ML20054J730
| ML20054J730 | |
| Person / Time | |
|---|---|
| Site: | Shoreham File:Long Island Lighting Company icon.png |
| Issue date: | 06/25/1982 |
| From: | Atomic Safety and Licensing Board Panel |
| To: | |
| References | |
| ISSUANCES-OL, NUDOCS 8206290499 | |
| Download: ML20054J730 (168) | |
Text
._ _
NCCI.IAR REGULATORY COM!i'SSICN q ,g- - -
w _uuJ . 3 BEFORE THE ATOMIC SAFETY AND LICENSING BOARD Is 16e Mat:::ar cf: :
LONG ISLAND LIGHTING COMPANY :
- DOCKET NO. 50-322-OL (Shoreham Nuclear Power Station) :
CA".T: June 25, 1982 PAGES: 5534 - 5700 AT: Riverhead, New York
.(,,
a
' 7'
,, 7, /i i l n , r. '
t'-
) /., 2 . , - c': ,
f .
- <7
=
[j , , , ' j g z,/
ALDER $0X
_r (REPORTLTG 400 Vi_T sia Ave., S.W. W==hd"g ==, D. C. 20024 Talaphene: (202) 554-2345 8206290499 820625 PDR ADOCK 05000322 T PDR
5534 O i UNITED STATES Or AnERICA 2 NUCLEAR REGULATORY COMMISSION 3 BEFORE THE ATOMIC SAFETY AND LICENSING BOARD 4 - - - - - - - - - - - - - - - - -x 5 In the Matter of 1 6 LONG ISLAND LIGHTING COMPANY a Docket No. 50-322-OL 7 (Shoreham Nuclear Power Station) 8 - - - - - - - - - - - - - - - - -x 9
10 Riverhead County Complex 11 Center Drive 12 Riverhead, New York 11901 13 Friday, June 25, 1982 0 14 The hearing in the above-entitled matter 15 reconvened, pursuant to recess, at 9:05 a.m.
16 BEFOREa 17 LANRENCE BRENNER, Chairman 18 Administrative Judge 19 JANES CARPENTER, Member 20 Administrative Judge 21 PETER A. MORRIS, Member 22 Administrative Judge 23 WALTER H. JORDAN, Assistant to the Board 24 Administrative Judge 25 O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE, S.W, WASHINGTON 0.C. 20024 (202) 554 2345
l 2
5535 l l
O i m EARxNCts.
2 On behalf of Applicants 3 ANTHONY F. EARLEY, Esq.
4 T.S. ELLIS III, Esq.
5 H. TAYLOR REVELEY III, Esq.
6 Hunton C Williams 7 707 East Main Street 8 Richaond, Va. 23212 9
10 On behalf of the Regulatory Staffs 11 RICHARD RAUSON, Esq.
12 Washington, D.C.
13 O 14 On behalf of Suffolk Countys 15 LAWRENCE COE LANPHER, Esq.
16 KARLA J. LETSCHE, Esq.
17 Kirkpatrick, Lockhart, Hill, 18 Christopher C Phillips 19 1900 M Street, N .W .
20 Washington, D.C. 20036 21 - --
22 23 24 .
25 O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5536
() I C9IIEEI2 l 2 WITNESSES: DIRECT CROSS REDIRECT RECROSS BOARD 3 George F. Dawe, l O- George Garabedian, 4
Pio W. Ianni, 1 Robert M. Kascsak, 5
Paul J. McGuire, 6 Paul W. Riegelhaupt and David J. Robare (Resumed) 7 By Mr. Ellis 5557 By Judge Brenner 5560 8 By Judge Carpenter 5564 g
By Judge Morris 5565 By Judge Brenner 5567 10 By Judge Carpenter 5576 By Mr. Lanpher 5579 11 By Ms. Letsche 5624 By Mr. Ellis 5627 12 (Afternoon Session..page 5632)
() 34 Robert M. Kascsak (Resumed)
Edward T. Burns and 15 vojin Joksimovich (Recalled) .
By Ms. Letsche 5633 16 EXHIBITS
,7 NUMBER IDENTIFIED RECEIVED 18 19 Suffolk County No. 21 5605 20 Suffolk County No. 22 5666 21 Suffolk County No. 23 5688 22 ES:
23
() 24 Morning - 5590 Noon - 5631 25 Afternoon - 5677 O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
l 5537 1
EEQEEEEIEEE 2 JUDGE BRENNER: Good morning. We have some 3 preliminary matters. At the end of the day yesterday we 4 were handed LILCO's Motions to Supplement Witness Panels 5 Plus Several Related Matters, a catchy title, despite 6 Mr. Reveley's belief when he handed it to us that we 7 would not read it last night, we did, or some of us 8 did. Some of us read some of it. Some of us read all 9 of it. We believe we should discuss scheduling of to responses, if any.
11 One of the motions is in response to the 12 Board's request that we hearck additional testimony on 13 procedures related to the water hammer contention, which O 14 is Suffolk County 4. LILCO proposes to file additional 15 testimony by July 9th, and also proposes that we hear 16 that as part of the litigation on human f actors 17 procedures, Suffolk County 19. That is acceptable to 18 the Board.
19 We would have to make some adjustments in -- I 20 think we should make some adjustments in the schedule 21 for cross examination plans. However, if we do that, 22 the other day, we scheduled the receipt of cross 23 examina tion plans on Suffolk County 19 for July 6th.
24 While we could keep that and require additional plans on 25 the additional testimony, we think it would be less work O
ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5538
() I for the other parties if we allow them to hold off and 2 file their plan jointly if they agree.
l 3 I do not know what the county's preference 4 is. This is my assumption, that the county would prefer 5 to be able to look at it all together.
6 MR. LANPHER: That is correct, Judge Brenner.
7 I did not have s.. opportunity to review all of the 8 filing yesterday, sa without prejudice, to take a 9 position on the various filings, yes.
10 JUDGE BRENNER: All right. Well, the 11 testimony was not filed yesterday, just the indication 12 that the testimony would be filed on July 9th.,
All 13 right, we will unschedule the filing of cross 14 examination plans on Suffolk County 19 from.the July 6th 15 due date. If we are going to try the issue in the same 16 sequence, I don 't know when . we would need the cross 17 examination plan by. That depends upo'n how things 18 develop between now and then, but it may be that we 19 would need the plan by July 13th , and tha t is very soon 20 after the receipt of the testimony, too soon, I would 21 submit.
22 Let's tentatively conside the filing of the 23 cross examination plan on Suffolk County 19 for July
() 24 20th, recognizing that that date is out of sequence with 25 the contentions scheduled before and after it. Now, it O
ALDERSoN REPORTING CoMPAN1f,INC, 400 VIRGINIA AVE., S.W WASHINGTON, D.C. 20024 (202) 554 2345
5539 lll 1 may turn out that we can keep that sequence anyway, but 2 as we get closer, I would like the parties to remind the 3 Board if we might have to alter the sequence because of 4 the adjustment we have just made.
5 Having made that adjustment, we would like to 6 add another contention to receive a cross examination 7 plan on July 6th by, and in looking ahead, I realized 8 that the next contention after human factor simulator 9 training, for which the cross plan will still be 10 scheduled for July 6th, is the SRV test program, Suffolk 11 County 22. I was struck by the oddity that instead of 12 immediately following that contention, the arguably 13 rela ted Suff olk County 28(a)(vi), reduction of SRV G 14 challenge.s, does not immediately follow. There is a 15 contention in between. I don't know why that is. I 18 thought it might make sense to require cross plans, not 17 necessarily combined cross plans, but at least for the 18 parties to prepare the two cross plans at the same time 19 for the SRV test program contention as well as the 20 reduction of SRV challenges.
21 MR. LANPHERa Judge Brenner, I think that was 22 the intention. If you will recall, I guess it was the 23 28(a)(vi) testimony was delayed until, I guess it was 24 the 8th of June, and we have put the SC 22 testimony at 25 the end of the previous filing with the intention of O
ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554-2345
5540 1 1
1
() 1 delaying hearing that, so that it can be heard at or 2 a bo ut the same time with 28(a)(vi). i 3 JUDGE BRENNER: All right. Well, consider my 4 comments, because I believe, unless I am mistaken, that 5 when we got the sequence of litigation, that they were 6 not sequenced together. Maybe what happened is, my list 7 ran beyond the sequence previously given to us, but the 8 parties should take a look at that.
9 MR. REVELEYa Judge, my understanding, 10 although I don't have my list in f ront of me, is that we 11 would put those two SRV contentions together, and I 12 didn't think we had sequenced as yet the two pieces of 13 testimony we filed last week, but that we do sequence O 14 them -- we will put the two SRV items together, and 15 perhaps even try them absolutely together, consolidate 18 the panels.
17 JUDGE BRENNERs All right. I think you have 18 just identified the problem, and Er. Lanpher has also, 19 not a problem, the reason for my confusion. That is, 20 Suf folk Coun ty 22 was. the last of the May 25th filings, 21 and Suffolk County 25 and Suffolk County 28 vere the 22 June 14th filings, and we had not discussed the sequence.
23 MR. REVELEYa Tha t is righ t. They have not
!( 24 been sequenced.
25 JUDGE BRENNERa All right. The only question l
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WAS'dNGTON, D.C. 20024 (202) 554-2345
5541
() 1 is, then, everybody agrees that those cross plans should 2 be filed together, even if they are not absolutely 3 consolidated. Can we wait until July 13th for the 4 filing of those cross plans? I think we probably can.
5 MR. LANPHER I think so.
6 MB. REVELEY Yes.
7 JUDGE BRENNER: In sum, there is now only one 8 cross examination plan due on July 6th, and that is 9 Suffolk County 20. As we get closer, we vill precisely 10 schedule cross examination plans for July 13th. You may 11 anticipate that it will be the SRV cross plans as a 12 minimum, and a t the time the parties prepare to discuss 13 that sequence, you can respecify the sequence of O 14, contentions for us beyond that. For example, you may 15 change the sequence between the ATWS contention and the 16 RPV integrity and testing contention, even though ther 17 came in diff erent batches, because they all vill have 18 been filed by then, so we vill valt to hear from the 19 parties as to the further sequence beyond the SRY 20 sequence.
21 Looking ahead, and this is just a comment, I
(
22 think what the Board might prefer is, put the SRV 23 contentions together, as everybody agrees, either 24 sequentially or consolidated, depending on what the 25 parties agree on, and have the RPV integrity and testing A
%)
l ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5542
() 1 contention, Suffolk County 25, either immediately before 2 those two or immediately after, and the reason I mention
{) 3 that is because of the assistance that the Board expects 4 to have on those three contentions.
5 I am not precluding the fact that you may want 6 to move the AT:!S contention up or something like that.
7 I just want to keep those three contentions together, 8 that is, Suffolk County 22, Suffolk County 25, and 9 Suffolk County 28(a)(vi) together in sequence.
10 In terms of the other aspect of LILCO's 11 motions to supplement which'were received yesterday, 12 that is, the addition of witnesses, that has been the 13 combined filing that we have been anticipating, and it O 14 is responsive to our concern and consistent with our 15 previous rulings. However, we want to give the county 16 or anyone else the ohportunity to object to particular 17 a'dditional witnesses as to particular additional 18 contentions, and an orderly way to do that would be as 19 to any cross examination plans which will have been 20 filed by July 6th, we would want any objections to be 21 received by July 6th.
22 A s to the proposals to add additional 23 witnesses, with respect to additional witnesses for
( 24 contentions on which we will receive cross examination ,
25 plans beyond July 6th, the objections would have to be O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W WASHINGTON, D.C. 20024 (202) 554 2345
5543
() 1 filed with the cross examination plans. Now, of course, 2 the objections should also be filed on the other 3 parties. If you have objections and can raise them 4 earlier, I think that would be appreciated, but we won't 5 require them earlier than the sequence we just 6 indicated, so as not to require the county to look ahead 7 too many weeks by July 6th.
8 Is this a good time to get the parties' joint 9 proposed schedule on discovery with respect to LILCO's 10 emergency planning ?
11 ER. EARLEY: Judge, the lawyers who are 12 involved in emergency planning have been in New York 13 this week. Both Mr. Lanpher and me have been in touch O 14 with those lawyers. A meeting is scheduled for next 15 Wednesday that will include all of the parties involved 16 in emergency planning. The purpose of the meeting is to 17 discuss the contentions, hopefully narrow some of the 18 disagreements about the contentions, and also to 19 finalize discussions about scheduling discovery.
20 My information is that there has been 21 tentative agreement that depositions will be allowed up 22 to the end of discovery, that the end of discovery will 23 be the final date for filing answers to interrogatories,
) 24 and on Wednesday they will discuss how far away from, 25 that final date they will back off the required time for O
ALDERSON REPORTING COMPANY,INC.
400 VIRGINIA AVE, S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5544
() 1 filing those interrogatories, and that discussion will 2 take place next Wednesday.
3 So, progress is being made. We don't have a 4 final schedule to give to the Board right now.
5 MR. LANPHER: If I could add to that, my 6 understanding is, out of this meeting they are intending 7 to prepare a status report to present to the Board when 8 we reconvene af ter the July 4 recess, or the week 9 recess. I guess that would be July 6 or 7, depending on 10 when we convene.
11 JUDGE BRENNERs It will be July 6th, at 10:30.
12 HR. LANPHER4 All righ t, July 6th , with the 13 intention of reporting to the Board on all of the 14 progress or lack of progress or whatever on narrowing 15 and stipulating to the wording of contentions, or 16 stipulating to objections, just what is going to be 17 objected to, and include in that a final discovery 18 schedule.
19 As Mr. Earley says, I have not been privy to 20 those discussions which have been going on, I think, 21 fairly intensively this week between people in Richmond 22 and Washington, D.C., so we are at a little bit of a 23 disadvantage here.
() 24 JUDGE BRENNER: Or maybe you are pt an 25 advantage being here. All right, that is acceptable.
O I
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554-2345
5545
() 1 The reason we had a ssumed that we might get the schedule 2 earlier than the status report was that the discovery
{} 3 time will end pretty quickly, close to the end of July.
4 I am not saying it has to be then, but it would be close 5 to that point. I would appreciate it if you would 6 convey the message, which I will reiterate right now, 7 that we expect the bulk of discovery to be in the most 8 ef ficien t means, which is often meetings of the experts 9 exchanging information, direct, face-to-face meetings, 10 and it may take several meetings, rather than many, many 11 interrogatories.
12 Now, I can understand why some information 13 would have to be documented, or might have to be
( 14 documented, but that is easy to do after you know what 15 the informa tion is. You can then just confirm it very 16 simply instead of having to have the process of 17 interrogatories and responses and objections and so on.
18 We are going to be busy with the hearing, and we are 19 just not going to be able to rule on extensive 20 objections. We think the parties have been quite good 21 in the past on discovery, and so this is in the vein of 22 asking them to work extra hard with respect to the LILCO 23 emergency planning discovery.
24 In addition, and this may be somewhat 25 premature -- you are likely to hear about it again after O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE, S.W., WASHINGTON D.C. 20024 (202) 554-2345 ;
1
55f4 6 O 1 ve hear where the dispute has been narrowed and what the 2 contentions are -- some of us have sat through an ,
)
O ero acr at aasaa 11tia tioa before ia oa c a citr or 4 another. I do not expect to sit here week fter week 5 and hea r arguments as to how many school buses there are 6 going to be, and matters of that nature. I am not 7 saying they are not relevant to concerns, but they are 8 not of the type of issue where we need to have expert 9 testimony and cross examination. I expect the parties 10 to be able to work out by admissions or more informal 11 means just wha t those underlying subsidiary factual 12 matters are, how many buses there would be and the like, 13 how many telephones there will be, what radio stations 14 will be involved. We don't have to hear that here, and 15 I am not going to be very patient about hearing it here. -
l 16 To the extent those f acts are important to 17 your other points or to your findings, we can get that 18 in by the way of admissions and then hear the experts' 19 testimony as to the significance of those f acts and what 20 they mean for the emergency planning process, which 21 could be very important. We are certainly not 22 denegrating that, but we don't have to sit here and 23 establish the underlying facts before us. 'The parties 24 can establish that among,themselves.
25 That is, of course, true about any litigation, O
ALDERSON REPORTING COMPANY. INC.
400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554 2345
5547 llh 1 but having been through emergency planning, I believe it 2 is particularly true in that area.
3 The other matters we have, before going into 4 the testimony, relate to Contention 7(b). Before I get 1 5 in to them , I will ask if there are any other 6 miscellaneous matters unrelated to Contention 7(b).
7 (N o response. )
8 JUDGE BRENNERa Yesterday we received two 9 documents for the first time, one of which is in 10 evidence and one of which is not. One is the LILCO 11 Exhibit 20, which is the memorandum for Mr. Michelson 12 from Mr. Denton. The other is the summary by GE of the 13 report with respect to the concern about failure in 9 14 vessel level sensing lines common to control and 15 protective systems, the Michelson concern, if you will.
16 It is in the form of a question and response.
17 We a re assuming, but do not know, that the 18 question was posed by the staff, and this is in a format 19 where FSAR's can be updated. I don't know if that is 20 correct, and I don't have to know that. Well, if you 21 know.
22 MR. EARLEY: I think Mr. Robare probably knows 23 the answer to that.
24 WITNESS ROBARE4 That was not a formal 25 question from the NRC. We posed the question to O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5548
() 1 ourselves as we anticipated they might.
2 JUDGE BRENNERa You sure copied their style ,
i 3 very well.
4 (General laughter.)
5 JUDGE BRENNERs It is our position that we 6 have seen those words before. All right. That explains 7 it. Well, there is, as Mr. Ellis correctly pointed out, 8 another contention involving that substantive issce, 9 detection of inadequate core cooling. These memos can 10 be useful, but they are not going to have much weight 11 the way they are now, that is, even as to the one in 12 evidence, LILCO Exhibit 20. There are comments on the 13 Michelson meno very informally stated by Mr. Denton's O 14 division, that is, NRR. It is very difficult to 15 ascertain exactly what they have in mind, and there is 16 no substantive support on the record for it here, so if 17 these documents are going to be important to anyone's 18 case, we expect them to be integrated in some fashion 19 into the litigation of the upcoming inadequate core 20 cooling contention.
21 We are ce rtainly not going to specify the 22 fashion. That would include a live body who knows 23 something about them.
) 24 ,
MR. ELLISs I think there has been a 25 substantial amount of testimony about the GE report, and O
ALDERSoN REPORTING COMPANY,INC.
400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5549
() 1 we supplied that as a result of the request.
2 JUDGE BRENNER: Yes, we appreciate your having
(} 3 supplied it, and I hope I didn't imply to the contrary, 4 but we can't do anything with it as is, and we do not 5 choose to take the option of putting it in evidence now 6 for a number of reasons, including the fact that 7 partially in recognition of the fact that another 8 contention is coming up, we are going to hold off a lot 9 of our questions about water level, this particular to problem, vis-a-vis the substantive problem as 11 distinguished from the methodology of classification of 12 structures and identification of systems interactions 13 for that other contention.
O 14 The Board has some further questions of this panel 15 as a followup to some of the answers the members of the 16 panel have given, but as you see, we are only going to 17 pursue it to a limited extent, and then we are going to i
18 await the other contention, and we thought we might as 19 well avait getting into these documents in any detail 20 until that point also.
21 (Whereupon, the Board conferred.)
22 JUDGE CARPENTER 4 I would like to ask a 23 question of staff. With respect to the water hammer 2,4 testimony that LILCO proposes to supplement, I would 25 like to ask whether staff expects to supplement their O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (2G2) 554 2345
5550
() 1 testimony.
2 MR. RAWSONs Judge Ca rpenter, may I have just )
1 3 one minute to consult on that? !
(]}
4 JUDGE CARPENTER: To be specific, I don't 5 recall the transcript verbatim. It relates to the area l
)
6 of staff witness testifying that he was not responsible 7 for reviewing procedures, operating procedures and so 8 on, whether he felt it should be done, but he wasn't the 9 proper witness to speak to it, and the Board -- I think 10 it would be very helpful to the Board to get some 11 additional informa tion in that direction, if you agree.
12 HR. RAWSON: Yes, sir. Let me check very 13 briefly right now. I might be able to give you a quick O 14 answer. If not, I will check into it.
15 JUDGE BRENNERa Wait a minute, Mr. Rawson.
16 Let me tell.you something you may not know. What Judge 17 Carpenter has in mind is.the excerpt from the transcript 18 cited on Page 3 of LILCO's motions, where we requested 19 the applicant and staf f consider, I guess it is f air to 20 say, even though that word isn't in there -- it was not 21 an order -- but consider providing additional testimony 22 on the matters of operating procedures and training and 23 the continuing things during the lifetime of the plant, O- ,
24 in service inspections and so on, with respect to water 25 hammer.
O ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5551
( 1 Mr. Bordenick some weeks ago told us the staff 2 was considering that. The time is drawing very late for 3 us to hear back, and I think that is the point of Judge
(])
4 Carpenter's asking.
5 MR . R AWSON : I may be able to tell you 6 something right now, Judge. If not, we will check on it 7 very promptly.
8 (Whereupon, the staff for NRC conferred.)
9 JUDGE BRENNEB. Yes. I note that Mr. Hodges 10 is here, and you are cc aferring with him.
11 MR. RAWSON: Judge Brenner and Judge 12 Carpenter, I an afraid we can't.tell you right now 13 whether a decision in fact has been made in that O 14 regard. We will let the Board know as early next week 15 as we possibly can, through a letter, if that is 18 acceptable, what the staff's intentions are in that 17 regard.
18 JUDGE BRENNER: That is acceptable. You can 19 see what the situation is. LILCO is going to file it as 20 soon as July 9th, and even as to that day we had to 21 adjust the schedule.
22 MR. RAWSON I will bring that to the 23 a ttention of everyone who needs to know about it, and we 24 will see wha t we will do next week.
25 MR. LANPHER: Judge Brenner, the excerpt from O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
l l
5552 i
l l
() 1 the transcript I know doesn't address Suffolk County 2 specifically in terms of supplemental testimony. We
, 3 just want people to know that we are going to, with the 4 Boa rd 's leave, reserve the right to file a motion at a 5 later time, whether the need to file rebuttal testimony, 6 since this testimony which will be coming in will be, I 7 assume, in the nature of rebuttal in itself, and we are 8 not intending to file additional testimony on the 9th, 9 but depending upon what comes in if new issues are 10 raised, we just want it clear that we will have to 11 evaluate that.
12 JUDGE BRENNER: Yes. And I don't remember if 13 we made the comment in the context of that particular 14 request as to water hammer, but there has been at least 15 one occasion when we made the general comment that when 16 matters are being supplemented in these particular 17 instances you can make the motion, giving the reasons 18 why you want to take additional action.
19 In that vein, although a somewhat different 20 situation, we expect to hear from you --
21 MR. LANPHER: On the seismic matter next week 7 22 JUDGE BRENNER: Yes, by July 1st, as I recall.
23 ER. LANPHER: Yes.
() 24 JUDGE BRENNER: And consistent with the spirit 25 of that type of thing, we would cartainly expect to see l
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE, S.W., WASHINGTON D.C. 20024 (202) 554-2345
5553 1 a request from you as timely as possible after reviewing 2 the testimony as to what additional action the county lll 3 wishes to ta ke.
4 MR. LANPHER: You are talking about on the 5 water hammer? I would expect t'o be able to at least 6 give you guidance at the time we would put in cross 7 plans. We would have had to focus on it by then.
8 JUDGE BRENNERs That is acceptable.
9 MR. EARLEY: Judge Brenner, it strikes me that 10 it might be more efficient that the county now knows 11 that the testimony that LILCO is going to submit 12 involves operating procedures on the issue of water 13 hammer, and that it would be more efficient if the 14 county feels it is necessary to supplement their 15 existing testimony, that they also file it on the 9th.
16 We have been having filings on all of the testimony at 17 the same time based upon the parties' perception of the 18 issues. I think the issue is clearly defined. If the 19 county has further testimony to submit, it might be much 20 more efficient to file it on the 9th rather than have 21 the county file something later on in the proceeding.
22 JUDGE BRENNER: I am not going to require 23 that. If the county feels they can, that is obviously 24 very acceptable, but this is in the nature of responding 25 to a particular focus. It is not strictly rebuttal, but O <
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5554
() 1 it is in that nature, and I think they are entitled to 2 see what you are going to say about the procedures.
3 I also have in mind the fact that as a
)
4 practical matter, some of the testimony has been that 5 these procedures are recently evolving. I am not saying 6 those procedures per se. I don't know. But the general 7 area of procedures. And I think it might arguably be 8 more efficient for the county to get a look at the 9 current state of aff airs as LILCO sees it and then 10 respond, but we will talk about the scheduling of a 11 response consistent with the extent of the additional 12 testimony and the schedule of the proceeding, so we are 13 sensitive to the schedule, but we are not going to O 14 require that they file something in advance of seeing 15 what you are going to file, partially because of the 16 subject area, as I have just indicated. -
17 Judge Morris has something to say with respect 18 to the ISEG.
19 (Whereupon, the Board conferred.)
s 20 JUDGE MORRISs As the parties know, the Board 21 received from LILCO yesterday some material relating to 22 the ISEG, including an organization chart a nd 23 qualifications of personnel. Obviously, the Board has
( 24 had very little time to really examine that material.
25 He also note that it is a draft in many regards, and O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE, S.W., WASHINGTON. D.C. 20024 '202) 554-2345
5555 h 1 that we wouldn't want to travel too far at this stage 2 knowing that there might be significant revisions at a 3 later time.
(lg 4 We also note that it is related to perhaps 5 several contentions, but certainly the operations OA 6 contention. I believe that is Suffolk County 13. So 7 that we don't plan to ask questions about it today, but 8 I would like to point out that the Board considers this 9 function of this group to be extremely important, and to that we vill have considerable interest in several 11 aspects.
12 For example, we vill want to f ully understand 13 the organization and chain of command, particularly 9 14 interested in the independence of this organization or 15 an aligned organization that operates the plant. We 16 vill, of course, examine the qualifications and 17 experience of the members of the group in some detail, 18 and finally, we vill vant to understand clearly the mode 19 of operation of this group. To perhaps oversimplify, we 20 vill vant to be able to dictinguish between those things 21 the group does which are reactive to situations that 22 develop or pieces of paper that flow into the office, as 23 opposed to being active. i 9 24 I use that shorthand of monitoring on a l
25 continuous basis the performance of the plant, the idea ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554-2345
5556 h 1 being that problems should be prevented , not mitiga ted.
2 So that is a brief description of the areas in g 3 which we will have considerable interest, and we will 4 get to that in the context most likely of the 5 Operational Contention Suffolk 13, but there may be 6 others as well.
7 MR. ELLISa Thank you, Judge Morris. LILCO 8 appreciates that. As you noted, it is in draft form, 9 and we will take all of these comments very much to 10 heart in our continuing review and revision of the 11 procedures.
12 JUDGE MORRIS: I don't know to what extent the 13 staff has had conversations or discussions with the 14 applicant on this matter, but we will be most interested 15 in the staf f 's discussions and comments in this area.
16 MR. RAWSONs We appreciate that as well, Judge 17 Morris.
18 JUDGE BRENNER: Okay. I think we are ready to 19 continue with redirect. If I had known it would have 20 taken this long, I would have suggested tha t the 21 witnesses didn't need to sit there that long, and I 22 spologize, although maybe some of the interchange was 23 good for you to hear also. I am not sure.
24 Mr. Ellis.
25 MR. ELLISs Thank you, Judge Brenner. Given O
ALDERSON REPORTING COMPANY. INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5557 h 1 your guidance r.his morning, I will try to omit a good 2 bit and focus more on the methodology. I will just have g 3 a few things. I will try to move as quickly as I can.
4 JUDGE BRENNER: What are you omitting in 1
5 response to what guidance? I 6 MR. ELLISt Well, I think the remarks that you 7 made concerning water level, I think I just have a few 8 matters that I want to clarify in that connection that 9 go more toward methodology than.the other.
10 JUDGE BRENNERa All right. I understand. We 11 are going to ask some questions about that.
12 Whereupon, 13 GEORGE F. DAWE, O 14 GEORGE GARABEDIAN 15 PIO W. IANNI,
- 16 ROBERT M. KASCSAK, 17 PAUL J. MC GUIRE, 18 P AUL W. RIEGELH AUPT ,
19 and DAVID J. ROBARE, 20 the witnesses on the stand at the time of recess, having 21 been previously duly sworn, were examined and testified 22 further as follows.
23 CONTINUED REDIRECT EXAMINATION 24 BY MR. ELLIS4 25 0 So that the matter is clear, Mr. Robare, and O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE, S.W, WASHINGTON, D.C. 20024 (202) 554 2345
5558 O 1 nr. ranni, and nr. ncGuire, we a ve heen ta1xine, 1 2 think, about two auto leg issues, boiloff and the break 3 or leak in the water leg reference. Those are two 4 different issues, are they not?
5 A (WITNESS ROBARE) That is correct.
6 0 With respect to the boiloff issue, was that 7 known to GE or others prior to the incident at Pilgrim 8 that has been referred to in the testimony?
9 A (WITNESS R0 BARE) Yes, it was, Mr. Ellis.
10 That was part of -- that consideration was evaluated in 11 the original design of the water level instrumentation 12 system at GE.
13 0 And was that also in the Shoreham design O 14 process?
15 A (WITNESS ROBARE) Yes, it was. It was -
16 analyzed in detail on the Shoreham as part of the 17 Shoreham process.
18 Q Mr. McGuire, did you Vant to add something?
19 A (WITNESS MC GUIRE) Yes, I did. I just wanted 20 to make the statement that it was known at Pilgrim -- I 21 don't know who else knew about it, but we knew the 22 possibility existed. That is how we identified, or how 23 it was identified, and we reported it.
24 0 You were not manager of Pilgrim, were you, at 25 the time it happened?
lO ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE, S.W., WASHINGTON, D.C. 20024 (202) 554-2345
ft"'
i 5559 .
() 1 A (WITNESS MC'GUIRE)-' No, sir. But we knew 2 about it long before it~ eve'r happened. In fact, steps 3 were being taken to improv'e the dryg udlcooling system 4 prior to the event. In fact, the event occurred on a 5 shutdown that started their refueling outage when those 6 modifica tions were made. c' 7 0 So that we are cicar on the sequence, were you 8 plant manager at Pilgrim before or after the incident?
9 A (WITNESS MC GUIRE) Before.
10 0 All right. Now, I guess, Mr. Ianni, with 11 respect to the water leg reference break or leak, was it i
12 known to GE that there could be an interaction between 13 the systems, protective and,the' control systems, prior
() -
14 to the Michelson memorandur,7 15 A (WITNESS IANNI) Er. Ellis, I checked with the 16 designers. They assure aes that.that point of having th e 17 static column and the only common' point between the 18 control and -- or rather the safety and the non-safety
! 19 systems, they consciously made that decision back in the I
i 20 mid-sixties, and it was a conscious decision made a t the 21 time the y we rs implementing IEEE 279, and as a result of 22 that, the? 3,r ' to do it that way, and it has been done 23 in that #4nne: for all subsequent BWR's which have been
) 24 licensed, and the key point here is that they are in 25 compliance with GDC 24, because of the last statement l
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5560 1 that we discussed yesterday in GDC 24.
2 Fu nd a men tally , the static column which is I
(} 3 common is considered a weak systems interaction, and 4 therefore in compliance. That safety is not S significantly impaired.
6 0 Now, when you say you checked with the 7 designers, I take it you don't know the f acts about 8 their knowledge from your own firsthand knowledge.
9 A (WITNESS IANNI) Yes, Mr. Ellis. As I 10 indica ted ye sterda y, I do not personally recall my being 11 involved, although I may have. It was a long time ago.
12 And so I checked with the people who were closer to IEEE 13 279 and they hava assured me of that f act.
14 0 I take it you did it since the testimony ended l
l 15 yesterday? -
16 A (WITNESS IANNI) Yes, sir.
17 MR. ELLISs We have no further redirect.
18 JUDGE BRENNERa I would like to follow up on 19 the last matters regarding GDC 24.
20 BOARD EXAMINATION 21 BY JUDGE BRENNER.
I 22 0 Mr. Ianni, yesterday, in response to a i 23 question, in talking about the analogy which, in 24 fairness to you, you were asked to assume about the 25 APRM's, and I believe you stated tha t if you only had O
ALDERSON REPORTING COMPANY,INC, I 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554-2345 l _ - _ _ _ _ _ _ _ _ _ . _ _ ,
5561 1 two APEM's, that is, channels, you were forced to use 2 only two APRM channels, tha t your approach would be 3 siallar to what you believe GE's approach has been with 4 respect to the water level or vessel level reference leg 5 design. Is that correct?
6 A (WITNESS IANNI) Yes, sir.
7 0 From your answer, I as a layman inferred that 8 you felt that it was not feasible to have more than two 9 reference legs. Is that a wrong inference?
10 A (WITNESS IANNI) The vessel was built, and you 11 simply can't put additional holds in a vessel which is 12 built.
13 0 Would it have been feasible originally to O 14 build it with a design that had an additional reference 15 leg? -
l 16 A (WITNESS IANNI) If one starts from scratch, 1
l 17 it would be feasible.
18 0 I am trying to get a better feel for the 19 design of these reference legs vis-a-vis penetrations to 20 the essel. Is there a separate penetration for each 21 reference leg, or if I trace it back, is there a single 22 penetration somewhere along the way for both reference 23 legs?
24 (Whereupon, the witnesses conferred.)
25 A (WITNESS IANNI) Yes. The water legs are O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554-2345
5562
() 1 simply teed into the vessel on opposite sides of the 2 vessel. You understand, Judge Brenner, tha t the ;
3 penetrations then go through the containment, from the 4 vessel through the containment on opposite sides to the 5 sensors, 6 0 Yes, but there are separate penetrations for 7 each one? What is confusing me is Figure 1, and maybe 8 of Attachment 9.
9 A (WITNESS DAWE) Judge Brenner, Figure 1 on to Attachment 9 is a schematic representing one reference 11 leg which can go to more than one instrument. However, 12 there a re two reference legs on the vessel. Each 13 reference leg has its own nozzle penetration to the O 14 vessel on opposite sides of the vessel, with the one 15 reference leg on each side then exiting the containment 16 with the instrumentation located outside the containment 17 in the secondary containment.
18 0 Yes. Thank you. And as I look closer, I see 19 that what the schematic shows as to the one reference s
20 leg is the wide range delta P meter and then the narrow 21 range delta P meter, and I think that is what confused i
22 me.
23 A (WITNESS IANNI) Also, Judge Brenner, you l
( 24 understand that from the sensor point of view, we are in 25 total compliance with IEEE 279. All of the entire O
ALDERSON REPORTING COMPANY. INC, 400 VIRGINIA AVE., S.W, WASHINGTON. D.C. 20024 (202) 554-2345
5563
() 1 electronic chain is. j l
2 0 I don't know how far I will pursue this at i 3 this point. In fact, not very far. But I understand
(])
4 your answer that you believe the design of the vessel 5 level system through the two reference legs complies 6 with GDC 24, because of the last sentence, which is, 7 " Interconnection of the protection and control systems 8 shall be limited to as to assure that safety is not 9 significantly impaired.
10 A (WITNESS IANNI) Only with respec t to the 11 reference leg.
12 0 Yes. That's right. Thank you. I take it 13 that you don't believe that last sentence is limited or 14 construed by the previous sentences in GDC 24? That is, 15 in order to understand wha t is meant by "shall be 16 limited so as to assure that safety is not significantly 17 impaired", do you have to meet the more specific 18 portions of GDC 24?
19 (Whereupon, the witnesses conferred.)
20 0 Let me indicate that I recognize that that is 21 a difficult question, or at least it is not susceptible 22 of an easy answer. It is easy to ask and difficult to 23 answer. It is something we would be interested in 24 getting into in greater depth in connection with the 25 contention on inadequa te core cooling, and I would be O
ALDERSON REPORTING COMPANY. INC, 400 VIRGINIA AVE., S.W., WASH!NGTON, D.C. 20024 (202) 554 2345
-=- _
5564
} 1 happy to leave it with the question outstanding at this 2 point, and get back into it later, recognizing that part 3 of the art, if you will, of construing statutes,
(])
4 regulations, and general design criteria is not just 5 looking at the words, but looking at the applications 6 over the years and the expert testimony as to why that 7 application makes sense or does not make sense.
8 BY JUDGE CARPENTERS 9 0 Mr. Ianni, in talking to the desginers, and I 10 get the sense that they recognize the things that are 11 being alluded to here very early, do they make any 12 special ef f orts to increase the reliability of this 13 particular item as a compensatory strategy to meet the 14 GDC 24 thrust that it should be limited, but I would 15 think also if you have for some reason that you can't 16 make the limitation zero, then some sort of compensatory 17 strategy would be used? Do you get any sense that there 18 has been any attempt in that direction, super safety 19 related, if you will?
20 A (WITNESS IANNI) Judss Carpenter, I am sorry, 21 I am not in the instrumentation field. So it is outside 22 my area, and I just haven't dealt in it that much to be 23 able to help you with that question.
24 0 In the spirit in whl=h Judge Brenner was 25 saying that these matters we will come back to, as also O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5565
} 1 a layman, I would be interested in that. Also, some 2 notion about what experience has shown is causing the 3 failures, what aspects of design apparently or
(])
4 operations is primarily contributing to these failures, 5 to get some perspective on how, if it is in conformance i
6 with the inst sentence, just exactly how it is, in terms 7 of present knowledge and possibilities for design 8 sen si tivity to the importance of this particular 9 component.
10 BY JUDGE MORRISs 11 0 While we are in the mood of expressing our 12 c urio si ty , with respect to the handout we received 13 yesterday, the question and answer f rom General 14 Electric, in the last paragraph on the first page it 15 starts out, "In the highly unlikely scenario," and then 16 it proceeds to f ail things in a very conservative way, 17 but again, I an afraid in every hearing I have.to quote 18 Lord Kelvin, who said, "I don't understand anything 19 unless I can measure it," so I don't have any sense of 20 quantification here, and I think we really need 21 something to put this whole process in perspective, and 22 the more quantitative it is, the happier I am.
23 (Whereupon, the witnesses conferred. )
24 MR. ELLIS: Judge Carpenter and Judge Morris, 25 we will bear those remarks in mind and have that l
I
()
i ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5566 O 1 information, unless any member of the panel wants to 2 respond at this time.
3
(]) WITNESS ROBARE: Well, just generally, the 4 statement " highly unlikely" had to do with a combination 5 of the reference leg break or leak and the worst' single 6 failure, which in this case was a total loss of AC and 7 DC on Division 2. We did not put a number on that, but 8 it was our judgment it was highly unlikely.
9 BY JUDGE MORRIS 4 (Resuming) 10 Q I understand what you did, and it is a 11 bounding situation, and you thought of all of the bad 12 things that could happen, but you know, it is sort of 13 like understanding margin. You need to somehow put it 14 in perspective in using some kind of quantitative s 15 concept. I won't tell you how to do it, but if you can 16 think about it before you come back.
17 (Whereupon, the witnesses conferred.)
18 (Whereupon, the Board conferred.
19 Jt'DGE BRENNER : We have a very few remaining 20 number of Board questions that are not vessel level 21 questions, if you will, and I think I will go ahead and 22 ask them now, before we inquire as to whether there are 23 further questions, so that any further questions can 24 include any responses, instead of having to go through 25 yet another round.
(
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5567
() 1 BY JUDGE BRENNER. (Resuming) 2 Q Mr. Dawe, some time ago, I guess, by now, you
() 3 4
were asked questions about Study D, which is the cable separation study done by Stone and Webster, and I 5 believe you indicated on the transcrip t ( c 1 hat time --
6 it was June 21st, but I don't have a citation -- that in 7 your view, advanced methodology was used in that study, 8 and although you did describe the study quite well, at 9 least if I understand what was done in the study -- I am 10 not sure I understand, particularly what you had in 11 mind, to which you would attach the label " advanced 12 methodology" and what that methodology was.
13 A (WITNESS DAWE) Yes, sir. Let me try to O 14 explain in a little bit more detail. With respect to 15 advanced methodology, I think I meant powerful 16 methodology. I think I meant detailed consideration 17 within the methodology. It is certainly not a 18 methodology that isn't a compilation of other 19 capabilities that we have always had. It was a 20 collection of things that we knew and had done and had 21 done before, and to that extent, I mean advanced.
22 Very briefly, we identified essentially four 23 input parameters from our shutdown model which we
(} 24 thought very carefully about for all the safety related 25 functions we needed from ho t opera tion to cold O
ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE S.W., WASHINGTON. D.C. 20024 (202) 554-2345
5568
( 1 shutdown. We looked at all of the various modes of 2 operation between and the various system capabilities we
() 3 had to perform those modes of operation, and developed a 4 network or logic to go from Point A to Point B.
5 We did it by function, then we identified the 6 systems performing those functions. Then we identified 7 the auxiliary support system supporting the primary 8 systems performing the functions,, Then we took the 9 systems in detail and identified every component, every 10 instrument we needed for the proper function and 11 operation of those systems, including operator 12 intelligence in the control room. That gave us our 13 shutdown equipment list.
O 14 We used our elementary diagrams to identif y l 15 all cabling needed for the proper perf ormance of that 16 shutdown equipment list. We used our raceway drawings 17 for tray and conduit throughout the plant to identify 18 the routing of cable by raceway. We used out computer 19 cable schedules, which identify cable in raceway, cable 20 in conduit, to be able to spatially locate the cable.
21 We took all of this information, coded it, loaded it 22 into a computer program that we developed to be able to 23 sort all of this information out. We added to it 24 certain manual knowledge that we needed as a result of 25 design review, for example, a motor control center which O
ALDERSON REPORTING COMPANY,INC,
5569
(} 1 sits on the floor as cable from bottom entry. Raceway 2 schedules would not tell you the cable was coming to 3 that area, but we looked at that sort of thing as well
(]}
4 to identify the component in the area as part of the 5 cable, as if the cable we re in the a rea we were looking 6 at.
7 With all of this information, and with the 8 spatial location in the plant, and with the computer 9 program ability to identify exactly the function of any to cable to the proper operation of any piece of equipment, 11 we were then able to look area by area for the total 12 cable layout of the building, and as we selected areas 13 to analyze, we were able to eliminate from consideration 14 any piece of equipment in or affected by that zone and 15 cable within that zone.
16 I call it an advanced methodology simply 17 because we looked at so much at one time in such a 18 detailed manner. I do think it is an advanced study.
19 Q I recall your testimony the other day, but 20 just to make sure, you also -- I think you used the ters 21 rotated the segments. In effect, you were able to 22 overlap the segments by doing that.
i 23 A (WITNESS DAWE) Yes, sir. Very simply, the 24 45-degree segments on an elevation would give us eight 25 study zones. At the edge of one, from Zone 1 to Zone 2,
)
ALDERSON REPORTING COMPANY,INC,
! 400 VIRGINlA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5570 i
- 1 f or example, you would have a boundary, and we asked 2' ourselves the question, is there something right on the
() 3 other side of that boundary that would significantly 4 affect our analysis, and to account for that, we stepped 5 the segments 22 and a half degrees after the first study 6 of an elevation, and did a second study of the same 7 ele va tion , which put the new boundaries in the middle of 8 the old zones and the old boundaries in the middle of 9 the new zones, so we ended up overlapping to make sure 10 there were no critical boundaries within our studies.
11 0 You described the study in the direct 12 testimony, and I think in response to a question as a 13 spatial interaction study, I submit for your comment 14 that in attempting to categorize these studies for 15 purposes of the f act that there are these abstract 16 categories of interaction studies, you may have not 17 presented the full picture, because, as I hear you 18 describe it, a f unctional stud y wa s necessa ry before, 19 during, and after in order to understand the 20 significance of the spatial situation.
21 A (WITNESS DAWE) With that description, I would 22 agree, there is functional study involved, but I 23 characterize it as a spatial study because we were l
() 24 looking at a spatial hazard, and the way in which any 25 spatial hazard could affect more than one system at the I
ALDERSON REPORTING COMPANY,INC,
( 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345 l
5571 1
same time.
2 0 Recognizing that there might have to be 3 adjustments made for other types of studies, would the
)
4 kind of approach you just described be useful for i
~
5 studies of some of the other matters listed in the 6 testimony, such as the high energy line break studies, 7 the control system failure studies, and in general the 8 others, even those ones tha t you described as spatial, 9 and I am suggesting also the ones that you described as 10 functional.
11 A (WITNESS DAWE) Yes, Judge Brenner, I think 12 they are useful, but they will not totally answer every 13 question. The additional studies that we do combined
) 14 with that study give us a great deal of confidence.
15 There are certain requirements for certain types of 16 studies that we have to assume on a regulatory basis.
17 for example, that would keep us from using that type of 18 a study to answer a regula tory question .
l 19 However, as I mentioned in my testimony, we 20 did not start that study to answer regulatory 21 questions. We were looking at the total building, and 22 had a lot of things in mind at the time we did it. For 23 example, the question of heavy loads was in our mind at 24 the time we did that study, as you do maintenance in the 25 plant and move things around in the plant. The question O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINtA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554-2345
5572
( 1 of pipe break was in our mind, although we had done the 2 pipe break study. We were looking at fire, and the 3 impacts or fire. That was in our mind. Basically, we
(])
4 were looking at all the types of hazards that can exist 5 in the res: tor building when we started the study.
6 Subsequently, we started to look at regulatory 7 questions that fortunately we were prepared to answer 8 because we had done the studies or we were well along to 9 completing them , a nywa y.
10 0 You say that that type of study wouldn't 11 answer all of the questions. I have the sense of it 12 being the other way around, and maybe you can tell me 13 where I am wrong. That is, in order to answer a lot of O 14 the other questions on the schedule in which you would 15 like to answer it, answer the particular questions, you 16 wouldn't have to undergo a study of this type, but if 17 you had for your own purposes performed such a study, 18 you wouldn't have to do very much else in order to 19 answer the questions.
20 I see the study as broader than answering the 21 question, so I am wondering what types of questions, if 22 you can generalize, and maybe you can't, keeping in mind 23 classifications and systems interaction, could not be 24 answered by the type of study you just described. It is 25 a very broad question, and if it is not susceptible of O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W, WASHINGTON, D.C. 20024 (202) 554 2345
5573
( 1 an answer, you can tell me that also.
2 (Whereupon, the witnesses conferred.)
3 (WITNESS DAWE) It is a broad questions, and I
(]) A 4 am not sure I am going to be able to answer it 5 completely without thinking about it a lot more than I 6 have done. Questions that I would have to ask to apply 7 it to any particular problem or hazard would be 8 something such as the hazard I am looking at and whether 9 my boundaries are big enough, or too big to be realistic 10 f or the haza rd I am looking at, assumptions I would make 11 in terms of other things happening at the same time in 12 combining the total loss.
13, For example, when you do the study, you really 14 give up a lot of equipment in the plant, when you take l
l 15 that much cable from any one hazard, and when you start 16 combining it with, have you had an earthquake. In this 17 case, I can combine it with, have I lost off-site power, 18 because I relied only on safety related items, powered 19 from on-site power sources, but that is the type of 20 questions, w hat are the assumptions for the study that I l 21 have to make, what is the size of the hazard. Things 22 have to be looked at in relationship to the hazard.
23 Again, the building was looked at, 45-degree 24 segments tha t we looked at are large segments, and in 25 some areas include an awful lot of equipment, and that O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5574
() I gives us a great deal of comfort that there is a great 2 deal of redundancy and separation in the building. To
() 3 ask me about a specific hazard on a specific regulatory 4 basis with all of the conservative assumptions, I would 5 have to look to see if it was answered by the study. I 6 think generally it would be.
7 0 All right. Let me try this as a followup. I 8 didn't mean to limit you to segments that large, or a I 9 study tha t b road. In other words, what I had in mind 10 was all of those approaches, that is, looking at the 11 plant, looking at particular segments that vould be 12 pertinent given the purpose of the study to look at, 13 looking at the logic, both on design and as it actually l
O 14 exists in the plant, analyzing the spa tial situation in 15 light of the functional situation, and that type of 16 thing.
17 The sense I get from your answer is, yes, you 18 could apply that, but you would want to make the 19 adjustments. That is, it is easier to do the broader 20 study, and then if you find out you have a problem, to 21 look closer as to whether that " problem" is a problem in 22 light of what you have to look at. That is, do the 23 f urther more detailed analyses, given the particular 24 hazard, when you are adapting it to a pa rticular 25 concern. Is that a reasonably fair summary?
O ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5575 1 (Whereupon, the witnesses con f erred. )
2 A (WITNESS DAWE) I think I understood wha t you 3 said, and I think the answer is yes, that this type of
(]}
4 methodology of spatial versus functional is applicable 5 and can be looked at, and I would also say tha t I think 6 that is just what we do.
7 0 On one of the other studies, Study E, and I am 8 using the shorthand in your testimony, it was the 9 failure modes and effects analysis study, I think around qg 10 Transcript Page 5116, but I don't think you have to look ,
11 at it, which was June 21st, you mentioned that as a 12 result of that study, you found that an isolation valve 13 needed some modification, and you indicated' a little bit ,
14 about what was involved, but frankly I have forgotten.
15 Can you very briefly refresh my recollection?
16 A ( WIT N ESS DAWE) Yes, sir. To the best I know 17 the de tails of the problem, which is not a great deal, 18 the valve that I was referring to was, I believe, in the l
19 reactability and closed loop cooling water system, which 20 is a two-train system, and on accident signals is to be 21 split into two trains to ensure the integrity of both 22 trains. The logic f or those valves required those 23 isolation signals. The elementary diagrams, I believe, 24 had an error in them which would not have assured that 25 isola tion, and it was the FMEA f rom the logic, and then O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W, WASHINGTON, D.C. 20024 (202) 554-2345
5576
() 1 the review of the elementary diagram f rom which the FME A 2 fault tree is completed that showed us that we had not
(} 3 done the elementary correctly for that one valve, and as 4 a result of that a change was made in the elementary 5 diagram to ensure that the valve would receive the 6 proper isolation signals at the proper time, and that 7 was what I was referring to.
8 0 So it wasn't the disclosure for the first time 9 that the design had to have that logic sepa ra ted so much 10 as a disclosure that as it was transferred to the 11 elementaries which, of course, would be used to perform 12 the construction and check on the construction, that 13 that intent was not carried forward and that correction O 14 had to be done?
15 A (WITNESS DAWE) That is true.
16 (Whereupon, the Board conferred.)
17 BY JUDGE CARPENTER (Resuming) 18 0 Mr. Robare, on Page 63 of the prefiled 19 testimony, Item M, which I believe you testified on 20 cross examination is a 1970 GE study --
21 A (WITNESS ROBARE) That is correct.
22 0 -- is that study prop rie ta ry?
23 A (WITNESS ROBARE) No, it is not.
24 0 Did it receive any sort of external peer 25 review?
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5577 1 (Whereupon, the witnesses conferred.)
2 A (WITNESS ROBARE) Do you mean external to GE?
3 0 Yes.
4 A (WITNESS ROBARE) I as trying to recall. If 5 it were a licensing topical report, it would have 6 received NRC review. Is that the sort of review you are 7 asking about?
8 Q Or ACRS. Or National Academy. I don't care.
9 A (WITNESS ROBARE) I don't have it with me 10 right now. I can at the break, I can find out.
I 11 Q Thank you. Turning to Item N, which I believe 12 you testified at cross examination was a GE report, 13 1976, is that report proprietary?
O 14 A (WITNESS ROBARE) That one is proprietary , yes.
15 0 Turning to Page 64 --
16 (Vhereupon, the Board conferred.)
17 Q Judge Brenner suggests I ask whether that one 18 received any sort of outside review.
19 A (WITNESS ROBARE) The scram reliability report?
20 0 Yes.
21 A (WITNESS ROBARE) Yes, that received extensive 22 review by many utilities and by the NRC in the context 23 of the ATWS generic issue.
24 Q Turning to Page 64, Iten 0, 'hich I believe 25 you identified on cross examination as a 1970 GE report, O
1 l
l ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE, S.W, WASHINGTON. D.C. 20024 (202) 554-2345
5578 I
() 1 I would like to ask again whether that was proprietary.
2 A (WITNESS ROBARE) That report is not 3 proprietary.
4 0 Did it receive any sort of outside review?
5 A (WITNESS ROBARE) Yes, it was a licensing 6 topical report, and it received NBC review.
7 0 And the last item that I have for the same 8 kind of questions is, Item Q, the THI 2 implications 9 item, is tha t proprietary?
10 A (WITNESS ROBARE) We have three categories of 11 reports at General Electric, non proprietary, and then 12 two categories that I would consider proprietary, or at 13 least for two different reasons. One is for a O 14 commercial reason. For instance, a group of utilities 15 paid for the report and would not like to have that 16 distributed without their permission, and the other is 17 the classical proprietary sense, and that is a 18 competitive concern at General Electric.
19 This particular one was sponsored by owners, 20 and it is proprietary in the sense that we could not 21 release it without their permission.
22 0 Has it received outside review or peer review?
23 A (WITNESS ROBARE) Yes. It received extensive
( 24 review from those owners that sponsored the study and 25 from the NRC.
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
i 5579
() 1 JUDGE CARPENTER: Thank you for making the 2 record clear.
3 JUDGE BRENNEHa All right. Any further cross 4 examination based upon questions since the last chance 5 the county had can be asked at this time.
6 RECROSS EXAMINATION 7 BY MR. lANPHER:
8 0 Mr. Robare, your counsel gave parties copies 9 of this question and response that GE prepared. Am I
~
10 correct that this is a summary of a larger underlying 11 study?
12 'A (WITNESS ROBARE) Yes, it is, Mr. Lanpher.
13 0 What is the name of the underlying study? Do O 14 you know?
15 A (WITNESS ROBARE) There is no study title, or l
i 16 formal report. This is the conclusion of the engineers 17 that studied this particular scenario, and the backup to 18 that, to these conclusions is included in their design 19 record files at GE.
20 0 When was this summary prepared, or conclusion?
l 21 A (VITNESS ROBARE) Approximately in November of 22 1981. That would be last November.
23 0 Now, am T correct that there isn't a single 24 pulled together document which suppotis these i
25 conclusions, but rather there are more likely a number
()
ALDERSON REPORTING COMPANY. INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2045
5580
() 1 of documents from which these conclusions were derived?
2 A (WITNESS ROBARE) That is correct. We formed 3 a task force of engineering and licensing engineers, and
")
4 we made this evaluation f or each of our product lines, 5 and the backup information is included in their 6 records. It was not a report that was formally 7 p ubli sh ed , nor was it requested.
8 0 Was this task force formed to analyze the 9 situation in response to the draft of the Michelson 10 memorandum, or what was referred to as the Michelson 11 memorandum?
12 A (WITNESS ROBARE) That is correct. I would 13 like to clarify that it was formed in response to O 14 anticipated questions from the NRC. It was not formed 15 in response to that situation being recognized as an 18 undisclosed design problem. We had considered that 17 scenario previously. We were merely anticipating the 18 need to document that conclusion with the NRC.
19 Q I believe you testified in response to Mr.
20 Ellis this morning with respect to the boiloff issue 21 that this had been known to General Electric and 1
22 evaluated in the original design and analyzed in detail 23 as part of the design process for Shoreham. Is that 24 correct?
25 A (WITNESS ROBARE) Actually, it was racognized O
Ame. son ee oe- co_. ,Nc.
l 400 VIRGINIA ave S.W., WASHINGTON. D.C. 20024 (202) 554-2345
5581
() 1 before Shoreham by our design engineers. Shoreham is 2 not unique in this concern.
3 0 So it was recognized even before 1967, 1968,
(])
4 when Shoreham . was intitially ordered f rom GE?
5 A (WITNESS ROBARE) I wouldn 't know exactly what 6 time to put on it, but I have discussed this with the 7 design engineers who designed the basic water level 8 system some time in the sixties.
9 0 Do you know why this design was chosen in 10 light of the fact that you knew that there'was a 11 potential interaction involved through the boiloff ?
12 A (WITNESS ROBARE) Yes. The design was chosen 13 for its high quality and reliability. The potential for O 14 boiloff was considered, and was considered to be -- it 15 was considered that that interaction was not er great 16 safety significance. In other words, the design was 17 fully suitable in consideration of that potential 18 interaction., We still feel that way.
19 Q Was any consideration given to designing this 20 system in a way to preclude the interaction?
21 A (WITNESS HOBARE) Do you mean back in the 22 sixties?
23 Q At any time. Let's start originally in the 24 sixties. If you know.
25 A (WITNESS ROBARE) Yes, in the sixties the O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON D.C. 20024 (202) 554-2345
5502
() 1 scenario was considered, and it was deemed to be an 2 acceptable situation, and no design change was 3 contemplated at that time.
4 Q Before we go to the more recent time period, 5 can I just follow up on that? What criteria does GE 6 apply to determine whether an interaction between 7 systems is acceptable?
8 (Whereupon, the witnesses conferred.)
9 A (WITNESS ROBARE) Well, first of all, we would 10 assure ourselves that the design iet all of the 11 appropriate codes and regulations, and then for specific 12 cases such as this, we would evaluate the potential 13 scenario and worst case possibilities, and assure O 14 ourselves that there is no safety concern.
15 Q What do you mean by safety concern, Mr. Robare?
16 A (WITNESS ROBARE) Well, in this particular 17 case we assured ourselves that the water level would be 18 maintained above the core, and there would be no fuel 19 failure.
20 Q Have you subsequent to the 1960's period that 21 we are talking about considered any change in design to 22 eliminate the boiloff problem?
23 A (WITiESS ROBARE) Yes, we have evaluated, we 24 have re-evaluated the design. We have also participated 25 in the BWR owners' group investigation of the design, O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5583
() 1 and in that context of those ef forts, there were or 2 there was consideration of potential modifications to 3 the design.
)
4 0 Is it fair to state that instead of design 5 modifications, it is GE's recommendation to ensure that 6 this interaction does not become a safety concern by 7 making sure that the proper emergency procedures are 8 followed and operators are trained to recognize the 9 problem, and act in a manner which will assure safe 10 conditions?
11 A (WITNESS ROBARE) Proper operator training and 12 alertness to the situation certainly is advisable and 13 has been implemented. The basic design, though, is O 14 still judged to be perfectly adequate.
15 (Whereupon, counsel for Suffolk County 16 conferred.)
17 (Whereupon, the witnesses conferred.)
18 0 Mr. Robare, in the prefiled testimony you have 19 spoken about the defense in depth concept for nuclear 20 power plant design. How is it consistent with that 21 defense in depth concept not to eliminate to the extent 22 feasible the potential adverse interactions relating to 23 the boiloff issue?
} 24 A (WITNESS ROBARE) Because I stated previously 25 there are interactions, particularly environmental O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE S.W., WASHINGTON. D.C. 20024 (202) 554-2345
5584
() 1 interactions, that a re a na tural occurrence. When you 2 postulate these types of events the question of safety 3 or the design question is whether or not th at 4 interaction poses a safety concern, and in this 5 particular case we have extensively evaluated it, and 6 f eel that there is no safety concern that requires a 7 design change.
8 A (WITNESS DAWE) If I might put that somewhat 9 in context as well, you will recall that the boiloff to problem in an accident situation as described in the 11 analysis requires many, many things to happen in terms 12 of operator error, turning off his ECCS, a long time 13 f rame af ter he turns it off, long time f rames of not O 14 turning it back on, despite the level indication that he 15 is getting, and so forth.
16 That is an extremely remote situa tion, but 17 within the defense in depth concept, I might add that as 18 I have testified before, we have looked at our plant and 19 we have looked at our dry well temperature conditions.
20 We have improved our dry well cooling extensively, and I 21 think that is well within the defense in depth concept 22 that we are talking about. We have taken those steps on 23 Shoreham. We have recognized that BWF 's have higher
( 24 temperatures than were previously predicted. We have 25 looked at our designs. We have doubled our capacity.
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554 2345
5585 l
() 1 We have improved our air flow, and I think that is very 2 much in our defense in depth concept.
3 0 You have not changed the design, however, of
)
4 the water level indicator system to preclude the boiloff 5 poten tial, have you? Or maybe that should go to Mr.
6 Robare.
7 A (WITNESS ROB ARE) No, we have not, since we 8 don't judge that as necessary.
9 0 Now with respect to the water leg break issue 10 as Mr' Ellis split up the issue earlier this morning, I 11 believe you stated that you have known about that 12 potential situation since the mid-sixties, correct?
13 A (WITNESS ROBARE) That is correct.
O 14 Q I believe you also said it was a conscious 15 desicion to design the system notwithstanding tha t 16 knowledge, correct?
17 A (WITNESS ROBARE) That's right.
18 0 Have you analyzed means to eliminate that 19 problem?
20 A (WITNESS ROBARE) Since we are postulating the 21 failure of -- I don't know how it is possible to 22 eliminate that problem.
23 0 Have you analyzed means to ensure that if 24 there is a reference line break, that control systems 25 vill not interact adversely with protection systems?
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) $$4 2345
5586
() 1 Maybe another vsy of asking this is, have you analyzed 2 means of uncoupling the control systems from the 3 protective systems in the sense that they come off the
{
4 same instrument lines?
5 A (WITNESS ROBARE) I don't know that analyzed 6 is a term I would use. Obviously, if all of the 7 instrumentation were separate, there would be no 8 interconnection. That is a simple fact.
9 Q Has GE considered --
10 MR. ELLISs Excuse me. I am not sure the 11 witness was done with his answer.
12 (Whereupon, the witnesses conferred.)
13 JUDGE BRENNER. Mr. Lanpher, we would like to O 14 take a midmorning break as soon as it is convenient, 15 somewhere in the next few questions.
16 BY HR. LANPHER: (Resuming) 17 0 Did you finish your previous answer?
18 A (WITNESS ROBARE) Yes, I have.
19 0 You said that you are not sure that my term 20 analyze was correct. Have you considered means to 21 eliminate this interconnection problem between the 22 control systems and the protection systems, the problem 23 that has been described in Suffolk County Exhibit 1?
24 A (WITNESS ROBARE) Do you mean, are we aware of 25 a design possibility that would eliminate that O
ALDERSON REPORTING COMPANY. INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5587
() 1 interaction?
2 0 Yes, sir.
{} 3 A (WITNESS ROBARE) Yes.
4 Q There is such a possibility?
5 A (WITNESS ROBARE) Yes, there is.
6 0 You have not implemented that?
7 A (WITNESS ROBARE) At Shoreham?
8 Q At Shoreham.
9 A (VITNESS ROBARE) No, we have not, 10 0 When did you become aware of the possibility 11 of the means to avoid it?
12 A (WITNESS ROBARE) Well, it is a simple fact.
13 I don 't think we had a revelation about that. If the O 14 instruments are on completely separate channels, there 15 is no interaction. It is not a matter of when did we l
16 become aware of that.
17 Q Is that what the design change would be, 18 putting the instruments on separate channels?
19 A (WITNESS ROBARE) That would be the simplest 20 way to eliminate any possible interaction, yes.
l 21 Q And that would be feasible?
l 22 A (WITNESS ROBARE) Not feasible for Shoreham, 23 no.
24 Q It would not be feasible because the design l
l 25 has already been implemented?
! CE)
ALDERSON REPORTING COMPANY,INC, l
400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5588
() 1 A (WITNESS ROBARE) No, because the vessel has 2 already been built. As Mr. Ianni testified, it is not a 3 fessible considerstion to add additional taps on the 4 reactor pressure vessel at this point.
5 0 It would have been f easible to im plement this 6 design change, in your opinion, if it had been made 7 initially back in the sixties before the pressure vessel 8 was manufactured?
9 A (WITNESS ROB ARE) Yes. That would have been 10 feasible. However, the design was evaluated and judged 11 to be adequate without those additional lines, as 12 suggested in this writeup on the reference leg break 13 analysis.
O 14 0 So would it be fair to state that you reached 15 tha t conclusion notwithstanding the fact that you knew 18 there were potential adverse interactions resulting --
17 which might result from a reference line break?
18 (Whereupon, the witnesses conf erred. )
19 A (WITNESS ROBARE) As the summary of the report 20 shows, we do not feel that it is an adverse systems 21 interaction. It is a systems interaction, but it has 22 been evaluated, and the consequence is, we do not 23 consider it to be adverse.
() 24 0 Do you have a copy of LILCO Exhibit 13, which 25 is the Denton to Michelson memorandum dated October 30, O
ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5589
( 1 1981, sir?
2 A (WITNESS ROBARE) Yes, I do.
3 0 Looking at the second sentence, and I will f])
4 read it, "We reconfirm that an instrument sensing line 5 malfunction could be the initiating event for adverse 6 control system action and simultaneously affect the 7 limited number of protective system channels." They use 8 the term " adverse control system action." Do "you 9 disagree with that characterization, " adverse control to system action?"
11 JUDGE BRENNERs Mr. Lanpher, you are talking 12 about the October 30th, 1981 meno from Mr. Denton?
13 MR. LANPHERa Yes.
O 14 JUDGE BRENNER: I think that is Exhibit 20.
15 MR. LANPHERt I am sorry.
16 JUDGE BRENNER: No, maybe I have the wrong 17 e xhibit . I am sorry, it is 13.
18 BY HR. LANPHER: (Resuming) 19 0 Do you recall my question, Mr. Robare?
20 A (WITNESS ROBARE) Yes, I do. It is certainly 21 -- it has an adverse effect upon the control system 22 which is rendered inoperative. I would agree with 23 that. My point is, it does not have an adverse effect 24 on the overall plant safety, because of the backup 25 systems tha t are a vailable , because you get an automatic O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554-2345
l l
5590 l l
l O i scraa, as 4.cumentee in the su.aarr or our steer. l
- 2 0 This interaction vill reduce the number of 3 channels which give you accerate protective action, 4 correct?
l 5 A (WITNESS ROBARE) Tha t is correct. ,
6 MR. LANPHER: This is a convenient place to 7 take a break.
8 JUDGE BRENNER: All right.' Let's break until 4
9 10:55.
10 (Whereupon, the hearing was briefly recessed.)
i 11 ,
12 13 O 14 l
15 16 17 18 19 20 l 21 22 23 l O 24 l
25 O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE, S.W, WASHINGTON, D.C. 20024 (202) $54 2345 l
t -_ _ _.
5591
() 1 JUDGE BRENNER: Let's go back on th e record.
2 Mr. Lanpher?
(} 3 4
BY HR. LANPHER: (Resuming)
Mr. Robare, has CE analyzed whether other taps Q
5 on the pressure vessel might be utilized as a feedvater 6 control mechanism, to have that as a tap for feedwater 7 control mechanism?
8 A (WITNESS ROBARE) I wouldn 't call it aa 9 analysis. We have -- we understand that that would 10 provide additional separation, yes. That is obvious.
11 0 I am not talking about new taps, existing 12 taps, whether they might be utilized f or that purpose.
13 A (WITNESS ROBARE) I as sorry. I am confused O 14 now. Could you please restate it?
15 Q I apologize, Hr. Robare. Has GE analyzed, and 16 I use the word loosely, given consideration to whether 17 there are existing taps on the pressure vessel which 18 might be utilized for the feedvater control mechanism?
19 (Whereupon, the witnesses conferred.)
20 A (WITNESS ROBARE) I just don't have personal 21 knowledge of that.
22 0 Mr. Robare, do you know whether there is any 23 o ther single f ailure event that will remove as many 24 protective system channels as will the reference leg 25 break that we have been discussing ?
( .
ALDERSON REPORTING COMPANY,INC.
400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5592
() 1 (Whereupon, the witnesses conferred.)
2 A (WITNESS ROBARE) Could you try that one more 3 time, please?
{} 4 Do you know whether there is any other single 0
5 failure event that will remove as many protective system 6 channels as will the reference leg break event that we 7 have been discussing?
8 A (WITNESS *ROBARE) I can give you my immediate 9 judgment on that. Obviously, to be precise would 10 require much more study. I believe if you lost one of 11 the reactor protection system MG sets, you would lose 12 the signals for many more protective ch,annels. However, 13 it is a f ailsafe design such that if that were to occur, O 14 you would scram.
15 (Whereupon, the witnesses conferred.)
16 (Whereupon, counsel for Suffolk County 17 conferred.)
18 A (WITNESS ROBARE) I believe I need to clarify 19 that. You would not get a full scram of the reactor, 20 you would get what is termed a half scram.
21 0 You referred to that as a failsafe design. Is 22 that correct?
23 A (WITNESS ROBARE) Yes.
() 24 0 With respect to the reference leg situation, 25 that is not a failsafe design, correct?
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W WASHINGTON, D.C. 20024 (202) 554-2345
5593
() 1 A (WITNESS ROBARE) I don't think I would 2 categorize it failsafe in the normal sense of that 3 term. However, as I have previously testified, and as 4 our report summary states, the overall effect on the 5 reactor is safe.
6 0 What did you mean by a failsafe design with 7 respect to your previous answer where you had referred 8 to a design as failsafe?
9 A (WITNESS BOBARE) Generally, failsafe implies 10 that the failure of the component would result in an 11 action that would be in the direction of safety or in 12 this case scram, scram signal.
13 0 Mr. Robare, can the water level reference legs i
O 14 be isolated from the reactor?
15 A (WITNESS DAWE) Mr. Lanpher, yes, they can -
16 be. In fact, they are automatically isolated from the 17 reactor by design outside of containment by excess flov 18 check valves. Inside the containment, to m y knowledge, 1
l 19 they are not isolable from the reactor.
20 0 Are there -- thus, there are some active 1
21 components on the reference leg lines. Is that correct?
! 22 A (WITNESS DAWE) Just like any other instrument 1
23 line coming from the high pressure source inside the
() 24 containment. Yes, sir.
25 0 Is it fair to state, then, that there can be O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON D.C. 20024 (202) 554 2345
5594
() 1 active failures as well as passive failures on these 2 lines?
3 (Whereupon, the witnesses conferred.)
4 A (WIT. NESS DAWE) No, I would not say that an 5 excess flow check valve would have that failure mode.
8 There is no flow in that line, and with no flow, the 7 excess flow check valve, which is a spring-loaded ball l 8 ' check valve, just has no failure mechanism th a t ca n i 9 close off that line.
I 10 0 Are you considering the only failure mechanism 11 to close that line? Is that the only failure you have 12 in mind?
l 13 A (WITNESS DAWE) There is no active failure
() 14 that can open that line.
)
15 0 Do you have a copy of Suffolk County Exhibit 16 1, Mr. Dava?
17 A (WITNESS DAWE) Can you describe that exhibit?
! 18 0 That is the Michelson memorandus. I would 19 like you to turn to Page 23.
20 A (WITNESS DAWE) We have that.
21 0 The bottom paragraph on that page refers to 22 Millstone Unit 1 incident, and it states, an i
l 23 investigation was -- an investigation of a field valve 24 that is normally used for filling the system was i 25 leaking. Do you consider that to be an active or
()
I ALDERSON REPORTING COV9ANY. INO.
400 VIRGINIA AVE., S.W WASHINGTON. D.C. 20024 (202) 554-2345
5595
() I passive failure, sir?
2 A (WITNESS DAWE) I would have to look at the 3 design to be sure of that, but I don't believe that that
{} 4 valve is on the reference leg. I think that is a fill 5 valve down at the instrument rack. So with respect to 6 your question about an active f ailure in the reference 7 leg, I would not have included that valve.
~
8 0 Doesn't the failure which is described there 9 for Millstone affect the reference leg, whether it is.
10 physically on that leg or not?
11 A (WITNESS DAWE) Yes, that failure can cause 12 the reference leg to d rain. However, I might note in 13 that type of a case with such a thing happening with the O 14 reference leg draining slowly through a valve, with the l
- 15 condensing changers, it is highly likely the reference 16 leg will stay full and not affect the instrumentation at 17 all. You might have a slow leak into the secondary 18 containment. If you got into a situation where you vore 19 leaking faster than you were making up through the 20 condensing chamber, you would not ha ve a ca tastrophic 21 failure, instantaneous failure in the instrumentation.
22 What you would have is a slow change of instrumentation, 23 which would be a far less severe event and it would give
( 24 the operator far more time to recognize the situation.
25 Things would not happen quickly at all in that
()
ALDERSON REPORTING COMPANY,INC, 400 VIRGINtA AVE S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5596
() 1 situation. I would not characterize that as anything 2 like, for example, a line break, where things would 3 change quickly as a result of the reference leg changing 4 quickly.
5 Q Can you turn your attention to the bottom of 6 Paze 24 and the top of Page 25 of the same document?
7 And I really want to turn your attention to the first 8 full paragraph on Page 25. The discussion starts on 9 Page 24, at the bottom, the last three lines. As I 10 correct that this description of the Brown's Ferry event 11 indicates -- excuse me for one moment.
12 (Whereupon, counsel for Suffolk County 13 conferred.)
O 14 0 Is it not correct that this event involved a 15 reference leg reading lowi
- 16 A (WITNESS MC GUIRE) That is correct.
17 Q And is it also correct that one of the 18 possible causes was inadvertent operation of an 19 equalizer or drain valve?
20 A (WITNESS MC GUIRE) That is possible, but 21 usually after the reactor has been shut down, which this 22 condition indicated cold shutdown. It is usually ICC's 23 -- Instrument and Control group would normally fill
() 24 those legs. It just so happens that this event was 25 recognized af ter the operators had started pulling rods, O
ALDERSoN REPORTING COMPANY,INC.
400 VIRGINIA AVE, S W, WASHINGTON, D.C. 20024 (202) 554-2345
5597
() 1 which required it to be reported then, but it is common 2 after a long shutdown to verify those things are filled ,
3 because the reactor is not operating and the conden sing 4 bars can't condense.
5 0 Mr. McGuire, in inadvertent operation of the 6 valves, would you consider that to be an active or a 7 passive failure?
8 A (WITNESS MC GUIRE) That would be an active.
9 (Whereupon, counsel for Suffolk County 10 conferred.)
11 Q Mr. McGuire, yesterday --
I don't have the 12 transcript page exactly, but I believe you stated that 13 maintaining water level is the highest priority for O 14 reactor operators. Do you recall that statement, or 15 something to that effect?
16 A (WITNESS MC GUIRE) Something to that effect, 17 yes.
18 Q Would you agree with that statement?
19 A (WITNESS MC GUIRE) In the context that as 20 long as water covers the fuel, the reactor is in a safe 21 condition.
22 Q In fact, you would want to keep the water 23 somewhat above the f uel, so you would keep a margin.
() 24 Isn't that correct?
25 A (WITNESS MC GUIRE) That is the way the plan O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5598 O is de=taae$- re -
2 0 As an operator, Mr. McGuire, if maintaining 3 vater level is a high priority, do you also not want to 4 have the most reliable instrumentation to determine the 5 nccuracy of that water level?
6 A (WITNESS MC GUIRE) Yes, sir, and from my 7 experience, the instrumentation that is there is b reliable.
9 0 You make that statement in recognition of the 10 events which are described in the Michelson memorandum?
11 A (WITNESS HC GUIRE) Yes, sir, because Mr.
12 Michelson even states in there that nothing has really 13 happened because of these events.
14 (Whereupon, counsel for Suffolk County i 15 conferred.) -
16 0 Mr. McGuire, the Michelson memorandum does 17 consider this to be a safety concern, doesn't it?
18 A (WITNESS MC GUIRE) That is what the subject 19 says, safety concern associated with reactor vessel 20 level.
21 0 You dissgree with that characterization. Is 22 that correct?
23 A (WITNESS HC GUIRE) I disagree with many parts O 24 of this memorandum. 1 thinx the testimony that I gave 25 yesterday indicates that -- those areas specifically.
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5599
() 1 It doesn 't appear to give designers any credit for 2 actions.
3 A (WITNESS DAWE) Mr. Lanpher, if I could add to 4 that just briefly, in the sense of safety concern, 5 certainly the Commission is concerned with this in terms 6 of the safe operation of the plant, and we would agree 7 with that. However, we have anaylyzed this situation 8 a nd we have looked at this safety situation, and we 9 believe that this design is safe, and it poses no undue 10 risk to the health and safety of the public. It is 11 highly reliable instrumentation. The operators are 12 trained. The procedures cover problems with level 13 instrumentation, but even given that, the analysis that O 14 GE has done has shown that given the event that you are 15 talking about, the design still protects the reactor.
16 So, I would not disagree with your 17 characterization of safety concern. What I would do is 18 differ with your characterization, I think, of an 19 unacceptable safety concern for this reactor, for this 20 plant a t Shoreham.
21 A (WITNESS MC GUIRE) I would like to add 22 something to my statement before about, you asked me if 23 the operator action was an active or passive f ailure, 24 and I think I said it was active. There are procedures 25 instructing the instrumentation controls technicians on
~
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
l 5600
() 1 how to properly isolate equipment from service when they 2 are calibrating it, and I don't know why Brown's Ferry 3 keeps popping up in all of these issues, but the plant
[}
4 that I managed, we never had a problem with that.
5 (Whereupon, counsel for Suffolk County 6 conferred.)
7 0 Mr. McGuire, yesterday you described an event 8 a t Ha tch Unit' 1, and I believe that is the event which 9 is described in the final pages of the Michelson 10 memorandum, starting at Page 34.
11 k (WITNESS HC GUIRE) That is correct. .
12 0 That event resulted from an erroneous high 13 water level indica tion, correct?
O 14 A (WITNESS MC GUIRE) Yes, sir.
15 0 That event then included failures of both HPCI 16 and RCIC, correct?
17 A (WITNESS HC GUIRE) I think, as I stated 18 y e s te rd a y , that is correct. Hatch has been having, I 19 would classify it as unusual problems with those two 20 systems since their initial startup of the plant, and 21 this was recognized in our evaluation of the plants in 22 '79. In fact, NUREG-0737 requires actions to be taken 23 to reduce that failure mode rate which had occurred due 24 to the fact that quick start of the turbines created a 25 pressure spike on the lines, which caused the elbow taps O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5601
() 1 which are there to detect pressure spike in the context 2 that it would iden tif y a break in that line, and as such 3 these pressure spikes caused the turbines to isolate.
4 Since that has happened, General Electric has made a 5 design change where they have incorporated a 6 three-second time delay which allows an ample time for 7 that pressure spike to subside, and the realibility of 8 the quick start has improved tremendously since then.
- 9. That design is applied to Shoreham.
10 Q At Hatch, af ter a period of time, 11 approximately five minutes, they were able to get the 12 HPIC operating, correct?
13 A ( WITNESS MC GUIRE) Yes, sir.
O 14 Q Were they ever able to get RCIC to operate in 15 tha t incident?
16 A (WITNESS MC GUIRE) I was involved in the 17 evaluation a t INPO, and no, they were not.
18 0 Now, the problems with the-HPCI and RCIC were 19 not limited to Hatch, were they?
20 (Whereupon, the witnesses conferred.)
21 A (WITNESS MC GUIRE) There have been other 22 isolations as described in the Michelson report.
23 0 Mr. McGuire, are you familiar with the
() 24 INP0/NSAC report which is referenced on Page 34 of the 25 Michelson memorandum?
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5602 In fact, I lll 1 A (WITNESS MC GUIRE) Yes, sir.
2 consulted with Mr. Rosen and Mr. Gillespie on this 3 subject when I was at INPO.
4 0 And isn't it true that that report states that 5 during 1979 there were approximately 100 occurrences or 6 operational difficulties with HPCI and RCIC?
7 A (WITNESS MC GUIRE) It says there was more 8 than 100 reportable items. That does not necessarily 9 mean that it af fected the operation of those machines.
10 By tech spec, if you had a switch that is supposed to 11 trip at two pounds, and it tripped at 2.1 pounds, which 12 to me is insignificant, you shouldn't have to report 13 that, because they use this as a data base to check the 14 reliability of components.
15 Q Well, isn 't it a fact that NS AC concluded that 16 the HPCI and RCIC did not meet the high reliability 17 standards expected of nuclear power plant systems?
18 MR. ELLISs Judge Brenner, we object to the 19 line of questioning. We don't see what it has to do 20 with the water level, which is what I thought we were 21 on, and certainly goes beyond the scope of redirect.
22 JUDGE BRENNERs Mr. Lanpher?
23 MB. LANPHER: This relates, I believe, to the lll 24 Hatch incident, Number One. Second, it goes to the 25 water level indicator because this is a high priority O
ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE S.W., WASHINGTON. 0.C. 20024 (202) 554-2345
l l 5603
() 1 system. These are the systems that tre supposed to back 2 up the water level indication when it gives signals, and 3 we were trying to establish that these are not reliable, 4 at least historically have not been reliable systems or 5 highly reliable systems, and in our view it makes it all 6 the more important that the water level indicator be 7 reliable.
8 JUDGE BRENNER4 If we were litiga ting the 9 inadequa te core cooling contention, I could understand 10 it. Do you want to tell me how it factors into this 11 contention?
12 HR. LANPHER: I think it relates to the 13 systems interaction concerns which we have been O 14 expressing, Judge Brenner, primarily in the sense that 15 these are some of the systems which must operate in the 16 event that the water level-indicators call upon them to 17 give the signals. If they are not highly reliable, this 18 is a further example of the failures that may result in 19 the event of spurious water level signals.
20 (Whereupon, the Board conferred.)
21 JUDGE BRENNER: We are not going to cut you 22 off, but you didn't make the best argument, and the 23 reason I say that is not a critique, but I want you to
() 24 keep the argument in mind in terms of your line of 25 questions. The purpose today is not to probe everything O
ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5604 l
l
() 1 about water level.
2 Hatch arguably -- well, not arguably, but it 3 relates to systems in*,eractions concern, or common cause 4 concern. I don't want to get hung up on the narrow 5 meanings of those words. And therefore what it tells us 6 about the previous methodology in terms of not 7 anticipating that design problem may or may not be 8 worthwhile in the context of Contention 7(b), and it is 9 that aspect that may be pertinent, not the aspect as to to water level, because I will tell you I could ask 11 questions for' two days about water level if I wanted to, 12 and I am sure you coald also, but you are going to have 13 that day or two some other time, and I an afraid --
0 14 well, there is certainly an overlap. I don't want to 15 have to repeat all of these questions then.
16 So, keep that in mind as your goal ra ther than 17 the water level by itself.
10 BY HR. LANPHER: (Resuming) 19 Q Mr. McGuire, do you recall the last question?
20 Do you want me to repeat it?
21 A (WITNESS MC GUIRE) I think the question was, 22 does NSAC think that these systems have poor 23 reliability. Is that the question ?
( 24 0 Well, let me repeat it, to be more precise.
25 That was it generally. My question was wbather NSCA
()
' ALDERSON REPORTING CGMPANY,INC, 4
400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5605 lhI 1 concluded that HPCI and RCIC did not meet the high 2 reliability standards expected of nuclear power plant 3 systems.
4 A (WITNESS MC GUIRE) In looking at the findings 5 and conclusions, you could speculate that is what they 6 were driving at, but unless you could show me 7 specifically where it says that, I don't think NSAC drev 8 that conclusion at all.
9 MR. LANPHER: J udge Brenner, I would like to 10 mark as Suffolk County Exhibit 21 for identification a 11 document entitled High Pressure Core Cooling System ,
12 Malf unctions a t Hatch. It is dated April, 1981. It is 13 by the Institute of Nuclear Power Operations, Nuclear 9 14 Safety Analysis Center, and I have provided copies 15 earlier to the Board and parties and witnesses.
16 JUDGE BRENNER: All right. It is so marked 17 for identification.
18 (The document referred to 19 was marked for 20 identification as Suffolk 21 County Exhibit Number 22 21.)
23 JUDGE BRENNER: I would note that, yes, you 24 provided copies earlier, and tha t was 9:00 o' clock this 25 morning, so we certainly haven't read it since ve have O
ALDERSON REPORTING COMPANY. INC.
400 VIRGINIA AVE S.W., WASHINGTON D.C. 20024 (202) 554-2345
5606 O ' heea ia e toa etace thea-2 MR. LANPHER: Judge Brenner, I would note that 3 the Hatch matter came up yesterday on redirect 4 examination, and I just'couldn't get it Xeroxed earlier.
5 JUDGE BRENNER: Yes. The reason I made my 6 comment was so that the record wouldn't imply that we 7 a re f amilia r with the contents of it at this time, and 8 your observation is also correct.
9 (Whereupon, counsel for Suffolk County 10 conferred.)
11 BY HR. LANPHER: (Resuming) 12 Q I would like you to turn to Page 1-2, Mr.
13 McGuire, the full paragraph on that page.
O 14 MR. ELLISs What page again?
15 MR. LANPHER: 1-2.
16 BY MR. LANPHER: (Resuming) 17 0 I will read that pa ra g ra ph . "This report 18 suggests that although public health and safety were 19 adequately protected by plant design and operation, the 20 HPCI and RCIC system did not meet the high reliability 21 ~ standards expected of systems in nuclear power plants."
22 A (WITNESS MC GUIRE) I think I made the same 23 statement myself a few minutes ago, where I said, if you O 24 reed the report, it suggest=, hut it didn t draw that 25 conclusion.
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2346
5607
() 1 0 That is the suggestion, though, of this 2 report. Is that correct?
3 A (WITNESS MC GUIRE) That is correct, and I 4 think GE could add to that somewhat.
5 A (WITNESS IANNI) Yes, Mr. Lanpher. GE and the 6 lead systems engineers who are responsible for these 7 systems are keenly awa re of these incidences, and have l 8 been since they have occurred, and as was pointed out, l 9 even some design changes have been made. Now, it is 10 perfectly proper for NSAC to point to the fact that 11 these systems can be made more reliable. And indeed 12 they can be, _and indeed a program is under way to make 13 them more reliable.
O 14 However, it is wrong to look at a single 15 component or a single channel of a multichannel system 16 and draw conclusions as to how effective ECCS is. The 17 facts are numerically that the reliability 'of the HPCI 18 need not meet high in terms of numbers like .99. The 19 reason for that -- excuse me. Let me back up. Even if 20 it is as reliable as .90 or thereabouts, the ECCS 21 network is not greatly affected.
22 The reason for that is that we have a 23 redundant and highly reliable ADS system which can bring
() 24 the system down and pressure rapidly, and then that 25 accesses not only the LPCI system with all of its pumps O ,
ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5008 lh 1 but other pumps as well. Even if you just look at the 2 LPCI and core spray systems which are low pressure, you 3 find that the overall protection to the core is still 4 very high. So, from that standpoint, you see, you have 5 to look at that from an overall systems network point of 6 view.
7 In addition to that, the BWR has access to 8 about 14. pumps that are not part of the ECCS network, 9 but which are scattered throughout the plant and which, 10 because we depressurize, if they are available, they can 11 be used to dump water into the core, and those pumps, 12 those pumps in excess of the LPCI and core spray pumps 13 are not even included in the reliability analysis, and I O 14 would like to suggest that there is no other reactor 15 around other than the BWR that has access to that many 16 pu.mps, and therefore we can say the core protection is 17 very high in spite of the failures that are recorded, i
18 and properly so, and fed back into the design.
19 A (WITNESS MC GUIRE) I would like to also add 20 that this report also suggests that Georgia Power could 21 do a lot more to make these systems more reliable if 22 they managed the plant and tested these components after 23 there had been maintenance done to them. That is also 24 implied. The other thing is that we have relied on HPCI 25 at Pilgrim. I consider it an extremely reliable system, O
ALDERSON REPORTING COMPANY,INC.
400 VIRG:NIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5609 O ' ad 1 thiax the coac1= io de wa, 1 evea de th t 2 st toment when I was at INPO to them, th t th e y were 3 heading la the wrong direction.
4 5
6 7
l 8 9
10 11 12 13 O 14 15 16 17 18 19 20 21 22 23 O u 25 O
ALDERSON REPORTING COMPANY,INC, 400 V.RGIMA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5610 They reached that conclusion or suggestion ll) 1 0
2 notwithstanding your suggestion to the contrary; 3 correct, Mr. McGuire?
4 A (WITNESS MC GUIRE) Yes. I guess political 5 pressure got to them.
6 (Laughter.)
7 JUDGE BRENNER: I guess you would tell me a 8 little more clearly what you mean by the fact that they 9 were hesding in the wrong direction. What conclusion 10 would you have drawn?
11 WITNESS MC GUIRE: Well, I think as I 12 men tioned before, I think one of the reasons we looked 13 at this system was specifically because of Hatch. It O 14 had a high failure rate on both of those systems. It 15 appeared to be a plant-specific piping layout problem.
to Really, I think the conclusions drawn in this report 17 that the systems were unreliable may apply to Hatch.
18 But as far as the rest of the systems that I 19 have operated, this is why you go through a testing 20 phase. Even during the tasting phase at Ha tch they had 21 problems with these things, isolating these two 22 systems.
23 And I just feel they did not go far enough at lhI 24 that time to correct the problem. And there va the same 25 problem over and over and over. And I just feel that O
ALDERSON REPORTING COMPANY, INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5611 lll 1 they could have done better.
2 And if you read the report, they did 3 maintenance on the controler, which actually controls 4 the quick-start of the turbine, and then turned around 5 and didn 't even test that component, which is, as far as 6 I am concerned, absurd in the operation of a nuclear 7 power plant. You do not work on a safety-related
~
8 component and then not test it.
9 JUDGE BRENNER: Well, now, in the testimony we 10 have heard from this panel, in part, it has been pointed 11 out that the f actoring in of operational da ta, including 12 and,in fact, perhaps particularly during the 13 startup-type testing phases is an important aspect, and 14 as a check, if you will, on a lot of things, including 15 whether or not systems classifications or systems 16 interactions have been done properly. A test of the 17 for-real as-built situation. Is that not correct?
18 WITNESS HC GUIRE: Yes, sir, that is correct.
19 JUDGE BRENNER: And is it not also correct 20 that procedures, whether required by technical 21 specificaions or commercisl interest of the plant 22 operator, are pretty much standardized in the nuclear 23 industry?
llh 24 WITNESS MC GUIRE: Ihat is also correct.
25 JUDGE BRENNER Well, what happened at Hatch?
O ALDERSoN REPORTING COMPANY,INC, l 400 VIRGINIA AVE., S.W., WASHINGTON, D.O. 20024 (202) 554-2345
5612 1 WITNESS MC GUIRE: That is what I would like
)))
2 to know, because these situations -- we had a similar 3 situation at Cooper when we first started Cooper up, and 4 that is where we made a design change to install the 5 ramp function generator, and that problem then went 6 away. That is as I described yesterday. This is one of 7 the things that bothered me more than anything else 8 about the situation down ther'e.
9 But specifically, they had the problem. They 10 probably contributed to half of these failures rates 11 themselves. And I just felt something should have been 12 done a lot sooner. And tha t is why I disagreed with the 13 report when I was at INPO, because they were basically 14 giving the systems a black eye when I didn't think it 15 was justified.
16 JUDGE BRENNERa Well, the question arises, why 17 should we place so much reliance upon the efficacy of 18 procedures of factoring in experience if you believe it 19 was ignored at least in part, at an operating reactor?
20 WITNESS MC GUIRE: I can't speak for Georgia 21 Power and why they took the course of action that they 22 did take.
23 JUDGE BRENNER: Then how do we know whether or lhI 24 not those pitf alls will be avoided at Shoreham?
25 WITNESS MC GUIRE: I think you should ask Long ALDERSON REPORTING COMPANY. INC, 400 VIRGINIA AVE., S.W., WASHINGTON D.C. 20024 (202) 554-2345
5613
() 1 Island Lighting Company what their position would be on 2 that, Judge.
3 WITNESS KASCSAKs Well, I think a certain 4 amount of feeling of credibility can be given to the 5 fact tha t we have already taken actions to address this 6 particular probles even before the Georgia Power 7 organization who had the problem. So obviously, we
~
8 would not only be addressing concerns at our own plant, 9 but we have been fed back information from other plants 10 and have taken that into account.
11 I do not think it is proper to necessarily 12 damn the system simply because some applications of that 13 information was not applied properly.
14 I also think that the atmosphere today and the 15 scrutibility and the review of how utilities implement 18 these types of recommendations and input operating 17 experience back into their programs is something that is 18 reviewed by organizations like INPO and reviewed by NRC 19 personnel that come to the site.
20 So we are not certainly left up to our own 21 means to do this, although we have a strong commitment 22 to do it, and I think our actions in almost every 23 instance we have brought up in the last couple of weeks
() 24 where operational feedback has indicated a problem, 25 actions have been taken at Shoreham and which I think O
ALDERSON REPORTING COMPANY. INC.
400 VIRGINIA AVE.,3.W., WASHINGTON. D.C. 20024 (202) 554 2345
r 5614
() 1 should giva you confidence that the system is being 2 applied and will be applied in the future.
3 JUDGE BRENNERa Has anybody, whether it be GE 4 or an owners group or INPO or an organization like the 5 ISEG for other utilities, examined what went wrong with 6 respect to procedures at Hatch to determine specifically 7 why precursors during the startup phase of this event 8 were not s51vad befora later events occurred?
9 WITNESS MC GUIRE: I think if you read the l
10 INPO report, Judge Brenner, it talked about cold l
11 startups and hot startups of the systems. And I think 12 Georgia Power was convinced that the system would
! 13 function properly, because they successfully started the O 14 machine when it was hot. And I think these failures 15 here were, as the report states, were during cold 16 con ditio n s. And why that happened, I don't know. I 17 sean like I said, I am just speculating.
18 BY MR. LANPHER: (Resuming) 19 Q Mr. McGuire, if you couln turn your attention 20 to page 4-2 of the INPO report we have been looking at.
21 A (WITNESS MC GUIRE) Yes, sir.
22 0 Looking at the bottor, the last sentence on 23 that page, is it not true that cold quick-start testing
( 24 had not been a common industry practice at the time this 25 report was prepared? s l C^)
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345 l
5615 l
() 1 A (WITNESS MC GUIRE) Yes, it is common industry 2 practice. But, for example, every time we at Pilgrim, 3 for example, came back on line from a refueling outages, 4 <e would test HPCI and RCIC at 150 pounds, and that is S cold condition. So when I use the word " common," I do 6 not know where they got that information from.
7 But like I said, this system layout at Hatch 8 v's a the problem specifically. And it was a unique 9 ssystem layout, which does not -- I don't think you can to apply that to Shoreham. ,
11 Q Mr. Robsre, yesterday I believe you testified
/
12 in response to Judge Jordan, that in order to have a 13 complete f ailure of the water level system in the sense O 14 of actusting protective system channels,.you would need, 15 in essence, a double LOCAs. you would need a failure on 16 one side and then a second' failure on the other. Do .,
17 yoyu recall that testimony? < ;
18 A (WITNESS ROBARE) Yes. .
19 Q Is it not true that a single LOCA on one side 20 could cause a heatup of the drywell which could cause 21 flashing in the nonbroken reference leg?
22 (Panel of witnesses conferred.)
23 A (WITNESS DAWE) I have not personally analyzed. ~
24 that. However, just speaking from knowledge of the tze
- li 25 of the containment and what I believe the scenario would
!! ~ ,o# 't
~
,1
/'
"' a ALDERSON REPORTING COMPANY,INC, I
400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345 t/ .-
5616
() 1 be, if you broke an instrument line, one side reference 2 leg, inside containment, long before you heated up the 3 containment enough to start flashing the leg you would 4 get the high drywell pressure scram at 2 pounds, you 5 would scras the plant and you would start your 6 initiation of ECC3.
( 7 This would be no different than any small 8 steam line break inside of containment. In fact, it is 9 probably less sev'ere than most small steam line breaks to inside containment because it is restricted by the 11 condensing chamber as well. This would just be a
, 12 small-break LOCA.
i 13 In fact, this kind of a scenario is far less 14 sever than a break of a steam line outside because you 15 have all of the rest of the protection system and its 16 other input signals to do the job for yous namely, the 17 drywell prassure bncking up the vessel water level.
18 Your ECCS will starts that is well within the design 19 basis of the plant.
l 20 Remember, the flashing of line inside the 21 containment takes a high temperature and a low vessel 22 pressure. And you are just not going to depressurize l
23 the vessel through an instrument line before you do
() 24 e ve ry thing else.
25 This is well bounded by the analysis that we ALDERSON REPORTING COMFANY,INC, i 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5617
() 1 have done for boil-off. And remember, that one takes 2 all kinds of time. No, sir, I don't think that the way 3 you characterize that would be correct at all.
4 0 That is your judgment? You have not performed 5 an analysis of that; is that correct?
6 A (WITNESS DAWE) I personally have not 7 performed that analysis. However, I think the analysis 8 that has been performed well bounds that case.
9 I think Mr. Ianni may want to add to that.
10 A (WITNESS IANNI) Such an analysis was done on 11 the BWR specifically for small-small-small breaks. And 12 you get high dryvell pressures rather quickly in the 13 event, because you do pass air into th e wetvell, and you 14 get high drywell pressure trip.
15 I personally know of this analysis. In fact, 16 I did something like that on Dresden 2.
17 (Counsel for Suffolk County conferred.)
18 0 Gentlemen, yesterday on redirect by Mr. Ellis 19 you were asked questions about the 1979 review, and you 20 described what you had done, and you concluded by'saying 21 you had also looked at the RHR system during this 1979 22 review. Do you recall that testimony?
23 A (WITNESS IANNI) Yes, si r.
() 24 0 Can you provide further details, Hr. Ianni, 25 about what you did in terms of looking at the RHR O
ALCERSON REPORTING COMPANY,INC, 400 VIRGINtA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5618
() 1 system?
2 A (WITNESS IANNI) We did the same things that 3 we did to the other systems. We looked at the 4 components within the system and checked them to see if 5 they were properly classified within the RHR system.
6 An1 then wa put them on the other list with the 7 components,and cross-checked them against the other 8 plants on the matrix to see if they were consistent.
9 The same kind of thing that we did for the ECCS 10 systems.
11 The RHR components also show up in the parts 12 list, so this was done part and parcel with the study.
13 0 Did you also go through the Shoreham plant l 14 itself to examine the RHR system? I believe you stated 15 with respect to the other systems you actually went 16 through the plant with a list of components as a guide 17 and checked them to make sure that -- I am not sure if -
l 18 you used the word " review of as-built drawings" -- but 19 you did a review of where they were in the plant, I 20 believe?
21 A (WITNESS IANNI) I am sorry, when I said that 22 ve went through each plant, I was thinking in terms of 23 ve went through each plant's parts list and drawings.
() 24 We did not physically walk through the plant. But 25 Shoreham was looked at specifically in the same manner.
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5619 l
() 1 0 Thank you for that :larifiestion.
2 Mr. Robare, yesterday you testified with 3 respect to the quality-assurance testimony that you 4 provided at pages 42 and 43 of your prefiled testimony.
5 I believe you stated that over the last weekend you
' 6 talked with persons to find out what quality standards 7 have been applied for the control rod drive system; 8 correct?
9 A (WITNESS ROBARE) That is correct.
10 0 Is there a specific list or document which 11 describes that?
12 A (WITNESS ROBARE) Yes, there is.
13 0 Is that in the FSAR?
O 14 A (WITNESS ROBARE) No. It is the QAR-2 plan 15 that I ref arence in my testimony. And the requirements 16 are generally shown on page 43.
17 0 Is that QAR-2 plan specifically applicable by 18 its terms to this control rod drive system?
19 A (WITNESS ROBARE) Yes, it is. It is generally 20 applicable to all nonsafety-grade equipment.
21 (Panel of witnesses conferred.)
22 A (WITNESS ROBARE) I might have missed -- let 23 se try one more time. I think I understand. There is a
() 24 purchase specification for the CRD p' ump. And basically 25 that generally describes the design requirements and the O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON D.C. 20024 (202) 554-2345
! 5620
() 1 quality-assurance requirements and the code requirements 2 that forms a part of the purchase order documentation, 3 that purchase specification. Referenced in the purchase 4 specification is the quality-assurance requirements. In 5 the case of this non-safety-grade component, it is the 6 QAR-2 requirements.
7 Does that answer your question?
8 0 Ihen is it your testimony that all 9 non-safety-grade equipment or non-safety-related 10 equipment gets 3AR-2 for precisely the same degree of 11 quality assurance?
12 A (WITNESS ROBARE) No, it is not.
13 0 Then I am confused. And I am not trying to be O 14 tricky on that. But you used the term "non-safety-grade 15 getting QAR-2." So if you can clarify that, I would 18 appreciate it.
17 A (WITNESS ROBARE) If I could characterize the 18 non-safety-grade QA program, compared with the 19 safety-grado program with respect to Appendix B, all 18 20 points of Appendix B are addressed in both programs.
21 The degree of application of those points would vary 22 depending upon the importance of that component to plant 23 reliability and any safety significance.
() 24 And what that means as a practical matter, if 25 the component is not significant for plant operation, O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D C. 20024 (202) 554 2345
f 5621 g 1 then the quality assurance and design engineer would 2 apply a lesser degree of some of the 18 points; for 3 instance, vendor inspections and audits, design 4 calculation backups would be less for the 5 non-safety-grade equipment.
6 And the records, especially the material 7 traceability records, would not be permanently stored 8 f or the non-saf ety-grade equipment. However, those 9 records are required by the General Electric program to 10 assure ourselves that the proper materials were 11 applied.
12 0 Thus, when you state that QAR-2 is applied to 13 a particular item, a non-safety-related item, in fact, I 14 that can represent a wide range of quality assurance, 15 quality control with respect to different items?
16 I mean for one item it means perhaps almost 17 all of Appendix B, for another item it may mean very 18 little of Appendix B depending upon the judgment of the 19 design engineer and the quality control engineers who 20 determine what, in fact, to apply to that item? Is that 21 correct?
22 A (WITNESS ROBARE) That is correct. And 23 Appendix B itself does not define that degree, even for I 24 the safety-related equipment.
25 (Panel of witnesses conferred.)
k ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5622
() 1 A (WITNESS ROBARE) I would also like to 2 supplement that, that because of a practial matter, 3 because of the scope of supply of the General Electric 4 versus the balance-of-the-plant, most of the General 5 Electric systems are related to safety and high plant 6 reliability. So is a practical matter, most of the
! 7 General Electric-supplied equipment receives a quite 8 high degree of quality assurance.
9 (Panel of witnesses conferred.)
10 JUDGE BRENNER: I do not know who is waiting 11 for whom, but I want to keep moving.
12 MR. LANPHER I apologize. I thought the 13 witnesses were conferring.
14 JUDGE BRENNEBs They were, and that is why I 15 expressed it the way I did. I just wanted to get a 16 sta tus check here.
17 WITNESS ROBARE: I have no further 18 clarification.
19 BY NR. LANPHER4 (Resuming) 20 0 The knowledge which you gave yesterday 21 reg a rding the control rod drive system, I believe there 22 are two other systems: the shroud head and separator, 23 and the steam dryer assembly. That whs knowledge which
() 24 you gained from persons in the GE quality assurance 25 department?
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5623
() 1 A (WITNESS ROBARE) Yes. I met with them 2 specifically last weekend on those componen ts.
3 0 And that was not information that you 4 specifically had prior to that time?
5 A (WITNESS ROBARE) That is correct. And as I 6 stated yesterday, it is not the type of information that 7 anyone would have in their head for numerous components 8 regardless of their discipline.
9 0 You were told that by the GE QA people; is 10 that correct?
11 A (WITNESS ROBARE) That'is correct. When I 12 asked them this question, the only way to answer it is 13 to delve into nunarous records and make a point-by-point O 14 determination.
15 0 GE does not maintain a list or master document 16 for non-ssfety-related equipment which documents what 17 degree of quality assurance is applied to each l 18 component?
19 A (WITNESS ROB ARE) No, there is not such a 20 list. The degree of application is applied commensurate 21 with the importance of those components to reliable and 22 safe operation of the plant.
23 0 But if someone wanted to determine that degree
) 24 of application, you would have to go through the records 25 as to each component to find out tne degree of quality l
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5624 l () 1 assurance which was spplied?
2 A (WITNESS ROBARE) That is correct.
3 MR. LANPHEH4 That completes my questions, 4 Judge Brenner. I believe Ms. Est, sche has a few.
5 BY MS. LETSCHEs (Resuming) 6 Q M r. Xascsak, I believe you stated yesterday in 7 response to a question from Mr. Ellis that Mr.
8 Yoksimovich had been retained as a consultant to further 9 sensitize the ISEG people on this systems interaction to issue. Can you tell me when he was retained for that 11 purpose?
12 A (WITNESS KASCSAK) Well, the specific 13 conversation we had relative to his expertise in this O 14 area and his use in conferring with the ISEG group and 15 the further involvement they would have in that program 16 was a number of weeks ago prior to this testimony being 17 given. He is a consultant to the company and has been 18 for a while.
19 Q In response to some questions from Mr. Ellis 20 about the startup and pre-op testing program and the 21 description of that in your testimony, you discussed how 22 thst was going to be able to detect adverse systems 23 interactions. Are some of the startup and pre-op tests
() 24 that are performed not single system tests?
25 A (WITNESS KASCSAK) Well, as I think I have O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5625
() 1 ind ica ted , there is a whole spectrum of tests, varying 2 from single component tests through individual tests 3 through integrated system tests where all of the systems 4 in the plant are tested based upon given transients and 5 initiating events.
6 Q It would be difficult to detect a systems 7 interaction in a single component or a single system 8 test, would it not?
9 A (WITNESS KASCSAK) No, because in many of to these systems there are support systems that are 11 involved in the actual operation of that system, and the 12 proper support of those other systems to the functioning 13 of that system would be detected through that individual 14 performance test; systems like lubrication systems or 15 cooling systems or air systems or power systems, for 16 that matter. All of these, obviously, each system needs 17 many support systems to function properly.
18 Q But in a single system test you would not be 19 able to detect a degraded performance in a functionally 20 effected system or a spatial *.y related system if it was 21 not one of these support systems you just mentioned; is 22 that not right?
23 A (WITNESS KASCSAK) Obviously, if a system is
() 24 not tested during that test, you cannot discover whether 25 that system has any function. Yes, I agree with you, I O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5626
(} 1 guess.
2 0 Mr. Dave, with reference to the cable 3 separation study -- that is, Item D in your prefiled O 4 testimony -- as it is described in your testimony, that 5 was not an analysis of cable separation performed to 6 demonstrate that i fire that disabled the cables and 7 raceways and associated equipment in particular areas 8 would not prevent safe shutdown assuming concurrent loss 9 of off-site power; is that not right?
10 A (WITNESS DAWE) Was an analysis, I think is
~
11 what you said, as oppposed to wasn't an analysis, which 12 I am not sure whether that's what you said?
13 0 No. I said it was an analysis.
14 A (WITNESS DAWE) Yes, it was an analysis to 15 demonstrata that disabling of all components associated 16 with cable in the fire area would not preclude safe 17 shutdown of the plant assuming additionally the loss of 18 off-site power.
i 19 0 Thank you.
20 MS. LETSCHE4 Judge Brenner, I do have some 21 additional follow-up arising out of Mr. Ellis' questions 22 concerning the Literick PRA, but I think it would be 23 appropriate to follow up on those when we get into the
() 24 PRA cross examination.
25 JUDGE BRENNER: Okay. Who would there be
. ()
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE, S.W., WASHINGTON D.C. 20024 (202) 554 2345
5627 )
l
() 1 questions of, Dr. Burns or Dr. Joksimovich?
2 MS. LETSCHEs I think they would probably 3 involve questions of the PRA panel, including Mr.
4 Kasacsak, and I would assume at least Mr. Burns might 5 have some pertinent information.
6 JUDGE BRENNER: I guess I did not realize Mr.
7 Kascsak was going to be on that panel. Is he?
8 MR. ELLISa Yes, sir, he is.
9 JUDGE BRENNERa All right, Staff?
10 MR. RAW 53N Staff has no further questions,
'11 Judge. Thank you.
12 JUDGE BRENNER: Any further redirect?
13 MR. ELLIS Judge, just a couple to clarify 14 one thing.
15 FURTHER REDIRECT EXAMINATION 16 BY MR. ELLIS:
17 0 Mr. Robare, as I understand the QA program, 18 the non-safety-related systems and components at GE get 19 QAR-2 as described in your prefiled testimony?
20 A (WITNESS ROBARE) I think I had better clarify l
21 that. The non-safety systems and components covered 22 under the QA program get QAR-2 type requirements. That 23 particular designation is not used throughout the
() 24 General Electric manufacture or make or by departments.
25 However, the points in my prefiled testimony ALDERSON REPORTING COMPANY. INC.
400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5628
() 1 on page 43 are the equivalent of QAR-2, and those points 2 are utilized for all aspects of the GE supply.
3 0 Now, you answered a question f rom Mr. lanpher 4 by indicating that you would have to refer to documents 5 for non-safety-related equipment as you did for the CRD 6 pumps to determine the level of each of the points that 7 a re part of the QAR-2 program. Do you recall that?
8 A ( WITNESS ROB ARE) Yes, that is correct.
9 0 Now, you would also have to look at documents 10 to see the levels for those points under Appendix B for 11 safety-related, would you not?
i 12 A (WITNESS ROBARE) That is correct.
13 0 So in determining the level of quality N
14 assurance applied in connection with the 18 points under 15 Appendix 9, and also the points raised on page 43 of 16 your prefiled testimony, there is an engineering 17 judgment made as to the level commensurate w'ith the 18 function of the system or component; is that correct?
19 A (WITNESS ROBARE) That is righ t, regardless of 20 whether it is saf ety-grade or not safety-grade.
21 Q Now, you indicated that you consulted with 22 persons this weekend. Did you also consult with 0A 23 persons in connection with the preparation of your l
() 24 prefiled testimony?
25 A (WITNESS ROBARE) Yes, I did.
l O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5629
() 1 MR. ELLIS I have no further questions.
We have no further questions, 2 JUDGE BRENNER:
3 unless there are sny questions stimulated by these last 4 questions.
5 (No response.)
6 JUDGE BRENNER. Well, we are finally at tha t 7 point, gentlemen.
8 (Laughter.)
9 JUDGE BRENNER I was going to thank you all, 10 which I will do. It has been s long 2 weeks, and we 11 appreciate that it gets harder as the week gets longer, 12 and it is such essier from this end of the table than 13 from your end.
14 I guess I will have to point out now that I 15 have learned, it was reminded that Mr. Kas:sak is going 16 to be here this afternoon, you are not going to get the 17 same break as the rest of the panel is, as I understand 18 it, unfortunately. But we do thank you very much for 19 your time and effort.
20 Mr. Robsre, do you have something to add?
21 WITNESS ROBARE: Yes, I just have one 22 follow-up to Judge Carpenter's question about one 23 particular report. It is on page 63 of the prefiled
() 24 testimony, the report having to do with the reactor 25 protection -- or the protection systems compliance with O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
l 5630 l
l l 1 IEEE-279.
2 That report was submitted to the NRC for 3 generic review at the time and, in that sense, has 4 received peer review. It is also referenced in the 5 FSARs.
6 JUDGE CARPENTER: Thank you.
7 MR. ELLIS: Judge Carpenter, or Mr. Robare, we 8 answered all of those. Have we now answered all of the 9 questions you had on those peer review, Judge 10 Carpenter?
11 JUDGE CARPENTER: Yes, you have.
12 MR. ELLIS: Thank you.
13 JUDGE BRENNERs The panel is excused.
14 (Witnesses excused.)
15 JUDGE BRENNERs After the lunch break, we will 16 come back with the PRA panel and begin with the County's 17 cross examination of that panel.
18 Let us stay on the record for a moment. Do we 19 have to meet off the record?
20 MR. LANPHER: We consulted and passed some 21 information back and forth. But we did not think that 22 there was any need to involve you all in it.
23 JUDGE BRENNER Fine. In that case, we will O 24 take e tunch dreak unt11 2 20.
25 (Whereupon, at 12 :0 5 p . m . , the hearing was O
ALDERSON REPORTING COMPANY, INC, 400 VIRGINIA AVE S.W WASHINGTON, D.C. 20024 (202) 554 2345
5631 h 1 recessed, to reconvene at 1:20 p.m., this same day.)
2 O 4 5
6 7
8 9
10 11 12 13 O 1.
15 18 17 18 19 20 21 22 23 O 24 25 0
ALDERSON REPORTING COMPANY,INC.
400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554 2345
5632 O an un991 m 1 2 (1s20 p.m.)
3 JUDGE BRENNER: Back on the record.
4 I guess we should swear two of the witnesses 5 in, and then we will allow you to introduca them.
6 MR. EARLEY: Judge, I think they were all 7 sworn in.
8 JUDGE BRENNER: Yes. Thank you. That was 9 some time sco. I recall the group overlapping the table 10 now.
11 Dr. Morris suggests there is some value in 12 redundancy, but we vill not carry that to a fetish in 13 this instance.
14 (Laughter.)
15 JUDGE BRENNER: Let us proceed.
16 Whereupon, 17 ROBERT M. KASCSAK, 18 the witness on the stand at the time of recess, resumed 19 the stand and was further examined and testified as 20 follows:
21 Whereupon, 22 EDWARD T. BURNS and 23 YOJIN J0KSIMOVICH O 24 were reca11.d as witnesses ane, havin, prew1ous1r seen 25 duly sworn, were further examined and testified as O '
ALDERSON REPORTING COMPANY,INC.
400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202)*.54-2345
m 5633 =
1 follows: --
2 CROSS EXAMINATION 3 BY MS. LETSCHE: -
4 Q Sentlemen, I would just like to go over what _
5 each of your participations were with respect to the __
m 6 Shoreham PRA.
7 Dr. Joksimovich, as I understand it, you are 8 on the peer review group that is performing the pee r 9 review of the Shoreham PRA; is that correct?
10 A (WITNESS J0KSIMOVICH) That is correct.
11 Together with Professor Rasmussen and Dr. Walton 12 Rodgers. ,
13 0 That review process is still under way; is 9 14 that correct? -
t 15 A (WITNESS J0KSIMOVICH) That is correct. We --
16 have five meetings so far, and the process is going to 17' continue until probably the year end.
18 0 And, Mr. Burns, I understand you were the lead -
19 analyst for SAI on the Shoreham PRA, is that correct? -
20 A (WITNESS BURNS) Yes, that's true. _
21 0 And I understand from looking at your resume ,;
22 that you were also the lead analyst on the Limerick PRA; 23 is that correct?
h 24 A (WITNESS BURNS) Yes, that is also true. ,
25 Q Rnd, Mr. Kascsak, I know you mentioned earlier O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE, S.W., WASHINGTON, D.C. 20024 (202) 554-2345 -
m 5634 h 1 that you had some involvement with the Shoreham PRA.
2 Perhaps you can tell us precisely what your involvement 3 was. _
4 A (WITNESS KASCSAK) Well, I think, as my 5 testimony states, my division -- well, the nuclear 6 engineering department, of which I am the manager of one 7 of the divisions, was the organization thst was involved 8 in an initiating the PRA and defining the scope and the 9 objectives of the program and prepared the procurement to documents which culminated in the award of contracts to 11 both SAI and PLEG.
12 And we have performed the role as project i
13 manager, in a sense, of that program and have ensured I
14 tha t the right technical information has been supplied 15 to these contractors and have worked with GE and Stone 16 C Webster in that vein and also performed a review ,
17 function in terms of reviewing the documents that were 18 generated by the contractors to ensure completeness and 19 accuracy in teras of the design of the plant.
20 0 Is your group also involved in reviewing the 21 PRA itself?
22 A (WITNESS KASCSAK) Yes, they are. -
23 Q Mr. Kascsak, I would like to direct your I 24 attention to page 120 of your prefiled testimony, ,in 25 which you state the initial goals of the Shoreham PRA _
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
l 5635 l
() 1 program. About midway through that first paragraph you 2 state that, and then you enumerate three goals.
3 My first question is are there any subsequent i 4 or different goals other than the three that you 5 indicate are the initial goals?
6 A (WITNESS KASCSAK) I think those are fairly 7 broad goals, and I would say that they enumerate quite 8 well still what is considered the present goals of the 9 program.
10 0 Ihat is to assess the Shoreham emergency plan, 11 to perform design verification, and to develop 12 reliability risk snalysis capability within LILCO?
13 A (WITNESS KASCSAK) Well, I think you I 14 summarized a little bit quickly and you lef t a few words 15 out here and there. But I would still stand behind the 16 words in the testimony.
17 (Counsel for Suffolk County conferred.)
18 Q Hr. Burns, I would like to direct your 19 attention to page 87 of your portion of the testimony.
20 And under the subheading 2.1 PRA Methodology, you state, 21 "The probabilistic risk assessment performed for the l 22 Shoreham piant provides basic data for a utility risk 23 management program." Can you tell me what that term
() 24 means, " utility risk management prog ra m " ?
l 25 A (WITNESS BURNS) I can tell you my l
l ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W4 WASHINGTON. D.C. 20024 (202) 554-2345
5636
() 1 interpretation of that. And I think it fits in with the 2 gosla that Shorehsm has in mind. But my interpretation 3 of " utility risk management" is a program whereby we 4 search for risk reduction; we try to identify if there 5 a re , in fact, significant risk outlyers or if there are 6 items that can be affected to improve safety.
7 Of course, the balance of the program has to 8 be the cost effectiveness versus the effect on safety.
9 But this is one part of that program.
10 A (WITNESS J0KSIMOVICH) Hay I add to that, it 11 is a relatively new term, and it also means maintenance 12 of the asses sed risks throughout the lifetime of the 13 plant. In other words, Shoreham is nearing completion, 14 and the plant is going to, hopefully, operate for 30 or 15 40 years,and there will be all sorts of changes which 16 ara going to be introduced at the plant. And the iden 17 is to use the PRA study as a living tool to update it 18 and keep it current, so that the planned risk as 19 assessed is valid on day one and on day -- I don't know 20 -- two, three, four thousand, et cetera.
21 Q Mr. Burns, you mentioned one of the purposes 22 was to identify risk outlyers. Can you tell us what you 23 mean by that term?
() 24 A (WITNESS BURNS) Whether there is any sequence 25 of events or identified safety concern that would be O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5637
() 1 estimated to be of much higher frequency than other 2 risks that may be evaluated.
3 0 Now, the SAI obligation or responsibility in 4 this program is to develop the probabilities of such 5 events occurring; is that right? That is at least one 6 of the responsibilities of SAI?
7 A (WITNESS BURNS) Task 1 or phase 1 of the 8 program is to identify frequencies of accident 9 sequences, yes.
10 0 And then phase 2 is to develop the releases 11 that would be associated, or the assumed releases that 12 would be associated with the events that you have 13 analyzed ; is that correct?
O 14 A (WITNESS BURNS) Phase 2 develops the releases 15 and the in-plsnt conditions that would occur or be 16 associated with certain accident sequences.
17 Q SAI is not involved in the evaluation of the 18 risk outlyers or the sequence of events or the assumed 19 released data in terms of its safety implications or its 20 design implications; is that not right? '
21 A (WITNESS BURNS) SAI must interface with 22 Pickard, Love & Gstrick, who is the contractor who is 23 actually running the ex-plant consequence evaluation in
() 24 terms of we must supply to them as input for their 25 calculations both the frequencies of identified accident O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE, S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5638
() 1 sequences and the radionuclide source terms and other 2 characteristics of the sccidents which include such 3 things as time to ration of release.
4 Now, when you say that that doesn 't have any 5 safety implication, in fact, there are several measures 6 of safety that have been used in the industry, although 7 there is at the present time, at least to my 8 understanding, there is no consensus of what safety 9 goals may be.
10 There is a number of terms that are used in 11 safety goals, and one of those is core-melt frequency, 12 and so, of course, we have a direct input to that. And
'13 ay own personal judgment is that that is a very 9 14 importan t goal . And strictly based on that goal or 15 strictly based on that measure of safety or risk, one 16 can isolate outlyers.
17 In addition, the source term evaluation that 18 is being done also identifies what the potential 19 radionuclide source term wo uld be . And based only on 20 that source term, one can identify whether that sequence 21 of events that would have produced that would be 22 considered to be important.
23 Then a step further would be coupling the 24 accident sequence frequency with the source term as 25 still another measure of potential safety implication to O
ALDERSON REPORTING COMPANY. INC.
400 VIRGINIA AVE S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5639 1
() 1 the public.
2 Then the next step, of course, with PLEG, they 3 would produce even further comparisons that could be 4 made to identify safety goals.
5 0 I guess the Shoreham PRA, as I understand from 6 your testimony, is intended to provide information to 7 LILCO regarding probabilities of events and associated 8 releases related to those from which LILCO could pursue 9 tha gosls that Mr. Kas:sak has stated in his testimony 10 with respect to using that information, is that right?
11 MR. ELLIS: I object to that question. I 12 think his testimony runs about 20 pages, awnd for her to 13 characterize it in two sentences seems to me to be O 14 unfair and inappropriate.
15 JUDGE BRENNERs Well, I think the witness can 16 indicate tha t to a reasonable extent. Now, if the 17 question cannot be answered, you can indicate that also, 18 and, if possible, attempt to say why it cannot be 19 answered. But let us give the witness a chance first.
20 WITNESS BURNS: I think, if I can clarify my 21 understanding of your question, it was were the 22 frequencies of accident sequences ;ind the soe ce terms 23 one of the results of the study? And I think, yes, that
() 24 is t. rue.
25 JUDGE CARPENTERS Ms. Letsche, Dr. Burns, O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5640
) 1 would you mini moving the microphone a little bit over 2 to your right?
3 WITNESS BURNS: Closer?
4 JUDGE CARPENTER: Not closer, but you are 5 looking in that direction, and I am having trouble 6 hearing you. I just want to hear every word you say.
7 (Discussion off the record.)
8 BY MS. LETSCHEs (Resuming) 9 0 Mr. Burns, are there any -- well, let me start 10 again. Have both phase 1 and phase 2 of the SAI portion 11 of this project been completed?
12 A (WITNESS BURNS) We submitted a draft for the 13 completion of phase 1 and phase 2. And so from that 14 standpoint, our initial calculations have been 15 completed. There is some ongoing work to refine some of 16 the points. But, in general, it is being completed, and 17 some comments frot both LILCO and the peer review 18 committee are yet to be incorporated.
19 0 The preliminary submission that you have made, 20 is that the one that is dated March 1982?
21 A (WITNESS BURNS) Yes, it is.
22 0 The refinements that you mentional th a t a re 23 still going on, I gather from your testimony that you do 24 not expect the results or the conclusions to change in 25 light of these refinements? I direct your attention to O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
l l
5641 l
() 1 page 107 of your testimony.
2 A (WITNESS BURNS) I am sorry, did you want me 3 to read the entire section?
4 0 I might not have asked a question just now, 5 and that might be your difficulty. Is it correct that 6 the refinements that you mentioned in your prior answer 7 will not enange the results or conclusions of the March 8 1982 draft, as you stated in your prefiled testimony?
9 A (WITNESS BURNS) I can't prejudge the answer 10 to that. But in general, the items that we have 11 identified thus f a r -- well, there are some -- in 12 general, they teni to make the assessment more 13 realistic. In other words, we have done some things 14 that may be considered overly conservative in the 15 original draft. However, on the other hand, there are to things that we are looking at that would add additional 17 sequences to what would be considered the dominant 18 contributors to risk.
19 So there are things going both ways. But in 20 general, I wouldn't say that they would change the 21 qualitative conclusions which I stated previously which 22 was in the original testimony.
23 0 Well, do you anticipate that the probabilities
() 24 that are stated in the results section of the draf t you 25 submitted, the probabilities with respect to the O
V ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5642
() 1 frequency of events occurring and core-vulnerable 2 conditions, do you anticipate that those are going to 3 change?
4 A (WITNESS BURNS) For the items that have been 5 currently identified to yet be incorporatei into the 6 study, I would estimate that those numbers are good to 7 within 10 to 15 percent.
8 JUDGE BRENNER: 10 to 15 percent, you said?
9 WITNESS BURNS: Yes, sir.
10 BY MS. LETSCHE: (Resuming) 11 Q Mr. Kasesak, when is it anticipated that phase 12 3 of the PBA vill be completed?
13 A (WITNESS KASCSAK) As I have indicated in my 14 testimony, we anticipate we are working towards 15 completing phase 3 by the end of September of this 16 year.
17 0 Have you received a draft of a report from 18 Pickard, Love C Garrick on the phase 3 study?
19 A (WITNESS KASCSAK) Yes, we have received draft 20 sections of tha t phase.
i 21 (Counsel for Suffolk County conferred.)
22 15. LETSCHE Judge Brenner, that particular j
23 -- well, never mind.
() 24 BY MS. LETSCHE: (Resuming) 25 0 Mr. Kasesak, when did you see tha t draft?
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554-2345
i -
a 5643 l
() 1 A (WITNESS KASCSAKh / I would have to refresh my 2 sesory, but a copy of that draf t was given to the Cou'nty 3 when we supplied the entire package.
O 4 scounsel for Suffolk County conferred.)
5 0 Do you snticipate that there will be 6 substantial changes from the draf t that you received of 7 phase 37 Or have you received any information to that 8 effect f rom Pickard, Lowe C Garrick?
/
9 MR. ELLISa Information to what effect?
10 MS. LETSCHE: Let me clarify my question,.
11 BY MS. LETSCHEt (Resuul'ng) ,-
12 0 Do you know whether Pickard, Love C Garrick 13 anticipa tes major revisions to be made in the draf t that O 14 you have received from them-on' the ph,ase 3 portion?
. -n 15 MR. ELLISa May I interpose;here an obj(ction, 16 Judge BEenner? There is no. testimony on phase 3, as I 17 understand it, in this Contention.
18 JUDGE BRENNER: Sure there is. We just had 19 questions about it. He testified not just today but, as 20 Mr. Kascsak probably indicated previously, that he was 21 looking towards the end of September as his estimated 22 date, not s guaranteed date, of course, necessarily.
23 And I guess I do not understand the
() 24 objection. Am I missing something, Mr. Ellis?
25 MR. ELLISa Well, I thought Ms. Letsche was I
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D C. 20024 (202) 554 2345
5644
() 1 going to examine these witnesses on methodology of phase 2 3. And perhaps I spoke too soon.
3 JUDGE BRENNERs Do you remember the question, d Mr. Kascsak?
5 WITNESS KASCSAKs The question I remember was l -6 whether or not it was my opinion that Pickard, Lowe C 1
7 Garrick believed there would be any significant changes 8 in the draft that we have been given. My response to
~
9 that would be that the draft is under review and we have
~10 no reason to believe that there would be significant
'11 changes, although I wouldn't say that would be precluded 12 based upon that review.
13 JUDGE B9ENNER M r. Kascsak, how about with
) 14 respect to significant changes in the methodology and 15 the application of the methodology as distinguished from 16 changes in the particular calculational results, could 17 you be more definite as to the former two categories as 18 distinguished f rom the possible changes in the 19 calculations or the results of those calculations?
20 I am trying to get a feel for where the 21 current review is focused in what areas and what areas 22 sight be more likely than other areas to undet s *hange 23 as a result of the current ongoing review.
() 24 WITNESS KASCSAKs I would characterize the 25 review of one of more applying methodology and the O
ALDERSON REPORTING COMPANY,INC, l 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
l 5645
() 1 Shorehsm specificity of that in the review and that the 2 methodology is far less likely to be subject to that 3 review.
4 (Counsel for Suffolk County conferred.)
5 BY MS. LETSCHEa (Resuming) 6 Q Mr. Burns, I would like to direct your 7 a ttention to page 97 of your prefiled testimony, the 8 paragraph that begins near the bottom of that page where 9 "SAI judges that f ault-tree / event-tree methodology is 10 the best available technique for augmenting the existing 11 deterministic evaluations and NRC regulations," and then 12 it goes on. Well, it goes on to say, "To ensure that 13 systems interactions are exposed and potential areas of 14 concern are identified."
15 I take it you would agree with me that there 16 are a number of methods, a number of techniques that 17 could be used to identify systems interactions. Is that 18 correct?
19 (Pause.)
20 A (WITNESS BURNS) Yes, there are a number of 21 techniques that can be used.
22 0 And would you also agree with me that the 23 various techniques have strengths and weaknesses with
() 24 respect to their ability to identify different types of 25 systems interactions?
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W WASHINGTON, D.C. 20024 (202) 554 2345
5646 (WIINESS BURNS) Yes.
(]) 1 A 2 (Counsel f or Suff olk County confe rred. )
3 0 Mr. Burns, do you have up there a copy of what 4 has been marked as Suffolk County Exhibit 19 for 5 identification?
6 A (WITNESS J0KSIMOVICH) I do.
7 0 I would like to direct your attention to page 8 53, the ps7e that is marked 53 in Suffolk County Exhibit 9 19. It contains Table 4.1, identified as a " Summary of 10 Systems Interaction Evalua tion Methods Review."
11 First of all, let me ask you, Mr. Burns, are 12 you f amiliar with the study performed by Lawrence l 13 Livernora Laboratory from which Suffolk County Exhibit 14 19 is taken?
15 A (WITNESS BURNS) I have read much of this 16 report. So to that extent, yes.
17 18 1
19 20 21 22 23 24 25 O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5647 ggg 1 0 I would like you to look a t Table 4.1 and tell 2 me if you agree that this does, in fact, summarize some 3 of the weaknessas and strengths of various methods of 4 identifying systems interactions?
5 A (WITNESS BURNS) It appears to do that at my 6 initial glance, yes.
7 (Counsel f or Suff olk County conferred. )
8 0 Mr. Burns, I would like to direct your 9 a ttention to the comment " Graph-based analysis," on 10 Table 4.1. Would you agree with me that 11 fault-tree / event-tree analysis, such as is performed in 12 a PR A, would be considered a graph-based analysis as 13 Lawrence Livermore's study uses that term?
14 A (WITNESS BURNS) I believe that's true. And I 15 don 't see sny other heading it would fit into here. I 16 would have to check the terminology. Do you want me to 17 do that? I believe it's true.
18 (Counsel for Suffolk County conferred.)
19 JUDGE BRENNER: Dr. Burns, if you look at page 20 Roman XVIII of Suffolk County 19 for identification, 21 they have a little one-paragraph blurb describing 22 graph-based analyses. Were you familiar with that 23 paragraph before you gave your answer just before?
lll 24 WITNESS BURNS: That is what I was recalling, 25 res, sir. That seems to confirm, so, yes.
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S W., WASHINGTON, D.C. 20024 (202) 554-2345
5648
() 1 JUDGE BRENNERs If you want a chance to read 2 it, we will give it to you now to read that one 3 paragraph.
4 (Pause.)
5 JUDGE BRENNERa Hs. Letsche, while he is doing 6 that, I cannot fit this line in with your revised 7 handwritten cross plan. And I am wondering if you could 8 orient me.
9 MS. LETSCHE: Jud'eg Brenner, this particular to line of questions is not explicitly stated on there, but 11 it is sort of an introduction to number 3. I do not 12 intend it to be extensive.
13 ' JUDGE BRENNER: The reason I asked you to look O 14 at it, Dr. Burns, is that you were somewhat tentative in 15 confirming whether or not the fault-tree / event-tree -
16 methodology would fit under the Livermore report's use 17 of the term " graph-based analyses." And I am certainly 18 not criticizing that tentativeness, since it is not your 19 report. But it may be, for all I know, that some line 20 of questions will go off in that direction based upon 21 that belief. And if it later turns out to be wrong, we 22 vill have wasted some time, perhaps, and received some 23 incorrect information.
O)
(_ 24 WITNESS BURNS: Y es, sir. I read the
- 25 paragraph, and in their terminology " graph-based l
l (2) i ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE, S.W., WASHINGTON, D.C. 20024 (202) 554 2345
56149 O a tv=1= $= ta=1=a veat tre aa < =1e tre-2 techniques.
3 JUDGE BRENNER I infer from the answer you 4 just gave me, though, that it is not synonymous with the 5 way you used the term "f ault-tree /even t-tree 6 methodology"?
7 WITNESS BURNS Well, I think that either 8 Lawrence Livermore or other people -- in other words, 9 the NRC has had se'veral contractors look at this 10 question of systems intersction and the tern 11 " graph-based analyses," I believe, has been used by 12 Brookhaven, Battelle, Sandia, and Lawrence Livermore, to 13 be a more general classification of techniques, O 14 including fault trees and event trees.
15 JUDGE MORRIS But, Dr. Burns, would you 16 distinguish what is being talked about here ?
17 Gesph-based analysis and the tables in'eled o "SI," safety 18 interaction evaluation methods, would you distinguish 19 these from your definition of a f ull probabilistic risk 20 assessement?
21 WITNESS BURNS I think, yes, there is no 22 question that in my mind the purpose of these 23 evaluations are to only look at systems in the context 24 of typical IREP studies, systems interactions, whereas 25 generally in the industry, I think, probablistic risk O
ALDERSON REPORTING COMPANY,INC.
400 VIRGINIA AVE., S W., WASHINGTON, D.C. 20024 (202) 554-2345
5650
() 1 assessment implies that one looks at systems 2 interactions potential radionuclides or source terms 3 that may be generated and potential ex plant 4 consequencas that may occur. At least, to me it does.
S So I am isot sure I answered your question.
6 Graph-basel analyses in this context from Lawrence 7 Livermore, using their definition at the beginning of 8 the report appears to limit itself to only event trees 9 and fault trees and not talk about diagram methods.
10 JUDGE H3RRIS: I guess I was anticipating a 11 little bit what I thought Ms. Letsche was going to ask 12 you, but I had better let her ask you.
13 BY MS. LETSCHEa (Resuming) 14 Q Mr. Burns, again referring to Table 4.1, at 15 the bottom of the column headed " Graph-based analyses,"
16 there is a statement, " Weak on inherent failures and 17 human interfaces." Would you agree that as compared to i 18 -- strik e t h at .
19 Would you agree that, for instance, as 20 compared to reviews of operating experience as a method i
21 of identifying systems interaction, that a 22 fault-tree / event-tree analysis would be weak in i
j 23 identifying human interface types of dependencies?
() 24 A (WITNESS BURNS) We can't do these techniques,
[ 25 event trees or fault trees, in a vacuum. In other i
j ALDERSON REPORTING COMPANY. INC, I 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
i 5651
( 1 words, we have to -- we have to be a ware what is 2 happening in the plants and how they are designed and
() 3 what the implications of certain plant arrangements or 4 logic systems are in each individual plant.
5 The event-tree / fault-tree therefore can't tell 6 you, it can't tell you the answer. It is only a 7 frsmework within which we hope we ca logically display 8 so that people can interpret the results.
9 One of the very important inputs to that logic to disgram is, in fact, operating experience data. And so 11 if your question is do reviews of operating experience 12 afford a method of identifying human interfaces and 13 effects on accident, potential accident sequences, I 14 think it is probably the most reliable and important 15 source of input to that information.
16 A (WITNESS J0KSIMOVICH) May I add to that, I 4
17 definitely di; agree with your characterization about 18 inferiority. I think the two techniques are 19 complementary.
20 0 But I take it you gentlemen would probably 21 agree that the most effective way of identifying systems 22 interaction would be through a combination of these 23 different techniques that you discussed in your O 24 testimony, things like reviewing ope ra ting experience 25 and fault trees and event trees and walk-downs and FMEAs
]
O ALDERSON REPORTING COMPANY. INC.
400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554 2345
5652 0 1 and dependency matrices and deterministic studiee? 1s 2 that right?
3 A (WITNESS BURNS) I think as much information 4 as you can bring to bear on the problem is appropriate 5 to do so, including those which you nentioned. And 6 again, the faul-tree / event-tree technique is that: it 7 is just merely a technique for displaying that 8 information.
9 A (WITNESS J0KSIMOVICH) If I may add, I think 10 to put these things in a proper perspective, as I 11 believe Bob Kascsak mentioned the other day, there are 12 two processes that we are talking about. One is a 13 design process, and the other one is an assessment O 14 process. There are techniques which are more effective 15 in the design process than the others. And there are 16 techniques which are much more effective than the others 17 in the assessment process.
18 What I as particularly claiming is that the 19 PRA as a technique is the best technique for the 20 assessment process.
21 (Counsel for Suffolk County conferred.)
22 0 Would you agree with Mr. Joksimovich, though, 23 that the assessment process that you just described is 24 dependent on the success of the design process?
25 A (WITNESS J0KSIMOVICH) Well, there is no O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5653
() 1 question about it, if the design process is perfect 2 there is no need for assessment. But that is rarely the 3 case. And in the assessment process, one has to use a 4 reverse type of technique to assure the complete 5 success.
6 And that is why a deterministic approach that 7 has been applied in the design of your nuclear power 8 plants is extremely well supplemented by the 9 probabilistic approach because it adds another diverse 10 assessment and leaves -- if the conclusions are 11 ide n tica l, then we have the highest level of assurance 12 thst the nuclear power plants have been designed 13 successfully.
} 14 (Counsel for Suffolk County conferred.)
15 Q Mr. Burns, certain externalities, or external 16 initiating events, were excluded from the Shoreham PRA; 17 is that right?
18 A (WITNESS BURNS) Yes.
19 0 Specifically, fire, flood, sabotage, and 20 earthouake were excluded; is that right?
21 A (WITNESS BURNS) Those are earth, wind, and 22 fire. Right. Those are the major categories. We did 23 do what was judged to be potentially important to
() 24 Shoreham. We did do an evaluation or included an 25 evaluation of internal floods, which is generally O
ALDERSON REPORTING COMPANY. INC, 400 VIRGINIA AVE, S.W, WASHINGTON D.C. 20024 (202) 554-2345
5654 -
l I
i
() 1 outside the scope of what at that time was called a 2 Level 3 PRA.
3 I guess as a little ba ckground , a t the tim e 4 that the original Shoreham PR A was contracted for, the 5 published PRAs that had been done had generally 6 concluded that external events were not a dominant j 7 contributor to risk.
8 And I think LILCO's initial intent was -- its 9 initial program was based upon that, and justifiably so, 10 since I think they were looking for the areas in which 11 they could make the greatest contribution to risk 12 reduction. And at that time it certainly appeared that l 13 that was a potential interplant interactions that may
() 14 occur.
15 - But as I said, this one event, internal 16 floods, were included in addition to that. The whole 17 program is comparable to what the NRC has set out to do, 18 at least initially in the IREP program which was focused 19 on those things that WASH-1400 had identified as 20 potential major contributors to risk.
21 (Counsel for Suffolk County conferred.)
22 JUDGE BRENNER4 Dr. Burns, I am just trying to 23 get a focus on the time frame you were talking about.
24 Could you tell me when the Shoreham PRA was originally 25 contracted for?
O ALDERSON REPORTING COMPANY. INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5655
() 1 WITNESS BURNS: I believe the official 2 placement of the contract was roughly April of '81, 3 although most of the discussions occurred from roughly 4 January '80 through '81.
5 BY 55. LETSCHE: (Resuming) 6 Q Mr. Burns, were you implying in your last 7 answer that there has been a change in perception about 8 the importance of these externalities since the time 9 that LILCO contracted for the PRA?
to A (WITNESS BURNS) I think -- well, certainly, 11 there have been -- and this is really very secondhand 12 information -- there are two studies that I know of that 13 have identified the earth, vind, and fire scenarios as
() 14 potential major contributors. Those were -- however, we
- 15 have to be a little careful because those were, in fact, 16 PWRs and have been characterized as totally different 17 than BWRs.
18 The specific reasons that external events 19 there were considered dominant contributors, I guess I 20 would have to defer, but I think that they are generally l
21 much different. A lot of it has to do with the 22 balance-of-plant on certain sites and some of it has to 23 do with the size of the containment of the PWR versus 24 the BWR.
25 0 But I take it that the analysis of those O
ALDERSON REPORTING COMPANY. INC.
l 400 VIRGINIA AVE., S.W WASHINGTON D.C. 20024 (202) 554 2345 l l
5650 1 externalities as initiating events is not beyond the 2 state-of-the-art? Is that right?
g 3 MR. ELLISa Which externalities are we now 4 talking about?
5 MS. LETSCHEa The externalities, well, maybe 6 M r. Burns can -- excuse me, Judge Brenner. Maybe let me 7 ask an introductory question.
8 BY MS. LETSCHEs (Resuming) 9 0 What are the --
10 JUDGE BRENNER Well, I thought we were 11 talking about earth, wind, and fire.
12 MS. LETSCHEs I did, too. Maybe I could ask 13 Mr. Burns, however, to relate the earth, wind, and fire -
14 to the specific externalities that were excluded.
. 15 BY MS. LETSCHE: (Resuming) 16 0 Do you mean the fire, flood, and earthquake, 17 and sabotage?
18 A (WITNESS BURNS) Those are generally the ones 19 that people talk about jost often.
20 0 okay. And in your prior answer you said that 21 you were talking about some knowledge you had that 22 earth, wind, and fire initiators in certain 23 circumstances that you had described in your answer were O 24 considered, more important, I guess, than they had been 25 earlier?
O ALDERSON REPORTING COMPANY,INC, 4M VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554 2345
5657
() 1 A (WITNESS BURNS) Yes. I think specifically 2 published reports that I have seen on Zion PRA indicate 3 that those seismic events are potentially contributors 4 in one measure of safety, and that particulsr measure 5 that I use there is early fatalities, which is what they 6 refer to it as.
7 In other measures of safety, other 8 contributors are important. For Indian Point 2 PR A, my 9 understanding is that they have a particular problem 10 with, or a potential problem, with on-site power, that 11 it could be affected by high winds. But again, this is 12 very secondary information. It is very plant-specific, 13 these types of things. And it appears to'be very even O 14 generic to certain types of reactors where you would 15 place the dominant risk.
16 A (WITNESS J0KSIMOVICH) Hay I add, I think that 17 there was an implication that the sabotage was also 18 within the sta te-of-the-art. I disagree with that. I 19 think we know how to analyze earthquakes, we know how to 20 analyze fires. The state-of-the-art has advanced 21 significantly since WASH-1400 was issued. That is not 22 the case with sabotage. I don 't know anybody who has 23 probabilistically assessed sabotage.
() 24 0 Br. Burns, do you agree that an analysis of 25 sabotage is beyond the state-of-the-art?
O ALDERSON REPORTING COMPANY. INC.
400 VIRGINIA AVE., S.W WASHINGTON, D.C. 20024 (202) 554-2345
5658
() 1 A (WITNESS BURNS) Well, I guess maybe I never 2 answered your question you asked about what the current 3 state-of-the-art was. I think in the original decision 4 to focus on in-plant consequences, in-plant systems 5 effects, part of that was the seismic effects and other 6 external externalities, the ability to assess those is a 7 developing technique and had, to that point, not been 8 demonstrated to be managable, a t least in my 9 estimation.
10 But as we have also stated, there have been 11 recent attempts to incorporate those. I still believe 12 that the uncertainties associated with those events and 13 the probabilistic standpoint is very large and, O 14 therefore, the conclusions that one can draw from the 15 evaluations have to be tempered with that large 16 uncertainty. My perception is that it is a large 17 uncertainty.
18 Sabotage, f or tha t particular aspect, I 19 certainly agree that, to my knowledge, that no one has 20 been able to probabilistically evaluate that. That 21 doesn't mean that the techniques of event trees and 22 fault trees haven't been applied for that. But it has 23 been in a qualitative sense.
() 24 0 But you can do event tree and fault tree 25 analyses of sabotage as an initiating events is that O
V ALDERSON REPORTING COMPANY. INC.
400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554-2345
5659 1 right?
2 A (WITNESS BURNS) People have tried to do 3 that. The success has been somewhat limited.
4 A (WITNESS J0KSIMOVICH) If I might add, we 5 lon 't know how to evaluate the probability of the i
6 initiating event.
7 JUDGE BRENNERa Dr. Burns, is tha t the main 8 problem with respect to seismic evaluation, too? Or is 9 it not knowing how to evaluate the probability of the 10 initiating event, or is there something else involved in 11 a greater difficulty with respect to a seismic external 12 event as compared to applying PRA techniques to in-plant l
, 13 events? -
0 14 WITNESS BURNS I can give you my opinion, and 15 then I think we should ask Dr. Joksimovich, also.
l 1
l 16 My opinion is that it is the initiating event 17 is one of the soutces. The other source is trying to 18 determine what effect that seismic event then has on l
19 in-plant systems. That determination is, in my l
l l 20 discussicas with people who have done it, is extremely 21 difficult and highly uncertain, I guess.
22 JUDGE BRENNER. So the implication in my 23 question is incorrect; that is -- well, what I had in
( 24 mind --
i 25 UITNESS BURNS: It is not only the initiator O
ALDERSON REPORTING COMPANY. (NC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5660
() 1 in a seismic event that contributes to uncertainty, 2 laroe uncertainty.
3 JUDGE BRENNER: Can you give me some 4 qualitative feeling for which part of it has the greater 5 uncertainty? Are they close?
6 WITNESS BURNS: Again, I haven't actually ever 7 done a seismic evaluation of a plant. But from my 8 conversations, I believe that both the initiator and its 9 effect on individual components, both of those effects 10 are about equal weight in terms of uncertainty, and then 11 they are additive or multiplicative.
12 WITNESS J0KSIMOVICH: The PBA procedures 13 guide, which has been prolifically referenced in my
() 14 testimony, contains a section and gives a recipe how 15 earthquakes can be treated. And the same is true for 16 fires.
17 It is true what Dr. Burns suggested, that the 18 uncertainties associated with these events are much j 19 larger than they are with internal events. And for very 20 good reasons we don't experience earthquakes of, I 21 don't know, 8.5 on the Richter or something, very 22 f requently, and hence it is very difficult to assign a 23 single value to that.
() 24 Hence, we have to assign a pretty broad
, 25 uncertainty band, but that is one of the virtues of the O
ALDERSON REPORTING COMPANY. INC, 400 VIRGINIA AVE, S.W, WASHINGTON, D.C. 20024 (202) 554-2345
_ _ _ - _ _ _ _ _ _ - ~_ ._. -
5661
() 1 PRA, because the PRA accounts for uncertainties 2 irrespective of how large they might be. And the whole f - 3 technique has an added value because as I said, these 4 are accounted for.
5 JUDGE BRENNERa I do not want to go too far 6 with this at this time, and maybe at no time. But you 7 do not need as f ar as the aspect of assessing the 8 effects upon systems, structures, and components as 9 distinguished from the uncertainty and the probability 10 of the initiating event, you do not need an earthquake 11 in order to be able to have some method of assessing how 12 things will shake out, so to speak.
13 WITNESS J0KSIMOVICHs It is a big shake that
~
14 we are talking about. In the studies which have 15 addressed the issue of earthquakes -- there are a number 16 of those around -- the typical type of a situation that 17 we are facing is that the field of seismology is 18 advancing and is advancing pretty rapidly. But it is 19 lagging behind where the assessment of internal events 20 is. And the kind of evaluation that is of importance is 21 not associated with the le9ols of SSE, it is associated
- 22. with levels of three times SSE.
23 BY MS. LETSCHE: (Resuming)
() 24 0 Mr. Burns, the Shoreham PRA also did not 25 include a consideration of operator errors of commission O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5662
() 1 as an initia ting events is that righ t?
2 A (WITNESS BURNS) I think there is no doubt 3 that operators of commission are included in the 4 Shoreham PBA as initiating events. There is no question 5 in my mind about that.
6 (Counsel for Suffolk County conferred.)
7 0 Mr. Burns, I would like to read to you from 8 page 4-1 of the Shoreham PRA, in which it says, "The 9 study also included losses of off-site power grid and to internal flooding as sources of failures, but 11 specifically excluded event sequences associated with 12 external events such as earthquake, tornado, and flood, 13 and internal events such as fires, sabotage, and O 14 operator errors of commission."
15 Would you like to change your prior answer?
16 A (WITNESS BURNS) No, that is true. And I
~
17 think my prior answer is also true, in the sense that 18 operator errors of commission which cause initiating 19 events, most of the initiating events that we use are 20 all based upon data that have occurred in the field and 21 therefore, operators of commission to lead to that 22 initiating event are, in fact, included in that data 23 base which causes the initiating event, which was your
() 24 question previously.
25 Now, if the question is, are they in general O
l ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5663
() 1 included everywhere in the accident sequences, no, ther 2 are not, as the statement says.
3 (Counsel for Suffolk County conferred.)
4 Q Thank you for that clarification.
5 Mr. Burns, I would like to talk a little bit 6 about the walk-downs that you describe in your testimony 7 that were performed in connection with the Shoreham 8 PRA. I think thst portion of your testimony is on pages 9 102 and 103.
10 You state in your testimony that there were 11 two walk-downs performed in connection with the Shoreham 12 PRA. Can you tell me approximately when those
- 13 occurred?
O 14 A (WITNESS BURNS) In June and August of 1981.
15 Those were the formal walk-downs with checklists. In 16 addition to the formal walk-downs, there have been 17 roughly three or four what I would characterize as plant 18 tours in which formal checklists were not actually 19 involved, but they were more of the plant 20 familiarization type which are recommended in the PRA 21 procedures guide. And they served as verification of 22 the earlier checklists.
23 0 Ihe detailed or the two walk-downs you
() 24 referenced, they were performed prior to the plant 25 tours?
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
56615 h 1 A (WITNESS BURNS) Yes.
2 MR. ELLIS: Let me just point out that he did 3 not testify to two walk-downs, he testified to two dates 4 upon which walk-downs occurred.
5 6
7 8
9 10 11 12 13 0 14 15 16 17 18 19 20 21 22 23 24 25 O l ALDERSON REPORTING COMPANY. INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554-2345
5665 ll) 1 JUDGE BRENNER: For all I know, he walked up 2 and down several times on those days.
3 (Laughter.)
4 BY MS. LETSCHE (Resuming) 5 0 Hr. Burns, did you participate in each of the 6 two walkdowns that you identify in your prefiled 7 testimony?
8 A (WITNESS BURNS) No. I was only on one of 9 those official walkdowns.
10 0 You state in your prefiled testimony that a 11 total of 65.1 man-days was involved in the p7rformance 12 of the two walkiowns you describe there. Did that 13 include the time put in by the LILCO engineers e 14 accompanying the fault-tree analysts?
15 A (WITNESS BURNS) Yes.
16 Q The walkdown that you participated in, did 17 that require a full 2 days of your time?
18 A (WITNESS BURNS) Yes, it did.
19 Q And during those 2 days, what systems did you 20 review? I am not trying to trick you at all. Maybe you 21 would want to refer to section 2.4 of the Shoreham PRA, 22 which I believe you reference in your testimony.
23 MS. LETSCHE: And at this time, Judge Brenner, llh 24 I would like to have marked as Suffolk' County Exhibit 25 Number 22 for identification Section 2.4 from the O
ALDERSON REPORTING COMPANY, INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5666
() 1 Shorehsm PRA.
2 JUDGE BRENNER: Let me make sure I understand 3 what that consists of, because as you know, you a ttached 4 more than one section. That would be pages 2-32 through 5 pages 2-36 of the draft Shoreham PRA; is that correct?
6 NS. LETSCHE: Yes, that is correct.
7 JUDGE BRENNERs That will be so marked.
8 (The document referred to 9 was marked Suffolk County 10 Exhibit Number 22 for 11 identification.)
12 BY NS. LETSCHEs (Resuming) 13 0 Nr. Burns, does referring to Table 2.4.1 14 ref resh your recollection as to what systems you 15 revieced during the walkdown you participated in?
16 A (WITNESS BURNS) If the question is which ones 17 I reviewed, it is the ones that are listed here. During 18 that walkdown, though, I also -- it was also necessary 19 for me to go to every one of these systems. The 20 exception to that, I think, is containment isolation, 21 which is kind of s broadly diverse term.
22 JUDGE BRENNER I am sorry, Dr. Burns, I 23 missed your last phrase, the containment isolation
() 24 system --
25 WITNESS BURNSa Is s very diverse system, and O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554 2345
5667
() 1 I don't think I could characterize that I walked that 2 system down. But in the other cases, it was necessary 3 for me to be aware of where all these systems were and 4 what their spatial locations were.
5 BY MS. LETSCHE: (Resuming)
~
6 0 Dr. Burns, did you complete checklists with 7 respect to all of the systems that you just indicated 8 you reviewed, those listed on Table 2.4.1, with the 9 exception of contsinment isolstion?
10 A (WITNESS BURNS) As I said, the checklist that 11 I was responsible for are listed there. The other 12 members of the team were responsible for the remaining 13 ones.
()
l 14 0 So the checklists were just on the service 15 water and the reactor building standby ventilation, I chilled water, and the reactor building closed-loop 16 17 cooling water systems?
18 A (WITNESS BURNS) The ones that I completed.
19 0 But while you were at the plant, you also 20 during that walkdown reviewed the other systems listed?
21 A (WITNESS BURNS) Yes.
22 0 I would like to direct your attention to Table 23 2.4.2, which is on page 2-35 of the Shoreham PRA. Is
() 24 this table an example of the form of checklist that was 25 used in connection with the walkdown?
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5668
() 1 A (WITNESS BURNS) It is a form of -- it is the 2 format that we used. It is not a particularly good 3 example. But it is the format we used, yes.
4 0 When you say it is not a particularly good 5 example, do you mean the information that is filled in?
6 A (WITNESS BURNS) Yes.
7 0 The heading fourth from the last, " Hazards 8 nearby," can you tell me what that means?
9 A (WITNESS BURNS) The intent of this and the 10 result of the walkdown were to identify if, in fact, 11 there were hazards located near equipment required for a 12 system to operate which would interfere with its proper' 13 operation, given normal sequences or given degree
() 14 accident sequence events.
15 For example, I guess the best example for this 16 particular discussion is where is the room ven tila tion 17 for each of the pieces of equipment? Is it located 18 adequately to supply cooling, and are there any 19 obstructions that might interfere with its proper 20 operation.
21 A second example would be for the internal 22 flooding evaluation which we ild as part of this PRA.
23 We needed to know locations of pipe and sources of water (3
V 24 that -- for the elevation 8 compartment in the reactor 25 building. And so that also was a major point of O
ALDERSON REPORTING COMPANY. INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554-2345
5669
() 1 investigation in these walkdowns.
2 0 In general, Dr. Burns, did the persons listed 3 in Table 2.4.1 as performing the walkdowns, perform them 4 individually? Let me see. I understand you say they 5 had some LILCO accompaniment. I am setting that aside 6 for a minute. Did they perform the walkdowns by 7 themselves?
8 A (WITNESS BURNS) Along with the LILCO 9 en71neering representative and whatever tour guide that 10 we had on that particular tour, yes.
11 0 There was one LILCO engineer with each person 12 in general?
1 13 A (WITNESS BURNS) Yes.
O 14 JUDGE BRENNERs Ms. Letsche, I do not know if 15 that is going to be important f or anything. But if it 16 is not, I hate to waste time with it. But I am not sure 17 if the number of people are perfectly clear now.
18 There is the PRA analyst; am I correct, Dr.
19 Burns?
20 WITNESS BURNS Yes, sir.
21 JUDGE BRENNER: Then there is the one LILCO 22 engineering representative; is that correct?
23 WITNESS BURNS: Yes.
() 24 JUDGE BRENNER: And then there is also a tour 25 guide?
O ALDERSON REPORTING COMPANY,INC,
' 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5670 lll 1 WITNESS BURNS: There may have been a tour 2 guide in aidition to the engineer, or just the LILCO 3 engineer.
4 JUDGE BRENNER: So it could be two people, it 5 coald be three people?
6 WITNESS BURNSa For the purposes of assessing 7 the proper operation of the plant or the arrangements or 8 any technical information as input to the PRA, there 9 were two people.
10 (Panel of witnesses conferred.) -
11 WITNESS BURNSt I guess I didn't want to imply 12 -- also in addition to the LILCO engineers, the what I 13 call tour guides were really technical personnel from O 14 the site and general startup engineers who could answer 15 any additional quastions that we might have about the 16 specific plant operation.
17 WITNESS KASCSAK I might just add, these 18 individuals are -- I felt bad that he categorized then 19 as " tour guides," because they are very technical 20 people, they hava inticata f amiliarity with these 21 systems, and that is one of the reasons why we tried to 22 get these people to go along with these particular 23 valkdowns.
h 24 (Counsel f or Suf f olk County conferred. )
25 BY MS. LETSCHE (Resuming)
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554-2345
5671 l() 1 0 Dr. Burns, could you describe the sources of 2 data used in the Shoreham PRA?
3 JUDGE BRENNERs You are interested in all of 4 the sources of all of the dats? Msybe it would help if 5 you could focus the question a little better, unless you 6 truly mean the question the way I phrased it.
7 BY MS. LEISCHE (Resuming)
/
8 0 I guess what I am interested in is the major 9 sources of data. You' talk about it in section 8.2 of to the Shoreham PRA. ' -
11 A (WITNESS BURNS) 8.2 only refers to a very 12 small portion.
13 0 Let me ask it this way. You testified in your O 14 dritten testimony', and there has been other testimony, 15 that there are a number of different data sources for 16 these sorts of studies. Can you -- and I guess I am 17 talking particularly about component failure data right 18 now -- I would like to know what sources of data were 19 used in the Shoreham PRA with respect to component 20 failures?
21 A (WITNESS BURNS) Okay. We are going to limit 22 the discussion to only the component failure rate data?
23 That is what you want to talk about?
() 24 0 Right now, yes.
25 A (WITNESS BURNS) Generally speaking, we used O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
4 T 5672
) <1 f our basic sources of da ta s the NRC studies that have 2 been done specifically by EGCG; those we used to 3 characterize major components like pumps and valves, and
{) 4' as an input to diesel f ailure ra tes, we also used some 1
5 General Electric operating experience data that was 6 a va ilable . Generally, we used those for specific
- 7 components that were applicable to only BWRs and 8 instrumentation mostly.
> 9- In addition to that one, there was not other 10 available sources. We resorted to WASH-1400 as a
, 11 basis. And I don't believe that we used IEEE-500, which
-12 was electrical components, although we quoted it for 13 comparison purposes.
O 14 O How did you determine which of these sources 15 to use for a particular component?
16 A (MITNESS BURNS) I guess I pretty much went 17 through those in the priority that I used that the NRC 18 data from EGCG was an evaluation of actual operating l
19 experience in operating nuclear power plants, and it was 20 the most recent source of data. So we chose that as the l 21 preferable source of data.
l 22 In general, these also yielded the highest 23 failure rates for components, although one has to temper t
l
() 24 th a t with the understanding th a t what is called a 25 failure rate in that context was generally things that l
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W WASHINGTON, D.C. 20024 (202) 554 2345
5673 (O
_) 1 were out of technical specification.
2 So tha Esilure rate was characterized by items 3 that may not classically be called an ultimate failure 4 of the component.
5 JUDGE BRENNER. Ms. Letsche, since we are 6 going to break at 4400, we could take a midaf ternoon 7 break any tine you want.
8 HS. LETSCHEs Just a second.
9 (Counsel for Suffolk County conferred.)
10 BY MS. LETSCHEs (Resuming) 11 0 Dr. Burns, with the exception of the pumps, 12 valves, and diesel generators and the BWR-specific 13 components, did the remainder of the data come from O 14 WASH-1400?
15 A (WITNESS BURNS) For individual component 16 failures, now we are talking about?
17 0 Yes.
18 A (WITNESS BURNS) I guess what remains after 19 that is a very small subset of the components of the 20 hardware components in th e pla n t. But, yes, the 21 remainder did come from WASH-1400.
22 (Counsel for Suffolk County conferred.)
23 Q Dr. Burns, do you consider pipe failures 24 different from component failures or pipe failures 25 included in what we have been discussing here?
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
. - _ . __._______..______._.__________.________.____.________.____________.___________J
5674
() 1 A (WITNESS BURNS) There are two aspects of the 2 pipe failure question. One is, as an initiating 3 frequency -- and that is discussed in section A.1 -- and 4 f rom tha t standpoint, we haven't talked about that yet, 5 as a contributor to other accident sequences, it was --
6 the data was generally taken from WASH-1400 f or tha t 7 particular failure mode.
8 0 Okay. Where did you get -- what was the 9 source of the pipe failure data in the first instance 10 you just described? -
11 A (WITNESS BURNS) As an initiating event?
l 12 0 Yes.
l 13 A (WITNESS BURNS) We tried to look at operating 14 experience that had occurred in nuclear power plants in 15 order to characterize the f requency of potential pipe 16 f ailures that could initiate an accident. One of the 17 problems with WASH-1400 was that there was a very 18 limited data base at the time for nuclear power plant 19 operation, and therefore most of the information on 20 LOCAs, which is really what we are talking about here, l
21 sost of that information was inferred from operation of 22 otner plants.
i
- 23 So there have been some studies which I
() 24 participated in, and there was another study by Spencer 25 Bush to look at operating experience and try to
>O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S W., WASHINGTON, D.C. 20024 (202) 554 2345
5675
() 1 determine if there have been incidents at nuclear power 2 plants that would allow us to characterize numerically 3 the values of pipe failures.
4 The conclusion was, for small breaks, yes,
( 5 indeed, there have been small breaks since that 6 containment. And so, from that standpoint, it was 7 relatively straightforward to characterize the frequency 8 of those events.
9 For medium and large breaks, it was more 10 dif ficult because there have not been -- at letst in the 11 operating experience that I have seen -- medium or large 12 LOCAs inside containment in the primary system piping.
13 And therefore, it was necessary to somehow O 14 bound that problem, and we used a technique similar to 15 tha t in WASH-1400, which was to estimate that the amount 16 of piping inside the primary containment was 17 approximately 10 percent of the high-pressure piping in 18 the plant, and that if we could find failures in the 19 secondary plant or in the entire nuclear plant, that the 20 piping quality was comparable to that in the primary 21 system, although probably in some cases it may not be 22 quite as good, that that would give us a representation 23 of the primary system piping failure rate, although it
() 24 might be conservative.
25 But because of the lack of data, we -- at O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5676
() I least I tended to be more conservative than realistic, 2 in some cases. So I did identify some cases where there f- 3 were pipe failures in the secondary plant of nuclear
(>) 4 power plants and, therefore, inferred that roughly since 5 that represented 90 percent of the piping in the plant, 6 that we could then estimate what primary system failure 7 rates would be.
8 It turns out that the large LOCA frequency 9 that we have used in the Shoreham power plant is higher to than -that which was used in WASH-1400, and is roughly 11 comparable to that used in the Zion PRA, which is more 12 recent.
13 (Counsel for Suffolk County conferred.)
14 A (WITNESS BURNS) It is probably useful to note 15 that the large LOCA initiating f requency for BWR turns i 16 out to be not the major contributor, that the sequences 17 following such an event.
18 MS. LETSCHEs Judge Brenner, this would be a r
19 convenient place to break, if that is what you were l
20 asking me before.
21 JUDGE BRENNER: Dr. Burns, I am not sure I 22 heard you correctly on the last answer. You said it i
l 23 turns out not to be a major contributor for BWRs?
() 24 WITNESS BURNS Yes, sir.
25 JUDGE CARPENTER: Dr. Burns, not following the
()
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5677 O tiae or aue=tioataa aut 1a ta r or ta-2 testimony I would like to ask a question which you might 3 sbout over the break as a source of relaxation. If you 4 would look on page 103 of your prefiled testimony, you 5 draw attention to the fact tha t you use systen 6 interaction analyses as starting materials for the PRA 7 analysis. And you list a number that you used, and I 8 recognize on page 104 Item 8 says "Special studies."
9 Turning back in the prefiled testimony to pages 56 10 through 64, there are Items A through R that have been 11 presented to us as system interaction studies.
12 I would like to know how many of those you 13 have used, which ones of those you found helpful.
O 14 And I say, if you want to think about that, I 15 have asked a very big question. There are about 17 16 items. And you might just take a minute to look at 17 that.
18 WITNESS BURNSs Yes. I can tell you at the 19 end of the break.
20 JUDGE BRENNER: Let us come back at 3s10.
21 (Whereupon, at 3 00 p.m., the hea ring was 22 recessed, to reconvene at 3:12 p.m., this same day.)
23 24 25 O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) $54-2345
5678
() 1 JUDGE BRENNER: Back on the record.
2 Immediately after we wen t of f the record, Mr.
3 Rawsom indicated he wanted to say something regarding 4 the SAI matter which we raised the other day.
5 MR. BAWSON: Thank you, Judge. It may well be 6 that I have been oversensitized to this question of 7 SAI's role by the Board in its comments yesterday, 8 although I sure that this is not a bad state of 9 affairs.
10 But I just wanted to be absolutely sure that 11 the Board is avara of and has focused on the fact that i 12 S AI's role as a subcontractor to Lawrence Livermore 13 axtends to~the report which has been marked as Suffolk O 14 County Number 19. That is indicated on the face of that 15 document.
16 But there were some questions from the Board 17 to Mr. Burns earlier which caused me to wonder if, in 18 fact, the Board had focused on that. And I wanted to be l 19 absolutely sure that we raised that while Mr. Burns is l
20 here.
21 JUDGE BRENNER We saw Mr. Kelly's name on the 22 report, and his affiliation.
1 23 HR. RAWSONa Fine. I apologize for taking up
( 24 the time. .
25 JUDGE BRENNERa The reason I asked th e O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
__ ~ _ _ . -
I l
5679 l
() 1 questions I asked is I was not sure, first of all, I did 2 not have the conflict question in mind at all then, and 3 the reason I asked the questions I asked was, as I 4 indicated, the initial somewhat tentative nature of Dr.
5 Burns' response. Just because Mr. Kelly was involved, I 6 did not mean that Dr. Burns at the time of that question 7 by Ms. Letsche or any time, for all I knew, had those 8 categories in mind. And that was the reason I asked the 9 question. But I do appreciate your raising that to matter. l 11 BY HS. LETSCHE: (Resuming) !
l 12 0 Dr. Burns, what was the data base used for 13 HPCI and RCIC failures in the Shoreham PRA?
O 14 JUDGE CARPENTERt Excuse me. Could I get the 15 answer to my question first?
16 MS. LETSCHE I am sorry, Judge Carpenter.
17 WITNESS BURNS: On page 104 there is a 18 reference made to special studies, and it cites DC power 19 reliability, diesel generator reliability, and 20 anticipated transients. And in the same line those 21 special studies were meant to imply that people have 22 done probabilistic evaluations or have identified )
23 potential systems interactions on a generic basis. And 24 tha t is what those,special studies refer to.
1 25 On page 101 I do say that the following aids (2)
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5680
() 1 have also been employed in the analysis of the Shoreham 2 plant. And under that I do say existing LILCO 3 deterministic analysis and referring to page 56, which 4 was the deterministic evaluations, the ones that were 5 reviewed in preparation for and in parallel with the 6 performance of the Shoreham PRA.
7 Number A, the pipe failure and internal 8 flooding was reviewed. However, we were forced to carry 9 it further than that particular study went. We did 10 additional work beyond that study.
11 The missile study, we used that evaluation 12 basically to eliminate missiles as a dominant failure 13 mode in the fault trees. The fire hazard analysis was O 14 not explicitly used, although we did review it. The 15 cable separation report did become an underlying 16 assumption in much of the PRA, in that there was, in 17 fact, separation of the cable. And we did also use that 18 in elevation 8 evaluation.
19 The f ailure-mode-and-effects analysis in E was 20 not available to us, although some of the summary 21 results were, and we reviewed those basically for 22 control circuits, although a detailed review was not 23 performed.
) 24 , Electrical bus failures -- I think I haven't 25 looked at that today -- I think that is the one that is O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5681
() 1 in the FSAR. And if i t is, then we looked at that, but 2 I can't recall if we explicitly used it other than as 3 background information.
4 G is ongoing, so we did not have any 5 information on that; nor H, which is ongoing. I, the 6 Limerick PR A informa tion a ssocia ted with that, from my 7 own personal knowledge was used. The BWR-6 standard 8 plant was done by General Electric. I have some 9 peripheral knowledge of what they were doing, but did 10 not use anything that they had done.
11 Heavy loads was not used. K, which is 12 analysis of industry experience, we did perform our own 13 LER reviews and reviewed documents in the literature O 14 with regard to LERs but not throgh LILCO.
15 Preoperational testing and startup testing, 16 thst, we dit not review that or use that. M, N were no t 17 used. O was reviewed but not directly used.
18 P, which is water level instrumentation study, 19 that study, any study that I have ever seen on water 20 level instrumentation has been made available to me 21 after the submitts1 of the draft, so it is not included
! 22 in the current draf t of the Shoreham PRA. We are nov 23 actively pursuing that for Shoreham, and we are also
( 24 . performing trying to put the water level instrumentation 25 problem in perspective, using event trees and fault O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554-2345
5682
() I trees for the BWR owners group who are addressing this 2 question separately on a generic basis.
3 TMI-2, the references that were given by General 4 Electric to those studies, we did, in fact, use in 5 conjunction with the Limerick PRA. So all of that 6 information that was applicable to Shoreham we also used 7 for Shoreham. Mostly, that was thermohydraulic analysis 8 that was done for degraded condition, degraded system i 9 conditions with multiple failures.
10 JUDGE CARPENTER: Thank you for your 11 response.
12 JUDGE BRENNER: Dr. Burns, on the 13 event-tree / fault-tree approach being worked on for water O 14 level, with respect to the BWR owners group, do you know 15 what kind of schedule that is on, and are you involved 16 yourself in working on it?
17 WITNESS BURNS: Yes, sir, I am. Before the 18 hearings, we were supposed to have the final draft July 19 Sth, but I don't foresee that happening. I think the 20 owners group is under time constraints, so I am not sure 21 what they are going to do. They have the LaSalle 22 hearings, but I know that we are targeted for July 6th l
23 right now.
() ,
24 JUDGE BRENNER: So it is an immediate future 25 time frame kind of thing?
l l
ALDERSON REPORTING COMPANY,INC, 400 VIRGINTA AVE., S.W WASHINGTON, D.C. 20024 (202) 554 2345
5683
() 1 WITNESS BURNS: Yes, sir. ,
2 JUDGE BRENNER: Proceed.
3 BY MS. LETSCHE: (Resuming) 4 Q Dr. Burns, what was the data base used in the 5 Shoreham PRA for HPCI and RCIC f ailures?
I 6 A (WITNESS BURNS) This is my favorite t 7 question.
8 0 Well, I am glad I could make you happy.
9 A (WITNESS BURNS) Several years ago, af ter THI, 10 there was a task force put together at General Electric, 11 of which I was a part, to identif y wha t was the 12 reliabilty of the ECCS network and in response to a i
i 13 small-break LOCA. One of the principal items of 14 interest was the reliability of the high-pressure 15 systems; in other words, avoiding rapid depressurization 16 was a key feature in understanding the response.
17 In the generic plant that we looked at, or 18 model, I guess it was really, a generic model, but from 19 the standpoint of HPCI and RCIC response, it is very 20 close to the existing Shoreham design.
21 We actually used Peach Bottom as the base 22 plant, but the systems are virtually identical. In 23 order to do that study, we did two thingsa we developed
( 24 fault tree models in detail of the individual systems, 25 and we used component-level failure rate data to O
ALDERSON REPORTING COMPANY,INC, j 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5684
() 1 characterize the conditional probability of individual 2 failure modes of the systems.
3 We then combined in the Boolean fault tree 4 f ormat the probability of HPCI and RCIC, and in 5 addition, we confirmed what the maintenance on 6 availability of RCIC and HPCI would be using operating 7 experience both from Peach Bottom and from industy in 8 2eneral.
9 Since this was a major issue in WASH-1400, we 10 f elt it was important to characterize it more correctly 11 than it had been done there. So we had one evaluation 12 that was a fault tree evaluation of the combined system 13 operating as a function of a demand requiring their O 14 operation.
15 In addition to that, there was a review of 16 operating experience up to that point by General 17 Electric and that review reflected the Hatch incident 18 that was talked about this morning, in which HPCI and 19 RCIC were both unavailable at the same time during a 20 challenge.
21 What we did was we tried to benchmark our 22 fault tree calculations versus the actual operating 23 experience that had occurred in the field. And as it
( 24 turns out, if you look at the individual HPCI operation 25 for comparable failure modes -- and we have to be a O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5685
() 1 little bit careful here because the overall calculated 2 una vaila bility of a system on demand, as used in the 3 Shoreham PRA, 'hereas in other PRAs includes many 4 things, it includes tests and maintenance, it includes 5 f ailure on subsequent starts, which turns out to be at 6 least in the fault tree evaluation that I have performed 7 and General Electric has ag reed to, to be a relatively 8 high contributor.
9 So what we actually compared was we compared 10 the initial --
the failure on initial start is what was 11 compared, failure on initial start versus operating 12 experience of failure on initial start, because in most 13 tests the -- and these are tests, most of the operating 14 experience are tests, not real demands -- because it is 15 very unlikely; there has not been a great deal of 16 demands, actual challenges, but more test operation.
17 And during test operation, these systems do 18 not operate for long periods of time, and therefore ther 19 are -- the most accurate comparison is between initial 20 start and run versus the model's initial start and run.
21 And the values that we got from the operating 22 experience indicated there was a roughly normal 23 distribution with roughly 4 percent unavailability for
() 24 this initial start and run failure probability.
25 Out in the tail was Hatch, which I think Mr.
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5686 A
(_/ 1 McGuire alluded to when he was talking about the j 2 reliability of systems. Of course, there are plants j
(} 3 that have extremely high reliability on the other end, 4 very, very high reliability of those systems. But most 5 of the plants were clustered sround roughly 4 percent )
6 unavailability for those systems.
7 The combined failure probability, when we 8 combine RCIC and HPCI together in the model in a Boolean 9 fashion to determine what the combined unavailability is 10 predicted by the model was, I believe -- and I would 11 have to go back and check my exact numbers -- but I l
12 believe it is roughly in the study we did, was 9 x
-3 13 10 . But you cannot hold me to that number.
14 In the operating experience data, we also 15 evaluated what the likelihood of seeing such an incident 16 as occurred at Hatch, which was a combined failure of I
17 both systems, and we determined that based upon the 18 operating -- let me interpret the model first. I am 19 getting a little flustered here.
20 The models, the number that we calculated for 21 unavailability using the model said using the operating 22 experience to the date thst we had examined it, that 23 roughly we should have seen between two, ro ughly two l
l
( 24 incidents, in the operating experience to that date in 25 which there was an actual challenge. And we actually O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5687
() 1 had a failure of both systems.
2 The operating experience indicated to that 3 date what we had available at that time indicated that 4 there had been one such event. My conclusion from that 5 comparison was that the model that we had constructed 6 using the input dsta that we had used for component 7 failure rates was a true representation of what was 8 going on out in the field, that, in fact, we were making 9 a very good approximation for the actual performance of 10 a plant during a challenge for those standby safety 11 systems.
12 Now, there are other things that the model 13 does incorporate that were not confirmed by operating O 14 experience. And those would bes failures on second 15 challenge during a sequence and the maintenance 16 unavailability. But that was a separate study.
17 Q Dr. Burns, is the study that you just 18 described the one that is ref erenced in the a ppendices 19 to the INP3 report on the Hatch event?
l 20 A (WITNESS BURNS) I believe when I looked at 21 the INPO report that I saw some of the same information 22 that we had developed in there, although I don't know if 23 they reference where that came from.
24 (Counsel for Suffolk County conferred.)
25 0 For the Shoreham PRA, Dr. Burns, did you use O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5688 l
1 the results of your model, or did you use the GE's 2 target reliability for those systems?
3 A (WITNESS BURNS) I always used the results of 4 the model because I need the model to accurately depict 5 in sequence the dependencies that may occur during the 6 sequence. In other words, there are certain failure 7 modes that occur in a fault tree that must be carried
. 8 through the Boolean algebra with other components that 9 may occur later on, 10 For exsaple, initiation logic for HPCI maybe 11 the same logic or the same components that initiate 12 other safety systems. Therefore, I need that 13 information, and I can't, if I just use one single O 14 number, I ca n't model that. So I use the operating 15 experience as a banchmark to model and then use the 16 model.
17 MS. LETSCHEa I would like to have marked as 18 Suffolk County Exhibit Number 23 for identification, two 19 pages taken from the Shoreham PRA. They are not 20 consecutive pages. One is page 3-172; the other is page 21 u-3.
22 (The documen t referred to 23 was marked Suffolk County 24 Exhibit Number 23 for 25 identification.)
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5689 O avoct antaxca: 1 texe it 7ou re sotaa to ==e 2 these for all of your purposes, at least together?
3 MS. LETSCHE Yes. That is why I have 4 attached them.
5 BY MS. LETSCHEa (Resuming) 6 0 Dr. Burns, I would like to direct your 7 attention to Table 4.1, which is on the second page of 8 what has been marked as Suffolk County Exhibit 23 for 9 identification. Does that table represent the 10 probabilities of the frequency of the occurrences listed 11 -- let me strike that.
12 Can you explain to us what the Table 4.1 13 indicates, that those results are?
14 A (WITNESS BURNS) The nuaerical value? The 15 numerical values or the purpose of the summary table? -
16 0 Well, this states the probability of these 17 types of events occurring per reactor-years is that 18 right?
19 A (WITNESS BURNS) The frequency of classes of 20 events occurring.
21 0 Can you explain, give us a general explanation 22 of what is in Class 1 on that table?
23 A (WITNESS BURNS) Certainly. In WASH-1400 the 24 evaluation of accidnts lumped most accident sequences 25 together. In other words, events such as ATWS or a LOCA O
e ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W WASHINGTON, D.C. 20024 (202) 554 2345
5690
() 1 or a loss-of-coolant-makeup are all treated as having 2 similar consequences in terms of containment interaction 3 and potential radionuclide release to the environment.
4 This table and the philosophy tha t we have 5 adopted here and in the Limerick analysis was to try and 6 make the estimate of radionuclide release more realistic 7 for those sequences that may have different potential 8 impacts on both containment and radionuclide release.
9 Loss-of-coolant-makeup are those accident sequences in 10 which we are unable, or it is postulated that we are 11 unable, to get sufficient coolant into the reactor 12 vessel.
13 These types of sequences are total loss of all O 14 ECCS injection due to any number of sources or causes, 15 why that may occur. So for those cases where that is 16 the problem, they are classified as Class 1 events. And 17 a containment analysis is done, is performed -- is 18 performed assuming that that is the problem.
19 Class 2 events are what WASH-1400 calls the TW .
20 events, which are an inability to remove heat from 21 containment; therefore, you challenge the integrity of 22 the containment, even though the core is maintained with 23 adequate core cooling up to the point where containment 24 may be challenged, ultimate capability of containment 25 may be challenged.
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5691
() 1 A LOCA is an extension of what is treated 2 generally in Chapter 15, carried now to the point where
() 3 4
ECCS systems are not able sufficient to keep up with that required cooling.
5 ATWS without poison injection, which is Class 6 4, are those accident sequences in which sufficient 7 reactivity using the control rods or the standby liquid 8 control system are not available, yet there is still 9 cooling to the core.
10 And Class 5 is an event postulated to occur if 11 the LOCA occurs outside containment and there is a 12 failure to isolate it, and there is a failure to make up 13 sufficient water to the reactor vessel.
O 14 0 The total core vulnerable frequency per 15 reactor year that is listed at the bottom of that Table 16 4.1, that is, the mean combined frequency of the 17 frequencies listed by class?
18 A (WITNESS BURNS) Yes.
19 20 21 22 23 24 25 ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5692
() 1 JUDGE MORRIS: I am sorry, I did not 2 understand the question the way it was phrased in 3 there. I am not sure of the answer.
[}
4 BY MS. LETSCHEa (Resuming) 5 0 Let me try it this way, Dr. Burns. Can you 6 explain what the number at the bottom of the table 7 represents, the one next to " Total core vulnerable 8 frequency per reactor-year"?
9 A (WITNESS BURNS) That is the summation of the 10 point estimates of each of the classes.
11 Q Now, the point estimates are those that are 12 listed hera. This is just one number. The point 13 estimates are the ones you are referring to up above; is 14 that right?
15 A (WITNESS BURNS) The point estimates for each 16 class, yes.
17 0 There are uncertainties associated with each 18 one of those, are thera not?
19 A (WITNESS BURNS) Yes, there are.
20 0 Are those uncertaintios the ones that are 21 summarized in Table 3.8.1 on the first page of what has 22 been narked as Suffolk County Exhibit 23 for 23 identification?
24 A (WITNESS BURNS) Yes.
25 (Counsel for Suffolk County conf e rred . )
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554-2345
5693
() 1 0 The uncertainties that are listed in Table l 2 3.8.1 I noticed are described as error factors. Does 3 that mean that the frequency point estimates in Table 4 4.1 can be off, can be in error, by a multiple of the 5 error factors listed in Table 3 8.17 6 A ( WITNESS BURNS) In this PRA we ira using 7 terminology in which the error factor is one of the 8 parameters of a log-normal distribution. That is 9 discussed in Appendix A, what parameter that error 10 factor is.
11 And in terms of Table 4.1, the point estimates 12 that are given there are mean values, and the mean value 13 of the log-normal distribution is plotted in the figure O 14 4.1. And the 5 parcent confidence bounds using the 15 error f actors from Table 3.8 1 are used to derive what 16 those confidence bounds are.
17 (Counsel for Suffolk County conferred.)
18 0 Let me make sure I understand. The figure 4.1 19 is a representation of sort of combining Table 4.1 point 20 estimates and the error f actors listed in 3.8 1 to show 21 sort of bands of arror?
22 A (WITNESS BURNS) Yes.
23 0 Dr. Burns --
() 24 MR. ELLIS: Excuse me. Might I have that 25 question and answer read back, please?
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S W WASHINGTON, D.C. 20024 (202) 554 2345
5694
() 1 (The reporter read the record as requested.)
2 WITNESS BURNS: I guess I would like to 3 loosely speaking error, possibly uncertainty, I guess, 4 would be a better terminology.
5 JUDGE MORRIS: Excuse me, Ms. Letsche, if I 6 might ask a question or two.
7 MS. LETSCHE: Certsinly, Judge Morris.
8 JUDGE MORRISs You described, Dr. Burns, what 9 you mean by " error factor," but I think there are 10 probably only three or four people in the room that 11 understand the use of that term in the probabilistic 12 distribution sense. Could you expand on the definition I 13 of that? It can be defined in terms of the probability O 14 distributions; right? )
l 15 WITNESS BURNSs Yes. It requires a 16 probability distribution to be characterized in terms of 17 the log-normal, which is what we assumed, or assigned 1
18 it, in Table 3.8.1. And I guess possibly the simplest 19 explanation would be if it was a median value, then the 1
20 error factor would indicate the multiplicative value of ;
21 both high and low of the 5 and 95 percent confidence 22 limits. j 23 In other words, if we had used in figure 4.1, ,
() 24 if we had used the median valu-3, then the triangles 25 would be approximately in the midband of all of the O
I ALDERSON REPORTING COMPANY. INC.
400 VIRGINIA AVE., S.W., WASHINGTON D.C. 20024 (202) 554-2345
5695
() 1 bands that are shown there. And the end of the bands, 2 as they are shown, would remain the same, and those are rT 3 the 5 percent and 95 percent confidence bounds on the V
4 probability distribution for that sequence frequency.
5 JUDGE MORRISs In figure 4.1 just below where 6 you have the triangle, that explains that is a mean 7 estimate. You say 5 and 95 percent confidence bounds.
8 That is an expression, I think, which is strange to some 9 of us. It really means 95 percent confidence, does it to not?
11 WITNESS BURNSs 95 percent confidence; right.
12 JUDGE MORRIS 4 And I guess I was confused on 13 error factor, because I have seen a different definition O 14 mathematically that sometimes it is called " error 15 factor," which is related to a hazards function. But 18 that is not what you meant here?
17 WITNESS BURNSs No, sir, I am sorry.
18 JUDGE CARPENTER: Let me be sure. Judge 19 Morris, I think, is trying to help me partly. What is a 20 5 percent confidence bound? I am not knowledgable in 21 this area at all. I am quite used to people talking in 22 terms of confidence bounds. I have just never seen 23 anybody interested in the bound, a 5 percent bound, 24 before.
25 WITNESS BURNS: Are you asking me O
ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WAS;i;t:0 TON, D C. 20024 (202) 554-2345
5696
() 1 q u a lit a tiv ely ?
2 JUDGE CARPENTER: Yes. I am just surprised.
3 I presume it is analogous mathematically to a 95 percent 4 confidence bound, and the way it is defined is 5 5 percent.
6 WITNESS BURNS: This is the 90 percent 7 confidence bounds on the probsbility distribution at the 8 5 percent.
9 WITNESS J0KSIMOVICH4 Is percentile something to better for you?
11 JUDGE CARPENTER: Of course. Thank you.
12 WITNESS J0KSIMOVICH: You can think in those 13 terms.
O 14 JUDGE MORRIS : To put in layman's terms, to be 15 sure we understand, taking the bounds, one of these
- 16 lines through a triangle, does it mean that you have 90 17 percent confidence that the true value lies within those 18 bounds?
19 WITNESS BURNS: Yes, sir.
20 JUDGE CARPENTER: Thank you.
21 BY MS. LETSCHEs (Resuming) 22 0 Dr. Burns, what is the error factor for the 23 total core vulnerable frequency per reactor-year that is
( 24 listed in Table 4.1?
25 A (WITNESS BURNS) We did not calculate that O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
5697
() 1 directly. The inputs to the ex-plant consequenca used 2 these uncertainty bounds for each individual class. The 3 ex-plant consequence calculation is performed by class 4 and by containment failure mode.
5 0 Is it safe to say that the error factor for 6 that total core vulnerable frequency would be in the
. 7 range of the uncertainties listed for the individual 8 classes in Table 3.8.17 l
9 A (WITNESS BURNS) I think a better 10 approximation would be the Class 1 would roughly bracket 11 the uncertainty bsnds. But that is very rough, though.
12 (Counsel f or Suffolk County conferred. )
13 0 Is that because Class 1 is the dominant O 14 contributor?
. 15 A (WITNESS BURNS) Yes.
18 (Counsel for Suffolk County conferred.)
17 0 Dr. Burns, with respect to Class 3, identified 18 as "LOCA," is there any contribution to that event from 19 ATWS or any other contributors?
20 JUDGE BRENNER: Ms. Letsche, I am sorry, it 21 was my fault, but I missed the beginning of your 22 question.1 23 MS. LETSCHEa I am asking about Class 3, which
( 24 is identified as "LOCA."
25 BY MS. LETSCHEs (Resuming)
{
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5698 v.
() 1 0 MY question is is there any contribution to 2 tha t class from ATWS or any other events?
3 A (WITNESS BURNS) There are ATWS events that we 4 have calculated could lead to a LOCA. If there was 5 failure of sufficient relief valve capacity during the 6 challenge from that standpoint, yes, there are, but very 7 low frequency. And they don't contribute a significant 8 amount to the net value.
9 (Counsel for Suffolk County conferred.)
10 0 Can you describe in a little more detail what 11 the ATWS events are that could lead to this LOCA Class 12 3?
13 A (. oESS BURN 3) What we are postulating here O 14 is that there is an anticipated transient, that there is 15 a failure to scraa, with its associated conditional 16 probability. And then, in addition to that, the 17 mitigating systems that would still be available, RPT 18 being one of them, if RPT fails, then the LOCA would 19 ensue, or we assume that a LOCA would ensue.
20 But it is not included in this LOCA frequency, 21 because the LOCA calculations here have lower source 22 term releases than ATWS. ATWS has one of the higher 23 source term releases, and that is why it has been
() 24 segregated out. So for the failure of RPT case, it does 25 cause a LOCA, but it does lead to higher radionuclide O
ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5699
() 1 sourcee,'cource release. Therefore, it's included in 2 Class 4.
3 There are other cases in which, as I said O 4 before, the relief valve ca pacity is either inadequate 5 or sufficient; number of relief valves f ailed to open and 6 those events ac'e -- lead directly to a large LOCA 7 event. That is an assumption that we make.
8 JUDGE BRENNER: Dr. Burns, what is "RPT"?
9 Reactor punctura?
10 WITNESS BU RNS: R ecirculation puncture. That 11 is a spacisl logic system that has been added to 12 Shoreham.
13 JUDGE BRENNERa Ms. Letsche, we would like to 14 break close to 4:00, but if you have a few more 15 questions that you want to ask today, I will s11ow you 16 some limited additional time.
17 MS. LETSCHE4 Just a minute, Judge Brenner, 18 please.
19 (Counsel for Suffolk County conferred.)
20 MS. LETSCHE: Judge Brenner, this is actually 21 s logical breaking point in my cross examination.
22 MR. ELLISs Judge Brenner, for planning 23 purposes, sa y we know spproxima tely how much to expect
() 24 on PRA methodology, so that the witnesses can make plans 25 as well next week?
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
- 5700 O i
< Wiratss aoxs1"ov1ca rae v ex seter, we 2 fi o p'e .
3 MR. ELLISs The week after.
4 JUDGE BRENNER: All right. Let me first 5 establi$h that we will be back in this room on July 6th 8 at 10:30 a.m. We are going to break early that day, 7 though,'at the end of the day, either 4 00 or 4:30. I 8 vill decide that day.
9 And I think we will get the answer to your 10 question. I think we could go off the record for that.
11 Is there anything else we need to do on the 12 record?
13 (No rasponse.) -
0 14 JUDGE BRENNER Okay, we vill adjourn until 15 10430 on July 6th.
16 (Thereupon, at 4: 00 p.m., the hearing in the 17 above-entitled matter was adjourned, to reconvene at 18 10:30 a.m. on Tuesday, July 6, 1982.)
, 19 20 21 22 23 24 25 O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE, S.W WASHINGTON, D.C. 20024 (202) 554 2345
NUCLEAR REGULATORY COMMISSICN
- O .
This is to certify that the attached proceedings before the
) BEFORE THE ATOMIC SAFETY & LICENSING BOARD in the matter of; Long Island Lighting Company (Shoreham Nuclear Power
$tation)
Da.te of Proceeding: June 25, 1982 I Docket flu:::ber: 50-322 OL Place of Proceeding: Riverhead, New York wore held as herein appears, and. that this is the origina.L transcript thereof for the file of :he Coc:=ission., ,
. Ray Heer Official. Reporter (Typed)
< O.
- w Officia Y eporter (Signature) e 0 '
3 O
3
- _-