ML20141P192
| ML20141P192 | |
| Person / Time | |
|---|---|
| Issue date: | 03/12/1986 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1495, NUDOCS 8603190131 | |
| Download: ML20141P192 (526) | |
Text
_.-_ - - - - --
' ORIGINAL -//pff' O UNITED STATES NUCLEAR REGULATORY COMMISSION IN THE MATTER OF: DOCKET NO:
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS SUBCOMMITTEE ON SAFETY PHILOSOPHY, TECHNOLOGY, AND CRITERIA O-
~
LOCATION: WASHINGTON, D. C. PAGES: 1- 396 l
DATE. WEDNESDAY, MARCH 12, 1986 s" ~m .
MSCECECOPY x l
Jo Not Remcale from ACRS 0 lice 4 !
O qP ACE-FEDERAL REPORTERS, INC.
officiat newrtm D\\ 444 North Capitol Street Washington, D.C. 20001 0go1" gift 1 a6031O (202)347-3700 F1495 P Dit NATIONWIDE COVERAGE
e CR26135.0 1 DAV/dnw
'1 . UNITED STATES OF AMERICA UI 2 NUCLEAR REGULATORY COMMISSION 3 ***
4 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS S SUBCOMMITTEES ON SAFETY PHILOSOPilY, TECHNOLOGY AND CRITERIA AND 6 EXTREME EXTERNAL Pl!ENOMENA 7
Nuclear Regulatory Commission 8 Room 1046 1717 11 Street, N.W.
9 Washington, D. C.
10 Wednesday, March 12, 1986 11 The raceting of the subcommittees convened at 8:50 a.m.,
12 Dr. David Okrent presiding.
13 ACRS MEMBERS PRESENT:
14 DR. DAVID OKRENT, Presiding.
DR. FORREST J. REMICK 15 DR. CARSON MARK MR. CARLYLE MICHELSON 16 MR. !!AROLD ETIIERINGTON 17 CONSULTANTS PRESENT:
18 DR. P. POMEROY M. TRIFUNAC 19 C. SMITl!
M. GRIESMEYER 20 RICIIARD SAVIO, Designated ACRS Staf f Member 21 22 23
( 24 rks i fleporters, Inc.
25
r PUBLIC NOTICE BY THE UNITED STATES NUCLEAR REGULATORY COMMISSIONERS' ADVISORY COMMITTEE ON REACTOR SAFEGUARDS WEDNESDAY, MARCH 12, 1986 The contants of this stenographic transcript of the ,
proceedings of the United States Nuclear Regulatory 1
Commission's Advisory Committee on Reactor Safeguards (ACRS), as reported herein, is an uncorrected record of the discussions recorded at the meeting held on the above ,
date.
No member of the ACRS Staff and no participant at O ,
this meeting accepts any respon'aibility for errors or . .
inaccuracies of statement or data contained in this transcript.
e O
l 1_
4 1350 01 01 2 DAVbw 1 PROCEEDINGS 2 DR. OKRENT: The meeting will now come to order.
3 This is a meeting of the Advisory Committee on 4 Reactor Safeguards combined Subcommittees on Safety 5 Philosophy, Technology and Criteria and Extreme External 6 Phenomena.
7 The purpose of this meeting is to review safety 8 goal policy, containment performance criteria and 9 containment venting, the impact of the 1986 Ohio earthquake 10 on the Perry plant and the proposed resolution of ASI A-17 11 systems interaction at nuclear power plants.
12 I'm David Okrent Subcommittee Chairman. Other
() 13 ACRS members in attendance at the moment are Dr. Mark and 14 Dr. Remick. ACRS consultants in attendance for this portion 15 of the meeting now are Dr. Lave, Dr. MacLean and 16 Dr. Griesmeyer.
17 Dr. Savio is the cognizant ACRS Staf f member for 18 I this meeting.
19 , The rules for participation in today's meeting i
20 have been announced as part of the notice of this meeting 21 previously published in the Federal Register on Thursday, 22 February 27, 1986.
23 A transcript of the meeting is being kept and
() 24 will be made available as stated in the Federal Reglster 25 Notice.
1 ACE. FEDERAL REPORTERS, INC.
cm rm ren - - --- --__.-- -
1350 01 02 3
(')
L/DAvbw 1 It is requested that each speaker first identify I
2 himself or herself and speak with sufficient clarity and 3l volume, so that he or she can be readily heard.
4 We have received no written comments from members 5 of the public. We have received no requests for time to 6 make oral statements from members of the public.
7 Before calling on the first speaker, I should 8 note that the Commission seems to be trying to complete some 9 sort of action on safety goals in the not-too-distant 10 future. And in fact, as the ACRS members know, we have a 11 special meeting scheduled with the Commissioners of those
_ 12 members who can make it, I guess, later in March. So that
'~'
13 there will be considerable interest and pressure on ACRS to 14 l try to arrive at some kind of position consensus or split, 15 j as the case may be, at the March meeting. And in this 16 regard, it will be useful for both members, and I would 17 think consultants, in this part of the meeting, to arrive at 18 I what I would call specific recommendations, as well as l
19 general comments, to the extent that that's practical. And 20 in fact, if you weren't going to endorse what the EDO has 21 set out or sent up to the Commission, is there anything that 22 you would endorse. If so, what is it.
23 I think the members will have to face that issue,
() 24 and the consultants. If you have such a proposal, I would 25 appreciate having it.
ACE FEDERAL REPORTERS, INC.
.--_-______-_____--_m_-_________m-___--m-_________-
E 1350 01 03 4 (L 'DAVbw 1 The meeting is scheduled to run until 11:30, in 2 case you've looked at the agenda.
3 (Laughter.)
l 4 In case you've looked at the agenda, we're 5 scheduled through with a shortened lunch break and no dinner 6 until 8 :00 p.m. , and I doubt that we'd want to run later.
7 In any event, why don't we begin the meeting.
8 The consultants should feel free to interject as they like, 9 while giving oach speaker a chance to make his points. At 10 the end of the session, I guess we would, among other 11 things, ask if the consultants have anything beyond the i
! 12 written reports. Two of you have turned in written CE) l
. 13 reports. Dr. MacLean has also supplied one.
l e
14 Does the Subcommittee want to raise any specific 15 issues at this time?
16 (No response.)
17 Well, they seem to be quiet. So let's call on 18 the Staff to tell us what the answer is to this complex 19 situation, and while they're doing it, if they have any 20 comments on the specific points raised by Commissioner i
21 Bernthal on the presentation with tha commissions by Mr 22 Stello, and so forth, I'd be interested in hearing where the 23 Staff's thinking is on some of those, because they're I
() 24 somewhat fundamental. l 25 Anyway, who's up for the Staff?
l ACE FEDERAL REPORTERS, INC.
.--- _ -_ _ -- _-_1 v m a n _--_ __ _ _ we -m _ _._ _rnu~~t___
1350 01 04 5 o/ DAVbw 1 MR. SNIEZEK: I'm Jim Sniezek, Acting Deputy 2 Executive Director for Fegional Operations and Generic 3 Requirement.
- 4 With me today is Matt Taylor of my staff, who 5 will do the main presentation, Mel Burns from the Office of 6 Research, Sid Fout, who is a cost analyst in the Office or l
l 7 Resource Management and George Sege from NRR.
l l 8 I think Dr. Ohrent correctly summarized the fact 9 that this is a very complex issue. It's been debated quite 10 heavily for the last five or six years in NRC. We ran a 11 two-year trial evaluation of the safety goals that were 12 published i,n 1983. The Fevised Policy Statement is a thick
.. O
! 13 report issued 'by the steering group that did the l 14 evaluation. Since that time there's been about 10 months of l
15 debate between the NRC Staff, the Commission and meetings 16 with ACRS to try to get to the truth of the matter, the 17 heart of the matter, the way we should be going.
18 l That was culminated by the Staff, at least, in 19 the February 14th paper to the Commission, which resulted in l
20 a February 20th Commission meeting, where the Staff 21 attempted to draw a reasonable. compromise to the various i
i 22 issues that were raised by the individual Commissioners, by l
l 23 the NRC Staff itself and by the ACRS.
() 24 Hopefully, we were able to do that in the 25 February 14th paper.
l l
? l ACE FEDERAL REPORTERS, INC.
--_-- ----_ -_ _ --_ r~wnant _ _ _ - - _ - _ _ _ _ _ - -
1350 01 05 6 DAVbw 1 I would like to mention at this time that all 2 office directors, the Director of Inspection and 3 Enforcement, the Director of NRR and the Director of 4 Research, fully support the February 14th Staff proposal to 5 the Commission, and the attempt of the February 14th 6 proposal to the Commission was to help the Commission reach 7 a decision. The Staff had asked the Commission about ten 8 questions for guidance, on which the Commission was unable 9 to reach a decision or even give the Staf f guidance on. So 10 the Staff tried to take it upon itself to try to ombellish 11 on those points and help the Commission reach a decision. j 12 l The Staff is basically of the viow that there is O 13 not mu'ch more that,can be done on this issue at this time',
l i
14 and it la just about time for a decision on whethe,r we want 15 a safety goal, what it's to look like and where we go f rom 16 here. j 17 With that, I would like to turn it over to Matt l 18 l Taylor, who will spend about half an hour briefing the 19 Committee on the February 14th paper that the Staf f 20 presented to the Commission to make sure the ACRS 21 understands that paper.
22 DR. OKRENT: Excuse me one minute.
23 Can I ask the Subcommittee members and Q 24 consultants, do you feel like you need a 30-minute briefing 25 on what was in the EDC paper to the Commission? !
1 I
ACE FEDERAL REPORTERS. INC,
______m--__--- __- N ----_---_ rTn m - _ ____ _ _ - -_---
=_
l 1350 01 06 7
'DAVbw 1 DR. REMICK: I think it would be helpful, Dave.
2 DR. OKRENT: I just wanted to go --
3 DR. REMICK: I find it complex.
4 DR. GRIESMEYER: Yes.
5 (Slide.)
6 MR. TAYLOR: As Jim Sniezek said, I'm Matt Taylor 1
7 from the Staff. We spent the last few months trying to put 8 together a resolution of many comments that both the Staff 9 and the Commission of the ACRS put forth over the past 10 number of months. We spent the past few months trying to 11 put together a resolution of many comments that both the 12 Staff and the Commission of the ACRS put forth over the past j
O 13 number of months.
14 The proposal we sent forward to the Commission on 15 February 14th -- is that readable? Okay -- had an objective 16 to assist the Commission in its deliberations on the best 17 course of agency action on the matter of safety goals.
18 In this 14th paper, the February 14th paper, we 19 believe we've accommodated the outstanding ACRS issues to 20 the extent possible. That derives from the July 1985 21 comments on the Steering Group report. In the paper, we 22 recommend that the Commission adopt in final form the 23 qualitative statemento. We provided the flexibility to
() 24 adopt provisionally, subject to further public comment, or 25 finally, the use of the integrated matrix, as a means of ACE. FEDERAL REPORTERS, INC.
m -
1350 01 07 8 DAVbw 1 implementing the qualitative safety goals.
2 As Jim Sniezek pointed out, the Program Office '
3 Directors were supportive of going forward with the February l 4 14th proposal. The comments that we received from the 5 office directors indicated they were reasonably satisfied 6 that their outstanding concerns on the Steering Group Report l
7 were resolved by the February 14th proposal.
8 Depending on the Commission's decision on the 9 14th proposal, we are prepared to go forward and implement 10 the Commission's decision. We would expect this to be l 11 primarily used in connection with resolution of generic 12 issues at this time.
O 13 DR. MARK Could I ask -- it's on that thing, I 14 think the third from the bottom - " Recommend authorizing 15 Staff to use the matrix on a trial basis." What is meant by 16 that word " Staff"? Is it the regional directors, the EDO, 17 or who?
18 MR. TAYLOR: As I mentioned, we would expect to 19 use this in the resolution of generic safety issues. Now 20 that would actually be the Staff at headquarters.
21 DR. MARK: !!eadquarters Staff.
22 MR. TAYLOR: Right.
23 MR. SNIEZEK Let me address that. " Staff" means 24 anyone who works for the EDO. The EDO would develop the 25 specific procedures to guide the use by his total Staff.
ACE FEDERAL REPORTERS, INC.
_____________ ___ __-_______ __ FVMW1 - _-______ ms .u a __ _ _ --___m
1350 01 08 9 l O
(J DAVbw 1 DR. MARK: So it would not be something at the 2 option of Region 37 3 MR. SNIEZEK If it is to be used, it would be 4 used by ovarycno in a uniform manner. It would not bo 5 optional.
6 DR. MARK: That's the point I was concerned 7 about.
8 DR. REMICK: Matt, whon you say "to uso," I don' t l 1
9 want you te go through the litany, maybe it will be coming 1
10 out on all the proposod usos, but one thing that is not 11 clear in reading this -- are you going to uso it in 12 licensing decisions on reactors, or are you going to uso it
'~
13 l for backfit, an propogo,d by tho Staf f now? I know about the 14 businosa of notting prioritios, and so forth, but what about 15 licensing matters, whether it be backfitting, whethor it bo i 16 a now licenso, or so on. That is not clear to me.
I 17 MR. TAYLOR: I would say that would be valuable 1
l 18 I in soveral areas. Backfitting, for one, in implomonting a 1
l 19 savoro accident policy. I would not soo that as being a 20 i licensing tool in itself, but only to supplomont safoty 21 procoodings.
22 DR. REMICK: Only for supplomontal information.
23 MR. SNIEZEK: I think wo stato quito clearly in i
I
() 24 the paper that licensees would be expected to comply with l 25 their requlations and commitments. This would givo us l
l ACE FEDERAL REPORTERS, INC.
- _ _ _ ___ _ _______ _ ____ mfg h71 - --- __ 1;rtrt^# arm 1 fn m .
l l l l
1 j 1350 01 09 10 ;
DAVbw I additional insight to help make judgments, but would not be h i
! 2 the overriding factor. l f
3 DR. GRIESMEYER: Would it be possible to i t
4 characterize the matrix as something that is just adding 5 perspective, but you don' t applyit directly anywhere? l t
6 MR. TAYLOR: I think it does provide an invisible i
7 template to assist the staff in making decisions and to find !
8 out where they might fall in that set of risk statements, if f 9 you willt yes. If they have a particular issues or l
l 10 particular PRA result, they could bear against that, the 11 matrix, as a template. f 12 DR. GRIESMEYER:
o But since it is not -- you don' t !
. i 13 make decisions based on how you come out on that template. :
i 14 You just use that with the other decision factors.
[
}
15 MR. TAYLOR: That's right; yes. So it would be !
16 one factor.
17 DR. REMICK: Matt, maybe you're going to cover it 18 later, but in case you don't, the one there where you are 19 basically taking the quantitative objectives out and put I l
C 20 them in the matrix, I guess. I don't under the purpose of l 21 that, because I think they now become hidden. That was one i
22 thing i thought that would help the public understand a l 23 quantitative sense of what the qualitative goals are by l 24 taking them out and now putting them in the matrix, which I 25 would claim are quite hidden in there, not clearly defined, ACE FEDERAL REPORTERS, INC. ,
t
___ _____ EYMEl___ _WP~Ir^1h__Rfn"n -
l
\
l l t l
1350 01 10 11
! DAVbw 1 and so forth, in my view.
1 2 I wonder why you did that. Why take out those >
l statenents, which I thought were fairly well-accepted.
3 I
(
1 4 don't know of anybody who's really objected to those. j 5 MR. SNIEZEK Dr. Remick, if it's okay with you, t 6 Matt's got that natrix. He is going to go through it later, [
I 7 and I think you can get the rossons then. We're starting to i
0 jump ahead.
l 9 DR. REMICK: Fine.
I 10 (Slide.) f r
11 MR. TAYLOR: AAs I mentioned, the February 14th l 11 proposal recommended to the Commission that they should O 13 issue, in final form, the Safet.y Goal Pdlicy, which,would 14 adopt, as final, the two qualitativo' statements on l 15 individual and societal risk. What we proposed to do was to i 16 I eliminate quantitative objectives, discrete statements of !
17 quantitative objectives and the benefit-cost guideline in l 18 l the policy itself and combine those into the integrated f
! t 19 matrix. It was not our intent to bury those, but to l 20 integrate those. So you can see the relationship between !
21 accident prevention, health effects and the cost-benefit 22 guidelines.
23 DR. REMICK: Why not have it both ways? I don' t 24 understand.
25 MR. TAYLOR: We basically do have those both ACE. FEDERAL REPORTERS, INC.
FGK960--_- 91T::I:DGuvtJ________CibEC~D
c350 01 11 12 m y
'DAVbw I ways, as currently structured in the February 14th paper.
2 DR. REMICK: But it says oliminato. In tho l
I 3 paper, it says oliminato. I know they are stated in thoro, 4 so I got confused. Aro they going to appear in the final 5 Policy Statomont as the Staff proposed or not?
6 MR. TAYLOR: That is, of courso, the Commission's 7 final docinion. Wo aro recommending that they appear an in the Fobruary 14th paper. j 8lI I
9 DR. REMICK: In that an explicit statomont thon?
i i
I 10 i Will they bo in thoro as an explicit statomont?
11 l MR. TAYLOR: They aro.
12 DR. REMICK: Why in "oliminats" thoro?
I )
'" 13 , MR. TAYLOR: Thoy aro in thoro an an oxplicit I 14 statomont, your howevor, wo mado the broak point, in tormn ;
L I 15 of adopting as final policy thin portion of the nafoty l
16 h goals, and in terms of tho quantitativo trostmont, wo would 17l liko this to bo .subjoct to furthor comment, provisional, or I
18l thoro would be an option for the Comminnion to adopt that an 19 its final policy.
20 Now wo do have a dincussion on each one of thono 21 quantitativo objectivos in the papor. When I nay 22 "oliminato," porhaps that in too otrong a word.
23 DR. REMICK: I got confunod about what it moann,
(~')' 24 becauso I do think you nhould clearly havo thono stated in u
l 25 the document, not junt bo incorporated in tho matrix.
ace. FEDERAL REPORTERS, INC.
f3pN3 _. N en.u, em enn un nu m _ _ _ _ _ _ . _ . _ _ _
l I
1350 01 12 13 l l
DAVbw 1 That's my own personal view.
l 2 MR. SNIEZEK: There is no question. I believe 3 they are clearly stated in the document, but their use is in !
l
! 4 the matrix. They are not used outside of the matrix. That i 5 is what it means. The reason why we did that is from at i
6 least the Staf f's review of everything that has been said 7 over the last five years, it seemed, at least until the 8 February 20th Commission meeting, that there was no f
9 disagreement on the two qualitative statements.
l 10 DR. REMICK: Dut can' t you say that about the two
[
11 qisantitative, maybe not the cost-benefit quidelines, that is !
12 a lot of controversy, in what you include, but the two cost
[
O 13 statement ones, I would think, are the sa*mo category as tho 1
l 14 qualitative.
- j 15 MR. SNIE2EK: I would say they were cloan, but l
16 there was some debate, as to whether they should be 1 ;
17 percent or 10 percent, but there was some debate. We tried [
t 18 to eliminate that area and just move forward with safety 19 goals. i r
20 DR. REMICK: Thank you. !
{
21 DR. OKREN1*: Before you leave that graph, j l !
l 22 Commission Bernthat has raised questions concerning the !
23 2trat bullet.
O 24 "n a do vou pron to tata abo"* *at au etoa?
25 MR. TAYLOR: I am not sure I can respond to his l
l Acs.FEocnAt. REPORTEMS, INC.
m nutm ma m
1350 01 13 14 DAVbw 1 question.
2 DR. OKRENT: In anyono from the Staff going to 3 address this quostion?
4 MR. TAYLOR: This quontion had to do -- as I 5 recollect from the February 20th mooting, he took some 6 issuo with the qualitativo statement that wo should bo 7 comparable on the technology.
O DR. OKRENT: That was one. ,
9 MR. TAYLOR: lie felt as though wo should havo 10 stated wo will do bottor. That was the fooling I had, at 11 least in convormation, that was the principal ono I recall.
12 DR. OKRENT: Doon the Staff have a comment?
13 MR. SN!EZEK The Staff han no specific comment 14 on his point. It is anmothing that has to bo dotiated by tho 15 Comminnion. It was put in and dovolopod by the Comminnion 16 originally. The Staff naw no reason to chango it. Tho 17 Commission had not dono a study of comparable technologion, 10 but in the original basin for tho papor, it was stated --
19 this to the original natoty goal paper -- it was stated that 20 nucioar wan nafor than comparable or at loant no nato as 21 comparable tochnologion.
22 The Staff has no information to refuto that. It 23 14 ron11y a basic policy docinion on the Comminnion'n part, Q 24 whether or not it wanta to comparo nuclair generation to l 25 other typos of oloctrical conoration.
ACEL?EDERAL RElioRTEns, INC. <
---_------f~ D UE -------- M Qr r~1---------fD C _------_-- --
1350 02 01 lb b
R DAVbur 1 DR. OKRENT Doforo we go away, it may be a basic 2 Commiosion policy, but originally the staf f proposed words 3 to the Comraission of that nort, some staff, and in conoral 4 the Commissioners at least looked to the statf for a 5 possible turn of words.
6 I just want to see whether you are standing l
7 firm.
l ,
8 MR. SNIE2EK The staf f lan' t studying that issuo I 9 at all. Staff would bo willing to go with either not of I
10 words, oither with it in or with it out. He don't think it l 11 makos that much difforonce.
12 MR. TAYLOR: My personal view is that wo have not O
13 soon a high lovol challongo to thono qualltativo 14 statomonts. This was tho first time. They sort of 15 surprised un that thoro was a challenlo at that timo.
16 DR. OKREt1T: I thought that was a high lovel 17 challongo.
j 18 HR. TAYLOR: Yos. I am saying up to that point 19 , wo havo not soon it. .
l i
20 DR. OKREllT Well, lot me dinngroo that you have 21 not soon a high level cha11onge beforo. l i
22 MR. TAYLOR: During my tenuro on the safety goal 23 team. i O 24 oa oxactir' a ta co==t toa r a rath t ad !
25 ' Commissioner Ansolatino expressed in ono form or anothor l
l ACE.FunI!nAL REPORTERS. INC. ,
1
- - _ __ _ _ _ _ 3~umssn_ _ _ _ -_ __ m --__--_m --:
1350 02 02 16 p I V DAVbur 1 what I think ir a similar point, that somohow there is 2 cither a written or an implied goal that the chanco of --
3 and now I don't know whether thoy both said the same thing 4 -- but a Three Milo Island kind of accident or a largo scale l 5 coro melt should be small in the next few decades, that one i
i chance in 50 in the next 15 or 20 years -- I am sorry --
7 50/50, not one chanco in 50. They fool uncomfortable with, i
8 I think.
9 They both indicated -- I am not suro I can in my 1
10 own mind without the benofit of somo timo consuming analysis
, 11 make theird figuros consistent, but nevertholons they havo l
( 12 both said that thoro in a nood to make the probability of 1
O 13 another coro damaging or coro molt accidont rathor low, and 14 in f act one of them has attachod this right to tho laws that l
l 15 sot up tho Atomic Energy Commission, and so forth, which l 16 says in the process of requiation you have to look out for 17 what you might call tho health of this industry, and they :
10 ! said that thoro would be no health at all.
i 19 DR. LAVE: Can I break in horo? :
l l 20 There are a couple of issuos running around. Lot 21 me just try to pull thom in.
I 22 The first isnuo has to do with tha relativo 23 safety of nuclear and others. What han gono on sinco the
] 24 work that I did and some other peoplo did in tho '70s on l 25 this doon not shake the banic proponition that nuclear in 1
ace 4 nounal RuronTEns, INC.
_ . _ _ _ _ _ _ _ _ _ _rvan _____.-__m_______e__________
1350 02 03 17
<^ I l L_NDAVbur 1 safer now than the othor means.
F 2 Uhat has gono on is a reinterpretation of some of 3 the hoalth offects of air pollution from other moans of 4 obtaining olectricity has gotten much more controversial, 5 and it is not that nuclear has gotten much less safe. It is 6 that some of the other means havo been interpretod as being 7 safor, since there is no resolution now as to how dangorous ,
l 8 air pollution from coal or oil is.
9 l
I interprotod the staff as backing off, saying i i
10 that wo can't definitivoly prove that nuclear is safor and 11 therefore let's got away with a statomont that says it isn' t 12 loss safe.
( 13 DR. OKRENT: Lot mo mako a coi.;mont.
! 14 Commissioner Dornthal communted, at loost in tho i 15 mooting with the staff -- and you may not have gotton that l 16 transcript, but I think that is where he olaborates on it.
i l 17 It was that ho didn' t think that the public would bo 18 satisfied at all with nuclear, which says comparable to 19 coal. They don' t think that coal in that good, and tho goal l 20 for nucloar should bo that it is bottor. ,
21 I don' t liko to put words in the Commissionor's 1
j 22 mouth, but I think he was questioning the uno of the word 23 "comparablo."
l 24 of courso, " comparable" is a moving targot.
I
(])
25 Some of thoso coal plants aro bottor than others, and so ACEfEDERAl. REPonTuns, INC.
_ _ _ ____ __ ___ _ - - _ _ _ -__1MND _ . f2TafbO:mm _ M----------------- -
h l l
[
1350 02 04 18 (
l DAVbur 1 forth.
l 2 But anyway, he had a fundamental question on the
[
3 " comparable," and this is a further complexity. l l
4 DR. LAVE: But from the staff viewpoint, if I l
5 were the staff and were asked to prove that nuclear were 6 safer than, for example, a state-of-the-art coal plant right ,
7 now, I would have a very hard time because of the amount of j 8 controversy on health effects of air pollution.
9 That is not part of the central mission of the 10 NRC just because the ranges of uncertainty have an overlap.
11 1 think it depends on whom the burden of proof is that 12 nuclear is safer.
O 13 The second issue, just to go on with that for a 14 second, is that when we were doing the work on how safe are
, r 15 nuclear plants, the amount o'f health offacts added by core 16 melts or other problems was extremely small. Those were not i
[
17 the sources of health problems, and so it is not that we are
{
18 concerned with future core melts like TMI at least in terms !
19 of increasing the health effect of nuclear reactors. Those l 20 really were very small contributions to the overall effecto 21 of reactors.
22 Go they are separable issues. I am not arguing 23 that you don't want to have a very, very stringent core melt Q 24 requirement, but that is not what is required to have l
25 nuclear reactors be safer. l ACE. FEDERAL REPORTERS, INC.
u-_____ __ r vwfn mv . . o e __
1350 02 05 19
/~T I L_IDAVbur 1 DR. OKRENT: I think I haven't soon in what i
2 cither Commissioner Aouolatino or Commionionor Bornthal havo 3 nald or writton that they arguo from the point of view of 4 health ef fects, that it is very important that the chance of the core molt, whether it is in the next 20 years or tho l 5
6 next century, be smaller than what was projected in one 7 estimato that the staff gavo some months ago, which I think l
l 8 was of the order of 50/50 by the end of the century, or somo 9 number like that.
10 MR. TAYLOR: That can be derived since 1975.
11 DR. OKRENT: These in a sonno are fundamental to 12 what you'might call the high order safety policy question,
() 13 and thoro in nomo interont in the Comminston in trying to
- 14 i adopt tho.high order safety policy and nay you recognian the l 15 ' details of theno quantitative objoctivon, and no forth, aro l
l 16 not all that easy to fix, and whatover. They might nond l
l l
17 something out for pubitcation and nay we will give it throo l 10 years or nomo number of tonto, and so forth.
19 Thero in a connidorablo interont, as I nay, in
[
! 20 thin.
l 21 "Difurcating" is the word I hoard. I didn' t look 22 it up in the dictionary, but there arn those in a nonne l 23 fundamental quantions on the natoty policy itnolf which tho
(]) 24 ntaf f didn't addrons in the memo to EDO but they heard 25 about.
1 1
ACl!.FliottnAL Rt!I'onti!ns, INC.
____ _ ___________ __ _ _ __________ ____ T % r3 m _ _ ww nm
! 1350 02 06 20 l !
1l l DAVbur I wondered if they have developed opinions on l ,
2 it.
3 MR. SNIEZEK Let me address that. Again, I !
j 4 think, Dr. Okrent, you expressed Commissioner Bernthal's [
l 5 coments at the meeting quite accurately. It was concern l l
6 not so much with the "less than" issue as with the i 7 comparability issue. ,
8 DR. REMICK: I didn' t read the "less than" f
9 issue. I think you read it out of context.
l 10 MR. SNIE2EX: The statf, I think, if you !
11 remember, was told several years ago by the Commission not 12 to do that study. Therefore, the statf wouldn't have a O 13 l specific viewpoint on it. I i
14 DR. OKRENT: See, this is a qualitative goal, and l
15 he wasn't asking for the moment that you show quantitatively 16 that something was the case. He, I think, was arguing I
17 society really doesn't think that whatever their perception )
i 18 of coal plant risk is, is good enough. j 19 MR. TAYLOR: They want to do better. !
20 DR. OKRENT: They want to do better for nuclear.
l 21 Paul 31ovic, who is not here for certain reasons, argues in 22 the same direction. ,
23 DR. REMICK I woulu like to make a point, Dave.
] 24 '
25 one, I think Commissioner Bernthal did read it out of context. lie read " comparable" and not "or less than."
t l
i ACE.FroERAL REPORTEk5. INC. (
135ma__ _ __ cn=rMm__ ____ _f3rra___ i
1350 02 07 , 21 I
(~')s
( DAVbur 1 Also, then the last part of that statement should j 2 not be significant additions to other societal risks. That I' 3 emphasizes the importance of coming along and putting in a 4 quantitative objective which defines what is meant by that.
5 That is what concerns me about now dropping those because 6 the quantitative objectivos were to define what significant 7' addition to other sociotal risk is.
8 You have now doemphasised that.
9, DR. OKRENTt Wait a second. I hate to say that I
l 10 he knows that Commissioner Bornthal modo a mistako. My i.
l 11 observation in that he has boon quito porceptivo in looking l
12 at what in in this, which is partly now to him sinco ho i O 13 wasn't here indood, I think, when this first came out in 14 ; '01.
l 15 So I have to assume ho did in fact road the whole i l
l 16 thing and has raised this issue.
17 I am not trying to support either way, but I am l 18 trying to find out has tho staf f boon trying to devolop pros 19 , or cons or alternatives or whatevor.
20 I know Stello was shocked. i 1
21 MR. UNIC2CK The answer to that in no. The 22 staff has not bean trying to devolop pros and cons and '
23 altornativen.
I 24 Now, I would like to mention, Dr. Romick, again,
(])
25 ! hopo as we go on we can annuro you that the quantitativo l l
l Act!.Filot!nAL. Rt!Poxitixs, INC.
r e oa-_------ m e m -- - - comr~n - - --- -- - --- - --- ----- --
I 1350 02 08 22 DAVbur 1 statements are to be used by the staff in determining I
2 whether or not the qualitative safety goals are being l 3 achieved.
4 DR. REMICK: I understand that, but I want the l 5 public to know what you mean by "no significant addition to !
6 rish."
7 MR. SNIEZEK I hope we will be able to show you 8 that. It is very clearly stated. l 9 DR. REMICK: The word " eliminate" disturbs me. '
10 MR. TAYLOM: That was an unfortunate word. What f 11 we wanted to do was to say perhaps " separate" qualitative (
12 and quantitative and do a better job on integrating those. l O 13 The quantitative elements are marked in the 14 February 14th paper. We certainly didn't throw those out. l l ?
15 DR. MAC LEAN: I think it is worth emphasising .
16 the point that Lester Lave made initially. However, even if 17 we look only at qualitative statements of the goals, we can l 18 distinguish the goal that nuclear power should be safer than 19 other forms of generating electricity from the goal, that 20 the risk should bo 10. 8
- wingle TMt event happening the 21 rest of the century.
22 Those can be treated separately and should be 23 treated separately.
24 DR. OKRENTI And they are two soparato points.
- 25 DR. MAC LEANT They did not distinguish enough ==
l ACE. FEDERAL REPORTEMS, INC,
! _ _ ___ ______ __ _______ __mn.Dr3 _ m .; i e_______
1350 02 09 23 i / DAVbur 1 in my own j udgment -- not distinguish well in Commissioner 2 Bernthal's statomont.
3 DR. OKRENT: But Commissioner Assolstino has boon 4 pushing the one about no accident in the next whatever.
5 Mike, did you have a comment?
6 DR. GRIESMEYER: No.
7 DR. LAVE: Can I just say that I think that the 8 staff might try to do a little artful rewording of that to i 9 omphasizo that nucioar power is at loast as safo as the 10 s ta te-o f- t he-a r t technology for other fuels and perhaps put 11 in something in there that at that point it is very hard to 12 prov,o that nuclear is safor just because it is so dif ficult i I)' l l 13 ! to figure out the health of fects for all tochr. ologies, l i 14 Thoro is some artful wording in there that will 15 got at what it is that the Commissioner is concorned with 16 and got you cack towards the original intent, which is that 17 , nucloar in fact is to be safer, but not got you into the box 18 l of having to prove that nuclear is safer.
I 19 MR. TAYLOR: As Mr. Sniosok pointed out, we havo 20 not actively sought out to reword this at this point li >
1 21 time.
22 DR. OKREtlT: Botwoen now and 12:00, if you can l
23 think of some artful wording.
() 24 MR. TAYLOR: Porhaps the committoo wouid help us i 25 4 on somo artful wording.
l l ACl!. FEDERAL REPORTERS, INC.
1
I i
1350 02 10 24 O
v DAVbur 1 Okay.
I i 2 (Slido.)
3 Just to be a little redundant here, these are the 4 qualitative safety goals, and we propose the Commission 5 could adopt them in final form as its policy statement.
6 These are unchanged from what has been proposed in l 7 NUREG-0880 throughout these deliberations.
8 The steering group did not propose any changos.
9 (Slide.)
l 10 Now, wo got to the integrated matrix that is sot J
11 forth in the February 14th paper. I believe we have 12 discussed this matter several times with the subcommittee in O 13 i
- the past. ,
1 .
I 14 What wo have dono is to take all, tho qualitativo 15 oloments that woro in the prior policy statomont and put 16 those together in a framework that would visibly show that 1
17 we are trying to emphasizo accident prevention and that wo 18 I are giving priority to health effoets. Basically, wo tio 1
19 those together with the cost /bonefit olomer.:.
l l 20 Now, the stooring group over a two-year 21 ovaluation period recommended inclusion of all important l 22 costs. In particular, thu most controversial area has been 23 the inclusion of avorted onsito costs -- among other l
(} 24 factors, cleanup costs, occupational exposuro costs, the 25 l cost of replacomont enorgy, and somo cost associated with I
l Acu FEDERAL REPORTERS. INC.
l 1350 02 11 5 DAVbur 1 capital lose. Those are the principal elements.
2 Uhat we have done -- in light of the controversy 3 over both the accident prevention or the emphasis on 4 improving the core melt and the averted onsite costs, what 5 we have done is we have tried to show that for future plants 6 we are striving to improve the core melt frequency. This 7 was in the policy statement.
l 8l We have also adopted the suggestion of the ACRS that current plants should strive to achieve a mean value 9
10 for core melt frequency of 10 to the minus 4 per reactor l l
11 year for all but a few of the small reactors. j 12 In the case of not meeting one of the health
! O 13 i
effects, ucually the prompt fatality objective of .1 I
J l 14 percent, which was usually the controlling objective given l
15 today's source t.orms, not the new ones, we have proposed '
16 where the health effects are not met that there is some 17 .dditional cost incentive or cost penalty, if you wish to 18 look at it, or some additior.al cost element for safety 19 improvement.
20 DR. OKRENT: Don' t run away from that viewgraph i 21 yet. There may be some discussion.
l 22 DR. REMICK: Are you asking a question, Dave?
l 23 DR. OKRENT: I asked .'or a pause.
] 24 DR. REMICK: 11ay I go ahead while you are 25 thinking?
I l
ACE.FEDCRAL REPORTERS, INC.
_ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ . - - -J~FJ.D93 __ . __ -___ h ( t r r ~ L _----_-_ _ 0 3 m _-_- _ _ _ _ _ - _ ---
1350 02 12 26 0)DAVbur
(_ 1 I guess I don't understand why the staff has 2 proposed a range of the averted onsite costs. If I were a 3 licensee, I would read that and I would see 10 to 1 4 percent. I know what that means.
5 But I wonder why the staff put that range in 6 there and if they thought who decides whether it is l
l 7 appropriate to use 10 percent or 1 percent or 5 percent, or 8 whatever. What is the purpose of that?
I 9 MR. TAYLOR: Frankly, the purpose is to give an 10 emphasis to the onsite -- make it a priority -- the of fsite 11 health effects.
l l 12 This range of 10 to 1 percent tracks with the l
( 13 tange of core melt frequency. The higher the core melt 14 frequency, the more cost expenditures you would be expected 15 to commit to safety programs to do a little bit better to 16 prevent the accident.
l 17 DR. REMICK: So 8 times 10 to the minus 4, I get l 18 8 percent.
19 MR. TAYLOR: That is correct.
20 DR. REMICK: I see.
21 MR. TAYLOR: So it does track with the front-end, 22 the core melt frequency, with an emphasis on accident
' l l
23 prevention. l
(} 24 Now, it is clear -- and I believe Mr. Stello 25 advised you at the last meeting -- that the choice of these ACE. FEDERAL REPORTERS. INC.
I 202 347 3700 Nationwide Cm _800 3 4 6646 _ _ _ _ _ . _ _ _ _ _
1 1350 02 13 27 I
(~'sDAVbur eJ 1 fractions is arbitrary. He wanted to set those sufficiently 2 wide from the 100 percent to show that when the health 3 effects are not met there is an additional chunk of 4 expenditures expected. But he wanted to emphasize that 5 health effects are the priority.
6 7
8 9
10 11 12 13 I .
14 >
i 15 16 17 l
18 19 ,
20 21 22 23 24 25 ACE-FEDERAL REPORTERS, I?4C. .
- 02-347-3700
. Nationwide Coverare 80G.336-6686
1350 03 01 28
(~ l
()'DAV/bc 1l DR. OKRENT: Again, before you remove it, I'm not 2 sure whether all these estimates and dollar values 3 associated with a matrix like this, or some other matrix, 4 seem to be consistent in my mind.
5 But, if I understand correctly, for looking at 6 the report that EDO sent to the Commission, that, for 7 example, Palisades, or the case of unique safety goals, if 8 you're a 10 to the minus 3, your lifetime expenditure to fix 9
that is on the order of $7 million for averted on site 10 l costs.
I 11 Is that correct?
_s 12 MR. TAYLOR: I have an example. I can get into N
13 that. T'en to the minus 3 core melt? .
I l
14 : MR. SNIEZEK: That is correct.
l 15 l MR. TAYLOR: That is, if you meet the health 16 effects.
17 DR. OKRENT: Is that the same number you get?
18 ! DR. LAVE: Yes. I was using slightly other 19 , numbers, but I got $9 million. I was also trying to point I
20 ' out that when you then went from 10 to the minus 4 to 10 to 21 the minus 5, the total amount of money available to fix the 22 core would be $900,000.
23 DR. OKRENT: Now, let me just say that's a matter f1 G
24 of arbitrary engineering judgment.
25 MR. TAYLOR: It's somewhat arbitrary. But, ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-3%%t6 _ _ _
1350 03 02 i 29
('i I L_/DAV/bc 1 again, you will note, if you don' t meet the health ef fects, 2 we do have the thousand dollars a person-rem included.
3 DR. OKRENT: I have two different questions.
4 What does the term " meet the health ef fects" mean 5 operationally?
6 MR. TAYLOR: If we have a risk assessment result 7 that would show that the point one objective was not 8 achieved, the mean value.
9 DR. OKRENT: The mean value. Whose mean value?
10 MR. TAYLOR: Ultimately, it would be the staff's 11 judgment on whether that was the correct mean value.
12 DR. OKRENT: It would be the staff estimate of 7_
(_ I 13 ' the mean value.
l 14 MR. TAYLOR: Right.
- l 15 DR. OKRENT: And this would not depend on how big 16 the uncertainties around the mean value were.
17 MR. TAYLOR: Well, a staff decision would of I
l l 18 course depend on estimating the uncertainties. I 19 DR. OKRENT: No, no, the uncertainties would be 20 folded into the mean value.
21 MR. TAYLOR: That's correct. And the mean value 22 general biological uncertainty would be skewed somewhat.
i 23 DR. OKRENT: This is for all initiators except
() 24 sabotage; correct?
25 MR. TAYLOR: That's correct.
l I
l ACE-FEDERAL REPORTERS, INC.
u EW1st Dia T& $rmthC@nTro N
1350 03 03 30 r
Ly)DAV/bc 1 DR. OKRENT: All right. Now, let's assume that 2 that's the case, that you're able to do it. Cou'ld a 3 licensee challenge you in court and say they've also done a 4 PRA and gotten a mean value that was better? Or 10 times 5 smaller?
6 MR. TAYLOR: I assume our licensee can challenge 7 ~ our decision in court at any time.
8 DR. OKRENT: And that would just have to play 9 out?
10 MR. TAYLOR: Whichever way.
11 DR. OKRENT: All right. Why is it the staff's 12 judgment that, when you have a core melt frequency of the i
() 13 order of 10 to the minus 3 -- and, by the way, this doesn't 14 really matter whether you' re Palisades, I guess, or Indian I
15 Point or Limerick -- and let's assume your mean value, and I 16 assuming with the new source term outputs they' re going to 17 come in, you meet the safety goals, that's $7 million. This 18 is S85. So I assume we' re talking about what a plant costs 19 l that is finished in '85.
I 20 What does a plant cost that's finished in '85, 21 about? Average?
22 MR. TAYLOR: Three to four billion.
4 23 DR. OKRENT: Three to four billion.
f'i) 24 : Why does the staf f feel that 7 million is a good s- l 25 ' engineering j udgment on the most one should spend to improve ACE-FEDERAL REPORTERS, INC.
____ ____ ESM_M ______ _ .___._ ._._N@NC@!Ka 7 _____ _._ ___ _ QbM_
1350 03 04 31 I l
/DAV/bc 1 the core melt frequency, when the mean is as large as 10 to l
1 2 the minus 3? j 3 MR. TAYLOR: If the mean is as large as 10 to the 4 minus 3.
5 DR. OKRENT: Not larger.
6 MR. TAYLOR: As large.
7 DR. OKRENT: Because 9.9 times 10 to the minus 4 8 fits this; right?
9 MR. TAYLOR: Yes. I believe one should look at i 10 the regulations.
I 11 DR. OKRENT: I don' t understand.
7_, 12 MR. TAYLOR: That's in regulations. If a current
') : I 13 l plant coming on line today is comi*hg on line at a core melt l
14 l; frequency of 10 to the minus 3, then our existing I
15 l regulations are not sufficient.
16 That's my personal view.
17 DR. OKRENT: That's a long, hard road. I'm not 18 l quite sure where that gets us.
l 19 ! MR. SNIEZEK: Let me mention, I don' t think we 20 l should refer on site costs; just completely separately that 21 out from the of f site costs based on the man-rem. If you 22 note that that S7 million for 10 to the minus 3 would be 23 , combined with a high population site of about $26 million, I 24 to somewhere up in the thirties some place for Palisades, e'
25 you might only come out with S1 million.
ACE-FEDERAL REPORTERS, INC.
202 E %ffRD Natinnwide faverage 800-33 4 616
1350 03 05 32
()DAV/bc 1 And the Paloverde site would be less, somewhat 2 less than a million.
3 DR. OKRENT: Let's stay with the Palisades site, 4 which is about a median site, an average site. It's not all 5 that unpopulated. It's not way out in the desert.
6 I'm trying to. understand the engineering judgment 7 that decided 7 million. I assume, if you have used full 8 averted on site costs, which Mr. Lave, I think, has 9 suggested in his report to us, then I guess you'd get about 10 10 times, which would be 70 million.
11 DR. LAVE: Wait a' minute. May I interject.
12 DR. OKRENT: Any time.
G 13 DR. LAVE: I did the following calculation, which 14 l 1s simple enough so we can all understand'it. You tell me 15 what's conceptually wrong with it.
16 I figured that the cost of a TMI was going to be i 17 S10 billion. Obviously, that's not the cost of the plant.
18 It's the cost of makeup power, it's the cost of 19 decommissioning the plant, et cetera, plus a few billion l
20 dollars for population exposure.
21 I've got everything in there. And, now, if you 22 do the simple calculation, if you multiply $10 billion 10 to 23 the tenth by 10 to the minus 3, you get 10 to the 7 th.
{} 24 Okay?
25 So that's S10 million. If you do that simple ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide CoveQ _ 800 336-6646
1350 03 06 33
/m l
, b_)DAV/bc 1 calculation, you don' t get into S30 million or $70 million.
2 That's the whole amount that would be available if you use 3 100 percent of on site costs, plus a thousand dollars a 4 man-rem.
5 Back of the envelop, first order approximation, 6 that's all that there is. My contention was I cannot 7 imagine a utility around that would not be willing to spend 8 more than S10 million to lower the chance of a large core 9 melt. Certainly, GPU would, from 10 to the minus 3 to 10 to 10 the minus 4.
4 11 And that, furthermore, when you then went from 10 12 to the minus 4 to 10 to the minus 5, it became a trivial O- 13 number.
. ~
14 So unless I misunderstand something, we' re never 15 talking about more than $9-10 million.
16 DR. OKRENT: This is where I have some little 17 problem in the arithmetic. And I didn' t do my own 18 l arithmetic, but my understandit.g is that the S7 million on 19 page 31, Table F-1, is where you meet the safety goals.
20 l And, therefore, you're only using 10 percent, 9.9 percent of I
21 averted on site costs. And that they were using 100 percent l 22 and they would get 70 million; whereas, you can' t get 70 l
23 million,[
() 24 DR. LAVE: That's correct.
25 DR. OKRENT: That's a nontrivial dif ference. And i 1
l ACE-FEDERAL REPORTERS, INC. I S PD7- f g . . NyMTitth@verage Ci%33M M6 l
1350 03 07 34 7-)
( / DAV/bc 1 i
it would be helpful to understand why, if we could in one or 2 two or five minutes, to see what causes it.
3 MR. TAYLOR: I have with me Sid Fout here. We 4 can get into how we went about the cost calculation if you 5 wish to get into details. But I will run through some of 6 the elements. I can do that now. Some of the elements.
7 And I'll use the Palisades site.
8 I can do that for a core melt of 10 to the minus 9 4 here, for example.
10 MR. SNIEZEK: Did you take that for every year?
11 Did you take the plant lifetime into consideration, the 12 remaining 25 years, in your calculations?
)
13 That may be the difference.
14 DR. LAVE: Let me just do it again and tell me 15 what's wrong because we' re off by an order of magnitade.
16 DR. OKRENT: Was it per year in your
,17 calculations?
l 18 DR. LAVE: No. Let me just tell you again. Ten 19 billion dollars seemed to me what you might be talking about 20 for the averted on site costs, plus off site radiation 21 exposure for a TMI.
22 Now if you're talking about something 10 times 23 worse than that, then you've got to be talking about 10 i
/~S 24 million person-rems exposure to the population. And that's V
25 crazy. I don' t know where in the world that's coming from.
ACE-FEDERAL REPORTERS, INC.
202 3_47_-3700 _______ _ __._ Nation @ Cong p_____ G M D C2^3
1350 03 08 35 l
- k]-DAV/bc 1 Fifty million. So I don't know how you're getting numbers t
l 2 that are a lot bigger than that, unless, somehow or other, l
j ,
3 the costs of makeup power are 10 times what I thought they l
l 4 were.
i 5 MR. FOUT: I think the explanation is that, in i i l 6 the staf f's calculation, we're taking the integrated risk 7 over the entire remaining life of the plant. And what 8 you've essentially done is looked at the consequences of an 9 accident occurring in a given year times the risk of that 10 accident occurring in that year.
11 In the staf f's calculation, we' re assuming a i 12 30-year remaining life, and we are present worthing the on O 13 site costs and we are leavin~g the off site costs and man-rem l '
14 exposure times a thousand dollars in an undiscounted 15 fashion.
16 l Therefore, our estimate effectively, well, it l 17 essentially takes the consequences of over a 30-year life.
l 18 DR. LAVE: I don' t understand. So, at the same 19 plant, we could have a core melt every year?
20 MR. FOUT: We' re looking at the risks of an 21 accident occurring in a given year, and we're saying that 22 those risks exist in each of the remaining years of the 23 remaining life of the plant.
() 24 DR. LAVE: I guess I'm not understanding 25 ' something, since it seems to me that TMI-II is gone. l l
l l
ACE-FEDERAL REPORTERS, INC.
_ . _ _ _ _ _ _ _ _ JIR4XF4M0______ __-- facitfh@mrn - _ _ r m # w " "'i-----.--- ---- _- --__
1350 03 09 36 l n
is AV/bc D 1 DR. REMICK: Lester, let me try. The figures we 2 use are risks on a per year basis. And I want to protect 3 against the risk of an accident in the 30 remaining years of 4 the plant. That's why they' re taking 30 times that year.
5 MR. SNIEZEK: The lifetime risk.
6 MR. FOUT: It can be deduced that we' re taking 30 7 times the probability of an accident in one year.
8 DR. LAVE: Okay.
9 DR. OKRENT: So we understand, I guess, then that ,
10 difference. All right. Let me then come back to my 11 engineering judgment question.
12 In a S4 billion plant, which has a core melt A
'U 13 l frequency of 9.999 times 10 to the minus 3, what is the 14 engineering judgment that says about S7 million is the right 15 amount? -
16 MR. TAYLOR: And this would be for the case --
l 17 : DR. OKRENT: It meets both. As I say, I'm I
18 assuming it's a calculation, a source term calculation using 19 l such knowledge as you have now.
20 MR. TAYLOR: Given that healtn effects are a 21 priority business with the NRC and we do meet these, there 22 is some risk of core melt that probably is unacceptable to 23 the licensee as well as ourselves.
24 Seven million dollars probably doesn' t go a very U^ l 25 big distance to fixing that. Now, if we were to say we will ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverase 800-3 4 6646
1350 03 10 37
)DAV/bc 1 use as a decision criterion 100 percent of the averted on 2 site costs to fix that core melt, again, that is an 3 arbitrary decision.
4 We have made that statement up front. We used 5 the 10 percent to emphasize health effects. But also to 6 couple this to the core melt, make this again a primary 1
7 emphasis, albeit at a 10 percent factor.
8 Now, one can argue that 7 million is not enough.
l 9 DR. REMICK: Dave, there's something I don' t 10 understand here. I thought Jim said the 7 million was the 11 averted off site cost portion.
12 MR. TAYLOR: That's on site.
l O 13 DR. REMICK: Excuse me. Averted on site. And 14 i th,at there was the additional $28 million for health 15 effects.
16 MR. SNIEZEK: But the 28 million would be for a I
l 17 high population site like Indian Point. For Palisades, it l
18 would be about one million.
19 MR. TAYLOR: I can go through that.
20 DR. OKRENT: No, that's all right. So, again, 21 this is an engineering judgment for which you're entitled, 22 as is everyone else, to express one.
- 23 Now, there is a philosophy which I read in l
(} 24 . Mr. Lave's comments that you're including on site costs, and 25 you don' t use 10 percent or 1 percent, or something. You ACE FEDERAL REPORTERS, INC.
ESD3NED 9ttm~c'bh N
l l
l 1350 03 11 38
! l
._ /DAV/bc 1' just make it the frequency per se.
2 In other words, this is a philosophy that all 1 3 cost benefits are practical to attribute, which, in effect, !
4 was the recommendation of the c riginal task force, or j 5 whatever you call it. I 1
6 MR. TAYLOR: The task force concern group l 7 recommended including the on site economic costs. Now there 8 i is the iscle of displaying the on site economic costs and 9 what extent or what practical decision rate you give to 10 those. l 11 DR. OKRENT: I agree. There are different ways 7
12 in which one can do this. But, right now, the staff is
'~
13 proposing a recipe with which I think it expects to l 14 influence strongly.
15 It says now " generic issues". It will be also 16 site specific issues. l 17 MR. TAYLOR: Yes. l 18 I DR. OKRENT: Using the bottom line of these --
19 now let me come back to that 10 to the minus third again, so l
20 you have the leverage of the order of S7 million, which, in 21 j fact, at some plants may go a long way. If you put enough 22 system in all of them and seismic is not a problem, that 23 might do it.
24 MR. TAYLOR:
( There is of course a man-rem 25 difference.
I ACE FEDERAL REPORTERS, INC.
202-34'.3700 Nation *ide Coverage 800 336-6646
1350 03 12 39
(')
( / DAV/bc 1 DR. OKRENT: Now you have what I would call 2 another philosophy that's been expressed in a couple of 3 different ways. One I could call the Minogue philosophy. ;
4 4 If you want to work up to 10 to the minus 4, forget cost 5 benefit. We have the Bernthal-Asselstine concern that we 6 really can' t af ford even core damage, let alone a large-
, 7 scale core melt with the kind of frequency that the staff 8 roughly estimated.
- 9 You've got to get some factor, whatever. The 10 ACRS, in fact, in July said we think we should work toward a 11 10 to the minus 4 mean for existing reactors. You could say 12 you are working toward it.
O 13 MR. TAYLOR: We have the language in there.
14 DR. OKRENT: But it's not clear how hard you' re 15 working toward it with this matrix. Do you see the 16 difference?
17 MR. TAYLOR: I understand.
18 MR. SNIEZEK: Dr. Okrent, I think you' re right. j i
19 That is a debatable issue. And I think, when this matrix 20 was first evolved, that Mr. Studall indicated that these 21 were very arbitrary numbers.
22 It was picked this way for a reason, using the I
23 decade approach, which gives primary emphasis to health 1
i
() 24 i effects. And then somehow still sways emphasis towards 25 movement to core melt for esisting plants, decreasing it 1
ACE-FEDERAL REPORTERS, INC.
_ -_ ____JMP35D_ _ - -_ __ Pot -W@'n~rn _ W-
1350 03 13 , 40
)DAV/bc 1 somewhat for existing plants.
2 And then, right up front, you could argue whether ,
3 it should go from 100 percent to 10 percent, you know. I 4 Everything would be one decade higher in those percentages.
5 It's hard to j ustify. We show that in one of our 6 figures that we have there.
l 7 DR. MACLEAN: Could I ask one question?
8 DR. OKRENT: Please go ahead.
9 DR. MACLEAN: In your example, it's right very i
10 close to 10 to the minus third. You are assuming that the l 11 health effects are met.
12 And, in that case, it's 7 million. If the health O 13 effects aren't met, then it looks like it jumps immediately I 14 to the 70 million. l l
15 DR. OKRENT: Right. '
16 DR. MACLEAN: That makes it less serious.
l l 17 I
i i
18 .
l . 19 l
l 20 l '
21 22 i 23
(} 24 25 l
ACE FEDERAL REPORTERS, INC.
DSC7 DE- NW C
1350 04 01 41 I
m JDAVbur 1 DR. OKRENT: Let me talk a little about health 1
2i effects. The staff said they would be the ultimate arbiter 3 of the need.
4 In my opinion, the ultimate need that the staff 5 arrives at is subject to some rather arbitrary choices by 6 the staff. They can truncato distribution. They can say we 7 really think the uncertainties are less than those 8 professors from Columbia said on seismic.
l I
9 In fact, I see there is a letter from Mr. Ernst !
i 10 to Mr. Hickman, who talks about treatment of uncertainties, 11 sensitivities of SARP and NUREG-il50, and he goes on and l - 12 ,, says, look -- I will paraphrase it -- you have really got to l l )
13 be a little bit cautious in how you treat all those 14 uncertainties and be somewhat pragmatic and put in, if you !
l 15 l will, your best objective judgments, not necessarily l 16 l calculate what I would call a full examination of uncertainties.
17 l 18 f If I have done you an injust'.ce, Mr. Ernst, fool I
19 l free to stand up and correct me. But as I read that memo, 1
20 i it sounded to me that you wanted Sandia to temper 21 appreciably what they might normally put in as their 22 estimate of uncertainties in the calculations that they 23 would be reporting.
-~ j
( '1 24 MR. ERNST: I will alt down and comment.
.J l
25 DR. OKRENT: Sure.
ACE FEDERAL REPORTERS, INC.
rio-t m Nen,M c~ tm sm im a
_. . . --- . .. = _ - - . - . _ - - .-. _ -- .-
l 1350 04 02 42 !
n L_/DAVbur 1 MR. ERNST: I think it is perhaps not quite l
2 paraphrased properly. What has to be taken in context, what 3 had been going on at Sandia was basically there is a number l i 4 of parameters that have uncertainty bounds around them. If :
i 5 you really wanted to sort it out, there's probably dozens or '
6 hundreds, particularly in the containment event tree, 7 whether they be 60 or 70 branch points, each one with maybe 8 multiple branches, of how the event could go.
9 What had been happening up to that point is there 10 had been what people thought might be reasonable ranges for 11 each of these branch points, and then in the sc-called 12 uncertainty analysis what they are'doing is walking through O ' 13 in t,he optimistic / pessimistic central pathw&ys to the 14 containment trees, which in my view meant you are really 15 taking perhaps each branch being reasonable but you are '
16 I really multiplying, for example, the pessimistic branch 17 point for each parameter together to come up with an overall 18 uncertainty range.
19 It seemed like that was undue pessimism, and if 20 you tried to put that in a 5/95 percentile kind of context, 21 you had the potential for greatly overstating the range of 22 uncertainties.
23 What is happening now is what is commonly called j 24 a limited hypercube approach, limited meaning selecting the 25 10 or so dominant uncertainty parameters that drive the l
l ACE FEDERAL REPORTERS, INC.
____ E MM*_ -------- NaionC@n~Q - _--.- ----- N-
1350 04 03 43 G
L_)DAVbur 1 uncertainties and using the stratified approach to sampling, 2 developing the uncertainty bounds using that scheme. ,
3 That is what that letter is all about.
4 The outcome we will see, because we are doing the 5 optimistic / pessimistic walk-throughs also just as a 6 comparison to see what difference this approach makes versus 7 the other, I think your bottom line is correct. It will 8 carrow the uncertainties compared to the other approach, but 9 I think appropriately so.
i 10 DR. OKRENT: By no means do I want to be 11 identified as on the side that takes extremes on the one 12 hand pessimistically and then extremes en the other hand
~
I 13 ' optimistically and just combines them all into a paper. '
14 Each has a probability of 1 of occurring. That is not a ;
i 15 thing I wish to be associated with.
16 ! But what you do say suggests a little bit of 17 pressure. It says since the imposition of our state of 18 knowledge and many of the parameters involved makes a 19 statistically preciso analysis unobtainable except in an j 20 extremely subjective sense, we seek reasonable engineering
- 21 approximations.
l 22 By the term "roasonable" we mean that the 23 judgment of the team and others -- that the true answer 24 likely lies within the specified range.
(}
25 Additionally, although the consensus of belief ACE. FEDERAL REPORTERS, INC.
- E!MM----- MM -- EM---
- -= . . _ _. . ____ _--_ _-. - - _ . . . _.
1350 04 04 44 rx L_sDAvbur 1 is not quantifiable in such a manner, that a reasonable 2 range should reflect the individual degrees of belief that 3 are neither unreasonably narrow, 70 percent, or unreasonably 4 wide, 99 percent.
l 5 This judgment, by its very nature, is i 6 subjective. Obviously, different analysts can differ on the l
l 7 range defined.
8 And then you go on.
9 It is hard to tell for me whether you are asking 10 them to do the best they can mathematically, given the 11 uncertainties, or to get what you call reasonable 12 engineering approximations, whatever that means.
l *13 I can' t tell when I read that memo.
. 14 MR. ERNST: Two comments, Dave.
i l
15 One, I think the word " pressure" I have to react i 16 to.
l 17 ! That letter was a process that took about two to 18 two and a half months to generate. It represented I
19 numerous -- and by numerous I mean four or five meetings, l 20 telephone calls, exchanges of draf ts between us and Sandia.
21 My perception is that Sandia appreciated the 22 exchange, fully support the letter, and I tried to make sure 23 that that was the situation before I signed it out.
{} 24 I think it represented a very dedicated and 25 useful hunt for some kind of process that would come up with I
I 1
ACE. FEDERAL REPORTERS, INC. .
202-347-3700 Nationwide Coverage 800-336-6M6 l
1350 04 05 45 r^s l L_jDAVbur 1 reasonable uncertainty bounds without getting too terribly 2 . enmeshed in the statictical preciseness of those bounds, but 3 some kind of a reasonable process.
4 I think it is a good letter. I think it 5 represents some work by Sandia and also that this is a good 6 letter.
7 I forget what my second point was right offhand.
8 I think the ranges of uncertainty will still be wide, and 9 let's say I think it does represent a reasonable process.
10 Oh, I know what the other point was, and it is a 11 point I wanted to make later on, also.
12 What one generates in this process is what I call 13 a range o,f means, which is a lot different than a simple 14 p'oint mean value with 5 and 95 percentiles on the i
15 distribution.
16 So it is a point I want to make. I was asked by 17 l Jim if I had any comments on this proposal, and I do have 18 some comments.
19 One of the comments that I had was there is 20 i substantial pressure -- I think I am using the word 21 properly -- by some to come up with a mean value, a single 22 value that you plug in to the safety goal kinds of 23 calculations, which tells me that the agency could well
{} 24 drift toward point values. l 25 I hope it doesn' t happen either, but I have seen ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Com 800 3fW""3
t i
i i
1350 04 06 46 i i
i
-( )DAVbur 1 the agency. It is a very deterministic, mechanistic kind of 2 a process many times when the agency gets working on t
- 3 problems.
I 4 There is the potential, is all I am saying, to i 5 take these numbers and make decisions based very largely on 6 the numbers. In this case, the narrative about accident j 7 prevention could be lost.
l 8 I do have that kind of a concern, and while 9 NUREG-ll50 is going to come up with a range of means, which !
10 one then could arguably say that I will take the upper 11 bound, the more conservative, as a regulator I will take the 12 more conservative value and use that as my mean for thore i 13 calculations. If that happens, I' don',t think I would have 1
) .
14 any serious problem.
! I i 15 If one, however, tries to develop the mean of 16 that range of means and use that, then I would have a lot 4
i 17 more concern.
18 MR. TAYLOR: I would just remark that the policy
}- 19 statement acknowledges that there are uncertainties that 1
l 20 need to be addressed and as part of addressing these, one is j
21 flawed. They must make an assumption in terms of the
- 22 distribution they use.
j 23 Typically, these are skewed. Typically, the 1
l 24 process itself tends to bias the mean in my observation.
i 25 I would agree with Mal that we shouldn' t J
ACE FEDERAL REPORTERS. INC.
202-347 3700 _ _ _ Nationwide C5Gep _ 800 336 % 46
1350 04 07 47 l
/T ksDAVbur 1 operate solely, singly, and to the exclusion of judgment on 2 single values.
3 MR. SNIEZEK I think there is no question that 4 there can always be misuse of information, misuse of the 5 process. That is one of the things you have to guard 6 against continuously, regardless of what the process is.
7 DR. REMICK: Dave, I have a couple of questions 8 on understanding the matrix.
9 Matt, if I understand, if I look at that bottom 10 box down there, core melt frequency of 10 to the minus 3, l
11 and I meet both health effects there, I would have the $1000 i
l ,_
12 a person-rom plus 100 percent averted of fsite costs.
( )
13 ; So if I have 1.'1 times 10 to the minus 3, I am in I
14 i that zone. -
15 If my core melt I calculate to be .9 times 10 to 16 the minus 3 and I don't meet both, am I correct that if I go l
17 i from 1.1 times 10 to the minus 3 to .9, I then have the l
18 ' S1000 per person-rem plus 10 percent?
1 19 MR. TAYLOR: There is a steep jump there. There 20 , is no doubt about it. A steep jump.
21 But I would say the mean values of 9 times 10 to l 22 the minus 4 and 1.1 times 10 to the minus 4 are quite 23 subject to conservative interpretation by the staff, I am I
(~) 24 ' sure.
v 25 DR. REMICK: Another question I havet if I ACE-FEDERAL REPORTERS, INC.
i 1350 04 08 48
(~) DAVbur
(,/ 1 recall, you decoupled the $1000 per person-rem away from 2 inflation in this NUREG-0880, December 1983 or January 1984, 3 but I think you dropped that in this proposal, is that l
4 correct?
5 I was wondering why.
6 MR. TAYLOR: The secondary influence? I l
7 MR. SNIEZEK: The $1000 per man-rem in inflation l 8 has not changed. It is the same as it was originally. We l 9 did not touch that.
10 DR. REMICK: I thought it said in 1983 -- l l
i 11 DR. GRIESMEYER: That was crossed out in the 12 comparable policy statement.
o 13 It had the new stuff underlined and the old stuff crossed out. The adjustment l l
14 for current dollars was crossed out.
15 -
MR. TAYLOR: We certainly could do that, the 16 adjustment of 1.12 or maybe -3 at present.
17 DR. REMICK: I was wondering -- unless it was 18 just an accident -- did somebody scratch it out? I was 19 curious about the reason why. in a year or two it doesn' t 20 matter. Ten years from now it will make a big difference in 21 that $1000 per person-rem.
22 I thought maybe you wanted to make it consistent 23 with Appendix I.
l
(} 24 MR. SNIEZEK: The S1000 per person-rem, the l
25 philosophy regarding that, we did not intend to change.
l l
l l ACE-FEDERAL REPORTERS, INC.
E3D37 DO ______9ttrCh r __ _ _ . _ N -- ---- ------ _ -
m- -.
1350 04 09 49 DAvbur 1 DR. REMICK: That I understand. But what year's 2 dollars? Did you change that?
l
! 3 MR. SNIEZEK: We converted to '85 instead of l 4 '83.
l 5 DR. REMICK: Well, '83 was the floating dollar.
6 Excuse me, in NUREG-0880 it was the floating. l 7 MR. TAYLOR: Yes, it was floating with 8 inflation.
9 Da. REMICK: I sure got the impression from 10 reading this -- and Mike did, too -- that you changed this 11 and now it is fixed dollars.
12 MR. TAYLOR: I don't think there was any intent.
O 13 ' We did inflate '83 to '85 fn our analysis. Maybe perhaps we i
14 i should put that in there.
15 DR. REMICK: I.think what you have got is changed 16 wording.
17 MR. SNIEZCK: It was not the intent to change the 18 philosophy.
19 MR. TAYLOR: We didn' t address inflating the 20 onsite costs either. I think it should say current 21 dollars.
22 DR. GRIESMEYER: Getting back to the first point 23 that Forrest mentioned, also relating to what Mal said, when
(]) 24 I we come up with these estimates we are trying to get a range 25 of means. It may not be fair to call that range of means ACE. FEDERAL REPORTERS, INC.
i ___ __ _ -__ __--_ _-----_ _ _ _ f M IP-E G_--- ------- h CM n n ------ -- e -
- .~ . - _ . . - _ - -
1350 04 10 50 DAVbur 1 a distribution of probability distributions, just kind of 2 calculating some averages of the distributions to get tho 3 mean anyway.
l 4 But there is going to be -- in the containment 5 performance there is going to be one or two orders of 6 magnitude at the very least, and probably in the core melt '
7 frequency there is at least one or two orders of magnitude 8 in the mean.
9 If there is going to be that much uncertainty, i
10 well, at least one order of magnitude -- [
11 MR. TAYLOR:' In the mean?
12 DR. GRIESMEYER: At least one order of magnitude O 13 in everything that I have seen. When we st; art sampling, l
14 when you take the base events and y'ou sample them, what were i 15 their ranges? You have got an order of magnitude.
- 16 If that is the case, then using any number with 17 an order of magnitude or two orders of magnitude uncertainty 18 in a decision matrix might be a little irresponsible, even 19 if you say that it is your responsibility to choose between 20 them.
21 So I am kind of uncomfortable with calling this a 22 decision matrix because you are not going to maka decisions 23 based only on this matrix because you are going to have the 24 other decision factors coming in.
25 So it in really a perspective matrix. It is l ACE FEDERAL REPORTERS, INC.
! E30%D m CrtntCO2co - CDOcc .- - - ---- _--. ---- - --
1350 04 11 51 DAVbur 1 going to add some perspective, and you are going to use it 2 somehow along with these other unstated decision factors.
, 3 I would be more comfortable if I saw also in this 4 matrix another table of what some of those unstated decioion f
5 factors are.
l 6 MR. TAYLOR: To account for two orders of 7 magnitude uncertainty? I am not sure this matrix will l 8 accommodate that. I just don' t know how to handle that. '
9 DR. LAVE: I think there sre two issues on the ,
10 table. Let me separate them. !
11 One of the issues on the table is that the f 12 probabilities are inherently uncertain, ano you are going to o
l l- . .
l 13 make judgments.
14 i Well, the world is like that. So you are alssya l !
i 15 going to make judgments for things that are uncertain, i i
16 sometimes by two orders of magnitude, but that is what you 17 have to do.
1 ,
18 The other one is to have a cliff built in there
! 19 which goes from 100 percent to 10 percent when you have got l 20 l an order of magnitude uncertainty, and that seems to me to 21 be just terrible.
22 ' What you are doing is you are setting yourself up 13 to be shot, and you are going to pull the trigger, because what is going to hapnen is somebody is going to come in --
JJ one of the vendors . one of the utility operators -- and is ACE FEDERAL REPORTERS, INC.
____.__ _ E~MS DO _ _ ____ __ Wm@m__ ______ flNEP"t
l q
1350 04 12 l 52 DAVbur 1 goinc to say, you mean to tell me that that really can't be i 2 .999 times 10 to the minus 3 instead of 1.00? And the only I 3 answer anyoody could ever give is, sure, it could just as 4 likely be tuat, in which case you may wind up with a 5 dif ferent decisiun.
6 And if you think you can insulate yourself that
! 7 much from either public comment or utility comment, you ere 8 crazy. Don' t build in that kind of cliff.
l 9 :1R. SNIEZEK: I don't think that is the intent, 10 to insulate ourselves. This is the philosophy that the 11 staf f will be using in helping it to arrive at decisions, 12 and if there is great uncertainty where it could os one way l
( 13 or another, the staff has to take that into account in 14 arriving at what type of actions we want the licensees to l
l 15 ' take.
t l 16 But you need a guiding philosophy. If you put l
17 too many caveate on it, you don't even have a philosophy.
j 18 DR. LAVE: But what I am going to argue as soon 1
19 as we get cut of this particular presentation is that you 20 :sught not to have these kinds of cliffs built into your 21 decision matrix, that it ought to be simply 100 percent 22 averted onsite costs, period, without the sliding scalu.
23 You are basically setting yourself up by having 24 those cliffs built into the scale.
(])
! 25 MR. SNIE2EK: There is no question that that is ACE. FEDERAL REPORTERS, INC,
___ _fBD7 FED _ ___ ___ _N --_ M M -___--___ -
l l
l 1350 04 13 53 I L / DAVbur 1 a view that some people could reasonably take.
i i
2 DR. REMICK: Jim, to go back to my comment before 3 to help you out, on page 21 is where you have scratched out 4 "This value should be modified to reflect challenges i n t?.to l 5 future." That is stricken out. i 6 MR. SNIEZEK: It wasn' t meant to change our I 7 philosophy as far as I know.
8 MR. TAYLOR: We meant to sta:e current dollars, i 9 l 10 i
l1 l 1
cs 12 ,
( ; i l
' l 13 l I
14 .
l 15 16 I l
17 I
18 l i
19 l I
20 l 21 22 23 c,
(
24 ,
o i l
25 i ACE FEDElb\L REPORTERS, INC.
202 mmm wmm u, cmem- om-
1350 05 01 54 1 DR. OKRENT: Would you please remember that?
{}DAVbw 2 Let's go on. I w..
1 3 MR. SNIEZEK: Put the matrix back up. I want to 4 make a couple of points that haven't been made, a couple of 5 points.
i 6 One, greater than 10 to the minus 3 and don't 7 meet a health effect, means basically you don't operate.
8 You fix it, before you operate.
9 The other point I want to make, right on the top j 10 of it, less than 10 to the minus 5 core melt, you meet both 11 health effects. You're basically home free, as far as NRC 12 backfitting, et cetera, you're done. That's where we would ,
() 13 expect future plantn, standardized plants to be less than 10, 14 to the minus 5, meeting both health effects.
15 DR. OKRENT: This range means that Mal was
)
16 talking about, at the moment I am not quite sure how he is 17 j going to develop the range, but let me assume there exists a 18 range that comes out of the Staff's analysis. How will you 19 judge that information? I can't tell.
20 MR. SNIEZEK: We don't have detailed procedures l
21 worked up on how to make all the judgments at this time. I ;
22 }
say this is a philosophy, a lot of detailed implementation '
23 work has to go in to even implement a matrix like this. No 24 question about it.
O 25 DR. OKRENT: If the range is as big as Griesmeyer i
ACE. FEDERAL REPORTERS, INC.
_____--_-----_----------- SM-DR3 . - - N - - -- -- N
i 1350 05 02 55 l
l
, DAVbw I suggests, I don't know if it is 10 each way.
2 DR. GRIESMEYER: Well, at least a total of 10.
3 s DR. OKRENT: It could certainly swing you out of 4 one category into the middle of another one.
5 'MR. ERNST: I don't want to try and prejudge what 6 the procedures might be. As Jim says, they haven' t been 7 , developed. My basic purpose for coming up with a range of 8 mear.s in NUREG 1150 is that one can get a better idea of 9 making different basically phenomenological kinds of 10 assumptions. One can get an idea of which of these 11 assuhptions drive the uncertainties, the most, then one can, 12 in a regulatory manner look at the various assumptions that n
() ,
13 are important and make a regulatory judgment about, well, do 14] I want to be that conservative or not? What is my belief as 15 to that particular assumption? So my intent is, basically, 16 to identify the drivers that are important to risk, express 17 these in terms of ranges of means. If you make different 18 assumptions on the phenomenology or whatever is driving and 19 then display the drivers so that people can visibly 20, understand and make decisions and not just make decisions l
21 based on what I would call a mean point estimate or 22l something like that.
23 DR. OKRENT: If I understand correctly, do you 24 anticipate that this range of means may largely result from 7~
25'
, differences in assumptions about phenomenology?
~
l For s
ACE-FEDERAL REPORTERS, INC.
- p;447 3700 Nationwide Coverage 20-31MM
1350 05 03 56
(] DAVbw 1 example, is it important that you might get sudden large
\J 2 scale dispersal of the core melt in the containment?
3 MR. ERNST: Direct heating? Yes. That will be 4 the principal kind of drivers. The things we have knowledge 5 about on distributions will already be treated in the normal 6 statistical fashion.
7 DR. OKRENT: All richt. That helps me.
i l 8 DR. REMICK: I have a nit question. Why did you 9 mixup -- well, it appears a mixup to me. You have l
10 prompt / latent at the top of that column that we usually talk 11 about prompt and delayed and acute and latent? Why did you 12 pick those two?
I- O- 13 na r^Yi o = oddtv .
ta e we et ene uooe tio= or 14 Dr. Okrent. We did have individual and societal there 15 originally.
l 16 DR. REMICK: I think you'd probably get to 17 societal perhaps.
18 MR. TAYLOR: That is latent mortality risk and l
( 19 prompt mortality risk.
l 20 DR. REMICK: I understand latent might be used 21 that way, but isn't it more correct to talk about prompt and l
22 l delayed or acute and latent? It is just a trivial point. I 23 just throw it out.
24 DR. OKRENT: I have no comment on this.
I l
25 MR. SNIEZEK: Delayed or latent; either way.
ACE. FEDERAL REPORTERS, INC.
_ mcsem m l
1350 05 04 57 1' DR. OKRENT: Maybe we will let you go by. this
(^}
s- DAVbw ,
- 2 one.
i 3 (Laughter.)
4 It has been suggested that we take a short 5 break. Short, beino my suggestion, since we have until 6 11:45, why don't we get back in seven minutes. All right.
{
3 7 (Recess.)
8 DR. OKRENT: Okay, Matt. Maybe we'll let you go 9 through the next one.
10 (Slide.)
11 MR. TAYLOR: I believe this slide addresses the 12 objectives or the parameters, if you will, that are
() 13 reflected in the integrated matrix.
To a large extent, this 14 parallels the quantitative objectives that were set forth i .
i 15 previously in NUREG 0800 and the previous policy statement.
I i 16 In the interest of time, I won't go down and read I 17 those, but prompt and latent cancer fatality objectives of
]
- 18 0.1~ percent, as in the past, we have, as noted in the 1 !
19 matrix, we have a sliding scale on the core melt frequency, 4
20 with impetus to improve. We also have a sliding scale on 21 benefit-cost guidelines that are coupled,to the core melt i
j 22 frequency.
l 23 DR. OKRENT: Before you take that one away, on i-'
24 the latent cancer fatalities, as it is now defined in your
() 25 document, if I recall correctly, it is proposed that this ACE-FEDERAL REPORTERS, INC.
__ _ _ _ _ _ _ ____ _______ _ _ _ r arn vila _____ _ ___ m .m , 1_ ._ _ _ _ _ _ .___ _ m
1350 05 05 58 1 be calculated out to ten miles, and after you get a number,
({}DAVbw I 2 you compare it as a 10 of a percent of the normal risk from 3 cancer.
}
4 MR. TAYLOR: That is correct. Previously, it was 5 50 miles.
i 6 DR. OKRENT: I have seen some differences of 7 opinion as to whether, when one does this, one has 8 individually cancer fatality risks, some kind of societal 9 latent cancer fatality risk or something that is neither 10 fish nor fowl, as it were, somewhere in between.
11 What does the Staff think it is, when it 12 calculates it that way?
() 13 MR. TAYLOR: Whatever it in, it's that at all 14 distances.
f 15 DR. OKRENT: Say that again.
I 16 MR. TAYLOR: Whether the integral distance is 10 l 17 miles or 50 miles, whatever it is, it is intended to i 18 represent an average.
19 DR. OKRENT: It's an average over some distance.
20 MR. TAYLOR: It's an average over some increment 4
l 21 out to that distance. I have Tim Margolies here, and as a 4 22 matter of fact, I asked him that particular question. How
! 23 do we do that averageing as a function of the interval cut
! 24 to the 10 miles or 15 miles. I think we averaged in the
! (:) 25 radius. We average one and we go out and average another i ACE-FEDERAL REPORTERS, INC.
______ rmPVAJEM_______he.. , t__ h _ __
1350 05 06 59 fl DAVbw 1 one, and we go out and average another one. So we got an
%J 2 average of functional distance.
3 Tim, would you speak to that?
4 MR. MARGOLIES: One way to think about it is, it 5 is the total number of fatalities out to some distance, 6 divided by the number of people that reside within that 7 area.
8 DR. OKRENT: This is my understanding.
9 DR. MARK: For that calculation, are you going to 10 take the U.S. national average of normal cancer fatalities, 11 or are you going down to Sun City, Arizona, where people are 12 all over 70 years old and they all have cancer? What are
- () 13 you Going to do?
14 MR. TAYLOR: U.S. national.
15 DR. MARK: You are going to take the U.S. average 16 and what will you use Baer 3 data? There is the average 17 cancer. You can establish that. If you take a countrywide 18 statistic, when you talk about rem or millirem or something, 19 you can make an assumption. You have no basis whatsoever 20 f,or knowing what you are talking about. For the number of 21 cancers which result from so many rem, what are you going to 22 do, use Baer 3, high, low, middle, or what?
23 MR. TAYLOR: I think perhaps Tim can best address 1
24 this. My understanding is that we do have, again, in the ,
1 CE) 25 CRAC-2 model, we do have a weekly dose-dependent 1
)
ACE FEDERAL REPORTERS, INC. !
_ _ _ _ _ - _ _ _ ____-___-_____- M ________ M _________ M _ ____ _ _
2
- 1350 05 07 60 1 relationship for cancer fatality rates. For example, at low f'}DAvbw v
2 does, one would have a low hazard rate. At higher doses, a 3 higher hazard rate, similar to an S curve approximation.
4 DR. MARK: CRAC-2 is a total construct which has 5 no particular relationship to any actual fact, unless you 6 feed data into it or some assumption into it.
7 What are you going to do? Are you going to use
'8 the Baer 3 linear quadr atic or the Baer 3 linear, or what?
9 MR. TAYLOR: It approximates the linear 10 quadratic, as I recall.
11 Tim, could you speak to that? Now there is an 12 effort under way to update the model right now.
() 13 MR. MARGOLIES: I think there are two issues.
14 You are saying what is the level which one uses to make this 15 comparison, in terms of background cancer risk. This is a 16 number which would come from vital statistics, national 17 statistics. That was the first thing you had asked.
18 DR. MARK: That is the denominator.
19 MR. MARGOLIES: Right. And now you are switched 20 over to this idea of, hey, how does one actually calculate 21 now, a number of latent cancers in actually implementing 22 safety goal. There are a lot of guidelines I would like to 23 suggest would have to be developed with regard to the 24 details, the calculations that we have performed to date to O 25 provide some perspective on the sensitivity of some of these ACE-FEDERAL REPORTERS, INC.
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ ----- f~WFMF70 ____ _ ____ h q 1 , t__ _ _ - - _ - - N________________ __
1 _ _ . - - - . .- _ ._- .. ._ _._ _ _ _ _ . _ _ .. . _ . __ _ . _
I i
l 1350 05 08 61 1 measures to risk, is based on our state of the art models,
{}DAVbw
! 2 and they do use basically a linear quadratic type of 1
3 relationship between dose and health effects.
l 4 You must be aware that there is ongoing now the 1
5 development of a new set of health effect models,'and in 6 fact, these are being reviewed, I think, by several of the 7 other committees. At which time that review'is complete, we 8 will incorporate any new revisions in our models. ;
9 DR. GRIESMEYER: You mean into CRAC-2 or~into 10 MACS, the new one? The new health effects model.
11 MR. MARGOLIES: Either. But right now, MACS is 12 our MELCOR off-site accident consequence model. That is our
() 13 intent now. F6r the future, it is to have much more 14 efficient setup models. This is where the new health !
1 l
15 effects models come in.
16 DR. GRIESMEYER: Another point is, in the MACS 17 model, when they have gone to the new health effects models, 18 they have considered that the person-rem was an extremely 19 crude thing, because it didn't loch -- they don't even use !
20 that unit any more. They icok at doses to specific organs 21 and calculate effects based on the doses to specific 22 organs. So they don't even use the person-rem any more, 23 which may make it a little bit awkward in the safety goal 24 policy.
O ~25 MR. TAYLOR: Is thet9 any reason why that could ACE-FEDERAL REPORTERS, INC.
I__________________ ________EMPRETL _ _ N ____ _ 030X ">________ _
i 1350 05 09 62
/ T DAVbw 1 not be translated into at least an equivalent man-rem?
V 2 DR. GRIESMEYER: What they do, when they are 3 asked to do it is, they calculate the bone marrow dose, and 4 it is roughly equivalent to the old person-rom, but it is 5 much more easy to relate bone marrow dose to a cancer than 6
it is person-rem, because the person-rem doesn't even tell 7l you what organs get it.
8 MR. TAYLOR: My understanding of the original 9 CRAC model also summed up all the doses. One would have to 10 still take account of the groundshine effects, and what have 11 you.
12 DR. GRIESMEYER: Right. But the calculated
() 13 person-rem depends upon a whole bunch of assumptions as to 14 which get puffed out, where they go and what pathways you 15 get and how they get into the body. I say I have this 16 person-rem, and it's got all those assumptions embedded into 17 it. You may want to change some of those assumptions, and 18 then the average effects per person-rem kind of never would 19 change.
20 MR. TAYLOR: That is possible that the health l l
21 effects modeling could cause some change. There is no doubt 22 about it. I do know that there is an attempt to upgrade the 23 .model, health effects model. My understanding is that that 24 is in process right now. The recommendation of, I believe, O 25 Dade Moeller's group. He is a key figure on that.
ACE. FEDERAL REPORTERS, INC.
__ _ _ _ ESN-DED _ __ __ N~f~rb CoC CBRDC"3
1350 05 10 63
[DDAVbw 1 DR. MARK: It seems to me that there is a real w/
2 need somewhere here for a statement in the implementation of 3 the gaa.j.a or whatever, to specify the way to which this is 4 going to be assessed. CRAC-2 is a very vague way of saying 5 anything. Baer 3 is pretty vague, and there will be CRAC-3 6 and Baer 4. And one will need, it seems to me, we're going 7 to assess this in the following manner.
8 MR. TAYLOR: I would agree that one does need a 9 certain amount of detail, procedures and descriptions.
10 DR. MARK: Not detailed, but just 11 specifications.
12 MR. TAYLOR: In the policy statement itself. ?
) 13 DR. MARK: I believe so.
14 DR. OKRENT: Could I ask just a question about 15 this. It suggests that the latent cancer fatality risk is 16 flat as you go from one mile to three miles.
17 MR. TAYLOR: This curve would seem to suggest it 18 is flat. It is reasonably flat.
19 DR. OKRENT: Why is that? Are people evacuated 20 faster?
21 MR. TAYLOR: I can't tell you the exact l 22 contribution, but this casi be affected by the heat of 23 release, which ecsentially moves the concentration out and 24 downwind more onto people that would be further downwind.
25 You are sensitive to emergency responses close at hand.
ACE. FEDERAL REPORTERS, INC.
_1 s va n - _m__-__-___
'6 1350 05 11 64 1 You are sensitive to, again, the heat of release,
('}DAVbw v
2 the plume height and what have you.
3 DR. OKRENT: What wculd you estimate as the 4 difference between averaging over 10 miles and averaging 5 over one mile?
6 MR. TAYLOR: About a factor of, I think, 5 to 7.
i 7 I think that is something Mal mentioned.
8 DR. OKRENT: All right. Let me get back to my 9 original question.
10 What do you call this risk? Is it a societal 11 risk? Is it an individual risk? What do you call it, the 12 Staff?
() 13 MR. TAYLOR: It is really our representation of 14 the average individual risk. It is an individual risk.
15 , DR. OKRENT: You told me that for the people 16 living within the first mile, the risk may be 5 or 7 times 17 greater.
18 MR. TAYLOR: That is correct, or it possibly 19 could be much lower, if the plume height were such.
20 DR. OKRENT: But averaged over this spectrum of 21 scenarios.
22 MR. TAYLOR: Sure.
23 DR. OKRENT: Okay. I won't belabor the point 24 now. It is just a little bit --
)
25 MR. TAYLOR: Again, this was a judoment, ACE-FEDERAL REPORTERS, INC.
__ 202 347 3700 _ _ _ _ _ _ _ NatinnwideW 800 336 6646
1350 05 12 65 f]/,DAVbw 1 Dr. Okrent, that 10 miles was a reasonable averaging x-2 distance.
3 DR. OKRENT: If you were to average over one 4 mile, would you find that some plants didn't meet it?
5 MR. TAYLOR: I donsa't think so. I don't think 6 there is any plant that would fail to meet this one.
7 DR. GRIESMEYER: Just one point. On the latent, 8 you said something about the impetus to improve core melt 9 frequencies, depending upon how much the frequency is over 10 10 to the minus 5 and the status of health effects. That is 11 in the second bullet there. I guess that is because of the 12 way you put it into the decision matrix, because you
() . 13 changed the amount of on-site costs, you should include. It 14 would seem you should have that impetus anyway, if you have 15 a higher core melt frequency, you are going to have higher 16 costs. So it is already built in there, and you wouldn't 17 need to have the decision matrix, in order to have that came 18 inference.
19 MR. TAYLOR: If you will recall the original 20 safety goals, they had no sliding scale. They had no 21 penalties necessarily indicated beyond $1000 per person-rem 22 for any core nelt frequency above 10 to the minus 5. We 23 I have tried to rectify that and give a very visible message 24 that we expect the future plants to strive for this lower O 25I core melt frequency. So my personal belief is that this is ACE. FEDERAL REPORTERS, INC.
- _ _ __-__-___ __- ___ mVikrin m _ (MWm
._. . . - . _ _ _ . . - _ =__ __ _ _ _ . _ _ _ . , _ _ .____._ _
.1350 05 13 66 .
DAVbw 1 an improvment with regard to the visible message we are 2 sending with the policy compared to the other approach.
3 4 :
5
! l i
6 7
- t 4
8 9
l i 10 11 '
1 :
i 12 0 12 4
14 15 16 i
17 i !
18 I i l i
4 19 20 21 i
22 I
l 23 1
24
- O 25 6
4 8
ACE FEDERAL REPORTERS, INC.
FmriMin _ M an t__-__ M . -_
n l
1350 06 01 67 l 1 DR. OKRENT: Are the future plants supposed to
{])DAV/bc 2 strive or are they supposed to achieve a 10 to the minus 5 3 d n the staff's eyes?
4 MR. TAYLOR: I would say achieve. If it's 1.1 5 times 10 to the minus 5 or 1.2 times 10 to the minus 5, for 6 all practical purposes, that's achievement.
7 Any further questions on this?
8 (No response.)
9 (Slide.)
10 I'll walk through the next slides fairly
- 11 quickly. Some of this will be redundant to our 12 discussions. But what I would like to do is just highlight
() 13 some key features that are in the policy statement that t
14 we're proposino.
15 Two qualitative safety goals have remained 16 unchanged. We have, let's say, downplayed as individually
, 17 discrete objectives the prior quantitative elements in the i 18 safety goal.
1 19 We have then put those in the integrated matrix.
20 We see the integrated matrix as being sort of a template 1 21 that's convenient, if you will, to accommodate decisions 22 that might arise for outlyers in the severe accident policy, 23 to assist in the decisions of backfit, and to help display ,
i 1
24 the safety cost tradeoffs.
, ( 25 DR. REMICK: I made my point on that second ACE. FEDERAL REPORTERS, INC.
_. - _ _ _ _ _ _ _ _ _ _ _ ___ __ _-. __ _ __ _ _InPTurm - --- N -----_e---------.-------_
'1350 06 02 68
(~)DAV/bc 1 bullet, but I won't make it again. But I fool strongly V
2 that's a mistake.
3 MR. TAYLOR: It's incorrect wording, I believo, 4 more than true substance of the policy in the statement. l 5 (Slide.)
4 6 In the policy statement, we made it clear that 7 the safety goals are not a substitute for existing
- 8 regulations. Licensees or applicants would still be i
9 expected to meet these, i 10 We have, in accord with the recommendations 11 derived from the ACRS review of the steering group report, 12 we have included words to exhort achievement of a mean core
() 13 melt frequency of less than one chance in 10,000 for current 14 plants.
15 DR. OKRENT: Where is that? Do you remember what 16 page?
17 MR. TAYLOR: I don't recall the precise pace, but 18 I it's in the policy statement. We have some other language 19 that has been included that recommends more than suggests, l
20 or sets an aiming point, if you will, for future plants to !
f 21 do botter than the current plants in terms of core melt 22 frequency.
)
23 We also, in the sense of trying to manage 24 uncertainties, we've also included some language in there to O 25 address the uncertainties, in particular that we should try '
ACE FEDERAL REPORTERS, INC.
_____m_______m____-__-_m_______---____
1350 06 03 69 l 1 to strive for a balanced design and not let single sequences l
{)DAV/bc l
2 and uncertainty about sequences drive the overall risk
, e i I 3 results or the core melt.
J
! 4 (Slide.) -
I 5 In effect, what we've done through the integrated t
6 matrix is established a de minimus risk statement, if you i 1
7 will. Also, thst's at the top of the matrix.
i 8 At the bottom of the matrix, we have indicated j 9 that that is a risk state that we would view as unacceptable i
10 in the long sense of long alterations.
1 i
) 11 There's a sense of urgency for safety l -
12 improvements, or shut the plant down. Again, that would be ;
( 13 a decision, that would be the Director of NRR's decision
- 14 that would be required there.
f j 15 We have given the flavor, if you will, of an i
l l 16 ALARA approach to try to drive towards improvements in the
- 17 core melt frequency. And also describe some of the cost i
18 tradeoffs involved.
7 19 Again, the matrix, as we pointed out earlier, we 20 have emphasized that the public health and safety is the
- 21 priority emphasis. Again, tying the core melt to the cost 22 benefit guideline. We give more emphasis to the accident 23 prevention, albeit one may argue about the total dollar i
4 24 amounts that would be coupled to certain levels of core
) 25 j melt.
1 i I i ACE. FEDERAL REPORTERS, INC. i
_ _ _ _ _ ___-----_h-_----- rJr+~Tfver r~t e_ ____ --
l 1350 06 04 70 1 We also acknowledge that there need to be care in
(}DAV/bc 2 operation, to have sound risk management programs aimed at 3 accident prevention. Again, we all recognize through l
4 accident precursors, probabalistic projections from that, 1
5 and we can drift away from the average. That is done l
6 periodically and one needs to keep an eye on that.
l 7 MR. LAVE: Excuse me. I would recommend that you 8 leave the first bullet up there. I think you're setting 9 yourself up again. The FDA's de minimus risk level is one 10 in a million lifetime. You're talking about a de minimis 11 level for core melt of 10 to the minus fifth per year.
12 That's several , orders of magnitude greater risk than the
() 13 FDA's guidelines. .
14 ,
EPA's guideline is not so explicit. They're -
15 either 10 to the minus 6 for the lifetime, or 10 to the 16 minus 5 per lifetime, depending.
17 I think I don't disagree with your goals. I 18 think the language that you'ro setting up there is going to ,
19 caus i you problema. Don't call it a de minimua risk because 20 it seems to be several orders of magnitude higher than that i
21 from other agencies. l 22 MR. TAYLOR: I would disagree with that.
I l
l 23 MR. SNIE2EK: Let me address that. You've ,
l l 24 addressed one of the key issues by what integrated the C) l 25 matrix. The de minimus level is not less than 10 to the l
l
! ACE. FEDERAL REPORTERS, INC.
_ _ _ . _ _ _ PnEiMTR _ - _ _ _ _ W @u L_______ENTD
1350 06 05 71 1 minus 5. The de minimus lovel is less than 10 to the minus
(~}
v DAV/bc 2 5 and you meet both health effects. That's the de minimus 3 level.
4 It's not just less than 10 to the minus 5.
5 MR. TAYLOR: That is a de minimus risk state, if 6 you will.
7 MR. LAVE: I'm sorry, I can't carry out the 8 calculation, but I suggest the staff do it and find out what 9 the risk per lifetime is. I will bet you that it is not 10 remotely close to 10 to the minus 6 or 10 to the minus 5 per 11 lifetime, because to begin with, you have to multiply the 12 numbers by 70 to go from an annual to lifetimo. And then O 13 we ve oot to sev somethino about the nember oc rem.
14 person-rems they're going to have because of core melt.
15 If you simply take the quantitative risk goal of 16 increasing cancer of .1 percent, that's one in a thousand, 17 10 to the minu's 3, if that's 10 to the minus 3 per year, 18, compare that to 10 to the minus 6 per lifetime and you're i
19i five orders of magnitude different.
20 MR. SNIEZEK: Without getting into the 21 calculations, de minimus, in our use here, means it's a 22 level below which nothing more is expected of the licensee.
23 That's what is meant by it. Maybe it's the wrong use of the 24, term.
O 25 Is that what you're suggesting?
ACE. FEDERAL REPORTERS, INC. l r~WTMYr4 - . h __ _ _ _____ _ _ fWFv""1
1350 06 06 72 1
1 MR. LAVE: I'm suggesting that the term de
(}DAV/bc 2 minimus comes from the 1980 Supreme Court benzene decision 3 which said that an agency had to find -- OSHA in that case 4 had to find a significant risk in order to regulate, and so 5 on.
6 There's a very specific legal context for that.
7 Get yourself out of that. I don't disagree with your goals 8 right now, just get rid of that language.
9 MR. SNIE2EK: Understood.
10 DR. REMICK: Nat, have you gotten any feedback 11 from the Commissioners on reincorporating the ALARA concept?
12 As you recall, they are the ones who took it out
() 13 originally. !! ave you gotten any kind of feedback from them?
14 MR. TAYLOR: There was an indication on not 15 having the ALARA concept, allowing the ALARA flavor in the 16 safety goal policy. Yes, there was that.
17 DR. REMICK: What kind of public feedback are you 18 getting?
19 MR. TAYLOR: At the moment, we have not put this 20 out for public comment. And that's the question right now.
21 DR. REMICK: Am I correct that NUMARC has a group 22 looking at what you're doing?
23 j MR. SNIE2EK: Yes.
24 ! DR. REMICK: Are they planning to come in i (2) 25 sometime?
ACE FEDERAL REPORTERS, INC.
PMGM _ ____ M ____ _ M _ - -
1350 06 07 73
(~';DAV/bc 1 MR. SNIE2EK: There hasn't been a lot of 2 communication with NUMARC. In fact, we w9re kind of 3 surprised to learn they had a commission going.
4 MR. TAYLOR: My understanding is that they are 5 trying to come up with an alternative proposal. That's my 6 understanding.
7 DR. MACLEAN: Can I comment?
8 My guess -- it's just a guess -- is tha- when 9 this is put out for public comment, you will find there is 10 strong support for keeping the ALARA concept in. I think 11 you will also find that this report will focus on the de 12 minimus context.
O)
(_ 13 DR. REMICK: But you don't plan to put it out for 14 public comment?
15 MR. SNIEZEK: That's a Commission decision. We 16 would envision that the Commission could have gone forward 17 with the quantitative goals without public comment.
18 However, it was up to the Commission whether or not they 19 wanted to go out for public comment on the matrix approach.
20 (Slide.)
21 MR. TAYLOR: Continuing with some of the other 22 key features.
23 DR. OKRENT: Excuse me a second. Let me 24 comment. I think when one puts this out for public comment, 7-( ')/
25 what one finds is, although there may be a fairly broad l
i I
l ACE. FEDERAL REPORTERS, INC.
rnf%ri:n e e______________
1350 06 08 74 1 interest in the subject, those who take the trouble to
( } DAV/bc
, 2 respond are those whose interests are immediately affected. f 3 And I'm not sure you would get a reasonable 4 cross-section of the public.
5 MR. SNIEZEK Actually, our attempt -- and 6 someone can conclude whether it was successful or not -- was 7 to accommodate all the comments from various extremes that 8 have been received so far. So there was a real policy 9 decision whether it would even have to go out for public 10 comment.
11 MR. TAYLOR: Continuing on with some of the other 12 key features of the proposed policy, it indicates that we
() 13
~
are about ready to apply or start implementing the 14 integrated matrix in terms of generic safety requirements.
15 Obviously, we do need some procedures, further 16 detailed procedures developed. What the matrix does is 17 enforces some sort of an accounting system where we should 18 keep track of what we project to be our safety improvement, 19 in terms of improvements in the risk state, and also some 20 accounting system for the sum total of the costs that we're 21 imposing.
22 We don't have that necessarily in place in any 23 disciplined way at this point in time.
24 DR. OKRENT: Can I ask, it was mentioned on one l ( 25 of your bullets, and somebody found the page for me --
ACE. FEDERAL REPORTERS, INC.
.------_ _ _ F M 1 E R - -- - -- -- --- M mNcMrr 1 ---------- FQ0TD -_-_------
1350 06 09
- 75 L
1 MR. SNIEZEK: Page 18, I believe.
}DAV/bc 2 DR. OKRENT: That there is an exhortation here to 3 further guide the staff. They should normally be lef t in 4 one in 10,000 per reactor year. You know, suppose the I
, 5 reactor meets the health goals, whatever you call the safety l
l 6 goals, but it's 9 times 10 to the minus 4, and they argue t
[ 7 that the measures that might bring them up to 10 to the 8 minus 4 fall outside of the dollar value that could be l
9 suggested by your matrix.
10 Which of these two guides predominates? How does 11 one decide?
l 12 MR. TAYLOR: Understand, this is only one factor l
l
() .
11 in the decision process. This describes what we would l 14 normally expect as the safety cost tradeoff. My hunch would i
15 be that if they are around 10 to the minus 3, that's j 16 probably some regulation that would apply. But I can't tell 1 */ you up front which argument would prevail. ;
18 I would guess that would be a management 19 l decision.
t i
20 MR. SNIE2EK It would really depend upon what l 21 type of fix could be made to improve the core melt ,
22 frequency. It would take into account how much would it l
l 23 cost. There's no equation that you would apply.
l l 24 DR. OKRENT: The words I see are policy statement
( 25 indicates you are about ready to apply safety goals i l
l t ace. FEDERAL REPORTERS, INC. I
___-_--__-_____-__---___M_----_Mw t - ----- M--
.1350 06 10 76 !
DAV/bc 1 (matrix).
2 It doesn't say "and we are about to apply the 3 exhortation on tho bottom of page 18 and 19." It docan't 4 say that. So it sounds like, as they said in 1984, some 5 people are more equal than others.
6 MR. SNIEZEK: Obviously, the staff has to l I 7 implement whatever is in the policy statement. What the 8 Commission will decide upon. And that will be implemented. ,
9 MR. TAYLOR: It is my belief in terms of the 10 generic issues the staff is giving heavy weight to core :
11 molting programs. I think also that there is not a high 12 favor if core melt is 10 to the minus 3 among many members 13 of the staff, or the management.
[}
l 14 . that's my personal observation. Certainly, not 15 without some compensating arguments to the contrary.
16 DR. OKRENT: Go ahead.
l 17 MR. TAYLOR: I would add, I think, that it was l
1 18 in the entire thrust of the steering group report..
19 DR. OKRENT: The fourth bullet just says the same l 20 thing. Through the integrated matrix.
l 21 MR. TAYLOR: Right.
22 DR. OKRENT: That's the thrust.
l 23 MR. TAYLOR: I think what we intended to l
l 24 emphasize here is that I believe, in connection -- and Jim i
25 may correct me if I state things wrong here -- I believe it 1
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide coversee soo 3364N6
l 1350 06 11 77 2
(') DAV/bc 1 was around $100 million. .
U 2 Jim, it was $100 million that has already been t 3 spent on post-TMI improvements.
l 4 MR. SNIEZEK: That's correct.
5 MR. TAYLOR: By any cost algorithm, we should 6 already have reached the safety level. These have not been l
7 cost-effective if we haven't.
8 DR. OKRENT: You know, I think that's really a 9 red herring that people have thrown into an argument.
10 That's my honest opinon. If, in fact, those changes have l l 11 not reached 10 to the minus 4, or you've been unable to i
12 quantify them, you know, the past is history.
l
() '
13 ,
MR. TAYLOR: That's true and the integrated l
14 matrix is intended to move forward from that point.
15 DR. OKRENT: Maybe all those things wore good, or 16 the majority or 50 percent, or something.
i t
- 17 Maybe they were aimed at something different. '
18 MR. TAYLOR: That could be. We have a lot to 19 lend them on the platter that is comparable, for the same 20 price.
21 DR. OKRENT: For example, money was spent on 22 fixing the control room. Money was spent on -- what do they l
l 23 call these little centers? So you don't have all these I i 24 people spilling into the operating room. And then there O 25 were the offsite centers, and so forth. I think you'd find i
ACE. FEDERAL REPORTERS, INC.
_ _ _ _ _ _ _ _ - _ _ - - _ _ . _ _ _ _ _ _ -___ ___rMmrin---____ m ~_________m_________
1350 06 12 78 1 it hard to cuantify. These end up being nontrivial. Not
{)DAV/bc 2 the full 100 million.
3 HR. TAYLOR: My point is that it's still a non-4 trivial amount, still.
5 MR. SNIE2EK The real issue is not the money.
6 It's the manpower of the utilities. Tf it's being expended 7 on things that are not improving safety, it's diverting it 8 from things that could be included in safety -- the money 9 and the manpower.
10 MR. TAYLOR In the last two bullets here, we 11 have some language in the policy statement indicating that 12 we would be working to explore the merits of payment, design
() 13 and performance. The goal or objective that is going on 14 right,now, I'believe that is going to be discussed in the 15 next session before you. We have included both internal and 16 external initiators within the safety goal and modo that 17 explicit.
18 DR. OKRENT: Do you feel that you have enough 19 information on the next to the last bullet to formulate 20 criteria, or that we really don't have that kind of 21 information?
22 MR. TAYLOR: I think there's a lot of tenhnical 23 information available. I think one element is what are we 24 really adding for that in the safety goal?
25 One would have to look at the implementation. Do ACE. FEDERAL REPORTERS, INC.
. . _ _ _ . EhMMM--- - -_--__ Wea'** W- ----------- ** M
1350 06 13 79 DAV/bc 1 we really need this in the safety goal? Is there merit to 2 incorporate it in there?
3 MR. SNIE2EK Let me add we heard from the Office 4 of Research yesterday, or the day before, a plan for 5 developing performance objectives that looked pretty 6 positive, that there could be some containment performance 7 objectives reasonably developed.
8 The exact scope and nature of them is 9 undetermined at this time, but there is a plan that's before 10 the EDO and being reviewed. And it will go down before the 11 Commission, in which they believe they can come up with the 12 proposal by the end of this calendar year.
O 3- . 1 d hate to nre3udos.
14 15 16 17 18 19 20 21 22 23 24 O 25 ACE. FEDERAL REPORTERS, INC.
- _ m - __ W ___ M _~
1 07 01 80 l1 l
DAV/bc 1 DR. OKRENT: You're really answering a different 2 question but, go on, I'm trying to see whether the 3 information is adequato, the technical information was l 4 adequate.
l 5 MR. TAYLOR: I'm not the one to answer that, I l
6 guess. I think we would be in a better position perhaps 7 when 1150 is pulled together. I think there's a lot of 8 information on hand that was not on hand several years ago.
9 Perhaps Mal would like to speak to that.
10 MR. ERNST: That's a very complicated question.
11 I really don' t know what " adequate" means. Clearly, there O 12 are going to be substantial uncertainties.. I think the real 13 question is trying to decide reasonable bounds on these 14 uncertainties and decide as a result how am I going to treat 15 the information and the decision-making process.
16 There is no precise answer, as you well know.
17 (slide.)
18 I'm afraid this one might generate more heat than 19 lig ht. But what we've tried to do is to explicitly address 20 uncertaintion in the policy statement, and try to take a 21 clear look at the uncertainties.
22 These did not exist because of PRA. These have O 2' a a rauad ror taaa 'i ao ta aa^ = taadataar iav l
24 provides a framework where we might better quantify these in 25 terms of core melt accidents and offsite health effects.
j ace. FEDERAL REPORTERS, INC.
_ - _ - _ - _ - _F~WM W __- M ____M O _ -
r .
,A !
sp .
1350 07 02 -
s 81 A
i ( JDAV/bc 1 DR. OKRENT: Is the fourth bullet mathematically 2 rigorous?
3 MR. TAYLOR: I guess that's my language thoro, so
! 4 that's my opinion. But I think you take two factors. If l
'S you have a large uncertainty to the input and you do apply 6 distributions that are skewed, such as log normal, you tend 7f' to find your mean so you can be biased -- I guess is the V .
U wrong word to be used there, but it cdn bo quite influenced 9 by the large range of sensitivities.
10 DR. OKRENT: Your words are " conservative bias".
11 I'd like to know whether you can justify those words 12 mathematically.
O 13 MR. TAYLOR: No, I guess I can' t.
14 DR. OKRENT: I would suggest you not use them or 15 put "It's my subjective opinion."
16 (slide.)
l l
l 17 MR. TAYLOR: What I was trying to convey here was 18 that with respect to the median, we are highsighted by a 19 1arge range of uncertainties. And we do approach the best 20 discrotion that we can make of the upper bounds.
21 Perhaps my languago did not correctly reflect 22 this figure in the Sandia document, but that's what I meant I
f e i 23 by a bias to the high side.
() 24 DR. OKRENT: What I'm getting at is, if you 25 really knew the uncottainty and computed it, it has a l
ACE. FEDERAL REPORTERS, INC.
_ _ __ _ _ _ N _--- N u ' 't-___ _____ N - _ _ _ -------
1350 07 03 82
(~T l L JDAV/bc 1 certain mathematical meaning. I'm not aware that there's a 2 conservative or a nonconservative bias.
3 MR. TAYLOR: I guess one can say that if we 4 arbitrarily choose right skewed distributions as a matter of 5 practice, that is a conservative bias, we could choose 6 normal distributions.
7 DR. OKRENT: The bias is in the distribution
, 8 you've chosen then?
9 MR. TAYLOR: That's correct.
10 And, traditionally, most PRA's that I'm familiar 11 with do use the right skewed distribution as an input 12 assumption. That's a general observation. There is a CE) 13 bias.
- 14 DR. OKRENT: It's true. Of course, Castellakas 15 has shown where some of these right skewed distributions 16 still fell outside the data range on the nonconservative 17 side.
18 ! MR. TAYLOR: That was my observation.
19 DR. GRIESMEYER: Go back to the last one. You 20 ' said you wanted to display the mean, the median and 1 21 confidence level ranges. I'm kind of concerned that you 22 don't know any of those.
l 23 MR. TAYLOR: That's correct.
l 24 i DR. GRIESMEYERt Because what happens is there is
(])
25 some place for data and there's a lot of place for judgment.
ACE FEDERAL REPORTERS, INC.
_________________ J ~V 1 m n ______-___ M _____ m _ __ __
F f ( '
l ..
- 1350 07 04 ,
.. 83 I
L_jDAV/bc 1 When you go to the expert to solicit his opinion, you're 2 lucky to get a number. Neither of these are related to any t 1 x s 3 distribution, and certainly not the mean of the 4 distribution.
5 And when your estimate is a propigation of all r
) 6 thesd' things, it may in s'sa o sense be central but it's not a 7 mean or a median or any confidence lovel. So you have to be 8 careful because I'm not sure that any of these things you 9 can get.
.\
10 MR. TAYLOR: Well, I can' t argue with your 11 assuming distributions on a lot of input variables. I can' t 12 argue with you.
13 DR. GRIESMEYER: If you assume point,ostimates, L
you're even worse.
I 14 6 \
'd 15 l MR. TAYLOR: In order to get a mean, you must in 16 some account take account of some distribution. You must 17 make some of your cutoff assumptions upper range, lower 18 range, and I think this is all we-intended to reflect here, 19 to make some judgment on where that mean value lies, whether 20 the median or the upper and lower bounds.
21 I don't think we intended to foster' e tensive, 22 complex uncertainty analyses by this policy statement. But 23' to'mak'e a reasonable, informed judgment.
s
(~S 24 DR. OKRENT: The Commission has still:not told us
()
25 whether, af ter displaying these uncertainties, how 3!
- ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646
1350.07 05 84 t'l l L / DAV/bc 1 philosophically they will affect decision-making.
2 In other words, maybe they should. But, up to 3 now, as you go around business, as you go around society, 4 one finds reactions to uncertainty. But, right now, one is 5 hardput to find any semblance of a guiding philosophy as to 6 what the NRC staff thinks ought to be done.
7 MR. TAYLOR: I think the staff has displayed the 8 way ycu want to go. Secondly, the staf f has to back up that 9 philosophy to display. We would like to go to improved core 10 melt frequencies. I think that's clear in the matrix. It's 11 clear from the reg.
12 Now, if you're asking me will management follow C:)- 13 to the letter what's there, I certainly. can't respond to 14 that. I think one has to have that philosophy approved by 15 the Commission. And I think there is a great deal of I
16 sympathy for that philosophy in the staff, and by the office 17 directors.
18 i I don' t know whether that answers your question. ,
19 DR. OKRENT: No. And, in fact, EDO asked that 20 the Commission not really get involved in this, as I read l 21 it.
22 MR. TAYLOR: It said yes. There's been a number 23 of arguments on the isolated quantitative objectives. We've
(} 24 tried to integrate those to visibly indicate what our 25 philosophy would be, and that that would be the j
> l ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Com 800-336-66 4
1350 07 06 85 T'; I L JDAV/bc 1 implementation.
2 MR. SNIEZEK: The Commission did not make an EDO 3 recommendation on that point.
4 MR. TAYLOR: The last slide I have here just 5 touches on some of the areas of implementation where we 6 would foresee possible effects on safety goal policy.
7 As I said, we think we can start introducing 8 those into the arena of generic issues, generic safety 9 issues, to help gauge the possible payback from resolution o i
10 some of these issues.
11 DR. REMICK: How do you see this fitting in with f 12 the backfitting policy? Any problems? Are there .
13 inconsistencies between the guidance?
14 MR. TAYLOR: I don't see any major 15 inconsistencies. Let me say it's not incompatible. One 16 could argue that it's incompatible perhaps insofar as we've 17 specifically included on site costs. That is not explicitly 18 mentioned in the backfit rules. But I think it's compatible 19 with all other elements.
20 DR. REMICK: So you don' t see a need for changing 21 the manual on the backfitting?
22 MR. SNIEZEK: There's no question, if the 23 Commission was to iraprove the safety' goal specifically with
(} 24 the use of the matrix, there would be some minor revision to 25 the manual chapters to reflect somehow how substantial ACE-FEDERAL REPORTERS, INC.
STR P37-3700 Nationwide CosGrm 800-336 6646
1350 07 07 86
(
(_\)DAV/bc 1 improvement in safety is tied to the safety policy. There 2 would have to be some change, but it is not inconsistent 3 with the rules.
4 MR. TAYLOR: I would propose to halt here, unless 5 there's further questions.
6 DR. OKRENT: That sounds like a good idea. Let's 7 maybe see what questions or comments our consultants have.
8 Thank you.
9 MR. LAVE: I guess I would like to take about 10 five minutes to say a few things if that's appropriate.
11 DR. OKRENT: Go ahead.
12 MR. LAVE: One thing which is just not in order, O 13 I finished some calculations about the implication of a .1 14 percent increase in cancers. And I'm not clear as to 15 whether that's interpreted as an annual or a lifetime goal.
16 DR. OKRENT: That's annual.
17 MR. LAVE: If so, that is 350 times the FDA risk 18 goal. I just point that out -- 350 times.
19 Okay, quickly, some comments. Number one, having 20 been on the outskirts of the process for about three years 21 now and not having had very much to do with it, I really 22 ,
think that the quantitative risk goals are very valuable.
23 We ought to try to find out what people agree on and get i 1
(~} 24 them out the door, s_/
l 25 They' re going to be imper fec t. They're going to l
l l
l ACE-FEDERAL REPORTERS, INC.
l EB@7-DER __ _ _ _ Nacmtfb@m c~) _ _ _ _ GNISc"3
~
l 1350 07 08 87
/~
L_T/DAV/bc 1 be revised because of changes in technology and changes in 2 perception. All of that's fine.
3 Look at them again in a few years. Indeed, build 4 an explicit provision. If there's a problem with the 5 document so far, it doesn' t explicitly say: These goals 6 shall be reviewed in five years.
7 If you say that, a lot of the problems disappear 8 because you're no,t building things in long-term.
9 MR. SNIEZEK: I think it does really say that 10 specifically in the document.
11 MR. LAVE: Great.
12 Secondly, I don't like the matrix. I think the 13 matrix is complicated. It is not explained in the document 14 and, even after being explained, it is relatively dif ficult 15 to understand. I want to hark back to ALARA and say that 16 the principle of ALARA is not bad. You use a thousand 17 dollars a person-rem to avert radiation exposure if it's l
18 ! greater than 10 rems, and five dollars. If it's less than 19 that, it's a thousand dollars a person-rem for all 20 exposures.
21 It seems to me, if you apply that principle to 22 the matrix, if what you use is full averted on site costs 23 plus $1,000 a rem, no sliding scale, it simplifies life
/~T 24 immensely.
U 25 And in a previous report that I did for ACRS, I ACE-FEDERAL REPORTERS, INC.
K % 347-3700 Nationwide Covern 80 4336-6686
1350 07 09 88 q,jDAV/bc 1 argued that all of thut was the only thing that really made 2 any sense.
3 Since then, I've had some discussions with people 4 who sort of indicate that perhaps the reason for the matrix 5 is utility concerns with the Commission getting into 6 economic regulation.
7 I don't know that that's the nature of the 8 concern but if it is there's a very simple solution.
9 Namely, all you have to do is say that if the external 10 benefits, that is, the benefits that are not related to 11 averted on site costs, are either less than some fixed 12 dollar amount or less than some percentage of total n
13 benefit's, then the benefit cost analysis is advisory, not 14 , conclusive.
15 That is, if the concern is that the utilities 16 would say, Look, there's a dollar in external benefits but
> 17 you guys are going to force us to do the following, that 18 essentially economic regulation and that's very easy to do 19 if the external benefits are not at least -- pick a number--
20 S100 million, or at least 10 percent of total benefits, then 21 any calculation is simply advisory.
22 That gets the utilities out of the box; if 23 they're afraid that all the Commission is going to do is
(} 24 economic regulation, it reinstates what is the eas'r 25 principle of using the full averted on site costs, plus 26 external health ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationw'de Coverage _ 800-336 6646 ___ _
I l
1350 07 10 89 q
(_/ DAV/bc 1 external health effects.
2 DR. REMICK: Excuse me, Les, what do you mean by 3 external benefits?
4 MR. LAVE: The reason why the NRC is worrying 5 about this is not because the utilities are going to damage 6 their property. If some utility wanted to come in and beat 7 up their reactor, while having no danger of spilling any 8 radiation to the public, then, presumably, the NRC would 9 stand aside and say: It's your money, you can do what you 10 want with it.
11 The NRC is concerned because you're going to be 12 spilling some radiation on to the public, or causing either O 13 i I
prompt or latent fatalities. That's where this $1,000 a 14 person-rem comes in.
15 DR. REMICK: So it's the risk to the public on 16 the basis of the S1,000 per person-rem. And that's what 17 you're calling --
18 MR. LAVE: Right. External benefits.
19 Next, it seems to me that Doug MacLean and I do 20 this sort of song and dance together, where we all say, Hey, 21 look, you guys, don't overestimate how much benefit cost 22 analysis is going to do for you. It's a tool. It's a very 23 helpful tool if you use it right.
{} 24 If you don't use it right, it's just terrible.
25 There seems to be more faith in this document in benefit ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage _ 80433 M 646 _ _
1350 07'll 90
(~h
(,jDAV/bc 1 cost analysis than I want you to have. And Doug is really 2 going to hit you with it.
3 If you use it as a tool to help you make 4 decisions, help you organize your knowledge, the ,
5 Commissioners know in detail what its litaitations are and 6 the staff are proficient, it's wonderful. If any of those 7 is not true, don't do it.
8 In particular, if you' re not going to change the 9 matrix, if you can use the matrix the way it is now, don' t 10 use benefit cost analysis because it's going to absolutely 11 mislead you. It's going to cause you to do things wrong, 12 not just to not do backfits when you should. But, when you 13 do backfits, it's going to tell you to do the wrong things 14 because it will tell you to spend too little money.
15 It is absolutely pernicious in the way it's in 16 there now if you use that matrix.
17 okay, that's part of the more general point.
18 Benefit cost analysis is certainly a two-edged sword. If 19 you don't use it exactly right, then probably you shouldn' t i
20 use it at all.
21 Core melt, I think we've covered all that stuff 22 very well. The last part of this is just sort of a general l
23 injunction, and I'll keep it short.
{} 24 The whole point of getting into quantitative 25 safety goals was to try and state things explicitly rather ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Covertr3 800 336 6M6
1350 07 12 91
)DAV/bc 1 than having surrogates for what it is that you want.
2 Get away from surrogates. I think all of the
(
3 dissention between the ACRS and the staff on means versus 4 medians had to do with uncertainties in whether the medians 5 took better account of the uncertainties than the means 6 did.
7 It gets everybody to misunderstand when you try i
8 and use surrogates. I understand that I sound like an 9 academic here, but what you have to do is try and set up the 10 structure as explicitly as you can, acknowledging 11 uncertainties. And don't start using surrogates for things 12 because they're convenient now.
I~')
L/
13 l I mean, the whole benefit of the quantitative l
14 i safety goals is to try and get the structure out on the 15 table so people can understand and use it. That's why I'm 16 all for putting out this document now.
17 Don't start getting into the various kinds of 18 surrogates. For example, in addition to $1,000 a person-
. 19 rem, you ought to be taking account of any property damage 20 offsite. Do not use the $1,000 a person-rem as a surrogate 21 for offsite property damage.
22 I don't believe that offsite property damage is 23 going to be very large, but include it explicitly, or at 24 least include some recognition of it explicitly.
(')!
s-25 That's the kind of thing I'm talking about. What ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6646
h 1350 07 13 92 rm L)DAV/bc 1 I'm saying, try to get the structure right even if it's 2 going to cause people some stomach aches in trying to figure 3 out exactly how yoa do it.
4 But, just the general injunction is: Let's get 5 on with getting out a document. Hopefully, before June.
6 7
8 9
, 10 11 12 O ;, ,
4 14 15 16 17 l l 18 ,) !
19 20 21 22 l
23
- 24 O
25 ACE-FEDERAL REPORTERS, INC.
~ . . - - , _
. . _ . . , _ _ M-M7 3700 _____ _ , Nationwide Coverage _ _
800-336-66.m _ __
1350 08 01 93
(,)DAVbw 1 DR. MARK: Off-site property damage can only be 2 measured, it seems to me, in terms of person-rems. I mean, 3 it doesn' t really hurt, if you have 5000 rem. You might 4 kill some fruit trees. That would be property damage.
5 otherwise, the property looks very much the same and the 6 only thing wrong with it is that people shouldn't be there.
7 So it's person-rem again.
8 DR. GRIESMEYER: That's not entirely true, 9l because what you have is, you have an interdiction criteria, 10 and you can directly trade property for man-rems.
11 DR. MARK: It's man-rem.
12 DR. GRIESMEYER: But there's a trade there, and 13 f you can eliminate man-rem for an awful 16t of interdicted 14 l property, and if you say, well, a,ll right, we don't have the I
15 man-rem, because we assume we're going to interdict a lot of 16 property, you've just lost a lot of the cost that's not 17 there.
18 DR. KARK: If you interdict the property, it is 19 because you have set a boundary on the man-rem.
20 DR. OKRENT: Well, if you throw away mile and 21 cheese.
22 i DR. MARK: I mean that little restaurant, Alice's 23 Restaurant, nobody can come, because there's too many
{} 24 man-rem going on there, but it is man-rem still.
25 DR. LAVE: But what you are going to find is, if ACE-FEDERAL REPORTERS, INC.
202 347 3700 _ _ _ ___. Nationwide Coverage . _ _ _ _ _
80 4 336 6646 , __
1350 08 02 94 (N I LjDAVbw 1 nobody goes in, there aren't any man-rems, and you've gotten 2 rid of a $100,000 restaurant, you are going to charge 3 yourself for the $100,000 restaurant; right?
4 DR. MARK: Because of where you set the 5 man-rems.
6 DR. REMICK: How far do you carry that cost into 7 consideration off-site in the secondary effects on people?
8 I can imagine someone who might die, because they don't have l 9 access to a home or something. How far do you carry out all 10 these costs? I mean the academic responses. You take into 11 account all costs, but that is not very pragmatic.
12 DR. LAVE: Yes, that would have set an academic O 13 responses precisely to do that. I think that in practice, 14 the limits of economics now are very firm, in terms of when 15 the signal disappears into the noise and you can't do 16 anything anymore. It seems to me that how far you carry it, 17 always has to do with whatever are the state of the art of 18 the analysis that is being practiced.
19 DR. REMICK: So it is a judgment call.
l 20 DR. LAVE: No, I think you go to the experts in 21 the field and say, what is it that we can quantify with the 22 current state of the art. Whatever they say you can't go 23 beyond that, you don' t go beyond that.
(} 24 I DR. REMICK: And who are these experts? Are they 25 economists?
ACE-FEDERAL REPORTERS, INC.
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __D3007-DED _ __ _ 9tc~It'b csaa~; __ ___________ GDEX"'D
h 1350 08 03 95 DAVbw 1 DR. LAVE: Yes. If you're talking about' economic 2 damages, yes, they are economists.
3 DR. MARK: There was one point in your comments 4 that I perhaps didn't follow correctly. S1000 a man-rem.
5 Do you mean to integrate that out to levels below natural 6 backgroun?
7 DR. LAVE: I do discount cancers in future, but 8 yes, well, look, we've got too issues. One of the issues is ,
9 the linear quadratic. Dose response relationship for 10 radiation. I don't mean to take a position on that.
11 However, that is, I think that, yes, you have to take a 12 count of exposures that then go below background levels.
O
- 13 There isn't any arbitrary cut-off, but I think that you also 14 discount exposures which occur a century from now or two 15 centuries from now as a result of the current releases. So 16 that you discount future health effects or future exposures, 17 in order to get current health effects.
18 DR. MARK: So you will integrate around the i
19 globe.
20 DR. LAVE: Yes, I would integrate around the 21 globe.
22 DR. MARK: I think that is nonsense.
23 DR. LAVE: Is it nonsense because we don't know 24 how to do it or because you don't believe it would be
(])
25 pragmatic.
ACE-FEDERAL REPORTERS, INC. '
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ ______ __ ___ _ FMP3-R __ -.__ _ WCHfb@K:r 3 . -_ _ _ _ _ BDDC"3
1350 08 04 96
<~ l
(,)DAVbw 1 DR. MARK: Because you are going to be counting 2 fractions of millirem about which you know nothing, neither 3 does the National Academy of Sciences nor anybody else.
4 DR. LAVE: But aren't you faced with the 5 following: we have to rely on some experts to tell us what 6 the dose response relationship looks like, if some person 7 gets, for example a thousandth of a millirem; right? I S don't propose to know what the answer is. If our expert 9 committee is going to be Baer 3 and we are going to use the 10 linear relationship from Baer 3, then I don't see any reason 11 why you want to cut it off arbitrarily, at, say, 1 millirem, 12 5 millirem or a 10 of a millirem.
O 13 ,
If you want to take Baer 3's linear quadratic and 14 say when we are down in the lower ranges then the effects 15 are less than that, that's fine, but whatever criteria you 16 want to establish, you can't just have some arbitrary 17 cut-off that says, if an individual receives less than 1 18 millirem that we are not going to count that.
19 DR. MARK: .I am not really indifferent on this 20 point. Whuc model are you going to use, because it makes a 21 vast difference on what answer you get.
22 DR. LAVE: I understand, and whoever is going to 23 make the judgment as to what model you are going to use, you
{} 24 '
25 then have to use it consistently and not put in an arbitrary That is all I am saying.
cut-off. If you want my personal ACE-FEDERAL REPORTERS, INC.
ll02-347-3700 Nationwide Coverage 800-336-6M6
1350 08 05 97 b DAVbw 1 beliefs, it is probably that the linear quadratic model is 2 better than the linear model at very low doses, but I am not 3 a radiation biologist, so don't pay any attention to what I 4 say about it, but whoever is, we've got to defer to the 5 experts.
6 DR. MARK: Please, no!
I 7 (Laughter.)
8 DR. MARK: But we do have to decide what we are 9 going to do.
10 DR. LAVE: Right.
j 11 DR. OKRENT: Doug?
l 12 DR. MAC LEAN: In my comments, I didn't get into
()
4 13 ,
the details of the matrix, beca'use I had a very difficult 14 time understanding them. I feel like I understand a little I
15 bit better how they play out today, but it is still a 16 difficult thing. I chose to comment instead on the form, 17 the structure of the safety goal policy and agree that the 18 numbers can be changed and debated by those who know better 19 than I do what they ought to be.
20 , It does seem to me that as I understand how the 21 numbers are being played out that there are certain problems 22 of consistency. In general, I think it is correct to insist 23 on some level of safety regardless of the cost and then
{} 24 l
beyond that, to look at increased safety, according to the 25 j cost of achieving it. But I think this is David Okrant's ACE-FEDERAL REPORTERS, INC. t 202-347 3700 Nationwide Csrca~a 800-3364wA6
1350 08 06 98 e'
(_)3DAVbw I concern. If you are going to insure that 10 to the minus 4 2 is the upper risk for the kind of serious accidents to the 3 core, then 10 to the minus 4, it ought to be at that level 4 in the matrix, not 10 to the minus 3rd.
5 So there is the problem of making that matrix 6 consistent with the other statements in dc,ument.
7 That is about all I have to say about that.
8 As far as costs are concerned, I do make some 9 remarks in my comments about how costs ought to be 10 considered, and I agree L1most completely, I think, with 11 Lester Lave, about the potential dangers of using 12 cost-benefit analysis. About the need to factor in costs, I O 13 would only disagree, I think, in whether it is acceptable to 14 use surrogates or not. It seems to me that if the including 15 of averted on-site costs is a way of putting added p ressure 16 to get safety to protect against core melt, then it is okay 17 to justify including those costs as that kind of a surrogate 18 to get that kind of result. It doesn't seem to me from what 19 I've heard today that you're creating a very great 20 incentive, given the estimate of less than $1 million at 21 some levels to move to greater safety 22 , I also make some remarks about the treatment of 23 uncertainties. I think that uncertainties need to be made
/~T 24 explicit perhaps, giving them greater emphasis than is in
'u_.)
25 the document, and it might mean, if we included reporting ACE-FEDERAL REPORTERS, INC.
EB347 3700 Nationwide Com 800 336-6646
l 1350 08 07 99
(\
L_)DAVbw 1 the uncertainties and where we are deal.ing with large 2 populations especially, we might want to move away from 3 using the mean, to using something more conservative than 4 the mean. Maybe or maybe not. But this would all be an 5 implication of making the uncertainties explicit and 6 treating them seriously.
7 My main concern, however, is in the statement of 8 the goals. We don't have here in the two qualitative goals 9 anything that is a truly societal goal. Both of them are 10 individual goals, and I think there is a need to include 11 something that more closely approaches a real societal 12 goal. I have tried to suggest what that means by describing
13 different perspectives one can take on safety. The 14 individual perspective which I think is reflected in both 15 the goals and more community perspectives would include not 16 only the average risk to the individual but the size of the 17 population.
I 18 I If you want to know what the expected deaths in 19 the population are, as well as the average risk to an 20 I individual, so that you might want greater safety for a 21 large population and a broader perspective still which would 22 look at the probability of an accident occurring anywhere in 23 the country. It is that perspective, I think, that is
/~T reflected in the comments of Commissioner Asselstine, the (l 24l.
25 l concern about the acceptability of perhaps a 50-50 chance of ACE-FEDERAL REPORTERS, INC.
202 347-3700_ __ Nat.onwide CKGC) __ _ ____ 800-33H646 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
1350 08 08 100
,-~
L /DAVbw 1 a core melt in this century. It seems to me that that can 2 be addressed by addir.g a truly societal qualitative goal.
3 That is, pick your number in the spirit of Harold Lewis.
4 Pick it arbitrarily and we can debate about it later. If 50 5 percent is too high, as I think it is, I can't vouch for the 6 calculations, obviously, then maybe 20 percent or 10 percent 7 or whatever number you think is acceptable risk of having a 8 single accident of this kind happen in the remainder of the 9 century. That number will have some important effects, I 10 think, on tne quantitative objectives.
11 For instance, it would work more towards ALARA 12 and against the de minimis level, because even though you 13 might still accept the 10 to the minus 4 level for each 14 individual plant,.once you look at the aggregate and look at 15 all the plants together, it might be the case that what you 16 allow for each individual plant in the aggregate doesn' t get 17 you down to the total level of safety you need.
18 I In that case, you would then look where further 19 improvements can be made most cost-effectively until you 20 reach that aggregate level. That seems to me -- by 21 formulating, taking seriously of real societal goals, it 22 would seem to me, to enhance the ALARA principle away from 23 the de minimis part of it.
(} 24 DR. MARK: You think of a societal mold. I can 25 l only agree, but it doesn't bother me in the least if the ACE-FEDERAL REPORTERS, INC.
- fBDT-Di@____ _______ NOCIfb_(9 meg ____ __ ___ CiNOX"'S
1350 08 09 101 L_jDAVbw I core melts here or there. It bothers me if people are 2 killed. How are you going to apply this societal goal to 3 core melts or to people killed or to man-rem or what?
4 DR. MAC LEAN: You can do it either way. I 5 thought that the working assumption in the document was that 6 we will assume that if there is a core melt thst will be a 7 release, because you can't at this stage distinguish.
8 DR. MARK: No, it isn't. If the core melts, 9 there's a few billion dollars down the drain, but there may 10 not be any lives.
11 DR. MAC LEAN: I can see arguing it both ways.
12 Even if there are no lives lost, there may be serious 7+
U 13 consequences, especially to the industry, of a core melt.
14 If there are lives lost, that's a different concern 15 entirely, more serious. Perhaps you may get a risk of core 16 melt with a release that results in some number of deaths.
17 However you set that, I am suggesting that there probably 18 should be a goal that takes this kind of accident that we 19 want not to happen and formulates it explicitly. As it.is 20 now, there is nothing to account for the added risk by just 21 adding the number of plants.
22 That I think is the concern -- it seems to me to 23 be the concern of Commissioner Asselstine, and I think that 24 concern can ce addressed in various ways. I don't want to
/~}
a 25 --
I think what you should be worrying about is the ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Cm 800 336-6M6
1350 08 10 102 g
(_)DAVbw 1 accidents that result in loss of life. But if other people 2 think that you should be worried about is just core melt, I 3 don't know. Perhaps there are good arguments for that.
4 DR. MARK: Look, I asked a question that didn't 5 have an answer.
6 DR. MAC LEAN: I know.
7 DR. OKRENT: Any another comments?
8 DR. MAC LEAN: No.
9 DR. OKRENT: Mike?
~
10 DR. GRIESMEYER: I must say, I share the concern 11 about cost-benefit analysis, because ostensibly you do it to 12 balance something, and if you leave things out, you're not C)
\' 13 balancing anything. And one of 'the problems s.ith the 14 cost-benefit analysis, traditionally, is that if it is 15 difficult, you don' t do it, and therefore, you screw up the 16 balance, and it becomes very questionable, whether or what 17 you' re balancing , and whether there is a benefit for doing 18 the analysis.
19 i So I agree, we might not want to include that l
20 stuff. There are some problems that aren't easily 21 monotizable, as has boon pointed out by the ACRS several 22 times. The societal resources. You contaminate a large 23 water supply, you might be able to jerry-rig some sort of
{} 24 dollar estimate for it, but it is probably not going to be i 25 that good. You wouldn' t want to make a decision without I
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Com 800 336-6 9 6
r 1350 08 11 103
(~^.
( ,)DAVbw 1 knowing roughly what that is. But you may not be able to 2 put a monetized dollar estimate on it. The fact that you 3 don't put a dollar estimate on it, doesn't mean you 4 shouldn't include it in your analysis in some way, or 5 somehow or other include ,it in the decision.
6 So I think you should at least have it out there 7 someplace in this group of other decision factors, which 8 brings up another point about the matrix. l 9 When I first saw the matrix, I thought that it 10 tended to be a good idea, in that it allowed you to admit 11 that there are several different decision situationsthat you 12 find yourself in and that you should treat them q
/ 13 differently. Upon looking at it, there are several probl'ms e 14 with the way it's set up, because it is going to be 15 extremely hard to implement. But the idea of having several 16 things out there on the table all at once, that you are 17 trying to consider, the three seemed like a good idea, in 18 fact. I am not concerned about the complexity of the 19 matrix. I think that what you are dealing with is something 20 that is fundamentally complex. In fact, simplifying it may 21 be dangerous, so you might want to add some more things to 22 the matrix. You might want to put in slots in this matrix I
23 for estimates of some of these quantifications of these
/~ 24 other dif ferent decisions factors.
L),
25 And if you were to do this, and you just have ACE-FEDERAL REPORTERS, INC.
202-347-3700 Narionmide Coverage 800 336 & l6
1350 08 12 l 104 r'3
(_/DAVbw 1 this matrix of things that you consider in the decision, you 2 can do sensitivity studies. You can try "arious 3 alternatives, and you can see how they might change some of 4 these slots. So you have a better idea of what you are 5 doing, when you make a decision.
6 If you were to have most of the things out there 7 on the table that you wanted to consider, then a little bit 8 of ad hocness in making a decision is probably all right.
9 If you fully quantify several things but leave out other 10 things, I am less comfortable, even if it is a well-defined 11 decision process. And if you have a bunch of things out 12 there and you have looked at them, and you can say, look, I
-) 13 i looked at them, I am much more comfortable with a decision 14 that includes everything than one that doesn't.
I 15 DR. REMICK: Includes everything, Mike. It can 16 get pretty ridiculous.
17 DR. GRIESMEYER: Right. So what you need to do 18 is come up with a list of things that you want to consider, 19 and that is what I would like to see in the matrix. I think 20 there are a few more things that you might want to consider 21 than what is in there at this time.
22 23 l
/~T 24 l
(_) i 25 l l
ACE FEDERAL REPORTERS, INC.
' )
202 347-3700 Nationwide Coverage 800-336-6646
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _J
1350 09 01 105
, ems U DAV/bc 1 If you want to do something in short order, 2 you're not going to be able to do this. You may just want 3 to clean up the ones you've got in the manner suggested by 4 Lester.
5 But, eventually, you're going to want to have i
6 some guidance in this matrix about how to deal with some of 7 these uncertainties, because the number in that matrix are 8 going to be extremely uncertain. t.nd the fact that you 9 calculated them and got a point estimate doesn' t really make 10 you feel all that good.
11 I'd like to have some guidance in the matrix 12 about uncertainty. This matrix with perhaps a containment
( 13 criterion, would look similar to the hazard states suggest'ed 14 by the ACRS several years ago. The reason they were put up 15 tLere was to have several things out there on the table that 16 are considered in the decisions. Explicitly considered in 17 the decision.
t 18 There was some defense in depth as well as some 19 attempt at accommodating uncertainties. And I think the 20 l matrix should have that sort of thing in it.
21 MR. SNIEZEK: Dr. Okrent, could I add something 22 i just as a point of clarity?
23 DR. OKRENT: Please.
() 24 MR. SNIEZEK: The matrix would be a philosophical 25 implementation of how we approach the issue. I agree with ACE-FEDERAL REPORTERS, INC.
P riu m . m ce m _______ m m m _ _ _ _ _ _ .
I 1350 09 02 106 m l bDAV/bc 1l you, there needs to be definition by the staff of what's 2 considered in the cost benefit analysis. We have several 3 staff documents that exist today, NUREG 0058, which is our 4 guidance for regulatory analysis; NUREG 3568, I think it's 5 coiled the Cost Benefit Handbook, which lists the factora 6 that we're supposed to consider in arriving at a decision.
7 So it's not just the matrix standing alone. It's 8 other documents which may have to be modified if the l
l 9 Commission would adopt the safety goal approach.
l l 10 DR. GRIESMEYER: I think a policy statement on 11 how you're going to approach safety goals should at least i l
12 some place in that policy statement, whether it be in the i
(
' ' matrix part of it or not, it should in the policy statement 13 l
14 somehow or other enumerate the things that are going to be 15 considered; because, otherwise, it's incomplete.
16 We're going to have those things. And then these l 17 ' other things that you can find by thumbing through all of l
18 our regulations, or all of our support documents. The 19 policy statement is, I think, a lot. The public relations 20 document, you want to make it clear what you're considering.
21 And if you only have the matrix in some of these 22 things, in this policy statement, the other decision factors i 23 aren't there and it's not really clear at all what's being
(] 24 considered.
25 MR. SNIEZEK: I would have to disagree. I think, ACE. FEDERAL REPORTERS, INC. l
_ _ _ _ - - - _ - - - - - - _ _ _ _ _ - . - - - - - - - - 2 3 DT-D G GtBcch CENEX~3
t l
l j 1350 09 03 -107
()DAV/bc 1 when you issue a policy statement, it's a guiding principle 1
2 that you're issuing, not a detailed implementation 3 document.
.I
- 4 DR. REMICK: I thought the things to be concerned ; '
l l 5 with in cost benefit are spelled out, but different people
! 6 dif fer whether it's adequate or not, or complete enough.
!.. 7 And I think, to try to spell it out, I don' t know ,
) 8 if you've ever been involved in litigating things that are [
t j 9 considered in NEPA consideracions, you know, it's the loss ;
i 10 of this historic house, access to the public, and all that ;
i i 11 kind of thing. It could go on indefinitely. [
i !
- 12 Every one of us, depending on the situation, can
(:)
i . 1 i
- 13 I imagine all kinds of things that are important to us to be ,
I
{ 14 considered, and it's not in there. :
i E 15 l You could not take into account all these things j 16 even if you tried. It is impossible. I think you have to -
r l 17 truncate it somewhere from a practical standpoint, or you'd i j ;
18 litigate forever. i
)
i 19 DR. GRIESMEYER: But, some place, these other _
l! 20 decision f actors must be spelled out. Are they? ;
i
- l. 21 MR. SNIEZEK: I mentioned the two NUREG documents j l 22 which guide the staff today, without cicsing the books, ;
i 23 whether they would have to be modified or not if the
(} 24 Commission went forward with a policy statement on safety 25 goals.
I ACE FEDERAL REPORTERS, INC.
202.W 3700 Naionwide Om:g_ 800yM4646_
1350 09 04 108 t'~ \
L )DAV/bc 1 MR. LAVE: Could I make one other point? I think 2 this portion of the averted on site costs seems to me to 3 have another problem with it. In the benefit cost analyses 4 I've done of the cost of increasing air safety or the safety 5 of dams or building codes, what you'll find in every one of 6 those cases is that they are, quote, " too safe", end quote.
7 That is, they are built much safer than any benefit cost 8 analysis would ever tell you to build them.
9 I just went through our National Academy of 10 Science's committee looking at dam safety, and the proposal 11 to spend billions of dollars there to make dams which have 12 failure probabilities certainly less than 10 to the minus 6 O 13 i or 10 to the minus 7 per year still safer. $
14 You try and take a look at the expected cost of life saved. You get into the hundreds of millions of 15 f 16 dollars. What goes on is that, in all of these other 17 aress--forget about nuclear reactore for a second--the 18 ,
public declares itself for low frequency, high consequence i
19 events, we want a lot more safety than would ever be 20 justified by benefit cost analysis. And that, therefore, 21 using a proportion of the averted on site costs there is 22 just not recognizing that the public is going to demand more 23 safety than any one of these calculations will ever do.
/' 24 My assertion is that when you're doing a benefit
)
25 cost analysis for nuclear reactors, the benefit cost ACE. FEDERAL REPORTERS, INC.
EMSSD Rtt:cC@nna CDHDC"3
'w
. r 1350 09 05 109 is'DAV/bc 1 analysis is going to give you a lower bound than the amount 2 of safety that the'public demands.
3 So I don' t see any reason to kind of water it 4 down by having a sliding'acale. Take a look at what is 5 being spent for air. safety, for dams or for buildings in 6 terms of the expected cocts per life saved.
7 You'll see that nuclear reactors are not too 8 safe.
9 DR. MARK: Lester, could you add to that, you 10 spoke of having been involved recently in a means for 1
11 improving dams. You said that the public demands something 12 or other.
13 What and phere is this public and in what form 14 l ,
does it demand anything, or even that it be heard?
~
15 MR. LAVE: I guess when I refer to dams, what I'm 16 talking about is the engineering professionals at the Bureau 17 l of Reclamation and the Corps of Engineers.
I 18 DR. MARK: That doesn't talk about the public to 19 i me.
20 MR. LAVE: But they're the ones who march off to 21 Congress with demands for very large public appropriations 22 to make those dams safer on the grounds that a failure 23 . probability of 10 to the minus 6 or 10 to the minus 7 per
() 24 year is hopelessly large, and that much has to be done.
25 I take your; point well about the public.
9 ACE-FEDERAL IREPORTERS, INC.
sesmsm _
1350 09 06 110
$ DAV/bc 1 DR. REMICK: It sounds more like job security.
2 ( La ug hte r. )
3 DR. OKRENT: Well, I must say the public 4 sometimes appears for an event. Their interest is aroused, 5 correctly or incorrectly. It certainly appears after an 6 event.
7 DR. MARK: Oh, absolutely. And the public 8 downstream from this dam might be very vocal.
9l :
DR. OKRENT: So it's a perplexing question which 10 I, myself, have never found the answer as to how one allows 11 for this public interest. You may recall, in the original 12 version of NUREG 0739, we had a risk aversion factor which 13 was way up.
14 i
\ DR. MARK: An exponent of 201.4 or something?
15 DR. OKRENT: An improved way of somehow allowing 16 for this public interest, which I think is real. And it 17l I does have differing effects on different technologies, and 18 f so forth. And that seems to be a fact of life. I don' t l
19 : know what else to say.
I 20 ' DR. MARK: Well, I would like to say only that I 21 understand what you're pointing at, Dave, and I know it's 22 real. I don' t know how to measure it. I would like to see 23 the same risk aversion factor applied to man-rem. If it's
) 24 man-rem at the rate of a millirem, it will have to be with a 25 factor of 1.4 with respect to one rem, or something of that ,
1 1
ACE-FEDERAL REPORTERS, INC.
8515-!SE Rttafbcesnca ENDc"3
1350 09 07 l 111 3
i
'. DAV/bc 1 sott.
2 And that would mean we would stop talking about 3 it. i<
1 4 DR. OKRENT: I can' t discuss it. In any event, l 5 are there any more comments on this aspect of today's ,
I 6 agenda? We're a little bit now beyond the appointed time.
7 (No response.)
8, DR. MARK: I don' t worry about that. I worry i
i 9 about the fact that our basic time is 8 o' clock tonight. !
10 , (Laughter.)
11 DR. OKRENT: Anyway, let me thank our consultants
,, 12 very much. We really appreciate your taking the time on
)
! ('-
13 l short notice to review this and provide another perspective
! l 14 ' on this than we usually have in our more parochial group.
15 Let's see. Why don' t we take five minutes even 16 though we're a little late, and start again in five minutes 17 1
on the next group. That will leave us 15 minutes behind the l
I 18 l agenda.
19 (Recess.)
I 20 DR. OKRENT: Going to the next part of the 21 agenda, I believe Mr. Sege is the spokesman for the staff.
22 And if we can, we'll keep this to 30 minutes. In fact, we 23 must.
(~') 24 i (Slide.)
s 25 MR. SEGE: Thank you, Mr. Chairman. The ACS has ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-346M4 _
1350 09 08 112 DAV/bc 1 recommended to the Commission, and the Commission took that 2 recommendation into account, that consideration should be 3 given to a containment objective along with the other 4 quantitative objectives as part of safety goals.
5 As Jim Sniezek mentioned, a plan to develop a 6 quantitative, obtainable performance objective has been 7 submitted to the EDO for submittal to the Commission in the 8 process now of review in the office.
9 The effort has been started. There's an active l
10 effort going on. There's an effort between Research and 11 NRR. We are working on a paper to the Commission for the 12 latter part of this year.
13 The work we have done so far includes the conduct 14 , of a number of interviews with people outside as well as 15 within the NRC-ACRS community to get a broader range of 16 viewpoints to start in our thinking that would be available 17 just within the resources of the staff working on the I
18 h project.
I 19 i We have identified for further study five I
20 quantitative containment performance objective options. I 21 will talk about those options in terms of descriptions, pros i 22 and cons. We have also given consideration to the general l l
23 merits of having a CPO option and what the general values of !
24 a CPO would be. And we are considering the comparison of 25 having a performance objective against a no action option ACE-FEDERAL REPORTERS, INC. ;
202 347 3700 Nationwide Coverage 800 336-M46 l
i 1350 09 09 113 l
(^ I h3/DAV/bc 1 and the possibility of a no action option and some 2 qualitative guidance that may be promulgated.
3 We have started an effort to make a quantitative 4 comparison with respect to these initial strawman options 5 against existing containment performance assessments without 6 any detail, but just to see generally what the problems of 7 applications would be and what sort of results might I
8 eventuate from such application.
9 This work has just begun. As far as 10 implementation of the CPO is concerned, it will be 11 implemented as part of the safety goals, as one factor among 12 others -- not in a go-no go fashion. To the extent that it
() 13 will influence future plants,where the cost benefit 14 relationships will be available, that unlike future plants, 15 for existing plants it might provide guidance that might 16 result in some limited fixes, the application, which will be ,
17 coordinated in the application will be the CPO action 18 policy.
19 (Slide.) .
20 The first of the options that we considered is an 21 adaptation of the containment performance goal that was 22 included in the 1980 ACRS document, suggesting for 23 consideration along with other approaches. We modified the
() 24 i ACRS approach somewhat to shed into the current general 25 nature of the safety goal policy definitions.
ACE-FEDERAL REPORTERS, INC.
- - - -_---_ --- Sms mnt _ _ __ mmeerc~t-_ ---____rwev-t
i l
1350 09 10 114 O
L /DAV/bc 1 That objective, as you may recall, although it 2 was stated somewhat differently, involved essentially an 3 conditionti probability of large-scale, uncontrolled 4 release, given large scale core melt of not more than .1.
5 There is in the ACRS objective also an aspiration 6 level which would be a factor of 10 lower than that. We are 7 stating that as a reduction in that conditional probability 8 for future plants to the extent justified on a cost benefit 9 basis.
- 10 There have been various reservations expressed 11 about cost benefit earlier this morning. I will not read
, i 12 the entire slide, but among the factors favoring such an
() 13 i
~
approach is that it documents the general basis, and'that,it 14 relates directly to the basic containment function of being 15 there, but it has some problems, some of which are inherent, 16 some of which may be fixable.
17 The single probability of a single release spec 18 may be too crude a measure for considering the various types 19 of releases that might be involved if the containment is 20 called upon to function.
21 So, in the event that this is the option that 22 would be adopted for use, I would think that a consideration !
1- l 23 would have to be given to outdoing some of its nice
(} 24 I
simplicity with some necessary complications.
25 DR. MARK: Before you go entirely away from that, ACE FEDERAL REPORTERS, INC.
- N 4hTQ. Patirt'bh - _ _ E0#7^:1_
1
.1350 09 11 115 DAV/bc 1 was the ACRS -- I'm sure it's possible -- ever so mad as to 2 stipulate 90 percent of the noble gases? First, that
.3 doesnt' buy you anything. Sacond, there's no possible way 4 of insuring it.
5 Ten percent iodine, I almost understand. Did we 6 ever say 90 percent of the noble gases? And how could we 7 have done so? On a Saturday morning maybe, after 8 everybody'd left, or what?
9 (Laughter.)
10 DR. OKRENT: I really don' t remember what's in 11 , NUREG 0729.
12 MR. SEGE: It does pose some difficulties in that O 13 , people calculate releases of 80 percent cf th'e noble gase,s I
14 i and we say that's not a major release. Clearly, the I
15 situation is more complicated than this rather simple 16 definition would properly reflect, and we recognize that, as 17 a practical matter, consideration would have to be given to 18 making it a bit more elegant.
19 i DR. MARK: It looks so simple up there. And it 20 seems so strange to me. ;
i 21 MR. SEGE: We haven' t revisited that yet. We 22 recognize the need to revisit it. I 23 ! (Slide.)
() 24 , A second option described on this slide 25 essentially defines the performance of containment in terms ACE. FEDERAL REPORTERS, INC.
-_____r%r%Em _ _______rcrv~T? h carr r n _ ___-_ M______________
l 1350 09 12 116
,O L /DAV/bc 1 of attenuating the release to the environment of those 2 fission products which would be airborne but would be ,
f 1
3 deposited as solids.
4 In addition to specifying a hundredfold desired r 5 reduction, the option also specifies how sure we ought to be 6 that the containment actually is there and fulfills its 7 function to this extent. And that is indicated by the 8 specification of the frequency with which the solids release ,
9 limit may be violating.
- 10 That is qualified with a figure with respect to 11 the nonsolids which is probably too strict, but we consider 12 that. What this does is characterize containment O 13 performance directly in terms of its basic function with 14 respect to radioactive material release.
I It has problems in that the assessment would be 15 l 16 subject to a maturing technology of source level i 17 predictions. And as we understand things better, or at 18 l least differently, we may add some uncertainty to the
{ 19 interpretation of this objective.
20 As with the previous option, the model on a 21 l large-scale release and the distinction only in two bins, 22 the solids and nonsolids, may be somewhat too crude for a 23 realistic assessment of the performance.
() 24 l
25 ACE FEDERAL REPORTERS, INC.
n . _ - ._ -
l 1350 10 01 117
.p P DAVbur 1 (Slide.)
2 There is a somewhat similar alternative 3 formulation as option 3, and we are considering pessibly a 4 single option that might pick up the better features of both 5 and try to avoid the pitfalls.
6 But this would specify the containment function 7 in terms of reduction of curies rather than reduction of 8 solids. It specifies that as within some selected number of 9 hours, some selected amount of time.
10 Weput72inquotesbecausewejustpickedsome 11 number as a marker without really having determined what 7s 12 length of time might be appropriate in this context.
i
(..) .
13 This emphasizes the radioac,tivity function of l
14 j containment in terms of more direct measurement. Even l
- 15 curies don' t relate to health ef fects that directly. It is 16 , a relatively simple statement that could easily be
! l
! 17 communicated to very knowledgeable members of the public.
1 l 18 However, it has some problems in that there is no l 19 provision here for low core melt probabilities, so that the I i l 20 ' consideration of this objective in trade-offs would not be 21 aided by this particular statement of the objective.
DR. OKRENT: Would you say there are no release 22 f 23 failures?
l
() 24 MR. SEGE: It would say that we strive for 25 containmenc to function this well when it is called upon to l
l ACE FEDERAL REPORTERS, INC.
u----------- - - - - - - - - - - - - - - - - - - - - - - - - _ - - - _ - - - - - - - - - - -_
1350 10 02 118
(~) l h / DAvbur 1' function. This objective is silent on the probability of 2 failure.
3 That is a defect, and that is clearly one of the 4 reasons why we thought that perhaps abandoning this as a 5 discrete option and considering it, some of its features, as 6 a variant of the previously presented option might be more 7 complete and therefore more useful to the CPO.
8 Incidentally, I would venture -- not all of these 9 options -- that these were developed with very little 10 background work. We qualified them somewhat on the basis of 11 a little more background work, but we have a lot more to do 12 ' before we will be able to judge the merits of this
[)
13 h particular mode of formulation. ,
f 14 (Slide.)
I.
15 We put in numbers to illustrate how a particular 16 framework would specify the CPC's work when we find a mode 17 of formulation that is attractive, and we will still have to 18 bid down the numbers better than we have so far. We still 19 think that having some numbers rather than X's and Y's is 20 useful at this stage.
I 21 The flexible approach involves two elements of 1 22 specification, both in terms of the frequency of a major 23 release from containment, of course, which is a product of
() 24 the core melt probability, and the initial probability of 25 containment failing.
ACE FEDERAL REPORTERS, INC.
rmmvert e cd me r , Inmmm ____
1350 10 03 119
( )
w DAVbur 1 But it also specifies a floor for containment 2 reliability, an absolute limit on the conditional 3 probability of failure that limits trade-offs within this 4 first part of the option.
5 We call it the flexible option because of the 6 degree of flexibility that these two parts of it suggest.
7 This clearly provides credit for low core melt 8 probability. It is tied to overall plant safety as well as 9 limiting trade-offs to assure an adequate degree of defense 10 in-depth detection.
11 It is structured entirely in terms of 12 probabilities, so one of its drawbacks is that it doesn' t O 13 make fine riskiness distinctions, the distinctions between a 14 major release and not a major release.
15 (Slide.)
16 Then, because of various difficulties in arriving 17 at probabilistic estimates and partly also to provide a sort 18 of intellectual defense in-depth against the weaknesses of a 19 PRA approach, it has been suggested to us and we are 20 l pursuing the possibility of a nonprobabilistic, or rather a 21 mechanistic kind of quantitative CPO, described on this 22 slide.
I 23 Again, the length of time involved in these Q 24 specifications -- if we decide to go this way -- these will 25 have to be given further consideration. Let us pause for a ACE. FEDERAL REPORTERS, INC.
. _ _ _ _ - - -_ --_ _. Smewen - - -__ _-- _----_ --- e _ _ _
I 1350 10 04 120
(~3 ;
' L /DAVbur 1 few seconds while we look at the slide. ,
i i 2 I mentioned the motivation and the advantagos of 3 the mechanistic approach. It also has some limitations and l 4 drawbacks in that the degree of assurance with which these 5 mechanistic goals will be met cannot be specified without a l 6 PRA-based approach. So we either leave it sort of vague or ;
I 7 we do contaminate the non-PRA approach with the PRA l I
8 approach.
9 I am afraid I am not using the word " contaminate" f
10 advisedly in this context.
I 11 DR. MARK: That first item under the pro list, t 12 you want a defense in-depth against oversights.
O 13 How does one go about that?
14 MR. SEGE: I am sorry, I didn't hear the l 15 question.
16 DR. HARK: Your first pro item, how dcas one [
i 17 imagine constructing a defense in-depth against oversights? ,
18 MR. SEGE: This is the intellectual defense 19 in-depth that I mentioned. If we say that our containment :
20 has to contain to certain limits regardless of what we 21 figure will come out with what probability if it fails to do 22 so, then we just said it has to be strong enough and tight 23 enough to do certain things, and if we refigure the source .
I
(]) 24 terms and refigure the probabilities of various accident t
25 sequences, it would not affect this mechanistic effect.
ACE. FEDERAL REPORTERS, INC. l
. -- rvTJM __ __ _ _We . E___ M _'
i 1350 10 05 l 121 l
DAvbur 1 So if we were wrong yesterday in how we figured 2 these PRA results and we figure them differently today but
, 3 are still wrong in a dif ferent way, then this would then, l
4 having this last barrier to release there, having a certain l
5 degree of assurance as a matter of eventual engineering 1
6 practice, provide this sort of intellectual defense 7 in-depth.
8 That is the concept that this is intended to 9 articulate. Some people may feel this is particularly 10 potent. Some people may not consider it particularly 11 potent.
12 DR. MARK: Okay.
O 13 (Slide.)
14 ! MR. SEGE: The Commission has not definitely 15 decided to include a containment performance objective, but i 16 rather we are preparing a CPO for the Commission's l
17 i consideration for inclusion. ;
18 In order to make the staff work complete, in 19 addition to searching for a suitable quantitative CPO, we 20 are also examining the merits of having one in the first 21 place. Possibly after we have seen how successful we can be 22 at devising a quantitative CPO, we may be able to develop a i 23 recommendation for the Commission which says this is the i best we can do and in the balance of thinking will be it
(]) 24 !
25 will be good for you to include it, or this is the best we l
l ACE.FEDER AL REPORTERS, INC.
! _ rm m m~n __ __ e _ - _ _ _ _ _ _ _ _ _ .h_ _.
1350 10 06 122 A
V DAVbur 1 can do and in balance perhaps you would be better off still 2 not to include it because there are still some problems with 3 it.
l 4 The "no action" option would be one quantitative 5 objective less, so it would be only three-quarters as 6 complicated as having three of them. It does not add 7 redundancy to the current quantitative objectives, in that 8 in a sense it multiplied. Combining the CPO with the core 1
9 melt frequency, it comes close to duplicating the health 10 effect, mortality risk objectives.
l 11 It retains greater design flexibility for the 12 industry. It avoids, of course, the difficulty of suitably O' 13 quantifying the containment goals. But it falls to provide 14 l reinforcement to defense in-depth and fails to focus on 15 containments as the last safety barrier that would be 16 intended by adding a CPO to our safety goals.
17 DR. OKRENT: If I can give you a tongue in cheek 18 comment, there could be one other alternative if the 19 evolving spirit of what I guess you might call l
20 self-regulation by the industry and you could just put out a 21 policy statement exhorting the industry to improve the 22 performance of its containment in the face of severe 23 accidents.
l 24 MR. SEGE: Undorstood.
j 25 DR. OKRENT: We are going to get that ACE. FEDERAL REPORTERS, INC.
_ _rmpTimm _ Wu , t- - _ _ ___e
1350 10 07 123
(~)
a /DAVbur 1 recommendation in the last item of the day. So it is not 2 that the staff is not thinking about this.
3 (Slide.)
4 MR. SEGE: I understand your comment.
5 A negative decision about a quantitative CPO or 6 simply a "no action" alternative or in connection with a "no 7 action" alternative, the Commission would still articulate 8 possible qualitative guidance that picks up what it believes 9 to be essential in the function and performance of 10 containment.
11 And we have a slide that would indicate this r 12 modification of the "no action" format position. It has O 13 some of the advantages and disadvantages of the "no action" 14 alternative. It clearly would not have the procision of a 15 quantitative CPO, but as compared with the completely "no 16 action" alternative, it would express Commission views that 17 would be taken into account in subsequently developing 18 specific and quantitative departments for containments, 19 l taking some Commission guidance into account.
20 (Slide.)
21 I mentioned the overall schedule of the project, 22 but my last slide identifies the next steps that we had in 23 mind if we are going to complete that limited simple-minded
() 24 comparison of the options against oxisting containment 25 assessments in the next few weeks.
4 ACE FEDERAL REPORTERS, INC.
rnfmssrm __
re>,vT++s .. i ___---__rmmy"~C _ _ ___.
1350 10'Ot8 124 DAVbur 1 We are reducing to writing the results of the 2 work to date.
3 Also, ir. the course of the next few weeks, we 4 have a workshop planned as outlined in the plan that is on 5 its way to the Commission, in which we will solicit the 6 views of approximately 15 selected participants drawn from a 7 broad range of backgrounds.
8 This is going to be a technical kind of workshop 9 in which the technical content will be substantial. So the 10 participants will be technical people who possess knowledge :
11 in various areas relevant to the subject and represent l 12 viewpoints in conflict with each other. So we can debate
' C:) 13 the major issues of the formulation of a CPO.
14 The workshop will have before it some limited i
15 number of options. We are thinking in terms of two to five 16 options, and we may possibly come up with something better 17 than we have thought of, and we will be requesting a meeting l
l 18 with this subcommittee just before the June ACRS meeting, at 19 which time we will be able to present quite a bit more of l 20 the results than we will be in a position to present now.
i 21 Also, that will precede by a month the ACRS
- 22 review in July, which has been put in the plan that is going '
- 23 to the Commission requesting an ACRS review for July and
() 24 review by the subcommittee a month earlier to allow some l 25 time between these two views. l i
i
- ACE. FEDERAL REPORTERS, INC.
._ _- __ rnriuirrL - _ _ mum i t _ _ _ M - - _ _ ___ __ .
1350 10 09 ; 125
('s L/DAVbur 1 A first draft of the Commission paper will be 2 attempted in May or June certainly if one takes the workshop 3 into account, and any results of the review by this summer 4 will be made immediately available to us, and we would like 5 to take it into account in a couple of weeks after that, two 6 or three weeks after that, to try for a Commission paper, 7 for an initial draft Commission paper.
8 And we thought that we might make it available at l 9 least to the subcommittee and the entire ACRS sometime 10 before the July meeting.
11 This completes the stated summary of the work of 12 the CPO.
13 We have here representatives of differant 14 organizations participating as far as the working group is 15 concerned.
16 DR. OKRENT: I have one question. It relates to 17 what you think is the depth of this comparison of options 18 versus existing containment resistance in the depth to which 19 you plan to do t'.is and whether the information concerning 20 containment behavior is adequate for your needs and what you 21 have time to do.
22 MR. SEGE: I don' t know whether " depth" is the 23 right noun to use. I would think that " shallowness" might
() 24 be more appropriate.
25 It is a very limited effort, and we have no ACE. FEDERAL REPORTERS, INC.
rmMumn .. e _ _ - . In my""t___________ _ m
~ - __ _ _ _ _
1350 10 10 126
^ I i
(V DAVbur 1l illusions about what can be accomplished within a few months 2 in such comparisons.
3 What we try to accomplish is to soo what the 4 practical obstacles may be to the implementation of theso 5 various options, and we have run Anto that and we have run 6 into problems in definition of the various terms used in the 7 description of the options.
8 It helps us pinpoint what those problems are that 9 have to be resolved. Of course, in doing that, there are 10 several options, and in doing it, half a dozen different 11 plants have different containment types. So the breadth l
12 will be there.
O . 13 We are more interested in exhibiting problems 14 ; broadly than in arriving at a real dependable evaluation in 15 these comparisons.
16 We also look for results that might be perverse
- 17 in the senso that one option may show a clearly unreasonablo 1
18 l result in evaluating a particular concainment in a i
19 , particular plant. If that is the caso, then that might be I
20 exhibited by these evaluations.
21 Those don't have the depth and the detail of the 22 trial evaluations that were dono over a two-year period for 23 the safety goals. It is a much more modest offort, but its O 24 ;
i ooe=t c ne ="o it ee== o8$ective we ent"x < a t"u to 25 be very helpful to us in sorting out the various approachos t
ACE FEDERAL REPORTERS, INC.
. _ _ _ _ _ _ - _ _ _ _ _ _ _ __________-__________smsmsen_-_______ m _ ____ m -
.. . . ~ . _ . . _ _ _ . _ . _ _ . - . . .__._ - .-
1 I
1350 10 11 127 '
l l DAVbur 1 to CPO formulation and to that aspect of the future 2 implementation guidance that might guide how we impact the 3 assessment.
4 l
l 5 6 i 7 ,
8 l 9
i 10 11 12 l 13 I , [
l 14 l l
15 i l
16 t
/
17 t
18 f i
19 .
20 ,
i 21 l 22 23 l l
l O 24 i
l 25 !
t l
l I
ACE FEDERAL REPORTERS, INC. !
1
- - _ __ _F~hmik%'b __ _____-_ M w , u . _ _ - _ _ _ .m________________l
1350 11 01 128 m I can sympathize with your doino
( )DAVbw 1 DR. OKRENT:
2 what you call a shallow, or what I micht call broad.
3 MR. SEGE: I should have said " broad" rather than 4 "doep." I should have emphasized the positive more than I 5 did, because we do think it is valuable.
6 DR. OKRENT: It sooms to me at some point there 7 will be a need for sufficient depth. I don't know if that 8 is built into your overall plans.
9 HR. SEGE: Cortainly, you are accusinted with the 10 problems that were encountered in the tria) ovaluations in 11 the impression on the safety coal, as a whole, there is no 12 expectation of CPO development would havo loss problems in b'/ 13 that rocard.
14 DR. OKRENT: I will leave the comment as it 15 stands. Yes, sir.
16 DR. HARK: This is the doncription of the status 17 of work on definino an objectivo, rouchly, nomothino liko 18 that. I know that in my own mind, perhaps not conerally, 19 discussino the nood of sayino somethina about containment, i
20 I which might belong in the safoty coal descrintion, there was 21 a tromondous cap in knowino anythino about what containmen*.9 22 could really be counted on to do. There is nothino in thin 23 plan that I happen to notico, which is coina to, lot's say, 24 anness the perform.snco of any containment acainst the
(~}
ts i 25 obioctivos. The containment 19 a complicated thina for ACE FEDERAL REPORTERS. INC.
_ - _ _ _ _ _ _ _ . _- m W _____-_____ m - _ _
l 1350 11 02 129
,( DAVbw 1 which we have really very little data to say how it really 2 performs. And what we need, as soon as we can, I think, is l 3 some assessment or some real measure of how a containment 4 would perform, if subjected to the stresses of an accident.
5 This plan does not, so far as I see, dio into that cuestion 6 at all.
7 Am I richt about that?
8 MR. SEGE: Zoltan Rosztoczy miaht respond to 9 that.
10 MR. ROS2 TOC 2Y: Zoltan Rosztoczy, NRR Staff.
11 Dr. Mark, we have two programs, two somewhat 12 separate procrams which are beino executed tocether. This i
O 13 is just the procram which directly feeds into the safety 14 coal and notential chances or additions to the safety coal,,
15 in terms of containment performance. The other program that 16 we are workina on is the severe accident procram, the
! / 17 implementation of the Commission's severe accident policy.
18 Under that procram, we are lookina at six reference plants.
i 19 These references are selected specifically to represent the 20 six basic containment types that we have in the country.
l 21 'Jach of those plants are beino evaluated in crest detail, 22 includino the containment performance of each of those l
23 plants.
1
(} 24 The information we are getting there is then fed 25 back into this work, as well, and also into the work of
)
ACE. FEDERAL REPORTERS, INC.
na > n. __ _ _ _ - _ A
l 1350 11 03 130
()DAVbw 1 tryina to decide if there is a need for containment
[
2 criteria, additional guidance, in terms of regulatory i
3 quidance.
I 4 DR. MARK: So this really is to be thought of in l
5 the context of an adjunct to the safety coal. statement.
l l 6 MR. ROSZTOC2Y: That is correct.
7 DR. MARK: And it could be that the no action 8 option is perfectly fine, because this containment you are 9 coing to examine may meet this.
- . 0 MR. ROSZTOCZY
- Yes.
11 DR. MARK: Or they may not.
12 MR. ROS2 TOC 2Y: If, for example, the Commission's 13 feelkna would be that some cuidance on containment is 14 needed, but not in this form, not in the safety coal form, l
15 but in some other form, then it would be done somewhere 16 else. -
t . 17 DR. MARK: Okay. We could decide this is what we I 18 want conteinments to do, a policy statement. You could l 19 decide that present containments meet the coal. Therefore, 1
20 no action.
21 MR. ROSZTOC2Y: There is a little more to that, 22 because you would like to see that the same will be true i 23ll also for future plants.
(~/
\_
) 24 DR. MARK: It will be easy for future plants, if 25 it applies to present ones.
i ACE. FEDERAL REPORTERS, INC.
- _ ____ _ -- _ -- _ _ _ _ ___ -__ - -- _ - _ _ _ F~VSFil'n M_. __M__---_
1350 11 04 131 i ( ) DAVbw 1 DR. OKRENT: Well, there may be different goals 2 occurring in future plants.
3 DR. MARK: Can we scrap this?
4 MR. SEGE: It is also possible that the same 5 objective statement might product different results in 6 future plants than in present plants, becaune of the
) ,7 workings of cost-benefit. It is so much cheaper, usually,
- 8 to make chances in the plant, so that there could be future 9 present plant distinctions in the results, even when there 10 is no distinction in the full statement.
11 DR. MARK: But as Zoltan just told us, t'h : real 12 cuestion that I quess I am interested in is how would 13 containments behave, is being approached under the sevore l 14 accident poli'cy and not under this.
15 MR. ROSZTOCZY: It is being fed into through that 16 little program that Georae has mentioned, and Dr. Okrent 17 asked the cuestion on, the comparison of these options 18 against the existing containment evaluations. What is being 19 done there is, we are checkinc for dose experience. We are 20 analyzing in detail. We are checking for these eix whether ,
4 l 21 these goals would be met, what would need to be done to 22 those plants, so these goals are met.
i 23 The two go hand in hand.
s l 24 DR. OKRENT: I think we'd better move along. One 25 last cuestion.
1 ACE-FECERAL REPORTERS, INC. I
- Inf5MTA -- __ _ _ W.,t _ . -_ _ - e _I
1350 11 05 132
, ( DAVbw 1 DR. REMICK: Georce, other than this committee, 2 are there strong components for a containment performance 3 objective?
i 4 MR. SEGE: Yes. In the interviews that we have 5 conducted, there were strong proponents, strono opponents, 6 and people who tended to be on one side of the fence tended l
7 to be on the other side of the fence, and I would think o there wasn't really a clear preponderance of views in the j 9 interviews clearly on one side or the other. These l
10 interviews were conducted before we did any work. And we 11 will have the workshop, where there will be an opportunity 12 to react to some initial work that had been done. And it l 13 will he interesting to see what the reaction of the aroup at i 14 .the workshop would be and what the range of the actions will 15 be, on the basis of a somewhat more advanced state of i
16 development.
17 DR. REMICK: Will you be coing to the workshop with this spectrum of options, or will you have narrowed it 18l 19 down by then?
20 MR. SEGE: I hope we will be able to narrow it 21 down to fewer options. We thought, we are not shooting for 22 having a single strawman recommendation to the Commission.
23 We are thinking in terms of two to five options, so we have 24 a nanageable number to address in discussion in the course 25 of a day and a half, so we don't cet too confused by tho l
ACE FEDERAL REPORTERS, INC.
_____ __ _ _ - - . - - _ _ - - - - rnewn- _ ___ m t_ _ _ _m___ ___
1350 11 06 133
("N
( ) DAVbw 1 results. So we will be seeking some narrowing of options in 2 the weeks to come.
3 DR. REMICK: Thank you.
l 4 MR. ROS2TOCZY: Mr. Chairman, along those lines, 5 I think it would be a benefit to us, if the Subcommittee, if 6 they have any thoughts on the narrowing down, could give us 7 the comments.
l 8 DR. OKRENTt Thank you. We are not going to take 9 time for that now, because we are coing to try to keep to 10 our schedule. Let's see what develons at some other timt.
11 Thank you. j t
12 MR. SEGE: But we would value anything you mioht l l
i
> 13 want to try to communicate to us in the next few weeks. ;
14 DR. OKRENT: Thank you very much. We'd better f i l
move on to discussions on the Status of the Development of a l l 15l 16 Containment Venting Emergency Procedure Guidelines.
17 We did receive a draf t response I assume the 18 Subcommittee members have seen, and if we can pick up a 1
19 minimum of 15 minutes on the original hour allotted here, we i
20 will still have half an hour for lui ch.
l l 21 MR. HODGES: My name is Wayne Hodaes. I am here ;
i l
22 to answer the cuestions that you had on the Status of the l
! 23 Development of a Containment Ventino Emeroency Procedure '
24 Guidelines.
25 I have a rather short presentation oreparod. I l
ace. FEDERAL REPORTERS, INC.
____--_-_ _ N ---____ N - t__ _--_m
1350 11 07 134
13l qualification limits of a particular valve.
14 And what we're more interested in is having a 15 reasonable assurance and less margin relative to the 16 operation.
17 MR. MICHELSON: I want to be reasonably assured 18l that the control system will still activate the valve, for 19.l instance, even though it may not be a safety-grade control l
20 system. Has it been lost by this adverse environment that 21 created this high pressure?
22 ,
If it has, then you'll never get the valve open.
23 Along the same lines, I think this might have
, 24 been pursued in the past, but it bothers me if it wasn't 25 pursued in the right way, the ductwork problem relating to ACE-FEDERAL REPORTERS, INC.
202-1471Mr0 N Coverage W36 6686
1350 12 14 154
(_mjDAV/bc 1 the vent valve.
2 The concern is, of course, that the entire system 3 of ductwork is internally pressurized to 60 pounds or 4 whatever. Now you suddenly open. Let's assume you are 5 successful in opening the vent valve. You will send a 6 depressurization wave back into'the ductwork, which, unless 7 it's very strong ductwork, will simply collapse. And you'll 8 never vent anyway.
9 I assume you have looked carefully at that and 10 are doing something, putting blowup panels in the ductwork, 11 or something, to take careoof that kind of a problem?
12 You've got to do something to keep the ductwork 13l from squashing into the valve as the depressurization wave 14 goes back.
- 15 MR. HODGES: I think the concern has been more 16 when you open the valve.
., 17 when you open the valve, you blow the ductwork 18 off the wall.
i 19 MR. MICHELSON: The ductwork, the whole system, 20 the whole containment is pressurized. But at the moment in 21 time when you start to open the valve, the very low pressure 22 wave moves back and flattens the ductwork into the valve.
23 You see what I mean, because it's still 60 pounds l 24 outside but, now, you suddenly have very low pressure inside as a dynamic depressurization wave that goes through 25l l
i ACE-FEDERAL REPORTERS, INC. l
__ .____ _ ______________ ________ ESCSED_ __ _ __ _ __ _ __ StCIOGbz~J-- - --- --_ _ _ CEEX"3 -_ _ ___ __ ___ _.______--- - _
1350 12 15 155 Q DAV/bc 1 the inside of the ductwork and then it collapses it because 2 the outside is at containment pressure, which is still what 3 it,we7 before.
4 DR. OKRENT: Are you concerned about any of the 5 temperatures that might exist under these conditions? !
6 MR. MICHELSON: There's another interesting thing 7 we went into. If it's galvanized ductwork -- and I don't 8 know that this is, but if it is, you've got a zinc problem.
9 I guess you outlawed zine on the inside of containment 10 though, didn't you?
11 DR. OKRENT: Yes.
12 MR. MICHELSON: What are you using? Cadmium?
( ~
13 What protective coating do you u,se on the ductwork? Just 14 paint?
15 MR. KRUGER: I think the ductwork is typically 16 galvanized ducting because it's beyond the containment.
17 18 1
19!
20 21 l
22 23 l
l' O 25 ACE-FEDERAL REPORTERS, INC.
.-.-___ _ ____T 45MFA -______ _ N ..., t -_____ _ _ h __ _
1350 13 01 156
()DAVbur 1 MR. MICHELSON: It is inside the containment.
2 You have to have duct work inside the containment.
3 Is that galvanized inside containment?
4 MR. KRUGER: It is a combination. There are some 5 elements designed to withstand the pressurization effects of 6 containment. There are others for normal distribution 7 purposes that would be assumed to fail.
8 MR. MICHELSON: For venting purposes, though, 9 your vent valve is attached to some kind of ducting 10 arrangement that vents around the plant.
11 Is that heavy gauce piping, or is it duct work, 12 light gauge, sheet metal?
( 13 MR. KRUGER: The safety grade that is designed to 14 essentially protect the valves against the imolosio'n that 15 you are discussing right now is the heavier gauge that would 16 be able to withstand, and then there is grating at the 17 termination of that safety grade piping such that you would ,
18 not suffer large debris within the duct. '
19 MR. MICHELSON: You don't want to block the duct 20 either. You got spiral one pipe inside.
l 21 I think that is the kind of questions that I 1 22 would ask. l l
23 DR. OKRENT: Well, we are eating into our lunch 24 time. So let me just note one kind of -- what is the 25 word -- expression of unhappiness, I guess.
ACE-FEDERAL REPORTERS, INC.
__ _ _ __ _____ ___ _ _ __ ___ __ OEF4ED ____. _ _ _ __ ___9tCICXB~;~c~2 - COODC"3 - __ _ _ __ _ _ _ __ _ _ _
1350 13 02 157 j DAVbur 1 I had thought that when the committee wrote its 2 paragraph on Grand Gulf that the ACRS wanted to be advised 3 that the staff and the Commission have had an opportunit/ to 4 comment generically and specifically, that in fact you 5 wouldn't bring it back.
6 I sort of find that something has been approved 7 for Grand Gulf.
8 MR. HODGES: The Revision 2 and I think probably 9 the Revision 3 guidelines were approved before Grand Gulf.
10 There have been no changes to the guidelines from the ll i vending standpoint.
12 DR. OKRENT: I doubt that that was done before we c
' ' 13 wrote the Grand Gulf letter. It seems to me it must be.
1 14 MR. HODGES: I don't recall for sure when thc 15 Grand Gulf letter was written, but it was issued in February 16l '83.
17 DR. OKRENT: The Grand Gulf letter was August 18 18th, '82. Okay.
19 So even though you haven't reached your final 20 position, it seems to me you have reached interim positions 21 and you didn't bring it back to the committee, and I find, 22 for one, that that is not in harmony eith the committee's 23 expressed request.
( 24I I don't knos ouite shy this has occurred, but it 25 seems to me it is something de need to reflect on. Let me ACE-FEDERAL REPORTERS, INC.
202.1J' 1*00 %!innatdm rm crane R(a.1%6Mn
i350 13 03 158 1 put it that say.
()DAVbur 2 MR. HODGES: Thank you.
3 DR. OKRENT: Anyway, I think se sill try to have 4 a subcommittee meeting ahere se have a presentation by -- I
- 5 guess it is NRR, by Research, by the Geners, and so forth in
, 6 the not too distant future, as soon as Savio gets the 7 existing information and se can talk about it, and se sill 8 need more time than today. ,
9 This eas really just an effort to quickly see 10 shere se are. We thank you for, in fact on short notice, 11 getting us an update.
12 Okay.
1 () 13 MR. HERNAN: Dr. Okrent, Ron Hernan, from the 14 staff.
15 We hope to have the formal signed version of this 16 response to you before the seek is over, eith the
, 17 attachments. We apologize for having to give you a draft 18 copy.
19 DR. OKRENT: It das helpful to have a draft to 20 save time on a busy day. So thank you.
21 MR. HERNAN: We sill also get you the other 22 material in ehich you have expressed an interest.
23 DR. OKRENT: All right.
24 We sill recess until 2:00 o' clock, ehen de start 25 sith Davis-Besse.
ACE-FEDERAL REPORTERS, INC.
_ . . _ _ _ _ _ _ _ - - 202 347 3700 _ Nationwide Coverage 800 336 6646 _ ._
J J 1350 13 04 159 3
(J DAVbur 1 (Whereupon, at 1:20 p.m., the subcommittee das 2 adjourned, to reconvene at 2:00 p.m., this same day.)
3 4
5 6
7 8
9 10 11 12 13 14 .
15 16
/ 17 18 19 20 .
l 21 22' 23 l
25 l
ACE-FEDERAL REPORTERS, INC. l muum w= . - c. __ = m su -~ !
1350 13 05 160
(_)DAVbur 1 AFTERNOON SESSION (2:05 p.m.)
2 DR. OKRENT: The meeting sill reconvene.
3 The next portion of the subcommittee sill deal 4 sith the revies of the 1986 Ohio earthquake and the 5 implications for the Perry Nuclear Poder Plant.
6 We have eith us as consultants to the committee 7 Dr. Pomeroy, Dr. Trifunac, and Dr. Smith, and members 8 present at the moment are Dr. Remick and Mr. Michelson.
9 We have scheduled until 6:00 o' clock for this 10 matter.
11 Do the members of the subcommittee sant to make 12 any remarks or comments at this time?
O
(~) 13 (No response.)
14 DR. OKRENT: I am going to ask the consultants to 15 just chime in as they find appropriate. Don't save 16 everything to the end of the meeting, please.
17 Why don't I get us a couple of minutes ahead of 18 schedule for the last time of the day and ask the NRR staff 19 to give its presentation.
20 (Slide.)
21 MR. STEFANO: Dr. Okrent, members of the 22 subcommittee, my name is John Stefano.
23 I am the project manager for Perry representing
(-)
U 24 NRC. Here eith me today are Mr. Hurray Edelman, Vice 25 President of Cleveland Electric Illuminating Company, who ACE-FEDERAL REPORTERS, INC.
EBEG-Di0 __ 9?~t:IfbCt:T1 ____m___________
l l
l l
1350 13 06 161 ;
()DAVbur 1 is the applicant for the Perry Nuclear Poder Plant, and his t 2 staff.
I 3 I understand they are prepared today to provide a 4 presentation on their findings, ehich eill follos my brief 5 opening remarks here.
i 6 With me here today are members of the NRC 7 Technical Revies Staf f sho have actually performed the 8 evaluation that se submitted to you in SER Supplement
(
! 9 No. 9. The people that sill be making the presentation of I 10 their respective sections are those people. ,
l l 11 As you can see on this slide, just a very quick ;
i 12 intro here, the earthquake occurred on January 31st, 1986 at O 13' 11:46 a.m. We have been able to pinpoint, I assume, the 14 epicenter of this cuake about 10 miles south of the Perry l
15 site. It eas reoistered as a magnitude of 5, modified
- 16 Mercalli 6.
17 We have determined that it das an extremely short 18 duration, high frequency type event, ehich essentially did 1 L ,
19 little or no damage physically to the plant, and you sill 20 find ehen de get into those specific aspects of our 21 presentation se sill tell you .4 hat se found and ehy de think r 1 ,
22 it did little damage. , l 23 With me making the presentation today 4i11 be l
{} 24 25 Ms. Phyllis Sobel. She sill present our geological and seismological evaluation of the event and describe any of j
ACE FEDERAL REPORTERS, INC. -
_M-_- m M_
1350 13 07 162 q
(_j DAVbur 1 the related confirmatory items ehich se have concluded need 2 to be done.
3 Mr. Arnold Lee sill be presenting our evaluation i
4 regarding the structural seismic design as well as the 5 seismic aualification of the equipment.
6 And se do have also in addition Dr. Jim Johnson, 7 from the Structural Mechanics Associates, eho das our 8 consultants on the seismic structural design; Dr. Leon 9 Righter of the NRC staff, eho sas our consultant in the area 10 of seismology and geology.
11 In addition, se have the folloeing 12 representatives from the U.S. Geological Survey: I k- 13 Dr. Kenneth Campbell, Dr. Robert Wesson, and Dr. Craig 14 Nicholson.
i 15 Just note. I understand that se recently signed a j 16 contract eith the USGS. They sill be assisting the NRC l 17 staff in the confirmatory sork shich se have laid out in the 18 SER supplement and shich 4i11 be discussed during 4
, 19 Ms. Sobel's presentation.
20 Let me just point one last thing out before se i 21 turn this meeting over to the utility.
22 It is our intent as a result of our deliberations 23 here today and before the full committee tomorros to obtain r~ 24 ACRS comments in agreement eith our action to license the D}
25 Perry Nuclear Poder Plant.
ACE. FEDERAL REPORTERS, INC.
. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ____ _ _ _ __ fVMG _ __ _ _ _ _ __ _ __ _h __ _ _ _ _ CDESC3 -__ __._ ____ _ _ _ __ _ _.___ _ _ _
~- ---
1350 13 08 163
_ DAVbur 1 By license, I mean issue a los poder license 2 shich 4111 enable the utility to load fuel and to operate up 3 to a 5 percent poder level, and in this particular case that 4 eould be apnroximately 178 megasatts of thermal. That is 5 our intent.
6 This is shat se hope to obtain from the full 7 committee tomorros and hopefully soon thereafter a letter 8 shich eill essentially supplement that, which eas provided 9 by the ACRS almost four years ago shen se first came to you 10 eith our safety evaluation report on the Perry Muclear Poser 11 Plant design.
,_ ,12 One other point you ought to be adare of: se 13 ha.ve received at least teo 206 petitions, one from Western 14 Alliance Reserve and one from the Ohio Citizens for 15 Responsible Energy, expressing their concerns over the 16 earthquake event and the impact that that event had on the i
17 design.
18 i We are in the process of responding to those two I
19 206 petitions, and de are prepared to respond prior to 20 licensina.
21 We have also received a motion from the Ohio 22 Citizens for Responsible Energy to reopen board hea;ings to 23 admit the earthquake event as a contention issue. They cite
(' , 24 as their contention that the design basis is inadequate as a 25 result af this cuake.
I 1
ACE. FEDERAL REPORTERS, INC.
% *n?.tti t?on Natinnute Internas EX1116.nM6
)
1350 13 09 164 o
biDAVbur 1 We have responded already to that motion on March 2 5th, and se are avaiting the board's decision on that.
3 If there are no opening remarks --
4 DR. OKRENT: There are questions.
5 I sonder if you sould remind me concerning the 6 status of a paragraph that sas in the July 13, 1982 letter 7 that the ACRS srote on Perry in #hich se said:
8 "We recommend that the Applicant 9 and the NRC Staff conduct studies 10 to evaluate the margins available 11 to accomplish safe shutdoen, 12 including the long-term safety O 13 i
j folloeing an earthquake of someshat 14 longer duration and greater severity 15 than a safe shutdoen earthquake. We 16 believe it is important that there
/ 17 should be considerable assurance that 18 the combination of seismic design 19 basis and margins in the seismic i 20 design are such that this accident !
21 source represents an exceptionally los 22 overall comoonent of risk for this 23 plant and recommend if any needed 24 ! modifications be made. We dish to be 25 kept informed on the progress resulting ACE. FEDERAL REPORTERS, INC. !
- - - _ _ - - _ _ - _ _ _ _ _ ______- _ _ _ ____ F M P W R n _ _-_ M M_
Vbl' t-1350 13 10 165 jx c-( ) DAVbur 1 from the studies."
.2 2 I just can't remember nog. Have you given us an 3 SER on this? Where does this stand?
'4 MR. STEFANO: No, sir, de have not. As se 5 understood that particular statement in the ACRS letter of 6 July 13th, '82, se sere concerned about looking at the 7 design in terms of a much larger type seismological event 8 than aas analyzed in the FSAR and also evaluated in the 9 SER.
10 As I understand it, sir, there das a generic 11 margins program initiated shortly thereafter or sithin that 12 time frame, during shich time matters such as this type of 13 an earthouake eere being considered.
14 To the best of my knowledae, neither se nor the 15 utility has responded specifically to that except being 16 available to provide any assistance that is required in that r
17 effort.
i 18 I have Dr. Leon Richter here eho would perhaps be 19 able to get further into that, beyond shat I can tell you at 20 this point if you eish.
21 DR. OKRENT: Well, if there is someone on the 22 staff sho can add further to what I have said, I am here to 23 listen.
24
- DR. RIGHTER: Dr. Okrent, I think the specific )
O 25 letter on Perry was one of those included in the package l
l i
ACE-FEDERAL REPORTERS, INC. '
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ______ _ ___ ___ _ _ ____ __fW5-9E0 -- - ._ __ Etat'hh ____ _ _ __ __ __ _ _ EZDOLT"1- ----- _ ------- - _ ----.-.---- - j
1350 13 11 166 JDAVbur 1 that helped stimulate the seismic margins research program.
2 That research is nos carrying on.
3 I think Mr. Stefano das correct. No specific 4 action sas taken. This sas dealt alth similarly to the 5 other letters that the ACRS has 4.ndicated this problem.
6 DR. OKRENT: In fact, my guess sould be that the 7 SSMRP program had already begun aben this letter sas 8 aritten. So I don't knoa --
9 DR. RIGHTER: Not the SSMRP, but the ahole 10 seismic margins program.
11 You have to remember the ACRS arote a letter and 12 said -- told the staff that se felt the staff aas not
> adecuately responding to that. This sas part of that shole 13{
14 package of a group of plants shich stimulated that part, .
15' shich aas a takeoff from the SSMRP.
16 DR. OKRENT: Can you tell me hos this paragraph 17 sas going to be addressed for Perry? If it is going to be 18 addressed generically, ahat is the study that does it 19 generically, and shy is generically adequate, and ahen sill 20 this be completed, and aho is doing it, and so forth?
21 I am a little vague in my mind nos just ahat the 22 staff is doing in all of this, f rar.kly .
23 MR. STEFANO: I can give a try on that, and Leon 24 can help.
(
25 DR. OKRENT: If you sould like to sait until l l
l ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336-6646
1350 13 12 167
'DAVbur 1 tomorros and come in eith a sell thought out, crisp 2 response, I am satisfied.
3 MR. STEFANO: Fine, sir, se #ill do that.
4 DR. OKRENT: I indeed sould like to have 5 ,
something that is to the point and not too general.
6
- MR. STEFANO: I understand, sir. We vill do 7 that.
8 DR. OKRENT: Let's make sure that is on the 9 agenda tomorros as part of their presentation.
10 DR. REMICK: Question: on the licensing matter, 11 I assume that the staff position in response to the motion 12 l is to deny the motion, am I correct? You did not state shat
(')
k 13 the staff's position is?
14 MR. STEFANO: I am sorry. That is correct, sir.
15 DR. PEMICK: I assume that. But is the licensing 16 action before the Commission, or can the staff issue the 17 license on their can? Have you gone to the Commission?
18 )
MR. STEFANO: 7 understand that in terms of the l 19 los poder license r N 1
- sue the license, that any 20 decision the board rutght u.ake is in the critical path on 21 full poser, and before de go to the Commission.
22 DR. REMICK: Then shy are you asking for an ACRS 23 latter?
3
(~} 24 MR. STEFANO: I guess the ansser to that question mj 25 is: shen se have a significant event such as this that ACE-FEDERAL REPORTERS, INC.
agr_ . 8@33&MM --
l l
l 1350 13 13 168 c
DAVbur 1 occurred it has been our policy to immediately advise the 2 ACRS of ahat se are doing, and in this particular case se 3 thought, because of the significance of this event and 4 because of the significant interest that has been placed in 5 this, not only from althin but from outside the agency, that 6 se could finally get you gentlemen to see ehat se have done 7 and comment on it.
8 DR. REMICK: So the letter is a staff request?
9 MR. STEFANO: Yes, sir.
10 DR. REMICK: Thank you.
11 MR. STEFANO: You are selcome.
12 DR. OKRENT: One last cuestion, and maybe it is I
?
13 Mr. Righter aho could ansaer it. .
14 Did the latest version of expe t opinion 15 estimates of seismic hazard curves by Livermore -- did that 16 include the Perry site?
17 DR. RIGHIER: Perry aas not one of the sites 18 ; included. The closest site is probably Braideood. I am not 19 quite sure about that.
20 , But as you kno#, if de look at the Livermore or 21 the EPRI program, many people have included seismic zones l
22 that include Perry, and of course the problem is they all 23 disagree.
r 's 24 So it is not a cuestion of there have been 1
1 25 earthcuakes in the past, but ehat is the cause of that?
1 ACE-FEDERAL REPORTERS, INC. ,
- l
( 20246 rmo Nationwide Coverage 800436-6M4
1350 13 14 169 LJ DAVbur 1 DR. OKRENT: I sas just curious as to shether 1
2 that had happened.
3 DR. RIGHTER: No.
4 5
6 7
8 9
10 11 12 O 13, 14 15 16 17 18 19 20 21 22 23 1
Q 24 l 25 ACE FEDbRAL REPORTERS, INC.
1 1350 14 01 170 e-(yjDAV/bc 1 DR. OKRENT: Any other questions for this 2 speaker?
3 (No response.)
4 MR. STEFANO: Thank you, centleman. Mr. Edelman, 5 the floor is yours.
6 DR. OKRENT: By the say, in connection eith the 7 question I posed to the speaker from the NRC about the ACRS 8 letter, tomorros, perhaps se might also hear shat Perry has 9 to comment on the paragraph ehich has been referred to.
10 MR. EDELMAN: Fine.
11 (Slide.)
12 Good afternoon. My name is Murray Edelman, Vice 13 President of Cleveland Electric, in charce of the Perry 14 Plant.
15 What I have here today is the agenda, ehat se 16 plan to cover this afternoon. Basically, I'll cover the 17 introduction and overvies briefly, a brief overvies from my 18 general supervisor of operations, Bob Stratman; shat 19 happened the day of the earthquake and our subsequent 20 actions to that time; the seismic instrumentation and data 2 ?. reduction by Dr. Chen, of Gilbert Commoneealth; seismic 22 design basis of the plant including the evaluation of 23 equipment done subsequent to the earthquake in regard to the 24 margine in our equipment; the earthquake design equivalency 25 by Dr. Stevenson; the earthquake analysis and seismicity by i
ACE. FEDERAL REPORTERS, INC.
mar.m mcwe i swsm
1350 14 02 171 7,
lu j DAV/bc 1 Mr. Holt from Weston Geophysics; the seismicity and 2 injection sells. The reason de bring this up is that at the 3 last subcommittee meeting se had, the question sas raised 4 from USGS of a sell-induced earthquake. We are addressing 5 that subject here today to the bottom line, that it is not a 6 problem at Perry.
7 And a final, overall conclusion by me at the 8 end. In addition to that, se have in the audience to 9 respond to any specific cuestions that any member of the 10 subcommittee or their consultants, se have people here from 11 General Electric sho looked at the equipment qualification 12 program of G.E. suppliod equipment.
O 13 We have Dr. Hall here. We have additional 14 consultants for Weston Geophysics aho have eorked on our 15 program, plus CEI technical people in our operations, 16 electrical and maintenance of our oen equipment.
17 So, hopefully, se have people here eho can 18 address your concerns as they come up this afternoon.
19 DR. OKRENT: Nobody asked the question, ahich is 20 of course a hypothetical question: Hoe long sould this 21 presentation take?
22 MR. EDELMAN: Our presentation is about teo 23 hours2.662037e-4 days <br />0.00639 hours <br />3.80291e-5 weeks <br />8.7515e-6 months <br />, ehich is shat se dere allotted on the agenda. That's
(^}
V 24 assuming that nobody asks a question.
25 (Laughter.)
ACE FEDERAL REPORTERS, INC.
s~vm ren - - - - -- _m m----_____________
1350 14 03 172
!,n!DAV/bc 1 DR. OKRENT: Let me note that the usual 2 instructions, I think, are like 50-60 percent of the time.
3 MR. EDELMAN: We have allowed for auestions in 4 our presentation.
5 DR. OKRENT: It sould be sell if people shorten 6 up a little bit so that there is ample time for cuestions.
7 MR. EDELMAN: I think ee've done that, 8 Dr. Okrent.
9 (Slide.)
10 A brief overvies. The earthquake occurred on the 11 31st. We responded. We initiated our emergency facilities 12 even though de aeren't in license. A subsecuent evaluation
's- 13 by everybody. And I mean the counties, the stato, the NRC 14 said that the precautionary action se took eas good and 15 helped alleviate concerns.
16 We've done subsequent evaluation of the plant
. 17 both from the physical and the design basis, the earthquake 18 analysis, all reaching our bottom line conclusion which se 19 presented a month ago, that it eas not a problem at Perry.
20 The staff has nos completed their evaluation and 21 you have their supplement to our safety evaluation report.
22 We did give you a brief overvios of this, the 23 operating events, at the full committee last month. Our 24 design is confirmed. One item, I did say at the last 25 meeting, that se dere about teo seeks asay from fuel ACE. FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336 4646
1350 14 04 173 I )DAV/bc 1 loading. We are nos complete and ready to load fuel. The 2 plant is sitting there saiting for a license so se can start 3 up operations. That is my brief overvies, consistent eith 4 our objectives to try and keep on schedule.
5 I'd like to turn the meeting over nos to Bob 6 Stratman to give you a brief summary on the plant status and 7 the responses following the earthquake. Mr. Stratman...
8 (Slide.)
9 MR. STRATMAN: It is my intention to briefly 10 described ehat occurred on January 31st and the plant's 11 initial response to that, the observations of the operatino 12 staff, the day the earthquake occurred.
L- 13 Prior to the earthquake ao had a number of tests 14 and other normal events going on at the plant. This 15 included calibrations, ISI testing, a major note shore se 16 sere beginning to move the startup sources into the reactor 17 vessel that morning. Actually, se sere going to do that 18 after lunch.
19 The other major event that eas undersay aas that 20l se aere making preparations to do our Division Teo Emergency 21 Response Testing. That's a major test se do every 18 22 months. Our technical specifications require lining up all 23 of our Division Teo ECCS and ESP eculpment, simulating a 24 LOCA signal and have that eauipment emercency start and be 25 fed from the diesels.
ACE FEDERAL REPORTERS, INC.
- fBDT-Di8 _ __ _ _ __ _ Ftrn3h___ __ _____ CD0DC"l __ _ ________ _ _ _ _
i
'1350 14 05 174 m
( )DAV/bc 1 All the equipment das lined up and de dere making 2 preparations to do that test ehen the earthquake occurred.
3 When the earthquake occurred it aas felt throughout the 4 building, standing in the control room.
5 We had approximately 39 systems, 39 safety 6 systems in operation at the time that sere in standby 7 readiness. I should explain briefly that standby readiness 8 is simply the status of the system shere it's ready to 9 receive an emergency start signal.
10 I have teo slides to give you a brief feel for 11 the systems that se had in operation that day.
12 (Slide.)
13 This is a list of 39 safety systems shich se had
. 14 in operation during the event or on standby readiness. It 15 should be noted that a number of these systems are not 16 completely safety systems, but they had safety-related
. 17 components which are designed to function during an 18 earthquake event.
19 Of these 39 systems, 21 of them have major 20 parameters ehich are monitored in the control room. Those 21 remaining systems that do not have a major parameter, such 22I as pressure or flow or some other parameter monitored in the l
23l control room, se do have enunciators shich indicate a local i
24 trouble alarm and you eould go to a local panel to determine
}
25 any problems sith those systems.
ACE-FEDERAL REPORTERS, INC.
_ _ ___ ____ _ ______ _ __ _ __ __ ____ ______ D 3 P3 -D B -_ _ ___ _ _ Lt r a t D h _______ ___ 0 3 fff % " 1 _ _ _- _ _ _ _
1350 14 06 175 (j
f}DAV/bc 1 Additional to that, the Aris, shich is our 2 computer ehich supports both SPES and our startup testing 3 program, sas in operation and monitors a large number of 4 plant parameters on these systems, as sas the process 5 computer also on line during the event.
6 Both those computers remained on line during the 7 event, and se had monitoring of all these systems during the 8 event.
9 (Slide.)
10 ...this slide just simply shoes the nonsafety 11 systems that I had in operation that day ehen the earthquake 12 occurred. A number of these, about seven of those, had
'- 13 their major parameters monitored in the control room. Like 14 the safety systems, those that don't generally have a 15 trouble alarm that says go to a local panel to determine if 16 you've got any other problems that may crop up eith a 17 system.
18 Additionally, though, for these systems, as sell 19 as the Aris and the prime process computer, monitor a large 20 number of points because se use the Aris during the poder 21 ascension program to do the startup testing.
22i MR. MICHELSON: Maybe I missed that. Is fire 1
23 protection on that list someehere?
{} 24 MR. STRATMAN: Yes, sir, it's on the first slide.
25 MR. MICHELSON: Do you consider yours ACE FEDERAL REPORTERS, INC.
202 147 1700 Nadoneida Coverase R(63364M6
1350 14 07 176 A
( ,)DAV/bc 1 safety-related?
2 MR. STRATMAN: Not exactly. In the quality 3 assurance sorld, it's an augmented quality system.
4 MR. MICHELSON: I sondered. The first slide 5 covered safety-related system and the next slide nonsafety.
6 And I eas looking for it apparently on the drong slide.
7 MR. STRATMAN: Right. It probably should be more 8 properly sheen. Parts of it are seismically supported in 9 the safety-related areas that they support.
10 MR. MICHELSON: I see ehy I missed it.
11 MR. STRATM AN : It's on the safety slide, right?
12 MR. MICHELSON: It eas charged and ready to go in
\~ 13l s11 cases.
I 14i MR. STRATMAN: Yes, sir, it eas.
15I During the earthquake se had no major anomalies I
16 eith plant equipment. All equipment continued to operate as 17 it should. None of the safety systems ehich were on standby 18 readiness, ready to start as part of the Division Teo test, 19 none of those initiated. They all behaved as se sould 20 expect.
21 There sere a small number though of pieces of 22 equipment ehich did in fact trip during the event, and I l 23 dant to go through those briefly.
24 We had the auxilliary boiler trips during the 25 earthquake. We never really figured out shy that happened.
ACE. FEDERAL REPORTERS, INC. l
_--_______-_FM%RER___-_-__ M __ _ M __ _ _U
1350 14 08 177 O)DAV/bc
( 1 One of the trips -- there's nine trips on the auxilliary 2 boiler. It's a nonsafety boiler that supplies house heating 3 steam. It tripped off durino the earthquake. We 4 immediately restarted the boiler.
5 We haven't had any problems eith it since that 6 are earthquake-related problems. The instrument air 7 compressor tripped. We have a highspeed centrifical 8 instrument air compressor ehich has a vibration trip on it.
9 The vibration trip functioned satisfactorily.
10 The earthouake tripped off the instrument air i 11 compressor. When instrument air pressure decayed, the 12 backup compressor came on around 90 pounds, which is the G
kJ 13 correct setpoint for the backup compressor to come on. So i 14 that functioned as designed.
15 We had a relay on the main generator protected 16 circuit that caused a trip. The reason this occurred is 17 that this relay did not have -- it is an under-excitation
, 18 trip on the main generator, ahich eas obviously not in
, 19 operation this day. It did not have potential applied to 20 the relay, so the relay eas free to flop back and forth due 4 21 to the vibration from the earthquake.
22 When it did that, se picked up a trip because the 23 DC to the trip circuit eas applied. We had an anomaly in
{} 24 the suppression pool level instrumentation. When you looked 25 at the recorder for suppression pool level, you had about a ACE FEDERAL REPORTERS, INC.
n s~uman m. e _m
1350 14 09 178
()DAV/bc 1 one and a half inch increase in level, ehich never came back 2 doen after the earthauake.
3 In other eords, se had a stepchange in the 4 suppression pool level and it just stayed there after the 5 earthquake. Subsequent investigations into that instrument 6 at the plant have shoen that se had air entrapment in the 7 sensing line. And the result of our anomaly was that air i 8 entrapment.
9 So, subsequently, se nos have design modification -
10 going into that instrument but a high point event on the 11 tubing run from the suppression pool because there's a high 12 point that you can't vent the air out of.
O 13 ,
MR. MICHELSON: Did you got differontial 1 14 displacement of the suppression pool sater?
4 15 MR. STRATMAN: We don't believe so, sir, not to 16 any great degree.
e 17 MR. MICHELSON: Why sould air entrap in the l 18 instrument line?
! 19 MR. STRATMAN: There must have been a little bit j 20 of splashing.
21 MR. MICHELSON: There must have been something.
22 MR. STRATMAN: Something happened. The shole 23 building moved a little bit. You knoe, se have no say of 24 saying ehat that das to you because of the say se monitor
(~%)
(_
25 it. There's no say to tell hoe much movement I had in that. l l
i ACE FEDERAL REPORTERS, INC.
- _ _ ____-__-_ _ - JmEikMn__ - -__ Ma. . ,i _M
1350 14 10 .
179 m
(,)DAV/bc 1 We had some movement in the eater but I couldn't quantify it 2 for you.
3 MR. MICHELSON: It shouldn't have been much.
4 MR. STRATMAN: No, sir, se don't believe it eas.
5 This eas a Rosemont transmitter. We did not see this 6 problem in any of the other Rosemont transmitters. We had a 7 number of them monitoring levels, differential pressures.
1 8 We didn't soo any kind of behavior in any of the other 9 Rosemont transmitters.
10 As a result of the immediate plant response to 11 this das that the shift supervisor implemented his 12 procedure, which is part of the E-plan shich 4111 be called
( 13 for in,an alert condition. He got a trouble alarm on his 14 seismic monitors ehen he sent back to the panel ehere the 15 lights read out from the seismic monitoring seitch. He had 16 an indication of an OBE back there, so he declared an alert 17 at the plant.
18 We did not notify the state and put the shole 19 E-plan into effect. We discussed that, but se santed to put 20 the plant into a condition ehere se sould have the Tech 21 l Support Center and the OSC available to support any efforts 22 that afternoon se santed.
23 So se, in fact, sent into our E-plan.
t
(} 24 MR. MICHELSON: Was your level instrument located i 25 on the floor, mounted on a rack, mounted on the suppression l
ACE FEDERAL REPORTERS, INC.
.-_ _------_-_-_ S n m u r n _ _-- _ m m 1______. non.na.ma
l 1350 14 11 180
'()DAV/bc 1 chamber, or on a sall, or ehere?
l 2 MR. STRATMAN: Perry is a Mark III design. I'm 3 not really sure. It's mounted on the sall at that drysell?
W 4 or on the containment eall?
5 I believe it's mounted on the containment sall, 6 sir. We have a Mark III, so se don't have a standard i
j 7 Taurus.
1
! 8 DR. REMICK: You indicated that the seismic I
! 9 monitor indicated that you had experienced an OBE? It had 10 calibrated for OBE?
l 11 MR. STRATMAN: I think, when they discussed the
)
12 seismic monitoring instrumentation, they'll be better able O' .
13 to anseer that ouestion. He got a series of red lights on 5
14 the back panel that indicated to him eith the procedure he 15 had that day that he had an OBE.
16 I think they sill be better able to describe the 17 interface beteeen the setpoint and the instrumentation later i
l 18 on.
t l 19 Subsequently into the event, se then put into 20 place the site area emergency. We did that just on the f
21 site. Again, se did not notify the state that se sere doing 22 it, but de did that to get accountability for our people.
23
)I When you reach the site area, then the security
{} 24 25 system ehich functioned throughout the event, the security computer, se sere able to get accountability. It eas a i
ACE FEDERAL REPORTERS, INC.
_ _ _ _ _ _ _ _ _ . .________-_ ______ ____- _ M ___ _ M (PWY"R -
1350 14 12 181 (o._)DAV/bc 1 eay to account for any people that might have been injured 2 or anything, and se sent into that part of our plan.
3 I'd like to say again that other than the 4 observations I've mentioned to you, se sae no equipment 5 anomalies during the event other than these five I mentioned 6 to you.
7 DR. REMICK: None of your computers demonstrated 8 any erratic behavior?
9 MR. STRATMAN: No, sir. Both computers behaved a
10 as they should. None of the safety equipment shosed any 11 erratic behavior. None of the nonsafety equipment ehich was
- 12 operating, other than these ones I have mentioned, shosed 13 any erratic behavior.
14 Immediately follo#ing the earthouake and ehen 15l things settled doen 15 minutes into it or so, se sent out 16 the roving operatori. They sere sent out specifically to :
17 start ealking doen the plant along eith the maintenance 18 personnel from the operation support center, since that das 19 noe manned.
1 20 They started to bring in some reports. We had 21 one report of a leaking flange and se had some reports of 22 small cracks in concrete.
23 As a result of those reports, myself and the
{} 24 senior operations coordinator, ahich is the man sho eorks 25 directly for me in the day to day operation of the plant, he ACE-FEDERAL REPORTERS, INC.
_ sm m w,n _ m- _ __ ___ _ r~w~m ..
l !
l !
1350 14 13 182 ;
d I se e o#e #d ee red the a1 e- I toered 11 are er i
l O o^vedo 1 l
l 2 the plant belos grade and I sent throughout the reactor l~
j 3 building. ,
i j 4 The kind of things I das looking for sere any ;
l l- 5 piping movement that shosed any paint scraped off of a pipe, ;
l 6 any interferences that se might have had, any leaking [
7 flanges, the kind of things you would look at. !
8 Also, I sent doen and looked at the report.s of 9 the damaged concrete. I'm not going to pretend to be a 10 concrete expert. That sill be addressed later. What I sas 11 sere small hairline cracks throughout the' concrete in i t I 12 various areas. l i
13 That sill be addressed a little later. Nothino 14 that I sad sould imply to me that it eas structural damage. l l
15 (Slide.)
i 16 After my tour of the building ehere I sae nothing j i
i 17 except one distorted hangar in the rad easte building, !
l 18 shich, ehen I sent and talked to my people in the rad easte !
19 control room, they told me,that a WR, a eork recuest, had I 20 been submitted on that prior to the earthquake because it j 21 had been distortedt and a leaky f:.ange on the hot eater i 22 heater, which is the other report I act in control. I 23 I sas no other damage. I sae no scraped paint. l 24 I sae nothing that sould indicate any real movement in the 25 plant any4here.
l l
ACE FEDERAL REPORTERS, INC. l
___ _ _ _ _ __ _ _ _ _ _ E D C # D G _ _ ______ _ h __ _ C~MDCD _ _J
1350 14 14 183
()DAV/bc 1 Subsequently, that evening, it eas decided to put 2 together a team of engineers and technicians to go out and 3 do a more thorough survey of the plant and document the 4 salkdoens.
5 The day se did this is de took the health physics 6 maps of the plant, ehich se have for every area of every 7 building, and provide a set of directions of the things 8 everybody eas to look for -- such things as concrete 9 fractures or cracks in concrete, bent hangars, scraped paint 10 on pipes, the kJnd of things you sould expect from an 11 earthquake.
12 We gave that list to everyone. And as a result 13' of that, there dere about 250 manhours spent on that that 14 evening. We had 473 observations. The slide pretty much 15 speaks for itself.
16 of the 473 observations, two dere thought 17 definitely to be caused by the earthquake event. That first 18 one, it should be noted, the heat exchange drain valve found 19 dripping. That valve das drippina through the valve, so it 20 dould be a seat laak and all that eas required saa a manual 21 valvo. It das just to take the handeheel on the valve and 22 tighten it and that leak stopped.
23 The transformer trip is ehat I previously 24 described on the generator detector relay. We had 95 25 indications that dere determined to be indeterminate. By ACE. FEDERAL REPORTERS, INC.
- _ _ __ .--_ _ _ _J'PSE1_ _ --_ mm.2 , CEY"3
1350 14 15 184 O
(,)DAV/bc 1 that, se meant se could neither prove nor disprove that that 2 condition existed prior to the earthquake. So they have 3 been labeled as indeterminate.
4 And 376 of them dere not felt to be related to 5 the earthouake. Of the total of 473 items, 140 required no 6 further action -- I'm sorry. All but 140 required no 7 further action. Of the 140 that did recuire action, they 8 dere dispositioned to be a dork request or a nonconformance 9 report. And se should characterize most of these as being 10 things like replacing lightbulbs. Some fuses, a couple of 11 meter lenses, tightening of flange bolts, tightening of 12 valve packing, that sort of thing. Some cosmetic drysall 13 repairs. Some of the NR's sere dispositioned "Use as is".
14 Others se had to go and do sork on. .
15 DR. REMICK Question. Was the spent fuel pool 16 filter empty at the time?
17 MR. STRATMAN: Yes, sir, the spent fuel pool das 18 tilled. The startup sources dere hanging from lanyards in 19 the pool. And se didn't have -- the levol indicators for 20i the pool ahosed no movement at all.
21 It sould be hard to tell from the level 22 indicator. I don't have a real recorder for that. It 23 didn't shoe any movement up there.
24 DR. REMICK: Nobody sae any slopping around? Was 25 there anybody in tho area?
ace. FEDERAL REPORTERS, INC.
. - - _ _ - _ . -_ __ F~VG4E'L _ __ _ _ _ _ _ _ M_ -_ ___W _-
1350 14 16 185 (m)DAV/bc 1 MR. STRATHAN: I believe there eas one man up 2 near the pool, but I don't think he was actually eatchino 3 it. I sent up probably an hour af ter the event. I didn't J
4 see any unusual setting on the eall or anything. Nothing 5 looked disturbed. ,
6 The refuel machine, there sere no tools or any i
! 7 pieces. Of course, se had all those off the rofuel l 8 machine. We had accountability, but I didn't see anything l r
9 out of place up there. In fact, I had staging and a couple
, 10 of tools and stuff laying on staging in the reactor 11 building, for instance.
i 12 None of that had fallen on the floor. I think [
13 some of it das still on the edge of the staging and stuff.
14 I really didn't see any evidence of any movement.
15
.) 16 17 18 19 ,
20 21 22 23
- 24 [
25 r e
i ACE FEDERAI. REPORTERS, INC.
___rvTurt -_ _ _ __ ev .
- t. m
1350 15 01 186
( )DAVbur 1 ! Other salkdoens and things that eere conducted 2 that day or the follosing day --
3 (Slide.)
4 -- consisted of a site survey.
5 hhat se meant by that das basically that eas an 6 external survey to the plant. We dere looking outside of 4
7 the buildings.
i
! 8 Perry is built on a bluff, basically, that is l-l That bluf f has had an erosion problem, ehich 9 over the lake.
10 de have discussed before. We have had to go into some 11 actions to stem that erosion. That das looked at. Just 12 generally, sindoes and that kind of stuff out in the plant 13 environs. .
14 We didn't see any indications of damage, and that l 15 included the bluff at the time.
I 16 Another survey that das done eas the plant l
17 settlement. We surveyed the monuments for the major 18 buildings. This is done monthly.
19 We decided to go ahead and do that right then to 20 determine if there had been any building movement. We sas 21 some small changes, but all these changes sere consistent l
l 22 sith variations ,se had seen in previous months.
23 The reactor building, for instance, had moved i r3 24 6/1000ths of an inch from the January survey of the V
25 monument. However, it should be noted that the reading se ACE. FEDERAL REPORTERS, INC.
_ . - - - - _ _ - - - - _ - - _ - - _ _ _ _ _ - - - _ _ - - DBOMD _ _--- _ N--____ CDU"C"3
-1350 15 02 187 O
(,y DAVbur 1 sae in February of '86 when se did this survey das identical 2 to the same reading that se had in February of '85.
3 So se don't feel there das any building movement i
4 from the survey that se did of the monuments.
5 We also did a survey of the cooling tower. We 6 have a 525-foot -- de actually teo of these -- two 525-foot 7 cooling towers. The one for Unit I has the basin flooded, 8 and de have been running the circulating sater system.
9 We d.dn't see any damage to the cooling toser.
10 It has great big columns that hold it up, a hyperbolic i
11 toder. We didn't see any damage to any of the structure of 12 the cooling toeer other than ehere the sall for the base of 13 the cooling toser joins the dall for the cireulating dater 14 pumphouse. There is a carbomastic seal in beteeen those teo l 15 pours of concrete. That das leaking a little bit of eater 16 and made about a 10-inch puddle on the ground.
f 17 We have seen this before after freeze and 18 thaes. It is not unusual for us to have to go out and 19 reseal this joint either in the spring or sometime during i 20 the sinter. In Ohio se tend to see severe temperature 21 variations rijht there on the lake.
i 22 So de don't believe this sas earthquake-related, 23 but de can't prove that it easn't either.
24 The last thing, se had seismic clearance
)
25 violations, which de have in the letter, that de have ace. FEDERAL REPORTERS, INC.
rwsm-__mu e w
1350 15 03 188 DAVbur 1 committed to clear out by 5 percent poser.
2 Also in that letter das a requirement to have all 3 seismic clearance violations in the reactor building 4 rectified prior to fuel load.
5 We still had 29 seismic clearance violations 6 outside the reactor building to be corrected the day the 7 earthquake occurred.
l l 8 It das decided at that point, after the 9 earthouake, to go take a look at all 29 of those. These 10 should be categorized mainly as interferences; i.e., a pipe 11 hanger doesn't have enough clearance from the next pipe over 12 or something like that and de are dorried about the movement 13 during an earthquake and an interference beteeen an adjacent l 14 pipe, an adjacent dall, an adjacent instrument, 4 hat have 15 you.
16 We sae no problem eith any of those 29 seismic 17 clearance violations that se had documented out of a number ;
18 of them that de had previously documented. It is still our j 19 intention to correct those. ,
l i
- 20 But I just santed to mention that se didn't see l
21 any problem eith those.
22 The last thing I would like to mention is so took 23 the electrical draeings for the plant and the list of i
24 eauf pment ehich was energis:ed that day and sent doen and did 1
25 a listing of just all the components, relays, seitchos, ACE FEDERAL REPORTERS, INC, l
r-p 1350 15 04 189 (m)DAVbur 1 instruments, et cotera, ehich soro energized during the 2 ovent, in terms of catocorios so 40 could cet a feel for hoe 3 many different pieces of equipment types soro onorgized, and 4 from all intents and purposes other than the five casos I 5 montioned to you had operated satisfactorily.
6 That is that set of numbers that you see right 7 thero.
8 The last thing that so did, so actually took this 9 action prior to the recolpt of a confirmatory action lettor 10 from Region 3.
11 (Slido.)
12 But the 473 items that I mentioned to you earlier 13 that 40 saa ehich 40 categorized as observations, that eas 14 dono immediately after the earthquake to conform eith our 15 confirmatory action letter.
16 The other thing that so did saa for any nos 17 identified sork items at Porry so had a computer driven 18 maintenance systom, shoroin any maintenanco activity ends up 19 on a eork request and ends up in the computer. And I mean 20 things from light bulbs, missing a faucot in a rostroom --
21 everything in Perry endo up on a PPMIS comouter system.
22 Wo committed to having all sork items reviewed by 23 our engineering staff at the plant for possible soismic r~s 24 relation to the ovents in other sords, to try to categorizo O
25 shethor the eork activity is the result of damago caunod by ACE FEDERAL REPORTERS, INC.
< - - - - - - - - - . EOC7DE) Rtrn D GL,u ; CDffrm __--_ _ _ __ _ _
1350 15 05 190 i Convbut 1 ehe earthauake.
2 So that das done and the total number of eork 3 requests which dere processed during the period of February i 4 3rd to February 28th, ehen the restriction sas lifted, eere l 5 2401, as shoen up there. i 6 Of those, 622 recuired evaluation. In other t
7 dords, the original engineering cut at those said these 8 could possibly be related to the earthauake. Of those 622, ,
9 459 eere determined not to be earthquake-related, and the 10 remaining 163 sere reviewed eith the NRC, Generally the l 11 resident inspector, prior to doing the repair.
12 Again, those are typified by the kind of things O 13 de see up there listed on the slide.
r l
14 The last thing I eould like to mention is se did 15 as a result of this earthquake note that se needed to make 16 some improvements in our procedures, both in the emergency 17 plan, the NUREG-0654 procedures, and in our operations ;
18 procedure dealing eith hoe de handle the earthquake event.
19 I have an off normal instruction that deals eith l 20 a seismic event. If I had a real emergency that sould l
21 affect core cooling or something, de have implemented the l
22 BWR Oener Group symptom-oriented emergency procedures, but I 23 still have some event-oriented procedures to deal eith l
l 24 specific events. I have an off normal instruction for a l 25 solomic event.
l Ace.FEDERAI. REPORTERS, INC. [
_ _ _ _--_--_ -- _ -_ _ FVS41n _-------- Nm a o _N _- - -- -
s l
1350 15 06 191
())DAVbur 1 The major problem de found eith these das that de 2 had not really clarified the entry conditions to assist the 3 operator in determining ehen to enter that procedure as 4 clearly as se could have. So de have gone back and improved 5 the entry conditions to both of those procedures to better i
6 line up eith the instrumentation that de have in the plant 1
7 to indicate the presence of a seismic event.
i 8 A long term action that se have in the procedures -
l l 9 that is not listed on the slide is se sant to incorporate in i
10 the earthquake procedure the aalkdoen methodology that se
{
i l 11 used at Perry that afternoon.
- 12 In other words, I am coing to have an attachment 13 to that procedure, the health physics maps of the building .;
j 14 and a list of instructions for hoe to immediately conduct
! 15 the ealkdoen so that you don't lose any information.
16 That eas ehat de dere most sorried about this
.> 17 last time, das losing some information or some 18 interference. So se sill have that built right into the 19 procedure so that that happens immediately after regaining l
20 stability of the plant.
21 DR. OKP.ENT: Have you ever looked in detail to i 22 see ehether the emergency procedures are still very much on 1
23 the mark if you had a severe earthquake, an earthquake more 24 severe than the design basis earthquake?
25 MR. STRATMAN: Yes, sir, I think I can anseer i
ACE. FEDERAL REPORTERS, INC.
E30T 98 -. N. ____N ____ -
1350 15 07 192 DAVbur 1 that.
2 I have been on the BWR Odners Group Emergency 3 Procedure Committee for approximately four years noe and as 4 chairman for teo years, so I am fairly familiar eith the 5 procedures and the guidelines that the BWR 04ners Group have 6 developed.
7 It is a purely symptomatic guideline all the 8 eay. There is no event methodology in there at all, so that 9 those procedures are broken doen into a containment set of 10 procedures and a reactor set of procedures and a secondary 11 set of procedures, and it soon sill be hitting the street.
12- I guess there sill be possibly gas portions of those 13 guidelines.
. 14 I have looked at the procedures since then. It '
15 is hard to make any parallel beteeen the Perry event because 16 de didn't see any findings, but in fact I firmly believe in 17 those procedures. They are a little long einded noe because 18 de made them symptomatic, in that they sill handle any 19 event.
20 I believe they 4111 still handle long term core 21 cooling and long term heat removal from a Mark III 22 containment. I absolutely believe that.
23 DR. OKRENT: All right. Let it go.
24 MR. STRATMAN: That is the end of my 25 presentation, sir.
s ACE. FEDERAL REPORTERS, INC.
_ _ ___ _____ _ _ ____ _ _ _ _ _ _ _ __N4E3 _ _ __ h- N - - _ _ _
?
1350 15 08 ,'193 (m/DAVbur 1 DR. POMEROY: Could I ask you one Question?
2 You said there eas an operator indication on a 3 back panel for the OBE.
4 EMR . STRATMAN: He gets a series of lights. They 5 are amber and red lights.- The seism 1c instrumentation sas 6 partially in'oervice, partially not in service. That sill 7 be discussed in detail.
8 He got a series of lights up there which sould 9 indicate to him by the procedure that he had in place that 10 day that he exceeded or noscibly had exceeded his OBE from 11 his understanding of it.
12 DR. POMEROY: But is there ancther indicator that C- 13 indicates excedance of an SSE? '
14 MR. STRATMAN: That can be done today. What you 15 end up doing eith the procedure modification noe is it 16 requires that he gets a series of lights, plus the STA has 17 to go doen and pull from the soismic instrumentation his
- j 18 chart report. Then he looks at the chart again. l 19 We hed to improve the procedure because de had to 20 -- and I am not a seismologist. That is Mr. Edelman's 21 line.
22 (Laughter.) l 23 MR. STRATMAN: But at any rate, you had to 24 convert it out into our building movenents ehere the seismic 25 instrument is located. The STA has been trained to do that ACE.FcotRAL REPORTERS, INC. l
--_ -_ -- - -_- r v a-a m - -------- m n m ~r , ---_ --- cmnc~n - ----_---- -_ !
1350 15 09 194 O He pulls together the tapes and makes a comparison.
( jDAVbur 1 noe.
2 We are putting together an operator aid, if you sill, to 3 help make that comparison.
4 That is ehat has come out of the event, that 5 there easn't a clear-cut day to make the call on the SSE, 6 and se are doing that noe.
t 7 Yes, sir. l 8 DR. SMITH: You mentioned that there sere 95 9 indeterminate items.
10 MR. STRATHAN: Yes, sir.
11 DR. SMITH: Were those issues such as light bulbs 12 and fuses?
13 MR. STRATMAN: Light bulbs, fuses, meter glasses, 14 some leaking flanges ehere you could not -- se did not have 15 a sork request in prior to the earthquake and you could not 16 determine ehether or not that flange das leaking as a result
. 17 of the earthquake or not. So it is indeterminate. We can't 18 make a call that say.
19 In some cases se had people tell us, oh, yes, I 20 knee that das leaking ahead of time, but unless de have a 21 sork request in there is no say to document that. So it 22 remained an indeterminate item.
23 DR. SMITH: If I follosed your presentation 24 correctly, you indicated there dere on the order of 7000 25 relays, of shich teo sere known to have cracked.
ACE. FEDERAL REPORTERS, INC.
E3347 3700- .Netionwide_W __ _ ____soo.3_3 u 646_______ ___ _ ___ _ __
1350 15 10 195 L
JDAVbur 1 MR. STRATMAN: Yes, sir. Those relays sere not 2 in an operating condition at the time of the event, though, 3 in that they did not have the potential applied to them as 4 they normally sould. l l
5 There sere none of those relays in any safety -
l 6 applications.
7 DR. SMITH: Right. But of the other 6900, some 8 fraction of those are monitored in the control room 9 indirectly or directly.
l 10 MR. STRATMAN: Yes, sir. They sould indicate the 11 trip of a piece of equipment or you aould pick up a flag.
( 12 DR. SMITH: A flag or an indicator light?
ir ,1
'- J 13 MR. STRATMAN: Right, other than the stuff that I 14 l mentioned to you, ahen se did the complete aalkdoan, se i
15 didn't see any abnormality.
16 DR. SMITH: I presume that not all 7000 of those 17 are monitored?
18 MR. STRATMAN: Oh, no, sir. I cannot pretend to
, 19 tell you that, no.
20 What se did again aas se took the prints and the 21 energi=ed equipment, the elementaries. They shosed shere se 22 had a relay in that circuit. Then se go to the master parts 23' list to determine ahether tnat is a repeat of one se have
(~') 24 already looked at on another system or if it is a nea one.
L/
25 DR. SMITH: So in your judgment, is it ACE-FEDERAL REPORTERS, INC.
702 347-3700 Nathnwide Coverage 800 336-6646 ____
\
1350 15 11 196
_'m
(,)DAVbur 1 conceivable that some of those relays momentarily changed 2 state or chattered an:1 sere not observed by the operators 3 themselves?
4 MR. ST'RATMAN: For the operating equipment I 5 can't absolutely rule that out, but I find it unlikely 6 because se didn't pick up any enunciators or anything else 7 that said de had a problem. So I really don' t believe se 8 did.
9 For stuff that sas not energized, your guess is 10 as good as mine, I guess. But for the energized equipment, 11 shich is monitored either in the control room or underneath 12 computers, I don't believe se did. I don't think se had any d l'3 , change in state, no, sir, because se sould pick it up.
14 DR. SMITH: I am not trying to press you for any 15 specifics, but in your best judgment sould you say that the 16 majority of the systems that sere energized you feel your 17 staff and people in the control room sould have had some 18 indication for the majority of thos'e systems?
I 19 MR. STRATMAN: Yes, sir, the majorit/ se sould 20 have one place or another, either a local panel or in the 21 control room or one of those two computers.
22 DR. SMITH: Has your staf f had time to look back 23 over the process computer records and printouts to determine r')
kJ 24 there is no indication there of any possible change in state 25 or of' chatter? l ACE-FEDERAL REPORTERS, INC.
202-349-3700 Nationwide Coveraar 800-)M
1350 15 12 197 1- DAVbur 1 MR. STRATMAN: I don't think I can really ansaer 2 that.
3 Dave, do you knoa if de sent back over the 4 process computers to look for any relay changing states or 5 anything like that?
6 VOICE: Over the process computer?
7 MR. STRATMAN: The process computer aill put out 8 alarms, sir. Other than that -- and se didn't see any of 9 those that indicated that ao had a problem ehen the systems 10 sere in operation. So I don't monitor relays on the process 11 computer.
12 So I guess I sould have to ansaer the say I did 7
( "/
~
13 .before. For the systems operating, no, sir.
14 MR. MICHELSON: If I, understood it correctly, you 15 had some 7000 relays -- is that the number?
16 MR. STRATMAN: Gene, do you sant to ansaer that?
17 The question eas: do se have 7000 relays, as 18 shoan on this?
19 VOICE: Yes.
20 MR. MICHELSON: Is Perry not using -- mostly 21 us'.ng solid state?
22 MR. STRATMAN: No, sir, se are not.
23 MR. MICHELSON: It is a Mark III?
24 MR. STRATMAN: But de are a relay plant, sir.
lu 25 MR. MICHELSON: Was the process computer ACE-FEDERAL REPORTERS, INC.
'ViS 1A* 1 *M \?ar. amu..f.t".u & ,a. QM 11Z IIJZ
l 1350 15 13 198 ;
1 DAVbur 1 actually online at the time of the event?
2 MR. STRATMAN: Yes, sir, both that and the Aris 3 computer.
4 We actually monitor more systems at Perry on the 5 Aris computer.
6 MR. MICHELSON: A third question: are there any 7 industrial type facilities near Perry, and ahat experience 8 did they have as a consecuence of the event?
9 MR. STRATMAN: We have a plant 15 to 20 miles 10 sest of us, and one of their boilers took a trip. We talked 11 to the people, eith the Morton Salt people, sho have salt 12 mines under the lake in Cleveland. They didn't have a p_
V problem there.
13 14 MR. MICHELSON: What do you mean, they didn't 15 have a problem?
16 MR. STRATMAN: They sent and looked at all their 17 mines after the earthquake and they didn't have a problem.
18 MR. MICHELSON: Physically? ;
19 MR. STRATMAN: Yes.
20 MR. MICHELSON: Are there any other processes 21 that sent on fairly near the site? Any seitchyard 22 experience?
23l MR. STRATMAN: We didn't see any problems in the 1
(} 24 switchyard, to my knosledge.
25 DR. STEVENSON: My name is John Stevenson. I am ACE-FEDERAL REPORTERS, INC.
msn.unn w-+c_ mtu -
. e e _. _ __. _, _ . _ _ _ _ - . - . . . . . . . _ _ . .
l l
1350 15 14 199 DAVbur 1 one of the three members of the Earthcuake Engineering 2 Research Institute that live in Ohio, ahich is not 3 sell-knoan for its earthquakes.
4 We are preparing a reconnaissance report for 5 EERI. We did do a reconnaissance of the area. There are 6 some small industrial facilities, some as close -- se did 7 look at one as close as eithin 3 miles of the epicenter. We 8 found no -- aell, there sere some indications.
9 For example, one plant reported a number of 10 relays tripped. We looked into it in more detail and found 11 that it aas because of electric changes, 12 One of the units that eas just mentioned, the f_.)
V
- 13l 530-megasatt turbine, tripped off line. It aas a
- 14 fluctuation in poser shich caused some relays in the area to 15 trip, but nothing se could determine sas caused, as far as 16 industrial facilities sere concerned sas there any anomaly.
17 There aere problems eith regard to structures.
18l We sas ceiling tiles had fallen, aindoas sere broken in !
19 regions closer to the epicenter, but se could detect no .
20 industrial facility changes induced by the earthquake.
21 There aere cases of some equipment shich sas not 22 bolted doen but simply floating ahich had moved as much as a 23 tenth of an inch, but this sas like eithin four or five i
(} 24 miles of the epicenter.
doen, there sere no malfunctions.
Any equipment that eas bolted 25 ACE-FEDERAL REPORTERS, INC.
I i
1350 15 15 200 k_)DAVbur 1 MR. MICHELSON: Did the voltage instabilities 2 reflect back into the Perry plant as sell?
3 DR. STEVENSON: That I have no idea shether they 4 sere or not. All I knos is there sas one facility that 5 reported a number of relays had tripped, but the conclusion 6 sas that it sas an electrical drop in voltage caused by the 7 drop of the 530 megasatt.
8 MR. MICHELSON: But ahen that sent out, I aould 9 have thought Perry might have seen a momentary drop, also.
10 DR. STEVENSON: That I don't knos. That sas the
. 11 only occasion ehere there eas any reporting during the 12 trip.
O , 13 MR. MICHELSON: Thank you.
14 MR. BOWEN: Jim Boden, eith CEI.
15 We had a vibration of the turbine trip on our 16 Eastlake 5 650-megasatt unit. We also had a backdoan in 17 poser of a 250-megasatt unit because of high vibration. I 18 telieve it aan on the teo-strap fan on our transmission yard 19 immediately adjacent to the plant.
20 Other then the breakers that tripped due to the 21 relay in the control room, nothing happened there. We have 22 a substation, 132 KV, in Leroy, shich I believe is credited 23 as the epicenter, and saa no effects shatsoaver.
() 24 We did, I believe, see a slight degraded 25 frequency on our system due to the Eastlake trip, but ACE. FEDERAL REPORTERS, INC.
m 1n.1,nn unn.s en-- m11s.uu
. -. -. . - . . - - - . .. ... . . . .. ~ . .. .. .-. .. . . - - .
1350 15 16 201
, DAVbur 1 nothing that eas of any major significance.
2 3
4 5
6 7
8 ;
9 10 11 12 O ,
13 -
14 .
15 16 ,
1 17 18 19 20 21 22 23 24 25 ACE-FEDERAL REPORTERS, INC.
wwnaa vu-- .c-__ _
1350 16 01 202 t- DAV/bc 1 DR. SMITH: Could you tell us approximately what 2 levels of response and motion were observed at the unit one 3 air compressor and the auxilliary boilers?
4 Or perhaps what elevation in the plant?
5 MR. STRATMAN: The air compressor is down on the 6 579 level of the plant. The trip-setting is 1.8 mils.
7 DR. SMITH: That's very low. _
8 MR. STRATMAN: Right. Grade level is 620 at 9 Perry. It's a very high speed machine so we've got a fairly 10 sensitive trip on that machine.
11 In terms of movement on the aux boiler, I don't
,_ 12 think I can really cuantify anything we saw there.
'^'
13 DR. SMITH: What elevation is that?
14 MR. STRATMAN: That's on 620, sir. There's two 15 package boiler units.
16 MR. EDELMAN: We'd like to continue with our 17 presentation with Dr. Chen. I made sure that everybody 18 should have a copy of the overheads. I may be getting a 19 little bit older but I couldn't read them from afar. So I 20 hope that you can refer to your package.
21 l I found sone difficulty in following the slides.
22 j DF. CHEN: My name is John Chen, Manager of the 23 Civil and Structural Department and Chief Structural 24i Engineer of Gilbert Commonwealth. I'm going to discuss the 25 reported earthquake versus the design of the Perry Power 1
l l
l ACE-FEDERAL REPORTERS, INC. i
1 1350 16 02 203
()DAV/bc 1 Plant.
2 First of all, I would like to show you the first 3 recorded data from the following instruments.
4 (Slide.)
5 We have two types of instruments in the plant.
6 They are the Kinemetrics Recorder and the Annual Response 7 Spectrum Recorder, shown on the first two pages. Those are 8 the list of those recording instruments.
9l The next two viewgraphs show you the locations of l
10 ' those instruments on the plant.
11 (Slide.)
12 This one is the ground view of those instruments' 13 locations and the next one shows you...
14 (Slide.)
15 ...thes elevation view of those instruments.
16 Now, the next one I'll show you the data 17 reduction.
18 (Slide.)
19 I think this viewgraph will answer one of the 20 questions which Dr. Pomeroy asked for the panel instruments 21 connected to the control room panel with amber lights and 22 red lights. The amber lights indicate two-thirds of OBE, 23 and the red lights indicate 100 percent of OBE.
24 So, during this earthquake, both amber lights and 25 red lights were lit. That was your question.
ACE FEDERAL REPORTERS, INC.
L- _ - --- ---_- --- _- --- ----- --------------- --- - ---- fatsRD --------s tes t e smx a ----- - exec"3 --- ---------- - ---- --- m_
l 1350 16 03 204
()DAV/bc 1 (Slide.)
2 The next one shows the data reduction of the 3 Kinemetrics data, the acceleration time, history reporter, 4 it was converted into strip charts and was digitized and 5 integrated twice to get velocity, then displacement.
6 Then the response spectro were calculated on the 7 recorded acceleration time history.
8 Now, before I show you the comparison of the 9 recorded time history versus the design time history, I 10 would like to summarize the result of the comparison.
11 (Slide.)
12 The 1986 Ohio earthauake has high frequencies, f}
13 short duration, low energy, low velocity and very small 14 displacement. And our design time history has broad band 15 frequencies, low in duration, high in velocity, large 16 displacement and high energy.
17 Now I'll show you how this conclusion was reached 18 by looking at the comparison of the two time histories.
19 ,(Slide.)
20 These two charts are on the same scale. The one 21 on the top is our Perry design time history, and at the 22 foundation of the reactor building. The one on the bottom 23 is the recorded time history at the same location.
24 As you can see, the design time has a much longer (G~]
25 duration, broader frequency content and multiple peaks. The ACE-FEDERAL REPORTERS, INC.
K4237-ETO FW36 6646
1350 16 04 205
() DAV/bc 1 recorded time history has a strong motion shift, part of the 2 record but less than one second. It has a very high spike.
3 Even in this case, the amplitude is about comparable.
4 This is the north / south component. A similar 5 comparison for east / west.
6 (Slide.)
7 East / west, the recorded amplitude is much less 8 than tric design time history. Also of short duration and 9 high frequency.
10 (Slide.)
11 The vertical component, same thing. Very short 12 duration. High frequency and small amplitude. The next one 13 shows you the one at- the containment elevation, 686 14 north / south component.
15 (Slide.)
16 This one, again, the reported time history has 17 very high frequency components, around 20 hertz. The 18 amplitude is somewhat higher in the design time history.
19 But it's of very short duration and high f recuency. We'll 20 come back to that point later on.
21 DR. POMEROY: Is that Kinemetrics?
22 DR. CHEN: Kinemetrics, yes, sir.
23 (Slide.)
24 This again is at containment elevation 686, 25 east / west component. Again, short duration, amplitude is )
l 1
l ACE-FEDERAL REPORTERS, INC.
f3347 3700 Nationwide h 8'10 3RY""3 "
1350 16 05 206 i DAV/bc 1 also less than the design time history.
2 The next one is a multiple component.
l 3 (Slide.)
4 At the same place. Again, it's of very short j l
5 duration and the amplitude is about comparable. High 6 frequency, 20 hertz.
7 Just because of those comparisons we reached a 8 conclusion earlier that --
9 (Slide.)
10 -- this is a very high frequency, small energy 11 earthouake. Now, the Perry design basis is based on a broad 12 band frequency design response spectra which was smoothed 7-E/ 13 with 84th percentile confidence level and a composite time i
14 history. .
15 It has also a long duration and high energy, as 16 shown earlier.
17 (Slide.)
18 The design spectra, everybody is familiar with.
19 (Slide.)
20li Now, the next one, we are making some ZPA 21 comparisons of the recorded carthquako versus design. ZPA 22 stands for Zero Period Acceleration. At the recorder, the 23 ZPA value varies f rom below OBE values to about 74 percent
( 24 of the SSE value, except at the containment vessel elevation Ls 25lj 686.
ACE-FEDERAL REPORTERS, INC.
FAMT/ ff7PD NN Coverage FTTf36-6646
1 1
1350 16 06 207
(-) DAV/bc t, 1 At that place, the relative displacement and 2 stress is very low because stress is proportional to 3 distortion. And the reason for that, again, is because this 4 is.a high frecuency phenomena.
5 DR. POMEROY: What was the value there?
6 DR. CHEN: I will show you the value.
7 (Slide.)
8 This is the value. The first column is the 9 comparison at the aux building foundation mat. The second 10 column is the reactor building foundation mat. The third 11 column is the top of the recirculation pump. The fourth 12 column is the reactor building platform. And the fifth 13 column is the containment vessel at 686.
14 We grouped these horizontally. The first row is 15 the recorded SSE, OBE, the recorded north / south component.
16 SSE, OBE. The second row is the east / west component. The 17 third row is the vertical component.
18 If you look at these components, the recorded 19 value in the first four colunns, they are all less than 20 either OBE value or SSE value, except the last column.
21 The last column only, the north, south and 22 vertical components were exceeded. Now we concentrate on 1
23 the fifth column by looking into the vertical displacement.
O 24 (Slide.)
V 25 Again, the first column is the displacement at l
l l
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwideh 900-33M5646 __-
1350 16 07 208
(_)DAV/bc 1 the top of the foundation. The second column is the 2 displacament at containment elevation 686. The third column 3 is the difference between these two columns. The relative 4 displacement, which is an indication of the stress level in 5 the containment.
6 If you look at this third column here, the north, 7 south and east / west component, the relative displacements 8 are less than OBE values.
9 In the other direction it's slightly higher. But 10 the net effect, in terms of stress, is ctill lower than OBE 11 design values. That's what I'm going to show you next.
12 DR. POMEROY: But the displacements in the A
13 vertical direction are greater than the SSE?
14 DR. CHEN: Relative displacements.
15 DR. POMEROY: The relative displacement is 16 greater?
17 DR. CHEN: Yes. The reaso1 we showed the 18 displacement comparision is because relative displacement is 19 an indication of stress. So we went one step further to 20 calculate the dynamic stress in comparison with design 21 stress.
22 (Slide.)
23 This table shows comparison at three elevations 24 -- 686,644 and 592. Dynamic forces were obtained from the 25 analysis of the containment sheer model with the recorded ACE-FEDERAL REPORTERS, INC.
202 347-0700 Nationwide Coverage 800 336-6646 l
1350 16 08 209 l
,jDAV/bc 1 time history at the top of the foundation as input. Wo have 2 axial force; MX, MZ are the components of the horizontal 3 access. The dynamic stresses were calculated from either P 4 over A plus MX over S, ac is the section modulus.
5 Here we picked the higher one of the two 6 horizontal moments. Also, one more conservatism. Those 7 P&M, XMZ's occur at the same time. We assume they occur at 8 the same time, in combination to increase the recorded 9 stress.
10 At 688 there is some slight accedence. If you go 11 through the next elevations, the design stress far exceeds 12 the dynamic stress. But t.1e important thing here to notice
-)
13 is the design is controlled at elevation of 592 because we 14 have a uniform thickne'ss shell.
15 So the thickness was controlled by the section 16 with the highest stress. So this slight accedence here 17 doesn't have any meaning. If you look at a section which 18 l controls the design, the design stress is 2.6 times higher i
19 than the dynamic stress. ,
20 li Another thing I would like to point out, for the il material we use here, the stress is 38 psi. So this stress 22 is very small in comparison with what we saw. I mean, in 23 comparison with real stress.
24 DR. REMICK: Something I'm missing, I need some 25 help, I think. Why is it that the 644 level, in fact, the ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 8001T *
- 3
1350 16 09 210 JDAV/bc 1 dynamic stress is above the design stress --
2 DR. CHEN: Do you mean 688 or 644?
3 DR. REMICK: 644.
4 DR. CHEN: 644, the dynamic stress is less than 5 the design stress.
6 DR. REMICK: You're right. It's the other way 7 around.
8 DR. CHEN: 688 is at a higher elevation. The way 9 we designed the containment, we have a uniform thickness of 10 the containment. So the thickness is controlled by the 11 section where you have a maximum stress. In this case, it's 12 Section 592. That's the section you have maximum stress.
13 Once we det' ermine the thickness from this ,
14 section,.the same thickness is applied throughout the 15 shell. So the design is controlled by stress at this 16l elevation. This slight accedence doesn't have much meaning, 17 especially if you look at the stress, one-half psi. It's 1
18 very small in comparison with the yield stress of the 1 19 material. l 1
20 (Slide.) l 21 Next, we will see some response spectra 22 comparison. And the conclusion of the comparison is that l
23 Perry design resconse spectra are far above the recorded 24 spectra in the frequency region below 14 hertz, the vast
{
25' majority. And there are certain recorded response spectra ACE-FEDERAL REPORTERS, INC.
202 341 1700 Nationwide coverage 80,k)M M46
1350 16 10 211 o
(_)DAV/bc 1 which exceed the design spectra around the 20 hertz' region.
2 But, even at those regions which correspond to 3 very small displacement, for example, the foundation bed, 4 the largest displacement was less than 700th of an inch, 5 which doesn't have much engineering significance.
6 We also compared the velocity spectra, which is a 7 measure of the energy content, and the area under the design 8 spectra is also much larger than the area under the recorded 9 spectra.
10 DR. SMITH: I don' t disagree with your conclusion 11 but I wanted to point out that figure D10 of the applicant's 12 submittal, which is the 686 foot elevation, shows the design 13 spectra exceeded from 11 hertz approximately on, I would
, 14 suggest that maybe you should be clear about the regions
, 15 where the design spectra exceeded. I don' t see that in the 16 conclusion. ,
17 DR. CHENs Okay.
18 (slide.)
)
19 As we say here, 14 hertz. Then we said.certain 20 spectra exceed. Okay. In the 608 north / south component, 21 there's a slightly lower frequency. Agreed. j l
22 (Slide.) l 23 This is the spectra comparison in the north / south i
24 reaction at the foundation mat of the reactor building. As
}
25 you can see,.in this region, the design spectra far exceeds ACE-FEDERAL REPORTERS, INC.
EBCF-DiB ha - - ------ CEELTD
1350 16 11 212
, DAV/bc 1 recorded spectrum except in the 20 hertz region. But, if 2 you look at that point, the spectral displacement at 20 3 hertz is less than 2.5. It's .025. Even at that point, the 4 spectral displacement is very small.
5 (Slide.)
6 And a similar comparison at the east / west 7 component. And, again, in the frequency region here, the 8 large margin exists except in the 20 hertz region, again, 9 over here, the spectral displacement at 20 hertz is less 10 than 200th of an inch.
11 A similar comparison --
12 (Slide.)
g i
/
13 -- at a vertical component, again, a large margin 14 , here and this at 20 hertz, the spectral displacement is 15 again less than 200th of an inch. I think this is the one 16 Dr. Smith was just referring to.
17, (Slide.)
18i Below the 11 hertz region. In the 11 hertz l
19 region here, the recorder 'ias f ar below the design 20 spectrum. This north / south containment elevation 686, if 21 you look at the east / west component at 686, again, the 22 spectral displacement is about less than 400th of an inch at 23 the high elevation 686.
24 v
25 ACE-FEDERAL REPORTERS, INC.
200-3c.3700 Nationwide Coverane 800-J 36-M*6
1350 17 01 213
(_,)DAVbur 1 A similar conclusion is reached for the vertical 2 component at the 686 elevation. So based on this, we 3 concluded that the design spectra far exceed the recorded 4 spectra except in the high frequency region around 20 hertz, 5 which presents a very small displacement on the order of two 6 or three hundredths of an inch.
7 (Slide.)
8 Here, we would like to summarize the lack of 9 engineering significance of the 1986 Ohio earthquake.
10 First of all, it is a proof test, as described 11 earlier by Mr. Stratman. The plant structure and safety 12 class equipment were not affected.
O b 13 MR. MICHELSON: Why do you call that.a proof 14 test? It is only a proof that if you get a unique 15 earthquake with a small high frequency distribution that 16 there isn' t an ef fect.
17 DR. CHEN: Yes, sir.
18 MR. MICHELSON: That is not very comfortable for 19 the rect of the seismic spectrum. Just this one unique 20 spot?
21 DR. CHEN: Yes, sir.
22 MR. MICHELSON: Now, that isn't a spot you would 23 normally expect this kind of equipment to not perform
(} 24 anyway?
25 DR. CHEN: Yes, sir. When we compare with the ACE-FEDERAL REPORTERS, INC.
202.}47-3700 9"etionwide coverase 800-3W
1350 17 02 214
/
_fDAVbur 1 other industrial criteria, usually during the pile driving 2 operation and blasting operation, the criteria used to 3 monitor the adjacent nonengineered structure is one inch per 4 second carticle velocity, and our recorded maximum velocity 5 onsite is here, .87 inch per second.
6 In c'.her words, if you look from the velocity 7 criteria point of view, what Perry Power Plant experienced 8 during an earthauake is less than the ordinary nonengineered 9 building would experience when there is a pire driving 10 operation nearby.
11 Another thing is that IEEE-344,'Section 7.5 would 12 not allow the vendor to use shop type testing to qualify for
( )
13 seismic demand because of the high frecuency, low energy .
14 content.
15 (Slide.)
16 The next slide, everyone has seen this a thousand 17 times, built-in general conservatisms for every power 18 plant. No need to go into detailed discussion.
19 In spfte of all those discussions here that the 20 earthquake doesn't have much engineering signficance, we 21 still went one step further to quantify the margins of 22 equipment design.
23 DR. REMICK: Before you go into the margin, what
~)
u; 24 , type of equipment would you think would be most susceptible l
25 to damage with a 20-hertz frequency? What type of things?
ACE-FEDERAL REPORTERS, INC.
T -141-1700 Nanonwde Coserane 800-136-6646
1350 17 03 215 s DAVbur 1 DR. CHEN: I think my presentation here is going 2 to cover your question.
3 DR. SMITH: Dr. Chen, in the applicant's 4 submittal there was a statement made that possibly the 5 recordings at 630-foot elevation and the 686 were influenced 6 by local effects, something in the plant or what have you.
7 I looked at the spectra, the time histories, and 8 so on and didn' t see any obvious signs on that.
9 Do you have any light to shed on that?
10 DR. CHEN: Yes. Wo have reached some conclusions 11 from an instrument at elevation 630. What happened was 12 after the earthquake we put in new plates afterwards. I 13 don't recall after how many days we put out the' plates -
14 again. There was no earthquake event, and,we found out the t
15 record at 25.4 hertz was higher than what was shown on the 16 recorded earthquake.
17 So that is an indication that at the 630 18 elevation that was really caused by people walking by intent 19 on the instrument itself. '
20 And we have suggestions to NRC to move the 21 instrument to an area where the instrument would be less
- 22 sensitive to the nearby impact.
23 As for the 688, 686 on the containment vessel, we 24 still believe that is some kind of secondary impact but we 25 have not reached a full conclusion. But in spite of that ACE FEDERAL REPORTERS, INC. 1
_ _ , 202-347 3700___ Nationwide Coverage ___ , _ _ 800 3364646 _ _ _ ,_ . _ _ _ ___ _ . _
1350 17 04 216
- DAVbur 1 fact we did not reach a full conclusion, we are taking into 2 account the recorded event at 686 to look for the margin.
3 So the outcome of that evaluation is fairly developed as far 4 as the conclusion here is concerned.
5 DR. SMITH: You end all instrument reporting at 6 the 630-foot elevation. The spectra, vertical and 7 north / south, are different and, interestingly enough, bear a 8 resemblance to the same spectra at the base of that 9 elevation.
10 Is there any possibility that those plates could 11 have been inadvertently exchanged, either in the actual 12 analysis of them or in the subsequent handling of the data?
13 DR. CHEN: Maybe Bob -- Mr. Paul Endo, the 14 i President of Endo Enterprises?
15 MR. ENDO: The plates arc serialized, and when 16 the plates are put into the recorder, the recorder's serial 17 number is recorded and the plate serial numbers are 18 recorded. There is a group of 12 that have a distinct 19 serial number. They are also marked 1 through 12, and they 20 are keyed so they can only go into position 1 through 12 in 21 the recorder.
22 So I don' t believe there could be a mixup among 23l recorders of plates or positions within the recorder.
~
/3 24 MR. MICHELSON: Is there any chance that there is v
25 a movement by the earthquake that would have reflected into ACE-FEDERAL REPORTERS, INC.
202 347-3'00 Nationwide Coverste 800 336 6616
1350 17 05 217
( )DAVbur 1 this?
2 DR. CHEN: That is one of the possibilities that 3 we thought could possibly have happened because of the 4 frictional coefficients between steel varying between .115 5 to .6. So when the G value reaches that value, there is 6 always such a possibility of secondary impact, yes, sir.
7 DR. TRIFUNAC: I cannot resist but to continue.
8 If I looked at a north / south analysis and I smoothed things 9 out by eye, I see that the earthauake is comparable to the 10 long train of vibrations that is used in the design. Then I 11 have looked at -- I think it is 680 plus, the recording --
12 and I see what I would interpret as perhaps 4 or 5, maybe 13 more, in the recording, and I see a responsive indication 14 of -- I don' t know. You may correct me there -- but perhaps 15 a factor of 2 in the calculation.
16 And to me, that is a lot of interference in the 17 signals as well. I would like you to comment on that.
18 DR. CHEN: This part of the discussion we were 19 just discussing about earlier at the 666 possible secondary 20 impact.
21 If you look --
22 DR. TRIFUNAC: What is the secondary impact?
23 DR. CHEN: It is caused by vibration, that
("T 24 something else was moved and comes back to a secondary short C'
25 duration impact on containment. one possibility just ACE-FEDERAL REPORTERS, INC.
- - - - - _ . - - _ - - - - - _ - - - - - - - - - - - - _ ------------- _ ---- f3PJ-613 -- -_ - -- ----- RECTt:DGb:~c^J - -- -- -- ----__ _ _ CDEX"3- --___ ___ _ - ___
1350 17 06 218 DAVbur 1 mentioned was the possibility of the frictional coefficient 2 being in the range of .15 to .6.
3 Now, to answer your question, if you look at the 4 north / south, yes, there is a very high amplification from 5 the ground level to 686, but if you look at the east / west 6 and vertical direction, which is not that high, so 7 north / south direction, based on what we have seen so far, is 8 still some anomaly despite of the fact that we took that 9 record as is into account in looking for the margins.
10 So whether there is really a secondary impact or 11' not would not influence our conclusion.
12 DR. TRIFUNAC: I am not looking at the
/~3 V , 13 conclusions from the point of view of the stresses of this i
14 particular earthquake.
15 You said engineering sidnificance. That bears on 16 all of your calculations as it was done originally.
17 How confident you are tnen that this data here 18 substantiates your procedures and calculations previously, I 19 think that is of significant engineering significance. That 20 is what I am asking you. l i
21 DR. CHEN: So you are talking about a j 22 verification of the recorded data versus -- 1 23 DR. TRIFUNAC: Wouldn' t you expect to even look
(} 24 at that among the topics that you are doing?
25 DR. CHEN: I will show you later on.
1 1
1 ACE FEDERAL REPORTERS, INC.
_______ ______.___ _ _ ___ ________ _ ________s n c m w m __ _ _ _ _ m ____________ e ___._ _ __ _ _ _ _ _
1350 17 07 219
( DAVbur 1 DR. TRIFUNAC: Why don' t you address it now, just 2 in one sentence?
3 (slide.)
4 DR. CHEN: I think this is the one you were 5 referring to.
6 MR. EDELMAN: These slides are not in the 7 handout.
8 DR. CHEN: That is right.
9 DR. OKRENT: They will be in our files.
10 MR. EDELMAN: They are not in front of you right 11 now, Dr. Okrent.
12 DR. CHEN: I did not intend to present this, but 13 since Dr. Trifunac asked this question, I will just use this 14 .to answer Dr. Trifunac's question.
15 This is the one I suppose you just mentioned in 16 the north / south direction. This is focus on spectra 17 recorded. The one at the top is the 686. The one at the 18 bottom is the top of the mat. Okay?
19 Now --
20 (Slide.)
21 --
the next one is east / west component. This is 22 the one at the bottom at the foundation mat. The one at the 23 top is at 686. This ratio is much less than what you saw in
'(^^ 24 the north / south direction. That is what I meant.
-) i 25 I DR. TRIFUNAC: But I don't understand. What does ACE-FEDERAL REPORTERS, INC.
202J07 fE0 WP^TP'h@ overage FE336-6646 l
i 1350 17 08 220
/
s - DAVbur 1 that prove?
2 I agree there is less, but what does that prove?
3 DR. CHEN: I didn' t prove anything yet.
4 (Slide.)
5 The next one is the comparison of the vertical 6 spectrum recorded at the base and recorded at the top. You 7 see the amplification at 20 hertz. That is what you were 8 referring to, the engineering significance, okay?
F 9 Now, we used a fixed base model of the steel 10 containment. We used a time history recorded at the top of 11 the mat as input, three components simultaneously, and we
,..s 12 again calculated the spectra on the base and at the top.
l 1 -
13 (Slide.)
14 This is the north / south components. It is right 15 in the shape as to the way it was recorded. The one at the 16 too is recorded. The one -- I am sorry, the one at the 17 bottom is the recorded and at the basement. The one at the 18 top is the calculated at the 686 elevation.
19 ! So this is a very good agreement in comparison 20 with recorded data in the north / south direction.
21 Now, let's look into the other directions.
22 DR. TRIFUNAC: Wait a minute. The calculated is 23 two times less than the recorded.
24 DR. CHEN: Yes, that is only in the north / south.
}
25 DR. TRIFUNAC: How is that in good agreement?
l l
ACE-FEDERAL REPORTERS, INC.
an?.t.17.1 an %rinnmuir renernse ROQ.11/L/ Win
1350 17 09 221 DAVbur 1 DR. CHEN: The agreement is in terms of the 2 shape.
3 DR. TRIFUNAC: I am not asking the shape. I am 4 asking the amplitude.
5 DR. CHEN: I will show you the agreement in the 6 next one.
7 (Slide.)
8 This is in the vertical direction. .4, .5, .56.
i 9 okay, this is much better agreement.
10 Actually, this ratio here is even higher than 11 what was recorded. That is what I call good agreement in 12 the vertical.
,m 13 And also let's look into east / west.
14 (Slide.)
15 Again, the recorded is .4, the calculated top is 16 1.1. Again, the ratio is even higher than recorded. That 17 is what I call good agreement.
18 ' DR. TRIFUNAC: I don' t understand what you mean 19 by agreement at all. You have a record at certain points.
20 ' You have calculations at certain points.
21 What is a good agreement? 50 percent, 100 1
22 percent, 1000 percent? What is a good agreement of what? l l
23 DR. CHEN: Can you tell me what is your
('~ 24 ' definition?
25 DR. TRIFUNAC: No, I am trying to describe i
ACE-FEDERAL REPORTERS, INC.
N2 347-3700 Nationwide Coverage 800-336-(416
l 1350 17 10 222 ts DAVbur 1 because you made a statement that the data is here 2 suggesting that there is no engineering significance of this 3 earthquake as far as your analysis goes, and I see that your 4 calculations are not capable of predicting the motions 5 within a factor of 2.
6 What is the engineering significance of that?
7 DR. CHEN: It seems to me we have answered the 8 wrong question. When I said that the nonengineering 9 significance is in terms of the high f requencies, low 10 displacement, you are interpolating the engineering 11 significance in terms of a comparison with recorded data
, 12 versus analysis data.
( )
, 13 That was not the intention of the original words 14 of nonengineering significance. That is in terms of high 15 frequencies, short duration, small displacement.
16 I believe some of your own papers were also 17 written that way.
18 DR. OKRENT: I am sorry, you are confusing me. I 19 want to get at this.
20 If I understand correctly, your prior analysis 21 underpredicts the measured amplification by a f actor of 2; 22 is that correct or not?
23 DR. CHEN: The original analysis, no.
24 DR. OKRENT: Or even this analysis underpredicts?
25 DR. CHEN: In the north / south.
ACE-FEDERAL REPORTERS, INC.
w . u .t m wm mi, em m., unt -
~-
1350 17 11 l 223 DAVbur 1 DR. OKRENT: Let's stay with the north / south. I 2 understand the rest.
3 DR. CHEN: Okay.
4 DR. OKRENT: Do you consider this difference 5 between what I will call calculation and observation to be 6 significant? Are you generally satisfied if you 7 underpredict or overpredict, as the case may be, within a 8 factor of 2?
9 DR. CHEN: When you say a factor of 2, you are l
10 probably referring -- I don' t know how you get a f actor of 2 11 here. Anyway, that is irre]evant.
12 The prediction here says in the north / south 13 direction the'reco,rded. data is somewhat higher than 14 calculated. That is why in an earlier discussion we 15 mentioned the possible secondary effect.
16 DR. OKRENT: It is a possible one, but it may be 17 a convenient supposition. At the moment what we have is a 18 l f actor of 2 dif ference. That is what we have, and I am 19 trying to understand whether this is just within what you l 20 i consider to be the working accuracy, or is there something 21 here that needs to be understood more deeply or.just what? !
22 DR. CHEN: I think before I answer your question, 23 I think we have to put ourselves in the proper perspective.
(~]
v 24 l i
What we are comparing here, okay, is the spectrum shape. We 25 know that is very sensitive to the time history itself, and ACE-FEDERAL REPORTERS, INC.
70]147-1700 Ngjopwide Coveqp 80fM16-6646 __
1350 17 12 224
/DAVbur 1 it is not unusual when you calculate the response spectrum.
2 We will see variations, and I recall at the 3 beginning when the response spectrum -- if you could for the 4 same time history -- if you had a different spectrum 5 calculated, you would anticipate some variation.
6 DR. SMITH: These calculations, let me see if I 7 can straighten this out.
8 The predicted values are what you have got by 9 taking recorded time histories, putting them back in your 10 bottle, and then calculating response spectra?
11 DR. CHEN: Yes, sir.
12 DR. SMITH: So these are not really predictions; o
13 ) they are calculations based on existing models.
14 DR. OKRENT: In fact,.it is a better measure of 15 the calculational techniques of the data than to have taken 16 some arbitrary time history that was approximate?
17 DR. SMITH: Right.
18 DR. TRIFUNAC: That is not the point. The point 19 is this, that you have a mathematical model that you stick 20 into a computer and you feed that model something and 21 something comes out of it, and you say this is the 22 response. This is what is done in the original case in OBEs 23 and SSEs, as well as by feeding the base margins here and
(~) 24 j seeing what comes out.
\_' ;
25 DR. OKRENT: I agree with you.
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6M4
I 1350 17 13 225 O DR. TRIFUNAC:
V DAVbur 1 Let me finish. And then comes the 2 earthquake, and the earthquake doesn't know about that 3 mathematical model. It simply records where the data is, 4 and that tells you about the engineering signficance of your 5 calculations.
6 That is what I am talking about.
7 8
9 10 11, 12 13 i 14 15 16 17 18 19 20 i
21 22 23 24 25 i l
l l
ACE-FEDERAL REPORTERS, INC.
__ __---- --_----_-_ _--_ __EBCF-DEL ---. - - - --_ _ Nh--_ _ _ _____ CIMOX"C1___ _ --- _----__ _ _
1350 18 01 226 A
tm,)DAVbw 1 DR. OKRENT: I'm trying to pursue the same 2 point. In other words, it's not, in the sense of the 3 moment, a question of, did the earthquake which occurred do 4 damage.
5 DR. TRIFUNAC: That's not the point at all.
6 DR. OKRENT: It's a question of how well you have 7 analyzed it --
8 DR. CHEN: Yes, sir.
9 DR. OKRENT: --
in this regard, and should we be 10 concerned in some broader way; all right?
11 DR. CHEN: Yes, sir.
12 DR. OKRENT: Just to make we're all talking about 13 the same aspect.,
14 DR. CHEN: Yes, sir.
15 MR. EDELMAN: I think we can get a little 16 amplification from Dr. Stevenson.
17 DR. STEVENSON: Yes. Dr. Okrent, I would like to 18 explain that recently I did a review for the ASME of the 19 ratio amplification from different power structures from 20 basement, as a function of time. Just to give you a feel 21 for the variation. The variation ranges from about 1-1/2, 22 and I am looking at some eight different nuclear stations 23 from 1-1/2 to about 6, with an average of about 2-1/2, so 24 you get a significant amount of scatter, depending on what
(~}
v 25 input you give as to what ratio of amplification might be.
ACE-FEDERAL REPORTERS, INC.
namn __m _ --__ _-_mmm
'4' 1350 18 02 227 (q
_/ DAVbw 1 I think a factor of 2 might bother you, but it is 2 certainly well within the scatter.
3 DR. TRIFUNAC: Excuse me? Is that for one plant?
4 DR. STEVENSON: No, this is for a number of 5 different plants.
6 DR. TRIFUNAC: That does not respond to his 7 question. We are talking about one plant, and that 8 particular plant might have amplification of 2 and the other 9 might have 3 or 7. The question is, how good is the model 10 for a given specific plant.
11 DR. STEVENSON: I would like to address that as 12 well. And this brings in this question of secondary O 13 effects. There was an explosion experiment done by 14 Mitsubishi sevsral years ago on a steel containment, where 15 they monitored the acceleration as a function of near the 16 equipment hatch, away from the equipment hatch, near the 17 dome, et cetera. They got very significant differences in 18 that acceleration, as a function of whether or not there wan l l 19 secondary effects. And I am talking about secondary l 20 effects, due to what they call " oil canning" or low bar 21 modes of vibration. My feeling is that this difference is 22 perhaps due to that low bar mode of vibration, but as i
23 Dr. Chen explained, that hasn't been proven yet. But that 24 may go somewhere in the direction of explainaing why we
{"/}
~
25 don' t get the amplification we get.
ACE-FEDERAL REPORTERS, INC.
_ _ _ _ - _ _ _ _ - _ _ _ _ _ _ _ ________--___F~W9746F71__- _ W su ,
t__ M --__ __ - -
i 1350 18 03 228
,e m
( )DAVbw 1 Certainly, we've seen this measured response of 2 steel containments to explosive charges. Whether that, in 3 fact, explains this anomaly or not, I dons't know.
4 DR. CHEN: To further add on to what John just 5 mentioned, after the earthquake, we verified that the crane 6 was parked in the north-side directioon and that this really 7 agrees with our observation that the east-west component and 8 vertical components were in very good agreement with our 9 analysis, but most of ours had some deviation.
10 DR. TRIFUNAC: Was the earthquake north-south or 11 eas t-wes t?
12 DR. CHEN: The earthquake is about 10 miles to n 13 the sou,th of the site.
14 DR. TRIFUNAC: Almost due south?
15 DR. CHEN: I would say maybe 5 degrees southwest, 16 almost' due south.
17 DR. TRIFUNAC: So you could have Rayleigh waves 18 coming at you.
19 Did you look at the difference between the 20 north-south and east-wes t?
21 DR. CHEN: I didn't look into that. Yes, sir.
22 DR. JOHNSON: My name is Jim Johnson.
23 Professor Trifunac, I did some independent 24 analysis of the containment shell, using the recorded
}
25 motions, but using the model of Gilbert & Associates, ACE-FEDERAL REPORTERS, INC.
.F%t%EG. _M___-___-_____M-_____-_____-_--_____
1350 18 04 229 r
(,). DAVbw 1 basically. I received their model. The original design 2 model contained a soil representation of the soil by soil 3 springs. I verified their model to be consistent, my 4 version of theirs to be consistent with theirs. Removed 5 those soil springs. Treated the recorded motion on the 6 foundation, as obviously the motion, after soil-structure 7 interaction, if any, had occurred, and analyzed the 8 containment vessel.
9 For the containment vessel and the other 10 structures, found at the reactor building foundation, 11 subjected to the recorded motions, I think a couple of 12 things are, fror my point, important to note. One is that 13 the original model developed by the utility had modes, two 14 modes in the containment vessel. One at about 20 hertz, 15 which was the second highest participation in response at 16 this location in the vessel. So from the standpoint of 17 validating the models, a model did adequately represent, in 18 my opinion, the frequency characteristics of the contaimment 19 vessel versus what was measured.
20 I did analysis on the vessel, probably similar, 21 and in a limited amount, at this point in time, similar to 22 what Dr. Chen did. And a couple of things I would like to 23 mention.
24 One is, there is at least one free parameter
(~)N
\-
25 involved here. That is damping in the structure. My ACE-FEDERAL REPORTERS, INC.
_ _ _ _ _ _ _ . __---_ --_ --- -__ _ EADMED C h ------_ CBESCD------___-- - - _ - _ _ _ _
1350 18 05 230
(_)sDAVbw 1 original calculations, and I believe Dr. Chen's, were at 4 2 percent damping for all modes, including the most important 3 mode responding to the vessel and this mode, at about 20 4 Hertz, which is the second most important response or mode 5 for response of the vessel.
6 Using the constant 4 percent as a starting value 7 to this point in time, the eas t-wes t response, ZPA-wise, are 8 very similar, calculated to measure.
9 Amplified responses, I recall, the east-west 10 calculated is slightly higher than the measured. In the 11 north-south direction, the ZPA is less, and it is my opinion 12 that either through system identification techniques or even 13 trial and error, one can arrive -- by adjusting parameters 14 such as damping, could arrive at a value where you match 15 well the calculated ZPA versus measured ZPA.
16 DR. TRIFUNAC: You're talking about this peak 17 discrepancy we're juct talking about.
18 DR. JOHNSON: I am talking about the ZPA first, 19 zero period amplitude. Then the question comes, is the 20 amount of amplification at 20 hertz that you observed in the 21 calculated response spectra and the measured response 4
22 spectra perhaps as a multiplier. These comparisons were 2 23 percent equipment response damping, and whether you see the
(" 24 same kind of amplification in the record up on the
(-}/
25 containment vessel, all of us that do seismic analysis and ACE. FEDERAL REPORTERS, INC.
EB8BDED GtczcGturo CM CC'D_ ___ _ ___
1350 18 06 231 DAVbw 1 calculate response spectra, that secondary response spectra 2 off of the equipment recognizes that there are uncertainties 3 in that that are influenced by the time history records and 4 by the response of the primary structure in that location.
1 5 And many times significantly in the north-south direction, 1
6 there is a large amount of amplification, I think, at 2 7 percent, at 20 hertz and 2 percent spectra. I believe it is 8 6.2 Gs versus .55 Gs.
9 I don' t believe that one would be able to get a 1 10 to 1 correspondence in that spectral amplification at that 11 level. If that is the criteria for excellent comparison 12 between calculations and measured response, I think we would 13 not necessarily be able to get closo in that degree, but we 14 do have good comparison in terms of the f requency
- 15 characteristics of the motion, using free parameters that 16 are available to us in the model, and we know our excitation 17 dependent, like damping.
18 We can reasonably predict what the ZPA of the 19 reference would look like and consequently, the ZPA of the 20 i spectra at that location. But I am not convinced that we 21 would be able to predict the kind of amplification or that 22 amplification factor to closer tolerance than perhaos a 23 factor of 1.5 or 2.
'^
~
x 24 There is another point here that I think is l
25l impcrtant, and that is, this reactor building is founded on ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coveraat M 336-6646
1350 18 07 232 o
(_)DAVbw 1 very stiff rock. The shear wave velocities are over 4000 l
2 feet per second.
3 Even so, I don't believe that soil structure 4 interaction is a significant effect relative to the response 5 of the reactor, but if there were a small amount of rocking 6 about the east-west axis, it may have some effect upon the 7 containment vessel and could affect the records.
8 Hence, I think we are fortunate from the 9 standpoint that this structure is found on rocks, so it is 10 not -- presumably SSI is not a significant effect here, and 11 we are able to make these verifying calculations. If this 12 were a soil-founded site, and if the foundation had
[")
%> 13 significant rocking, then we would need more records than i
i 14 what we have to even get close to the kinds of calculated 15 response that we measured.
16 DR. TRIFUNAC: I just have a very brief comment.
17 Damping is an interesting parameter, but the concern that I 18 have is that the duration is so short, that the feedback 19 between the oscillator is not there to take time to build it 20 up. So I doubt this may be the key paramenter. But the I
1 l 21 thing that concerns me is what we have seen in Mexico. And '
l 22 the thing is -- that's concerned me for a long time -- that ,
I l
23 there is no soil interaction here significantly. But you '
i l
{} 24 are in the direction of the earthquake, where you might be 25 looking at Rayleigh waves, and they may be producing l
l
, ACE FEDERAL REPORTERS, INC.
_ _ _ - _ _ - _ _ - _ _ - _ - . .D___ __ __ _ _FT3-D2 _ rsT t d b @ r r n C%ffXn___ _ _ __ _ __ --__ _ __
i l
l 1350 18 08 233 1
(3,fDAVbw 1 rotational acceleration that doesn' t get into any 2 calculations period.
3 Mexico City shows that very clearly. And rather 4 than just trying to work our way around this and saying 5 maybe this is banging something against something else, it 6 might be a good idea to go back and look at the engineering 7 significance of our calculations and see whether we are 8 missing something. I 9 This may be a good opportunity.
10 DR. JOHNSON: When I said, soil structure 11 interaction was not perceived to be important, that comment, 12 directly, is intended toward inertial interaction, if there
' 13 are some other wave-scattering effects oE some other 14 wave-field effects.
15 DR. TRIFUNAC: I agree with you completely, but 16 what I am talking about is the fact that the ground is also 17 rolling as the waves are going by. We don't put that in.
18 DR. CHEN: Okay. Thank you.
19 (Slide.)
20 Next we are going to try to identify the !
21 equipment margins, irrespective of whether 688 has a 22 secondary effect or not, because we took that into account 23 This is a representative of the components which we use. We 24 will show you, this has a variety of equipments involved
}
25 here. There is a GE motor, Byron Jackson pumps, and in the l
ACE FEDERAL REPORTERS, INC. I m _____f~VSfi 3- x_ ___ --_ Mm a _ _ __ _ ENITD _ _ _ _ __ _ ___ ___ _ _ _ _ l
1350 18 09 234 m
(,) DAVbw 1 quantification analysis, these two pieces of equipment are 2 covered together for the margin identification, and we have 3 control switch gears. We also have power switch gears. And j 4 these two pieces of equipment, we have relays, and we also f
i 5 have blowers. We have pressure transmitters, flow 6 transmitters, valves, actuators and instrument ranks. So it 7 covers a variety of equipment types. It is in a variety of
- 8 baildings at different elevations.
9 We look into the equipment at the point which we
- 10 just looked at. The reactor building is 688. That is what 11 our discussion was concentrated on. Different verification 12 methods and different vendors. So we have covered a wide 13 spectrum of equipment types, building height elevations and 14 qualification methods.
15 MR. MICHELSON: Is there supposed to be a point 16 that you are making on the slide you just had up? Is there 17 a message? What are you trying to prove?
18 DR. CHEN: Sorry. In spite of the fact which we 19 have just described, that this earthquake has a very short 20 duration, high frequency, it didn' t have an ef fect on the 21 plant equipment and structures.
22 i MR. MICHELSON: Did we expect it to have an l
23 effedet?
4
/~T 24 DR. CHEN: We did not.
O 25 MR. MICHELSON: And we didn' t see any, and t
ACE. FEDERAL REPORTERS, INC.
____ _ _ ___ _ -- _ EVG-DRL - . mh ___ _ _____ CIMDCD
1350 18 10 235
()DAVbw 1 therefore, it simply verifies that plant impulses don' t 2 bother the equipment. But that is not a proof test.
3 DR. CHEN: No, it is not a proof test; no, sir.
4 We just went one step further to show that this kind of 5 earthquake doesn' t have an ef fect on the equipment. That is 6 why we picked a variety of equipment to identify the 7 margins.
8 DR. REMICK: It is obvious for tomorrow's 9 presentation, you will want to scratch engineering
- 10 significance and proof test.
11 (Laughter. )
12 MR. MICHELSON: There is no demonstration of
/^%
(-) 13 margin on that slide either.
14 DR. TRIFUNAC: Excuse me. Just a quick
! 15 question.
) 16 What is the frequency band at which your original 17 OBEs and SSEs were put in? What is the high frequency 18 cut-off?
19 DR. CHEN: In calculating the spectrum?
20 DR. TRIFUNAC: In the original design?
21 DR. CHEN: 33 hertz.
22 DR. TRIFUNAC: That can' t be.
23 DR. CHEN: The original spectra in calculating 24 the desing? ;
1 25 DR. TRIFUNAC: What is the frequency band in ACE FEDERAL REPORTERS, INC.
EscFDe cerm e e ----__ e -
1350 18 11 236 (O_)DAVbw 1 your dynamic populations in your original analysis of OBE 2 and SSE?
3 DR. CHEN: You mean the dominant frequency band?
4 DR. TRIFUNAC: No, I mean the frecuency band of 5 the calculation, the high frequency calculation. What was 6 it?
'7 DR. CHEN: 33 hertz. Unless I misunderstand your 8 question.
9 DR. TRIFUNAC: Nobody produced a statement.
10 DR. CHEN: In the calculation, we went as far as ,
11 33 hertz.
12 DR. TRIFUNAC: Waht was the cut-off figure for 13 the kinemetrics analysis?
14 DR. CHEN: The instrument is good up to 50 15 hertz. -
16 DR. TRIFUNAC: I am asking you about the analysis 17 not the instrument.
18 DR. CHEN You mean the spectrum calculation?
19 DR. TRIFUNAC: All calculations.
20 DR. CHEN: The spectrum calculation we went out 21 to 33 hertz or 50 hertz, irrespective of calculation.
22 VOICES- 35 hertz.
23 DR. TRIFUNAC: How many points per second?
24 VOICE: 250.
) 25 DR. TRIFUNAC: Well-maintained all throughout?
1 ACE. FEDERAL REPORTERS, INC.
FinSSSWD _ ._ _ M _ _ _ - M-
1350 18 12 237
()DAVbw 1 VOICE: Yes.
2 (Slide.)
3 DR. CHEN: This Vugraph shows the locations of 4 those instruments. We look at containment at 688, 5 intermediate building 620.
i 6 MR. MICHELSON: Are these devices that you looked 7 a t?
l 8 DR. CHEN: Yes, sir. This is just showing the 9 location of this equipment.
10 MR. MICHELSON: You took these particular pieces !
11 and examined them and verified that these, indeed ---
12 DR. CHEN: Yes, sir.
13 DR. OKRENT: If I can just ask a question.'When '
14 you examine a transmitter to ascertain that nothing I
15 happened, are any of those being sealed hermetically against l
16 steam environment?
l 17 DR. CHEN: I cannot answer that question. l 18 Can Dave or Jane answer that question? Are those 19 transmitters sealed against a steam environment?
l j 20 VOICE: I don' t know the answer to that.
21 DR. OKRENT: If you don' t know the answer, how do 22 you know whether this might have been a requirement, which 23 is very hard to check, by your inspection and whether you
(} 24 have lost it, if it was a requirement?
25 DR. CHEN: I think when you are talking about l l
I ACE. FEDERAL REPORTERS, INC.
__ rmPwe:n ma - . c .m- - _ _ _
1350 18 13 238
()DAVbw I sealed against steam, I suppose it is part of the 2 environmental qualification.
3 DR. OKRENT: It was originally; yes. You 4 mentioned earlier there was something that was developed --
5 that would develop a leak here and a leak there. I am 6 trying to see -- it is not an easy thing to check by eye.
l 7 In fact, it is quite difficult. You have to run a special 8 test. I don' t know that this equipment was so sealed, but l
9 you don't know either.
10 MR. EDELMAN: Dr. Okrent, one, most of this 11 equipment is located outside containment, so therefore, it l 12 would not be subject to the steam qualification in that 13 environment.
l 14 Second, what Dr. Chen is trying to do with this l
15 equipment in the subsequent Vugraphs, is not taking any 16 specific piece of equipment and saying we expected that, but 17 looking at the design seismic response spectra that we did 18 yet to qualify the equipment and then analyzed are shown on 19 subsequent graphs where we are, relative to the actual 20 measurements of that equipment that we would have been l
i 21 exposed to in our readings in the plant.
i i 22 i 23 ,
I 24
(~}
\/
l 25 l l
i l
! ACE FEDERAL REPORTERS, INC.
. _ ----- - --- -- _ _-- fBfS9iB - -- -- --- - G2TCfDh - C --_-- -----
1350 19 01 239
()DAVbw 1 DR. OKRENT: Again, you're really answering a 2 different question than the one I am meaning, and what I am i
3 curious about is the possible, I don' t know, that the high 4 frequency component led to leaks in small piecen of l
5 equipment of the kind I have just indicated. It is not that 6 there is something pouring out. It isn't important, unless 7 you get into an accident situation. And by the way, it l 8 is not clear to me that only the roactor building is the ,
9 building where you hear about the steam environment.
10 MR. EDELMAN: There are some other parts.
11 DR. OKRENT: Yes. I don' t accept your prior 12 answer. Maybe you can convince me, but those words were not 13 enough.
14 MR. MICHELSON: Did you'look at your heating and
! 15 cooling systems, which are often in higher reaches of the 16 building, where you have more amplifications, which are
. 17 pieces of equipment? Did you actually physically examine 18 any for possible rubbing or whatever?
19 MR. EDELMAN: That equipment was working before, 1
20 during and after. We've had our inspections as described by 21 Mr. Stratman. We've noticed no problems.
22 MR. MICHELSON: I don't know that Mr. Stratmena 231 described examining blowers, for example, which is a good >
l l {} 24 25 place to look to see if things have really been knocking around.
l ACE. FEDERAL REPORTERS, INC.
________ - _ _-__ M _--___ _ M _ ___ __ _ M
'. 1350 19 02 240 i
()DAVbw 1 MR. STRATMAN: To my knowledge, unless a piece of 2 equipment has subsequently broken, we did not go up and do 3 as you suggested.
4 MR. MICHELSON: Those fans can just bend. You'll 5 just bend the blades, unless you are standing nearby. They 6 don' t stop. Not too likely. It is an indicator of how much 7 effect this kind of event could have.
, 8 MR. LEE: As part of the seismic qualification i l
9 review at the Porry site, we actually went through the ,
10 l seismic qualification review, as part of the sequential 11 testing, sequential qualification. "nat included the 12 environmental qualifications that ou just talked about.
<]'
1-3 First of all, we did all the OBE testing, SSE 14 testing, after that, followed by EB.
15 So all the environmental effects have been 16 included in the sequential testing.
j -
17 MR. MICHELSON: In the testing program?
18 MR. LEE: Yes.
j 19 MR. MICHELSON: You don' t test your large ;
20 blowers, for instance.
21 MR. LEE: Transmitters.
22 MR. MICHELSON: We're talking about ventilating
, 23 blowers. i
! l
{} 24 MR. LEE: Some of the equipment is not here, if l
),
25 this is in the severe environment. l i
i
! ACE. FEDERAL REPORTERS, INC.
_ _ _ _ _ _ - - - - _ _ - ---__ __--__--__--- __ rn a vi n -- _ _ m _- -- - h - - - _ -_1
1350 19 03 241 I ,JDAVbw 1 MR. MICHELSON: This is an opportunity to look at 2 equipment. It is still a rather mild inpulse, but it starts 3 to show significant damage, when we can only find that out 4 on a blower by examining it. It will probably go through 5 the storm all right, and it wasn't a high G value area of 6 the building, probably. It is an interesting piece of 7 equipment to look at.
8 VOICE: I believe he was trying to answer the 9 question on possible leakage in the transmitters.
10 MR. MICHELSON: Okay.
11 VOICE: 10 Gs on the shake table 12 MR. MICHELSON: I am not sure -- you don' t shake
(~}
k' 13 them just before you put them in on autoclave, do you? I 14 mean, .you can' t prove that the environment caused by the 15 earthquake affects the environmental qualifications, because 16 I don't think you shake just before you do the environmental 17 qualification.
18 MR. LEE: You shake.
19 MR. MICHELSON: It is a separate test at a 20 separate point in time.
21 MR. LEE: It is a separato test.
22 MR. MICHELSON: So I am not sure you can ever 23 draw the conclusion that earthquakes don't break the
/~T 24 hermetic seals.
V 25 MR. LEE: But I think after the SSE tests, all ACE FEDERAL REPORTERS, INC.
. EnPREM ____N---.___h-____________
1350 19 04 242 m
I DAVbw 1 the anomalies have been observed. If this is an anomaly, it 2 will have to be fixed before it goes through the 3 environmental test.
4 DR. OKRENT: I am sorry. You've left me really 5 feeling like I am not quite sure whether the point you made 6 was valid. Originally, I think you were trying to reassure 7 me that, in fact, the equipment had been tested in 8 conditions more severe than was the case here, and that it 1
I 9 was subsequently checked in the environment.
10 MR. LEE: The DBE case, for example.
1 l
l 11 DR. OKRENT: What I heard a moment ago was that, 1
l 12 in fact, these tests are done by separate or may be done by l (~~)
13 separate vendors. It is not a requirement that the 14 environmental tests be done after the shaking on the same 15 piece of equipment.
16 MR. LEE: It will be on the same piece of 17 equipment, all right. I am not sure of the time.
l l 18 DR. OKRENT: If you're not sure about the time, l
l 19 then I think ycu'd better go home and do some homework l 20 before you complete the answer to the question.
21 MR. LEE: We will do that i
l 22 MR. MICHELSON: Mayne the utility can do that 23 when they describe their program.
1 (~ 24 DR. OKRENT: You see, I don' t know that there is V]
25 a problem here. I am only probing, because I have been so l
ACE FEDERAL REPORTERS, INC. '
F v m in- M r. t m _______________ _
1350 19 05 243
( ) DAVbw 1 reassured.
2 MR. CHEN: You can be reassured.
- 3 (Laughter.)
1 4 DR. OKRENT: There's probably something they l
5 missed. ,
j 6 MR. GRAHAM: Dave Graham, from Cleveland l 7 Electric.
8 I don't think I have a direct answer to your ,
9 question, but the way the program goes is to age the 10 equipment, to put it through temperature, humidity and 11 radiation age. Put it on a shake table and then to r
l 12 functionally operate it after it has been on a shake table. !
l l i
13 DR. OKRENT: That wouldn' t then tell me whether
(
14 the shaking was effective. f' l
15 MR. GRAHAM: I understand that. ;
16 DR. OKRENT: Okay.
17 (Slide.) i i
i 18 MR. CHEN The next one shows the other locationn '
! 19 of those equipments. i l >
10 DR. OKRENT: By the way, just for my own benefit, 21 where do you think we are in the presentation from Perry?
22 HR. EDELMAN: I think there are certain things, !
23 Dr. Okrent, that we could cite one example rather than go i
24 through all the Vugraphs, to expedite Dr. Chen's talk in the l
25 next few minutes.
l l l
! l ACE FEDERAL REPORTERS, INC. '
____--__ _ 1 3fS 2 3-_ ----- _ N --- _ _ N __
1350 19 06 244 (3 DR. OKRENT: According to the agenda i
V DAVbw 1 Le t's see.
2 I have, there are four more presentations.
3 MR. EDELMAN: They are a lot shorter. ,
i :
4 DR. OKRENT: How much more time would they take, 5 uninterrupted?
I 6 MR. EDELMAN: I would say 40 minutes.
! 7 DR. OKRENT: How much more time does Dr. Chen l
8 think ho needs?
I 9 MR. CHEN: I can speed up and skip a few of the 10 Vugraphs.
11 MR. EDELMAN: He can be done in about five 12 minutes.
i 13 DR. OKRENT: We are interested not --
14 MR. CHEN: I am going to remove some of the 15 pattern.
16 DR. OKRENT: We just want you to, in fact, show l
- .> 17 us -- try to think very hard of what could be the problem l
18 and show that, in fact, you have looked at various points l 19 at the ottier 8000 components.
20 MR. CHEN: All those equipment qualifications --
21 t'll just show you the one which I think is critical.
22 (slide.)
23 This one is the qualification curve for those ;
j 24 transmitters in the instrument racks. The orange line is 25' the origini qualification curve and the green line is the ACE FEDERAL REPORTERS, INC.
_ rvman r n~m
1350 19 07 245
()DAVbw 1 amplified spectrum at the location where those transmitters 2 are located inside the instrumentation. As you can say, the 3 resonance frequency of the rack is right around 20, 10, 20, 1
4 30, between 20 and 30 hertz. That is where I think, in the 5 sense of combining resonance with the frequency, it is the 6 most critical case.
I 7 Even with this case, the transmitters, the flow l 8 transmitters, the pressure transmitters,.were all qualified 9 to this curve and the amplified spectra at the location of 10 the transmitters is the green curve.
11 Now I am going to move to other material again.
12 DR. SMITH: Dr. Chen, on that curve, I am puzzled O* 13 by the amplitude of the amplified spectra. Is my l 14 understanding correct, my figure shows that that is a 5 !
l 15 percent damped spectra showing 7 Gs, that green curve.
16 DR. CHEN: Yes, sir.
17 DR. SMITH: Now from our measured 2 percent 18 spectra, which is about 6 Gs, this would seem to correlate 19 to 10 or 12 Gs at 2 percent. Again, I don' t know where this 20 is.
21 DR. CHEN By the way, I am looking at a 22 different location. I think the one you referred --
23 DR. SMITH: This just seemed very high.
24 DR. CHEN: Yes. This is very high, because the 25 way the amplified spectra and the location of tho flow ACE. FEDERAL REPORTERS, INC.
PRJiVJ3 m m---.--- e _ _ _
1350 19 08 246
()DAVbw 1 transmitters was buried. That is why this is very high.
2 DR. SMITH: Could you walk me through how you got 3 that?
4 DR. CHEN: This curve? Okay.
5 DR. SMITH: What was the starting point?
6 DR. CHEN: The starting point is the recorded 7 spectrum by Endo at that eleration. That is the aux 8 building, 586. That is the basement of the auxiliary 9 building. We used that recorded spectrum as the starting 10 point. We used that as input to the instrument racks, and 11 we did an in situ test to determine the transfer function 12 between the support location of the rack and the locations 13 of -- the transmitter locations.
Okay? And we combined the -
14 input spectrum with the transfer function derived from the 15 amplified spectra at the flow transmitter locations. That 16 is how it was derived. So it is much higher than the way it 17 was located, because it is amplified once already.
18 DR. SMITH: Then you chose to use a 5 percent 19 spectra for the amplified.
20 DR. CHEN: The reason we used 5 porcent was 21 because the original qualification curve was 5 percent. If 22 you want to derive the 2 percent curve, that yellow line 23 would move up.
j
{) 24 25 DR. SMIT!!:
DR. CHEN:
I want to compare apples and apples.
Yes. It is 5 to 5. The green line is ACE FEDERAL REPORTERS, INC.
--_____ ________ f~WiFiR_____ __M . . __ m
1350~19 09 247 O
t/DAVbw I also 5 to 5.
i' 2 DR. SMITH: So this is the amplified in cabinet.
. 3 You chose a very high application in the cabinet.
t 4 DR. CHEN: It is very high, because the way it 5 was derived was a,very conservative approach.
6 DR. SMITH: Thank you.
7 MR. MICHELSON: How did you get the red line 8 again?
9 DR. CHEN: The red line is the original 10 qualification curve supplied by the vendor.
11 MR. MICHELSON: That is not the design curve.
12 ' That is the one supplied by the vendor.
13 MR. EDELMAN:
14 DR. CHEN: Yes.
15 MR. MICHELSON: Just by happenstance in this 16 case.
- l 17 DR. CHEN: We also keep the one in the instrument ;
i l l 18 which is right in resonance with the frequency spectrum.
19 DR. REMICK: Is that the answer to the question,
] 20 that you would get to transmitters or some of the j 21 equipment you would expect to have the greatest response?
l 22 DR. CHEN: Yes, sir. Because we pick the ,
I 23 equipmont which is right in resonance with the frequency
{} 24 I
content.
25 DR. REMICK: Are there any other items of l
I ACE FEDERAL REPORTERS, INC. I s m r m r; n W t________ m___ _ ____ __ !
i l 1350 19 10 248
/~~T V DAVbw 1 equipment besides the transmitters that you would suspect 2 would give you that type of response, your most susceptible 3 response?
4 DR. CHEN: I think this is the one we have found l
l 5 so far, but we are in the mode of confirmatory analysis, and 6 we are concentrating on high frequency work.
7 That will be the last Vugraph I am going to show 8 you, per Dr. Okrent's request to remove all the petty 9 material.
! 10 (Slide.)
11 In spite of the fact we have performed all those 12 analyses, CEI has committed to perform further analyses, and 13 those are the criteria we are going to follow. This is the 14 next to the last one of my package.
I 15 ' We concentrate on the active safety class l 16 equipment required for cold shutdown. The second criterion 1
17 is based on the report written by Lawrence Livermore l
l 18 National Lab. That lists a set of equipment for Perry Power 19 Plant with frequencies higher than 40 hertz and with HCLPF 20 value lower than .05 G. HCLPF means high confidence low 21 probability value. In this criteria, we have taken the most 22 critical group of material to perform further equipment 23 modifications, quantificationof the margins. And the rest 24 are those equipments which we listed earlier.
25 And so far, we have completed 75 percent of the 1
l l ACE. FEDERAL REPORTERS. INC.
______-______________ m _________ m .. -
_______m__
1350 19 11- 249
('S
(_/ DAVbw 1 equipment types in our confirmatory probe.
2 DR. SMITH: On some of the graphs you did not t
3 show, I have to come back to your Figure ~6, which is the 4 switch gear on this on, if I interpret it correctly, you 5 compare the measured amplified response at 3 percent damp to 6 the 1 percent test spectrum.
7 MR. EDELMAN: S 8 DR. CHEN: Switch gear. Okay. Le t me see.
9 DR. SMITH: Figure 6.
10 DR. CHEN: Yes. I should have pointed that out.
11 DR. SMITH: I only bring it to the attention, 12 because if what I surmise is correct, those figures are O 13j actually very close together., The measured response would 14 still be under the to'st spectrum, but I think for the 15 record, we should note that you are comparing apples and 16 grapefruit here.
17 DR. CHEN: Yes, sir. Let me qualify that point.
18 The reason we compare this orange line is the 1 percent 19 curve. The reason we compare it with this curve is because 20 that was the only curve that was available.
21 DR. SMITH: Wait. I am sorry to interrupt, but 22 s that is not a true statement. That curve has already been 23 corrected, adjusted from 2 percent measured down to 3 i
(} 24 percent, as stated eight in your test. ,
25 -
DR. CHEN Are you talking about this one?
ACE FEDERAL REPORTERS, INC. i m_____________ _______-_____W. t --
1350 19 12 250 i
')DAVbw 1 Okay. We could have, but the reason we didn' t do 2 it is because of the margin here. Even if you do it, you 3 will still show a large margin here.
4 DR. SMITH: My point is, if you do it right, that 5 is not a large margin. That is a very small margin. It
- 6 appears that someone is trying to make it look better than l
! 7 it is. It is good enough. You don' t have to change it.
8 DR. CHEN: This one. If you make a proper 9 adjustment, this will go down. But it was not our intention 10 to show you a better margin than it really is, because I can 11 show you the other case.
12 (Slide.)
13 The testing spectrum is 5 percent. We are I
14 comparing 5 percent with 2 percent in class spectra.
15 DR. SMITH: I understand that.
16 DR. CHEN: So this is the oth9r way around. If i 17 you adjust for the 2 percent, this will move up. You will i 18 show even more margin.
t .
19 DR. SMITH: But the switch gear are the critical l
l 20 things. That's got the relays in it.
i 21 l
22 23 1
1 24 1
(^}
G l
, 25 l
ACE FEDERAL REPORTERS, INC.
s~ W;w m ______ x m .m-
i 1350 20 01 251
(_)DAVbw 1 DR. CHEN: Even af ter we make the adjustment, you 2 will still have the envelope.
3 DR. SMITH: I see it will still be under; 4 however, I think there is an important point to bring out.
5 Also, this is based on amplification of a factor of 2. That 6 should explictly be made clear.
7 MR. EDELMAN: I think, Professor Smith, what you 8 are saying is that all the curves we have all show below the 9 design. If you taake the amplification or the f actors the 10 same, you will still show that. In some cases, it will be 11 closer, but in all cases, it is below; is that correct, sir?
12 DR. SMITH: That's correct.
(/ 13 MR. EDELMAN: Thank you.
14 (Slide.)
15 DR. CHEN: This will be the conclusion. -
16 DR. OKRENT: We will assume we have seen that.
17 , Okay.
1 .
I 18 Now let's see if there are any other questions 1 19 for this presentation, while it is fresh in people's minds.
20 DR. TRIFUNAC: Would yo take the Vugraph that 21 Dr. Smith was looking at.
22 What was the reported intensity at the site? I 23 DR. CHEN: 6, epicenter.
24
(~] DR. TRIFUNAC: I am asking at the site.
25 DR. CHEN: At the site?
ACE-FEDERAL REPORTERS, INC.
207 347-3700 Nationwide Coverage 800-3NHW86 - - - - -
1350 20 02 252
)DAVbw 1 VOICE: A maximum of 5.
2 DR. TRIFUNAC: And the peak acceleration was?
3 DR. CHEN: The peak acceleration at the site 4 recorded at the top of the map, .18 G north-south, east-west 5 .1, vertical .11.
6 DR. TRIFUNAC: Say .1 to .2 G. What is the 7 design intensity at the site?
8 DR. CHEN: What is that 77 9 DR. TRIFUNAC: I am asking you.
10 DR. CHEN: I was asking Dick. 7? Was it 7?
11 Yes. 7.
12 DR. TRIFUNAC: What sort of factor do you go to
\~' -
13 l from intensity 5 to intensity 6 and from intensity 6 to 14 intensity 7, for,.say, a peak acceleration?
15 DR. CHEN: What would I expect to go from 16 intensity 5 to 6? I do not want to venture into that 17 discussion. I prefer -- Dick, would you prefer to answer 18 that question?
19 MR. EDELMAN: Dick Holt, our 20 seismologis t-geologis t from Weston Geophysics.
21 MR. HOLT: The question, as I understand it, 22 Dr. Trifunac is, what would I expect to increase the peak 23 acceleration?
/^)
v 24 DR. TRIFUNAC: What are the different kinds of l 25 recording that are typically observed between'different ACE-FEDERAL REPORTERS, INC.
__2iaaanittcman sane 1
1350 20 03 253 1_)DAVbw 1 successive levels?
2 MR. HOLT: If I were to use a normal relationship 3 of setting a Reg Guide response spectrum of 1.6 --
4 DR. TRIFUNAC: I am not asking for results; I am 5 asking about observations about measured differences. If 6 you have peak acceleration at MMI 5 or so much, how much 7 bigger is it at 67 How much bigger is it at 7 or 8? What 8 is the number that you multiply things as you go up from one 9 intensity to another?
10 MR. HOLT: I think if I used the Trifunac and 11 Brady curve, it would be about a factor of 2.
12 DR. TRIFUNAC: Any curve is about a factor of 2.
g
?
13 So at the site, you have intensity 5 and a peak acceleration 14 of, say, .1 G. How much would it be if the intensity were 15 6?
16 DR. CHEN: Dick, could you answer that question?
17 ! MR. HOLT: To associate an intensity and 18 acceleration value, I certainly would not expect to see 19 something in the longer period lower frequency and .2 G 20 associated with intensity 5.
21 DR. TRIFUNAC: Look. This is the same kind of 22 question that we went through on er.gineering design. There 23 is a theory, and there is a measurement. Okay. A lot of
(")
K.;
24 people publish papers which are theories. So you have here l
{
25 a case of intensity at the site is 5. And the peak ACE-FEDERAL REPORTERS, INC.
3 2-347-3700 Nationwide Coverage 80;L3M-NH6 _
1350 20 04 254
/
i/ DAVbw 1 acceleration is like 10 percent G. So under the same 2 conditions and the same geology, the same tectonics, the 3 intensity is not 5 but 6. Could it be factors of 2 bigger?
4 MR. HOLT: Sure it could be a factor of 2 5 bigger. It could also be a factor of 2 smaller, because the l 6 range and acceleration value for a given intensity is at l
7 least an order of magnitude.
8 DR. TRIFUNAC: That is why I went from 18 percent 9 to 10 percent, just to allow that factor of 2 there. Now if 10 it went from 6 to 7, which is your design level, it could go 11 another factor of 2. !
i 7, 12 MR. HOLT: It could also go down.
( l 13 DR. TRIFUNAC: Are we talking about the same 14 thing?
l 15l MR. HOLT: I think we are. The scatter of the i
16 acceleration.
1 17 DR. TRIFUNAC: Okay. Let's talk about the mean !
1 18 value of the data, and let's assume that this particular f 19 point is not an unusual point.
20 MR. HOLT: Well, the point is an unusual point 21 DR. TRIFUNAC: The current earthquake is an 22 unusual earthquake?.
23 MR. HOLT: With respect to associating
() 24 acceleration with intensity, the answer is yes.
25 DR. TRIFUNAC: Why is this earthquake unusual?
ACE-FEDERAL REPORTERS, INC.
- m- _
1350 20 05 255
)DAVbw 1 MR. HOLT: If I look at the scatter of 2 acceleration values with respect to an intensity 5, 3 certainly, if I looked at .2 G, it would be at the high end 4 of the scatter, and the means value would be well below 5 that.
6 DR. TRIFUNAC: This is the data for California.
7 We are talking about a totally different region. It is just 8 like having a lot of theories in the calculation of the 9 response and then actual measurement in actual building.
10 Actual measurement and actual measurement is always much 11 more accurate than a theory or measurement in some other 12 building. This is the earthquake that was recorded here in
/^)
\' 13 this province with all the geological and other features 14 'that are there.
15 MR. HOLT: The mean value for intensity would be 16 substantially more than .2. The reason is that the i
17 intensities vary substantially with respect to foundation l 18 and soil mechanics, and this particular intensity was 19 i recorded on rock, which is typically low in intensity and l l l 20 retains its high frequency from the soils to the plant.
21 DR. TRIFUNAC: I am asking you such a simple 22 question. I know that it isn' t the rock and the recording l
23 was at the site already, as it is. So there is no need to i
('; 24 qualify it.
l 25 MR. HOLT: I would not take the .2 and double it ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nananwide Coverase M 13Minen
1350 20 06 256 7 I
( ,)DAVbw l' for each intensity after that. I am saying that the scatter 2 with respect to intensity is too great.
3 DR. TRIFUNAC: So California data suggest that 4 the mean levels go up by, say, a factor of 2 or so from one 5 level to another. And here you have a record of, say, 30 i
6 percent G.
7 MR. HOLT: One record does not constitute a 8 mean.
I 9 DR. TRIFUNAC: Of course not.
10 DR. OKRENT: Do you think we are going to have 11 more than a standooff on this, Mike?
12 (Laughter.)
7- ,
\#
- MR. HOLT: Probably not, Dr. Okrent.
13 ]
14 DR. TRIFUNAC: I'm just trying to have a simple 15 question answered.
16 DR. OKRENT: I think we'd better try another 17 one.
18 MR. RIGHTER: Excuse me. Leon Righter, from the 19 Staff.
20 This is a perfect example of why we really don' t 21 think peak acceleration, particularly related to intensity, 22 is really a meaningful way of designing a nuclear power 23 plant.
~
r 24 DR. OKRENT: I think Craig had a question. Did v
25 you have a question?
l I
1 ACE-FEDERAL REPORTERS, INC. l 00 -)17-3700 Nationwide Coverage 80013W I
1350 20 07 257 t,/DAVbw 1 DR. SMITH: No more.
2 DR. OKRENT: No more on equipment. All right.
3 Other questions.
4 DR. TRIFUNAC: I didn' t get an answer to this 5 one. I'll stop here.
6 (Laughter.)
7 DR. OKRENT: Do you feel it is an important one?
8 DR. POMEROY: I think we are going to come back 9 to it.
10 DR. OKRENT: Put it another way.
11 DR. POMEROY: Dick is going to give a 12 presentation.
'# 13 DR. OKRENT: You want to wait until Dick is 14 standing in front of you, so you can shoot at him more 15 effectively.
16 (Laughter.)
17 DR. OKRENT: Okay. Are there any other questions 18 for this speaker now?
19 (No response.)
20 DR. OKRENT: I am going to suggest that we take 21 about a seven-minute break. We've already gone about 2-1/2 22 hours.
23 (Recess.)
(
. 24 MR. EDELMAN: Our next speaker is Dr. Stevenson.
t I
25 It is a very short presentation.
ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Covera e 800 336-6646
I 1350 20 08 258 l
,e
( / DAVbw 1 DR. STEVENSON: This will actually bear directly 2 -- it will be so short, I won't even try to put it down.
3 (Slide.)
4 But basically, this bears somewhat on the 5 question that was asked just before the break. What is the 6 acceleration that we would expect? What I tried to do is 7 independently evaluate for this earthquake the actual 8 measured data, what the information in the general 9 literature would have told us that this acceleration was 10 equivalent to.
11 What I am suggesting here is that there was work 12 done by a number of different authors summarized back in '81 s
by Kennedy on, as he put it, the peak acceleration as a 13l 14 measure of damage.
15 What I tried to do is fit this earthquake to the 16 methodology used by Housner and a number of other authors to 17 come up with an equivalent design acceleration, as a measure 18 of damage, where they effectively looked at the energy input 19 from the earthquake, determined power and, in effect, came 2C ! up with a procedure to come up with an equivalent design.
21 This would be associated with 0 ground 22 acceleration. I used the basic design assumption that was 23 used by Gilver, which would imply that this was a .15 G free
{} 24 field, where they measured .18G at the base of the 25 containment. And using the methodology that is contained in ACE-FEDERAL REPORTERS, INC.
,m.m.m w. _ _u. m = ms-
1350 20 09 259 DAvbw 1 that original report that was given to the ACRS a month ago, 2 we determined for this particular earthquake, that we get an 3 equivalent free field acceleration of .073 G, on this 4 basis. That was one approach we took.
5 A second approach was again to use a procedure 6 somewhat independently, developed again by Kennedy in his 7 paper, and more recently, also used in NUREG CR3805 --
8 (Slide.)
9 -- where he attempts to come up with an equivalency or 10 damage measurement potential. In that one, we took some 11 four frequencies and the amplification spectra for Reg Guide 12 160 and the spectrum from the Ohio '86 earthquake and O
V 13 compared them and determined that the Ohio earthquake had an 14 equivalency with regard to comparing the spectrum and 15 specific frequency values.
16 We used the same four frequency values that 17 Kennedy had used, plus we added 4 hertz frequency, because 18 this is one of the major areas of energy input from the Ohio 19 carthquake and the 22 hertz, which was the dominant energy 20 content of the Ohio earthquake, and found that on a design 21 Reg Guide basis, the Ohio earthquake, in terms of the energy 22 in those six frequencies, the average ene:gy through the six 23 frequencies came out as little as .02 G, again, using the j 3 24 methodology suggested by Kennedy in his earlier Saper and (G 1 25 more recently in the 3508.
l ACE-FEDERAL REPORTERS, INC.
202 347 3700 Natbnede Coverase 800-33H646
1350 20 10 260 uj DAVbw 1 Really, what we are saying here is that using the 2 techniques that are available in the literature --
3 DR. OKRENT: Before you take that away, I wonder
. 4 if it that is really a meaningful thing to put up, in the 5 sense that if you have something that is vulnerable either 6 4 or 22, averaging it, gives you a feeling of calm maybe you 7 shouldn' t have.
8 DR. STEVENSON: It is meaningful for those 9 components, 90 percent of which are below 10 hertz. Now 10 whether it is meaningful for components that are 11 high frequency-sensitive, the answer is, obviously, it may 12 not be.
( ,
13 DR. OKRENT: But put a qualification like that on 14 the Vuc 'oh, and this is what bothers me.
15 DR. STEVENSON: Well, I said they were average.
16 I am not trying to represent this as anything but what it 17 is. I am simply taking the techniques and methodology that 18 are available in the literature and applying it to this 19 particular time history. ,
20 DR. OKRENT: Okay.
21 (Slide.)
22 DR. STEVENSON: The conclusion I reach is that 23 the Ohio '86 earthquake has a response equivalency to the f3 24 l design earthquake Reg Guide 160, well below the design OBE.
l 25 l Therefore, no damage to plant should have resulted. Again, '
l ACE-FEDERAL REPORTERS, INC. I
?O* 147.1700 %rinnwide Cnverane Adik11&AMn
1350 20 11 261 k_)DAVbw 1 this is in relation to the design for one earthquake being 2 equivalent to the other. This does not mean that there is 3 not the possibility of high frequency-sensitive components, 4 which may or may not affect the value of the eartnquake.
5 That is really 'l I have to say.
6 DR. OKRENT: Mike?
7 DR. TRIFUNAC: Do you think that we can safely 8 assume that the shape of the spectra in the eastern and 9 western United States might be the same or similar?
10 DR. STEVENSON: It is hard to say that, because I 11 think probably, first of all, it is a question of the shape 12 of the near field and far field earthquakes. We know they
^
(~#)
k' 13l are not.
14 DR. TRIFUNAC: Given the same conditions, would 15 they be similar or different or what?
16 DR. STEVENSON: I don' t think we have an updated 17 answer to that question, frankly.
18 DR. TRIFUNAC: So the answer is that you don' t 19 know?
1 20 DR. STEVENSON: I would say right now we don' t 21 know, since we have not measured any far field strong motion 22 earthquakes in the East.
23 DR. TRIFUNAC: I can only as what is in those
'} 24 papers in Kennedy and Housner and others that you have 25 mentioned, because I haven' t seen them, but if I speculate l l
I ACE FEDERAL REPORTERS, INC.
202-347 3700 Nanonwule Coverase aufk33&dM6
1350 20 12 262
(~)S
(_ DAVbw 1 what might be in those papers, would their analysis apply, 2 or would you use them if the spectra in the eastern United 3 States did, indeed, have a different shape?
4 DR. STEVENSON: They apply, to the extent that 5 near field, shallow earthquakes are characteristic of the 6 earthquake we had in Ohio.
7 DR. TRIFUNAC: Is this earthquake we are looking 8 at a near field earthquake?
9 DR. STEVENSON: Yes. Well, I would judge it is a 10 near field earthquake, 11 DR. TRIFUNAC: Do you think that seismologists 12 would call this a near field earthquake?
h') 13 DR. STEVENSON: I consider a ten-mile epicentral 14 distance as near field. If there is another definition that 15 a seismologist wants to use, then it is not a near field, 16 but I consider within ten miles is near field.
17 DR. TRIFUNAC: Does your conclusion really have 18 to depend in invoking near field at all?
19 DR. STEVENSON: No. It depends on using, as 20 examples, earthquakes with similar energy magnitudes from 21 similar epicentral distances.
22 DR. TRIFUNAC: My question is this, again, I'll 23 repeat it.
(} 24 Given the methodology of what you referred to, 25 the names of Kennedy, Housner and a few others came out, i
l ACE-FEDERAL REPORTERS, INC.
ww.voo y. a==u= ev.=== m1u -
l l
l 1350 20 13 263 i
(_/ DAVbw 1 would that methodology allow you to make those conclusions, 2 if you think the spectra in the eastern United States were 3 different?
4 DR. STEVENSON: If they are different than the 5 measured spectra at the Melody Ranch and those types of 6 earthquakes in the West, the answer would be no, they would 7 not be comparable, but I have no basis to assume that one 8 way or the other.
9 DR. TRIFUNAC: So your conclusions would go only 10 if they are cimilar?
11 DR. STEVENSON: No, what we measured, certainly, 12 in Ohio is similar.
O 13 DR. TRIFUNAC: To what?
14 DR. STEVENSON: To Melody Ranch and similar 4.8 15 to 5.3 magnitude epicentral distances from 8 to 10 miles.
16 They are similar. I mean, similar by what measure. They 17 tend to be high frequency. They tend to have a low energy 18 content. They are similar. I mean, there is no, to me, 19 definitive parameter relationship that would say similar or 20 not similar. But they are similar in appearance. They are 21 similar in duration. They are similar in frequency 22 content.
23 l DR. TRIFUNAC: Let me rephrase the question.
(~ 24 Reg Guide 160 has a certain shape; doesn't it?
% )/
25 DR. STEVENSON: Yes.
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverame 800 33Hn46 . . . . _
l 1350 20 14 264
_J DAVbw 1 DR. TRIFUNAC: That shape comes from where?
2 DR. STEVENSON: It comes from 22 earthquakes, 3 mostly far field. 22 horizontal and 11 verticai.
4 DR. TRIFUNAC: Okay. So that is based on some 5 earthquake data; right? And what if the shapes of the 6 portions that you are recording here did not fit those that 7 built up 1609?
8 DR. STEVENSON: Well, they definitely don't. 160 9 are earthquakes rich in low frequency content, because by 10 and largo, they were large earthquakes, damaging earthquakes 11 at some distance from the site at which they were measured.
12 So you are quite right. This earthquake does not relate
' ~ '
)
13 very well to Reg Guide 160 earthquakes. And you really -
14 wouldn' t expect it to cause any damage that one coul.d
~
15 determine.
16 Your question is well-put, that Reg Guide 160 17 doesn't cover this kind of earthquake. That is right. All 18 I am trying to do is, using the technique and methodology 19 used by other researchers for similar kinds of earthquakes, i
20 to relate the equivalency.
21 DR. TRIFUNAC: Small California earthquakes of 22 ccmparable spectral shape were not damaging either.
23 DR. STEVENSON: That is correct.
( ') 24 DR. TRIFUNAC: So I don' t understand what you are 25 relating.
l ACE FEDERAL REPORTERS, INC. 1 I
n,.u .vnn ymnn.a, em,.u, un num
1350 20 15 265
_ DAVbw 1 DR. STEVENSON: What we are trying to show is, 2 using those same rclationships that were used on the smaller 3 California earthquakes that didn' t cause any damage as well, 4 to try to relate them to Reg Guide acceleration effective 5 design values, which is, again, simply following a procedure 6 done by other people.
7 Now that is all I am trying to show you. I am 8 not trying to say that this is equivalent to any particular 9 value, just simply showing what other people have done when 10 we ecmc up with cimilar relationships for this specific 11 earthquake.
DR. OKRENT: I dons' t think their main argument
. 12l I t )
13 hinges on this point. Do you know, Mike?
,14 DR. TRIFUNAC: I don't quite understand it.
15 16 17 18 19 20 21 22 23 l l
(~') 24
%J l 25 ' ,
l ACE-FEDERAL REPORTERS, INC.
r w.rm Numnwufe cm er2nc am.1tsu
1350 21 01 266 DAVbur 1 DR. MARK: Out of deep ignorance, what kinds of 2 things focus their sensitivity on 20 hertz?
3 DR. STEVENSON: What kind of components?
4 DR. MARK: Components, structures, whatever.
5 DR. STEVENSON: Primarily components that are 6 high frequency sensitive.
7 ( Laughte r. )
8 DR. MARK: Well, that I was able to guess.
9 ( Laugh te r. )
10 DR. MARK: Are they pencils or desks or what?
11 DR. STEVENSON: Usually we consider in that
_ 12 category relays, switches --
s /
4 13 DR. REMICK: We were told transmitters.
14 DR. STEVENSON: Transmitters.
15 DR. MARK: Not shear walls, not pumps or valves?
16 DR. STEVENSON: Most pumps and valves --
17 certainly, you can find pumps and valves that have 18 frequencies in that range, but they are not sensitive to 19 these kinds of things.
20 DR. MARK: They are small things. They include 21 small pieces of equipment. They don't include structures.
22 DR. STEVENSON: In general, yes.
23 DR. OKRENT: I think well-supported small 24 piping.
(~^
25 DR. STEVENSON: Although I think German showed a ACE-FEDERAL REPORTERS, INC.
- . ~ _ _
1350 21 02 267 l DAVbur 1 study that less than 10 percent of the piping in the Newport 2 plant, talking about feet, have frequencies above 10 hertz, 3 fundamental frequencies of about 10 hertz. So there is some 4 data. You do have a band of information.
5 DR. MARK: At Perry, then, things of that sort 6 have been looked at with this question.
7 DR. STEVENSON: That is what I think that really 8 Jay Chen was discussing.
I 9 DR. OKRENT: Yes. In a sense, that is what he 10 was discussing.
11 DR. MARK: Thank you.
12 DR. OKRENT: We had better go on to the next
" 13 presentation now. Let's go on. You know there is another 14 topic tonight besides Perry. I am not sure when.
I 15 MR. EDELMAN: Dick has shown you some of these 16 slides before. He will go through those very quickly, and 17 , there are some new ones we should dwell on here.
18 ; DR. OKRENT: Okay.
19l MR. EDELMAN: So we will not skip a whole lot.
l I
20 1 DR. OKRENT: Don't skip anything that you think 21 is important.
22 MR. EDELMAN: Yes, sir.
23 (Slide.)
- 24 l, DR. OKRENT
- In a sense, what I mean is *he last i
25 j time it was an information meeting, this time we are I
ACE-FEDERAL REPORTERS, INC.
202 147 1 700 Nationside Coveraae 800-3 % 6646
1 l
1350 21 03 268
()DAVbur 1 reviewing it.
2 MR. HOLT: I will cover several topics. The 3 first is background information with respect to the 4 seismicity in the Perry plant, information with respect to 5 ene Leroy earthquake, which is a discussion of the township 6 in which it occurred, discussion of the response spectra, 7 and the conclusions.
8 This slide shows the Perry Nuclear Power Plant in 9 the center, designated as PNPP, a 50-mile and a 200-mile 10 radius around the plant site.
11 The plant was licensed under Appendix A in the 12 concept of a tectonic province, wherein the largest O
k' 13 earthquake in the tectonic province is assumed to occur at 14 the site.
15 Let me review for you those earthquakes.
16 The Attica earthquake of 1929 -- for sure I am 17 going to get the date -- 1929, magnitude 5.2. The Anna, 18, Ohio earthquake, magnitude slightly under 5. And 1
, 19 subsequent to the submission of the PSAR an earthquake which 20 occurred in Sharpsburg, Kentucky, magnitude 5.2.
21 The intensities carried for Attica, it was a very 22 localized area of Attica that had intensity 8.
23 There was the Anna, Ohio earthquake, is 7 to 8.
24 And the Sharpsburg, Kentucky is intensity 7.
25 DR. OKRENT: Has anyone drawn some kind of old ACE-FEDERAL REPORTERS, INC.
202-347-3700 NationwideC% a 800 336 6646
1350 21 04 l 269
/
(
.)DAVbur 1 fault line from Attica to Anna?
2 MR. HOLT: Has someone drawn a fault line from 3 Attica to Anna?
4 DR. POMEROY: Yes. Bullard in 1959 drew it all 5 the way from New Madrid.
6 MR. HOLT: I was going to say he certainly didn' t 7 stop at Anna, nor did he stop at Attica.
8 DR. POMEROY: Dick, before you go away from 9 tha t -- I am sorry to hit you on the first slide, but could 10 you just characterize for me quickly the scismicity within 11 the 50-mile radius? What is the oldest event that shows up 12 there in terms of a date? Do you know?
w/ 13, MR. HOLT: I will do that while I have the slide 14 that has the 50-mile radius. Perhaps that will be more 15 convenient. -
16 l (slide.)
i 17 f Information with respect to the Leroy earthquake, 18 l- which occurred on January 31st, 1986, it is approximately 19l 10.5 miles south of the plant site. I will show you a plan 20 map in a few minutes. It occurred about 4:47 -- I am 21 sorry -- 11:47 a.m.
22 Based on a U.S. model for location, the best 23 location is 41.650 and 81.162. I will show that again on a
'~; 24 , detailed map after this.
~-) l 25 ' Focal depth is approximately 5 kilometers. MB ACE-FEDERAL REPORTERS, INC.
WMG.%Kn W1 Coverage (W15W""5
1350 21 05 270
._,iDAVbur 1 magnitude of 4.9 from ten worldwide stations.
2 DR. MARK: What does that " constrained" mean on 3 your focal depth? With focal depth, you say 5 kilometers 4 constrained.
5 MR. HOLT: The constraint, the modeling with 6 respect to the focal depth, if you constrain the model and 7 fit the data to it in this particular case there was a much 8 greater fit. Otherwise, if you let the data run without 9 constraining it, you get a large variation with respect to 10 the focal depth.
11 In this particular case, if the earthquake were 12 constrained at 5 kilometers deep, it fit the data from
,o 13 several stations much better.
14 3 DR. MARK: Thank you.
i 15' (Slide.)
16 MP. HOLT: I repeat with respect to the 17 magnitude, is 4.9 MB. The modified Mercalli intensity is 18 i 6. I will show you some isoseismals with respect to that.
19 That was listed as preliminary when we gave the 20 designation of the Perry. I think at this point it is 21 pretty final. We have over 800 reports.
22 The questionnaires that were circulated returned 23 and reported maximum effects of a few damaged chimneys, 1
24 cracks in walls cracked and fallen plaster, broken windows, 25 and well water effects, the well water effect being ACE-FEDERAL REPORTERS, INC.
202-347-1700 Nationwide Coverage 800-336- 4 86
1350 21 06 271
()DAVbur 1 silting of water supply wells.
2 (Slide.)
3 Based on the intensity data, we constructed this 4 isoseismal map. In the middle of the map you see the main 5 shock shown on a square. That is the epicentral location 6 for the earthquake from instrumentation.
7 The isoseismal around the square is intensity 6.
8 There are small circles of lower or higher intensity, as 9 expected, through the region.
10 I might, one, say a very brief look at the 11 surficial geology of this area, the larger intensities do 12 seem to correlate to the softer or deeper overburden. That, f8 .
\ 13 however, will be checked.
14 (Slide.)
15 within 50 miles of the site, this plan is 16 actually an FSAR drawing with two additions.
17 The earthquake of 1983, January 22nd, I believe 18 the magnitude is about 3, and of course the earthquake of 19 January 31st, 1986. The intensities are shown on this 20 slide.
21 Previous -- in answer to your question, Paul --
22 previous to the occurrence of the 1986 earthquake, there was 23 a 1943 event approximately 12 miles west of the location of r3 24 the 1986 event, with a circle of confusion on it. That V
25 would be 12 miles. So there could be a coincidence.
ACE-FEDERAL REPORTERS, INC.
. - _ - . _ _ _ _ _ _ _ _ - - - - - - - - _ _ ---__.._ _ _ __ ----- E S C T-E B __ _____ Jt r I O Ss G n -----__-- E -- _ _ _ _ _
1350 21 07 272 JDAVbur 1 The magnitude of that event is 4.7.
2 DR. POMEROY: Just to clarify it for me, the 3 other events that you show there, the smaller events, is 4 that in 1858 or 1958 that I see above there?
5 MR. HOLT: This is 1858, 1943, 1948, and two in 6 1955.
7 DR. POMEROY: I can' t see the rest of the 8 numbers, but is there a significant amount of activity prior 9, to 1940 or so?
10 l MR. HOLT: I apologize. Let me go through the 11 dates for you.
12 1836, 1858, 1898, 1906, 1928. These are everts I :
x' 13l that cccurred in the vicinity of Cleveland. That was 14ll addressed apecifically, I might add, in the FSAR.
15 1885 and 1932, 1934, 1857, and of course the 1983 16 event.
17 i (Slide.)
18 With respect to conclusions in the 50-mile 19 circle, we have relatively low seismicity. We have no 20 capable faults. That is why the licensing procedure was a 21 tectonic province. The highest intensity is a modified l 22 Mercalli 6, and the highest magnitude is the 1986 event, 23 4.9.
3 24 (Slide.)
25 As mentioned in the introduction, a question ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-66*6
1350 21 08 l 273 l()DAVbur 1 arose with respect to some injection wells south of the 2 plant site, and on this plan view is shown the Perry Nuclear 3 Power Plant, again in the upper righthand side of the 4 slide -- the Calhio injection wells, which are approximately 5 2 kilometers deep and about 2600 feet apart.
6 The situation with respect to the injection and 7 the potential for causing carthquakes will be addressed by 8 Dr. Talwani after my presentation.
9 And the series of aftershocks with respect to the 10 1986 event -- there are 11 aftershocks at the present time.
11 The largest one is of magnitude 2.4, and the latest one 12 occurred last Saturday.
)
ss 13 The total distance range over which they occurred 14 is approximately 4 kilometers.
15 DR. POMEROY: And the approximate depth range?
16 MR. HOLT: The depth range, as I recall -- and 17 somebody in the audience I am sure will correct me if I am 18 wrong -- is 7 kilometers to about 2 kilometers.
19 ( S li,de . )
20 There are a number of people that deployed 21 instruments subsequent to the main shock. Without trying to 22 name them, because for sure I will leave someone out, but 23 there were a significant number.
<S 24 Shown on this slide is the background topography Q
25 on a scale of 1 to 250,000. The blue dots represent the ACE-FEDERAL REPORTERS, INC,
__ __ _ __ ._ ___ __. _ _ __202-347-3700_ __ _ ___ Nationwide C&CIp __ __ ___ _800 336 6646_____ _ __ __ ____ ___
1350 21 09 274 l
iDAVbur 1l instrument locations that we used for focal plane solution 2 and for the solution for our depth calculation.
3 The main shock, of course, is shown as a red 4 square in the middle of the slide, and if you have really 5 got good eyes, it shows the aftershock sequence, all of them 6 pretty much contained within the red square.
7 (Slide.)
8 The focal plane solution for the earthquake is 9 shown in two focal point solutions on the slide. The 1
10 l triangles are potential or pull-away. At any rate, the 11 things that look more like circles are compressional. The 12 best solution is a north 20 degrees, east with a right
'- J -
13; lateral motion, or 70 degrees, west with lef t lateral 14 motion.
15 We see very little thrusting or normal faulting 16 in this. It is mostly strike-clip.
17 DR. POMEROY: Is that consistent with other I
18 l solutions in this part of the country?
i 19 MR. HOLT: We don't have many solutions in this 20 part of the country. I would have to look it up. I really 21 don' t know.
22 (Slide.)
23 The safe shutdown earthquake for the purposes of
(~1 24 l licensing was characterized as modified Mercalli intensity Em j 25 7. Peak ground acceleration associated with that carthquake ACE-FEDERAL REPORTERS, INC.
202-147 3700 Nationwide Coverage 800 33MW6
e 1350 21 10 275 :
(_) DAVbur 1 was .15 g, and the regulatory guide 1.60 spectra.
2 Subsequent to the submission, NRC staff asked us 3 to do a site specific spectra with respect to the Perry 4 plant. They picked a mean magnitude of 5.3, plus or minus 5 half a magnitude, and as a subset to represent Anna they 6 suggested we look at a 5.5, MBLG plus or minus .3.
7 I might add that in the slides I am going to show 8 you I have the site specific spectra. The actual magnitude 9 that we achieved in computing these spectra I think was 10 5.5. When we got through with all of the components, the 11 average turned out to be higher than the 5.3.
12 (Slide.)
13 Shown on this slide is the design spectra, shown 14 as a solid line in the upper part of the graph, and 14 15 horizontal components that were selected to represent the 16 5.3 magnitude. As I said, it is 5.5 on rock foundation 17 conditions which are similar to Perry and, in addition, in a l
18 distance range from 0 to 20 kilometers, which we took as l 19 representative of the definition at the site in Appendix A.
20 (Slide.) !
21 The 50th and 84th percentile to that data set 22 looks like that.
23 (Slide.)
24 Again, a slide that was shown to you by Jay Chen,
}
25 although he plots frequency and I plot periods. So you have l ACE-FEDERAL REPORTERS, INC.
_ _ _ _ _ _ _ _ _ _ _ _ - - - - - - - - - - - - - - ----- _____f'MS-ER3 _ _____.-- Nm L___._ _ _ EEY""L _
1350 21 11 276
)
,DAVbur 1 got to think of it backwards from what he plotted.
2 The design structure again is shown in the solid 3 line. The earthquake of January 31, 1986, the Leroy 4 earthquake, is shown in the two horizontal components, and 5 as you can see, there is an excedance of the design spectra 6 in the 20 hertz, approximately 20 hertz.
7 DR. OKRENT: I am sorry. Could you mention that 8 20 kilometer bit again? My mind must have been wandering.
9 A moment ago you used the te rm "20 kilome ters. "
10 MR. HOLT: I did?
11 DR. MARK: It was on that curve, the slide before 12 this, with about 15 different spectra. You said they were 13 spaced out 0 to 20.
14 MR. HOLT: That is right.
15 To achieve the definition at the site, which is 16 required by Appendix A for a tectonic province, we selected 17 distances for earthquakes in the magnitude 5.3 average range 1 l
18 at 0 to 25 kilometers away, as representative of a maximum 19 intensity at the site.
20 l DR. OKRENT: Average then over 0 to 10 21 kilometers? l 22 MR. HOLT: No. We took all of the worldwide data 23 and set in strong motion. Out of that data set we selected 24 , only those earthquakes that were recorded on rock, only 25 those earthquakes that represented on the average a ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 80043MM6
i 1350 21 12 '277 Oo^ veer 1 eo nitea e 5.2 , ena o e1v eao e e erea2me kee i.aet e re in e 2 sacrt distance range of 0 to 20 kilometers.
3 DR. POMEROY: .Were any of taose in tae Eastern 4 United States?
5 MR. HOLT: Taey were not.
6 7
8 /
9 i 10 11 i
12 13 .
14 15 16 17 18 19 4 20 )
I !
21 22 23 O 25 s ,
1 1
ACE-FEDERAL REPORTERS, INC. I 202 347 3700 ._ NationwideCoverase____ _ ____ 800 336 6646 ___ . _ _ _
1350 22 01 278
(
- i. ,)DAVbur 1 DR. POMEROY: Can you clarify for me exactly the 2 period or the frequency at which the design basis earthquake 3 or the Reg Guide 1.60 number is achieved?
4 Everybody keeps talking about 20 hertz, but I am
.5 getting a little -- the peak is at 20 hertz, but where is 6 the actual point?
7 MR. HOLT: I think I looked at it a couple of 8 weeks ago, and I think the band is from something like 21 9 hertz to 24 hertz. It is in an error band.
10 You are talking about this large peak?
11 DR. TRIFUNAC: Where it crosses. Where does it 12 cross?
(_ 13 MR. HOLT: Or where does it cross the reg guide?
14 I am sorry.
15 VOICE: 12 to 14 hertz.
16 '9. POMEROY: And this isn't a question to you, 17 but jut .or my own i erest. We were talking earlier, I 18 [ think, where are ft quencies that people consider to be 19 significant in terms of engineering significance.
20 ( Laughte r. )
21 DR. POMEROY: That is a question to somebody 22 else. I am talking about the structural components.
23 MR. MICHELSON: There are pieces of mechanical 24 equipment at 10 hertz. There are some. It just depends on 25 the design of the support structure.
ACE-FEDERAL 'IPORTERS, INC.
202-347-3700 _ _ _ Nationwide Coverage , _ 800 336-6646 _ __ _ _ _ _ _ _
l
1350 22 02 279 s . DAVbur 1 DR. STEVENSON: One piece of data you should 2 know. The Ge rmans , for example, cut their frequency --
3 acceleration as a function of frequency at 16 hertz. Above 4 16 hertz they do not consider acceleration as a parameter to 5 be used to determine dynamic requirements.
6 MR. MICHELSON: Isn't that for structural 7 purposes?
8 DR. STEVENSON: No, it is for equipment as well.
9 MR. MICHELSON: For relay chatter and things of 10 that sort?
11 DR. STEVENSON: That is 16 hertz. They say 12 whatever the acceleration is at that point from the
(8 -
- I k# 13 earthquake becomes the des'ign acceleration for that plant. ,
14 They don't look at higher acceleration associated with pool 15 swell, pool dynamics, or airplane crashes. That is a 16 relatively recent change.
17 They effectively say above 16 hertz we no longer 18 use acceleration as a design parameter.
19 DR. POMEROY: Let me ask one more l l
20 nonseismological question.
21 I assume that if we exceeded the SSE 1 or 2 hertz 22 the utility would be very concerned?
23 MR. EDELMAN: Yes.
~
(N
\_/
24 DR. POMEROY: At 20 hertz we are not concerned, l
25 is that correct?
Ace FEDERAL REPORTERS, INC.
_ _mrmum avm _-
1350 22 03 280
) DAVbur 1 DR. CHEN: As we have just shown, based on the 2 comparison between the recorded and design spectra, that 3 exceeds 20 hertz.
4 You also mentioned even with excedence at 1 or 2 5 hertz was that significant. Based on the recent experience 6 data collected by the seismically qualified utility group,
'7 their conclusion is for ground G value equal to or less than 8 .303 G, this kind of earthquake was not challenging the 9 plant equipment no matter whether it is I hertz or 20 10 hertz.
11 DR. POMEROY: Thank you.
12 DR. OKRENT: One grain of salt with that
(
k- statement.
13 )
14 MR. ANDERSON: Newton Anderson, of the staff.
15 I don't think the SOUAT group concluded that 16 everything was good. They said the structural integrity is 17 , okay, but they didn't say anything about relay chatter qr 18l equipment operability.
19 MR. MICHELSON: They only looked at selected kinds of equipment, too.
20 l 21 MR. ANDERSON: The range of equipment has been l
l 22i expanded considerably. l l
23 MR. MICHELSON: Are they looking at heating and
( 24 ventilating fans, for instance?
25 MR. ANDERSON: Yes, sir. j ACE-FEDERAL REPORTERS, INC.
D vtrage 800-336-6646 _
1350 22 04 281 m
Q( DAVbur 1 MR. MICHELSON: Where do you think they start 2 worrying about those?
3 MR. ANDERSON: We haven't gotten to the final 4 answer on that yet. They are under consideration.
5 (Slide.)
6 MR. HOLT: The slide that is up on the screen now 7 shows the 14 components along with the Leroy earthquake of 8 131 averaged in. As you can see, when you average it in 9 with the 14 components you have a slight excedance.
10 DR. TRIFUNAC: Could you put the previous slide 11 on that has the Reg Guide 1.60 on it?
12 (Slide.)
x 13 This earthquake was 10 miles south?
14 MR. HOLT: Right.
15 DR. TRIFUNAC: It was magnitude what? 5?
16 MR. HOLT: 5.
17 DR. TRIFUNAC: If this earthquake had magnitude 18 l 5.3 and happened in exactly the same place with exactly the j 19 same mechanism and the same sort of stress, where would you 20 put the spectra, up or down and by how much?
21 MR. HOLT: That is a good question. 5.3 again, I 22 would average with an average data set if I could get it 23 because particularly earthquakes, small earthquakes, the FI
(~T 24 earthquakes are going to have a very wide range with respect i V
25 to peak acceleration and with respect to frequency, and I l
ACE-FEDERAL REPORTERS, INC. ,
l 202-34?-3700 Nationwide Coveragg 80(L6 6(H6
1350 22 05 282 r's (_)DAVbur 1 think that the answer would be that I would not necessarily 2 scale up the high frecuency range. As a matter of fact, 3 subsequent to this slide, I will also show you one from New 4 Brunswick that was also recorded on rock and has a high 5 frequency. 6 DR. TRIFUNAC: I will repeat my question. I said 7 here is a spectrum recorded, and all of a audden there is an 8 earthquake in the same place, the same focus, the same 9 magnitude. It just doesn' t happen to be magnitude: 5, but 10 5.3. 11 Do you use the standard seismological procedures 12 which you have used many, many times in all these 13 calculations? Where would you put this? . 14 MR. HOLT: I don' t know what you mean by standard 15 seismological procedures. We are looking at the present 16 time at one component, and you are asking me to predict what 17 the next component would look like if we increased the 18 earthquake to a magnitude of 5.3 and kept everything else 19 the same. 20 Unfortunately, nothing stays the same when the 21 next earthquake happens, and if everything stayed the same, clearly the spectra would probably go up slightly. 22 l 23 DR. TRIFUNAC: By how much? 24 MR. HOLT: Slightly. It is very difficult. 25 DR. TRIFUNAC: Excuse me, Mr. Holt. How can you ACE-FEDERAL REPORTERS, INC. M 3700 NationwideM 800 3C
1350 22 06 283 r~8 l (-)DAVbur 1 calculate percentages and what these curves on the diagram ! 2 and everything is in terms of slightly? 3 I am asking you a simple question, a very simple 4 question. Can you give me an answer? 5 MR. HOLT: Are you asking me to predict one 6 component of motion, or are you anking me to look at a 7 number of 5.3 earthquakes that have occurred at that site 8 and to average all of those components? 9 Those are quite different questions. 10 DR. TRIFUNAC: I will ask a third time. 11 MR. HOLT: I feel comfortable with the second. I 12 do not feel comfortable trying to predict one component or A J 13 two horizontal components of motion from one earthquake. 14 That is the reason that we averaged. 15 DR. TRIFUNAC: I will ask you the question. The 16 earthquake occurs at the same place with the same mechanism, 17 with the same geology, st the same site, and i,t is recorded 18 by the same instrument and by the same component of that 19 instrument, for that matter. 20 So everything is the same. The only thing that 21 is different is that it is magnitude 5.3, not 5.0. 22 MR. HOLT: When you say the 5.3, you are saying 23 that things are different. Because I have a different fault {} 24 , line failure, I perhaps have a different depth to increase 25 the size of the earthquake. I have changed a whole number ACE-FEDERAL REPORTERS, INC. 202.347-1700 Nattanwide Coverase M336 "I'
1350 22 07 284 ())DAVbur 1 of things to get that to 5.3. 2 DR. OKRENT: Look, Mr. Holt, don't bend so much. 3 Why don't you say, okay, taking your assumptions, this is 4 what I would get and let's be done with it? 5 Because you have already made the point that an 6 earthquake is going to be different from one to another. 7 But he is trying to say, suppose in fact that it 8 is as similar as it can be within the limitations that one 9 has a slightly larger magnitude, okay? 10 You know, because we are just using valuable 11 time. 12 MR. HOLT: I already answered the question. I 13 said it would increase if I used an average value. 'He asked 14 me how much it would increase. I could increase it by the 15 amount that I show when I average the value -- 16 (Slide.) 17 -- which is some 5 or 10 percent. ! l 18 DR. TRIFUNAC: What is the antilogarithm of .37 19 1 MR. HOLT: Whatever the 10 to the 5.3 versus 5.0 20 is. If you are talking about energy output of the 21 earthquake? 22 DR. TRIFUNAC: No, I am asking about what is the l 23 antilogarithm of .3. 24 MR. HOLT: I don' t know. 25 DR. TRIFUNAC: Slightly bigger than 1, right? i ACE-FEDERAL REPORTERS, INC. 833479Ei3 _W M
I l l 1350 22 08 285 (m,DAVbur
) 1 DR. REMICK: Dr. Trifunac, I am having a little 2 bit of a problem. This is not a candicacy exam. If you 3 feel that the answer you have heard is not what you think it 4 should be, tell the committee what you think the answer 5 should be.
6 DR. TRIFUNAC: The committee knows what the 7 antilogarithm is of .3. 8 DR. REMICK: No, not that question. 9 (Laughter.) 10 DR. REMICK: But really, if you feel that the 11 answer -- I am sure curious to know as to what you think the 12 answer should be. O' 13 MR. MICHELSON: What is the point leading to? 14 DR. TRIFUNAC: The point is that the magnitude 15 measured in the period range, which is around 1, 2, 3, 4 I 16 seconds, is much more stable in acceleration, so it is not 17 subject to all this, and the difference between magnitude 18 5.3 and 5.0 is the antilogarithm of .3, which is about a 19 factor of 2, and the whole thing goes up by a factor of 2. 20 I don' t see why it is so difficult to say that. 21 I was hoping that Mr. Holt would say that. 22 DR. REMICK: Obviously, he is not going to. 23 , MR. MICHELSON: What is this eventually leading 24 to? So it is a factor of 2. So you are really worried
}
25 about an even larger earthquake, I assume? ACE-FEDERAL REPORTERS, INC.
. __ _____ S%PT1R __ _ _ _____N u , - t_____________TMfr^t*1_ _ _
1350 22 09 286 ()DAVbur 1 DR. TRIFUNAC: What will happen at the site i'f 2 you have an earthquake within this distance range, 5 to 5.3? 3 Here is an experiment. 5.0 is not so different 4 from 5.3, and it is at 10 kilometers, which is between 0 and 5 20 miles, the range that they have looked at. It is a 6 perfect calibration point, a curve exactly in the right 7 region, the right distance range, the right depth range, the 8 same geology, just the right propagation path. 9 DR. OKRENT: Let me put it this way. I know that 10 as part of their seismic consideration the French have 11 decided to consider what I will call not very strong l l 12 magnitude 5, 5.5, shallow earthquakes, in trying to decide 13 what spectrum should be used in evaluating a plant, and I 14 have to assume that in a sense Trifunac's question is aiming 15 the same way. , 16 If we have a 5.3 or a 5.5, might it not give you 17 two or three times more of the component in this area? 18 And the next question is: even if it did, is it ! 19 tolerable, or do you just shrug your shoulders and say you 20 don' t care how much you get in this area? l 21 And I don't know whether we have had an answer. 22 By here, I mean the frequency range. l 23 MR. MICHELSON: Really a frequency range above 24 about 12. 25 DR. OKRENT: In other words, suppose it were l l l i ACE-FEDERAL REPORTERS, INC. EBOT 3700 _ _ _ __ ___ NW _ _ __ _ _ GDWX"3 _ . _ _ _ . _ _ _ _ _
1350 22 10 287 i o)DAVbur s_ 1 about twice or three times what they got. How bad would 2 that be? 3 DR. TRIFUNAC: This thing times 2. 4 DR. OKRENT: Is that in effect partly where you 5 are aiming? 6 DR. TRIFUNAC: It is obvious I am not aiming very 7 well. 6 DR. OKRENT: All right. I don't know that we 9 have got the answer. 10 MR. MICHELSON: I don' t know if that helps me s 11 any. 12 MR. HOLT: I am at a loss for the question at 13 this point. 14 DR. OKRENT: Why don' t you go on? 1 15 It is not a question that relates to whether the 16 earthquake which occurs has damaged the plant. It is 17 related, in a sense, to the question that I raised at the 18 beginning: 19 Are there adequate margins in the design to deal 20 with events beyond the design basis? 21 It so happens this particular one would focus on 22 the high frequency. , 23 MR. HOLT: That is right, Dr. Okrent. 24 (Slide.) 25 Speaking to the question, as I mentioned earlier, ) I ACE-FEDERAL REPORTERS, INC. ' __ _ ._ _ _ .______ ____ __ _ ____ ___ f3f!PER3- _ _ _ _ _ _ 3t2~IOCw 1_ _ . CEEY"D - _ _ __ __ _ _. _ _ _ __. _b
l 1350 22 11 288 DAVbur 1 there is an order of magnitude scatter in the acceleration 2 range, in particular with 'espect to any given size 3 earthquake, and in a frequency range where I am somewhat 4 familiar with, say blasting effects, lasting effects, if I 5 repeated the blast at the same spot at the same location at 6 the same distance, the same energy, I would still have an 7 order of magnitude scatter with respect to the resultant 8 acceleration. 9 There must have been a reason that I reput this 10 slide back up. I think what I forgot to say is that the 11 84th percentile of the Perry spectrum is shown in dash and 12 the 84th percentile -- 50th percentile with the January
,im
')
13 event added in is shown in solid. 14 . MR. EDELMAN: Dr. Okrent, I think I would like 15 Jack Chen -- if he would double that input on the 16 l antilogarithm -- if we would double the part, I think Jack
. 17 l Chen would like to answer what would that impact be in our 18 rtructure, seeing the same frequency distribution that you 19 have here.
20 DR. TRIFUNAC: The question was purely 21 seismological. 22 MR. EDELMAN: Okay, but the impact of the 23 seismological one is what would it do to the structure of l' 24 ; the plant. 25 DR. CHEN: I can give you some slides. ACE-FEDERAL REPORTERS, INC. i 202 147-1700 Nationwide C.werage R00-11M484
1350 22 12 289 ()DAVbur 1 DR. OKRENT: By the way, it is the safe 2 shutdown. Don' t use the word " structures," first, in some f 3 constrained way, but we are really interested in safe 4 shutdown. 5 MR. EDELMAN: Yes. Can you answer that 6 succinctly? 7 DR. CHEN: Can I show you some viewgraphs? 8 ( Laughte r. ) 9 DR. CHEN: We did some preliminary analyses to i 10 find out what would happen if you increased the magnitude of 11 the acceleration of the recorded spectra and the extended 12 duration of the recorded time history.
\J 13 Again, here all we did is purely based on elastic 14 analysis, which is a very constructive approach.
15 so by simultaneously increasing the amplitude of 16 the peak by 30 percent of all three components and extending 17 the strong motion part of the recorded time from less than 18 one second to two seconds, and we repeat the elastic spectra 19 calculations. 20 What we found was, in essence, there is no change ; 1 21 in the spectra in the 20 hertz region. What it did was 22 increase the amplitude of the spectra in the lower frequency 23 region, by 4 hertz to 6 hertz in that region, as we show (~N 24 earlier. The recorded spectrum was enveloped by a large L] 25 margin by the design spectrum to begin with. I ACE FEDERAL REPORTERS, INC. rmpaara - N _ _ _ _ __ _ _ __ _ _. m
1350'22 13- 290 ()DAVbur 1 So the bottom line here is based on w' hat we have 2 seen so far. 3 If you increase the peak acceleration by 30 4 percent simultaneously, double your duration, there is 5 really no effect. It doesn' t change our conclusions. 6 MR. EDELMAN: That is specifically for the 7 spectra. 8I DR. OKRENT: I don' t want to spend any more time, 9 but I have a feeling he answered a different question. 10 Go ahead. 11 (Slide.) 12 MR. HOLT: The next event is from the New 13 Brunswick series of earthquakes that occurred in 1982, this 14 particular one in March. l t 15 It was on rock. It is the Mitchell Lake Road i 16 site, for those familiar with New Brunswick. The magnitude
. 17 is 5, and I believe the distance is 3 to 4 kilometers.
18 The two horizontal components are shown here with 19 the peak shown at about 30 hertz. 20 (Slide.) 21 If we would average those two components -- I am 22 sorry -- those four components, now, the two from New 23 Brunswick and the two that were recorded on the foundation : 24 map at the Perry plan, into the site specific spectra, this 25 is what it would look like. Again, the dashed line shows ACE-FEDERAL REPORTERS, INC.
- __- _ _ __iDCT@ G________ M L_____ -_ CDCCD _ _ _ _ __ _ _ - _ _ __ - -
1350 22'14 291 DAVbur 1 the input. The percentile has the site specific offect, and 2 the solid line shows the addition of the four components. 3 1 4 5 6 7 J J 8 ; 9 J 10 11 12 ) , 13 14
- 15 1
16 e 17 18 1 i 19 . t 20 i 21 I 22 23 l 25 1-i ACE FEDERAL REPORTERS, INC. _.__-_-_-__-______m_______-_m______m
l 1350 23 01 292
)DAVbw 1 (Slide.)
2 In conclusion, the tectonic apprach is valid, 3 certainly, from what we hae looked at, geologically. We 4 have looked at the epicentral area. We have examined a 5 number of photos that were taken since then. We still have 6 some more to do as confirmatory work. We certainly have not 7 seen any capable faulting. We have not, to this date, seen 8 any tectonic structures which would leave the tectonic 9 province approach valid. Safe shutdown earthquake intensity 10 is a modified 7. The design intensity versus the 11 site-specific response spectra of the SSE were 5.5 and 5.3. 12 As mentioned, the excedence of the safe shutdown earthquake w 13 was short duration, less than half a second at the 20 hertz 14 range. The high frequency and the excedence, of course, 15 above the 84th percentile, if you are compositing events, is 16 not unexpected.
, 17 (Slide.)
18 The last slides, I'll go through very quickly. 19 This basically is confirmatory information with respect to 20 the Perry plant intensity evaluation, initial activities and 21 programs and studies that are going on are shown. If you 22 organize -- I won' t read though this -- basically, a great 23 deal of this work was done in earthquake instrumentation. 24 The aftershock data and the locations were seen with respect 25 to ground motion. We are looking at the strong motions 1 l ACE. FEDERAL REPORTERS, INC. ;
--_---------EBCT E D-- N .CMDCD l
1350 23 02 293 ()DAVbw 1 records, worldwide, again. 2 (Slide.) 3 I showed you the ones from the eastern United 4 States, namely, New Brunswick and Leroy. 5 With regard to the geology, we have looked at 6 geological anomalies that were cited in the FSAR, PSAR in 7 epicentral areas. We have responded to NRC questions on 8 geology. At the present time, we are reviewing the 9 geological data in the literature since 1978, which is the 10 FSAR, reviewing remote sensing data, side-looking radar, 4 11 high-level photography, low-level photography, LANDSAT, 12 mapping geology in the epicentral area, to determine whether 13 or not there are a number of gas logs that have been taken 14 since 1978. I think, some 1000 of th.em. 15 We are looking at a number of the factors listed 16 on there -- right now I won' t bore you with reading them -- 17 as confirmatory data. As mentioned previously, there was a 18 question with regard to the potential for the injection 19 wells of Calohio, causing the main shock and the 20 aftershocks. That will be addressed by Professor Talwani. 21 DR. OKRENT: Just let me see. Are there 22 questions for Mr. Holt at this time? 23 DR. POMEROY Dick, would you agree with the 24 statement that Staff made that it is not unusual in an 25 earthquake to have a high amplitide, high frequency peak , ACE FEDERAL REPORTERS, INC. f~Vmffa-----____- N ii t _ -----__ e
1350 23 03 294 ()DAVbw I accelerations with limited duration? 2 MR. HOLT: I would agree with that. 3 DR. POMEROY: Then they make another statement 4 that says something like the presence of high frequency 5 ground motions is more likely a local site phenomenon than a 6 source condition. < 7 MR. HOLT: I think I would like to distinguish on , 8 that question. The high frequency components are a measure 9 of function of three things -- the source, the propagation 10 path and the site. 11 DR. POMEROY: Do you know of any eastern 12 U.S. earthquake were high frequencies have not been 13 observed, where there have been strong motion recordings in 14 this distance range, where high frequencies have not been 15 observed? , 16 MR. HOLT: The ones from New Hampshire that were 17 on deeper soil did not have high frequency content and the 18 number of earthquake aftershocks that were recorded in New 19 Brunswick were all in very solid glacial till, which has 20 shear wave velocity on the order of 2000 feet a second. It 21 is more rocklike than soil like. And to the best of my I knowledge, I had all of them, high frequency components. 22 l 23 High frequency motion, I'm sorry. {} 24 25 DR. POMEROY: MR. HOLT: Are you familiar with Arkansas? I am not familiar with Arkansas. ACE. FEDERAL REPORTERS, INC. ___ _ __ _ _ ___r vvnw~t- __ _ __ m __ m
1350 23 04 295 O (_j DAVbw 1 DR. OKRENT: Okay. 2 Before we get to the next talk, is there anyone 3 -- do you want to comment here? 4 DR. RIGHTER: I want to point out that the Staff 5 would differentiate between earthquakes recorded at Perry 6 and earthquakes recorded elsewhere and would not amalgamate 7 them in one set and treat them the same, statistically. 8 Similar, we would not characterize this ground motion as 9 " expected." 1 10 DR. OKRENT: With regard to the question of 11 whether deep injection was a cause, first, let me ascertain, 12 is there anyone in the room who thinks that there is a O 13 likely connection between the deep injection and this 14 earthquake? 15 DR. POMEROY: I intend to investigate that 16 question. 17 VOICE: What probability would you say? 18 MR. RIGHTER: That is the perfect question. 19 DR. OKRENT: Dr. Talwani. Do we need d ust a 20 one-minute summary or not, or .ad you better give your 21 presentation? 22 DR. TALWANI: I have reviewed some of the 23 available data on the calchio injection wells and the {} 24 current seismicity of the area. I then compared these with 25 the other documentary cases of well injection, which , l ace FEDERAL REPORTERS, INC. , ___- _ _ _ __ - _ _ __ _ -_-____-_-_r n M ui n ___________ m ___-_-__--_ m _______-_-_-__
1350 23 05 296 r~) ( / DAVbw 1 inducted seismicity, and reached some conclusions. 2 (Slide.) i 3 Again, this is the location of the two wells. 4 They are located about 11 kilometers from the epicenter. 5 These are the various aftershocks. A point to note is that
~
6 even after instrumentation, there has been no seismicity 7 observed from the vicinity of the wells or along this path. 8 DR. POMEROY: Just for information, could you 9 show us where the solution mining takes place, the salt 10 solution mining?' 11 DR. TALWANI: I am not sure. 12 VOICE: That occurs over where it says Grand O 13 River, and from that point over along towards the site along 14 the lake. Approximately three or four miles, I would say, 15 from Grand River over to the site. So it is not in 16 between. The solution salt mining does occur in the F Unit 17 of the saline, which is 2000 to 2200 feet below the surface i 18 versus this injection well, which is down 5000 to 6000 19 feet. , 20 DR. POMEROY: How long as the solution mining 21 been going on in the area? 22 VOICE: Solution mining started on at the turn 23 of the century and continued until about 10 years ago, then 24 stopped. 25 DR. POMEROY: I presume there were larger t ACE FEDERAL REPORTERS, INC. < _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ ._______________________N_______m_____m_____
l l l .1350 23 06 297 i ( DAVbw 1 injections at greater pressures during the later years. 2 Is that presumption fair? l 3 VOICE: I don' t have that information. 4 DR. POMEROY: Thank you. l 5 DR. TALWANI Next, I am going to show you a 6 cross-section along this line. 7 (Slide.) 8 Again, this is the plant and the wells. A 9 closeup of this area is shown here. The water is injected 10 in the Maynardville formation. Also the Mt. Simon, which 11 lies above the relatively impervious Precambrian basement, 12 and this is on the same horizontal and vertical scale. This (:) 13 is the location of'the main shock. This is the aftersho,ck. ! 14 The point of the slide is that the water is all l 15 pumped in right about here at a depth of about 1.8 16 kilometers, about 3 kilometers or so above the main shock 17 and about 11 kilometers. And the injection in well No. 1 18 has been going on since '71 and in well No. 2, since about 19 '74 and about '82. The earthquake occurred in '86. ' 20 (Slide.) ! l l 21 To summarize some of the points, there is no 22 known seismicity near the wells, before or after January 31, 23 '86. In most other cases where seismicity has been observed {} 24 in the vicinity of wells, there is always a whole bunch of earthquakes right in the vicinity of the wells. 25 Even after ACE FEDERAL REPORTERS, INC. _-_--___-___--_ m __---_ m _ _ M ___
1350 23 07 298 (m_)DAVbw 1 the earthquake there was no seismicity observed. The 2 earthquakes were about 11 kilometers south of the wells, and 3 no seismicity has been seen between them. 4 In other cases where seismicity has been 5 associated with injection wells, it has been propagated 6 along faults or fault zones and leaves regular traces in the 7 zone of seismicity, starting at the wells and propagating 8 out or in the vicinity of the well and propagating out. 9 Here there seems to be no seismicity in the vicinity of the 10 well, on which we know the epicenter is. And there has 11 been -- as Dick showed a little while ago, historically, 12 there has been low level seismic activity in this general O- 13 area. - 14 Another point is that the time lags observed 15 between injection and seismicity at other known locations is 16 in terms of days or weeks, and here if we can take the '82 17 injection as the causative one, it has been almost four 18 years. And if you include '74, almost 12 years before the 19 earthquake. 20 DR. REMICK: Did you see observed time lags 21 between injection? You mean the initiation of injection? 22 DR. TALWANI Be' n the starting of injection 23 and the initiation of seismicity. 24 DR. REMICK: But they are still injecting water (]} 25 into the wells currently? ACE FEDERAL REPORTERS, INC. r,am m. < - - - . - - _ _
1350 23 08 299 ()DAVbw 1 DR. TALWANI: I believe they are. 2 DR. POMEROY: Would you say that statement is 3 true.for the Rocky Mountain Arsenal well? 4 DR. TALWANI: Yes, there was a delay between the 5 start-up of seismicity and the injection. 6 DR. POMEROY: Was that on the order of days and 7 weeks? 8 DR. TALWANI: Let me check. 9 (A pause.) 10 DR. TALWANI It says a few days to a few 11 months. 12 DR. POMEROY: Talking about induced seismicity 13 rather than seismicity associated with injection of water 14 wells, can you tell me what the total volume of water or 15 whatever has been injected here'is over that time period we 16 discussed? 17 DR. TALWANI I think the estimates I saw were, 18 over 200 million gallons of water were injected into to well 19 No.1 and over 80 million gallons of water injected into i 20 well No. 2.
- 21 DR. POMEROY
- So that is comparable to the Rocky
- 22 Mountain Arsenal situation?
23 DR. TALWANI: I believe it is less than that. I {} 24 am sorry. I take it back. I am not sure about Rocky 25 Mountain. < 1 ACE. FEDERAL REPORTERS, INC.
- - _ _ _ _ _ _ . _ _ _ - ______-____ _-____ r w in n --____ m .;__-__-__m___ _ _ _ _ _ _
1350 23 09 300 (A_) DAVbw 1 DR. POMEROY: In the case of the induced 2 seismicity associated with the Aswan Dam in Egypt, for 3 example, were there earthquakes between the water itself and 4 the point source of the seismicity? 5 DR. TALWANI: They were long faults, which were 6 more or less at right angles. 7 DR. POMEROY: Could I hypothesize in this case a 8 fault that existed in or above the Precambrian basement, 9 which was simply feed by the water from the injection well, 10 and the water from the injection well at the pressures that 11 it was injected at, was sufficient to trigger an earthquake 12 along that fault? O 13 DR. TALWANI: You can hypothesize. The only 14 catch against that being so is the complete lack of 15 seismicity along the path, even at present. 16 DR. POMEROY: I don't see where that is a 17 problem. I think a porous medium, which certainly you are 18 injecting into, and if the nearest fault is that far away, 19 the water is going to go to that fault, presumably, and that 20 is going to be a preferential point for any initiation of 21 future activities. 22 DR. TALWANI So you are presuming a very nice 23 clean availablo fault system that extends from the wells and ; 1 {} 24 25 goes about 11 kilometers. DR. POMEROY: No, I am assuming that there is a l l ACE. FEDERAL REPORTERS, INC. r~PiuTn m _ ___ m ._
1350 23 10 301 l DAVbw 1 fault that is 11 kilometers away, and you are injecting 2 material into a highly porous formation. I don't know what 3 it is, and that flow is simply taking place in that 4 formation until it happens to reach the nearest available 5 point where there is a release mechanism that is built in. 6 DR. TALWANI: I would expect to see some 7 seismicity. Normally, it requires some sort of a fJow, but 8 the question is, why would it just flow in one direction? 9 DR. POMEROY: I think it would flow in all 10 directions. I am emphasizing that there is one particular 11 fault. 12 The other point I would like to make, and I would 1 13l also like to ask, if possible, that you be around when the 14 Staff's consultants make their presentation, because I would 15 like to hear some discussion about the question of historic 16 seismicity activity in the area, before the wells. 17 It is my understanding, on Dick's slide, that in 18 the vicinity we are talking about, that the earthquakes 19 essentially all occurred after 1943 or so. 20 The question is, are there possible ways that the 21 solution might affect the stress concentrations in that 22 area, prior to the time of this injection? 23 DR. TALWANI And I think, if I remember right,
<~'T 24 there were some earthquakes before 1943.
25 DR. OKRENT: I have a pragmatic question. ACE FEDERAL REPORTERS, INC.
?n? w.1 m winns un envenn, am us 66u
r 1 I l 1350 23 11 302 [~)DAVbw (_, 1 Suppose there is a chance, which is not l
-2 negligible, that the deep well injection, in fact, caused l
l 3 this earthquake. l
- 4 DR. POMEROY
- What is its significance? It is l
5 reasonably straightforward. If there is a real connection 6 between the injection, such that it has modified the 7 conditions of the occurrence of earthquakes in that area, so l 8 that you could expect a change in the rate of activity or in l 9 the maximum size of the activity, either way. 10 DR. OKRENT: Now if I get -- let me postulate -- 11 one a year just like we had this year, but no worse. 12 Presumably, each time, Mr. Chen will come in and say, "No O 13 , significance." 14 (Laughter.) 15 After while, in fact, he would just report, "Same 16 as last time." 17 (Laughter.) l 18 Therefore, you have to suggest that they might be 19 significantly larger, large enough that they pose a threat. 20 Is that, in your opinion, a possibility? l 21 DR. POMEROY: If there is a direct connection -- 22 I don' t know the answer, but it's certainly a possibility at i 23 some level. l (} 24 DR. OKRENT: I am just trying to understand. 25 DR. POMEROY: I would like to hear the other side l ACE. FEDERAL REPORTERS, INC. ! L____ _ -----_ - __ ___ _ __ _ _ ___ ______ - r~vme;~t _ _ _ _P - e __ m _..
1350 23 12 303 , A Cl DAVbw 1 of this picture. 2 DR. OKRENT: Anything else? 3 MR. EDELMAN: I think when I put my start up, I 4 would like Dave Green from my company, who did a little bit 5 of research on the environmental testing, to give you a 6 30-second update on the test equipment. i 7 Dave? 8 MR. GREEN: I'd like to clarify the question that f was asked earlier with regard to components after seismic 10 and then operating in a harsh environment. Those devices 11 have been tested in a harsh environment, following the 12 seismic event, following the seismic testing. So that they 13 have demonstrated to the required spectrum to operate in a 1 14 harsh environment.. 15 DR. OKRENT: That helps. 16 j 17 18 19 20 l 21 i 22 2 23 , 24 25 l l ace. FEDERAL REPORTERS, INC.
---- r,m.m .m. m- - - - - - - - - m_
1350 24 01 304
) DAV/bc 1 (Slide.)
2 The word on the slide, and I guess I would modify 3 the first word, the plant structure certainly has been 4 unmodified by the earthquake. 5 We've tried to present today in our report to the 6 staff that there was no change to our conclusions on our 7 geology and seismology, that the design earthquake is l 8 bounded by the January '86 event, that this inclusion of 9 this recent event together with the other data would not 10 substantially change our design spectrum. 11 The seismic capability, it's capable of i 12 accommodating the earthquake as it did. We've used the 13 i words lots of times "short duration", "high frequency", " low 14 energy". We're going to continue in our confirmatory 15 program and will demonstrate that our equipment has l 16 sufficient margin, even including the high frequencies at th 17 20 hertz level. l l 18 With that, that concludec our remarks for the l 19 day. I would ask that either now or later, Dr. Okrent, that l 20 you give us some indication of the level of detail that you ( 21 wish us to cover tomorrow. ; l l 22 MR. MICHELSON: Will you explain for just a i l 23 moment, first, though, again, you're using this as a proof 24 test. 25 MR. EDCLMAN: I started out by saying that all it ACE. FEDERAL REPORTERS, INC. _ - _ _ _ _ - _ _ _ _ __ _ __--_ r m usn _ _ _ m m -_.
i t 1350 24 02 , 305 DAV/bc 1 proves is that what we measured from this earthquake. 2 MR. MICHELSON: That's all it proves, and this 3 was not a substantial challenge to the plant to begin with. 4 MR. EDELMAN: No. But it was the first time that 5 I'veknownth'atapre-optestshasahtuallybeenperformed 6 on an earthquake. 7 (Laughter.) 8 MR. EDELMAN: I agree with you, it is not a proof 9 test. The earthquake did not come anywhere near what the 10 platit ' is designed' for. I agree with you. , Wezwill eliminate 11 the word " proof". 12 MR. MICHELSON: I would hope so. O 13 DR. OKRENT: I assume you are going to stay here 14 to hear what the staff says, so we'll come back for question 15 for tomorrow. Let's let the staff get up. 16 MR. STEFANO: Can I just introduce these people
. 17 to you? 1 18 ' DR. SMITH: One more fast question.
19 DR. OKRENT: By all means. 20 DR. SMITH: Would you be committing to some i 21 cooperative efforts with industry and EPRI? 22 MR. EDELMAN: Oh, yes, we have done that already 23 in our earlier presentation. We're part of the Seismicity 24 Owners Group, and all our efforts are being factored into l 25 that. We've worked very closely with EPRI and their staff. ACE FEDERAL REPORTERS, INC. l _____ _ ___--_ e _-__ _ . m -- <---- ' m___
1350 24 03 306 i ()DAVfuc 1 EPRI was invited here today to our presentation. l 2 I believe Dr. Ewald is here. He'll address the issue with 3 Perry Data being accounted into the Owners Proposed Group on 4 the seismicity Work'done by EPRI. , 5 DR. OKRENT: That's tomorrow. 6 MR. EDELMAN: That's fine. But, Dr. Ewald and 7 Dr. Stepp are both here to address that issue. 8 MR. STEFANO: Okay, Dr. Okrent, the first NRC 9 staff presenter here will be Dr. Sobel. She will discuss 10 the findings that we have documented regarding the ~eology 11 and seismology and the objective of the confirmatory matters 12 we've concluded from that review.
.,]
13 DR. OKRENT: By the way, could you give us a . I 14 summary of what areas the other speakers will also, cover? 15 *MR. STEFANO: Yes. I said this before, but I'll - 16 tell it again. Dr. Lee, the other presenter, will discuss 17 the equipment qualification seismology in the structure 18 itself. We do have as backup Dr. Johnson. 19 DR. OKRENT: Are you going to comment in some way 20 as to whether the recent earthquake in any way requires an l 21 evaluation of the current design basis? l 22 MR. STEFANO: I think that's what Dr. Sobel is i 23 going to address, yes. 24 DR. OKRENTt Thank you. [ 25 MR. POMEROY: Could I just ask, is it ACE FEDERAL REPORTERS, INC. 1 . ___ _ _ _ _ _ _ m _ _ _ - _ _ m t______ _ m _-
1350 24 04 307 j DAV/bc 1 contemplated that your USGS advisers are going to give any 2 presentations? 3 MR. STEFANO: It is not contemplated that they
'4 would give any presentation. They are here, however, to 5 comment and answer any questions that you may have.
6 DR. OKRENT: From now on, you can pose questions 7 to NRC or USGS. Please. 8 DR. SOBEL: My name is Phyllis Sobel. I am going 9 to summarize the staff seismology evaluation of the 10 January 31st earthquake and discuss the staff's seismology 11 confirmatory items. 12 I won' t go into the Elrst figure in great detail ( :
13 because you've heard most of the information already from 14 the applicant. The epicenter shown here is the same one 15 shown on the applicant's slide. It's the latest USGS l 1
16 Epicenter. The maximum intensity for this event was a 17 modified Mercalli 6. 18 This event is perhaps rema,rkable in the number of 19 institutions that rushed to the epicentral area to record 20 the aftershocks. Among them were some of the NRC's 21 contractors, such as St. Louis University and Memphis State 22 and, of course, the USGS and Lamont.
]
1 23 So far, there have been about 10 epicenters, the { (~') 24 largest magnitude 2.4, only some of the largest aftershocks v 25 were felt. The aftershocks occur in a cluster of about one ACE-FEDERAL REPORTERS, INC. CQYgmac 800-33646M
1350 24 05 308 (~'N (_) DAV/bc 1 to six miles deep. The USGS, the staff's consultant, is 2 just beginning to assess the free field recordings from 3 those earthquakes. As we heard this morning, the implant 4 seismic instruments triggered, and the entire history shows 5 that ground motion was of short duration and high 6 frequency. 7 MR. POMEROY: Phyllis, is that your maximum 8 intensity? 9 DR. SOBEL: That's from NEIS. 10 MR. POMEROY: Do they have a plot of the 11 intensities? 12 DR. SOBEL: I understand the USGS has an 13 isoseismograph but I haven't seen it yet. 14 (Slide.) 15 This figure summarizes the staff's review from 16 the SER which was published in 1982. Most of the structures 17 at the site are founded on Devonian shale bedrock. There 18 were no capable faults found in the site region. During the 19 plant site excavations, faults were discovered in the 20 excavations and these faults were found to be not capable. 21 They're probably of glacial origin. 22 e And faults were also discovered in the intake and I 23 discharge tunnels. These were found to be not capable and 24 probably not techtonic in origin. {} 25 The Perry site lies in the central stable region ACE-FEDERAL REPORTERS, INC. 202-147 3700 Nauonwide Coverase 800-336-6(pl6 l
1350 24 06 309 O 't) DAV/bc 1 techtonic province. The largest earthquake that could not 2 be correlated with geologic structure was of a magnitude 5.3 3 event, similar to the Ohio earthquake of 1937. 4 So the size of this 1986 event was consistent 5 with what we had known from historic and instrumental 6 seismicity in this area. 7 As was explained by Dr. Holt, the site-specific 8 response spectrum for a nearby magnitude 5.3 event were 9 compared to the design spectrum. And the Perry SSE was 10 found acceptable. 11 (Slide.) 12 This figure summarizes the confirmatory seismic O 13 activities. As I mentioned, it looks like the aftershocks 14 occurred in a cluster. When the fault planes solutions for 15 this event are available, we'll be comparing the stress 16 directions inferred from those fault plane solutions with 17 the regional maximum stress direction. 13 Now, at present, it seems from the tentative 19 solutions the regional stress direction, which is 20 east /northeact, is approximately the same as that found for 21 the' composite all plant solution for the aftershocks. 22 We will also be reassessing the faults in the site area which I mentioned earlier believed to be induced 23 l 24 by Pleistocene glaciation. We'll be examining the effect of b) 25 the stress directions on these faults. ACE-FEDERAL REPORTERS, INC. . M M .1?nn Natwinmida Cmstune 200.116.66d6 l
1350 24 07 310 ()DAV/bc 1 The USGS and the applicant are both exploring the 2 possibility that the earthquakes were related to injection 3 of chemical wastes in two wells about seven miles north of 4 the earthquake. 5 Past experience with induced seismicity has shown 6 seismicity beginning near the wells and later spreading out 7 from the wellhead. However, in this case, no seismicity has 8' been detected and we don't know whether there were any 9 events. Some were cmaller than about magnitude 2, but no 10 seismicity had been detected near the wells. l'1 In addition, there had been previous seismicity 12 in the area, including the 1943 magnitude 4.5 event, that 13 was mentioned earlier by the applicant. 14 MR. MICHELSON: Can I ask a question at this 15 point? In the earlier presentations, it was postulated that 16 there seems to be a straight line between Anna and Attica, 17 which Perry seemed to be right on the line. 18 Then I think there was a statement, something 19 about indeed there is some kind of an extension of faults 20 , from the New Madrid area through there. 21 Could you shed any light on those comments? 22 DR. SOBEL: There is no known fault line 23 connecting Anna, Ohio with this region. 24 MR. MICHELSON: Maybe I misinterpreted what they (-)) 25 were saying. ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverase 900-33Md46
l 1 l 1350 24 08 311 /
.)DAV/bc 1 DR. SOBEL: The zone around the plant is one of 2 diffuse seismicity.
3 MR. POMEROY: In a 1959 paper, Bullard projected 4 a line. ( 5 MR. MICHELSON: It's just his imagination? 6 MR. POMEROY: Based on historical seismic 7 activity from the St. Lawrence region all the way through. 8 MR. MICHELSON: There seems to be something 9 because of the distribution of the events. 10 MR. POMEROY: Of the seismicity. But that's a 11 matter of subjective evaluation. 12 MR. MICHELSON: I was kind of mixed up. As far
,,3
,13 as you know, there was riothing?
14 DR. SOBEL: As far as we know, there's no fault 15 line. - 16 DR. OKRENT: Before we leave the point, and this 17 is a question addressed to the USGS as well as to the staff, 18 if you want to assure yourself that, in fact, there was not 19 some connection below the surface between the New Madrid 20 area, from the large earthquakes on the St. Lawrence, is 21 there some kind of research program profiling, or whatever, 22 that could give you such information? i 1 23 Is it possible that the profilir.g has already
/
24 been done for other reasons that would shed light on this
}
25 theory? It has ramifications no less than moving ACE-FEDERAL REPORTERS, INC.
,n,. w. rw un na. cn m., muuu
1350 24 09 312 (_) DAV/bc 1 Charleston. 2 Let me ask both the NRC staff and the USGS. 3 DR. RIGHTER: Dr. Okrent. Andrew Murphy of the 4 Department of Research is here, and I'm sure he can tell you 5 the large amount of money wo*ve expended investigating just 6 such a possibility in the northeast extension of the Madrid 7 zone. 8 I don't think there's anything definitive that's 9 come out of that. The thing seems to change in character 10 and it's very vague, but it is an ac"'.ve research program 11 that we're pursuing and using geophysical means, geological 12 means, recording data to see whether there is, how far the fi . 13 i Madrid zone extends. 14 . Maybe Dr. Murphy can expand on it. But I don't 15 think we found anything at this point that points any 16 flaming guns that would say that Perry, that we have this 17 connection between the New Madrid instant. 18 DR. OKRENT: What does the USGS currently think 19 is the kind of evidence that exists one way or the other on 20 this question? 21 MR. STEFANO: I don' t think they' re going to 22 answer. I 23 (Laughter.) 24 MR. STEFANO: Can we give an answer to that? (~)T
\_
25 MR. WESSON: I'm Rod Wesson from the USGS. ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nauanwule Coverase B(D336 6646 . .
1350 24 10 313
- DAV/bc 1 DR. OKRENT: You should say your specialty is 2 deep well injection.
3 (Laughter.) 4 MR. WESSON: That's what I came more prepared to 5 address. This is, of course, an important generic question 6 and one that is being thought about at some level. But I 7 don't really have anything to add to what Dr. Righter said. 8 If your question is what could be done, what more 9 could ce done to try and answer the question, then, of 10 course, this kind of question requires long-term operation 11 of seismographic networks so that we can keep track of the
-, 12 small earthquakes that occur throughout Central America and i < '^'
13 requires a regional geophysical study such as deep crustal 14 , profiling and gravity and magnetic type surveys, because the i 15 clues are not in the surface geology. 16 If there's anything to be seen and anything 17 . there, it's in the mid to lower crust. 18 DR. OKRENT: Has the mid to lower crust been 19 looked at? 20 MR. WESSON: It's hard to see, Dr. Okrent. ) 21 (Laughter.) , 1 22 DR. OKRENT: And you say you live in Ohio. And I l l 23 used to live in Ohio and I never saw a bit of it. (~S 24 MR. WESSON: But one must use geophysical n 25 techniques such as recording small earthquakes and seismic
-ACE-FEDERAL REPORTERS, INC.
l -o,,.. , m m.~_4.rm.~. --
l i 1350 24 11 314
, )DAV/bc 1 reflection profiling, and so on.
2 And these are expensive kinds of techniques and 3 they have to be done on a regional scale; they can't be done 4 just looking at Perry, say, if you're asking the whole 5 question. You have to understand the broad regional 6 techtonics. 7 DR. OKRENT: I now live in California and, of 8 course, we have a reactor out there where it turns out 9 people were looking for other reasons and found things that 10 turned out to be relevant to our reactor. 11 I'm just wondering, have people been looking for 12 other reasons all along the St. Lawrence and the Great .\ 13 Lakes, in oil exploration, and so forth? 14 And is there a potential source of information 15 there that has not been tapped? I'm only trying to open a 16 dialogue. 17 MR. WESSON: With regard to the existing seismic 18 reflection data that's been obtained for oil and gas 19 prospecting, the basic problem is that, for example, in this 20 area near Perry, we have about 6,000 feet of Palaeozoic 21 rocks. Those constitute the targets for oil and gas 22 exploration. 23 So most of the data that's gathered is ('] v 24 essentially focused in that depth range. If we're going to 25 find, if there were such a lineament, my impression would be ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-3M-M ,
1350 24 12 315 tj DAV/bc 1 that the important evidence would be at greater depths of 2 several kilometers. 3 DR. OKRENT: Okay. 4 DR. SOBEL: Incidentally, a large part of the 5 research effort on the northeast extension of the New Madrid 6 zone has been the modeling of gravity magnetic data. To 7 date, they have not hypothesized a structure in the area of 8 the Perry site. 9 The final item on the confirmatory issues, which 10 staff believes is the most important, we're assessing the 11 free field ground motion recordings with respect to 12 worldwide data, and we're lookir.g for an answer to the ,o 13 , question: 14 What part of the high frequency excedence is 15 observed in the Perry site, and what is the transmission 16 path for the possible side effects? 17 MR. POMEROY: Phyllis, is this the place to ask 18 questions about injection? 19 DR. SOBEL: Yes. 20 MR. POMEROY: Can you or Rob Wesson, or somebody, 21 address the question of what the current state of their 22 opinion is with regard to the relationship between 23 earthquakes and injection? Not only the chemical wastes l) v 24 that are taking place at the Calohio site, but also any 25 solution mining effects and any effects of the possible 1 l ACE-FEDERAL REPORTERS, INC. 202 347'3700 Nationwide Coverage 8043 % 6646
1350 24 13 316 DAV/bc 1 secondary recovery. I don' t even know if it's done there, 2 with regard to the oil and gas. 3 And I understand there have been some 4 calculations of the stresses in the region as well. And I 5 wonder if we could hear those as well. 6 DR. SOBEL: I'll defer that to Robert Wesson of 4 7 the USGS. 8 9 10 11 12 13 . 14 15 16 17 18 19 20 21 22 23 24 25 l l ACE-FEDERAL REPORTERS, INC. '
,,,, o, ,,m m.. . 4. c.- - - , ~ . ~ l
11350 25 01 l 317 L_jDAVbur 1 (6:00 p.m.) 2 MR. WESSON: The direct answer to the question, 3 as Leon said, is there is no smoking gun implicating the 4 Calohio wells as contributing to earthquake. The question 5 of what is the likelihood that the well may have triggered 6 the earthquake, my personal feeling is it is on the order 7 of -- at this point from what we know, it is on the order of 8 1 in 5 to 1 in 3, or something like that. It is not 9 insignificant. 10 These are the reasons why that would be. If we 11 undarstood -- if we had a cluster of small earthquakes at 12 the bottom of the well, that would constitute a smoking gun, (, l 13 and I think this whole discpssion would be in a much 14 different light. - 15 Because the earthquakes are so far away from the 16 well, 10 kilometers or 11 kilometers, and we see no trail of 17 l seismicity back, it raises a lot of questions in your mind, i 18 l Now, the other way of asking a question about 19 could this earthquake be induced is to actually estimate or 20 measure the state of stress, say at the base of the 21 i! Paleozoic sediments, at the top of the Precambrian, and then 22 to in effect do a calculation of the effective stress and 23 the change in effective stress owing to the injection of the
' l ' _ '; 24 well. I ~/
25 In order to do that calculation, we need to know ACE-FEDERAL REPORTERS, INC. F 4 347-51700 Nationwide Coverage 800-3M
1350 25 02 318 d i L 'DAVbur 1 two components of horizontal stress. l 2 Okay, we know from the hydrofract measurements at 3 the time that the 1971 well was completed -- and I think 4 that is confirmed by the 1981 well -- we know the least 5 principle compression. We don' t know the maximum principlc 6 compression. We assume that it is at least the weight of 7 the overburden, okay? 8 If we assume that the maximum compression is at I 9 least the weight of the overburden, we calculate the stress 10 state at the o~ottom of the well, we are close to failure, 11 okay?
- r. 12 Now, if one were able to extrapolate from Western
~
13 New York and beyond the maximum compressive stress, then I i 14 think we would be close to failure. I think we would be l 15 l above failure at the bottom of the well. 16 But we really can't talk about failure at the 17 l bottom of the well. We have to talk about failure in the i 18 [ hypocentral area. 19 Okay, we are currently carrying out calculations 20 to try and determine what the change in pressure in the 21 hypocentral area was as a result of the 15 years pumping and 22 l 300 million gallons of fluid injected at Calchio. 23 If one assumes a radial flow model -- in other
; 24 words, if we have kind of an isotropic aquifer that is about 25 500 feet thick at the base of the Paleozoics, then after l
l l ACE-FEDERAL REPORTERS, INC. l 202 347-3700 Pat NTrwidefoverage fXibO3666J4
~
F~ l 1350 25 03 319 ( / DAVbur 1 15 years pumping at the rate that they pump, we have about a
- 2) five-bar stress, about a five-bar pressure increase at the 3 top of the Paleozoics -- excuse me, at the top of the 4 basement 10 or 11 kilometers away.
5 Now, what we are currently doing is making 6 assumptions along the lines of the ones you indicated. 7 Suppose there is a more confined channel of some sort that 8 would in effect direct the flow, say, along the fault zone. 9 Obviously, that will give a higher, larger increase in the 10 pressure change in that area. 11 So these various studies are underway. 12 What we need to do,.the most important thing, to b'" 13 try and sort this out, I think, is to get a better estimate 14 of the maximum principle or the maximum compressive stress, 15 and once we did that then we could understand the range of 16 pressures that would be required to induce earthquakes at 17 that zone. 18 Now, one of the other issues that is important is 19 the depths of the earthquakes. Currently, there is It 20 discussion within the community about the depths of the 21 earthquakes. They range from 2 to 7 kilometers. 22 Well, if they are 2 kilometers, then the question 23 of induced earthquakes are very important. If they are at 7 24 kilometers, then, you know, it becomes more -- the (~/h
\-
25 probability becomes more attenuated. So that is an ACE-FEDERAL REPORTERS, INC. _. - _ - _ - - _ _ -las RD-- _-_ a s:a n @ m m n- - - - - - - - - - - -c m e c m ----- -- ---- L
1350 25 04 320 ('/ Ls DAVbur 1 I important line of pursuit. l l i 2 Another issue is the time lag that Dr. Talwani 3 talked about is an arguable issue. On the one hand, one 4 would like in the best of all worlds to fit this model, 5 earthquakes propagating along zonos as the pressure front 6 moves out. 7 But it takes 10 or 11 kilometers away. It takes 8 almost 15 years to get up to the five bars that wo are 9 talking about. 10 Now, an additional problem is the question of 11 what was the level of seismicity of very small earthquakes 12 in the Perry or in the Calohio zone in the years preceding O 13 the 1986 earthquake. , 14 j Well, we don't have very good data about that. 15 The closest seismograph station is the one, as I understand 16 it, at John Carroll College. John Armbruster from Lamont 17 has taken the records of the larger aftershocks, the 18 magnitude 1.9 and 2.4 earthquakes, and gone through two 19 months preceding the January earthquake and examining to see 20 if there are any foreshocks. 21 Two things have been learned from that. First, 22 there are two suspicious events, one in late January and '
- l 23 another in December, that could at this time possibly be l
() 24 foreshocks. That remains to be confirmed. 25 But, secondly, it seems clear from his work that ACE-FEDERAL REPORTERS, INC. FBD37-DED SteEC@sGrJ---.______ CMOCD
1350 25 05 321
.s L_/DAVbur 1 the station at John Carroll is sufficiently insensitive that 2 its threshold of detection is probably just less than 3 magnitude 2.
4 So we could have -- it is possible that we could 5 have many earthquakes of the size of the smaller aftershocks 6 that could have been occurring in the months beforehand. 7 So to conclude, we are trying to do the best we 8 can of indicating whether the well can be implicated or not 9 or is responsible or not. At the present time it is not 10 obvious that there is going to be a clear-cut conclusion. 11 My personal feeling is that the probability is 12 not insignificant, but it is not overwhelming either.
~
(:) 13 i i Yes. 14 DR. MARK: If there were in fact any foreshocks, 15 doesn't that almost knock out the possibility that a big 16 event was due to :he well because the pressure you are 17 talking about would have been dissipated. 18 MR. WESSON: Well, in a way that gets back to the 19 question that Dr. Pomeroy posed earlier. As the pressure 20 front spreads out, in effect you get a larger and larger 21 fault length that is above a given failure criterion or 22 above a particular threshold. 23 So the trouble with induced seismicity ('} v 24 l earthquakes is that once you get it in sequence it is not 25 obvious what the maximum magnitude is going to be, and it ACE-FEDERAL REPORTERS, INC. 202 347 3700 ___ _ , Nationwide Coverage . , , _ _800-33M646 , _ . . , _ ___
1350 25 06 322
'DAVbur 1 is conceivable that the magnitudes could get larger through 2 the sequence.
3 This is observed sometimes with cases of 4 reservoirs that induce earthquakes, and to some extent it is 5 true with the Denver Arsenal case, where it started out with 6 small earthquakes and built up to magnitude 5, a couple of 7 magnitude 5's, and then when the well was closed and no 8 longer pumped and the earthquakes continued for several I 9! years, as the pressure front dissipated smaller earthquakes l 10 continued for several years. 11 DR. POMEROY: Are you also going to look, Ron,
-s 12 though, at the question of the other activities that are f )
Y 13 going on there? 14 , MR. WESSON: Conceivably. Of course, we know i 15 i that salt mining operations and the injection associated l I 16 l with that has caused earthquakes in western New York State r i 17 ' in the Attica Dale area. 18 l So it is conceivable that some of the pre-salt l 19 l mine conclusion, some of the activity before they closeo the I 20 salt mines could have been associated with that mining 21 activity. 22 We frankly don' t have a plan at the moment to 23 look at that, although certainly that is a legitimate i f,;
~
24 l question of pursuit. 25 I DR. TRIFUNAC: Has anything been calculated on ACE-FEDERAL REPORTERS, INC. 202-M9-1700 Wationwide Coverage 800-336-M
i 1350 25 07 323 (~S I L >DAVbur 1 the stress drop of this particular area? 2 MR. WESSON: That work is underway. I don't know 3 of a calculation at the moment for the stress drop at the 4 main shock. Work is also underway to calculate stress drops 5 from aftershocks. To say just a couple of words about the 6l 7 aftershocks, among the recorders that were put in the 8 epicentral area, there were some very wide dynamic range, 9 wide frequency band instruments. So it has been possible to 10 obtain very high quality digital recordings that have been 11 processed to date on about six of the aftershocks.
,- 12 The USGS group at Menlo Park is at this moment O 13 completing'an open file report that should be available by 14 the beginning of next week that will include the time 15 histories and the spectra for those.
16 It is fair to say that even the aftershocks are 17 characterized by a considerable enrichment in high 18 l frequencies as compared with similar sized earthquakes in 19 the Western U.S. They may be somewhat less high frequency 20 than New Brunswick, but there is a lot of high frequency 21 there. 22 DR. POMEROY: Would that be true, Ron, even for 23 the events that the USGS recorded in California on hard rock () 24 sites close to the source? 25 MR. WESSON: That is my understanding. ACE-FEDERAL REPORTERS, INC. F4237-3700 Nationwide Covi~n 800-336-6646
1350 25 08 324 . 'DAVbur 1 DR. OKRENT: Okay. 2 MR. STEFANO: Other questions of Phyllis? 3 (No response.) 4 MS. SOBEL: I am finished. 5 MR. STEFANO: Next we have Dr. Arnold Lee, who 6 will be presenting our findings on the equipment 7 qualifications equipment and also the confirmatory issues 8j related to that. 9 (Slide.) t 10 MR. LEE: My name is Arnold Lee, Staff Reviewer 11 of the Engineering Branch of NRR, Division of BWR
,_ 12 Licensing.
)
l 13 I
~
My presentation will cover structural sbismic 14 l design and seismic qualification. 15 First of all, structural seismic design. A 16 number of walkdowns have been conducted by NRC staff since 17 l the occurrence of the earthquake. On the days after the l 18 earthquake an NRR inspection team was sent to the site to 19 conduct a review and audit, including a review of 20 preliminary seismic recordings. 21 l A special safety inspection was also conducted by l 22 l NRC Region 3 staff on February 5th through 7th. This i 23 l included a post-earthquake walkdown and detailed regional ( , 24 , inspection of structures, pipings, HVAC, electrical and
/
l 25 structural systems, including anchor bolts and the bed ACE-FEDERAL REPORTERS, INC. I 202 3F-3700 Wationwide Co~erage . _ _ _ 8%33/>6M6
r-1350 25 09 325
,a V DAVbur 1 plate.-
2 A total of about 90 inspector hours were spent on 3 these activities. 4 In addition to this, the seismic qualification 5 review team, which consisted of a staff member and staff 6 consultant from each, also conducted a site audit on 7 February 6th, primarily to investigate the effect of an 8 earthquake on the safety-related equipment. 9 Questions have been raised regarding the adequacy 10 of the original reactor building dynamic model in predicting 11 high frequency response, and other related questions also 12 were raised regarding the similarity of the recorded 13 l foundation motion and the free field motions.
. 14 In the work performed by the staff consultant, 15 Dr. Jim Johnson, of Structural Mechanics Associates, the 16 utility's re.etor building dynamic model was first reviewed, 17 and then an independent eigenvalue analysis was performed 18 which verified the adegaacy of the original building 19 frequency characteristics obtained by the utility.
20 A good correlation of major structural response l 21 and the response calculated using the recorded foundation 22 motion and the fixed based structural model was obtained. 23 This confirms a lack of rocking motion, which is 24 an important part of soil structure interaction phenomena. (]) 25 This lack of soil structure interaction phenomena in turn ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Cm 8 % 33 M 616
1350 25 10 1 326 I
.c'DAVbur 1 leads to a conclusion that the recorded foundation motions 2 are similar to free field motion in frequency content of 3 about 20 hertz. Free field ground motion recently recorded 4 during one of the af tershocks has further confirmed this 5 high frequency counting at the 20 hertz.
6 With the fixed base dynamic model and the 7 recorded foundation accelerated time history as input, our 8j consultant has further calculated the response spectra at a i 9 higher elevation of reactor building and found out that it 10 has got similar amplifications as the measured response 11 spectra at a similar location.
-s 12 1 All of the major foundation response spectra, of
(' ) i * . 13 ' course as Dr. Johnson just stated a little earlier, is 14 subject to variation. 15 DR. SMITH: This last point is a little bit 16 different than what we heard earlier, where there was some 17 discussion as to reasons why the responses measured at a 18 I higher level in the building might not be valid. 19 I wonder if we could settle on this 20 interpretation. I think certainly I would desire seeing the 21 record closed and say, yes, we made these measurements, we 22 believe them, they are real, and it is a "no, never mind." 23 MR. LEE: To answer your question, sir, you have m 24 to understand that the recorded motion is a low energy m-25 earthquake. Because of that it produces very low stress, ACE-FEDERAL REPORTERS, INC. 202 M 7-F 00 N ainnwid ? Coverne RIV) .1Fyf646
P' 1350 25 11 327 rh l LJ DAVbur 1 and the corresponding pertinent value associated with it is 2 small. 3 So in the earliest stage of our investigation, we 4 did use 4 percent eigenvalue like the utility did, and we 5 didn' t find very good correlations. 6 So we changed it a little bit, and we reduced 7 that value, and we reached pretty good agreement, as 8 Dr. Johnson just stated. 9 DR. SMITH: Would one suspect that with low soil 10 structure interaction, high frequency motion, damping values 11 being low, that they are really not working, the structure 12 is not dissipating the energy; it is just like hitting it 13 with a hammer and having it ring? 14 MR. LEE: That is right. 15 DR. OKRENT: I am a little confused. It seems to 16 me what I heard was using the input from the measurement,
/
17 using in fact the same damping value on two out of three, 18 the calculations agreed with the measurements. This was 19 what the utility's analysis got. 20 Is this not what I heard? 21 DR. CHEN: Yes, sir. 22 DR. OKRENT: I agree you can change damping from 23 one direction to another, but I don't see that that is () 24 meaningful. That is not what you are talking about. 25 So what is it that you are doing that gives it a ACE-FEDERAL REPORTERS, INC. _______ - - _ _ ____ __ _ - . FMS-Jfd3 -_ ---Er* rP1 h -----_----.E WPX"M
i 1 1350 s 25 12 328 l' DAVbur 1 fit, and do you get a fit to all three concurrently; that 2 is, vertical, east / west, and north / south, with the same 3 physical assumptions? 4 MR. LEE: Dr. Okrent, I will let Dr. Johnson 5 answer that question. 6 7 8 9 10 11 12 13 14 15 16 17 18 19 l 1 20 21 22 l 23 l 24 25 ACE-FEDERAL REPORTERS, INC.
,m.m.m. m 1_m.c_ = m ,- ~
l l 1350 26 01 329 Ii /'IDAV/bc 1 DR. JOHNSON: I'm not sure exactly what point you 2 are asking, but let me respond to Dr. Okrent's point. The 3 first thing is that we calculated a response from the three 4 directions. As you recall, in one direction there's a CPA 5 and a foundation of about a 10th g. In the other direction 6 there's a ZPA of about 1.8 g's. 7 So, assuming that the vessel will behave somewhat 8 nearly elastically and we can uncouple the motion for 9 calculational purposes in those two directions, we probably 10 would induce stress along those two principal axis that are 11 different. So you could surmise that you have different 12 damping values, depending on which direction you're exciting .O . 13 it and what the amplitude is. 14 The calculations showed that, in just a first cut 15 of 4 percent damping, the east / west direction in vertical 16 had a reasonable comparison. North / south was underpredicted 17 by the calculations, and 4 percent damping in those modes. 18 So what I said earlier was a simple thing to do 19 , and as I'm going is to investigate the effect of different l 20 ' damping values for those roads in the north / south direction, 21 and calculate the response. 22 DR. TRIFUNAC: What physical mechanism would 23 explain the difference between the north / south and east / west 24 directions? The building is round. f' }) l 25 DR. OKRENT: Anyway, why would the damping be 1 ACE-FEDERAL REPORTERS, INC. F%347 3700 NatMwide m 800-336-6M6
1350 26 02 330 b) 1 'DAV/bc 1 less if the motion is greater? Which is what I understood. 2 DR. TRIFUNAC: That is okay for long periods of 3 time, but why should the north / south versus the east / west 4 dynamics have different diameters at such small levels of 5 response? 6 DR. JOHNSON: That's a good question. At small 7 levels of response. If we're talking about very low 8 stresses, which we are in fact, because of the stress levels 9 that Dr. Chen showed earlier, they are very low stresses. 10 That's a good question. It may be related to the 11 polar crane issue, but I don' t have an answer for that. 12 Maybe, Dr. Chen? Os ' 13 DR. CHEN: May I say a few words here? It seems 14 to' me I think the important thing we have overlooked was the 15 applicant did an analysis, and the NRC consultant also did 16 an independent analysis which verified the applicant's 17 analysis that two out of the three components are in very 18 good agreement with the recorded data. 19 And I think the most_important thing we should 20 not forget that is a very good verification of the model. 21 As to the third component, in both the 22 consultant's analysis and our analysis, it's somewhat 23 different from the recorded data. () 24 It seems to me it's not too important as far as 25 the Perry license is concerned. There may be another ACE-FEDERAL REPORTERS, INC.
- r~wasm . _ m __-. .m _ -
1350 26 03 331 (_ f)/DAV/bc 1 generic issue we can pursue on another occasion.- I don't 2 think we should dwell on this point too much. 3 DR. SMITH: I understood you to say earlier that 4 one of the differences in your analysis was using 2 percent 5 damping versus 4 percent used by Gilbert. 6 You came up with a similar amplification. 7 DR. JOHNSON: The structural analysis was done at 8 4 percent damping. The equipment response vectors were 9 generated at 2 percent damping. Both the applicant and our 10 first pass at it made those assumptions. I said that there 11 could be at least two possibilities of why for this first 12 pass we didn't see the kind of amplification in the ZPA that O 13 was observed in the measured response. 14 One was damping. Another was some possible 15 effect of rocking in one of the directions being 16 introduced. And that's not easy to discern from the data 17 that exists, since there's only three instruments. Or, one 18 set of instruments in three orthogonal directions on the 19 foundation. 20 I think, to repeat again, we can predict 21 reasonably well the ZPA up at that level. I don't think 22 that we'll be able to consistently predict closely the 23 spectral amplification in the 2 percent spectrum. That it's (~S, 24 , a lot more difficult to predict that amplification. s/ 1 l 25 DR. OKRENT: I think it's getting late. And ACE-FEDERAL REPORTERS, INC. . __ 202-347 3700 __ _ _ Nationwide com 800 336 6646
1350 26 04 332 1 DAV/bc 1l we're going to have to go on. We're about a half an hour 2 beyond our agenda. 3 (Slide.) 4 Lee, on the previous analyses I just mentioned, 5 we established the opinion that the original Perry reaccor 6 building model is indeed adequate in predicting high 7 frequency response. The high frequency, short duration 8! earthquake has an insignificant energy content to cause the l 9 damage to the structures. Based on the concept of a plastic 10 l response for the reactor buildings, the recorded motion 11 would need to be scaled by a factor of more than 2 in order
-, 12 to achieve the deformations corresponding to the design I
(_/ , 13 ! level of the forces. A quantitative assessment by the applicants has 14 l 15 l further indicated that the dynamic stresses in the i 16 l containment building induced by the recorded foundation time I 1 17 l history is well below the design value. I 18 l The next slide is the equipment and seismic i 19 qualifications. 20 (Slide.) 21 I have stated previously a number of plant 22 walkdowns have been conducted by the staf f, only to reveal 23 l no observable damage of any significance to the equipment I 24 ' itself, the supports and the mounting configurations. 25 The high frequency, short duration earthquake ACE FEDERAL REPORTERS, INC. 202147_m) Nationwide Coverse 800-116-6646
1350 26 05 , 333 o- 'DAV/bc 1 has indeed very insignificant energy content to cause damage 2 to the equipment. All of the 39 safety-related and the 36 3 nonsafety-related systems that were energized during the 4 earthquake have functioned as designed. 5 A quantitative assessment by the applicants of a 6 sample of equipment revealed so far has further indicated 7 that the original seismic qualification is adequate. 8 Included in the reviews are the equipment items 9 which were qualified originally by analysis, and those l 10 originally qualified by testing. For those qualified by 11 analysis, the recalculated stresses and forces are shown to
~~ 12 be well below the allowable.
( l 13 l For those items which were originally qualified 14 by testing, the test results factors have been shown to be I 15 f well above the response spectra corresponding to the i 16 recorded motion. 17 My conclusions -- 18 MR. MICHELSON: Are you satisfied that the 19 applicant looked at the right variety of components? Or did 20 l you just look at his list and judge those? 21 MR. LEE: Yes. For tha review that the staff 22 l conducted, yes, we are satisfied for the purpose of low I 23 power operation. o ' ( 24 MR. MICHELSON: Do you think there are no other 25 types of components? ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage R00-336 6646 _ _ _
l 1350 26 06 334 p h- DAV/bc 1 MR. LEE: There will be other types, as I will 2 show later. 3 (Slide.) 4 My conclusions, there is no significant safety 5 impact of the earthquake on the equipment and structures at 6 Perry that has been identified. The design basis 7 earthquake, although it was exceeded at a high, narrow 8 frequency region, the original plant seismic design was, 9 however, not affected. 10 The staff conclusions as stated in the previous 11 SER and its supplement regarding the adequacy of the seismic 12 qualification program and the structural seismic design 13 remains valid. - 14 My last slide will present some of the 15 confirmatory actions which are now taken by the applicants 16 in order to satisfy the long-term staf f requirements. 17 ' (Slide.) l 18 First, to perform an additional quantitative 19 assessment on the seismic qualification of a broader sample 20 of equipment types located in different buildings on various 21 elevations. 22 l Second, perform a generic evaluation of a high 23 frequency, short duration earthquake with regard to the I () 24 energy content and potential safety cignificance of 25 equipment and structures at Perry, and using this result to ACE-FEDERAL REPORTERS, INC. 202 347 1700 Nananmde Cmarane 800 3Eddd6 . .
~
1350 26 07 335 m,
, a- 'DAV/bc 1 assess the seismic capability of the Perry plant.
2 If another earthquake of similar characteristics 3 but with higher magnitude and somehow longer duration should 4 occur at the site, the applicant is progressing quite well 5 in this direction. I believe that a final and positive 6 confirmation can be expected. 7 I should also add that our staff cc,nsultant, 8 Dr. Johnson, is also pursuing along this line in 9 investigating the energy content aspect of the short 10 duration earthquake. 11 i MR. MICHELSON: What is the approximate schedule s 12 that you expect the applicant to get this work done on? ( ) . 13 l MR. LEE: About June. 14 MR. MICHELSON: Relative 'to power generation, 15 what's the schedula? 1 16 j MR. LEE: Before they exceed 5 percent power. 17 MR. MICHELSON: Thank you. l 18 MR. LEE: That concludes my presentation. , 19 DR. OKRENT: Any questions? 20 ' (No response.) l 21 (Slide.) l l 22 [ MR. STEFANO: Gentlemen, I will not ruminate on I 23 < all the conclusions. I think you've heard them several
,, (
( -
) 24 I times from the individual presenters today, and also by the 25 utility. I think the bottom line is that we think we've ACE-FEDERAL REPORTERS, INC.
w.24717po. - e .
l 1350 26 08 i 336
,N I i a o DAV/bc 1 gotten to the point in our evaluation to issue a low power 2 license for this plant.
3 We have identified a number of confirmatory 4 matters that need to be further pursued, which we will be 5 pursuing both in the building ourselves with the USGS and 6 the utility has already committed to do certain things which 7 are well-documented in the SSER we have issued. 8 The target date for all of this input is around 9 mid-summer this year. June, I believe, is the date that's 10 been spelled out. We will first have to look at that and 11 the results of that to determine whether it is permissible
, - .s 12 to go beyond that point in time.
L) ' 13 We certainly will have to convince the Commission 14 and yourselves of this as well. Right now, we are -- I'd i 15 ! just like to reiterate what I had asked this morning. We 16 would like to get a comment and agreement from this panel, 17 hopefully, tomorrow, so that we can at least license this 18 ! plant on the 14th, which is Friday. 19 We see no technical reason at this point why that i 20 cannot occur. 21 That concludes the NRC's staff presentation. Is 22 i there any further question? 23 (No response.) I I 24 DR. OKRENT: I don't see any problem. Let me ask 25 the consultants what principal points they would say ACE-FEDERAL REPORTERS, INC. 202-347-3?00 Nationwide Coverne 8001 % M -_
1350 26 09 337 A l JDAV/bc 1 should be called to the full committee's attention from 2 their own perspectives, if any. 3 Why don't we start with Dr. Smith. 4 DR. SMITH: What's the significance of bringing 5 something to the full committee? 6 DR. OKRENT: Well, in fact, the full committee is 7 going to hear the Perry earthquake situation tomorrow. It's 8 the full committee that makes decisions. the subcommittee 9 only tries to develop information and, to some extent, 10 organize the time for the full committee, since they have 11 less time. But they can al' ways reorganize it.
?2 They have all the freedom in the world. So, in O' 13 fact, I'm* going for those of the consultants that are here 14 tomorrow, I'm going to ask at the beginning of the 15 presentation rather than at the end for them to call out 16 what they think are things to pay attention to, things that i l 17 should be done perhaps, if there are any that are not on the 18 staff's list, or whatever.
19 Okay. So if one is satisfied with what the staff 20 has concluded, one can say that, for example. IC viere are 21 things that you haven' t heard and you don' t know whether 22 they're on the list, you can mention that. Okay? l 1 l 23 MR. MICHELSON: Aren' t there two problems that we j l (~) 24 have? One is what do we need to consider in the process of l tj l 25 making the decision of whether or not to go to 5 percent. ACE-FEDERAL REPORTERS, INC. M -3700 _ Nationwide @CT~J 80r) 3)(H5M6 i
1350 26 10 338 1 DAV/bc 1 The other question, I think, is the interesting 2 one. Is there something new, unique or unusual about this 3 event that we ought to be thinking about in terms of our 4 overall understanding of seismic conditions, and 5 particularly of future threats to the site. 6 DR. OKRENT: I agree that's a possible point. I 7 will wait to hear what our consultants have to say on this 8; and other things. r l 9l DR. SMITH: The two pages of confirmatory actions 10 l that the NRC staff has suggested, would that be part of the 11 full committee? 7-~ s 12 l DR. OKRENT: The full committee is going to hear m Il 13 that. . 14 ; DR. SMITH: Well, I would not add to the 15 confirmatory actions of the staff. I would only clarify the i 16 first point about additional quantitative assessment of a 17 l broader sample of equipment. I think it would be of great i 18 - benefit to the industry, and this is in effect a separate l 19 l issue from the licensing issue. - 20 But it would be a great benefit to the industry to have a careful evaluation of the types of relays, where 21 l 22 they're located, the levels of response and experience. 23 l Granted, they have no problems. 24 l That tells all of us something. And I would hope 25 to see that be done, but I would certainly not want to make ACE-FEDERAL REPORTERS, INC. w .u .t nn w, mo w, e m u m, nu -
1350 26 11 l 339 L /DAV/bc 1 that a requirement for a license. 2 The second thing, I would be very happy to hear 3 everyone say that the levels measured at the higher 4 elevations in the plant are real. We believe them. End of 5 statement. 6 On the other hand, if that isn't done, then it 7 seems to me there are four or five things that have to be 8 done to say why they're not believable. 9 I would want to see supports for the 10 instrumentation analyzed. I would like to see shell modes, 11 i some other modes of response of the containment analyzed. I
,, 12 I'd like to see the issue that Prfessor Trifunac raised
('~ ') , 13 l about Rayleigh waves considered, and appropriate issues of i 14 l damping, and so on. I 15 So my solution would be avoid doing all that and 16 say, fine, we believe it. That's all I have to say. 17 DR. OKRENT: I really am not sure I understand 18 what you mean when you say, fine, we believe them. 19 DR. SMITH: Well, the statements have been made 20 several times today that these measurements are not real, 21 they're influenced by other things. 22 DR. OKRENT: If you said , Finc, we believe them, 23 in fact, they're accurate? (~', 24 DR. SMITH: That they're real.
~
25 ' DR. OKRENT: Wouldn't you still be in the ACE FEDERAL REPORTERS, INC. EBCT4ED _ NM=h - - EC _- - __ _ _
I 1350 26:12 340 DAV/bc 1 situation of wanting to understand why they occur? i 2 DR. SMITH: Personally, no. I'm perfectly happy 3 3 to say they happened and they're not important. 4 DR. OKRENT: Just that they're not important. I 5 DR. SMITH: Right. 6 7 8 9 i l 10 i 11 12 i
- O 13 1 14 15 16 17 l t 18 19 ;
20 21 22 ' 23 24 25 I ACE-FEDERAL REPORTERS, INC.
- _ - -- --. - -- - Ese mia ------ gtmnenrs___m _ _ _ _ _
l l 1350 27 01 341 (~T - J i_JDAVbw 1 DR. OKRENT:' You will have your chance tomorrow 2 to make your points. 3 Mike? 4 DR. TRIFUNAC: Nothing further of significance. 5 Nothing to do with this site, necessarily. 6 I think we are definitely going to see small 7 earthquakes with high frequencies and short durations. I 8 think it is ridiculous to waste so much time on analysis of l 9 every one from the beginning. I think there are some l l 10 fundamental questions about the adequacy of the whole l 11 concept. People who live and work with Reg Guide 160, as l 12 you know, it is impossible. And if you exceed it here and l (' 13 there, everybody gets nervous about it. l 14 So I think it would be very helpful to address the fundamental questions, what could be done to help these 15 l 16 people not go through all these details, when these things l 17 are going to happen again for sure, on the one hand, but l [ 18 then on the other hand, I feel that the fact that we record i I I 19 something in the building, tells us about the engineering 20 just as well. And I think we should look at that. I think 21 we should learn the lessons. 22 Of course, we have a different frequency range l 23 than sore other places, I know, but in Mexico City, things {} 24 . happened that shouldn't have happened. We should have done 25 this 20 years ago, and we just didn't look at it. And if i l ACE-FEDERAL ) REPORTERS, lNC.
- - _ _ _ _ _ _ _ _ _ _ _ _ - - . _ _ _ __ ___ _ __ _ _ _ ___ ___ __ 202 3c.37m _ N_m_w W W ___ __ _ _ __ %3 3m6_ _ __ __ ___ _ _ _ _ _
I 1350 27 02 342 l DAVbw . I we have a recording like this, it is fantastic. We just 2 look at it and learn what we did wrong and not try to mold 3 it into something that it cannot be molded into, but do it 4 right and do it okay next time. In this case, no problem. 5 But these earthquakes are raising some serious 6 questions, and we don't have the machinery to handle it. 7 The other thing I am concerned with is the types 8 of analyses that we have in the recordings are not going to 9 sufficiently high frequencies. There may be some 10 fundamental differences between the East and West in the 11 United States. We should go to high frequencies of 12 recording and data processing. We see those spectra plotted 7-i (_)/
- 13 up to 30 hertz, 25 hertz, really. No data has more than 25 14 hertz, because it has been filtered.. Somebody puts a 15' computer program that goes to 33 hertz, and they think 16 they've d.ine some calculations, but I think we should extend 17 this, so that we can see what is going on in the Central 18 Eastern United States. i 19 The hues are different here than they are in 20 California, and we know that, but how much does this bear on 21 the recorded motions for engineering calculations? Nobody 22 f knows that. And next year we are going to have another one 1
23 of these. Maybe two years from now.
) 24'j DR. OKRENT: Paul?
25 DR. POMEI)
- I think, first of all, that we l l
l y . z. .. ACE FEDERAL REPORTERS, INC. smrum .m _-- m __ _
i .1350 27 03 343 O is DAVbw 1 need to investigate what the Staff has initiated with USGS, 2 in terms of the injection question. If there is a 3 relationship that can be established, and there is a 4 significant probability that a relationship exists, then 5 there is a possibility that it gats as high as the SSE and 6 perhaps higher. l 7 If it does, then the second question relates to 8 high f requency, and that is, is it of any significance? l 9 In this case, we significantly exceeded the SSE 10 at a point principally in the range of 20 hertz, but that 11 has no significance, and it has no significance, if we 12 exceed it by a factor of 4, instead of 2 or 8 instead of 2. 7-)
%J 13 , Then I think we are worried about the wrong ,
14 thing. Otherwise, I think we ought to really pull together i l 15 all of the high frequency information that we have, because 16 we have a lot, and there is a lot that we could learn by 17 just simply looking at all the records in the Eastern United 18 States, all the ground motion records. And we are not doing 19 that right now. 1 20 ' I think the Staff should be pursuing that. 21 DR. OKRENT: Let me ofLer a comment. l [ 22 I hav3 to assume there is some magnitude at which i 23 20 hertz is important. Okay? l () 24 { i Now whether it is 2 or 4 or 8 or something 1 25 bigger, I don't know, but it would be useful, in fact, ' ACE-FEDERAL REPORTERS, INC. _-- r~vT9Tn ---.---N'.- o- -----_ e -_ _ __ _ _
1 I l 1350 27 04 344 l t') ' L_/DAVbw 1 maybe to back into the answer, to make some estimate as to 2 just what magnitude would be interesting and why. 3 DR. POMEROY: I would address that to the Staff. 4 DR. OKRENT: It is not certainly this issue. 5 DR. POMEROY: The other issue that I have, Dave, 6 is really that 1 am concerned that everybody, as we've 7 heard, everybody and his brother are out in the Cleveland 8 area immediately after the earthquake. I think maybe there 9 may be two teams of people out there. In three or four 10 months, there isn' t going to be anybody out there. 11 I really have a strong feeling that we should : 12 have some monitoring capability, longer term than the next A 13 few months. I don't know whether it is the Staff o'r the 14 USGS or the Applicant, but that should be monitoring -- not 15 necessarily it should be monitoring in addition to the 16 strong magnitude events, but it should also be monitoring 17 the high sensitivity, so we can look at those strong 18 magnitude events. 19 DR. OKRENT: If you think this is important, it 20 should be in your written comments that I assume you are 21 preparing and think about it for tomorrow. 22 Any other points? 23 (No response.) (} 24 MR. STEFANO: Also, Dr. Okrent, one last thing. 25 We will have a brief presentation on the seismic origins ACE-FEDERAL REPORTERS, INC. m _ _ ____ _ _ _ _ _ ___ __ . _ _ _ _ _ _ _ _ _ _ _. __ _ _ _ _ fBEF-DED _N rtfb@~rO GBCC'3
- . . - - - - ~. -
1350 27 05 345 A
\ !
4 'DAVbw 1 here tomorrow that we could not answer today. Yes. i 2 MR. HERNAN: We are prepared to address that 3- tonight, if you would like. 4 DR. OKRENT: I have other things to do tonight, 5 besides eat dinner. So I am going to give you my 6 guesstimate for the agenda tomorrow. 7 It is tomorrow, isn't it? 8 2:00 o' clock tomorrow, I am told. 9 The first 15 minutes will be for a subcommittee 10 report and comments by those of the ACRS consultants who are 11 here tomorrow. 12 And by comments, I mean in the same way that I 13 asked you today to point out what you think are the 14 principal issues. And I would say in this case, things 15 related to Perry itself. You have made your point on the 1 availability, Dr. Smith. 16 l 17 Then we would go into a report by the Staff, 18 I which is their full report, not just an introduction, and i 19 generously allocated 20 minutes, plus 10 minutes, for 20 questions. 21 We would then have the report by Cleveland 22 Electric Illuminating Company and their consultants. And I 23 have again generously a2 located 30 minutes for the () 24 presentation and 15, minutes for questions. 25 We would then have a 10-minute presentation from l l ACE FEDERAL REPORTERS INC.
, rnearm _rc w t- __----m_________
1350 27 06 346 10 L_sDAvbw 1 EPRI on generic questions concerning this earthquake and the 2 Central U.S., or whatever it was, plus 5 minutes for
-3 discussion. And I have allowed myself 45 minutes in my hip 4 pocket for discussion after this, because I assume none of 5 you are going to stick to the time I have given you.
6 But I will feel free to close you off shortly 7 after you've reached your time. 8 Do you understand? 9 MR. EDELMAN: Yes, sir. 10 i DR. OKRENT: So that will use up all the time 11 the Chairman gave to me. 12 So af ter the subcommittee report, the Staff will ' O 13 be on, and in your report, you would include a few minutes 14 on seismic margins, a minute or whatever you have. And 15 also, we would like to hear from Cleveland Electric 16 Illuminating to get their response to that particular 17 par: graph in the prior letter.
- l 18 l MR. STEFANO: Very well.
I 10 DR. OKRENT: And we start at 2:00 o' clock. 20 (Recess.) 21 DR. OKRENT: Is the speaker for the Staff here? 22 Please, let's begin. And by the way, assuma 23 we've read the document, the draf t document. () 24 Is that okay? 25 MR. MICHELSON: Oh, we've read it. That's not ACE-FEDERAL REPORTERS, INC.
. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ ___ ___ ___ ______ EDOSDia _ ___ _ _ _ __ G2~D~ OC% > _ _ _ CWKf3 _ _ _ _ _ _ . _ _._ _1
- a. : .m- .sb . e A 4 *-e _.4 -4 Am A 1350 27 07 347 b DAvbw
- c. 1 the problem.
2 We want to know more why one arrived at a certain 3 approach than a history; okay? 4 MR. THATCHER: DO you want me to skip the history 5 and the general overview? I'll skip. Background, I'll 1 6 skip. 7 MR. MICHELSON: Is there a handout of this? 8 DR. OKRENT: You've got one, I think. 9 Starting at " Draft Resolution, August '85. You 10 had a different position then than now. Perhaps you could 11 try to tell us what it was then, what i t is now, and why. gs 12 MR. THATCHER: Okay. I think the best place to 13 do that is probably to go to about the sixth slide then. I 14 might have to back up to some of the findings to explain. l . 15 l MR. MICHELSON: What I would like to hear, l , 16 though, is a current explanation of the Proposed ! 17 Resolution, but that starts on page about 3. The bullet 18 says " Proposed Resolution." 19 MR. THATCHER: That bullet says it on about the 20 ' third slide, but I think if we go to the slide on the l ' l 21 Resolution, " Division Comments," maybe that would be the l 22 , best place to start, because that is the point where I was 23 going to get into the differences between what you say in 1 () 24 the August draf t anc what is in the package now. Okay? 25 (Slide.) ACE FEDERAL REPORTERS, INC. r n u m __ m. . ___ m ____ _
. _ _ _ - - - - - - . . . . ~. - 1350 27 08 348 DAVbw 1 Major comments on here. 2 DR. OKRENT: Why don't you first say what it was 3 you recommended? 4 MR. THATCHER: In August, we had a 5 recommendation for a number of actions. First and probably 6 mest significant was a requirement for a walk down, with 7 certain spatial interactions at existing plants. In 8 addition to that, we proposed a generic information letter 9 on about five functional types of interactions. We also 10 proposed that for future plants, we were proposing a new 11 SRP section that was focused in on mostly spatial type 12 interactions, including some of the review criteria, and so 13 forth, that we intended to use for the evaluation of the l'4 walk down to be performed on existing plants. 15 We also proposed guidance to a plant during a PRA 16 type of analysis, and we also stated that ongoing operating 17 experience were, indeed, doing a adequate job, and at one 18 time, dissemination of certain information, as I mentioned 19 before, the five function areas, was all we thought we 20 needed to do in that area. 21 Now that proposal went to all the divisions in i 22 ' NRR. 23 DR. OKRENT: Excuse me one minute. The generic j () 24 ' studies were what again? You said they were plant-specific 1 25 walk downs. ACE. FEDERAL REPORTERS, INC.
1350 27 09 349 DAVbw 1 MR. THATCHER: They would have to be 2 plant-specific walk downs. 3 DR. OKRENT: Then you said some sort of generic 4 -- 5 MR. THATCHER: The generic information letter to 6 all utilities. 7 DR. OKRENT: Okay. 8 MR. THATCHER: It would be for information only. l 9j Okay. The first comment here has to do with the I 10 l new SRC. As you see the Resolution, we have dropped the SRP j l I 11 section. Both the Division of Systems Integration and the 12 Division of Engineering believe, basically, that their 13 existing requirements and guidance, as spelled out in I 14 : various places in the Standard Review Plan, were adequate I i 15 l for the treatment of the adverse systems interaction that we 16 i were proposing to put really into one section. 17 DR. OKRENT: May I ask a question in that regard, 18 with regard to, let's say, water, do they feel that it is 19 l just that water is mentioned, but that the Staff has been 20 ! following it up correctly, or what? What do they mean when 21 they say " existing requirements and guidance are adequate"? 22 MR. THATCHER: In this kind of .omment, they 23 are really only addrecting the need for a new SRP. 24 Okay, now there are more specific guides in the 25 SRPs, which they fed back to us, with regard to water ACE FEDERAL REPORTERS, INC. rmsvmen m . . - . .m
1350 27 10 350 DAVbw 1 communications. It is not that it is just mentioned in the 2 SRP. In fact, there is a Branch Technical Position that has 3 a lot to do with pipe break and water. 4 DR. OKRENT: How about seismic effects due to 5 nonseismic components? Is that specifically in the current 6 SRP? 7 MR. THATCHER: It is throughout the SRP in a 8 number of sections 9 DR. OKRENT: It is not backfit; it's front fit. 10 Is that it? 11 MR. THATCHER: The SRP evolved around 1975. 12 DR. OKRENT: I am sure Perry, for example, has O 13 not done a seismic walk down like Diablo Canyon has. You 14 don' t list them as having done it, but they have done some 15 kind of a seismic walk down. 16 MR. THATCHER: I can't answer for Perry, 17 < specifically, but it is a new plant. I can answer for 18 Millstone, I think it was. I am not sure. But I am aware 19 of a number of plants going through the licensing review 20 right now that do quite extensive walk downs on that exact 21 subject. Waterford is one that I know. I was down there, 22 j but it comes mostly from Reg Guide 129. And admittedly, the 23 Standard Review Plan of Reg Guide 1.29, which were () 24 instituted at a certain time, and there may be older plants, 25 that that kind of a walk down wasn't then. Now the walk ; ACE. FEDERAL REPORTERS, INC.
-_ - - _ - _ ___----.-_____--__ N ---------__ M o-- ------ m
1350 27 11 351 ('): L- DAvbw 1 down I am talking about, I donsa't think they are probably 2 as extensive as was done at Diablo Canyon, but they still 3 were looking for the kinds of things that you mentioned, 4 nonseismic interactions. 5 DR. OKRENT: Your own report says only that two 6 plants have done seismic walk downs -- SONGS 3 -- two I 7 stations, SONGS and Diablo. 8 MR. BOSNAK All of the new plants do what they 9 call seismic two for one, and in a lot of cases, it may be 10 purely analytical. In other words, we look at seismic-r'onseismic. 11 Piping. Nonseismic equipment and .i 12 statial interaction with respect to seismic, but it may not () 13 l always be done. It can be done in company with a seismic 14 walk through. 15 DR. OKRENT: What I am trying to understand is 16 that these people apparently said existing requirements and 17 guidance are adequate.
; 18 MR. ANDERSON: Our idea in proposing a new ; 19 Standard Review Plan was that the current guidance in the 1
20 Standard Review Plan is stated in a number of different 21 sections and areas, and we felt that it would be much more 22 ef ficient and ef fective for review, if we bring it all 23 together in one place. So there would be a specific () 24 requirement for a systems interaction review rather than 3 25 have it scattered out throughout all the various sections. ace. FEDERAL REPORTERS, INC. l _ - _m ma. - w - - - __ m- _ ____
l I i l L1350.27 12 352 L.f)DAVbw l-V '1 I don't think that the intent was to add anything i l 2 to the actual requirement, but just to make a more efficient l , i l ' 3 licensing process. l 4 DR. OKRENT: So you think that by now Gary's , l 5 things.are sprinkled throughout? , i 6 MR. ANDERSON: We called their attention to these 7 means, and they pointed to the various standard review plan 8 sections, and it.is really there. How uniformly it's been 9 applied. ! 10 MR. MTcHELSON: I think you have to look 11 carefully at what is there, in view of the kinds of system ) 12 l o 13 interaction that you are working with, for example, in the pipe break sections, 3.6.1, I' guess it is. They deal there I i l 14 with the old letter. If you go back and look at those I 15 things carefully, you will find, for example, that you can 16 postulate all you want on pipe breaks at a plant, for
- 17. instance, and you go through the analysis for the pipe l ,
l 18 breaks at those point in time, the plant can still safely ! 19 shut down. Now what it doesn't address is the problem that 23 bothers me a little bit. 21- For instance, these seismic events, that's a i 22 common challenge to the entire plant. It is a common 23 challenge to all nonseismically qualified equipment, and it i
~h 24 is difficult for me to see why now only one pipe of some l 25 size breaks during that time. You know, we are in the l
I , ACE. FEDERAL REPORTERS, INC. '. - - _ - _ _ _ _ _ _ _ _ - . .-5~WDKn _M W
i' ( 1 i 1350 27 13 353 DAVbw 1 process of shaking 20-inch pipes. You are going to rip one l l ( 2 off each instrument line. Things of this sort. 1 3 Unless you are very sure from all your analyses. l 4 Now you know, if I break off one instrument line, l 5 hopefully, the analysis at 3.61 shows that, but it didn' t 6 show what happened, when I broke off six instrument lines, 7 because of a common seismic disturbance. 8 Now how are you handling that kind of system \ I 9 interaction? l 10 11 12 0 13 - 14 15 16 17 18 19 20 l 21 22 j 23 l 24 25 l l l ACE FEDERAL REPORTERS. INC. I _______________ m ____________ N v.c , t__ _ _ _ m .
l {}JO2801
- c. 354 1 DAVbur 1 MR. ANDERSON: I can' t answer that. I don' t 2 know how it was specifically handled under that standard.
3 MR. MICHELSON: I don't even find a discussion of 4 it. I find a good discussion of the fact that you are happy , 5 with the pipe break because you have gone through the 6 thing. I think that is not an unreasonable approach, but it 7 doesn't address the common cause seismic influence and 8 'doesn't address a number of other issues, pipe break, for 9 example, doesn't address the point failure. 10 What happens if a big 20-inch bellows breaks? 11 That is not addressed under the pipe break criteria. You O 12 can't go thre agh leak rate arguments and that sort of thing 13 at all. 14 Yet the system interaction only speculated -- l 15 well, it didn't even speculate. You said -- the NRC said 1 j 16 pipe, and, by golly, that is what the utility did. They did 1' 17 pipes. 18 But there are a number of weak components around I 19 the system besides, and pipes are not necessarily the most 20 serious of fender, and you don' t speculate here as to how you 21 would treat those, as a "for example." 22 M3. THATCHER: I don't know if we said pipe or i j () 23 not, but I think the event that occurred at Quad Cities that 24 resulted in the Giambusso letter -- i 25 MR. MICHELSON: Indeed, that was a bellows ACE FEDERAL REPORTERS, INC.
-- ------ ____ _ ____ rmeTuTn __ ___ m -
t .- ___--.____m_______________
1350 28 02 355 O'DAVbur e- 1 failure, but the rules that came through didn't address 2 bellows. 3 Even the Cities' DDDC-4 doesn't recognize the 4 f act that bellows were probably one of the weaker spots, and 5 some people put a lot of bellows in those systems, some 6 people haven't put any, and some plants are having a little 7 difficulty at the present time. 8 These are big bellows. So we still got the same i 9 amount of blinders on. It is a little narrow.
, 10 MR. ANDERSON: With regard to this particular 11 issue, in looking at a standard review plan, as I said, 12 there very well may be inadequacies in the old standard O 13 review plan and we didn't question them. We allowed
- 14 ourselves to be convinced by those people from whom we 15 needed signatures on our package that we do a good job.
i 16 And I think Bob has made a note of that here if l 17 there is anything that really needs to be done. 18 MR. MICHELSON: That issue we can pursue 19 informally with the staff. I think I know what the staff 20 position is now. 21 It is not perhaps an unreasonable one for plants i 22 that are already built, but when you start opening up 1 23 things, if you are deriving comfort from having done systems ( ), 24 interaction studies, don't derive too much comfort. They 25 are still rather narrow studies. I ACE. FEDERAL REPORTERS, INC. j
- - - _ - _ ----.--_---.-- rn mlren _-------- m ___--____-- m _
1350 28 03 356
!?
L /DAvbur 1 Another example which I can' t find too much 2 guidance on is how you are handling the degraded conditions 3 of air and power, and so forth. Sometimes the worst thing 4 to do is to lose a little bit of power, not a lot of power. j 5 In other words, it almost drops to 60 percent. 6 The resistance of the line goes up and as a 7 result you have terminal voltage dropping and no relays kick 8 out, but strange things happening when you start to use 9 several components in a degraded way instead of a total 10 way. 11 MR. THATCHER: I think that area is covered in 12 our functional discussion of the concept of fail-safe. 13 MR. MICHELSON: Maybe it is, but I want you to
, 14 point out to me how you are handling this under system 15 interaction because I couldn' t find the discussion. It was 16 a lot of paper, and I think it is something that you have to 17 show me in how you are handling degraded air and degraded 18 power.
19 MR. THATCHER: Degraded air is specifically 20 mentioned as a problem that occurred under the general topic 21 area of fail-safe. 22 MR. MICHELS0tle But you didn' t provide the 23 guidance on how we are going to now handle system () 24 interaction studies, or you didn't provide a dialogue so 25 that the utilities understand these problems. I hope they ACE. FEDERAL REPORTERS, INC. o ,t----__---_ m __ _ _
1350 28 04 357
,O) kdDAVbur 1 do.
2 Anybody that has experienced degraded air begins 3 to understand. 4 DR. OKRENT: The comment is existing requirements 5 and guidance are adequate, and I think Carl is questioning 6 that the existing guidance and requirements are adequate in 7 some areas and has tried to give some examples, and it is 8 not necessarily a complete list. 9 MR. THATCHER: I don't know if wo aro talking 10 about the guidance or whether we are talking about the 11 implementation of that guidance. The staff -- 12 MR. MICHELSON: The guidance isn't even there. . 13 MR. THATCHER: There is a gap between what gets
- 14 written down in the GDC and what gets done.
15 MR. MICHELSON: How can we begin to understand 16 solid state systems today when their losses are degraded? 17 And even under total loss of power they didn't r 18 understand how solid state devices seemed to fail in various I i
- 19 ways, and under elevated temperature conditions you don' t 20 know today.
21 In fact, there is a very tough problem in 22 pre-predicting the behavior of solid state equipment under i 23 elevated temperature, because if you talk to the experts ;
- () 24 ' they will tell you that devices do not fail under the same 1
- 25 mode every time. !
{ t j ACE FEDERAL REPORTERS, INC. r~vmrrn e m
1350.28 05 i 358 A L_)DAVbur 1 MR. ANDERSON: I don't know that we can address , 2 degraded conditions under systems interaction. 3 MR. MICHELSON: Isn't that what we are dealing 4 with? 5 I thought we were dealing with interaction 6 between the cooler that was trying to keep the room cool and 7 it fails, and, lo and behold, they have got two trains of 8 equipment. 9 MR. ANDERSON: That may be a whole subject area 10 in itself. I i 11 MR. MICHELSON: Isn't that systems interaction? ! 12 MR. THATCHER: Yes, it is, but can I day my 1 13 piece? 14 But your example there was -- we are going to get 4 15 into this debate over examples -- do you want to use my - 16 examples or do you want to use your examples? 17 You said the room cooler didn't work in a 18 degraded redundant safety systems. Okay, that is a spatial 19 interaction. 20 But the way to solve it, you are coming after it f 21 from an equipment qualifications standpoint, and that is 22 another kind of common cause problem. I am coming after it 23 that you ought to have two coolers, one for each train, 24 (]) redundant and separate. And the story, that is it, no more 25 system interaction. : ACE FEDERAL, REPORTERS, INC. . ; rymri:n _ _ _- . xa w ___________ m - _ __- _ _'
1350 28 06 359 f'T L /DAVbur 1 Now, if you want to argue about did we qualify 2 those two redundant things to high enough temperature, that 3 is a different story. 4 MR. MICHELSON: Or did you even realize that the 5 cooler was needed? Because you thought that the cooler -- 6 MR. THATCHER: I don' t. 7 MR. MICHELSON: Let me extend the example. You 8 have got one room, one train, one cooler. It all sounds 9 reasonable. 10 But what you didn' t understand is when you lose 11 the cooler in that room and degrade the electrical equipment 12 in that room, although it is one train, it feeds back and O 13 causes various kinds of actions around the plant, and, lo i 14 1 and behold, it got into the second train. 15 You say, well, that is not possible. But what it 16 does is it couples back through nonsafety-related equipment 17 that causes adverse effects on the safety grade equipment in 18 another area. 19 For instance, I am not saying this is a good 20 example, but let's say that for some reason the fact that 21 this controlled equipment degraded in the room, actuated the 22 fire protection over in the other area, which maybe it did. 23 I don' t know. () 24 See, that is the kind of system interaction I am 25 talking about. I think that is the kind that someday is ACE FEDERAL REPORTERS, INC.
- - __ --- _ -- _ ------- _--.------- - ----- Nm - -- - - - men-rY*m -- - --- _---- e ---------- i
1350 28 07 360 r~N (J DAVbur 1 going to get you, and this program has somehow going to get 2 a handle on that, and I think it has got a ways to go. 3 I think what you are doing is good. You are 4 looking at as many examples as you can find, and you are 5 studying history and these techniques of trying to find out 6 about these funny things. 7 I think, though, that it isn' t all going to be 8, history. You are going to have to do several kinds of 9 testing, for instance, on electrical equipment to find out 10 how it responds to elevated temperature. 11 MR. THATCHER: We do that all the time. There is 12 a whole environmental qualification program. O , 13 MR. MICHELSON: Now, environmental qualification, 14 you look carefully at your environmental qualification 15 program. The purpose for it was to assure'that the given 16 piece of equipment could perform its functions under adverse 17 environmental conditions. It never went in there to see 18 what the failure modes in the equipment would be under 19 adverse environmental. If it will, it will work. If it 20 don't work, fix it. 21 That was the approach, and that was only on a 22 limited set of equipment. 23 Now, we are dealing with a lot of other equipment 24 which didn't perform essential functions in that particular (}} 25 case and which would give you unwanted actions which could ACE FEDERAL REPORTERS, INC.
------- _ __ fBDT ED - _ _ - N h _ -__---- 0 3 C --_----- - ---._----_ ---
l 1350 28 08 361 r~x ,
\,_)DAVbur 1 get you in trouble.
2 The fire protection one certainly could givo you 3 unwanted action and constrain the wrong equipment at the l 4 wrong time. l 5 MR. THATCHER: Of course, that aspect is clearly 6 covered in Appendix R. 7 MR. MICHELSON: Yes, I will admit. But what they 8 didn't harp on was the way in which the fire protection 9 might be actuated by other apparently nonrelated effects, ; 10 like loss of a room cooler that might heat up the 11 equipment that is monitoring and controlling fire l 12 protection. O 13 MR. ANDERSON: I dont really know how you find 14 all of those, but I think we went through this discussion 15 the last time we were down here, and there are a number of 16 scenarios. Mr. Ebersole had a number of them. ! 17 We can' t solve all the problems in the world, and 18 we have been trying to put together what we thought was a [ 6 19 very reasonable resolution of the issues, and there are l [ 20 areas like degraded equipment, degraded voltage, degraded ! 21 air systems. 22 I don' t know that we can address that in an I 23 effective manner. I don't know what you do. () 24 MR. MICHELSON: Here is what bothers me. You 25 named it just a moment ago. I think you honestly believe l ACE. FEDERAL REPORTERS, INC. mana mcemr, ------ rmunc~n '
1350 28 09 362 ; (~h L/DAvbur 1 you have resolved the issue. 2 I don't honestly believe that your program has 3 reached the point where you can claim you resolved the 4 issue. 5 MR. ANDERSON: Mr. Michelson, on our technical 6 findings, I think the first bullet under there says that we 7 cannot resolve systems interaction as a subject area. We l 8 had to drastically reduce the scope of this problem in order l 9 to get our arms around it and come up with what we think is 10 a reasonable resolution. 11 Now, there are a lot of peripheral areas and 12 so-called common mode failures, manufacturing errors, , ( 13 calibration errors, and mechanisms to search for various l 14 scenarios and plants that we just don't feel that we can 15 handle. 16 MR. MICHELSON: I think you are quite right, but 17 now are you claiming the issue is resolved? 18 l MR. THATCHER: We are not claiming that there is 19 l never ever going to be a problem. l l 20 MR. MICHELSON: In a sense, you are claiming it 1 21 is sufficiently resolved that you can take it off your 22 list. l 23 That bothers me because I don' t think it is ready () 24 to go off the list. l 25 l MR. ANDERSON: No, no. What we are saying is we l l ACE FEDERAL REPORTERS, INC. f%f64E0 . Gt t a f b h - -____.___ I ! M C C D -
1350 28 10 363 - ID LJ DAVbur 1 'think that based on the work that we have done and the 2 experience and our assessment, as qualitative as it may be, 3 of the safety significance, that we know enough about it 4 that we can take it off the list as an unresolved safety ; 5 issue. 6 We think there may be some residuals out there, 7 but we think that the safety significance isn't such that it 8 should continue to be a USI. 9 MR. MICHELSON: Then I have to judge that i 10 decision. Is this thing ready to be removed from the USI? 11 MR. ANDERSON: Yes, sir. 12 DR. OKRENT: That is one of the things you need I O 13 I to j udge. I 14 l MR. MICHELSON: When I look at what you have dono I 15 so far, what you have published, what guidance you have
- 16 provided, and so forth, on the basis for trying to make a l 0
judgment, is this thing ready to be removed from the USI l 17 l 18 list? 19 , MR. ANDERSON: That is the question. 20 MR. MICHELSON: I was hoping to hear moro, and I 21 don' t think we are giving it a fair shake with this process d 22 at 7:00 o' clock at night, what Wolfmann refors to as 23 nonprime time and complains bitterly whenever his subject is 24 l stuck on at the end like that. 25 DR. OKRENT: I am just as interested in this ACE. FEDERAL REPORTERS, INC.
-_ _ _ mmm_ - wemr, e
1350 28 11
^
364 , l L-)DAVbur
/ 1 subject as you are.
2 MR. MICIIELSON: I don't think it is a fair way 3 for the staff. I think it is not a fair shake for them. I 4 think you have to have some prime time to do it on, and I 5 think you need to go through and carefully try to reach a 6 conclusion: is this subject now ready to be removed from 7 the USI? 8 MR. BOSTAK: I think removing something from a 9 list doesn't mean that the whole thing -- that you can 10 forget it. 11 MR. MICilELSON: Experience says it does. 12 MR. BOSTAK: We don't look at it that way. I O'~ 13 think the fact'that the issue would be publicized to the i 14 ; industry, the fact that you are going to have to do 15 li something, and whether or not each and overy possible 16 interaction can be covered, say, in a matrix that we put in, 17 you just can't do that. 18 So we are trying to make the industry aware of 1 19 l the fact that this is something they have to contend with. I 20 ' MR. MICHELSON: I think it is only fair that you 21 have this opportunity to present it. 22 23 l () 24 25 ACE. FEDERAL REPORTERS, INC. r = w s s ta en 1 -------- mmrm -
1350 29 01 365 l f3 b DAV/bc 1 DR. OKRENT: Carl, let me mention what seemed to 2 be a few different futures. One is, if we were satisfied 3 that the items identified in the staff paper were a 4 sufficiently complete list, are we satisfied with the 5 proposed means of handling it, which would now mean in fact 6 no SRP on those? And no specific requirement for plans to f i i 7 do a limited focus study. ! t 8 In fact, they' re weren' t in August going to ask ) 9 for a functional focus. I guess they thought that was too 10 difficult.
; 11 MR. ANDERSON: There's no payoff in it.
l
! 12 DR. OKRENT: So, now what they propose, as I i
l O 13 understand it, is to have an information letter, and also to i . l 14 recommend that somehow in this statement that the Commission ; I l 15 said in their safety policy that plants will do some kind of i i t 16 systematic study, that this would include the focused j 17 studies that they would be talking about. ; l 18 MR. MICHELSON: But no requirement. 19 DR. OKRENT: Well, in fact, that's.something ! 20 still in the future. j l 21 MR. MICHELSON: And not much guidance from the ; } t 22 staff. 23 I DR. OKRENT: In fact, EPRI in developing some 24 kind of quasi-surrogate approach for doing it. In the first j 25 place, this is certainly nothing you do with a surrogate i i ACE FEDERAL REPORTERS, INC. R.P3 m ______________Grt~r^JC u n__ ____ m _ _____-_-__
1350 29 02 366
/s Ls DAV/bc 1 plant.
2 So one decision is, as I say, if you think what's 3 in their proposal is enough, is the approach that they're 4 proposing, which is self-regulation, and you know, if peoplo 5 would really take self-regulation seriously on all sides, 6 I'm all for it. 7 MR. ANDERSON: Dr. Okrent, pardon me for 8 interrupting you but with regard to that, we did have some 9 discussions with the AIF subcommittoo on systems 10 interaction. Also, we got somo good understanding, a little 11 better understanding of how the INPO operatos in regard to 12 their information letters and other documents that g,o out to O 13 the utilition. 14 We found, first of all, in the area of flooding, i 15 what they propose is ossentially what wo woro getting ready I 16 to proposo. And we think that we understand now that it t 17 will be policed by the industry, and also audited by NRC 18 resident inspectors. That makes us fool much more 19 comfortable with issuing the information letters. 20 So I don't think just by tolling the industry 21 about the problems means that they look at it and shrug 22 their shoulders and walk away from it. 23 MR. MICHELSON: Aro you happy, for instanco, with () 24 your resolution of water hammer and what's happoned since 25 the resolution of the water hammor? ACE. FEDERAL REPORTERS, INC. SmPTuvn -e. ---_--m_____---_
l 1 1350 29 03 367 s i ( t'J L- DAV/bc 1 MR. ANDERSON: The water hammer was much the same , l 1 l 2 type of issue. There's nobody that's ever going to do away i 3 with water hammer. Every timo you flush a toilet or turn on 4 the faucet, you're going to get one. i j 5 MR. MICHELSON: The industry is supposed to keep l 6 an eye out for these problems themselves. What did they do l l 7 when they realized that checkvalves weren't checking and so 8 forth? I 9 Well, they took appropriate steps. 1 i 10 MR. ANDERSON: Unfortunately, the incident you're i I 11 referring to, I think they took the steps after the
)
12 incident. l O 13 MR. MICHELSON: This is the same thing. Unless p l i 14 really serious system interaction occuen, they may or may . l 15 not do anything. Is that good enough? Maybe it is. 16 MR. ANDERSON: I think what they will do is they 17 will take the guidance or the information seriously. I f 18 don't think they can be totally effective, precluding never 19 having another one. l f 20 I'm convinced and I think most of the people l l 21 associated with this program are that there's really no way i l l 22 that you can eliminate syster.s interaction. l l 23 MR. MICHELSON: Do you believe that the agency (]) 24 now will do anything more about system interaction once it's 25 determined that it's resolved? Will there be any more money l l ACE. FEDERAL REPORTERS, INC. _ _ ______ _ ____- _rVmi:n __-_ m ___ _ _ _ __ m _
t 1350 29 04 368 (~') l i L /DAV/bc 1 to do work on the walkdowns? Do you thi.nk this is going to 2 be an item of much interest on the part of the staff once 3 it's resolved? s ! 4 MR. ANDERSON: I don' t think there'll be any ! 5 intensive effort. 6 MR. MICHELSON: I think you're proposing that , 1 7 once you resolved it, this is it. This is all the NRC is ! 8 going to do. 9 MR. THATCHER: I would say that I think more is l 10 going to be done in the area of severe accidents. ; 11 MR. MICHELSON: This is how you get into a severe l l 12 accident. I'm not sure that your accident program is going () 13 to help this. ! 14 MR. THATCHER: That's exactly it. 15 MR. MICHELSON: I don't think they're worrying 16 about this. I think they're worrying about once you've got 17 a severe accident, how do you take care of that? 18 MR. BOSTAK I just want to add to what Knut l 19 Anderson has said here. I think industry is going to be ; 20 looking at things like the San Onofre event. 21 MR. MICHELSON: I'm a little acquainted with how 22 INPO evaluates these events. Do you think INPO is primarily 23 interested in pursuing things like system interaction l (]) 24 i studies? Or do you think they're more interested in 25 correcting equipment once it's f ailed or malfunctioned? ACE. FEDERAL REPORTERS, INC. __ __m _ __ _ _ _ _ _ __ Mv .t_______-_e_______-------.-_--
r 1350 29 05 369 L >DAV/bc 1 MR. BOSTAK: I think they're interestod in 2 looking at events that have potential for doing somothing to 3 plant operational capability, either safety or economics. ! 4 MR. MICHELSON: You've looked at their SER's, I'm 5 sure. Rarely do they deal with those sort of hypothetical 6 oxtrapolations of experienco. They deal with fixing the 7 problem. 8 MR. BOSTAK: I don' t think they' re going to bo 9 trying to look for possible hypothetical accidents, but they 10 cortainly are going to try to follow through on things that l 11 have occurred. l l 12 MR. MICHELSON: This is what we' re doaling with O 13 though on the pipe breaks in this room. When the water gets 14 in the duct and it runs down and starts spilling into here 15 an'd there, that's a typical system interaction? ! 16 If you got two trains of equipment? 1 17 MR. THATCllER: I think the SER on that subject 1 . j 18 j approached it the way the staff would approach it. Now I 19 don't caro if the fire protection system or what drips on 20 some pump casing, but I do caro if it gets in my logic 21 ' cabinets. That's when you start to look back. 22 That's what INPO, as I underctood, had in their 23 proceduros. The samo as I would have proposed. In fact, I l
) 24 was in the process of writing one at the timo.
l 25 MR. ANDERSON: They como to the samo conclusion l l ACE. FEDERAL REPORTERS, INC.
-_____ _ ----- ImPttm __ m___ _ _ __ __ __m _ __.
1350 29 06 370 g L/DAV/bc 1 that we did with regard to flooding, that we probably missed 2 the mark when wo worried about large volumos, because that 3 isn't tho way it happens. 4 DR. OKRENT: Well, it did happen that way. 5 MR. ANDERSON: Well, it happens both ways. 6 DR. OKRENT: But that really wasn' t even treated 7 really adequately. It's sort of half of a fix that way. 8 But the thing you're talking about now that INPO is, say, 9 proposing something. 10 MR. THATCHER: A sorios of events have occurred. 11 DR. OKRENT: So I'm not sure if this meets Carl's l' 12 point about whether they are attacking with the samo vigor. () - 13 l MP. MICHELSON: I think they're moro reactivo. . l . 14 DR. OKRENT: It's much easier to tell other 15 people about their own plan. It's easier to toll other l 16 l peoplo that this is a problem which has occurrod onco. I 17 moan, I can understand, you know, the position they'ro in. 1 18 And you can only push a cortain number of times. l l 19 Woll, look, lot me ask a procudural question. 20 Whora does this thing stand as far as you're concurned? 21 I notico that what you gavo us was a draft. That 1 22 means that it's not gono to SRGR yet. 23 MR. ANDERSON: No. What that means is that, as () 24 usual, wo woro crowdod for time that wo wanted to got this j 25 packago to you and wo woro preparing it to send it up to l l l
- ACE. FEDERAL REPORTERS, INC.
SpV9x t ma i t ______m-________-__
1350 29 07 l 371 l OL JDAV/bc 1 l Mr. Denton. It's now in Mr. Denton's of fice. We have had a l l I 2 meeting, I guess, with Darrell Eisenhut and the division l l 3 directors in NRR, where we went over essentially the same { 4 thing that Dale was presenting here. And we got essentially I 1 5 concurrence with the exception of one item, which we j 6 knuckled under on. l 7 And we corrected it. So, well, every once in a f 8 while -- { 9 MR. BOSTAK: That was a perception problem. ( 10 MR. ANDERSON: It was a perception problem. They 11 thought we were wrong, and we thought we woro right. [ 12 ! 13 l l 14 i , f 15 16 ! 17 i i 18 19 [ l 20 l l 21 l 22 l l t 23 f l 24 ' i 25 ! l l l l ACE. FEDERAL REPORTERS, INC. ! '___ _ ___ __ _ _ _ _ _ Nm__ _N_____. h _. i
1350 30 01 l 372 3 (v DAV/bc 1 (Laughter.) 2 f1R. ANDERSON: But the package which you have is < 3 ossentially tho name one that wont up for their signaturo. 4 And that signaturo is a transmittal letter to the CUGR 5 Committee. We're asking the CUGR Committoo to give us a 6 recommendation for public commont. l 7 MR. MICllELSON: llavo you seen your schodulo yet? l l 3 MR. IIERNAN: We havo to send it back. Wo' d 9 oxpect it goes to the CRGR in the next month. 10 DR. OKRENT: Suppose the ACRS thinks that they 11 want to look at it seriously? And, in other words, thoro's i 12 reason that they at least soo smoke. Okay? Do you know O 13 what I moan? l 14 It may be a fairly strongly committee but not in l 15 support of this. In your opinion, the best thing is for you 16 to proceed the way you are in the full committoo and ACRS 17 should act separately? Or, if this is the case, suppose tho 18 ACRS sont a memo to the EDO saying: 19 Wo havo soon this draft. The subcommittoo had a 20 mooting with the ntaff, but wo think tho full committee may l 21 have somo serious roservations with this draft. If you aro
- 22 willing to work with us, we would try to deal with the 1
23 matter via subcommittoo and full committoo in the next O 24 co"ot or "oataa-25 Do you think that would be a good way to go? l ACE. FEDERAL REPORTERS. INC. rvan m _rm n ~ m _ _ _ ______ _
1350 30 02 373 l (~)/ L- DAV/bc1 MR. ANDERSON: I soo no problem with that. What 2 we' re asking for now is for concurronce to go out for public 3 comment. It would be a 60-day public comment periad, after 4 which we'll be resolving comments and preparing the package 5 again. , 6 So we'ro a good nino months to a year. l ! 7 MR. 00STAK: Some of your comments could be 8 brought in in the public comment period. 9 DR. OKRENT: I'll toll you, in a sonso, I look at 10 the ACRS as sort of the onos who roquestod the information 11 on this one. I really don't look upon us as one of the 50 , 12 members of the public on this issue. 13 MR. ANDERSON: I understand. I 14 l DR. OKRENT: I would profor, mysolf, although I 15 don' t know how the other committoo membors would fool, but 16 at least for tho full committno to have a chanco to say l l 17 whether or not it wants to offor a comment before the thing 18 goon too far. 19 MR. DOSTAK Whether you have some comments that i 20 ' might influence tho whole document. l l 21 DR. OKRENT: The whole documont enough, yon. i 22 Right. 23 MR. HERNAN: May I quoto f ran the Memorandum of () 24 l Understanding that's boon signed f rom EDO and Fraloy on 1 25 satoty-rolatod guidanco such as this? ACE. FEDERAL REPonTuns, INC. ___ r~v w :n m __-__ m _ ______
I 1350 30 03 374 l DAV/bc 1 It says: 2 "Normally, ACRS comments on safety-related 3 comments would be provided in the following stages: i 4 One, af ter CRGR review. But, prior to issuance 5 of public comment." 6 It seems to me, the way to go in this case is to 7 continue with CRGR processing as we are, but before we 8 actually go out for public comment. But I do think the ACRS 9 should put out comments. 10 DR. OKRENT: I find that acceptable. 11 DR. SAVIO: Dave, we sent a separate note to EDO 12 on this requesting ACRS have a chance to comment. Ray has O. 13 I tentatively had time scheduled for ACRS consideration at the 14 April meeting. I don' t think this gets ahead of your Cuger 15 discussion. 16 DR. OKRENT: April is when? This month or next 17 month? 18 ' (Laughter.) 19 MR. ANDERSON: We'd be happy to talk to you any 20 time, as long as the meeting doesn't start after 2 o' clock. 21 (Laughter.) 22 DR. MARK: It has to start earlier than 7 o' clock 23 at night. O 24 "a ^"ocaso"' " tt '21 co tat e bue - l 25 won't be happy. ACE FEDERAL REPORTERS. INC. ________ ____ N ____ M - ... m. u
, 1 !- l l I I i l F ' i 1350 30 04 375 ,' DAV/bc 1 DR. OKRENT: So you say there's tentatively time ' l l 2 on the full committee in April? I L 3 DR. SAVIO: Yes. In anticipation of some i 4 subcommittee review tonight, and others. 5 DR. OKRENT: Will you be available the Wednesday l 6 before? In the morning, for example? How are you booked? 7 Or first thing in the afternoon. 8 I think we need to have four or five hours. l 9 MR. ANDERSON: To spend a lot of time on it. ; 10 DR. OKRENT: Furthermore, we ought to organize i 11 the different aspects of it. Now, today, I knew it was late 12 and I literally' wanted to get to the heart of the matter () , 13 quickly, which you did b'y putting that.on. And I apologize. i 14 MR. THATCHER: That's fine. We want to go home, j 15 too. 16 DR. OKRENT: I think we maybe made more 17 progress. ! i 18 MR. MICHELSON: I'sa not available at that time at l 19 allt my daughter to getting married. l l 20 DR. MARK: Do you feel as though you should be t 21 there? 22 (Laughter.) f 23 MR. MICHELSON: I think so. l () l ! 24 DR. OKRENT: So, the Wednesday before the April ; 25 meating, you are unavailable. I won't get in until t I t i ACE. FEDERAL REPORTERS, INC. I _______MGa __--- K . J Yw o--_- N _ -- -----_ ---- J
1350 30 05 376 r" l bb DAV/bc 1 Wodnesday afternoon. l 2 DR. OKRENT: And you don' t want a Wednouday night 3 mooting? 4 MR. MICllELSON: That is what it would amount to, 5 yes. I 6 DR. MARK: I am way behind in this discussion. I l l 7 wondor if you could interpret for me that second paragraph 8 on the loft. f i 9 MR. TilATCllER: DSI and DE? That ono? Okay. I l 10 think basically they said we didn' t have onough ovidence. i 11 This comment was in our August packago. At that timo, wo t l 12 woro proposing a walkdown requirement. O 13 I DR. MARK: At this timo, you said you didn' t have I i 14 onough evidence to justify a walkdown. Then, what doos tho 15 last aantonce mean? You justify not doing something. 16 MR. TilATCllER: That's exactly what wo got in the 17 comments. If you think you justified a spacial ono, you 18 may or may not havo. But, not only that, they didn't think 19 we justified not doing it for functional. 20 DR. MARK: So thuy thought you should do it for 21 functional? I 22 , MR. TilATCllER: They didn' t say that. That's very 23 different. Nobody said that. i 1 () 24 DR. HARK: You didn' t justify skipping the 25 functional. They didn't nay you should includo it. So that l l l ACE. FEDERAL REPORTERS INC. _____ ___ _ _ ___ _ _f4PWhT1 Dr i~%. 1 m -
1350 30 06 377 O 1 U DAV/bc 1 really moana just what it says. 2 MR. T!!ATCllER: Yon. 3 DR. MARK: Okay. Now, in what fashion. When I 4 think of a walkdown, I think, for instance, of discovering 5 that an air conditioner is tootoring on the edge of the 6 shelf. And whon it falla off, it will fall on an instrument 7 airlino. 8 MR. Tif ATCilER: Possibly, yon, sir. 9 DR. MARK: Thore is no way in the SRP whero l' 10 you' ro over going to discover that. Evor. 11 MR. TilATCllER: By looking at the GRP7 12 DR. MARK: By looking at the drawingo of the e ! 13 ! airlinos, you wouldn't realize thoro'n an air conditioner 14 oitting up thoro near tho roof. 15 MR. TilATCilER: That's why a walkdown would be 16 part of a soarch for spacial problems. 17 DR. MARK: Now, at pronont, it is not proponod 18 that thoro be a walkdown with that kind of question in mind. 19 DR. OKRENT: That's a requiromont. They're going 20 to recommand to tho industry...woll, they'ro going to point 21 out cortain problems in the information lottor. And I 22 guono, and this isn' t cloar to me, I guons they think that 23 as part of novoro accidont policy and the systomatic I
] 24 oxamination of what I'll havo to cay is mont plant -- I 25 don' t know what they'll do with plants that already havo ACE. FEDERAL REronrEns, INC. ;
.ssse scrar c e m o cerca
1350 30 07 378 l (~)DAV/bc L_J 1 some kind of PRA. That they would recommend that thin 2 include focused walkdown. 3 DR. MARK: I would like to hope that a focusod 4 walkdown is something that would have a high probability of 5 happening. 6 MR. ANDERSON: Dalo, on our last -- you cin' t 7 road it thero down on the bottom, on the right. When wo 8 talkod about the nood for a solomic intoraction walkdown, wo 9 considorod what wo boliovo in going to be dono under the 10 implomontation of USIA-46. Wo think that tho scope of ! 11 equipment in ossentially the samo. 12 And they are preparing detailod review procoduron O . 13 for a walkdown in conjunction with USIA-46, which includes 14 limited syntoms intoractions review, which would look for 15 air conditioners. 16 DR. MARK Look, I don' t really caro whether it's 17 under USIA 17 or 46, but I would liko to son that somowhoro 18 ' it doonn' t got lost. 19 l . l 20 t 21 22 23 ( 25 ! i Acc FEDERAL REPORTEns, INc. <----------- ----- ____- E3MM3-__ _ _MW CD EC 1 _____'
- f I
l l l t l 1350 31 01 379 i DAVbw 1 DR. OKRENT: Let me tell you that things have a l 2 habit of getting lost, when they deal with the next 3 unresolved safety issue, or they are given to some committee f 4 or commission or panel to study. If you look at history, I 5 think more get lost than not, that way. i l 6 MR. MICHELSON: A means of reducing USI. That l 7 is all it is. l 8 MR. ANDERSON: No, it is a matter of not 9 duplicating the effort. ! 10 DR. MARK: Well, it shouldn't be duplicated, and f 11 it shouldn't get lost. Something that has that effect is 12 fine with m,e. l
.O 1.1 DR. OKRENT: What is the title of A177 14 MR. ANDERSON: Systems Interaction. f 15 DR. OKRENT: What is this title of A467
[ 16 MR. ANDERSON: Seismic Qualification of Q ; 17 in Operating Plants. l l 18 DR. OKRENT: Qualification. And it is really for ; 19 the things that need to be qualified. i 20 MR. ANDERSON: Well, it is for equipment in the [ t 21 plant, and it is required to bring the plant to a hot l 22 shutdown condition. ! i 23 DR. OKRENT: Yes. In other words, it really C 24 doesn't mean an it was originally written to include f 25 nonseismic equipment. t i ACE. FEDERAL REPORTERS, INC. l
-__ r v nt n -_--___--.--- m - 1 - __m
1350 31 02 380 LS DAVbw 1 MR. ANDERSON: No, it includes .t only to the 2 extent that you have to ensure that you can protect the 3 equipmont that you are trying to got soismic verification 4 for. So it does, indeed, include a noismic systems 5 intoraction oviow for proximity problems, sitting too clono 6 together. It looks at support systems for connections and 7 flexibility. So it is not too different than wo would 8 specify for a focused surge on A17. And the utility group l 9 has committed to do this roviow. Although our final 10 approval for it isn't comploto yet, and wo are progroaning 11 rapidly to develop the proceduro to do it. 12 I DR. OKRENT: Is this liko what Diablo Canyon did, O ,13 would you say? . 14 MR. ANDERSON: No, it in a far cry from what 15 l Diablo Canyon did. They hit it with a stodge hammer. Maybo 16 3 this in more of a tack hammer. But what it in in 17 specifically looking at equipmont to bring the plant to a ,
*t 18 hot shutdown condition. I think Diablo Canyon wan much l 19 1arger in scopo. i l 20 MR. MICllELSON: Dut very littlo related to tho 21 system interaction, because they are focusing on the noismic 22 thing. Ilot shutdown, undor noismic conditions. And ad 1 23 understood it --
l O 24 , "a ^"ocaao"' rn e t cer et-25 MR. M1CitEL90N Now wo are dealing with many, 1 Acu FEDERAL REl'ourEns, INC. __ r mm n m_ _
?
1350 31 03 38 1 i DAVbw 1 many other issues. The potential for adverse interaction 2 under seismic conditions. That is really the issue. ; l 3 MR. ANDERSON: That is one of them. l 4 MR. MICHELSON: That might won' t be covered by l 5 46. The system interaction potentials. I i 6 MR. ANDERSON: Not all of them. ; 7 MR. MICHELSON: Just some of them. I gather ( I 8 those related to heightened pressure boundarien. ! 9 MR. ANDERSON: For the seismic part, what A46 is [ 10 going to do to ensure protection of the equipment within ! i 11 their scope isn't too far different than where we landed for i 12 our focused surge on A17. i (:) 13 , MR. MICHELSON: Now let's take the case of t 14 degraded power. If you put it over in A46, they are not l l 15 even going to know what to do with it. They have a ( 16 different kind of talent over there. They won't even know 17 what to do with it. 18 MR. ANDERSON: No, just that part of it. f I 19 MR. MICHELSON: You are going to break up the i 20 program and move the pieces and put some in A45, some in 21 A467 ! 22 MR. ANDERSON: We are going to avoid duplication ' 23 of requirements, if we can l i l () 24 In answer to the question == f l l 25 MR. MICHELSON: The electric power question. { l i l ! l l l ace.Fctn!nAt. REPORTERS, INC. 1____ _ ._rywn _me.. ;z t- c'e~n
1350 31 04 382
<m LJ DAVbw 1 Whero are you going to put it?
2 MR. TilATCilER: If you want to talk about a 3 degradod power condition -- 4 MR. MICllELSON: Ne,, degraded power within the 5 plant. 6 MR. T!!ATCi!ER: The Millstono ovont in 1976 went i 7 into the plant, put in the second lovel of under voltago l l 8 protection. ! 9 MR. MICHELSON: My oxample was within the 10 plant. That isn't to any that that is the only degradod , 11 powor that could be usod as an examplo. i 1 12 MR. TilATCilER: To addrosa your question, I havo l ( l 13 ' ,to broaR tt into oloctrical powor. If you want to talk 14 about I&C power nupply -- you have said you are dealing with ! 15 those noparatoly. 16 MR. MICilELSON: Where are you dealing with thom 17 separately? 10 I MR. TilATCllER: It in not at at USI. It is a 19 generic insuo. The problom became ono -- A17 went through ! 20 ! and did its naarch with its definition. I, for one, and l 21 other people bocamao concernod that thoro should be moro I&C 22 power kinds of thingn in this list. So wo decided, let's 23 nearch it dif forontly. Lot's como after it a difforont () 24 way. So wo soarched. Tho I&C power supplion -- and when ! 25 nay I&C powor supplios, I am talking about DC buson, AC I Act!.Fliot!RAL Rt!!'onnins, INC. r,rnen c4 mam, - -- -- -- -- r n~
1350 31 05 383 A b DAVbw 1 busos, all this kind of thing. I And wo became a little bit disturbed by the 2l 3 number of ovonts, not nocessarily adverse system 4 interaction, just the shoor number of ovents, some of which 5 were adverso systems interactions, in our definition. 6 Subsequent to that investigation, wo started pursuing what l 7 is going on already. That is when we started finding out, 8 there are a number of issues dealing with I&C type power 9 supplies, one being Generic Issuo A30, which is the l 10 l safety-related DC bus. I i 11 l Another issue is Generic Issue 48, which is l 12 limiting conditions of oporating on vital buses. Wo decided [ O 13 ,' at that point that sincp the concern seemed to bo larger l - 14 l than just the adverso systems interaction, that maybe what 15 l wo nooded was a gonoric pulling together of thoso, and in r 16 ' the reorganization, we woro able to do that. The Division l 17 < of Safoty Roview and oversight inhoritod those gonoric I 18 issues that wore being worked by various parts of the 19 i Staff. Wo brought them all under one. issue, and we are 20 proposing to havo one task action plan to pull all that 21 togethor and expedito what wo think noods to be dono first l l 22 and work on other aross as nocoscary. l 23 lift . MICllMLSON: Maybe I misunderstood. I 24 thought you wore now proponing to break it all up and send 25 it back out to whoro it camo from. I thought the wholo 1 l l l ace FEDERAL. RnponTuns, INC. __ _ ___________ . . _ __ f3 M M] _ _ _ __ ctrzcGmco - 0Mcc3
1350 31 06 384 DAvbw 1 object was to get this system interaction in the one 2 package, because it involves mechanical, electrical 3 interaction with differnt systems, and so on, and you 4 couldn't divide it up into electrical and mechanical, and so 5 forth. 6 Now you are telling me -- I thought you were 7 telling me to break it up again. 8 MR. ANDERSON: I dons't think this is a matter of 9 breaking it up. The issue that he is talking about, I&C 10 power supplies, what Dale discovered is that there are a i 11 number of other issues on the Staff, some have been around I 12 for many, many years, which all deal with the, same area. O 13 Some of them, the implementation had started, and for some ' 14 l reason, stopped, and there are what, eight or ten issues, l 15 Dale? 16 MR. MICHELSON: And some of the issues don' t 17 deal with the potential action or potential ef fects on the power supply or loss of ventilation or things of this sort. : 18 l l 19 l MR. ANDERSON: They deal with I&C power supplies, l r 20 , as well as this issue. And what we decided to do as to tie ! I 21 all these old issues together and intograte them into a ; 22 single entity and work that problem. 23 MR. MICHELSON: Is that what you are doing now? () 24 MR. ANDERSON: That's the way we are proceeding. ' 25 MR. MICHELSON: But you aren't going to move I l ACE. FEDERAL REPORTERS, INC.
-- m fr~1- ___---_.m---___m__------____
1350 31 07 385 (3 w/DAvbw 1 anything back to A46. 2 MR. ANDERSON: Oh, no. We are talking about I&C 1 f 3 power supplies. 4 MR. MICHELSON: Just in the IEC power supply 5 area? [
- 6 MR. ANDERSON
- So the I&C power supply concerns 7 that we have better fit in this overall program of T&C j 8 power, which we think we can clean up a lot of these old 9 issues and get some of these resolved.
10 MR. MICHELSON: Do you think they will handle 11 the heating and ventilating aspects for AC power systems? 12 MR. THATCHER: I think from one point, inverters
- l 13 are sansitive to heating problems. ,
14 MR. MICHELSON: Yes, they are. Very sensitive. I 15 MR. THATCHER:
~
l That is proposed to be addressed. i j 16 MR. MICHELSON: They also don't like to get i l 17 wet. 1 18 i 19 1 20 21 22 23 1 l 0 24 1
- 25 I
i t ACE. FEDERAL REPORTERS, INC.
- __---- __ ----- --- _----- -----_ f3DT DO - - - - N N-
i l 1350 32 01 386 DAVbur 1 MR. THATCHER: Particularly when you put a fan 2 and blow water over it. l l 3 MR. MICHELSON: That was overlooked. 4 MR. THATCHER: B&W went so far as to annunciate 5 when they lost IEC power. 6 MR. ANDERSON: I was surprised to learn that 7 sometimes they put large floor fans in front of the 8 inverters to keep them cool. Of course, they found they 9 shouldn't put them next to eye wash fountains. I 10 DR. OKRENT: Look, I am trying to think of how we 11 can accomplish future review effectively and efficiently. 12 It seems that we can't have a meeting the day before the O 13 April meeting. 14 Isn' t there something else going on early in l 15 April? 16 MR. MICHELSON: The day before there is also an l 17 ECCS meeting, UHI and Westinghouse. l 18 DR. SAVIO: The April 8th is tho Tuesday before l 19 the full committee. 20 DR. OKRENT: But Carl won't be available Tuesday 21 or Wednesday, it seems, and I am not really very anxious to 22 have a meeting without Carl because then we would have , 23 another one with him. 1 24 (]) So it looks like you had better figure on the day 25 before the May meeting for the subcommittee. 1 ACE. FEDERAL REPORTERS, INC. ___ _ F B P3 G 3 _ _ - _ M_ ____ - C2mM - - _ .
c i i i 1350 32 02 387 ('j% L DAVbur 1 MR. MICHELSON: Wednesday is Safety Research, as 2 I recall. 3 DR. OKRENT: Why don't we assume that we can go 4 there for half the day? How is that? 5 MR. MICHELSON: It will depend on what is coming 6 up. 7 DR. OKRENT: It is hard for me. 8 MR. MICHELSON: It is hard to get Chet to move 9 his committee around, too. 10 DR. OKRENT: I have classes Monday and Wednesday, 11 and I can't miss them. That is a problem. 12 DR. SAVIO: There is time scheduled for the April I 13 i full committee, two to three hours. - 14 DR. OKRENT: No, that wouldn't work. I think l'5 that is probably not time well spent. 16 Well, let's assume -- I will talk to Siess. We 17 will see if we can get him for half a day or some other 18 people for half a day. 19 , Aut even so, it seems to me we need to do some 20 preliminary work to make the meeting more effective. 21 For example, we just had time to understand if 22 you move something into USI A-46, just how much of the 23 seismic problem would it solve? Would it really be all of (} 24 it or part of it? 25 And I think you have to tell us, and maybe you s ACE-FEl ERAL REPORTERS, INC.
, 202-347-3700 Nationside Com 80 4 336 4 646
I l 1350 32 03 388
/ i l 1 'DAVbur 1 can think on that.
2 Well, in other words, we might in effect, without 3 too much effort now, since you have gone through all of this 4 within the staff, indicate which parts of systems 5 interaction are going to be dealt with in other activities 6 and bound them. Don' t j ust say seismic. If it is part of 7 seismic, which parts will be dealt with via, for example -- 8 well, go back and looksee what the staff has in mind. 9 When you get around to getting Speiss' agreement 10 as a specific thing for each and every plant, you may get a 11 no-no.
-- 12 MR. ANDERSON: He is aware of what we are doing.
13 MR. THATCHER: I think he is aware of it, but.I 14 do have to admit that there may be some discontinuity. That 15 bothers me. 16 MR. ANDERSON: We have met with Speiss and IDCOR I 17 i just last week on this very subject. l 18 DR. OKRENT: What did he say? 19 l MR. ANDERSON: I don't know whether I should 20 repeat it. They are not too happy because one of our basic 21 conclusions is that systems interaction reviews be plant 22 , specific, and you are not going to find them on a generic I 23 basis. 24 MR. THATCHER: We don't understand how you can be [) 25 l doing a search for outliers and not deal with it. ACE-FEDERAL REPORTERS, INC. 202 307.2700 mwide fwerage MM356646 l
l l l 1350 32 04 389 L>DAVbur 1 DR. OKRENT: I don't either. Nevertheless, 2 there seems to be some trend to try to implement this 3 policy. 4 That is why I said surrogates, and so forth. You 5 can do certain things with surrogates, but this you can't 6 do. 7 MR. ANDERSON: That is one of our basic 8 findings. 9 So I think what we need to do is go through our 10 technical conclusions and findings and the basis for them 11 before we get started on this, and that way you will 12 understand. O 13 I DR. OKRENT: Maybe you can do that. O'n the other 14 hand, Carl, maybe you could take a look at their summary I 15 recommendations. 16 MR. MICHELSON: That is what I did before I came 17 here. j 18 l DR. OKRENT: I know, but make a list then of the l 19 l things that it is not clear to you are being covered. I I ! 20 MR. MICHELSON: That will be a very long list and 21 will- take a lot of time. It really needs discussion as you 22 go because things can drop off the list, depending on what 23 the answer is to the previous question. 24 DR. OKRENT: (]} Then at least categories, and so 25 forth. l l l ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nuionwide coq M
1350 32 05 390 l . JDAVbur 1 MR. MICHELSON: I can give you many examples. 2 DR. OKRENT: All right. I don' t care, even by 3 examples. Get it to Savio. 4 MR. MICHELSON: This is more than I have got time 5 to do right now. If you want anything in detail, I think we 6 need to sit down with these gentlemen and talk about it. 7 DR. OKRENT: Is there something you can do in the 8 next four weeks? I 9 MR. MICHELSON: I am not sure because about two 10 of those weeks I am going to be gone to a wedding. I 11 l DR. OKRENT: Are you flying anywhere?
,_ 12 MR. MICHELSON: I am going to be down there, and '~'
13 I am not about to be worrying about systems interaction. I 14 DR. OKRENT: Well, all right. 15 ' What I have in mind is the sort of thing that you 16 should be able to do in a couple of hours with that document 17 in your hands, okay? I don't have in mind a month's work. 18 l MR. MICHELSON: I can give you a list with the { 19 flavor. 20 DR. OKRENT: Because what I want to be able to do 21 is transmit it to the staff so again they come into the 22 subcommittee meeting prepared to say either it is here or it l l 23 is there or we don't think or we can't do it or it's too l l
^ I
[' 24 i hard, or whatever, with some more words than that, again to 25 try to get a fairly effective meeting. ACE-FEDERAL REPORTERS, INC. 202.M7 fiit NottrTPMfoverage N336 6646
1350 32 06 391 I /) DAVbur l- 1 Okay? Is that all right? 2 MR. MICHELSON: We are talking about another 3 month? 4 DR. OKRENT: Yes, and we are shooting for a 5 meeting the Wednesday before the May meeting, and I will 6 have to talk to Siess. I will tell him if he wants me to 7 come to a part of the meeting he is going to have to give up 8 part of the meeting, and so on. Otherwise, I won't come in l l 9 Wednesday at all, and it won't be scheduled beginning 10 supposedly at 6:00. 11 Okay? 12 MR. HERNAN: What about us taking this to CRGR? (1J 13 DR. OKRENT: Well, your memo said that in fact we 14 have agreed that you should. On the other hand, it also 15 says that we have to notify you that we shouldn't send it 16 out for public comment if we want to make a comment? 17 MR. HERNAN: I think we can agree. 18 l DR. OKRENT: Can we agree? Can we just agree 19 now? 20 All right, because it is clear that there is 21 interest, rig ht? l l 22 Okay. So then we will assume if CRGR gets to it, 23 okay, you will send it there but you won't send it out for () 24 ! public comment, and we will try to expedite. 25 MR. BOSTAK: Before the meeting, we will tell ACE FEDERAL REPORTERS, INC. 202-347-ET0 Nationtifh Coverage GiMEM""3
1350 32 07 392 u /DAVbur 1 them what happened here at this meeting. 2 DR. OKRENT: Okay. 3 MR. MICHELSON: As a simple example, just to give 4 you one more thing to think about, if the staff makes a very 5 strong claim that the pipe break analysis.-- and they went 6 through that whole ritual and everybody is happy, fine. 7 If you look back carefully, though, at the rules, 8 one of the rules that we do, for instance, is that you do 9 not analyze for adverse system interaction. You only
'10 analyze for loss of function.
11 If the thing gets wetted down, take it out of 12 your functional availability and make sure you can still 'O 13 shut down. We have never required that you go in and 14 determine the effect of wetting that piece of equipment down 15 in terms of what it might now do. 16 Think about that. Maybe I am wrong. If I am 17 wrong, show me. 18 ! But I don't think I am, and I don' t believe that 19 is the kind of analysis that was done, and I think that is 20 one of the things that has got to be done for system 21 interaction, and it is not even addressed in here other than 22 to say the job is done. 1 23 The pipe break analysis under the Giambusso () 24 letter, everybody is happy. 25 DR. OKRENT: If I could add -- ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverc~] 800-336-5646
1350 32 08 393 ,\ E )DAVbur 1 MR. MICHELSON: I don' t think it was ever done. 2 MR. THATCHER: Can I ask a little bit on that, 3 Mr. Michelson? 4 How many failure months do we look at? 5 I mean, I can take an air system and I can have 6 an infinite spectrum of failure modes. 7 MR. MICHELSON: Think of the example. I just 8 said I broke a pipe and I wetted down a piece of equipment. 9 Let's say it is an inverter. 10 The analysis did not require that you determine 11 the effects of the inverter being wetted but rather that you 12 remove l't from the functional availability and assure you U-s 13 still have enough functional equipment left to shut down l 14 safely. , l 15 All I am saying is what is the effect of wetting 16 down that inverter, and it is a potential system 17 interaction. 18 < MR. THATCHER: And I say that wetting down that 19 inverter -- you could put three drops on it, it is going to 20 'do this; you put four, you can do this; you put five -- how 21 , do yov do this? Y : 22 e MR. MICHELSON: You have not addressed system 1 23 ' interactions if you are going to ignore such examples. If (} 24 you don't know how to handle it, that is a different 25 matter. ACE. FEDERAL REPORTERS, INC. 202 347.F00 Nationwide Om 900 33MM6
1350 32 09 394
/ I h DAvbur 1 I just say, just say we don't and forget about 2 doing it.
3 One way is to protect the equipment. 4 MR. THATCHER: I think that was our approach. I 5 thought that is what we were talking about. 6 MR. MICHELSON: Not unless that piece of 7 equipment turned out that you could not safely shut it 8 down. 9 But you didn't look at it from the viewpoint of action, just loss of function. If you couldn' t af ford the 10 l 11 loss of function, you have to protect it. If you could 12 , afford the loss of function, you didn't protect it. O 13 But you never analyzed it for anything I want the 14 function for. 15 That is what we call systems interaction, and you 16 will say, okay, do that loter. You guys have got a USI, you 17 figure it out. 18 j I don't find this addressed in the document at 19 all. 20 [ DR. OKRENT: If I could request one more thing, I 21 think the first thing was this comment that you don't need a 22 new SRP; everything is already in the SRP. 23 I think it would be helpful, if it is not asking () 24 too much, to be shown this. 25 Is that a fair request or even only a little ACE-FEDERAL REPORTERS, INC. - _ _-___ __-_- __-__-_ ___ _ FRPT1MR____--__ M ___-___-_ M ___--_-_---_- -
I 1350 32 10 395 /~s l DAVbur 1 unfair? 2 MR. THATCHER: I missed the question. What is in 3 the SRP? 4 DR. OKRENT: Is it already in the SRP but just 5 needs to be pulled together? 6 MR. ANDERSON: I don't know that we can in any 7 reasonable time go through every SRP, but we can at least 8 give you a sampling of it, a heavy sampling. 9 DR. OKRENT: In fact, that might be useful 10 because when one looks at the problem, when one looks at the 11 words of your samples and j udge, in effect, the adequacy as 12 the words in the SRP. So do a reasonable effort here. O 13 MR. ANDERSON: What worries me is that we really 14 weren't intending to add a whole lot of new guidance. What 15 we were intending to do is to replace that guidance through 16 all the standard review plans in a single place, which might 17 have some merit, but we weren't addressing it from the 18 i standpoint really of taking care of it. 19 DR. OKRENT: Have you pulled those together, 20 though? Did you have a draft o* the new SRP section ever? 21 MR. THATCHER: Yes. l 22 DR. OKRENT: Show what is done. You have done it 23 then,' in effect. Supposedly, that is either the old SRP or 24 it is not. 25 MR. MICHELSON: I think we have to decide whether , l l ACE. FEDERAL REPORTERS, INC. ; 202 347 4 700 Nationwade Coverase 900-13dWinds i
l 1 l l 1350 32 11 396 i m- 'DAVbur 1 or not this issue has been resolved. 2 DR. OKRENT: That is the number one question. 3 MR. MICHELSON: I would think it is going to take 4 a little while and some careful dialogue to see if we 5 understand. 6 I think you guys have done a good job, but I 7 think it is almost an impossible problem. But that doesn' t 8 remove it f rom the list just because it is im possible . It 9 just makes it a little more difficult for future 10 resolution. 11 DR. OKRENT: Look, wa said we were going to 12 adjourn around 8:00 o' clock. It is around 8:00 o' clock. t . .' ' ' 13 , I want to thank you for being so cooperative at i 14 this late hour, and we wish the Division directors were as 15 supportive of your work as we are. 16 l (whereupon, at 8 :05 p.m. , the subcommittees were 17 adjourned.) 18 19 j 20 l 21 22 I l 23 l l , 24 l 25 I l l ACE-FEDERAL REPORTERS, INC. 202 447 1 709 Nationwide Coverage 8004 4 % M
(-_...---...-_-.-.--.-----._--_._--..-.- 1 4
/
i l@ l l l l' l i i 4 ' i l ! 1 i !' i i I !
; i 1
a l i
' I DISCUSSION WITH THE ACRS SUBCOMMITTEE 1 .
! i 1 ON , I t SAFETY GOAL POLICY ! 9 ! MARCH 12, 1986 ; i : 6 l ? , f k i b e i. h I e l I b i i I l i l I l.
l () HISTORY OF PRA AND SAFETY GOALS ' e REACTOR SAFETY STUDY (WASH-1400) i e KEMENY COMMISSION AND NRC RESPONSE ;
! e PUBLIc WORKSHOPS 1981 e PROPOSED SAFETY GOAL POLICY STATEMENT FOR PUBLIC COMMENT FEB, 1982 e FOUR PUBLIc MEETINGS 1982 l
e REVISED POLICY STATEMENT MAPCH 1983 i ! e 2-YEAR EVALUATION PERIOD o STEERING GPOUP REPORT i l e ACRS .iND STAFF COMMENTS / MEETINGS t e FEBRUARY 14, 1986 STAFF PAPER , i l
O STAFF PROPOSAL OF FEBRUARY 14, 1986 e PURPOSE - ASSIST COMMISSION IN ITS DELIBERATIONS REGAPDING THE BEST COURSE OF AGENCY ACTION e BELIEVE IT ACCOMMODATES ACRS COMMENTS AND ALL STAFF COMMENTS ON STEERING GPOUP REPORT () e RECOMMEND ISSUANCE OF OUALITA IVE STATEMENTS IN FINAL FORM AS COMMISSION SAFETY GOAL POLICY e RECOMMEND AUTHOPlZING STAFF TO USE INTEGRATED MATRIX ON A TRIAL BASIS AS A QUANTITATIVE MEASURE OF QUALITATIVE SAFETY GOALS e ALL PROGRAM 0FFICE DIRECTORS (NRR, RES, IE, NMSS) SUPPORT THE FEBRUARY 14 PROPOSAL e STAFF PREPARED TO IMPLEMENT COMMISSION DECISION O
STAFF CONCLUSIONS i e ISSUE IN FINAL FOPM, AS THE COMMISSION'S SAFETY GOAL POLICY, i Two QUALITATIVE STATEMENTS REGAPDING INDIVIDUAL AND S0cIETAL i RISK. ELIMINATE QUANTITATIVE OBJECTIVES AND BENEFIT-COST GUIDELINE O e AS INDIVIDUALLY DISCPETE STATEMENTS. COMBINE THEM INTO AN l INTEGRATED MATRIX WHICH THE STAFF CAN USE AS A QUANTITATIVE i l MEASUREMENT OF THE COMMISSION'S QUALITATIVE SAFETY GOALS. 1 1 l i l i ,
- O i i
1 _ _ _ _ . - . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ - - - _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ . _ _ _ - ___________]
'1 O QUALITATIVE SAFETY G0ALS 4 INDIVIDUAL RISK G0AL - INDIVIDUAL MEMBERS OF THE PUBLIC SHOULD BE PROVIDED A LEVEL OF PROTECTION FROM THE
; CONSEQUENCES OF NUCLEAR POWER PLANT OPERATION SUCH THAT INDIVIDUALS BEAR N0 SIGNIFICANT ADDITIONAL RISK TO LIFE AND i HEALTH l
O : 0 SOCIETAL PISK G0AL - SOCIETAL RISKS TO LIFE AND HEALTH FROM NUCLEAR POWER PLANT OPERATION SHOULD BE COMPARABLE TO OP LESS THAN THE RISKS OF GENERATING ELECTRICITY BY VIABLE COMPETING TECHNOLOGIES AND SHOULD NOT BE A SIGNIFICANT l I ADDITION TO OTHER SOCIETAL RISKS l I b '
o o i O i l IllTEGRATED SAFETY GOAL KCIS1011 MATRIX M ELT. EALTII trtELis AIIB COST-BENEFIT)* COST BENEFIT IAEALTH EFFECTS 30.11/RY ($1,000/e-a + AVERTED ,; LARGE-SCAR.E CORE l Tti FnEcuENCY (pea RY1 PROMPT / LATENT ONSITE COST)
.i e,
HEET seTM NO Flx
<10-5 FIx ($1,000/r-a) d I Don *T MEET ONE i
i MEET sotu Fax ($1,000/r-a + 1 -> 01 AOSC) ., 10-9 5 Flx ($1,000/r-a +1001 AOSC)
~ son *T nEET ONE HEET Bo1H FIx ($1,000/P-a + 10 --> 11 AOSC) 10-3 % Flx ($1,000/P-a +1001 AOSC) nou'T MEET ONE I
\ ~ MEET toTM Fsu ($1.000/P-a +1001 AOSC)
>10-3 son'T MEET ONE Flx (Cosi no LintT) l)
1 1
'ALL VAltlES AltE TAKEll AS EAll VALIIES t _ _ - _ _ _ _ _ . __
a
%)
PARAMETERS REFLFCTED IN MATRIX e HEALTH EFFECTS PROMPT FATALITY RISK - THE RISK TO AN AVERAGE INDIVIDUAL IN THE VICINITY OF A NUCLEAR POWER PLANT OF PROMPT FATALITIES THAT MIGHT RESULT FR0n REACTOR ACCIDENTS DOES NOT EXCEED ONE-TENTH OF ONE PERCENT (0.1%) 0F THE SUM OF PROMPT FATALITY RISKS RESULTING FROM OTHER ACCIDENTS TO WHICH MEMBERS OF THE U.S. POPULATION ARE GENERALLY EXPOSED. LATENT CANCER FATALITY RISK - THE RISK TO THE POPULATION IN THE AREA NEAR A NUCLEAR POWER PLANT OF CANCER FATALITIES THAT MIGHT RESULT FROM NUCLEAR POWER PLANT ([) OPERATION DOES NOT EXCEED ONE-TENTH OF ONE PERCENT (0.17) 0F THE SUM OF CANCER FATALITY RISKS RESULTING FROM ALL OTHER CAUSES, e LARGE-SCALE CORE-MELT FREQUENCY - THE LIKELIHOOD OF A NUCLEAR REACTOR ACCIDENT THAT RESULTS IN A LARGE-SCALE CORE-MELT IS ALLOWED TO VARY. THE IMPETUS TO IMPROVE THE CORE-MELT FREQUENCY IS DEPENDENT UPON HOW MUCH THE FREQUENCY IS GREATER THAN 10-5/RY AND THE STATUS OF THE HEALTH EFFECTS. e BENEFIT-COST GUIDELINE - THE BENEFIT OF AN INCREMENTAL REDUCTION OF MORTALITY RISKS AND IN THE FREQUENCY OF LARGE-SCALE CORE-MELT ACCIDENTS IS COMPARED WITH THE ASSOCIATED COSTS ON THE BASIS OF $1,000 PER PERSON-REM AVERTED PLUS A PERCENTAGE OF THE ONSITE RADIOLOGICAL, INCLUDING ECONOMIC, COSTS AVERTED.
~m l (v
4 KEY FEATURES OF SAFETY G0AL POLICY STATEMENT 4 1 0 TWO QUALITATIVE SAFETY G0ALS REMAIN UNCHANGED
~
0 THE QUANTITATIVE DESIGN OBJECTIVES AND BENEFIT-COST GUIDELINES ELIMINATED AS INDIVIDUALLY DISCRETE j STATEMENTS OF EXPECTATION l 0 QUALITATIVE SAFETY G0ALS INCORPORATED INTO THE
) REGULATORY DECISIONMAKING PROCESS THROUGH AN J
INTEGRATED SAFETY G0AL MATRIX ! O MATRIX COMPRISED OF QUANTITATIVE PARAMETERS FOR FREQUENCY OF CORE-MELT ACCIDENTS, PROMPT AND LATENT CANCER FATALITY RISKS AND A BENEFIT-COST GUIDELINE I l 4
- i I' ;
($) ; i ! i i KEY FEATURES OF SAFETY G0AL POLICY j i I ! . r O POLICY STATEMENT INDICATES SAFETY GOALS ARE NOT A j
- SUBSTITUTE FOR EXISTING REGULATIONS i
i j t POLICY STATEMENT EXHORTS ACHIEVING A MEAN CORE-MELT i i i FREQUENCY OF LESS THAN 1 IN 10,000 PER REACTOR YEAR FOR i CURRENT PLANTS IN ACCORD WITH 7/85 ACRS RECOMMENDATIONS i ! i . ($) 0 THE POLICY STATEMENT INDICATES THAT FUTURE PLANTS WOULD l , i BE EXPECTED TO ACHIEVE A CORE-MELT FREQUENCY LOWER THAN j i , CURRENT PLANTS i \ I J 0 POLICY STATEMENT INDICATES THAT THE STAFF SHOULD s j ATTEMPT TO ENSURE THAT CORE-MELT FREQUENCY AND l MORTALITY RISKS ARE NOT DRIVEN BY A SINGLE ACCIDENT 1 SEQUENCE (MANAGEMENT OF UNCERTAINTIES) 5 I ($) l i
KEY FEATURES OF SAFETY G0AL POLICY POLICY STATEMENT, THROUGH THE INTEGRATED MATRIX, ) 0 ESTABLISHES A DE MINIMUS RISK LEVEL AND ALSO A RISK LEVEL ABOVE WHICH A FACILITY SHOULD HAVE A SENSE OF URGENCY FOR SAFETY IMPROVEMENTS OR SHUTDOWN,
! O POLICY STATEMENT, THROUGH THE INTEGRATED MATRIX, 4
ESTABLISHES AN ALARA APPROACH TO AID STAFF SAFETY DECISIONS 4 BENEFIT-COST GUIDELINE EMB0 DIED IN MATRIX EMPHASIZES '[]) THAT ACHIEVEMENT OF THE MORTALITY RISK OBJECTIVES IS THE PRIORITY CONSIDERATION IN THE SAFETY G0AL CONCEPT (PUBLIC HEALTH AND SAFETY EMPHASIS) 0 BENEFIT-COST GUIDELINE EMPHASIZES A SLIDING EXPENDITURE SCALE TIED TO CORE-MELT FREQUENCY AND DRIVES TOWARDS A REDUCTION IN CORE-MELT FREQUENCY (ACCIDENT PREVENTION , EMPHASIS) ^ 0 POLICY STATEMENT RECOGNIZES THAT MAINTENANCE, OPERATIONS AND MANAGEMENT MAY HAVE A SIGNIFICANT IMPACT ON THE STATE OF PLANT RISK (RISK MANAGEMENT EMPHASIS) ()
4 i f C')
KEY FEATURES OF SAFETY G0AL POLICY 0 POLICY STATEMENT INDICATES THAT WE ARE ABOUT READY SAFELY G0ALS (MATRIX) IN THE REVIEW 0F GENERIC SAFETY REQUIREMENTS - STILL NEED TO DEVELOP METHODOLOGY FOR l ACCOUNTING FOR INTEGRATED SAFETY IMPROVEMENT IMPACT CORE-MELT FREQUENCY AND DEVELOP AN AGGREGATE COST SYSTEM, OTHER AREAS WILL BE IMPLEMENTED GRADUALLY J
0 POLICY STATEMENT INDICATES THAT USE OF THE SAFETY G0AL INTEGRATED MATRIX WILL BE USED IN CONJUNCTION WITH D MINISTIC DECISIONMAKING 7 O POLICY STATEMENT RECOGNIZES THAT UNCERTAINTIES MUS 9 ! INTO ACCOUNT IN THE REGULATORY DECISIONMAKING PROCESS S POLICY STATEMENT, THROUGH THE INTEGRATED MATRIX, EMPHASIZES THE NEED TO STRIVE FOR SAFETY IMPROVEMENTS WHICH ARE EFFECTIVE 9 POLICY STATEMENT EXPRESSES FUTURE INTENT TO ESTABLISH OUANTITATIVE OBJECTIVES TO GUIDE CONTAINMENT DESIGN AN PERFORMANCE, STAFF SHOULD PROPOSE THEM BY EARLY FY '87 9 INCLUDES INTERNAL AND EXTERNAL INITIATORS
l I i TREATMENT OF UNCERTAINTIES (]) o UNCERTAINTIES NOT GENERATED BY PRA METHODOLOGY , 2 e PEER REVIEW 0F PRA RESULTS i l, o UPPER B0UNDS AND MID RANGE VALUES CAN BE INFERRED FP0M EXPERIENCE l
- e USE OF MEAN INTRODUCES A CONSERVATIVE BIAS AND MAY l l APPROACH THE UPPER CONFIDENCE LIMIT i
! () e l'SE WITH UNDEPSTANDING 0F MAGNITilDE OF UNCERTAINTIES
- INVOLVED - LOOK TO ROOT CAUSE CONTRIBUTORS j e DISPLAY MEAN, MEDIAN AND CONFIDENCE LEVEL RANGES i
e STRIVE FOR BALANCED CONTRIBilTIONS FROM DOMINANT SEQUENCES AND MINIMIZE UNCERTAINTIES BEING DRIVEN BY SINGLE SEQUENCES i i s l l
- O AREAS OF IMPLEMENTATION l
! i J
- i I
i e GENERIC ISSUES - ABOUT READY J 1 l
- i l
l e PLANT-SPECIFIC BACKFITTING/ EXEMPTIONS i i [, 1 l e SETTING REGULATORY PRIORITIES i i I 1 1 !O l
- GENERIC issues l
t l .
- RESEARCH PROJECTS i !
l l l REGULATIONS / REGULATORY GUIDES :
- INSPECTION ACTIVITIES 1 t ; f i
<lO l i ; l ! i
ti O NRR STAFF PRESENTATION TO THE ACRS .
SUBJECT:
CONTAINMENT PERFORMANCE OBJECTIVE (CP0) DEVELOPMENT: STATUS
SUMMARY
'0F NRC STAFF WORK DATE: MARCH 12, 1986 tC3RESENTER: GEORGE SEGE PRESENTER'S TITLE / BRANCH /DIV: SENIOR REACTOR SYSTEMS ENGINEER REGULATORY IMPROVEMENTS BRANCH 1 DIVISION OF SAFETY REVIEW 8 OVERSIGHT l PRESENTER'S NRC TEL. NO.: (301) 492-9640 SUBCOMMITTEE: SAFETY PHILOSOPHY, TECHNOLOGY AtlD CRITERIA O 1 l l
Me O WORK STATUS 4 l
- PLAN SUBMITTED TO ED0
- RES/NRR EFFORT, BNL SUPPORT ,
PAPER TO COMMISSION EARLY FY 87 . 15 INTERVIEWS
- 3 NRC, 3 ACRS, 4 INDUSTRY, 2 CRITICS, 3 OTHER 5 00ANTITATIVE CPO OPIIONS FOR STUDY
[ ALSO: NO-ACTION OPTION; OUALITATIVE OPTION
- PROS, CONS, MODS i
STARTED COMPARISON OF OPTIONS VS. EXISTING CON- ; l TAINMENT PERFORMANCE ASSEESSMENTS (BY INSPECTION; j NO DETAIL) IMPLEMENTR110N APPROACH PART OF OVERALL SG IMPLEMENTATION: ONE FACTOR AMONG OTHERS; NOT "G0/NO-GO." i
- INFLUENCE FUTURE PLANTS; LTD. FIXES FOR EXISTING PLANTS M EMETkTIN 1
l t l l 1
() OPTION 1: SIMILAN 10 1980 ACRS . l DESCRIPTION: l CO'NDITIONAL PROBABILITY OF LARGE-SCALE UNCONTROLLED RELEASE *, l GIVEN LARGE-SCALE CORE MELT: l - FOR EXISTING PLANTS: d; 0,1 ' FOR FUTURE PLANTS: LOWER, PER L/B
*) DEFINED BY ALHS AS 10% IODINE, 90% NOBLE GASES f
PR0: RELATES DIRECTLY TO BASIC CONTAINMENT FUNCTION SIMPLE, EASY TO UNDERSTAND HAS DOCUMENTED GENERAL BASIS
~~
CON: PROBABILITY DIFFICULT TO ESTIMATE ([) SINGLE PROBABILITY AND SINGLE RELEASE SPEC ARE INSENSITIVE TO IMPORTANCE OF RELEASE TIMING & COMPOSITION, CONTAINMENT DESIGN, POPULATION LENDS TO DISCOURAGE REDUCTION OF FREQUENCY OF j MODFRATE-SCALE RELEASES l l I =
\
l 1
O OPITON 2: FISSION-PR00ULI ATTENUATION DESCRIPTION: REDUCTION FACTOR DUE TO CONTAINMENT FOR AIR-BORNE RADIONUCLIDES THAT WOULD DEPOSIT AS SOLIDS OUTSIDE 2
- CONTAINMENT: 2m 10 MEAN FREQUFNCY OF MAJOR RELEASE OUTSIDE' CONTAINMENT:
OVER 1% OF SOLIDS es10-6fpy l OvER 1% OF NON-SOLIDS d;10-5/RY l PRO: CHARACTERI7ES CONTAINMENT PERFORMANCE DIRECTLY IN TERMS OF ITS BASIC FUNCTION OF CONTAINING THE RADIOACTIVE FISSION PRODUCTS. ! INTERNAL FP REMOVAL MECHANISMS RECOGNIZED. CON: ASSESSMENT SUBJECT TO CHANGES IN MATURING TECHNOLOGY OF SOURCE-TEPM PREDICTIONS. (]} * ! RELIES ON MODELS OF UNCERTAIN FIDELITY FOR CONTAINMENT LEAKAGE AND FP PLATE-0UT.
~
SOLIDS /NON-SOLIDS DISTINCTION MAY BE TOO CRUDE A MODEL FOR RISK CHARACTERISTICS. NON-SOLIDS SPEC MAY BE TOO STRICT. i l l l s v l -
Q OPTION 3: CURIE REDilClION l>ESCRIPTION: LIMIT CURIES RELEASED WITHIN 72" HOURS AFTER LARGE-SCALE CORE MELT:
- fEl% OF TOTAL CURIES THAT WOULD BE RELEASED WITHOUT CONTAINMENT. ,
PR0: EMPHASIZES RADIOACTIVITY-CONTAINING FUNCTION BF CONTAINMENT r RELATIVELY SIMPLE STATEMENT THAT CAN BE EASILY COMMUNICATED J TO THE PUBLIC. . 9
- l CUN: NO PROVISION FOR CREDIT FOR LOW CORE-MELT PROBABILITIES.
KILLS TRADE-OFFS. _. NO CLEAR NEXUS TO OVERALL LEVEL OF PLANT SAFETY O I a l l -( 85 I
+
.l
- O
_ _ _ . . _ _ _ _ _ - ~ _ - , . _ _ _ _ - - _ _ _ _ _ _ , , - _ _ . . - _ . _ _ _ . _ _ _ __.
i! '
. s .
' *~
gss .s UFlION 4: FLEXIBLE APPH0ACH v ,. v (. DESCRIPTION: , ea, FREQUENCY OF MAJOR RELEASE FROM CONTAINMENT da10-6/RY CONDITIONAL PROBABILITY OF EARLY CONTAINMENT FAILUPE, GIVEN LARGE-SCALE CORE MELT: 450.1 REGARDLESS OF OTHER FACTORS.
;I PK0:
PROVIDES DESIGN FLEXIBILITY, WITHIN LIMITS INTENDED TO PRESERVE DEFENSE IN DEPTH. CREDIT IS GIVEN FOR LOW CORE MELT PROBABILITIES. TIED TO OVERALL PLANT SAETY GOALS. 5RELATIVELY SIMPLE TO ASSESS. (FEWER PARAMETERS TO ASSESS.) CON:
.. STRUCTURED ESTIRELY IN TERMS OF PROBABILITY, WITH NO
' RISKINESS DISTINCTIONS ABOVE " MAJOR RELEASE" THRESHOLD o ADDRESSES ONLY EARLY CONTAINMENT FAILURES; DEPENDS ON
(]) ,
' INDIRECT IMPLICATION FOR LONG TEPM.
t
.)
\ &
=
i t 4 O ' 4 1
'i
( l OPTION 5: MECHANISitt DESCRIPTION: NO SUBSTANTIAL LEAKAGE FOR "24" HOURS NO GROSS STRUCTURAL OR MAJOR PENETRATION FAILURE FOR "72" HOURS
~
ABILITY TO INTERVENE WITHIN "72" HOURS MAXIMUM PRACTICAL ASSURANCE THAT SYSTEMS ON WHICH CONTAINMENT EFFECTIVENESS DEPENDS ARE ALWAYS OPERATIONAL. PRO: DEFENSE IN DEPTH AGAINST WEAKNESSES OF PRA DEPENDENCE, ESPECIALLY OVERSIGHTS. SUBJECT TO EVALUATION BY CONVENTIONAL ENGINEERING ANALYSIS ~ PARTLY TESTABLE: MAJOR LEAK PATHS EASILY RELATED TO LEAKAGE MONITORING WITH SLIGHT POSITIVE OR NEGATIVE PRESSURE IN NORMAL OPERATION. CON: DEGREE OF ASSURANCE NOT COVERED BY NON-PRA-BASED APPROACH. DIFFICULT TO RELATED TO EXTENT OF FISSION PRODUCT RELEASE AND TO MORTALITY RISK OBJECTIVES. PRA STILL NEEDED ON CRITERIA FOR NEGLECTING LOW FREQUENCY EVENTS. l l 4 O
() NO-ACTION OPTION
~
DESCRIPTION: NO CPO WOULD BE PROMULGATED. . PR0: RETAINS RELATIVE SIMPLICITY OF CURRENT SET OF QUANTITATIVE OBJECTIVES. . DOES NOT ADD REDUNDANCY TO CURRENT QUANTITATIVE OBJECTIVES. RETAINS GREATER DESIGN FLEXIBILITY., AVOIDS DIFFICULTY OF SUM ABLY QUANTIFYING CONTAINMENT GOALS. - CUN: FAILS TO REINFORCE DEFENSE IN DEPTH. FAILS TO FOCUS ON CONTAINMENT AS LAST CLEAR BARRIER TO PUBLIC IMPACT AND ONLY MAJOR FEATURE THAT IS THERE EXCLUSIVELY FOR PUBLIC SAFETY. O t - _ _ _ _ _ - _ _ -
l I i O UPil0N 6: QUALITATIVE DESCRIPTION: CONTAINMENTS SHOULD SURVIVE WITHOUT GROSS FAILURE ALL BUT THE VERY UNLIKELY OF THE SEVERE ACCIDENTS. CONTAINMENT LEAKAGE AND VENTING DURING AND FOLLOWING SEVERE
. ACCIDENTS SHOULD BE LIMITED TO SPECIFIED LEVELS. <
THE PROBABILITY OF OCCURRENCE OF EVENTS WHICH BYPASS THE CONTAINMENT SHOULD BE VERY LOW.
. PR0: ARTICULATES POLICY BUT AVOIDS DIFFICULTY OF SUITABLY QUANTIFYING CONTAINMENT GOALS;
- DEFENSE IN DEPTH AGAINST WEAKNESSES OF PRA DEPENDENCE, ESPECIALLY OVERSIGHTS.
i P.10VIDES GENERAL GOALS FOR DETAILED CONTAINMENT PERFOR-MANCE CRITERIA (00 ANT. & QUAL., WITH RECOGNITION OF t ALTERNATIVE CONTAINMENT DESIGNS). ~ O CON: LACKS PRECISION OF A OVANTITATIVE CPO. CONSISTENCY AND COHERENCE OF CONTAINMENT REQUIREMENTS LESS CLEARLY GUIDED. l l O I 1
() NtXT STEPS
~
COMPLETE LTD. COMPARISONS (h- UPil0NS VS EXISTING - APRIL CONTAINMENT ASSESSMENTS: DRAFT ON OPTIONS, IMPLEMENTATION APPROACH, ISSUES: APRIL WORKSHOP: HARPERS FERRY, MAY 12 - 13 t ACRS SUBCOMMITTEE: JUNE 4
- DRAFT COMMISSION PAPER FOR STAFF COMMtNi~: JUNE i
ALSO TO ACRS i ACRS REVIEW: JULY 10 OR 11 , ) _. i ] i 4 b l
1 . - ' ~ - O O O ,
. . . . ..~ . . -. _ - . _ _ _ _ . . . . . . . . _ . . . . . . . . . _ . . _
f' STATUS OF DEVELOPMENT OF CONTAINMENT VENTING EMERGENCY PROCEDURE GUIDELINES D
. _ _ _ . _ _ _ _ - . - - . -, - - - -~a - r - -- -w m--
U; l 0, O, C H R O N O _.O GY ;-
+ BWR OWNER'S t,ROUP FORMED A COMMITT EE TO DEVELOP GENERIC GUIDELINES FOR EMERGENCY -PROCEDURES IN AUGUSF 'l 9 /9
+ STAFF ISSUE D FIRST SER Af' PROVING REV 2 IN FEBRUARY, I 983 ONE OPEN ITEM WAS CRITERIA FOR DEFINING CONTAINMENT VENTING PRESSURE
+ REV 3 APPROVED IN NOVEMBER,1983. CRITERIA FOR DEFINING CONTAINMENT VENTING PRESSURE STILL OPEN
+ SEVERE ACCIDENT WORK ON LIMERICK FIRST SHOWED NEED TO VENT WET WELL RATHER THAN DRY WELL
+ DRAFT REV 4 CONSIDERS CONTAINMENT VENTING FOR HYDROGEN CONTROL AND PRESERVATION OF CONTAINMENT INTEGRITY. NOT CURMi l'TE D
O O O . CHR~ONOLOGY ;- i
+ BWR OWNER'S GROUP FORMED A COMMITTEE TO DEVELOP ,
GENERIC GUIDELINES F'OR EMERGENCY -PROCEDURES IN AUGUST ~19 79 ,
+ STAFF ISSUED FIRST SER APPROVING REV 2 IN !
FEBRUARY. 'l 983 ONE OPEN ITEM WAS CRITERIA FOR DEFINING CONTAINMENT VENTING PRESSURE
- REV 3 APPROVED IN NOVEMBER,1983. CRITERIA FOR DEFINING CONTAINMENT VENTING PRESSURE STILL OPEN . .
- SEVERE ACCIDENT WORK ON LIMERICK FIRST !
l SHOWED NEED TO VENT WET WELL RATHER THAN i DRY WELL t
+ DRAFT REV 4 CONSIDERS CONTAINMENT VENTING ,
FOR HYDROGEN CONTF:OL AND PRESERVATION OF i CONTAINMENT INTEGRITY. NOT CURMITTED l l
o
~
O. O STATUS I
$REV 3 LATEST APPROVED VERSIOrl. VENT TO PREVENT CONTAINMENT FAILLIRE. l'A! LURE PRESSURE POORLY DEFINED '
+ RECENI WORK BY OWNEl<'S GROUP, NRC CONTRCATORS/
S I~ aft, IDCC.R SHOW SEVERAL VENTING CONSIDERATIONS:
- 1. VENT THROUGH POOL FOR SCRUBBING
- 2. ASSURE OPERABILITY OF ISOLATION VALVES
- 3. ASSURE OPERARILi'IY OF ADS VALVES
<1 . DEGRADATION OF EQUIPMENT OR PERSONNEL ACCESS DUE TO DUCT FAILURE
- 5. HYDROGEN CONTROL
, + REV 4 'WIL.L INCLUDE THESE CONSIDERATIONS
- PGP's IMPLEMENTED BASED ON REV 3 PLUS FURTHER i PLANT SPECIFIC INFORMATION ,
l + POST--lMPLEMENTATION REVIEW FOR OR'S l
AGENDA INTRODUCTION AND OVERVIEW M. R. EDELMAN PLANT STATUS / RESPONSE AND FOLLOW ON INSPECTIONS R. A. STRATMAN SEISMIC INSTRUMENTATION AND DATA REDUCATION DR. C. CHEN GILBERT COMMONWEALTH SEISMIC DESIGN BASIS DR. C. CHEN GILBERG COMMONWEALTH EARTHQUAKE DESIGN EQUIVALENCY DR. J. STEVENSON 1 STEVENSON & ASSOCIATES () EARTHQUAKE ANALYSIS AND SEISMICITY R. HOLT WESTON GEOPHYSICAL hEISMICITYANDINJECTION WELLS DR. P. TALWANI UNIVERSITY OF SOUTH CAROLINA
SUMMARY
AND CONCLUSIONS M. R. EDELMAN e () I 4
OVERVIEW I i i
- SEISMIC EVENT 01/31/86
- PLANT RESPONSE
- SUBSEQUENT EVALUATIONS
- PHYSICAL PLANT I - DESIGN BASIS
^
- EARTHQUAKE ANALYSIS l
- PRIOR ACRS MEETINGS t
l
- DESIGN CONFIRMED :
- PLANT READY TO LOAD FUEL
\ ! l i i i t i J l i I I " i O I I l
O . PLANT STATUS PRIOR TO SEISMIC EVENT ONGOING TESTING, CALIBRATION, WORK ACTIVITIES PREPARATION FOR DIVISION II DIESEL GENERATOR TESTING
- i STARTUP SOURCES NOT YET MOVED 1
SYSTEMS ENERGIZED (IN OPERATION AND STANDBY MODE) SAFETY SYSTEMS NON-SAFETY SYSTEMS < l O i t 4 ee j l C:) ; I
l CE) . SAFETY RELATED SYSTEMS ENERGIZFD DURING THE SEISMIC EVENT { i SYSTEM DESCRIPTION ( Cll CONTROL ROD DRIVE i C41 STANDBY LIQUID CONTROL C71 REACTOR PROTECTION SYSTEM D17 PLANT RADIATION MONITORS E12 RESIDUAL HEAT REMOVAL E21 LOW PRESSURE CORE SPRAY E22 HIGH PRESSURE CORE SPRAY G41 FUEL POOL COOLING AND CLEANUP M15 ANNULUS EXHAUST GAS TREATMENT M23 MCC, SWITCNGEAR, 6 MISC. AREA HVAC M24 BATTERY ROOM EXNAUST M25 CONTROL ROOM NVAC M26 CONTROL ROOM EMERGENCY RECIRCULATION M32 ESW PUMPHOUSE VENTILATION M40 FUEL MANDLING BUILDING VENTILATION M43 DIESEL BUILDING VENTILATION l
- Pll CONDENSATE TRANSFER AND STORAGE P22 [
j MIXED BRD DEMINERALIZER P41 SERVICE NATER ( P42 EMERGENCY CLOSED COOLING ! P43 NUCLEAR CLOSED COOLING P45 l EMERGENCY SERVICE WATER P47 ! CONTROL COMPLEX CHILLED WATER
- P49 ESW SCREEN NASH P52 INSTRUMENT AIR l
i P54 FIRE PROTECTION C95 ' EMERGENCY RESPONSE INFORMATION SYSTEM P51 SERVICE AIR , R14 110 V AC VITAL INVERTERS R22 METALCLAD SWITCHGEAR , R23 400 V LOAD CENTERS R24 MOTOR CONTROL CENTERS R25 DISTRIBUTION PANELS - 120, 208 & 480 VOLTS R42 D.C. SYSTEM R43 STANDBY DIESEL GENERATOR (SDG) R45 . SDG YUEL OIL R46 SDG JACRET WATER COOLING R47 SDG LUBE OIL l R61 MAIN CONTROL ROOM ANNUNCIATOR l 1
. ~ ___ _ _ _ __-. .... - . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ , _ _ _ , _ . _ _ _ - _ . _ _
! i iO . NON-SAPETY RELATED SYSTEMS ENERGIZED DURING THE SEISNIC EVENT i SYSTEM DESCRIPTION I F42 FUEL TRANSFER EQUIPMENT i G33 REACTOR NATER CLEANUP Mll CONTAINMENT VESSEL COOLING M13 DRYNELL COOLING M21 CONTROLLED ACCESS AREA NVAC M27 COMPUTER ROOM NVAC M35 ! M36 TURBINE BUILDING COOLING & VENTILATION i OFF-GAS BUILDING EXKAUST ' M41 NEATER BAY VENTILATION M45 i N21 CIRCULATING NATER PUMP MOUSE VENTILATION i CONDENSATE M23 l
! CONDENSATE FILTRATION N24 CONDENSATE DEMINERALISERS !
N32 M71 TURRINE CONTROL (ENC) I CIRCUL4 TING NATER l P20 NATER TREATMENT P21 TWO BED DEMINERALIBER () P44 : j P55 TURBINM BUILDING CLOSED COOLING ! BUILDING NEATING P61 AUXILIARY STEAM PS2 ; P72 AUXILIARY BOILER FUEL OIL l l PLANT UNDERDRAIN ! C91 PROCESS COMPUTER i C94 NEALTH PHYSIC COMPUTER P56 SECURITY R11 STATION TRANSFORMERS i R15 l R36 TECNNICAL SUPPORT CENTER UPS R44 NEAT TRACING & ANTI FREEEE PROTECTION SDG STARTING AIR R51 : RS2 INTRA PLANT COMMUNICATIONS ' R53 MAINTENANCE & CALIBRATION f l R57 EECLUSIOW AREA PAGING SYSTEM ! i R71 RADIO & IN-PLANT ANTENNA SYSTEM l LIGNTING sli PONER TRANSFORMER 8 4 841 8TEP UP STATION ' i l
.. 1 \
1 O INITIAL INSPECTIONS AND FINDINGS FOLLOWING THE SEISNIC EVENT OPERATOR SURVEY (NO STRUCTURAL DAMAGE) WALMDOWNS BY PLANT MAINTENANCE PERSONNEL (NO STRUCTURAL DAMAGE) SYSTEMATIC PLANT WALKDOWNS BY 65 ENGINEERS AND TECHNICIANS (NO STRUCTURAL / EQUIPMENT DAMAGE)
- 250 MANNOURS i -
LTILIIED DETAILED CNECRLIST 1~
, - RESULTSt
- 473 OBSERVATIONS j TYPIFIED BY: MINOR MAIRLINE CRACES IN CONCRETE, BURNED OUT LIGET l BULBS, LEARING VALVE FLANGES, LEARING PIPE FLANGES i
- 2 THOUGHT TO BE CAUSED BY EVENT i
O -NEAT EXCHANGE DRAIN VALVE ) FOUND DRIPPING
-TRANSFORNER TRIP
'I
- 95 INDETERMINATE 1
1
-TYPIFIED BY BURNED OUT LIGHT
/ BULBS, LEARING VALVE FLANGES, t LEARING PIPE FLANGES, CRACRED , METER LENSES 1
- 376 NOT RELATED TO EARTHQUARE I
i i 1 f L
t r [ O . SUBSEQUENT SURVEY, INSPECTION, AND STUDY ACTIVITIES SITE SURVEY PLANT SETTLEMENT SURVEY COOLING TOWER WALKDOWN f SEISMIC CLEARANCE INSPECTION ENERGIZED ELECTRICAL EQUIPMENT STUDY RELAYS 6968 1
) -
SWITCHES 6962 1 l - INSTRUMENTS 4721 l POWER SOURCES 6493
} -
TRANSFORMERS 1885 1 I l t I i i 4 I ,I
() FOLLOW ON CONFIRMATORY ACTIVITIES ON GOING SVI's SYSTEMS ENERGIZED AND NOT-ENERGIZED DURING EVENT NEW WORK REQUEST REVIEW PROCEDURE ASSURES CEI/NRC REVIEW OF ALL POTENTIAL EARTHQUARE-RELATED ITEMS TIME FRAME: FEB. 3 THRU 28 TOTAL WR's DURING PERIOD: 2401 622 REQUIRED EVALUATION 459 DETERMINED TO NOT BE EARTHQUAKE RELATED 163 WERE REVIEWED WITH THE NRC PRIOR TO IMPLEMENTATION
' TYPIFIED BY: BURNED OUT LIGHT BULBS,
() BLOWN FUSES, CRACRED METER LENSES, LEARING VALVE FLANGES, LEARING PIPE FLANGES, VALVE SEAT LEAKAGE, SUPERFICIAL DEFECTS IN DRYWALL WALLS, AND HAIRLINE CRACKS IN CONCRETE PROCEDURE REVISIONS ADDITIONAL GUIDANCE TO OPERATORS ON USE OF INFORMATION FROM AVAILABLE INSTRUMENTATION EVALUATION OF SETPOINTS AND ACTIONS BASED ON EXPERIENCE AND RESULTS OF ALL ENGINEERING EVALUATIONS l l 1 I l l 1
() FOLLOW ON CONFIRMATORY ACTIVITIES ON GOING SVI's SYSTEMS ENERGIZED AND NOT-ENERGIZED DURING EVENT NEW WORK REQUEST REVIEW PROCEDURE ASSURES CEI/NRC REVIEW OF ALL POTENTIAL EARTHQUAKE-RELATED ITEMS TIME FRAME: FEB. 3 THRU 28 TOTAL WR's DURING PERIOD: 2401 622 REQUIRED EVALUATION 459 DETERMINED TO NOT BE EARTHQUAKE RELATED 163 WERE REVIEWED WITH THE NRC PRIOR TO IMPLEMENTATION '
' TYPIFIED BY: BURNED OUT LIGHT BULBS, *
() BLOWN FUSES, CRACKED METER LENSES, LEAKING VALVE FLANGES, LEAKING PIPE FLANGES, VALVE SEAT LEAKAGE, I SUPERFICIAL DEFECTS IN DRYWALL WALLS, AND HAIRLINE CRACKS IN CONCRETE PROCEDURE REVISIONS ADDITIONAL GUIDANCE TO OPERATORS ON USE OF INFORMATION FROM AVAILABLE INSTRUMENTATION EVALUATION OF SETPOINTS AND ACTIONS BASED i ON EXPERIENCE AND RESULTS OF ALL ENGINEERING EVALUATIONS I I
s Page 1 oT2 PERRY NUCLEAR POWER PLANT UNIT NO.1 SEISMIC MONITORING INSTRUMENTATION. TA8LE IV Instrument Type Manufacturer /Model Number Location References hber Reactor Building . ' Foundation Mat Figures A and B D51-N101 (1) Kinemetrics/ SMA-3 Elevation 575*-10* Azimuth 175* Reactor Bu!!dingPage
;;nemetrics / SMA-3 Containment vessel Figures A and C D51-N ill (1) Elevation 686* Azimuth 174*
Reactor RecirculationPump (Inside Drywell, Reactor Building) Figures A and D D51-R120 (2) Engdahl/ PAR-400 Elevation 605-0 ( Approximately) Azimuth 145* D51-R130 (2) Engdald / PAR-400 0UT O F 5 E R ViC E Auxiliary Building Foundation Mat Figures A and E D51-R140 (2) Engdahl/ PAR 400 (HPCS Pump Roon) Elevation 568*-4* I Trianial Time-History Accelerograph
- 2. Trianial Peak Accelerograph
- 3. Triasial Response Spectrum Recorder
o o
~
Page 2 cf 2 O PERRY NUCLEAR POWER PLANT UNIT NO.1 SEISMIC MONITORING INSTRUMENTATION TABLE IV f i
'" ' Location References
," Type Manufacturer /Model Number Reactor Building
'- D51-R160 (3) Engdahl/ PSR-1200-H IV-12A Figures A and F E i 7
- 10 j Azimuth 225*
I t i Reactor Building 630* Platform D51 Rl70 (3) Engdahl / PSR-1200-H I V II ' Figures A and G D tm60 l. Azimuth 238* { ) } i Auxiliary Building
" 8" Figures A and H D51-R180 (3) Engdahl/ PSR-1200-H /V $ mp R )
! Elevation 568*-4* l ; Auxiliary Building F at Ma Figures A and J D51-R190 (3) Engdahl/ PSR-1200-H t V p p l Elevation 568*-4" 1 l l l 1 i i I 1. Triasial Time-flistory Accelerograph } 2. Trianial Peak Accelerograph i 3. Triasial Response Spectrum Recorder l .. _ . -. .. . . . . - .
(o)
- 0 u. L n.
mme - 1 gr n ,, m a p. .E . .A L .1 , == 4-
; 1
- g --t u== t- '. --
j a- w.d.r.,__ Ej -
. e- m..
1
- _p=.
5w WL ~ p ..@ Q ,., IT - m j . .*
, . @w {'
g ,j-- c a -
.,=, , s
$. 9 e.- b:J$$k m,_.-u: .y - _ .
& NWG..in R j' !. A m.~~ \
t-; -
, og- n "q< ,,..
. ,. a .
. eg C.. . ."_T?
=_g g
. ~
ei.e- _
,:,:. . gp
, j.
- 1. :.. ,3 2gh ., e e <
h~ '.,!,; l r, -
%_ i :.
A m3) s e., -(9, i e .- . ., , e ~l - . . .
- =
i t y ,
- i2 - > ,
e.rW #.,
'm- %EJ ~~ "1 (; -?,h T'" h. . '~
0m Zee $ ,i t l J
- r. . e ' ,f"-0% e5% = ._
WD , @ M -_ p it 302 M g " .e h' gh i . _
. .m^M @ f.r -
F
$ m, ,' x
* -g " -= 3.fst- ..
l .i S. [ , ,=
,o :
l e 4- _. L u..,cg o 4r .. ~ . ( . 5 CJ a G p :PQ F- . ~~ .= G
-1 a c r- -.. .. . ;
.g G t- - i g ::
m.-
-. 8 peg.y 2
" 5 -.
i p2 n r- byaf :w.mgg , ;m; m .e a.
.,3 1) m .a-..
J a. r , e.J..c tam
,;' . m... , 4dn.., :g
'"y f ci ,i e
[. k , b , ,f? % n. , m m(;4/ j)*e..p,Q. ... . a - y===. v;m.m w> m 1 wg . -
, s ;. y c.
e y_ r. . , 1.r_ ip;, " ifL.;f , .g;r.t-y mj n .-y.D v
.R .lN a.m ):z-
&e , , %. ,*P d**
h$,
,/ == j #y *- I 1\
--= [ g
;. s, , . - gam,- , % ;,,~,,~
- . a. =wsntu , . ,.
~~ -i
.,"d. @.
. mM,,,,,, .
m
~ ^^
_t _4'i
'^-^~~^'\,
--,.k =b ..,,h.5*
Yb
-..r-{g -
f g , ,., j . I , pre 7 a L. M N ( 7,, , r._____,
.e-m
*' w
.rlg -Q 5 a b B- u :
- f_ .*
I I
.I -.
_ t.
,I 1
1,
- a. 1 Bis -
G Y g
@ Instrument #D51 1101 7 3
@ Instrument #D51-fl111 l
@ Instrument #D51-R120
'V @ Instrument #D51-R140
@ Instrument #D51-R160
@ Instrument #D51-R170
@ Instrument #D51-B180 -
@ Instrument #D51-R190 .
I l 9 e 3
= .. j "h__ % ,. .;;==n k ,,n s, w l I
'. gr *. ,,=_
1 * * ~ g a g 3.==,, ,
-u ,,
up ^ "' " g.57 man i
, T .g _ _ W.
- _, _%e t , , , , , , , , 'm ..'..'m ^ ....J
- 1. . !
Jh.a.m.A . - - -, - . - . - - -
,,,g,, ,,j ,
g_ _
" " ~
k
; . .. - n -. -- T d N
.. ,anam; 4 ._
JJJ- -ia,ff") hL l3, .,. . 7 ' 3
,,. N .
fi),ll ise4 l a"a ye v t-* rem m {
.n- . , "M
> ,.. -g.,
r _==o 6
,. m.w
= L ._
s - .I r, .I e/ 2 r.- q; a%
.,J .
8 .::; u
,,ar _y= .1 , - p M,M. 9. lj',, p +. j ) g , jd
% e, x -
q _ .J. .t y ' , Q s n=>j f* ;J *, s 1
.7,7, 7 ff**T
*Q.
a. T ' .
*M, =' ' . c _=.r pi .. ,--
ar - ~
- i r-fla'.*; ; p - . . . . , . . . . . . t.
l** (*Y Tj l l
$b b~ "$' !
G
},, , , . l -
~
+
I' i J( . , , , _
. "".~(t.J j '.:.,.
-c r,. + mm ,I .. .....J. .
9 ,i . m .
...m 7 +. 4,
, ,i ,
gb.b.. gI . . ~ f, k3..J1- jf',J;. y L L.Jh E J.L_M 0121. 1"_Ngh.y'
. . .. t i
fiUic .......
!!( L @_.
y
~~ ; ' .' .*
t= lt.m. p l l: , ., \.m. .-
.~k= ... ,k.n..t .y t
i .. ! / . !\ / \. .. . I
-, I \ / \
- G i @
' OB
- 1. #D51.N101 R/B Foundation Mat, El. 575', Az.175'
- 2. #D51.N111 R/B Containment Vessel, El. 686', Az.174'
- 3. #D51.R120 Reactor Rectre Pump, El. 605', Az.145' p
- 4. #D51.R140 A/B Foundation Mat, El. 568 gu.s o 5. #D51.R160 R/B Foundation Mat, El. 574' Az. 2251
' ) 6. #D51.R170 R/B Platform, El. 630' Az. 238'
- 7. #D51.R180 A/B Foundation Mat, El. 568'
- 8. #D51.R190 A/B Foundation Mat, El. 568'
i ' t I i i ENGDAHL INSTRUMENTATION DATA REDUCATION 1 TRIAXIAL RESPONSE SPECTRA RECORDER 8 AND l TRIAXIAL PEAK ACCELERATION RECORDER 8 i DIRECT MEASUREMENT OF TRACES U8ING l MICROSCOPE l
- TRACE 8 RANCED .002" TO .215", AVG. .030" l l
- PERFORMED BY PLANT STAFF ,
- CONFIRMED BY P. ENGDANL ONE RESPON8E SPECTRA RECORDER EQUIPPED WITH l SWITCHE8 li
- CONTROL ROOM PANEL W/ AM83R AND RED LIGHTS [
i
- TWO THIRD 8 OBE AND 08E FOR CACH FREQULNCY ;
r i
'l e ,
l r I l i
i I l l l l t l KINEMETRICS INSTRUMENTATION DATA REDUCTION l l 5 i l TINE NISTORY ACCELEROGRAPH { ON-SITE - CONVERSION OF CA88ETTS TAPE TO STRIP CHART l l
- PEAR ACCELERATION MEASUREMENT f RINEMETRICS OFFICE - CONVERSION TO DIGITIEED PORMAT i
- CORRECTED TINE HISTORIE8 !
l i
- VELOCITY AND DISPLACEMENT TIME HISTORIES j l - RESPONSE 8PECTRA l l
i f i i t [ l i i i i l
.e O
I O , CHARACTERISTICS OF THE 1986 OHIO EARTHOUAKE l e HIGH FREQUENCIE5 i e SHORT DURATION e LOW ENERGY
.. i I
e LOW VELOCITY f
.l l e SMALL DISPLACEMENT
- l l
O CH AR ACTERISTICS OF THE PERRY DE5IGN BASIS i EARTHQUAKE (55E) ! l i i ! e BROAD BAND FREQUENCIES f I
- e LONG DURATION l l
l e HIGH VELOCITY i I e LARGE DISPLACEMENT l i
.
- HIGH ENERGY I
I
.. 1 o ;
r i
" = = -
' s' * ... .. . .U.. ' .. j.i. iC.. .;i lW..;.i. 3.: s;. . . ". ..._I. l = dst. I. . 7.~Y.1.". iw. I @. . f., .=,*t.; .!!'r4.eim .. . j e m...m.e.
4
,. ~ .=a t ..i. i..4 ';..t','., .
[ ' S. - l _. l _ .
. t ..
s ! - g . . 2, ,, i
) l \ '
l p .-.' - ._. . - . - .. a g l . l L l s ! I H
, ,.6 .. ._ . _
(,.- i l .
., , l. .
y . - r ..
.-. 1. - ..
{n .
* ,J &
I 8 Ii f l .t . . _ . . . .
. . . . . . . . . . . , . . . . _ - _ ._.. . . ma I a l g
l E t .1 v .. .... ... ... ... .. ,u. .u .... ... pe ,, , m'"m.10DCWCC=3 . . ., . .... .; . 114,.ie ssgs
)
,.u3 ~ :m.A..M,,.=.nh a
@ .. . . . ..... .-. l a.w.., ,=dw. . ...:.T.;T.,g-'tt ... ... l6,t oI9!.T."ihW"r'a" , le. ...m.e. ..
e
,s . . . . . . . . . . . . ._ _ . . . . .-. . . . . .. . .. . . . . -
1 ' s . . . . . . . .. .. ... _. s \
- e Gd I '
I
,m, g %*,vW .f ': _ .
^ - -
. . ~ ... . . . . .. -
i
*! I
\l
.n, y ... .. ..... .. . . . . . . . - .. .__ ._ . . ~ - .. ... .. . . .
a j L , H
. . . . _ ... .. . . . . . . . .. .. ... .. . .. . . .d I f.( 1
( . . . . . . ...... . . . . .. . . . . .. . . . . . . .. .. . l
.41
, .. e
,;N. I f
( e. . . . . . . O s
'.e.., - .u. r.. .. . .i..;;. . . ;;- . ;.-- r .; - ; ;. ... .. ...
. e ,,, *
. . . . n. d.i urA. ns ..
4
i.
,A, a @?.*.:. :"",:.'.:.'" .... llN:Wt'.dd?'th,'.!,, I!k;}Ct: 3[.I.,f l$4;.$;'$';fW**.;$,'"' l'.,T".U.
. I r : s -
- v. .
f ( 3
, . ... . _ i.. - .
s. I .. _ ... 1.
' l l
- s -e.-.
~ -
g .
. g .
, p. . . . . . . ' . _
l . I f ; l ( l *p' . . . ,. _ .. i - . . -
-l ,
t
. lsI.
s m_ _- { { l!! .-
.4
' j E j jg !' *
- l g 't
' .a i .+ .-
; E I
I 1
.I. .
l N
.t.
o . S . er$ l ,y . . . . . _ ._ ! . . _ , 1
-. .._ a:a l .
a
's . m e.w O)
(v.- gg;ptp.'N SE C ,
- s. n .. se e'.se e.se e.ee Ilad L esla841$808 i n . s e.se a n.m .6. ie it. m it. se m
si.se Te.es 1:...~.
. . . .~. . . ... ... . ... II,p,;P, .J'.11w...;.I..T.)1:
,. . . . . , ... ", .. ((tLaa,l..")r1:. , .. ., ".,'a>= l;,;,r%,..,) i A; 7.,.;;,.,L . li.=.i.a.".. .. .
.s l
s .. .... t
; a g . . . .... . - .-. . . . . ._. . _ . . . . . . . . . ..
,s . . . . .....'. _. .. . ._.._ . _ _
p . ...
.p.. s s:
. i so. -
g tme.ic ,q ,sy* to-se:: = -
- - = - -
2 .._ -- 1 3 y; .. . . . .. _ . . . .-_ . . . . . . . . . . _.. .... . _ . . .. _ __. . . . . ,
\ . E.,
s
- I.I 9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . _ . . . .. . .. .. . . . .
. 3, a 'l j
9 . ... . . . . . .. .. ..
, I n
(. *. '.i. I. x"- 4 e
.* 9 . . . . .. ... . ..*. . . . . . . . . . . . . .. .. u i
s Y *
, <[ e. . e, w iI.ei./. . u e A, .7"""i e e'i ~', Je" i i3~~".. . ...w . .e.i.'..
e..n . .. 347 . . . ..-. .
, ,e e . w
'i e ,,
- r. v t I[g_ .. ..... . ~.. . .. . . . p, . . . ;t.g,i .. . ..,;.2.
..r N- . ;.; ; . , i ., . . . ... ,. . ., l t' .. .;;E. . . . .. r . .p.
., y.W,m). )y. ,. .;;,,.
. i.,... . . .,,,. .j .i ..
. .... ... l t
n =
\
1 $, i . .
i f *I l i 6 ..
g , I . i .
. i s v *
. I
,E ,
- l. . t .
l, fy &f&W)f%&dfkk\f I
+ _ _ .. _ . . . . _ -- .s....-
i ii4 .
- n. ,
,+ . . . .
',1 1
1
+ .,
- h
.t
.I l __.j_ i I .
s I .
. I m
- s. . .. .. ... . . . .
_ ..........d..i . . . in sul e5 .... ... __......9.__ k' (97 .=.: . . IMEiWW,','.. lE0!iHlr."" inGE'linsie l::=:: I 5 _ E 2 e, ._ . . . . . _ ___ . - _ . . . . . . . _ _.__ . _ .. .. _ . . . _--. . . . . . . E l ji$s 9 pp,s=;; . = -
= 1 0,
g _. . . . . . . . ... . _ . . . . __ . . . _ . . . . . _ . . . . . .. . . . . . ...... .. ., 3.
- s . ss 4 ..._ .. .. _. _... .... . . . _ . ..... .. . . . _ . . . . . . . . . . . a u
3 't 4 . .. . .. ... . . . . . ._ . . . . . . . . .. .. .. ... . s
.. ,I /"N) I I
( i
. . . . .. . . .3T 1.,A- .. ; ;r. . . . . . . ,
.. ... , . v ,.L.,,,
,, ,r ,r... - ,r.;r ,. ...,.
.....,.t,,..
, ,v--
. s. ._ ._ . ......is....._................
. ._ n.:: . = . ... i m...... ::::. .:.
. .;..... ...... . i ..,...... . . =,.... . ......,.
m =. , mi.. . . ... _
.. i i
l .. ! g .
._ ._i,. _ l
! J, . .. i _ . . _ _ ._
i
- l. -l l . i : .
; , j .
. i j l i .- g . I g
- i : 1 j _ _. . _. _ _ .r _.
I
- i
. l l i I l
_2
,; I '., _
,,,, , '. . l _. . .
[ _. ..
~ ' '
I l h'jy f
~
ll* ]f f , Y L_ hhi)df{ _. y N . Y%
; i
- p. _ . _ . . - _ . . . . _ .,..
ll i i i. i . i l 3 E l- . .: l g
+
-Y --
i l
==
l
\
E ! i . h 8 t i i i .
}- - -
4 ,1 l 17rs s t 1 l l --- - l: '
*:8 i
S t 5.es i.no a.se s'. no s.se 7.3 e.se i e. se i t. se ss,se it.se se.se me.se es.se an.se r- r1;;=.v T 1 as.se
,,I.lR.E 188 SECD8:05 ,, ,,
'8
.... =. . . . l n..!.n,';;;t
'i i..
o l;,g'i.,U,T..h. !"n,,:,.; ;;. .,.. l .m..i. . . .
. LcJ.....
. . - . .. .. i..t!.N;U..". :;; ", ... .T,.,.!,:n,.e.=;;r.
. , . , lL , .,.e... "..m.'" . , , , , . . .
3 . _. _. .._. .. ... __ . . . .. ... _ _ . . ___ ._ -. __. 2, 4 E _ . . _ _ _ _ _ _ _ . . _ _ . . . _ _ . . . . . . _ . . _ _ . _ _. ._..
, , w 6j3 G4' :i lt @Nh.Irh.f(fpkwnunwkW-tm , N .l '. R. ,
ji _ _ . . .. . ._ g p . ... . l . . . .. _.. .. _. ..... . . _ . ll: 6-
.K l s p,,9 . . . _ _ . . .. __ . _ . . . _ . . _ . . . . . je L '.
ll
= :.
* '1 l2 l
+.
* '775
**:E f 1
,I
$ l e.ee i.w 3 ee e.se s._se p.se s. ee ie.se it se i s.se i b.co t s.se 18. e0 it.w_ . . s' oe si .se s a. * : .s a s.w .a ---s 1IMf IN %f4ANO% $8. ==.T1
l
.. .. . .. . .i m.. .. a ....z. w . . .. u ..,... ...
. La. .: c. ,. ...:a.... m... q c. a ,w... m,,.. a i. a.. u. . .i.. ... .... -1 i . .
. i, : 1 i .
i i l ' i _ l: ' i !, i a i , I i
, J.
. ._ .. - __ _ l, , ,
= __. . . . _.
, J
!- 4
., i ;
. I : .
E
, _ . _,I_ _I .!. ! . .
, r l
l
- p. [_ . '. ,
I !.! i A J n.- y 'tm i .1914 i l} h' 11 i i q p( 1 p p.i d-/ I pp T
. s '
v i i gi t 4 ! ( , l i ;'
; f . _ _!
i I i
, _j_
_ i
. 6
.__ g t
. i l l i !. I. j H
+ _ _. ._ u
.__,._.i ...
i f.a. d. - s 4 .L_ , m:x en
, . i
! t i
- a. ! t
- , ,, tamt in st.conos ,
is s i er,. u m .ar. suitslu. a ei ( , s
$'.".:"' WFi'*4 -. hgg s .r. a.3 l:i s 6 :s s g*at t e. Es.C; *i !.:',:. D*;!,,.u lEt!!;i :: t:t;.. , l t' at t Ci n T.is,"" ?H gis.v,. . i,r,J".2'- lgaggg g,:
=....
I 3 . t a
- g. . _ _ _ _ _ _ .. _ _ _ . _ . _ _ _ . . _ _ ._. . . . .
i E. g . . . _ . .. __ _ _ = . _ . . __ e_1 . _ . _ . . . . ._ __ - . . _ -. . _ . . . . . _ h/f*'M ., I I
's j jNNIfj @ M M M- "Y-W++ M N - - - - - - " " - - - - '
s .: w. l
. w-p . . . . .
3. i .8 .. . . . . . . ._._. . .. . . . . . . . . . _ . . _ _ _ . ._.__. _ . . . __. . . . .
- i. .:
3 4:
,g .
4 . .. . ...__. _._ .. .. . .
...., .. . . . . . . . 8 r.:. .:t I
n .- t V) .
- 8
. .. w a.u
..._w__. a. . . .,n. u l
i.
- ...u....,
i.w
...,_..u_, .
- e. w ..,i,...,,.>.. .s n ns
- n. ...a
. . .s.
i
. .r
~
T. i - .:~. .*
- p. g. . . ~ . .. ..3..- -~.~ . :~ '==: p> .$.e:.::;t ~.M.:.?..:~.' *- j!.i!)y.::!'..p.* W.G.. l""'"'""-
. 4.' W..::.;2.
.u'u l)'"L :':7. . . s=:.:J 0:e- -- =-
.l I
i I I
. 3 l l-.
l l I'. i l ,' I : l i. i i .
- . . ... ..i......
- i i
- j-l i . i , i i i .
. j e i l
_=g, 't :
=
t
. ._ . . - r, ._.
----.y-=
l [
}l
. =:a.
. .J
. l
- I b 2
d. I. '. i
- i at M
. I **
l - g, n }
. . . , . 3 ,
3 i.
' .r7 4 .._ .- 3:3
\
.)
* -'i ~
e.se s . s. s. .e s.se s.se 3.se .. es i s. se is. se i t. s. as.es is.n. as. i s. n. si.se sa.as.,
' rm
~- n .i . u . . . i. = i :
Tint Im Sttancs 04 .a . .-- .=. .-.. I 5,:!W...:'.,2 i... I.k.!.4.;;.;..'M
. . .. . f., id t '"T, T,r,.l..w"..."""
l:.!.n'i.'*.f, . .;':'.,:"1,,1:',:'.N......
. , . . .m l ea.a.i.= a.. .
e . . _ . ...._ . . . . - . . . ._ . . . . _ . .._ _. . . _ _ . _ . . _._.. .- - ._. . . . . J' s ...
- a. _ . . . _ . . . . - ._ - . . . -. . - . _ . -- _ - - . . . _ _ . . . ...
e
- g. . . _ . . . _ . . _ . . _ . _ . _ . . _ _ ___ . . . . -. .. . . . . .
r l!
= >Y),WW0 f ,Ii t,W&~?fh%===^^- i
. = - x- : - - - -- - - --- -- - ----
- m. lr J
& : I m t g . . . _ _ . .. . . _ _ _ . _ . .
. _ . _ _ _ . . . _ _ . . _ = . _ . . . . . _ . . .
g
.8.. ...l .. . . _ . . . ._ __. _ _ . . _ _ . . . . . _ . _ . . . . Ei.
i . .. . . . . . . . . . . . . . . . _ . . . . _ . . . . .._. ..._. . . . . .. . _ _ . . !=
. i ,::n:::
( (m; e . . . __ . _ . _ . . . . _ _ . . . . . _ _ . . . _ . . . . . _ . . . . . . . . I
=:
t ;
=
- b. .
.-,,.u...... i.--.
..u .. i.u . . a.
s i n, ..i. , y , n .n$
,, . .. u . . ,6.
. .... u ... ii.u r
,. _ o.
_. n
,,. u
O . SEISMIC DESIGN BASIS F e BROAD BAND FREQUENCY DESIGN RESPONSE SPECTRA e SMOOTHED,84 PERCENTILE
~
SPECTRA e COMPOSITE TIME HISTORIES WITH LONG DURATIONS AND HIGH O
=
ENERGv I i 4 O l l l l __-,-_,---____._.___.__.___...-...______,_.__..___.._.__.._______._______J
~
[ A - s- x M' PING ACTOR,
^ s) s X>
I
- j. .
"\s K ,
b W 4; [ . 5
" / Nh %
i
~
CE?Odh30^x^lO ^> - i
! xX / - h> .A cc I )Obb A' M Db
/
5 s ^ x e %^
/ \ S.
/ N / '
o \
~
0.1 [Ob IObbi f 0.2 C.5 1 2 5 to 20 50 100 FREQUENCY, eps FIGURE 1. HORIZONTAL DESIGN RESPONSE SPECTRA - SCALED TO 1g HORIZONTAL OROUND ACCELERATION . oe
- O( s.
- , - - - . , . , _ . , , . , - _ _-_-.---.._,.,,-.,,.._.r..,,,,,,-,,,_-_,-g ,.,,,,,,wm_,, --m m .- - w _. .m.- m -_ ,,.. -. . - , , . ,
O - ZPA COMPARISON e ZERO PERIOD ACCELERATION
.g e RECORDED ZPA'S VARY FROM BELOW OBE VALUES TO 74% OF SSE _ VALUES e EXCEPT AT CONTAINMENT VESSEL ELEVATION 686'
~ ~
e BUT RELATIVE DISPLACEMENTS
^" 5'"55555 "5"5 (^5 *"" ^5 O -
ALL OTHER LOCATIONS) ARE LOW REASON: e HIGH FREQUENCY AND LOW
~
i ENERGY OF THE 1986 EARTHQUAKE 5 l l O l-l - _
O ' O O i I TABLE II l Comparison of Design ZPA's' VS Recorded ZPA's i (Expressed in g values) Auxiliary Building Reactor Building Reactor Building Reactor Building Reactor Building - Foundation Mat Foundation Mat ' Recirculation Pump Platform Containment Vessel Elevation 568' Elevation 574'-10* Elevation 605' Elevation 630' Elevation 686* PAR 400(Engdahl) SMA-3 (Kinemetrics) PAR 400(Engdahl) Inside Drywell SMA-3 (Kir.emetrics) 051-R140 D51-N101 D51-R120 P5R 1200(Engdahl) D51-tJ111 D51-R170 Recorded .l? .18 .32 .09 .55 1 55E .17 .18 1.06 .48 .40 ) NS OBE .10 .10 .86 .40 .24 a Recorded .06 .10 .11 .1C .18 EW SSE .20 .18 1.06 .48 .40 $ OBE .30 .30 .86 .40 .24 l l Recorded .03 .11 .05 Note 2 .30 ] VERT. SSE .20 .18 .47 .28 .24 l OBE .I0 .30 .38 .16 .15 l
, f. 1 j, ,
. Recorded .18 !.23 1341 s Note 2 - .65 1
SR55' 55E' .33 .31 4 1.57 '. K......_, ' I ,l
'.7 3 2
.62
- r.n OBE. .17 .17 ~- 1.27 s n~597 .37
- s s l
! t Zero period acceleration ! 2 ZPA indeterminable from available data j 3 Square-root of-the-sum of the squares 4 Iscensing basis is $$E
O ' O O
~
T ABLE I j Comparison of Design Displacements V5 Recorded Displacements' i ( Expressed in centimeters /one inch = 2.54 cm) COLUMN 1 COLUMN 2 C O L U M N 2 minus C O L U M N 1 Reactor Building Reactor Building Relative Displacements i n , Containment Vessel for the 4 Foundation Mat Elevation 574'-10" Elevation 686* Containment Vessel SMA-3 ( Kinemetrics) SMA-3 ( kinemetrics) D51-N 101 D51-N111 Recorded 0 09 0.17 0.08 l ,! $$E O044 0.28 0.24 N5 i OBE O 023 0.17 0.15 Recorded 0.16 0.21 0 05 0.044 0.28 0.24 EW SSE OBE O 023 0.17 0.15 Recorded 0 05 0.07 0.02 VERT. SSE O 02 0.37 0.017 OBE O 013 0.022 0.009 i I l
~
Recorded -' -! .
- 0.1 SRSS' SSE - ' Shkh ' ~ '
,l :- 0.34 OBE -
;]} ' : 0.21 I ,
, . n. .c
]
- 1. Displacements based on same time-step to determine relative displacements
- 2. Square-root of the-sum of the squares i --- _. - _ -- - - - _ _ --
O . CONTAINMENT STRESSES COMPARISON E!.EVATION DYNAMIC FORCES DYNAMIC STRESSES DESIGN STRESSES P Mx Mr P/A + Mx/S or P/A + Mr/S (K) (Ft-K) (Ft-K) (K/In2) (K/in2) 688'-6" 1,339 44,220 31,820 0.414 .398 644'-6" 1,589 46,970 44,820 0.464 .802 592*-3" 1,674 38,000 53,670 0.510 1.320 O . r i l e 6 , O
O ' RESPONSE SPECTRA COMPARISON l e PERRY DESIGN RESPONSE SPECTRA ARE FAR ABOVE THE RECORDED SPECTRA IN THE FREQUENCY REGION BELOW 14 H::. o CERTAIN RECORDED RESPONSE SPECTRA EXCEED DESIGN SPECTRA VALUES IN THE REGION AROUND 20 Hz. e CORRESPONDING SMALL DISPLACEMENTS O (EXAMPLE: 7/100 INCHES OR BELOW AT FOUNDATION MAT)
- o NO ENGINEERING SIGNIFICANCE
- . e RECORDED VELOCITY SPECTRA SHOW i
MUCH LESS ENERGY THAN THE DESIGN RESPONSE SPECTRA l e D 8 O
.c.J '
EARIHOJA.%t. JAhUARY Ji. I966 11AE001 PERRY NUCLEAR POWER *LANT COMP SOUTH SMA35/N 165-il I l ( DAMP 1NC VALUES ARE 2 PERCENT OF CRITICAL FREDUENCY - HZ .. 10 1 10 3 6... . . . . .. - . 3.. ... . . . .....;10
. 5*
g . ~
~ . PSA - G SD - IN' -
q . 10 : .
' ~
.0 18 2
'. ES)
'R. S. M A L.$14 s18 T iD 3,s.t.Homt,@,s 2. - SD _ CM -
~
. ,- A- -
i0 b0 o . . n - .
! _ l. 80 :
1 _, - . z - N - 10 en . 1 . -
- n .
a y
. / [' 0*
~
[ 4
~
'10 ,
' -l?GCbRDE 1
i : g-2 3 l_ - .
~ ~
'l 0
- l
- . -3 -
?
{ ~ O ! g
-1 , i . ,,,.t . , , , ,.,t . , , ,..,.I ~,
-i 10 1 10
_. PERIOD - SEC
ML 5.0 EARTHOUAKE JANUARY 31. 1966 11A8001 PERRY NUCLEAR PCER PL ANT COMP WEST SMA35/N 165-1T DAMPING VALUES ARE 2 PERCENT OF CR*TICAL C FREQUENCY - H . 10 1 10 3 '
. . . g... . 6 . . . .. . . . . . . g... . ; i- 0 t- - g j ~
- d ,
f. a PSA - G SD - IN - t - 1 10 -
] -
id 0 1 .,
.z
. StGN $18 '] ' g I x.c. mar En..WC' g9 - cn s.tt.uaa.,f .t1 I .
10s _ l- . e (hv -
. r} 0 sy GC'y
- ~
-t
, 1 .0
! 3 .
z D
~
. s1 . Q 10
-g .
-2
- 1 ,0 _
F?EcCRDE -
~
- '10 ' .
1 ,
~ -
i -2 -3 I 1 . ,80
- 1. _
9
-. -2 .'
- '1 0 . -
l
! -3 r
) -
( ..
~
/' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' '
10
-4 10 1 10 i PERIOD - SEC ( (
PERRY NUCLEAR POWER PLANT COMP UP 5.9A3E/N iss-tv IIA 8001 ( DAMPlNG VALUES ARE 2 PERCENT OF CRITICAL FREQUENCY - HZ 10 s 10 1
. 10 a . .. s.... , . . . s..s s . . ,...3
~
d -
.PSA - G ,
SD - IN
../.-
1 10 -
' ~
; - \ ,a 1 e, -
4 - s . ,2 _ 1o' ESIG)4 50 - cM n.e.ya r 5. m.s, .
- P.s.t.VranCAL,6817 10,s ,
i -
- r ~ . s, o Q.g.
0'
,' : 1
- = -
lo
,. ~,, .
u,, o - 1 2 1 40
~
-i REcoR b '. [
, s3 g , ,
t l' -
. : x-2 iG:
i - ,
. _ _2 3
'i 0 _
_3 _1 - l .. ! O ;o
-2 e , , , ,,,,, , , , , , , , , , . . , , ', . . . .
-l 10 1 10 PE8100 - SEC 'I/..[//6'j/ /
~~
j%Rsds
_-.m....i: . s v .. rs t. d.h a v . . K Y 31. i956
, 11AE002
?ERRY NUCLEAR l'cWER PL ANT Co.MP SOUTH s.MA3s/N .65-2l
( DAMPINC VALUES ARE 2 !'E8 CENT OF CRITICAL O
'J -
10
. REQUENCY - HZ .;
1
... . .. . . 10
.... . .. . . g... . ;i t .
i -
,g .
- ' PSA - G ' 50 - IN i 1 0: - DES N - t O.' ff.
.8e
- ,. 0
. SSE //0 -
}e'Efe 1,
~
~ - .. .
if : 1.C so - en : 1 i r , s . . No - O- (!2 i _g -
. - 1 .0
- t. .
C::' m
. m -
I - 10 C3 . - [ -1
~
1 -2 - 1
- RECORDED o
-2
- . yo -
3 _.
, - 2 n i -s
.. .o i -
- 2
'10 -
3 , _
. - 1 .
Cd : , , .
\,,..., , , , , , , , . , , , , , , . . . ,
' 16' i 10
.i PCR100 - SEC ,
/ '
11AE002 PERRY NUCLEAR POWER PLANT COM.* WES! SMA35/N 165-2 ( DAF. PING VALUES ARE 2 PERCENT OF CRITICAL FREQUENCY - HZ C*j s 10 1 6... . . . . . 10
.....;tt
~
- . I j . .
L . PS A - C . SD - IN
!<i, to 1
S N - Ed, Z8f Wh ht"
' ~
0 1
- ~
l ! .
~. ,
'st g -
lC
- SD - C".
~
I s .
** i -
(a
$10 O (,. ' -
1 6 ' e . Z . ., s.,, . s1 . .i. m . - o . 2 - l' .
'I . ; .
,i G .
i
. RacoRb -
3
~
. '10 . .
1 2 -s [ IV 0 l I-s1 g2 k;
; -3 -
1
'1 ~
O *' 10 '
-6 10 1 10 L , PERIOD - SEC ld%/
rhen/
- 's O.U t A n t .9 U U A M. JANJA.0,Y 31. ISE6 11A8002 PERRY NUCLEAR P0i'ER PL AN! C O M.* U." SMA35/N 165-4 Ov (( -
DAP.P;NG VALUES ARE 10 g... . . . 2 PERCENT OF CRITICAL FREQUENCY - HZ 1 10 i - - - 6 g
- PSA - C . -
SD - IN . 8 10, - i! .
; g 0 . 1d 9 .
ES\ - 4 , . . gg . . i o- . h[g[jfg
- ~
t
, SD - CM
, a
. 10,s
. s10 -
= I .
.c . -
't -
1 so: l -
~ .
.Zs . .
I,,n .
. s1 a .-
-2
- 1. 1 so .
t RecoPb b 1 1
'10 ,
]
~
- * -2. -3
, l 0
- -2 -
'1 D .
,.3
( O { ,0
.,. . . , ,..., . . . , . . . . , /. .
.N,...., : .
-i i 10 1 10 t .. PERIOD - SEC - /
l
l VERIFICATION OF THE LACK OF ENGINEERING SIGNIFICANCE OF THE 1986 OHIO EARTHQUAKE e PROOF TEST- PLANT STRUCTURES UNAFFECTED
- PLANT EQUlPMENT UNAFFECTED e OTHER INDUSTRIAL CRITERIA CONFIRM THE CONCLUSIONS .
~ '
o PILE DRIVING R BLASTING OPERATIONS - ) USE 1 INCH /SEC.P. ARTICLE VELOCITY AS THE ' , j NON-DAMAGE THRESHOLD FOR NON-O ENGINEERED STRUCTURES e THE RECORDED MAXIMUM VELOCITY
, . ATTHE PERRY FOUNDATION MATis ONLY 0.87 INCH /SEC.
1 - . , e IEEE 344 PROHIBITS QUALIFICATION BY SHOCK-TYPE TESTING (LOW ENERGY, HIGH _FREQU ENCY, SHORT DURATION) O . i l . . - - - - - . , - _ _ . - _ . - - - _ . . . - _ _ - - . - . , . . _ _ _ _ . - . . . . - .
BUILT IN CONSERVATISMS OF PLANT DESIGN Q e BROADENING AND ENVELOPING PROCESS OF FLOOR RESPONSE SPECTRA. e EQUIPMENT QU ALIFICATION BY GENERIC TESTS. e STRAIN HARDENING NOT ACCOUNTED FOR AND STATIC ALLOWABLE 5 USED FOR DYNAMIC LOADS. O . e LOADING COMBINATIONS. e PRIMARY STRESSES VS. SECONDARY STRESSES. e DAMPING VALUES.
~
e RESPONSES OF EL CENTRO STEAM PLANT.
,,w _,-,-- --,-- y9 y,yww&myg., ,----%-y-.-,,--9 y-e,w-w,
- - - - - - - - - - -----m.---- -.- r-e . ,_y--
.. .= . - . - __
REPRESENTA'2IVE EQUIPMENT EVALUATED FOR SEISMIC QUALIFICATION MARGIN Qualification Components Bldg. Elev. Methods Vendor Motors Auxiliary 568' Test / GE Analysis Pumps Auxiliary 568' Analysis Byron Jackson l Control Intermediate 620' Test GE Switchgear Blowers Auxiliary 620' Test GE Pressure Auxiliary 568' Test Rosemount Transmitters l Power Control 620' Test Brown Boveri
- Switchgear Auxiliary 568' Test Rosemount O FlowTransmitters Valves Reactor 688' Test / Henry Pratt Analysis Actuators Reactor 688' Test / Bettis Analysis Limitorque Instrument Auxiliary 568' Test GE Racks I
l
- C:)
1
7 . , g - 1
~m (J ) n v (v)
Zij VAcuaM REL &f val V& LIMIT ToRooG AcTdAToR di oR46 f VALV&, N6MR[ fRATT Ta&oHATe c AcTv4TsR, 8G7715
$.u(, y v MGTAL CLAD SWITCH 4EAFs * - -
Byogy 60VERL y e,_,,,,,. E 1 uc.... ....:u r t.i"iPX"".u' o uu , m =,:m" r
=
M&&I- ]*
- w"fNr a b ""'Tf_.!
I _._. _ !'l _f _ y . _ u Et o
-a- a - - -"
; f" ;/ d_. O ji fig -...m..-
Y _7a,o,, m. f ww- """btw, s _. 1 r ;. . n - - . m- g g 1 DIII L I II;l I I d_ llM[ i e'" "** Y _-)wb""*_'"L I tikL. fP. C l [ i ' M{ A I. J b 7 i ?, titt %. h1 i
- .- c=4 Q l ..
x -7 4 .'
$1,2.- V '~TOL;g ..c [
. I i,q
~
il. v.rlf~hiniiG5Iin g l f]2M i 4 . n . ,4.
', , I"A' ,"1 i Ikq
~g ;([ ,,-_ _.
!1 n ..
g ~ [j g y.,..,,, j . . . :n.. ... . ,g g I = ,b 4 gg v
; 1 I i
ll 11 I mL.l O03 OJ u-ml b r-b,,J-
- Ir.
f m ,.
, m. '
jlli3s i -m.,- < l . h ~ ~ u >. u s. gi a. t r r_1 wl ra, u i
~L ,,..x m.. -
' ~~J,,,,\~--J, ',,,\ ,,, ,,,, / m. J f.= ,,,
l =. e .l;, , i
.w r
,,1 n e u
e i m.,, i i i
il o i
-m e- f' h by O V
'f*07';
!! / ;
u b. .:ll .?- - g!lI;.g,p/ I
. j - - - -- -
ry- {**g+ i , -La y R ;N#4,, j - D,,:,:,;,,,;_;y
- _it.it A ; ,';,, ..... . ,.
m f= = -
- .. c. - = , m
- .'. f _' E 9.3 J
@ d4 5 5,,,,.- am.
'Pg-=i~,-}7 . GGyL. - t. '
*a * - . - f7 J
- im. s ..,
,yg
' ' ' " ' " ~"f' y
..;, p3 m, -
plipp "' ;~. "' =- f 1 --
-,:. s5tyg9
. m ._l .~,...
- .1 = a 4 t .- . 2 . -
e -n - . . .-.._. - r . . - . y .g [= g my m Wfg f 74 =., Tis <:- _ q.m....
.y JL pm -., L,, . ..l ,
.- 3_ ,
m t
-g
" ,> , n ... r : w . _ .x.
L yIL tk ._. -- o : m:
'~
- ~ ..
W ..r.f. . .,- eL 4g.2y.j.. }_=i=. - ; ' #"."j'
-(, . _.L}, .1 C, -
d.n
-r ;
f spl.
- ..- t w
i=,- . .-r _.
- wq (i ,
t' (n3. w,.l)]ll i
-- --<.,'s,'}<re -'
S.., - -4. . w,
,- l y m.-7 A .i 7A -.
7.=-,- --
- '('; = -t-. a.t r' r 4
jg. y=- m'-r-\ . . =- j- . .1 :-
@ _0 .iy
_g- 2' 11 r.
.12 i_.. k. "T
;,. a L
g -- g T M a M3 pu,lit. sJ=.-- t-j . ,. E t r -j-.
- u. =@L,Tl#.
- :: w
- M~ r . - - ,
'V, -y-- L ym,.-
,ga_ :iinc u H
one a;q. l _ 2 LL m -u , g-- -- ,,q - ,,, ,[ ,,,. . , j M"; :$.bg
- : a
$I ima ir*~ l .
'I
{ ; 4 < 4,,, k l
=. -
== l ,, .=- ==
. . . .- =.-_ , ...- ...- c ~ .. .- ,. . . , , . . ., _ _ , , , _
., j
- RHR fuMp#MoTon. LasTRUMEnf racks -
~ RECIRc0L ATsod fdMy Lpc5 funf4 hot *N, INSTRUMENT Racks
*[esf CoNTsoL SWITCH 4 EAR HPC .5 fumy i mot *R , wsin0 MENT Racks Cz & DsFFGnEuTIAL PREssuRG TRA"5M'MRS fgss30nE TRAasusTi&R.S FLou TPAN.suur128s GE, SyRod Jacused, RoseMead[
Am. 16 (2-PERRY HUCLEAR ( ,,N p ,, THE CLEVELAN
TABLE 2 m*J **
.nas.
mun,
..m j unu.u..n .m g, '.y.=.;5;; . . ~ .. l9g .;3;;,
.... m
. n o i i.n .
- i. n . i .. .. . .. m , ,,,,, ,,,,, ,,,,, ,,,,,
.. m , .. .. n , n.a g.y . . .n .. ... ,., ...
1 n.
<= . .. . .. . :::;; .. .
. .m.
. . .n .a
..so
.. =
. . . n.
.. . . .m
.on n ,.. . g ,. ..... t= t;;; .. . .. . :
.. . - ,< .= .. = .. =
. =
,3
.in .= ..is . . u.
i .
=
1 ,. i o . i,i i . . ... .. .. . .. m .
- 1. .- n io., . .. m ..,,.
.. m . .
- i. um ,i
.. = ~ , n . iu . .. .... ..,,, ..
... . . n.
-a
... .. .. . .= :;;; tg ;,y
..- t= .:
_a n iu . .I.Itt
..~
... .. ~
- w. wi u n io .... .. . ..,,, ,,,,,
..i= . .. .
- n. .
- u. . . . ,i . tm ::;;; :.=
. .. ~..
.. m u a. .. ..n. .m .,. i.
. m.,
- u. . .. ..
.. m.
.. m t
- u. t .. .
...m
.. m ,. ..
..m ...m .. m i$- a ~ -a ~. .= .. . .. = .. . i. .. l i.. .n . i i-o .. .. . .. . .... ..., .. .
- n. ~n .. m .. m .. m .. m ,., ..m i n w u i. .. .. .. .. . , .. m
- u. - -a. <- - .i,i w. m .. . . .n = m
. =wu .... .. .
. i..ue<ui . .. ...,, .. .
- n. w.- .. . .. ..... ..
.. . tg t;;;
a ==*-a-a= -*-a. tilt lni i tW in t,=,' ta
. m.i_ w
- n. i .. .i .. m .. ... . ,, . 1=
t li. , i
- u. i... i n,n .- .. . .. . ...,, .... 1 ii., . ,
- r. ..if. .. pis ..Hf ...e t .. 4 N. i ei.,.t t
- f. i M. . $ .t em ft .. N. .... .. .. .,y ..
.ei. t.t 27 t e t .. ..33 I8 b '*8
' ..M8 .. 4,
.e e.t
- m. Fa f,. $ (.t .. 6 em
.n :i= .. . . ..
. .ci .. ..
1 O
. - . , _ -,.,__n__.-. ---- , .,--.-- - ,,_ ,---.. _ , _n,,,,.,,-_--,..-,_nn_ , - - , . , - , - - -,-,n-
N' Ialtf R op &W R A c. KS - l
, d MP s 1N c1 P0001 ,
I ~ gg' Q" lH Lt.P00l7 g.b
~
9411 P0018 " s.
-.t c., w n; m 11.Pocu I
~ (H 2.2, P005I
- *.S si c 4 L 0 6. v.
-* v : t 4 L Ds .P.64 0 dp
- w. t En.t m 5 i 4
.Ji. 4
. t. -
. e
--- m (,
I i, a 1
+
1 m "1 *
._ a.
/\ _I A M. - # % yA ., q sun -
x m- f._ s <
-a,$ st - -
= ;t .
-f-
-f_ g:, :. .r..e .g i
.s
. - 1 1- , ,
m . .__. '
,a___ _-
t=- - 5==- uM t-r,s . n e, l p ,- .,: -. .s t v _~ - _ l V
}
_-i F ll
~/ ~
t- . 1 is s .,. ! I e r r , , . l
. - , I 4
9 _--
~.
l'
.~ _
f ' m m - IE 5 5 I I'. A i . TL If IE i V i -
. . . . ..i i
, , . . . . . . . . i . g. . . . . . ..
e k, 10 1C0
\ F r.tutwu (W!)
Reconse R.s .
; 7,?
ren. Aux b 6GS 7, , gg g g q, ,
'9
* , O8 e
l l
. - . . . . . . . ... .. . . . . . ._. ..w . . . . . ..
. e.
VA *,* .I A - . ?5 O 7 e. 4::
'2;;8 i::c:s a a::l*..
~~l 8.4"* --t
-l i
- p
?.l..e 1:;;4 .g . l.*Uiie
; ':s IX. g
e- ~..g... 3..!!!J
**I*' i"* !I l .I
- 4 e."*.;l i! v.3 g 4
- !!i.I
.h.! ::::::: t;*!;;l I *pa 1 N
/
8 W.
/.*', I w
'.' w
* . .' e eI -
- l. a. g.,
. e ,.-. .
,4... .m
)9
-."- y
- F lia, g
.. "*~..
,* .; 4
. %,,,,o
. . . sc
- e. ..
y M
' ' ' f4
- n:'. o l cc
... i 2
.,i i l i
C:*! :tri i. k , q q
.. . a. ...
. . . . .....e
"**..l s:6.i
*::: N %l g O
.* .::r' g Cl.a
- i;.. I l......
'.' w e,
4
^ *
\ ;
. et 6
. - j A
' l i
:.*?.l g bl o %, .
- s i.r*: ,
' 4
;) 6.llj;l
.I t.. .. i . N
'.. 'g
'9 ..
% := ,
r, '.'. i.!!.. ' l '.... ::::.. i e. i .
. ',, ,i:!';:s
, t, , y
- t. 6.I.
.8 . . . - . . . . , ::::;;
, .a , t.. .,,..I i::!;::
**"*8 i 3 ..":. . .
r.
,. :,""1
.i 1. : .. ... .. , . t.:...'i
. , . .-4.,
. t.....! .- t .,
l e .. l
.i :::i 5 i,
....c. .::ri
- i..n:. I i : l-- e <:t:: i:.:, ci e t,, ,.. ... , ;
w__________________________.______ __ - _ ___ _ _ . _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ - - _ _ . _ _ . _ _ _ _ _ _ _ _ _ _ - - _ _ _ _
. ....-- ...... .... a u. s; .a . . 6. i :..s . . : n . . w.. , :. . . . - .
O U Test Program No. 43250-1
. Run No. 60. ECA,1* (SSE) Page 26 og 63 Report No. 37-51955-S5
{ ce a. recr e t - 4.16 k V MrfAL Cue SwfemmR
?
, t e e. sis e u . Ceaws ba. E. L. G 20'
- ~ s yr vs; % .
4..lG KV lic-stisveied SvfrFM Figureh sP- %9 a l I
= 100-- __
80 - 1 SPECTRUM DESCRIPTION o m ,i -
,,,i + 11 , ; .
,, , . r.,, i 60-Hation Input : Front-to-Back/ Vertical } q 7,S, Accel. : 100 g, Feak 5 .
4 Damping : 1.% . .E. ' 3 _ , , Dash Curve ' Test Program Beri:.SSE RRS 9 30 4 Point Curve : Horiz. TRS, analyzed at 1/3 i a octava frequencies by Boriz. E j 8 Control Accelerometer ;E 20 -- 1_ Solid curve : Customer's Boriz. SSE RR$ 3 ; 3- ,
, . _ ,., v. m ..,.,,..,.,m,,,,,....,~,,...-..-
r I1 ] i . I'. I j ( 133' 8 I I 88I. . I I I I I '
] Ii ; : I i,
, ,, , , n , , ... i . . .. i ... ..... ..o . . . . .
o , . n .. .o . i in o., g i . . . . . i . ...ui.. . . . m 8 - -' - ' '
,j
.= , i
,1, ..
o, ,m,.., , m, ,
..m, m..,
, . =
, 1 - , , ,,,om
,i .
. ,,m,, m , ,. , , , ,, u. - o, , ... . .. m .o. ,, m , . .. .
= 6 - -
- n. -
.___ 2=
.6 w
a -- c 4- - - -
' ^
2 y ' ' ~ 3_= . g e me - . i
; e' e _-
*' 2 -
ll .. . . I l'
., c r ,.
; J
, ,, ;l
-[i, P r j ],a 7I , N *1 i i ! L i LT . i 1 1
,,, _..d' a 'i 1 I , 11 IJ n 6 U tiq. i i t. I , ..!
i, ,l i ' l {! !!.7 3
!l r 3 .t :1 7 I 'L i Fi 1 A isi i i .i#l 11 , i i ! . .
u.i . . .r i i i na .a ..a . . . i .n 1.. .!..m...- i i. .o. ... . . . .... e 1.0 - - - - - 0.8
, y , ;-j y ; , , h
.i: 1I . I a !Ii1aiii. 1 ii ,i1- : 1 i ii11 1 f 1'I 11iEi t.L' ! Ii, 'I! I~I I I I i g*
- 0. 6 f'
0.4
-- r -'
-- -- I 1 . -i 2- -^ '- F ' l ; h1 [.
0.3
=
t' 0.2 -- !. - t.
. - e _ _ .--= '
M T .j e -,
, u ,
i r- - - so
- r _-
1
. i ,i sm i n ,
w ii, , i a
.mi ll,, I .
tili t I 151 i ff J ii i !! !!' 11i I I il - 3
' lI l
,ll, I 4 llIlli I . Ii *l . J l J '4i e il 3l*+ tlI I . .,e . i j.
0.1 . . . , . . . . . . . ..,.g,. . . . . . , . . . . { 0.4 1.0 10 ., 100 Frequency (Es) 8RcWA! 30 VERI EE.CCT. IAlb. - Mo>gL
- 5NK-3gn Figure (p .
J'8' , COMPARISON OF THE CUSTOMER'S HORIZONTAL SSE RRS TO THE HORIZONTAL
,f j
FRoh"I-TO-BACK TRS OF THE SEK-350 SWITCHGEAR TEST SPECIMEN j G A Z. MPL
- IR22S006> IK 22 9007- IR22.'scob .
l
CoMc M Eu r - RC ente PvMP TIIP DMAAL $wrtMM ~ pl FULL SCALE SHOCK SPEC [ RUM (g Peak) (V ' 1.0 0 10 0 100X 10'00 0 i - hem't em- bs r*% svsrru :
} f.Wnew. .G.tesmm Ra E.LcAMPmo C to' l2 %]
__.-_......o m.w
- i. -.
----.> d _... .._____..... 4-xx-,- . .4_4
- Em_rgy . e.__ M .& --,..-
----. 4_=,,,.,+_ j.4 _ , m -MTM3iyeh.%M,ww.A.,,W
.... a-.- - ,
, em ,,_ _ t ,
w.i-i+i"-f e -H 61 - i-- a--ip.i Vig-7-;6 , . .
- - =.__...== _.
w ou.
.= _ =: =:== u == ..=,= =..sW-ff'Efth,'~~"Ab~~@u m_= : = : == =.-m==.-,-_=:.=._.._-=...==._=__:===>=
- - - - ==:2=:n l
= == . == i == == =_=_ ==_ _: c -== =.::.:: : ::,
b _-._.:==-- :.- - _.- _.__.._ .. _ _ . . . _ . _ . _ _
": H _ . E_R, _.7h. . W, ..i.V.. ._=W_ S. ...SZ..=a 1..__.w. . ..,.__ _..a.iz..K_W=_W._M
..=..=.._=.6. .. . . ., . .. 2: . , 0. ..,.&.j : g,.,,. u&. _
=. =...=. .~.=.=. ,. . ;
,,.,____..,.,.,_-__._,,.fg..
o . gg,_. ,, ...
._. ,4 _. _> .. . :_. 4. . . ., .,. ..+ . % v v _ _ _ , i n. . ., .
. ., 4:.(.m_.,.........._..._._......_.....f._._.
. e. 4.
- 1. .. .- ... :._. =. . .. . .. p. .. . . . .j .-
s
=_. : ., .= ==.= : ==._= = ===
. :_: L.=. e. _.j _,.4 :.. a___==.=_
___.. _ _ _. . . . . , r_ .. ., _- -- . : = a = = :: _=_.=_=.n...=.==._.su._.
. . _ ._ . . . _ i . _ _ a. . =. .. ...-
_.._..._.==.=.s...-. _ _ _-...._2...._..
.__._._-_...=_a=_..,.=..__.s.__L...__
.___.._..__..._.a..__3.
_ _ _ _ _ . -- ___ _- __, _ _ : w -
.__.__._,..____.__..__..__.__.,_._.A..._.4._.m___....,._.
... _J._ _ ..,
_ _...4__ m . ,.. __ ,., _ ~.- _4. _ . _.1 . _ . . .._
..__.w. _. __
,.r.n . g , _. . . .
- m. .u
.u
.. x... m.3.__
w .s w. . . .
, . ._ _a ..
.. u . e. :
, , . ._.u s . _ _ _ . E
. . 4
- . . .h.v
.m '?
# ~9 '
h.4 ~.aI. ' i . M.1 .mN.a *l.*v.f. i
.h. .i.h: .i.' .i
, o.Ii.. o. ._mii ,iO..
. ~ . .
9.' .-tai
.i *M_ Pn' .J ,2.
.,._._ W. EiTi.i-l_ii.id...
.7
* :.u. :.: '1" u. ..x, o .:.o.,
_i_a. ,
- w. . .a.
...Ti,C_U.L'.
n.1.. i.y. w P .
', . .t.' 1' .:_w*"1 . . . . . . . . . 'i n"D.T-.u". n .. .m.I iii.1 i
._i*
'ro
.u a
i Li Ti .:.u. -
- 4:, .7'1.1. 11 nm e e s.tnn. e u .>.:. ., . . . . . . .:w*$x "~9:
.i.
s.e __
._.L,...-
. _ . . , . . . _ . . - _ . , . .. 4
; w..,
4_.,...
.t ..,..w,_m_,_,,4 _
s
. ._ ., . . a. ..3 _,q. e .4 - m___
-9
. . . . .E _ .. _3- .iihi _. _-i._mb i d 1 e 4.
--- .- i i 94 4
""1 __..---r...._....
.--i".H@-iii. .-H_...44 g i4> +i--4 QMHi'Hii--ji
- .s. . . . . -: . -
- - - Hii:--jhg=-b
- .=. == : ,.-H.-:..--
W.:===:n:--
- __ _i-;i ; '=.--::: s ::-= =i! -A: l e.-=i f -
.e . . .'A : v-- : : n. *:: .. - ::= .:b. . . . - -
. .x os r: .. = - = ::=- =W= . &::~ = * ' 3 . U+X _x--------~? I .=- - - ~ -
.. _ _ _. ._~ - -- =_.. . _ _,.. . =. .:::E. .. .!_. 47
- c. _ _
0 _. ~ -
; = - - -_-- " _ -
_ __. ... . . . . . . . _ . . .. . . . .. _.4. . . . . _w
'G__.:.:=_._._:.t .
.!. _- :_4
. = . - _.__~e*:=.
. F.
_= _. . _ L_ _ _ . _ ..
.- i
_:x, CN._-.._.,._._.___.,.,.._._____.._u__.< . _ . _ _ _ _ -.----
-.....3.._._._
._. ~ _
g . , . . _,_._ . . , . - , . . .u,__-_.,-_..V _.... _ - , . .... .. g!
- _..... _ _. X. . L . ,
, 6. ,
. _ 4.. , .. :
.7 t-
- = ! = = =. . . . .-.-: _- -:.- . = 8: d::=-:. .:n . "_ . . .,., . .W :.. . -=- .:::= :r.!:. . . . .:: :: .
I=.: 2. . ..:.:
=::-: . . .:.:: ..
- j ..:m =:_::_je.:= . .=_:h. ..:-:- _ -_._ 3 2. 3 ::::: ~ ::-:.:::.:. ===r =g: =2 -: -.= .t: -. -.:.-: .n_-- . . -._--_ -_ -. -_ ". - - - . - l q
- 5 - - .:1.." Q: - : 2::n==. .: -.: *. :.:_. x__~_M, .._.:.<=. _.....
pc n m:. x =:.r 3. c: .c.,.,.
- ... y ... Pf
._a.lu_- ..!
-:==.
. . . .:... X, .. n:..::::::.:.r.:::
. ..1. . . .. ._. .2...::: . . .: .2.. :.J.:.:.:.:
.. .. -1. . - _. . . _ _.
..__.u ._. e. .u_.=.=,t. . . . . . .
.._9_._. _.* _ _.. _ _...e. _- _
~
. ...... i
.nt =....,.,_ _-.,.
...2.,_.._. _. _-_m..... 3/ __. _ _...
_o r -,. .__. 1._.... 1.. . .. . . p.
.- . ! __"L
-~~,A._.
n___...,, _o ,g _ 7 ,, _..__,_.,.2... g,,. , _ _ m_
. o y _..,,,,.._
.v s_._.. ...
_.. '... ...a._..-_. 1 d ., ._. _- _.......t_.m. _ -. -, _. pu:_
.; .a
.s.4 k....q....p._ ..
- . s
- ___a.
- t. <o n n. .
- a. t.l.a.w.u. . V,7;a. D.1_.a. i.Q .. . . . ._* . . .. .. .. . . ,1 .1. ..a. . i
_ a..u.a.. 3. . . .
- 1. . , w. u.:: a.u. ..a. 1_q. .u. .
,_ _..: u n..n...e :.m.1',ni,1:=.. 1. m :1.a .L.t. s.. a. . .
l It.____._. -__. ._..... , _._- p_ _ _ _ _ _ _ _\ _+ 1.., ._ __.n._. m _.4. . - .. _ _ _ _ _ . . . _ _ _ _ . . . _ _ . _ . . _ . _ _ _ _ . .
; ', /--- _-- V / i i _ .r. = m . , - ._ ,_ __ . . &
.a .a. .a_ a _4 4 N.. y ; -.. _x.___ _ _;._- , ,
j.. ._... . .u . _ w ....__ . _._. . ._ _ _ . . - =. _. =._=. - - =. _::._ l =_._.
. . _ . . . . . . _ . ._ .: 1=. :.1.-==. . . . . _ . . . .
. :.. ., : u_.:_:::. .. : : . : =_.u. :m. _i :. . . . . . . . . _ . ._ _ _: =. 3 :a =_ =_.=_=. .
- ' n._. .u. : .: : .:.._ .n._.r.: =_ _ .:. :. =__:._n,............
_ >~ ...:.. . . =. . ra.a .
= =:. . . = := _..
2_2..=. : - .....g:.2__==_.._._._._ . . .. .=_ _ .:.=..=._=..=._.
. . . . . . . . . _= ,: _=_. .=._:r . =__ : u.
......:...1..__
=._=_____._=_ __=. . .
. . . . _ .t: ..
.i
._.b.....,._..s..4..___....._..s_.x___._......
a g g . :s 5 y = g p y . _ _......
.__._.__ _ . .-_.2.._. ..._ . _..._. _ . _ _ _._.__
. . . . . . . . ._ .... ......._.._.._._.__....._.+.r.r...en .1 . .
. - =... ==
.wy.z.w. p w:a.:n. . c.ra. ;- .1 y .p: @: . :n_; uw:.
- : _:yyy.u. := .s. .,n,: .: .: ._...
; w= : : =
u :: .-.-:== r-==.. _= u_ .-: r_=_ =_= =-=::. =_=_=
..._.:=_=_=_.:=_.a___._._=_.__._.._:=.=.=_=.=..=
a:. .:. .n. .
. ,:1n...a._.
_. .:..._. _. ..__._ . _ . _ . ._. ..:a.__. . . . . . _ . _ . .
=5==
CK.. L=V.../. 0. 9..t.L.i.
, ;MG F_R_QN.D. ..T..e..:1M.
_ _" '. ". .. .. .. .. ._. _. _ _ .. .. =._M _ g. ..s.... 4 /
.am. .. . . -- - , .,. .:-B.. ..
- e. m:. _z=> . . . ....... =_ _ ._
2.2_. .a .c. . . . . _ 9. .4. 4. ... _ _ . ,e........
. _ _s .. __ . a.., .a
..n . _n ..,.. , .. .
- . 4 J . L. :.
u sa.: h.1,.
. .. .t
.._.i
. . .__... _ . . ._ nL . . . ._ .a . .. ....L . . ._ s w uu ~~:...: rs ..q-..L4...
r
.:a:L A:wua. :A , m. . ..j. . .;. a. ..-
a.- .
. . a . 4 y a:. w a . .. . u . ,v, n .:.m. . . . .w .. p . s..: :.2 .: , .m 4 - .
a . 'n.:p: m.m:h.
~a a .u ;4, u.r :
n
;m n
- w. . . n.a,.j. b..u..
L
. . 2. .' O u -. . .u. . . . .m, ._...,<
- a. . .
= . .d ,t. -h D. : . ... .
. . v r.3.
.u v :2.'., H,. u. m, , -,m. . -. . . .u,... ,.;,o.
v, , . , -- - - . - - w -m t u- _. .__ - ,a en.. ..
. i,r.--
= :,- .- .
, a s s , ..
, ,-s u.ud , , a s . , . . . . ,
ice C, to too . .
.A f" O _> ' facMtAL GMT ~ Mehn.
- Pwer < r m encyNhc (Hz) _. g;, f/
GA1 MPL # # Q.2.$00lt 'l lM.ts 008 S i m , o q \ q c nl 7~ P L
; Ansool+ k 1A115045'- L OCATION NO. _H.c a.
Fwe8 w _ _ - -
, b2 of 3 .'
cc M 9CM E NT' FEDE-302V2 MS,I \/ LgAK A(g (.odTReL bYSTEM bi-dWER i.-...
.- -T, . m m o s u em e ,,, >.>.... m --., ,,,,
PAELfB STWT DT.aaEIC L.63tas (888) LOCAT*Ie M
- boxILtAt,y bLb4r $.l*
KYSTre MSIV ts. mA4 ar eserna6 SN-
.< == -i = =am ,
S. , i e i l l lk l %. ~ t..
,/ l i
l-i
~
i e i l 1 l i.' : -____ _______i---'-- : i e i s- _ _ _ _ _ f ,,_ _ g 7 , g _, i.s, - -_- -
.... __ __ _ _ .: _ M6.e_ade _ ___'
eu.are74 ! l ' ! non st.n <tel -- e e , i . .....i. is s. = ....... rimw.n. =
&cuttht. GLscr. loMFbc McDec # 2 C H-G -MI-lU ggr M PL.
- I E3 2-C 0001, IE32 coco 1B. IE32 Co002 F.
g COV 110 .c .,- u. . :, CO% PYl .
, , , a,.....,
e8 O FreaNe 7
l O O O I I l l l TABLE 181 ! CONTA8NMENT VALVE 5 AND ACTUATOR 5 COMPARISON DATA 1 ) l EXTRAPOLATED NATURAL EN" DE54GN V" ACTUATOR V VALVE l FREQUENCf OF RE DED LERATION O N NN j PIP 9NG SYSTEM ACCELERATION UEAN O UEA M WT W E RECORDED l EARM OUAKE l l j ' urge Valves (42" Henry Pratt) of M14 System wnth Bettes NS 0.55 0.48 0.63 0.72 1.4 3.0 6neumatic Actuator 41.6 Hz EW 0.18 0.48 C.48 0.18 1.5 3.0 l
- MPL Nos. M14-F040 V 0.30 0.28 0.54 0.58 0.57 3A l
! M14-F090 58t15 0.M $3t55 2.13 5R55 52 Model No. T-420-5R2 l i 1 l sacuum Releef Valve (24* 1 *enry Pratt) of M17 System N5 1.76 1.M 0.74 0.67 8.M 5A i j with Lemet Torque Actuator 32 Hz EW 0.46 1.M 0.73 0.17 8.73 5A i MPL Nos. M17-7015 V 0.50 0.73 0.53 0.36 8.53 5A M17-F025 5R55 0.30 SASS 1.17 1R55 8.7 M17-F035 M17-F045 Model No. SMS.0015-H3SC l f i } i l l 4
. t ta
\
- j. AJ ,
1 EQUIPMENT SELECTION CRITERIA ACTIVE SAFETY CLASS EQUIPMENT REQUIRED FOR SAFE SHUTDOWN. EQUIPMENT LIST COMPILED BY LAWRENCE LIVERMORE NATIONAL LABORATORY, WITH FREQUENCIES HIGHER THAN l 14 Hz AND HCLPF VALUES'LESS THAN 0.5g. l SUPPLIED BY MULTIPLE VENDORS. l l ACTIVE COMPONENTS QUALIFIED BY ANALYSES. i VALVES & MOTOR OPERATORS SUPPPORTED BY PIPING ' l SYSTEMS. t f ELECTRICAL SWITCBGEAR & INSTRUMENT RACKS. VERTICAL PUMPS. BATTERIES & BATTERY RACKS. CRITERIA RESULT IN MORE COMPREHENSIVE SAMPLES ' OF EQUIPMENT. i s APPROXIMATELY 75% OF SELECTED EQUIPMENT TYPES i HAVE ALREADY BEEN EVALUATED WITH AMPLE QUALIFICATION
, MARGINS.
BALANCE TO BE COMPLETED BEFORE JUNE 1986. T I k i y I 'se i
, - , _ _ _ . . . ~ , _ _ , . _ _ _ _ - - . . _ _ , _ _ _ _ _ _ . . . . _ , _ , , , . . _ , - , - , , _ _ . . , , _ , , , , _ , , _ _ - . - . _ , , . _ _ , _ , . .
l O . CONCLUSION THE 1986 OHIO EARTHQUAKE: HIGH FREQUENCIES L SHORT DURATION LOW ENERGY LOW VELOCITY SMALL DISPLACEMENT I STRUCTURAL & EQUIPMENT DESIGN HAS SUBSTANTIAL MARGINS OF SAFETY RELATIVE TO LOADS & STRESSES INDUCED BY THE EARTHQUAKE O 't I 1 1 l t O
i O DETERMINATION OF DESIGN GROUND ACCELERATION AD^ COMPARED TO A yp FOR THE OHIO 1986 EARTHQUAKE o MAXIMUM A p = .15g FREE FIELD o A* 9
= .073g PREE FIELD
- BASED ON PROCEDURES DEVELOPED BY HOUSNER, ARIAS, MORTGAT, MCCANN AND SHAH, VANMARCKE AND LUI i
l e O
. O . COMPAR'.2ON OF AVERAGE SPECTRAL RESPONSE USING THE R.G. 1.60 SPECTRA AS THE BASIS OF THE COMPARISON. j R.G. 1.60 0.15g OHIO - 1986 0.02g a i
- AVERAGE OF 4 FREQUENCIES USED BY KENNEDY, 2.1, 3.2, 5.3, 8.5 Hz PLUS TWO DOMINANT FREQUENCIES OF OHIO 1986 EARTHQUAKE, 4.0 AND 22.0 HZ i
b 4 I i 1 O 1 I i
O i CONCLUSION: A THE OHIO 1986 EARTHQUAKE HAD A RESPONSE EQUIVALENCY TO THE DESIGN EARTHQUAKE (R.G. 1.60) WELL BELOW THE DESIGN OBE. ; THEREFORE NO DAMAGE TO PLANT SHOULD HAVE RESULTED.
; O '
] . l i I 4 4 O l' i
a4* 7e* o ~].NQ^; x' 4;=l ;
%. V o
e c
. o w-e Y l
C '
\
a _ _ e c Em *8
- c a
e ATTICA o o e 1 -
- e 1/31/86
==
O { cli
* ,p 7 o a a e 3C
~
a o e ANNA e e
, g me p g .
- C: O 8
a O c , ea e 200 " 0 o p fi~
?
g Q 7/27/80 e, t, e . 9 4 . t g
) :
D
~ . ~ ..
C
..* C N* co M ~ h. e.e .
. LA e se too m as EPRI DATA BASE
- - _ _,_ .-_.,,,___,..~,,,_,__.___,.__..,____,_-._.__,,.__--.,-.___m ,_-_ -. ,,,__
- g ,
MAINSHOCK JANUARY 31,1984 LOCATION : TOWN OF LEROY
- LAKE COUNTY, OHIO l
DISTANCE TO PLANT : A PPROX.10.5 MILES l APPROX.17.0 K M. ! I i ORielN TIME : 18hr dem 42.3 s U.T. i l U.S. MODEL JEFFREYS-EULLEN i O TABLE l LATITUDE : 41.6 5 0 N 41.8 4 9 N LONGITUDE : 81.18 2 W 81.10 5 W I j FOCAL DEPTH : 5 KM (CONSTR AINED) ) Mg (TELESEISMIC) 4.9 (10 ST ATIONS) l l l SOURCE: NATIONAL EARTHQUAKE INFORM ATION SERVICE (uses) O
-------n-e-.---~~._--,v----- - - , , .-ye.yw.,--,--,_ _ ,_- .--e
O . EARTHOU AKE M AGNITUDE & INTENSITY l l 1 ., r 4.9 M, USGS (EQUIV ALENT RICHTER SC ALE) I 1 MODIFIED MERC ALLI INTENSITY VI ,
- PRELIMIN ARY-WESTON GEOPHYSIC AL i
O j . ! e QUESTIONN AIRE
- l l
e REPORTED EFFECTS > i FEW DAMAGED CHIMNEYS CRACKS IN W ALLS I CRACKED / FALLEN PLASTER , BROKEN WINDOWS I -
. WELL-W ATER EFFECTS b r
- O
! l ( l l
O O O g g,g. 81.38* S1* l - I l ,
, w-v w_w ne Lake Erie w
- v . ass PNPP 8'
' y ev-v,,, "
,,,, 8 - v "-81 .ar 8 av 8v ev-v V av
/ , , .
- y ev-v W w- see-ev 41.75* - $ ,,
e ese 'see w-v V ,1 w-v "-',w-v ev-v i ly , w-v "-' y yg '
,l.L.
--1, v w-v , ,,_ ,
v u. g.,, " w-v v v v l v me-w v-vi v ' y w-vv -n n-,, vi eVI',88 ,l l
'-" a "
-a - , ,v y '
v _p ll = w
,, me ' LAKE f.
, n-v Mainshock ll m-avl.
1/31/86 y 'l a u-. ,I~ -= v. l . v ueGs mEwEv3 .v iv IV av v ' w -. me l v "-' lol IV . --.a 8'-' av ! v ev-v V
,, ..y I av v-vi 8v a t
vi me-w I.
" - ~
- ,, .v -
av en . l ::: av 4 1.. -
,, .m+
l _ _. . . . . .. j .
,v . ,
. ...m.
l CUYAHOGA , GEAUGA 8' E A um I w-ev
- O
. =
0 o O 1 J ' i
\b
$ o . g4
. s -
o nuonez f h 3E *
- I SITg O enman R-RO 1
EE .\
~
ore ~"" : os ' ws
\ omgrr :
- N g
.. o /.
i
, cf gr -
Mm .. ?
. 1
! d i l O
);
i ammes lose ==e en mm.. .se , m O O a.e-se 0 ee ' C no se o oa , O se sQ Os n . O O ..
,)
'3 W*
" en se
..[ l O 60'vClfuct es'n e,* , l l
l l l l Figure 4 Seisetaity Map
O .
, FIFTY-MILE SEISMICITY l
l RELATIVELY LOW SEISMICITY
.i i NO CAPABLE FAULTS lO HIGHEST INTENSITY -
MODIFIED MERC ALLI VI i
! 1
.; HIGHEST M AGNITUDE 4.9 M g
. 1 I
i l \ - e8 ,O l l w-w yw_ W g w g-_
A O . PNPP LAKE Erit I paamy n CALMIO INJECTION WELLS IW-1e _4 g e43' j ename 84W88
, elW-a
" $mi O
PAImaevILLE J L3meV i i e site e g,g N - .ro m! usesteswan .
,,e b ,,,,,. ,
LAK LC0JNTJ,_,_,_,,,,,,,,._,__._._._.,..-.- -.-.- F SEAUSA COUNTY 8M, %ss l l l 81'13'30"
@ Melashoek steeAH wekeene 9 j MILE e in)eettee we#e O MM
a -u ,- -_a__ ___a- _ _ _ -~ ---_ _ __.____ _____,.-..- * # .,,----__-,_.___._.-,.a- _ _ _ _ _ - - _ g wme,.mh,n, s -
.- - -n .n '
! ^ 5%
~
l n-;s 1 - x.glsWt . w :w nu7
' NbhM) F .s ice
- 4.*woh
- .M ,#. _
-- 7~ .g .
m, , h, gy mm s. , _ .-m . w r y J e .: . 9:. Maif
- sgy)l
'fhh&._ .
* '4 g "R ~ M M M a s
., Qgg,y4, s.
O S. ' kh k"Pi s 3. ' gf-
.. _ _ . . . . . \w -
w -- A
\
O . CFF5 - CHIO 86 SE2 - 3 INCON5!ST - RUN 3/7/85 - N i Oo O O O '
\ ; O ^ ^^
f b O __ __ O i A ! A Og O Qo . I e i ..
' c" "'c "" ' ""
O
3
- e O .
5AFE SHUTDOWN FARTHOUAKE I I MODIFIED MERCALLIINTENSITY = Vil PIAK GROUND ACCELERATION 0.15 g REGULATORY GUIDE 1.60 SPECTRA - I . SITI SPECIFIC SPECTRA O - l . j e 5.3 Msto i .5 j e 5.5 M ata * .3 l l 1
- i i
I i
- 4 l
1 i l . , i 1
.-_ . _ . - - - , _ _ ~ _ _ . -._- _ __ - - - . _ . --
I l O . l 2 PNPP - BASIC SSRS 14 Horea. Comps. Damping = 0 05 10 66 i i i i i i i i ii il 6 i I i i , M M m m muun LM E f . y 0 08 - I d _ 7 E -0 3 10 > y ~ a , t v j x= eum m m W
-1 '
10 -' l ! !!' ' I I f '!' ' ' ' '!'
-2 -1 0 '
1 10 to 10 10 , ll]$1ll l1ll;i:08
#'t
.lst i:
p l } s ssas - 14 nonIzourAL conrowrNTs DIAonAx vs na 1.6o 18 h ', .--
l O . PNPP - BASIC SSRS Ci4 Components 3 Camping = 0 05 2 10 i i i ii Il I i I i i i-
- i I i ii 16 g -
W M 1 i >= - 84th _ c - O d, - I C ' j C -0 \50th t 2 10 / \
- W _
i m - 4 g
~
~
l - 4 I - ! 10
-l
/1 I I I 'll I I I ! ! ' ' ' ' '
- -2: .
-1 0 1 10 to 10 10 ,
PERIOD (SEC)
- o iEEi4 I 88R8 - SOTH & 84TN PERCENTILE V8 RG 1.60
I
~
- O .
2 PNPP REACTOR FOUNDATION 31-JAN-1986 Demo.ng = 0.05 , 10 i ii i I ii
- 1 I I II II I i Il I i; [
l l _ { _ 1 j u ! 1 s 10 a } .
, >= _
>! u
- e -
4
' O lA,l
_i 1 e i o -0 s
! 3 10 \
\ l
* =
l 1 _ l _ l ! -1 tii ! i i i,i 10 t i i sit i i i i l -2 -
-1 0 10 1
- 10 10 10
- PERIOD (SEC) l) o !B5 4
l ms - Lamor ao vs na 1.so
O O . 2 PNPP - BASIC SRSS + PNPP (1/ 31/86 R e e s .1 Damping = 0 05 10 i i i i i il I i i i1 -
- 1 I i ii 11 _
g - W W l N IO 2.
=
Y N
~
w
= -
Q .'
5 _
C -
,. m,34ig m
7
,,, a -
c ,/ g . M. Sots %.. C =0 10 / .
~
sasic
/
,/ .,....,, _
/
j' /j .,n>
/
-1 i i i eit ,, t , i ii' j to i f I i !!I i
~
-2 -1 0 1 i
10 . 10 to 10 i PERIOD (SEC) 3 511:L' 3 R" ..
= ;id, O
l l SSR8 & PNPP (LEROY) - 50TH & 84TH VS REG GUIDE
O . 0 .p.ng = 0.05 2 M1TCHELL LAKE RD., N.B. 31-MAR-82 10 i i iii la i i i i i i i i Ii II g - 3'A 1
=
10 V N
>= _
g _ Oc t,a 6 c c -0 3 E 10 V N N l h : N 10
-1 I I I I Iil [i 1 I ? ' ' '
-1 0
-2 10 IC 10 10 PERIOD (SEC)
O -rt:"::::: Sm m R8 - MITCHELL LAKE ROAD - NEW BRUNSWICE
O . 2 PNPP - BASIC SSRS + MLR + PNP 1/31/96 Demping = 0 05 1 10 ii l
* - 1 I i i i li i i i ii ii i i-
~
i l lue M De
.nup f.A,1 M 1 10 U
M "
>=
O .* ' Q ,' *.....N., ,m 84i6 4 './ % oy l p s \ C ... % $50th o -o ., O 10 / \ w - 3. fj \ gasic s
/ ,
- / . .. ...
i )
, / _
j s' \ to
-1 li I I_ tif i i i i !ti e i i i iil 1 1
1
-2 -1 0 1; 10 -
10 to 10 PERIOD (SEC) ,
,, 1.:N.'
O 1::R" SSR8 - BASIC & NEW BRUNSWICK & LEROY l l l
,_..__-_.___.._____,,__,____....__e._,____._, _ _ . _ _ , . . . _ . , _ . . _ _ _ _ _ . _ _ _ _ . . _ _ _ _ _ , , . . _ _. _ _.. ., - - _ ______ _ _,__
f h
- O .
CONCLUSIONS
, TECTONIC PROVINCE APPRO ACH STILL V ALID
- NO C AP ABLE FAULT
- NO TECTONIC STRUCTURE 1 i SAFE SHUTDOWN EARTHQUAKE INTENSITY l
1
- MODIFIED MERC ALLI Vil vs. VI (1/31/C6)
SITE SPECIFIC RESPONSE SPECTR A FOR SSE I
- 5.5 2.3 v s. 4.9 Mg (1/ 31/ 8 6) 5.3 2.S vs. 4.9 Me (1/ 31/8 6) ;
I l 1 EXCEEDANCE OF S AFE SHUTDOWN E ARTHQUAKE l l i
- Sid O RT D U R ATIO N ( < 0.5 s e c.)
- HIGH FREQUENCY (20 Hz) li ..
< O e EXCEED ANCE ABOVE 84% EXPECTED i
l
-"em - --- __u _4- __ - p _...__- - _ , ___
a, _na,,a__ _ i i O l l - i il " $t j i
.g (i
i I( i 1 i l l
. , -1 ,
.3
& I I s (
{ ! s
-l O
$11-l inat
';h -1 ;
1 1 '
. IL i
~
p il ! e
~
l Y ' _ $eee ee m ' 1 ll.,
, i O
l l ~~ l I l ' I l'i-i ..t j
O O O ATTeCassiur i SilsamasassiCat. 3, amsaran BSNSTSCatles Mast InsCtaat feME MdNE 88TE penEness er 3153855 le4Tl44. ACT8vsfles . I. SALIM
- a. Seaeseaay SeeI.et8en Amstroie and eeeleottaa of felt reportes ergestee the cellectles ed desage and felt reporte; Preposettes ed leeseisest espe med GI gesetteneestes detesseesties of epicesters receemet latervleus ( M & MI): asotew ed heeteticet esseeacacy:
ether M t cettected dose (tamis, ashoots,etc.): asepamos se MC asepenas to E W e to f$at B. Eatt' - " Isetrmeeterie Isotett portable seteese errey to Iwate of terehache; Centl===d emetyens ei efterehack date: notere&as af teseheck lacettees and campere to heterseestles ei lacettene, feelt ee&eedneck; yleen seleties , J to See ei 8estresete1 eelee8c8ty; ProIlatesty reepenee ae meC es testrussetat serthgeeks dete P to MC W e FSet at. enesis sonnem aseems iregeoecy centest of eser-ito84 Aseems and eseIeena meeeered aremed est8ese and .eetthgeshes; reepense spec &re aeietIse te dee8as weIese; W 8e and seeIysee ai1 apptIcable j
- Seatrummetaties (ce1Ebeoc1ene eed locaties) ets ; rles dote fee this earthgeeke Aseems lepect of bagh fregemecy. hi p lapst to GI roepease en ef fect ei selease forces empittote, ehest derettaa eeteetc metten; em Perry Flent; j Seepease to MAC
-; " to mac update FEAE w - ___ - _ _ _ , _ _ _ _ _ _, _ _ - . _ - _ - - . . . - - - - _ _ , , , , _ - , . _ _ _ - - - -___,_,---,_e _, - - - - - - _ __ ,_ _ , _ , _ , _ ,_ , _ _ , _ _ , _ -
_ _ _ _ _ _ - _ _ _ _ _ _ - . _ . . . = - - _ _ - ____-. - __ _ ___. _ _ - --. _ _ _ _ . . ___._ _-. _._. -. _ -.__ .-._ _ _ _ _ _ - - - ) O O O i i i . ATTAC3emart I SSIsasssastCAA. & EdesastCat. ImuusTIG4TBW ' Puest MEAMAR ruMER Fl.Asrt SSTE l 1 letYIAL ACTEviT853 poemas er SW 1I1. Emasasy I l j Check ett repeated geolegte seemettee; Seetow gesteste date & L8terosere elece ' j Check femendettee esed8stene of damaged hamees FSea (8930) etsurtesee, ete.; EseSee resses remote seemleg; Check gestegy le opteestret eres (bedteek & eettet phetestaphy (les locol). rader serileiet); (SEAS) teedest;
~
toepend to enc gneet8ene se geology Map gestasy 8e apteesttoI erse ier be8eala fahrte se surfleet estattela; } toterates opteestret eres structures j threagh ametrate ei ett & gee melt tege.
~
i emetye8e ei esteele reflectlee serveye
- o t ,te,e1.te,, e.e ,,apsi.ter,); i j Assess strees condittees Se epicentret ones as seideered by meI1 be d w a &
byerefrecture dete; l l i Frepete tap of foreet8ee espe to better ) seeses otracteree;
^ Seeses of meter unit lege for sett conditsees
{ 1 . and Sedteetteme of bedrock samee of ha d potenabtIity; Campete & 8etospret estettag geophysteel dete; [ Seelante regleest etreeteres t Briefly esam8ee enjer fracteree le PSas/F548 Frepete structata! & esteenteeteste eyetheets; Beepend to M C l g ydete FSAR l 1 i l i ~ l } . j 1 ; 1 i _ . _ - -,_ _ . _ _ _ . ___ _ -,,_4_.,,-__---_w_
, - ._r_, % , .._ __, ., _,._ __ ._ , _ __ w , _ _ ,,,
t O . SEISMICITY AND CALHIO INJECTION WELLS i .f I NO RNOWN SEISMICITY NEAR WELLS
- BEFORE/AFTER JANUARY 31, 1986 EVENT . EARTHQUARES - 11 RM S OF WELL8.
l NO SEISMICITY BETWEEN THEM i i
- I
; HISTORICAL SEISMIC ACTIVITY IN AREA BEFORE WELLS l
I
, OBSERVED TIME LAG 8 BETWEEN INJECTION & SEISMICITY !
AT OTHER LOCATIONS IS IN DAYS /WEER8.
. HERE ar 4 YEARS ,
SWARMS AT OTHER RNOWN LOCATIONS OF INJECTION l RELATED SEISMICITY. NOT 8EEN HERE !
- O *
, LACK OF HYDMOLOGICAt. CONNECTION BETWEEN WELLS !
I l I I : i I i 1 4 1 I I i t t O i ! l t I !
4 PNPP LAKE ERIE penny CALHIO INJECTION WELLS IW~ 1 " - 41845' J l enAno Riven TA . iw-
' - S Mi.
PAINE $ViLLE ! LEnoy m
- s. ic M ;,
USGS(DKWEYJ
.st/ *si t o !
l ._.,.g_c.p;g.,._.._._._._._._._._._.2a _._._._._
\ .
1240"
@ Mainshock sis
- Af tershocke '6 1 MILg O o lejoetion wolle 0
1 MM e b. e
O . I l
.i..
t- gl,
;I
* "L ,
- l. .
i ! I. 1-
.l .
I l8 i l- l8
?
5 , O 1 l-III 1-li$ l11l1 p i i
._i.,..II ..,
I h I. ;
=-e . . .../.
1 I li l - O , ,
! y!
l I l-Il'! ,, I \ - - - - - - - _ . _ _ _ _ _ _ _ , _ _
O -,
SUMMARY
AND CONCLUSIONS PROOF TEST OF SEISMIC DESIGN
- JANUARY 31, 1986 EARTHQUAKE
INTENSITY VI, 4.96 RICHTER
- PLANT STRUCTURES AND SYSTEMS
(. UNAFFECTED BY EARTHQUAKE f, NO CHANGE TO' CONCLUSIONS ON GEOLOGY AND SEISMOLOGY
- DESIGN EARTHQUAKE BOUNDS JANUARY 1986 EVENT (MAGNITUDE, ENERGY ETC.)
- INCLUSION OF RECENT EVENT WOULD NOT CHANGE DESIGN SPECTRUM PLANT SEISMIC CAPABILITY ABLE TO ACCOMMODATE 1
JANUARY EARTHQUAKE
- SHORT DURATION, HIGH FREQUENCY, LOW ENERGY O - MEASURED RESPONSE EXCEEDANCES OF NO ENGINEERING SIGNIFICANCE EQUIPMENT EVALUATIONS CONFIRM ~ DESIGN ADEQUACY e
4 e t 8 9 O ; t $>
'g i
..t, f . . ;i i . NRR STAFF PRESENTATION TO THE O ACRS
SUBJECT:
PERRY EARTHQUAKE DATE: MARCH 12 - 13, 1986 PRESENTER: J. STEFANO, P. SOBEL, A. LEE Co. . PRESENT5R'S TITLE / BRANCH /DIV: DIVISION OF BWR LICENSING 1 PRESENTER'S NRC TEL. NO.: J. STEFANO, X29473 SUBCOMMITTEE: EXTERME EXTERNAL PHENOMENA 1 l q O ; 1
" * * * * * * * ' * * * * *e. e
1 O l J. STEFANO - SUMMARIZE EARTHOUAKE EVENT, PLANT RESPONSE, STAFF ACTION / FINDINGS, POST-LICENSE CONFIRMATORY ISSUES P, SOBEL - GE0 LOGY / SEISM 0 LOGY FINDINGS AND OBJECTIVE OF RELATED CONFIRMATORY ITEMS TO BE PURSUED POST-LICENSE A. LEE - EQUIPMENT QUALIFICATIONS AND RELATED CONFIRMATORY ITEMS 4 STRUCTURAL FINDINGS AND OBJECTIVE OF RELATED CONFIRMATORY ITEMS POST-LICENSE J. STEFANO - CONCLUDING REMARKS !O
^ c::, ; STAFF FINDS (SSER N0, 9 - MARCH 5, 1986) NO OBSERVED SIGNIFICANT DAMAGE TO PLANT FROM EARTHOUAKE 1 DESIGN OF PLANT STRUCTURES AND EQUIPMENT HAVE SUBSTANTIAL MARGINS OF SAFETY RE LOADS / STRESSES INDUCED BY EARTHQUAKE N0 BASIS AT THIS TIME TO REVISE SEISMIC DESIGN. BASES FOR PLANT SEVERAL CONFIRMATORY MATTERS NEED FURTHER ANALYSIS AND REVIEWS BY CEI/ STAFF (LISTED (]) BELOW) J i
O CONFIRMATORY ITEMS t MOST COMPLETED PRIOR TO FULL POWER LICENSE 4 (ALL EXCEPT LONG TERM GENERIC ITEMS) 4 FAULT PLANE SOLUTIONS INJECTION WELLS FAULTS AT PLANT SITE l - ENRICHED'HIGH FREQUENCY HIGH FRE0, SHORT DURATION EARTHOUAKES i ($) RELOCATION OF SEISMIC INSTRUMENTS SEISMIC QUALIFICATION OF EQUIPMENT 1
4 4-i O CONCLUSION PLANT SEISMIC DESIGN ADEQUACY HAS BEEN REAFFIRMED CONFIRMATORY WORK NOT EXPECTED TO RESULT.IN ANY I DESIGN CHANGES , 1 ! 5% LICENSE ON MARCH 14 1 l 1 l i 1 O l ' i 4 l t i h l 4 i 1 b' I . O 11 l l r
-,.-,--n--..nna,...,,,.----,-.-,-----,-.-.~_... , , _ _ - - - - -_
O JANUARY 31, 1986 EARTHQUAKE f LGCA T I ON - 41. 65
- N , 81.16
- W ABOUT 10 MILES SOUTH OF PERRY MAXIMUM INTENSITY VI (MM)
AFTERSHOCKS - ABOUT TEN
- 1 TO 6 MI DEEP
- LARGEST WAS MAGNITUDE 2,4 ;
i .) - FREE-FIELD RECORDINGS i INPLANT RECORDINGS OF MAIN SHOCK SHOW EXCEEDANCES OF i SSE AND OBE AT HIGH FREQUENCIES (AB0VE 15 Hz).
, O 1
i l 1 4 i I j l' I
i ^ i i ) STAFF REVIEW (SER-1982) MOST SEISMIC CATEGORY I STRUCTURES FOUNDED ON DEVONIAN SHALE BEDROCK. . NO CAPABLE FAULTS IN THE SITE REGION. l FAULTS IN THE INTAKE AND DISCHARGE TUNNELS AND IN THE Pl. ANT SITE EXCAVATIONS ARE NOT CAPABLE. 1 PERRY SITE IN CENTRAL STABLE REGION. SITE SPECIFIC RESPONSE SPECTRA FOR A NEARBY MAGNITUDE 5.3 EVENT COMPARED TO THE SSE (0.15G, RG 1.60). O 1 T 4
- .O
_.___,____.___U
CONFIRMATORY ISSUES LOCATION OF MAIN SH0CK AND AFTERSHOCKS. FAULT PLANE SOLUTIONS AND STRESS DIRECTION. SEARCH FOR ASSOCIATED GEOLOGICAL STRUCTURE (S). ASSESSMENT OF FAULTS WHICH WERE BELIEVED TO BE INDUCED BY PLEISTOCENE GLACIATION. POSSIBILITY THAT EARTHQUAKES ARE RELATED TO INJECTION OF CHEMICAL WASTES IN TWO WELLS 7 MILES NORTH OF 1 THE EARTHQUAKES. ASSESS FREE-FIELD GROUND MOTION RECORDINGS WITH
' () RESPECT TO WORLD-WIDE DATA BASE; ASSESS SOURCE OF HIGH FREQUENCIES AND POSSIBLE SITE EFFECT.
1 0
I t g 4 V STRUCTURAL SEISMIC DESIGN PLANT WALKDOWNS REVEALED NO SIGNIFICANT STRUCTURE DAMAGE. REGION lil, SQRT, AIT A GOOD CORRELATION OF MEASURED IN-STRUCTURE RESPONSE AND THE RESPONS$ CALCULATED USING THE RECORDED FOUNDATION MOTION AND A FIXED-BASE REACTOR BUILDING DYNAMIC MODEL WAS OBTAINED. THIS CONFIRMS A LACK OF ROCKING RESPONSE; AND HENCE, INSIGNIFICANT S0ll-STRUCTURE INTERACTION. RECORDED FOUNDATION MOTIONS ARE SIMILAR TO THE CORRESPONDING FREE-FIELD GROUND MOTIONS IN FREQUENCY CONTENT; ABOUT 20 HZ. WITH A FIXED-BASE 3-D MODEL AND THE RECORDED FOUNDATION ACCELERATION TIME HISTORY AS INPUT, THE CALCULATED RESPONSE SPECTRA AT HIGH ELEVATION OF REACTOR BUILDING INDICATE SIMILAR AMPLICATIONS AS THE MEASURED RESPONSES OVER THE MEASURED FOUNDATION RESPONSE SPECTRA, AT 20 HZ REGION. 4
-._-_.m_. __ _ _ , . _ _
- J ~ - w ,"
t ( THE ORIGINAL PERRY REACTOR BUILDING DYNAMIC MODEL IS, THEREFORE, ADEQUATE IN PREDICTING HIGH FREQUENCY RESPONSE. THE HIGH-FREQUENCY, SHORT-DURATION EARTHQUAKE HAS AN INSIGNIFICANT ENERGY CONTENT. FOR REACTOR BUILDING, THE RECORDED MOTION WOULD NEED TO BE SCALED BY A FACTOR OF HIGHER THAN TWO TO ACHIEVE DEFORMATIONS CORRESPONDING TO THE DESIGN LEVEL OF FORCES. A QUANTITATIVE ASSESSMENT INDICATES THAT THE DYNAMIC () STRESSES IN CONTAINMENT BUILDING INDUCED BY THE RECORDED FOUNDATION TIME HISTORY IS WELL BELOW THE DESIGN VALUES. i l 1 1
)
\ \m)
EQUIPMENT SEISMIC QUALIFICATION PLANT WALKDOWNS REVEALED NO DAMAGE TO EQUIPMENT ITSELF, , THE SUPPORTS, AND MOUNTING CONFIGURATIONS. THE HIGH-FREQUENCY, SHORT-DURATION EARTHQUAKE HAS AN INSIGNIFICANT ENERGY CONTENT TO CAUSE DAMAGES TO EQUIPMENT. ALL OF THE 39 SAFETY-RELATED AND 36 NON-SAFETY RELATED SYSTEMS THAT WERE ENERGIZED DURING THE EARTHQUAKE HAD FUNCTIONED AS DESIGN. THE TRIPPING OF THREE NON-SAFETY RELATED SYSTEMS WAS i CONSIDERED TO BE EITHER INSIGNIFICANT OR EXPECTED BY DESIGN. (])
- A QUANTITATIVE ASSESSMENT OF A SAMPLING OF EQUIPMENT REVIEWED SO FAR INDICATES THAT THE ORIGINAL SEISMIC QUALIFICATION IS ADEQUATE.
O CONCLUSIONS NO SIGNIFICANT SAFETY IMPACT OF THE EARTHQUAKE ON EQUIPMENT AND STRUCTURES HAS BEEN IDENTIFIED. THE DESIGN-BASIS EA'RTHQUAKES MAY HAVE BEEN EXCEEDED AT A HIGH, NARROW FREQUENCY REGION OF THE RESPONSE SPECTRA, THE ORIGINAL PLANT SEISMIC DEFIGN WAS NOT AFFECTED.
'TNE" STAFF CONCLUSIONS AS STATED IN THE PREVIOUS SER AND ITS SUPPLEMENTS REGARDING THE ADEQUACY OF THE SEISMIC QUALIFICATION PROGRAM REMAIN VALID.
1 i I 1 i I } I O l
1 () CONFIRMATORY ACTIONS BY THE APPLICANT
~
PERFORM AN ADDITIONAL QUANTITATIVE ASSESSMENT ON THE SEISMIC QUALIFICATION OF A BROADER SAMPLE OF EQUIPMENT TYPES, LOCATED IN DIFFERENT BUILDINGS ON VARIOUS ELEVATIONS. PERFORM A GENERIC EVALUATION OF A HIGH-FREQUENCY, SHORT-DURATION EARTHQUAKE WITH REGARD TO ITS ENERGY CONTENT AND POTENTIAL SAFETY SIGNIFICANCE OF EQUIPMENT AND STRUCluRES AT PERRY. USING THE RESULTS OBTAINED, ASSESS THE SEISMIC CAPABILITY OF THE PERRY PLANT, IF ANOTHER EARTHQUAKE OF SIMILAR CHARACTERISTICS, BUT WITH HIGHER MAGNITUDE AND/0R LONGER DURATION SHOULD OCCUR NEAR THE SITE. d 1 q s__ _ _--.._-- . -- r - , _ _ , , . , ,m .
,_y,_ . - - - - - - _-w-,--,-,,ym....y _ _ -,y,- ,-- -., - - -,w--. --m,j4---v
. . *-[
/' NRR STAFF PRESENTATION TO THE O
ACRS l
)
l l
SUBJECT:
USI A " SYSTEMS INTERACTIONS" PROPOSED RESOLUTION DATE: MARCH 17, 1986 , PRESENTER: DALE THATCHER Q - PRESENTER'S TITLE / BRANCH /DIV: TASK MANAGER USI A 1h ENGINEERING ISSUES BRANCH DIVISION OF SAFETY REVIEW AND OVERSIGHT PRESENTER'S NRC TEL. NO.: 492-8358 SUBCOMMITTEE: SAFETY PHILOSOPHY, TECHNOLOGY AND CRITERIA .
. AND EXTREME EXTERNAL PHENOMENA e
- O ,
M ** eM* .=,
1 r
- O A_17ea0Gi'AM ,
l OVERVIEW RE-DEFINED /SCOPED THE ISSUE (FALL '83) . PERFORMED A NUMBER OF TASKS BASED ON DEFINITIONS FORMULATED A NUMBER OF TECHNICAL FINDINGS , 4 CONSIDERED A NUMBER OF ALTERNATIVES FOR RESOLUTION
*~FOR EXISTING PLANTS
- .
- FOR NEW PLANTS
- ~
- FOR A PLANT DOING A PRA lO
- F R C NTINUED REVIEW 0F OPERATING EXPERIENCE DEVELOPED A PROPOSED RESOLUTION
- NO BACKFITS l
- GENERIC INFORMATION LETTER l
- GUIDANCE FOR SEVERE ACCIDENT POLICY IMPLEMENTATION
- REVIEW GUIDANCE FOR FUTURE PLANTS J .
e O
([) BACKGROUND 1974 - ACRS CONCERN - - STANDARDIZED PLANTS VAGUE DESCRIPTION OF THE CONCERN NRC STAFF ACTIVITIES USI IN 1978 .. BRANCH FORMED ABOUT 1980 LAB CONTRACTS TO STUDY 1980-81 CONFLICTING DEFINITIONS METHODS INVESTIGATION O NEW TASK ACTION PLAN - 1983 WORKING SET OF DEFINITIONS MORE FOCUSED PROGRAM DRAFT RESOLUTION - AUGUST 1985 CIRCULATED FOR DIVISION COMMENT SIGNIFICANT COMMENTS FROM DE & DSI
~ REVISED RESOLUTION - JANUARY 1986 .
PF,0 POSED RESOLUTION - JANUARY 1986
- s e
O i
- _ - - _ __ _.._ ___,-_.,_ _ _ _ _ . _ . . _ , , . _ - _ . , . - _--___.__m.__- -
. l. . .
DEFINITIONS /SCOPF LOOK
- SCREEN SIGNIFICANT T ~
EVENTS 1
=
CLASSIFY COMMON CAUSE ' i MULTIPLE FAILURE EVENTS INDEPENDENT FAILURE EVENTS OTHER ADVERSE COMMON
' SYSTEMS CAUSE -
INTERACT 10'N l . , ; EXTERNAL SYSTEMATIC, NONRAND0M, FAILURE.IN ONE. FAILURE IN ONE FAILURE IN ONE l PHENOMENA CONCURRENT FAILURES IN . SYSTEM WHICH SYSTEM WHICH SYSTEM WHICH i WHICH CAN ' REDUNDANT SAFETY SYSTEMS IS FUNCTIONALO IS SPATIALLY SIMULTANE00Sli . HINDERS THE COUPLED TO COUPLED TO OPERATORS-
; CHALLENGE DUE T0-SAFETY SYSTEMS SAFETY SYSTEMS RESPONSE BY I REDUNDANT -DESIGN -MAINTENANCE DUE T0: DUE 70: -ERR 0NE0US -
SAFETY SYSTEMS -MANUFACTURE -TEST- -PHYSICAL -SPATIAL. INDICATION (E.G. WIND, -INSTALLATION -0PERATION CONNECTION INTERTIES -PR0HIBITED l EARTHQUAKE) -SHARING OF -SHARING 0F ACCESS . SUPPORT LOCATIONS SYSTEMS 4 t g .
TASKS SEARCH OPERATING EXPERIENCE AND ESTABLISHED TRENDS, PATTERNS - REVIEW SYSTEM INTERACTION STUDIES DIABLO CANYON INDIAN POINT 3 MIDLAND
- REVIEW WORK ON METHODOLOGY PROBABILISTIc RISK ASSESSMENTS METHODS COMPARISON AT IP3 DIGRAPH AND FEMA / FAULT TREE OTHER CANDIDATE METHODS REVIEW PAST AND ONGOING ACTIVITIES ACRS LABORATORY STATE-OF-THE-ART REVIEWS i POST-TMI REQUIREMENTS I
' INDUSTRY INITIATIVES s
h e O - 1 i
---.+.,,,,y , ,,,-3_-- p,.. ,,w. ,,,,,ym_,,,4,,,,*,w_.wg,--.mw w-m..., -,p.# 9,y._,-wn. -- %,7-- -,m , p.9 g,-g%.,,_- ,-,
f ($) . TECHNICALFIN0INGS ~[ SENERAL [
- DIFFICULT TO ADDRESS BECAUSE IT IS A SUBJECT AREA RATHER THAN A FOCUSED ISSUE !
- NO CONCRETE, GENERIC SOLUTION TO ELIMINATE CONCERNS
\
SEARCH METHODS i l
- MANY AVAILABLE ! . 1 - SOME BETTER FOR ONE TYPE ASI (FUNCTIONAL, SPATI AL)
() - TO DO OVERALL PLANT SEARCH NEED COMBINATION
- A COMPREHENSIVE SEARCH IS COSTLY AND NO GUARANTEE ALL ASIS
. WILL BE FOUND l
- A PRA IF DONE WITH SUFFICIENT DETAIL CAN FIND ASIS
- OPERATING EXPERIENCE DOES UNCOVER THEM AND THEY D0 RECEIVE ,
ATTENTION
- MOST COST EFFECTIVE APPR6ACH APPEARS TO BE TO FOCUS A SEARCH O
e l
TECHNiCALFINDINGS(CONT.) ,
~
ASI EVENTS - , l
- PLANT SPECIFIC IN NATURE
- PLANT SPECIFIC IN RESULTS/0VERALL IMPACT
- DO NOT LEAD TO CORE MELT ALONE, TYPICALL-Y RECOVERABLE
- SOME PATTERNS; WHERE OCCUR, CAUSES
(, - LARGE NUMBER OF Asis PREVIOUSLY ADDRESSED
~
- SOME Asis INVOLVE COUPLING OF EVENT INITIATION AND MITIG OION O
- RESIDUAL CONCERNS IDENTIFIED IN THE lac POWER AREA f
W e O . w - - - ,. ,- - -- - -,_ ,_ ,.-___. __,_,, _ , _ ,
ALTERNATIVES CONSIDERED
)
EXISTING PLANTS - REQUIRE A COMPREHENSIVE SEARCH . DEVELOP A PRESCRIPTIVE SOLUTION REQUIRE AN INDEPENDENT SHUTDOWN SYSTEM REQUIRE A FOCUSED SEARCH 1 FUTURE PLANTS PROVIDE FOR A SEPARATE REVIEW PROVIDE ADDITIONAL GUIDANCE / CRITERIA PRA ([) - PROVIDE ADDITIONAL GUIDANCE ENDORSE A SPECIFIC SEARCH METHOD ONGOING OPERATING EXPERIENCE REVIEWS PROVIDE NEW RECOMMENDATIONS FOR HOW TO TREAT ASIS PROVIDE INFORMATION To ONGOING REVIEWS , l l O et 1 l 6
e . O ReSot.uriou or Division Consen1S... COMMENT RESOLUTION , DSI AND DE BELIEVE EXISTING DROPPED PROPOSED NEW SRP REQUIREMENTS / GUIDANCE ARE SECTION. ADEQUATE FOR TREATMENT OF ASIS. DSI AND DE DID NOT BELIEVE WE THE PROPOSED RESOLUTION JUSTIFIED THE REQUIREMENT FOR TREATS BOTH EQUAL NOW, A FOCUSED SEARCH FOR SPATIALLY I.E., NO NEW REQUIREMENT -- COUPLED ASIS AND THEY DID NOT FOR EITHER BUT BOTH BELIEVE WE JUSTIFIED NOT INCLUDED IN THE GENERIC O SEARCHING FOR FUNCTIONALLY INFORMATION LETTER. COUPLED ASIS. DS! AGREED WITH A NEED FOR THE RESOLUTION HAS DROPPED ACTION ON OLDER PLANTS FOR THIS REQUIREMENT BUT THE INTERNAL FLOODING. BASIS INCLUDES RELIANCE ON SPECIFIC INDUSTRY EFFORTS ! l IN THIS AREA., l DE RECOGNIZED A NEED FOR THE RESOLUTION ACKNOWLEDGES C0 ORDINATION OF USI A-46 THE USI A-46 ACTIVITIES. AND THE SEISMIC ASPECTS OF , A-17.
- . _ - - - _ . - . - - - _ _-_- - , . _ __ _ - . _ _- .- .- _ = , PROPOSED RES01.UT10N . .
Q . GENERIC LETTER FOR INFORMATION ONLY ,
~
i ACKNOWLEDGE NRC/ INDUSTRY EFFORTS IN TWO AREAS SEISMIC CONCERNS AT OLDER PLANTS FLOODING (INTERNAL) EVENTS ,i
; COORDINATE AND EXPEDITE THE RESOLUTION OF Ist POWER
~
CONCERNS /lSSUES 4- ACKNOWLEDGE THE SEVERE AcclDENT POLICY AND ITS l RELATIONSHIP ^ o
- FOR FUTURE PLANTS, SUPPLEMENT THE REVIEW CRITERIA REGARDING SINGLE FAILURE l I
; i l
i .
- r l
O. . 1,
-. .-._._._-_._.______.__~,--,.--,-.._.n.-,_ .._....,__..._..._,,,__._,.,,-.w.---____n- .s - , , ,--,.,
I
~
j GENERIC LETTER O o DESCRIBES AND DEFINES THE ISSUE - i o HIGHLIGHTS SEVEN SPECIFIC ASI AREAS f ELECTRIC POWER SYSTEMS SUPPORT SYSTEMS IN GENERAL FAIL SAFE DESIGN PRINCIPLES PREFERRED FAILURE MODES lac POWER SUPPLIES SEISMICALLY-INDUCED SPATIAL INTERACTIONS INTERNAL FLOODING AND WATER COMMUNICATION j --- 0 RECOMMENDS , Q - REVIEW THESE AS PART OF OPERATING EXPERIENCE 1 EVALUATION I IF A UTILITY DOES PLANT SAFETY REVIEW WORK (SUCH AS , PRA, WALKDOWN, ETC.) FOR ANY REASON, THEY SHOULD FACTOR TREATMENT OF ASIS INTO IT 1 I - CONSIDER THESE AREAS AND ASIS UNDER THE SEVERE j ACCIDENT POLICY i l 5 s 4
- _ - - . . - . . ~ . - _ _ - - - . . _ - . - . _ , - . _ - _ _ . _ _ - . . _
l j - l
- ADDED GUIDANCE FOR APPLICATION OF A SINGLE FAILURE l l0 WHERE A SINGLE FAILURE CAN CAUSE AN EVENT ,,
~
l kN2 - i CANALSOPREVENTPROPERACTIONOFTHEPLANTSAFETY50VIPMENT DESIGNED TO PROTECT AGAINST THAT EVENT; THEN--THE REMAINING SAFETY EQUIPMENT SHOULD BE CAPABLE OF
; MITIGATING THE EVENT (1. E. PREVENTING CORE MELT) EVEN WHEN !
DEGRADED BY AN ADDITIONAL SINGLE RAND 0M. FAILURE. . I NOTE: IF THE INITIATING FAILURE IS UNLIKELY (E.G., A LARGE l ): PIPE BREAK) THEN THE ADDITIONAL SINGLE FAILURE CONSIDERED ; l j '" MAY BE LIMITED TO MORE LIKELY FAILURES AND CONSIDERATION OF lO ADDED PIPE BREAKS, ETC., MAY NOT BE NECESSARY. h i a ! l !l
- 1 l ' l 1
.s
! O. I i l
CERTIFICATE OF OFFICIAL REPORTER (~ . C' This is to certify that the attached proceedings before the UNITED STATES NUCLEAR REGULATORY COMMISSION' in the matter of: NAME OF PROCEEDING: ADVISORY COMMITTEE ON REACTOR SAFEGARDS SUBCOMMITTEE ON SAFETY PHILOSOPHY, TECHNOLOGY, AND CRITERIA DOCKET NO.: PLACE: WASHINGTON, D. C. ( DATE: WEDNESDAY, MARCH 12, 1986 were held as herein appears, and that this is the original transcript thereof for the file of the United States Nuclear Regulatory Commission.
\s e (sigt)
/s v ,,
(TYPED) DAVID L. HOFFMAN
'i Official Reporter ACE-FEDERAL REPORTERS, INC.
Reporter's Affiliation i}}