ML19289F812

From kanterella
Jump to navigation Jump to search
Summary of ACRS 790510-12 Meeting Re Review of Accident at TMI
ML19289F812
Person / Time
Site: Crane Constellation icon.png
Issue date: 05/16/1979
From: Carbon M
NRC OFFICE OF PUBLIC AFFAIRS (OPA)
To: Hendrie J
NRC COMMISSION (OCM)
Shared Package
ML19289F809 List:
References
NUDOCS 7906200035
Download: ML19289F812 (6)


Text

.'

i h #8 gk UNITED STATES

+

[Y (7 i NUCLEAR REGULATORY COMMISSION

k. / j t

?-

ADVISORY COMMITTEE ON REACTOR SAFEGUARDS

/

waswisoros, c. c. rosss Fay 16, 1979 Ronorable Joseph M. Bendrie Chairman U. S. Nuclear Regulatory Commission Washington, DC 20555

Subject:

IhTERIM REPORT NO. 3 ON 'nIREE MILE ISL'.ND N'ILEAR S'IATION UNIT 2

Dear Dr. Hendrie:

During its 229th meetirg, May 10-12, 1979, the Advisory Ccrraittee on Reactor Safeguards continued its review of the recent accident at Eree Mile Island Nuclear Station Unit 2 (DiI-2), including implications drawn f rom the occurrence of this accident. The Cocmittee has several a$di-tional recocraendations to make at this time.

Reactor Pressure Vessel Level Indication ne Committee believes that it would be prudent to consider expeditiously the provision of instrumentation that will provide an unambiguous indica-tion of the level of fluid in the reactor vessel. We suggest that licens-ees of all pressurized water reactors be requested to submit design pro-posals and schedules for accocolishirg this action. his would assure the timely availability of reviewed designs if the Staff orgoirg studies should indicate that early implementation is required. %e Cortraittee believes that as a minimum, the L vel indication should range from the bottom of the hat leg piping to the reactor vessel flarge area.

Operator Trainire and Qualification The NRC Staff should exandne operator qualifications, trainirg, and li-censire to determine dat changes are needed. Consideration should be given to educational background, to trainirs rethods, and to content of the trainire program. Attention should also be given to testing methods, with specific concern for the ability of the testirg methods to predict operator capability.

Examination of licensirg procedures should deter-nine whether they are responsive to new inforcation that is developed about plant or operator perforrance. Effort should also be rrde to determine whether results of examirations can be correlated with oper-ator ability. Requalification trainirg ard testirg should be similarly 7906200035 233 009

Honorable Jose;h M. Hendrie May 16, 1979 examined to insure that they the account of informaticn that 10 devel-oped by operation in the plan', and to determine that relevant informa-tion atout other plants is made available to operators, arx! is made part of the training arx3 requalification progrun. As part of this and of other core extensive stixlies, continuing attention must be given to the amount of information which an operator can assimilate arri use in normal and in emergency situations and to the best method of presentirra the in-formation to the operator. We use and limitations of simulators for operator training should receive careful consideration.

Evaluation of Licensee Event Reoorts Because of the potentially. valtable information contained in Licensee Event Reprts (LERs), the Committee recoranends that the NRC Staff estab-lish formal procedures for the use of this information in the trainirg of supervisory arx1 raintenance staffs and in the licensirq and requali-fication of operating personnel at commercial nuclear power plants.

We inforration in LERs may also be useful in anticipatirg safety problems.

At the present time some utilities routinely request that they be pro-vided capies of all LERs applicable to plants of the type they operate or to specific systems and components in a given class of plants similar to their plant. Certain reactor vendors have made similar requests arx5 use the LERs to review and evaluate the parformance of their plants.

In addition, the NRC operator licensing staff has indicated Eat they use LEF in reviewiry operating experience at ccxtmercial facilities.

h e large number of LERs that attribute the cause to personnel error would tend to indicate that a formalized program of IIR review would be useful in the training, licensirg and requalification of ntx-lear pwer plant personnel. We extent to which such a program could be used to anticipate sefety problems shoulc: also be considered.

Operating Procedures Safety aspects of individual reactors durirq normal operation and under accident conditions are reviewed in detail by the NRC Staff and discussed with the ACRS. Acceptable limits for normal operations are formalized by Technical Specifications, sutnitted by the licensee arx3 approved by the NRC Staff. Operating procedures for severe transients have received less detailed review by the NRC Staff.

It appears that such procedures would benefit frce review by an interdisciplirary team which includes personnel expert both in operations and in system behavior. Also, for the longer tern, there may be merit in considerirg the development of more standard-ized formats for sach procedures.

233 010

Honorable Joseph M. Bendrie May 16,1973 Reliability of Electric Power Supplies During the past several years there have been several operatirg exoeri-ences involvirg a loss of AC power to important engineered safeguards.

The ACRC believes it important that a comprehensive reexamination be made by the NRC and the reactor licensees of the adequacy of design, testirg, and maintevnce of offsite and cnsite AC and D' power supplies.

In par-ticular, f silure codes and effects arelyses should be made, if not al-ready perft.tmed, core systematic testing of pwer system reliability, in-cluding abnormal or anomalous system transients, should be considered, and improved gaality assurance and status monitorire of power supp: y systems should be sought.

Analysis of Transients The ACRS recernends that each licensee and holder of a construction permit be asked to take a detailed evaluation of his current capability to with-stand station blackout (loss of offsite and onsite AC power) including additional complicating factors that might be reasonably censidered. The evaluation should include examination of natural circulatim capability, the continu1rg availabil'.ty of components needed for long-term cooling, and the potential for improvement in capability to survive exterded sta-tion blackout.

The AGS also recocmends that each licensee and construction permit holder should examine a wide rarge of anomalous transients and degraded accident conditions which might lead to water hanner. Methods of controlling or preventirg such r,ditions should be evaluated, as should research to provide a better basis for such evaluations. The Committee expects it would be appropriate to have such studies dona generically first, for classes of reactor designs and system types.

Emeroency Plannivg An effort should be undertaken to plan and define the role NRC will play in energencies ard what their contribution and interaction will be with the licensee and other emergency plan participants including other govern-ment agencies, industry represmtatives, arxl rational latnratories. Sr.h olannirq should consider:

assurance that formal documentation of plans, procedures and organization are in place for action in an emergerry, designaticn of a technical advisory team with names ard alternates for the anticipated needs of an emergency situation,

Honorable Joseph M. Hendrie May 16,1979 compilation of an inventory of equi;raent and materials which may be needed for unusual conditions includity its description, location, availability and the organi-zation which controls its release.

We Cecraittee recocraends that each licensee be asked to review and revise within about three renths:

his bases for obtainirn offsite advice ard assistance in emergencies, from within and outside the coruny, current bases for notifyirg ard providirg information to authorities offsite in case of emergency.

This review ard evaluation should be in tems of accidents havirg a broad rarge of consequences. The results of this review should be reported to the NRC.

Decontamination and Recovery The Comittee wishes to call attention to the in.portance of a program de-signed to learn directly about the behavior, failure modes, survivabil-ity, ard other astects of component and system behavior at IMI-2 as part of the long-term recovery process. his program should also exanine the lessons learned at IMI-2 to determine if design charges are necessary to facilitate the decontanination and recovery of major nuclear power plant syste=s.

Safety Review Procedures ne TMI-2 accident has imposed large new pressures on the availability of manpower resources within the NRC Staff.

If progress is to be expeditad on the new questions which have ariscn and cn existirg unresolved safety.

issues, the ACRS believes that new mechanisms should be sotght and imple-mented. For those safety concerns where such a mechanism is a;propriate the Cwmittee reconmerxis that the Conraission should request licensees to perforra suitable studies on a timely basis, includirg an evaluation of the pros and cons, and proposals for possible implementatica of cafety improvements. The NRC Staff should concurrently establish its own capa-bility to evaluate such sttxiles by arranging for suppo. t by its consult-ants ard ccintractors. In this fashion, the Coccittee anticipates that the informtion cn htlich jtx3 ::ents will be based can be developed nr-h 9

trore expeditiously, ard an earlier resolution of many safety ecncerns may be achieved.

733 0'i2

Honorable Joseph M. Herdrie Ksy 16, 1979 Capability of the NRC Staff he Comittee recomends that the capability of the NRC Staff to deal with basic and engineetirs probleras in what may be temoed broadly as reactor a2d fuel cycle chemistry be aupented expeditiously. 21s should inc1tde establishment of expertise withir the KRC, with assis-tance arrarged frcra consultants and contractors, in such important technical areas as the behavior of M ard DWR coolarrts ard other ma-terials under radiation conditions; generation, harrn frvy and disposal of radiolytic or cther hydrogen at nuclear facilities; performance of various chemical additives in contairnent sprays; proczessing and disposal technigaes for low and high level radioactive wastes; chemical operations in other parts of the nuclear fuel cycle; ard in the chemical treatment operations involved in recovery, decontamination, or ciecomissioning of nuclear facilities. We Ccranittee wishes to emphasizes the importance of providirg this expertise in both the research and lier =2sirg maragenent element, of the NRC.

Single Failure Criterion The NRC should begin a scudy to determine if use of the sirgle failure criterion establishes an appropriate level of reliability for reactor safety systens. Operatirs experience suggests that axiltiple failures a.1 coxon mode failures are encountered with sufficient frequency that ti.cy need rore specific consideration. Eis study should be accompanied by concurrent considerat :an of how the licensing prow can be modified to take account of a new set of criteria as appropriate.

Safety Research The ACRS believes that, as a result of the 'IMI-2 accident, various safety research areas will warrant initiation or much greater emphasis, as ap-propriate. De Corraittee suggests that consideration be given to an atg-mentation of the NRC safety research budget for FY 80.

Also, the Ccenittee believes that a larger part of the safety research program should be oriented toward exploratory research as contrasted to confirmatory research, with scoe degree of freedcm from immediate licens-irg regairements. We ACRS plans to have a Subcomittee meetirg cn this subject with representatives of. tne NRC Office of Nuclear Regulatory Research in the near future.

De Ccmittee is continuirg to review these matters and will report fur-ther as additional recocmendations are developed.

'g3 013

Honorable Joseph M. Hendrie May 16, 1979 Additional connents by Messrs. H. Lewis, D. Maeller, D. Okrent, and J Ray are presented belnw.

S neerely, Fax W. Carbon Chairman Additional Coments by Messrs. H. Lewis, D. Moeller, D. Okrent, an6 J. Ray The potential for i redtx: tion in risk to the p blic in the case of a ser-ious reactor accident by the ieplementation of a neans for controlled, filtered ventiry of s containment which could retain prticulates and the bulk of the iodine has been recognized for more than a decade. We concept was reconrnended for sttx3y more recently in the American Physical Society Reprt on light-water teactor safety and in the Ford Foundation-Mitre Report, ' Nuclear Power - Issues and Choices." It is a high pri-ority item in the NRC plan sutnitted to Congress for Research to Improve the Safety of Light-Water Nuclear Power Plants (NtraEG-0438). We sttx3y performed for the State of California on underground siting concitxled that filtered, vented containment was a favored optim to explore in con-nection with possible means to mitigate the consequences of serious re-actor accidents. Ibever, little progress has been made on the develop-ment of sufficiently detailed design information on which to evaluate the efficacy and other factors relevant to a decision on Fossible implementa-tion of such consequence amelioratirq systens.

The 'IMI-2 accident suggests that the probability of a serious accident in which a filtered vented containment could be useful is larger than many had anticipated.

We recormeM that the Co mission request each power reactor licensee and.

construction permit holder to perform design sttx3ies of a system which adds the option of filtered ventirg or purgirq of containment in the ever.t of a serious ace! dent. 'Ihe system should be capable of withstand-ing a steam and hydrogen environment aM of recoving and retainirs for as long a time as necessary radioactive particulates and the great bulk of the iodine for accidents involvire degraded situations up to and in-citx31ryg core melt. Such sttx3ies could be done generically for several reactor-contairnent types, and should evaluate the practicality, pros and cons, the costs, aM the patential for risk reduction. A period of about twelve months for a report to the NRC by licensees and constructim permit holders appears to represent a possible schedule.

233 014