ML20246F955
| ML20246F955 | |
| Person / Time | |
|---|---|
| Issue date: | 04/06/1989 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| References | |
| REF-GTECI-101, REF-GTECI-NI, TASK-101, TASK-OR ACRS-2633, NUDOCS 8905150150 | |
| Download: ML20246F955 (6) | |
Text
.
DATE ISSUED: 4/6/89 CORPM33 l
f])):.5/uk/fY 1
ACRS MEETING MINUTES /
SUMMARY
OF THE INSTRUMENTATION AND CONTROL SYSTEMS SUBCOMMITTEE MARCH 29, 1989 BETHESDA, MD PURPOSE The purpose of this Subcommittee meeting was to review the NRC staff's proposed resolution to Generic Issue 101, "BWR Water Level Redundancy."
ATTENDEES ACRS pg W. Kerr, Chairman R. Baer, RES H. Lewis, Member W. Minners, RES C. Michelson, Member D. Persinko, NRR C. Wylie, Member A. Szukiewicz, RES P. Davis, Consultant D. Thatcher, RES W. Lipinski, Consultant M. El-Zeftawy, Staff Others B. Collins, INEL M. Check, NUS W. Smith, Bechtel h0l I
MEETING HIGHLIGHTS, AGREEMENTS, AND REQUESTS 1.
Dr. Kerr, Subcommittee Chairman, stated the purpose of the Subcom-mittee meeting and introduced the other present ACRS members and consul,tants.
i Dr. Kerr indicated that he would like to know the criterion by which the NRC staff determines when an issue has been resolved or DESIGNATED ORIGINAL k [ Aggy 890406 certified By 2633 PDC i
i Instrumentation'and Control Systems Meeting Minutes March 29, 1989 what determines a fix. He also asked the NRC staff for their raisonaletoignorecommonmodefailuresintheriskanalysis.
Dr.:Kerr also, added that he would be interested to know when the utilities do the IPE, will it be assumed that this issue has been resolved, or will they also be asked to look at sequences that might involve the water level.
2.
Mr. R. Baer, NRC/RES, summarized the staff's resolution of this GI-101.. He indicated that this generic issue had a postulated water level instrument line failure event. The failed instrument line can provide a false hi-level signal that results in a.re-duction of main feedwater flow and also for certain designs can defeat the automatic signal that starts the RCIC or the HPCI depending on which particular level line is broken.
If the failure is outside containment there would be no increase in containment pressure, and for certain plant designs, there would not.be an automaticdepressurizationsystem(ADS) signal.
Mr. Baer indicated that, the contractor, INEL, perfonned a PRA-type analysis and not a deterministic single failure analysis. The conclusions were that core melt frequencies attributable to sensing L
line break were relatively low (less than 10-6/RY).
In the cost-benefit analysis, the staff used a $1000 man-rem as a guideline to make a decision. The staff determined that there is no favorable cost benefit ratio, and concluded that no backfit or other action can be justified.
l 3.
Mr. A. Szukiewicz, NRC/RES, indicated that Generic Issue 101 l'
focuses on the ability of BWRs to mitigate an instrument sensing line leak or break that could affect both the control and pro-tection systems. A previously resolved issue, GI-50, " Reactor Yessel Level Instrumentation in BWRs," addressed several areas of
- _ _ _ _ = _ - _ _ _ _ _ _ _ _ _ - _ _ _ _
=.-
a.
Instrumentation and Control Systems Meeting Minutes March 29, 1989 concern that also involved BWR level instrumentation.- Resolution of GI-50 involved improvement to reactor vessel water level instru-mentation. These improvements included preventing overheating'of the reference legs, reducing the sensing length in drywell and, improving level indication accuracy. The potential for break or leak in an instrument sensing line in conjunction with a single failure was excluded from GI-50 and designated as a' separate issue, i.e., GI-101.
BWR water level measurement systems have provided adequate water level information for various conditions of reactor operation.
I However, there have been several incidents where interactions between plant control systems and protection systems have occurred due to the level measurement systems. To various degrees, BWR designs are vulnerable to these interactions since sensors and instrument sensing lines that monitor vessel level are comon to both the protection systems and the non-safety-related control systems. A search of operating experience has not identified the occurrence of any instrument line breaks. However, there have been 10 instances of small instrument line leaks. A break or signif-icant leak in the instrument sensing line connected to the refer-ence leg would cause a false reactor vessel high water level indication. This would automatically reduce the feedwater flow into the vessel and cause the actual water level in the vessel to decrease. The presence of this false high level indication may also prevent automatic operation of the emergency safety systems, such as HPCI/HPCS or RCIC in some designs, and could confuse the l
operator in assessing the actual water level in the vessel.
To address this concern, an evaluation was perfonned by EG&G (Idaho) on the reactor water level instrumentation systems. All the technical findings are documented in NUREG/CR-5112.
I i
x.
[ l*
}
Instrumentation and Control Systems Meeting Minutes March 29, 1989 The NRC staff concluded, based on the low probability of core melt
-and the high cost / benefit ratio to reduce public risk, that the resolution of GI-101 does not call for any additional-actions for BWR licensees and applicants. The staff believes that emergency procedures do exist on all plants to mitigate the consequences of instrument line breaks.
Although the staff is not proposing any actions, the staff is issuing a generic letter to all licensees and applicants of BWR plants. The generic letter and NUREG/CR-5112 are provided for information only.
Each BWR plant licensee is expected to review the information to verify that the design of its facility has been-correctly. represented and maintain appropriate procedures and operator training.
4.
Mr. B. Collins, Principal Investigator /INEL, summarized the study that was performed by INEL. The study was designed to evaluate cil current BWRs for consequences of postulated instrument sensing line i
breaks associated with the water level instrumentation, combined with an independent single failure in the reactor protection system. All BWR plant designs were placed into one of five groups, based on their system characteristics (i.e., BWR 2, 3, 4, 5, or 6).
A plant from each group was evaluated assuming an instrument i
sensing line break combined with an additional single failure. The l
study was performed to determine if a break and a single failure would obstruct:
(a)reactorshutdown,(b)decayheatremoval,or (c) reactor coolant inventory control. Six scenarios were iden-tified to be of some potential safety concern. These six scenarios were analyzed for frequency of occurrence and the potential of leading to core melt. Those events are considered applicable to BWR-2 and BWR-3 designs and to the older vintage BWR-4 designs. No significant failure sequences were identified for the more recent BWR-4 designs or for the BWR-5 and BWR-6 designs. Those six
Instrumentation and Control Systems Meeting Minutes March 29, 1989 sequences were then subjected to a value/ impact analysis, where possible plant modification alternatives were examined on the basis of risk reduction and cost.
Mr. Collins indicated that based on the technical evaluations and value/ impact analysis, none of the alternatives identified has shown a favorable cost / benefit ratio.
5.
As a result of the Subcommittee discussion, some of the members and consultants expressed some concerns in regard to the following:
Dr. Kerr indicated that he is interested to know whether or not the NUREG-1150 plants looked at this issue, and'if there is a risk contribution from malfunction of the water level indicator that one can identify in the 1150-plants. The staff responded that they will get a definite answer regarding this question.
Mr. Davis expressed some concern regarding the methodology.
He indicated that BWR PRA's do not generally examine instru-ment line breaks as an initiating event. They use small break and large break LOCA's only.
In view of that, he questioned whether the IPE's will ever pick this issue up and resolve it.
1
" Mr. Michelson expressed some concern in regard to the limita-tion of the scope of the issue.
However, he added he agrees with the staff's resolution for this narrow scope. Dr. Lewis L
also expressed some discomfort from narrowness of scope.
I Mr. Wylie indicated that earlier studies that were performed l
on water level indication and effects on control production l
systems did not consider common-modo failures. Dr. Kerr shared the same concern.
L_______--__
V, o
i Instrumentation and Control Systems Meeting Minutes March 29, 1989 I
_
- Mr. Davis expressed some concern that the staff in analyzing the failures did not consider environmental concerns.
Mr. Lipinski stated that the generic letter proposed by the staff should ask the individual plant-owners to. address the common-mode failure issue.
Dr. Kerr commented that the current analysis is based on backfitting and for existing plants, with no recommendations being made for designers of new plants.
Dr. Lewis questioned the sensitivity study that was performed by INEL.
It was used by the NRC staff cs an uncertainty
{
study.
He cautioned that sensitivity study and uncertainty study are two different things.
- Dr. Kerr indicated that it is not clear that how the people who perform IPE's and attempt to settle generic issues can deal with this issue.
FUTURE ACTION 6.
The Subcommittee members recommended that a letter be written on this issue concurring with the NRC staff's proposed resolution to' this generic issue, but with some comments. This matter will be presented to the full Committee on April 6-8, 1989, that includes a I
brief presentation from the NRC staff and the ACRS Subcommittee Chairman.
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, or can be purchased from Heritage Reporting Corporation, 1220 j
L Street, N.W., Suite 600, Washington, D.C. 20005,(202) 628-4888, i