\ '^^.^'< *bv" •^^.^ ^<*. .*^''-^*. .• .*^'*. \ "9- -4. 'fO' ■>.•.;■.• .«■ f--^ .*'°-' V\» % %.«* •>o^ 'bV -^^0^ ^^c,^^ 1^ /j^ o.-^' '•^.* >; ^-.c,^^ lA^ ,^^ "^. V ,^^ ^^ .♦^"^. -...• ,0* \ • :..>' ,s_ ''tr i'i' ♦J J> \ ..„ "ol-' / ^*'*- •I o .* r>' . •.;»• v^ A ... •«- 4 * <^ %..<,^^ * v/'V *• ^^ v^ (5°^ : o*. '.i ^Ho. •;♦ j'^ "V "" ^s" <- '^vT* .&«" % *-».s*° .o*".v..\ .'*X PV **o--T^r'\** \--p?.\/ \--?^-'y% -.^r/ /\ •.^^.' **^ ^ •^"°"- ^--^^^^X' /°*^-V' /t-^S. /^'--So''^^^'^^^^^^^ '.■ .*""*. ** /\ .•^ : .«""*. -. -^^0^ Ao^ IC ®®^^ Bureau of Mines Information Circular/1982 ) > Underground Coal Mine Power Systems Proceedings: Bureau of Mines Technology Transfer Seminar, Pittsburgh, Pa., September 16, 1982 Compiled by Staff-Bureau of Mines UNITED STATES DEPARTMENT OF THE INTERIOR Information Circular 8893 Underground Coal Mine Power Systems Proceedings: Bureau of Mines Technology Transfer Seminar, Pittsburgh, Pa., September 16, 1982 Compiled by Staff-Bureau of Mines UNITED STATES DEPARTMENT OF THE INTERIOR James G. Watt, Secretary BUREAU OF MINES Robert C. Norton, Director ^\\ A^ <^^13 This publication has been cataloged as follows: Bureau of Mines T( chnology T ran- fer Seminars (1982 : Pittsburgh, Pa.) Underground coal mine power syste TIS. (Information circular ; 8893) Includes bibliographi cal references. Supt. of Docs, no.: I 28.27:889 3. 1. Coalmines and min 2. Electricity in mining Mines. II. Title. III. Bureau of Mines) ; 8893. ing — United States— Power Supply — Congresses. — (Congresses. 1. United States. Bureau of Series: Information circular (United States. TN295.U4 [TN343] 622s [622' .48] 82-600227 f^ PREFACE This Information Circular sununarizes recent Bureau of Mines research results con- cerning improved mine electrical power systems for America's underground coal mines. The papers are only a sample of the Bureau's total effort to improve coal mine safe- ty, but they do delineate the major concerns of the mine electrical power programs. Some of the technology discussed has applications in other types of mining. The six technical presentations reproduced here were made either by Bureau ^"Researchers or by personnel representing Bureau contractors at the Technology Trans- fer Seminar on Underground Coal Mine Power Systems given in September 1982 in Pitts- burgh, Pa. The content of the other presentation not included here can be found in the Bureau of Mines Handbook, "Application Notes — ^Mine Electric Power Systems." \ Those desiring more information on the Bureau's mine electrical power safety programs in general, or information on specific situations, should feel free to contact the Bureau of Mines Division of Health and Safety Technology, 2401 E Street, NW, Washington, D.C., 20241, or the appropriate author. ^ Was] ^^ ^b^O CONTENTS iii Page Preface 1 Abs tract 1 Introduction 2 Design Practices To Minimize the Probability of Shock During Control Box Maintenance, by Thomas Novak and George J. Conroy 4 A Design Guide for Explosion-Proof Electrical Enclosures, by P. A. Cox and Lawrence W. Scott 11 High-Voltage, Explosion-Proof Load Centers, by George Conroy, Randy Berry, and Robert Gillenwater 29 Demonstration of the Discriminating Circuit Breaker (DISCB), by Michael R. Yenchek 55 Intermittent Duty Rating of Trailing Cables, by George J. Conroy and Herman W. Hill 74 Semiconducting Rubber as a Low-Voltage Shield for Personnel Protection, by J. N. Tomlinson and L. A. Morley 82 UNDERGROUND COAL MINE POWER SYSTEMS Proceedings: Bureau of Mines Technology Transfer Seminar, Pittsburgh, Pa., September 16, 1982 Compiled by Staff-Bureau of Mines ABSTRACT This Bureau of Mines publication presents an overview of mine electri- cal power systems research currently being conducted by the Bureau. The papers, given at a Technology Transfer Seminar, emphasize the increas- ing importance of research related to the safety considerations of underground coal mine electrical systems. Selected topics are included that summarize results of research in the areas of electrical shock prevention, explosion-proof enclosures and load centers, discriminating circuit breakers, and trailing cables. Other topics discussed at the seminar are published as separate sections of the Bureau of Mines Hand- book "Application Notes — Mine Electrical Power Systems." INTRODUCTION Mine power systems (MPS) research encompasses the following areas: per- missible equipment, power distribution, dc power systems, cable usage problems, and intrinsic safety. In general terms, MPS research attempts to identify and attack electrical problems which pose a significant hazard to the safety of the individual miner. This is accomplished by one of several methods. In studying shock prevention, for exam- ple, the problem must be correctly iden- tified. To this end. Mine Safety and Health Administration (MSHA) accident statistics were studied to build a history of electrical accidents both fatal and nonfatal; then innovative but practical solutions were proposed and evaluated. Some of these solutions are contained in the first paper. Shock pre- vention research will continue to receive a high priority in future MPS research. A recent MSHA publication^ attributes 60 pet of the mine fires investigated through 1972 and 21 percent of all meth- ane ignitions to electrical ignition sources and inadequate design of electri- cal equipment. A Bureau study2 of elec- trical violations cited by MSHA indicated that the second most common electrical violation was failure to comply with reg- ulations for permissible electrical face equipment maintenance. In fact, the improper use of permissible equipment accounted for 17 pet of the total elec- trical violations studied. The study of permissible equipment and enclosures is an area requiring investigation into materials, methods of construction, and testing criteria. At present, the explosion-proof enclo- sures used in underground mines are constructed to rigid design requirements that contribute to the difficulties in ""msHA IR-1018, "Electrical Hazards in Underground Bituminous Coal Mines," 1975, 5 pp. ^BuMines IC 8725, "Mine Inspection Rec- ords Study," 1976, 11 pp. maintaining the enclosures in a permissi- ble condition. If designers were per- mitted more freedom, it would be possible to construct enclosures that are easier to maintain in a permissible condition, and to eliminate aluminum alloys if they are deemed a safety hazard. These new enclosures would be submitted to perform- ance tests to insure that they offer the same degree of safety as the enclosures presently being constructed to the design requirements. It is the objective of this research area to determine the safety factors in the present design requirements, and to develop and demon- strate the feasibility of performance tests. The second paper deals with the results to date of this research area. Another paper deals with the need for ever-increasing power requirements of larger and larger face equipment. The paper on explosion-proof load centers presents tentative specifications and a request for interested persons to comment . At least 80 of the 127 coal mine fires involving the trolley system, which were investigated by Federal personnel from 1952 to 1977, could have been prevented if circuit protection systems capable of responding to low-current arcing faults had been available and universally in- stalled. If one considers only non- reportable fault conditions on haulage- ways, the annual worth of a low-level fault protection scheme is estimated at $21,000 to $52,000 per mine. Thus, even on a cost-saving basis, without mention- ing saved lives, sensitive trolley wire protection can be shown to be of value to mine operators. This does, however, require physical demonstration that the method performs as required. To meet this objective, the Bureau sponsored research that led to the development of the discriminating circuit breaker (DISCB) described in the fourth paper. This device may find future applications in electrified mass transit systems. The DISCB can distinguish between low level ground fault currents and legitimate high-current loads on a mine haulageway, responding (instantaneously) only to the former. It can be employed in any exist- ing haulageway with a minimum amount of modification of the haulage equipment, A scaled demonstration is described in the paper. Cables contribute more to downtime and to personnel injury than any other electrical component used in mines. Accident classification by injured activity from 1975 to 1979 has shown that in 19 pet of all accidents contact with energized cables occurs while splicing a supposedly dead line. Proper ratings, construction, and application of cables is very important. The last two papers discuss these subjects in more detail. The increased use of electrical power in the mining environment dictates that research into specific mine electrical hazards be focused on those areas which will most significantly reduce fatal and nonfatal accidents among miners. MPS research will continue to identify elec- trical hazards and propose innovative, yet practical, solutions to alleviate those hazards. DESIGN PRACTICES TO MINIMIZE THE PROBABILITY OF SHOCK DURING CONTROL BOX MAINTENANCE By Thomas Novak! and George J. Conroy2 ABSTRACT The use of dead-front construction and other means of segregating higher-voltage portions of control box circuitry with carefully planned use of test points, an in-place schematic diagram, and indi- cator lights, are explained in this paper. Sensitive circuitry for personnel protection is discussed. The reception by manufacturers and mine personnel of the ideas as embodied in a demonstrator unit is summarized. Recommendations to improve maintenance safety are listed and discussed. INTRODUCTION Mine Safety and Health Administration (MSHA) accident statistics for both coal and metal-nonmetal mines show that main- tenance personnel have the highest fre- quency of electrical accidents. These statistics might be expected since re- pairworkers have the greatest exposure to electrical hazards. Yet electrical- safety features of equipment have been directed primarily toward protecting the electrical wiring, the machine, and the mine from heat and fire damage. Consid- ering the evidence, present emphasis should be placed on protection of mainte- nance personnel. The harsh operating conditions common to most mining processes contribute greatly to the electric shock hazard. As mineral or rock is extracted, the elec- trical machines must advance, followed by their sources of power. During these moves, both equipment and cables are fre- quently stressed by pulling over rough surfaces and being twisted or impacted. As all those who work in mines can attest, gentle handling of equipment is not the rule but the exception. The extreme abuse increases the amount of "• Professor of electrical engineering, Pennsylvania State University, Fayette Campus, Uniontown, Pa. ^Supervisory electrical engineer (re- tired). Bureau of Mines, Pittsburgh Re- search Center, Pittsburgh, Pa. maintenance while at the same time it renders insulating qualities, etc. , ques- tionable, all of which increases the repairworker's exposure to electric shock. The environmental conditions of the mine also enhance the probability of electrical accidents. Wet conditions are often encountered which decrease the con- tact resistance between a person and earth. Thus, a person becomes more sus- ceptible to electrocution from contacting an energized conductor. Space on mining machines is extremely limited, and electric control boxes and panels are crowded with parts. Trouble- shooting with the circuit energized is permitted under present regulations, leading to a distinct probability that, either through ignorance, inattention, or an Inadvertent slip, an elbow or hand will contact an energized member. If normal production is interfered with by a breakdown, which is frequently the case, the repairworker will probably be rushed. Especially in mines where roof and rib might be unstable because the machine cannot do its roof-bolting job, the cramped working conditions and the foreman's frequent encouragement to com- plete the repair, can create sufficient confusion to add to existing potential hazards. There is not much hope that the working conditions will change; nevertheless, the accident statistics can be improved. Electric shock prevention can be afforded to the maintenance worker as follows: 1. Minimizing exposure to energized members. 2. Providing sensitive, rapid-response electrical protective devices. Both of these approaches will be discussed. The first approach can be implemented by having the machine manufacturer or, in some cases, the mine maintenance group itself, incorporate certain physical- mechanical design modifications that are simple in concept but so beneficial in their consequences that their added first cost can easily be shown to be balanced by reduced maintenance costs and improved safety. The second approach is, in some cases, only now becoming fully realiz- able, with the entry into the mining field of rugged, reliable solid-state devices. PHYSICAL DESIGN MODIFICATIONS The design concepts to be considered in this paper are dead-front construction, interlocks, segregation of high-voltage circuits, and lockouts. They may be employed singly or in combination, with the objective of avoiding inadvertent contact with energized members by intro- ducing elements which, while restraining from contact, facilitate or at least do not significantly hinder rapid trouble- shooting. The designer should also give consideration to secondary aspects of protection, as explained when discussing interlocks which remain, even when the primary protective elements are bypassed. Dead-Front Construction It is possible to perform a large num- ber of troubleshooting operations with- out making a manual approach of any kind to the vicinity of the high voltage (600 V, 480 V, etc.) portions of a con- trol circuit, if test points are supplied on a panel that is interposed as a phys- ical barrier between the circuit and the external world. For application to explosion-proof enclosures, a hinged pan- el can be installed in a position slight- ly recessed behind the outer cover of the enclosure. The test points are installed to penetrate the surface of the hinged panel to allow important voltage measure- ments. The concept requires little or no increase in overall size of the control case. The dead-front, troubleshooting concept can be applied to almost any type of electrical-control case. A simplified unit was constructed to demonstrate the concept and is shown in figure 1. The internal components comprise a simple, two-motor starting circuit. Since this prototype was built for use as a demon- stration unit, the electrical con^jonents are housed in a portable, lightweight, aluminum enclosure rather than a heavy, explosion-proof box; however, the ulti- mate usage was kept in mind at all times. The unit is designed to simplify troubleshooting procedures, as well as to reduce the hazards of electrical shock. Figure 1 shows the dead-front panel with the outer lid of the enclosure removed. The incorporation of test points into an actual diagram of the circuit schematic is an important aspect. This feature could greatly simplify troubleshooting techniques and possibly reduce machine downtime. In addition, the schematic provided on the panel is more permanent than paper schematics, which deteriorate quickly in the moist mine atmosphere and are susceptible to being mislaid and lost. The hinged panel provides a com- plete barrier between the repairworker and the electrical components. The test points are pin jacks and therefore re- strict access to dangerous voltages re- quiring the use of test probes. A key is required to gain access to the internal FIGURE 1. - Dead-front panel with outer lid of enclosure removed. FIGURE 2. - Panel as removed after internal electrical controls are unlocked. electrical controls. After being un- locked, the panel can be opened, as shown in figure 2; however, opening the panel results in the operation of two interlock switches which are in series with the undervoltage release of the circuit breaker. The opened interlock switches interrupt power to the undervoltage re- lease, tripping the circuit breakers. Thus, all internal components, with the exception of the incoming leads of the circuit breaker, are deenergized. A guard with a warning light is placed around the incoming connectors of the breaker to prevent accidental contact should the repairworker fail to deener- gize the incoming line. On an actual ac machine used in coal mining, the inter- lock switches would usually be connected in series with the pilot circuit of the ground-check system rather than to an internal circuit breaker. Opening the panel would trip the upstream circuit breaker, located in the section power center; thus all power to the control case would be shut off. In that appli- cation, the warning light would be re- tained, whether or not the internal cir- cuit breaker was included. Many variations of the basic dead- front concept can be applied to control centers. One possibility is to use a dead-front panel made of transparent plastic, such as polycarbonate. Since the electrical controls are located directly behind the panel, the physical movement of line starters, relays, and contactors can be observed while voltage measurements are made at the isolated test points. The transparent panel can be so constructed that permissibility is not compromised, with only intrinsically safe test points used if any are needed. With this design, a relatively thin, steel-hinged outer cover protects the plastic panel from most sources of scratching or other damage. Whether or not the transparency of an unprotected panel can be maintained under actual operating conditions depends upon the in- dividual application. A further refinement of the dead-front concept is to replace or supplement the test points with visual indicators, such as LED's or lamps. In effect, the built- in indicators would serve in place of a voltage tester for many routine trouble- shooting observations, and a quick assessment of the condition of the con- trol circuit could be accomplished simply by visual inspection of the panel. Testing and changing control-circuit fuses are among the most frequent electrical-maintenance procedures. For a dead-front approach, the fuses can be mounted behind a separately interlocked explosion-proof cover. The interlock would operate an internal contactor which, because of its very low duty cycle with regard to circuit interruption, could be relatively small. The fuses could thus be removed and replaced with no risk of having personnel contact ener- gized metal clips. A refinement of this idea is to include test points under the interlocked cover connected to each end of each fuse. With the circuit opened by the interlock, fuse continuity can be checked with an ohmmeter without removing the fuse. Interlock Disable Precautions Interlock design should always be ori- ented toward prevention of accidental reclosure of the interlock. Where the interlock is a simple jump-link arrange- ment, the energized contacts should be female, recessed in insulating plastic. If a sensitive switch is used, the pres- sure to close the switch should be ap- plied by a protrusion operating through a hole in the panel behind which the switch is mounted. Any safety feature can be circumvented by a determined maintenance worker if it interferes with his or her troubleshooting function. Since some interference by an interlock is unavoida- ble, the means to circumvent the device should be thought out beforehand by the designer and not left to be jury-rigged in a manner that permanently damages or disables the interlock. In its most desirable form, the disabling means will automatically clear once the necessity for its use is over. An example would be a latched interlock switch arranged so that cover closure results in mechan- ically resetting of the latch. Also, a highly visible warning should be included telling when an interlock has been dis- abled, to remind the repairworker that circuits remain energized. point on a terminal strip must be num- bered to correspond with the circuit's schematic, and all con^jonents within the case should be labeled for easy identifi- cation. The test points should also be segregated according to their voltages. In other words, control-circuit test points of 120 V should be located in an area away from the power-circuit test points of 440, 550, or 950 V. The re- pairworker would then know the approxi- mate level of the voltage with which he or she is dealing. The color coding of conductors might also be helpful in dis- tinguishing different voltage levels. The main disadvantage of using isolated test points or voltage indicators for troubleshooting is that the amount of wiring is increased. The additional wir- ing increases as the size of the circuit increases. With very large circuits it may not be feasible to isolate all possi- ble test points. A circuit analysis would have to be performed to determine which test points are the most critical. The critical test points would then be brought out to the dead-front panel or terminal strips. The preceding discussions primarily deal with machine circuitry. However, the concepts are not limited to this application. Dead-front, troubleshoot- ing panels can also be applied to the low- and medium-voltage circuits of power-distribution equipment in a similar fashion. Lockout Features The following example illustrates a maintenance-related electrical accident that has occurred numerous times: Segregation of Different Voltages For some controls, isolation of the en- tire circuit with a physical barrier may not be possible or desirable. One means of improving electrical safety in these situations is to include one or more ter- minal strips for use as test points. The terminal strips should be located such that accidental contact with energized conductors would be minimized. Each test A repairworker deenergizes the faulty circuit at the power center prior to per- forming repair work. While he or she works on the faulty circuit, another worker, who is unaware of the situation or mistakes the cable of the faulty cir- cuit for another cable, energizes the faulty circuit and thus subjects the repairworker to electrical shock or electrocution. Mine power center manufacturers have made available lockout features on cable couplers to prevent this situation. Locking-type dust covers on equipment- mounted receptacles along with keyed couplers are supplied as a means of re- ducing the possibility of this hazard. With keyed couplers, the plug of each outgoing circuit is matched to fit only one receptacle. The dust covers are con- nected to the coupler or the equipment by a chain in order to prevent their loss while not in use. A loosely hinged cover would serve the same purpose with in- creased operational facility and greater assurance against damage. Locking-type dust covers for the cable- mounted plug can afford even more protec- tion for maintenance personnel. Once the plug is locked, connection to any recep- tacle is impossible. Chain-connected, locking-type dust covers are presently available for high voltage plugs. How- ever, consideration should be given to the development of hinged dust covers for all plugs energized at higher than 30 or 40 V. Electrical lockout features utilizing the ground monitor circuit can also be incorporated to give the mechanic working at the machine the ability to prevent reestablishment of power from the load center, until he or she is fully ready to allow it. A typical circuit would employ an interval timer, a small relay, and a reset switch. When lockout features — and strict procedures for their employment — are combined with other cable and machine- related safety provisions, a very signif- icant improvement in the count of fatali- ties and injuries may be observed. Sensitive Electrical Protection This subject will be summarized here rather than explored in detail, as a very thorough treatment will become available in the final report of a Bureau contract (J0113009) during November 1982. Many aspects of the foregoing discussion are also included in the report. The present discussion will center on the physiologi- cal basis for relying on sensitive elec- trical protection and will simply mention a few of the findings of that study and other research projects. Electric Shock Threshold Ventricular fibrillation is usually the most dangerous shock hazard, as it occurs at relatively low values of current. Disabling or fatal burns require larger currents, and cardiac arrest due to high current flow can, in some cases, be coun- tered by appropriate physical action by another person after the current has been turned off. When fibrillation occurs the pumping motions of the heart become dis- organized and, finally, the pulse ceases. Death occurs within minutes. Dalziel (_1_)3 presented an equation extrapolating statistical data obtained from experi- ments on animals, which predicts the min- imum threshold for fibrillation for a body weight of 110 lb (50 kg) at a power- line frequency of 60 Hz I = 8.3 msec < t < 5 sec, ft - - where I = minimum current in milli- amperes through major extremities. and t = duration of the shock, in seconds. The equation is represented graphically by the solid curve of figure 3. Accord- ing to Dalziel (J_) , the area underneath and to the left of the solid curve is considered the "safe area" with respect to fibrillation. Although it is impossible to eliminate fatal reactions to electrical shock for all cases, a reasonable degree of safety can be achieved by limiting the maximum ground-fault current to a low value (say 500 mA) and by matching the characteris- tics of a ground-fault relay to the ■^Underlined numbers in parentheses refer to items in the list of references at the end of this paper. 1 r- 1 1 - Fibrillation curve - \ Operating characteristic for a definite-time relay -\ Fibrillation Pick- up Safey \area "'x,^^ Operating time V ^^ ^ ^ 100 200 300 400 500 CURRENT, mA FIGURE 3, - Fibrillation curve. fibrillation curve. If the Dalziel equa- tion is accepted, a definite-time relay with an operating characteristic similar to the dashed line of figure 3 would be required to meet the necessary sensitiv- ity. The pickup current of 50 mA was selected since this value is generally considered the minimum threshold of fib- rillation. The operating time of 0.1 sec is based on worst-case conditions and takes into account system capacitance and the reaction time of molded-case circuit breakers. Ventricular fibrillation can also be caused by direct current, but related re- search has not been nearly as con5)rehen- sive when compared with alternating cur- rents. However, Daziel (O indicates that the fibrillation level for dc cur- rents is approximately five times the threshold of ac currents. Therefore, the pickup current of 250 mA at an operating time of 0.1 sec is suggested for dc applications. Ground Fault Current Interrupters (GFCI) The GFCI concept was originally devel- oped by Fuchs-Westinghouse, Ltd. 4 in South Africa, for use in residential wir- ing, and is presently commercially avail- able from a number of sources in the United States. It is being increasingly incorporated into single-phase systems in both residences and industrial buildings. The system depends on having a toroid with a "square" hysteresis loop, through which all power conductors pass, so that the flux created by the current out the one conductor can be balanced by the flux generated by return current in the other conductor. Any unbalance indicates the presence of a leakage current through the earth or external grounding circuits and if the output of a secondary winding on the toroid is utilized to trip a sensi- tive relay circuit, leakage currents as small as 5 mA will result in circuit interruption. In the investigation conducted under Bureau of Mines contract JOl 13009, no commercially manufactured GFCI was found for the high-current applications pecu- liar to mining. Generally, the highly sensitive response to fault currents was accompanied by a similarly high sensitiv- ity to electrical "noise" pulses con- ducted through the mine wiring or trans- mitted through space. Frequent nuisance tripping resulted, which would make the devices intolerable to operating person- nel. It is possible, however, to include devices of this type in control boxes, to be enabled when the outer cover is opened but bypassed in normal operation. In this way, the maintenance worker would have highly sensitive protection at the expense of susceptibility to nuisance '^Reference to specific manufacturers does not imply endorsement by the Bureau of Mines. 10 tripping, whereas the noise immunity of the machine would not be conqjromised dur- ing routine operation. Sensitive Earth Leakage (SEL) System The SEL system has been used for over 10 years in the United Kingdom, and has been recently introduced into the U.S. mining operations by the Consolidation Coal Co., a subsidiary of DuPont Indus- tries {2). Full details of the funda- mental circuit have not been revealed. For U.S. mines, the British circuit was modified by adding an input filter, shielding the current transformer, and providing a short time delay, all to min- imize nuisance tripping. The modified system limits the maximum ground-fault current to 500 mA, while the relay picks up at 140 mA., operating about 0.17 sec (10 cycles of 60-Hz current) after initi- ation. While these values of fault cur- rent and time are not yet comparable to the safe levels indicated in figure 3, there is a strong promise that future improvements will result in a relaying system that matches the desired charac- teristic. Meanwhile, the use of the sys- tem improves the probability that perma- nent damage will not result from contact with energized members. CONCLUSION MSHA statistics show that maintenance personnel are the primary victims of electrical accidents in the mining in- dustry. This high incidence can be attributed to the worker's frequent ex- posure to electrical hazards. Reduced electrical exposure, through the use of dead-front construction and other physi- cal barriers between the control circuit and the troubleshooter, in combination with test points, interlocks, and circuit lockout features, can play a significant role in decreasing electrical accidents. Sensitive ground-fault relaying, which is capable of detecting and isolating a ground-fault prior to electrocution, should be developed to provide protection if preventive measures fail. Dalziel's (1) ventricular fibrillation curve could define the ultimate time-current char- acteristic of such a relaying system. It is hoped that the illustrated examples provided in this paper serve as a start- ing point in a campaign to enhance elec- trical safety and that new approaches suggest themselves as these few sugges- tions are tried. REFERENCES 1. Dalziel, D. F. Electric Shock 2. Dolinar, K. D. Improved Ground Hazard. IEEE Spectrum, February 1973, Fault Protection System for Low and pp. 44-50. Medium Voltage Trailing Cables. IAS of lEEEProc. , 1982, pp. 122-127. 11 A DESIGN GUIDE FOR EXPLOSION-PROOF ELECTRICAL ENCLOSURES By P. A. Cox"! and Lawrence W. Scott2 ABSTRACT A guide Is being prepared for the design of explosion-proof electrical enclosures. It will address those as- pects of design which affect enclosure strength and ruggedness. Material selec- tion will be covered only for materials such as polycarbonates, adhesives, and sealants which can be seriously degraded by the mine environment. The guide will not address electrical function, nor will it duplicate or supplant the re- quirements set forth in the Code of Fed- eral Regulations (CFR), Title 30, Part 18 (Schedule 2G). If followed, the design guide should assure that the strength and ruggedness requirements of Schedule 2G are met. INTRODUCTION This paper represents a status report on the preparation of a design guide for explosion-proof electrical enclosures. It is proposed that the guide be pub- lished under the auspices of the Federal Bureau of Mines and that it be made available to both the Mine Safety and Health Administration (MSHA) and enclo- sure manufacturers. Use of the design guide will be optional. Its purpose is to provide guidelines for the design of explosion-proof enclosures which, if fol- lowed, will assure that the enclosure will both pass MSHA certification testing and fulfill its intended service life underground. The guide will cover only those aspects of design that affect en- closure strength, ruggedness, and service life. It will not address electrical function such as connector design, light- ing, electrical leads, or electrical pen- etrations. An attempt has been made to keep the guide general, rather than spe- cific, so that it will continue to be useful as MSHA requirements evolve. For example, ruggedness is not now a specific requirement in Schedule 2G. Ruggedness ^Senior research engineer, Southwest Research Institute, San Antonio, Tex. ^Technical project officer, Pitts- burgh Research Center, Bureau of Mines, Pittsburgh, Pa. has been achieved in the past by minimum thickness requirements and by design to withstand internal explosions of methane and air; however, future regulations may eliminate minimum thickness requirements and also permit the venting of enclosures to reduce internal overpressures. If this occurs, a ruggedness requirement may be imposed to insure that the enclosure will withstand the mine environment. Thus, a section on ruggedness will be included in the design guide. The concept for the design guide and much of the material in it originated under Bureau of Mines contract H0377052. This contract included a survey of design and fabrication practice in the manufac- ture of explosion-proof enclosures. The survey showed that a design guide could be useful to the mining industry to unify and formalize procedures already being followed and to introduce to the industry new data and procedures generated in this contract. Data are also being taken from other related research, such as Bureau contract H0387009 (4_)3 and from numerous other sources as cited in the bibliography of this paper. ■^Underlined numbers in parentheses refer to items in the bibliography at the end of this paper. 12 ORGANIZATION AND CONTENTS OF THE GUIDE The guide is organized to address the various steps in the design process in the approximate order in which they usu- ally occur. It is divided into many chapters, each of which covers a separate topic. It is hoped that through this organization of material the designers can easily locate and use the parts that they need or follow step by step through the entire design process. To facilitate the use of the guide by designers, as well as by analysts, many of the design process steps will be given in graphical format. When the guide is first used, it is recommended that the designer review each chapter for content and then read care- fully or work through examples (to be presented in the guide). The examples will cover the procedures described in the guide and will acquaint the designer with proper use of the equations, graphs, and data. Because the guide will cover a number of different topics, some extensively, not all of the material can be presented in this paper. Instead, a brief review of the contents of the guide by topic is given, and following sections give de- tails on three of the topics: "Design Pressures," "Design for Static Pressure," and "Design for Dynamic Pressure." Graphical solutions, which will be part of the completed design guide, have not yet been developed. Design Requirements . - This chapter will cite MSHA requirements that affect the strength of explosion-proof enclo- sures and discuss new requirements that are being considered for adoption by MSHA at the time the guide is published. MSHA requirements are subject to change, and current regulations must always be checked. Design Pressures . - This chapter will give procedures for choosing static and dynamic design pressures. Usually the static pressure will control the design. but dynamic effects must be checked also. Methods will be given for estimating the maximum value, rise time, and decay time of the dynamic pressure. The design pro- cedure assumes that pressure piling does not occur, and guidelines will be given that will help the designer avoid pres- sure piling in his or her enclosure. The "Design Pressures" section of this paper gives the major details of this chapter except for those which pertain to pres- sure piling. Design for Static Pressure . - Proce- dures will be given for sizing rectan- gular plates, circular plates, and cyl- inders to withstand the design static pressures without excessive distortion. Much of this chapter is included in this paper. Design for Dynamic Pressure . - Proce- dures will be given for checking dynamic effects of the loading for both elastic and elastic-plastic material behavior. For small deformations, such as now permitted by MSHA, the elastic proce- dure can be used with good accuracy even though plastic behavior is ignored. For larger permanent deformations, the elastic-plastic procedure is recommended. This material is covered in the related section of this paper. Design for Ruggedness . - MSHA does not now have an explicitly stated ruggedness requirement; however, a ruggedness test is being considered for adoption. This chapter will contain procedures for de- signing the enclosure to withstand exter- nal impact loads that can occur in the mine and which will most likely be in- cluded in a ruggedness requirement if one is adopted. Guidelines for Windows and Lenses . - This will be an extensive chapter that covers the design of windows and their mounting arrangements for use in explosion-proof enclosures (9^). Design procedures will be given for both glass and polycarbonate windows and lenses. Thermal Stresses in Windows. This brief chapter will give procedures for evaluating thermal stresses that are pro- duced in glass and polycarbonate windows and lenses by the MSHA thermal shock test with one-side quenching. Window mounting arrangements for minimizing thermally induced stresses will also be given. Welded Connections . - Weld joint design and joint efficiency will be covered in this chapter. Material is taken primar- ily from American Welding Society (AWS) Welding Code DW14.4, and this code is referenced for welding procedures, welder qualification, and weld joint inspection. Bolted Connections . - This chapter will contain design procedures for bolted con- nections, primarily for enclosure covers. Cover restraint associated with external loads and prying action produced by in- ternal loads will be treated. Connec- tions will be designed to avoid permanent deformation in the bolts which could in- crease the flange gap. 13 Reinforcement of Openings . - Simple formulas for area replacement and stiff- ness continuity will be given for use in designing reinforcement around openings in the enclosure. Materials of Construction . - This chap- ter will list materials that are suit- able for explosion-proof enclosures. It will include metals, polycarbonates, glasses, and adhesives. Service life re- strictions will be noted where applica- ble, and procedures will be given for qualifying new polycarbonates, adhesives, and sealants. Examples . - Examples that demonstrate the use of the material in the guide will be included in this chapter. These examples will not cover the steps a designer must follow which lead up to the selection of an enclosure geometry, but they will address the design details that assure proper enclosure strength. DESIGN PRESSURES Static Pressures Unless higher than expected pressures occur during the explosion test, the structural performance test will subject the enclosure to 150 psig. Because the magnitudes of pressures that can result from pressure piling cannot be predicted, it is proposed that the enclosure be designed under the assumption that pres- sure piling will not occur and that the guidelines for avoiding pressure piling (to be given in the design guide) will be followed. Using this approach, the de- sign static pressure will be 150 psig. from the work of Perlee (7^). To analyze the response of an enclosure for the ex- plosion pressure, the pressure's mag- nitude, rise time, and duration must be known. Methods for estimating the mag- nitude and rise time of the pressure are contained in following sections. Because the enclosure is usually tightly closed during the explosion test and in service, the pressure does not decay rapidly; therefore, its duration can be treated as long relative to the response time of the enclosure. Peak Pressure Dynamic Pressures Dynamic pressures are produced in an enclosure during the explosion test. These pressures are routinely measured by MSHA during testing and have also been measured by other investigators. For a spherical enclosure, a typical pressure- time history is given by Zabetakis ( 12 ) and shown in figure 1. Conqjarison be- tween experiment and theory is also given The peak pressure that theoretically can occur in a closed volume from the ignition of a mixture of methane and air at the stoichiometric4 ratio is 117 psig. '^Ratio at which there is, theoreti- cally, just sufficient oxygen in the air for complete combustion of all of the methane. This ratio is 9.5 pet for meth- ane and air. 14 120 100- o) 80 — 60- 40 - 20- I \ r Bulletin 627 KEY Calculated Experimental 10 20 30 40 50 60 70 TIME, msec uu 80 60 C3) CO 40 Q. n" Q- 20 o^ UJ 10 8 CO 6 CO LU CC 4 a. 1 1 1 1 \j^_ - Rl 7839 (J ~ " n = 4.3 A^ // — / / — - ^ KEY y / • Calculated o Experimental 1 // 1 1 , ,J 1 — L_ 8 10 20 30 40 50 70 TIME, msec FIGURE 1. - Pressure in a 9-liter spherical chamber produced by ignition of a 9.6 voi-pct CH^-ai mixture. This pressure is produced by complete combustion and assumes that no heat loss to the walls of the enclosure and, of course, no pressure piling occur. A more realistic upper limit for the pressure is the measured value shown in figure 1. This pressure is for a 9.6 vol-pct methane-air mixture, which is slightly rich and in practice gives the highest measured value of the explosion pressure. Also, this pressure was measured in a spherical chamber with central ignition, which is an idealized condition relative to most enclosure tests. Typical values of pressure measured by MSHA in the explosion test are 60 to 80 psig. Thus a peak explosion pressure of 100 psig is suggested as a reasonable design value unless the designer has valid reasons for increasing its magnitude. For example, the dynamic design value might be in- creased if the particular geometry of the enclosure is expected to produce pressure piling. Rise Time The rise time of the explosion pres- sure, as well as its magnitude, is needed to calculate the response of the enclo- sure to these dynamic loads. Further, a short rise time generally produces a greater response in the enclosure than a long one. Therefore, an estimate of the minimum rise time is made for design purposes. To determine the minimum rise time, it is convenient to use the maximum rise rate of the pressure. Again, the data in figure 1 for spherical enclosures will be used. Using the minimum rise rate deter- mined for spherical enclosures will not guarantee that a minimum rise time for all enclosures will be obtained, but us- ing a minimum estimated rise time is con- servative, and so the approach is appro- priate for design of nonspherical enclosures. 15 An approximate relationship between the time from ignition to peak pressure and the enclosure volume is given by Zabetakis (12) as 75 /V, msec. (1) where V is in cubic feet and t is in milliseconds. This is not a minimum rise time, but equation 1 can be used to scale a minimum rise time or maximum rise rate determined for one enclosure to enclo- sures with different volumes. A mini- mum rise rate for a 9-liter enclosure can be estimated from figure 1. Taking the tangent to the steepest part of the curve in figure 1,4 gives a pressure rise rate of MAX 9 7 psi „ „ ./ 1 n c = 9.2 psi/msec. 10.5 msec (2) This same rate is given by the upper part of the curve in figure 15. Com- bining equations 1 and 2, the maximum rise rate for enclosures of other vol- umes is = 9.2 psi/msec 75 V0.1352 ft3 75 3/v"Tt3 4.72 psi/msec. Thus, the minimum rise time is (3) •"min MAX W 4.72 (4) where Pmax is in pounds per square inch, gage, and V is in cubic feet, or for a maximum explosion pressure of 100 psig 21 "*/¥, msec (5) where V is in cubic feet. Equation 5 is used unless the expected dynamic pressure is other than 100 psig. DESIGN FOR STATIC PRESSURE The sizes and shapes of enclosures can vary greatly, but basically the geometry of the sides, bottom, and cover can be categorized as rectangular, circular, and cylindrical. Approximate design proce- dures are given in the following para- graphs for treating each of these basic geometries. Some of these procedures include provisions for permanent defor- mations; others do not. If permanent deformations are included in the design, they must not exceed the limits set by MSHA. Use of these procedures will be illustrated by example problems in the design guide, but ultimately the designer must use his or her own judgment in the application of these methods. Rectangular Plates Solutions for rectangular plates under uniform static loads are taken from the work of Jones and Walters O ) . These solutions are based on rigid-perf ectly plastic theory and give the relation- ship between the applied pressure, P, and permanent normal deflections of the plate, Wq. For the geometry of fig- ure 2, equations 6 through 11 give the solutions for clamped and simply sup- ported plates. FIGURE 2. - Geometry of rectangular plates. 16 Clamped Plates P 1.236 w, — = 0.618 + P. h where P, 3ay (3 - 2Co) 4hi 3 = a/b ^ 1, and ?o = e(/3 + 02 - 3). (1 + 3^) /TT^ - 3(2 + 3^) /3^rF (6) (7) (8) (9) Simply Supported Plates where 3a, (3 - 2CJ 2h- 3h^ Pc 4 w2 I 5^+ (3 - 2Co)2 T^ = 1 + (3 - Co) for j;- < 2": (10) (11) The solution for clamped plates has been shown by Jones and Walters (5^) to give good agreement with experimental results for plates with 3 in the range of 1/3 to 1 and a/h in the range of 57 to 161. Equations 6 and 10 both should be applied to conditions for which w^/h is less than 1/2. The current MSHA require- ment that the permanent deformation be less than 0.040 in/ft will usually give w^/h « 1/2. Boundary conditions for the walls and covers of typical enclosures are not well known. This occurs because some junc- tions between side walls are formed by bending the plate and some are produced by welding. Further, the strength of the weld joint will vary depending upon the weld joint design; so the designer must use his or her own judgment in defining the boundary conditions of the plate and, thus, the proper equations to be used. If there is uncertainty about the proper choice of boundary conditions, the assumption of simple support will give a conservative results. (The choice of boundary conditions will be further illustrated in the design guide by the use of worked examples.) 17 Circular Plates Solutions for circular plates are based on rigid-perf ectly plastic theory and de- fine the pressure for which a fully plas- tic collapse mechanism first develops in the plate. Lower bound collapse pres- sures for uniformly loaded plates of radius R and thickness h are given by Wood (11), and are as follows: Clamped Edges P, 1.877 ayh2 Simply Supported Edges P =^ (12) (13) where Oy is the yield stress of the mate- rial. Note that these equations are independent of displacement. They do not include the stiffening effects of mem- brane forces produced by large displace- ments and in-plane restraint at the boundaries. The stiffening effects of displacements have been accounted for in simply supported plates by Sawezuk (8). For plates with in-plane restraint at the boundary, he gives ^ = l^l^^^^ (14) P 3 \h J center displacement of the plate and P^ is given by equation 13. As stated for rectangular plates, the permanent deformation now permitted by MSHA is so small that membrane effects are negligible. Thus equations 12 and 13 are satisfactory for enclosure design. Also, in sizing the cover, it is recom- mended that the clamping effect of the bolts be neglected and that equation 13 for simple support be used to determine cover thickness. Cylinders Cylinders designed for 150 psig will have thin walls, so that standard elastic formulas for thin wall cylinders can be used for design. Stresses will be uni- form across the wall thickness, t, and the design conditions can be based on the circumferential wall stress just reaching some prescribed design stress. For cyl- inders, a design stress just below the yield stress (95 pet ay) is chosen be- cause, once yielding occurs, the deforma- tions cannot be accurately predicted. Beyond the yield stress, the deformations will be controlled only by the effect of end restraints and the increase in strength of the material produced by strain hardening. Thus, the following pressure-thickness relationship is recom- mended for cylinders with a mean radius, R: Thin Wall Cylinder p ^ 0.95 Oyy ^ (15) This formula should be used with the min- imum yield stress of the material to assure that permanent deformations do not occur. Local yielding may occur at the ends of the cylinder or around penetra- tions, but the deformations will be well within the current MSHA requirement of 0.040 in/ft. (The design of penetrations will be covered in another part of the guide. ) DESIGN FOR DYNA^aC PRESSURES Dynamic effects produced by transient loads can be important when the rise time of the loading is less than three times the fundamental period of the structural element. Methods for including dynamic effects in the design of enclosures for the explosion pressures are given in this section. 18 Elastic Behavior If the response Is elastic, that Is, no permanent deformations occur, the effect of the dynamic loads can be esti- mated from figure 3. This figure gives the dynamic effect in terms of a dynamic load factor (DLF). (DLF),^;^^ is the max- imum deflection produced by the dynam- ic loading divided by the deflection pro- duced by a static load of the same magnitude. figures contain formulas for the funda- mental frequencies of rectangular plates, circular plates, and cylinders. Note that for a cylinder, the fundamental mode corresponds to a uniform radial expansion and is not a bending mode. A uniform radial expansion is the type of response that a uniform internal pressure will produce in the cylinder. The next sec- tion also contains procedures for esti- mating fundamental frequencies of struc- tural coii5)onents. I To determine the DLF, one must know the rise time of the loading and the funda- mental frequency of the structural ele- ment. (The rise time of the explosive loading is determined from data given in "Design for Dynamic Pressures" section. ) In general, the shortest rise time will produce the greatest dynamic effect; how- ever dips in the DLF do occur at even multiples of t^/T,^. Because the rise times can vary, the minimum rise time should be estimated and the largest DLF for this rise time or any longer rise time should be chosen as the appropriate value. The fundamental frequency of the structural component can be calculated from equations in figures 4 and 5. These Once the DLF has been determined from figure 3, equations 6 through 15 can be used to calculate the required plate or wall thickness. Note that the design pressure is now the maximum explosion pressure as shown in the section "Design Pressures" multiplied by DLF. This procedure is approximate because the formulas for plates in the "Design for Static Pressure" section are based on some material plasticity and thus the response is not coii5)letely elastic. The error will be small. A method that accounts for material plasticity, and which can be used if larger permanent deformations are permitted, is given in the next section. FIGURE 3. = Maximum response of 1-degree elastic systems (undamped) subjected to constant force with finite rise time (9). 19 boundary'" wb /ph/D FOR VALUES OF b/a CONDITIONS 0.4 0.6667 1.0 1.5 2.5 11.45 10.13 12.13 22.58 23.65 14.26 10.67 17.37 23.02 27.01 19.74 11.68 28.95 24.02 35.99 32.08 13.71 56.35 26.73 60.77 71.56 18.80 145.50 37.66 147.80 F V = 0.3 / / / / / / / / / / / / / / / / / / / / / / F ' //////// V = 0.3 / / / / / / / / / ^ ' / / / / ^ / /////// /^('^///^ KEY CLAMPED SIMPLE SUPPORT FREE D = Eh- 12 (1- v) p = mass density FIGURE 4, = Fundamental circular frequencies for rectangular plates (derived from data in reference 3), 20 GEOMETRY CIRCULAR FREQUENCY 10.22 ^ / D S.S. V = 0.3 ■ -f -v^ © p(l - V ) (breathing zone) Eh- 12 (1- V ) mass density FIGURE 5. - Fundamental frequencies of circular plates and long cylinders (derived from data in reference 3)., Elastic-Plastic Behavior For elastic-plastic behavior, par- ticularly If larger permanent defor- mations are permitted, procedures devel- oped by Biggs (2^), Nemark (j6 ) , and Beck (_1_) are recommended. These methods are based upon the assumption of a defor- mation pattern for the structural ele- ment. The usual assumption for elastic behavior is the static deformed shape under the same distribution of loading. For plastic behavior, a hinge mechanism is postulated. Once the deflected shape has been chosen, the system (beam or plate) can be transformed into a one- degree-of -freedom (dof) system for which the dynamic response can be easily coiq)uted. 21 Based on the assumption of the static deformed shape and the formation of plas- tic hinges, figures 6 through 10 give transformation factors, resistance func- tions, spring constants, and shear reac- tions for uniformly loaded rectangular and circular plates. These quantities are given for different boundary condi- tions, plate aspect ratios, and material behavior. Definitions of these quanti- ties follow. Maximum Resistance . - This is the total load at which the plate will develop a fully plastic hinge mechanism. For the elastic solution of a clamped rectangular plate, it is the load at which a plastic hinge forms at the fixed boundary. For elastic-plastic and fully plastic behav- ior, it corresponds to the development of a hinge at the center as well as the fixed edge. The moments used to calcu- late the maximum resistance are M° = negative plastic bending psb moment capacity pev unit width at the center of edge b. psb total negative plastic bending moment capacity along edge b. pfb = total positive plastic bending moment along the midspan section parallel to edge b. M = same pf a J edge as above, but for and ^pc ~ positive plastic bending moment capacity per unit length at the center of the circular plate, Mp2 = negative plastic bending moment capacity per unit length at the edge of the circular plate. Positive and negative moments need be considered only for concrete slabs in which these values may differ from each other. For homogeneous materials, there is no distinction between the positive and negative values. Spring Constant . - This is the spring constant for the plate expressed in terms of the elastic modulus, E, and the moment of inertia per unit width, Ig. It is calculated as the total load on the plate divided by the static center deflection. Dynamic Reaction . - The dynamic reac- tion is the shear at the boundaries ex- pressed as a fraction of the instanteous values of the applied load, F, and the resistance, R. To accurately determine the maximum shear, the time-history of the one-dof system must be calculated; however, because the loading from inter- nal explosions is idealized as a ramp function to a constant load, calculating the shear reaction at t,,, when the response is a maximum, and after the load has peaked, should give the maximum val- ue. An upper limit is found by taking the peak value of F and R to calculate the dynamic shear. Load Factor, K|^ , The load factor is the ratio of the load applied to the equivalent one-dof system to the total load applied to the plate. When the static deformed shape is assumed for the deformation pattern (as it is in figs. 6-10), it is also the ratio of the spring constant of the equivalent system to the spring constant of the plate as given in the figures. Mass Factor, k^ . - This is the ratio of the mass of the equivalent one-dof system to the total mass of the plate. Load-Mass Factor, k|^n^ . - The factor is the ratio k^/k ^ . load-mass 22 Simple Suppo .y Strain Range a/b Load Factor Mass Factor •s, Load-Mass Factor ^M Maximum Resistance Spring Constant k Dynamic Reactions \ \ Elastic 1.0 0.9 0.8 0.7 0.6 0.5 0.45 0.47 0.49 0.51 0.53 0.55 0.31 0.33 0.35 0.37 0.39 0.41 0.68 0.77 0.71 0.73 0.74 0.75 ¥(Va-p.) i(l2.0M^^^ + 11.0M^J i(i2-° Va"^°-^ Vb) i(l2.0M^^^+ 9-«"pfb) i(^^-°Va- ^-^Vb) 252 El^/a^ 230 El^/a^ 212 El^/a^ 201 El^/a^ 197 El^/a^ 201 El^/a^ 0.07 F + 0.18 R 0.06 F + 0.16 R 0.06 F + 0.14 R 0.05 F + 0.13 R 0.04 F + 0.11 R 0.04 F + 0.09 R 0.07 F + 0.18 R 0.08 F + 0.20 R 0.08 F + 0.22 R 0.08 F + 0.24 R 0.09 F + 0.26 R 0.09 F + 0.28 R Plastic 1.0 0.9 0.8 0.7 0.6 0.5 0.33 0.35 0.37 0.38 0.40 0.42 0.17 0.18 0.20 0.22 0.23 0.25 0.51 0.51 0.54 0.58 0.58 0.59 ^(Va-Pib) I (1^-° "pfa + ^1-° "pfb) i(l2-°"pfa + l°-3 Vb) i(l2.0«^^^. 5-8 Vb) i(l2.0M^^^. 9.3 Mp,,) i(l2.0M^^^. 9-°Vb) 0.09 F + 0.16 R^ 0.08 F + 0.15 R^ 0.07 F + 0.13 R^ 0.06 F + 0.12 R m 0.05 F + 0.10 R m 0.04 F + 0.08 R 0.09 F + 0.16 R^ 0.09 F + 0.18 R^ 0.10 F + 0.20 R^ 0.10 F + 0.22 R^ 0.10 F + 0.25 R^ 0.11 F + 0.27 R FIGURE 6. - Transformation factors for two-way slabs. Simple supports— four sides, uniform load; for Poisson's ratio = 0.3 (10). The purpose of the transformation fac- tors is to define a one-dof system that can be easily solved for its response to dynamic loads. The resulting equation is The fundamental period of the plate, T^, is also given by the one-dof system because the system is kinematically equivalent to the plate. The period is calculated as KmK-^ + KLkx = KlF (t). (16) where M, F, and k are the total mass, total load, and spring constant, respec- tively, of the plate, and K^ and K^^ are the transformation factors. Dividing equation 16 by Kl yields 2Tr 'Kk,M K.k = 2. /M. (18) Klu/^ + kx = F(t). (17) Equation 17 shows that only K^^^ is needed to obtain an equivalent one-dof system. Thus, only K^^, the spring constant from the figures, and the plate mass are needed to calculate T|^^ Numerical or closed-form solutions can be obtained easily for equation 17, but graphical solutions are also available. The solution for a ramp loading to a 23 '"•'^ Strain Range a/b Load Factor h Mass Factor Load-Mass Factor ^M Maximum Resistance Spring Constant k Dynamic Reactions ^.A ^B Elastic 1.0 0.9 0.8 0.7 0.6 0.5 0.39 0.41 0.44 0.46 0.48 0.51 0.26 0.28 0.30 0.33 0.35 0.37 0.67 0.68 0.68 0.72 0.73 0.73 ^°-* ";sa — ;sa*^v — ;sa*^v '■^»;sa.-^Vb «-^";sa*--i^Vb -«;sa*^v. 575 El^/a^ 476 El^/a^ 396 El^/a^ 328 El^/a^ 283 El^/a^ 243 El^/a^ 0.09 F + 0.16 R 0.08 F + 0.14 R 0.08 F + 0.12 R 0.07 F + 0.11 R 0.06 F + 0.09 R 0.05 F + 0.08 R 0.07 F + 0.18 R 0.08 F + 0.20 R 0.08 F + 0.22 R 0.08 F + 0.24 R 0.09 F + 0.26 R 0.09 F + 0.28 R Elasto- Plastlc 1.0 0.9 0.8 0.7 0.6 0.5 0.46 0.47 0.49 0.51 0.53 0.55 0.31 0.33 0.35 0.37 0.37 0.41 0.67 0.70 0.71 0.72 0.70 0.74 i[^^-°(^fa-Va)*l^-°^fb] i[^^-°(Va*Va)^^^-°^fb] i[l^-°(Va^Va)*^°-^^fb i[^^-°(^fa*«psa)^ ^-^ Vb_ - 12. of M ^ + M 1 + 9.0 M ,, a [ \ pfa psa/ pfb 271 Zlja^ 248 EI /a" 228 tlja- 216 EI /a^ 212 El^/a^ 216 EI /a^ 0.07 F + 0.18 R 0.06 F * 0.16 R 0.06 F ^ 0.14 R 0.05 F + 0.13 R 0.04 F ^ o.n R 0.04 F + 0.09 R 0.07 F + 0.18 R 0.08 F + 0.20 R 0.08 F + 0.22 R 0.08 F + 0.24 R 0.09 F + 0.26 R 0.09 F + 0.28 R Plastic 1.0 0.9 0.8 0.7 0.6 0.5 0.33 0.35 0.37 0.38 0.40 0.42 0.18 0.20 0.22 0.23 0.25 0.51 0.51 0.54 0.58 0.58 0.59 l[l^-°("pfa*"psa)*l^-°^fb] i[^2-°(Va^Va)^^°-^"pfb] i[^^°(Va^Va)* ^-'^fb] i[l^-°("pfa^Va)* '-^Vb] i[^^-°(Va^%sa)* '-"Vb] 0.09 F + 0.16 R 0.08 F + 0.15 R_^ O.OT F + O.n R^ 0.06 F + 0.12 R 0.05 F + 0.10 R 0.04 F + 0.08 R 0.09 F + 0.16 R 0.09 F + 0.18 R_^ 0.10 F + 0.20 R 0.10 F + 0.22 R^ 0.10 F + 0.25 R 0.11 F + 0.27 R^ FIGURE 7. - Transformation factors for two-way slabs. Short edges fixed, long edges simply sup= ported; for Poisson's ratio = 0.3 (10). 24 Simple Support. ^ Fixed Strain Range a/b Load Factor Mass Factor Load-Mass Factor Maximum Resistance Spring Constant Dynamic Reactions \ % Elastic 1.0 0.9 0.8 0.7 0.6 0.5 0.39 0.40 0.42 0.43 0.45 0.45 0.26 0.28 0.29 0.31 0.33 0.34 O.bl 0.70 0.69 0.71 0.73 0.72 ^°-^ «;sb 2° -2 «;sb "■2 «;sb 575 EI /a^ 600 El^/a^ 610 El^/a^ 662 El^/a^ 731 El^/a^ 850 El^/a^ 0.07 F + 0.18 R 0.06 F + 0.16 R 0.06 F + 0.14 R 0.05 F + 0.13 R 0.04 F + 0'. 11 R 0.04 F + 0.09 R 0.09 F + 0.16 R 0.10 F + 0.18 R 0.11 F + 0.19 R 0.11 F * 0.21 R 0.12 F + 0.23 R 0.12 F + 0.25 R Elasto- Plastic 1.0 0.9 O.S 0.7 0.6 0.5 0.46 0.47 0.49 0.51 0.53 0.55 0.31 0.33 0.35 0,37 0.39 0.41 0.67 0.70 0.71 0.73 0.74 0.74 i[^^-°%fa^^^-°(Vb^Vb)] i[l2.0M^^^.10.3(Mp^^.M^^^)] i[l2.0M^^^. 9.b(mp^,.H^J] i[l=-°"pfa^ ^-^iVb^^fb)] - 12. OM^ + 9.0|M ^+M,J a [ pfa ^ psb pfb;_ 271 El /a^ 248 EI /a^ 228 Elg/a^ 216 El^/a^ 212 EI /a^ 216 El /a^ 0.07 F + 0.18 R 0.06 F + 0.16 R 0.06 F + 0.14 R 0.06 F + 0.13 R 0.04 F + O.ll R 0.04 F + 0.09 R 0.07 F + 0.18 R 0.08 F + 0.20 R 0.08 F + 0.22 R 0.08 F + 0.24 R 0.09 F + 0.26 R 0.09 F + 0.28 R Plastic 1.0 0.9 0.8 0.7 0.6 0.5 0.33 0.35 0.37 0.38 0.40 0.42 0.17 0.20 0.22 0.23 0.25 0.51 0.51 0.54 0.58 0.58 0.59 l[l^-° Va"^^-°(Vb"Vb) i[l^-°%fa^l^-°(Vb"Vb) i[l^-°%U*l°-^("psb*^4 i[i^-°Va^ '•«(^sb*^4 i[l^-°"pfa^ ^-^(Vb^^fb) 1 [l2.0 M^^^ 4 9.0 (m^^^ . M^^^) 0.09 F + 0.16 R__^ 0.08 F + 0.15 R 0.07 F + 0.13 R^ 0.06 F + 0.12 R__^ 0.05 F + 0.10 R 0.04 F + 0.08 R_^ 0.09 F + 0.16 R___ 0.09 F + 0.18 R^ 0.10 F + 0.20 R 0.10 F + 0.22 R 0.10 F + 0.25 R 0.11 F + 0.27 R_^ FIGURE 8. - Transformation factors for two-way slabs. Short sides simply supported, long sides fixed; for Poisson's ratio = 0.3 (10). 25 w ^ / / ( ( ^ D Dynamic Reactions Strain Load Factor Mass Factor Load-Mass Factor Maximum Spring Constant Range a/b \ \ Sm Resistance ^ V ^B 1.0 0.33 0.21 0.63 "•^ ''psb 810 El^/a^ 0.10 F + 0.15 R 0.10 F + 0.15 R 0.9 0.34 0.23 0.68 "•* ";.b 742 El^/a^ 0.09 F + 0.14 R 0.10 F + 0.17 R 0.8 0.36 0.25 0.69 "•* "psb 705 EI /a^ 0.08 F + 0.12 R 0.11 F + 0.19 R Elastic 0.7 0.38 0.27 0.71 2'- 2 «;sb 692 El^/a^ 0.07 F + 0.11 R 0.11 F + 0.21 R 0.6 0.41 0.29 0.71 "•^%sb 724 EI /a^ 0.06 F +-0.09 R 0.12 F + 0.23 R 0.5 0.43 0.31 0.72 ^°-2 "psb 806 El^/a^ 0.05 F + 0.08 R 0.12 F + 0.25 R 1.0 0.46 0.31 0.67 l['^-°(Va^ Va)*l^-°(Vb^ ^j] 252 El^/a^ 0.07 F * 0.18 R 0.07 F + 0.18 R 0.9 0.47 0.33 0.70 i[^^-°(Va^ Va)*^^-°("pfb" ^sjl 230 EI 1^ 0.06 F + 0.16 R 0.06 F + 0.20 R Elasco- 0.8 0.49 0.35 0.71 l[l^-°(Va^ Va)*^°-^(Vb* %=^)j 212 El^/a^ 0.06 F + 0.14 R 0.06 F + 0.22 R Plastic \ / \' 0.7 0.51 0.37 0.73 i[l^-°(Va^ Va)^ '-nvb^ ",..) 201 El^/a^ 0.05 F + 0.13 R 0.06 F + 0.24 R 0.6 0.53 0.39 0.74 ih°(Va^ Va)^'-KVb^ V.)] 197 El^/a^ 0.04 F + 0.11 R 0.09 F + 0.26 R 0.5 0.55 0.41 0.75 l[l^-°(^ta- Va)* '-"(Vb^ V.)] 201 EI /a^ 0.04 F + 0.09 R 0.09 F + 0.28 R 1.0 0.33 0.17 0.51 \ 12.0 ^Mpj^ + "„.)•"■» (v.- vj) 0.09 F + 0.16 R 0.09 F + 0.16 R_^ 0.9 0.35 0.18 0.51 l[^^-°(Va^ "psa)^ll-°(Vb^ "»0] 0.08 F + 0.15 R 0.09 F + 0.18 R Plastic 0.8 0.37 0.20 0.54 i[i^-°(Va^ Va)^^°-^(Vb^ "..) 0.07 F + 0.13 R 0.10 F + 0.20 R 0.7 0.38 0.22 0.58 \ 12.0f Mpj^ + Va)^ '■«(Vb* %.\ 0.06 F - 0.12 R_. 0.10 F + 0.22 R__^ 0.6 0.40 0.23 0.58 i[l^°(»pfa^ «psa)* '-^("pfb^ V.)] 0.05 F + 0.10 R^ 0.10 F + 0.25 R^ 0.5 0.42 0.25 0.59 ^ 12.0 (m j^ + Va)* '"("pfb- ".«)] 0.04 F + 0.08 R 0.11 F + 0.27 R_^ FIGURE 9. - Transformation factors for two-way slabs. Fixed supports, uniform load; for Poisson's ratio = 0,3 (10). 26 Fixed Edges Simple Supports Edge Condition Strain Range Load Factor Mass Factor •Si Load-Mass Factor Maximum Resistance Spring Constant Dynamic Reaction Simple Supports Elastic 0.46 0.30 0.65 18.8 M pc 216 El/a^ 0.28 F + 0.72 R Plastic 0.33 0.17 0.52 18.8 M pc 0.36 F + 0.64 R Fixed Supports Elastic 0.33 0.20 0.61 "•1 V 880 El/a^ 0.40 F + 0.60 R Elasto- Plastic 0.46 0.30 0.65 i«-«(v"v) 216 El/a^ 0,28 F + 0.72 R Plastic 0.33 0.17 0.52 i«-«(v"v) 0.36 F + 0.64 R m FIGURE 10, = Transformation factors for circular slabs, Poisson's ratio = 0.3 (10). constant value is given by figure 11. Figure 11^5 gives the maximum displace- ment, x^, relative to the displacement at yielding, Xg, and figure ll5 gives the time, tj^, at which the maximum response is reached relative to the rise time of the load, t,.. Both of these quantities are given as functions of the maximum resistance, R^, the maximum total applied load, F, and the fundamental period of the plate, T,^, To use figure 3, one must determine the following quantities: Tjg - calculated by equation 18, the rise time of the load, the maximum value of the load on the plate which is the product of the maximum value of the pressure and the plate area. R^ - maximum resistance from figures 6 through 10. Xg - deflection at which the plate yields. It is calculated as the maximum resistance for elastic be- havior divided by the spring con- stant from figures 6 through 10. Note that to use figure 4, the only transformation factor that is needed is Klm, used in equation 18 to calculate Tf^. Also, it is necessary to know in advance how the structure will respond; that is, will the structure remain elastic, experience mild plasticity, or undergo gross deformations? This determines the choice of parameters from the fig- ures. If elastic behavior is expected, choose the parameters that correspond to elastic behavior; if gross plasticity is 27 r t, \£L:\2i Jr Load Resistance function Displacement function .4 .6.81.0 2 tr/tn 4 6 810 20 Load Resistance Displacement function function . ■ ■ ' I L_L .4 .6.81.0 2 tr/tn 4 6 810 20 FIGURE 11, - Maximum response of undamped single-degree-of=freedom elastic=plastic syster to step pulse with finite rise time (TO). expected, choose the parameters for plastic behavior; if mild plasticity is expected, average the values for elastic and plastic behavior. Another alterna- tive is to solve equation 17 for the response of the structural element. When solving equation 17, note that the parameters that define the one-dof approximation mist change as the system yields. As explained previously, the maximum response is usually produced by a loading with the shortest rise time; however, dips in the response do occur at even multiples of the t^/T,^ ratio. Therefore, it is recommended that the response be taken as the maximum value that occurs for all rise times equal to or greater than the minimum rise time as determined in the "Design Pressures" section. CONCLUSIONS The authors hope that the design guide that is being developed by the Federal Bureau of Mines will be of great benefit to the manufacturers of explosion-proof enclosures. This can only happen if the guide contains the appropriate informa- tion in a suitable format and if the guide is accepted and used by the indus- try. Although this paper does not show the final format of the guide, it does give an overview of its contents. The format will be similar to that used in this paper, but it will be con?)lemented by charts in some areas to make the solu- tions simpler. Comments on the scope, contents, and format of the guide are solicited by the authors in order to tailor the guide to the mining industry. 28 BIBLIOGRAPHY 1. Beck, J. E., B. M. Beaver, H. S. Levine, and E. Q, Richardson. Single- Degree-of-Freedom Evaluation. Air Force Weapons Lab Rept. AFWL-TR-80-99, March 1981, 306 pp.; available from Kirtland Air Force Base, N. Mex. 2. Biggs, J. M. Simplified Analysis and Design for Dynamic Load. Ch. 7 in Structural Design for Dynamic Loads. McGraw-Hill Book Co., Inc., New York, 1959, 453 pp. Urbana, 111., Tech. Rept. on Off. Naval Res. Contract N60-RI-71, December 1950, pp. 34-44. 7. Perlee, H. E., F. N. Fuller, and C. H. Saul. Constant -Volume Flame Propa- gation. BuMines RI 7839, 1974, 24 pp. 8. Sawezuk, A. Large Deflections of Rigid-Plastic Plates. Proc. 11th Inter- nat. Cong. Appl. Mech. , Munich, Germany. Sprenger-Verlag, 1964, pp. 224-228. 3. Blevins, R. D. Formulas for Natu- ral Frequency and Mode Shape. Van Nostrand Reinhold Co. , New York, 1979, 492 pp. 9. Scott, L. W. Some Design Factors for Windows and Lenses Used in Explosion- Proof Enclosures. BuMines IC 8880, 1982, 9 pp. 4. Francis, P. H. , and J. Lankford. Recommended Acceptance Testing Criteria for Adhesives and Sealants for Explosion- Proof Electrical Enclosures (Contract H0387009, Southwest Res. Inst.). BuMines OFR 129-80, Jan. 9, 1980, 67 pp.; NTIS PB81-128738. 5. Jones, N. , and R. M. Walters. Large Deflections of Rectangular Plates. J. of Ship Res., June 1971, pp. 164-171. 6. Newark, N. M. Methods of Analy- sis for Structures Subjected to Dynam- ic Loading. University of Illinois, 10. U.S. Army Corps of Engineers. Manual EM 1110-345-415, Design of Struc- tures To Resist the Effects of Atomic Weapons. 1975, 136 pp.; available from Defense Documentation Center, Logistic Agency, Cameron Station, Alexandria, Va. 11. Wood, R. H. Plastic and Elastic Design of Slabs and Plates. The Ronald Press Co., New York, 1961, 344 pp. 12. Zabetakis, M. G. Flammability Characteristics of Combustible Gases and Vapors. BuMines B 627, 1965, 121 pp. 29 HIGH-VOLTAGE, EXPLOSION-PROOF LOAD CENTERS By George Conroy,'' Randy Berry, 2 and Robert Gillenwater2 ABSTRACT To attain future underground mine pro- duction considered to be necessary by the Department of Energy, voltages higher than the presently permitted 4,160 V must be carried to the close vicinity of the mining machines; otherwise, trailing cable size and weight creates handling problems. A project has been running for the past year to delineate approval tests and acceptance criteria for explosion- proof load centers that will permit oper- ation inby the last open crosscut. The investigation includes fabrication and testing of a load center. Tentative specifications are presented in this report. INTRODUCTION Present methods of developing some un- derground coal mines and some extensive noncoal mines require the transfer of large amounts of electric power through long lengths of trailing cable. A prac- tical alternative to specifying very large diameter cables, used to reduce voltage drop, is to transfer the power at high voltage and relatively low current levels. The voltage might then be trans- formed to a working level by equipment located near the mining machine. In cer- tain circumstances, either inherent in the mode of operation or because uncon- trollable methane "bursts" can occur, the stepdown transformer and associated switching apparatus may become surrounded by methane-air atmosphere, leading to the possibility of a mine explosion if the equipment has not been constructed to be permissible. Verifying permissibility is the respon- sibility of Mine Safety and Health Admin- istration's (MSHA) Approval and Certifi- cation Center in Triadelphia, W. Va. The ^Supervisory electrical engineer (re- tired) , Pittsburgh Research Center, Bu- reau of Mines, Pittsburgh, Pa. ^Technical staff consultant, Foster- Miller Associates, Waltham, Mass. ^Senior engineer, Foster-Miller Associ- ates, Waltham, Mass. most important guidelines for the deter- mination are the set of regulations con- tained in Title 30, Code of Federal Regu- lations (CFR), Part 18. Part 18 gives very little guidance for test and accept- ance of equipment to be energized at voltages from 1,000 to 4,160 V, and does not even consider voltages above 4,160 V. In an effort to facilitate the use of higher voltages, the Department of Energy (DOE) funded a Federal Bureau of Mines project to develop acceptance tests and criteria that MSHA could use to supple- ment current Part 18 regulations. At the same time, DOE has funded MSHA to develop an explosion test gallery capable of accommodating and testing the large en- closures necessary for high-voltage transformers and switchgear. At MSHA's option, this gallery could be transported to a high-power test facility to conduct explosion tests on load centers using a high-voltage source ranging up to the 15- kV maximum now contemplated. These tests would measure the internal pressures and temperatures resulting from the occur- rence of an arcing electrical fault in the presence of an explosive methane-air mixture. At other times, the MSHA gal- lery would be utilized at Triadelphia, W. Va. , for explosion testing as pres- ently conducted on all permissible equipment. 30 This paper examines the following spe- cial considerations concerning high volt- age load centers: • Power rating: maximum levels • Maximum current to an arcing fault • Maximum arc duration • Electrical clearances • Insulating materials Each of these five topics is dis- cussed separately in the following sections. SPECIAL CONSIDERATIONS CONCERNING APPROVAL AND TESTING CRITERIA FOR HIGH-VOLTAGE PERMISSIBLE LOAD CENTERS Power Rating of Load Centers Maximum power requirements for under- ground coal mine section load centers, whether or not permissible, are estimated to be within the 2,000-kVA upper limit addressed in the ongoing Bureau program. One limiting factor is the physical size of such equipment as con^sared with the dimensions of existing mine entries. Load centers capable of delivering more than 2,000 kVA are likely to pose major mobility problems in underground coal mines. This may be less of a problem in such noncoal settings as oil shale mines. However, the question of permissibility has not yet been resolved -in that appli- cation, nor have particular equipment rating and size requirements been specified. Another limiting factor for lower volt- age permissible load centers is cable size. Present Part 18 requirements for three-conductor trailing cable for use at voltages up to 5 kV specify a maximum ampacity of 305 A for 350 MCM shielded cable. This places an upper limit on the power rating of permissible load centers, assuming power input through a single trailing cable, as shown in figure 1. As indicated by the curve, a 2,000-kVA load center could be supplied at any line voltage level above 3.8 kV. No more than 500 kVA could be utilized in a load cen- ter designed for 1.0-kV operation. How- ever, a 350 MCM cable is much larger than most mines would care to use for supply- ing power to individual sections. Sup- posing the use of a 4/0 cable, with a maximum ampacity of 220 A, yields the lower curve shown in figure 1. Using this cable, a 2,000-kVA permissible load center could be supplied at any line voltage level above 5.1 kV. A 500 kVA load center would require a minimum oper- ating voltage of 1.3 kV. Adoption of cable rating tables that allow a higher conductor temperature would result in higher current levels and correspondingly higher allowable kilovoltampere ratings. This is not likely to occur without 2,000 III/ 350 MCM / cable-y / M/'l 1 1 1 1 1 1 1 1 1,000 - // - 800 - 1 - 600 - II - 400 11 - / / cable - 200 "/ " 100 1 _ 80 - 60 - 40 - Present 30 CFR 18 voltage limitation 4.16 kV - 20 - - 10 1 1 1 1 1 1 1 1 1 1 1 1 1 1 I 2 3 4 5 6 7 8 9 10 II 12 13 14 15 LINE VOLTAGE, kV FIGURE 1, - Regions of practical interest, high- voltage permissible load centers. 31 substantial justification. Therefore, the practical range of interest in pre- paring special acceptance criteria for very high voltage peinnissible load cen- ters includes input powers from 500 to 2,000 kVA, and line voltages from 1.0 kV to a maximum of 15 kV, the highest volt- age expected to be used underground in the forseeable future. Although load centers operating at less than 4.16 kV could supposedly be examined under the existing regulations, a dearth of prece- dents in the region above 1 kV justifies interest with regard to new test and acceptance criteria. Maximum Current to an Arcing Fault One of the principal questions regard- ing permissible load centers is whether or not their explosion-proof charac- teristic can be compromised if the energy released by an arcing electrical fault is added to that generated by a simultaneous methane-air explosion within the enclo- sure. The energy contributed by the arc- ing fault depends on the fault current. Because the arc has a finite resistance, this fault current will be less than the current possible in a "bolted" (zero resistance) short circuit. Thus, the worst case situation can be conserva- tively considered as the occurrence of an arcing fault with a current flow equal to the maximum available current through a bolted fault at the internal terminations of the trailing cable. The shortest length of trailing cable connecting the load center to the substation supplying its power would usually be 500 to 1,000 ft, while the maximum might be 21,000 ft. The cable impedance at the shortest dis- tance is so small as to be a negligible factor in limiting the fault current, so that substation transformer impedance must provide the worst case limit. At the longest distance, trailing cable impedance along with the arc voltage might limit fault current to a rather low value, thus creating a sensitive situa- tion with regard to the ability of primary circuit protection devices to interrupt the current rapidly enough to prevent a serious buildup of pressure in the enclosure. This problem is explored in more detail in the next subsection. Continuing the conservative approach, the substation is assumed to be connected to an "infinite bus" at the point of supply from the utility. This substation is presumed to be dedicated to the single load center concerned so that its trans- former impedance may be related directly to this circuit. This impedance is assumed to be 5 pet of the impedance that would yield rated load current, and therefore rated apparent power, for any given line voltage. Taking some discrete values of line voltage and power rating, transformer phase-to-phase impedance would appear as shown in table 1. This value can be used to calculate the current available to a bolted fault (neglecting cable impedance). The im- pedance per thousand feet of the smallest size cable that could be used for the selected values, per Part 18, is also listed in table 1. This value will be used for estimating the current limiting effect of long lengths of cable. Table 2 and figure 2 show the current available to ^bolted faults as calculated using the preceding assumptions. The largest available fault current is 8,175 A in the 1-kV, 500-kVA unit. The availa- ble current in the 4.16-kV, 2,000-kVA unit is only slightly less. A worst-case test as suggested by the analysis is to be one with 10,000 A available from a 4.16-kV source, supplying an arcing fault in a 2,000-kVA load center. As the intent is to verify only that the enclo- sure can withstand any developed internal pressures and temperatures, actual test- ing would be performed using separately introduced test electrodes rather than actual component members. An arbitrary electrode spacing of 6 in is recommended, with the arc being started by means of a connecting filament. 32 TABLE 1. - Primary cable values Line voltage, kV Power rating, kVA Line current at rated power, A Transformer 5 pet impedance, ohms Mine cable size Cable impedance, ohms per 1,000 ft 1.0. 2.3. 4.16. 7.2, 12.5, 15. 500 500 1,000 1,000 1,500 2,000 1,000 1,500 2,000 1,000 1,500 2,000 1,000 1,500 2,000 289 126 251 139 208 278 80 120 160 46 69 92 38 58 77 0.173 .916 .458 1.499 .999 .749 4.490 2.993 2.245 13.351 9.021 6.794 19.485 12.990 9.742 350 MCM AWG 2 300 MCM AWG 1/0 AWG 4/0 350 MCM AWG AWG AWG 2/0 AWG AWG AWG AWG 6 AWG 6 AWG 4 ^Smallest size permitted by 30 CFR 18. TABLE 2. - Peak available current to a bolted fault 0.116 .386 .125 .255 .151 .116 .296 .153 .102 .468 .296 .237 .468 .468 .296 33 1 \ — I — \ — I \ \ r I I I I I I I I I I I I I I I I I i 2 3 4 5 6 7 8 9 10 II 12 13 14 15 LINE VOLTAGE, kV FIGURE 2. - Peak available current for various power ratings. Maximum Arc Duration Modern Industrial practice with regard to power systems In the 1- to 15-k.V range calls for the use of circuit protection devices capable of Interrupting the cir- cuit within 50 msec (3 cycles at 60 Hz). Circuit Interruptions In this short Interval would occur after a fault to ground activates a sensing relay, which signals the circuit Interrupter to Initi- ate action. The total Interrupting time on a ground fault might therefore be as much as 83 msec (5 cycles). If, for any reason, the ground fault relay should fall to act. It Is Important that circuit overcurrent protection can act to Inter- rupt the current; otherwise, an arc of long duration and extreme destructlveness could result. The question therefore arises as to whether or not the combined effect of transformer Impedance, Im- pedance of the longest trailing cable likely to be used, and arc voltage drop can limit the circuit current to a value that Is too low to permit reliable set- ting of overcurrent relays without nui- sance tripping on motor startup. Arc voltage Is known to be a decreasing function of arc current, becoming fairly constant for currents lower than about 300 A. Arc voltage then slowly Increases with Increasing current, probably never exceeding 1,000 V under the conditions described. Some additional Investigation Is justified In this area. If an arc voltage of 1,000 V Is assumed (except In the case of the 1.0-kV load center), this value can be used together with the Im- pedances listed In table 1 to calculate the ratio of minimum overcurrent to rated current for various line voltages and power ratings. These ratios are listed In table 3. The ratio at 1.0 kV line voltage Is uncertain, as 1,000-V arc voltage obviously cannot be assumed. However, this value of line voltage Is already covered by 30 CFR 18, and no Instances of long-continued arcing have been reported. From table 3, the smallest calculated Ip/Ip ratio Is 3.5, where Ip Is the fault current and Ip^ Is the rated current. This occurs for a 2.3-kV, 500-kVA load center. In order for this value to be reached by a motor startup motor larger than 425-hp must directly online, under load, starting current that Is 5.5 current. The more common starting cur- rents for mine equipment are about 2.5 times rated current. Motors larger than AOO-hp powered from a 2.3-kV line can certainly be equipped with step starters. If necessary, to avoid nuisance Inter- ruptions even when started under full load. Therefore, the overcurrent protec- tion for the load center primary circuit can be set below 3.5 times rated current, with only a very small time delay — less than 2 cycles — to allow for unavoidable line transients. From this analysis, circuit protective devices can be presumed capable of Inter- rupting an arcing fault, either by ground fault relaying or by the circuit over- current relaying, within 7 cycles. This permits calculation of the maximum energy In the fault as shown In the last column of table 3. The free volume within the enclosure should be sufficient to avoid excessive buildup of pressure when this energy Is combined with that released by a methane explosion. current, a be started to give a times rated 34 TABLE 3. - Arc currents and energies at various line voltages Line voltage, Power Calculated arc Ratio, Maximum fault kV rating, kVA current, A If/Ir energy, kJ 1.0 500 (0 (0 (O 2.3 500 440 3.5 118 1,000 1,162 4.6 312 4.16 1,000 1,500 992 790 7.1 3.8 481 383 2,000 1,600 5.8 776 7,2 1,000 1,500 577 996 7.2 8.3 485 837 2,000 1,409 8.8 1,184 12.5 1,000 491 10.7 715 1,500 753 10.9 1,098 2,000 975 10.6 1,422 15 1,000 1,500 476 612 12.5 10.5 833 1,071 2,000 875 11.3 1,531 lvalue uncertain, 1,000-V arc voltage cannot be assumed Electrical Clearances A special laboratoiry investigation^ was conducted to determine the effect of a methane-air explosion in facilitating the occurrence of an arc between electrodes sustaining a high potential difference. The mixture at which arcing is most likely to occur is 9.8 pet methane, the same mixture that gives maximum pressure and highest flame temperature. Using this mixture, the critical voltages for arc initiation were determined for a number of gap distances. A linear relationship was found to exist between electrode spacing and the minimum initia- tion voltage under explosion conditions. As presented in figure 3, curve B indi- cates that arcing can occur during the explosions at a gap distance more than 18 times as great for a given voltage as the spacing for an arc in air, shown by curve A. For design purposes, a factor of 1.5 times the curve B spacing is considered ■^Scott, L. W. / and Joseph G. Dolgos. Electrical Arcing at High Voltage During Methane-Air Explosions Inside Explosion- Proof Enclosures. BuMines TPR 115, 1982, 9 pp. to be the minimum recommended clearance between energized members. Insulating Materials Used in Explosion-Proof Enclosures It has long been known that certain organic insulating materials, when decomposed by the action of an electrical arc, can liberate large quantities of hazardous gases. This has never been a problem in the United States for per- missible equipment operating at less than incidents of enclosure this phenomenon have Canadian and European use of high-voltage equipment in explosion-proof enclosures is commonplace. 4 1,000 V. However, failure caused by been reported' in mines where the ^Barbero, L. P. E. H. Davis, and H. Lord. Hazards Resulting From the Volatilization by Electric Arcing of Insulating Materials in Flameproof Equip- ment. Pres. at 15th Internat. Conf. on the Safety in Mines Research, Karlovy, Vary, Czechoslovakia, Sept., 18-21, 1973, 8 pp.; available for consultation at Bureau of Mines Pittsburgh Research Cen- ter, Pittsburgh, Pa. 35 z^ Curve A -Curve B FIGURE 3. 10 20 30 40 70 80 90 100 no 50 60 SPACING, mm Minimum arc of voltages versus air-gap spacings of electrodes. Curve A, in air; Curve B, in 9.8 pet methane-air mixtures. Owing to the widespread use of organic plastic insulating materials in the elec- trical industry, a conplete ban of such materials in high-voltage explosion-proof enclosures is not practical. However, the use of such insulators must be accom- panied by special efforts that minimize the amount of such material used and insure that the materials that are used have been tested and found highly resist- ant to the destructive effects of elec- trical arcs. In addition, protective devices inside a permissible enclosure can be used to detect arcing, overpres- sure, or excessive temperatures and dis- connect incoming power before a hazardous condition develops. CONCLUSIONS The preceding discussion describes some of the problems and reasoning in- volved in the development of acceptance criteria for high voltage permissible load centers. The product of this devel- opment to date is the document attached as appendix A. This document has been reviewed by various government and indus- try personnel and comments received up to now have been incorporated. Further views and comments are invited as the opportunity for revision will continue for some time. 36 APPENDIX A. —RECOMMENDED APPROVAL AND TESTING CRITERIA FOR HIGH-VOLTAGE PERMISSIBLE LOAD CENTERS! INTRODUCTION The objective of phase I of this pro- gram is the development of criteria for use by the Mine Safety and Health Admin- istration (MSHA) in testing and approving permissible load centers with maximum voltage ratings of 15 kV and maximum power capacities of 2,000 kVA. A draft of the recommended criteria follows this brief introduction. The distinguishing feature between the existing requirements of Title 30, Code of Federal Regulations (CFR), Part 18, and the requirements of 30 CFR 18 as sup- plemented by these criteria, is that the latter permits approval of permissible equipment consisting, in full or in part, of high-voltage circuits and con5)onents. The criteria address only those factors that are influenced by the higher volt- age. These factors are discussed in num- bered paragraphs that follow. The organization of the recommended criteria follows closely that of Part 18. This organization has the obvious advan- tage of familiarity and is the natural result of "extending" Part 18 to cover the type of equipment of interest in this program. For ease of reference, selected sec- tions of Parts 18 and 75 are reproduced in appendix B. These sections are appli- cable to permissible load centers with one recommended change. That change is to Section 18.47 — Voltage Limitation. It is recommended that the 4,160-V limita- tion found in Section 18.47(d) be de- leted. Recommended changes to the texts of Sections 18.47(d), 18.47(d) (3), and 18.47(d)(5) can be found in the comment following paragraph 14. With the 4,160-V limitation removed. Part 18 will continue to be the sole source of requirements for approval of low- and medium-voltage permissible elec- trical equipment. However, Section 18.47(d)(6) reserves for MSHA the right to require additional safeguards for high-voltage equipment. These criteria represent those additional safeguards re- quired for the approval of load centers, transformers, switchgear, and related equipment with maximum voltage ratings of 15 kV and maximum power capacities of 2,000 kVA. Similar criteria should be developed for other high-voltage applica- tions, as needed. Comments and suggestions from interest- ed persons are welcomed. Written com- ments are especially appreciated. PART A—GENERAL PROVISIONS 1. Purpose The purpose of these criteria is to specify the design and testing require- ments to be used by MSHA in approving load centers, transformers, and switch- gear as permissible for use in gassy mines or tunnels. These criteria are a ^This work was performed under Bureau of Mines contract H0308093. supplement to the existing requirements of 30 CFR 18, all of which apply unless specifically modified or replaced by parts of these criteria. These pro- visions are applicable to load cen- ters, transformers, switchgear, and related equipment operating at maximum voltages of 15 kV. For transformers, the maximum secondary voltage is 4.16 kV, and the maximum power rating is 2,000 kVA. 37 2, Definitions a. Corona (partial discharge). A type of localized discharge resulting from the ionization of gas in an insula- tion system when the voltage stress ex- ceeds a critical value. The ionization is localized over only a portion of the distance between the electrodes of the system. b. Corona inception voltage. The lowest voltage at which corona occurs as the applied voltage is gradually increased. c. Corona extinction voltage. The highest voltage at which corona no longer occurs as the applied voltage is grad- ually decreased from above the corona inception voltage. d. Phase segregation. The isolation of each phase conductor of an electrical circuit by means of a conqjletely sur- rounding grounding metallic covering or enclosure. 3. Quality Assurance The factory inspection form required by 30 CFR 18.6(k) shall specify a system- atic checking sequence designed to assure the quality of each load center. The form shall include, but not be limited to, a detailed checklist in the following areas: • Explosion-proof construction — with special attention paid to flange gap dimensions, surface finishes, cable entrances, plugs and receptacles, joints, covers, fasteners, and welding quality. • Components — a visual inspection for damaged or faulty components prior to assembly. • Assemblies — insure that all conpo- nents, subassemblies, and assemblies are fitted in accordance with appropriate drawings and specifications. • Operation — check for proper opera- tion of all mechanical devices and link- ages. Also check for proper installation and operation of pressure relief, venti- lation and drainage devices, pressure rise detectors, over-temperature sensors, and other protective devices. • Ancillaries — ^where applicable, all preceding checks will be carried out on ancillary components, assemblies, and enclosures. Emphasis should be placed on systematic organization of the inspection form to insure that all checks are made at appro- priate times. PART B~CONSTRUCTION AND DESIGN REQUIREMENTS 4. Limitation of External Surface Temperatures The temperature of the external sur- faces of mechanical or electrical load center conq)onents shall not exceed 150° C (302° F) under normal operating conditions. 5. Electrical Clearances Minimum clearances posed electrical conductor explosion-proof enclosures listed in Table A-1. between surfaces shall be ex- in TABLE A-1, Minimum clearances COMMENT: No change is recommended to the current requirements of 30 CFR 18.23. Load centers must be designed so as to adequately dissi- pate the heat generated by the transformer without exceeding the surface temperature requirements. Voltage range, V 8,000 to 15,000 , 5,000 to 8,000 , 2,000 to 5,000 , 1,000 to 2,000 , Clearance, in 7 4 3 2 38 6. Insulating Materials Used in Explosion-Proof Enclosures a. Inorganic insulating materials shall be used, where feasible, in prefer- ence to organic plastic insulating materials. b. Insulators conqjosed of organic plastic materials shall not be used as bushings or supports for bus bars or in other locations where potentially danger- ous short circuits might occur. c. Where the use of organic plastic insulating materials cannot be avoided, the following conditions shall be met: (1) The volume of such materials used shall be kept to a minimum. 7. Gaskets and Sealed Enclosures a. Gaskets shall be used in accord- ance with 30 CFR 18.27. b. Hermetically sealed (welded) en- closures, pressurized with inert gas or other special atmosphere, may be used for transformers if (1) Means are provided for sens- ing the loss of an effective seal and automatically disconnecting the power supply from the enclosure if the seal is lost. The methods se- lected for this purpose shall pre- vent reapplication of power to the transformer until the proper atmos- phere and an effective seal are restored. (2) Those materials used shall be highly resistant to electrical tracking and arcing. (3) A detection device shall be provided that will operate to remove the power incoming to the enclosure before decomposition of the insulat- ing material due to • an electrical fault leads to hazardous conditions. COMMENT: A material will be deemed highly resistant to electrical tracking if it has a con5)arative tracking index (CTI) of not less than 250. Materials will be deemed highly resistant to elec- trical arcing if they pass the fuse wire arc test — described by J. N. Hardwich in "An Improved Fuse Wire Arc Test Including a Pro- posed Specification," The Elec- trical Research Association, Report No. 5078, 1964~or an equally effective test recognized by MSHA. The detection device specified in paragraph c(3) may operate on pres- sure rise, temperature rise, detec- tion of the products of insulator decomposition, or other effective means. In either case, the ground monitor circuit can be used to re- move the power incoming to the enclosure. (2) Means are provided for pre- venting or relieving overpressuriza- tion of the enclosure caused by accidental overfilling with gas, operation at high temperatures, or internal electrical fault. (3) The enclosure is of substan- tial design and construction so as to prevent damage that may lead to a loss of seal. The minimum thickness of material for the walls shall be 1/4 in. c. Enclosures designed in accord- ance with paragraph 7(b) need not be designed to withstand the minimum inter- nal pressure of 150 psig as specified in 30 CFR 18.31(a)(1). d. Enclosures designed in accord- ance with paragraph 7(b) need not be explosion tested as specified in 30 CFR 18.62 and paragraph 19 of these criteria. 8. Explosion-Proof Enclosures a. The requirements of 30 CFR 18.31, 18.32, and 18.33 must be met by all explosion-proof load center and auxiliary con5)onent enclosures. All welds shall be made in accordance with American Welding Society Standard AWS D14.4-77. 39 b. MSHA may Impose additional re- quirements for the use of high-voltage conq)onents in potted enclosures. COMMENTS: The problems associated with placing high-voltage compo- nents in potted enclosures should be explored, and adequate testing and approval criteria developed, before this type of equipment is accepted. 9. Access Openings and Covers Access openings in esqjlos ion-proof load center enclosures will be permitted where necessary for proper maintenance such as tap changing and circuit breaker adjustment. The provisions of 30 CFR 18.29 must be met. f. ICEA standards for derating am- pacities for cables wound on reels, and ICEA recommended minimum bending diam- eters, shall be observed. g. No temporary splices shall be used. All permanent splices and ter- minations shall be made in accordance with manufacturer's specifications by qualified personnel familiar with the techniques required for proper high- voltage cable installation, operation, and maintenance. 11. Lead Entrances; Cable Connectors and Plugs a. The provisions of 30 CFR 18.42 — Explosion-proof distribution boxes — shall apply to explosion-proof load centers. 10. High-Voltage Power Cables High-voltage power cables used as portable cables, or located where the use of permissible equipment is required, shall conform to the following: b. High-voltage cable connectors and plugs used in areas where permis- sible equipment is required shall meet the requirements of 30 CFR 18.41 and the test requirements specified in table A-3. a. Have each conductor of a current-carrying capacity consistent with the Insulated Cable Engineers Association (ICEA) standards (see table A-2). b. Have current-carrying conductors not smaller than No. 6 (AWG). c. Have flame-resistant properties (see 30 CFR 18.64). d. Have short-circuit protection at the outby (circuit-connecting) end of underground conductors. The fuse rating or breaker trip setting shall be included in the assembler's specifications. e. Have nominal outside dimensions and tolerances consistent with ICEA standards. c. Tests specified in paragraph (b) shall be performed on each high-voltage connector or plug intended for use on permissible equipment or on approved cables in areas where permissible equip- ment is required. Equipment shall be designed so that plugs and recepta- cles can be completely assembled and tested before mounting on the permissi- ble enclosure. Cable connectors that have been tested for use in permissible areas shall be clearly marked and identified. COMMENTS: The tests listed in ta- ble A-3 are based on work performed under Bureau of Mines contract H0377043. 40 TABLE A-2. - Ampacities for portable power cables, amperes per conductor Power conductor size Single conductor 2,001 to 8,000 V,l shielded 8,001 to 15,000 V,l shielded Three-conductor, round and flat, to 5,000 V, nonshielded Three conductor, round to 8,000 V, shielded 8,001 to 15,000 V, shielded AWG, copper: 8 6 4 3 2 1 1/0, 2/0. 3/0. 4/0. MCM, copper: 250 300 350 400 450 500 550 600 650 700 750 800 900 1,000 NAp 112 148 171 195 225 260 299 345 400 444 496 549 596 640 688 732 779 817 845 889 925 998 1,061 NAp NAp NAp NAp 195 225 259 298 343 397 440 491 543 590 633 678 NAp NAp NAp NAp NAp NAp NAp NAp 59 79 104 120 138 161 186 215 249 287 3^0 357 394 430 460 487 NAp NAp NAp NAp NAp NAp NAp NAp NAp 93 122 140 159 184 211 243 279 321 355 398 435 470 503 536 NAp NAp NAp NAp NAp NAp NAp NAp NAp NAp NAp NAp 164 187 215 246 283 325 359 NAp NAp NAp NAp NAp NAp NAp NAp NAp NAp NAp NAp N^ NAp Not applicable. ^Based on single isolated cable in air operated with open-circuited shield. NOTE. -These ampacities are based on a conductor tenq)erature of 90° C and an ambient temperature of 40° C. TABLE A-3. - Standard dielectric tests for high-voltage cable plugs and connectors used in areas where permissible equipment is required Test Test voltage, kV 8.7-kV class 15-kV class 27 15 75 7 40 35 6-hr withstand value do Impulse withstand peak value do Corona (partial-discharge) extinction 25 95 11 15-min dc withstand value do 50 41 the tech- 1.38 may be between the to prevent 12. Leads Through Common Walls Between Explosion-Proof Enclosures Insulated bushings with proper volt- age rating and current-carrying capaci- ties may be used in the common wall between two explosion-proof enclosures. When insulated wires or cables are ex- tended through a common wall between two explosion-proof enclosures, niques described in 30 CFR II en5)loyed provided the seal two enclosures is sufficient propagation of an explosion from one en- closure to the other. Wires and cables shall be mechanically secured in open areas of the enclosure and in passageways between enclosures to prevent excessive movement in the event of high current flows. 13. Openings Through Common Walls Between Explosion-Proof Enclosures As provided in 30 CFR 18.38(e), un- sealed openings through common walls between explosion-proof enclosures shall be large enough to prevent pressure piling. Partitions subdividing single enclosures shall not be used. Internal coiq)onents shall be arranged so as not to effectively divide the interior of the enclosure into pockets joined by re- stricted passages, COMMENTS: Pressure piling is a conplex phenomenon that can oc- cur when gas in a portion of an explosion-proof enclosure is com- pressed before being ignited. The resulting pressure rise may be much greater than would normally be ex- pected from a methane-air ignition. Subdividing enclosures into com- partments connected by narrow pas- sages, either intentionally or by careless component placements, can result in pressure piling and should be avoided. It is not always possible to predict exactly when, or to what degree, pressure piling may occur. However, its presence can be detected in the explosion tests specified in 30 CFR 18.62. When a pressure exceeding 125 psig is developed during explosion tests (which would indicate that pressure piling has occurred) , MSHA can reject the enclosure unless con- structional changes are made that result in a reduction of pressure to 125 psig or less, or the enclo- sure withstands a dynamic pressure of twice the highest value recorded in the initial test. 14. Voltage Limitation Load centers with nameplate ratings in excess of 4,160 V, but less than 15,000 V, may be approved as permissible if the applicable requirements of 30 CFR 18, and the additional requirements con- tained in these criteria, are met. COMMENT: The following changes to 30 CFR 18.47 are recommended: • In 30 CFR 18.47(d), delete the words "but not exceeding 4,160 V." • 30 CFR 18.47(d)(3) should be changed to read: "All high voltage switchgear and con- trols for equipment having a nameplate rating exceeding 1,000 V are approved (permis- sible) for use in gassy mines or tunnels, or are certified as suitable for incorporation in a machine to be submitted for approval, or are located remotely and operated by remote control at the main equipment. Potential for re- mote control shall not exceed 120 V." 42 • 30 CFR 18.47(d)(5) should be changed to read: "Portable (trailing) cable for equip- ment with nameplate ratings greater than 1,000 V shall include grounding conductors, a ground-check conductor, and grounded metallic shields around each power conductor and shall be adequately con- structed and insulated for the applied voltage," 30 CFR 18.47(d)(6) reserves the right for MSHA to require "addi- tional safeguard" for high-voltage equipment. These criteria repre- sent those additional safeguards for high-voltage load centers (up to 15 kV and 2,000 kVA). Similar criteria should be developed, as needed, for other types of high- voltage, permissible equipment (for example, motors and motor starters) . 15. Electrical Protective Devices a. High-voltage circuits connected to permissible load centers shall be protected by modern, high-speed circuit breakers equipped with devices to pro- vide protection against undervoltage, grounded phase, short circuit, and overcurrent. b. Upon detection of a ground fault in a circuit supplying power to high- voltage, permissible equipment, the cir- cuit shall be deenergized and remain so until the ground fault is cleared. In no instance shall such a circuit be ener- gized while a phase conductor remains grounded. c. All equipment intended to break current at fault levels shall have an in- terrupting rating sufficient for the sys- tem voltage and the current that is available at the line terminals of the equipment. Equipment intended to break current at other than fault levels shall have an interrupting rating at system voltage sufficient for the current that must be intermapted. d. The overcurrent protective de- vices, the total impedance, the con5)onent short-circuit withstand ratings, and other characteristics of the circuit to be protected shall be so selected and coordinated as to permit the circuit pro- tective devices to clear a fault without the occurrence of extensive damage to the electrical components of the circuit. e. Additional coordination shall be provided between the electrical circuit characteristics and protective devices and the design parameters (for example, strength, free volume) of the explosion- proof enclosure to prevent damage to the enclosure should an electrical fault oc- cur. Precautionary measures may include (1) Limitation fault energy. of available (2) Provision of adequate free volume within the enclosure, (3) Inclusion of devices which detect hazardous pressure and tem- perature rises, ozone, or visible arc radiation, and/or (4) Use of devices or tech- niques designed to vent or limit in- ternal pressure. COMMENT: High-voltage ac circuit breakers are available with inter- rupting speeds of 1, 2, 3, 5, and 8 cycles. A description of various types and a history of the develop- ment of such breakers can be found in the "Standard Handbook for Elec- trical Engineers," Fink and Beaty, 11th ed. In addition, vacuum cir- cuit breakers with interr^ipting speeds of less than 1 cycle are available in the voltage range of interest to this program. These breakers are in common use through- out the mining industry today and offer other advantages (for exam- ple, compact size, low maintenance, enclosed contacts) which are desir- able for mine electrical equipment in general, and permissible equip- ment in particular. I 43 Paragraph 15(b) is included in recognition of the fact that some mine electrical systems have, in the past, been allowed to operate for long periods (hours) with one phase of the power system grounded. This is usually done to avoid shut- down of an entire mine while a sin- gle ground fault is being located and repaired. However, continued operation with a grounded phase increases the electrical stress on the system insulation and, with the occurrence of a second fault, can give rise to a double phase- to-ground fault situation. This defeats the purpose of ground shields, barriers, and other forms of phase segregation provided in shielded cables and permissible high-voltage enclosures. If per- mission is ever granted to energize and operate a power system before clearing a ground fault, it should be conditioned on the removal from the circuit of equipment of the type covered by these approval criteria. Devices other than high-speed breakers are available to limit fault energy. Most notable of these is the current-limiting fuse. This type of protection is used extensively in the utility industry to prevent overpressurization of transformer enclosures due to internal faults. However, the current-limiting characteristics of these fuses does not come into play unless several (20 to 30) mul- tiples of rated current is availa- ble from the circuit in the event of a fault. In most mine high- voltage power systems, the overall impedance of the circuit will limit fault current availability to a few thousand amps. Only in locations where abnormally high fault cur- rents are available (for example, very close to the main mine sub- station) would current-limiting fuses actually provide the kind of protection for which they are designed. Data from testing to be performed later in this program may shed additional light on proper coordi- nation of circuit electrical char- acteristics and enclosure design specifications. , High-Voltage Circuit Design a. High-voltage circuits and compo- nents in permissible enclosures shall conform with accepted practices and stan- dards of high-voltage design for the appropriate voltage class. b. Ground barrier and shields shall be used when possible to minimize the occurrence of phase-to-phase faults with- in enclosures. 17. Component Placement a. High-voltage electrical compo- nents located in explosion-proof enclo- sures shall not be placed in the same plane as the flange gap. b. Components located in tight, explosion-proof enclosures shall not be placed in such a way as to effectively subdivide or create "conq)artments" or "pockets" within the enclosure which might give rise to pressure piling upon ignition of a methane-air mixture. PART C~INSPECTIONS AND TESTS 18. Inspections Inspections specified in 30 18.60 and 18.61 shall be required CFR for Additional requirements of parts A and B of these criteria shall also be performed. These include a. Examination of items listed on permissible load centers. inspections which take into account the the factory inspection form. 44 b. Examination for the use proper insulating materials. of c. Examination for adequacy and proper installation and operation of all electrical and mechanical protective devices. d. Examination for areas of possi- ble excessive electrical stress. (4) Fault current — fault cur- rent shall be increased in succes- sive increments of 1,000 A until an excessive pressure rise (>125 psig) is measured in the enclosure, or to a maximum of 10,000 A. The enclo- sure shall be approved for use in circuits with available fault cur- rents less than the maximum fault current used in this test. COMMENT: In examining for areas of possible excessive electrical stress, MSHA may want to use one of a number of standard corona (partial discharge) detection and measurement techniques (for exam- ple, IEEE Standard 454—1973 or ASTM Standard D 1868-73). Such testing might also be required by the manufacturer as part of the quality assurance provisions (see paragraph 3 of these criteria). 19. Tests To Determine Explosion-Proof Characteristics a. With exception of the hermet- ically sealed enclosures referred to in paragraph 7 (b, c, and d) of these cri- teria, all permissible enclosures used with load centers or related equipment must pass the explosion tests specified in 30 CFR 18.62. c. All or any part of the addi- tional tests specified in paragraph 19 (b) may be waived by MSHA for equipment meeting all other requirements of these criteria and of 30 CFR 18, if such equip- ment is provided with one or more of the following design features: (1) An enclosure containing more than one phase of a high- voltage circuit is designed and constructed so as to preclude the possible occurrence of a phase-to- phase arcing fault (for example, complete phase segregation or shielding is provided). (2) An enclosure is equipped with approved vents or pressure re- lief devices such that no pressures greater than 15 psig are measured in the methane-air explosion tests re- quired by 30 CFR 18.62. b. Enclosures containing more than one phase of a high-voltage circuit must meet the requirements of 30 CFR 18.62 when the methane-air mixture in the en- closure is ignited by a high-voltage, phase-to-phase arcing fault. The fault used in this test shall have the follow- ing specifications: (1) Fault duration — 15 cycles (0.250 sec). (2) Source voltage — rated tem voltage. sys- (3) Fault arc length — fault arc length (electrode spacing) shall be equal to the minimum clearances specified in paragraph 5 of these criteria for the appropriate system voltage. (3) An enclosure containing more than one phase of a high- voltage circuit is provided with a minimum free internal volume of 1 m3. COMMENT: The design of high- voltage, metal-enclosed switchgear so that each phase is enclosed in a separate metal housing, with an air space provided between the housings, is considered to be the safest, most practical, and most economical way of prevent- ing phase-to-phase short-circuit faults through construction meth- ods. Therefore, the additional testing described in paragraph 19(b) is required only for enclo- sures containing more than one phase. 45 Other design techniques may also provide a high degree of safety; they are listed in paragraphs 19(c) 1, 2, and 3. Paragraph 19(c) gives MSHA the prerogative to waive arc tests for such designs. It should be noted, however, that MSHA re- tains the right to verify, by test- ing, the degree of protection afforded by these design features, and is free to exercise that right until sufficient experience with such testing is gained. subject to revision, pending re- ceipt of data from testing to be performed later in this program. The use of pressure relief vents [paragraph 19(c)(2)] makes field inspection more difficult than the simple feeler-gage test now used for explosion-proof housings. It may be desirable to develop a new test to insure that the vents have not become clogged once the unit is in operation. The minimum free volume specified in paragraph 19(c)(3) — 1 m^ — is 46 APPENDIX B.-CFR 30, SUBCHAPTER D-ELECTRICAL EQUIPMENT, LAMPS, METHANE DETECTORS; TESTS FOR PERMISSIBILITY, FEES; PARTS 18 AND 75 PART 18— ELECTRIC MOTOR-DRIVEN MINE EQUIPMENT AND ACCESSO- RIES Subpart A — General Provisions Sec. 18.1 Purpose. 18.2 Definitions. 18.3 Consultation. 18.4 Equipment for which approval will be issued. 18.5 Equipment for which certification will be issued. 18.6 Applications. 18.7 Fees. 18.8 Date for conducting investigation and tests. 18.9 Conduct of investigations and tests. 18.10 Notice of approval or disapproval. 18.11 Approval plate. 18.12 Letter of certification. 18.13 Certification plate. 18.14 Identification of tested noncertified explosion-proof enclosures. 18.15 Changes after approval or certifica- tion. 18.16 Withdrawal of approval, certifica- tion, or acceptance. Subport B — Construction and Design Requirements 20 Quality of material, workmanship, and design. 21 Machines equipped with powered dust collectors. 22 Boring-type machines equipped for auxiliary face ventilation. 23 Limitation of external surface tem- peratures. 24 Electrical clearances. 25 Combustible gases from insulating material. 26 Static electricity. 27 Gaskets. 28 Devices for pressure relief, ventila- tion, or drainage. 29 Access openings and covers, including unused lead-entrance holes. 30 Windows and lenses. 31 Enclosures— joints and fastenings. 32 Fastenings— additional requirements. 33 Finish of surface joints. 34 Motors, 35 Portable (trailing) cables and cords. 36 Cables between machine compo- nents. 37 Lead entrances. 38 Leads through common walls. 39 Hose conduit. 40 Cable clamps and grips. 41 Plug and receptacle-type connectors. 42 Explosion-proof distribution boxes. 43 Explosion-proof splice boxes. 44 Battery boxes and batteries (exceed- ing 12 volts). 45 Cable reels. 46 Headlights. 47 Voltage limitation. 48 Circuit-interrupting devices. 49 Connection boxes on machines. 50 Protection against external arcs and sparks. 51 Electrical protection of circuits and equipment. 52 Renewal of fuses. Subpart C — Inspections and Tests Sec. 18.60 Detailed inspection of components. 18.61 Final inspection of complete ma- chine. 18.62 Tests to determine explosion-proof characteristics. 18.63 Tests of battery boxes. 18.64 Tests for flame resistance of cables. 18.65 Flame test of conveyor belting and hose. 18.66 Tests of windows and lenses. 18.67 Static-pressure tests. 18.68 Tests for intrinsic safety. 18.69 Adequacy tests. Subpart D — Machines Assembled With Certi- fied or Explosion-Proof Components, Field Modifications of Approved Machines, and Permits To Use Experimental Equipment 18.80 Approval of machines assembled with certified or explosion-proof compo- nents. 18.81 Field modification of approved (per- missible) equipment: application for ap- proval of modification; approval of plans for modification before modification. 18.82 Permit to use experimental electric face equipment in a gassy mine or tunnel. Appendix I Appendix II Subpart E — Field Approval of Electrically Operated Mining Equipment 18.90 Purpose. 18.91 Electric equipment for which field approvals will be issued. 18.92 Quality of material and design. 18.93 Application for field approval; filing procedures. 18.94 Application for field approval; con- tents of application. 18.95 Approval of machines constructed of components approved, accepted or certi- fied under Bureau of Mines Schedule 2D, 2E, 2F, or 2G. 18.96 Preparation of machines for inspec- tion; requirements. 18.97 Inspection of machines; minimum re- quirements. 18.98 Enclosures, joints, and fastenings; pressure testing. 18.99 Notice of approval or disapproval; letters of approval and approval plates. Subpart A — General Provisions § 18.1 Purpose. The regulations in this part set forth the requirements to obtain MSHA: (a) Approval of electrically operated ma- chines and accessories intended for use in gassy mines or tunnels, (b) certi- fication of components intended for use on or with approved machines, (c) permission to modify the design of an approved machine or certified compo- nent, (d) acceptance of flame-resistant cables, hoses, and conveyor belts, (e) sanction for use of experimental ma- chines and accessories in gassy mines or tunnels; also, procedures for apply- ing for such approval, certification, ac- ceptance for listing; and fees. § 18.2 Definitions. • • • "Approval" means a formal docu- ment issued by MSHA which states that a completely assembled electrical machine or accessory has met the ap- plicable requirements of this part and which authorizes the attachment of an approval plate so indicating. "Approval plate" means a metal "Certification" means a formal writ- ten notification, issued by MSHA, which states that an electrical compo- nent complies with the applicable re- quirements of this part and, therefore, is suitable for incorporation in ap- proved (permissible) equipment. "Component" means an integral part of an electrical machine or acces- sory that is essential to the function- ing of the machine or accessory. "Distribution box" means an enclo- sure through which one or more port- able cables may be connected to a source of electrical energy, and which contains a short-circuit protective device for each outgoing cable. • • • "Explosion-proof enclosure" means an enclosure that complies with the applicable design requirements in Sub- part B of this part and is so construct- ed that it will withstand internal ex- plosions of methane-air mixtures: (1) Without damage to or excessive distor- tion of its walls or cover(s), and (2) without ignition of surrounding meth- ane-air mixtures or discharge of flame from inside to outside the enclosure. "Flame-arresting path" means two or more adjoining or adjacent surfaces between , which the escape of flame is prevented. "Gassy mine" means a coal mine classed as "gassy" by MESA or by the State in which the mine is situated. "Incendive arc or spark" means an arc or spark releasing enough electri- cal or thermal energy to ignite a flam- mable mixture of the most easily ignit- able composition. 47 "Intrinsically safe" means incapable of releasing enough electrical or ther- mal energy under normal or abnormal conditions to cause ignition of a flam- mable mixture of methane or natural gas and air of the most easily ignitable composition. • • • "Permissible equipment" means a completely assembled electrical ma- chine or accessory for which a formal approval has been issued, as author- ized by the Administrator, Mining En- forcement and Safety Administration under the Federal Coal Mine Health and Safety Act of 1969 (Pub. L. 91-173, 30 U.S.C. 801 or, after March 9. 1978, by the Assistant Secretary under the Federal Mine Safety and Health Act of 1977 (Pub. L. 91-173, as amended by Pub. L. 95-164, 30 U.S.C. 801). "Portable cable", or "trailing cable" means a flame-resistant, flexible cable or cord through which electrical energy is transmitted to a permissible machine or accessory. (A portable cable is that portion of the power- supply system between the last short- circuit protective device, acceptable to MSHA, in the system and the machine or accessory to which it transmits elec- trical energy.) "Portable equipment" means equip- ment that may be moved frequently and is constructed or mounted to fa- cilitate such movement. "Potted component" means a compo- nent that is entirely embedded in a so- lidified insulating material within an enclosure. "Pressure piling" means the develop- ment of abnormal pressure as a result of accelerated rate of burning of a gas- air mixture. (Frequently caused by re- stricted configurations within enclo- sures.) § 18.6 Applications. • • • (e) Drawings, drawing lists, specifica- tions, wiring diagram, and descriptions shall be adequate in number and detail to identify fully the complete assembly, component parts, and subas- semblies. Drawings shall be titled, numbered, dated and shall show the latest revision. Each drawing shall in- clude a warning statement that changes in design must be authorized by MSHA before they are appliea to approved equipment. When intrinsi- cally safe circuits are incorporated in a machine or accessory, the wiring dia- gram shall include a warning state- ment that any change(s) in the intrin- sically safe circuitry or components may result in an unsafe condition. The specifications shall include an assem- bly drawing(s) (see Figure 1 in Appen- dix II) showing the overall dimensions of the machine and the identity of each component part which may be listed thereon or separately, as in a bill of material (see Figure 2 in Appen- dix II). MSHA may accept photo- graphs (minimum size 8" x 10 'A" ) in lieu of assembly drawing(s). Purchased parts shall be identified by the manu- facturer's name, catalog number(s), and rating(s). In the case of standard hardware and miscellaneous parts, such as insulating pieces, size and kind of material shall be specified. All drawings of component parts submit- ted to MSHA shall be identical to those used in the manufacture of the parts. Dimensions of parts designed to prevent the passage of flame shall specify allowable tolerances. A nota- tion "Do Not Drill Through" or equiv- alent should appear on drawings with the specifications for all "blind" holes, (f) MSHA reserves the right to re- quire the applicant to furnish supple- mentary drawings showing sections through complex flame-arresting paths, such as labyrinths used in con- junction with ball or roller bearings, and also drawings containing dimen- sions not indicated on other drawings submitted to MSHA. (j) The applicant shall submit a sample caution statement (see Figure 3 in Appendix II) specifying the condi- tions for maintaining permissibility of the equipment. (k) The applicant shall submit a fac- tory-inspection form (see Figure 4 in Appendix II) used to maintain quality control at the place of manufacture or assembly to insure that component parts are made and assembled in strict accordance with the drawings and specifications covering a design sub- mitted to MSHA for approval or certi- fication. § 18.11 Approval plate. (c) The approval plate identifies as permissible the machine or accessory to which it is attached, and use of the approval plate obligates the applicant to whom the approval was issued to maintain in his plant the quality of each complete assembly and guaran- tees that the equipment is manufac- tured and assembled according to the drawings, specifications, and descrip- tions upon which the approval and subsequent extension(s) of approval were based. S 18.12 Letter of certification. (b) A letter of certification will be accompanied by a list of drawings, specifications, and related material covering the details of design and con- struction of a component upon which the letter of certification is based. Ap- plicants shall keep exact duplicates of the drawings, specifications, and de- scriptions that relate to the compo- nent for which a letter of certification has been issued; and the drawings and specifications shall be adhered to ex- actly in production of the certified component. Subpart B — Construction and Design Requirements S 18.20 Quality of material, workmanship, and design. (a) Electrically operated equipment intended for use in coal mines shall be rugged in construction and shall be de- signed to facilitate inspection and maintenance. (b) MSHA will test only electrical equipment that in the opinion of its qualified representatives is construct- ed of suitable materials, is of good quality workmanship, based on sound engineering principles, and is safe for its intended use. Since all possible de- signs, circuits, arrangements, or combi- nations of components and materials cannot be foreseen, MSHA reserves the right to modify design, construc- tion, and test requirements to obtain the same degree of protection as pro- vided by the tests described in Subpart C of this part. (d) Flange joints and lead entrances shall be accessible for field inspection, where practicable. • • • S 18.2,'J Limitation of external surface temperatures. The temperature of the external surfaces of mechanical or electrical components shall not exceed 150° C. (302° F.) under normal operating con- ditions. § 18.24 Electrical clearances. The clearance between live parts and casings shall be sufficient to minimize the possibility of arcs striking the cas- ings. Where space is limited, the casing shall be lined with adequate in- sulation. 8 18.25 Combustible gases from insulating material. (a) Insulating materials that give off flammable or explosive gases when de- 48 composed electrically shall not be used within enclosures where the materials are subjected to destructive electrical action. (b) Parts coated or impregnated with insulating materials shall be heat- treated to remove any combustible solvent(s) before assembly in an explo- sion-proof enclosure. Air-drying insu- lating materials are excepted. • • • § 18.27 Gaskets. A gasket(s) shall not be used be- tween any two surfaces forming a flame-arresting path except as follows: (a) A gasket of lead, elastomer, or equivalent will be acceptable provided the gasket does not interfere with an acceptable metal-to-metal joint. (b) A lead gasket(s) or equivalent will be acceptable between glass and a hard metal to form all or a part of a flame-arresting path. § 18.28 Devices for pressure relief, ventila- tion, or drainage. (a) Devices for installation on explo- sion-proof enclosures to relieve pres- sure, ventilate, or drain will be accept- able provided the length of the flame- arresting path and the clearances or size of holes in perforated metal will prevent discharge of flame in explo- sion tests. (b) Devices for pressure relief, venti- lation, or drainage shall be construct- ed of materials that resist corrosion and distortion, and be so designed that they can be cleaned readily. Provision shall be made for secure attachment of such devices. (c) Devices for pressure relief, venti- lation, or drainage will be acceptable for application only on enclosures with which they are explosion tested. § 18.29 Access openings and covers, in- cluding unused lead-entrance holes. (a) Access openings in explosion- proof enclosures will be permitted only where necessary for maintenance of internal parts such as motor brushes and fuses. (b) Covers for access openings shall meet the same requirements as any other part of an enclosure except that threaded covers shall be secured against loosening, preferably with screws having heads requiring a spe- cial tool. (See Figure 1 in Appendix II.) (c) Holes in enclosures that are pro- vided for lead entrances but which are not in use shall be closed with metal plugs secured by spot welding, brazing, or equivalent. (See Figure 10 in Ap- pendix II.) § 18.30 Windows and lenses. (a) MSHA may waive testing of ma- terials for windows or lenses except headlight lenses. When tested, materi- al for windows or lenses shall meet the test requirements prescribed in § 18.66 and shall be sealed in place or pro- vided with flange joints in accordance with § 18.31. (b) Windows or lenses shall be pro- tected from mechanical damage by structural design, location, or guard- ing. Windows or lenses, other than headlight lenses, having an exposed area greater than 8 square inches, shall be provided with guarding or equivalent. § 18.31 Enclosures— joints and fastenings. (a) Explosion-proof enclosures: (1) Cast or welded enclosures shall be designed to withstand a minimum internal pressure of 150 pounds per square inch (gage). Castings shall be free from blowholes. (2) Welded joints forming an enclo- sure shall have continuous gas-tight welds. All welds shall be made in ac- cordance with American Welding Soci- ety standards. (3) External rotating parts shall not be constructed of aluminum alloys containing more than 0.5 percent mag- nesium. (4) MSHA reserves the right to re- quire the applicant to conduct static- pressure tests on each enclosure when MSHA determines that the particular design will not permit complete visual inspection or when the joint(s) form- ing an enclosure is welded on one side only (see § 18.67). (5) Threaded covers shall be de- signed with Class 1 (coarse, loose fit- ting) threads. The flame-arresting path of threaded joints shall conform to the requirements of paragraph (a) (6) of this section. (6) Enclosures shall meet the follow- ing requirements based on the internal volumes of the empty enclosure. Less than 45 cu Volume of empty enclosure 45 to 124 cu. in.. +- Minimum thickness ol matenal lor walls Minimum thicKness ol material lor flanges Minimum thickness of material for cover Minimum width ol |Oint— all in one plane Maximum clearance— loinl all in one plane Minimum width ol joint, portions ol which are in different planes— cylinders or equivalent Maximum clearances— joint in two or more planes. cylinders or equivalent (a) Portion perpendicular to plane (b) Plane portion Maximum bolt =• spacing— loinis all in one plane Maximum bolt spacing — joints, portions of which are in different planes. Minimum diameter of bolt (without regard to type of loint) Minimum thread engagement' Maximum diametncal clearance between bolt body and unthreaded holes through which it passes". Minimum distance Irom interior of enclosure to the edge of a bolt hole: Joint — minimum width 1" Joint- less than 1" wide More than 124 cu. Cylindrical Joints Shafts centered by ball or roller bearings: Vi" y." 0.015". 010" 0125" Shafts through journal beanngs: "> Other than shafts. Minimum length of flame-arresting path 0.003" 0015" 002 ' ' V3 2-inch less is allowable for machining rolled plate. 2 ViB-inch less is allowable for machining rolled plate. ' If only two planes are involved, neither portion of a joint shall be less than 'A-inch wide, unless the wider portion conforms to the same requirements as those lor a joint that is all m one plane If more than two planes are involved (as in labynnlhs or tongue-and-groove joints) the combined lengths of those portions having prescribed clearances will t>e considered 'The allowable diametrical clearance is 008 inch when the portion perpendicular to the plane portion is V, inch of or greater in length If the perpendicular portion is more than 'A incti but less than 'A inch wide, the diametncal clearance shall not exceed 006 inch > Where the term "bolt" is used, it refers to a machine bolt or a cap screw, and for either of these studs may be substituted provided the studs bottom in blind holes, are completely welded in place, or the bottom of the hole is closed with a secured plug Bolts shall be provided at all corners ^ Adequacy of bolt spacing will be judged on basis of size and configuration of the enclosure, strength of matenals. and explosion test results ' In general, minimum thread engagement shall be equal to or greater than the diameter ol the bolt specified "Threaded holes for fastening bolts shall be machined to remove burrs or projections that affect planarity of a surface forming a flame-arresting path. "Less than Vn-inch ('/<-inch minimum) will be acceptable provided the diametncal clearance for fastening bolts does not exceed Vsz inch '"Shafts or operating rods through journal bearings shall be not less than '/.-inch in diameter. The length of (it shall not be reduced when a pushbutton is depressed Operating rods shall have a shoulder or head on the portion inside the enclosure Essential pans riveted or bolted to the inside portion will be acceptable in lieu of a head or shoulder, but cotter pins and similar devices will not be acceptable. A9 (b) Enclosures for potted compo- nents: Enclosures shall be rugged and constructed with materials having 75 percent, or greater, of the thickness and flange width specified in para- graph (a) of this section. These enclo- sures shall be provided with means for attaching hose conduit, unless energy carried by the cable is intrinsically safe. (c) No assembly will be approved that requires the opening of an explo- sion-proof enclosure to operate a switch, rheostat, or other device during normal operation of a machine. § 18.32 Fastenings — additional require- ments. (a) Bolts, screws, or studs shall be used for fastening adjoining parts to prevent the escape of flame from an enclosure. Hinge pins or clamps will be acceptable for this purpose provided MSHA determines them to be equally effective. (b) Lockwashers shall be provided for all bolts, screws, and studs that secure parts of explosion-proof enclo- sures. Special fastenings designed to prevent loosening will be acceptable in lieu of lockwashers, provided MSHA determines them to be equally effec- tive. (c) Fastenings shall be as uniform in size as practicable to preclude improp- er assembly. (d) Holes for fastenings shall not penetrate to the interior of an explo- sion-proof enclosure, except as pro- vided in paragraph (a)(9) of § 18.34, and shall be threaded to insure that a specified bolt or screw will not bottom even if its lockwasher is omitted. (e) A minimum of Va inch of stock shall be left at the center of the bottom of each hole drilled for fasten- ings. (f) Fastenings used for joints on ex- plosion-proof enclosures shall not be used for attaching nonessential parts or for making electrical connections. (g) The acceptable sizes for and spacings of fastenings shall be deter- mined by the size of the enclosure, as indicated in § 18.31. (h) MSHA reserves the right to con- duct explosion tests with standard bolts, nuts, cap screws, or studs substi- tuted for any special high-tensile strength fastening(s) specified by the applicant. § 18.33 Finish of surface joints. Flat surfaces between bolt holes that form any part of a flame-arrest- ing path shall be plane to within a maximum deviation of one-half the maximum clearance specified in § 18.31(a)(6). All metal surfaces shall be finished in manufacture to not more than 250 microinches. A thin film of nonhardening preparation to inhibit rusting may be applied to fin- ished steel surfaces. § 18..35 Portable (trailing) cables and cords. (a) Portable cables and cords used to conduct electrical energy to face equipment shall conform to the fol- lowing: (1) Have each conductor of a cur- rent-carrying capacity consistent with the Insulated Power Cable Engineers Association (IPCEA) standards. (See Tables 1 and 2 in Appendix I.) (2) Have current-carrying conductors not smaller than No. 14 (AWG). Cords with sizes 14 to 10 (AWG) conductors shall be constructed with heavy jack- ets, the diameters of which are given in Table 6 in Appendix I. (3) Have flame-resistant properties. (See § 18.64.) (4) Have short-circuit protection at the outby (circuit-connecting) end of ungrounded conductors. (See Table 8 in Appendix I.) The fuse rating or trip setting shall be included in the assem- bler's specifications. (5) Ordinarily the length of a porta- ble (trailing) cable shall not exceed 500 feet. Where the method of mining requires the length of a portable (trailing) cable to be more than 500 feet, such length of cable shall be per- mitted only under the following pre- scribed conditions: (i) The lengths of portable (trailing) cables shall not exceed those specified in Table 9, Appendix I, titled "Specifi- cations for Portable Cables Longer Than 500 Feet." (ii) Short-circuit protection shall be provided by a protective device with an instantaneous trip setting as near as practicable to the maximum start- ing-current-inrush value, but the set- ting shall not exceed the trip value specified in MSHA approval for the equipment for which the portable (trailing) cable furnishes electric power. (6) Have nominal outside dimensions consistent with IPCEA standards. (See Tables 4, 5, 6, and 7 in Appendix I.) (7) Have conductors of No. 4 (AWG) minimum for direct-current mobile haulage units or No. 6 (AWG) mini- mum for alternating-current mobile haulage units. (8) Have not more than five well- made temporary splices in a single length of portable cable. (b) Sectionalized portable cables will be acceptable provided the connectors used inby the last open crosscut in a gassy mine meet the requirements of § 18.41. (c) A portable cable having conduc- tors smaller than No. 6 (AWG), when used with a trolley tap and a rail clamp, shall have well insulated single conductors not smaller than No. 6 (AWG) spliced to the outby end of each conductor. All splices shall be made in a workmanlike manner to insure good electrical conductivity, in- sulation, and mechanical strength. (d) Suitable provisions shall be made to facilitate disconnection of portable cable quickly and conveniently for re- placement. [33 FR 4660. Mar. 19, 1968; 33 PR 6343, Apr. 26, 1968] § 18.36 Cables between machine compo- nents. (a) Cables between machine compo- nents shall have: (1) Adequate cur- rent-carrying capacity for the loads in- volved, (2) short-circuit protection, (3) insulation compatible with the im- pressed voltage, and (4) flame-resis- tant properties unless totally enclosed within a flame-resistant hose conduit or other flame-resistant material. (b) Cables between machine compo- nents shall be: ( 1 ) Clamped in place to prevent undue movement, (2) protect- ed from mechanical damage by posi- tion, flame-resistant hose conduit, metal tubing, or troughs (flexible or threaded rigid metal conduit will not be acceptable), (3) isolated from hy- draulic lines, and (4) protected from abrasion by removing all sharp edges which they might contact. (c) Cables (cords) for remote-control circuits extending from permissible equipment will be exempted from the requirements of conduit enclosure pro- vided the total electrical energy car- ried is intrinsically safe or that the cables are constructed with heavy jackets, the sizes of which are stated in Table 6 of Appendix I. Cables (cords) provided with hose-conduit protection shall have a tensile strength not less than No. 16 (AWG) three-conductor, type SO cord. (Refer- ence: 7.7.7 IPCEA Pub. No. S-19-81, Fourth Edition.) Cables (cords) con- structed with heavy jackets shall con- sist of conductors not smaller than No. 14 (AWG) regardless of the number of conductors. § 18.37 Lead entrances. (a) Insulated cable(s), which must extend through an outside wall of an explosion-proof enclosure, shall pass through a stuffing-box lead entrance. All sharp edges that might damage in- sulation shall be removed from stuff- ing boxes and packing nuts. (b) Stuffing boxes shall be so de- signed, and the amount of packing used shall be such, that with the pack- ing properly compressed, the gland nut still has a clearance distance of Vs inch or more to travel without meet- ing interference by parts other than packing. (See Figures 8, 9, and 10 in Appendix II.) (c) Packing nuts and stuffing boxes shall be secured against loosening. 50 (d) Compressed packing material shall be in contact with the cable jacket for a length of not less than '/2 inch. (e) Special requirements for glands in which asbestos-packing material is specified are: (1) Asbestos-packing material shall be untreated, not less than ^A 6-inch di- ameter if round, or not less than Vifi by ■Vi6 inch if square. The width of. the space for packing material shall not exceed by more than 50 percent the di- ameter or width of the uncompressed packing material. (2) The allowable diametrical clear- ance between the cable and the holes in the stuffing box and packing nut shall not exceed 75 percent of the nom.inal diameter or width of the packing material. (f) Special requirements for glands in which a compressible material (ex- ample—synthetic elastomers) other than asbestos is specified, are: (1) The packing material shall be flame resistant. (2) The radial clearance between the cable jacket and the nominal inside di- ameter of the packing material shall not exceed V32 inch, based on the nominal specified diameter of the cable. (3) The radial clearance between the nominal outside diameter of the pack- ing material and the inside wall of the stuffing box (that portion into which the packing material fits) shall not exceed V32 inch. § 18.38 Leads through common walls. (a) Insulated studs will be acceptable for use in a common wall between two explosion-proof enclosures. (b) When insulated wires or cables are extended through a common wall between two explosion-proof enclo- sures in insulating bushings, such bushings shall be not less than 1-inch long and the diametrical clearance be- tween the wire or cable insulation and the holes in the bushings shall hot exceed Vie inch (based on the nominal specified diameter of the cable). The insulating bushings shall be secured in the metal wall. (c) Insulated wires or cables conduct- ed from one explosion-proof enclosure to another through conduit, tubing, piping, or other solid-wall passageways will be acceptable provided one end of the passageway is plugged, thus isolat- ing one enclosure from the other. Glands of secured bushings with close- fitting holes through which the wires or cables are conducted will be accept- able for plugging. The tubing or duct specified for the passageway shall be brazed or welded into the walls of both explosion-proof enclosures with continuous gas-tight welds. (d) If wires and cables are taken through openings closed with sealing compounds, the design of the opening and characteristics of the compounds shall be such as to hold the sealing material in place without tendency of the material to crack or flow out of its place. The material also must with- stand explosion tests without cracking or loosening. (e) Openings through common walls between explosion-proof enclosures not provided with bushings or sealing compound, shall be large enough to prevent pressure piling. 18.: Ht >nduit. Hose conduit shall be provided for mechanical protection of all machine cables that are exposed to damage. Hose conduit shall be flame resistant and have a minimum wall thickness of ^16 inch. The flame resistance of hose conduit will be determined in accord- ance with the requirements of § 18.65. § 18.40 Cable clamps and grips. Insulated clamps shall be provided for all portable (trailing) cables to pre- vent strain on the cable terminals of a machine. Also insulated clamps shall be provided to prevent strain on both ends of each cable or cord leading from a machine to a detached or sepa- rately mounted component. Cable grips anchored to the cable may be used in lieu of insulated strain clamps. Supporting clamps for cables used for wiring around machines shall be pro- vided in a manner acceptable to MSHA. § 18.41 Plug and receptacle-type connec- tors. (a) Plug and receptacle-type connec- tors for use inby the last open crosscut in a gassy mine shall be so designed that insertion or withdrawal of a plug cannot cause incendive arcing or sparking. Also, connectors shall be so designed that no live terminals, except as hereinafter provided, are exposed upon withdrawal of a plug. The fol- lowing types will be acceptable: (1) Connectors in which the mating or separation of the male and female electrodes is accomplished within an explosion-proof enclosure. (2) Connectors that are mechanical- ly or electrically interlocked with an automatic circuit-interrupting device. (i) Mechanically interlocked connec- tors. If a mechanical interlock is pro- vided the design shall be such that the plug cannot be withdrawn before the circuit has been interrupted and the circuit cannot be established with the plug partially withdrawn. (ii) Electrically interlocked connec- tors. If an electrical interlock is pro- vided, the total load shall be removed before the plug can be withdrawn and the electrical energy in the interlock- ing pilot circuit shall be intrinsically safe, unless the pilot circuit is opened within an explosion-proof enclosure. (3) Single-pole connectors for indi- vidual conductors of a circuit used at terminal points shall be so designed that all plugs must be completely in- serted before the control circuit of the machine can be energized. (b) Plug and receptacle-type connec- tors used for sectionalizing the cables outby the last open crosscut in a gassy mine need not be explosion-proof or electrically interlocked provided such connectors are designed and construct- ed to prevent accidental separation. (c) Conductors shall be securely at- tached to the electrodes in a plug or receptacle and the connections shall be totally enclosed. (d) Molded-elastomer connectors will be acceptable provided: (1) Any free space within the plug or receptacle is isolated from the exterior of the plug. (2) Joints between the elastomer and metal parts are not less than 1 inch wide and the elastomer is either bonded to or fits tightly with metal parts. (e) The contacts of all line-side con- nectors shall be shielded or recessed adequately. (f) For a mobile battery-powered ma- chine, a plug padlocked to the recepta- cle will be acceptable in lieu of an in- terlock provided the plug is held in place by a threaded ring or equivalent mechanical fastening in addition to the padlock. A connector within a pad- locked enclosure will be acceptable. § 18.42 Explosion-proof distribution boxes. (a) A cable passing through an out- side walKs) of a distribution box shall be conducted either through a packing gland or an interlocked plug and re- ceptacle. (b) Short-circuit protection shall be provided for each branch circuit con- nected to a distribution box. The cur- rent-carrying capacity of the specified connector shall be compatible with the automatic circuit-interrupting device. (c) Each branch receptacle shall be plainly and permanently marked to in- dicate its current-carrying capacity and each receptacle shall be such that it will accommodate only an appropri- ate plug. (d) Provision shall be made to relieve mechanical strain on all connectors to distribution boxes. S 18.43 Explosion-proof splice boxes. Internal connections shall be rigidly held and adequately insulated. Strain clamps shall be provided for all cables entering a splice box. • • • § 18.47 Voltage limitation. • • • (d) An alternating-current machine shall not have a nameplate rating ex- 51 ceeding 660 volts, except that a ma- chine may have a nameplate rating greater than 660 volts but not exceed- ing 4,160 volts when the following con- ditions are complied with: (1) Adequate clearances and insula- tion for the particular voltage(s) are provided in the design and construc- tion of the equipment, its wiring, and accessories. (2) A continuously monitored, fail- safe grounding system is provided that will maintain the frame of the equip- ment and the frames of all accessory equipment at ground potential. Also, the equipment, including its controls and portable (trailing) cable, will be deenergized automatically upon the occurrence of an incipient ground fault. The ground-fault-tripping cur- rent shall be limited by grounding resistor(s) to that necessary for de- pendable relaying. The maximum ground-fault-tripping current shall not exceed 25 amperes. (3) All high voltage switch gear and control for equipment having a name- plate rating exceeding 1,000 volts are located remotely and operated by remote control at the main equipment. Potential for remote control shall not exceed 120 volts. (4) Portable (trailing) cable for equipment with nameplate ratings from 661 volts through 1,000 volts shall include grounding conductors, a ground check conductor, and ground- ed metallic shields around each power conductor or a grounded metallic shield over the assembly; except that on machines employing cable reels, cables without shields may be used if the insulation is rated 2,000 volts or more. (5) Portable (trailing) cable for equipment with nameplate ratings from 1,001 volts through 4,160 volts shall include grounding conductors, a ground check conductor, and ground- ed metallic shields around each power conductor. (6) MSHA reserves the right to re- quire additional safeguards for high- voltage equipment, or modify the re- quirements to recognize improved technology. § 18.48 Circuit-interrupting devices. (a) Each machine shall be equipped with a circuit-interrupting device by means of which all power conductors can be deenergized at the machine. A manually operated controller will not be acceptable as a service switch. (b) When impracticable to mount the main-circuit-interrupting device on a machine, a remote enclosure will be acceptable. When contacts are used as a main-circuit-interrupting device, a means for opening the circuit shall be provided at the machine and at the remote contactors. § 18.49 Connection boxes on machines. Connection boxes used to facilitate replacement of cables or machine com- ponents shall be explosion-proof. Port- able-cable terminals on cable reels need not be in explosion-proof enclo- sures provided that connections are well made, adequately insulated, pro- tected from damage by location, and securely clamped to prevent mechani- cal strain on the connections. § 18.50 Protection against external arcs and sparks. Provision shall be made for main- taining the frames of all off-track ma- chines and the enclosures of related detached components at safe voltages by using one or a combination of the following: (a) A separate conductor(s) in the portable cable in addition to the power conductors by which the machine frame can be connected to an accept- able grounding medium, and a sepa- rate conductor in all cables connecting related components not on a common chassis. The cross-sectional area of the additional conductor(s) shall not be less than 50 percent of that of one power conductor unless a ground-fault tripping relay is used, in which case the minimum size may be No. 8 (AWG). Cables smaller than No. 6 (AWG) shall have an additional conductor(s) of the same size as one power conductor. (b) A means of actuating a circuit-in- terrupting device, preferably at the outby end of the portable cable. Note: The frame to ground potential shall not exceed 40 volts. (c) A device(s) such as a diode(s) of adequate peak inverse voltage rating and current-carrying capacity to con- duct possible fault current through the grounded power conductor. Diode installations shall include: (1) An over- current device in series with the diode, the contacts of which are in the ma- chine's control circuit; and (2) a block- ing diode in the control circuit to pre- vent operation of the machine with the polarity reversed. S IS-.S! Electrical protection of circuits and equipment. (a) An automatic circuit-interrupting device(s) shall be used to protect each ungrounded conductor of a branch cir- cuit at the junction with the main cir- cuit when the branch-circuit conductor(s) has a current carrying ca- pacity less than 50 percent of the main circuit conductor(s), unless the protec- tive device(s) in the main circuit will also provide adequate protection for the branch circuit. The setting of each device shall be specified. For headlight and control circuits, each conductor shall be protected by a fuse or equiva- lent. Any circuit that is entirely con- tained in an explosion-proof enclosure shall be exempt from these require- ments. (b) Each motor shall be protected by an automatic overcurrent device. One protective device will be acceptable when two motors of the same rating operate simultaneously and perform virtually the same duty. (1) If the overcurrent-protective device in a direct-current circuit does not open both lines, particular atten- tion shall be given to marking the po- larity at the terminals or otherwise preventing the possibility of reversing connections which would result in changing the circuit interrupter to the grounded line. (2) Three-phase alternating-current motors shall have an overcurrent-pro- tective device in at least two phases such that actuation of a device in one phase will cause the opening of all three phases. (c) Circuit-interrupting devices shall be so designed that they can be reset without opening the compartment in which they are enclosed. (d) All magnetic circuit-interrupting devices shall be mounted in a manner to preclude the possibility of their closing by gravity. S 18.52 Renewal of fuses. Enclosure covers that provide access to fuses, other than headlight, control- circuit, and handheld-tool fuses, shall be interlocked with a circuit-interrupt- ing device. Puses shall be inserted on the load side of the circuit interrupter. Subpart C — Inspections and Tests § 18.60 Detailed inspection of components. An inspection of each electrical com- ponent shall include the following: (a) A detailed check of parts against the drawings submitted by the appli- cant to determine that: (1) The parts and drawings coincide; and (2) the minimum requirements stated in this part have been met with respect to materials, dimensions, configuration, workmanship, and adequacy of draw- ings and specifications. (b) Exact measurement of joints, journal bearings, and other flame-ar- resting paths. (c) Examination for unnecessary through holes. (d) Examination for adequacy of lead-entrance design and construction. (e) Examination for adequacy of electrical insulation and clearances be- tween live parts and between live parts and the enclosure. (f) Examination for weaknesses in welds and flaws in castings. 52 (g) Examination for distortion of en- closures before tests. (h) Examination for adequacy of fas- tenings, including size, spacing, secu- rity, and possibility of bottoming. § 18.61 Final inspection of complete ma- chine. (a) A completely assembled new ma- chine or a substantially modified design of a previously approved one shall be inspected by a qualified representative(s) of MSHA. When such inspection discloses any unsafe condition or any feature not in strict conformance with the requirements of this part it shall be corrected before an approval of the machine will be issued. A final inspection will be con- ducted at the site of manufacture, re- building, or other locations at the option of MSHA. (b) Complete machines shall be in- spected for: (1) Compliance with the require- ments of this part with respect to joints, lead entrances, and other perti- nent features. (2) Wiring between components, ade- quacy of mechanical protection for cables, adequacy of clamping of cables, positioning of cables, particularly with respect to proximity to hydraulic com- ponents. (3) Adequacy of protection against damage to headlights, push buttons, and any other vulnerable component. (4) Settings of overload- and short- circuit protective devices. (5) Adequacy of means for connect- ing and protecting portable cable. § 18.62 Tests to determine explosion-proof characteristics. (a) In testing for explosion-proof characteristics of an enclosure, it shall be filled and surrounded with various explosive mixtures of natural gas and air. The explosive mixture within the enclosure will be ignited electrically and the explosion pressure developed therefrom recorded. The point of igni- tion within the enclosure will be varied. Motor armatures and/or rotors will be stationary in some tests and re- volving in others. Coal dust, produced by grinding coal from the Pittsburgh coal bed to a fineness of minus 200 mesh, will be added to the explosive gas-air mixtures in some tests. At MSHA's discretion dummies may be substituted for internal electrical com- ponents during some of the tests. Not less than 16 explosion tests shall be conducted: however, the nature of the enclosure and the results obtained during the tests will determine wheth- er additional tests shall be made. (b) Explosion tests of an enclosure shall not result in: ( 1 ) Discharge of flame. (2) Ignition of an explosive mixture surrounding the enclosure. (3) Development of afterburning. (4) Rupture of any part of the enclo- sure or any panel or divider within the enclosure. (5) Permanent distortion of the en- closure exceeding 0.040 inch per linear foot. (c) When a pressure exceeding 125 pounds per square inch (gage) is devel- oped during explosion tests, MSHA re- serves the right to reject an enclosure(s) unless (1) constructional changes are made that result in a re- duction of pressure to 125 pounds per square inch (gage) or less, or (2) the enclosure withstands a dynamic pres- sure of twice the highest value record- ed in the initial test. • • • § 18.67 Static-pressure tests. Static-pressure tests shall be con- ducted by the applicant on each enclo- sure of a specific design when MSHA determines that visual inspection will not reveal defects in castings or in single-seam welds. Such test procedure shall be submitted to MSHA for ap- proval and the specifications on file with MSHA shall include a statement assuring that such tests will be con- ducted. The static pressure to be ap- plied shall be 150 pounds per square inch (gage) or one and one-half times the maximum pressure recorded in MSHA's explosion tests, whichever is greater. • • • § 18.69 Adequacy tests. MSHA reserves the right to conduct appropriate test(s) to verify the ade- quacy of equipment for its intended service. Subpart D — Machines Assembled With Certified or Explosion-Proof Components, Field Modifications of Approved Machines, and Permits To Use Experimental Equipment § 18.80 Approval of machines assembled with certified or explosion-proof com- ponents. (a) A machine may be a new assem- bly, or a machine rebuilt to perform a service that is different from the origi- nal function, or a machine converted from nonpermissible to permissible status, or a machine converted from direct- to alternating-current power or vice versa. Properly identified compo- nents that have been investigated and accepted for application on approved machines will be accepted in lieu of certified components. (b) A single layout drawing (see Figure 1 in Appendix II) or photo- graphs will be acceptable to identify a machine that was assembled with cer- tified or explosion-proof components. The following information shall be furnished: (1) Overall dimensions. (2) Wiring diagram. (3) List of all components (see Figure 2 in Appendix II) identifying each according to its certification number or the approval number of the machine of which the component was a part. (4) Specifications for: (i) Overcurrent protection of motors. (ii) All wirjng between components, including mechanical protection such as hose conduits and clamps. (iii) Portable cable, including the type, length, outside diameter, and number and size of conductors. (iv) Insulated strain clamp for ma- chine end of portable cable. (V) Short-circuit protection to be provided at outby end of portable cable. (c) MSHA reserves the right to in- spect and to retest any component(s) that had been in previous service, as it deems appropriate. (d) Fees for testing under this sub- part shall be consistent with those stated in § 18.7. (e) When MSHA has determined that all applicable requirements of this part have been met, the applicant will be authorized to attach an approv- al plate to each machine that is built in strict accordance with the drawings and specifications filed with MSHA and listed with MSHA's formal ap- proval. A design of the approval plate will accompany the notification of ap- proval. (Refer to §§ 18.10 and 18.11.) (f) Approvals are issued only by Ap- proval and Certification Center, Box 201B Industrial Park Road. Dallas Pilie, Triadelphia, W. Va. 26049. £5]^....., '^'^^-(^ftP :,.rE 53 >!^JS,^ c'esrsvr f/Sum 9- MUM -^^-^ PART 75— MANDATORY SAFETY STANDARDS— UNDERGROUND COAL MINES Subpart I — Underground High-Voltage Distribution 75.800 High-voltage circuits; circuit break- ers. 75.800-1 Circuit breal DISCB internal components, FIGURE 4, - DISCB control installed underground. 59 HAULAGEWAY MODEL Background The management of Federal No. 1 Mine of Eastern Associated Coal Co. , Grant Town, W. Va. , expressed an interest in utiliz- ing the system to protect a 1-mile sec- tion in the oldest but still actively used area of the mine. Prior to commit- ment they requested a laboratory demon- stration of the DISCB basic functions using prototype hardware and a simulation of the particular haulage section. The Bureau of Mines, therefore, has recently constructed and successsfully operated a lumped parameter simulation of the rail section, protected by the actual DISCB equipment. empties, connect the rotary dump area with the active sections of the mine. At the No. 1 substation a 500-kW mercury arc hewittic rectifier was tied into the system through a circuit breaker having an overcurrent setting of 2,500 A. It has since been replaced with a solid- state unit. The positive No. 9 section copper trolley is paralleled part of the way by a 1,590 kcmil aluminum feeder cable, tied to the trolley at 200-ft intervals. The track conductors consist of 85-lb double- bonded rails. The distance between trolley and feeder is 12 in; between trolley and rail it averages 72 in. Federal Haulageway The Federal No. 1 Mine was visited to gather data on a portion of the rail haulage fed from a single 300-V source shown in figure 5. Two parallel track entries, one for loads and the other for The available locomotive loads are: Two 50-ton locomotives with four 160-hp motors, six 37-ton locomotives with four 120-hp motors, and two 15-ton locomotives with two 150-hp motors. Numerous utility vehicles of 150 hp and less are also used. ^ P^^ \^ FIGURE 5. - Portion of Federal No. 1 haulage used for model. Theoretical Analysis The rectifier can be represented by the equivalent circuit shown in figure 6. Mine rectifiers generally are found in one of two configurations: The three- phase bridge and the six-phase double wye (12). It can be shown that the operation of both of these circuits is equivalent (14) . The steady-state regulation curve of either circuit is shown in figure 7. The effective source resistance, V/I, is not constant but is lower in the over- load range than for the short circuit. The source resistance, Rg, may be calcu- lated given the per-unit reactance and resistance of the transformer rectifier. For a 500-kW unit, typically percent R equals 1.1, percent X equals 7.5, and percent Z equals 7.6. 60 Overload range ^SOURCE AAAAr 1-SOURCE -onrvnn — FIGURE 6. - Direct current mine power supply. Short circuit FIGURE 7. - Rectifier voltage regulation. ^— I Assuming an infinitely stiff source feeding a 500-kW three-phase bridge rectifier, the ac impedance can be calculated as (3, pp. 12-17) 300 = 128 V, and Therefore, and •LINE - NEUTRAL I.35/3 1.35/3 Iline = 0*816 Idc = 0'816 (1,666) = 1,360 A, ZgASE = 128/1,360 = 0.094 ^. Rac = (0.11)(0.094 n) = 1.03 mn, Z^Q = (0.076)(0.094 fi) = 7.14 mfi, Xac = (0.075)(0.094 fl) = 7.05 mfi, X/377 7.05 (1C"3) 377 'AC ^ ^/~>'/ TT=7 18*7 yH, For the overload range the equivalent dc circuit impedance is ( 14 ) ^SOURCE = 6 fL;^c + 2 R^C = (360)(18. 7)10-6 + 2( 1 .03) (10-3) = 8.79 ma. For the short-circuit case RSOURCE = ^ ^AC= /3 (7.14)(10-3) = 12.37 mfi The equivalent source inductance is essentially constant and equal to (14) 'SOURCE 1.65 L.p = 1.65(18.7)10-6 = 31 ^h. Since the DISCB detects relatively low levels of fault current, the equivalent source resistance for the overload range was chosen for the model. The theoretical dc resistance at 20° C for 400 kcmil, figure 9 hard-drawn copper trolley wire is (1) 0.02687 n/1,000 ft. For the 1,590 kcmil aluminum feeder it is (J^) 0.01091 fi/1,000 ft, or roughly equivalent to 1,000 kcmil copper. So the paralleled trolley and feeder resistance is 0.00755 n/1,000 ft The resistance of two 85-lb rails cross-bonded at 200-ft intervals and hav- ing 33 bonded joints per rail per 1,000 ft is (6^) 0.0064 fl/1,000 ft. Actual measurements (9^) of unbonded joints indicate that their resistance averages 50 times that of a well-bonded joint. Resistances of unbonded 85-lb rail joints have been measured (2^) to be 0.025 Sl» In simulating poor bonding for a pair of 85-lb rail it is assumed that 70 pet of the joints are unbonded. Thus the dc resistance becomes 0.335 fi/1,000 ft. Because the DISCB imposes a 3-kHz sig- nal directly onto the haulage system con- ductors, the importance of skin effect was considered. Let R' be the effective ac resistance for a linear cylinderical conductor and R the dc resistance; then the aluminum feeder x = 14.92, k = 5.53, so ac resistance is 0.0637 fi/1,000 ft and, for trolley and feeder is parallel, R kR, where k can be determined from standard references O, p. 4-29) in terms of X = 0.0636 /^ where f = frequency in hertz, \i = magnetic primability of the conductor (assumed constant), and R = dc resistance at 20° C. 3kHz 0.0373 fl/1,000. For steel rails the value of y, and thus R' , will vary and should be deter- mined by test. Measured ( 16 ) values of ac resistance versus current indi- cate that between 500 and 800 A, R' is almost constant and a maximum. As this range is of interest for the DISCB, an approximate extrapolation of the curves yielded R'3kH2 = 0.3273 J^/1,000 ft for 85-lb double-bonded track. The inductance of any trolley system configuration may be calculated theoret- ically by several methods (2_, _7) with the following assumptions: 1. All conductors are nonmagnetic. 2. All conductors are cylindrical. 3. Constant spacing exists between conductors. 4. Rail self -inductance is negligible. 5. The cross-sectional area of feeder is added to trolley and/or rails. Accurate field measurements of system inductance yields results in substantial agreement with the theoretical values. Therefore, it was not considered neces- sary to choose inductance values for the haulage model based upon rigorous theo- retical calculations; instead, they are reasonable estimates from field surveys (_5, pp. 9-1, 9-13) of systems similar to Federal No. 1. Thus For the 9-section copper trolley at 3,000 Hz. = 0.0636 M^^ . 9.25. K = 3, / 0.142 60, so the resistance of the trolley to a 3- kHz voltage is 0.09734 fi/1,000 ft. For L9s&85# - 0'^ mH/1,000 ft, Xl = 9.3 fl/1,000 ft at 3 kHz and L9s|JAi&85# " ^.3 mH/1,000 ft, Xl = 5.7 J2/l,000 ft at 3 kHz. 62 In general, the use of parallel feeder conductors decreases inductance while greater conductor separation increases it. The shunt capacitance between the sys- tem conductors can be determined by indi- vidually calculating capacitance to neu- tral points and combining the resultant values in series and parallel as neces- sary. The equation that is used is ( 15 , pp. 77-83) 0.0388 log (Di/R;) pf /mile, vyw^ FIGURES. vwv 22 ii (Includes lights) Electrical model of mine haulage locomotive. where ^fi - the capacitance of a conduc- tor to a neutral point, R j = the radius or equivalent radius of the conductor. such as pumps and lights distributed along the haulage were simulated using 90 fi per 500 ft. Construction of the Model and Dj = the distance to the neutral point between conductors. The values arrived at by these calcula- tions are, for the trolley and or feeder and track, Cfg equals 0.016 pF/1,000 ft; and for the trolley and track, C^ equals 0.005 pF/1,000 ft. The respective shunt capacitive reactances at 3 kHz are X_ The actual haulage system routing was rearranged, as shown in figure 9, to fit on a 4- by 8-ft plywood board. It was subdivided into sections and simulated as shown in figure 10 where L is the system inductance per section length. The par- allel combination of R AC and ^DC in series with R simulates dc resistance, and R/\Q + R, the ac resistance; L< is equals 3.3 kn/1,000 ft and X^, equals 10.6 sufficiently large to approximate skin kj^/1,000 ft. For modeling purposes the shunt capacitance was neglected. Large mobile haulage loads on dc mine systems utilize series field dc motors. Empirical relationships for 300-V-dc motors show that the effective inductance can be approximated by ( 12 , pp. 4-18) La = 190/hp rating (mH). The circuit simulation is shown in figure 8. The starting resistance, R^, can be varied to produce up to triple full load current. Stationary loads (11, p. 16) effect at 3 kHz, represents distrib- uted stationary loading and R3, the high resistance of poor bonding (normally jumpered) . Owing to power the lab, loading did not exceed 100 copper magnet wire to form inductors, ance values were source limitations in and fault simulations A dc. Number 8 square was wound on the lathe Appropriate resist- obtained with nickel- chromium wire noninductively wound. The demonstration board is shown in figure 11. 63 LEGEND = 9S trolley II 1.5 MCM aluminum feeder 9S trolley ^-*^ 1.5 MCM aluminum feeder ++++• 85Hb double bonded track -O- Rectifier No. I, Hewittic 300 V, 500 kW ^^ ITE circuit breaker ra Dump FIGURE 9. - Federal No. 1 haulage model. / 1 I I I I I I I I I I I I I I I I I I I I I M I I I I I + »■ — •- Rdc Lsk AAAAr ^ac VNAA/^ Rs FIGURE 10. - Lumped haulage simulation. 64 z "H '*^4r.;;^##H%i^ *****^4**44##..ff#i*#*jH4w#^l^ .s mzm Miummum *tf©ft *tcrjr»f« ^ tm MfWITTIC^ 30© y^ S©0 KH iTf cmcyif »ii«AKfii FIGURE 11, - Haulageway model. LAB DEMONSTRATION Current and Voltage Detection Upon completion of the model the dis- criminating circuit breaker controls were connected to impress the 3-kHz signal on the system at the rectifier location as shown in figures 12 and 13. The 3-kHz current flow with no external mobile load or faults connected was 1.17 A as mea- sured by the current detector. Referring to figure 9, with a 1.5-fi resistive fault at point B, the rectifier, the 3-kHz cur- rent increases to 4.42 A: the current detector relay is activated and the cir- cuit breaker trips. A simulated 15-ton locomotive placed at B drew 2.30 A at 3 kHz and did not trip the breaker. Applying the fault at point A, 3,450 ft from the source, the total high frequency current increases slightly over the no- load value, to 1.24 A. This point is past the protective range of the current detector where audio current magnitude remains relatively unchanged for resis- tive faults remote from the substation. 65 FIGURE 12. - Laboratory setup. It is here that the DISCB voltage under normal, abnormal, and no-load con- detector is needed and a simple exam- ple will illustrate this. Referring to table 1 the high frequency voltage was monitored (fig. 14) at six locations ditions. Location while A, G, U, T, it. B is at the substation and Q are remote from TABLE I. 3 kHz voltage variations Load condition Location B A G u T Q No-load 8.0 6.1 7.1 7.8 7.9 7.9 7.9 6.7 5.0 5.9 .2 1.4 6.6 6.6 6.3 4.8 5.7 6.1 6.1 .1 1.3 6.8 5.2 6.0 6.6 6.7 6.7 6.7 6.9 5.3 6.1 6.7 6.8 6.8 6.8 7.1 1 . 5-J2 fault at B 5.4 15-ton locomotive at B.... 1.5-J^ fault at A 6.3 6.9 15-ton locomotive at A.... 1,5—0 fault at G.... 7.0 7.0 15-ton locomotive at G. . . . 7.0 66 FIGURE 13. - DISCB controls at substation. 67 ^.-n T'l FIGURE 14, - Voltage measurements on model. No-load is defined as that time when only distributed stationary loads such as pumps and lights are connected on the system. The high frequency voltage is a maximum at the rectifier and drops by 22 pet at the remotest point. With a fault near the rectifier the 3-kHz volt- age throughout the system decreases 24 pet from the no-load value. The voltages at B for a fault or a locomotive differ by 15 pet. Since this margin between legitimate and illegitimate loads is in- sufficient for discrimination a voltage detector located near the no purpose. source serves Away from the substation, high current loads and faults substantially alter the 3-kHz voltage distribution. With the fault at A the signal voltage there drops to 3 pet of the no-load value. It also drops substantially with a legiti- mate locomotive load there. However, now there is an 86-pct difference in the two voltages, large enough to adjust the set- ting of the voltage detector to protect 68 against resistive faults. It is of in- terest to note that the voltage magnitude remains relatively unchanged at locations remote from the fault and the rectifier. DISCB worst-case performance is illus- trated in figure 15 with a voltage detec- tor located 2,875 ft away from the recti- fier at A. Through judicious placement of the voltage detectors it is possible to protect the entire system. Active Impedance Multiplier As described in the first section, the 3-kHz impedance of vehicles rated 25 tons and larger must be raised sufficiently to prevent nuisance tripping. This is accomplished by mounting an active imped- ance multiplier (fig. 16) on board large mobile loads. Laboratory testing of the multiplier with a simulated 37-ton loco- motive yielded satisfactory results. \ ' 1 1 _ ^\s\ ^-Current due to - VX/ 1-5 a fault - - \\\ Location of voltage detector 'V^ Voltage due to 15-ton . locomotive^ Jrip . __,^levels Current due to \~ " ]5-ton locomotive n. -____~------.^ ^V/ Voltage due to "^,^_^-l.5ii fault / Current detector ^y/ detector 600 1,200 1,800 2,400 3,00C DISTANCE FROM RECTIFIER, ft FIGURE 15. - DISCB protection. f FIGURE 16. = Active impedance multiplier (AIM) v^^ith power supply. 69 Signal currents and voltages were mea- sured with the load at the rectifier. Using the multiplier the current drawn was 1.3 A. Without it current increased to 2.5 A. The voltage at remote points remained unchanged. Moving the locomotive to point A the current level was not changed by the mul- tiplier's exclusion. However, the volt- age decreased from 5.0 to 1.0 V. Figure 17 illustrates the effect graphically. Poor Track Bonding Effects of Arcing A series of arcing fault tests were conducted to note any effect on DISCB operation. A resistive fault was applied at G in series with two steel electrodes, 0.5 inch in diameter and separated by an air gap. Arcing was initiated by bridg- ing the gap with several strands of a 19- strand No. 12 AWG wire that vaporized upon energization. The air gap was var- ied from 3/32 to 5/8 in. The presence of the arc did not affect the flow of 3-kHz current or DISCB operation. Poorly maintained or disconnected track bonds will insert an additional impedance in the rail circuit and slightly reduce the 3-kHz voltage measured at remote points. For example, with a poorly bonded track simulated between the recti- fier and G, and the 15-ton locomotive at G, there was a 10-pct reduction in the signal voltage at G over the good bonding value. 600 1,200 1,800 2,400 3,000 3,600 DISTANCE FROM RECTIFIER, ft FIGURE 17. - Effects of active impedance mul- tiplier (AIM). Rectified Versus Generated Input The DISCB and the demonstration board have been used with both a 30-kW genera- tor and a 200-kW rectifier. No differ- ence in operation could be detected. Satisfactory operation was obtained for input voltage fluctuations from 200 to 350 V dc. Further Study At present sufficient hardware is available in prototype form for small- scale demonstrations to interested coal operators or for consideration by a manu- facturer as a marketable product. It is intended to install the system at the Federal No. 1 Mine on the portion of rail haulage modeled in the laboratory. Technical advice will be furnished by the Bureau as required throughout the in- stallation and initial demonstration phases of the single-section system. The equipment will remain installed for a sufficient time to accumulate an extended performance history. The Bureau intent is to show that the unit can be operated for a 3-month period with no more than one nuisance interruption and no instance of any failure permitting the trolley line to remain energized for a sustained ground fault greater than 200 A. Typical trolley haulage systems in coal mines are powered by multiple dc sources, typically about 1 mile apart. The 3-kHz DISCB signal is impressed upon the system 70 at these substations through an oscilla- tor and power amplifier. Since the sig- nal can be applied at several separate locations, means is provided to minimize circulating audio frequency currents by selection of a master frequency and phase. The power amplifier contains a synchronizing unit that locks onto the nearest outby oscillator and disengages its own master oscillator. If for any reason the outermost master oscillator controlling the system is unavailable the next outby oscillator automatically takes over the master role and sets the fre- quency and phase of the 3-kHz voltages. By this means the integrity of the dis- criminating system is maintained even when several substations are out of com- mission. It is this interaction of DISCB power source controls that remains to be demonstrated in the Bureau's laboratory with a multisource system. strung alongside the trolley wire to carry signals for the system. For the substation breaker to close, proper data must be received through the cable. For example if the cable is broken by a roof fall the dc power cannot be energized. Also, if the detector units indicate a faulty condition, both inby and outby breakers are prevented from closing. The pilot wire carries signals to synchronize the master oscillators and provides the power to operate relays contained in the voltage detectors. Finally, it can be used to reintroduce the high frequency tone onto the trolley at points remote from the substation. Thus, the wire serves a number of vital functions. How- ever, it does require additional labor expenditures for installation and mainte- nance. So it is desirable to explore substitute techniques, such as multiplex- ing, to eliminate the pilot wire. Upon agreement with a cooperating mine, the DISCB system will be installed to protect a haulage system having at least three branches protected by separate cir- cuit breakers and fed from more than one dc source. This larger demonstration and long-term usage test will prove to the mining industry that the system is fail- safe, reliable, and effective. The present design requires that a lightweight cable con5)rising three twist- ed pairs of insulated 20 gage wire be As the 3-kHz voltages and currents are present on the system even when dc power is interrupted it is possible to detect the location of a fault by walking along the wire with ac voltmeter and noting where a minimum occurs. It appears fea- sible that the fault location can be pin- pointed automatically by sampling data from the current and voltage detectors. Ultimately, this information could be fed into a con5)uterized mine monitoring sys- tem for readout on the surface. REFERENCES 1. American Society for Testing and Materials (Philadelphia, Pa.). Standard Specification for Figure 9 Deep-Section Grooved and Figure 8 Copper Trolley Wire for Industrial Haulage. ASTM Bl 16-64, C 7.11, 1965, p. 195. 2. delong, C. P., and W. L. Cooley. Measurement of Rail Bond Impedance. Proc. of the Fourth WVU Conf. on Coal Mine Electrotechnology , Morgantown, W. Va., Aug. 2-4, 1978, pp. 6-1—6-8. 3. Fink, D. G. , and J. M. Carroll (ed. by). Standard Handbook for Elec- trical Engineers. McGraw-Hill Book Co., Inc., New York, 10th ed. , 1969, pp. 12- 17, p. 4-29. 4. Hall, P. M., K. Myers, and W. S. Vilchek. Arcing Faults on Direct Cur- rent Trolley Systems. Proc. 4th WVU Conf. on Coal Mine Electrotechnology, Morgantown, W. Va. , Aug. 2-4, 1978, pp. 21-1—21-19. 5. Helfrich, W. , P. M. Hall, and R. L. Reynolds. Time Constants of Direct Current Trolley Systems. Proc. 5th WVU Conf. on Coal Mine Electrotechnology, Morgantown, W. Va. , July 30-31, Aug. 1, 1980, pp. 9-1—13. 71 6. Jones, D. C, M. E. Altennis, and F. W. Myers. Mechanized Mining Electri- cal Applications. The Pennsylvania State University, University Park, Pa., 3d ed. , 1971, p. 211. 7. Koehler, G. Circuits and Net- works. Macmillan Publishing Co., Inc., New York, 1955, pp. 198-202. 8. Laboratorie du Centre D' Etudes et Recherches des Charbonnages de France. CERCHAR Pub. No. 1306, 1963, 8 pp. 9. Myers, K. G. Open Rail Bond Re- sistance Measurements. MSHA Investiga- tive Report C102979, 1979, 4 pp. 10. Ohio Brass Mining Equipment (Mans- field, Ohio). Line Materials. Equipment Catalog No. 70, Sec. 100, 1974, p. 1601. 12. Paice, D. A. , A. B. Shimp, and R. P. Putkovich. Circuit Breaker Devel- opment and Application. Phase I (Con- tract H0122058, Westinghouse Electric Corp.). BuMines OFR 103(l)-75, Mar. 12, 1974, 165 pp.; NTIS PB 248 310. 13. . Circuit Breaker Develop- ment and Application. Phase II (Con- tract H0122058, Westinghouse Electric Corp.). BuMines OFR 103(2)-75, Mar. 12, 1974, 75 pp.; NTIS PB 248 311. 14. Schaeffer, J. Theory and Design. Inc., New York, 1965, Rectifier Circuits John Wiley & Sons, 265 pp. 15. Stevenson, W. D. , Jr. Elements of Power System Analysis. McGraw-Hill Book Co., Inc., New York, 3d ed. , 1975, pp. 46-61, 77-83. 11. Paice, D. A. Portable Calibrator for DC Circuit Breakers (Contract H0122058, Westinghouse Electric Corp.). BuMines OFR 73-79, July 1978, 43 pp.; NTIS PB 297 732. 16. Trueblood, H. M., and G. Wanchek. Investigation of Rail Impedance. Elec- trical Eng. , December 1933, p. 905. 72 APPENDIX. —RECOMMENDATIONS FOR FIELD INSTALLATION At present, there are no guidelines covering the installation of the DISCB in an underground mine. Since haulage sys- tems vary in size and shape, they must be analyzed individually. Once a cooperative agreement is reached with mine mangagement, an up-to-date mine electrical map should be obtained. It should show the routing and size of the trolley haulage conductors and the loca- tions of all power sources and circuit breakers. Additional design data and special sys- tem features can be determined on the initial mine visit during a nonproduction shift. Incandescent lamp distribution and pump locations should be noted as well as modifications to enhance trolley phone performance such as coiled leads at substations or capacitors across dead blocks. The number of active impedance multipliers needed can be determined by tabulating the sizes and horsepower rat- ings of the larger locomotives. The heavy current welders for bonding rails must use inductive resistors to prevent nuisance tripping. technique ( 12 , pp. 2-21) shown in figure A-lA , the oscillator frequency is adjust- ed for resonance (V^ is in phase with V^) and the values of C, f, V^, and Vg are recorded. Another approach O, p. 9-5) is shown in figure A-lB. The current is recorded upon fault through the test resistor. The time constant of the sys- tem can be determined by measuring the time it takes the current to reach 0.637 of the peak value. By connecting a portable 10-V, 3-kHz oscillator and monitoring the current, the no-load effects of pumps and lights can be measured. A signal voltage dis- tribution similar to figure 17 can be obtained by taking voltage readings on board a small vehicle of about 100 hp as it traverses the system. Battery pow- ered voltage recorders can be installed at key locations underground and left running during production time. This information is helpful for establishing voltage detector protection zones and settings. Finally, a computer model of the sys- tem can reinforce the analysis and can be During this initial visit, installation details of the DISCB system can be dis- cussed. The controls will be located nearby the rectifiers so these areas should be inspected. The pilot wire can be conveniently supported using existing communication wire hooks if available. Typically, electrical noise on mine trolley systems is less than 0.1 V at 3 kHz ( 11 , p. S).! However, since substan- tially higher values occasionally have been recorded measurements should be made. Resistance and inductance can be ap- proximated theoretically, given conductor size and separation. Underground tests can yield more exact values. In the Underlined numbers in parentheses refer to items in the list of references pre- ceding this appendix. Trolley wire Oscilloscope ( / Trolley wire 5il FIGURE A»1. Rails Testing high frequency characteristics. 73 updated as the system changes. In this manner simultaneous loads and faults can be simulated easily. Including the ef- fects of distributed stationary loading the circuit consists of lumped u sections representing 500 ft of trolley wire and incorporating nodes for calculation pur- poses as shown in figure A-2. -orwv .i'WW _rvw\ /WY^. J'WW FIGURE A-2. - Computer simulation. 74 INTERMITTENT DUTY RATING OF TRAILING CABLES By George J. Conroy 1 and Herman W. HiH2 ABSTRACT Federal Bureau of Mines sponsored research conducted at Penn State Univer- sity and West Virginia University has resulted in recommendations to the Mine Safety and Health Administration (MSHA) concerning the maximum current ratings for trailing cables used at various duty cycles. If approved, the ratings would permit smaller size cables than those presently required by 30 CFR 18, yet would provide equivalent safety when pro- tected by circuit breakers which include overload trip capabilities. Computer and calculator programs for calculating allowable ampacity (current capacity) are presented. INTRODUCTION If conductor size for a continuous miner cable is chosen on the basis of continuous-duty ampacity, the result is a very large-diameter, and consequently, very heavy trailing cable. This is a hardship in the manual handling of the trailing cable, particularly as the miner backs out to permit cleanup and roof bolting. The usual solution has been to use as small a cable as the local inspec- tor will permit, down to AWG 4/0 or smaller, without resorting to any partic- ular references or guidelines other than a general idea as to what acceptable insulation temperature should be. This temperature depends on the duty cycle of the machine. Consequently, an enforce- ment and compliance problem exists in that changes in mining pattern or strata, or even changing the machine operator, can transform a safe cable choice to an unsafe one without there being any standard available by which the unsafe condition can be judged. This paper describes a method of determining an intermittent-duty ampacity for trail- ing cables. The ampacity value can be ^Supervisory electrical engineer (re- tired). Bureau of Mines, Pittsburgh Research Center, Pittsburgh, Pa. ^Assistant Professor of Electrical Engineering, West Virginia University, Morgantown, W. Va. updated with changing mine conditions so that dangerous situations can be anticipated. At the request of MSHA, a 2-year se- ries of cable tests were performed by the Pennsylvania State University (PSU) to study the effects of varying duty cycles on cable temperatures and to find what modifications of the circuit pro- tection devices would be necessary in order to maintain safe operation if intermittent-duty ratings higher than the continuous-duty ampacities were permitted. The data from these tests were analyzed by both PSU and West Vir- ginia University (WVU), and specific conclusions have been reached. An effective summation of the findings concerning circuit protection is to be found in the doctoral dissertation of George Luxbacher of PSU entitled "Evalu- ation of the Effectiveness of Molded- Case Circuit Breakers for Trailing-Cable Protection, " November 1980. An impor- tant conclusion is that by including thermal overload elements in the breakers it is possible, with proper adjustment, to adequately protect the cable despite large variations in the duty cycle. Please note that the inclusion of these elements is essential for safety, if rat- ings are based on intermittent duty in a situation where the duty cycle can vary. 75 METHODS OF CALCULATION } WVU's analysis of the PSU data, in com- bination with previous Bureau research on conductors by Derek Paice of Westing- house and others, provided reconnnenda- tions for sizing conductors, in the form of equations, nomographs, and a calcu- lator program. A good representation of the thermal data, vrLthin the limits of experimental error, is given by the equation: 0.13 /A (1) where and A = the cross-sectional area of the copper conductor, 3 in circular mils. the thermal minutes. constant, in A single time constant is considered suf- ficiently accurate with regard to both heating and cooling. The relationship which yields a new cable rating for int^nnittent duty is then ^int duty where cont ratine 1 - exp (-T,/c) 1 - exp (-T2/c) ' (2) ' , = the total time in minutes of a cycle of operation, T2 = the operating time in min- utes or "on" time within a cycle, during which the current flows; ^A convenient relationship, relating AWG size to conductor area, is the expression: A = 105500 exp (-0.232 W) where W is the wire size and A is the area in circular mils. This holds true for all AWG wire sizes with the proviso that the larger sizes are represented as 1/0: 2/0: 3/0: 4/0: and cont rat i n< i nt duty the time constant from equation 1, the continuous current rating, the intermittent rating. duty Further explanation is required regard- ing the flow of current during a particu- lar operating time. It is rare that a mining machine would be operated in a manner such that current drain would be- have as a step function, as shown in fig- ure 1. The more usual behavior is as seen in figure 2, where many peaks and valleys of current drain occur and there 500 2.5 5.0 TIME, min FIGURE 1. - Ideal 50 pet duty cycle. 2 4 TIME, min FIGURE 2. - Realistic current trace for continu- ous miner. 76 are also multiple levels of even the average current, during each cycle. Therefore, while the preceding calcula- tions yield an intermittent duty rating for any given on time, it may not be a simple matter to determine whether this rating is being exceeded, on the average, during the on time. The choice of an on time, itself, may not be totally simple, as there may be low current drains, such as from headlights or idling pump motors, throughout the cycle. These small cur- rents contribute almost insignificantly to the heating of the trailing cable; yet their presence con5)licates the defining of the operating cycle. Some arbitrary ground rules become necessary, such as deciding that a single root-mean-square (rms) average current will represent the drain and that the machine will be con- sidered to be on whenever the short-term average current exceeds 25 pet of the continuous-duty rating of the trailing cable. Thus, for a machine powered by an AWG 4/0 cable (continuous current rating of 180 A) and yielding the current drains shown in figure 2, T2 would be 2.25 min out of a total cycle time of 5.0 min. This would result in a cable current max- imum rating of A int = 180 duty by equation 2. exp (-5.0/60) ^ exp (-2.25/60) ^^^ ^ It is possible to instrumentally deter- mine the true rms value of the current. However, it is probably adequate for present purposes to take the envelope of the peak reading at each major step, as they appear on an instrument having iner- tia in its movement, and estimate an "average" value from this. Then Note that all of the current drains have been included in the average, even though the duration of the on time was set by ignoring the lowest values of drain. Comparing this true value with the cal- culated rating, it appears that the cable is adequate for the observed load. A simple assumption for the initial calculation of the rating might be to agree that intermittent duty will be arbitrarily defined for the rating pur- pose as some cycle such as 60 min com- prised of 50 pet on time, 50 pet off time. For the preceding example, this would have yielded an intermittent duty rating of 252 A. It would still be necessary to use the rms averaging tech- nique on the actual usage data in order to check operation against the calcu- lated cable rating. Just what the most representative arbitrary cycle might be — 50-50, 25-75, etc. — has never been decided. INSTANTANEOUS TRIP SETTINGS Thermal-magnetic trip circuit breakers with the thermal overload units sized at or below the cable's continuous duty ampacity may allow use of higher currents with providing adequate protection with regard to routine loading at all inter- mittent duties. However, nothing in this paper should be construed as suggesting that the size of fuses or the instanta- neous trip setting of circuit breakers may be increased above the values per- mitted by present regulations for a par- ticular size of cable. These values are determined by short-circuit considera- tions, not by long-term thermal effects. z (InTn) Z T, (3) ^8002 X 0.05 + 652 X 0.35 + 5752 x 0.20 + 2752 X 0.85 + 4752 X 0.8 + 152 X 2.75 5.0 = 263 A by equation 3. NOMOGRAPHS 77 Figures 3 and 4 conprise a set of nomo- graphs that may be used to determine the thermal time constant of the cable and then, directly, a cable rating factor, K, which is equivalent to the entire radical expression in equation 2, such that i nt duty = K I IPCEA* (4) where I|pcea ^^ '-^^ continuous ampacity rated by the Insulated Power Cable Engi- neers Association (IPCEA). Equation 4 can be used to find cable rating factors directly. Nomograph 1 (fig. 3) is used with the on time, t2, and the cable time constant (or the cable heating time constant) to deter- mine a heating factor, H. The total cycle-time, t,, is used with the cable time constant (or the cable cooling time constant) to determine a cooling factor C on the same nomograph. Using the values of H and C on nomograph 2 (fig. 4), one obtains the cable rating factor, K. One example is worked out on figures 3 and 4 for No. 6 round cable Cooling factor H Heating factor Cable rating factor with = 30 1. 2 — ^ "" " ) 1- 1 — min, thus K = 1.97, 20 min, and time FIGURE 3, - Intermittent duty rating nomograph 1. PROGRAMS The appendix to this paper gives HP41C4 and HP97 programs for calculating individual rating values. In addition, it gives a BASIC program listing for obtaining tabulated intermittent duty ampacities applicable to 30 CFR 18 appen- dix I rating tables. DRAG CABLES The rating factors obtained by apply- ing the described techniques, with 30 CFR 18 continuous ratings as a base, are not at this time approved by MSHA but do represent Bureau of Mines best judg- ment on realistic values for cables that may or may not be on reels during part of the working cycle. If only drag cables are considered, the Title 30 con- tinuous duty ampacities are probably '^Reference to specific programs does not imply endorsement by the Bureau of Mines. excessively conservative, and even higher intermit tent -duty ratings for drag cables would be possible, if based on revised 30 CFR 18 tables. This exception is im- portant because the greatest need for higher ratings may occur as handling dif- ficulties are experienced with large- diameter drag cables, and it may be pos- sible to use smaller cables for the given currents. The solution to this problem must be in modification of the continuous duty tables of 30 CFR 18, not in altering the intermittent-duty calculations. 78 ■f-H \ h H h 4 3 2.5 2 1.5.1.3 1 .9 0.70.60.50.40.3 0.2 0.1 0.050.01 C, cooling factor / H, heating factor Cable size 40 FIGURE 4. - Intermittent duty rating nomograph 2. VOLTAGE DROP As the stated calculations are based on thermal considerations (that is, maximum permitted temperature) and do not take into account the effect of the higher currents on voltage drop in the cable and other circuit elements, it is recommended that the trailing cable intermittent-duty current not exceed about 250 pet of the continuous rating, whatever the calcula- tions may give. CONCLUSIONS This discussion has presented a practi- cal method of determining an acceptable intermittent-duty loading of trailing cables, although further consideration by MSHA would be required before ratings which result from the calculations could be considered as applicable to the coal mining industry. APPENDIX. —PROGRAMS 79 HP41C Program, Cable Ampacity Rating Table Buildup — Cycle Time in Minutes, Operating Time in Percent 01 LBL AMP 02 LBL 00 03 CYCLE T? 041 AVIEW 051 STOP 06 LBL A 07 STO 03 08 PCT ON? 09 AVIEW 10 STOP 11 LBL B 12 ENTER+ 13 .01 14 * 15 STO 01 16 WIRE? 17 AVIEW 18 STOP 19 LBL C 20 STO 02 21 249 22 ENTER+ 23 RCL 02 24 XY? 30 GTO 05 31 * 32 GTO 05 33 LBL b 34 SQRT 35 .13 36 * 37 STO 04 38 RATED? 39 AVIEW 40 STOP 41 LBL D 42 STO 05 43 2.5 44 * 45 STO 06 46 RCL 05 47 EXECUTE 48 AVIEW 49 STOP 50 LBL E 51 RCL 03 52 RCL 01 53 * 54 STO 09 55 FIX 1 56 ON T= 57 ARCL 09 58 AVIEW 59 RCL 09 60 RCL 04 61 / 62 CHS 63 EX 64 CHS 65 1 66 + 67 1/X 68 ENTERf 69 RCL 03 70 RCL 04 71 / 72 CHS 73 EX 74 CHS 75 1 76 + 77 * 78 SQRT 79 RCL 05 80 * 81 STO 11 82 ENTERf 83 RCL 06 84 XY? RCL6 PRTX SPC SPC R/S 2 3 2 CHS 072 *LBL5 073 R/S 074 075 076 077 078 079 080 081 082 083 084 085 086 087 088 089 1 5 5 X RTN R/S -21 36 03 36 04 -24 -22 33 -22 01 -55 -35 54 36 05 -35 35 06 -21 36 02 16-34 36 06 -14 16-11 16-11 51 21 05 51 -62 02 03 02 -22 -35 33 01 00 05 05 00 00 -35 24 51 SAMPLE RUNS 5-min cycle time 50 pet on (enter 50") "0" then 2" for AWG 2 90 A from 30CFR18 New I value "0" then "1" 30CFR18 value New I value "0" then ")" 30CFR18 value New I value Same cycle Vary percent on to "10" AWG #0 (same) 30CFR18 (same) New I value 5.*** A 0.50 *** B 66335. ***C,R/S 90 *** D *** E 125. 83656. ***C,R/S 100. *** D 239. *** E 105500. ***C,R/S 120. ***D ***E 167. 5, *** 0.10 *** B 105500. *** C 120. *** D *** E 83656. *** C Vary AWG # to "4" 30CFR18 100. *** D New I value *** E 250 pet of 30CFR18 rating 300. 250. .1 pet on 1.000000000-03 *** B AWG #1 83656. *** C 30CFR18 100. *** D New I value *** E 250. (NOTE; 250 pet is maximum allowed.) For additional values, enter only the changed parameters, depress proper letter, then "E". Input Values A - Cycle time B - Percent on C - Cable area1 D - 30CFR18 rating E - Execute known, enter "0" (zero) depress then enter AWG nixmber, 81 Conyuter BASIC Program AMPCAP 19-FEB-81 08:06:03 5 DIM 12(60,10) 6 REM CALCULATES AREA FROM WIRE SIZE OR USES AREA DIRECTLY, TO OBTAIN AVERAGE 7 REM THERMAL TIME CONSTANT OF CONDUCTORS. 10 PRINT "AMPACITY TABLE CALCULATION" 11 PRINT "THERMAL RATING ONLY. VOLTAGE DROP SHOULD ALSO BE CONSIDERED." 20 PRINT " " 30 PRINT " " 40 PRINT "ENTER WIRE SIZE, RATING FROM 30CFR18, AND MAX TIME IN MINUTES." 41 PRINT "FOR SIZE 00 USE -1, FOR SIZE 000 USE -2, FOR SIZE 0000 USE -3" 50 INPUT W,R,L 55 R1-2.5*R 60 IF W<250 GO TO 190 70 A1=1000*W 80 C=.13*SQR(A1) 90 FOR T=l TO L 100 FOR P=l TO 10 110 P2=.1*P 120 Nl=l-(2.71828-(-T/C)) 130 Dl=l-(2.71828-(-P2*T/C)) 140 K=SQR(N1/D1) 150 I2(T,P)=INT(K*R) 155 IF I2(T,P)>R1 THEN 12(T,P)=INT(R1 ) 160 NEXT P 170 NEXT T 180 GO TO 240 190 REM CALCULATES AREA 200 Al=105500*(2.71828-( .232*W)) 210 GO TO 80 220 REM PRINTING FORMAT MAY BE IMPROVED BY BASIC 'PRINT USING' STATEMENT 230 REM IF YOUR COMPUTER SUPPORTS IT. PERCENT OPERATING TIME" 30";" 40";" 50";" 60";" 70";" 80' 240 PRINT "CYCLE";" ";" PERCE 245 IF R<90 GO TO 250 246 PRINT "(MIN)";" 10";" 20"; 90";" 100" 247 GO TO 255 250 PRINT "(MIN)";" 10";" 20"; 90";" 100" 30";" 40";" 50";" 60";" 70";" 80";' 90";" 100" 255 PRINT " " 260 FOR T-1 TO L 270 PRINT T;" ";I2(T, 1) ; I2(T,2) ; I2(T,3) ; I2(T,4) ; I2(T,5) ; I2(T,6) ; I2(T,7) ; I2(T,8) ; I2(T,9);I2(T,10) 280 NEXT T 290 PRINT " " 300 PRINT " " 310 PRINT "next?( Y or N )" 320 INPUT Q$ 330 IF Q$="Y" GO TO 20 340 PRINT " SIGN OFF" 82 SEMICONDUCTING RUBBER AS A LOW-VOLTAGE SHIELD FOR PERSONNEL PROTECTION By J. N. Tomlinson'' and L. A. Morley2 ABSTRACT Semiconducting rubber insulation is used in high-voltage distribution cable to provide a gradual transition of poten- tial, in order to avoid the occurrence of harmful corona. Metallic shielding is used in both distribution and trailing high-voltage cables to assure that any fault to a phase conductor will be a rel- atively low current phase-to-ground fault through the shield rather than a highly energetic phase-to-phase fault. A strong desire of users of mining machines em- ploying cable reels has been to find a way to make a cable with semiconducting shielding take the place of metallically shielded cable, which is so much heavier, larger, and less flexible. Research shows no semiconducting formulation capa- ble of conducting the fault current of several amperes that is usually consid- ered necessary for operations of current- operated ground fault protection devices. A voltage-sensing protection method has been found, however, employing an addi- tional conductor in the cable. This method is so sensitive that, as discussed in this paper, the circuit can be inter- rupted before the semiconducting rubber in the vicinity of a phase-to-ground fault is carbonized. Therefore, it should be possible to provide cable reel installations with short-circuit protec- tion almost equivalent to that provided by metallic shielding. INTRODUCTION Shielding of low-voltage trailing ca- bles, such as those used on shuttle cars, offers improved personnel safety but is not easily implemented simply because the usual shield designs cannot withstand the excessive flexing. Under Federal Bureau of Mines Contracts G01883063 and J0199106,4 evaluations have been made of shield designs using copper-copper, copper-cotton, and copper-nylon braid combinations as well as copper stranding with semi conductive rubber materials. None of those proved to be satisfactory mainly because of copper fatigue. How- ever, semi conductive rubbers are able to withstand the flexing and might provide a ^Instructor of Mining Engineering, Pennsylvania State University, University Park, Pa. ^Professor of Mining Engineering, Pennsylvania State University, University Park, Pa. suitable shield if materials having suf- ficient electrical conductivity can be used. This paper is based on a portion of the research conducted under Bureau of Mines contract J0199106 in which the authors have addressed the resistiv- ity requirements for conductive -rubber shielding materials in conjunction with fault-sensing equipment sensitivities and power-interruption times. Morley. Mine Trailing Cables and Ca- ble Splices: Shielded Cables (Contract 60188036, Pennsylvania State Univ.). BuMines OFR 81-80, Feb. 29, 1980, 69 pp.; NTIS PB 80-208135. ^Pennsylvania State University. Mine Trailing Cables and Cable Splices. September 1979, 109 pp.; available for consultation at the Bureau of Mines Pittsburgh Research Center, Pittsburgh, Pa. 83 THE PHYSICAL CASE A typical example of how shielding might improve personnel safety is illus- trated in figure 1. In this highly sim- plified case, a nail has pierced the ca- ble and made contact with an energized conductor. At the same instant, a person is in contact with the nail and also with the grounded machine frame. The result is current flow through the person as well as through the shield to the ground- ing conductor. (A similar situation could result if, for instance, a mechanic were to cut into an energized cable in preparation for making a splice repair. ) For this simplified and strictly resistive case, the portion of current through the worker is ^" ^ (Rs + V Rg + ^sK (1) where V is the phase-to-neutral voltage, Rg is the resistance through the shield from the nail to the grounding conductor, R is the ohmic value of the grounding resistor, and R^ is the worker's effec- tive resistance which can vary apprecia- bly depending on contact resistance, body weight, and so forth. SHIELD RESISTANCE The situation for the nail pierc- ing the semiconducting sheath is shown in figure 2, and also in figures 3 and 4 when the sheath is unwrapped from the insulated conductor to form a uniform layer of semiconducting material as illustrated. The condi- tion of the current flow in figure 4 between the nail and one side of the grounding conductor is equivalent to the heat flow from a heated cylinder^ to an exposed wall as illustrated in Penetrating object Im — ti Current through shield Is-i- ^ Grounding resistor Rs Resistance to grounding conductor (conductive rubber) Rm Resistance associated with person Grounding conductor FIGURE 1, - Parallel current paths through person and shield for a simplified circuit. 'Incropera, F. P., and D. P. Dewitt. Fundamentals of Heat Transfer. John Wiley & Sons, New York, 1981, p. 140. 84 figure 5. The analogy for electrical re- sistance, R, is thus Nail Semiconducting shield FIGURE 2. - Nail piercing cable and shorting semiconducting shield to power conductor! Nail Grounding conductor FIGURE 3, = Shielding material unwrapped from insulated conductor. ^in^:^\ 2TrL Vd / (2) where py is the volume resistivity, L is the thickness of the semiconducting layer, h is the distance from the center of the cylindrical probe (or nail) to the grounding conductor, and D is the probe diameter. Empirical and theoretical values of the resistance, R, have been obtained and are plotted in figure 6 as a function of h where h is given in terms of probe diameters. The difference between the two curves is attributed to the finite limits used in the experimental eval- uations. The curves in the figure have been normalized for p^ =1 fl-cm and L = 1 cm. Since the piercing object will in effect be somewhere between two grounding conductors (fig. 3), the resistance is more like that shown in figure 7 and a value on the order of 6 ^ might be ap- proximated if Py were equal to 1 fi-cm and the shield thickness, L, were on the order of 1 mm. MAXIMUM CURRENTS VERSUS TIME Values for I^ from equation 1 are plotted in figure 8 for three different voltages. Nail FIGURE 4. - Shielding further simplified for analysis of current between pow- er and grounding conductors. The concern in electrical shock is that the victim's heart will lose its normal rhythm and go into ventricular fibrilla- tion. The current level and time expo- sure at which this statistically occurs for a typical human has been given by DalzielS as I = 116 where amp I is in minimum current and t is duration of in milli- shock in ^Dalziel, D. F. Electric Shock Hazard. IEEE Spectrum, February 1973, pp. 44-50. 85 Piercing object ^°h Semiconducting sliielding FIGURES. Electrical path from piercing object to grounding conductor via the semicon- ducting shield material. Grounding conductor Theoretical = A_ ln(4h/D) ''v =1 -TL cm L =1 cm ■ 20 40 60 80 100 120 140 160 DISTANCE FROM PROBE TO GROUNDING CONDUCTOR, probe diameters FIGURE 6. - Theoretical versus experimental resist- ances between probe and grounding con- ductor in a conductive medium. seconds. The values of t are scaled against I^ in figure 8 and provide an indication of the time requirements for power interruption in order that the shield is effective for the operating conditions considered. VOLUME-RESISTIVITY TEST DATA Volume-resistivity measurements were made on semiconducting shield materials used in three prototype cable designs and another cable used in a similar mining application in the United Kingdom. Also, measurements were made on the conductive material used in a high-voltage cable application, and some semiconducting tape was also evaluated. The sample config- urations are illustrated in figures 9 and 10. Table 1 presents the laboratory values. TABLE 1. - Volume resistivity values for various semiconducting material samples Typical volume resistivities (fi-cm) Sample Along extrusion direction Across extrusion direction Prototype 1 Prototype 2 Prototype 3 Drill cord 750 600 280 300 12,500 240 415 500 180 130 High-voltage cable. . . Semiconducting tape. . 5,600 610 86 1.0 CJ 0.5 LU O Empirical Grounding conductor pv = 1 -TL-cm L=1 cm 20 40 60 80 DISTANCE FROM PROBE TO GROUNDING CONDUCTOR, probe diameters FIGURE 7. = Resistances with probe placed be- tween two conductors. Applying some of the values from ta- ble 1 to the curves in figure 8, it is obvious that power-interruption times would have to be extremely short for the high system voltages where the maximum ground-fault current is limited to 25 A. If R was set to limit maximum ground-fault current to 750 mA, the re- quired power-interruption times might be more practical, especially in the low-voltage cases. Materials with vol- ume resistivities orders of magnitude lower than the best values listed in table 1 would, of course, provide a much better safety margin for the shield design. The advantage of higher-ohmage ground- ing resistors is quite obvious from the curves presented in figure 8. The higher values also limit ground-fault current in the shield which might otherwise volatize the shield material and provide no shunt protection for the victim. This latter advantage is also true should copper braid bemused rather than semiconducting materials. The curves in figure 8 are generated using a resistance value of 500 SI for the victim. Volume resistivities of 100 fi-cm would produce about equal current flow through the shield and through the vic- tim, all other resistances being equal. It is also important to note that while the shield is actually being cut into, the contact resistance is much higher, and this would reduce the initial shunt- ing effect even more. FAULT-TEST EXPERIMENTS Numerous tests were conducted in which a nail was driven into proto- type cables having semiconducting rub- ber shields of the SHD configuration. Both ac, 295-V line-to-neutral and dc, 245-V line-to-neutral systems were used. Grounding resistors limited the maximum current to 15 A. Fault sensing used a zero-sequence method for the ac system and saturable-reactor sensing for the dc system. 87 1,000 500 100 1 50 o > I o o I 10 5 - ^^^^ - 950 V- ^V^ - 7 - 480 V-^ /^^^ - 208 V-^vV/ 950 V-^^.y/|^^^'^ ^^^^^^^ _ f/^-AQO V - M. - 1/ ^^^208 V ~ / - / ^^^-lm = 750mA ^^— lm-25A - / Rm = 500^ - / 1 1 I 25 S LU 50 2 100 ^ 250 3 _j 500 CD LL 1,000 ir < -J 2,500 O 5,000 z LU > 0.1 1.0 10 100 1,000 VOLUME RESISTIVITY OF SHIELD MATERIAL, ii-cm FIGURE 8. - Currents and ventricular fibrillation times versus shield resistivities for various operating voltages and ground resistor. 88 Copper paint Circumferential test sample FIGURE 9. - Sample configurations taken from semi- conducting shielded cable used to de- termine volume resistivity, pv. Razor blade FIGURE 10. - Sample configurations used in measur- ing pv for semiconducting tape samples. The results were in two general cate- gories. For the first group, the time required to interrupt the power ranged from slightly less than 1 sec to as high as 2.5 sec, depending on sensing method, puncture location, and so forth. The resistance between the nail and the grounding conductor was time dependent and decreased as the I^r heated the mate- rial. In most cases, the semiconducting material actually carbonized before the power was interrupted. In the second group, the immediate current flow vola- tized the semiconducting shield material in contact with the nail, and the power was not interrupted at all. Approxi- mately 1 out of every 10 experiments pro- duced this result. The second group in which the shield was volatized obviously provided no personnel protection. However, a higher grounding resistance would limit the current to a less destructive value and so provide an improvement for these cases. A higher grounding resistance for the first group would tend to lengthen the already-too-long time between initial contact and power interruption since the i2r dissipated in the semiconducting shield would decrease; thus, the rate of resistance change between the nail and the grounding conductor would be les- sened. The solution here is high- resistance grounding along with lew- resistivity shielding materials. Safety limitations would simply be the current- sensing abilities of the ground-fault system and associated power-interniption delays. CONCLUSIONS The results of this work suggest the semiconducting shielding might have application in low-voltage trail- ing cables under the following circumstances. (2) High-ohmage grounding resistors are required to allow more time to sense and interrupt the power and to limit current levels which might otherwise volatize the the shield material. (1) Volume resistivities on the order of 1 fi-cm must be applied to the cables of interest. It is important to note that volume resistivities measured on laboratory slab samples are usually much lower than that measured for sam- plings taken from an actual extruded shield. 1982 - 505 - 002/70 (3) High-sensitivity ground-fault de- tectors are needed. Combining semi- conductive shielding of trailing cables, very low ground-current limits, and high- sensitivity ground-fault relaying should substantially reduce electrocution haz- zards on portable or mobile mining equipment. INT.-BU.OF MINES, PGH., PA. 26334 ^D 221 * 4^ O <> ♦'TVV« .0 1' *-^„ .♦" .. ^3, 'o , . • A ^ •'TVS' .0^ ^ '" • ' * A <► *' TVT* -0^ ^oV" ^'J" > „4q. »- V '/ *^' V* ..JJil'* <^*. %.♦* /^^\ V.s^" • :-^•''\:•■^w,\.^'\^^.•J\}^\A•.^^,•,,^'■''\•' T* ,G^ o, 'o.T" A <. ♦/TV*' ,0^ *3, 'o,.'» ^ <» ♦^TTi' .0^ <^, '»..* .'V <^^ *in %,^' -ov^^ ■• X/ O J* • •. '^^ rr.^' .^^' ^'= "^..^^ « : ^v^% ^.W9K: .^^^^^ '^•i' .V* %••; ^° .'i-^ ,•1* . «^ .■ •^u..-^ v" .. **• .%- • • - .0 s^^^^ -■ ^' ^^ ^v-. '• ^"•^^. QBBS BROS. I .0^ <.'JL'*. o ."^o* V' A .*'% -.^ AUGUSTINE FLA. ' 3Z084 .<> ■% %. .• .*^\ \ •^0 '^..*^ .^^'■v.