ML20151D256
| ML20151D256 | |
| Person / Time | |
|---|---|
| Issue date: | 02/10/1988 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| References | |
| TASK-A-45, TASK-OR ACRS-2550, NUDOCS 8804140068 | |
| Download: ML20151D256 (29) | |
Text
r-
, g gy -
DATE ISSUED: 2/10/P8 gg$ cR656 Advisory Comittee on Reactor Safeguards Decay Heat Removal Systems (DHRS)
Subcommittee Meeting Minutes January 28, 1988 Washington, D.C.
PURPOSE: The purpose of the meeting was for the Subcomittee to continue its review of the NRC Staff Resolution Position for USI A-45:
"Shutdown Decay Heat Removal Requirements."
ATTENDEES: Principle meeting attendees included:
ACRS NRC D. Ward, Chairman J. Mazetis, RES J. Ebersole, Member W. Minners, RES C. Mark, Member E. Chelliah, RES C. Michelson, Member C. Siess, Member ERC T '1ric .
C. Wylie, Member D. Ericson I. Catton, Consultant P. Davis, Consultant P. Boehnert, Staff MEETING HIGHLIGHTS, AGREEMENTS, AND REQUESTS
- 1. Mr. Ward noted that resolution of USI A-45 has been underway for a long time, He said the Subcomittee has not yet received any documentation supporting a resolution position, although the ACRS has had some inkling of the NRC Staff position.
The Chairman said, depending on what we hear today, he may recom-mend ACRS revieti at the March 1988 Meeting, even if documentation is not available, eg41 60 080210 g
2550 pop
, e DHRS Meeting Miriutes January 28, 1988 Mr. Ward referred to a Nuclear Safety Analysis Center - NSAC/W OG (0wners Group) Report (NSAC-113) which was an Industry review of the Point Beach Plant Analysis Report perforwed for NRC by Sandia Laboratories. He said DHRS Subcomittee Consultant Mr. P. Davis reviewed this document at the Subcomittee Chairman's request. The Subcomittee has received a copy of Mr. Davis' written report detailing his critique.
Mr. Michelson requested NRC to define what they mean by a "dedicat-ed DHR system," as no one has really defined such a system to his satisfaction, at least. Mr. Ebersole said such a dedicated system will have performance boundaries that should also be identified.
- 2. W. Minners (RES) provided introductory remarks for the NRC Staff.
He indicated that the Staff is providing a preliminary resolution position at this time. Mr. Minners also sai,d.the infomation on the RES review of installation of a the NSAC-113 report is also
- preliminary. In response to Mr. Michelson, Mr. Minners said the issue of installation of a dedicated DHR system is whether or not it is worth the expense involved.
- effort. The background, scope, and objectives of this USI were reviewed (figs. 1-3). Points noted during the above discussion included:
- In response to Mr. Ebersole, Mr. Mazetis said that the Staff's current resolution approach (see below) was chosen because of
- the plant-specific nature of DHR vulnerabilities.
- In response to Dr. Siess' question on an objective of USI A-45 (establish new requirements, if necessary, to reach DHR l reliability compatible with CDF (core damage frequency) goal),
l t
s .
DHR$ Meeting Minutes January 28, 1988 Mr. Minners said NRC cannot distinguish a conditional proba-bility of core damage versus core melt as the PRA's don't address this point. RES said the objective of core melt research is to develop procedures for accident management. In response to Mr. Ward, Mr. Mazetis'said the CDF being used by the Staff for A-45 is 10-5/RY. Mr. Ward said the safety goal has(tentatively)adoptedaCDFof10'4/RY. Mr. Mazetis said the 10-5 number is compatible with the IPE (Integrated Plant Examination) risk parameters.
Dr. Siess asked if better guidance is expected from the Comission on just what core melt frequency should be used in Staff safety evaluations. Mr. Mazetis indicated that such guidance would be welcome, as the Staff is struggling with this problem. Dr. Siess said it seems that the "Safety Goals" are being set "ad hoc" by different Groups in NRC.
Mr. Davis indicated that there are few PRA's (if any) that show CDF's at or near 10-5/RY and this will give NRC problems. Further discussion suggested that some plant modifications may be required if a CDF of 10-5/RY is a "real" goal.
- Mr. Michelson raised concerns with the NRC analysis of sabo-tage. Mr. Mazetis said details of this concern would be better discussed at a future Subcomittee Meeting. Mr. Ward said the Subcomittee should revisit this Item at the end of this Meeting for potential follow-up.
RES discussed the six PRA DHR Plant Analysis Reports (Case Studies) perfonned for NRC by Sandia. These Reports are now all complete.
The Report Sections dealing with Sabotage will be available, in camera, in the near future. In addition, an overall Sumary Repore
- .* e DHRS Meeting Minutes January 28, 1988 -
for the Case Studies will also be issued in the near future.
Figure 4 lists the key findings from the Case Studies. Overall, the Studies show that the relative importance of vulnerabilities is plant-specific. In response to Mr. Michelson, Mr. Minners indi-cated that some fixes (for fire threat for example) may deserve a relook, given our present knowledge base.
The proposed NRC draft Regulatory Analysis for A-45 listed six Alternatives for resolution. These are:
(1) No Action - Would be acceptable j_f,f NRC analysis results are found to be overly conservative (e.g., NSAC-113 for Point Beach).
(2) Limited Scope PRAs Severe Accident Program IPEs.
(3) Specified Systems Modifications - USIs and G!s.
(4) Depressurization and Cooling - PWR (feed and biced), and BWR (containmentventing).
(5) Dedicated hot shutdown capability. -
(6) Dedicated cold shutdown capability.
It was noted that Alternative 1 (No Action) may be acceptable if the CDF goal was increased to 10-4/RY (from 10-5/RY), or, if upon NRC Staff review, the EPRI/)'0G Report did find that the Sandia Case Studies results were overly conservative. The above work in this area is still on-going. Regarding Alternative 4. Mr. Ebersole noted feed and bleed for PWRs is not as comprehensive as BWR containment venting (since BWR's can inherently "feed and bleed.").
Mr. Ward asked if the variability of F&B to impact the CDF in the
DHRS Meeting Minutes January 28, 1988 Case Studies was due to the system limitations or the accident scenario of concern. Mr. Mazetis said it was more the latter.
The Staff's value impact analysis that accompanied the above Alternatives was performed three different ways: (1) averted offsite costs only, (2) averted offsite plus onsite costs, and (3) all the above, plus effects of special considerations (e.g.,
sabotage, moratorium, resolution of other generic issues, and "unquantifiables"). Depending on the approach used, the cost effectiveness of the Alternatives varied, (e.g., Alternatives 5 and 6werecosteffectiveusingapproach(3)). Mr. Ebersole asked if the costing of Alternatives 5 and 6 looked at such cost cutting approaches as "no QA" or no "Safety Grade" requirements, etc. NRC said no, since the V/I analysis requirements dictated that the above Items be considered. Dr. Mark questioned the rationale of saying such items as sabotage, nuclear morit.oria, etc., can be "cost effective" as these are really subjective value judgments.
NRC endorses Alternative 2 as a resolution approach for the follow-ing reasons:
- A-45 case studies showed most risk contributors to be plant- -
specific.
- Use of "Method 3" (credit for "moratorium avoidance") goes beyond value/ impact methods previously used for USIs/GSIs and therefore Alternatives 5 and 6 can not be justified on the basis of conventional value/ impact analyses.
- Insights gained from six case studies and EPRI-}!0G analysis (plus NRC/Sandia review of same) will become guidance to
[
licensees, as the severe accident program advances.
DHRS Meeting Minutes '
January 28, 1988 In response to Mr. Ward, NRC said they have not done any additional work in support of the technical findings supportino the Regulatory Analysis; rather, the basic conclusion has been changed. The next revision of the Regulatory Analysis should be available to the Subcomittee in 30 to 60 days. In response to Dr. Siess, Mr.
Mazetis said the Regulatory Analysis will follow the dictates of the Backfit Rule. NRC also said the Comission will be presented with all six Alternative approaches for consideration. In response to Mr. Ward, Mr. Minners said Method 3 of the V/I analysis is not outside the dictates of the Backfit Rule requirements.
Details of the implementation of Alternative 2 were noted. Plant-specific PRA's would be required. A point of debate in the Staff is whether the PRA should be part of the IPE approach for severe accident analyses or be a "stand alone" PRA. Presently, the Staff is leaning toward incorporating the "A-45" PRA into the IPE Program.
NRC is reviewing the EPRI/WOG Point Beach PRA (NSAC-113) for two central reasons: (1) it provides guidance for a Utility performing a PRA to resolve A-45; and (2) it may result in justifying Alterna-tive1(NoAction) fora-45. (However, NRC does not at this time believe (2) will turn out to be the case.) NRC and Sandia will discuss the details of the on-going review. Mr. Ward expressed concern that Item 2 is really "doable". RES said they need to be convinced that Item 2 is doable as well: i.e.,we(RES) don't think Alternat.ive 1 can be justified.
- 4. E. Chelliah (RES) discussed the results of the NRC/RES review of the NSAC-113 analysis of the Sandia Point Beach Study. The results l
for the various accident sequences analyzed were detailed (Figs.
! 5-8). Mr. Michelson questioned the bases for the LOCA frequency
- value used in the EPR!/)'0G and Sandia Studies. Mr. Davis said the n
, i DHRS Meeting Minutes January 28, 1988 NRC has the most complete data base for detemining the frequency of SB LOCA's.
In response to Mr. Davis, Mr. Mazetis indicated that the ultimate decision on the acceptability of the EPRI values of event fre-quencies shown (Figs. 5 and 6) versus the Sandia values has not beenmadeyet,pendingfutureconsultation(s)withEPRI,Sandia, NRC, etc.
Details of the differences between the NRC and NSAC-113 Reports were provided. For many of the specific sequences, NRC agreement with the 113-Report is pending receipt of additional details of their analyses. Mr. Ebersole asked if one can have much confidence in values of unreliability estimates of 10 10-8 as cited in the 113-Peport. Mr. Chelliah indicated there is a concern in this regard. ,,
In conclusion, Mr. Chelliah indicated that the "real" value of core melt frecuency for Point Beach probably lies somewhere between the NSAC-113 estimate and the Sandia Report estimate. This conclusion is, as noted above, pending receipt of additional infomation from EPRI/W0G on the details of their Report.
In response to Mr. Ward, Mr. Mazetis said NRC and Sandia will meet with EPRI/WOG on this matter. Mc. Mazetis indicated that be sees the goal of this interaction as providing insights to prospective Utilities that may use this approach for IPE analyses.
Prior to the next presentation, Mr. Ward indicated to the Subcom-mittee that the process described here and outlined in the Sandia Plant Analysis Reports needs to be factored in to the overall evaluation of the A-45 resolution approach,
t DHR$ Meeting Minutes January 28, 1988 l
- 5. ThecomparisonoftheSNLandEPRI/g0GPointBeachanalyseswas discussedbyD.Ericson(EPCIInc.-formallyatSNL). Key points noted by Dr. Ericson were:
- EPRI/EOG vsed the Risk Management Query System Approach to test the validity of their PRA. Overall, the approach appears acceptable.
- The results of the NSAC-113 reanalysis (as compared to the SNL analysis) were: (1) a factor of 30 reduction in core melt frequency. (2) an additional factor of 7 reduction in offsite consecuences, and (3) a 50% to 400% increase in cost estimates for proposed fixes. The main reasons for the differences in the results of the two reports are given in Figure 9.
- General comments on the above differences noted by Dr. Ericson were:
- The SNL Study was performed in 1984-85, theEPRI/EOG results were based on 1986 data (i.e., these Studies are "snap-shots" in time).
i
- Current methodology does not provide a "correct" answer.
i
- Difficult to establish effectiveness of human performance i
(particularly given use of new emergency procedures).
4 i
- Difficult to ascertain exactly what data, models and assumptionwereusedbyEPRI/EOG.
DHRS Meeting Minutes January 28, 1988'
- Specific conenents on the Reports' differences were:
- EPRI/W0G reduced the unreliability of human recovery time by a factor of 50; this is questionable based on past real experience (e.g., TMI, Browns Ferry).
- EPRI/MOG model of recovery was more plant specific and allowedmoreoptionsthantheSNLReport(Figures 10-11).
l
- Figure 12 lists the reasons for differences in the I
, modeling of common cause failure rates. Dr. Ericson said l theywishtoexplorethedetailsherewithEPRI/EOG.
- The assumptions used by EPRl/EOG for the seismic analysis were indicated by Dr. Ericson to be optimistic. In some instances, there are basic disagreements between SNL and '
EPRI/MOG(e.g.,possibilityofRWSTfailurerecovery), i
- The internal flood analysis difference turned on the calculation of the pipe break frequency. Modification of ,
, the correlation tsed in NSAC-113, the restricted pipe l 1e99th considered, and definition of tenns used in the 113-Report are all ltems of contention.
- The EPRI containment success criteria has not yet been evaluated by the Staff.
l l
- EPRI/EOG argues that differences in cost estimates are due to:
I l
- Failure to consider all design requirements (seis- ;
mic) for specific aspects of modification, f t
.o .
l DHRS Neeting Minutes January 28, 1988 ;
- Failure to account for existing structures and/or buried piping or cabling. ;
- Failure to account for iteration between initial design and final installation.
- Cost experienced in installing new batteries.
Dr. Ericson is puzzled by criticisms in this area, as SNL used an A/E costing expert for their analysis.
- Regarding upgraded E0P's (Fig.13), Mr. Ward said infor-mation he has been raises questions as to the ultimate usefulness of E0Ps in accident situations.
Dr. Ericson also noted the following: ,.
- Assumptions, data, and arproach used in SNL A-45 Reports have been often miscuoted.
- Best estimate analyses claimed by EPR!/WOG were presented with no discussion of the associated uncertainties. In response to Mr. Ward, Dr. Ericson said the SNL analyses were based on j
point estimates derived from generic data. Further discussion I
noted that the EPRl/WOG used various data bases for different parts of their reanalysis.
l
- EPRI/EOG Report comented on presuraed NRC positions vis-a-vis resolution of A-45; the Point Beach Report, only addressed a specific plant Case Study.
- 6. Mr. Ward asked RES for details on the resolution schedule and what RES would ultimately like to get from the ACRS via A-45 resolution.
RES said their schedule includes discussion with NUMARC in early l .. - - - - . . - - _ - - -- .
.~ . .
DHRS Meeting Minutes January 28, 1988 March, coeurrent with completion of their Regulatory Analysis. The Analysis should be available in 1-2 months. RES would appreciate ACRS coments on their resolution approach.
In response to Dr. Siess, RES indicated the nature of the Staff's reviewofthePRA'ssubmittedbyLicensees(ifany)isyettobe detemined. Mr. Minners acknowledged the difficulties of reviewing a PRA, but indicated that the alternative is a costly generic l hardware fix of possible dubious usefulness. He also indicated I
that the DHR issue seems to "naturally" fall into the severe accident concern.
! Mr. Davis indicated agreement with the proposed resolution ap-proach, but was troubled that there seems to be no cost-effective fixes. Mr. Minners disagreed, saying some fixes will be found cost
, effective and, in any case, he believes offssite and on-site costs should be considered. In response to Dr. Siess Mr. Minners 3 indicated that the "real" cora melt frequency probably lies somewhere between the above two Reports. Dr. Ericson agreed with Mr. Minners' sentiment. Mr. Minners also cautioned against being 4 swayed by the "bottom line" numbers noted above.
- 7. The Subcomittee met in open Executive Session. Mr. Ward referred to a discussion Paper he put together (Figs. 14-15). Mr. Michelson j asked if new plants should be addressed under A-45. Mr. Ward said A-45 does not address new plants. After further discussion, Mr.
1 Ward suggested the issue of DHR for new plants be postponed to a j future discussion.
l j Turning to the discussien Paper, the Chatman noted the six resolu-tion Alternatives and asked if there are others to consider and which, if any, the Subcommittee favors. Key points noted were:
DHRS Meeting Minutes January 28, 1988 -
NRC Proposed A-45 Resolution Alternatives
- TherewasdiscussionofAlternate2(limitedscopePRA). It was suggested that we (ACRS) reconnend every plant do a full-scope Level 1 PRA. Nrther discussion noted that there 4
is the problem of lack of Comission guidance vis-a-vis the frequencyofcoredamage(10~4/RY)versuslagerelese (10-6/RY). ,
' Dr. Mark suggested the ACRS address accidents not severe accidents, etc. He suggested that one list of accidents be established that would be addressed by the IPE's; i.e.,
address "accidents" as a coherent whole. This approach would eliminate "limited scope" PRA's for each specific US! and/or Generic Issue. HesaidtheACRS(andNRC)needstodevelopan integrated resolution approach to the USI/GI problem.
- There was discussion of the PRA vis-a-vis the safety goal and large release geal. There was general concern that the NRC i stating a 10-5/RY frequency for loss of assured core cooling 4 is inconsistent with the NRC Safety Goal (10'4/RY for core damage).
- There was general agreement that there are no other logical Alternatives to consider other than the Six noted above.
, Debate shifted to which Alternative (s) the Subcomittee favored. Mr. Ward suggested that the Subcommittee endorse Alternative 2, but that a full scope level 1 PRA (up to core melt) be required. In addition, for the Severe Accident Policy, the licensee would have to perfonn an analysis of the containment system. The cost benefit analysis would be abandoned in lieu of meeting the Safety Goal. Further discus-sion noted that sorre plants have initiated fixes independent of being told to do something by NRC (e.g., Point Beach
DHR$ Meeting Minutes
- January 28, 1988 installed seismic-capable battery racks after the SNL Plant Analysisinvestigation).
Considering the uncertainty associated with PRA's, it was agreed that judginent must be exercised by the reviewer; 1.e.,
when something ununsual is seen in a PRA result, additional action (s) (fixes) may be necessary.
Cost / Benefit (?/B) Analysis to Support A-45 Resolution
- Mr. Wad said C/B should be ignored if you adopt the approach noted above; however, C/B was used to consider and develop the abme Alternatives. Thus, the ACRS needs to address C/B somehow. The Chairman suggested the ACRS should cemnent on C/B.
What Inforwation is Needed for ACRS to Make Decision?
' Dr. Siess said the ACRS would need to understand the practical aspects of using a PRA if we endorse Alternative 2.
- Dr. Mark said NRC would need to set out cific, uniform methodology for performing PRAs so one as is the arguments like the Subermittee heard today. '
- Mr. Davis indicated that there is no prob'iem with a lack of Industry expertise for performing PRAs. Over 35 PRA's have been performed on a large enough variety of plant designs to assure a good product.
Mr. Ward said he would recomend that the Subcomittee hold another Meeting when PES issue their Regulatory Analysis Report. ACRS review would follow. The Subcomittee decided that it should
.DHRS Meeting Minutes January 28, 1988 further review the EPRI/WOG reanalysis of the Point Beach report following the NRC/NUMARK consultation on the differences between the Reports.
- 8. The meeting was adjourned at 3:45 p.m.
- +**********
NOTE: Additional meeting details can be obtained fre,n a transcript of this meeting available in the NRC Public Document Room, 1717 H Street, N.W., Washington, D.C., or can be purchased from ACE-Federal Reporters, 444 North Capitol Street, Suite 402, Washington, D.C. 20001,(202)347-3700.
. e.-
4 l
,. . -.n - - - . . , - , - - - , . . , . , , , - , . - , , , , - , , - . - - . - - . , - ,
' . , 'i'-.
BACKGROUND CONCERNS ON RELIABILITY OF DHR FUNCTION:
- PAISED BY TMI EXTENDED TO SPECIAL EERGENCIES (E.G., FIRE, FLOOD, SEISMIC, SABOTAGE)
ACRS, TASK FORCES, AND SPECIAL C0ft11SSIONS REC 0ftENDED TliAT HIGH PRIORITY BE GIVEN TO STUDIES OF IPPROVING SHUTDOWN DHR FUNCTION (POST-TMI EFFORTS)
UNRESOLVED SAFETY ISSUE (USI A-45) APPROVED DEC, 24, 1980 (SECY-80-325)
KEY QUESTIONS:
D0 CURRENT DESIGNS PROVIDE THE RELIABILITY N.'EDED TO EET CDF GOAL?
- ARE THERE IPPROVEiWS TO DHR FUNCTION IN OPERATING PLANTS hiilai ARE COST-BENEFICIAL?
I __
lFM.)]
USI A l45 SPECIFIC OBJECTIVES
- DETERMINE THE SAFETY ADEQUACY OF DECAY HEAT REMOVAL IN EXISTING POWER PLANTS FOR ACHIEVING BOTH HOT SHlHDOWN ANDCOLDSHlHD0hhCONDITIONS .
- DEVELOP AND EVALUATE ALTERNATIVE METl10DS FOR IMPROVING RELIABILITY OF DECAY HEAT REMOVAL FUNCTION,
- ASSESS THE VALUE AND IMPACT OF ALTERNATIVE MEBODS
- '
L l'
USI A-45 SCOPE EVALUATED DHR SYSTEMS NEEDED TO RESPOND TO TRANSIENTS AND SMALL-BREAK LOCAs (DID NOT INCLUDE LARGE BREAK LOCAs OR ATWS)
ALSO EVALUATED SUCH SYSTEMS' VULNERABILIU TO FIRE, FLOOD, SEISMIC, INSIDER SABOTAGE SABOTAGE CONSIDERATICNS:
- CONSIDEFID VULNERABILITY OF DHR SYSTEMS TO "INSIDER"
~
SABOTAGE
- DID NOT CONSIDER "PHYSICAL SECURITY" ETIK)DS ("OUTSIDER")
3 f_f/A 3
Slft%RY OF FINDINGS FROM SIX CASE STUDIES PROBABILITY OF CORE ELT DUE TO DiiR FUNCTICN FAILURE IP(CM)DHRI AVERAGES 2 TO 3 x 10-4 PER R-YR (INCLUDES ,INERNAL AND EXTEPML CAUSES)
SUPPORT SYSTEM FAILURES (E.G., EE RGENCY POWER, SERVICE WATER, C0f'PONEE COOLING) CONTRIBUTE SIGNIFICAELY TO P(CM)DHR LACK OF REDUNDANCY AND SHARING OF SYSTEMS, PARTICULARLY AT SUPPORT SYSTEM LEVEL, CONTRIBUTE SIGNIFICANT RISK FOR SOME PLANTS LACK OF INDEPENDENCE, SEPARATION AND PHYSICAL PROTECTION OF REDUNDANT SAFEGUARD TRAINS CONTRIBUTE SIGNIFICAE RISK ,
RISK FROM FIRE, FLOOD, SEISMIC, SAB0TAGE IS SIGNIFICANT BOTTCM LINE IS TilAT RELATIVE IMPORTANCE OF VULNERABILITIES IS PLANT-SPECIFIC 5
/m g)
1 P014T PEACH 2HR/3RA l
l SEQUENCE FREQUENCY :/RY'
SUMMARY
DIFFERENCE S2MH1'H2' 4.7E-5 SNL
- REVISED SMALL LOCA FREQU 5.8E-7 [EPRI
- MODIFIED CCW SUCESS CRITERIA
- UPDATED INITIATOR FREQUENCY 7.7E-7 [EPRf.
- N0 NEED FOR RECIRCULATIONye :"
NA EPRI:
T2M0H1'H2' 3.5E-6 SNL'
- REVISED PORY PROBABluTY 1.9E-7 EPRf S2MD1D2 8.7E-6 SNL'
- REVISED SMALL LOCA FREQUENCY 9.5E-8 [EPRf
- MODIFIED CCW SUCESS CRITERIA l
h&5)
a PC' INT BEACH DHR/39A SEQUENCE FREQUENCY /RY:
SUMMARY
DIFFERENCE ,
- NEW METHOD FOR FLOOD FREQUENCY 1.0E-8 'EPRI
- UPDATED HPIS PERFORMANCE SEISMIC 6.1E-5 SNL
- CREDIT FOR SECOND~ TRAIN EVENTS 6.3E-8 lEPRI: HALON SYSTEM
- REVISED HALON SYSTEM FAILURE PROBABluTY e CREDIT FOR REC 0VERY OF AFWS FOR SWITCHGEAR ROOM FIRES lN ff
POINT BEACH DHV3RA SEQUENCE DESCRIPTION S2MH1'H2' - SMALL LOCA FOLLOWED BY THE FAILURE OF LOW AND HIGH PRESSURE RECIRCULATION SYSTEMS T1MLE - A LOSS OF 0FFSITE POWER EVENT FOLLOWED BY THE COMMON MODE FAILURE OF THE DIESELS, RESULTING IN EARLY CORE MELT T30H1'H2' - A TRANSIENT INVOLVING A STUCK OPEN PORY AND FAILURE OF RECIRCULATION SYSTEMS. 'THE MFW SYSTEM IS ASSUMED TO BE INITIALLY AVAILABLE.
T30D1D2 - A TRANSIENT INVOLVING A STUCK OPEN PORY AND FAILURE OF BOTH HIGH AND LOW PRESSURE INJECTION SYSTEMS. THE WFW IS ASSUMED TO BE INITIAU.Y AVAILABLE.
T2M0H1'H2' - A LOSS OF FEEDWATER EVENT FOLLOWED BY A STUCK OPEN PORY AND FAILURE OF RECIRCULATION SYSTEM.
S2MD102 - A SMALL LOCA EVENT FOLLOWED BY THE FAILURE OF
! BOTH HIGH AND LOW PRESSURE INJECTION SYSTEMS.
30 NT BEACH 2WPRA SW FLOOD - A FLOODING EVENT IN SERYlCE WATER PUMP ROOM CAUSING lil0PERABillTY OF ALL SW PUMPS, RESULTING IN LOSS OF CORE C00UNG SYSTEMS.
THE FLOOD iS ASSUMED TO RESULT IN A TRANSIENT.
SEISMIC EVENTS - SEISMIC EVENTS Of LOW TO MODERATE INTENSITY RESULTING IN:
(A) LOSS OF FEEDWATER SEQUENCES INVOLVING INJECTION FAILURES, (8) SMALL LOCA SEQUENCES INVOLVING INJECTION FAILURES, AND (C) TRANSIENTS INVOLVING THE FAILURE OF BOTH .
PRIMARY AND SECONDARY C00UNG SYSTEMS.
FIRES - 1RANSIENT COMBUSTIBLE FIRES WERE CONSIDERED.
(A) A RRE IN AFW PUMP ROOM FOLLOWED BY PRIMARY C00UNG SYSTEM FAILURES.
f (8) A RRE IN 4.16 KY SWITCHGEAR ROOM FOLLOWED BY PRIMARY AND SECONDARY C00UNG SYSTEM FAILURES.
f
.o Reasons for Differences i
HPI does not require CCW SBLOCA and transient-induced LOCA frequency Low Point Beach-specific transient frequencies j AC and DC bus cross-connects l Miscellaneous operator recovery actions or
- human error in performing normal actions 1
h N
i l USI A-45 Recovery .
1 i
l RA-1 Loss of offsite power l RA-2 Loss of main feedwater Battery common cause i
RA-8 RA-9 Battery fault RA-10 Diesel ~ common cause
~
j RA-11 Diesel fault RA-6 Other failures from control room
- RA-7 Other failures locally l
lN i k lk .
EPRl/WOG Recovery
! RWST refill using spent fuel pool or chemical and volume control system Cross-connect AC or DC buses Manual control of turbine-driven AFW pump .
Provision of backup supply of feedwater Use of charging system for loss of feedwater
! Balancing loads on the service water system i
L Recovery from co'mmon mode failures l
o -
v
,4 i
)
Common-Cause Failure Rates I
Reasons for Changes .
Design review for determining beta factors l
The design review process has yielded a factor of two lower in EPRrs application than the beta factors calculated using all i
industry experience. {le, Factor of 3 on pump failure)
Recovery from common-cause failures A review (AEOD report C504) indicates that roughly 60% of both human and hardware failures were recovered within one hour .
Used data from Millstone-3 PRA hik i
i w - - - - - - - -
er F
Upgraded Emergency Operating Procedures i,
i There is evidence NUREG/CR-4617, March 1987 I that EOPs markedly improve operator performance (CMF coservatively estimated to be reduced by a factor of 1/8)
~
Synergisms am'oung transient analysis, upgraded training, and management overview i
l N.
, .. 1. s Jen. 28, t988 DAW Staff has oroDosed six alternatives for the resolution of A-45:
Are there others?
Which do we like?
Il No action -- require nothing additional of licensees 21 Limited scope PRAs, perhaps coupled to IPEs somehow 3] tieke some specified system modifications; use resolution of other USls and Gls to specify exactly what 4] Designete some upgraded standards or requirements for depressurization and cooling for PWRs -- feed & bleed (but, what about conteinment cooling) for BWRs -- conteinment venting (credits existing ADS) 5] Provide e new dedicated hot shutdown cepebility et each plent EFW end limited RCS makeup cepebility with dedicated services
- 6) Provide e new dedicated cold shutdown cepebility et each plant EFW end limited RCS makeup capability plus RHR with dedicated services Staff hos orocosed 3 elternatives for cost / benefit enetusis:
Should others be considered?
Which do we like?
For which alternettves?
(Presumably all also include the cost of mods as e negative.]
\
gnid Al Consider only everted.en-site costs B} Consider everted on-site plus everted off-site costs Cl Consider, with attempts to quantify as possible sabotage e ' nuclear moratorium" of some sort resolution of other USis and Gls other f actors -- perheps some totelly impossible to quantify l
W
!Ie
. .4 What information does ACRS need to make a decision?
Do we need e repeat of information we have heard over last 5 ye
-Why? Can it be summerized somehow? Or indexed?
Are differences in industry and NRC PRA results important?
For which ellernatives?
f/4 /5 ~