I* . » • 
 
 
 > 
 
 
 
 
 
 ^P* »i*£* ^ V fc ^A^* ^ aP 
 
 
 
 ^ v t ^xik> ^ aP 
 
 
 ,** V .«^;. % c,0*".i^."* o o " ' /''•^i^ ' ' ,0*°,; "-•• 
 
 A ^ 
 
 
 
 
 d* 
 
 • ; ^x : -.w- : /X l ™' j^** 'iffls s*\'W. : s*x •: 
 
 X°-7V - 
 °4, ° " ° A ^ 
 
 „4o* 
 
 %* A 
 
 
 
 v> 
 
 
 
 A ^ 
 
 ** * 
 
 r o V 
 
 
 
 
 
 ^°o 
 
 ^ CU. * . « o • A J 
 
 . » • A <v *?^VV* G v <^ *« . » * A ^ 
 
 •# 
 
 ^ 
 
 ^^ :MM° %a* .-M&\ %.^ .-isSfcji-. *«.«*♦ .'Jfe*v %.>* .-isafei-. %.** .1 
 
./VliP •/% "-.w.- /\ •«••/% -IK- /\ -W-' /'** • 
 
 .% ^.-s^V /,c:^.^o /.fi,\ o°*.^.% >*.•&*,'< 
 
 >>- ■*► 
 
 
 , Ok 
 
 
 
 £ ^, - 
 
 **.«< 
 
 'o K 
 
 ^ 
 
 3* ^ 
 
 *0, ^.0^%^ rP*V '-'-!• *<> 
 
 
 
 
 xi5 ^ 
 "V-s v °^^: 
 
 C, vP 
 x,^ ^ ° 
 
 
 
 & V 
 
 
 
 4 O 
 
 vV N 
 
 . » • A <. 
 
 
 V " 1 x^> v °^ 
 
 V^ A 
 
 ^^ 
 
 
 V 
 
 -*m*; <f\ °*-^ : +**+ ''&g?S /\ 
 
 V ^ 
 
 •r 
 
 ^ 
 
 ^ v 
 
 
 ' ^ 
 
 
 
 
 
 • xv\ % X ' 
 
 
 •&g&s <? • .^ v -4P*' ^ •• .^ v ^v "-SB* 4 ,^ 
 
Bureau of Mines Information Circular/1985 
 
 Research on the Reliability of Mine 
 Electrical Equipment: A Status Report 
 
 By Dean H. Ambrose 
 
 UNITED STATES DEPARTMENT OF THE INTERIOR 
 
 75? 
 
 ^/NES 75TH A^ 
 
h, , &JU4A ■' t^l 
 
 Information Circular 9050 
 
 Research on the Reliability of Mine 
 Electrical Equipment: A Status Report 
 
 By Dean H. Ambrose 
 
 UNITED STATES DEPARTMENT OF THE INTERIOR 
 
 Donald Paul Hodel, Secretary 
 
 BUREAU OF MINES 
 Robert C. Horton, Director 
 
\ 
 
 
 
 Library of Congress Cataloging in Publication Data: 
 
 Ambrose, Dean H 
 
 Research on the reliability of mine electrical equipment. 
 
 (Bureau of Mines information circular ; 9050) 
 
 Bibliography: p. 11-12. 
 
 Supt. of Docs, no.: I 28.27: 9050. 
 
 1. Coal mines and mining— Electric equipment. 2. Electricity in min- 
 ing. I. Title. II. Series: Information circular (United States. Bureau of 
 Mines) ; 9050. 
 
 TN295.U4 622s [622'. 334] 85-600151 
 
CONTENTS 
 
 Page 
 
 1 
 
 2 
 
 2 
 
 3 
 
 4 
 
 9 
 
 10 
 
 11 
 
 13 
 
 14 
 
 Abstract 
 
 Introduction 
 
 Reliability terms 
 
 Reliability research program 
 
 Reliability data base 
 
 Application of reliability data 
 
 Conclusions 
 
 References 
 
 Appendix A. — Field data collection technique and components' sample size 
 
 Appendix B. — Field data conversion to failures per unit-year 
 
 ILLUSTRATIONS 
 
 1 . Ground fault protection system 4 
 
 2. Reliability characteristics of circuit breakers using theoretical 
 
 analysis 5 
 
 3. Reliability characteristics of circuit breakers using field data 5 
 
 4. Reliability characteristics of undervoltage relays using theoretical 
 
 analysis 6 
 
 5. Reliability characteristics of solid state undervoltage relays using 
 
 life test data 6 
 
 6. Reliability characteristics of electromechanical undervoltage relays 
 
 using field data 7 
 
 7. Reliability characteristics of ground check monitors using field data 
 
 and maintenance records 7 
 
 TABLES 
 
 1. Molded-case circuit breaker data 4 
 
 2. Undervoltage relay data 5 
 
 3 . Ground check monitor data 7 
 
 4. Splicing rates of trailing cables on mine machines 8 
 
 5. Replacement rates of trailing cables on mine machines 8 
 
 6. Motor data 8 
 
 7. Control circuitry data 8 
 
 8. Failure rate data source comparison 9 
 
 9. Maintenance intervals for mining electrical equipment 10 
 
 10. Probability of hazardous voltages of mining operations 10 
 
 A-l. Components' sample size 13 
 
 ^ 
 
UNIT OF 
 
 MEASURE 
 
 ABBREVIATIONS 
 
 USED 
 
 IN THIS 
 
 REPORT 
 
 pet 
 
 percent 
 
 
 
 yr 
 
 year 
 
 V 
 
 volt 
 
 
 
 
 
RESEARCH ON THE RELIABILITY OF MINE ELECTRICAL 
 EQUIPMENT: A STATUS REPORT 
 
 By Dean H. Ambrose 1 
 
 ABSTRACT 
 
 This Bureau of Mines publication summarizes the status of mine elec- 
 trical equipment reliability research. Failure rate data were estimated 
 from theoretical, laboratory, and field studies. Tables are presented 
 listing reliability characteristics of molded-case circuit breakers, un- 
 dervoltage relays, ground check monitors, cables, motors, and control 
 circuitry. Application of failure rate data is shown in tables listing 
 maintenance scheduling intervals of components and components that are 
 most sensitive to the probability of hazardous voltages on power dis- 
 tribution systems in the mines. Complete, accurate, and appropriate 
 failure rates on mine equipment are of the utmost importance for relia- 
 bility evaluations. 
 
 Electrical engineer, Pittsburgh Research Center, Bureau of Mines, Pittsburgh, PA, 
 
INTRODUCTION 
 
 In the past, reliable design could be 
 described simply as picking good parts 
 and using them correctly. Today, the 
 complexity of systems and the demand for 
 good reliability in many applications 
 means that sophisticated methods based on 
 numerical analysis and probability tech- 
 niques have been accepted on determining 
 the feasibility (from the reliability 
 standpoint) of systems. Classic examples 
 of ultrahigh reliability exist in systems 
 built for the military and the National 
 Aeronautics and Space Administration 
 (NASA). System availability and time 
 needed for maintenance are also important 
 to many systems. 
 
 Electrical systems developed for mech- 
 anized underground coal mines over the 
 past 25 yr have been well accepted by 
 the mining industry. Most people agree 
 that improving mining equipment reliabil- 
 ity, at least from the design viewpoint, 
 is not an easy task. Despite the need 
 for higher reliability, electrical sys- 
 tems of today are deteriorating faster 
 than necessary because of the poor qual- 
 ity of maintenance provided during the 
 life of the system (l). 2 Federal Regula- 
 tions (30 CFR 75 and 77) mandate timely 
 
 electrical preventive maintenance — 
 equipment examination, testing, and main- 
 tenance. The preventive maintenance pro- 
 grams at most mines are generally hap- 
 hazard and neglected (_2, p. 455). 
 
 A nationwide survey of electrical 
 equipment reliability, completed by the 
 Institute of Electrical and Electronic 
 Engineers (IEEE) in industrial plants 
 such as chemical, petroleum, cement, 
 etc. , revealed that inadequate mainte- 
 nance is a significant cause of equipment 
 failures (3, p. 268). In response to 
 this survey, the Bureau of Mines funded 
 several studies to address the problems 
 relating to inadequate maintenance; these 
 studies were aimed at increasing produc- 
 tion and improving mine safety by im- 
 proving mine electrical equipment 
 reliability. 
 
 In a series of investigations, the Bu- 
 reau gathered and analyzed failure rate 
 data for mine electrical equipment. This 
 report discusses the reliability data ob- 
 tained from these studies, the most im- 
 portant applications of this new informa- 
 tion, and future research that would 
 build upon the present reliability pro- 
 gram results. 
 
 RELIABILITY TERMS 
 
 Availability . — Capability of an item, 
 considering its reliability and mainte- 
 nance, to perform its required function 
 at a stated instant in time (4^, p. 333). 
 
 Confidence interval . — The range of val- 
 ues within an estimate of the failure 
 rate that will be supported (will not be 
 contradicted) by the data (_5, pp. 16-19). 
 In this paper the confidence interval of 
 A has the property: lower limit < A < 
 upper limit > 0.95; i.e., the probability 
 that the confidence interval covers or 
 contains the true value of A is at least 
 95 pet. 
 
 2 Under lined numbers in parentheses re- 
 fer to items in the list of references 
 preceding the appendixes. 
 
 Failure . — The termination of an item's 
 capability to perform its required func- 
 tion (4, p. 334). 
 
 Failure rate (A) . — The rate at which 
 failures occur in a certain time inter- 
 val; i.e., failure per unit-time. The 
 time units may be hours, years, shifts, 
 etc. (_5, p. 11). 
 
 Hazard rate . — The instantaneous failure 
 rate; the probability that a component 
 will fail in a small interval of time (_5, 
 p. 11). 
 
 Maintainability . — A characteristic of 
 design and installation expressed as the 
 probability that an item will be retained 
 in or restored to a specific condition 
 within a given period of time, provided 
 maintenance is performed in accordance 
 with prescribed procedures and resources 
 (4, p. 335). 
 
Mean time between failures. — For a par- 
 ticular interval, the total functioning 
 life of a population of an item divided 
 by the total number of failures within 
 the population during the measurement 
 period (4, p. 336). 
 
 Reliability . — The characteristic of an 
 item expressed by the probability that it 
 will perform a required function under 
 stated condition for a stated period of 
 time (4, p. 337). 
 
 RELIABILITY RESEARCH PROGRAM 
 
 The failure rates of mine electrical 
 power system components for coal mine op- 
 erating and environmental conditions were 
 estimated using one or more of the fol- 
 lowing: 
 
 1. A theoretical model based on a 
 thorough understanding of the physics of 
 failure mechanisms was used to predict a 
 failure rate. 
 
 2. A statistical estimate of the fail- 
 ure rate was made using accumulated field 
 data on failure or information from main- 
 tenance records. This technique is wide- 
 ly used in other industries. 
 
 3. Accelerated life tests were per- 
 formed by conducting magnified load tests 
 in a laboratory environment. In this 
 type of accelerated test, each sample was 
 tested at higher than normal stress lev- 
 el. It is possible to obtain the life 
 distribution of the devices and their 
 failure rates from the data collected by 
 performing these tests. This technique 
 is useful in obtaining a rough estimate 
 of the failure rate in a very short 
 time. 
 
 Technique (2) (statistical estimation 
 using field data) is by far the best and 
 most accepted approach to obtain a real- 
 istic failure rate for a component (4, p. 
 37). The field data collection technique 
 and components' sample size are given in 
 appendix A. 
 
 The selection of components for relia- 
 bility evaluation were protective devices 
 and certain mine power equipment. The 
 protective components of the mine power 
 distribution systems are typically the 
 molded-case circuit breaker, the under- 
 voltage relay (UVR) , the ground check 
 monitor (GCM), and the ground-fault relay 
 (GFR). (See figure 1.) These components 
 
 are responsible for disconnecting power 
 circuits to mining equipment under abnor- 
 mal system conditions, such as circuit 
 overcurrents , ground faults, system un- 
 dervoltages, or open ground conductors. 
 To ensure the protection of personnel and 
 equipment, abnormal conditions must be 
 detected and cleared promptly. This re- 
 quirement implies that protective compo- 
 nents must be extremely reliable. A high 
 degree of reliability also assures in- 
 creased productivity by maximizing the 
 availability of the equipment. The four 
 possible undesirable modes of operation 
 of protection devices are — 
 
 (a) failure to trip when desired, 
 
 (b) undesirable tripping, 
 
 (c) failure to close, and 
 
 (d) catastrophic failure. 
 
 The first mode of failure is the most 
 significant from the safety point of 
 view, while the other three modes affect 
 productivity (6^, p. 97-98). 
 
 The other mine power equipment investi- 
 gated were cables, motors, motor start- 
 ers, and control gear. A cable can fail 
 and result in electrical hazards to per- 
 sonnel, an ignition source for fires, or 
 loss of production due to any one the 
 following causes: 
 
 (a) Any one of the three phase conduc- 
 tors or the ground conductor or the 
 ground-check conductor can break open; 
 
 (b) short circuits due to insulation 
 failure can occur; 
 
 (c) shielding of phase conductors can 
 fail; and 
 
 (d) jackets can fail. 
 
 The causes of a motor's failure are nu- 
 merous. However, mechanical and electri- 
 cal failures and starter-control gear 
 failures can result in loss of production 
 and equipment, and in electrical hazards 
 to personnel and mine fires (6, p. 152). 
 
480 V 
 3 phase 
 
 J 
 
 War 
 
 480V/I20V ™gg^. 
 
 Mine power center 
 
 Mine face equipment 
 
 Breaker 
 
 UVR 
 
 #4H 
 
 GFR GCM 
 
 GFR- 
 
 rOi 
 
 imor 
 
 *\ 
 
 $< 
 
 Filter 
 
 £-£ 
 
 Filter 
 
 FIGURE 1. - Ground fault protection system. 
 
 RELIABILITY DATA BASE 
 
 Table 1 summarizes the results obtained 
 from theoretical analysis, accelerated 
 life tests, and field data evaluation of 
 molded-case circuit breakers. The relia- 
 bility characteristics obtained from the 
 theoretical analysis (fig. 2) show that, 
 for the breaker to have 95-pct reliabil- 
 ity, it must be removed from service be- 
 fore 1 yr of continuous service. Field 
 data (fig. 3) also indicate a removal 
 time of less than a year for greater than 
 95-pct reliability. Replacement cost 
 would be an expensive item for the mine 
 operators. With good maintenance proce- 
 dures, such as periodic inspection, cor- 
 rect examination, and calibration re- 
 quirements, the circuit breakers could 
 have >95-pct reliability for a longer 
 service life (_5, p. 74). 
 
 Table 2 summarizes results obtained 
 from theoretical analysis, accelerated 
 life tests, and field data evaluation of 
 both electromechanical (EM) and solid 
 
 TABLE 1. - Molded-case circuit breaker 
 data 
 
 
 Theo- 
 
 Lab 
 
 Field 
 
 
 retical 
 
 test 
 
 data 
 
 Failure rate ( A) 
 
 
 
 
 failues per 
 
 
 
 
 year. . 
 
 1 0.1607 
 
 2 0.30 
 
 3 0.12 
 
 95-pct confidence 
 
 
 
 
 interval of A: 
 
 
 
 
 
 l 0. 05975 
 1 0.3084 
 
 ND 
 ND 
 
 '♦o.oee 
 
 
 "+0.204 
 
 5 MTBF (1/A)...yr.. 
 
 4 6.22 
 
 ^3.33 
 
 ^8.33 
 
 ND No data 
 
 Wenkata (_5, p. 75); extrapolated (aver- 
 aged). 
 
 2 Hill and Collins (_7, p. 17). 
 
 3 Hill and Collins (7, p. 16). 
 
 ^Calculated. 
 
 5 Mean time between failures. 
 
 state (SS) UVR's. Theoretical part fail- 
 ure rates were extrapolated from rates 
 
uu 
 
 K ' ' ' ' 
 
 i 
 
 ' I 
 
 1 
 
 
 V> N 
 
 
 
 
 
 \\ v 
 
 
 KEY 
 
 
 
 \ > N. 
 
 
 
 
 90 
 
 -1 \ v 
 
 & 
 
 Upper limit 
 
 — 
 
 
 \ « > 
 
 o 
 
 Failure rate 
 
 
 
 - \ °\ X 
 
 • 
 
 Lower limit 
 
 
 80 
 
 
 \ 
 
 
 - 
 
 70 
 
 \ 
 
 V 
 
 v 
 
 ~ 
 
 
 V \ 
 
 
 *>. 
 
 . 
 
 
 \ ^ 
 
 
 v 
 
 
 s: 
 
 \ ° 
 
 \ \ 
 \ \ 
 
 \ \ 
 
 A ^ 
 
 
 V 
 
 ti 
 
 50 
 
 \ \ 
 \ x 
 \ \ 
 \ \ 
 
 \ N 
 
 
 
 - 
 
 40 
 
 
 
 
 
 
 
 
 V 
 
 
 30 
 
 
 
 
 " 
 
 
 
 
 
 
 
 
 
 
 20 
 
 
 
 
 -6 
 
 10 
 
 i i i i 
 
 1 
 
 1 
 
 i 
 
 4 6 
 
 TINIE,yr 
 
 FIGURE 2. - Reliability characteristics of cir- 
 cuit breakers using theoretical analysis 
 
 <^ 60 
 x* 
 
 UJ 
 
 a 
 
 > 50 
 
 10- 
 
 KEY 
 a Upper limit 
 o Failure rate 
 • Lower limit 
 
 4 6 
 
 TIME.yr 
 
 10 
 
 FIGURE 3. - Reliability characteristics of cir- 
 cuit breakers using field data. 
 
 TABLE 2. - Undervoltage relay (UVR) data 
 
 
 Theoretical 
 
 Life test 
 
 Field 
 
 
 EM 
 
 SS 
 
 EM 
 
 SS 
 
 EM 
 
 SS 
 
 Failure rate ( A) 
 
 failures per year. . 
 95-pct confidence interval of A: 
 
 1 0. 15 
 
 4 0.0498 
 
 '♦0.159 
 
 4 6.67 
 
 io.ne 
 
 ND 
 
 ND 
 
 1+ 7.35 
 
 ND 
 
 ND 
 ND 
 ND 
 
 2 0.49 
 
 2 0.36 
 2 0.73 
 
 2 2.04 
 
 3 0.13 
 
 t+ 0.0715 
 
 '♦0.221 
 
 4 7.7 
 
 ND 
 
 ND 
 ND 
 
 
 ND 
 
 EM Electromechanical. SS Solid state. ND No data. 
 
 ^-Venkata (_5, p. 85); extrapolated (summed). 
 
 2 Collins (8, p. 61). 
 
 3 Hill and Collins (_7, p. 16). 
 
 H Calculated. 
 
 5 Mean time between failures. 
 
 tabulated by the United Kingdom's Atomic 
 Energy Agency (UKAEA). A comparative 
 analysis shows that the SS UVR performs 
 (theoretically) better than the EM UVR. 
 One reason might be that the SS UVR has 
 
 fewer mechanical components. However, as 
 
 shown in figure 4, both UVR's degrade to 
 
 the 95-pct reliability levels within 6 to 
 18 months of operation. 
 
100 
 
 100 
 
 90 
 
 80 
 
 70- 
 
 x 
 w 
 
 Q 
 
 60 
 
 ^50- 
 
 < 
 
 40- 
 
 30 
 
 20 
 
 iv — r | i | i | i i i 
 >. N « N KEY 
 
 \\ \ a EM upper limit 
 — \ ^ N O EM failure rate _ 
 
 \ c\ »^ • EM lower limit 
 
 \ \- \ # A SS failure rate 
 \ \\ \ 
 
 A \ A ^* 
 
 
 \ " \ v - 
 
 \ % *\ ^ 
 
 — \ \ •. *x — 
 
 \ ^ ^ 
 
 \ \A ^ 
 
 \ Q\ ^ " 
 
 \ \\ ^J' 
 
 A "\ v 
 
 \ V\ 
 
 \ W 
 
 \ v - \ 
 
 X °^ x * 
 
 \ \ A 
 
 \ * X 
 
 \ °^ ^ 
 
 \ v A 
 
 \ > ^ 
 
 \ x cx v 
 
 \ ■> A^ 
 
 \ ^ X - 
 
 \ VV " 
 
 \ X °0> 
 
 
 \ ^ N 
 
 1 . 1 1 1 ■ 1 
 
 2 4 6 8 
 
 TIME.yr 
 
 FIGURE 4. - Reliability characteristics of under- 
 voltage relays using theoretical analysis. 
 
 x 
 
 UJ 
 Q 
 
 CD 
 < 
 
 UJ 
 
 or 
 
 KEY 
 A SS upper limit 
 o SS failure rate 
 • SS lower limit 
 
 4 6 
 
 TIME,yr 
 
 "I 
 
 -A 1 
 
 8 
 
 =8 
 10 
 
 FIGURE 5. - Reliability characteristics of solid 
 state undervoltage relays using life test data. 
 
 Life test data (fig. 5) showed a de- 
 gradation of 95-pct reliability of SS UVR 
 within 2 months of continuous service, 
 while the failure rate determined from 
 field data for an EM UVR was less than 
 the failure rate predicted in a theoreti- 
 cal analysis. This suggests that the 
 part failure rates used for the predic- 
 tions were not too pessimistic. As shown 
 in figure 6, the EM UVR falls below the 
 95-pct reliability level after 3 to 8 
 months of continuous service. This means 
 that if the UVR's were to be extremely 
 reliable, they would need replacement be- 
 fore completion of a continuous service 
 of 1 yr. The level of reliability of UVR 
 could be raised about 95 pet with suit- 
 able modifications in the design or main- 
 tenance of the relay (8, p. 84). 
 
 Table 3 shows reliability characteris- 
 tics of GCM's based on field data and 
 maintenance records. Both show the reli- 
 ability fall below the 95-pct level with- 
 in 2 months of continuous service (fig. 
 7). The lower limit of the confidence in- 
 terval for field is of the same magnitude 
 of failure rate established from the 
 maintenance records. 
 
 Trailing cables feeding shuttle cars 
 and continuous miners in underground coal 
 mines regularly are subjected to harsh 
 treatment. The failure frequency and the 
 associated splicing rates are very high 
 compared with the failure rate of compo- 
 nents such as circuit breakers and UVR's. 
 Table 4 shows splicing rates on trailing 
 cables feeding shuttle cars and continu- 
 ous miners, as estimated from field data. 
 
100 
 
 90- 
 
 80- 
 
 701- 
 
 . 60 h 
 x 
 
 _ 
 
 z 
 
 >- 501— 
 
 40- 
 
 30- 
 
 20- 
 
 10- 
 
 ■ , 1 1 , 1 , 1 
 
 \\y KEY 
 
 _\ * \ £ EM upper limit - 
 
 \ x \ 
 
 \ ° v » o EM failure rate 
 
 \ \ \ • EM lower limit 
 
 \ i \ 
 
 - V \ ^ 
 
 A V N 
 
 \ ° S- 
 
 \ N • 
 
 \ « N 
 
 \ ' \ 
 
 — \ » • 
 
 \ v \ 
 
 \ Q . N 
 
 V \ •- 
 
 \ v \ 
 
 \ v * ^ 
 
 \ \ - 
 
 \ ^ ^ 
 
 \ *•» 
 
 \ °v *V 
 
 \ \ N> 
 
 \ \ 
 
 \ N 
 
 \ °N 
 
 
 \ V 
 
 \ X 
 
 \ X 
 
 
 
 
 
 \ "° % x 
 
 
 7\. ** 
 
 
 N. O v _ 
 
 
 
 
 X T 
 
 ^v 
 
 ^■^^ 
 
 f , 1 
 
 4 6 
 
 TIME,yr 
 
 10 
 
 KEY 
 & Field upper limit 
 O Field failure rate 
 • Field lower limit 
 a Records failure rate 
 
 gj n— ■&. 
 
 FIGURE 6. • Reliability characteristics of electro- 
 mechanical undervoltage relays using field data. 
 
 2 3 
 
 TIME,yr 
 
 FIGURE 7. - Reliabil ity characteristics of ground check 
 monitors using field data and maintenance records. 
 
 TABLE 3. - Ground check monitor (GCM) data 
 
 
 Theo- 
 
 Lab 
 
 
 Mainte- 
 
 
 reti- 
 
 test 
 
 Field 
 
 nance 
 
 
 cal 
 
 
 
 records 
 
 Failure rate ( A) 
 
 
 
 
 
 failures per year. . 
 
 ND 
 
 ND 
 
 11.6 
 
 2 0.75 
 
 95-pct confidence interval of A: 
 
 
 
 
 
 
 ND 
 
 ND 
 
 3 0.93 
 
 ND 
 
 
 ND 
 ND 
 
 ND 
 ND 
 
 3 2.72 
 3 0.625 
 
 ND 
 
 "MTBF (1/ A) yr. . 
 
 3 1.33 
 
 ND No data. 
 
 -Hill and Collins (7, p. 16). 
 
 z Hill and Collins (_7, p. 17). 
 
 'Calculated. 
 
 u Mean time between failures. 
 
 Each cable failure resulting in the need insulation failure with a safety hazard 
 for a splice was considered to be a cable involved. 
 
A number of field data samples were 
 isolated from the above data, in terms of 
 the time interval in shifts, starting 
 from a new or rebuilt cable until it was 
 replaced. Table 5 shows the replacement 
 rate for shuttle cars and continuous 
 miners. Comparing the mean time between 
 replacement of 96.6 shifts for trailing 
 cable on a shuttle car with the mean time 
 between splice of 3.01 shifts reveals 
 that over 30 splices are commonly made 
 before a cable is replaced in the partic- 
 ular mine under study. 
 
 A large number of ac and dc motors are 
 used on underground machines. The motors 
 suffer frequent failures (electrical or 
 mechanical) with resultant safety hazards 
 and downtime. The failure rate in table 
 6 represents all motors used in the mine, 
 irrespective of size, location, and ap- 
 plication. Different application of mo- 
 tors in the mine are expected to generate 
 different failure rates. 
 
 Failure rates of control circuitry used 
 for continuous miners and shuttle cars 
 are shown in table 7. The control cir- 
 cuitry includes the various contactors, 
 motor starters, switches, and the general 
 wiring on individual machines. Consider- 
 ing the complexity of the control cir- 
 cuitry and available data, it was not 
 possible to estimate individual failure 
 rates of the various components of the 
 control circuit. 
 
 TABLE 4. - Splicing rates of trailing 
 cables on mine machines 
 
 TABLE 5. - Replacement rates of trailing 
 cables on mine machines 
 
 Splice rate ( A) 
 
 splice per shift. . 
 95-pct confidence in- 
 terval of A: 
 
 Lower 
 
 Upper 
 
 4 MTBS (1/ A).... shifts.. 
 
 Shuttle 
 car 
 
 !0.332 
 
 L 0.292 
 3 3.01 
 
 Contin- 
 uous 
 miner 
 
 2 0.072 
 
 3 0.0504 
 
 3 0.1022 
 
 3 13.9 
 
 1 Venkata (_5, p. 102). 
 
 2 Venkata (_5, p. 106). 
 
 Calculated. 
 
 ^Mean time between splices 
 
 Replacement rate ( A) 
 
 replacements per shift. . 
 95-pct confidence in- 
 terval of A: 
 
 Lower 
 
 Upper 
 
 *MTBR (1/A) shifts.. 
 
 Shuttle 
 
 Contin- 
 
 car 
 
 uous 
 
 
 miner 
 
 0.01035 
 
 0.00714 
 
 0.005 
 
 0.0023 
 
 0.0176 
 
 0.0146 
 
 96.6 
 
 140.1 
 
 Mean time between replacement. 
 
 Source: Venkata (5, p. 107). 
 
 TABLE 6. - Motor data 
 
 1 Failure rate ( A) 
 
 failures per shift. . 
 2 95-pct confidence in- 
 terval of A: 
 
 Lower 
 
 Upper 
 
 3 MTBF (1/A) shifts.. 
 
 ac 
 motor 
 
 0.056 
 
 0.00381 
 
 0.02436 
 
 2 17.86 
 
 dc 
 motor 
 
 0.0379 
 
 0.0234 
 
 0.06064 
 
 2 26.39 
 
 1 Venkata (_5, p. HO). 
 
 2 Calculated. 
 
 3 Mean time between failures. 
 
 TABLE 7. - Control circuitry data 
 
 
 Shuttle 
 
 Contin- 
 
 
 car 
 
 uous 
 miner 
 
 Failure rate ( A) 
 
 
 
 failures per shift. . 
 
 1 0.022 
 
 2 0.32 
 
 3 95-pct confidence in- 
 
 
 
 terval of A: 
 
 
 
 
 0.01815 
 0.0561 
 
 0.0176 
 
 
 0.0545 
 
 l+ MTBF (1/A) shifts.. 
 
 3 45.5 
 
 3 31.3 
 
 Venkata (_5, p. 110). 
 
 2 Venkata (_5, p. 111). 
 
 Calculated. 
 
 4 Mean time between failures. 
 
 No failure data are available for such 
 equipment as transformers, cable cou- 
 plers, contactors, and other items us- 
 ually found in the mine power system. 
 
However, some failure rate data are 
 available from IEEE (_9, pp. 17-41) and 
 UKAEA ( 10 , p. 72) , and suitable weighting 
 or multiplication factors could be appl- 
 ied to certain mine operating and envi- 
 ronmental conditions (_5, pp. 111-114). 
 Table 8 lists the failure rates, convert- 
 ed to a common unit — failures per unit- 
 year — for many electrical components and 
 gives comparative data from IEEE and 
 UKAEA, where such data exist. Although 
 
 several components can be compared using 
 these data, no strong conclusion can be 
 made concerning any general multiplica- 
 tion factors that would be applicable to 
 the mine environment, largely because of 
 the unavailability of comparable data for 
 many of the components. The conversion 
 processes for presenting research data in 
 failures per unit-year are shown in ap- 
 pendix B. 
 
 APPLICATION OF RELIABILITY DATA 
 
 Preventive maintenance traditionally 
 means performing various tasks, on a pe- 
 riodic basis, to extend the life of a 
 component. An example is the periodic 
 lubrication of motor bearings. However, 
 a more general definition is the process 
 of prolonging the useful life for a com- 
 ponent, and when failure is imminent, re- 
 placing the component with minimum down 
 time and personal hazard. 
 
 Problems and concerns of implementing a 
 preventive maintenance program could be 
 discussed at great length. Preventive 
 maintenance is the key to improved pro- 
 ductivity and safety in mines (2_, p. 
 488). While the mining systems pose 
 
 unique problems, these difficulties are 
 no more severe than those of other 
 industries. 
 
 Research (_5, pp. 10-42) determines 
 optimal maintenance schedules for certain 
 mine power equipment. The information 
 used to obtain the results is based upon 
 failure rate from field data and es- 
 timation from other sources, discussions 
 with mine operators, and engineering 
 judgment. The preventive maintenance 
 intervals for various mining electrical 
 equipment are shown in table 9. These 
 intervals seek to maintain a 95-pct level 
 of reliability. 
 
 TABLE 8. - Failure rate data source comparison 
 
 Failure rate, failures 
 per unit-year 
 
 Research 
 
 UKAEA 
 
 IEEE 
 
 Molded-case circuit breaker. . 
 
 Undervoltage relay (UVR) 
 
 Ground check monitor (GCM)... 
 
 Ground fault relay 
 
 Cable splicing: 
 
 Shuttle car 
 
 Continuous miner 
 
 Cable replacement: 
 
 Shuttle car 
 
 Continuous miner 
 
 ac motors 
 
 dc motors 
 
 Machine control circuitry: 
 
 Shuttle car 
 
 Continuous miner 
 
 ND No data. 
 
 0.0086 
 
 .0093 
 
 .1143 
 
 None 
 
 35.174 
 12.708 
 
 1.096 
 1.513 
 1.059 
 .5918 
 
 5.825 
 5.648 
 
 0.0175 
 
 .0438 
 
 ND 
 
 ND 
 
 ND 
 ND 
 
 ND 
 
 ND 
 
 .0876 
 
 .0876 
 
 ND 
 ND 
 
 0.0052 
 ND 
 ND 
 ND 
 
 ND 
 
 ND 
 
 ND 
 
 ND 
 
 .0109 
 
 .0556 
 
 ND 
 
 ND 
 
10 
 
 Research ( 11 , pp. 106-156) developed 
 reliability models for the electric power 
 systems of surface and underground mines 
 capable of predicting the probability of 
 electrical shock hazards existing on dis- 
 tribution systems in the mines. Relia- 
 bility data are used to predict hazard 
 rates through fault-tree analysis. 
 
 Predictions of mine electrical accident 
 probabilities using reliability modeling 
 techniques have produced several impor- 
 tant results. Table 10 summarizes the 
 results for the various mining operations 
 analyzed. The most sensitive components 
 are defined as the components which, if 
 reliability is increased, will provide 
 better protection against electrical 
 shock hazards. 
 
 TABLE 9. - Maintenance intervals 1 for 
 mining electrical equipment, months 
 
 Cables: 
 
 Shuttle car , 2.5 
 
 Continuous miner ^ c 
 
 Roof bolter 5^0 
 
 Couplers: Low-voltage 36.0 
 
 Molded-case circuit breaker 7.0 
 
 Motors: 
 
 ac motors 12.5 
 
 dc motors 21.5 
 
 iThe following equipment is subject to 
 emergency repairs only: 
 
 Ground check monitors 
 
 Ground fault relays 
 
 High-voltage couplers 
 
 Power center transformers 
 
 CONCLUSIONS 
 
 The mine electrical reliability re- 
 search program evaluated molded-case cir- 
 cuit breakers, undervoltage relays, 
 ground-check monitors, cables, motors, 
 and control circuitry. Failure rates 
 were estimated from theoretical, labora- 
 tory, and field studies. Results showed 
 that to maintain 95-pct reliability, many 
 
 components would have to be removed from 
 continuous service within 2 months to 1.5 
 yr. Failure rate data were used to cal- 
 culate optimal maintenance scheduling on 
 various mining equipment. For example, 
 the optimal maintenance interval is 7 
 months for circuit breakers and 12.5 
 months for ac motors; whereas ground- 
 
 TABLE 10. - Probability of hazardous voltages of mining operations 
 
 Operation 
 
 Probability of 
 hazardous voltage, 
 hazards per year 
 
 Most sensitive 
 component 
 
 Underground coal mine: 
 
 0.0047 
 .1871 
 
 .01403 
 
 .005946 
 
 .00003 
 
 .00153 
 
 .00409 
 .00003 
 
 .02897 
 
 Ground diode. 
 
 Underground M/NM mine: 
 480-V system 
 
 Safety ground system. 
 
 Open or short grounding resistor. 
 Cable insulation and leakage detector. 
 Open ground wire. 
 
 Ground check monitor and grounding 
 
 resistor. 
 Ground wire. 
 
 300-V system 
 
 120-V system 
 
 Dredging: 
 
 2, 300-V system 
 
 
 120-V system 
 
 Open ground wire. 
 
 Grounding resistor and overhead ground 
 wire. 
 
 Open pit M/NM mine: 
 
 Distribution system.... 
 
 M/NM Metal-nomental. 
 
 Source: Hill and Collins (11, pp. 123-147). 
 
11 
 
 fault relays require emergency repair 
 only. Prediction of mine electrical ac- 
 cident probabilities using reliability 
 modeling techniques has produced several 
 important results to improve mine dis- 
 tribution systems for various mining op- 
 erations. The most sensitive components 
 to probability of hazardous voltage for 
 underground coal mine operations are the 
 ground diodes for shuttle cars and the 
 safety ground systems for continuous 
 miners. 
 
 The need for complete, accurate, and 
 appropriate failure rates for mine compo- 
 nents is of the utmost importance. Sev- 
 eral of the failure rates used for appli- 
 cation purposes are extrapolations of 
 failure rates for similar devices from 
 other sources. The failure rates for the 
 same devices in a less hostile environ- 
 ment than a mine are also suspect. These 
 facts point out the importance of com- 
 plete data from field tests and mainte- 
 nance records. Complete data on failure 
 rates of mine electrical equipment should 
 include components installed in an oper- 
 ating coal mine power system. These com- 
 ponents include ground-check monitors, 
 ground-fault relays, molded-case circuit 
 breakers, electromechanical and solid 
 state undervoltage relays, ac and dc 
 motors, control circuitry, cables, cable 
 couplers, and transformers. 
 
 A suggested field data collection tech- 
 nique could include the following: 
 
 1. Select several underground coal 
 mine sites to represent the worst mine 
 environment. Experience from previous 
 in-mine tests shows that working sections 
 are phased out periodically; therefore, 
 several mines would be needed to provide, 
 on a continuous basis, an adequate compo- 
 nent sample size for failure rate data 
 analysis. 
 
 2. All electrical components identi- 
 fied for evaluation would be tagged with 
 unique identification stickers. 
 
 3. Maintenance records would be main- 
 tained for each tagged component. 
 
 4. Postcards would be provided to the 
 mine personnel for notification of in- 
 stallation and failure dates for each 
 component. 
 
 5. As tagged components fail they 
 would be removed from service by the mine 
 personnel and returned for laboratory 
 analysis to determine failure modes and 
 possible failure mechanism. For larger 
 components, e.g., cables, motors, and 
 transformers, mine personnel would pro- 
 vide their best judgment on the cause of 
 component failure. 
 
 Failure data collection and failure 
 rate estimation has only begun to help 
 achieve mine power system reliability. 
 After complete data on failure rates are 
 assembled, the overall system reliabil- 
 ity, availability, maintainability, and 
 safety (RAMS) models could h~ developed 
 and quantitative results can be obtain- 
 ed. Suitable component replacement, 
 preventive maintenance, and optimum 
 maintenance-crew deployments can be 
 achieved to maximize both safety and pro- 
 ductivity of the mine. 
 
 Long-range findings of the RAMS studies 
 of existing systems and equipment can be 
 used in the design of future systems and 
 equipment, particularly because large- 
 scale automated or semi automated systems 
 of mining are evolving for use in the 
 near future. Therefore, a high degree of 
 RAMS is an essential attribute of these 
 systems to justify the cost and complex- 
 ity of such automation. The system must 
 always be designed to be fail-safe, yet 
 cost-effective. These aspects cannot be 
 overemphasized, as already proven by re- 
 cent advances in space and other sophis- 
 ticated technologies. Coal mining, 
 though it looks like a mundane applicat- 
 ion, is an equally challenging task. 
 
 REFERENCES 
 
 1. Syska and Hennessy, Inc. Engineer- 
 ing Management Series. V. 6. Applied 
 Preventive Maintenance. Eng. Manage. 
 Div. , New York, 1981, 49 pp. 
 
 2. Morely, L. A. Mine Power Systems. 
 Volume II (contract J0155009, PA St. 
 Univ.). BuMines OFR 178(2)-82, 1980, 532 
 pp.; NTIS PB 83-120386. 
 
12 
 
 3. Heising, C. R. Quantitative Rela- 
 tionships Between Scheduled Electrical 
 Preventive Maintenance and Failure Rate 
 of Electrical Equipment. IEEE Trans. 
 Ind. Appl. , May/June 1982, 220 pp. 
 
 4. Reliability Analysis Center (RADC 
 Griff is Air Force Base, New York). Reli- 
 ability Design Handbook. Cat. No. RDH- 
 376, Mar. 1976, 342 pp. 
 
 5. Venkata, S. S., M. Chinnarao, E. W. 
 Collins, and E. U. Ibok. Transients Pro- 
 tection, Reliability Investigation, and 
 Safety Testing of Mine Electrical Power 
 Systems. Volume II. Reliability of Mine 
 Electrical Power Systems (grant G0144137, 
 WV Univ.). BuMines OFR 52-78, 1979, 197 
 pp.; NTIS PB 81-166779. 
 
 6. Stanek, E. K. Digital Computation 
 of Transients and Safety Testing of Mine 
 Electrical Power Systems (grant G0144137, 
 WV Univ.). BuMines OFR 52-78, 1977, 200 
 pp. NTIS PB 283400/AS. 
 
 7. Hill, H. W. , and E. W. Collins. 
 Mine Power System Safety and Reliability 
 
 Improvement. Volume II (contract 
 J0199119, WV Univ.). BuMines OFR 183-84. 
 1983. 19 pp.; NTIS PB 85-109288. 
 
 8. Collins, E. W. Reliability Analy- 
 sis, Maintenance Scheduling of Solid 
 State Undervoltage Relay. M.S. Thesis, 
 WV Univ., Morgantown, WV, 1980, 123 pp. 
 
 9. Institute of Electrical and Elec- 
 tronics Engineers. IEEE Recommended 
 Practice for Design of Reliable Indus- 
 trial and Commercial Power Systems. IEEE 
 Standard 493-1980. Wiley, 1980, 224 pp. 
 
 10. Green, A. E. , and A. J. Brourne. 
 Reliability Technology. Wiley, 1972, 233 
 pp. 
 
 11. Hill, H. W. , and 
 Mine Power System Safety 
 Improvement. Volume 
 
 E. W. Collins. 
 
 and Reliability 
 
 I (contract 
 
 J0199119, WV Univ.). BuMines OFR 46-85, 
 1981, 186 pp. 
 
 12. Stanek, E. K. Enhancement of Mine 
 Power System Safety and Reliability 
 (grant G0188097, WV Univ.). BuMines OFR 
 116-80, 1979 207 pp.; NTIS PB 81-125361. 
 
13 
 
 APPENDIX A.— FIELD DATA COLLECTION TECHNIQUE AND COMPONENTS* SAMPLE SIZE 
 
 A. Data collection technique for the 
 molded-case circuit breakers with UVR's, 
 GFR's, and GCM's: 
 
 One underground coal mine agreed to 
 cooperate with a Bureau contractor 
 in providing an operation mine site 
 for field testing. The components 
 identified were tagged with unique 
 identification stickers. Postcards 
 were provided to the mine personnel 
 for notification to contractor of 
 installation and failure dates for 
 each component. As tagged compo- 
 nents failed, they were removed 
 from service by the mine personnel. 
 No GFR units failed. Components 
 involved in field testing were 
 phased into operational service on 
 a new section power center from May 
 1981 to September 1983. 
 
 B. Collection technique for trailing 
 cables, motors, and control circuitry: 
 
 One underground coal mine agreed to 
 provide the Bureau contractor with 
 maintenance foremen's reports for a 
 
 1-yr period. The data available 
 over the year contained the dates 
 and shifts during which trailing 
 cables, motors, or control cir- 
 cuitry were repaired or replaced. 
 C. Table A-l lists components' sample 
 sizes. 
 
 TABLE A-l. - Components' sample size 
 
 Molded-case circuit breaker 14 
 
 Undervoltage relay (UVR) 14 
 
 Ground check monitor GCM) 14 
 
 Ground fault relay (GFR) 14 
 
 Cable splicing: 
 
 Shuttle car 10 
 
 Continuous miner 6 
 
 Cable replacement: 
 
 Shuttle car 10 
 
 Continuous miner 5 
 
 ac motors 56 
 
 dc motors 68 
 
 Machine control circuitry: 
 
 Shuttle car 6 
 
 Continuous miner 4 
 
14 
 
 APPENDIX B.— FIELD DATA CONVERSION TO FAILURES PER UNIT-YEAR 
 
 To obtain failures 
 determine the numbe 
 year, then divide by 
 For some components, 
 year proceeded duri 
 the total shifts (1 
 in this case, it is 
 the number of shifts 
 number of shifts pe 
 fraction of a year o 
 to divide this numbe 
 failures to obtain f 
 
 per unit-year, first 
 rs of failures per 
 
 the number of units. 
 
 operation during the 
 ng only a portion of 
 ,059) during a year; 
 
 necessary to divide 
 
 in operation by the 
 r year to obtain the 
 f operation, and then 
 r into the number of 
 ailures per year. 
 
 Molded-case circuit breakers: 
 
 Failures per year.. 0.12 
 
 Units v 14 
 
 Failures per unit-year 0.0086 
 
 Undervoltage relays (UVR's): 
 
 Failures per year 0.13 
 
 Units t 14 
 
 Failures per unit-year 0.0093 
 
 Ground fault monitors (GFM's): 
 
 Failures per year 1.6 
 
 Units t 14 
 
 Failures per unit-year 0.1143 
 
 Cable splicing — Shuttle cars: 
 
 Failures 93 
 
 Fraction of year in operation 
 = (280 shifts v 1,059 shifts 
 
 per year) v 0.2644 
 
 Failures per year 351.74 
 
 Units t 10 
 
 Failures per unit-year.. 35.174 
 
 Cable splicing — Continuous miners: 
 
 Failures 36 
 
 Fraction of year in operation 
 = (500 shifts v 1,059 shifts 
 
 per year) ^ 0.4721 
 
 Failures per year 76.255 
 
 Units v 6 
 
 Failures per unit-year.. 12.708 
 
 Cable replacement — Shuttle cars: 
 
 Failures 10 
 
 Fraction of year in operation 
 = (966 shifts f 1,059 shifts 
 
 per year) -s- 0.9122 
 
 Failures per year 10.963 
 
 Units v 10 
 
 Failures per unit-year.... 1.096 
 
 *U.S. CPO: 1985-505-019/20,110 
 
 Cable replacement — Continuous miners: 
 
 Failures 5 
 
 Fraction of year in operation 
 = (700 shifts x 1,059 shifts 
 
 per year) i 0.6610 
 
 Failures per year 7.564 
 
 Units v 5 
 
 Failures per unit-year.. 1.513 
 
 ac motors: 
 
 Failures 28 
 
 Fraction of year in operation 
 = (500 shifts v 1,059 shifts 
 
 per year) -5- 0.4721 
 
 Failures per year 59.309 
 
 Units 7 56 
 
 Failures per unit-year.. 1.059 
 
 dc motors: 
 
 Failures 19 
 
 Fraction of year in operation 
 = (500 shifts t 1,059 shifts 
 
 per year) v 0.4721 
 
 Failures per year 40.246 
 
 Units v 68 
 
 Failures per unit-year.. 0.5918 
 
 Control circuitry — Shuttle cars: 
 
 Failures 16 
 
 Fraction of year in operation 
 = (500 shifts t 1,059 shifts 
 
 per year) v 0.4721 
 
 Failures per year 33.891 
 
 Uni t s -i- 6 
 
 Failures per unit-year.. 5.648 
 
 Control circuitry — Continuous miners: 
 
 Failures 11 
 
 Fraction of year in operation 
 = (500 shifts t 1,059 shifts 
 
 per year) -r 0.4721 
 
 Failures per year 23.300 
 
 Units -5- 4 
 
 Failures per unit-year.. 5.825 
 
 INT.-BU.O F MINES, PGH..P A. 28112 
 

 U.S. Department of the Interior 
 Bureau of Mines— Prod, and Distr. 
 Cochrans Mill Road 
 P.O. Box 18070 
 Pittsburgh, Pa. 15236 
 
 OFFICIAL BUSINESS 
 PENALTY FOR PRIVATE USE. S300 
 
 ] Do not wi sh to recei ve thi s 
 material, please remove 
 from your mailing list* 
 
 ] Address change. Please 
 correct as indicated* 
 
 AN EQUAL OPPORTUNITY EMPLOYER 
 
V v > 
 
 ^ ^ sate' \ * s**** m * * 
 
 *o 
 
 ^f 
 
 
 
 o' ; 
 
 ■c? 
 
 
 "of 
 
 
 
 .'^ 
 
 ^ 
 
 " N ° * <^ ^ . t * . , ^ Q A % c o « o , ^ 
 
< SSk' /^dkS y k .-^i-%. J'ZBkS j*.-$ik.\ o»*.i^-> 
 
 . 
 
 v °* 4? •!' V .*1^L> Oft A 0' •> V V • • 
 
 'lyM. -SK'* ^"V °*ww** ** v "\ '•«!