ML20153F896
| ML20153F896 | |
| Person / Time | |
|---|---|
| Issue date: | 05/05/1988 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1666, NUDOCS 8805110083 | |
| Download: ML20153F896 (155) | |
Text
.
b
~
3R l' \\ A UNITED STATES O
t NUCLEAR REGULATORY COMMISSION 1
.-=============================================.=============
UNITED STATS liUCLEAR REGULATORY COMMISSIOli ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 1
In the Matter of:
)
)
)
337th ACRS MEETING
)
l l
AFTERNOON SESSION t
LOCATION:
Washington, D.,C.{3I"
/ ' ' T'O f J ' ?" '
E PAGES:
112 through 229 5~ t I
.-j j
1 30 f
0i i~h'E'S h'@'i /tM0 7 h DATE:
May 5, 1988 3
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
't I
l HERITAGE REPORTING CORPORATION j
Opdsf Raporter O
1224 L Street, N.W., Seite 640 l
Wanhington, D.C. 20005 (202).6M r
'Ik' 3 880505 T-1666 DCD
i PUBLIC NOTICE BY THE 2
UNITED STATES NUCLEAR REGULATORY COMMISSION'S 3
ADVISORY COMMITTEE ON REACTOR 3AFEGUARDS 4
5 6
7 The contents of this stenographic transcript of the 8
proceedings of the United States Nuclear Regulatory 9
Commission's Advisory Committee on Reactor Safeguards (ACRS),
10 as reported herein, is an uncorrected record of the discussions 11 recorded at the meeting held on the above date.
12 No member of the ACRS Staff and no participant at 13 this meeting accepts any responsibility for errors or 14 inaccuracies of statement or data contained in this transcript.
15 16 17 18 19 20 21 22 23 24 25 Heritage Reporting Corporation (202) 629-4888
.e 112 1
UNITED STATE" NUCLEAR REGULATORY COMMISSION 2
ADVISORY COliMITTEE ^3.;dACTOR SAFEGUARDS 3
4 In the Matter of:
)
5
)
)
. 6
)
337th ACRS MEETING
)
7
)
8
- Thursday, AFTERNOON SESSION May 5, 1988 9
Room 1046 10 1717 H Street, N.W.
Washington, D.C.
20555 11 The meeting convened, pursuant to notice, at-1:03 12 p.m.
13
()
BEFORE:
DR. WILLIAM KERR, Chairman 14 Professor of Nuclear Engineering Director, Office of Energy Research 15 University of Michigan
. Ann Arbor, Michigan 16 ACRS MEMBERS PRESENT:
17 1
DR. FORREST J. REMICK 18 Vice Chairman Associate Vice President for Research 19 Professor of Nuclear Engineering i
The Pennsylvania State University 20 University Park, Pennsylvania 21 DR. IIAROLD W.
LEWIS Professor of Physics 22 Department of Physics i
University of California 23 Santa Barbara, California 24 (Appearances Continued on Page 2) 25
(
Heritage Reporting Corporation (202) 628-4888
_--,, ~ -.
,,... _, -.,, -.,.,,, _,, ~,
,--.as
d 113 0
0'1 1
1 APPEARANCES (Continued):
2_
ACRES MEMBERS PRESENT_(Continued):
1 3
DR. DADE W. MOELLER Professor or Engineering in Environmental Health 4.
Associate Dean for Continuing Education
. School of Public Health 15 Harvard University Boston, Massachusetts
.g DR.-PAUL G. SHEWMON 7
Professor, Metallurgical Engineering Department Ohio State _ University, 8
Columbus, Ohio 9
CHESTER P.
SIESS Professoi Emeritus of Civil-Engineering 10 University of Illinois Urbana, Illinois 11 DR. MARTIN J.
STEINDLER 12 Director, Chemical Technology' Division Argonne National Laboratory.
13 Argonne,' Illinois
.O 14 MR. CHARLES J. WYLIE.
Retired Chief Engineer 15 Electrical Division Duke Power Company 16 Charlotte, North Carolina i
17 MR. DAVID A. WARD Research Manager on Special Assignment 10 E.I. du Pont de Nemours & Company Savannah River Laboratory i
19 Aiken, South Carolina 20 MR. CARLYLE MICHELSON Retired Principal Nuclear Engineer i
i 21 Tennessee Valley Authority l
Knoxville, Tennessee, and, 22 Retired Director, Office for Analysis and Evaluation 7
of Operational Data 23 U.S.
Nuclear Regulatory Commission Washington, D.C.
[
24 25
)
l Heritage Reporting Corporation I
i l
(202) 628-4888 L
i e
i
114
('/)
1 A F_ T_ g R N Q Q N g g g g I, Q N
'~
2 (1:03 p.m.)
3 CHAIRMAN KERR:
The integrated safety assessment 4
program, Roman Numeral II, Report of ACRS Subcommittee 5
Chairman, Mr. Siess.
6 DR. SIESS:
Tab 4, gentlemen, has all the information 7
you will need to follow the discussion.
It starts off with a 8
summary report, status report prepared by Sam.
For those of 9
you who have followed ISAP and ISAP-II, I will start out 10 briefly by saying that the staff has prepared a SECY document 11 providing recommendations to the Commission and that their 12 recommendation now is that they not go ahead with 1 SAP-71, not 13 try to implement it.
b 14 For those of you that are not that familiar, Jet me 15 give you a little background.
A number of years ago, the staff 16 undertook something called Systematic Evaluation Program which 17 was to look at the older plants to see how they compared with 18 the current level of safety as provided by the current 19 regulations.
20 Now, there is an implicit assumption that safety is 21 related to regulations.
22 They picked nine or ten plants.
I never can forget.
23 I think it was ten originally, and ones of them got shut down, 24 Humboldt Bay, of the old plants, the Yankee-Rowe, Lacrosse, San 25 Onofre 1, that class of plants, plus a few more recent ones O
Heritage Reporting Corporation (202) 628-4888
\\.
115 O
1 that-didn't have full term licenses.
2 And they developed a group of, a list of issues, of 3
requirements that these plants might not meet.
I forget.
4 There were several hundred of them.
And then they narrowed 5
that down to what, 127 items, 147, and they looked at each of
'6 these plants to see how well they met those requirements and
'7 when they didn't they looked at a PRA or a simulation thereof 8
to see whether it made any difference.
9 And that was finished a few years ago, I mean that 10 phase, and the original intention was to go back through all 11 the plants, to start with the oldest ones and work their way up 12 and compare them against the current regulations, using risk 13 analysis of some sort to decide whether fixes need to be made O
14 and backfit them if necessary.
15 But when they finished the SEP, they decided that it 16 really wasn't worth it, going through that same process with 17 all the plants.
They hadn't found that much of a problem, and la they thought the plants would be getting better aa time went on 19 anyway.
So they took that long list of issues and narrowed it 20 down to 47, I believe it was, which were the ones they had 21 found were most important in the old plants, and said now for 22 the next round let's just look at those issues.
23 And then they decided rather than continuing with the 24 SEP with tnose issues, they would look at a more voluntary 25 program which they called ISAP -- Integrated Safety Analysis O
lieritage Reporting Corporation (202) 628-4888 i
116 g
U 1
Program.
2 MR. DURAISWAMY:
That's "Assessment."
3 DR. SIESS:
Safety Assessment Program, because the 4
last stage of the SEP was an integrated safety assessment.
And 5
they said we'll just take these 47 items and any plant that
'6 wants to will look at these and will use risk analysis to see 7
how it works.
They never inaugurated it.
All the Commission 8
ever approvrd was a pilot ISAP program to see whether it would 9
work and what resources would be required.
10 Now, they got two quick volunteers for the pilot ISAP 11 or one quick volunteer with two plants.
Northeast Nuclear 12 volunteered Haddam Neck and Millstone 1, both of which had been 13 SEP plants, for this continuing ISAP thing.
7,I t'~#
14 The Commission, as I said, this was a pilot program.
15 You've heard a report from Millstone and Haddam Neck, from 16 Northeast Nuclear.
They were extremely enthusiastic about it.
17 They wanted it intended to Millstone 2 and 3.
They liked the 18 idea.
l')
So the staff was asked to develop an ISAP program to 20 try again I guess and this was called ISAP-II.
I may not have 21 all the details straight.
22 MR. DURAISWAMY:
That's a follow-on program.
23 DR. SIESS:
Yes, it was the follow-on program, called 24 ISAP-II.
And the staff developed a program which was somewhat 25 different than the ISAP-I.
It didn't include the SEP items.
O Heritage Reporting Corporation
[
(202) 628-4888 i
i
.j
-117 l
1 It did require PRA and it said we'll get together with the 2
licensee and all the improvements he wants to make to his plant i
3 and all the improvements we want_him to make to his plant, and 4
that was USIs and GIs, plant specific-requirements or whatever, 5
will all be looked at in terms of risk, they will be assigned 2 6 priorities, high, medium or low, and we'll negotiate a schedule "7
with the licensee.
And for the major items, we'll put that 8
schedule in a license amendment.
9 Now, this is very similar to the integrated living 10 schedule except that the use of PRA would permit some of these 11 items to be eliminated if they could be shown not to be 12 significant to risk, just that some items were eliminated from 13 SEP.
That was part of the carrot.
It still required a PRA, O
14 still required a license amendment.
15 That program, a description of it was sent out to 16 everybody in a generic letter, asking them whether they were 17 interested.
And no commitment, just are you interested,.would 18 you like to have more information?
If you don't like it, what 19 don't you like about it?
And so forth.
I think most of you 20 have seen that questionnaire.
21 And the staff got back responses that go something as i
22 follows.
Six utilities -- well, let's face it, there were 56 23 utilities out there and 50 of them replied, responded.
I think 24 by now you've got more than 50.
And you know, some that didn't 25 respond were like La Crosse.
Since they were shut down, they
(
i Heritage Reporting Corporation I
(202) 628-4888 2
118 e
1 didn't see much point in it.
2 Six utilities representing 18 units said they were 3
definitely interested and 29 utilities representing 52 units 4
said they were possibly interested and would like to have more 5
information.
Eighteen utilities indicated they were not 16 interested.
And some of them returned just the questionnaire
$7 with some comments and some wrote really long letters 8
explaining why they were or were not interested.
Some of the 9
lack of interest was they didn't like the license amendment, 10 which has since been taken out of it.
11 Some of the concerns I think are simply uncertainty 12 about how it would be run, about how the PRA would be reviewed 13 and how the schedules would be negotiated.
It is a completely 14 new concept to most people, and I think with a lot of 15 uncertainty.
16 People that were possibly interested, in two cases 17 the staft met with people and cleared up some of those 18 uncertainties.
19 As I read between the lines, I think some of it is 20 just plain suspicion as to how it would be, or fears about how 21 it might be implemented.
And as a result of that survey, a 22 decision was made that the continuing of ISAP-II would require 23 an expenditure of resources by NRC that would not be justified 24 in view of the number of utilities that would participate and 25 benefit.
O Heritage Reporting Corporation (202) 628-4888
119
('
1 The actual wording I hava in front of me, the number 2
of utilitios that ultimately would participate in and benefit 3
from ISAP-II do not represent a majority, a statement I find 4
hard to believe because nobody knows now who would ultimately 5
participate and the idea of majority sort of bothers me.
,6 And the conclusion is, because of the expenditures 7
and the likely relatively low number of ultimate participants 8
less than a majority, ISAP-II becomes a lower overall priority 9
item among the NRR's programs and it is recommended not to 10 continue.
11 That's the situation.
This is going up to the 12 Commission as a recommendation from the staff and the question 13 now is what advice to we want to provide to the Commission at 14 the time they receive this and how do we go about it.
15 It seems to me there are three possible courses of 16 action.
One is to say the staff is the best judge of their 17 resources and not disagree with their conclusion.
18 Another would be the other extreme, say that we still 19 think ISAP-II is a great idea, it's the way we ought to run our 20 business, it's the way we ought to make plant changes and that 21 it's worthwhile even if only two utilities participate and we l
22 think that actually more would, and try to persuade the 23 Commission to continue to pursue this.
24 (Continued on Page 10) i 25 l
Heritage Reporting Corporation (202) 628-4888 l
E
120
(~}
1 (Continued from previous page) 2 DR. SIESS:
That would lead to what I think would be 3
sort of an intermediate position that would say please go ahead 4
with the workshop and the educational process, explore this 3
further with the industry, and find out just what number you 6
might get to participate with a revised program, and one 7
without the license amendment, which they really haven't had a 8
chance to say, vote on.
9 I think they are giving up too soon.
I think that 10 there are a lot more utilities that would pick up on this, and 11 if they understood it thoroughly, a lot more would too.
12 Now, staff is not affected in this, they think if 13 some of them understand it, they might go the other way, since O
14 they are facing numbers that are going to stay small.
15 We made no decision in the subcommittee meeting, we 16 are not supposed to.
We didn't reach a consensus.
We will be 17 coming before you in connection with the IPE, some proposals 18 that are not in the draft letter that you have with the stuff l
19 Dave Ward has proposed.
20 There are a couple of things that would I think have 21 a bearing on the ISAP question.
One is, if everybody had a PRA 22 I think that with the resistance to ISAP would be greatly 23 diminished.
24 And two, is that ISAP can't be related to the IPE, it l
25 can be used as a means of deciding what needs to be done and in
(
Heritage Reporting Corporation (202) 628-4888 1
l 121
(~S I
what order.
2 I don't believe, this is a very personal thing, and I 3
am not sure that staff realizes the full implications of a 4
broad ISAP program.
It is a completely different way of going 5
about your job, it is a completely different way of doing G
business.
- 7 It places a lot more emphasis on risk and a lot less
- D on the regular requirements of the regulations and standard 9
review claim.
We saw this in SEP, it worked in SEP with a 10 dedicated review group.
Dedicated SEP group, with their own 11 project managers with a certain amount of authority.
12 I assume that somebody at NRC has a feel for this, I 13 am not sure.
I haven't seen any evidence that this O
14 understanding of how major this is in a way of operating.
15 I suspect that commissioners don't have the slightest 16 idea of what is involved.
Now I am not at all sure that if 17 they understood it, it would make any difference.
There might 18 be even more opposition of the staff among commissioners, if 19 they understood how the traumatic, well traumatic maybe isn't 20 the word, although I think there would be some trauma involved 21 in the early years of it.
22 I have trouble understanding why so many people are 23 against it.
The only outfit that's ever tried it is very 24 enthusiastic about it, on these new, they are begging for them.
25 MR. SHEWHEN:
I'm sorry if I missed it in your Heritage Reporting Corporation (202) 628-4888
122 1
introduction, but what is the Wason's-(ph] amendment that you'd 2
earlier said that it was in and is now out?
i-3-
DR. SIESS:
Well the original scheme was, and it's 4
the-same thing in the integrated living schedule, that once a 5
schedule is arrived at, that schedule is incorporated in a
'6 license amendment.
And any changes in that schedule now must
'7 go through this process.
That is not in there anymore.
'8 MS. MILLER:
Let me-just clarify.
In the. original 9
ISAP -, we had talked about only portions of the policy, 19 significant items being included in the license amendment, a
11 The license amendment in general would talk about the 12 utility complied with improved methodology, and as we_ adhering 13 to the schedule of the methodology develops, not after we
(
14 accrued the fault of the schedule in the license amendment, the 15
-- amendments, and this is going to change.
4 16 DR. SIESS:
You did say that certain items would be 17 included.
4 18 MS. MILLER:
That's right.
I 19 DR. SIESS:
And let's face it.
If the utilities have 20 any experience in the integrated league schedule everything was 21
-- isn't it?
Only certain items again.
22 MR. CAPUCCI:
Tony Capucci of the staff.
The only 23 way I'm familiar with integrated schedules is the way -.
24 DR. SIESS:
You know if they have a license 1
25 amendment, but it just, -- offset, okay.
How they changed the, i
Heritage Reporting Corporation 4
(202) 628-4888
)
o 1
. + -
a.
123
. /j
(
1 just the process then.
2 Again, I think the license amendment feature was 3
objected to partly on suspicion, partly on the fact that you 4
have to come back to the staff if you want to change it.
5 It wasn't all that clear to me, and I suspect to some 6
of them as to just what they had to come for the license 7
amendment change.
But anyway, that's out, but it was in at the 8
time they got the responses.
9 Melanie Miller, who has been shepherding this thing 10 through the staff is here to answer any questions that you 11 might have.
I have made no recommendations yet.
12 CHAIRMAN KERR:
Any comments?
13 DR. SIESS:
Let's see, who was at the subcommittee g-D 14 meeting.
Michelson, Wylie, Remich ar.d Ward.
Let's hear from 15 them.
I'll start with Mr. Ward first, since--
Can you address 16 this independently of the other matter?
If not, go ahead.
17 MR. WARD:
No, I don't have anything in particular to 18 add to what you say, except that I think I believe more 19 strongly that this program coupled with the IPE program really 20 provides an opportunity to bring some real coherence to the 21 regulatory program.
And I think it will be almost a tragedy if 22 the opportunity is passed by.
23 DR. SIESS:
I think that, you say this program, you 24 mean a formal ISAP?
You mean that using risk is a basis for 25 deciding things?
O Heritage Reporting Corporation (202) 628-4888
4 124 0
1 MR. WARD:
Well, I guess I don't separate the two.
I 2
mean the latter is the important element, but it has to be done 3
in some formal way.and the ISAP proposal suggests a way that 4
that be done, and it would have to be some systematic way, 5
other than a formal way.
6 DR. SIESS:
In our generic items, one of the last 7
ones we used questioned the ISAP type approach.
18 MR. WARD:
Yes.
I
')
DR. SIESS:
That.is really what you are talking 10 about.
11 MR. WARD:
Yes.
12 DR. SIESS:
Forrest?
13 DR. REMICK:
I don't have anything specific to add.
O 14 I think it, in my mind is, well one, I think it is unfortunate 4
15 that there has not been more interest on the utility side for 16 an ISAP program, i
17 But I realize that there have been some recent staff i
18 modifications that might make it more saleable and my 19 understanding is that that has not been broadly disseminated.
20 So perhaps there would be more interest.
I 21 And I think it is the way to handle things in the l
22 future, to put them into that risk perspective.
I guess I'd 23 like to see as a long-term way of doing business at the NRC, so 24 I am somewhat disappointed that it apparently did not go 25 further.
l.
(:)
Heritage Reporting Corporation l
(202) 628-4888
125 1
And I am also disappointed, and I.think you expressed 2
and David expressed, that the agency-isn't grasping this as a 3
way of integrating a large number of activities that are l
?
4 ongoing and which we do it piece meal, depending on which l
4 5
office is handling it, and which department or division is 6
handling it.
7 And no where in their agency is there that broad j
O statemanship, wisdom that pools all these things together where l
9 they can be pooled together, and therefore for the thousands of 10 licensees out that are trying to make sense of what we are 11 doing, are completely lost from time to time.
L
'1 12 So I feel it is unfortunate that we are not grasping l
13 the opportunity to make it an integrated approach.
O 14 DR. SIESS:
Let me comment on something you said.
I 15 said something at the subcommittee meeting about possible more 16 favorable response if there was a workshop and people had a 17 chance to discuss it.
18 And I think that, who was it?
Thomas?
At that time, 19 said based on I think a feeling of having talked to these 20 people he didn't think we would gain that much interest from l
21 workshops.
So might lose some of the people that already 22 expressed an interest, and maybe gain some.
I 23 So, you know.
I would almost have to defer to his 24 judgment there, but you could always suggest the--
l 25 MR. WYLIEt Could I ask a question?
Forrest are you
(:)
Heritage Reporting Corporation (202) 628-4888 1
126
(~%
\\2 1
basically supportive of the idea that Dave promoted in his 2
letter to Curt about the, regarding the severe accident 3
considerations in IPE?
4 DR. REMICK:
Yes, that's certainly one part of it.
5 MR. WYLIE:
Well what you've heard of -- I can't add 6
to much more than simply, Dave has already said and Forrest has 7
already said, and I support both those.
8 DR. REMICK:
Yes, as I said to the subcommittee, I 9
really think that we are at the point, and I think this 10 committee is at the point where it should face the question.
11 And I sure hate to place requirements on the utilities.
12 But I am so convinced that a good PRA done with the 13 utilities, or by the utility can add a tremendous understanding 14 to the risk from that plan.
15 I think we should face that with a PRA, I think that 16 you have to face this for ISAP and a good IPE program, and it 17 will, I just see an opportunity to integrate many of these 18 things that we are doing that are unrelated.
That appear to be i
19 unrelated, anyhow.
20 DR. SIESS:
Last month in our generic items letter we 21 commented on the compartmentalization of disaggregation, or l
22 whatever you want to call it on USI's, GI's, IPE's and so 23 forth.
24 And again, there are differences, and that is that 25 some of these things are based on the regulations, and some are O
i Heritage Reporting Corporation (202) 628-4888 L
127 1
based on policy statements.
2 And I think there is a real problem in trying to get 3-people to look at this thing, in a logical fashion.
They look 4
at whether it is in compliance with the regulation of GDCs, or 5
whether It is going to meet the safety goal, or if it is i
6 something else.
7 And we are now defining safety at least three 8
different ways, I think, or adequate protection.
I have read 9
that SECY on the draft of the backfit rule that defines 10 adequate protection in terms of the regulations.
And 11 presumably there is a safety goal in mind in some other way.
12 One'of the advantages that the staff has always cited 1
13 I think in terms of ISAP is the PRA.
And everybody that has O
14 ever made a PRA thought it was useful.
15 And as proof of that, EPRI has just finished a study 16 headed the Practical Application of Probabilistic Risk 4
s 17 Assessment.
18 They interviewed utilities, they interviewed NRC l
19 staff people.
It said all utility interviewers stated that the 20 advantages of PRAs awarded development costs benefits very 21 considerably.
j 22 Utilities that perform PRAs primarily to satisfy NRC 23 requirements, achieve their original objective but with few 24 subsequent benefits.
25 Those that use PRAs for other purposes, such as
)
O lieritage Reporting Corporation (202) 628-4888
128 O
1 identityin9.wre cos t of fective design alternatives and 2
procedures accrued continued benefits.
3 Dut EPRI made this survey and everybody thought the i
4 PRA was a good idea.
They can commend them.
They'suggest that l
5 maybe everybody should have one.
And I think if everybody had 6
one the response to ISAP.would be a lot.
7 They didn't have a use for it.
One of the uses is 8
arguing with the staff, and maybe the staffs doesn't want that.
J 9
Charlie, you wouldn't need it.
10 MR. WYLIE:
Well I just thought I'd, I can't add too-11 much more than what has already been expressed by Forrest and 9
12 Dave.
I thought the two programs should be combined, should-it 13 require PRE on the state of the art, the level two and three
(
14 would include, to do it right.
I think Dave's words were let's 15 do it right.
I'm not sure it has to go to level two and three 16 but that was another point.
Carl?
17 MR. MICHELSON:
Yes I agree with what has been said 18 around the table.
I believe that this is a real opportunity to 19 serve two in a greater thinking concerning design basis l
20 excellence and beyond design basis, and do it as a part of one 21 picture at one time.
t 22 DR. SIESS:
But now that's not just ISAP.
That's IPE 23 plus ISAP.
1 24 HR. MICHELSON:
That's right.
IPE plus ISAP.
25 DR. SIESS:
IPE is what the board needs to be fixed, 3
Heritage Reporting Corporation (202) 628-4888
129 1
and ISAP is deciding on what basis you, yeah--
2 MR. MICHELSON:
And they can do it as two, aspects 3
could be treated a little differently when it comes to our 4
view.
It shouldn't be treated differently in looking at the 5
problem.
6 DR. SIESS:
What I hear now is a consensus of at 7
least for the next hour or two hours, of integrating the ISAP, 8
our consideration of ISAP into our consideration of IPE.
Is 9
that a fair statement?
10 CHAIRMAN KERRt Let me, without disagreeing with 11 anything that has been said, raise some questions.
We heard a 12 rather lengthy and detailed presentation this morning of fire 13 research.
Which is treated currently in most PRAs certainly in O
14 the 1150 with one or two exceptions as an externally initiated 15 event.
16 The current IPE says we are going to do internal 17 initiators and we are going to work out a method for doing 18 external initiators, fairly soon.
Or else, decide what to do 19 with it.
20 Now, if the evaluations we heard this morning had any 21 quantitative substance, one, and I don't mean this in a 22 pejorative way, one could conclude that seismic risk and fire 23 risk for outweigh risk due to internal initiators, or at least 24 well.
A hundred percent of the risk almost, if you do the 25 addition not too carefully.
What are we going to do about this O
Heritage Reporting Corporation (202) 628-4888
130
')
1 if we do an ISAP II and IPE?
2 MR. WARD:
Well I think you know, it is when I note 3
that you suggested, they shouldn't be kept separate.
I mean 4
that is saying internal and external initiators is really just 5
kind of an artificial, an historical artifact, or it really 6
doesn't have any, there is not particular logic behind it.
7 There perhaps, as some people claim that have got o 8
better handle on the understanding, the probability of 9
distributions are entirely initiators and external.
I don't 10 whether that is universally true or not.
Perhaps that is some 11 reason to treat it differently.
12 But whether what we heard this morning is precisely 13 right or not, you know, it certainly apparent that the class of O
14 things which has been, we have unfortunately called external 15 initiators seems to be a class that contributes much of the 16 risk.
Half of it? Seventy percent, thirty percent, I don't 17 know what it is.
18 But in addition to that, it can cause somewhat of a 19 different nature of response in the plant systems because of 20 the characteristics of some of the external initiators.
21 MR. WYLIE:
But I think what you heard this morning 22 too is that you are lumping everybody into the same, all plants 23 into the same mode, in a way.
{
24 And you are not recognizing whether there are 25 differences among plants.
And that is where the individual O
Heritage Reporting Corporation (202) 628-4888
131
('s
\\-
1 plant examinations will be a big benefit.
2 I think you've got extremes.
I think you've got a 3
nice bell curve, you've got some real good, safe plants over 4
here, and you've got some lousy ones over here, and you've got 5
a whole bunch of others in between.
6 MR. WARD:
Yeah, but should the individual plant 7
examination be done in this sort of series proposal, with the 8
external, the internal initiators now and the external
'9 initiators vaguely sometime in the future.
10 MR. WYLIE:
I don't think so.
11 MR. WARD:
Right.
Oh, okay, then that is fine.
12 MR. WYLIE:
I think that is the wrong way to do it.
13 DR. SIESS:
At one time they knew it had some basis 14 for separating internal and external, whether it was a good 15 basis I don't know, I won't comment on.
But there is 16 absolutely no reason to separate these.
17 You have to look at anything that can cause risk and 10 include it.
It's ridiculous to try to divide this in some 19 fashion.
20 The uncertainties may be greater in one than in 21 another but so what.
We get to the back end the uncertainties 22 are so large it doesn't make any difference.
I think it is 23 ridiculous.
24 MR. WYLIE:
I think to go this way, to a large extent 25 there is no point in even pursuing what we heard this morning.
O Heritage Reporting Corporation (202) 628-4888
132 Os 1
MR. WARD:
What do you mean, Charlio?
I don't 2
understand that.
3 MR. WYLIE:
I say if you go this way, I mean and 4
provide these programs and look at internal and external 5
events, as part of this program why pursue much more than with 6
this research.
7 MR. MICHELSON:
Well, the reason is that you have to 0
have some info data to go with your PRA model in order to get 9
the numbers out.
10 MR. WARD:
Maybe for that.
11 MR. MICHELSON:
Yes, that is what you would have to 12 do.
And also learn better how to better model.
The human 13 factors, and is another element in here that if not done well, O
14 if at all hardly the PRA's in terms of the input data, and may 15 be a major contributor to risk and somewhat able to really show 16 good PRA.
17 And it might come up with a major competitor risk, I 10 don't know.
But I think the human factor has to be thrown in 19 along with the internal and external events.
20 DR. SIESS:
It was the major contributor to the only 21 two accidents we have had so if the PRA doesn't say it is a 22 major contributor something is wrong with one or the other.
23 MR. MICHELSON:
The major considerations are one and 24 the same.
25 DR. SIESS:
Yeah, it works both ways now.
O Heritage Reporting Corporation (202) 628-4888
133 O
1 MR. WYLIE:
Dut it needs to be in there, and talk 2
external internal.
3 DR. SIESS:
You know, I don't think there is any 4
question but this stuff has been compartmentalized.
It may be 5
according to the discipline or maybe who is looking at what, 6
and who knows enough to ask questions about something.
7 Dut do you realize how often in the course of looking 8
at something we have seen different segments of the staff every 9
two hours in here.
10 And I don't know whether they talked to each other or 11 not.
They don't stick around here to hear what the others have 12 to say.
13 CRAIRMAN KERR Some of them probably have to do some O
14 work rather than just starting a meeting.
(Laughter)
It is 15 going to fail.
How far shall we pursue this?
16 DR. SIESS:
Well, as I said, the one I heard as to 17 the comments I got from the subcommittee attendees, the next 10 step is to go on with the IPE and continue our consideration of 19 ISAP in connection with the overall larger problem, let's say.
20 And then depending on what comes out of that, if ISAP doesn't 21 become a part of our recommendations on IPE then we will go 22 back to the subject letter or something on ISAP, which I can 23 prepare very quickly.
24 But I think there is another problem here, gentlemen.
25 You can talk to the staff up to a certain point, but these are O
Heritage Reporting Corporation (202) 628-4888
134 i'
t major issues, and how do we go about educating the commission.
2 Not advising them, unless we just wanted to take our 3
advice blindly, which they is certainly not that much 4
inclination to do.
S MR. WYLIE:
You think we ought to schedule a meeting?
6 MR. WARD:
I think the best thing to do is to start 7
off by writing a thoughtful letter that just doesn't give some, 8
you know, shots from the hip but explains our reasoning.
9 DR. SIESS:
Good step.
10 MR. WARD:
It's a worthwhile step.
11 DR. SIESS:
We frequently come out with such a letter 12 as a first draft, and by the time we get finished with it, it 13 isn't.
How about a short letter, of about three sets of O
14 additional remarks.
15 MR. WARD:
No, I think we can write a letter.
16 DR. SIESS:
Well I'll turn it back to you, Mr.
17 Chairman, to discuss IPE letter.
Or if we can find a broader 10 definition for it.
19 C11 AIRMAN KERR:
Any further comments on ISAP II?
20 (Continued on following page.)
21 22 23 24 25 Heritage Reporting Corporation (202) 628-4888
4 4
135 Okl 1
CHAIRMAN KERR:
The agenda shows us beginning a 2
discussion of the IPE latter at some later time, but does the 3
staff have people here that are in a position to start this 4
earlier?
5 You will recall that the individual plant examination 6
was first mentioned in the Commission's policy statement issued 7
in August of 1985 and I will read from that at least one 8
paragraph:
"Accordingly, when NRC and industry interactions on 9
severe accident issues have progressed sufficiently to define 10 the methods of analysis, the Commission plans to formulate an 11 integrated, systematic approach to examination of each nuclear 12 power plant now operating or under construction for possible 13 significant risk contributors sometimes called outliers that O
14 might be plant-specific and might be missed absent a systematic 15 search."
16 The committee commented to the Commission in a letter 17 dated June 9, 1987 and expressed some reservations about the 18 state of that particular draft.
Since then, there has been a 19 considerable change and I believe at least from my point of 20 view, a considerable improvement in the letter, draft letter 21 that we see now.
And I should cal your attention to the fact 22 that this is in Tab 5 along with a number of other things.
23 I will remind you that I believe you have a copy of l
24 the letter in Tab 5.
You might want to refresh your memory on 25 the comments we made at that time.
(
Heritage Reporting Corporation (202) 628-4888
136 O
1 The Subcommittee on Severe Accidents, and I think 2
there were about six members of the committee present for that 3
meeting, met on April 26 and received a presentation and 4
considerable discussion of the draf'; letter.
5 I have asked that some parts of that discussion be 6
presented to the committee in somewhat abbreviated form with 7
emphasis particularly on.what is called the back end analysis O
cuggested in the IPE letter.
9 I have no further introductory remarks to make but I 10 would ask any of you who were present at the subcommittee 11 meeting if you would like to make any comments.
12 Thank you.
The results of the subcommittee meeting 13 and the discussion that followed, I have prepared a draft IPE O
14 commentary which is at your position, this green sheet.
You 15 might want not to refer to that during this presentation, but 16 you might want to look at that as we go into further 9
17 consideration of this possible joining or co-joining, or 18 whatever -- conjoining is the word, of the IPE and the ISAP-II.
19 I will at this point turn things over to the staff.
20 And I'm not quite sure what the order of presentation is, or 21 the protocol.
But I can see Mr. Beckjord is here and protocol 22 would suggest that I ask him to begin things.
23 MR. DECl;JORD:
Thank you, Mr. Chairman, 24 We are here today for two purposes.
The first is to 25 make a presentation on the integrated plan for severe O
Haritage Reporting Corporation (202) 628-4888
137 O
1 accidents.
That is a short presentation.
I believe the j
N l
2 subcommittee has already heard the presentation on this.
3 Dr. Themis Speis will make that presentation.
4
~
The second and main purpose is to bring before you 4
t 5
the independent plan examination, the IPE, and provide you with_
6 any final information which is needed for what we hope will be j
7 the preparation of your letter of approval.
8 on that, we're ready to move out on the IPE.
Its l
9 purpose, as you well know, is to identify vulnerabilities in 10 operating plants and to detect any outliers among other things.
11 I think it is important to get on with this job, 12 presently, because if there are risk outliers there, we want to I
13 discover them as soon as possible.
O 14 The IPE has had review of the CRGR and we have a i
l 15 letter of their approval with some comments which we are r
16 responding to.
i i
j 17 I'm satisfied that we're ready to move out with it.
f j
18 Tom Murley is also satisfied that the time has come.
And so J
19 that's what we wanted to discuss with you in the second 20 presentation.
[
1 l
21 DR. SIESS:
Will somebody define what you mean by an 22 outlier in the presentation?
2 23 MR. BECKJORD:
An outlier in the dimension of risk l
24 that is substantially above what we believe is the state of the l
25 drt today among operating plants on core damage frequency.
(
()
Heritage Reporting Corporation
(
(202) 628-4888 l
t 138 O
1 DR. SIESS:
Risk has at least two dimensions.
One 2
is core melt and the other is large releases.
And core melt 3
you get with what they call the front end and large releases 4
you need both the front end and the back end.
5 Are the outliers we're talking about related to both 6
of these dimensions and to both of these ends?
7 MR. DECKJORD:
I think they are, yes.
I tend to 8
think in terms of the front end first because it is the front 9
end.
We're going to address this in the presentation.
10 DR. SIESS:
Now, since there is no safety goal that 11 relates to core molt, is an outlier just something that 12 contributes a large fraction or is it something that 13 contributes more than we thought it should?
O 14 Sometimes people use the word "dominant 15 contributors."
And a dominant contributor in my mind is not an 16 outlier.
17 MR. BECKJORD:
No, no.
I'm thinking of the final 10 number on core damage frequency.
That is substantially -- for 19 example, 1150 is showing that of the five plants for which we a
20 have reports completed now, core damage frequencies of several 21 times ten to the minus four.
I would regard an outlier as 22 something which is substantially higher than that.
23 DR. SIESS:
You've taken 1150 then as a base.
74 Anything bigger -- that is acceptable and anything bigger than 25 that or substantially bigger than that is an outlier.
O Heritage Reporting Corporation (202) 628-4888
139 1
MR. BECKJORD:
Well, you asked me for my frame of 2
reference and that is my frame of reference.
3 DR. SIESS:
Okay.
There's been a safety goal frame 4
of reference, the 1150 is your frame of reference.
5 MR. BECKJORD:
Well, the safety goal, we are working 6
on the, how to implement that policy statement, now, and we 7
have a recent letter giving your advice and we have been 8
working on that actually for some time now and I expect that 9
that activity is going to come to a conclusion, at least our 10 conclusion, late this year.
11 DR. SIESS:
liow, what would then be a measure of an 12 outlier for something in terms of releases?
Again 11507 13 MR. BECKJORD:
We've had long discussion on that in O
14 our office, and I'm not really prepared to address that 15 specific question this afternoon.
I think there are a number 16 of considerations, and it's not simple, 17 DR. SIESS:
One reason I ask, Eric, is that when 18 people have done PRAs in the past and have found what were then 19 called outliers, every one I can think of was discovered from 20 the front end, not from the back end.
It was an outlier in 21 terms of possible core damage and not an outlier for 22 containment failure or releases.
?3 At least one of our consultants in one meeting 24 questioned whether you need the back end with all its 25 uncertainties to identify outliers the way he thought outliers O
Heritage Reporting Corporation (202) 628-4888
~
140 /141'
/~%
U 1;
had been called l'n the past.
-2 MR. WARD:' Are you'sure it isn't the Mark I problem, 3
an' outlier on the back end?
4 DR..SIESS:
I'm not sure.
If the probability of core 5
melt is low enough, the probability of a large release from an 6
early containment failure, may be well within thes safety goals.
7 It might be a dominant contributor.
8 MR. WARD:
Well, the numbers that are used to' express 9
for probability of core melt and Mark I BWPe are quite low, 10 generally.
11 CHAIRMAN KERR:
Are you pre-empting Mr. Beckjord's 12 answer?
13 DR. SIESS:
Okay.
I've got answers to at least O
14 enough to let Themis start speaking.
15 CHAIRMAN KERR:
Mr. Spels, I want to correct a 16 statement.
We did not hear Mr. Speis' presentation earlier.
17 We did hear something that involved some slides, but it did not 18 have the Spels flavor.
And so we are hearing that for the 19 first time.
20 (Continued on Page 31) 21 22 23 24 25 O
Heritage Reporting Corporation (202) 628-4888
142
-1 MR. SPEIS:
Thank you, Mr. Chairman.
I assumed th'at 2
you wanted to hear my accent.
You are not going to be 3
disappointed.
4 Was I supposed to be listening when I was in the back 5
a half an hour ago?
6
. CHAIRMAN KERR:
Were you supposed to be listening; 7
yes, sir.
8 MR. SPEIS:
I want to make'two points.
I heard 9
Professor Remick say that everybody at the staff level was free 10 to send requirements to the industry or to the utilities.
11 DR. REMICK:
What did I say?
12 MR. SPEIS:
That everyone is free at the staff level 13 to send out letters, and impose requirements, and things of O
14 that sort.
Maybe I misunderstood you.
15 DR. REMICK:
No, I certainly did not make that 16 statement.
17 MR. SPEIS:
Okay.
Because there is only one office.
18 Everything goes through the NRR Office.
And you know lately, 19 the last three or four years, any requirement is --
20 DR. REMICK:
No, My statement was something to the 21 effect that there is no office of wisdom that is pulling all of 22 these things them and integrating them.
I am not just talking 23 about NRR.
We spent time yesterday with NMSS and Research, 24 which they were talking about below regulatory concern.
And I 25 think that it is a wonderful opportunity to tie together the O
Heritage Reporting Corporation (202) 628-4888
143 o
il safety. goal, but we do not think that way.
That is a NRR 2
' matter.
That is not a NMSS matter.
And we have opportunities 3
all of the time to think as an agency, but we do not, and that 4-was the point that I was trying to make.
5 MR. SPEIS:
Well, we talk too much to each other, you 6
know.
That is one of the difficulties, you know.
We talk to 7
death.
8 CilAIRMAN KERR:
Maybe you do not listen.
9 MR. SPEIS:
You have viewgraphs in front of you.
10 (Slide.)
11 MR. SPEIS:
What we have been attempting to do is to 12 put all of these programs together and see what sense.they 13 make, how they fit together, how they interaction, wtit is the 14 input from one program, what does the output from one program 15 go, and so on and so forth.
And I am talking specif.Jally 16 about the issues that are related to the severe accident 17 program.
'8 Let me give you some background before I proceed any 19 further.
You are familiar with the severe accident policy 20 statement, which was issued back in 1985.
Back in February of 21 1986, we presented to the Commission an implementation plan, 22 which really addressed what was in the policy statement at that i
23 time.
24 And if you recall, there were three elements in the i
25 policy statement.
One, the individual plan examination dealing O
Heritage Reporting Corporation (202) 628-4888 l
i
144
- 1 with the existing plans.
The other one was what are you going 2
to do for future plants as far as severe accidents are-3 concerned.
'And the third area was at that time, see, you have 4
done all of this resource for the last five or ten years, and 5
maybe it is time to look over some of the regulations,'some of 6
the more' relevant regulations that have some relation to severe 7
accidents and see what changes can be undertaken. 'So this 8
implementation plan that we. presented to the Commission dealt 9
with those three areas at that time.
10 Back at the end of December.of 1986, the new 11 Chairman, there were other issues that had come to his-12 attention at tnat time, including the resource program itself,
~
13 and he wanted to see an integration of the program.
O 14 In February of last. year, we put some of these issues.
15 together in some pseudo-integrated fashion.
That was not 16 complete though. 'Back in July of 1987 when we briefed the' 17 Commission, we gave them a three-prong approach o resolving l
18 severe accident issues.
I think that Dr. Murley has discussed 19 this with you.
The IPE containment improvement, plant t
20 operations and improved plant operations.
So that wcs our j
21 briefing to the Commission.
22 And back in December of 1987, we went to the 23 Commission a paper that really addressed the Mark I issue.
And j
24 at that time, the paper itself attempted to discuss to some the relationship of that program to some of the other 25 extent
(
l Heritage Reporting Corporation l
(202) 628-4888
M N
'145 1-ones.
2 Coming back to 1988, it would.have been' preparing 3
with the industry, and we had discussions with you over-the 4
last few-years about the so-called IPE effort, individual plant 5
examination.
At the same time, we have in parallel working on 6
Mark I iss;as.
The 1150 has come upon us.
It means different 7
things to different people.
Unfortunately, the resource 8
program, many of you and many of us have questioned how does it 9
fit inte this overall effort.
10 So the EDO called all of us who have something to do 11 with this issue to Baltimore for a meeting back in 12 February of 1988 to discuss all of these things.
And out of 13 that deliberation or discussion came this thing that we are O
14 doing right now.
15 So that is enough of the background.
What we are 16 trying to dc is --
17 DR. St.ESS:
You keep referring to severe accident 18 issues, and it is not quite clear to me what the scope is.
19 Could you give me an issus that is not a severe 20 accident issue?
21 MR. SPEIS:
Well, very shortly, I will give you this 22 list here.
23 DR. SIESS:
Okay, I will wait.
24 MR. SPEIS:
The purpose is to get all of these things 25 together and see how they fit.
Let's make sure that they make O
Heritage Reporting Corporation (202) 628-4888
~
146
~1 some sense.
In fact, you know,-in this coming year or maybe in 2
the next two years, we will be going to the Commission with 3
some recommendations on some of these issues.
And we cannot do 4
it all at once, unfortunately or fortunately.
S' And one of the things that we want to do is_present 6
to them what is our plan, where do things stand right now and 7
what will they be going through in the next few years,.and how 8
do those separate pieces fit together basically.
So that is 9
the purpose of this effort.
10 The objectives, I think that I have talked about some 11 of them already.
To provide an understanding of our activities 12 that are underway to implement the severe accident policy.
To 13 assure that these activities are consistent with the 14 Commission's policy and strategic goals.
To assure that our 15 activities are consistent among themselves, and that is a very 16 important part in itself.
17 To have a common goal of ultimately leading to 18 improved plant safety, are properly coordinated among the 19 responsible NRC organizations.
And there will be a lot of talk 20 and discussion as far as that goes.
And'to assure that the 21 Commission is aware of the key technical and policy issues, 22 some of which need their guidance and approval basically.
23 Let me put a list up of some of these things, and 24 let's see what you think about them as far as following under 25 the umbrella of severe accident activities.
O Heritage Reporting Corporation (202) 628-4888
7.
~
4 147 O',
1 We are ' talking basically about the individual plant 2
examinations, which is the key part of the Commission's policy 3
statement, where they said work with industry and come up with 4
.a good method, and go out and make sure that every plant 5
undertakes this individual examination for individual things, l.
6 plant specific vulnerabilities or outliers.
And~I think that
~7 we are talking about both of them.
And Brian in his 8
presentation discussed it.
9 We are talking about container performance 10 improvements starting with Mark 1.
And here, we are talking 11 really about the generic aspects of containment performance 12 improvements.
Logic says that you should not have these. things 13 here.
Individual plant examination is okay by itself, because O
14 it does include both the front end and the back end, the back 15 end meaning the containment performance.
16 So the question is why are you pursuing this.
I am 17 not so sure that I can give you a good reason.
I guess that 10 there were Chernobyl, and there were some preliminary findings 19 from NUREG-1150 and other statutes that indicated that possibly i
20 some containments are less robust than others, and maybe we 21 should accelerate the effort.
22 So that is why it is here really.
But those two 23 complement each other.
We want to make sure that the industry 24 or un do not duplicate anything.
But really here we are j
l 25 looking at the generic aspects of it, okay.
So in the l
()
Heritage Reporting Corporation (202) 6%3-4888 l
l
e V
148-il individual. examination part, the plan will have to be examined
'2 front and back end.
And if we come up with any generic 3
requirements, they have to fit.
They have to fit in the plan 4
itself, okay.
5 So the examination itself and any generic 6
requirements,-they have to complement each other.
But logic 7
again says that it should go further.
8 Improved plant operations.
This is the continuing 9
part of doing everything that Murley does and the regions do to 10 improve the plant operations, which is really the preventive 11 part of accidents.
12 Now all of us know that PRAs and other statutes do 13 not treat human actions and human interventions very O
34 accurately.
So there is the parallel effort that precedes.
15 You have this program SARP dealing with management 16 effectiveness and all of these things that you people are s
17 familiar with.
18 As part of these activities, a key part, is the 19 severe accident research program, and what type of support that 20 it provides to these activities.
Also, what type of efforts 21 should continue into the future on a generic basis, as far as 22 searching and gaining more understanding of not only things 23 that we do not know well enough right not, but may be 24 uncovering things.
25 We are talking about an accident management program, O
lieritage Reporting Corporation l
(202) 628-4888 i-
149
- 1 cuul Dr. Sheron will talk about it more specifically.
There is
'2 just-NUREG-1150, the reactor risk reference document, which is 3
supposed to be an update of 1400.
4 As far as I am concerned, it is a document that 5
updates all of the knowledge of the last three, or four, or 6
five years.
It gives a good assessment of five plants, and 7
those insights and the lessons from it will be extremely 8
important and useful in proceeding with decisions in the IPE 9
framework and the container performance improvements.
10 Generic safety issues are important.
Again these are 4
11 the things that we have found from experience that lead us to 12 scrutinize them and examine them very carefully to see if our 13 regulations should be changed, if we should do something on a 14 plant specific basis.
And to report on this whole activity, 15 because there are additional efforts in the prevention 16 category.
17 External events is an important part.
I heard you 10 say that maybe we should not proceed with~the IPE.
If it stays 19 on course rignt now, because right now we are recommending that 20 we proceed with internal events.
And I do not think that we 21 are ready to do that right now. But I think that we have some 22 good reasons why we should separate internal and external right 23 now.
We will talk to you later on about that.
24 DR. SIESS:
You do not even have internal events on 25 your list.
O lieritage Reporting Corporation (202) 628-4888
~ _,
-150
("1
- 1 MR. SPEIS:
Excuse me.
2 DR..SIESS:
You do not even have internal events.
3 MR. SPEIS: 'These-internal events'are included here.
4 LDR. SIESS:- I see.
5' MR. SPEIS:
I am sorry.
6 DR. REMICK:
Will you tell us why it makes sense,'and 7
maybe it is to separate out external, could somebody explain to.
8 me why fires and things are called external?
9 MR. SPEIS:
We will do the best we can.
But I think 10 that based on what I heard earlier that we might have to have a 11 separato discussion with you.
Because some of our experts like 12 Joe Murphy and others are telling us that we are not going to 13 lose any efficiencies by stqrting out with the internal events
- O 14 and maybe later on proceeding with the external events.
We.
15 will tell you what we can right now, and maybe we will have to 16 discuss this further.
17 Because we feel very strongly that we are ready to-4
[
18 proceed with the internal events. And if we have to wait for 19 the external events story to get together and agree and do the 20 work that we think is required, that is going to be another 21 year and a half, and I do not think that we are serving safety 22 by waiting for another year and a half.
But we will talk to 23 you more about it.
24 DR. SHEWMOll:
Did you hear before that ISAP and IPE 25 are integrated?
(
Heritage Reporting Corporation (202) 628-4888 i
r
0 1
151
(
~1 MR. SPEIS:- I only came about ten minutes ago, ten or 2
fifteen minutes.
But my view is that ISAP is the machine that 3
-you grind through, and what.comes out of it is-what is really 14 important, and then you schedule it.
The IPE provides the
.5 framework to do that assessment, to come up with the hierarchy.
6 uso that is really the PIUL.
That is the way that I look at it.
7 DR. SHEWMON:
Did you say that ISAP produces or 8
grinds through those things, and out comes what is important?
9 MR. SPEIS:
That is right.
10 DR. SHEWMON:
It seems to me that that is a fair 11 description of what IPE'is supposed to do.
12 MR. SPEIS:
Well, ISAP goes a step further.
In the 13 ISAP machine, you dump licensing issues and you dump many other 14 things.
15 CHAIRMAN KERR:
Both the important and the 16 unimportant.
2 17 MR. SPEIS:
Yes.
18 CHAIRMAN KERR:
I seem.
19 MR. SPEIS:
There is licensing and regulatory things.
20 So you do more.
21 DR. SIESS:
But there is one very important thing.
22 None of these things contribute anything in safety until they 23 get done.
24 MR. SPEIS:
Right, that is right.
25 DR. SIESS:
And ISAP is a means of assuring that the O
Heritage Reporting Corporation (202) 628-4888
i e
152 1
most important things get done first.
2 IPE can identify all sorts of things, and we.can have
'3 USIs like ATWS: that identify all of the-important: things, but 4
how many plants have fixed it yet?
5 MR.'SPEIS:
We all agree that we can create the 6
masterpieces of safety requirements. But if they do not go out-7 there, it does not mean anything.
-8 MR. BECKJORD:
But in the case of the things that 9
were uncovered during the 1150 work on those five and now six 10 plants, there were a number of things that were disclosed.
And 11 it is my understanding that action was taken on every one of 12 those.
13 DR. SIESS:
But they had to be important enough.
O 14 They found them more important than ATWS, and ATWS has not been 15 fixed on them and something else has.
I do not know.
16 Were they fixed because they were high priority 17 safety items?
18 MR. BECKJORD:
My understanding is that the ones that 19 were identified on 1150 were fixed because they were considered 20 to be important, so the action was taken.
21 DR. SIESS:
But the suggestion is that a lot of USIs 22 may be unimportant.
I make the suggestion, I am not implying 23 it, I am stating it. Because they have not been through a risk 24 assessment some of them have not.
ATWS may not be a good 25 example.
l lieritage Reporting Corporation l
(202) 628-4888 i
l
t L
153 1
MR. SPEIS:
. Ell, I think that some of the USIs W
2 probably-are not as important as some high priority cases. Some 3
of the USIs that were generated as.USIs ten years ago, we did 4
not have a very rigorous process of going through and seeing 5
how important they were risk-wise or contributing to core melt,
~
6 or whatever criteria that you want to use.
That is a fair 7
~ tatement.
s 8-DR. SIESS:
The thing that sets the ISAP item off 9
from all of the others there is that it is involved with the 10 actual implementation, not the identification.
It is the stuff 11 that gets you done, to get something done.
12 MR. SPEIS:
Well, Cecil is here, and I think the 13 staff.
But I guess that you have heard enough about it.
14 There is the severe accident policy for future 15 plants.
We want to make sure that it is consistent with the 16 other activities.
And this thing here is very important.
What 17 type of criteria do we use, how do we define closure.
You 10 know, are we going to have this Damocles sword over the 19 industry for the next ten years or so.
What is the process, 20 what are the things that they have to do that can bring this 21 process to some recognition of closure.
22 And of course, the safety goal is very crucial.
We 23 have got two letters from you, and we are looking at them very 24 carefully.
We have some good ideas on how to apply the safety 25 goal.
We will be coming to you again, and we will be making O
Heritage Reporting Corporation (202) 628-4888
154 1-recommendations to the Commission in the next'few-months-or so 2
after we talk to you once more.
3 But I think at this point in time that I do-not think 4
that-we are ready to discuss in_ great detail.
And personally, 5
most of_us at Eric's organization think_that the way to go is 1
6 the way that you have suggested it.
7 MR. WYLIE:
One of the things on there is recognizing:
~
8 an outlier.
Apparently, everyone feels that they will-know one 3
when they see one.
'10 MR. SPEIS:
Well, it is part of this, and part we
. ill talk about.
11 w
12 MR. BECKJORD:
Dr. Sheron is going to talk about the 13 implementation.
14 DR. SIESS:
Before you take that off, if every plant 15 out there had a PRA Level 3, would that list look any 16 different.
These are their activities. ;
17 It seems to me that the first two items would not 18 have to be on there, am I right?
19 MR. SPEIS:
Yes, yes.
20 DR. SIESS:
Is there anything else that would come 21 up?
22 MR, SPEIS:
Well, you know, having a complete PRA.
23 By having this accident management program, a complete PRA 24 gives you lots of wisdom and insights on how to sharpen your 25 procedures.
You know, it gives you some idea of what is O
Heritage Reporting Corporation (202) 628-4888
155
)
1 important to fix in a cost effective way.
So that takes care 1
T2
- of this.. -It is a way to clean up most of these things here 3
basically.
~
4 DR..SIESS:
In other words,'several of your listed 5
activities up there would be subsumed into a PRA.
6 MR. SPEIS:
The other question is if one does a.PW4, 7
what do you with the results, okay.
I have a diagram here.
8 All of us make diagrams.
9-CHAIRMAN KERR:
Are you sure that it is not upside 10 down?
11 MR.-SPEIS:
No.
I guess that it is not a very good 12 picture.
It is not upside down.
I do not think so.
i 13 The main activities are the IPE, the container 14 performance, improved plant operations, severe accident 15 research.
You see all of them all really fit into the accident 16 management.
The resource program fits into all three of them,
'17 IPE, container performance, improved plant operations, and of 18 course it continues.
l l
19 I had ISAP here. I prepared this viewgraph I guess 20 before NRR made its recommendation, but that does not mean that 21 ISAP is out forever.
It is at the point where the grinding can 22 take place, and the hierarchy can be developed with the help of l
L 23 the PRA, and then ISAP does its thing.
24 Some of the backfit and safety goals again are 25 guidance.
And some of the v:her activities are input and Lo Heritage Reporting Corporation (202) 628-4888
e 156 t"'\\
~(/
1 output.
2 DR. SIESS:
I do not see anything on there that says
-3 where you do-something about a plant improvement other than 4
containment.
The box says containment performance 5
improvements, and I assume that that means that you do 6
something.
7 MR. SPEIS:
Yes.
8 DR. SIESS:
Make more reliable sprays.
9 MR. SPEIS:
Yes.
10 DR. SIESS:
But suppose that it is not containment, 11 suppose that it is front end, the probabilities from the front 12 end.
13 MR. SPEIS:
It is an individual plant examination.
O 14 DR. SIESS:
But that is just an examination, but 15 where is it that you do something?
16 MR. SPEIS:
Out of that process, you identify plant 17 vulnerabilities, okay.
18 DR. SIESS:
Okay.
19 MR. SPEIS:
And then next development improvement 20 that could be procedures, hardware or whatever.
l 21 DR. SIESS:
But develop them means do them?
1 22 MR. SPEIS:
Excuse me.
23 DR. SIESS:
What does develop mean?
24 MR. SPEIS:
To do them.
l 25 DR. SIESS:
To do them?
(
Heritage Reporting Corporation (202) 628-4888
157 1
MR. SPEIS:
Yes.
2 DR.-SIESS:
Okay.
l3 MR. SPEIS:
Again, it'is Greek writing in the English 4
language probably.
That is the difficulty.
5 I do not think_that we will talk about the next 6
viewgraph, NUREG-1150.
You heard enough.
I want to say 7
something about the issues program.
Because you people in the 8
past based on your letters had some difficulty in understanding 9
the rationale for some of the things.
10 What we are trying to do now is identify what are 11 really the issues, what are the severe accident issues.
We are 12 trying to put them in this category, and take them a 13 containment at a time.
Take the last, for example, and focus O
14 on failure modes, the important ones.
And then to try to 15 understand what are really the issues, what are the things that 16 drive and challenge the containment.
1 17 Also develop a relative prioritization of these 18 issues. And then see if we have the right issues program to 19 address those issues.
And that is not a very easy task.
20 Because you know, some of these programs have been around for a 21 long time, and they have developed a life of their own. So now 22 we are trying to orient them in terms of focusing on specific 23 issues.
24 So when you see this integrated planning detail, 25 there is a big chapter there on resource which tries to O
Heritage Reporting Corporation (202) 628-4888
158
()
~
1 identify the issues, and then list below what type _of resource 2.
we are doing to address those specific issues.
So I am giving E,3 a.very cryptic example here of containment, for example.
4.
DR. SHEWMON:
We have had a fair amount of criticism 5
and it may continue on any coherence of that program and the 6
relevance to other NRR issues, and I am glad at least'that you-7 are looking at it.
But before you leave it, I have one.
8 question.
9 Do you have any idea whether in that direct 10 containment heating that anything.is being done about the 11 failure of the primary system at anyplace other than spring 12 without a nozzle?
I 13 MR. SPEIS:
We have a program, and I have not 14 personally looked at it in detail.
I am trying to put this 15 into some coherence right now.
The program, it is a four prong 16 program. One of them is to understand the probability, okay.
17 And that involves that maybe the primary system could fail 18 someplace else before you blow down through the vessel.
And 19 that is in a sense a practical program.
20 And the other one is maybe we can utilize existing 21 equipment or procedures to ameliorate this issue by 22 depressurizing the system.
Another one then is looking at the l
23 consequences themselves.
And another one is looking at the 24 geometry.
l 25 DR. SHEWMON:
That is the one that sprays it out l
O Heritage Reporting Corporation (202) 628-4888 l
l
r 159-I tubes?
2 MR. SPEIS:. Yes.
- 3 DR. SHEWMON:
Those groups had different meetings in 4
different months and did not talk to each other.
5 MR. SPEIS:
I think that we are trying to address 6
this program in its essential elements in a final effort right 7
now, because it is kind of open-ended to some extent.
I agree 8
with you.
9 DR. SIESS:
There is an amount of comfort that I get 10 from that slide, but what is the assumption that I make about 11 how complete it is.
12
.MR. SPEIS:
Well, as I say, it is cryptic.
You will 13 see it in detail when we put the plan together.
O 14 DR. SIESS:
Let me ask you.
I see an item there at 15 the bottom of the page, analysis of performance tests on 16 concrete containments.
For some reason, I do not see a 17 corresponding item of performance tests on steel containments.
18 And we do have steel containments in large dry PRWs.
And we 4
19' have made tests on steel containment, and it did perform in a 20 very interesting manner, especially compared to the concrete.
21 But yet I do not see that up there.
22 MR. SPEIS:
This is not complete.
I am sorry.
That 23 will be there.
I put this up to show you the concept, that we 24 are trying to identify the issues and the resource, and 25 unfortunately it is not complete.
O Heritage Reporting Corporation (202) 628-4888
160
_1 DR-SIESS:
It would give me a lot more comfort if' 2
you would give us a complete.one.
3 MR. SPEIS:
We are trying to do this, and it is not 4
easy, because some of these problems have been around.
The is 5
vested interest.
But Eric believes in this.
6 CHAIRMAN KERR:
I do not think that there is an 7-obligation.
8 MR. SPEIS:
I think that I am taking too much away 9
from Brian's time, because I think that it is very crucial.
10 Let me see.
I do not think that I have any others.
I think 11 that I am almost done.
12 CHAIRMAN KERR:
Do not hurry.
i 13 MR. SPEIS:
So the schedule is that we will have i
O 14 something to the Commission in the next three or four weeks on 15 this subject, maybe even sooner than that.
16 CHAIRMAN KERR:
This subject being an integrated 17 approach?
18 MR. SPEIS:
An integrated approach.
We will try to l
19 address all of these issues.
We can pass a letter for 20 information.
I have a letter which is an outline of the things 21 that we will be addressing in this integrated plan.
And we 22 will just pass it around for your information. It was 23 pre-decisional, but it left our office.
And the NRC got it 6
24 about a month ago, so I think that it is appropriate that you 25 gentleman can have a look at it.
O Heritage Reporting Corporation (202) 628-4888 I
s
-r-+
-m w
-n--
re
,v.
,nu y - _. - -.,, --.
g q-a 8
161 0
1 CHAIRMAN KERR:
We are the last to know.
2 MR. SPEIS:
The last to know.
-3 (Continued on next page.)
4 5
6 7
8 9
10 11 12 13 O
14 15 16 17 18 19 20 21 22 23 24 25 O
Heritage Reporting Corporation i
(202) 628-4888
, _ -_ _.. - _ -- ___... _ _-- _ _ _.___ ~..__ ___
162-O 1
DR. SHERON:
Mr. Chairman, my name is Brian Sheron, 2
I'm in the Office of Research, and I will be telling you now 3
about the IPEs.
4 I was down here last week, if you recall, and I guess 5
'a number of you heard the presentation.
I'm going to try and 6
summarize it putting emphasis on the areas which I think raise 7
the most questions and Dr. Kerr suggested that I emphasize.
8 (Slide presentation) 9 I would like to go through the summary first, very 10 quickly if I could.
One is that we will be issuing the IPE 11 letter as a 50.54(f) request to the licensees.
12 This is a mechanism where the staff can ask for 13 information from licensees in order to determine if a license O
14 should be suspended, revoked, modified, and the like.
15 It has seemed to be the appropriate vehicle to do 16 this.
There were other vehicles one could consider, like an 17 order, but on balance we felt that this was the best.
18 Again, the basis for the 50.54(f) request is the 19 commission's severe accident policy statement.
And just to 20 sunearize real quickly,.the policy statement stated at the 21 plant specific PRAs exposed relatively unique vulnerabilities 22 to severe accidents.
23 That the undesirable risk from some of these unique 24 features could probably be fixed by some relatively low cost 25 changes to either procedures or hardware.
O Heritage Reporting Corporation (202) 628-4888 i
t 163 0
1 And that an analysis should be made of any plant that 2
has not yet undergone a systematic examination, when the NRC 3
and the industry have sufficiently progressed to define the 4
methods of analysis.
F Keeping in mind, the commission did not expect the-6 industry ro go off and do full blown PRAs but rather to look 7
for some sort of an abbreviated cheaper method, you.might say, 8
to do these examinations.
9 DR. REMICK:
Why is that, Brian?
Why is that 10 decision made?
11 DR. SHERON:
I'm not real sure other than that I l '.
think that they felt that the costs to perform such a PRA were 13 not in keeping with perhaps what they felt the benefits were, 14 they could get.
15 DR. REMICK:
Does the staff feel that inevitably 16 though, people will do PRAs or should do PRAs?
17 DR. SHERON:
When we went through our letter 18 developing the letter, I think we concluded that there were 19 certain benefits that a utility would find if they did a PRA 20 and in the generic letter there is actually a short section 21 which describes some of the potential benefits of doing a PRA.
22 But we have not ruled out the option of using the 23 IDCOR methodology, for example.
And we are not saying that the 24 IDCOR methodology is an unacceptable way.
25 DR. REMICK:
I haven't heard about your action native O
Heritage Reporting Corporation (202) 628-4888
m 164 k/
1 program and I guess we will hear about it.
2 DR. SHERON:
Yes.
3 DR. REMICK:
Can that be done fully without a PRA?
I 4
don't know what you Laven't hit, but--?
5 DR. SHERON:
Yes.
6 DR. REMICK:
It can be done without a PRA?
7 DR. SHERON:
It could be done without a PRA, as long 8
as it, it would be based on the IPE, okay.
And the methods 9
that are used for the IPE would have to satisfy the intent of 10 that policy statement, which is namely that there is sufficient 11 to find the outlyers, or the vulnerabilities.
12 (Slide) 13 Since the policy statement was issued in 1985, the 14 staff has had rather intense interactions with the industry, 15 and I'm sure the IDCOR program.
16 And we feel we have reached a point now where a 17 methodology, this abbreviated methodology has been developed to 18 the point where it is sufficient to go forward with the 19 individual plant examinations for internal events.
20 The purpose, again let me just repeat, is for 21 utilities to identify and understand the most likely severe 22 accident sequence, that could occur at their plants, and to 23 evaluate and implement means for improvements that might be 24 necessary and I think most importantly, is we would like to 25 make sure that the staff at the plant that operates the plant b
J Heritaye Reporting Corporation (202) 628-4888
165 Cs
')
1 develops an awareness for severe accident behavior in the 2
plant.
3 (Slide) 4 This slide is not in your viewgraph package.
It was 5
in the package that we gave to the subcommittee.
I think it 6
would be worthwhile to just briefly go through it on the 7
external events.
8 First off I just want to point out that the 9
commission severe accident policy does not exclude external 10 events.
When they said to look for vulnerabilities, did not 11 say for internal initiators only, it's just to look for the 12 vulnerabilities.
13 At this time, we are asking licensees only to proceed (s
I 14 with their IPE's for the internal initiators.
The industry, 15 the NUMARC group, well IDCOR is now under NUMARC, but the IDCOR 16 group did not really develop any methods for external 17 initiators during our interactions with them.
18 What we are doing is we intend to work with HUMARC to 19 develop an acceptable methodology for the external events.
The 20 question that really is being raised with regard to external 21 events is how does one best treat them.
22 We have had a program underway right now for some 23 time at Lawrence Livermore and I am sure we will be down here 24 shortly and I'll tell you the reason why.
To talk to you about 25 the program we have in place, to decide how best to treat the b
uJ lieritage Reporting Corporation (202) 628-4888
166
()
^
1 external initiators.
2 One is that what we have learned so far is that there 3
are some events for which need to be looked at for all the t4 plants and I think seismic is the one that falls into that
'5 category.
6 Some external events are unique only to a few plants 7
and.it is not necessary to require all plants to consider them.
8 And some. external events may be in fact acceptably treated by 9
the existing design base, and in fact nothing has to be done.
10 DR. REMICK:
Brian, could you say that there is some 11 of the internal and that it has been only applied to a few 12 transfer, however you wouldn't require the plant so it doesn't 13 make sense.
14 While you are saying that especially for external-15 events, certainly some internal events at sone plants wouldn't i
j 16 make sense to do, it doesn't apply.
17 DR. SHERON:
Well, for example if there is an event 18 that a certain plant, because they have a unique system, and 19 that other plants didn't have that system, of course they 20 wouldn't have to.
21 DR. REMICK:
Right.
So I guess I don't think that is
(
22 much as argument.
s l
23 DR. SHERON:
Well, one of the, let me give you a l
24 couple more examples, okay.
One is that for example, on i
25 seismic.
There are a number of programs going in the j
l C:)
Heritage Reporting Corporation (202) 628-4888 l
l l
t 167 1
commission in that area right now.
2 There is the eeismic design margins program, eastern 3
seismicity, A-46, and the concern that the industry raised, and
- 4 I think we were sympathetic and in agreement with it.
+5 And that is that it we were going to'ask the' industry 6
to do anything on seismic, either for the IPE or any of those
~
7 other programs, whatever it is we were going to ask them, we 8
ought to ask them once'and once only.
9 If they were going to have to walk down their plants.
10 to look for seismic vulnerabilities, they shouldn't have to 11 walk the plant down three or four times, for each one of those 12 programs.
They should do it one time, and one time only.
13 In order to do that, we have to make sure that O
. 14 whatever we ask the industry to do, will satisfy all of those 15 needs, whatever they are, for all those programs.
16 Another thing is, is that the seismic--
17 MR. WARD:- Brian?
18 DR. SHERON:
Yes, sir?
19 MR. WARD:
I need to interrupt there, just for a 20 minute or two.
,I can take what you just said and substitute it 21 with a few words and make that an argument for not doing 22 internal on external in series.
23 DR. SHERON:
I'm sorry, for not doing?
24 MR. WARD:
Well, I mean you are going to have them 25 going through here for both, and then have them go through an
(
l Heritage Reporting Corporation (202) 628-4888
,w,
-.---y_ms,
,--..r-
,--_m%%-
nm
,.,,,m,..yv w.-,---,r,-
,,,m-,-,,y--,--
--,www~, - - -
168
' ()
1 IPE process of some sort, considering only internal initiators.
s 2
DR. SHERON:
For'now, correct..
3 MR. WARD:
Yes.
Then some time in the future, you 4
are going to have..nem go through a very similar process, 5
considering now tru,rnal initiators.
~6 DR. SHERON:
Correct.
7 MR.. WARD:
Okay.
O DR. SHERON:
But using, perhaps using different 9
methods.
10 MR. WARD:
Perhaps?
11 DR. SHERON:
That's what I was going to say.
That is 12 that for example the seismic experts, and I'm not going to 13 claim to be one, are saying that using a method like a margins 14 approach may be superior than to try to do a PRA and try to 15 come up with actual numbers of probability, mainly because 16 there is large uncertainties in there, there is large 17 conservatisms that are inherent in that, and in fact a margins 18 approach may get you a much better answer.
19 MR. WARD:
Yes, but part of the problem of that is 20 that thu things you decide that has to be fixed in a plant 21 because of whatever an analytical approach, if they can, it 22 might turn out to be very closely related to something you have 23 already fixed, because of the IPE using internal issues.
24 LR. SHERON:
Well I don't know for certain, I am not, 25 you know.
I don't make that--
Heritage Reporting Corporation (202) 628-4888
9
,p i
1 169 1
MR. WARD:' 'Isn't that kind of likely that that will 2
happen?
3 DR. SHERON:
I don't know to tell you the truth.
I 4
don't-think so.
- 5 MR. WARD:
All right.
6 DR. SHERON:
I think what THEMIS suggested, you know t
-7 I would like to just echo, and that is that I think it is 8
appropriate that the staff comes down and gives the appropriate 9
subcommittee a full briefing on exactly what rur approach is to 10 the external initiators and why it is not going to cause any 11 problem to proceed with internal first, and then pick up the 12 external later.
13 DR. REMICK:
Brian, one thing that is brought up when O
14 THEMIS was up is, I still don't see the justification of why we 15 can consider some internal events like fires and floods as 16 external.
Why we continue t at irrationality, is it necessary?
17 Is it just historical?
18 DR. SHERON:
I would presume that is the latter.
19 DR. REMICK:
The point I was trying to make, Brian, 20 is maybe you e n convince me that the seismic, there is a 21 reason, maybe neve is a better way.
A..d if there is then I l
l 22 think it should ba considered.
f 23 But if yeo can give me a justification for seismic, l
24 can you give me a justification for each of the external and 25 particulary those that I consider--
[
(
Heritage Reporting Corporation (202) 628-4888 1
s 170 0
1 CHAIRMAN KERRt. Do you mean right now?
2 DR. REMICK:
Well, no.
3-CHAIRMAN KERR:
In the subcommittee menting--
4 DR. REMICK:
Yes, that's right.
I'm just trying, to 5
go to--
6 CHAIRMAN KERR Let's hear it now, Brian.
Or we do 7
it in the subcommittee.
8 MR. MICHELSON:
How does the human factor into enter 9
this consideration, it's the third factor that has got internal 10 events, external events and the human factor, which can 11-interject with the errors of commission or omission and make 12 things in the nature of severe acci<!nnts.
13 DR. SHERON:
At the subt tee meeting I think we
()
14 made that a bit to your satisfaction but the approach taken. was 15 that we were not trying to invent the new wheel here, and that 16 we would basically expect that the operator actions of omission i
17 or commission as they are currently treated in state of the art 18 PRAs, that is how they would be treated here.
19 Again, where operators are taking action and the IPEs I
20 are taking credit for operators to take certain actions, for i
21 xample better to mitigate an event, we would expect that the l
22 utilities would produce a program ~
I.e.,
the accident 23 management program, that would help give that assurance, that i'
these operators indeed had the proper training and could in 24 25 fact take these type of actions.
1
(
Heritage Reporting Corporation (202) 628-4888 l
l l
i 171 0
1 As for the operator errors of commission, pushing 2
the wrong button and so forth, I think I would have to defer to 3
the PRA experts, because I don't think those are actually 4
considered to the degree that you'r9 talking even in state of 15 the art PRAs.
There is a big uncertainty on the probabilities, G_
for example you would assign.
1 7
Let me just quickly point out that Mr. Beckjord, in I 8
believe it was December, established a senior steering group 9
for external events.
The chairman is Larry Shao of NOR, l'
10 members are Tom Novack, from AEOD, Guy Arlotto and myself from 11 the Office of Research.
12 This four-member steering group has as a charter to 13 provide recommendations to the senior NRC management on how
(
14 best to treat the external events for the purposes of the 15 severe accident policy.
I 16 There have been a number of meetings, and I think we l
17 just concluded two days of meetings.
There are some 18 subcommittees that have been established with the experts I
19 within the commission, and they have been tasked to come up 1
l 20 with proposed methods, will be interacting with NUMARC, and I 21 think right now the schedule is to produce final reports with 22 recommendations within about 18 months, i
23 So what I'm proposing, I think THEMIS was, is that
(
24 sometime in the near future I would expect that perhaps Larry 25 Shao and some of the subgroup chairmen come down and give a
- 10 Heritage Reporting Corporation (202) 628-4888 i
s m
4 172 l)
'/
1 rather intense presentation on what their efforts are in this 2
area.
3 CHAIRMAN KERR:
My agenda calls for a break at about 4
this point.
Do you have any objections to our taking a break 5
at this point?
6 DR. SHERON:
No.
7 CHAIRMAN KERR Fifteen minutes. We'll be back at ten 8
of.
9 (Whereupon, at 2:35 p.m.,
a brief recess was taken, 10 to resume at 2:50 p.m.)
11 (Continued on following page.)
12 13 O
~
14 15 16 17 18 19 20
(
21 1
l 22 23 24 25 3
i Heritage Reporting Corporation l
(202) 620-4888
ll
[
173 f
1 (Continued from previous page)
.s 2
DR. SHECON:
Okay.
What I would like to talk about 3
now is the examination process.
And before I go into this, 4
just let me say that I will be going into a little more detail 5
on the back end, so if you could maybe refrain from the 6
questions, the type that you had at the meeting, until I go 7
through that, because I think hopefully I will answer some of 8
those.
9 One is that, and let me, I can't over-emphasize this, 10 and that is that regardless of whether a utility does a PRA or 11 an ISEM or something else, we think that the biggest benefit 12 that-will be attained from implementing the severe accident 13 policy is that the licensee staff should actively participate 14 in all aspects of the IFE.
So that the knowledge gained 15 becomes an integral part of the operating, training and 16 procedure program.
17 And I think Dr. Siess had a question, was that well, 18 if utility staff t rns over every several years, then is this 19 the benefit going *' be lost.
And I guess our feeling is that 20 this is where the concept of accident management comes in, and 21 accident management programs.
22 And that is if these insights that are gained from 23 the IPE are manifest in an accident management program, and 24 that new personnel as they con.e to the plant and go through 25 their training programs are trained in this accident management O
Heritage Reporting Corporation (202) 628-4888
'174 r
1 program, then these insights and the like would indeed be 2
passed on.
3 The licensees, as I said before, are expected to 4
. conduct a systematic examination of the plant for design 5
operation,.and maintenance emergency operation, and to identify
<6 plant specific vulnorabilities to severe accidents for-both
- 7 core damage and containment performance.
8 Both internel and external' initiators are to be
.9 considered, however, separate from the IPEs or the internal 10 events.
4 11 We want them to quantify their results of the 12 examination for sequences that contribute the most to the total 13 core damage or the large release frequency.
And we will be
(:)
14 defining-large release.
1 15 As I said at the subcommittee meeting, we are working 16 on a otfinition of large releace, and by the time this letter 17 is out, and has to be implemented, then we should have a 1
18 definition for the industry.
i 19 I'm sorry?
20 MR. WARD:
Is there a possibility that there is a 21 need for a definition of the large release of the safety goal 22 policy?
23 DR. SHERON:
Yes, and that is 'here it is being 24 developed right now.
l 25 MR. WARD:
Can it be consistent with that, or are you Heritage Reporting Corporation (202) 628-4888 i
L'
e i
175
/^3 52 1
talking about the same thing?
2 DR. SHERON:
Whatever, what is developed for the 3
safety goal policy will be used for this.
Again, just to 4
reemphasize, we want the utilities to understand where it can 5
possibly go wrong in their plan, and to identify and evaluate 6
means for improving the plant done in containment performance, 7
either via hardware additions modifications, changes in the 8
procedures, changes in training.
9 We would also like the industry, once they have gone 10 through and gained this understanding, they should decide, and 11 I emphasize they should decide, which improvements will be 12 implemented and the schedule for their implementation.
13 1 don't have any viewgraphs right here.
I did at the 14 subcommittee meeting.
We have screening criteria which have 15 been proposed.
These screening criteria are not any sort of an 16 acceptance goal or anything.
17 The screening criteria are primarily only as a 18 threshold for identifying those sequences that should be 19 reported to the commission in their IPE results report, and 20 nothing more.
21 Again, the utility is to determine what things ought 22 to be fixed and take the appropriate action.
We will be 23 reviewing the results, and I will discuss that in a little bit.
24 But if we see something where wo disagree with the utility, the 25 staff's ultimate recourse would be through the 5109 backfit O
Heritage Reporting Corporation (202) 628-4888
y 176
(~)'
c 1
process.
2 DR. REMICK:
Brian?
3 DR. SHERON:
Yes.
4 DR. REMICK:
Would the Jicensee have the flexibility 5
when he does this, he finds out that let's say if we fix this 6
or perhaps for some other information he obtains as a result of 7
doing the IPE, that a requirement from the staff resulting 8
let's say from a USI or something else, just doesn't make sense 9
from a risk significance.
Does he have that option of 10 eliminating that other requirement?
11 DR. SHERON:
Say from a USI?
12 DR. REMICK:
No, let's say it's USI.
13 DR. SHERON:
If the requirement of the USI for O
14 example, is manifest let's say, the maximal I think, or would 15 be in a new rule, okay.
If there is a sound technical basis 16 that licensee can request an exemption to the rule.
And I 17 think the staff would have to look at it on a case by case 18 basis.
19 DR. REMICK:
Okay.
20 DR. SHERON:
If it wasn't a rule, but perhaps just a 21 new interpretation of an existing rule or regulation, the 22 licensee has that same option.
But again we are not trying to 23 have a result here where the licensee is doing something dumb.
24 DR. REMICK:
No, I think it is at this point that 25 some is thought that the ISAP would make sense in the O
Heritage Reporting Corporation (202) 628-4888 O
177 1
'O
~
1 implementation of these things in scheduling and determining i
2 what is the most important, eliminating things that were not
)
3 important so far.
It would not even be at this point that ISAP I
e 4
would make sense.
I j
5 CHAIRMAN KERR:
As I read the draft letter I thought i6 it'said that if it could be demonstrated that something is in
'7 effect satisfying a generic issue, or close to satisfying, that i
8 the staff would consider or request that this satisfy that 9
issue.
I 10 We gave the utilities an option which, and in the
{
11 letter, which says that if you feel that as a result of your 12 IPE -- be it a PRA or the IPEM or something else -- if you 13 believe that you have satisfactorily addressed a generic issue l
O 14 or USI, for example, either by showing that the USI or generic 3
15 issue, that nothing needs to be done to your plant because the 1
16 concern or the risk probability of core melt is low enough, 17 okay.
18 And there is nothing that is cost effective that can j
19 be done to improve it, or if you make a certain change to your I
l 20 plant as a result of your IPE and decide that that chango 21 indeed satisfies one of the concerns of a generic issue, we 22 have told the industry, as a matter of fact we would encourage i
23 them to come in and tell us speci11cally that this is the
~
24 conclusion they have reached and we would review it.
25 And presumably if they used sound reasoning, we would O
Heritage Reporting Corporation (202) 628-4888 l
__ __ r 178 1
find it acceptable and consider that issue resolved for that 2.
plant. 'So the option is there.
3 DR. REMICK:
Okay.
So it sounds like you would have 41 incorporated some of the potential benefits from an ISAP 5
- program, i
6 DR. SHERON:
'From the standpoint that the ISAP'
'7 program treats implementation of generic issues, part of.it,you 8
might say is there.
9 CHAIRMAN KERR:
Mr. Ward?
10 MR. WARDt Bri.n, at the other end of the spectrums, 11 as I recall the discussion in the subcommittee meeting, I think 12 something you may have said or sor. cone said that if the IPE 1
13 identified something that the staff thought the licensee rhould 0
14 fix, but the present regulations didn't require it, you can 15 resort not only the backfit rale, but if necessary you would 16 resort to rulemaking.
17 DR. SHERON:
That's a possibility.
Particularly if 18 it's a generic type of concern.
You would probably go to 19 rulemaking through.
20 Let me go through very quickly if I could the methods 21 of analysis that are acceptable.
The generic letter specifies 22 several options to the utilities that could be used.
23 One is the IDCOR IPEMs front end with the staff 4
24 enhancements that are identified in an SER that would go out to 25 the utilities, plus a containment performance analysis that is Heritage Reporting Corporation (202) 628-4888
c 179 g~)
(_)
I consistent with the guidance in appendix 1 to the generic 2
letter.
3 Another option is they can do a level 1 PRA and then 4
also do a containment performance analysis, again consistent 5
with the appendix 1 guidance.
6 Or they can do a level II or they can submit a level
'7 II or a level III PRA, provided the back end is consistent with 8
the guidance in appendix I.
9 or they can use other systematic evaluation methods, 10 and we are not trying to define any, but for example some 11 utility may have something that is different from a PRA and 12 different from the IPEM.
And we don't want to preclude 13 accepting that.
14 (Slide) 15 The IDCOR IPEM evaluation takes advantage of 16 available IDCOR reference plant analysis.
Similarities among 17 plants, up to date knowledge of severe accident phenomena, but 18 wo acknowledge that the system's analysis is indeed less 19 rigorous than a full scale PRA.
20 The IDCOR IPEM method was submitted to the NRC in May 21 of 1986, it was initially modified by IDCOR in December of 1986 22 to accommodate some comments we sent them and we received 23 further modification in April of 1987.
24 We have completed our evaluation.
And we will be 25 issuing, we have like an SER, an evaluation which hopefully O
Heritage Reporting Corporation (202) 628-4888 1
4 100
(~)
'l 1
will be going out shortly, which will contain these 2
enhancements that I discussed, which would bring it up to a 3
level that we find is acceptable.
4 Just to get a reference point here, at a level of 5
effort we believe that is required of the utility uses, the 6
IDCOR IPEM is equivalent to a level I PRA but is less than full
'7 scope level III PRA.
8 (Continued on following page) 9 10 11 12 13 0
14 15 16 17 18 19 20 21 22 l
23 24 i
25 O
Heritage Reportirg Corporation (202) 628-4888
s 181 (3
kJ 1
(Slide) 2 For the front end, our key areas of concern with the 3
IPEM method is the emphasis on bottom line numbers, namely, the 4
core damage frequency.
We emphasize that the real emphasis 5
should be on identification of plant specific vulnerabilities, 6
the listing of potential areas for improvements, and then 7
selecting the most promising improvements for the 8
implementation.
9 Some other concerty are that all the test 10 applications have the benefit of a full scale PRA which could 11 have influenced the IPE.
12 There was no test application of the revised IPEMs.
13 The BWR IPEM does not assure a complete evaluation of 14 the net impact of the system design and operational 15 modifications on the overall plant risk estimate.
Which is if 16 they make a change to the plant, there's no way they go back 17 and reestimate how that affected the overall risk.
18 And also extreme care, we think, is needed in 19 mod 911ng support states, the criteria for determining when more 20 detailed analysis of support states is needed is left to the 21 analyst which leaves a little bit of a question mark.
22 The staff evaluation, these enhancements as I called 23 them before, address these areas of concern, and hopefully if 24 utility follows the enhancements we've proposed, these concerns 25 would go away.
O lieritage Reporting Corporation (202) 628-4888 l
182
.(')
s/
1 (Slide) 2 Let me:get on to the back end now if I could.
The 3
concerns we had with the IDCOR IPEM'on the back end is that 4
they're too narrowly focused.
We don't think that it provides 5
utilities with the information they would need to examine the 6
containment performance for the full range of accidents.
<7 Again, emphasis on bottom line numbers, namely,
+
- 8
' release fraction.
We emphasize that identification of 9
containment failure mechanisms and timing would allow for 10 identification and implementation of a recovery procedure were.
11 important.
12' The analysis also relies very heavily on the IDCOR 13 views of severe accident phenomenology and the map code.
O 14 And as you can guess from this, some of these views 15 are not in agreement with at least the Staff's views or other 16 experts in.the area.
4 17 There's no recognition of uncertainties in the 18 phenemenology.
i 19 An optimistic assessment of correct operator actions i
20 to maintain containment integrity during severe accidents.
And 21
'I guess this is part of Carl's concern about the human factors.
i 22 And a loose interpretation of the IPEM questions by 23 the utilities.
These are questions that are in the IPEM method 24 that a utility must address.
Could result in a lack of i
25 sufficient attention to measures to mitigate consequences.
(
Heritage Reporting Corporation l
(202) 628-4888 i
l l
I
t 183 1
(Slide) 2
- Let me just -- I'm sorry?
3 CHAIRMAN KERRt Loose attention to-almost anything 4
could result in that, it seems to me.
5 DR. ~ SHERON:
Well, we just felt that it could have 6
been beefed up a-bit.
7 Very quickly, I don't want to spend a lot of time.
i8 These are the enhancements to the front end.
I don't want to 9
read through these but as you can see, these are designed to 10 address the concerns we had for the level of detail and 11 symmetry concern for systems.
12 (Slide) 13 Now, let me get in if I could to the guidance on 14 containment system performance which I understand you had a 15 number of concerns about at the last subcommittee meeting.
16 One is the guidance on performing a containment 17 assessment is indeed available.
NUREG CR-2300 as supplemented 18 with the insights of Appendix 1 of the Generic Letter we think 19 provide a fairly comprehensive guide for doing a back end 20 analysis.
21 If one takes existing PRAs, for example, Zion, Indian 22 Point 3, or the NUREG 1150 plants that have containment 23 performance assessments that are consistent with the guidance 24 in the NUREG 2300, as well as the guidance of Appendix 1, we 25 would consider those as acceptable back end analyses.
O Heritage Reporting Corporation t
(202) 628-4888
?
184
/~T
(_/
1 I'm going to walk through in a second in the next 2
view graph an example of an examination for a PWR which is 3
derived from 2300 and Appendix 1.
4 Just in principle again the IPE should provide a 5
realistic evaluation of the containment performance.
The aim 6
is to develop event trees to cover the range of uncertainty and 7
our guidance that we're giving utilities therefore focuses on
- 8 containment failure mechanisms and timing, just using releases 9
from tables.
These are look-up type of release.
10 And integrate systems response probabilistically into 11 simplified but realistic containment event trees to allow for 12 recovery or other accident management procedures.
In this to 13 examino all classes of sequences that have a significant
}
14 probability, and to recognize and account for the 15 uncertainties.
16 Now, this is a key bullet here.
That is, fo those 17 phenomena that are not well understood, that are difficult to 18 quantify or there's considerable uncertainty in their results, 19 no specific containment performance analysis is required at 20 this time.
However, consideration should be given to 21 strategies to deal with those issues while awaiting the generic 22 resolution.
23 (Slide) 24 DR. SHEWMON:
Would direct containment heating come 25 under that sort of thing?
O Heritage Reporting Corporation (202) 628-4888
185 f
1 DR. SHERON:
Yes.
And for the BWR it would be minor 2
melt through issue.
3 CHAIRMAN KERR:
Excuse me.
What is meant by the IPE 4
should develop event trees that cover the range of uncertainty.
5 What does that mean?
It's about midway.
Aim the IPE to 6
develop event trees that cover the range of uncertainty.
What 7
is meant by that?
8 DR. SHERON:
Well, for'every event tree, there's the-9 numbers and so forth, you know, for the branches.
You cannot 10 assign a specific number.
Each thing has uncertainties to it.
11 CHAIRMAN KERR:
But who knows what the range of 12 uncertainty is?
13 DR. SHERON:
Is that in 2300, do you know?
O 14 HR. SPEIS:
Well, what is meant there actually that 15 is referring to two or three of the generic issues where there 16 are a variety of views, a wide range of views and depending on 17 what assumptions are made, you can get extreme --
18 CHAIRMAN KERR:
I recognize that.
I'm saying, who 19 knows what the range of uncertainty is?
20 MR. SPEIS:
Well, they're discussed in a number of 21 documents.
22 CHAIRMAN KERR Do you give references in the letter?
23 MR. SPEIS:
Yes.
We have discussed these issues with 24 IDCOR, for example.
25 CHAIRMAN KERR For example, there's some documents
(
d Heritage Reporting Corporation (202) 628-4888
186
()
I that would say the range of uncertainty is from zero to one on 2
probabilities.
I've seen such documents.
Is that what is 3
meant?
4 MR. SPEIS:
Well, on some of the decisions I hope 5
were better than that, but --
6 CRAIRMAN KERR:
It just does not seem to me that that 7
statement, if it appears in the letter as such, and I don't 8
remember, gives much guidance.
Cover the range of uncertainty.
9 MR. SPEIS:
Give some more specific reference or more 10 11 CHAIRMAN KERR:
Maybe you can't be more specific than 12 that, but if you can't, then I wouldn't know what you meant.
13 DR. SHEWMON:
There's no meaning to the term unless
()
14 you specify a confidence level.
15 MR. SPEIS:
Well, unfortunately on those two or three 16 issues, the range is anywhere from ten percent to 100 percent, 17 or 90 percent.
It's almost like zero to one.
It's almost like 18 that.
But there are ways of getting around them, and maybe 19 Brian will talk about that, especially on the Mark I liner.
20 DR. SHERON:
What we've done here, you have not seen 21 this before at the subcommittee meeting, is we've gone through, 22 if you go through NUREG 2300, and you can see that these steps 23 here are very much akin to what is provided in Chapter 7 of 24 flUREG 2300.
And we've included in these descriptions here some 25 of the guidance that's also provided in Appendix 1.
lieritage Reporting Corporation (202) 628-4888
(
n 4
187 t
i k/
1 I don't want to read these all to you for the sake of 2
time.
I can go through the main steps.
3 One is the plant familiarization step.
4 Step two is sequence grouping which is where you get 5
stuff from the front end analyst.
6 Step three is determination of the containment 7
failure modes.
And a list of these is provided in Table 7.1 of 8
And you don't have to do a full blown analysis of 9
the containment capability if, for example, you can show that 10 the margins, the safety margins from another analysis of a 11 similar containment, or if you just want to rely on for example 12 what the ASME code or other building codes require of the 13 containment in terms of structural margins.
,-U 14 (Slide) 15 Just quickly, this is not in your package but this is 16 Table 7.1 of NUREG 2300, and you can see it does provide 17 guidance in terms of what kind of failure modes and mechanisms 18 indeed exist for the PWRs.
19 CHAIRMAN KERR:
Is internal failure defined 20 somewhere?
I asked this question for Mr. Siess, and I kept 21 waiting for him to ask it, and he didn't.
22 DR. SHERON:
Do we have a definition?
23 DR. SilEWMON :
Failure to contain.
24 MR. SPEIS:
We have.
I don't have it with me.
25 Costello has given us.
(Q
%i fleritage Reporting Corporation (202) 628-4888
a f
188.
O 1
CilAIRMAN ' KERR:
But it would be readily'available to 2.
an analyst.
3 MR. SPEIS:
Yes.
4 DR. REMICK:
.A hundred percent a' day?
5 MR. SPEIS:
Excuse me?
6 DR. REMICK:
A hundred percent a day?
I 7
MR.'SPEIS:
I think it's even higher than that, yes.
i8 DR. SHERON:
Step 4 is development of containment 9
event tree.
You can do it by analogy with a similar plant 10 design or selecting event tree headings used in Nodal! questions 11 in table 7-3 or 7-4 of NUREG 2300.
You want to divide the 12 accident into major time periods which are events before core i
13 melt and events after core melt, and then also divide the O
14 events related to in-vessel and ex-vessel phenomena into high 15 pressure sequences and low pressure sequences.
16 Step 5 is determination of the containment 17 challenges.
18 (Slide) 19 Step 6 determination of the source term magnitude, i
20 again, by analogy to releases from reference plant analysis or i
21 you can use an integrated code analysis and calculate your own t
22 source term.
l 23 Quantify event trees, perform sensitivity studies to i
24 identity parameters that are likely to have a large effect on I
25 the likelihood or time of containment failures, and vary those e
Heritage Reporting Corporation l
(202) 628-4888
)
i
/
189 il parameters over a feasible range.
And again, this is I guess 2-your question on uncertainty, Dr. Kerr.
3 Professor Shewmon?
4 DR. SHEWMON:
Do you ever sort of wonder whether it's 5
all cost ef fective if any believe that the list plant, as the 6
Commission has said, that the plants are really not an undue 7
risk to the health and public safety, or health andLsafety of
- 0 the public?
9 DR. SHERON:
Yes.
I think that this is cost i
10 effective to do this analysis.
I can't quantify it.
Again, we 11 think the biggest benefit that we're going to derive from this 12 whole process is, one, that if there are any outliers or 13 vulnerabilities out there, we're going to catch them and get O
i 14 them fixed.
15 Ilumber two is that we hope that utility personnel are i
16 going to get this much better understanding of the way their 17 plant is expected to behave durir) a severe accident and to be i
18 aware of what these uncertainties might mean in terms of how 19 they have to respond.
20 And thirdly that they would be able to take these l
21 insights that they gain from this analysis, not only the front l
22 end but the back end, and put them into some sort of an
)
j 23 accident management program at the plant.
So that if indeed, 24 God forbid, they were faced with a severe accident, they would 25 have a structured program somehow that would allow them to deal l
(E)
Heritage Reporting Corporation
]
(202) 628-4888 f
0
-+w--
-,,-m,-
---+,-,-,v-
,+-,my----e,-w-m-e-n
190 0kl 1
with it, rather than sort of by a seat-of-the-pants approach.
2 So I think from a qualitative argument, I think these 3
are well worth the money.
4 MR. SPEISt I w6nt to add something that I think is 5
very crucial.
If you read Appendix 1 very carefully, you will 6
see that one of the things we're trying to bring out is the 7
margins that these containments have beyond the design basis 8
which are very crucial to accommodating some severe accidents.
')
Maybe not all of them, but you have to understand them.
And if 10 you don't understand that margin, then you're not able to 11 utilize it.
12 And what we're trying to do with this accident 13 management and with this type of containment performance O
14 analysis, that you go through this analysis, understand the 15 margins, understand the failure times, and see what are the 16 appropriate things that can be done.
And if you don't kriow 17 that, then you have no way of being able to cope or manage.
18 And this is one of our key differences.
19 And I don't think maybe it's not a difference with
..ially by emphasizing the bottom line 20 IDCOR anymore, but 1 21 number, the release fraction, and not stressing the 22 deterministic aspect of the capability of the containment, when 23 does it fail.
For example, as all of you know, large dry 24 containments are able to accommodate the initial pressure spike 25 that can occur by interacting one hundred percent of the column O
Heritage Reporting Corporation (202) 628-4888
191
()
I with water, okay.
2 But as I say, there are some failure mechanisms that 3
the~ containment cannot cope with.
But there are some of them 4
that they can cope.
And it's important that those utilities-5 understand their containment, understand their existing 6
margins, and then put them to use.
i 7
DR. SHEWMON:
Yes, but that's an argument that says i
8 there's just no such thing as a reactor that's too safe.
9
.That's an argument we've been going back and around on for a 10 couple of years.
And we started out this exercise looking for 11 outliers.
And now you end up going through all this long list i
12 of drills that they have to do assuming there's going to be an i
13 outlier found, and if not, whether they find one or not is sort 14 of coincidental to all these various things they have to do for j
i 15 the good of their souls.
l 16 DR. SHERON:
We've discussed accident management a 17 bit.
I don't want to go through the slide just for the sake of j
18 time.
I just want to point out that we're telling the industry i
19 in the letter that we believe accident management is something l
20 that they need to take -.
1 21 Yes, I said I just want to skip that slide just for 22 the sake of time.
But I just want to point out very quickly i
23 that we are not asking the industry to produce an accident I
j 24 management program as a requirement that must be submitted h
j 25 along with their IPE results.
What we are telling the industry 4
(
i Heritage Reporting Corporation (202) 628-4888 l
192
/O k/
1 is that when you do your IPE, you will be gaining a lot of 2
insights which will most likely be able to manifest themselves 3
in an accident management program.
4 The staff is working with NUMARC right now to define 5
the scope and how best to implement a severe accident 6
management program at plants.
And that somewhere down the 7
road, you're going to have to do this.
However, if you come 8
across something as part of your IPE that could be construed I 9
guess in the context of accident management for which you 10 believe a substantial improvement in safety can be made. then 11 we would certainly expect the utility to implement it at that 12 time, and not defer it just because you know somewhere down the 13 road, there'll be this thing called accident management.
O 14 What they conclude needs to be done could be 15 integrated into a larger accident management program later on.
16 That's all I wanted to say about the accident management.
17 (Slide) 18 I also just want to point out that we looked at USI 19 A-45, because we were coming to a conclusion on A-45 right 20 around the time the IPE letter was being developed.
And the 21 conclusion on A-45 in a nutshell is that there was no generic 22 cost effective fix that could provide a substantial improvement 23 to decay heat removal across the board for operating plants.
24 What we found is that the vulnerabilities in decay 25 heat removal systems were in fact plant unique and were O
Horitage Reporting Corporation (202) 628-4888 i
l
f
/
193 O
1 different, depending upon the plant.
For the six reference 2
plants that we looked at, about half of them appeared to have 3
the majority of the core melt probability coming from external 4
initiators like seismic.
The other half roughly from internal 5
initiators and the like.
6 What we concluded was that if for example the IPE 7
program did not exist, what we would be going forward with as a 8
recommendation would be for each plant to conduct some sort of 9
a systematic search or a limited scope PRA on their decay heat 10 removal system to search for these vulnerabilities.
Given that 11 we had a program in place that in fact did that, not only for 12 the decay heat removal system but for the whole plant, coupled 13 with the fact that if you did for example a PRA on a decay heat 14 removal system, we think that's about 80 percent of a complete 15 PRA.
16 We concluded that we would subsume the A-45 program 17 and its resolution into the IPE.
So the utilities by 18 conducting their IPE will in fact have looked at decay heat 19 removal system and we are proposing that as our resolution.
20 (Slide) 21 Let me just quickly run through the IPE results 22 review.
What are we going to do with this stuff when we get 23 it.
Just for some logistics, there's 109 licensed plants.
If 24 we look at replicate plant, we would expect the IPE submittals 25 that we would see to be around 80.
O Heritage Reporting Corporation (202) 628-4888
194 s
1 We are proposing, we think that we can do a review of 2
cach submittal in about six person months.
And this is going 3
to be just sort of an overview type of review.
We're not going 4
to sit there and get down and look at all the numbers up to 5
five dccimal places.
But we would spend about six person 6
months on this review, including the review of any proposed 7
modifications that the utility makes.
8 We also just for logistics expect these submittals to 9
be coming in over a period of about three years.
We're 10 estimating that the total effort to do this review is going to 11 be about 16 person years effort per year.
We're proposing to 12 do it with eight contractors and eight staff, four staff coming 13 from NRR, four from RES.
The lead for those reviews would O
14 remain in RES.
15 The four staff coming from NRR and probably the four 16 from research would be on a rotating basis.
In other words, 17 some staff member might spend a year working on some reviews 18 and then they would come off and another person would come in.
19 So over the three or four year period the stuff comes in, we're 20 going to have a substantial number of staff that have gotten a 21 good year of experience doing these type of reviews.
22 At about 175 K per person year at the labs, we're 23 talking approximately $1.4 million a year expenditure to 24 support this effort, And as I said before, in case there's a 25 disagreement with what a utility proposes either to fix 0
Heritage Reporting Corporation (202) 628-4888
195 1
something or if they propose not to fix something, the staff 2
would pursue it in accordance with the backfit rule or perhaps 3
an order.
4 We'd like to propose here that these reviews be done 5
on a team concept.
We would have an NRC person be the leader 6
of each team.
And each team would have several plant systems 7
specialists, a PRA specialists.
Also, these teams could draw U
upon the expertise within the staff as necessary for
-9 specialized areas.
It's not like they're restricted.
If 10 there's a structural problem, they may have to go to a 11 structural expert within the staff.
12 We would propose probably two teams would do only 13 large dry containments.
One team would do the Mark I and Mark O
14 II containments.
And one team would do the Mark III and the 15 ice condenser containments.
16 DR. REMICK:
Over three years, when would you 17 envision the first ones coming in?
Would you state that?
18 DR. SHERON:
I would guess that the first ones would 19 7ome in within several months of when the letter goes out.
Por 20 example, Susquehanna's already done an IPE.
And it's very 21 possible that once they receive the letter, they could turn 22 around and submit something.
23 DR. REMICK:
But those who had not done one, when 24 would you expect, how long do you expect them to take?
25 DR. SHERON:
We're not real sure.
I guess an IPE O
Heritage Reporting Corporation (202) 628-4888
196 1
might take a year, a year and a half, depending on what type of 2
internal review they have to do.
A lot of it depends on 3
funding.
4 DR. REMICK:
And suppose they decide to do a PRA and 5
haven't done it.
Wouldn't it take longer than that?
4 6
DR. SHERON:
Yes, I guess several years perhaps.
7 DR. REMICK:
So we're talking about a process here 8
that could maybe take six years altogether maybe?
9 DR. SHERON:
No, I hope not.
It depends if they all, 4
10 I think somebody once characterized it, if they all try and 11 treat us like the IRS, and all come in on April 15th, yes, we 12 may have a big backlog in terms of what we have to review.
But 13 we are going to be encouraging the plants that have dcne these O
14 systematic examinations to get them submitted, and hopefully we i
15 con find a way that we can balance our resources for review 16 with the IPEs coming in.
17 DR. REMICK:
Do you expect that they would all be in 10 within roughly three years after the letter went out?
Is that i
19 what you meant by that?
20 DR. SHERON:
- Yes, j
21 MR. SPEIS:
The majority of them.
l 22 DR. SHERON:
Yes, there may be some that for certain
'3 hardship reasons or whatever request more time, and we'd have 1
24 to look at those on a case by case basis.
25 (Slide)
O Heritage Reporting Corporation (202) 628-4888 I
197 1
What are:we going to do with these results when we 2
- g e t t h e m,-- b e s i d e s r e v i e w t h e m.
Well,-one to review, hopefully 3
is we would-insure that an adequate analysis of plant design.
4 and operations was done to discovery.any vulnerabilities to 5
core damage.or unusually poor containment performance.
6 Dwe're going to act a little bit as a clearing house 7
here in the sense that we'll be looking for consistencies in 8
the identification and treatment of leading core damage 9
sequences.
If I have five plants that are identical or nearly 10 identical, and four of them come in and identify a certain 11-sequence with a significant. probability and the fifth guy 12 doenn't hnve that identified, we're going to have to go back 13 and ask that plant, h]w come you didn't get this.
Maybe they O
14 missed it by an oversight or maybe they have a reason.
But 15 we'll be basically the ones assuring this consistency.
16 We want to make sure that the sequences and the 17 sequence frequencies that they identify are indeed appropriate 18 and they don't look out of line.
We'd like to independently 19 conclude that the decisions the utilities make on whether or 20 not to make improvements are indeed justified.
21 And I think overall, we want to allow the Commission 22 to conclude that severe accident policy has been responsibly 23 implemented by the industry.
24 Our consideration will include, when we look at their 25 results, the consideration will be both for quantitative O
fleritage Reporting Corporation (202) 628-4888
- -. _ _. _ _ _ _ _ _ _ _ _, _. - _. ~..
.m
U
'198 I
measures and.non-quantitative judgment to determine if'a. plant 2
is acceptable or not.
3 If NRC consideration indicates that plant-design or 4
operation could be enhanced by additional' protection, if it's 5
beyond the regulations, as I said before, implementation would 6
be in accordance with 50.109.
If we see something that 7
indicates a plant design or~ operation should be changed'just to 8
meet tb4 regulation, then again, implementation would be 9
without regard to cost except to look perhaps among the 10 selected alternative ways to meet the regulation.
11 The IPE results will be used to identify severe 12 accident vulnerabilities that are generic to a class or several 13 classes of plants.
When we get all of these IPEs in, we're 14 going to be looking to see if there's something that keeps 15 cropping up that's almost generic.
16 And we're going to ask ourselves to question is this i-
~
17 a deficiency in the regulations, why is this allowed to occur 18 in all these plants.
If a generic deficiency is identified, 19 this is where we would use the safety goal to determine if a 20 regulation modification needed to be made.
l y1 So this is basically how we would propose using a l
22 safoty goal with the IPE results.
And I think that is 23 consistent with the guidance and the recommendations of the l
24 Committee.
i 25 I think that ends my presentation on the IPE program.
(
Heritage Reporting Corporation (202) 628-4888
"'T vf
=w g
?,,.
er
1 199 1
I would just summarize by saying what Eric told you 2
before, and that is we feel comfortable that there's enough 3
information and we've worked long and hard with the industry to 4
the point where we're ready to go forward with this and get-it 5
going.
6 We recognize that the letter may not be perfect.
7 There's always' areas where one could improve it a little bit
'O here or there, but we think it's good enough-to get going and 9
start the process with the industry.
We recognize that once 10 the letter goes out, we're going to hold workshops with the 11 industry, listen to their questions, listen to their concerns 12 about the letter, try and resolve them as they come up.
13 We'll be back to the ACRS, both the subcommittee and O
14 the full committee, giving you progress reports on what's going 15 on and what we're learning from this whole thing.
16 So with that, if there are any other questions, that 17 concludes my presentation.
18 CHAIRMAN KERR:
Any questions?
19 (No response) 20 CHAIRMAN KERR:
Brian, would you be willing to say a l
21 little bit about the guidance that is mentioned in several.
I t
22 gather there's going to be some additional guidance prepared in 23 the form maybe of a description of the review process?
l 24 DR. SHER 0ll:
Yes.
One of the things we are doing 25 right now is preparing a review plan for the Staff.
This would O
Heritage Reporting Corporation (202) 628-4888 1
l
200 7
t i
\\/
1 be akin to the standard review plan that we use for licensing 2
purposes.
But this would provide guidance for these review 3
teams on what to do when they get an IPE in.
We'll be 4
developing that.
5 We'll be back down to the subcommittee discussing 6
that review plan with them.
We will be sending it cut to the 7
industry so that they can see what the Fraff is going to be 8
looking for when they make their submittals so that they can 9
assure that they're touchlag on the right points, the points 10 that we're looking for.
11 CHAIRMAN KERR:
Any questions?
12 (No response) 13 CHAIRMAN KERR:
No further questions, I guess.
()
14 Mr. Joe Colvin is here from NUMARC.
I asked if he'd 15 be willing to say a few words about NUMARC's current position 16 on this, because I thought it'd be helpful to the Committee to 17 have that.
18 MR. COLVIN:
Gomd afternoon.
19 I'm Joe Colvin.
I'm Executive Vice President of 20 NUMARC, and I'm here today with you primarily because my key 21 technical people conveniently found themsalves on travel.
And 22 after I came in about the time that Dr. Spies was opening up, 1 23 can understand why.
24 So let me go on.
I think that, gentlemen, as you 25 know, we've had a tremendous amount of worx and attenti,n given
("%
k_)
Heritace Reporting Corporation l
(202) 628-4888 l
l
201 t'
1 to severe accidents since the occurrence of the accident at 2
TMI.
And we believe that the work that's been done through the 3
NRC research program and through the industry,thr) ugh both the a
IDCOR program and the EPRI research program has greatly
~5 extended the state of the art knowledge in the severe accident 6
arena.
7 You've had many briefings on this work over the past 8
years, and I don't intend to cover that.
I certainly don't 9
have any view graphs or slides.
And the work done by the NRC 10 Staff and the industry has really culminated in the issuance of
' 11 the policy statemsnt in 1985.
And since that time, both the 12 Staff and the industry have been working on the methodology to 13 implement the Commission's policy statement.
14 The industry through the IDCOR program developed the 15 IPEM, Individual Plan Evaluation Methodology, because we felt 16 that that would provide a standard, cost effective, easily 17 reviewable method for performing the IPEs called for in the 18 Commission's policy.
19 It's probably appropriate to digress a second and 20 respond to a question perhaps answer the question Dr. Remick j
21 had or to support the comments by Dr. Sheron.
We went after l
l 22 the IPEM not only because it was cost effective but because we l
23 desired that each of the plant operators were involved in the 24 process so that they could carry that forth once knowing their 25 vulnerabilities, and carry that forth as Dr. Sheron indicated, O
Heritage Reporting Corporation (202) 628-4888 l
l l-
1 202 1
into the plant operations, improvements in plant operations, 2
procedures, and so forth.
3 In addition to the development of the methodology of 4
the IPEM, we conducted eight trial applications of this 5
methodology for verification and demonstration.
And both the 6
methodology and the trial applications were reviewed by the NRC 7
Staff.
The interactions between the industry and the Staff and 8
this proposed methodology yielded significant improvements, I 9
think, in the methodology and substantial agreement on the 10 front end portion.
11 We still have come differences that remain, as 12 indicated by the comments and the slides you saw earlier, 13 principally on the adequacy of the back end evaluations for ft~'1 14 source term analysis.
However, we believe that the industry 15 and the Staff have a good understanding of the relative 16 positions and differences regarding that methodology.
17 The industry hasn't seen the draft generic letter, 18 given its predecisional status, but we've had discussions with 19 the Staff regarding the contents of the letter and believe that 20 we understand the Staff's intentions.
And I think we've 21 participated in the ACRS subcommittee meeting in the audience 22 and listened to the discussions and had some of the same 23 concerns that were expressed or raised at the meeting.
l 24 However, given that the severe accident p.licy 25 implemeatation is going to be a substantial undertaking l
(
x l
\\
l Heritage Reporting Corporation (202) 628-4888 l
l l
,4 203 p/, -
's 1
requiring both a continuing dialogue between the industry and 2
the Staff, we feel that the generic letter should be issued 3
without any further delay.
We'd like to get that moving and 4
get this process going in the industry.
5 The industry's been working on this with'the Staff 6
for nearly eight years.
It's time to move forward.
We 7
understand that the letter will allow IPEs to be accomplished.
8 either using the IDCOR IPEM, a level 2 PRA, or other 9
appropriate analyses as agreed to by the Staff.
10 All these analyses are significant undertakings by 11 the individual utility involved, and there are many concerns 12 with some of that, the state of the art of such methodologies.
13 But we feel that that state of the art is adequate for 14 understanding the individual plant responses to transient and 15 accident conditions.
16 We realize that no methodology can be perfect in such 17 a difficult and complicated area, and so we'd like to go forth 18 and put this in progress and work with the Staff to resolve 19 some of these difficulties.
20 The IDCOR program, as the IDCOR program, has come to 21 a close.
It's satisfied its intended purpose that we started 72 back in 1980.
NUMARC's presently forming a working group, as 23 indicated by Dr. Sheron, carry on the work in severe accident 24 areas.
25 The two initial activities of this working group will Heritage Reporting Corporation (202) 628-4888
204 1
be to consider the questions of' accident; management programs 2
and the extent of external events with regard to severe 3
accident conditions.
Particularly with regard to accident 4
management we consider it vital that the IPEs whether done by 5
the IPEM or done by a PRA analysis proceed forward at this 6
time.
Such evaluations will be necessary to provide the 7
detailed information to each of the plant operators on the 8
response and vulnerabilities of his plant and the adequacy of 9
his procedures in the event of a severe accident or transient 10 condition.
11 We feel that this information must be known before 12 the industry can proceed with consideration of the adequacy or 13 improvement of accident management plans and strategies.
We O
14 further feel that plan evaluations can move forward with the 15 issue of external events even being considered subsequently.
16 Industry intends to work with the Staff to define the concerns 17 in external' events to assess on going efforts to address that, 18 and to determine what further action may be necessary to 19 satisfy external event concerns from a severe accident point of 20 view.
21 As I stated, the industry and the NRC have been 22 discussing these for a long time.
We feel that the generic 23 letter should be issued and we'll work with the Staff in 24 continuing activities to work out these other issues and we 25 think that through that process, we'll make substantial gains O
Heritage Reporting Corporation
(?02) 628-4888
t 205 i
h
' l ~-
in safety.
2 Thank you.
3 CHAIRMAN KERR:
Thank you, Mr. Colvin.
4 Are there any questions?
5 DR. SIESS:
It seems to me that you_ visualize this as 6
a-continuing of effort, start with this and just build up and 7
eventually somewhere reach a solution to the severe accident 8
issues.
Is that about the picture?
9 MR. COLVIN:
We].1, I'm not sure that I'm as 10 optimistic that we'll ever reach the end point at this.
I 11 think that as we continue to learn more, we're going to 12 continue to change that process and make appropriate 13 improvements.
'( )
14 DR. SIESS:
But you talked about this being a good 15 first step?
16 MR. COLVIN:
I think this is a good first step.
i 17 We've got to get this moving so that the plant operators, the 18 IPEM process, if we use the IPEM process, what we do is we put 19 in place a methodology to look for vulnerabilities at the 20 plant.
And we can use the plant people to do that.
It's about 21 a two man year effort process to do that.
l 22 The outcome of that process, though, are really l
23 things that the plant operators themselves can apply to making 24 improvements in safety at their plant.
When they find an 25 issue, an area in which they have some vulnerability, then they 1
l l
()
Heritage Reporting Corporation (202) 628-4888
i 206 1
can put in place actions to reduce that vulnerability.
2 As we go forth, as Dr. Sheron' indicated, in the
~
-3 external _ event area, one of the things that we are doing is 4
working with the Staff in the area of seismic. 'And we're 5
incorporating the work of the Seismicity Owners Group that's
.6 been at work through the Electric Power Research Institute, and 7
the Seismic Owners Equipment Qualification Group and 8
incorporate all the' seismic issues into one, so as we move 9
forth, we can address those all at once.
So_I think this is a 10-continuum of effort, and as we uncover new issues and new 11 areas, we'll address those in the same appropriate manner.
12 DR. REMICK:
Joe, would you have an idea on how many s
13 of the utilities might choose the IPEM versus doing a PRA?
Can O
14 you hazard a guess?
15 MR. COLVIN:
Crystal ball, I'm not.
To be quite 16 honest, I'm not sure.
17 John, do you have any insight into that perhaps, in-18 discusslons?
I think it's going to be determined for us by 19 where the utilities may want to go in the future.
And what 20 resources they have available to them.
l 21 A number of utilities have completed Level 1 PRAs.
32 And the question they're going to be faced with is can they now 23 go back and look at that and see if that will be an acceptable 24 means of satisfying a portion of the IPE generic letter.
In 25 any case, even though you've done a PRA in the past, you're O
Heritage Reporting Corporation (202) 628-4888
6 207
(,)
1 going to have to go back and revisit those issues and make sure you've covered all the bases, even if it was a Level I,
Level 3
II or Level III.
4 So I think the utilities are going to have to assess 5
that when we see the generic letter, see what Appendix 1 really 6
contains and make a determination and base that upon the 7
existing resources and what you've done in the past.
I don't 8
have a sense of the numbers of the 54 utilities as to which way 9
they'd go at this point.
10 DR. SIESS:
You mentioned seismic, but you didn't 11 mention any of the other external initiatives.
12 MR. COLVIN:
I was using seismic only as an example.
13 We have to deal with all external events, and we're working
, -)y
\\.
14 with the Staff in those areas.
15 DR. SIESS:
It seems to me it's a little different 16 though because there's a very good chance with our seismic 17 margins work and the SQOG work that seismic can be treated 18 almost outside of PRAs.
And I don't think the other external l
19 events can, flooding, fire, and so forth.
And I think it's a 20 much smaller part of the problem.
i 21 CHAIRMAN KERR:
Other questions?
l l
22 (No response) l 23 CHAIRMAll KERR:
Thank you very much, Mr. Colvin.
l 24 MR. COLVIll:
Yes, sir.
Thank you.
l 25 CHAIRMAll KERR:
Well, it's now 4:45 and hence it is l
l Heritage Reporting Corporation (202) 628-4888
208
.<j.
k/
time for a ten minute break before we begin the next session on 1-1 2-containment systems.
3 (Brief recess is taken.)
4-CHAIRMAN'KERR:
Mr. Ward, do you have any comments?
5 MR. WARD:
Under Tab 6_there is~a good report G
furnished by Mr. Houston.
The issue of concern about the' Mark 7
I containments was originally discussed at least in the formal-8 committee meetings' back in 1986 when Bob Bernero came in.
9 Then in late December, Jerry Hulman, who's a speaker 10 today, came in and discussed the proposed staff action plan to 11 resolve these issues.
One of the important steps in that 12 action plan was to hold a workshop in which interected parties 13 in the industry and the Government and elsewhere, if there is O
14
'an elsewhere, were invited.
That was held in February in 15 Baltimore.
16 I attended that and Mr. Houston did.
And in fact he 17 again provided a good summary of that dated March 22nd and sent l
18 it out to all of you.
At that time, the Staff indicated they 19 had an interim plan for a resolution of the issue.
I think the 20 schedule for that may have changed a little bit.
But at any 21 rate, Mr. Hulman has volunteered to come into the meeting today 22 and provide us a briefly and I guess some preliminary 23 conclusions about the Staff's analysis of the situation.
24 I don't think we will want to write a letter on the 1
25 subject at this meeting, but if you're so inspired by what you O
Heritage Reporting Corporation I
(202) 628-4888 1.
~
209 I
hear from Mr. Hulman or-from each other, we can do that.
'2 Any other comments from other members?
3 (No response) 4
.MR. WARD:
Mr. Hulman, please go ahead then.
5 MR. HULMAN:
First, I'm not Bob;Bernero.
- Secondly, 6
the.' issue 'of a letter, when I originally suggested a schedule 7
for this presentation, I had hoped that we would have an 8'
interim report to the Commission with some recommendations and 9
be able to discuss that with the Committee.
10 That hasn't happened.
What has happened is that we 11 have prepared a status report and it is in the concurrence 12 chain now, and I expect it to reach the EDO and the Commission 13 within the next couple of weeks.
If after you see that status O
14 report, you'd like a further briefing, I'll be more than happy 15 to come back down.
16 My present intent is to try and hold to the schedule 17 that we told the Commission in December on Mark I's and that 1
18 is, have a final report and recommendations by the end of 19
. August, 20 (Slide) 21 With that as an introduction, let me go through a Q2 little summary of where we're at and start off by saying, why 23 are we concerned with Mark I's?
We are concerned because they 24 are relatively small volume containments with suppression pools 25 that result in a close coupling of the containment performance OV Heritage Reporting Corporation (202) 628-4888
210-i (i/
4 s
1-with the reactor systems, as opposed to large drive.
2-Another reason is that in Draft 1150, there was a
.3 wide range of conditional containment failure likelihood 4
primarily due to the possibility of liner melt thre"qh.
5 Even without the issue of liner melt through, Draft 6
1350 indicated a relatively high early conditional containment 7
failure probability.
And also there's a wide range of core 8
melt probabilities based on the informaticn we have.to day.
9 (Slide) 10 And this is a summary of the PRA core damage 11 information that we have today for Mark I.
You can see there's 12 quite a variation.
13 DR. LEWIS:
Could you explain the number in the O
14 footnote on the bottom left of that?
15 MR. HULMAN:
There is a mistake, a typographical 16 mistake.
17 DR. LEWIS:
Ah, that explains it.
Thank you.
18 MR. HULMAN:
I apologize for that.
There is a 19 typographical mistake.
It should be.002.
20 DR. LEWIS:
Thank you, f
21 MR. HULMAU:
The range of core melt damage 22 probabilities that have been estimated is troubling to me.
And l
73 I'm hopeful that the IPE process is going to shed a little more L
24 light on it.
l 25 MR. WARD:
What's the reason for the hopeful?
O Heritage Reporting Corporation (202) 628-4888 l
1 211 6,,I
!1 MR. HULMAN:
Pardon me?
RJ 2
MR. WARD:
How?
I mean, I'm curious about why you're 3
hopeful of that.
4 MR. HULMAN:
Well, I think the use of either the IPEM 5
or PRA techniques should give us a better understanding of the 6
range of core melt probability.
For example, the top value in 7
this table was done as kind of a quick study and a conservative 8
one for TVA.
9 CHAIRMAN KERR:
Just a minute.
Do you want to make 10 your sermon again, or should we print it and?
11 MR. HULMAN:
I'm sorry?
12 CHAIRMAN KERR:
You used the term, conservative, in 13 connection with the PRA.
(}s f
14 DR. LEWIS:
Let's just stipulate my usual speech 15 about conservative.
We should tape it some day so I don't have 16 to go over it.
Best estimate.
17 MR. HULMAN:
I understand that comment.
I'm saying 18 that in the particular case of this one value in the table, I 19 think that was conservatively done, not a best estimate.
20 DR. LEWIS:
I'll stipulate my speech about best 21 estimates.
22 MR. WARD:
We sort of got side tracked there.
You 23 didn't finish with your answer.
24 MR. HULMAM:
Well, I guess my answer to your question 25 is I think a review of all the Mark I's is liable to produce Heritage Reporting Corporation (202) 628-4888
r 1
212 O
bwl
-l1 closer values at lower levels, best estimate.
I think the 2
process of doing the IPEs will produce. lower values.
Not only 3-the evaluations themselves, but the process.
4-MR. WARD:. And this is --
5 MR. HULMAN:
I'm a bit troubled by high values, as 6
high as 2 times 10 to the minus 3, for example.
7 MR. WARD:
And this is independent of the nature of:
8 the IPE, I mean, whether it's a full blown PRA or abbreviated 9
PRA?
10 MR. HULMAN:
My judgment is it will produce lower 11 values just by the process itself.
I think the utilities will 12 find that there are some vulnerabilities that will reduce the 13 core damage probabilities when they fix them.
And they will O
14 fix them.
Some of them are relatively easy to do.
15 (Slide) 16 The intent of this containment performance program, 17 starting with Mark I's, is to look at generic challenges to 18 containment.
Those challenges that we believe are generic to a 19 particular class of reactors, starting with Mark I's.
We're 20 looking at backfits, specific backfits.
Small group of what we 21 think are reasonable backfits, either singly or in combination, 22 to try and attack the containment challenges and 23 vulnerabilities to these generic kinds of challenges.
24 And we think that either a backfit will result and 25 result in rulemaking or we'll be able to conclude that not O
Heritage Reporting Corporation (202) 628-4888
I 2
213 l1 backfitting is necessary.
At this point, I can't say what the 2
conclusion is, but we would.not go with the generic letter 3
approach at this point that was proposed a couple of years ago.
4 (Slide) 5 The tasks in the program were to issue a. paper on 6
what we thought these generic challenges were, what-the
~
7 potential failure modes were.
That's been done.
That was 8
done, if I remember, early in January.
9 We were to do some parametric calculations.
Those 10
-have been done.
11 We held a workshop at the end of February in 12 Baltimore.
We had 150 people there, and in a moment, I'll try-13 and summarize what we learned from the workshop.
O 14 We're doing technical assessments of each one of the 15 proposed improvements.
16 We will do cost benefit analyses.
There may be other I
17 criteria besides cost benefit.
18 We will prepare an interim report.
That's the status 19 report I mentioned earlier.
That's in the concurrence chain.
20 And we hope to issue the final report and 1
21 recommendations to the Commission at the end of August.
j 22 (Slide) i.
~
What are the generic challenges we're looking at?
23 24 This is an attempt to summarize them as we understand them at 25 the moment:
combustion, early overpressure, overtemperature, O
Heritage Reporting Corporation (202) 628-4888 1
v 214
()
il early overtemperature, the so-called liner or shell melt 2'
through,' late overpressure, overtemperature, core concrete
~
3 interaction, some bypass sequences that we may not be able to 4
consider in this particular generic study, failure to isolate 5
is of the same type, and rapid overpressure and missiles which 6
we think has a low probability and need not be considered.
7 (Slide) 8 Now, this is a summary that we've prepared of the 9
workshop in February.
There's a mistake on it.
I want to 10 clarify that for you.
But let me go through it, and I'll 11 clarify it when we get to it.
12 It was a three-day meeting with 150 representatives 13 from industry, the research community, the Staff and there were 14 interested members of the public there, as well.
15 The industry emphasized prevention, and I think 16 that's probably the right emphasis.
They also said that any 17 fixes should be plant specific and should flow from the IPE.
18 I'm not sure I agree with that.
19 There are a variety of views on the probability of 20 liner or shell melt through that are important.
For example, 21 the manner in which the vessel may fail and release debris is 22 important.
If there is a relatively small pour over a long 23 period of time, that's one degree of challenge for the shell. 1 24 If it fails suddenly and massively, it's another challenge.
25 The industry believes water can prevent the liner O
Heritage Reporting Corporation (202) 628-4888
215 1
1 melt through.
That's not'a view that's held by the entire 2
research community, however.
3 DR. SHEWMON:
Now, this is water so that the core 4
runs into water and then the question is whether it would under 5
water.go through the liner?
~
6 MR. HULMAN:
Whether you would dissipate enough heat 7
and prevent the debris from piling up on the liner and causing 8
a melt through, right.
9 DR. SHEWMON:
I'm trying to get an idea of --
10 MR. HULMAN:
Under water.
Under water.
Right.
11 There's no consensus from the NRC researchers on 12 whether the liner fails and when that I can perceive at this 13 point.
That may change in the next month or so.
At this
()
14 point, there's no consensus.
15 The next bullet I think is a little misleading and 16 that's the one I want to correct.
It's a mistake.
There is no 17 consensus that water will prevent shell melt through.
There is I
18 a consensus that water will attenuate fission products.
The 19 case that we're looking at is a station blackout where you 20 can't put water into the containmant spray system at present.
21 So if you were to fix the containm nt spray system, you would 22 have water in the floor of the drywell.
23 There was general agreement that ADS reliability is 24 important, even in a station blackout.
And improvements may be l
25 achievable at modest cost.
In fact, one utility has already l
O Heritage Reporting Corporation (202) 628-4888 1
e 216
()
!1 implemented a fix-for station blackout.- That's Susquehanna.
!2 My understanding is that they have added another portable 3
diesel generator and a battery charger-to extend the life of
~
4' the batteries.
5 Our study so far of some of these improvements that 6
we've been considering, considered, indicate both positive 7
benefits for accident prevention and mitigation.
This is 8
something that Bob Bernero did not concentrate on.
He only 9
looked at the mitigation benefits.
We want to look at both.
10 The last conclusion to come out of the meeting was T
11 that New York State Power Authority asserted that their version 12 of the Code package, the source term code package and the 13 manner in which they used it yielded significantly lower source G
5'l 14 terms.
We are in the process of evaluating what changes 15 they've made to that code package, and how they've used it.
16 We're trying to see whether they've got a point or not.
17 (Slide) l 18 Now, in looking at some of the proposed improvements, 19 we have come up with a preliminary set of positive and negative 20 benefits, and whether we get core melt prevention and whether 21 we get mitigation.
22 And this is the list.
Venting.
The hardened pipe 23 from the outboard isolation valve of the containment to the l
l 24 stack is one of the improvements.
And based on what Boston l
25 Edison has done and what's been done in Europe, we think that
( )'
Heritage Reporting Corporation (202) 628-4888 i
(
l
217
('
il some kind of protection against inadvertent operation isa 2
probably in' order.
3 There are positive benefits and negative benefits.
I 4
think yop can read the table as well as I can.
5 These are the improvements that we're evaluating both
~
6 from a technical standpoint and a cross' benefit standpoint.
7 DR. REMICK:
Could you tell me how the venting might 8
induce core damage, the negative comment you have up there?
9 MR. HULMAN:
There is a class of sequences where you 10 can, if you vent, eliminate the net positive suction head 11 available to the pumps in the suppression pool.
You can cause 12 cavitation and pump failure.
So there are situations in.which 13 venting may help you, and there are situations in which venting O-14 may not help you.
15 DR. REMICK:
I understand what you're saying, but you 16 would never vent so you'd be subatmospheric and aren't those pumpscapableofoperatingwithatmosphericpressurei$sidethe 17 18 containment?
I guess I still don't understand.
19 MR. HULMAN:
You're taking suction from the 20 suppression pool.
21 DR. REMICK:
Right.
2 MR. HULMAN:
And trying to pump water from the j
23 suppression pool into the vessel.
l 34 DR. REMICK:
Right.
25 MR. HULMAN:
And then --
O Heritage Reporting Corporation (202) 628-4888 i
i a
-.-,s+---
V-218
()
- 1 DR. MOELLER:
The failure to control, does that mean 12 you vented when you didn't want to?
13 MR. HULMAN:
Where -- I'm sorry?
4 DR. MOELLER:
Negative, the second-bullet under the
-5_
top-negative column.
6 MR. HULMAN:
No.
That means that you, the failure to 7-control means that.you may have opened the vent unnecessarily, 8
-not inadvertently, just unnecessarily.
9 DR. MOELLER:
Oh, unnecessarily, okay.
10 MR. HULMAN:
Okay?
11 (Slide)
-12 The workshop we thought was useful in narrowing some 13 issues.
I'm back to the previous one.
Let me go to the other 14 one now.
15 (Slide)
- 1. 6 We see a balanced approach to the resolution of 17 containment issues, these generic contniitect issues that 18 involves the possibility of improved hardware features, the 19 accelerated implementation of some existing requirements, like 20 the ATWS rule.
21 Hydrogen control, we know there are still issues out 4
.22 there on hydrogen control related to the rule and the 23 possibility of the station blackout rule.
24 And we see the need for improved emergency operating 25 procedures and training to cope with the containment O
Heritage Reporting Corporation (202) 628-4888 i
I L
t t
219 l
challenges.
I'm not spedking about core damage, just the i
4 12 containment challenges.
And there's a need to monitor and 3
potentially improve plant operations through the SALP process, 4
through improved tech specs, through diagnostic reviews, 5
basically the regulatory oversight that's done by NRR.
6 MR. WARD:
What do you know about the status -- I 7
mean, you've said you think there needs to be improved 8
procedures, improvements in procedures'and training, so you 9
must know something about the status of the procedures that are 10 out there now.
What's your judgment about the quality and 11 usefulness of the procedures?
12 MR. HULMAN:
I think the procedures are pretty good j
13 as far as they go, but there are many possibilities that could u
(k 14 be encountered in severe accidents that I'm not enie are 1
15 covered adequately by the existing owners' group guidelines for 16 emergency operating procedures, and I suspect for the actual i
17 emergency operating procedures at the plants.
18 MR. WARD:
What do you mean by severe accident in 19 this sense?
I mean, I'd be surprised if there's anything out 20 there explicitly to deal with that.
21 MR. HULMAN:
There are some guidelines. They're not 22 called severe accidents.
They're symptom-oriented procedures, 23 for example, or guidelines for procedures.
I can't speak to 24 the procedures themselves, but to the guidelines.
But I don't 25 think they are necessarily adequate, nor do I think they O
Heritage Reporting Corporation (202) 628-4888 i
[
-220 t')
~ k/
1 emphasize trying to prevent the accident.
If you can't-ptuvent T2 it, keeping the fuel in the vessel, and as a last resort,
'3 managing that containment.
4 CHAIRMAN KERRt Have you thought about how many
'5 people need to be trained for this' sort of thing?
Do you 6
foresee that all licensed operators need this training or only 7
a special crew that would be available on call?
8 MR. HULMAN:
I haven't thought about that at all.
9 By the way, this is not just a research program.
10 This is a research and NRR program, and we haven't gotten to 11 that point, yet.
It's a fair question, just haven't gotten 12 that far.
13 MR. WARD:
For the Mark I's, what's the status of
(:)
14 containment venting as far as existing EOPs?-
15 MR. HULMAN:
Is there anybody here from NRR?
NRR has 16 conducted a survey of emergency operating procedures at some 17 Mark I's.
18 Are you familiar with it?'
19 MR. WARD:
No, I'm not familiar with it.
1 20 MR. HULMAN:
They have done a survey.
They're not 21 here to address it today, and 1 would be reluctant.
I do know 22 that it's spotty, but beyond that I can't speak to it.
i 23 CHAIRMAN KERR:
But there are some BWRs that use 24 venting as part of their emergency operating procedure, are l
25 there not?
l O
Heritage Reporting Corporation (202) 628-4888 l
221
,pV
-1 MR. HULMAN:
That's correct.
12
. CHAIRMAN KERR:
Is that what you meant or did you
- 3 mean --
4 MR. WARD:- That's.what I'm trying to find out.
It 5
had~ conflicting reports on it.
Did the EPGs provide'for 6
venting?
7 MR. HULMAN:
Yes, both of the dry well and the, wet-8 wall.
9 MR. WARD:
The EPGs do in both cases.
10 HR. HULMAN:
Now, whether they have been implemented 11 specifically that way at a plant is not clear to me.
But I do 12 know that NRR is conducting a survey on it at each one of the 13 Mark I's, I believe.
Now, whether it would say in each of O
14 their emergency procedures, open this two inch vent in the wet 15 valve as a precaution, and you don't buy much with a two-inch 16 line.
17 DR. REMICK:
Are the human factors people you have 18 reviewing that going to have --
19 MR. HULMAN:
Systems people as well.
20 DR. REMICK:
Understanding of the process or are they l
21 going to be --
l 22 MR. HULMAN:
My understanding is it will not only be 23 human factors people, it will be systems people.
And I think 1
24 when we come back in August, NRR will be prepared to speak to 25 that.
Heritage Reporting Corporation l
(202) 628-4888
222 1
DR. REMICK:
In the back of my mind, it seems to me I
~2 heard once, too, that there is venting of containment pressure 3
in the absence of any radioactive materials in the containment, 4
and then the other question is venting high pressure when there 5
are radioactive materials.
And so I think one has to be 6
careful when you say that they have procedures for venting, if 7
they cover both cases, or it's just the case when there's no 8
release of radioactive materials into the environment.
9 MR. HULMAN:
That's true.
The EPGs, the present 10 version of the EPGs speak to venting regardless of the amount 11 of radioactivity, and indicate that you could even use the 12 drywell as a last ditch effort to save the containment.
13 DR. REMICK:
I'm sorry?
Use the drywell?
O 14 MR. HULMAN:
Drywell vent.
15 DR. REMICK:
Oh, the drywell vent.
16 (Slide) 17 HR. HULMAN:
There is an interim Mark I report to the 18 Commission that was scheduled on April 30th.
It's not April 19 30th but the status report is in the concurrence chain and I 20 expect it out within the next week or ten days.
21 The final Mark I report with recommendations is due 22 to the Commission the end of August of this year.
We still 23 intend to meet that.
There is the possibility of some carry 24 over issues relating to shell melt through.
The consensus 25 building process is a very difficult one.
We may have to O
Heritage Reporting Corporation (202) 628-4888
l 4
223 I
continue and do some more research and evaluation beyond August
- 2 of this year, but I'm hopeful that the whole process relating
- 3 to Mark I's can be completed by a year from August, in the form 4
.of either the issue is not important or let's-see whether we 5
can afford to prevent it, whether there's cost benefit 6
justification or other justifications.
7 And then lastly, it's not only Mark I's that we're 8
looking at, we're looking at the generic challenges to each 9
containment type.
We'll be back here to talk to you about ice 10 condensers, Mark II's, III's and V's.
11 MR. WARD:
Was the first date missed or what?
12 MR. HULMAN:
I missed it by a week or two.
13 MR. WARD:
You are going to give us a report, is that 14 it?
15 MR. HULMAN:
A sta'tus report.
16 MR. WARD:
Second one, the recommendations, the final 17 report with recommendations, might those be recommendations for 10 research or resolving the issues, or are those going to be --
19 MR. HULMAN:
They'll be of three types in my 20 judgment.
One, nothing needs to be done for certain kinds of 21 challenges, or what could be done is not cost beneficial or 22 necessary.
Second, we want to back this and it's justified, 23 that's a recommendation.
And third, we may have to do some 34 more research and study.
25 So it's either not a problem or not justified, it is O
Heritage Reporting Corporation (202) 628-4888
224 O-justified, or we need to do more work.
1 42-MR. WARD:
And this is on some elements of the:whole 13 issue, or --
i-4 MR. HULMAN:- It's on the improvements.
Are the 5'
improvements justified.
6 MR. WARD:
That's the whole set of improvements?
7 MR. HULMAN:
Individually or in combination.
8 MR. WARD:
So in August of this year, you might come 9
out with a short list of things that we ought to do now, and 10 another list of things that are --
11 MR. HULMAN:
Right.
Yes, and a third list of things 12 that are not worth doing.
And I'll give you an example.
Let 13 me go back to the list --
0 14 MR. WARD:
You have some preliminary conclusions that 15 I think just blew by us.
16 MR. HULMAN:
That was the conclusions from a 17 workshop.
18 No, I think you're right.
This is ths one.
Let me 19 go over that one.
20 Closure appears likely for hydrogen control.
For 21 example, there are two sub-issues under hydrogen control that 22 we looked at.
One is the adequacy of tech specs that now allow 23 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> of operation at start up or shut down without 74 inerting.
The question is, is that too risky?
Should we 25 shorten the allowable time.
The judgment is at this point, O
Heritage Reporting Corporation (202) 628-4888
a 225
/~s
' \\-)
.1
. preliminary judgment is it's not too risky.
No change is
- 2 necessary.
I 3 There's another =ub-issue the$ was brought about by a 4
review of Boston Edison's proposal on Pilgrim, and that is 5
whether there are sneak sources of oxygen that after an 6
accident could result in hydrogen ignition or detonation.
And 7
they found some, and they proposed fixing them as part of their j
8 prc ram on containment safety improvements.
9 However, they're was a generic letter, 8409, that was 10 issued in 1984 that said you ought to fix this, utilities, and 11 this is the guideline for doing it.
So they're apparently IP complying with the generic letter but they haven't done it, 13 yet.
And there may be other plants out there the same way.
I O
14 see that as compliance, not as a new requirement.
15 MR. WARD:
You mean the 1984 generic letter dealt 16 with the sneak sources of oxygen that they just discovered?
17 MR. HULMAN:
No, I didn't say, just discovered.
We 18 knew that there were sneak sources of oxygen.
We told the 19 utilities in 1984, we think you ought to fir. them.
We presumed i
20 that they had been fixed.
Boston Edison came in last summer 21 and said, we're just fixing them.
22 DR. MOELLER:
Do the resident inspectors play a role 23 in monitoring the degree of responsiveness cf a utility or 24 licensee?
?
15 MP. HULMAN:
In what area?
(
Heritage Reporting Corporation (202) 628-4888
226 O( /
1 DR. MOELLER:
The one you were just covering.
This
.2 was issued in '84 and this is '88?
- 3 MR. HULMAN
My understanding is that NRR through the 4
SIM system and contacts between the project managers and the 5
utilities and the resident inspectors have been attempting to 6
verify the implementation of all the requirements that we've 7
laid on industry: su they are involved.
8 Any other questions?
9 MR. WARD:
I'm interested in what you mean by the 10 second item there, the primary interest.
11 MR. HULMAN:
This one?
12 MR. WARD:
Yes.
13 MR. HULMAN:
Well, I think that's a little subtle.
O 14 Let me see if I can clarify it.
Industry's interested in 15 prevention; so are we.
But we're also interested in trying to 16 strike some kind of balance between prevention and mitigation.
17 These proposed improvements have attributes of both in some 18 cases.
19 So we see this as an attempt te strike that balance 20 for Mark I's with potential improvement.
21 MR. WARD:
Do you have some sort of a goal there in 22 mind?
I r. san one I guese the Committee has suggested is for 23 implementation of the safety goal thet some sort of balance 24 between prevention and mitigation can be struck for 25 implementation of the safety goal.
We even suggested some O
Heritage Reporting Corporation (202) 628-4888
9 227 (kl 1
quantitative expression, in that area?
2 MR. HULMAN:
Yes, I would hope we could drive the 3
probability of containment failure to less than 10 to the minus 4
6.
5 MR, WARD:
Containment failure?
6 MR. HULMAN:
This is core melt accident and 7
containment failure, probability of a large release less than 8
10 to the minus 6.
9 MR. WARD:
Okay.
Now, that's not an expression of 10 defense in depth, yet?
11 MR. HULMAt:
Might be.
12 MR. WARD:
No, no it isn't.
I mean, if a licensee 13 comes in with a claim that his probability of core melt is 10 7-)
V 14 to the minus 7, that would satisfy your requirement of 10 to 15 the minus 6.
16 MR. HOLMAN:
I didn't say it was a requirement.
17 MR. WARD:
Uell, your goal, whatever?
18 MR. HULMAN:
It's a preliminary one, yes.
It's not a 19 requirement.
20 MR. WARD:
I mean, the point is, are you willing to 21 let a licensee put all his eggs in the prevention basket?
22 MR. HULMAN:
I haven't faced that issue, yet.
I'm 23 not sure.
I'm not sure we should.
24 MR. WARD:
Well, I hope you don't.
I think we're 25 getting some coe: plaint from some quarters O
Heritage Reporting Corporation (202) 628-4888
228
(
1 CHAIRMAN KERR:
Are you leading the-witness?
2 DR. SHEWMON:
As a mildly relevant and slightly
<3 irrelevant question, there was this brouhaha a while back when 4
Pilgrim came in and said, gee, we want to improve our Mark I, 5
and the NRC said, no, you don't want to improve it, and we 6
won't let you.
Where is that now?
7 MR. HULMAN:
That's an NRR issue.
And let me see if 8
I can shed a little light on it.
9 First of all, NRR did not tell the utility, we won't 10 let jou.
They asked a number of questions because of concern 11 over the potential negative aspects of venting.
And some of 12 the other potential improvements that impacted safety-related 13 components and systems.
O 14 My understanding is that they have agreed to most of JE what Boston Edison has proposed.
But are reserving judnment on 16 the use of the vent, which is already in.
Now, I can't speak l
17 to what NRR's views are on the issue right now.
But when we 18 come back in August, I'm sure that NRR will be prepared to do 19 that.
20 DR. REMICK:
Are there any ADS reliability issues 21 other than tbo long term DC supply?
22 MR. HULE 1:
My understanding is, no.
23 All right.
Hearing no other comments --
24 MR. WARD:
Yes.
Are there any other questions?
4 25 (No response)
(:)
Heritage Reporting Corporation (202) 628-4888
i v
229
()
~MR. WARD:
I' guess not.
So thank you very.much,'Mr.
1 2
Ilulman.
3 Mr. Chairman, ur.less other members think-there's some 4
reason the Committee should discuss this further at this time 5
and perhaps develop some comment, I don't propose anything, so 6
I'll turn the meeting back to you.
7 CilAIRMAN KERR Are there comments from the rest of 8
the Committee?
9 (No response) 10 CHAIRMAN KERR:
I see no~ ground swell' support for 11 such a position.
T2 Thank you very much,'Mr. Hulman.
13 This brings us to time for a ten minute break.
O 14 (Whoreupon, a+. 4:35 p.m.,
the committee was recessed 15 for the da'>,
to reconvene the following day, Friday, May 6, 16 1988, at 8:30 a.m.,
in the same place.)
17 18 19 20 21 22 23 24 25 O
Heritage Reporting Corporation (202) 628-4888
e 1
CERTIFICATE CE) 2 3
This is to certify that the' attached proceedings before the 4
United States Nuclear Regulatory Commission in the r.atter of:
5 Name:
337th ACRS MEETING 6
7 Docket Number 8
Place Washington, D.C.
9 Date:
May 5, 1988 10 were held as herein appears, and that this is the original 11 transcript thereof for the file of the United States Nuclear 12 Regulatory Commission taken stenographically by me and, 1,?
thereafter reduced to typewriting by me or under the direction 14 of the court reporting company, and that the transcript is a
()
true and accurate recor'dfaf the fore ' ing proceedings.
15
.t,). l' b, r <![
16
/S/
17 (Signature typed):
Donna Cook 18 Official Reporter i
19 Heritage Reporting Corporation 20 21 22 23 24 25
(
Heritage Reporting Corporation (202) 628-4888
O f,n
- {y:l-
' ()
SEVERE ACCIDEllT IflTEGRATED PLAN BY THEf11S P. SPEIS, DEPUTY DIRECTOR OFFICE OF flVCLEAR REGULATORY RESEARCH O
FOR ACRS f1EETING MAY 5, 1988 O
O BACKGROUhD o
AUGUST 8, 1985,'"SEVERE ACCIDENT POLICY STATEMENT,"
FEBRUARY 28, 1986, "IMPLEMENTATION PLAN FOR SEVERE ACCIDENT POLICY STATEMENT," SECY-86-76 o
DECEMBER 1, 1986, STAFF REQUIREMENTS MEM0, CHAIRMAN REQUESTED PAPER ON INTEGRATION OF SEVERE ACCIDENT ISSUES l
0 FEBRUARY 17, 1987, MEMD FROM EDO TO COMMISSIONERS ON Q
PRELIMINARY PLAN FOR INTEGRATION OF SEVERE ACCIDENT ISSilES o
JULY 15, 1987, STAFF BRIEFED COMMISSION ON A PLAN FOR CLOSURE 0F SEVERE ACCIDENT ISSllES o
DECEMBER 8, 1987, "MARK I CONTAINMENT PERFORMANCE PROGRAM PLAN,"
o FEBRUARY 9-11, 1988, BALTIM0RE MANAGEMENT MEETING ON SEVERE ACCIDENT ISSUES O
l l
i m-y.
.-, - - -~.., - - -..,-__.. _ _.. __,-_,
m._,-_.
'O SEVERE ACCIDENT INTEGRATED ELMi
~1 o
PURPOSE:
'TO PRESENT STAFF PLANS FOR INTEGRATION AllD CLOSURE OF SEVERE ACCIDEllT ISSUES 0-OBJECTIVES:
TO PROVIDE AN UNDERSTANDING 0F THE STAFF ACTIVITIES THAT ARE UNDER WAY T0 IMPLEMENT THE COMMISSION'S SEVERE ACCIDENT POLICY TO ASSURE THAT THESE ACTIVITIES ARE CONSISTENT l
WITH THE COMMISSION'S POLICY AND STRATEGIC G0ALS O
TO ASSURE THAT THE STAFF ACTIVITIES ARE CONSISTENT AMONG THEMSELVES, HAVE A COMMON G0AL OF ULTIMATELY LEADING T0 IMPROVED PLANT SAFETY, AND ARE PROPERLY C0ORDINATED AMONG THE RESPONSIBLE i
NRC ORGANIZATIONS l-TO ASSURE THAT THE COMMISSION IS AWARE OF THE KEY TECHNICAL AND POLICY ISSUES, SOME OF WHICH i
NEED COMMISSION GUIDANCE OR APPROVAL l
l l
l O
o
U,,
'O SEVERE ACCIDENT ACTIVITIES o
INDIVIDUAL PLANT EXAMINATIONS (IPE) o CONTAINMENT PERFORiiANCE INPROYEMENTS (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 GENERIC SAFETY ISSUES g
o EXTERNAL EVENTS o
INTEGRATED SAFETY ASSESSMENT PROGRAM (ISAP) o SEVERE ACCIDENT POLICY FOR FUTURE PLANTS o
SEVERE ACCIDENT CLOSURE /USE OF SAFETY G0AL i
i I
i O
l t
Y
O O
O m
ISIS g
i G51s 1
i Stfety reais 12 RIG-1I50 BacLfit Iblicy l
Individual Plant i
Examinations 3L Contairunent Identify levcim
&cident
. cre Perform 3nce Plant Improvements fbnagement Accident Strategy. Procedu Proc, ram Closure Iciprovenents Vulnerabilitics d
l (li.irdwa re J
i l
26
.{
1 t
SMP i
5 t., Reviews IST
)-
e t i..
1 Isrproved Plant
,r 1
Operations
,I Cemtisening j
Severe Sevesc Accident
' -Accident P.csca rtli Itc<ca rt h I
FIGURE 1.
SEVEPE ACCIDErlT IriTEGRATED PLAfi i
)
Rolo of NUREG-1150
~
O O
Objective:
i l
Provide up-to-date assessment of severe accident frequencies and risks for five plants, reflecting data obtained since Reactor Safety Study i
I t
j Roles in Regulatory Process:
Independent staff assessment of risks l
Technical data base as input to:
IPE i
accident management containment performance initiatives safety goal implementation j
generic issue resolution Prioritization/ focus of research
SEVERE ACCIDENT RESEARCH BEGINNING IN.1980, AFTER THE TMI-2 EVENT, RESEARCH HAS PROVIDED A DATA BASE AND MODELS FOR:
l 0
FISSION PRODUCT RELEASE, TRANSPORT, DEPOSITION, & REVAPORIZATION i
o CONTAINMENT LOADING BY HIGH PRESSURE MELT EJECTION (HPE) o HYDR 0 GEN DETONATION AND BURNING o
CORE / CONCRETE INTERACTIONS (CCI) o CONTAINMENT PERFORMANCE TESTING o
EFFECTS OF NATURAL CIRCULATION ON THE PRIMARY SYSTEM l
o CORE NELT PROGRESSION (EARLY STAGES)
FUTURE RESEARCH EFFORTS WILL FOCUS ON SPECIFIC ISSUES SUCH AS:
O o
CONTAINMENT FAILURE PROBABILITY BY DIRECT CONTAINMENT HEATING (DCH) INCLUDING EFFECT OF NATURAL CIRCULATION o
MELT SPREADING AND-CONTAINMENT SHELL FAILURE IN MARK l'S o
RESEARCH DATA IND MODELS TO ASSESS ACCIDENT MANAGEMENT STRATEGIES 1
o LONGER TERM CONFIRMATORY RESEARCH ON:
4 DCH CONSEQUENCES REFINEMENT OF HYDROGEN BEHAVIOR MODELS j
i CORE MELT PROGRESSION (LATE STAGES)
L0i;d/ CONCRETE INTERACTIONS FURTHER MODEL ASSESSMENT AND REFINEMENTS O
...l 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 o
DCH PROBABILITY OF HIGH PRESSURE MELT EJECTION (NATURAL CIRCULATION)
CUT 0FF PRESSURE FOR HPE MANAGEMENT THROUGH DEPRESSURIZATION CONSEQUENCES ANALYSES OF PERFORMANCE TESTS ON CONCRETE CONTAINMENTS (1/6 SCALE)
T Q 4~.)
a cy)
C)
ACRS MEE"3G
+ loa J
ON THE IMPLEMENTATION PLAN FOR THE SEVERE ACCIDENT POLICY STATEMENT INDIVIDUAL PLANT EXAMINATION (IPE)
GENERIC LETTER O
THEMIS P. SPEIS DEPUTY D: RECTOR FOR GENERIC ISSUES OFFICE OF NUCLEAR REGUIATORY RESEARCB BRIAN W. SHERON, DIRECTOR DIVISION OF REACTOR AND PLANT SYSTEMS OFFICE OF NUCLEAR REGULATORY RESEARCH l
r O
MAY 5,1988 1
O
- 1.
SUMMARY
o A 10 CFR 50.54(f) LETTER REQUESTING THE INDIVIDUAL PLANT EXAMINATIONS FROM AIL UCENSEES HOLDING OPERA UCENSES AND CONSTRUCTION PERMITS FOR NUCLEAR POWER BASIS FOR REQUESTING INDIVIDUAL PLANT EXAMINATIONS IS o
THE COMMISSION SEVERE ACCIDENT POUCY ISSUED ON AUGUST 8,1985 (50 FR 32138)
- PLANT SPECIFIC PRAs EXPOSED RELATIVELY UNIQUE VULNERABILITIES TO SEVERE ACCIDENTS
- TILE UNDESIRABLE RISK FROM TIIESE UNIQUE FEATURES HAS BEEN REDUCED BY LOW-COST CIIANGES VIA PROCEDURES OR MINOR DESIGN MODIFICATION O
- ANALYSIS WILL DE MADE OF ANY PLANT TilAT HAS NOT YET UNDERGONE AN APPROPRIATE EXAMINATION WHEN NRC AND INDUSTRY SUFFICIENTLY PROGRESSED TO DEFINE THE METHO OF ANALYSIS 31NCE ISSUANCE OF TIIE POUCY STATEMENT THE STAFF INT o
EXTENSIVELY WITH INDUSTRY AND DEVELOPED APPROPRIATE DOCUMENTATION FOR THE IPEs PURPOSE OF IPE IS FOR THE UTIUTIES T0:
o IDENTIFY / UNDERSTAND THE MOST UKELY SEVERE ACCIDENT SEQUENCES THAT COULD OCCUR AT THEIR PLANTS:
EVALUATE /lMPLEMENT MEANS FOR IMPROVEMENTS O
EVELOP AN AWARENESS FOR SEVERE ACCIDENT BEHAVIOR 2
(D
- 2. EXAMINATION PROCESS t
o LICENSEE'S STAFF SHOULD PARTICIPATE IN AIL 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 PROCEDURAL) TO SEVERE ACCIDENTS (FOR BOTH CORE DAMAGE AND CONTAINMENT PERFORMANCE); BOTH INTERNAL AND EXTERNAL INITIATORS ARE TO BE CONSIDERED, HOWEVER SEPERATE FROM THE IPEs 0F INTERNAL EVENTS
- QUANTIFY RESULTS OF EXAMINATION FOR THE SEQUENCES THAT CONTRIBUTE THE MOST TO TIIE TOTAL CORE DAMAGE OR LARGE RELEASE FREQUENCY
- UNDERSTAND WHAT COULD PROBABLY G0 WRONG IN A PIANT.
- IDENTIFY AND EVALUATE MEANS FOR IMPROVING PLANT /
CONTAINMENT PERFORMANCE (VIA HARDWARE ADDITIONS /
MODIFICATIONS, ADDITIONS TO PROCEDURES, TRAINING);
O
- DECIDE WHICH IMPROVEMENTS WILL DE IMPLEMENTED AND SCHEDULE FOR IMPLEMENTATION 3
C
- 3. M3TH0]S OF AXALYSIS o THE GENERIC Irnu SPECIFIES SEVERAL OPTIONS THAT COULD BE USED TO SATISFY THE mMINATION REQUIREMENTS IDCOR IPEMs FRONT END WITH STAFF ENHANCEMENTS +
CONTAINMENT PERFG3MANCE ANALYSIS CONSISTENT WITH APPENDIX 1 IEVEL-1 PRA + CONTAINMENT PERFORMANCE ANALYSIS O
CONSISTENT WITH APPENDIX I ( OR LEVEL-II OR -III PRA YITH "BACK-END" CONSISTENT WITH APPENDIX 1) 0THER SYSTEMATIC EVALUATION METHODS (STAFF PREV t
MIGHT BE NECESSARY) l l
l 4
O IDCOR IPEMs EVALUATION THE IDCOR IPEMs TAKE ADVANTAGE OF AVAILABLE IDCOR o
PLANT ANALYSES, SIMILARITIES AMONG PLANTS AND UP-TO-DATE KNOWLEDGE OF SEVERE ACCIDENT PHENOMENA. THE IPEM ANALYSIS IS LESS RIGOROUS THAN A FULL SCOPE PRA.
THE IDCOR IPEMs WERE SUBMITTED TO THE NRC MAY 1986 o
INITIALLY MODIFIED BY IDCOR IN DECEMBER 1986 TO o
ACCOMODATE NRC COMMENTS. FURTHER MODIFICATIONS WERE RECEIVED IN APRIL 1987 NRC EVALUATION OF THE IDCOR IPEMs IS COMPLETE o
IT IS EXPECTED TIIAT THE LEVEL OF EFFORT REQUIRED IF A o
UTILITY USES TIIE IDCOR IPEM IS EQUIVALENT TO LEVEL-I PRA BUT IS LESS TilAN IF A FULL-SCOPE LEVEL-III PRA WERE USED l
l l
5
0 IDCOR IPEMs EVALUATION (CONT.)
[ FRONT-END)'
o KEY AREAS OF NRC CONCERNS:
- EMPHASIS ON BOTTOM-UNE NUMBERS, NAMELY CORE DAMAGE FREQUENCY.
THE STAFF EMPHASIZES IDENTIFICATION OF PLANT SPECIFIC VULNERABIUTIES, USTING OF POTENTIAL AREAS FOR IMPROVEMENTS AND SELECTION OF MOST PROMISING IMPROVEMENTS FOR IMPLEM
- ALM 0E ALL TEST APPUCATIONS HAD TIfE BENEFIT OF A FULL SCOPE PRA YBICH IN THE STAFF'S OPINION COULD HAVE INFLUENCED TII
- NO TEST APPLICATION OF THE REVISED IPEMs
- IDCOR BWR IPEM DOES NOT ASSURE A COMPLETE EVALUATION OF TH IMPACT OF SYSTEM DESIGN AND OPERATIONAL MODIFICATIONS ON T OVERALL PLANT RISK ESTIMATE EXTREME CARE NEEDED IN MODELING SUPPORT STATES. THE CRITERIA FOR DETERMINING YBEN MORE DETAILED ANALYSIS OF SUPPORT STAT IS NEEDED IS LEFT TO TIIE ANALYST STAFF EYALUATION PROPOSES SEVERAL ENHANCEMENTS TO ADDRESS o
O THE AREAS OF CONCERNS IN THE FRONT-END 6
IDCOR IPEMs EVALUATION (CONT.;'
n 1
V (BACK-END) o THE IDCOR BACK-END IPEMs ARE TOO NARROYLY FOCUSED AND TILL N PROVIDB UTILITIES YlTH INFORMATION NEEDED TO EXAMINE CONTA PERFORMANCE FOR THE FULL RANGE OF SEVERE ACCIDENTS o EMPBASIS ON B01 TOM LINE NUMBERS, NAMELY RELEASE FRACTION. TIE STAFF EMPHASIZES IDENTIFICATION OF CONTAINMENT FAILURE MECHANI TO ALLOW FOR IDENT&lCATION AND IMPLEMENTATION OF RECOVERY RELY VERY BEAVILY ON IDCOR VIEYS OF SEVERE ACCIDENT o
O PIEN MEN L Y AND TBE I C R MAAP CODE o SOME OF IDCOR VlEYS AND METHODOLOGY ARE IN DISAGREEMEhT YITH CURRENT RANGE OF ASSrWPTIONS NO RECOGNITION OF UNCERTAINITIES IN TIE PHENOMENOLOGY o
o OPTIMISTIC ASSESSMENT OF CORRECT OPERATOR ACTION: TO MAINTAIN CONTAINMENT INTEGRlTY DURING SEVERE ACCIDENTS LOOSE INTERPRETATION OF TIIE IPEM QUESTIONS BY THE UTILITIES o
COULD RESULT IN A LACK 0F SUFFICIENT ATTENTION TO MEASURES TO MlTIGATE TIE CONSEQUENCES OF AN ACCIDENT 7
m O
STAFF ENHANCEMENTS (FRONT-END) o LEVEL OF DETAIL
- SUPPORT SYSTEMS FAILURES USED IN TBE SUPPORT SYETTEM TREE NEED TO BE EXPANDED
- BETTER TREATMENT OF COMMON CAUSE FAILURES (BOTH Q AND QUANTITATIVE)
- BETTER TREATMENT OF HUMAN FAILURES AND IIUMAN RECO
- SPECIFIC VULNERABILITIES SHOULD BE IDENTIFIED AND TREAT A SEGMENT OR COMPONENT LDTL
{
- SENSITIVITY AND IMPORTANCE ANALYSIS RECOMMENDED FOR MA CONTRIBUTING ASSUMPTIONS
- RECOMMEND THE USE OF FAULT-TREES INSTEAD 0F THE PROP IPEM APPENDIX D METHOD o SYMMETRIC CONCERN
- FOR NON-SYMMETRIC PIANTS IPEM SUGGESTED SUPPORT SYSTEM l
END STATES ARE NOT ADEQUATE
- FOR NON-SYMMETRIC PLWTS SUPPORT SYSTEM END-STATES MUST BE BASED ON INDEPENDENT SUPPORT SYSTEM SEGMENTS (TRUNS, COMPONENTS,... etc )
i 8
O GUIDANCE ON CONTAINMENT SYSTEM PERFORMANCE SCOPE /0BJECTIVES o GUIDANCE ON PERFORMING CONTAINMENT ASSESSMENT IS AVAllABLE IN NUREG/CR-2300 AS SUPPEMENTED WITH THE INSIGHTS OF APPENDIX 1 o EXISTING PRAs, (e.g., ZION, INDIAN POINT 3. NUREG 1150 PIAh7S) THAT IIAVE CONTAINMENT PERFORMANCE ASSESSEMENTS CONSISTENT WITH NUREG/CR-2300 AND THE GUIDANCE OF APPENDIX 1, YOULD BE CONSIDE ACCEPTABE o EXAMPE OF EXAMINATION FOR PWR IS PROVIDED o THE IPE SHOULD PROVIDE A REALL%C EVALUATION OF CONTAINMENT PERFORMANCE o AIM THE IPE TO DEVELOP EVENT TREES THAT COVER THE RANGE OF UNCERTAIN 17Y o HENCE STAFF GUIDANCE FOCUSES ON HAVING UUURES:
- EVALUATE CONTAINMENT FAILURE MECHANISMS AND TIMING
- USE RELEASES FROM TABLES PREPARED ON THE BASIS OF REFERENCE PIAh7 IhTEGRATE SYSTEMS RESPONSE (PROBABIllS71CALLY) INTO SIMPLIFIED, DUT REALISTIC, CONTAINMENT EVENT TREES (ALLOW FOR RECOVERY OR OTHER ACCIDENT MANAGEMENT PROCEDURES)
- EXAMINE ALL CIASSES OF SEQUENCES TITil SIGNIFICAhT PROBABIl]TY
- RECOGNIZE AND AS APPROPRIATE, ACCOUNT FOR UNCERTAINITIES o FOR THOSE PilEN0MENA HIAT ARE NOT NELL UNDERSTOOD, DIFFICULT TO QUANTL"Y, AND THERE IS CONSIDERABE UNCERTAlhTY IN HIEIR RESULTS, NO SPECIFIC CONTAINMENT PERFORMANCE ANALYSIS IS REQUIRED. HOWEYF9, CONSIDERADON O
v SIl0ULD DE GIVEN TO STRATEGIES TO DEAL WITH THOSE ISSUES WHIE AWAIUNG TIIE GENERIC RESOLUTION.
9
i EXAMINATION GUIDANCE STEP 1 PLANT FAMIUARIZATION SYSTEM DESIGN: ESF OPERATING LEV 2, AVAIIABIUTY PLANT LAYOUT; CAVITY CONFIGURATION, CONNECTION TO SUMP, FLOW PA1 TERNS BETWEEN COMPAIUMENTS STEP 2 SEQUENCE GROUPING RECEIVE FROM FRONT-END ANALYST A SET OF TIIE IIADING ACCIDENT SEQUENCES GROUP SEQUENCES WITH SIMILAR EFFECTS ON CONTA!NMENT SYSTEMS " PLANT DAMAGE STATES"
- TIME OF CORE MELT (EARLY OR LATE)
- PERFORMANCE OF CONTAINMENT SAFETY SYSTEMS ALL CONTAINMENT SYSTEMS AVAILABLE NO SPRAYS OR COOLERS COOLERS ONLY SPRAYS ONLY IDENTIFY ANALYSIS REQUIRED FOR EACH (OR GROUP) LEADING SEQUE STEP 3 DETERMINATION OF CONTAINMENT FAILURE MODES LIST OF POTENTIAL CONTAINMENT FAILURES MECHANISMS (TABII 7.1)
BY ANALOGY WITII SIMilAR PLANT OR USING BUILDING CODE REQUIREMEh7 ( E.G., A FACTOR OF 2 TO 3 DEPENDING ON CONTAINMENT TYPE)
O 10
(
t EXAMINATION GUIDANCE (CONT.)
STEP 4 DEVELOP CONTAINMENT EVENT TREE BY ANALOGY WITH SIMIIAR PIANT DESIGN OR BY SELECTING EVENT TREE HEADING USING NODAL QUESTIONS IN TABLE 7-3 OR 7-4
- DIVIDE THE ACCIDENT INTO MAJOR TIME PERIOD EVElRS BEFORE CORE MELT EVENTS AFTER CORE MELT (IN-VESSEL & EX-VESSEL)
- DIVIDE TIE EVENTS RELATED TO BOTB IN-VESSEL AND EX-VESSEL PHENOMENA INTO:
HIGH PRESSURE SEQUENCES O~
LOY PRESSURE SEQUENCES STEP 5 DETEWINATION OF CONTAINMENT CIIALLENGES DETERMINE THE TIME AT WHICH THE CONTAINMENT FAILURE IS REACHED FOR THE FOLLOWING CHALLENGES (0.'.PROPRIATE COMBINATIONS).
USE AVAIIABLE INFORMATION WHERE APPROPRIATE, ;.g., PRAs NUREG-1150 OVERPRESSURJZATION STEAM AND NONCONDENSABLES COMBUSTION PROCESSES (H2)
DIRECT CONTAINMENT IIEATING CORE-CONCRETE INTERACTION BASEMAT PENETRATION STRUCTURAL FAILURE 1
11
1 >
EXAMINATION GUIDANCE [ CONT.)
STEP 6 DETERMINATION OF SOURCE TERM MAGNITUDE ESTIMATE THE FISSION PRODUCTS THAT CAN BE RELEASED TO TIIE ENVIRONS:
BY ANALOGY TO RELEASES FROM REFERENCE PLANT ANALYSIS OR INTEGRATED CODE ANALYSIS STEP 7 QUANTIFY EVENT TREE (T0 GET RELATIVE IMPORTANCE OF THE CONTAINMENT CHALLENGES) 7 USING DATA FROM STEPS 5 & 6 (PROBABIllTY OF VARIOUS CONTAINMENT FAILURE MODES AND SOURCE TERMS)
STEP S SENSITIVITY STUDIES IDENTIFY PARAMETERS THAT ARE RELY TO HAVE LARGE EFFECT ON THE LDGLDI0OD OR TIME OF CONTAINMENT FAILURES (e.g., FUEL COOLANT INTERACTION, HYDR 0 GEN GENERATION, DEBRIS C00 LABILITY) OR SOURCE TREMS VARYING THESE PARAMETERS OVER FEASIBLE RANGE STEP 9 IDENTIFY AREAS FOR 1MPROVEMENTS UTIIJZE RESULTS FROM AB0VE, IDENTIF WHAT CAN BE DONE T0:
-REDUCE CHALLENGES TO CONTAINMENT
-REDUCE THE SOURCE TERM O
12
[
O 4
ROLE OF SEVERE ACCIDENT Mt.RACEMENT SEVERE ACCIDENT MANAGEMENT IS A PROCESS IN WIUCH ACTIONS TH o
CAN PREVENT CORE DAMAGE OR MITNATE TIIE CONSEQUENCES OF ACCIDENT ARE IDENTIFIED, EVALUATED, INCORPORATED INTO A STRUCTURED PROGRAM, IMPLEMENTED AT A PLANT SITE AND ARE AVAILGLE TO THE OPERATORS AND FIANT MANAGEMENT IN THE EVENT OF SEVERE ACCIDEN SEVERE ACCIDENT MANAGEMENT ENCOMPASSES HARDWARE, HUMAN, AND o
ORGANIZATIONAL PACTORS o IT PROVIDES DECISION MAKERS AT THE PLANT A STRUCTURED PROGRA FOR MANAGING A SEVERE ACCIDENT O
PROPOSED GENERIC L. ?rER ADDRESSES SEVERE ACCIDENT MANAGEME o
AS FOLLOWS
- UTDITIES ARE EXPECTED TO DEVELOP AN ACCIDENT MANAGEMENT PR FOR PREVENTION OR M]TIGATION OF RISE IMPORTANT SEVERE ACCIDENT IDENTIFY MEASURES THAT PIANT PEPSONNEL CAN AND SHOULD TAKE IN CASE OF SEVERE ACC1 DENT. ASSESS AGAINST THE CRITERIA 0F 10 CR 50.59 AND, IF APPROPRIATE, SUBMIIT FOR NRC REVIEW IN ACCORDANCE WITH 10 CFR 53.00 UTIllTIES SHOULD PREPARE TO SUCCESSFULLY EXCUTE MEASURES TI FOUND T0 (1) PREVENT CORE DAMAGE, (2) PREVENT CORE DEBRIS FROM PENETRATING THE VESSEL OR, (3) MA!! AGE TIIE C0 SEQUENCES RTA57 AND NUMARC DISCUSSINC THE NEED TO DEVElhP AND IMPLEMENT o
ADDITIONAL SEVERE ACCIDENT MANAGEMENT PLtN E
~
_\\
d i
IPE RESULTS REVEY TERE ARE 109 LICENSED PLANTS, REPLICATE PLANT YOULD REDUCE o
THE IPE SUBMIITAIS TO 80 o 6 PERSON-MONTH PER PLANT REVIEW 0F TBE IPE SUBMI1TALS INCLUDING REVIEY OF ANY PROPOSED MODIFICATION o IPE SUBWI1TALS ARE ESTIMATED TO BE OVER A PERIOD OF 3 YEARS APPR0XIMATELY 16 PERSON-YEARS EFFORT PER YEAR,8 CONTRACTORS, o
8 STAFF ( 4 FROM NRR & 4 FROM RES ). RES WILL HAVE Tile LEAD 0
APPR0XIMATELY $1.4M PER YEAR ( 8 CONTRACTOR STAFF YEARS )
o IN CASE OF DISAGREEMENT, THE STAFF WILL PURSUE FIX IN o
ACCORDANCE YlTH THE BACEFIT RULE OR ORDER TEAM CONCEPT-EACll PLANT REVIEY TO BE TILE RESPONSIBIIJTY OF o
ONE NRC TEAM LEADER, SEVERAL PIANT SYSTEMS SPECIAUSTS, AND PRA SPECIALIST TWO TEAMS WILL D0 ONLY IARGE DRY, ONE TEAM VILL DO MARE I & 11 o
AND ONE TEAM YlLL D0 MARK 111 AND ICE CONDENSER O
14
f$U 8TAFF USE OF IPE RESULTS REV[EW 0F IPE RESULTS FOR THE FOLLOWING:
o
- TO ENSURE ADEQUATE ANALYSIS OF PIANT DESIGN AND OPERATIONS TO DISCOVER PARTICULAR VULNERABillTY TO CORE DAMAGE AND UNUSUALLY POOR CONTAINMENT PERFORMANCE
- FOR CONSISTENCY IN THE IDENTIFICATION AND TREATMENT OF LEADING CORE DAMAGE SEQUENCES
- TO ENSURE SEQUENCES AND SEQUENCE FREQUENCIES ARE APPROPRIATE
- TO INDEPENDENTLY CONCLUDE THAT DECISIONS ON THETHER TO MAKE IMPROVEMENTS ARE JUSTIFIED
- TO ALLOW COMMISSION TO CONCLUDE THAT THE SEVERE ACCIDENT POLICY HAS BEEN RESPONSIBLY IMPLEMENTED BY INDUSTRY O
CONSIDERATION Witt iNCtuDE 80Tu QUiNTITiTivE MEASURES AND NON o
QUANTITATIVE JUDGMENT TO DETERMINE IF A PLANT IS ACCEPTABII OR NOT o IF NRC COSIDERATION IDICATES THAT PLANT DESIGN OR OPERATION COULD BE ENHANCED BY ADDITIONAL PROTECTION BEYOND NRC REGULATION, IMPLEMENTATION WOULD BE IN ACCORDANCE WITH 10 CFR 50.109 IF CONSIDEPATION INDICATES PIANT DESIGN OR OPERATION SHOULD BE o
CHANGED TO MEET NRC REGULATION, IMPLEMENTATION WOULD BE WITHOUT REGARD TO COST EXECPT TO SELECT AMONG ALTERNATIVES IPE RESULTS WILL BB USED TO IDENTIFY SEVERE ACCIDENT VULNERABILITY o
GENERIC TO CLASS OR SEVERAL CIASSES OF PIANTS
- THIS INFORMATION ON GENERIC VULNERABRITIES WOULD DE USED TO EXAMINE 1F DEFICIENCIES IN TIIE REGULATION EXIST IF GENERIC DEF1CIENCIES NERE IDENTIFIED SAFETY GOAL WOULD BE USED TO DETERMINE IF REGUIATION MODIFICATIONS WERE NEEDED
..o s
( k h !!/ 4 O
- 5. RELATIONSHIP TO USI A-45 o A-45 ANALYSES HAVE SH0TN THAT DECAY HEAT REMOVAL FUNCTION FAILURE ARE SUFFICIENTLY P! ANT SPECIFIC AND WOULD REQUIRE SYSTEMATIC EXAMINATION o PROPOSED STAFF RESOLUTION OF A-45 IS TO SUBSUME ISSUE INTO IPEs o THE PROPOSED GENERIC LETTER STATES THAT THE IPE Sil0ULD IDENTIFY THE VULNERABLE ASPECTS OF DHR FUNCil0N
{
o THE PROPOSED GENERIC LETTER PROVIDES INSIGHTS GAINED FROM SIX LIMITED SCOPE PRA PERFORMED BY NRC UNDER THE A-45 PROGRAM l
l
~
1 15 73
t 3,k 4 / 4/Y f'
i *,
ylUt.c f (. Ilfk nqj
)
t r
+
l-FARK 1 CONTAlHNT INTEGRITY UNDER SEVERE ACCIDENT CONDITIONS i'
t i
4 MAY 5,1988 I
l l
t l
O SEVERE ACdlDENT IS9ES BRAtOI
- t. G sutras. cater NUCLEAR REGULATORY BESEARCH l
l e-
\\
}
l-
\\
i s
i l
t r
j.
1 c
t i
i I
l I
i O
i I
1 l
i
-J
l i
- 'O f
PARK i CONTAIMeli CONCERIG RELATIVELY SFALL VOLlFE CONTAlttBTS WITH SUPPPESSION POOLS ESULT IN CLOSE COUPLING OF CONTAlWENI PERFORPANCE WITH REACTOR SYSTDE DRAFT NUREG-1150 WIDE RAIEE OF CONDITIONAL CONTAlttENT FAILUPE LIKEllH00D, PRlPARILY DUE TO POSSBILilY OF LIER ELTTHf0VGH EVEN WIT 10fT liter ELTTHROUGH, DRAFT-1150 INDICATED RELATIVELY HIGH EARLY CONDITICHAL FAILURE PFOBABILITY O
WIDE RAfEE OF CORE DAPAGE FREQUDiCY O
2
-m O
O O~
l
)
i rSTIMATES* OF CORE DAMAGE FREQUENCY CORE DAMAGE FREQUENCY PER REACTOR YEAR (/RY)
AFTER A-45 l
PLANT (STUDY)
BASE **
MODIFICATIONS _
INTERNAL EVENTS EXTERNAL EVENTS f
Brown's Ferry (Draft - TVA) 2x10-3,
x x
~4 f
Brown's Ferr- (IREP) 2x10 r
Cooper (A-45) 4x10~4 10~4 x
x
~4 Millstone 1 (IREP) 3x10 x
-5 Peach Bottom (RSS) 3x10 x
w Peach Bottom (IOCOR) 4x10-6 x
-6,,,
I Peach Bottom (NUREG-1150 Draft) 8x10 x
Quad Cities (A-45) 2x10~4 10-4 x
x l
l 1
- Values estimated at different times for different purposes. For example, the TVA Draft
-Study was intended as a conservative bounding analysis for use in directing more detailed analyses of the final PRA. Since these studies were perfonned over a span of 12 years,
'significant improvements in emergency operating procedures and in ATWS protection in
)
recent years are not reflected in some studies.
- A factor of two reduction in mean Station Blackout core damage frequency due to A-44 modifications has been estimated.
- Statign Blackout contribution estimated at 7x10~0/R-Y
+
2x10' =.003 1
i
, ow C
o.
I f
4-e
- i i
3 PARK I C0hTAltf0ff PERF0WANCE PROGPM l
I I
1~
f i
4 I
EVALUATE NEED FOR IffR0eBIT FCR GENERIC CHALLENGES i
~
i FAY RESULT IN BACWITS i
~
i THAT LIELY WILL RESULT IN
~
lO Rutermm k
i
+
h i
1 t
i 1
r i
f i
i tr-l O
l 4
[
4 i
-r-
_--a,,n
-n m.
I.
O PROGRAMTASKS PA R R ON CHALLENGES a P0 N IAL FAILURE MODES PAP #ETRIC CALCULATIONS WOR 13 HOP TEONICAL ASSESSWhTS OF PONIAL IWROVBES COST /BEEFIT ANALYSES; OTHER CRITERIA IKIERIM REPORT FINAL REPORT i
i f
k i
I 1
l f
9 i
,i 4
i i
i e
l O
l i
5 a
o i
i j
1
__,,-.n
.--,,n--,---,.n,,,,
'~
- )-
e
(
N )
'w
t' ARK I OIALLBES & ELATIVE LIELiliOOD OF FAILUE ICDES WHAlleBK C0fRAlfFE!H ELATIVE O !All B E E C0tRAlfM3R FAILUE MODE FAIWE TIE LIKEllH000 CasuSTI0tr DRWELL RUPTURE PRIOR TO OR AFTER VESSEL BREACH LW EARLY TEAR IN WETWELL AIRSPACE &
PRIOR TO, AT, OR AFTER Hlat OVERPRESSURE /
LEAKS THROLGi DRWELL UTER VESSEL BREA01 OVERTEMPERATURE HEAD SEAL **
EARLY LINERITLTTHROUGH AFTER VESSEL BREACH VARIABLE ***
DVERTEMPERATURE LATE LEAKS THROUGH DRWELL UPPER SEVEFAL HOURS ++ AFTER START Hlai DERPRESSUREI HEAD SEAL WITH POTENTIAL FOR OF CCl+++
OVERTEMPERATURE A TEAR IN WETWELL AIRSPACE **
CORE CONCRETE BASEPAT PENETRATION MANY HOURS AFTER START LW litTERACT10tl 0F CCI cn BYPASS INTERFACING SYSTEMS LOCA AT ACCIDENT INITIATION LW FAILURE TO ISOLATE CONTAINMENT BYPASS AT ACCIDENT lillTIATIori LW RAPID STEAM EXPLOSIONS / MISSILES PRIOR TO OR AT VESSEL BREACH LOW 0/ERPRESSURE/
MISSILES
- FOR DEltERTED CONTAI! PENT
~
- BASED ON CHICAGO ERIDGE & IRON ANALYSIS OF PEAOI BOTTOM
- LIKELIHOOD OF LINER MELT THROUGi MAY BE DIFFERENT FOR HIG1 PRESSURE SEQUENCES
++ TIME TO FAltURE IS SHORTER FOR LIMEST0tE (HIGH GAS CONTEtTT) AS COMPARED TO OTHER LESS GASEOUS CONCRETE AGGR
+++ CCI - CORE CONCRETE INTERACTION L
t
'O StJt%RY OF FEB, 24-26, 1988 BWR MARK 1 WORKSHOP THREE DAY WEilNG WITH 150 It00STRY, ESEAR04ERS, STAFF NO PUBLIC REPRESBRATIVES INDUSTRY DPHASIS ON PPEVENTION, ANY FIXES SHOULD BE PL#ff S E CIFIC FROM IPE VARIETY OF VIDG ON PROBABILilY OF LIER ELT-TliROUGH
- MAlfER OF VESSEL FAILUE AfD RELEASE OF DEBRIS Iff0RTANT
- INDUSTRY BELIEVES WATER CAN P8 EVENT LIER ELTTHROUGH
- WATER BBEFICIAL, BUT NO CONSENSUS FROM NRC ESEAROERS ON WHETHER lifer FAILS At0 WHEN GBERAL AGREDENT -- WATER IN DRWELL USEFUL TO DELAY /PREVBR SHELL O
FAILUE Ata TO REDUCE FISSION PRODUCT ELEASES AGREDENT THAT ADS PELIABILITY ltPORTANT.
IPPROVEWNTS ACHIEVABLE AT t0 DEST 0)ST SUSQUEHAfMA LICENSEE TAKING ACTIONS tKM, POTB RIAL POSITlW AND NEGATIVE SAFETY ltFACIS OF VENilNG
- REDUCE CDRDELT LIKEllH000, EDUCE 00NSEQUENCES, BUY TIE
- POTENTIAL FOR UN!EESSARY PELEASE, INCEASE COPEELT LIKEllH00D FOR SOE SEQUBKES NYPA ASSERTS THAT THEIR VERSION OF STCP YlEWS SIGNIFICANTLY LOWER SOUR TEPPS, HAS OFFERED THAT NRC EXAMINE f0DIFICAT10NS. RES PUPSulNG, O
7
O O
O SlM%RY OF POIDRIAL I!PROV9ENTS POTB RIAL INPOVSTNT BENEFITS LUtt itLI ALLittNI POSITIVE EGATIVE PIBBRION MITIGATION VENTING CONTROLLED RELEASE
- MAY INDUCE CORE D4%GE X
X
- HARDENED PIPE PREVENT CONTAltNENT FAILURES
- FAILURE TO C0fRP.0L
- PROTECTION AGAINST INADVERTENT REDUCE CORE D#%GE FREQUENCY
- INADVERTENT OPERATION OPERATION
- DEINERTING HYDROGEN CONTROL REDUCE ELIMINATE C PBUSTION
- NONE X
1
- REDUCED DEINERTING PEP.10D
- ASSURED LONG-TEIN INERTING CONTAltfENT SPRAYS REDUCE LIKEllH00D OF
- MAY DEGRADE EQUIPE NT X
X
- BACKUP WATER SUPPLIES CONTAltNENT FAILURE BY
- SYSTEMS INTERACTIONS
- ASSURED OPERATION DURING SB0
- OVERPRESSURE /0VERTEffERATURE EFFECTS
- LINER N LT-TP. ROUGH
- IMPACT DURING DBA REDUCE CORE NLT FREQUENCY
- REDUCE ACCIDENT CONSEQUENCES CCRE DEBRIS CONTROL REDUCE LIKELIHOOD OF LINER
- EttiANCE DtWMARD CCl*
X
- REFRACTORY CURB NLT-THROUGH
- PDTENTIAL CHEMICAL
- SACRIFICIAL BEDS INCrNPATIBILITY ADS EreiANCEMENT REDUCE LIKELIHOOD OF HIGH
- NONE X
X
- AIR / NITROGEN SUPPLY EttiANCEE NT PRESSURE SCENARIOS
- ALTERraTE BACKUP POWER REDUCE LIKELIHOOD OF EARLY
- CABLING PROTECTION PRESSURIZATION LOADS l
REDUCE CORE DM%GE FREQUENCY j
TRAINING AND PROCEDURES GREATER LIKEllH00D OF
- NONE X
X COREELT PREVENTION NJD MITIGATION
- CORE CONCRETE INTERACTION
-m
+
e
- 4 O
PRELIMINARY CONCLUSIONS WORKSHOP USEFUL IN NARROWING SO E ISSUES INDUSTRY'S PRIMARY INTEREST IS ACCIDENT PPEVENTION CLOSURE APF3RS LIKELY FOR
- HYDROGEN CONTROL AS C&PLIANCE ITD1 UNDER EXISTING RULES
- RELIABLE WATER SUPPLY IN STATION BLACKDUT
- ENHANCED ADS REllABILilY
- VENTING
- HARDWARE PELATED ACCIDENT MANAG9ENT (TRAINING 8 PROCEDURES)
STRATEGIES f0RE FOCUSED RESEARCH NEEDED ON O
- VESSEL FAILURE Af0 DEBRIS RELEASE
- LIEllH000 0F LINER ELT-THROUGH WITH WATER & DEPRESSURIZATION STAFF INCORFORATING FINDINGS IN UPCOMING C0ft11SS10N PAPEPS POTENTIAL POLICY ISSUES RELATED TO VENTING, SAFELY GOAL LARGE RELEASE PROBABILITY, DEFENSE-IN-DEPTH i
1
[
I E
O 9
i
a s
i,
- O i
4 i
t e
l BALANCED APPROA01 TO RESOLUTION i
r t
1 I W ROVED HARDWAPE FEATURES i
~!
1 ACCELERATED IWLEENTATION OF EXISTING REQUIREENTS l
(ATWS, HYDROGEN CONTROL, POSSIBLITY SB0 ETC.)
1 4
I e
IWRWED E0P'S & TRAINING t
- O IWRWED PLM OPERATIONS E
S 1
h
+
d 1j k
's 9
L 4
i i
f
[
r t
f i
i O
3 t
i 10 1
j
.. _ _,_ ~ _
.:i i
b
~
O i
i SO E ULE 1
i INTERIM VAhK 1 EPORT TO COWi - 4/30/88 i
FINAL PARK I EPORT W/ECOWB0ATIONS - 8/30/S i
POSSIBLE CARRYOVER OF IN-CONTAlttENT CORE DEBRIS C0fRROL IWROVBENT - 8/30/89 O
~
OTHER CONTAltf06 TYPES - 8/30/89 l
I r
i f
i
)
i s
l I
l 11 l
\\
I