ML20151U971
| ML20151U971 | |
| Person / Time | |
|---|---|
| Issue date: | 04/26/1988 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1660, NUDOCS 8805020225 | |
| Download: ML20151U971 (267) | |
Text
cRS7' /460 i OR G \ A_
O UNITED STATES 1
- NUCLEAR REGULATORY COMMISSION 1
r e ss as sa mm am am en ea ss en em as sm an am es ma m m an - m us am me ns as as as em am ma ss e . . as em as sa mm as em an am ul um as s a m m mm ms am m a l Iil THE MATTER OF: )
)
MEETIllG REGARDIliG PROPOSED )
GEtJERIC LETTER FOR )
ItJDIVIDUAL PLAllT ) J EXAMItJATIOllS (IPEs) )
i i
l i
j l
4 i
- O 1
i l
i l
i l
l 1
l LOCATIOtJ Washington, D.C.
i l
PAGES: 1 through 231 , _ ,
,. s m a f7 l DATE: April 26, 1988 . , c, - A P Y" s- <
j [ '. P t
- -~
r ! l * ,., p
\ #
,,)
.................................................klii.'e*
" .s j
I j 'g i ; ,' { ; si ; !d., " .'- ~
i i
HERITAGE REPORTING CORPORATION l
Q == n-112e L Serest, N.W., See ett .
l 2
Weekington, D.C. 20005 l
- oso2022s es:o w"
! PDR CCRs (292) M l T-166o DCD
-w. _ . v v w an e- ,. -r-- 'n vT v **
t p 1- PUBLIC NOTICEJBY.THE 2 , UNITED STATES NUCLEAR REGULATORY COMMISSION'S '
3 ADVISORY COMMITTEE ON. REACTOR SAFEGUARDS
{
4 :
1 5
6 l 7
.The contents of this stenographic transcript of.the 8 proceedings of the United Stctes1 Nuclear Regulatory
?
9 Commission's Advisory. Committee on Reactor Safeguards (ACRS),
10 .as reported herein, is an uncorrected record of the discussions I? recorded at'the meeting held on the above date.-
12 No member of the ACRS Staff and no participant at ;
13 thic meeting accepts any responsibil.ity for errors or
(]) 14 inaccuracies of statement or data contained in this transcript.
1 15 '
16 -
17 !
18 i 19 I 20 i 21 22 23 I 24 ,
t 25
.() Heritage Reporting Corporation (202) G28-4888
- ., ,. M ,
?. , ,<,
1 i
C) 1 UNITED STATES NUCLEAR REGULATORY-COMMISSION l
2- ADVISORY COMMITTEE ON REACTOR SAFEGUARDS ;
3 SEVERE ACCIDENTS SUBCOMMITTEE 4 <
In the Matter of ) ,
5 ) :
MEETING REGARDING PROPOSED ) .
i 6 GENERIC LETTER FOR INDIVIDUAL )
PLANT EXAMINATIONS (IPEs! )
7 )
8 Tuesday, April 26, 1988 9
Room 1046 10 1717 H Street, N.W.
Washington, D.C. 20555 11 The meeting convened, pursuant to notice, at 8:30 l 12 a.m.
() 14 BEFORE: DR. WILLIAM KERR, Chairman Professor of Nuclear Engineering Director, Office of Energy Research 4 15 University of Michigan i Ann Arbor, Michigan 16
! ACRS MEMBERS PRESENT:
17 CARLYLE MICHELSON i
18 Retired Principal Nuclear Engineer Tennessee Valley' Authority 19 Knoxville, Tennessee, and, Retired Director, Office for Analysis and i 20 Evaluation of Operational Data U.S. Nuclear Regulatory Commission 21 Washington, D.C.
' 22 PAUL G. SHEWMON Professor, Metallurgical Engineering Department 23 Ohio State University i
Columbus, Ohio
! 24
- (Appearances Continued on Page 2) 25 1
+
Heritage Reporting Corporation (202) 628-4888 s-
r t
, 2 1 APPEARANCES (Continued):
2 ACRES MEMBERS PRESENT (Continued): l 3 CHESTER P. SIESS Professor Emeritus of Civil Engineering 4 University of II,\inois Urbana. Illinois 5
DAVID A. WARD 6 Research Manager on Special Assignment E.I. du Pont de Nemours & Company 7 Savannah River Laboratory Aiken, South Carolina 8
CHARLES J. WYLIE 9 Retired Chief Engineer Electrical Division 10 Duke Power Company Charlotte, North Carolina 11 i ACPS COGNIZANT STAFF MEMBER:
12 ,
DEAN HOUSTON ,
r- 13
(_)s 14 CONSULTANTS:
I. CATTON 15 P. DAVIS J. LEE 16 M. CORRADINI 17 PRESENTERS:
18 BRIAN SHERON, et al. (Of fice of Res > tch Staf f)
J. HULMAN (Research) 19 20 21 22 23 24 25 Heritage Reporting Corporation (202) 628-4888
, \.
~a i
)' .
1 PRQCEEQINgg !
2 DR. KERR: The meeting will come to order. This is a ;
meeting of the Advisory Committee on Reactor Safeguards,
.3 4 specifically the Subcommittee on Severe Accidents. My name is-t i 5 Kerr I am Subcommittee Chairman. In attendance today as i 6 members of the Committee are Mr. Michelson, Mr. Shewmon, Mr.
7 Siess, Mr. Ward and Mr. Wylie.
8 As consultants we have today Messrs. Catton, 9 Corradini, Davis and Lee.
10 Our purpose in meeting today is to review a version ,
11 that we are told is near to being a final version of the NRC 12 staff's proposed generic letter which will describe the 13 procedure to be followed in carrying out individual plant j 14 examinations.
i 15 Dean Houston is the cognizant ACRS staff member for 16 the meeting.
17 Rules for participation in the meeting were announced i 18 as part of the meeting notice published in the Federal Register 4
19 on March 30, 1988.
20 A transcript of the meeting is being kept. I ask i 21 that each speaker identify himself or herself and use a 22 microphone.
. 23 We have received no written comments nor have we had 24 requests for time to make oral statements.
1 25 The individual plant examination was first officially
(:)
i Heritage Reporting Corporation (202) 628-4888
f 4 ;
^ -
1 mentioned I believe in a policy statement of the Commission i 2 that was issued in August of 1985, and I will read from the !
3 policy statement: "Accordingly, when NRC and industry 4 interactions on severe accident issues have progressed 5 sufficiently to define the method of analysis, the Commission 6 plans to formulate an integrated, systematic approach to an 7 examination of each nuclear power plant now operating or under
{
8 construction for possible significant risk contributors, ;
9 sometimes called outliers, that might be plant specific and l 10 might be missed absent a systematic search."
11 This subcommittee has discussed the process of the
{
12 formulation of a procedure and has reviewed several versions at 13 a number of meetings.
O 14 The report made by the full committee of ACRS 15 comments on an earlier draft appeared on June 9, 1987, and I 16 think that was the last time that we commented on this process j 17 and on the draft version that existed at that time.
] 18 I would call your attention particularly to the last i
19 two paragraphs in our letter, for those of you who at least 20 have copies of the letter, and I think those of you around this 21 table do have.
. 22 And in those paragraphs, we expressed some
- 23 reservations about what was proposed at that time.
24 I'm told that the staff has made revisions which they i 25 believe respond appropriately to our concerns as expressed in
() Heritage Reporting Corporation
! (202) 628-4888
- 4. .
6 1 the letter.
2 This topic is scheduled for discussion by the full 3 . committee at-our May meeting-and I anticipate that the 4 committee will prepare a report at that meeting.
5 Consequently, I ask those of you who are here as 6 members of the committee and those of you who are here as -
7 consulcants to give me your comments at the end of'the meeting.
8 I think we'll have some time for comments to be made orally.
9 [f for some reason we are pressed for time, I do need comments 10 in order to try to prepare at least a draft letter which will 11 serve as a basis for the beginning of the discussion of the 12 full committee.
13 I don't know whether -- Dean, can you remind me when 14 the discussion by the full committee is scheduled?
15 MR. HOUSTON: It's 1:30 on Friday. Friday, May 6th, 16 I believe it is.
17 DR. KERR: Thank you.
18 Are there any comments or questions by members of the 19 subcommittee?
20 (No response) j 21 DR. KERR: Consultants?
22 (No response) 23 DR. KERR: Then I guess I have a few more comments l 24 before we hear from the staff.
25 First, since considerable emphasis is placed upon the
}
Heritage Reporting Corporation (202) 628-4888 l
l se,,,. .vm. - -- ~, y. , ,. -_ _ _ , - _ . . - , . , . , ,
, r r-j 6-kL
~
1 use of PRA in the procedure, I would be interested in comments 2 from'the staff about whether there now exists a PRA which the 3 staff-would find acceptable as a basis for the IPEM. And if 4 so, I'd appreciate having it identified.
5 Second --
6 DR. SIESS: Excuse me, Bill. What do you mean by 7 IPEM?
8 DR. KERR: Individual Plan Examination Methodology.
9 DR. SIESS: As I recall, that acronym referred to the 10 IDCOR method.
11 DR. KERR: I don't think it did.
12 DR. SIESS: That's why I asked.
13 DR. KERR: No.
14 DR. SIESS: Because I've gotten that impression 15 somewhere, that IPEM was what they referred to as IDCOR.
16 DR. KERR: Well, IDCOR, but you don't mean they made 17 an effort to develop something that would be acceptable to 18 satisfy the requirements for an IPEM, as I understand the 19 jargon.
12 0 DR. SIZSS: Okay. What you meant was it would 21 satisfy the generic letter?
22 DR. KERR: Well, to satisfy the Commission's 23 commitment in the safety policy statement or the severe e4 accident policy statement.
25 A second question which perhaps I could find if I
() '
Heritage Reporting Corporation (202) 628-4888
i l
I 7
t'l
'> 1 looked further but I haven't.foundlis on what basis changes 2 will be made if they are made in response to the examination.
3 For example, will new systems be required to be safety grade;
'4 w!.11 license amendments be required? And I gather they may be, 5 because reference is made to 50.59 being a basis for suggested 6 changes.
7 And this puzzles me a bit because my impression is 8 that one judges whether something has safety significance or 9 not by whether it requires a change in tech. specs or whether 10 it violates some particular rule and regulation. And this 11 would raise a question as to whether changes expected to be 12 made as a result of this do require license changes, tech.
13 spec. changes, on what basis they will be reviewed under O 14 existing rules.
15 A third question has to do with the way in which the 16_ unresolved safety issue on removal of de cay heat will be 17 treated.
18 I gather it is to be subsumed within the individual 19 plant examination process. And I would be interested, for 20 example, in how whatever is done by a licensee to satisfy USI 21 46 impinges on the requirement to satisfy 45.
22 Certainly much of 45 has to do I think with the fact 23 that under certain assumed sequences one has loot all electric 24 power.
25 And my question really has to do with whether, if 46
() Heritage Reporting Corporation (202) 628-4888
8 g.
\- 1 has been satisfactorily resolved for a plant, it is now 2 assumed 3 that electric power is satisfactory or if under the
,4 satisfaction of 46 electric power is only satisfactory for non-5 severe accidents, and now one has to have another look for 6 severe accidents. This is not clear to me.
7 I am also interested to find that the staff considers-8 direct containment heating to be one of the more significant 9 risk contributors, in the discussion on contairaent analysis.
10 I would be interested in the basis for that 11 conclusion, whether it is simply a matter of engineering 12 judgment or wnether it is based on numerical estimates of-13 contribution to release or risk or whatever.
O 14 And I would also be interested just as a basis for 15 comparison as to whether the staff thinks that that contributes 16 more risk than, say, catastrophic vessel failure, which I think 17 from some of the numbers appearing in the screening criteria, 18 perhaps has about the same likelihood of occurring.
19 Mr. Siess?
20 DR. SIESS: One thing I would like to add. There is 21 a reference in the IPE letter to the use of the safety goal 22 policy as a factor in assessment. And I would like to hear 23 what portion of the safety goal policy that refers to -- the 24 qualitative, the quantitative, the large release, the severe 25 core damage -- as well as how the staff expects to use the O Heritage Reporting Corporation (202) 628-4888 L- .
e =
s 9
'-- 1 safety goal policy in relation to an examination that deals 2 only with internal events.
3 I don't think there is anything in the safety policy
- 4. that allocates the risk between internal and external events.
5 DR. KERR Did you have a comment?
6 MR. WYLIE: I assume the comment you made a moment 7 ago-regarding 46, the reference really was to 44, I think.
8 DR. KERR: It may be. I was thinking of electric 9 power and I may have confused the numbers. The USI to which I
'10- resolved has to do with reliability of electric power. So if 11 somebody will transpose the numbers, if they need transposing.
12 MR. WYLIE: I think it's 44.
.. 13 DR. KERR: Thank you. Other comments?
\"# 14 (No response) 15 DR. KERR: In terms of the schedu e that you have 16 which shows Mr. Spels as making a first presentation, Mr. Spels 17 is ill. I don't know whether his illness has anynhing to do 18 with his consideration of appearing here or not but I assume 19 that he has some sort of virus and I hope for him a speedy 20 recovery.
21 In the meantime, Mr. Sheron will try to cover the 22 material that he was going to discuss, but since he needs some 23 time to get the relevant slides, we will postpone what would 24 have been Speis' presentation on severe accident issues to a 25 later period today and will begin with Mr. Sheron's O Heritage Reporting Corporation (202) 628-4888
p 10 V
1 introduction and discussion of the IPE Generic Letter.
2 Mr. Sheron.
3 (Slides being shown) 4 MR. SHERON: I hope you don't mind that I took.my 5 jacket off. I have a' feeling I'll be up here_a while.
6 First,-let me introduce myself. I'm Brian Sheron 7 with the Office of Research staff. And I have with me Farouk' 8 Eltawda, Bill Beckner, Tom Cox, John Chin and Glen Kelly from 9 the staff. Hopefully among us we'll be able to answer all your 10 questions.
11 Let me just before I start on my actual presentation 12 just give you a little history on where we are with this.
13 As you know, the interactions with the industry have 14 been going on with IDCOR I guess some seven years now. We
'15 finally completed that interaction process. I understand that 16 IDCOR doesn't exist and it is now, the whole program is under 17 the NUMARC organization. <
18 We've got to the point, we've been down here a number 19 of times. If you recall, I think the last time I remember was 20 last July or so. We brought you up to speed on where we were i
21 with preparing the whole letter. We've worked on it 22 substantially since then. We've worked very closely with NRR, 23 in the new organizatic7. We think we've reached consensus with 24 NRR such that the joint offices of Research and NRR are 25 satisfied with this letter. And that goes from Dr. Murley, and O Heritage Reporting Corporation (202) 628-4888
9
/~s 11 k) 1 Mr. Beckjord, on down, that this package is in good enough 2 shape to go.
3 We have met with the CRGR. We met last week. We 4 pointed out, I talked to Denny Ross and he said we set a record 5 for the length of the CRGR meetings so far in this agency. It 6 took a span of something like-three days with the staff and 7 there was an hour and a half conference call among the CRGR 8 members after that. We are still awaiting their letter.
9 I will a little bit later tell you what the results
' 10 of that meeting were.
11 And I hope, if I for some reason don't address your 12 questions, Dr. Kerr, Dr. Siess, please let me know.
13 This is an outline of right now what we would propose 14 the generic letter to consist of. There is a summary, 15 discussion of the examination process, discussion of how 16 external events should or should not be treated, discussion of 17 acceptable methods of examination, discusses the relationship 18 to USI A-45.
19 There is a short section which points out the 20 benefits of doing a PRA. It talks about the severe accident 21 sequence selection. There is a section on severe accident 22 management.
23 There is another section which discusses the severe 24 accident prevention and mitigation features reports. We 25 discuss how we would like the industry to document their O Heritage Reporting Corporation (202) 628-4888
12
- - . 1 examination results, what the licensee response should look 2~ like and the sche'dule. And then there's a short discussion on 3 .the regulatory basis for this request.
- 4. Now, there are five appendices to the letter. .The 5 first is the Guidance on the Examination of the Containment 6 Performance.
7 Appendix 2 is our Criteria for Selecting, or the 8 Screening Criteria for Selecting the Severe Accident Sequences.
9 Again, a short appendix discussing our views on 10 accident management and how we think they relate to the conduct 11 of the IPE.
12 Appendix 4 discusses in more detail the type of 13 documentation we would like the licensees to provide.
O -14 Appendix 5 is something we added most recently based 15 on the decision to resolve A-45 and that is to add a section 16 providing insights on decay heat removal vulnerabilities.
17 And then Attachment 1 is just a list of references 18 which are called out throughout the letter.
19 This is a summary basically of the letter. It points 20 out that we are requesting the information under 10 CFR 21 50.54(f). Just to refresh your memory, 50.54(f) allows the 22 staff to go out and collect information it deems necessary to 23 make a determination whether a license needs to be suspended, 24 revoked or modified.
25 It is information collection only. It is not the O ,
Heritage Reporting Corporation (202) 628-4888
1 examination results, what the licensee response should look 2 like and th* sche'dule. And then there's a short discussion on 3 the regulatory basis for this request.
4 Now,.there are five appendices to the letter. The 5 first is the Guidance on the Examination of the Containment 6 Performance.
7 Appendix 2 is our Criteria for Selecting, or the 8 Screening Criteria for Selecting the Severe Accident Sequences.
9 Again, a short appendix discussing our views on 10 accident management and how we think they relate to the conduct 11 cr the IPE.
1: Appendix 4 discusses in more detail the type of 13 documentation we would like the licensees to provide.
O 14 Appendix 5 is something we added most recently based 15 on the decision to resolve A-45 and that is to add a section 16 providing insights on decay heat removal vulnerabilities.
17 And then Attachment 1 is just a list of references 1
18 which are called out throughout the letter.
19 This is a summary basically of the letter. It points 20 out that we are requesting the information under 10 CFR 21 50.54(f). Just to refresh your memory, 50.54(f) allows the 22 staff to go out and collect information it deems necessary to 23 make a determination whether a license needs to be suspended, 24 revoked or modified.
25 It is information collection only. It is not the
%.) , Heritage Reporting Corporation (202) 628-4888
s ,
13 5)( 1 imposition of any requirements. It just says that.we are 2 collecting this information so that we can perform our 3 regulatory functions.
4 The basis, the. underlying basis for the 50.54(f)
'5 -request-has been determined from our legal staff to be the 6 Commisalon's severe accident policy. We did not have to go 7 through and do any formal cost benefit analysis to show why 8 this is something that passes cost benefit analysis and so 9 forth.
10 The plant specific PRAs, if you remember the policy 11 statement, said that they expose relatively unique 12 ' vulnerabilities to severe accidents, sometimes, and that in
~
13 general, though, plants didn't pose any undue risk to the 14 health and safety.
15 As~you know, it pointed out though that when you do a 16 PRA or a plant specific examination you typically find that 17 there are unique aspects that may pose some undesirable traits 18 which you'd like to fix.
19 And the Commission said that these examinations 20 should be performed so that any undesirable risks can be j 21 reduced by low cost changes via procedures or minor design l
22 modifications.
23 Any plant that has not undergone a;propriate l 24 examination should do an analysis once the NRC and the industry l
25 has sufficiently progressed to define the methods of analysis.
l O Heritage Reporting Corporation l (202) 628-4888 f
J
T 14
\J 1 And we think we've reached that point.
2 Since the issuance of the policy statement, the staff 3 has been interacting extensively with industry through the 4 IDCOR organization, and we think we have now developed an 5 appropriate method and documentation for the IPEs.
6 In essence, the purpose of the IPE is for the 7 utilities. And I stress the word "utilities." It is not for 8 the staff, it is not for their contractors. It is for the 9 utility people themselves to identify and understand the most 10 likely severe accident sequences that could occur at their 11 plants, to evaluate them, to decide if there are ways that they 12 can make the improvements that the Commission had in mind, and p
13 to make those improvements, and to develop an awareness for
() 14 severe accident behavior.
15 MR. SHENMON: One of the thingo that that is sort of 16 working on the sidelines with this stuff has to do with USIs.
17 Do you see any difference in the ability of an IPE or a PRA to 18 cope with these sorts of specific but not precisely defined 19 items or items that don't have precisely defined solutions?
20 MR. SHERON: No. I think we decided that the IPE or 21 a PRA would be capable of looking at a USI or a GSI.
22 You will note though that the only USI that we are 23 proposing to resolve is A-45. And I will discuss that in more 24 depth later.
25 MR. HOUSTON: Excuse me, Brian?
Heritage Reporting Corporation (202) 628-4888
t .
jg_ 15
.V 1 MR. SHERON: Yes.
2 MR.-HOUSTON: I have a comment on the "Implements v
3 Means for Improvement" part-of that slide.
4 On Page 2 of the letter it says: "It is exp,ected 5 that the achievement of these goals, IPE goals, will bring the 6 overali industry severe core damage and large-radioactive 7 release probabilities to values more consistent with the 8 . Commission's safety goal policy statement."
9 It seems to me that statement implies that as they 10 currently exist the plants do not or ar not consistent with the
-11 safety goals, which seems to be inconsistent with the safety 12 goal policy statement which says that the NRC feels they are, 13 the plants are safe.
14 And I don't think doing an IPE is going to bring the 15 plant to a safer state unless you actually implement some means 16 of-improvement.
17 MR. SHERON: I agree, doing an analysis and not 18 changing anything on the plant is not' going to make it safe, 19 except --
20 MR. ELTAWDA: Brian?
21 MR. SHERON: I'm sorry.
22 MR. ELTAWDA: It seems like that has been changed 23 because that was a typographical error. I am going to pass 24 right now a revision to two pages of the gencric letter.
25 MR. SHERON: So now it will say that the IPE will O Heritage Reporting Corporation (202) 628-4888
p 16 0 1 verify.
2 MR. ELTAWDA: Will verify that the program is in 3 compliance.
4 MR., DAVIS: We recognize that one'could interpret'it 5 exactly how you did which is that the plants _today don't meet 6 the safety. goal, and we were not trying to leave that 7 impression.
8 MR. WARD: You say you are going to pass that out?
9 MR. ELTAWDA: Yes.
10 DR. SIESS: Is there a safety goal for internal 11 events versus one for external events?
12 MR. SHERON: I don't ti. ink the safety goal 13 distinguishes.
14 DR. SIESS: I don't quite understand how then an 15 analysis for internal events is going to tell you anything 16 about how a plant meets a safety goal.
17 MR. SHERON: We are not proposing to, when we're 18 finished, doing this evaluation with the internal events, to 19 stand up and raise our right hand and say all these plants meet 20 the safety goal.
21 DR. SIESS: This item on Page 2 refers to sometime in 22 the future after you've done external events?
23 MR. SHERON: Right now, the Commission has, we 24 understand the Commission's guidance system and even the ACRS' l
25 is that we should not be applying safety goals to judge l O)
\-
l Heritage Reporting Corporation l (202) 628-4888
1
(~g 17-ul . -
1 individual plants but.rather we should be using it to judge the 2 adequacy of the regulations under which these plants were
.3 designed.
4 And as I get into this a little later, you will hear 5 how we intend to apply the safety goal in that manner.
6 DR. SIESS: Okay. I'll wait.
7 DR. KERR: At least as far as the ACRS comments go 8 and I think to some extent these were endorsed by the 9 Commission, was also an injunction that PRA not be used to 10 judge the adequacy of an individual plant.
11 MR. SHERON: In total, correct. Other factors. And 12 we are not proposing that these examinations, whether they do a 13 PRA or an IPE, would be the sole determinant. In other words, 14 we are not -- what we are trying to do is not stress bottom 15 line numbers as being the determinant or when you do or don't 16 fix something.
17 DR. KERR I do not get that impression from the 18 letter, but I perhaps misunderstood.
19 MR. SHERON: The examination process itself, I guess 20 the biggest thing that we see as the benefit of this is that 21 the licensee's staff should participate in all aspects of the 22 IPE so that the knowledge gain becomes an integral part of 23 operating, training and procedures programs.
24 We are not encouraging the licensee to go out and 25 hire a contractor to do the whole thing for him, package it up O Reporting Corporation Heritage (202) 628-4888
i' .
18 b
(~/.
N- 1- ar.d-just'have a vice president sign it off and send it in and 2 say I'm finished.
-3 We don't think that you have really achieved what the 4 real benefits are of this unless the utility staff has actually 5 participated-in the walkdowns, in understand the plant and 6 understanding where the vulnerabilities might lie. '
7 MR. CORRADINI: Can I ask a question about that?
8 MR. SHERON: Yes.
9 MR. CORRADINI: Will you get into later on how you 10 are going to ensure that's going to happen?
11 MR. SHERON: Yes. A little bit.
12 MR. CORRADINI: Because I guess it was a year ago 13 when we had first seen this, the comment was, at that time, if CT
l' 14 I remember correctly, that there seemed to be an inconsistency 15 between timetable and this goal because if the timetable is 16 relatively short, the utilities opt for the contractor doing 17 most of the work. Simply, they don't have the staff or the 18 time to build up staff to do it themselves.
19 And so I'm curious. I think this sort of purpose is 20 quite good. I'm just curious how it's going to be ensured.
21 MR. SHERON: Well, I think the first thing is we're 22 going to ask the utilities when they submit their plan for 23 implementing this to tell us. You know, we would like to see 24 from them at that time what they propose to be the degree of 25 involvement of their staff.
O Heritage Reporting Corporation (202) 628-4888
19 L 1 If they come in and wring their hands and say gee, 2 you guys have so many other requirements on me I can't do this 3 right away, you'know, we may have to consider allowing them to 4 delay, if there's valid reasons, or we'll negotiate with them.
5 But I think that if this doesn't take. place, then the 46 principal benefit is going to be missed. ,
'7 DR. KERR: Brian, since the term "vulnerabilities" 8 has been used by you now a number of times and is used even 9 more in the report, is it expected that somewhere in a glossary 10 the term will be defined so that people will know what it is 11 they are looking for?
12 MR. SHERON: No. There is no real definition of the 13 term "vulnerability" in the sense that one can point to like a O
\~ 14 speed limit.
15 And I'll get into a little bit more of this when we 16 talk about the screening criteria and the like.
17 DR. KERR: My impression is that when this area was 18 first defined, and the term outliers was used, that it was 19 anticipated that one would not look at every sequence in every 20 plant or every sequence having a probability bigger than 21 something but rather, one would look for things that weren't 22 just unique to a plant but that rather, contributed a rather, 23 large, disproportionate even share of risk.
24 My impression is that as this process has developed, 25 that aspect has been lost and that now one is being asked to Heritage Reporting Corporation (202) 628-4888
g~s 20 Q l look at all sequences that make significant contributions to 2 core melt, and not just look for initially were called 3 outliers.
4 Do I have a misapprehension or has there been a
'S change in philosophy?
,6 MR. SHERON: I don't think it's been a change in-
-7 philosophy. We would hope that this process will very 8 quickly -- you know, one of the things we want to do is get 9 the letter out so that if there are any ten to the minus two or 10 ten to the minus three sequences out there, they get identified 11 and fixed. Those are the true outliers, you might say.
12 There are other sequences that may be of lower 13 probability. We are asking them to identify t' hose and they are
}
14 going to have to take a look at them and say is there something 15 that can be fixed, as the Commission pointed out, very 16 inexpensively or in a pragmatic sense, that can either 17 substantially reduce that probability or eliminate it.
18 So while the first goal is to get the big outliers, 19 you might say, identified and out on the table, we still want 20 to look a little bit further down, because we're recognizing 21 there are uncertainties in a PRA.
22 And we had, as you heard, on the A-45 for Point 23 Beach, I'm sure, the industry came in with a n mber that was a 24 factor of 30 than the staff and one might draw totally 25 different conclusions regarding whether or not something needs O Heritage Reporting Corporation (202) 628-4888
p 21 L) I to be fixed.
2 And you could trace, we had extensive meetings with 3 the industry-and we were able to trace many of these 4 differences back to such nebulous things as assumed probability 5 that an operator will take an appropriate action and so forth.
6 And these are areas where there is still considerable 7 uncertainty and disagreement, not only between the industry and-8 the staff but among just general PRA practitioners.
9 So one at this time doesn't want to preclude and base 10 all of their decisions on bottom line numbers by saying gee, if 11 it's less than ten to the minus four, I'm not going to fool 12 with it.
13 I think we want to take a little bit closer look and
\
14 understand that even if it is down in the lower range, that if 15 there is something that is a practical fix, one should probably 16 make it.
17 DR. KERR: I ask the question, because we have asked 18 the same sort of question in connection with our review of 19 NUREG 1150, namely, can one really judge whether a reactor is 20 safe or not without doing an individual plant PRA?
21 And the answer that we received from some members of l 22 the staff, I don't think it's an official position of the 23 staff, but the answer most closely associated with PRA at the 24 time was no, we don't believe you can tell whether an l
25 individual plant is safe without doing a PRA.
l
() Heritage Reporting Corporation (202) 628-4888 1
I
- O: -1 I could interpret what.I read in this letter-as_being 2 an effort to persuade all operators-to do an individual plant 3 PRA in order to determine.whether'the plantris safe or not.
4 I don't know whether that's what is being suggested.
5 MR. SHERON: When you say PRA, do you mean PRA or IPE 6 or interchangeably?
7 DR. KERR: 1 me'an PRA, because a significant-8 preference is expressed in the letter for a PRA, indeed perhaps 9 even a Phase 3 PRA, a complete PRA. ;
10 MR. SHERON: Well, obviously, a PRA has benefits, 11 some benefits, over an IPE in terms of perhaps depth of 12 analysis and so forth.
{ 13 Keeping in mind though that when one weighed the 14 costs against the benefits of doing this, the Commission :
i 15 instructed the staff to work with the industry to find 16 simplified methods.
17 We think that the methods that have been developed by 18 the industry, coupled.with the enhancements that we are 19 proposing, are suitable for determining whether or not the t 20 plant has any outliers.
21 We are not claiming that this method is foolproof and 22 it's going to find everything. Again, it's based, for example, 23 the IPEM is based on like having reference plants. So it's not 24 perfect, but we think it's good enough to meet the intent of -
25 the Commission's policy.
O Heritage Reporting Corporation '
(202) 628-4888
23 L
f')3
( 1 MR. SHERON: Very quickly, the benefits, and I guess 2 that you have heard this before. We think that doing plant 3 specific PRAs during IPEM will yield insights to unique severe 4 accident vulnerabilities, if they exist, that could lead.to low 5 cost modifications that would eliminate them.
6 Again, I cannot stress enough the utility involvement 7 in understanding the possible range of severe accident behavior 8 in the plant. And thus, would be better prepared to prevent or 9 mitigate an accident progression sequence. And I will get into 10 a little more detail on this when we talk about the accident 11 management.
12 And again, once you finish with an IPE, we would hope 13 that the insights that are gained by the utility. namely
\l 14 recognition of the role of the prevention and mitigation 15 systems and any operator actions that need to be taken, as well 16 ~ as improvements like hardware and procedure changes resulting 17 from this, would be fed back into a severe accident management 18 program.
19 MR. HOUSTON: Brian.
20 MR. SHERON: Yes.
21 MR. HOUSTON: I have three questions on item two. I 22 notice that the CRGR letter of April 25th says, if I interpret 23 it correctly, item one, that using an existing PRA would not be 24 acceptable, because the utility personnel would not be 25 adequately involved in the assessment, and that that would not O Heritage Reporting Corporation (202) 628-4888
- rl 24 C'
1 meet the intent of'the IPE.
2 MR. SHERON: Right.
3 MR. HOUSTON: But do you not think that it would be 4 adequate if the utility did a careful screening of the plant to 5 make sure that an existing PRA either was representative of'the.
6 plant-or could be modified to be representative of the plant, 7 plus adding what advances have been made in PRA to bring it up 8 to the current state of the art, would you consider that as an 9 acceptable alternative?
10 MR. SHERON: My personal answer is yes. If the 11 utility demonstrated that their staff became heavily involved 12 in confirming say the adequacy of an existing PRA and its 13 update.
)
14 One of the biggest concern is that the plant is a 15 constantly changing thing, okay. And you want to make sure 16 that if a PRA was done three years ago, that it may not reflect 17 the configurations that are in the plant today.
l 18 So we would want the utility (a) to confirm that the 19 PRA that it is submitting, if it is on record before, has been 30 updated to reflect the current design. And number two, would 21 be to demonstrate how their staff was involved in either this 22 update or review or whatever, that gives us the assurance that 23 it is the plant utility staff that understands what was done.
24 MR. HOUSTON: I concur with that, but the CRGR letter 25 seems to exclude that. And maybe I am misinterpreting that
- (
l Heritage Reporting Corporation (202) 628-4888
. , ~ - - - - - - , - , - - - - . . , , , . - - - ,.
a 25
(~)
V 1 item one, but it seems to be fairly definite. Thank you.
2 DR. KERR: Are you referring to Gary Quitschriver's 3 memo when you talk about.the CRGR letter?
4 MR. HOUSTON: 'Yes. I am sorry, that is right.
5 DR. KERR: Okay.
6 MR. SHERON: Okay. Let me discuss now, if I could, 7 the treatment of external events. First off, I think that we 8 have to recognize that the Commission's severe accident policy 9 does not exclude external events. When the industry embarked 10 on their methods development, it was restricted to the internal 11 events. There was a plant developed, I forget the number, 12 sometime ago that proposed treating external events, a plan for 13 treating them.
14 We have since been working to develop a method for 15 treating external events. We have had our contractor at 16 Lawrence Livermore and Mr. Budniks at Future Resources working 17 on this program with us.
10 And we have recently, and I will tell you in a 19 second, expanded our work in this area. But the bottom line is 20 that we are not ready yet to go forward and ask *.he industry to 21 spend considerable resources examining the external events with 22 regard to possible initiators. And they are restricted to 23 proceeding only with the internal events.
24 MR. MICHELSON: Brian, I need a clarification.
25 MR. SHERON: Sure.
O Heritage Reporting Corporation (202) 628-4888
f- 26 (s 1 MR. MICHELSON: In the proposed letter on page three, 2 it talks about externally initiated events.
3 And my question is is a pipe break an externally 4 initiated event or an internally initiated event?
5 MR. SHERON: If it is initiated by seismic, it would ,
6 be externally initiated.
7 MR. MICHELSON: How about if it is initiated by
'8 corrosion or whatever?
9 MR. SHERON: I would call that internally initiated.
10 MR. MICHELSON: And if that type of event going to be 11 in the first? ,
12 MR. SHERON: I doubt it.
13 MR. MICHELSON: Well, some internal events as well 14 are going to be excluded from this examination presently then.
15 I thought that you said that all internal would going to be 16 done and all external later. And now you are telling me that 17 if a pipe breaks from internal causes.
10 MR. SHERON: I am saying that the utility is going to 19 have to examine its system and walk it down.
20 MR. MICHELSON: I understand.
21 MR. SHERON: One of the things that they are going to 22 have to do is look at operational experience, and take that 23 into account.
24 MR. MICHELSON: But what are they looking for if you 25 are asking to look for internal events; are they looking, for f\
NJ l Heritage Reporting Corporation ,
i (202) 620-4888 L
b .
^
27
,[a 3_
1- instance, at pipe' breaks caused by internal conditions?:
2 MR. SHERON: I think that they should be looking-3 for them'. Whether they.actually come up.with.something that-4 passes their screening criteria.
5 MR. MICHELSON: That is part of the first look, under 6 this letter you are going to look for pipe breaks?
4 7 MR. SHERON: Yes..
38 MR. MICHELSON: Okay. That is.what I wanted to make
'9 sure of.
10 MR. SHERON: We have discussed this with NUMARC. I 11' guess that we first approached it back in November with the 12 IDCOR steering committee meeting. I have gotten a commitment 13 from Cordell Reed sometime ago that the industry would indeed 14 take on a commitment to work with the staff to address how best 15 to treat external events from the standpoint of coming up with 16 a defensible means of saying that we have met the intent of the 17 Commission's policy statement.
( 18 The question that we are struggling with right now is 19 how does one go about doing external events. As you know, 20- dealing with something like seismic on a probabilistic basis, 21 it has a lot of uncertainties. . People always raise the 22 question of how do you deal with low probability high 23- consequence events and so forth.
24 Right now, as I said, we have an examination process 25 going on to see which external events need to be treated, and O Heritage Reporting Corporation I
(202) 628-4888
28
,e...
'$ I 1 which ones may already be adequately treated-say-with a design 2 base.
u 3 And our work to date'shows us that there=are some 4 external events, and the one that comes to mind is seismic, 5 .that really need to be looked at by all plants. There are some 6 ' external events that are unique only to a few plants. For
- 7. example, if a plant is very low compared to its heat sync let's ,
8 say, a river or a lake, flooding might be a problem for that 9 plant. Where say it was situated relatively high, it may not ;
10 have that same concern.
11' And then there are some external events, as I said 12 here, that may be acceptably treated by an existing design 13 basis. There is concern,.for example, that transportation
~ 14 accidents may have been adequately considered in the design !
't 15 base from the standpoint of all credible cazardous materials 16 that might be transported in the vicinity of the site. And 17' there is nothing that is changed that says that one has to look ,
18 at something that may have a different probability. ;
19 And lastly, which again is a rather major concern I 20 guess not only on the part of the staff but the industry 21 expressed this as well, and that is that we do have a number of 22 ongoing programs dealing with external events primarily 23 seismic. There is the seismic design margins program. There 24 is A-46. There is the eastern seismicity program.
25 And the last thing that we wanted to do was to go O Heritage Reporting Corporation (202) 628-4888
r o .
9
_rd .
- 1 forward and'tell the industry, for example, to go off and do-a 2 seismic PRA or a seismic margins walk-down, and then come at
.3 -them two other olmes, and tell them to do the same thing in 4- order to solve the specific concerns of say A-46 or.STNP.
-5 So before we go out and ask the' industry to do 6 something, we want to make sure that what we want to ask them 7 to do is that they only have to do it one time, you know, if we
,8 can solve all of these problems.
9 MR. MICHELSON: Can you tell us just briefly your 10 thoughts about internal fires thus far?
11 MR. SHERON: I am sorry, towards internal fires?
12 MR. MICHELSON: Yes. The one listed here has an 13 external event.
14 MR. SHERON: Yes, and that will be studied.
15 MR. MICHELSON: Would you be looking at fires that 16 burn longer than the aesign basis, like if it is in a one hour 17 prot.ected area that the fire burns an hour and a half, is that 10 kind of the way that you look at it?
19 MR. SHERON: John, you may want-to reflect what your 20 people are coming up with right now.
21 MR. CHEN: I am John Chen from the Research staff.
22 In regard to the fire, we now have a study, a study 23 to identify certain things which were not located in the past.
24 So we are having some consideration right now within Research 25 to determine what aspect we want to look at.
O Heritage Reporting Corporation (202) 628-4888
Q p 30 V
1 At the same time at current, what we are thinking 2 about with regard to fire risk, we talk about the PRA itself, 3 and it will probably will give us sufficient. However, we also 4 recognize that the current Appendix R still will pose some kind 5 of residual risk at the plant site, and we are studying that.
6 We are having another committee, which will look into it in 7 much more detail.
8 With regard to how long these will last, for 9 instance, a fire burning one hour or one and a half hours, what 10 the impact will be, what the methodology there should be used, 11 and how to address these sort of things. In our subcommittee 12 meeting, that is going to be discussed.
13 MR. MICHELSON: You talked about the fire scoping 14 study. The one that I am aware of is not dealing with severe 15 accidents, but it is dealing with design basis.
16 MR. CHEN: That fire scoping study is not dealing 17 with -- itself addresses from a probabilistic point of view, 18 and tries to identify what is the fire risk associated with an 19 unaccounted for scenario.
20 MR. MICHELSON: But it is dealing with design basis, 21 I believe.
22 MR. CHEN: But from the probabilistic point of view.
23 MR. MICHELSON: I did not find severe accident in 24 there.
- 25 MR. CHEN
- No. Severe accident itself, like the O Heritage Reporting Corporation j (202) 628-4888 e
m
k
~
31 1 past, when we talk about the size. We say if you satisfy 2 Appendix R, then we consider that design basis. You just 3 determine what kind of design and how you are going to minimize 4 fire risk. But if you use the probabilistic point of view, you 5 still have some residual fire risk.
6 And right now, we are trying to identify this 7 residual fire risk, whether that is going to pose any problems 8 to us, but that is the assessment.
9 MR. MICHELSON: That is going to be later then?
10 MR. CHEN That is going to be later.
11 MR. MICHELSON: Thank you. '
12 MR. WYLIE: Let me ask a question. In the [
i 13 consideration of the seismic, or fire risk, or whatever, are i 14 you going to include interactions as part of that application? ,
15 It is not now. That is the reason that I.am asking the ,
16 question. It is not now in the programs that you have got now, t 17 in A-46.
18 MR. CHEN: As far as interactions and those 19 interaction problems, there are certain things that we will l r
20 definitely address, and there are certain things that we will -
21 not be able to address because of the scope itself. The system 22 interaction in regard to the spatial aspect, we are definitely 23 going to address.
24 MR. WYLIE: By spatial, do you mean falling down in 25 something?
l Heritage Reporting Corporation (202) 628-4888 i
m~
32 1 MR. CHEN: For' instance, you have the equipment and 2 components within that, given that the ceismic event itself can
.3 interfere against ec h other, and that will come out.
4 MR. WYLIE: Suppose that it is a tank that falls and
, 5 spreads its content throughout the building, are you going to 6 look at that?
7 MR. CHEN: Yes.
8 MR. WYLIE: That $will be looked at?
9 MR. CHEN: Yes. But certain things like if we talk 10 about seismic induced and seismic inde.ced suppressor systems, <
11 that we could still evaluate.
12 MR. MICHELSON: Are you going to look at the !
1.1 migration of heat and smoke that will set off unwanted fire 14 mitigation?
15 MR. CHEN: That is a little too much detail for me. I 16 am not too sure about that. .
\
17 MR. SHERON: Let me skip, if I could, this viewgraph.
l
, 18 I think that it is pretty self-explanatory. ,
l 19 MR. MICHELSON: If you want to do it later, that is 20 fine.
21 MR. SHERON: Well, the next viewgraph, this basically ;
22 says why we are holding e'f. Let me tell you right now what we j 23 are doing, and I am going to make a suggestion to the committee i i
l 24 here.
l 25 With regard to this whole question of how we treat l
i
( Heritage Reporting Corporation l
- (202) 628-4888
,o 33
(.) 1 external events from the standpoint of severe accident policy 2 and stuff. We recognize that this is a big issue, and it is a 3 controversial issue.
4 And to address that, Mr. Beckjord, I guess that it 5 was about last December, appointed a steering group, a senior 6 steering group. And the chairman is Larry Chou from the Office 7 of NRR. The members are Tom Novak from OEOD; Guy Arlato from 0 Research; and myself from Research.
9 The purpose of the steering group is to come up with 10 and recommend to senior management how best to treat external 11 events in the context of the IPE and the severe accident 12 policy. i 13 And another goal is to again ensure that whatever we i
14 recommend as the treatment is properly coordinated with all of 15 the other agency programs that are ongoing with regard to 16 external events. And the key here is that we are shooting to 17 complete this task in about eighteen months.
18 DR. SIESS: And the task that you are going to 19 complete in eighteen months is to decide what to do?
20 MR. SHERON: The proposal right now is -- and what I 21 was going to suggest is that it might be appropriate for the i
22 subcommittee to ask Dr. Chou periodically to come down, whether 23 it is this subcommittee or another one, to come down and to
! 24 discuss in detail what the steering group is doing, and what 25 its plans are and schedules.
O b
Heritage Reporting Corporation (202) 628-4888 5
() 1 This is being formulated right now. There was a memo 2 that was just issued by Dr. Chou which made assignments or 3 three separate subgroups. One to deal with seismic, which is 4 headed up I believe by Leon Wrighter. There is one dealing 5 with fires and floods. And another one dealing with high 6 winds, tornedoes, and everything else.
7 So there are three subcommittees. Each one is 8 composed of about four or five members with experts from within 9 the staff. The continuity is provided by John Chen, who is a 10 member of all three subcommittees.
11 Each subcommittee is supposed to come up with a 12 workable plan for establishing what they believe needs to be 13 done from the standpoint of the severe accident policy. The 1~ intent is in eighteen months to have a report, a NUREG report, 15 which will basically provide guidance. It will be the guidance 16 document that says we have examined external events from the 17 standpoint of severe accident policy. Here is what we the 18 staff thinks needs to be done by the industry in order to 19 effectively satisfy the Commission's policy intent.
20 DR. SIESS: You know, you repeatedly talk about how 21 you treat external events. And yet as I sit in these meetings, 22 I hear more questions about how you define an external event.
23 Please do not leave that out.
24 MR. SHERON: Okay.
25 DR. SIESS: And I think that somebody needs to think O Heritage Reporting Corporation (202) 628-4888
O I about how we ever got into this business of-distinguishing 2 between internal events and external events. All of them end 3 up affecting the plant, whether they come from one source or 4 another.
5 I think that we got into this because there is one 6 group of people who understand how the reactor systems operate 7 and another group of people who think about floods and 8 earthquakes. But it is an absolutely artificial distinction 9 .that is getting us nowhere but in trouble. There is no such 10 thLng as an external event that causes a core melt. That core 11 is inside, not outside.
12 MR. SHERON: Brian, as a follow-up on that point and 13 as a comment on one of the earlier replies that you gave me, I 14 -think that you ought to go back and look very carefully at pipe 15 break. The reason is that we have gone to great lengths to 16 define certain areas in which we will accept leak before break.
17 We did this on a probabilistic basis.
18 We said that for design purposes that the 19 probabilities were acceptable. But it you go back and lock at 20 those probabilities, you will find that they fall well within 21 the severe accident range. And you have go to back and look 22 now.
23 Because some of these pipe breaks cause whips, if 24 they were to occur with that very low probability, which 25 penetrate containment as part of the initiating event, and you O
Heritage Reporting Corporation (202) 628-4888
. 36 1 end up with-a potential for very severe consequences. Because 2 .for one' thing, the penetration of containment-leads to the loss -
3 of the equipment outside of containment, which can get to core 4 melt very.quickly.
'5 And I think that you-need to define beyond design 6 basis pipe breaks as external events,.and look at them more 7 carefully before you decide'what the utility ought to do. And 8 that is why I baited you a little Llt to find out whether you 9 really meant pipe breaks. I think that within a design basis 10 that pipe breaks should now be a part of the study.
11 But beyond the design basis pipe breaks, the ones 12 that we deal with in leak before break and so forth, clearly 13 there is some kind of another class of events. And you are 14' going to need a lot of thought. Because I believe that I can 15 walk you right into a trap, if you are not careful.
16 MR. SHERON: Let me skip now to Section 4 of the 17 Generic Letter, which are the methods of analysis. The letter
, 18 itself specifies that several options can be used ts satisfy 19 the examination of requirements.
20 Let me preface this before I get into this by 21 pointing out that if we are to issue the letter under i
22 50.54(f), that my understanding from the legal staff is that 23 the letter cannot proscribe the method of information l
24 collection.
- 25 We can ask that the information be collected and
! (
l Heritage Reporting Corporation j (202) 628-4888 i
I i
c' 37 L,)3 I provided to the staff, but my understanding is that we cannot 2 proscribe very rigidly that this is the way that you must 3 collect it. Otherwise, you go beyond the 50-54(f).
4 So what we have said in the 1c Lter is that there are 5 basically three methods, and it is the third one which I think 6 gives us that flexible out, I would call it.
7 But the first one is that the IDCOR IPEM method, the <
0 front end, which coupled with the staff enhancements which are 9 in a SER that we plan to issue, plus a containment performance 10 analysis that is consistent with Appendix 1 of the Generic 11 Letter, and I will be talking about that a little bit later, 12 would provide an acceptable analysis.
13 A second acceptable way to perform the IPE is to 14 either do a Level-I PRA. Or if you already has a Level-I PRA, 15 to update it and involve the plant staff, as we discussed 16 previously, to submit that. Plus a containment performance 17 analysis consistent with again Appendix 1.
18 And by inference, you can see that if a plant has a 19 Level-II or Level-III PRA, that would be acceptable in the same 20 sense provided that the back end again was done consistent with 21 the guidance in Appendix 1.
22 MR. MICHELSON: Brian, when you talk about a 23 Level-I PRA, I can have quite a range of Level-I PRAs, as I 24 understand it, which deals with the completeness with which you 25 have done the modeling and so forth in the process of doing a O Reporting Corporation Heritage (202) 628-4888
1
. . 38
=U: 1 Level I-PRA.
2 System interaction, for instance, how well have-you 3 modeled it in your PRA. Because that is the sort of thing.that 4 begins to pick up some of these out-liers.
5 So to say that you have got a Level-I PRA alone is 6 not quite an adequate proscription, is it? You have got to 7 talk about completeness or omissions, and you have to talk 8 about human factors integrated into it.
9 MR. SHERON: We have provided a reference. I believe
-10 we referenced, was it NUREG 2300, is that it?
11 MR. ELTAWDA: we reference NUREG 45.50 and 12 NUREG 2300, and NUREG 28-15.
13 MR. MICHELSON: I think that those were kind of 14 general methodology descriptions. And you have got quite an 15 option though as to what you do.
. 16 Are you saying, for instance, that all system
~
17 interactions would be a part of a Level-I PRA?
18 MR. ELTANDA: Yes, I think so.
19 MR. MICHELSON: They are not presently. I mean there 20 are various attempts that I have seen, for instance, to model 21 system interactions, vatious attem"ts to model the human 22 factor. Some of them are rather crude, and some of them are a
- 23 little better. But they are all Level-I PRAs, as I understand 24 the definition.
I 25 MR. SHERON: It was not our intent in doing this to i
Heritage Reporting Corporat!.on (202) 628-4888 l
I 1 go in and provida a very proscriptive guide on how to do a 2 Level-I PRA.
3 MR. MICHELSON: But when I read your letter though, 4 it seems like in places that it talks all about human factor 5 inputs and so forth, like it is really going to be in there. I 6 am just cautioning.
7 MR. SHERON: We are asking the industry that they 8 should think about and consider this, and factor it in. But 9 you know, again please recall that this is not -- you know, the 10 letter that is going out is requesting them to provide the 11 information. It is not telling them that you just do a PRA and 12 you must do the following.
13 DR. KERR Brian, there is a reference in the letter 14 to state of the art PRAs.
15 Does the staff intend to give an example of one or 16 two state of the art PRAs?
17 MR. SHERON: We are proposing to --
18 DR. KERR: My question is does a state of the art PRA 19 exist?
20 MR. SHERON: Yes, there are state of the art PRAs.
21 DR. KERR: So in your final letter, you will list 22 these, that this is what we mean by state of the art PRA?
23 MR. SHERON: I am not sure that we were intending to 24 list in the letter what an acceptable PRA is.
25 DR. KERR No, I did not say acceptable. I said O Heritage Reporting Corporation (202) 628-4888
I
<~w ,
40 U 1 state of the art. Because the term is used, and presumably the 2 staff has something in mind.
3 MR. SHERON: You will hear a little bit later that we 4 are proposing workshops.
5 DR. KERR: If you know what you mean by state of the 6 art PRAs, do you have some in mind, unless you think that the 7 state of the art is going to change to rapidly.
8 MR. ELTAWDA: You find that in the Generic Letter on 9 page four, as to acceptable PRAs.
10 DR. KERR: Now wait a minute. That is not the 11 question that I asked. I am asking whether there exists 12 somewhere a state of the art PRA.
s 13 For example, are the PRAs in 11-50 state of the art
)
14 PRAs?
15 '
MR. ELTAWDA: Yes. That is NUREG 45-50.
16 DR. KERR: So if one did a PRA the way that they are 17 done in 11-50, that would be in your view a state of the art 13 PRA?
19 MR. ELTAWDA: Yes.
20 MR. MICHELSON: And you think that those include l
21 system interaction?
22 MR. ELTAWDA: The system interaction that you are 23 describing, I do not know what you mean.
j 24 MR. MICHELSON: Well, you use the words in your 25 Generic Letter here, too. And I am saying that I am not sure Heritage Reporting Corporation l (202) 628-4888 l
1 that there is a state of the art PRA that includes adequate 2 modeling of system interaction.
3 MR. SHEWMON: State of the art means the best 4 currently available, rather than the best that we can imagine, 5 does it not?
6 MR. SHERON: Correct.
7 MR. MICHELSON: There is not any then.
8 MR. SHEWMON: There is not any that does what you 9 wish it would do.
10 MR. MICHELSON: And that is what I am try<.no .o find ,
11 out, if that is intended to.be upgraded to such a state or not. .
12 MR. CORRADINI: If I could ask a follow-up question 13 to go back to what was asked at the beginning.
14 If under NUREG 11-50, the five of the six reactors 15 there are being looked at in that methodology as state of the 16 art, is there an acceptable PRA that meets this? -
17 MR. ELTAWDA: There are PRAs. The Indian Point PRA 18 is acceptable. We have a design PRA. We have the Limerick 19 PRA. There are a lot of PRAs. But the point that Brian is 20 trying to make is that the utility has to look at these PRAs 21 and upgrade them, to sure that they reflect the plant as 22 designed.
3 MR. SHERON: If a utility, if they took the
- l 24 11-50 PRA, and came in and told us that the went through it, 25 and that they cross-checked it against their design, and that ,
O Heritage Reporting Corporation 7 (202) 628-4888 .
i i
_ _ _ _ _ , - _ . _ _ _ _ _ __,..____ . - - - . - - . - - _ ~ _ . - -- - - - . . ---.
b 42
~
1 they upgraded it where it was appropriate, where the design may 2 have changed, that their plant personnel had been through it, 3 and were heavily involved in the review and upgrade and so 4 forth, such that we walked away-with confidence that they met 5 the intent of the Commission's policy; yes, that would be an l 6 acceptablo PRA.
7 DR. SIESS: Suppose that they came in and said all of (
8 that, except that Pickett, Lowe & Garrett had done it with 9 their own staff, would that be an acceptable state of the art 10 PRA?
11 MR. SHERON: The PRA may be an acceptable state of 12 the art, but satisfying the Commission's policy would not' ;
13 acceptable.
O 14 DR. SIESS: In other words, they could not satisfy
. 15 this letter unless they themselves with people that they had i
16 employed had reviewed the PRA?
17 MR. : HERON : Correct. !
18 DR. SIESS: Now those people still have to be on !
19 their ataff? [
i 20 MR. SHERON: Well, I cannot control their hiring.
f 21 DR. SIESS: Suppose they hired somebody for three [
i 22 years, reviewed the PRA, and then let them go. I mean you ere 23 making a point as to what is acceptable. I am a utility, and I I 24 have to have a PRA that is going to cost me a lot of money and I
25 a lot of time, how do I do it.
Hericage Reporting Corporation (202) 628-4888
43 1 MR. SHERON: I guess that my feeling is~that if there 2 are utilities out there that are going to be so devious as to 3 try to circumvent the intent, okay, they will be able to do it.
4 DR. SIESS: That is trying to.get the job done.
5 MR. SHERON: Well, I cannot put in this 6' letter --
~
7 DR. SIESS: You told them-that they have to do it in 8 three years. And the chances that.they are going to do it in 9 three years with people that they already have on their staff i 10 is very' slim.
11 MR. SHERON: We did not say that they have to do the 12 entire thing with people on their staff. The people on their 13 staff have to be involved. They can hire Pickett, Lowe &
. 14 Garrett or anybody else that they wish to perform an IPEM, or 15 a PRA, or the like. But we have said that their staff must be 16 involved in that process and work with those people. But they 17 do not have to do all of the work themselves.
18 DR. SIESS: And you expect those staff to say around 19 until the accident happens.
20 MR. SHERON: You know, I cannot predict what the 1
21 future is for the plants.
I 22 DR. SIESS: I mean presumably you want that 23 experience within the utility.
24 MR. SHERON: Yes.
25 DR. SIESS: Now the corporation does not really know IO i Heritage Reporting Corporation j (202) 628-4888 L ---
,_s 41 b 1 anything, but it is the people.
2 So if you want that experience, how are we going to 3 know that experienco'is still going to be there so many years 4 from now,.have you thought about that?
5 MR. SHERON: Right now, the Commission did not say 6 that the plant was supposed to maintain that expertise. But 7 let me tell you how it will be maintained I think, and you will 8 hear a little bit later.
9 One of the biggest outgrowths I think of doing an 10 IPE, we believe, is going to be a severe accident management 11 program in the plant. Hopefully, that will be something that 12 is documented and put into a training program and the like for 13 plant personnel. And the insights that one gains from the IPE O 14 will be embodied in that accident management program.
15 And as new personnel come on board at that plant, I 16 would presume that part of their indoctrination process would 17 lua training in a severe accident management program. And those 18 insights would be transferred at that time.
19 DR. SIESS: That would be nice.
20 MR. SHERON: That is our intention.,
21 DR. SIESS: I woulo hope. i 22 MR. LEE: These two options that you are describing, 23 do you see any difference between IPEM and Level-I PRA as far 24 as Appendix 1 fitness is concerned?
j 25 MR. SHERON: From Appendix 1?
Heritage Reporting Corporation (202) 628-4888
F~ ~ ,
[
-s 45 l) 1 MR. LEE: Right.
2 MR. SHERON: Probably not.
3 MR. LEE: So Level-I PRAs the way that you have 4 defined should be equivalent to IDCOR IPEM plus some of these 5 enhancements, should be equivalent to Level-I PRAs?
6 MR. SHERON: Yes.
7 MR. LEE: But then back end treatment would have to 8 come in either way, in either option?
9 MR. SHERON: Right. Again if the industry does 10 propose, if someone in the industry does propose another 11 methodology, again we would probably want to take a look at it 12 and review it, to make sure that we would find it acceptable 13 before they went ahead and spent a lot of money, but we are not 14 precluding any options at this time.
15 The IDCOR front end method, we have been down here 16 and discussed this, so I will just kind of briefly run through 17 it. The IPEM method cover sequences initiated by internal 18 events, loss of offsite power, and internal floods.
19 And it jnvolves the basic following elements of the 20 Level-I PRA. First off is the plant familiarization and 21 systems notebooks. You know, these are basically all of the 22 data that one needs to put together to perform either the PRA 23 or an IPEM, schematics, interface data, etcetera.
24 The accident sequence definition and quantification, 25 recognizing the important role played by support systems. The O Heritage Reporting Corporation (202) 628-4888
46
.O '1 Interdependence among support systems should be analyzed to 2 identify support states.
3 MR. MICHELSON: Brian, on that particular bullet, 4 again it is indicative of the same question on system 5 interaction. We think that we know that if.y>>u lose, for 6 instance, instrument air completely, that we know how devices-7 are supposed to respond. But we are finding that if we only 8 lose the pressure partially but not totally, that we are 9 finding a different set of responses. Such responses were not
- 10. included in the PRAs of the past for various reasons.
3 11 Is it intended now that we go back and take a sharper 12- look at over voltage and under voltage, and over pressure and l 13 under pressure on control air and things like that, is that a
() 14 part of this look to see if we get into a severe accident
! 15 scenario?
- 16 MR. ELTAWDA
- No, we are not going to look at these
, 17 in more detail in the existing PRAs.
18 MR. MICHELSON: And there is some basis to believe 19 that it is a non-problem, or you just do not know how to do it, 20 or what is the reason for not including that sort of thing?
l 21 MR. ELTAWDA: I think that the reason now is that it 22 is not state of the art.
l 23 MR. MICHELSON: Well, we will never get it into the i 24 state of the art, if we do not first of all decide whether it l
25 is a problem.
) Heritage Reporting Corporation (202) 628-4888
o 47
- ~# 1 Are you trying to decide whether it is a problem as a 2 part of this program?
3 MR. ELTAWDA: No, not at this time.
4 MR. MICHELSON: Is it some other program where it is 5 being decided?-
6 MR. SHERON: I would say let us take a look at it 7 with regard to how feasible it is to include it.
8 MR. MICHELSON: Traditionally, you are just looking
'9 at loss of voltage, or loss of air, or loss of water, but you 10 never look at what happens if it is only partially. And you 11 can read LERs every day, and you can see how things really 12 happen, what really happens if they lose partial air pressuro 13 and that sort of thing. It is not the set that you necessarily
( 14 envisioned at all.
- 15 MR. SHERON
- Again I would point out that the 16 industry, one of the key parts that they are supposed to do in 17 this IPE is to look at the operational data. And if they 18 determine, for example, that partial loss of air, or voltage or 19 something has happened at the plant, I would think that they 20 should consider it then, as just something that is identified 21 by their operational data review.
22 MR. MICHELSON: Well, it may not have happened at 23 that particular plant, but it may have happened at another 24 plant. But it is a generic problem, and it should be looked at 25 by all plants. But it does not seem to be a requirement. I
.O Heritage Reporting Corporation 2 (202) 628-4888 v -. - --- - - . ~ - . - , . - . - - - - , < . -
w 48 1 thought that we were dealing now with severe accidents beyond 2 the design basis. And indeed, the design basis was you design 3 for loss of air, and you design for loss of voltage. But there
. 4 _has not been a design basis so'far to consider degradation 5 versus total loss.
6 DR. KERR Well, let me make a suggestion. I believe 7 that you indicated that you are going to get together with the 8 utilities in workshops, and give some indication of what the 9 staff at least would find acceptable.
10 Are you going to tell the ACRS at some point about 11 this sort-of thing?-
12 MR. SHERON: We are' going to be developing a review 13 plan for how the staff and its contractors would go about 14 reviewing the submittals, and what we will be looking for.
15 DR. KERR: It seems to me at that point that one 16 might well introduce these things.
17 MR. MICHELSON: Except it might be getting late 18 already for the utilities to be including it, unless we see 19 this review plan very closely at the front end.
20 DR. KERR: I say that at least the impression that I
! 21 get from reading this is that workshops are going to be held, i 22 and they would be held fairly soon, because the utility has to i 23 come in.
24 MR. SHERON: Without two to three months after the 25 issuance of the letter.
! O Heritage Reporting Corporation (202) 628-4888
4 49 r'g.
'# 1 MR. MICHELSON: But this would not be a subject of a 2 workshop discussion if the staff does not bring it up. I 3 suspect that the utility will not bring it up.
4 DR. KERR: What I am saying is if they bring this to 5 us at some point, that it is at that point that we can have 6 input into the process.
7 MR. MICHELSON: Oh, yes. Just a forewarning, I 8 guess.
9 MR. SHERON: We had intended to come back down to the 10 subcommittee with our review plan as we develop it.
11 Well, the last two items were data assessment and 12 parameter selection. And as I guoss I pointed out, Carl, 13 analysis of plan specific experience in terms of failure rate O 14 and initiator frequencies. I would hope that if a plant had 15 some problems with thoso type of systems, voltage or undex 16 voltage, or partial loss of air, that that would show up at 17 this point.
18 One thing also that we are going to be doing as we 19 review these is if we start seeing some plants, for example, 20 identifying initiators that they feel are something that they 21 need to consider in their plant, and we find that there are 22 identical plants that do not, we may ask them the question as 23 to why they are not considering it, if there appears to be no 24 valid reasons why they are unique.
25 You know, one thing that I guess that the staff is O Heritage Reporting Corporation (202) 628-4888
,., 50
- 1 going to be performing in an unofficial manner is like a 2 clearinghouse.
3 DR. KERR: Would you anticipate that some stations 4 might be able to make the case that nothing need be done?
5 MR. SHERON: Some might. Typically, what we found 6 out is that when a plant does a PRA or an examination, that 7 they will most likely fix something right then and there, and 8 then reanalyze the plant as though that fix were there. We 9 might not even learn about it, for all that I know.
10 You know, I am sort of convinced that a plant is not 11 going to come in and say here is all of the vulnerabilities 12 that I found in my plant, as though I was sitting on a ragged 13 edge of a cliff for the past umpteen years.
14 My guess is if anything that they will identify 15 enhancements, I would call it, that they found, ways to improve 16 the plant.
17 DR. KERR: Let me go a bit further. 11-50 has 18 analyzed a number of plants without saying whether those plants 19 are acceptable, but simply has said here are the results.
20 Is the staff going to make a judgment that those 21 plants would or would not be acceptable without further 22 changes?
23 MR. SHERON: I will answer the question. And that is 24 that we will be taking a look at the results that are submitted 25 by every utility. And we will be basically looking over their O Heritage Reporting Corporation (202) 628-4888
51 1 shoulder regarding whet their judgment was.
- 2 DR. KERR: I am trying to get a po.i.nt of reference.
3 Because I can look at 11-50 and see what war found there, and 4 see what the numbers are, and what they have done. And I am 5 not asking if you have looked yet.
6 But do you plan to look at cay this set of plants cnd 7 decide that yes, we would find those acceptable, any of those 8 changes, assuming that the staff is involved and understood 9 what was going on, or that we would not find this acceptable, 10 that something would have to be done. I mean that sort of 11 guidance, it seems to me, would be quite helpful to people, 12 MR. SHERON: We had not planned to go and to
-m 13 specifically draw those kinds of conclusions on the 11-50 U 14 plants. That would involve us making a determination, for 15 example, of how much it cost to make a tix.
16 DR. KERR: But you have got to do something like this 17 before you tell your contractors what to look for. And if you 18 cannot give them some for examples, I do not see how they are 19 going to know what to look for.
20 MR. SHERON: Well, we will be giving them for 21 examples. And I am sort of jumping ahead into the review 22 process.
23 DR. KERR I will wait.
24 MR. SHERON: Okay. But the review is not joing to be 25 done by contractors alone.
O Heritage Reporting Corporation (202) 628-4888
52
( !
1 DR. KERR: I know that, but a significant amount
,2 apparently will be, and they have to have some guidance.
3 MR. SHEROll: Yes. We are going to have to go through 4 two or three reviews, lead plants if you went to call them.
5 And we are going to have to discover ourselves what it is what 6 we will or will not accept, what we should be looking for. And 7 my guess is that after we have gone through~three or four of 8 these that we will be in a better position to_tell the 9 contractors, for example, what they should be looking for.
10 DR. KERR: I keep coming back to 11-50. Not because 11 I think that it is perfect, but because it represents I think a 12 rather significant effort in terms of manpower and money. And 13 it was initially at least described as a baseline of risk.
14 Now since that risk has been baselined for those 15 plants, it seems to me that it would be useful to try'to look 16 at those and say we find these acceptable or that we would not 17 find them accoptable. I am not suggesting that you give me an 18 answer at this point, but it seems to me that if you do not do 19 something like that.
20 MR. SHERON: Let me give you a little bit of the 21 dilemma that we are in. In other words, if I just say that
! 22 they are acceptable or not. They may come in and they may have 23 certain core melt sequences that are down to the 10 to the 24 minus 5th. We, for example, have not looked at those and said 25 gee, is there any way that I can eliminate that sequence or put O Heritage Reporting Corporation (202) 628-4888
,% 53
- )
\'
1 it down to the 10 to the minus 7th range or something.
2 If I look at it, I may say I can do that, because it 3 requires installing a $25,000 piece of equipment. In which 4 case, I then have to do a cost benefit analysis in accordance 5 with 50-109 and the like, and make a judgment right then and 6 there whether based on all of the competing factors with the 7 significant improvement to safety, that the cost benefit ratio, 8 etcetera, the ultimate reduction in risk, or the core melt 9 probability, I would have to make a decision would I or would I 10 not go forward and propose to backfit this plant.
11 That is a considerable effort on the part of the 12 staff to do that for the 11-50 plants for those sequences. Now gg 13 what we are asking the utilities to do, and again I am jumping V 14 ahead here, is to look at the leading sequences that contribute 15 to core melt or poor containment performance.
16 And we are asking them to make a judgment on their 17 part, it is their plant, as to what do they think about these 10 sequences. Do they think that something practical can be done 19 to eliminate or reduce their frequency. And what we would like 20 to know from them is what are the results of their judgment on 21 this.
22 And what we are doing is that we are not trying to 23 second-guess them or anything, but to review what they have 24 done and to say that yes, your logic makes sense, you made the 25 right decision, and we agree with you.
i O Heritage Reporting Corporation (202) 628-4888 l
- . t: s 54
() .
1 DR. KERR: I read that, and I think that it is 2 commendable, except that it would seem to say that you need to 3 do this no matter what you think the present state of the plant 4 is.
5 MR. SHERON: No. We have nothing against them coming
, '6 in and saying all of the core melt sequences, for example, are 7 in the 10 to the minus 6 range.
8 DR. KERR I am talking about the total plant behavior, not individual sequences. The thing that makes the 9
10 plant safe or not, it seems to me, is how the total plant 11 performs, not how individual a6.quences perform.
12 MR. SHERON: Correct.
13 DR. KERR: So one therefore it seems to me is'being 14 asked to look at individual sequences and make judgments about 15 those independently of what they think about the risk posed by 16 the tctal plant performance.
17 MR. SHERON: If the total plant performance is 18 acceptable, then the individual sequences that make it up, I i
19 think by definition that they would find them acceptable.
, 20 DR. KERR: Well, one is not given a recipe or even i
l 21 asked to look, as far as I can tell, about what is the status 22 of the total plant. One is rather asked to look at individual 23 sequences, and screening criteria is given in terms of accepted 24 probability of core melt.
l 25 MR. SHERON: You get the overall status of the plant
( )-
Heritage Reporting Corporation (202) 628-4888
lv: . i 55-
- p. -
\_)
1 by looking at all of the individual sequences and'then adding l
2 - them up. And you come up with a' bottom line number that says 3 this is what the total core melt frequency'is. .
4 DR. KERR But there is not anything-given in hero 5- that says here is an acceptable total plan number. f 6 MR. SHERON: That is correct.
7 DR. KERR: Hence, it seems to me~that what you are 0: t
'8 asking is that they do the exercise and reach conclusions about
-9 individual sequences independently of'what the total plant 10 performance-is.
3 11 MR. SHERON: There is no criteria for the total plant 12 performance. l 13 DR. KERR Tnat is right. And therefore, what they 14 are being asked to do is independent. It has nothing to do 15 with the total' plant performance, because no criteria are given !
E 16- for total plant performance. ;
17 MR. SHERON: That is correct, j
- 18 DR. KERR
- Now it does not seem to me that it makes 19 sense to treat sequences in a plant that is very low risk in 20 the same way that I would treat sequences in a plant that maybe
- 21 had a risk ten times or a hundred times that. l
! 22 MR. SHERON: And thia process will follow that. T 23 DR. KERR It is not clear to me that it does from ,
24 what I have read, but perhaps. ,
t 25 MR. SHERON: Because we are saying that if one l
[ (
Iforitage Reporting Corporation I (202) 623-4888 !
-w,-.,_,---r,-m,- w,-- -,--+n,,---- c, - - =-. ,---m-m,-.. -,,w---,,g-- -.,-,--.---n.--,-,v, - -
-,-----,r-
4 56 1 utility comes in and says, suppose it is in the same system, 2 and let's even suppose that it is down to the same valve or 3 something, and they say if I replace this valve with one of 4 higher reliability or something, it changes this sequence from 5 one times 10 to the minus 5th or four times 10 to the minus 6 6th.
7 And another plant may come in and say if I change 8 that same valve, my core melt goes from one times 10 to the 9 minus 3 to two times 10 to the minus S.
10 And the first utility says based on that reduction, 11 it does not appear to be cost effective for me to fix that 12 valve, so I cm not going to do it. And the staff may say, yes, 13 we agree, that makes sense.
14 The next guy comes in and says, you know, 1 am not 15 going to fix that valve. And we are going to say, you know, wo 16 do not agree with you.
17 DR. KERR: Suppose a utility hac done a PRA, and 18 there are some, and come in and say a core melt freqlency, or 19 risk, or whatever they want to chose is low enough so that we 20 do not think that we need to do anything more.
21 Is that acceptable?
22 MR. SHERON: We will have to ree what their number is 23 and what their leading sequences are.
24 DR. KERR: So you have a number in mind.
25 MR. SHERON: No. We have to look at their sequences i
Heritage Reporting Corporation (202) 628-4888 i
l-_.
g 57 !
I 1 and decide. '
i 2 DR. KERRt I do not see why you have to look at- l i
=3 sequences. What you really want to know is what the total !
4 plant performance is. Now if you do not balleve that_ number, !
5 then you have to look at sequences. But if there is some f 6 performance criterion that covers the performance. -
7- MR. SHERON: The Commission did not give us any, t 8' MR. HOUSTON: Now wait-a minute. ,
- 9 Does not the safety goal give you that in terms of
- i 10 overall risk from the plant which is a total plant performance
, 11 index? !
- 12 MR. SHERON
- But I am already getting guidance from f
- e 13 the ACRS which says do not use the safety goal to judge ;
O 14 individual plants.
1 i
15 MR. HOUSTON: Yes, but your IPE says that you will ;
i 16 use a safety goal in conjunction with other. ,
17 MR. SHERON: It does, yes. f i
! 18 RDR. KZRR: As one measure. l 19 MR. .SHERON: As one measure, and I will explain that. ;
i t l 20 MR. HOUSTON: And that seems to capture the overall ;
i !
21 plant performance concept. l L 22 MR. SHERON: No. I will explain how we are going to
! 23 use the safety goal. [
24 MR. HOUSTON: Okay.
l 25 MR. SHERON: And it is not in terms of an overall l f
Heritage Reporting Corporation !
(202) 628-4888 i
- h I _ __ _ _ _ _._ __
,- w Ry ,
Ivy l, ? .-x 58 1 measure of performance. Now it is very difficult to
,2 understand, but we are not proposing to' establish a speed limit 3' or an overall measure of performance for plants, which says if
. 4 your number is this low. We are trying to stay away from 5 bcttom line numbers.
6 MR. HOUSTON: But certainly, that number is a 1
7 consideration.
8 MR. SHERON: Yes, it is a consideration.
4 9 DR. SIESS: If it is a consideration, you have got to 10 believe it. And a lot of people calculate it, but nobody 11 believes it, especially when they put the uncertainties on it.
I
! 12 That is the dilemma.
I
! 13 MR. SHERON: Okay, evaluation. The IDCOR IPEHs take O 14 advantage of available reference plant analyses, similarities i 15 among the plants, up-to-date knowledge of severe accident 16 phenomena. And as I pointed out before, it is a less rigorous 17 method than e full scope PRA. But nevertheless, we think that 18 it is acceptable to identify any potential vulnerabilities, t
19 MR. MICHELSON: Brian, what do you mean by a fuli 20 scope PRA in this case?
i 21 MR. SHERON: State of the art.
l 22 MR. MICHELSON: A Level-! PRA can also be a full l
l 23 scope?
1 l 24 MR. SHERON: Yes. I am sorry. When I used the words l
, 25 full scope, I mean state of the art PRA.
l Heritage Reporting Corporation (202) 628-4888 l
i 59 1 MR. MICHELSON: Level-I.
2 MR. SHERON: Obviously, the IPEM goes up through [
3' Level-II, you might say.
4 MR. MICHELSON: Well, okay, i
5 MR. SHERON: Where I guess norma ;y that one defines 6 full scope as through Level-III.
7 MR. MICHELSON: Not necessarily. I do'not know that 8 that is the case.
l 9 DR. KERR: That last sentence is a throw-away.
10 MR. MICHELSON: Okay. It is confusing, because I am 11 not sure. I'was just trying to find out what you thought that i
12 you had in mind. [
13 MR. SHERON: IDCOR submitted the IPEMs back in May of l O. 14 1986. They were modified by IDCOR-in December of 1986 t f
15 following staff comments. And then they made further [
16 modifications and resubmitted them in April of 1987. We have i l
17 completed our evaluation. And based on that, we would expect l 38 that if a utility was to use the IDCOR IPEM, that the level of [
F 19 effort that they would expend is equivalent to a Level-I PRA, [
20 but would be less than if a full scale Level-III PRA was used, i 21 Just briefly going through the key areas of the t
22 concerns that we had, One is that we felt that the IPEM had a l
23 little too much emphasis on bottom line numbers. Namely, the 24 core damage frequency. !
t 25 DR. KERR: By the way, what is meant by core damage j l
i
() Heritage Reporting Corporation l
J y
(202) 62d-4888 :
F i
r
60
'(>' 1 'and core. melt as used in the Generic Letter, is that going to 2 be defined?
3 MR. SHERON: What is meant by it?
4 DR. KERR: Yes.
5 MR. SHERON: I do not think that we are defining it, 6 'are we?
7 DR. KERR: Well, I think that you should. Because it 8 is defined in various ways all the way from incipient core 9 damage to core on the floor. And for example, if you mean the 10 same thing by those two states, I think.that you should say so.
11 MR. SHERON: I guess what we can clarify it. To me, 12 it means --
13 DR. KERR: I am not suggesting what the definition 14 should be, but I think that there should be one. Because it is 15 a fairly important criterion, screening and everything else.
16 I:R . SHERON: Let's take a look and see if we can add 17 something there.
18 Again we have emphasized the identification of plant 19 specific vulnerabilities, and listing of potential areas of 20 improvements. Again, you know, we feel that one should be 21 staying away from trying to pick out a bottom line, but rather 22 using it as a relative measure of where the vulnerabilities 23 might lie.
24 And I might point out that you might have heard at 25 one time Dr. M.rley talk about Susquehanna and the IPEM that O Heritage Reporting Corporation (202) 628-4888
61
.q
%- 1 they used. We took a little bit of a look at that. And :
2 guess that my feeling is that their philosophy and their 3 approach is to stay away from bottom line numbers and basically 4 -just saying_I'am looking for defense in depth and I am looking 5 for eliminating vulnerabilities wherever I can, and not try to 6 sit there and draw all of my conclusions based on getting a 7 number that is one times 10 to the minus 5th versus three times 8 10 to the minus fifth or something like that. Mike.
9 MR. CORRADINI: I guess that I am thinking again 1
l 10 about a year age when we went through this. I sense a 11 difference. You said the IDCOR evaluation and emphasis on 12 bottom line numbers.
13 But historically, was that not the first reasons that 14 you would do the IPEM, and that was to -- I guess that it goes i 15 back to definitions, as to what a vulnerability is in l
! difference to an out-lier.
16 17 I mean that the first presentations that we had on 18 this, the way that my impression was that the IPEM was to look 19 for out-11ers, things that kind of poked out above some cort of 20 envelope that was acceptable. And the way that I read it now, t
! 21 or at least the way that I am listening to this, that 22 vulnerability is not the same thing at all. That is that you l
l 23 may find a vulnerability which does not poke out outside of the 24 envelope, so that you may still require the plant to fix it.
u 25 MR. SHERON: We have not defined what the envelope is l'
O Heritage Reporting Corporation (202) 628-4888
62
'0 1 yet. So we used the words out-lier and vulnerability 2 interchangeably.
'3 MR. CORRADINI: But are they interchangeable? ,My 4 impresulon is that'they are not.
5 MR. SHERON: One may want to define them differently, 6 but to us.
7 MR. ELTAWDA: I think that they are used 8 interchangeably. I think that what you were thinking about is 9 the draft policy statement looking at what sticks out of that 10 ' envelope. The final version of the policy statement, the 11 published one, does not talk about that. It talks about plant 12 vulnerability and the means for fixing it.
13 DR. KERR: I am sorry, but August 8th refers to 14 possible significant risk contributors. It does not talk about 15 vulnerability. I am quoting, I think, directly.
16 MR. ELTAWDA: If you go on page 18.
17 DR. KERR: But at least, as far as you are concerned, 18 out-liers and vulnerabilities mean the same thing?
19 MR. SHERON: It is something I guess that would 20 warrant being fixed, if it was identified. There are things 21 that may pop up that may be a significant, I hate to use the 22 words significant contributor, okay. But if it is a dominant 23 sequence in terms of the total core melt. It may not be cost 1 24 effective to fix it, in which case then in my mind, that is not l
25 a vulnerability.
O Heritage Reporting Corporation (202) 628-4888
e 63
/ 1 DR. KERRt .Okay, go ahead.
2- MR. SHERON: You know, if we go through and find 3 something that, for example, maybe the: total core melt 4 frequency came out to be 10 to the minus 5th, and there was a 5 contributor there that was five times 10 to the minus 6th or 6 half of it, that is a dominant contributor to the total core 7 melt frequency. But if the only way to fix that~ contributor is 8 to add a second containment or something, which costs hundreds 9 of millions of dollars, that is not cost effective in the 10 context of 51-09, and I would not call that a vulnerability 11 then.
, 12 DR. SIESS: If it only cost a dollar to fix it, it 13 would be a vulnerability?
14 MR. SHERON: If it is not fixed, I would call it a
- 15 vulnerability.
i 16 DR. SIESS: That is strange. You better get that 17 down in writing. Because that is the strangest definition of 18 vulnerability that I have heard yet.
19 MR. SHERON: Well, it certainly would not behoove 20 staff to run around saying that there are plants that have j 21 vulnerabilities out there that were not fixed.
22 DR. SIESS: If you want to call them improvements 23 rather than vulnerabilities. You find something that would L 24 really improve safety, but you needed a knot and it only cost a ,
l 25 few thousand dollars, which I do not believe, you know, then i
(
() Heritage Reporting Corporation (202) 628-4888 l -e Mwy
64
~
l .fix it. .But if it would improve safety and you do'not need it, 2 if it is a few million, then you do not do it. If it would 3 improve safety and you do need it.
4 MR. SHERON: Assuming that there is some acceptable 5 limit.
6 DR. SIESS: I am using the word improvement, and you 7 are using the word vulnerability. And I do not think that your 8 word is right in the way that you are using it.
9 MR. LEE: You are giving us an example to satisfy the 10 statement in the Generic Letter that the IP is intended to 11 verify certain things, rather than improve. On page two of the 12 Generic Letter as revised, we emphasize the word verify.
13 MR. ELTANDA: Verify that all plants are consistent 14 with the Commission's safety rules.
15 MR. LEE: Right. That this plant may require a 16 second containment satisfy the statement?
17 MR. SHERON: It would satisfy it if the analysis 18 showed that it was not cost effective.
19 DR. KERR: It seems to me that you are introducing 20 something that the Commission I think eventually said not to L 21 include in the safety goals policy, and that is an ALARA l
22 criterion, which says that you fix things depending on how much 23 they cost. And you do this independently of whether they are i 24 within the safety goal or not.
l
! 25 MR. SHERON: It aise did not include a core melt
( Heritage Reporting Corporation (202) 628-4888 l
I. , - . . _ . __ __ --, . _ _ _ _ _ ,
65 7s U 1 criteria or a condition criteria.
2 DR. KERR: Initially, an ALARA criterion was in the 3 first draft, and the. Commission decidcd not to put it in the 4 final draft. It appears to me that what you say is 1:itroducing 5 an ALARA criterion. I am not necessarily opposed to it. But 6 you are saying whether you fix it or not depends on how much it 7 costs.
8 MR. SHERON: I am just using the criteria 50-109.
9 DR. KERR: 50-109 has to do with backfit.
10 MR. SHERON: That is right.
11 DR. KERR: It does not have to do with ALARA.
12 MR. SHERON: But whether you call it a backfit, or an 13 enhancement, or whatever, again you are pushing us down this 7-
"' 14 road of what is the minimum acceptable. You know, if the core 15 melt is 10 to the minus 4, okay, then anything that they do to 16 improve that is an enhancement. And if it is less, anything 17 that they do is fixing a vulnerability. We have not done that.
18 DR. KERR: So you are not going to require that 19 anybody do anything, is that right?
20 MR. SHERON: If you read the Commission's policy 21 statement, it says that we would expect the industry to fix 22 vulnerabilities that they find. It does not say that the staff 23 has to make a determination and go fix them. We are going to 24 look over their shoulder and see that they are doing a 25 competent job and making competent decisions, but we do not Heritage Reporting Corporation (202) 628-4888 l
66
-(j.
'l intend to get_into the business.
2 DR. KERR: But eventually, if this expect means.
3 anything, you are going to have to decide whether they have in 4 your view done a good job.
5 MR. SHERON: Yes.
6 DR. KERR: Now you are either going to have to do 7 that on an ad hoc individual basis, or else you are going to 8 have to have some criteria.
9 MR. SHERON: I think that right now we-are saying 10- that we would like to go on an ad hoc individual basis, because 11 there are a lot of factors that go into it. The quality of the 12 analysis that they have done, the previous plant performance, 13 and operating history.
p/~
\.
14 DR. KERR: But a whole lot goes into an ad hoc 15 individual basis, too. The geality of the reviewing team, the 16 viewpoint of the reviewing team, hidden agendas that may be 1
17 involved. It is just as risky for you guys to do it that way.
d 18 MR. SHERON: I would just point out that we decided 19 that if we tried to come up with some bottom line number.
20 DR. KERR: I am not suggesting that it has to be a 21 bottom line number, but it has to be something with which you
! 22 use to make decisions.
23 MR. ELTAWDA: If the staff feels that a fix would be 24 necessary, it is going to go and do a cost benefit analysis, 25- and try to justify that fix. And that is the only alternative O Heritage Reporting Corporation (202) 628-4888 l
67
]
v 1 that we have, if the Commission backs the policy.
2 MR. SHERON: If a plant comes in and says my plant is 3 acceptable, I do not have to fix anything, the staff may say
'4 that we disagree with you, but the alternative that we have to 5 fix that plant is the backfit rule.
6 DR. KERR: The backfit rule has to do with existing 7 rules and regulations. It does not have to do with something 0 outside of the rules and regulations.
9 MR. SHERON: Yes, it does.
10 DR. KERR: What?
11 MR. SHERON: Anything. And on. top of that, if I want 12 to fix the plant for a severe accident, I may have to 13 promulgate rulemaking.
14 DR. KERR: I agree. But I do not see how you are 15 going to backfit something if all of the rules and regulations 16 are being satisfied.
17 MR. SHERON: Like I said, I will have to promulgate a 18 new rule if I have to. If we decide that a plant has to make a 19 fix.
20 DR. KERR: Okay. You are saying that rulemaking may 21 be necessary in addition to backfitting.
22 MR. SHERON: If I cannot justify that fix as needed
- 2. to satisfy the existing ru]es and regulations, then I would 24 have to promulgate a new rule.
25 DR. KERR: Suppose you find something that does not
(
Heritage Reporting Corporation (202) 628-4888
68
'rx i' ' ')
1 satisfy the existing rule, does a change have to fall under 2 backfit?
3 MR. SHERON: If we can show that it is not a new 4 staff interpretation of that rule, then that plant would have 5 to fix it regardless of cost.
6 DR. KERR: And that is independent of the numerical 7 vulnerability.
8 MR. SHERON: Correct.
9 DR. SIESS: That is independent of IPE.
10 DR. KERR: Yes.
11 MR. SHERON: If we decide that they can make an 12 enhancement and it can be fixed by say a new interpretation of 13 an existing regulation, then one just goes ahead and does it r]
6
%/
14 via 51-09.
15 MR. CORRADINI: Could you go over that again.
16 MR. SHERON: All right. This is in a viewgraph 17 somewhere, I think.
18 MR. C_ORRADINI : Then we will wait until you get to 19 it.
20 DR. KERR: I would suggest five minutes of 21 uninterrupted presentation, by me.
22 MR. SHERON: Okay. I will go back to a few areas of 23 NRC concerns. Let me see, I went through this. Test 24 applications. Most of the test applications that they did had 25 the benefit of a full scale PRA, which might have influenced (ah Heritage Reporting Corporation (202) 628-4888 l
o .
1
'69 1 the IPE results. But I guess that there is a corollary, too, 2 and that says that you can show then that the IPE picked up the 3 sequences that the PRA did, so it is a mixed blessing.
1 4 The IDCOR BWR IPEM does not assure a complete 5 evaluation of the net impact of system design and operational 6 modifications on the overall plant risk estimate.
7 In other words, if they modify the design as a result 8 of their IPEM and add something, this would not assure-that 9 they will go back and look at the total impact of that addition 10 on the plant.
11 Extreme' care is needed in modeling the support 12 states. And the criteria for determining when more detailed 13 analysis of support states is needed is left to the analyst.
14 DR. KERR: What is a support state?
15 MR. SHERON: A support system might be a better word.
16 DR. KERR: Oh, okay.
17 MR. LEE: I do not understand your statement, able to 18 perform Level-I or Level-II PRA, that the judgment would not be 19 left to the analyst, but some other formal structure that can 20 take care of all of these system interdependencies.
21 MR. SHERON: No. We are saying right now that their 22 IPEM leaves that decision to the analyst.
23 MR. LEE: Instead of IPEM methodology, I would use a 24 Level-I or Level-II PRA, and then staff would not have any 25 concern in this regard for some reason?
O Heritage Reporting Corporation (202) 628-4888 I
70 N
(/
1 MR. ELTAWDA: We are not endorsing the method.
2 DR. KERR: -Which method are you not endorsing?
- 3. MR. ELTAWDA: We are not saying a method, a Level-I
,4 PRA. We are going to use the. data and the' analysis of the 5 Level-I PRA. We are not going out and endorsing any method, 6 that is to say that the method as exists is acceptable. We 7 have implied a deficiency in that method. So when the 8 utility --
9 DR. KERR: I do not d. ink that you understood 10 Dr. Lee's question. He was asking'whether you think the 11 judgment of the analyst is not useful in a PRA.
12 MR. ELTAWDA: It is useful in the PRA.
13 DR. KERR: And indeed, has to be used.
14 MR. ELTAWDA: It has to be used.
15 DP. GERR: The implication here is sort of that a PRA 16 analyst does not have to use any judgment, whereas the IDCOR 17 guy does.
16 MR. ELTAWDA: But the reason for that is that the 19- history of the IDCOR IPEM is that for the utility staff to use 20 the method. And not all utilities have PRA experience. So 21 that is why we have highlighted that. It is the difference 22 between a staff member doing and a PRA experienced person doing 23 the analysis.
24 MR. LEE: The intent behind again the IPE is to let 25 the utilities person be involved, even if they were to choose n'
v
, IIeritage Reporting Corporation (202) 628-4888
l 71 1: 'to go to the Level-I-PRA route.
2- MR. ELTAWDA: That is correct.
3' 'MR. LEES- So what'is the distinction that is being 4 made in this statement? I am still puzzled.
5 MR. ELTAWDA: That we.want the utility involvement.
6 But the utility involvement does not mean.that they have to do
- 7 everything-themselves.
8 (Continued on next page.)
9 10 11 L 12 l
13 14
-15 16 17
'18 19 20 21 22 23 24 25 l
Heritage P.eporting Corporation (202) 628-4888
. _ . . . ~
, 72
( ,).
1 DR. KERR: Brian,-at this point the schedule shows a 2 break. Is-this a reasonable place for a ten-minute break? Or 3 are you within one or two. slides?
4 MR. SHERON: Well, I think the back end analysis is 5 going to generate a little bit of discussion, so this might be 6 a good place to break.
7 DR. KERR: Ten-minute break.
8 (Whereupon, a brief-recess was taken.)
9 DR. KERR: You may begin when ready.
10 (Slides being shown) 11 MR. SHERON: I'm going to talk now about the back end 12 analysis, or beyond, once you've proceeded past the core melt 13 and the vessel failure.
14 The IDCOR back end IPEM method we felt was a little 15 too narrowly focused and did not really provide utilities with 16 the information needed to examine containment performance for 17 the full range of severe accidents.
18 The back end ant lysis as they present it relied very 19 heavily on the IDCOR views of severe accident phenomenology and 20 the map code.
21 And needless to say, there were disagreements between 22 the staff and IDCOR regarding methodology and phenomenology, 23 particularly with regard to the current range of assumptions.
24 The IDCOR methodology did not recognize uncertainties 25 in the phenomenology, and we feel, had a rather optimistic i
O Heritage Reporting Corporation L
(702) 628-4888
~ r- , ,e-n, - . - .-a n - - , , -,._.w,. ,.-,, w --.-.n - ~
. - . , , _ - -----4
/~s 73 1 assessment of the correct operator actions that were assumed to 2- be taken to maintain containment integrity during severe 3 accidents.
4 I would point out here by the way that one of the 5 CRGR comments was that with regard to operator action assumed 6 in doing an IPE is that if the utility assumes an operator is 7 -going to take a specific action to mitigate or prevent a 8 certain sequence, we would expect to see procedures in place 9 and some understanding of what kind of training that operator 10 had received that would assure us that that type of operator 11 action was indeed credible.
12 DR. KERR: I noticed, and my impression, correct me 13 if I am wrong, is that up to now the staff has reviewed 14 guidelines for emergency operating procedures but has not 15 generally reviewed the procedures themselves.
16 This represents a change then and you will review the 17 individual procedures that go into the severe accident ,
18 management.
19 MR. SHERON: It's not clear right now to what extent 20 we're gofng to actually go out and physically review the 21 procedures.
I L 22 What we would most likely like to see is a 23 confirmation that the utility, for example, has put procedures 24 in place. My understanding is that NRR is initiating a program ,
i l
25 which will actually go out and examine the in place procedures
! Heritage Reporting Corporation (202) 628-4888
<%- 74
- U l' at the plants.
2 I would have to defer to the NRR staff. They could 3 probably come down and give you a briefing on their program on 4 this.
5 DR. KERR: What you have in mind is that somebody-6 will review these procedures? I'm trying to find out what you 7 mean by "in place."
8 MR. SHERON: If a utility comes in and says we assume 9 the operator will take such and such an action, will go down 10 over in this part of the plant and turn this valve or whatever, 11 we would like them to confirm in their submittal that there are 12 procedures or there is a program in place at the plant so that
. 13 we have assurance that the operator would be --
14 DR. KERR: So as long as they say "we have 15 . procedures" --
16 MR. SHERON: They may choose later on down the road 17 as part of accident management, for example, to go out and take 18 a look at those procedures.
l 19 And as I said there is a program --
20 DR. KERR But "in place" doesn't mean that. "In
- 21 place" means they say to you, "we have procedures"? Is that 22 the meaning of "in place"?
23 MR. SHERON: Yes. They need to confirm that they
! 24 have in place.
l 25 MR. MICHELSON: Along the same lines, certain of
! (
Heritage Reporting Corporation (202) 628-4888 l
75 f 3-NJ 1 these procedures call for the use of non-safety grade equipment 2 for the severe accident mitigation, and that equipment isn't 3 necessarily under any kind of surveillance of any sort or 4 verification of operability of any sort? .'I'm thinking S particularly of manual valves that will sit in one position for 6 years and it might be the valve that you have to now close for 7 the particular event.
8 How do you do a PRA for such a situation? What do 9 you use for the reliability, the number to be used for the 10 operability of that manual valve that hasn't been turned for 11 ten years? You.certainly can't use standard manual valve 12 reliability numbers because they may not pertain at all.
~
13 Is there some magic way you're going to ascertain 14 that non-safety related equipment is sufficiently reliable for 15 this purpose? If so, how are you going to do it?
16 MR. SHERON: I think the utility is going to have to 17 make the case why they can take credit for it. They'll have to 18 answer the question. We'll review it.
19 MR. MICHELSON: Yes. And they will have to put in 20 some kind of a number for the probability that you can 21 successfully close the valve and it will include not alone a 22 certain operator response number but also a number for the 23 actual operability of the valve under this condition.
I
! 24 MR. SHERON: If it's appropriate, yes.
25 MR. MICHELSON: And you will review those numbers.
l O Heritage Reporting Corporation (202) 628-4888 l
t.
76 73
(,) 1 And I sincerely hope you don't think it's the same number that 2 might be appropriate for a valve that's operated once a week or 3 month or something.
4 MR. SHERON: Please keep in~ mind, too, that we are 5 not proposing to review these. And I'll get into this much 6 later, the review process. .
7 Dut we are not going to be reviewing these IPEs down 8 to the nth detail.
9 MR. MICHELSON: But you would expect the utility to 10 have performed the calculation though.
11 MR. SHERON: Yes.
12 MR. MICHELSON: And if you went to look, you would 13 find it?
-14 MR. SHERON: Correct.
15 I would point out, loose interpretation of the IPEM 16 questions by the utilities could result in a lack of sufficient 17 attention to measures to mitigate the consequences of an 18 accident and what the staff has done is we have to address what 1 19 I guess we think are the concerns or deficiencies in the IPEM 20 methodology.
21 We have prepared Appendix I to the IPE letter which 22 we hope provides additional guidance to the utilities on how to 23 assess containment performance. Mike?
24 MR. CORRADINI: Is this the right time to ask about 25 Appendix 1, or should I wait?
O Heritage Reporting Lorporation (202) 628-4888
77 1 MR. SHERON: There was going to be a presentation on 2 Appendix 1. .I have a little more information on it I was going 3 to give. This might be an appropriate time.
4 MR. CORRADINI: A general thing. A year ago when'we 5 saw this, the equivalent of what Appendix 1 was at that time
- 6 was a little bit disjointing. In reading Appendix 1 this time, 7- it's not disjointed, but I don't see it as being that clear.
8 If were -- again, I'll put myself in the role -- if I got the 9 letter and I went to Appendix 1 to see what do I have to do 10- more than IDCOR, IPEM for the back end, I don't know if I could 11 glean that from Appendix 1?
12 That is, I really don't see there a clear discussion 13 or a clear specification of what one should do.
( 14 I mean, if I were to do it in the simplest terms, it 15 says you're going to have to do some sort of simplified 16 containment of entry, but as to what is acceptable or non-17 acceptable, is it purposely vague?
18 MR. SHERON: Again, we're getting into this what's 19 acceptable and what's not acceptable like we're dealing with an 20 Appendix K rule or something, and we're not going that.
21 MR. CORRADINI: No, I don't mean it like that. I'm 22 viewing it from the utilities standpoint of if they've got to 23 do it, they have to have a warm feeling somewhere that they 24 know what to do. And reading Appendix 1, if 7 were them, I I
25 wouldn't know exactly what to do. And let me just show you why l Heritage Reporting Corporation l (202) 628-4888 1 ,
3
. I 78
.q-t,,/ 1 I mean that.
2 If I go through Appendix 1, it says, go to Chapter 7
- 3 of the Procedures Guide, which we've got here, and I was in on 4 this six years ago when this was put together and this is very 5 general and a lot of what's in Chapter 7 of the Procedures 6 Guide is essentially out of date to the back end. So I don't
.7 know if that's an appropriate reference.
8 MR. ELTAWDA: It is still an appropriate reference.
9 What Appendix.1 does is to introduce into the process that in 10 those issues that your SER 2300 is supplementary information.
11 So in all the procedures of the SER 2300, if you come across an 12 issue that says do an analysis or something like that, the 13 tendency right now is to go to Appendix 1 and find this
) 14 information.
15 And we did not want to reproduce Chapter 7 in the 16 Appendix 1. I think if you follow Chapter 7 and whenever you 17 need some information about the phenomenology, go to Appendix 18 1.
19 MR. CORRADINI: Okay. Then, if I were to put it in 20 25 words or less, Appendix 1 says do some sort of simplified 21 containment of entry and think about these things when you're 22 doing it and no more. That's literally what Appendix 1 is 23 saying.
24 DR. KERR:
25 MR. ELTAWDA: No.
O)
\~ Heritage Reporting Corporation (202) 628-4888
e 79 1 MR. CORRADINI: Well, let me say it again, because I 2 want to make sure that I understand the thrust of what you are 3 trying to do and given that, if what you've written 4 accomplishes it, is the thrust literally to say, to do a 5 containment of entry back end of some type following the PRA 6 Chapter 7, considering these new things or these current things 7 that you discussed in Appendix 1?
8 MR. ELTAWDA: Right.
9 MR. CORRADINI: Okay. And then my concern about what 10 is acceptable or was not acceptable you are saying again is 11 going to be addressed on an ad hoc basis. When you see what 12 they come up with, you'll decide whether or not it is of 13 sufficient quality or preciseness that you can accept it or 14 understand it or use it?
15 MR. SHERON: Yes. We've struggled with this back 16 end, because there is considerable uncertainty, as you've seen 17 from the 1150 analyses, for example. Once you get into the 18 back end, there is considerably uncertainty in many phenomena.
19 But for the most part, we see it right now that there 20 are two right now that are still hanging out. For the BWRs, 21 it's the liner melt-through problem and for the PWRs it's the 22 direct containment heating.
23 I was going to point out that I think what we're 24 trying to emphasize on the back end analysis is not again to 25 focus on being able to calculate these phenomena down to the O Heritage Reporting Corporation (202) 628-4888
. , , , - , -n- - _ . _
/
y g 80
~
1 nth degree and coming up with some bottom line number for 2 containment failure but rather what we want is the utility to 3 go through and to understand what the phenomena are that the 4 containment is going to experience or is expected to experience 5 should one melt down and penetrate the vessel, to understand 6 what the uncertainties are in those phenomena today as we best 7 understand them. And I think this is one thing we thought that 8 the IDCOR methodology really didn't address too well, and that 9 is that there are wide diversities of opinion, as I know you 10 are aware, regarding the probability of certain phenomena and 11 even whether it exists or not.
12 And we want them to understand that these phenomena,
-s 13 that there are uncertainties, that there may be things that 14 they can do or should be aware of should a severe accident 15 occur in their plant.
16 One cannot tell them today gee, you can ignore this 17 because we haven't resolved it. I don't think that's a prudent 18 thing to do. I remember someone once said, when Einstein's 19 theory of relativity surpassed Newton's theory of gravity, 20 apples didn't stop in mid-air while scientists debated it.
21 If a core melt happens, whatever t.he appropriate 22 phenomena are are going to occur and the utility personnel are 23 going to be faced with it at that time whether it's tomorrow or 24 next year.
25 And whether one group says no, that can't happen or O Heritage Reporting Corporation (202) 628-4888 I
'x 81
- n. 1 another group says it can happen or gee, it's big in 2 uncertainty, we think that the utilitp personnel need to 3 understand what that uncertainty is right now and what it means 4 with regard to what they may be faced with should they have to 5 cope with such an accident.
6 They need to know that if they get a high pressure 7 scenario that if they don't depressurize, there is this thing 8 called direct containment heating that may or may not occur and 9 they have to understand what the consequences are, both if they 10 depressurize the plant or if they don't depressurize the plant. 1 11 If, for example, something happened tomorrow, they 12 would have to make that decision based on the available 13 knowledge. One can't wait around for research to solve all the 14 issues.
15 So we want them in their back end analysis to 16 consider these uncertainties, to look for practical solutions, 17 be it in an area of accident management or the like, but to 18 basically understand their containment performance and 19 understand where things can fail and why.
20 DR. KERR: Brian, I would say that it's a good idea, i 21 if the staff hasn't been able to solve a problem, to ask a 22 utility to do it. But there is at least implicit in this 23 document some guidance which it strikes me could be misleading.
24 For example, in Page 10 of this, I read: The core 25 molt sequence in high primary system compression is often due t
( Heritage Reporting Corporation (202) 628-4888 i
. c.
4 82 h 1 to a station blackout sequence. And then the sentence: The 2 high pressure scenario also represents one of the most 3- significant contributors to risk.
4 That is a pretty significant statement, I must say.
5 It doesn't say anything about uncertainty or whether people 6 disagree ~. It says it is a significant contributor, one of the 7 most significant contributors to risk. Which means, it is 8 either relatively high probability or high consequence or both.
3 Then on the next page, the quote is made from the 10~ Brookhaven report which finally says: An early study of 11 relative merits of the possibilities available would be 12 valuable.
13 Then the statement: The staff is in favor of O 14 adopting the panel recommendation. Is the panel recommendation 15 that an early study of the possibility would be valuable? I 16 guess. So the utility should study the possibility.
17 'Then, on Page 14, and I think this is the same thing,
' 18 I read: Unless the licensee can demonstrate the primary system 19 can be reliably depressurized, a low probability of early 20 containment failure should not be automatically assumed.
21 How can a licensee demonstrate that the primary 22 system be reliably depressurized without a rather extensive 23 research program? Because nobody has demonstrated up to now 24 that it can be reliably depressurized with existing equipment, 25 that i know of, for PWR, O Heritage Reporting Corporation (202) 628-4888
t k
83 1 MR. .SHERON: Wnat we're saying is that -- and'they 2 - can't, okay? They're saying fine. I can't demonstrate that I 3 can reliably depressurize because I haven't done the analysis
'4 and the like. And all we're saying is then don't discount the 5 direct containment heating scenario at this time.x 6 DR. KERR: But what does "don't discount" mean? Do
+7 you mean to recognize that it may occur or does it mean you've
'8 got to do something with your system?
9 MR. SHERON: Recognize that it may occur.
10 MR. CORRADINI: Can I try another tack on this?
11 Because it's really maybe a philosophical sort of thing. I see ,
12 what you're trying to do in Appendix 1. The way it's written, 13 again I'm putting myself in the role of reading as if I knew
) 14 nothing about this, I would really get uncomfortable. I'd get 15 nervous if I were the utility person. Because it would leave 16 me with the sense that I'm entering in an area where you have 17 all this research work that has gone on and is going on, and 18 there's a lat of uncertainty or at least debate that there's 19 uncertainty, and here I am now trying to do my rudimentary 20 calculations and I'm trying to get an idea, and here I am doing i 21 this -- is it to become appreciative of the uncertainties? Is
- 22 it to contribute to the solution? Ia it to understand how my 23 plant is particularly vulnerable or not vulnerable to these 24 events? I don't know what the purpose is.
25 When IDCOR presented their stuff last year, it was
' Heritage Reporting Corporation l
(202) 628-4888
J 84
~
1 clear what the intent was and the intent.vas toft ake their 2 methodology and'look for. geometrical, I.mean'at the bottom,-at 3 the baseline,-geometrical differences in particular plants and 4 their containments, and see, you'know, does this plant have an 5 unusual geometry that would either contribute to this physical i
6 process or mitigate it or something like that. That seems like
+
? a very reasonable thing to do, at the very least.
8 But a lot of what I read in Appendix 1.goes beyond 9 that and I'm just wondering, do you really need to emphasize 10 that because I don't know if the utility people really want to 11 do this at this point or can. Am I making myself clear about 12 that?
13 MR. SHERON: I understand what you're saying. I 14 think the work, again, I would reiterate that you know, if 15 it's -- we don't think it's appropriate for utility personnel
- 16 or whatever that's involved in this that's going to be t
r 17 operating the plant to go down some sort of a rosy path that l 18- says you know, well direct containment heating is a very low l
19 probability thing and therefore I need not worry about it.
20 If it occurs, they should understand it, they should 21 understand what it is and how it can affect the way the 22 containment performs and they may want to on their own do 23 something like initiate a study.
i 24 DR. KERR: Do you think anybody understands it at 25 this point?
O Heritage Repor ing Corporation l (202) 628-4888 1
.--,,,,w,- - , , , _ ,
,, 85 1 MR. SHERON: I don't know. We understand parts-of-2 it.
3 DR. KERR You're saying a utility should understand 4 it and I'm trying to grope for the level of understanding.
5 Because I would not be sure that there is anybody wno would say 6 that he thoroughly understands the sequence and its phenomena 7 and the results.
8 MR. SHERON: I didn't think it was our job to tell 9 the utilities to what depth they have to understand their 10 plants in order to be acceptable.
11 MR. ELTAWDA: We don't want them to discount the 12 sequence if it happens in their plant.
rs 13 DR. KERR Let's suppose that I find out or conclude
\
14 that the probability of catastrophic vessel failure is about 15 the same as the probability of direct containment heating.
16 I'll just hypothesize.
17 Are you asking that utilities be prepared to deal 18 with catastrophic vessel failure, to have procedures in place 19 to do that?
20 MR. ELTANDA: Yes.
21 DR. KERR: Because I don't ace anything in here about 22 catastrophic vessel failure.
23 MR. SHERON: It's not going to do us any good not to 24 tell them about the existence of these --
25 DR. KERR But why not put in catastrophic vessel ,
O Heritage Reporting Corporation (202) 628-4888
1 failure? It's a pretty serious consequence. And the 2 probability is probably in'the same region as some of the 4
3 things in this screening criteria.
4 MR. SHERON: We don't know that yet.
5 DR. KERR:' We've had a lot of studies of the 6 probability of catastrophic vessel failure which give numbers, 7 and by the way, there is some basis for some of these numbers.
8 MR. SHERON: Yes, but in my understanding they are .
9 all down around 10 to the minus sixth or seventh. ,
10 DR. KERR: Well, do you have reason to believe that 11 the direct containment heating is not in the same ballpark?
12 MR. SHERON: I don't believe the uncertainty or the i 13 probability of vessel failure is so unknown as direct 14 containment heating.
15 DR. KERR: I don't, either, but catastrophic vessel 16 2ailure may well be in that region. Why not put in 17 catastrophic vessel failure then as something that they ought 18 to consider if it could well be in the same ballpark?
19 I'm groping for a reason for putting in certain
- 20 phenomena.
21 MR. ELTAWDA: Because previous study has not f
22 identified that as the accident sequence that you have put in.
23 We are not going beyond what are the studies for research have j 24 said.
25 DR. KERR: One of the screening criteria I think is O Heritage Reporting Corporation (202) 628-4888
[ -- ._
87 1 that sequences that have a probability less than two times ten 2 to the minus sixth must be considered.
3 MP. ELTAWDA: Greater.
4 DR. KERR: I'm sorry. Greater. That's-awfully close 5 to one times ten to the minus sixth. And that may be about the 6 probability of catastrophic vessel failure. Hence, it seems to 7 me it would be easy to make a case on the part of some 8 reviewer that a licensee should have considered in detail the 9 results of catastrophic vessel failure and should have 10 procedures in place.
11 MR. SHERON: That shows up as something that falls 12 within the screening criteria but right now, it is not an area q- 13 for example, of controversy between the staff and the industry.
%)
14 I think the. metallurgists would probably pretty much be agreed 15 on what the probability of that is.
16 But we have a situation here where we felt the 17 industry was taking a very optimistic position on direct l
10 containment heating and that is that it was either very low l
l 19 probability or it was not a problem even if it did occur.
20 And we're saying we're not convinced of that yet and
! 21 we're not saying it's just wee, the staff, but there is 22 division among the technical community at this time, and it is l 23 not appropriate to discount it.
24 MR. CORRADINI: I see what you're saying. But I 25 guess I'm not so much bothered by a lot of u at is said in l
Heritage Reporting Corporation (202) 628-4888
4 .
88 f- l' Appendix 1 as the organization in the sense that I' read it and f 2- I get the sense that you focused on direct containwont heating 3 and BW MARC-I liner meltthrough. But those are just two 4 possibilities of a whole range of thinga that may happeri. And 5 I'm afraid that if you -- well, there's a number of things that 6 I'm worried about. But if you leave it as written, if I were 7 the utility person, I would tend to focus on those simply
- 8 because that's where I was directed to focus. And that may not ,
9 be the proper focus.
10 MR. ELTAWDA: No, that's not true, because these are 11 the only two issues that have been identified since.the new SER 12 2300. There is a table in (l' apter 7 that identifies all the 13 phenomenological, all the threats to containment integrity.
, () '4 Only two things ar' left out of it.
15 MR. CORRADINI .11 right. I see your point. I see 16 your point there. One side thing. The table you mention is l
17 the wrrng table. You Jay rable 72 and I'm trying to say --
18 MR. ELTAWDA: I ;annot remember.
I 19 MR. CORRADINI: But I mean in the document. You say t 20 Table 72 and I turn to Table 72 and you've got bin 21 characteristics. So that can't be the right table. >
i 22 MR. ELTANDA: That might be. .
23 MR. Cr ?.RADINI : So you're saying that the only reason that . . cua and discuss those at Appen d ix 1 is because they 2 * :ns to what was already discussed in Chapter 7 of the Heritage Reporting Corporation (202) 628-4888 i i
[.
G *
".a v- r - - - - - - - - ---*-a w v 'w e y ?
^m
m, -. - . . _ _ -
d: + ,
1 89
. As
' 1 PRA Guide.
2 MR. ELTAWDA: That's correct.
3 MR. CORRADINI Okay.
4 MR. MICHELSON: Brian, there is a class of events 5 which you traditionally call I think "Event V" which is an 6 interface between high and low pressure piping which could 7 undergo potential rupture under certain circumstantes.
. ,8 There are other types of external, other types.of 9 pipe breaks'outside of containment that can also lead to 10 potentially catastrophic events that have to be looked at such 11 as in the case of the boiling water reactor, the reactor water 12 cleanup system, if it ruptures, and you fail to isolate it, you 13 can be in very deep trouble. If the steam line, and it's
() 14 always pressurized and looking directly at the reactor, if it 15 fails, it's a bypass of containment. And if it is unisolated, 16 it's a bypass of containment. And then of course it 17 jeopardizes equipment required to keep the core cool.
J 18 How are those brought into this? I don't find them 19 flagged out like I do Event V and some of the others and I 20 don't find them as a part of the discussten of hypass of
't 21 containment and so forth. How are those flagged out?
22 MR. SHERON: The utility -- they're not flagged out 23 in the sense that we're giving the utility this detailed road 24 map on how to follow it through. As you know, there's a 25 section on here that talks about under containment of entries,
( Heritage Reporting Corporation l (202) 628-4888
^90
-0 1 you know, you should identify those sequences that result in 2 bypass of containmert.
3 MR. MICHELSON: Yes. Well, it's well recognized I 4 quese on the part of everybody that does this sort of thing S that thowe are something to look at in a severe accident. Is 6 that correct?
7 MR. SHERON: We're presuming that the people that do 8 this work are not going to be, you know, novices with regard to 9 doing a PRA or an IPEM. ,
10 .MR. MICHELSON: Well, I haven't seen these flagged 11 previously or beyond the design basis. Now you're dealing with ;
12 the newer probabilitics, but you're not worrying about the 13 operability of valves -- will they indeed close. I have the 14 same problem on containment bypass from the viewpoint of the 15 containment isolation valves which are normally open.
16 There's no history to go on. You have to use tests, 17 since the people haven't he.d been LOCAs yet, to test their 18 containment isolation valves with. So history means nothing, 19 routine operation under nominal load conditions essentially 20 means very little if anything. And the test data doesn't exist .
21 for these big valves under dyncmic loading of this sort. Some i 22 tests have been done under somewhat lesser conditions which 23 indicate that there could be indeed some interesting essential 24 problems with them.
25 I hope that when this is done that this is taken into O Heritage Reporting Corporation (202) 628-4888 ,
b I
- a. ,_ ,..._-_,_y . . _ , , , _.,. _ _
91
- o._ -1 account when they come up with the magic numbers.
2 MR. SHERON: This is not unique to-the IPE core.'What 3 you're' bringing up is a question that is generic to all PRAs.
4 MR. MICHELSON: Clearly the integrity of containment, 5 though is one of your main thrusts.
6 MR. CORRADINI I don't meaa to beat on this,'but 7 there's one thing. I'm looking through the organization of 8 Appendix'I and I guess that's what I'm most concerned about, is 9 as you give them background, you give them a history and you 10 talk about insights. And then you hule a section, "General 11 Guidance on Containment Performance." And before you come up 12 with, it appears to be the minimum acceptable -- I don't want 13 to use those words, but I guess the minimum acceptable things 14 that have to be contained in the back end analysis, which you 15 designate as 1 to 6 -- you bove a discussion of high and low 16 pressure sequence.
17 And I guess that's partly again for educational 18 purposes. But there are some things stated in there that 19 there's no references made as to how they come up with these.
20 For cxample, on the bottom of Page 15 there's a 21 paragraph that starts talking about debris beci coolaoility. It 22 says the staff views this as an area of uncertainty and 23 recommends assessmenta be based on available cavity area and as l 24 assumed maximum coolab)o depth of 25 centimeters. For depths 25 in excess, both cool and non-cool outcomes should be O Heritage Reporting Corporation (202) 628-48--
~
H92 i 1- considered.
2 Now, that might be right but I mean, I have two ways i
3 to look at this.
f 4 Either if you're going to give the utility in 5 Appendix 1 an education, if you want to say it that way, I 6 would reference it so that they say gee,'where the heck did 7 this come from? Because I don't know where this came from.
8 And I don't.know if it's even justifiable. But under the 9 assumption it is, it ought to be referenced.
10 .MR. SHERON: Yes, we can certainly get a reference.f 11 MR. CORRADINI: Okay. But there's about four or five 12 pages, before you come to your list of 1 through 6, what 13 appears to be the guts of Appendix 1. But if you're going to ,
() 14 give them an education, and that's your purpose in the 15 preceding pages, I would at least reference them back to- [
16 somewhere if they want to read in depth to undqrstand this, 17 they have somewhere to read. Because some of these statements 18 are stated, are made, and I don't really know where they come 19 from. >
i 20 MR. SHERON: Well, again, like I said, we are 21 assuming that the people that perform this are not going to be 22 novices.
l 1
23 MR. CORRADINI: I realize that. But for debris bed 24 coolability, for example, I will look to Ivan. If he knows 25 where 25 comes from.
l O Heritage Reporting Corporation -
(202) 628-4888 j
{
93
(. 1 MR. CATTON: We did calculations for Zion a' number of i 2 years ago and I think it was deeper than-thht. But when I read
-3 through this sometimes it sounds like they're talking ~about 25 4 4 centimeters of molten material and snother time-it's 25 l
5 centimeter debris bed. Which'is it? And they're also talking t
6 about sprays and that the sprays will cool it. It's confusing 7 and inconsistent, t
8 MR. ELTAWDA: We are not trying to describe the case l -9 that you have debris, it will cool-it but it will not hurt. In i 10 the appendix --
11 MR. CATTON: I don't recollect page and paragraph, 12 but it does say that the sprays will cool it.
13 MR. ELTAWDA: Might cool it. The whole idea of the O -14 appendix, as Brian said, is that when they are faced with a i i
15 problem they just won't sit there. We have to think about L
l 16 maybe adding additional water source so they pour it on the 17 debris bed. Whether it's going to cool it or not, at least ,
18 they've done something. It's accident management rather than 19 trying to cloud the issue.
I 20 MR. CATTON: I'm concerned about these poor guys in 21 the utility, that the DCH, I think right within this room we 22 can get two or three people who will argue all three corners of 23 the issue. And it seems to me then all the utility would have 24 to do would be to pick one and reproduce what this person has i
25 argued and ther we could get into negotiations for the next O Heritage Reporting Corporation (202) A28-4888
- - - -. . -- -. .. - - - . - -- , . -- ~
,- 94 l_) i five years.
2 This business of the 25 centimeters and cooling or 3 r.at cooling, the meeting that was in Baltimore, the experts who 4 are doing rescarch for research disagreed on coolability of 5 this stuff. There was an experiment that we were shown that 6 shows there is a possibility tha* it lo coolaole if you put 7 water in it.
8 The utility can pick either one.- I think you could 9 take some of this here and within about a half a day I could 10 give you all of the consideration you need and have ammunition 11 to argue with you for the next five years, and get nowhere.
12 MR. SHERON: All we're trying to say is you know, I fm s 13 sort of get this feeling that gee, we don't know the back end
( /
14 very well so the utility should be totally exonerated from 15 having to consider it.
16 MR. CATTON: I didn't say that at all.
17 MR. SHERON: Well, I apologize, but that's the 18 impression I get. And what we're trying to say is that we 19 understand that there are a lot of uncertainties that still 20 remain with regard to containment analysis. And all we are 21 trying to do is point out to the industry that they need to 12 consider and undecstand what these uncertainties are and they 23 have to decide for themselves whether or not there is something 24 they should do or wai, to do or is beneficial for them to do.
25 But to ignore it just because somebody says well, gee, the O Heritage Reporting Corporation (202) 628-4888
o 95
-(:) 1 staff hasn't finished all their research on this so we don't 2 have to think about it is unacceptable. Okay? We pointed out 3 that we don't know, okay, whether or not one gets minor 4 meltthrough due to debris beds in the BWR, but we do know that t
5 if.you pour water on top of it or if there's water in that lower cavity, we seem to think our initial assessment is that's G
f 7 beneficial. But why does the staff have to go through and do e 8 the analysis out to the nth degree before the utility even 9 decides to do something? It's their plant, they should decide I 10 it. ;
i 11 DR. KERR: Brian, the point I think is that the staff 12 at this point after having labored for a number of years over 13 an extremely difficult problem, namely, should there be
'O 14 containment performance criteria, has not yet been able to come t
j 15 up with any.
16 And one of the reasons it hasn't, I think, is because 17 it is difficult to know what to do about existing containments. {
18 Now, in order to decide what to do about existing 19 containments, one has to have a better understanding apparently 20 than we have now about some of the phenomena. And you are ;
L 21 suggesting that an individual utility who probably is confused 22 as some of us are by the results of some of the research is 23 going to make a rational decision about what to do about some 24 of these extremely low probability, severe accident sequences.
25 Maybe this is the case. -
lieritage Reporting Corporation (202) 628-4880 r
_ ._ - - , - , . - , , _ . . . , - . . - , , _-_-m. . . , - _ _ - - - - - - - - - - - - - - - - - - - -
i 96
() 1 But I don'.t see on what basis they are going to 2 decide.
3 Incidentally, concerning this degree bed coolability,.
4 you might want to have somebody look at the-SER on the S IPEM/BWR analysis because it_contains a statement that the 6 staff is not yet convinced that the presence of water is 7 sufficient to ensure coolability of any depth of core end 3 debris. And I don't know whether that's in contradiction to 9 the other statement or not. It probably is not. But it might 10 be confused by someone who wasn't careful in reading.
11 MR. ELTAWDA: If you look at Page 19 in Appendix 1, 12 you find that we are taking out the two controversial issues of 4
13 containment heating and liner meltthrough and we say that there
( 14 is no consensus among the researchers about how to handle this, i '
15 and we are not expecting the utility to come with their own 16 analysis. What we expect them to do is be aware of the 17 phenomena, the possible outcomes, and if they can look at means 18 to mitigate its conscquences er preventing it. So like if they 19 want to study the pressurization system and see if they can 20 eliminate the high pressure sequences, we are not dictating, I but we want them to be aware of the phenomena.
21
- 22 DR. KERR
- Okay. Let me aak about a situation which 23 arose several months ago when Pilgrim came in with a venting 24 system which they thought would ameliorate a severe accident 25 situation. And the reaction from the staff, you've got to go a Heritage Reporting Corporation
- (202) 628-4888
97 1 way and answer a lot more questions.
2 Is that the sort of thing? -Suppose-someone does look 1
3 at the containment heating, comes up wita a solution, then as 4 another iteration, the staff says, here are questions we have 5 about this,.go answer those questions.
6 MR. SHERON: Well, I think number one, Pilgrim was in 7 a unique situation.
8 DR. KERR: I realize-that.
9 MR -. SHERON: I understand that. And I also think 10- that --
11 DR. SIESS: What was unique about Pilgrim?
1 12 MR. SHERON: I beg your pardon?
13 DR. SIESS: What was unique about Pilgrim? That they 14 wanted to do something, or what they wanted to do?
15 MR. SHEh0N: They were shut down, okay, number one, i
16 and they were proposing-these kind of modifications and the-17 like as I understand this part of the startup program.
18 But the question, you know, any modification that the 19 utilities propose. In this case the utility came in and said 20 we propose venting. Nid I wasn't very close to this. But for 21 er. ample, Jf they are proposing to violate a GDC, which is the 22 one that probably requires containment isolation, they need to 23 come in and submit the justification for that under 50.59.
24 That's a regulation, they have to do it. And tar staff has to 25 make the decision cs to whether or not they agree, which means O Heritage Reporting Corporation (202) 628-4888
, 98 1 you have to weigh the pros and the cons. And that would be 2 true no matter what was pronosed regarding a back end analysis.
3 If 50.59 is triggered, that utility has to come in and present 4 an analysis.
5 DR. KERR: Yes, but you see, when you're talking 6 about ac cidents that are beyond the design basis, and hence are 7 beyond axisting regulations, how are you going to deal with 8 accidents that are beyond the existing regulations by uppealing 9 to existing regulations?
10 MR. SHERON: If what the utility is proposing is a 11 violation of existing regulations, regardless of whether 12 they're in the severe accident space or not.
13 DR. SIESS: But a lot of the fixes are going to have 14 to be that.
15 MR. SHERON: They may be. And if the utility 16 proposes them themselves, great. They can do that.
17 But if the staff ultimately decides to do it on their 18 own, through the backfit process, then we will have to 19 promulgate a regulation to support whatever it is we want.
20 DR. SIESS: You mean that Pilgrim would be in 21 violation of the regulations if they had a core on the floor 1
22 and wanted to verit, because the regulation says that you don't 23 open the container, even though that regulation never 24 considered the core on the floor? Is that your argument?
25 MR. SHERON: I don't know, because I was not involved O Heritage Reporting Corporation (202) 628-4888
99 73 kl 1 in that decision. You would have to ask Dr. Murley or his 2 staff.
3 DR. SIESS: But there are going to be hundreds of 4 such things to be involved in in connection with the IPE. And 5 you're going to wait the length of a month to decide how to 6 handle them?
7 MR. SHERON: Because each one is liablo to be unique, 8 DR. SIESS: The first time it comes up it's unique.
9 MR. ELTAWDA: They have to look at both the positive 10 and unique impact of that strategy, not because venting can 11 help in certain situations can you propose venting without 12 looking at the other negative aspects of it and the question 13 that was asked of Pilgrim is related to that. Have you looked O
N' 14 at the negative aspects of venting. And I think it's a logical 15 question to ask when you propose something.
16 DR. KERR: I was not being critical. I was using 17 that as an example of something that occurred and I was asking 18 is that same procedure going to be followed if a utility 19 suggests something as a result of the IPEM? That was my 20 question.
21 MR. SHERON: If they can do it within the framework 22 which doesn't require staff approval then they can do that.
23 DR. SIESS: Any fix they can make that doesn't j 24 violate a regulation, fine. But if you can find somewhere in 25 the regulations that that is violated, then you have to review Heritage Reporting Corporation j (202) 628-4888 l
I
rs 100
( a 1 it. Is that right?
2 MR. SHERON: They make the determination. They have 3 a plant safety committee that is supposed to review the changes 4 and determine if the criteria of 50.59 are violated or tripped.
5 If they're not, then they can make the change themselves and 6 all they are supposed to do is document what they've done and 7 put it in the files so it can be audited by an inspector, let's 8 say.
9 DR. SIESS: Now, 50.59 talks about an unaddressed --
10 MR. SHERON: If they create an accident of a 11 different kind, if they increase the frequency of an accident, 12 or if they make the consequences of an existing accident more 13 severe.
14 DR. SIESS: Yet all of those accidents that were 15 envisaged in 50.50 are design basis events. Right?
16 MR. SHERON: Yes.
17 DR. SIESS: And we are looking at things beyond the 1
18 design basis?
l 19 MR. SHERON: That's right. But if they propose a 20 criteria that could affect a design basis event and one of the
! 21 questions Dr. Murley had, I know, was the inadvertent venting.
l
{ 22 And if there was a possibility, for example, of venting l
t 23 containment, as a result of this new system or whatever they i 24 were putting in, such that one could vent during a design-basis i
l 25 event, the staff has to look at it. So one has ta look at the l
I /~3 l LJ Corporatian l Heritage Reporting (202) 628-4888
101 (s
)
1 impact of what they do on the design basis as well as beyond 2 design basis.
3 MR. LEE: I was wondering if there is some way of 4 getting a little more guidance from IDCOR studies and reports, 5 especially in IPEM, again coming back to the question 6 specifically of early containment failure or low probability of 7 early containment failure that IDCOR people assume.
8 Is there any possibility of resolving that issue in 9 some way that an individual utility personnel could handle that 10 problem in a manner that would be satisfactory to the staff?
11 That issue or tilat astumption makes the containment 12 event tree methodology approach very simple in IPEM, the wsy I fs, 13 understood it, the way I remember. And the staff is proposing
() 14 that you cannot use that basis.
15 MR. SHERON: Well, number one, we interacted 16 extensively with IDCOR and I think on several issues we agreed 17 to disagree.
18 Understand that when a utility submits their
'9 analysis, they can come in and they can say yes, we have 20 considered all this stuff and we choose not to do anything 21 about it right now. That's their decision. Now, again, as I 22 said before, if the staff decides that their decision, that we 23 don't agree with it for some reason, supposing they way we've 24 looked, we have a BWR here and we've decided that yes, we 25 understand that if you put water in the lower cavity it can O Heritage Reporting Corporation (202) 628-4888
102 0 .
1 help scrub fission products even if you fail the liner, and the 2- like,-but we choose not to do that. Fine. They've considered 3 it. That's their consideration.
4 If the staff doesn't like that consideration, our 5 recourse is to say yes, you should put water in the cavity and 6 we will pursue that through a backfit rule, and if we can 7 demonstrate that there is a substantial increase in safety and 8 it is cost effective, then fine, good for us. We do that. And 9 perhaps we have to promulgate a regulation in the process. But 10 what we are trying to do !s make sure that the utilities are 11 aware and purposely go through and think this through so that 12 there is -- you know, nobody can ever come back and say you 13 never told us about this. Gee, I just had a severe accident O 14 and I had a high pressure meltthrough and golly, it failed 15 containment and you guys knew about depresssurization but you 16 never told us about it. Shame on you. And we don't want to be 17 in that situation. Okay? We are imparting our knowledge to 18 them and we are basically saying we want to force you to read 19 it and understand it and acknowledge it, and then if you choose 20 to do nothing, fine. But you are aware of it. You can't sit 21 beck and say gee, you didn't solve the proble:a so we didn't 22 have to worry about it.
23 DR. KERR: If they did propose depressurization by 24 blowdown and PWR, then they would have to demonstrate that what 25 they were using is safety grade equipment?
O Heritage Reporting Corporation (202) 628-4888
103 t )
1 MR. S!!ERON: No.
2 DR. KERR: What?
3 MR. SilERON: No.
4 DR. KERR: Well, why not?
5 MR. S!!ERON: limm?
6 DR. KERR: Why not?
7 MR. SilERON: It's beyond design basis.
8 MR. CORRADINI: They would have to go at least to 9 show that they weren't violating 50.59, 10 DR. SIESS: It would have to be safety grade 11 equipment to prevent spurious actuation, Bill, but not safety 12 grade to reduce --
-- 13 MR. SifERON: PORV is required in steam generator
(")'
14 tube rupture for about 70 percent of the operating PWRs and it 15 has never been safety grade in those plants. And there is a 16 generic issue right now, generic issue Number 70, which is 17 addressing that very problem.
18 MR. SilEWMON: What is required, did you say?
19 MR. S!!ERON: The industry has never said that a PORV, 20 previous to TMI, has never said that a PORV needs to be a 21 safety grade device because it does not perform a safety 22 related function.
23 After TMI and after the Genet steam generator tube 24 rupture, we asked a number of PWRs how do you depressurize tho
?S plant in 30 minutes to meet the assumptions you make in your n
k) ~
lieritage Reporting Corporation (202) 628-4888
7~s 104 U 1 analysis? And they go, gee, we used a PORV. And I said, and 2 we asked them, we said, well doesn't that constitute a safety 3 related function? And they said yes. And we said well, why 4 isn't it safety grade. And they kind of scratched their head.
5 And every PWR tha.t ever re.'ied on a PORV to meet the steam 6 generator tube rupture since 1982 has a safety grade PORV.
7 MR. SHEWMON: The argument is they've got a block 8 valve behind it, isn't it?
9 MR. 3 HERON: No. They are required to open that 10 valve to meet Part 100. And they sever recognized that prior 11 to 1981. And there is a generic issue that addresset all the 12 plants that were licensed prior to 1981 that were licensed 13 without a safety grade PORV.
14 DR. SIESS: Did they have to replace the PORVs?
15 MR. SHERON: The generic issue was design.
16 DR. SIESS: All of those that have been licensed he.ve 17 a safety grade FORV, Are those different from the other 18 plants?
19 HR. SHERON: Yes.
20 (Continued on Page 105) 21 22 23 24 25 T
(t-'J Heritage Reporting Corporation (202) 628-4888
m.
. :h '
l< .
105 F 1 DR. SIESS: Are the valves different, or do they ;
- 2 'j ust have a different label'on them?
3 MR. 6 HERON: I really do not know. 'But they have to l 4 have the end stamp and the QA, and whatever else goes with :
e 5 safety grade. !
6 DR. SIESS: But you do not know about the valve 7 itself?
8 MR. SHERON: I.have not looked into the details of 9 the requirements.
10 DR. SIESS: All of those plants that do not have that-11 valve with all of those things on it are not as safe as the 12 ones that do have it?
13 MR. SHERON: I did not say that they were not as O 14 safe. I just said that the staff requirement --
15 DR. SIESS: They do not meet the regulations.
4
, 16 MR. SHERON: I did not say that they do not meet the 17 regulations either.
18 DR- KERR: So if an applicant proposed a blowdown of 19 PWR primary with non-safety grade equipment, the staff would 20 not question it?
i-l 21 MR. SHERON: We may question it, but not from the 22 standpoint of whether or not it is safety grade. One may have j 23 to look at such things as operability in an adverse 24 environment. For axample, we just had a preeentation yesterday l 25 from INEL which showed that for the depressurization option, O Heritage Reporting Corporation (202) 628-4888
'M 106 1 that you get substantial high temperatures if the pressurizer 2 valvo is open.
3 DR. KERR: I am not advocating this. To me, it has a 4 number of negatives. But it is suggested in here, at least 5 implicitly as an option, which would do away with the high 6 _
pressure sequences. So I am sort of asking what would happen ;
7 to a utility that came in and said we are going to depressurize t
8 and we are going to use KORVs, what would the staff then do. l 9 And you are telling me that they would do some sort of analysis 10 of what is being proposed-to determine on some basis that it 11 was reliable. Not on the basis of safety grade or not, but 12 something else.
13 MR. SHERON: Well, from the standpoint of e 14 reliability, it would only be -- what I was trying to point out ,
1 15 is that, for example, if for some reason the valve went shut 16 partway through the venting, would that produce a worse +
17 situation than if it stayed open the whole time. And there is i 18 some consideration of that, because one gets very high i i
19 ter.peratures in the upper dome of the pressurizer during a 20 venting. !
i 21 MR. LEE: Do you envision at this stage that there 22 exists a methodology that can be used to perform this kind of i
, 23 analysis to the satisfaction of the staff? ;
i 24 MR. SHERON: Yes.
i 25 MR. LEE: What would be such a methodology, where do
(:) Heritage Reporting Corporation j
(202) 628-488E
107 O 1 I go forward; for example, if the MAAP code is not acceptable +
2 to staff? ,
3 MR. SHERON: No, no, we did not say that. We said 4 that the IDCOR methodology is acceptable, but it would have to 5 take into consideration the information that is in Appendix 1.
6 MR. ELTAWDA: If it is unacceptable, it is because it 7 ignores something. And that is what we are saying.
A MR. SHERON: What I am saying is that.they would have 9 to consider the phenomenonology identified in Appendix 1.
d' 10 -MR. ELTAWDA: There are no certain set of codes that 11 are acceptable. It is the assumption that the utility comes >
12' and uses the right assumption, and that is the assumption that 13 .comes into the analysis.
O 14 MR. LEE: Again if I were go to a set of codes, you
! 15 would just have to accept the methodology.
16 MR. ELTAWDA: It depends on the assumptions that have 17 been used in these codes. You cannot just simply say these are .
18 the codes and they are acceptable, no. We want to know what ,
19 are the assumptions.
. 20 MR. CORRANDINI: Let me try it another way. I think 21 that I know what John is getting at, but let me ask. Or let's
, l 22 take an example that you used, Rango. What is Rango came back i t
23 and said, gee, we updated NUREG 11-50 at the front end, so that j 24 it matches up with what is really in the plant and walked it I
25 down, and we want to submit that as our solution to the front
(:) Heritage Reporting Corporation (202) 628-4888 i l '
l
. - - - , . . . . _ _ ~ , . - . _ _ _ . _ _ _ . . . _ _ _ . _ - _ _ . - _ _ _ . . _ _ _ _ _ _ _ _ _ ___._ _
108 !
O 'l end part of the IPE. And now for the back end, we want to 2 submit.NUREG 11-50 analysis for the back end. .
3 Is the NUREG 11-50 back end analysis acceptable? I :
. 4 am putting you on a spot purposefully though, because I am i
5 asking what examples can you provide to the utility people to ,
t 6 'say gee, if you were to do something. Truly, maybe this~is not 7 the thing to do because of cost or whatever. At least that is 8 some indication of what falls in the range of acceptable.
9 MR. ELTAWDA: What falls in the range of acceptable
. 10 is any method that gives one guidance.
11 MR. CORRANDINI: But would 11-50 follow? I mean it .
12 appears to me that it follows Appendix 1 guidance.
13 MR. ELTAWDA: If it follows it, then we review it at
()
, 14 that time. I cannot answer whether it is acceptable or not l 15 acceptable without knowing the assumption.
16 DR. KERR: Well, earlier you said that the 1150 PRAs 17 would be considered state of the art PRAs. .
18 MR. ELTAWDA: I would like to correct myself on that.
19 It ends at 11-50. We have not discussed it.
20 DR. KERR Okay. So now, I am down to Zion, Indian 21 Point, and Limerick as being state of the art.
22 MR. ELTANDA: At the back end. The back end always j 23 has to follow the front. ;
24 DR. KERRt So there does not now exist a state of the 25 art PRA? ,
o
(:)- Heritage Reporting Corporation .
l (202) 628-4888 i 1
1-
l 109
} 1 MR. ELTAWDA: For the back end.
2 MR. SHEWMON: So-it has changed. I did not think 7
3 there was. I s
4 MR. SHERON: We would.most likely, if someone came 'I 5 and said we are submitting the 11-50 back end, I hate to be !
6 presumptuous, okay, but we would look at it and most likely [
7 find it as generally acceptable provided that we could follow 8 through the reasoning that the utility went through regarding 9 what assumptions were made, and.whether it agreed or disagreed, .
10 or wriat type of conclusions that they drew from it.
11 MR. CORRANDINI: Dut I want to follow this logic, 12 because I think that it might just be a useful way to think it !
13 through. So let's say that you did that. And I understand O 14 where you are coming from that you have not really looked at 15 it. But under the assumption that you looked at it and you ..
16 found if acceptable, now let's go the next step further. We 17 have not gotten to your screening criteria. ,
18 What if the utility was not Grand Gulf but was a 19 MARC-3 or looked at the MARC-3 or the Grand Gulf analysis and 20 said well, we as a utility, the best that we can do at this 21 point is look at the geometrical differences in our containment 22 and in our primary system, etcetera, and given all of that we L 23 still fit in the back end, we still fit into that sort of back 24 end analysis.
[
25 Thst would be essentially not what you are giving in ,
O Heritage Reporting Corporation l (202) 628-4808 m . , _ _ _ . - . .._, _ . . _ _ . , _ _ _ _ _ _ _ _ _. _ _ . _ _ _ _ _ _
110 o
k-) 1 Appendix 1. It would be essentially saying we will use what 2 was in the MARC-3 back end analysis, and we will look at the 3 variations off of the back end analysis for our particular 4 MARC-3.
5 Granted that you cannot tell me whether it is 6 acceptable or not, but would that at least be a method which 7 they could use to do the back end analysis, take a 8 NUREG 11-50 analysis and look at, what shall I call it, 9 variations off of that?
10 MR. ELTAWDA: I think that we are saying that in 11 Appendix 1, that they can use other analysis.
12 MR. CORRANDINI: You will look at it?
13 MR. ELTAWDA: We will 1cok at it.
14 MR. MICHELSON: Brian, I have got a couple of 15 questions which I think that we dealt with a little earlier, 16 but I would like to tell you where I am a little concerned.
17 One of them again is the question of the kind of failures of le say low energy systems that can be experienced outside of 19 containment.
20 The traditional analysis within the design basis is 21 to look at a certain size of break in low energy as being the l
22 maximum break in energy that would occur. The problem that I 23 have now if you want to go beyond the design basis is to think 24 about the possibility of low energy billows failing. And they 25 do not fail as a critical crack, but they fail ae a n
Heritage Reporting Corporation .
l (202) 628-4888
. e i
111 :
'O 1 cate trovaic reetere. '8e 8111e 81 terv " seee eed ie thet l
2 regard.
3 There are a number of safety systems in this country -l 4 -in water particularly where a large number of billows are t
5 ~ involved. I'never find them in the PRAs as a consideration. !
6 But maybe they are there, and I have not found them. But you f
7 have to now think for the back end analysis of what kind of low 8 energy pipe ruptures are you going to accept and component. j 9 failures. And bolt enclosures being another one. Do bolt f
10 enclosures over fail? Well, there is a very low probability, t
, 11 but I think finite that they can fail. .
12 And I do not find any discussion of those kinds of 13 approaches at all, and that is interesting. I do find a ,
O 14 diecuseion -f meintenence erocedures. and 1 em woneering if ;
- 15. you have in mind there analyzing the consequence of a t
16 maintenance man leaving a valve's top works off isolated, and :
17 the next day the operator opens the line and floods the system.
18 This has actually happened. It is not an imagination.
19 And is that the kind of maintenance procedure that 20 you have in mind, and are you going to search out these kind of i l
21 potential maintenance failures from the viewpoint of l l l 22 catastrophic events?
- 23 MR. BECKNER: I do not think that what is going to 24 come out of the IPEM or the PRA is going to be beyond the state
! 25 of the art. I think that the types of things that you are :
l' f
! O neritege Regerting cergeretien l (202) 328-4888 !
i I
-v -
112 7
() 1 bringing up, if they occur, is someone finds them somewhere, 2 that we NRC as a clearinghouse would bring those types of 3 things to the notice of plants that might be vulnerable to a 4 similar type of thing.
5 But I do not think that the methods that we are 6 proposing here would be unique, to specifically identify those 7 types of unique things in the plant. It is only as they are 8 found for whatever purpose, and we try to make that information 9 generally available.
10 MR. MICHELSON: Well, these are found. They are well 11 known. I do not necessarily notice the agency highlighting ,
12 them with bulletins or something. But the experience of 13 leaving the top works off and having the line repressurized n
\_) 14 happened not too long ago, and it flooded a fair amount of 15 equipment.
16 In another case at another plant, that same kind of 17 situation might have indeed led to core melt for all I know.
18 You have to look to see if you have the vulnerability. And I 19 do not recall if that was highlighted in any strong way, so 20 that every utility in the country looked to make sure that they 21 did not have a problem.
22 MR. DECKNER: I do not think that it would be.
23 MR. MICHELSON: I thought that it would be done undor 24 an IPE.
25 MR. BECKNER: Under the course of this program, if l Heritage Reporting Corporation (202) 628-4888
r 113
\_/ 1 someone came in and identified, for instance, the cover being 2 left off, if they identified that as a major vulnerability of 3 the utility, then we may question, well, does that possibly 4 occur at other utilities; and if so, why have they not found 5 it.
6 MR. MICHELSON: Why do you believe why it will even 7 be brought up by any utility, if it does not appear to be 4
8 within the scope of what we expect them to do?
9 MR. SHERON: Carl, let me try and address that if I 10 csuld. You are raising a rather specific issue.
11 MR. MICHELSON: As an example of maintenance.
12 MR. SHERON: Yes. What I am saying is that you have 13 also raised a number of others. For example, partial loss of 14 air systems, partial voltage, and the like.
15 And I guess that we did not feel that it was 16 appropriate for the staff to try and list every conceivable 17 type of failure that one might be able to think of in the plant 18 and list it, and then go off and tell the industry you have got 19 to think about all of this stuff.
20 First off, if you remember, the IPEM is derived from 21 a PRA, I guess what we consider to be more or less state of the 22 art PRAs and the like. If these were significant contributors, 23 number one, I would hope that the PBAs, that we are identifying 24 them. If they were not, I think that there is a more generic 25 question that I think needs to be addressed.
I O Heritage Reporting Corporation (202) 628-4888
, k 114 7)
(_
1 MR. MICHELSON:
Why would the PRAs identify them when
- i i '2 they are not in the models to begin with? j 3 MR. SHERON: But you are challenging the fundamental.
4 If we said that --
5 MR. MICHELSON: It can be done. The PRA' people tell [
6 me yes, if that is a requirement, we know how to set up models J
7 'and do it. It is a lot of work, no doubt, but it can be done.
8 It is not beyond the state of the art. It is beyond the state 9 of present modeling procedures, not techniques. The techniques 10 exist for modeling such events. But_it is not in the PRA, 11 because it is a lot of work. And how far do you go. That is 12 what I thought, that this was to proscribe, 13 MR. SHEWMON: What he is saying is that if they think ,
() 14 that it is a significant contributor, then it could be there.
F 15 MR. SHERON: We certainly will go back and take a e
16 look.
17 MR. MICHELSON: But how do you decide a significant ,
l 18 contributor, if you have not modeled it in at least one or two l
19 cases to kind of search out whether these are significant 20 contributors. We are claiming that we are looking at 21 maintenance in here as a potential for severe accidents. That
. t l 22 is in your very general and vague words. And I am just giving 23 you a specific case.
24 Are you looking at that kind of maintenance?
25 HR. SHEWMON: Well, you heard the answer, and the !
t Heritage Reporting Corporation I (202) 628-4888 !
115 A> 'I answer is.no.
. '2 MR. MICHELSON: The answer is no. I find that a
'3 significant shortcoming.
4 MR. SHERON: You may find even more significance in 5 that we are proposing to spend about six staff months reviewing 6 each of these IPE submittals, which-is far less-than what we 7 normally spent in the past when we reviewed a utility submittal 8 of a PRA.
9 Again it is a matter that the Commission has stated 10 that they do not think that there are out-11ers out there, and 11 we are doing this more or less to confirm that and the like.
12 That we are not going to be doing very detailed reviews again 13 to look at each and every, to make sure that they have O 14 identified every possible sequence and the like.
15 We want to make sure that they have done a credible 16 job, and that they have done a competent job. And that what 17 they are coming,up with and the like appears to be reasonable.
18 But we do not have the resources, nor do I think do we believe 19 that it is necessary to go into that detail on each and every 20 IPE that is submitted. Dr. Siess.
21 DR. SIESS: What is your definition of out-lier 22 again? You have just used the word. Is it something that 23 certainly dominates the vulnerability, or is an out-lier 24 something that happens that nobody ever thought would happen 25 like Carl is talking about?
C:) Reporting Corporation l
Heritage (202) 628-4888 l
116 O
x- 1 MR. SHERON: What we are talking about as an 2 out-lier is that in general we understand where the risks come 3 from plants, but we recognize that there are plant unique 4 features. And that when one does a plant specific PRA or an 5 IPE, that there is a potential that you can identify individual 6 weaknesses in the plant as a result of that examination of the 7 individual plant.
B DR. SIESS: But only if you look for it. And 9 usually, you only look for things that you have reason to look 10 for, because they happen.
11 Would a PRA have spelled out the TMI incident as a 12 probability, or would that be an out-lier?
13 MR. SHERON: In terms of the specific sequence?
14 DR. SIESS: Yes, did anybody ever put in a PRA that 15 the operators would turn off the pumps?
16 MR. SHERON: I seriously doubt it.
17 DR. SIESS: I seriously doubt it, too. And that 18 caused an accident.
19 Was that an out-lier that we did not find or would 20 not have found or what?
21 MR. SHERON: I would argue that it is some sort of an 22 out-lier. We found it the hard way.
23 DR. SIESS: You know, PRAs calculate probabilities of 24 various ways to get into trouble, and yet none of them would 25 caiculate the one incident that has already happened. It is O Heritage Reporting Corporation l (202) 628-4888
r 117 -
1 pretty poor calibration, is it not?
2 MR. SHERON: Not necessarily. I did not scy that j 3 they could not calculate the one that happened.- I just said ;
4 that they could not calculate that sequence. One would not ,
i 5 calculate the Davis-Bessie sequence, but one could probably -
6 calculate that the probability of a loss of all feedwater might 7 have been pretty closo.
8 HR. SHEWMON: You mean that they could get a good ;
9 calculation on the probability of all feedwater without <
10 including that sequence?
11 MR. SHERON: That specific sequence. That was one 12 specific sequence. As an argument, I said that there is -l 13 10 to the 29th ways that I can lose auxiliary feedwater. I am O 14 sorry. If I take a specific sequence like Davis-Bessie, it may ;
15 individually, each failure may have a combined probability of
(
16 10 to the minus 27th in terms of getting to the end point. But ;
17 I might argue that there is 10 to the 24th ways that I can get ,
18 to that same end point wnich gives me an aggregate probability ,
19 of say 10 to the minus 3 or 10 to the minus 4 of losing [
20 feedwater. I cannot calculate that specific sequence with f
21 every particular nuance in terms of a failure.
22 MR. SHEWMON: It sounds good when you say it fast, 23 but I still do not take it.
24 MR. SHERON: That is why I said it fast.
25 MR. SHEWMON: I am still wondering whether I can look O Heritage Reporting Corporation (202) 628-4888
118
, 1 at experience to calibrate PRAs with any confidence.
2 MR. SHERON: I think that the real question that you 3 are asking is more or less one to tall to the PRA practitioners 4 in the state of the art PRA. We are not asking the industry to 5 forge new ground in the sense of expanding the current way that 6 we do PRAs. I mean we have already said that we think that the 7 11-50 PRAs are state of the art. And if they are not 8 addressing this, you know, maybe that is something that should 9 be taken up with the staff at a separate meeting. But I am not 10 suggesting that the industry should go off and forge new 11 ground.
12 MR. SHEWMON: Your answer is partly convincing in 13 terms of something like Davis-Bessie and in terms of PRA. But 14 one of the major objectives in the program that you are talking 15 about here is accident management.
16 MR. SHERON: Yes.
17 MR. SHEWMON: Now certainly, accident management is ,
18 very sequence oriented.
19 MR. SHERON: I am not sure that I agree with that.
20 MR. SHEWMON: You do not think so.
21 MR. SHERON: We just have not determined that yet.
22 It inay be symptom oriented.
23 MR. SHEWMON: I would think that it would be much 24 more sequence oriented than the 10 to the minus 29, or 10 to 25 the 27, whatever, you have given me on Davis-Bessie, o) i Heritage Reporting Corporation (202) 628-4888 1
119
(_) 1 MR. SHERON: Okay.
2 MR. SHEWMON: Because you might have to do things 3 that depend on what failed.
4 DR. KEER: Why do we not lot him talk for awhile.
5 MR. SHERON: Well, I have basically gone through what 6 I intended to say on the back end here. So I would propose to 7 move on to another noncontroversial subject which is A-45.
8 (Continued on next page.)
9 10 11 12 13
")
(/ 14 15 lb 17 18 l
19 20
! 21 l
l 22 23 24 25 Heritage Reporting Corporation (202) 628-4888 l
1 l
i l
l l
120 ;
( 1 MR. MICHELSON: Are you going to talk about errors l
. 2 of omission, errors of commission and so forth? What do you !
l 3 expect the utility to'do with that sort of idea on their
(
4 knowledge, which maintenance is a part of the same problem?
5 MR. SHERON: I don't think we're imposing that the f
4 6 utilities consider errors of commission. As I said before if 7 they assume an operator is going to take an action, which 8 enhances the ability of the plant to mitigate or withstand a 9 severe accident.
10 From a human factor standpoint we would probably like 11 to see some confirmation on the part of the utility that the i 12 operator isn't properly trained to take this stop.
13 MR. MICHELSON: Let me read your words that confused ,
14 me, because your answer is already highly divergent. Item 3 on
- 15 page 18, which is one of those magic six or eight that was
16 referred to earlier. Item 3 says, okay page 18.
l 17 MR. SHEWMON: This was page 18?
18 MR. MICHELSON: Yes.
^
j 19 MR. SHEWMON: Of appendix one?
20 MR. MICHELSON: Yes, appendix one. i l
, 21 MR. SHEWMON: I got a lot of page 18's.
- i 22 MR. MICHELSON: Oh, I see. I guess there, let me
^
23 double check.
24 MR. SHEWMON: I've got it.
25 MR. MICHELSON: Okay. Item 3 tsiks about system l .
C:) Heritage l Reporting Corporation 3 (202) 628-4888 .
k
,..- -,_e.,_ .__, - - m. y,-.-< - . - - , , . - ~ - . , _ _ _ , ._..- .-
121
() I human response should be integrated probabalistically with 2 phenomenological aspects and to simplify but realistic 3 containment of entries for plants being examined.
4 I think you are saying you expect that that's going 5 to be done. Is that right? Is that what that is supposed to 6 be saying?
7 And I thought you just got done telling me now we 8 really aren't going to be able to do much of that, so this 9 sounds great.
10 And If you are going to do that in your modeling I 11 think that would begin to get the answers I would need.
12 MR. SHEWMON: This is through co6tainment.
13 MR. MICHELSON: Yes. Yes.
14 MR, SHEWMON: But you were talking about before what 15 did containment. ,
16 MR, SHERON: Yes, well it is part, yes. It's 17 containment bypase.
18 MR. SHEWMON: Well with the extent on which some 19 malfunction; due to maintenance or other things, is already 20 factored into the PRA if they have performance or reliability 21 data for equipment, isn't it?
i 22 MR. MICHELSON: No not, those are not--
23 MR. SHEWMON: As to how it's liability equipment?
24 MR. MICHELSON: No, this is the problem where the 25 fellow goes to maintain the equipment.
[ Heritage Reporting Corporation (202) 628-4888
, 122
( ,)
'~
1 MR. SHEWMON: Well that's a particular part, and when 2 you get reliability integrated over a hundred plants over ten 3 years you don't talk about whether the maintenance guy screwed 4 up, or it's rusted shut.
5 MR. MICHELSON: Only, you talk about if only if the 6 equipment subsequently fails to function properly. Here is a 7 case where the equipment was disassembled and not reassembled 8 before the energized the system.
9 MR. SHEWMON: Somebody might say it failed to 10 function. Now how that, if you record it in your data base, 11 it's--
12 MR. MICHELSON: It wouldn't be. But yet I thought 13 those kinds of things were going to be considered, in reading r-)
\")
14 item 3 it sounded like pretty good scope, and you're really 15 going to do that.
16 MR. SHEWMON: You're just an eternal optimist.
17 MR. SHERON: I think you're reading a little bit more 18 into this than what's intended.
19 MR. MICHELSON: That perhaps is. What should I have 20 read into item 3, then?
21 MR. SHERON: Well just, if you read the second 22 sentence in there, you know it's what I was saying before, and 23 that is that--
24 MR. MICHELSON: No, the second sentence is ancillary 25 to it. The first sentence is the, what you think is going to Heritage Reporting Corporation (202) 628-4888
123 !
~
1 be'done, and the second is in addition. It should be made for i
2 recovery of, for other accident management procedures and I 3 think that myself. t l
"4 MR. SHERON: Oh, ace're saying that the operator is 5 going to interact. The operator is expected to interact, when 6 looking at containment' performance you need to consider that.
t 7 You know, that's another way, just a way of saying 8 that.
9 MR. MICHELSON: If the operator accidentally does the l 10 wrong thing and opens the containment isolation valve during an 11 accident, for instance, is that included? Is that a constant 12 error? Is that the, it might be in the PRA as opposed to 13 isolate?
() 14 MR. SHERON: No.
15 MR. MICHELSON: All right, I'm sure emphasized.
16 That's mentioned under containment, a 17 MR. SHERON: It will tell you to isolate--
18 MR. MICHELSON: Well if you'd open the valves, it'll 19 tell you to isolate.
20 MR. SHERON: It's an error omission, he opens what he i 21 shouldn't have been fiddling with, but he did anyway. And he 22 reached for the wrong handle.
23 MR. MICHELSON: Well under appendix 3 in the Occident 24 management, the third paragraph says nothing is required at 25 this time. So what does that mean?
Heritage Reporting Corporation (202) 628-4888 1
124
() 1 MR. SHERON: Hit and delay.
2 MR. MICHELSON: Well I assume it means that a 3 licensee doesn't get-- When we get to the accident management 4 section, I can explain, it's a part of accident management.
'S MR. SHERON: Well I can explain, it's more of what we 6 were talking about, appendix one 7 DR. KERR: Proceed.
8 MR. SHERON: A-45. I'm not sure whether, I guess 9 most of you at one time or another heard numerous presentations 10 on the A-45 issue, which la requirements for decay heat 11 removal.
12 The results, if you remember this issue was started 13 back in around 1981 I believe it was and it is probably one of
() 14 the most thoroughly examined generic or USI's that the 15 Commission has ever worked on.
16 The analysis itself was done by doing a limited scope 17 PRA on six plants. To look at the decay heat removal systems 18 and their contribution to risk, etc., on six different plants.
19 In addition, consideration was given to sabotage and 20 the like. The conclusion that was reached by project manager 21 on the A-45, Andy Markacy, was that if one was to, if one were 22 to come up with a truly generic solution, that the generic 23 solution would be to backfit essentially on to most operating 24 plants a dedicated decay removal system.
25 When one goes through a cost benefit analysis, on O Heritage Reporting Corporation (202) 528-4888
_- _ , . . _ . . - . ~ . . . ._.
1 4
125 q
() 1 such a system, the only way one can come up with a favorable ,
2 coat benefit ratio, to justify such a system would be to 3 include not only-on-site costs, but also other considerations 4- like a moratoria on nuclear power plant operation and so forth.
I 5 Which we felt, the staff felt, was well beyond'the 6 5109 considerations and totally, also not consistent with 7 previous use of backfit analysis and cost benefit analysis.
8 So we did not endorse backfitting a dedicated system i 9 on plants, but what we did conclude was that when we looked at 10 the six plants that were analyzed for A-45 we found that there 11 were unique contributors to decay heat removal vulnerabilities.
12 And that these vulnerabilities were as I just said i 13 unique to each of the six plants. In some I think the rough ;
() 14 rule of thumb we have is about half the plants or so, I think 15 the majority of the risk was coming from external events, 16 primarily seismic, and the other was, the other half coming i
17 from internal events. l t
18 And what we concluded was that in order to look for 19 vulnerabilities in the decay heat removal system one would have ,
20 to look at each plant's decay heat removal system individually, 21 and looked for these vulnerabilities. ,
i !
) 22 In the event that we did not have this severe l 1 !
i 23 accident policy or an IPE, the staff would have probably
- 24 endorsed conducting an individual examination for decay heat ,
, 25 vulnerabilities in each plant. j i
' ([) lieritage Reporting Corporation l (202) 628-4888 1
I --.---- ... - - - . _ - _ -
126
() 1 Howaver, because we had the severe accident policy 2 which was proceeding to implement these individual plant 3 examinations, again from the standpoint of both redundancy and 4 not asking the industry to do the same thing twice, it appeared ,
r 5 that we could, that through the IPE process that we hope that 6 decay heat removal vulnerabilities then would be identified as 7 part of the IPE process.
8 Because of that, we concluded that if one 9 successfully completed the IPE, one in fact did the examination 10 that we would have required of A-45. So for this reason we 11 proposed to subsume A-45 into the severe accident policy.
12 And I use the word subsume as opposed to resolve 13 because to me subsume means that A-45 has gone away. One need
() 14 not show later on down the road that the issue is resolved, but 15 one just has to'say that, conclude that the IPE process will in j 16 fact accomplish the same objective as A-45 intended to ;
j 17 accomplish, and therefore whatever the staff concludes on the 18 IPE is satisfactory.
i j 19 MR. MICHELSON: How do you build an IPE on a new j 20 plant, since new plants don't have to meet the resolution of A-21 45. Got resolved by--
22 DR. KERR IPE was never designed for that.
! 23 MR. MICHELSON: No, but I am wondering, what I am ,
24 getting to and then how do you address A-45 for new plants, if f
i 25 you don't face up to it here.
f Heritage Reporting Corporation
, (202) 628-4888 i i l i
_. 127 1 MR. SHERON: Well number one, the CP rule I believe 2 requires that plants do a PRA, number one. Number two is that 3 there is a program on-going right now, which is implementation 4 of the severe accident policy for future plants.
5 MR. MICHELSON: That's under, you say that's under 6 development now?
7 MR. SHERON: Yes, Billy Morris' division is working 8 on that.
9 MR. MICHELSON: That's where I think that you would 10 address it.
11 MR. SHERON: And that's where that would be picked 12 up. Dr. Siess?
13 DR. SIESS: You use the expression of A-45 being 14 subsumed into the IPE. If I eliminate ATWS, and event B, what 15 severe accidents are there that aren't subsumed into A-45?
16 MR. SHERON: We estimated somewhere about that A-45 17 encompassed somewhere 80 to 90 percent.
18 DR. SIESS: Okay. So really-19 MR. SHERON: Or, if you did a PRA on a decay removal 20 system, you've done 80 or 90 percent of a PRA.
21 DR. S7.ESS: So you can subsume in either direction.
22 MR. WARD: In speaking of the other direction, I 23 think I could build a case where every existing USI or generic 24 issue could be subsumed into the IPE program.
25 MR. SHERON: That is correct. Let me address, if I O Heritage Reporting Corporation (202) 628-4888
4 I
s t
SR5 128 l
( 1 could that,-I kinew that question and we have anticipated it and 2 had answered ourselves. !
3 What we did is we went through and we looked at every l 4 generic issue and every USI. And tried to determine which
'S USI's or which GSI's might he able to_be resolved via the IPE 7 6 process. ,
7 And what we concluded is that there were indeed a 8 number of them that might be resolved. We also recognized i 9 though that in most of the GSI's if you take a look at them, 10 number one all had a resolution schedule that halted their 11 resolution within about two years.
i 12 So we concluded that most of these would probably be
- j. . . :
- 13 resolved before the generic letter went out and before the ;
O. 14 industry started to really come in and submit their IPEs.
l l'
15 DR. SIESS: That was USI's?
16 MR. SHERON: Yes, and GSI's. !
I 17 DR. SIESS: You said with a two year schedule. Was !
i 18 that all of them or just USI's?
. . t 19 MR. SHERON: I think that's both. I think all of 20 them really have a schedule within about, I'm not saying all of i t
21 them, but 90 percent of them I think are scheduled to be ;
I 22 resolved within about two years.
j- 23 Secondly, was that if you look at most of the GSI's
- 24 for example, they deal with a very specific issue, sometimes l 25 just on a class of plants, and for the most part a resolution i
() i Heritage Reporting Corporation i (202) 628-4888 i
129 C
\,_) I has already been identified, and the majority of the work is 2 putting together the regulatory analysis that either justifies 3 that resolution or somehow would rationalize a modification to 4 it.
15 So there is usually in place some sort of a 6 resolution already, and for that reason we decided that it 7 would not be appropriate to include all of these in the IPE.
8 Now another reason, is that the industry was very 9 adamantly opposed to including resolution of USI's and GSI's in 10 general in the IPE process.
11 At one time it was, if you remember, A-45 was 12 becoming the catchall. When ever anybody raised an issue about 13 something, we said well that will be solved in A-45.
() 14 It got to the point that people were then saying well 15 don't worry about that we'll solve that in IPE. And pretty 16 soon the IPE was starting to fall over on its own weight. We 17 were solving everything in it.
18 And the industry was basically saying is that if you 19 make us solve everything in it, you'll never get a solution.
20 So they very strongly asked us to not include a lot of other 21 baggage in the IPE. And I guess we took a long look at it, and 22 we agreed that it was not appropriate.
23 Now one thing we did do in the letter though, is to 24 put in the option. In other words if a utility did their IPE 25 or the PRA, whatever, and they concluded that either an O Heritage Reporting Corporation (202) 628-4888
e I "
130
() 1 enhancement that they identified resolved an existing USI or 2 GSI, or in fact they did their analysis and concluded that for 3 that USI or GSI as it relates to their plant, their plant was 4 acceptable as is and nothing more needed to be done.
5 They could come in and volunteer a resolution. And 6 we would certainly welcome it and review it. And if we agreed 7 with their conclusion, then we would declare that issue solved ;
8 for that plant. ,
9 DR. SIESS: You have two thoughts there, one of which 10 I see stated in the letter, and the other one I don't. The l
11 first was that if the utility found a fix that would take care 12 of a USI or GSI as well as some other vulnerability.
13 The second was if they thought it wasn't necessary as
() 14 a result I don't see the second one in here.
i 15 MR. S!!ERON : I'm not sure that is. If they found 16 that there was no fix necessary, we can certainly add that, if 17 that's not there.
18 DR. SIESS: But once you've got a good PRA you might 19 find out that some of the generic issue resclutions really are l
20 trivial. That was one of the carrots that was being held out 21 in ISAP too. And I didn't find it in here the way you said it.
22 MR. S!!ERON: Well we can certainly add that, if it's 23 not there.
24 DR. SIESS: Where you have on page 6, I found very 25 difficult to understand. It's referring to a notable residual Corporation I lieritage Reporting (202) 628-4888
131
,O' '
1 vulnerability.
2 And you roa).ly need a glossary with this thing.
3 There are words like that, there are words like unusually poor 4 containment performance.
5 MR. SHERON: I think that has been fixed. Oh, I'm 6 sorry, I apologize here. The policy statement uses those 7 words, for unusually poor containment performance.
8 DR. SIESS: So that's why you use them.
9 MR. SHERON: We had a chance to fix that. Well, I 10 mean since we are supposed to implementing the commission's 11 policy we felt that it was appropriate to try and use the 12 language that they used.
13 DR. SIESS: Well that means you don't really need to 14 define large release then. That came out of a policy statement 15 that didn't define it either. I mean, this is nice.
16 MR. SHERON: Anyway, what we have done is we have 17 said that since we are subsuming A-45 into the IPE's we would 18 expect that the IPE that the utility conducts would take a 19 specific look at the decay heat removal function.
20 And what we've done is we've added, as I said this 21 appendix where we have tried to summarize the insights very 22 briefly from the six limited scope PRA's that we did for A-45 23 in terms of identifying what type of vulnerabilities we found, 24 what systems they were found in and the like.
25 Which is basically just some useful information to O Heritage Reporting Corporation (202) 628-4888
132 1 the analysts that are going to do this. Again it is just 2 passing on and imparting knowledge that we learned from the six 3 case studies.
4 MR. LEE: In that regard, in this appendix five you 5 are referring to the fact that decay removal issue is highly 6 specific and at the same time, there is a statement to the 7 effect that also the issue is highly dependent upon human 8 factors.
9 Which seem to be more generic in nature in my 10 opinion. So to me, they are total contradictory to some 11 extent.
12 MR. SHERON: It is not contradictory, because when we l 13 went through the Point Beach analysis that Westinghouse and
) 14 EPRI did, and when we compared it to analysis that the staff 15 contractor did, at Sandia, what we found was that that factor 16 of 30, difference, in the bottom line results , a lot of it was 17 attributed to assumptions regarding operators, operator 18 actions.
19 And we are in the process right now of preparing a 20 more extensive document, which summarizes these interactions 21 with the industry on the Point Beach and we are going to 22 reference that document from the IPE letter, so if anybody 23 wants to learn more about the details of the industry analysis, 24 and the staff analysis and why there were differences and the 25 like they will be referenced through this document.
() Heritage Reporting Corporation (202) 628-4888 l w -w ----r -w ,mnyw-
w--
---,,4y w -- --~ - - - e- ,, m y -w-m -e ---
e 133 '
(%
Ll 1 MR. LEE: Maybe I didn't phrase my question 2 proporly. But if there is indeed a factor of 30 difference in 3 risk or whatever, not when you get due to the difference in 4 assumptions that are made, then wouldn't it dominate any I 5 hardware plant specific differences?
6 DR. KERR Particularly if this is the human factor's ;
i 7 assumption. What you seem to be asking for is not a plant '
8 specific one, but a generic one.
9 Because here are three, on two different groups on 10 the same plant, they arrived at completely different 11 conclusions which you felt were unreasonable because they 12 weren't generic.
13 MR. SHERON: In this instance, okay. In other words, O, 14 number one, I think there are two levels here. One is thac 15 when you look at the staff analysis, okay. Let's not even talk 16 about the industry yet.
17 If you look at the staff analysis of the six plants, 18 what you find is that those things that are contributing to 19 risk are highly plant unique, and very variable. ;
20 Some plants, the big contribute is to risk, or the I 21 seismic, the external event. The others it's internal 22 initiators.
23 Superimposed on that is when the industry goes off 24 and does the same analysis they get a different answer. And 1
25 what we found is that the reason for that different answer had Heritage Reporting Corporation (202) 628-4888 i p
1
134
'J 1 to do with human factor assumptions.
2 Now we accepted some of the industry assumptions, and 3 we didn't accept others. Now from the standpoint of what that 4 means. What does it mean, accept. I don't know, because we 5 are not approving it. ,
6 DR. KERR Well, you fail to accept it because it was 7 generic to the point bucause you felt it was generics.
8 MR. SHERON: Unjustified.
9 DR. KERR: But generically unjustified, 10 MR. SHERON: Not justified with respect to the 11 assumntion made for Point Beach.
12 DR. KERR: But it might have been justified for 13 another plant.
(n 14 MR. SHERON: It might have, I don't know. The review 15 that we did was specifically, you know. We were not trying to 16 sit there and say this was a generic problem. But we were j 17 trying to understand where the differences between the analysis 18 to Point Beach did in Westinghouse.
19 DR. KERR: But the assumption about the human factors 20 must surely not have been based on the studies in specific to 21 Point Beach operators. They were based on assumptions about 22 operators generally, weren't they?
23 MR. SHERON: Do you want to say something John?
24 MR. CHEN: Maybe I'll give some examples, of the 25 uniqueness in the A-45's plant analysis. Some of those plants O Heritage Reporting Corporation (202) 628-4888
W l
135 1 will be the risk came out from seismic. Some came out from the 2 plant.
3 DR. KERR Can we talk about the human factors 4 problem first?
5 MR. CHEN Okay, I'll also keep the human' factors 46 apart.
7 MR. SHEWMON: Can we back up then about six inches 8 from that?
9 MR. CHEN Okay. Let's talk about one of the related 10 human factors that they talk about the decision. There is in 11 analysis talk about, there is a potential for flood induced 12 damages.
~ 13 And that flood was triggered by a dam maybe 500 miles 14 upstream. And the analysis assumption itself stated the 15 operator would not take actions. That this one part tie in 16 with the human factors.
17 The industry feels for 500 miles away a dam failed.
18 The operator at least got to have some kind of warning 19 somewhere, that he can take proper action.
20 DR. KERR: Our question has to do with whether the i 21 assumption about operator action was based on general 22 considerations of operator action or about plant specific 23 operator actions.
24 MR. CHEN: This is plant specific.
25 DR. KERR: So those operators wouldn't have taken O Heritage Reporting Corporation (202) 628-4888
136 1 action, but an operator in the plant next door might have. Is 2 that what you are saying?
3 MR. CHEN: This is just an assumption in analysis, 4 assuming the operator did not take the action. It doesn't 5 mean--
6 DR. KERR: But would you have assumed that about any 7 operator or just about the--
8 MR. CHEN: Just for that particular plant, had that 9 kind of particular failure phenomenon.
10 DR. KERR Now wait a minute. We are talking about 11 operators.
12 MR. SHERON: There is a dam, 500 miles up river, that 13 fails. We would assume in general that the operator--
14 DR. KERR Okay. That's what I was trying to get at, 15 to me that is a generic consideration. I'm being critical of 16 it, but it's generic for a specific scenario.
17 MR. SHERON: Yes, but it wasn't a consideration of 18 plant specific characteristics of those operators. It was a 19 consideration of generic characteristics of operators.
20 DR. KERR But they are, of the six plants that we 21 looked at, there may have only been one plant that had that 22 unique vulnerability of having a dam fail 500 miles upstream, 23 you know. In which case then, we would call that plant 24 specific.
25 DR. KERRt Now John's question was whether these O Heritage Reporting Corporation (202) 628-4888
p 137
() I human factor things, which apparently are important, are really 2 expected to be plant specific, or are they expected to be 3 generic?
4 MR. LEE: If the human factors can make so an order 5 of magnitude type difference into risk calculation, risk 6 estimate. Then that is--
7 MR. SHERON: This is what Westinghouse and EPRI came 8 in and told us they assumed for the operators. I don't know 9 what somebody else would come in and tell me. Problem? I l 10 can't argue--
11 MR. LEE: This is in your appendix to the generic 12 letter. That is why I am trying to understand this particular 13 point. ,
() 14 And especially when later on, in other appendices you 15 are saying that accident management, or human factors related 16 to accident management also should be included.
r 17 And then the operator errors of commission and 18 omission, all of these cannot be so plant specific, they have 19 to be, it has to be generic, and it can make a big difference 20 in my opinica.
21 MR. SHERON: I'm not going to argue, I'm not going to 22 say that they are all generic because I am trying to point out 23 that this was one utility, or one vendor that walked in and 24 said, these are our assumptions.
I 25 And we happened, when we looked at that one vendors j
Heritage Reporting Corporation
- (202) 628-4888 i
i
130 ud h I assumptions, and their analysis with that plant, okay, we 2 disagreed.
3 I can't, you know, I can't say you know, that the 4 next guy that comes in, if combustion walks in or GE walho in 5 and does an equ4 valent analysis of one of our A-45 plants, that 6 any differences that arise are going to be due to human factor 7 considerations.
8 We may agree a hundred percent on the way the 9 operators are assumed to conform, do you follow?
10 MR. IEE: Yes.
11 MR. SHERON: So I'm just trying to, we look at this 12 that was a unique situation.
13 DR. MERR I don't believe that the J i s c u s s .i.on is pm
() 14 converging, John.
15 MR. SHERON: I'm not, maybe I am not.
16 DR. KERR I don't believe that it is likely to 17 converge.
18 MR. SHERON: We are supposed to stop for lunch 19 somewhere around noon. Is there a more convenient stopping 20 place for you than right now?
21 MR. SHERON: Let me just do six, which is benefits of 22 PRA, and then the next item is to screening criteria, which I 23 presume will also--
24 DR. KERR: Okay, all right. Now that we have seen 25 that slide, let's go to lunch. We will return at approximately (n
L) Heritage Reporting Corporation (202) 628-4888
. . . ~ . _ _ _ _ _ . . _ _ _ _ . . _ _ _ _ - . . - _ _ _ ._ _
4 ,
4 ,
l 1
l 139 !
- 1 1:05.
2 (Whereupon, at 12:05 p.m., the hearing was recessed, !
l 3 to reconvene at 1:05 p.m., this same day, Tuesday, April 26, j l l j 4 1988.) !
i s 5 , ,
4 ;
4= '
! 6
! 7 i i 8 i
4 9
, 10 l 11 i
! 12 l
! 13 14 i
! 15 16 17 18 19 20 21 t
l 22 i
} 23 24 25 O Heritage Reporting Corporation (202) 628-4888
.m. ._. ~ .
ft .
. 140 1 A F T E R-N_ O O N_ E E_ S g-I__O N_-
2 (1:05 p.m.)
3 DR. KERR: Whenever you are ready, sir,.
4 (Slides'being shown)
MR. SHERON: Let me go back to Item 6, which is the 5
6 Benefits of a PRA, very quickly. I want to point this out, ;
7 that.as I said, we think that the other USIs and GSIs that-8 might be resolved, if there is a PRA in place, this 1s-not to 9 say tbst the IPE could not do the job, but I think we would 10 _have to take a harder look to see if it was capable of picking 11 up specific vulnerabilities associated with the issue here.
12 MR. DAVIS: Excuse me, Brian.
13' MR. SHERON: Yes.
( 14 MR. DAVIS: Let me see if I can clarify something on 15 that relative to Dr. Siess' earlier comment. The sense here is~ .
16 that they have to do something to eliminate it, but wouldn't it 17 also be acceptable to show that it's not a problem for the ,
18 plant? !
19 MR. SHfRON: Correct. Yes.
20 MR. DAVIS: And I think that's the sense that this 21 should be worded, rather than --
22 MR. SHERON: If you recall, I said I already agreed i
)
23 that we would change that wording.
24 MR. DAVIS: Thank you.
25 MR. SHERON: License renewal. One of the things l
\ Heritage Reporting Corporation
! (202) 628-4888
. - _ _ . . . _ _ . _ _ . . _ . . - _ - . - ~ _ _ _ . . _ _ . _ . . . _ _
i
141 1 we're considering right now is that for age-related 2 degradation, the components, we need to know which components 3 we should worry about and which ones don't deserve a lot of our.
4 attention.
5 One way to do it is we should identify the risk 6 significant components. PRA 50 used to do that. PETS, which 7 is probabilistic evaluation of tech, specs. An associated 1 8 program might be a risk. management program deeling with how to 9 continually assess the safety of a plant and included in that 10 might be justifying tech, specs that need to be fixed, changed 11 or what have you.
12 A similar situation might be just on licensing 13 actions. Many times plants come in looking for an emergency
') 14 tech. spec, change. A lot of times we are struggling around 15 looking for a rationale. PRAs coulc. help provide that 16 rationale.
17 And then with the ISAP-II program, Son of 1 SAP, we i
18 didn't want to call it, which is designed to optimize the total i
19 safety and expedite schedules to implement fixes, again a PRA 20 would be useful in that prioritization scheme.
21 MR. LEE: Can you explain a little bit more about 22 what the ISAP-II program does?
23 MR. SHERON: I'm going to turn to Tom Cox.
! 24 MR. LEE: This is something new here that is being 25 accomplished.
Heritage Reporting Corporation (202) 628-4888 l
i i
- 11 .
142
() 1 MR. COX: Tom Cox. The ISAP-II program as designed 2 would be somewhat of a follow-on to'an original ISAP program.in 3 which plants -- it's a voluntary participation kind of program, 4 which, for ISAP-II, the plant owners would produce a PRA that 5 would be reviewed by the staff and with that PRA there would be 6 the opportunity to resolve a number.of existing, outstanding 7 issues for the plants, and the PRA would become a living kind 8 of document or analysis for the plant that would go on, in 9 succeeding years would be able to be updated and would provide 10 a basis for resolving issues as they come up in the future.
11 And issues tha; were prioritized and decided to have 12 some value for resolution as a result of this work would be i 13 scheduled on a plant's living schedule and both the utility and
()- 14 the NRC would participate in determining when-those' issues 15 would be actually implemented in the plant. That is, the 16 resolutions of those issues would be actually implemented.
17 MR. LEE: It sounds almost like IPE. I thought the 18 IPE documents are sort of supposed to be living documents as f
. 19 well.
20 MR. COX: The IPE is actually kind of a one-shot 21 treatment to look for plant unique vulnerabilities. The ISAP L 22 approach would have somewhat of a longer term. It is intended i
23 to have a longer term effect.
24 MR. SHERON: As long as NRC keeps identifying issues, 25 ISAP would continue. Theoretically, you could finish O Heritage Reporting Corporation (202) 628-4888
,- 143
(~)
's / 'l everything up. But I don't think that's likely.
2 DR. KERR: Continue, please.
3 MR. SHERON: The next section of the generic letter, 4 Section 7, addresses the screening criteria. Let me start out S before I go into this and what they specifically are and 6 explain what their purpose is.
7 These criteria are in no way whatsoever related to 8 safety goals or anything else. They ar strictly what I would 9 call a threshold type of criteria which basically tells the 10 industry these are the sequences we want you to put on the 11 table and tell us about and these are the ones, if they don't 12 meet the screening criteria, you don't need to report them, put 13 them in your back pocket.
14 But anything that passos the screening criteria, we 15 want you to identify in your submittal. It does not mean you 16 have to fix anything. It just says identify it. ,
17 These are the things we would like to see. These are 18 the things you ought to be screening for maybe further 19 consideration as to whether or not something needs to be done.
20 But they are in no way related or tied to the safety goals.
l 21 The sequences are those that would contribute more I
22 than 5 percent to the predicted core damage frequency, those 23 for which the core damage frequency is greater than ten to the 24 minus sixth per reactor year -- yes, sir?
25 MR. LEE: The Brookhaven Report, I should have O Heritage Reporting Corporation (202) 628-4888 l
e 144
'( ) 1- memorized the report number, but it's NUREGs ER 4902. Two C times ten to the minus sixth per reactor year one of them has 3 and the in the generic letter it's two times ten to the minus 4_ sixth. Maybe there is a small conflict there.
5 MR. ELTAWDA: In.USCR 4920,-the Brookhaven Report, 6 this is a number that was just a strange criteria that was used 7 on a high sequence. That would be diluted in the report.
8 MR. LEE: Okay. Then I was wrong. The Brookhaven 9 Report has 10 to the minus sixth but the Appendix 2 of the 10 generic letter is two times ten to the minus sixth.
11 MR. ELTAWDA: There is some inconsistency.
12 MR. LEE: The other occasion is the second bullet 13 there, the 5 percent, if a sequence comes to 5 percent to the
( 14 overall core damage frequency, is that what you are implying?
15 MR. SHERON: Yes.
16 MR. LEE: That seems like a perpetual game that one 17 could play. You can eliminate one this 5 percent and then of i
l 18 course you reduce the overall frequency and you end up with the 19 second one that could contribute an extra 5 percent.
20 MR. SHERON: You don't eliminate it from your
- 21 consideration. All we're saying is you don't need to tell us l
22 about it. Do you follow?
23 MR. LEE: Well, I mean there were a lot of i
l 24 implications that once you recognized a risk-dominant sequence 25 you bester do something about it.
l
() Heritage Reporting Corporation (202) 628-4888 l
1
~.]-
145
-1 MR. SHERON: The-utility goes through, okay, let's 2 supposing they_come up with 100 sequences, all of which have 3 some contribution.
4 They add them all up and they get the total _ core melt 5 ' frequency.
6 All we want them to do is tell us of all of those 7 hundred sequences which ones contributed more than 5 percent to 8 that total number?
9_ The one s that ' did not, you don't have to tell us.
10 You put them in your file somewhere, okay, and hang on.to it so 11 you can derive how you got that number.
But in terms of what 12 you come in and report, in your report to the. staff on how you 13 did your IPE, you only tell us about those which contribute O 14 more than 5 percent.
15 MR. LEE: Regardless of the absolute frequency value 16 that you end up with?
17 MR. SHERON: Right.
I 18 MR. LEE: Then somehow I still feel that the point of 19 the generic letter is once you find something that is risk 20 dominant, utilities would como and fix it.
21 MR. SHERON: It's up to the ut!.lities.
22 MR. LEE: So they don't have to do that? You don't 23 expect that they would fix anything?
24 MR. SHERON: No, I would hope they would fix the 25 stuff that they think needs to be fixed.
O Heritage Reporting Corporation (202) 628-4888
A 146
, 1 MR. LEE: So if they fix it, then you are left with 2 residual core damage frequency, which would be considerably 3 lower than what you started off with, but there's another one 4 that comes to. 5 percent to the residual core damage. frequency.
5 MR. SHERON: If they make the fix as part'of their 6 identification, then fine, they come in, they say we evaluate 7 our plant, here are the things that we find, and maybe then 8 what they are going to say is and here is the stuff we fix and
'9 when we fix it, here's what the new answer is.
10 Another way they might do it is to go through and do 11 their IPEM or their PRA and they'll say gee, we better fix 12 these four things. And they fix them. And then they submit 13 those results with the stuff that's fixed.
i V 14 MR. LEE: And the results are one sequence that will 15 count for 5 percent or more. Then what happens?
16 MR. ELTAWDA: That's nothing. It's up to the 11 utility. It's only to record these sequences. Any action is 18 up to the utility completely, if they want to do anything about 19 the sequence or not.
20 MR. LEE: What I would like to see is somehow an 21 intersection between the second criterion and the third i 22 criterion of some kind.
23 MR. ELTAWDA: The reason for the second criterion is 24 there are some methods out there that split the sequences so 25 find that they don't have any dominant sequence and they can O IIeritage Reporting Corporation (202) 628-4888
1 all be ten to the minus sixth or something like that. So the 2 can skip.this criterior. So to prevent that you want them to 3 report other_ sequences even if you are low, because of the 4 method that you are adopting. Tell us which one is going to be ;
5- 5 percent. You can split the sequences so fine that every'one 6 of them is ten to the minus sixth.
7 -MR. LEE: I understand that point, but if for that 8 particular dominant sequence, five times ten to the minus 9 seventh, but you are still expecting them to analyze the 10 sequence. ,
11 MR. ELTAWDA: They have already analyzed it.
12 MR. SHERON: They've already analyzed it and they 13 have come up with five times ten to the minua seventh. All O 14 we're saying is that on this criteria, they would not have to 15 report it but if that five times ten to the minus seventh 16 was great than 5 percent of the total frequency, they would 17 have to report it. they don't have to do anything with it.
18 All we're saying is tell us, put it in your report, tell us.
19 MR. LEE: But the word "include" means in the 20 analysis you have to start counting those sequences somehow.
l 21 MR. SHERON: You are misinterpreting the word 22 "include." Include in the report. That's what that means.
23 Include in your report. Not include in your PRA or in your I 24 IPEM.
t 25 DR. KERR: I think, John, he is saying that one has l ( Corporation l Heritage Reporting l (202) 628-4888
- - - - n
149 ss 1 to analyze all' sequences.
2 When you only include in your report those that 3 eithe r are' bigger than or contr:1.Ue more than ten to the minus 4 sixthper reactor year in core melt or core damage or those
.5 that contribute more than 5 percent. -But you.have.to analyze a 6 much bigger number than that.
7 MR. LEE: I'm still -- I'm confused by the words 8 screening sequence selection criteria, it says. I interpreted
.9 it'only those sequences need be selected and analyzed. There 10 is no mention of the report requirement.
11 DR. KERR: No, I think a much larger number has to be 12 analyzed. In fact, there's no guidance given on how large that 13 number is. '
'O-14 MR. SHERON: They might have a thousand, I don't 15 know, a hundred. All we are saying is that once you have 16 analyzed all your sequences, the only or.es we want you to tell 17 us about are the ones that fall above these two criteria or 18- these three or four criteria here.
19 MR. LEE: Thank you.
20 MR. DAVIS: Brian, excuse me. I have a related 21 question.
22 I thought I heard you say that this screening 23 criteria has no relationship to and is not inspired by the 24 safety goals.
25 MR. SHERON: Correct.
l
() Heritage Reporting Corporation (202) 628-4888
9 149
\/ 1 MR. DAVIS: But the first paragraph of Appendix 2 2 says that these screening values shall provide information on 3 plant characteristics and behavior that can be used to 4 determine consistency with the Commission's safety goal 5 policies.
6 MR. SHERON: Let me tel you how -- when I get a 7 little bit further in here I'll tell you how we're going to use 8 safety goal and you'll see how these numbers would be used.
9 MR. DAVIS: Okay.
10 MR. SHERON: Okay? Not the screening numbers, 11 though. But I'll tell you how we're going to use safety goal.
12 MR. DAVIS: This working perhaps then may need to be
,, 13 looked at. Because you say you are going to use the screening k_ 14 goals to --
15 MR. ELTAWDA: The staff is going to look at these 16 sequences to see how we as a staff can reach conclusions on the 17 consistency of the industry with the safety goals. It's not 18 for the utility to go and say I comply with the safety goal in 19 that sequence or not.
20 DR. KERR: That will mean that the staff is going to 21 look at the bottom line and do something or other with it.
22 MR. SHERON: Possibly, yes. Yes. I'll get into that 23 a little bit later.
24 DR. KERR: Mr. Shewmon?
25 MR. SHEWMON: That next to the last bullet is the way
('T U
Heritage Reporting Corporation (202) 628-4888
150
,1
) I the report, the Appendix 2 is going to read, or'the comparable 2 item in Appendix 2. Because that tells'when 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> is' 3' starting.
4 MR. SHERON: No, we are going to change that a little 5 bit, okay. And the reason is that there are'some containments 6 for which 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> is not an appropriate number.
7 MR. SHEWMON: Well, 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> there is better than in 8 the report because it doesn't say when the clock starts in the 9 report. I thought it was maybe when the rods went in but now 10 you tell me it's when the vessel penetrates.
11 MR. SHERON: Well, then, containments fail in less 12 than 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />.
13 We'.7 not saying the sequences that are greater than
() 14 ten to the minus sixth --
15 MR. SHEWMON: My point is an editorial one.
16 MR. SHERON: Okay.
17 MR. SHEWMON: If 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> is then the time the vessel 18 is penetrated, that defines it, whether it's the right or wrong 19 answer, and it defines it better than Appendix 2 does with the 20 draft we have. Go on. Forget about it. Somebody else in the 21 room understands it.
22 MR. ELTAWDA: As Brian said, we are changing that 23 criteria.
24 DR. KERR: You are changing that criteria? What will 25 it read?
U Heritage Reporting Corporation (202) 628-4888 L
, _. . _ _ _. ~ . _ _
151 1 MR. SHERON: Well, for a BWR it may be more 2 appropriate to use three hours, for example.
3 What Sn! are trying to distinguish here is between the 4 short, the early containment failure where you would have a 5 -large release and the late containment failure where you may 6 have had a lot of settling and played out stuff and the release 7 is not very large, comparatively speaking.
8 And for a BRW, that, you know, a failure after three
, 9 hours may result in a much lower release than if you had a 10 failure before three hours.
And I think what we're going to 11 propose is to change the words here so that we recognize 12 differences in containment performance, and not just lump 13 everybody into this 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />.
O 14 DR. KERR The assumption being that even though the 15 containment has failed the water will, wherever it is, will 16 still retain a lot of fission products?
17 MR. SHERON: For some instances, yes, it may be the 18 water that retains fission products. In other cases it just i
19 may be that the stuff settles out and everything, like in a
! 20 large dryer or something.
, 21 MR. SHEWMON: Another way to do this, as you know at 22 least as well as I do, is 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> from when the rods go in, 23 figuring there's a lot more activity in the first 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> than 24 there is after the 12th day.
25 But you've got it, it could be two weeks after the Heritage Reporting Corporation l (202) 628-4888
_ _~ . _. _ ._ - .- - - _ - . _ . . _
i 152 1 rods went in but you still want to know from when it came-2 .through the vessel.
3 MR. SHERON: Yes, that's what we're talking about,'12 ,
4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> from the vessel penetration,'since that's when the 5 fission products ultimately get into the containment. Dr. f
'6 Siess.
7 DR. SIESS: I apologize if I didn't hear the answer-8 earlier, but is it important that we know the difference 9 between a function sequence and a systemic sequence?
10 MR. SHERON: I think it is. We put a definition in 11 there. Farouk, do you want to -- ?
12 DR. SIESS: I read the definition. Could you give me [
13 a definition that I could more easily understand than the one 14 that's in there?
15 I mean, don't repeat that one, because I didn't 16 understand it.
- 17 MR. ELTAWDA
- How about if I give you an example?
18 DR. SIESS: That would help.
19 MR. ELTAWDA: There are some PRAs that look for 20 example at the sequence and split that at the sequence into, 21 that was due to failure of the insertion and they call that a 22 sequence, that was due to failure of the system and that's a 23 sequence. So these are systemic sequences. The failure of the 24 system itself. What we are saying is that you have to combino 25 all the sequences that lead to a loss of the function which is O Heritage Reporting Corporation (202) 628-4888
153 (m) 1 the failure to scram.
2 DR. SIESS: Thank you.
3 DR. KERR: That is so the sequences will be bigger or 4 easier to handle or what consequence do you base that on?
5 MR. ELTANDA: The consequence is that it is a 6 sequence.
7 MR. SHERON: The remaining one is also to identify 8 sequences which lead to containment bypass and have a 9 probability of occurrence greater than 10 to the minus seventh.
10 MR. LEE: Regardless of whether the sequence 11 constitutes 5 percent or more to the core damage frequency, the 12 back end analysis will be retained?
13 MR. SHERON: Yes.
(m) 14 MR. LEE: I guess that was my question. So the back 15 end analysis would be the same. You have to go through the 16 same kind of analysis.
17 MR. SHERON: Yes. But not for every sequence. We 18 noted if you read in Appendix 1 it says that even though you 19 may have a large number of sequences that lead to core melt, 20 they may all manifest themselves in just a few different 21 containment analyses.
22 MR. LEE: Those that contribute to 5 percent or more 23 will be analyzed in more detail in the back end?
24 MR. SHERON: No. Just the 5 percent to core damage, 25 not to containment.
Heritage Reporting Corporation (202) 628-4888
e
\
d 154-() . 1 MR. LEE: Right. That's the point that I was trying 2 to understand.
3 MR. SHERON: Yes. Right.
4 MR. LEE: So after you identify-that kind of 5 sequence, then what do you do?
6 MR. SHERON: Well, they have to look now at the 7 containment performance, but again it's this matter of lumping 8 the sequences and doing just, you know, a handful of 9 containment performance, and I think we said look tables.
10 MR. LEE: So the back end sequence can be different? ,
11 MR. SHERON: Yes.
12 MR. LEE: Depending upon whether a typical sequence 13 meets this criteria or not?
() 14 MR. SHERON: Yes.
- 15 MR. LEE
- Thank you. ;
4 16 MR. WYLIE: Let me ask a question just for
- 17 clarification on your sequence for core damage frequency. The I 18 appendix I'm looking at says it's two times ten to the minus 19 sixth, and not one time. Is that correct? l i 20 MR. SHERON
- Yes, because that was probably an 21 earlier number and that's going to be corrected to be 22 consistent.
23 MR. WYLIE: The wording, you have it greater than, :
24 and here it says equal to or greater than. Is that correct?
I 25 MR. SHERON: Equal to or greater than is the correct i
l i
O- Heritage Reporting Corporation (202) 628-4888 !
k
9 155 1 wording. We_will correct that.
2 DR. KERR: Various spots, an analyst is admonished to
'3 consider uncertainties. ,
4 How does he. consider uncertainties in making these 5 selections?
6 MR. ELTAWDA: Hopefully this-selection, we have done 7 the analysis at that time and studied it with the important 8 paramecers --
9 DR. KERR Let me be more specific then, hecause he 10 knew the estimate value, the so-called 95 percent confidence 11 value, whatever that means, or none of the above.
12 MR. ELTAWDA: In the Appendix 2, it means the mean 13 value.
14 DR. KERR The screening numbers are the mean value?
15 MR. ELTAWDA: Yes.
16 DR. KERR Jmd that is independently of what the !
1 17 uncertainty is? The mean value has uncertainty in it? You got 18 me there.
19 MR. SHERON: It's not exclusively considering the r
- 20 uncertainty. I think it's to input it in a number.
21 DR. KERR: But here is the mean value independently 22 of the distribution?
23 MR. SHERON: Yes.
24 MR. CORRADINI: I think the mean is weighted.
i 25 DR. KERR: A weighted mean? I'm not going to pursue 1
() Heritage Reporting Corporation (202) 628-4888
)
156 1 this any further. It's a mean value.
2 MR. SHERON: Yes.
3 DR. KERR: Thank you.
4 (Continued on the next page.)
5 6
7 8
9 10 11 12 13 14 15 16 17 18 19 20 l 21 r
22 23 24 25 I
Heritage Reporting Corporation (202) 628-4888 i
4 4
157 my s/ 1 MR. SHERON: Okay. .Very quickly, let me go through 2- what we see as the role of severe accident management in th'e
- 3. IPE process here.
4 Accident management is -- at least in our definition 5 -- a process in which actions that can prevent core damage or 6 mitigate the. consequences of a severe accident are identified, 7 evaluated and incorporated into a structured program. Note I 8 didn't say procedures, but structured program.
9 Implemented at a plant site, and would be available 10 to either the operators or the plant management in the event of 11 a severe accident.
12 And accident management is rather broad in our 13 perspective, and encompasses hard work, human and O' 14 organizational factors, and hopefully it will provide the i
15 decision makers at the plant a structured program for managing 16 an accident.
17 In other words, you don't want to get into a seat of 18 your pants operation if you had a severe accident where you are 19 doing a bunch of back of the envelope calculations to determine 20 what kind of decisions you have to make, what pressures should 21 I vent, the containment if I'm going to vent it, who makes that i
22 decision.
23 What factors have to be considered before I make that
, 24 decision, who needs to be notified.
25 DR. KERR: Is this discussion from a perspective of
(:) Heritage Reporting Corporation (202) 628-4888
1 1
1 158 l l ()' 1 Appendix 3 or something'else? Because this third paragraph in 2 Appendix 3 says nothing is required ~at this time.
3 MR. SHERON: That's right.
4 DR. KERR: What does that mean?
5 MR. SHERON: Okay. This is basic.- The simplest term 6 I can use is.this is a warning shot, okay. And that is, that 7 the staff--
8 DR. KERR: I'm not trying be coy, Brian.
9 MR. SHERON: I know.
10 DR. KERR: It says nothing is required at this time, 11 and I've got all these other requirements. What are--
12 MR. SHERON: All right, that's what I'm trying to 13 explain.
() 14 MR. WARD: An advanced warning.
15 MR. SHERON: We are going to be pursuing with the 16 industry a need for them to establish accident management 17 programs at their plant.
18 We originally wanted to establish these as part of 19 the IPE letter. The industry again was vehemently opposed to 20 doing that, because they said you are putting more junk into 21 this IPE again.
22 And don't tie the IPE program or the completion of an 23 IPE to having accident management. We agreed. We said fine, 24 we will not tie it to the IPE but you have to recognize that 25 the staff is going to be pursuing this and eventually you know, Heritage Reporting Corporation (202) 628-4888
159 I we won't be requiring you to implement accident management, 2 severe accident management programs at your plant, sometime 3 down the future.
4 And you should understand that number one, everything 5 that comes out of this IPE is probably going to be the biggest 6 or probably form the majority of the information you need to 7 develop an accident management program.
8 As I said before, one of the biggest things we have
'I 9 is that when one does an IPE one will be making assumptions 10 that the operators do good things or take certain actions.
11 How do we know the vparator is going to do that. Yes 12 sir?
13 DR. SIESS: Since you mention that, why use it more
-O 14 important that the operators not take specific action that 15 would have adverse effects than that the operators take 16 beneficial actions?
17 MR. SHERON: We changed that word to I believe to 18 eliminate what, more important?
19 DR. KERR More important.
20 DR. SIESS: Thank you. You kney, it seems to me that 21 for something that is not required now, it sould have been 22 better to put that comment somewhere up front rather than in 23 the third paragraph of an appendix.
24 DR. KERR: I would certainly urge that you just leave 25 that out.
Heritage Reporting Corporation (202) 628-4888
=- -
e A b i
160-(). 'l MR. SHERON: What is it?
2 DR. KERR: The appendix 3. If you wanted to publish 3 it, publish it somewhere else. If it is not part of the IPEM, 4 why put it in?
5 MR. SHERON: Well the same reason that says if we 6 don't, one of the things that we are going to add to this 7 letter by the way in external events, is a statement that says 8 somewhere'down the road, you are going to be acquired to 9 address external events.
10 And one of the considerations was is that well gee, 11 if you are not asking them to do external events now, why tell 12 them about it, okay. Why muddy it up.
13 But we decided to do it because if you talk to a PRA 14 analysis they will say, gee, if you do a seismic PRA, you may 15 not want to do your fault trees, event trees and the like the 16 same way.
17 That if you were going to include an external event 18 PRA in the future, because you wouldn't collapse, say if you 19 used an inlet and an outlet valve to a pump, you may want them 20 as one valve in the PRA. Whereas you wouldn't do it.
21 DR. KERR: I understand that while I think I consider 22 it valid, I don't see what it has to do with this.
23 MR. SHERON: What we are saying is that the results 24 that they get from the IPE are going to be identifying the 25 areas where the operators are being assumed to do things and O Heritage Reporting Corporation (202) 628-4888
s 161 a l' the like.
'2- DR. KERR This may be, Brian.
3- MR. SHERON: Okay.
=4 DR. KERR: But the' fact that you're going to have'to 5 do an' accident management scheme later on is not going to 6 influence this IPE.
7 It will be the same IPE whether you have to do an 8 accident management scheme later on or-not I would assume, 9 because this is an effort to get at the risk contributors.
10 And the accident. management is something separate.
11 Whether you do that or not, won't influence or 12 shouldn't influence the IPE.
13 MR. SHERON: We are trying to tell them that 14 although, you know if we tell them accident management is 15 coming later, okay.
16 Then one thing that a utility might do is say, well I 17 won't do anything to enhance the operational procedures let's 18 say at this time because I don't know what NRC has in store for 19 me for this down in the future, but I'm not going to do 20 anything now. And we don't want to encourage them that way, to 21 do that.
22 DR. KERR: But you don't tell them what you are going 23 to do later on anyway, so they are still in that position. You 24 just tell them you are going to do something.
l 25 MR. SHERON: That's right. But I think if we know 1
(
Heritage Reporting Corporation (202) 628-4888
162
() 1 1 roughly what it is we are thinking about, we owe it to them to
=2 just tell them what our thinking is at this time. And that is 3 all this doen.
4 MR. LEE: But it is somewhat contradictory also, 5 reads appendix 3 to the effect that nothing is required now, 6 and then in the section 9, like with page 9 and 11 of the, 7 that's a proper, you are saying utility should keep this 8 information report to us.
9 MR. SHERON: Page 12.
10 MR. LEE: The documentation submitted to the NRC 11 should include this information together with the steps the 12 licensee, that they can insure that the personnel are properly 13 trained, and so on.
() 14 So that implies that there is certain documentation 15 required. So you ought to require the utilities to do 16 something about accident management.
17 MR. SHERON: All we are saying is that if they take 18 credit for an operator doing something, we'd like to know what l
19 it is that gives us the assurance that the operator can indeed 20 do that.
21 But there are other things, with regard to accident 22 management besides just an operator taking a certain step or 23 something, it is a much bigger--
24 MR. LEE: I am not disputing that point, I am just 25 trying to clarify the statement between, the statement on page
( Heritage Reporting Corporation (202) 628-4888 i
t 163 f]~
N 1 -12 and the statement in appendix 3.
.2- The first phrase of the third paragraph says although 3 nothing.is required at this time. But that is contradictory to ,
4 what is being said in the part that--
5 MR. SHERON: I disagree, Phil.
6 DR. KERR: I think that sentence means nothing is 7 required in accident management at this time. It doesn't mean -
8 that utilities aren't required to do anything,.they just aren't 9 required to establish an accident management program. Isn't
, 10 the sense of it?
11 MR. SHERON: That's right.
12 MR. LEE: I'd like to suggest that you clarify it for .
13 us. :
O 14 DR. SIESS: It seems to me there is some kind of r 15 distinction you are making between the things that they would 3 16 assume in the analysis as being done, and additional things 17 that might be done, and also possibly a distinction between 18 operator actions and hardware changes, as a part of accident 19 management. Is that the distinction you are making here? I 20 MR. SHERON: That's, I think that is fairly close to j 21 it.
22 DR. SIESS: It seems to me you could say it a heck of 23 a lot better. And you know, back where you talk about external 24 events, you have a note that says deferred item.
1 25 And I come up to accident management and I don't see !
E C)
Heritage Reporting Corporation (202) 628-4888 ,
f I
164
(-
(_) 1 any deferred item note by it, and I have to get over into the 2 appendix to find out that it is a deferred item.
3 MR. SHERON: We can probably end there. Do we have 4 one?
5 MR. ELTAWDA: I hardly think deferred, that the 6 utilities are meant to for the IPE or the procedure and the 7 accident management -- what we are asking utilities if they 8 discover anything that they found that is important for use, 9 there is from the plant, go ahead -- I meant it. But don't 10 submit all your procedures at this time. That's all what we 11 are trying to say.
12 DR. SIESS: You are not saying it very well.
13 MR. ELTAWDA: Okay we'll try to clear that up.
b~d 14 DR. SIESS: And you're saying it right now a heck of 15 a lot better than somebody said it in writing.
16 MR. SHERON: Okay, well let us take a look at the 17 words and see if we can clean that up a little bit. But the 18 intention was that if you find something that is boaeficial to 19 do as a result of your IPE please go ahead and do it and don't 20 wait because you think the staff is going to come down and make 21 you change it all later on or something.
22 We have a commitment right now to work with NUMARC to 23 define the scope of accident management programs and to also 24 define the best way that we can get these things properly 25 implemented at the plants.
O O Heritage Reporting Corporation !
(202) 628-4888
165 -
4 1 NUMARC has established a committee c. ave e 2 accidents and they told us this is their T hat " er of 3 business. And so we'll be working with th'. in the near future 4 to try and put this accident management stuff into place.
5 Let me quickly touch on the severe accident 6 prevention and mitigation features report which is New Reg CR 7 49.20. I think you've heard this before.
8 These reports were prepared by B & L. In essence 9 what they do is they try and summarize and distill insights and 10 information that has been gained from all different various 11 sorts of PRE's and so forth from 1150, from PRES that were 4
12 submitted, from the IDCORE IPEMs etc., into one concise set of i 13 documents.
14 There is a document for each containment type. Again 15 they highlight plant features and operator actions that have 16 been found in the past to be important to rink.
17 And these, I won't dwell on these, these just show 18 some of the things that were found. .One of the things that 19 came out of the CRGR review and I guess the staff has agreed j
20 that we will not include these reports as part of the letter i
21 properly.
l 22 But rather they will be referenced. There was a 23 concern that people may read the insights in there as being, if
!. 24 you call it staff guidance the utilities may say well gee, l
25 that's a requirement.
1 Heritage Reporting Corporation (202) 628-4888
s; ,
i I
l 166
- 1 And even though we can deny it, and claim that there :
-2 was no intention to make these reports requirements or 3 anything, but.merely as guidance or help to the industry, it 4 was_ felt that to eliminate any confusion on the part of the 5 industry, we would not include these as part of the letter, but !
6 rather reference them and make them available to the industry- ;
'7 as separate documents. ;
8 DR. KERR: Does the staff endorse the suggestions i 9 that are contained in those reports? If somebody came in and i
10 said we are going to do this because it's in the recs, CR 11 .49.20. i 12 MR. SHERON: I don't think we were in a position to !
13 say we endorse. Some of them we may feel are beneficial, but i O 14 all we were doing was reporting insights that were gained from 15 dther PRA's.
16 DR. KERR: Well an insight to me means an a
17 understanding. To you it just means an ldentification.
l l 18 MR. SHERON: So one should not interpret anything in 19 these reports as saying this will be beneficial. All it's
! 20 saying is look at this, and decide whether you think it's 4
21 beneficial or not.
22 And say that this is what other people have learned I 23 and identified in their PRA's or the like.
24 DR. KERR: But we don't know--
, 25 MR. SHERON: What they have concluded.
4 Heritage Reporting Corporation (202) 628-4888 4
--c -
..-,,,..,,-._,,,_.n_ , _ _ _ _ _ _ _ _ _ , _ , , _ _ , _
167 ,
1 DR. KERR: But we don't know whether it is any good
<2 or no't.
3 MR. SHERON: I don't want to get in the business of 4 telling the utilities how to fix their plants, that is my 5 problem.
6 DR. KERR I am not being critical, I am trying to ,
7 understand how. 7 8 DR. SIESS: We are not endorsing, we are just saying l 9 that this is what other people have found, and you know you 10 have to look at it, and, yes? !
11 DR. KERR: As long as everybody is drawn to 12 understand the ground rules, I think that it's perfectly 13 reasonable.
O 14 MR. SHERON: Well that is one of the reasons we ;
15 pulled it out of the letter proper. To make sure it wasn't 16 misinterpreted as some staff recommendation. ;
17 The last item is the documentation, I'm sorry, not l 18 the last item. What we would like in the documentation is to !
19 find a basis for the findings that the utility draws in a ;
1 20 traceable manner.
- 21 We certainly envision that perhaps it is part of I
22 looking at their results that they submit. We may have to ;
23 interact further with the utility to understand some of the !
24 conclusions that they drew, and the like. ;
- i j 25 It may involve having meetings with them out at their l
() Heritage Reporting Corporation
- (202) 628-4888 i
- _ _ _ ._ _ _ _.-_, _ .-_- . _ _._ _ _ _ _ _ . _ _ _____w
+ . .- . .- - . .
168 1 plant. site, and certainly hopefully will be able to go back and ,
2 find the detailed-information that was used to support their 3 conclusions.that were reported to the staff._ ,
4 DR. KERR: Now earlier, someone asked you a question- ,
5 and someone responded, perhaps not you. That the IPEM was -
6 meant to be a snapshot. That's how the plant is now, and it !
7 may not be like'that three years from now in fact.- ,
8 And yet as I read this, the licensee is expected to 9 keep some of this stuff for the life of the plant. How come?
10 MR. SHERON: Or until the staff I think says they 11 don't have to. .
i 12 DR. 1; ERR: But if it is reasonable to suppose that 13 the thing may be out of date in five years.
( 14 MR. ELTAWDA: We are changing the list in this area, l
15 to see'if the list we will see it in five, if it is pertinent !
16 information. So if it is outdated, then we may have more i 17 information to update it for IPE. [
18 DR. KERR: Oh wait a mintite. You mean if they are 19 going to be required to update it. [
20 MR. ELTAWDA: No, there are no requirements. That is i
l 21 preceded by other, superceded by other information.
! 22 IIR . SHERON: The thing is we didn't want to get 23 ourselves to say well, keep it for five years and then throw it l 24 away.
4 25 And if we hadn't finished our review, for whatever
(
l Heritage Reporting Corporation I (202) 628-4888 4
t k
i
~
.g i
- 169
) 1. reason, and a lot'of it is beyond our control, like resources, 2 staff resources.
3 One didn't want to get in the business of then doing 4 a review and finding out that there is--
5 DR. KERR:- What if you review something that is five 6 years out of date. LSurely you aren't going to'do that.- Well, 7- it's irrelevant.
8 I don't see the point of keeping something for.the 9 life of the plant that.is more than likely to be outdated in 10 five years.
11 MR. SHERON: No, we discussed this at length with-12 CRGR and the like, and this was, seemed to be an acceptable 13 compromise. That they should keep it then unless it is 14 superceded.
15 MR. SHEWMON: You would see a good reason for keeping 16 it for five years, if it is only valid for five years, though, 17 I assume, Dr. Kerr, 18 DR. KERR: I haven't looked at that question. I 19 would want to give it serious consideration before answering it 20 though.
21 MR. CATTON: Isn't this supposed to be a living 22 document? Which we'd say would keep updating.
23 DR. KERR: No, it is not a living document.
24 MR. CATTON: No?
25 DR. KERR: Nope. They don't expect it, they don't O Heritage Reporting Corporation (202) 628-4888
=_
1.70
,n.
k.._) 1 expect it to be updated.
2 MR. CATTON: I guess I share the view of why do it 3 then.
4 DR. KERR: If CRGR said it was the thing to do, I 5 would be reluctant to disagree.
6 MR. CATTON: Okay.
7 MR. SHERON: I don't think the commission's policy 8 statemenz implied thet it was st.pposed to be a living document 9 either, see.
10 (Continued on the following page.)
11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 O Heritage Reporting Corporation (202) 628-4888
171
,s
(-) 1 MR. SHERON: There should be two tiers of 2 documentation. The first one would be the reporting of the 3 results of the examinations, NRC.
4 This would be telling us the leading sequences, and 5 you know, the ones that pass the screening criteria -- core 6 damage frequency and the likelihood of large release.
7 MR. CORRADINI Can I ask a question?
8 MR. SHERON: Yes.
9 MR. CORRADINI: So kind of how tfils works is they get 10 the letter, and the work is started, and then they report back 11 based on the screen criteria you went over -- everything that 12 includes five percent or above, per mal frequency (ph), plus 13 all of the other caveats; plus all of the other inclusions 14 below.
l 15 And then they give you the core damage frequency 16 cumulative? The second line is -- for all sequences? Core 17 damage frequency for all frequencies? I'm sorry. I don't know 18 what that means.
19 MR. LEE: Sum total, you mean.
20 MR. CORRADINI: Sum total? And then the likelihood l
21 of large release. What's a large release?
22 MR. ELTANDA: We will be able to define at the 23 workshop. The Commission is working on the definition of the 24 large release, and you'll have this information during the 25 workshop.
Heritage Reporting Corporation (202) 628-4888 l
[
172 1 MR. SHERON We tried a bunch of different ones, 2 okay? We tried to substitute containment failure in there.
l 3 You'know, and then some -- the Commission.uses the term; and
-4 we're struggling right now to come up with a definition.
5 And we're presuming that by the time the letter goes 6 out to the industries ready to start this, we'll have that 7 definition for them.
8 MR. CORRADINI: Okay.
9 MR. SHCRON: And then the main contributors to the
- 10 core damage. frequency containment failure. And then lastly, a 11- disposition of contributors.
12 And by that I mean that they may come in with 30 13 sequences that pass the screening criteria that they have to 14 document.
15 We're not telling them that they have to do a cost-16 benefit analysis, and tell us why they did or didn't fix each 17 of those.
18 What we're saying is, just tell us which ones you 19 propose to fix.
20 MR. CORRADINI Question there.
21 Why would they want to fi;: any?
22 MR. SHERON: They may not. They may say, I don't 23 have to fix any. But we're saying that they may come in and 24 say for all of them, we've decided that we don't have to fix 2 'i any. Good-bye, good look, okay?
O Heritage Reporting Corporation (202) 628-4888
j t
. 173 r 1 The other thing is they may come in and say we've j 2 decided we've looked at these four, and these four we're going 3 to do the problem. Okay, for this one we're going to add some 4 procedures over here. For this one we're-going to change over 5 a control system on a valve or something like that. I don't 6 know.
7 MR. CORRADINI: Okay. Good. I guess the reason I'm 8 asking this is, again, historically, the picture I had was that 9 there was some envelope that was going to be drawn. I don't 10 know how.
11 But let's say it was drawn. And then wo look for 12 things that poke out of the envelope. -In this case, the screen 13 criteria really isn't the envelope. It's just an arbitrary O 14 criteria, and you're kind of leaving it to the utilities to 15 come back and say whether they choose to or choose not to do 16 anything when they identify these things that pass your 17 screening criteria.
18 DR. KERR: In that same connection, I think, on page 19 9 of something or another, it is expected that the licensee 20 will move expeditiously to correct any identified 21 vulnerabilities that were determined to be important to 22 correct.
23 The dctormination is made by the utility. And moving 24 expeditiously means they tell you what they plan to do, and 25 then you review it before they can do it. Is that right?
O Heritage Reporting Corporation (202) 628-4888
174
'~
,3.
end__21 'l MR. SHERON: If it trips the 50.59 critbethlathey 2 have --
' :) DR. KERR But they're in a region of great unknowns, 4 right?
5 MR. SHERON: I understand. But if they change 6' something that affects the design basis, the licensing design 7 basis,.they have to tell us.
8 DR. KERR: I'm not trying to argue with what you're 9 telling them. I'm trying to understand what you're telling 10 them. I'll wait to argue after I understand, maybe.- So moving 11 expeditiously doesn't mean you make'a change in equipment, it 12 means you analyze this, submit something to the staff and say 13 we proposed to do this, we have determined that it either does 14 or does not pass the 50.59 criteria and then wait for staff 15 approval before doing anything.
16 MR. SHERON: They don't have to tell us. If it 17 doesn't pass the 50.59 criteria, they don't even have to tell 18 us. They just fix it.
19 DR. KERR: So I would guess again in uncharted waters 20 like this they will not be sure and in order to be on the safe 21 side and avoid fines, which I find occur more and more i
22 frequently these days, they will submit almost anything they 23 plan to do to you.
l 24 HR. SHERON: They might. I can't spot them.
! 25 DR. KERR: This is not something that falls clearly O Heritage Reporting Corporation (202) 628-4888
175
() 1 into design basis accidents and stuff so I don't see that it's 2 obvious whether it's an unresolved safety issue or not.
3 MR. CORRADINI I agree. I don't think it's going to 4 be obvious enough what to do.
5 MR. SHERON: Susquehanna went through and did an IPE 6 and made numerous changes to their plant and I don't believe 7 that they came in and begged our approval of everything they 8 did.
9 MR. CORRADINI: So you mean the minor ones they just 10 will go and do. So there may not be a problem.
11 MR. SHERON: I mean, just a simple example would be 12 if they found out that an HPI pump was not reliabl,e enough and 13 so they decided that they had to change it out. Suppose the
() 14 new pump they used had a much higher reliability but a 15 different head versus flow curve. They might have to analyze 16 the ECCS which would be a 50.50 trigger.
17 MR. DAVIS: Brian?
18 MR. SHERON: Yes.
19 MR. DAVIS: A comment on that. I have a bit of 20 concern that without doing the external event, it may be 21 difficuJt to define what kind of improvements are most 22 beneficial, and you will not have the answer to the question, 23 should I qualify my improvement to some seismic margin or 24 something like that. I don't know if this will be a big 25 problem or not but it's certainly something that the staff O Heritage Reporting Corporation (202) 628-4888
176 1 maybe ought to consider.
-2 I could even envision adding a new system that is not i 3 seismically qualified that would increase plant risk from 14 external events because it could impact another system that you ,
'5 need for external events or something like that.
6' MR. SHERON: With regard to seismic, the staff if l 7 going right now in a direction towards seismic margins which 8 will not give you a bottom line number. And-I think the 9 utility, if they make a decision that they want to add a piece 10 of equipment, they can make the judgment right then and there 11 for what seismic criteria they might want.to design that, 12 depending upon how they envision its ability to reduce risk or 13 its vulnerability to seismic events.
() 14 But I agree. If it comes to a question where they 15 say I really can't decide until I do my seismic work, maybe we 16 just say fine, just hold off until you do it.
17 MR. CORRADINI But you don't think that will be a 18 big, you don't think a good majority of this will be a problem, 19 that the external events are not considered at this time? I 20 guess my question was going to be duplication of effort. How 21 much of this has to be duplicated when external events finally 22 come around?
23 MR.~SHERON: That's why I started to say before that 24 we're trying to put a warning in there for utilities that when 25 they set up their event trees and the like, again, I guess the C) Heritage Reporting Corporation (202) 628-4888
l 177 i
() 1 example we use is that if there's a pump with an inlet and ,
f
- 2 outlet valve, you may for simplicity first lump those two i
3 valves inte one with one failure probability whereas one valve
.4 may be way up high and one valve may be way down low in a 5 plant, and when you do your seismic review you want to keep i 4-
?
6 them separate.
7 And so we're saying be careful because we don't know 8 where we're coming from right now on external events, whether 9 it's seismic margin, but it may be seismic PRA, you want to 10 keep in your fault trees and stuff the two valves separate and 11 you don't want to collapse them right away.
12 So we're trying to give them that guidance that says, I
13 you know, so we don't duplicate effort down the road and that
() 14 they can go back into their internal events PRA and add the 15 seismic without any major grief.
16 Let's see. The TIER-2 documents, these are basically P
17 all the stuff that the utility uses to develop the documents ,
i 18 that would be submitted. We wouldn't want to see that. We 19 would expect they would retain it in their files.
20 We would only be evaluating the first tier reports, 21 the stuff that is submitted, and like I said, if we needed to 22 meet with the licensee, and the like, we might have to go out 23 to the plant or something and we would expect that they could 24 produce some of the basis documents that supported their 25 submittals. ,
i Heritage Reporting Corporation
, (202) 628-4888 ,
t i
i
178
() 1 MR. LEE: Question,
- 2 MR. SHERONa~ Yes.
3 MR. LEET You also.had, as part of the documentation 4 requirement, some kind of independent field review. Could you 5 comment on that a little bit?
6 MR. ELTAWDAt: We are changing the peer review over 7 to, as we say in the generic letter, an independent group in 8 the utility, rather than a peer. review. Some other people in 9 the utility that were not involved in the PRA can review it, i
10 But we are not going for a peer review.
11 MR. LEE: Thank you.
12 MR. SHERON: Review of the results. As I said 13 before, since this is sort of our first time out with the IPE
() 14 and we're a little bit on a learning curve in terms of we 15 really haven't done an IPE review other than what we've seen 16 from the IDCOR submittals way back during the preparation of 17 their methods.
18 What we plan to do, and we've already started doing 19 this by the way, is to prepare a review document which would be 4
20 akin to a standard review plan, for the staff and the 21 contractor reviewers.
22 And it would include the areas of review they should 23 look at, determination of adequacy of IPE results and I'm not a
24 going to tell you what they are. because I don't know yet, 4
25 action levels, and what I mean by action levels is, what I l Heritage Reporting Corporation
, (202) 628-4888 1
1
-. .-.-_.c .. __ . _ _ , _ , , _ _ _ . , .
^! 4 4
1 179 h( ) 1 envision is, if I have a contractor or even a staff person 2 looking at this, they're going to be looking to make sure 3 things look reasonable and make sense and they're not making 4 some wild assumptions regarding for example operator action or 1
5 something like that and so forth, i
6 If we see something that looks out of line that might 7 be an action level, in which case, when I say action it means 8 we may call up the utility and say hey, we got_some questions
! 9 we'd like to get clarified.
j 10 How to interpret the results and sample evaluations.
11 Again, this is the, I think what was being looked at before and i
12 that is, are there any good examples out there.
! 13 What we propose, and if I could skip right now -- as
() 14 you'll see here, it says the IPE review document would be made
! 15 available to all utilities shortly after the issuance of a 16 generic letter.
j 17 I bellese you alt got a copy of a -- I'm sorry, Dean l 18 is passing it out now -- on the CRGR recommendations, and this i 19 may be a little helpful in determining where we're going to go 20 with this.
I 21 First, we would issue the generic letter with all its 22 apoendices. Both front end and back end analyses would be 23 included in the request.
l 24 As I said before, NUREG/CR-4920 would be issued l
l 25 separately so it would not be considered as a requirement but l~
Heritage Reporting Corporation (202) 628.4888
180 1 rather as just another useful guidance document.
2 We would prepare our review document, our standard 3 review plan, if you will, for the IPEs.
4 And we are proposing to conduct workshops for the 5 industry. And at those workshops we would discuss this review 6 plan, so the industry could see and understand exactly what it 7 is the staff is going to do with the results when it comes in.
8 Dr. Siess.
9 DR. SIESS: How long do you think it will be before 10 you have those well-developed guidance for the workshops?
11 MR. ELTAWDA: At the end of May we'll have the first 12 draft and then discuss it with the licensee in the workshop and 13 after the conclusion of the workshop we will just see if we 14 need to refine this guidance further or not.
15 MR. SHERON: That may be a little optimistic. That's 16 the first draft from our contractor.
17 MR. ELTAWDA: That's what we want to discuss.
18 DR. SIESS: All right.
19 MR. SHERON: There's an internal review process.
20 DR. SIESS: This well-developed guidance goes beyond 21 the 4920 and all the other paper we've seen, and you are going 22 to develop that in the next month?
". J MR. SHERON: It's already under development. But 24 this is guidance on how the staff is going to review their 25 results.
Heritage Reporting Corporat ion (202) 628-4888
9 181
() 1 DR. SIESS: No. This is well-developed guidance on 2 how to do the analysis. I'm reading right off your slide.
3 Purpose cf the workshop -- presents the licensees 4 well developed guidance on how to do the analysis, particularly
'S the back end.
=6 And I said when do you expect to have that well-7 developed guidance ready?
8 MR. SHERON: You've heard it.
9 DR. SIESS: I've heard it?
) 10 MR. SHERON: We are going to go out and explain it, 11 discuss it with the industry.
12 MR. CORRADINI: So the logic is that the letter will
- 13 not be issued but you'll start these workshops for the
() 14 utilities to see.the draft review document?
15 HR. SHERON: The letter will be issued. The letter 16 will be issued. About two months later we would anticipate 17 conducting workshops in the regions as well as headquarters, 18 with the utilities, invite them, whoever wishes to come to the 19 open meetings.
! 20 At that time, we would, as I envision it, we would be 21 walking them through the letter, answering the questions that 22 they may have after they've had a chance to read it, digest it I
23 and see what's being asked of them.
7 24 In addition we would present to them our review 4
25 document. In other words, here's how we intend to review your
, Heritage Reporting Corporation (202) 628-4888
182
() 1 results when you submit them? Here's the stuff we'll be 2 looking for and the like. Get feedback from them. We're not 3 claiming it's going to be perfect. We'll be down here to the 4 subcommittee again with this review document. And we'll be 5 showing you what we think we want to look for.
6 Again, once we have these workshops we want to get 7 feedback from the utilities where they think there are hard 8 spots, where they don't understand things, where they think we 9 can sharpen up our guidance a little bit. And what we've 10 committed to do to CRGR is to issue the review document in 11 final form and at that point we're saying we would start the 12 clock for the utilities now to go forward and start their IPEs.
13 MR. CORRADINI: Start the clock for the 120 days to 14 respond to you or start the clock for the three years?
15 MR. SHERON: Well, it would be to start the clock for 16 the 120 days. But we're thinking about cutting that maybe down 17 to 60 as a proposal.
18 MR. CORRADINI Oh, 19 MR. SHERON: Because they've already had the document 20 for two months.
21 DR. KERR: Yes, but they haven't had your explanation 22 of what it means.
23 MR. SHERON: Yes, but they've had two months to think 24 about it.
25 (Laughter)
O Heritage Reporting Corporation (202) 628-4888
183 1 MR. SHERON: If the utilities come back and say hey, 2 60 days is too short, we can't do it and here's why, we'll 3 reconsider.
4 But again I think most people, you know, even the 5 industry has stood up and said hey, we have to get on with this 6 and get started. We're not making the plant safer by 7 negotiating. You're only going to make them safer when they 0 start to do this work sud start to understand where the 9 vulnerabilities in their plants.
10 DR. KERR: You might make the plant safer by 11 negotiating if going in and fixing things makes it worse.
12 Since the staff is not telling utilities what to do, it's 13 conceivable that the suggestions that come out of the utilities O>
\- 14 will make things worse with the background that the utilities 15 have.
- 16 MR. SHERON
- I guess that's conceivable but --
17 DR. KEaB: It's very conceivable.
18 MR. SHi?RON : I'm not sure what you're asking us to 19 do.
20 DR. KERR: I'm simply -- you said that negotiating 21 wouldn't make plants safer and I'm suggesting that it could if 22 negotiating keeps people from going in and doing the wrong 23 thing.
24 MR. SHERON: But I'm only talking about writing a 25 letter that says when they're going to start their analysis and O Heritage Reporting Corporation (202) 628-4888
c ,
184 1 how they're going to'do it. That's all.
2 DR. KERR It'a a hypothetical situation, let's hope, 3 sir.
4 MR. SHERON: Okay. Let me just go through here. I 5 think, yes,.the results review again. Let's go through numbers 6_ real quick. There are 109 licensed plants. If you look at 7 replicate plants, we'd probably reduce, we're guessing, about 4
8 80 IPE submittals.
9 We have estimated that on the average, and please i
10 note on the average, about six person months per plant review 11 time is proposed which would include any review of-a 12 modification.
13 As I said, on the average. I think the first several 14 plants that can come in are going to take a lot more time than 15 six person months. But hopefully once we get a couple reviews 1 16 under our belt and feel a little more comfortable, I think the 17 reviews are going to be a little more routine. We'll know what 18 to look for, what not to look for and the like.
! 19 As I said, we don't intend to spend a lot of staff 20 resources. We don't have a lot of staff resources to spend on 21 this. We don't think we need to look at each of these in the 22 same depth that we've looked at PRAs in the past and again, you 23 know, the primary beneficiary is the industry on this.
24 We are guessing that IPE submittals would be made l 25 over a period of about three years. Some plants are probably 4
Heritage Reporting Corporation
! (202) 628-4888
185 going to submit things very early, if they already have a PRA
/( ) I 2 that's on the books that's fairly up to date. They would just 3 have to do the review with their staff and make sure the staff 4 understands it and then submit it.
5 Other plants don't have anything and are going to 6 have to go out and start from scratch.
7 We're looking initially about 16 person years effort 8 per year which is staff effort and this comes from eight 9 contractor staff and eight in-house staff. NRR will provide 10 four and the Office of Research will provide four. This is 11 again a firs *. cut.
12 The lead for the reviews will be in the Office of 13 Research, with people from NRR, and maybe even physically
() 14 located over in the research building. Dr. Murley has 15 indicated that he would like to rotate the people on one-year 16 assignments. We would most likely do the same.
17 So that at the end of the three or perhaps four year 18 period both the Office of Research and NRR would have upwards 19 of perhaps 15 or 16 people that participated in these reviews 20 and have this experience.
21 If you can crank out eight contractors, that's 22 something like about 1.4 million a year in money that we would 23 have to spend.
24 And again, we're thinking about breaking up into four 25 teams where you would have two NRC staff and two contractors O Heritage Reporting Corporation (202) 628-4888
1
~
186 j 1 working on a review team for certain plant types. <!
2 And for example I think we had the MARC-1s and the [
3 ice condensers -- I'm sorry,.we have it here. Down here. Yes.
4 -Two teams would only do large dry containments; one team would ;
5 do the MARC-1 and MARC-2s and one team would do the MARC-3 and- [
6 ice condenser plants.
i 7 So there would be some continuity as well as the t 8 teams would not be bouncing between different. containment 9 types.
10- The team concept, as I just said, each plant review 11 would be the responsibility -- there would be team leader. .
12 That would be the person that's on the chopping block who would !
13 be responsible for that review. And then we'd have plant i O 14 systems people, a PRA specialist, and these teams would be able l
15 to go out to the staff or their, you know, say contractors, to
]
16 get specialized expertise if we had to. Dr. Shewmon?
17 MR, SHEWMON: Would you tell me again what the ,
18 schedule is? You were talking about 60 and 120 days, but some f 19 of these are going to come in three years from now, others much ;
20 sooner apparently.
21 MR. SHERON: The plan was, well, the original plan 22 was that we would issue the letter and the letter would say 23 please tell us within 120 days of your plans and schedu3es for 24 completing your examination.
25 After 120 days, the utilities would come in with a O Heritage Reporting Corporation (202) 628-4888
187 I 1 letter that says okay, here's what we're going to do. And they 2 may say, and we plan to have our submittal-two and a half years-3 from now or we're going to use our existing PRA and we're going 4 to update it and fix it and we'll have the submittal in one 5 year.
6 But in essence we would expect that once, at the end 7 of the 120 day period, and the letters come in from the 8 utilities, that they ,ould all indicate that they would 9 complete their submittals within a three-year period.
10 MR. SHEWMON: And you hope they are more or less 11 uniformly distributed in around t.wo years or something?
12 MR. SHERON: Yes. I can't guarantee they're going to >
13 be uniform. Somebody said they may use the IRS approach and 14 commit it to deadline. I don't know.
15 MR. MICHELSON: This idea of having some lead plants 16 to kind of solidify your process, how does that play into this?
17 MR. SHERON: We're going to be looking at whether we 18 can review some plants on a lead plant concept. I'm not sure 19 we have any legal authority to require any one utility to 20 submit their IPE before another.
21 What we may be doing is just taking the first couple 22 of plants that come in of each containment type and calling 23 them a lead plant and reviewing them, you know, probably in a 24 14ttle more detail than the remaining ones.
25 DR. KERR In connection with that, on Page 14, I O Heritage Reporting Corporation (202) 628-4888
388 1 find the statement that: In iniplementing a severe accident 2 policy the Commission enco.urages those plants that have not yet 3 undergone any systematic examination to promptly initiate the 4 examination.
5 How does the Commission encourage them to do that?
6 MR. SHERON: It says so in a letter, basically.
7 There isn't any way that we can -- ,
8 DR. KERR: Well, what doos it mean? I thought you 9 were going to give them some special concession maybe if they 10 did.
11 MR. SIESS: doesn't it say back on Page 1 something 12 that requires them to do it?
13 MR. ELTAWDA: They are required to do it but --
O 14 HR. SHERON: They are required to submit information 15 on the vulnerabilities, on their plants vulnerabilities to 16 severe accidents.
17 DR. SIESS: This is back on Page 1.
18 DR. KERR: Then on Page 2, we encourage each licensee i 19 to use its staff. How are you going to encourage them to do 20 that?
21 DR. SIESS: "You are requested to perform an 22 individual plant examination and submit the results to the .
23 NRC." That's the requirement of the generic letter, right?
24 MR. SHERON: Yes.
l 25 DR. SIESS: You are requested to. That means you do !
O Heritage Reporting Corporation (202) 628-4888
s 189
( [ 1 ~1t.
2 MR. SHERON: Yes.
3 DR.'SIESS: What does it mean then that you encourage 4 them to do it?
5 DR. KERR the encourage is in a different context.
6 Encourage is first used on Page 2 where each licensee is 7 encouraged to use its staff.
8 DR. SIESS: I'm not talking about that. There is a 9 requirement on Page 2. Now, the encourage on Page 14.
10 MR. ELTAWDA: The encouragement on Page 14 is to have 11 those utilities that they have not done a PRA or IPE to. start 12 immediately, not to wait until the end of the three years.
13 They still have to do it, but don't wait until the end.
() 14 DR. SIESS: Are there some utilities that have done 15 an IPE?
16 MR. ELTAWDA: IPE and PRA are synonymous. They do a 17 nyctematic examination of the plant.
18 DR. SIESS: Wait a minute. An IPE and a PRA are 19 synonymous?
20 MR. ELTANDA: That's in the policy statement.
21 DR. SIESS: Aren't you calling for more than a PRA?
22 You are asking them to evaluate this and evaluate that and come 23 in with fixes and so forth? Doesn't the list of documentation 24 call for more than submitting a PRA?
25 MR. SHERON: Of course it does.
O Heritage Reporting Corporation (202) 628-4888
190 (3
(_) 1 DR. SIESS: Then the IPE is not synonymous with a 2 PRA. The IPE is a PRA plus, isn't it?
3 MR. SHERON: Yes. Okay, yes.
4 DR. SIESS: In that sentence, I have the same 5 question Dr. Kerr had. After you've required it, to say you 6 encourage it. You're trying to-say that some of you have a 7 head start, but that's in the next paragraph.
8 MR. SHERON: What page are you on right now?
9 DR. SIESS: Page 14.
10 MR. SHERON: Of the generic letter?
11 DR. SIESS: Yes. The last paragraph says, those with 12 an existing PRA, you know, we expect to get in before three 13 years. But this, you know, those that haven't ought to get O
\/ 14 started. But this comes before that. It says those plants 15 that have not yet undergone a systematic examination to 16 promptly initiate the examination. I don't know how else to 17 interpret it. You've required something, and then you 18 encourage it.
19 MR. SHERON: No. We have required them to make a 20 submittal. We are encouraging plants that have not done a 21 systematic evaluation to do it, to start it right away and not 22 wait .
23 DR. KERR: But you aren't really encouraging them to, 24 you're just saying it would be nice.
25 HR. SHERON: I can't require them to.
4 Heritage Reporting Corporation (202) 628-4888
191
() 1 DR. KERR: Well, but "encourage" carries with it 2 generally that they get some reward for doing it.
3 DR. SIESS: What does the statement on Page 2 -- 1 4 don't want to be a lawyer for goodness' sakes, but on Page 2 it 5 says you are requested to perform an individual plant 6 examination and submit the results to NRC , and you say that's 7 not a requirement? '
8 MR. SHERON: That's a requirement.
9 DR. SIESS: Then why did you say you can't require 10 it?
11 MR. SHERON: All I can say is that they are required 12 to submit the results of their examination.
i 13 DR. SIESS: Yes.
() 14 MR. SHERON: And to give us a schedule. So I said 15 some utilities --
l 16 DR. SIESS: Encouraging them to start early.
17 MR. SHERON: The ones that don't have it we're saying 18 start early. Don't wait.
i 19 Some utilities may wait and say gee, the staff 20 doesn't want this stuff for three years so I'll wait until the ,
21 end of year two and then I'll start it.
22 DR. SIESS: They might say we'll wait until the end i
- 23 of three years and maybe the staff will know what it wants. ;
24 (Continued on the next page) 25 Heritage Reporting Corporation (202) 628-4888
192
() 1 MR. MICHELSON: How long before you think that you 2 are going to be ready to tell the utilities how to handle 3 external events?
4 MR. SHERON: Eighteen months.
5 MR. MICHELSON: I beg your pardon.
6 MR. SHERON: Eighteen months.
7 MR. MICHELGON: Okay, so it is eighteen months. So 8 even if they started today, they cannot finish obviously in 9 less that eighteen months, because they will not even have 10 their instructions until eighteen months for the external 11 events part. So they cannot finish before twenty-four months 12 or something at the earliest.
13 MR. HOUSTON: But the external events are separate.
() 14 They will have another schedule.
15 MR. MICHELSON: But it is a part of the package that 16 they have to finally submit.
17 MR. CORRANDINI: No, it is not. That is the reason 18 that they are nodding their head no behind you.
19 MR. MICHELSON: You mean that external events will 20 never be in the package?
21 DR. KEP.R Not in this one.
22 MR. SHERON: A separate package, a separate 23 submittal.
24 MR. MICHELSON: You mean another request letter and 25 so forth?
O' Heritage Reporting Corporation (202) 628-4888
193
[
1 DR. KERR: Yes.
P 2 MR. MICHELSON: Okay. I misunderstood. I thought !
3 'that you were going to kind of make it an addendum to this one. f 4 MR. HOUSTON: Excuse me, Brian. On the first bullet, l 1
i 5 what do you mean by replicate plan?
6 MR. SHERON: Sister unit. -The same thing, the same
- 7 plant, you know. [
l 8 MR. HOUSTON: Does it have to be at the same site, '
9 the same utility? [
t 2
10 MR. SHERON: Yes, most of the unit site, the same j t
11 architect-engineer.
12 MR. HOUSTON: The same architect-engineer? !
?
13 MR. SHERON: Duablo Canyon with a mirror image. !
O '14 DR. SIESS: How-about Gravewood and Firing?
I 4
15 MR. SHERON: I.do not know. !
i 16 DR. SIESS: Those were licensed I thought as t f 17 replicates. f 18 MR. HOUSTON: They would not meet the requirement :
J.
j 19 that the utility do it if they belonged to a different utility, i 20 DR. SIESS: Those belong to the same utility. [
21 MR. HOUSTON: Would you not like to see them enow l l
22 that the plants are indeed replicate? Because we have seen !
23 examples where plants were supposed to be identical really are
- 24 not, because they are built at different times, and there have .
25 been some changes made. !
O Heritage Reporting Corporation (202) 628-4888 i
i
- _ - . . _ _--_L_..-___,_,_-_._..__... . . , _ . - , _ . . , . --
4 194 ,
t
(_,) 1 MR. SHERON: The utility would have to certify that !
2 these plants are identical.
3 MR. HOUSTON: You might want to discuss that a little 4 bit in your letter, because it may not be clear to a utility 5 that he does not need to do anything for one of the plants.
6 MR. SHERON: This is our review plan. The industry 7 does not know anything about this. This is a resource 8 estimate. If a utility does an IPE and says 900, the plant 9 next door looks exactly like this, I can use it, fine, they can 10 submit it. But they have to make that determination. I am not 11 going to tell them.
12 MR. HOUSTON: What do they submit to you, se that ycu 13 are satisfied that they are identical?
15 MR. HOUSTONt On both of them?
16 MR. SHERON: If they make a determination that the 17 IPE that they did on one is the same as the one that they would 18 do on another, okay, they submit it. I do not have to make 19 that determination for them.
20 MR. HOUSTON: But you have to be satisfied that the 21 plants are identical.
22 MR. SHERON: If the licensee comes in. These things 23 have to be submitted under oath and affirmation. When that 24 vice president signs a piece of paper that says here is my 25 IPE and it represents this plant, I mean he is making a O Heritage Reporting Corporation (202) 628-4888
195 1 statement there that says that if he-is wrong, that he goes to-2 jail, in simple terms.
l 3 MR. HOUSTON: But you are not asking him to show any ;
' 4 avidence that the plants are similar? !
r 5 MR. MICHELSON: They have to audit, if they wish. !
i !
6 The audit is up to the NRC, if they have any question or doubt. l l 7 MR. SHERON: What is the utility supposed to show,
<8 you know, are they supposed to submit drawings to show that ,
9 everything is the same. I mean I do not know what I would look j
] ,
10 for, other than the utility saying we have looked at our plants ;
f 11 and they are indeed the same, or that we have made the ;
I 12 following changes. j i
DR. SIESS:
13 Are you going to send out a list of 14 questions?
i 15 MR. SHERON: What? ;
16 DR. SIESS: Are you going to send out a list of !
17 requests for additional information? }
18 MR. SHERON: No. i 19 DR. SIESS: That is how the staff usually gets at j i
20 these things. !
! 21 You do not expect to do that?
i 1
22 MR. SHERON: Not really. !
i 23 DR. SIESSt Your review process here will not be the f i
4 24 same as the review process has been in the past say for the j
! 25 safety evaluation report? i
( Heritage Reporting Corporation l j
- (202) 628-4888 ,
1
4 196 MR. SHERON: I guess that we have not decided yet if
{} 1 2 we review what they have submitted and we need some additional 3 information or clarification how we would go about getting'it.
4 It may just be that we want to have a meeting with the people 5 'that did the work just informally.
6 DR. SIESS: But if the_ technical reviewer decides 7 that he needs more information or he questions whether the 8 plants are identical and he passes that along to the project 9 manager, and the project manager sends it out to the applicant 10 or the licensee, do you expect to have somebody in the process 11 with these teams that filters that out?
12 MR. SHERON: Well, number one, if anyone goes-to the 13 licensee and reaiests them to expend resources, it would be
() 14 through NRR.
15 DR. SIESS: That is where it is now.
16 MR. SHERON: That is right. If the review teams -
17 requested additional information, just like if we were 18 reviewing a topical report, you would go through the project 19 manager and decide on how best to obtain the information that 20 you need. New the project manager might set up a meeting with 21 the utility and the review team, and get the appropriate people 22 just across the table.
23 The intent is not to put any big burden by going 24 through sending out a hundred questions and all of this stuff.
25 You know, you cannot do that within a six month review O Heritage Reporting Corporation (202) 628-4888
- F '
l' .
197 y
. (V -1 schedule.
2 Last but not least is the staff review of the IPE 3 results. And this I hope will answer the_ question what are we 4~ going to do with this stu'ff'when we get it..
5 First off, we are going to review the results for the 6 following. We want to assure that the utility did an adequate 7 analysis of plant design and operations. To make sure that 8 they discovered particular vulnerabilities to core damage, and 9 to use the Commission's word on poor containment performance.
10 And as I mentioned before, we will probably act as an 11 ad hoc clearinghouse to look at consistency in the 12 identification and treatment of leading core damage sequences.
13 Like I said, if I get plants that are very similar in design, (N # 14 and one utility comes in and says I have got a bad actor here 15 and nobody else identifies it, we may ask questions as to why 16 did these other plants have that same bad actor.
17 DR. SIESS: How will you handle this chronologically.
18 Suppose Plant A comes in first, and ten plants down the line 19 Plant B comes in with what you just said.
20 Do you go back now and get Plant A7 21 MR. SHERON: If we think that is significant that i
22 should be brought to the attention of that plant, I think that 23 we would do it.
24 DR. SIESS: My point is that I think that your idea l
! 25 of what is consistency is likely to vary with the tide. If you I
() Heritage Reporting Corporation (202) 628-4888 l-
198
() I have ten reviews coming in sequentially, what you thought was 2 consistent at the beginning and what you think is consistent at 3 the end may be different.
4 MR. SHERON: Like I said, two years down the road, if S we determine that some plant identified a bad actor and we did 6 not even look at it on the five plants that we looked at the 7 year before, we will have to determine the significance of it.
8 And I think that as a minimum that we would most likely, you 9 know, unless we were going to take some action on that plant, 10 we would just go back and let those other plants know about it.
11 DR. SIESS: Now your review teams are back end 12 oriented, or they are PWR versus BWR.
13 MR. SHERON: Well, it was not even so much that. We (ms_) 14 have four teams.
15 DR. SIESS: If you have got them separated bv 16 containment, they are back end oriented. You have them listed 17 by containment.
18 MR. SHERON: Some people have said, gee, maybe you 19 should lump them by architect-engineet. I do not know what is 20 the right way. Because someone will say then they are front 21 end oriented. And you know, gee, why is that right compared to 22 lumping them by containment type.
23 DR. SIESS: So you really do not have any intention 24 there?
25 MR. SHERON: It would be a logical way to do it.
k- Heritage Reporting Corporation (202) 628-4888
- ( ). 1 DR. SIESS'
- A large dry, whether it is a steel
'2 containment or a concrete containment, is a large dry. Okay.
3 I will take back what I said.
4 MR. SHERON: Let me see where was I here. Again, you 5 know, consistency and identification of treating of leading 6~ -core damage sequences, to assure that sequences and sequence 7 frequencies are appropriate. This is basically _ sort of like 8 looking over their shoulder. Does everything look reasonable.
9 You know, stuff that we have experience in on what the range of 10 numbers are for certain sequences for that kind of plant, does
- 11. this look reasonable or is it out of line.
12 And to independently conclude that the decisions that 13 the utility makes on whether to make improvements are
() 14 justified.
15 DR. KERR: How are you going to do that?
16 MR. SHERON: Judgment. You know, was it a prudent 17 judgment that they made. We do not have some specific 18 criteria.
19 DR. KERR: And this will be done by the team?
20 MR. SHERON: Yes. This is another argument. You 21 know, if I want a bottom line number, what is going to happen 22 is just like Appendix K. If we tell the industry that the 23 speed limit is 2200 degrees, I get peak clad temperatures that 24 all come in at 2199.
25 DR. KERR: I am not being critical at this point.
Heritage Reporting Corporation (202) 628-4888
r , a 200 s-) 1 MR. SHERON: I understand.
2 DR. KERR: I am trying-to understand what is that you
^
3 are planning to do.
4 MR. SHERON: I want to point out though that there is 5 another reason why one does not want to use an absolute 6 criteria. Because if we told them that the bottom line is a 7 core melt to the 10 to the minus 4, everyone would shoot for 8 it.
'9 DR. KERR: Somebody is going to use some criteria 10 though. You have to have criteria to_make decisions. You may 11 be unable to tell me what criteria will be used at this point, 12 and you may never be able to tell me what criteria you would 13 use, but there would have to be some. Otherwise, you get no 14 decisions. But if you are going to independently conclude and 15 you disagree, then you get into the backfitting mode.
16 MR. SHERON: Ultimately, if we disagree.
, 17 DR. KERR But first do you feel that you have to 18 have a rulemaking, or can you backfit without a rule?
19 MR. SHERON: You can backfit without a rule if the 20 Commission so chooses by an order.
21 DR. KERR: If you conclude that changes are made, 22 does that constitute a backfit and subject to the backfit rule?
23 MR. SHERON: If the staff concludes it and the 24 utility disagrees.
25 DR. KERR But there is already a disagreement. I O Heritage Reporting Corporation (202) 628-4888
201
(~'i
(/ 1 postulated that.
2 MR. SHERON: Okay.
3 DR. KERR: You have made the independent conclusion, 4 and you disagree with the utility, and then what do you do?
5 MR. SHERON: We would first more likely try to 6 dialogue with the utility and understand.
7 DR. KERR: No, I am assuming.
8 MR. SHERON: Okay, that is all done.
9 DR. KERR: What happens then? ,
10 MR. SHERON: Presuming that we agree to disagree.
11 The staff would then have to determine if (a) we-think that the 12 scenario needs to be fixed, and if it needs to be fixed solely 13 for severe accident considerations. In other words, the plant 14 totally meets the licensing design base and all of the 15 Commission's regulations.
16 Then we would (a) have to do a backfit analysis to 17 show that whatever improvements that we were proposing to 18 backfit on that plant was a substantial improvement to safety 19 and cost beneficial. And number two is that the imposition 20 would either be by the staff developing an associated rule to 21 require that, or by asking the Commission to issue an order.
22 MR. CATTOM: Brian, on the same, independently 23 concluding.
24 What are you going to do about DCH and the liner 25 penetration? As far as I can tell, all of your contractors Heritage Reporting Corporation (202) 628-4888
'i 202
..() Edisagree at this. time, that there is a lot of disagreement.- It 2 .seems to me that one utility might get one guy, and the other
.3 utility gets the other guy. ,
4' MR. SHEROli: Well, like you said, if there is 5 inadequate data right now, and we recognize that there is 6 inadequate information for one to draw any definitive 7 conclusions. Otherwise, we would be doing it right now. You 8 know, the liner melt, you will be hearing from Jerry Hulman, 9 not today, but sometime in the future on his container
- 10. performance initiative, which is specifically addressing ~that 11 very issue.
12 MR. CATTON: But he has to finish first before you 13 can independently conclude.
14 MR. SHERON: We are going to independently conclude 15 whether direct containment heating is or is not a problem for a 16 plant. But what we told the industry, and I do not want to go 17 through this again, but all we told them to do is that they 18 need to understand and be aware of the issue and understand its 19 consequences on their plant.
20 And lastly here, hopefully when we are done with the l 21 review, it would allow us to go down to the Commission and tell 22 them that the industry has responsibly implemented their severe 23 accident policy.
24 How consideration will include both quantitative 25 measures and non-quantitative judgment to determine if a plant Heritage Reporting Corporation (202) 628-4888 I
}
203 O\_/ 1 is acceptable or not. And as I'said here, as I just said, if 2 the NRC consideration indicates that the plant design or 3 operation could be enhanced by additional protection beyond the 4 regulations, implementation would be in accordance with the 5 backfit rule.
6 If we find something that says that they need to 7 change the plant to meet the regulations and it is not a new 8 interpretation, then implementation would be without regard to 9 cost except to select among various alternatives.
10 IPE results will be used to identify severe accident 11 vulnerabilities. generic to a class or several classes of 12 plants. Now this is where we get into the safety goal.
13 Information on generic vulnerabilities will be used
-'-/f i 14 to examine if deficiencies in the regulations exist. If 15 generic deficiencies were identified, the safety goal would be 16 used to determine if regulation modification were needed.
17 What we are going to do is that we are going to look 18 at all of these IPEs or PRAs, or whatever comes in, and we are 19 going.to say is there anything that is coming out of these that 20 appears to be generic in terms of a deficiency in the plant.
i 21 Is there a vulnerability that keeps showing up in the same kind 22 of plant or something, or in the same system, and does that 23 indicate to us that there is a deficiency in the licensing i 24 regulations that allows that system with that vulnerability to l
l 25 occur in all of these plants.
() Heritage Reporting Corporation (202) 628-4888
204
.m
.s. ) 1- And if that is the case, then we would take a look at 2 that'. And we might.even look at the bottom line numbers that 3 the utilities were giving us, and compare that in an aggregate 4 sense against the safety goal, and determine if there were 5 something that implied that we needed to fix the regulations.
6 So that if by fixing the regulations, it would fix 7 these plants to the extent that we could say that they meet the 8 safety goal. That is.how we intend to use the safety goal in 9 terms of using it in conjunction with the IPE results.
10 MR. LEE: Would the review process be prejudiced 11 depending on whether the utility staff or personnel were 12 involved in the IPE or not? This is a point that was brought 13 up, I believe, earlier. The Generic Letter encourages staff
- 14 invo3vement.
15 MR. SHERON: I think that typically if a utility, if 16 we find out that a utility did not involve its staff, you know, 17 we would try to learn that early on, and I am sure that there 18 would be meetings with the appropriate senior officials of the 19 NRC to further encourage the utility vice presidents to involve 20 their staff.
l 21 MR. LEE: You do not feel the need to specify that 22 beyond what is on page two?
23 MR. SHERON: Under 50.54(f), I do not think that we 24 have any legal authority to require them to involve certain 25 people on their staff. We encourage it, okay. Keeping in mind l Heritage Reporting Corporation l_ (202) 628-4888
4 205 f'T
- (/- 1 _that_this is.a unique requirement in the sense that the 2 Commission issued a policy which said that.they wanted these 3 examinations done.
4 There were two choices to implement ~this. One is 5 50.54(f), which I guess 1s a milder form, which is requesting 6 that the utility' submit certain information, so that the staff 7 can make a determination.
8 The other alternative is just to issue an order ,
9 which says that you will do this, this, and this accord 1ng to 10 this, this, this in this way, and we chose not to do that. But 11 we could have gone that route as an alternative, which would 12 have said you will involve the utility staff, you will use this 13 method, you will submit this,'and there is no flexibility 14 there.
15 And we felt that the 50 .54(f) gave us a little more 16 flexibility in trying to accommodate the way that the industry 17 would go about preparing the information.
18 That concludes my presentation on the letter. If you 19 want, I can go through now Dr. Speis', or if you want to take a 20 short break.
21 DR. KERR: Why do we not take a short break. Let's 22 come back at ten of.
23 (Whereupon, a recess was taken.)
24 DR. KERR: This may look like the same speaker that 25 wo have heard, but this is not actually. This is somebody O Heritage Reporting Corporation (202) 628-4888
206 1 . named Spels'.
2 MR. SHERON: This is a. presentation which Dr. Speis
'3 was going to give, which is the discussion of the integrated 4 plan for severe accidents and hopefully closure.
5 (Slides being shown.)
6 MR. SHERON: Very briefly, the background is that in 7 August of 1985 that there was a severe accident policy 8 statement, which is the basis for going forward with the IPEs.
9 On February 28th of 1986, the staff in 86-76 submitted the 10 implementation plan for the policy statement. On .
11 December 1, 1986, the Chairman sent down to the staff a request 12 to integrate issues on severe accidents.
13 On February 17th of 1987, a memo went back from the 14 EDO to the Commissioners with a preliminary plan for 15 integration of severe accident issues. On July 15th, the 16 staff, both the Offices of Research and NRR, briefed the 17 Commission on a plan for closure of severe accident issues.
18 On December 8, 1987, the staff sent down to the 19 Commission a SECY 87-297, their plan for MARK I containment 20 performance. And in February of this year, on the 8th and the 21 lith, there was the infamous Baltimore meeting on severe 22 accidents, which Mr. Stello got together with management in the 23 Commission is involved on the severe accident issue, and 24 discussed the overall plan for how we reach closure on the 25 issue.
O Heritage Reporting Corporation (202) 628-4888
207.
() 1 And the basic outcome of that meeting was that the 2 staff would prepare a-plan for integrating and closure of all 3- severe accident issues. And this shows the objectives of that
-4 plan,-which would be to provide an understanding of staff 5 activities that are presently underway. To implement our 6 . severe accident policy, to assure that these activities are 7 consistent with the Commission's policies and strategic goals, 8 to assure that staff activities are consistent among themselves 9 with a common goal of ultimately leading to improved plant 10 safety and are properly coordinated.
11 And lastly, to assure that the Commission is aware of 12 the key technical and policy issues, some of which need 13 Commission guidance and approval.
14 MR. WARD: You sure cannot get that, unless you are 15 aware of the, right?
16 MR. SHERON: Yes. Just to give you an idea of the 17 severe accident issues that are currently underway within the l
18 agency. You just heard ad nauseam the individual plant l
l 19 examination process. You would have heard later this afternoon 20 from Jerry Hulman the container performance improvement 21 program.
22 The Office of NRR is conducting an ongoing program of 23 improved plant operations. The Office of Research, as you 24 know, is conducting a severe accident research program. We 25 have just initiated in this fiscal year -- yes, sir.
l O Heritage Reporting Corporation L (202) 628-4888 l
I
208 1 MR. SHEWMON: Have there been any changes in the
[G^)
2 severe accident research programs that would lead you as 3 somebcdy not immediately connected with it that it is 4 converging any faster than it has in the last year or two?
5 MR. SHERON: I really am not qualified to answer 6 that, because I have not followed the program in that detail to 7 make a judgment.
8 MR. SHEWMON: Thank you.
9 MR. SHERON: The accident management program. As I 10 said, that is a new initJative that we just started up this 11 fiscal year. The lead is in my division. We are in the 12 process right now of putting together a research program plan.
13 There is a first draft on my desk. Once we get the internal e
(,n; 14 review, my guess is that we would be down here discussing it 15 with you in terms of what we propose to do and areas that we 16 will be looking at.
17 This will probably be, right now we are talking about I
18 an expenditure of maybe of upwards of $3 million a year in this 19 area. We have already had one meeting with NUMARC and EPRI to 20 learn about what their approach is in this area.
21 NUREG 11-50, which you have heard about, which 22 hopefully will be coming out, I guess that it is scheduled for 23 this fall, which is the reactor risk reference document. The 24 Commission has promulgated safety goals. And the staff under 25 Wayne Hausman is putting together a safety goal implementation
( Heritage Reporting Corporation (202) 628-4888
e 209
()
I 2-program.
We have, as you heard before, the generic issues.
3- One thing that I guess that I forgot to address for you is the 4 relationship with 844. Let me do that as soon as I finish 5 this.
6 External events, which you heard about also before.
7 The ISAP program, Tom Cox gave you a little bit of background 8 on before. There is the severe accident policy implementation 9 for future plants which Billy Morrison's division is 10 responsible for implementing. And then there is the general 11 . question of severe accident closure. When do we say that we 12 have closed on severe accidents for the operating plants.
13 MR. MICHELSON: External events real fast, what did
() 14 you say that you were doing on that or what comment did you 15 have?
16 MR. SHERON: That I discussed it previously today.
17 MR. MICHELSON: Yes, but there is more to it than 18 that. You discussed it as it relates to the IPE program 19 somewhat, like you are still thinking about it for IPE.
20 Is that all that is going on, is that the only thing 21 going on?
l 22 MR. SHERON: For external events?
23 MR. MICHELSON: Yes, external events.
24 MR. SHERON: Right now, I guess that is all.
25 MR. MICHELSON: That is the only activity going on?
O Heritage Reporting Corporation (202) 628-4888
b l
210 M R '. SHERON: Well, there are other seismic. You
(( ) 1 2' know, there is the fire scoping study.
3 DR. KERR: The seismic margins program is still 4 underway.
5 MR. SHERON: The seismic margins program. There are-6 a number of programs ongoing, but they are not necessarily 7 geared towards severe accident closure. There is a
-8 relationship there, but the only thing that is directly related 9 to severe accidents from the standpoint of risk, you might say, 10 is the IPE.
11 MR. MICHELSON: That is the only thing. .Thank you.
12 MR. SHERON: This chart here --
13 DR. KERR: Why do we not stipulate that chart.
O)
(,_ 14 MR. SHEWMON: Do you want to subsume it someplace?
15 MR. SHERON: No, I just want to explaia very quickly.
16 If you follow this through, you will see that hopefully these-17 programs all fit together.
18 DR. SIESS: The box down at the bottom on the left, 19 and I see an arrow going off to the right, way off to the 20 right, and I see a line going up from that box.
21 What kind of arrow, is that leading in or out?
22 MR. SHERON: Here?
23 DR. SIESS: Yes.
24 MR. SHERON: Leading in.
25 DR. SIESS: No, the bottom, down at the bottom. You O Heritage Reporting Corporation (202) 628-4888
211
(_ ) 1 do expect the severe accident research to feed in somewhere.
2 MR. SHERON: To feed into container performance 3 improvements.
4 DR. SIESS: But you also expect it to continue after 5 the severe accident closure?
6 MR. SHERON: Yes. What we are saying is that when I 7 talk about closure, what I am not saying is that we would have 8 solved all of the questions associated with severe accidents.
9 But we will have addressed them to a satisfactory degree. That 10 we feel confident that we are not going, for example, be 11 requiring the industry to do more at this time. It is more 12 confirmatory research, you might say.
33 DR. SIESS: You are in research, are you not?
p kJ 14 MR. SHERON: Yes.
15 DR. SIESS: I do not see any arrows leading into the 16 severo accident research from some of this other stuff. And I 17 would certainly expect that there would be some input to the 18 direction of the severe accident research coming out of these 19 other actions up there.
20 MR. SHERON: Well, for example, 11-50 is a feed-in.
21 DR. SIESS: I just asked you the way that the arrow 22 went, and you told me that it went up.
23 MR. SHERON: I am sorry. Well, from severe accident 24 research, it goes up. But NUREG 11-50 insights help guide 25 where you do your research. So there is an arrow in this A
V Heritage Reporting Corporation (202) 628-4888
~
' 212
}()
1 direction. There is a feedback process, okay.
2 DR. SIESS' Okay, go ahead.
3 MR. WARD: Brian, where do the external events, 4 picking up on external events, come in here?
5 MR. SHERON: Well, that is incorporated in-the 6 individual plant examinations. I am talking about this in 7 total.
8 MR. MICHELSON: You do not take any research'to pick 9 up on external events?
10 MR. SHERON: Well, there may be some. That is what 11 Larry Chou's task is.
12 MR. MICHELSON: So eventually, there may be some 13 arrow going down to severe accident. ,
() 14 MR. SHERON: Ultimately, there is always a feedback 15 in all of this, but I think that it would totally complicate 16- this diagram if we kept showing that.
17 MR. MICHEL3ON: It is pretty complicated already, to i.
18 understand at least, i.
19 MR. HOUSTON: Brian, would you be willing to hazard a
]
20 guess as to what date might be associated with that last thing?
21 MR. SHERON: If I do, you will hang me with it, 72 right?
23 MR. HOUSTON: No, no. I would like to know how you 24 envision the time required to do all of this.
25 MR. SHERON: Well, as I said, the individual plant O Heritage Reporting Corporation (202) 628-4888 i
213 e
(}- 1. ' examinations, we are guessing will be initiated within a year.
~
- 2 Hopefully, we would get.the information in within three years.
3 The safety' goal and backfit policy are just going to be ongoing 4 considerations. It is no something that they are going to 5 produce something that is going to have to be implemented by 6 the industry.
7 Improve plant operations is also something that is 8 going to feed in, but also continue. USIs and-GSIs, again they 9 will continue to.be identified and resolved. There is nothing 10 that we can do to actually stop that.
' I would guess, just hazard a guess, that somewhere on 11 12 the order of five years. I know that Dr. Spels has said a 13 couple of times that he does not think that issues such as
() 14 direct containment heating or the liner melt-through question 15 may will be resolved for the next five years.
16 MR. HOUSTONt Thank you.
17 MR. SHERCN: There is another reason that I said-
[ 1 73 that. And that is if you look in this agency, there is no 19 person who says in the same job for more than five years. Five 20 years when the time is up, I will probably not be here. Someone 21 else will have to suffer with it.
22 Let me just point out, I know that a lot of people 23 has asked why 11-50, what are we using it for, what is its 24 purpose. I think that a lot of people have tried to read into 25 11-50 or interpret its use more than what it really was O Heritage Reporting Corporation (202) 628-4888
. .. . . . _ - - . =- . . .
214
., 1 intended for.
2 My recollection was that when we got done with our ,
3 source term research and source term code package and 4 everything, what we recognized is that you cannot draw any.
5 generic conclusions just by source term research, that you had 6 to apply it to a plant. And the way that you did that was 7 through the 11-50 type plants.
8 DR. KERR: But I gather that the 11-50 analyses are 9 not state of the art for containments. And indeed, there is no 10 state of the art PRA for containment. I thought that was the r
11 answer that I got to the earlier question.
12 MR. SHERON: I would think that they are about as F 13 state of the art as we are going to get right now. t
() 14 DR. KERR: Well, state of the art always means right 15 now.
16 IIR . SHERON: But there are a lot of questions. You 17 know, as you know, there has been a lot of criticism and stuff. '
18 And I guess that the jury is still not back on it. So I do not 19 want to hazard a guess to say that this is something that the 20 staff would accept.
21 DR. KERR: Well, it is an up-to-date assessment of 22 severe accident frequencies of plants, which certainly includes 23 containment.
24 MR. SilERON : Right, that is correct. !
25 DR. KERR: But it is not good enough for the IPEM.
I lieritage Reporting Corporation (202) 628-4888 ,
h
D 215 Because it would not be acceptable if you just followed the'
_ ( ): 1 2 ~NUREG 11-50-back end.
3 MR. SHERON: NUREG 11-50 accounted for the 4 uncertainties. And as you saw, there is a big spread.
5 DR. KERR: I am not trying to be critical. I am 6' trying to understand. My statement may be wrong. But from 7 what I heard earlier, I interpreted the answer to say that just 8 following the 11-50 example would not be considered to be an 9 adequate treatment of containment for the submittals that you 10 expect to get.
11 MR. ELTAWDA: The important thing that the utilities 12 have is that they accept the NUREG 11-50 results. And let's be 13 honest about it, you will accept the NUREG 11-50 results.
() 14 DR. KERR: Do you?
15 MR. ELTAWDA: Well, I work for the Office of 16 Research.
17 MR. SHERON: I would say that we consider those to be 18 state of the art. If the utility came in and submitted the 19 11-50 results and said this is our best shot at containment 20 performance, I think that we would find that as acceptable as 21 anybody else would.
, 22 MR. WARD: What are you going to do with the peer 23 review?
24 MR. SHERON: What do you mean, what am I going to do 25 with it?
! /~T l
k/ Heritage Reporting Corporation (202) 628-4888
(
216
) 'l MR. WARD: You said that you had to follow it.
2 MR. .SHERON: Oh, okay.
3 MR. CORRANDINI: Can I_get a different question in, 4- so that I understand. This is, I guess,.a question from your 5 past presentation.
6 But what role can 11-50 play in essentially helping 7 you go through the process of review, that is could 11-50 play 8 a role by using it as a straw man for your review process?
9 MR. SHERON: Well, for example, in the 10 NUREG CR 49-20 reports, the insights from 11-50 were factored 11 in right then and there.
'12 MR. CORRANDINI: Right. But I guess that I was going 13 even a step further and saying let's assume that NUREG 11-50 is J
14 finiehed, and you have get five plants out there with ,
15 essentially a front end and a back end.
16 MR. SHERON: Right.
17 MR. CORRANDINI: Can you use them as essentially your .
18 first set of surrogates to hone your review process?
19 MR. SHERON: If the utility proposed to submit the 20 11-50 as their IPE, we would accept it provided that (a) they 21 showed that the 11-50 analysis indeed represented the plant as 22 it exists today; and (b) that the utility staff somehow was 23 involved in the review, et cetera, so that they are totally 24 familiar with and have that understanding that we asked for; 25 and if they endorsed and drew their conclusions say from the !
O Heritage Reporting Corporation (202) 628-4888
~ , - .. - . .
217
(). tl- back end analysis, you know, recognizing the uncertainties, and 2 made competent conclusions from it, I~think that we would find 3 that acceptable, if they wished to adopt it.
4 MR. CORRANDINI: Right, that is from their end. But 5 I am looking from your end in the review process. There is 6 some stuff that you have to go through that you had listed in 7 one of your previous slides about looking at it and deciding 8 what improvements may have to be done. Because part of.11-50, l 9 if I understand it correctly, is improvements and doing a cost 10 benefit analysis.
11 So there might be usefulness gained from looking at I 12 11-50 as surrogate examples for the review process, not only t
13 from the standpoint of the utilities submitting to you, but
( 14 also your process of deciding what is the cost benefit- arialysis 15 that you are going to have to do to make a case for five '
16 individual plants.
17 MR. SHERON: For those five plants, you might be able 18 to use it. But you know, I cannot make the argument that what .
19 you glean from SURREY analyses applies to North Anna or some 20 other plant.
21 But anyway, the role in the regulatory process, I 22 think that this kind of sums it up right here. One is that it i
23 gives the staff an independent assessment of the risks from i
24 nuclear power. And although it has been subject to a lot of 25 criticism, particularly in the uncertainty analysis and the way l
Heritage Reporting Corporation ,
(202) 628-4888
218 il that'it treated uncertainties, that I think that it does
.( )
2 represent a state of the art of assessment of risk, at best as
^
3 the technical community out there at least at the laboratories 4 understands it.
5 And as we said before, it has a number of uses. It 6 provides a technical data base for us for the IPE. We will be 7 looking at it for accident management.
8 DR. KERR: When you refer to a data base, are you 9 referring to the results or to the data that you used?
10 MR. SHERON: I think both, both. We would look at 11 the results.
12 DR. KERR: I personally would not think of data as
. 13 being the output of a PRA.
() 14 MR. SHERON: No. But I am sure that to do these that 15 they had to collect a lot of data on the plant design. ;
16 DR. KERR: What you are referring to. ,
17 MR. SHERON: Both, again. ;
18 DR. KERR: Okay. I just wanted to know what you 19 meant by data.
20 MR. SHERON: We would use both, both the results as 21 well as the data that was collected. Again it is being used on i
- 22 a container performance initiative, the safety goal 23 implementation, and generic issue resolution. And it also j 24 helps us prioritize and focus our research in particular severe 25 eccident research, knowing what to work on and the like.
l O
v Heritage Reporting Corporation (20?.) 628-4888
219
(} 1 MR. MICHELSON: Would you refresh my memory again as 2 to how you are intending to handle external events from the 3 . viewpoint of 11-50. You have not included external events.
4 MR. SHERON: The final version of 11-50 will include 5 external. events, is that correct?
-6 MR. HOUSTON: For two plants.
7 MR. MICHELSON: But see, you can draw from false 8 conclusions if you have not included external events. Until 9 you know what contributions that they can potentially make to 10 this, you could conceivably draw false conclusions from the 11 results thus far.
12 So how are you going to make sure that you are not 13 led into that kind of a trap?
() 14 MR. SHERON: I guess that I do not have an answer.
15 MR. MICHELSON: For two plants, you might not find a 16 real problem when you do external events on two plants. You 17 might begin to get some indication of where the problems can !
1,8 be. But if yon just simply. ignore external events, you have l ,
19 ignored a potentially large'contrirutor to risk.
20 MR. HOUSTON: I do not think that they are proposing 21 to ignore them, j I
22 MR. MICHELSON: They are going to include them on two 23 plants, 1 thought you said.
24 MR. HOUSTON: But the IPE will get them on all plants i
, 25 eventually.
l Heritage Reporting Corporation (202) 628-4888 l
l O.
1 l
'l 220-() 1 MR. MICHELSON: Well, yes. But I am just looking at-2 11-50 now, because that is the current subject.
3 MR. SHERON: I do not know, other than to be cautious 4 when you use the results at this time.
5 MR. MICHELSON: Well, it could be. You do not know 6 how to interpret the results unless you have looked at the full 7 spectrum of severe accidents, for instance. ,
8 MR. SHERON: I would not go that far. I mean I think 9 that there is a lot to be learned from internal initiators.
10 MR. MICHELSON: How are you going to consider human-11 factors? That is the same problem.
12 MR. SHERON: You lost me. ,
t 2
13 MR. MICHELSON: Are you considering human factors in 14 11-50? You are plugging them in, I think using some rather 15 generic numbers. [
16 MR. ELTAWDA: In certain cases, we have done actual.
17 MR. MICHELSON: You have no data base for a lot of 18 what you are doing on human factors in 11-50. >
19 DR. KERR: Carl, since I want to get to some 20 discussion on IPE later on.
21 MR. MICHELSON: Okay. When we hit 11-50 again, we 22 will get there.
23 MR. SHERON: This is just a real quick summary of 24 severe accident research. I do not want to get into it in any 25 depth, because this is really not my area. I am doing this for O Heritage Reporting Corporation (202) 628-4888
221' h 1 Dr. Spels.
2 You know- we just wanted to point'out that after the 3 TMI-2 accident, that research has indeed provided the data base 4 and models for a number of important issues, fission product 5 release, containment loading, hydrogen, poor concrete 6 interactions, containment performance testing, natural 7 circulation effects on the primary service, and core melt.
8 progression.
9 And to point out the future research efforts, we will 10 focus on specific issues. These are the direct containment 11 hearing issue, including the effects of natural circulation.
12 The liner melt through issue, the melt spreading and 13 containment shell failure in MARK-Is. And research data and
() 14 models to assess accident management strategies. And then as I 15 said before, the longer term confirmatory research on direct 16 containment heating sequences, refinement of hydrogen behavior 17 models, core melt progress, core concrete, refined fission 18 product, temperature properties, and further model assessment 19 and refinement.
20 (Continued on next page.)
21 22 23 24 25 O Heritage Reporting Corporation (202) 628-4888
i l
222 (Slide.)
1 2 MR. SHERON: My last viewgraph is an example an issue 3 and how long-term and near-term research interact with it. And 4 you can take the issues that are associated with a large rack 5 PWR which are the' containment failure modes which are basically 6 direct containment heating, hydrogen-burning detonation and 7 late failure by core concrete.
8 And then there is the issue container performance and 9 accident management strategies such as the depres'surization of 10 the primary system.
11 There is a lot of other accident management 12 strategies which basically I think right now in the back-end 13 deal with recovery actions that an operator may be faced with.
() 14 But for this exernple research to address the issue for direct ,
15 containment heating is the probability of high pressure melt !
4 16 ejection, and this is whether you get natural circulation and a 17 failure someplace else in this system.
18 This business of a cutoff pressure for high pressure i
19 melt ejection. You know, is there such a thing as a cutoff 20 pressure? Is it something that's a more gradual decrease in 21 terms of the energy available to push the core out of the 22 vessel?
23 Management through depressurization and the 24 consequences of this. We just had a presentation yesterday, 25 and every time I look at it, it gets a little more complicated O Heritage Reporting Corporation (202) 628-4808
. ~
223
. ) 1 and hairy as to the benefits of depressurization.
2 And then analysis of performance tests on concrete 3 containments, the.one six-scale tests that were done.
4 So I think what we're trying to point out here is
~5 that there are a number of-Issues that are going to be around-6 awhile, and the research program is going to be directed to try ,
7 and focus in on those issues that we need to get answers to.
8 Dr. Siess.
9 DR. SIESSa' I find it interesting that the reference 10 there to the tests of the concrete containment, but no '
11 reference to a test on the steel containment, and we've got a 12 lot of steel containment out there.
13 MR. SHERON: Yes, it's an example and I am not trying (O
,/ 14 to, you know --
15 DR. SIESS: You mean the list is not complete?
4 16 MR. SHERON: No, it's just an example of an issue.
S.
17 DR. SIESS: I'm talking about the information that's 18 up there on research to address this issue of direct ;
19 containment heating. That's not a complete list of re' search
) 20 that relates to direct containment heating.
1 21 MR. SHERON: No, I agree. There is the steel r
! 22 containment issue here and the test that was done on that.
21 DR. SIESS: It's mislaading.
24 MR. SHERON: Okay.
25 DR. SIESS: I have a feeling people are trying to i
(
l Heritage Reporting Corporttion l (202) 628-4888 t
l'
224 1 sweep that steel containment test under the rug for some
( })
2 reason.
3 MR. SHERON: No, I don't think so.
4 DR. SIESS: Anybody else who wants to talk about it?
5 MR. SHERON: Well, I can certainly pass that on to 6 Mr. O' Lotto and let him come down.
7 DR. KERR Does that complete your presentation?
8 MR. SHERON: Yes. Did you want me to speak very 9 briefly on the A-44?
10 DR. KCRR: I think first I will ask if there are 11 further questions generally from members of the subcommittee, 12 and I think probably, Brian, the answer is no at this point 13 with the schedule we have.
/~
(_) 14 MR. S!!ERON : Okay.
15 DR. KERR: Do you have further questions of Brian on 16 the IPE?
17 (No response.)
18 DR. KERR I hear none. I assume you have dealt with 19 uncertainties, and there are no questions.
20 MR. SHERON: If I could just close by saying I 21 understand the --
22 DR, KERR: By all means.
23 MR. SHERON: It's the intention of the full committee 24 to write a letter to the staff. I guess I would close by 25 saying that we have struggled with this now a number of years, N-) Heritage Reporting Corporation (202) 628-4888
i l
225
( ). 1 both with trying to reach agreement with IDCOR on the methods [
-2 and the like, and then in light of the reorganization in the 3 staff to get everybody back up to speed and on board, Dr.
4 Murley is satisfied with this package. Mr. Beckjord is 5 satisfied. We have been to the CRGR.
6 I think we, in general, agree with their conclusions 7 which you have seen, their recommendations, and I think the 7 8 bottom line is that we think it's appropriate at this' time to 9 go out with a letter and get the interaction with the industry 10 started.
11 I'll be the first one to admit that there is f 12 unanswered questions with this package on how we are going to 13 do certain things and the like. I can't look into the crystal 1-() 14 ball and anticipate every outcome or every conceivable 15 question. But I think the only way we are really going to get -
16 this going is to start it and to get the industry involved and 17 interact with them, you know, and try and answer their ,
s 18 questions and the like.
19 It's not our intent to make them do work twice or to 20 spin their wheels. But, you know, they need to get started and 21 they need to think about it so we can start dealing with them 22 on a technical level. .
l 23 And so for that I would hope that the committee 24 would, you know, consider recommending that we get the letter 25 out, get the process started and encourage the continued (
I
) O , Heritage Reporting Corporation (202) 628-4888 i
1
q_ .__
226
( )' 1 interaction both with the ACRS as well as the industry to sort 2 out these remaining questions.
-3 Dr. Siess.
4 DR. SIESS: I would like to look into your crystal 45 ball just a little bit and estimate three or four years down 6 the line. 'ssuming this goes out, three or four years down the 7 line how many plants would you expect to have made -- to have 8 done a PRA?
9 MR. SHERON: The last estimate I saw from -- I think 10 NRR did a survey a couple of months ago, and the numbcr I saw 11 was we would expect about 50 plants out there have some sort'of d
12 a PRA either in progress or already completed.
13 DR. SIESS: That's out of the 80?
14 MR. SHERON: Out of the 80. So we are guessing 15 initially that there would be approximately 50 PRAs and maybe j- 16 30 IPEMs or something else.
17 DR. SIESS: Thank you.
! 18 MR. CORRADINI: According to your schedule then, when
'~
19 were you planning to issue the letter?
20 MR. SHERON: The schedule right now I guess would be i
! 21 that we are in the process of revising it to reflect the CRGR 22 comments and also the comments we got today from the 23 subcommittee. We present this to the full committee on May 24 6th. Any comments, any further comments we receive there we'll
- 25 try and reflect in the letter.
(
Heritage Reporting Corporation (202) 628-4888 l
l
i 227
) 1-
~
I guess.we would-like to try and getLthis down to the 4 2 Commiss1on some time by about the middle or latter part of May <
r 3 for their consideration. 'Whether they need a briefing'on it or j 4 the like, I don't know. ,
,5 I guess, if everything goes we11, I would expect j
- 6 hopefully to see the Commission give us approval to see the_ !
7 letter by about maybe mid June, and then we would be proposing 8 to conduct the workshop sometime say around the middle of 9 September, early September, have them finished.
10 MR. CORRADINI Then 60 days from that is when the
, 11 clock -- !
12 MR. SHERON: We're going to propose 60 days from that 13 would be when they would come in and tell us what their plan is
- O 14 for conducting their IPE. i 15 DR. KERR
- Any further questions of Mr. Sheron? ;
i.
16 Mr. Davis.
1 17 MR. DAVIS: Brian, do you think it's possible that [
18 the back-end phenomenological is ms will become so 1
- j. 19 controversial that this whole process might be held up until
! 20 those get resolved?
21 MR. SHERON: No. As a matter of fact, I hope it was f
3 22 the opposite. And that was that we did not want to see the t i 23 whole IPE process get held up because of back-end uncertainties
[ 24 and technical differences among the experts around the country.
25 So for this reason is why we left the back-end !
i.
(:) Heritage Reporting Corporation i
(202) 628-4888 t
228-
.( ) 1 analysis sort of purposely nebulous or open to the utilities.
2 In other words, all we are saying is -- again, you know, go 3 through the back-end analysis, understand it, understand where 4 the uncertainties are, understand that there are technical 5 differences of opinion, okay? But you have to be the judge of 6 what it is you think is the right thing to do for your plant.
7 Farouk reminded me during the break, you know, the 8 whole purpose is, you know, the staff is going to be looking at 9 these. We have a research program. Somewhere down the road we 10 hope we will get more definitive information that people can 11 make a judgment at that time.
12 But if there is something that you see that you can 13 do or want to do in the interim, okay, without waiting for the 14 NRC to do its thing with its research, you know, by all means 15 go ahead. We are not stopping you. We want you to do what's 16 safest for your plant.
17 But we're not saying that if you choose not to 18 depressurize your plant for high pressure direct containment 19 heating that that's unacceptable at this time. We're saying we 20 recognize, we understand there is a big uncertainty. But as 21 long as you acknowledge that you recognize that that 22 uncertainty exists and this is what, in your value judgment, 23 Mr. Utility, that you felt is the best thing to do at this 24 time, you know, fine.
25 MR. DAVIS: Well, briefly, I am a little bit O Heritage Reporting Corporation (202) 628-4888
m .
229
'() _ 1 concerned about since you tied these two together there may be 2 continuing delays in getting the Level 1 analysis underway.
3 You know, I've looked at about 35 Level 1 PRAs, and
'4 in about five or six cases, some rather significant 5 vulnerabilities have been found at plants, and they have been 6 fixed. And those vulnerabilities were found without doing 7 anything about the containment. They were the front-end 8 problems, and there may be more plants out there that have 9 those kinds of problems, and it seems to me like we want to get 10 out there and find them as soon as we can, and we don't need 11 the back-end analysis to do that.
12 MR. SHERON: We're not saying that they should stop 13 doing things to fix the front-end vulnerabilities if they find O( / 14 them just based on the back-end analysis.
15 If they find something, we're saying, you know, it's 16 your plant. You should fix it right away. You know, as we 17 said, we would expect you to expeditiously fix anything that 18 you found.
19 Also keep in mind in this back-end analysis, there 20 are a lot of concerns with the back-end analysis. For example, 21 on containment heat removal systems and the like which we want 22 them to look at, which may not be associated with any phenomena 23 that we have uncertainties on, but it may just be a hardware 24 vulnerability and the like.
25 That's part of the back-end analysis that we want O Heritage Reporting Corporation (202) 628-4888
230
( ) I them to look at. Okay, it's.not just all these exotic 2 phenomena that we don't have a handle on, but also some of the 3 hardware and the systems that are used to make sure the 4 containment stays in tact. And if they find vulnerabilities 5 there, that's also a good thing to do in a back-end analysis.
6 MR. DAVIS: But they don't need the resolution of 7 these issuee to do tuat.
8 MP, SHERON: That's correct.
9 DR. KERR: Any further questions of Mr. Sheron?
10 (No response.) ,
11 DR. KERR: Well, again thank you for an excellent 12 presentation and for your durability.
13 I don't think we need to record any further part of
( 14 this, but I do want to go into Executive Session.
15 (Whereupon, at 3:30 p.m., the meeting was 16 concluded.)
17 18 19 20 21 22 23 24 25 O Heritage Reporting Corporation (202) 628-4888
1 CERTIFICATE ,
2 3 This is to certify that the attached proceedings before the 4 United States Nuclear Regulatory Commission in the matter-of: ,
5 Names MEETING REGARDING PROPOSED GENERIC LETTER FOR INDIVIDUAL PLANT EXAMINATIONS (IPEs) ~'
6 7 Docket Numbers f 8 Place: Washington, D.C.
9 Date: April.26, 1988 10 were held as herein appeais, and that this la the original 11 transcript thereof for the file of the United States Nuclear i
12 Regulatory Commission taken stenographically by me and, [
13 thereafter reduced to typewriting by me or under the direction 14 of the court reporting company, and that the transcript is a
() 15 true and accurate record o the foregoing proceedings.
16 /S/ o% l>dk L l
17 (Signsture typed): Joan Rose i 18 Official Reporter 19 Heritage Reporting Corporation i 20 ,
- 21 ,
, 22
[
, 23 r l 24 ;
25
($) Heritage Reporting Corporation L
l (202) 628-4888 i i
i
INTRODUCTORY STATEMENT BY THE SEVERE ACCIDENTS SUBCOMMITTEE CHAIRMAN APRIL 26, 1988 g The meeting will now come to order. This is a meeting of the Advisory Comittee on Reactor Safeguards Subcomittee on Severe Accidents.
I am W. Kerr, Subcomittee Chairman.
The other ACRS Members in attendance are: C. Michelson, P. Shewmon, C. Siess, D. Ward and C. Wylie.
Also in attendance are ACRS Consultants: I. Catton, M. Corradini, P. Davis, J. Lee.
The purpose of this meeting is to review the intended final version of the NRC Staff's proposed generic letter on Individual Plant Examinations (IPEs).
Dean Houston is the cognizant ACRS Staff Member for this meeting.
The rules for participation in today's meeting have been announced as part of the notice of this meeting previously published in the Federal Register on March 30,1988(FR10309).
A transcript of the meeting is being kept and will be made available as stated in the Federal Register Notice. It is requested that each
! speaker first identify himself or herself and speak with sufficient clarity and volume so that he or she can be readily heard. .
We have received neither written coments nor requests for time to make oral statements from members of the public.
(Chairman's Coments - if any)
We will proceed with the meeting and I call upon Themis Speis.
l O
ika
, ge t >
AC3S BETIXG ON THE IMPLEMENTATION PLAN FOR THE SEVERE ACCIDENT POLICY STATEMENT INDIVIDUAL PLANT EXAMINATION (IPE)
GENERIC LETIER O THEMIS P. SPEIS
~
DEPUTY DIRECTOR FOR GENERIC ISSUES l
1 OFFICE OF NUCLEAR REGULATORY RESEARCH BRIAN W. SHERON, DIRECTOR DIVISION OF REACTOR AND PLANT SYSTEMS OFF"CE OF NUCLEAR REGULATORY RESEARCH l
l APRIL 26,1988 1
,~
1 >
GENERIC Irnu CONCERNING INDIVIDUAL PLANT EXAMINATION FOR SEVERE ACCIDENT VULNERABILITIES
- 1.
SUMMARY
- 2. EXAMINATION PROCESS 3.
EXTERNAL EVENTS (DEFERRED ITEM)
- 4. MBTIIODS OF EXAMINATION
- 5. REIATIONSIllP TO USl A-45
- 6. BENEFITS OF PRA
- 7. SEVERE ACCIDENT SBQUENCE SELECTION
- 8. SEVERE ACCIDENT MANAGEMENT
(]
0.
SEVERE ACCIDENT PREVENTION AND MITIGATION FBATURES RBPORTS
- 10. DOCUMENTATION OF EXAMINATION RESULTS
- 11. LICENSEE RESPONSE
- 12. REGULATORY BASIS APPENDIX 1, GUIDANCE ON TlIE EXAMINATION OF CONTAINMENT SYSTEMS PERFORMANCE (BACK-END ANALYSIS)
APPENDIX 2, CRITERIA FOR SELECTING IMPORTANT SEVERE ACCIDEhT SEQUENCE APPENDIX 3, ACCIDENT MANAGEMENT APPENDIX 4, DOCUMEhTATION
("l APPENDIX 5. DECAY IIEAT REMOVAL VULNEPABILITY INSIGHTS ATTACIIMENT 1,11ST OF REFERENCES 2
I
? !
n U 1.
SUMMARY
o A 10 CFR 50.54(f) LETTER REQUESTING TIIE INDIVIDUAL PLANT EXAMINATIONS FROM ALL LICENSEES Il0LDING OPERATING LICENSES AND CONSTRUCTION PERMITS FOR NUCLEAR POWER PLANTS o BASIS FOR REQUESTING INDIVIDUAL PLANT EXAMINATIONS IS THE COMMISSION SEVERE ACCIDENT POIJCY ISSUED ON AUGUST 8,1985 (50 FR 32138)
- PLANT SPECIFIC PRAs EXPOSED RELATIVELY UNIQUE VULNERABILITIES TO SEVERE ACCIDENTS
- THE UNDESIRABLE RISK FROM THESE UNIQUE FEATURES IIAS BEEN REDUCED BY LOW-COST CHANGES VIA PROCEDURES OR MINOR DESIGN MODIFICATION.
n - ANALYSIS WILL BE MADE OF ANY PLANT THAT HAS NOT YET
. UNDERGONE AN APPROPRIATE EXAMINATION WHEN NRC AND INDUSTRY SUFFICIENTLY PROGRESSED TO DEFINE TIIE METHODS OF ANALYSIS o SINCE ISSUANCE OF THE POLICY STATEMENT TIIE STAFF INTERACTED EX"fENSIVELY WITH INDUSTRY AND DEVELOPED APPROPRIATE DOCUMENTATION FOR THE IPEs o PURPOSE OF IPE IS FOR THE UTILITIES TO:
IDENTIFY / UNDERSTAND TIIE MOST LIKELY SEVERE ACCIDENT SEQUENCES THAT COULD OCCUR AT THEIR PLWTS; EVALUATE /lMPLEMENT MEANS FOR IMPROVEMENTS DEVELOP AN AWARENESS FOR SEVERE ACCIDENT O DEHAviOR 5
s C-
- 2. EXAMINATION PROCESS o LICENSEE'S STAFF SHOULD PARTICIPATE IN ALL ASPECTS OF THE IPE S0 THAT KNOWLEDGE GAINED BECOMES AN INTEGRAL PART OF OPERATING, TRAINING AND PROCEDURE PROGRAM, o LICENSEES SHOULD CONDUCT SYSTEMATIC EXAMINATION OF PLANT DESIGN, OPERATION, MAINTENANCE AND EMERGENCY OPERATION T0:
- IDENTIFY PLANT SPECIFIC VULNERABILITIES (DESIGN AND O PR CEDURAL) TO SEVERE ACCIDENTS (FOR BOTH CORE DAMAGE AND CONTAINMENT PERFORMANCE); BOTH INTERNAL AND EXTERNAL INITIATORS ARE TO BE CONSIDERED (EXTERNAL INITIATORS ARE DEFERRED)
- QUANTIFY RESULTS OF EXAMINATION FOR THE SEQUENCES THAT CONTRIBUTE TIIE MOST TO THE TOTAL CORE DAMAGE OR LARGE RELEASE FREQUENCY
- UNDERSTAND WHAT COULD POSSIBLY GO WRONG IN A PLANT.
- IDENilfT AND EVALUATE MEANS FOR IMPROVING PLANT /
CONTAINMENT PERFORMANCE (VIA IIARDWARE ADDITIONS /
MODIFICATIONS, ADDITIONS TO PROCEDURES, TRAMING);
- - DECIDE WillCH IMPROVEMENTS WILL BE IMPLEMENTED AND SCIIEDULE FOR IMPLEMENTATION l
a
C IPE BENEFITS o PIANT-SPECIFIC PRA's YIELDED INSIGHTS TO UNIQUE SEVERE ACCIDENf VUIRERABIllTIES MADING TO IAW-COST MODIFICATIONS o UTILITY INVOLVEMENT IN UNDERSTANDING POSSIBB RANGE OF SEVERE ACCIDENT BEIIAVIOR IN ITS PIANT(s) AND THUS IS BLUER PREPARED TO PREVENT, AND/0R MITIGATE ACCIDENT PROGRESSION AND CONSEQUENCES O INC RPORATE THE PIANT SPECIFIC INSIGHTS (e.g., RECOGNITION OF ROLE OF PREVENTION AND MITIGATION SYSTEMS AND OPERATOR ACTIONS) AND IMPROVEMENTS (e.g., IIARDWARE, PROCEDURES)
RESULTING FROM THE IPE INTO A SEVERE ACCIDENT MANAGEMENT PROGRAM O
5 i
O 3. EXTERNAL EVENTS o COMMISSION SEVERE ACCIDENT POLICY DOES NOT EXCLUDE EXTERNAL EVENTS, o LICENSEES ARE ONLY REQUESTED TO PROCEED WITil THE EXAMINATIONS FOR INTERNAL EVENTS o STAFF IS WORKING WITII NUMARC FOR DEVELOPMENT OF ACCEPTABLE METIl0DOLOGY Oo QUESTION IS Il0W TO BEST llANDLE EXTERNAL EVENTS IN SEVERE ACCIDENT POLICY IMPLEMENTATION STAFF CURRENTLY EXAMINING EXTENT TO WHICII EXTERNAL EVENTS MUST BE TREATED WORK TO DATE INDICATES:
o SOME EXTERNAL EVENTS SIIOULD BE LOOKED AT BY ALL PLANTS o SOME EXTERNAL EVENTS ARE UNIQUE ONLY TO A FEW PLWTS o SOME EXTERNAL EVENTS MAY BE ACCEPTABLY TREATED BY EXISTING DESIGN BASIS.
o TIIE STAFF INTENDS TO EFFECTIVELY INTEGRATE ALL ONG0ING PROGRAMS DEALING WITIl EXTERNAL EVENTS (e.g., A-46 SDMP)
O S0 NO DUPLICATION OF EFFORT DY INDUSTRY WILL OCCUR 6
)
o b
EXTERNAL EVENTS (CONT.)
EXTERNAL EVENT ISSUE o MANY POSSIRLE SOURCES OF HAZARD o
CIIANGING CRITERIA (PIANTS NOT UNIFORMLY DESIGNED) o LEGE UNCERTAINITIES IN ESTIMATING FREQUENCY OF PRI EVENTS o NEED TO DETERlIINE PROPER MIX OF PROBABillSTIC AND DETERMINISTIC APPROACII FOR VARIOUS llAZARDS f) o NO PARAUTL OVERAIL INDUSTRY EFFORTS (i.e., SIMILAR TO THAT FOR INTERNAL EVENTS) YET IN PLACE O
i
O EXTERNAL EVENTS ' CONT.)
STEERING GROUP MISSION o RECOMMEND HOW BEST TO TREAT EXTERNAL EVENTS IN CONTEXT OF THE SEVERE ACCIDENT P0llCY o ENSURE THAT RECOMMENDED TREATMENT IS COORDINATED WITH OTHER AGENCY PROGRAMS RELATED TO EXTERNAL EVENTS AND N DUPLICATION OF EFFORTS RESULTS O
o COMPLETE TASK IN APPROXIMATELY 18 MONTHS
! O i
i 8 l 2
1
() 4. METHODS OF AXALYSIS o THE GENERIC Irnu SPECIFIES SEVERAL OPTIONS THAT COULD BE USED TO SATISFY THE EXAMINATION REQUIREMENTS IDCOR IPEMs FRONT END WITH STAFF ENHANCEMENTS +
CONTAINMENT PERFORMANCE ANALYSIS CONSISTENT WITH APPENDIX 1 LEVEL-I PRA + CONTAINMENT PERFORMANCE ANALYSIS.
CONSISTENT WITH APPENDIX I ( OR LEVEL-II OR -III PRA WITH "BACK-END" CONSISTENT WITH APPENDIX I)
OTHER SYSTEMATIC EVALUATION METHODS (STAFF PREVIEW MIGHT BE NECESSARY) i I
O l 1
e 0 IDCOR METHOD FRONT END l 0 IDCOR IPEMs COVER SEQUENCES INITIATED BY INTERNAL EVENTS, LOSS OF 0FFSITE POWER AND INTERNAL FLOODS.
O IDCOR IPEMs INVOLVE THE FOILOWING BASIC ELEMENTS OF LEVEL-I PRA:
- PIANT FAMIUARIZATION AND SYSTEMS NOTEBOOKS INCLUDES SUBSTANTIAL AMOUNT OF DESIGN. OPERATIONAL, AND PRA INFORMATION (SCHEMATICS, INTERFACES, DATA, CUT-SLTS (PWR IPEM ONLY))
O - ACCIDENT SEQUENCE DEFINITION AND QUANTIFICATION RECOGNIZES TlIE IMPORTANT ROLE PLAYED BY SUPPORT SYSTEMS. INTERDEPENDENCE AMONG SUPPORT SYS7 EMS IS ANALYZED TO IDENTIFY A SET OF SUPPORT STATE.
DATA ASSESSMENT AND PARAMEI'ER ESTIMATION INVOLVES INVESTIGATION OF DATA APPllCABILITY, AND ANALYSIS OF PLANT SPECIFIC EXPERIENCE (FAILURE RATES AND INITIATOR FREQUENCIES)
INTERPRETATION OF TIIE RESULTS i
10
.v ;
1
)
t > IDCOR IPEMs EVALUATION o TIIE IDCOR IPEMs TAKE ADVANTAGE OF AVAILABLE IDCOR REFERENCE PLANT ANALYSES, SIMILARITIES AMONG PLANTS AND UP-TO-DATE KNOWIEDGE OF SEVERE ACCIDENT PHENOMENA. TIIE IPEMs SYSTEM ANALYSIS IS LESS RIGOROUS THAN A FULL SCOPE PRA.
o TIIE IDCOR IPEMs WERE SUBMITTED TO THE NRC MAY 1986 o INITIALLY MODIFIED BY IDCOR IN DECEMBER 1986 TO ACCOMODATE NRC COMMENTS. FURTIIER MODIFICATIONS WERE q p RECEIVED IN APRIL 1987 o NRC EVALUATION OF TIIE IDCOR IPEMs IS COMPLETE o IT IS EXPECTED TIIAT THE LEVEL OF EFFORT REQUIRED IF A UTILITY USES TIIE IDCOR IPEM IS EQUIVALENT TO LEVEL-I PRA BUT IS LESS TIIAN IF A FULL-SCOPE LEVEL-Ill PRA WERE USED 0
u
.o t >
IDCOR IPEMs EVALUATION (CONT.)
(FRONT-END) o KEY AREAS OF NRC CONCERNS:
- BMPHASIS ON BOTTOM UNE NUMDERS, NAMELY CORE DAMAGE FREQUENCY.
TIIE STAFF EMPilASIZES IDENTIFICATION OF PLANT SPECIFIC YULNERADIIJTIES, IlSTING OF POTENTLAL AREAS FOR IMPROVBIENTS AND SELECTION OF MOST PROMISING IMPROVEMENTS FOR IMPLEMENTATION
- NO TEST APPLICATION OF Tile REVISED IPEMs
- ALMOST ALL TEST APPUCATIONS HAD TIIE BENEFlT OF A FULL SCOPE PRA WHICH IN THE STAFF'S OPINION COULD HAVE INFLUENCED THE IPE IDCOR BWR IPEM DOES NOT ASSURE A COMPIITE EVALUATION OF Tile NET IMPACT OF SYSTEM DESIGN AND OPERATIONAL MODIFICATIONS ON THE OVERALL PLANT RISK ESilMATE EXTREME CARE NEEDED IN MODEUNG SUPPORT STATES. Tile CRITERIA FOR DETERMINING WilEN MORE DETAILED ANALYSIS OF SUPPORT 57ATES IS NEEDED IS LFTf TO THE ANALYST o
STAFF EVALUATION PROPOSES SEVERAL ENilANCEMENTS TO ADDRESS O 1 TiiE ARtAS Or CONCERNS iN Ti!E rRONT-END 12 i
.D t >
IDCOR IPEMs EVALUATION (CONT.)
(BACK-END) o TIIE IDCOR BACK-END IPEMs ARE T00 NARROWLY FOCUSED AND WILL NOT PROVIDE UTILITIES WITIl INFORMATION NEEDED TO EXAMINE CONTAINMENT PERFORMANCE FOR THE FULL RANGE OF SEVERE ACCIDENTS o RELY VERY HEAVILY ON IDCOR VIEWS OF SEVERE ACCIDENT PHENOMENOLOGY AND THE IDCOR MAAP CODE o
SOME OF IDCOR VIEWS AND MET 110D0 LOGY ARE IN DISAGREEMENT WITH CURRENT RANGE OF ASSUMPTIONS w
I o NO RECOGNITION OF UNCERTAINITIES IN Tile PRENOMEN0 LOGY o OPTIMISTIC ASSESSMENT OF CORRECT OPERATOR ALTIONS TO MAINTAIN CONTAINMENT INTEGRITY DURING SEVERE ACCIDENTS o
LOOSE INTERPRETAT[0N OF Tile IPEM QUESTIONS IlY TIIE UTILITIES COULD RESULT IN A IACK OF SUFFICIENT ATTENTION TO MEASURES
' 3 MITIGATE THE C014 SEQUENCES OF AN ACCIDENT o APPENDIX 1 TO TIIE GENERIC LETI'ER PROVIDES ADDITIONAL GUIDANCE
() ON IIOW TO ASSESS THE CONTAINMENT PERFORMANCE N
- - -- - y - - - - -
t >
SCOPE /0BJECTIVES-STAFF GUIDANCE lBACK-END) o TIIE IPE SIIOULD PROVIDE BASIS FOR UTILITY'S
- APPRECIATION OF SEVERE ACCIDENT BEllAVIOR RECOGNITION OF ROLE OF MITIGATION SYSTEMS DEVELOPMENT OF ACCIDENT MANAGEMENT PROCEDURES o
TIIE IPE SIIOULD PROVIDE A REALISTIC EVALUATION OF CONTAINMENT PERFORMANCE o
!!ENCE STAFF GUIDANCE FOCUSES ON !!AVING UTILITIES:
(} -
EVALUATE CONTAINMENT FAILURE MECIIANISMS AND TIMING SIMPLY LOOK UP RELEASES FROM TABLES PREPARED ON Tile BASIS OF REFERENCE PIANT EVALUATIONS
- INTEGRATE SYSTEMS RESPONSE (PROBADIllSTICAILY) INTO SIMPLIFIED,BUT REALISTIC, CONTAINMENT EVENT TREES (A110h' FOR RECOVERY OR OTIIER ACCIDENT MANAGEMENT PROCEDURES)
- EXAMINE ALL CIASSES OF SEQUENCES %TF11 SIGNIFICANT PROBAlllLITY
- RECOGNIZE AND AS APPP,0PRIATE, ACCOUNT FOR UNCERTAINITIES O
14 j
w_-- - - . - ------.- -, ----.--
G i
- 5. RELATIONSHIP TO USI A-45 o A-45 ANALYSES IIAVE Sil0NN TEAT DECAY IIEAT REMOVAL FUNCTION FAILURE ARE SUFFICIENTLY PLANT SPECIFIC AND NOULD REQUIRE SYSTEMATIC EXAMINATION o PROPOSED STAFF RESOLUTION OF A-45 IS TO SUBSUME ISSUE INTO IPE o
TIIE PROPOSED GENERIC LETTER STATES TIIAT'TIIE IPE SIIOULD IDENTIFY TIIE VULNERAllLE ASPECTS OF DlIR FUNCTION
( ') o Tile PROPOSED GENERIC LETTER PROVIDES INSIGilTS GAINED FROM SIX LIMITED SCOPE PRA PERFORMED BY NRC UNDER TIIE A-45 PROGRAM O
15
.o 1 6
- 6. BENEFITS OF PRA o OTHER USIs AND GSis IP IPE IDENUFIES RESIDUAL VULNERABIUTY THAT IS TYPICALLY ASSOCIATED WITil USl OR GSl AND ITITUTY PROPOSES A MEASURE ACCEPTABLE TO TffE STAFF TO EUMINATE OR TO SUBSTANTIAlLY REDUCE TIIE VUIRERABIUTY, THE USI OR GSI MAY BE CONSIDERED CLOSED ON A A PLANT SPECIFIC BASIS o UCENSE RENEWALS PRA COULD BE USED T0 IDENTIFY RISK-SIGNIFICANT COMPONENTS Ah3 q SYSTEMS F0" WIUCII AGE-REIATED DEGRADATION CONCERN MUST BE ADDP2SSED D DURING TIIE UCENSE RENEWAL PERIOD o PETS HISE MANAGEMENT PROGRAM TIIAT CONTINUA 11Y ASSESSES TIIE SAFETY OF THE PIANT PROVIDES A POWERFUL TOOL TO Tile PIANT MANAGEMENT o SUPPORT FOR UCENSING ACTIONS PRA MlGilT BE USED TO JUSTlFY TECIINICAL SPECIFICATION CIIANGES o INTEGRATED SAFETY ASSESSEMENT PROGRAM ll OPITMlZES Tile TOTAL SAFETY AND EXPEDITES SCllEDULE TO IMPIIMEhT FTXES O
a _.
4 >
- 7. SEVERE ACCIDENT SEQUENCE SELECTION (SCREENING CRITERIA) o USED TO DETERMINE TIIE POTEhTIALLY IMPORTANT FUNCTIONAL SEQUENCES WlUCll REQUIRE CONTAINMENT EVENT TREE ANALYSIS. NOT TO DETERMINE IF PLANT IMPROVEMEhTS ARE REQUIRED o INCLUDE SEQUENCES WHICH CONTRIBUTE MORE THAN 5% TO TIIE PREDICTED CORE DAMAGE FREQUENCY o
INCLUDE SEQUENCES FOR WHICH THE CORE DAMACE FREQUENCY
( ) IS GP. EATER TIIAN 1E-6/ REACTOR YEAR o INCLUDE SEQUENCES USEFUL FOR IDENTITYING SYSTEMS WEAKhTSS OR OPERATOR ACTIONS BASED ON DLTERMINISTIC EVALUATIONS, ENGINEERING JUDGEMEhT AND CONSIDERATION OF UNCERTAINITIES IN TIIE PROBADluSTIC ANALYSIS o
INCLUDE SEQUENCES FOR WIIICII TIIE CDP IS GREATER THAN 1E-6/
REACTOR YEAR AND CONTAINMENTS FAIIS IN LESS TlIAN 12 !!0URS FROM TIME OF VESSEL PENETRATION o INCLUDE SEQUENCFS YlIICH IIAD TO CONTAINMENT BYPASS AND
() IIAVE A PROBADi!JiY OF OCCURRENCE GREATER TilAN lE-7/RXY u
l
~
' e, i I
- 8. ROLE OF SEVERE ACCIDENT MANAGEMENT o SEVERE ACCIDENT MANAGEMENT IS A PROCESS IN YluCH ACTIONS TilAT CAN PREVENT CORE DAMAGE OR MITIGATE TIIE CONSEQUENCES OF SDT ACCIDENT ARE IDENTIFIED, EVALUATED, INCORPORATED INTO A STRUCTURED PROGRAM, IMPlIMENTED AT A PIANT SITE AND ARE AVAllABLE TO TIIE OPERATORS AND PIANT MANAGEMENT IN THE EVENT OF SEVERE ACCIDENT o
SEVERE ACCIDENT MANAGEMENT ENCOMPASSES IIARDWARE, HUMAN, AND ORGANIZATIONAL PACTOPS o
IT PROVIDES DECISION MAKERS AT Tile PLANT A STRUCTURED PROGRAM FOR MANAGING A SEVERE ACCIDENT C
o PROPOSED GENERIC LETTER ADDRESSES SEVEP2 ACCIDENT MANAGEMENT AS FOLLOWS
- UTIUTIES ARE EXPELTED TO DEVELOP AN ACCIDENT MANAGEMENT PROG FOR PREVENTION OR MITlGATION OF RISK IMPOISANT SEVERE ACCIDENTS IDENTIFY MEASURES TilAT PLANT PEPSONNEL CAN AND SHOULD TAKE IN CASE OF SEVERE ACCIDENT. ASSESS AGAINST TILE CRITEPJA OP 10 CFR 50,59 AND, IF APPROPRIATB, SUBM117 FOR NRC RD'IEW IN ACCORDANCE VITU 10 CFR 50,90 UTIUTIES S110ULD PREPARE TO SUCCESSFULLY EXCUTE MEASURES TIIA FOUND T0 (1) PREVEST CORE DAMAGE, (2) PREVENT CORE DEBRIS FROM PENETPATING THE VESSEL OR, (3) MANAGE TIIE C0 SEQUENCES O o STAn' AND NUMARC DISCUSFINC TIIE NET.D To DLTEIDP AND IMPLEMENT ADDITIONAL SEVERE ACCIDENT MANAGEMENT PLtN IS
+
- 9. SEVERE ACCIDENT PREVENTION AND MITIGkTION FEATURE REPORTS (NUREG/CR-4920) o Tile KNOWLEDGE AND INSIGHTS GAINED FROM TIIE REVIEW AND ANALYSIS PERFORMED BY IDUSTRY AND NRC NERE COLLECTED TOGETIIER, AND SORTED ACCORDING TO APPIlCABILITY TO REACTOR TYPES o
IHGilLICIIT (NOT TO SPECIFY) PLANT FEATURES AND OPERATOR ACTIONS THAT WERE FOUND TO DE IMPORTANT TO RISK IN PREVIOUS STUDIES o
PilNT FEATURES AND OPERATOR ACTIONS o
V
- AVAIL 4DILITY OF SUPPORT SYSTEMS AND IDENTIFICATION OF NECESSARY COMPONENTS IDENTIFICATION OF [MPORTANT OPERATOR ACTIONS IDENTIFICATION OF PARAMETER FOR INITIATION OF MITIGATING SYSTEMS AND OPERATOR ACTIONS SURVIVABIlJTY OF EQUIPMENT EQUIPMENT CAPABILITIES, CAPACITIES, AND DURA 110N OF OPERADillTY
- ACCESSIBILITY OF EQUIPMENT o
NUREG/CR 1920 SUMMARIZING PAST EXPERIENCE IIAVE DEEN PROVIDED TO ACRS O
19
+
i >
- 10. DOCUMENTAION OF IPE RESULTS o PROVIDE THE BASES FOR THE FINDINGS IN A TRACABLE MANNER o TWO TIERS OF DOCUMENTATIONS o TIER-1 REPORTS THE RESULTS Of TIIE EXAMINATION TO NRC
- TILE LEADING SEQUENCES AND SCREENING CRITERIA
- TIIE CORE DAMAGE FREQUENCY AND THE LIKEllH00D OF LARGE RELEASE lHE MAIN CONTRiljuTORS FOR CDF AND CONTAINMENT FAILURE
- DISPOSTION OF CONTRIBUTORS o TIER-2 DOCUMENTS THE EXAMINATION THAT WOULD DE RETAINED BY LICENSEES
! o NRC WIII EVALUATE TIER-1 REPORTS AND MAY AUDIT TIER-2 O
20 l
..e i
O IPE RESULTS REVIEW o
PREPARE AN IPE REVIEW DOCUMENT FOR THE STAFF AND CONTRACTOR REVIEWERS TO INCLUDE:
- AREAS OF REVIEW DLTERMINATION OF ADEQUACY OF IPE RESULTS
- ACTION IRELS INTERPRETATION OF THE RESULTS SAMPLE EVALUATIONS O o IPE REVIEW DOCUMENT WILL BE MADE AVAILABLE TO ALL UTILITIES SHORTLY AFTER THE ISSUANCE OF TIIE GENERIC IATr I
21 l
+ ..
O IPE RESULTS REVIEW (CONT.)'
o TIIERE ARE 109 LICENSED PIANTS, REPLICATE PLANT WOULD REDUCE THE IPE SUBMI'ITAIS T0 30 o
6 PERSON-MONTil PER PLANT REVIEW 0F THE IPE SUBMflTALS INCLUDLNG REVIEW 0F ANY PROPOSED MODIFICATION o IPE SUBMflTALS ARE ESTBfATED TO DE OVER A PERIOD OF 3 YEARS o APPR0XIMATELY 16 PERSON-YEARS EFFORT PER YEAR , 8 C0hTRACTORS, 8 STAFF ( 4 FROM NRR & 4 FROM RES ). RES WILL PdVE Tile 12AD O
. o APPR0XIMATELY $1.4M PER YEAR ( 8 CONTRACTOR STAFF YEARS )
o IN CASE OF DISAGREEMENT, Tile STAFF WILL PURSUE FIX IN ACCORDANCE WITH THE BACKFIT RULE OR ORDER o TEAM CONCEPT- EACll PLANT REVIEW TO BE THE RESPONSIBILITY OF ONE NRC TEAM LEADER, SEVERAL PLANT SYSTEMS SPECIALISTS, AND PRA SPECIALIST o TWO TEAMS WILL DO ONLY IARGE DRY, ONE TEAM WILL DO MARK I & II AND ONE TEAM WILL 00 MARK llI AND ICE CONDENSER
- O i
i 22 I
+
a J,
i '
STAFF USE OF IPE RESULTS o
REVIEW 0F IPE RESULTS FOR THE FOLLOWING:
- TO ENSURE ADEQUATE ANALYSIS OF PLANT DESIGN AND OPERATIONS TO DISCOVER PARTICULAR VULNERABILITY TO CORE DAMAGE AND UNUSUALLY POOR CONTAINMENT PERFORMANCE FOR CONSISTENCY IN THE IDENTIFICATION AND TREATMENT OF LEA CORE DAMAGE SEQUENCES
- TO ENSURE SEQUENCES AND SEQUENCE FREQUENCIES ARE APPROPRIAT
- TO INDEPENDENTLY CONCLUDE TllAT DECISIONS ON NHETHER TO MAXE IMPROVEMENTS ARE JUSTIFIED
- TO ALLOW COMMISSION TO CONCLUDE THAT TILE SEVERE ACCIDENT POLICY HAS BEEN RESPONSIBLY LVPLEMENTED BY INDUSTRY O o CONSIDERATION YILL INCLUDE BOTH QUANTITATIVE MEASURES AND NON QUANTITATIVE JUDGMENT TO DETERMINE IF A PIANT IS ACCEPTABLE OR NOT o
IF NRC COSIDERATION IDICATES THAT PLANT DESIGN OR OPERATION COULD BE ENilANCED llY ADDITIONAL PROTELTION BEYOND NRC REGULATION, IMPLEMENTATION WOULD DE IN ACCORDANCE WITH 10 CFR 50.109 o
IF CONSIDERATION INDICATES PLANT DESIGN OR OPERATION SHOULD BE CHANGED TO MEET NRC REGULATION, LMPLEMENTATION WOULD BB WITHOUT REGARD TO COST EXECPT TO SELECT AMONG ALTEPSATIVES o
IPE RESULTS WILL BE USED TO IDENTIFY SEVERE ACCIDENT VULNERABIL]TY GENERIC TO CIASS OR SEVERAL CIASSES OF PLWTS
- THIS INFORMATION ON GENERIC VULNERADILITIBS WOULD DE USED TO EXAMINE 1F DEFICIENCIES IN THE REGULATION EXIST IF GENERIC DEFICIENCIES WERE IDENTIFIED, SAFTIT G0AL WOULD BE USED TO DETERMINE IF REGUIATION MODIFICATIONS WERE NEEDED 23
)
L$ect its O
l CRGR RECOMMENDATIONS !
o ISSUE THE IPE GENERIC LETTER WITH ALL ITS APPENDICES. B0Til l FRONT-END AND BACK-END ANALYSES ARE TO BE INCLUDED.
o ISSUE NUREG/CR-4920 SEPERATELY S0 IT WILL NOT BE CONSIDERED AS REQUIREMENTS.
o STAFF'S REVIEW DOCUMENT WILL BE DISCUSSED IN DRAFf FORM Q WITH LICENSEES AT WORKSB0P(S) o PURPOSB 0F THE WORKSHOP (S):
- PRESENTS TO LICENSEES ( NOT NEGOTIATE ) WELL DEVELOPED GUIDANCE ON HOW TO DO TIIE ANALYSIS, PARTICULARLY TIIE BACE-END INFORM THE LICENSEES HOW THE STAFF IS G0ING TO REVIEW TIIE IPE SUBM11TAIS
- SIIARPEN THE NRC GUIDANCE, WIIERE APPROPRIATE o ISSUE THE REVIEW DOCUMENT IN FINAL FORM, TO START THE CLOCK O
\
4 (N ,.
<b, SEVERE ACCIDENT INTEGRATED PLAN l BY THEMIS P. SPEIS, DEPUTY DIRECTOR l
l OFFICE OF NUCLEAR REGULATORY RESEARCH FOR ACRS SEVERE ACCIDENTS SUBCOMMITTEE MEETING APRIL 26, 1988
Y')
U BACKGROUND o AUGUST 8, 1985, "SEVERE ACCIDENT POLICY STATEMENT,"
50 FR 32138 o FEBRUARY 28, 1986, "IMPLEMENTATION PLAN FOR SEVERE ACCIDENT POLICY STATEMENT," SECY-86-76 o DECEMBER 1, 1986, STAFF REQUIREMENTS MEMO, CHAIRMAN REQUESTED PAPER ON INTEGRATION OF SEVERE ACCIDENT ISSUES o FEBRUARY 17, 1987, MEMO FROM EDO TO COMMISSIONERS ON PRELIMINARY PLAN FOR INTEGRATION OF SEVERE ACCIDENT ISSilES o JULY 15, 1987, STAFF BRIEFED COMMISSION ON A PLAN FOR CLOSURE OF SEVERE ACCIDENT ISSUES o DECEMBER 8, 1987, "MARK I CONTAINMENT PERFORMANCE PROGRAM PLAN,"
SECY-87-297 o FEBRUARY 9-11, 1988, BALTIM0RE MANAGEMENT MEETING ON SEVERE ACCIDENT ISSUES
!O
m ~
d
( )
SEVERE ACCIDE_NT INTEGRATED PLAN o PURPOSE:
TO PRESENT STAFF PLANS FOR INTE CLOSURE OF ALL SEVERE ACCIDENT IS o
OBJECTIVES:
TO PROVIDE AN UNDERSTANDING OF ACTIVITIES THAT ARE UNDER WAY T COMMISSION'S SEVERE ACCIDENT POLICY TO ASSURE THAT THESE ACTIVITIES WITH THE COMMISSION'S POLICY A (x, TO ASSURE THAT THE STAFF ACTIVITIES CONSISTENT AMONG THEMSELVES, HAVE A C OF ULTIMATELY LEADING TO IMPROVED PLAN AND ARE PROPERLY C0ORDINATED A NRC ORGANIZATIONS TO ASSURE THAT THE COMMISSION IS KEY TECHNICAL AND POLICY ISSUES, SOME OF NEED COMMISSION GUIDANCE OR APPROVAL
l
.O !
SEVERE ACCIDENT ACTIVITIES, o INDIVIDUAL PLANT EXAMINATIONS (IPE) o CONTAINMENT PERFORMANCE IMPROVEMENTS (CPI) o IMPROVED PLANT OPERATIONS (IPO) o SEVERE ACCIDENT RESEARCH PROGRAM (SARP)
. o ACCIDENT MANAGEMENT (AM) PROGRAM o NUREG-1150, "REACTOR RISK REFERENCE DOCUMENT" o SAFETY G0ALS o GENERIC SAFETY ISSUES o EXTERNAL EVENTS ,
o INTEGRATED SAFETY ASSESSMENT PROGRAM (ISAP) 0 SEVERE ACCIDENT POLICY FOR FUTURE PLANTS o SEVERE ACCIDENT CLOSURE 4
O
m t
ne eer rd i i
/ ei s g vco cl n AC l th i
nc tneer i rda t ei c nvcs oecc CSAI ' t m'
t.
t ;
n &m eva dgr i ag cno
&!!ar P t '
N A
L P
h c
c D soe E t rr T ipa c A o uh n R l
m syr t
G t cvgai E
. i y T t s f c Li e/ rt(
l paoef N el f a cl it mr I ao ab dI S T SC N E
P D
' ' I A _
s C
S I C A
E s R i
e E y t V f i E i l S tti u
^
' nnh eas
> dl r .
IPe 1 l
n u s E
_ V w e R i U v G e I rR l
r, F A ,.
S .c Ji' tt -
ae g
s s M n tet ncn l o enc a s a i ma n d
u t a nmcs d e i o
i n irv aoo v t vti in= tf r ot r nra nrp pae dan nl IPE oee cpl ml p IPO 7 i' 0
5 1
1 I
s sI C th nc eer S
L G M A
rda eic I
vcs ece SAi t 4
- ) l1 i llii1 , i e ill3!i
, 1l14 1ij!!) !ljj .A j l41A 1
O "
A st'ty"*:' '" O~
Objective:
Provide up-to-date assessment of severe accident frequencies and risks for five plants, reflecting data obtained since Reactor Safety Study Roles in Regulatory Process:
Independent staff assessment of risks Technical data base as input to:
IPE accident management containment performance initiatives safety goal implementation generic issue resolution Prioritization/ focus of research .
L . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
SEVERE ACCIDENT RESEARCH je- S
\--)
- BEGINNING IN 1980, AFTER Tile TMI-2 EVENT, RESEARCH HAS PROVIDED A DATA BASE AND MODELS FOR:
o FISSION PRODUCT RELEASE, TRANSPORT, DEPOSITION, a REVAPORIZATION o CONTAINMENT LOADING BY HIGH PRESSURE MELT EJECTION (HPE) o HYDR 0 GEN DETONATION AND BURNING o CORE / CONCRETE INTERACTIONS (CCI) THERMAL 8 CHEMICAL o CONTAINMENT PERFORMANCE TESTING o EFFECTS OF NATURAL CIRCULATION ON THE PRIMARY SYSTEM o CORE MELT PROGRESSION (EARLY STAGES)
FUTURE RESEARCH EFFORTS WILL FOCUS ON SPECIFIC ISSUES SUCH AS:
o CONTAINMENT FAILURE PROBABILITY BY DIRECT CONTAINMENT HEATING l (DCH) INCLUDING EFFECT OF NATURAL CIRCULATION o MELT SPREADING AND CONTAINMENT SHELL FAILURE IN MARK I'S o RESEARCH DATA AND MODELS TO ASSESS ACCIDENT MANAGEMENT STRATEGIES o LONGER TERM CONFIRMATORY RESEARCH ON:
DCH CONSEQUENCES REFINEMENT OF HYDR 0 GEN BEHAVIOR MODELS CORE MELT PROGRESSION (LATE STAGES)
CORE / CONCRETE INTERACTIONS REFINED FISSION PRODUCT HIGH TEMPERATURE PROPERTIES FURTHER MODEL ASSESSMENT AND REFINEMENTS S
d
O O O~
AN EXAMPLE OF AN ISSUE AND ITS ASSOCIATED NEAR AND LONG-TERM RESEARCH CONTAINMENT TYPE LARGE DRY PWR ASSOCIATED ISSUES o CONTAINMENT FAILURE MODES DIRECT CONTAINMENT HEATING (DCH)
HYDROGEN BURN / DETONATIONS LATE FAILURE BY CCI LOADS (OVER T8P) o CONTAINMENT PERFORMANCE o ACCIDENT MANAGEMENT STRATEGIES DEPRESSURIZATION OF PRIMARY SYSTEM RESEARCH TO ADDRESS ISSUE I o DCH i
PROBABILITY OF HIGH PRESSURE MELT EJECTION l (NATURAL CIRCULATION)
CUTOFF PRESSURE FOR HPE MANAGEMENT THROUGH DEPRESSURIZATION CONSEQUENCES l
l ANALYSES OF PERFORMANCE TESTS ON CONCRETE CONTAINMENTS (1/6 SCALE)
-