ML20248A726

Summary of ACRS Ac/Dc Power Sys Reliability Subcommittee Meeting on 890607 in Bethesda,Md Re NRC Proposed Resolution of Generic Issue 128, Electrical Power Reliability

ML20248A726
Person / Time
Issue date:	07/03/1989
From:	Advisory Committee on Reactor Safeguards
To:	Advisory Committee on Reactor Safeguards
References
REF-GTECI-048, REF-GTECI-049, REF-GTECI-128, REF-GTECI-A-17, REF-GTECI-A-30, REF-GTECI-A-44, REF-GTECI-A-47, REF-GTECI-EL, REF-GTECI-NI, REF-GTECI-SY, TASK-048, TASK-049, TASK-128, TASK-48, TASK-49, TASK-A-17, TASK-A-30, TASK-A-44, TASK-A-47, TASK-OR ACRS-2649, NUDOCS 8908090010
Download: ML20248A726 (13)

Text

- _ _ _ . _-

00$$~A8

/ -/r l

I i

ACRS Meeting Minutes / Summary of the l AC/DC Power Systems Reliability Subcommittee June 7, 1989 Bethesda, Maryland i

Purpose The purpose of this Subcommittee meeting was to review the NRC staff's i

proposed resolution of Generic Issue 128, " Electrical Power Reliabil-ity."

Attendees ACRS NRC C. Wylie, Chairman R. Baer, RES J. Carroll, Member D. Thatcher, RES W. Kerr, Member P. Davis, Consultant Others M. El-Zeftawy, Staff R. Bell, NUMARC R. Haroldsen, INEL T. Pietrangelo, NUMARC S. Sharron, SERCH Meeting Highlights, Agreements, and Requests

1. Mr. Wylie, Subcommittee Chairman, stated the purpose of the Subcom-mittee meeting and introduced the other present ACRS members and consultants.

Mr. Wylie briefed the Subcommittee regarding Generic Issue-128. He indicated that GI-128 was identified in NUREG-0933, "A Prioritiza-tion of Safety, " Revision S March 1987, as the integration of three separate safety issues, namely:

' Generic Issue 48 (GI-48), Limiting Conditions for Operations l (LCOs) for Class IE Vital Instrument Buses, which deals with a 1

2 50/

a onen 8908090010 890703 j fjp ACRS g Certin ce y M g _'[I .

WORKING COPY DATE ISSUED: 6/15/89 ACRS Meeting Minutes / Summary of the AC/DC Power Systems Reliability Subcommittee o June 7, 1989 Bethesda, Maryland Purpose C The purpose of this Subcommittee meeting was to review the NRC staff's proposed resolution of Generic Issue 128, " Electrical Power Reliabil-ity."

Attendees ACRS NRC C. Wylie, Chairman R. Baer, RES J. Carroll, Member D. Thatcher, RES W. Kerr, Member P. Davis, Consultant Others M. El-Zeftawy, Staff R. Bell, NUMARC R. Haroldsen, INEL T. Pietrangelo, NUPARC S. Sharron, SERCH Meeting Highlights, Agreements, and Requests

1. Mr. Wylie, Subcommittee Chairman, stated the purpose of the Subcom-mittee meeting and introduced the other present ACRS members and consultants.

Mr. Wylie briefed the Subcommittee regarding Generic Issue-128. He indicated that GI-128 was identified in NUREG-0933, "A Prioritiza-tion of Safety, " Revision 5, March 1987, as the integration of three separate safety issues, namely:

' Generic Issue 48 (GI-48), Limiting Conditions for Operations (LCOs) for Class IE Vital Instrument Buses, which deals with a

________-_______--___-____-___________-_____-__--_____----_A

.7 m - - - - - - - - - - - - -

AC/DC Power Systems Reliability Mtg Minutes June 7, 1989 safety concern that some operating nuclear power plants do not have administrative controls or technical. specifications {

governing operational restrictions for their Class 1E 120 y ac. /

vital instrument buses and associated inverters that may result in the failure of the plants' safety systems to perform their' safety functions when required.

• Generic. Issue 49 (GI-49), Interlocks and LCOs for Class IE Tie Breakers, which deals with a safety concern that independent, redundant Class 1E ac or de buses can be interconnected via tie breakers which may be left closed by mistake. When left j closed, the tie breakers can compromise the independence of {

the redundant safety-related buses and, in some cases, may prevent the emergency diesel generators from supplying emer-gency power when needed.

• Generic Issue A-30 (GI A-30), Adequacy of Safety-Related DC Power Supplies, which deals with a safety concern that some plants may not have adequate provisions (e.g., monitoring, maintaining, and tests) for assuring that the de power sup-plies are available and capable of performing their safety  ;

functions when needed.  !

The staff's proposed resolution of GI-128 is to issue two separate generic letters with related information requests pursuant to 10 CFR 50.54(f), " Conditions of Licenses," to inform the licensees of related concerns and to obtain infonnation for 'the staff to assess 4 whether necessary actions have been taken to resolve these con-cerns. One generic letter would address GI-48 and GI-49, and the other would address GI A-30.

2. Mr. R. Baer, NRC/RES - Branch Chief / Division of Safety Issue Resolution, outlined GI-328 current status. He indicated that on

1

.. I J

AC/DC Power Systems I Reliability Mtg Minutes 3- June 7, 1989- l May 24, 1989, the CRGR had a meeting on the subject issue and basically concurred on the proposed resolution with minor changes.

The staff is ready to brief the ACRS in June and July 1989 (if l needed). Mr. Baer indicated that he expects the proposed generic letter to be issued in Augur.t 1989.

3. Mr. D. Thatcher, NRC/RES-Task Manager, p asented a brief background regarding Generic Issues-48 and 49. He indicated that the primary concern of GI-48 was identified when it was found that some operat-ing nuclear power plants do not have any administrative controls or technical specifications governing operational restrictions for their Class 1E 120 V ac vital instrument buses (VIBs) and associ-ated inverters. Without such restrictions, the normal or alternate power sources for one or more VIBs could be out of service indefi-nitely. This could place certain safety systems in a situation where they could not meet the design basis, including loss of off-site power or the single failure criterion.

Nr. Thatcher indicated that the designation of " Vital Instrument Bus" may be interpreted differently for different plants. The staff designation refers to the ac buses that provide power for the instrumentation and centrols of the engineered safety features (ESF) and the reactor protection system (RPS) and are designed to provide continuous power during postulcted events. T S breakers are devices used to cross connect either redundant Class IE buses l' in one unit or Class IE buses in different units at the same site.

The staff.is concerned that the VIBs may be subject to power failure modes that may not have been considered during the safety analysis of the plant, such as in the event of removal of one or more of the normal or alternate power sources for the VIBs from service for repair or maintenance.

i AC/DC Power Systems Reliability Mtg Minutes June 7, 1989 The concerns of GI-49, " Interlocks and LCOs for Class 1E Tie

. Breakers," were raised by an incident that occurred at the Point Beach Unit P plant. On June 9,1980, it was discovered that a tie breaker between the safeguards buses at the plant was improperly left closed after a plant shutdown. The root cause of this event was attributed to personnel error. Approximately five weeks elapsed before the improper closure at Point Beach was discovered.

With the two breakers closed, the two redundant buses were con-nected and the independence of the buses was lost. If there had been a loss of off-site ac power with the tie breaker closed, interlocks would have prevented automatic closure of the diesel generator output breakers. This event was evaluated by the NRC staff, resulting in the identification of GI-49. GI-49 deals with procedural controls to reduce human error of the type that occurred at Point Beach Plant. The staff also noted that the tie breaker interlocks to prevent manual paralleling of standby power sources had not been implemented at the Point Breach Plant.

Mr. Thatcher indicated that the interlocks raised as a concern were to help protect against the potential for an operator making an error and inadvertently closing a tie breaker between either:

' two operating diesel generators which are potentially out-of-phase, or

• an operating diesel generator and an incoming feeder line -

which are potentially out-of-phase.

Although interlocks can crovide an additional degree of assurance for some infrequent situations, they can also have a potential negative impact on safety. PRA analyses have shown that cross connecting can allow for options that can prove to be beneficial.

f

)

AC/DC Power Systems. j Reliability Mtg Minutes June 7, 1989  ;

I The NRC staff after evaluating the concerns of GI-48 and GI-49 has concluded that these concerns can be generally resolved by the verification or implementation of appropriate limiting condition of  ;

operations in the plant technical specifications and by inclusion d of associated administrative controls in plant procedures for the l Class IE buses and tie breakers. The staff's primary objective is l to verify that plants are not being operated in violation of the design criteria of 10 CFR Part 50. Conditions identified by the q staff's evaluation suggestco a strong possibility that the single )

failure criterion may be violated for substantial time periods in some plants. The staff is proposing to issue a generic letter with related information requests pursuant to 10 CFR 50.54(f) to verify that the plant's safety design basis is being satisfit.d. The information requests would be related to the plant technical specification to include LCOs and surveillance requirements for vital instrument buses, inverters, onsite power sources, and tie breakers. If the licensees' response to these information requests is not satisfactorily, then the staff is requesting an explanation from the licensees. Mr. Thatcher indicated that this explanation can be accomplished by eithen (a)providinri '

ation and supporting analyses, or (b) by submitting an at % ..ent request proposing that appropriate LCOs be incorporated in the plant Tech Specs. Mr. Thatcher indicated that the concerns of GI-48 and GI-49 have been resolved in the recently licensed plants by implementation of Standard technical specifications and current licensing practice.

4. Mr. R. Haroldsen, Idaho National Engineering Lab (INEL), summarized the technical findings as a result of the study that was performed at INEL. The technical findings for GI-48 are as follows:

• Disagreement on the specific time limit that should be appli-cable l

l l

4

p .. ,.

g . .

(I N ,~

AC/DC Power Systems Reliability Mtg Minutes -6 . June 7, 1989 l ' The PRA analysis does not provide generically applicable guidance on the optimum time limit that should apply

• The current STSs for W and CE plants include restrictions derived fror.. practical considerations. These provisions are believed to be reasonable and adequate to resolve concerns.

• Administrative restrictions should apply; there should be a limit on the time that a vital instrument bus should be allowed to operate from a non-preferred power source.

'I The technical findings for GI-49 are as follows:

' Interlocks could provide backup to administrative control:

- Could prevent inadvertent paralleling of onsite power sources

- No action recommended due to:

-- belief that plants already contain interlocks related to tie breakers,

-- competing risks.

• Current STSs for W and CE plants include restrictions derived '

from practical considerations. These provisions are believed j to be reasonable and adequate to resolve concerns.

• Administrative restrictions should apply; there should be a l

limit on the L - a tie breaker should be allowed to be closed.

s N' L . .

L AC/DC Power Systems Reliability Mtg Minutes June 7, 1989 Mr. Haroidsen indicated that a survey of the technical specifica-tions revealed that:

-' Current Standard technical specifications for W and CE reac-tors include practical time restrictions on VIBs connected to non-preferred power sources and require tie breakers to be open as a limiting condition of operation. -

• A survey of plant technical specifications indicates.that 35 of 113 plants have the same restrictions on operations with VIBs connected to non-preferred sources as the Standard technical specifications.

• Twenty-seven plants of the 113 plants also require tie break-ers to be open during operation.

• The status of operating restrictions at other plants is unknown.

A more detailed information of the technical findings are docu-mented in the INEL Report EGG-NTA-7727, Rev. 3, that was issued on March 1989.

5. Mr. Thatcher briefed the Subcommittee in regard to GI A-30. "Ade-quacy of Safety-Related DC Power Supplies." The specific area of concern of Generic Issue A-30 is the adequacy of the safety-related de power in operating nuclear power plants, particularly with regard to multiple and common cause failures. Risk analysis and past plant experience support conclusions that failure of the de power supplies could represent a significant contribution to tha unreliability of shutdown cooling. Analysis indicates that in-adequate maintenance and surveillance and failure to detect battery

AC/DC Power Systems Reliability Mtg Minutes June 7, 1989 l

l- ,

unavailability are the prime contributors to failure of the de power systems.

Safety-related DC power is used for the overall operation of the safety-related portions of the electrical system includirg circuit breaker control for the AC power. It is typically also a source of vital AC power (via the. vital inverters) for safety-related in-strumentation and logic systems as well as operator indications.

During normal operation, the battery chargers supply the load requirements and maintain the batteries fully charged to be avail-able during loss of offsite power. For a loss of offsite power event, battery power is particularly important during the time period when the diesel generators are starting and immediately thereafter, because the circuit breaker control to sequence loads and the excitation of the generator field windings is entirely dependent on DC power.

Under a postulated station blackout (USI A-44) the batteries would be the only source of electrical power available. DC power, during this postulated event, is needed for operability of the steam driven auxiliary feedwater system in PWRs and for the RCIC and HPCI systems in BWRs. These are typically the only systems available to mitigate a station blackout event. All of these systems require DC power for operation of valves and for system control. Further, i instrumentation to inform the plant operators of plant status and to allow them to diagnose and correct the cause of the station blackout would be entirely dependent on the DC power supply system.

Mr. Thatcher cited three significant A-30 events:

(a) Palisades (1981) - in which human error caused tripping  !

of both vital batteries,

,~  :.-

AC/DC Power Systems Reliability Mtg Minutes June 7, 1989 4

(b) Hillstone 2 (1981) - loss of one vital DC power source resulting in loss of control room indication and alarm, loss of operability of switching equipment and one of the emergency generators failed during the event,  ;

(c) Zion 2 (1976) - loss of one vital DC bus resulting in a series of events (lot s of indicators and annunciators, loss of all cooling pumps, loss of plant controls, and a firedamagedoneofthediesels).

Mr. Thatcher indicated that the staff believes the most cost-effective approach to resolve this issue is for the staff to request certain information from all plants in order that the NRC can establish that adequate measures have been or will be taken'at all facilities. Only a portion of the measures would be reflected in a plant's technical specifications. The responses may indicate that in some cases improvements in DC system surveillance, mainte-nance and procedures are necessary.

The NRC staff is requesting a response from all holders of operat-ing licenses to the list of questions, supplied by the NRC, within 180 days of the date of the generic letter. If the staff review of l the licensee responses indicates that additional actions are l required, they will be backfit on a plant specific basis.

1 Mr. Thatcher outlined the relationship of GI-128 to other issues (e.g., USI A-44/ station blackout, USI A-17/ systems interactions, and USI A-47/ safety implication of control systems). For example, coping with station blackout relies on stored de power, de to ac conversion (inverters) needed for critical ac loads, and the capacity of batteries to be addressed by coping analysis. The relationship with USI A-17 and USI A-47 involves the number of significant events that have losses of vital ac and de, losses led l

w____-__--_-_-__ - _ _ _ _ - _ _ _ _ _ _ _

j: y ,

i AC/DC Power' Systems Reliability Mtg Minutes June 7, 1989 1

to plant transients and loss of safety equipment, and the improve-ments in de with vital ac and crossties can help reduce frequency H. 'and impact of power loss.

6. Mr. R. Haroldsen, INEL, summarized the technical findings as a result of the study that was performed at INEL for A-30, as fol-lows:

• NUREG-0666: Concluded that de power can be a significant contributor to shutdown unreliability:

- Better battery testing and maintenance

- Don't use tie breakers

- Stagger testing and maintenance

- Reverify adequacy of shutdown cooling with failed de power source

• NUREG/CR-4470: ORNL reviewed 251 instrument events (50 events involveddepower,9eventscausedtransients):

- More control room indicators

- More alarms

- Indication of bypassed de power source

• SOER-83-5: INPD reviewed past events and made recommendations:

- Review adequacy of de systems in all plants

- More alarms and monitors

- more maintenance and surveillance

• IEEE STD 450: Revised to include more maintenance, testing l and surveillance

AC/DC Power Systems

' Reliability Mtg Minutes June 7, 1989 A more detailed information e' the technical findings is contained ,

in the INEL Report EGG-NTA-8197 that was issued in March 1989,

7. As.a result of the Subcommittee discussion, some of the members and consultant expressed some concerns in regard to the following:

' Mr. Wylie commented that it is not clear what the NRC staff will do with all the information that they will obtain from the licensees in response to the proposed generic letter.

' Mr. Wylie indicated that the staff does not have a specific or a well defined acceptance criteria to judge the licensees performance in areas like maintenance, testing, and surveil-lance.

• Dr. Kerr commented that the staff should focus its efforts on developing and applying reliability theory to important plant system instead of obsolete criteria such as the single-failure criterion. Dr. Kerr cited as an example, is the resolut' ion of ,

GI-48, in which the staff concentrates on establishing LCOs to be incorporated into the plant technical specifications rather than on specifying a reliability goal.

l

• Dr. Kerr pointed out that the staff should consider an inte-grated approach resolution with interrelationships among other GIs and USIs, such as that proposed for the IPE program, instead of the fragmented approach. Mr. Davis shared the same sentiment.

• Mr. Carroll commented that the staff in its proposed resolu-tion of GI A-30 that deals with the adequacy of safety-related DC power supplies did not review in great details and defined a specific acceptance criteria for items such as the specific

~ - ~

. ,, :x .

- ~

,.._ +,

u.

1. AC/DC Power Systems Reliability.Mtg Minutes' ~ June 7, 1989 l

1 gravity in batteries,-ambient and battery temperatures, and DC power' supplies failure probabilities.

• Mr. Davis indicated that the' staff should focus on the overall' reliability of the system rather than the.AC/DC power supplies only. He comented that some plants can remove decay heat ,

without relying on the AC/DC system.

~

Mr.' Dav1s indicated that some' pump seal small LOCA analysis 1 depends on- the failure probabilities of the batteries, how-ever, most of the available PRA work that has been performed

.is of a generic nature and not a plant specific which could create a problem.

• Dr. Kerr commented that if the staff's proposed resolution to GI-128 relies on a plant specific nature, then that makes it not a generic issue.

1 Future Action

8. The Subcommittee members decided to write a letter on this issue concerning with the staff's proposed resolution, but with some l

, comments. This matter will be presented to the full Comittee on i 1

June 8-10, 1989, that includes a brief presentation from the NRC l

staff and the ACRS Subcommittee Chairman.

1 NOTE: Additional meeting details can be obtained from a transcript of this meeting available in the NRC Public Document Room,  ;

2120 L Street, N.W., Washington, D.C. 20006, (202) 634-3273,

~

l- or can.be purchased from Heritage Reporting Corporation, 1220 L Street, N.W., Suite 600, Washington, D.C. 20005,(202)  !

628-4888.

l e= __