ML20100F881
| ML20100F881 | |
| Person / Time | |
|---|---|
| Site: | North Anna  |
| Issue date: | 03/29/1984 |
| From: | Dalton K, Kresser J, Savage J LAWRENCE LIVERMORE NATIONAL LABORATORY |
| To: | NRC |
| Shared Package | |
| ML20100F862 | List: |
| References | |
| CON-FIN-A-0442, CON-FIN-A-442 UCID-20053, NUDOCS 8412070065 | |
| Download: ML20100F881 (127) | |
Text
l s,
UCID. 20053 S
TECHNICAL EVALUATION REPORT ON THE SEVEN MAIN TRANSFORMER FAILURES AT THE NORTH ANNA F0WER STATION, UNITS 1 AND 2 (Docket Nos. 50-33,8, 50-339)
Kerry J. Dalton, Jean V. Kresser, Jack W. Savage, James C. Selan March 29, 1984 6
- g.g
,,,x
,gg_ m-- m.7
-s s. n
--=m m.z K
gy b,hl 4$h
\\
-s This is an informal report intended primarily for internal oe limited external distribution.
[.4 The opinions and conclusions stated are those of the author and uay or may not be those of the Laboratory.
s-This work was supposted by the United States Nuclear Regulatory Commission under a Memorandum of Understanding with the United States Department of Energy.
j l
NRC FIN No. A-0442 s
8412070065 841127 PDR ADOCK 05000338 S
4
[
l ABSTRACT This report documents technical evaluations on various aspects pertaining to the seven main transformer failures at the North Anna Power Station, Units 1 and 2.
These reports cover the subjects of Probability Risk Assessment (PRA), Failure Modes and Effects Analysis (FMEA), Root Causes, Protection Systems, Modifications, Failure Statistics, and Generic Aspects.
The PRA determined that the contribution from a main transform r failure affecting plant safety systems so as to increase the risk to the public health and safety is negligible. The FMEA determined that a main
=
transformer failure can have primary and secondary effects on plant safety system operation. The evaluation of the " Root Causes" found that no single common cause contributed to the seven failures. Each failure was found to have specific circumstances for initiating the f ailure.
Both the generator and transformer primary protection systems were found to perform correctly and vere designed within industry standards and practices. The proposed modifications resulting from the analyses of the f ailures will improve system reliability and integrity, and will reduce potentially damaging effects. The failure statistic survey found very limited data bases from which a meaningful correlation could be ascertained. The statistical compar-ison found no appreciable anomalies with the NAPS failures. The evaluation of all the available information and the results of the separate reports on the main transformer failures found that several " generic" concerns exist.
.T 4 - - - - -
i TABLE OF CONTENTS Page ABSTRACT i
TABLE OF CONTENTS iii POREWORD xi BACKGROUND.
. xiii TASK REPORT 1 -
REPORT ON THE LICENSING DESIGN BASIS, PROBABILISTIC RISK ASSESSMENT, AND FAIIERE MODE AND EFFECTS ANALYSIS ASSOCIATED WITH THE MAIN TRANSFORMER FAILURES AT THE NORTH ANNA POWER STATION, UNITS 1 AND 2: Jack W. Savage and James C. Selan 1
1.
INTRODUCTION 1-1 2.
SYSTEM DESCRIPTION.
1-2 3.
TRANSFORMER FAILURE CHRONOLOGY.
1-2 l
4.
LICENSING DESIGN BASIS.
1-5 5.
FAILURE MODE AND EFFECTS ANALYSIS 1-7 5.1 PRIMARY EFFECTS.
1-7 5.2 SECONDARY EFFECTS 1-8 6.
PROBABILITY RISK ASSESSMENT.
1-13 6.1 COMPARISON OF SIMPLIFIED EMERGENCY POWER j
SYSTEM DIAGRAMS.
1-13 6.2 HOW NIFS MIGHT CONTRIBUTE TO WASH 1400 FAULT l
TREE EVENTS.
1-13
{
6.2.1 LOSS-OF-NET.
1-16 6.2.2 INSUFFICIENT POWER TO EMERGENCY BUSES 1-20 l
6.3 WHERE MTFS APPEAR IN THE FAULT TREES.
1-22 I
6.4 ESTIMATES OF NAPS MTF CONTRIBUTIONS TO FAULT TREE QUANTIFICATION.
1-26 1
7.
CONCLUSION.
1-27 1
1 REFERENCES 1-28 APPENDIX 1 1-29
)
i iii
TASK REPORT 2 -
EVALUATION OF THE " ROOT CAUSES" FOR THE MAIN TRANSPORMER FAILURES AT THE NORTH ANNA POWER STATION, UNITS 1 AND 2:
Kerry J. Dalton and James C. Selan 2
1.
INTRODUCTION 2-1 2.
FAILURE REVIEW DOCUMENTATION 2-2 3.
SYSTEM CONNECTION 2-2 4
TESTING AND MAINTENANCE OVERVIEW 2-4 5.
" ROOT CAUSE" EVALUATION.
2-6 5.1 TRANSFORMER FAILURES.
2-6 5.2 VEPCO'S TASK FORCE INVESTIGATION.
2-10 6.
SUMMARY
2-10 7.
CONCLUSION.
2-11 REFERENCES 2-12 TASK REPORT 3 -
VIRGINIA ELECTRIC AND POWER COMPANY NORTH ANNA POWER STATION, UNITS 1 AND 2, MAIN TRANSFORMER AND GENERATOR PROTECTIVE SYSTEMS:
Jean V. Kresser 3
1.
INTRODUCTION 3-1 2.
GENERAL DISCUSSION 3-2 3.
NORTH ANNA UNITS PROTECTIVE SYSTEM.
3-3
3.1 DESCRIPTION
OF THE RELAYING SCHEMES.
3-3 3.1.1 GENERATOR PRIMARY PROTECTION.
3-3 3.1.2 GENERATOR BACKUP PROTECTION.
3-5 3.1.3 GENERATOR LEADS 3-5 3.1.4 ISOLATED PHASE BUS DUCT BACKUP GROUND 3-6 3.1.5 GENERATOR BREAKER FAILURE (G12).
3-6 3.1.6 MAIN TRANSFORMER 3-6 l
4 TRANSFORMER FAILURES 3-7 5.
CONCLUSIONS 3-7 APPENDIX A 3-9 REFERENCE DRAWINGS 3-10 REFERENCES 3-11 SUPPLEMENTAL REPORT (REPORT BASED ON ADDITIONAL NRC STAFF QUESTIONS).
3-13 v
TASK REPORT 4 -
EVALUATION OF THE LICENSEE'S PROPOSED MODIFICATIONS: Kerry J. Dalton and James C. Selan 4
1.
INTRODUCTION 4-1 2.
EVALUATION OF THE PROPOSED MODIFICATIONS 4-1 2.1 TRANSFORMER REPLACEMENT.
4-1 2.2 SURGE ARRESTERS.
4-2 2.3 GENERATOR GROUNDING.
4-2 2.4 ISOLATED PHASE BUS SEPARATION 4-3
~
2.5 INSTALLATION OF MONITORING RECORDERS 4-3 2.6 MAIN GENERATOR BREAKER 4-3 2.7 REFUELING OUIAGE INSPECTIONS 4-4 2.8 FIRE PROTECTION MODIFICATIONS 4-4 2.9 PROCEDURAL CHANGES 4-5 3.
CONCLUSION 4-5 4-6 REFERENCES TASK REPORT 5 -
REPORT ON SURVEY CONDUCTED OF ELECTRIC POWER INDUSTRY'S EXPERIENCES WITH EXTRA HIGH VOLTAGE TRANSFORMER FAILURES AND THE COMPARISON WITH THE MAIN TRANSFORMER FAILURES AT THE NORTH ANNA POWER STATION, UNITS 1 AND 2: James C. Selan 5
1.
INTRODUCTION 5-1 2.
SURVEY.
5-1 2.1 SURVEY FIELD.
5-1 2.2 SURVEY RESULTS 5-3 2.2.1 EDISON ELECTRIC INSTITUTE FAILURE STATISTICS.
5-3 2.2.2 DOBLE ENGINEERING COMPANY FAILURE STATISTICS.
5-8 2.2.3 IEEE NUCLEAR DATA RELIABILITY MANUAL.
5-9 3.
SURVEY COMPARISONS.
5-12 3.1 NAPS MAIN TRANSFORMER FAILURE RATES / MODES 5-12 3.2 NAPS MAIN TRANSFORMER FAILURES VERSUS EEI DATA 5-12 3.3 NAPS MAIN TRANSFORMER FAILURES VERSUS DOBLE ENGINEERING DATA.
5-13 3.4 NAPS MAIN TRANSFORMER FAILURES VERSUS IEEE RELIABILITY DATA.
5-14 4
SUMMARY
5-15 5.
CONCLUSION.
5-16 REFERENCES 5-16 v11
- -- - - ~.
.~--e
TASK REPORT 6 -
EVALUATION OF GENERIC ASPECTS OF THE MAIN TRANSFORMER FAILURES AT THE NORTH ANNA POWER STATION, UNITS 1 AND 2: James C. Selan.
6 1.
INTRODUCTION 6-1 2.
PRIMARY ASPECTS.
6-1 2.1 TRANSFORMER FIRES 6-1 2.1.1 FIRE PROTECIION.
6-2 2.1.2 OVERHEAD CONDUCTORS / BUSES 6-4 2.1.3 CABLE TRAYS.
6-5 2.1.4 STORAGE OF SPARE EQUIPMENT 6-5 2.1.5 STORAGE OF FLAMMABLE MATERIAL NEAR POTENTIAL o
FIRE HAZARDS.
6-5 2.1.6 OIL FILLED TRANSFORMERS IN GENERAL 6-5 2.2 TRANSFORMER MAINTENANCE AND OPERATIONAL PRODEDURES 6-6 2.3 EXCESSIVE SHIPPING AND HANDLING.
6-6 3.
SECONDARY ASPECTS 6-7 3.1 CASCADING EFFECTS 6-7 3.2 EXTENSIVE ELECTRICAL / MECHANICAL DAMAGE 6-7 3.3 MISSILES / EXPLOSIONS.
6-7 4
CONCLUSIONS 6-8 REFERENCES 6-9 OVERALL
SUMMARY
OF THE TASK REPORTS ' CONCLUSIONS 7
7
SUMMARY
7-1 OVERALL CONLCUSION OF THE EVALUATION OF THE SEVEN MAIN TRANSFORMER FAILURES AT THE NORTH ANNA POWER STATION, UNITS 1 AND 2 8
8.
CONCLUSION.
8-1
)
I e
ix
FOREWORD This report is supplied for the U. S. Nuclear Regulatory Commission, Office of Nuclear Reactor Regulation, Division of Licensing, by Lawrence Livermore National Laboratory.
i The U.S. Nuclear Regulatory Commission funded the work under the authorization entitled " Evaluation of Main Transformer Failures at North Anna Power Station," B&R 20 19 10 12 1, FIN A-0442.
W 1
4 i
m 0
4 l
t xi
i BACKGROUND o
There have been seven (7) Westinghouse main transformer faults in 26 months at Virginia Electric and Power Company's (VEPCO) North Anna 4
Power Station (NAPS). Prior to the latest two faults occurriag in the North Anna Unit 1 (NA-1) main transformers on November 16 and December 5,1982, all previous transformers f aults had occurred at NA-2.
All faults to date
}
are listed below in chronological order:
(1)
NA-2: MT "A", November 29, 1980 i
(2)
NA-2: Mr "C", June 19, 1981 (3)
NA-2: MT "B", July 3,1981 4
i (4)
NA-2: Mr "C", July 25, 1981 (5)
NA-2: MT "B", August 25, 1982 (6)
NA-1: MT "C", November 16, 1982 (7)
NA-1: MT "B", December 5,1982 4
This document contains six separate Task Reports which provide a technical evaluation on specific concerns identified by the NRC staff i
as a result of the seven main tr:insformer failures at the North Anna Power i
Station, Units 1 and 2.
Specifically these Task Reports are:
i f
Task Report 1 REPORT ON THE LICENSING DESIGN BASIS, PROBABILISTIC i
RISK ASSESSMENT, AND FAILURE MODE AND EFFECTS ANALYSIS
(
ASSOCIATED WITH THE MAIN TRANSFORMER FAILURES AT THE NORTH ANNA POWZR STATION, UNITS 1 AND 2 Task Report 2 EVALUATION OF THE " ROOT CAUSES" FOR THE MAIN TRANSFORMER FAILURES AT THE NORTH ANNA POWER STATION, UNITS 1 AND 2 i
Task Report 3 VIRGINIA ELECTRIC AND POWER COMPANY NORTH ANNA POWER j
STATION, UNITS 1 AND 2, MAIN TRANSFORMER AND GENERATOR PROTECTIVE SYSTEMS Task Report 4 EVALUATION OF THE LICENSEE 'S PROPOSED MODIFICATIONS i
Task Report 5 REPORT ON THE SURVEY CONDUCTED OF ELECTRIC POWER INDUSTRY'S EXPERIENCES WITH EXTRA HIGH VOLTAGE TRANS-FORMER FAILURES AND THE COMP /LRISON WITH THE MAIN TRANSFORMER FAILURES AT THE NORTH ANNA POWER STATION,
~
UNITS 1 AND 2 l
l Task Report 6 EVALUATION OF GENERIC ASPECTS OF THE MAIN TRANSFORMER FAILURES AT THE NORTH ANNA POWER STATION, UNITS 1 AND 2 j
Each Task Report evaluates the concern and presents a conclusion j
based on the available information.
i t
xiii
..m
.,-n.
m,n
->---s
,-n~
n.--.--.v
---en--,
m-m m w,m m y m we,- -,-
ww..-
,--,----w.,-
TASK REPORT 1 REPORT ON THE LICENSING DESIGN BASIS, PROBABILISTIC RISK ASSESSMENT, AND FAILURE MODE AND EFFECTS ANALYSIS ASSpCIATED WITH THE MAIN TRANSFORMER FAILURES AT THE lt}RTH ANNA POWER STATION, UNITS 1 AND 2 s
Jack W. Savage James C. Selan S
S 4
5 1
E' ABSTRACT-i i
1 This report documents the probabilistic risk assessment on the frequency of main transformer failures at the North Anna Power Station. A
{
study.was made to determine if the frequency of main transformer failures and accompanying plant trips, resulting in increased challenges to the plant's protective systems, is a departure from the design basis and if such a departure 4'
could significantly increase the risk to public health and safety. A limited failure mode and effects analysis was also made to identify the affect of a i -
sain transformer failure on plant systems and equipment.
The increased frequency of.asin transformer failures does not signifi-cantly affect or degrade the overall reactor protection or coolant systems nor contribute an increased risk to the public health and safety.
Based on the l
i adequacy and reliability of the detection and protection system, the effects of l
a main transformer failure on plant systems and equipment can significantly be limited.
l l
FOREWORD i
3 This report is supplied for the U. S. Nuclear Regulatory Commission, Office of Nuclear reactor Regulation, Division of Licensing, by Lawrence Livermore National Laboratory, i
The U. S. Nuclear Regulatory Commission funded the work under the i
authorization entitled " Evaluation of Main Transformer Failures at North Anna
}
Power Station," B&R 20 19 10 12 1, FIN A-0442.
4 1
b I
i I
t i
l i
I i
1-1 i
I
_ _ _..-_ _,.._ _ _.. _.. -... _,... _ _.__ _,_ -,_ _,_..___..~.-_.,_,_ _ __......-- _ _._.._
TABLE OF CONTENTS M
1.
INTRODUCTION.
1-1 2
SYSTEM DESCRIPTION 1-2 3.
TRANSFORIER FAILURE CHRONOLOGY 1-2 4
LICENSING DESIGN BASIS 1-5 5.
FAILURE MODE AND EFFECTS ANALYSIS.
1-7 5.1 PRIMARY EFFECTS 1-7 5.2 SECONDARY EFFECTS.
1-8 6.
PROBABILITY RISK ASSESSMENT
. 1-13 6.1 COMPARISON OF SIMPLIFIED EMERGENCY POWER SYSTEM DIAGRAMS
. 1-13 6.2 HOW NIFS MIGHT CONTRIBUTE TO WASH 1400 FAULT TREE EVENTS
. 1-13 6.2.1 LOSS-OF-NET
. 1-16 6.2.2 INSUFFICIENT POWER TO EMERGENCY BUSES.
. 1-20 6.3 WHERE MTFS APPEAR IN THE FAULT TREES
. 1-22 6.4 ESTIMATES OF NAPS MTF CONTRIBUTIONS TO FAULT TREE QUANTIFICATION
. 1-26 7
CONCLUSION
. 1-27 REFERENCES
. 1-28 APPENDIX 1
. 1-29 6
1-111
TABLE OF ILLUSTRATIONS M
FIGURE 1 North Anna Power Station Electrical One-Line Diagram.
1-4 FMEA-1 MTF Short Circuit Fault Propagation 1-9 FMEA-2 MTF to Generator Failure.
1-12 l
FIGURE II 5-2 Simplified Emergency Power Block Diagram.
1-14 FIGURE 2 Simplified NAPS Emergency Power Supply 1-15 FIGURE II 5-4 PWR Emergency Power System Feult Tree (Sh 8).
1-17
.i
+
FIGURE II 5-4 PWR Emergency Power System Fault Tree (Sh 9).
1-18 FIGURE II 5-4 PWR Emergency Power System Fault Tree (Sh 10).
1-19 FIGURE 3 Contributions to Quantification of WASH 1400 IP Fault Trees 1-21 FIGURE II 5-1 Surry 1/2 - Single Line Diagram (Sh 1) 1-23 i
FIGURE 4 NAPS FAULT PATHS.
1-25 1
l 1
t-1-v
REPORT ON THE LICENSING DESIGN BASIS, FROBABILISTIC RISK ASSESSMENT, AND FAILURE MODE AND EFFECTS ANALYSIS ASSOCIATED WITH THE MAIN TRANSFORMER FAILURES AT THE NORTH ANNA POWER STATION, UNITS 1 AND 2 (Docket Nos. 50-338, 50-339)
Jack W. Savage James'C. Selan Lawrence Livermore National Laboratory 1.
INTRODUCTION The North Anna Power Station (NAPS) was licensed to operate by the U.S. Nuclear Regulatory Commission (NRC) in April,1978 for Unit I and in August, 1980 for Unit 2.
Both units are pressurized water reactors (PWRs) designed by Westinghouse (W). The plants are under operation by the Virginia Electric Power Company (VEPCO), the licensee.
The output of each unit's generator is connected to a 500 kV transmission system through a main step-up transformer bank consisting of three single phase main transformers (MT) rated 22 kV/500 kV.
Seven of these main transformers failed in a 26-month period from November 29, 1980 to December 5, 1982. The first five failures occurred at Unit 2; the last two failures at Unit 1.
This report documents the study nade to evaluate the frequency of asin transformer failures (NTFs) to determine if the accompanying plant trips result in increased challenges to the plant protective systess which may result in a departure from the design basis.
The study also seeks to
~
determine whether such a departure could significantly increase the risk to the public health and safety.
Included in this report is a limited failure mode and effects analysis to determine NTF effects on plant systems and equipment.
0 1-1
2.
SYSTEM Di!SCRIPTION An electrical one-line diagram of the North Anna Power Station, Units 1 and 2, is shown in Figure 1.
This figure and the following system information was taken from the NAPS updated final safety analysis report (UFSAR) [Ref. 1].
Each of the two main generators is connected to the 500 kV switchyard through three single phase (19) 22 kV/500 kV main transformers (MTs). The switchyard in turn serves two 500 kV/34.5 kV service transformers which supply power to the three 34.5 kV/4.16 kV onsite reserve station service transformers (RSSTs). The output of each main generator also serves three unit station service transformers (USSTs). These USSTs in turn supply power to each of six 4.16 kV normal station service buses (three per unit) and their associated 480-volt buses.
Each unit has two 4.16 kV emergency (Class 1E) buses.
For Unit 1, these 4.16 kV buses (1H and 1J) can receive power from either the RSSTs, the USSTs, or the emergency onsite diesel generators. At Unit 2, the 4.16 kV Class 1E buses (2H and 2J) receive power from either the RSSTs or the onsite emergency diesel generators. The bus tie breaker between 2H and 2J is racked out, and is under strict administrative control. The dotted lines shown in Figure 1 represent the future plans for supplying power from the USSTs to 2J and' 2H, which will remove the administrative control tie breakers. Each 4.16 kV Class 1E bus feeds two 480-volt Class 1E buses, which supply power to various loads and motor control centers (MCC). The 120-volt vital AC system (4 vital buses) is normally supplied from static inverters and batteries with an alter-nate supply through a regulating transformer from a 480-volt Class 1E MCC.
The normal source of power for each unit's auxiliaries is the main generator through the USSTs. The preferred source of power to the Class 1E equipment is from the switchyard through the RSSTs with the diesel generators as the standby emergency source. Unit i utilizes a main generator breaker which allows for backfeeding through the MTs and USSTs to the station auxil-iaries as an alternate source. The installation of a main generator breaker at Unit 2 is in progress. For loss of the preferred souce, automatic transfer to the diesel generators occurs with manual transfer control to the alternate sources.
f 3.
TRANSFORMER FAILURE CHRONOLOGY The main generator sta, up transformer bank used at the North Anna i
Power Station, Units 1 and 2, consists of three-19 22 kV/500 kV transformers.
Since November 29, 1980, when the first 19 MT failed, six more MTs have failed with the last occurring on December 5, 1982.
Of these seven failures, five have occurred at Unit 2, and two at Unit 1.
A simplified chronology of the MTFs [Refs. 2 through 4) is as follows:
l 1-2 1
7 i
FEBRUARY, 1971 Serial Nos. 1994, 1995, 1965* and spare 1993*
delivered to Unit 1 MARCH-APRIL, 1974 Serial Nos. 2098, 2099, 2100 delivered to Unit 2.
APRIL, 1976 2098, 2099, 2100 sent to Georgia Power (Bowen Plant) where 1526 and 1527 had failed (purchased 1968, installed 1970).
MAY 5, 1976 2099 failed at Bowen Plant - 2098, 2099, 2100 sent to Muncie E for repair / test.
APRIL, 1978 2098, 2099, 2100 sent back to Unit 2.
NOVEMBER 29, 1980 2099 failed (A9) - 1993 spare installed.
Failure of HV lead to LV coil.
JUNE 19, 1981 2100 failed (C9) - 1527 installed (rebuilt af ter failure at Bowen Plant).
JULY 3, 1981 2098 failed (89) - GE transformer from Surry 1/2 installed.
Failure of HV bushing to LV coil and tank.
JULY 25, 1981 1527 failed (C9) - 2099 installed af ter rebuild by E; GE transformer replaced with 2100 af ter E rebuild.
Failure of HV line coil to LV coil.
AUGUST 25, 1982 2100 failed (B9) replaced with a McGraw Edison trans forme r.
Failure of HV bushing, corona shield, and LV winding.
NOVEMBER 16, 1982 1995 failed (C9) - to be replaced with a McGraw Edison transformer.
Failure of HV bushing shield to LV coil.
DECEMBER 5, 1982 1994 failed (89) - to be replaced with a McGraw Edison transformer.
Failure of IIV lead, LV coil, and turn-to-turn fault.
O Serial nos.1965 and 1993 have experienced no problems 1-3
soo-34.5nv RESERVE STATION TOswlTCHYARD TO SwetCHVARD SERylCE TR AN570RNERS d
ANN 0a5 $
M V W
IEANonS
[] NC NC [ ]
54 suv 395av SWITCHYARO SwtTCMYARD
&#ES (3 NC
'NC[l No
[3NC AREA i
i sus 2G g aus iG p
F__
__ q NCb
,**'[RA s. V Y
( NC NO Y a,dA, l
M T AN 0.1
"* * *' O cower rowsR S T A fl0N j
station unit aEsERvE UraT RESERVE UNIT RESERVE AAE8 l
AREA sTASERv y Q STA SERVy V STA SERygg SI A SERy y y STA SERV y,g STA SERVy9 NC C TRANS-TRANS-1RANS.
1RANS TRANS-TRANS-
,. 3
, g 4
vonutR voRutR A voRute 70RufRB rannts ronutR C NCL2 I
22-4.86mv 34.5-4 16mv 22-4 e6mv 34.5-4.I6 mv 22-416KV 34.5-4.16 sv I-34 l-3 i
-34 l-5 l-34 1-3
[ NC 4 sov
[] NC NC 4 sov []NC NCE]
4H OV NC
,s, uAin I sussA susO I sus is nos E
- susic, ISU$ F_
u AIN GEtotR ATOR GEMEA ATOR s
1 NO. g
[
M0.2 I
NO s NO
]NC NO NO C
NO
] NC C
L___I_
~
Yusin "T"
5 IJ NO
'j NO f
)
i DIESEL y
y y
a y
]
GENERATORS NChl h}NO O
0 aus 2A I sus e aus 2c 1
8 r5 I
'.c
- NO l
e 1
}
{jN NC (,} l l NO aus 2H NC l
l
}NO CB RACKE0 FIGURE 1 l
OUT NORTH ANN A POWER STATION i
FUTURE UNITS i AND 2 i
ADMIN i
CONTROL i
i
4.
LICENSING DESIGN BASIS A synopsis of the plant operating modes and accompanying plant trips at the time of MTF is as follows [Refs. 5 to 8]:
FAILURE CONSEQUENCE November 29, 1980
. Unit 2 at 100% reactor power
~
. Turbine / reactor trips June 9, 1981
. Unit 2 at 100% reactor power
. Generator / reactor trips July 3, 1981
. Unit 2 at 17% reactor power
. Generator / turbine / reactor trips
. Safety injection for 2 minutes as a result of low-low Tav and spurious high steam flow signals July 25, 1981
. Unit 2 in shutdown for preoperational testing
. hts on backfeed from 500 kV system at time of failure August 25, 1982
. Unit 2 at 30% reactor power
. Turbine / reactor trips November 16, 1982
. Unit 1 in hot shutdown (mode 4) at time of failure December 5, 1982
. Unit 1 at 30% reactor power
. Generator / turbine / reactor trips
. Safety injection for 7 minutes as a result of 2 channels of high steam flow inoperable, combined with low-low Tav signals Of the above seven HTFs, five resulted in reactor trips (RTs) for a NAPS reactor trip rate of 2.3 per year (5 MTF/ reactor trips over 26 months).
In this period of transformer failures, Unit i experienced 13 reactor trips and 6 power-ramp downs, which resulted in 4 reactor trips, bringing the total to 17..
Unit 2 experienced 21 reactor trips and 2 reactor trips from 11 power ramp downs, which brings the total to 23 [Ref. 9]. These total RTs pro-duce a trip rate per year of 7.83 and 10.60 for Units 1 and 2, respectively.
Combining the total reactor trips produces a NAPS reactor trip rate of 18.4 per year (40 reactor trips over 26 months).
1-5
The contribution of induced reactor trips to total reactor trips due to NTFs at Units 1 and 2 is one and four respectively, which produces the ratios shown below.
- L = 0.0588 = 5.88%
Unit 2: RTgyp Unit 1: RTNTF' 4. o,1739 17,39g
=
R' TOTAL 17 RtTOTAL U
Table 5.2-3, " Summary of Reactor Coolant System Design Transients,"
in the NAPS UFSAR defines the number of occurrences for four transient condi-tion categories. Under the upset condition category (those of a moderate frequency of occurrence), reactor trips from full power have a design frequency of occurrence of 400. This converts to 10 full power reactor trips per year 4
j over a 40 year design plant life per plant.
T A cogparison of'the NTF/ reactor trip rates, total reactor trips, and 4
those NTF/ reactor trips at full power, indicates that induced reactor trips i
due to an NTF are insignificant on challenging the reactor protection and coolant system designs. This is based on the fact that only two NTFs/RTs occurred at full power and two other NTFs challenged the reactor protection and coolant systems (safety injection for 2 and 7 minutes). A note must be made that this j
conclusion is based only on the information available on the number of RTs j
during the failure period. It does not include detailed information on each trip occurrence as to reactor power level, causes, effect, or other systems 4
challenged.
1 The NAPS NTF/ reactor trip rates are also compared to two other data bases, the EPRI NP-2230 [Ref.10] and the Zion Probabilistic Safety Study
[Ref. 11]. The EPRI report presented frequency of occurrence' data on PWRs for 2093 events occurring over 213.4 plant years. Table S-1 in the report shows an occurrence rate per year of all transient categories for Westinghouse j
PWRs of 9.71.
Comparing this figure to the RTs per year from all causes (7.83 and 10.60 for Units 1 and 2, respectively) shows that NAPS alone has an RT rate
{
approximately equal to the rate of occurrence for all transients in the EPRI l
PWR data base.
l The Zion Probabilistic Safety Study data base included 30 power plants
)
with 131 years of total operating experience.
This study shows tLat for the l
l PWR generic population, the number of reactor trips from all causes identified j
was 13.17 per year. The comparison of this RT rate to the RT rates at NAPS, Unita 1 and 2 (17 and 23 RTs over 26 months, respectively), shows NAPS has a lower RT rate as shown in the ratios below.
UNIT 1 RT/YR = 7.83 = 0.595 = 59.5%
UNIT 2 RT/YR = 10.60 = 0.805 - 80.5%
l PWR POP RT/YR 13.17 PWR POP RT/YR 13.17 i
{
Based on the comparison of the available 'information, the RT rates at i
NAPS (Units 1 and 2), with or without NTF contribution, are withi4 the mean rate l
of occurrence identified in the above data bases.
2 i
l I
1-6 i
j
.m.
._... - _.~ ~.,_ _._
5.
FAILURE MODE AND EFFECTS ANALYSIS A failure mode and effects analysis (FMEA) was conducted to identify the effects of a MTF on plant systems and equipment. The FMEA was also used to support the methodology used in the probabilistic risk assessment.
NAPS has a MTF rate of 3.2 per year (7 failures over 26 months).
The effects of an MTF can be categorized into primary effects and
~
secondary effects. Each category below only identifies in general the systems and equipment which may be influenced by the MTF.
No attempt was made to analyze in detail overall effects on plant availability or safety or to assign a degree of probability or seriousness to the effects.
5.1 FRIMARY EFFECTS An MTF can be of several types and can occur in either the high-i voltage or low-voltage windings or bushings as:
(1) Opens (2) Shorts (3) Conductor to ground faults (4) Conductor to conductor faults (5) HV to LV faults (6) Single phase faults 8
(7) Multi phase faults (8) Combination of the above The effects of such faults can be manifested in the transformer or in the systems attached to the transformer.
Some examples are:
(1) Damage to the transformer windings, bushings, tank, and periph-eral equipment.
f.
j (2) External effects to the plant such as fires, oil spills, missiles (tank rupture), overhead conductor (bus bar) damage.
(3) External effects to the power system such as abnormal voltages (low or high, transients, e.g. dips, spikes) and potential thermal effects on motors or the main generator due to unbalanced voltages.
The effects will be influenced by the plant operating mode and the time of i
duration. The effects will also be influenced by the response or lack of re-sponse of the detection and protection systems which are intended to isolate 1-7 I
I
the transformer to protect it and the systems from the effects of the failure.
It is important to recognize that the longer a MTF exists on the system, the greater the possibility that there will be significant secondary effects.
These detection and protection systems are being evaluated in a separate interim report.
5.2 SECONDARY EFFECTS In general, the effects of an NTF will vary with the specific equip-
- ~
ment involved and may result in:
CAUSES VOLTAGE EFFECTS 0
Low High Dips Spikes Unbalanced Failure to operate when wanted x
x Failure during operation x
x x
x Unwanted operation x
x x
x x
Examples of some equipment which might be af fected are:
protective relays, circuit breakers, control circuit fuses, reactor protection system, containment system, main generator system, and trip systems (turbine, generator, reactor).
What follows is presented as a source from which areas for further investigation can be selected.
No attempt was made to determine their relative impo rtance.
(1) Figure FMEA-1 illustrates how an MTF will " propagate" through the NAPS emergency systems (see also Figure 4 in Section 6 for the electrical one-line). As it proceeds, any of the secondary effects 7
previously listed might affect or influence the system and equip-ment. Any loads connected to any of the buses may also be affected.
Note that the figure includes only the emergency power system.
(2) Unbalanced voltages which exist long enough on the system are a potential source of overheating or thermal damage to motors, the main generator, and the unit station service transformers.
Unbalanced voltages can come from. equipment with phase-to phase, open phase, single phase, and partial faults. A main generator can quickly overheat from unbalanced loading if system relaying or circuit breakers fail to clear a fault from the generator system. A motor can overheat from operating single phase or at reduced voltage.
(3) Voltage transients (spikes of sufficient magnitude) might damage semi-conductors and rectifiers (e.g., instrumentation, battery l
charges, inverters, computers, RPS). Voltage transients might l
also cause inadva.rtent or incorrect operation of sensitive equip-l ment, depending on time duration.
1-8 l
l t
.. q MTF SIIORT CIRCUIT FAULT " PROPAGATION" TRAIN B TRAIN A MTF1 FIRE MTF2 g
j g
g
- OIL SPILLS
CBG1 CBI TRANS. RELAY CB2 MISSILES CB3 CB4 500 kV PROT.SYS g
500 kV 500 kV MAIN MAIN CEN1 BUSl~
OTIIER INT'LKD BUS 2 BUSl CBS GEN 2 1
1
"" ' S'S 1
1 540 kv MAIN GEN 1 FAILURE RSS-1 RSS-2 RSS-1 BUS 2
- UNBALANCED LOAD GEN TURBINE l
I k
- MOTORING
~
o 1
34.5 ky. REACTOR OTHER 34.5 kV TURBINE TRIP
- BUSL BUS 2 o
f CB8 I
o o
1 RPS MALFUNCTION CB9 CB10
.C 1
SIMILAR SMiE j
{
{
TO MTF1 AS MTF1
.RSS-C RSS-B
.RSS-A T
CB27 CB12 1
1
-BUS 2C
.BUSF
.CB23 CBiS CB30 CN16 CB25 1
1 NO BUSlB BUSE BUS 2B BUSD BUSlA 1
NO NOTE: EMERGENCY POWER SYSTEM ONLY o
o o
o BALANCE OF PLANT OMITTED CB17 CB18 CB42 CB19 CB43 CB20 NO CO3 NO CN36 CN34 CN44 CN37 NO 1
V 1
N 1
M BUS 2J g, g BUS 1J BUS 2J IMERGENCY BUSES FIGURE FMEA-1
+
a e
g
~
i
)
(4) Overvoltages of long duration or repetitive application may cause equipment overheating followed by f ailure.
(5) Figure FMEA-2 illustrates a scenario for a main generator failure traceable to an MTF. An MTF results'in a voltage tran-sient. A voltage transient of suf ficient time may result (with a spurious RPS actuation) to initiate turbine trip. A turbine trip causes relay protection systems to operate.
Should CB G-1 fail to operate (cpen), the main generator would begin to motor and given sufficient time, f ail.
i (6) The f ailure of the transformer relay protection system or iso-lating circuit breakers to operate following an MTF can place a distribution f ault on the high voltage buses with potentially insuf ficient power on the emergency power system which may affect any permanently connected emergency loads.
(7) A high voltage to low voltage MTF will probably cause damage to the generator system equipment (e.g. buses, generator windings, transformers, etc.) due to the finite time required for system relays and circuit breakers to operate correctly. Failure of j
some equipment to operate properly will result in more damage.
(8) If an MIF exists on MT-1, and it is cleared from the high voltage buses (CB 1 and 2) and the generator (CB G-1) but not f rom the 1A,1B, or 1C buses (CB 22, 24, 26 as shown in Figure 4 of Section 6), any attempt to re energize will be affected because the MTF will appear as a distribution fault on any uncleared station service bus.
i (9) An MTF will activate the main transformer relay protection system, which in turn will activate other systems with which it is inter-locked (e.g. reactor, generator, turbine trip system, and other systems ).
If plant power operation is above the preset trip point, an MTF will result in a reactor trip (scram). Five of the seven NAPS MTFs resulted in scrans for a rate of 2.3 per year (5 MTFs in 26 months) compared to WASH 1400, Table III 4-2
[Ref. 5] transformer f ailure rate of 10-6 per hour or 8.76 x 10-3 per year. This is a ratio of NAPS MTFs to WASH 1400's of 260 MTFs.
~
(10) A f ault of sufficient magnitude in conjunction with transformer pressures, can cause the transformer tank or bushings to rupture and create oil spills, fires, and possible missiles. These ef fects can generate potential f ailures in other auxiliary supporting systems and structures.
t 4
1-10 4
(11) MTFs resulted in the following consequences from missiles, oil spills, and fires.
Item Source Consequence (A) Missiles...... MT Bushings
....... None (B) 3-011 Spills.. Tank Rupture....... See Item (c)
(C) 1-Fire........ Tank Rupture....... RSST cable riser fire, turbine building damage and overhead bus bar meltdown O
Explanation:
(A) The rupture of the bushing porcelain did not generate any observed missile damage. The bushing fragments fell to the tank top or to the surrounding ground.
(B) Three of the seven MTFs caused tank rupture allowing oil to spill to the ground and spurt from the source bay to an adjoining bay and structures. One of the three oil spills resulted in a fire.
(C) The resulting fire from the oil spill caused damage to i
the RSST cable riser, turbine building structure, and and overhead bus bar meltdown.
Based on engineering judgement, the possibility of missile (porcelain) damage to safety related equipment or systens (e.g., diesel generators, motors, control room, etc. ) seems remote because of the reported NAPS experience and the exis-tence of various structural barriers. These barriers consist of the concrete transformer bay walls, turbine building walls and floors, turbine and generator foundations, and other equip-men t.
The fire resulting from the MTF is the only one in NAPS oper-ating history, and no data from other sources were found upon which to base a meaningful comparison with industry experience.
Since the system configuration was practically unchanged, a similar fire could result in similar consequences. The physical configuration of the 500 kV lines to the main transformers of both units is such that there is a remote possibility that a fire of sufficient magnitude and duration at one unit might cause a fault in a 500 kV line at the other unit. This possi-bility has not been examined.
WASH 1400 did not include consideration of MTF caused missiles, oil spills, and fires, and the scope of this report precludes a more extensive analysis.
1-11
i MTF If VOLTAGE TRANSIENT If SUFFICIENT TIME' U
RPS MALFUNCTION If TURBINE TRIP i
1 If PROTECTION RELAY SYSTEM If CB G-1 DOES NOT OPEN If MAIN GENERATOR MOTORS u
SUFFICIENT TIME i
i lf MAIN GENERATOR FAILURE l
l t
I l
4 FIGURE Fl!EA-2 i
- 1-12 i
I 6.
PROBABILITY RISK ASSESSMENT The NAPS has an MTF rate of 3.2 per year or 3.7 x 10-4 per hour.
The basis for the probabilistic risk assessment (PRA) is to determine if the increased frequency of MTFs and accompanying plant trips result in a depar-ture from the design basis and if such a departure could significantly increase the risk to the public health and safety. WASH 1400 [Ref. 12] was used as the data base on which the Surry Power Station, Units 1 and 2, electrical system was modeled. A transformer failure rate of 10-6 per hour or 8.8 x 10-3 per year is defined in WASH 1400. All tables and figures referenced in this section are from WASH 1400, except where noted.
6.1 COMPARISON OF SIMPLIFIED EMERGENCY POWER SYSTEM DIAGRAMS Fi ure II 5-2 illustrates the power system which was analyzed in E
WASH 1400. RSS-B has been added to the WASH 1400 simplified diagram to make the comparison with NAPS more complete. Figure 2 illustrates an equivalent system which applies to NAPS.
Functionally, both systems are similar through the 34.5 kV buses and up to the 4160-volt transfer buses. The WASH 1400 4160 volt emergency power system includes a dedicated on site DG-1 for the Unit 14160 volt 1H bus and a
" swing" DG-3 which can serve Unit 2 or Unit 1 on a priority basis, but not both simultaneously.
In contrast, NAPS has four DGs, one for each emergency 4160-volt bus, which is a more conservative design expected and is to be more reliable than the WASH 1400 design.
At the 4160 volt emergency bus level, the NAPS system appears to have more options in the alignment of offsite power sources to the emergency buses and is expected to be more reliable than the WASH 1400 design.
It therefore appears that the use of the WASH 1400 fault trees will be acceptable for this review and will be conservative.
6.2 HOW MTFS MIGHT CONTRIBUTE TO WASH 1400 FAULT TREE EVENTS An MTF can potentially contribute to the fault tree failure rate calculation in two ways:
(1) As an influence on " Loss-of-Net" (offaite power).
(2) As a contributor to " insufficient power" to the emergency power system sources (offsite or onsite power).
It should be noted that " Loss of-Net" is included in " Loss of-Of fsite Power,"
since the of fsite emergency power system continues f rom the 500 kV buses to the 4160-volt emergency buses (Class 1E) H and J.
1-13
E TRAIN B TR AIN A 230KV 500KV SWITCHYAR D SWITCHY AR D BUSES BUSES AUTO TRANSFORMERS x
22KV 22KV
~
'syd5 <
>a5 UEr[w.
1 YiY
's'es Y MAIN GENERATOR MAIN GENERATOR UNIT I UNIT 2
\\\\NU NYS
\\\\NU
- , :H"?"
- in".'"
- s %""
OFFSITE 4 iso y 4 iso v 4 iso y Tgy,srga inggsrga inggsryn d
V 4isc
TO UNIT 2 SYSTEM 4'so 'M U
ONSITE
/EvEEu <
v y
n, TO BALANCE OF n,
v EMERGENCY POWER SYSTEM v
DIESEL GENERATOR DG-3 DIESEL GENERATOR DG-1 NOTE : SWING DIESEL CAN SERVE UNIT I ON UNIT 2 ON PR!nRITY BASIS BUT NOT BOTH SIMULTANEOUSLY.
(UCLD-iG661 PGS FIGURE 115-2 SIMPLIFIED EMERGENCY POWER BLOCK DIAGRAM 1-14 L
TR AIN ' B TR AIN A SOOKV SOOKV SWITCHY AR D SWITCHY AR D 500KV BUSES BUSES
)
i 1
I I
TO 90P TO BCP
- SYSTEM +
+ SYSTEV RSS 34.SKV Xp R XFMR M AIN GEN M AIN GEN UNIT I 34.5KV 34.5KV UNIT 2 BUS BUS 1
RSS RSS RSS XFMRA XFMRB XFMRC 4160 416 0 4160 XFMR XFMR XFMR BUS D BUS E BUS F I
I 4160 4I60 4160 416 0 BUS 28 BUS 18 BUS 2C BUSlA
~
/
/
~,,. ' '
Q FUTURE 416 0 4160 416 0 41GO IJ IH 2H 2J 4160V E MER GE NC Y BUSES BALANCE OF BALANCE OF E MER GE NC Y E MER GE NC Y P,0WER SUPPLY POWER SUPPLY DG-CG i
FIGURE 2 SIMPi_IFIFIED NAPS EMERGENCY POWE:i SUPPLY (AFTER SURRY - WASH 1400 FIG ll S-21 1-15 l
l l-i
,.m.
6.2.1 LOSS-OF-NET For this repert, " Loss of-Net" is defined as being caused by system instability due to a NAPS MTF which results in loss of generation. The NAPS UFSAR [Ref.1, page 8.2-4] states that the interconnected network can successfully withstand the simultaneous sudden loss of both NAPS main gene-rators with no serious effects. Therefore, a NAPS MTF should not cause loss-of net, and the values used in WASH 1400 estimates need not be changed due to NAPS MTFs.
" Loss of-Electrical-Net" appears in the WASH 1400 fault tree in Figure II 5-4, sheet 9 (page 18) as diamond event 2000020F, which contributes to " Insufficient Power on Bus 34.5-6 (single cut sets only)." Transfer tri-angle 18 moves this event to Figure II 5-4, sheet 10 (page 19), where it contri-butes to event ";P04, Insuf ficient Power (IP) on Bus 34.5-6. "
"JP04, " in turn, contributes to event "JB07, Loss of Offsite Power (LOSP) to 41601H."
Transfer triangle 17 moves the " Loss of-Net" event to Figure 5-4 sheet 8 (page 17), where it also contributes to event "JB00, Insufficient Power on Bus 4160 1H. "
" Loss of-Net" is important as it relates to the calculation of releases which involve the TMLB' sequence [V 4.3, page V-39].
The T = P1 term involves loss of offsite power and represents an interruption of the main feedwater delivery provided by the plant power conversion system (PCS).
The T = Pt term might be affected if an MTF caused a generator trip which re-suited in " Loss of-Net. " The WASH 1400 probability of such a trip is qnet 10-3 [III 6.3 page III-71). The NAPS MTF rate is 3.2 per year. Therefore, 3
x 10-yobability of " Loss of-Net" due to an MTF is 3.2 x 10- per year or 3.7 the p per hour, which is negligible compared to the value of Py = 0.2 per year for the loss of offsite power from all causes used in the MB' calcu-lation [V 4.3.1, page V-39].
The MLB' terms involve non recovery of of fsite power and loss and non-recovery of onsite power to the emergency power system. The WASH 1400 analysis included the contributions of MTFs to the insufficient power events which were used in evaluating the loss and non recovery items above. These quantities should not be significantly affected by MTFs.
The WASH 1400 " Loss of-Net" contributiov to "IP on Bus 34.5-6, (JP04)" and "LOSP to 41601H, (JB07)" is A = 2 x 10-5 per hour [ Table II 5-3 page II203].
The A = 3.7 x 10- per hour NAPS MTF contribution to " Loss-of-Net" due to a MTF is also negligible compared to the WASH 1400 value.
G 1-16
l i.
H.1 P.sulfacount A = 3.6 x 10-6 In
- Oa TB'0d '"
O ir nw P
oa*=g*g0a Og d,"",
(Sheet 8A - Original Tree)
Onlyt L*88 dreakers Tage 601H r3
( 3 e
5HT.3 deans INSUFF Power Froni
(,,og ogg,,
/ Feun On Desse GEN. Systern p
y, g,,
Ceoen seen (Serge Cut See 0%
41601H (Seagle Cut LoeJ Breemer Fasis To Open
[ed)
Sets Ons l v
$HT. 9 17 Bremer
% Loss 2C010CSF 15H3 ZGE 10010 EDG el m 1"rnn CONO.154 ZC810r3F isMg xastn 41MH 5 tub 8us ICS10077 15M6 ZSS100JO Cheryng PP e CM P IC ZC#1011F 15H4 25510110 STA. GEN AUX FD P P.
IC810125 ISMS 25510120 Charyng PP i CM P l 4 ZCa1007F B5MS ACdiOO70 Ca,sse ZC21008F 15H9 XCdl0003 g,,
ZC81011F 15H4 XCS10110 i g,,,,,
ZC81012F 15H5 XCa10120 XCa1013J, Shoru ZC81013F 15M3 ZGdJ0tBP 15H 7 XT R300 53 S$1H XCA30023 CIA wr 13 C2A ACB1002F e5He xca1020 SER.15HG Short XCA tM10 C11A Mast 0020 T-F C01003 1Et RCal m l C31004 15F1 THRu '
4 SERS C81005 25C1 XC810060 -
CB1006 25JS XCA10030 ' CONO To CB 2SC1 XCA100eO
" To C8 2SJS Insufficent Control DISTR 18 Breaker 1$ H3 Seester P w To h he 1
15H3 F asas Breaker l$ H3 When Wettun U"
Traos Open To Close Otferte Power is Lost DSL GEN SYS OW XCA10020 Conductor C15A l
ZC8tt130 Short eER.15H3 Stede' Carcuet in 8asu* eat Oween 1 58* 3 Brusher 8AT T SYS Power Oa sus Generator Pient weaerstor F e='
INADVERT 15H31NADVE R T tOpen 8 ATT DC 1 A l$*98 F asis To Start Or To $uppey Load Tre Opens Oue To
, o, op.,
cut Sea ontvl Faies Te Accept After A succentua Ocea Opere*
Land if Start & Loma ZCalotJY Xav90010 10 ZCS1007X ZGE100t A ZGE100tF SMT. 3 1
FIGURE 115-4 PWR Emergency Power System Fault Tree (She 8) II-261 1-17
"\\
l SHT. 8 Loss Of Offste SHT.10 Power To Bus 41601H (Single Cut Sets Only)
NOTES:
1.
Open circuit failures including transfer Sus F. tnat result is loss of offste power to 41601H.
' 2.
Distritmstion faults, including transfer l
/18 Sus F. that resurt en loss of offste power to 41601H 1 Open encuit fasaures, including high SHT.10 Greaker 15H8 8"Suffic'ent Power vontage sw.tchyard that result in Inmently S"
see t.P. on aus 34.5 6.
Note On Bus 34 54 (Single 4.
O strioution faults. including hsgh N'1
Owns 2
Cut Sets Only) vostage sw.tchyaro, tnat ren.it,in f.P. on Bus 34.5-6.
[\\
XCA10018 C11A XC810020 15H8 f
T XTR10018 RSS C XCA10010 C11A I
XCA10068 C13A XCA10050 C12A ZC81016F LT26 XTR10010 RSS-8 ZC81004F 1501 XC810160 LT26
- 5 XCA10058 C12A XCA10060 C13A X8510028 T.F XC810040 X8S10020 T.F XCA10030 To C8 25-C XCA10040 To C8 Operator Breaker g
g Inadvertently 15HS gag g
ggy Opens Circuit INADVERT 3
Of Electrical 4
Breaker Opens n,,
15H8 2X10-5 ZC810028 X8S10048
- 34. 54 8us X8S00180 5042 EC81002X ZC80041K LT22 Z000020F
~ XC800570 XT562 XCA00148 C14A XC801590 G202 XTR00168 MT2 ZC811170 56 XC880160 LT26 XC810170 56 XC810190 LT16 XC810200 R.22 X8S10040 34.5-5 XC800410 LT22 XCA00140 Cl4A XTR00160 MT2 XC800430 L202 ZT R10180 500 KVA-A ITR10190 500 KVA 8 ZTR10200 500 KVA C XCA10100 CTA XCA10110 CT8 XCA10120 CTC NAPS - MTF is not a single failure FIGURE II 5-4 PWR Emergency POwOr System Fault Tree (Sh. 9) II-262 I
l-18 i
i
\\
t.m of ot*=se A = 3 x 10-5 (Tablo II 5-2, WASH 1400)
- - rx'ao 'a u.On O
-s L.e of ONw o**-'*
gr,47,'"
,a;;j,'g'=
7"T, 'li'-"
A = 2.9 x 10-5 (Table II 5-2) o**
(NAPS MTF Contribution Negligible) upoa
/u\\
b 4.4 x 10-7'((NAPS) 7*9 x 10-8 WASH 1400)
Double Faults 3.6 x 10-8 (WASH 1400) to a.is o'",. 'l',U',",s7,=
con o-**--
t es c s.
'-**=a"--
4.0 x 10-7 (NAPS)
-*=
oc owe 7 0.en
.s 08
/g. '\\
g3 5
4.3 x 10-8 1*7 x 10-5 C-*'='*" '='
sT (Both)
'"",N,,,',.
bs.c.a.
(Loss-of-Net)
"C'"*"
raau Twau
,c, Biard
?!A'ET-mg33!ryo s;s 2a27ma 1 o.esne=ama xcaicoJo iscs assi0aso XCa t0210 1 F ums R
THRU 10 suas o.n e w
xcsiaa0o ie0 ic e.,
c,,,,,,
r.
o, sa
- o. r.
o.=
u.w.a s-e, ct XC8iO18o s 2c nae e 2CB f0tGO LT26 E*AiEll E,.""
cun. nes sse ZCs tot 9o of16 Ret to190 s.
Sr MIS-'EEr xTaimio Tr.n o..
r RfA1GQ5Q Corusucter i.f C12A
- .r Sus 4160 tG
{CB ici90 trfs Ec A f 00 N3
.clicua 00 m Te.8KH $6 acs00no L.T ts
.r.
sc 5
Desreveson F.uns e
la **In VOR ', s.
IC900070
- omee Due f. Osen sWYo R*'eu8t*n c.ru oa ?M 4 24500040 34 SS
'as*"u*s 34'"5'4'"*'
ZT m00180
'5(X) KV A A Anis 500 2 8=ari Ta ZT RODI90 500 K VA 4 "c""x 'Do EA"***
IT "g";'g g 3.6 x 10-8 (WASH 1400) 4 9.0 x 10-11 4.0 x 10-7 (NAPS)
(Both)
EaM 'd a s
-s zcsiO20o a 2a i
I
!.i'.*sa.a E.*,".70s.= a
= gg, c3,0,,,
c7 o**='*
o.r un v
x1w2 i av seiecow.130 o.
s so.esavor.
230 4 5m 1.8 x 10-8 (WASH 1400) 9.0 x 10-9 E E E IE' EE" Edr 3
1 1.8 x 10-8 (NAPS)
(Both) 9x10-9 6' ASH 1400dSEE E$'l RCA00230 Line gy 3.8 x 10-7 (NAPS)
,,s1,,,
is t...
.s1...
r o.m
%g -
XT.b[8.i's 8
G *M e
an L207 8 1 -
,rs, c2n 10-3/D UWa U"!
"aPJ EL re' =,,,
=c=="o 10-3=/o ic ~
"c^=2"
- *3
""u~'**'
ce=w c>=
M'%E IE'.,
(NAPS MTF) s MTF + G102 fails to open FIGURE II 5-4 PWR Emergency Power System Fault Tree (Sh. 10)11-263 1-19
6.2.2 INSUFFICIENT POWER TO EMERGENCY BUSES Iusufficient power to an emergency power system bus can be caused by:
(1) Bus opens (2) Bus shorts (3) Cable opens (4) Cable shorts (5) Distribution f aults (6) Power source insufficiencies (a) Onsite diesel generators (b) Offsite sources (net, high voltage 500 kV and 34.5 kV buses)
These were all considered in the WASH 1400 fault trees with Tables II 5-4 through II 5-41 identifying the values used in the quantification of the reduced f ault tree events. The published trees were reduced from the original trees and do not include events with insignificant contributions. Table II 5-2 in WASH 1400 summarizes the individual emergency power system events and quantities used and Table II 5-3 explains how the quantities were calculated and describes the relationships and the dependencies among the events.
Figure 3 (page 21) graphically illustrates these relationships. The original trees included both the "H" and the "J" train events.
l All of the WASH 1400 dominant accident sequences and release category characteristics are based on the event and fault tree quantities in WASH 1400, Table 5-2, Main Report (page 79) and Table V 3-14, page V-25.
t l
l Figure 3 demonstrates that the inquiry can be limdted to the effects of the NAPS MTF to the events " Loss of-Net," "JP04," and "JB00," since they are the bases for the other events in the figure.
l l
l 4
1-20
CONTRIBUTORS TO QU ANTIFIC ATION OF WASH.1400 IP FAULT TREES (TABLE II 5-3)
JV03 JVol A
~
JH00 JF00 JD00 5
JB00 o
- 4. x id IP ON 4160H A = 3.s x ids O!ESEL FAILS tjpl7 O
0 = ig5 ar IP 7
TO START orrsiTE to 4: sos x. 3 x i65 DURING LOCA PsV !AB E 7 A A
JPO4 a = ii' 2 DIESELS FAIL 0THER IP ON 34.5-6 x = r.s x id' TO START (P)
I I
LOSS OF o, id3
^'#**'
CCNTRI TION DOMINANT CONTRIBUTION 1-21
---INSIGNIFIC ANT C ONTR IBUTION j
FIGURE 3
- 6. 3 -
WHERE PffFS APPEAR IN THE FAULT TREES (1) The location of the generator MTF branch was found in the WASH 1400 electric power fault trees, and an estimate was made of the branch contribution to the original results using WASH 1400 failure rates. This value was then compared to the contribution made using the NAPS MTF rate. The MIFs would contribute in the WASH 1400.f ault trees at "JPO4, Insufficient Power on Bus 34.5-6" as distribution faults not cleared by a circuit breaker (Appendix II, Figure II 5-4, sheet 10, page II-263 and Table II 5-7.,
page II-202] as follows:
s WASH 1400 Event "JPO4" A = 2.9 x 10-5
. WASH 1400 contribution from Contribution including failure of distribution faults and other circuit breaker to clear NAPS MTF events MIF = 4.4 x 10-7 DF = 7.9 x 10-8 A
A i
The ratio,1MTF = 5.6, but the net effect of either of the contri-A DF butions to "JPO4," A = 2.9 x 10-5, is negligible.
(2) It is conceivable that an MTF could cause a loss of net under some circumstances. However, VEPCO system stability character-istics can withstand the simultaneous sudden loss of both NAPS main generators without loss-of net [UFSAR, page 8.2-4].
There-fore, NAPS MTFs should not alter the loss of-net probabilities from those used in WASH 1400 provided the circuit breakers clear 4
the faulted transformer from the buses. -If the breakers do not clear the fault, an MTF becomes a distribution f ault on the system. The complete WASH 1400 fault trees included the distri-bution fault events related to main generator system f aults, i
including MTFs, but the WASH 1400 reduced trees omitted them because of their negligible contribution to the results.
l (3) Locations on the WASH 1400 fault tree where MTFs influence calcu-lations of " Insufficient Power on the 34.5-6 Bus" are as follows:
i LOCATION
- Event "JP04, IP on 34.5-6 Bus" [ Figure II 5-4 Sheet 10 and Figure II l
5-1 Sheet 1]
1-22
s 500 SOS 2 230 4 L202 M I S'd a
Glog di4s4 Jl/ ldtJ02 N
C202 34.5
{
20
[
N wit C1
(
b Cito?
G2?lF'
' [] CIT 24u
?'41Ti 2 21..
7237263 2
L L[] E [] E C)
E t_
t_
C,2n
"'" Q 4002 3102 f202 Mc?
Cas e no Lti230 Y
$H Reasses 4I R
b iP= C 14e C14A MT500 22 R12 "I' LT12 R22 ()LT22 i
= $ L_^
8 M
E
) I*
u NL I '"
Uma 1 C338 7 UN2
- 3... nia. A y+
ma.
4 P"'" C'3"a"Ai C hAug hC120 d
Lluaet 21 33 TR A g mL tums 21 hRANS N ?5Ci I
9' 4,,0, A,3,,A 2
'r;) i ta
(,gy se 1C C12A LJ 3
... 3, 2
',:' )
.C v=
w
,,q ".
n.!)..II 1
A y,,n..
), L_.,_
2,., h L_^ _. !
i e_
i a ;.
2,
""'E "A.. C.n i,Ci g
n C,,,
m-C"^
l 15Je )
is.n.3 )
41601Jf5TUS SUS) l 418 1J dl641H l
al601MISTUS BUS) l)
(
)
3 i
)1$J3 ISJ7
)
^
ISM 7 ggw P
C,..
~
l P-ggg NL C1lE Sk
<F CISA CIS -
gg a
4 2SJ3 4 & '180 My SS T ANSIJ M&
w y y $5TRANSIM g,y, C2.,,
,7-- C2 A o
)t4J1
)14H1 4841J l
g egggy 34JG
)
14J10 14MIO g
gy,g l
C38 "b r-C3A l
l NL
$L SL NL Cee-gi
<p-Cea WCC1J12 WCC lHl.2 s
NL St SL NL A
B (Sh 2)
(Sh 2)
FIGURE il 5-1 Single Line Diagram (Sh.1) II-250 Surry 1/2 I
i 1-23
COMMENT The MTF appears on the HV buses as a multiple f ailure distribution f ault in which certain circuit breakers f ail to open on demand to clear the MTF from the system. These fault paths are as follows:
For 34.5-6 Bus (a) MT-500 via CB G-202, MT-2, CB LT-22 (b) MT-230 via CB G-102, CB L-202, MT-2, CB LT-22 For 34.5-5 Bus l
(a) MT-230 via CB GlT 240, CB 24002, CB L-102, MT-1, CB LT-12 L
(b) MT-500 via CB G2T 577, CB 55702, MT-1, CB LI-12 This type of f ailure was considered in the original f ault trees via an event called " Generator System Fault." It was omitted from the f ault trees published in WASH 1400 because its contri-bution was insignificant.
In no case does the output of the main generator and transformer serve directly the emergency power system [page II-82, II 5.1.3].
The station service main 7
l generator system and balance of plant (BOP) systems were included in WASH 1400 fault trees only as they affect the IP tree events as distribution faults The estimated contribution of MTF distributior f aults can occur, as shown in Figure II 5-4 (Sheet 10) page 11-263 via two event 7
paths.
Path 1
" Distribution Fault Through BKR L 202 in 230 kV Switchyard."
Circle Item " Distribution Fault on 230-4 BS of BKR L 202" Diamond Item ZCB0043D "BKR L 202 Fails to Open."
Path 2
" Distribution Fault Through BKR G 202."
Circle Item " Distribution Fault on Cable Side of BKR G 202" Diamond Item ZCB0059F "BKR G 202 Fails to Open."
(4) Corresponding MTFs at NAPS are described below (Figure 4):
LOCATION IP on 34.5-1 Bus IP on 34.5-2 Bus 1-24
-e--
v
M lOL OTHI AN NORTH sys7gy FIo$k 2-13
"*"S 8
V IR GIN I A L ADYSM ITH -E MER GE NCY POWER M
system IEPSI 500KV -i
---rU TU R E 6l l3l I 4 i l2 SI 500KV -2 L
\\
d RSS-2 NW WW RSS-I TR AIN 8 l-30 l-30 TR AIN -1 34.5 -2 l 8 lNC l 7 lNC 34.5 - l d
lTIl l II lNC l10 NO 9
NC NC 3io#
"E-ion NW W
4.iskv
- 4. tex /,NC
- '8#"
l 16 lNC l15(JC l14 l l 13 lNCll2 lNC D
E
,F 20 lNC If NC T2l-NCl18 l l 17 lNC uT-t
^*
NO zav ss 22KV MT-2 uss i^.4 u uss.'AfN USS-IA4 N NW NCl GI INC Nb NCl24 (cl23 l NC l22 lygl 21 l 2
^
[lGKVIqlNO IC
[--.4.16 K V 4.16 K V LWN O UNIT I W
,j
^4NU S S - 2 l
WW uss-2 32 31 30 29 NC NC 28 27 NO NC NO NO 2A 2C 4.16 K 4.16KV 4.16KV,~~ 1 43 NO
.INO
??
r--,
NCip2; h
hNC bh S
b
.No NC l44lNOl36 l3Sl
!34 3l NO NC N0 NC UNIT 2 4.16XV IJ { 4.16KV IH { 2H{ 4.16KV 3;[ {
] 4.16KV 4,
3g 2J{
N01 gpgj NO gygj NO
_,W W wZ NO DG D3 00)
FIGURE 4 i
NAPS FAULT PATHS 1-25
COMMENT The MTF appears on the HV buses as a multiple f ailure distri-bution fault in which certain circuit breakers fail to open on demand to clear the MTF from the system. These fault paths are as follows:
For 34.5-1 Bus (a) MT-2 via CB-3, RSS-1, CB-7
~
(b) MT-1 via CB-1, RSS-1, CB-7 For 34.5-2 Bus (a) MT-1 via CB-2, RSS-2, CB-8 (b) MT-2 via CB-4, CB-5, RSS-2, CB-8 NOTE - Paths other than those listed above could exist, depending on the operating alignment of the HV circuit breakers.
1 6.4 ESTIMATES OF NAPS MTF CONTRIBUTIONS TO FAULT TREE QUANTIFICATION j
There are five event paths in which a NAPS MTF contributes to the fault tree quantification of "JP04, Insufficient Power on bus 34.5-6."
See Appendix 1 for the accumulative contribution derivations.
Path I contributes to the event " Distribution Fault Thru Breaker L 202 in 230 kV Switchyard" via distribution faults on bus 230-4 on both sides of breaker L 202, which f ails to open, and an added event of the 3
f ailure of breaker G 102 to open.
Path 2 contributes to the event " Distribution Fault Thru Bkr G 202" via distribution f aults on the cable side of breaker G 202 and the f ailure of breaker G 202 to open.
Path 3 contributes to the event " Distribution Fault Thru Bkr XT562 (in 500 kV switchyard)" via distribution f aults on the cable side of breaker XT562 and the f ailure of breaker XT562 to open.
Path 4 contributes to the event " Insufficient Power Due to Open i
Circuits on 230-4 and 500-2 Buses" via open circuits on the 230-4 and 500-2 buses.
Path 5 contributes to the event " Distribution Fault Through Load Breaker" via distribution faults on the cable side of the breaker and the load breaker f ails to open.
l 1
l 1-26 l
1
The total contribution of the above five paths to the quantifi-cation of "JP04, Insuf ficient Power on Bus 34.5-6" is 4.4 x 10-7 for NAPS MTF and 7.9 x 10-8 for WASH 1400.
Since the combined effect of all five contribution paths to "JP04" plus the single cut sets. (transfer item triangle 18) is 2.9 x 10-7, contribution of the increased NAPS MTF rate is negligible.
7.
CONCLUSION Based on the analyses of the present information available on the main transformer f ailures and f requency of occurrence at the North Anna Power Station, Units 1 and 2, it is concluded that:
(1) The increased f requency of nain transformer f ailures and accom-panying plant trips increases the overall challenges but does not appear to significantly degrade the plant protection and reactor coolant system designs. The total number of reactor trips experienced in the 26 month period of transformer failures (with or without MTF contribution) is consistent with that shown in the data bases compared in this report.
(2) The main transformer f ailures can have primary and secondary effects on plant systems and equipment depending on the adequacy and reliability cf the detection and protection systens for iso-lating the faulted transformer. The adequacy of these systems will be evaluated in a separate interim report.
(3) Based on the WASH 1400 data bases, the contribution from a main transformer failure affecting the plant safety systens so as to increase the risk to the public health and safety is negligible.
f i
1-27
REFERENCES 1.
Virginia Electric Power Company's North Anna Power Station, Units 1 and 2, Updated Final Safety Analysis Report.
2.
VEPC0 Report of Task Force to Investigate the Failures of North Anna No. 2 Transformers, Olin R. Compton, dated September 30, 1981.
3.
North Anna Cenerator Step up Transformers Report of Failures in 1980-82 (draf t testimony), later incorporated into a main report, dated January 29, 1982.
4 Notes from Westinghouse and VEPCO meeting held at Muncie, Indiana on February 9, 1983.
5.
VEPC0 memorandum, D. E. Thomas to W. R. Cartwright, undated.
6.
NRC memorandum, L. B. Engle through R. A. Clark to T. M. Novak, dated July 6, 1981.
7.
NRC memorandum, L. B. Engle to R. A. Clark, dated December 17, 1982.
8.
Telecon, L. B. Engle (NRC) to J. C. Selan (LLNL), dated April 21, 1983.
9.
Telecon, L. B. Engle (NRC) to J. C. Selan (LLNL), dated May 3, 1983.
10.
EPRI NP-2230, "ATWS: A Reappraisal, Part 3, Frequency of Anticipated Transients," dated January, 1982.
g 11.
Commonwealth Edison Report, " Zion Probabilistic Safety Study," dated September, 1981.
12.
WASH-1400 (NUREG - 75/014), Reactor Safety Study, An Assessment of Accident Risks in U.S. Commercial Nuclear Power Plants, dated Octobe r, 1975.
1-28
APPENDIX 1 In Path 1, the estimated contribution of the MTF will be via an event added to Figure II 5-4, sheet 10, from the original tree, " Generator #1 System Fault, JP22." This item transfers the MTF to other parts of the original tree and combines as event "JP29" with the failure of breaker G 102 to open:
WASH 1400 NAPS MTF Path 1 (6 faults) 1.8 x 10-5 Gen 1 system fault incl.
3.7 x 10-4 NAPS MTF (JP29) and L 202 fails to open 10-3 G 102 fails to open 10-3 1.8 x 10-8 3.7 x 10-7 6 faults 1.8 x 10-5 1.8 x 10-5 L 202 fails to open 10-3 1.8 x 10-8 Path 2 (3 faults) 9.0 x 10-6 3 faults 9.0 x 10-6 and C 202 fails to open 10-3 MTF rate 3.7 x 10-4 9.0 x 10-9 3.8 x 10-4 G 202 fails to open 10 -3 3.8 x 10-7
.e A third path, via " Distribution Fault through BKR XT562 in 500 kV Switchyard," contributes to the distribution fault with a value of 9.0 x 10-9 The combined contribution of these three paths to the "JPO4" quantity ist Path 1 1.8 x 10-8 1.8 x 10-8 Path 2 9.0 x 10-9 3.8 x 10-7 Path 3 9.0 x 10-9 9.0 x 10-9 Paths (1)+(2)+(3) 3.6 x 10-8 4.0 x 10-7 1-29
A fourth path via " Insufficient Power Due to Open Circuits on 230-4 and 500-2 Buses," contributes to the."JPO4" quantification as follows:
WASil 1400 NAPS MTF Bus 230-4 (2 faults) 6.0 x 10-6 Same Bus 500-2 (5 faults) 1.5 x 10-5 Same Path 4 9.0 x 10-11 9.0 x 10-11 at A
Path 5 via " Distribution Faults through Load Breaker" is another set of double faults which contributes as follows:
10 Cable Side Shorts 3.0 x 10-5 Same 6 CB faults 6.0 x 10-6 Same 1 Bus fault 3.0 x 10-6 Same 4 Transformer faults 4.0 x 10-6 game 4.3 x 10-5 Same Load Breaker fails 10-3 Same to open Path 5 4.3 x 10-8 4,3 x 10-8 The total contribution of Paths 1, 2, 3, 4, and 5 is:
1,2,3 3.6 x 10-8 4.0 x 10-7 4, 5 4.3 x 10-8 4,3 x 10-8 Total 7.9 x 10-8 4,4 x 10-7 Path 4 was neglected since its contribution of 9 x 10-11 is negligible.
NOTE:
t a
1-30
9
' TASK REPORT 2 EVALUATION OF THE " ROOT CAUSES" FOR THE
~
MAIN TRANSFORMER FAILURES AT THE j
NORTH ANNA POWER STATION, UNITS 1 AND 2 Kerry J. Dalton and James C. Selan
\\
2
ABSTRACT This task report documents the evaluation of the " Root Causes" on the seven main transformer failures at the North Anna Power Station, Units 1 and 2 The evaluation was based on the available information and operating histories surrounding each transformer. The evaluation determined that there was no known single common cause which contributed to the seven f ailures. The f actors which were predominant contributors to the f ailures were excessive shipping, handling, repeated installations, and improper storage of components when not in use.
l FOREWORD l
This report is supplied for the U. S. Nuclear Regulatory Commission, Office of Nuclear Reactor Regulation, Division of Licensing, by Lawrence Livermore National Laboratory.
The U. S. Nuclear Regulatory Commission funded the work under the authorization entitled " Evaluation of Main Transformer Failures at North Anna Power Station," B&R 20 19 10 12 1, FIN A-0442.
.=
4 t
2-1
4 TABLE OF CONTENTS Page 1.
INTRODUCTION.
2-1 2-2 2.
FAILURE REVIEW DOCUMENTATION.
3.
SYSTEM CONNECTION.
2-2 4
TESTING AND MAINTENTANCE OVERVIEW 2-4 i
5.
" ROOT CAUSE" EVALUATION 2-6 5.1 ' TRANSFORMER FAILURES.
2-6 5.2 VEPCO'S TASK FORCE INVESTIGATION.
2-10 6.
SUMMARY
2-10 7.
CONCLUSION 2-11 REFERENCES.
2-12 i
TABLE OF ILLUSTRATIONS r
Figure 1 - Main Generator Step-Up Transformer 2-3 Connections l
Diagram 2-3
)
i l
l 2-111 i
i
(
l
EVALUATION OF THE " ROOT CAUSES" FOR THE SEVEN MAIN TRANSFORMER FAILURES AT THE NORTH AN!!A POWER STATION, UNITS 1 AND 2 (Docket Nos. 50-338, 50-339)
Kerry J. Dalton and James C. Selan Lawrence Livermore National Laboratory, Nevada 1.
INTRODUCTION 4
Virginia Electric Power Company (VEPCo), the licensee, experienced seven main transformer failures from November 29, 1980 to December 5, 1982 (a period of 26 months) at the North Anna Power Station, Units 1 and 2.
Of the seven f ailures, the first five occurred at Unit 2 with the remaining two at Unit 1.
These transformers are single phase units. The output of the main generator supplies power through isolated phase buses to a bank of three single phase units where the generator output is stepped up from 22 kV to 500 kV for connection to the switchyard.
The transformers were all manuf actured by Westinghouse. The units were delivered to Unit 1 in 1971 and to Unit 2 in 1974 Westinghouse and VEPCo conducted joint and independent investigations to determine the cause of the failures. Each failure was analyzed by establishing an accurate operating history from which a hypothesis of the failure mechanisms was formulated. From this the licensee identified actions required to prevent the reoccurrence. At present, a single common cause has not been identified for the seven f ailures.
This report documents an evaluation of the " Root Causes" of the seven main transformer f ailures based on the information supplied by VEPCo, Westinghouse, and the NRC.
2-1
2.
FAILURE REVIEW DOCUMENTATION Numerous amounts of information in the form of memoranda, instruc-tion leaflets, letters, trip reports, and investigation reports was supplied.
A listing of only those documents used in the evaluation is as follows:
(1) VEPCo memorandum, D. E. Thomas to W. R. Cartwright, undated
[Ref. 1).
(2) VEPCo memorandum, R. E. Bridges, Jr. to W. R. Cartwright, dated December 11,1980 [Ref. 2).
(3) VEPCo memorandum, R. E. Bridges, Jr. to W. R. Cartwright, dated December 29, 1980 [Ref. 3].
(4) VEPCo Report of Task Force to Investigate the Failure of North Anna No. 2 Transformers, Olin R. Compton, dated September 30, 1981
[Ref. 4].
(5) NRC memorandum (November 29, 1982 meeting), L. B. Engle to R. A. Clark, dated December 17, 1982 [Ref. 5].
(6) NRC memorandum (December 9,1982 meeting), L. B. Engle to R. A. Clark, dated December 17, 1982 [Ref. 6].
(7) North Anna Generator Step up Transformern Report of Failures in 1980-82 (draf t testimony), later incorporated in a main report, dated Janaary 29, 1982 [Ref. 7].
(8) Westinghouse letter (D. A. Yannucci) to James C. Selan (LLNL),
dated March 28, 1983 [Ref. 8].
(9) Westinghouse letter (R. A. Wiesemann) to the NRC (L. B. Engle),
dated May 6,1983 [Ref. 9).
(10) VEPCo letter (W. L. Stewart) to the NRC (H. R. Denton and R. A. Clark), dated August 17, 1983 [Ref. 10).
(11) VEPCo letter (W. L. Stewart) to the NRC (H. R. Denton and J. R. Miller), dated December 16, 1983 [Ref. 11].
3.
SYSTEM CONNECTION l
A simplified diagram of the connections for the main generator and main generator step up transformers is shown in Figure 1 (typical of both units). The generator output. (each phase - 22 kV to ground) is connected in a 9 Wye" configuration with the neutral grounded through a grounding transformer and a resistor. The generator output at the main transformer bank (3-19 units) l.
is connected in a " Delta-Wye" configuration where it is stepped up to 500 kV 2-2
A j'
A B
C AB C
A MAIN A
B C
TRANSFORMERS W
W W
d i
_L
=
C B
TO UNIT STATION SERVICE TRANSFORMERS sigi.
- nyi, t
GENERATOR
\\___
BREAKER
/
(UNIT 1 A
ONLY)
A B
C
'T" GENERATOR C
B l
.1.
7 GENERATOR NEUTRAL GROUNDING
?
RESISTOR TRANSFORMER
~f FIGURE 1 - MAIN GENERATOR STEP-UP TRANSFORMER CONNECTION DIAGRAM 2-3
(each phase - 288 kV to ground). A unit station service transformer (a 30 unit) from the generators isolated phase buses is connected " Delta-Wye" to step down the voltage from 22 kV to 4.16 kV.
A generator breaker is utilized only at Unit 1.
4 TESTING AND MAINTENANCE OVERVIEW With an ever increasing emphasis on power system reliability and the avoidance of power interruptions, proper maintenance of all system components is a must. Maintenance schedules will vary depending on the size, complexity, and relative importance of the unit to the system.
Since transformers require removal from service to accomplish main-tenance (except for some visual inspections), it is important that thorough maintenance steps and practices be established. This includes having trained personnel and the proper equipment. Manufacturer-recommended procedures should be followed. These procedures, namely shipping, handling, installation, main-tenance, testing, etc. will be found in the manufacturer's instruction leaflets (ILs) for the component.
VEPCo's procedures call for the following routine tests (with accep-tance guidelines) to be performed before energization or af ter any major incident:
TEST ACCEPTANCE GUIDELINE Insulation Power Factor Less than 1%
Bushing Insulation Power Factor Less than 1%
Transforcer Excitation Less than 120% of previous data Transformer Turns Ratio Less than + 0.5% deviation from previous data Bushing Capacitance Less than 110% of previous nameplate data Lightning Arrestors Doble Engineering gudelines Oil Dielectric more than 26 kV ASTM
- 1816 40 mil gap Oil Moisture Content Less than 25 ppm
- Oil Power Factor Less than 0.1%*
Dissolved Gas-In-011 Less than 240 ppm H
- 2 160 ppm CH
- 4 115 ppm C H
- 26 190 ppm C H
- 24 trace C H
- 22 580 ppm CO*
The values in the acceptance criteria are the limit values which require further investigation.
They are not indicatve that a problem exists, but do require an analysis for being outside the acceptance range.
1
- American Nuclear Insurer's Guide 2-4
l i
In addition to the routine tests above, induced voltage tests were performed on all the units at Unit 2 af ter the fourth f ailure and on all units at Unit 1 after the sixth failure. VEPCO states that the results of these tests were all satisfactory [Ref.10].
Corona. tests were also performed on serial Nos. 7001993,_7002099, and a spare (GE) prior to the fourth failure where the results of these tests (20-35 pV at both.150% and 135% test voltage) were all acceptable and well below the manufacturer's acceptance level of 150 pV.
All of these tests were independent of those performed by Westinghouse at the f actory at the time of repair.
By [[letter::05000339/LER-1983-077-03, /03L-0:on 831120,fire Damper in Safeguards Ventilation Sys Found Failed in Closed Position.Caused by Failed Fire Damper Fusible Link.Link Replaced & Fire Damper Restored to Operable Status|letter dated December 16, 1983]] [Ref. 10], VEPCo states that their standard procedures for the ins.tallation and maintenance for the Westinghouse transformers were designed to meet or exceed those contained in the applicable Westinghouse instruction leaflets.
Several specific procedures highlighted were:
(
(1) Monthly tests for dissolved gas beginning in September,1980.
This was changed to quarterly test af ter 7 months. Due to the failures in 1981 and 1982 the frequency was increased.
j-The increased frequency was not documented.
(2) Daily recording of the temperature gauges.
1 (3) Daily noting of the operation of the pumps and f ans.
(4) Quarterly inspections in greater detail of the auxiliary equipment and devices.
(5) Special precautions are undertaken to assure that the trans-former is not contaminated during tank work.
This includes following all recommendations by Westinghouse as addressed i
I in the instruction leaflets.
(6) Storage, handling, and transportation procedures are in accordance with those outlined in the instruction leaflets.
4 t
i l
l J
2-5
5.
" ROOT CAUSE" EVALUATON 5.1 TRANSFORMER FAILURES An evaluaton of each failure based on the available information is as follows:
Failure No. 1 NA-2 9 A Serial No. 7002099 (November 29, 1980)
Type:
High Voltage (HV) Line Lead to Low Voltage (LV) Winding i
Cause:
The initial problem was an inoperative winding hot spot temperature indicator. The indicator is used to turn the second half of the cooling devices on when the hot spot temperature reaches 70*C.
The first half of cooling comes on when the transformer is energized.
Since the second stage of cooling did not come on, the top oil temp-erature device gave an alarm, excess N2 pressure caused the sudden pressure relay and mechanical relief device (MRD) to operate with a high oil level alarm. An inertaire valve was closed by operating personnel to lower the N2 pressure and let the MRD reset.
It was noticed that the second cooling stage was not operating and the winding hot spot indicator read O'C.
The operator turned on the coolers manually, but did not open the inertaire valve. The cooling action of the additional vacuum in the gas space caused evolution of bubbles in the insulating oil. The bubbles greatly reduced the dielectric strength and allowed the flashover between the HV line and LV winding.
Secondary Factors:
The unit failed at Georgia Power Co. in 1976 and was rebuilt. The unit was over-filled with oil, leading to the MRD operation on high temperature. The unit was over excited to 141% and 114% for undeter-mined periods of time. The unit was subjected to excess shipping and handling. The unit was stored in N2 for two months.
Conclusions and Corrective Action:
i The operating personnel were not properly trained in the functioning philosophies of EHV forced oil cooled with forced-air cooler (FOA) generator step up transformers and their auxiliary devices. The people who operate the equipment should be aware of the following:
I (1) FOA transformers do not have a self cooled (OA) rating.
l (2) How the coolers and pumps are energized.
(3) A 0' winding hot. spot temperature is not normal.
2-6
(4) All auxiliary devices should be monitored on a frequent regular basis.
The personnel should be f amiliar enough with all devices to know what to look for.
(5) The annunciator should have more than one remote closure to identify what is wrong.
(6) Maintenance and monitoring records need to be kept current and
~
supervised.
(7) If the cooling controls are _ not checked, then a second winding hot spot device should be installed to provide redundant control.
Top oil temperature control is not recommended for F0A transformers.
(8) Replace the inertaire oil preservation system with the constant oil preservation system (COPS). This will eliminate the bubble formation.
(9) Improve relaying to reduce the instances of over excitation.
Failure No. 2 NA-2 9 C Serial No. 7002100 (June 19, 1981) 1 Type:
Improper storage of the HV bushing caused the dielectric breakdown of the internal insulation which resulted in the f ailure.
Inadequate and conflicting manufacturers ' instruction leaflets [Ref. 4] on how to store EHV condenser bushings precipated the improper tushing storage. This also led to the improper positioning of the bushing during its extensive transportation. The HV bushing was shipped 4 times, installed 3 times, and stored with a 3*-4' angle of r :aose for 5 months.
Secondary Factors:
The unit was in service in the same transformer bank when 7002099 failed at Georgia Power and at North Anna Unit 2.
The unit was shipped excessively.
The unit was stored in N2 for 5 months. The unit was over excited at least 2 times at 25 kV and once at 31 kV (generator voltage base).
Conclusions and Corrective Action:
Bushings must be stored properly (20* above the horizontal).
It is recommended that the manufacturer provide a shipping container that 4
will not allow the bushing to be -stored in an improper manner.
The warranty would be voided if the bushing is stored in any other way.
The manufacturer also needs to produce instruction literature that applies strictly to EHV bushings.
General literature that is out of date should be eliminated.
Excessive shipping and repeated instal-lations of the high voltage bushing must be miniedzed.
2-7
Failure No. 3 NA-2 0 B Serial No. 7002098 (July 3.1981)
Type:
HV Bushing to LV Winding and case (ground)
Cause:
Improper bushing storage caused the bushing to fail and flashover to the LV winding and ground. The same discussion as for Failure No. 2 applies.
Secondary Factors:
The unit was in service in the same transformer bank when 7002099 failed at Georgia Power and when 7002100 and 7002099 failed at North Anna.
These disturbances put overvoltage (% not documented) on the unit which hastened this failure. Also numerous over excitations occurred (see Failure No. 2).
The unit was stored in nitrogen for six months.
As with the previous unit there was excess shipping and repeated installation (see Failure No. 2).
Conclusions and Corrective Action:
Same as f ailure No. 2.
Failure No. 4 NA-2 0 C Serial No. 7001527 (July 25. 1981)
Type:
HV Line Coil to LV Windiag Cause:
An incipient fault was in the LV winding adjacent to HV line high-low space.
This transformer was in service in the same transformer bank when three separate failures occurred. The.overvoltage produced by one or all of these f ailures may have caused the LV turn-to-turn failure. This failure generated carbon particles which reduced the dielectric strength of the oil between the HV line coil, and the nearby LV winding ground and flashover occurred. These faults produce short circuit forces that are quite strong; hence, coil movement. The previous failures also produced f ault currents causing winding movement which could have produced turn insulation damage.
Secondary Factors:
The transformer was shipped 3 times.
When unit was shipped from Georgia Power, vacuum was not pulled, as required, prior to filling it with dry air. During oil filling at NAPS, water and percent gas content data was not available. Tests made later on the oil from the oil processor used in the filling indicated a high probability that the oil met specifications [Ref. 4].
The utility did not know how many pumps and coolers were operating when the f ailure happened.
Conclusions and Corrective Action:
Reduction of transportation, handling, installation, and removal is necessary. EHV transformers cannot be treated in the same manner as 138 kV substation transformers. Maintaining records and knowing actual system conditions are necessary. When overvoltages are experi-l enced, additional surveillance on the connected equipment and on the other phases should be implemented.
2-8 l
Failure No. 5 NA-2 0 B Serial No. 7002100 (August 25, 1982)
Type:
Mechanical failure of HV Line Bushing and flashover to LV Winding and Tank.
Cause:
Bushing had a fatigued bolt which was found at the point of failure.
The fatiguing was due to many shipments (5), installation and removals.
Failure No. 6 NA-1 0 C Serial No. 7001995 (November 16, 1982)
Type:
HV Bushing Corona Shield to LV Winding Cause:
This transformer f ailed f rom the HV line bushing corona shield via the insulating cylinder " stovepipe" to low-voltage coil No. 14 Due to cooling of the ambient (30' C change), the nitrogen originally in solu-tion in the oil may have become insoluble and formed bubbles in the oil prior to being energized and f ailing. When the transf ormer was ener-gized and the first stage of cooling care on and started circulating the bubbles, a sufficient number of these bubbles became entrapped between the stovepipe and the HV bushing to reduce the dielectric strength and precipitate the failure.
Conclusions and Corrective Action:
A possible solution, when a transformer is energized under " cold" condi-tions, is to preheat and circulate the oil by running one stage of cooling prior to energizing the transformer.
Installation of a COPS oil preservation system would eliminate N2 bubble generation.
Failure No. 7 NA-1 0 B Serial No. 7001994 (December 5, 1982)
Type:
HV Line Lead to LV Winding to Ground causing LV turn-to-turn failure.
Cause:
This unit underwent maintenance and modifications prior to the f ailure.
The unit was retrofitted with the COPS oil preservation system. The
" stovepipe" was removed f rom around the HV bushing. The cooling pumps were removed and refurbished due to metallic contamination (bearing failure) found in the oil. The unit was flushed, oil filtered, and refilled following Westinghouse cold weather re-impregnation procedures (Ref. 10]. Fault arc burns were found on a LV winding at a distance of 41" from the HV lead.
VEPCo fo]Iowed Westinghouse's new procedures for startup by circulating the oil prior to energization. Westinghouse has been unable to produce bubbles f rom the HV lead except under vacuum conditions. Therefore, it is felt that foreign material may have caused this failure.
Either all the metallic particles were not flushed out, or due to the internal work done, some foreign material was introduced.
No cause for f ailure could be identified by the factory engineers af ter the teardown.
2-9
.i S cnndary Factora:
Unit was in the bank when 1995 f ailed.
Conclusions and Corrective Action:
There is no guarantee that the presence of foreign material can be totally eliminated. However, conservative procedures must be used when personnel are working inside transformers to prevent the intro-duction of foreign material.
When overvoltages are experienced, additional surveillance on the connected equipment and the other phases should be implemented.
5.2 VEPCO'S TASK FORCE INVESTIGATION Based on engineering judgement of the results of VEPCO's investigation of the seven failures submitted in References 4, 7, and 10, we find that VEPCO has established reasonable root causes and/or hypotheses of the f ailure mechan-ises. They have identified corrective actions and/or recommendations to prevent the reoccurrence which we find are in accordance with our evaluation.
See also Task Report 4
" Evaluation of the Licensee's Proposed Modifications," for addi-tional evaluations.
6.
SUMMARY
A summary on the evaluation of all of the available f acts and operating histories for each of the seven transforner failures is as follows:
Failure No. I was caused by a winding temperature indicator failure and operator error.
Failures No. 2 and No. 3 were caused by bushing f ailure due to improper storage. Clear and non conflicting instruction literature must be produced by the manufacturer.
Failure No. 4 was caused by an incipient fault.
No comment can be made other than to exercise care in shipeent and installation since the unit was stressed by being in a bank with several other f ailures.
Failure No. 5 was caused by mechanical f ailure of the bushing due to excessive shipments and handling.
Failure No. 6 was caused by dielectric breakdown of the insulating oil which may have occurred due to the formation of gas bubbles.
Installa-tion of a COPS oil preservation system, pre circulating, and heating of the oil may have prevented this f ailure. Turning on the pumps and no coolers would provide a heat source for preheating the oil.
l Failure No. 7 was believed to be caused by oil contamination.
It is l
possible that more flushing might have eliminated the foreign material.
l 2-10 i
'l
6.
CONCLUSIONS Based on the available information on the seven main transformer failures at the North Anna Power Station, Units 1 and 2, it is concluded that:
1.
Excessive shipping, handling, and repeated installation were contributors to failures No. 2, 3, 4, and 5 and should be eliminated. Conservative means should be undertaken to assure that the units and bushings are not subjected to vibrational stresses during shipping and handling. Also great care should be exercised to avoid the entrance of moisture and/or contami-nants during installations, and repairs.
2.
Improper storage of the HV bushing was a contributor to failures No. 2 and 3.
It is imperative that manuf acturers supply accurate, up-to-date information and instructions on their product. A storage container should be provided to ensure that the 20* angle of elevation is maintained.
Oil contamination may have been a contributor to f ailure No. 7.
3..
Conservative processing practices must be followed.
Strict control of entry into the units must be enforced to prevent the introduction of foreign material or moisture.
4 Operator error was a contributor to f ailure No.1.
It is imperative that the operating personnel be thoroughly instructed in the operating philosophies and peculiarities of EHV transformers and their auxiliary devices.
5.
Gas bubble evolution may have been a contributor to failure No. 6.
Conservative means should be undertaken when extrene ambient temperature changes are experienced between the tire of processing and the time of energization to ensure gas bubble evolution does not occur.
6.
Our evaluation found no cause related to the Westinghouse trans-former design. The installation of a constant oil preservation system along with the revised procedures for precirculating and heating of the oil will help to prevent the formulation of gas bubbles and subsequent dialectric breakdown. The removal of the " stove pipe" was directed by Westinghouse as a precautionary measure to prevent possible bubble accummulation in the area of the HV lead. Redundant winding temperature indication should be considered if cooling controls are not regularly inspected.
7.
Conservative maintenance practices and procedures should be exercised to assure system reliability. Schedules and surveil-lance frequencies should be based on the relative importance of the system and on past operating experiences. Thorough and accurate maintenance and surveillance records are a must for assuring full, functional opergting status.
2-11
8.
When overvoltages are. experienced, additional surveillance on the connected equipment and the other phases should be implemented.
9.
VEPCO has established a determination of the root causes and/or failure mechanisms with corrective actions or recommendations to prevent the reoccurrence of failures.
REFERENCES 1.
VEPCo memorandum, D. E. Thomas to W. R. Cartwright, undated.
2.
VEPCo memorandum, R. E. Bridges, Jr. to W. R. Cartwright, dated December 11, 1980.
3.
VEPCo memorandum, R. E. Bridges, Jr. to W. R. Cartwright, dated December 29, 1980.
4 VEPCo Report of Task Force to Investigate the Failure of North Anna No. 2 Transformers, Olin R. Compton, dated September 30, 1981.
5.
NRC memorandum (November 29, 1982 meeting), L. B. Engle to R. A. Clark, dated December 17, 1982.
6.
NRC memorandum (December 9,1982 meeting), L. B. Engle to R. A. Clark, dated December 17, 1982.
7.
North Anna Cenerator Step up Transformers Report of Failures in 1980-82 (draf t testimony), later incorporated in a main report, dated January 29, 1982.
8.
Westinghouse letter (D. A. Yannucci) to James C. Selan (LLNL), dated March 29, 1983.
9.
Westinghouse letter (R. A. Wiesemann) to the NRC (L. B. Engle), dated May 6, 1983.
10.
VEPCo letter (W. L. Stewart) to the NRC (H. R. Denton and R. A. Clark),
~
dated August 17, 1983.
11.
VEPCo letter (W. L. Stewart) to the NRC (H. R. Denton and J. R. Miller),
dated December 16, 1983.
I 2-12
. =.
TASK REPORT 3 VIRGINIA ELECTRIC AND POWER COMPANY NORTH' ANNA POWER STATION, UNITS 1 AND 2, MAIN TRANSFORMER AND GENERATOR PROTECTIVE SYSTEMS O
Jean V. Kresser 2
i
].
f i
e t
i 5
3
. - - - ~
p.-
ABSTRACT This report documents the technical evaluation of the adequacy of the protective systems for the generator and transformer. The evalu-ation is based on design experience and good engineering and industry practices.- The evaluation found that the protective systens are adequate
.and operated as designed with the exception of several backup relays.
These relays require optimizing zone coordination.
FOREWORD This report is supplied for the U. S. Nuclear Regulatory Commission, Office of Nuclear Reactor Regulation, Division of Licensing, by Lawrence Livermore National Laboratory.
The U. S. Nuclear Regulatory Commission funded the work under the authorization entitled " Evaluation of Main Transformer Failures at North Anna Power Station," B&R 20 19 10 12 1, FIN A-0442.
6 3-1
TABLE OF CONTENTS Page 1.
INTRODUCTION.
3-1 2.
GENERAL DISCUSSION 3-2 3.
NORTH ANNA UNITS ' PROTECTIVE SYSTEMS.
3-3
3.1 DESCRIPTION
OF THE RELAYING SCHEMES 3-3 3.1.1 GENERATOR PRIMARY PROTECTION.
3-3
~
3.1.2 GENERATOR BACKUP PROTECTION 3-5 3-5 3.1.3 GENERATOR LEADS 3.1.4 ISOLATED PHASE BUS DUCT BACKUP GROUND.
3-6 3.1.5 GENERATOR BREAKER FAILURE (G12) 3-6 3.1.6 MAIN TRANSFORMER.
3-6 4
TRANSFORMER FAILURES.
3-7 5.
CONCLUSIONS 3-7 APPENDIX A.
3-9 REFERENCE DRAWINGS.
3-10 REFERENCES.
3-11 SUPPLEMENTAL REPORT (REPORT BASED ON ADDITIONAL NRC STAFF QUESTIONS) 3-13 O
TABLE OF ILLUSTRATIONS NAPS UNIT 1 TRANSFORMER / GENERATOR PROTECTIVE ZONES 3-4 FIGURE 1 3-111
VIRGINIA ELECTRIC AND POWER COMPANY NORTH ANNA POWER STATION, UNITS 1 AND 2, MAIN TRANSFORMER AND GENERATOR PROTECTIVE SYSTEMS (Docket Nos. 50-338, 50-339)
Jean V. Kresser for Lawrence Livermore National Laboratory 1.
INTRODUCTION This report is the result of the work performed as defined under
" task 4 sub-task 2" of the document titled " Evaluation of Main Transformer Failures at North Anna Power Station," dated February 24, 1983 The work was performed at the request of the Lawrence Livermore National Laboratory.
It consisted of the study and evaluation of the North Anna Power Station, Units 1 and 2, generator, transformers, (main, auxiliary, and reserve), and associated connections protection.
The North Anna Power Station of the Virginia Electric and Power Co.
(VEPCO) has two units (No.1 and No. 2), each consisting of a 1,088,600 kVA, 22 kV generator and three single phase 350 MVA 22/500 kV (Delta-Wye connected) main transformers.
Each generator has an auxiliary (station service) 22/4.16 kV transformer connected directly to its terminals (through an iso phase bus).
In addition, there is a reserve station service transformer, 34.5/4.16 kV, connected to the 500 kV buses through two 34.5 kV buses, each fed through a 500/34.5 kV transformer.
The main transformers are connected to 500 kV buses of the " breaker and a half" configuration. This is a fairly common industry practice. Unit No. I has a generator breaker between its terminals and the LV terminals of its main transformer.
However, Unit No. 2 is " unit connected" in that the generator is directly connected to the LV terminals of its main transformer without an intervening generator breaker. This is also a common industry practice, and is acceptable provided both the generator and the transformer are adequately protected.
3-1
2.
GENERAL DISCUSSION Protective relaying is more of an art than a scie'nce and there are no industry standards. However, a well designed protective system must be based upon certain necessary and/or desired attributes:
(1) Reliability: The protective system must operate at all times that it is supposed to operate for every type of abnormality in the power system.
(2) Security:
The system must not operate falsely when it is not supposed to operate.
(3) Selectivity: The system must isolate the minimum portion of the power system required to clear the abnormality.
(4) Response Time:
The system should operate in a minimum or predic-table amount of time depending upon the particular application and desired protection.
(5) Simplicity: The system should involve a minimum complexity of elements and circuitry that is compatible with the function to be performed.
(6) Economy:
The system should involve a minimum investment that is compatible with the type and degree of protection that are i
justified by good engineering practices.
(7) Protective Zones:
In order to achieve selectivity and reliability, the power system must be divided into protective zones. Adjacent zones must overlap so that no portion of the power system is lef t unprotected.
While there are no industry standards for protective systems, certain practices, based upon operating experience over many years and the development and availability of protective relays, have become generally accepted by electric utilities.
Interchange of ideas, developments, and experience through technical meetings, particularly those of the IEEE Power System Relaying Committee, and others, have been extremely fruitful in developing commonly accepted protective relaying systems.
D
{
l l
l 3-2 l
3.
NORTH ANNA UNITS' PROTECTIVE SYSTEMS The VEPCO North Anna Units have seven protective zones (for Unit No. 1) as follows, and are shown in Figure 1:
Zone 1:
Generator windings Zone 2:
Generator leads and station service transformer HV leads Zone 3:
Main transformer windings Zone 4:
Main transformer HV to load side of 500 kV breakers Zone 5:
Station service transformer Zone 6:
Reserve station service transformer Zone 7:
Reserve station service transformer and 34.5 kV breaker All adjacent zones overlap and there are no " blind" areas. As previously stated, Unit No. 2 does not have a generator breaker.
However, its protective zones are similar, except that Zone 2 does not include a generator breaker and the reserve service transformer is common to both units.
A detailed description of the protective relaying scheme follows.
3.1 DESCRIPTION
OF THE RELAYING SCHEMES Protection provided for the North Anna generators is not exclusively for fault protection, but includes protection for conditions that will harm the prime mover, overheat the generator rotor and stator, or affect the system.
Sone of these conditions require immediate tripping, while other conditions need only to have action taken by the plant operator while the machine is in operation.
The following is a description (relay, type and function) of the relay protection at NAPS.
3.1.1 CENERATOR PRIMARY PROTECTION (1) There are three single phase, high speed, differential type CFD relays.
These relays will detect phase-to phase and three-phase f aults in the stator winding by a comparison of the current entering to the current lea *,ing the winding.
(2) The C0Q relay is a three phase, negative sequence, induction disc overcurrent relay.
It 4111 detect negative sequence current in the generator stator.
(3) The GCP relay consists of a three phase, power directional unit and a single phase, overvoltage timing unit.
This relay detects a motoring condition of the generator by reverse power. Although planned, this protection is not presently installed on Unit No. 2.
3-3
3___'
q'
.c
- w w
.r______________4_.__v_s________________,i Ece.E 4 i
, r;p i
., c;n i
.,4 u
i.
.a i________,
l i
,--------------- i
'8"4w"1F.!
9 I
dL l
C. sa =.,.
.i mae r,
I o
/m l
i l u?!
- --i----["t,------- i L
.3 i
e
@i na 8
I I
8 sia fo1 d
0 l
l
==u i
I WW I
0 gal l
.m.i l
8 vv g
34 gre.:
I I
zer.z a c
l W
t l
g l
i e
i g i i
l I
5
. e e
i e
C.-
g t
4 1
I l
___1_____!
L__ _ '.______Li__J i
l c _..
. _ _ _ _ _ _ _.._m______
20NE 6 e
a
_9s.ee?F av 17' e
,3
.-~
l 9
l l
l l
s_
j_______________________,
,i c..-
l m,.
l== s e
t e
_________2_____________________________,
li 1
1
-")
o 3
c-i
.-c,,.
[]
,I W**"'
f f-_s usu e
..,4 e
l EHO-]h~_.rw--'
I hk s
n pE2 p:
l
,:H9
,o,,,
,aLLpf CL g
9P i
e i
i e
i b"')
.i e
@ l D-<3hI'3*w k __ _ _ _ _ _ _ _ !_ _ _ _ _ _ _ _I__ _ _ _ _
l o
l anar g
a, e
I d"
8 e
ICNE f e
i l
e e
s l
l c;
e e____________4.___,_____.L____________________________i A
g 5-i 9
a.t_ut,optan ga
)
ZCt:E 1 9_;
M _.3 3 3 l
)
l l,%
l 9 ~_ _ _ _ _ _.,' ?2,.
FIGURE 1 - NAPS l' NIT 1 TRANSF0DfER/
n i
'____S
- 7 GENERATOR PROTECTIVE 3
ZONES 3-4 t
1 (4) The KLF-1 provides loss of field protection which would result in thermal damage to the generator and instability to the rest of the system. This relay has the single phase impedance unit and directional unit applied.
(5) The STV is a single phase static overvoltage per hertz relay with a time delay function. This relay protects the generator (and main transformer on Unit No. 2) from overexcitation during off line conditions.
(6) The protection for ground faults in the stator of the generator, the isolated phase bus duct, the low voltage winding of the main
~
transformer, and the high voltage winding of the station service transformers is provided by the CV-8.
The CV-8 is a single phase overvoltage relay with a very low setting.
3.1.2 GENERATOR BACKUP PROTECTION (1) The KD41 is an impedance relay that is time delayed for backup protection. This relay detects phase-to phase, and three phase faults in the generator, the isolated phase bus duct, and 80%
through the main transformers. This relay is supervised by a CFVB voltage balance relay which blocks false trips due to loss of relay potential.
(2) The generator ground fault protection is provided by a C0-8 relay. This relay is an induction disc, inverse time, single-phase, overcurrent relay. This relay detects ground faults in the generator, the isolated phase bus duct, the low voltage winding of the main and the high voltage windings of the station service transformers.
(3) The generator leads out to G12 (G202 and G2T575 on Unit No. 2) during startup are protected by three single phase, instantaneous overcurrent SC relays.
Those relays protect for phase-to phase and three phase faults (and 500 kV ground faults on Unit No. 2) until the generator is connected to the system. These are the only relays that respond correctly to low frequency current during start up.
3.1.3 GENERATOR LEADS (1) Three CFD single phase, high-speed, differential relays make up the protection for the generator leads differential.
These relays detect phase-to phase and three phase faults on the isolated phase bus duct from the generator line side bushing to the low side bushings of the main transformers and to the high side bushin8s of the station service transformers.
3-5
i 3.1.4 ISOLATED PHASE BUS DUCT BACKUP GROUND
'(1) This protection is provided for by a CV-8 single phase, voltage relay. The CV-8 provides ground protection on the i
isolated phase bus duct until the unit is connected to the system (not applicable to Unit No. 2).
u 3.1.5 GENERATOR BREAKER FAILURE (G12)
(1) The SBC relay provides breaker failure protection.
It consists
{
of a fast-reset current detector with two independently adjust-able pickups for phase and ground faults and built-in timers
-t to provide a delay to allow primary protection to operate.
}
This relay provides backup protection to open the surrounding i
breakers if G12 fails to open during fault and nonfault condi-tions (not applicable to Unit No. 2).
3.1.6 MAIN TRANSFORMER l
(1) The transformer differential protection is provided by three, j
high-speed, harmonic restraint, percentage differential HU l
relays. They compare the current entering to the current j
leaving the transformer with a percentage error factor to trip for all types of internal faults.
.f 1
l (2) The Westinghouse transformers are protected by the pressure sensitive SPR relay. The SPR detects internal arcing faults i
by the resulting sudden pressure changes of the nitrogen gas I
blanket that covers the transformers. The Qualitrol RPR relay provides similar protection on the modified Westinghouse trans-i
]
formers with COPS (Constant Oil Preservation System) and on the i
new McGraw transformers. The RPR operates on a sudden change j
in oil pressure that is created by internal arcing faults.
I (3) The reserve station service ground protection is provided by j
sn induction disc, very inverse time, single phase overcurrent
[
j type Co-9 relay. This ptcvides backup protection for any ground faults on the low voltage windings of its transformer and the j
i I.
s j
l i
4 2
3-6 i'
i s
m....
4 TRANSFORMER FAILURES There have been seven transformer failures from November 29, 1980 to December 5,1982--a two year period. All of the failures involved an HV bushing or an UV winding lead.
It appears that all of the failures originated from an HV bushing or HV winding lead and that some of them involve the LV winding.
In all cases, the failures were line to ground. The fault currents are quite substantial. The sum of the measured 3I contributions is in the o
neighborhood of 19 kiloamps at 500 kV.
In most instances two protective zones were involved and, in some cases, high speed backup relays, as well as primary relays, operated. Relay currents were quite high and relay response time was quite short (in the neighborhood of one cycle or less). Since it is impossible to coordinate high speed relays, in some cases, backup relays operated unneces-sarily--but not incorrectly. This is not necessarily undesirable providing that the operations are correctly analyzed and do not cause confusion. A pro-tection engineer should be able to interpret operations correctly. The causes of the failures, or comments thereon, are beyond the scope of this report.
5.
CONCLUSIONS There is no standard generator / transformer protective system to which the VEPC0 North Anna Power Plant, Units 1 and 2, protective system can be com-pared. However, there are manuf acturers recommendations, general electric utility industry practices, and system protection engineers ' technical ability and experience. The subject of generator / transformer protection has been a subject of discussion at numerous IEEE Power System Relaying Comnittees and industry associations ' meetings and conferences.
Based on the evaluation the VEPCO generator / transformer protection at North Anna Unit 1 is as good as can be provided, based upon the requirements listed in Section 2.
North Anna Unit 2 requires special consideration due to the fact that it is a unit connected generator--with no generator breaker.
For a transformer f ault, the gererator remains connected to the transformer.
Therefore, the gene-rator excitation must be removed.
It is presured that this is donei otherwise, the generator would keep feeding the f aults until backup relay operation.
From the f ault recorder data on Failure No. 7 at Unit 1 (general drawi g No. 5
" Fault Recorder Information"), it can be determined that the protective system performed within expected time limits. The fault current lasted for 3 cycles in the 500 kV leads and about 41/2 cycles in the trans-former (3I ).
VEPCo states that the transformer was cleared from the 500 kV n
nystem in 3 cycles and f r'om the 22 kV system in 7 cycles (Ref.15 Appendix A).
These times consist of relay tien plus trip and lockout relay time plus circuit breaker interrupting tire.
The record f rom the fault recorder is for Unit No. 1 3-7
and therefore, the generator current was interrupted by the generator breaker.
The fault duration is considered to be reasonable. The transformer HV f ault duration of about 41/2 cycles is somewhat longer than would have been expected (the record is of the sum of the 3-500 kV phase currents or 31 ).
However, it o
is not considered to be excessive, and is within nominal limits.
In addition to the generator and transformer protection and the other protective zones listed in Section 3, the protection of the 500/34.3 kV transformers, the 34.5 kV buses, and the 4.16 kV station service buses were studied and evaluated.
It was determined that the protective system for these elements was satisf actory and adequate.
The conclusion is that part of the VEPC0 North Anna Power Station protective system which was studied, analyzed, and evaluated conforms to good engineering practice, and is not deemed to warrant justifiable modifications in any way.
J e
i I
i i
I l
1 i
ee i
I I
(
i i
3-8
_L - __-.- _--- - - -,_ - ~_ _.
APPENDIX A Documents used in Evaluation:
1.
" Evaluation of Main Transf ormer Failure at North Anna Power Station,"
dated February 24, 1982, NRC Document A-0442.
2.
Response to NRC information request, dated January 7, 1983. VEPC0 docunent prepared by System Protection Department.
3.
" North Anna Cenerators Step up Transformers: Report of Failures in 1980-1982," undated, presumed to have been prepared by VEPCO.
4.
.'EPC0 memorandum:
North Anna #2 Main Cenerator Transformer "A" Phase Failure and Resultant Damages Incurred, undated, D. E. Thorns.
5.
Update on Transformer Fire at North Anna Power Station, July 3,1981.
NRC memorandum by Leon B. Engle, July 6,1981.
6.
Trip Report to North Anna Failure of B Phase Main Transformer and Subsequent Fire in the Transformer Area.
NRC memorandum, Matthew Chiramal, July 14, 1981.
7.
North Anna Power Station, Units 1 and 2 - Observations on July 3, 1981 Unit 2 Main Transformer Fire.
NRC memorandum, Leon B. Engle, July 22, 1981.
8.
Unit 2 Transformer Failure and Fire on July 3,1981.
Virginia Electric and Power Company, North Anna Power Station, NRC Report.
9.
Safety Consideration Resulting f rom the North Anna No. 2 Transformer Fire.
NRC memorandum, R. L. Baer, December 3,1981.
10.
Trip Report Regarding the North Anna Unit 2 Output Transformer Failures.
NRC memorandum, W. C. Marsh, September 4, 1981.
11.
Report of the Task Force to Investigate the Failures of North Anna No. 2 Trans f o rme rs.
Orin R. Corpton, Task Force Chairman.
12.
Site Visit to obtain Briefing f rom VEPCO on Latest Rocurring Transformer Fault at the North Anna Power Station (NAPS).
NRC memorandum, Leon B. Engle, December 17, 1982.
13.
Trip Report NRC Meeting (Region II Of fice) on November 29, 1982, to review VEPCO's Investigation Into the Causes of Repeated Main Transformer Faults at the North Anna Power Station, Units 1 and 2 (NA 1/2).
NRC memorandum Leon B. Engle, December 17, 1982.
14 Task Force Review of the Recurring Transformer Fault at the North Anna Power Station.
NRC report, Thomas A. Ippolito, undated.
15.
VEPCo letter (W. L. Stewart) to (H. R. Denton), dated February 10, 1983, 3-9
REFERENCE DRAWINGS i
GENERAL Figure No. 1 Simplified One-Line (Unit #1)
Figure No. 2 Simplified Fault Recorder Connections Figure No. 3 Simplified Three-Line (Unit #1)
Figure No. 4 Simplified One-Line (Unit #1, Unit #2 and Switchyard Figure No. 5 Fault Recorder Information l
VEPCO DRAWINGS 1
Switching Station One-Line l
2 Switching Station One-Line 67' Switching Station Elementary C.B. G102 48 Switching Station Elementary C.B. G102 49 Switching Station Elementary C.B. G1T568 50 Switching Station Elementary C.B. G1T568 i
l 129 Switching Station Elementary C.B. G202 l
130 Switching Station Elementary C.B. G202 131 Switching Station Elementary C.B. G202 132 Switching Station Elementary C.B. G2T575 133 Switching Station Elementary C.B. G2T575 STONE AND WEBSTER DRAWINGS (Unit #1) 11715-FE-1A Main One-Line Diagram 11714-FE-21A A.C. Elementary (Gen. and Mn. Transformer) 11715-FE-21B A.C. Elementary (Station Service Transformers) 11715-FE-21D D.C. Elementary (Gen. #1 and Transformer Protection) 11715-FE-21G D.C. Elementary (4160-Bus LA Bkr.15A1,15A2, and Bus 1B Bkr. 1581, 1582) 11715-FE-21H D.C. Elementary (4160-Bus 1C Bkr.15C1,15C2, and Bus 1G Bkr.15G1, Bus 1H - Bkr.1541 11715-ESK-8AD Elementary Diagram (22 kV Gen. Circuit Breaker G12) 11715-ESK-8AE Elementary Diagram (22 kV Gen. Circuit Breaker G12) 13050-FE-21Y D.C. Elementary G12 Breaker Failure STONE AND WEBSTER DRAWINGS (Unic #2) 12050-FE-1A Main One-Line Diagram 13050-FE-21A A.C. Elementary (Gen. and Main Transformer) l 12050-FE-21B A.C. Elementary (Station Service Transformer) 18050-FE-21D D.C. Elementary (Gen. #2 and Transformer Protection) 13050-FE-21G D.C. Elementary (4160.us 2A Bkr. 25A1, 25A2 and Bus 2B Bkr. 25B1, 2582) 12050-FE-21H D.C. Elementary (4160-Bus 2C Bkr. 25C1, 25C2, and Bus 2G Skr. 25G1, Bus 2H Bkr. 25H1).
3-10 t
REFERENCES The following references are included for general interest only.
They serve as guidelines, because any particular application may entail some special considerations not considered in the references. The protection engineer's judgement should prevail.
9 1.
Applied Protective Relaying, Chapters 6 and 8.
A book published by Westinghouse Electric Corporation.
2.
The Art and Science of Protective Relaying, Chapters 10 and 11, C. Russell Mason. John Wiley and Sons, Inc.
3.
Protective Relays Application Guide. The English Electric Corporation, Ltd.
4.
The Principles and Application of Relays and Relaying Schemes for the Protection of Electrical Apparatus and Systems, Jean V. Kresser.
Privately published by Westinghouse Electric Corporation.
5.
Technical Lectures on Protective Relays and System Protection by Jean V. Kresser.
Unpublished.
9 3-11
T
- ~ ~ ~ ' '
~ ~~
^ ^ ^ ^ ' ' ~ ~ ~ ' ~
~ ^ ^ ^ ^ ^ ~ ^ ^ ~ ~
^
^
1 I
iy i.
4 s.
5
?
6 e
i 1
4 1
1
!'f i
d 4
5 4
6 4
a t
4 1
i 1
1 3
4 i
,i i
i i
4 4,
h 1
L i
1, b
1 l
'r i
l a
t a
3-12
~....
.~...
SUPPLEMENTAL REPORT I-t I
VIRGINIA ELECTRIC AND POWER COMPANY l
NORTH ANNA POWER STATION PROTECTIVE SYSTEM Jean V. Kresser r
September 27, 1983 e
I
,4 l'
O i
3-13 l
l
I.
INTRODUCTION This report supplements the report dated May 16, 1983.
It has been prepared to comment, discuss, and answer comments and questions brought forth by the original report. The numbering under Section II of this report conforms to that of the attachment. The subsection headings refer to the matters involved.
II.
COMMENTS AND DISCUSSION Complete documentation for all seven faults was not made available to the writer of this report. The following pertinent data were available for the indicated faults.
Fault location and type:
Faults No. 1 through No. 6.
Fault currents and distribution:
Faults No. 1 through No. 4.
Relay operations:
Faults No. 1 through No. 4.
Oscillograph records:
Faults No. 4 and No. 7.
1.
Analysis and Evaluation of Relay Performance Fault No. 1 (11/29/80):
Flashover from A phase HV winding lead to LV winding.
Fault current from summation of measured currents in circuits feeding the fault:
18,893 amps.
Fault duration (on HV side):
3.5 cycles.
l Relay operations:
See attached sheets.
l Contact between the HV to the LV in the transformer, through an arc, imposed 500 kV (289 kV to ground) on the 22 kV ISO phase bus causing flashover to ground on all three phases. Thus, the fault affected l
the transformer differential relays and the 22 kV generator leads' differential relays. On the HV side, the fault was on A phase.
A and B phase relays operated due to the fact that since the trans-former connection is delta-wye, for an A phase fault there is fault current in A and B phases in the LV leads. All three generator leads differential relays operated, since all three phases were involved in i
the fault.
The operation of the generator C phase differential relay was incorrect. However, as noted for this f ault, it was due to damage l
to the C phase current transformer in its circuit during the fault.
The operation of the 500 kV leads' backup ground relay (67N) was incorrect since it should not have operated if properly coordinated with the primary relays (87/PW2). However, this incorrect operation was not detrimental.
It is not known whether this relay has an instantaneous trip element.
If it has, it should be disabled in order to allow for proper coordination.
3-14
Fault No. 2 (6/19/81):
Failure of C phase HV bushing.
Measured currents summation:
20,013 amps.
Fault duration:
3.5 cycles (on HV side).
Relay operations:
See attached sheets.
The fault (HV bushing f ailure) evidently involved both the C phase transformer differential zone and the C phase 500 kV leads ' dif fer-ential zone, causing A phase and C phase differential relays of both zones to operate correctly. Again, the backup 67N relay operated incorrectly due to lack of coordination.
Fault No. 3 (7/3/81):
Measured currents summation:
20,292 amps.
Fault duration:
3.5 cycles (HV side).
Relay operations:
See attached sheets.
In this case also, both transformer's and 500 kV leads ' dif ferential zones were involved and the 67N backup ground relay operated incorrectly.
Fault No. 4 (7/25/81):
C phase HV winding to LV winding.
Fault currents summation:
13,974 amps.
Fault duration:
Not available.
Relay operations:
See attached sheets.
The transformer was energized from the 500 kV system only. The A phase and C phase transformer differential relays operated correctly.
The i
backup distances ' relay (21) operation cannot be explained f rom the insufficient available data. This relay is set 80% into the trans-former, and is time delayed.
Its operation and the fact that A phase and C phase relays operated indicate that the generator must have been connected to the transformer, contrary to a report statement that it was not.
A probable conclusion is that since Unit No. 2 does not have a generator breaker, it continued to feed current into the fault until its terminal voltage decreased sufficiently so as to no longer sustain the current, and that this length of time was no longer than the time delay of relay 21.
For Faults No. 5 through No. 7, a request for relay operations data was made but these data were not made available.
Since these faults were similar to at least one of Fault No. I through No. 3, it is to be expected that the relay operations were the same for the sane kind of faults.
The f ault current magnitude is an inverse function of the " driving point impedances"--zero, positive, and negative for line-to ground faults.
These, in turn, are a function of the impedances of all cir-cuits feeding into the fault. No data are available on these imped-ances, but using typical impedances for generator and transformer, the measured f ault currents are within an expected calculated range.
i 3-15
2.
Evaluation of the Protective System Item No. 1 of'this section discusses the relay performance for the four f aults for-which' relay operations data were available.
In each case the primary relays operated correctly and the f ault clearing time from the HV system of 3.5 cycles (58.3 milliseconds) was well within the time range to be expected.for the type of protection involved. Assuming a two cycle breaker time and a half cycle trip and lockout relay time, the primary relays operated in one cycle--
which is as good as can be expected.
No data are available for f ault clearing from the 22 kV side. However, as previously stated.
for the Unit No. 2 faults, the fault duration time is much longer since there is not a generator breaker.
Backup relays, (67N for the first three faults, and 21 for the fourth f ault) operated when they should not have--being backup relays. How-ever, this is not detrimental, although it may cause some confusion.
Their operation is attributed to lack of coordination for 67N and too short a time delay for 21.
Current transformer performance can be evaluated f rom oscillograph records. Two of these were available.
One has traces of Unit No. 2 500 kV currents.
These traces show that the 500 kV current transformers output and performance were very good.
The other record, for Fault No. 7, has traces 500 kV lead currente and transformer 3Io (fault contribution) current--which is the sum of the 3-500 kV line currents.
In all cases the current transformers output and performance are very good.
3.
Reason for Only One Generator Failure The equipment which is damaged by a fault is determined by the f ault location, current magnitude, and time to clearance.
Faults Nos. 2, 3, and 5 were HV bushing failures which were cleared f rom the HV side in 58.3 milliseconds. The generator contributed some current to the fault, but its magnitude and duration were far below the thermal capa-bility of the generator.
Faults Nos. 1, 4, 6, and 7 were from an HV lead to the LV winding, imposing through the are 500 kV to the 22 kV system.
For Fault No. 1, all three phases of the 150 phase bus flashed over to ground. This shielded the generator windings from the high voltage.
For Fault No. 4, the transformer was energized from the HV system only--the generator not being connected. There are no detailed data for Fault Nos. 6 and 7; these were on Unit No. I which has a generator breaker.
It is evident that, for Fault No. 7, the generator failed due to very high overvoltage before its breaker could separate it from the system.
4 Generator Protection l
The generators are protected with high speed relays, in conformance with industry practice. The oscillograph for Fault No. 7, where the generator was also f aulted, shows that the 500 kV lead currents were interrupted in three cycles (50 milliseconds). This is very good I
3-16 i
l i
+
performance and can only be~ slightly improved by using solid-state relays.
Furthermore, imposing 500 kV on a 22 kV generator is bound to cause extensive damage in an extremely short time.
5.
Protection Against High Voltage to Low Voltage Faults The North Anna transformer protection for HV to LV faults consists of sudden pressure relays and differential relays.
The sudden pressure relay operates when there is an are under the oil. The differential relay operates when there is a fault within its protective zone.
Either or both protection relays will operate for an HV to LV fault.
Both did operate for Faults No. I and 4.
No data are available for Faults No. 6 and 7.
The sudden pressure relay also operates for Faults No. 2 and 3 which were HV bushing failures, and which presum-t ably caused arcing under oil.
The sudden pressure relay operated correctly in all four faults for which relay operations data have been made available.
III. CONCLUSIONS 1.
Primary protection performance was correct and operated at high speed.
2.
Some backup protection relays (67N and 21) operated before they should have. This requires optimizing zone coordination.
3.
Current transformer performance was very good.
4.
The generator protection is very good, and operated in 50 milliseconds.
5.
Protection against high voltage to low voltage faults is very good, and operated at high speed (58.3 to 70.8 milliseconds).
r 3-17
RELAY OPERATIONS FAULT NO. 1:
11/29/80 Main Transformer Protection Transformer Differential A9 Unit 2 Main Transformer 87-T Transformer Dif ferential B9 Unit 2 Main Transformer 87--T Transformer Differential Lockout Unit 2 Main Transformer 86-T Sudden Pressure Relay A9 22 KV Generator Leads Protection Unit 2 - Generator Leads Differentiel A9 87-GL Unit 2 - Generator Leads Differential B9 87-GL Unit 2 - Generator Leads Differential C9 87-GL Unit 2 - Generator Leads Differential Lockout 86-GL Generator Protection t
5 Unit 2 - Generator Differential C9 87-G Unit 2 - Generator Differential Lockout 86-0 Note: The C9 current transformer in the generator differential circuit was damaged during the fault and accounts for this operation. There was no fault damage inside the generator protection zone.
500 KV Leads Backup Protection Generator 2 500 kV Leads Backup Ground 67N Generator 2 500 kV Leads Backup Lockout 86 L3-BU i
, FAULT No. 2:
6/9/81 i
Main Transformer Protection 1
l Transformer Differential A9 Unit 2 Main Transformer 87-T l
Transformer Dif ferential C9 Unit 2 Main Transformer 87-T Transformer Dif ferential Lockout Unit 2 Main Transformer 86-T j
Sudden Pressure Relay C9 500 KV Leads Protection Generator 2 500 kV Leads Differential 87-PW2A Cenerator 2 500 kV Leads Differential Lockout 86-PW2A j-Generator 2 500 kV Leads Backup Cround 67-N Generator 2 500 kV Leads Backup Lockout 86 L3-BU 3-18
r=
FAULT NO. 3:
7/3/81 Main Transformer Protection Transformer Differential B9 Unit 2 Main Transformer 87-T Transformer Differential C9 Unit 2 Main Transformer 87-T Transformer Differential Lockout Unit 2 Main Transformer 86-T Sudden Pressure Relay B9 500 KV Leads Protection Generator 2 500 kV Leads Dif ferential 87-PW2A Generator 2 500 kV Leads Differential Lockout 86-PW2A Generator 2 500 kV Leads Backup Ground 67-N Generator 2 500 kV. Leads Backup Lockout 86 L3-BU FAULT No. 4:
7/25/81 Main Transformer Protection Transformer Differential A9 Unit 2 Main Transformer 87-T Transformer Differential C9 Unit 2 Main Transformer 87-T Transformer Differential Lockout Unit 2 Main Transformer 86-T Sudden Pressure Relay C9 500 KV Leads Backup Protection Generator 2 500 kV Leads Phase Backup 21 Generator 2 500 kV Leads Backup Lockout 86 L3-BU O
l 6
3-19
t L.
I-l l
TASK REPORT 4 l
l l
EVALUATION OF THE LICENSEE'S PROPOSED MODIFICATIONS AT THE NORTH ANNA POWER STATION, UNITS 1 AND 2 Kerry J. Dalton James C. Selan E
e F
i L
I i
I i
v e
e h
I 4
ABSTRACT This task report documents the evaluation of the licensce's proposed modifications which stem from their analyses of the seven main transformer failures at the North Anna Power Station, Units 1 and 2.
The evaluation finds that most of the proposed modifications are of significant value to the upgrading of the generator system design. Several procedural changes should be incorporated to ensure that conservative means are undertaken to reduce failures.
Several design modifications should be implemented in the areas of fire suppression, transformer bay segregation, and drainage.
FORWARD l
This report is supplied f.r the U. S. Nuclear Re8ulatory Commission, Office of Nuclear Reactor Regulation, Division of Licensing, by Lawrence Livermore National Laboratory.
The U. S. Nuclear Regulatory Commission funded the work under the authorization entitled " Evaluation of Main Transformer Failures at North Anna Power Station," B&R 20 19 10 12 1, FIN A-0442.
l 4-1 f
TABLE OF CONTENTS Page 1.
INTRODUCTION.
4-1 I
i 2.
EVALUATION OF PROPOSED MODIFICATIONS.
4-1 2.1 TRANSFORMER REPLACEMENT.
4-1 4-2 2.2 SURGE ARRESTERS.
l 2.3 GENERATOR GROUNDING.
4-2
(
2.4 ISOLATED PHASE BUS SEPARATION 4-3 i
2.5 INSTALLATION OF MONITORING RECORDERS 4-3
~
2.6 MAIN GENERATOR BREAKER.
4-3 2.7 REFUELING OUTAGE INSPECTIONS 4-4 2.8 FIRE PROTECTION MODIFICATIONS 4-4 2.9 PROCEDURAL CHANGES.
4-5 3.
CONCLUSION 4-5 REFERENCES 4-6
[
l l
l l
4-111 l
t i
EVALUATION OF THE LICENSEE'S PROPOSED MODIFICATIONS AT THE NORTH ANNA POWER STATION,' UNITS 1 AND 2 (Docket Nos. 50-338, 50-339)
Kerry J. Dalton James C. Selan Lawrence Livermore National Laboratory, Nevada 1.
INTRODUCTION Virginia Electric and Power Company (VEPCo), the licensee, perforced -
numerous analyses and evaluations on the seven main transformer failures at the North Anna Power Station, Units 1 and 2.
Based on these analyses and evaluations, the licensee has made, or is planning to cake, modifications to various systems associated with the main transformers.
The purpose of this task report is to evaluate the licensee 's proposed modifications to determine the value of these modifications for reducing the frequency of main transformer f ailures at the station or to eliminate them.
2.
EVALUATION OF THE PROPOSED MODIFICATIONS This section identifies the licensee's modifications [Ref.1] that have been made or are planned to be made based on the evaluations of the MTFs.
Following each identified modification is an evaluation on the value of that modification.
2.1 TRANSFORMER REPLACEMENT VEPCo has replaced the Westinghouse manufactured main transformers at Units 1 and 2 with McGraw Edison units and General Electric units, respectively.
As stated in the " Root Cause" evaluation (Task Report 2), we found no cause related to the Westinghouse transformer design used at NAPS. Our evaluation found that there were several factors which contributed to the seven failures.
Unless these f actors are minindzed or eliminated, the potential for 4-1
experiencing similar type failures with these new transformers still exists.
In general, these f actors are:
(1)
Eliminate excessive shipping, handling, and repeated installations.
(2)
Exercise conservative maintenance procedures and practices and employ accurate and up-to-date record keeping.
(3)
Minimize overexcitations.
-(4)
Ensure manfacturers instruction documentation is updated periodically and explicitly followed.
9 (5)
Ensure operating personnel are thoroughly knowledgeable in the operating philosophies of EHV transformers and their auxiliary devices.
2.2 SURGE ARRESTERS VEPCo plans to install 24 kV surge arresters on the 22-kV isolated phase buses f rom the main generator at both units. This modification is being engineered by VEPCo.
Surge arresters are protective devices for limiting surge voltages on equip nent by discharging or bypassing the surge current.
For types of trans-former f aults experienced (HV to LV windings), elevation of the 22 kV system, the planned surge arresters will help prevent potentially damaging voltage reflections on the buses and onto the generator.
2.3 GENERATOR GROUNDING VEPCo has completed the installation of several modifications with respect to the generator neutral grounding cable, transformer basic insulation level (BIL), and generator frame grounding at both units.
The generator neutral grounding cable (original size was 4/0) was replaced with a shielded cable with stress cone terminations.
The purpose of a shielded cable is to distribute the dielectric stresses evenly along the entire length of the cable. The stress cone terminates L
the shield at the cable end to ground. This evens out the stresses at critical.
(
points during transient events.
Since the neutral is grounded via a transformer /
l resistor, the neutral ground cable is subject to higher potential elevations, l
thus increasing the dielectric stresses on the cable.
l The analysis indicated that the generator neutral grounding transformer l
BlL rating should be at leas.t -125 kV.
The original BIL rating of the transformer was not documented; however, it is unlikely.that it was lower than 95 kV at the voltage level.
The BIL rating is indicative of the insulation characteristics 4-2
so that the insulation will not break down on voltage surges of high magnitude and short duration. This 125 kV BIL rating is standard for this system voltage and will provide the insulation coordination in the 22 kV system.
The generator frame was grounded using one 1000-MCM copper conductor at two separate points. The generator frame was then grounded to the neutest enclosure using one 2000-MCM copper conductor. All the ground conductors were terminated with compression lugs.
This upgrading of the grounds indicates that the original f ault current analyses probably resulted in lower f ault currents than were experienced in the f ailures, thus requiring the increase in current carrying capacity.
4 2.4 ISOLATED PHASE BUS SEPARATION l
VEPCo's analysis indicates that the separation between the isolated phase bus flexible elbows and the bus duct should be increased to the manu-facturer's recommended value.
These modifications have been completed on both units.
This new value as well as the present separation distance was not docu ment ed.
It is not clear why the manufacturer's recommended bus duct clearances were not followed in construction. These clearances are not only set for voltage jump, but for heating and losses produced in the bus ducts.
2.5 INSTALLATION OF MONITORING RECORDERS The installation of continuous monitoring recorders to record the electrical parameters of the generator and on the isolated phase buses at both units is being engineered by VEPCo.
Recorders are important instruments and useful tools in the electrical power field. They provide concurrent and historic records of various functions 1
and operating conditions af fecting the performance of a power system. The instal-lation of monitoring equipment is certainly beneficial in having a real time status for an operating system.
2.6 MAIN GENERATOR BREAKER A main generator breaker was installed at Unit 1 after the sixth i
failure (first failure at Unit 1).
VEPCo is currently evaluating whether to install a generator breaker at Unit 2.
A main generator breaker like any circuit breaker is a protective device.
It must be coordinated with the other protective zones. Again, a definite time is involved before tripping occurs. The experience with the seventh f ailure was such that the generator was damaged due to the magnitude 4-3
of the fault currents before the generator breaker tripped and isolated the generator. The damage to equipment before breaker tripping is dependent on fault location, magnitude, and the time to clear. Therefore, the decision regarding whether to install a main generator breaker or not is both an economic f actor and a f actor of its importance to the overall system protec-tion. Thus, all f actors surrounding the root causes and their corrective means and system upgrading as proposed, with proper protective coordination, indicate that the installation of a main generator unit breaker may not be warranted at this time.
2.7 REFUELING OUTAGE INSPECTIONS VEPCo is considering that during refueling outages an inspection and cleaning of the isolated phase buses, ducts, and insulators be performed.
A properly instituted preventive maintenance program will contribute to the increased reliability and integrity of components and systeus.
2.8 FIRE PROTECTION MODIFICATIONS VEPCo has completed a study of possible fire protection modifications which are being considered for installation in 1984 These modifications include:
(1)
Increase the height of the common dike which surrounds the trans-former bays.
(2)
Add firewalls at the end of each transformer bay.
(3)
Increase drainage flow rate capacity.
The height of the common dike surrounding the transformer bay area should be adequate to contain the volume of oil plus the volume of water dis-charge from the deluge system. The drainage system must also be adequately sized to accommodate this volume for discharge. Our evaluation finds that each transformer bay should be segregated so as not to allow water / oil flow from one bay to the next.
This means that individual reservoirs and drainage systems must be utilized for each bay.
We recommend that the addition of a firewall to enclose the transformer area should not be made. This would not permit proper cooling of the transformers and it will prohibit easy access.
Instead of the firewall, reconstruct the diked area to form individual separate diked areas for each transformer. This would require that the drainage system he redesigned. These factors are also addressed in Task Report 6, " Generic Aspects."
4-4
1 2.9 PROCEDURAL CHANGES VEPCo's evaluation has determined that there are no procedural changes required.
Based on our evaluations we find that several procedural changes should be made as applied to the main transformers.
Specifically, they include:
(1)
Conservative maintenance procedures to ensure that during trans-former work foreign contaminates are not introduced into the transformer tank.
(2)
Additional emphasis on regular scheduled inspections with up-to-date record keeping on transformer and auxiliary devices operating status.
(3)
Additional emphasis on regular scheduled reviews for adequacy of fire fighting plans, personnel training and fire fighting equipment.
(4)
Stricter control on shipping, handling, and proper storage of components.
3.
CONCLUSION Based on the available information on the seven main transformer failures at the North Anna Power Station, Unit 1 and 2, with respect to the licensee 's proposed modifications, it is concluded that:
(1)
Shipping, handling, repeated installation, and overexcitation must be cdnimized.
(2)
Conservative maintenance procedures with thoroughly knowledgeable personnel must be followed.
(3)
Installing surge arresters on the isolated phase buses will help to prevent the exposure to overvoltages.
(4)
Replacing the generator neutral ground with a shielded cable, increasing the grounding transformer BIL rating, and increasing the generator frame ground conductor size will improve the system's capability for protecting against the potential f ault-currents analyzed.
(5)
Following manufacturer's recommendations on any component or-system should be implemented as a minimum.
(6)
Installing monitoring recorders can only be beneficial to the functional status of an operating system.
4-5 b
~
(7)
It cannot be definitely ascertained that the installation of a main generator circuit breaker would prevent generator damage. The important f actor is trip time.
(8)
A properly instituted preventive maintenance program can enhance the integrity and reliability of components and sys t e ms.
(9)
Modifications should be made to prevent the flow of water /
oil mixture from one transformer bay to another and to increase the holding and drainage capacity of each individual bay.
(10)
Several procedural changes should be made to ensure that; a) good conservative maintenance procedures are practLced; b) regularly scheduled inspections are made with accurate and up-to-date record keeping; c) shipping, handling, and repeated installations of equipment are minimized; d) components and equipnent are properly stored when not in use.
REFERENCES 1.
VEPCo letter (W. L. Stewart) to the NRC (H. R. Denton and R. A. Clark),
dated August 17, 1983.
b J
4-6
TASK REPORT 5 5
REPORT ON THE SURVEY CONDUCTED OF THE ELECTRIC POWER INDUSTRY 'S EXPERIENCES WITH
' EXTRA HIGH VOLTAGE TRANSFORMER FAILURES AND COMPARISON WITH THE MAIN TRANSFORMER FAILURES AT THE NORTH ANNA POWER STATION, UNITS 1 AND 2 James C. Selan l
=
4 5
ABSTRACT This report documents the results of a survey condacted in the electrical power industry on experiences with extra high voltage transformer failures. The survey was made in order to provide a statistical base f rom which a comparison with the North Anna Power Station transformers could be
~
made. The statistical base was very limited, as the f ailure data could only be compiled from three sources.
The comparison results showed that only limited correlations could be made with the available data, specifically in the areas of number of f ailures, type of failures, consequences of the f ailures, and time to the first f ailure.
FOREWORD
-This report is supplied for the U. S. Nuclear Regulatory Commission, Office of Nuclear Reactor Regulation, Division of Licensing, by Lawrence Livermore National Laboratory.
The U. S. Nuclear Regulatory Commission funded the work under the authorization entitled " Evaluation of Main Transformer Failures at North Anna Power Station," B&R 20 19 10 12 1, FIN A-0442.
e 5-1
TABLE OF CONTENTS Page 1.
INTRODUCTION.
1-1 2.
SURVEY 1-1 2.1 SURVEY FIELD.
1-1 2.2 SURVEY RESULTS 1-3 2.2.1 EDISON ELECTRIC INSTITUTE FAILURE STATISTICS.
1-3 2.2.2 DOBLE ENGINEERING COMPANY FAILURE STATISTICS.
1-8 2.2.3 IEEE NUCLEAR DATA RELIABILITY MANUAL.
1-9 3.
SURVEY COMPARISONS
. 1-12 3.1 NAPS MAIN TRANSFORMER FAILURE RATES / MODES
. 1-12 3.2 NAPS MAIN TRANSFORMER FAILURES VERSUS EEI DATA.
. 1-12 3.3 NAPS MAIN TRANSFORMER FAILURES VERSUS DOBLE ENGINEERING DATA
. 1-13 3.4 NAPS MAIN TRANSFORMER FAILURES VERSUS IEEE RELIABILITY DATA. 1-14 4
SUMMARY
. 1-15 5.
CONCLUSION
. 1-16 i
REFERENCES
. 1-16 ILLUSTRATIONS TABLE 1 - Similar NAPS Transformer Classification Failure Statistics 1-4 TABLE 2 - 1979 Failure Statistics.
1-5 TABLE 3 - 1980 Failure Statistics.
1-6 TABLE 4 - 1981 Failure Statistics.
1-7 TABLE 5 - Transformer Failure Modes
. 1-10 TABLE 6 - Generator - Main Power Transformer Failure Rates 1-11 5-111
REPORT ON.THE SURVEY CONDUCTED OF THE ELECTRIC POWER INDUSTRY'S EXPERIENCES WITH EXTRA HIGH VOLTAGE :
TRANSFORMER FAILURES AND COMPARISON WITH THE MAIN. TRANSFORMER FAILURES AT_THE NORTH ANNA POWER STATION, UNITS 1 AND 2
-(Docket Nos. 50-338, 50-339)
James C. Selan Lawrence Livermore National Laboratory, Nevada 1.
INTRODUCTION The North Anna Power Station (NAPS), Units 1 and 2, has experienced seven main transformer failures (MTFs) in the time period from November 29, 1980 to December 5, 1982 (26 months).. Surrounding each failure, a record of known
. facts which encompassed the transformer's operating history was established.
From these known facts, the cause of failure was.lto be determined. A detailed evaluation of each failure will be addressed in'a separate Lawrence Livermore National Labortory report. However, the purpose of this report is to present a statistical base on transformer failures of similar type used at NAPS in which the number of failures (rates), type of failure, over what period of time, causes and consequences and other important aspects can be correlated.
The statistical data base was to be achieved by conducting a survey of the experience with extra high voltage (EHV) transformers in the electric power industry. The statistical base was to provide as much pertinent infor-mation as possible so that a meaningful comparison could be made.
2.
SURVEY 2.1 SURVEY FIELD This section presents the organizations associated with the electric
)
. power industry that were contacted for any statistical information on transformer failures.
Specifically, those contacted were:
Electric Po' er Research Institute (EPRI)
(1) w (2) National Electrical Manufacturers Association (NEMA) 1 5-1 u
J
t (3) Institute of Electrical and Electronic Engineers (IEEE)
(4) Nevada Power Company I
(5) Electrical World (an electrical utility magazine published by McGraw-Hill)
(6) Edison Electric Institute (EEI)
(7) Westinghouse *
(8) General Electric *
(9) Federal Energy Regulatory Commission (FERC)
(10) North American Electric Reliability Council (NERC)
(11) Doble Engineering Company **
(12) Bonneville Power Authority (BPA)
(13) Oak Ridge National Laboratory (ORNL)
(14)
U.S. Nuclear Regulatory Commission (USNRC)
From the above list of contacts, only three (Contact No. 3, 6, and 11) were able to supply limited statistical information (see Section 2.2).
Edison Electric Institute was referred to several times as the only known source for the requested information. The Oak Ridge National Laboratory supplied a listinge of 184 License Event Report (LER) abstracts as a result of all types of transformer failures in the nuclear power generating industry. Only two of the seven NAPS MTFs were listed in these LER abstracts, and they contained no other entries involving similar type transformers.-
1 i
i
~
l Denotes organization which has detailed failure information that is i
classified " proprietary" and was, therefore, not available upon request.
l I
Denotes organization which has detailed failure information that is j
available only to its clients. Virginia Electric Power Company (VEPCo) l is a Doble client and was requested to provide this information.
5-2
2.2.
SURVEY RESULTS The classification of the main. transformers utilized at NAPS during the period of failures is as follows:
(1) Transformer type power (2) Application generator step up (G.S.U.)
(3) Phases single (4) Type construction shell form 4
(5) Tap changer none (6) Size - 330 MVA (7) Voltage - 22 kV (low side) to 500 kV (high side) 2.2.1 EDISON ELECTRIC INSTITUTE.(EEI) FAILURE STATISTICS The transformer failure statistics (type similar to NAPS) shown in Table 1 are from three EEI Reports [Ref. 1] for the years 1979 through 1981.
The information in these reports was collected by the Electrical System and Equipnent Committee of EEI.
The reports cover transforcers 2.5 MVA and larger from all manufacturers. The 1979 and 1980 reports cover transformers that were installed in the last 10 years only, while the 1981 report covers the last 11 years. The 1979 report contained a first-tine installation data base on 10 years with the succeeding reports correcting for the number of retirements and additions which occurred during the reporting year.
In addition to the transformer classifications listed in Table 1, the reports contained a detailed itemized listing on the troubles reported and is shown in Tables 2, 3, and 4 for the years reported.
These tables are repro-ductions of those in Reference 1, and contain several numerical changes in the
" ANNUAL TROUBLE RATE" and " AVERAGE SERVICE LIFE" rows due to apparent mathemat--
ical errors in the original data submitted. These changes are indicated with an asterisk (*).
The formulas used for these calculations were defined in the report.
It should be noted that other statistical information may be avail-able for earlier years; however, due to changes in data compiling, the most detailed and current failure information available was provided for the years stated above.
5-3
TABLE 1 - Similar NAPS Transformer Classification Failure Statistics 1979 1980 1981 ITEM NUMBER PERCENT NUMBER PERCENT NUMBER PERCENT TOTAL REPORTED TROUBLES 103 100 82 100 77 100 TRANSFORMER fYPE:
POWER 80 77.7 71 86.6 65 84.4 TYPE CONSTRUCTION:
SiiELL 7
6.8 9
11.0 6
7.8 Y.
s TAP CllANGES N/A N/A N/A N/A N/A N/A TYPE APPLICATION:
GEN. STEP-UP 7
6.8 6
7.3 7
9.1 MVA SIZE:
100 TO 499 14 13.6 9
11.0 14 18.2 VOLTAGE CLASS KV:
451-650 4
3.9 2
2.4 2
2.6 PilASES
SINGLE 2
1.9 2
2.4 3
3.9 i
t 2
e
'n e
se em 95 men am
- e as se es e*eo e
e e
WB 9
es e oe Se en op se es a de e
em em to se e se #
ee4 se m eg e ee en m 8me e.
go u o
e e
W U
em W
en W
en N
6 en W
Go O 0e e i
e me e
-; e-up ce b De e he8eso de
- es -ee an se e ememe en eam as e de em e sem e sem anaam ee ne se e en and,ee e
og on -
-- en se emme e hee es as e
e em
.en+e e
e e
e S
e ag. e se e
se e se es se e
- se e en e m -
e se e esea w*
S e
- are W em **
e e,
ee -
e se e en me e
am see as e e,
e en ome.e o ee
= =, =. =.
e a ** e e
e se se preauen en seen an anSeesen e4 se e se e e sc en e e en e se gesm an -amen - mene se as ge e e
arde s
e p -
e emeh%
M eseem
-e
.o.
e.em e
.e m @e.-eGe
-ce o
4
.e.e 4 hMM,ehe*ee.e em
.e #e g3.5 e
o e o e
e e
a e
g an ene e se e e p Go en som e e
en
,e
, se en e spe g, en e e4 gm Y
e4*e U
e e4 M
e w
g e en
.m.e n
--m.e
.e m. -e e.e=
me
=e e.
e......--
e e
me.e e.pecome,=
e so -
and e e.s ee
..e e
W
=-
.o.
en M
e**
se e N
E e
es W
W We-e m meg t/3
-e M
M M M48 M
W--%
e90%
in a
- 8**
- m E4 e
O w
e-o
=>4 e
4 gae
=
w e
b S
m o
gg g ******
==4
% m444 em e e -
e e
o e
s.
ce --==....
o.-e me
~ 2e e
-e"*-
3
. e e*-
e-so-
% 8'D e eoee e Men ee
@- e eD% e h um We eee e%@eeeheee eeg*
e e
ae 4
h en o
b en e a
so em 8m en e he se op ano eau ep 3
A e.*
ec ce n.
se se
.o..
e me e on..e g
ee e
-e R. e.
e e
8 e me 8"
ee 95 se et h
e em A Se e e
.e s, se e.
e
- e es3 e
=n; y
e no We ga
=
====
=
.e nee =
-ee C
e e.p. e C.
se as ee e "* e a
ese ed o
Q
=e O
no N
e
-
e
-e se-
=.
m.-
es se me-
.e.
emn ee a
e
.e se
- Gee, e
am e **
- e
. e
=,
e
. ee 3
e hE es --
ee.
em..-.*
es e===-
em e e e se e-e e
an. e =en
.. G ee...
.e a
e se o
e-as enee b
en en as en a
w Og
- b
. m g
m
&~
3 e 3
- e. e.
e e
esse up 4 meme me e e en-em en meo
--**se-me eq e
s e
m.e e..,
Q e
.o
-ee+
as-e e e M
4 m
e N
o e
e.
.4
)
W U
O me
- I"O s e.
se e=
y z==
we.
i
.e en..
m e
.e m e s w s
.a.e.B.e.g...
- w=G e
3
... ee = *
=e W
w==t c
-ase
.se e e =
w o
a
- .n..
m ag
. =.
a..
.e-
.e.e.a O
e ce s
z w m s s e = =...e.e m..ee e.
-e
...---e-est.....e e. -.e ree. ebe.
ee. em o.
c m
- 2
=
e es 3.
se = = e m e... e.
M ae r-e-
t
- e..e e.o g e.
em e
eeeeeeee=e em. m.
a.
w 6
.S.3 E = =3 3 3 - S e %D w =- a s - e n e
.ee mmes e en e
a e
e e
$.e e U S en e a e E
==
===msee e
e
=-Esta ed e ge3 geWe eeE.
S==
w greememwms
.p 333 seeev m
B.
Se om me ee o.me Pd e c e c a g a w.& = = 8 we e e
eeE e
O e
- - a E. e =. e-e R e e w g m. a m33 we-w e a
w we e.
med
,4 Se.e.e...
ne e-4 z
e, saar-e ame mEm3*
p 5-5
.e.
- s.*
e o.
w==
eeema-=-
=
-es
.==
g egeee44 e-e446 e
- - eema e
p
-
EeerEeeWewee*aB-wee ee 3 & en e== m.me.s
=
e.3 yee e e =>,,, se e
e O
ema 5 em an-e D am eo m ee e s
.I.g.4 e
S 1
g a
w
-.e e,,s t b.e.
em e*.me en P
M es em 3m e
om
-#.R,-
m, l
W e
3 gr g
e W
e e an e
e e
~*ee g
e
=.
e.-
e G
j es m.
e
.w e
me O
1
gg naea==~~oon-o oo..
==.a.o--n....g g gunanos. ng g=g
.......ong gggg
-u
" Sad N
no
.o--ooo-o......o
-.-o-o.o......
o.n.---o-,
--oononoon
.-o~
g-Q n;**
g g
o-o..e-.oooo..
noo.
-oo.ooooooo.-~
- nooooo-o..
on-ooooo-n-o.
-..g.
u M *:
5 o..aeo.ooooooo..ooo
.o.a.ooooooooo oooooo..oo ao oo.
........o.
. oo
- R*
. oo g
C,C
--o-ao-o....oo.....
oo.ooo-.oo..-.
nea n...--
o.
.ao.on--n-24J g
aa=~=genoaano.
goa.
.g......no..n=
g gn..aog.
~g ggg...... 3.g gggg g
2,4 A ao.e-o..oa-oooooooo no...oooo.-a o
ooo.ao-oo.
on
~~a oooo a.oo.
.gng "dA gg
....--....o
...oooo
.e--..oo.e-...
....o...on oa.-~
-ooo-o-o..
ggg; u
- d4 8
u
- g
...-.oo.oo....o...o
...o....o
.oo.
..-ooo..
e.-
ooooooooo-gg a a-add w
- 2
~
uc ggg
...ng.nne.....
goo.
gg.n-a--no.on.
gn...og.
- g=g n......g.
.ggg u
n..
m J
g n.
nooo--oooonoooooooo
-ooooo.o.one-n
-o-o-onoon
.on nooooo----
-oo~
p UR
- 9*
g it
.-4 W
e5 "*********"******** **********"*"" ****"***** ** **"
- aaa
"=~"
c R
":95 g
w n.
a gg e-..--ooooooo...o.-
n-...o-o.o-o-e
-.-o--oona o.
e-.
.-no-onne.
g.n~
- 9.*
m
-o p
fo.
neo
-n-nooooe.o.o..
o..nooo-nono.e oo.n--oon-
-e-
- o enn.m..p.*
- m. e. n.
g n-~
~
i
":44 e,cc M
=
.oonna-ooo--oo.-...
n---..
ooono.=
.n-o-o-o-.
.---~.
---on-on--
on..
c A
00":
Lua no.
u 3
oggg g
osoooooooooooo oooooooooo ao ooo oooooooooo
.b.o.oooooooooooooooo m
-t
~=4.~
c a
P*
- 2
- g eg oooooooooo-ooo-oooo oooooooooo.co-oo-ooo-ooo on nao
..-oooooo.
gggg "4A c
w Neo n-co-.o.oo-oooo-ooo oono-o-ooo-one oo-ono-o.n o.
~~n
...ooonnoo
-e.g d.0 t.
c o*
n
-ooon.oooooo..o.oo.
.nooooo-oo-o-.
.on---oon-
~~
.ne on-n-on-~~
g.gg c
=
- n. o 2 t.~
a
-o.
g "3
e.
-na gnanoo-o.oogoo.
.o.noooanono.~
mogo.
-ag.
g..g n~nnn.....
U so e
o.onn.~oooo-oooeoo-nn-woooooonoo.
--n-aeoomo e.
.-e
---on-enan gggg m
";d4 u
k gtg M
66 e --
r sagadsg.s;gg ag ".,
g u
n c e s::
3 e
g en n
.g-B e
g.-
a
-nn.
.-..o n
u h
g uh6:euss 5m m at,:c o.
3
- g. {.
- - -l
- d ":: E W P 3 E E T a
M C
o 3
g v e u. 2;; a.v !. v 8
-s n.2 n""5
- e. s s. s.. = 2 38
" "n d
u - I" " s-58
= a.s a c e a b c., ya s a a a :G " s..g g 3 s h,? u-Q :: a
" V" sa n
--. ~
~-
e g:
=,
s a
2 g o ",.
- n s a s s H u.s h,s..
Gr g n,. o -
l y
W re d 1
M sgsy;=
g
=
c s us it :.: s s.
m sa g
=
-:b5 l
.w s
a u m..
n(
war u
g 5
0 g
g C
a e
8-g*s g
. e. h -
o EEa3 a
3 3
2 C'
F 5-6
.e u..n-..ne..u u
.e.n.on.
u...-
n.
.non.
...n ne e.
-....n I:U r.
\\
n ~.; -
g ;... ~
1.
3 o.oa w....
.wawu.
g.
~~no.....u....
-..= uu -
u.
.-u-gg
.. g ;. g u.
.. e,.
j.a....
......a.o..
t
.~..........
....o.~
a.
.e
~a n.~
B I..................o.
~ 4 *: 9 oooo..........
.oo.......
.o
..o.......o g
.:;g G
.on-e....~.....
..nooo.ao....e
-.e..~o.o.
...n--n..n.
g*
g.~...na.e--o-o.g...
,{
gg.;-en.
..og.
- g. gen...gg 4
~~
g.=
..~.-
g....
.gg,
....-oo.........o..
"$6J
...-........o.
.o-..o....
o..-.......
- 22 2 g=...-...oo..........
,.2
..oo..-o......
oo-..oo...
-.o
....e.~.o.o 4i
- 33 e
.g4 u
- 3...................
= -
oo-...o...
..oo..o-o.*
w, u
- 3 a
.C..
4g W
N u
n..
n.
.n-
~~~...o
..n~
ss
~~
--n
- f R. ^,
y y.... o........~.
o.....o o
no
..o..
e.~~.
G g
e u
3 85 ***-*******=*******
- 5
~~~~a********* ~********~ ** ~~~
d****-**~**
?. : 0
<e
";4 g
g:
.~ene
...-.-...o a.
.-nn.a.o....-,
.oe.o-..e-
.... -..~~~~....
4 ggy a
p...-..-n-...-...--...
m a.
.....o m
M n a.
q w.
e.,
l e
.o
..o.
.oo..
c
~
n.o o o o -. o.. e.
e
.e.e n o. ~. a -
.e
-e.
.C 4
......o............
A O..".3 0
.........o....
.......o..
.o 223 g
oc
. o b
-G
..~....o.........~. *.~o..oo.oe...
..~o.e..on
.n
.o.-e...
o.
.-no 3
w
"**7 0
fno oooneen..o.....oo.n u.
W v..
en.
oo-ooooo.n e.e.on.oo-
-o..--m.n e
N. a eggg c
osg.
.nn.-..o.....~...
~~~~.a..o..on.
.o.....onn
.ao..~-~.-.
m ggg
.g.....a-a...--......
um
....-.....n...
...~....;
g g..
g.;;
g
..--oo
.oo.....
.o*
g
..o.....
U
........ ~
og
.=
Q pg. ;
m
- e. p a4 9t-,:
C.
gJe a
--.o.
u y..
-.g
- 2 R.EOUSESE3w)$I.E. + :..:. :~;
.v..e..e..n..u. n.. M
=
u SQ U
N
$*2
- 3... =
~~~*-*-** :
m M
6 li. 2 0. j
.s.
""J*.
M u t '.5 ".;.3 2 :;C R
. -. 7 7.1!. n! 7 3. d..",
I"Tr" 355j 033M.7o..
5 3 $ 3.;.
3 2..*,"..k.'
2
!.*. *. U.2.' '.
- 8
.s=.
.-v geo.i...
5.
U. E. :
50.:0 *t a a,
- T tt. u n 3
p.h
.,,. 3, 3
o.
.. U.g.
o n
gg yn l u.
m
.,a
-.g g
w.
. 8
- 4. m.
g gg
.g g
s:
.e.-
...3 3 - W.3
- a. a..a.L.
W Q em e o
a w., oEa o,
y e
s J
Q t.
o.e 8
g g
g lg g Q
.a u
.4 wt U. ". " *a o
w g
g u
g
.W
.m
--4 c.
82 n$
R E s ?.p n*
3" u
c
.p sa C
5-7
l 2.2.2.
DOBLE ENGINEERING COMPANY FAILURE STATISTICS Extracts of the.f ailere statistics compiled by Doble Engineering Company which were requested and supplied by VEPCo ' [Ref. 2] are as' follo'ws:
1980 HIGH SIDE @
LOW SIDE @
. TYPE OF
,i FAILURE NO.
BIL (kV)
BIL (kV)
MVA AGE PHASE CONST.
1 500 @ 1300 22 @ 150-369.6 7'
1 Shell 1981 4
1 500 @ 1675 20.7 0 150 400 7
1 Core 2
500 0 1425 23.8 @ 150 310 11 1
Shell i
3 500 @ 1300 22 @ 150 330 8
1 Shell 4
500 @ 1300 22 @ 150 33'O 8
1 Shell 1982 a
22 @
369.6 8
1 Shell 1
500 @
i 369.6 9
1 Shell 22 @
2 500 @
3 500 0 22 @
369.6 9
1 Shell The 1982 failure statistics contained the added colum of ' Failure Mode '.
The failure modes were idendified as a) coil, b) core, and c) other (leads, etc.).
The f ailure mode for the three 1982 f ailures listed were:
i Failure No.
Failure Mode 1
a, c l
2 e
4 j
3 e
5-8
2.2.3 IEEE NUCLEAR DATA RELIABILITY MANUAL This manual (IEEE Std. 500-1977) [Ref. 3] applies to reliability data of electrical, electronic, and sensing components for nuclear power generating stations.
Its intended use is for system reliability analyses.
The collection of data was f rom a wide range of applications and limited to actual f ailure counts in nuclear power plants. The data was not collection method utilized a Delphi survey procedure in which several questionnaire iterations were made to synthesize and refine information
^
collected from the previous questionnaires.
The data bases were predominantly from the following areas:
(1) Operating data from nuclear and non nuclear plants and other large industries (2) Published data (3) Maintenance control records (4) Manufacturer records From these data bases, the f ailure modes and failure rates were established.
The failure modes were categorized into three classes:
catastrophic, degraded, and incipient. Details of each of the failure modes for the generic listing of transformers are presented in Table 5.
A main power generator or unit transformer similar to NAPS is generically listed in Section 7.2.1 (Ref. 3].
The failure modes are defined for observable effects on the system resulting f rom a change in operating characteristics or state of the component. Thus, the f ailures are identified at a systems level rather than at the specific com-ponent level.
Factors such as aging, environmental, human error (operational),
misapplication, and cascade f ailure were included in the modes.
The environnental ef fects used in the failure data bases were tempera-ture, radiation, and humidity. Each of these effects defined a multiplying factor to be used in conjunction with the failure rate.
For the generic listing of Section 7.2.1, the multiplying f actors for high temperature, high radiation, and high humidity are 1.6,1.07, and 1.13, respectively.
The failure rates for similar type NAPS transformers are shown in Table 6 [Ref. 3].
The failure rates are shown in failures per million hours of operation. The various failure rate values are defined as follows:
Low and High:
These values represent the best and worst estimates under normal operating conditions.
Rec:
This is the recommended value to be used in system probabilities and other numerical analysis results.
The low /high values can be used as lower and upper bounds.
Max:
The upper bound value to be used under all applications.
5-9 i
TABLE 5 Transformer Failure Modes Generic Listing Catastrophic Incipient Degraded 7
No Output Mechanical damage Contamination
~
~
(a) Due to automatic removal by protective Incorrect output due circuit ry to faulty tap changer Overheated (b) Due to manual removal (delayed Output less tha.1 Corona removal) rated capacity (c) Due to open circuit (d) Removed because of shorts 7.1 No Output Mechanical damage Contamination 7.2 (a) Due to automatic 7.3 removal by protective Incorrect output Overheating 7.4 circuit ry due to faulty tap 7.5 (b) Due to manual removal changer Corona (delayed removal)
(c) Due to open circuit Output less than rated capacity 7.6 No Output Mechanical damage Contamination (a) Removed because of shorts (b) Open circuit 7.7 No Output Mechanical damage Contamination (a) Due to automatic removal by protective Output less than Overheating circuit ry rated capacity (b) Due to manual Corona removal (delayed removal)
(c) Due to open circuit Note: This table is a reproduction of Table D14 in IEEE Std. 500-1977.
5-10
.~ -
TABLE 6 - Generator - Main Power Transformer Failure Rates C% apter: ? - Transformers s,ce
- y,2 Majn power = generator er unit Subwerion
- 7.2.1 Liquid filled - IS Part Descripaken f ailure klo.le Fa.luse itete Date of Souere
- 3. ant (Il (2)
(3) denmon (Sp 14)
Failures /10'llours railuresit 0*Cy cles Low Itee llegh
&las Low Hee liigh
&fas
- 7. 2.1. 6 347 550kV ALL W) DIS
.743 1.622 ?.668 2.985 Celphi CATASTROPHIC
.532 1.162 1.912 2.139 No output a.
due to automatic removal
.478 1.043 1.716 1.92 t,y protective circuitry b.
due to manual renovel
.0421
.0920 151
.169 (delayedremoval) u c.
due to open circuit
.0124.0271
.0446
.0498 i
H CCCPADED
.0941
.205
.338
.378 w
Mechanical damage 0247.0540 0888
.0914 Incorrect cutput due to
.0247
.054
,0388
.0934 faulty tap changer Output less than rated
.0446.0973
.160
.179 capacity INCIPl[Ni
.116
.254 418 468
3.
SURVEY COMPARISONS This section presents to the extent possible a limited comparison with' the information available on the main transformer failure rates and failure modes of NAPS to the three statistical data bases.
3.1 NAPS MAIN TRANSFORMER FAILURE RATES / MODES
^
The seven main transformer failures at NAPS occurred over a 26 month interval from November, 1980 to December, 1982. This period equates to a failure rate of 3.2 per year or 3.7 x 10-4 per hour. The three areas of failure were in the HV bushing (4 f ailures), HV lead (2 f ailures), and HV coil (1 f ailure) with all faults going to the LV side (coi1, tank, winding, or ground).
3.2 NAPS MAIN TRANSFORMER FAILURES VERSUS EEI DATA An overview of the statistical data presented in Tables 1, 2, 3, and 4 indicates that the largest percentage of the transformer f ailures reported are outside the design classification of the NAPS transformers. These failures encompass lower voltage ratings, smaller MVA size, core-form type construction, and primary use in substations.
It should be noted at this point that it is difficult to make an accurate comparison with the numbers available with no specific column cross referencing. Only generalf zed correlation can be achieved.
An extraction of the statistical data for the number of failures in percent of the total inservice for the particular NAPS transformer classification is shown below.
i NAPS NUMBER OF FAILURES IN PERCENT OF TOTAL INSERVICE TRANSFORMER CLASSIFICATION
- 1979 1980 1981 Gen. Step up 1.0
.80
.88 e
100-499 MVA Size 1.0
.62
.91 1
Shell Form 0.7
.80
.51 451-650 kV Class 2.1
.93
.82 Comparing tM::e annual f ailure rates to those in other transformer classifications generally shows that transformers used for generator step up, in higher voltage classes and larger MVA sizes have a larger failure rate in The EEI statistics do not categorize transformers as either single or i
three phase units.
5-12
i percent of total transformers in service.- This means that one can expect more failures in these classifications than others.
The statistics show that trans-formers of shell-form type construction have a considerably lower annual failure rate than those of core-form type construction.
An overview of the failure modes (Tables 2, 3, and 4) for all trans-former classifications, indicates the windings (both high and low side) and the load tap changers (LTCs) were the dominant locations of failure, whereas at NAPS the HV bushing and HV lead were the dominant locations. The dominant sources shown in the tables for these failure modes were design (mechanical and elec-trical) and external (other and unknown). As a result of the above, the major contributing factor to the failure cause was dielectric breakdown. This falls in line with the failure modes at NAPS.
The other area of comparison which needs to be addressed is failures causing fire and/or explosion.
In the 3 years of troubles reported for all transformer classifications, a total of 5 failures resulted in a fire and/or explosion. Two of the five transformers involved were of the generator step-up classification. The following comparison shows the correlation of the resulting single fire from a NAPS transformer failure to those reported in the EEI statistics.
Gen. Step-up No. of Fires and/or Explosions 1.00 5.00 2.00 Total No. of Failures Reported 7.00 262.00 20.00 Percent of Total Failures 14.29 1.91 10.00 Failure Rate /Yr.
0.46 1.67 1.00 3.3 NAPS MAIN TRANSFORMER FAILURES VERSUS DOBLE ENGINEERING DATA A review of the three years (1980-1982) statistical data shows that eight transformers of the NAPS classification failed.
Seven of the eight failures reported correspond to the same time period of the NAPS failures (one in 1980, three in 1981, and three in 1982). The eighth failure reported occurred in 1980. The only major comparisons that can be made are that the voltage class, the size, single phase, the age and type construction are essentially the same as those at NAPS.
These reports did not identify detailed failure information.
Statistical data covering the age (approximately ten years) of the transformers were requested from VEPCo but were not provided.
The Doble Engineering statistics, like the EEI data, show that shell-form type transformers have a lower failure rate than those of core-form construction.
These failure rates are not based on the total population of the type in service. A comparison of these parameters is shown below:
5-13
\\
1980 1981 1982 Total Failures percent percent percent Reported 194 of total 228 of total 283 of total Shell 26 13.40 23 10.10 30 10.60 Core 168 86.60 205 89.90 253 89.40 3.4 NAPS MAIN TRANSFORMER FAILURES VERSUS IEEE RELIABILITY DATA The worst case failure rate (for all modes) for similar type NAPS 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />. From main transformers presented in Reference 3 is 2.668 failures /10 this the following can be stated:
106 hours0.00123 days <br />0.0294 hours <br />1.752645e-4 weeks <br />4.0333e-5 months <br /> = 114.16 years Failure Rate = 2.668 failures /114.16 years Failure Rate = 0.0234 failures / year For a single transformer this failure rate of.0234 failures / year transforms to a mean time of 42.74 years to the first failure. Applying this to a bank of 3-10 transformers would give 42.74/3 or 14.25 years to the first failure.
It must be noted that these failure rates assume 100% continuous operation from the time of initial in-service operation. A more reasonable value would be at most 75%.
This would produce an increased time anticipated to the first failure for a bank of 3-10 transformers of 17.81 years.
The exact operating time for the NAPS transformers is not available at this time. The following is a brief outline on the transformers' operating history.
February, 1971
. Transformers delivered to Unit 1 March / April, 1974
. Transformers delivered to Unit 2 Fe brua ry, 1974
. Transformers began servicing Unit 1 April, 1976
. Unit 2's transformers sent to Georgia Power May, 1976
. One of Unit 2's transformers f ailed at G.P.
April, 1978
. Transformer repaired and sent back to Unit 2 -
placed into service with initial synchronizing in Augur', 1980.
November, 1980 to August 1982
. First five failures at Unit 2 November, 1982 to December, 1982. Last two failures at Unit 1 5-14
Based on the above operating history, a period of approximately 6 years lapsed at Unit 1 before the first failure occurred. At Unit 2, a period of approxi-mately 21/2 years lapsed until the first failure occurred (second f ailure of the same transformer). Applying, at best, the very conservative value of 75%
continuous operation shows that the worst case anticipated time to the first f ailure is a factor of three times sooner than expected at Unit 1.
For Unit 2, the anticipated time to the first failure is approximately seven times sooner.
If the maximum upper bound for f ailure rates is used (2.985/106 hours), the anticipated tice to the first failure would be reduced by approximately 10%.
4
SUMMARY
This report documents an attempt to compare the f ailure statistics of the NAPS MT to those in the power industry and other areas where high voltage transformers of similar classifications are utilized.
As can be seen from the available statistics presented, the comparison was highly limited and not very conclusive. This is largely due to the two main factors encountered of
" proprietary" and/or " protected" information and the lack of detailed transformer failure records and operating histories, although, the second factor is probably for the most part included in the " proprietary" and " protected" information category.
The lack of detailed transformer operating histories and records causes many misinterpretations and too many assumptions in using f ailure numbers only. With just numbers and no detail available many questions can arise as pointed out below:
(1) Overloading: Was the transformer overloaded, how much, how many times, for how long?
(2) Use:
What percentage of use is continuous, how much down time,,
was there misapplication?
(3) Storage:
How was it stored when not in use, what are the storage conditions, how long in storage?
(4) Transportation / Shipping / Handling:
How many times, how were they they handled?
(5) Maintenance:
Were normal routine preventative maintenance procc-dures used, was any extra maintenance required and how of ten?
The questions above are only a few that would have to be answered in order to compile accurate meaningful statistics.
Both the EEI report and the IEEE standard statistics did attempt to take into consideration some of these pa ra me te rs.
One last point that must be made with respect to these f ailure statis-tics is that the data for the most part are only from a limited representative sample and does not reflect the majority of the transformer failures experienced.
5-15
5.
CONCLUSION Based on the ' three limited statistical data bases comparisons, it can be concluded that:
(1) As the operating voltage becomes higher and the size becomes larger, higher failure rates in percent of the total type of transformer in service are expected.
~
(2) Transformers of shell-form construction have a significantly lower f ailure rate than those of core-form construction, irrespective of voltage class, size, application, type, and phases.
(3) The dominant contributing f actor in transformer f ailures is dielectric breakdown.
(4) The dominant locations of transformer f ailures are the windings (both high and low side) and load tap changers, unlike the high voltage lead and bushing locations with the NAPS transformers.
(5) NAPS has a lower failure rate per year for fires and/or explo-sions caused by generator step up transformer f ailures.
(6) NAPS experienced a sooner-than expected time until the first f ailure (three times for Unit 1 and seven times for Unit 2) than the reliability data show.
REFERENCES 1.
Edison Electric Institute Report (Jacei C. Goellner) to Jaces C. Selan (LLNL), dated June 16, 1983.
2.
VEPCo letter (W. L. Stewart) to the NRC (H. R. Denton and R. A. Clark),
dated August 26, 1983.
3.
ANSI /IEEE Std. 500-1977, "IEEE Guide to the Collection and Presentation of Electrical, Electronic, and Sensing Component Reliability Data for Nuclear-Power Generating Stations. "
l l
l 5-16
I TASK REPORT 6 EVALUATION FOR THE GENERIC ASPECTS OF THE MAIN TRANSFORMER FAILURES AT THE NORTH ANNA POWER STATION,-UNITS 1 AND 2 James C. Selan e
e i
6
~
1 i
ABSTRACT This task report documents the evaluation of the main transformer f ailures at the North Anna Power Station, Units 1 and 2 to determine if the f ailures present' a' ' generic' concern in other nuclear power plants.
The scope of the evaluation included such areas as maintenance procedures and practices, protection systems, root causes, transformer fires, f ailure con-sequences, etc.
The evaluation finds that as a result of the seven failures, several generic concerns exist in the areas of offsite power routing and sepa-ration and fire protection systems.
The evaluation also found that several other concerns exists which may require additional reviews or analyses.
i FOREWORD This report is supplied for the U. S. Nuclear Regulatory Commission, i
Office of Nuclear Reactor Regulation, Division of Licensing, by Lawrence Livermore National Laboratory.
The U. S. Nuclear Regulatory Commission funded the work under the authorization entitled " Evaluation of Main Transformer Failures at North Anna Power Station," B&R 20 19 10 12 1, FIN A-0442.
1 r
6-1
TABLE OF CONTENTS Page 1.
INTRODUCTION b-1 2.
PRIMARY ASPECTS.
6-1 2.1 TRANSFORMER FIRES 6-1 2.1.1 FIRE. PROTECTION.
6-2 6-4 2.1.2 OVERHEAD CONDUCTORS / BUSES 2.1.3 CABLE TRAYS.
6-5 2.1.4 STORAGE OF SPARE EQUIPMENT.
6-5 2.1.5 STORAGE OF FLAMMABLE MATERIAL NEAR POTENTIAL FIRE HAZARDS 6-5 2.1.6 OIL-FILLED TRANSFORMERS IN GENERAL..
6-5 2.2 TRANSFORMER MAINTENANCE AND OPERATIONAL PROCEDURES.
6-6 2.3 EXCESSIVE SHIPPING AND HANDLING.
6-6 i
f.
3.
SECONDARY ASPECTS 6-7 3.1 CASCADING EFFECTS 6-7 3.2 EXTENSIVE ELECTRICAL / MECHANICAL DAMAGE.
6-7 3.3 MISSILES / EXPLOSIONS.
6-7~
1 1
4 4
CONCLUSIONS.
6-8 REFERENCES 6-9 i
TABLE OF ILLUSTRATIONS l
1 j
Figure 1 NAPS Transformer Fire Suppression and Drainage Systems.
6-3 1
i j-6-111
EVALUATION FOR THE GENERIC ASPECTS OF THE
. MAIN TRANSFORMER FAILURES AT THE NORTH ANNA POWER STATION, UNITS 1 AND 2 (Dccket Nos.- 50-338, 50-339)
James C. Selan
~
Lawrence Livermore National Laboratory, Nevada A
1.
INTRODUCTION The North Anna Power Station, Units 1 and 2, experienced seven main t ransformer f ailures over a 26 month period. With each occurring transformer f ailure new operational data and specifics were being identified. This accum-ulation of information was the basis for which numerous evaluations on various aspects concerning the failures were performed.
The subjects of these evalu-ations are identified in Task Reports 1 through 5.
The purpose of this task report is to determine if any aspects surrounding the transformer failures present a generic concern to all other nuclear power plants. This determination will be made based on all the available information and the evaluations documented in each of the separate task reports.
2 PRIMARY ASPECTS This section attempts to identify prieary aspects of the main trans-f ormer f ailures which may present a generic concern with respect to other nuclear power plants. The generic aspects listed below are not prioritized in order of importance. There may exist an "effect and consequence" relation-ship between the various aspects.
2.1 TRANSFORMER FIRES The available failure information indicates that out of the seven l
f ailures, three resulted in a ruptured transformer tank.
Of these three tank ruptures, two fires occurred, but only one was attributed to the ruptured i
tank itself and to the spilling of transforcer oil. The other fire (very ' minor) 6-1
which occurred was limited to some scaffolding below the phase ducts which caught on fire due to hot' metal fragments and burning cable from the elec-trical fault.
The resultant fire from the ruptured tank and the spilling of transforrer oil was extensive. The fire was not contained within the trans-former bay and spread beyond to the turbine building and other components.
The ' extent ' of this fire could not be correlated with the fire related failure statistics presented in Task Report 5 (Interim Report on the Survey Conducted of the Electric Power Industry Experiences with Extra High Voltage Transformer Failures and the Comparison to the Main Transformer Failures at the North Anna Power Station, Units 1 and 2) as the data was not of sufficient detail. However, the statistics (EEI data) did indicate a lower fire related failure rate at NAPS.
A review of the events surrounding this fire indicates that a fire of sufficient magnitude (location dependent) has the potential for degrading plant safety equipment and safety systems.
The following sections identify those relevant areas of concern of transformer fire with regard to plant safety systecs and operation.
2.1.1 FIRE PROTECTION SYSTEM It is clearly evident from this incident that mitigating the ef fects of a transformer fire is not only necessary to preserve the integrity of the safety systems but to prevent severe damaging effects.
A review of the events and available information finds four areas encompassing the transformer fire that need to be addressed.
Specifically, these are the deluge system, drainage system, fire barriers, and fire fighting and related procedures.
The deluge system itself demonstrated that it is a very important integral part for an overall ef fective fire protection system. This system eliminated the potential fire damage which could have occurred to the other two units (phases A and C).
The spare transformer bay (empty at the time of the accident) is not equipped with a deluge system.
If the spare bay had not been empty (no deluge system), the cransformdr would have conceivably suffered damage due to the intensity of the fire. Conversely, had it been equipped with a deluge system, the spread of the flammable oil / water mixture could have been increased due to the additional flow of water (oil being lighter in weight than water). This then introduces the subject that the drainage system must be adequately sized to accommodate the flow of water and oil.
The fire suppression and drainage systems at NAPS are illustrated in Figure 1.
The common bay area for the transformers consists of approxi-mately five feet of crushed stone with a 6-inch retaining wall above the grade and the crushed stone.. There are two 6-inch drainage inlets used for the collection and discharge of any accumulated liquid.
6-2 l
This drawing was adapted from USNRC Report Nos. 50-338/81-19 and 50-339/81-16.
~ %. ^ C ^ Y UNIT 1 UNIT 2 TURBINE BUILDING WALL
(
)
CABLES g ETE
,-9 FIRE SUPPRESSION SYSTEMS (r
/
i
~.
5
[
N Q
3 { [-:
It}
bh g
'y s
a 1e
, zu x
t
/
M' L,.
-s
~
^
~ " ~ ~
HAIN TRANSFORMER
- _. ~~
~~
g::,,
CPARF RAY.
- e'
~
g r
~ ~'
~
RAINS COMMON
_ u s
s.
SIDE VIEW RESERVE STATION C BUS BARS STATION SERVICE TRANSFORMERS TUR8tNE BLDG llALL
//N 1
i g
4 1
re=a=,r c
A 8
p 1 ~ -w 'm---
.%.. < -c.
,q,
~
'~
~ Qu~y: ~,.*h-.
~~,
~
~
s Y.
u O %.6" DRAIN %O _----
1_3 N ' '~
A'., 5 o
-~~
~
. -,7
-e,.__ _
TOP VIEW q'
I l
FIGURE 1 - NAPS TRANSFORMER FIRE SUPPRESSION AND DRAINAGE SYSTEMS t
6-3
. The amount of discharge from the transformer deluge system combined with the water used by the fire fighters approaching from the east directed the water / oil flow toward the spare transformer bay and the turbine building where it accummulated and overflowed. The two drainage inlets became ineffective due to this directional flow. Had the fire fighters been able to approach predomi-nantly from the west, the station service transformer and possibly Unit 2 equip-ment might have suffered damage rather than the Unit 1 turbine building, cables, and overhead bus bars. Transformer bay drains should be strategically located to complement fire fighting plans.
The spread of the flammable mixture could have been greatly reduced if the transformer bay segregation was increased and each had individual drains.
With the use of a three wall fire barrier, the water / oil mixture was allowed to flow around the end of the separation wall to the adjacent bay and surrounding area.
A review of the limited information available surrounding the July 3, 1981 fire concerning fire fighting and related procedures finds that an independent evaluation cannot be made. However, an NRC memorandum dated Au gus t 21,1981 [Ref. l} does recommend six items pertaining to the fire at NAPS as possible generic concerns. A review of these items finds we concur that they should be considered for possible generic concerns.
In general, the subjects of these concerns are as follows:
(1) Increase access to building roofs which exceed ladder heights.
(2) Fire hose stations for building roofs where height is a factor.
(3) Fire plans for non safety areas which have a potential for a fire hazard.
(4) Common radio communications for all fire fighting organizations.
(5) Increase moveability of fire fighting equipment.
(6) Increased overall personnel training to fire potentials, perfor-mance, equipment, etc.
t 2.1.2 OVERHEAD CONDUCTORS / BUSES The bus bars, which supply offsite power f rom reserve station service trans forner C (RSST-C) to the non-Class 1E equipcent, were routed overhead near the spare transformer bay. Offsite power supplying the Class lE equ!pment from RSST-C is by underground cable. With the spread of fire from the directional flow of the water / oil mixture into the spare bay and onto the turbine building, the bus bars melted causing an electrical f ault and tripping of the reserve station service transformer (RSST). Just prior to the RSST trip, Unit 2 saa in a hot shutdown where the Class 1E equipnent was being supplied by the RSST.
This electrical f ault propagated by means of an undervoltage spike into the plant 's Class 1E buses. This undervoltage spike caused protective relaying on the Class lE buses to operate and spurious alarms and trips.
This combination
~
of relay actuation and spurious alarms / trips caused a spurious safety injection to occur. This was terminated af ter approximately 2 minutes.
(
6-4
~
. ~ -
~ _. -.
I The concern'here is that the' location and the separation of the General Design. Criterion 17 (GDC 17) preferred offsite power source ' circuits not be near potential significant fire hazards.
Design requirements should
.be integrated into GDC 17 to review for loss of the preferred offsite power sources due. to a common ' fire.
The review should evaluate such factors as-overhead versus underground,; location, separation of the preferred offsite power source circuits, locations of transformer bay drains (capacities),
. possible accumulation of flammable liquids, and fire fighting accessibility.
- At NAPS Unit 2, fire fighting was predominantly. from the east thus directing the water / oil flow towards the overhead RSST buses, 2.1.3~
CABLE TRAYS The cables involved in the fire incident were associated with the overhead buses from the REST.
These cables were located in vertical-trays on the turbine building. Again the directional flow and accumulation of the water /
oil mixture and a subsequent-fire started the cables burning. The fire spread up L
_ the tray to the roof of the turbine building, where fire fighting was hindered due to inaccessibility and lack of equipment. The same factors addressed in the previous section (2.1.2 Overhead Conductors / Buses) should be evaluated in the design requirements for cable trays where a fire may affect plant safety systems.
2.1.4 STORAGE OF SPARE EQUIPMENT i
The burning oil was carried to the spare transformer bay by the water from the deluge and firefighting systems.
Fortunately the spare transforner bay was empty, since the spare had been removed earlier to replace the C-9 f ailed trans f o r mer. As stated previously, the spare main transformer bay at NAPS is not equioped with a fire suppression deluge system for protection from potential fire hazards. Extensive damage.to the spare transformer could have resulted had it been located in this bay.
Although fire damage to spare equipment such as the main transformer is not safety related, it does have an economic impact on continued plant operation.
4 i.
2.1.5 STORAGE OF FLAMMABLE MATERIAL NEAR POTENTIAL FIRE HAZZARDS l,.-
Documentation on the fire incident indicated that during the fire fighting operation some construction material was damaged in the vicinity between Unit 2 and the area for Unit 3.
Since this reference was only brief, it is' con-cluded-that this material did not contribute to the spread of the fire. However, the storage of flammable material should not be near potential fire hazards.
2.1.6 OIL-FILLED TRANSFORMERS IN GENERAL
'It is clearly evident that electrical faults in liquid-filled trans-formers (oil) can be of sufficient magnitude to cause tank rupture and flashover 6-5 y
,,.-----.,-,-m,..,-
.y,,e e
e
to ignite the insulating media.
Tank rupture caused by internal pressure buildup can spray the oil over considerable distance and area.
Since oil is lighter in weight than water, the spread of the flammable liquid can be greatly increased uy a water flow. Therefore, it is of utmost importance that, for a transformer tank rupture, the spread of flammable liquid be contained to a relatively small area by the ef fective use of fire barriers and drainage systems.
In addition, since the liquid is highly flammable, the use of an effective fire suppression system (deluge system) is also of importance. These factors of containment and fire suppression must be considered where equipment location is such that a potential fire hazard could affect plant safety sytems.
2.2 TRANSFORMER MAINTENANCE AND OPERATIONAL PROCEDURES Lack of adequate transformer maintenance and poor operational pro-cedures and practices can contribute to increased transformer failures and unreliability.
Following manuf acturer's recommended maintenance and opera-tional procedures can effectively reduce the chance of f ailure. This should be supplemented by developing good preventive maintenance programs and prac-tices. The Key to any effective program is having well-trained and knowledge-able personnel.
Conflicting manuf acturer's instruction leaflets on bushing storage led to f ailures No. 2 and No. 3 (see Task Report 2, pp. 2-7, 2-8).
Operator error on the inertaire system contributed to failure No. 1, and the possibility of foreign matter lef t from transformer work may have contributed to f ailure No. 7.
All of these incidents reflect that proper and precise maintenance practices and procedures are a must. Taking short cuts or deviating from established procedures may contribute to increased f ailures.
2.3 EXCESSlVE SHIPPING AND HANDLING Power transformers, like most other electrical apparatus, are skillfully engineered and carefully manufactured. They consist of integral components assembled to form a useable operating piece of equipment.
Power transformers should receive scheduled inspections, tests, maintenance, and be carefully handled in installation and shipment.
Since these transformers consist of large masses of iron and copper they can undergo substantial physical stresses f rom temperature changes (expan-sion and contraction) and vibration (transportation, handling, etc. ).
A common factor in the first five MTFs is that they were all subjected to excessive shipping and handling and repeated installations.
Due to their physical size, the Westinghouse GSU transformers are designed to be shipped on their side. This requires additional shipping braces to support the internal structures. VEPCo did indicate that they do utilize f actory supplied supports and instructions in their shipping procedures (Ref. 2].
6-6
3.0 SECONDARY ASPECTS This section identifies three secondary aspects of the main trans-former f ailures which are not considered to be a generic concern for new requirements but rather a concern for the operational effects that a trans-former f ailure may have on plant safety systens.
3.1 CASCADING EFFECTS Failures No. 1, 3, and 7 exemplify how an electrical fault occurring in a transformer can cause a cascading effect in plant systens. The magnitude can be limited by the protective systens.
Since a definite time for overlapping zone coordination is involved in protective systems, the initiating event in most cases cannot be isolated instantaneously. Manual intervention into the automatic protective systens may compound the effects. A means to minimize cascading ef fects is to ensure that proper coordination between automatic protective systems is established and will operate as designed.
3.2 EXTENSIVE ELECTRICAL / MECHANICAL DAMAGE Extensive electrical / mechanical damage can result from transformer f ailures similar to those at NAPS. Again failures No. 1, 3, and 7 depict those instances where cascading effects caused extensive electrical / mechanical da mage.
Failure No. 3, resulted in a fire which caused structural damage to the turbine building and transformer bay fire barriers.
The importance here is to minimize the extent of damage that may result from a transformer f ailure. The principal f actor in accomplishing this is to optimize the protective systens.
Isolation and time are the key elements of the protective systems for reducing the extent of dama;e.
3.3 MISSILES / EXPLOSIONS Based on the observed transformer f ailure experiences at NAPS, there were no resulting missiles or explosions.
Further observation f rom a plant trip indicated that for a resulting missile or explosion from a main transformer f ailure at NAPS, the possibility of the plant 's safety systems being af fected is remote. The plant's protective systems are separated f rom the main transforners by a fire wall, a 10 foot dead space, turbine building wall and structural members, floor elevations, auxiliary piping, and equipment.
However, the possibility of a udssile or explosion from a transformer f ailure cannot be discarded.
If the location of a transformer is such that plant safety systems may be jeopardized by potential missiles from a f aulting trans-former, then design requirements should be evaluated.
6-7
4 CONCLUSION Based on the available information on the seven main transformer f ailures at the North Anna Power Station, Units 1 and 2, it is concluded that the following items have sufficient importance to the integrity of plant safety systens that they be considered a " generic" concern.
(1) Design requirements should be established to prevent the spreading of a potential fire from the tank rupture of liquid filled (flammable) transformers.
These requirements should pertain to all transformers whose location is such that plant safety systems would become vulnerable from a potential trans-former fire.
Specific requirements to the containment of a fire should include:
(a) Individual fire suppression deluge systens (to include any spare equiprent if located in common area).
(b) Individual fire walls or barriers to prevent the spread of a fire to adjacent transformers or to other vital equipment.
(c) Drainage systems (preferably individual) to adequately accommodate the sources of liquid from the deluge system, transformer tank, and fire fighting equipment.
(2) Design requirements should be established for the routing and separation of of fsite power source feeds where a potential trans-former fire could cause loss of offsite power. Transformer fires should be considered in GDC 17 requirements.
(3) Design requirements should be established for the routing and separation of other cables or conductors whose failure due to a potential transforcer fire could impact on plant safety systems.
(4) A review program should be established to periodically evaluate the adequacy of the fire protection systems and fire fighting procedures for both safety related and non safety related areas where a fire hazard potential exists for impacting plant safety sys te ms.
This review should include such factors as accessibility, adequate equipment, fire fighting personnel (training) and communications.
In addition to the above " generic" concerns are concerns of a generic nature which may require additional reviews or analyses.
These concerns are:
(1) As the operating voltage becomes higher and the transformer size becomes larger, f ailure statistics show that larger f ailure rates in percent of the total in service are expected. These failure 6-8
rates may be reduced by ensuring that a high degree of reliability of the transformers and auxiliary systems are maintained. This may include increased maintenance schedules, better trained personnel, periodically updating manuf acturer's product information (instruc-tion leaflets), and better operational practices (installation, handling, record keeping, etc.).
(2) Increasing component reliability can be achieved by minimizing the amount of shipping and handling it receives, particularly with higher operating voltages (EHV).
(3) Cascading effects can be minimized if automatic protective systems ' coordination is optimized.
Operator errors can only compound the effects.
Therefore, it is of utmost importance that the personnel be thoroughly trained and knowledgeable in the operating philosophies of the system and its components.
(4) Extensive electrical and mechanical damage can result from a transformer f ailure, and conservative means should be exercised to minimize the damage.
(5) Missiles or projectiles from a transformer failure impacting plant safety systens is remote, but the potential does exist since it is location dependent.
REFERENCES 1.
NRC memorandum, R. C. Lewis to James H. Sniezek, dated August 21, 1981.
s 2.
VEPCO letter (W. L. Stewart) to the NRC (H. R. Denton), dated August 17, 1983 O
4 6-9 L
/
OVERALL
SUMMARY
OF THE TASK REPORTS ' CONCLUSIONS James C. Selan O
e 2
e
{
i i
i f
7
l 7.
SUMMARY
A summary of the results of the six Task Report evaluations are presented as below:
(1) Based on the WASH 1400 data bases, the probability of a main transformer failure adversely affecting the operation and capability of. the emergency systems so as to increase the risk to the public health and safety is negligible.
(2) A limited f ailure mode and effects analysis showed that a main transformer failure can have primary and secondary effects and the extent of those effects depends on the adequacy of the protection systems for isolating the f aulted transformer.
(3) The increased f requency of main transformer failures accom-panying plant trips increases the overall system challenges but does not appear to significantly degrade the plant pro-tection and reactor coolant system designs.
(4) The following factors were found to be significant contri-butors to the cause of the seven failures:
(a) Excessive shipping, handling, and repeated installations.
(b) Improper storage of components when not in use.
(c) Gas bubble evolution reducing the dielectric properties.
(d) Overexcitations.
(5) Our evaluation found no cause for failure related to the Westinghouse transformer design used at NAPS.
(6) Conservative maintenance practices and procedures are a must, especially with the higher operating voltages.
(7) The transformer / generator protection system's performance was very good and does not require modifications except for optimizing protective zone coordination (backup relays). The design of the protection systers follow good engineering practice.
1 i
7-1
.(8) The following licensee proposed modifications are considered to be of significant value for upgrading the fault protection systems for the generator.
~
(a) Installation of 24 kV surge arresters on the isolated phase buses.
(b) Replacing the generator neutral ground with a shieldedL cable with stress cone terminations.
i B
(c) Increase conductor size for the generator frame grounds.
(d) Increase the BIL rating of the neutral grounding transformer.
(9) Excessive shipping, handling, and repeated installations of the transformers and components must be eliminated if these contributing f actors to increased f ailures are to be significantly reduced.
(10) All manufacturer's recommendations should be explicitly followed unless justifiable analyses show otherwise.
(11) A properly instituted preventive maintenance program can only contribute significantly to the increased reliability and 4
integrity of systems and components.
Complementing a good program is the implementation of accurate and up-to-date record keeping, having thoroughly knowledgeable and trained personnel in EHV systems and their supporting devices, regular scheduled inspections, and practicing good conservative maintenance procedures.
(12) Modifications to the fire protection system in the area of increasing individual transformer bay segregaticn, holding capacity, and capacity of the drainage system to discharge the volume is considered necessary.
(13) With only three statistical data bases available, the limited comparison resulted in the following correlations:
(a) As the operating voltage becomes higher and the size becores larger, higher failure rates in percent of the i
total type of transformer in service are expected.
(b) Shell-form transformers have a significantly lower f ailure-a rate than those of core-form construction.
(c) The dominant factor contributing to a failure is dielectric breakdown of the inst 1ating media.'"
(d) The dominant locations for failures are the windings (high and low side) and the load tap changers, unlike the HV lead and bushing of the NAPS ' f ailures.
7-2 l
C
o (e) NAPS has a lower percentage rate for resulting fires and/or explosions from generator step up transformers f ailures.
(f) NAPS experienced a sooner-than expected time to the first failure than the reliability data would indicate.
(14) The following aspects of the MTFs have sufficient importance to the reliability of plant safety systems to.be considered primary
~
" generic" concerns as applied to the main transformers:
(a) Individus1 fire suppression deluge systems for spare equip-
~
ment should be installed if located in a common area.
(b) Individual transformer bay segregation should be increased to prevent the flow of a flammable water / oil mixture to adjacent or surrounding equipment. This would also encompass the adequacy of the area to contain the volume of liquid from the tank and deluge systems and the adequacy of the drainage system to discharge this volume.
(c) Cable and bus separation and routing f rom the preferred offsite sources near a potential fire hazard which could impact on the operation of plant safety system should be incorporated into GDC 17 criteria.
(d) Additional emphasis should be placed on regular scheduled reviews to evaluate the adequacy of all phases of operation surrounding fire fighting plans for both safety and non-safety related areas.
This should include such items as adequate equiprent, common communication networks, adequate personnel training, accessibility to the fire, etc.
(15) The following generic aspects of the MTFs may require additional reviews or analyses for potential upgrading as applies to the main transformers:
(a) Conservative procedures and practices and better trained personnel are required as the operating voltages becomes higher to ensure system reliability.
(b) Excessive shipping, handling, and repeated installations must be minimized.
(c) Protective system's zone coordination must be optimized to adnimize the damage from cascading eff ects.
(d) Potential threats from missiles or projectiles f rom f aulting transformers impacting plant safety systems are remote, but may exist depending on the location of the transformer.
7-3
u OVERALL CONCLUSION OF THE EVALUATION OF THE SEVEN MAIN TRANSFORMER FAILURES AT THE NORTH ANNA POLDER. STATION, UNITS 1 AND 2 James C. Selan -
e e
9 t
I 1
8 f
h
,g,
.-.-.,e e,,
--,--,..r.,
,.n.,, -, - L ee,,,,,~
,,-e---v->
r
E-8..
CONCLUSION
' Based on the six Task Reports ' evaluations of the main transformer f ailures at the North Anna Power Station, Units 1 and 2, it is concluded that:
(1) No known single common cause contributed to the seven failures.
(2) No cause for failure was found to be related to the Westinghouse transformer design of the NAPS transformers.
o (3) Generic concerns resulted in the area of fire suppression for transformer fires and their impact on plant safety systens.
(4) The root causes of Failures No. 1, 2, 3, and 5 are known.
(a) Failure No. I resulted because of dielectric breakdown of the insulating oil due to the evolution of N2 bubbles from the inertaire system.
This condition was caused by operator error (inertaire system turned off) and an inoper-able winding hot spot temperature indicator which led to manual initiation of additional cooling.
(b) Failures No. 2 and 3 resulted because of dielectric break-down of the bushing condenser oil.
This condition was caused by improper storage due to conflicting manufacturer's instructions and extensive transportation.
(c) Failure No. 5 resulted because of mechanical f atigue of a bushing condenser support bolt. This condition was caused by extensive handling and transportation.
(5) The root causes of Failures No. 4, 6, and 7 are not prei sely known, however, the f ailure mechanises are known.
(a) Failure No. 4 resulted f rom an incipient f ault in the trans-former prior to energization and failure. The cause of the incipient fault is not known..The transformer had been in the banks when three separate failures occurred where it was exposed to overvoltages.
The overvoltages may have caused the LV turn-to-turn f ailure and the generation of carbon particles which produced multiple arcing paths and subse-quent dielectric breakdown. The presence of extensive car-bon coating internally was found during transformer tear down.
(b) Failure No. 6 resulted from dielectric breakdown of the insulating oil.
The breakdown may have occurred due to the formation of gas bubbles as a result of an ambient tempera-ture change of 30'C since processing, causing the nitrogen in the solution to become insoluable. As the transformer 8-1 W.
was' energized' and the ' coolers / pumps activated per VEPCo-Westinghouse established procedures, the-circulation caused the bubbles to accumulate in the " stovepipe" near the HV bushing.
(c) Failure No. 7 resulted from dielectric breakdown of'the insulating oil. The breakdown may have occurred due to oil contamination from metallic particles or the intro-duction 'of foreign material during transformer work.
Prior to energization, the " stovepipe" was removed, a constant oil preservation system was installed and the cooling pumps refurbished because of bearing f ailure.
The coolers were activated per new Westinghouse proce-dures to eliminate gas bubble evolution into the oil.
(6) VEPCo has established a reasonable determination of the root causes and/or f ailure mechanisms and identified acceptable corrective actions and/or recommendations to prevent failure reoccurrence.
O i
0 e
8-2
DISCLAIMER This document =es prepared as an account of work sponsored by an agency of the United States Go,ernment.
Seither the United States Government nor the University of California nor any of their employees, makes any warranty, espress or implied, or anunees any legal liability or responsibility for the accuracy, com-pleteness, or uwfulness of any information, apparatus, product, or process disclowd, or reprewnts that its uw would not infringe privately owned rights. Reference herein to any specific commercial products, proccu, or wrsice by trade name, trademark, manufacturer, or otherwiw, does not necessarily constitute or imply its enderwment, recommendation, or favoring by the United States Government or the University of California.
The views and opinions of authors espressed herein do not necessarily state or reflect those of the Lnited States Government thereof, and shall not be owd for advertising or product enderweent purpows.
9 1
rnneed in the t:n.ied se,ies. Ament.
4. laie er.n.
N,e..i Te< haw.i i.e.rm.e.. eier,.c.
U.S. Department et Commerce 9299 Port R. van R.e4 Spnngheid, s4 22161 Prwr Prsated Cep,8
- stscreewho itse Domestic Domestic Page Range Price Page Range Price 001 025 5 7.00 326-350 5 26.50 026-050 8.50 351 375 28.00 051 075 10.00 376-400 29.50 076 100 11.50 401 426 31.00 101 125 13.00 427-450 32.50 126-150 14.50 451-475 34.00 151 175 16.00 476 500 35.50 176-200 17.50 501 525 37.00 201 225 19.00 526 550 38.50 226 250 20.50 551 575 40.00 251 275 22.00 576-600 41.50 276-300 23.50 601 up' 301 325 25.00
' Add 1.50 for each additional 25 page increment, or portion thereof from 601 pages up.
' s i