ML20236N260

Transcript of ACRS 331st Meeting on 871105 in Washington,Dc Re USI A-44, Station Blackout. Pp 1-160.Supporting Documentation Encl

ML20236N260
Person / Time
Issue date:	11/05/1987
From:	Advisory Committee on Reactor Safeguards
To:
References
REF-GTECI-A-44, REF-GTECI-PV, TASK-A-44, TASK-OR ACRS-T-1622, NUDOCS 8711160091
Download: ML20236N260 (184)

Text

{{#Wiki_filter:yg # g g _gn ~ ~All}',-l, M. ~ Olig \\ A_ p..w WitD STATES z y .N. .l,.,........... U. CLEAR REGULATORY COMMISSION ......................=........... 1 IN THE' MATTER OF:- DOCKET NO:' ADVISORY COMMITTEE ON REACTOR' SAFEGUARDS 331st ACRS MEETING 'USI A-44 STATION BLACKOUT. . Af J, j ~.. --t

LOCATION:',

PAGES: 1 0 :- ' Washington, D.C. 1 through-160 c; Y' DATE: x.._ -November 5,'1987- ....................w... .y ~ lk c) 39fM A0W@ *,"y9*NyQy hh' .Y'L

i..

l Heritage Reporting Corporation Official Reporters 1220 L. suest. N.w. i f. Washington. D.C. 20005 l } , m,.- 1 8711160091 871105 PDR ACRS T-1622 PDR l J

.- < O.[^ ik(ek M' 'i/ '.y ll L Sk iM. tin t- + ,) nj j y{,[.:l, :>.v.3 g s: .,3.. L t. ; 'H:-Q-t s ~y; n-7, 2-y g, . I L. '~ FUBLIC NOT1CE BY THE' ^ u;a; 2 i. UNITED S' TAT S NUCL' EAR' REGULATOR COMMISSION'S. .jj a.,.c);f 3 ADVISORY COMMITTEE ON. REACTOR SAFEGUARDS: '4 J SDM, OCMER 5 a 1W '^][' y I5; v 6 3 4 '7 The: contents of this stenographic transcript of oj - '8' th'e ' proceedings :of : the United States ' Nuclear Regulatory j i. 9' ' Commission's Advisory' Committee on Reactor' Safeguards (ACRS), '!10 .as,reportediherein, is an uncorrected record of the- ,.c (11 ' ' ) discussions:' recorded at ' the meeting held on the above date. L12 No : member of' the ACRS Staff and no participant at l-13-this meeting accepts any. responsibility for errors or 14 '

inaccuracies ~ of statement or data contained in thic

' 15 -transcript. ') i.: yg

p.'

e4 a. 174 i 1 18

19

'20 l 4 s

22

!~ E - 23 .71 24 .,M.. ' 25 4 1 t Acme Reporting Company I i an,, u..... ] _____...._.._1_____ - - _U

W,Q

• 1 l

1 1 ' UNITED STATES NUCLEAR REGULATORY COMMISSION i Y)

M ^

2 ' ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 3 [)- ) 'I 's 4 In the Matter of: y ) s ) ) 6 331st ACRS MEETING ) 7

Thursday, g

November 5, 1987 9 Room 1046 1717 H Street, N.W. 10 Washington, D. C. 20555 II The above-entitled matter came on for hearing, c. 1 12 pursuant to notice,Lat 10:30,a.m.

es

) 13 (,/ .l Chairman BEFORE: 'DR. WILLIAM KERR l l4 Professor.of Nuclear. Engineering Director',' Office of Endtgy Research University of Michigan Ann Arbor, Michigan ACRS MEMBERS PRESE'NT: MR. JESSE C. EBERSOLE 48 Retired Head Nuclear Engineer Division of Engineering Design l I9 Tennessee Valley Authority Knoxville, Tennessee DR. HAROLD W. LEWIS 21 Professor of Physics y Department of Physics 22 / -} University of California 'k Santa Barbara, California ()' f .24' 25 1 Heritage Reporting Corporation (202) 62s.4sse __.. _ _ _ _ _ _ _. _ _ _ _ _ _ _ _ _ _ _ _ _ _. _ _ _ _ _ _ _. __________.__ _ _____J

ifk ,2. 'i 1 'l, 'ACRS MEMBERS PRESENTo(CONTINUED:) c ;~ 2 Q L {M 5 L DR. CARSON MARK ~ 2-Retired" Division Leader j '3 l{ Los Alamos. Scientific Laboratory-Los Alamos, New' Mexico 'i I MR.iCARLYLE MICHELSON 5 Retired' Principal Nuclear-Engineer Tennessee Valey. Authority ) 6L Knoxville, Tennessee, and., Retired Director,- Of fice. for Analysis and Evaluatior t 7-of Operational Data: t U.S.. Nuclear Regulatory Commission: ?! L 8 Washington, D. C. i 1 ? 19" DR. DADE W. MOELLER b Professor'of Engineering in' Environmental Health' 10 Associate DeanLfor Continuing Education School of Public Health-R 11 Harvard' University' Boston, Massachusetts 1 t. 12' ^ fr MR. GLENN Ai REED '} 13 Retired;. Plant'Mana$er L . Point Beach Nucle'ar'Po'wer Plant. li 14 Wisconsin Electric. Power Company j 1Twb Rivers, Wisconsin j / 15 ,s a, <r\\ DR.~ PAUL G. SHEWMON* 16 Professor, Metallurgical Engineering Department Ohio State. University-j 17 Columbus, Ohio j g 18 DR. CHESTER P. SIESS Professor Emeritus of Civil Engineering 19 University of Illinois Urbana, Illinois 20 I DR. MARTIN J. STEINDLER 21 Director, Chemical Technology Division q Argonne National Laboratory J /- 22 Argonne, Illinois ( 23 MR. DAVID A WARD Research manager on Special Assignment ,. () 24 E.I. du Pont de Nemours & Company Savannah River Laboratory 25 Aiken, South Carolina i Heritage Reporting Corporation ne u. .I

l '3' I ACRS MEMBERS PRESENT (CONTINUED:)

Q 2

MR. CHARLES J. WYLIE Retired Chief Engineer 3 Electrical Division 'ggg Duke Power Coinpany 4 Charlotte, North Carolina 5 ACRS COGNIZANT STAFF MEMBER: 6 Raymond Fraley, Executive Director 7 NRC STAFF PRESENTERS: 8 Mr. John Opeka Mr. W. Minners ) 9 Mr. Brian Sheron Mr. Jerry Mazetis 10 Mr. Marchesa 11 12 33 14 j 15 16 17 18 19 20 21 '2 khI 23 lll 24 l l 25 1 1 Heritogo Reporting Corporoflon a n..<m l

4 1 'PjR OL.C,E,E_ D I N_ G_-S ~ j. 4 p/ -x; h_ l 2 ' CHAIRMAN KERR: Station. blackout. Mr.-Wylie. 3 MR.'~WYLIE:. Thank you, Mr. Chairman. ?/"i f.d

4~

The backup'information'for today's meeting-is.at-q 1 5 Tab 3 of the' folder. The purpose of the meeting isttoLhear } 1 Ev ~ j . - 6 Presentations'from,the staff and the NUMARC representatives l in' regard to.the proposed resoluti'on of USI A-44L Station' 67 .7 8-Blackout. l 9-If-you recall,-during our May'ACRS meeting we had I 10 ; presentations by the1 staff'of NUMARC regarding the.resolu- .f gj tion lof USI A-44. And we' wrote a letter on' June.9th of this

12

- year, and' in the -letter we did not recommend; the issuance l [h - .g3 of-the final rule'at that timerbut: recommended the staff .y/ continue-.to work with NUMARC on:the technical' aspects:offthd? 14 q 15 industry initiative, t' hat'iffby Sept $bber of 1987 the staffL hadnotreachedafsatisfabtbr'resolutionatNUMARCthatthey 16 .j g7-proceed.with the lssuancetof the rule. -t 18 Todap we have representatives of'NUMARC.and the. stafftotell'uswhat}astakenplace,giveusthe' status 39 20-report of the resolution and other.information as appropriate. 21 The agenda is in your folders and-Mr. John Opeka, i ,l 22 I belieVe, is going to present the NUMARC presentation at this time.- So I'11' call Mr. Opeka. l 23 MR. OPEKA: Good morning. My name is John Opeka. I ) 24 I'm an Executive Vice President for Engineering and Operations 25 Heritage Reporting Corporation <=>m... _______E______

d 5-i- L I-for Northeast Utilities and'Qhairman of the NUMARC Station (,r Ns 21 Blackout Working Group. And'with me this morning I have Al ] t 3 Cutter who is the Vice President of Nuclear Engineering De- ,6 t'~') 4 partment for Carolina Power and Light, and he's a member of 5 the NUMARC Station ~ Blackout Group. And I have several people 6 from the NUGSBO organization which'is_the Nuclear Utilities Group on Station Blackout whic'h has been providing technical 7 8 support for our committee. And they are here to help me out, 9 to respond to any questions you might have on my presentation. 1 1 10 I do not have any slides today, but I did pass'out 1 11 a handout'which I will be using as a guideline in my presen-12

tation, c

] l I'%) ^ 13 And just as a reminder, since the last' presentation i .!s. 14 that we had made at the ACRS meeting on May 8th, we had be-i 15 fore that on June 17th, had submitted comments on the proposed - ) 16 Rule for Station' Blackout, and as part of those comments we 1 17 committed to four initiatives. NUMARC committed to imple-18 menting four initiatives. And since that time, and also 1 19 since the meeting in May, we have been working through l 20 various public meetings with the NRC staff to come to some 21 agreement on how we should implement those four Station 22 Blackout Initiatives as well as meet the objectives of the ) (-)) '1 \\_ 23 proposed rule. j (RJ~) 24 And we have a document called "NUMARC 8700." Here 25 ia a draft copy. It's fairly thick. It has the methodology i Heritage Reporting Corporation ) l om m-me

6 1 and guidance that we would be following in order to achieve 2 those four initiatives as well as meet the objectives of the 3 proposed rule. g 4 With our various meetings with the NRC, it appears 5 that we are close to resolution on many of the concerns that .6 we had with the proposed ~ rule that was submitted last year L 7 and' basically a technical agreement with many of thewe issues 8 that we had and we-expect that we'-ll come to final resolution l ~ 9 on the wording very soon on those. 10 Two other things as a result of the last-ACRS meet-11 ing is that industry has--consistent with your comments-- 12 does seek a method of addressing the station blackout issue f 13 with a minimum of NRC and licensee resources. And that's the 14 whole purpose of the initiatives and also working very close 15 with the NRC, will be coming up with the specifics on how to 16 meet tite proposed rule. 17 And also consistent'with your comments is that we 18 don't want to just address the issue of blackout but also 19 want to address other generic issues such as the reactor 20 coolant pump seal issue and the emergency diesel generator 21 reliability issue. And those two are referenced in the 22 NUMARC 8700 document. 23 I'm going to page 2 very briefly. Some of the lll 24 concerns that we had, and I'll just go over a couple of 25 these. We have a proposed rule that was written over a year Heritage Reporting Corpora +1on <na.u.au

7 l 1. ago was that if you were not getting any credit for usefof 2 on-site or off-site backup AC power sources, and also we 3 weren't getting credit for any. severe weather procedures, e 4 and it appears that in both cases n'ow we'll be getting credit 5 for backup sources and also use of procedures in certain 6 areas. 7 We. were~ very concerned about the coping demon-1 8 stration being open ended and that has been eliminated. 9 And also there was concern about other regulations 10 impacting a station blackout issue and we feel that working 11 with the NRC that industry has come to agreeable resolution 12 of those specifics. 13 DR. SHEWMON: Mr. Opeka. 14 MR. OPEKA: Yes. 15 DR. SHEWMON: Could you define a backup source? 16 I take it this is not the usual diesel but certainly are in 17 tech specs'and diesel generators. 18 MR. OPEKA: It could be one of the diesels but 19 what is more referred to is that if you have a fossil plant 20 or a dam or a gas turbinethat's not tied to the emergency 21 buses that could be used as a backup power source, that's 22 what we're talking about. Getting credit for backup AC 23 power sources. lll 24 Then the next page just has a brief summary of 25 the four initiatives that we agreed to back in June. This s Heritage Reporting Corporation non m...

8 1-the industry. And again, very briefly, Initiative 1, what lk 2 we agreed to do then was use the matrix that was in the pro-3 posed rule that the NRC issued last year and determine where g. 4 our plants are as far as coping capability. That is, look at 5 the emergency diesel generator reliability, the configuration, 6 and the off-site, power reliability and configuration and us-7 ing those parameters, determine how'long;a plant would have 8 to cope in order to not provide undue risk to the plant. 9 And any plant that was found to be in the eight-10 hour category we would either procedurally or through modi-11 fications get that plant down to a four-hour category. 12-Initiative 2 was to come up with implementing 13 procedures that.each plant would know how to cope--have a 14 procedure on coping with a station blackc.nt event, restoring 15 of AC power and also preparing for severe weather conditions. 16 The third one was to address the cold fast starts 17 of emergency diesel generators such that any plant that had 18 a requirement to do cold fast starts of their en ergency diese] 19 generators on a frequency greater than once per refueling 20 to that get changed so it wouldn't be any greater than that 21 frequency. 22 And then Initiative 4 was to start a monitoring 23 program of the emergency AC power unavailability. lll 24 Now, the status of those four Initiatives is that '25 Initiative 1, this is the one using the matrix to find cut Heritage Reporting Corporation (202) assasse

77_,-- __ _ .J 9^ l( l I where1you-are as far as coping category. We' had had a num-l 'rh l ~ 2 ber of--we developed a' procedure. We had a number of work-l shops with all of the. utilities.to make 'ure that they were 3 s O'm I i ) i 4 using the procedure consistently and then we verified the .) s data after that'was done. That was corapleted. But in the l 6 meantime, based on comments that the NRC had received from i 7 us and other members, thep'used that information and modified 8 the matrix, such that the work we had done was not really j 1 ) 9 valid any more. 10 We had decided, in fact, NUMARC had a meeting on l l 11 October 22nd to approve a new Initiative called 1-A which } a 12 basically would do the same thing the initial Initiative did iy /"N 4 ( jf 13 but using the.new NRC matrix criteria. So that is another i 14 requirement that we--another Initiative that we as an indus-15 try agreed to abide by. And that work will be ongoing in the 'i c 16 future. 17 Initiative 2, talking about developing procedures 18 .on coping, severe weather and also recover from AC power. l 19 We've been working on that and those guidelines and require-20 ments are in this NUMARC 8700 document that we have put to-gether and have been working very closely with the NRC on. 21 i /~4 22 Initiative 3, the one talking about the cold fast l V 23 starts, right now the situation is that there aren't any I~). 24 plants that either have the requirement or--all plants (- c il 25 either have the requirement or are in the process of making Heritage Reporting Corporation p m u..

lo i i I . technical specification changes such that there won't:be any [ /"N t 2 pl' ant that has'a' requirement for cold fast starts at a fre-3 'quency that'is greater than once per refueling. So with that N' 4 Initiative basically is near completion. 5 And Initiative 4, on. monitoring emergency diesel 6 generator unavailability, IN.PO agreed to do that as part of l 7 their Performance l Indicator Program and'that is being done' 8 now. So that's completed. 9 Now, another. Initiative on'the next page, we agreed 10 to implement as an. industry and this was agreed to unanimous-11 ly at the October 22nd NUMARC meeting, was that we would also 12 agree to do a new Initiative called Initiative 5. And at the 1 /- 13 last ACRS meeting there was a concern that even though we had 14 an Initiative where we would determine what coping-category 15 you are and then take action to get those plants into a four-l 16 hour coping category, there was still a question on how do 17 we know that each plant can cope for four hours. And at that 18 time I indicated that we were considering making that another 19 Initiative. And on October 22nd the industry has voted to 20 make that a new Initiative, and as a matter of fact, we have 21 been working to put together a simplified coping procedure i l ,r3 for each of the plants to follow such that they can determine f 22 (_/ l 23 whether the plant can cope for four hours or greater. If l i( ) 24 they can't, what actions they would have to take in order 25 to get into that four-hour coping category. i 1 Heritage Reporting Corporation l l o n man lE_______--- J'

.m.. .- q. g. y 1

t. m

>4 11a

u

_i a 1-IThat! simplified methodology and-detailed procedure Tj ._. y V~Y j EM 2' Iis.aga'in part oflthis NUMARC 8700 document. l l [ 23- [ CHAIRMANeKERR: Why do-you: refer.to this-as a. 1 S- [' f.4 ! simplified; methodology? ? 5; I

MR. OPEKA

';Well{ it's s,implified such that'we

^.. 't / + -: wanted ito.use-asiinuch of the' exis'tirig. data as possible. r.In 1 7. - other words_,'ube?. technical'specificationfdata, FSAR-data,: J e 7 such that'you don't have'to spend a'l'oU o$ moneyjdoing'a-f I -8 .as 9 .detailyanalysis toydetermine'if.youlcan cope. I 10 For example,2 there's really five: factors,thatfwe're , i d 11 considering on coping. Condensate inventory for decay heat; 3 d

12..

removal. .'We came up w,ith a simplified procedure.thatisays '! lj 13' ..if you have a certain megawatt plant you can determine'.how y- } ' '14 much: heat-you havelto'take away in four hours assuming.that 15-the plant is operated at 100 percent' power for one hundred: ~ 16 hours. 'And then shuts down. You would go to tech specs to- -l 17 find out what your tech spec requirement is'for condensate l i 18. storage tank level, use that number, and then use some' simpli-J ' l' _ 'l l 19. fied cookbook procedure to find out whether you have enough .l.j a 20 water-or not for four hours. If you don't, then there's an 21 1 option to look'at. backup sources of water that don't require' ] 1 .p " 22 .use of AC. power to see if you will meet that four-hour re- .y 23 quirement. l J .. tO 24 That's what I mean by simplified procedure. In a %/. 25 cookbook fashion it's simplified, but it's based by I think H Heritage Reporting Corporation (to2162s 4ase

12 1 .a good detail analysi's. .h 2 CHAIRMAN KERR: Thank you. 3 MR. MICHELSON: Does'the' procedure include verify-O 4 ing whatever valves that are to be operated are indeed opera-5 able under the blackout condition or-- 6 MR. OPEKA: Yes, that's correct. There are three 7 areas we were very concerned with. The condensate water-8 inventory for decay heat removal purposes, the battery capa-9 city for four hours, the~ compressed air capacity, the effects l l 10 of loss-ventilation and containment isolation. So in a 11 containment isolation case, if the valves failed in a closed 12 position or were found to be locked, or were three inches or 13 less, we felt that those were a good enough criteria for 14 meeting the containment isolation requirement. But there are 15 other valves that are not on those categories. We might 16' want to ensure that there is a procedure that says that within 17 a certain time frame you want to go through a specific valve 18 and c_;se it to maintain continued integrity. So that's the 19 type of thinking process that we went through. 20 MR. MICHELSON: To remove the water you are going 21 to have to open some potentially closed valves or whatever. 22 MR. OPEKA: That's right. 23 MR. MICHELSON: Or valves that might have failed h 24 in the closed direction on the loss of power. 25 MR. OPEKA: Right. And if they us DC prder, then Heritage Reporting Corporation non m.m.

13 I we took that into -account in the battery capacity situation. 1 2 If they used air, we took that'into account, compressed air 3 analysis. O 1 4 MR. MICHELSON: Generally are you trying to protect l i 5 compressed air in order to maintain operability? 6 MR. OPEKA: Well, we want to make sare we have l o 7 enough compressed air for four hours to operate all the 8 necessary valves in order to keep.the core covered. That's E 9 basically--and if we find that we don't, then we would look 10 into areas where we'could shed some air valves so that we 11 can have that additional air capacity for the needed valves. 12 DR. MARK: At time zero you need water at a cer-13 tain rate. Four hours later you perhaps only need half as 14 much. Do you take account of that drop off in the require-15 ment for water over the four-hour period or do you just run. 16 it full. force throughout? 17 MR. OPEKA: No. We take into account the decay-- 18 that the need for the heat source--the decay heat decays in 19 a four-hour period. 20 DR. MARK: And your valves then, as Carl raised 21 the question, have to be capable of adjustment from time to 22 time? 23 MR. OPEKA: That's right. -l 24 DR. MARK: Spin things out to cover the four hours. 25 MR. OPEKA: That's right. Heritage Reporting Corporation (202) 42s-4sse

H a , m. .g y- -

14; 1

,..,z.. ,L q ~" 1 i .-MR. L EBERSOLE : Can you depend.onirejection~:of decay, 1 %yy, g /t, .2. energyisoleiy'byfthe process of evaporation'on PWRs for the ~ .p. 3 -secondarie s ?! 7 >p ^ On PWRs'?' + Y 4 JMR. OPEKA:, f, : -5 MR.- EBERSOLE: 'Yes. That's -- generators.- q q tt 6 .MR..OPEKA: ThatL's. righti. We are using ! thef atmos -: 1

)

.g =

'. 0-i.] pheric dump system.-

R i 4. ,_1. . i 'f ~t' 8.' MR.',EBERSOLE: "It'Jtakes"ailot'less water. l '4 MR. OPEKA: Right. The atmospheric dump.s'ys' tem. 1 9 o .'j 4 10-MR'. EBERS' OLE: 'Btit what Niil_you do for the boilers I Lil 'which are pulled up in their containers?' ,. i. 12 =. MR...~OPEKA':. That's right. ] 1 -.] i 13' MR. EBERSOLE: :You can't' count'on evaporation.for-. s_.. l '14- -those.. 15 MR. OPEKA: Well, in a PWR, the condensate storage. j 16 tank. level would-be dependent'on using the' water from:the. i 'i 17 auxilliary feedwater system to the steam generators.and then jt 18' the steam g'oing out'through'the atmospheric routes. 19 MR. EBERSOLE: ' That takes a modest amount o.f-power. .20 MR. OPEKA: On BWRs it varies. If you have an

21-isolation condenser then you are concerned about'the makeup

) 22 water to the isolation condenser. 23 MR. EBERSOLE: That's also a modest makeup power ] l O 24 capacity. .V i 25 MR. OPEKA: Right. If you have the RPCI System q l Heritage Reporting Corporation m u.4 I

m . r

t. (

N l te m 15L U ~, ,- A I u l + ..i 1-then the makeup would be--the water would-be'goingLto the ,yy' 1 vessel but then the water:would be. going ~to:theLSRVs'back1 l Jsf f2-L J

3 Ed.owni to the ~. torus, - so-that

would ' e thel.requirementEthere.

j J L /~ y <V~ j MR. EBERSOLE: But then how' are; you; going' to get the w IS-heatiout of the containment?- i y J 6 MR. OPEKA: ;The' heat outi..of the' containment, as: o ^ 4 _f, 7 farLas--we don't,think that's,a problem-because we'think that 3 ,E.-3.A . f M.i 8- ,.the LOCAaccidentLis more-' severe 3temperaturewise as a: station j x 9 . blackout accident.fortfour hours that we'didn't have.to- 'h t' ' t . i 10' really analyze temperature effectsIon the containment'basedi q ~ d 11 on a, station blackout.because the LOCA.provides,m th.1at proper 1

g n,

12 analysis.

.I think that's-the case, j! ~ ,[/[ .13' 'MR. MICHELSON: Are.you saying'that;.you are going s g f 14l .'to store the heatin the~ containment:for the four' hours for- 'q 15 the boilers? 'MR.'OPEKA:' That's,right.< l 16 17. MR. EBERSOLE: Youi are going - to resid'e ' on. just -stor-i 1 9 18 age of heat in the- - .] 19 ' MR. MICHELSON: It'gets pretty hot in four hours. 20 ' MR. OPEKA: Yes, but we believe that the LOCA '21 accident provides an envelope over that four-hour period. o '22 MR. MICHELSON: You really don't need to invite' 4 .qj 23 heating up the containment and everything that much if you f 'l 24 don't'have.to. 25 - MR. OPEKA: What we're trying to show though is i i Heritage Reporting Corporoflon s (202) s2s.4ase q i ri__________________

m-e n; % - 31 a + n A 1/; '16. 1 A l f i f ) i l-that?we havela;four-hour coping capability. (And put, emphasis jr ; ] .L, g .2 : 'on' improving;our reliability /and availability _of our.off-site w(r ws 3: -power: sources and'on-site emergency' power sourcesito prevent-yqy., _ c g^ 4' a blackout rather than emitigate'one..

I

%c. ~ 1 5 .MR. MICHELSON: "You take that' approach and you'are D 1 ~ 6i

goingfto heat-up the buildings significa'ntly from the torust q

E '7 if 'it's' uninsulated.L And you "are i' cluding; that !in your n q < l '~ 8 lwhole analysis. ) Depending upon?RPCI.foriHPCI-to operate. 7 o r., v. ~.. 1 u d 9 It's right'.'next;to the torus. 1 .. v i 10 ( MR; OPEKA:.' That's'right.;_'As far as the--there's: 1 11 three dominanb; areas as:far as temperature is concerned.' A: + - -

12 :

lossiof ventilation would.be'RPCI HPCI~ room, steam tunnel,- 13' those are PWRs,-and auxilliary heat water. pump' room and' heat ~ ~ - 14 .up. yards that'we would have.to ensure.tha't with the tempera, i i . 35' .ture that'we expect to be reached after four. hours was less 16.-

.than'what'any of the. required equipment-that'hadzto; operate' i

,17 is:really/ designed.for.- '18 MR. MICHELSON: -A boiler has just one. pressure ~ 1' 9 vesssel to. supply steam. Sticking of any SRV will depres- ] 20 surize to.the point where the steam driven machine' won't 21, work. Are.you counting on SRV operation to keep the steam 22 pressure up to.run the HPCI RPCI complex? 23-MR. OPEKA: We were, yes. .q 24 MR. MICHELSON: Well, what if one sticks? You _g. 25 lose your steam pressure and then you must resort to motor-Heritage Reporting Corporation (202) usasse -_.l-.._--._...-

~ ~ - .. - ~. i ',,: r f'- t 1 , i vi! i + ci! .., a,. 17; -c 4, y n o ,3 .( y, .,. !lr -1: ? driven : equipmenti: to; get ' water lin. And:that steam-driven- =d W

l y V[ 4 s.

y l.y, T2 , equipment is. notoriously unreliable. m .w ..c N ;MR..OPEKA:- We11, we don't assume that.onelsticks' q c E ';, y gNI -4L 'open. <m l 5! MR.:MICHELSON: Well,'I don't think'theirLreputa, .q o a- .q 64 tion'is worth that--well,jthat's something:that.can'be: picked 4 7. up. You're; not assyming' any singl'e f ailure Jaf ter having lost' ] ' 8. : some of your'stieam power? 1 3 e 9. MR' j0PEKA: b Th.at 'srighti.- \\ 10. .MR. EBERSOLE:e No singlen.failuriiafter loss of' { ,.a 11 AC power? i' yt g 12' JMR.cOPEKA:- Except the--yes,-that's~right. Except' sx 13 that if we have multiple sites.and we.want to take~ credit" .14 .for an emergency diesel generator:on a' shared basis-- 15

- MR. ' REED :

Well,in the transient,. Jesse, that '.16. . ' occur s, ~ would an SRV be l'ikely.to operate?.. j .l i CHAIRMAN KERR: _ Gentlemen, mayLI'sug[ rest if'we are: 17; 1 18 going to go :into this much detail on an issue which is ob-1 p 19, viously important, we ought to schedule a subcommittee be-20 cause I don't-think we have time enough to trace out each of ] 1 1 21 these scenarios ir detail today. .l 1 t 22 .MR. EBERSOLE: Well, Bill, these things hang on one 1 ..'23 . hook, you know, here and there. 24. CHAIRMAN KERR: Jesse, I'm not denigrating the im-25 portance of the issue. I just don't think we have time to i Heritage Reporting Corporation an> n. 1 8 8 mu - 1

y,3-. 8.

V

.) M c

18 N

1]%{5-s 3 -s 0 p g." ,n .m ' 1: fexplore it inithis'much detail'today.. .3 t .\\.) 'd l p ti, J21 MR.LEBERSOLE: Okay. <3' MR.'OPEKA' But'that was. basically'my. presentation.^.- 'I ,[ DIt Lwasj a. status. report on what.we 've done - since-. the /last ; ACRS-l ..l J4. 1 3 + .e 5 .i'n May. I' worked;very. closely;in open; meetings, public meet-

6~

ings,;lwith the'NRC'in trying to come to a? reasonable resolu-l .r-7 ' tion on.the' station :blac'kout! issub',jarid we feel as an lin'dustry + .8. that we have a' package that; properly addresses:the revised ~.. ~ q ~c e < c., s .4 - g

I

9..

. proposed rdle of'the.NRC in that it'sr anL. adequate resolution. l m -10 of the' issue.-- .g j .e 11 CHAIRMAN KERR: Questions?: -{ 12 Is it your view that there has been ::a' significant. + . ' 13 reduction or will=be a'significant reduction'in' risk as a- < 14 result.of this' set of. Initiatives?- (! i r 15 MR.'OPEKA: Yes..But I' don't'know'the specific 16 number. I know we'did some analy'ses on that. Yes. q . CHAIRMAN: KERR: There are no' questions?- Thank ~ ,j 18 you. j 19-MR. OPEKA: Thank you. =20 MR. MINNERS: Well, I just want to say that I 21 think our staff and I know that the industry NUGBO staff 1 1 22 .have' worked very hard to come to a resolution of this issue u ~ a 23 .and we appreciate that hard work. And I don't think I really l 24 have very much to add to what John said. 25 We have come to an agreement on the resolution of Heritage Reporting Corporation l n m u.4 i n

19 l' l 1 this issue. As I said, we have worked pretty hard~doing this, r~ 8 -') 2 We are now in the process of what we think is just a pro i 3. forma NRR and OGC review of our package which we're going to (~') b 4 submit to the EDO. That is based on our continued work with 5 NUMARC. We are going to go to the EDO and say industry 1 6 agrees with us. And we agree with industry. And part of 7 that is that they have provided Initiative 5 which gives a 8 coping assessment which was really the heart of the written i 9 rule. They'have developed a very detailed document and have 10 worked with their people and I think that's a good idea, 11 Then the industry will have some very detailed guidance that 1 12 they developed themselves which they can say that the NRC (9,/ 13 agrees with. And that document will be submitted to us and 7 14 it will be referenced as part of the resolution of A-44, 15 I think the changes that have occurred in this 16 process--we've modified the rule a little bit. Originally 17 the rule had a Rule 63 and some revisions to GDC 17. And I 18 always thought that was kind of redundant and the industry 19 said that gives the implication that it's a general science 20 criteria and therefore all of the DBA kind of things apply. 21 Which isn't what we meant. We don't have single failures. ) 22 We're doing a little different EQ. We have a modified QA ,3 %) l 23 program and things like that. i l f( ) 24 So the rule now consists only of 5063. It doesn't 25 have the GTC. Heritage Reporting Corporation ] nm m. a _-_-__D

? jj; i ( n 20. J 4 y

y'

• [N

'l- -Complementing that,Lwe?are going to. issue the red'- d 1 .t ' Q-F ; 12.[ gu'ide'which provideslthe' details.."And_that's' going'tolbe.a 3 red' guide which has all tne' details and we'll also' reference ~. s ) 4' Lthe appliciablef s;ections of. NUMARC - 8700'. Land so.-they.'are) + .s. identical. And as you have in]your,packag'e there,.there's J 6- .a comparison of the Table of Contents of the.sectionsLof;the 7 red guide with the' sectionsin'NUMARC. So'we:have a corres ' if .J s 8 -pondence.,, w ? T$1espresent s'tAff dchedule'is._t try.to'get it'out" 10 this month to the EDO. We'h've a few editorial word changes .j a 7 H 11 'and; word' engineering going'on-I; understand. 'And,we'lllget n ( i 12L it.over.to.EDO--we'.11 give.him a month to concur and. send it 2 - to ' thelCommission _ and, hopefully _we 'll be iabl'e - to -have the - 13-14 rule' issued'and b'egin implementation in March. So that's .15. -our sched'ule. -l '16 CHAIRMAN KERR'; Any. questions? ?!7 MR..MICHELSON: Yes. One. Could you tell meLjust ] 18 briefly_the extent to which you are comfortable with the-19 assumption'that the relief valves will never stick open' , j - 20 during the four hours of operation if they are called upon - 21 to. operate? i _ 22 MR. MINNERS: Well, I think-- 23 MR. MICHELSON: They have to cycle. 24 MR. MINNERS: We're not doing a'DBA, okay. We 25 are doing a probability assessment and the probability of 7 Heritage Reporting Corporation <mnu 'c

.21 1 l 1 losing off-site power and losing the diesel generators and 1 I l <~c 2-then having the valves stick open, we think is low enough l l 3 that we don't have to worry about it. I /N, 4 MR. MICHELSON: You've done the calculations and 5 are satisfied with that conclusion? 6 MR. MINNERS: I don't know if there's a specific 7 calculation. 8 MR. MICHELSON: Well, you inferred there must be 9 a calculation. You are saying'the probabilities are all J 10' this low. You must have done some kind of--you must have 11 some basis for making the statement. a 12 MR. MINNERS: Well, there's a calculation, and I .,, - ~. 13 don't have it with me, that assumes what the unavailability ( s/ ~ 14 of the coping capability is. Yes. There were some assump-15 tions made there. ) 16 This is not something that's come up recently, 17 Carl. 18 MR. MICHELSON: No, no. 49 MR. MINN1'RS : This has always been the staff's 20 plan. As we went t hrough the public comment period and 21 we've gone through two ACRS reviews with this, and so are g you bringing up a new issue or is this-- 22 \\_/ 23 MR. MICHELSON: No. I was just asking about-7) (j 24 MR. MINNERS: Just asking that old question. Okay. 25 MR. MICHELSON: It's still a valid question I i ) Heritage Reporting Corporation j i ( m u..uu l

j 22 l~ l think. ("T 2 CHAIRMAN KERR: Mr. Minners, I have a question 3 which I think is not directly related to this resolution, 1/~y l t '/ 4 which seems to me is a good resolution.- But do you have.an. l 5 ~ idea of the likelihood of-loss of AC power for more than ( i 6 four hours once these things are in place? Has there's.any- ~ l 7 body made some sort of estimate of the improvements that 8 have occurred or what one might give as a number for-- 9 MR. MINNERS: I don't have the--s i 10 CHAIRMAN KERR: It's okay. If it exists, I can i 11 'get it. 12 MR. SIRKUS: Al Sirkus. The staff's technical im[) 13 findings and detail on those kind of matters are in NUREG ~. - 14 1032 and that has been out for public comment before and I 15 think it's been down to the committees and/or subcommittees. e 16 A lot of detail and that type of cross referencing of the-17 relative effects. i 18 CHAIRMAN KERR: I wasn't asking for relative 19 effects. I just wanted to know if someone had made an esti-20 mate of the likelihood of loss of all AC for more than four 21 hours. 22 MR. SIRKUS: Yes. It's in 1032. NUREG 1032. <-u) '( 23 MR. MINNERS: And that depends on your--which con- [~) 24 figuration and those kind of things. s_e i 1 25 MR. EBERSOLE: When I look at the entire position Heritage Reporting Corporation (202) 628 4888

,/c,; s il .., 4 , > {.#_ ql 43 9 234 m L '. lt w. fn ll that?you takelon AL44-fwith these graded blackout = time' inter-b val's--I--think.that'was a-smart piece ~of'bu'ainess.because it-v 12 L y.- l 1.1 3 took.into'~ account the~ spectrum of reliability o'f power of ) >f 44. - various plants which is -a very broad' range.. 5 . When--you collapse'all these downLto four hours l l o ~1 6- -it strikes meit has some'of the elements that mustDhave.gone l 1 7 into _ the engineering decision' of the (Titani'c,, like' putting; i ~ -8 . bulkheads up'to the deck., ~ ~ .j 9. MR. MINNERS: Mr. Ebersole, I can't hear the a c y,i j p i s 10 question. j ..n 11 MR..EBERSOLE:-1 I'm,saying-- . ( j 12 .. CHAIRMAN-KERR: He hasn't asked.one'yet. -i 4 13 MR. EBERSOLE: I'm saying_the net ~ result of this: 14 ' resolution.is the collapse,.'these. extended intervals from 15' lthe. worst of our plants down to four hours for everybody. 16- "Right?- '17 MR. MINNERS: I don't know quite what you are id !8 saying, Jesse. I think as John has said, there are some= 19' plants that may have to make changes to ge't'from an eight-i 20 - hour category to a four-hour category. This is not just a 21 paper study, .22 MR. EBERSOLE: Well, so you say, well, we can look l s

O'.

23 at.the changes. 24 MR. MINNERS: What are the changes? .25 MR. EBERSOLE: Yes. q l Heritage Reporting Corporation j m.a. _ b

m:~ ; v (. a, h, I I \\' e, .qc' ga 1:

MR.' MINNERS

- L I ! don ' t k'now...

Maybe John-knows them. ~ .f q M LV [2" i t I[ don't know~any: specifics what people'have!done. ? m yy. U, 3; L -MR.~EBERSOLE:.Will we.in' fact.get al distribution-- 'q a 'V I4-LIJunderstand,;.'asDBill's' aid a while ago that we'll;get the 5.!

details. l Will',we:-understan'd: the' details.:or get the details

~ O 6? of:these~ changes? LI'm afraid it'.sl going to' fall 1intofa' big, G' L* 7 black = hole. We'llinever really root out.the details of the-W '.' 8 ' cha'nges that have to;take place. 9: . MR.~ LMINNERS : Well, we intend-to review what'thet e ./ 3 ) 10 ' utilities hitve; d'onel and they., are going [t'o come back and say.; 1 11 .we're in;a four-hour category.and we're going to ask the. 3 . e.' 12 .questi~on,.~well, how'did you get.yourself in that category, l13 and'they'll going to say, because I've'got this severe wea- ] 14-ther, and.I'.ve got this switch crisis. We'.re_ going.to,.you l 15 ~know, review that like we do. 'i 16 So,.it seems to me that if you'are-willing to J 17 accept.that there are some. plants now that are designed to j 18. a four-hour category, okay, it certainly must be possible to 19-take any plant that are eight-hour categories or greater 20 and redesign them and-make them four-hour plants. 23 MR. EBERSOLE: ~All right. And you intend to find 1 these pinch points on time and fix them? / 22 MR. MINNERS: Pinch points on time? .23 l i ( '24 MR..EBERSOLE: Yes. Where you say something goes 25 bad in less than time X. You will fix these, particular Heritage Reporting Corporation i j cmna. l

w s,. 4 l ,.)' -l. l

25.

(..; 2 b,.[ g. '.1 Lfeatures,of the' plant. p 2l' MR. MINNERS:.LThat's part'of'the coping-- h L - E ,4 f, i3~ 'MR.' EBERSOLE: Yes.' Part of the coping.

y s

M-14f MR.:MINNERS: -- demonstate that we can indeed-- d 'S

MR. EBERSOLE:

In the ' essence of L the: coping analysis '6 I. don't knowlanything about.'this resolution. 7 MR. MINNERS:- We're not doing<aiPRA on.these plants, 4; 8. okay. 'We'reLdoingLa^simplifiedjp'robablistic method'on whichs 9 we have looked at;it in-detail,;okay, and then we said, hey,. ~ y if' you have Ithese certain* fea6ures,- this ; kind of1 a' switch 10 -

t.,

y. t card' and~ this kind of weather ' frequency and.this kind.of-- 11 tt 127 that puts youJin a,certain category.- So.it's[a rough cuti x, n.

[

13' offthings.. W'ecare'not doing a, detailed PRA. Alth'ough the? I'4. overall goalffor M 44 was to have core melt' frequency'or' 15-core damageifrequency, whichever term you want.to use, be 16 about!10lto the.minus 5 per reactor; year. -_ Because..'of the ^- l8 17-simplified method', there may be some plants that h' ave'a -18 higher frequency. The intent was to have an industry average 19 to the'10-to the minus 5 after we got all the changes in: 20 of the. coping capability. So one plant may be a little - 21. higher but the other plants will be a litt?a lower and on W-22 the average we're okay. And that comes about because'we've N )^ 123 .got the/ simplified method. And when you fall into this I 24. matrix, ~ you could be at one edge of the box, okay. You may.. 25: just get into the matrix. ' Heritage Reporting Corporation (202) 42sases Q=:-_-_--____._-- i

u -F ,(, ,q 3 1 TI?t ' . 2 6, 4: r; n. ,y 'l m s 1

11 lMR.1EBERSOLE:: 'On this core-damage potential,;it's:-

7 '7,- '

2:

also containment' data,Jpotential coincident with'it, because'

.i. 9.. N "the heat--the absence or; rejection of containment will"b' ring- .e L.) ~ o 4: .thatlright on. .As.a matter of fact,.that might.precedef ~ '5 core damage. 6 ' CHAIRMAN'KERR: 'Mr. Ebersole, whatlis.it that you .7 would like to see that you don't have?- ,Bk.ll,alliI.cansay.isI'.m= blind ^ 8' s, MR. EBERSOLE: i.. i 9 until. I..see' the copin'g L studies as tio whether this resolutions i 10. is - what.it 'should be.- 7' i g' 7-4 4 7 i o .v 11 CHAIRMAN-KERR: And the coping-studies-will be 1 12 : made 'available to ' the staf f ancT[ presumably.if we. want tol'see. 13;

t. hem-- is.there.al time schedule'for. submission at this point?

14-MR. MINNERS: Yes, there is. And'I' forget what ~15: the numberLof;that is. Al can'look it'up'if you remember ~it. '.It's probab.ly.too.short a time.- You always'put'the ninety t 16 17- 'or--120-day-kind of thing.. Do you. remember.what it is,,Al? 18. CHAIRMAN KERR: Well, you can get that for us. 19 MR.-MINNERS: Sure. He'll find it.for you. 20 MR. EBERSOLE: Until we see the patch work to.fix 21 this, I don't know what we've got. ,j l-1 1 22! MR. MICHELSON: Do you know offhand how hot the n -V. j 23 suppression chamber gets after four hours of isolation with ) ) 24 no cooling? Do you know roughly the number? I-just ( 25 wondered how effectively the relief valves can still discharge { I l Heritage Reporting Corporation { j cman E i_ _n____._____.__.1_.____________

1-1

1

4 27- 'l DJ Ri q .. e,7 .1; to;thestorus-after four hours of no cooling.

That 's pretty T/"T a

2

much determined lby the water' temperature.

S ~ 03 .MR. MALONEY:-,It's-consistent with what the-- my_ .n .L4- .name is. Steve lMaloney. The analyses that I've seen on,thisL , 5; ^ issue"shows;thatyou:donbthaveanycool-instabilities 6: over the four-hour ~ period, ca. cool, heat is not an issue. J, 7 MR.~MICHELSON: I can' agree with'that real quick 8 if I; knew what temperature;we,were,; Talking:about. i

9 MR' MALONEY: ' boblinsthbilitynumbersthat'are H

10 ~used range from 20.0ito; upwards toi.220, degrees- -? .e 9 m c; s 11 i MR'. MICHELSON: Oh, no, no. Not;under.these condi- ' l? tions,Lno.c.170 tol180Jdegrees noicooliinstability,_ I agree'.- 1 ~ "] - 13' (But now you are defending a problem..that. relief-valves con ' 14 itinue'to'.functioniproperly discharging under?waterxthat is' 15- . reaching 200 degree whatever?'LI don't even-know if.we have 4 16 e x p e r i e n c e 'w'i t h ' t h a t.- .I just wondered. 17 MR.~MALONEY: I'm just' recalling from studies I've 18' looked at in the past. 19 - MR. MICHELSON: If this is 170 degrees, I wouldn't 20 worry about it. If it's 240,;I would-- 21 MR. MALONEY: I don't believe we're getting up that -l ge-{: . 22 high, and even there, I think those temperatures you'll be U. looking at'well late in the outwards of six hours. 23 g ( f.- 24 ' CHAIRMAN KERR: I assume this is treated somewhere 25- 'in the staff's review of the situation, since that's the Heritage Reporting Corporation n.v.a

s i 28 L l I I f assumption that we're going to remove those and then you 1 2 know that you've good condensing stability from the relief 3 valves in four hours. '~ 's '/ 4 MR. EBERSOLE: But I don't know that the condensing f 1 1 5 stability is the target of interest anyway. It may well be ] 6 that-- 7 MR. MICHELSONL: Well, it's one of them. 8 MR. EBERSOLE: Yes, but the loss of MPSH is 'another-l 9 the atmospheric component. i 1 f 10 CHAIRMAN KERR: Gentlemen, I think these are im-11 portant questions and they need-some attention. I. don't i 12 believe the group we have here this morning is likely to have n(/ 13 these details readily available and if we want these kinds j ~ 14 of details I think we ought to tell SAB what it is we'd 15 like to see-- 16 MR. MICHELSON: Well, are we writing a letter on 17 this issue? 18 MR. MINNERS: Well, the usual ACRS technique is l 19 they like to see the first submittal or something like that. 3 20 CHAIRMAN KERR: It sounds to me like we're going 21 to inherit a plant-by-plant examination. i 22 MR. WYLIE: Well, let's hope we get a plant-by-g3 'j ~ 23 plant examination, because otherwise I don't think the pro-(%) 24 blem will be solved. 1 1 25 MR. SIRKUS: The rule is currently written and will Heritage Reporting Corporation m.n.au

1 29 I be . submitted to the EDO. It calls for an information sub-g) 4 2 mittal 270 days'after the issuance of the rule. 3 CHAIRMAN KERR: And an information submittal means ,C) 4- .what? 5 MR. SIRKUS: Coping analysis. The results of cop-6 ing analysis on a plant-by-plant basis. 7 CHAIRMAN KERR: At what point, for example, will 8 an eight-hour plant have to become a four-hour plant? 9 MR. MINNERS: That's not a staff requirement. 10 The industry offered that they would go out and make all 1) plants four-hour plants.- That's not our requirement and l 12 won't be our requirement. Our requirement only is that you '( f 13 classify yourself-by the coping capability and then provide 14 that coping capability-- 15 CHAIRMAN KERR: I see. In other words, if the 16 plants automatically become four hours, then you will require 17 only that they demonstrate-- 18 MR. MINNERS: That's right. 19 CHAIRMAN KERR: The coping capability is a de-20 scription of what one does to cope once one gets within the 1 21 four-hour category, let's say. I l i , ~.s 22 MR. SIRKUS: Four, eight or twelve or whatever they N-] 23 submit. But the industry initiatives or at least the dis- ,() 24 cussions that we've had with the NUMARC people indicate that s_ 25 the industry's intent is to get all plante into the four-Heritage Reporting Corporation oni u..

m, .. 3 y -5 ',':y:, 1 30 .y 1 'i ' hour' coping and; recovery capable.; I (*(

I

2~

CHAIRMAN KERR:,'By whe'n?.'Do-you have any. idea?

y

e..

'3

Maybe. I.' sh'ouldi 'ask' Mr ' ' Opeka. -

4

- MR. :..OPEKA : 'On our\\ Initiative 1-A',.that's;the new. 1 5 Initiative as.a' result of'new matrix. It says thatfif we 6. .-could:use procedures in order to get the plant.'into a four-M 7

hour category, it1would.be one year'from the" time-that'the.

'j 1 '8 rule!becomes effective. If it requires hardware..ch$mges. I -( 4 9' .then it ' would-.be done ;within a reaso'nable amount of. time, ' 3 '10 WhichDI'm sure.that'we would have to work with the'NRC to j g c, . H, -t r 9 idetermine -what,t ressonable 'is,.

< +,

y .,c v '12 ' CHAIRMAN KERR: .,q. 'Okay. o j (f 13' DR.'SHEWMON: ' Sir;'I would*like to ask one of you, ] ~ I 14- 'and maybe you are better. ~

15 As.I understand, whether a plant is a four or

~ 16 - eight-hour site has to do with how fast--how power off,Lsite 17 .is likely to be lost and what changes can be made td getLit 18 back. What sort of things do you have in mind to take a 19 plant that's sixteen down to four? Is this more lines in, a 20 of putting a dam nearby? Or what? j 21 MR. OPEKA: Well, you can do a number of things. ' 22 The coping category is based on the reliability of outside u L ' power source,.which is configuration and also could be due' . 23 - '24 to weather where it's located. And also the configuration 1 25 emergency on-site diesel generators and also the reliability i Heritage Reporting Corporation f (202) 62s.4ase .j l L=_ _ = =- - _-___-___ ---. A

n ,s -s.. W 31: u.' jLfr.l2; 1 1 los so i'f you have#a low rel'iable. -l' ofltho'se diesel generators, P . ~., .q.4 '2- diesel generator, you can increase.'the reliability. That's' ~ 2 .s f [' 3D one way'of,getting up. You can'do some analysis 'looking at1 / 4, .your--iflyou.are-a. multi-unit' site and'have:~fourLdiesel- ~ r f, lM 4 l$' . generators 'on a shared' basis, if you could: show:.that ' oneIL61ese L' {- I lt r t 6 can provide all the station blackout modes.for both plants,t you can:tak credit for that diesel as an alternate AC-5 7: e A t, 8; source. 'So it's those kind of methods that-you canLuse in s ) 0rder)to getlthe plant down into,an eight-hour category. M r ~ 9 6 ~ a 10~ .Or-you.can alsolu'se" alter'nate AC'asJa'me'nsJof e.x Lik 'll - .getting'the/ plant down'ito a:.less'than>four-hour' category. 5is O F L,. 12 As longias.you can show that.that alternate AC source is g .-] l. .13' either 'available wiEliin...an hour, then you'd have to show one-hour coping capability or if that source is available 14 .15 Ewithin ten minuten,.then you don't really have to'show.any .16 . coping capability. 17 MR. SHEWMON: So you don't have to show'that.the 18 ' grid on the south tip of Florida is really the most stable .19 in the country? Or'something of that sort. 20 MR. OPEKA: That's right. That's' correct. 21 MR. SHEWMON: Yeah. Okay.-- '22-MR. EBERSOLE: I guess you would show that.you. 23 could put the peaking plant, gas burning or oil burning, 2 4 -- that's one of the plants and enhance the AC power. 25 MR. OPEKA: That's right. If you use alternate Je. Heritage Reporting Corporation m usa

a o .c .32 m g, .i ~ q I "AC, then you have.certain requirements-on-the QA forgthat: y .().l 1 2 . alternate 'AC' source. And also it has tolhave'the rel'iability- ) .3 o'f l95' percent. So';you k'now, you'have some-strin' gent re. i '4-quirements there. ] i 75 MR.'EBERSOLE: I think it'would'be importantIfor .the staff not re' ally to'ask simply'for a coping capability M 6 7' analysis'but to delineate-the details of how many systems _l 8-a.. -you expectEto-remain functional and'their degree in time. 4l iL y of degradation. In other,words, assign some! details to the- ]- 9 i ?lu . coping analysis you expect' froni' the< industry, pointing: out a u 4 11 i to them whatcyou'fexpect.c.ertain systems to;do.. In other.- + f 12 .words, develop ~a' chronology of failures. And so you could. [ 13 survive it in the critical'syctemsJto execute a safe shutdown. q 14 MR. SIRKUS: Well,'isn't that plant dependent i

]

15 though?- j 16 MR.'EBERSOLE: Oh, of~ course. That would-be plant: ] 17: Ldependent. 1 1 -18 MR.!WYLIE: Each plant is different in this regard. '19 .MR. EBERSOLE: But, you know,_for instance, l l 20 essential water. Yoil start with that. ] 21 ' MR. WYLIE: It seems like to me it's got to ori- . W 22 .ginate with the plant, in plant analysis. I don't see how y 23 you can prescribe this kind of thing. 1 24 MR. EBERSOLE: Well, it seems like there should be 25 some general language, Charlie, that identifies the composite Heritage Reporting Corporation <mnaa w=_- . _ _ _ _ _ - - _ - - _ = _. _

~ .] y. L33 r. . g

of. crippled

systems.

I v. .\\ 1 i 2'. MR.~WYLIE: Well, the general language.now is.that. 3' one has to show' capability to cope with- ~ q f* - 41 MR..EBERSOLE:,'That's.too close to saying;.do. good.- CHAIRMAN'KERR: Jesse,'it'isn't. Because'the:re-5 .6 liability of.this equipment exists in the literature.. One l .7: knows, for example, the, likelihood that a' relief valve will' 8' stick:open. It's not zero. And it's, from'whatJI-heard-J 9 ' earlier, it is.takentinto consideration that_with the pro-10 :' .bability-that'existsJfor<. loss lof Eff-site power for more than i four hours,jEliis[ additional likelihood,.will make the. total' jg like'lihood of:c re mb.., lt /somewhere arouridi10 to.the minus 5-i ~ 12 %( 33 or not greater than.10...to.the minus.5 on the'. average. So 34 - these_ things are not'being ignored. 15 - MR. EBERSOLE: ;Again, Bill, I think we're looking' 16 at the composite',.not merely of core damage but-containment. 17 Even one preceding the other in the unexpected-- so_10 to= 18 the minus_5 may not be a' good number. We're talking about no containment and a damaged core. 39 20' CHAIRMAN KERR: Well, maybe. i 21: MR. EBERSOLE: I don't think that's considered j

• i 22 for the safety goal.

i ,O CHAIRMAN KERE: It's not necessarily this unless ~ 23 l 24 one gets a set of failures. 25 MR. EBERSOLE: Sure. l ? Heritage Reporting Corporation < man. _._____._______._____________._____u_w

N _1 d !z .P- -: 3 4 ; ( Is CHAIRMAN-KERR: 'And if you'are going.to:go beyond "2 this in:more detail?it1seems;to me one.has to' decide'on the

31 reliability;that;you are going to demand.-

.b ^ MR; EBERSOLE:. Just to look at core damage'is not. V 4 l' 5 by~itself'enough..You"may not-have'a. containment even before 6 '- you: lose the core. [.,s 7 That'.s all';I:have to say. 8-CHAIRMAN:KERR: Any further' questions?- 9 We'thank:you, gentlemen, and it-seems.'to,me--this. 9 '10 ~ appears to'be.a c'ase:in which'thefstaff and the: industry. 11 - work together rather effectively to achieve a: desirable: goal. t . l'2 - - I think we 'sho,ul~d ^comminhi both-gro,ups,. f} 13. .You do.not anticipate a need for'a letter, Mr.. ~ 14 Wylie. Is there'any'-- { i 15 MR. WYLIE: Not unless the committee-- J 16-CHAIRMAN.KERR: Is there'any feeling that-we need' u ^ 17 to write a lettier? .I assume'then that we will not. 18 CHAIRMAN KERR: Mr.Sheron,.we're about fifteen -) I . 19 minutes ahead of schedule. -Is there any particular reason .i i 20. that your group can't go ahead with'this? j 'l 21 MR. SHERON: We're set. 1 d p-22 CHAIRMAN KERR: Okay. Are there preceding remarks 1 A._/ ; 23 Mr. Ward wants to make before Mr. Sheron begins? O 24 an w^ao= ree-25 CHAIRMAN KERR: Are you willing to listen to Mr. I Heritage Reporting Corporation m nus i _.2

2_1_.1____________._..

_3

J 1 -:) < L35 ,1 .a: 12 EWard-for a few minutes, Mr. Sheron?- ~ .2-MR..SHERON: Certainly. ) 233-MR.- WARD: Weve been hearing about the A 45 Pro-Sh" The Decay Heat Removal' Subcommittee. .j 4 gram forJseveral years. t '5, has sche'duled, I guess, two.or,three times.in-the last year. 6' .a meeting to hear.about the staff's' proposed. final or near'- -7 final' resolution of.the Program. ;That meeting-has been. post- -1 81 poned several times and.I guess the subcommittee arid' ther - j ~ i9' ccommitt'ee as a whole has'been getting'a little' concerned

10 L about'the' schedule'and whentthere is' going'to be some sort 11 of resolutionEof it.

12' SoLlast; month'we' asked the; staff'to come down;and ~~ ~ 13 just tell usJwhat the schedule was. We didn't really expect

1. 14

_there woul'd be any particul'ar: technical content to their 15 presentation today. Rather we'd just'like to hear some dis-16 cussion of what the problems are and what the prognosis is; 17 for a firmer schedule and a plan for resolution of the issue. 18 Now, Mr. Sheron agreed to come down and do that' 19 ' and not only-that, I just learned from-Mr. Bainert that'Mr. 20 Sheron has a bomb to drop-on us.- And in fact, has some s' ort 2! of resolution _ proposal which I really don't have any idea l: ~ 22 what it is. This may be wonderful. This may be--we'll all j t-l 23 sing _"Hossanna" after we hear this. On the other hand, we 1 u h 24 may be irritated.that it's presented to the committee in, 25 you know, just sort of dropping it on us without letting the Heritage Reporting Corporation m u.4 -i

g ri. .1. .r* l';' (" 'I 36. l s s vg a. .m H', , ' ; I s! .. learned' persons on theLDecay' Heat' Removal CommitteefhaveLa. I y I ]s. 2 chance to reviewLit, study it, and comment--on?it.first.. l 2 3 So with'that-- ^a. .9_)1 y '4 -.MR. ' REED : -I'd like to note"that the preceding. 1 \\ A fdiscussionon.theLA-44'somepeople. felt;thatA-45ehould ,5: 6o Lcome in along with.orsbefore'A-44. resolution. )And-I think ~ 77 that addresses someiof' Jesse's concerns:abouh details.such n 'I 85 as safety' relief. valves staying-- 9

• CHAIRMAN.KERR:

Does that end your-- 'a' 10 ' MR. WARD ~:.That'sithe end<of my remark's. 1 11J Well,.the suspense.is killing me. Let'c: get on; a 12< with it. t i

t. t 13 M R '. EBERSOLE:

Defore we open this up, I'm going 14 to have to say I: suspe'eti ominous conclusions. Because the 1 15 drift has been that way. 1 16 MR. SHERON: I would ask that the committee ~ members 17 . keep their~ hands above the table. i 18 Thank you for the opportunity to.come down and 19 talk to you about this. I must. apologize. This was not 20 .something we were saving up to. lay on you as a bomb. This 21 was a solution that evolved only over-the last several' 22 weeks. And'as a matter of fact, I really did not even have-- 3,:. 23 I would guess I would call my management's blessings to come j (. 24 'down and tell you until probably a' couple of days ago. So 25 .let me just point this out that this is the solution that Heritage Reporting Corporation mu t ; r- ____mm__._mmm_[___._mm._

( t A 37, .h p u o v~. l' we'velagreed to.. JWe aresnow' developing the details and.-Il y n ,2'

would'certainly recommend andLhope that we.would come back

,j c l. ~

3 fto.the Decay.~ Removal Subcommittee and present this-in more; pa

,1 a ~ !4 ! detail before-we come to the; committee'with anything finalt L .5 Thi's'is:just happens to.be classic of, poor timing. i v '6 So we'll'take.it'from there.1 4, . 7 l-I've.got just'some! background here'real quick to? ' j 18-refresh your_ memory.>on this issue.. And that is-that this- - ~ 9 L these are Vu-graphs I think you've..seen.in the' past in one. formoranotherfrom~Mr.LMarchesa,bhichshows.thehistory ~ f how we got to. where we 'are with A-45, whati the concerns.

11 o

~12 areLand whati tlie Jkey questions'are thattweLwere>trying:to- [a. ~ s. ..j 13: ' answer which are; are "the' current designs adequate arid are improvementstodecayremovpiisisteihsinoperatingplants- .- 14. 2 15 cost' beneficial. q 16 MR. EBERSOLE: Can I comment on that:last:line? 17 MR.'SHERON: Certainly.- 18 MR..EBERSOLE: . Implicit in that is perhaps'a funda-I 19 - mental and guiding presumption that what's called ~ adequate ~20 safety has already.been obtained even before you started-21-this study, and all you are. dealing'with is improvements =i 22 beyond adequate safety, because otherwise you wouldn't have

23-the cost beneficial aspect--

24' MR. SHERON: Well, let me speak to that a little 25: bit and I hope I'm not practicing law without a license up Heritage Reporting Corporation m g r I I

d 38 i 1 here. We had a real question come up on that very issue g) l 2 and that is that.if one was making a conclusion as a part of this as to whether one was dealing with adequate safety 3 7s kJ-4 or whether one was dealing with improvements beyond adequate s safety, I guess you are all familiar with the backfit rule 6 and the recent court ruling on that. We went over and spoke 7 with Mr..Treeby, one of the attorneys in OGC about that very 8 subject. Namely, that if the staff were to conclude that we l 9 did not believe that the current decay removal requirements j 10 .Provided adequate safety would we then be required to do a l 11 backfit analysis or not. 1 12 The' answer we got was that right now the staff has (v) 13 no speed limid inLterms of what is adequate safety. And so 14 from a legal opinion that we got, adequate safety is defined 15 as those requirements and those regulations that were in 16 place at the time the plant was licensed. And anything that 17 goes beyond that is considered an improvement beyond adequate 18 safety unless the Commission so decides that the current 19 regulations or the current situation does not provide ade-20 quate safety. 21 Because if one says, gee, the current regulations 22 don't provide adequate safety, then one has to ask the ques-23 tion of, well, what set of regulations do provide adequate [) 24 safety and in whose mind? And it can obviously only be the se 25 Commission's. Not any one staff member or any group on the Heritage Reporting Corporation <mna au

\\l ~ l if 39- 'J e .i ,r _n c 'l " staff.. p,' J+ t i .So we've checked that with'the attorneys and.that's 2 '3 + th'e definition of adequate' safety and-improvement to safety A i ~4 'that we're working on. ~ -5 MR..EBERSOLE: Well, you have.a law" basis.for .6 s'afety which:is'in'the' regulations and requirements but I.

7..

think.we all know that'adherance;to regulations. requirements 1 8 edon't buy' safety. You can cite any number.of-regulations x, ~ 9: which'are'so; ambiguous; '10

MR.-SHERON:

1. understand.

t. MR. EBERSOLE: fButithe interpretations that.;you. =12;

applp'to them can mean a terribly poor plantcor a veryLgood?

j

.i y '..f . 13 .one and the rangeLof...c' considerations?thad one makes in'these-- a u .14 'MR.-SHERON: All;I'm.trying to say is that from. i .l ~ 15 / our discussions.with the atto'rneys, the staff does not have 16 ' the authority to just arbitrarily decide what constitues 17 - safety. j E 18 MR. REED: What you are automatically doing'is H i) 19. saying that you never learned anything since the plant was l 20 ' licensed about-- 1 1 J 21 MR. SHERON: No, we're not saying that. j )J 22. MR'..EBESOLE: Oh, yes, you are. (23 .MR. REED: I thought you were. ( )L 24' = CHAIRMAN KEER: Look this is a regulatory agency - 25 and it has to abide by the laws of the country. And what Heritage Reporting Corporation (202) 62s.4ess i

[ 40 .j i 1 Mr. Sheron is trying to tell you is what the law is. And i l /~N, \\/ i hat's all he can do. He cannot break outside the law. t 2 1 3 If there needs to be a change in the regulations, then the q \\/ 4 regulations have to be changed. Or if there needs to be 5 something else-- t 6 MR. EBERSOLE: But the bendwith of interpretation 7 inside the law is so fantastically broad. 8 CHAIRMAN KERR: Jesse, that may be true, but the 9 regulatory agency has no choice ~but to try to operate within 10 the laws. I don't see how they can avoid this. I would hope 11 they wouldn't try.to avoid this. 12 MR..EBERSOLE: I think our discussion is really n ~ I) 13 soupy. s-l 14 CHAIRMAN KERR: Mr. Ward. i 15 MR. WARD: ,Yes. Brian what you said about the 16 definition of adequate safety or adequate protection seems 17 to be a little different--it's different at least from my 18 understanding of what--I guess it was Mr. Newbury and Mr. 19 Crockett told us last month. They are from the OGC I guess. 20 And they seemed to say--and I mean in this point--you seem 21 to say that the definition of adequate safety is whatever -es 22 you get from the existing body of regulations. And we asked l N. 23 them that question rather specifically and they said, no, (( ) 24 that's not right. That there is--I mean, in particular whe-25 ther the Backfoot Rule applies or whether cost benefit Heritage Reporting Corporation osvsu.a.

y L i% m" 41 3; .i l1: ' analysis could be used~in'an~. argument about. compliance, they p -Q,wl. i 2.. .saidfelearly cost-benefit" analysis arguments do not comelin. i H .3 iwith1 regard to' bringing a plantiinto compliance with the M(3' . 4-regulations. ..i. 1 MR.-SHERON: Correct. 5 t 6 = MR.; WARD: And they do noti come in to.bringingTa 37 plant L to a, level of. adequate protection, blit those aren' t the. ,8: same thing. -9 LMR.:SHERON: IfLa: plant does not comply with'the 10 regulations, then--you are correct.. It'is my understanding L 7 j .11:

that'one would;notJhave;to take into, account cost' benefit 'tolbringait-'into complibnce.

4 1 M )- ~ 13 i 'MR. WARD: Yes. 1 = s ,14 MR~. "S HERON : The problem you get into, and.I !!5 suffered through this back' Ohen :Iiwas Chief of Reactor Sys- ~ 16' tema is the>following. And'that'is.that when a licensee" 17 comes in.and presents his design, and claims that it;:has i analyzed that plant against the existing regulations and 18' .19 ' criteria or whatever, the staff reviews it and writes an 20 - SER and approves it. Ultimately the Commission ~ licenses-it-- ' 21_ based on the staff saying we have reasonable assurance that o 22 they've met the. Commission's rules and regulations. t - t 23L Subsequently, you can go back and find out that l 1 24f perhaps there are areas where they did not do an analysis, 25 for example, that we now seem to be requiring more current - j 2 5 Heritage Reporting Corporation - j (mna. J =D, 1

n. l42 W-c 'plantscand you,can argue, well,. gee, they-didn't meet ~the .2 regulations,-therefore they have to upgrade. 'We did this -)=y on the PORV. issue. We found out'after-the: fact--but you're- ~3- 'kI "4 . member after TNI,'all dhe-licensees stoodiup'and said,PORVst 5' do notuhave to be: safety' graded.' They almosti raiseditheir; 6. 'right" hands and' told us that. .7 : Subsequently,.we looked.into.the. steam generato'r: 8' tube rupture' event and asked them, do you rely on the.PORV to ~ 7 i .50 'depressurize the. plant in order to meet the regulations?: j q q 1 ulo" PAnd'they all said, oh, yes.. And so we said, well, gee, fhere: c 11 is.a-case whereLthd regulations obviously would imply'thatyai ~ t + i .12 PROV needs 'to be safety'~ grade but they-are not. Therefore. U 131 we-should be able to backfit ortwe should be able to:tell-14' these utilities, you' ~missad'it 'when ' you licensedL your plant. ) ,.l 15 .You told us a false ;s'tatement; that ~your PORVs were ' not safety a 16 ' grade.. They really should'be. Go upgrade. 17 What'we were told'is that when the plants were 18' . licensed, the older plants, the staff did not require those- -19 licensees'to perform a system analysis of a steam generator 20 . tube rupture. Strictly a radiological analysis. 'They made-j i 21 .very gross assumptions regarding depressurization time. -If 1 '22 you went back you would say, each plant assumed that the 'y .1 b ,r q %[ 23 primary to secondary pressure was equalized in thirty minutes. D 24 And therefore the argument, and this strictly a legal type l ~{ f? I, 25 of argument. And so I don't defend it or anything. And l i Heritage Reporting Corporation j l [ m.u. 0'

43 4 1 that is that'the staff has no basis to assume that the plant l r - u - 2 cannot meet the licensee requirements with the existing de-3 sign. In other words, we have not demonstrated that that ,il plant must have a safety grade PORV. That it must use the 4 5 PORV to depressurize. Therefore, the burden is on the staff 6 because we approved that plan. And so we have to demonstrate 7 through cost benefit backfit analysis why that PORV needs 8 to be1 upgraded to rafety grade. And as a consequence all-9 the older plants in.the United States may indeed rely on a 10 PORV to perform a.ssfety-related function but the PORVs are 11 not safety grade. And there is a generic issue right now, 12 Generic Issue 73, tha't specifically.is looking into what to s_ [v) 13 do about that. 'That is the result of our. attempt to try and g4 upgrade PROVs. 15 So I don't know if that. helps, but it shows the 16 dilemma we're sort of caught in. 37 MR. WARD: Okay. I don't want to spend any more 18 time on it now. But I've still got a little bit of dis-19 connect between the position you're taking and what my under-20 standing was of what the lawyers told us. But I'll hold 21 that for another time. 22 Go ahead. ] 7(_) 23 MR. SHERON: Okay. ['S 24 CHAIRMAN KERR: Now you can take off your legal f us) l l 25 hat and put on your engineering hat. I Heritage Reporting Corporation om mau

p. 3 :., :

- ~ v p I ~ H s .,a '? p 'l

MR. : SHERON: - lThank you'.

j: H

1..J 2.

.:The question.now.is what were thefobjectives:of.- !m 3 this task that was'to. determine the-safety' adequacy of the-N..WM: 6 4 decay removal'.in. the. existing power plants? for achieving. 3 b6th hot shutdown.-and coldcshutd5wn conditions ~.. Tolevaluate, 4 6 the._ feasibility o E ' alternative methods.; for Iimproving re " liability of dbcay heat; removal especihllyfa dedicSted decay l, ~

.7

'8H heat.: removal.: system; ;To assess theivalue andlimpact'o'f the- >9: most promising ~ alternative methods.' And to develop a plan l-LIO - for implementincf any new licensing 'requ'irements?that 'may jcome-i 11-rout-of thistassessment, y j

• • • E.*~*******?

d 12' -t (q) ;, a r 13 (Go to next page).j

• ,[

14 i s ,s j 4 15 -16J .17 q -18 -19 _1 0

' 21 1

p);

22 v. 23 - f5 24 25 Heritage Reporting Corporation om. __._._________i___.__________._ _ _ _ _ _

}l i 45 1 MR. SHEWMO.N: Is a complete rupture ~of the Steven l 2 Generator Tube, a small break? 3 MR. SHERON: It's not classified,as one with 4 respect to the SBP, but in reality, I would say, yes. It's 5 a loss of primary coolant. ,6 MR. SHEWMON: I wasn't thinking so much of that 7 as just that size is still small, small from the standpoint 8 of what size break? l 9 MR. SHERON: A small break is defined as any break 10 size for which the make up. system is not capable of main-taining the primary system in the inventory. 11 l 12 Typically that runs around, if I remember, 3/8ths of an inch diameter. 13 MR. MICHELSON: For a PWRT 14 MR. SHERON: For a PPR. A lot of it depends on 15 how the HBI--I'm sorry, about the make up capacit y too, but 16 that is the definition-- 17 MR. REED: Tell me again what you told me. A small 18 break is anything that the-- 19 MR. SHERON: That the make up system--anything 20 that will be pressurized in the plant to start the ECCS is 21 22 a small break. 23 MR. REED: Is a small break, okay. 24 MR. SHERON: Anything that does not depressurize 25 it such that the ECCS is actuated is considered a leak that Acme Reporting Company l202s M6 G 888

F 46 I 1 is made up by~the breaker system. . (y (_) 2 MR. SEEIEN: - Okay. I thought Paul was looking for the 3 definition on the other end of the small break, not between (~h r (~) 4 leak and small break, but between small break and medium break ) 5 or large break, isn't that correct? 6 MR. SHEWMON: Yes, it was. 7. MR. SHERON: That's really what he was looking for. 8 Okay. There really is'no absolute definition as 9 far as I recall. My. recollection is'the way it is analyzed, 10' is you carry the small break analysis by increasing the break 11 size and you carry the'large break analysis by decreasing it-12 and looking for that cross over point in terms of, say peak (J') 13 lock temperature. You may get different results with the same break 14 15 size and typically you would pick the larger. 16 MR. SHEWMON: The large break is clearly a primary 17 pipe or something that: is big but then you can find some that 93 are smaller-- MR. SHERON: Instrument lines, pipe splits and so 39 forth. 20 MR. SHEWMON: Thank you. 21 MR. SHERON: The vulnerability of the systems 22 ki requires the decay removal under special emergencies, for 23 example--well the external events as vell.gm; sabotage was lo.ket .r 3 24 (m/ at. As I said here, " Sabotage by an insider" as well as special 25 Acme Reporting Company 120 0 628 4886 E1______._

3' i s 47 I events. My comments concentrated on the system design . 'T (~ / 2 aspects to reduce the vulnerability to the inside rather than 3 physical security measures. ,\\ (,/ 4 This vu graph just is a summary to show you really 5 the amount of work that was done in a study for A-45. 6 There were 6 case studies done, 6 different plants. 7 These are all published. Where we are now is we have a 8 Technical summary and a-Technical Findings Report that is 9 being prepared for publication and the regulatory analysis 10 which is also being prepared. These are some of the findings, some of the main 11 findings that came out of these case studies. One is the 12 ('~J l probability of core melt due to decay or system failures 13 G ~4 can be greater than 10 to the for some plants and I wanted 14 to bring your attention that we just received an EPRI study 15 16 done by EPRI and the Westinghouse Owners Group on Point 17 Beach case study and we just got it and their bottom line was that they got a core melt frequency reduction by a factor ) 18 4 of 31 compared to our analysis. ig Other areas were that support system failures 20 were significant contributors to core melt frequency, emergency 21 22 power service, water component cooling. /-s) Concerns were raised as a result of these assess- ~' 23 1 ments regarding the redundancy of components and systems, i (~T 24 L,/ .s ar ng of systems, particularly at the support system level h i 25 Acme Reporting Company (2021 6P8 4R88 __.M_.._______ ]

$g 48 1 in some plants and then there.were also questions and concerns 2 raised about the overall general arrangement of ecruipment 3 from a safety viewpoint, lack of independence, separation, a physical protection were among the safeguards raised. x And then there is always that question of the 5 6 residual risks. For example, there are things that you'll 7 miss. There are things that PRA's may not cover, such as sabotage,' operator errors and so forth and not really covered 8 9 very well in PRAs and then there are, just like I said, the question that you will not get all of the events per 10 11 possibility. tie have not had a chance here to review the 12 EPRI Westinghouse Owners Group study in detail. This is 13 just a--we reproduced this yesterday out of their report to 14 show you whether NRC does the calculation or PRA, whether it 15 16 goes PRA for decay removal. This was the core melt frequency. We got--they get, you can see-that there is substan-17 tial disagreement in terms of the frequency of initiators. 18 Specifically you'll notice that internal events 19 contributed to, Point Beach. We thing the contribute roughly 20 half of the core melt frequency where as the EPRI study 21 showed that it could have been dominated by seismic as opposed ' 22 (,! to the internal events. 2'3 And another significant area, we estimated the buy O 24 L/ -5th contribution was 3.2 times 10 to the and theirs was down 25 l l Acme Reporting Company l l avv n..... _ __ _ __ - D

y*

. y- .,i .49-1-8 q l' to 10 to the~- . range. So there are some~ differences,;you VI 1 .know,,which would show that, you know, they' conclude they y 2 3 get a:much; lower. core melt probability. l. x ' %j-

d

As I said, we haven't had time at all to-go through-5

'these and1really:try and understand the differences, so I'm 1 6 not prepared to explain them.- 1 y '7, DR. MOELLER: What'is that last column, " reduction; 8 '. factors,"--oh that's'tlie difference-- 9 MR.'SHERON: :Just the ratio of--- (1 10' DRl:MOELLER: --of NRC over ETh.; !I 11 MR. SHERON: Yes, sir... 12 DR. - ;MOELLER :. 11 right. 13' MR. EBERSOLE:- Brian, isn't there some sort of a ~ i ~ 14 general thesis.that once you get up arotind 10 to the ~ 15 10 to the and on up, up,-up, you'.reireally in a cloud and D. 16 .you're dreaming? - 1 17 MR. SHERON: Well, once.you get up around.10 to the ~ 18' in that' area,.I think, there's'an uncertainty. 19 MR. EBERSOLE: It makes the numerical value just . absolutely, I can't see any further thbn that. 20 '21' MR. SHERON: Well, it's just difficult. I don't . I -6 p. . 22 distinguish between between say, 10 to the and 10 to the 1 M- .-7 10 to the. very well. 23. O' - 24 MR. EBERSOLE: That's right. .L y 25 MR. SHERON: I remember Tom Early, a long time ago, j Acme Reporting Company (2021 628 4886 __in___ M _ _:________-._

k af t ..f';' , ) z, y .1 %~ - (.p s q;7 cc p i ~50-1

e jd;;

a 3

onceitoldime, he said, always-reme'ber the age'of.:the earth is

~ m 2 J10.to the.9th= years. -3 These were the alternative'. resolutions'that were pb. 4 ., proposed and assessed for A-45'. A further one was to do: x . Alternative L5 'nothing, everything is: good enough the way'it is. 4 LG 2 was'was to perform limited. scope-PRAs and to' resolve the

7

issue.

Three was the!near accident program, individual plant-8 examinations. - Alternative 3 was to; specify systems modifica - ~ ~

e

'9 tions that resulted'.from USI and GSI resolutions. Alternative + 10' ' 4 was to enclose the-depressurization.of cooling / for; PWR :through-i t 11 .a' feeding, bleed capability and from PWR:through'the' payment -12 vending.. Alternative 5 was. to require a; dedicated hot 1 shut ' ~ s 13) down: capability. Alternative 6 was to require a dedicatedI 14 coal shut down capability.. '15. DR. MARK:- Couldlyou' help me a little. 'I'msur{ 16 the subcommittee knows this and I' don't. A: dedicated hot' s 17'

shut down capability.would require what, in addition;to what i s%

1 18-in'some!of the. plants now?. 19 '- MR. SHERON: Basically.I;' believe. it' would. be+ a 20 '.

separate.tuilding which would house heat exchangers like--I '.,

~ don't know--Warren, Jerry, do-you~want to expand on that?.:. 22 -MR..MAZETIS: First one is to have another means of feed water injection which would be secondery and then make 23 24 up capability is the primary. That's the design that was E 25 used'to cost out:that case. Acme Reporting Company aoa,.,.a...

(,. ..n= a 4 e r '51 If.. 'l j t y '1-

DR.' MARKS

: Now you need these-additional;dedica'ted-h...

2 sotirces :of water, pumps. That's;an additional. building or-1 3.' can1 additional pipe:or what? { mU: 4-MR. MAZETIS: Well, it wou'ld be--there isn't. room-- s5' in'looking through the plans, there wasn't room to put'this J6 equipment'inside the plant,:so an. additional building was 7, constructed.and it was' required that-some'high' pressure e> '8 - piping go into that buildint and there would be dedicated 9 . tanks in there.. 'DR. MARKS: 'If youLwere building a new-plant, you 1 10 -. 1 might not need^an a'dditional-building because7you should 31 i ) leave room to' install that. ,j 12 MR. MAZETIS: One of the bigger causes of the ( 13 -dedicated-system and I'm surprised at this, was the cost of .14 ~ ~ 15 the high pressure pipe in the dedicated building, that was one l

1 16 of'the major costs.

MR. EBERSOLE: Warren, you're sticking to the 37 18 worst' problem which is the BWR. If you look at the i boiler,it could mean more than a low pressure water source a:. 19 through'a low pressure pipe going to open cycle boiling. 20 21 . What was the cost--I think you ought to separate these two-l concepts because one is vastly easier than the other to cope 22 s with., 23 .) The boiler needs just an incoming supply of water ~24 and that'.s about all it needs because you can carry off the 25 i Acme Reporting Company 1207a 428 4888

-.-.L.-_.

--.-.L

p Cf, 52 1 heat with ventilation. V-) 2 MR. MAZETIS: Jerry Mazetis. The alternate 5 and 3 6 that are being presented here, it's the system that was just / '/ discussed, the auxilliary-- ~ \\ 4 5 MR. EBERSOLE: But that's the BWR case. 6 MR. MAZETIS: BWR, right. For the BWR it was 7 basically a make up system, RCI's system. 8 MR. EBERSOLE: High pressure. 9 MR. MAZETIS: High pressure for Alternate 6. It 10 included boiler pressure language. MR. EBERSOLE: Would it require a methodology 11 for' removing heat' pump containment with exchangers or did 12 I' it just blow the steam to wherever you need to take it? 13 x,J MR. MAZETIS: It was just interfacing with the 14 15 Primary system. CHAIRMAN KERR: Gentlemen, we are scheduled to 16 17 stop this session at 12:00 noon. I have missed the bomb 18 shell so far. MR. MICHELSON: Can we clarify the scope of this 19 20 thing. I thought the original scope was to include small 21 break LOCA on your scoping, MR. WARD: No, it was not, it absolutely was not, 22 g-R.) it never was. 23 MR MICHELSON: I was just looking at his slide on ( 24 25 scope, that's all. It says, "Small break LOCA and I was Acme Reporting Company i sa w.>..... 4 _________a

/t 53 1 just'trying'to figure it out and now we seem to have lost that 2 You do not intend to address any LOCA? 3 MR. MINNER: The scope of the issue, as it currently _) 4 stands is documented in the case work up and it includes 5 transients, small breaks LOCA--small break LOCAS up to the 6 size of approximately 1 and a half to two' inches. 7 Included in that, for example, alternatives 5 and 8 6, are make up systems and associated-- 9 MR. MICHELSON: It says small break LOCA and that 10 was what I wanted to get cleared up. 11 MR. EBERSOLE: How did you get the heat out of the 12 boiler containment? [,.,) 13 CHAIRMAN KERR: Jesse, this is a progress report. v 14 MR..:EBERSOLE: Okay. 15 CHAIRMAN KERR: We're not hearing a final report 16 at this time. 17 MR. MINNER: If I may add that the details of each 18 of'those designs are in the references that Brian put up on 19 the board. 20 MR. MICHELSON: We don't have those yet, do we? 21 MR. WARD: No. Let's go ahead, Brian. 22 tMR. MINNER: The details are available because x_) we have sent all:the case studies in and you can look it up 23 and see what the details are. 7'N 24 s.) 25 MR. SHERON: The value impact analysis that was Acme Reporting Company l j iron n2s-4see

gy-, a i.' t j p 5' d - ^ g s ?ll .doneJforfthelA-45 studies, was..performedcin 3Jdifferent# ways. , y.. ;, 3 a -2_ lIfam'sure?you'reifamiliar with this, i3 First,.only the inverted off. site cocts;were ~ m-a d'. -included..The second way was, in addition to;^theLoff. site- '5 costs,,we included!Lthe inverted'on site costs land third, } c6 .I think, was'to take the method 2, which is off sitejplus. ' on' site'and tojinclude, the effects of other special con ~- 8- .sid$ rations such as things are. sanitized, moratoriums, l 9l resolution of. issues and quantifiables.. t. 10 DR..MOELLER: What is'~the moratorium'for?'

11-

.MR. SHERON:.. I believe that' assumes thatLif.one A 12 Lhas a core me'lt accident in this country, that: the. Congress ~ j 13 ..wouldLtake: action to, eventually phase out nuclear. power. jy/ M 14' DR. : MOELLER: 'Thank pou. 15 MR. SHERON: The results showed that if I used the 16. firstymethod,:.which is=only inverted-off site costs, t alternatives 3 and 4 were#.s the only ones that came out-as being 17 -- s 1 18. cost / effective. :With method'2 which'is the-inverted off 19 ' s i t e ',. p l u s'_ o n s i t e, alternatives 2,-3 and 4.are cost effec-20-tive and if I use the third method, which included the special 3. 21 . considerations, then the alternatives 5 and 6 also became cost 22 - effective _which.were the vacated systems. O.3 23 DR. MARK: Cost effective always means a $1,000 24, per man rem-off site. 25 '. MR. SHERON: I beg your pardon, I'm sorry. Acme Reporting Company 12021 6284898 . -) ; z ^ _.n..,_--.-..--.----.-.__---.__i._-~.-_.------..,.---_.-_---------_____-

(V 55 1 DR. MARK: Cost effective, does it always mean 'C 2 a $1,000 per man rem off site? 3 MR. SilERON : That's what we used, I believe, as - ) 4 our criteria, yes. 5 MR. SHERON: Okay, I guess this is the bomb shell, 6 if'you want to call it a bomb, that is that the P.esearch 7 Management from Mr. Betcherd on down and now I also believe 8 Dr. Merley, is in general agreement.with this, although we sti ll 9 have to give his office a specific briefing but the concept, ~ 10 I know, I believe is acceptable to them. We have decided to endorse Alternative 2 which is 11 12 to seek decayed mobil improvements via the implementation ) 13 of a severe accident policy, individual plant examinations. x_,- 14 And why did we cg> this route? What was the basis for our 15 decision? Well first off, A-45, we looked atocase studies 16 17 and-they showed the main contributors to plant risk were P ant specific and so we could not conclude that a universal l 18 19

fix, i.e., a dedicated system, was necessarily the right 20 solution for all plants.

T1.e other thing was that when one uses the third 21 method for back fitting, which includes special considera-22

c-)

tions, mortatoria and so forth, this goes well beyond the 23 back fit analysis methods that were conditionally used for ) 24 25 all previous USI's and GSI's. Acme Reporting Company (202l 628 4888 I

,,l 5) 56 1 Also, we have concluded, we are right now working () 2 on this rather actively, is that with the IPE program that 3 would implement the severe accident policy, we are looking (_) 4 to try and use that program to resolve as many generic issues, 5 on an integrated basis, as we can. 6 We are right now trying to put the right words 7 in.the geniric letter that would go to-the industry offering 8 this opportunity for them to solve a number of generic issues 9 on an integrated basis. The reason is as follows and I could probably best 10 it explain it by an example that Cordell Reed raised at a 12 recent meeting and that is he said, look, you guys have an ~) 13 IPE process and somewhere down the road you're going to make a me look at seismic as part of external events and I am going g4 to have to do a seismic margin study or something and you also 15 have A-46 which is going to make me do something, walk down 16 ' plants, do a margin study or something on my older plants 17 for seismic and you have an East Coast Seismisity Program 18 where you're worried about the earthquakes that were used to 19 design a lot of plants on the East Coast. 20 He said, I really hope you don't make me walk 21 down my plant for seismic three times in a row and we said, 22 n, we don't want to do that. What we will do is that if 23 y v have to walk down your plant to look for seismic vulner- [') 24 Lj abilities, we'll do it one time. We will somehow. 25 Acme Reporting Company Q021 628 4888

o 7-7y.. W:s s (. .r

!; < j"A g

.s '~ f ;:7 1 r x c;, wy e e c 5 7 y, 4 :p R-E m W .'Y ~ f.;<, c s kq/L, L 4 _1' gv y g., -j

Soanywayfthatiis,7 sortlofca[b$ sis forfthe.fap-lq.

a..... v i h-Q: f' proach-istthat.we: recognize thatiifTwe dhfan'IPE that:wel } 2; -may very well--if..wetdid.'not include generic' issues, wefmayi N.;~i,. J3 - y l n 1, , (q: J; 0

very well have'Lto turn around and ask utilities to do similar 4

a yearLor)two11ater as.'these generic.'issuesiand.USI's wo ?5 6 - become resolved, which we will.then hear complaints fromc ] ( .the!. industry that you.' re ' making me do the-- same wo: k. over and' 7

1. +

4 iIj ' ver andlover.- l ~ > 8 q 9l .So whatEwe are doing"right.now,1 is wocare examining H ~ '0 all" f the; lists of active USIs and'GSIs and.trying to deterp 1 mine which~ones are most amenable to resolution via a process .yy-such' as the :IPE,.and 'to offer to the industry the ' opportunity 7

12 i

that'if-they specifically address the-issues' associated'with L13 - j Of .-yg .those' generic issues'ob USIs as-part of their.IPE, and'in Ji {

15 their reports, specifically come back to us and tell'us whati 1

4 16' .their, conclusions'are.regarding. vulnerabilities and risk. j <3 M that the staff is willing?to write.off those generic-issues .gv1 77 j Ifor those plants and we felt that because'.the contributors y ~ t -18 l to risk' appear to be plant specific, there was no one major .f l& 39' q contributor for any plant that stood out'and said if I fix . l. -20 .~ this contributor for every plant decayed, mobile or satis-y 4 . 21 ( ' A. factory, 'iti was different ones for different plants. l '22 j M-MR. WARD: Brian, let me interrupt, you here. I ci l 23 s nk we.get the drift of what you've got to say and I would 24. like.to give, unless you got'maybe one final statement you q . 25 ) i v j' i Acme Reporting Company i aw sua..a

y g. r -. _ y y I p; M _{ l ~ ) .g q / U 3;;" a w w Tgg u, I' R? ' m Q ', 's e S 'b (1) T would.iliks /to.. wrap" up. I would like to.; 4 m j~,. 3 y b MR.ISHERON: 'I *do ' have ~ oneR fi.nalb-l a " 2 (.-

v O3'

.' M R. ; W A R D :r Okay.. And then1I1want.l toLgivef.Andy; l ' j]:l ' 'i L,- .4 Marchesa aEfe,w-. minutes. j . 5. - 1M1. SHE.RON: Okay. J s 6 MR.-WA'RD: I think the committee could' benefit 7 ' from. hearing his view's. ~- J 1 -8 .MR.-SHERON: That was what I-was really^just. + . going. tor'say.

.What 'I: was : going to tell you and ' that is

.D' I 4 that.' Andy, e as you..kriow, thas, been l working on-this for Ethe l l'0 11' -past.6.' years'or so.;.Andy: reached.the conclusion on what ~ 12 : .he-believed,was the most' appropriate solution'to the issue. 13 'As'I,said.it w'as presented'to the... managements.in;the Offi)e1 of Research.,'After deliberation--a little;. painful u 14-15 . deliberation.,-.;we felt-that~we could.not support'that particu 1 16 lar solution 7 ~ 17 ; I do recognize, and I have said. this'a number of 18' ' times and I'll say it now and that is.that there is not-39 question ^with respect to the technical adequacy or content .25 of the work Andy did. J. I think it is of the highest technical quality m-.

21 22 and what we are really discussing here is a policy decision.

h.. It's a policy decision on back fit analysis and how one goes . 23-about doing a backfit, what. considerations one takes into 24 , account when one does a backfit. Are there acceptable values 25-4 Acme Reporting Company 'M (202) 628 4R08 I ,, Q r.ly __l'i_____.._1.._'_.._.

7 6-L

59. l i

1 that one should achieve in resolving a generic issue with I ) respect to'large release of a core melt frequency. 2 3 There are a number of other issues which went (_) 4 well beyond-what I would call a technical decision, but more 5 or less, it's a policy decision, and we recognize that and 6 we struggled with it because we realized that we don't have 7 very good guidance on a number of those areas, so what we 8 are proposing is that as we go forward with this resolution 9 that the management has decided is most appropriate, mainly 10 the IPE approach, that we will afford, either Andy would wish 11 to address the committee, CRTR and the commission regarding 12 why he thinks his resolution is the most appropriate. ) 13 MR. WARD: Okay, thank you Brian. Andy, we really u-have just got about 6 minutes and the Chairman is going to 14 15 bang the gavel at noon, but I would just kind of like to get 16 a summary of your views. 17 MR. MARCHESA: Thank you, Mr. Chairman, for giving 18 me the opportunity. I don't have any planned remerks. I 39 just found out late yesterday that they wanted me to come down 20 here and basically make some comments. 21 As y u know, I have been working on this program 22 for 6 years, but I feel that something far more comprehensive ,s 1 ,.l 23 than the IPE program is required to resolve this issue and (^T 24 the basis of that, briefly, is as follows: %) I think for most plants the core melt frequency is 25 Acme Reporting Company (2021 628 4689

t i f. 60 / ~ 1 greated than 10 to the and in order to satisfy the L, ij 2 reasonable assurance guideline that is in the commission's 3 Policy statement of no core melt in the U. S. pop 61ation 73 4 of plants, that the. only way that you're going to get a factor of 10 reduction in core melt frequency is through the 5 6 dedicated system option. In addition I feel that most plants are not' going 7 -6 8 to meet the 10 to the large release criteria that is in I 9 the safety goal policy statement because the only way you're going to meet that is to get core melt frequencies down 10 around 10 to the ~ range unless you place, you know, undue gg reliance on a container system. 12 In addition, there are significant--I.would call them (') 13 ,,/ deterministic weaknesses in these plants, that is the physical 14 arrangement of plant equipment is very poor from a safety 15 standpoint in that you have redundant trains of safety 16 equipment sitting side by side in a common area which makes 17 18 them very vulernable to common mode events like fire, flood, seismic and sabotage. gg Basically I feel that the only way to improve these 20 21 plants, because they have broad vulnerabilities is that the resolution has to be broad based and comprehensive and 1 22 think the IPE is going to result in debates over numbers 23 and POA studies, it's obvious to me that the industry is goinc (~l 24 ') to come in with PRA that say core melt frequency is less 25 Acme Reporting Company (2025 626 4886

n {$81 ^ 'yf, .(' 61 -. " ~, _5 ~. ..np W..,' than 10.;to %he

and nothing'has{to be done.

I-thinkLwhattis c

g 1

12-Lgoihg?toEresult lfrom this IPE sttidylis ' endless.:debatest on 1 y ;..q ~ ' '.probalistic numbersiand'no real improvement in. safety.- 3J ~ ' h. - That 's - 'it. - l p 4-- '5 DR. SHEWMON: Could you'back'up-for;a minute'and: i i' 6 tell me why you. feel:a significant improvement in safety.is 7;

needed with regard

to the decay-.

8 MR.- MARCHESA: I'think therregulations'are in-adequate lanci cost should not be an ' overriding concern 'in re - 9 $10-solving this-' issue. LI think.they'are-inadequate because i t'. they. result in high core melt frequencies. They result in - i . arrangement of safetyseguipment where you got safety trains 12. J.O . sitting' side by side in'the same area.. -They;should be ~ e 13 o V separate', they should lie' independent,jthey should ~ not' be '14 15-vulnerable to commonfmode' failures. { And.I['thinkthesefnewgenericissues'aregoingLto

)

16 keep popping.up[andEthat the only way to get rid. of the 17 uncertainties involved is'through'a dedicated option that will 18. resolve a whole host of th'ings that you'can q,antify, but'the 19 20-things that you cannot quantify, DR. SHEWMON: I have another question. I just 21 w nder if this dedicated system consists of a separate build-22' 'O O' ing n the edge of the containment that brings in water or 23 (] 24 has it got a design that would be acceptable? 25 MR. MARCHESA: We have looked at several designs. Acme Reporting Company (202) 628 d688

7, y, Dj ~ ' - ~ x u y gj ) + %A, 4 7 4 s a c I62. L,, s. . ; y" I Q y g

-t.

k.; _.

(" .1L ...Welooked.,atethe Miller designs including ltheJdesign that Mr. H ^ a ~, s N%'Pv. j s Ih 2- ~'Rhese suggested.,. Basically I'm not advocating:one designior: n;; m 3 (another. I thi'nk that the dedicatsd: system concept =provides' j .&f i h, .x d l4 '=

systeins to make up to the primary -system and:cthel secondary; 5

. system along.with1 dedicated water supplies'.and all:thefsup- ..] , n, .4 ~ -l 6. / port systems-that go -along ;with1 satisfying; thoseLfunctions ] 7 like AC-DC power, essential; service'. water and componenti: - ~ 8 cooling.; water, a system'that is totally l separate'and in - 9 dependent from;theinormal plantT. systems-because right now!the io-normal plant'systemslare,so inter; dependent,~there-is.so much l P [ 11 sharing goes on', there is so much sitting side;by, side, that-12; .they're very: vulnerable. '13 DR. SHEWMON: Lokay,' fine. 3 1 '14 DR..MOELLER: I wondered'what.the? ratio.was', what, 1 15-the cost ofLthe fix proposed by RES versustyour co'al shut-4 w 4 16l down, dedicat'ed-- 0= 17 'MR. MARCHESA: The cost, depending.on the plant, 18 is anywhere.from $60 to $100 million. 19 DR. MOELLER: For yours? J 4 1 s20 MR. MARCHESA: For my system. .t 21. .DR. MOELLER: And what for theirs-- 22 MR. MARCHESA: Well not my system, but for a l ..,)'u.

j A r-23

' dedicated system. ' 24 ' Now, it's true if you go ahead with this cause 25 benefit analysis approach, which is, by-the way, in a state u l l Acme Reporting Company m .w,,.a -on u_ = < - ~

0 3 63' i 1 of evolution, you know. This area has been evolving for the (%_) 2 Past 2 years and there's no real one way to do a' cost benefit 3 analysis. ) I feel that it is appropriate to do the cost benefit 4 5 using the method 3 that Dr. Sheron has outlined, that unless 6 you include these other considerations like sabotage, un-7 quantifiables, relationship to other generic issues, you're 8 not complete, you're short' changing the values. 9 CHAIRMAN KERR: Gentlemen, I realize this is an 10 extremely important issue, but we simply do not have time 11 to pursue it in the depth required. I think we have time 12 f r maybe one more question. l I ') 13 MR. REED: I have a question. l %s' CHAIRMAN KERR: Go ahead. 14 15 MR. REED: Do you visualize this system, this 16 dedicated system, as a back up system in case the other n rms fail? 17 MR. MARCHESA: Yes. 18 1 I MR. REED: And that's the basis of the way it .] 19 l 20 will be designed? ) MR. MARCHESA: Yes. ) 21 l 22 MR. WARD: Thank you very much, Andy. Mr. Chairman, -s ( } v 23 the meeting is back to you. 24 CHAIRMAN KERR: One hour break for lunch. (^)\\ \\_ 1 25 (Whereupon, at 12:02 p.m., the hearing in the above-entitled matter was recessed for a luncheon break.) l Acme Reporting Company L202l 628 4888

1

l CERTIFICATE 2.

g^e y j 'd 3'.This'is toicertify that the attached proceedings before the. '4 -United States Nuclear Regulatory Commission in the matter of:- Names _- 331st ACRS MEETING-6 7' ~ Decket Number: 8 Plece: Washington,.D. C. 9 Datet. Thursday, October 5, 1987' 10 were held as herein appears, and that this is the original 11 transcript thereof for the file of the United States Nuclear 12 Regulatory Commission taken stenographically by me and,. 13 thereafter reduced to typewriting by me or under the direction 14 of the-court reporting company, and that the transcript is a 15 } true and accurate-record of.the foregoing proceedings. 16 /S/ 1I W 18 17 (Signature typod): 19 Official Reporter 19 Heritage Reporting Corporation-20 21 22 23 24 25 I ) Heritage Reporting Corporation (202) 628-4888

v ;\\ v w )

r

,w. c I'l [ s. 'E O 3 H st W. ? ? m e q 64' j ?' yi:

A F'T-E RJN'.O.O N S E S 'S 'I O N :

j 1 LI ? hI ' ~ s '2 -CHAIRMAN KERR .We now come toifurther consideration B b 3' ' of the'. Tennessee Valley Authority,.the.present subcommittee 0 -l

4

. chairman is Mr. Wylie,LIEdo;believe. j 5 MR.' WYLIE. All right. Thank you, Mr. Chairman.-- The i 6- ' subcommittee on the TVA. organizational issues met: yesterday,; 7. . and had reports by the staff and TVAfon the issues. It'was 8 quite a long subcommittee _meetingsand:quite excitingLand in 9 depth. It lasted.from 8:30Luntil about 7:00 yesterday. nSo,.it 1 -10 was-quite'a long meeting.- There'was quite a11arge exchange of 1 i 11' . information. The staff covered the status of'their activities' 12: and their: schedules, and TVA covered responses 1to.the- 'l >l '13 recommendations'which the ACRS had'ma'de'in their lettbr of. 1 August 12, 1986c end gave-the status of each'of those items in. - i -14 15 -'some-depth. 16. The staff covered their report of their integrated! j 17 design inspection of the1essentialJraw-water cooling system-at 18 TVA, again, in some depth there-and1their investigation:and' j 19 their findings and their. schedule.for resolution of issues. 20 TVA covered their safety review board and its activities. l

21 There.was quite a bit of discussion in that regard.

-l t ' 'b 22 From the meeting yesterday, we selected several items < j.. E 23 to bring to the full committee today. The staff will cover the ) ~ 24 status report and discussions of the integrated design 3 l 25 inspection of the essential service water system at TVA, and Heritage Reporting Corporation (202) 628-4888 1

a- =- 2 e, 1

k.,,

b 1 l.65; t] l' 7"NJ, 1 sM f1 then. TVA" will; cover their overview of> the TVA management: 1 ? 2, reorganization'anditheir' discussion of the'r. Nuclear Safety? E i P3-Review Board structure and-staffing 1and activities,;and a-Li 4 discussion of the-Sequoyah' restart' issues, l l 5: The agenda is in'. tab 6 of your, book, and the backup- "J .6 Information also is contained-therein. That's about all I'm-ju L 7 going to cover about the meeting yesterday, unless some;ofo H 8 -other'other subcommittee members want'to add to that. t Mr. Ward? i l 10 MR. WARD: Yes. I would like to say something. ,I 11' ' guess I was kind of critical of the. staff yesterday afternoon. j 12 ~ .towards the!end of the' meeting. I think it's an important-i 13 topic. I think the criticism was appropriate. I think.it's.an. 7 ~

14 important topic, and maybe we can handle.it mostLefficientlyLif -

15 I state my; concerns to the rest.of.the committee now'up front, 16 and kind of;get'it.out o>f the way. ] )) . 17: Maybe not out of the way, but to put some perspective Ng3 18. on what they hear.- 1 19 The issue is this: in our letter of August of last i - 20 year on the TVA programs, we made a number of recommendations. -: j i 21 'I'll just read one of-them -- it's a short paragraph. This was 22 one of the recommendations in our letter. "The NRC staff is 1 3 23 currently confronting the difficult task of reviewing.a large j 24 organization with complex and '.ong-standing organizational j 25 problems, and judging if adequate reorganizational steps have ( Heritage Reporting Corporation (202) 628-4888 j i l l 1

a ~ ,'4 .] , y. l 4He 1 i y' 66 ) Ll: beenitaken.- The NRC staff in'the~past-has'been oriented:to 2- -hardwareErather than management' problems, and perhaps has not 3 given enough1 attention!to developing;the= capability 1to perform 4 . management reviews.. We believe that improving the' management-5 of-activities related to nuclear power plants continues to be 6-one'of the most effective means forfimproving theLsafety of i 7 operations, and we. recommend that the-Commission give high ' priority.tojdeveloping NRC capability for conducting l management-9 freviews.- 'Emphasiscshould;be place ontidentifying management (10 ' problems before they.. lead to dif ficulties on the. scale-l o- , 11 encountered'at.TVA."

12

.Well,fI think that the staff and.the. commissioners 1 L13 have'done nothing in response'to that. recommendation-of ours'-- TO. 14- .15; months:ago. LThat leads-to'the'following: problem!' TVA.has - 15' taken whatiseemsito.me from-what I heard yesterday a numbe'r of c -.e{fectiveprogramstodeal_.withwhat I will call the " hardware 7 16; 17 l problems ", Ewhich' we heard ' a lot about. But also,fto deal'with. ?l8 thelmore~ general problem of' management organization -- there 19 ,seems to'be general agreement that that was the root problem, .i 4 20 and, in fact, the ' hardware problems tended just to be symptoms 4 ~ 21-of the management organization problems.

• 22 Now, my problem is that that staff a year ago, and x

23 "for many years, has had the capability to review and make 1 24 competent, professional judgments about the hardware and hard 25. technology programs or proposals that licensees are bringing l Heritage Reporting Corporation (202) 628-4888 I =L_________:_

R Q. ~ m; ; y 3 ', J -67 ^ [

1(

before'them. 2 . The 7caff,.back a year ago,.did not seem~to have~the 3 similar' capability lto.make professional, competentireviewsand: ~ 4

judgmentsiabout the' managerial and organizational, changes that-a 5;

this licen'see is making. And despite the. fact that we called n '6L that to-their-attention, and I think that they agreed with it, i -7_' .they have done nothing in the fifteen months since then that 8'

was apparent to us - 'it was apparent.to me, anyway, in the l

91 . meeting yesterday -- to.put themselves in a position to.make; 1 10 'more professional-judgments and reviews of the very important. H 11 managerial and organizational changes that TVA_is proposing. 12l before they restart.their reactors. ~ 13- 'I think that's a serious deficiency. I am very O ~14 l disappointed that there~hasn't been'something;done.along those 15 ' lines. Now, I'm not proposing that the staff TVA how to manage .16 'or' organize themselves. I wouldn't propose that the.staffLtell' 17 TVA' how to design their plants.- I'll repeat my point'again, 18 because some people don't seem to get it -- some of the members j ~19 don't seem.to' understand --'I guess, because I'm not making it l 20 very clear. I think the staff should, though, be in a position 21 to make competent, professional judgments about the 22 organizational and managerial matters that are confronting 23 them in the TVA situation. I don't think they are. 24 Okay. I've bad my six minutes. That's it. 25 MR. REED: Well, I think I have to rebut a little to j fr~N th l l Heritage Reporting Corporation i (202) 628-4888 l 1

s y-)k ~; 68 x_ 1 that. I wrote.: additional comments on the previous letter 2 saying that I didn't think it was a good idea for the staff to 3 tool up and become experts or anything in the management area. 4 My reasoning is that what the staff should be using and looking 5 for is perhaps management breakdown through performance j 6 indicators or through the SALP process, and then there is 7 always IMPO, that is supposed to concentrate on good management l 8 and how to achieve good management. They have forums and ~ 9 workshops and all these types. 4 10 So I really don't think that the staff has to prepare l l 11 themselves to deal with the management issue other than to say 12 that there is something wrong. u-13 MR. WYLIE: Any other comment? fD ~' 14 Okay. Let's proceed, then. The first item on our j 15 agenda today is the NRC status report, and the discussion on 16 NRC's integrated design inspection. Bob Hermann of the staff l 17 is going to cover that for us. ] i 18 MR. HERMANN: Good afternoon. I'd like to talk to j 19 you and provide a synopsis of'the independent design inspection 20 that was done at Sequoyah Unit 2. I'd like to start with a j i 21 little background on why the staff though it necessary to do an 22 integrated design assessment at Sequoyah. 23 At Sequoyah there were a number of programs by TVA -- 24' such things as equipment qualification, design-base 25 verification programs, calculations programs -- and a very long n us Heritage Reporting Corporation (202) 628-4888 l

En7n ~-- L,-

h,'

7' -;p 7); y w

69' r

Mp li ' laun' dry list'of-programs:that essentially were appiled across One of.the: concerns of:the staff-..is interface; 24 - the board.- <between those. programs, andlone option initially that was. 3 c 4, ; provided to TVA wasito do an independent-design; inspection of .5-themselves -- so-called verticai slices. EverybodyfhasLheard 6 about them'-- say, across.one system or two systems to look at nterfaces. i -7 8 TVA chose not to do that,.and the. staff itself chose 9 'to conduct the' independent design inspection. j 10 DR. 1911U(: You specify Unit 2. Is there any. 11-difference between Units 1 and 2? 12' MR..HERMANN: In' terms of' designs? 13. DR. MARK: In terms of'your discussion. 14" MR..HERMANN: No, not particularly talking'about' Unit i 15

2. : There are some areas that haven't been looked at.

I would 16' think that the work that's been'done in most cases is 17 appropriate to Unit 1 also. But we are specifically talking l l 18 .about Unit 2 at this time. 19 MR. WYLIE: Unit 2 is the first unit to come back2cni 20 ' line. . 21 DR. MARK: That I realize. But does not the work on j . 22 Unit 2 cover Unit 17 l 23 MR. WYLIE: ' I would think so. 1 l ' 24-MR. HANNUM: There may be areas that are identified-l 25' as problems for one year that I'm sure the utility would do (T J r~ / Heritage Reporting Corporation I (202) 628-4888 i L_____z______.-.-_--_-

s I L 70 I f ~~'s i t_) 1 something abont'on the other one. e 1 2 DR. MARK: Well, there may be, but I doubt if'there j 1 j 3 are. The concrete came out of a different bin -- 4 CHAIRMAN KERR: No, just the welds were made on a l 1 5 different day by a different man, j 6 MR. HANNOM: I would suggest, Mr. Mark, if there is 7 anything in principle that is deduced from any of these i 0 inspections, that would certainly apply to both units. If they ) 9 are specific, isolate oversights, those would apply to only i 10 one. We certainly believe that in terms of re-establishing a j 11 design base and re-establishing a credibility -- doing the 12 inspection on one unit should suffice for both units. 13 DR. MARK: That's what I hope. Now, it's true that g ~s ) 14 the welding was done by different people and the concrete was 15 made out of different sand -- things like that. 16 MR. HANNUM: But not at a greatly different time, and 17 certainly not under different controls. 18 DR. MARK: Thank you, that's enough. 19 MR. HERMANN: Okay. The integrated design inspection 20 was performed by a team of 14 people with NRC inspectors with ] 21 contractors as assistants. The programs were over five I 22 disciplines -- essentially mechanical systems, mechanical G 23 components, civil structural, electric power, and s 24 instrumentation and control. l 25 MR. EBERSOLE: Let me ask you a question about the s rn \\w.] Heritage Reporting Corporation (202) 628-4888 I b 5 i

.4 's. i.!Il

U a

1 1 i ^ {4 ' [11 - term L "inspec tors "i. I-fcan't'get: rid of the coiinotat th

2-inspector. isi the:L man that : examine's i the : materialization, th' idesign Litsel'f.-

f3~. .. design, but not T e fMR.;HERMANN I It?or, engineers =---people-withidet.

4'

- 5 ~

Jand) engineering' backgrounds'. The' consultant'Js generallyicome:

6-from' areas'where'they had. Worked.for AEs or had' worked"for-7 utilities. LI.have'a back-up slideLifiyou want toisee the.

~s. L 8; composition of the group, r, 9 .MR.'EBERSOLE: No, I.want you to broaden the' 3; i -1 10 connotation. '11 MR.;HERMANN: Well,.it's in the term that-we.use ---- o

12;

-inspectors. The inspection emphasis was on implementation of

g

'13. the FSAR commitments, technical adequacy of design documents,

14 design. interface control, design change control, s.ince the. pre 15 O/L days.

The problem at Sequoyah that was' covered,by the'DBVP./,

16 was the fact that there had.been a-lot of changes:to the' plant' 17 for' modifications and problems with the design control process.

' I '18 That problem'was from O/L.on. L

19 This program did take a look at some of the pre-0/L The activiti's were in preparation for-l20 parts of the program.

e '21 the inspection, Inspections were at the TVA design offices and' 22 at the site. When that was completed, identification of a set l' of restart issues were issued to TVA on October 9. The final L . 23' l

24~

report preparation, which is due.to go out tomorrow, is to L L. 25 -follow up on the findings that are in the report. i O Heritage Reporting Corporation (202) 628-4888 ) L l .__-____-___________________-__-______D

l l I 72 1 I 1 The topics that were covered in the inspection were l 2 the validity of design inputs and assumptions, the validly of i 3 and conformance to the design specifications, the validity of 1 4 analyses, proper implementation of system interface i 1 5 requirements, proper component classification, and control j -6 design changes. l 3 .7 MR. EBERSOLE: I would just guess that the committee is l 8 would be interested in the character of the design inputs, and ) Q 9 principally the assumptions. I had sort of a limited contact 10 with that plant early on, and one of the drastic problems was 4 11 this horrible flood that was postulated to occur from upstream j i 12 in the river and come down fairly fast and raise a 40 foot head ) 13 against everything and knock over Chickamauga Dam, and thus h-14 leave the plant stranded with'some kind of a muddy stream 15 behind it. 16 The original concept was to capture a body of water 1 17 clear and capable, and resort to evaporative cooling through 18 tower operation rather than a heat transport open cycle, which l l 19 represented a minimum use of water, because of the tower, 20 There was no attempt to use full water for sensible heat i 21 transport. I understand all that has been abandoned now, and l 22 you are going to pull water out of the thread of the river. 23 That faces the pre-existing questions of what is the caliber of 24 that water, the quality, will it still be there, will it be mud l 25 instead of water. Heritage Reporting Corporation l (202) 628-4888 1 I

'[ s! ~ T -73 l 1-It'gets71nto a' pretty' hairy' picture being unable 1to l2. ' assess what willibe there. I don't know how those are.

3

. overcome. I think the committee'might be interested in that. 4 MR.-HERMANN: Okay. I was basically'. going to' finish j 1 5 one more allde'on the' criteria for sample' selection'and get_to 1 ~ ~ '6 the RCW, and'then perhaps we:could' talk about'it for a little 7 bit'. 8 (Slide) 9 .MR. HERMANN: The. sample system, this.ls basically-10 <the selection criteria for picking a system. And one was that 11 the system be essential'to. plant safety, be designed by.'the AE, 12 which in this~ case was TVA, lx3 representative of safety 13 features in other systems, and be a design which involves

j_A

^ s 14 ' internal' interfaces between' functional areas, external 15-interfaces with nuclear steam' supply' vendor, component vendors '16 and engineering service organizations. q 17 Jusse was.getting a little ahead of me'. What was -] 18 finally picked was his favorite system.that they used to 19 disappear'from when the Chickamanga Dam disappeared. 20. (Slide) 21-MR. HERMANN: I believe one of the things that'was 22-changed, and the utility maybe could help me with some of the 23 details on this one, the system used to be, as he indicated, a I .2 4 ' closed loop system with cooling towers, and I guess most of the 25 ' water for other systems was picked up out of the normal service j () Heritage Reporting Corporation (202) 628-4888 i

s. s 0 - ~~ m 4 ll ,\\ s 1 y '74 i .:/M. _, N) - ,1 bay. ow 2 They' changed'the design of the system to extend thei j 3' system'out into a different'placeEin-the> river.where if you had i 4 a dam break you.would still have a reliable' source of water:to , be picked up.for the essential all-water cooling system. 65 ~Basica11y'what they.did was build a bunch of, I call them i 7 little' houses, as I'did yesterday, that are bounded on. rock, 8 that go out into the river and the pump suctions are 9 essentially out.in an area of the river which, if you get the 10 dam break,-is supposed to have water. 11 I: don't know if I can say much more about the details 12. of design, other than that, just off the' cuff.- AL3 Let me go through a little-bit about I guess .( 3 'I ) =, 14 schematically'what we'are talking about in here. It's minimum 1 15 river channel and ERCW, and taking the pumping station. This 16 is the new part of the system. 1 a 17 The function of the system is. essentially to provide 18 cooling. work, a lot of things, component cooling. work. This is j a 19 on'the raw water side. Room chillers and coolers, control 20 . room, computer generator building, RCM, react coolant pump 21 motors and upper containment coolers. Essentially that is the 22 scope of the system. l 23s Oh, yes. One of the other things fs the auxiliary 124 control air compressor. 25 (Slide) s s-Heritage Reporting Corporation (202) 628-4888

i 75 i_3 (- )- 1 MR. HERMANN: The largest number of problems that i 2 were identified as part'of the IDI were in a civil / structural. 3 area. Those findings, as we said earlier, were identified in a j 4 letter from Jim Keppler-to Admiral White on October 9, 1987. i .5 That particular area is part of the review of another 6 program, the design calculation program. The civil / structural 7 area hadn't been reviewed by TVA to that point, so I guess 8 personally I'm not. surprised that c large number of the 9 findings are in the civil / structural' area. I 10 Examples of the kind of things that were found were 11 calculational assumptions, not representative of as-built. I 12 As-built fundamental design considerations were not evaluated. 13 There were discrepancies between analyzed and f_-) 14 installed support configurations, questions on rebar cutting 15 and placement, and development of the vertical respond spectra 16 to the steel containment. 17 There were some problems on a computer code and there 1 1 10 was a question of whether or not adequate response had been I l 19 developed in the vertical direction. l 20 I think the significance of that one would probably 21 be mostly on things that would be high in structure in the 22 containment, such as may be sprays, if there were anything. 23 CIIAIRMAN KERR: Are these the things that one would 24 normally expect might be discovered in the procots of a review 25 for an operating license? Or are these things that the staff ('h V Heritage Reporting Corporation l (202) 628-4888 l l e__--_

o 1 - 4 .- t i r 76- ..,.3 L-M 1' normallyfdoes notflook'at?. l ~ MR..HERMANN: I think the scope of the calculations l 2 t 1 '3 review, the design-based review, and the IDI:are'a. lot more in.' i '4

depth" reviews in. terms ot details >of how analyses were1 5

performed)in the licensing reviews. 1 1 6 Licensing. reviews,1I-generally'say,-more 1ook at .. \\. n7f ? criteria-and whether somebody' commits'to do something. And; 8' this wasta. lot more of a.look at anMimplementation. ~ 9.. CHAIRMAN'KERR: -Is that based on the. fact that most 10 organizations are considere'd so competent that one doesn't'.have 11' 'to look in this much detail?- i 12 I'm1trying to understand why.one needs,to do this for 13 TVA and doesn'tDneed-to do it in reviewing other-plants?- " 4 14; MR. HERMANN. I think-in the more recently licensed 15

plants, having.an independent designnins'pection, either by.the

- 16 staff or by a third' party organization,.is getting to'be a 17 fairly' current. practice. i ~ 18 -CHAIRMAN KERR Yes, but this doesn't sound-1ike it l .19 lwas just independent. It sounds to me like just a lot more 20-depth. 21 Independence can simply mean that you get a different 22 ' group to do the same sort of analysis. Unless I misinterpreted 1 23 what is going on here, this was not an independent; it was a 24 much more thorough analysis.

25 MR. HERMANN:

I don't think this was a more in depth ( Heritage Reporting Corporation (202) 628-4888 =--. _ - - -.

d,- 1 I 0 77 ~ /\\ 1 ,) 1 ' review by the staff-thancan independent design 11nspection that. 12

would have been performed by a third party, in terms of eithe~r-3.

staff or time.. 4 CHAIRMAN KERR: Are you telling me that the 5 independent: design review was not more in depth than'theTstaff I 6 'would normally do or it is more.in. depth?. 7-MR. HERMANN: I'm telling _you it is more in depth than '8 the staf f would ck). I guess what I am.saying is,_on a new 9 plant, there.have been cases.where the staff has done this kind. ~ 10 of review or.an independent third party ~has done this type of l 11 review. 12 When this review has been~ conducted,.when an c 13 independent design review by.a third' party or the staff is 7-5-' L14 conducted, this one is, I wouldn't say, much different in-l 15 detail than the'one'you would expect-at another facility. ] q 16-CHAIRMAN KERR 'But this particular on was in much think you said. 17_ more depth, I ' 18 MR. HERMANN: No '. You asked me about licensing at-19 the time of the plant. What I said is that the licensing 20 review at the time of the plant, at the time Sequoyah was 21

licensed, back in that time frame, the depth of that review 22 probably would not be as detailed as this review was.

23 CHAIRMAN KERRt So if the stamp were now licensing an 24 operating license for Sequoyah, they would probably catch thece 25 things. A(/ q Heritage Reporting Corporation (202) 628-4888

"s 3 y .j ? .j i 78 V' - 1E DR.'STEINDLER: As:a'new plant., 2-MR..HERMANN: As a new plant,'today. 1

1 3

CHAIRMAN;KERR I. don.t have any preconception.- I'm 4' justetrying"to find-out'. 5. MR. WYLIE:

Okay, 6-CHAIRMAN KERR:

Do you think the staff would have y ~7 caught these things? 8 MR. HERMANN: If they had done this kind of a review l I 9-at the time of' licensing. ,q 10 CHAIRMAN KERR: Sure. What.I am trying to find out: 11 is whetherlthey_would do this kind of review as a normal j 12 operating licensing review. j 4 13 MR. HERMANN: And-I'am telling you it depends on the l p' ' 14 time. frame. A plant that was licensed ten years.ago, no, you 15 would not have. Today, yes. { -i 16 MR. HERNAN: Dr. Kerr, I think the answer is no,-.the . 17 staff would not do this~ depth of review as a routine part.of 18 the OL process. 'But there have been several recent' plants; i 19 where we specifically did an IDI in addition to the normal l l 20 process. 21 -CHAIRMAN KERR: Was the IDI just independent or was 22 it in more depth than the normal review process? i 23 MR. HERMANN: It is always in more depth than the {' i 3 24 normal review process. Normal review is an audit review, i 25 DR. MARK: These items you listed are all paper c i Heritage Reporting Corporation L, (202) 628-4888 l

79 m ()a 1 . problems, such as lackLof records or lack of.having made the' 2' calculations. ILdon't think it is really the quality of the-1 y I '3 building that yousare looking at. 4-MR. HERMANN: Questions on'rebar placement and i I 5' questions on discrepancies between analyzed and installed i 6 support configuration-for. components I-think'are'probably 1 m 7 relating to hardware and not to just records. 8 MR. WYLIE:' I'think in all fairness, the staff did 9 find a considerable number of hardware discrepancies. One .10 glaring example is, and I don't'think he has gotten to it yet, 1 11-in the-mechanical area, pump design and cooler design, where i - 12 they were just not adequate. l l-I)J - 13 . DR '. LMARK :. Well, that'is the kind of thing which I ) 1 14' would think of as interesting and important. 'This doesn't seem-t - 15 to have that quality. 16 MR. HERMANN: This is strictly just'the' civil. area.' 17 There ista laundry list in.that-October 9th. report of examples 18 in-the civil area, things like problems with some tanks from 19 the way the tanks are supported. There is a flexibility =20- -problem with a floor, floor or wall. There is a variety of 21 things. 22 MR. MARK: Look, I'm only looking at that exact list. 23 You are talking of things which are real and need attention, 24 and:like you said, not on the board, but more laterally. 1 25 MR. WYLIE: I might add, too, I think, that the staff I l L(I Heritage Reporting Corporation (202) 628-4898 l

wa x - Ll 1 80- ! /'s ~ j l Tsj 1. has issued ~their findings and'TVA has responded..But we have .g

2' not had the opportunity to review these.

3 So we are a little at a-disadvantage toTknow what TVli 4~ 'is going to say about'these' things. 1 ~ 5 MR. HERMANN: .I believe it wasoan October 29th 6 . submittal,-and I don't think the. staff has. looked at it. The 1 7 ~ other thing that is going on is that this. week.the team isJdown.

j 8

at Sequoyah or Knoxville -- I'm not quite.sure'where_they are j 9 this week -- having a meeting on theso items,.on resolution and i 10-further information~that TVA.is providing'on the issues. 11 It was a set of things that was identified for TVA'- ~12 They have responded on'them. They have responded in writing to .13 them, and the staff is looking at that response now.- And where '(] 14 .it's going to go from there, I can't say. -15 DR. HANNUM: Mr. Mark, if I may.perhaps help:a bit, 16 TVA.having:seen in the civil' area'the questions that'were 17' ' raised, we believe that the' bulk of these are'in fact paperwork 18 concerns. i 19 There are several where we have not completed our '20 analyses yet, and there may in fact be some physical l 21 shortcomings that it would be appropriate.to correct. l 22 Wo believe those are a few. But this is still under ) 23-discussion, and NRC has not completed their review of our 24 response. And so I would not want to mislead you as to whether 25 there are many, a few, or a few possible hardware problems that h-x Heritage Reporting Corporation (202) 628-4888 j i w-_

- lf!' w .y 81 ' ,.h () l. need to be. addressed.1' i 2" DR'.' MARK: Thank you. 3' MR. HERMANN: I think a better flavor of.the kind of-c 41 1 things,:for-the details of:the things,fif.you are= interested in '5 them,'the' October'9 letter. essentially outlines theideficiency 61 'in'an area, discusses the problem, its generic implication as 7 Lto.whether it'might?be applied to other systems. 8- - ItisLabout that thick, and it certainly could not be 19' covered in the time'available. We were just trying to' pick' 110. some; examples.for you. 11: >(Slide) 12 MR. HERMANN: With regard to_the mechanical systems,. 13: the thing wectalked~ about earlier, there were questions: on.the j 14-Ldesign' pressure of'the ERCW,_the calculation methods to'come to 15; 1those design pressures, assurances that components meet greater ~ 16 ' design pressures. ~ 17 Another item is:there is a staff disagreement with -18: TVA's. position _on code interpretation for design pressure to 19-31.1. 20 A hardware problem if you want to get to that area, 21-is related to the coolers for the air compressors. Essentially j 22 the: jackets for the coolers have a pressure of one half of the '23 system, the normal system operating, or the system design 24 pressure which is 160 PSIG, where the jackets for the coolers o 25-are in the range of 57 PSIG. 6d) Heritage Reporting Corporation (202),628-4888 b

i 4 82

7'i.

'There also'were some-concerns on the effects,of non -(_/- 1-2- - safety related'ERCW piping,.and there are some; questions on i 3 mild environment temperature of-some of the: equipment in.the t 4L ERCW' pump house.' 1 -5 MR. EBERSOLE: Were thereLany questions about the 6 ' basic' nature: of the PNRD? 'Itknow everybody pays a' lot of I 7_ attention to getting water to critical components, but we find: I d 8 sometimes people-don't take the water away, by the same kind of '9 ' rationale.- They may manifold it,.they may take.it to pinch ~ 10 ' points-wheretyou have a possibility of stopping the flow and 11' stopping the;whole-process. 4 12 The exit path, in short, is just as important as the . }f s. L13 ' inlet path. And I wondered'if you examined the exit' path-here \\-) 14 to be sure you.had a clear one under all the. conditions faced-15' by the designer. 16 MR. HERMANN: I'm not positive of the extent of the 17 review of the. system design-people, but there were basically a 18' nuclear' type and a mechanical' type systems component, with a 19 systems engineer on the team. Part of their job was to look at 20 functional system flows and the pressures. 21 MR. EBERSOLE: But that is really not in the rule 22 book. It is just something you do. And my question is, is i 23 that done? The answer has to be yes. 24' 14R. HERMANN:- I can't answer it any more positively 25 than I know people that looked at the system design. Y l Heritage Reporting Corporation ~ (202) 628-4888 l,

( q ,l. q 'o y Dm 03 o l' (' Slide) 2. Regarding'the mechanical-piping components,.it's 3 probably a-little bit more general, components in: general,-'and '4 one of the areas.that tended to be problematic was1there:wasLa f i S; problem' identified;with how TVA'had reviewed a vendors record 6. for seismic qualification of some of the components..OneLof; p . 7; the big ticket items which has come out of'the: calculations 4 j 8 program, and I guess some'more problems identified'here,fthat 9 .there were' problems with pipe' supports. j 10 There was~aLbig-program,fas part of the calculation 11 ' regeneration program, there1were essentially'five thousand 12 fmissing calculation drawings. And.they were reviewed under 13 another program. And this again, I think, points out some'more- ,.d 14 problems in.the pipe support area. '15 (Slide) i 16 The last thing I'm going to put up is some discussion-j i 17 .on some electrical problems.- As.I undarstandLfrom Gene Embro,. 18 the team leader,.the problem they were most concerned about'in ) =19 the electrical area was the lack of groundfall protection on 20 the 480 auxiliary power systems. The supplemental information: 21 had been received from TVA, and I think that issue is basically i .22 closed. J23 I'm not quite sure-of the status of the rest of 24 these. These items again are questionable things that were R 25 looked at as part of the calculations program. And there are a L Heritage Reporting Corporation (202) 620-4868 f

i l l 84 ) p (,) 1 few things that motor starting voltage questions and some of 2 the voltage drop calculations I know were questions that came 3 .up as part of the calculational program review itself. And 4 again, there's some discussion on these questions in the 5 electrical area. 6 I think these are probably on the way to resolution. 7 I'm not sure of the status of what's transpired this week as ) 8 part of the closecut items, but it's the best of my 9 understanding that the electrical area is part of the IDI 10 really wasn't an area that the team had big problems with. 11 And that summarizes the IDI program. 12 MR. WYLIE: Any questions? 13 After the resolution or analysis, what do you plan to z., i ) ' ~ ' 14 do with this after you get all~the responses from TVA? 4 l 15 MR. HERMANN: I think first you know you have a 16 technical interfaces to see which of the problems will go away 17 from supplemental information and where you stand with them. 18 The action is that the items have been identified as restart i 19 items. Whatever items don't get closed from an analytical 20 viewpoint or providing supplemental information, of the ones 21 that are in that October 9th letter, will be expected to be 22 fixed before the plant restarts. 23 MR. WYLIE: Well, now, there's a number of these that 'l 24 have generic implications. What do you plan to do to those if j ?S there are still items that are valid that have to be resolved j cx s I ,\\_/ Heritage Reporting Corporation (202) 628-4888 l l l l a

7 g r .m --.\\3 i I t Oi 85-1

1 for restart?!

2 MR. HERMANN:. You're talking about generic with L 3 . regard to other systems at TVA? 4 MR.' WYLIE: Yes. -5 MR. HERMANN: I think they'd be-expected to address ~ 6 -them. '7 MR.1WYLIE: Any other questions? l 8- '(No response.) 9 MR '. WYLIE: Okay, thank'you, Mr. Hermann. l q '10 Let's move-along, and now TVA's Dr. Hannum will cover 11 the overview of TVA management reorganization' discussions -12 objectives and assignments and discussions as t'o nuclear" safety. 13 review board. structure and staffing. 14 I might point out that Dr. Hannum is filling in for 15 some of the people that were here yesterday that had to' leave 16 and go back to a meeting at the NRC, so he's pinch hitting to 17' some extent, although-I'mysure he's quite competent to talk in 18 these areas. 19 MR. HANNUM: Thank you,.Mr. Wylie. I guess I'am 20 pinch hitting. Mr. Gridley presented much of this material to 21 the subcommittee in much more detail than I will have the 22-opportunity to do today. But we do appreciate the opportunity ~23-to address the full committee to appraise you of our'overall 24-status with regard to restart. 25 Let me begin by saying on behalf of Admiral White j ) .C. Heritage Reporting Corporation (202) 628-4888 l

s 86' ' -)' 1 that we are very upbeat in terms of our prospects for 2 completing the, recovery actions and getting TVA back into the 3 nuclear power generation business. And I hope after I've gone through some of this material, you will understand the basis 4 5 for our optimism. 6' (Slide) 7 What I would like to do today is go through the basic 8 issues that have been identified and give you a status report 1 9 as to where we stand. As I'm sure this Committee is well 10 aware, several years ago we had problems where Sequoyah was 11 ' shut down because of the inability to meet a commitment for 12 environmental qualification. 13 Brown's Ferry was shut down because of problems in 73V 14 operator qualification. And Watts Bar was shut down because of 15 a series of employee concerns which required investigation. 16 The three units all being shut down for diverse and apparently 17 serious causes. Someone made the observation that perhaps i 18 there was a general management problem here which ought to be J l 19 addressed. That seems in retrospect wise and correct. And 20 when Admiral White was brought in, he summarized the basic root i 21 cause of TVA's problems as a lack of a sufficient number of ] 22 experienced managers to provido leadership and direction to l 23 TVA's nuclear program. 24 Therefore, in line with what Mr. Ward said in his 25 introductory comments, we agree completely that the fundamental bl v Heritage Reporting Corporation (202) 628-4888 _-____---_ L

77,-

~

~ = a ip os o i y: _g7 s !. s% ' Q

15 problem that TVA hat;-been.facin'g and must'~ address'is the basis

~ '2 management problem. ~The specifics.are in fact symptomsiof;this- '3- ' underlying cause. 4 LSo what I would like.to addressutodayLis, first, 15 where'we stand.with? regard ~to our corrections to the--managements s T6 system,.and that is_ basically summarineduin~.these first:three? ~7 Litems.- Establishing a.stableLorganization, putting-a. strong' 18 --management teamlin' place, and establishing.a proceduralizedLand' 9 ~disciplinedLway"of'doing' business. 10 You can't just fix'the: root cause. We.also;obviouslyE 11 have.to fix the symptoms and'.that's with regard.to the 12-technical < integrity of the plant,'andLJoe Ziegler from our- ~ L13 ' licensing;staf f cwill update you cn1 the. status on.some of;those. 14 .(Slide) ~ 15 'And of course, also, we must reestablish confidence

16

that1was apparently lost because of the large number of L

'17 employee concerns. So I plan to speak'to each of these areas 18 as part of this general update. j 19 First, with regard to establishing a. stable j o' organization, that is done. This is-the organization that is 20 21 in place. This organization has been in place in terms of its -22. structure for over a year now. It is redrawn from some things L l

23 you may have seen before,. lumping a number of the staff 24 functions into a single block so it doesn't look quite so

-s 25 cluttered. pg (L) Heritage Reporting Corporation (202) 628-4888 i s> n _ _. - =. _ _ - - _ _ - _

88 /~T "._) 1 It's been strengthened in terms of people. There are 2-now two deputy managers in terms of Mr. Fox and Mr. Mason, 3 whereas a year ago, there was only one. We have recently 4 brought in a new assistant manager, Joe Bynum who has a 5 considerable TVA background to help us oversee the operations 6 area. 7-This is supported by a specific policy and 8 organization manual and position descriptions have been 9 prepared for all personnel. The keys to th-organization, no 10 particular organization is ever right or wroag. It's an 11 organization that works, or one that doesn't work. The basic 12 principles of this organization are strong central direction, 13 competent people throughout, clearly defined responsibilities 7s 14 and accountability, and serious attention to management 15 training, motivation and evaluation. 16 (Slide) 17 The second piece once there's a structure is to put a 18 strong management team in place. I'll ask him just to flip 19 through a series of a few graphs hero that list the 47 new 20 senior level managers that have been brought in since January, 21 1986. 22 As she flips through, note the right hand column, and 23 you will see that this does represent a very substantial 24 backlog of experience. Mr. White has said, TVA today has the 25 strongest group of managers in the nuclear business in this (~)h Heritage Reporting Corporation (202) 628-4888 l

.) i l ,/-, 89 v' 1 country. .I don't like to quibble with Mr. White, but I would '2 correct him on one thing on that, it is not the strongest team i 3 in this country, it is the strongest team in the world. 4 Of these people, only nine are currently filled by l 5. line managers so the concern that has been expressed of a 6 -temporary team coming into patch that is essentially behind us 7 now. There's a strong effort also to identify within the l-8 existing TVA organization those managers with strong potential: 9' and they are being developed into excellent managers. 10 DR. MARK: It was said a year ago that you couldn't-i 11-get the kind of people you needed because of salary '12 limitations.' Has that bottleneck been broken? 13 ~ MR. HANNUM: Yes and no, Mr. Mark. There apparently + \\ 14 are many fools such as myself who are willing to work at i 1 15 government salaries. Hopefully, this is something that will 16 continue. These people have all signed on. We have not gotten 17 any relief from the government pay cap. 18 The problem has been solved in spite of the fact that 19 it has not been corrected. 20 (Slide) 21 One of the other pieces of having a competent 22 management team having the proper people in and of itself is 1 23 not sufficient. There is a very strong component of training. 24 I'd just note here for numbers, the training courses here are l 25 orientation to nuclear supervision, supervisory development C ( Heritage Reporting Corporation q (202) 628-4888 l 1 E j

,7,,--- ..y 1 90-QT% }g,j o 1 course,.sk'111sJassessment workshop, managing for excellence.. 2 Astyoufcan see, we.ran'many people through that during the- ~ ~ 31 first four months of th1s. And' during ' the next year,: extremely '4 ! ainbitious. plans ' for putting all of our managers through some '5-form of management development. 6' MR. REED: I didn't' comment'on this yesterday, but' 7 you add 1800 to 500,Lyou get.2300. Are'you telling me'that 13 there are 2300 manager level people 1n the TVA nuclear. ~ 9 . activity?. 11 0 MR. HANNUM: Yes, that's approximately correct. 11 MR. REEDS. Wow. 12-MR. ZIEGLER: 'Let me' correct that a little bit. The ,13 _ 1800 that are going to the.first segment are not necessarily '( l 14 managers. 'They're front line supervisors and in some cases ~15 that would to be senior level engineers.. Especially out in the ol'6 . field work so some of the senior level engineers would also be' 17 going to this training. 18 MR. HANNUM: Basically, anyone with supervisory 19 responsibility 20 MR. REED: Nuclear related? 21 MR. HANNUM: Yes. Total employment is about 14,000. 22 MR. REED: In the total TVA, you mean, or just 23 nuclear? 24 MR. HANNUM: Nuclear. 25 MR. REED: My God. Those numbers stack up way beyond / .(~y v Heritage Reporting Corporation j (202) 628-4888 l

_____:_=__

j i 4 91 (^T ( (f 1 some of the others I've heard, and this proliferation of people ] l 2 surely is strange, i 3 MR. EBERSOLE: Do you see why it's like being a J A barnacle on the side of the Queen Mary. 5 MR. HANNUM: It is not at all clear that after we 6 complete the restart efforts that the numbers will be anything 7 like this. But we still with the second unit 1 at Sequoyah, 8 three units at Brown's Ferry, two units at Watts Bar, we have 9 to plan for several years of very intense rebuilding. 10 MR. REED: And these probably include construction l 11 design people? 12 MR. HANNUM: Oh, yes, 13 MR. REED: And you do most of your construction? gs 14 MR. ZIEGLER: TVA's does their own engineering and 15 construction, so this would include all the construction people 16 at Bellefonte and Watts Bar. 17 MR. REED: Okay. It becomes more obvious when it's 18 that way. Thank you. 19 (Slide) 20 MR. HANNUM: Basically, the third with regard to 21 management is established procedures, disciplined way of doing .) 1 22 business. A great many things have been put in place in this 23 in terms of a talkdown procedure system, a tracking system, a 24 unitized tracking system for the entire office of nuclear power 25 to make sure we meet commitments, and formal and somewhat (3 (/ Heritage Reporting Corporation (202) 628 4888 ( l i

92 ,-f i (_) 1 ' complex system'for tracking conditions adverse to quality to be 2 sure that things are not simply noted and put aside, but that 3 they are responded to and fixed in a timely manner. -4 Configuration management system, the structure is'in l 5 place doing away with a rather clumsy two-drawing system as l 6 built and as designed drawings. The design baseline has 7 captured the design requirements, changed control boards are 8 put in place. j 9 I'would not represent that all of these procedures i 10 and disciplined way of doing business are totally effective 11 yet, but the structures are in place. 12 That in fact would be my summary overall for the X 13 management system. The structures are in place. An extremely 14 talented experienced group of individuals has been brought '15 together and this is being developed -- and I use the active. '16 tense of the verb deliberately -- is being developed into a 17 smoothly functioning team. 18 The next area, I will not myself -- skip on -- speak l 19 to, Joe Ziegler will speak to that in just a moment in terms of 20 insuring the technical integrity of the plant where he will 21 address the environmental qualification program, design 22 baseline program, the electrical issues, that is, cables and 23 the calculational regeneration, some of which Bob alluded to in 1 i 24 the IDI. l 25 The final area listed on here is reestablishing trust A lieritage Reporting Corporation (202) 628-4888

= - - ~ - - - 4 y j ,a ~ ^93 1 ,and confidence'in. management. The key reason this has.been a-2 major problem and-a' major program at TVA'of-course is'the I wealth of employee.c'ncerns generated.'at' Watts Bar. There were' - 3. o r_ 4 some five thousand individuals interviewed'during the' process 5.- of trying to capture all of the concerns ~. These-five thousand 6' interviews led to some 5300 concerns expressed, representing 7' some two'thousa'nd different issues that could be of safety '8 significance. Of.these, approximately half ---which is a very q y 9.- surprisingly'large number -- approximately half of these were L101 'coilfirmed'to have a degree of' validity. ..1 11 These.will when we.have closed;the. program, will have' -t 12 resulted in 1400 actual corrections to hardware system or paper-13 work as a result of-these, employee concerns. 'None of them i R

14-would have resulted in shutdown of an operating plant.- Some of k

15 them would have been reportable under NRC rules. 16 These are all captured and corrective action plans l 17 have been developed, are being reviewed, and the group which-i 18 oversaw.this effort.is continuing in force to make sure that .19 ' . appropriate corrective action plans and commitments to do that I J -20 is in place. That offort is rapidly winding down in terms.of ] f 21 the old program.- Since February, 1986, we have had a somewhat i 22 more proactive response to employee concerns, trying to track ) \\ 23 them back into the management chain, and the indications, if 24-you look at the indicators, suggest that that is very c 25. effective. j-I L lD-1 Heritage Reporting Corporation L (202) 628-4888 l L_ i_ _ _ _ _. _ _.

+ Q Y 5 r 94 '/ 6 1 'We are now down to approximately twelve' employee 2 concerns:being; expressed-per month,;which-is'a fairly nominal 3- . amount,.and suggests that we are gettingLthings backlinto the 4 proper.'line. 5. .MR. MARK Does this.-12: refer to Sequoyah?- D 6.. MR..HANNUM: No, that is' corporate. All of our sites 7. -and considering there are 14,000 employees, that in'factLis'ai 8L very small number. '9 Basically,'overall in the management,.let me"say once' l10 - more, we believe.'the basic. problems have been' addressed, are- -i -11 under control. We still have a. lot of work to'do to get a ) 12 smoothly functioning. team and to get'the. discipline and 13 procedura11zation throughout the organization.that we expect:to; = 14; see. 15.. I personally would characterize the situation of the 16 TVA today has a good management = structure, but that that is not. -17 acceptable to Admiral White. Admiral White will settle for 18 - nothing less than an' excellent organization, and.he is -- I ' 19. . don't think any-of us would disagree with that -- that the 20 visibility TVA has, we should not return to operation until we 21 have a first class sharp operation. 22 The one area that remains of some concern because it-S 23 is not addressed by any of these programs to date is the matter. 1 24 of operational control as we return to operation. The .25 operating staff, themselves, have not operated these plants for 4 Heritage Reporting Corporation .(202) 628-4888 - =-

i 95 () I two years. And that is why we are anxious to proceed with a i >2 non-nuclear heatup to get back into an operating mode and work i 3 out some of the rough edges that are likely to be present in 4 that. 5 And with the substantial extra management attention 6 that will be placed on operation, we do.not believe that either 7 that or the initial nuclear operation will constitute any 8 unucual risk. 9 Let me proceed now -- well, first of all, are there 10 other general questions on the status of the management 11 recovery program of TVA? I propose to proceed at this point to 12 the Nuclear Safety Review Board and the review functions if 13 there are no more questions on this. f. (J 14 CHAIRMAN KERR: No questions? 15 DR. MOELLER: When you go to these nuclear safety 16 review boards, is there a separate one for each plant? There 17 is. Now, is there overlap? f 18 MR. HANNUM: Let me speak to that very directly. 19 Nuclear safety review boards, first of all, have two rather 20 different functions. The Techspec review board that is 21 required to review the items as specified in the Administrative l 22 Tech Specs. We are also, by direction of Admiral White, his 23 ' principal advisor in all matters dealing with nuclear safety. j 24 In terms that the ACBS has used, I am TVA's nuclear j l 25 conscience. In doing that second function of being the 1 /~T l V l Heritage Reporting Corporation (202) 628-4888 L l l l l k. - ~ ~~ ---_-_ - - _ - _ - _ l

i 96 l (_) I conscience, the nuclear safety review boards are not concerned 2 with schedules or costs. We are concerned with the 3 restructuring, the development of an excellent system, 4 procedure, attitude and record with regard'to nuclear safety. t 5 As far as the boards, themselves, yes, there are 6 three active nuclear safety review boards,-one for Sequoyah, 7 one for Browns Ferry and one for Watts Bar. I have not 8 organized a board at this point for Bellefonte. That will be 1 9 done as we approach the application for an operating license. 10 For each of these boards, I chair all three of them. 11 I have no other function within TVA other than the nuclear 12 safety boards. This is a full time function for me. Each 13 board has its own dedicated technical secretary and I have a l -,s 'V 14 very small clerical support staffing serving the three boards. 15 For each of the boards, I do have subcommittees, a 16 subcommittee addressing unresolved safety question 17 determinations when the plant determination is that there is no 18 unresolved safety question. So of course if it is an 19 unresolved safety question or determined to be so, that goes 20 direct to the Board and not to a subcommittee. And then other 1 21 boards on generic topics and security matters. 22 Each of the boards, as I say, I chair all three 23 boards. The boards include the site director in each instance. 24 They include a senior manager from the division of Nuclear 25 Engineering, they include a senior quality assurance manager b,m Heritage Reporting Corporation (202) 628-4888 l' t L___z__ - _ _ -

+ 'k:

97-

[ li Land'other senior managers as appropriate. 'Each board has five. 2. senior'non-TVA: advisors. These are called advisors only 3 ~ .3 'because-of the peculiarity of the Federal Advisory Committee j 4

1. Act which.I'm sure you are aware of, andLwe do not feel 1that it 1

5 21s within the' intent of thelFederal' Advisory Committee Act that. '6-this beLa' Federal advisory committee. '7-And if'it's a Federal' committee that-advises ~, it 8- 'can't-have non-employees on it, so these are advisors. They-9. review all material as. members, they participate in all 10 meetings, their voice is heard'and I can assure while forma 11y' I 1 - 11' they do-not vote, their opinion is as important to me-as any of 12-the members and'is so represented.to the managers. - sq 13 MR. EBERSOLE: Tom, I'd like.to'ask you a question.- 0 14 You'say you'd like~to get in a lead role ~from where-you'are. 15 There are large numbers of unresolved safety issues as a' class i m 16 of things to do, there are generic issues, there'are' lots of-i 1 r l 1 - 17 things hanging in front of NRC. Would this group of people f 18 stare at these issues and be less-dependent than the general 19 industry at large is on NRC coming to points of resolution, ,.,c a 20 realizing that some of them can be done with a'more or.less

)

21 -impact on economics than were' thought to be by other utilities. ) J 22 Do you follow me? 3 1 23 I'm asking you whether you will take these stony j i 24 matters and stare at them with your group of competent people 1~ 25 and say, ah ha, this organization is going to fix that one. L -k - Heritage Reporting Corporation L (202) 628-4888 l? _---x_

i i 98 T_j 1 MR. HANNUM:.That is a credible scenario, 2 Mr. Ebersole, but I would not propose to commit that we are 3 going to solve the problems for the industry. 4 MR. EBERSOLE: Even on your own basis? ) 5 MR. HANNUM: It is certainly within our charter to 6 look at whatever we wish to look at and one of my boards has 7 had an initial look at alternate containment requirements.such 8 as are discussed in NUREG 1150. We have not come to any 9 conclusions or any recommendations to the manager. 10 The'only practical constraint that I would suggest 11 applies here is these boards are by charter by forum advisor to l 12 the manager of nuclear power only to-help him in formulating .13 TVA policy.. As to whether TVA wishes to take a lead, that is a ] , ~ ~,. q 14 matter that is beyond the scope of an advisory board. l 15 We can advise him on the technical aspects, the 16 technical considerations from a safety point of view, but since 17 we do not consider costs or schedule or other impacts, other 18 management impacts, it is properly the management's role to 19 weigh those recommendations in light of other matters. 20 I don't know whether that addresses your suggestion 21 here, or not. I guess the most constructive positive way I can 22 respond is key unreso]ved safety questions that are floating 23 around in the industry, these boards are going to be interested j 24 in them and concerned with them. Whether we.do anything i 25 constructive with regard to them, I wouldn't want to make any q em - L.) j Heritage Reporting Corporation ) (202) 628-4888 J l ______o

m,. 3. -t- .q. ; i.l _.r s , 9 9 -- j 8 .i-promise.. 2 MR. EBERSOLE: You say you're waiting for the 3' . regulatory people to find the solutions? d i s-4i ~ MR. ' HANNUM a ' No, sir,lwe'are'not waiting'for.the i] -m .] 15 regulatory authority for anything. j q [ L6 ; .MR. EBERSOLE Well,'USIs'and. generic matterstare i i 3

7. '

items that are;home, Mr. Chairman. y 5:

8 MR. HANNUM:-

These boards =believe there is something' 9-that. seriously affects:the safety of'a TVA nuclear "1 10 ! installation, we.will addres's'those, not from a regulatory SC', 11' point.of view, but from a safety point of view. TVA isJ '121 responsible for;the safety of these plants, not the NRC. -7 113 MR. EBERSOLE: 'I'll give you just a little bit of:an. M .14' ' example. You know the French come'out and solved their seal .15 problems, main pump. seals'by putting systems'in which are 16 patches admittedly to keep the. seals cool ~and lubricated, and 17 'we're still hassling about that. How long we'll hassle'about-o

18..

that, about TAP ~45, about anything else is' unknown.by anybody. 19 Certainly, I would think some fraction of these -20 issues would be picked up by the more-conservative utilit'ies 21 and solved by them, if necessary, unilaterally, p 22-DR. MOELLER: Back on the board here, this Mr. 23 Wilson, the secretary, is the secretary for all the Boards?' ![ 24- .MR. HANNUM: No, sir. He is secretary-for this board 25 only and that is his only function, that is a full time job for p Heritage Reporting Corporation (202) 628-4888 l: y LL =_

^^ w. i i J i' ' 4 100' 'a i f"y 1

him.

q f.

2 -MR. MOELLER: So you're the only one that's" common? j 3 MR. HANNUM: At the moment, I amithe'only that is LTo'try t'o.be sure that we don'ttbecome too provincial, 4 common.' 5 I do invite either members or advisors.to attend the other. 6L boards from time to time and.have anLinformal. sort of' rotation.. s 7 DR. MOELLER: Now, when you say the advisors.atten'. I d 8 all meetings, how frequent are the. meetings? j 79 MR.'HANNUM: _The Sequoyah and Browns Ferry boards q 10 meet approximately every two months. 11 HDR. MOELLER: For how much time? 12 'MR. HANNUM: Two days. 13 .DR. MOELLER:' Two. days. 14 'MR. HANNUM: 'The Watts Bar board, because we're on a 15 somewhat more elongated schedule at this point, is meeting for 16 two days every three months. i i 17-DR. MOELLER: Other than yourself, these are all j j ' 18 people with two hats, right? 19-MR; HANNUM: Other than myself and the technical 20 secretary. '21 DR, MOELLER: Yeah, right. .22 MR. HANNUM: -These are people of two hats, yes, j i . 23? I discussed at some length with the subcommittee 24 yesterday, and I won't bother trying to repeat that today, a -J 25 ' number of examples where advice that these boards have provided ) Heritage Reporting Corporation j (202) 628-4888

i 4 J e i >;2 '101 p: TJ 'l-toLAdmiral'. White have influenced'him, or.'he'has perhaps i I 2 ~ coincidentally or perhaps in1 response has made changes in the 13 review process in terms of a -number of specific technical 4 issues, in terms-of the internal lTVA organization and in, terms 5 of his personalicommitments to. insuring that we do.closeiopen: 6 ' issues and respond to corrective' actions. 7 Ferhaps it would have happened without the' nuclear 8-safety review-board, but I like to think we.have at least-9 l helped the overall organization by providing visibility to a. 10 ~ number of; key issues, and certainly where the boards have had 11 concerns and'made recommendations, the management structure 12 starting'with Admiral White has been extremely responsive to 'i 4 21 "13 these recommendations. h 14 I'd like to say just a word beyond.the' functions of 15 the nuclear safety review board. A review board, whether it be 16 TVA's nuclear safety review board, or whether it be the NRC, 17 cannot impose safety on an operating line organization.- L18 -Diffuse safety where it is safety is everybody's business but I 19 it-must be somebody else's responsibility doesn't work either. I 20 We are trying very hard within TVA to make. clear 21 accountability for safety. And what we have at this point is j l 22 there are very clearly defined safety responsibility and 23 accountability. For the actual operations and maintenance of .24 the plant, it is the plant manager who is responsible for-that. lj 25-He knows that and he has his review boards, and a review staff l x } I Heritage Reporting Corporation l (202) 628-4888 l

yy p-3- r "9 _ "'i 's g / E_ sp' s y. 102' to assistLhim in making intelligent-decisions. .1- > 21 We' oversee.that to be sure thatithey are.in fact [P y" l3 -concerning themselves with safety and advising him in that- ~ ~ !4; 3 regard.y a 5 As far-as the technical integrity of the' equipment, 6~ within the: plant, that is the responsibility of a project .7 engineer. The project engineer'who'is physica11y'on site'at' 8 the plant andfhas delegated staff from the division of' nuclear'- 9 engineering'to support him in that. L10 1The core itself is the responsibility of the fue'1s.

11. ' branch and:we are working to be sure that there is an adequate i

12 interface'and exchange among.all three of these: groups. .g .13 '. There are other review groups within TVA which.I. 14J should' ment' ion in passing. We do have an.IS'EG, independent 15 safety. engineering group at Sequoyah, and one is being formed' 16 at Browns Ferry and will be formed at Watts Bar. 17-We have a nuclear managers review' group which is the 18, ' follow on the carrier over from the old nuclear safety review 'q ] 19 staff, which is available to the manager to perform in depth j 20 reviews and studies as assigned including areas covering 21L nuclear safety. 22 The manager has other ad hoc means at his disposal. -] 23 'The division of nuclear quality assurance does special in depth j j 24: audits in areas and we are of course working very closely with j 1 25 INPO, N RC,' and other external review bodies. O' Heritage Reporting Corporation (202) 628-4888

.O, i, g m.. ~ 103 F-4

1

In conclusion,1with regard ~toLthe nuclearLaafety

'2-review-board and this. general function,.there is'a safety c3: conscience. That's me. I'm alive and. technically.well, 1 1 4

physically not in too bad a shape, functioning and I believe

.5 ' effective in! terms of TVA's operation.

l 6'

.The line accountability for safety is clear..The

7 Individuals know that theyJare. responsible-and they are o

d 8 accountable. And the procedures to implement-that,, to: give~ 9 .them'the support they~need,.are being put in place. Again,iI'- ~ 100 'usenthe-active tense of the verb'there. 1 11 Within the review functions themselves,~there are a- '12 numberJof' redundancies. These are.deliberato. 1TheyJare . <~s- '13. ' reasonably coordinated. I: meet with~the heads of the other U-114 . review l groups to make sure that.we complement each other'.rather L. .15 than compete. 16 That concludes my introduction on the nuclear safety j 17. review board.and:its function. I'd'be glad to answer any 18 . questions on'that if there are, before turning the microphone IV Lover to Joe Ziegler.to fill us in on the specific TVA restarts 20 issues. 21 DR. MARK: You mentioned, but didn't elaborate Tom on 12 2 ' security provisions. And I know you don't want to elaborate in i ? 23: the open, either. l L24 You have, I presume', to follow the normal rules about 2 51 fences and lights and guard numbers and all that sort of thing, 4 O Heritage Reporting Corporation 1 (202) 628-4888 l .l

J j d ] 104 .?"y 1 1,f. 1 and what in. addition to that can you say.about surveillance or. j '2' surveymof: insiders. l 3 MR. HANNUM: 'Let me try to address that in-this:way, l 4 Mr'. Mark. I'm not qualified to give you a. detailed' description ~ j 'S even.if it were proper to.dosso in an open forum.. I am not.the I i L6 proper person tongive you a' detailed briefing on that. We l 7, could arrange that if you wish.- i 8' Let me say that a year ago,1the security was part of 9 .TVA's.overall.public' safety office, and was not to the standard 10 ~ that Admiral White believed was appropriate for a nuclear 11 operation. He has made arrangements through the board of' 12 -directors to have his own security director at each site who 13 has functional responsibility for all of :the security forces. -sy g 14. We have an overall corporate security man who is' now on ther 15 staff. 16 We are undergoing a number of upgrades in that area l 17-so as to have what we trust will be a fully satisfactory 18 -program. There have been problems in that area, but those are 19 recognized and being corrected. 20 As far as your specific question on what we are doing 21 relative to insiders, other than a pretty strongly implemented I 22 drug program, which is of course only a small piece of that, 23 I'm afraid I'm not-competent to respond. 24 DR. MARK: Well, you have, I'm sure, the usual i 25 surveillance of people carrying things in and out? l (2} Heritage Reporting Corporation (202) 628-4888

i i 1 p -g ,,} l 4 si l 105' l e $1' MR. HANNUM: .Oh,.-yes. ~Oh,-yes. 12. DR. MARK: :And you.-have. background 1.._.;s on'tho' 3 + j p 3~ Lpeople who are really in~the sensitive areas?. Y 'J 'MR.;HANNUM ' I believeLso. Joe, can.you help;me onL j 4 55 tliat? l 'i 1 MR. ZIEGLER: I'believefthe regulations now require { i .71 an' FBI 1 check'on all' the unescorted: nuclear plant' access, but' 1 d 8 we've done'that all along being;a F'ederal. Agency, so yes,.we j 9~ do.: -j .10 DR.LMARKt.LWell, that's probably.enough.-- Thank you. ] a 11, ~DR. SHEWMON: Could you onlarge a little bit.onLan .j -l 12 FBI check?- I hope you don't go tofa Q clearance checktfor-all V. - 13 of these? l 1, l 21 4 ' MR.,HANNUM: No. 1 i 1 f15. 'MR.' ' Z IEGLER : No. I'm not exactly sure of'the D 16 terminology used, but -- I l 17 '- MR. HANNUM: It's an indices check. R 18-MR. ZIEGLER: The FBI has done TVA's security checks ~ l 19 .for as long as I can remember, even in the non-nuclear' i .20' employees. 21 DR. SHEWMON: But you don't know how to calibrate 22. ~these or compare these with a Q or an L or what it is? i 12 3 MR. ZIEGLER: No, I don't. l 1 -24 DR. SHEWMON: Thank you. 25 MR..HANNUM: My impression is it is certainly not a }l' Heritage Reporting Corporation (202) 628-4888 l E.. 1

p '106 G But it's in that ball As - 1- .Q, andHit may'be?as; strict as an L.' 2 ' park. Which side of that it falls on, I'm not sure. 3 DR. SHEWMON: Thank-you.- L 4 MR. EBERSOLE: Dr. Hannum, I guess I need to get my? L 'S-head cleared here about what is'the meaning of a nuclear safety 1 ~6 conscience. -7 Let me'try to tell you what.I.think it-is. I : take" it-8 .your charge-is to dig up, explore, analyze, make findings about 9 ' safety issues,-and bring.them before the line, and make a' clear .101 - presentation of the. severity, etcetera of these issue's, and. 11 hand it;to them, and that is the conscience. I-ratheriam. 12 inclined to think that conscience really resolves in. Admiral 13 White'and the several people that1you call the project h-14 engineers.at the plants, the latter,being subject to White's 15 policy-directions as to what they'can and cannot do. 16 Am I correct? k 17 MR. HANNUM: Let me correct one detail in what you i 18 ' describe, Mr. Ebersole, and then elaborate on one point on 19 that. -20 The detail is we bring those matters-to the attention' l -21 of the manager, so that's more precise than to the attention of i 22. the line. ] q l23 Now, incidentally, because line management takes part l H24 in all our meetings, they know about it, but formally the i '25 recommendations, the advice, the matters wn bring to the .p gi Heritage Reporting Corporation .j (202) 628-4888 i ,1

4 4 i j i I 107 1 , '~N i (_) 1 attention of the managers. And formally, you are quite right, j i 2 if he chooses to ignore that advice, all we will do, and we i 3 will do it, is we will continue to track the item and we will-4 continue to remind him every two nonths in writing with copies q 5 to the general manager and the board of directors on the public 1 6 record that we still are concerned, and Mr.. Manager, you still l 7 'have not addressed the problem. j 1 8 MR. EBERSOLE: Well, as a means of assessing your 9 success or lack of it in this endeavor I think it would be i 10 important to us to see at least a part of that stream, at least 11 in the first months or years of your operation, to see the i 12 caliber of the issues, the solutions. We have to look at the 3 13 totality, not where you stop, but where he stops. And I don't i 7s '_] 14 know any other way to do that. 15 MR. HANNUM: Of course. You are absolutely right. A 16 recommendation from a review board in and of itself means f 17 nothing unless there is some action, and it is the action which 18 should be judged. That is recorded in our minutes and I would 19 be glad to keep you posted in whatever manner you feel so that 20 you can judge the adequacy of the responses. 21 MR. EDERSOLE: Thank you. \\ d 22 MR. HANNUM: Mr. Ziegler, if you can help us with i j 23 this. ) 24 MR. ZIEGLER: The first thing I want to talk about is l t 25 just give you a very brief indication of where we are. I think 1 1 I f~f \\-) l Heritage Reporting Corporation (202) 628-4888 1 I l

4 108-li l' you're fully familiar withethe problems that got TVA in.the. 2 position we're in with all our plants' shut down. i ] 3 Our get well plan, if you will, we call it our 4 nuclear performance plan, it's several. volumes. The first u 5 vo'lume deals with TVA corporate. Volume.2 deals.with Sequoyah_ t 6 in particular. Volume 3, Browns. Ferry and there's a volume:4 l 7 not yet been published that will deal.with! Watts Bar. j 8 (Slide) 1 9 Basically, you can divide what's.in those' volumes- )

l 10 into three particular areas, organizational improvements, 11 programmatic improvements, specific plant condition j

12 improvements particularly in the volumes dealing.with the' f 13 plants. O 14 , I want to give you a.brief discussion on four of the 15 technical more programmatic technical issues which the ACRS 16 subcommittee'had expressed some interest in. 17 (Slide) 18 The first one, environmental qualification of 19 electrical equipment per 10 CFR 50.49, I'll go through, you're 20 all familiar with it. I'll go briefly through the components i 21 of the program at sequoyah. 22 (Slide) i 23 We begin with a complete review of all accident l 24-scenarios, all possible harsh environments, all 1E equipment 25 that comes under the scope of 50.49 and identified the () l Heritage Reporting Corporation L (202) 628-4888 o (1 b l

l_ 109 rt ,; V (l' equipment-to.be.qualifled. That will be the' safety related a , 2; electrical equipment'under 50.49 that is' in the harsh .i 1 j 3 environment as defined by the regulation. 4' (Slide) 5-We. develop qualification guidelines, then,. basically '6_ >proceduralized, the guidelines by which we would do our 7 qualifications so that we.would be consistent throughout the- '8 program. -The documentation'of that we put in an EQ' binder 9, system which -luis all the requirements, comprehensive records of '10-what is required.both to-originally qualify 1the equipment and 11-to. keep it-' qualified. 12' lAt that point in time, once we have all the- -13 documentation. records ano know what we have out there once the 14 evaluation has been done, then we would implement any hardware 15 changes or other repairs, modifications, or replacements as 16 necessary to make sure that what we have in the plant is indeed 17 qualified. 18 The final step in the' program is the qualification .19 and maintenance data sheets which is the. process we'use to 20 . identify the maintenance requirements to insure that-we keep 21 the equipment qualified throughout its plant life, and that 22 includes both corrective maintenance requirements and 23' preventive maintenance requirements. 24 DR. MARK: Now, Sequoyah was in an operating status l 25-at one time. This program is subsequent current? C Heritage Reporting Corporation (202) 628-4888 l

N.(' 1 /> g 'r: y ~ og 'p c t c M n N 110 i o 1A ia: MR.-: ZIEGLER. Th'e program '~as.;I'm ~ laying it. out right ' <i"+ G! : hererisisubsequentito:the' time Sequoyah shut 1down. q F c3L -DR.: MARK: Have you in.theIcourse'of this! subsequent

b/ w; f 4 1 program found: things-which demanded'changeLcr. replacement?.

15[ ' MR..ZIEGLER: There.have.been hardware changes. .j Mt 6: There've been, Ildon'tEknow all-the examples.- I know that we

7' changed out someitables'forfsubmergence reasons.

.I know:a' lot 8 'of theJadditional reviews that'were-done since August'of~'85 9-when we shutLthe plant down. includes-some identification of-n. ( 10 harsh 1 environments due to otherLthan design basis' accident 11' scenarios. Itlwas accidents but they.were_not the Chapter 15 z 12 ' accidents ~in.FSAR. l13 DR.' MARK: 'So you-would~believe'that the plant is in 14.' better shape now than.it had:been?- ) g n 15 ' ' . MR. ZIEGLER As far as: meeting the regulations,- '16 .we're definitely in better. shape. 17~ DR. - MARK : : No,-I mean, in~ meeting the real life. The -18 regulations are -- j 19 MR. ZIEGLER: I'would1 hope to think that we won't 20L have.thesefaccidents,lbut should we have them, yes, we're in 21' mucn better shape.. 22 DR. MARK: Fine. '2 3 ' HMR. EBERSOLE: 'However, you did not get rid of the I

24-terminal boxes and boards that got wet as a result of the a

1 ( 25 .conde~nsation? Heritage Reporting Corporation (202) 628-4888 u L

m,-. < w--

• (

))

c.

s q_ t x .s . ;a ; "wc s ? r .u 4:ff f-f' : " - ~- a . 111' 9ess i "Q, '.1 1 MR. ZIEGLER:. I_?think that was; discussed yesterday. l2<

I'm not the ae'signiengineer,;buttwhat we'have has been 3L qualified.I~think.

Now,; whether it's; qualified to ever'ybody's? q 4-satisfaction,fI don't know. i 5: MR.:EDERSOLE: In other words, it has been passed on- .t '6' by?the Staff? '7 '501. ZIEGLER: I believe that's true. ~EU MR. EBERSOLE: Whether-it's. wet or. dry. 9- . MR.- WYLIE:-.Well,.I think they answered!. yesterday 10'

thatfit's been environmentally; qualified.-

11-MR. MICHELSON: Yeah, they' claim'they'did. il2 MR.'EBERSOLE: In the wet condition. 4 13 MR. MICHELSON: Yes.; The.only; question:was the. )U E14- ' nature'of:the test and.they'didn't.knowLthesdetailsLenough, tot 15. assure us I think one.way or the'other. 16 MR.-WYLIE: Yesterday? -But~~resumably an appropriate. 17-MR. MICHELSON 'Yes. p ~18 ' . qualification test was performed and the' equipment. passed the' 19 . test. 20 MR..WYLIE: We could get a copy of their 21 qualification test, if you want to look at it. f 22 MR. MICHELSON: Yes, that would be the.next step. I 1 ?23: believe that would be appropriate. 24 MR. WYLIE: Does he know which equipment you want to '2 5' -see. i k 7 Heritage Reporting Corporation (202) 628-4888 l --____o___.u____ d

.s n ? ] i 112 P 'y. MRe: MICHELSON:, Terminal' boards:inside'the. "a -. containers. 3-MB. WYLIE:.Inside the containers. 4 MR. MICHELSON: 7 hat are in the boxes _with vents. E i' 5~ MR. EDERSOLE:' With holes bored in them. 6' MR. HANNUM: We'll be glad to-provide you with 7 detalls on that if it would be of interest to the committee. 8: MR. MICHELSON:- Now, the details we want are the test i9 that was used to.do the verification. 10 MR. HANNUM: Yes. There have been qualification 11-tests run on terminal blocks, but precisely _what they are, we'll have to get you that information. 12 '13 MR._ZIEGLER: Right. ~ Y h 14: _ Program status to date on the EQ program.the 15_ -implementation is complete. The binders have all been issued.- 16 There's somewhere between 90 and 100 binders. They're very -17 voluminous. There are'_a very few hardware changes that remain. 18 I think the ones that do remain'are some additional Raychem ) 19 splice changeouts. 'l 20 And that's about.it on the program. TVA has 1 21 certified to the NRC that the program is complete with the .I 22~ exceptions that I just mentioned, and informally at least we've .23 been notified by the staff that they think our program is very 24 acceptable. 25 (Slide) .) Heritage Reporting Corporation (202) 628-4888 i i

1 .,u.

.e s

113- [ 1-Next slide', I want to talk a little bit about'our n 2. ' design baseline'and'verificationiprogram. This program;was ^ 3 initiated to' correct problemsfassociated with design control l a .4-process at Sequoyah, actually'at all'TVA plants,-but here'in[ 5 .particularLat'Sequoyah, associated with the. plant:sinceLthe-

6 operating license was issued, the. modification process 7-basically.

~8' The scope of the program'was identified after- .} ~: L9-evaluation by.TVA staff to be 37 systems and those are.the: 1 l 10 systems required for safe shutdown of the plant and the'FSAR. q - 11L accident. mitigation systems and to insure their' i 12' 'functionability. The components of the. program were to conduct -l 13' . system: functional walkdorns. basically to; insure that the plant-Qi,. -N*= 14 'and the drawings of'the safety-related portions of these' H . lb systems' matched.. We want to know what we'have iu the plant, 16-develop the baseline and.make sure it's' correct. y d 17 The topic reconstitute design criteria, was not an l d 18-effort to change the design criteria but basically to put one 1 19 set of design criteria in a place where it could easily be 1 20 retrieved and found so that we would know what that' design 21 criteria was. 22 There were reviews of all modifications since the 23 operating license as well as any modifications that were in j 24, process and any modifications that for whatever reason the 25;. priorJty had changed and had not been implemented to make sure Heritage Reporbing Corporation (202) 628-4888 1 -.__u.m___.____.._._.______m

l 114 ,.y (._f 1 that we knew exactly where we are and where we were going, in ] 2 order to pick everything up. I t 3 There were also system technical evaluation reports, J 4 we' termed it, STERs, on each of the systems once all the 5 .information had been put into place'to assess the results and 6 address any discrepancy from the results of the walkdowns and l l 7 other reviews that were done. And to make sure that these j l 8 systems would function when called upon. i 9 The program status to date is that all reviews are 10 complete, all STERs for the system evaluations have been 11 issued. We have a punchlist closure system trying to close out ] I 12 all remaining open items. I think there's about 100 now that 13 either have not been closed out or require some further t T s~/ 14 evaluation. 15 And the bottom line here is that in our view, there i 16 were no significant hardware problems although there were quite 17 a few discrepancies identified in design drawings and the 1 18 actual configuration of the plant. And we believe this effort l 19 has put us in a better position to continue operations. ) 20 I might mention that we have modified our design I i 21 control process so that the loop is closed when we make j I 22 modifications that the designers design the changes, the 1 23 changes go through a work package review process that the i 24 designers are involved in, the modifications are made and the ] 1 25 designers get back into the process to upgrade the drawings and (^h l A_/ i Heritage Reporting Corporation (202) 628-4888 1

1 3 w 115 1. make sure that,the design was actually implemented as wasi 2L intended., So.that the' loop is closed:there,-and that's not 3' 'necessaril'y th fway it's always beenlat TVA. 4-lMR.!WYLIE l Joe, how much more' time do youothink -j 5: .it'll take to finish? '6 Y2. ZIEGLER: It depends on how many, questions I'get 7 'on cable' testing. 8~ MR. WYLIE: Mr. Chairman, whats your desire? 9 MR. KERR: My desire,is~that he have one1more minute. l10 MR. WYLIE: One more minute. We better skip'to the: 111 -civil calculations. L1'2 MR. ZIEGLER: Let me just quickly go ahead on the .13 ' civil calculations' program-scope. Basically TVA went'through a. - 0 l 6 1. review of all engineering calculations at Sequoyah 100 percent- ? 4 15 on electrical, mechanical and nuclear calculations. We did,not 16 include 100 percent of the civil originally, and I think now 17-we've gone back to do that. -18 There were 5800 rigorous-pipe supports that were done 19 by a vendor during construction of the plant, and through an 20 error, those calculations were destroyed. It was TVA's error 21 in allowing the vendor to destroy them. We have regenerated 22 those calculations. We've done a comprehensive consistent 23, design criteria covering the original designs specs, B31-1, and - 24 eny - ptions to it, plus IE bulletin 79.02 79.14 requirements 25. that came after that. O Heritage Reporting Corporation (202) 628-4888

~, j s K ~ y y [ 116 M' .1 'We've! identified'57.0Ltotal support modifications: 1 42 'that'll need to be made. The status ~toLdate is thatLthe1 design

3u -criteria have been submitted to the staffi.

I think-we're still': r 4 . waiting for formal. acceptance of that.: We've identified based: i 5 'on that design criteria 240 modifications 1 required;for post 'l i 6 restart. Based on a restart criteria which'iscan interim =7

criteria on 79.02 and.14 not.unlike'other utilities have'had.

8 So the question of safety factor I think' align forf oth. b 9 There are'330 modifications that'are required for i 10-restart. 'We think we'reJon schedule here. And that.we'll..have 11' them in time.for' restart. -Well, we will have them in time for. 12 restart whatever:that time: takes to be, but we-think we'reTon 13 Lthe way;and will have them complete in a. matter of a few ~ ' n". 14-months.- 7 15-That' basically is the program. That's it. Any 16 questions? ' 17, MR..WYLIE: Any. questions? 18 MR. REEDS. It's not a question,. it's a comment. I '19 was impressed in the subcommittee meeting yesterday by the get 20 well. activities of TVA. And you've gone a long way towards 21 implementing our letter comments, previous letter comments. i 22 One thing that I was impressed by is they've gone to 23 aptitude testing and it looks to be on a fairly sophisticated j 24 basis. I think that's going to help them a lot. q 25 I do think perhaps we might study further this issue j q Heritage Reporting Corporation (202) 628-4888 I

1 ~,, u-S .f'. x: 117' 1}"$ f); 1 ~ of cable testin~ -and.whyl.the high voltage was used,;andwhether' g ,2 .it.shouldfever'bejdonelagain.: i o o 73 L ' . CHAIRMAN KERR:.Any"other questions or comments?- 74: .(No: response) 5' ' CHAIRMAN.KERR:- ~Thank:you,7 sir. ' 6 '- -MR.oZIEGLER:' Thank ; you.:- s i7 ' CHAIRMAN KERR: Somebody here,;one of the older-

8; members of

the Committee-is suggesting a break..Maybe that's"a 9

good' idea.- Let's have ten minutes.. s 10-

(Brief. recess is taken)_ CHAIRMAN KERR Gentlemen,.I refer you to tab'7.

.) 12 ~DR. LEWIS: Sir? 13 CHAIRMAN KERR: Yes, sir. a pi ' I14 DR. LEWIS:..May I'make a~ point before you even begin,,

15l this71s.a point of personal privilege.

On page 23 of-tab 7, it 16 , quotes me in the transcript'as saying virtually nobody other f i 17 than the: Nuclear Regulatory Commission understands thelwords, L18 ' confidence level.- It is unthinkable that I would say'anything '19 111ke:that, and therefore,.I want everybody to change,their page d 20 '2'3 of tab 7 to what I really said which is, nobody around the. I ' 21 : . Nuclear Regulatory Commission understands the words, confidence -22 level. You will surely agree that that's a valid point of "23 personal privilege, won't you, Mr. Chairman. 24 MR. ZIEGLER: How about nobody. 25 DR. MARK: Are there some people somewhere who do? i LC:) i Heritage Reporting Corporation (202) 628-4888 i____L_L_...__

m,c-7 i

U h.- k 4 r- ,y - 118 -- 1 DR; LEWIS:' Yes. c2 'DR.? MARK:- I' don't.-see this.on.page 27. 3~ DR.-LEWIS: Page 23'. 4L CHAIRMAN KERR: Any other-Irrelevant comments. On-an I- 'S .early.page under tab 7 there is a letter that we wrote;on Julyf 6 15th, in which we commented on NUREG 1150. ~And I refer you'to-7- .the last paragraph, which I'm sure all'of.you remember, but.in 1 I 8: case you don't, you might want to refresh your memory. L 9; Subsequent to'that,.we received a memorandum over Mr. l -10 Stella's signature which was meant to correct any false l 11: . impressions that we might have gotten from the subcommittee'in ' 12 the committee meeting and as indicated in our letter',.and. 11 3 subsequent to that letter, Mr. Halston. decided that.it;might be~ obu

14 well for h'im and or Mr.. Murphy.to come talk to us about'what 15'-

th'ey perceived as a possible. misunderstanding on our.part. And 16; they're here. ] - 17 And'I turn the floor over to one-or both of them. 18 MR. MURPHY: Mr. Chairman, I believe I won't need.all 19 Lthe time that's been allocated to say what I want to.say. 1 20 CHAIRMAN KERR: We'll be glad to take it back if you 21 don't need it. .l 22-MR. MURPHY: Basically, what I'd like to do at the 23 start is to bring the Committee very briefly up to speed on 24 what we're now doing in NUREG-1150 during the interim period. i 1 25 It's been a few months since we've discussed it and we've had l ): Heritage Reporting Corporation l (202) 628-4888 1 l

-em- 'i- , q 1{ 3 i 'l T 'y, d A: .119-1 s ) l, ' jq, Then'I.do want to:get 'nto just~ ) Q 51, lquite a bitiof work going'on. - i i l2_ briefly' some commentis son the ~ letter and ACRS' suggestions ~~3

]

"3 .(Slide) r '4 : I think-the whole thing boils down in many ways to 1 SL the comment I.had'at the top, thati the use-of PRA.in' regulation 'i s 16 -requires that.PRA'. findings and.'concl6sions be able to be ] 7' communicated.to'a broader audience"than those of usL'who did the: y J 8L . study. -And this has been one of'the_ problems with_many PRAs. ] 1 19 The work'that1we have to complete 1150 is' going.to focus on. a 'i L10 'scrutability and' traceability and hopefully the' credibility of Lil the methods and results used. 12 (Slide)~ 13. I_'ll say a few words on the adequacy of~our' 1 (h' 14 supporting data base and then we're also addingJrisk:of-j ~ y 15 external events for;two plants. 'l .l 16 A number of the comments we've received, as-I'll-show. j '17 you in a: minute,.have dealt with our expert et teitation 18

process.

We have gone back and involved decision theo'ists who,i r '19 are expert in the area of eliciting expert opinion. 'We ' re. i 20 redoing that part of the analysis in considerably more detai1 ~ 21 than we did in the draft report. This is a substantial amount 22 of effort and will extend over a several month period. 23 In terms of we're trying to document -- .24 Cl! AIRMAN KERR: Excuse me, Mr. Murphy, there is a i .25 question here. LOL Heritage Reporting Corporation } (202) 628-4888 = =

1 1 g, jl^ yf '120 g' __g T1, .DR. SHEWMON :' Are'you going to:say-something more 2 "aboutlthat:-or have~you said what you're' going to say about that-3: first thing.. ,.MR. MURPHY:. I've1 pretty well said what.I'm. going to-i [' 5- .say about that.

6-DR. SHEWMON:

.Now, _are you doing this by sort:of-contemplating'what'you already.got from the ex'perts'or going ~ ~ 8 back and talking-to the experts again? 39-MR. MURPHY:'.The process, the way we're.doing is.- s 10' number one,-we'reLinvolving many more experts;than we had last a 11 time. 12. DR. SHEWMON: Okay, so.you feel you're getting~new

13-data rather than sortfof asking --

4(l)~ f 14. MR. MURPHY: : Hopefully,.we're getting new-data. If 15: you want,'Izcan' describe the process.at some' length.

.16.

.DR.'SHEWMON: You've answered my question. "17 (Slide). m MR. MURPHY: Okay..In terms of documentation, we,had g 19 a number of computer codes that we're used in the' draft report 20 that were well documented or did not receive complete quality 21. assurance. We're going back and trying to do a better job on ~ 22-both of those. 23 Because some of these processes are still evolving, I 24 won't say they will meet the quality assurance standards that -25

we might want when we're all-done, but they will be h.

m-Heritage Reporting Corporation (202) 628-4888

-r-- - i i , h b' l',li;

• q i

a s i 1 121 i ,m( ). 1 substantially better than what we'had a year ago. 2 We will present our.results in terms of output 3 distributions. Indicated on that will 'lx3 the 5 and 95 percent 4 bounds, the medium, the mean and the most~.. In other words, we d i 5 plan to present all the information so that it's easily 6 understood. 7 DR. SHEWMON: Well, now at least two ways you can j 8 work with experts is to say, here's the spread and in some way 9 get yourself what you feel is a mean or a mode or whatever. l 10 Another way is to do things sequentially and see if you can ] .11 evolve amongst the experts some commonality of opinion. You're 12 going to use generally the same procedures as before where you 13 sort of see what the spread is'and then try to put it up on a 14 line? 15 MR. MURPHY: We:re using somewhat the same approach i i 16 as before. What we're doing is we've identified the major 17 issues affecting the uncertainty. We're calling together the l 18 groups of experts, and first we're trying to state the problem i 19 very clearly and have discussions both with the experts who are j l 20 on the expert panel who will be hold, as well as allowing a i 1 21 broad based opportunity for others to present evidence, for i 22 want of a better word, in this process. 23 So for instance, we would expect that the people from 24 Towski and Associates who were active in the ECORE movement 25 will be presenting information to the panelists as well as the /T 'LJ Heritage Reporting Corporation (202) 628-4888 L L

{ l 4 3 122 ) 1 NRC_ contractors. We actually sent letters out to everyone we 2 .know doing work in the area, both in the U.S..and 3 internationally to provide us synopses and bibliographies that 4 we can point the experts toward. 5 Once we've defined the problems'for the experts, we 6 allow them six weeks to study the problem, roughly six weeks. 7 And then they come back in with their degree of beliefs in various outputs, and'they define while we will suggest the i 9 decomposition of the issue as to how they can get to the 10 outputs, they're allowed to decompose them anyway they want. 11 They will then for each one of them essentially have 12 a distribution that relates to their degree of belief of what 13 the outcome of the issue is. And we will combine these / i j 14 distributions.from the individual experts. I should say after j 1 15 they come in, they'll give their distributions and they'll do l -16 so in a blind manner, and I guess what I mean by that is, blind i 17 may be the wrong word. They.will cast their votes privately. 18 They will not share those votes with the other experts. 19 But then having voted, there will be an interchange 20 between the experts on a technical level rather than one what 21 my distribution was versus what yours is but rather how I see I 22 the problem and what I think the important points are. 23 At the end of that they will again have the 24 opportunity to change their votes, and they will also be asked 25 to provide a justification for what they have. (~h () Heritage Reporting Corporation (202) 628-4888 l i ) i

p 123. a ()' l' ' DR '. = SHEWMON : : Thank you.

2 LMR.-MURPHY:--We then are going to take.that 3

information and. combine it weighing the:experte uniformly.. In 4 other words,.we won't say expert one.is better in this1 field 5 than' expert 2. We're going to weight them uniformly and 6-combine,those' distributions. That final distribution is what 7 'we'lluse in our analysis. 8 As I say,.in terms of the presentation of the. i 9 outputs, we-will identif,y them, and the medium:and the'mean of 10 the distribution, the mode will show because we'll show the-11. whole distribution,.and then the bounds, JL 2 'We are looking now at structural changes in how'we . +. 13. ' document the. report. We have found there's a tremendous amount /N

V-41'4 L of information--in NUREG 1150 draft and we find that an awful i

15~ lot of what people haven't found. It's some of the important 16-points are buried 1500 pages into the report. -And we're trying- .17 - very hard right now to figure out.how to change the layout of i 18' the report so we get the important information up front and .19 convey the meaning in a little bit better way than we.did in '20 the draft. 2 11 (Slide) 22 In terms of the adequacy of our supporting data base, 23 here I need two types of information. The characterization of L 24: plant design and operation, we find that we have new l 25 information on the plant design that indicates that some of the O Heritage Reporting Corporation (202) 628-4888 L

q I. F 124 I m 1,) 1 modeling assumptions we made the first time were incorrect, i 2 In addition to that, we find that in some cases the 3 plant has changed in the two years since we began the,last 4 analysis, and in some cases, these changes relate to changes 5 that'were made.because the utility looked at lessons learned 6 from what was important in the-draft report and has corrected 1 1 7 those problems. l 8 In addition,-in some cases we now have better-data, i 9 the utility has suggested to us that we were using improper 10 ' data. A good example is the loss of off. site power data that 11 we-used in the draft report for Sequoyah. We now have new 12 information from TVA, and we'll be utilizing that information. 13 .And it indicates that the frequency of loss of off site power '~' 14 is significantly lower than what we used. a 15 We also have more phenomenological data.

Again, 16 there's been a lot of work done in the severe accident research 17 area over the last two years, and that information is being fed 18 back into.the analysis, again, information not only from the

-i 19 NRC's contractors, but also from what's been going on in 20 industry. And we've also asked similar information from 21 international sources. 22 MR. WYLIE: What was the extent of the difference in 23 the loss of off site power, the numbers? 24 MR.. MURPHY: The difference in frequency of loss of 25 off site power I believe was a factor of 3. And what it is is ( ) Heritage Reporting Corporation (202) 628-4888 1 l l A

q e

a l ] 125 4 -.. I} 1 that in the draft report, we'used the average numbers for the 1 l $s 2' Southeast reliability district or reliability area, whatever l k 3-it's called. The general average for loss of off site-power i '4 numbers'for the southeastern United States. 5 TVA has shown us that their grid is such and their 6 plan is such with up and coming lines, that's an improper 7 number to use on that plant. l r 8 We're going back in and fixing that. 9 An example of one of the design modifications, our A0 understanding of design and how it factors in, one of the key 11 features in the Sequoyah analysis in the draft report was an 12 assumption that upon loss of component cooling water that you 13 would lose the charging function. We now have some information 14 we're still evaluating that indicates that the charging pumps 15 have been tested for 24 hours without component cooling _ water.. 1 / f 16 Assuming, let's first look at the evaluation. 17 Assuming that when we get done with it, that we agree with 18 that, that's going to substantially change the results of the 19 Sequoyah analysis, because that happened to be one of the 20 dominant accident sequences. And that would have a large 3 I 21 effect on both risk and the malfrequency for Sequoyah. 22 As I said, in terms of phenomenological data, we're 1 23 trying to do a much better job of improving the documentation 24 of what we considered. We've asked both our contractors as j 25 well as the international sources and our contacts in foreign ] l { ~ l w/ Heritage Reporting Corporation l (202) 628-4888

i 126. .n. () 1 governments as-well as the industry'to provide us:with N 2 bibliographies and data. And in terms of the experts as they 3 provide their expert judgment, they're going to be asked to 4 ' document the bases upon which they made it. 5 (Slide) 6 We are adding consideration of external events. The 1 7 slide says, seismic and fire. It'll also include flood for 8 Surry and Peach Bottom. And we plan to do the other NUREG-1150 9 subsequent to the publication of the report. 10 (Slide) { i 11' There's one point here that we haven't decided j l 12 entirely yet, and that is a number of the comments we've i 13 received have said, this is a new-body of work. The external () l ,s '~ 14 events, and it should be published as a draft for comments 15 similar to the way draft 1150 was. We have not made a final j 16 decision on that, but it's possible that this effort external a 17 event analysis of these plants and the risk significance of .18 them propagating that information through the containment of 19 entries will ne published at approximately the same time as ) 1 20 final NUREG 1150, but with a different number and as a draft ] J 21 for comment. Alternatively, it may be in the final NUREG 1150. 22 That determination hasn't been made yet, i 23 I personally am leaning toward accepting the comments ) i 24 we've received, and publishing it as a draft for comment. But i 25 there hasn't been a decision made yet. m i L) l Heritag0 Reporting Corporation ) ] (202) 628-4888

g, 'g.,- ,. 7*, ,. g., .y, y r ,[ y ' %%M 9;- [, % m ',';3 P l @f M. m,..'M m v 3 ' #y, m ., m. p i g-.m Q T y, 7 e o.- g7 r.. h; 4 i '9 c.- .' DR.-:i MARK : ! /Is fire t risk 'an. external eventi?1 1 Qj 1 yq jh} :P 2 L - MR. L. MURPHY : she.way we.usedithe term.in PRA,1we tend.. r6' 3. LtoicalliitJanl external: event. ,It?sjatpoor.:use;ofJtfie Englishi

t m

g. [ 41 l'anguage to call' Lit external'. 'o n q, L. 5 * ' f DR.: MARK: But thatc doesn' t.stop you.: I

S:

MR. MURPHY: 'The'same thing is>true,:in.ternal floods.-

g. 7

'are? regarded:as external. events by the PRA community, and,that;' p w '.8L 'again is.a misnomer. I 9 MR' EBERSOLE: TVA came forth and gave you atibetter. a 3 10 estimate'of their'llabilityof;off site power, didn'tEthey?: p'. 11 MR. MURPHY: Pardon?L 112 MR.'EBERSOLE: Did.TVA give.you this'better; estimate. ' m 9... , o'f - thel'r liability of. of f : site. power? - 13' A.i ,H'" 14-

- MR. MURPHY Yeah,mthey provided the' data for.thet

-{ 15 .plantiover-'a ten or~ twelve year period. ll 16 MR. EBERSOLE: Were there any.. utilities that'came.. O l .j . forward to: give ;you a ' worse picture of their of f cite - failures?' 17 u 18-I!: doubt it. .19 MR. MURPHY That.hasn't happened. s 20 MR'. EBERSOLE: So there's both ends of the. spectrum. j 21 How do you work that?

q 22.

MR. MURPHY: The way we have worked it is as we did- ) i- .23 our initial data analysic, we asked our people to look f>r, the j i <s 24 data analysis that was done for NUREG-1150 was sparse'. It was '25 more than has been done on a lot of the earlier FRAs but not Heritage Reporting Corporation (202) 628-4888 1 F I lI ~~

1 1 i 128 (x_) 1 what'I'd call the. equivalent of a state of the art PRA today. 2 We did ask our analysts to.look for' failure rates 3 that were higher than-they'd appear, you know, to go through-4 the OERs and the maintenance logs of the utility and see if S they.saw some reoccurring patterns that indicated equipment was 6 failing more often'than we anticipated. 7 That search was not as complete as you would do in a 8 full scope PRA today, but we did attempt to look for out liers 9 in that' regard. 10 (Slide) 11 The principal comn.ents received fall in the areas of I i 12 which I've been talking about the question of excluding l I 13 external events in the draft, some questions on how we handled q s-ss 14 operatos: error, significant comments on the way we did our ] 15 uncertainty analysis, and the use of expert opinion. Questions 16 on how we did the phenomenological calculations where we had to 17 extrapolate.beyond the source term code package and many 18 comments to the effect that the documentation was-difficult to 19 understand. 20 (Slide) 21 As I think I've said, we're trying in our rework to 22-address those. The general nature of the comments are that 23 we've received a total of comments from 36 different 24 organizations, with 39 authors and 46 1etters. They add up to 25 about 800 pages as of yesterday. I've been informed but have ~T (G Heritage Reporting Corporation (202) 628-4888 l L

h t

q w mv,

[ i,

129:

._c e *d; 1; 'not yet.seen we have; received a compilationcof comments that' l* ._ ~ c (f ' P2' !was$ gathered-by IEEA from: member countries that I-believe came h" < Kin?this~ morning,'and:from.what I understand is fairly thick. 13. ~ 4 >ButEIchave.not seenLthat. So the 800 pages'doesn't include ? 5 that. 6 'We will-include in our' final report something much-'in 7 the style.of the.old Appendix 11 of Watch 1400, a' report-on the' l E 8' ' comments. We are in the process now of. annotating the-comments. j H 9 and trying to consolidate them into topic area. These will-be-j given tofboth'the staff and our contractors 1to try to get .10 ' 11 ' responses-inLthese general areas. These'11Lbe' combined and 1:2 published either as a summary chapter or.as an~a'ppendix to ~ 13- 'fina1 ;1150. - 3 2([)' 1 14

( Slide) i d

15. Moving on,'I'd like to-just basically discuss the .s 1 . 16. ACRS letter. _I guess.two questions come to mind'from the. 17 comments of-the ACRS'.__ Number one is I think'we had somewhat of. 18 a misunderstanding'on the~1ast paragraph-of the let'ter. And j ( 19 perhaps that' misunderstanding has even gone so far as to lead 1l 20 'us here, because I think our understanding was that you were j 1 21 unhappy with the answer you got from Mr. Stello and that was, j 22 the reason we were here. I think that may be incorrect. .i 23 But I think we do need to know to understand better 4 1 ) 24 what the ACRS suggests in this area. I would also like to i 25 explore if we have the time some ideas as to what the ACRS ) l ( l Heritage Reporting Corporation I (202) 628-4888 'l l r l l _ _A

m 3_ q 7 ,( ~ wv + s.- 4j y ,..y,, y; w } 'l ,4 w

130.

-i j s 1? ' meant..inLi s comments I guess on thefsevere.-accident W 2 -implementationiin terms.of the need to conductJfuture PRA; audit, , l ~ 31 activities.: Exactly-lwhatithe committee' hadfin' mind,1and- /4' iwhether you'would do.a' completely random. audit'"or1whether.you -1 .5; would want'to fill in the. gaps'of:the plants 7thatihave notihad l h g ? .PRAs or how you would.suggest. proceeding inithat-area. .I think-L .6-7-

7 yourlinput could'be very valuable to us if.-we have the time-to; q

,') 8.'-discuss;that'beforetI leave the podium. i1 d 1 '9

CHAIRMAN ' KERR': 1Which letter was that, Joe?- Do'you"

i

.g 10 . h a v e.i t ?'- ] 11 'MR. REED:., July:15th. jl .g 3 12 CHAIRMAN KERR: 'Okay, go ahead. ] g 11 3 ; (Slide)- 4 =4 L14 MR. MURPHY: What I have in my; final' view graphs.is 15 . pretity much a reiteration of what was. in the letter. that came-

- j l16-

.from Stello. The-' family of.PRAs that's available to us now. a el '17 which in' terms of-what the NRC has access.to.is on'the. order of ~ 4 1~8 25 to.30 PRAs. I think the actual. number.that's been done.is L19.- ? approaching,40. In this. family, it' indicates that the risk can .20! vary considerably as we go from plant to plant, and the risks q o 21 are often dominated by interactions that are plant specific, ? '22J and these tend to come from the supporting-systems, things like I23' AC\\DC power, and component cooling water, service water, things g i ^24-of that sort. { v k s25-Thus, the quantitative results in terms of core 10; i Heritage Reporting Corporation (202) 628-4888 j I

gj .c 131-1- damage' frequency;and the health consequenc'es from.one reactor 2 are:not.appliceble to another one, even though.it's of the.same: -3 general. type. But we do' find that the important' accident-j 4 sequences tend to have a generic.' commonality. .In other words,. ) .t ~5 virtually every boiling water reactor.that's been looked at,'it 6 is found that station black out are~significant accident-1 4 7 sequences. 8 In the early days, we found that the long term loss l l 9 of decay heat removal:was important.until we started giving 10 credit for venting and other-operations that have now come 11 through the'new EOPs that are available. 12 .That type of. generic information is available and'is-I 13l useful. t l 14 DR. LEWIS: Could I'ask, I apologize, I missed'.the-p 15 beginning'of your talk ~ kind of. But this whole question of-l ~ l 16 what one can learn from these.PRAs, we've had some conversation g 17 about before. 'And the question is not really whether a' 18' specific accident sequence is duplicated in several different-19, reactors. The question is whether one has sampled accident 20 sequences. 21 For example, if you look at a given type of airplane, 22 you have different accident sequences from the ones you have in 23 another type of airplane, yet there is a certain accident rate 24 because the sampling brings up different things float to:the l l 25 surface with different-things. But you still learn a lot about A) i Heritage Reporting Corporation (202) 628-4888 l I

c FOi, ^132 ,{m r -l 1 the safety of airplanes. ~In'the same sense that.here,cyou J .2 flearn-about the safety of. reactors by, sampling ~adm'ittedly 1 3-different. accident sequences. 1 4~ It seems.to me..to'be far too restrictive'to look for i J. .5

sequences which are duplicated from oneLreactor toLthe'next in.

6 order to.get_information about the population from this

7 sampling process that you're doing.' And it seems to me;that

.8 all'these. things are. written in.such a restrictive way that ,1 19 you're denying yourself reasonable inferences about-the j '10 population. .11 ' MR.' MURPHY: I don't mean it that way,:but~I can see-j 12^ Lwhere I've givenLthat impression. One of.the. questions I asked '13

before you'came back-into.the room, Dr. Lewis, is: I'll mentions

-I ) I leave-14, again,that I would like if we have time today before I 15 -the podium t'o discuss with the committee a little, the '1 16 suggestion the' committee has made about future PRA audits.- q 17

Exactly how the staff, you know, what thefcommittee's q

'18' suggestions would'be for how the staff proceeds in this area,. j 1 19 as to whether.we do totally random audits or whether we try to j 20 fill in gaps of plants that have not had PRAs and what approach ,j 21' you'd suggest in.the going through that kind of thing. ') 22 DR. LEWIS: Well, we'll have to' talk about this some l l 23 more because I also missed your comments about how to improve l 24 expert opinion. But I've read the thing and it hurts my head. l 4 25 So, we'll have to talk about that more. O Heritage Reporting Corporation (202) 628-4888 l i -)

. c. ( 133 l' MR. MURPHY: What I'want to impl'y here is I think-l 2 iwhat you just said is there's a useful body:of:information and- ~3 it's this' generic information on general things that have been l 1 4 found important. They tend to group in-general categories and 5 we suggest that these are applicable to other. plants,'and they. 6 could be used in getting information on the other plants. 7 The information can be used in.two ways. Certainly, 8-this collective PRA information serves as a foundation if 9 you're doing a. future plant specific analysis. They're.not 10-adequate in and of themselves, but you've got a-tremendous leg 11' to build on on the work that has gone before. .12 You still:have to go back and make.sure that there 13 are no features which are plant specific which are going.to 14

compromise.the redundancy and diversity assumptions that-were 4

'15 .made perhaps in the-earlier-PRAs and~which are-required by the 16 . regulation. ) 17. MR. EBERSOLE-I don't know of any diversity 18 requirements in the regulations, redundancy's required. But l 19 find.me where diversity is required? Just line out one 20 statement where that's required. And while you're doing it, 21 think about Davis, Bussey. 22-MR. MURPHY: Point's well made. 23 MR. REED: Jesse, I believe there's diversity ) 24 required in the general design criteria for containment 25 cooling, as an example. It says diversity. ?} Heritage Reporting Corporation (202) 628-4888 _ _ -_ ___D

^i y v 134 [ )- l' MR..EBERSOLE: That may bc an isolated. case, though. '2 .-I'm1 talking about in'a general ~ context. 1 ,, CHAIRMAN KERR:- Jesse, in the, general design criteria. 4: .that have to 'ck) with the' rapid shutdown system, it'does not' l 5 require diversity but it mentions redundancy and diversity as a 6 .way;of. achieving a highilevel of reliability.' { 7 MR. EBERSOLE: It's an open end, but it's not a 1 8-requirement.- _This was in the context of a requirement. ;If 0 9 they're made as statements of options, C.ey ' re not - 10 requirements. 3 -l 'll MR. MURPHY: I think in terms of the way I used the j l 1 ' phrase, it would'have been better stated if I said that as- - ~ 13 'we're looking at using'.the collected PRA information we go.to a 14-plant specific analysis, you don't compromise the redundancy LS 'and diversity assumed in earlier studies'as.you go forward'to u -l 16 'the new ones. i 17 We also:want to evaluate what additional' measures 18 exist at the' plant which may make the plant different than what' l 19 you'd got if you'd just added reading the absolute letter of 20 the regulation. j 1 21 Ard I think our bottom line conclusions in a nutshell' l ) 22 are'that the PRA analyses have identified details which may D 23 have been compromised some of the intended redundancy that were -24 found in earlier studies. An example is the interfacing system 25 LOCAs, or the AC dependencies on the turbine driven feed water ~i d's/ l Heritage Reporting Corporation (202) 628-4888 1 1 L_.____

p 1 1' ,135 ,(,f: 1-pumps that showed up i..n the' auxiliary feed water pump; study i 2 that was1done roughly at the same time'as the Three-Mile Island i 3-

accident..

i A'nd inLthose cases where that kind _of thing s'been' l 4 .i 5-identified,-the-regulatory practices have been changed in terms '6 of'the way it was handled in a standard' review plan or branch 7 technical positions or reg guides or whatever. And'thus we l 8-think PRA'has contributed to the current effectiveness _'of the 9 . regulation, but we'd say that as we continue,.you still have to' ~ 10 continue to search for out liers and therefore we need to do 1 11 some' plaint specific analyses to conform the existing judgment -l 12 that the plants'are safe. I 13 This.is what I tried to say last time. I'm not sure + ): 'A-14 that: message clearly got across. 15 With that, I'm open to questions. That's.the points 16 I! wanted to make to you gentlemen. '17 MR. WARD: I guess one thing I'd be' interested in, l 18 and I don't know if you expect that there's any way for this l 19 sort of general insight to come out of this stuff or not. But' 20 if you find, and I'm thinking of this and the results you get 21 opposite the safety goal requirements outline, and you're 22 finding that these plants -- I mean, you're not alarmed by the 23 results for these particular plants, when you think about what j 24 the Commission wants to do as far as regulation, that is, to 1 .25 set a safety goal. You aren't getting a danger message or Heritage Reporting Corporation ] (202) 628-4888 1 1 }c

p q [" i 136 j g ?! /~< something for-these requirements. L Aj Liu i ~ 2 .But I wonder what'you are learning,"or whether you're. 1 ,3 . going to be able to. learn something.about th'e'importance'_of -- 1 J ~4 -I mean~the body of regulations that'we have? I:mean, in d l -5 general, these plants seem to be coming.somewhere close'to l 6 meeting the safety goal and that could be for a couplof I 7. different reasons.- One is-the results are not really j l 8 objective,. or the definition of the safety goal-wasn't j

l 9

completely objective, and this is all sort of a self-serving 10 analysis. - 11: Another one is, the regulations have caused the 12 plants to have.a.certain level'of~ safety performance that gee 13' whiz'seems to be in conformance with the safety goals. And 1 ' /] 14 another possibility is that it doesn't have an awful lot.to'do 1 15 ~ with the' regulations, but it's got something to'do'with.the 16-inherent characteristics of LWRs'and the traditional design-17 " practices of the power industry or somethingElike.that. 10 ' And that's the reason the plants are coming out with 19 safety performance which is reasonably consistent with the f 20 safety goals. .21 Are you ever going to learn from this nort of work. 22 anything about those sorts of general perspectives? l l 23 MR. MURPHY: Well, in two of what you mentioned, I -24 think the answer is yes to, but I don't think we're ever going 25 to learn how to qualify it. One is, I guess I have an inherent s ($) l Heritage Reporting Corporation (202) 628-4888 l r __m-un.---

4 .137 l ) 1 s.- bias, I've been a regulator over 25' years, and I,would l'ike'to 7-2' think we've had some impact on'the safety of the plants. So. 1 i 3 that my own personal view is that the body of regulations we i '4 have has to some extent at least has caused the accidents that 5 are within the design basis envelope to be low risk 6 contributors, and I think to'some extent at least that comes 7 from the regulatory activities. .] ~ { .) 8 By the same token, I think you have to say a lot of j j 9 that also comes from the fact that the industry itself is ] l 10 interested in safety'and I don't want to imply that the core j 11. melt frequency would be extremely high if it wasn't for the j t 12 regulator because I think there is an emphasis on the part of 13 virtually all'the plant owners and operators that'I've come in 'J 14 contact with they have a real interest in safety, j 2 15 Where the current level safety got to whether it was ] 16 because of the interest of the regulated industry or the 17 interest of the regulator, I don't think we'll ever be to 1 18 separate out who had the most influence. Nor do I think it's j 19 important. I think the LWRs have some very nice i i 20 characteristics. And those obviously factor into it. 21 I think as we go on and on in PRAs space as we do 22 sampling programs, as we look at plants individually, we will 1 23 get a better feeling for what. if anything can cause a plant to 24 be an outlier from the general trend of the industry as a 25 whole. That's not to imply that an outlier necessarily means (~h L-) Heritage Reporting Corporation (202) 628-4888 m~Ax----- - - - - - - - _ _

7g, 3.y m 138 1-Jthat it's' unsafe and has to.be corrected. If I had.to make my . projections on:the final output of 1150, I'would'say,that our 2 3 uncertainty' bands are going to' drop a little bit, at leastLthe. 4

contral' measure, the mean-'or the median, depending.on how you'

.5 look at it, will be somewhat lower than exists today. !^ 6' 'Aiid that information, the distributions are ' givenLin 7-the report; but they're buried in'an_ appendix in there, and-L L8:.most'; people haven't'found'them yet. Those I expect are going -- 9 ' .to' shift down a bit. It's hard to say how much while-we're- .10-still going through the' analysis. 11' -And'I'm being:veryfspeculative as I say,that because' 12 the' analysis hasn't been completed yet'. I suspect our upper 13 bonds may not change too much, but the safety goal is couched I 14 'in terms-of means, and even with the plants we have now in 15' . terms of the' quantitative. safety goals and the quantitative . 16' design objectives, the mean value is in general.I-thinkJat-17-

least an order of magnitude below the safety goal.

I think 1 .18 that that's very we look at five different-reactors, and we 19 came to that conclusion'in all of them, and I think that that's 20 .an important conclusion. 1 21 I'll.put it differently. I think if we.found that 12 2 - three of them were substantially above it, we'd be having a L t L '23 substantially different discussion today than we are. You 24~ can't arbitrarily draw.the conclusion that that means every-25; plant in the country meets the safety. goal, but it ' indicates I" ) Heritage Reporting Corporation (202) 628-4888 L __...n u

C ,. v 1 .a e P 1394 0 L( / 1 that,wo've 1ooked at.five plants.- ~ . e also have a number-of W .,_t -2 other PRAsiat variousl levels of detail and we' looked.Jat what-L l- '3 their'results have been'. y L 4- -And:we can start with Use'information we have now and' L l '5f say we think"we know many of the> features that could.leadito a ~ f-g

6 higher. level of risk, and we know now how to go

back and look.

I 7. at'aniexisting. plant or an. existing analysis, and try tojsee:if, 8- 'any of those features exist, and-try to implement it into=it. 9 'I'm reasonably confident.that.we're safe, and<I'have? l j -10 no. problem with.the Commission's statement; but'by the same p 11 token, I~think thatLthe netural follow onato it to do an. l-12; individual' plant examination is a worthwhile consideration. ~ 13 MR.'EBERSOLE: What were the plants you examined? ' I . O,q 1 .'14 forget at the moment? .15 MR. MURPHY: Pardon? J16 - -MR. EBERSOLE: What were the plants you examined, 17 which ones? 18' MR. MURPHY: Surry and Peach Bottom,-Sequoyah, Grand 19 Gulf, and Zion. 20-MR. EBERSOLE: Let me pick. Peach Bottom. We just~got 21-rid of the TVA. team.this morning. They've been doing a lot:of' 22 - work and have been. shut down for two years. Was your' analysis 23 based'on.the as-expected condition of the plant.when they get ~24-done with all the corrections, or the as-was condition of the 25-plant, and can you identify the statistical difference in L LOL Heritage Reporting Corporation e-(202) 628-4888 o k l-.- .--.--.---L.--.:---_.

i 140 m l l' 1 safety because of all these actions that have been taken? x/ 2 MR.-MURPHY: We've based our analysis en the plant on 3 the drawings that were available to us that we got from the 4 plant. ) 5 MR. EBERSOLE: Then? 6 MR. MURPHY: As of a year ago, year and a half ago, I 7 guess. a 8 MR. EBERSOLE: I don't know whether they'd dredged up 9 all the problems then, or not. 10 MR. MURPH'i: I think it's in areas where we've had 11 problems, I think TVA has told us where we've made assumptions 12 that have been untowa-'d, if I can say that. We have had very i i j 13' good discussions with the utility. We had a team at the plant (,_l 4 N' 14 'for'close to a week, and have had several meetings on the j l 15 . subject. It's something I really appreciate because TVA has I l 16 had its own troubles at this time, and they've taken the time l 17 out to be very cooperative with us. 18 CHAIRMAN KERR: Going back to the question of our 19 letter, I will try at least to provide my interpretation of the 20 sense of what we were trying to say, and I think that's j l 21 probably legitimate if one does it with all the Committee l 22 present. 1 23 But I will read from the first sentence of that last 24 paragraph. "One might conclude, both from the report and from 25 comments made by the Staff, that the NRC regulations are i t" (-)/. Heritage Reporting Corporation (202) 628-4888 1 l m._... _ _ _ _ _.... _ _ _ l

+ 3

f)

"4 y i 141~ Lil inadequateLto determine plant equipment and procedures ~ 4 - 1 - 2i nece'ssary to. protect public health and: safety." Now,jimplied' 13L 'in that is'to. protect public: health and-safety from risk, I (4 L thisik, !at least that -was my : implication. -5 .Now,.ifione goes to page 57:of-'the transcript of,the' ~ 6 subcommittee meeting, and:I realize that-Mr.LLewis has'already 17 ' discovered 1that it completely' misquoted.him so.perhaps it =8' misquoted you as well. -I asked th'e' question','I thought'ther 1 9 report 1was telling me.that one:could not judge the riskiof a -10 . plant without doingua PRA' LThat to me. tells me that- ~ 11- ' regulations do not.. permit one.to'. judge a risk of'a-plant. 12: -And Murphy is. quoted as:saying,."no, it; tells me.that-13 I cannot'do.a_ quantitative estimate of the risk without doing a' g; l'41 PRA."' 15 Now, if'you can't do a quantitative estimate of~ risk, '16 then I don't'know how you. judge the risk of a plant. And it i 17 .was that point that we were.trying to makelin the letter. If 18 one can't say that without a PRA the~ risk of a plant is .i 19 sufficiently low as determined, for example, by regulations, j 20-then it would follow, it seems to me, that the regulations do 21 not tell you whether or not the risk of a plant-is acceptable. 22-And that was the point I think that we were.trying to 23 make. Now, we didn't say that we had understood you. It just i 24 said that at least one could conclude from the discussion that 'i 25 that was what you were saying. () Heritage Reporting Corporation H (202) 628-4.888 I q: i

y 10 is 142 '1 MR. MURPHY:- .I'm a little troubled by'the comment,'.I- .21 guess..To me,,to say you'can't have a quantitative estimate of. .3 the risk' without doing a'PRA isslike saying I'can't know what-4 -the-sum of:.a' column of numbars is if I don't add them up. .5 The process of getting a quantitative estimate of the-6 risk is basically doing.a PRA. Not having a-PRA, can I make 7 .the' finding that's required by-law -.that is no undue risk.to 8 the' health and~ safety of-the public. 9- 'Obviously, we'd be making that based on-our 310 . understanding of what was important from a risk standpoint for 11. . years going-back to the sixties. 12 ^ CHAIRMAN KERR: We did thst before we had the PRA 13 technique available. What one might have hoped would occur, it. 3 I 14 seems to me,-once the PRA technique became available, is that- .l 15 the PRA would indeed show that the regulations had provided an , 16 appropriate level of risk. 17 What I thought we'were being told, in both the 18 Subcommittee and the Full Committee, was that no,.you can't do 19 it. Unless you do a PRA on every plant, there isn't any way 20 you can draw the conclusion that the regulations have permitted l 21 one to achieve an acceptable level of the risk. 22 Now I may have misunderstood -- we may have 23 misunderstood. 24 MR. MURPHY: No, there are a lot of subtleties in -25 . phraseology involved here that can lead us into traps. To get 10 Heritage Reporting Corporation (202) 628-4888 i

i s 143 'r[ N 11 Lan~ absolute quantitative number of risk I-think you have to do V 2-Jan analysis.on that plant. !3 CHAIRMAN KERR: But nobody asked'for'an absolute' ] 4-quantitative.

5'

' MR' MURPHY: WE can look'at the body of PRAs'that' 6

have'been done, and the body of regulations, and those 7'

indicated'that.the plants we've looked at'to date meet the: l ~8' . safety goals. 9: That's evidence'to me that the body'offregulations. .10 that we have has'been effective in terms of limiting the risk. 1 Et .Where, over time, where we have found something;in a PRA'-- or 12: in any other way.~ 13 It doesn't.have to be a PRA -- an operational data, bOJ -14 or.some further analysis of different types. and an example 15: I'll use is the Event-V from WASH 1400 that was not scmething_ 16 we looked at prior to that time. 17 There the regulations were effective -- not the 18 regulation, per se, but the implementation of the regulations 19 in the way the plant was analyzed by the regulatory staff -- 20 'was modified to reflect that new information coming-in. 21 CHAIRMAN KERR; But see, this is one of the problems 22l -- maybe not the right word -- that I had with a memo from Mr. 23 Stello. L 24 It seemed to say that PRA had to be used to correct-4 25 the regulations, because the regulations, for example, having Jo Heritage Reporting Corporation (202) 628-4888 e LL__'-.-..

,_7-_-___._. l> or -j i 144 h!. -i {(J 1 to do with auxiliary feedwater; those having to do'with Event-V -l 2 .had;been inadequate up till that time, and they had.to be i L !3 corrected once.one did a PRA 4 That could lead me to believe -- you know, these y 5 regulations may, on each new plant'you do, you may. find 6 something that will require you to correct the regulations. .7 So I didn't find that memo to be a.very convincing 8 argument. i 9 MR. MURPHY: I think what it is is, the' evolution -- the regulations evolve with' time as we gain.more information. 10 a L 11 It doesn't-have to be'from PRAs. i 12 We're getting a lot of.information from. operational 13 experience. We get information from other analyses. These can -,A. : A 14 lead us to believe that in some areas that the conservatism 15 that was put into the regulations 20 years ago may.not be 16 . required anymore; we may want to relax them. 17 -But in other cases, you can find something and say. 18 that it effectively says there may be some sort of a gap in the 19 regulatory that we want to correct. 20 CHAIRMAN KERR: It wouldn't be difficult to sell me 21 on the idea that indeed every plant maybe should have a PRA 22 because maybe the existing regulations are inadequate. 23 I would not find that criminal or anything, because 24 aftet all, the process is evolving. We did the best we could 25 with regulations. Maybe they're inadequate. O Heritage Reporting Corporation (202) 628-4888 1

i l 145 [) /1 I don't know. But at least I hope you at least have v 2 some idea of what we were trying to get at as a. committee when 3 we wrote that last paragraph. j 4-Now on a related but rather different subject, I find i 5 that you are doing more work on Peachbottom, and it has been 6 subject to a lot of analysis up till now. 7' Peachbottom is now shut down. As far as I know, it i 8 is the first plant that has been shut down by a commission 9 order because of something that had nothing to do with the 10 plant equipment, as far as I can tell. i 11 I am unaware of any predictive-mechanism and existing 12 PRAs that would have predicted that shut down. And yet, the 13 Commission apparently considered this one of the more serious fr $/ 14 r'isk contributors that it has encountered -- at least judged by i 15 what it did. I 16 And this makes me wonder about PRAs in terms, at 17 least, of the Commission's appraisal of risk' contributors. And 18 I'm not trying to be critical of PRAs; they do what they can 19 do. 20 But here's something in which, it seems to me -- in 21 your view, would a PRA have -- 22 MR. MURPHY: Well, I have not been part of that 23 decision process, but I'll offer one observation. There've 24 been studies that have been done many years ago, and I think 25 they're verified in virtually every PRA that has been done to O %.) Heritage Reporting Corporation (202) 628-4888

'n ? 146

[)

1 date that if your human error probability goes up substantially-2 from what was -- 3 CHAIRMAN KERR: No, but I'm trying to take a real 4 plant now which has been studied perhaps more than any other 5 plant, except maybe Surry. 6 It's been studied; it's been restudied; it's been 7 reevaluated; it's been QA'd. And.I'm trying to ask myself, has 8 there been anything about any of those studies that would have l 9 predicted what happened? 10 And the answer I come up with is, I' don't see any, H i 11 but am I missing somt. thing? -12 MR. MURPHY: Well, only in terms of a general sense 13 that I think on any plant that if you believe tnat you have a- \\l 14 significant degradation of operative performance, there's a 15 good chance the risk is going to rise substantially. 16 CHAIRMAN KERR But voe, that hasn't shown up in any 17 Peachbottom study that I've ever seen -- anything that would i 18 say -- and I don't think it's true that you haven't gone to 19 Peachbottom and talked to the operators and talked to people. 1 20 That's been done repeatedly, I think. I l 21 MR. MURPHY: No, what I'm saying is that if you were i 22 to say that the operator performance that was assumed on the l J 23 PRA -- thero was a Brookhaven study, and I can't give you a I l 24 reference to it right now, that was done perhaps three years l l 25 after the reactor safety study was published. l- /^l \\ l Nu Heritage Reporting Corporation (202) 628-4888 l )L_= __

147 () 1 It used that information to say if you increase the 2 ~ human error frequency, the risk goes up. 3 CHAIRMAN KERR I'm familiar with the study,1 Joe, 4 because one of my doctoral students was the one vdio carried it i 5 out. And I'm not talking about that. 6 I'm talking about this plant which was studied and 7. restudied, and yet it seems to me that at least in terms of 8 what action that the Commission took, had perhaps one of the 9 more significant risk producing situations that has been 10 encountered. 11 DR. SIESS. Still that study doesn't -- I mean, it 12 doesn't take much of a research project to say if people make 13 errors, you increase the probability of a plant problem. TMI . (~') k' 14 taught-us that, j l 15 But I didn't see any study that showed me how much I-16 could reduce the risk by feeding all my operators no-doze l 17 pills. I 18 MR. MURPHY: No, I agree. The PRAs that I have seen. I 19 don't differentiate at that level. 20 DR. SIESS: But the Commission does. So what can the 21 PRA tell us about safety goals? 22 MR. MURPHY: I think there's a basic assumption that l 23 goes into the PRA results in terms of the way you assume the j 24 plant is being operated and the operational readiness, if I can l 25 call it that, of the plant. l x l Y-) l I Heritage Reporting Corporation (202) 628-4888 l

148 () 1 To the extent that those assumptions that go into the 2 PRA are wrong, you could have problems. 3 DR. SIESS: Do you know which ones are wrong? I 4 mean, if I knew which ones were wrong I'd be a lot happier. 5 MR. MURPHY: No, I don't know which ones are wrong'. 6 And in fact, I don't know any of them are wrong. But there's a 7 basic assumption that 's made in the PRA in terms of the 8 general level of the performance of the operational crew. 9 DR. SIESS: I_just got somewhere away from the bottom 10 line I know that. 11 CIIAIRMAN KERR: Mr. Lewis? 12 MR. LEWIS: I couldn't sit here while this was going-13 on. I apologize; I was trying to put out a fire at another ~s) 14 agency. 15 But the question of wrong and right is the wrong 16 question for PRA. This comes up all the time, just like safe 17 and unsafe is the wrong question for PRA. 18 In a PRA, you're supposed to take as many things as 19 you know, assign them the probabilities 1as best you know them - l i 20 ' - there's no question of whether wrong or right. j 21 The probabilities may be high; they may be low. The 22 uncertainties may be larger than you think. Very often they're 23 less than you think they are because you're trying to i 24 accommodate idiot views at both ends of the spectrum, and avoid 25 trouble for all sorts of complicated reasons. i l ~% l l {d Heritage Reporting Corporation l (202) 628-4888 1 1 i

W:. .,L 149' l 1L f'/p 1 But'inithe end, the PRA gives you the best possible s-i L 2 estimate of..the probability of something happening. In'that

.).

3. sense,,it does'tell you something'about safety goals. J 4 DR. SIESS: Supposed -- supposed. L 1 L l5: MR. LEWIS:- Pardon? f 6 DR. SIESS: -Supposed.to give.yourthe best possible 1 l 7 Lestimate. l1 MR. LEWIS: Well, if you do it properly it gives you-3 t> sest possible'-- possible is the key word. -10 CHAIRMAN KERR: It gives you the best ' stimate thatJa e i 11 PRAJcangive you. 12' MR. LEWIS: No, it's the best estimate that humans p 13 can give you if it's done properly. 14 CHAIRMAN KERR: Well, if that's true, what explains 15 that fact that three different groups can do a PRA on the same 16 plant and come.up with three answers. i 17 MR. LEWIS: Actually,_that's much healthier than if I

i 18 they came up with'the same answer.

19 CHAIRMAN KERR: Which one of those is the best 12 0 possible estimate? 21 MR. LEWIS: No, each one has a certain uncertainty 22-associated with it. It's like-giving kids three exams; they 23 come up with the same exam. 24 It happened to me last quarter. It tells you 4 25 something about how they took the exam. First of all, the PRA I 'r ( ) IIeritage Reporting Corporation (202) 628-4888 I i ) 4 j

i i i 150 f ) 1 should not give a number; it should give a number and an v 2 ' uncertainty -- a distribution of probabilities, 3 If the distributions of probabilities that people 4 give have no overlap at all -- for example, if I were to k 5 measu,,re your height, and say it's four feet plus or minus a 6 half inch, and he were to measure it and say it's six feet plus .7 or minus a hal'f inch, then by golly one of us has a bad ruler. 8 MR. HARK: No, he did it by PRA. 9 (Laughter) ? 10 MR. LEWIS: Well, you guys are throwing away the 11 hammer. But the point is that an intrinsic part of a PRA is 12 some understanding of the uncertainty associated with the 13 result -- that's an intrinsic part. c'% kl 14 In that sense, it can't be right or wrong. If the i 15 uncertainty associated with the result is such and such and 16 later' experience shows you that you get 35 events in a row 10 17 times sooner than you should, then for God's sake, something 18 was wrong with the input assumptions. 19 But the PRA is just an exercice in logic -- putting l 20 the things together. And as people keep saying, it is really I 21 true, it forces you to do your analysis in a disciplined way. l 22 That's all it does. 23 And the people who argue it are people who much 24 prefer to do their analysis in an undisciplined way. There are 25 some things I prefer to do in an undisciplined way, too, but /^% k-) lieritage Reporting Corporation (202) 628-4888

1 151 {} 1 they're none of your business. l 2 (Laughter) 3 MR. MARK: Did you say the PRA gives you the best 4 humanly possible estimate, or something or another? 5 MR. LEWIS: That's a slight overstatement, but -- 6-CRAIRMAN KERR: Excuse me. I think this is an 7 interesting discussion, but I don's see why we need to keep Mr. 8 Murphy here for the discussion. 9 So may I suggest'that if we have any further 10 questions for him, we get those out of the way and then we 11 carry on this discussion. 12 Mr. Moeller? 13-MR. MOELLER: I had a question. Did the Subcommittee ( '~i x' 14 consider the Stone & Webster comments on NUREG 11507 I 15 MR. MARK: The Subcommittee did not have those 16 comments when it met. i i 17 MR. MOELLER: Oh, okay. Well, you know, listening to 18 the presentation, it sounded like the staff was systematically i 19 moving forward and so forth. { 20 And in reading, yesterday, the Stone & Webster i l 21 comments, I found them -- if you believed everything that was 22 said they could almost be rather devastating. l 1 23 Does the staff -- L 24 CHAIRMAN KERR: What Mr. Murphy is saying is that 25 they have moved beyond those original comments, taking into 1 I (^h l %/ l Heritage Reporting Corporation i (202) 628-4888 l l l

3 152 (} 1 account, I presume, the Stone & Webster comments. You 2 certainly had'those. 3 MR. MURPHY: The Stone & Webster comments are being 4 factored in as we go forward. { S MR. LEWIS: Do they agree that their comments have 1 6 been f actored in properly and they've dropped their concerns? 7 MR. MURPHY: Well, we've received a large number of 8 comments. And the way we handled them was, quite frankly, is .9 we thought we knew where the weaknesses of the report were i 10 better than anyone else did, j 11 So we started the work before the comment period was 12 _over. As we get the comments,'we're going back and looking at j i 13 what we're doing and.seeing what has to be modified. [ (^T \\/ 14 MR. LEWIS: Yes, but Dade asked specifically about 15 the Stone & Webster comments and I wonder whether they know l 16 what your resolution of them is. 17 (Pause) 1 18 MR. MURPHY: Who do you mean by they? i 19 MR. LEWIS: Stone & Webster. 20 MR. MURPHY: Stone & Webster. Stone & Webster, as a 21 corporate entity, we haven't met with, but Ed Warman who tends 22 to present most of their comments is on a -- 23 MR. LEWIS: So he knows what you'vo done with them? 24 MR. MURPHY: He knows, right, I think, what the work 25 going on is. L L Heritage Reporting Corporation i (202) 628-4888 l \\ l _21-_ j

L l~s .j l ,;j 153 'l U (( )

1' MR. MOELLER:. Then.you fully-interacted with him and-

'2, so.forth? : 3 MR. MURPHY:' I'd say he's been informed of what;we're 47 doing.- 5 LMR. MOELLER:. Well,'I thought you just'said that1he' 6 came-to the1 meetings or something. ~ -j 7

' CHAIRMAN KERR:- No, he said he was fully. aware.of

.j 8-what was going on,LI think,;not.that he-came to the meetings. 9 'MR. MURPHY:- 'I don't want to put words in Ed Warman's 10 . mouth. We have'given him - he~1s a member-cf an--ANS committee'~ < 11 - that;has.been formed to review-NUREG 1150, and we've-given 12 them, I guess, atthree to four hour briefing on what we're-p '13 doing and where,we're going. 14-He has that much information. I don't want to imply; -j 15'

that we've sat down with him1for' weeks and gone over.every one 16 of his comments in detail -- that'would.be an unfair 17 characterization.

18 MR. LEWIS: No, nobody said that. But is he' 19 comfortable now? 20' DR. SIESS: Joe, could I check something to 1x3 sure I

21. 'you're not' misquoted?

Did you say that you thought you knew 22 better.than anybody else what the weaknesses were in the w 23 report? 24 MR. MURPHY: No, I think I said -- If I said that, I 7.- .25 said it wrong. What the assumption was was that we had two O Heritage Reporting Corporation (202) 628-4888 1

'a l + 154 i s 7) 1 options: one is to stop all work and wait until we had all.the r- ) 2 comments in; the other is to start work based on what we know l 3 were the weaknesses in the work with the assumption that the 4 comments that would como in from outside did not cause us to go 5 back'and restart,.and rework. 6 You know, there was something so drastic, some fatal 7 & law that would cause us to have to go back to "go" again. 8 I think that has been true. In other words, I don't 9 think the work we've done between February and now -- the 10 comment period just ended October 1st, and we still have one 11 .other organization that has told us they're going to give us-12 extensive comments that has not submitted them yet. That's the Illinois Department of Nuclear Safety, or Division of Nuclear 13 4 14 Safety -- I may have the name of the organization wrong. 15 DR. SIESS: You were right -- you were right the 16 first time. 17-MR. MURPHY: But with that exception, we have all the 18 comments we're anticipating. In terms of the way we're redoing 19 the expert opinion elicitation and trying to gather the 20 information for the changes that we've made in analyzing the 21 accident frequency -- those based largely on our discussions 22 with the utilities involved. 3 23 The decision back then was that we needed to meet 24 with the utilities on the design of their plants and factor any 25 changes in those designs into the analysis, rather than any (~\\ \\.J Heritage Reporting Corporation (202) 628-4888

l 155 [ )i. I decision made at the minute level in terms of what those n-2 changes should be. 3 CHAIRMAN KERRs Excuse me. Do you-have an answer to 4 your question, Mr. Siess? Okay. 5 MR. LEWIS: Could I ask one just quickle, again? I 6 apologize because I was out of the room; you probably covered 7 it. This very elaborate proced: Ire for what is called the 8 elicitation of expert opinion, is that an NRC invention, or did -9 that come from consultants? Where did that particular 10 procedure come from? 11 MR. MURPHY: The way we did it or the way we're going 12 to do it? 13 MR. LEWIS: The way you're going to do it. .p h-14 MR. MURPHY: Yes, the wcy we're going to do it, I'd 15 say the most elaborate use of it the.t I'm aware of was done by j 16 the Electric Power Research Institute on the seismic program. I 17 MR. LEWIS: And this is a duplicate of that? i 18 MR. MURPHY It would be unfair to call it a 19 duplicate. We're using what they did, in many ways, as a 20 model. We're scaling back somewhat from it because the effort 21 that they expended in the seismic program was enormous. 22 We believe we can -- we've involved several of the i 23 same consultants that they used. Lnd we believe we can 24 accomplish something that's very similar in end product quality 25 to what they did with somewhat lowered expenditure of %(J i. Heritace Reporting Corporation (202) 628-4888 l c

..g. 1156 'l resou rc es.:' 2, What:we're doing'is notias elaborate'as what they-3. ' d id., But'we arectrying to follow the general trend; we're. 4 using'many of the same people.- 5 LMR. LEWIS ' I see. .And did'they in turn,get ittfrom. 6: 'these same consultants,.or did they invent it in-house? Par'on? 7 .MR. MURPHY: d 4 1

R.. LEWIS

Did they get this' procedure from these-8 M 9 same consultants that you're sharing, cn: did they invent it?: 10 MR. MURPHY: I'm not that familiar with_all the .11! detail. But,my opinion in talking to some-of the-Effry staff 12.. was.they. relied heavily on the initiation on some of these-13- . consultants. .14 MR. LEWIS: Okey, fine. V 15 I don't want to characterize exactly how they do

16 _

their analysis -- L 17 MR. LEWIS: No,.I' understand that. I can-find out. '18. MR. MURPHY: -- I don't know enough about it to~ p 19 Lreally say. j =i 20 CHAIRMAN KERR: These are decision-theory type 21 people. 22 MR. MURPHY: What? CHAIRMAN KERR: These are de

ion-theory type 24 people.

l 25 MR. MURPHY: Yes. Essentially, there's a group at U-s. Heritage Reporting Corporation (202) 628-4888

a -l.' 157 1 LosfAlamos'that.we've been funding.for some timc.- In fact, I ' :(] ?? 2 there's Jane Booker and Mary Myers, and then consultant-Steve. 3. 'Horer fromLthe. University of Hawaii at Hilo'.. And' Ward. Edwards 4 and --'can't remember;the. gentleman's.first name. I think his H 5 name is~Keeney from USC. - ~ 6 And they.have been involved'in.the procesa'-- at- .j j 7- 'least, Edwards and Keeney,<'Ilbelieve, have been involved in a, j . process ith Effry, and I'm told.that they1are experts inLthe j 8 w 9' fi' eld. ] i 10 'I must admit, I know~not very much'about the experts 11L fin expert elicitation.- l 12 MR. LEWIS: The reasonoI-ask-is that'I do know many L13: of.the' people;who are experts in decision-theory which is;a la E 1 14 non-trivial subject. '15 I would be very interested, just for my own 16 edification, to see a list of these names that-you have with-17-what'their background is in doing this kind of thing. 1 18 MR. MURPHY: I'd be glad to get'that for you. i 19 CHAIRMAN KERR: Other comments or questions?-

{

20 (No response.) 21 Do you have any further questions for us, Mr. Murphy?- 22 MR. MURPHY: N o,- I think we heard the issues. 23 CHAIRMAN KERR Okay. We look forward to the further 24 development of NUREG 1150. I -i i 25 MR. MURPHY: I'm sure I'll be back again on this I'T l %/ l Heritage Reporting Corporation i (202) 628-4888 { l E_---_--___--_---

l - ?! 158-

~

/*% ' [].- 1 subject. l. 2 MR. MARK: IEhave.a phil'sophical' question that I o would like to hear Hal's comment ~on.

l y

3 l 4-CHAIRMAN KERR: We have 15 minutes during.which we 5 .can discuss.philosophicaluquestions. i 6 MR. MARK: It shouldn't take 15 minutes; it might l L 7 take aboutione.and a half minutes. 8 CHAIRMAN KERR: By the way, you're free to listen to l 9 this philosophical discussion if you want to but -- goLahead, i 10 Mr. Mark. 11- 'MR. MARK: PRA is an exercise in logic, you told me. i 12 There are little leakages-in there.. 13-CHAIRMAN KERR: Excuse me. Did you tell.iULm that? [ l 14-MR. lOMU(: He said it right here. i ~ 15' MR. LEWIS: I said something like that. ' 16-CHAIRMAN KERR: Okay.. 1 17 MR. LEWIS: .It's a reasonable misinterpretation of - 18 what I said as compared with an unreasonable misinterpretation. 19 MR. MARK: It was simply an exercise in logic, and 20 you got the truth when you got through with it. There are 21 little links in there. l 22 I could understand if one really knew and had .23 correctly, in each link, the probability of the go, no-go step -l ) - 24~ that we're looking at, and I really knew that curve, and 25 applied it correctly, then I think I would have at the end a p. G Heritage Reporting Corporation (202) 628-4888 i ~ t

h J i L 159 l l(} ) i probability curve for success or failure -- whatever the 2 outcome is to be -- of the whole reactor system. ) 3 MR. LEWI3: With certain caveats, but go on. 4 MR. MARK: One of my caveats then is there are many -5 places in here where I really don't know the probability 6 distribution of the go, no-go feature, and I' invent it. 7 And the place where I do that mostly11s on the human 8 interaction. 9 MR. LEWIS: Oh, you do it everywhere, but yes. 10 MR. MARK: I do it everywhere; there are a few plcces 11 .where I even know the distribution -- about two out of 100, 12 The human one, I don't really know it. I used somebody's text 13 book on -- it's either .1 or .5 or something else, es 14 I don't know distribution at all. I don't even know 15 the number .1. So I put it in; I put in something. And then-I 16 go through the whole process, and I get this exercise in. logic 17 -- the best human answer, I believe you told me. 18 CHAIRMAN KERR: Excuse me just a minute. I see that 19 the young lady is very carefully taking all of this down. Do 20 we -- 21 MR. LEWIS: That's not necessary. 22 MR. MARK: It's certainly not necessary. 23 CHAIRMAN KERR: Is it okay if we end the -- 24 MR. MARK: Absolutely okay. 25 CHAIRMAN KERR: Thank you. v Heritage Reporting Corporation (202) 628-4988

\\' 160 j 1 (Whereupon, at 5:05 p.m., the hearing was ' / 2 concluded.) 3 4 5 6 '7 8 9 10 11 12 q l 13 14 15 '16 1 17 ] 18 1 19 20 h 21 l 22-l 23 24 25 i I Heritage Reporting Corporation (202) 628-4888 j i

I l 1-CERTIFICATE 2 (3 \\_/ 3 This is to cortify that the attached proceedings before the 4 ' United States Nuclear Regulatory Commission in the matter of: 5 Namet 331st ACRS MEETING (AFTERNOON SESSION) l 0 j 7. Docket Number: 8 Place: Washington, D.C. 9 Date: November 5, 1987 10 were held as'herein appears, and that this is the origir.a1 l 1 11 transcript thereof for the file of the United States Nuclear j 1 12 Regulatory Commission taken stenographically by me and, l q 13 thereafter-reduced.to typewriting by me or under the direction 14 of the court reporting company, and that the transcript is a 15 true and accurate record f the foregoi g proceedings. 16 /S/ / 17 .(Signature typed): Donna Cook 18 Official Reporter 19 Heritage Reporting Corporation 20 -21 22 23 24 25 f ()- Heritage Reporting Corporation (202) 628-4888 e _________._____m._.-_

hYQtA- ,p S n .v "\\, INDEPENDENT a %,,1 DESIGN INSPECTION SEQUOYAH UNIT 2 ROBERT A. HERMANN O V O

O y m r t o a p ng o ,e d u en R I e e ni ,i e k ol, o pm u M mo m xF a poH o mo A oC Rm uo C s om mPR r noo o e oRo o.t R Rgn l i o tp he t o am ppCm r n y o e C rum m.ru sg my r l tPu u a o sn na m e P Ptt r ei ir i rl apH o nR nS l po tS i en x o mo n o eHS I R PRC SCI u C oC o A C C G N 1 W DL t O u B R LOR E l T o C N rm O G. H f to/ C l no L D ORA d C NQ S srl p .e~ , ttl w nno o oeo t CVC l W p m s sr L 3f urr g .e r..* Poe D. t l ett tl G pnn' .oo R.n o N e eo poe UCVC C.M o CVC D a gn C i k R L t no k I ndii A W npt O U I I l i l l B ama j.5U W ut R C P S, O R T E CA E R

Jt; TVA IDI'(INTEGRATED DESIGN INSPECTION) 40- - PERFORMED BY TEAM OF NRC INSPECTORS WITH CONTRACTOR ASSISTANCE j j (TEAM OF 14)- i -. INSPECTIONS ADDRESS FIVE DISCIPLINES' l - MECHANICAL SYSTEMS 1 - MECHANICAL COMPONENTS - CIVIL / STRUCTURAL - ELECTRIC POWER i - INSTRUMENTATION AND C0f4TROLS - IllSPECTION EMPHASIS - IMPLEMENTATION OF FSAR COMMITMENTS, - TECHNICAL ADEQUACY OF DESIGN DOCUMENTS - DESIGN INTERFACE CollTROL - DESIGN. CHANGE CONTROL (PRE-0/L) - INSPECTION ACTIVITIES lNSPECTICN PREPARATION - DIRECT' INSPECTION AT i - TVA DESIGN OFFICES - PLANT SITE - IDENTIFICATION OF RESTART ISSUES - INSFiCT1014 REPORT PREPARATION - FOLLOWy 0F INSPECTI0tl REPORT FINDINGS O-

1; y ^ s INSPECTION TOPICS - VALIDITY OF' DESIGN INPUTS'AND ASSUMPTIONS - VALIDITY l0F AND CONFORMANCE TO DESIGN SPECIFICATIONS - VALIDITY OF ANALYSES . PROPER IMPLEMENTATION'0F SYSTEM INTERFACE' REQUIREMENTS PROPER COMPONENT CLASSIFICATION - CONT'ROL OF DESIGN CHANGES O duus i O O O i

l' Q.{ -} '.. ~ j 93 ? j e) l SELECTION CRITER'IA'FORoTHE: SAMPLE SYSTEM j () s l) 1 LTHE SAMPLE SYSTEM;WILLLHAVE.SOME OiUALL dF;THE*FOLLOWING L 5 CHARACTERISTICS:' a - SE:ESSENTIALiTO PLANT SAFETY LDESIGNED BY THE ARCHITECThEN'GINEER-(TVA) t - BE REPRES'ENTATIVE OF SAFETY 1 FEATURES IN'OTHER SYSTEMS -[BE A DESIGN WHICH !NVOLVES' INTERNAL INTERFACES BETWEEN FUNCTIONAL ARFAS:AND EXTERNAL INTERFACES WITH'THE NUCLEAR STEAM SUPPLY' \\' - SYSTEM 1 VENDOR,LCOMPONENT VENDORS,1AND ENGINEERING SERVICE d. ORGANIZATI"ONS d ESSENTIAL; RAW COOLING' WATER SYSTEM HAS BEEN. SELECTED FOR REVIEW. '[ 4 I j 1 r l u l i s L 9 f1 L1

^ .l CIVIL / STRUCTURAL 1 a ITEMS IDENTIFIED IN LETTER DATED OCTOBER 9, 1987 - KEPPLER TO WHITE i CONCERNS IDENTIFIED FOR CALCULATIONS SUPPORTING DESIGN OF SAFETY-RELATED BUILDINGS AND EQUIPMENT FOUNDATIONS EXAMPLES CALCULATION ASSUMPTIONS NOT REPRESENTATIVE OF AS-J BUILT ] AS-BUILT FUNDAMENTAL DESIGN CONSIDERATIONS NOT l EVALUATED n DISCREPANCIES EXISTED BETWEEN ANALYZED AND i INSTALLED SUPPORT CONFIGURATIONS l QUESTIONS ON REBAR PLACEMENT l l l DEVELOPMENT OF VERTICAL RESPONSE SPECTRA FOR STEEL 1 CONTAINMENT VESSEL TVA TO PROVIDE ADDITIONAL INFORMATION - INSPECTION - WEEK OF NOVEMBER 2 - 14 0F 19 ITEMS p)

..[* t ^ MECHANICAL-(SYSTEM) b -EXAMPLES - DESIGNL RESSURE OF ERCW P CALCULATION METHODS FOR: SYSTEM DESIGN PRESSURE i ASSURE-COMPONENTS MEET RATED DESIGN PRESSURE-STAFF DISAGREES WITH-TVA-POSITION ON CODE -INTERPRETATION FOR DESIGN. PRESSURE B 31.1' i RATED DESIGN PRESSURE O' AIRLCOMPRESSORCOOLERSDESIGNFOR1/2:0FSYSTEM- . PRESSURE (50 - 70 PSIG VERSUS 160 PSIG) l i g ' CONCERNS ON NON-SAF 'Y RELATED ERCW PIPING QUESTION ON MILD ENVIRONMENT TEMPERATURE FOR EQUIPMENT -ERCW PUMP HOUSE O

g. ,2 1 0h MECHANICAL-(PIPING COMPONENTS)- TVA' REVIEW 0F VENDOR. RECORDS FOR SEISMIC QUALIFICATIONS ANALYSIS OF PIPE SUPPORTS CONDITIONS ASSUMED IN THE. ANALYSIS DID NOT MATCH THOSE .) -SHOWN.ON DRAWING q q q l 1 .U. t LO 1 l

q P-,; a: [ ,. 's ELECTRICAL 6li$j + j 'LA'CK 0F GROUND FAULT DETECTION ~0N 430V-AUXILIARY POWER L# SYSTEM .i SEPARATION ON MAIN CONTROL BOARDS 1 -SEISMIC QUALIFICATIONS OF. ELECTRICAL EQUIPMENT i MOTOR. STARTING AND RUNNING-VOLTAGE CAPABILITIES VOLTAGE DROP' CALCULATIONS FOR' LOW VOLTAGE CIRCUITS .1 4 .o

c.,N[ s f Uk pre. ps y of . (~ G r Y s USE OF:PRA;IN REGULATION J, A; MURPHY e PROBABILISTIc-RISK. ANALYSIS BRANCH blVISIONOF.RECTORACCIDENTANALYSIS i

O c

l l II l O s w - _-___._-._.---- - -. - - -

p 'i s ,+ 4 --n .Use of PRA iniregulation requires 1PRA findings and conclusions be. communicated to a broad audience. WorkLto complete NUREG-1150 will focus on key O shortcomings of draft analysis

• Scrutobility, traceability, and credibility of methods and results j
• Adequacy of supporting data base
• Risk'from external events o

O I

L.. f ?Q.;. 1 Scrutability, Traceability, and Credibility of Methods and Results j Improved expert judgement elicitation process and documentation Documentation and quality assurance of models used 1 Characterization of output distributions k . Documentation structural changes l l l O 1

i; f

(O o i (- Acecuacy o' Suapor:inc Ja:a 3ase 1 Characterization of plant design and operation ' design and operational modifications ..^o data l. Phenomenologica! data

• • • , improved documentation of data considered incorporation of new data i

s g l L

n v Ts < ""o "n _x~e "1c ve T:s. ~ I I Consicera: ion o' seismic, ' ire ris< i l nv or 'ina \\ REG ' 50 ) u l Surry a eac, ao::om l 'l i I L 0:ler 3 ants suasecuen:y 1 .O I.w__

r o om ,'D ] - Principal comments on draft NUREG-1150' reinforce the need for-l -change in'previously-indicated areas. Scope ofLAnalyses: - Excluded Sources of Risk - Unanalyzed-Equipment / Operator Actions Uncertainty Analysis 1 g - Methods- - Display LUse' of-Expert Opinion 4 Phenomenological Calculations- - Supporting Data Base - Traceability L - Credibility of Methods l Documentation C l L - _-_ 2

m .g g i.., it D

Comments on: Draft NUREG-1150 and. Staff Response m

Affiliation-

Number of Number of Organizations Letters j

e Individuals: -7. 10 Industry Groups 9-11-Licensees-10-14 Government 6. 7 Public interest Groups 3 3 y Academia 1 1-u 36 Organizations 39 Authors O 46 Letters 800'Pages I I 1 Staff Response

l

. Annotation of comments and consolidat on nto top c areas (Staff) i i i Dissemination of topic areas and responses (Staff / contractors) Compilation of responses in NUREG-1150 appendix (. Staff) Summary chapter in NUREG-1150 main report (Staff) o 1 r 1 .j

q_

i;-

o ! [ '. ',:: T r 'N, ' i .t r ) .-{- 1 t 3,., w ;I ^ v;9 .o 10. - NEED BETTER UNDERSTANDINGTOF LAST PARAGRAPH OF.ACRS 7/15' '4= LETTER; l., 1 I Lo - NEED TO EXPLORE ACRS THOUGHTS 0N FUTRE PRA AUDITJACTIVITIES. "h 1 a O b'

m.. s ,1. ? y c /j ; y. 1 p ti g

s y

w tf-c. 2 j 4 = 0' 1 FAMILY F PRAS-AVAILABLELINDICATE RISK CAN VARY CONSIDERABLY l J 'FROM PLANT-TO-PLANT. 01 RISKS OFTEN' DOMINATED BY lNTERACTIONSTTHAT ARE PLANT SPECIFIC-1 AC/DC POWER 4' COMPONENT COOLING < WATER l SERVICELWATER. 11 LTHUS, QUANTITATIVE RESULTS (E.G.,. CORE DAMAGE, FREQUENCY,- I I HEALTH' CONSEQUENCES) FROM ONE REACTOR MAY NOT BE APPLICABLE TO:ANOTHER OF SAME' GENERAL TYPE. g v. 01 HOWEVER,? !MPORTANT ACCIDENT : SEQUENCES HAVE BEEN FOUND TO ~ HAVE GENERIC COMMONALITY. i i 1 I i,. ij :~ L l 4 I k 1 1 0'

PRALANALYSES HAVEsIDENTIFIED. DETAILS WHICH MAY HAVE COMPROMISED INTENDED REDUNDANCY (E.G., INTERFACING SYSTEM

'LOCAS', AC' DEPENDENCIES ON' TURBINE. DRIVEN AFW PUMPS.) AND REGULATORY PRACTICES HAVE BEEN iiODIFIED. 4 10 THUS, PRA HAS. CONTRIBUTED T0' CURRENT EFFECTIVENESS.0F REGULATIONS. . l O HOWEVER, TO SEARCH FOR OUTLIERS,' PLANT-SPECIFIC ANALYSES ARE-NEEDED'TO CONFIRM JUDGMENT THAT OPERATING Pl. ANTS ARE SAFE. a 1 a Lo i-f. i}}