iL"'i.t- ~- -t:. ,".~ ~~ _~ . - - _. '~~'~- 'ttO .... ~..~~' .~L:!~~ '\~--....., ';... _..: ~~~,~~4ir~~r-illI~¥l~1fE~l2:t~~1,rt01~l~j~';;fL :-..r: ~!-'","I..-.., .:;.,~.:t.¡~"ü..".~ '~:f' l,-: ' lh'~J'~"'~"¡.,'-"',;~, s."t;')'~' 'c~ . .j.""''':- -,i. 7' .\....' ...-:~..~ ~:- ",_l' t.\ è'''~ . l-_..4.._-......f'..~~"i~~-.., _ ~~,~..-, .;~;;.If~~i'lt,';4'1 ~~~..,~.t..~~~~.,~".- ~.~...~.(i\.~(--'~~';,l~'.J.. ~..l,;'.!.~..~l,: ~ ,¡"",j. ..'~...'!.:i..y,. .:~. t:.,'''-i.....,:.".. (i-I,t....ii....'" '-i,,:i".-l..!l ..." ~,;"'''l-i.'!~...ii''' ~~'""-i'~ ;\!,lli.i:'....-,...... -;a.-.'..t."'ot'~".,,,~,,-:..'J;¡. Ò" . ~::~.~.;:~"" ~"~"~~'~~.i.~.' ...~ ~.:\ ,.;;./... ~"~~~~fA" ,:.t ;i~~..~.... .",,r:'~ : ~6., ::X-',~'r:~ i.' ft'c.::' 'ji stb--,...-."-,-;.., ",. .*-i-!"" I~~. l-~.r.'$ '¡~ . .~.; ~~,JJ~-"-"'~f.i ,.;.1.,...;,..,'i.J.:..:,1iI...-, ~:v...1..;'" ',~ .~:'~r.;.. .l:...~'":- ...J'~~.Jt, '~)lt,;,'?i.',"r'" .. n'r....~~k-~..!J 'G-'''~~ tl ~ li. ~~.,"":t'" jlA.~).?k1'~."C;;~i. ..~- ~~::::-+;C\~'l~.-:.,"". ~~!,' S .!.:-':t:, -::":.i;w~.i ~.,"'ž:!'~ ~"'~..'\ -; ...~::~. .. """..~t\.. ..-:. ~ .;l.~'-:'~ ll\;' l....~...ii'-.,~-(~~..~.~.. ¿ol#,' ..~',ri..,.. .'~""-;' :., ;."*~....-l-~l""~~.~;'~..i~,,\~~-'i.i~ ~,"'ii-' .--.!.~.i~.'l'" '-..-'.......~...: '~S-i"" ., "-. ~ .~! i::....\ ..............~~~;.¡...l.......~, ~.".... .~~-:~A;..~"'~.,~t..-,,-l ....'v.;,.~ ',..;........- '~ ".,.... )~~i~.:;.~.;,:.. ~1J~:t:,~!l,,-~~,~~~~";"~i~~ ...,..... ~~,;~l~1--~...#,~r~-2.~t~~i~~¡~~. ~R-....#~-. ."..~..,.l.... .'v_-:;i..;..l1.t~4~~ ?:.;;-L..-~;i""J#_,..'"...,'-~.'-,~,'" .-)f..:.~_'''- ...,~. R~~-,~ "l'.-f ..: ....,~ l -..,. '-.. ). '~~'''Ll''' _ ":"~ .,-; C';. - ~~._., u. .;',l-" '-. ~..~î-~._ 1~::Ffi,. :;~'~~~~~" :J.~~~~':'",.;'.,,~;"~,='~;~:~K~t:;~i:;;, 'l;:¡'¿~~~l1~.':', .,.i -rst:~Yi;'~.7'.~,~~ -:,, ~ '''',-. ,.~..,...... ~ ..."Z.'"'.'''''' ~r',"" '''C'..,'' ~ .' .-l' ~_'l _ ...._"".:l' ~,,,,,~.., Joil.. . ...... -ê,',,_,., .""..' .."...~..."'f'.J.... ,4-~'. -"'-~ ~-4' ..... _ ~Jß . -. "')~:'''.~;.''-.i 4'_ ~ ~ :,,.."":......~'!_. ...._....' ' J.~'_..~ -:,;. l - ,~-II''" ~:~ ~'"'j~'''' ;'!~l;'1~\ 'l~:'';:\"~~.:~l "J! ~:,,-~~.~.. ,..'. ..'1. ';j?:" l"~\.~/~\."'",,~ - _ '-."-".' ...,..',..1..;..'..--.~~~ .,.:.. ~',"'" ....J.....".f'-..--..;ft ,. ..-.~'-. \.. ,.¡.~.. .. "¡ l :"'.. "! ..j" :...~R:;¡. ~..l..~-.~ ':'~ -";- -,. a .~ to: . '... .. ~ #' If ...~',.~...~_ .,0"_1.~ .."_,..... ...."'," '. . __'J~-' .._.. ~~'" ..... ~ ..... ..'t.;"-, ""..' .. .~_ .. ......" ii "_-'..-~.--.._.'!.. ~-,.~)i.,.'"i'!'.:,~.--:'I..i..'_'._~"-"g_ ...'...~.i..": ;:",..-.#,¡.__.,....."". -:-'A .. -l.+.-,L,.~).~.-.r~_~.~.",;. ..t~-it~..~..... --.:... -:.:._."...~~t""-i'.-Tõ:...,. l"~...,......:\.,. d61':~.'" .'~,,-'''-~r.I''A. _...\('~..,;).'-"'!t.r. -., .'!..., ~ t,_ .c.\l.:~. ';'~$r.. __Ai ~:, :~..."~"-,....._l...l\.:.~..~:#f~;::;."" ..~~~..~:/;'~. ..,-i..¡#_.p1..:i' ~:.-r('\,~,- .,)~.. r~-" L.. ~.. '--::: .~-.-..~i 1I.. .. ..~...~!...1'.;..r__-I'~JI". -',"': ,~. ..¡~ .-..~;;" ~.. _~., f~p.l-,~;''-~, .. ...~ l' "'~""1o...i :. r." Fiinre 18. Very reiiar, contiuous bedi¡i contact between lower, fie.wained lamination characterd by scattered, aninlar quarz wain (bri¡iht white) in a very fie-waied matr of muscovite (speckled yellow/pin), gmding upward into lamination with abundant, aniiar quar wain. Some quarz wain touch alon¡i tangential contacts, whie most ar isolated by suroundig muscovite. Discontinuous str¡iers of iron hydroxide (black to very dark brown), which may represent dia¡ienetic alteration of origi biotite fles, are ali¡ied parallel to defie beddin¡i. Field of view 2.4 mm at 40X, taen under crossed polars. Sample 3045475 (Fiiiures 19 and 20) This sample is characterized by a matrx composed of very fine-grained, randomly oriented muscovite lathes that are arranged in diffse beding laminations that are gradational, based on changes in grain size and mineral content Finer-grained layers host scattered grains of angular quart, which are isolated by the surrounding, muscovite-dominated matrix. Coarser-grained layers are dominated by angular quartz grains, which touch along angular, tangential contacts. Thin stringers of iron hydroxide are abundantly distributed and aligned parallel to beding laminations. The stringers have a crystal habit similar to mica suggesting that they represent diagenetic alteration of original biotite. Other strngers are very continuous and follow beding laminations and pre-xisting micro fractures, representing precipitation of iron along open-aperture planes. Despite the presence of abundant iron hydroxide, a powerful magnet is not attacted to the sample. Appendix P - Page 26 of 28 Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 21 I The sample contains approximately 17% quart, which ranges in size from 0.01 mm r ("fine silt') to 0.1 mm ("very fine sand"). The remaining approximately 83% of the rock's volume is dominated by muscovite, which ranges in size from 0.003 mm ("clay") to 0.1 mm ("very fine sand"). I ,.._~' -.,_'O '~~. . ,~;~,"d~_'''''l~... .... ~ · " ' ..'j#"" .' '?:" . P'II 'l.~--"~' '::~-.. ""'t~... ,...~,. ,..,.~ . -.. ~ ... io" .;0.' ::c'J':' " " " '. "'''t . ")¡i' :... "~"i'.. ... . .......--~ '~-. ,~..~~"S.~",,~_, ....~. L,~..,-,r.,..~~~..'7,,,l ,!.~.. .l.. . í..l.'.: -'.~ '1.~~p'"'''i')!;:'', ..:i....r.o.~ ,.~:~ ~.:~;'=..l., _,d;;~_~,_,,,!~,:;,,~: ....'i ~~'~, "d:"~~ ~":., "."-c~.r-~..! .~' .,,:;-~'~.-. \.,~ ." '-"~'.f..'" ,,' .......... "...r....--,._ J ,""~'.r'~';" .. ," ,.4 '" '. ~,:~l. ~1' "'.." .~~..).. "'~ -l., .. ;. ~.' ,'; ..,... ~.. ~ ,-, .',,'... ,,~~.:r"""'" '.. .."' '~"l',, _1....: '..-....~ ~..~,-..-r,"l"-""'O ..,._., ....-y...-' ;-...... 'f_ __.. "",.i, .1,. -: ~....". ''t. '.' r "P ".-.. ;'... to. tl . , ~~.., .... .-" OL" -i:t-'" ..t.:..... .t~\....: ~..~.,..-_l._..~. .._. ',.... " .'. #.,~.( :....~ ~ ~ . ..~.. ~ ....,.~, l/'" ~,,\ ~ ~.;\.......--'' ". ......~-~... . ,.it ........,~"\.~.~'....:r.\....!It:_~.~..t.. i ..-..;."'..r.~~._. ~~,' .. ...Y".. ....:.."A'-.. . .. '.... ....-"..'..'..) ~''lI'. .,' . '. I~. ",.'.t' ,. .LV.-- 4f.'..- . ...,.. ...-..., lt..__"._t~-_~_ _,'t~..i "~';~~-'1-~". .. L_~. .. -,-""i_~ .... ,..~.-. '.~,'.'F'.'.. K ~......., .~ "'ío'~" ,-,",t"' . ,:"~(, "...".,.~,., ;... ':4 ..'._~41,..: ' ...."Ç ~~-l-:'~"" ...(..;,:.~~..,,-;r-i.y..!" J~'t'_"".."~~~__':",~"~"'..~""/"-' .. .~.. --- #-.-'... ;¡ ..-. t. Ii.. ~If~; ....v... ....... '-¡~ ..', 'f .1. "')~ .........-l ... .... _."~Yi' .-...._ .'.l. .',' ,,._.'__"'.."" ~.'~ ....-¡..... ".. ~.~ .. PO ~ '., ,.. ¡.,i'.... . _~.,. ,l¡....s"........~, 'l~,:'-' ;,s ......., ..."" "''' :-....~ "'~.'.fl~."'). ....... ...l ..'.... l'_ . .. ... . ..-, . ..,. .. -' - -!', '.. - - .. l' -'.J. .._ .. .. -:. t.¡..... .. ~'.' ",-" ....~i: ., .....~ .... ... __"~ .l.. ' I. r... ..' '.... .... ,'_..., ..; '.\~ ~..-. __' 1- '!... ,'_"" ... .. )'" ..t... . '",. ""J'- ...1,...; .,' c___;:' ll. l \i-'.. . ..' .-.' -:.;'''-'''-..'\..~..:'''., ...".. ':, '," '" .. .. "It '\ ~ -I.... ~... ~'.. ." ";. ... ..~ _... ! l, '..~ . ': .í..' _ ,_ " i.' ~ ~ l, _ ...-. :':-. "".", ...,. " . .... ~', - ".~'.' _"..... I/' .' oI.ro:~ ~u . .~.... '" . 'I. , , .~... ..''' '.-".. _." .t '.,..--; ....."', ."';1' ...... L: -.-.,... ~ ~ ~ .'-' .. , ~ , ..... .. l- ~ .,... .., .......:-.. . ... ~t).' ......"..... - ,c. ,- , '... ..J-.: ~ Fi¡ne 19. Lowest level of shale imedite roof exposed at mine roof horion, showi¡i aniiar quar wain (bri¡iht while) scattered thou¡ihout a matr composed of fine-waied musovite lathes (speckled yellow /blue/ pin/ ween). Wispy strn¡iers of irn hydroxide (black to very dark brown) are ali¡ied along beddin¡i laminations, and may reprent di¡ienetic alteration of origial biotite flakes. Field of view 2.4 mm at 40X, taken under crossed polars. Appendix P - Page 27 of 28 i Appendix P - An Evaluation of Features & Description of Features Observed Inby Spad 4010 22 Fi¡i 20. An¡iular quar wain (white and way) touch alon¡i ta¡iential contact nearly isolated in a matr of randomly oriented musovite lathes (yellow). Field of view 1 mm at 100X, taken under crosed polar. Appendix P - Page 28 of 28 Appendix Q - Mine Dust Survey Sago Mine Explosion Investigation Sago Mine - Wolf Run Mining Company - Mine ID# 4608791 SURVEY #1(a): Sampling Area: Mains Lab No. Bag No. Sample Type 681608 681609 681610 681611 681612 681613 681614 681615 681616 681617 681618 681619 681620 681621 681622 681623 681624 681625 681626 681627 681628 681629 681630 681631 681632 1A21 1A21 1A22 1A22 1A23 1A23 1A24 1A25 1A25 1A26X 1A26X 1B1 1B1 1B2 1B2 1B10 1B10 1B11 1B11 1B13 1B13 1B20X 1B20X 1B21 1B21X Floor Floor Floor Floor Band Band Rib/Floor Floor Floor Floor Floor Band Band Band Band Band Band Band Band Band Band Rib/Floor Rib/Floor Floor Floor Appendix Q - Page 1 of 15 Collected 1/30/06 - 2/03/06 by Clay Location in Mine 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 + + + + + + + + + + + + + + + + + + + + + + + + + 5326 5326 1/2" Sample 5400 5400 1/2" Sample 5500 2/02/06 GI 5500 2/02/06 GI 5625 2/03/06 GI 5728 1" Sample 5728 1/2" Sample 5852 2/01/06 GI 5852 2/01/06 GI 00 1" Band 00 1/2" Band 520 1" Band 520 1/2" Band 4186 1" Band 4186 1/2" Band 4426 1" Band 4426 1/2" Band 4700 2/02/06 GI 4700 2/02/06 GI 5285 5285 1/2" Sample 5326 5368 1/2" Sample 1/2" Sample 1/2" Band Page 1 of 9 Rec. 2/17/06 from Cook/Hicks Dust Analysis Coke Content 79.1 79.0 78.3 80.2 80.2 79.9 76.9 77.7 76.8 70.2 68.3 45.2 45.6 69.3 71.4 74.1 77.8 72.0 73.8 78.0 76.5 73.5 77.1 73.2 74.2 Trace Trace Trace Trace Trace Small Trace Small Small Small Small None None None None None None None None None None Trace Trace Trace Trace Appendix Q - Mine Dust Survey Sago Mine Explosion Investigation Sago Mine - Wolf Run Mining Company - Mine ID# 4608791 SURVEY #1(a): Sampling Area: Mains Lab No. Bag No. Sample Type 681633 681634 681635 681636 681637 681638 681639 681640 681641 681642 681867 681643 681644 681645 681646 681647 681648 681649 681650 681651 681652 681653 681654 681655 681656 681657 681658 681659 681660 681661 681662 1B21X 1B22 1B22 1B22X 1B22X 1B23 1B23 1B24 1B24 1B24X S1B24X 1B25 1B25 1B26 1B26 1B26 1B26X 1B26X 1C1 1C1 1C2 1C2 1C5 1C5 1C9 1C9 1C11 1C11 1C22 1C22 1C23 Floor Floor Floor Roof/Rib Roof/Rib Rib/Floor Rib/Floor Floor Floor Band Band Floor Floor Floor Floor Floor Floor Floor Band Band Band Band Band Band Band Band Band Band Rib/Floor Rib/Floor Roof/Rib Appendix Q - Page 2 of 15 Collected 1/30/06 - 2/03/06 by Clay Location in Mine 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + 5368 1/2" 5400 5400 1/2" 5430 2/02/06 GI 5430 2/02/06 GI 5500 2/02/06 GI 5500 2/02/06 GI 5625 2/03/06 GI 5625 2/03/06 GI 5660 5660 1/2" 5728 1" Sample 5728 1/2" Sample 5822 1" Sample 5822 1" Sample 5822 1/2" Sample 5852 5852 1/2" 00 1" Band 00 1/2" Band 520 1" Band 520 1/2" Band 2000 1" Band 2000 1/2" Band 3946 1" Band 3946 1/2" Band 4426 1" Band 4426 1/2" Band 5400 5400 1/2" 5500 2/02/06 GI 1/2" 1/2" 1/2" Page 2 of 9 Rec. 2/17/06 from Cook/Hicks Dust Analysis Coke Content 72.4 73.0 71.1 69.7 71.5 76.6 73.2 74.6 74.8 68.8 73.5 71.4 72.1 66.6 66.8 66.7 59.4 58.9 47.2 42.9 57.9 60.8 62.4 45.1 88.8 88.3 62.0 60.4 70.5 71.6 72.5 Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Small Trace Small Large Large Large Large Large None None None None None None None None None None Trace Trace Trace Appendix Q - Mine Dust Survey Sago Mine Explosion Investigation Sago Mine - Wolf Run Mining Company - Mine ID# 4608791 SURVEY #1(a): Sampling Area: Mains Lab No. Bag No. Sample Type 681663 681664 681665 681868 681666 681667 681668 681669 681670 681671 681672 681673 681674 681675 681676 681677 681678 681679 681680 681681 681682 681683 681684 681685 681686 681687 681688 681869 681689 681690 1C23 1C24 1C24X S1C24X 1C25 1C25 1D1 1D1 1D2 1D2 1D5 1D5 1D7 1D7 1D8 1D8 1D20 1D20 1D21 1D21 1D22 1D22 1D23 1D23 1D24 1D24 1D24X S1D24X 1D25 1D25 Roof/Rib Rib/Floor Band Band Floor Floor Band Band Band Band Band Band Band Band Band Band Band Band Rib/Floor Rib/Floor Rib/Floor Rib/Floor Rib/Floor Rib/Floor Rib/Floor Rib/Floor Band Band Floor Floor Appendix Q - Page 3 of 15 Collected 1/30/06 - 2/03/06 by Clay Location in Mine 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + 5500 2/02/06 GI 5625 2/03/06 GI 5657 5657 1/2" 5728 1" Sample 5728 1/2" Sample 00 1" Band 00 1/2" Band 520 1" Band 520 1/2" Band 2000 1" Band 2000 1/2" Band 2982 1" Band 2982 1/2" Band 3464 1" Band 3464 1/2" Band 5255 5255 1/2" 5326 5326 1/2" 5400 5400 1/2" 5500 2/02/06 GI 5500 2/02/06 GI 5625 5625 1/2" 5658 5658 1/2" 5728 1" Sample 5728 1/2" Sample 1/2" 1/2" Page 3 of 9 Rec. 2/17/06 from Cook/Hicks Dust Analysis Coke Content 70.7 65.8 74.2 74.6 56.9 60.4 51.6 49.9 74.1 75.7 58.9 64.3 49.9 57.7 85.1 78.3 73.0 69.5 74.3 73.6 72.4 70.2 67.9 66.1 64.6 64.8 74.0 74.1 60.1 59.4 Trace Trace Small Trace Small Small None None None None None None None None None None Trace Trace Trace Trace Trace Trace Trace Trace Trace Small Small Small Small Small Appendix Q - Mine Dust Survey Sago Mine Explosion Investigation Sago Mine - Wolf Run Mining Company - Mine ID# 4608791 SURVEY #1(a): Sampling Area: Mains Lab No. Bag No. Sample Type 681691 681692 681693 681694 681695 681696 681697 681698 681699 681700 681701 681702 681703 681704 681705 681706 681707 681708 681709 681710 681711 681712 681713 681714 681715 681870 681716 681717 681718 681719 1D25X 1D25X 1E1 1E1 1E2 1E2 1E3 1E3 1E5 1E5 1E6 1E6 1E8 1E8 1E10 1E10 1E20 1E20 1E22 1E22 1E23 1E23 1E24 1E24 1E24X S1E24X 1E25 1E25 1E25X 1E25X Rib/Floor Rib/Floor Band Band Band Band Band Band Band Band Band Band Band Band Band Band Rib/Floor Rib/Floor Rib/Floor Rib/Floor Rib/Floor Rib/Floor Rib/Floor Rib/Floor Band Band Floor Floor Floor Floor Appendix Q - Page 4 of 15 Collected 1/30/06 - 2/03/06 by Clay Location in Mine 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + 5770 1" Sample 5770 1/2" Sample 00 1" Band 00 1/2" Band 520 1" Band 520 1/2" Band 1050 1" Band 1050 1/2" Band 2000 1" Band 2000 1/2" Band 2522 1" Band 2522 1/2" Band 3464 1" Band 3464 1/2" Band 4186 1" Band 4186 1/2" Band 5255 2/02/06 GI 5255 2/02/06 GI 5400 5400 1/2" 5500 2/02/06 GI 5500 2/02/06 GI 5625 5625 1/2" 5665 5665 1/2" 5728 1" Sample 5728 1/2" Sample 5768 1" Sample 5768 1/2" Sample 1/2" 1/2" Page 4 of 9 Rec. 2/17/06 from Cook/Hicks Dust Analysis Coke Content 51.6 53.1 96.3 97.0 55.4 64.8 84.4 72.1 87.7 86.9 87.1 91.2 80.5 88.8 87.0 81.2 83.0 80.4 65.4 63.0 69.2 65.7 65.1 64.1 78.7 81.7 57.4 57.4 52.5 54.2 Large Large None None None None None Trace None None None None None None None None Trace Trace Trace Trace Small Small Small Small Small Small Large Large Large Large Appendix Q - Mine Dust Survey Sago Mine Explosion Investigation Sago Mine - Wolf Run Mining Company - Mine ID# 4608791 SURVEY #1(a): Sampling Area: Mains Lab No. Bag No. Sample Type 681720 681721 681722 681723 681724 681725 681726 681727 681728 681729 681730 681731 681732 681733 681734 681735 681736 681737 681738 681739 681740 681741 681742 681743 681744 681745 681746 681747 681748 681749 1F1 1F1 1F2 1F2 1F3 1F3 1F4 1F4 1F5 1F5 1F6 1F6 1F7 1F7 1F8 1F8 1F10 1F10 1F13 1F13 1F14 1F14 1F15 1F15 1F16 1F16 1F17 1F17 1F18 1F18 Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Appendix Q - Page 5 of 15 Collected 1/30/06 - 2/03/06 by Clay Location in Mine 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + 00 1" Band 00 1/2" Band 520 1" Band 520 1/2" Band 1050 1" Band 1050 1/2" Band 1474 1" Band 1474 1/2" Band 2000 1" Band 2000 1/2" Band 2522 1" Band 2522 1/2" Band 2982 1" Band 2982 1/2" Band 3464 1" Band 3464 1/2" Band 4186 1" Band 4186 1/2" Band 4700 1" Band 4700 1/2" Band 4780 1" Band 4780 1/2" Band 4851 1" Band 4851 1/2" Band 4934 1" Band 4934 1/2" Band 5011 1" Band 5011 1/2" Band 5100 1" Band 5100 1/2" Band Page 5 of 9 Rec. 2/17/06 from Cook/Hicks Dust Analysis Coke Content 90.4 90.3 70.7 72.7 89.9 85.5 86.0 87.5 76.1 81.2 86.8 84.3 80.9 66.9 88.7 88.3 86.3 86.8 79.6 77.9 83.5 80.3 71.6 78.4 79.6 81.6 74.8 78.1 73.4 76.4 None None None None None None None None None None None None None None None None None None Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Appendix Q - Mine Dust Survey Sago Mine Explosion Investigation Sago Mine - Wolf Run Mining Company - Mine ID# 4608791 SURVEY #1(a): Sampling Area: Mains Lab No. Bag No. Sample Type 681750 681751 681752 681753 681754 681755 681756 681757 681758 681759 681760 681761 681762 681763 681764 681765 681766 681767 681768 681769 681770 681771 681772 681773 681774 681775 681776 681777 681778 681779 1F19 1F19 1F20 1F20 1F21 1F21 1F22 1F22 1F23 1F23 1F24 1F24 1F24X 1F24X 1F25 1F25 1F25X 1F25X 1G1 1G1 1G2 1G2 1G3 1G3 1G4 1G4 1G5 1G5 1G6 1G6 Band Band Band Band Band Band Band Band Band Band Band Band Band Floor Band Band Floor Floor Band Band Band Band Band Band Band Band Band Band Band Band Appendix Q - Page 6 of 15 Collected 1/30/06 - 2/03/06 by Clay Location in Mine 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + 5176 1" Band 5176 1/2" Band 5255 1" Band 5255 1/2" Band 5326 1" Band 5326 1/2" Band 5400 1" Band 5400 1/2" Band 5500 1" Band 5500 1/2" Band 5625 1" Band 5625 1/2" Band 5658 1" Band 5658 2/03/06 GI 5728 1" Band 5728 1/2" Band 5771 1" Sample 5771 1/2" Sample 00 1" Band 00 1/2" Band 520 1" Band 520 1/2" Band 1050 1" Band 1050 1/2" Band 1474 1" Band 1474 1/2" Band 2000 1" Band 2000 1/2" Band 2522 1" Band 2522 1/2" Band 1/2" Page 6 of 9 Rec. 2/17/06 from Cook/Hicks Dust Analysis Coke Content 76.5 76.1 73.6 73.9 77.5 70.9 72.1 73.8 67.2 68.0 72.8 73.7 76.3 77.7 64.9 66.9 52.5 53.2 58.4 59.8 75.6 76.0 56.4 60.4 61.8 54.7 66.4 66.6 76.2 75.7 Trace Trace Trace Trace Small Trace Small Trace Trace Small Small Small Small Small Small Small Small None None None None None None None None None None None None None Appendix Q - Mine Dust Survey Sago Mine Explosion Investigation Sago Mine - Wolf Run Mining Company - Mine ID# 4608791 SURVEY #1(a): Sampling Area: Mains Lab No. Bag No. Sample Type 681780 681781 681782 681783 681784 681785 681871 681786 681872 681787 681873 681788 681874 681789 681875 681790 681876 681791 681877 681792 681878 681793 681879 681794 681880 681795 681881 681796 681882 681797 681883 681798 681884 1G8 1G8 1G9 1G9 1G10 1G10 S1G14 1G14X S1G14X 1G15 S1G15 1G15X S1G15X 1G16 S1G16 1G16X S1G16X 1G17 S1G17 1G17X S1G17X 1G18 S1G18 1G18X S1G18X 1G19 S1G19 1G19X S1619X 1G20 S1G20 1G20X S1G20X Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Appendix Q - Page 7 of 15 Collected 1/30/06 - 2/03/06 by Clay Location in Mine 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + 3464 3464 3946 3946 4186 4186 4780 4813 4813 4851 4851 4886 4886 4934 4934 4974 4974 5011 5011 5046 5046 5100 5100 5130 5130 5176 5176 5208 5208 5255 5255 5287 5287 1" Band 1/2" Band 1" Band 1/2" Band 1" Band 1/2" Band 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" Page 7 of 9 Rec. 2/17/06 from Cook/Hicks Dust Analysis Coke Content 51.8 61.8 61.8 65.1 72.3 67.6 75.6 72.6 68.4 75.5 75.0 69.6 70.8 75.5 75.2 64.7 63.9 76.2 72.8 73.6 71.7 74.4 74.1 63.0 69.1 72.3 74.8 66.9 63.2 72.1 72.2 68.1 67.5 None None None None None None None Trace None Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Small Small Appendix Q - Mine Dust Survey Sago Mine Explosion Investigation Sago Mine - Wolf Run Mining Company - Mine ID# 4608791 SURVEY #1(a): Sampling Area: Mains Lab No. 681799 681800 681801 681802 681803 681885 681804 681886 681805 681806 681807 681808 681809 681810 681811 681812 681813 681814 681815 681816 681817 681818 681819 681820 681821 681822 681823 681824 681825 681826 681827 681828 681887 681829 681830 681831 Appendix Q - Page 8 of 15 Bag No. 1G21 1G21 1G21X 1G21X 1G22 S1G22 1G23 S1G23 1G24X 1G24X 1G25X 1G25X 1H1 1H1 1H2 1H2 1H3 1H3 1H4 1H4 1H5 1H5 1H6 1H6 1H7 1H7 1H9 1H9 1H10 1H10 1H15 1H15 S1H15X 1H18 1H18 1H19 Sample Type Band Band Band Band Band Band Band Band Rib/Floor Rib/Floor Floor Floor Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Collected 1/30/06 - 2/03/06 by Clay Location in Mine 5326 5326 5361 5361 1/2" 5400 5400 1/2" 5500 5500 1/2" 5658 5658 1/2" 5768 1" Sample 5768 1/2" Sample 00 1" Band 00 1/2" Band 520 1" Band 520 1/2" Band 1050 1" Band 1050 1/2" Band 1474 1" Band 1474 1/2" Band 2000 1" Band 2000 1/2" Band 2522 1" Band 2522 1/2" Band 2982 1" Band 2982 1/2" Band 3946 1" Band 3946 1/2" Band 4186 1" Band 4186 1/2" Band 4851 1" Band 4851 1/2" Band 4891 1/2" 5100 1" Sample 5100 1/2" Sample 5176 1" Sample Page 8 of 9 Rec. 2/17/06 from Cook/Hicks Dust Analysis 67.5 67.3 67.4 66.3 66.3 66.6 59.4 61.8 72.3 74.1 56.9 5 4.4 64.0 66.0 46.3 55.4 70.8 67.9 54.1 45.2 62.9 45.6 68.4 68.0 70.4 85.7 78.4 74.1 52.8 64.3 77.4 76.9 77.6 75.4 74.5 75.8 Coke Content Small Small Small Small Small Small Large Small Large Large Large Large None None None None None None None None None None None None None None None None None None None Trace Trace Trace Trace Small Appendix Q - Mine Dust Survey Sago Mine Explosion Investigation Sago Mine - Wolf Run Mining Company - Mine ID# 4608791 SURVEY #1(a): Sampling Area: Mains Lab No. 681832 681833 681834 681835 681836 681837 681838 681839 681840 681841 681842 681843 681844 681845 681846 681847 681848 681849 681850 681851 681852 681853 681854 681855 681856 681857 681858 681859 681860 681861 681862 681863 681864 681865 681866 Appendix Q - Page 9 of 15 Bag No. 1H19 1H20 1H20 1H21 1H24 1H24 1H24X 1H24X 1H25 1H25 1H25X 1H25X 1I2 1I2 1I3 1I3 1I4 1I4 1I5 1I5 1I6 1I6 1I18 1I18 1I19 1I19 1I20 1I20 1I21 1I21 1I22 1I24 1I24 1I25 1I25 Sample Type Band Band Band Rib/Floor Floor Floor Floor Floor Floor Floor Floor Floor Band Band Band Band Band Band Band Band Band Band Band Rib/Floor Rib/Floor Rib/Floor Rib/Floor Rib/Floor Rib/Floor Rib/Floor Rib/Floor Floor Floor Floor Floor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Collected 1/30/06 - 2/03/06 by Clay Location in Mine 5176 1/2" Sample 5255 1" Sample 5255 1/2" Sample 5326 1" Sample 5625 2/02/06 GI 5625 2/02/06 GI 5655 5655 1/2" 5728 2/02/06 GI 5728 2/02/06 GI 5760 1" Sample 5760 1/2" Sample 520 1" Band 520 1/2" Band 1050 1" Band 1050 1/2" Band 1474 1" Band 1474 1/2" Band 2000 1" Band 2000 1/2" Band 2522 1" Band 2522 1/2" Band 5100 1" Band 5100 1/2" Sample 5176 1" Sample 5176 1/2" Sample 5255 1" Sample 5255 1/2" Sample 5326 1" Sample 5326 1/2" Sample 5400 1" Sample 5625 5625 1/2" 5728 5728 1/2" 1/2" 1/2 " Page 9 of 9 Rec. 2/17/06 from Cook/Hicks Dust Analysis 76.1 72.6 71.9 77.1 75.6 74.4 86.1 82.3 64.8 66.3 66.3 66.4 63.0 59.9 56.4 59.5 55.9 46.5 44.9 48.5 69.3 67.4 74.4 73.6 75.4 75.0 72.7 73.2 78.2 77.9 77.8 79.3 77.0 65.5 64.1 Coke Content Trace Small Trace Small Small Small Small Small Large Large Large Large None None None None None None None None None None Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace Small Small Appendix Q - Mine Dust Survey Sago Mine Explosion Investigation Sago Mine - Wolf Run Mining Company - Mine ID# 4608791 SURVEY #1(b): Sampling Area: Mains Lab No. 681888 681889 681890 681891 681892 681893 681894 681895 681896 681897 681898 681899 681900 681901 681902 681903 681904 681905 681906 681907 681908 681909 681910 681911 681912 681913 681914 681915 681916 Bag No. Appendix Q - Page 10 of 15 1A25X 1A26 1A27 1A28 1A29 1A30 1B25X 1B27 1B28 1B29 1B30 1C25X 1C27 1C28 1C29 1C30 1D26 1D27 1D28 1E26 1E27 1F26 1F27 1F28 1G26 1G27 1G28 1H26 1H28 Sample Type Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band Band 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Collected 2/16/06 by Cook/Hicks 5748 5822 5880 5980 6043 6135 5748 5880 5980 6043 6135 5748 5880 5980 6043 6135 5822 5880 5980 5822 5880 5822 5880 5980 5822 5880 5980 5822 5980 JC JC JC JC JC JC JC Location in Mine JC JC JC JC JC JC JC JC JC JC JC Page 1 of 1 Rec. 2/17/06 from Cook/Hicks Dust Analysis 92.0 70.0 69.1 62.4 60.6 56.3 66.3 58.7 58.5 54.0 60.0 63.4 50.9 51.3 58.1 54.6 50.1 56.3 57.2 59.2 59.3 54.3 54.6 56.9 56.9 50.8 53.3 59.3 45.0 Coke Content Trace Large Large X-Large X-Large X-Large Small X-Large X-Large X-Large X-Large Large X-Large X-Large X-Large X-Large Large X-Large X-Large Large X-Large X-Large X-Large X-Large X-Large X-Large X-Large X-Large Large Appendix Q - Mine Dust Survey Sago Mine Explosion Investigation Sago Mine - Wolf Run Mining Company - Mine ID# 4608791 SURVEY #2: Sampling Area: 1st Left Lab No. Bag No. Sample Type 681917 681918 681919 681920 681921 681922 681923 681924 2F1 2F1 2G2 2G2 2H4 2H4 2H5 2H5 Roof & Ribs Roof & Ribs Ribs & Floor Ribs & Floor Band Band Ribs & Floor Ribs & Floor Appendix Q - Page 11 of 15 Collected 1/30/06 by Sparks Location in Mine 0 0 0 0 0 0 0 0 + + + + + + + + 00 00 1/2" Sample 474 474 1/2" Sample 1424 1424 1/2" Sample 1898 1898 1/2" Sample Page 1 of 1 Rec. 2/17/06 from Cook/Hicks Dust Analysis Coke Content 87.3 93.6 89.8 86.0 58.5 66.4 47.7 36.8 None None None None None None None None Appendix Q - Mine Dust Survey Sago Mine Explosion Investigation Sago Mine - Wolf Run Mining Company - Mine ID# 4608791 SURVEY #3: Sampling Area: 2nd Left Collected 1/30/06 - 2/03/06 by Ison/Sturgill Rec. 2/17/06 from Cook/Hicks Lab No. 681925 681926 681927 681928 681929 681930 681931 681932 681933 681934 681935 681936 681937 681938 681939 681940 681941 681942 681943 681944 681945 681946 681947 681948 681949 Bag No. Appendix Q - Page 12 of 15 3A1X 3A1X 3A6X 3A6X 3A13X 3A13X 3A14 3A14 3A14X 3A14X 3A15 3A15 3A15X 3A15X 3A16X 3A16X 3B8 3B8 3B13 3B13 3B14 3B14 3B15 3B15 3B16 Sample Type Floor Floor Floor Floor Floor Floor Floor Floor Floor Floor Floor Floor Floor Floor Floor Floor Floor Floor Roof & Floor Roof & Floor Floor Floor Floor Floor Floor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 + + + + + + + + + + + + + + + + + + + + + + + + + Location in Mine 40 40 1/2" 408 0" to 1/4" deep 408 0" to 1/4" deep 908 0" to 1/4" deep off floor 908 0" to 1/4" deep off floor 945 0" to 3/8" deep 945 1/2" 973 0" to 1/4" deep on mine floor 973 1/2" 1030 0" to 1/4" deep on mine floor 1030 1045 0" to 1/4" deep on mine floor 1045 0" to 1/4" deep on mine floor 1125 0" to 3/8" deep on mine floor 1125 0" to 3/8" deep on mine floor 526 0" to 1/4" deep 526 1/2" 877 0' to 1/3" off mine floor 877 0" to 1/3' deep off bottom 945 0" to 1/3" deep off mine floor 945 1/2" 1013 0" to 1/4" deep on mine floor 1013 1083 0" to 3/8" deep on mine floor Page 1 of 3 Dust Analysis 85.7 85.8 71.4 70.1 65.9 65.6 64.7 66.2 64.6 61.8 59.7 61.1 59.8 59.4 89.1 89.4 81.7 83.1 71.9 71.6 83.9 75.1 91.5 91.7 66.3 Coke Content Trace Trace Trace None None None None None None None None None None None Trace None None None None None None None None None None Appendix Q - Mine Dust Survey Sago Mine Explosion Investigation Sago Mine - Wolf Run Mining Company - Mine ID# 4608791 SURVEY #3: Sampling Area: 2nd Left Collected 1/30/06 - 2/03/06 by Ison/Sturgill Rec. 2/17/06 from Cook/Hicks Lab No. 681950 681951 681952 681953 681954 681955 681956 681957 681958 681959 681960 681961 681962 681963 681964 681965 681966 681967 681968 681969 681970 681971 681972 681973 681974 681975 681976 681977 681978 681979 681980 Bag No. Appendix Q - Page 13 of 15 3B16 3C6 3C6 3C7 3C7 3C8 3C8 3C13 3C13 3C15 3C15 3C16 3C16 3C16X 3C16X 3C17 3C17 3D1 3D1 3D12X 3D12X 3D13X 3D13X 3E17 3E17 3G1 3G1 3G1X 3G1X 3G2 3G2 Sample Type Floor Floor Floor Floor Floor Floor Floor Floor Floor Floor Floor Roof & Floor Roof & Floor Roof & Floor Roof & Floor Floor Floor Ribs & Floor Ribs & Floor Ribs & Floor Ribs & Floor Floor Floor Floor Floor Ribs Ribs Ribs & Floor Ribs & Floor Ribs & Floor Ribs & Floor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Location in Mine 1083 1/2" 378 378 1/2" 453 453 1/2" 526 526 1/2" 877 877 1/2" 1013 1013 1/2" 1083 1083 1/2" 1125 1125 1/2" 1152 1152 1/2" 00 00 1/2" 845 2/01/06 GI 845 2/01/06 GI 1/2" 908 1/31/06 908 1/31/06 GI 1/2" 1152 1/31/06 GI 1152 1/31/06 GI 1/2" 00 1/30/06 Intake GI 00 1/30/06 Intake GI 40 1/30/06 Intake GI 40 1/30/06 Intake GI 80 1/30/06 Intake GI 80 1/30/06 Intake GI Page 2 of 3 1/2" 1/2" 1/2" Dust Analysis 55.2 70.1 73.5 69.0 65.8 68.7 70.6 71.6 68.4 75.0 73.9 73.7 73.7 50.7 51.0 69.3 75.2 82.5 82.7 67.8 60.0 74.6 71.3 75.5 75.5 78.0 78.7 71.2 72.4 71.7 70.2 Coke Content None None None None None None None None None None None None None None None None None Trace Trace None None None None None None None None Trace None None Trace Appendix Q - Mine Dust Survey Sago Mine Explosion Investigation Sago Mine - Wolf Run Mining Company - Mine ID# 4608791 SURVEY #3: Sampling Area: 2nd Left Collected 1/30/06 - 2/03/06 by Ison/Sturgill Rec. 2/17/06 from Cook/Hicks Lab No. 681981 681982 681983 681984 681985 681986 681987 681988 681989 681990 681991 681992 681993 681994 681995 681996 681997 681998 681999 682000 682001 682002 682003 682004 682005 682006 682007 Bag No. Appendix Q - Page 14 of 15 3G2X 3G2X 3G3 3G3 3G3X 3G4 3G4 3G4X 3G4X 3G5 3G5 3G5X 3G5X 3G6X 3G6X 3H1 3H1 3H2 3H2 3H3 3H3 3H4 3H4 3H5 3H5 3H6 3H6 Sample Type Ribs & Floor Ribs & Floor Ribs & Floor Ribs & Floor Ribs & Floor Ribs & Floor Floor Floor Floor Floor Ribs & Floor Ribs & Floor Ribs & Floor Floor Floor Ribs & Floor Ribs & Floor Ribs & Floor Ribs & Floor Ribs & Floor Ribs & Floor Floor Floor Ribs & Floor Ribs & Floor Ribs & Floor Ribs & Floor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 + + + + + + + + + + + + + + + + + + + + + + + + + + + 110 110 150 150 180 215 215 265 265 293 293 343 343 408 408 00 00 80 80 150 150 215 215 293 293 378 378 Location in Mine 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/30/06 1/31/06 1/31/06 Intake Intake Intake Intake Intake Intake Intake Intake Intake Intake Intake Intake Intake Intake Intake Intake Intake Intake Intake Intake Intake Intake Intake Intake Intake Intake Intake GI GI 1/2" GI GI 1/2" GI GI GI 1/2" GI GI GI GI 1/2" GI GI 1/2" GI GI 1/2" GI GI 1/2" GI GI 1/2" GI GI 1/2" GI GI 1/2" GI GI 1/2" GI GI 1/2" Page 3 of 3 Dust Analysis 76.3 76.4 72.0 74.8 85.9 69.4 71.5 76.5 76.0 69.7 72.7 64.4 66.1 61.6 70.7 74.7 73.1 64.7 71.7 66.8 67.1 67.2 68.1 69.9 68.5 59.4 60.2 Coke Content None None None None None None None None None None None None None None Trace Trace None None Trace None None None None None None None None Appendix Q - Mine Dust Survey Sago Mine Explosion Investigation Sago Mine - Wolf Run Mining Company - Mine ID# 4608791 SURVEY #4: Sampling Area: 2nd Left Mains Collected 2/01-16/06 by Sparks/Hicks Rec. 2/17/06 from Cook/Hicks Lab No. 682008 682009 682010 682011 682012 682013 682014 682015 682016 682017 682018 682019 682020 682021 682022 682023 682024 682025 Bag No. Appendix Q - Page 15 of 15 4B1 4B1 4B2 4B2 4B4 4B4 4C1 4C1 4C2 4C2 4C5 4C5 4C6 4C6 4D2 4E1 4E2 4G2 Sample Type Ribs & Floor Ribs & Floor Floor Floor Floor Floor Floor Floor Floor Floor Floor Floor Floor Floor Band Band Band Band 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 + + + + + + + + + + + + + + + + + + 00 00 83 83 242 242 00 00 83 83 324 324 401 401 83 00 83 83 Location in Mine 1" Sample 1/2" Sample 1/2" 2/01/06 GI 2/01/06 GI 1" Sample 1/2" Sample 1" Sample 1/2" Sample 1/2" 1/2" 1/2" Page 1 of 1 Dust Analysis 73.3 74.1 71.8 75.5 71.0 76.7 73.7 72.8 70.5 71.5 73.1 72.4 76.6 71.8 52.1 55.7 55.8 53.2 Coke Content X-Large X-Large Large X-Large X-Large X-Large Large Large Large Large Large Large Large Large Large X-Large X-Large X-Large 2 North lvlains 0 0000000 0 . 0 A2 Panel Al Panel SEALED 2 No?h Moms . l?O? . O_r . No 4 O_r lstLe? No. 5 Belt - . . IO. . - IO-2nd Left Parallel 006669 0.00LEGEND No coking Trace quantity of coking (DO . Extra Large quantity of coking Proposed sample location where a mine dust sample could not be taken Percent lncombustible 6) Entry Number Crosscut Number Appendix Sago Mine, MSHA ID 46-08791 Wolf Run Mining Company Map Showing the Location of all Intended Mine Dust Sample Locations and Results 150? 300? Appendix S - Executive Summary of "Inspection of Sago Mine Voice Communication Equipment" Department of Labor Mine Safety and Health Administration Industrial Park Road RR1, Box 251 Triadelphia, West Virginia 26059 December 21, 2006 MEMORANDUM FOR RICHARD A. GATES Manager, Coal Mine Safety and Health, District 11 FROM: JOHN P. FAINICW Chief, Approval and Certification Center SUBJECT: Executive Summary of Inspection of Sago Mine Voice Communications Equipment Coal Mine Safety and Health, through Robert 1. Phillips, requested that (a) the mine phones at Wolf Run Mining Company's Sago Mine, J.D. No. 46-08791, be identified by model and (b) a brief description of how they were interconnected with each other be prepared. Table 1 describing the telephones is attached as is Table 2 showing unused pager connections, and a diagram of their locations. Multiple communications systems were in place at Sago Mine at the time of inspection. These included: • paging loudspeaking telephones located in various areas, both underground and on the surface; • a distributed antenna radio system allowing for communications between the surface and mobile underground equipment (trolleyphones); • a commercial telephone system on the surface; and • portable two-way radios. These systems were interconnected on the surface. Hardware used for connection of the paging system to an extension of the mine's telephone system were provided. An additional interface was used to connect the paging system to a radio transceiver, which allowed for two-way communication with portable VHF radios used on the surface. Portable two-way VHF radios were apparently used for point-to-point communications underground, but this was not observed during the post-accident investigation. Portions of the hardware associated with these systems were evaluated and inspected to determine operational status. It was determined that: • When inspected between January 27, 2006, and February 4, 2006, the underground portion of the paging telephone system featured eighteen (18) Appendix S - Page 1 of 8 Appendix S - Executive Summary of "Inspection of Sago Mine Voice Communication Equipment" individual telephones. Three (3) of these were not connected to the system; two of these were in the area of damage caused by the explosion and the third was found on top of a piece of mobile equipment. As detailed in attached Table I, the functionality of the units varied from normal to non-functional. The two units found in the area of explosion damage were not tested. The pager line was found to be intact except in the area of explosion damage. Leading from the surface, the most inby end of the undamaged line was located near the 1 Left Section track switch in the 2 North Main track entry. Additionally, the pager line was not damaged from a point near the #4 crosscut of the No.6 belt on the 2 Left Section, and leading inby. In the damaged areas, the cable was found to be cut or pulled apart, especially where it traversed crosscuts, exposing it to the apparent forces from the explosion. Repairs had been affected to these areas by splices or replacement of the cable with twisted-pair wiring. Additionally, at least nine (9) unused facilities for connection to the underground pager line were found. It is not known, for specific locations, if telephones were present at the time of the explosion, if they were moved during the mine rescue, or if telephones were ever connected. Not all of the paging telephones found connected to the system were permissible. The only devices found in areas where permissibility was required were assumed to have been installed during mine rescue. • The underground trolleyphone system consisted of the signal line, the track as a return line, a repeater, terminating resistors, and trolleyphones carried on the track-mounted mobile equipment. The repeater did not function when inspected; laboratory testing of the unit is the subject of another report titled "Gai-Tronics Corporation Trolleyphone Carrier Repeater, Exhibit No. GH-91P." The signal line was severely damaged in the area affected by the explosion. It had apparently been repaired to allow for communications before the inspection occurred. The repair consisted of termination of the line to the track approximately 20 feet inby spad 3854, at the 50 block of the No.4 belt. This was the most outby undamaged area. The trolleyphones found on the #6 and #8 mantrips were found with depleted batteries, and were not tested for function. They appeared to be complete, and with minimal damage. It should be noted that, if the signal cable had been damaged and the line was not terminated, the trolleyphones would most likely not have been able to provide communications with the surface. Appendix S - Page 2 of 8 Appendix S - Executive Summary of "Inspection of Sago Mine Voice Communication Equipment" • None of the conductors associated with the trolleyphone system or the paging telephone systems showed any signs of burning or charring associated with excessive current. However, it should be noted that ignition-capable sparking can occur without leaving marks on conductive elements such as these. Comprehensive test results can be obtained from the Chief of the A&CC, RR 1, Box 251, Industrial Park Road, Triadelphia, West Virginia 26059. Appendix S - Page 3 of 8 Appendix S - Executive Summary of "Inspection of Sago Mine Voice Communication Equipment" TABLE 1. SAGO MINE UNDERGROUND PAGING TELEPHONES INSPECTED, PAGE 1 of 3 Location Identifying Marking I Approval I Marking #1 Belt, #1 Block #1 Belt, 13 Block #2 Bclt, 22 Block Comments Receive Page? Provide Page? Talk to surface? Listen to surface, handset? Battery Voltage Top: 12.93 Bottom: 9.95 Top: 11.1 Bottom: 11.1 Top: 10.4 Bottom: 10.45 Case is green and yellow Stain less steel case Audible hum from handset I Case: None PCB: WBA ISOlA Femco Telephone, PCB: WBA3422A Femco Telephone, Model 821301, PIN AM7021, SIN 307003 None Yes Yes Yes Yes None No Yes Yes Yes 9B-155-1 Yes Yes Yes Yes #3 Belt head PCB: WBA 1598 None Yes Yes Yes Yes #3 Belt drive starter Pyatt-Boone Page Boss, Model I 12 PCB: 005-0077-003 Pyatt-Boone PageBoss PCB: 005-0077-003 RevQ Pyatt-Boone PageBoss, Model 112, SIN 12927, PCB: 005-0077-003 RevQ PCB: WBA3422A 9B-I02-2 Yes Yes Yes Yes Inside: 10.47 Outside: 10.47 9.47 No No No Yes 9.51 Yes Yes (muffled) Yes (muffled) Yes 8.82 Yellow and black case Gulton Femco Division, Permissible Paging Telephone, Model 821301, pin AM7011, SIN 045291 'Spruce', AEI Paging Phone, PIN 755-1 9B-155-0 Top: 11.31 Bottom: 11,31 11.61 Top: 7.88 Bottom: 10.17 Yellow and black case, Page Speaker missing #3 Belt, 9 Break #3 Belt, 17 Break #4 Belt, I Block 'Supply hole,' #4 Belt, 9 I Block #4 Belt, 40 Block Appendix S - Page 4 of 8 I Stainless steel case I 9B-I02-2 I I Yes Yes Yes Yes Yes Yes Yes Yes No No No Yes Appendix S - Executive Summary of "Inspection of Sago Mine Voice Communication Equipment" TABLE 1. SAGO MINE UNDERGROUND PAGING TELEPHONES INSPECTED, PAGE 2 of 3 Receive Page? Provide Page? Talk to surface? Listen to surface, handset? Battery Voltage Comments 'Sago', pcb WBA3422 No No No No 11.84 "A687JK", pcb WBA3422 Yes Yes Yes Yes 11.22 N/A N/A N/A N/A 11.12 Stainless steel case, dirty (earpiece and mie holes are filled with dirt) Yellow and black case, clean, installed in close proximity to unit detailed above Unit not tested for voice function because pager line was I disconnected, but remnants of I wiring presumed to be associated I with pager line found in terminals; audible click heard when page switch operated. Unit covered in soot and found in rubble; not connected to pager line; Handset was missing; handset cord was flexible and appeared to have been mechanically separated from I handset; an audible click heard when page switch operated; interior of unit elean and apparently undamaged. When first inspected, voice communications with this unit were not possible. After a break in the pager line at 21 Block of #5 belt was located and repaired, the unit worked. Location Identifying Marking #4 Bclt, 49 Block, near spad 3845 #4 Belt, 49 Block, near snad 3845 #4 Belt, 57 Break Approval Marking Femco Model 741301/402 Pcb 3422 9B­ 34(illegible)5 I Crosscut near #6 Belt drive Calibration sticker "Date 10-5-05 by RH"; Pcb WBA3422A ~ N/A N/A N/A N/A 12.27 Yes Yes Yes Top: 11.99 Bottom: 11.96 I I I I I Left Section, at Power Center Femco Model No I Illegible (illegible); Serial No. I 23(illegible); Two Battery Telephone Permissible: pcb WBA4097 Repair Job 35867 Date Rec'd 10-10-05; Date repaired 10-12­ 05; Hughes Supply I Co. Appendix S - Page 5 of 8 Yes I I Appendix S - Executive Summary of "Inspection of Sago Mine Voice Communication Equipment" TABLE 1. SAGO MINE UNDERGROUND PAGING TELEPHONES INSPECTED, PAGE 3 of 3 Location Identifying Marking Approval Marking Receive Page? Provide Page? Talk to surface? Listen to surface, handset? Battery Voltage I Left Section, #3 entry, Near old #7 belt drive Gulton, Fernco Division, National Mine Service, Gulton Permissible Paging Telephone, Model 821301, PIN AM7020, SIN 028020, 2 battery permissible phone, PCB: WBA4097 Rev B PCB: WBA4097, 'Spruce' Illegible No Noisy No No Top: 8.90 Bottom: 10.59 Yes Yes Yes (low volume) Yes Top: 10.54 Bottom: 10.38 Yes Yes Yes (low volume) Yes Top: 11.45 Bottom: 11.47 2 Left Section, Entry #4, near spad 4276 On top of shuttle car canopy near 2 Left power center I PCB: WBA 3422A, 'Sago' I I Appendix S - Page 6 of 8 Comments I Phone was located at end of twisted pair cable that was apparently added by rescue teams from end of mine phonc cable at Dower center Phone was not connected, but was believed to have been the phone connected at the power center before the explosion and subsequent rescue. Phone was connected to line for testing Appendix S - Executive Summary of "Inspection of Sago Mine Voice Communication Equipment" TABLE 2. UNUSED PAGER CONNECTIONS, SAGO MINE, FEBRUARY 4, 2006 Location #3 Belt, 12 Break, Belt entry #3 Belt, 18 Break, Belt entry #3 Belt, 27 Break, Track entry #3 Belt, 28 Break, Track entry #3 Belt, 31 Break, Track entry #3 Belt, 37 Break, Track entry #4 Belt, 13 Break, Belt entry #4 Belt, Between 21 and 22 Break, Track entry #4 Belt, 25 Break, Track entry Appendix S - Page 7 of 8 Comments 6 inches long Branch line drop. Ends appeared to be cut out of the jacket. Cable spliced into main cable. The ends of the cable had been stripped of insulation and covered with black tape. Pigtail connector for branch line. Pigtail connector for branch line, covered with black tape. Cable splice in track entry was clean, appearing to have been new. Bare ends of branch line in belt entry. Wires covered by tape. Branch line drop with bare ends. Mr. Denver Wilfong indicated that he thought he used phone at this location on morning of accident. Appendix S - Executive Summary of "Inspection of Sago Mine Voice Communication Equipment" aID Pager locations as defined Table 1 moo Doom an a DDUUDUD DECEASED 5 BED ?5 <7 BUSES emmgw [3369393 DDID 5 Q7 <70 ,9 ShieldedCable 33x0 0 >30 Dum?g 15OKVAPowerBox 0 Dd Cl 0 a as DPPD steal: a ago no DOES QUEBEC QED 0 <7 Eur: Manama a Egg-om 6 SEEDS EDD I <7 1? 00?? Edi? DQDG -- 0 . a o. No.2 MOtor Chargers 1St NE Moms [3 575 VAC ?fm? su'mp Pump 1] 0 C) 575 VAC Belt DUSter Belt 7200 VAC Line No. 9 110 VAC ump emce emce A 900 pfli?r?ko'l?t?? No. 4 No. 2 trickle Dicey?0;: VAC 480 240?120 VAC P?Wer f?r PMS Shop 8 A.W.G. Conductor Shielded Cable VAC ist Left Belt Starter Shuwe car 575 VAC 1 0 7200 480 VAC 7 Airlock Door_11o VAC Airlock Door_Not 8 A.W.G. coo?3 Distribution Box 480 VAC No. 14 Stanco Pump 7 ,l nun-u @3426 UDEDDQ a ?f DU 3 <7 @0170 Disconnect Switch I: '7 a Airlock Door Transformer (Could not find) VAC - I No. 4 Belt Take-up-575 VAC 1 No. 8 Stanco Pump Starter (Shown at x?cut 31 on electrical map) 2 North Mains 1O slurry Pump?480 VAC 7 VP 7 Airlock In Use inpservicpe DD SQ ?m?w?mw NWP ?wwmm wmml::017055 Legend VCD ??mmw E3 :3 . E: :3 C: DQDEQQU E7 1 Belt Starter?575 VAC 8 Ki?4120 VAC Start/Stop Switch 120 VAC GFI Circuit for No. 3 Sump Pump 575 VAC Pump Not Fan starter 4150 VAC No. 3 Pump?110 VAC E: I: Sn my pump?575 VAC D??r ?0 in Use Feeder 575 VAC Scoop Charger ln Intake En ry No. 3 Pump 1OHP-575 Montrip Charger Original Location (Blown down trook entry)-75 VAC - No. 1 Belt Power Center No. 1 Airlock Door?110 VAC a. a eCted to Power) 4 Belt P?Wer Center 500 KVA 21SC_ShUttle car No. 8 Pump 10 VAC Shielded Cable Welder?575 Connected to Power I gig: . . all"? 5 ea 0C3 No. 3 Belt drive motors :3 Qg?BOttew Locomotive (3 i No. 1 Montnp Chorger?480VAC No:m ruse Dm?m:s PumpgiguOmZAgn Electrical) :imgleitsriaYAfnap) No. 41 Roof Bolter 575 VAC . 2nd Left Belt Power Center I i 110 VAC sump VAC Transformer for a?rlock doors 8&8 Scoop Charger?480 VAC E, . diam center 300 E6 a 8 A.W.G. condUCtor Shielded cable 8 KV 4/0 Gee?3 conductor Shielded cable I Conductor E): a Shielded Cable No. 2 Airlock Door?110 VAC No. 3 Drive Take?up?575 VAC DUDE mm DUDDUDD NOTE: Some portable water pumps may have been relocated. No. 2 Stonoo Pump 480 VAC No. 3 Belt Starter Box?575 VAC No. 3 Belt Power Center No. 3 Distribution Box 480 VAC 500 KVA 7?200_600 VAC 5 3? 5 M?t?r Chmgers No. 4 Piston Pump?10 HP-575 VAC mama QUDU DEED DEE DD . 120 VDC Battery Pod Duster DDUDQDDUD Amxmme1 Sa 0 Mlne, MSHA ID 46-08791 DD El Wolf Run Mining Company [j [1 El DUE Map of the Electrical System, l:l Cl I: Equipment and Associated Items a DEED EU a 250 500 LEGEND Water Pump Wire Mesh Pump Cable Track Belt Seal Location Appendix Y-2 Sago Mine, MSHA ID 46-08791 Wolf Run Mining Company Map of the Electrical System, Equipment and Associated Items 2nd Left Mains and 2 North Mains Inby Crosscut 57 Appendix - Executive Summary of "Portable Gas Detector Testing" US. Department of Labor Mine Safety and Health Administration Industrial Park Road RR1 Box 251 Triadelphia, West Virginia 26059 January 26, 2007 MEMORANDUM FOR RICHARD A. GATES District Manager, Coal Mine Safety and Health District 11 FROM: JOHN P. FAINI 99? Chief, Approval and Certification Center SUBJECT: Executive Summary of Investigation of Portable Gas Detectors Recovered from the Sago Mine The Approval and Certification Center as requested by Coal Mine Safety and Health, conducted a laboratory investigation of gas detectors recovered from a fatal explosion at Wolf Run Mining Company?s Sago Mine, Mine ID. No. 46?08791 on January 2, 2006. These devices were: 0 two (2) Industrial Scientific Corporation (ISC) Model Multi-Gas Monitors; 0 three (3) ISC Model LTX310 Multi~Gas Monitors; and a seven (7) CSE Corporation Model 102LD Methane Detectors. The two ISC Model devices and one of the ISC Model LTX310 units, with Exhibit Numbers beginning with were recovered separately from the others. They were apparently not in the mine at the time of the explosion, but were taken into the mine by mine personnel during attempted rescue operations. The investigation identified several permissibility discrepancies that were attributable to improper maintenance (overdue calibrations, carrying strap grommet displaced and holes in instrument case allowing dust to enter the instrument, and missing case securing screws) or manufacturing discrepancies that deviated from the approved design. There was no evidence that any component of any of the pieces of evidence would have produced conditions that would have provided enough energy to ignite a flammable methane-air mixture. The following sections summarize the testing and inspection of each of the twelve instruments. Appendix - Page 1 of 12 Appendix - Executive Summary of "Portable Gas Detector Testing" ISC Model Multi-Gas Monitor, This ISC Model Multi?Gas Monitor, Serial No. 0408001-374, carried Unique Identifier Exhibit Number ACC-Z was assigned to this exhibit by personnel. The unit was marked with MSHA Approval Number It was inspected and compared with approval documentation. Operational and performance tests were also conducted. The Software Version displayed by the monitor during startup was and the display indicated that the battery was nearly fully charged. The following instrument Peaks were displayed during startup: I CH4 7 2 The "Peak" oxygen reading stored in the monitor "as-received? is a minimum value that had been measured by the monitor; the reading indicated that the monitor was exposed to low concentrations of oxygen. The ?Peak? methane reading stored in the monitor indicated that the monitor was not exposed to high concentrations of a combustible gas. The monitor reported ?No Data Available to download,? indicating that it was not configured to log data. This means that no periodic readings of methane, oxygen, and CO were recorded during use. Therefore, it could not be determined when the ?Peak? readings occurred. FRESH AIR READINGS: Oz .0 OR (flashing) Prior to calibration, the unit indicated over range conditions at oxygen concentrations greater than 19.15%. At lower oxygen concentrations, the monitor read much higher than the sampled concentration. For example, at the lowest sampled concentration (13.04%) the monitor displayed 20.9. The monitor could not detect various concentrations of carbon monoxide (CO) as-received; it would only display The manufacturer?s representative said that this most likely indicated that the monitor?s calibration settings were adjusted in an environment that contained a high concentration of CO in air, resulting in a significant offset in the zero calibration. The methane display readings were lower than the sampled concentration. Readings of 2 Appendix - Page 2 of 12 Appendix - Executive Summary of "Portable Gas Detector Testing" methane concentrations that were 2% and greater were not within the limits of error specified in 30 CFR Part 22. After calibration of the monitor, it: detected methane within the allowable limits of error; - detected oxygen within the requirement; and accurately detected carbon monoxide. The last calibration date for the monitor was given as 11-16-05. This was 47 days before the accident. The time reading from the monitor was approximately one hour and 41 minutes ahead of the actual time. There were minor discrepancies between this monitor and the documentation file. There were bar code labels on the various assemblies of the monitor that are not specified on documentation. There were two cylindrical pieces of foam used as ?dummy? sensors in two unused sensor slots in the monitor that are not shown on the documentation. ISC Model Multi-Gas Monitor, ACC-3 This Model Multi-Gas Monitor, Serial No. 0309270-042 was assigned Exhibit Number The unit was marked with MSHA Approval Number The unit was inspected and compared with approval documentation. Operational and performance tests were also conducted. The Software Version displayed by the monitor during startup was The battery indicator status gave an indication of fully charged. The following Instrument Peaks were displayed during startup: READINGS CH4 Oz co The ?Peak? oxygen reading stored in the monitor, as received, indicated the monitor did not measure low concentrations of oxygen since the last calibration. The ?Peak? methane reading stored in the monitor "as?received? indicates that the monitor was not exposed to high concentrations of a combustible gas. Since the monitor reported ?No Data Available to download?, it could not be determined when these Peak readings occurred. Appendix - Page 3 of 12 Appendix - Executive Summary of "Portable Gas Detector Testing" FRESH READINGS20.9 FAIL Prior to calibration, the unit detected four of the sampled oxygen concentrations within the i0.5% requirement. The only reading that was greater than i0.5% tolerance was the reading of 13.6 when sampling 13.04%. The monitor could not detect various concentrations of carbon monoxide ?as-received?. It would only display The methane display readings were significantly lower than the sampled concentration. For example, the monitor displayed 3.1 when sampling 4% methane. After calibration of the monitor, it: - detected methane within the allowable limits of error; 0 detected oxygen as it did before calibration with the only reading greater than i0.5% tolerance was the reading of 13.6 when sampling 13.04% Oxygen; and accurately detected carbon monoxide. The last calibration date for the monitor was given as 3-1-04. This was 672 days before the accident. The time reading from the monitor was approximately two hours and 6 minutes ahead of the actual time. There were minor discrepancies between this monitor and the documentation file. There were bar code labels on the various assemblies of the monitor that are not specified on documentation. There were two cylindrical pieces of foam used as ?dummy? sensors in two unused sensor slots in the monitor that are not shown on the documentation. Model LTX310 Multi?Gas Monitor, This Model LTX310 Multi-Gas monitor, Serial No. 9710027-116 was assigned Exhibit Number ACC-1. The unit was marked with MSHA Approval Number 8C-65-2. The unit was inspected and compared with approval documentation. Operational and performance tests were also conducted. As-received, the monitor did not operate due to a depleted battery. After charging, the monitor was in operational status. However, it was programmed to display the reading of combustible gas concentration as percent LEL (Lower Explosive Limit). The 4 Appendix - Page 4 of 12 Appendix - Executive Summary of "Portable Gas Detector Testing" following Instrument Peaks were displayed during startup: . PEAKREADDKB r_ Oxygen Toxic 18.4 . The ?Peak? oxygen reading stored in the monitor ?as-received" indicated the monitor was exposed to a low concentration of oxygen since the last calibration. The ?Peak? reading stored in the monitor "as-received? indicated that the monitor was exposed to a high concentration of a combustible gas. The ?Peak? carbon monoxide reading indicated that the monitor was exposed to a high level of carbon monoxide. It could not be determined when these Peak readings occurred. FRESH AIR READINGS: Toxic CQ-15 02 i i 209 0 Prior to calibration, the unit detected five sampled oxygen concentrations within the i0.5% requirement. The monitor could accurately detect the two sample concentrations of carbon monoxide "as-received" even with the ?15 offset in fresh air. The methane display readings were significantly lower than the sampled concentration and were displayed as percent LEL. The monitor was programmed to display the combustible gas readings as methane by volume, and the monitor was calibrated. After calibration of the monitor, it: detected methane within requirements at 0.25%, 0.50%, and 1.00% only. It read significantly lower at the higher sampled concentrations; - detected oxygen within the i0.5% requirement; and accurately detected carbon monoxide and no longer had the zero offset. A calibration label could not be found on the monitor. It could not be determined when the monitor was last calibrated. There were minor discrepancies between this monitor and the documentation file. There were bar code labels on the various assemblies of the monitor that are not specified on documentation. There were labels on the monitor that were probably applied by the mine for identification purposes that are not on the documentation. Appendix - Page 5 of 12 :l Appendix - Executive Summary of "Portable Gas Detector Testing" ISC Model LTX310 Multi-Gas Monitor, KLH-4 The serial number on this unit is 9609008-244. The unit was marked with Approval Number The unit was inspected and compared with approval documentation. Operational and performance tests were also conducted. As-received, the monitor did not Operate due to a depleted battery. After charging, the monitor was in operational status. The following instrument Peaks were displayed during startup: PEAK READINGS CH4 Oxygen Toxic 1.7 18.7 59 The oxygen ?Peak? reading stored in the monitor ?as?received? indicates the monitor was exposed to a low concentration of oxygen. The CH4 "Peak" reading stored in the monitor ?as-receivec indicates that the monitor was exposed to a high concentration of a combustible gas. The carbon monoxide "Peak" reading indicates that the monitor was exposed to a high level of carbon monoxide. It could not be determined when these Peak readings occurred. FRESH AIR READINGS: CH4 02 7 i Toxic 20.4 7 Prior to performance testing, the monitor no longer displayed a reading for the oxygen sensor and oxygen accuracy testing could not be conducted. Prior to calibration, the monitor could not accurately detect two sampled concentrations of carbon monoxide. It gave a display reading of 198 with 50ppm of CO applied and a display reading of 402 with 100 of CO applied. The methane display readings were higher than the MSHA limits of error at 0.50%, 1.00%, and 2.00%. After calibration of the monitor, it: 0 did not detect methane within MSI-IA requirements at 0.50%, 1.00%, and 2.00%. It read higher at all the other sampled concentrations; - accurately detected carbon monoxide; and the oxygen sensor could not be calibrated due to the blank display for oxygen. 0 Appendix - Page 6 of 12 Appendix - Executive Summary of "Portable Gas Detector Testing" A calibration label could not be found on the monitor. It could not be determined when the monitor was last calibrated. There were minor discrepancies between this monitor and the documentation file. The part number of the combustible gas sensor (1704-6269) did not agree with the part number (1704A1856) shown on the approval documentation. The part number 1704? 6269 sensor assembly is approved in other ISC instruments and the 1704A1856 sensor is a sub-assembly of the part number 1704-6269 sensor assembly. One of the case securing screws was missing. ISC Model LTX310 Multi-Gas Monitor, KLH-15 The unit was marked with MSHA Approval Number The unit was inspected and compared with approval documentation. Operational and performance tests were also conducted. As?received, the monitor did not operate due to a depleted battery. After charging, the monitor was in operational status. The following instrument Peaks were displayed during startup: PEAK READINGS Oxygen I Toxic? 14._6 The "Peak" oxygen reading stored in the monitor is a minimum value that had been measured by the monitor; the reading indicated that the monitor was exposed to a very low concentration of oxygen. The ?Peak? CH4 reading stored in the monitor indicated that the monitor was exposed to a high concentration of a combustible gas. The ?Peak? carbon monoxide reading indicated that the monitor was exposed to a high level of carbon monoxide. It could not be determined when these Peak readings occurred. FRESH AIR READINGS: CH4 Oz Toxic 21.3 7 COH1 Prior to calibration, the monitor detected the five sampled oxygen concentrations within the i0.5% of reading requirement. The monitor could not accurately detect the sampled concentrations of carbon monoxide. It gave a diSplay reading of 102 with 50ppm of CO applied and a display reading of 191 with 100 of CO applied. The methane display readings were lower than the sampled concentrations and not within MSHA limits of error at all sampled concentrations. After calibration of the monitor, it: 7 Appendix - Page 7 of 12 Appendix - Executive Summary of "Portable Gas Detector Testing" 0 detected methane accurately at all sampled concentrations; - accurately detected carbon monoxide; and - accurately detected the various oxygen concentrations. A partial calibration label was found on the monitor with only the month (7) and day (22) legible on it. It could not be determined when the monitor was last calibrated. There were discrepancies between this monitor and the documentation file. The identifying part number on the buzzer in the monitor (PB-1220P) did not agree with the buzzer part number specified on the documentation. There was a jumper wire on the bottom side of the main PCB that is not shown on the documentation. One of the case securing screws was missing. The part number on the oxygen sensor in the unit (1703-5114) did not agree with the part number (1702-3516) specified for the sensor on the documentation. CSE Model 102LD Methane Detector, KLl-l-2 The serial number on this unit is 4421. The unit was marked with MSHA Approval Number 8C-37-7. The unit was inspected and compared with approval documentation. Operational and performance tests were also conducted. As-received, the instrument had sufficient charge on the battery and was in operational status. Prior to calibration, the instrument read 0.0 in fresh air and 2.3 with 2.5% calibration gas applied. It detected all sampled methane concentrations accurately except for the 4.00% methane concentration. It read low (3.6) at the 4.00% concentration. After calibration, the instrument detected methane accurately at all sampled concentrations. A CSE calibration label was found on the instrument with a calibration date of 12/ 21/ 05. At the time of the accident, this instrument had a valid calibration meeting the MST-IA requirement of being calibrated every 31 days. There were minor discrepancies between this Detector and the documentation file. The revision level of the main PCB assembly in the instrument was marked Revision C. The latest revision level of the documentation on file for the main PCB is B. The markings on R3 on the main PCB in the instrument are 4310R-1 02-124 which disagrees with the documentation which specifies part number 8 Appendix - Page 8 of 12 Appendix - Executive Summary of "Portable Gas Detector Testing" CSE Model 102LD Methane Detector, The serial number on this unit is 1870. The unit was marked with MSHA Approval Number 8C-37-4. The unit was inspected and compared with approval documentation. Operational and performance tests were also conducted. As-received, the instrument had sufficient charge on the battery and was in operational status. Prior to calibration, the instrument read 0.1 in fresh air and 1.9 with 2.5% calibration gas applied. The only sampled methane concentration that it read accurately was the 0.25% methane concentration. It read low at all the other sampled concentrations. After calibration, the instrument detected methane accurately at all sampled concentrations. A CSE calibration label was found on the instrument with a calibration date of 11/ 7/05. This was 56 days before the accident. There were very minor discrepancies between this Detector and the documentation file. CSE Model 102LD Methane Detector, KLI-I-21 The serial number on this unit is 4277. The unit was marked with Approval Number 8C-37-7. The unit was inspected and compared with approval documentation. Operational and performance tests were also conducted. As-received, the instrument had sufficient charge on the battery and was in operational status. Prior to calibration, the instrument read 0.1 in fresh air and 3.4 with 2.5% calibration gas applied. The instrument did not read any of the sampled methane concentrations accurately. It read high at all sampled concentrations. After calibration, the instrument detected methane accurately at all sampled concentrations. A calibration label could not be found on the instrument. It could not be determined when the instrument was last calibrated. The revision level of the main PCB assembly in the instrument was marked Revision C. 9 Appendix - Page 9 of 12 Appendix - Executive Summary of "Portable Gas Detector Testing" The latest revision level of the documentation on file for the main PCB is B. The markings on R3 on the main PCB in the instrument are 4310R-102-124 which disagrees with the documentation which specifies part number The grommet that surrounds the carrying strap was displaced from the case and was not located as shown on the documentation. CSE Model 102LD Methane Detector, GH-45P The serial number on this unit is 2064. The unit was marked with MSHA Approval Number The unit was inspected and compared with approval documentation. Operational and performance tests were also conducted. As-received, the instrument had sufficient charge on the battery and was in operational status. Prior to calibration, the instrument read 0.0 in fresh air and 1.4 with 2.5% calibration gas applied. The only sampled concentration that the instrument read accurately was the 0.25% methane concentration. It read low at all other sampled concentrations. After calibration, the instrument read accurately at the 0.25%, 0.50%, and 1.00% methane concentrations. It read low at the other higher concentrations. A CSE calibration label was found on the instrument with a calibration date of 06/10/05. This was 206 days before the accident. The revision level of the main PCB assembly in the instrument was marked Revision C. The latest revision level of the documentation on file for the main PCB is B. The markings on R3 on the main PCB in the instrument are which disagrees with the documentation which specifies part number The two screws that secure the detector assembly to the instrument are flat head instead of the specified Phillips head. CSE Model 102LD Methane The serial number on this unit is 4588. The unit was marked with Approval Number 8C-37-7. The unit was inspected and compared with approval documentation. Operational and performance tests were also conducted. As-received, the instrument had sufficient charge on the battery and was in operational status. Prior to calibration, the instrument read 0.0 in fresh air and 2.6 with 2.5% calibration gas 10 Appendix - Page 10 of 12 Appendix - Executive Summary of "Portable Gas Detector Testing" applied. It read accurately at all sampled concentrations. After calibration, the instrument continued to read all sampled concentrations accurately. A CSE calibration label was found on the instrument with a calibration date of 12/ 12/05. At the time of the accident, this instrument had a valid calibration meeting the MSHA requirement of being calibrated every 31 days. The revision level of the main PCB assembly in the instrument was marked Revision C. The latest revision level of the documentation on file for the main PCB is B. The markings on R3 on the main PCB in the instrument are 4310R-102-124 which disagrees with the documentation which specifies part number CSE Model 102LD Methane Detector, The serial number on this unit is 4961. The unit was marked with MSHA Approval Number 8C-37-7. The unit was inspected and compared with approval documentation. Operational and performance tests were also conducted. As?received, the instrument did not operate due to a depleted battery. After charging, the instrument was in operational status. Prior to calibration, the instrument read 0.2 in fresh air and 2.2 with 2.5% calibration gas applied. The only sampled concentrations that were within the MSHA limits of error were the readings at 1.00%, 2.00%, and 3.00%. It read low at the other measured concentrations. After calibration, the instrument read all sampled concentrations accurately. A CSE calibration label was found on the instrument with a calibration date of 10/ 18/ 05. This was 76 days before the accident. The revision level of the main PCB assembly in the instrument was marked Revision C. The latest revision level of the documentation on file for the main PCB is B. The markings on R3 on the main PCB in the instrument are which disagrees with the documentation which specifies part number CSE Model 102LD Methane Detector, GH-87P The serial number on this unit is 4843. The unit was marked with Approval Number 8C-37-7. The unit was inspected and compared with approval documentation. 1 Appendix - Page 11 of 12 Appendix - Executive Summary of "Portable Gas Detector Testing" Operational and performance tests were also conducted. As-received, the instrument had sufficient charge on the battery and was in operational status. Prior to calibration, the instrument read 0.0 in fresh air and 2.4 with 2.5% calibration gas applied. The only sampled concentration that was not within the MSHA limits of error was the reading at 4.00%. After calibration, the instrument read all sampled concentrations accurately. A calibration label could not be found on the instrument. It could not be determined when the instrument was last calibrated. The revision level of the main PCB assembly in the instrument was marked Revision C. The latest revision level of the documentation on file for the main PCB is B. The markings on R3 on the main PCB in the instrument are 4310R-102-124 which disagrees with the documentation which specifies part number 24. One of the two screws that secure the detector assembly to the instrument was a flat head instead of the Specified Phillips head. Comprehensive test results can be obtained from the Chief of the RR 1, Box 251, Industrial Park Road, Triadelphia, West Virginia 26059. 12 Appendix - Page 12 of 12 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports Aug 11 2006 6:46:06 PM Nicole Ferro Thank you for using Vaisala's STRIKEnet® to validate the referenced claim. Your report was generated using data from Vaisala's National Lightning Detection Network®, the most comprehensive archive database in North America. STRIKEnet Report 168384 Claim Number: Insured/Claimant Name: Approx. Claim/Loss Value: Items Damaged/Loss Type: Claim Address: Search Period: Search Center Point: Search Radius: Jan 1 2006 10:00:00 PM US/Eastern Jan 2 2006 10:00:00 PM US/Eastern 38.940940° N (Latitude), 80.202310° W (Longitude) 15 mi/25 km around the given location. Comments: 162 strikes were detected by the National Lightning Detection Network for the given time period and location. Thank you again for selecting STRIKEnet. If you have any questions please contact us at 1 800 283 4557 or thunderstorm.support@vaisala.com. Best Regards, The Vaisala STRIKEnet Team Vaisala Inc. Tucson Operations 2705 E. Medina Road Tucson, AZ 85706, USA thunderstorm.vaisala.com Tel. +1 520 806 7300 Fax +1 520 741 2848 thunderstorm.sales@vaisala.com Appendix AA - Page 1 of 31 Aug 11 2006 6:46:06 PM GMT Page 1 of 18 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports STRIKEnet Report 168384 Report Title: 60-06MR-308 Total Lightning Strokes Detected: 162 Lightning Strokes Detected within 15 mi/25 km radius: 128 Lightning Strokes Detected beyond 15 mi/25 km whose confidence ellipse overlaps the radius: 34 Search Radius: 15 mi/25 km Time Span: Jan 1 2006 10:00:00 PM US/Eastern to Jan 2 2006 10:00:00 PM US/Eastern Location Points For Lightning Strokes Lightning data provided by Vaisala's NLDN® and/or Environment Canada's CLDN. Vaisala Inc. Tucson Operations 2705 E. Medina Road Tucson, AZ 85706, USA thunderstorm.vaisala.com Tel. +1 520 806 7300 Fax +1 520 741 2848 thunderstorm.sales@vaisala.com Appendix AA - Page 2 of 31 Aug 11 2006 6:46:06 PM GMT Page 2 of 18 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports STRIKEnet Report 168384 Report Title: 60-06MR-308 Total Lightning Strokes Detected: 162 Lightning Strokes Detected within 15 mi/25 km radius: 128 Lightning Strokes Detected beyond 15 mi/25 km whose confidence ellipse overlaps the radius: 34 Search Radius: 15 mi/25 km Time Span: Jan 1 2006 10:00:00 PM US/Eastern to Jan 2 2006 10:00:00 PM US/Eastern Confidence Ellipses For Lightning Strokes Lightning data provided by Vaisala's NLDN® and/or Environment Canada's CLDN. Note: These ellipses indicate a 99% certainty that the recorded lightning event contacted the ground within the bounds of the ellipse. Vaisala Inc. Tucson Operations 2705 E. Medina Road Tucson, AZ 85706, USA thunderstorm.vaisala.com Tel. +1 520 806 7300 Fax +1 520 741 2848 thunderstorm.sales@vaisala.com Appendix AA - Page 3 of 31 Aug 11 2006 6:46:06 PM GMT Page 3 of 18 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports STRIKEnet Report 168384 Area Of Study With Center Point Vaisala Inc. Tucson Operations 2705 E. Medina Road Tucson, AZ 85706, USA thunderstorm.vaisala.com Tel. +1 520 806 7300 Fax +1 520 741 2848 thunderstorm.sales@vaisala.com Appendix AA - Page 4 of 31 Aug 11 2006 6:46:06 PM GMT Page 4 of 18 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports STRIKEnet Report 168384 Report Title: 60-06MR-308 Total Lightning Strokes Detected: 162 Lightning Strokes Detected within 15 mi/25 km radius: 128 Lightning Strokes Detected beyond 15 mi/25 km whose confidence ellipse overlaps the radius: 34 Search Radius: 15 mi/25 km Time Span: Jan 1 2006 10:00:00 PM US/Eastern to Jan 2 2006 10:00:00 PM US/Eastern Lightning Stroke Table (Note: Earliest 50 events shown. Events ordered by time.) Peak Distance From Date Time Current (kA) Center (mi/km) Latitude Longitude Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 4:14:03 AM 5:33:58 AM 5:35:41 AM 5:35:41 AM 5:36:14 AM 5:36:14 AM 5:36:14 AM 5:43:55 AM 5:43:55 AM 5:51:33 AM 5:55:25 AM 5:57:41 AM 5:57:48 AM 6:00:09 AM 6:04:01 AM 6:04:01 AM 6:04:01 AM 6:04:12 AM 6:04:12 AM 6:05:30 AM 6:06:16 AM 6:06:16 AM 6:07:26 AM 6:08:29 AM 6:08:29 AM 6:09:23 AM 6:09:23 AM 6:09:23 AM 6:10:32 AM 6:10:32 AM 6:10:32 AM 6:10:32 AM -5.3 26.5 -29.1 -16.7 -61.8 -12.1 -8.3 -5.7 -4.0 12.6 -5.8 -7.8 25.1 12.5 -60.5 -11.9 -15.8 23.9 -33.9 -6.0 -7.5 -9.3 -48.8 -88.7 -15.3 -23.7 -7.1 -15.3 -12.7 -5.9 -8.9 -8.6 22.1/35.7 15.4/24.8 15.7/25.3 15.7/25.3 14.3/23.0 13.0/21.0 13.4/21.6 15.8/25.6 15.7/25.4 12.9/20.8 15.3/24.7 20.1/32.4 6.5/10.6 10.0/16.2 15.9/25.6 16.0/25.8 15.6/25.2 13.4/21.6 14.1/22.7 14.6/23.5 17.1/27.5 15.2/24.6 15.9/25.6 9.9/16.0 9.5/15.3 16.1/25.9 16.3/26.4 14.0/22.5 15.7/25.4 15.4/24.9 15.4/24.8 13.8/22.3 38.6437 38.7260 38.7178 38.7181 38.7462 38.7684 38.7619 38.7334 38.7347 38.9451 38.7407 38.7366 38.8487 38.8131 38.7107 38.7090 38.7150 38.9601 38.9427 38.7303 38.6943 38.7200 38.7122 38.9486 38.9408 38.7083 38.7045 38.7388 38.7151 38.7177 38.7190 38.7413 -80.3571 -80.2798 -80.1439 -80.1427 -80.2931 -80.3012 -80.2993 -80.0757 -80.0765 -80.4429 -80.3269 -79.9362 -80.2310 -80.2920 -80.2097 -80.2116 -80.2163 -80.4506 -80.4652 -80.1803 -80.1740 -80.2049 -80.1721 -80.3875 -80.3788 -80.1837 -80.1816 -80.2225 -80.2443 -80.1791 -80.1787 -80.1827 Vaisala Inc. Tucson Operations 2705 E. Medina Road Tucson, AZ 85706, USA thunderstorm.vaisala.com Tel. +1 520 806 7300 Fax +1 520 741 2848 thunderstorm.sales@vaisala.com Appendix AA - Page 5 of 31 Aug 11 2006 6:46:06 PM GMT Page 5 of 18 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports Peak Distance From Date Time Current (kA) Center (mi/km) Latitude Longitude Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 6:10:32 AM 6:12:16 AM 6:12:16 AM 6:12:16 AM 6:13:08 AM 6:13:08 AM 6:13:08 AM 6:13:08 AM 6:14:27 AM 6:14:27 AM 6:15:13 AM 6:15:13 AM 6:15:13 AM 6:15:13 AM 6:15:14 AM 6:15:22 AM 6:15:23 AM 6:17:13 AM -15.6 14.0 -27.8 -8.3 -19.1 -17.6 -16.0 -10.8 -35.8 6.3 20.5 -133.9 -35.9 -5.4 -8.1 -34.8 -11.1 -14.2 12.3/19.8 11.8/19.1 13.0/21.0 13.0/20.9 13.7/22.0 13.7/22.1 6.3/10.2 7.2/11.6 9.7/15.7 14.2/23.0 9.4/15.2 9.4/15.2 9.4/15.2 10.6/17.2 11.2/18.1 13.6/21.9 13.5/21.8 12.1/19.5 38.7698 38.8615 38.8467 38.8431 38.7436 38.7427 38.9792 38.9956 38.8361 38.7359 38.8770 38.8775 38.8751 38.8620 38.8588 38.7465 38.7469 38.8164 -80.1395 -80.3974 -80.4129 -80.4086 -80.1817 -80.1775 -80.3097 -80.3168 -80.0813 -80.1713 -80.3572 -80.3573 -80.3565 -80.3729 -80.3835 -80.1619 -80.1628 -80.0430 Vaisala Inc. Tucson Operations 2705 E. Medina Road Tucson, AZ 85706, USA thunderstorm.vaisala.com Tel. +1 520 806 7300 Fax +1 520 741 2848 thunderstorm.sales@vaisala.com Appendix AA - Page 6 of 31 Aug 11 2006 6:46:06 PM GMT Page 6 of 18 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports STRIKEnet Report 168384 Report Title: 60-06MR-308 Total Lightning Strokes Detected: 162 Lightning Strokes Detected within 15 mi/25 km radius: 128 Lightning Strokes Detected beyond 15 mi/25 km whose confidence ellipse overlaps the radius: 34 Search Radius: 15 mi/25 km Time Span: Jan 1 2006 10:00:00 PM US/Eastern to Jan 2 2006 10:00:00 PM US/Eastern Lightning Stroke Table (Note: Closest 50 events shown. Events ordered by distance.) Peak Distance From Date Time Current (kA) Center (mi/km) Latitude Longitude Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 6:26:35 AM 6:26:35 AM 9:30:44 AM 6:38:51 AM 6:38:51 AM 8:30:44 PM 7:36:46 AM 6:38:51 AM 6:13:08 AM 5:57:48 AM 7:22:01 AM 7:22:01 AM 6:13:08 AM 6:29:42 AM 7:11:49 AM 7:03:33 AM 7:52:21 AM 6:15:13 AM 6:15:13 AM 6:15:13 AM 6:08:29 AM 6:14:27 AM 7:09:31 AM 6:08:29 AM 6:00:09 AM 10:52:47 AM 6:15:13 AM 6:57:04 AM 7:35:11 PM 7:03:33 AM 7:33:54 PM 10:56:18 AM 101.0 38.8 27.5 85.7 -12.6 18.1 -5.7 -86.0 -16.0 25.1 23.7 -19.4 -10.8 19.3 87.8 -20.9 47.4 -133.9 20.5 -35.9 -15.3 -35.8 178.8 -88.7 12.5 -5.4 -5.4 -9.3 -6.0 77.4 -14.6 32.5 1.9/3.1 3.4/5.5 3.6/5.7 4.4/7.1 4.9/7.8 5.4/8.7 5.8/9.3 6.2/9.9 6.3/10.2 6.5/10.6 6.6/10.7 7.0/11.2 7.2/11.6 7.5/12.1 7.9/12.8 8.3/13.3 8.3/13.4 9.4/15.2 9.4/15.2 9.4/15.2 9.5/15.3 9.7/15.7 9.8/15.8 9.9/16.0 10.0/16.2 10.4/16.8 10.6/17.2 10.7/17.3 10.9/17.5 10.9/17.5 10.9/17.6 11.1/17.8 38.9260 38.8968 38.9003 38.9805 38.9748 38.9289 39.0048 38.9954 38.9792 38.8487 38.8570 38.8500 38.9956 38.8782 38.8344 38.8292 38.8949 38.8775 38.8770 38.8751 38.9408 38.8361 38.8130 38.9486 38.8131 38.8012 38.8620 38.8845 39.0867 38.8354 38.9679 38.8218 -80.2331 -80.2313 -80.2431 -80.1380 -80.1227 -80.3014 -80.2719 -80.1113 -80.3097 -80.2310 -80.1420 -80.1457 -80.3168 -80.0886 -80.2577 -80.2578 -80.0595 -80.3573 -80.3572 -80.3565 -80.3788 -80.0813 -80.2805 -80.3875 -80.2920 -80.1279 -80.3729 -80.0159 -80.1257 -80.3530 -80.0018 -80.0641 Vaisala Inc. Tucson Operations 2705 E. Medina Road Tucson, AZ 85706, USA thunderstorm.vaisala.com Tel. +1 520 806 7300 Fax +1 520 741 2848 thunderstorm.sales@vaisala.com Appendix AA - Page 7 of 31 Aug 11 2006 6:46:06 PM GMT Page 7 of 18 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports Peak Distance From Date Time Current (kA) Center (mi/km) Latitude Longitude Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 6:15:14 AM 7:07:03 AM 7:03:33 AM 7:54:44 AM 7:35:55 PM 7:54:44 AM 7:54:44 AM 6:57:04 AM 10:57:53 AM 6:12:16 AM 6:17:13 AM 10:56:18 AM 6:17:13 AM 6:17:13 AM 6:37:59 AM 10:56:18 AM 6:17:13 AM 7:37:00 PM -8.1 -198.4 -35.7 -9.0 -8.6 -23.3 -17.8 -20.7 -51.1 14.0 -10.9 -47.4 -9.3 -7.5 -27.8 18.1 -14.2 -8.4 11.2/18.1 11.3/18.2 11.3/18.2 11.6/18.7 11.7/18.8 11.7/18.9 11.7/18.9 11.7/18.9 11.8/19.0 11.8/19.1 11.9/19.2 12.0/19.4 12.0/19.4 12.1/19.5 12.1/19.5 12.1/19.5 12.1/19.5 12.2/19.6 38.8588 38.7912 38.8191 38.7732 38.9710 38.7714 38.7714 38.8925 38.8009 38.8615 38.8195 38.7952 38.8183 38.8170 38.9342 38.7945 38.8164 38.9826 -80.3835 -80.2878 -80.3433 -80.2165 -79.9878 -80.2190 -80.2191 -79.9923 -80.0760 -80.3974 -80.0443 -80.0790 -80.0427 -80.0436 -79.9771 -80.0782 -80.0430 -79.9816 Vaisala Inc. Tucson Operations 2705 E. Medina Road Tucson, AZ 85706, USA thunderstorm.vaisala.com Tel. +1 520 806 7300 Fax +1 520 741 2848 thunderstorm.sales@vaisala.com Appendix AA - Page 8 of 31 Aug 11 2006 6:46:06 PM GMT Page 8 of 18 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports STRIKEnet Report 168384 Report Title: 60-06MR-308 Total Lightning Strokes Detected: 162 Lightning Strokes Detected within 15 mi/25 km radius: 128 Lightning Strokes Detected beyond 15 mi/25 km whose confidence ellipse overlaps the radius: 34 Search Radius: 15 mi/25 km Time Span: Jan 1 2006 10:00:00 PM US/Eastern to Jan 2 2006 10:00:00 PM US/Eastern Lightning Stroke Table (Note: All events shown. Events ordered by time.) Peak Distance From Date Time Current (kA) Center (mi/km) Latitude Longitude Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 4:14:03 AM 5:33:58 AM 5:35:41 AM 5:35:41 AM 5:36:14 AM 5:36:14 AM 5:36:14 AM 5:43:55 AM 5:43:55 AM 5:51:33 AM 5:55:25 AM 5:57:41 AM 5:57:48 AM 6:00:09 AM 6:04:01 AM 6:04:01 AM 6:04:01 AM 6:04:12 AM 6:04:12 AM 6:05:30 AM 6:06:16 AM 6:06:16 AM 6:07:26 AM 6:08:29 AM 6:08:29 AM 6:09:23 AM 6:09:23 AM 6:09:23 AM 6:10:32 AM 6:10:32 AM 6:10:32 AM 6:10:32 AM -5.3 26.5 -29.1 -16.7 -61.8 -12.1 -8.3 -5.7 -4.0 12.6 -5.8 -7.8 25.1 12.5 -60.5 -11.9 -15.8 23.9 -33.9 -6.0 -7.5 -9.3 -48.8 -88.7 -15.3 -23.7 -7.1 -15.3 -12.7 -5.9 -8.9 -8.6 22.1/35.7 15.4/24.8 15.7/25.3 15.7/25.3 14.3/23.0 13.0/21.0 13.4/21.6 15.8/25.6 15.7/25.4 12.9/20.8 15.3/24.7 20.1/32.4 6.5/10.6 10.0/16.2 15.9/25.6 16.0/25.8 15.6/25.2 13.4/21.6 14.1/22.7 14.6/23.5 17.1/27.5 15.2/24.6 15.9/25.6 9.9/16.0 9.5/15.3 16.1/25.9 16.3/26.4 14.0/22.5 15.7/25.4 15.4/24.9 15.4/24.8 13.8/22.3 38.6437 38.7260 38.7178 38.7181 38.7462 38.7684 38.7619 38.7334 38.7347 38.9451 38.7407 38.7366 38.8487 38.8131 38.7107 38.7090 38.7150 38.9601 38.9427 38.7303 38.6943 38.7200 38.7122 38.9486 38.9408 38.7083 38.7045 38.7388 38.7151 38.7177 38.7190 38.7413 -80.3571 -80.2798 -80.1439 -80.1427 -80.2931 -80.3012 -80.2993 -80.0757 -80.0765 -80.4429 -80.3269 -79.9362 -80.2310 -80.2920 -80.2097 -80.2116 -80.2163 -80.4506 -80.4652 -80.1803 -80.1740 -80.2049 -80.1721 -80.3875 -80.3788 -80.1837 -80.1816 -80.2225 -80.2443 -80.1791 -80.1787 -80.1827 Vaisala Inc. Tucson Operations 2705 E. Medina Road Tucson, AZ 85706, USA thunderstorm.vaisala.com Tel. +1 520 806 7300 Fax +1 520 741 2848 thunderstorm.sales@vaisala.com Appendix AA - Page 9 of 31 Aug 11 2006 6:46:06 PM GMT Page 9 of 18 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports Peak Distance From Date Time Current (kA) Center (mi/km) Latitude Longitude Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 6:10:32 AM 6:12:16 AM 6:12:16 AM 6:12:16 AM 6:13:08 AM 6:13:08 AM 6:13:08 AM 6:13:08 AM 6:14:27 AM 6:14:27 AM 6:15:13 AM 6:15:13 AM 6:15:13 AM 6:15:13 AM 6:15:14 AM 6:15:22 AM 6:15:23 AM 6:17:13 AM 6:17:13 AM 6:17:13 AM 6:17:13 AM 6:17:14 AM 6:18:11 AM 6:18:11 AM 6:18:11 AM 6:19:55 AM 6:21:23 AM 6:22:15 AM 6:22:15 AM 6:22:15 AM 6:26:35 AM 6:26:35 AM 6:29:42 AM 6:34:55 AM 6:37:59 AM 6:38:51 AM 6:38:51 AM 6:38:51 AM 6:49:31 AM 6:49:31 AM 6:51:41 AM 6:51:41 AM 6:53:39 AM -15.6 14.0 -27.8 -8.3 -19.1 -17.6 -16.0 -10.8 -35.8 6.3 20.5 -133.9 -35.9 -5.4 -8.1 -34.8 -11.1 -14.2 -9.3 -7.5 -10.9 -9.1 -17.9 -14.9 -4.8 -7.9 8.8 -9.4 -19.5 8.3 38.8 101.0 19.3 -116.3 -27.8 -12.6 85.7 -86.0 -13.4 -23.4 -5.9 -8.0 -12.7 12.3/19.8 11.8/19.1 13.0/21.0 13.0/20.9 13.7/22.0 13.7/22.1 6.3/10.2 7.2/11.6 9.7/15.7 14.2/23.0 9.4/15.2 9.4/15.2 9.4/15.2 10.6/17.2 11.2/18.1 13.6/21.9 13.5/21.8 12.1/19.5 12.0/19.4 12.1/19.5 11.9/19.2 12.9/20.8 13.9/22.5 14.0/22.6 13.4/21.7 13.4/21.6 13.7/22.1 12.9/20.8 13.1/21.1 12.9/20.9 3.4/5.5 1.9/3.1 7.5/12.1 15.6/25.2 12.1/19.5 4.9/7.8 4.4/7.1 6.2/9.9 14.9/24.0 14.8/23.9 14.3/23.1 14.8/24.0 14.9/24.0 38.7698 38.8615 38.8467 38.8431 38.7436 38.7427 38.9792 38.9956 38.8361 38.7359 38.8770 38.8775 38.8751 38.8620 38.8588 38.7465 38.7469 38.8164 38.8183 38.8170 38.8195 38.7586 38.7477 38.7492 38.7555 38.7612 38.7715 38.7832 38.7797 38.7808 38.8968 38.9260 38.8782 38.8325 38.9342 38.9748 38.9805 38.9954 38.8643 38.8650 38.8741 38.8539 38.8956 -80.1395 -80.3974 -80.4129 -80.4086 -80.1817 -80.1775 -80.3097 -80.3168 -80.0813 -80.1713 -80.3572 -80.3573 -80.3565 -80.3729 -80.3835 -80.1619 -80.1628 -80.0430 -80.0427 -80.0436 -80.0443 -80.1463 -80.1268 -80.1164 -80.1257 -80.1087 -80.0681 -80.0733 -80.0738 -80.0763 -80.2313 -80.2331 -80.0886 -79.9468 -79.9771 -80.1227 -80.1380 -80.1113 -79.9435 -79.9439 -79.9501 -79.9492 -79.9316 Vaisala Inc. Tucson Operations 2705 E. Medina Road Tucson, AZ 85706, USA thunderstorm.vaisala.com Tel. +1 520 806 7300 Fax +1 520 741 2848 thunderstorm.sales@vaisala.com Appendix AA - Page 10 of 31 Aug 11 2006 6:46:06 PM GMT Page 10 of 18 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports Peak Distance From Date Time Current (kA) Center (mi/km) Latitude Longitude Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 6:55:33 AM 6:55:33 AM 6:55:33 AM 6:57:04 AM 6:57:04 AM 6:57:04 AM 7:02:39 AM 7:03:33 AM 7:03:33 AM 7:03:33 AM 7:07:03 AM 7:09:31 AM 7:11:49 AM 7:22:01 AM 7:22:01 AM 7:25:06 AM 7:25:35 AM 7:34:50 AM 7:35:26 AM 7:36:15 AM 7:36:15 AM 7:36:46 AM 7:41:00 AM 7:42:35 AM 7:42:35 AM 7:42:35 AM 7:42:35 AM 7:42:35 AM 7:42:35 AM 7:42:35 AM 7:50:24 AM 7:52:21 AM 7:54:44 AM 7:54:44 AM 7:54:44 AM 7:56:59 AM 8:03:36 AM 9:16:58 AM 9:19:08 AM 9:30:44 AM 9:32:24 AM 9:45:50 AM 10:43:53 AM -57.6 -8.0 81.5 -20.7 -9.3 -12.8 -10.5 -20.9 77.4 -35.7 -198.4 178.8 87.8 -19.4 23.7 -11.3 -9.9 -8.6 -4.9 -16.7 -14.9 -5.7 -8.5 -5.6 -8.7 -9.8 -12.0 -21.5 -9.1 10.6 -9.1 47.4 -17.8 -23.3 -9.0 35.3 69.4 -12.5 -30.1 27.5 -10.5 -11.4 -18.1 14.3/23.1 13.7/22.1 15.8/25.4 11.7/18.9 10.7/17.3 12.8/20.6 18.7/30.2 8.3/13.3 10.9/17.5 11.3/18.2 11.3/18.2 9.8/15.8 7.9/12.8 7.0/11.2 6.6/10.7 13.4/21.6 14.8/23.9 14.1/22.8 18.0/29.0 14.9/24.0 15.0/24.2 5.8/9.3 15.3/24.7 13.3/21.4 14.9/24.0 15.6/25.2 15.6/25.2 15.7/25.3 12.7/20.5 13.7/22.1 17.5/28.2 8.3/13.4 11.7/18.9 11.7/18.9 11.6/18.7 15.1/24.4 14.9/24.1 13.7/22.0 13.9/22.5 3.6/5.7 15.8/25.4 30.4/49.0 13.1/21.1 38.8105 38.8093 38.7927 38.8925 38.8845 38.8824 38.6788 38.8292 38.8354 38.8191 38.7912 38.8130 38.8344 38.8500 38.8570 38.7715 38.8894 38.8245 38.7189 38.8337 38.8287 39.0048 38.7197 38.7489 38.7257 38.7148 38.7145 38.7133 38.7567 38.7425 38.6882 38.8949 38.7714 38.7714 38.7732 38.7267 38.8919 38.7963 38.8256 38.9003 38.8046 38.8504 38.7571 -80.4102 -80.3942 -80.4256 -79.9923 -80.0159 -79.9765 -80.1120 -80.2578 -80.3530 -80.3433 -80.2878 -80.2805 -80.2577 -80.1457 -80.1420 -80.0796 -79.9337 -79.9853 -80.3784 -79.9613 -79.9632 -80.2719 -80.2313 -80.2137 -80.2184 -80.2158 -80.2161 -80.2154 -80.2043 -80.2052 -80.1707 -80.0595 -80.2191 -80.2190 -80.2165 -80.1431 -79.9315 -80.0283 -79.9888 -80.2431 -79.9663 -79.6481 -80.1397 Vaisala Inc. Tucson Operations 2705 E. Medina Road Tucson, AZ 85706, USA thunderstorm.vaisala.com Tel. +1 520 806 7300 Fax +1 520 741 2848 thunderstorm.sales@vaisala.com Appendix AA - Page 11 of 31 Aug 11 2006 6:46:06 PM GMT Page 11 of 18 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports Peak Distance From Date Time Current (kA) Center (mi/km) Latitude Longitude Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 10:46:01 AM 10:46:01 AM 10:46:01 AM 10:46:02 AM 10:46:02 AM 10:50:58 AM 10:51:58 AM 10:51:58 AM 10:51:58 AM 10:51:58 AM 10:52:47 AM 10:54:41 AM 10:54:41 AM 10:56:18 AM 10:56:18 AM 10:56:18 AM 10:56:18 AM 10:57:53 AM 10:57:53 AM 11:33:33 AM 7:22:07 PM 7:24:01 PM 7:26:41 PM 7:30:10 PM 7:33:54 PM 7:35:11 PM 7:35:55 PM 7:37:00 PM 7:37:00 PM 7:38:14 PM 7:38:37 PM 7:43:10 PM 7:43:10 PM 7:43:10 PM 7:45:43 PM 7:51:16 PM 7:53:38 PM 7:53:38 PM 7:55:27 PM 7:56:00 PM 7:56:00 PM 7:56:00 PM 8:00:54 PM 58.3 -23.3 19.0 -26.7 -8.0 -6.0 -6.6 -13.2 -7.3 -8.0 -5.4 -31.1 -7.2 -47.4 18.1 -12.5 32.5 -51.1 -9.7 34.0 -10.6 -20.7 -34.0 -9.9 -14.6 -6.0 -8.6 -8.4 -4.2 -20.4 -6.2 -6.4 -5.7 -8.3 -14.1 -8.3 -22.7 -15.1 18.0 -51.1 -13.4 -6.9 22.7 14.7/23.7 13.5/21.7 14.8/23.9 14.6/23.6 14.5/23.5 14.3/23.1 13.1/21.1 13.1/21.1 13.2/21.3 13.5/21.7 10.4/16.8 13.5/21.8 14.3/23.0 12.0/19.4 12.1/19.5 12.7/20.4 11.1/17.8 11.8/19.0 14.3/23.1 16.1/25.9 15.7/25.3 16.0/25.7 14.6/23.6 17.0/27.5 10.9/17.6 10.9/17.5 11.7/18.8 12.2/19.6 12.3/19.8 12.5/20.1 13.8/22.3 14.6/23.5 14.5/23.5 15.0/24.1 15.6/25.2 14.8/23.8 15.0/24.2 16.1/25.9 24.6/39.6 15.9/25.6 15.9/25.6 16.0/25.8 13.8/22.3 38.7283 38.7464 38.7261 38.7287 38.7300 38.7415 38.7591 38.7585 38.7568 38.7511 38.8012 38.7668 38.7557 38.7952 38.7945 38.7894 38.8218 38.8009 38.7766 38.7174 39.1332 39.1406 39.1298 39.1779 38.9679 39.0867 38.9710 38.9826 38.9890 39.1042 38.9805 39.1090 39.1083 39.1151 39.1224 39.1187 39.1313 39.1319 38.7418 39.1394 39.1434 39.1445 39.1238 -80.1940 -80.2287 -80.1922 -80.1995 -80.1976 -80.1264 -80.1325 -80.1346 -80.1335 -80.1430 -80.1279 -80.0861 -80.0844 -80.0790 -80.0782 -80.0693 -80.0641 -80.0760 -80.0399 -80.1189 -80.3585 -80.3531 -80.3258 -80.2922 -80.0018 -80.1257 -79.9878 -79.9816 -79.9819 -80.1014 -79.9491 -80.0371 -80.0370 -80.0359 -80.0278 -80.0490 -80.0657 -80.0306 -80.5816 -80.0525 -80.0613 -80.0597 -80.0953 Vaisala Inc. Tucson Operations 2705 E. Medina Road Tucson, AZ 85706, USA thunderstorm.vaisala.com Tel. +1 520 806 7300 Fax +1 520 741 2848 thunderstorm.sales@vaisala.com Appendix AA - Page 12 of 31 Aug 11 2006 6:46:06 PM GMT Page 12 of 18 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports Peak Distance From Date Time Current (kA) Center (mi/km) Latitude Longitude Jan 2, 2006 8:30:44 PM 18.1 5.4/8.7 38.9289 -80.3014 Vaisala Inc. Tucson Operations 2705 E. Medina Road Tucson, AZ 85706, USA thunderstorm.vaisala.com Tel. +1 520 806 7300 Fax +1 520 741 2848 thunderstorm.sales@vaisala.com Appendix AA - Page 13 of 31 Aug 11 2006 6:46:06 PM GMT Page 13 of 18 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports STRIKEnet Report 168384 Report Title: 60-06MR-308 Total Lightning Strokes Detected: 162 Lightning Strokes Detected within 15 mi/25 km radius: 128 Lightning Strokes Detected beyond 15 mi/25 km whose confidence ellipse overlaps the radius: 34 Search Radius: 15 mi/25 km Time Span: Jan 1 2006 10:00:00 PM US/Eastern to Jan 2 2006 10:00:00 PM US/Eastern Lightning Stroke Table (Note: All events shown. Events ordered by distance.) Peak Distance From Date Time Current (kA) Center (mi/km) Latitude Longitude Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 6:26:35 AM 6:26:35 AM 9:30:44 AM 6:38:51 AM 6:38:51 AM 8:30:44 PM 7:36:46 AM 6:38:51 AM 6:13:08 AM 5:57:48 AM 7:22:01 AM 7:22:01 AM 6:13:08 AM 6:29:42 AM 7:11:49 AM 7:03:33 AM 7:52:21 AM 6:15:13 AM 6:15:13 AM 6:15:13 AM 6:08:29 AM 6:14:27 AM 7:09:31 AM 6:08:29 AM 6:00:09 AM 10:52:47 AM 6:15:13 AM 6:57:04 AM 7:35:11 PM 7:03:33 AM 7:33:54 PM 10:56:18 AM 101.0 38.8 27.5 85.7 -12.6 18.1 -5.7 -86.0 -16.0 25.1 23.7 -19.4 -10.8 19.3 87.8 -20.9 47.4 -133.9 20.5 -35.9 -15.3 -35.8 178.8 -88.7 12.5 -5.4 -5.4 -9.3 -6.0 77.4 -14.6 32.5 1.9/3.1 3.4/5.5 3.6/5.7 4.4/7.1 4.9/7.8 5.4/8.7 5.8/9.3 6.2/9.9 6.3/10.2 6.5/10.6 6.6/10.7 7.0/11.2 7.2/11.6 7.5/12.1 7.9/12.8 8.3/13.3 8.3/13.4 9.4/15.2 9.4/15.2 9.4/15.2 9.5/15.3 9.7/15.7 9.8/15.8 9.9/16.0 10.0/16.2 10.4/16.8 10.6/17.2 10.7/17.3 10.9/17.5 10.9/17.5 10.9/17.6 11.1/17.8 38.9260 38.8968 38.9003 38.9805 38.9748 38.9289 39.0048 38.9954 38.9792 38.8487 38.8570 38.8500 38.9956 38.8782 38.8344 38.8292 38.8949 38.8775 38.8770 38.8751 38.9408 38.8361 38.8130 38.9486 38.8131 38.8012 38.8620 38.8845 39.0867 38.8354 38.9679 38.8218 -80.2331 -80.2313 -80.2431 -80.1380 -80.1227 -80.3014 -80.2719 -80.1113 -80.3097 -80.2310 -80.1420 -80.1457 -80.3168 -80.0886 -80.2577 -80.2578 -80.0595 -80.3573 -80.3572 -80.3565 -80.3788 -80.0813 -80.2805 -80.3875 -80.2920 -80.1279 -80.3729 -80.0159 -80.1257 -80.3530 -80.0018 -80.0641 Vaisala Inc. Tucson Operations 2705 E. Medina Road Tucson, AZ 85706, USA thunderstorm.vaisala.com Tel. +1 520 806 7300 Fax +1 520 741 2848 thunderstorm.sales@vaisala.com Appendix AA - Page 14 of 31 Aug 11 2006 6:46:06 PM GMT Page 14 of 18 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports Peak Distance From Date Time Current (kA) Center (mi/km) Latitude Longitude Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 6:15:14 AM 7:07:03 AM 7:03:33 AM 7:54:44 AM 7:35:55 PM 7:54:44 AM 7:54:44 AM 6:57:04 AM 10:57:53 AM 6:12:16 AM 6:17:13 AM 10:56:18 AM 6:17:13 AM 6:17:13 AM 6:37:59 AM 10:56:18 AM 6:17:13 AM 7:37:00 PM 7:37:00 PM 6:10:32 AM 7:38:14 PM 10:56:18 AM 7:42:35 AM 6:57:04 AM 6:22:15 AM 5:51:33 AM 6:17:14 AM 6:22:15 AM 6:12:16 AM 5:36:14 AM 6:12:16 AM 6:22:15 AM 10:51:58 AM 10:51:58 AM 10:43:53 AM 10:51:58 AM 7:42:35 AM 6:19:55 AM 6:04:12 AM 5:36:14 AM 7:25:06 AM 6:18:11 AM 10:51:58 AM -8.1 -198.4 -35.7 -9.0 -8.6 -23.3 -17.8 -20.7 -51.1 14.0 -10.9 -47.4 -9.3 -7.5 -27.8 18.1 -14.2 -8.4 -4.2 -15.6 -20.4 -12.5 -9.1 -12.8 -9.4 12.6 -9.1 8.3 -8.3 -12.1 -27.8 -19.5 -6.6 -13.2 -18.1 -7.3 -5.6 -7.9 23.9 -8.3 -11.3 -4.8 -8.0 11.2/18.1 11.3/18.2 11.3/18.2 11.6/18.7 11.7/18.8 11.7/18.9 11.7/18.9 11.7/18.9 11.8/19.0 11.8/19.1 11.9/19.2 12.0/19.4 12.0/19.4 12.1/19.5 12.1/19.5 12.1/19.5 12.1/19.5 12.2/19.6 12.3/19.8 12.3/19.8 12.5/20.1 12.7/20.4 12.7/20.5 12.8/20.6 12.9/20.8 12.9/20.8 12.9/20.8 12.9/20.9 13.0/20.9 13.0/21.0 13.0/21.0 13.1/21.1 13.1/21.1 13.1/21.1 13.1/21.1 13.2/21.3 13.3/21.4 13.4/21.6 13.4/21.6 13.4/21.6 13.4/21.6 13.4/21.7 13.5/21.7 38.8588 38.7912 38.8191 38.7732 38.9710 38.7714 38.7714 38.8925 38.8009 38.8615 38.8195 38.7952 38.8183 38.8170 38.9342 38.7945 38.8164 38.9826 38.9890 38.7698 39.1042 38.7894 38.7567 38.8824 38.7832 38.9451 38.7586 38.7808 38.8431 38.7684 38.8467 38.7797 38.7591 38.7585 38.7571 38.7568 38.7489 38.7612 38.9601 38.7619 38.7715 38.7555 38.7511 -80.3835 -80.2878 -80.3433 -80.2165 -79.9878 -80.2190 -80.2191 -79.9923 -80.0760 -80.3974 -80.0443 -80.0790 -80.0427 -80.0436 -79.9771 -80.0782 -80.0430 -79.9816 -79.9819 -80.1395 -80.1014 -80.0693 -80.2043 -79.9765 -80.0733 -80.4429 -80.1463 -80.0763 -80.4086 -80.3012 -80.4129 -80.0738 -80.1325 -80.1346 -80.1397 -80.1335 -80.2137 -80.1087 -80.4506 -80.2993 -80.0796 -80.1257 -80.1430 Vaisala Inc. Tucson Operations 2705 E. Medina Road Tucson, AZ 85706, USA thunderstorm.vaisala.com Tel. +1 520 806 7300 Fax +1 520 741 2848 thunderstorm.sales@vaisala.com Appendix AA - Page 15 of 31 Aug 11 2006 6:46:06 PM GMT Page 15 of 18 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports Peak Distance From Date Time Current (kA) Center (mi/km) Latitude Longitude Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 10:46:01 AM 10:54:41 AM 6:15:23 AM 6:15:22 AM 6:13:08 AM 9:16:58 AM 7:42:35 AM 6:21:23 AM 6:55:33 AM 6:13:08 AM 6:10:32 AM 8:00:54 PM 7:38:37 PM 6:18:11 AM 9:19:08 AM 6:09:23 AM 6:18:11 AM 6:04:12 AM 7:34:50 AM 6:14:27 AM 10:54:41 AM 5:36:14 AM 6:51:41 AM 10:57:53 AM 6:55:33 AM 10:50:58 AM 7:43:10 PM 10:46:02 AM 6:05:30 AM 7:43:10 PM 7:26:41 PM 10:46:02 AM 10:46:01 AM 7:51:16 PM 6:49:31 AM 10:46:01 AM 7:25:35 AM 6:51:41 AM 6:53:39 AM 6:49:31 AM 7:42:35 AM 7:36:15 AM 8:03:36 AM -23.3 -31.1 -11.1 -34.8 -19.1 -12.5 10.6 8.8 -8.0 -17.6 -8.6 22.7 -6.2 -17.9 -30.1 -15.3 -14.9 -33.9 -8.6 6.3 -7.2 -61.8 -5.9 -9.7 -57.6 -6.0 -5.7 -8.0 -6.0 -6.4 -34.0 -26.7 58.3 -8.3 -23.4 19.0 -9.9 -8.0 -12.7 -13.4 -8.7 -16.7 69.4 13.5/21.7 13.5/21.8 13.5/21.8 13.6/21.9 13.7/22.0 13.7/22.0 13.7/22.1 13.7/22.1 13.7/22.1 13.7/22.1 13.8/22.3 13.8/22.3 13.8/22.3 13.9/22.5 13.9/22.5 14.0/22.5 14.0/22.6 14.1/22.7 14.1/22.8 14.2/23.0 14.3/23.0 14.3/23.0 14.3/23.1 14.3/23.1 14.3/23.1 14.3/23.1 14.5/23.5 14.5/23.5 14.6/23.5 14.6/23.5 14.6/23.6 14.6/23.6 14.7/23.7 14.8/23.8 14.8/23.9 14.8/23.9 14.8/23.9 14.8/24.0 14.9/24.0 14.9/24.0 14.9/24.0 14.9/24.0 14.9/24.1 38.7464 38.7668 38.7469 38.7465 38.7436 38.7963 38.7425 38.7715 38.8093 38.7427 38.7413 39.1238 38.9805 38.7477 38.8256 38.7388 38.7492 38.9427 38.8245 38.7359 38.7557 38.7462 38.8741 38.7766 38.8105 38.7415 39.1083 38.7300 38.7303 39.1090 39.1298 38.7287 38.7283 39.1187 38.8650 38.7261 38.8894 38.8539 38.8956 38.8643 38.7257 38.8337 38.8919 -80.2287 -80.0861 -80.1628 -80.1619 -80.1817 -80.0283 -80.2052 -80.0681 -80.3942 -80.1775 -80.1827 -80.0953 -79.9491 -80.1268 -79.9888 -80.2225 -80.1164 -80.4652 -79.9853 -80.1713 -80.0844 -80.2931 -79.9501 -80.0399 -80.4102 -80.1264 -80.0370 -80.1976 -80.1803 -80.0371 -80.3258 -80.1995 -80.1940 -80.0490 -79.9439 -80.1922 -79.9337 -79.9492 -79.9316 -79.9435 -80.2184 -79.9613 -79.9315 Vaisala Inc. Tucson Operations 2705 E. Medina Road Tucson, AZ 85706, USA thunderstorm.vaisala.com Tel. +1 520 806 7300 Fax +1 520 741 2848 thunderstorm.sales@vaisala.com Appendix AA - Page 16 of 31 Aug 11 2006 6:46:06 PM GMT Page 16 of 18 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports Peak Distance From Date Time Current (kA) Center (mi/km) Latitude Longitude Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 Jan 2, 2006 7:43:10 PM 7:36:15 AM 7:53:38 PM 7:56:59 AM 6:06:16 AM 7:41:00 AM 5:55:25 AM 6:10:32 AM 5:33:58 AM 6:10:32 AM 6:04:01 AM 7:42:35 AM 7:45:43 PM 6:34:55 AM 7:42:35 AM 7:22:07 PM 5:35:41 AM 5:35:41 AM 7:42:35 AM 6:10:32 AM 5:43:55 AM 6:55:33 AM 9:32:24 AM 5:43:55 AM 6:07:26 AM 7:56:00 PM 7:56:00 PM 6:04:01 AM 7:24:01 PM 7:56:00 PM 6:04:01 AM 11:33:33 AM 7:53:38 PM 6:09:23 AM 6:09:23 AM 7:30:10 PM 6:06:16 AM 7:50:24 AM 7:35:26 AM 7:02:39 AM 5:57:41 AM 4:14:03 AM 7:55:27 PM -8.3 -14.9 -22.7 35.3 -9.3 -8.5 -5.8 -8.9 26.5 -5.9 -15.8 -9.8 -14.1 -116.3 -12.0 -10.6 -16.7 -29.1 -21.5 -12.7 -4.0 81.5 -10.5 -5.7 -48.8 -51.1 -13.4 -60.5 -20.7 -6.9 -11.9 34.0 -15.1 -23.7 -7.1 -9.9 -7.5 -9.1 -4.9 -10.5 -7.8 -5.3 18.0 15.0/24.1 15.0/24.2 15.0/24.2 15.1/24.4 15.2/24.6 15.3/24.7 15.3/24.7 15.4/24.8 15.4/24.8 15.4/24.9 15.6/25.2 15.6/25.2 15.6/25.2 15.6/25.2 15.6/25.2 15.7/25.3 15.7/25.3 15.7/25.3 15.7/25.3 15.7/25.4 15.7/25.4 15.8/25.4 15.8/25.4 15.8/25.6 15.9/25.6 15.9/25.6 15.9/25.6 15.9/25.6 16.0/25.7 16.0/25.8 16.0/25.8 16.1/25.9 16.1/25.9 16.1/25.9 16.3/26.4 17.0/27.5 17.1/27.5 17.5/28.2 18.0/29.0 18.7/30.2 20.1/32.4 22.1/35.7 24.6/39.6 39.1151 38.8287 39.1313 38.7267 38.7200 38.7197 38.7407 38.7190 38.7260 38.7177 38.7150 38.7148 39.1224 38.8325 38.7145 39.1332 38.7181 38.7178 38.7133 38.7151 38.7347 38.7927 38.8046 38.7334 38.7122 39.1394 39.1434 38.7107 39.1406 39.1445 38.7090 38.7174 39.1319 38.7083 38.7045 39.1779 38.6943 38.6882 38.7189 38.6788 38.7366 38.6437 38.7418 -80.0359 -79.9632 -80.0657 -80.1431 -80.2049 -80.2313 -80.3269 -80.1787 -80.2798 -80.1791 -80.2163 -80.2158 -80.0278 -79.9468 -80.2161 -80.3585 -80.1427 -80.1439 -80.2154 -80.2443 -80.0765 -80.4256 -79.9663 -80.0757 -80.1721 -80.0525 -80.0613 -80.2097 -80.3531 -80.0597 -80.2116 -80.1189 -80.0306 -80.1837 -80.1816 -80.2922 -80.1740 -80.1707 -80.3784 -80.1120 -79.9362 -80.3571 -80.5816 Vaisala Inc. Tucson Operations 2705 E. Medina Road Tucson, AZ 85706, USA thunderstorm.vaisala.com Tel. +1 520 806 7300 Fax +1 520 741 2848 thunderstorm.sales@vaisala.com Appendix AA - Page 17 of 31 Aug 11 2006 6:46:06 PM GMT Page 17 of 18 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports Peak Distance From Date Time Current (kA) Center (mi/km) Latitude Longitude Jan 2, 2006 9:45:50 AM -11.4 30.4/49.0 38.8504 -79.6481 Vaisala Inc. Tucson Operations 2705 E. Medina Road Tucson, AZ 85706, USA thunderstorm.vaisala.com Tel. +1 520 806 7300 Fax +1 520 741 2848 thunderstorm.sales@vaisala.com Appendix AA - Page 18 of 31 Aug 11 2006 6:46:06 PM GMT Page 18 of 18 AWS Convergence Technologies, Inc.AWS Convergence Technologies, CONFIDENTIAL Appendix AA - Vaisala Group and Inc. Reports SAGO MINE EXPLOSION January 2, 2006 Investigative Review, Research & Findings Contacts: Jim Anderson – Director Professional Services janderson@aws.com c: 202-302-7008 o: 301-250-4016 Shawn Cook – Manager Professional Services scook@aws.com c: 301-943-8666 o: 301-250-4040 - Confidential Appendix AA - Page 19 of 31 1 AWS Convergence Technologies, Inc.AWS Convergence Technologies, CONFIDENTIAL Appendix AA - Vaisala Group and Inc. Reports Overview On January 2, 2006 at approximately 6:30am, an explosion at the Sago coal mine in Tallmansville, West Virginia filled a mine shaft with poisonous gas, killing 12 miners and leaving another in critical condition. Although, the cause of the explosion has not yet been determined, it has been widely reported that lightning near the mine could be one of the contributing causes of the incident. Lightning in the area could have detonated elevated methane levels in the mine caused by changes in the barometric pressure that are more common in the winter months or other factors. The United States Precision Lightning Network (USPLN) detected a single, powerful lightning stroke at or near the mouth of the Sago mine at 6:26:36 (6:26am and 36 seconds). Through additional research initiated by WeatherBug it was discovered that Dr. Martin Chapman, PhD, a research assistant professor from Virginia Tech, analyzed the seismic data and found that two independent seismic sensors read a minor seismic event, possibly from the explosion, two seconds after that stroke at 6:26:38 (6:26am and 38 seconds). The lightning stroke held a particularly strong positive charge of 35 kAmps, compared to a typical stroke of 20 kAmps. Overall, the USPLN detected 100 lightning strokes in the region within a two hour time period around the explosion (6:30am plus or minus one hour). The USPLN network has a verified accuracy of 250 meters on average. The documents and findings in this report represent our data and analysis of the Sago Mine Explosion. It is our hope that this information will help your investigation into this matter and that WeatherBug can serve as a resource for determining the cause of this accident and for any preventive measures that may prevent future incidents of this nature to protect lives and property. Appendix AA - Page 20 of 31 2 AWS Convergence Technologies, Inc.AWS Convergence Technologies, CONFIDENTIAL Appendix AA - Vaisala Group and Inc. Reports United States Precision Lightning Network (USPLN) About the Lightning Detected and the Lightning Network • The USPLN, which is owned and operated by TOA Systems and Weather Decision Technologies, consists of 100 sensors deployed throughout the U.S. o These sensors are antennas that detect the radio wave pulse generated by lightning strokes o On average 9 sensors detect an individual lightning stroke, and only 3 are needed to accurately determine the location of a stroke – providing redundancy and excess capacity • The USPLN uses a Time of Arrival technology similar to GPS (used for OnStar or other navigation systems) and advanced signal processing to determine the time, location, strength and charge of the lightning strokes. o The USPLN utilizes a fully redundant and fault tolerant IT infrastructure o USPLN uses newer and more advanced technology than that used by the competing lightning detection network o The USPLN is capable of differentiating between cloud-to-cloud lightning and cloud-to-ground lightning o Individual lightning “Flashes” often contain multiple branches called “Strokes”. The USPLN can detect these strokes in real-time • The USPLN has a verified accuracy of 250 meters on average (RMS), the competing network is reported to have accuracy of 500 meters. • The USPLN detected 100 strokes within 1 hour before and after the explosion within a 35 mile radius of the mine. o The flash/stroke that struck closest to the mine is estimated to have hit 450 meters from the mine entrance. It carried a charge of +35 kAmps. Positive strokes are often more destructive than negative strokes. This was a very powerful stroke. The average stroke is about 20 kAmps. It takes about 100 Amps to run all the appliances in an average home, so this would be over 200 times more powerful than that and all the energy is delivered in a millisecond. Appendix AA - Page 21 of 31 3 AWS Convergence Technologies, Inc.AWS Convergence Technologies, CONFIDENTIAL Appendix AA - Vaisala Group and Inc. Reports o Individual storms have different ratios of positive and negative strokes, but typically only 10% of strokes are positive. The stroke closest to the mine was an unusual positive stroke. The image below shows the location of the lightning stroke in relation to the mine as reported by the USPLN. Appendix AA - Page 22 of 31 4 AWS Convergence Technologies, Inc.AWS Convergence Technologies, CONFIDENTIAL Appendix AA - Vaisala Group and Inc. Reports Within a 10 mile radius in the 2 hour period around the explosion, the USPLN detected 59 lightning strokes. It should be noted that only the USPLN detects strokes in real-time. The NLDN Network detects Flashes and can count the number of strokes but not locate them in real-time; they provide stroke data to their clients only after further signal processing and usually delayed by a day. There is some uncertainty on our part about where the exact entrance of the mine is, so in the Table below distances from the mine entrance are calculated to three locations (A, B, C) where the latitude of each are: A (38.941N, 80.202W), B ( 38.906N, 80.219W), and C (38.851N, 80.159W). The Table includes all lightning STROKES detected by the USPLN within a 10 mile radius of either A, B or C. USPLN Strokes Detected Within 10 miles of locations A, B, or C for 2 hour Period Centered on Time of Explosion Distance (miles) from Time Date (UTC) ms Latitude Longitude kAmps A B C 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 10:35:41 10:35:41 10:36:14 10:36:14 10:36:14 10:57:48 10:57:48 11:00:09 11:04:01 11:04:01 11:04:01 11:06:16 11:07:26 11:07:26 11:07:26 11:08:29 11:08:29 11:09:22 11:09:23 11:09:23 11:09:23 11:09:23 308 333 404 504 581 353 375 219 373 415 455 272 446 465 493 431 445 982 3 67 72 247 38.7256813 38.7263298 38.7643547 38.7682266 38.7627296 38.8658638 38.8514175 38.8187141 38.7195168 38.7145233 38.7198029 38.8094749 38.7319336 38.7228661 38.7229614 38.9610901 38.9581146 38.7323303 38.7115021 38.7313423 38.7474098 38.7327232 Appendix AA - Page 23 of 31 -80.1397018 -80.137291 -80.2599258 -80.30616 -80.2898941 -80.2489243 -80.2336044 -80.2847443 -80.199173 -80.2081985 -80.2114258 -80.2173615 -80.1406555 -80.1519928 -80.1611481 -80.3887711 -80.3610382 -80.1612625 -80.1793137 -80.2371216 -80.215271 -80.1151581 -29.2 -17 -49.8 33.4 -7.8 0 24.6 13.7 -55 -10.3 -14.9 8.2 -42.1 -11.6 -16.6 -70.7 -13.7 -88.9 -20.5 27.6 15.6 0 15.2 15.2 12.6 13.2 13.2 5.8 6.4 9.5 15.3 15.6 15.3 9.1 14.8 15.3 15.2 10.2 8.6 14.5 15.9 14.6 13.4 15.1 13.1 13.1 10 10.6 10.6 3.2 3.8 7 12.9 13.2 12.8 6.6 12.7 13.1 13 9.9 8.4 12.4 13.6 12.1 10.9 13.2 8.7 8.7 8.1 9.8 9.3 4.9 4 7.1 9.3 9.8 9.5 4.2 8.3 8.9 8.8 14.5 13.1 8.2 9.7 9.3 7.8 8.5 5 AWS Convergence Technologies, Inc.AWS Convergence Technologies, CONFIDENTIAL Appendix AA - Vaisala Group and Inc. Reports USPLN Strokes Detected Within 10 miles of locations A, B, or C for 2 hour Period Centered on Time of Explosion Distance (miles) from Time Date (UTC) ms Latitude Longitude kAmps A B C 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 1/2/2006 11:10:32 11:10:32 11:13:08 11:13:08 11:14:27 11:14:27 11:15:13 11:15:13 11:15:14 11:15:22 11:15:23 11:17:13 11:17:13 11:17:13 11:17:13 11:17:14 11:18:11 11:18:11 11:19:55 11:22:15 11:22:15 11:22:15 11:26:35 11:29:42 11:29:42 11:38:51 11:57:04 12:03:33 12:03:33 12:03:33 12:09:31 12:11:49 12:13:49 12:17:56 12:17:56 12:22:01 12:22:01 102 428 30 243 704 763 768 783 106 844 106 30 81 92 119 728 419 443 468 388 425 580 522 454 938 846 758 399 441 445 753 823 886 482 498 325 364 38.708683 38.7793159 38.7285919 38.9870758 38.7603951 38.7609596 38.887516 38.8016663 38.866375 38.757637 38.7530136 38.8184853 38.8194199 38.8250198 38.8213539 38.7708473 38.756115 38.7548027 38.7682266 38.7882881 38.7905159 38.7889099 38.9071693 38.8838577 39.0450668 38.9996719 38.8792191 38.8381348 38.8451958 38.8097878 38.8004189 38.8285065 38.8299713 38.8618698 38.8562851 38.8690491 38.8736839 Appendix AA - Page 24 of 31 -80.1775665 -80.1447983 -80.1767654 -80.3248367 -80.1751938 -80.1472244 -80.3489304 -80.0149536 -80.3815384 -80.1463165 -80.1476288 -80.0413284 -80.0407715 -80.0415115 -80.0394287 -80.1356812 -80.1195526 -80.1096191 -80.1063004 -80.0673065 -80.0675964 -80.0668335 -80.220871 -80.0839996 -80.0777817 -80.042984 -79.9833374 -80.2485428 -80.2446899 -80.3446274 -80.0074463 -80.2458344 -80.250267 -80.1765823 -80.1635132 -80.1388321 -80.1124039 12.6 -14.3 28 0 0 7.8 0 63.7 -6.9 -32.1 -10.2 -12.3 -8.8 -6.2 -10.1 -8.1 -16.6 -12.5 -7 -8.7 -16.7 9.2 35 20 -15.8 93 -11.6 0 12.8 6.3 -73.4 0 -22.3 0 -9.8 -15.3 22.7 16.1 11.6 14.7 7.3 12.5 12.8 8.7 13.9 11 13 13.3 12.1 12.1 11.8 12 12.3 13.5 13.8 13 12.8 12.7 12.8 2.5 7.5 9.8 9.5 12.5 7.5 7 11.9 14.3 8.1 8.1 5.6 6.2 6 6.7 13.8 9.6 12.4 8 10.3 10.7 7.1 13.2 9.2 11 11.2 11.3 11.3 11.1 11.3 10.3 11.6 12 11.3 11.5 11.4 11.5 0.1 7.4 12.3 11.5 12.9 4.9 4.4 9.5 13.5 5.5 5.5 3.8 4.6 5 6.2 9.9 5 8.5 12.9 6.3 6.3 10.5 8.5 12 6.5 6.8 6.8 6.8 6.6 6.8 5.7 6.9 7.2 6.4 6.6 6.5 6.6 5.1 4.7 14.1 12 9.7 4.9 4.6 10.4 8.9 4.9 5.1 1.2 0.4 1.7 3 6 AWS Convergence Technologies, Inc.AWS Convergence Technologies, CONFIDENTIAL Appendix AA - Vaisala Group and Inc. Reports Seismic Data Coincident with the reported lightning strokes, WeatherBug brought in expertise from the Virginia Tech Department of Geosciences to examine whether there was seismic activity in the mine region at 6:26:38 that may have been caused by the explosion. The seismic readings have a timing error of plus or minus 3 seconds. The evidence suggests that the lightning stroke could have caused the explosion due to the correlation between the timing and location of the lightning stroke and seismic activity. USPLN versus Vaisala’s NLDN Network It is important to note that two separate lightning networks reported lightning data related to the Sago Mine Explosion – The USPLN and Vaisala’s NLDN Network. These networks are different, based on different technologies, and do not have the same accuracy. Since lightning is a potential cause of the explosion, it is important to note the differences between these networks and evaluate their validity. Appendix AA - Page 25 of 31 7 AWS Convergence Technologies, Inc.AWS Convergence Technologies, CONFIDENTIAL Appendix AA - Vaisala Group and Inc. Reports It was reported in the Charleston Gazette that, Vaisala, a federal government contractor, reported lightning strikes within 1.5 miles of the mine. “Three lightning strikes hit within five miles of the Sago Mine within a halfhour of Monday morning’s deadly explosion, according to a federal government contractor that monitors thunderstorms. Two of the strikes, including one that was four to 10 times stronger than average, hit within 1 1/2 miles of the center of the Upshur County mine, according to the contractor.” http://www.wvgazette.com/section/News/2006010439 • The USPLN can detect both the Flash from the main bolt of lightning, and the individual Strokes, all the little forks in the lightning bolt, some of which can strike the ground many miles from the main Flash. • Visalia’s published accuracy of 500 meters on average for the National Lightning Data Network (NLDN) versus 250 meters for the USPLN. Appendix AA - Page 26 of 31 8 AWS Convergence Technologies, Inc.AWS Convergence Technologies, CONFIDENTIAL Appendix AA - Vaisala Group and Inc. Reports About WeatherBug • WeatherBug’s mission is to protect lives and property by providing the most precise weather available • WeatherBug owns and manages the largest and most advanced weather network in the U.S. --- totaling 8,000 • WeatherBug technology can provide advance warning of all types of weather threats, including lightning • Only WeatherBug offers live, neighborhood level weather – vs. hourly weather reports from area airports • WeatherBug partners with local TV broadcasters, the National Weather Service, government agencies and private organizations APENDIX Dow Jones Article http://www.aws.com/aws_2005/releases/2006/release_01062006.asp WeatherBug Press Release http://www.aws.com/aws_2005/releases/2006/release_01062006b.asp Pittsburgh Tribune Article http://pittsburghlive.com/x/tribune-review/trib/regional/s_412305.html Appendix AA - Page 27 of 31 9 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports The Occurrence of Lightning near Lat: 30.2002997 Lon: -85.6244055 West Virginia For the Period 5:00 AM EST January 2, 2006 to 5:00 AM EST January 3, 2006 Prepared by Matt Gaffner Meteorologist Weather Decision Technologies, Inc. 1818 W. Lindsey St, Bldg. D, Suite 208 Norman, OK 73069 405-579-7675 Ext. 239 mgaffner@wdtinc.com For Dean Skorski P.O. Box 18233 Pittsburgh, PA 15236 412-386-6949 Skorski.dean@dol.gov Date Prepared: January 11, 2006 Appendix AA - Page 28 of 31 Page 1 of 4 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports INTRODUCTION This report describes the identified cloud-to-ground and cloud-to-cloud lightning activity within a 10 mile radius centered on the location of interest in West Virginia (Lat: 30.2002997, Lon: -85.6244055). Expert meteorologists at Weather Decision Technologies, Inc. (WDT) have carefully examined the archived record of cloud-toground and cloud-to-cloud lightning strikes within this area of interest for the time period 5:00 AM EST January 2, 2006 to 5:00 AM EST January 3, 2006. This report describes the results of our investigation. LIGHTNING ANALYSIS/CONCLUSION The purpose of this investigation is to determine the closest lightning strike to the location of interest. The source of lightning data for this investigation is the United States Precision Lightning Network (USPLN) which is maintained and operated by WDT and TOA systems Inc. The USPLN lightning data archive consists of identified cloud-toground lightning strikes since May 28, 2004, and the location accuracy of cloud-toground lightning data detected by USPLN is 250 meters (.076 miles). An examination of the lightning strikes during the 24 hour period of interest reveals that seventeen cloud-to-ground lightning strikes and six cloud-to-cloud lightning strikes occurred within the 10 mile radius centered on the address of interest (Figure 1). In addition, the closest cloud-to-ground lightning strike occurred 2.5 miles south-southwest of the address of interest at 6:26 EST on January 2, 2006. All other identified lightning strikes are shown in Figure 1 as well as in Table 1. Appendix AA - Page 29 of 31 Page 2 of 4 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports Figure 1. Map centered on the location of interest. The identified cloud-to-ground lightning strikes are depicted with red “bolts”. The identified cloud-to-cloud lightning strikes are depicted with a blue “X”. The light blue star depicts the location of interest. The areal extent is 2 miles by 20 miles (400 mi2). (Lightning data source: USPLN) Appendix AA - Page 30 of 31 Page 3 of 4 Appendix AA - Vaisala Group and AWS Convergence Technologies, Inc. Reports Appendix 1. Cloud-to-ground and cloud-to-cloud lightning strikes for the period of 5:00 AM EST January 2, 2006 to 5:00 AM EST January 3, 2006 and within 10 miles of the address of interest. Time is in 24-hour Eastern Standard Time (EST) format. Bearing is relative to due north from the location of interest. For example, 90 degrees = east, 180 degrees = south, 270 degrees = west, 0 degrees = north. (Lightning data source: USPLN) Date-Time (EST) 1/2/2006 5:57 1/2/2006 5:57 1/2/2006 6:00 1/2/2006 6:06 1/2/2006 6:08 1/2/2006 6:13 1/2/2006 6:15 1/2/2006 6:15 1/2/2006 6:26 1/2/2006 6:29 1/2/2006 6:29 1/2/2006 6:38 1/2/2006 6:38 1/2/2006 7:03 1/2/2006 7:03 1/2/2006 7:11 1/2/2006 7:13 1/2/2006 7:13 1/2/2006 7:17 1/2/2006 7:17 1/2/2006 7:22 1/2/2006 7:22 1/2/2006 7:52 Amplitude 24600 0 13700 8200 -13700 0 0 -125100 35000 20000 -15800 93000 -35800 0 12800 0 -22300 -22700 -9800 0 -15300 22700 43500 Appendix AA - Page 31 of 31 Latitude 38.851 38.866 38.819 38.809 38.958 38.987 38.888 38.889 38.907 38.884 39.045 39 39.007 38.838 38.845 38.829 38.83 38.838 38.856 38.862 38.869 38.874 38.898 Longitude -80.234 -80.249 -80.285 -80.217 -80.361 -80.325 -80.349 -80.368 -80.221 -80.084 -80.078 -80.043 -80.288 -80.249 -80.245 -80.246 -80.25 -80.222 -80.164 -80.177 -80.139 -80.112 -80.057 Page 4 of 4 Bearging (°) 195 206 208 185 278 296 245 248 203 122 43 65 315 199 199 197 198 188 160 165 145 134 111 Distance (mi.) 6.4 5.7 9.5 9.1 8.6 7.3 8.7 9.6 2.5 7.5 9.9 9.5 6.5 7.5 7 8.1 8.1 7.2 6.2 5.6 6 6.7 8.4 .- II I'll - -. a- . yeti h; :r . . .41, :?fu II-. mini:- fl 1 iI Explanation Location of lightning strike reported by Vaisala?s National Lightning Detection Network (NLDN). Number to left of symbol represents the peak current in kilo-amps; number to right of symbol represents the time that the peak current was recorded. Location of lightning strike reported by Weather Decision Technologies, lnc.?s U.S. Precision Lightning Network (USPLN). Numberto left of symbol represents the peak current in kilo-amps; number to right of symbol represents the time that the peak current was recorded. Sago Mine workings. Locations of power line poles, with trace of power line (dashed where main line is projected). Locations of telephone poles and junction boxes, with trace of phone line. Location of large poplar tree shattered by lightning. Appendix BB Sago Mine MSHA ID 46-08791 Wolf Run Mining Company Sago Mine in relation to recorded locations of lightning strikes, a lightning-damaged poplar tree, and the mines phone and power lines. Appendix CC - Results from Analysis of Seismic Data Results from Analysis of Seismic Data for the January 2, 2006 event near Sago, WV Martin Chapman Department of Geosciences VPI&SU Blacksburg, VA ph: 540-231-5036 email: mcc@vt.edu Introduction The author examined regional seismic network recordings for the time interval around 6:30 AM, EST January 2, 2006 to determine if the event at the Sago mine was seismically recorded. A small amplitude signal was identified on records at broadband station MCWV, near Mont Chateau, WV, the nearest seismic station to the mine. This station is part of the U.S. Geological Survey Advanced National Seismic System (ANSS) which is designed to record world-wide seismic activity as well as to monitor shocks in all regions of the U.S. The signal was also recorded at larger distances by three stations to the south: FWV, ELN and BLA. These more distant stations use short period sensors and are operated by Virginia Tech as part of the ANSS. The following is a summary of the results pertaining to the location and time of the eve nt that generated the seismic signals. Data Figures 1 through 4 show the data recorded at stations MCWV, FWV, ELN and BLA respectively. The signals have been bandpass-filtered using a 3 pole Butterworth prototype with corner frequencies 1.0 and 5.0 Hz. The signal/noise ratios of these data are small, however, measurement of arrival times for P and S waves was possible. The estimated arrival times are given below in Table 1, in Eastern Standard Time. The coordinates of the recording stations are as follows: BLA: ELN: FWV: MCWV: Appendix CC - Page 1 of 8 37.2113 deg N 37.2805 deg N 37.5810 deg N 39.6582 deg N 80.4202 deg W 80.7517 deg W 80.8118 deg W 79.8457 deg W 1 Appendix CC - Results from Analysis of Seismic Data Results Figure 5 shows the epicenter estimated using the arrival time data in Table 1. The locations were determined using the velocity model in Table 2, in conjunction with the computer program Hypoellipse. Table 3 gives hypocenter and origin time estimates for 3 cases. The first case assumes that the focal depth of the source is near the ground surface, consistent with a mining-related source, but not necessarily located near the Sago mine. Latitude, longitude and origin time are treated as unknowns to be determined from the arrival time data. The origin time estimate in this case is 06:26:38.29 EST with standard error 1.65 seconds. The 68% confidence ellipse for the epicenter determined from the seismic data includes the Sago mine location (Figure 5). A 68% confidence interval for the origin time is 06:26:36.60 to 06:26:39.94 EST, assuming no systematic bias due to uncertainty associated with the velocity model in Table 2 or in phase arrival time measurement. The second case is a completely un-constrained location, in which the latitude, longitude, focal depth and origin time are treated as unknowns to be determined. The computed epicenter is very near the Sago Mine location in this case (figure 5). The estimated focal depth is shallow (2.5 km) but very poorly determined (68% confidence: 0 to 34 km). The 68% confidence interval for the origin time is 06:26:35.35 - 06:26:41.21 EST. The third case assumes that the source occurred at the Sago mine, (Latitude 38.9407°N; Longitude 80.2030°W) with zero focal depth. The only free parameter to be determined is the origin time. The 68% confidence interval for the origin time is 06:26:36.46 - 06:26:40.00 EST. Conclusions The seismic signal recorded on January 2, 2006 at approximately 06:26 EST was caused by an underground disturbance at or near the Sago mine. Assuming that the source was at the Sago mine, a 68% confidence interval for the origin time is 06:26:36.46 - 06:26:40.00 EST. Simply put, the event most likely occurred within a 4 second interval centered at 06:26:38.2 AM. This estimate assumes no systematic error in phase arrival time determination, and/or bias in the seismic wave velocity model used for analysis. It is possible that the origin time estimate is slightly late, due to the very emergent nature of the P and S wave arrivals because of low signal/noise ratios at all the recording stations. Appendix CC - Page 2 of 8 2 Station MCWV FWV ELN BLA Appendix CC - Results from Analysis of Seismic Data Table 1 P arrival* S arrival* Hour Minute Second Hour Minute Second 06 26 52.6 06 27 3.5 06 27 5.1 06 27 24.1 06 27 9.0 06 27 32.7 06 27 9.7 06 27 32.2 * All times are Eastern Standard Time. P wave velocity (km/sec) 5.63 6.05 6.53 8.18 Table 2 S wave velocity (km/sec) 3.43 3.52 3.84 4.78 Layer thickness (km) 5.7 9.0 36.0 - Table 3 Depth constrained Depth unconstrained Depth and location constrained Standard Error of Origin Time Azimuth of Error Ellipse SemiMajor Axis Major Axis Length Minor Axis Length Latitude Longitude Focal Depth Origin Time* 38.9243°N 80.1169°W 0 km (fixed) 06:26:38.29 1.65 s 286° 23 km 4.4 km 38.9465°N 80.1920°W 2.45 km 06:26:38.28 2.93 s 289° 23 km 4.0 km 38.9407°N (fixed) 80.2030°W (fixed) 0 km (fixed) 06:26:38.23 1.77 s * All times are Eastern Standard Time. Appendix CC - Page 3 of 8 3 Appendix CC - Results from Analysis of Seismic Data Figure 1. Waveforms recorded at station MCWV, 85.4 km from the assumed epicenter at 38.94065 degrees N, 80.20295 degrees W. Appendix CC - Page 4 of 8 4 Appendix CC - Results from Analysis of Seismic Data Figure 2. Waveforms recorded at station FWV, 160.1 km from the assumed epicenter at 38.94065 degrees N, 80.20295 degrees W. Appendix CC - Page 5 of 8 5 Appendix CC - Results from Analysis of Seismic Data Figure 3. Waveforms recorded at station ELN, 190.5 km from the assumed epicenter at 38.94065 degrees N, 80.20295 degrees W. Appendix CC - Page 6 of 8 6 Appendix CC - Results from Analysis of Seismic Data Figure 4. Waveforms recorded at station BLA, 192.9 km from the assumed epicenter at 38.94065 degrees N, 80.20295 degrees W. Appendix CC - Page 7 of 8 7 Appendix CC - Results from Analysis of Seismic Data Figure 5. Map showing as a black diamond the assumed location of the Sago mine event (38.94065 degrees N, 80.20295 degrees W). The red diamond shows the epicenter determined using the arriva l time data in Table 1 with focal depth fixed at the ground surface. The red line indicates 68% confidence ellipse for the epicenter location. The blue diamond is the epicenter estimated with the depth unconstrained. The blue line shows the corresponding 68% confidence ellipse. Seismic stations used in the location are indicated by the red triangles. Appendix CC - Page 8 of 8 8 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine SANDIA REPORT SAND2006-7976 Unlimited Release Printed April 2007 Measurement and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Matthew B. Higgins and Marvin E. Morris Contributing Editors: Michele Caldwell and Larry X. Schneider Prepared by Sandia National Laboratories Albuquerque, New Mexico 87185, and Livermore, California 94550 Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy’s National Nuclear Security Administration under Contract DE-AC04-94AL85000. Approved for public release; further dissemination unlimited. Appendix DD - Page 1 of 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Issued by Sandia National Laboratories, operated for the United States Department of Energy by Sandia Corporation. NOTICE: This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government, nor any agency thereof, nor any of their employees, nor any of their contractors, subcontractors, or their employees, make any warranty, express or implied, or assume any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represent that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government, any agency thereof, or any of their contractors or subcontractors. The views and opinions expressed herein do not necessarily state or reflect those of the United States Government, any agency thereof, or any of their contractors. Printed in the United States of America. This report has been reproduced directly from the best available copy. Available to DOE and DOE contractors from U.S. Department of Energy Office of Scientific and Technical Information P.O. Box 62 Oak Ridge, TN 37831 Telephone: Facsimile: E-Mail: Online ordering: (865) 576-8401 (865) 576-5728 reports@adonis.osti.gov http://www.osti.gov/bridge Available to the public from U.S. Department of Commerce National Technical Information Service 5285 Port Royal Rd. Springfield, VA 22161 Telephone: Facsimile: E-Mail: Online order: (800) 553-6847 (703) 605-6900 orders@ntis.fedworld.gov http://www.ntis.gov/help/ordermethods.asp?loc=7-4-0#online Appendix DD - Page 2 of 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine SAND2006-7976 Unlimited Release Printed April 2007 Measurement and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Matthew B. Higgins Electromagnetic Qualification and Engineering Department Marvin E. Morris Electromagnetic and Plasma Physics Analysis Contributing Editors Michele Caldwell Larry X. Schneider Sandia National Laboratories P.O. Box 5800 Albuquerque, New Mexico 87185-1152 Abstract This report documents measurements and analytical modeling of electromagnetic transfer functions to quantify the ability of cloud-to-ground lightning strokes (including horizontal arc-channel components) to couple electromagnetic energy into the Sago mine located near Buckhannon, WV. Two coupling mechanisms were measured: direct and indirect drive. These transfer functions are then used to predict electric fields within the mine and induced voltages on conductors that were left abandoned in the sealed area of the Sago mine. Appendix DD - Page 3 of 104 PAGE 3 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine ACKNOWLEDGEMENTS A complex project of this type could not have been undertaken without the funding, coordination, and hard work of the MSHA staff that were involved. We wish to thank MSHA staff William Helfrich, Richard Gates, Robert Phillips, Harold Newcomb, Russell Dresch, Dean Skorski, Joseph O’Donnell, and Arthur Wooten for their support of this project. Jurgen Brune and Eric Weiss of NIOSH generously provided their Lake Lynn facility for initial trials of the measurement techniques used at the Sago mine. We thank the ICG staff, Chuck Dunbar, Al Schoonover, Johnny Stemple, Larry Dean, Kermit Melvin, and Brittany Bolyard, for their generous help in arranging access and for providing the services we needed to accomplish the measurement tasks. In spite of our obvious interruptions to their operations as well as extensive demands on their time, they generously provided the services we needed in a timely manner. We are grateful for the support of Dr. E. Philip Krider and Dr. Martin Uman, who independently reviewed the lightning database information for this report. We also wish to thank the consultants, Dr. Tom Novak, Dr. E. Philip Krider, Elio Checca, and Dr. Martin Uman for freely sharing their thoughts on this project. Monte Hieb and John Scott of the State of WV Office of Miners’ Health, Safety, and Training provided additional useful information and help to accomplish the work. Finally, we would like to thank the property owners of the land above the sealed area, Mrs. Goldie Gooden, Tim and Chris Leggett, Bill Patterson, and George Roessing, for generously allowing us access to their property in order to drive ground rods, string wires, and operate our equipment despite obvious interruptions to their lives. Most importantly, we would like to thank our measurement team, Dawna R. Charley and Leonard Martinez, for their extraordinary work in the field. Without their hard work and long hours the measurements could not have been completed. Appendix DD - Page 4 of 104 PAGE 4 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine TABLE OF CONTENTS ACKNOWLEDGEMENTS ........................................................................................................................................4 LIST OF FIGURES.....................................................................................................................................................7 LIST OF TABLES.....................................................................................................................................................10 EXECUTIVE SUMMARY .......................................................................................................................................11 ABBREVIATIONS, ACRONYMS AND INITIALIZATIONS.............................................................................12 1 INTRODUCTION ............................................................................................................................................13 1.1 MOTIVATION FOR RESEARCH AND MEASUREMENTS...................................................................................14 1.2 OBJECTIVES OF MEASUREMENTS ................................................................................................................14 1.3 PREVIOUS WORK ON LIGHTNING INDUCED MINE EXPLOSIONS...................................................................14 1.4 MEASUREMENT METHOD AND ANALYSIS ...................................................................................................14 1.4.1 Direct Coupling Transfer Function Measurements and Analysis..........................................................15 1.4.2 Indirect Coupling Transfer Function Measurements and Analysis .......................................................16 1.5 SOIL AND ROCK SITE DATA ........................................................................................................................17 1.6 LIGHTNING EVENT INFORMATION...............................................................................................................17 1.7 OTHER SITE INFORMATION .........................................................................................................................18 1.8 FIDELITY ISSUES OF STUDY ........................................................................................................................21 1.8.1 Current Flow on the Surface from a Real Lightning Stroke and the Indirect-drive Test Setup .............21 1.8.2 Physical Changes to the Sago Site after the Accident............................................................................22 1.9 POTENTIAL FURTHER AREAS OF STUDY .....................................................................................................22 1.9.1 Nonlinearities ........................................................................................................................................22 1.9.2 Coupling from Vertical Pipes near Sealed Areas ..................................................................................22 1.9.3 Distributed Drives for Metallic Penetrations ........................................................................................22 1.9.4 Amplification Effects of Wiring Resonances..........................................................................................23 1.9.5 Effect of Grounded Roof Meshes ...........................................................................................................23 1.9.6 Coupling Paths Not Present in Sago Mine ............................................................................................23 1.9.7 Geologic Irregularities Affecting Coupling ...........................................................................................23 1.9.8 Lightning Current Return Path Assumptions.........................................................................................23 2 ELECTROMAGNETIC COUPLING PHENOMENOLOGY MODELS...................................................24 2.1 DIRECT COUPLING VIA METALLIC PENETRATIONS INTO MINE ...................................................................24 2.1.1 Localized Drive Transmission-line Theory............................................................................................24 2.1.2 Distributed Drive Transmission-line Theory .........................................................................................25 2.2 INDIRECT ELECTROMAGNETIC COUPLING VIA SOIL AND ROCK ..................................................................25 2.2.1 Static Coupling Model for Current Injected into Homogeneous Half-Space.........................................25 2.2.2 Infinite Line Source above Homogeneous Half-Space...........................................................................26 2.2.3 Infinite Line Source at Surface of Homogeneous Half-Space................................................................28 2.2.4 Uniform Magnetic Field at Surface above Homogeneous Half-Space ..................................................29 3 MEASUREMENT METHODS .......................................................................................................................30 3.1 DIRECT DRIVE ............................................................................................................................................30 3.1.1 The Differences and Similarities between Conductive Penetrations .....................................................30 3.1.2 Setup/Equipment Layout with Photos ....................................................................................................30 3.1.3 Results....................................................................................................................................................34 3.2 INDIRECT DRIVE .........................................................................................................................................36 3.2.1 Setup/Equipment Layout with Photos ....................................................................................................36 3.2.2 Results....................................................................................................................................................39 3.2.3 Results Compared with Diffusion Model ...............................................................................................45 4 RESULTS COUPLED WITH LIGHTNING .................................................................................................48 4.1 DIRECT DRIVE TRANSFER FUNCTIONS COUPLED WITH LIGHTNING STROKES .............................................49 Appendix DD - Page 5 of 104 PAGE 5 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 4.2 INDIRECT DRIVE FROM NLDN AND USPLN POSITIVE STROKE 1-3............................................................52 4.3 INDIRECT DRIVE FROM HYPOTHETICAL STROKE DIRECTLY OVER SEALED AREA.......................................54 4.4 INDIRECT DRIVE FROM A HYPOTHETICAL CLOUD-TO-GROUND STROKE WITH A CURRENT CHANNEL OVER SEALED AREA ..........................................................................................................................................................56 5 CONCLUSIONS ...............................................................................................................................................59 5.1 5.2 DIRECT COUPLING ......................................................................................................................................59 INDIRECT COUPLING ...................................................................................................................................60 6 RECOMMENDATIONS..................................................................................................................................62 7 REFERENCES..................................................................................................................................................63 8 APPENDIX A — ANALYTICAL AND NUMERICAL MODELS FOR VOLTAGE AND CURRENT USED TO DETERMINE ELECTROMAGNETIC COUPLING INTO THE SAGO MINE.............................65 8.1 INTRODUCTION ...........................................................................................................................................65 8.2 STATIC CURRENT DRIVE MODELS ..............................................................................................................66 8.2.1 Homogeneous Half-Space......................................................................................................................66 8.2.2 Two Layer Half-Space ...........................................................................................................................67 8.3 EDDY CURRENT, INFINITE HORIZONTAL DRIVE WIRE MODELS..................................................................68 8.3.1 Homogeneous Half-Space......................................................................................................................69 8.3.2 Two Layer Half-Space ...........................................................................................................................72 8.4 EDDY CURRENT COUPLING INTO HOMOGENEOUS HALF-SPACE FROM UNIFORM MAGNETIC FIELD AT SURFACE ..................................................................................................................................................................75 8.5 EDDY CURRENT, INFINITESIMAL AND FINITE LENGTH HORIZONTAL DRIVE WIRE MODELS .......................77 8.6 REFERENCES FOR APPENDIX A ...................................................................................................................77 9 10 APPENDIX B –CALIBRATION DOCUMENTATION OF MEASUREMENT EQUIPMENT ..............78 APPENDIX C – COMPILATION OF MEASURED DATA....................................................................82 11 APPENDIX D – LIST OF UNDERGROUND SEALED AREA COAL MINE EXPLOSIONS SUSPECTED OF LIGHTNING INITIATION.....................................................................................................102 12 APPENDIX E – MEMORANDUM FROM DR. KRIDER.....................................................................103 Appendix DD - Page 6 of 104 PAGE 6 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine LIST OF FIGURES FIGURE 1-1 APPROXIMATE LOCATION OF INITIATION OF EXPLOSION IN SEALED AREA OF SAGO MINE.......................13 FIGURE 1-2 LOCATION OF LIGHTNING STROKES AT SAGO MINE CONTEMPORANEOUS WITH SEALED AREA EXPLOSION. .........................................................................................................................................................17 FIGURE 1-3 VERTICAL PIPES IN VICINITY OF SEALED AREA OF SAGO MINE. ..............................................................19 FIGURE 1-4 AC POWER DISTRIBUTION LINES AND TELEPHONE LINES NEAR POSITIVE 101 KA STROKE. ....................20 FIGURE 1-5 ROOF MESH AND CABLE IN SEALED AREA WHERE EXPLOSION WAS INITIATED. THE RED LINE REPRESENTS A CABLE FROM A WATER PUMP LOCATED AT THE TOP OF THE FIGURE. THE GREEN LINES REPRESENT METALLIC ROOF MESH. ........................................................................................................................................21 FIGURE 2-1 EQUIVALENT CIRCUIT OF A SECTION OF TRANSMISSION LINE. .................................................................24 FIGURE 2-2 DC CURRENT DRIVE WITH HOMOGENEOUS CONDUCTING HALF-SPACE. .................................................25 FIGURE 2-3 INFINITE LENGTH, HARMONICALLY TIME VARYING HORIZONTAL CURRENT DRIVE OVER A CONDUCTIVE HALF-SPACE........................................................................................................................................................26 FIGURE 2-4 SKIN DEPTH, δ1, AS A FUNCTION OF FREQUENCY FOR RESISTIVITIES OF 10, 100, AND 1000 OHM-M. .......27 FIGURE 2-5 AMPLITUDE AND PHASE OF ELECTRIC FIELD AS A FUNCTION OF FREQUENCY AT DEPTH OF 100M WITH RESISTIVITIES OF 10, 100 AND 1000 OHM-M. ......................................................................................................28 FIGURE 2-6 HARMONICALLY TIME-VARYING MAGNETIC FIELD DRIVE OVER CONDUCTIVE HALF-SPACE. ................29 FIGURE 3-1 DIRECT DRIVE CONCEPTUAL DRAWING. ..................................................................................................31 FIGURE 3-2 DIRECT DRIVE MEASUREMENT LOCATIONS. ............................................................................................32 FIGURE 3-3 (A.) CURRENT PROBE ON TROLLEY COMMUNICATION CABLE. (B.) CURRENT PROBE AND VOLTAGE CONNECTION ON CONVEYOR BELT STRUCTURE. (C.) VOLTAGE PROBE ON POWER CABLE. (D.) CURRENT PROBE AND VOLTAGE CONNECTION ON RAIL...................................................................................................................33 FIGURE 3-4 INDIRECT DRIVE CONCEPTUAL DRAWING. ................................................................................................37 FIGURE 3-5 PARALLEL (A.) AND PERPENDICULAR (B.) SURFACE CURRENT DRIVE FOR INDIRECT DRIVE MEASUREMENTS. .................................................................................................................................................37 FIGURE 3-6 ELECTRIC FIELD MEASUREMENT LOCATIONS. ...........................................................................................38 FIGURE 3-7 SANDIA DIPOLE ANTENNA IN HORIZONTAL AND VERTICAL POLARIZATIONS INSIDE PREVIOUSLY SEALED AREA....................................................................................................................................................................38 FIGURE 3-8 COMPOSITE ELECTRIC FIELD ALONG P-DIRECTION WITH PARALLEL LINE DRIVE ON SURFACE..................41 FIGURE 3-9 COMPOSITE ELECTRIC FIELD ALONG X-DIRECTION WITH PARALLEL LINE DRIVE ON SURFACE. ................41 FIGURE 3-10 COMPOSITE ELECTRIC FIELD ALONG P-DIRECTION WITH PERPENDICULAR LINE DRIVE ON SURFACE. .....42 FIGURE 3-11 COMPOSITE ELECTRIC FIELD ALONG X-DIRECTION WITH PERPENDICULAR LINE DRIVE ON SURFACE. ....42 FIGURE 3-12 INDUCED VOLTAGE ON PUMP CABLE (~300 M OR 984 FT. LONG) DUE TO WIRE CURRENT DRIVES ON SURFACE. .............................................................................................................................................................43 FIGURE 3-13 P-DIRECTED ELECTRIC FIELD ALONG P-DIRECTION WITH PARALLEL LINE DRIVE ON SURFACE. ..............44 FIGURE 3-14 P-DIRECTED ELECTRIC FIELDS MULTIPLIED BY AN EFFECTIVE CABLE LENGTH OF 120 M (394 FT) COMPARED WITH THE INDUCED VOLTAGE ON THE PUMP CABLE...........................................................................44 FIGURE 3-15 P-DIRECTED ELECTRIC FIELDS COMPARED WITH THE DIFFUSION MODEL WITH AN EFFECTIVE RESISTIVITY OF 80 Ω-M...........................................................................................................................................................45 FIGURE 3-16 AVERAGE OF P-DIRECTED FIELDS FROM P2 TO P8 COMPARED WITH DIFFUSION MODEL. ........................46 FIGURE 3-17 INDUCED VOLTAGE ON PUMP CABLE DUE TO PARALLEL WIRE CURRENT DRIVE ON SURFACE (WITH 60 HZ AND HARMONICS REMOVED) COMPARED WITH ANALYTIC DIFFUSION MODEL OF 120 M (394 FT) LONG CABLE AND AN EFFECTIVE SOIL RESISTIVITY OF 80 Ω-M AND THE DC RESISTIVITY TERM......................................................47 FIGURE 4-1 BASIC POSITIVE AND NEGATIVE LIGHTNING WAVEFORMS USED AS INPUTS FOR ANALYSIS. ......................48 FIGURE 4-2 LOCATIONS OF RECORDED LIGHTNING STROKES WITH RESPECT TO THE SEALED AREA, WITH DISTANCES AND ANGLES. .......................................................................................................................................................52 FIGURE 4-3 VOLTAGE INDUCED ON PUMP CABLE (USING AN EFFECTIVE LENGTH OF 120 M OR 394 FT.) DUE TO THE THREE POSITIVE LIGHTNING STROKES RECORDED ON THE NLDN AND USPLN...................................................53 FIGURE 4-4 VOLTAGE INDUCED ON PUMP CABLE (LENGTH OF 61 M OR 200 FT.) DUE TO THE THREE POSITIVE LIGHTNING STROKES RECORDED ON THE NLDN AND USPLN.............................................................................53 FIGURE 4-5 INDUCED VOLTAGE PULSE ON PUMP CABLE (USING AN EFFECTIVE LENGTH OF 120 M OR 394 FT.) DUE TO A HYPOTHETICAL POSITIVE AND NEGATIVE 100 KA CLOUD-TO-GROUND LIGHTNING STROKE 100 M FROM DIRECTLY ABOVE SEALED AREA. .........................................................................................................................54 Appendix DD - Page 7 of 104 PAGE 7 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine FIGURE 4-6 INDUCED VOLTAGE PULSE ON PUMP CABLE (LENGTH OF 61 M OR 200 FT.) DUE TO A HYPOTHETICAL POSITIVE AND NEGATIVE 100 KA CLOUD-TO-GROUND LIGHTNING STROKE 100 M FROM DIRECTLY ABOVE SEALED AREA....................................................................................................................................................................55 FIGURE 4-7 INDUCED VOLTAGE PULSE ON PUMP CABLE (WITH AN EFFECTIVE LENGTH OF 120 M OR 394 FT.) FROM HYPOTHETICAL HORIZONTAL CURRENT CHANNEL FROM A CLOUD-TO-GROUND +100 KA STROKE, H IS DISTANCE OF THE CURRENT CHANNEL ABOVE THE GROUND. .............................................................................56 FIGURE 4-8 INDUCED VOLTAGE PULSE ON PUMP CABLE (LENGTH OF 61 M OR 200 FT.) FROM HYPOTHETICAL HORIZONTAL CURRENT CHANNEL FROM A CLOUD-TO-GROUND +100 KA STROKE, H IS DISTANCE OF THE CURRENT CHANNEL ABOVE THE GROUND...........................................................................................................57 FIGURE 4-9 INDUCED VOLTAGE PULSE ON PUMP CABLE (WITH AN EFFECTIVE LENGTH OF 120 M OR 394 FT.) FROM HYPOTHETICAL HORIZONTAL CURRENT CHANNEL FROM A CLOUD-TO-GROUND -100 KA STROKE, H IS DISTANCE OF THE CURRENT CHANNEL ABOVE THE GROUND. .............................................................................57 FIGURE 4-10 INDUCED VOLTAGE PULSE ON PUMP CABLE (LENGTH OF 61 M OR 200 FT.) FROM HYPOTHETICAL HORIZONTAL CURRENT CHANNEL FROM A CLOUD-TO-GROUND -100 KA STROKE, H IS DISTANCE OF THE CURRENT CHANNEL ABOVE THE GROUND...........................................................................................................58 FIGURE 8-1 DC CURRENT DRIVE WITH HOMOGENEOUS HALF-SPACE GEOMETRY. ....................................................66 FIGURE 8-2 DC CURRENT DRIVE WITH TWO LAYER HALF-SPACE GEOMETRY. ..........................................................68 FIGURE 8-3 INFINITE HORIZONTAL CURRENT DRIVE, EDDY CURRENT COUPLING GEOMETRY. ..................................69 FIGURE 8-4 SKIN DEPTH AS A FUNCTION OF FREQUENCY FOR RESISITIVITIES, τ1 = 10, 100, 1000 Ω-M. .....................70 FIGURE 8-5 AMPLITUDE OF ELECTRIC FIELD FROM A LINE SOURCE PLACED AT HEIGHTS, H = 0M, 100M, 200M, 500M, AND 1000M, AT Z = 100M WITH τ1 = 80 Ω-M........................................................................................................70 FIGURE 8-6 PHASE OF ELECTRIC FIELD FROM A LINE SOURCE PLACED AT HEIGHTS, H = 0M, 100M, 200M, 500M, AND 1000M, AT Z = 100M WITH τ1 = 80 Ω-M. ..............................................................................................................71 FIGURE 8-7 AMPLITUDE OF THE ELECTRIC FIELD AT Z = 100M FROM A LINE SOURCE PLACED THE SURFACE OF A HOMOGENEOUS HALF-SPACE WITH τ1 = 10, 100, 1000 Ω-M................................................................................71 FIGURE 8-8 PHASE OF THE ELECTRIC FIELD AT Z = 100M FROM A LINE SOURCE PLACED THE SURFACE OF A HOMOGENEOUS HALF-SPACE WITH τ1 =10, 100, 1000 Ω-M ................................................................................72 FIGURE 8-9 INFINITE HORIZONTAL CURRENT DRIVE, TWO-LAYERED, EDDY CURRENT COUPLING GEOMETRY. ........73 FIGURE 8-10 AMPLITUDE OF THE ELECTRIC FIELD AT Z = 100M FROM A LINE SOURCE AT THE SURFACE OF A TWOLAYERED HALF-SPACE........................................................................................................................................74 FIGURE 8-11 PHASE OF THE ELECTRIC FIELD AT Z = 100M FROM A LINE SOURCE AT THE SURFACE OF A TWOLAYERED HALF-SPACE........................................................................................................................................74 FIGURE 8-12 GEOMETRY FOR EDDY CURRENT FIELD CALCULATIONS IN HOMOGENOUS HALF-SPACE DRIVEN BY UNIFORM MAGNETIC FIELD AT THE SURFACE. ....................................................................................................75 FIGURE 9-1 CALIBRATION FREQUENCY RESPONSE OF FIBER-OPTIC TRANSMITTER/RECEIVER PAIR. ..........................78 FIGURE 9-2 CALIBRATION FREQUENCY RESPONSE OF CURRENT PROBES USED...........................................................79 FIGURE 9-3 CALIBRATION FREQUENCY RESPONSE OF SANDIA DIPOLE ANTENNA. .....................................................79 FIGURE 9-4 CALIBRATION FREQUENCY RESPONSE OF NANOFAST HIGH-IMPEDANCE PROBE......................................80 FIGURE 9-5 CERTIFICATE OF CALIBRATION FOR 4395A NETWORK ANALYZER...........................................................81 FIGURE 10-1 DIRECT DRIVE CURRENT TRANSFER FUNCTION OF TROLLEY COMMUNICATION LINE WITH A LOCAL GROUND. .............................................................................................................................................................82 FIGURE 10-2 DIRECT DRIVE CURRENT TRANSFER FUNCTION OF TROLLEY COMMUNICATION LINE WITH A FENCE GROUND. .............................................................................................................................................................83 FIGURE 10-3 DIRECT DRIVE VOLTAGE TRANSFER FUNCTION OF CONVEYOR STRUCTURE WITH A LOCAL GROUND...83 FIGURE 10-4 DIRECT DRIVE CURRENT TRANSFER FUNCTION OF CONVEYOR STRUCTURE WITH A LOCAL GROUND...84 FIGURE 10-5 DIRECT DRIVE VOLTAGE TRANSFER FUNCTION OF CONVEYOR STRUCTURE WITH A FENCE GROUND. ..84 FIGURE 10-6 DIRECT DRIVE CURRENT TRANSFER FUNCTION OF CONVEYOR STRUCTURE WITH A FENCE GROUND. ..85 FIGURE 10-7 DIRECT DRIVE VOLTAGE TRANSFER FUNCTION OF RAIL STRUCTURE WITH A LOCAL GROUND.............85 FIGURE 10-8 DIRECT DRIVE CURRENT TRANSFER FUNCTION OF RAIL STRUCTURE WITH A LOCAL GROUND. ............86 FIGURE 10-9 DIRECT DRIVE VOLTAGE TRANSFER FUNCTION OF RAIL STRUCTURE WITH A FENCE GROUND. ............86 FIGURE 10-10 DIRECT DRIVE CURRENT TRANSFER FUNCTION OF RAIL STRUCTURE WITH A FENCE GROUND............87 FIGURE 10-11 DIRECT DRIVE VOLTAGE TRANSFER FUNCTION OF POWER CABLE SHIELD WITH A LOCAL GROUND...88 FIGURE 10-12 DIRECT DRIVE CURRENT TRANSFER FUNCTION OF POWER CABLE SHIELD WITH A LOCAL GROUND. ..88 FIGURE 10-13 DIRECT DRIVE VOLTAGE TRANSFER FUNCTION OF POWER CABLE SHIELD WITH A FENCE GROUND. ..89 FIGURE 10-14 DIRECT DRIVE CURRENT TRANSFER FUNCTION OF POWER CABLE SHIELD WITH A FENCE GROUND....89 FIGURE 10-15 DIRECT DRIVE VOLTAGE TRANSFER FUNCTION OF RAIL STRUCTURE WITH A LOCAL GROUND...........90 Appendix DD - Page 8 of 104 PAGE 8 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine FIGURE 10-16 DIRECT DRIVE CURRENT TRANSFER FUNCTION OF RAIL STRUCTURE WITH A LOCAL GROUND. ..........90 FIGURE 10-17 DIRECT DRIVE VOLTAGE TRANSFER FUNCTION OF RAIL STRUCTURE WITH A FENCE GROUND. ..........91 FIGURE 10-18 DIRECT DRIVE CURRENT TRANSFER FUNCTION OF RAIL STRUCTURE WITH A FENCE GROUND............91 FIGURE 10-19 DIRECT DRIVE CURRENT TRANSFER FUNCTION OF TROLLEY COMMUNICATION LINE WITH A LOCAL GROUND. .............................................................................................................................................................92 FIGURE 10-20 DIRECT DRIVE CURRENT TRANSFER FUNCTION OF TROLLEY COMMUNICATION LINE WITH A FENCE GROUND. .............................................................................................................................................................92 FIGURE 10-21 NORMALIZED COMPOSITE ELECTRIC FIELD FOR P-DIRECTED SURFACE CURRENT DRIVE AT POSITIONS FROM P2 TO P8. ...................................................................................................................................................93 FIGURE 10-22 NORMALIZED COMPOSITE ELECTRIC FIELD FOR P-DIRECTED SURFACE CURRENT DRIVE AT POSITIONS FROM X1 TO X9...................................................................................................................................................93 FIGURE 10-23 NORMALIZED VERTICAL ELECTRIC FIELD FOR P-DIRECTED SURFACE CURRENT DRIVE AT POSITIONS FROM P2 TO P8. ...................................................................................................................................................94 FIGURE 10-24 NORMALIZED VERTICAL ELECTRIC FIELD FOR P-DIRECTED SURFACE CURRENT DRIVE AT POSITIONS FROM X1 TO X9...................................................................................................................................................94 FIGURE 10-25 NORMALIZED P-DIRECTED ELECTRIC FIELD FOR P-DIRECTED SURFACE CURRENT DRIVE AT POSITIONS FROM P2 TO P8. ...................................................................................................................................................95 FIGURE 10-26 NORMALIZED P-DIRECTED ELECTRIC FIELD FOR P-DIRECTED SURFACE CURRENT DRIVE AT POSITIONS FROM X1 TO X9...................................................................................................................................................95 FIGURE 10-27 NORMALIZED X-DIRECTED ELECTRIC FIELD FOR P-DIRECTED SURFACE CURRENT DRIVE AT POSITIONS FROM P2 TO P8...................................................................................................................................96 FIGURE 10-28 NORMALIZED P-DIRECTED ELECTRIC FIELD FOR P-DIRECTED SURFACE CURRENT DRIVE AT POSITIONS FROM X1 TO X9...................................................................................................................................................96 FIGURE 10-29 NORMALIZED COMPOSITE ELECTRIC FIELD FOR X-DIRECTED SURFACE CURRENT DRIVE AT POSITIONS FROM P2 TO P8. ...................................................................................................................................................97 FIGURE 10-30 NORMALIZED COMPOSITE ELECTRIC FIELD FOR X-DIRECTED SURFACE CURRENT DRIVE AT POSITIONS FROM X1 TO X9...................................................................................................................................................97 FIGURE 10-31 NORMALIZED VERTICAL ELECTRIC FIELD FOR X-DIRECTED SURFACE CURRENT DRIVE AT POSITIONS FROM P2 TO P8. ...................................................................................................................................................98 FIGURE 10-32 NORMALIZED VERTICAL ELECTRIC FIELD FOR X-DIRECTED SURFACE CURRENT DRIVE AT POSITIONS FROM X1 TO X9...................................................................................................................................................98 FIGURE 10-33 NORMALIZED P-DIRECTED ELECTRIC FIELD FOR X-DIRECTED SURFACE CURRENT DRIVE AT POSITIONS FROM P2 TO P8...................................................................................................................................99 FIGURE 10-34 NORMALIZED P-DIRECTED ELECTRIC FIELD FOR X-DIRECTED SURFACE CURRENT DRIVE AT POSITIONS FROM X1 TO X9. ................................................................................................................................99 FIGURE 10-35 NORMALIZED X-DIRECTED ELECTRIC FIELD FOR X-DIRECTED SURFACE CURRENT DRIVE AT POSITIONS FROM P2 TO P8.................................................................................................................................100 FIGURE 10-36 NORMALIZED X-DIRECTED ELECTRIC FIELD FOR X-DIRECTED SURFACE CURRENT DRIVE AT POSITIONS FROM X1 TO X9. ..............................................................................................................................100 FIGURE 10-37 INDUCED VOLTAGE ON PUMP CABLE (~300 M LONG) DUE TO WIRE CURRENT DRIVES ON SURFACE. 101 Appendix DD - Page 9 of 104 PAGE 9 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine LIST OF TABLES TABLE 1-1 LIGHTNING DETECTION NETWORK DATA, JANUARY 2, 2006.....................................................................18 TABLE 3-1 DIRECT DRIVE MEASUREMENT LOCATIONS...............................................................................................32 TABLE 3-2 SUMMARY OF CURRENT TRANSFER FUNCTIONS, USING POSITIVE LIGHTNING WAVEFORM, FOR CONDUCTIVE PENETRATIONS WITH CURRENT MINE GROUNDING ...............................................................................................34 TABLE 3-3 SUMMARY OF CURRENT TRANSFER FUNCTIONS, USING POSITIVE LIGHTNING WAVEFORM, FOR CONDUCTIVE PENETRATIONS WITH FORMER MINE GROUNDING .................................................................................................34 TABLE 3-4 SUMMARY OF CURRENT TRANSFER FUNCTIONS, USING NEGATIVE LIGHTNING WAVEFORM, FOR CONDUCTIVE PENETRATIONS WITH CURRENT MINE GROUNDING..........................................................................35 TABLE 3-5 SUMMARY OF CURRENT TRANSFER FUNCTIONS, USING NEGATIVE LIGHTNING WAVEFORM, FOR CONDUCTIVE PENETRATIONS WITH FORMER MINE GROUNDING ...........................................................................35 TABLE 3-6 SUMMARY OF FIGURES FOR DRIVE CONFIGURATIONS .................................................................................39 TABLE 4-1 CHARACTERISTICS OF POSITIVE AND NEGATIVE LIGHTNING WAVEFORMS USED IN ANALYSIS ....................48 TABLE 4-2 DIRECT DRIVE MEASUREMENT LOCATIONS...............................................................................................49 TABLE 4-3 PEAK CURRENTS AND VOLTAGES FROM A POSITIVE 100 KA LIGHTNING STROKE, FOR CONDUCTIVE PENETRATIONS WITH OLD MINE GROUNDING .......................................................................................................49 TABLE 4-4 PEAK CURRENTS AND VOLTAGES FROM A POSITIVE 100 KA LIGHTNING STROKE, FOR CONDUCTIVE PENETRATIONS WITH CURRENT MINE GROUNDING ...............................................................................................50 TABLE 4-5 PEAK CURRENTS AND VOLTAGES FROM A NEGATIVE 100 KA LIGHTNING STROKE, FOR CONDUCTIVE PENETRATIONS WITH OLD MINE GROUNDING .......................................................................................................50 TABLE 4-6 PEAK CURRENTS AND VOLTAGES FROM A NEGATIVE 100 KA LIGHTNING STROKE, FOR CONDUCTIVE PENETRATIONS WITH CURRENT MINE GROUNDING ...............................................................................................50 TABLE 5-1 CURRENT AND VOLTAGE AT THE 2ND LEFT SWITCH DUE TO A 100 KA PEAK, POSITIVE CLOUD-TO-GROUND LIGHTNING STROKE AT THE ENTRANCE OF THE MINE ...........................................................................................59 Appendix DD - Page 10 of 104 PAGE 10 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Executive Summary This report documents measurements and analytical modeling of electromagnetic transfer functions to quantify the ability of cloud-to-ground lightning strokes (including horizontal arc-channel components) to couple electromagnetic energy into the Sago mine located near Buckhannon, WV. These transfer functions, coupled with mathematical representations of lightning strokes, are then used to predict electric fields within the mine and induced voltages on a cable that was left abandoned in the sealed area of the Sago mine. If voltages reach high enough levels, electrical arcing could occur from the abandoned cable. Electrical arcing is known to be an effective ignition source for explosive gas mixtures, and corona discharge has been postulated to be so as well. However, given the time scale of lightning (~100 μs), it is unlikely that corona would develop before an electrical arc. Corona is due to ionization of surrounding air and usually a precursor to arcing, given sufficient voltage. Two coupling mechanisms were measured: direct and indirect drive. Direct coupling results from the injection or induction of lightning current onto metallic conductors such as the conveyors, rails, trolley communications cable, and AC power shields that connect from the outside of the mine to locations deep within the mine. Indirect coupling results from electromagnetic field propagation through the earth as a result of a cloud-to-ground lightning stroke or a long, low-altitude horizontal current channel from a cloud-to-ground stroke. Unlike direct coupling, indirect coupling does not require metallic conductors in a continuous path from the surface to areas internal to the mine. Based on the direct coupling measurements, lightning currents attenuate rapidly on the conductors as a function of distance into the mine. It is highly unlikely that a worst-case lightning stroke could generate sufficient voltage on a cable within the sealed area to cause concern – even if the lightning stroke directly attached to physical conductors at the entrance to the mine. Results from the indirect coupling measurements and analysis are of great concern. The field measurements and analysis indicate that significant energy can be coupled directly into the sealed area of the mine. Due to the relatively low frequency content of lightning (< 100 kHz), electromagnetic energy can readily propagate through hundreds of feet of earth. Indirect transfer function measurements compare extremely well with analytical models developed for the Sago site which take into account measured soil properties. Lightning stroke data recorded by the National Lightning Detection Network and the United States Precision Lightning Network at the time of the explosion does not support the conclusion that high enough voltage to provide a source of ignition could be generated in the sealed area. However, analyses of credible hypothetical scenarios (an undetected stroke closer to the sealed area or a horizontal arc channel of a recorded stroke above the sealed area) indicate voltages large enough on the abandoned cable in the sealed area to be of concern for electrical arcing. Eyewitness accounts of simultaneous lightning and thunder above the sealed area at the time of the explosion lends further credence to these hypotheses. This work was sponsored by the Mine Safety and Health Administration. Due to the complexity of lightning interactions with large multi-path structures and the limited duration of this project, it was not possible to address the full intricacies of potential lightning interactions at the Sago mine. However, results cited in this report can be considered as a significant indicator of the potential for lightning to couple energy into underground mining structures. Significant follow-on research would be required to address the complexity of mining structures to an extent to fully characterize these energy coupling mechanisms. Once achieved, it is reasonable to expect that mitigation techniques and safety standards could be developed to secure mining structures from future lightning threats. Appendix DD - Page 11 of 104 PAGE 11 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine ABBREVIATIONS, ACRONYMS AND INITIALIZATIONS CW dB DOE FFT IFFT NLDN USPLN Continuous Wave deciBel Department of Energy Fast Fourier Transform Inverse Fast Fourier Transform National Lightning Detection Network United States Precision Lightning Network Appendix DD - Page 12 of 104 PAGE 12 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Measurement and Modeling of Electrical Transfer Functions for Lightning Coupling into the Sago Mine 1 Introduction On January 2, 2006, an explosion was initiated in a methane-air mixture within a sealed area at the Sago underground coal mine near Buckhannon, WV that resulted in the deaths of twelve miners. The approximate location of the initiation of the explosion is shown in Figure 1-1. Figure 1-1 Approximate Location of Initiation of Explosion in Sealed Area of Sago Mine. Because of the fraction of a second simultaneity of the explosion and nearby lightning strokes recorded by the National Lightning Detection Network (NLDN) and the United States Precision Lightning Network (USPLN), lightning is strongly suspected to have caused the explosion. Additional eyewitness reports of other lightning not recorded by NLDN and USPLN further these suspicions [21]. If the timing of the recorded lightning strokes and the underground mine explosion are considered independent statistical events, then the probability that such a combined event would occur at random in a given year is extremely low. When this highly improbable event is coupled with the fact that at least eleven underground coal mine explosions have occurred since 1990 (see Appendix D) in which lightning is suspected of being the cause, it further supports the need to understand the potential role of lightning in the Sago disaster [1-4]. The coupling mechanisms that may have brought lightning energy into the sealed area at Sago were unclear and complicated by the fact that there were no known metallic penetrations into the sealed area of the Sago mine, unlike other sealed area explosions. Prior to 1990, lightning location Appendix DD - Page 13 of 104 PAGE 13 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine and timing data was unavailable, leaving the possibility that many earlier mine explosions would also be correlated to lightning events. The goal of this project was to perform field measurements at the Sago site and to develop analytical models to quantify potential lightning coupling mechanisms that are capable of delivering significant energy into the sealed area of the Sago mine. 1.1 Motivation for Research and Measurements Over the last decade, Sandia National Laboratories (Sandia) has developed unique capabilities to characterize and mitigate lightning effects on high value assets within the Department of Energy (DOE) and other agencies as part of a national security mission in nuclear weapons stockpile stewardship. Additionally, the history of potential lightning induced mine explosions suggested that a program using modern electromagnetic measurement techniques and analysis could be valuable during the investigation at the Sago mine. These modern lightning coupling measurement techniques were developed by DOE/NNSA specifically for the evaluation of the performance of lightning protection systems on buried, explosive storage structures, nuclear weapons assembly and dismantlement facilities, and at tunneling systems at the DOE Nevada Test Site. These Sandia developed techniques have been compared and validated using rocket-triggered lightning measurements [5-7] and have undergone significant technical review within the DOE and by the Defense Nuclear Facility Safety Board, an independent federal agency established by Congress in 1988. 1.2 Objectives of Measurements The principal objectives of this program were to identify, characterize, and quantify the electromagnetic paths of lightning electrical energy into the sealed area of the Sago underground coal mine. These paths include direct coupling through metallic penetrations into the operating area of the mine and indirect coupling through the earth overburden to conductors in the sealed area. Measurement results are compared with basic analytical models to confirm the validity of proposed lightning coupling mechanisms. The measured transfer functions were then used to predict the voltages generated on a cable left abandoned within the sealed area from the lightning stroke locations and amplitudes determined by the NLDN and the USPLN. In addition, the raw lightning event data from the NLDN and USPLN was analyzed to ascertain if there were any instances of data at the correct time that did not meet all of the criteria to be recorded as a lightning stroke. 1.3 Previous Work on Lightning Induced Mine Explosions Recent previous works by Novak and others [8,9] have utilized commercial, numerical electromagnetic codes to calculate the voltages on metal-cased boreholes connecting the surface with the sealed areas in mines. They have postulated corona discharge as an initiating mechanism based on experimental work by combustion researchers [10,11]. Berger, Geldenhuys, Golledge, Zeh, and others have analyzed the specific situation of lightning-caused explosions in shallow South African underground coal mines [1216]. The Australian, German, and Chinese literature on lightning initiated underground coal mine explosions has not been thoroughly explored. 1.4 Measurement Method and Analysis The coupling mechanisms of lightning energy into the Sago mine have been divided into (1) direct coupling via metallic penetrations from the outside of the mine that are terminated immediately outside Appendix DD - Page 14 of 104 PAGE 14 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine the sealed area, and (2) indirect coupling through the soil and rock overburden above the sealed area. The metallic penetrations analyzed and measured were the AC power shields, the coal conveyer system, the transportation rail system, and the mine trolley communication cable. The primary focus of this study was to determine electric fields within the mine and the resulting induced voltage on a cable within the sealed area due to both the direct and indirect coupling mechanisms. Electrical arcing is known to be an effective ignition source for explosive gas mixtures, and corona discharge has been postulated to be so as well. However, given the timescale of lightning (~100 μs) it is unlikely that corona would develop before an electrical arc. Corona is due to ionization of surrounding air and usually a precursor to arcing, given sufficient voltage. Lightning coupling mechanisms were characterized by driving potential pathways with low-level, continuous sinusoidal signals and measuring the resultant signals at distant locations. The resultant data when divided by the input signal produces a transfer function that can be coupled with a mathematical representation of lightning strokes to calculate a resultant signal at points inside the mine. The advantages of using this technique are as follows: • • • • Measurements can be made without waiting for a natural or triggered lightning in the vicinity. Safety is not compromised due to use of low-level signals and interference with ongoing mine operations is minimized. The frequency content of the low-level drive signal can be tailored to that of natural lightning. Many data points can be taken with this method which enhances the precision of the transfer functions. The disadvantage is that the nonlinear effects of high-voltage arcing cannot be taken into account. 1.4.1 Direct Coupling Transfer Function Measurements and Analysis Because all metallic conductors into the Sago mine were terminated outside the sealed area of the mine, current cannot be injected from outside the sealed area directly into the sealed area. However, currents flowing on conductors inside the mine, but outside of the sealed area, may be able to induce voltage on a cable inside the sealed area through electromagnetic coupling. To determine the amplitude of these currents, attenuation on each conductor entering the mine was measured using transfer function techniques. Low-level direct coupling transfer function measurements were made by injecting current onto metallic penetrations at the entrance to the mine and then measuring the voltage and current levels on these penetrations at various points within the mine, up to immediately outside the sealed area. The voltage induced on conductors inside the sealed area could then be calculated based on the projected current level on conductors outside the sealed area and an analytical estimate of the electromagnetic coupling between this current and the conductors inside the sealed area. The measurements were made over a frequency range from 10 Hz to 100 kHz, corresponding to wavelengths in air of 3x107 meters to 3000 meters respectively. We are able to use very small signals because our instrumentation is very sensitive and has a large dynamic range. We demonstrated that we could measure input currents and at some distance and from the source even with significant attenuation. Because the direct-drive measurements are taken as a function of frequency, the mathematical representation of a lightning stroke is transformed as a function of frequency. To use the data, the directdrive transfer functions were multiplied by the frequency representation of a lightning stroke. The product was inverse Fourier transformed to represent the resultant signals inside the mine from a lightning event outside the mine, as a function of time. To represent the worst-case scenario input for the purpose of these calculations, an assumption was made that the lightning stroke attached to the metallic penetrations at the entrance of the mine. Appendix DD - Page 15 of 104 PAGE 15 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 1.4.2 Indirect Coupling Transfer Function Measurements and Analysis The large currents in a lightning stroke have an associated magnetic field. When a lightning stroke attaches to the earth, this creates a magnetic field tangential to the ground. For a fully developed lightning stroke, it is reasonable to approximate this magnetic field as H (r ) = I 2π r over a distance of 30 m – 1000 m, where I = lightning current and r = distance from stroke attachment. For distances within 30 m of the attachment, magnetic field calculations are more complex and this approximation is incomplete. To a first order of approximation and as a bound, the magnetic field is calculated above a perfectly conducting ground plane, as above. This approximate tangential magnetic field is used as a drive to generate current in the finitely conducting earth. The calculations in this report do not deal with magnetic fields generated in the immediate vicinity of lightning strokes; therefore, these interactions will not be evaluated here. For distances greater than 1000 m from the attachment, the approximation for the magnetic field at the surface may be an overestimation, but can be considered a reasonable upper bound. When lightning attaches to the ground, the magnetic field tangential to the ground creates currents not only on the surface, but deeper in the earth as well. It is a fundamental principle of electromagnetics that magnetic fields on the surface of a conductor can generate currents within the conductor of some depth. For frequencies sufficiently low that displacement currents can be neglected, this is called the skin effect and is dependent upon the resistivity of the conductor. When displacement currents are neglected, the electromagnetic coupling phenomena are called diffusion coupling or, equivalently, eddy current coupling. The skin depth characterizes the exponential decay of these currents in planar geometries. Resistivity measurements have shown the soil in the vicinity of the sealed area of the Sago mine to be a fairly good conductor; therefore, it is reasonable to assume that some electromagnetic energy can propagate from the surface of the earth into the sealed area of the mine. This effect is similar to propagating radio waves through seawater, also a fairly good conductor, and communicating with submarines. The methodology used to measure the electromagnetic coupling through the earth is to simulate magnetic fields in the earth by connecting a frequency variable voltage source via straight wires on the surface between ground rods at either end of the wires. The ground rods are placed a significant distance from each other, approximately 100 m on either side of the region where the electric fields are measured, or where voltage is induced on an insulated wire. The electric field and the voltage on a cable are measured over a frequency range from 10 Hz to 100 kHz. At this point we have the electric field and voltage response on the pump cable in the sealed area from a known linear current distribution on the surface. Two steps are involved in calculating the response of a lightning stroke attachment at a distance from the sealed area. The first step involves estimating the magnetic field (or surface current) above the sealed area from a lightning attachment to the ground at a distance from the sealed area. The second step involves calculating the electric field in the sealed area of the mine due to the uniform magnetic field (or surface current) on the surface using the parameters determined from the coupling measurements. Once these connections are made with data in the frequency domain, then the Fourier transform of the lightning stroke can be multiplied by the transfer function. The inverse Fourier transform of the product can be taken to determine the peak electric field and peak voltages that would be caused by a lightning attachment of a given amplitude at a given location with respect to the sealed area. If the peak induced voltages are significant, arcing between conductors could occur. A few tenths of a milliJoule of energy in the arc would be a sufficient ignition source for a combustible methane-air mixture [17]. This amount of Appendix DD - Page 16 of 104 PAGE 16 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine energy is readily available from almost any arcing process envisioned in a lightning induced event. Bulk air breakdown in small gaps (several millimeters) occurs at average electric field values of approximately 10 kV/cm with standard lightning waveforms [18]. Surface arcing can occur at electric field values in the 5 kV/cm range. 1.5 Soil and Rock Site Data The soil and rock resistivity play a major role in determining the amplitude and frequency dependence of indirect coupling into the sealed mine area. Several studies provide resistivities measured with different techniques and equipment. The resistivities determined by the different measurements appear to be somewhat inconsistent. However, resistivities in [19] match the numbers that give us the best fit for our analysis of electromagnetic coupling through the ground. The resistivities in [19], using a best fit to electromagnetic sounding data, are 100 Ohm-m from 0 to 40 feet, 10 Ohm-m from 40 to 120 feet, and 100 Ohm-m from 120 to 350 feet deep, yielding an average of 77.3 Ohm-m above the sealed area at the borehole. In this study an average resistivity of 80 Ohm-m is used to characterize the soil and rock overburden atop the sealed area of the Sago mine. 1.6 Lightning Event Information Three positive polarity lightning strokes were identified by the NLDN and the USPLN that were coincident with the Sago underground coal mine sealed area explosion. Their location, polarity, and amplitude are shown in Figure 1-2. Figure 1-2 Location of Lightning Strokes at Sago Mine Contemporaneous with Sealed Area Explosion. Appendix DD - Page 17 of 104 PAGE 17 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Table 1-1 gives the location, polarity, and amplitude of the identified strokes. Also provided in the table are the distances from the stroke locations to the sealed area, and the angle that a line between the stroke location and the borehole above the sealed area makes with the pump cable in the sealed area. It should be noted that physical evidence of only stroke number 3 was found after several searches of each attachment area. An analysis of the USPLN and NLDN data strongly suggest that stroke number 1 and 2 in Table 1-1 represent a single stroke, and not two separate events [20,21]. Table 1-1 Lightning Detection Network Data, January 2, 2006 Stroke No. Time 1 6:26:35.522am 2 6:26:35.522am 3 6:26:35.680am Longitude/ Latitude N38.897/ W80.231 N38.9071693/ W80.2201 N38.926/ W80.233 Angle with Cable (Degrees) Detection System (kA) Distance to Borehole (km) Positive 38.8 5.44 52.8 NLDN Positive 35 4.02 49.3 USPLN Positive 101 2.91 85.5 NLDN Polarity Amplitude The accuracy of the NLDN is shown in general by the confidence ellipses drawn around the most probable locations. The ellipses give the probability that the lightning is actually inside the ellipse. The estimated 99% location uncertainty for both strokes detected by NLDN was better than 1.1 km (0.7 miles). The fact that the tree was found damaged approximately 197 feet (59 m) from the most probable location of the 101 kA stroke further demonstrates the NLDN location accuracy near the Sago mine [20, 21, 35]. Recent validation experiments on the NLDN have shown stroke detection efficiencies between 70 – 85% and flash detection efficiencies of 90 – 95% [34]. (Lightning flashes are typically comprised of multiple strokes.) It is believed that the two strokes (1 and 3 from Table 1-1) at Sago were part of the same flash [35]. Several other possibilities exist that were not, or could not, be confirmed by the lightning detection network data. Although quite reliable and accurate, the possibility exists of strokes not being detected. Simultaneous thunder and flash were reported by residents living on top of or nearby the sealed area [21]. In addition, the lightning detection networks are designed to locate the ground strike points of cloud-toground strokes and do not provide information about the channel geometry above those points, such as if a stroke had a long, low horizontal component that could be important in radiating fields into the mine. Also, upward discharges that are initiated by tall vertical structures will not be detected by the systems unless the initial continuous current phase is followed by at least one leader-return stroke sequence [20, 35]. There were several tall communication towers (the tallest being ~ 200 ft.) within a mile of the sealed area, the closest being approximately 0.5 miles. 1.7 Other Site Information Measurements discussed in this report were made on the most likely coupling paths into the sealed area. Other potential conduits of lightning energy are mentioned in this section, but were not characterized due to the limited budget and schedule of this project. While they are mentioned here for completeness, the lack of measured data on them does not change the conclusions in this report. If it is desired to develop an overall lightning protection scheme specific to the Sago mine, it would be useful to characterize these potential conduits in the future. Appendix DD - Page 18 of 104 PAGE 18 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine All vertical pipes in the vicinity of the sealed area are shown in Figure 1-3. The vertical pipe closest to the sealed area of the mine is the gas well pointed out in Figure 1-3. It is unlikely that any field enhancements due to the vertical pipes would induce a significant amount of voltage onto the pump cable in the sealed area because the cable is orthogonal to the pipes. However, as potential conduits for lightning energy, they are mentioned here for completeness. The horizontal gas pipes that are in the vicinity of the sealed area are also shown in Figure 1-3. These pipes are in general buried at a depth of 2 feet from the surface. The response on the pump cable, or electric fields in the sealed area, due to the current drive of the horizontal gas pipes was not characterized because it was not planned for and was not characterized because of liability issues. The gas pipes, if driven locally to the sealed area, would have similar coupling characteristics to the pump cable as that of the indirect drive experimental setup. If the gas pipes were driven remotely, the amount of attenuation from one point on the pipes to another point is mostly dependent upon the resistivity of the soil surrounding the pipes. If the soil surrounding the pipes has low resistivity, a majority of current injected onto the pipes would attenuate in a short distance. However, if the pipes are either not in contact with the soil or the resistivity of the soil is large, then the pipes would act as insulated conductors. Attenuation on the pipes in this case would be much less. Closest Gas Well Figure 1-3 Vertical Pipes in Vicinity of Sealed Area of Sago Mine. Both telephone wires and AC power lines were in the vicinity of the 101 kA positive stroke and could have provided metallic conduction paths into the Sago mine AC power system, or the telephone communication system, or to other metallic penetrations into the mine. The location and routing of this wiring with respect to the stroke are shown in Figure 1-4. The direct-drive measurements discussed in Appendix DD - Page 19 of 104 PAGE 19 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Section 3.1 lead to the conclusion that even if the power and telephone lines were conduits of the lightning energy, they would not be a plausible source of energy to cause high voltage in the sealed area. Figure 1-4 AC Power Distribution Lines and Telephone Lines near Positive 101 kA Stroke. The presence of metallic roof mesh and pump cabling and its relationship to the approximate location of initiation of the explosion are shown in Figure 1-5. The pump cable is shown as the red line and the green shaded area depicts the metallic mesh. The pump cable is noted because indirect coupling measurements are made on it. With these measurements, the voltages induced on the pump cable due to lightning strokes on the surface are calculated in this report. The metallic mesh is noted because it is used in some of the measurements for grounding purposes. It was not considered a plausible receiver or antenna of the electromagnetic energy that propagates underground because it appears to be well grounded at regular intervals to the roof of the sealed area, and, therefore, would not support a large voltage potential. Appendix DD - Page 20 of 104 PAGE 20 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Figure 1-5 Roof Mesh and Cable in Sealed Area Where Explosion Was Initiated. The red line represents a cable from a water pump located at the top of the figure. The green lines represent metallic roof mesh. 1.8 Fidelity Issues of Study To have confidence in the measured results, several fidelity issues were addressed to ensure that the measurements could be used to calculate a realistic natural lightning response. 1.8.1 Current Flow on the Surface from a Real Lightning Stroke and the Indirect-drive Test Setup Consideration was given here to two issues that could limit the applicability of the indirect measurements. The first consideration is whether the current flow pattern in the earth is sufficiently similar to lightning. The electric and magnetic fields near rocket-triggered lightning have been measured, and the current flow in the soil out to 30 m distance from the attachment can be inferred [22,23]. Nonlinearities at the lightning attachment point often cause arcing either on the surface of the soil or into the soil that are not duplicated by the low-level drive current measurement method. Because these arcs are limited to the attachment area, they do not affect the overall current flow at large distances to a significant degree. We are not modeling the stroke attachment region, as stated previously. A second, more significant consideration is that the near-surface current flow pattern produced by these measurement techniques may not accurately represent natural lightning current flow patterns. This is possible because either the current flow at large distances from the attachment point is not duplicated due to the use of ground rods as a return current path during the measurement, or because resistive inhomogeneities in the soil and rock overburden can perturb the flow pattern. However, good correlation between the measured results and the homogeneous earth models suggest these deviations are negligible for this particular project. Appendix DD - Page 21 of 104 PAGE 21 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 1.8.2 Physical Changes to the Sago Site after the Accident Physical changes were made at the site after the explosion occurred and before Sandia researchers arrived at Sago. These changes do not impact the validity of these measurements, but they are included here for completeness. A three-inch borehole was drilled into the sealed area of the mine immediately above where the explosion is likely to have been initiated. The borehole has only fourteen feet of steel casing from the surface downward, which should not affect the measurements significantly. Also, two eighteeninch steel casings were added to connect water pumps in the north end of the sealed area to the surface. Because these pipes are a large distance from the region of the sealed area where the explosion originated and are orthogonal to the pump cable, they are not expected to affect the measurements significantly. The pump cable in the sealed area was modified for the indirect drive measurements. The pump cable was spliced with 12-gauge wire to recreate the length of pump cable believed to have been there during the explosion 1. For the measurements, the pump cable was connected with 12-gauge wire to the ceiling mesh and the exposed conductors were placed underwater approximately four crosscuts from the back of the sealed area. The approximate total length of the recreated cable was 300 m (984 ft). 1.9 Potential Further Areas of Study The following items are potential areas for further study. Their effects on the coupling mechanisms characterized in this report are unknown, but believed to be of minimal effect. Evaluating these areas will not change the basic conclusions in this report. 1.9.1 Nonlinearities Surface arcing and arcing through soil and rock are well-known phenomena that can propagate lightning energy over a distance of a hundred feet or less. Because these phenomena occur only at the full amplitudes of natural or triggered lightning strokes, their behavior and effect on coupling could not be studied using the low-amplitude transfer function measurement techniques of this study. There is no evidence an arc can travel a distance of 300 feet through soil and rock, therefore, it is unlikely this would have any effect on this analysis. 1.9.2 Coupling from Vertical Pipes near Sealed Areas The effect on the coupled electromagnetic field caused by direct drive of the vertical gas well that passed near the sealed area was not measured or modeled in this study. Because we could not guarantee that damage to cathodic protection systems or other instrumentation would not be caused by our drive system, the owners of the system would not allow attachment to the pipe without indemnification. Direct drive of the vertical pipe could have caused some enhancement of the coupled electric fields in the sealed area, but would not change the conclusions of this report. 1.9.3 Distributed Drives for Metallic Penetrations Although the localized drive at the entrance to the mine of all metallic penetrations to the mine was studied (except pager communication line), the propagation of voltages and currents on these penetrations 1 As a note, there is some disagreement as to the length and positioning of the pump cable at the time of the explosion. The test team used information provided at the time of the measurements, which was that the pump cable was intact and the cable shield was grounded to the submerged pump. Appendix DD - Page 22 of 104 PAGE 22 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine can be enhanced by current flow on the surface of the earth above the penetrations. Simple considerations indicate that the voltage and current amplitudes are not enhanced significantly. The measurement that could have elucidated this phenomenon was cancelled because of the physical and political impracticality of stringing a wire from the entrance of the mine through dense forests and livestock-occupied pastureland to a location above the sealed area. 1.9.4 Amplification Effects of Wiring Resonances Several coupling resonances were identified on the mine trolley communication and power wiring that could enhance lightning coupled voltages in sealed and unsealed areas of the mine. The characteristics of the resonances were so small that the enhancement would not be significant; however, the maximum extent to which this factor could amplify voltages in sealed areas was not studied. 1.9.5 Effect of Grounded Roof Meshes Voltages induced between sections of roof meshes in the sealed area were not measured because the substantial grounding of these meshes via rods driven every three feet or so to provide roof support was thought to prevent buildup of voltages. We found at the site that the use of nonconductive epoxies may prevent good contact between the epoxy bolts and the rock. The voltage buildup between sections of roof mesh and the effect of the roof mesh on electric fields and voltages within the sealed area was not studied in this project. [36] 1.9.6 Coupling Paths Not Present in Sago Mine Lightning coupling paths into sealed areas that are common in other underground coal mines but are absent from the Sago mine, such as coupling along metal-cased boreholes that extend from the surface into the sealed area and coupling through other metallic penetrations used for monitoring or other instrumentation were not studied in this project. 1.9.7 Geologic Irregularities Affecting Coupling The extent to which geologic irregularities such as faults and mineral deposits that affect the coupling of lightning energy into underground coals was not quantified in this study. 1.9.8 Lightning Current Return Path Assumptions The analysis used in this report assumes that lightning current is uniform in the radial direction. The extent to which large-scale inhomogenieties affect the current paths, and the extent to which the variation with depth affects the coupling, were not quantified in this study. Appendix DD - Page 23 of 104 PAGE 23 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 2 Electromagnetic Coupling Phenomenology Models Modeling was included in this project to compare the measurements with theoretical calculations. The results for mathematical modeling of coupling of electromagnetic energy into the mine by direct coupling and by indirect coupling are now given. The details of the derivations and derivations of more complicated models are given in Appendix A. 2.1 Direct Coupling via Metallic Penetrations into Mine The conductive penetrations into the mine can be modeled as transmission lines, or lines of distributed impedance (i.e., the combination of resistance, capacitance, and inductance). It is helpful to model the penetrations as transmission lines, because then their behavior over a wide frequency range, such as the measurements made here, can be analyzed. The classic theory of transmission lines is detailed by King in [24]. Useful formulas for calculating the transmission-line parameters in situations similar to those at the Sago mine are given by Warne and Chen in [25]. 2.1.1 Localized Drive Transmission-line Theory Using the differential circuit representation in Figure 2-1, the equations of transmission-line theory can be developed [24]. Iz I z+Δz rΔz + − lΔz gΔz Vz cΔz + − +z Figure 2-1 Equivalent Circuit of a Section of Transmission Line. The transmission-line equations are given by d 2V dz 2 d 2I dz 2 = yzV z = yzI z y = g + iω c z = r + iω l The complex propagation constant is given by γ 2 = yz = ( g + iω c )( r + iω l ) Appendix DD - Page 24 of 104 PAGE 24 OF 104 Vz++Δz Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine These equations along with current or voltage source terms corresponding to localized current or voltage drives and appropriate loads have been used to develop a formal theory of transmission lines [24], which, along with properly determined transmission line parameters, is appropriate to the study of the propagation of lightning currents along direct coupling paths on metallic conductors into the Sago mine. Note that the variable z is used both as the distance along the transmission line and as the impedance parameter for the transmission line. 2.1.2 Distributed Drive Transmission-line Theory In many situations the current and voltage sources driving the transmission line of Figure 2-1 are not localized to a small volume but are distributed incremental current and/or voltage sources generated along the length of the transmission-line. An appropriate theory for this type of drive has also been developed in [24]. This type of transmission-line treatment is appropriate if the stroke does not directly attach to or is not conducted via metallic connections to the transmission-line. 2.2 Indirect Electromagnetic Coupling via Soil and Rock To calculate the electric fields in the earth induced by a current on the surface, the problem is simplified by representing the earth as a homogeneous material with a constant resistivity. Section 2.2.1 calculates the simplest case given a finite-length, DC current drive. The calculations become more complex in Sections 2.2.2 and 2.2.3 as the current drive is assumed to be of infinite length and time-varying, as more appropriate for lightning currents on the surface. These results are used to compare to the indirect measurements of the electric field in the sealed area as a function of the drive current on the surface. 2.2.1 Static Coupling Model for Current Injected into Homogeneous HalfSpace The geometry for the simplest model for DC current coupling is shown in Figure 2-2 and is analyzed in [26]. z y region – 0 ε0, μ0, τ0= ∞ I -I (-b,0,0) region – 1 ε1, μ0, τ1 (b,0,0) r1 x r2 (x,y,z) -z Figure 2-2 DC Current Drive with Homogeneous Conducting Half-Space. Appendix DD - Page 25 of 104 PAGE 25 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Current I is driven into the conductive half-space at Cartesian coordinate (-b,0,0) and the current is extracted at Cartesian coordinate (b,0,0). The upper half-space, region-0, has infinite resistivity and the lower half-space, region-1 has resistivity, τ 1 . From simple considerations, V(x,y,z), the potential at Cartesian coordinate (x,y,z) with respect to infinity is given by ⎛ V ( x, y , z ) = τ1I ⎜ 2π ⎜⎜ 1 ( x + b) ⎝ 2 + y2 + z2 − ⎞ ⎟ 2 2 2 ⎟ ( x − b ) + y + z ⎟⎠ 1 The electric field at point (x,y,z) is easily calculated from E ( x, y, z ) = −∇V ( x, y, z ) And calculating the x-component of interest E x ( x, y , x ) = − ⎛ ∂ V ( x, y , z ) ∂x ⎞ ⎟ − ⎟ 3 3 2 2 2 2 ⎤2 2 2 ⎤2 ⎟ ⎡ ⎡ ⎜ ( x + b) + y + z ⎟ x − b) + y + z ⎦ ⎣( ⎦ ⎠ ⎝⎣ ( x + b) τ I⎜ = 1 ⎜ 2π ⎜ ( x − b) The next coupling models to be considered are generalizations where the current is time varying say as with eiω t and the displacement currents are neglected because region-1 is assumed to be a good conductor. This generalization turns out to be more difficult than one might expect because the current in the earth depends on the geometry of the current path above the earth. A simpler model that corresponds to the electromagnetic coupling below an infinitely long, horizontal wire grounded at a large distance away and driven by a voltage source is, however, developed in the next section. 2.2.2 Infinite Line Source above Homogeneous Half-Space The current drive geometry of an infinitely long, horizontal wire placed a distance, h, above a conductive half-space is shown on the left side of Figure 2-3. A side view is shown on the right side of Figure 2-3. Similar configurations are analyzed in [27-31]. z z y −∞ I region – 0 ε0, μ0, τ0=∞ I ∞ h h x x region – 0 = ε0, μ0, τ0∞ region – 1 ε1, μ0, τ1 region – 1 ε1, μ0, τ1 (y,z) (y,z) -z Figure 2-3 Infinite Length, Harmonically Time Varying Horizontal Current Drive over a Conductive HalfSpace. Appendix DD - Page 26 of 104 PAGE 26 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine The current drive is harmonically time-varying and is directed along the positive x-axis at height, h, above it. The upper half-space has infinite resistivity and the lower half-space has resistivity, τ1. If one neglects displacement current and relates current density, ix(x,y,z) and electric field, Ex(x,y,z), in region-1 through, Ex(x,y,z)=τ1ix(x,y,z), then the current density in the lower half-space, region-1, can be determined to be Ex ( y, z ) = ikς 0 ∞ eqz e − uh cos uydu π ∫0 u + q where k = ω μ 0ε 0 q = u 2 + ip 2 p2 = ωμ 0 2 = 2 τ1 δ1 2τ 1 ωμ 0 Numerical calculations of this integral are carried out in Appendix A. δ1 = Note that the skin depth, δ, plays an important role as a parameter in all diffusion coupling calculations. For convenience it is plotted for resistivities of 10, 100, and 1000 Ohm-m in Figure 2-4. At a given frequency, the lower the resistivity the smaller the skin depth, meaning a majority of the current is contained closer to the surface. Hence, there will be better coupling deeper underground for ground with resistivity of 100 Ohm-m than for ground with resistivity of 10 Ohm-m. 10 4 τ1 = 10 Ohm-m τ1 = 100 Ohm-m τ1 = 1000 Ohm-m Skin Depth, δ 1 (m) 10 10 10 3 2 1 0 10 1 10 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 2-4 Skin Depth, δ1, as a Function of Frequency for Resistivities of 10, 100, and 1000 Ohm-m. Appendix DD - Page 27 of 104 PAGE 27 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 2.2.3 Infinite Line Source at Surface of Homogeneous Half-Space If the line current source is brought to the surface of the conducting homogeneous half-space, where h=0, integrating this result for y=0 to get the horizontal electric field immediately below the current source yields ⎧⎡ ⎫ ⎤ ⎪⎢ ⎪ ⎥ z ⎡ ⎡ τ1I 1 ⎪⎢ z⎤ z ⎤⎪ 1 1 ⎥ − (1+ i ) δ1 1 + − i 2 K 0 ⎢(1 + i ) ⎥ − (1 + i ) Ex ( y = 0, z ) = e K1 ⎢(1 + i ) ⎥ ⎬ ⎨ (1 + i ) π δ12 ⎪ ⎢ δ1 ⎦ δ1 ⎦ ⎪ ⎛ z ⎞ ⎛ z ⎞2 ⎥ ⎛ z⎞ ⎣ ⎣ ⎢ ⎥ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎪⎢ ⎪ δ δ ⎥ ⎝ 1 ⎠ ⎝ δ1 ⎠ ⎦ ⎝ 1⎠ ⎩⎣ ⎭ where K0 and K1 are modified Bessel functions. Note that we are now using positive z in the downward direction in the formula. A plot of the electric field at z=100 m depth for resistivities of 10, 100, and 1000 Ohm-m is shown in Figure 2-5. 10-1 z y −∞ Abs[Ex] (V/m/A) 10-2 I region – 0 ε0, μ0, τ0=∞ ∞ h x region – 1 ε1, μ0, τ1 (y,z) 10 -z -3 10-4 h=0m Tau1 = 10 Ohm-m Tau1 = 100 Ohm-m Tau1 = 1000 Ohm-m 10-5 10-6 101 2 3 4 5 67 2 102 3 4 5 67 103 2 3 4 5 67 104 2 3 4 5 67 105 104 2 3 105 Frequency (Hz) a.) Amplitude of Ex 200 z y −∞ I region – 0 ε 0 , μ 0 , τ0 =∞ ∞ h x region – 1 ε1, μ0, τ1 100 Arg[Ex] (Degrees) (y,z) -z 0 h=0m Tau1 = 10 Ohm-m Tau1 = 100 Ohm-m Tau1 = 1000 Ohm-m -100 -200 101 2 3 4 5 67 102 2 3 4 5 67 103 2 3 4 5 67 4 5 67 Frequency (Hz) b.) Phase of Ex Figure 2-5 Amplitude and Phase of Electric Field as a Function of Frequency at Depth of 100m with Resistivities of 10, 100 and 1000 Ohm-m. Appendix DD - Page 28 of 104 PAGE 28 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 2.2.4 Uniform Magnetic Field at Surface above Homogeneous Half-Space Assume that a uniform y-directed magnetic field of intensity, H0, is instantaneously applied above a conducting half-space, as shown in Figure 2-6. z z y H = H0yy region – 0 ε0, μ0, τ0=∞ H = H0yy dl h x x region – 0 ε0, μ0, τ0=∞ y region – 1 ε1, μ0, τ1 region – 1 ε1, μ0, τ1 (y,z) (y,z) -z Figure 2-6 Harmonically Time-Varying Magnetic Field Drive over Conductive Half-Space. If displacement current is neglected, the horizontal electric field below the surface of the conductive halfspace is given by Ex ( z ) = τ 1 (1 + i ) − (1+ i ) H0 y e z δ1 δ1 Note that positive z is used in the downward direction in the formula. Also note that this formulation describes the electric field due to the uniform surface current produced by a cloud-to-ground lightning stroke. Appendix DD - Page 29 of 104 PAGE 29 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 3 Measurement Methods 3.1 Direct Drive The goal of direct drive is to characterize the attenuation or decrease in signal from the entrance of the mine to various distances into the mine. This is accomplished by directly injecting a current on various conductive lines going into the mine and measuring the current at points further in the mine. Ideally, the transfer functions of each conductive line going into the mine would be measured in the same configuration as it was during the time of the explosion. However, the grounding at the entrance of the mine was changed following the explosion. Changing back to the old grounding state required the power to the mine be removed, thus stopping all mining operations. A small set of measurements were made while power was disconnected on Sunday, November 5th. It was not possible to conduct all measurements in the one day when the power was disconnected, and stopping mine operations for three days was not feasible. Therefore, the rail, trolley communication line, and conveyor belt structure were measured at six locations at a later time (November 8th and 9th), with the mine grounding system in its current state. The transfer function of the shield on the power cables could not be measured while power was energized. 3.1.1 The Differences and Similarities between Conductive Penetrations The four conductive penetrations going into the entrance of the mine that were measured were 1) the shield on the power cable, 2) the rail, 3) the trolley communication line, and 4) the metallic structure of the belt conveyor. The conveyor structure and the rail both appeared to be grounded frequently (the rail by surface contact with the ground and periodic bolts), and the conveyor structure by periodic legs bolted to the ground. The shield on the power conductor appeared to be grounded at each power center. The trolley communication cable was an isolated wire running the length of the mine and was only grounded at the entrance. Because of this, at low frequencies the attenuation on the trolley communication line is quite small. 3.1.2 Setup/Equipment Layout with Photos The principal measurement method used to characterize the coupling through metallic penetrations into the mine is shown in Figure 3-1. Current is driven onto the metallic penetration with an audio amplifier and is returned through either a local ground or a "fence" ground. The local ground consisted of three 18inch long conductive ground rods driven into the top soil. Each rod was approximately five feet from the other rods and 20 feet away from the driving point. The "fence" ground was long wire attached to the chain link fence that runs along the hillside above the entrance of the mine. The reasoning for the two grounding techniques was to help show the difference between a local point source drive and a distributed current source drive. A lightning stroke drive could be a distributed current source. The fence drive provided a distributed source for at least several hundred meters. The fence ground also provided a lower ground resistance, which in turn allowed more current to be driven on the lines. By driving more current on the line, the dynamic range of the measurement system was increased. The resistance of the local ground with respect to the rail was 90.2 Ω and the resistance of the fence ground with respect to the rail was 3.68 Ω. It is easy to see from this DC measurement that 20 to 30 times the amount of current could be driven on the fence ground than the local ground. The resistance of the local ground with respect to the conveyor structure was 97.5 Ω, and the resistance of the fence ground Appendix DD - Page 30 of 104 PAGE 30 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine with respect to the conveyor structure was 10.08 Ω. All DC ground measurements were made with a Megger DET 5/2 Earth Tester. The direct-drive system is broken into two parts, the drive end and the measurement end. The drive end consists of a 12 V marine battery and a 120 VAC inverter to provide isolated power for the measurement equipment, a fiber optic receiver, an audio amplifier driven by a network analyzer, connecting wires to the conductor being driven, and wires to a ground (local or fence). The drive signal produced by the network analyzer is optically coupled to the audio amplifier allowing the signals to be phase-locked to increase the sensitivity of the measurements and allow for phase measurements. The technique of phaselocked detection allows measurement of voltages as low as 10s of nanoVolts. The measurement end consists of a 12 V battery and a 120 VAC inverter for isolated power for the measurement equipment, a fiber optic transmitter, a network analyzer, and current and voltage measurement probes. The voltage measurements on the rail, power, and conveyor were made with respect to the roof mesh. Voltage measurements were not made on the trolley communication line because it was isolated without an exposed conductor. rail Current probe to measure signal Few hundred feet to 2 miles 40 V, 1 A max 10 Hz – 100 kHz Network analyzer (for measuring signal) Fiber-optic line to drive the audio amplifier and keep the Network analyzer phase locked. Audio Amp Fiber-Optic Rx Fiber-Optic Tx Ground rod + - 120 VAC inverter + ground - 120 VAC inverter 12 V battery Electrical power for network analyzer 12 V battery Electrical power for amplifier Figure 3-1 Direct Drive Conceptual Drawing. The measurements were conducted at seven locations along the left rail, trolley communication line, and the conveyor belt structure as they proceeded into the mine. Measurements were conducted at the first three locations for the power cable shield. The power cable shield measurements were completed on Sunday, November 5th while the power was turned off. The power cable shield was not measured during regular mine operation due to safety concerns. Table 3-1 lists mine features at each measurement location, the approximate distance to the entrance (drive location), and the conductors measured. The measurement locations are also shown on the map of the mine in Figure 3-2. Appendix DD - Page 31 of 104 PAGE 31 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Table 3-1 Direct Drive Measurement Locations Conductors Measured: Voltage (V) & Current (I) Location Mine Feature Break Number Approximate Distance from Entrance 1 #1 Power Center Belt 1, Break 1 30 m (98 ft.) 2 #2 Power Center Belt 2, Break 1 459 m (1506 ft.) Power Cable Shield (V&I) Trolley Comm Line (I) Rail (V&I) 3 #3 Power Center Belt 3, Break 1 669 m (2195 ft.) Power Cable (V&I) Trolley Comm Line (I) Rail (V&I) st 4 1 Right Spur Belt 3, Break 16 1076 m (3530 ft.) nd 5 2 Right Spur Belt 4, Break 11 2178 m (1.35 miles) st 6 1 Left Switch Belt 4, Break 50 3255 m (2.02 miles) nd 7 2 Left Switch Belt 4, Break 59 3491 m (2.17 miles) Grounding System in 2 Configuration 1 Grounding System in 3 Configuration 2 Power Cable Shield (V&I) Trolley Comm Line (I) Rail (V&I) Conveyor (V&I) Figure 3-2 Direct Drive Measurement Locations. 2 3 Mine grounding system similar to the grounding scheme in place during explosion. Mine grounding system in current state. Appendix DD - Page 32 of 104 PAGE 32 OF 104 Trolley Comm Line (I) Rail (V&I) Conveyor (V&I) Trolley Comm Line (I) Rail (V&I) Conveyor (V&I) Trolley Comm Line (I) Rail (V&I) Conveyor (V&I) Trolley Comm Line (I) Rail (V&I) Conveyor (V&I) Trolley Comm Line (I) Rail (V&I) Conveyor (V&I) Trolley Comm Line (I) Rail (V&I) Conveyor (V&I) Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Three current probes were used for the various measurements: a Pearson model 110A; a Pearson model 4688; and a LEM-flex model RR3035 current probe. The voltage was measured with a high-impedance voltage probe model P601 made by Nanofast. The current and voltage probes are shown on the various conductive penetrations in Figure 3-3.The calibration curves for each probe versus frequency are located in Appendix B. (A.) (C.) (B.) (D.) Figure 3-3 (A.) Current probe on trolley communication cable. (B.) Current probe and voltage connection on conveyor belt structure. (C.) Voltage probe on power cable. (D.) Current probe and voltage connection on rail. Appendix DD - Page 33 of 104 PAGE 33 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 3.1.3 Results The results from the direct drive measurements were consistent with expectations. All of the processed spectral, or frequency-domain, voltage, and current transfer functions for each conductive penetration at each location can be found in Appendix C. For clarity, only a summary of the results is shown here. The summary tables show the attenuation of the peak amplitude of a positive and negative lightning-like pulse attached at the entrance of the mine. This quantity is calculated by multiplying the spectral representation of the current of a positive/negative lightning pulse by the current transfer function measured of a given conductor at a given location (that was measured in the mine). This product is then transformed to the time-domain by an inverse fast Fourier transform (IFFT). The attenuation listed in the tables is then simply the peak output divided by the peak input. The peak current output to peak current input attenuation, for the various conductors at the measured locations, given a positive lightning waveform, are shown in Table 3-2 and Table 3-3 and, given a negative lightning waveform, are shown in Table 3-4 and Table 3-5. Table 3-2 and Table 3-4 show the attenuation with the mine grounding system in its current state, while Table 3-3 and Table 3-5 show the attenuation with the mine grounding system like it was during the explosion. The darkened cells of the tables indicate no measurements were recorded in the given locations. Table 3-2 Summary of current transfer functions, using positive lightning waveform, for conductive penetrations with current mine grounding Power Cable Trolley Comm Line Conveyor Rail Shield Local Fence Local Fence Local Fence Local Fence Location Gnd Gnd Gnd Gnd Gnd Gnd Gnd Gnd 1 1.7x10-3 2.9x10-3 2.2x10-2 2.9x10-2 8.9x10-2 1.4x10-1 2 3 1.8x10-3 2.8x10-3 3.2x10-3 4.9x10-3 3.6x10-4 9.2x10-5 4 1.7x10-3 2.8x10-3 5.6x10-4 1.1x10-4 3.8x10-4 9.4x10-5 5 1.4x10-3 2.2x10-3 7.2x10-4 2.7x10-4 3.9x10-4 3.0x10-4 6 1.2x10-3 1.9x10-3 4.0x10-4 1.1x10-4 5.5x10-4 4.2x10-4 7 1.2x10-3 2.0x10-3 3.0x10-4 9.3x10-5 2.3x10-4 3.5x10-4 Table 3-3 Summary of current transfer functions, using positive lightning waveform, for conductive penetrations with former mine grounding Power Cable Trolley Comm Line Conveyor Rail Shield Local Fence Local Fence Local Fence Local Fence Location Gnd Gnd Gnd Gnd Gnd Gnd Gnd Gnd 1 4.6x10-2 6.2x10-2 2 2.6x10-4 3.7x10-4 1.8x10-2 2.8x10-2 1.3x10-3 1.6x10-3 3 1.4x10-4 1.7x10-4 1.6x10-2 2.5x10-2 1.3x10-3 1.5x10-3 4 5 6 7 Appendix DD - Page 34 of 104 PAGE 34 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Table 3-4 Summary of current transfer functions, using negative lightning waveform, for conductive penetrations with current mine grounding Power Cable Trolley Comm Line Conveyor Rail Shield Local Fence Local Fence Local Fence Local Fence Location Gnd Gnd Gnd Gnd Gnd Gnd Gnd Gnd -3 -3 -2 -2 -2 -1 1 2.4x10 4.3x10 2.2x10 2.9x10 8.4x10 1.4x10 2 3 2.7x10-3 4.7x10-3 3.7x10-3 5.2x10-3 3.2x10-4 1.3x10-4 4 2.4x10-3 4.2x10-3 5.7x10-4 1.4x10-4 3.1x10-4 1.7x10-4 5 2.0x10-3 3.4x10-3 8.1x10-4 3.1x10-4 4.3x10-4 3.4x10-4 6 1.8x10-3 3.2x10-3 4.4x10-4 2.9x10-4 5.3x10-4 5.4x10-4 7 1.7x10-3 3.0x10-3 2.9x10-4 8.7x10-5 2.6x10-4 1.9x10-4 Table 3-5 Summary of current transfer functions, using negative lightning waveform, for conductive penetrations with former mine grounding Power Cable Shield Trolley Comm Line Conveyor Rail Local Fence Local Fence Local Fence Local Fence Location Gnd Gnd Gnd Gnd Gnd Gnd Gnd Gnd 1 4.7x10-2 6.2x10-2 2 2.2x10-3 3.0x10-3 4.0x10-4 6.0x10-4 1.8x10-2 2.7x10-2 -3 -3 3 2.2x10 2.8x10 1.6x10-4 2.2x10-4 1.6x10-2 2.4x10-2 4 5 6 7 Appendix DD - Page 35 of 104 PAGE 35 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 3.2 Indirect Drive 3.2.1 Setup/Equipment Layout with Photos The method used to characterize indirect electromagnetic coupling into the sealed area is shown in Figure 3-4 and Figure 3-5. The current from the audio amplifier (which is driven by the output from the network analyzer) is driven on to a long wire above the ground which is terminated at each end with ground rods. The ground rods are placed so as to produce a current distribution in the ground that simulates a linear current drive. Two configurations were used for the indirect drive measurements. One configuration was through ground rods placed so as to drive the current parallel to the sealed area of the mine and over the area where the explosion occurred as shown in Figure 3-5A. The surface drive wire was approximately 500 m long. A second configuration was through ground rods placed so as to drive the current perpendicular to the sealed area of the mine and over the area where the explosion occurred as shown in Figure 3-5B. In this case the surface drive wire was only 200 m long. Two types of measurements were made to characterize the indirect coupling into the sealed area. The more time consuming of the two was the electric field mapping measurements made in the vicinity of the explosion ignition area, where the core hole is located. The other measurement was the induced voltage on a spliced intact pump cable going from the back of the sealed area to the location of the core hole. The pump cable was spliced with 12-gauge wire to recreate the length of pump cable believed to have been there during the explosion 4. The end of the pump cable at the back of the mine was originally attached to the pump which was submerged underwater and chained to the ceiling mesh. For the measurements, the pump cable was connected with 12-gauge wire to the ceiling mesh and the exposed conductors were placed under water approximately four crosscuts from the back of the sealed area 5. The approximate total length of the recreated cable was 300 m (984 ft). The electric field at various locations in the sealed area of the mine was measured with an active dipole antenna connected to a receiver via fiber optics. The fiber-optic receiver is connected to the network analyzer measurement port so that the signals are phase-locked in order to measure very small signals in the microVolt/meter range. The three polarizations of the electric field were measured at a total of 15 locations for both the parallel and perpendicular wire current drives. The three polarizations measured were the vertical, P-directed (parallel to the length of the sealed area), and X-directed (transverse to the length of the sealed area). A photo of the dipole antenna in horizontal and vertical polarization is shown in Figure 3-7. The exact locations of the measured electric field are shown in Figure 3-6 where the distance between locations was approximately 10 m. The figure shows 17 total locations; however, positions P1 and P9 were not measured due to water hazards. Because of the amount of data taken and the spacing between measurement points, the lack of these two points does not impact the results. The induced voltage measurements were taken on the pump cable with both a parallel and perpendicular surface wire drive. These measurements were also conducted using the instrumentation system shown in Figure 3-4. The induced cable voltage was measured with a Nanofast high-impedance probe in the vicinity of the core hole, and transmitted to the surface with fiber optics. 4 As a note, there is some disagreement as to the length and positioning of the pump cable at the time of the explosion. The test team used information provided at the time of the measurements, which was that the pump cable was intact and the cable shield was grounded to the submerged pump. 5 Test team was unable to reach the back of the sealed area where the pump would have been (it was removed after the explosion) due to water. Appendix DD - Page 36 of 104 PAGE 36 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Earth Overburden Figure 3-4 Indirect Drive Conceptual Drawing. (A.) (B.) Figure 3-5 Parallel (A.) and perpendicular (B.) surface current drive for indirect drive measurements. Appendix DD - Page 37 of 104 PAGE 37 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Figure 3-6 Electric field measurement locations. Figure 3-7 Sandia dipole antenna in horizontal and vertical polarizations inside previously sealed area. Appendix DD - Page 38 of 104 PAGE 38 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 3.2.2 Results The purpose of the electric field mapping of the explosion area was to first look for any field inhomogeneities due to geological features, and second to compare to the analytical model. An added benefit was the ability to verify the induced voltage on the pump cable by integrating the parallel component of the electric field across it. The electric field measurements are shown first and then the induced cable voltage is plotted. The electric field results did show an enhanced electric field at the P5 and the X7 locations (this is noted for general interest,) but do not impact the cable results. The cable integrates or averages the fields over the cable length to build-up a potential difference or voltage. The data collected for the indirect drive tests from the dipole antenna was only usable above 100 Hz. This was due to a very large 60 Hz clutter signal from surrounding power lines and the high noise level from the network analyzer below 40 Hz. Both of these factors were overcome for the long wire voltage measurement by reducing the IF bandwidth of the network analyzer from 10 Hz to 2 Hz. The reduction of the IF bandwidth lowered the noise floor considerably and reduced the sensitivity of the transfer function to the 60 Hz clutter; however, the time for a single swept measurement increased from ~1.5 minutes to ~10 minutes. With the large number of measurements desired for characterizing the electric field in the sealed area, the higher IF bandwidth was used for the majority of the data collected. The overall effect on the data was minimal. As a result, only data from frequencies above 100 Hz are plotted for the dipole measurements in the body of this report. The full spectrum of the data collected can be found in Appendix C. The normalized composite electric fields from the dipole antenna at various locations are plotted in this section. The composite electric field is simply the root-sum-square or amplitude of the electric field vector. The measured electric field is normalized by the current in the drive wire on the surface, so that the units are V/m/A. The normalized electric fields due to the wire current drive parallel to the P-direction, measured at locations P2 through P8 and X1 through X9, are shown in Figure 3-8 and Figure 3-9, respectively. Similarly, the fields due to the wire current drive perpendicular to the P-direction, measured at locations P2 through P8 and X1 through X9, are shown in Figure 3-10 and Figure 3-11, respectively. This information is summarized in Table 3-6. Table 3-6 Summary of figures for drive configurations Drive Configuration P-directed Current Drive (Parallel) X-directed Current Drive (Perpendicular) Electric Field at P locations Electric Field at X locations Figure 3-8 Figure 3-9 Figure 3-10 Figure 3-11 Referring to Figure 3-8, note that the composite electric fields measured in a path parallel to and immediately below the drive are about the same amplitude. The presence of metal objects near the antenna affects the local fields somewhat. The measurement at P5 was made in the area between unconnected sections of roof mesh. The slight resonance at about 60 kHz in the P5 measurement was probably caused by a resonance of the cable that was attached for the voltage measurements. This cable was not removed for the electric field measurements, and high electric fields may have been induced on the disconnected end of the cable at resonance. Referring to Figure 3-9, note that the low-frequency amplitude tended to decrease as the electric field antennas were moved away from the center line immediately below the drive line. Appendix DD - Page 39 of 104 PAGE 39 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Referring to Figure 3-10, the composite electric fields measured in a path perpendicular to the drive cable are reduced significantly from the field due to a parallel drive cable. Again, a slight resonance was seen at P5. Referring to Figure 3-11, the fields measured parallel to and below the perpendicular drive are comparable in amplitude to those shown in Figure 3-8. Because of the closer spacing of the ground rods on the surface, more variation was shown in the individual measurements. Appendix DD - Page 40 of 104 PAGE 40 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Amplitude (V/m/A) 10 10 10 10 -1 P2 P3 PX4 P5 P6 P7 P8 -2 -3 -4 10 2 10 3 10 4 10 5 Frequency (Hz) Figure 3-8 Composite Electric Field along P-direction with parallel line drive on surface. Amplitude (V/m/A) 10 10 10 10 -1 X1 X2 X3 PX4 X5 X6 X7 X8 X9 -2 -3 -4 10 2 10 3 10 4 10 5 Frequency (Hz) Figure 3-9 Composite Electric Field along X-direction with parallel line drive on surface. Appendix DD - Page 41 of 104 PAGE 41 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Amplitude (V/m/A) 10 10 10 10 -1 P2 P3 PX4 P5 P6 P7 P8 -2 -3 -4 10 2 10 3 10 4 10 5 Frequency (Hz) Figure 3-10 Composite Electric Field along P-direction with perpendicular line drive on surface. Amplitude (V/m/A) 10 10 10 10 -1 X1 X2 X3 PX4 X5 X6 X7 X8 X9 -2 -3 -4 10 2 10 3 10 4 10 5 Frequency (Hz) Figure 3-11 Composite Electric Field along X-direction with perpendicular line drive on surface. Appendix DD - Page 42 of 104 PAGE 42 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine The voltage on the pump cable was measured with a high-impedance voltage probe and the network analyzer set to a 2 Hz IF bandwidth. The normalized results of the voltage amplitude plotted relative to the drive current on the surface wire, with units of Volts per Amp (V/A), are shown in Figure 3-12. There is a spike at 60 Hz due to stray signals from power lines. The data is skewed by noise only below 20 Hz. 10 0 Parallel to Drive Wire Perpendicular to Drive Wire Amplitude (V/A) 10 10 10 -1 -2 -3 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 3-12 Induced voltage on pump cable (~300 m or 984 ft. long) due to wire current drives on surface. To compare the induced voltage measured on the pump cable with the field measurements, we will look only at the parallel surface drive induced fields from P2 to P8. Furthermore, we will only look at the horizontal polarization directed along the P-axis, parallel to the direction of the drift. The horizontal polarized electric fields are shown in Figure 3-13. The normalized electric fields are in units of Volts per meter per Amp (V/m/A) while the normalized induced voltage on the pump cable are in units of Volts per Amp (V/A). If we integrate the electric fields over the length of the pump cable, we should obtain the induced voltage from Figure 3-12. Assuming a simple uniform distribution we can simply multiply the electric field by a length. The effective length of cable (similar to the effective area of an antenna) needed to match the electric fields measured with the induced voltage measured was found to be approximately 120 m (394 ft). This means that only the 120 m (394 ft) closest to the measurement end of the cable contribute to the induced voltage. The comparison between the measured induced voltage and the electric field multiplied by the effective length of 120 m (394 ft) is shown in Figure 3-14. Appendix DD - Page 43 of 104 PAGE 43 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Amplitude (V/m/A) 10 10 10 10 -1 P2 P3 PX4 P5 P6 P7 P8 -2 -3 -4 10 2 10 3 10 4 10 5 Frequency (Hz) Figure 3-13 P-directed Electric Field along P-direction with parallel line drive on surface. 10 Amplitude (V/A) 10 10 10 0 -1 P2 P3 PX4 P5 P6 P7 P8 Pump Cable -2 -3 10 2 10 3 10 4 10 5 Frequency (Hz) Figure 3-14 P-directed Electric Fields multiplied by an effective cable length of 120 m (394 ft) compared with the induced voltage on the pump cable. Appendix DD - Page 44 of 104 PAGE 44 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 3.2.3 Results Compared with Diffusion Model The model for diffusion coupling from an infinite current source above a homogeneous half-space was presented in Section 2.2.3. This model is compared with the measured electric field and induced voltage on the pump cable. Using an effective soil resistivity of 80 Ω-m, the analytic model plotted in Figure 3-15 matches very closely the horizontal (P-directed) electric field measured with a parallel current drive. The correlation between model and measured data is extremely good from 10 to 100 kHz. This confirms that the major coupling mechanism from the surface to the sealed area is field diffusion coupling. The measured data is contaminated by 60 Hz resonances and clutter below 1 kHz for this polarization. The data deviates from the model of coupling beneath an infinite line at frequencies below 1 kHz. The measured data stays at a constant level of approximately 0.0006 V/m/A, whereas the analytical model predicts a downward slope. Much of this deviation can be attributed to the field caused by the DC component from the finite spacing of the ground rods. An estimate of this component of the electric field is shown below 1 kHz, where the skin depth is much larger than the depth to the measurement antennas. A comparison of the average of the P-directed electric field measurements from P2 to P8 with the analytic diffusion model is shown in Figure 3-16. The average field is a more meaningful value to compare since it has local variations removed. The amplitude and shape show amazing correlation. Amplitude (V/m/A) 10 10 10 10 -1 P2 P3 PX4 P5 P6 P7 P8 Analytic Model DC Resistivity Term -2 -3 -4 10 2 10 3 10 4 10 5 Frequency (Hz) Figure 3-15 P-directed Electric Fields compared with the diffusion model with an effective resistivity of 80 Ω-m. Appendix DD - Page 45 of 104 PAGE 45 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 10 -2 Amplitude (V/m/A) Average of P2 through P8 Analytic Model DC Resistivity Term 10 10 -3 -4 10 2 10 3 10 4 10 5 Frequency (Hz) Figure 3-16 Average of P-directed fields from P2 to P8 compared with diffusion model. To compare the analytic model with the measured induced cable voltage, we simply integrate over the effective length of cable discussed in the previous section, 120 m (394 ft). Again, the model shows excellent agreement with the measured voltage from 1 to 100 kHz. There is a deviation from the model of coupling beneath an infinite line at frequencies below 1 kHz, where the measured data stays at a constant level of approximately 0.1 V/A. Much of this deviation is caused by the field caused by the DC component from the finite spacing of the ground rods. An estimate of this component of the electric field is shown below 1 kHz where the skin depth is much larger than the depth to the cable. The measured data has been processed to remove the 60 Hz clutter signal and its harmonics. Appendix DD - Page 46 of 104 PAGE 46 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 10 0 Measured Voltage, Parallel Drive Wire Analytic Model with 120 m Wire DC Resistivity Term Amplitude (V/A) 10 10 10 -1 -2 -3 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 3-17 Induced voltage on pump cable due to parallel wire current drive on surface (with 60 Hz and harmonics removed) compared with analytic diffusion model of 120 m (394 ft) long cable and an effective soil resistivity of 80 Ω-m and the DC Resistivity term. Appendix DD - Page 47 of 104 PAGE 47 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 4 Results Coupled with Lightning The results from the direct drive measurements, and the indirect coupling measurements and analysis are coupled with recorded and hypothetical lightning strokes in this section. The analysis performed in this section uses the recorded amplitudes, when appropriate; however, for all other cases, nominal amplitudes of 100 kA were used. The value of 100 kA was used for two reasons: first, there was a cloud-to-ground stroke recorded close in time and distance to the explosion area on the order of 100 kA; and second, the value of 100 kA is easy to scale. It should be noted that the voltages presented in Section 4.2 and 4.3 were calculated using the uniform magnetic field excitation formulation shown in Section 2.2.4. The voltages from a hypothetical long, low altitude horizontal current channel from a cloud-to-ground stroke of Section 4.4 were calculated using infinite line current source above a half-space shown in Section 2.2.2. The basic lightning waveforms used in this section as inputs into the transfer functions are shown in Figure 4-1. The negative lightning waveform was created using a double exponential formula found in [32]. There is no analytic or mathematical model for a positive lightning waveform found in published literature. Hence, a positive lightning waveform was created using a 15th order polynomial of the author's design and appending a 100 ms tail on the backend. The positive lightning waveform characteristics were tailored from values found in [33,20]. Some pertinent waveform characteristics of the modeled lightning waveforms are shown in Table 4-1. 100 +100 kA Stroke -100 kA Stroke 90 80 Amplitude (kA) 70 60 50 40 30 20 10 0 0 50 100 150 200 250 300 Time (μs) 350 400 450 500 Figure 4-1 Basic positive and negative lightning waveforms used as inputs for analysis. Table 4-1 Characteristics of positive and negative lightning waveforms used in analysis Amplitude (kA) -100 +100 +30 Full Width at Half Maximum, FWHM (μs) 68 69 69 Appendix DD - Page 48 of 104 PAGE 48 OF 104 dI/dt (kA/μs) 16.7 6.5 2.0 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 4.1 Direct Drive Transfer Functions Coupled with Lightning Strokes If we assume that lightning directly coupled onto the conductive penetrations into the entrance of the Sago mine with either a positive or negative 100 kA stroke, then peak voltages and currents can be calculated. Only the direct drive transfer functions measured with the fence ground are used in this analysis because the fence ground is more representative of a current distribution due to a real lightning stroke. The peak currents and voltages on the trolley communication line, conveyor, rail, and power cable shield were calculated using the following procedure. First, the lightning waveforms shown in Figure 4-1 were transformed into the frequency domain with a fast Fourier transform (FFT). Then the lightning data was multiplied by the complex transfer function of a given conductor to a given location. The resulting frequency waveform was then transformed back into the time domain with an inverse fast Fourier transform (IFFT). The peak voltage or current was recorded for each waveform. This was then repeated for each conductor at each location measured. Voltage was not measured on the trolley communication line because it was an insulated cable. The measurement locations cross-referenced to break number and approximate distance from the entrance are summarized in Table 4-2 for convenience. Since the transfer function of the shield of the power cable was only measured out to the #3 power center, an extrapolation was performed to estimate the voltage and current at location 7 (at the 2nd Left Switch). The extrapolated values of voltage and current on the shield of the power cable are shown in the green highlighted cells of the “Power Cable Shield” columns of Table 4-3 and Table 4-5. The voltage was extrapolated using an exponential curve fit, while the current was extrapolated using a simple logarithmic curve fit. These extrapolations were matched with the trend of the first three points, and are a best-guess speculation. The peak current and voltage from a positive 100 kA lightning stroke attached directly to the entrance of the mine for each conductor at each location are shown in Table 4-3 and Table 4-4. The peak currents and voltages due to a negative 100 kA stroke are shown in Table 4-5 and Table 4-6. Table 4-2 Direct Drive Measurement Locations Location Mine Feature Break Number 1 2 3 4 5 6 7 #1 Power Center #2 Power Center #3 Power Center st 1 Right Spur nd 2 Right Spur st 1 Left Switch nd 2 Left Switch Belt 1, Break 1 Belt 2, Break 1 Belt 3, Break 1 Belt 3, Break 16 Belt 4, Break 11 Belt 4, Break 50 Belt 4, Break 59 Approximate Distance from Entrance 30 m (98 ft.) 459 m (1506 ft.) 669 m (2195 ft.) 1076 m (3530 ft.) 2178 m (1.35 miles) 3255 m (2.02 miles) 3491 m (2.17 miles) Table 4-3 Peak currents and voltages from a positive 100 kA lightning stroke, for conductive penetrations with old mine grounding Trolley Power Power Conveyor Conveyor Rail Rail Comm Cable Cable Line Shield Shield Location Imax (A) Imax (A) Vmax (V) Imax (A) Vmax (V) Imax (A) Vmax (V) 1 6213 8369 2 162 37 643 2841 3229 3 154 17 233 2547 1582 4 5 6 7 480* 1* * Extrapolated values. Appendix DD - Page 49 of 104 PAGE 49 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Table 4-4 Peak currents and voltages from a positive 100 kA lightning stroke, for conductive penetrations with current mine grounding Trolley Power Power Conveyor Conveyor Rail Rail Comm Cable Cable Line Shield Shield Location Imax (A) Imax (A) Vmax (V) Imax (A) Vmax (V) Imax (A) Vmax (V) 1 293 2884 10931 14087 136693 2 3 279 495 881 9 996 4 279 11 62 9 436 5 220 27 11 30 1079 6 190 11 2 42 321 7 198 9 1 35 106 Table 4-5 Peak currents and voltages from a negative 100 kA lightning stroke, for conductive penetrations with old mine grounding Trolley Power Power Comm Conveyor Conveyor Rail Rail Cable Cable Line Shield Shield Location Imax (A) Imax (A) Vmax (V) Imax (A) Vmax (V) Imax (A) Vmax (V) 1 6193 7989 2 295 60 668 2711 3078 3 279 22 218 2417 1438 4 5 6 7 280* 1* * Extrapolated values. Table 4-6 Peak currents and voltages from a negative 100 kA lightning stroke, for conductive penetrations with current mine grounding Trolley Power Power Comm Conveyor Conveyor Rail Rail Cable Cable Line Shield Shield Location Imax (A) Imax (A) Vmax (V) Imax (A) Vmax (V) Imax (A) Vmax (V) 1 434 2926 11279 13606 143340 2 3 467 515 1052 13 1615 4 417 14 77 17 934 5 343 31 13 34 1367 6 320 29 3 54 650 7 301 9 1 19 62 An item of interest is the relatively high current on the shield of the power cable (480 A) at the 2nd left switch (location 7) in Table 4-3. The power cable does not stop at the 2nd left switch, but turns approximately 90 degrees to the left and travels down the 2nd Left Main and onto the 2 Left Power Center. This presents a similar coupling mechanism as the indirect case where a long line current drive on the surface produces electromagnetic fields that propagate through earth. Only in this case, instead of the lightning currents on the surface being the drive, the induced current on the shield of the power cable inside the mine provides the drive mechanism for coupling onto the pump cable. Assuming a direct 100 Appendix DD - Page 50 of 104 PAGE 50 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine kA positive stroke onto the shield of the power cable at the entrance to the mine, an analysis of this scenario results in <50 V peak induced on the pump cable, too low to be of concern. Appendix DD - Page 51 of 104 PAGE 51 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 4.2 Indirect Drive from NLDN and USPLN Positive Stroke 1-3 The locations of the three recorded lightning strokes, on the NLDN and USPLN, are shown in Figure 4-2 with the calculated distances and angles. Note that it is highly probable that the 38.8 kA and 35 kA strokes represent a single stroke with a location discrepancy, as discussed in Section 1.6. The angles shown are the angles between the line made up from the lightning stroke to the center of the pump cable, and the line formed by the direction where the pump cable lay. The first stroke analyzed is the 38.8 kA positive lightning stroke, 5.44 km (3.4 miles) away from the sealed area and an angle of 52.8 degrees. The second stroke has an amplitude of 35 kA at a distance of 4.02 km (2.5 miles) away from the sealed area and an angle of 49.3 degrees. The last stroke has an amplitude of 101 kA, a distance of 2.91 km (1.8 miles), and an angle of 85.5 degrees. The resulting induced voltage pulses on the pump cable (at the end of the cable nearest the explosion area) are shown in Figure 4-3, with peak amplitudes of 25.7 V, 33.8 V, and 16.2 V for the three strokes, respectively. The effective length of 120 m (394 ft.) was used for the length of the pump cable in Figure 4-3. Since there is concern about the actual length of intact pump cable present at the time of the explosion, analysis was performed on a pump cable with a length of 61 m (200 ft.) to account for the length of the cable piece found closest to the explosion area. The resulting induced voltage pulses on the 61 m (200 ft.) length of pump cable are shown in Figure 4-4. None of the induced voltages from these recorded strokes have the necessary amplitude to cause an arc inside the sealed area. It should be noted that taking the indirect coupling model approximation out to 3 km and beyond represents an upper bound on the coupling. Figure 4-2 Locations of recorded lightning strokes with respect to the sealed area, with distances and angles. Appendix DD - Page 52 of 104 PAGE 52 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 40 Stroke #1: 38.8 kA Stroke #2: 35 kA Stroke #3: 101 kA 35 30 Amplitude (V) 25 20 15 10 5 0 -5 0 100 200 300 400 500 Time (μs) Figure 4-3 Voltage induced on pump cable (using an effective length of 120 m or 394 ft.) due to the three positive lightning strokes recorded on the NLDN and USPLN. 20 Stroke #1: 38.8 kA Stroke #2: 35 kA Stroke #3: 101 kA Amplitude (V) 15 10 5 0 -5 0 50 100 150 200 250 300 Time (μs) 350 400 450 500 Figure 4-4 Voltage induced on pump cable (length of 61 m or 200 ft.) due to the three positive lightning strokes recorded on the NLDN and USPLN. Appendix DD - Page 53 of 104 PAGE 53 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 4.3 Indirect Drive from Hypothetical Stroke Directly over Sealed Area If we assume a 100 kA negative or positive lightning stroke attached within 100 m (328 ft.) from directly over the center of the pump cable in the sealed area on the surface, it could induce a sufficiently high voltage in conductors in the sealed area to cause an electrical arc. This effect would be maximized if the stroke were directly inline with the pump cable direction at an angle of zero degrees. The induced voltage on the pump cable (with an effective length of 120 m or 394 ft.) from a 100 kA positive and negative cloud-to-ground stroke is shown in Figure 4-5. The maximum voltages are 23.8 kV from the positive pulse and 22.3 kV from the negative lightning pulse. Again, since there is concern about the actual length of intact pump cable present at the time of the explosion, analysis was performed on a pump cable with a length of 61 m (200 ft.) to account for the cable piece found closest to the explosion area. The resulting induced voltage pulses on the 61 m (200 ft.) length of pump cable are shown in Figure 4-6. The maximum voltages expected on the shorter cable length are 20.5 kV from the positive pulse and 19.1 kV from the negative lightning pulse. Lightning currents as low as 20 kA (either positive or negative), which is closer to the statistical average peak current of cloud-to-ground lightning strokes, can produce thousands of Volts on the pump cable. This level of voltage is more than capable of initiating an electrical arc under the right conditions. The peak voltage amplitude expected on the pump cable scales linearly with the peak current amplitude of the driving lightning stroke. The results from the 100 kA case shown in Figure 4-5 can be scaled to the 20 kA case by dividing the peak amplitude of the voltage on the cable by a factor of five. 30 +100 kA Stroke -100 kA Stroke 25 Amplitude (kV) 20 15 10 5 0 -5 0 100 200 300 400 500 Time (μs) Figure 4-5 Induced voltage pulse on pump cable (using an effective length of 120 m or 394 ft.) due to a hypothetical positive and negative 100 kA cloud-to-ground lightning stroke 100 m from directly above sealed area. Appendix DD - Page 54 of 104 PAGE 54 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 30 +100 kA Stroke -100 kA Stroke 25 Amplitude (kV) 20 15 10 5 0 -5 0 50 100 150 200 250 300 Time (μs) 350 400 450 500 Figure 4-6 Induced voltage pulse on pump cable (length of 61 m or 200 ft.) due to a hypothetical positive and negative 100 kA cloud-to-ground lightning stroke 100 m from directly above sealed area. Appendix DD - Page 55 of 104 PAGE 55 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 4.4 Indirect Drive from a Hypothetical Cloud-to-Ground Stroke with a Current Channel over Sealed Area If we assume a 100 kA positive cloud-to-ground stroke with a long, low altitude horizontal current channel directly over the sealed area and inline with the pump cable direction at an angle of zero degrees, it could be capable of inducing voltages on the pump cable sufficient to produce electrical arcing. Pump cable (with an effective length of 120 m or 394 ft.) voltages are shown for a positive cloud-to-ground stroke with horizontal current channel at heights (H) of 0, 100 m (328 ft.), 200 m (656 ft.), 500 m (1640 ft.), and 1000 m (3281 ft.) above the surface in Figure 4-7. The maximum voltages from the positive current channel at the heights given are 15.3 kV, 7.2 kV, 4.6 kV, 2.1 kV, and 1.1 kV, respectively. Induced voltages for a negative cloud-to-ground stroke with a current channel directly over the sealed area are shown in Figure 4-9. The maximum voltages from the negative current channel at the heights given are 14.3 kV, 6.7 kV, 4.3 kV, 2 kV, and 1.1 kV, respectively. Again, since there is concern about the actual length of intact pump cable present at the time of the explosion, analysis was performed on a pump cable with a length of 61 m (200 ft.) to account for the cable piece found closest to the explosion area. The resulting induced voltage pulses on the 61 m (200 ft.) length of pump cable are shown for a positive cloud-to-ground stroke with horizontal current channel at heights (H) of 0, 100 m (328 ft.), 200 m (656 ft.), 500 m (1640 ft.), and 1000 m (3281 ft.) above the surface in Figure 4-8. The maximum voltages from the positive current channel at the heights given are 7.8 kV, 3.7 kV, 2.3 kV, 1.1 kV, and 0.6 kV, respectively. Induced voltages for a negative cloud-toground stroke with a current channel directly over the sealed area are shown in Figure 4-10. The maximum voltages from the negative current channel at the heights given are 7.3 kV, 3.4 kV, 2.2 kV, 1 kV, and 0.5 kV, respectively. 15 H= 0 m, V H = 100 m, V H = 200 m, V H = 500 m, V 10 =15.3 kV max =7.2 kV max =4.6 kV max H = 1000 m, V Amplitude (kV) =2.1 kV max =1.1 kV max 5 0 -5 0 50 100 150 200 250 300 Time (μs) 350 400 450 500 Figure 4-7 Induced Voltage Pulse on Pump Cable (with an effective length of 120 m or 394 ft.) from Hypothetical Horizontal Current Channel from a Cloud-to-Ground +100 kA Stroke, H is distance of the Current Channel above the Ground. Appendix DD - Page 56 of 104 PAGE 56 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 10 H= 0 m, Vmax=7.8 kV H = 100 m, Vmax=3.7 kV H = 200 m, Vmax=2.3 kV H = 500 m, Vmax=1.1 kV H = 1000 m, Vmax=0.6 kV Amplitude (kV) 5 0 -5 0 50 100 150 200 250 300 Time (μs) 350 400 450 500 Figure 4-8 Induced Voltage Pulse on Pump Cable (length of 61 m or 200 ft.) from Hypothetical Horizontal Current Channel from a Cloud-to-Ground +100 kA Stroke, H is distance of the Current Channel above the Ground. 15 H= 0 m, V H = 100 m, V H = 200 m, V H = 500 m, V 10 =14.4 kV max =6.7 kV max =4.3 kV max H = 1000 m, V Amplitude (kV) =2 kV max =1.1 kV max 5 0 -5 0 50 100 150 200 250 300 Time (μs) 350 400 450 500 Figure 4-9 Induced Voltage Pulse on Pump Cable (with an effective length of 120 m or 394 ft.) from Hypothetical Horizontal Current Channel from a Cloud-to-Ground -100 kA Stroke, H is distance of the Current Channel above the Ground. Appendix DD - Page 57 of 104 PAGE 57 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 10 H= 0 m, Vmax=7.3 kV H = 100 m, Vmax=3.4 kV H = 200 m, Vmax=2.2 kV H = 500 m, Vmax=1 kV H = 1000 m, Vmax=0.5 kV Amplitude (kV) 5 0 -5 0 50 100 150 200 250 300 Time (μs) 350 400 450 500 Figure 4-10 Induced Voltage Pulse on Pump Cable (length of 61 m or 200 ft.) from Hypothetical Horizontal Current Channel from a Cloud-to-Ground -100 kA Stroke, H is distance of the Current Channel above the Ground. Appendix DD - Page 58 of 104 PAGE 58 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 5 Conclusions The conclusions made in this report are specific to the geometry of the Sago mine site where measurements were taken. The results cannot and should not be generalized to any other mining systems. 5.1 Direct Coupling The current and voltage on metallic penetrations into the mine were calculated given the direct drive transfer functions and a mathematical representation of a positive-polarity, 100 kA peak cloud-to-ground lightning stroke. This calculation assumes that the lightning stroke attaches directly onto the metallic penetration at the entrance to the mine. While there is no evidence that lightning struck the entrance of the mine, this assumption represents the worst-case placement of an attachment for this analysis. The farthest point into the mine that the direct drive measurements were made was at the entrance to the 2nd Left Parallel, 3,491 m (or 2.17 miles) into the mine, as close to the seal that was breached by the explosion as possible. At this location, the peak currents and voltages calculated at this location given the input of a positive 100 kA peak lightning stroke attaching at the mine entrance are shown in Table 5-1. The voltage was not measured for the trolley communication line because it was insulated and not an exposed conductor. Table 5-1 Current and voltage at the 2nd Left Switch due to a 100 kA peak, positive cloud-to-ground lightning stroke at the entrance of the mine Metallic penetration Trolley Communication line Conveyor Structure Rail Shield of Power Cable 6 Current 198 A 9A 35 A 480 A Voltage Not measured 1V 106 V 1V The voltages and currents on the conveyor, rail, and shield of the power cable outside the sealed area are incapable of coupling sufficient energy into the sealed area to cause an electrical arc in the sealed area. The voltage on the trolley communication line is not anticipated to be significantly larger than those of the conveyor, rail, and power cable shield. • It is highly unlikely that direct drive coupling, even under a worst-case scenario, could have initiated electrical arcing on the cable in the sealed area. Because of the substantial initial grounding of metallic penetrations that enter the mine, and because of the multiplicity of grounding points of these systems as they penetrate into the mine, the lightning current is divided sufficiently so that only a relatively small amount of current is injected into the mine near the sealed area. All metallic penetrations were intentionally terminated outside the sealed area. Consequently, the amplitude of current flowing on conductors outside the sealed area is insufficient to generate adequate voltage on the cable inside the sealed area to cause arcing. At low frequencies, the parallel nature of the multiplicity of grounding points is sufficient to divide the lightning current. At higher frequencies, the metallic penetrations can be treated as non-ideal (lossy) transmission lines with periodic grounding that attenuates the high-frequency components of the current even more than lower frequencies. Although this coupling mechanism is likely insufficient to cause arcing, the voltage and 6 The current and voltage for the shield of the power cable were extrapolated from measurements made at the Power Centers 1, 2, and 3. Appendix DD - Page 59 of 104 PAGE 59 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine current is sufficient to cause electrical shocks to personnel contacting these metallic penetrations, even miles back into the mine. 5.2 Indirect Coupling Three things are needed to conclude that indirect coupling of lightning energy into the sealed area produced high voltage and an electrical arc that could have been the initiation source of a methane-air explosion in the sealed area of the Sago mine on the morning of January 2, 2006. They are: • • • lightning energy propagating from the surface through the overburden into the sealed area; an antenna, or receiver (such as a cable), of this energy present in the sealed area; and lightning of sufficient magnitude and proximity to the sealed area at the time of the explosion. The indirect measurements coupled with analytical models discussed in this report confirm that electromagnetic energy with the frequency content of lightning driven on the surface penetrates the ground into the sealed area. Measurements and analyses also confirm that the pump cable acts as a receiver of this energy and is the most likely coupling agent in the sealed area. Two cloud-to-ground lightning strokes were recorded in the vicinity of the Sago mine within one second of the explosion in the sealed area. Based on the results in this report, these lightning strokes were too far away to have generated enough voltage on the pump cable to create an electrical arc in the sealed area. A thorough, expert analysis of the raw data provided by several lightning detection databases did not uncover evidence to support the detection of another cloud-to-ground stroke in the correct timeframe. • It is unlikely that indirect drive from the vertical components of the recorded lightning strokes (recorded amplitude and location) around the Sago mine could have initiated electrical arcing on the cable present in the sealed area. The simultaneous events of recorded lightning strokes and the explosion in the sealed area of the mine; the multiple personal accounts above the sealed area describing simultaneous flash and thunder [21] (indicating extremely close lightning); the lack of data from the lightning detection networks from upward positive lightning initiated from tall structures [20, 35]; the inability of the lightning detection networks to resolve the presence of horizontal lightning arc channels [20, 35]; and the unlikely, but possible, scenario of an undetected cloud-to-ground lightning flash [34] of sufficient magnitude and proximity to the sealed area at the time of the explosion led to the investigation of various hypothetical lightning stroke events. The expected voltage on the abandoned cable was calculated for each scenario using the indirect coupling models developed in this report. The first hypothetical case explores the possibility of the presence of a horizontal lightning arc channel acting as a source of energy. For this scenario, a 100 kA-peak horizontal arc channel is assumed to be parallel to the pump cable in the sealed area at distances of 100 m (328 ft), 200 m (656 ft), 500 m (1,640 ft), and 1000 m (3,281 ft) above the ground above the sealed area. For a positive-polarity flash, the resultant voltages on the pump cable were 7.2 kV, 4.6 kV, 2.1 kV, and 1.1 kV, respectively. For a negative-polarity flash, the resultant voltages on the pump cable were 6.7 kV, 4.3 kV, 2 kV, and 1.1 kV, respectively. While these calculations use favorable coupling circumstances (high peak arc-channel current and parallel orientation of the arc channel to the pump cable and 120 m cable effective length), this hypothetical scenario presents a reasonable case for high-voltage electrical arcing. Appendix DD - Page 60 of 104 PAGE 60 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine • It is reasonable to assume that if a horizontal, low-altitude arc channel occurred from one of the lightning strokes recorded by the NLDN (or USPLN) or from an unrecorded lightning stroke, it could have initiated electrical arcing on the cable in the sealed area. The second hypothetical case explores the possibility of an undetected cloud-to-ground stroke of sufficient magnitude and proximity to the sealed area. Applying a 100 kA-peak, cloud-to-ground stroke of optimum orientation to the pump cable (61 m length) within 100 m (328 ft) of the sealed area, the results are peak voltages on the pump cable of 19.1 kV for a negative-polarity flash, and 20.5 kV for a positive-polarity flash. For the same conditions, the induced voltage decreases as distance of a lightning stroke from the sealed area increases. • It is reasonable to assume that if an average or above average cloud-to-ground lightning stroke occurred above the sealed area at Sago, that it could initiate electrical arcing on the cable in the sealed area. Recent discussions led to a third hypothetical case, which is not examined in detail in the report, of upward-going positive lightning initiating from tall structures. Four tall communication towers (heights of approximately 200 ft or less) are within approximately 1 mile of the sealed area, the closest being about 0.5 miles. If we hypothesize an upward-going positive lightning stroke from the closest tower, (recalling that these type of events are not typically captured by the current lightning detection networks), the induced voltage on the pump cable would be 763 V. The conclusions of this report are that lightning of sufficient magnitude and proximity to the sealed area would create high voltage on the pump cable to create an electrical arc. The simultaneity in time of recorded lightning strokes and the explosion occurring is very strong evidence of cause and effect. Furthermore, eyewitness accounts of simultaneous lightning and thunder at the time of the explosion, plus the analysis of credible hypothetical scenarios which cannot be confirmed by lightning detection networks, lend credibility to the idea that lightning-induced electrical arcing was not only plausible, but highly likely. Appendix DD - Page 61 of 104 PAGE 61 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 6 Recommendations The results of this short-term project demonstrate the usefulness of transfer function measurement techniques and analytical modeling to evaluate lightning effects in mining environments. The effects described in this report are significant. A more comprehensive research and development program should be conducted to expand on this work to extend this research for use in other underground coal mining operations. The research program would be conducted using similar transfer function measurement techniques, experiments at other sites with rocket-triggered and natural lightning, and analytical and computational modeling using validated state-of-the-art codes adapted for this application. Once completed, it is reasonable to expect that mitigation techniques and safety standards could be developed to secure coal mining systems from future lightning threats. Appendix DD - Page 62 of 104 PAGE 62 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 7 References 1. Checca, Elio and D. R. Zuchelli, Lightning Strikes and Mine Explosions, Proceedings of 7th US Mine Ventilation Symposium, June 5-7, 1995, pp 245-250. 2. Checca, Elio L., Investigative Report No. C-042094–Oak Grove Mine Methane Gas Ignition, U. S. Department of Labor, Mine Safety and Health Administration, Pittsburg Safety and Health Technology Center, Pittsburg, PA, April 6-12, 1994. 3. Scott, Doniece S., E. Larry Checca, Clete R. Stephan, and Mark J. Schultz, Accident Investigation Report (Underground Coal Mine) Non-Injury Methane Explosion, Oak Grove Mine (I.D. No. 01-00851), U. S. Department of Labor, Mine Safety and Health Administration, District 11, January 29, 1996. 4. Scott, Doniece S., and Clete R. Stephan, Accident Investigation Report (Underground Coal Mine) Non-Injury Methane Explosion, Oak Grove Mine (I.D. No. 01-00851), U. S. Department of Labor, Mine Safety and Health Administration, District 11, July 11, 1997. 5. Morris, Marvin E., Richard J. Fisher, George H. Schnetzer, Kimball O. Merewether, and Roy E. Jorgenson, Rocket-Triggered Lightning Studies for the Protection of Critical Assets, M. E. Morris et al., IEEE Transactions on Industry Applications, Vol. 30, No. 3, pp 791-804, May/June 1994 (1994 Prize Paper Award from IEEE Power Systems Society). 6. Chen, Kenneth C., Kimball O. Merewether, Tom Y. T. Lin, andParris Holmes, Jr., Final Report: U12g Tunnel Lightning Evaluation, Sandia National Laboratories Report, SAND2004-1619, Sandia Nationa Laboratories, Albuquerque, NM, April 2004. 7. Dinallo, Michael A., and Roy E. Jorgenson, Recommended Lightning Protection Practices for Operations Being Conducted in G-Tunnel at the Nevada Test Site, Sandia National Laboratories Report SAND2006-1049P, Sandia National Laboratories, Albuquerque, NM, February 2006. 8. Novak, Thomas, and Thomas J. Fisher, Lightning Propagation Through the Earth and Its Potential for Methane Ignitions in Abandoned Areas of Underground Coal Mines, IEEE Transactions on Industrial Applications, Vol. 37, No. 6, Nov/Dec 2001, pp1555-1562. 9. Sacks, H. K., and Thomas Novak, Corona Discharge Initiated Mine Explosion, IEEE Transactions on Industrial Applications, Vol. 41, Sept/Oct 2005. 10. Liu, Jian-Bang, Paul D. Ronney, and Martin A. Gundersen, Premixed Flame Ignition by Transient Plasma Discharges, 11. Ronney, Paul D., Technical Progress Report on Corona Discharge Initiation, University of Southern Californa, Dept. of Aerospace and Mechanical Engineering, Los Angeles, CA, Sept 12, 2003. 12. Berger, K., Protection of Underground Blasting Operations, edited by R. H. Golde, in. Lightning – Vol 2, Lightning Protection, Academic Press, New York, NY, 1977, pp633-658. 13. Geldenhuys, H. J., A. J. Eriksson, W. B. Jackson, and J. B. Raath, Research into LightningRelated Incidents in Shallow South African Coal Mines, Proceedings of the 21st Internal Conference on Safety in Mines Research, 1985, pp775-782. 14. Golledge, P., Sources and Facility of Ignition in Coal Mines, in Ignitions, Explosions, and Fires, 1981, pp2-1–2-12. 15. Geldenhuys, H. J., The Measurement of Underground Lightning-Induced Surges in a Colliery, Symposium on Safety in Coal Mining, South Africa National Electrical Engineering Research, Pretoria, South Africa, October 5-8, 1987.. 16. Zeh, K. A., Lightning and Safety in Shallow Coal Mines, 23rd International Conference of Safety in Mines, 1989, pp691-700. 17. Staff-Mining Research, Methane Control in Eastern U. S. Coal Mines, Proceedings of the Symposium of the Bureau of Mines/Industry Technology Transfer Seminar, Morgantown, WV, May 30-31, 1973. Appendix DD - Page 63 of 104 PAGE 63 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 18. Insulating materials for design and engineering practice, John Wiley and Sons, 1962, Library of Congress Catalog Card Number 62-17460. 19. Rucker, Dale, Marc Levitt, Shawn Calendine, John Fleming, and Robert McGill, Geophysical Survey for the Old 2 Left Section of the Sago Mine, Buckhannon, WV, hydroGEOPHYSICS, Inc., Tucson, AZ, August 18, 2006. 20. Martin A. Uman, University of Florida (private communication). 21. West Virginia Office of Miners' Health, Safety, and Training, Report of Investigation into the Sago Mine Explosion which occurred January 2, 2006, Upshur Co. West Virginia, December 11, 2006. 22. Fisher, R. J., G. H. Schnetzer and M. E. Morris, Measured Fields and Earth Potentials at 10 And 20 Meters from the Base of Triggered Lightning Channels, 22nd International Conference on Lightning Protection, Budapest, Hungary, September 19-23, 1994. 23. Schoene, J., M.A. Uman, V.A. Rakov, V. Kodali, K.J. Rambo, G.H. Schnetzer, Statistical Characteristics of the Electric and Magnetic Fields and Their Time Derivatives 15 m and 30 m from Triggered Lightning, J. Geophys, Res, Vol. 108, No. D6, 4192, doi:10.1029/2002JD002698, 2003. 24. King, Ronold W. P., Transmission-line Theory, Dover, New York, NY, 1965. 25. Warne, Larry K., and Kenneth C. Chen, Long Line Coupling Models, SAND2004-0872, Sandia National Laboratories Report, Sandia National Laboratories, Albuquerque, NM, March 2004. 26. Smythe, William R., Static and Dynamic Electricity, A Summa Book, Albuquerque, NM, 1989. 27. Wait, James R., Electromagnetic Waves in Stratified Media, The Macmillan Company, New York, NY, 1962 28. Tegopoulis, J. A., and E. E. Kriezis, Eddy Currents in Linear Conducting Media, Elsevier, New York, NY, 1985. 29. Stoll, Richard L., The Analysis of Eddy Currents, Clarendon Press, Oxford, UK, 1974. 30. Krawczyk, A., and J. A. Tegopoulis, Numerical Modeling of Eddy Currents, Clarendon Press, Oxford, UK, 1993. 31. Kaufman, A. A., and P. Hoekstra, Electromagnetic Soundings, Elsevier, New York, NY, 2001. 32. Cianos, N., and Pierce, E. T., A Ground-Lightning Environment for Engineering Usage, Technical Report 1, SRI Project 1834, August 1972. 33. Rakov, Vladimir A., and Martin A. Uman, Lightning, Lightning Physics and Effects, Cambridge University Press, New York, NY, 2003. 34. Cummins, Kenneth L., et. al., The U.S. National Lightning Detection Network: PostUpgrade Status, Proceedings of the Second Conference of Meteorological Applications of Lightning Data, 86th AMS Annual Meeting, Atlanta, GA, 29 January - 2 February 2006. American Meteorological Society. 35. E. Philip Krider, University of Arizona (private communication and memorandum, see Appendix E). 36. Phillips, Robert, Resistivity Measurements, personal communications from Robert Phillips Appendix DD - Page 64 of 104 PAGE 64 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 8 Appendix A — Analytical and Numerical Models for Voltage and Current Used to Determine Electromagnetic Coupling into the Sago Mine Marvin E. Morris Consultant, Sandia National Laboratories, Department 1652 February 11, 2007 Abstract The purpose of this appendix is to document the relevant analytical models to be used to predict the voltages and currents produced in the Sago mine by current drive sources used to simulate the effects of a lightning stroke attachment near the mine or on the surface of the earth above the mine. Also considered are horizontal arcs above the surface of the mine. 8.1 Introduction The purpose of this appendix is to document the relevant analytical models to be used to predict the voltages and currents produced in the Sago mine by current drive sources used to simulate the effects of a lightning stroke attachment near the mine or on the surface of the earth above the mine. Subsequent measurements corresponding to these models will be used to identify coupling paths and quantify coupling amplitudes of the lightning energy into the sealed area of the mine where the explosion was thought to have been initiated. The initial section of the appendix documents the DC drive current models for both a homogeneous half-space and for a two layer half-space. The next section of the appendix documents the eddy current models for an infinite length horizontal drive wire over both a homogeneous half-space and a two layer half-space. The next section documents the eddy current coupling into a homogeneous half-space from a uniform magnetic field at the surface. The final section of the appendix references the literature for eddy current models for an infinitesimal length and a finite length horizontal wire over both a homogeneous half-space and a two layer half-space. Computer codes have been implemented in Fortran and Mathematica to calculate the resulting potentials and fields and the resulting voltages generated within the earth. Appendix DD - Page 65 of 104 PAGE 65 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine z y region - 0 ε0, μ0, τ0=∞ I -I (-b,0,0) region - 1 ε1, μ0, τ1 (b,0,0) (b,0,0) r1 x r2 (x,y,z) -z Figure 8-1 DC Current Drive with Homogeneous Half-Space Geometry. 8.2 Static Current Drive Models The simplest model for current coupling into a conductive earth is the DC conduction of current into a conductive half-space. The models for this are well known. 8.2.1 Homogeneous Half-Space The DC or very low-frequency situation to be modeled is shown in Figure 8-1. Current I is driven into the conductive half-space at Cartesian coordinate (-b, 0, 0) and the current is removed at Cartesian coordinate (b, 0, 0). The upper half-space, region-0, has infinite resistivity τ0 and the lower half-space, region-1, has resistivity τ1. From simple considerations, V (x, y, z), the potential at Cartesian coordinate (x, y, z) with respect to infinity, is given by ⎛ ⎞ ⎟ − 2 2 2 2 2 2 ⎟ ( x − b ) + y + z ⎟⎠ ⎝ ( x + b) + y + z τI V ( x, y , z ) = 1 ⎜ 2π ⎜⎜ 1 1 The difference in potential between two points can be calculated by taking the difference of the potentials at the two points calculated with the above formula. Appendix DD - Page 66 of 104 PAGE 66 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine The electric field at point (x, y, z) is easily calculated from E ( x, y, z ) = −∇V ( x, y, z ) and calculating the x-component of interest ⎛ ( x + b) ( x − b) τ1I ⎜ ∂ ⎜ − E x ( x , y , z ) = − V ( x, y , z ) = 3 3 ∂x 2π ⎜ 2 2 2 2 ⎤2 2 2 ⎤2 ⎡ ⎡ x − b) + y + z ⎜ ( x + b) + y + z ⎦ ⎣( ⎦ ⎝⎣ ⎞ ⎟ ⎟ ⎟ ⎟ ⎠ 8.2.2 Two Layer Half-Space Because there is often a less resistive layer of topsoil above the more resistive layer, which includes the mine, it is necessary to generalize the above homogenous half-space model to a two layer half-space model. The DC or very low-frequency situation to be modeled is shown in Figure 8-2. Current I is driven into the conductive half-space at Cartesian coordinate (-b, 0, 0) and the current is removed at Cartesian coordinate (b, 0, 0). The upper half-space, region-0 has infinite resistivity τ0 and region-1, the layer of thickness, a, has resistivity τ1. The infinitely thick layer region-2 has resistivity τ2. From more complicated considerations, V (x, y, z), the potential at Cartesian coordinate (x, y, z) with respect to infinity, is given by I V ( x, y , z ) = 2π ⎛ ⎛ 2τ 1τ 2 ⎞ ⎜ ⎜ ⎟ ⎝ τ 1 + τ 2 ⎠ ⎜⎜ ⎝ ⎞ ⎟ − 2 2 2 2 2 2 ⎟ ( x + b) + y + z ( x − b ) + y + z ⎟⎠ 1 1 in region-2. The difference in potential between two points can be calculated by taking the difference of the potentials at the two points calculated with the above formula. The electric field at point (x, y, z)is easily calculated from E ( x, y, z ) = −∇V ( x, y, z ) Appendix DD - Page 67 of 104 PAGE 67 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Figure 8-2 DC Current Drive with Two Layer Half-Space Geometry. and calculating the x-component of interest ∂ V ( x, y , z ) ∂x ⎛ ( x + b) ( x − b) I ⎛ 2τ 1τ 2 ⎞ ⎜ = − ⎜ ⎟⎜ 3 3 2π ⎝ τ 1 + τ 2 ⎠ ⎜ 2 2 2 2 ⎤2 2 2 ⎤2 ⎡ ⎡ x − b) + y + z ⎜ ( x + b) + y + z ⎦ ⎣( ⎦ ⎝⎣ E x ( x, y , z ) = − ⎞ ⎟ ⎟ ⎟ ⎟ ⎠ 8.3 Eddy Current, Infinite Horizontal Drive Wire Models The next obvious generalization of the above model is to make the current injected into the earth time varying, say I = I 0 eiωt xˆ and to neglect displacement current. This generalization turns out to be more difficult than one might think because the current in the earth depends on the geometry of the current path above the earth. A simpler model that corresponds the electromagnetic coupling below a long, horizontal wire grounded at both ends and driven by a voltage source can, however, be developed. Appendix DD - Page 68 of 104 PAGE 68 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 8.3.1 Homogeneous Half-Space The current drive geometry of an infinitely long, horizontal wire placed a distance, h, above a conductive half-space is shown on the left side of Figure 8-3. A side view is shown on the right side of Figure 8-3. The current drive is harmonically time-varying and directed along the x - axis at height, h, above it. The upper half-space has permittivity ε0and infinite resistivity and the lower half-space has permittivity ε1 and resistivity, τ1. Both regions have free space permeability, μ0. If one neglects displacement current and relates current density, ix(y, z), and electric field, Ex(y, z), in region-1 through, Ex ( y, z ) = τ 1ix ( y, z ) , then the current density in the lower half-space, region-1, can be determined to be ix ( y, z ) = − ∞ iωμ0 I ∞ e qz e − uh i 2 I e qz e− uh = cos cos uydu uydu πτ 1 ∫0 u + q π δ12 ∫0 u + q Ex ( y, z ) = τ 1ix ( y, z ) = ikς 0 ∞ e qz e − uh cos uydu π ∫0 u + q where z z y −∞ I region – 0 ε0, μ0, τ0=∞ I ∞ h h x x region – 0 = ε0, μ0, τ0∞ region – 1 ε1, μ0, τ1 region – 1 ε1, μ0, τ1 (y,z) (y,z) -z Figure 8-3 Infinite Horizontal Current Drive, Eddy Current Coupling Geometry. k = ω μ 0ε 0 q = u 2 + ip 2 p2 = δ1 = ωμ0 2 = 2 δ1 τ1 2τ 1 ωμ 0 This or similar expressions are given in [A1-A3]. Appendix DD - Page 69 of 104 PAGE 69 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Note that the skin depth, δ1 = 2τ 1 , plays an important role as a parameter in all diffusion coupling ωμ0 calculations. For convenience it is plotted for various values of resistivity. Integrating this result for y =0 and for h =0 yields the closed form result: ⎧⎡ ⎫ ⎤ ⎪⎢ ⎪ ⎥ z ⎛ ⎛ τ1 I 1 ⎪⎢ 1 1 ⎥ −(1+ i ) δ1 1 z⎞ z ⎞⎪ + − i 2 K 0 ⎜ (1 + i ) ⎟ − (1 + i ) Ex ( y = 0, z ) = ix ( y, z ) = e K1 ⎜ (1 + i ) ⎟ ⎬ ⎨ (1 + i ) π δ12 ⎪ ⎢ δ1 ⎠ δ1 ⎠ ⎪ ⎛ z ⎞ ⎛ z ⎞2 ⎥ ⎛ z⎞ ⎝ ⎝ ⎢ ⎥ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎪⎢ ⎪ δ δ ⎥ ⎝ 1 ⎠ ⎝ δ1 ⎠ ⎦ ⎝ 1⎠ ⎩⎣ ⎭ where K0 and K1 are modified Bessel Functions. Figure 8-4 Skin Depth as a Function of Frequency for Resisitivities, τ1 = 10, 100, 1000 Ω-m. Figure 8-5 Amplitude of Electric Field from a Line Source Placed at Heights, h = 0m, 100m, 200m, 500m, and 1000m, at z = 100m with τ1 = 80 Ω-m. Appendix DD - Page 70 of 104 PAGE 70 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Figure 8-6 Phase of Electric Field from a Line Source Placed at Heights, h = 0m, 100m, 200m, 500m, and 1000m, at z = 100m with τ1 = 80 Ω-m. Figure 8-7 Amplitude of the Electric Field at z = 100m from a Line Source Placed the Surface of a Homogeneous Half-Space with τ1 = 10, 100, 1000 Ω-m. Appendix DD - Page 71 of 104 PAGE 71 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Figure 8-8 Phase of the Electric Field at z = 100m from a Line Source Placed the Surface of a Homogeneous Half-Space with τ1 =10, 100, 1000 Ω-m 8.3.2 Two Layer Half-Space The current drive geometry of an infinitely long, horizontal wire placed a distance, h, above a conductive half-space is shown on the left side of Figure 8-9. A side view is shown on the right side of Figure 8-9. The current drive is harmonically time varying and directed along the x - axis at height, h, above it. The upper half-space has permittivity ε0and infinite resistivity, the layer of thickness h1 has permittivity ε1and resistivity, τ1, and the lower region has permittivity ε2and resistivity, τ2. All regions have free space permeability, μ0. If one neglects displacement current and relates current density, ix(y, z), and electric field, Ex(y, z), in region-2 through, Ex(y, z) = τ2ix(y, z), then the current density in the lower half-space, region-2, then for h =0 and y =0 can be determined to be Ex ( y = 0, z ) = − i 4τ 2 I 1 π Appendix DD - Page 72 of 104 δ ∞ u1eu2 h1 e − u2 z du − u1 h1 + + − − u e u u u u e ) ( )( ) 1 2 1 1 2 ∫ ( u + u )( u 2 2 0 1 u1 h1 PAGE 72 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Figure 8-9 Infinite Horizontal Current Drive, Two-Layered, Eddy Current Coupling Geometry. u1 = u 2 + ip12 p12 = δ1 = ωμ0 2 = 2 τ1 δ1 2τ 1 ωμ0 u2 = u 2 + ip22 p22 = δ2 = ωμ0 2 = 2 δ2 τ2 2τ 2 ωμ0 Similar expressions are developed in [A1-A3], but I am aware of no closed form expression for the above integral. The formula must be integrated numerically to obtain results. Note that the variable of integration is on the positive real axis and that no singularities are present on the positive real axis. As the skin depths get longer and longer, the branch cuts get closer to the real axis. If we consider the asymptotic behavior of the integrand as uÆ∞, then ⎡c ⎤ u1eu2 h1 e − u2 z du ⎥ ⎢∫ − u1 h1 u1h1 i 4τ I 1 ( u + u1 )( u1 + u2 ) e + ( u − u1 )( u1 − u2 ) e ⎥ Ex ( y, z ) = − 2 2 ⎢ 0 ⎥ π δ2 ⎢ 1 1 i ⎢ + E1 ( cz ) + 2 ( 2h1 − z ) E2 ( cz ) ⎥ c 4δ1 ⎢⎣ 4 ⎥⎦ to the first two terms in e− uz e− uz and 2 where c » max[δ1, δ2]. u u Appendix DD - Page 73 of 104 PAGE 73 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Figure 8-10 Amplitude of the Electric Field at z = 100m from a Line Source at the Surface of a Two-Layered Half-Space. Figure 8-11 Phase of the Electric Field at z = 100m from a Line Source at the Surface of a Two-Layered HalfSpace. Appendix DD - Page 74 of 104 PAGE 74 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 8.4 Eddy Current Coupling into Homogeneous Half-Space from Uniform Magnetic Field at Surface If we consider the geometry shown in Figure 8-12, z z y H = H0yy region – 0 ε0, μ0, τ0=∞ H = H0yy dl h x x region – 0 ε0, μ0, τ0=∞ y region – 1 ε1, μ0, τ1 region – 1 ε1, μ0, τ1 (y,z) (y,z) -z Figure 8-12 Geometry for Eddy Current Field Calculations in Homogenous Half-Space Driven by Uniform Magnetic Field at the Surface. with uniform harmonic magnetic field with time harmonic variation eiωt , H = H 0 y yˆ , in the y-direction, then the electromagnetic field equations, neglecting displacement current can be developed directly from Maxwell’s equations. ∂jx ( z ) ∂y = iωσ 1μ0 H y ( z ) where jx(z) is the current density in region-1 and Hy(z) is the magnetic field in region-1. The second equation is given by ∂H y ( z ) ∂y = jx ( z ) Substituting one equation into the other yields ∂2 H y ( z ) ∂y 2 = α 2H y ( z) where Appendix DD - Page 75 of 104 PAGE 75 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine α= (1 + i ) δ1 2 δ1 = ωμ0σ 1 where δ1 is the skin depth in region-1. The general solution of the above equation is H y ( z ) = K1eα z + K 2 e −α z Choosing the properly decaying solution as z Æ -∞, and using the boundary condition at the surface of Hy(0)=H0y H y ( z ) = H 0 y eα z jx ( z ) = ∂H y ( z ) ∂y = α H 0 y eα z Because Ex ( z ) = τ 1 jx ( z ) where τ1 = 1 σ1 E x ( z ) = τ 1α H 0 y eα z is the only component of the electric field in region-1. Because a surface current density is related to the magnetic field immediately below a perfect conductor by the relationship j0 x = −2nˆ × H 0 y the above solution could also be considered to be the electric field of a homogeneous half-space excited by a uniform x-directed current flowing on the bottom surface of a perfectly conducting sheet on the surface of the homogeneous half-space, but electrically isolated from it. The exciting current on the sheet to produce the field in the equations would be j0 x = −2 H 0 y in the x-direction, or alternatively Appendix DD - Page 76 of 104 PAGE 76 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine H0 y = − 1 j0 x 2 in the above formulas. 8.5 Eddy Current, Infinitesimal and Finite Length Horizontal Drive Wire Models Eddy current models for infinitesimal and finite length horizontal drive wires over a homogeneous halfspace have been developed in [A4-A7]. The x-directed electric field immediately below the wire can be expressed in closed form for the infinitesimal length dipole [A7]. Expressions for the electric field in a two-layer half-space excited by an infinitesimal horizontal drive wire have been developed in [A8]. These models are quite complicated and were not further developed for this program because of lack of time and resources. 8.6 References for Appendix A A1. Wait, James R., Electromagnetic Waves in Stratified Media, The Macmillan Company, New York, NY, 1962. A2. Tegopoulis, J. A., and E. E. Kriezis, Eddy Currents in Linear Conducting Media, Elsevier, New York, NY, 1985. A3. Kaufman, A. A., and P. Hoekstra, Electromagnetic Soundings, Elsevier, New York, NY, 2001. A4. Goldstein, A. A., and D. W. Strangway, Audio-frequency Magnetotellurics with a Grounded Electric Dipole Source, Geophics, Vol. 40, December 18, 1974, pp669-683. A5. Sommerfeld, Arnold, Electromagnetic Waves Near Wires, Wied. Annalen, Vol 67, 1899, pp233-290. A6. Sommerfeld, Arnold, Partial Differential Equations in Physics, Lectures on Theoretical Physics, Vol. VI, Academic Press, New York, NY, 1964. A7. King, Ronold W. P., Margaret Owens, and Tai Tsun Wu, Lateral Electromagnetic Waves, Springer-Verlag, New York, NY, 1992. A8. Riordan, John, and Erling D. Sunde, Mutual Impedance of Grounded Wires for Horizontally Stratified Two-Layer Earth, Bell System Technical Journal, Vol 12, pp162-177, 1933. Appendix DD - Page 77 of 104 PAGE 77 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 9 Appendix B –Calibration Documentation of Measurement Equipment 10 Fiber Optics Transfer Function 9 8 Amplitude (dB) 7 6 5 4 3 2 1 0 1 10 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 9-1 Calibration Frequency Response of Fiber-optic Transmitter/Receiver Pair. Appendix DD - Page 78 of 104 PAGE 78 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 10 0 Pearson Model 4688 LEMflex Pearson Model 110A Amplitude (V/A) 10 10 10 -1 -2 -3 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 9-2 Calibration Frequency Response of Current Probes used. 10 2 Effective Height (m) Sandia Antenna Calibration 10 10 10 1 0 -1 10 1 10 2 3 10 Frequency (Hz) 10 4 Figure 9-3 Calibration Frequency Response of Sandia Dipole Antenna. Appendix DD - Page 79 of 104 PAGE 79 OF 104 10 5 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 5 Nanofast High Impedance 1x Probe 0 Amplitude (dB) -5 -10 -15 -20 -25 1 10 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 9-4 Calibration Frequency Response of Nanofast High-Impedance Probe. Appendix DD - Page 80 of 104 PAGE 80 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Figure 9-5 Certificate of Calibration for 4395A Network Analyzer. Appendix DD - Page 81 of 104 PAGE 81 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 10 Appendix C – Compilation of Measured Data Direct Drive Transfer Function Data: The following transfer functions were measured with the mine grounding system in current state. Current Transfer Function Amplitude 10 10 10 10 10 10 10 0 1 3 4 5 6 7 -1 -2 -3 -4 -5 -6 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-1 Direct Drive Current Transfer Function of Trolley Communication Line with a Local Ground. Appendix DD - Page 82 of 104 PAGE 82 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Current Transfer Function Amplitude 10 10 10 10 10 10 10 0 1 3 4 5 6 7 -1 -2 -3 -4 -5 -6 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-2 Direct Drive Current Transfer Function of Trolley Communication Line with a Fence Ground. Voltage Transfer Function Amplitude 10 10 10 10 10 10 10 0 1 3 4 5 6 7 -1 -2 -3 -4 -5 -6 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-3 Direct Drive Voltage Transfer Function of Conveyor Structure with a Local Ground. Appendix DD - Page 83 of 104 PAGE 83 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Current Transfer Function Amplitude 10 10 10 10 10 10 10 0 1 3 4 5 6 7 -1 -2 -3 -4 -5 -6 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-4 Direct Drive Current Transfer Function of Conveyor Structure with a Local Ground. Voltage Transfer Function Amplitude 10 10 10 10 10 10 10 0 1 3 4 5 6 7 -1 -2 -3 -4 -5 -6 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-5 Direct Drive Voltage Transfer Function of Conveyor Structure with a Fence Ground. Appendix DD - Page 84 of 104 PAGE 84 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Current Transfer Function Amplitude 10 10 10 10 10 10 10 0 1 3 4 5 6 7 -1 -2 -3 -4 -5 -6 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-6 Direct Drive Current Transfer Function of Conveyor Structure with a Fence Ground. 10 Voltage Transfer Function Amplitude 10 10 10 10 10 10 2 1 3 4 5 6 7 1 0 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-7 Direct Drive Voltage Transfer Function of Rail Structure with a Local Ground. Appendix DD - Page 85 of 104 PAGE 85 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Current Transfer Function Amplitude 10 10 10 10 10 10 1 1 3 4 5 6 7 0 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-8 Direct Drive Current Transfer Function of Rail Structure with a Local Ground. 10 Voltage Transfer Function Amplitude 10 10 10 10 10 10 2 1 3 4 5 6 7 1 0 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-9 Direct Drive Voltage Transfer Function of Rail Structure with a Fence Ground. Appendix DD - Page 86 of 104 PAGE 86 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Current Transfer Function Amplitude 10 10 10 10 10 10 1 1 3 4 5 6 7 0 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-10 Direct Drive Current Transfer Function of Rail Structure with a Fence Ground. Appendix DD - Page 87 of 104 PAGE 87 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine The following transfer functions were measured with the mine grounding system similar to the grounding scheme in place during explosion. Voltage Transfer Function Amplitude 10 10 10 10 10 10 10 0 1 2 3 -1 -2 -3 -4 -5 -6 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-11 Direct Drive Voltage Transfer Function of Power Cable Shield with a Local Ground. Current Transfer Function Amplitude 10 10 10 10 10 10 10 0 1 2 3 -1 -2 -3 -4 -5 -6 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-12 Direct Drive Current Transfer Function of Power Cable Shield with a Local Ground. Appendix DD - Page 88 of 104 PAGE 88 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Voltage Transfer Function Amplitude 10 10 10 10 10 10 10 0 1 2 3 -1 -2 -3 -4 -5 -6 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-13 Direct Drive Voltage Transfer Function of Power Cable Shield with a Fence Ground. Current Transfer Function Amplitude 10 10 10 10 10 10 10 0 1 2 3 -1 -2 -3 -4 -5 -6 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-14 Direct Drive Current Transfer Function of Power Cable Shield with a Fence Ground. Appendix DD - Page 89 of 104 PAGE 89 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 10 0 Voltage Transfer Function Amplitude 2 3 10 10 10 10 10 10 -1 -2 -3 -4 -5 -6 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-15 Direct Drive Voltage Transfer Function of Rail Structure with a Local Ground. 10 0 Current Transfer Function Amplitude 2 3 10 10 10 10 10 10 -1 -2 -3 -4 -5 -6 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-16 Direct Drive Current Transfer Function of Rail Structure with a Local Ground. Appendix DD - Page 90 of 104 PAGE 90 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 10 0 Voltage Transfer Function Amplitude 2 3 10 10 10 10 10 10 -1 -2 -3 -4 -5 -6 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-17 Direct Drive Voltage Transfer Function of Rail Structure with a Fence Ground. 10 0 Current Transfer Function Amplitude 2 3 10 10 10 10 10 10 -1 -2 -3 -4 -5 -6 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-18 Direct Drive Current Transfer Function of Rail Structure with a Fence Ground. Appendix DD - Page 91 of 104 PAGE 91 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 10 0 Current Transfer Function Amplitude 2 3 10 10 10 10 10 10 -1 -2 -3 -4 -5 -6 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-19 Direct Drive Current Transfer Function of Trolley Communication Line with a Local Ground. 10 0 Current Transfer Function Amplitude 2 3 10 10 10 10 10 10 -1 -2 -3 -4 -5 -6 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-20 Direct Drive Current Transfer Function of Trolley Communication Line with a Fence Ground. Appendix DD - Page 92 of 104 PAGE 92 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Normalized Composite Electric Field Amplitude (V/m/A) Indirect Drive Transfer Function Data: Surface current drive in the P-direction 10 10 10 10 10 0 P2 P3 PX4 P5 P6 P7 P8 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Normalized Composite Electric Field Amplitude (V/m/A) Figure 10-21 Normalized Composite Electric Field for P-Directed Surface Current Drive at Positions from P2 to P8. 10 10 10 10 10 0 X1 X2 X3 PX4 X5 X6 X7 X8 X9 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-22 Normalized Composite Electric Field for P-Directed Surface Current Drive at Positions from X1 to X9. Appendix DD - Page 93 of 104 PAGE 93 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Normalized Vertical Electric Field Amplitude (V/m/A) 10 10 10 10 10 0 P2 P3 PX4 P5 P6 P7 P8 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-23 Normalized Vertical Electric Field for P-Directed Surface Current Drive at Positions from P2 to P8. Normalized Vertical Electric Field Amplitude (V/m/A) 10 10 10 10 10 0 X1 X2 X3 PX4 X5 X6 X7 X8 X9 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-24 Normalized Vertical Electric Field for P-Directed Surface Current Drive at Positions from X1 to X9. Appendix DD - Page 94 of 104 PAGE 94 OF 104 Normalized P-Directed Electric Field Amplitude (V/m/A) Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 10 10 10 10 10 0 P2 P3 PX4 P5 P6 P7 P8 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Normalized P-Directed Electric Field Amplitude (V/m/A) Figure 10-25 Normalized P-Directed Electric Field for P-Directed Surface Current Drive at Positions from P2 to P8. 10 10 10 10 10 0 X1 X2 X3 PX4 X5 X6 X7 X8 X9 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-26 Normalized P-Directed Electric Field for P-Directed Surface Current Drive at Positions from X1 to X9. Appendix DD - Page 95 of 104 PAGE 95 OF 104 Normalized X-Directed Electric Field Amplitude (V/m/A) Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 10 10 10 10 10 0 P2 P3 PX4 P5 P6 P7 P8 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Normalized X-Directed Electric Field Amplitude (V/m/A) Figure 10-27 Normalized X-Directed Electric Field for P-Directed Surface Current Drive at Positions from P2 to P8. 10 10 10 10 10 0 X1 X2 X3 PX4 X5 X6 X7 X8 X9 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-28 Normalized P-Directed Electric Field for P-Directed Surface Current Drive at Positions from X1 to X9. Appendix DD - Page 96 of 104 PAGE 96 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Normalized Composite Electric Field Amplitude (V/m/A) Indirect Drive Transfer Function Data: Surface current drive in the X-direction 10 10 10 10 10 0 P2 P3 PX4 P5 P6 P7 P8 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Normalized Composite Electric Field Amplitude (V/m/A) Figure 10-29 Normalized Composite Electric Field for X-Directed Surface Current Drive at Positions from P2 to P8. 10 10 10 10 10 0 X1 X2 X3 PX4 X5 X6 X7 X8 X9 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-30 Normalized Composite Electric Field for X-Directed Surface Current Drive at Positions from X1 to X9. Appendix DD - Page 97 of 104 PAGE 97 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Normalized Vertical Electric Field Amplitude (V/m/A) 10 10 10 10 10 0 P2 P3 PX4 P5 P6 P7 P8 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-31 Normalized Vertical Electric Field for X-Directed Surface Current Drive at Positions from P2 to P8. Normalized Vertical Electric Field Amplitude (V/m/A) 10 10 10 10 10 0 X1 X2 X3 PX4 X5 X6 X7 X8 X9 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-32 Normalized Vertical Electric Field for X-Directed Surface Current Drive at Positions from X1 to X9. Appendix DD - Page 98 of 104 PAGE 98 OF 104 Normalized P-Directed Electric Field Amplitude (V/m/A) Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 10 10 10 10 10 0 P2 P3 PX4 P5 P6 P7 P8 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Normalized P-Directed Electric Field Amplitude (V/m/A) Figure 10-33 Normalized P-Directed Electric Field for X-Directed Surface Current Drive at Positions from P2 to P8. 10 10 10 10 10 0 X1 X2 X3 PX4 X5 X6 X7 X8 X9 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-34 Normalized P-Directed Electric Field for X-Directed Surface Current Drive at Positions from X1 to X9. Appendix DD - Page 99 of 104 PAGE 99 OF 104 Normalized X-Directed Electric Field Amplitude (V/m/A) Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 10 10 10 10 10 0 P2 P3 PX4 P5 P6 P7 P8 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Normalized X-Directed Electric Field Amplitude (V/m/A) Figure 10-35 Normalized X-Directed Electric Field for X-Directed Surface Current Drive at Positions from P2 to P8. 10 10 10 10 10 0 X1 X2 X3 PX4 X5 X6 X7 X8 X9 -1 -2 -3 -4 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-36 Normalized X-Directed Electric Field for X-Directed Surface Current Drive at Positions from X1 to X9. Appendix DD - Page 100 of 104 PAGE 100 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 10 0 Parallel to Drive Wire Perpendicular to Drive Wire Amplitude (V/A) 10 10 10 -1 -2 -3 10 1 10 2 3 10 Frequency (Hz) 10 4 10 5 Figure 10-37 Induced Voltage on Pump Cable (~300 m long) due to Wire Current Drives on Surface. Appendix DD - Page 101 of 104 PAGE 101 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 11 Appendix D – List of Underground Sealed Area Coal Mine Explosions Suspected of Lightning Initiation 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Mary Lee #1 – August 22, 1993, Walker County, AL Oak Grove Mine – April 5, 1994, Jefferson County, AL Gary 50 – Between June 9 and 16, 1995 Oak Grove Mine – January 29, 1996, Jefferson County, AL Oasis Contracting Mine # 1 – May 22, 1996, Boone County, WV Oasis Contracting Mine # 1 – June 15, 1996, Boone County, WV Oak Grove Mine – July 9, 1997, Jefferson County, AL Soldier Canyon Mine – July, 2001, Wellington, UT Pinnacle Mine – September 1, 2003, Wyoming County, WV Pinnacle Mine – August 30, 2005, WV, Wyoming County, WV Sago Mine – January 2, 2006, Tallmansville, WV Appendix DD - Page 102 of 104 PAGE 102 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine 12 Appendix E – Memorandum from Dr. Krider Appendix DD - Page 103 of 104 PAGE 103 OF 104 Appendix DD - Measurements and Modeling of Transfer Functions for Lightning Coupling into the Sago Mine Distribution Internal: 5 MS1152 2 MS1152 2 MS1182 3 MS1152 1 MS1152 1 MS1152 2 MS9018 2 MS0899 M. B. Higgins, 1653 M. E. Morris, 1652 L. X. Schneider, 1650 M. Caldwell, 1653 D. R. Charley, 1653 L. Martinez, 1653 Central Technical Files, 08944 Technical Library, 04536 External: 20 William Helfrich Mine Safety & Health Administration Pittsburgh Safety & Health Technology Center P.O. Box 18233 626 Cochrans Mill Road – Bldg. 151 Pittsburgh, PA 15236 1 E. Philip Krider Institute of Atmospheric Physics The University of Arizona P.O. Box 210081, Rm. 542 1118 E. 4th Street Tucson, AZ 85721-0081 1 Martin A.Uman Department of Electrical and Computer Engineering University of Florida P.O. Box 116200 311 Larsen Hall Gainesville, FL 32611 Appendix DD - Page 104 of 104 PAGE 104 OF 104 Appendix EE - Report on the Investigation of the Well Heads and Gas Pipeline System Appendix EE - Page 1 of 12 Appendix EE - Report on the Investigation of the Well Heads and Gas Pipeline System Appendix EE - Page 2 of 12 Appendix EE - Report on the Investigation of the Well Heads and Gas Pipeline System Appendix EE - Page 3 of 12 Appendix EE - Report on the Investigation of the Well Heads and Gas Pipeline System Appendix EE - Page 4 of 12 Appendix EE - Report on the Investigation of the Well Heads and Gas Pipeline System Appendix EE - Page 5 of 12 Appendix EE - Report on the Investigation of the Well Heads and Gas Pipeline System Appendix EE - Page 6 of 12 Appendix EE - Report on the Investigation of the Well Heads and Gas Pipeline System Appendix EE - Page 7 of 12 Appendix EE - Report on the Investigation of the Well Heads and Gas Pipeline System Appendix EE - Page 8 of 12 Appendix EE - Report on the Investigation of the Well Heads and Gas Pipeline System Appendix EE - Page 9 of 12 Appendix EE - Report on the Investigation of the Well Heads and Gas Pipeline System Appendix EE - Page 10 of 12 Appendix EE - Report on the Investigation of the Well Heads and Gas Pipeline System Appendix EE - Page 11 of 12 Appendix EE - Report on the Investigation of the Well Heads and Gas Pipeline System Appendix EE - Page 12 of 12 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 1 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 2 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 3 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 4 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 5 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 6 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 7 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 8 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 9 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 10 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 11 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 12 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 13 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 14 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 15 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 16 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 17 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 18 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 19 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 20 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 21 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 22 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 23 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 24 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 25 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 26 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 27 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 28 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 29 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 30 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 31 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 32 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 33 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 34 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 35 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 36 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 37 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 38 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 39 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 40 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 41 of 42 Appendix FF - Geophysical Survey of the Old 2 Left Section of the Sago Mine Appendix FF - Page 42 of 42 ,x 1 Explanation Hen.26:35 I?r' ?7 a" "l a" Ir Is". I A. "5.4.4354. 1- (fin-V. .r ax 511Location of lightning strike reported by Vaisala's National Lightning Detection Network (N LDN). Number to left of symbol represents the peak current in kilo-amps; number to right of symbol represents the time that the peak current was recorded. Location of lightning strike reported by Weather Decision Technologies, nc.'s U.S. Precision Lightning Network (USPLN). Numberto left of symbol represents the peak current in kilo-amps; number to right of symbol represents the time that the peak current was recorded. Sago Mine workings. Keyspan gas lines and wells. Great Oak gas lines and wells. Dominion gas lines and wells. Chesapeake gas lines and wells. Trubie Run gas lines. Mike Ross gas lines and wells. Eastern American Office gas lines. Eastern American Energy gas lines and wells. Location of large poplar tree shattered by lightning. Appendix GG Sago Mine MSHA ID 46-08791 Wolf Run Mining Company Sago Mine in relation to recorded locations of lightning strikes, gas wells, and gas lines. Location of gas lines and gas wells courtesy of U.S. Department of Labor Appendix HH - Observation and Sampling Collection Methodology Mine Safety and Health Administration Pittsburgh Safety Health Technology Center P.O. Box 18233 Pittsburgh. PA 15236 Roof Control Division 06AA23 August '31, 2006 MEMORANDUM FOR RICHARD A. GATES District Manager, Distri 11 THROUGH: KELVIN K. WU Acting Chief, Pittsburgh Safety and Health Technology Center W?ygf M. TERRY HO Chief, Roof Control Division ?tl?xmfiidfw ?V?l FROM: E. PHILLIPSON 5/ t? Geologist, Roof Control Division SUBJECT Observations and Sample Collection Methodology at Wolf Run Coal Company, Sago Mine, MSHA I. D. No. 46?08791, on March 20?21, 2006 Observations As requested by the Accident Investigation Team, several rock and water samples were collected from the Sago Mine. Several water samples were collected from the Backhannon River, Trubie Run, and an tin?named tributary off Trubie Run. Additionally, observations of features were conducted in the Vicinity of spad 4064. March 20, 2006 Sample Collection Sample collection began on the track entry of the Main between crosscuts 32 and 33, where arches rivere installed through difficult ground conditions beneath Trubie Run. Dripping water was collected in a small glass vial, and labeled ?Trubie? (Figure 1). Appendix HH - Page 1_of_15_ Appendix HH - Observation and Sampling Collection Methodology Figure 1. Collection of dripping water in the track entry of the Main, between Crosscuts 32?33. Observations resumed inby the former 211d Left Seals. The underground sample collection effort was prompted by an interest to document the cause of the blue haze? that has been observed throughout portions of the 2"d Left Mains. Figure 2 represents a map of the 2?d Left Mains showing the locations of water and rock samples collected on March 20, 2006. Observations proceeded inby along the #7 Enhy of the Main, where ?blue haze" was observed on several broken rock faces. This entry had been benched, and these locations were not accessible. "Blue haze? was observed in the vicinity of spad 3986, where the crosscut had not been benched. Two samples, one of rock that hosts the ?blue haze? on its surface, as well as a water sample, were collected from the roof. of the crosscut between spads 3986 and 398'}. Water droplets were collected with a plastic dropper by drawing each individual droplet into the dropper, and then expelling the water into a glass vial (Figure When all available water droplets that were in contact with the "blue haze? had been. collected, the glass vial was approxi mater half full. The plastic dropper was used only at this collection site, and was labeled with the sample number before being separately stored to avoid cross contamination with other droppers, which were carried in a sealed plastic freezer bag. Mine personnel assisted in the collection of a slab of immediate roof rock that hosted blue haze? on a fracture surface. When the slab was retrieved, it was broken in. half, with one half retained by MSHA personnel and the other half provided to mine personnel, who were present. The MSHA sample was stored in a freezer bag and labeled whereas the sample provided to the mine was stored in a freezer bag labeled duplicate?. Photos were taken of the sample in place in the roof, after it had been split in half, and of each half in. its reSpective bag (Figure 4). Appendix HH - Page 2 of 15 Appendix HH - Observation an? Sampling Collection Methodology 5 \meA-m Wm a? x? Xxx}; Riv} . f) ?x 10!] 208 m: feet 5; Bl?huag/iv j: :2 {fig/3&gig/Rx], i 8< 6% if 3 1 me 7&2 ?x f/ ?1 . /\\lum?l Figure 2. Map of the observed area of 2?d Left Mains in which rock and water samples were collected on March 20, 2006. 5.3 ,i/V l; Appendix HH - Page 3 of 15 Appendix HH - Observation aild Sampling Collection Methodology Figure 3. Collection of water droplets on the roof and fracture faces where "blue haze? was observed between spads 3986 and 3981. Figu re 4. Photo of Sample with [he mine?s duplicate sample. The sample was collected Erom the roof approximately 28 feet into the left?hand crosscut from sped 3986. The distance was measured with a laser range ?nder. A_11pendix HH - Page 4 of 15_ Appendix HH - Observation and Sampling Collection Methodology 5 Observations then proceeded along the #2 Entry of 2ml Left Mains, and across to the vicinity of spad 4028. "Blue haze? was observed on the rib of the right inby crosscut in the spad 4.028 intersection. 0 water was present at this site, and so no water sample could be collected. However, a slab of rib that hosted the blue haze? on the exposed face was collected, and Split in half, with. each. half provided to the mine and MSHA personnel present (Figure 5). As the MSHA sample was being further split to obtain a suitably sized sample to place in a bag, the rock separated along a prominent bedding plane, exposing a fossil of a fern leaf attached to a very well developed central stalk, preserving the diamond?shaped, nearly serrated texture of the bark. Figure Samples collected from the right inby crosscut in the sped 4028 intersection. Half of the sample was retained by MSHA in a bag labeled "BH40?28ric" and the other half was retained by mine personnel in a bag labeled "Bl-14028ric duplicate." Observations continued. across to the spad 4010 intersection, which had to be accessed. by proceeding inby from spad 4028 to a more accessible crosscut due to benching. No water was observed in this sample collection locality, so no water sample was collected. A slab of ?blue haze" coated rock was obtained from the right rib of the #6 Entry, where the floor ramped down to the beginning of the benched area (Figure 6). Appendix HH - Page 5 of 15 Appendix HH - Observation and Sampling Collection Methodology Figure 6. Slab of "blue haze" coated rock retrieved from the right rib approximately 55 feet inby spad 4010. This slab was subsequently split in half, with each half retained by mine and MSI-IA personnel present. Observations continued inby along #7 Entry, from the adjacent crosscut from spad 4010 to the crosscut inby spad 4063. This crosscut connects with terminated #8 Entry, where a barrier was left for a gas well. Observations were made of the intersection of the crosscu't and the #8 Entry, the segment of the #8 Entry inby spad 4064, and the segment of the crosscut A sample of water seeping from the right rib of the #8 Entry was collected with a plastic dropper, which was labeled with the sample number ?W4064?3l? to mark it from the dropper used previously for sample collection at the site (Figure 7). The labeled dropper was photographed. Appendix HH - Page 6 of 15 Appendix HH - Observation and Sampling Collection Methodology 7 I ?ma uni Figure 7. Water sample in glass vial collected from right rib of solid coal block in #8 Entry, 31 feet inby spad 4064. Plastic dropper is labeled with mple number to distinguish it from the separate dropper used in collection of sample til-1398628. The intersection is characterized by a series of wavy brown streaks, which extend into the adjacent #8 Entry and crosscut. Each wavy brown streak has a linear, although undulatory trend that was measured with a Brunton compass. The streaks in the outby side of the #8 Entry exhibit a trend of approximately 13-15013, such that they project into the right inby corner of the intersection (Figure 8 and 9). Traversing from the #8 En try inby to the left?hand crosscut, the brown wavy streaks change their orientation, and continue to trend toward the right inby corner of the intersection. Thus, the brown wavy streaks define a radiating pattern with a center point about the right inby corner of the intersection. These long, roughly linear, but wavy brown streaks are developed in flat, planar portions of the exposed roof. Several brown wavy streaks were observed that are approximately perpendicular to those developed on the planar roof horizon. These perpendicular streaks were deveIOped along protrusions or small brows from the roof, and represented the exposed ridges of bedding asperities, and are generally less than 4 feet long. Comparatively, the longer linear, but wavy streaks are generally 7-?12 feet long. The brown wavy streaks do not represent surficial dust, and instead represent actual linear exposures of rock. The brown shale is masked by a very thin bedding parting of black, carboniferous shale, except where it has been. removed to expose the brown wavy streaks. The removal mechanism gave the appearance of sand. blasting or scouring, rather than gouging. There were no depressions associated with the streaks that would suggest a geologic origin such as scour marks or trace fossil imprints. Despite the straight rib profiles observed even in the benched portions of this Appendix HH - Page 7 of 15 Appendix HH - Observation and8 Sampling Collection Methodology area, characterized by an absence of sloughing, several large (1 2 feet) blocks of coal had been removed from the face of the discontinued #8 Entry. This intersection, and the short segment of the crosscut and entry, had not been benched. While in the intersection, gas could be heard seeping from the right rib, and the sound of gurgling water was also apparent from the right rib. The sound was similar to that of water running down a drain, rather than pouring into a standing body. The intersection was characterized by standing water that ranged in depth from nominally 3 inches to at least 7 inches. Underground observations concluded in this intersection and the mine was exited. Figure 8. Photo of linear, wavy brown streaks in the terminated #8 Entry, viewed looking inby along #8 Entry. Streaks are actually the exposed based of the brown shale that overlies the thin. black carbonaceous shale bedding parting. Appendix HH - Page 8 of 15 Appendix HH - Observation and Sampling Collection Methodology 9 coal blocks separated I feat from prominent sleet plane 100 4 ll A Hammers/I, -. @De/ Figure 9. Detailed map of observations in the 2"3 Left Mains, showing positions of brown streaks? and sample collection sites with detailed geological observations from February 21, 2006. March 2?1, 2006L0bservations Observations and sample collection continued on the surface the following day. Water samples were collected from a segment of the Buckhannon River, a segment of Trubie Run, a small tributary that drains into Trubie Run, and from Trubie Run at a position directly over the Main where sample "Tm bie? was collected on the previous day (Figure 10). Appendix HH - Page 9 of 15 Appendix HH - Observation and Sampling Collection Methodology ?ue-Ii - yea?s?vfax. . i x? . was- - ?x a i ?M.A?l . I a 'h?I 1 _-:eur-?rr2-. he1.1-. I I. ?Ida?? .l ?ylrIglI AV. Jr5-55{ublBMaln I: I aI? J. I {g ?15?If! l. . '?dllfi JIIJ LCI): ..-I I fix": - I Gd'3qanI,/r I '3 a" - "inFigure it). Map of surface water sample locations in relation to Sago Mine workings, lineaments, and hication of lightning strike. The first sample was collected from. the Buckhannon River along a segment north of the position of a reported lightning strike, and the point where Trubie Run enters the river. The sample, labeled "W?Buck,? was collected from the edge of the stream bank, in slow moving water, with no visible organic material present. The location of the sample site was determined with a hand-held GPS unit, and recorded for diSplay in the 618 from which Figure 10 was produced. The next sample was collected from Trubie Run between the Buckhannon River and. the un-named, northeast trending tributary, and labeled ?Trubie? (Figure A sample with the same name was collected on March 20, 2006, from the underground workings on the track entry in the Main beneath Trubie Run. However, the two samples have different exhibit numbers. Sample Appendix HH - Page 10 of 15 Appendix HH - Observation andISampling Collection Methodology "TrubieNEtrib? was collected from the un-named tributary that trends northeast from Trubie Run. No other access could be found for this small stream, although the valley was paralleled by a county road to the head of the valley. Sample ?TrubieMain? was collected from Trubie Run above the point where the Main passes beneath Trubie Run. This location was determined with a hand?held CPS based on coordinates provided by the mine. All locations were confirmed by MSHA personnel using their own hand?held GPS unitGig? 4- -11 - i - Figure 11. Photo of surface water sample collection at location "Trubie." Not to be sample named ?Trubie? collected {rem the track entry of the Main, underground. omsed'wiui the Upon completion of the surface water sample collection, a prominent tree that was reportedly struck by lightning on the day of the Sago Mine explosion was visited. The position of the tree was determined. using a hand-held GPS unit, and plotted on the RCD (315 maps (Figure 12). Using a powerful hand-held magnet, a small piece of magnetic material was found at the base of the tree. The material is interpreted to be magnetite, and exhibits surface oxidation. The genesis of the magnetite is not known. The magnetite could represent detrital material that was originally incorporated in eroded sedimentary rock, or if evidence of lightning striking the ground, would have to represent a change from original hematite (F6203) to magnetite (F8304). Appendix HH - Page 11 of 15 Appendix HH - Observation andZSampling Collection Methodology :I.- leh It! \il Figure 12. Location of lightning?struck tree, determined by RCD personnel with hand-held GPS unit. Location is 300 feet from plotted location of lightning strike, indicated by small red star in center of cross- hatched area. Locality Pole 1" represents a twin?pole power line support that reportedly exhibited lightning damage, although the timing of the damage is unknown. Discussion and Analysis Results The purpose of the samples collected on March 20, 2006, was to document the ?blue haze? and determine its composition. The rock samples and water sample, collected at ?BI-I4028ric,? and were submitted for X-ray diffraction and Inductively Coupled PlasmaAMass Spectrometry to determine the composition of the blue film that coats the rocks. Due to the inability to obtain suf?cient sample material for analysis, only the iridescent coating from sample ?BH4028ric? could be analyzed "by X?ray diffraction. The iridescent "blue haze? scraped from sample ?BH4028ric? was determined to be calcite, based on the X-ray spectrum of diffraction peaks. Due to the lack of limestone in the mining horizon and the absence of water precipitates and ?drippers,? it seems most likely that the source of the calcite diffraction peaks is from. lime rock dust applied to the coal ribs. Whole-rock geochemical analysis was also conducted of the "brown streaks," and the samples collected at each "blue haze" location to determine if there were differences in the rock. The results of analysis by" Inductively Coupled Plasma, reported as major Appendix HH - Page 12 of 15 91 J0 91 aged - HH xlpuaddv 9.13M Imam mu 1mg agqml spumq 3,12,qu a?mump wumu ?daaqs masaxdm Sleaddu pm? pads 50 apps 31}; 19M) 14:13-3de ll 3N6 mammaugj c133,, SB 04 pauagal 3qu go Snumm Lu 13112;? mmpanpuon [1331113913 put? Hd 91.11 5535512 512m (Hammin) mew mp 50 3313.111 am mm; puts mew 3,131 WE GO Anna 8# 31.13 JO qu 11131.1 :3qu mm; puma @003 ?15 113mm uo smeaus campus mm; papanoa 3916111123 .13; 12m amp }0 asodmd aql ?Uam 593 page: am 139412301 3; qamm pumiaq 233:2; ziuu-g 8# am 50 mums qseaquou 0; wasde 1mm waned 19mm 12 mm; gamut 1110353 9119,3133 ummq? 31.1.1 412ng 32101.1 ?upsalaquy sy 31 '59,wa pamms 39M goo; am 30 Bude Bugppaq SHOQDEHOQJED >13121q ?ugqn, mp 313wa 3113115 mqu BLQKIJQAO am Buysodxa 121035 Antenna aw ?33129145 aqi ?guuaqmam 03, mumsgsal 91; 03 s>pm :{wmauupas my {mama}: irossaane umuums L9 s; ?apgxo mummy; 13 ?amng 'Spaq [1303 pue ?samqs ?sr?lup u; punt); {mamm Mumauupas t? 51 ?amueqma L103 ma ?amapgg 'Lezum agaoasmu 30 xmum 12 moq?nmqq 13313111233 511112.113? 21191111 quepunqe 310 anuasaxd pamaumnop 4131.11 .3001 agegpamun am 30 saldums 131.140 LEO papnpuoo sgsA" [131.19 :3;d0350.t3;m are 51111531 ?aqu 30 smauodmoa papadxa an? anunoml me ?anzxoasmu ?zqn-mb JO an; i 911.1111 puB ?aquapgs ?alyuumbi ?23,.mnb memos 03 Aim?X Aq sum ?33,ng 94? paddmp 3qu u; ?means am JO aldums 31119qu ?31 Lu papajim Saldums >130; .10} ?sa;>gxo mfmu 512 pagmdaJ musnld pmdnaj) MBAQDHPUI Jo ginseng '1 alqu] 900i 1623i 600 693? 293 990 't?i?D ELSE 189$? t- ;gqp?g 25"86 921. 9?0 LEVI 690 890 350 ?000 i 90': VS 11$? 8? qg 0275-48 3388 180 5.981 9?30 9170 SH) 300 0 932 1701-2 89917 IIQIUXE '99?0 LOVHE 8:586 1300! I 29?s 136 HT) Gi?'l 179$ 96 8 2190 EEO W?l ZVL 890?!) 800 E9122 41023 SE 273 $99 9899 1 Swans ummq c3 d3! 'Snd d3! d3} 'S?j ?S?d d3! 'S?i d3! d0! d3! ?S?j d01?SfH dOl d3! "?S?zi :pougaw aauaxawa LCD ?75 - L00 5% L00 L000 ?fa 10'0 L00 ?is: ED 0 We L00 L390 0/21 L00 of; ?00 incl LOO .0, I iliUJl'} Housman :suun 1930i 903d ZOLL 08M OZEN C390 05w OUW A (320385 EOZIV ZOIS Lxuewega ?umas [1203 go ands .IQMOI put? .?Iaddn am 19111 11021101.] nos-oaled 1-3 mm; papanm mam gg-Q [U?qg mm; saldmes semaqm 111939 11203 :3ka 50 nomad snolamgssog 11101; 1194331103. Stem aysggo?qg aldums 33333.: ?121.11 9901017143 pure mm; sagdtues qu 331.13 uaamgaq 113115311; 51 aqusoqd anqm ?gg?O m?qg '40 qu mp 1110.1; papanoa .mpuyq uozuoq am 333mm an?. pue 08W 1,13qu u; swank): aqusoqd pm? ?LLIms;Smod ?Lumsau?em am smuammp pagwm papaya: 03112 311-3 asaq 'qndu; A. 31131213 :10 >139] [3 ?1213 xiq ?uyoagax ?sanlm eunume .Iaq?yq pure gamma gangs .191?on ({pueagm?gs 151m (nugzo?qg {39?01;ng3) cm 9qu mm; panama.) saldums uI "311mg sq; gnoq?nmq; sum?? 24.191111 quepnmqe 911.] Bugdaax LU ?eugumlv .Iamoy pure 123mg go a?emaalad 1.29143ng 13 my? (swans mxqu pm; gZ?gg?gHg) apeqs 500,1 uaaimaq Saauammp q'q?nq?yq uma put-3 jizmuatugpas 30 aAnqulasaldm aq 04 waddle 1311:1331 SISKIBLIB all pazmmuums an; ?Sapgxa E31 ABOIOpomaw uoyaanog Sundweg pue uogezuasqo - HH xypuaddv Appendix HH - Observation and Sampling Collection Methodology '14 collected in segments that run from the area of the prominent lighthing?damaged tree to the spad 4010 area to assess the potential that electrolytic solutions might occur in fracture zones between these two points. Results of pH. and electrical conductivity tests are Summarized in Table 2. The pH values of all samples, both surface and underground, are nearly neutral, ranging from 6.8 to 7.5. The pH of water samples collected from the surface, from the Buckhannon River and Trubie Run, are consistently 6.8. The pH of water samples collected underground is more alkaline, at 7.1?7.5. Electrical conductivity values are reported in units of microuSiemens per centimeter. A micro-Siemens is the same as a micro-intro, denoting that conductivity reported in ?rnho? units is the inverse of resistivity reported in ?ohms.? Samples collected from Trubie Ru and the northeast-trending tributary of Trubie Run is very similar, ranging from 423?47 9 micro?iernens/ cm. These conductivity values are within a ran ge comparable to that for ?good city drinking water.? Upon entering the Buckhannon River, the electrical conductivity values exhibit a marked increase and nearly double to 83.5 microwSiemens/cm. This value is comparable to a range greater than that of ocean water. Water samples collected underground exhibit a much higher conductivity, with a sample collected from the fire tap at the junction with the new 2Dd Left Main characterized by a value of 196 micro?Siemens cm and the sample of dripping water collected from the track entry Main directly beneath Trubie Run (Figure 1), characterized. by a value of 295 micro-Sien?iens/ cm. The sample of dripping water collected from the right rib of the #8 Entry (W 4064?31) is characterized by a much higher electrical conductivity of 708 micro?iemens/crn. For example, this value is comparable to the electrical conductivity given. by one reference as a 30% solution of nitric acid. It should be noted that Sample was collected from the rib of the #8 Entry near where the cased gas well is located. I Conductivity pH Units: gig/cm pH Detection Limit: 0.01 0.1 Reference Method: ISE pH Jim.? lap 2nd Left; Exhibit "l 196 7?1 Trubie; Exhibit ?1 295 7.2 union?3?1, Exhibit 6 708 7.5? W~Buck; Exhibit ?1 83.5 6.8 Trubie; Exhibit 2 47.9 6.8 Trubie NE trih; Exhibit 3 44.8 6.8 Trubie Main; Exhibit 4 I 42.3 6.8 Table 2. Electrical conductivity values (in micro-Siemens "per centimeter) and pH values for water samples collected underground and from surface streams. Chemical analyses were also conducted on samples Wiltloilmfil and Blrl3986-28 by the Inductively Coupled Plasma?Mass Spectrometry method, to identity approximately trace elements. The most abundant material in both samples was sodium, which occurred in the water droplets collected at locality at a concentration of 447 ppm, and at locality at a concentration of 135 ppm. The next most Appendix HH - Page 14 of 15 Appendix HH - Observation anlgl) Sampling Collection Methodology respectively; silica at 12 and 9 ppm, respectively; potassium at 10.4 and 9 ppm, respectively, magnesium at 2.2 and 0.6 ppm, respectively; aluminum at 2.2 and 0.2 ppm, respectively; and less than 1 of cobalt, zinc, arsenic, and bromine with even lower concentrations: of remaining trace elements. Based on the neutral pH values for the water samples, it would be difficult to consider i-vater in the vicinity of the mine as ?acid mine water.? Based on the electrical conductivity values for the water samples, surface water appears to have little likelihood for representing a good conductor, although the sample from the Buckhannon River exhibited greater conductivity than that of ocean water. The water samples collected underground exhibit much greater values for electrical conductii-rity, with i-vater dripping from, the roof and rib exhibiting the highest conductivity values. The sample with the highest electrical conductivity value was collected from the rib adjacent to the cased. gas well (Figure 9). Based on a review of available literature, the value of 708 micro?Siemens/ cm is comparable to the conductivity of a 30% solution of nitric acid. Because the pH of the sample is shown to be neutral, the high conductivity "value suggests that some other ionic substance is present and has dissociated. Strong acids are considered good electrical conductors because they easily dissociate into charged ions in solution. The presence of high concentrations of sodium may be an indication of a strong ionic solution in this vicinity. ?that the sample with the very highest value of electrical. conductivity is localized in the vicinity of the gas well may be significant, considering that the radial pattern of scour marks is located in the same entry within a few tens of feet, and radiate out from the entry corner beyond which is located the cased gas well (Figure 9). Of all the water samples collected, it appears that Sample W4tl64-31 would be the most likely to represent an electrolytic solution capable of conducting electricity. it should also be noted that during the sample collection effort, a significant amount of water was heard draining through the rib, which is an indication that this is not a solid, impermeable coal rib. If you should have any questions or if we can. be of further assistance, please contact Sandin Phiilipson at 2304-5412015. Appendix HH - Page 15 of 15 Appendix II - Executive Summary of "Submersible Pump Parts Recovered from Sago Mine" Appendix II - Page 1 of 3 Appendix II - Executive Summary of "Submersible Pump Parts Recovered from Sago Mine" Appendix II - Page 2 of 3 Appendix II - Executive Summary of "Submersible Pump Parts Recovered from Sago Mine" Appendix II - Page 3 of 3 Appendix - Sago Mine Pump Cable Test U.S. Department of Labor Mine Safety and Health Administration Pittsburgh Safety Health Technology Center P.O. Box 18233 Pittsburgh. PA 15236 Mine Electrical Systems Division April 4, 2007 MEMORANDUM FOR RICHARD A. GATES District Manager District 11 THROUGH: HocH Chief, Pittsburgh Safety and Health Technology Center FROM: WILLIAM J. LFRICH Chief, Mine Electrical Systems Division SUBJECT: Sago Mine Pump Cable Test Attached is a copy of the subject report which details the testing that was conducted on a section of a pump cable removed from the sealed are of the Sago Mine (ID #46-08791), located in Upshur County, West Virginia. If you have any questions, please contact William Helfrich at (412) 386-6959 or email at Helfrich.william@dol. gov. Attachment cc: R. Phillips, District 2 Appendix - Page 1 of 11 Appendix - Sago Mine Pump Cable Test bcc: M. Skiles, TS w/ attach L. Zeiler, TS w/ attach T. Hoch, PSI-ITC w/ attach R. Stoltz, VENT w/ attach W. Helfrich, MESD w/ attach D. Skorski, MESD attach MESD Files w/ attach T: MESD Lab Files 2007/ Sago Pump Cable Test memodoc Appendix - Page 2 of 11 Appendix - Sago Mine Pump Cable Test UNITED STATES DEPARTMENT OF LABOR MINE SAFETY AND HEALTH ADMINISTRATION PITTSBURGH SAFETY AND HEALTH TECHNOLOGY CENTER MINE ELECTRICAL SYSTEMS DIVISION (MESD) REPORT NO: INVESTIGATION DATE: January 31, 2007 LOCATION: NIOSH Mine Electrical Laboratory Pittsburgh Research Center INVESTIGATORS: William J. Helfrich Chief, MESD Dean P. Skorski Supervisory Electrical Engineer, MESD WITNESSES: Robert L. Phillips Acting District Manager, - District 2 Monte Hieb WV Office of Miners? Health, Safety, 8: Training John Hall WV Office of Miners? Health, Safety, Training William Hutchens Attorney Jackson Kelly SUBJECT: Inspection and Testing of Sago Mine Pump Cable INTRODUCTION On December 4, 2006, the Mine Electrical Systems Division (MESD) was requested to obtain a section of a pump cable from the sealed area of the Sago Mine. Arrangements were made with Coal Mine Safety and Health District 3 and the mine operator to retrieve the cable on December 7, 2006. MESD took possession of the cable on that day and has securely maintained the section of cable. A water bath testing of the cable, which provides information on the dielectric strength of the conductor insulation and jacket, was conducted on January 31, 2007. Appendix - Page 3 of 11 Appendix - Sago Mine Pump Cable Test 2 CABLE TESTING PROCEDURES The following cable tests used the test procedures in NEMA ICEA 5-95?658- 1999, titled ?Ethylene-Propylene-Rubber Insulated Wire and Cable - Nonshielded 0-2kV Cables,? Section 6.10.1 - ?Voltage Tests? as a guideline. General These tests consisted of voltage tests on a Tiger Brand Cable, AWG 3-conductor, type G-GC, 2000 volt cable. Each conductor was tested separately, and a voltage was applied between individual conductors and the grounded water tank. The multiple conductor cable was immersed in water for at least 12 hours and tested while still immersed. Resistance Test After the cable has been immersed in water for the determined time, a resistance test was made between each conductor and the tank frame, as well as between each of the conductors. A continuity measurement was also made on each conductor in the cable. Alternating Current Voltage Test This test was made with an alternating potential from a transformer of ample capacity, which was in no case less than 3 kilo-volt?amperes. The frequency of the test voltage was nominally between 49 and 61 hertz, and had a wave shape approximating a sine wave as closely as possible. Single-phase voltage was applied separately across each of the four (4) insulated cable conductors and the two (2) ground conductors, and tank frame. The order was red, white, black, yellow (ground check), ground 1, and ground 2. The tank frame was grounded to the power system ground. The rate of increase from initial applied voltage of zero (or previous test voltage) to each following applied voltage for the specified test voltage was approximately uniform and was not more than 100 percent in 10 seconds nor less than 100 percent in 60 seconds. Appendix - Page 4 of 11 Appendix - Sago Mine Pump Cable Test 3 The duration of the alternating-current test voltage to the immersed cable was as follows for each conductor: Voltage Duration 100 volts 5 minutes 200 volts 5 minutes 300 volts 5 minutes 400 volts 5 minutes 500 volts 5 minutes 600 volts 5 minutes 700 volts 5 minutes 800 volts 5 minutes 900 volts 5 minutes 1000 volts 5 minutes 1200 volts 5 minutes 1400 volts 5 minutes 1600 volts 5 minutes 1800 volts 5 minutes 2000 volts - 5 minutes (maximum rating of cable) End of Test Measurements were made of voltage and current flow in the ground path for each of the voltage levels. If a current of a least 100 milli-amperes was detected at any voltage level, the test was terminated, and NO additional testing was conducted on that conductor. It was possible that a catastrophic failure of the conductors in the cable could occur during this testing. The power supply was provided with ground fault tripping, which was not adjustable and was approximately 5.4 amperes. If a current of a least 100 milli-amperes was not detected through the 2000-volt test level, NO additional testing was conducted on that conductor individually as it had passed this part of the test. Test Setup (see Figure 1 below) Electrical testing was conducted in a metal tank, with each side and depth of approximately 3 feet. The tank was grounded to the power system ground and filled with tap water. The test cable was immersed in tank water, which was maintained at room temperature for at least 12 hours, and the water level was kept constant. Appendix - Page 5 of 11 Appendix - Sago Mine Pump Cable Test 4 The length of the test cable was approximately 190 feet. The cable was rolled onto a fabricated reel to permit the cable to fit into the tank in its entirety. The outer jacket of the first 6 inches of each end of the cable was stripped away. The three-phase conductors, ground check wire, and two ground wires were isolated from each other on each end. Approximately 1 inch of conductor insulation was removed from each insulated conductor (one end only) to facilitate connection of test leads. A 1-foot portion (minimum) of each end of the cable was kept above water as leakage insulation. A diagram of the test setup can be found at the end of this document (Figure 1). Test Procedures Resistance Test (No Power/ Conducted with a Fluke Model 87 VOM) 1. Visual observation of the test cable. 2. Measure continuity of each conductor in cable (see results in Table 1 below). Connect test leads to cable and tank. 4. Measure resistance of each conductor in the cable. a. Each conductor to the grounded water tank (see results in Table 2 below). b. Each conductor to all the other conductors (see results in Table 3 below). 9? I Table 1. - Continuity Measurements of Each Conductor Red White Black Ground Groundl Ground2 Check 0.1 ohms 0.1 ohms 0.1 ohms 0.2 ohms 0.5 ohms 0.5 ohms Appendix - Page 6 of 11 Appendix - Sago Mine Pump Cable Test 5 Table 2. - Resistance Measurements of Conductors in Cable (measurements recorded in meg-ohms) Red White Black Ground Ground 1 Ground 2 Check 1.351 6.4 1.304 1.889 1.014 1.014 Tests were conducted with one lead of ohm? meter connected to each conductor and the other lead connected to the metal tank. Table 3. - Resistance Measurements between Conductors in Cable Red White Black (21:33:! Ground 1 szund Red 5 3.45M 78K 0.945M 31 .2K 31 .8K White I 9.65M 12M 8.74M 8.8M Black i 0.968M 40K 40K - I I. 29.7K 30.5K Groundl . I 0.3K Ground2 designates Meg?ohms (millions) designates Kilo-ohms (thousands) Alternating Current Test Procedures (results are shown in Table 4 below) (.4 Visual observation of the test cable (no power). Connect test leads to cable and tank in the order described earlier (red, white, black, ground check). Remove all observers from test area. Engage main power switch. Switch power source on. Increase voltage to 100 volts at predetermined rate. Hold voltage constant for 5 minutes time. If current ?ow was below 100 milli-amperes, increase voltage to 200 volts at predetermined rate. 9. Hold voltage constant for 5 minutes time. 90519.9?th Appendix - Page 7 of 11 10. 11. 12. 13. 14. 15. 16. 17. 18. Appendix - Sago Mine Pump Cable Test 6 If current flow was below 100 milli-amperes, increase voltage to 300 volts at predetermined rate. Hold voltage constant for 5 minutes time. If current flow was below 100 milli-amperes, continue increasing the voltage in 100?volt increments (or 200-volt increments, see table below) up to the 2000 volt limit. NOTE: If current flow exceeded 100 milli-amperes at any voltage level, testing was over for that conductor. Decrease test voltage to zero. Switch power source off. Keep key on person. Bleed off any static charge on water tank with shorting stick. Hang shorting stick on tank. Connect test lead to different conductor. Remove shorting stick. Repeat from Step 5. Table 4. - Cable Testing Results (individual conductor to grounded tank frame) Current Level Through Conductors (milli-amperes) Voltage Level (volts) Ground Red White Black Check Ground 1 Ground 2 100 6 7 24V 107mA 24V 206mA 200 1000 94 41 36 1200 125 78 41 1400 116 100 40 1600 127 4310 1800 25011 03 GFT 2000 Notes: U?lr-PmNr?i ammeter briefly read over 100 milli-amps 3 times during test ammeter brie?y fluctuated over 100 milli-amps 2 to 3 times ammeter read over 2 amps twice during test ammeter read over 2 amps twice during test ammeter read over 100 milli-amps numerous times during test Appendix - Page 8 of 11 Appendix - Sago Mine Pump Cable Test 7 6 - ammeter ?uctuated number times over 500 milli-amps and once over 1 amp 7 - ammeter read over 4 amps twice during test 8 - initially, ammeter read over 100 milli-amps 9 - ammeter averaged over 100 milli-amps for duration of test 10 - ammeter ?uctuated over 100 milli-amps several times during test 11 ammeter averaged over 250 milli-amps at beginning of test GFT - ground fault trip on power supply (current level at least 5.4 amps) The white phase conductor passed the testing at the voltage rating of the cable (2000 volts). The investigators decided to test this cable further to determine the voltage level at which it would fail. The additional testing procedures followed the previous procedures with the exception of the duration at each voltage level. A 1-minute hold at each voltage level would be used to determine the level of voltage for the white conductor to meet the failure criteria. The following table illustrates the results of this additional testing (Table 5). Table 5. - Cable Testing Results (voltages above the 2000-volt rating of the cable under test) Current level through white phase conductor (milli?amps) Voltage Ground Level Red White Black Ground 1 Ground 2 (volts) Check 2500 18 3000 21 ,5 . 3500 24 4000 27 4500 30 5000 36 5500 GFT Appendix - Page 9 of 11 Appendix - Sago Mine Pump Cable Test 8 METAL TANK T0 -- @rk? 0?1 20V STEP-UP AC SOURCE TRANSFORMER I 'l 2937 Figure 1. - Test Set-Up VISUAL OBSERVATIONS Following the removal of the cable from the water bath testing, the cable was allowed to dry. The cable was then removed from the cable reel and all damaged areas and splices were documented. The table below (Table 6) provides the locations of damaged areas and splices with respect to the coupler (cat head) end of the cable. The measured length of the cable was 192.5 feet. Appendix - Page 10 of 11 Appendix - Sago Mine Pump Cable Test 9 Table 6. - Visual Observations of Pump Cable (all measurements (in feet) are from coupler end of cable) Spliced Sections (included Damaged Sections permanent and . . Taped (includes temporary . slices, cracks, etc.) splices) semons Temporary P: Permanent 11.5 27 87.3 17.7 37.7 46 132.8 33.5 65.9 87.1 146.7 93.4 94.7 182.9 96.9 99 100.3 105.4 108.1 108.4 110.9 111.7 112.1 114.9 120.4 121 128.1 131.3 132.1 135.2 138.7 139.5 141.7 144.6 150.7 162.4 168.8 171.9 174.3 180.3 185.7 186.5 186.9 187.8 SUMMARY The water bath testing of the pump cable revealed that the insulation on three of the four insulated conductors in the cable (red, black, and ground check) failed prior to the test voltage reaching the rated voltage of the cable (2000 volts). The white conductor reached a voltage of approximately 5,500 volts before a failure of the insulation occurred. The design of the cable did not provide insulation for the 2 ground conductors. As expected, they both failed at a very low voltage. There were numerous locations on the cable jacket where significant damage was apparent. There were three permanent splices and one temporary splice in the section of cable tested. Based on the overall condition of this relatively short section of cable, the test results confirmed the anticipated failure of the insulation surrounding the conductors. Appendix - Page 11 of 11 Appendix KK Sago Mine, MSHA ID 46-08791 Wolf Run Mining Company Earth Resistance Measurement Values Appendix LL Sago Mine, MSHA ID 46-08791 Wolf Run Mining Company Mine Map Detailing the Extent of Flame and the Direction of the Primary Explosion Forces