ML20209E939
| ML20209E939 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 07/10/1985 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1426, NUDOCS 8507120355 | |
| Download: ML20209E939 (87) | |
Text
- * " '
OR>IGINAL me
('S UNITED STATES OF AMERICA (j
NUCLEAR REGULATORY COMMISSION In the matter of:
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS Subcommittee on Diablo Canyon Plants Docket No.
Location: Washington, D. C.
Date: Wednesday, July 10, 1985 Pages: 1 - 86 ACRSOFF10ECDPY"*
I
~
Jo 10~:lemove " rom ACRS Vice ANN RILEY & ASSO
'ES V]
f Court Reporters Ob 1625 I St., N.W.
Suite 921 L
Washington, D.C.
20006 g
~IA2b PDR
1 1
UNITED STATES OF AMERICA 2
NUCLEAR REGULATORY COMMISSION 3
4 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 5
SUBCOMMITTEE ON DIABLO CANYON PLANTS 6
7 8
Room 1046 9
1717 H Street, N.W.
10 Washington, D.C.
11 Wednesday, July 10, 1985 12 The Subcommittee on Diablo Canyon Plants of the 13 Advisory Committee on Reactor Safeguards convened, pursuant to O
14
' notice, at 8:30 a.m.,
Chester Siess, Chairman of the 15 Subcommittee, presiding.
16 PRESENT:
17 C.
SIESS, Chairman 18 D.
OKRENT, Member 19 J.
EBERSOLE, Member 20 C.
MARK, Member 21 ALSO PRESENT:
22 E.
IGNE, ACRS Staff Member 23 24
(
25
2 1
PRESENTERS:
'2 H.
Schierling 1
3" S.
Brocoum i
4 L.
Reiter i
5 S.
Israel 6
L.
Cluff 7
R.
Fray 8
D.
Brand 9
10 11 12 O
14 15 16 4
17 10 19 20 21 22 23 24 25 h
e'-
3 1
P ROCEED I NG S p\\
'- (,/
2 MR. SIESS:
The meeting will come to order.
3 This is a meeting of the ACRS Subcommittee on the 4
Diablo Canyon plants.
I am Chester Siess, Chairman of the 5
Subcommittee, and the other members of the Subcommittee that 6
are present today are Jesse Ebersole, Carson Mark, and we 7
expect Dave Okrent sometime this morning.
In fact, I expected 8
him at 8 o' clock.
9 The purpose of this meeting is to review the 10 evaluation by the NRC Staff of the Licensee's long-term 11 seismic program plan.
12 The ACRS Staff member for the meeting is Elpidio
,m
(
13 Igne, seated on my right when he is sitting down.
-%L 14 The rules for participation in the meeting have been 15 announced as part of the notice in the Federal Register of 16 June 28, and a transcript is being kept.
I request that each 17 speaker please identify himself or herself and speak with 18 sufficient clarity and use your microphone so that they can be 19 recorded.
20 We have received no written comments from members of 2:1 the public and no requests for time to make oral comments from 22 members of the public.
23 For the benefit of the Subcommittee, there are 24 several documents involved.
The Long-term Seismic Program A
25 Plan is this volume, which we saw several months ago.
It was
F~
4 i
submitted, I think, as required by the license condition by Akj 2
January 31st.
3 We had an early review of this document at a meeting 4
in Los Angeles on March 21.
That is the meeting that was 5
supposed to be March 21 and 22 and ended up at 9:30 on March 6
21.
At that time, PG&E gave us an overview of the program 7
plan.
8 We had a large number of consultants present at that 9
time and had the benefit of their comments, both orally and, 10 some of them, subsequently in writing.
At that time the Staff 11 had not finished its review of the program plan.
12 Subsequent to that meeting, the Staff has reviewed 13 PG&E's program plan, has submitted to PGSE a list of questions 14 and comments, and those were responded to by PGSE.
15 You have or should have from what Igne sent out a 16 copy of the Staff's evaluation that is attached to the July 17 2nd letter from Novak to Furley.
The evaluation itself is 18 only five pages.
And also attached to it are the Staff 19 comments and PG&E's responses to those comments.
So that is 20 the material we have.
21 What we plan to do today is hear from the Staff 22 regarding their review and their conclusions, to hear from the 23 Licensee either as indicated on the agenda following the Staff 24 or, if it turns out to be useful, during the Staff 7g 25 presentation if there are questions that come up that we want
5 1
an immediate answer to.
And then near the conclusion of the em 2
meeting, we need to decide what should be taken to the Full s,
3-Committee and how we should handle that.
4 Now, to set the stage for that, et me s u g g e r. t that 5
there are two possibilities, two ways of handling this with 6
the Full Committee.
If we come out with a fairly clean 7
recommendation, something to the effect that the Staff has 8
reviewed the plan and has found it acceptable and we agree, 9
then I see no reason for the Full Committee to hear any 10 significant amount of presentation.
-11 I would propose if it is clean to handle it as a 12 subcommittee report to the Full Committee and give the Full
/ \\
13 Committee a short letter that they can endorse that fact.
14 If there is anything controversial, if the letter is 15 going to include any recommendations beyond those that have 16 been made by the Staff and agreed to by the Licensee, then 17 obviously we will have to let the Full Committee hear the 18 arguments, So we will settle that at the end of the meeting 19 after we see how we have come out.
20 But it seems to me that spending an hour and a half 21 which we have scheduled of Full Committee time may not be 22 necessary.
23 Does anybody have any problems with that or with 24
'those possibilities?
We will settle on the approach later in 25 detail.
6 1
Now, this meeting is going to adjourn at 12
?
(,/
2 o' clock.
.There is another meeting scheduled at 1 o' clock in 3
this room, chaired by one member of this Subcommittee, and I 4
suspect it will be attended by others.
So we are going to 5
have to make 12 o' clock pretty close.
We could stretch it, I
6 guess, to 12:30 if we had to.
We could stretch it to 1 7
o' clock if Dave doesn't want to eat lunch.
8 Are there any comments or questions by members of 9
the subcommittee before we get started?
10
[No response.]
11 Any comments by the Staff on the proposed agenda?
12 MR. SCHIERLING:
No, there are none.
13 MR. SIESS:
All right.
Then we will hear from the 14-Staff, and I guess Hans Schierling is going to lead that off 15 as project manager.
16 MR. SCHIERLING:
I would like to give you first of
, 17 all a very brief summary where the Staff currently is standing e
18 with regard to the licensing of Diablo Canyon, Unit 2,
and 19 then turn the meeting over to Steve Brocoum.
20 As you will recall, Unit 1 was issued a full power 21 license in November
'84, and since May
'85, the unit has been 22 at full power operation.
Unit 2 received a low power license 23 in April
'85.
The fuel loading was completed in May.
24 Currently the Licensee's schedule is that the unit 25 will go critical by the end of July and would be ready to go
7 1
about 5 percent in early August.
2 At this time a Commission meeting has been scheduled 3
for July 30th, at which time the Staff will brief the 4
Commission regarding its recommendation for issuance of a full 5
power license.
6 MR. MARK:
Excuse me.
For information which I 7
should know but don't, are Unit i and Unit 2 in absolutely all 8
respects identical?
9 MR. SCHIERLING:
In all respects and absolutely, 4
10 no.
They are essentially identical. There are some 11 differences. One difference is there is a slight difference in 12 power level I think it is less than 5 percent, and that is
~13 due to a more advanced design of the turbine generator.
There
}
14 are also some small differences in the reactor internals, as I 15 recall, and I am sure there are certain differences in the 16 plant layout, in equipment layout.
17 You will probably recall that the two units are a 18 mirror image of hower.
However, there are still some 19 differences due to fuel run piping and things of that nature.
20 MR. MARK:
Amongst these differences, are there any 21 which would obviously require separate and different 22 consideration of being suitable for operation?
23 MR. SCHIERLING:
No, there are none.
24 MR. EBERSOLE:
Let me ask a question.
Yesterday we 7g 25 were talking about reactor operations instance, and a
8 1
fascinating event was discussed.
I think it's based on the
[x
")
.\\%j 2
Westinghouse units at Sequoyah where the flux scanning 3
mechanisms presented a seismically vulnerable possibility of 4
having excessive water drawdown from the bottom of the reactor 5
vessel through the flux scanning tubes due to a disruption of 6
those tubes from a seismically-induced fall of equipment on 7
top of them at the deck.
8 Is Diablo subject to that problem?
9 MR. SCHIERLING:
I could address that right now, or 10 I have it on my list.
11 MR. EBERSOLE:
Okay.
If you have it on your list, 12 forget it.
f 13 MR. SCHIERLING:
I will talk about that later.
N 14 Let me first introduce other members of the Staff.
15 We have here Steve Brocoum from the Geoscience Branch, Sandy 16 Israel, we have here. our consultants and advisers, 17 Dr. Slemmons and Dr. Brown.
Unfortunately, we do not have 18 here Jim Knight, the Director of the Division of Engineering.
19 You probably know that he is part of the Staff today with the 20 Commission testifying before Congress.
21 As I mentioned earlier, we issued the Unit 2 low 22 power license in April
'85, and had documented our safety 23 evaluation in a number of reports that address specific 24 issues.
One of them was allegations, design verification, gg 25 piping and supports, and the last one was No. 31, which
9 1
addressed numerous and miscellaneous licensing items.
2 As a result, the Unit 2 low power license is a 3
relatively clean license with only a few license conditions 4
that are out of the ordinary.
We are currently working on a 5
report.
In all likelihood it will be again a safety 6
evaluation report to provide the bases for our recommendation 7
to the Commission on full power licensing.
8 That report, again, will address allegations, will 9
address the two items that we had identified earlier as 10 needing resolution prior to full power operation.
One was the 11 evaluation of the shear capacity of a construction joint in 12 the turbine building for Unit 2.
The other one was the 13 seismic design of the pipe way structure for both Unit 1 and 14 2.
15 We will report on the current status of 16 modifications and also on emergency preparedness with respect 17 to the recent decision, GAR decision, I call it, regarding 18 medical facilities.
l 19 You might recall that the Appeal Board, the Atomic 20 Safety and Licensing Appeal Board had reached a decision 21 regarding the design verification only for Unit 1 and had s
22 specifically stated that they would make its findings at a 23 later time.
That was recently done, I think, in late June, in 24 ALAB 811.
25 The Board reaffirmed the NRR licensing authority to
10 1
issue a full power license, of course, after briefing the im 2
Commission.
3 I mentioned earlier that the Commission and some of 4
our staffers today are testifying before the Markey 5
Subcommittee on matters of emergency preparedness and 6
consideration of earthquakes, and that is the reason why Jim 7
Knight is not here.
8 Dr. Ebersole, you mentioned the potential for a 9
systems interaction where potentially a part of the flux 10 mapping system, the INCO flux mapping system 11 MR. SIESS:
Hans, I don't think that all of this 12 information is particularly pertinent to our review of the
/
13 long-term seismic program.
D) 14 MR. SCHIERLING:
Do you want me to skip that 15 particular item?
16 MR. SIESS:
Yes. Being updated on the licensing 17 status of Unit 2 is really not part of the agenda, and I would 18 like to keep the meeting focused on the long-term seismic 19 program today.
20 MR. SCHIERLING:
Okay.
Let me then go right into 21 that.
22 MR. SIESS:
Okay.
23 MR. SCHIERLING:
Steve Brocoum will make the 24 presentation for the Staff However, just to lay the
\\
25 foundation, as you mentioned earlier, currently there is a
11 1
condition in the Unit 1 full power license for the seismic b)
(,e 2
reevaluation program.
3 MR. SIESS:
Can we assume that that will also be a 4
condition in the full power license for Unit 27 5
MR. SCHIERLING:
The Staff does not intend and does 6
not see any need to include that particular license condition 7
in the Unit 2 full power license.
We do not have it currently 8
in the Unit 2 low power license --
9 MR. SIESS:
The units are close enough that the 10 seismic hazard for one will apply to the other.
11 MR. SCHIERLING:
Our rationale is that most of the 12 information that is being generated is not unit specific but a
13 site specific.
Secondly, the Units, as you pointed out, are 14 essentially identical, and we expect that the results would be 15 equally applicable to Unit 2.
We do not, therefore, intend to 16 include that as a license condition in the Unit 2 full power 17 license.
18 We did send to you our evaluation, I think it was by 19 July the 2nd, of the proposed program plan.
At this time the 20 plan has not been approved by the Staff.
The Technical Staff 21 has completed its evaluation.
However, it has not gone 22 through a detailed management review.
23 We expect to approve the program probably by the end 24 of July, depending upon the comments that we will receive from O
25 you.
d
~ - -. _. - _
l 12 1
I would,like to point out one important element in i
f
.2 the program, the way it currently stands, and that is 3
flexibility.
The program has flexibility to account for 4
changes that might become necessary as a result of information 5
being available.
6 As you know, PG&E currently is engaged in a scoping 7
study that will be completed by the end of November.
We again 8-will factor that into our review process.
9 In accordance with the license condition, the 1
10 Licensee has three years to complete the program, is expected 11 to complete the program in three years.
12 Once the program has been approved, that would mean, 13 if we are talking July
'85, August
'85, we expect the program 14 to be completed by mid
'88.
15 These are my comments, Steve.
Unless there are any 16 questions, general questions to the current status of 17 licensing or the program in general, I would ask Steve to go 18 ahead.
19 MR. SIESS:
One question:
20 The license condition is three years after the 21 program is approved?
22 MR. SCHIERLING:
That's correct.
23 MR.'SIESS:
And your approval is essentially the
'/..
24 Staff review we have, subject to our comments, and management 1
\\
25 review or whatever?
i
3 13 1
MR. SCHIERLING:
That is correct.
2 MR. SIESS:
Okay, and you expect to have that by 3.
July?
4 MR. SCHIERLING:
By the end of this month.
5 MR. SIESS:
Okay.
Let me suggest something.
I will 6
use the term to expedite things, and I don't want it to at all 7
appear to be railroading anything.
I have read the Staff 8
evaluation, as I believe the other members have, and as I read 9
the Staff evaluation, the Stcff is satisfied with the 10 program.
There is a statement in the thing that the Staff 11 concludes that the amended program plan is responsive to the 12-license conditions for the evaluation of the seismic design 13
' basis.
14 Now, that is as close to approval as anything I have 15 over seen.
16 Now there is, I believe, one important qualification 17 and that is that the program should be kept flexible.
The i
18 Staff concludes that the program is sufficiently flexible to 19 allow revisions that may be necessary to accommodate new 20 information obtained during the course of the program.
21 And I also got the impression that the Staff thinks 22 that the program is flexible and would approve the degree of 23 flexibility that would permit the program to be completed 24 within the three years specified.
And I think that is
\\
25 important, because this is a very extensive program and yet
14 1
three years is the license condition.
2
'Without the three-year condition, this is a program 3
I think that could go on forever because as soon as a 4
geologist gets new information, you know, we've got new 5
questions, should have new answers.
But usually we have new 6
questions.
7' But in view of the general agreement of the Staff 8
with the program, or the general approval of the Staff of the 9
program, I would like to propose that we not go into detail as 10-to why the Staff approves the program.
11 I would like to suggest that we not go through the 12-discussions that have been held between the Staff and the 13 Licensee, the Staff comments, the Licensee res'ponses, the 14 c h t.n g e s, explanations and so forth, but simply see if we are 15 sattsfied that'the Staff knows what they are doing approving 16 4t, and see if there are any open questions that the 17 suhecmmittee has.
18
.And I think one question the subcommittee definitely 19 wants to review is the' question of the PRA level, and that is 20 l i !c e l y to take some time.
21 Now, Dave, are you satisfied with this, that you
\\
22 don't want to hear a lot of reasons as to why the Staff s,
23 approved?
24 MR. OKRENT:
If the Staff has any fairly significant
\\
t 25 concerns after its discussions with some specific parts of the
f 15 1
-program that maybe_after the three years, that have been 2
adequate for the purpose, so we're left with an awkward 3
position over this sort of thing.
4 MR. SIESS:
Put your mike on.
-5 MR. OKRENT:
I would like to hear what those are.
6 If they don't have any, they can just say they don't have-7 any.
That would be a way of getting to the heart of whether 8
there is anything important, or what you might call an open 9
area between the Staff and Licensee.
10 If we could do that first, then we could proceed on 11 to-see whether the subcommittee has questions in those areas, 12 and at some point we would like at 13 MR. EBERSOLE:
Chet, I agree with you, that this is 0i 14 no place to take up specific vulnerabilities of the Diablo 15 Canyon Plant.
16 For instance, the seal vulnerability or this new 17 phenomenon here.
18 MR. SIESS:
That wasn't on the agenda.
19 MR. EBERSOLE:
Yes, I understand, and I understand 20 this is a more general topic that we are working on.
21 Now one of the things that always bothers me about 22 this type of thing we are undertaking here, I guess I will 23 call it the degree of relevance of all this vast amount of 24 information to the specific mechanical issues that this plant j
25 faces.
The probability of a given earthquake of a certain I
- - -. ~ - - - - -
r 16 1
magnitude, I see literally piles of a variety of
(,_j/
2 informational types, and I have a great deal of difficulty 3
getting any relative understanding of the degree of importance 4
these bits of information have to the ultimate problem at the 5
plant site.
6 If we can find a method somehow that says this as you said, 7
particular phenomenon that we have dug up here is or is not relevant to the 8
it might go on forever 9
ultimate mechanical problem the plant faces, I think we should 10 do so.
11 MR. SIESS:
Well, that is the objective of the long 12 term plan, Jesse.
At the end of three, years from the end of
[
13 this month, the Licensee will have reviewed essentially all V}
14 those factors that relate,to the seismic hazard at the site.
15 MR. EBERSOLE:
- Well, I'm talking about waiting for 16 those factors.
17 MR, SIESS:
I know.
And the fourth item in the i
18 license condition was that PG&E shall assess the significance 19 of the conclusions drawn from the seismic evaluation studies, 20 and that's where it comes down to what does this mean.
i I
L 21' If it turns out that the seismic hazard is larger 22 than the plant was designed for, or larger than was thought 23 about at the time the plant was designed, they are to assess 24 the significance, and that is the fourth item in the license 25 condition.
17 1
Incidentally, that license condition goes on to say 2
shall assess the significance, utilizing a PRA and 3
deterministic studies as necessary, to assure adequacy of 4
seismic modeling.
5 MR. EBERSOLE:
Yes.
Well, my problem is, of course, 6
the coupling of the importance of an event that may be found 7
or some phenomenon found to the actual ultimate problem at the 8
plant.
The fact that we dig up some distant fault somewhere 9
five miles away, or wherever, I have terrible difficulty 10 relating that to the specific condition that may ultimately 11 exist at the plant.
12 MR. SIESS:
Well, this is the object of the
(~
13 program.
Go from that fault to try to find its significance
(
14 in terms of the groundmotions at the site, and then after they 15 get that,.they are to look at the significance of the plant 16 safety.
And it is all in there.
17 Now as you read this thing, it looks like everything 18 is related to the geology and the seismology, but the bottom 19 line.is the significance to the plant.
20 MR. EBERSOLE:
Yes.
And how you arrive there is a 21 weird process, indeed.
22 MR. SIESS:
That is the bottom line.
All of this 23 stuff we go through to get there dominates the physical 24 effort, or at least it will for the first two and a half or L
25 three years, i
I
4 d
18 1
Carson?
'}
s,/
2 MR. MARK:
Nothing.
3 MR. SIESS:
Okay.
Would the Staff like to respond 4
then to Dr. Okrent's question?
Are there any issues that 5
bother you?
6 MR. BROCOUM:
Steve Brocoum with Geosciences Branch.
7 I gather the subcommittee doesn't want the 8
presentation which we had planned, which was just a summary of 9
our evaluation.
10 MR. SIESS:
We trust you.
11 MR. BROCOUM:
Thank you.
12 However, I would like to bring to the subcommittee's 13 attention two items which we think are important.
The first
- (O) and it is shown in one of the viewgraphs --
14 is the major 15 one of the major concerns of the Staff was the emphasis in the 16 program plan on the collection of data and information as 17 opposed to the emphasis on what the major issues are, and what 18 their priorities are.
19 That, I think, was the major technical, in a 20 nutshell, concern of the Staff.
21 In our meeting with PG&E on May 9th, where they 22 responded orally to our concerns, and in their submission on, 23 I think it was, June 11th, PG&E did discuss briefly what in 24 their opinion were.the major issues and they did make a very l - (v) gg t
25 gross prioritization.
l l
, = -
19 i
However, PGAE has deferred discussing these in great
.f~
l 2
detail until their scoping study is completed in November of 3
this year, so that that response to the Staff's concern will 4
not be forthcoming in detail until November.
5 So our approval or recommended approval, I should 6
say, is contingent on that scoping study.
And a lot of the 7
responses to a lot of the other less important, if you like, 8
concerns also will be included in the scoping study.
9 So our only concern, I guess, at the moment would be 10 that PG&E will complete and do an excellent job on that 11 scoping study.
12 MR. SIESS:
Now the issues you are talking about
/~ )
13 here are geological, seismological?
14 MR. BROCOUM:
Geological, seismological, tectonic 15 and, I guess, would include, I guess --
16 MR. REITER:
I think we have addressed the issue 17 that Dr. Ebersole raised about significance.
We discussed 18 that with PGSE, that that indeed is a very critical 19 phenomenon, what is the significance of all these issues.
And although, as Dr. Ebersole said, there may be some 20 if 21 interesting fault out there, but if it has no bearing on 22 safety of the plant, it's not worth investigating.
23 MR. BROCOUM:
Presumably this logic tree analysis 24 that PG&E is doing will help assess significance, for example, O
25
'of tectonic features maybe 30 or 40 miles away from the plant,
20 1
and whether any major effort should be expended on studying
/
k 2
them, as opposed to studying some other feature nearer the 3
plant or in the plant, for that matter.
4 So I guess what I'm saying is our approval isn't 5
wholehearted.
It is contingent on their scoping study.
But I 6
should also add to that, that based on all -- we have had at 7
least five meetings with PGSE during the fall and spring, and 8
when-they have promised us things, they generally have 1
9 delivered on those promises, t
10 So we are fully expecting that PGSE will have an 11 adequate scoping study in November.
So that is what our 12 approval is based on.
13 MR. OKRENT:
At the end of the scoping study, are
~s 14 you supposed to be provided both with information as to what 15 among they had planned that they now deemed to be significant 16 in dealing with the primary issues, and what may be 17 peripheral, and an estimate of what may be missing from the 18 program as it was originally planned, that may be necessary to
\\
l 19 treat the primary issues?
20 MR. BROCOUM:
That is correct.
That is our 21 understanding.
22 MR. MARK:
I think it is almost on the same point, 4
23 and I am sure you have answered it already.
In the Staff's 24 questions or comments on the plan, there were a very 25 considerable number of faults which were mentioned, a fair i
e v-u a,-
.n-,en, n-n.-,, -.,,.,.,,,
,.--n-.-e--.-,,--
21
'1 number of them ones which I at least was unaware of mI
')
(
f 2
previously, and I had the question as to whether all of these 3
indeed need to be studied.
4 I take from what you said that the scoping study 5
might say that the Santa Maria and the Santa Margherita and 6
the Santa Miscellaneous 7
[ Laughter.]
8 are --
9 VOICE:
That's sacrilegious.
10
[ Laughter.]
are not really either large enough or 11 MR. MARK:
12 close enough or tied in enough to have to be investigated in 13 detail.
14 We know enough about them to be able to lay them 15 aside.
Or is it really going to be necessary, in your view, 16 that they really mix some borings and some seismic soundings 17 for each of those things that were mentioned in the comments?
18 MR. BROCOUM:
I don't know if they will have to do 19 it for each of the faults that are mentioned, but for some 20 faults it may be possible to 21 MR. MARK:
Some you recognize that you really do 22 want more information on?
23 MR. BROCOUM:
Yes, and one of the people we have to 24 help us in this is Dr. Slemmons.
And I should also point out 25 that his contract is now in place, and he will be working in
m 22 1
that area this summer.
And I think between PG&E's work and b
(,/
2 Dr. Slemmons' work, we will be able at the time of the scoping 3
study or shortly thereafter to help resolve which are the 4
important tectonic features near the plant and which are not.
5 Maybe Burt Slemmons might want to add a little to 6
that.
7 MR. SLEMMONS:
Burt Slemmons speaking.
8 The things that we will attempt to do during the 9
summer will be to look at both imagery and do angle 10 reconnaissance, and aerial photography, to look at the faults 11 away from the site, 10, 20 or so miles, to determine whether 12 or not they show signs of activity and whether they will give 13 slip directions and belong to the general pattern.
And so wo 14 will have a look at that particular relation.
15 MR. MARK:
And this might curtail the list of total 16 possibilities?
That is, some might drop out.
17 MR. SLEMMONS:
Yes.
18 MR. MARK:
Thank you.
19 MR. SIESS:
Now can we assume that at some point 20 there will be an agreement between the Staff and the Licensee 21 on those things which are not significant enough to be 22 investigated further?
23 MR. BROCOUM:
We hope so.
24 MR. SIESS:
I mean Staff is willing to agree at some 25 point that this you can put aside and concentrate on something 4
--,-m
--ve,-.
,,--,,,,---.-,,p.-
e v.,--m,
-,----v--,-
,,,e-n--,-,e.
~,,,,, -
e-,
23 i
else?
i 2
MR. BROCOUM:
We would like to agree with PG&E as to 3
what the important issues are, and what our priorities are for 4
those issues, yes.
That, I think, would be a reasonable goal 5
fairly early in the program.
6 MR. SIESS:
Is what they are proposing sufficiently 7
comprehensive as of now that you think everything is there and-8 that nothing is likely'to be added, only something is likely 9
to be put to one side?
10 MR. BROCOUM:
Well, we think that the program is, as 11 we stated in the past, it is a comprehensive program.
And in 12 some instances it may have to be modified in order to be 13 completed in three years.
But there may be other instances g
14 where something may have to be added, some particular type of 15 investigation which wasn't obvious to us when the program was 16 planned, but may become obvious later on.
17 So I think we have looked into that carefully, and 18 we feel that the program is flexible enough to accommodate 19 those types of things.
4 20 MR. SIESS:
Now your interaction with the Licensee 21 on this will be at what frequency?
it is a 22 MR. BROCOUM:
The interaction will be 23 requirement of the license condition that we have meetings 24 twice a year and they submit quarterly progress reports.
)
/
25 Based on our past experience in less than a year, we have had
24 1
five meetings with them and many other types of interactions.
(V 2
MR. SIESS:
Okay.
Because meetings twice a year 3
doesn't sound quite frequent enough to get this, and if it is 4
a written question, written response type thing, that takes 5
six months to go through a cycle of that.
6 What I am looking for is some assurance that the 7
Etaff will work closely enough with the Licensee and vice 8
. versa that the interaction won't introduce delays.
9 MR. BROCOUM:
I should point out that these are 10 formal meetings.
We have had numerous weekly, several times 11 weekly informal conversations by telephone meetings and other 12 settings with the Applicant or Licensee for over a year.
to say we had 100 contacts would 13 I would say k
14 probably not be exaggerating the. number.
We have constant 15 informal -- I was talking about formal meetings, open 16 meetings.
But in terms of infermal meetings, we have constant Lloyd Cluff and I talk on the phone several times a week.
17 18 MR. EBERSOLE:
Chet, when you find in your searches you have agreed to the fact that it is 19 that some phenomenon 20 Irrelevant to the problems of seismic design at Diablo Canyon, f
21 will you document that finding of irrelevance for the finding?
22 I have often wondered, you know, who was it that 23 ever said you didn't need to look to the sea as you looked at 24 faults?
25 MR. BROCOUM:
That's a very good point, and we have
25 1
been very careful to document everything to do with the (A) 2 license condition to date.
For example, when we sent out 3
Staff comments to PGSE, they came in and gave us a verbal 4
response, and then we asked them to formalise those responses 5
so that those would be documented.
We have been very careful 6
to document everything up to date, and I think we are very 7
careful to do that.
8 I would like to maybe cdd one thing that is not in 9
the Staff evaluation which might be of interest to the 10 Subcommittee.
We had a site and regional geological 11 reconnaisance on June 24 and 25 of this year, what I look at 12 as a kick-off meeting for the investigations to do with this 13 and it's in one of the Vu-graphs in the package we gave out
'kJ 14 here.
'15 The purpose was-to get all the important characters 16 it is a Vu-graph that is about six Vu-graphs back.
The 17 purpose was to get all the geological, tectonio and 18 seismological participants both for PGSE and its advisers and l
19 consultants and from the NRC and its advisers and consultants i
20 together so that we could all view, if you like, the tectonio 21 features in the vicinity of the plant so we oculd all kind of 22 be starting from the same perception, if we could, and begin-23 the evaluations, 24 It was also to familiarize people like Dr. Slemmons, 25 who had never been there, with the plant.
It was to get the i
.--, - - - _ - - -. _ - - - - - - _.. --- -- - - - - -. - - - - - - ~-
26 1
USGS to give us their insight.
So the participants included O
'(,/
2 PG&E and its consultants, the NRC, which included Dick 3
McMullen, the geology reviewer, Bob Rothman, the seismology 4
reviewer, myself, the USGS, which was Bob Brown, the 5
University of Nevada, which was Bert Slemmons.
6 We spent two days in the field.
The first day we 7
made a series of overflights throughout the area as far east, 8
about 50 miles east to the San Andreas Fault.
We looked at 9
that fault and all the other major faults in the region under 10 low sun angle conditions to see from the geomorphic features 11 if we can get some feeling or understanding as to how active 12 they are.
l\\
13 What we did is we looked at the San Andreas Fault 14 first, which we know is active and is the largest fault and 15 the most active, I guess, in California, and then we went to 16 look at other faults and we used the San Andreas as a 17 benchmark.
On the second day, we went up to the area where 18 the San Simeon Fault near Ragged Point comes on shore and 19 spent some time out there looking at the fault.
20 So we had this kickoff meeting at looking at the 21 geology and tectonios around the site, and that is in the 22 Vu-graphs but not part of your evaluation.
23 The date on the Vu-graph is incorrect.
It was the 24 24th and the 25th, not the 17th and 18th.
So that is another b)
\\\\'
25 piece of information that is new that is not part of the
27 1
evaluation.
',o N,)
2 MR. OKRENT:
You didn't throw any cigarettes i
3 overboard, did you?
4 CLaughter.]
5 MR. BROCOUM:
It was just a week before all these 6
fliers.
7 We will be issuing, of course, a trip report on this i
8 reconnaissance trip to Diablo Canyon.
9 MR. SIESS:
Any other comments that the Staff wants 10 to make?
11 MR. BROCOUM:
No, we have no other comments at this 12 time.
13 MR. OKRENT:
Could I ask one question?
There was a 14 question that the Staff asked of the Licensee.
I don't 15 remember the exact wording, but it was to the effect of how do 16 you get enough diversity of opinion in trying to estimate the 17 seismic hazard since it is relatively well known that 18 different experts who end up, if left but to their thoughts, 19 with differing opinions?
And the Licensee responded you would t \\
20 have a panel of efforts for this purpose, which on the surface 21 sounds okay, but from the experts that I know, I could pick 22 two panels seemingly at random and predict in advance that 23 their mean or average prediction would vary by an order of 24 magnitude in the direction in which --
25 MR. SIESS:
An order of magnitude on what?
Not on i
28 1
A
/
)
,/
2 MR. OKRENT:
On frequency of earthquake of a certain
(
3 size or so forth for the same site, and predict in advance 4
which way the panels would go.
So having panels of itself 5
doesn't necessarily introduce what I would call a breadth of 6
opinion.
That is the first point.
7 MR. SIESS:
A full range of uncertainty.
8 MR. OKRENT:
The second point is different.
If one 9
looks back in history -- let's say first let's think about 10 nonseismic plants.
You can recall that from time to time, 11 there was a large body of opinion, let's say, in the 12 scientific community, a prevalence of opinion with one or two 13 who thought something different.
And if at the time you took 14 a poll of the experts, and even if you brought in these one or 15 two outcasts to bring in their opinion and averaged it all, of 16 course you would still get the prevalent opinion as the 17 answer.
And as you are well aware, from time to time the 18 prevalent opinion was quite wrong.
19 I suspect that if someone were to go back and look in fact, I'm fairly confident that if someone were to go 20 21 back and look at the matter of both the amount of seismioity 22 and so forth, that one would expect in the west part of the go back 20, 25 years 23 San Andreas or even on the San Andreas 24 or so, look at prevalent opinion, or if one were to look at N
25 the estimates of the risks to a nuclear power plant from the
T 29 l
1 seismic initiator and looked at opinion from around the time i
t
\\
2 of 1970 to '75 leading up to and including WASH-1400, again j
3 there was a prevalent opinion that in fact has proven to be 4
wrong.
5 The question I'm introducing -- which I don't is how, when you try either 6
pretend is an-easy question 7'
within your own group or when the Licensee tries within his 8
group, how do you see to it that enough of a diversity in 9
ideas as well as opinions are brought into this ultimate 10 estimation of what the seismic hasard at the site is?
11 Is the question clear?
12 MR. BROCOUM:
Yes, and it's a very good question.
I but I want to make one comment first.
13 am going to ask Leon 14 This is about that science, and particularly earth sciences in 15 particular are full of examples where people who were outcasts and a famous example is Alfred Wagner, who came 16 at one time 17 from Germany in the 1920s to the American Geophysical Union 18 meeting and gave papers on plate tectonics, conical drift, 19 and he was hooted, literally, out of the meeting, and 20 ultimately, of course, he ended up being correct.
21 So that is a concern that we always have, is if 22 someone who is on he outskirts from the normal curve, is he 23 the genius that we are all missing out on. So we are very 24 aware of that.
That question of yours does not surprise me at 25 all.
So I am going to speak primarily right now from geology
30 1
tectonios.
(
2 We always look at all the opinions.
We read all the 3
papers and we evaluate all of them.
The difficulty is when 4
you have a paper that is so different than the norm, is this 5
person "off the wall" or does he really have some new insight 6
that nobody else has ever seen?
7 I think you have to handle that on a case-by-case 8
basis.
You have to keep your mind open and you have to make 9
an effort to keep up with not only the mainstream of what is 10 going on but with diverse opinions from people who are outside 11 the clique, if you like, of a normal group of people.
12 The only thing I'm trying to tell you is the Staff 13 is aware of that and makes an effort to do these things.
I 14 will use an example.
The Crouch paper was a little bit out 15 from the mainstream on the Hosgri Fault, and we recognized 16 that immediately, that it was a new interpretation, and we 17 cidn't dismiss it, as I think the Subcommittee knows, as being 18 off the wall.
We considered it, we realized it was a good 19 paper, and I think we acted appropriately.
20 I think the Staff does make an effort to keep an 21 open mind on these things.
22 Now, in terms of the range of opinions, the first 23 part of your question about setting up the panels and the 24 groups, I think maybe I should let Leon discuss that a little 25 bit.
31 1
MR. SCHIERLINO:
Steve, let me add one thing.
I O
( /
2 think it is exactly in response to such a concern that we are 3
now engaging on this program to see if and what new 4
information currently should be considered, so I think that is 5
one of the reasons why we have the program.
Again, as Steve 6
indicated, the Crouch paper was an example where PG&E and the 7
Staff acted almost immediately to consider that new 8
information and evaluated it regarding its applicability and 9
relevancy.
10 MR. OKRENT:
By the way, is it still thought to be 11 largely relevant or not?
Is the Crouch paper still thought to 12 be largely relevant or not?
13 MR. BROCOUM:
It's an important paper, but I would 14 say that since the Crouch paper came out, there have been 15 three or four newer papers already published or in press on 16 the same region with somewhat different interpretations.
l 17 MR. OKRENT:
But not diametrically opposed 18 MR. BROCOUM:
Well, some of the new papers are still 19 suggesting tens or hundreds of kilometers of strike / slip 20 movement, so in that sense they are opposed.
21 MR. OKRENT:
Yes.
Well, if I can carry the 22 conversation a couple of minutes further, with the permission 23 of the Subcommittee Chairman, I think we are not going to have 1
24 major topics that need the full 12.
I have recently seen a j
i k
25 few papers for the east of the Rookies, one about new faulting l
l l
l 32 1
somewhere in Oklahoma.
I 2
MR. BROCOUM:
The Meers Fault.
'3 MR. SIESS:
That was discovered in NRC research,
-4 wasn't it?
5 MR. OKRENT:
It was an NRC report.
6 MR. BROCOUM:
It wasn't discovered in NRC research, 7
but we support all the studies on it since it was first 8
mentioned in the literature.
9 MR. OKRENT:
And I saw one that suggested that maybe 10 New Madrid in all its glory does link up into Kentucky and so 11 forth and so on, as-has been suggested from time to time.
12 MR. SIESS:
Just don't move it north.
13 MR. OKRENT:
North and east is the direction.
I 14 can't tell when the NRC gets things like this whether they
{
15 respond with what I would consider an appropriate emphasis on 16 the possible both local and generio implications of this new 17 suggestion, new measurement or so forth.
There certainly has 18 to be on the part of the Staff a very understandable 19 reluctance not to find that some value of safe shutdown that 20 was approved at the construction permit stage, reapproved in 21 some different form but safe level at the operating license 22 stage, a reluctance to have this upset.
So one can go rather 23 slow about trying to see how important some new piece of 24 information is, k
25 I don't know if you do.
On the other hand, I don't
33 i
i see big outbursts of activity that follow up on some of these
\\,/
2 things.
So I don't have a good feeling for the way the Staff 3
decides to respond and how it decides whether or not it has to 4
set high priorities on certain things, and that could enter in 5
this review. I don't know in what way, but it might.
6 MR. BROCOUM:
I can make several comments on that.
7 First of all, in terms of the program plan, one of the things 8
we said early on in our meetings with POSE -- and I think Don that we should have a 9
Brand will remember this 10 comprehensive enough plan that surprises, if you like, such as 11 the Crouch paper, are very unlikely to happen.
Of course, you 12 can never guarantee that you will never be surprised, but we 13 should have enough understanding, a review of all the tectonic 14 models that are proposed, certainly the ones that seem 15 credible by a good part of the scientific community, to avoid 16 the kind of a surprise we had just as Unit 1 was being 17 licensed.
18 So that is a specific goal of designing the program i
19 plan in this case, 20 MR. OKRENT:
If I could interrupt one moment, from 21 what little I understand of the history, had people decided 22 that we didn't really know all we needed to know about Hosgri l
23 and so forth and that the USGS paper saying there was a Hosgri l
24 and then really wasn't Commandment take a magnitude, 7.5 25 No. 11 written on a tablet, one might well have been trying to
- ~ _ - _.,
~
34 1
see is there information from proprietary sources -- as one
(_j 2
knew very well there is lots of proprietary work going on in and'does it tell us anything or suggest anything 3
the area 4
markedly different?
5 And in a sense, as far as I'm concerned, the chance I will call it the chance -- publication of a paper by 6
7 Crouch seems to me to have not been the way I as a member of and I'm not talking about being a member of the 8
the public 9
ACES, but a member of the public would hope that either the 10 Licensee or the NRC dealt with that particular seismotectonio 11 situation.
12 MR. BROCOUM:
I think I should point out that wo 13 were informed of that paper by PGAE, and we were informed of 14 it prior to its publication.
15 In fact, as soon as PGSE learned of that paper, it 16 was presented orally in a meeting, and I assume PG4E attended 17 that meeting on the West Coast 18 MR. OKRENT:
But that is not my point, though.
My 19 point is that the information on which Crouch and co-authors 20 finally decided to develop a paper wasn't obtained the night 21 before they gave that talk, and it is known, there has been 22 lots of interest by oil companies in the whole area, and i
i 23 information must be getting obtained, you know.
24 MR. BROCOUM:
Yes, you are correct.
I think the U
I showed a slide indicating 25 subcommittee about a year ago
35 1
that literally hundreds of lines off the coast of California
(),)
,r-2 near the Santa Maria Basin and north.
So there was a lot of 3
information.
4 But even those types of interpretations are very 5
interpretative.
6 In other words, two seismic experts can, looking at 7
the same data, come up with two different interpretations.
It 8
depends in part on the interpreter's preconceptions of what 9
the tectonics are, and if you think it is a thrust fault, you 10 will tend to see thrust faults.
And if you think it is 11 vertical strikes, you may see those.
12 So I don't think, although the Crouch paper was an 13 excellent paper and the data was very well presented, I think
(}
\\_J 14 that until he put it together, I am not sure that another 15 interpreter or one of the Staff would have necessarily come up 16 with his interpretation.
I 17 MR. OKRENT:
But you are not addressing the question 18 of did the Staff, recognizing that there was lots of new data 19 since 1977 or whenever it was they adopted a position on 20 Diablo, did the Staff say, " Gee, it might well be worth our 21 while trying to gain access and maybe we can get it for free 22 since we are a government agency, and promising not to divulge 23 anything," and to have looked at this with experts, if 24 necessary?
\\
25 I am not aware that this was done.
36 1
MR. BROCOUM:
I am not trying to evade your
-n 2
question, but the Staff does make an attempt to go to all the y,j 3
major scientific meetings and does make an attempt to stay in 4
contact with the soientific community within the limits of our 5
travel budget.
We do try to get proprietary information when 6-we know about it, but having worked for an oil company, oil since they are competing with each 7
companies are very, very 8
other, are very, very secretive about their proprietary data, 9
and we were not aware of that data until it came out through 10 Crouch.
11 And had we heard about something, or had we been 12 aware, I think we would have made some effort to learn about
/
\\
13 it through our contacts.
I think we would have.
U 14 I think the other examples you brought up on the i
15 Meers fault in eastern Kentucky, when we first heard about 16 the Meers fault, we immediately asked Dr. Slemmons to go look 17 at it.
This was within weeks of learning about that fault.
18 He did make a several-day trip there.
19 We have since funded the Oklahoma Survey to study 20 it.
They, through our funding, through the Oklahoma Survey,
.t 21 we have funded the USGS to study the Meers fault.
We have had 22 an ongoing program probably of 10 years duration on Kentucky i
23 fault zones.
And there was a recent symposia on the Meers 24 fault, the Seismological Society of America.
About half the 25 people at the symposia were people being funded by the NRC.
i l
37 1
Those are examples, I think, where the NRC has, if
(
\\
2 you like, aggressively looked into these things.
But there is 3
a difference between aggressively looking into new i
4 information, of which we could be inundated, and having a 5
reaction that you have to shut down this plant or impose a 6
license condition.
7 I think in that case, in terms of having an action 8
on the plant, we are cautious.
But in terms of following up 9
new information as soon as we learn about it, we are cautious, 10 as we should be, in the first instance, I believe, or else we 11 would be cautiously reacting without knowing if there is any 12 merit to what we are reacting to.
13 But in the second instance, I think the Staff has
\\.
14 been assertive or aggressive in following up new information 15 about any part of the United States that we learn about that 16 has anything to do with nuclear power plants.
17 MR. OKRENT:
Let me make one more try at this.
18 MR. MC MULLEN:
My name is Richard McMullen.
19 Prior to knowing about the Crouch paper, we had 20 already begun preparing elements to go as a part of the 21 licensing condition under direction from our management.
We I
22 recognised there was a lot of oil exploration going on out 23 there, so we were preparing or had already been preparing I don't know if we knew the exact vehicle it 24 attaching a 25 was going to be made known, but we were going to ask PGSE to i
(
e
,_.m
38 1
take a look at the oil company data prior to knowing about the D) i
(_/
2 Crouch paper, because it had been nearly four years since the 3
Appeal Board hearing.
So this was already underway, under 4
direction from our management.
5 MR. OKRENT:
Had it gotten through CRGRt 6
MR. MC MULLEN:
No.
It was a memo from Jim Knight, 7
I believe, to our branch.
8 MR. OKRENT:
Under the new backfitting law, would 9
you have been allowed to ask that they undertake a substantial 10 study costing them perhaps millions of dollars with no way of 11 showing in advance of this some kind of a net benefit?
12 MR. MC MULLEN:
We probably would have had to go
[
\\
13 through that process.
14 MR. OKRENT:
Yes, but what answer would you have 15 gotten?
16 MR. BROCOUM:
Well, the way we handle something like 17 that is we do it through our technical assistance or through 18 Research, because that way we don't have to go through the 19 backfitting regulations.
20 MR. OKRENT:
So you would not ask the Licensee?
21 MR. BROCOUM:
Under the new backfitting, we will 22 have to go through CROR first.
That is not our decision.
23 MR. OKRENT:
Yes.
What some people have called the 24 black hole, 25 In any event, thank you.
That is interesting.
I am
39 1
not so optimistic as you, as to whether you could have m
)
2 achieved that purpose, had there not been a licensing 3
condition.
And at least my impression is the reason for this 4
license condition is probably not because the ACRS recommended 5
back in '78 that this thing be looked at in 10 years, but 6
because the ACES was asked to re-review this after a lot of 7
allegations and so forth and so on.
8 The Committee reminded the Commission that 10 years 9
ago, or whatever, eight years ago or whatever, six years ago, 10 the Committee had made such a recommendation, and for reasons 11 which only those in higher elevations understand, it was 12 decided that this would be a good licensing condition.
13 Let me get to the same kind of point via another 14 route.
15 Certainly some years ago there was very little data 16 on common mode failures.
I'm talking about seismio now.
17 Valves not starting, diesels not starting, diesels shutting 18 off, whatever.
So it was hard to estimate risks where those 19 kinds of things were important.
20 The Commission could have said, well, that's life.
21 Actually they invested a substantial amount of research money
\\
22 in trying to get better information on common mode failures 23 different ways.
24 The information is still not good, but it's better 25 than it was.
Okay.
In fact, it was done not because it was
40 1
known that there_was some particular plant for which it was
(%( )
2 important, it was recognized that it was a problem of general 3
interest for LWRs.
4 I am not sure the uncertainty in the rate of common 5
mode failure, after we had a little bit of data, was radically 6
different than our current uncertainty, or uncertainty eight 7
years ago s ia ou t just what was the situation at Diablo Canyon 8
with regard to seismology and geology.
Since in the case of 9
common mode failure components, redundant components, or 10 common cause failure, which in that event it was deemed 11 relevant to try to learn more, it seems to me there was reason 12 not to sit back on the licensing decision that the Staff
}
13 adopted in '77 or '78 for Diablo, but in fact recognizing 14 certainly by 1980 or '81 that there was data lying out there 15 that other people had gotten somehow, that it behooved one to 16 see how good that original judgment was.
And although you 17 mention, Mr. McMullen, some intent to do this, I am not aware 18 of anything that the Staff or the Licensee initiated prior to 19 the discussion of this long range plan.
20 Maybe the Licensee did some things on his own, !
21 don't know.
So I think it would be well -- we cannot undo the 22 past, but it may be well for the seismological and geological i.
23 bunch to ask itself sort of a general question, you know, what 24 should be our approach to taking advantage of possible new O
25 information, even though from time to time it may upset our
,y.-
__m_,._.,..,.-,c-.4 7--y-.
41 1
prior interpretation of what was okay at some site.
We have (O) 2 to buy that risk, because keeping up with the literature is 3
fine, I agree we need to do that, but I am talking about 4
something different, something that you initiate in sufficient 5
depth that it doesn't mean you do a broad program like PG&E is 6
going to do.
That's not what I'm talking about.
But is 7
sufficient to at least get a feel of, well, gee, could there 8
he something there, should we go on, et cetera.
9 Well, that is enough.
10 MR. SIESS:
Mr. Brocoum mentioned a little while 11 back that the long term seismic program plan was expected to 12 be sufficiently flexible to take care of surprises.
I think 13 that was the term he used.
I think of surprises in two 14 timeframes:
t 15 One of them is in the frame of the three years of 16 long-term seismio. program; but the other is surprises that may 17 occur at some time after the long-term seismic program,is 18 completed.
And surprises could be new geological evidence, 19 new geological hypotheses, or might simply be allegations.
20 To what extent do you think successful completion of 21 the long-term seismic program will provide a tool to handle 22 such surprises that may come during the life of the plant?
23 MR. BROCOUM:
I think it will provide an excellent 24 tool, and I think it is PG&E's intent, and I think they should 25 correct me if I'm wrong, that they will keep a level of 1
- _ ~ -...
~,_. -,_ _ -,,.-.-..._ -,_
~,.,-..._.-.
m,-_
-.. _ _. - ~..
s 1
v 42 4
t
(
1 expertise even beyond t h er three year program to handle new O
'ii s
V 2
information which will, of course, come along.
3 So we fully expect new-information and new theories 4
and new ideas on the California margin to be occurring for the 5
next 30 or 40 years, and we expect that PG&E will maintain j
6 their capability to evaluate that information for its impact s
i 7
on thear plants.
s 8,
That has, I think, been discussed between us and
'. M
\\
7 9/
them.
s 10 PR. OKRENT:
'You earlier mentioned that you thought s
i 11 it was well'to get a good spread of opinion so that some ides 12
- hy somewhat vague recollection of the subcommittee and Full 13 Committee meetings back in 1977 and '78 includes the fact that 3
14 there was a Prof. Brune from, I think, San Diego at the time,
',1f'I'remeub'er oor e c '. l y, who had some ideas that differed from 15 f
S s.
16 the bulk of the seismologists and geophysicists there.
t i
17 My impression is he certainly is a very intelligent s'
18 g e o ptijg i o l s),
or whatever he calls himself.
But he had some 19 thoughts whloh might not have been raised had he not been i
l 20 there, whatever their disposition.
i r
21 At the moment!!t is not clear to me that either i
1 22 within the expert group that is being assembled by the s
23 1,1 o e nis e ar.d Jh*y all, to the extent that I know them, look i
Q.14, [ ' l i k e and the group that the reputable people in the area l
y' s
+
25 \\
Staff is collecting, I can't tell whether you are in fact i
3 I
4 i
=
,r f
43 1
going to consciously.get the benefit, even if not as a s j/
2 consultant but as a participant in some workshop or whatever, 3
of what I will call the Prof. Brunes who are out there.
Okay?
4 MR. REITER:
A particular example of that -- yes, we 5
do have a great deal of respect for Prof. Brune.
And a lot of 6
the students have done very important work.
For instance, of the three-man panel on the 7
8 theoretical groundmotion model that NRC is putting together, 9
two of them are his students, former students of Dr. Brune.
it is hard for me to 10 And I don't know to what extent 11 believe that the utility, in their groundmotion modeling, 12 would not reach out and use some of Dr. Brune's students also 13 in some of their work, and so I think we try to be familiar 14 with all of those out there who are doing good work, and 15 certainly a lot of the work that Dr. Brune does and the 16 students do is excellent work.
well, let me just pursue it.
I am glad 18 to hear what you said.
I am in no way, as I said, trying to 19 belittle the names of the people that are on the panel, but l
20 some times a man like Prof. Brune is thinking some years ahead 21 of his students, because otherwise he doesn't have areas to 22 suggest for Ph.D.
research or whatever.
And I am sure not all
?
23 of my former students would have the same point of view on a i
I 24 specific problem as I do, but I used him as an example and I 25 don't know, you know, who others may be.
i l
. ~ _ -
44 1
I am providing a suggestion that, one, make a 2
deliberate effort to pursue what you said you were going to 3
do, and just introduce this one recollection from history, 4
okay?
S MR. BROCOUM:
Of course, as the panel is aware, we 6
do have the USGS, Bob Brown is their representative, and he 7
can call on resources from the East.
Not just Bob Brown, but 8
he has the resources of the USGS to call on for any 9
information they have and, of course, we have Dr. Slemmons.
10 I should point out on the very last viewgraph, we do 11 have a list of the people who will be on the advisory panels
- 12 and theoretical numerical modeling of groundmotion.
That 13 panel is being put together by Lawrence Livermore National t
14 Laboratory, and it consists of Kay Archy of the University of 15 Southern California, R.
J.
Archilete, University of 16 California-Santa Barbara, and S.
M.
Day of S-Cubed.
And the 17 second panel is being put together by Brookhaven National 18 Laboratory for soil structure interaction.
It consists of 19 C.
J.
Constantino of CCNY, G.
Gesettis of Rensselaer 20 Polytechnic Institute, A.
S.
Valetos of Rice University, and 21 R.
V.
Whitman of MIT.
22 These panels are not put together at the moment, 23 The contracts are presently being processed.
24
. Finally, I think we do intend to have technical
~
25
' assistance for.the PRA, but in what form and who and through
45 1
what laboratory or group that will be through at this time is 2
not known.
3 MR. SIESS:
Thank you.
4 Has the Licensee heard anything in that discussion 5
that they would like to comment on?
6 MR. CLUFF:
My name is Lloyd Cluff with PG&E.
If I 7
might just briefly respond to a couple of points that have 8
been made, the question about the dialogue and cooperation in 9
terms of exchanging information, I think in the past year, 10 prior to my coming with PGSE and since I have been there has 11 been a very good open dialogue, and I fully expect that this 12 process will continue and certainly increase as the program
/~'T 13 gets fully under way.
b 14 We are very pleased to see the quality of the people 15 from the NRC Staff, their interest, as well the people working 16 with Bob Brown at the USGS and Dr. Slemmons and the group that 17 he is working with to help us understand the significance of 18 some of the field relationships.
19 If I might comment on a few of the points that most 20 recently have been talked about here, the range of opinions or 21 interpretations that various people might have, as I mentioned 22 at our meeting in Los Angeles, we adopted a technique or a 23 tool to help us look and understand and keep track of the 24 diversity of. opinions and interpetations, and that is the 25 logic tree tool where it is a formal, graphic way of
46 1
representing various interpretations and making assessments of 2
the degree of belief or probability that one interpretation 3
might be more correct than another, but at least representing 4
the full spectrum of the opinions that have been published 5
that people have mentioned personally or that we might conjur 6
up ourself in playing some "What if?" games to take a look a 't 7
potential su prises and so forth.
8 It's a framework within which I believe we are using 9
right now to help us in the scoping effort in setting 10 priorities to address the issues.
The most important part of 11 this is there are two answers to the breadth of opinions that 12 might exist.
One is the basis for the opinion, and by setting
/
)
13 this up in a very systematic way, it allows us to look at the 14 weight of the evidence that allows one opinion to maybe be 15 favored over another and to reexamine new developments in the 16 evidence and so forth.
17 The other part of that, which I think is more 18 important, is: So what?
What difference does it make?
How 19 relevant is it to the solving of the seismic design of the 20 plant itself?
To me, that is where we are going to focus our 21 attention, to examine in generic or general way the range of 22 opinions, but we will focus on what difference does it really 23 make.
24 This logio tree that we are developing helps us 25 systematically think about that, ask questions and then focus
.y y -, _ - - - _ -.. _.
47 1
on what difference it really makes, and it will be a good way 2
of keeping a record, and as later information becomes 3
available or someone wants to hypothesize another idea or 4
whatever, we can reenter that structure and say, well, given 5
this new understanding, then, what difference does it really 6
make?
7 It might be in many cases that it's a matter of 8
academic discussion and has no real relevant bearing on the 9
issue that's at stake here, which is the adequacy of the 10 design of the structure.
11 MR. EBERSOLE:
I'm glad to hear you say that.
I can and I think I have sat for many, many hours 12 recall 13 listening to academic discussions about matters that I was 14 personally sure had nothing to do with the fundamental issue, 15 so this leads to some efficiency that I haven't yet seen in 16 this process.
17 MR. CLUFF:
But let me qualify that with sometimes 18 there are some rather esoterio and academic exercises that are 19 useful to understand the behavior of a fault that may be at 20 some distance that has no direct relevance to the plant itself 21 but may give us some insight into understanding a fault 22 that's closer.
You may be able to look and understand the 23 behavior of a distant fault and apply that knowledge to a 24 fault that is closer and say, given their interrelationships, O
25-we have a better understanding of that fault there.
So we
1 48 1
intend to carry out some of those studies to help us with
~w
(,)
2 greater insight.
3 I might point out also, having recently joined PG&E 4
and am still learning about the history and the past 5
development of the geologic, seismic and earthquake 6
engineering aspects of this program, that Crouch is purported 7
to have come up with this so-called radical idea, 8
interpretation or whatever, and that is really not true.
9 In my going back and looking at the work that has 10 been done by PG&E and the USGS and others, the concept of a 11 thrust or a reverse slip component on the Hosgri Fault or 12 other related faults is not a new idea; it has been in the (d~)
13 literature, the oil companies have considered it, 14 Dr. Hamilton, who has been a consultant for PG&E for many 15 years argued in some of his testimony about the importance of on the Hosgri Fault.
16 a significant thrust compcynent 17 So I think the more radical interpretation or idea 18 that comes out of Crouch's paper 1s that all of the faults are 19 dominantly thrust, low angle thrust.
They may be tied to a 20 very shallowly dipping decollement at some depth.
That is 21 kind of a popular idea that is going around in tectonics 22 throughout the world.
23 What it really comes down to is as we are going 24 through the scoping study looking at the issues, it may turn
-s 25 out that it doesn't matter, so we are taking a look at that
49 1
and really assessing what significance is there to a pure b)
(,
2 strike / slip or a combination of a component of dip / slip, 3
strike / slip on whatever fault we are worried about, and then 4
what difference that really makes in terms of the strong 5
groundmotion as it is attenuated from the source to the plant, 6
and what we are finding in going through some of this, that in 7
a lot of cases it really doesn't matter.
8 What reatly matters is when you look at the 9
topography of the region that is there because of the 10 tectonics and the active tectonism that is going on is that if 11 there is a significant long-term thrust component on some of 12 those faults, and given the slip rates that are being (m
13 purported to some of these faults, if it is primarily reverse 14 slip on these faults, one would see tremendcus mountain ranges 15 in that area that are not there.
16 So my view is right now the real answer will come 17 out that there is probably a significant combination of both 18 strike / slip and dip / slip, and I'm not so sure it will make 19 that much difference when it comes down to the final analysis.
20 In terms of keeping track of ongoing work, whether 21 it be the published literature, a lot of times we find that it-22 takes a number of years after someone gets an idea and does 23 some research before they finally get it in print and we have 24 access to it.
We have a program that we are developing and we gs k
25 have a list of some 15 or 20 different agencies or
50 1
universities or whatever that we know are conducting work in 2
that area, including people from Europe and so forth.
f 3
Imperial College has been doing work in that area.
4 It is our intention to keep track of not what they 5
are publishing but to actively have some dialogue with these 6
researchers so that we at least are alerted to new 7
interpretations.
And then we plug that information or these 8
interpretations into our logic tree to see what significance 9
they have. And while one can't guarantee that one eliminates 10 surprises, at least one has a way of managing surprises so it 11 doesn't become a panic situation.
12 That is our intent, to keep track of this new I /' )
13 information that develops.
And while we have talked about the 14 three-year program, it is our intent to get to the heart of 15 all of this and come to confident conclusions at the end of 16 this program in three years.
Certainly with the information 17 that we will have developed, we will have the knowledge and 18 the ability to incorporate as time goes on at any time in the 19 future new information or new ideas as they come along and 20 will be able to assess what relevance they have to the design 21 of the plant.
22 I guess that is all the points that I have right 23 now, and those that I have taken as I have listened to the 24 dicussions.
25 MR. SIESS:
Thank you.
51 1
Any questions for Mr. Cluff at this point?
-q 2
[No response.]
3 MR. SIESS:
Gentlemen, I am going to declare a 4
break, but when we come back, I think we could concentrate on 5
the issue that was rais*G not by the Staff but by the 6
Subcommittee, a s1J that had to do with the level of the PRA, 7
and we will probably devote the rest of the time we need to 8
that subject after the break.
Be back about 10:15, please.
9
[ Recess 3 10 MR. SIESS:
The meeting will reconvene.
11 Is there any objection to proceeding now to 12 discussion of the PRA levels?
D 13 Dr. Okrent raised this issue, so I'm going to let-14 him take the lead on it.
15 We have had a response from the-Licensee and a Sta:f 16 response to the Licensee response -- no, I guess the Staff 17 response is to the ACRS.
18 MR. BROCOUM:
Yes.
19 MR. ISRAEL:
Sandy Israel, with the Staff.
20 In our previous deliberations with the Licensee, we 21 indicated to them that we had found acceptable all they had to 22 provide us was a Level 1 PRA, and I guess unfortunately that 23 has been misconstrued by a number of people to believe that 24 the Staff has now abandoned risk, and that's not really the O
25 case. Let me go into that a little bit.
52 1
MR. OKRENT:
Excuse me, but before you start, would 2
you define what you mean by a Level 1 PRA, what you mean by a 3
Level 2 PRA?
I'm not sure everyone --
4 MR.
ISRAEL:
I will go with Level 1,
and everything 5
after that is the back end anyway.
Level 1 in my terminology 6
would be one that would give me core melt frequencies for 7
various plant damage states.
Plant damage states refer to the 8
condition of the containment with respect to its cooling or 9
being bypassed, so I can have core melt without containment 10 cooling, I can have core melt early or late, I can have core 11 melt with containment cooling.
12 I can also have those other r.ituations where the 13 containment is bypassed prior to core melt, such as Event V,
14 or such as are shown in the Indian Point study where seismic 15 took out the containment all together and that then initiated
-16 the core melt from there.
17 So that it's really a systems condition on the 18 plant, starting from initiating situation.
It does not refer 19 to phenomenological effects that would necessarily go on after 20 core melt, such as overpressure failure or hydrogen burns or 21 missiles, the alpha-type things.
22 MR. OKRENT:
What woula Level 2 be?
23 MR. SIESS:
Wait, still on Level 1.
You say 24 containment failure. I think what you are saying is a breach 25 of containment due to pressure or tsaperature.
53 1
MR.
ISRAEL:
No.
What I'm talking about in terms of (sv) 2 Level 1,
we are talking about containment conditions, core 3
melts -- I can have it with containment cooling or without 4
containment cooling, and then subsequent failure of the 5
containment would depend upon phenomenological situations 6
inside the containment downstream.
7 MR. OKRENT:
So that's not in Level 1.
8 MR.
ISRAEL:
That's correct.
9 MR. SIESS:
And if the containment already has an 10 opening in it, that's not a Level 1.
11 MR.
ISRAEL:
If the containment already has an 12 opening in it, such as the failure to isolate purge valves or
[%
13 that sort of thing, that would be in Level 1.
If the
\\
14 containment failed such as a seismic event where the 15 containment tilted and broke away from the auxiliary building 16 such that you failed the containment, you also have lost 17 cooling to the reactor.
You would have containment failure 18 essentially prior to core melt, and that would b e' in Level 1.
19 MR. SIESS:
But anything that happens after core 20 melt is not in Level 1 except Level 1 does tell yo@ what is 21 the status.
22 MR.
ISRAEL:
That's correct.
23 MR. SIESS:
Okay, I understand that.
24 MR. OKRENT:
And what is Level 2?
25 MR.
ISRAEL:
Level 2,
I guess, would be the
54 l
1 containment analysis, so it would give you the split fractions
(~%
(,,
2 for the various containment failure modes, whether it be 3
hydrogen burn or slow overpressure or fast overpressure, 4
. basemat meltthrough, no failure at all, that sort of thing.
5 And out of Level 2 you would also get the source terms 6
associated with each one of those failure modes, and then 7
Level 3 would be the consequences offsite.
8 MR. SIESS:
Would Level 1 give you the source term 9
for what is in the containment at that time?
10 MR. ISRAEL:
No, it would not, but let me address
- 11 that.
We have gotten ahead of my talk.
12 There are several practical reasons why we think i
13 that stopping at the Level 1 is appropriate.
One of these i s 14 that 'we have alternate means of estimating the offsite 15 consequences.
At present, if you consider the old source term 16 work, a great deal of effort went into the siting study by 1.-
17 Sandia, a strip report. I don't have the NUREG numbers with-18 me, but for each one of the sites, they have indicated what 19 the potential early fatalities would be, latent fatalities, i
20
- person rem as a function of probabilities.
This could be a 21 source of the Staff making quick estimations of what the 22 offsite consequences would be given the plant damage states 23 that would be obtained in the Diablo Canyon study.
i i
-s 24 MR. OKRENT:
Excuse me. I am missing something in V
l 25 your logic.
Please help me.
w w-
,,p-w
-w
,-w,-
m4
-g. - - -
,m
--r,------------+-----,--m
,-w, e
m w
-v--
w-
55 a
1 From a Level 1,
you don't get the release from the f\\
\\ j/
2 containment.
3 MR. ISRAEL:
That's right.
4 MR. OKRENT:
If it were stripped, you would have to 9
5 have releases from the containment.
What is this thing that 6
Staff has referred to that it lets it quickly extrapolate?
7 MR.
ISRAEL:
The Staff has had a great deal of 8
experience now with large dry containments.
It has been 9
through a large effort on Indian Point and Zion and through an 6
10 effort on Millstone, and the results of these efforts have 11 indicated that in terms of offsite consequences -- and we will 12 talk about early fatalities first -- that those are dominated 13 by containment failure prior to core melt, such as the Event 14 V or potentially by the seismic upset of the containment.
15 The source term for that would be the highest source 16 term, the SST-1 that is in the strip report, and the 17 conditional containment failure for that is 1 for that type 18 of event.
So it is a way of estimating early fatalities right j
19 off the bat without much difficulty.
20 In terms of latent fatalities, those based on work 21 we have done before are dominated by long-term overpressure 22 failures of the containment, and the Staff has now had 23 experience with large dries and is able to estimate within a 24 reasonable success of what the split fraction would be, and 7-s 25 they have ways of estimating the source term based on those f
.--,,-n n
56 1
source terms that the strip report is based on.
'3
)
2 So, using prior experience in the strip report, one 3
could then estimate in a very rapid manner what the early 4
fatalities could be and the latent and person rom.
5 MR. OKRENT:
Just one minute while I find certain 6
documents.
7
[ Pause 3 8
I have a memo from Mr. Denton to Mr. Dircks dated 9
June 12th.
The subject is the safety goal evaluation work.
I 10 will quote from it, the second paragraph on the first page.
11 "As highlighted by our recent answer to a question from the 12 Congress, the probability of a core melt accident, even if the 13 proposed guideline were met, is high.
The proposed guideline 14 is 10 to the minus 4 per reactor year for a large-scale core 15 melt.
16 "If just met in the median by all plant systems, it 17 would result in an approximate 50 percent probability of one 18 or more serious reactor accidents in the next 20 years in a 19 population of 100 plants.
Also, there would be about a 10 l
20 percent probability of two or more such accidents.
21
" Calculations of public risk from these accidents 22 rely heavily on the knowledge of fission product behavior and 23 containment performance beyond conditions for which mitigation l
24 systems and structures are designed or tested.
25
" Computer simulations and limited small-scale test l
?
l I
i
57 1
data need to be applied with caution.
~
k/
2 "The accident prevention guideline is a quantitative 3
corollary of our defense in depth concept and should include a 4
margin to allow for the imperfection in the current methods of 5
predicting core melt, fission product behavior and containment 6
performance."
7 I would suggest Mr. Denton does not feel that there 8
is a straightforward path going from the prediction of core 9
melt frequency to the estimation of release categories; would 10 you?
11 MR.
ISRAEL:
No, but that is quite right.
You have 12 now talked about overall core melt frequency.
We are now
/
13 talking about plant damage state, which is the decomposition 14 of the overall core melt into various situations, and some of 15 those situations are much more apt to give you significant 16 offiste releases than other situations.
17 MR. OKRENT:
Excuse me.
You think Mr. Denton is 18 unaware of plant damage states when he writes this?
19 MR. ISRAEL:
I can't comment on what Mr. Denton 20 wrote and what was in his mind.
21 MR. OKRENT:
Did the plant damage states tell you 22 just how to decide what is the correct description of the 23 phenomena that go on given a core melt?
24 MR. ISRAEL:
No, it does not, and that is probably 25 one of the problems here, is that the containment analysis is
58 1
highly sensitive to certain assumptions that go on in terms of O
3 J
2
.the-phenomenological phenomena, and this has been a source of 3
investigation for the severe accident program and the source 4
term work.
5 That 2ets to my second comment or reason for 6
stopping at Level 1,
is that a great deal of this knowledge is f
7 now being' generated currently in the source term area and the 8
severe accident area.
In fact, there will be a report out 9
shortly, NUREG 0956, dealing with source terms, which will 10 talk to Surrey, at least, in terms of what the new source 11 terms mean in terms of consequences at Surrey, and in about a 12 year, the Staff will be putting out another report, NUREG 1-3 1150, that will deal with about five surrogate plants dealing 14 with offsite releases from typical plants.
15 This is exactly the area or exactly the forum that 16 probably should be going into phenomenological considerations 17 inside the containment.
It doesn't quite seem appropriate to i
18 burden Diablo Canyon and make that a test bed for development 19 work in the severe accident containment area, and I think that
)
20 is one-of the reasons we are willing to stop there and 1
21 estimate.
The Staff is able to estimate, at least right now, 22 using old source term work, using the strip report, and in the 23 future, based on this new work that should be coming out 24 within the next year or two, incorporate what the new source O
25 term work may mean.
59 1
MR. OKRENT:
Well, you were the one who first said
(
2 you thought you could estimate from the plant damage states, 3
and what I tried to read was something that questioned whether 4
in fact you had'such straightforward an ability as you 5
suggested you had.
6 Now I think you are conceding that there are some l
7 complexities as to what is going on in containment.
In fact, 8
I think there are some complexities, and in fact, not all of 4
9 the possible failure modes have been given-equal and I
10 sufficient emphasis up to date.
That is my own personal l
11 opinion.
12 However, I see no basis at all for your statement 13
. that you would be burdening Diablo Canyon with doing something 14 that you suggested would somehow go far beyond what was done
'15 for Indian Point 2 and 3,
that ~was done for Zion, that has 16 been done for several other recent plants.
Hopefully, by the 17 time they were doing such a thing for Diablo Canyon, that 18 would be at least two years hence they would have the benefit 19 of this elegant or expensive Staff research program.
20 MR. SIESS:
Did you say extensive or expensive?
I 21 MR. OKRENT:
Both. And the industry program. And
.22 maybe they could do things that were better than were done for 23 Zion and Indian Point, et cetera.
But just going from a Level 24 1 to a Level 2 does not say go ahead and put in a $100 25 research program, to from Level 1 to Level 2.
~~-
-,w
-n, n-.
a
--n,,.e.,-
60 1
MR.
ISRAEL:
Well, but I think the analogy isn't
/~
(,T) 2 quite apt, Dr. Okrent.
The Indian Point and Zion work were 3
risk motivated, and they were containment oriented.
This 4
particular action with Diablo Canyon is not there; it is in 5
the seismic area.
6 I want to give you an example of the effort.
I have 7
12 volumes of the Indian Point PRA in my office. Three volumes 8
deal with the internal event core melt analysis.
One volume fire, seismic, whatever.
One 9
deals with external events 10 volume dealt with a summary of the results, and seven volumes 11 dealt with the containment analysis and consequences analysis.
12 That seems like totally the wrong emphasis for this 13 particular action here, which is a seismic action and not a f}
\\J 14 risk action.
There was a different forum there and therefore 15 you had a different emphasis.
16 What we are saying is we have an alternative for 17 estimating the offsite consequences.
That seems quite 18 reasonable and adequate at that time.
19 MR. OKRENT:
Let me go back to what you just said.
20 In the first place, I heard the representative from PG&E say 21 he wanted to look at various pieces of seismic and geologic 22 information in terms of what is their impact on the overall 23 answer, meaning the risk.
And if something was interesting if when you 24 and even if there was another possible model O
25 did it you ended up with about the same amount of shaking, it
61 4
1 might not be that importani to pin it down between one and the 4
p
()
2 other.
3 For example, why?
Not because the shaking was the 4
same but because the risk to the plant would be different.
If 5
you could show that, then you proceed accordingly.
So this 6
is, in fact, what we are interested in, not just what is the' 7
change in seismology and geology at the plant.
8 MR.
ISRAEL:
But he gives me that when he gives me 9
the plant damage state.
He is shaking --
10 MR. OKRENT:
He does not give you the risk when he 11 gives you the plant damage state, and I only have to pull out 12 two papers that pointedly emphasized that the thing you have 13 to be careful to look at is not core melt frequency but --
14 MR.
ISRAEL:
And the Staff is on the hook for doing 15 that.
As a matter of fact, the way this whole thing will play 16 out is when they finish with their work, they will come to the 17 Staff and say, hey, we think they ought to modify XYZ, or they 18 may come to us and say, hey, we want to sit pat, we don't 19 think we have to do anything to the plant.
Then the action is 20 with the Staff to review whatever they have submitted and make 21 a finding whether additional modifications have to be made.
22 And then making that finding, you have to go through the 23 backfit, and Staff is on the hook for doing offsite 24 consequence analysis as part of that backfit.
O 25 MR. OKRENT:
I don't understand why -- I am at a l
l i
u 62
+
1
- loss,
- a. complete loss to understand why you think the 2
applicant should try to identify the release categories and 3
the frequency thereof.
4 MR. ISRAEL:
We are not requiring it.
He can do it 5
if he'wants to.
We are not requiring it.
6 MR. OKRENT:
But why do you think it is not relevant 7
to the evaluation that the Commission thinks is relevant?
8 MR.
ISRAEL:
Because these calculations are very 9
expensive.
We have spent over $500,000 just reviewing the 10 Indian Point back end analysis.
We are talking about big 11 dollars.
'12 MR. OKRENT:
How much does this total package cost?
4 13 MR.
ISRAEL:
I don't know what the total. package is, 14 but I know the back end analysis is very time consuming.
15 MR. OKRENT:
Well, I think it is relevant.
How much 16 is this package costing the Licensee?
How much is the total 17 package costing the Licensee?
Minimum number for this 18 long-term evaluation.
19 MR. CLUFF:
Oh, for the total program?
Well, it is 20 less than $10 million right now. It depends on what we come 21 out of the scoping and what we do.
22 MR. OKRENT:
I would suggest that if they had to 23 spend $500,000 --
24 MR. SIESS:
That was for the Staff review.
O 25 MR. OKRENT:
No, no, no.
I'm sorry.
1 63 1
Look, PLG, who happens to be their contracter, has
)
2 been through a few Iarge dry containments.
It has been l
3 through a few large Westinghouse dry containments.
This plant i
4 is not identical, but nevertheless, it doesn't enter the l
5 picture having to ask for money to learn how to begin.
They I
6 want to think of what it would do differently or better, and 7
'i t certainly has to allow for differences between Indian 8
Point and Zion and some of the in-plant protective features 9
and so forth, 10 But it is not, I think -- if i t-is a major 11 undertaking for PLG to go from plant damage states to a Level 12 2,
then there must be something important that you need to
[\\
13 understand and to do, in my opinion, because --
14 MR.
ISRAEL:
No.
Let me hypothesize there that what 15 happens is that you run into long-running computer programs 16 that chew up money, and these computer programs can tell you 17 what the containment loading is, and it still boils down to 18 coming up with split fractions.
And the things that get you 19 are not these split fractions with containment cooling; the 20 thing that gets you is probably going to be they wape out the 21 containment as part of the seismic event directly.
22 You know, I have the source term for that and you 23 already have the split fractions, one.
You have that type of 24 experience already.
There is no need to start reanalyzing.
O 25 When you get into the back end analysis, you start a
4 64 1-reanalyzing cases that suddenly are not really important.
,o 2
MR. SIESS:
Excuse me, Dave.
We asked the Licensee l
~3 how much the long-term seismic program is envisaged to cost, 4
and he gave us an answer.
Have you got any idea of how much 5
the additional cost would be to go to Level 2 on the PRA and i
6
_ whether that would require additional time over the three-year 7
time limit you have?
8 MR. BRAND:
My name is Donald Brand, with the 9
Utility.
10 We have looked at the relative cost and cost-benefit 4
11 of going beyond Level 1 and going to Levels 2 and 3.
Our 12 current estimate is an. additional expenditure of $1 million to 4
13 82 million for going all the way to Level 3,
and in our 14 evaluat' ion we thought it was more prudent for us to be looking i
15 at seismic and nonseismic initiators in terms of full scope of 4
16 activity at Level 1 than it was in moving beyond that Level 1 17
.- p l a t e a u.
18 That is not to say that we are insensitive or that 19 we are not interested in risk per se.
It was simply our 20 evaluation, lookingunt the license condition, that Level 1 21 satisfied that requirement.
22 MR. OKRENT:
Did you say or can you say what it i
23 would cost to go to Level 27 h
24 MR. BRAND:
I would offer probably a third of the 2
25 number I quoted perhaps have, but I would be thinking probably i.
i'
- v... - -..
.m_,...,.
65 1
a third would be a better number.
-g 2
MR. OKRENT:
Yes, I would think so, too.
Have you 3
ever done an evaluation of the failure pressure and 4
temperature and point or points under containment assuming a 5
slow increase in pressure?
Has that calculation been done?
6 MR. BRAND:
We did some earlier work several years 7
ago.
Let me ~ ask our staff.
8 MR. OKRENT:
Remind me; what kind of containment do 9
you have?
10 MR. SIESS:
Reinforced concrete with diagonal 11 rebar. Probably the strongest containment 12 MR. OKRENT:
And a liner inside?
13 MR. SIESS:
If it fails due to seismic load, I am 14 going to be very much surprised.
15 MR. OKRENT:
Me, too.
16 MR. SIESS:
That one I think we could rule out, but 17 that's beside the point.
18 MR. FRAY:
We have done the normal FSAR licensing 19 analyses looking at the containment.
Last fall there was some 20 scoping work on the containment, the loads in it due to 21 pressure buildup.
I think you could better describe the work 22 that was done on the containment.
23 MR. SIESS:
Could you tell us what the bottom line 24 was?
At what pressure would you expect the containment to no
-s
%/
25 longer be able to contain leakage?
That is, that the liner
66 1
would fail or penetration would fail.
,m i
b g
j 2
MR. WHITE:
My name is Bill White.
3 We have done some preliminary calculations which 4
we haven't had an opportunity to bring out all the aspects 5
that'we would like to, but preliminary indications are that 6
the containment can withstand greater than 100 psi.
7 MR. SIESS:
And design pressure is what, 8
50-something?
9 MR. WHITE:
Fifty-something, or 47, 10 MR. BRAND:
I remember 47 being the number.
11 MR. SIESS:
Well, 100 is a pretty good number.
12 Some people presented 100 and 42 and so forth. It's not an 13 uninteresting number in trying to go from accident states to
,(
14 releases, and I don't know how the Staff, in fact, would be 15 able to go very simply if they didn't have some reasonably 16 reliable estimate of the containment capability. That is an 17 example, it seems to me, cf information that one would get as 18 part of going to Level 2.
19 MR. SIESS:
That's comparable to Zion, as I recall.
20 Zion was in the neighborhood of 100 to 110.
21 MR. OKRENT:
Well, they showed a higher number. I 22 don't know what their real number is, but they showed more 23 like 140 for Zion.
i 24 MR. SIESS:
You know, I hate to sound like a lawyer l
f and if there are any lawyers present, they can accept that i
25 I
67 l
l 1
2 ELaughter3
,j
.\\
3 The license condition says that the program is to 4
reevaluate the seismic design bases, and of course, a strict 5
interpretation of that, which I don't think anybody is making, 6
is just to reevaluate that sero period ground acceleration and 7
the shape of the input spectrum, together with soil / structure 8
interaction type of stuff.
9 The last item of the detailed items says assess the 10 significance of the conclusions drawn from the seismic 11 reevaluation.
This is really what we are talking about.
How 12 do you ~ assess the significance?
Do you interpret significance 13 strictly in terms of core melt probability, which really is of 14 concern to the owner of a plant because it is going to cost
-15 him a Ict of money no matter what the next step is, or do you 16 assess the significance in terms of the health and safety of 17 the public?
18 Now, I'm not sure that has ever been debated. This 19 language is not as specific as it could be I know the ACRS 20 recommendation, the ten-year review was i r-the context of 21 seismic design bases, although those basen don't mean anything 22 unless you are concerned about the health and safety of the 23 public, which is what we are supposed to 9e concerned with.
24 Now, in your response to the i t.e m s the Staff asked 25 you to respond to, the last paragraph c on:mi t s PG&E to
68 1
.considering the merits and cost-effectiveness of extending the
. g)
(
2 PRA to Level 2 and/or 3 and evaluating that after they have 3
completed the long-term program and done the PRA to Level 1,
4 and you said this evaluation and possible extension will be 5
performed outside of the long-term seismic program.
6 That was one reason I asked earlier not just about 7
the cost of going to Level 2 but whether that is something 8
that would add significantly to the time required to complete 9
the program where you have got a three-year limit.
I assume 10 it has to be done sequentially, although some parts of it 11 could be done in parallel.
12 MR. BRAND:
Our comments in that paragraph were not
- 1. 3 intended to be an express commitment to do a Level 2 and Level
(^}
/
t 14 3 but only that that consideration would be made in the 15 future, and as pointedly outside the long-term seismic 16 program.
17 I have been advised by our staff that we could 18 probably perform the Level 2 and the Level 3 activity within 19 the three-year time frame that we currently have, although we 20 do not have an excess of time available to do that.
In other 21 words, it will fit but it is tight.
22 MR. OKRENT:
And if you did only the Level 2,
it 23 would fit more easily.
24 MR. BRAND:
Certainly.
I N
25 MR. SIESS:
What is the difference between Level 2
69 1
and Level 3?
(O) 2 MR.
ISRAEL:
Level 2 would give you source terms for 3
the various release categories and split fractions.
Split 4
fractions are the probabilities that the containment will fail 5
because of long-term overpressure or early hydrogen burn or 6
whatever, basemat melt-through, et cetera.
7 MR. SIESS:
And Level 3 would do 8
MR. ISRAEL:
Consequences.
9 MR. SIESS:
Level 2 involves source terms. If you 10 did a Level 2,
would it come near the end of this thing where 11 presumably we would know more about source terms than we do 12 now?
-s) 13 MR. BRAND:
Yes, although based on what I heard 14 Mr.
Israel mention as his time schedule, I don't know that the 15 Staff work will be out in time for us to take full advantage 16 of it and complete
,u r work within the three-year time frame 17 we have imposed on this program.
It is possible, but again, 18 not assured.
19 MR. OKRENT:
Well, you can always do source terms 20 parametrically.
The Staff does it, and they set an example.
21 You know, they choose larger numbers and smaller numbers and 22 say if it's possible to get results.
23 MR.
ISRAEL:
But the new source term work becomes 24 very important, especially for the long-term 25 overpressurization failure, which was a major contributor to J
,p
-h 70 7 '-
w 1
latent fatakities, cancer fatalities and person rem.
CQ
/
t
-s s
the' plate-out within the reactor and even within the 1(v)+
2 Evidently g
3 contIinment may be higher than what the old WASH-1400 work
~'
l l
(
4 was.
see, we are now caught right in the
'5 So that 6
middle h4re.
I have old source term work which I can use to estim$te of what the offsite consequences are, and 7
make an 8
potentially, maybe. in about two years, I may have generic 9
Creports that would gsve me information for handling new source 10
. term.
I don't know.
f
~
11 MR. SIESS:
Well, going back to the objective of a 12 license condition, suppose you go through this whole project 13 and you end up t h a t' the Hosgri spectrum, the Hosgri design 14 basis a n'd the other t!ings that were done, the tower effect,
(
I hate to use the term -- a 15 et cetera, it turns out to be 16 reasonable value,4which I assume has to be expressed i
17 probabilistically some way, that it's two sigmas or something.
4 18 What now do you learn?
Do you go beyond that and sq 19 look at the' plant state?
Do you have to do a PRA now?
You 20 haev confirmed the design. basis.
MR. BRAND:
Presuming we simply confirm the design 21
^
22
' basis, we would not at that time have to embark upon a PRA per 23 se.
At the same time, understanding the dynamics and the
%+
24 interactions between the various components of the program, we
)
25 are proposing embarking on a PRA presently rather than waiting
~
,i
+
+
71 1
the year and a half or two years before making that r~N i. ()
2 determination.
3 MR. SIESS:
Now, suppose you ended up with a 4
significantly larger design basis earthquake, seismic design 5
basis, and you did the PRA and your core melt frequencies by
{
i 6
some measure, because there are lots of different core melts'
)
7 and which ones lead to consequences, were still within the 8
lower range of the quantitative safety goal of 10 to the minu.
9 5,
say, or below 3 x 10 to the minus 5,
which is a number that 10 has been kicked around?
How would you feel at that point?
11 Would you feel that the significance was not significant?
12 MR. BRAND:
Well, we are going to be making both 4
13 deterministic as well as probabilistic assessments of this. I 4
14 would not want to simply prejudge what we would be wishing to 15 do in this regard.
At the same time, one of the reasons we 16 are doing the probabilistic work is to put into perspective 4
17 the relative benefit, shall I say, of a particular design i
~
18 base, and should that be slightly higher than our existing one 19 for any particular reason, then we would have to look at the 20 relative merits of that.
21 I would not want to speculate too much at this time 22 about that.
I would say, though, that as we are entering into 23 this, we have got, I think, a good confidence that our current a
24 design basis is quite conservative and we are not expecting 25 per se major increases in the design base to come out of the
i o
72 1
program two to three years from now.
(O) 2 At the same time, I would not want to say we are 3-ruling out that possibility, and we are embarking on this 4
program with that consideration in mind as well.
5 MR. SIESS:
Has the Staff given any thought to how 6
they would' interpret the words " assess the significance of the 7
conclusions," whether they are going to assess them at the 8
level of the seismic design basis, at the level of core melt 9
or at the level of consequences?
10 MR.
ISRAEL.
I don't think we have worked out a 11 decision logio yet.
Let me give you some perspective, though, 12 that may help, and let me t a'1 k to the proposed safety goals, 13 which are just proposed and not requirements, but let me just 14 put them on-the table.
15 Yesterday I quickly went back and looked at the 16 Indian Point estimates, the individual risk numbers for 17 Indian Point, and for early fatalities, you would exceed --
18 and these are rough, so I could be off by factors of 2 or 3,
19 but I'm going to throw them out anyway.
You would exceed the 20 early fatality value for core melts that fail the containment 21 first.
i-1 l
~22 If I had a seismic event that took out the l
23 containment, tilted it away from the auxiliary building, et 24 cetera, took out the containment building first, if the 25 frequency of that type of sequence were around 2 or 3 x 10 to
73 1'
the minus 6,
that would be about the early fatality value on j%
}
)
2 the safety goal, one-tenth of one percent individual risk.
s.-
)
3 The other sequence that could potentially add to 4
that is an extended station blackout, and you get this to a 5
seismic event by wiping out obviously the offsite power and 6
failing the diesels one way or another so you have extended 7
blackout, and you would have to have frequency for this type 8
event of up to around 5 x 10 to the minus 4.
9 MR. SIESS:
This is at the Zion site?
10 MR.
ISRAEL:
This is individual, and it would happen 11 to be at Indian Point. These are sort of rough numbers.
They 12 will vary from site to site.
13 Now, it turns out that the
}
J 14 MR. SIESS:
You are talking about populated sites 15 versus Diablo Canyon, 16 MR.
ISRAEL:
No, these are individual risk values.
17 MR. SIESS:
Okay, these are individuals.
18 MR.
ISRAEL:
Now, what would affect the individual 19 is probably aside from whether because it rains right after 20 the event.
Obviously that could affect this situation, but 21 also it is probably the distribution.
The closer the people 22 are to the plant, the more likely, the higher dose levels.
23 It's sort of interesting.
My recollection of going 24 to the Diablo Canyon plant, I had to drive about seven or ten 25 miles, I recollect, from the gate to the plant, and being at
74 1
Indian Point, Zion and Limerick, less than a mile between (j
2 population.
So there may be a bias.
Indian Point may have 3
essentially higher potential individual values than would 4
Diablo Canyon because of the distribution.
5 You also brought up population, which is another 6
aspect of the safety goal, and the safety goal says look at 7
if you violate whatever these targets are, go to a 8
cost-benefit and look at the averted monetized risks to the 9
public. And it turns out that at 50 miles, Diablo Canyon has 10 roughly 200,000 people.
The average site in the United States 11 has about a million people.
Indian Point had about 17 million 12 people.
So in terms of societal risk, Diablo Canyon is well
(
13 below Indian Point, roughly one percent of whatever Indian C/
14 Point was, and below the average site.
15 So in terms of looking at offsite consequences, I 16 tried to give you a judgment as to those types of plant damage 17 states that could lead to early fatalities that would give you 18 about the safety goal and also give you a judgment in terms of 19 person rem, which somehow affects our monetization of offsite 20 consequences.
This plant doesn't have very much population.
21 MR. EBERSOLE:
Sandy, you mentioned either 22 seismically-induced or otherwise induced long-term AC power 23 failure.
Originally, you remember, that problem was due to a 24 depreciation of station batteries, and thus the failure of the s
25 aux feed systems, and then it was compounded when it was 2
75 1
learned that the Westinghouse systems in particular had seal I
(
2 failures.
3 Which of these phenomena -
4 MR.
ISRAEL:
Well, it happened to be the seal 5
failure. That was the worst situation.
E MR. EBERSOLE:
Is it contemplated as you know now 7
that anything will be done about seal failure as it is being 8
done in Europe?
9 MR. ISRAEL:
I will have to beg off.
I know that 10 Westinghouse Owners Group is working on the problem, but where 11 they stand, I really don't know. I would not want to 12 conjecture.
I presume that once they come up with a fix 13 MR. EBERSOLE:
What was the relative time difference 14 between the onset of damage with the seal failure versus the 15 battery drawdown?
Do you remember?
16 MR.
ISRAEL:
Okay.
I think an end point was 17 probably about an hour or two, as I recollect.
We talked 18 about core melt in about an hour or two.
19 MR. EBERSOLE:
For the seal failure.
But for the 20 battery it was many hours.
21 MR.
ISRAEL:
I would say about five hours.
But the 22 problem is when does containment failure occur, because that 23 is when the offsita dose occurs, and the most recent numbers I not for Indian Point, for another plant -- they 24 have seen 25 said, well, we get up to about 18 or 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br />.
76 i
Now the problem is that with the new source term (O) 2
' work, it may be that even if the containment fails in 18 to 20 3
hours, things may have precipitated out into the containment l
4 or in the primary system.
I'm not familiar, that's not my 5
area of expertise, but it could have a significant effect.
6 MR. EBERSOLE:
You could stand a nonrejection of 7
heat from the containment for 18 to 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br />?
8 MR. ISRAEL:
It could be.
9 MR. EBERSOLE:
Just by storage.
10 MR.
ISRAEL:
On the other hand, t! s source term work 11 may be uncovering other types of containment failure modes 12 which are generic again, and when that is put on the table, 13 certainly that would then influence how we would come up with
's 14 our estimate, the Staff estimate of the offsite consequences.
15 MR. EBERSOLE:
Does this containment failure include 16 the particular failures of penetrations due to high 17 temperature rather than just overpressurization?
18 MR.
ISRAEL:
No.
None of the work so far has done a 19 very careful study of --
20 MR. SIESS:
That is still research.
21 MR.
ISRAEL:
That's still research.
Now, in' terms 22 of the source term work, there is a containment penetration 23 working group, evidently, that is coming up with a report 24 shortly dealing with --
25 MR. SIESS:
10377
77 l
1 MR.
ISRAEL:
I don't know what it was.
I 2
MR. SIESS:
No. The research on penetrations under (m) 3 severe accidents is under way, and it is not a containment 4
penetration working group; it's a containment performance 5
working group, and it is coming up with a report that 6
addresses six plants, but they are pretty p1&nt specific. You 7
can't draw any general conclusions from that.
8 Leon?
9 MR. REITER:
Dr. Siess, I really can't address Level 10 1 versus Level 2 or 3,
but I would like to maybe give a little 11 insight into some of the reasoning behind putting in that 12 fourth condition, some of the discussions that went around 13 it.
I think the basic reasoning has to do with what 14 Dr. Ebersole said.
We didn't want to come up with abstract, 15 seismological or even structural engineering evaluations that 16 would give us little impact as to how these things interacted i
1 7-and what they meant for the plant.
18 A typical kind of scenario we were trying to really 19 get hold of was suppose we did this analysis and we found out 20 that in a vertical response spectra, that if we had to do it 21 over today, instead of having a.5g response spectra derived 22 by this particular criteria, we might be higher in certain 23 frequency ranges.
24 We didn't want that to trigger a hole extensive 25 reanalysis or a fixing program without getting some feeling as 1
)
1
78 1
to the significance.
By significance, I mean what did it mean 2
in terms of the engineering and what does it mean in terms of 3'
the systems of the plant.
Did it mean anything?
Was the 4
fragility such that, yes, these could be easily accommodated?
5 We didn't really think out exactly how that 6
significance would be measured, but quite honestly, we were 7
not thinking of a whole PRA at that point; we were just 8
thinking of a seismically-oriented PRA.
We went into 9
discussions with the Commission, and many of the Commissioners 10 felt that perhaps we were putting too much emphasis on the PRA 11 since there was a great deal of uncertainty about the use of 12 probabilistic estimates, and they urged us to make sure we had 1.3 some possibility to do deterministic analysis.
b 14 And therefore the condition reads "probabilistic 15 and deterministic as required."
16 I think the Utility came to the ACRS, and I may not 17 be correct, but I think at the ACRS' urging but not 18 necessarily as a requirement, the Utility agreed to do a full 19 scope FRA to look at internal and external events.
In other 20 words, I think the Utility may have argued and may have argued 21 successfully that in the context of significance of.75 versus 22
.8g, that a seismic PRA in its narrow sense might have helped 23 us answer that question.
24 Again, as Mr. Brand pointed out, they are even not if the 25
_ required to do a PRA, only if the seismic design a
v.-- ", --,,,
w-
.--,c--.-,_,
e
,w.-.
a, m,.-,-,
,,e.
.,m-,_
.--mw-
79 1
analysis shows to be in excessof what was originally assumed.
2 But I think they said that they can't wait till the last 3
minute, and we realize there is going to be a lot of 4
uncertainty, so they are going to go ahead and start work on 5
the PRA immediately.
6 So our narrow perspective at that point was not in 7
generating what is the most correct risk number or the 8
absolute values but rather in determining what is the 9
difference between some new seismic " design basis" or design 10 level that would be determined as a result of this new 11 analysis to what was assumed in the past.
That sort of 12 relative kind of judgment.
[")
13 MR. SIESS:
Thank you, Leon.
- h 14 I would like to express my opinion, and then I will 15 go around the table.
16 As far as I am concerned, I believe that the Level 1 17 PRA that will be done will probably be adequate to assess the 18 significance.
I think how adequate it is will depend on how 19 significant any change in the design basis might be.
I am 20 pleased that a full scope PRA is being done because I think 21 it's important to know the relative contribution of seismic 1
22 and other events to risk, but I would be perfectly willing to a
23 wait and see where you come out assessing the significance and 24 your Level 1 PRA to see.
g
\\
25 Now, that is my personal opinion.
80 1
Dave, I will give you the last word.
,)
2 Carson?
y 3
MR. MARK:
I think I would agree very much with 4
you.
I would not like it written down that Level 2 or 5
enhanced PRA may not come to seem indicated, but that the 6
seismic design base, if within the present envelope, that 7
viewing the need for an additional PRA, with that in sight, 8
would make a lot of sense.
And I think that is the position 9
that the applicant has somewhat indicated, anyway.
10 If they are within an envelope, then the present 11 plans should cover.
If they find that there are things 12 sticking out that are worrisome, then they might have to go 13 further, and I think they should be willing to, at some future
\\s 14 date, and would when they knew what.
But I believe I am in 15 agreement.
16 MR. EBERSOLE:
It is entirely reasonable to me, 17 Chet.
I have no problem with that.
18 MR. SIESS:
Dave?
19 MR. OKRENT:
Well, I am sorry to see that the 20 Subcommittee members are wrong.
21
[ Laughter) 22 The Staff is completely wrong in its logio, in my 23 opinion, and I think the Utility would make a mistake in 24 letting the Staff go the way from the Level 1 to the Level 3 O
\\'
25 and not itself defining what the Level 2 was.
If I were the
81 1
Utility, I would shudder at the thought of being at the hands O
( )
2 of the Staff and its consultants and who said what the 3
releases were.
4 What you are proposing would not evaluate the 5
containment properly.
You would not have a handle on what its 6
capability was.
You would not look at how various features 7
that might be on during containment would run during a 8
postulated core melt accident, and this is the question at 9
some plants.
Not the same as Indian Point with regard to the 10 safety features.
11 And I think all around it is a fundamental mistake 12 and it is, in fact, a modest amount of money, and the time to 13 plan for Level 2 is at the beginning.
Get it into what you
(h G
14 are doing, get the necessary information developed, and not, 15 in my opinion, because something turns up later, try to add it 16 on.
17 I'm just astonished.
I will put it that way.
18 MR. SIESS:
Now, you didn't really address what 19 basis you think is necessary to evaluate the significance of a I
20 possible change in design basis.
21 MR. OKRENT!
I think the question is public health 22 and safety, and if, in fact, their evaluation were to come out Staff, Licensee, ACRS -- that the 23 that everyone agreed 24 design basis should have been less than it is already designed y
\\-
25 for, I think they could then at that point say we have done
82 1'
what you asked for.
/(uJ 2
But as has been stated, since you can't anticipate 3
that, begin now doing a PRA, and furthermore, having a PRA, 4
you are in a good position to look at individual 5
perturbations, seismic or otherwise, that may arise during the 6
next three years or thereafter.
And if they don't do a PRA 7
for this reason, the severe accident policy statement says 8
that they would have to do some kind of systematic evaluation 9
of their plant, and there aren't very many plants that are 10 good precursors for Diablo Canyon.
So I think sooner or 11 later, they are going to do something like a PRA and it will 12 have to include a good seismic component.
13 So all in all, I think the Staff position, in a
{
N/
14 sense, is wrong.
I think it is to the disadvantage of the 15 Utility ot wait, for a variety of reasons, because I would 16 like, as I say, to have done my own evaluation because the f
17 Staff may hand you one that you don't like, and I am sorry the 18 Subcommittee members don't see the light.
19 MR. EBERSOLE:
Could I ask for clarification?
Due 20 to your vehemence in this matter, Dave, let me say I want a 21 clarification.
You say that the Level 1 will not do a 22 containment integrity analysis; is that right?
23 MR. OXRENT:
There is no need to do it in a Level 1.
24 MR. EBERSOLE:
Oh, yes, there is, f
25 MR. OKRENT:
No.
83 1
MR. SIESS:
No, not in Level 1
^
i s j 2
MR. EBERSOLE:
Well, let me say why there is, in my 3
view.
4 MR.
ISRAEL:
Let me just clarify that.
They will do 5
integrity so far as system isolations --
6 MR. EBERSOLE:
No, no, no.
Let me tell you why I 7
took my position so we all understand each other.
I took it 8
there would be a containment integrity analysis for the 9
following reason.
If you look at a seismic event and find 10 that the centainment is, in fact, damaged, it was almost 11 axiomatic there will be damage to the heat removal systems for 12 the core, t
[
1-3 MR. SIESS:
I think that is included.
Any seismic
(
14 effects on the heat removal systems, any seismic effects on 15 MR. EBERSOLE:
Due to containment failure?
16 MR. SIESS:
Yes.
17 MR. EBERSOLE:
See, I am saying if you have 18 structural failure of the containment, it is axiomatic that 4
19 you will have damage to the heat removal systems from the 20 core.
21-MR. SIESS:
They have to do a seismic analysis of 22 the containment.
23 MR. EBERSOLE:
Well, we did have, in fact, a 24 substantial containment analysis.
25 MR. SIESS:
But that's not telling you whether the
84 1
containment will fail.due to overpressure.
i
(
2 MR. EBERSOLE:
No.
3 MR. SIESS:
I think Okrent has stated his position 4
-very' clearly.
What I said was I don't think anything beyond 5
the Level 1 -- depending somewhat on what happens to the
(
l 6
design bases.-- I think Level 1 is sufficient to determine the 7
significance at that stage.
If there is significance, you are 8
going to have to go beyond.
j 9
Okrent's warning about putting yourself at the mercy 10 of the Staff is another issue that I might well agree with.
j.
4 j'
11 If-I were doing it, I would go farther, but I don't think 12 there is a requirement to.
13 MR. BRAND:
I had one thing, and that is we will, of t
4 14 course, continue to take Dr. Okrent's views in mind as we are 4
15 going forward with this program.
I have been assured by our 16 staff as well as by John Gerrick that as we proceed with our t
17 program, we are not precluding or interfering with later work 18 on subsequent Levels 2 or 3; that the work we will be doing in 19 and on this three-year program will certainly feed into a 20 later effort on 2 or 3 should it be desirable.
21 As well, we will certainly keep your thoughts in 22 mind with regard to simply being dependent solely on staff as 23 we make our evaluations on any subsequent backfits or plant 24 improvements that may be considered.
25 MR. MARK:
I would like to add to any impression
~.
85 1'
that I tried to indicate before.
I can think of reasons why
/%()
2 they might be well served to do a PRA, and that there may 3
even be reasons outside this program which will make it seem 4
necessary or worthwhile.
I think insofar as the seismic 5
program is concerned, those reasons aren't embedded in that 6
and that they meet their license condition, probably, or with 7
some hope, with the simpler thing.
But by the time they are 8
through with source term work and severe accident analysis, 9
you may wish you had done the PRA full-scale planned from the 10 start.
11 MR. SIESS:
Gentleman, what I propose is now I think 12 this goes to the Full Committee.
The Subcommittee can report 13 that the Staff has found the program acceptable and that wo 14 have found the Staff's findings acceptable; that the issue of 15 the PRA level still exists, and I assume Dave would like to 16 put additional remarks in any letter unless we get a majority 17 of the committee agreeing with him.
18 So I think the Full Committee should hear some 19 arguments on the Level 1 versus higher level PRA involving, I 20 assume, Mr. Israel, Dr. Okrent and the Licensee, and then we 21 will prepare a letter based on that.
The letter will either 22 recommend a higher level PRA or not.
If it does not, Dave can 23 add his remarks.
24 Is that acceptable procedure?
25
[No response]
r.
~ _ _ - -.,--_ _ -
4
86 1
MR. SIESS:
Okay.
O()
2 I don't see any way of getting this scheduled 3
earlier on tomorrow unless we could swap some time with 4
Quantitative Safety Goals.
Would you have any objection if 5
-that was swapped, Dave?
6 MR. OKRENT:
Yes, I would object.
7 MR. SIESS:
Why?
8 MR. OKRENT:
Because-I think it is a more difficult 9
issue.
10 MR. SIESS:
Safety Goals is 1:00 to 3:00, and this 11 is 3:00 to 4:30.
12 MR. OKRENT:
Yes.
I would prefer not to have it i
13 changed.
(
14 MR. SIESS:
Okay.
That's the way we will do it, 15 then.
16 Any other comments?
17
[No response.]
18 MR. SIESS:
Thank you.
The meeting is adjourned.
19
[Whereupon, at 11:20 a.m.
the meeting was 20 concluded.]
21 22 23 24 25
O
\\'
1 CERTIFICATE OF OFFICIAL REPORTER 2
3 4
5 This is to certify that the attached proceedings 6
before the United States Nuclear Regulatory Commission in the 7
matter of:
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS S
9 Name of Proceeding: Subcommittee on Diablo Canyon Plants 10 11 Docket No.
()
te piace:
Washington, D. C.
13 Date: Wednesday, July 10, 1985 14 15 were held as herein appeaEs and that this is the original 16 transcript thereof for the file of the United States Nuclear 17 Regulatory Commission.
18 (Signature) g (TypedName'6fReprter) [Suza ge B. 56dng-20 21 22 23 Ann Riley & Associates, Ltd.
24 25
>