ML20235A988
| ML20235A988 | |
| Person / Time | |
|---|---|
| Issue date: | 12/20/1988 |
| From: | Kerr W Advisory Committee on Reactor Safeguards |
| To: | Zech L NRC COMMISSION (OCM) |
| Shared Package | |
| ML20235A877 | List: |
| References | |
| ACRS-R-1341, NUDOCS 8901050196 | |
| Download: ML20235A988 (2) | |
Text
- _ _ _ _ _ _ _ - _ _ _ - _ . _ _ . _ __ - __ _ __ . _ - _ . . _ _ _ _ . _ - - _ _ _ _ _ _ __ _
e 8o* aseg'e,, UNITED STATES
! , e NUCLEAR REGULATORY COMMIS$10N
& . ,I ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 0,, WASH WGTON,0. C. 20666
.*g -
December 20, 1988 The Honorable Lando W. Zech, Jr.
Chairman U.S. !!uclear Regulatory Commission Washington, D.C. 20555 Deer Chairnan Zech:
SUBJECT:
E0VIPitEllT QUALIF10ATION-RISK SCOPING STUDY During the 344th meeting of the Advisory Comittee on Reactor Safe-guards, Decenber 15-16,19P8, we considered a report from our Subcomit-toe en Reliability Assurance pertaining to its review of the Equipment Ovalification (EQ)-Risk Scopine Study, performed by the Sandia National Latcratories (Sill) for the NRC Office of Fuclear Regulatory Research (RES). This matter was also discussed with representatives of RES and SNL during our 339th reeting, July 14-16, 1988 and during previous meetinos of nur Reliability Assurance Subcommittee on December 16, 1987; Jurt 14, 1988; and December 12, 1988. W'e also had the benefit of the docur,ert s referenced.
The purpose of this study was to assess the risk significance and risk uncertainties associated with current EQ requirements for safety-related electrical eovipment. The approach was to use information from existing TRAs to deternine what electrical equipment would be needed to prevent er r.itigate the consequences of a severe accident and, at the same time, would be exposed to a harsh environment related to that accident.
For the most part, the conclusions and recommendations from this study are plant specific. For this reason, the NRC staff proposed, and we i agree, that the insights from this study can be used in two ways:
- As items to be considered further in the Individual Plant Examina-tien and the Accident Management programs.
- As a means to limit or better focus the EQ inspections at existing plants.
One conclusion from the study is that the importance of the accident radiation dose in EQ is overemphasized. We believe that this warrants a review of some of the current requirements in Regulatory Guide 1.89,
" Environmental Qualification of Certain Electric Equipment Important to Safety for Nuclear Power Plants."
Another observation of general significance is that existing PRAs utilize equipment failure rates derived from experience in normal Attachment 10 g } @g <5 blQ ' 4
e l
The Hercrable Lando W. Zech, Jr. December 20 1988 operating environments.
There is no basis for confidence that these failure rates are realistic for equipment required to operate in a harsh envirennent, whether resulting from a severe accident or a design-basis accident.
The significance to risk of this observation is potentially irportant and deserves further study.
TFe follewirg comments do not relate specifically to this study but do relate to the preccss of managing research.
We believe that review of the study by a four-person peer review panel contributed sigr.ificantly to the credibility of the conclusions and the quality of the final report. We were favorably impressed with the enertise end variety of the panel members and the way they interacted with the rescerch team. '
Finally, we sucgest that had a risk-hsed scoping study been performed befere the EQ research program was begun, rather than after it was corplcted, the nature and scope of the program might have been differ-ent , and presurebly better. While such a study should not be expected to dictate all aspects of a large, ccmplex.research program, it should help tc distinguish between the clearly important and the clearly unimportant, end perhaps ever between what is knowable and what is not.
Sincerely, William Kerr Chairman i
References:
- 1. EC-hisk Senpine Study, Oraft Final Report - L. D. Eustard, Sandia I;ational Laboratories, A. M. Kolaczkowski, G. T. Nedford, and J.
Clark, Science Applications International Corporation, dated May 1988
- 2. Letter from George E Sliter, Electric Power Research Institute, to Meni Dey, NRC Dffice of Nuclear Regulatory Research,
Subject:
EQ-Risk Scoping Stud /' Peer Review, dated June 10, 1988 :
- 3. Letter from K. S. Canady, Duke Power Company, to Moni Dey, NRC Dffice of Nuclear Regulatory Research,
Subject:
EQ-Risk Scoping Study Peer Review,idated June 28, 1988 4 Letter from S. P, Carfagno, Franklin Research Center, to L. D.
Bustard, Sandia lational Laboratories,
Subject:
EQ-Risk Scoping Study, dated June 15, 1988
- 5. Draft Final Comments, Peer Review of the Sandia National Laboratory Equipment Qualification-Risk Scoping Study, A. J. Wolford, H. L.
Fagleby, EGl.G Id4ho, Inc., dated June 1988
- 6. EQ-Risk Scoping Study: ,
dated)
Discussion of Peer-Review Coments (un- l i
a ..
i., . .
[p neg'o UNITED STATES y -
^
NUCLEAR REGULATORY COMMISSION 3
I ADVISORY COMMITTEE ON REACTOR SAFEGUARDS
. ! wAsmorow. o. c.nossa gv /
....+
January 19, 1989 1
The Hencrable Lando H. Zech, Jr.
Chairman U.S. Iluclear Regulatory Commission Washington, D.C. 20555
Dear Chairman Zech:
SUBJECT:
P. ARK I CONTA1WENT PERFORMANCE IllPROVEMENT PROGRAM During the 345th meeting of the Advisory Committee on Reactor Safe-guards, January 12-14, 1989, we heard a staff presentation on their tref t preposed paper entitled, " Mark I Containment Performance Improve-ment Program." This metter was also discussed during our 344th meeting, Decerber 15-16, 1988, and by our Subcennittee on Containment Systems on Decenber 6, 1988. The Committer- riso discussed this issue during neetirgs held in July and December 1986 and Decenber 1987. We also had the benefit of the documents referenced.
Park I BUR plar.ts have beer, singled out for special attention because of indications of a high probability that their containments may not be effective in nitigeting severe core melt accidents. A conditional failure probability as high as 90 percent has been estimated for some accident secuences. Even if this estinate is shown to be valid, indi-cations are that these plants de not present a risk to the public that is larger than average for LWRs, because the probability of core melt (i.e. , the need for containment) is calculated to be lower than is typical fer PWRs.
The package of improvements for Mark I plants being proposed by the staff is primarily directed toward lowering the probability of core melt withcut changing ncticeably the conditional probability of containment failure. The staff has documented estimates of a factor of five to ten l- in the reduction of core melt probability due to internal accident initiators for plants that incorporate the proposed recommendations.
Estinates of improvement in containment performance have not been calculated, although there are statements that the probability of failure will be reduced. He were told in an oral presentation that the !
improvements might reduce the conditional probability of failure to less i I
than 50 percent. It was emphasized that this was only an estimate.
We have previously expressed our opinion that the Cocrnission's safety goal is an appropriate standard for establishing how safe plants should be. We also have suggested in our letter dated April 15, 1986, that an implementation plan for the safety goal should provide a framework for Attachment 11
c .
The Honorable Londo W. Zech, Jr. January 19, 1989 assuring that plants have adequate defense in depth as well as tssuring that they neet quantitative risk standards. As a class, Mark I plants, es indicated by several PPAs for particular plants, appear to conform to the quantitative risk standards. These plants may not have an appropri-ate belance between prevention and mitigation.*
On the basis of a linited analysis of the potential costs and benefits of the proposed inprovenents, the staff concludes that the improvements are gertrally cost beneficiel and are thereby justified for all 24 Mark I plants. We do not agree. A number of assumptions used in the analy-sis seen not to provide e fair and balanced comparison of potential ccsts and berefits. It appeers to us thet there would be a wide varia-tion ir. the conclusions if the ar.alysis were done for each individual plant.
We cercluda that no risk-br. sed reason has been identified which justi- i fits sincling cut l' ark I plants from the general population of LWRs.
There is e procran to look at all plents to identify any pessible " risk cutliers." This is the Ir.dividual Plant Examination (IPE) pregram. We ,
believe that Mark l plants should be analyzed as a part of tFis prcgram art thet vulnerabilities in individuel plants can thereby be identified, analy2rd, and corrected where necessary.
Pe reconrrrd that the preposed irprevement plar for Mark I containments Le dropped se that liter.see and llRC resources can concentrate on the mort effective IPE approach.
Sincerely, Forrest J. Remick Chairman
- For this discussion, we define prevention as those activities intended I to keep the core from melting, and mitigation as those activities intended to keep fission products released from a melted core away from the public.
References
- 1. Draf t Proposed Paper from V. Stello, Jr., Executive Director for Operations, to the Commissioners,
Subject:
Mark I Containment Performance Improvement Program (Predecisional), received January 13, 1929
- 2. V. S. Nuclear Regulatory Commission, Generic Letter 88-20, "Indi-vidual 51rnt Examination for Severe Accident Vulnerabilities - 10 CFR 50.Sa(f)," Novenber 23, 1988 l
_ _ __-___-___________ - ____- __ _ __ _