Text

Transcript of ACRS Subcommittee on Systematic Assessment of Operating Experience 871103 Hearing in Washington,Dc Re Role of Aeod.Pp 1-178
	ML20236K162
Person / Time
Issue date:	11/03/1987
From:	; Advisory Committee on Reactor Safeguards
To:	;
References
	ACRS-T-1620, NUDOCS 8711090128
	Download: ML20236K162 (182)
	v • d • e

{{#Wiki_filter:.. e U.NLt ED STATES NUCLEAR REGULATORY COMMISSION ...........................................................=......................= IN THE MATTER OF: DOCKET NO: SYSTEMATIC ASSESSMENT OF OPERATING EXPERIENCE O R G \\'A,_

t i

I i 9-LOCATION: WASHINGTON DC PAGES: 1 - 178 ) DATE: NOVEMBER 3, 1987 .....;............................................................................=.. ( i %ypagom ADRS0% Heritage Reporting Corporation Q(Scial Reporters 1220 L Street. N.W. O wufungton. D.C. 20005 (202) 628-4888 8711090128 871103 PDR A CFiS T-1620 PDR _____zi

/ ) A PUBLIC NOTICE BY THE J-

2, UNITED-STATES NUCLEAR REGULATORY COMMISSION'S

'1 3 P.DVISORY COMMITTEE ON REACTOR SAFEGUARDS i '%J g n. 5 TUESDAY, NOVEMBER 3,.1987- 'l j .6-l 7 The contents of-this stenographic transcript-of 8. .,the' proceedings of.the-United States Nuclear Regulatory j '9 l Commission's-Advisory Committee on Reactor Safeguards (ACRS), 10 as reported herein, is an uncorrected record of the 11-discussions recorded at the meeting held on the above date. 12 No member of the ACRS staff and no' participant at g f() 13 .this meeting accepts any responsibility for errors ~or 14 inaccuracies of statement or data contained in this 15 transcript. I 16 17 i 18 19 1 20 21 22 l- ,g , L.) - 23 24 .U 25 Heritage Reporting Corporation (202) 420 4848 .]

av

UNITED STATES NUCLEAR REGULATORY COMMISSION b~

2' ' ADVISORY SUBCOMMITTEE ON SYSTEMATIC ASSESSMENT OF EXPERIENCE 3 4-1.)_/ In the Matter of: ) ? 4 ) l 5 THE ROLE.OF AEOD- ) ) 6

Tuesday, 7

November 3, 1987 8 Room 1046 1717 H Street 9 Washington, D. C. 20555 10: The above-entitled matter came on for hearing, 11 pursuant to notice, at 1:00 p.m. l 12- 'ACRS-MEMBERS PRESENT: 7^ j;,T- /' 13 DR. HAROLD W. LEWIS Professor of Physics 14 Department of Physics { University of California l 15 Santa Barbara,. California-16 MR. JESSE C. EBERSOLE Retired Head-Nuclear Engineer 17 Division:of Engineering Design i Tennessee Valley Authority' 18 Knoxville, Tennessee I 19 MR. CARLYLE MICHELSON Retired Principal Nuclear Engineer 20 Tennessee Valley Authority Knoxville, Tennessee, and, 21 Retired Director, Office for Analysis and Evaluation of Operational Data j 22 U. S. Nuclear Regulatory Commission I fs(_) Washington, D. C. l 23 / 24 b 25 1 Heritage Reporting Corporation j mm u

] 2 1 ACRS MEMBERS PRESENT (CONTINUED): . h-2 MR. DAVID A. WARD l Research Manager on Special Assignment 3' E. I. du Pont de Nemours & Company Savannah River Laboratory 4 Aiken, South Carolina 5-MR. CHARLES J. WYLIE Retired Chief Engineer 6 l Electrical Division l Duke Power Company 7 Charlotte, North Carolina 8 NRC STAFF PRESENTERS ] 9 MR. ED JORDAN MR. THOMAS NOVAK 10 MS. KATHLEEN BLACK MR. JACK CROOK 11 MR. JACK ROSENTHAL MR. VICTOR BENEROYA 12 MR. MARK WILLIAMS MR. BOB DENNING 14 is 16 17 18 19 20 21 llh 23 llh 24 1s Heritage Reporting Corporation (302) S28-4SSS

[ i Ys y,

4 M-

.c .'l f3I 1 g&y x

:P U

+ m+ 7 s k[q, P R-O C'E'E D.I N G'S- ~~~~~~~~ ~ e gww.., -(1 : 0 0. 'p. m. )} s. q. . g g, o 7 n This.'is a:mee' ting'of'the ACRS Sub : d3;

g...

DR'.4 LEWIS: ... yy, 1o- [d 24; fcommittee'on1 Systematic.. Assessment of Experience. 'I am e Ai; " jf . ilaroldiLewis. ~ I am Chairman ' of thsl Subcommittee' andi the, Vr, d 4 ,..J 16/

others. here are Charlie Wylie, ' Jesse Ebersole,- Dave 1 Ward '

W v w 1~ 17 .'and Carl Michelson. ;I think.that's all we~ expect, isn't:it?f h y ,T 1 8 Rich Major,cwho.-is sitting next to me, is really ,9: runn'ing the shok.

I am: required,to'say that'anyone who 1

N _. ,l '10: speaks should. speak'up clearly and say who he'is:and I will 1 g. j ,11 ...not: read.the" rest!of the1boilerplate.if we're'willing.to? ~ 12' LatipulateL.that'i,t's beentread. q _y. .+ [,f' 1 y 13- 'So our purpose here is really to-try to; understand-L!4i how the NRCoincorporates~ experience.- Not,necessarilyfoperat-I y ~ l 215 1 ing experience,: but Principally operating experience into 1 ' 16 ' tits structure : so thcSt one makes the.'best. use of the thingsL i 17 that do happen. There.:aretlots-of things that happen and 18-in general wiien accidents occur, not only in nuclear plants u-1 19 but elsewhere people always'look back'with sorrow'at the j 1 ~ i 20 fact that they should have been -able.to prevent them had they l r 21' only learned from what went before, and it's easy to do that -f 1 3 I 122 with football games, nuclear accidents and that sort of s - - '23 thing. ( 24 And so what we are trying to do is to get a better 25 understanding of how NRC is doing this hopefully systematically 1 Heritage Reporting Corporation o m u.. ,l l _ 1 1 ---_ - i

o ltyl[ ' ' ', m ? A ((};W o ' 4 :. 4 7[' y 4

s gfi)d 'm:n -

_J3 1 n, ft, 1

1:

.and>'obviously'AEOD. plays'a. central' role there so we'arechappy M,q J2 (youfare (willing to. come: and J.tell' us 'h'ow you keep the NRC : y w n'

3?

! straight:throughlyour. operationL and-'other than that we are [q_M J J4' yours. s 'I notice that'our agenda..on which.I've-just. saved' q 15 ~ t 6 ;. ius".;eight minutes has'a'wholefstring of ten-minute presenta-7 -t' ions'and!if each of them were to last' fifteen minutes,Jwe'd .8. be here until;the middle of the night. So we'll have to b'e 29- . reasonably-tough?about'.either' acceding to.your' wishes and f 10 j scratching things;l that you don? t think ' are important or else: ,o O:: 8 a;. .g_ ' : 10 -try to? keep pretty.-much-on schedule if.that's agreeable. i 1 .1 N i So'We'reLyours. 12. t"

1 13)

'MR. JORDAN: lThank.you very much, Dr. Lewis. I 14" cappreciate the opportunity-to be here. Jij Q; ~ 0:l [=15) 'I havelbrought.down with me some of the management '~ -m il61 i from AEOD.and.I*d:like to+ int'roduce them-first..This is. Tom 17 1 N o v a k o n fm y.. l e f t ). D i v i s i o15: D'irector. Mr. Williams, the'

118,

' Branch Chie,f. ' And' Jack 'Rosedthal, a-Branch' Chie f. Vic. 0 / 3 19-(Beneroya, an' Assistant.n AndrKathleen Black, Jack Crooks, j 20' and - Bdb -Denning. And so,they'11 all participate at various N. t - 21 -points. q s W ' ' l 22' DR. LEWIS: You have us outnumbered, a A

) '

.23' MR. JORDAN: We'll bring in the heavyweights later. 24 I'd like to first just mention something about the 25 reorganization. I think it's sort of in kcoping with the 'I' Heritage Reporting Corporation 1

q. :

~

cr- ~ y t; 5 1 letter-that Vic Stello replied to to the ACRS on follow-on p 3'sl ' i 2 actions from the 327th meeting and that is that the-April I 3 1987 NRC reorganization actually enlarged and strengt hed y y' l/ 4' AEOD and place'd increased emphasis in the Agency on'the re-' 5 view of operating plants both in NRR and AEOD, so I look on t 6 the reorganization as having improved the Agency's ability, .l 7 some streamlining and some increase in resources. i 8 ~ The NRC.has developen the stragetic plan tLat forms ] 9 the basis'for the Five-Year Plan which is of course a part I 10 of the budget process. That strategic plan clearly enunciated. j 11. the increased emphasis on operating experience assessment so 12 the Agency strategy is to continue and to in fact increase l r- /x), 13 the emphasis that we place on operating experience assessment, v 14 I had a recent trip that I think helped focus some 15 of my views on where we should put more emphasis. I attended 16' a conference in. Vienna and met with the French, both the EDF 17 and their Atomic Energy Commission'personel and the emphasis 18 that I want to place in the relatively near future is on -19 linking tighter the feedback process so that when we find a i 20 problem and it's recognized in the Agency that it gets commu-21 nicated and we follow up and assure that appropriate actions l ,w 22 are taken. And by we, I am talking about the whole Agency, !,Y 23 not AEOD. 24 And in the interactions we've had with INPO, we're (} 25 stressing having this -- closure, that finding a problem Heritage Reporting Corporation m m.a.

l 6 isn't near enough. You really have to assure that the right 1 2 . people get notified and that the right people then take .3 ' act' ion and it's completed, done and that problem won't / kJ 4 bother us anymore. '5 So I just wanted to give you a capsule where I'd 6 like to be headed and any advice that the ACRS can give us 7. to help fulfill that would be, of course, appreciated. 8 Do we have the handouts available? That we could 9 pass out?. 10 What we would propose to do is to indeed follow j i 11 your outline that you provided us and I will give the dis-12 cussion description of the organization.and the reason that r's / 13-Lee Spesard is not here who is the Division Director for the l 1 y- ) 14 other Division is he is working on a report that his hot in 'j 15 his hands and most of the presentation is going.to be about 16 the other Division, so I'll cover Lee Spesard's Division and 17 I'll start with the first slide then. j l 18 Go ahead and give me the next one. l 19 Just to identify, AEOD now has two divisions, the 20 Division of Operational Assessment and Division of Safety 21 Programs. The Division of Safety Programs is much the same l 22 as the AEOD before the reorganization. That is, the per-j-) \\J. 23 sonnel in it are largely the same, and I'll explain the minor j t)T 24 changes. 'w 25 There is an added function of the Committee to Heritage Reporting Corporation onw..u.. II

s p .g ] ~~ ~- ~ - m lr ', ;; in 3 zy! y% .j b.. .1 JReview Generic Requirements that is residing-in'the Office' i Q ' yy < itj' 3 -2 LoffAEOD.: 1 i 3-tCould.I have the.next slide please? UJ-} T 4: MR.-WARD:- Let's see. Which office'doesl Tom-headL ~ q 15 -now?.And,which 'does. Lee Spesard. head?.- .] ) - i q-DMR.: JORDAN: Tom heads the'Sdfety Programs and 1

61 i

7i Lee S'pesard heads.the Operational. Reassessment and I'. hope-the q - 81 titles aren't; misleading. The Operational Assessment Divi- .I ~9 .'s' ion that Lee Spesar'd:is the Division Director for has three 101 ' branches. We've had'recent presentation-'related to two of ~ e 1 11 those branches. The' extreme.'right organization, the Incident 12: Response Branch. You got a briefing Octob'er 8th on the FFE b 13- .' lessons.learne'd'and so'.I would ju'st touch.on.a<few things l 14 that are of' current interest and are receiving I think some' ] .] 3' 15 _ intense-Congressional interest at this time'. <1 l 16.. The, Emergency Response ~ Data System is one.. There / y 17~ ~-is'a bill, H.R. 1570', that Congress is looking:at that pro-18 poses to~ require every utility to feed data to the Nuclear 19 . Regulatory Commissioriland:the NRC'would then make that data /, l 20. available to'the states and the NRC has been planning to 21 implement'an Emergency l Response Data System. We've funded a 22 for it. And we have a contract selection in the process ] 23 right now. The nuance that this bill adds is that the system f '24 would be either be continuous or would be aided by the NRC j 25 or possibly have a black box recorder so that data would be Heritage Reporting Corporation (202) 428-4884 \\

g =J y 4, j

~

~ ~~ ]. ~- ~ T-T - ~' ~ '~= s k m e M 8 ]Kiai 9.o l Lyy } l/ '], Q Lavailable'.goinglback to time ~zero!of'the; event.

y j My '

E2 s..The'NRC'.s proposalLand the-one we're funded.for: <y. j i T3 ; would require the l'icensee to turn it.on when he' notifies- ] 7 ,, (p..; > -- y 'j .us of1an' event. So there1's - a little bit of dif fererice ] 4.. - ., 7 .._ n

there.

3 6; 1MR. WARD: It's. kind of' late, isn'.t"it? .t.:. j 7T MR.' _' JORDAN: - ;The NRC.is not goirig to be ablel to 1 ) 8: diagnose' the-event on the front.end but we' want to be. able ' j .t ]

9 t o.' k n o w ' w h e t h e r,. t h e: p l a n t, condition is deteriorating,.whether i,

1 '; 1 .l

protective 1 actions in advance of some release are needed and 10.

7 11 so.the, utility will have-data in house that one can analyze j i i it'in.tihe ' event after the' fact. But'in. terms of in.real' 1 l2 h g3-time, getting a great benefit out o'f that. front end, our: ,L 3 34 emphasis.is on'the back;end. And I admit I.would like to'- -15 -have the front'end, but-when I-- 16. - MR. WARD: .The interest in.the moment is managing- { l the-actual event that exist.. 17 18 .MR. JORDAN: (That's~ correct. j MR., WARD: I understand. 19-i MR. JORDAN: Unless somebody.has'some questions. -l 20 l f .about the FFE and how that's unfolding, the report is going 'l n- .21 . '2 2 final now and has a number of recommendations for future I I 'b activities and incident reponse. j 23 I guess the only other thing I would mention there 24 is that research has substantial effort now in providing 25 Heritage Reporting Corporation <mnua -I mL

g y-7 -- c g J 1 W 7 ; > yj;: .k y):gy w i q l;} w h-3yc. d L 0* 5..1.j r 4

1-i Jstrategiesiforiresponse"to degraded _ core condition. fAnd'this pp s

y .c L3p M,

2-fisjboth' fore the NRC and forz industry, andof. ' course, theflin-l

'cident response' program'will' accommodate thos'e as soon"as: .i "3: ~ ' ~ }.[M ~ 4/ 64 ? /thatjdataJlis..-available. So we're': intimately involved:with1' 5^ .. research -in -- that-activity. 'y'

6

MR.TEBERSOLE: ;Could!you-comment <on wh'atLtriggers-i (

7 !.

and'ident'ifies, quote " emergency"?. 1 ..m. j 8: MR. JORDAN: There'are reporting requirementsfin ~ l ,1 9= 507.2,250731and in the 5047 that identify variousElevels..of: e /10 - ' emergency,Lunusual event,. alert,.siterarea'andLgeneral.. 5 g-q 11-

So there are criteria'that are fairly clear'in. terms-ofiwhat

.j w l12 severity'one'should-call a particular event.,:So,the NRC: ,j R 1 s 13. Operations Cent'er'gets1 called on events that are..not emer-- { = 2 .. a 14

.genciesLplus emergencies-and we have the opportunity for I

a - 15: second site'upon being called to identify or<ask questions ^ q.j H c .E 16 :

that would: lead.to subsequent notifications of a under-~

V .17 - classified event. i ) j

18-DR. LEWIS: - There are two' things I don't entirely e

'.19 understand'or I just want to be clear. One is is there any 1 + 20 - ambiguity aboutTwho is managing the? plant during' the course i 1 21. of.an event? l g 22 MR. JORDAN: Absolutely none. , = - p 23 DR. LEWIS: Okay. 24 MR. JORDAN: The utility is. 25 MR. LEWIS: Okay. But you will be monitoring what Heritage Reporting Corporation oon um

p, a-u g y y ;, r + ? q; + m,, - ~ 10 - W

g, X

x- + j.yl' ^:) fF i ,theyg do--

j p/

h

2 ":

. MR. JORDAN: ,Yes,' sir. .g ,?3'

DR., LEWIS

.In real' time.. s 4

MR.. ' JORDAN

Inlreal time.. 5-DR.LLEWIS:' And communicating your feelings'7about, t 6-what they do to them in real time? p .7 : ,MR.. JORDAN: Our role'is'really to understandJwhat -8 lis. going onein the plant to be able to' advise the utility 'and support them if we can. 9 1 '10; .DR. LEWIS: No, I. understand that. ,;g jg. MR. JORDAN: And to make sure that'the protective. J -12 action. recommendations are made to thelstate.and locals. And v .n 13 ', if..the utility.is making recommendations,-that.we support

o

j 14 their. recommendations.

) -1 ,,.DR. LEWIS: Theonly reason I ask.is that I hav'e j i s -16 a certain amount of cynicism built in to me and some-years 17, 'ago a person now1 prominent in NRC was asked'this question. 18' He said, "No, no, we wouldn't interfere with the utilities' ~= ) g9 operation management o'f an'~ accident unless, of course, we 20 thought they were doing something wrong."- And that intro-21 duces enough ambiguit'y'into what would really happen to make ' 22 ' one a bit nervous. But that isn't your view? o MR. JORDAN: Right. And we have communicated our 23 fl '24 role I think so that the NRC staff understands. We have l V 25 - the periodic exercises and I believe that the only way the Heritage Reporting Corporation <mnua I' ._-.._--_1

.2 m.y.4., _.. f ~4 -y- ,E. f 1 ~s fi, 9;;;, 4 R 'llE g

): '

L--j, g' a, s -} Yv j = t m /k, (10 fNR kcould/inTfactjdirectly-involve.itself would.beflater on l

b;fy J! [2 '

ythey!would have a siteiteam that'are able:to see the. data, j n Etalk;tib the people and[come up with aft'otally like view.- 3: .j 4' - Butlfrom' Washington ~,.I don't:believe we.'could-q ,.j i $ DR. LEWIS: 'Okay. :I'm comfortable with that. 3 l'* >J '6. ' . The second questi'on. I just want to: be clear about-- -l H ~ .is in view.of~the reorganization, does,this put AEOD in the p: (7 .: \\ ,$8 / driver's; seat in terms of.NRC's response to an incident;.that" -j. 1 ' '.' i 9 .is,3are you in'a support role or are you.really managing the' 1 .\\ 102 ' NRC1 response? .). 'MR.. JORDAN: 'We are doing, and I hatelto'give you' g g :. 12 a: fuzzy' answer. We.have a managing role in terms of-develop-J

13
ingland maintaining the Agency's response capabilities.

A_.e - 1y j;; f l 3 4.. DR. LEWIS:L I.see. , j 15; MR., JORDAN:.When.there is a. response though, the j ti 16 l

team isJconstitu'ted from'the entire' Agency' including-the j

lf ,,1 ~ 17 ' Chairman or one of the Commissioners and the-office directors 18- 'and senior engineers, you-know, specialists, electrical con-i 19; tainment, et cetierai.thaticonstitute the various teams. t D R'. LEWIS: So.your job is more of a staff job, 20- 'to make sure.th$ resource is available--- 23 ' 22' MR. JORDAN: Exactly. Yes. l e l DR. LEWIS: Okay, very good.

23 MR. JORDAN:

We're not running the response. We 24 25 are providing the backbone and the resources and the facilities Heritage Reporting Corporation mu

y 3 -. =- ,. a.c pN l , Sif. j / 's

12; '

7 i q+ _., g~ c;m-q 3 ~ x . ~. '_,-{. _. g7 ! <;(.

li DR.jLEWIS:

I understand, d] ~ ~ l ,.y, 'Edi let'me hypothesize.something' q i .2 l MR..EBERSOLE: t s 3'. that-hastjust,been mentioned here. IWhat roleiwoul'diy~ou plap;- .(,, .w ".L ! + )" c x - 74 ' Llet's.'say=ifeI have'an old boiler with rising, pressure,=.be - l \\

n 5

' cause L I couldn ' t 2 get rid of the exhaust steam from- = gj ,__3 a 6 'MR.' JORDAN: We 've lh'ad an exercise like' that. ~ ' - ] ~ 1 X 7 .MR. EBERSOLE: Yes. That's a dirty question to ifq y 81 .ask.you, but who' decides when you'll; perform.the dirty.act'of 4 9 releasing.some pressure With a little radioactivity in-it? i MR. JORDAN: Right. We have a team constituted' ti 10 n 'd' 11 -called,the Reactor Safety'. Team. There's another' team called '12 the< Protective Measures Team..So the' Reactor Safety: Team j 13 'has?on'its" membership a containment specia'listJa'nd_aytherm. ~ 14 dynamitist and electrical and so on, and so those people, m j is . based on the-continuing increase'in pressure, based'on'not-1 q 16 having sources if you 'need-removal, would postulate that con-- ] f >l7 -tainment is going to fail in' blank hours. realistically, you' ] 1 ,\\ 18 know, not just theidesign pressure is going to be exceeded, j .j 19 - but containment, it.would: fail likely at 2.7 or whatever. -l 1 i 20 And they wou'ld identify that there,may be.a need to vent in 1 advanceofthdt; And tnen the decisions would be made by '21 l 22 the Executive Team. This would:be the Chairman with the } 23 office'direc. tors supporting him. And the decision would be 24 with regards to the recommendations not with regards to the ~ 23 . action. Heritage Reporting Corporation (202) 62s 4see

+,m-Q }y ?

, Q 4

[',J -' ^ w , as .v, + qi;. Y-

13. 7 g

N <g

~ .f g' i g; 4 n oj g x1 y g. n X. 'L, il ' MR... EBERSOLE : : LAll.right. So they would recommendL 1 A m '?f f / ' i, j+ i ] h1 ~4 .i. 12 to;the--: (.{ c o X 3? -MR.fJORDAN: ! Theclicensee wouldibe'the decision < L u. ?' X4 'makerFinfthisicase.. The,NRC.would be'~ recommending'withcre-i .i : ,t -i D,

spect(to'the concern for offsite radiologic.

. q ,m ,l q. '6' S'o*the, decision'is?always the li-MR.iEBERSOLE: ,ni l 7. - ' censeeJinithis case? X i8 MR.: JORDAN: ~Yes. Let me make one more. point [about ' ]!

9

- the Incident: Response' Branch. That does have within :it ::the - l q 10: , operations; of ficers who are theD twenty-four-hour duty station -] ?ll~ iintheTNRbito.receiveevents' calls. So that: set'of'engineerE i 12: Dare within'thisiorganization. They provide 'much of the ;sup-- ( 13.

portJto. keep 1the. operation'Csnter functioning. [And: they ethen e

1 3 i 14l xareltheipeopleLwho first; receive' notification of any reactor l l Jj " 15-L Tevent, <a'nd then their responsibility is-to recommunicate Lit' la 1 -? !!6 .and' so ' AEOD is an office that's involved in: those front end I 1 - 17 L ! communications and'we'11. sort of trail that through our dis-1 ~18- -cussion in terms of how we're involved. 4 19 - I'd like to jumpfin to the Technical Training Cen-20: ter. ~ Ken Raglin is the Branch Chief and as.I think most of 21 you-:know, that,facilityJis located inkChsttanooga. It's a 22 'little-jewel of a facility.,.I. don't know how many of you hyC q p 23 have been to that facility, but it is very nice. It's a 1 ) ' 24 ' professionally well-run organization with licensed operators 1 l' 25 and training personnel with previous training experience in j Heritage Reporting Corporation von uwee

gw 99 -o r.. s. ~7-y - W

s

= - - - ..o.- 3 - : (; 4 q g )@E E ( j d{ @;m,;,

W N y'6 S cp l

o 1 1 a'7 t f 14;

S i:
My ', ;

b j+ s, 4 Ib, r.A 2 j!.. $1ridustry who[tha n;the NRC's personnel. in -technical areas. c, l b 1 , P ' 2J ?I./.wotildj urge 7 you"t'o f takb.;an-opportunit'y to go; visit.'. Wefhavei y t. Ii g l . /; ej :j ' . r... Vi ?:3, [tNo J Ni'mula t[o r s ( there [f unc tioning '. A GE simulator and a' West-y ^ ? u ,f v i-N. m,. i .n v. 'inghousel simulator. We areJplanning'on usin'geth'elengineering Vj ( 14i t s .) , e, M 15 - analyzer [throughteitheriBrook'Havenfor Idaholasya=part;of-di.{ lj l the. s,. ~,.. <, training program and-the. people have accumul ted'a. number-1;

m. 6.

J 11 '9 H/[y l

"g d

of; training.,:aidsLthat are reallyfterrific. So(thah_.is a; 7. .. 3 j,

g,

~ 'y J ~ 82 resource?for;the4 Agency'that';Ifwould like to urge you.}'tfo:look-M] L;. ] jy; ' bi N n 7 A F . 9.- a t.'. 7 3:, O, C";j 6 .? 1

( 7 L

to You.'did have;afbriefing March 12th,Iubelf.ove. I: 4 3 v'. y V 11 was;not,here:for th'at. That'was be' fore the< reorganization.. j ~ f 'l ,3f

12L And~sothe5only.otherthingI'would'sayabout!theTraininkp[

a as ) N$f, {g. 3 A 13. Center is that thereil's'a. larger role'in.the' reorganization - d &6 4 1 -14 for theftechnical training activity. Because the: Agency'is. 1 15 .convertingLto' operations,~thereiis no more newhilcensing: j .sa

161 essentia'llk,anditisreviewing.operatingplants.

There'is j s 4: ' ~' 17 trainin'g and qualification that needs.to be done within the ~ t i 18 Agency.and..the AEOD Office is providing stimulusand then 19 We'11 provide the' basic course work for that'traiping and a t 20 qualification program. And that does extend:also to non-21 reactor training. So with the NMSS Office, we are developing J 4 N-22' nonreactor training courses to support them. So there's a { v 23 very large effort there. There is an increase in staff of 1 24 two for this coming year and we're very proud of that opera-25 tion. , e i Heritage Reporting Corporation I i (202) 628 4494 i

fNN l$,

Ry.

0 Y ~ q^ A*

a r

^ T35; hk q m p](V r npg7), Q c } g g_ g g f' f3 i " ~ '^ sq d' Al% r 7 ~n p y Ma-n- ma i n,i'i i x ct[h 4;,. 2 n l.W l 7.4)b NO W Ed, formerly;youlhad access;to;the' TVA ' A

1T MR.' WARD:

C gl . 'l training 1 simul'a' tors ; up. near Sequoia. q: l t '4

j Sq2. 7

e q=:v. }. on. ? 4 ';q G , %,- a;) },., J h[.f i g ",) (( q [f lMR. c JORDAN} That'sl correct'.- kJ Q. J qp f@ g ','E41 1< c[ e, MR '. i WA'RD : 'But'.now there.arefactuallyytwo3simula-1 g ,y, u. M;w, p; No y 4. - 54 .,.itors-- H s e, x V p .e 4 4 a .n ...>H C7 :tg. ga _g; iMRT'JDRDAN: We have two1 simulators.- Se'e, ! the 9

p

- y

6 L

/,, ' ; V ()_, i y .r '".~,7 l problem;thatiwe had was that withLthei ncreased, training o 'l q+v '81 Jand-qualification.at: commercial' power reactors, the.ir-simu-s g$ ..I;>,q ilgtor timeL.wasibooked'up. 3 w

An'd.so the best we couldido.would:

n W .jjf ; 9 ~ ' l-, In, , SM g J. ei ( [.)

, [ S, lo :

.'be occasi,onally'toiget;some back shifts to getimidnightito;

r.!

&p .1 fb h Q,'

11

,eightfor whatever. time and in many cases"'e>were unable to w 4 (y ;%-..,. <2 x. 92, % getPpimulators,atiall and then had to go.to ano'thercsimul'ator ~ [ 3[;s f.IN $htoshaveailittle'more;timeonit. .So our7 instructors were' W. 'I4 Aconstantly having-toLchange from one' simulator'-to another.. , W 2 R- ~ m m 4 Cho'with'cancelledplants,.thereweresomesimulatorsthat .15 ' s '.= ;., F7'< -( ?wereLavailable-reasonably.on the market so we.got lease-X" L 16 g ( 1,. Ek. mi l7 1 purchase. agreements onithe two simulators we have'that are -L g fv.-. : 18 iequivalent in price' to our lease-only ',costh that we -were pre-4 ^ a,

1' 19' "viously'being charged and we.have as much time as there is in i

joi ~ a year hyailable on the simulators, so we.are able to' offer ] (t -n ' O [ 3, 'a. l - 21 -a great deal.back tor.the! Agency.<> i h} h y,, c .e e, o 22 MR. WARD: Well, back when you were using the TVA, j g L'L/: / l 04 ^ 1

23 I took one of the weekend night'-shift classes and it was

' 24.- very helpful to me. I-thought it was an excellent approach. 2Si MR. 2BERSOLE: To me, a simulator comes in several os i m Heritage Reporting Corporation Md '/ (202) MS 4848 i ay n M,, ..,.?V !..... ~ 1

&x xld a ;@$$ j!($NfM 0'c'N '1 ? ? nj' M ,f I' ~ v v .w M y,g 7 7 \\. yp, e m, ,, Q.

i16, W

w,n, A ~ .n a-gM" * >, m 4 9,<

p.

u

< v.;f q. 6

,0,39> .i-W:i l~ f ,,.y;' '. forms.'Gaeof)themcan~7just;pkayon.aMape'backwhat'.'sbeen , yk : ", ( l ' ., y y~ x s y 7,7 } .g f. t i Eput?into it. "Anotherldoes.f'fer,]:, simulation'. I V.s --21 e g 4g r. c g' -MR.EJORDANO.., h. is[ gel s'a'simtilation that is-jy [: f f[/. 'g.g -3'- -) t ) y < P,( ,5 i ] 9 g 3 d, 13 MR.EBERSOLE: t e a'l 's i m u l a t i o n ? ' I . i. 4 1

4

e. r

~Q

C

't if ,, 4 ~ ,:p s 4 / ~ Ye s L. siil.i : }6, ' t is a' complete > mockup--- / y3 g MR;DJORDAN: 'd d, .{ IK. l,"J s t $, yn S6i i for instance i ? the Eb1'ac/c;b'ox-s . y <,, 1 ., = p '( / T% ... i/it. ,4 I a' " : 7n,. M iMR.cEBERSO,LEr WSo it will develop'information.as? , a\\ m 3- ,.s; m.v. + Il 7 jus't; plan i back fdom; previous

e..: d-g j

hf[ Y 'A; T M R.' JORDAN ~ I t. 'doesn ' t..go into a severe ( accid'ent'.. o M:h{m ' ' fl01 'MR2EMICHELSON: 'It;doesn't huve'rsal' fancy thermo D 11 p 'g y- { 3 ? f hydrauliefm'odels of L the : latest var'iety neebssarily, but' it's Lit. o M% ho y;.. ...got pretty ]gocid-'- l121 v v. 1 (130 MR. EBERSOLE. But!it i real ' simu'lator'? ( l 3 t W ' 14 L MR. JORDAN: Yes,= sir. ,J u 3 1 31 ' ys. ' 'hl7 MR. WYLIE:- 'Could I~ask Y question, Ed? In this' l 3 g ,e i 16 > training) these areas.'that you've':ido.stified, is this all s u .n r < 37;.O encompassing for the wholef RC or-y 1 W .t -'g ir .4 f .I " 9, )y j a 'l8 } MR.' JORDAN: Ppig is tephnical training for the l 3 +{k )I e J 19 F whole NRC; ( ,: 5 2% *{* MR. WYLIE:lFor"the whole.~NRC. 1 s lI /\\ 'f' 1 p-MR. JORDAN: M ps,,siry '7 s L_ ' 21 J t. -t y i i c w, 'E MR.JWYLIE: Soly'sb've got the whole ball of wax. v - e. e .}', i - -h 22 g'... {My, 1 p, hy.- h , MR. JORDAN:"~ Technical training. I "'Inu MR..WYLIE: 7echnical trhining. The whole ball'of s ,' 35/": wax, t ( / %l Heritage R)eporting Corporation ~ I > A v q, m# f <=> u.4 .. '}f3..]k) ?g_ f' ,1 l_

% $ilg % N < g, g > 4 a ~~ ~~ T 4 a J, o. m-e Ge , V +u. 1

w,

17' m;.

gyg m p%g3^ qw 8 ] + L .l l MR M, JORDAN:, Yes, sir. LJ 3 #_. E ,,. ~ 1 g ,.~ 2 i MR. EBERSOLE: LA3 case:in point. Would?it' simulate 4 2 j.

.t

.!3r alBWRiatlasEwith failure to initiate operator. action?

t v.,

p 1/{ 4'g V MR. L JORDAN : ~Only into'the event. Not to. core' d ~

]

, f: 3; " a[ ~

damage.1 -It;doesn'.t go'to core damage at.all.- Butithe' front j

y 5 l~. -y

3. -

e 4 mf, 4 e M would'beisimulated. So:you wouldlhaveLthe-I ~ s f. ,9;f " ?7s stransientsoffiscale and- ' 1 land.that's wh'ere the' engineering analyzer in' con-i + g; y ~ + .n-

}

,a f - .-: 9-

junctionswith.the simulatorTis.a good training' tool.

The c ,. ~ m y .s i j W l10j ' engineering analyzerswill[then--it's'not fancyLwith.all of-

'.n

m t y y.k i

7 the nice'sinterco' nections and controls, but you: can simulate--

> b _

11 : n 5

q m

4, [12" 1 MR. ' EBERSOLE : 'Would'it simulate operatoriresponse ,.p q 4 i.to curb-such:an't. event?. Iti's a good exercise, by the way.:- K Q' L13 '41 -One of the-best; 1 1 t D. h! 155 'MR. JORDAN: 'If you don't go too far into the !16 event,(yes. t -17 LOkay. The next branch is the Diagnostic Evaluation 18' and Incident Investigation Branch. And.Stu Reuben is the j .]

19. -

. Acting, Branch Chief at this time. ^

ji

'20 The programs there'are some new and old.

21 '

MR.~MICHELSON: Excuse me. Could I ask.you--on

there. 8 MR. JORDAN: Yes, sir, i E MR. MICHELSON: Thank you. 1 10 MR. JORDAN: .The Diagnostic Evaluation Program is f Il a new' program..The Agency found that there are cases and l 12 plant reviews in which--and this has been at a managment .,m. ( ). 13-level--which we're not really able to put our finger on what 14 the plant's problem is'or whether indeed there is a real 15 problem or perceived problem. And we've applied our normal 1 16 inspection program tools. This normally would come about 17 through semi-annual meetings with office managers and Vic 18 Stello to go over plant performance. This particular tool 19 is one that would be applied, that would give a relatively 20 . independent look which is an advantage that AEOD has in not 21 involving the regional people in which this plant is located 4T 22 and. selecting'from the Agency appropriate technical skills V ~23 to fill out a team that wouldn't have any previous involve- '24 ment in that particular plant. 25 MR. MICHELSON: You are referring now to the Heritage Reporting Corporation (202) 428 4844 c_ 2_. .= tj. ,i' y> Li ~19" i p, c

;.w, '

a-y q .H et. flL Diagnostic Evaluation Program. 3 , f /N h. 'h2 12 JMRi) JORDAN:~~Yes.. \\ ;0 ;jV -MR. MICHELSON:' Land there are. slides later. Are '3-4 Eyoutgoing'tolgo into.more detail'later? W f5( ~ MR. JORDAN:: I'11 go.into more detail. In fact,- i 6[

why" don'tLwe- '

j s 1 - 7,. MR..MICHELSON: I just want'to'make sure'if now' l i 8-was the-time'to!ask or later. j us, H f '9i MR. JORDAN:..~Let's go to Slide 5. And.I. pretty-

t l

/ much discussed the first three bullets and.the last b'llet u 11. isLthat,this' program supplements theEinformation'that the 12 - Systemat'ic3 Assessment.of Licensee Performance and the per- ,.,g 'e. A. ] s t] 13 fformance' indicator programs and other inspection data have t 1 '14 already provided. 15- 'MR. MICHELSON: Well, I'm not sure that you told ~ 16, ' me much about what this program' entails though. How does 'i for instance, to performance. indicators? '17 this relate, 4 1. 18 MR.~ JORDAN: Performance indicators are--'I have j i 19 to say this--merely a tool. They are data..This is an. 20-inspection--I should say an evaluation on site by twelve .i 21-tried-and-true engineers with a manager team leader in 1' L 22-- .accordance with s[ program that.we'.ve recently developed. l O.- l 23 They are on site for about two weeks. / 24 MR. MICHELS.ON: How are these teams made up though? y' 25 Is this all from AEOD or around the Agency or what's the p Heritage Reporting Corporation m us.a ._L_-_2___.__-._ s 4 .g. <g L /20 'L ?. %)ff g s 9 y e .,v ) i m,, (;ig g,s s. .1 ., Z (. 1: composition?i r.x v '(g.h .l2 I MR.. JORDAN: 'The teams are madecup,from(AEOD and, ~ ,.T.. 3- 'around the Agency. , _ ;? w o M 4L MR.TMICHELSON:- But I thought you'said:you1.didn'tL d'sturb the regional offices.~in,doing this? 15 i g ,,O LMR.-JORDAN:: No. JWhat'I'said was we'.did~,n'ot use' s o

f. '

S ) ~.,.the! regional; office in which this particular reactor!isilo - ~7l 8; zcated. I UE 9' ..MR.:MICHELSON: ,So.yousget'them fromlthe'other< .10 l- , regions'.1 .; n y e ( "' I 11: .MR.? JORDAN: So we get.them from the..other regions. ]

c. -

c 12' We have.:recently completed'such an effort at the Dresden I

e (,.j .13 facility:and so we used resident-inspectors from two other-i ( -14' facilities.~ u 15~ MR. : MICHELSON:' Is there a manual chapter for,this I ti ' program? 171 MR.' JORDAN: Yes, there is. .c 18 MR..MICHELSON: .I'd like to hear more about it and j, n 19' I don't think today.is the day. But I read some.of the 20. reports'for these, evaluations and Igfound them rather in-21 teresting - but I di'dn' t know the basis or the composition, you f ~ 1

22' know, how the program got; started.or an'yt ing else.

23 DR. LEWIS: Today is going to be an over-- ' 24. MR. MICHELSON: ' We 're, into the middle of it and j 25 I'm surprised though that a program like this has been Heritage Reporting Corporation m u. 1 .1 ')f m 21 7' 4,, 4 ' ) r y E;. m 1-developed', lis in 'operationJ and the ACRS never heard of' it, m L2 ,at least I-h'ad ne'ver heard:of..'it. 1 y.. ,3 MR. JORDAN: We are going to brief you.on thet T?O*. ~ i '4 LDresden findings las'soon as their report is published. 4 5; MR. MICHELS'ON:. Oh, yeah, I'd like to hear about j 6 --

the programithough first'that brings'these findings about 7

before:I judge-the quality of the findings. -8 'DR. LEWIS: ~We'11 have'to.dec'ide where to go from

t >

9 here but today is'to get a general view. '10 MR. MICHELSON: What's the' mechanism though by

u 11 which significant programs list their--they somehow escape

=12

our attention.

Where.did we break down in not' knowing'about .,6

(fj

.13 it. At.least I didn't know about it. I asked around a 3 14 ,little bit. I couldn.'t find anybody else who'had.ever-heard m 15 of.it. And.yet it's in operation because ILstarted getting 16 the reports. i l'7 MR. JORDAN: It was another of'those things.that 18? were in the strategic plan for the Agency and.AEOD was 19 assigned the responsibility.for develeping it. 20 .MR.-MICHELSON: Generally in something like this 21 we see a manual chapter f'irst and then we ask, if we wish, l 22' to hear about~it. Otherwise then we start seeing the reports, V. [. 23 But I never even saw tNe manual chapter on it. I

(

24 DR. LEWIS: Well, there must be some other things ,25 we don't know abott it, but I can't-- 1 i Heritage Reporting Corporation ~ <m.a. l Co.. y y. T: 3c , g.. 4' . 2 '2 ' ~ s

q

[, d i icp-f 4 -[. .I~ .- MR. JORDAN:. I hope.that we can'brief you.suffi " 1 pfT s-kli - 0 2- ?ciently'.that;there-won't be many surprises out:o.f AEOD-after- 'W r' .i L3 Lthis.. {[. ..MR.LMICHELSON: .We'.11 find out.- But you mayihave 'e 4-s a 1 1 15.-- ?some;along theLway. '6 1MR',' tJORDAN:'[We'are.not sitting-still. L.. m. y h 7f SoEthe-obj'ectLis to go'to the. site and' collect. xm ..8? Information.using-the.bes't initial information.you could 9 'havel-in' terms ofLth'e. perceived problem.at a plant; t a l g-10' MR. MICHELSON: 'But these_are not inc' dent acti-l ~ i ..>4 L' ill-vated--' ,[ 12~ MRi' JORDAN: These are'not incident activated. .p c r 13' .They..are performance activated or.perceivedLperformance j ~14-fachivated;or questionable-- s r; i 15 MR. MICHELSON: .So you~are in the business of Q 16.- examining performance now, not just incident-- ~17 MR. JORDAN: This is exactly correct. So the 1 .v 18~ role of'AEOD includes plant performance through developing I L, 19 performance indicators and through doing these diagnostic

i.

20. evaluations. 21 MR. MICHELSON: How do you visualize your role in c 22 incident investigation in. situations where you have earlier 23 gone in and'done the diagnostic evaluation and maybe you came ~24 out that everything'was fine'so you had some predetermined-r 25 notions and then this same plant experiences a serious Heritage Reporting Corporation can.n. 2_x_____-__---_.-____---__-_-.___- a L,~ 7 l W$ $q.'. q c.g u -

o. -

3 o' e-23-i e t \\l, ( 4 q ) s. n >.h' _ ? ' { A

l.,

t 3 -g - i2, 'MR.~ JORDAN: I'm glad you asked.. No,tsir. LThe i s a 19-9 . Incident Investigation Program is.one that AEOD has a steward- ^ ,3 '6" M, "41 Jehih over"bilt we, don't run it as1such, fIf you recall, the ~ l .F -5' EDOJidentifies'the' team'directorLfor aniincident-investiga- [6 ? ' tion. The't'eam is~ constituted from.among the offices. :Once. [7-again'an. air.ofiindependence.with respect to' previous' involve-j o ment; "H 8!, 1 ?9 , MR. - MICIIELSON : Always'AEOD isione of1.the: members, l 'sn't it?'. 'i I i 10; d cic.y 'jg .MR'... JORDAN :. Yes. 1 A t.12 - 'MR. WARD:.And you' furnish' sort of the secretaries-- a p s. Il3 - MR. JORDAN: We furnish the support. We're not. ,t o e <t' ( -?', ', j 14' ' steering the team. We ' re not managing the ' team. A senior ~ oe =,.15

SES manager is selected who may be from AEOD or one of the p

.J 16.-

regions.

.i MR..MICHELSON: Let's take a-lesser incident that i 17: x 18-you look into, and again, of course, you are looking into- ' ncidents and situations where maybe you have predetermined j i 19-20 how~well the utility.was doing. Presumably the case would ~ 1 .21.. .e . be where you thought eyerything was fine and then you start 322 seeing' incidents. This doesn't--this isn't supposed to p :, (23 cloud yourithinking in reexamining how well you think the 24 utility is doing because you do have a little bit of a con-25 flict there I think. Heritage Reporting Corporation (202) 62s-4ses q. M'L____.L_._____-_---.-.__ y

r.

~ =, ~ ' ~ ~ ~ ~ ~ ~ ~~ - -- - R ,o . 3 i: e = u '. 2 4 9 4 .7.. I -, ] _; y i .f.

f., ',. p >,

it 'DR. LEWIS: It can't_be made as clean asthe'.s; D u (: 1 M' L2l Emaking 16. 4 [

i.

MR; JORDAN: It's a step;. shot in.tdme.h'owever'. 3-E g?() We'tak' eda' sample. 'We provide:the. finding. ' And;.then the. y 4' ,, ~. [ -x ' . offices, the region'al office and-NRR'are. responsible for 1 'S', I 6 . followfup;of,.anything that comes out of~it.- 1 , f. -7., ' DR.; LEWIS: -Butithis-is.:onelof.the classicLdeep e -7 - e

I 8

iqtiestions that 'have - to do with -the Agency. _ You-are investi-' L 9 'gatingjthefthing'.that it1 regulates. And'_ clearly we'll'be 9 s +; 10 ' talking'more'about that. t- .J ' ll ' MR. JORDAN: Yes. 12? MR. MICHELSON: I'justLwanted*to put it'into con-1 13 . text'of understanding what this was... .y 14L .MR..EBERSOLE: Ed, in doing this sort of. thing, i y^ 15- 'there are'two fields in which you'can have performance One 6 9: 0 16 .isithe: actual manipulation or, operation of a given plant, E 17 -where'you say the-problem-l's in the operational aspect, and 18 is the other in the design deficiencies or too severe duties l 19' required of operators? 'Do you discriminate? Do you put 20' these and identify: either -one selectively? .21 MR. JORDAN: Our goal'is'to try to get to a root 4 l

j. :

h 22 cause for good or bad performance however it occurs in that 23 .particulariplant. L24 MR. EBERSOLE: So you don't hesitate to call it a 25- -bad design? Heritage Reporting Corporation non um l ~ m___ ____....m m_.._ I 25 l [ l h l 1 MR. JORDAN: And if the Agency's programs are on-2 tributing to a problem at a plant, then we're directed and l 3 powered to speak to those as well. So that's the air of (^y 4 independence in terms of-- l 5 MR. EBERSOLE: Well, classically there is a reluc-6 tance to call a design a bad design, and innumerable patches l l 7 are put on it and operational efforts are put together to try 1 8 .t.o fight it out. But in the long run, I think we have a 9 tendency to let a bad design ride too long. I hope it's one 10 of our products. 11 MR. JORDAN: I wouldn't argue. 12 I think I've said about as much about the diagnostic ,s(,) 13 as I need to. 14 The Incident Investigation Program you've been 15 previously briefed on and I won't go into any further details 16 That's still alive. We did just have a training session down 17 in Charlottesville this past two weeks. In retrospect I 18 should have invited an ACRS member for the last day and I'll 19 just briefly mention something about the last day. 20 This is a training program for team leaders and 21 team members. And so we constituted five teams or five peo- ~N 22 ple each and two of these were materials teams and three of t.J ' 23 them were reactor teams. They were given a four-day instruc- /' 24 tion course. We gave them the outline or the background of ) %j 25 what we were supposed to be doing and then they had a one-Heritage Roporting Corporation (102) 426 4448 i 26 I week workshop. The workshop was to simulate their response / 2 to an actual event, to interview players of this particular 3 problem,.such as there were people playing plant manager m_. k~ 4-and operation superintendent and operators and so on. And 1 5' .to put togetherla story, write a brief report, and'to assem-6 ble then a presentation for the Commission as their gradua-7. tion eFercise. And so last Friday, Jim Sezick-and.myself 8 and Dick Cunningham constituted a mock commission and heard' 9 their presentations. Each had forty minutes and our role .10 was to be tough on them. To press them in areas where'they 11 Edidn't'seem to have good information. To interrupt them. . l 12 To cause them to have to respond as though they were giving ,m 4) 13 a real presentation. And I must say that it was a very 14-satisfying experience. The teams had come up with very good 15 analyses of the information that had been provided. And provided excellent presentations. And so we have a cadre a 17 of about now forty-five people, who have gone through this 18 kind of a training force from which to pick any subsequent 19 investigative team. And we'll be putting on retraining ) 20 courses as well as continued training. 21 So that was a real success I'd like to advertise, 22 and in thinking about it, it would be a good opportunity to ,f~g V 23 ask one of the ACRS members next year to come down and serve (} 24 as an examiner. 25 MR. MICHELSON: Could I ask a question, since I Heritogo Reporting Corporation m aam 27 7 i 1 guess you're not going to inform IIT any longer in this dis- \\^ 2 cussion. 3 I've asked on a couple of occasions about what I \\ 4- 'does an AIT mean as opposed to an IIT. And I get varying 1 5 answers..Sometimes I get the answer, "Well, an AIT is just 6 a-smaller IIT." Other times I get more than the manual 4 7 chapter definition where an AIT is an augmented inspection j 8-as opposed to an incident investigation. i -9~ Could you tell me what your concept is of an AIT? I 10 What is it' supposed to be? And further explain to me why l l l 11 Surrey, for instance, was an AIT and not an IIT. -i 12 }1R.' JORDAN: Okay. First of all, the AIT is a 1 Gj,) 13 way.for the Agency to put resources together from the -14 various offices to provide expertise to support a particular q 15 issue. And it's normally a response to lesser significant 16-e' vents. That is, there was not a. declared energency, it- ] f '17 wasn't a declared site area emergency. And there were no 18 radioactive releases and from first precepts, things worked 19 reasonably well. There weren't compound failures. And so 20 the reverse of that is an IIT. 21 Now, there are clearly-- 22 MR. MICHELSON: The reverse of that is an IIT. \\s 23 Are you saying that AIT is-- ( 24 MR. JORDAN: The AIT is managed directly by the 25 region. And the region gets assistance from the other Heritage Reporting Corporation mvnum m 4 ' h t1 N, 'q 4 t fs W ?: Le 28. u . j.' r . office's':to5put this; team together. p; I'd (2 T$1e IITEis managed fromJheadquarters, through'AEOD,

3' riot" managed by7them.

A Freudian slip.- It's managed by the q

Oi i

4l .,EDO wholhasfappointedfa team-leader and.. supported'by,AEOD. 5: so1that's really the big difference,.:.is the level of inde-- ,d L ' '61 -pendence that! ' occurs. I 17- .. MR. EBERSOLE: What he's'.wonderingJis.what's the q I8 basis-'for,the. decision to form an AIT instead of:an'IIT? J9J ' Isn ' t tha't closer to what you are-4 J10 ; MR. MICHELSON: Well, it's'only'part-ofLit.. 1 11 '. ,. MR. ; EBERSOLE : ,Oh, excuse me. A 12 MR.-MICHELSON: I really haven't got'to the: full' 'If '13' - point yet. [ -l E14' 'MR. JORDAN: The basis for the' decision is proce- 'l ?' ,{ 15 dures have been developed for both and then the senior j ..-16 managers, the regional' director and the office directors' l ] + ' 17 who agree or' concur in each case. Once it's an AIT to cause. 18 a consideration to be made and that's really my responsibi- - 19 lity now as to whether an IIT ought to be considered. The 20' decision'is then'Vic Stello's if there'is a recommendation 21 for an IIT. And so the office directors have the respon-L- 22 sibility in each(case to decide based on the criteria and E t

3

-23 based on the particular event which of the tools the Agency 24 ought to apply. -25 MR. MICHELSON: What does AIT stand for? Heritage Reporting Corporation l- <mna.. 7. p. 14 ,m 3 ~ 2 91 l' s ( ', I MR. - ' JORDAN :. Youla'lready stated it.. Augmented a. l ?y - L2h (Inspection' Team. ], N L3; MR.~MICHELSON:.Well, I read the manual chapter' 1 ?): i N.- .4L and'it seems to read that it's just'a regional inspection y a. q ' H [5 ' which is required-augmented assistance. -AndiI'look'at the i 6. -reports and'they read like region'al inspection reports,'not' di -7 like-IIT-type reports. ;They,are. enforcement oriented reports as-opposebto-moottcauseLkindof. oriented.reportslikewe- ~ 8. 9) have seen from'the.IIT,;so my concern is that they seemLto .j. 10 L .befapproaching-AITs'from.a.. regional enforcement viewpoint. j ~ i 11L .and.not from"a true, incident. investigation-viewpoint. '12 So correct me if I'm wrong. Or is the. manual' .13-chapter. going to be; revised or is that your.' intention? That 14 'an AIT be just'an augmented.. inspection? 4 . ~ i 15 MR. JORDAN: The AIT manual chapter..is one that 1 16: NRR is responsible for maintaining. And.the goal i's~that it~ f 17 - 'be a regional. tool, a regional. directed tool with help'from 18 headquarters-as appropriate, and the orientation in terms -l l r 19 of it being'an inspection--an inspection is really supposed

l 20 to be fact finding.

And I have been pleas'ed with most of .i 21 "the~AITs I've' read,in terms of the facts. That they have o <v":

22 really searched out and gotten.what appears to be the root

) 23 cause issues.,When they find noncompliance, they identify -24 it. And-do take enforcement action. And that is a dif-25 ference in the IIT. Heritage Reporting Corporation (292) 628 4400

i 30 I

MR. MICHELSON: Yes, but investigating an incident \\n) c 2, is not necessarily focusing on whether or not regulations 3 -were met but rather focusing on what happened, what were the ? 4 ' root lcauses of these happenings, and what kind of corrective 5 actions are needed. 6 MR. JORDAN: I'm in an awkward position. I'm de-7 fending a program that is not mine. 8 MR. MICHELSON: No. But I have a little difficulty i I 9-in that we seem to be taking rather significant events 1 i 10 and you were going to comment on Surrey, I believe. We are 11 .taking significant events and still looking at them from the 12-inspector's eyes because of the AIT program which is focus- }/. 13 ing differently and you read the reports. Maybe you are 14 happy with them. Some of them I'm not overly happy with as 15 being incident reports, but rather good inspection reports, 16 fine from that viewpoint. And yet serious events are being 17 handled by an inspection process, rather than an incident 18 investigation process. i l 19 MR. JORDAN : The Surrey pipe break was one that 20 we clearly considered sending an IIT on and the personnel 21 injuries and death were certainly regrettable. The decision 22 at the time on Surrey was that this was not a severe chal-j gs ( -) 1 i 23 lenge to the plant in terms of its ability to continue to ['J) 24 be cooled and safely shut down. It was a balance of plant I x 1 25 event in which personnel were injured and so the thing that l Heritage Reporting Corporation am m_m. J s 31 i l 'needed to be understood was why did the_ pipe fail and was it \\/ 2 being inspected, and so it was a conscious decision at the 3 time that this was not one that we would send an IIT on. '_) 4 I think-one can second look at that and say that 5 was a very close call. l 6 MR. MICHELSON: Well, any time that safety related 7 equipment is potentially challenged by things like fire 8 protection eqitipment setting off and by things like door 9 locks throughout the plant locking out the operator, that's I 10 sort of--it didn't take_very much to see that a good investi-H l 11 gation of the event was needed and now if AIT will give us 12 a good event investigation, that's great. .But it doesn't ] I /~N _ () 13 seem in the manual chapter to be oriented that way and the \\ 14 reports that come out don't seem to be oriented that way as I 15 I read them. They certainly don't seem to be looking into i 16 such questions--the fundamental arrangement of the fire l l 17 protection and why it set off so expensively and so forth 18 as a result of a pipe break in the therma building. 19 DR. LEWIS: I think that in the end they'll have 20 to spend a lot of time talking about the IITs and the AITs 21 and how the Agency is configured to look at operating events f-22 and I wonder if your description of what the offices do may ( )g 23 not be the right time to go that deeply into it. (} 24 It's also clear to me that we're not going to be 25 able to adhere to our agendas. I'm going to turn this over Heritage Reporting Corporation < m) u.. 3 S: 4: 32 l "5 r i U b ~ to you to manage the topics'and the time. And the only thing 1 4 - ~/ I will do as my authority is to give'us a fifteen-minute f, 2 3 break at three.o' clock. And other than that, you are respon-1 /~s t 2: 's/ 4 sible for covering the subjects with your people. 5 MR. JORDAN: Good. All right, sir. 6 Then I would say to finish off the Diagnostic 7 . Evaluation and Incident Investigation Branch review--it's a 8 very small' group. There are a total of only ten people in 9 that entire branch including branch chief. They are the core, 10 the center, of resources throughout the Agency. And so they 11 .are maintaining capability. When they are not.themselves .12 participating in the diagnostic or an investigation or trr.in-[ _ ;.m 13 ing, they then are farmed back to provide support bc an AIT '(( J' l !l 14 4 or'some other inspe ion-type effort in the Agency, Sc tnat l those fields are maintained, 15 l 16 So we have a constantly going matrix of that 17 particular set of people. I 18 And'then I would like to transfer over to slide 3 19 and talk about Tom Novak's division. And that's where we 20 would dwell the rest of the presentation. And I'll just give 21 a brief overview and then allow the questions that you pro-22 vided to explain how we function. 23 DR. LEWIS: Could I ask you one question? It 24-suddenly occurs to me that you've been describing the func-as tions of AEOD as it's now--since the reorganization. It's l l' Heritage Reporting Corporation l non u.4u. O-_-__-______-_ 33 I quite:different from what they were before. x r'S. ' k / 2 There are a number of bills on the Hill that would. I ^' 3 . transfer the functions of AEOD to some new conjectual board. k'~ 4 Do the people who wrote those bills, have they been in con-1 5 tact to find out what the functions of AEOD are? 6 'MR. JORDAN: No. 7 DR. LEWIS: That they are transferred? h i 8 MR.' JORDAN: They have not. -9 DR. LEWIS: I see. ;That's interesting, isn't it? 10 MR. JORDAN: it's interesting but not'that sur-11 ,prising, right? Yes, weve followed'that with interest and .t 2 I've been down to some of the' hearings.

. /"3 t

(_/- 13 l MR. EBERSOLE: Much of this is fairly recent-- i 14 jl MR. JORDAN: Since April y i ' [ (5 ll DR. LEWIS: Since April, but so are the bills. 16 MR. EBERf0LE: Yes, but the thinking that probably 3 U (7 went into their drafting was-- 18 DR. LEWIS: I think we know what happened. 19 MR. JORDAN: The Division of Safety Programs, as 20 I mentioned earlier, is pretty much the staff that was in the 21 organization of AEOD before the reorganization. The Trends 22 and Patterns Analysis Branch, Mark Williams is the Branch .es 'U 23 Chief. That is a systematic review of data that are collected '() 24 in the number of data basis and the communication of infor-25 mation. And they'll go into their functions in much greater Heritage Reporting Corporation (102) 626 ded w_____.__________...__ m \\, < a), \\ 34 ii. I Ldetail. \\ '" 2 The Operations Analysis Branch is by event screen-3 ing reviewing data, performing case studies, engineering

/^%

\\d 4 evaluations and. technical reviews, plus the Performance 5 Indicator Program. And that is an additional feature. It ~ dwells in the Reactor Operations Analysis Branch because.dn 6 7 my view if you are going to look at plant performance, you l 8 really have to look at it on a fine. level and review the data 9 .and do. analyses of what it means with regard to that parti-10 Vq cular plant. This program is one that we've reported to 6 i 11 ACRS on previously and it continues to evolve, and we'll have a.little more discussion on it. i 13 MR. WA RD : Who is the branch chief there? i f 14 MR. JORDAN: Jack Rosenthal. 15 The Nonreactor Assessment Staff, Kathleen Black i 16 is the leader of that group, and that is the focus AEOD has 17 in other.than reactors. For instance, the hospital mis-18 ' administrations is one of the studies that we point to 19 proudly. And in human factors type reviews are also products 20 that Kathleen has been responsible for. And her people, j f 21 MR. MICHELSON: That's human factors including I f 22 application to reactors? ( fs (_) 23 MR. JORDAN: Yes, it does. 24 MR. MICHELSON: The nonreactor organization is [} 25 reactor criented as well? l s i Heritage Reporting Corporation l cm uun.. f i l i o Y g' .b i :.f : t M/ b: 3,5 ~

q.,,

m \\ w. ' J. + 4

MR.) JORDAN:

It also has human factor: reactor .1 1 (. ..( >2 oriented,.that's correct.

-1;
)

i o l .. f$ . 3. I've mentioned the-'CRGR already, so I. won't talk e. 0 4-any.further about.that. (,x I 4- ' Si I think.it's:probably time for me to turn to. Tom-Novakian'dito ask him to begin runningidown.the response to. )1 .i. ~ ', '6 .. o 17 the questions, and.that.would put us;only. ten m'inutes'behind l '8-. j IJbelieve. t -9 MR. NOVAK: Let me have Slide'10. :And let'me t'ake .]:! ~ 10 ' asfe'w minutes'togexplain what I'.ve tried to do to be respon-4 11 sive'to-your questions. _If you'll glance ahead.at the next coupleiof '12 ( 13 slide's, Slides 11, 12, 13 and so forth, you'll see bull'ets . [ ,14 with some. statement and then some numbers..For example, 2 =i '15 discussion ~ topic: 3 ' and then 8. 16 ' 'What I did is I went back through-the memo'and-17 I tried to. capture the question that was being asked. For

18. -

~ example, 3 had to do with how do we focus on which issues-- 19 itlhas'to-do with how can we be sure, for example, that we. 20 . are.looking at the right issue. 21 .So I've rearranged your. questions, but I think 22 ' 'they'll flow and also I would like to--then when you look at-the record, you'll see that the question that I'm trying 23 24 . to answer, and that will help you. 25 Now, before we start to answer your questions, let Heritage Reporting Corporation <mun 36 l l I me just spend some time on this slide. Slide 10, 7. 't 2 'I think it's important that you understand where q 3 we start from in the Division of Safety Programs in terms 4 of' reviewing oper'ating experience. And I've listed a few 1 1 5 things here.. And let me just. start with the' top. 6 Each day there is a conference call between members I 7 of the Reactor Operations and Assessment Branch, Jack Rosen-8 thal takes part in it with section leaders and individuals. 9 There is also an individual from the Incident Response 10 Branch. He is the fellow that.is really taking down the 11 notes. Those people are the operations officers. They sum t 12 up what happened. What came in to the Incident Response () 13 Center last night. So they are going to tell the staff 14 across the board. Now, that conference call then has we 15 as listeners, and participants from the Division of Safety i 16 Programs, the Incident Response Branch people telling us 17 what happened. NRR has a number of pecple, Carl Berlinger, 18 for example, and other people who work in NRR are there, as 19 well as members from the EB0 staff. 20 Now, that's a telephone conversation which-kind of 21 goes over the highlights of what was reported under 5072. 22 What did the utilities call in and tell us happened to their gy ~V 23 plant. And I think that's very good for us, because that's (l 24 where we get a lot of oral communications with regard to %) 25 what happened and what we can recall, which may be similar or Heritage Reporting Corporation 1 non su a.. , mr-o ,.b5 I .'.t f t M7,.. r t..b 3 < ( 3 7.. 1 ip ' U-t 'l' .actually'aalso it; helps you to go.-back when necessary to get-t.... M./TM*' < 2' moreiclarification of the event. Sometimes'it comes in' U i ~ -] ~ ieces and>when you try td put it all.together.itidoesn't' j ~ L3 p f.! i 4

4 quite fit, so-there's a sense of, okay,'if we.can ask~these

~ n is? .tiwo:orfthree~or four. question's, we!11 have:a'better under - a. .F .g -sta,nding of:the event. 61 .7 And'I think we have,a.real' input into-that. Jackf -{ ~ 8 'also'workedLbefore coming to AEOD, he1 worked for I&E and he ,m ,l + 9-pretity; much ran that show an'd has a good ' first-hand knowledge a 1 - 10' 'it. 'In_-fact', we'Were-justJtalking today about some' events n 111l t.hatiwe are trying to1 follow 'nd go back to NRR with. a o q l12 - .It'.s-interesting.that,when you change' jobs,:you. 1 1 ~ 13'--,Lare:the same person,-but you cc.n look.at the problems'dif-' 14. > ferently, and I think that's' healthy, and I.think that's' h 15' ' working to-our advantage. 16 I:know coming.from NRR after many years and now 17 being..in-AEOD, you do look at a-problem a little differently. '. You~ don'tL feel that you have perhaps. the same pressure as 19 you would in.NRR, and I think our comments are intended to 20 .;be helpful and really looking at the problems as we see them. l21 We do that as part of our job and I think the feedback'that .j i l 22 we;get from the Incident Response Branch is very effective 23 in this working relationship, and this is one example where,. !O 24 I think AEOD strengthened?because we have that direct feed-25 back from the day-to-day operations. Heritage Reporting Corporation (202) 428-4400 -c .o N :1 i, ' 1:,' '[ ' \\ .M 38 %, :/ - + f IL All-right..Now, thosel are the' things-that1 happen-a 2' each!and'every ' day.4 Youlhear about.them and you: talk'about' 3' .them andLyou. pick up on whether or'~not.there'is:an'act'ivity C-thEtiwe would;look.at.. It' may occur that through that dis-4.- 2 - 5 . cuss' ion ~'we.were asked todo a'searchLor'look up.and see'if 6 'there isfany' data. So we'are responsive-toithat. Ifathe' V.7.

question-1comes.up, well,-did they have an event'like.that

- 8' lbefore?.-Weican'say, we'11 find outlfor you; And.so that, p 9 'feedbackLis done-just that way. We'can,go.back and I?ll' talk.

10' y
about cour. data base,: but that'sL the kind of thing. thatI' 11:
think'.: has been; helpful'in tieing us; closer'in with.theLopera-12 l','g

.tionslof!the plant on a day-t'o-day'. basis. l'3 Now,-Licensing Event Reports. _5073. This is -i e . 14 'nothing more than the-classical LER reporting. But I think' E 15 it's important to understand.just'what we do with them and 16.' . hat is the. volume. w L 17 There are about 3,000 Licensing Event ~ Reports a 18 . year. This is since the'new' rule. Since 1984 when the new 19 rule came in. That transition point, the number of report's

20

- came down, but the detail in which they had to be reported -;21' . ent.up. But since that point in time, they are reporting w r-s 1 22 about 3,000 reports. l. rj ~' 23 Now, in our office, in Jack Rosenthal's branch, 1 l I [ 4 24 - at least two people read that licensing event from cover to j 25 cover. That's to understand.it. That's to try to digest Heritage Reporting Corporation <mnua g >.j ~ -~ ~ (39. '] % u n,, s l:W< %,:.. -o-3 , - i y. y ,. [ [ (,.y.. iit. 2Thatls:part ofLour-job. Just to go through.those re-I I ' Wr~Y ~ i [ M >4 2 ports. I i ( /3; i .Now,: obviously there '.s a :large variety ~ in the. nature. ] 4 ,Q7 )4 ~ of the reports..: fYou get a. lot'of them that areijust a-i s 1 t 5" spuriods0 actuation \\of an engineering' safety feature. It must j 6 beJreported by the'. regulations. .But you can.really digest '4 g L7 'itivery quickly7 and setilt'aside. i . s-X 8' LOther reports take some time. I mean a detailed.. i <fevent willEtalde.somecmore time'.and it-has-to be' digested.. i 9J

j

,y 4 ^ og fl0i Buti we do 'a two-person' review because. we categorize 'it. Ini ~11 .otherwordshthe! reason'we' redoing:.this'.isito~ decide'how: at ; ~.12 .important;is that event. And we don' t rely on just one :in-- 13) 'dividual.to makefthat decision. A reviewer does'it and then-1 14 his section'gets..it'. f 16 MR.'EBERSOLE: I remember a lot of' argument about h 16 - reorganization in the context that you would not report an 1

[/

l17 LER unless you-had>a functional failure of some parameter. i 18 In short, you would not report a degradation of a function - 19 ' .down to'the last line of defense. That would appear in.the i i L20 other data base. And, you know, there was a terrible fight "4 21 ' about that. Of course, you wouldn't get as many' reports i 22 i ? because you were waiting ~until you got a full functional' -. 23 failure. Has that been a problem? Or have you overcome it l24 by .a new subsidiary rule that make you get this data anyway? -25 I'm' talking about getting down to the last line of defense. v. Heritage Reporting Corporation m.n. L2 _ 'k [>*' p YQ .y x 4 n ., ' p' L 40-a ,'C3 q- ~ e. 3 s 9" 1 Ih O ', ' 'I" MR.:NOVAK: By giving.up LCO action statements, Il l i 7 2:[ Some of us lament that'information. thinktis what we gave up. ~ a.Weidid'deci'de that the.NPRDS,would'be able to provide it;and-i N c "3' ) 1 .p ? y" p ':. .d44 ->'.that' has'not-been true. ~ u o J

5

, MR. - EBERSOLE: That's what I've'been afraid of. . l ? 4 a q v. ' MR. 'NOVAK : It-does give'us component, failures 1 4 + 6i 4-f j 7-early_but'it's'hard to establish trained-failures fromithe-- E '8 combine'd. complement 1 failures. p. L ,-MR.'EBERSOLE: What bothered'me'wasLit'wouldn'tf '9 mi 10 tell.you how-close you-were to big trouble. cThat.is, you,. ~ q III know, you1would3 get--well, ' you would get thel reali f ailures,- [ t ; s124 ~ . the : serious. ones. , E, ; j '13 : .MR. NOVAK: Yes. ..y i 14 MR, EBERSOLE: The-serious ones. What it wouldn't t i15i t' ell you.is.the vastly.more' numerous approaches to diff1-d '.Q 16 culty'which is=wherelthe statistical base probably:is in: ] o .i I - 17 fact. How close you'get'to'it.. JORDAN: Right. We are looking.at that from 18 MR. ~ 19 ' the standpoint of performance indicators and we may have I 20 more to say about-it.when we're done. d

21

.MR. NOVAK: 'Very briefly, with regard to cate- ) - 22 gorization, let me leave that and then we'll-move on.. When ' ~ 23 these people read these, they decide just how important is W '24 that event. We basically;have:got four categories and I

y 25 won't spend.a whole lot of time on it.

5 Heritage Reporting Corporation (202) 426 4404 l l s 2 ,l m e. 1. o. ,3 t s v .g o y. _ v e y - r s '}; e,$- F A' w3; ,'t i k [1j . Category'I is.something very important,:let's get g '2: .on;ift. -It has.an immediacy. ,c ,1 .L ! 3-Number.two is one which appears.Lto have safety i [ C; n ~ '4. Significance. In other,words, fit's worth looking at but no 1- -5 immediatetaction is required.- ,c c 1 ?, 'l 6-Three,is-one that we're'not really sure of. I' d W~ ! ~' .7 mean,,there are. things that weijust can't tell from first? I <i %, - <o l' o; 8 cflush. In other words, it could go-to be more important or. j + j ~ 'ticouldLbe dropped. 'And.we have a Category III and I'll .33, 9 i 'sf. 10 get back. to that in a moment. i 1 111 Category'IV;is'one that we can look'at'and say..it-e. .I v 12 .is'not a significant safety issue. We' can pretty'much dis c ~ 13s miss'.'it on'.that basis. We' read it and s'et it asidessaying_ 114 'it.doesn't tell'us anything more than what is there. p 15 But III requires'us to go back and look'again. We 16 don't leave anything in III. III either goes up to II or. D

)

( L 17 - it goes to IV. But you.ha've got to do some more work. III 18 is a transition if I could put it that way. c .19 DR. LEWIS: When'you say safety significance, that 20 can mean two things. One, it can mean threatens the plant.- 21-And the other could mean it gives you information you didn't c,_ y '22-have before. Which of those? Is it more plant safety? cs 23 MR. NOVAK: I'd say it's the former, plant safety. 24 DR. LEWIS: Even though the purpose is to develop 25 information that you didn't have? I'm really very interested 'Herttoge Reporting Corporation <m. i.. Y- y- .. --- - r - ---

,5 l

.! '#II sl[ t-ftg ] g ,y, y y;;,, v 'y 'e ,N ^ -.i.( q ) ; gg y y ; c..9'1 (, 1 st-f 4 ^ ]. j p". K W~ 6( g# y'y; ;4( 's t

c

',m ,= ~j ,. j f jin" criteria; for :when: things; are'.significant. f,[hl(f c. x a [.n.fr _3 (UV', ' 1 '2- 'MR!LNOVAK:

Let me ask' Jack'to answer.-

I/mean g 1 y? q ( 3 whatiI ;wasi repe'ating here. is'. pretty..much ;ba sic ' rules, ' but f Q[.%./ L r - Mlj' I w, ,1 'alwaysgwhensyou'Edeal with-issues the= rules;;should.be be.nl a' s 3 J' s. y'il, . r '/ s. 3 . [, cl3 l ilittle bit if you feel there'sya basis' to look at onega little j q;. y q

6L cmore., But.what' we're looking?atsis'weLseefan event,?whbtlis 1

"g. [ w f,,. v. ^ hefsaLfety significanceiin termsfof-~ ax ,f '7' t , 4 7.,.. '3.t sh 'l'm trying to' understand what , 4' .8, -MR. LEWIS: ' Wait. 3 y ~ !y Jd ,4 ., ) (

9'
theswordi" safety? significance" means: and maybe,;Ja6k cany.

m n o' j, 4,; .J:' 4

f) 210.-

-explain 4it.> d a . g e. -11: MR.~ROSENTHAL: LJack Rosenthal.t ' Reading tNefLER 'f ~ H .1 > u;.n a 4 -12

youJ should' have an' eventlwhiclic has one.or more f alluresiend."

i /-, [ 13l 'couldihave' led'tofa'.significant' event, e can lookraN 3.. r 4 [s. l4T failuresofsafety..: systems:wheretherep<sM(noinitiativea'ndsEy 1. w ~ 4s< .. elli, what happened if-there was an-initiative.Then when we 15 q o .getintomaybe'aninternalfisodandyouhMVwhatadditional 4 .16 : H 17_ ' things.would have had toJhave happened-- a more' serious i 18 event.. We see a lot under'II which we send on because there l N, r,- 19 : are other programs and not to take action. But - I would1 call. - g '20 if.you find that. licensing reports is-in and you find that y. y a ,1 t f21_ his motor operating belts are not environmentally qualidied,' ' y* w n 22 I would call that a significant piece of information even 7 23 though I.didn't have an initiating event, I didn't have a cj i 24 challenge. l: 25 DR. LEWIS: I see. I tend to, believe it or not, Heritage Reporting Corporation m sna ,i b 5g W f m mq'-3p 19,4 -e c a-m t w-3 'W 4 " ', q m. x }/- ,0., - s .s y p =t s - w mr ~y - s ,g sqq .f, Y. 7 ,,p.. 43-p b.,,% *. 'W sg ~ ~ -,y(

.. s

,. m.1 N'.,,,, m.think instermsJofsfall: trees and eventitrees.

YouLknow,

g. -4 (1;

1 f gh # 9d' J2' ,(theimind; boggles.dt theiconcept. .But~ILcan': imagine-things 1 G., y d ithatitell)'yousomethingabout:aplant.yethaveno' safety }W ,,a i M [4 8 4l significance:indthe. sense thatethere's noLthreat to.the plant. f U , y. 5 I L. That isjtogsay,fsome' thing that!-you tnought had a probability' ?! ! c ' i l f of 110.tol-the mi'avn'((pbppens :four? times inua row,.. but 1' E 6~ .s ' actuallyf it's': nd['all,t at significant. Allit teaches'_you [ 7 a ~,s .6 m QT@g 5-

is thatly~ou hsye somehow.a bad-way of estimating that, pro-m

.J,' h j9 bability nd?I uld-. regard.tNht as extremely significant-Ko, h g 4,, y.8 i to -eve'nKthough in;the sense,of threateningithe future of the" V i-fjg- .c pjk d 1 ,.plantOit has.no reallimplicatiopt,. That's..what.I'm trying ~ W :> f E12 -

to'.gropeJ or,;becauseTin ihe' end-it:would.be nice--and'wel

%q.3.. ,a' 3 r

.b..,n

' ' g-( )t M b3 'will'beitalki~ng abd,ut how you judge the significance of-y: "a vu R;, 7 $4; ithin'gs.; -'I' alwaysLipel' that,the significance of somethingl , fs n v. g ]15' il , q; '.N.f ' is"in direct proportion to the que.stion of whether it-was ~ ', -q! ? 1 ..,7 f N d ' 26 9; ulienpected. That is, if things which were' expected happened' J kf,f+lt> l

j, y,;

en 1 ff }R " jyg "a.. 17-( even dhough~they threatenedL.the plant, it's not particularly ^ g.p y ~ ^- 4 r E j f .v i8 -significant in.the sense that you haven't learned very much >n . I; fi Ltc st- % 1,,f 19 4 even though it may. threann the plant. ' h j.ki t 9e u (r Qj 20-So I'm oriented toward learning.rather than pro-Q n, i -21 tecting the plant. It's somebody else's-job to protect the j i iJ t i 22' . plant. And I'm,trying to make that distinction. I will be j mar p' G I ' t VJ 23 trying to as we"go along. p - 24 MR. NOVAK: Let me just make one point. I under-I &4s [d k 25 stand your comments. What's important though is that ~ u n. m M Heritage Reporting Corporation Q . m ma t t j \\ i t>.,-A 8 - s-, j .-$W m f "Y UdS v s a > <, 's 44 1 A i / K5, ' }-) } 1 ( b;% s: l 3, J ydu reqognizef what J K is that we rch6. ' In other words, not { o ( ') l ( ) j. ,/ j' ) , disagreeing'h.ithyou;'butnotallunexpectedthingsare J 2 ,3 .necercarilysrihortaple. ' {l ,9 v ol 4 DR. LEWIS: I understand.that. )i. N t. s g f(h f; 5's-( ,.MR. iiOVAK: We(fal,andI .] think-- ri,,t i - q .g, 6 't DR. LEW 34 We like to think that, I think. Ity t ,.s ? MR. NOVAK: /it may be -you may want to comment on .; i ; t j ~' f t e 1 the next proposed rule changq,. but what I'm saying is and I 8 I !]:, # don't maan to disagrqe, I'm jt.sk'saying that we're dealing t t 10 with a special data bpse her-a is,what is required-- 11 DR LEWIS: I understand is t. 12 MR. NOVAK: --and reported by,the regulations. / 13 And-t/dat 'is really the starting point of our knowledge to ] G,)y; ( yi 1 4 { f g" 14 look,at this data and.to judge the significance. // r j 15 MR. JORlhN: yhd I can help by saying that we j, 16 takeslicesofthi[) thin) from many different directions. ./ / -f >s 17 And' Aha direction we are ta'lking about ;here is reading by t' N i y 18 9taf f membeth the report i'tse16 and trying to get lessons l 1 .s 19 ous' of it as you indicattd' : hat are applicable to other plants or to that plaru'h;te,1h. But we don't throw the l 20 [21 report away when we're done reading it. It's in the data w -S , 22 base and then can be searched several different ways in ) ) )\\ 23 connection with otteer events, put together a story that we didn'tseereadingkhat report alona. So there is always [; 24 25 second sicjn't in this abaa.' And we'are always trying to ] 4 I ] jh!ritage Re@or9ng Corporation l 3 (\\ , n.n wa.n l .f 9 l 2 ___ _ 's P;. , q,3 ,bf # ( @ q,' 4 My J e.. q1 i %, w k,,, ' ' 61 ./ - '4 .s y- + 4 L M I hg .45 i W[$g{.g j" ~ < s g q t g m 49 .j. m , 1 1' ' fassemble;throu'ghithe. Trends'and. Patterns th'e variouslinfor-- ,d UN,, r NZ 2j$ .mation i'nto^nowisomething tangible as a. potential b'ene' fit.. i i + g , ~ _,y 3 'SCSI(thinklour: significance.is~.probably different than-yours" i l i >W4

.-. 41 Lwith1 respect'to.the reading'.of'the' reports.

.i L t" I5" p DR.(LEWIS: ' No',.I-understand--that. And.I'm only o f6 i revealingf my own bias'here' because I think the 'whole opera - 9 T J.E

7. '

' tion.'is designed.to give us information-- 48L M R '. JORDAN: Yes. 19 DR. LEWIS: -- and;there really are different 10~ criteriarin:the search forlinformation and the search for safety significance-as I understand you've defined it. [ . l l, 12-lI'm having < trouble thinking of examples from - .y } l13! ~readtor.sa'fety. But I knoV a great example from physics, j y 14 'whichlis..my real job'be'cause:a' bunch of my friends got"a) q 15 Nobel Prize'for discovering"something in which everybody in n

j 1!

16 ) .the physics community knew'that if you built a certain ma- ] 117; chine and operated it in a certain way that thing would be j m ..'18 ) observed. And so it was-absolutely expected. There was no-4 19' addition to the information we had in the world. And they + 20. shouldn't have gotten their Nobel Prize for it, but in fact 7 '.21' ' it was an important event even though it was expected. So \\ ..* K). 22 b it's that kind of distinction I'm groping for. Maybe'it's

i A

23 - 'just subtle. L4 MR. ROSENTHAL: Let's take station blackouts which 2 i,p-(.i. a 25 has received an enormous amount of Agency attent. ion. An Heritage Reporting Corporation <=>u .o_ ,..t [ ' 4 4 j a _ {:o: a<>.Tj, .,.6 c. yl >, ;M:, w o af .; q q

,[ L. ; )

,Y [+ ' y :. :: ; ; e t ' 5,s '46' hy j + fi -1, n g((f - alloti.ofLresources-on-it. If on the other hand,'we find'a-1; Devent" involving lthe simple loss ~of power. I'wouldn't' spend; 1 Ml j .W -2, 3: rea l-[f ailure-- -

P i 1

4 L - 41 .'R.. LEWIS: '"That's'important. D b -nMR IROSENTHAL: Then.we'11 bite:off!on that.

5, s'

y 7-lNowF on1the'other end of-the. spectrum,,iffweThave 6 u L:- 17.; tan;evenb with-is';.outside the design basis of the plant >as 8 iwegunderstand it, even.though nothing happened, we:would'.- ) y a 19J

spend ~some' resources.to'look..at that.

j "\\ 10' DR.. LEWIS: Oh, sure.. And'you should. But shation f e

11.

blackout.is.a good example because.a station'b'lackout does-12-threaten the' plant. !And.we know there will be power. 13-Ef ail'ure s. 1 '14

MR. NOVAK

I'm' going to move on-- 15: MR.='EBERSOLE: Well,ibefore you move on,Llet',me ) l 16 get a couple of answers here to some quest' ions. The'first ^ i O Ll7 - 6ne is under the new LER system where you point out there's 18. 'about 3,000 LERs a year, can you tell me roughly the dis-i 19 tribution amongst these categories on a percentage basis? l )a 20 How many are-- ) i 21. MR. NOVAK: There are 3,000 LERs come in, and ( 4 22 there's about 200 in Category II. l 23 MR. EBERSOLE: About 5 percent. f;. 24 MR. NOVAK: IIIs get arbitrated away so they don' t i 25 count. 'And the rest are IVs. Heritage Reporting Corporation r <mn i. -1 m ~~ 7 , g. .a, <r . m,. . v. .c i L, .9 ,7'p ; 7 w 'u Y. .. y ^ [' N ^ .7 o br, u. <j ,4 ,. h s. I - I' MR.lEBERSOLE: 'Have you ever gone>back toilookLto; y0 ,gy 2 ' see under/the:oldELEN rul'e'whether this.percentagehin^each i j.: i. cate' gory has? chang'edisi'gnificantly~as a result o'f presumably: .. lyg. Q ,] .4j ithr'owing"out[noisesand just' concentrating more on real: events? n ip 53 lIs[it still running the.'same' percentage?' If it'is, it's kind j a. db 6 gf surprising,q.but-- ~ cy I' 07-L MR..NOVAK: -I don't:know. Y .~ MR. \\ EBERSOLE :.You might want to look sometime, just -j 'E Philosophically, t'o see why'when we thought.we got rid of.all' -l [' l 10 i .the-Dnoisesthat we're still~s'eeing the same percentage amongst a ' I I ;- all'the? paper. 'I'in just curious. i 7 v ..12' ] .Now, the other question.is INPO does'the;same q ~

j p'

0 i.. f... '.13 things. They. produce their list of.significant events and-i so they grade.it. You excha'nge'those lists I think with ' 14 ' l 15" IINPO periodically, don't:you? 16-MR. ROSENTHAL: We have. periodic meetings with f7 17 ' INPO and describe what we're doing and what work they..are 18- .doing.; 19' MR.:EBERSOLE: Well, that wasn't really'my ques-20 tion. j.. 21 [. MR. ROSENTHAL: But we haven't been swapping to 22 the best of my knowledge lists of specific events. n1 23 MR. EBERSOLE: Well, what I was really wondering is l 24 ': .do you' receive their list of grading on events and compare 25 7: with how your grading works out? I just wondered if you were Heritage Reporting Corporation <mnu } '. >. e w ~; j . g 3.g } , y/ o a m

.g+ yy y.

y m 3, y .n 48. ,' u

(

.y ywy La. i w !c s' r ,, [ 5 3 f f

r

.' I t doing1that. - r .,3 .m. . i; g -. t 4.

27-i-MR iROSENTHAL:

,No. .7 t 1 .3 l MR.iEBERSOLE:. Is:it because you haven't' indicated. s 1 W 24! i nJ. interest'orlis'itlbecause of-nonavailability?' .5 -MR..ROSENTHAL:. The.latter. The materialfis. H 1 zu .6- ..availableTro us., 7- 'MR. EBERSOLE : ;I thought. lit was.- You just ' haven' t-- -~ R 8 . 'M R. JORDAN:We just' haven't'done it. 9: ~ MR. EBERSOLE: 'Okay. Thank you.- j 1 10. MR. WARD: 'I would think that would'be of interest -l oll-to you. :: I mean they write a relatively small number..of-SOERs [ they; call:them. :But-Inshould think that would be- -

12-13 MR. EBERSOLE:

They go through a grading process-I14' .ontevery.LER just'.like AEOD does and they grade them.

.15 '

MR. WARD:. Shouldn't a' grading-process not;corre-lj 16 '- late with yours.for some' reason? ' I'mean'do they have some ..171 .different emphasis? l 18'- .- MR. JORDAN: The criteria are a little different. l i 19 But'I think in the; main--maybe I have to explain something '20

else, h

f21' Performance indicators has an i~ identification of c. 22 significant. event. It is a criteria that was applied largely 1 h-1 pQ l l 23 based on the work they were doing for preparing for the

.3-24 meeting of the ACRS essentially.

And so there are criteria 25 there that are similar but not identical to this significant l t Heritage Reporting Corporation q <nua i u n.... 7. 'Y r ;-.$< ~]: ' u, + k_ ? ',, ,,'u s c f f ,jV, 49 / m 4 p,p q 3 J L1 egrading1. criteria:thatu'is'used'.invthe review'of events.- We: [

2 d1M compare the,.significant events there.with INPO because-

'\\ - [ L i .thsy have"also performance indicator' types. And if'we'had. 'c., l3 f. '[ ' '4 , airea'sonable overlap with a little' spillage;on either side' m .o + q f' - 5.!

ofiit, so eng'ineers, reactor.: people,-come;out fairly.near e-y e

6 -- to-the'~same with simila~r~ criteria.- So'we'have done that. , ' :9 .l i 7 particular comparison.- ~. r 8' MRi EBERSOLE : - But it would'be_ fairly.important- ,'to. knowLif:they judged.'a.particular event as'being'signifi-9 1 '10l

cant:;and' you didn' t.

I was just kind!of curious as to why 'lli you' missed 'it.'or,why would you think they'would think it was u, -i 12 ,important. q [ 13: 'MR. JORDAN:. It's-certainly in',the; fuzzy areas ~~ ] ^ 114 where there's:a difference. -You know, if there'is a North' q: (15 2 Anna tube 1 failure or a Surrey pipe failure-- d L16 'MR. EBERSOLE-You won't miss those. i y .m ~,{ .17 'MR.: JORDAN: We're not going to miss those. M' '18 But it's in the gray areas where-- ,s 19 MR. EBERSOLE: It's this few percent, that 5 per-i 20 ' cent, more or less-- t! 21 MR. JORDAN: Right. j i 22 MR. EBERSOLE: I just wondered how you compared U '23 with their particular--I just wondered also if you even knew i 1 y, i j;

24 -

what percentage-- l 25 MR. JORDAN: That's a question that we'll look at. i Heritage Reporting Corporation y m us.m. ____a____-_______- y -- I 50 l 1 I appreciate it. l i 2-MR. EBERSOLE: It's interesting. l 3 MR. NOVAK: Okay. I'll move on. i m). 'v' 4 Now, that we've read all these LERs, we'also have. 5 Oak Ridge under.' contract reading them also. Now, what they. 6 do is they digest the event so that it can be now placed 7-into a data base, but it's done in a way that actually 1 ') 8 breaks up the event. I'm not going to spend a lot of time i 9 but'it's a thorough digestion of the event in terms of se-I l 10 .quence. And they put that in and then they also look at a j l 11 good description of the event which you might look for if I 12 you were say, doing a key work search. So separately again- .1 r 1 ( 13 these people do that. And everyone of these are read and' l 14 they are put into the sequence coding system. -15 Now, what we have is a total of 26,000 LERs in our 16 data base, but since '84 we've got about 10,000 with that 17 level of detail that permits us to really do more thorough l 18 review of events. And I just wanted to put that in there. I 19 So that goes on. 20 MR. LBERSOLE: You put 24,000-- 21 MR. NOVAK: I said 26,000. ) -5 22 MR. EBERSOLE: Yes, but all 26,000 are already w) 23 coded? f^) 24 MR. NOVAK: Yes. They are all in the data base. l \\._/ 25 MR. EBERSOLE: Okay. l l Heritage Reporting Corporation can. l = ,a; ,, p ~ >; J 'b ~ r ['t. 3 )..

l]51~,

l M c ?- y N....;. j, q r). 1 w g.

1

,g 4 e s a m.g s, a. . v. q Is i:1': MR..NOVAK-:'.Now,,'itLdoesn't'just stopithere, andiI (, ~ "A s , f it d- ~ ,[. m: y, as D2; thi'nk1probably Dr. Lewis-~can'. recall from the, Reactor Safety th y , : 3'

Study.

We also' haveLOakLRidge looking'at specific. events [()K 97 A. : , 4;

underlwhatiwe" call,an Accident Sequence Precursor. Program.

e i 1 ~5i )Whad thislisJintendedLto do is look at some of'.these' events d ha, ~ i> 5; c ,. = _ 6' 3 that'have.som'ething.that smacks.you', either it:was;an event s' ' ~ ~ 1, ' fo,C g.. ,3 .p .v: J e 7: .with'some'. accompanying failures. TAnd wejlook.at it'to see- ' J. l ch , L8i . :what1 accident sequence precursor it might suggest. ~And1we.. f .-/'. m l9~ /yook atJthem'andJbasically what we're'doin'g'is,trying to. L: L. u f f Llo-identify.where they occurred..And so if you were to ask:us,' j q it. isay,cfor,.1986,,we added about thirty-five cr; forty events; -3 4 .Il2 - of'that' interest such'that'if you wanted to go.back to a Y ~ ,-[ L131 specific. plant and ~say, okay,-over'it's five'yearstof opera-2 [ 14 tion, howimany, events of real. interest, and.if youlwould u'se 1 L15! the Accident' Sequence PrecursorrProgram as a threshhold,,we ?.

16 wouldLbe able to tell you' exactly lwhich events'they are.: ~So 17

~it's:really trying.to' capture the very important events that-5 18 occur at..a plant. And-we've~been that now since-- i

191

'MR. WILLIAMS: '69 to the early eighties.- Lj 1 1 ,20 MR. NOVAK: There has been some criticism about ,21 the absolute values that are suggested by this work, but I w 1 l { r 22 think.it's more important to look at the trends. In other 1 >.3 '23 words,'what you want to know is is a particular mode of ' 24 l operation of the plant, are they having more than their share 25'. of these types of events. And we're beginning to use that ^ Heritage Reporting Corporation (202) 62s.4 ass I l my y; y s w .$D 'n L yn ; 0: :.h. e 3p,,~ m* r 'f s. g i ., c. a, l 1 C ^'

: ' n,

.r, E .sWp' < <, ..1

more and more now when.we'flookcat-plant performance.. And,we.

,f~ 5,

y 3 f go back candisay just :how many of; these. events'have: they -had J2!

g. [h:;,y L3 over the"last::-five;or"six' years.So'that's-another thing -m . 3 4 'that;we're doing. ,, a:. LSI RMR. EBERSOLE: Tom. o e

7. f, y

y, I 6. t M R'. NOVAK: 'Yes, 1 i7; 'MR;:EBERSOLE: There'snan area which I'think is; .e i L8' quite' interesting.from the Niewpoint of learning from ex-9' L perience. and thati 'is-basically ' the problem:; of : system ; inter - ',X 4< 10: . actions. lCan you tellius1what you're"doing?in'. terms of.. . g' L .g' ^ -l 11: looking at.LERs and' thinking'about. system interaction'Jand

12.

' how are - you--usually ' you: don ' t' see. this in one nice. event.- f$ 13. You see~itfin hundreds"of'little events. : Are youimaking E t 14 ) an;y' particula'r effort to pull together the systems-inter-- 3; '15l ^ . action question? a 3l -16: MR.ENOVAK: -I'11 let Jack, add to my comment. .I n 17. do'.know-that we have.some people specifically.wa'tching"it. ] f 18 In other words, I.have.a very. talented engineer, Dr. Pete 19. Bland, who spends a lot of time. .If there's any system -. 20 interactions, he-asks that he be shown a copy so he can.look 1 -21 at' it. , 22 What we're trying to do, and this is may be one 5. ; l 23-example of what we call a watch list. It's kind of a soft j i 24- . interest level. In other words, we have a watch list, and n l 25 Jack will talk a little bit more about that, which kind of f] Heritage Reporting Corporation (202) 424 4804 'fi s nm..,, 2 9 J ' w}i:._ / j .., 9 f, "_., .,}. s' 1 - 4 g '53: M L, - /,, e 7 \\ pc > <t 'c, "M 4 4 t? N.;4 l' And. systems interaction: -n[ fI Lidentifies: things ofsinterest. i.,. ~y ^ f p-a\\ + 9e ....t r, < e s.- , s s i , y .- 2 ' icould.be:onithe watch ~ list.- Now,' what triggers the'studyfis 33 p. > f' f 3 'more or$10ssithe!i'ntegration of a nuinber of events.j When you ,l y T V] ~ . s a y,. h e y,'I we are seeing'enough;ofithe'se. Can we= structure; a 14r y ..y 1 ~5-some sort;offaniengineering revi'ew tohbetter define what"are ] 4[ ~ a . 6= fsome ofithe prob 1 ems.of syst'emslintbraction. wy. 9 7i I think..that's' the wayjILlook at>it.right'now. 4 8L l Jack,'is'there!anything you want-to 'dd?j a T -9 LMR...ROSENTHAL: :Well, I can't,take on systems i .g ' M 10' interaction. IBut I.~can take'on itsisub sets., So we:are R ..-11: looking at-- q i l12-MICHELSON: What1do you mean you-can't take-it'on?. 4.. 3

LI mean clarify =--

w. i t' il4 MR. ROSENTHAL: I'd'rather.ftry to'do discreteL -l !] t . 15 volumes:and'get work products.out.rather than write the 16-encyclopedia,.'and so, for example, interactions.of the fire s 17.- Iprotection. system 1with safety' systems and reconsidering-that-- 'i 18D and'all that.was done.to'the appendix R is a fair game topic l .o t 1 l 19c ~and that's on_a large listing. But that's not the entire-- n. 30 We had one in which there was a lot of chlorine being moved 21 on': sight in connection with new scrubbing and stuff, and y 22 there's;a fair amount of hydrogen on the side-- _ 23 MR. MICHELSON: But there's a number of LERs on

24!

almost a daily basis where water has gotten on the floor and n i L 25 it's gotten into here and there and it's caused some very I i o L,4 Heritage Reporting Corporation 7 <mnm. r n- -._-2..._- ~{' .g 14 a ,T ,,_j. _ -] ~ h" 1 y-7y 1549 1 A3 r s w ci, 7, bin 5 .l.- interesting system effects, and not always aggravateG'by. ~ l yG; " n-L ~ 3

, N-: 2 !

[delugef. system,:forTinstance, sometimes by otherfthings. AndL ll ja, N 4 3 Iljusth. wondered 7if.-systsm interaction.was"getting'a. systema'- a o l ' f', -l L4 ticilodk because.it's going to comeJa.'little bitsjand. pieces.. Aif g, Si -It'.sYnotLgoing'to jump'out to you~in one< big' event.. lIf it .o .does,.it's::probably going to be a'really; interesting.one.. ' 61 o,, -But.-do you attempt,to pull-that concept together-n7 l i <;g W

8-

.or-- 1 i S 9' 'MR. NOVAK: No. Not as-an integratedLprogram. ta 10 .I: think realistically it would take a substantial amount'of 'E i c ~ 11 ' our resources. .I would also look and see what other offices I knowing Research and NRR_have struggled with. '12:

are doing'.

13 i -! systems Jinterac tions.. g'z ' d i14 : MR.-MICHELSON: They areijust thinking about:the ra '15- . concepts. You are not getting theLdata into.the thing-- l1 16: 'MR.'NOVAK: Maybe'we can assist,them. j I

17' MR. MICHELSON:

The data are spread all over, you 18-know.

i

'19 MR. NOVAK: Yes, they are. 1 20 MR. MICHELSON: It's not going to be easy to 21 retrieve even with the retrievable system because it was. 22 never--are you presently identifying when you see an event 23 that you think is a systems. interaction? Are you coding '24 ' it with a.particular retrievable capability? You know, we n. l

25

.used to think of putting a watch list as a special comment. Heritage Reporting Corporation (202) 428 4004 I r-x .t l \\' r y-4 M i ' 155' t. 3c7 .+ y( pr l :.; r + 3 4- .e i 4 5 1 1 LArecyou_ commenting--a spe~ial commenting' systems' interaction? .J ,l' c , p- '

Ql' MR. JORDAN: > Ye s ', ) I-think:we do.

I'n other'words,7 ,2 u-33l when wellook for' events.similar that happened at Ma'cGuire,- G: 4 W .j MQ-. . 4-

we;look-forLanJair. system problem in one' unit,,even'in;one c.

W- .i.. r ~ 5 . unit: from!another; unit, that wouldicausei aL trip or 'some other i

6~

impactfin another unit.- So we still havelthe coding lcapa- ..c ?71 ' bility.. E .8 MR2::MICHELSON: My specific question.was if I"came 1 w, to'youian'deasked you to give me all the'LERs that you think: 10I have' system interaction effects, can you call up the computer ~ 4

3., -

iIl tolgive you'that.with a.particular word or whatever? q .i j 12 MR.'NOVAK: Carl, I guess I would.say we would-131 ..try. How successful we would be in getting you-- > 14 ' MR. MICHELSON: Because-the " comment"'section'that L15 you had.-available on the old sequence coding allowed you:to e. 0-16: put inathe word if-you wished system interaction, andLyou j 17 could search:'on'.that comment to see where it' appeared'. I 118 just wo'ndered if you'd'been asking Oak Ridge to do that be-19 - !cause it has to be don'e manually as the-LER comes through 20 - 'the process. I can't be done later at all. 21' MR. NOVAK: I don't know. They are doing some of ] 22 that. I don't recall, you know, whether systems interaction '23 is-one of them. 24 MR. MICHELSON: You know, you read them and you say ~ 25 ' gee, here's a neat systems interaction. Now, how can I Heritage Reporting Corporation

=

g .q-p,- 5

C7g. ,.f

7

[,p - .q: g 79 y, 'i . 5 6; 1 1 _ ~ x. Y I .n J sJ. F 3 qe lf 1 recoverit-twoheaNslaser, No't remeinbering the particular .j"'V ..planb arid 'all that$, A' df[it wis ~ intiended that ' the " comment" l "N 12; ~ n j m3l sectionJdo that you remember. That's where you put the word. ~ .nc .. t}.

4T

Anylwatch l'ist : word. you wan't would. go in there. and t',o

'S-ryears;fromLnow!I ca'n:.go:back a'nd. recall all-the systems 6~ . interactions.;ILthought I'd get.. 'It's-one way of getting. 7 L

a handle'on.that.

8- . MR.'. NOVAK : Well, let(me'look.and see, Carl,:andi 9 we'll call;you.back and let you know what we'do have. T10 'MR. MICHELSON: See'what you do have.. If'you~ ^ 'll would. l 12' MR.!NOVAK: Okay.. Mark will mention later,on ini 13 ' (the; discussion,.but we alsoLtake the same data and put.it. 14: into other data basis basically :at Oak ' Ridge to do trends 15 -and: patterns studies. There.we.will be looking at the, trip 16 behavior,-'engi~neer safety feature actuations,-'and: safety. 17 system failures, and test spec violations. Those are just'- 18 kinds of things that'are trended and I think some rather q 19 useful information comes out of some of the trends. They 20 do' point out certain overall. behaviors of plants over an 21 extended period of~ time. '22 I just want to finish our data base. You'll see 23 'we have something on foreign events. -Actually through the L24' incident reporting systems sponsored by NEA, we review about 25 one hundred events a year. They are not very exciting events j 3 I Heritage Reporting Corporation j on was. 1 N 1

" f g. '.,_ [.. Ci R j1 1 e x, ,3 WT l Sj; '4 3 ( d ,s .t "4- 'l'. i 4 j. s .l i i n 1 3 I '. "i iwea.d51se?mylbottomilinez /You reaNthem We have a-file. c[q .\\ i 12 TYou V r recal1ithem up..-We do,have a file orla data basesfor [ 73; $ ~foreigri (.snts. lit's managed 1for-us.at Oak Ridge. -We read-I T L 4 themfthough to see'if there's'somethingLwe'can learn about-J ongoing plants. And I'.ll; fess up. ;We= missed one awhile 5 .a jl 6-J ag'oL Iflyourgo back-recently, they' inspected the, pumps-J 7. "af Palo Verde,.'and.they.found cracksxin'the reactor pool '8 . pump shaft.. Well, there l's~at least two or three, pieces of: 9 data in'the foreign event. file going back to 1985'where you. l 10-had complete fai' lures'of the shaft at KSB pumps.

Well, 02 Lil.

fo.r better -f or worse,-~we have a certain. screening criteria a

12 which suggestedJ1ook for generic implications from. foreign' b

13) events. In other.words,-our people went'back. They saw it. 14. They. -asked' for.' more.information. 'They confirmed that'it was-3 1

15

.a' single utility. application. We had seen events of that I16-typeJbefore. - And we' judged it.to be a Category IV in: terms. / Ll7 of safety-. significance. It happened before. It's-covered i i e t ,18 by our analysis. 19 What we're doing now though is we're going to bring n 20 up those kind of events. If we see something like that, we're c h i< 21 going to make it known to the proper program office. In A; - 22 -this case it would be NRR so that they can sit down and dis-A l ,23 cuss'these events with the licensees. ((N 24: So we do tend to want to learn what we can from d, l-25 these foreign events. And we do spend time looking at them. Heritage Reporting Corporation m ma J__--__-.--___-_L_-_-

y Q3 lv. w % " J '

ig 2:,

~ 3- . c--- o u 3-

58.
l j ;.

~ ^

q 'f i e

zy,< v.

e

,e ; w ~ ,E ' - a

[

bi i i,

.: a

[g 'Ll; WButias I.-say,9there'are aboutiL a# hundred-!a/ year. ^ ~ :;yn; 3 - A )- 2 [n .MR.'EBERSOLE: Tom,-are those-tips.above.or.below, w-9 m7 itheirevolution coEnter--are'thev'abohe'or below the revol'u-a 3' rk i Y onE counter 'for'.theStrip ? system? You.know,.they usel gear; .a

J

'4; ti e !/ ' 'l .,.. ~, ~ 5- '. teeth Jthat. pass a Epickup..:

6

.MR; ROSENTHAL: .On those plants there'is an opti-7 4<. q 6 { r 7l / / cal',digitallreader of'the' pump.. speed'which,is.off the motor, J ..i-fj ; 4' ^1

s

so yod ~wouldn't pick up shaft; failure--

R o .'9: MR. EBERSOLE: Right. a i i 10 2 MR.?.ROSENTHAL: --but the speed.. indicated-- l11 MR.qEBERSOLE: Sat you are still. turning; y !2-MR. ROSENTHAL': However, there'sa' Delta--' ] 1 p. '.MR. ' EBERSOLE : But:it'.s slow, l /,( ~13

'MR.

ROSENTHAL: that'szelbowed in. primary systems-- j (15[ MR. EBERSOLE: ' But it's slow. 16 MR. ROSENTHAL: --to. pick up the degree of flow. .i-l h, s j 17 'MR. EBERSOLE: But that's a secondary pickup an'd 18' 'I think it's slow. Otherwise they wouldn't~have this one. l 19 'Are they ready for damage from loss of the first signal? i 20 Do you follow me? l 21 MR. ROSENTHAL: Yes. I think it's a calculated a 22 risk.

23

'MR. EBERSOLE: I don't know what the extent of the 24' differential is. The differential damage. 25 MR. ROSENTHAL: My point being is, you know, you Heritage Reporting Corporation m u. .3

.( 59 I are looking at critichl power ratios-, . f'Y. + 2 MR. E8ENSOLE: And core damage hf some. level which ] '~) ' 3 is not real'ly all'that fundamental.' l ~['T i'I b' '4 MR.'ROSENTHAL: But they.did have these events in~ 5 this case and they went back and actually analyzed the true 6 performance. They.still had substantial margins. 7 MR. EBERSOLE: So it's commercial loss, isn't it? i 't 8 MR. ROSENTHAL: Yes, I would think. And I think 9 in'this case Arizona Power is very concerned about the 10 shafts and they are looking at this. 1 i 11 MR. MICHELSON: Let me ask. What are you doing. l . ith the IEEA IRS reports. Are they a part of what you are 12 w ( 13 referring to here as the one hundred events per year? 14 MR. NOVAK: Oh, yes. ~15 MR. MICHELSON: That's included in that? 16 MR. NOVAK; Yes. 17 MR. MICHELSON: Thank you. 18 MR. NOVAK: Lastly, and certainly not of least 19 importance is component failure data that we get to look at 20 through the NPRDS system. Now, we'll talk about that some 21 more, but that's an INPO managed system, just for informa-22 tion..There are about 50,000 component failures today in ,S .V 23 NPRDS. And there are about a half a million engineering /~) 24 records. Now, later in the discussion I'll give you some [ :, %.) 25 more details. But this line here then is intended to give I [ Heritage Reporting Corporation y von.... ___.L:._..

U 60. 1-you kind of a thumbnail understanding of-that's our data. ['Y ' ~'o "k/' - 2' .That's what wo get to s'ee and from. hat now comes about our t .n Mifi, - 3 engineering studies' an'd what;we'want to do with the data. q,,q 1 ;- ,32 . 4-This isijust what comes in the. shops, so to speak.

sow,

,. i MR. EBERSOLE: Carl, you mentioned A-17. -Let me 6 mention one.in-the more confined areas. A-47. And we'll be 7 talking about it irt the full' comraittee this week. You know, 8-the interaction between control and safety systems. What - 9 part do you play, for' instance, in getting'into that particu-to lar area where we are talking about the centrol band width tj . 21 'and the rate of change within it. .This is a control system-- 12 ' implications of a control system on-- [7 13 MR.~ROSENTHAL: I would like to think that AEOD x-I j4 efforts of several years past would help _ spur on that study. 15 The~ones that went to A-47, it's been followed by AEOD. 16 But it's Research's job to work it. And about a week ago or 17 two weeks ago we got to comment on-- 4 18 MR. EBERSOLE: Well, it converged down to two i 39 things. 20 MR. MICHELSON: Well, I'd like to hear what you 21 were going to say. Dr 22. MR. ROSENTHAL: I did want to say that once the i .V 23 issue is found, and it comes a USI or GI or whatever, it's ('T 24 turned over to Research. They'll work it. We'll let go of

-)

25 it. Go off into other things and try to monitor what's Hesitage Reporting Corporation j mv u.am = _-__ - 1

} . F. 61 I: -l going on. If they are addressing the. problems that'we thought 3,- 'kl 7 '2 but we 're surely not going 'to 'd';plicate the ef forts of Re-3 search.

rm

'4 MR. EBERSOLE: Will you go back.and evaluate your 5 own view of the--the conclusions they reach which is really 6 only.two things. Too much feed water or too little feed 7 water. That's all they found. And I have difficulty think-8 ing that's all there is in that very large topic. Well, I'm 9 asking you. Do you go back and.look at their conclusion? t 10-Can you comment on it? 11 MR. JORDAN: If it's a'part of the. Agency review 12 process, AEOD has that: responsibility. Ag) - 13 MR. EBERSOLE: You do have another shot at it? 14 MR. JORDAN: Yes. l 15 MR. EBERSOLE: Okay.. 16 MR. MICHELSON: Let me ask on NPRDS. Are you 1 17 still monitoring the program to determine whether or no we 18 are getting adequate reporting then on our need to change the 19 LER system? 20 MR. NOVAK: Yes, we are. 21 MR. MICHELSON: In view of the fact you've been I 22 doing this now for two or three or four years or whatever, 23 what is the current situation and which direction are they ('T 24 heading and do you have any other comments? 1 'u) 25 MR. NOVAK: Okay. I'm going to answer them later l Heritage Reporting Corporation mm m.m. q l

62 l' on-in the program. \\ 2 MR.~ MICHELSON: Later? Okay . Good, good. ~ 3 MR. NOVAK: I've got a specific section and it will ,~,) -'s 4 give you a few more ' minutes-to think about it. MR. MICHELSON: Good. 6 MR. NOVAK: The next slide please. 7 Okay, now, we're going to get into some of the 8 questions or all of the questions that you asked and let 9 me sort'of--what'I'll do here, I'll start with basically 10 capturing what I think was the essence of your-question and 11 then trying~to answer it as we see it. 12 There was a concern over deciding which issues to

ry

(,/ 13 ' review and'it was sparked by your Question 3, which says,. 14 " Where does the focus come from when deciding which issues 15 to pursue?" and then also a part of Question 8, "How can 16 AEOD be sure that they are concentrating on the right issues?' 17 Well, you've heard a lot of that already. Our i ~18 ' daily conference calls help us to focus on the problems. ] 19 What should we be looking at? We do take off from these 20 conference calls. They are not long duration studies, but 21 they can be quick turn around. So that focuses the actual 22 event of interest. ,f-g i..) 23 DR. LEWIS: That only tells you what your friends (]) 24 think is important. It doesn't give you any objective 25 criteria. Heritage Reporting Corporation ) (202) 424 4849 I

63-1 MR.>NOVAK: Well, we make a judgement. _,rx g \\ 2 DR. LEWIS: Sure. q 3 MR. NOVAK: We may decide to look at it indepen-(..; /& In other words--and it does happen where the program 4 .dently. 5 office, maybe NRR in this case, decides that they do not plan 6 to do anything further. They are satisfied that the actions 7 taken by the licensee are a basis'for continued operation, l 8 for example. We may want to look a little more, to look at l f 9 some of.the earlier history of an event of that type. 10 Either in another' reactor or other events at that reactor. 11 So.we make prompt a study of that type just because we think-- j 12 we are not quite sure in our own minds that we know every- /~}. i .(, thing about~it and there's no need to follow up. i 13 14 Now, generally these are not major studies. We. i 15 call them technical reviews. If everything turns out right 16 and most'of the time NRR made the appropriate decision, then 17 our additional looking confirms their judgement that it was 18 acceptable. 1/ So focus in this case, you have a short term pro-20 blem and then you can have the longer ones which I'm going 21 to talk a little bit more about. In other words, a more 22 substantial review. Where you are going to spend perhaps 23 a man year of engineering review developing an understanding '({]) 24 of some actions regarding the operation of a plant. 25 DR. LEWIS: Now, I understand that. But the sense Heritogo Roporting Corporation om niam _._____.________.___J

Mu j< ~ n' r y, S + ye m%. .;;c.. 4 .. u s y p t 4

3.

s'- x-w," , -ij n.y ' 16 4 : j%d J >< -l p. ;; i t ~- -~. a-g e 't[.3j, d c j.; f( " -- 1 of,the question,iif.I' understand it,-was are there criteria -i 4 7' Y" Wf}, ."for (deciding,wh'ahlsJIImporNanti? Now d$ sNone; make - the 'deci2 k 4%@:g "2 2( ,e 7- .y, g g" *, c '4 ; L3!

sion? lAnd.yourfanswer h'as.'been,.well, we.make-the-decision ~.-

~ .. m

n C'

.4 . -' [< ~ h +' I 4 y ; "And/th'ere'snothinhN.fniprincipleiwrong with that.: 4 .) s 3[ 5 -. tMR.yNOVAK: It is7a, lot engineering. We don't! j' ~ 6: 'have7a'--letlme1say'it this'way. I don't'have a firm.ruld i ,t t -7 or screenin'g' processI thatEis so strict--iti's freally: the p'eo ' ple{tstjyou manage'and do your. job.. 8 ~ 9 .D.R.jLEWIS: -Who:have.to' have! guidelines.

10

, MRW N'OVAKiL'Yes. 1 x 11 5DR.: LEWIS: I'm not in favor of"a computer program u e e.c ' , 'to-makk the' decision',;but we, touched;on the issue.before' (12 '~ 13 'wheniwe were.: talking,about safety significance. !And what is. .j q I' (14 l implicated.- 15 MR. ~ ' JORDAN : - There are several levelsiof review ] 2 2 16 ' - though. The one.that Tom is talking ~about is the various l 17-

l. prompt review based.on the' event that happened yesterday 18 or the day.before and'whether AEOD ought to start based on 3..

l 'that information.: Dut there is the much longer-term review ~ 19 r 1 -in terms of reviewing the LERs and the criteria for signi-201 21-fican'ce of that which.is the second shot at the same thing. .l l

22 And then the trends and patterns that are constantly-sweep-

/ i 23 'inglthe data, looking for combinations of information from l l L24 different plants. So there are several levels and the one 25 that has the criteria is really the second level, I'll call Heritage Reporting Corporation 'i< (102) 428 4484 f YN v.

77 -, j j kI .h ' 74 st 165' l 6e o 1 f ) [i gg, [ 1 -- tit,;4where they'are>ccreening!for'the. significance'3one, two",-

j

'j s %<J Y2T /threefor four. Q^,, p. m:. ', ;ej f_ + [ 7 E'.3= DR; -- LEWIS : ' Whatever significance;means.. l ~ fight. .j 4 J4L MR. L J0ftDAN : ; Right, p 5.- MR.-WILLIAMS ~ As far as= events informa ion;goes,. 3 ,g 1 6 Iawanted to'take/a quick shot-at it. I think~we'have a 1 ~ d fairlyluniform: handle on addressing that issue.. There is i.7 0 8' Jfor.: abnormal occurences-a set of criteria'as to what consti-9. tutes anc'AO.' There'is a handbook. There is manualichapter. f?- .[ ! 10 [ There is Commission-approved guidelines. 7 i11 . DR. LEWIS: Right. 'm 12 ' ' MR. WILLIAMS: As far as the: Accident' Sequence -1 ~ 13,. Precursor [goes,-the' report iston.the desk.here and in itL Ti4$ you'll' find a' written set of criteria,,mult'iple failures, .i 3; ,f15' particulars of events, that it'would meet to have signifi, 16-cance,.from an event-significance perspective. 5 a '17 There~are other definitions of significance and 18 'we're attempting to drive these together to some. degree 19 where it seems reasonable to drive them together to have j 3, 20 some uniform definition as far as an event's significance ,s '21 - in some respect. 1 ') p. Then there's the other part that you are talking 22 V 23 about which is serendipity of finding from scooping the ( 24 ' data and kind of coming up with something that's not expected it: 25 a failure rate for motor operating valves or--generally Heritage Reporting Corporation m sua.

my.. m 1~ Ji].\\ n:7 {p, s' ,p: 66-1 o.. r + S v ) i l. d ., j. .I. '[ ^1. thesef. kinds of things'we1 call'NPRDS~for with some' success ' p. n ~4f, .2 : 'sometimes.-"And we?ve beenga'sked t'ofstudyjthe~ feasibility;of s i..,. 1 - l. i...,- .3; monitoring--benchmarking and~ monitoring., reactor protection y, ; -J L 4 ' system reliability:Lusing!these"kihds.;'of data basis,'and

~

5 'that's-ajdifferent kind of significance,that we'dibe'looking-q at; relative tolexpected reliability'versus' operating ex ' j 6~ 1 .) .7: .perience. d + 4 1 -8 .So:we're trying to-ge't.it from that perspective 9-to butEas far'as events'go, I think there's a lot of~ stuff - 10' favailab_le. 11 DR. LEWIS: Well, that's why I'm pressing for il 9 criteria because there is'a lot of stuff'available and it-1 12 y } l[) 'i -tj 13 always' takes much more wisdom'to decide what not to worry J 0 '14 aboutfthan to' decide.what to worry about. You worry about j s '15 everything and delude yourself and there are other agencies j 1 1 16; at which they.have lists.of thousands of. things that they: i 17 regard as uniformly important. 18 But I can--you mentioned some criteria. I can 19 imagine three criteria. One is threatening the safety of 20L the plant. The other is serving as a precursor to a se- ' 21; quence which could threaten the. safety of.a plant. And the 22 third would be unexpected. That is to say, even though it 23-doesn't threaten anything, it still conflicts with our pre-24 conception of what should be going on in the plant. A thing } 25 like that may not'be of any safety significance in the sense l Heritage Reporting Corporation oav u..a

r-- 67 1 that we talked about before, but.still if it reveals the I'\\ } \\ 2 factLthat there is something-you don't understand, i't's 3 worth knowing abo't. -And so'Ircan imagine three things of u .And obviously there is nothing articulate here. 4 '^ 4-that sort. 1 5 You are making judgements about what is important. That's 6 what I draw from this conversation. ~ 7, MR. WARD:.Well, do you--just to follow up on that. 8 Do you react or.think about the last sort of thing? I think 9 that's what psychologists call a fundamental surprise as 10 opposed to just an ordinary surprise. That is something to 11 look for. I 12 MR. NOVAK: I think we do. Clearly I think we ,a l j/ 13 respond to the unexpected. We-don't like to see the unex-l 14 pected. And it bothers us when something like that happens. ~i 15 We certainly go in and find out why did it happen? How was 16 it missed?.Did we assume it was so unlikely that it would 17 be neglected and then it pops up? 18 There are a number of unexpected things that hap-19 pen and they are not, as you say, they are not that safety 20 significant, but they do raise the question in a sense, the 21 broader question, of how much do you really know about the js 22 performance of that machine-- s 5 )' 23 I think--not to make it sound too run of the mill, () 24 but recognizing that we look at a number of events.. We 25 force ourselves to read every licensee event. We screen Heritage Reporting Corporation l m cu. l

.w , 3! -'. \\.) ~d {.'. 4 u

j.? ?

l$ l -r. 4 s s.. i 68: u o, n 7 -p s s 7 r 'i 't f .j (f* 3 _, out.95fpercentEvsrylquickly. LI:mean if we can dism'iss<it!'

I

..s n .n,j(E.c 1 .v. 1 =E. l2l Twe'lljput it away. NotEtotally. We cat.egorize U[t..' It[goes. n .2 <4.,

3"

into the datajbase.,'But we[refnot; going to spend:arlot'~of

[.\\ '~1 4 time'on it. 'Welthink there are.more important'thingsoto-look w. 3 LS 'a1.;~~Tden'we look ateabbutJ150-a' Year which are significant-U 6;

i. bu t '. then1 we{askLourselves who:elselmight'be lookingeatiitJinJ 7:

the"; Agency?)- LIf t NRR ~ has got 'a f pro' gram - thatiis reasonablyJ sr .,1 i 8- ..followi'ngjthat kind"of an-effort,.we!11 make a decision there 9\\ toidrawfup--to provide the'information and to make sure they. i-y 10

know what's going.'on.

But not start anythingCup.ourselves.- ~ . 11 i ' So we' really. go 'down Ethat order in a: sense tof then' ~ i '12 . decide where"do"we'. feel.we can make.an impact' from-a' piece 13)- Jof/workithattwe.do'. It's a'value impact. ,We want to get ~ .- t

14 a returnLon'our. investment of resources...We want,to-take on.

i 15' the)stu'dy and'then hopefully find something'that's worthy.ofi ,16 sharing it'with.the~ staff and' industry. 17 DR.' LEWIS: Well,.if'I can muddy the waters even 18= a little more than they are already. muddied. One criterion 19 you mentioned was something happens that you don't expect 20 and you want to know why it happened. But there's the con-21 verse of that kind. Sherlock Holmes once solved a mystery j l 22 'because the dog didn't bark. And I wonder if you are alert 23 to non-events. That is to say, we're carrying--let me invent i I 24 - something. We are carrying a failure rate 10 to the minus 25 '3 per activation for something or other and somehow it p ~ Heritage Reporting Corporation m m.= +

69

'4 doesn't,seem to have.ever happened. Earthquakes,,for ex-I 'j 2' ample. That's the dog.not barking. And yet it's,signifi-3 cant.in the. sense that! you nave ' earned something but it-l { ); V 4 hasn't threatened the' safety of the plant. In fact, that's 5 the information. 6 Is there any mechanism for going back and looking-7 for stuff'like that? Or is it really only one sided? No-8' body ever asks that question. I 9 MR. JORDAN: It's really one sided. You know, from 10 time.to time people would postulate a certain failure or 11 . event or something like that. And we'll inevitably go to the 12 data base and if you don't find any events of that nature, n[f 13 we'll say, well, then isn't that a concern from an opera-3 14 tional standpoint._ The data doesn't support it. 15 DR. LEWIS: We obvious).y need a subcommittee 16 'on inexperience. 17 MR. EBERSOLE: A case in point is water overfill. 18 You know, it's one of our' tragedies that we think might 19 happen. It hasn't occurred enough to--four times I think 20 is all it's ever occurred. 21 MR. MICHELSON: That's a very good comment. They 1 L f3 22 just didn't think about-- f 'wi 23 MR. NOVAK: I guess the--we have enough to do f] 24 looking at the events that happen. And that's true. The 25 things that you are talking about are certainly worthy of I Heritage Reporting Corporation o.n u.a.u I

a p;k( y~)Y ~' or lV 'c C 1 l ?. +i' a ' b,: ya 1[ QQ 1,.;(;q{ l > y [Qu 4-s wa .a Ep ,.70s T, &g, fy ',, ' 0lM. 3'Nk}j E J f. i x. s w' w, a o ;: p a. a y N.j,m g' q .I' tt11'nking ? a' bout.. l '. W:,r ' M .A, i u p s W e. _t . Q,, 2 't mi]ssfi. j; .Myp, primary; goal isnto mako.sure;that..I!M. y 2 g7 y F 2.; pg j s a,. t, .9

9 4

i - <eam 0 J y'm , 33 c o , ?p m _ c~,}:;n Tf,. j Z ']; ,MW " ? somettling.-,- '..-4= " g 'O

s n

3f " 3 DR.-LEWIS: .I wish:you hadn't~madelthis11ast'com-T ~ ' W4 w. wu w'l. p ~ , 3. 4 -t-: 1 finenti 3 ( (l f ( j-f butIam' going]toiletitgoF k... \\ i g ',a( ( ~, T .o (4 i c' 4 .I1 r However,Ehave you thought of the ne$tL i t g l@ MR. NOVAK: H +q ? S jbiglacc'identJthats going to' happen y.: N, J y 6i ss cIchave a'little different.qu'stion ' i /MR.-MICHELSON: n e m- <g y M' . ;h y-along theisame-line?though and.that is the original'missioni L .e 10 W, of AEODLwa's to-.analyzeTand evaluate operational $ experience. H o At' shat? time'itJdid not get involved.in the day.to-dayj,M - 11 f, < /.. : (:::n,, o 1 .'J.'

(

'gg-s ' 51' 3 Itwaited.ulltilthingssettled.downaliftie' ' /j + ' happenings. 33 . ~ U(s t are'happeni'g~.is rather manpowerfintensive.' I's it necessary-~ l n py c E .do you.think in'the whole process $s.it necessary'to i o dhi I ~ h n-18 the. day-to-day happenings or do you--can'you.still analyze. o ' I I-your operating. experience by starting the LER. What dd^ 9 20-you gain by looking'at the day to day 4-what. additional .y3 q. i v, things are being added then that would redi'ly enhance the ) ' I .y i 22 ( process? I'm thinking, you know, if sorgdday Congress does .. j

23
? decide to have a separate safety board, could~they start m,...,,, i 24: adequately with just LERs, or do they really need'to be i 25 9 . involved in your day-to-day operations? Heritage Reporting Corporation (202) 674 4440 m LL^2 _ _ ~.. -_-.s
-= P f %y N-f kt ,q d D l ^ y W ; $ -l hg % m. do w n @Mgm i 3; <
>e l.
g w;f - s' " 71 ? ?L: Nikpf$ # I s p,djW..- M y-r1 ,$,t... t qe e
n..
!j e( y;p am .sf, (J e u. ~ Usk. t <LTl~n; il t 9{ L E '.) g" # ~ MR.iJORDAN q ';)I'l'1[ answer (that.- 'There:isTaineed ,qj ff;' y y yq:t ~ ~ k p~$rithe A$encyftol.use SilNitsdesources as iwisely;asEpossi-WM,s c2- ~ i.-4 m ,n. )f. _\\I, ... Mi y . n. -:.#.hl'an, d maybe something < we lhaUen ' t: said L a's. clearly as ; we " j.;g iC.,. 3' .,j s i ,Wb1 4 i' j
m. ( s.,
m:WM'
?c qk b
1 ,u, pNppbg."c ' "shoulbis that the: responsibility"foriimmSdiate follow.up of-74-
%n&
=ipu. J Q ty 3 -M- -5? an:eventiinLwith<theiregionaliofficel'and with NRR. And:: they. t y ,f vt- '$m _ x9 %! 4:; t 3 :j pM[fx.- h 'ido,th$ assa Way of ife. lAnd so when a.II failure" occurs' ~ ^ .- g 7. A ,jat Morth Anna,ithen the: region.haslgot-people on the~way.Last . n' h,,T (7; 4 t,j '81 Jsoon as'theylhear about it and they'll belfollowingLup.and. '.?y m m 1 1 j f . i9 a;they!llfbeLasking.'for.supportLfromiNRR; c ~

n,
~ w 10 The role that,"AEODiplays in those kinds ~of. things: r a-g ...[ h '. ' ' 11.. fisjonii fee'dingobac'k of what~we'veflearned in.the past. And Q" e- + .tn l ig:y ~ ,y:4 7 !f providing input'fromLthe~datasbase..The?analsyis that; occurs, L
4 i%
yc a. j ~ c Q 13Y ?for/ instance, in-tr ' morning calls'Is-the' call. runs thi'tp.) r y' W L:m. 114 'minutesynow?-.. ThaQ i's?a' feedback.procdss.that is-working qq-i: ~ N 15 both wa r/p 1. Y.. tr 1.6 i MR.?EBERSOLE: On individual' events.- -y '17 E MR. JORDAN: On the. events:that happened'last ] 4' w 'y I. f,C ) 4f,Jp (18 ' night,'the discussion this. morning will say that, well,' gee, R s a qq, n - that aix4 system failure is very similar to something I recall ] 19 i y l ,146: 'last.. week? and the synergism among these people is marvelous. [ M Eg Wp m' .o 'y 21 If you haven't listened in on the phone call, I would ask i 9: "b fd,J: ,M 27 'that!ytau consider doing that. e,, p~' .1 ~%. 3 j MR. MICHELSON: Well, I have in the past. The t. } ) (24 regional office can benefit considerably, but the question 25 really is pitched for you to do the job of analyzing operating l t:< O.. I he ' Heritage Reporting Corporation g4; m *aa py-c(
gn y w - g.; a F O l ' '7., '^ v. i '.O. 4 I,: 'y .lj ; 'g"/ i n;4..{f y, y 3.! [g]"jQ ,? ) y m 4 oc
m h
n3 k 's: b fl. ehhlhncing,Ul' earning lthcheigher ;1evel,i crha tingith'e : hi@her.? f y Q S. ile,ve Lkisds'of" final < recommendations and so)ffortN -which/ iso n.x 2 y - ;g ,. {: c g. '} -(yourmaing.ll'j f + <put' 7:%d y,r how m' chidoesq the ' ay-to-day opera- ~ .j 4 A.y 3 ' b u &}jyz ^W <W l j jn FW F d y ;dn, help' yo{ h 'arr^1@ihg st thosel;kindh (of firiall higher-7 (i', f ti di . l ,+ y- } 1e,vdl, conclusion's[Mslitiancessentialingredient,for.(isLit: -d '5 ,y 5 [$j%;. I .A A OD ' T C ihsN- %heping(inlaind" the' amountL o'f time fit takhs. you to q~ s sj no ...~.a p' a
7 pdre.{cipAhelhTadallybasis'.thdtacouldbe.dev6tedto)doingL QF p 1
,M
/
y,., .. g 'm'oreanalysbshndSv,auahion:asfopposed;to. worrying;about--L fI I .8i ,u t-hnking { about the :[ddy--to-day l events ?' I '3 9 q .. i lm' -10 E'1 'MR. JORDAN: JIn;termsiof the' responsibility /for--c J 4.,. m W Unil ,t. hat's1the point 0Iiwasltrying to make. abo.ut.who is responsi-- +l 'c riti pp Q- [ A Ap': , f 'AEODJ 1 ,,7 ; e, vents;.-is not AEOD. 1 . ble 4. for, the. day-to-day ? rev ' ewi 4, 4 12
1. (
T1 (i.'s?t recipient'of th i for , Ion'.ank we're-the transmitter' [There'{a f- - g! lY t ,o e d /, f f.T 's -) t o o 1,I. thought'your people were-15? MR.'fMICHELSON: I A 16 setting /up the' calls'and'so-forhh each. day,.aren't'you?.
n 4.
d 17 And it ' ayk con on.through your. Incident. Center. 'h/ q c 18-
Md. JORDAKt The calls come.to

thel Incident Center.
1 4 g 19. Those people were added1to the existing 'staf'f so that'sDnot 8 ;f 20 -:a dilution of' thh ef fort!. Its rea,l'ly an addition'to it.' f< y3} .) / so that the peojdMdat are from the previous AEOD staff [ 21 )
/l1 22 that;are listening apd participating in that morning call..

q>
3 in' addition to the OMCenter people are feeding back.the- ~ 23 '
Dy.
L h. 24 ' results of'what's been done in the past so that somebody 25 doesn't start down a wroncJt athway and they then are n guy).!; j. t
rg 5
) g/ Heritage Reporyrig Corporation x + y' ' m;t, v%y + r i.. ..c q g
[ I I'- '[.- ) Y 73i h,s d i c, y-a s -I + ym t i + .1 'k. . l triggered'intosself' starting onan issue more promptly.-th'an y y(y vn b 4 a c s 4 v.~ l2.
waiting-thirty' days or waitincj.) for; an analysis--
m ~ 1 ,3g j. ljR.JMICHELSON: I cjuess I di'dn ' t' mak.e ' my-- bi -3l m ,,. j. w 41 ,r MR ~. JORDAN: LSo mygviewjisithatthererisa-very '"y s w'3 8 , 4
5.
. strong: synergism,ja very beneficial.effect'of'that joint. s,= s + + 1 4 . )' l 6.. t. m A 6 MR.';MICHELSON: I' guess'I didn't make the; thrust
I 8
i< c7
1 adequately clear.
ILwas'mainlyfinterested.in~ order 1to:'ana-j _1 lyze and' evaluate operational' experience,:is'it necessary Vd 1 81 o q 91 . to; participate. in.the lday-to -d' y experience. which is a_ new. 'l a 4. ,3 N',, 10: cadditionJof:the last year,.the.:last several months in-fact? m s '11 ,It was-done;without it in the past .Now you:arel.doing-it J 12) with it. Howfmuch do you,thi'nE it_ helps the' process? J d (A l a /- 13- - MR. NOVAK: I think.you could do the job without; l '14' it, but--- I t 15-MR.. JORDAN: I.'d like to finish that answer.- a 16 MR..NOVAK: Jack can explain the answer--- v 17 MR..MICHELSON: Your boss wants it first. 18 MR. JORDAN: .I may change my mind. 1 19 MR. MICHELSON: You understand the thrust of my-20 question? I'm really trying to figure out is this'really_ 21' a helpful additional wrinkle? Or what's your experience? I 1 22 MR. JORDAN:- In my view it is a helpful wrinkle. i 4 Oj j .There.is a tendency for reviewers of data only to build 23 24 themselves into a tower that they don't really keep up with 25. -what's happening day to day and understand the pressures of Heritage Reporting Corporation (202) 426 4sta L O _-_.-_-_ _.. _ ~ _ - l
.j 74 1 plants and thetfrequency with which these kinds of events
)
2 occur, so there is a realism or reality of getting involved 3 and. staying closely in tune with the other people in the. l f') i J' / '4 Agency'that arefdoing-'similar. work. That there is an inter-l
S change'that I couldn't possibly evaluate.
It is so bene-l 6 ficial. And I believe that if you'took a poll in the Agency i 7 you would find the remarkable value that people feel out of 8 those interchanges. 9 MR. MICHELSON: In terms of your final product-- I 10 DR. LEWIS: Are the conference calls' taped? 3 11 MR. JORDAN: They are on the tape but they are 12 not maintained long. Because it's done through the Opera- )
3
( j/ 13 tions' Center and it's automatically taped. 14 DR. LEWIS: And it's erased everyday? 15 MR. JORDAN: And so it's erased after a week or 16 so. 17 MR. MICHELSON: So you think that what you are 18 learning from the day-to-day operations is really enhancing 19 the quality of your case studies and so forth? j 20 MR. JORDAN: What we're jointly transferring be-l 21 tween the program office that's responsible on a day-to-day 22 basis and AEOD, I feel is of great value. . g w] N. 23 MR. MICHELSON: For your work? () 24 MR. JORDAN: For our work as well as for their 25 work. Heritogo Reporting Corporation on).u.
Qy q ] c g,> c
75'
,q 3 'i V 'T 5 l 2 to-get a, feel for'how'much'it's'adddd to your: product.- 1 E.g ,,y j. Q ' 3 ;- ll-l;( -... MR.jJORDAN: I think that's a point I want tosmake. q 1,. ,'.i f4 We're.real'1y"acpartyofcthe NRC,andfso"it's an overall NRC l-8 - t ,e n, j r 5 i reffort'and if by>doing things'in a cooperativeifashion'we v e ,m. -, ?, I
6" i get a greater prodtIct outi "then ~that's--
l [ [7-MR. MICHELSON:: There is'the-otheriside'of the d a H 8-- . coin that.might be! argued'and.that is if you get.too.;ihti~-' l U:. 9J lmately. involved too :early,you lose your. perspective? to l'ater 4 -10' sit back and really look and see what happened. .You-develop j 1 11 too many predetermined. conclusions too early.in the. game and-o .12. you' kind ofilose your ability to give a fresh: view. ~ ~ WK That's why'I'm trying.to make the \\ / 13 MR. JORDAN: 3 14 ' point that the successive levels of review don't--the fact j -l 15 fthat~you. don't start a review based on a morning call'-doesn't 16' preclude you from starting-one' based on the licensee and'that 17 isLa different person; generally than is reviewing in'the . 1 'i 18L tail of that particular event. 19 MR. MICHELSON: The question was though if you 20 - start with that first morning call.and you start getting-j j21 engrossed along the way into the details, the core, the whole Ng 22 thing is congealed, sometimes it involves.thi'nking to take-ft 23 a fresh.look at.the whole perspective problem. I just wanted .24 to know how you viewed it. Your feeling is that you want to - 25, get involved early and in detail? Heritage Reporting Corporation omu.an
W'h = rs s n ~ b, ' '}}.,_., a g%. 2 W > 7.6' g,,;p.- f yie 'f..,t.L..
1
'MR.UJORDAN: Well~, early in the'necessary detail.- .j ~ 3p:~ Lej se m. 4_.x ,2
vourknow,.if"NRR is going-;to fol-low;a particular: event, 4
4 ~ l
-- 4
) k 3: tliereis i no. point. in us foll'owing it, and so -wel would support l l thembyLdobigi.thksearch'ing'tha[t"wouldtNe'n'contributeito. U d, c.4 ' ./ d i C.4 7 5l -the s6lutionE,of that; problem, and then.we're out of it-Tand t[ l _ t '
y
[ k 6
we.doosomething"else.3 f f f (,'
-7' MR'. L. MICHELSON : Okay.: ILthink'I understand.' '( s l8 .MR.' JORDAN: So there-is a'relativo partnership. 9, that 1stworking.:very.well. j g l, '10-MR.ENOVAK: Okay. I:think we have' finished' dis- 'II cussing;on is~ sues. q '12" MR. JORDAN: Now,.we're an; hour behind. I started. s, ' 13., 'you ten minutes =behind. 14 'MR.' NOVAK: lI've got a lot of help'in~ answering-15 sometof these. questions. !.But I think it's an important j 1 j '16 point.and'my personal belief on are we concentrating on. fine n 17 ' l issues. IJ1ike-to believe I'm never sure. That's why we i 18 keep asking:ourselves are we looking? We do tr.y and re-l 19 assess the priorities, the studies. Are they leading us to-120 ' -worthwhile results in ' terms of information, or are we just '21 g spending a lot more time on them. 22 MR. MICHELSON: On that question, let me ask one L 23 other. I,believe that water hammer, for instance, got a '24 fair look by the Agency sometime back as there was a generic 25 or unresolved safety issue. And then a decision was made Heritage Reporting Corporation (202) 626 4sse r
7 g p. q.
7,7.
, y; .i. X'~
I-for a? resolution was reached--whenever you-see a-water hammer,.
[-L 72a Edo.youLlook at it as lf the Agency--as.if it..were.a fresh i q yl 4 l'
1 y.
., l 3. problem'andinot'thatrthe-Agency has"some predetermined notions .. ~.. % q t W-4f on it?'cIn.other/words, how independent'are you in digging ? ?
5
-into problems / tlia t" the - gency'lias 'alr' eddy tried to put. to 1 64 ' bed? '. Technically and politically ;they've tried to put it to 4 m 7~ bed. 1 8 MR.-ROSENTHAL: We take fresh ~looks. We-take ' freshL looks.' ! And 'when ' NAD.. tried ' to ' challenge. the regulatory ! 9? jl ' 10 - basis,,that's not to say thatLwe're looking now at water 11 hammer--Ifjust wanted to get in'.a' word about--we've got'a 2 [' 12' study lnowLon~certain~ breakers in Atlas-..because we don'.t .] W; i -h/ .13 think'those-breakers are performing as well as expected. ' 3; 14 Check valves--the' Agency is doing'relatively.little-on check .i 15-valves. What we would like to see is ultimately when the. 16
data shows-~a decline in failure: rate of check valves, and
.] 17 if not-we'll step ~in again'and work on HPAC.- We'll likely' i 18-challenge Appendix R. I think we are challenging the regu-19 .-latory. basis. 20; Now' ultimately I may choose not to work on a task 21 ~ that'is not in conformance with let's say an instrument draft p L22 on something or a draft that I don't perceive it to be-- 23 it works both ways. 24~ MR. EBERSOLE: Jack, you mentioned recyc trip. 25 ' I would hardly think you would look at that. In preference to i Heritage Reporting Corporation uen nu i
e. n o. ....... y n "3 - jqt, %pn:4 Al, i ~ 7l8) o
,'; J. -
( g.. s ? %,i 1 'l' b
. '..p,
'ill PWR breaker! trips':vhich.has-to do their thing onlevery L2L .MR.'ROSENTHAL:.EA-recyc trip on a. boiler has'to ^
e i
'L. 3.' ' perform onlyYif-lthe" scrim systems fail. And.that's a-rare pN 4 " chS1'lenge.,,'Bdt.these,other things.havento hit.it everyday.. ^ J., -y w 9 q s t' '5 iDo-you. follow m'e?e X 6- .MRT EBERSOLE:
Ye s. ?
s 4 ,d; a '7 'MR.' ROSENTHAL: The challenge. frequency.is so: j o, y 8 low?for those functions. Whereas it's very high for the y 1 4 r 9, ' breaker trips 1for the rods ~on PWRs..:And you know McGuire andall2[hatstuff1that's,comingout. (zy 10: '.11 ' . MR. ~.EBERSOLE : I! just wondered why you-mentioned .12 .thatLparticular' aspect. it pb ?[ .I ' 13 MR. ROSENTHAL: We're looking'at.that'because we R 14-care concerned thatithose; breakers which were.not required to l'5 receive ~the flow are'in fact improperly servicedLand main-16- ' tained.' 17. 'MR. EBERSOLE: Yes. n K 18 MR. ROSENTHAL:. Now,~the other areas of the NRC 19 are ' paying a lot of attention to the PWR breakers and so 20 we'wouldn't. ' 21 MR. EBERSOLE: Originally they were unclassified-o, 22 breakers. ,V 23 MR. ROSENTHAL: ' Bear in mind, we are well tuned 24 into what NRR is doing. We are well tuned in what I think 25 to what RAS is doing. And we won't-- and we attempt not to Heritage Reporting Corporofion (te2} 624 4ses ( C-.. -- r 1_.__ u- .J
n.g g'i m ..' ~ ' ' I ', t j.$ .g.' i ,4 3 s .,m... i
n p
.79' 'M] T;- g" 4 ' ~ $,.o 4 6 ,v_ n-4& M y. l >, ^ "duhlicate efforts. "]. j. q g. 7 ; ' l2; .DR' : LEWIS :: Our polite' member, Mr. Ward, had~a .R: ) ?3:
question..
( 1r 7, ,j
m
~ f y, ir . Qi 1. ? ' (4 MR. WARD: Let me.go'down and just try.;'to. keep'it 1 n k .,!brief,.a;litle-;tanyekhaNotkter[hamm'e s and maybeTsort' SJ ( a, '6-p of a probef to, see whether we can' learn certain 's' ort's ;of 1 ,'[3, ' y _ v. O. } ( 70 things from !operatiing' esperience information base. -l 1 ,a ,'08 'There's.a particular point"about-the actual' nature l m q, ' li': That a
y<
-\\ . ~ .. }
of water hammercevents.which I'd like tocunderstan'd..

9 o
J ?. .f. e i 10 a' number'of: people have raised-the/ question that maybe:the , 11 . water?hammerEis,more safety significant'than we'veLbeen-12. trea'tiing and' we ought : to; learn more about -it, this, that: or ' i ...k.. !!3 .the other. thing. p. =14-But.I guess it seems_to me that water. hammer.may. ,n is - be'important,. uniquely important, or may not be. depending on ~ wheth'r'it's a type of' phenomena-that tends to affect-- 16' e 1' commonly?. affect-redundant. systems in a nuclear power: plant. .17 1 a 18 .I mean we really keep plants safe by having a redundant. 19 system'. That's sort of.the core' requirement. And,'you-know, -.i 20 ' seismic events are bad news'because they tend to' ignore the 21 fact that we got redundancy-there and take them all out. c - 22 And I've asked the systems kind of guys, the thermohydrau-p 7: 23 lics guys, what-they can say about that and they kind of said, (} 24 well',.maybe they'll come back to us some day, but is there 25 something in the operating information base-- ,v: Heritage Reporting Corporation m us.au
==_ 2 2.
]n'W a y -[ q 5 w. _ 'l e ~ ~ p?.py % j --n s s j - a Q
80 d
' M, g j 4 a t y
[y 2,,
-i ?
y
' q & ;.
l1 MR. ROSENTHAL: You know,"we5 worried about mode- . p r.. -2: Lsteam bonding of auxilliary feed pumps.. 'Be they. motor'or-- i wm
3
' and one can end;up with--we 've sesn" data wh'ere you have 1 -4 ~ 1 +' a -( 4-leaking check; valves and you, end up putting ' steam; back into ' x ~ .j 1 + e > 3 g.comon h$ader /annsb' tiiSts an e$amplel of-- fl "~ u
n o,
,s-a.. 'b. l ? 6., Land the'le'aking'checkival'ves and steam-migration. W m, Lsurelythas proximity t'o1watsr hammer type' issues.L .7 , au .8: MR.LWARD: ' But';if we were. to ask-- X w ,t MR.- ROSENTHAL: Andiwe are working-'on'that'one. 9. m m ') .t
10i lMR.' WARD:'
L If we ~.were ask. to. ask the particulai-j question.is thel experience.:from a thousand reactor! years j 11: w
12
,theoryfever'got--where'there.are water-hammer events,'do they 7 [ 13 or do.they.fnot. tend to involve more than one' system.or a
j p
.i gl 114 redundant set?. Is that: scrt of' question answerable?- 1 11 15J
MR..NOVAK:
'Well, we can' answer it from the data. e -16.- See,.andlI think you've got a good point. We focus on what [ 17: happened, what was reported, and if you have a~ water hammer i '18~ and you'had a bunch of valves fail because of the water -l o q -19 hammer' it'll all be reported. If only.one valve failed, -l s 1 l ' 20! that's all that would be reported. So you can worry about l 4
21' the configuration.
3.- l ': ~.22 We'do not take that next step to see how close i i ' F- '1 't h .23 were we to a common mode. In other words, the dynamic re- ) 1 l.3 :}. r 24 sponse, the analysis that would go into it, we are not pre-25 pared to do that. We don't use those tools day in and day Heritage Reporting Corporation l~ <mna l L" .k
81 I out.- We are seriously thinking about that as something we 3 1 \\> 2 should be doing. That is coupling the actual behavior with 3 going and looking at the plant performance to see how close + 1^ J. 4 we could have comt to something more significant, if you 4 5 . follow me.' s 6 But currently the way we do business now is to 7 look at more or:le'ss.the factual'results and just use our' 8 judgement to take the next step. 9 MR. WARD: What I'm really asking is do water 10 hammers tend to affect redundant sets or don't tend to? 11 MR. JORDAN: I think from the previous study that l
\\
12 they didn't tend to-- ../N ( ) 13 MR. ROSENTHAL: I think if it relieves itself 14 from the first failure, you are probably all right. 15 MR. WARD: If that's the law of nature, that's 16 good news I guess. You need a weak link I guess. 17 MR.'ROSENTHAL: Let me answer on the high or low 18 because I just don't remember all the facts. And that is that. 19 AEOD did do a study on water hammer events. Eric Leeds wrote 20 a paper on that which he'll be giving. And that information 21 was used in the resolution of those merit issues on the ?~% 22 water hammer. \\_J - 23 MR. MICHELSON: The real concern is if you lose n 24 j the presssure boundary as a result of the water hammer. Now i 25 you can involve other equipment, but fortunately water hammers Heritage Reporting Corporation cm ete.au 1
.,1 4 i a 80 4,q 6o, g; q u,-, d [t. so-far( generally; haven' tL ruptured pipes ' or_ with few excep-' h: . e' e e 2 tions'. .But.if they do: rupture.the(pipe, that's when they'll' y e get-intio the otherf train.of equipment because their. physical; r 3' a i .A ? t ry - -4. Kseparation;is'often-- ~ l .g y 3 .l p s
5. '
'MR.fWARD:; Yes,jbut(it goes.beyond'that. I mean y . 4 (- 'm xw four.. systems 1 are 'g. supposed-to accommodate a failure of the' 'l 4 6 +u- .: 7 ' .pressurervslve. ;I "mean th'at's whyp hd have a ECCS' systems. j .y y8' DR.' LEWIS: 'Well, given a choice, you'd'rather-not q 'l 9. ,have it h'ppen. .] a i.$ MR.! WARD:- Well,.no', but I want tio'know~if it's .e -10 i 11-something special. -I_want to know if there's a. fundamental-12 surprise' lurking.there. y g i
13 DR.x LEWIS

I. understand. I understand. g(/: .i '14 MR. WARD: You-know, a break in one place isn't ,s .15 news. That's dog. bites man. 14 MR. MICHELSON: No, but that's not a design basis. 4 \\ 17-because we haven't designed our plants'that'we just break i ~ f 18 pipes'everywhere. It's been rather specific about where they 7 - 19' can break and we haven't designed for breaks in other loca- ~ 20' L. tion s. If water hammers could cause breaks in other loca- .21 tions, they;could be potential-- 4j 22 MR.. WARD: Well, to me that's another--it's not ~ e. 23-just compromise or redundancy. 24 MR. MICHELSON: That's why they get more than one 25 type of equipment. ) i. Heritage Reporting Corporoflon j m m.m.
p.-

I, r
' 1 - f Mj j'6 1 't-t (;, ~ r - 8 3 ^j g Q ,o' cf m'fE I? MR. JORDAN:.Maybe the right answer'is to.sayjthat'.
k
~\\ % + j p. G-2 Lwe've nevericlosedfour books on'an issue-andEthat the data o 1 '3; ~ remains and.the nextiwater hammer event that captures;our j p f[y - 4 b fancywillcause)us,-perhapsItd(golokback'atall;thatdata ,i Si again,-tofseefif.there'is.now'a common thread or a further ? 2 'g 1, 1), / 4 i Q, -.6 threat that weioughtitoipursue;andDI think that's the beauty-n a u, 7- .of therprocess. 2 8,- iMR.1MICHELSON: You.'ve kind of. assured me that-9
jg
.pou'arefreally?not; influenced'.by.the= fact that-the issue:has'- ' 10 ' Lbeen resolved. W
.,. 'MR. JORDAN:. No.
In fact, that's an important^ ..] 12 point. fOne-of8 our roles is to assure.that' issues.in" fact ~' 13' .are. resolved. -We got. tasked recently by the EDO with re-j - 14' zgards toJa' technical" specification change to provide an ~ l 15L
evaluation of<whether or.not'in fact that was' justified 1
16"
based oneongoing experience..And Ilthink~that's an-important u
17 .. thing that we have to do. Is make -: sure a decision that we ~ 18 make today with the best information available really is ~i 19 ' sustained by'the operating experience. 20 MR. MICHELSON: And you've got to make sure that 21 new experience doesn'ti indicate that an old resolution is t hxp jH:l 22 invalid and that you need to reopen the-issue. I (L3. h-d 23 DR.' LEWIS: In either direction. g 24 With the power vested in me, by golly, we're going, . 25 to have a fifteen-minute break. (Whereupon, there was a short recess.) Heritage Roporting Corporation mu o
h-A I b!b 1 DR.-LEWIS: All right, let's go. You were in mid I . ) 2 sentence, I believe,. Tom. 3 MR. NOVAK: .We're going to pick up now talking about c.,. $_). 4 the communication paths. 5-In Question 5 you ask: What are the communication 6 paths between AEOD and other offices within the NRC, for 7 example: between them and all our research entities and so 8 forth. 9 How are good ideas dissin 'nated, for example, a promis-10 ing area for research, the need for a new regulation, correct ~ .11 ing the poor regulation, how formal are the paths. 12 I think you have recognized that we do have a number o ,. m 13 paths that we can use. We have the daily communications that (v). 14 we ta,1ked about earlier. There are weekly events briefings 15 that we sit in with. NRR goes over the events of interest. 16 That's another area for us to communicate. 17 I would say generally though, we try to write out 18 thoughts down. If we've got an idea and we think it's worthy, 19 we ought to take the time and I think it's a good discipline. 20 I prefer to see the persons views written down so that we can 21 support them. I think it's very difficult for me to support a 22 view if I'm trying to interpret his view. I would rather have ,s ] 23 him write it down and then we can sit down and understand it and we can support him. (~]' 24 w 25 So, generally, our ways of communicating are to write Acme Reporting Company 42021 628 4888
k V 1 memos, do a report. Certainly our reports are quality p o (q,) 2 ducts. A number of times since I have been in AEOD, I have 3 heard, certainly AEOD before the reorganization and hopefully ,m i: ') ,g through this reorganization, had a quality of reports that'has 5 stood up and I think that this is useful in the sense that if l 6 you do the job right, if you write a good report, you are-communicating and you are getting your thoughts across and we 7 8-do strive.for that. D. DR. LEWIS: The next step is to get someone to read 10 - them. MR. NOVAK: There's another point and I'm glad you gy got--I.got a point here to talk about that and we are trying 12 to improve on getting people to read them. 13 v I'll give you an example. Yesterday we met with the g 1 cal nuclear steam supply representatives in Washington. 15 There's one for each of the main steam generators and a couple 16 of their manufacturers. And you really need to know how do 17 they look'at our reports, what are their priorities and it's 18 g enlightening because they are operating as a service company today. Most of those people operate when they have a job to 20 do, they'll read the material, but if they don't have a job 21 to do, in other words, if they don't have a contract or some 92 ! (^ ~ thing to charge against, they're not reading a lot of materia:_ 23 because it's really not related to something they're getting f3 ,y ~ NJ paid t do and it's important for us to recognize that, that 25 Acme Reporting Company 12Okt 628 4886
,Y j 00 we have to really sharpen up our reports. l 1 x._) 2 In other words, take the time to get a good punch 3. line to your reports so they'll take a few minutes.to read 1, 4 it. They may not' read the whole report,'but.they*11 get enough 5 of it to know what we're trying convince them from what we have 6 learned and I think these discussions are useful. 7 I'm meeting with the owners groups starting the first 8 of the year. We need to have this communication. People 9 are telling us that there is a lot of written material, but 10 there aren't that many people available to reai them. There's li a lot more material than poeple and we we're face with really 12 trying to structure a good report, something that's going to (qj 13 be picked up by the right people and read and that is a .g 14 challenge. I 15 DR. LEWIS: Cne of our great interests, certainly 16 for me, one of our abiding interests is what influence does 17 AEOD have on NRC and on reactor safety and writing a 18 report,. getting something to read is one channel through which one exerts influence. Hopefully there are others and I hope 19 20 we will get into that in some substance at some point. 21 MR. NOVAK: Okay. 22 MR. MICHELSON: Clarification on something that you ,_+) v 23 said, Tom. You said you asked the engineers to write ' ("4 24 these concerns down somewhere. s_) Is that even before these technical reviews and so 25 Acme Reporting Company (2021 620 4886
L ( L 1 1 \\ CYs (. I forth forth, kind of written documents or is that what you ) L r s,, i m) 2 had in mind? 3 MR. NOVAK: I think they first think out a'problen. 'J 4 I think Jack can speah to it. l 5 MR. ROSENTHAL: Let me say that we-recently instituted 6 a Watch List--we have a Watch ' List we keep up to date as of 7 1 month and we have a-- 8 DR. LEWIS: By a watch list, you mean these are 9 the things you're watching for? 10 MR. ROSENTHAL: Right. Then we have a report that's 11 roughly monthly on what we're working on and we have asked 12 people now to put down abstracts, to clearly define what /m (Vl' 13 is the problem, what are they looking for as a means of 14 focusing their studies. 15 MR. MICHELSON: These are just internal documents? 16 MR. ROSENTHAL: That's an internal document and then 17 bi-monthly, we'll issue that to the Office Directors so 18 that they will know what we're working on. 19 MR. MICHELSON: I haven't seen one of those yet. 20-MR. NOVAK: I just concurred-- 21 MR. MICHELSON: Oh, you just started to do it--okay. 22 DR. LEWIS: You know the next thing is that we will v 23 ask for a copy. /'N 24 MR. MICHELSON: Send us a copy \\) 25 DR. LEWIS: When you ask people what their-- Acme Reporting Company 42026 628 4888
y i ds l 1 'MR. MICHELSON: He was going to tell us the rest 1 2 'of the price l'ist. 3 DR. LEWIS: I'm sorry. ,~. MR. ROSENTHAL: So early on, there is an emphasis (j-4 5 that developes,; hey, what is the issue and what are we looking and then let's even try to identify potential resolution and g 'then to. write,a'short abstract and over time that could-- 7 MR. MICHELSON: The engineering evaluations though, 8 i are not a part of the process you just described? 9 MR. ROSENTHAL: No. An engineering evaluation would 10 ultimately come out of-- gi MR. MICHELSON: --a particular item when you expand 12 ^ upon it enough to even write on it? .f ) 13 v/ MR. ROSENTHAL: Right. 74 MR. MICHELSON: And you're still going to do that 15 process? gg MR. ROSENTHAL: Yes, we still do TR's and these case 37 studies. We would like--I, for one, as the manager and I know 18 my boss is going to what to know what we're spending our time gg n and are we headed down a proper path, and not throw 6 20 months of time into something. 21 22 MR. MICHELSON: Okay. ? 23 DR. LEWIS: My concern was just that, that sometimes the best way to find out what people should know is not to ./O 24 V 25 ask them what they're looking for because the best service i f l Acme Reporting Company a w .2.....
(y, ,,i dS I anyone can provide anyone,. advisory ' committees too, is to tell 7 (/
2 them= things they don't want to know that they should know so 0-3 acceptability of the product is sometimes a negative indicator
(~) l ( /. 4 'of its value and.which leads me to wonder how obnoxious you 5 and be and still survive. You know what I am asking. 6 MR ROSENTHAL: Yes. There was an Air Systems Study 7 that was started when Mr. Michelson was head of AEOD and 8 ultimately went out and ultimately went out and prioritized 9 this whole issue. 10 That study was worked and it must have been 3 or 4 11 documents issued on'the air system and there is just a 12 recent 1987 Air Systems Report and I think that's a good n(p) 13 example of being tenacious, going back and hounding that 14 you thought that those air systems were bought and were proper ly 15 maintained. It's a good example. 16 MR. JORDAN: Correct. 17 DR. LEWIS: AEOD is an office, of course, so you 18 report to the Commission through the EDO. 19 MR. JORDAN: That's correct. 20 MR. LEWIS: Not that he would, you understand, but 21 that means that he can, in fact, block your channels, 22 the Commission, if he wants to. s, V 23 MR. MICHELSON: He can cut off your water if he wants /7' 24 to. O 25 MR. LEWIS: I'm doing it right now. Acme Reporting Company 12021 628 48Rs
w-f 8U MR.-JORDAN: It's hypothetical. In reality, I 1 4 , 'T t,/ 2 get a great deal of support from EDO? 3 MR. LEWIS: I know that, I know that, and that's why, (/ 4 I say, he wouldn't do it. I'm thinking of the structure, 5 you know, EDO's come and go. 6 MR. JORDAN: So do Office Directors. 7 MR. LEWIS: And even Office Directors come and go 8 and there are some whose communications channels should be 9 cut off, hypothetically, of-course. 10 I'm thinking only of the structure. The structure 11 is that the EDO does sit between you and the Commission. 12 MR. JORDAN: The EDO does not review our case (f 13 studies or our engineering reports. In other words, they're 14 not issued through him. They're issued through my office 15 with a recommendation to the EDO and I would not have a problem 16 if the EDO disagreed with a recommendation I felt very 17 strongly about, to go to the commission. That would not 18 bother me. DR. LEWIS: I understand that you and the EDO get 19 20 along very fine.. I'm probing this for a very specific reason 21 because ACRS sometime a few years ago recommended that AEOD l 22 with a separate office be set up reporting directly to the L, ') f ru Commission and supported a recommendation in that direction. 23 l~ It never happened and I'm not convinced that it was a bad (~'-) . 24 l 25 idea, but we'll see. 1 Acme Reporting Company 12OJi 62B4868
0) 1 MR. JORDAN: And there's also a Nucelar' Safety Board legislation in connection with the single administra-2 1 3 tor and-- ~ ,/ 3 (,j 4 DR. LEWIS: Oh, I know that. MR. JORDAN: So there are a number of permutations .5 6 of different organi~zational arrangements. .7 DR. LEWIS: That's right. I have a conflict of ' interest here because I think that's a great idea. 8 9 MR. JORAN: I realized your previous interest. 1 DR.' LEWIS: But please go on. 10 MR. NOVAK: Okay. Well, just to clost out, I do think 13 we try to write down our ideas. We don't have to polish them ] ~ 12 7'O ff to the point that we spend a lot of time, but I do 13 GJ encourage our people, when we have an issue or a method we
4 want to communicate to another Program Office, let's do it 15 by letter.
Ed is very supportive. It does come over at the gg right level and I have just seen very good response to it, so 17 I think the quality of our work and the fact that we are 18 judicious in what we go after. We try to understand it and 10 then go after it and we have had good success. 20 I don't know of anything where they have said, you 21 guys are all wet and turned us down. 22 MR. EBERSOLE: Well, Tom, let's take the case of the 23 Air System itself. When we heard that report on that, I O 24 G personally churned out a letter for ACRS to pass on I was 25 Acme Reporting Company 12OJi 628 dues ]
n. a k. h-O [ s J1 e 7, 4_ i yJ \\ ' 3. '1._ ' unsuccessful;asLoffthe. moment, but I'm as stubborn as you' o 4 hi L2 are,fI. reckon,'so.I'm not done yet,_but.are you. satisfied L3: Lwith the.actionsLtaken on the-air systems. I' thought that f'; we.could envisionLthat they.'hadLa.potentialsfor simultaneous-L,I_ 4 3 -, J, '5 , common;old failure, especially if you.shakeithem up a little '6' bit and. Unloosed debris intoithe'various receptors, it could- .7 cause some< pretty serious problems.- a Are you? satisfied withithe degree of. attention'bEing h: 8' . paid,by~the industry at'large'to the' air: systems? I gather '9Y '10 'lyou'didn'tLand I wasn't. MR.;ROSENTilAL: I'am not satisfied with the degree 11 .12 ~ of attention that' industry has paid to: air systems'in the 13
past.
I am' satisfied with the attention that the current ~ J14-case study, C-701, is getting. 15 .Now, that went out as an IN, effectively. announced l 16 the publication,- that study, a number of l licensees have 17 written in and had asked'us for copies of;that' study. It's 18 not.over in RES for prioritization-as'a generic issue. _19' Each side has used that study as a basis for some 20 special inspections at some of their plants. Regions are a; 21 well aware of the document, so it's receiving a fair amount y-22 of attention. O. ir months f rom now, ' we may be screening them 23 24 differently, but that's it's current status. { 25 MR. NOVAK: I think also that you have picked a very Acme Reporting Company (2021 638 4886 _'._____i_'__.__________.__________.__.____ _ _. _ _ _ _ _ _ _ _ _ _ _. _ _. _
10 SU-1 tough one, it's a broad issue, it's got a lot af parts to it. (-) 2 MR. EVERSOLE: I understand that. They way I 3 sense it is that'we're just waiting for a double failure, p - ()- 4 MR. WYLIE: The nature of your suggestion or your 5 recommendations in that study, in there it says-- 6 MR..ROSENTHAL: I know it's all very pragmatic and 7 that is that the air systems need care and feeding, you have 8 to' clean and dry them and the particle sizes in the air have 9' to be smaller than what the designers assumed, but the air to quality has to be at least as good as what it actually means 11 in. terms of air quality. 12 MR. WYLIE: These were statements of good practice [) 13 basically, in principle? v 14 MR. ROSENTHAL: Right. 15 MR. WYLIE: But there was no recommendation that the 16 industry take some action? 17 MR. ROSENTHAL: No. 18 MR. MICHELSON: Was an alternate suggestion noted? g) MR. ROSENTHAL: Well there was no-- 20 MR. JORDAN: So it was a recommendation within the agency as being prioritized by research now for regulatory 21 99 action. (- MR. WYLIE: As a generic? ~~~ 3 2 MR. JORDAN: Yes. (T 24 (.) 25 MR. WYLIE: All right. l l l Acme Reporting Company L2021 628 4888 l __._____________________._____________J
. It l 3( 1 MR. NOVAK:- We'did brief the commission. Al Molinski y3 im) 2 who did the work, was here and I think it's the kind of work 3 that we should be doing. () 4 Just recently, in following up on the concern of the 5 Mid Loop Operation, you remember and I checked on'that and 6 we see again that we have gotten good response from the 7 program office. Regulatory letters were sent to each of the 8 PWR, licensees. They are responding. 9 So I am encouraged that those kinds of actions 10 are showing a sensitivity to the things we have picked up. 11 - MR. EBERSOLE: But you know how long these mean times 12 .become?. ( ) 13 MR. NOVAK: Well, you're right and there is the 14
communications as well.
We have tried to alert people and 15 get them to be responsive. I don't--I know it takes time. 16 We've looked at the pace at which generic issues are being 17 resolved. Weve got a number of recommendations that are 18 being addressed through generic issues. 19 The Agency, as a whole, is looking at what it can do 20 to speed up that process, i 21 MR. EVERSOLE: You can imagine such things as somebody l-hitting a pipe with a hammer and collapsing two sensitive 22 I Q,). pieces of apparatus because of entrainment of debris. It's L 23 1 ~y easy to do. (m-) l MR. NOVAK: I'm going to go on. 25 l } Acme Reporting Company 62O2l L2H 4988 1.__...._
D ( dl) i i ~ 1' MR.-EBERSOLE: All right. p- .. (j 2 MR. NOVAK: There was a question on independence of jud g-3 ment. Ilow does AEOD insure that they.are exercising indepen-() 4 dence of judgment? ~ 5 Let me first say that I am convinced that management a has a sensitivity to that. In the time that I have been 7 operating with the Director of this division, I have had no 8 suggestion--there has been no suggestion that our decisions 9 were compromised or affected by other offices. There is a lin a 10 organization. 'We work through it and we're very independent 1.1 in that sense. 12 DR. LEWIS: How many oxen have you gored? The only ( 13 test of this is when you have really gored some oxen. r,). 14 MR. NOVAK: Well. I think Ed's reaction to this 15 tiid Loop Operation, the idea of how are these people treating 16 the containment. 17 There was a question as to when you are operating with 18 liid Loop Operation, you are particularly vulnerable, you should know the question of whether or not a containment can 19 20 or cannot be buttoned up quickly. 21 I think we wrote a very direct answer 22 to the Office Director at NRR saying that there was action L/ 23 that should be taken. I think they picked up on that as an 24 example and we sent you, in fact, a copy of that memo sometime back which is an illustration where we do direct our concern 25 Acme Reporting Company , m,s.... y
nm , rr 1/ $" 3 db 'directlyyto the Program' office concerned with it. l I w./ 2 I don't sense, personally, a reluctance. If you take 3 the things that we deal in, we recommend a number of information \\g_J 4 notices. We'll do a study we think people should informed o 5 about. We've had a very. good track record of success that G we have drafted and with very few changes, they go out.. So, 7 I think there is an acceptance on the part of the Program 8 Office to look at what we're saying very carefully. 3 MR. EBERSOLE: One of our veteran members maintains-10 very stoutly that that problem is due to mal-design and not 11 malpractice and he's very vehement about this. 12 MR. NOVAK: I hesitated to bring it up because I (') 13 thought there might have been-- . <_ / 14 MR. EBERSOLE: It was one of those cases were I noticed 16 in your discussion you didn't mention the design aspect which, 16 you know, I mentioned earlier, that you generally steer clear 'of the accusation that are directed toward desian defects. 17 He thinks otherwise and I'm rather inclined to agree 18 .with him. He wants a vent in it, of course. 39 20 MR. NOVAK: Yes. I'm not going to second guess the' logic of our thoughts. I'think we have enough in our 121 22 report to fix, personally, a good fraction of that problem. 7-(- Changes are being made, implementation changes are being 23 /~'*> 24 made, so perhaps what we looked at are the things that we V saw could be fixed. Not to say there isn't question a questic n 25 Acme Reporting Company Q00 6,0 4806
pt' 9 "t' l 1 of design, but would could be achieved in a shorter time L 2 period. 3 MR.-MICilLLSON: On your slide, what is the meaning of 1 ,~v k_) 4 the "no peer review of the recommendations"? 5 MR. NOVAK: What that is effectively saying, when we t 6 put out a case study, Carl, we ask that the other. offices 7 and industry comment on the technical content, but we are not 8 asking for any comments on the recommendations. 9 MR. MICHELSON: The recommendations are already in 10 the document that they are reviewing. j l 11 MR. NOVAK: I don't know that'that is the case. If 12 it is, that's fine, but we don't-- ( ) 13 MR. MICl!ELSON : I assume that's still the case. _14 MR. NOVAK: Yes, but-- 15 MR. MICIlELSON: You just ignore the--any comments to on the recommendations you ignore-- l 17 MR. NOVAK: Are ignored. { { 18 MR.MICHELSOM: But, you do include them yet. There j j 19 was a time when we attempted to not include them and we got ] I 20 a lot of static because nobody could figure out where we 21 were coming from cn: where we were headed for and so there is j i 22 no objection to putting them in, there's just--I wanted to ysD 23 see if you were still including them or-- MR. JORDAN: Normally the recommendations are in a (~ 21 \\_- covere transmittal that we send to Vic Stella, so we picked 25 f Acme Reporting Company I im, m m. 1 l-J
w a. _, = -. k ' Y /! 'g *.. '; i o' ?: >.f
I
,3' . e'f.1 .$ ',q'* , 4,. ~ 4 .i-- ~,,..,3 ..p <.4 7 .f. 'y c n: ~ ,/ .j,' , c $n + , ( "- [/ ' f Q 'i t e 1 p. t w f@y (, ; 1 9' 'w;, 4 43 g m, n v,' W WA ~ y (&j! F., n'o;ut those ' 61 items. of ; things 1.that oucrht t tof be ?done. - c : s w
1,
,q a '1 p-
w(
g-i ..A.,9-.. t m .2 <;MR..MICHELSON ?They're'in;'the casO studyi.itself.- .j'
p
.y + y. y ' [3 : e' '[MR.[ JORDAN: Nhey'reiburiedLintthecase, study g.. h w 't M( / m.4 -MR'.,. MICHELSON : They're not(burie'd, therN,.'s a"sectionL ] 1 .. f} therei calledo " Recommendations- -conclusions ' $, nb recommendations." u 2 5 16 J w ? 2, M /W ^ ~
MR.L JORDAN:' f Yes.-
But thereCib :notland }r in fsc&," 3 1 1 4 c , y g ,,yn,.A. p,# ,' u. 4 y. ,.. g q,. nq s.
when

wetdiscussed'tliiscwith;Vic Stella', we arc nottoffering d( ', s ' -
4 J7. ., W ,,3- - d]y m.- ,3 ~,7
;{
\\.< ' ?.. t i f,a iC 28' /for.peerJcomment.t -3 {dr 4,,' 7 .M,R. MICHELSON: Yes,that's'theLwholeg'x.oint. et. < 9 y o m ~10 e 7 L'Yes. ?- j 1MRXJORDAN: 4 4 m 1: Zo T11) 'MR.7MICHELSON:.Yes,okay. 3y q 3 e. [ 12 9
p MR. NOVAK: LAt the last timp thNtill;we -(ibLE with L
eTcom-! [ f13 ~ [ mission,?there'wasadiscussion.on;just:howdowe'go$$'ont ] e O a 'citinglon what.s'ouldibe a recomme'ada'tida and what should be j h' f14 k W ,A q J f 15 .a, suggestion.. a a 16.- 5 Recommendations, oficourse, are. our[ highest :levelLt ipf , ' \\' r ll. > 4: 17 - conc e rn'.- We believe-that alrecommendNbiohy we track a ';b a s 3, 7, recommendationcand follow it'tiirough and the suggestion-we I (' i .18: l o -Yf but we don't'see it as; a safety concerny' m - .jg believe has merit, .C a y, c m.- 3 i, !q a 20 of(the same level that would'be. applied to a! recommendation. 21 .The question then l's, just.how do you decide what is a i w recommendation and what is a suggestion.-.AEOD has-had an in r L 22 5 yO; w ,ternal' procedure which provides a guideline for what.would' j 23
fp 4
\\i 24 meet the test. ' ?. .? e In other words, we ask our people to look--wtten they 25 '\\ r .-{ w y1 Acme Reporting Company' e .h,k, (200 6de-dece 1 3 ( P l; l s 'b IJ
A h kk $/V - M , :ffh SQ K' 1-look at a.particular issue, if there is any reducti'on in any ,3
Cl 2'
safety margins, severe accident recurrences or unrelated safety 3 problems. It's those' kinds'of things. It doesn't rise to (/ 4 that. level. '4. Oh 5 , A:nd Jack,:of',7ourse, is doing that day in, day out, 1 s b V' O .so I'll ask him and you really look at the particular thing-- i, t ? y~ f 4 as l you're working a problem up, you'll have a particular 8]L point that you want to make,.you don't know what to call it, I 9 call it George, until you go back and you try to weight it, 10 how important is it and if if it mecaures up to one of these, 11 .then we're going to decide that it should be'a recommendation M ?) 12 and that suggests then that we're going to follow through on p 3. l 'i f If it doesn't have that safety significance, we want 1 -14 3 15 t'o convey that we believe there is something here that people 16 should pay attention to and then we'll classify it as a sug- 'Y gestion. 17 gg MR. MICHELSON: Jack, I would like to follow up with 39 a question. It's my vague recol ection that only case studies 20
have recommendations, technical reviews and engineering evalua-t Yet, many times you find s'07m very needy, im-q n, ions do not.
i t n ie \\ 22 portant issues treated in an engineering evalaation or a / ) ^ ech'nich! review, but none of them can ever be sent out then l 23 r ( f (^) 24 iSt. ha rd recommendations, they can only be--I think you never v 25 send,those out with recommendations,c oply 'f th suggestions no i 1 Acme Reporting Company l 4 cm,u.ou 1 i L -- GL
~ i;m .m. 4 ' 3 '. _$( gj jj h,p h., h j 5lq 7 3 l .-].; _13.. I{. _ ll:fihli,j$? Y.lMip? ' h g., k 7 g 'N:llf&.fiM(W l('W W ,r a" ^ T /)ffl Kh .g f;)f :M W 1 1# qh 4 'n7L fif pfM[D y lly.b. ,} j -- w: Ol }y ] matterfliow ineaty$ theiissue sehna :to be.; e Li [W[ N' T @[;i[/12 . hk i j O . i That'sEwhat decides-1 ~ CouYd M L2 lMR.~ JORDAN Lf'g g,LI help? ~Q,J .-q M ,,7 4? ,o, .y ' I. wantNS/ shy, A-'a'. case. study. . d
whether:'orlrrd 'it' N 4datido
+ .3, y..g + r p tibecombd*a case stu'dy kf it._has; recommendations.. ,:j ~ c 1..)? 4 -k. p. ( .MR.EMICHE SO'N:
Jo,.-I: don'l yhink that.'.s1the.way~it hll llf (
ff 6: wo rqs. - 7.s-a [p
.g :i
/ y; 1 ) [..:OL'MR..' JORDAN: That'<s the way ih vorks'now, .a ,[u,f, p c .7 ..S q. m, y-3, q., 4 f N h 'MICHELSON: bo.[s. it?- h, < [ 4[hh(gs i/s t, 7 y
/t.
t i r . w q n, . m l <V. 9 T1.MR.-JORDAN' wyes.- : ? ,i.i Q t .C /. t. q = , [ /! 4., .i (,, . thj You mdanTif; yoti, write a two page 'MR'..MICHELS 10 ( i 91 .. T .s .1 M n. .techn1 cal 7pview tmdlyou decided thatJyou;.'really-got;at Jy 11l yf 3 p o fred /lpumEndati3n,th'eycancall'itacashstudy? h (l NM -12'
kl[hTHALify 9
Weihave?anoth]br) categorization' called I j 13, ROSL MR. e&p, y '; 5) / ru l t{ "Special Stud $fs."'There~are no regaldtions*onithisiihat.says t ~ J+ A rjo 'l$ c case study lias to be 82 --pages in, Npgtil or 3 pages in length ~ [;\\ e ~ w '_ d F1.- 16' Ygare' tequired that.it had af peer reviewjand we. can1 .g rs d 'h ihave a:3.tage16dse, >g
hc n

.y h _a ~ R g j
MR (MTCHELSON: ;: hit l oply on a case study'are you doing y
a Q } peer reviews. y/)'. p 3., t / .w y ~ Mp. ROS.ENT)iAb: So w'e have'a 3!page case study. [p <.x. 20 - 4 e j 34 j [D 5" MR. MICHILSON: So, I guess what it amounts to is that g z s 3 ,a -c you never make a rec' commendation unles you send it off for q j~ 22 q -N 4 l C' ' pder review, that's what it boils (p n to. l I o3 1y ~ ,r 3: MR. JORDAN: That's our rule, yes. What it is is l,. 24 dependent on'.what's contained in it. 25 Acme Reporting.Cqmpany s '4200 628 4888 d [b - 1 O < 'h
m;x , bb 1 -
p A
.t e i e I' b @lfh:p:t &v
. c y Gr.
G. @pt%[, j.). 'cli MR.gMICHELSON: ~ 'I've never'seenLmany'shortlones al- @f s. .(; 1
y,. :e Ji d '
2; though there.areLaffewi but 'mostfofothem take several' months- [ 's. ' JWR 3.'
. or!:tyears --:
>7 r You' set the pattern for us. JQ>.S _3.' .4 -MR-' JORDAN: 3 .!MR'.MICHELSON: 'I m a little' concerned--I read some 1 5.- + ,t i ( 6_ ire'al; good-technical reviews'andiengineering evaluations,.. j jf n ' '.7) 1.butfyou' don'tiever make recommendations, Land you seem..to f m 4 M I8 give me all theibasis'thatlyou need-for a recommendati'on. 1 b It.? still',; you' know,,. j ust is submitted suggestions not reconmendations,. U 0 o .... I'think weJrecognize that to some degree, MR. NOVAK: i l10: 733; Scarl'. ;. hatLwe're.trying-to do~is. pin; point our-audience. In W jOd ithergwords,;now-that we have put"out' engineering evalutions, t 12 o 1 ,33; 1we(writeLdirectly to a Regional. Administrator. When.there was 1 ${]- ,y %f}' -+ faLSalem; event,'acoupleofyearsago.,'itwasanengineering; , c 141 'W 9,
.}.
15'
evaluation,j:WeRthought there.were some rather important

ypf J
116 ,p'oints(to b'e'madeand by breaking it out and sort o% w;iting. W, 17 sit' direc:tly tolthe Regional Administrator, writing it to the; 18.- ap'propriate' level of management in NRR you sort of put the' i e %' ~ onus on them. You: guys tell us what you are going to do about 19' 120' it,,we:think it's important tolfollow up with,it. N ; ,t . a i. 21 Now, that's. kind of influencing. We're not going to ...g fg 22 jump out.the window if you don't this, but we certainly- ~ Q;% I think'you ought to pay attention and with some discussions ' ' Mi ' 23 D) ' 24.: withithe utilityl you can probably turn this thing around N.4 v. ' 25. ~ andall parties'.should be better for 4.t, so I think that's what y>t Acme Reporting Company aori sie dese h
mmmm = u- ,4MQt t .n 4 w, , qs >~ n$g9]7 [:... ?,, [y ' ' [M:.h a ,s a w W,,F \\,
h. '
, Mew 3 .10 E, M, % l1 (were?trpingitkdo.. h .x
g

2 fiffyoutwrite!an'e'ngineeringTevaluationiand;you puti i
,4, q:. ~ >,o
3- /everybody'onYthe distributi'onflist;tyoufletl people off; thel s4
.hSok). I th' ink what'~we're;Ltrying toldo-isl kind:of} pinpoint .i: no y- .5 lthat7recipientiand try,todo better in getting;them to;be'more 6;. at'tentive L to - our ' problem'. s ndl J. f Okay--! ?> v '8 DR..: LEWIS:: ;One. question.- -ri 1. Sure. 's . MR'..' NOVAK : s 8 m ~ '10'- ~. DR.. LEWIS : Your; recommendations--you would actually. o p.: ; 11- ,make recommendations.and:they'could be--I just want to be sure ?l2 about t:he f system.' They can; be pocket vetoed in the sense I ((Q., '13 -- thatjthereTis;no.' responsein the rules--in any ofL-the. rules: ' i i..- <l -y i 4 il4- ,requir.ed:to a'. recommendation,.is.that-correct? r s b ~ v
15 MR. NOVAK:
A.recommandat' ion'is made to the staff. \\ / i> > ; j 'DR." LEWIS: 'Right. And they canisimply say, how: '16 n y. I 'I . 1'. interesting'and'go onfabout their. business. 7 1 '{ t s [ -MR. NOVAK: And wecan appeal that'to the EDO. l s
18_
MR.' MICHELSON: But they have the document. 10 i MR.' NOVAK: 'They have the document and-- ,20- -t s
y e
[ 321 'DR.' LEWIS: That was the question. If they don't y m ..[ 22- . accept'it, they have to document their reasons for not ..y ).. J' 23.- caccepting it. >( 24' MR. NOVAK: Right. And if we find-- 2DR. LEWIS: In what length of time--what period of j Y D: 25 l i j i1 IV Acme Reporting ' Company n 4 47021 $i?8 4888 s a -,.W. .n. ' : _ 'u .f
p.. <. g,., g;o;j ',m' - a-y, g.
v%
jp ~ 8 u. ?,); (b ' ~ ./ ; t N
~ l
.' Q 3 - o ,m ,y
N7( y i..1 ;
ftime?-- I .MR.[NbhAK: 'I: thinn if Lwas 6 months., , a.:. . i.- + 131 J MR. ': JORDAN : - 30Idays.: g. 7 { ,4l LDR. LEWIS:,jAnd.-then[it's your decision whether to c ~ [5j ftake[itfup'the.line;, g _f 6 MR.;NOVAK: .That's right.-' ,w 4 ,V- ._ 7 'DR. LEWIS:- ThatL's what I didn't know. v m .MR.fMICHELSON: ' -Only?a-recommendation wasLtreatedi ,@ 7 8 g.m, M ID: 's.that'way. 10; MR. NOVAK: Right.. " [ 5.' jp p . ' DR. : LEWIS : 'I. understand. W . ~ 12 J
MR.'NOVAK

i Okay, I wanted,to:moveion then andLgo on to[the AEOD 's influence- '.'What ~ influence ~ 'can AEOD:-bring to
-.13
g47 l$earion[its' recommendations.: offered'in, reports....How is?AEOD's 15 1 advice.-accepted.bythe] agency?.

~
16y Again,fIithink;itjgoes.to the quality of our reports. lOne'of th'e';thingsi hat we'restryingsto,do is sharpen.up those t L '17 ! t 1 81 ? recommendations.: LIn'other wbrds, we think can improve if we can be.very' specific-in'what our: recommendation is' intended-to ~ 19 do.and I have ask'ed our people to do just that and I think: .20 + we will have more, success if.we can be very specific andLdraw. 21:- the line. I don't'want to go.back over the. air system, but-22 . Q;. it mayTnot be that easy,.but we ought to be searching for that 1, - 23 p i simplicity in terms of a recommendation and then you can decide
24.
25 - jyea or nay. .y, f -Acme Reporting Company 4202) GiB 4888
[ Mis '.Wi"' 4 u., z, Tl x s t U / [Qf ((,,,' v yy; pw 3
[
mWW q ,3 n -..u .l-
R f..,.,'
,: 1 1 e Yoti :hadsa question',_ "Howfdoes AEOD measure. their owrr ,7jw[ T, ' l ,, g.. ; -ff 2; l effectiveness? aWhere does the feedbackh.comeLfrom?" ' : That 's ' q:. ]9 ie1 3; ]questionL8. 4 I t}iink, just as1we mentioned aMnumber;of\\ things,, ..v h. M 7 [5; we're:gettingfaLnumber of'feedbacksLfrom industry. We didia .c ~ g studylon'!BWR Cleaning. Systems and it was'taken very-well:and' ' C.C nj 4 theyput[it[into(theirftrainingcurriculum,forexample.- ,? 7 'y< V8 .A number'of' things,.Inthink, determine the' feedback m .9-through the fact'that bulletins are issued based'oncour f. .recommehdationsk information noticesLare issued ba' sed)on.our-- 10: ,E l331 recommendat' ions). so those are positive: feedbacks. b' S ~as far_as.the staff:is' concerned,1we'.ve'got somei
12;
.v c 7 !13, Every clear: ways'of' gauging our. effectiveness.- Industry,<we .y ! continue to. work'at it. Just most recently'Ed went'down;to ,1 - 3. }14-c 3 a 15' .tAtlanta andcwe. talked'with INPO about a new-plant study;that L J ?ly we had done.'.- Mark's crew" looked at a number..of. plants-c l'icensed ini ' 84 - 85 andf out' of.'that: welsuggested to.INPO g 38-it would be appropriate to have a work shop'and:they agreed g'g -and the next_ month, they'were invitin;g 15.. utilities who have. plants just coming on the line to discuss with-INPO and with j -; 20
. [.
d'" us someLof' the lessons learned with regard _to new plant per-l 21 r 22 -formance, so I see that as a positive feedback and suggestion i f a s 'p ~ 23 - as:to how effective we are. E 24 Again, I think that new plant report.was a quality r25 report. We worked at it. It didn't get there just--there was ) i l ~ Acme Reporting Company sacas ais.4ase
z. T gflly~ ' < 7),.. Y.
+ :& '
~ ^i h ^ w g.p ? ' + /, f '} : ' 3 fy " b o.. la g ' 1j iterations;~and Ifthink that's.what we're? lea'rning, how best 2[ be. ef fective.;. 'Butlonce it'..:.was outi, LI :think it was'.acc'epted ? '3-L bylindustry. as Ih'avingf some' very J good 'le'ssons?.to - be. learned E ~
7 l{ (>r
.4- -and that'sJwhat:we're doing. m '5 So that's the way I think'we're;st'arting to look at: l ' p 6' how effective we.~are.. ,k o 7 j '7 M'R. MICHELSON:- 'I think ther.e was'ailarger question-8. ' involved 1here though thantjust AEOD.- I-thinkithe basic ques l9' EtionyinvolvedLwas - how is.-the industry viewing operational-10' - texperien'cefin feeding it ba'k into their~ operations because. c Ell: 'weLread your case' study of a year or--a year:and a. half ago .i 112[ .or morelnow inqwhich it-painted a somewhat, bleak picture of ~ 5f l 13 - lourfindustry which was usingsa' lot of this operational; ,J Is ? experience'and I:wondsred.--if the picture has2. brightened up 4 15. 'anyisince:because.I think you did some. follow on with one or -16' .two utilitiies and-I just wondered how you viewed today.the 6 J .17c way'in which industry lis viewing ~ operational, experience in _- g J'L 18' terms.of enhancing the safety of their own operations. 19 MR. JORDAN: Maybe I would just kind of restate that 2 20 is'that I do feel that the biggest problem that the U'. S.- 21-industries have is its lack of the closure on the overall 2 22- . feedback and we're looking for ways it's positive enough to.be ' f
O.
f 23 imore'effecrive. /]t l24' MR..MICHELSON:- As I recall, in reading the report $./ - 25 .though, you found that many of the utilities didn't even read Acme Reporting Company ] an n.- n 3
ap n:: 'v y ,7 %k.,. p kJ H-q u a ~ j -1061 R o +
y 11L l things likelAEOD. reports.
They> filed themlwith'their' O q 4 M.g y 1 Li'ensing Manager or.something and-. plants.'never'saw the,
y. M.p; i2; c
l p 0 " pf,. ?3- .reportsfandItheffeedback'that you.wereEtrying'.to giveltoEthe? D J S -industryj:was'.not getbing(to thef u'sers---the real users. 74. .v.- .'And'that's'why,Lfor instance,.with this. 5:
MR. --JORDAN

k,3 ; U: } t ~ i 6' new plant study, rather..than. send it out as a broadcast' or; re- + i. j ' 7I L ' publish lit'as a'new reg, we also have-been sending.La specific '8 letter.to each new; plant.-licensee, those who.were.in the first l 9-two years ~of operation,-with a copy of-the report recommending 10 thaththey. review that:and utilize its lessons. .11, -So we are trying'to focus the'correspondenceLto the 12 right. place. The utilities, you=can draw a circle.'toLthe 3 2( =13 utilities.andithen" identify all-.of the interest points,or 14' linformation' points.'.Each' individual,-utility has literally ,15 ithousands-of.: issues a year that are sent in to sort out and -16 there'is'a.need to somehow emphasize, focuse on'the ones'that- ~ "jf '17
are'important to them.
!18 !!R. MICHELSON:- But I senses in reading your report d
19'.
'that they were. pretty much depending upon.INPO to worry about a -20 it. r-21 MR. JORDAN: That's right. 22 MR. MICHELSON: 'And no depending upon AEOD or its cbJ hundreds of other inputs that they might be getting. It 23 ,i L 24 - appears they were pretty much falling back on, if INPO tells J25 me I got a problem,.I'll look at it kind of an approach. Has Acme Reporting Company 120 0 628 4888 r__.__=_'___
n: cr., , ci fc( .[' s. 107 n 1
that changed.any or is that really a good perception of the.
,3
k. L 2
situation. 3: MR. JORDAN: Jack Crooks was responsible for that i s/ 4-initial study. There has-been a subsequent study. 5 Jack, maybe you would like to make someJcomments on 6 that. 7-MR. MICHELSON: Have you issued a subsequent study? 8' MR. JORDAN: A subsequent study was done from INE.and 9 has been. issued and-- 10 MR. MICHELSON: I haven't seen it. It wasn't an AEOD 11 report then. 12' MR. JORDAN: No, it was not. [ )k 13 MR.'MICHELSON: So that's why I missed it. s-14 MR. CROOKS: Jack Crooks from AEOD. The follow up '5 study was a study done by Pacific Northwest' Lab. It is not 1 16 yet through the final completion of the report stage. They 'found that--what.they basically looked at was the effectiveness s 17 of the NRC communications and they were to make recommendation s 18 for ways of improving the systems where the systems needed im-19 20 Provement. It currently--we have gone through a draf t report-- 21 a draft report has been entered and they are basically finaliz-22 ,l C) - ' \\J ing it. 23 (~/T 24 They went to an additional 5 utilities. They found-- \\_ they confirmed findings that we have come up with. They also 25 Acme Reporting Company (20 0 628 4980
ygQ,W, ,P %in ; nf ' & ~ ~ ~ - p: 9.r ze o.- e s @&ola;h .A L L s hv. c g ;a "10s; } n> un ifoundNhatthe"situationwas_ch'nhing,lthattherewas l J[ g a [f+ Mkf;l(,g 2'
recognition,of"the{ things.that EdLis talking.about that: there-
.l + ' [-; ~ \\ fp_ " '. ~.. ~ needsvtolbeincreasedifocus. 3- .) 4-One of the follow'up' actions that I took was that*I. L , >. ~ idid go.back.to INPO and; discuss'with them whether they,had' rl' 5-6- nmade. program.chan'ges: based on theirEfindingsLbecause they? m E Lalso wereLlook'ing aELthe same.. things.and they havejfediback. 7. W 8 ithrough'theirLgood-practices some'specifi'c changes.. 1 1>,' e "9' 'In other'words,.made recommendations.to their. members-1 To that]they ought to focus on, particularly the things,11ooking. j i -( 11l 'at the" effectiveness of?their' program. In other words, they're '12 .f, -getting?all.the?information, they know what the objectivesis. i 1 q 13 The. objective'is to, you know, feed back the things.that1will 14 ' leadfto impr'ovement'and safety andithen assess is that,actually-P -L orjis everybody.J earning.the information'by not' f r i 1 ~ 15 happening,.
16 too many things happening to improve'the situation.

17..
MR. MICHELSON: One of the basic falacies of the INPO' j 4 i ,18 approach,;of course, is they try to view the experience.in'more 19 of a generic ~ sense and send out instructions to the utilities. 20 Events that'are occurring on a particular plant might 9 21 be extremely"important to another plant that had just that m similar situation, but when INPO looks at it,.they say, gee, 22 TD this is just one. plant specific, not realizing that there 23 might be another particular plant having the same difficulty. 24 ~ There is no substitute, of course, for the plant looking
c 25 Acme Reporting Company 12OJD 628dPLS
hhh, s, { ~ ~ ,m: M .[ + gym [j '. 7 3 {Q 4 .y w 'I-..' g- ' N..':'tl ~ , at anjexperiencetand;thenfrelating-it to'a situation,.but l 1 S.h
apparentl'y they're?not2doingfth't._

They're.looking atlit a
2 W = thro
ugh. generic examinations.such as, I-!think, basically-7 g,
'(] LINPO.does'.. ~' ~ j 5 'MR. CROOKS.- 'I.: don't want to. : leave the' impression - , n,, E< I6
that,we.were totally ~de' pendent on what.INPO is!doingLbecau'se, y'
lin h'ouse,.'we also,:back through the regions and also through 1 3 8-NRRan'd inspection programs', arecshifting the focus somewhat ,R -g' '~to how well is the operating. experience being used. L10' LMR..MICHELSON: How can you shift it to the utility
n
? , by that process? . ig, 1 .12 .MR. CROOKS: We're trying to evaluate their response. >.2 u / 13' I~didLhave--what we're doing is, say assessing how well they-W-c are reacting to theifeedback and whether or not our. feedback 14 15 .-is-havinglthejimpact that we expect it to have.- ,a 16 'MR..NOVAK: Carl, I think we're starting to look-- MR..MICHELSON: Thafreportwas-issued, it's a 17 18 .real. good report. I'm sure~that th'e regional offices..got it. Are they then responsible to pick up--you can't enforce.that ig -20 report, because there's no requirement. If there was a 21-requirement-- MR. WILLIAMS: We clearly recognize the problem. [ h_ 22 l MR. MICHELSON: Beg your pardon. 24 MR. WILLIAMS: We clearly recognize the problem. I l21 think Ed has earmarked it as a major one and we're revisiting-- h 25 l-Acme Reporting Company I au,.
4...
E 1
r-w tj: ; br~ ; y, = +- v3 4 i 7.... i i 4'. 4 t i 3.ji S...j 211Qi ,y , ; x. q 1. ~ Tom,Lin his' previous ~ discussions,3 n.. + . intends-to pursue aL ~ [ i /n I different/ feedback / mech'anism'tryingitoiunderstand it. i ,,.r 2L J' 3 pg J3
We"know 'that there"is'.a failing of processing qN.
if 4'
.evsn information~from the NRC in preferenceit'o:INPO's'ER's. 7 s 1
5 So.wehaveLspeci'fic utilities and events and we.know that-q N'
LG Ethese'.. things?are happening. ,5 ( .7 So, it's.a recognized problem, it's a failing and it's L8 .something we've'got to work on. y 4 ( 9' ' :MR. MICHELSON: ,Well', it was a real good report. 'It m 10 left me' quit'e' disturbed though, _1' l - MR. JOR' DAN: -.It was a disturbing report. ~ ' .MR'../MICHELSON:' It.was a' disturbing report.; \\- [( 13 - <MR. JO'RDAN: I-wouldLmake a st'atement that there are m. 14: more ga' ins. to be ma'de in utilization of'the' existing evaluations' .s 15; and. data:that 1s.-being generated by industry and'the NRC than thereareLimprovements-that.can:bemade-in.the'actuAb.evalua-16 4 17 'tions'and analys'is of the data.- ~18. But somehow getting the information to the' ultimate user and assuring that there is disclosure on' solving the-19 20 Problem so we can make more benefits in a short time frame'. 4 21 22 MR. MICHELSON: There's no use in developing'more 10 recommendations when you can't get what's out there now. 23 lM 24 MR. JORDAN: We really need to focus on that operative end of it and the idea of targeting the information, sending 25 Acme Reporting Company (2021 6264868 2 v -,.e i
mn:. ? W;)b ,' (b ~'
v.
+i,' a +,3y $$[ fY Y',, 5 .itkin -af more'.ldirecte'd' fashion and s being selective. about the-r i +. 7; 3 J (' 2' .inf rmationithAt's in and'that means substitution because a n.-. s i# ' ,Ag
decision! between 4 recommendation and a suggestionLreally-
&~ ~ y changes the' emphasis a greatadeal on what'happens andlyou; i 1 n a+ q Y 5 - w uld.like tojsend!everything out as a: recommendation, but 6 ,when you do,'you? inundate the process,;you won't getLany-4 of it don'e--accommodated,.so.we're carefully weighing-them and- ~ 7 u
7
. e have the benefit of always being able to re-examine, l W !8 w t
g
, 'If-we;put out'a technical review this year and next ig,
year'that particular' issue'comes-back to us and we're. con-cerned about it,'it's no' problem at.all. making-a-case. study
,33 p i2- -onfit,and so.we"have:that continued ability to reexamine. l 73 4 !MR. NOVAK: I would like to. move on.to the question s. 34 - you2hadLo'n ho*1do we followLup on the. recommendations andi
15 uggestions.
p .Well,'we do have a tracking scheme. .We look'at'each-16; recommendation-that-we make to any. specific facility'and it 17 d. is tracked in the sense that we have a formal tracking. '18 We can tell you exactly the status of these. We update it 10 20 twice a year. We're going to be looking at them just about j this time now since it's been about 6 months since the last l 21 4 l 1 22' t ime'. e LO..: I thing this is something, again, that we want to look 23 r at because what we found is that it's an agency problem. Most 21 of our recommendations, we have about a 100, roughly, that are K> 25 Acme Reporting Company 62041 67H 4898
--- m ):! l: j _ ~I' l l w, W S p: m; ; ym 112L A. ,. n i i i -: i, - ><799 i ?containe'dlwiShin? gerieric issues,1 so [it's)the pace at which
i. 6
. 7.c .( T' 1 1 - M -j f2f lthos'es ^ generic : issues :are f resolved that ; will:'see. sonie f closure R gg .g + l % a:
3l Jonithese recommendations.
They.maylhave.sparkedLaLgeneric $4; l issue orithey~maylhave been folded in'with.something that was .j .5 started' befor'e: our:. recommendation came to' the - surface, but -t s '6 we certainlyidon't loose track'. We are-interested and we- .7 -do-follow most of the offices and we,are--I:would say1we-are: '. 8 'having;some success in trying.to focus lon that. 'We hav'e 4 x 9_ got some Work-to do'and'I'think it will take a combined agency y j 10' effort byVResearch, NLR andisuch.. ~ J
11 qMR..MICHELSON:
.Wh'at do you do if you feel..that it.'s. .] 1 12
just not going'along fast enough, what is the method =of. opera-
[4Y i
13
' tion atithat point. A3-l . jq .MR..NOVAK: Well, Icthink what..we would do, first of- . ' 5: ..all,Lwe would communicate with.the office and if he felt'the-d 1 ,4 .g j
n 10.
pace:was? going-to be.what'it was;and'we felt it should be j ~ 17; accelerated','we would. write the EDO'but we'would havesto have .<e a +% a good technical basis for that. i '18- ~ (: U 19 MR. MICHELSON: Have you ever done that? b 20. MR. NOVAK: No. s 4 s 1 21; MR. ROSENTHAL: Yes. A' week or two ago, there was a 1, l
22
, report.-on a case study on an operation and it was sitting over j . O*. 3 2" 'in research. 'It.had become PI something and it was sitting 23 .24 there. 25 And, Ed Jordan, in.this case, wrote a letter to - Acme. Reporting Company $2026 629A488
, g x ' ;n @p-rum. p l 43 - 1;1j, 0 9:n y~ p- ,, '[: lff, y('r,1Q. [* ',. i ~ themi hattsaid[ we.' studied it in-the past, INPO_has worked
1

? .t m . ;3 $/ e d o n < i t,- e v e n t s c o n t i n u e ? t o.L h a p p e n,; i t s ' j:A]p 4 o i > 2.: on ' it ',{we ' ve: w' rk. 1 x l' 4 i time 'for ' regulatory.' action.. l 1 '3-i ' 4,9 o l l. .4L MR. MICHELSON:
Good, W
s 5
MR. JORDAN

Land they.did respond.
-i ~
i 6

MR.
ROSENTHAL: We don't do this>very often.but j A LM 7t ,we.are certainlyLprepared to do it.. 8 'MR. JORDAN: 'And so the action out'of it wasya generic .;( 9 Lletter that: went to the utilities for their action and that' 10. has'to be the~ goal. It's.not fixed until siti goes ' to the i 11-
utility and(the utility takes som'e physical action'in the
~ ' , 12 '. pl' ant, :a review and physical action. That's'really where QO
13; we're1 going.

s. ;
WJ G I 14' MR. WYLIE: Let.me ask a~ question... This particular l .15 , oneCon-the: air. systems, there was a whole. bunch of various 16 perscriptive recommendations there. How d5 you follow up-(on those? -You've said that you think there are some things 17, E hat'are much improved or improving in that area. Who takes t 18: s-19 that' action. Those are all good practice actions. l practice actions. -l 20 MR'. ROSENTHAL: They are al. 1 21 The ultimate way that we.know the plant, of the pro - q S 22 gress'of the plant i's that you look 5-years from now at the 1 ":O 23 operating experience, you should see a diminished number of 24 air system problems. That's the ultimate way you sco that 25 it is. fixed. ] ) Acme Reporting Company s l 12021 628-4888 .J. l ___________._._._______._._..___._____._.m.
my=T rw-- mr =- - - - z- = .h f E' (
h. '
-f + i, q t <4 ,,m < ml':- .(' iy ' N', .m y, y i gm n,py-1 tNow,. 'the 3trainingfinitiativ,e in1there,-it'.got; turned " jff b. l 2' overLtohINPO.1 fWe'.llHhave.to monitor *that.- jud q f3;~ j 2 w1 x [Other issues.in that' report are being prioritized Thlg J' 3 L... % ?f% }4 ain:the..RESLgeneric resolution' process'. Whenlthey do their, 3.Nf, i f ' L, .L ) ' ~ y _prioritization','we.ll.get anotherJshotD'at it, butiright'now-- 1 j i ,.. %^ ' ' ' This-is research?: j H ?gl MR'. WYLIE: n, 'MR'. ROSENTHAL: Research. ;We' continue to see: events ['l s 7 s y [ -8; involving _-the air systems.andfI keep' sending.them'overito: i [c. 9' reminds them that?we saw'an event yesterday" involving; a 1 -veryfslow:-MSIB2 closure and that was a good: example.where 10 a tit: -we.should participate.in the morninglreport because we l 12'.
staffed it and'said, hey, it'lo'oks like you have. system prob-13l:
1 ems and I got-called-out of a meeting earlier today,_.the; man'. q Twho wrote thatireport'would be going to:an'AIT"at' Perry. j 1
tomorrow,Jso we'rs keeping the pressure'on.
15 -
MR. EBERSOLE:. But you're' hoping.these events'will ig 4
gr occur inf singular style,, but these are parallel systems and some untoward l event'.like-aifork lift. bumping into a pipe, 18 'can create coincident events'that'can'get more wicked than just 19l 20
a= singular failure-somewhere.
1 _l1 "21- .MR..ROSENTHAL: Sure. r '22 MR.-EBERSOLE: I think-- Gp .y8 '23- 'MR',ROSENTHAL: That's the basis'for fixing the air N , 24 systems. Any. single valve'failing, you know, what you're y %h -worried about is common mode failure. 1 21 .v. > !. I.. Acme Reporting Company i a n.,..... h___z_2____________.___:_____ ___._____._u_m_ _ _. _ _ _ _ _ _ _. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _.. m m
I i n j h g, * ' a + [/ I[, s ,, & ( l .t '1- ? MR.. 'EBERSOLE :' Yes. i t,9. y} 'q[ "(f 2 6 MR.[ROSENTHAL: And it's a;very difficult' area to
g_p 3
"a ,33% (3} quantifp.;.~If(ERA.were out,_RES'would have to function' s f; + c4s Mien. 'We'have relatively new. data on: common mode. failure , &_~/ i w. 5' Lof equipment'in the air. systems. -"1 "6; '[We'vefgot some examples,.but from:an engineering-f f _t 7'
stan'dpoInt, we-know[that thatfpotential'is out:there for-R.,
3= ~ . gL ihavingjmultiple failures. That's what makes.it a tough: 9 ' issue. 10; ' MR.!MICHELSON: Wehn we talk-to TVA, we',11 ask; .i 11_ them~about,their water system and the_under design lof.the. ) 1'2 .after coolers on the air compressors and with theirebeing h 13 over pressurized and their higher _than the air pressure-- ~-).1 H 14'
the'. water pressur'e' is,high~er thanJthe1 air' pressure,-you' bust:
v 15-that 'af ter cooler tube, you got. water in every. piece _ of equip-S. r 16-ment inithe. plant'if it's big enough and. fast enough to--if yo u m t -want'to. worry,' start worry'ng about things like that. They're ] 2 c-. i e b.a L17s w T( 19 18' supposed to be--the water pressure'is. supposed to'be lower-j t. lthan~the air pressure and,.you know, Calvert Cliffs had'that- ' 19 ' ',3 20.- , problem,a few years back and they got quite a' bit of. water q 21 around the components before they got it under control. 22 There are a lot of things to start worrying about '23-if you really..want to worry and that's a very practical prob-g,. f 24 : 1em"and here is where they over pressurized it way beyond R 25 ~ design Some of those were only designed for 50 and 60 pounds l' n Acme Reporting Company (202) 626 4888 ..?.w M r. __id_____________._____ ._m_ _ _. _
nj m, d a, sn-e c ~,a c n i S ' JSA 4 4 f, y a origy a q; _s 29 .%e ' '} ) ' s, 4 .i \\ ,W, o - t. t., a fandJth'ey'reLrunnihg3a-1601 pound waterfinLthere'with a~100-
1'
~ i W o r... Mi [2( ppoundiair. ~ ,2 T '3 - ' MR.-WYLIEi, Solbasicallycyourifollow up on those; l l[N .at research, and'what actions they ?. .j l4 recommendationsLis"look.. s I5' ' take) on th'oselr'ecommendatdions? 1 0 MR.cROSENTHAL:: On that particular issue. :.NowLlet,me
give another'exampleLof.a case; study'on motor operated valves J
jj- _8
whichLwas forwarded.
Vic'Stello forwarded to'to'INPO for 19j
action ~andl tasked us.with following'up on that action'and.
Twelare,doingithat.- ', 10' p1 '117 'We.had'our-opening meeting with Hugh. Marks ascouple 12-of months 1agot A couple of weeks ago we had another~ meeting L13j Lin, house of several. offices, RES, NOR, and AEOD wasninvolved-n (g4 f a$d we'llibe' discussing ~it again shortly., [ '15 -In that case, we were tasked for follow up. We'have 161 ibeenitasked with follow up in terms of. resources. i c ^ 17 MR. WYLIE: Where you make a recommendation.and an IN-y. 18: goes out, what~ follow up is done on that?-- o 19 MR. JORDAN: Well, we specifically, an IN is followed ,; up'by'the: region and-- 21' MR. WYLIE: You don't follow up to see what is happen-L/ a
22 ing in--
y c(]. ~ MR. JORDAN: Not in that specific example. We would 23 ' 24 be satisfied if a new commission notice had been issued. MR. WYLIE: That satisfies you? 25 Acme Reporting Company o e.,...... i..
=- y ,,m 117 1 i l I 1 MR. NOVAC: 'Yes, for that example. But where it is ,m. s._) 2 something else, we keep it open until we see it physically 3 change at the plant. A recommendation, in our tracking 4 system, stays open until.the completion--just saying it went 5 to generic issue blank blank, we don't' quote the book on it. 6 We wait until it comes out and what had been done, then 7 it will stay open until we see a physical change in the plant. 8 MR. MICHELSON: Not in a generic sense though. You 9 don't monitor each and every plant to see that they have made 10 that particular change? MR..NOVAK: No, but we would-- 11 MR. MICHELSON: You're satisfied that the requirement -12 ,m has been -- ( ) 33 q,..- MR. JORDAN: Right. The generic water system, the 14 15 previous issue, that satisfied our immediate concern. We'll continue to look at data to see if that is effective now, 16 MR. MICHELSON: Now, where you're tasked to follow g7 ig INPO, which I gather you were in this one case, you will write it off once you have seen INPO issue something, I don't know 39 whether it'would be a good practice or if it will be--they 20 can't issue requirements to licensees. 21 But you're satisfied that they have done all that is 22 %) necessary to take care of your particular problem? 23 MR. JORDAN: Yes. rT g Q.) MR. MICHELSON: And then you write it off? 25 Acme Reporting Company 1202) 628 4888
gg k,- 7 .f I L 118 l' ) 1 MR. JORDAN: Yes. i Le g s_/ 2 MR. MICHELSON: You don't have to do that personally, to-speak, as opposed to NRR doing it. s 3 (3, J 4 MR. JORDAN: I have to' emphasize again that writing 5 it off doesn't necessarily mean the end because we really want 6 to see:a change in the frequency of that particular type of incident. 7 8 MR. ROSENTHAL: Right. We have discussed this and g have plans to go back and work at MOV's, look at it and check it. 10 .Over the years, the programs have been effective 33 12 and the rate of failures.of those devices ought to be coming () 13 down and that will be the ultimate judgment. v MR. WARD: Have you actually followed anything through 34 15 and observed that effect yet? MR. ROSENTHAL: I have not. 16 DR. LEWIS: Could I explore why you want the event g7 18 frequency to come down. Is there a general rule that all event frequencies should come down-or is it because, just 39 taking this example, because it has previously been under 20 estimated in the past and discovered through events that it's 21 m re f a threat to people than you thought before? 22 23 MR. ROSENTHAL: The failure rate of MOV's and the /- 24 failure rate of check balances is too high. E.]' 25 DR. LEWIS: It's too high compared to what, that's wha: Acme Reporting Company 120U 628 4888
o ?Of 1 119 1' I'm groping for? You can't make them not fail. ,,.( ); 2 MR. JORDAN: Too high compared to the safety studies 4 3. 'or what we expect:from PRA-- y_f 4 MR. NOVAK: Chr too high with respect to your design 5 assumptions. For example the check valves originally were 6' or the-. equivalent system with check balances is based on 7 the assumption that those check balances would act as like g pass and failures with very low failure rates. You just 9 don't see that. 10 DR. LEWIS: Very low meaning a number? I didn't know that a failure rate-- 3; 12 MR. ROSENTHAL: Nobody uses a number but if they classi-- (J)' fled it as a pass or failure, then there must have been a 13 sense that it had the. sort of failure rates of other pass y 15 and failures such as pieces of pipes. 3g What, in fact, you see is that those check out values introduce all different sorts of scenarios that 77 13 may not have been thought about and it clearly indicated to me that they do need some care. yg DR. LENIS: I'm not questioning your judgment. I'm 20 trying to find out whether there is a criterion for the 21 optimal failure rate in the vales which the experiences shows 22 -'t l' v. 23 is exceeded and that's why they need to come down. /~% 24 The thing I'm--well let me start wi th that for a
5. J 25 moment.
Let me tell you what worries me is how people react Acme Reporting Company 12 O 26 6?n 48 A6 - - _ ~.
e U2 r 120 .) 1 to everything and shouldn't. That is to say some failures ? t () 2 are. expected failures within the design basis of the system. j 3 You should' notice that they're happening at the right rate (_.) 4 and let it go at that. 1 5 But, we have a tendencey to fix anything that happens o and so there has got to be some criterion when you say the 7 failure rate is too high. There has got to'be a quantitative 8 criterion, somewhat better than assuming that the failure 9 rate should be low.and we see it high because that's not a 0 10 really a criterion. Where is it? 11 MR. JORDAN: I have to react to that, I guess, My 12 view of the U. S. nuclear industry and its failure rate is 5 f' ) 13 that practically every failure rate is too high compared to 1 1-14 the capability of the piece of equipment. 1 i 15 Inherently and what a good maintenance practice will ) s 16 get you, which reduces your challenge rates to an accident 17 and it increases your reliability rate. 18 DR. LEWIS: I believe that they are too high. j 19 MR. JORDAN: Compared to-- 20 MR. WARD: Compared to what, Hal, you believe they're 21 too high? ) 22 MR. LEWIS: I believe-- G 23 MR. WARD: I think it's a good auestion. (~s, 24 DR. LEWIS: This is a copyrighted guestion. us' 25 MR. WARD: They can't be zero, you know. Acme Reporting Company an, su a. 1
g. 4,3, / 121 l' DR. LEWIS: You know what'I'm driving at. ,m ~ 2 MR. JORDAN: I'll compare it to the Japanese experience, ,_) 3 to the French experience-and to what we thought were the failure , -- 3 (/ 4~ rates in some of'the PRA's that we have done some years ago. 5 DR. LEWIS: So it's compared to the PRA's that we have 6 done some years ago, because, you know, there may be too low 7 a failure rate. 8 MR. JORDAN: Well, in terms of their plant availability y, 9 I.think they feel they do, they're spending more on maintenanc e, 10 Preventive maintenance than they should, but in terms of 11 safety and frequency of failures, their failure rate is very .12 very low. Their reliability of equipment is extremely good. / ) 13 I wish we had that problem, v 14 MR. WARD: Well we-- 15 DR. LEWIS: Why--I'm sorry. 16 MR. WARD: No, no, it's your question. 17 DR. LEWIS: You go ahead, you're doing just fine. 18 MR. WARD: No, no, go ahead. ID DR. LEWIS: I was trying to find a quantitative L 20 criteria and-- 21 MR. JORDAN: From the work that you did in the liability 22 and related performance indicators, the idea of having a low G 23 challenge rate, a low arrival rate of transients and to have /^^ 24 high' availability of safety systems is, in my view, a very %-]' 25 important part of safety. Design is another part. l i Acme Reporting Company ) i mu, uo u.. t L__: _ _ _ l
, t, ' .L,n meM a l 1 -DR. LEWIS: Low and high are not particularly in-i f rmative words. They are like beauty, they are in the eye 2 3 of the beholder and, you know, since you cannot achieve a ,x()- 4 zero failure rate, there has got to be a criterion for what it is,that excites you and-- 5 G - MR. ROSENTHAL: As you well know, we don't have a safety ~ goal. 7 DR. LEWIS: There's not a safety goal? g MR. ROSENTHAL: There isn't a parsoned of safety g systems' reliability from systems or trends other than maybe. 10 some work in trying to parse that. There is no safety 3 assurance program that is recommended from the end~ point side 12 (V'^) 13 and" investigatory hearing, as you are aware. 14 So we don't have firm numbers. Now on the other 15 hand, we have a fair amount of PRA expertise in the British. 16 We do have the ASP Program, SARA, PRISM, these PRA based 17 tools and we'can play games with.them, serious games. 18 DR. LEWIS: Sure. MR. ROSENTHAL: I know they're not event free. 19 .20 Pilgrim has had 3 losses of all types of power in 2 years. ~ 21 I.know that's not the 10 to the type number that is assumed in the black out rule and that something ought to be 22 7 ) done there. L 23 L And to do that without having a firm acceptance /] 24 %s 25 criteria, I do feel that, at least for the time, that we have to take the perceptions that we have or the assumptions, Acme Reporting Company (202) 620 4868 l (~
c, . 1 y),',
.s
,a f;4 Q $..... Ib f, 128[ m a -I m n y ' yl Quantitative t assumptions" we use :in generic issue; resolutions,.wath - f^ t m -,- ' /2 somefofItheIERAs that.we rely on,.we want'the kinds:-of data f3 thatEwe're;seeing-to.make some judgments on and not'be firm.; ~ Mh@ :,. ,t L.- 'Quihe often:the.OER System'or the.MPRDS. System, ~ 4 o J L5- 'weLcan't come.up withlfallure-rates, but what"we can do is to ,i 7 lsayfthdt underlying >some" issue resolution or underlying some '7; gPRA,:.was the' assumption of.a failure rate of."x.""and then'you ~g ' -look and ;you find >6 events and you say, what - is 1 the likelihood. q ~ I 9 _that thelvalueLof x:is the true value given-that I have f. ' ' 9 l' i That's> observed;6 and.seetif you wor'kit:in that sense. -_10 - 'she're we are.now. n 11:
DR.. LEWIS

That['s a1 very. sophisticated ' statistical -12. a. ' question which.would take us a long' time, but I have a'.funda- ~ 13 n., cli ymentaldproblem:with.usingJPRAs because' PRAs:are descriptions 15_ Lofisystems:as they!are,r not as they should be and'having' l6:
a--hopefully their descriptions of the systems arefas.they; u
!are'andinot as they1should be:. l ~ 17 i;.. For example, a PRA may assume a certain failure psA 18:l Frate for. something,: but in all honesty, it could be 10. times 19 .asibig and the safety of-the plant might notlbe compromised.. 20f H Then noticing that a failure rate.is larger than C 21 s .'is' expressed inithe-PRA.is not a good reason to go back and j - 22 'St .'3 try;to fix that component. 2 MR. ROSENTHAL: I'm sorry. If I have diesels--we 24' [ 25 are expecting diesels to perform as.92, .95 type devices. o, g c. L Acme Reporting Company a n.a. <.e. n
x + ~ y%., ..g V; y em l- "i + 4 T 134 r
q ',.
m vog, LEWIS:: When you say expect'to-be,!for whati ] 'f m 1 s .y ~~ ' i,. [..
2-
' purpose? 4 j a:c. 3 ;- 'HR. ROSENTHAL: It.'s used'.in our2 resolution of, ,4 ,flet's1say4A-4.4.-l Buried in'the: assumptions of newi Reg.1150-c which w'll form the' basis for.Jother. regulatory. decisions,- i a i .51 s-I 6.. .solthere i-is an ' agency expectation':n'ow incits,.either almost. +- '7' rulesmaking'orfmutual. rule' making, that diesels are' going;to-g uf
n e
p y 8'
Operate ~at better than.92,
.,95: type availability on, demand.. 9 Jokay. [ ;w , _ to' ~ That=.'becomes now-a' goal _for me.,;Now if I see ai .i %[
e plant ! that 's operating - at 1. 8.'ori. 7,- there 's som'ething wrong
. Ils y , L,;. l12 and they. oughthto fix the. diesels, and forget.about the. ~. L. 13 numbers at.that point and say what's wrong'with theLdiesel.. 14 DR.ELEWIS: Yes, but wait.'a minutej; wait'a minute'. - 15 .Now~you're describing the" system as in, support'of the regula-- f.
16-tory (process rather.than'as$in1 support of safety..
17' 'That'is to say you're-sayinglthat-the. fact that-the .95 for diesel' reliability is built into 1150 and built = 18.. .92 - into the resolution or whatever it is, +'.iat ' somehow makes 19 20 'it.an essential number for the safety of the plants,.but--I 21 think that's what you're saying. g 1 .g He's nodding yes and you're noddinq no. You guys 22 t ~ L23 ' settle that among yourselves. y y MR. JORDAN: Let me fix it then. WHat he was .%J ' describing in terms o.f a reliability of.8 number, we're really p 25 M L R Acme Reporting Company (202) 628 4888 'e
y mh k.-[g ' y o ^g. ..n y%, L : J' FJ'. g j' . ]' 4 t*,/, a m <n u y j,. 4 r
p. '
o r lA :. "Y, Y f,_, ^ W< ' Y': hY\\ 4 %, i'? < l* ~.
.Lus),
+ F .w: yR b 1 l .~ n G V. ".y Lwo."ried;aboutjthe.1's.;j'ILmean,..I'm sorry,'.butLwetgot plants r ,. g '[ LoufthereTtha'trare:notperforming;:well.at'allandwe're' i +~ E I, 5 ,r L3 -
w rried'about"those plants.
't l (- p [ -I don ' t[ doubt'l that: for. a moment'. D DR'. LEWIS: m, e p -l n 5I MR.-fJORDAN: And so if-we,get outlto where we're'in ,j .y R; 3 6; the.8]sland?.9's'fordiesels,thenthat's: nice,:there'ssome .,, hat-. maybe can be done, buti we ' re worried ' about the j 7 tuning 7 t e y ? guys'thatha've.2 diesels that won',t startibeca'uselthe air'sys-
8 1
'l tem lis'n.'t: functioning. -9_ q i DR.' LEWIS: Yes, bjt'he's worried at.8.and you're' y 10
1w rried at'.8Kand you; guys have got to compromise at.4 Lor
, e-ill. 1 ] '12- .5.' Y'
13 MR. JORDAN:
We wony about_ the same' things. He N L q[v.; 7pl madesacbad example. e. m
15'.
DR. LEWIS: But'that's'just'the. issue that I am y 167 .trying.to' address,-the issue of'using-. criteria that are s safetly. based rather thanLotherwise based.in fin'ingout how d c 3 17 : much--I'm'not criticizing, I'm trying to find out where they . '18. are., 39 MR. JORDAN: I understand, r wish we were in W l20 - '3-r le f gilding;a lily over performance,1but we're not. ?21 We're really right down at the level of being either acceptab3 e1 227 Q' l. or' unacceptable. 23 DR. LEWIP: When you gild a lily you kill it, you ct.t 24 l 4 25 off it's air supply, so it's not a good idea. i Acme Reporting Company 3, (202) 620 4888 d Y _.J :__' _ L _ L__
Qryy[37:is ' ',k?[ - + w]m .Q fht;n y y ',1'M y A.j" ' 'E ju .%pp d WTp ^ y y 1 't 4 hhm w. W j V l hI f ' (;:.) (. N 4 + xa y -y / > ./c s'],.) 7 Mc 's g;,,, I =) 4 ,/ fphe syJ j I r , r'f' LMRP. JORDAN:h Butiswe'r; e, i not even close- -we 're9not I N{A~, .?it -j t w 6:, .n: c + m [hM Leven cldseyto: gilding [the.Ililyis:what?I'm[sayin.. j#
bh
?23 Lik' W t.;DR.1 LEWIS:JDI'mLsolglad.- f V l 3_. 3 y n g-- p.x y p 7-q q.y 4 ' k b p. , ~47
MR;; JORDAN:01'm not'. 'Ifwish(wk were:. gilding the-h i
.~ .() 3 7. y ~ ' I wi.tsh:that<performancefwas"suchi.that We'.woreEtuning yj,j
1-(
15- ' 111 T* - - s ,o 1 Y-q-
e g
., ~. -s (. .. g "; t 1 .3
t

y; - 6 'ailittleLbit.; l.. > f, 1 yw -t sq 1 a., + , - g M M, .3 3
You'know,ifI.c$hmix"a!metaphorhthenf Y,~d

DR.1 LEWIS:
.p m ,3v pp* '8 jyouLw uld be killing the,, goose that l', aid:the goldengegg,Jbut-- e> J' , \\ .. e 'g ' MR.LJORDAN: .I.would like'to Vork somewhere else .M =v ,y ^
10 lbecause(. hat'siwhattitrs been.
~ t m g f, -I l k'. 6 Okay,.I have pushed;this-far enough. I-a DROLEWIS: 4 rg-T.f ' -11 p e m 1 21 see wh'ere you're coming'from. LJesse. j i 2 .a rO. .MR. EBERSOLE: 'Youtwere talking about check'. valves 13,' %[ :, . j ..,x,) i, -y .gf -
beingpassihefthings-andlI[canjustenv.isionLtpatjyod e
thinkingiabout5those"on.the injection < lines;thatYjust sit' }}1 1 s,, c. e q 15 - Z Lthere-with:noLflow'on'.them,5 butLthere are:some'vdr,ae'y active .16-check valves'likeiin)the main feed pumps andxthe.y face.some 17 -
18 -
very interesting potential problems,.-unless'we want to rub,out ! I the thesis of pipe failure, they f ace-the pr9blem of com'ing 49 j;' down on them like a cannon, you know,.without a. recoil L 20 . mechanism and shattering themselves and ' following on up and 21 ); .at discharging,.for.ginstance, a boilder into the turbine:hqll E '22 before the motor' operator valves can close and when you talk' / - 23 about valves operability, the valves are-put there as isola-24 tion devices for the hypothesis that you want to sectionalize l<
25

Acme Reporting Company l
a o,,. >. a... L,
N$5 &s w n e (7 r m q~ M wo TW x' lt, = 8 u QL.
q !
% ';?{ f $: @, w L1} bro,kenisystems, but weiknow very little:about how(theyLwill:.. g j q%g;. y ~ ?!N29 : iwork when thbyj,are e re'allyLloaded' andal'1 youristatistics = } Mb ~ w>. A,y s.. i v 3 .are?baseil{on; waving.themlbackiandiforth-with no.real:cdynamicj 4 yma< o,t-a n, ~ 1,' t 9 9 x, -o. r,,.. .ui EL load [ason.themn I lo~ok c) ' a h...,4'#, 4 pue. areas,jyou know,iofcmal-s m >,..% i .. N, 'ys
p ML az,g e
p ij ' ',
cc p
5.
?infoimatib;r as being ai.ser' ous problem".- hh, ^ You: don't.seelthIkingER's,becausewedon'tbreak S EG. 7- . Oipes,buthwehave'blindspots(l'n?ourfb'dsiness.: 57 .8 ' MR*. L JORDAN : And AEOD is1not'! presently.seeing'those y)(A ('@,ih md a S ENd spots.: L' 9 10 E. - EBERSOLE: : No', that's right. 3 .t V, n 11 MR. : MICHELSON : That's true. ] I DR.ELEWIS:.That's)got to be a mixed _metaphbr. 12!
h. -. 9..
1 .i o I.suggest that'.we move-on,-' G. ' ' .-y .qMR.'NOVAK: i K 131 , G v: L " $p.8 y%,.. v QgY[ ] R ' ' Q '. <g M DR.-LEWIS: Well,cI've got,to.let Dave [in, He had a s m.1 s. A mm g ','s o 3 9,'3 'G h '/15' Jh'is hand.:up, I think.. i m .y At 4.it % Y-[. 3x fli-
MR~..NOVAK:
.Oh, I'm.sorry. o . q m., p 7,., p1. [7 MR. WARD: No,. I'just. wrote.it.down'to myself. I'll g J y 1 9 m; m..3 bs/ appp with that. ] M yi ig. h j u i 19 DR. LEWIS: In'that case, roll 'n. o q W v. go' MR. NOVAK: Mark Williams is going.to take a few ) minutes and talk'about:the roleYof trends an'd patterns.
j. 21,
( @y 1: hesitate, but.I'm going to say that the '? 22 '..Q. gy' questions in 12'and 13 refer to either statistical analysis M . 23 1 . h.... '. 24 t that can be used to identify important events. These might ,.,n ,h 25' include.thonk$ hat are important to safety, and have an p,y c,. <i.. Acme. Reporting Company )j gy, V i G . m s u n.. p,' m~. ,.* o,
,y kih fph) ) j ,.( 'p. 128 4
u. '

()4 1
unnecessarily.high frequency of occurrence, n-f 2 . Then question 13,, "What.is the role of. trends and j q-3 patterns analysis in the evaluation of operating events?" ,3 p (f'f 4' Mark has been the Branch Chief of the Trends and \\ 0l 5 Pattern Branch and he's goibg to try and catch the thread of I
6 your questions and answer them.
7 MR '. WILLIAMS: I thought the best way to look at g_ 'it was just to take a snap shot of the things that we have 9 done recently and the thin @ that we're working on today. 10 Some of the things that I'll talk about we're '/ 11 actually working on today and we don't have any answers yet, 12 but they are illustrations of the areas that we're working .,9 / 13 on. u 3 14 When IJ1ooked at the things that.we were working 15 on within the last couple of months or few months, they did \\> 16 fall out into there arpas: general monitoring of plant-per- \\ ( L 17 formance, revisti.3 requiatory requirertents, stimulating J l 18 11.dustry and their programs and generally this is throvefb 1 10 acquired expertise ih an area and participation in. industry, csners i e 20 group: activities meetings md such and then applying the 21 operating, experience to actually making a difference in plant ] l 22 saNty and de.veloping and improving the data base system to LV support all this activity and operating experience. - 23 l (N 24 On the first one, on monitorir g plant performance, \\.,,) the trends and pattern program is the generator of all the 25 Acme Reporting Company (202) 628 4886 il_ _ __
Jgf 129 Ib 1. information. that the performance indicator progr m 'that Jack /m-(_): 2 will. talk about.in awhile has been utilized. 3 Back.in the. days of past development, AEOD was ,e n (_) ' 4 doing a number of different trends and pattern analysis to 5 try and pick up on what constituted good performance or ~ 6 tends in performance, so we generate all the information 7 except for., I guess one indicator, but most of the indicators 8 come from trends and patterns activities. 9 We take the 3,00 LER's and we code them and process '10 them externally'and internally in house and then we provide 11 ' coded information to.the PI program people who put together 12 the reports. (~~j). 13 We'also do plant specific monitoring on plant w 14 performance. When there is out lyers, we do detailed analysis. 15 An example of this would be, Tom. shot a letter off te NRR 16 a few months back about Vogel's performance and we had 17 meetings with NRR and Vogel's performance which ultimately 18 led to meetings with the senior VP of Operations at Vogel and 19 his staff. We went down there'and showed them our data and 20 briefed them and at that point in' time, their understanding 21 .of their performance was pretty much average. They thought 22 that they were average and based on our data, we got the ' message across that their performance wasn't average, 23 I think we made a clear cut difference in the r'T 24 (_J ' 25 operations at Votel, at least in the eyes of the people there Acme Reporting Company (202' 62d 4 988
&D 130 1 at the-staff by using our trends and patterns analysis to (,s1 2 characterize exactly where they stood in the world of new 3 Plant experience. 4 We also provide input to senior management for the 5 senior management meetings. NRR had some meetings to try to 6 assess the performance of all the plants and come up with 7 potential problem plants to discuss at their senior management 8 meetings semi-annually. 9 Then Tom and Jack and I went over and all carried 10 bundles of information, just about all of which was gleaned 31 from trends and patterns programn--trends and patterns data 12 bases.and we also carried with us some other program informa- [' 13 tion like the accident sequence precursor program information which 14 gave us a perspective on trenos and patterns of the more 15 significant events at plants. 16 Then we provided that input to that process and 17 I think we made a difference in characterizing what we thought 18 were the problem plants and we had differences of opinions 19 and we ironed them out in that process of meeting with NRR. 20 We also assessed performance relative to the 21 quality of L"R's. One of the programs that we have had for a 22 couple of years is actually going through the LER's to assess 7-) s_) 23 the quality and then feed that back to the regions and the 'N 24 regions send it back to the utilities to improve their LER (O 25 reporting. Acme Reporting Company 12OJa 628 4888 L_
cTc T' 131 s We're also responsible for abnormal occurrence 1- ~ ,\\ s_) 2 identification and we write the report to Congress and we 3 coordinata-that effort.throughout the staff. fm( (/ 4 In-the future, we're probably going to be getting 5 into more plant aging performance and how the plants perform O' -over their age and identify generic' issues as they go through 7 -their life cycle. 8 In the area of revisiting regulatory requirements, 9 there are a couple-of areas we're. working on currently. Just 10 recently NRR has asked us to research the basis for the 11 ASME Section 11 Testing Frequency for Pumps and Valves, so 12 we'll be looking through our data bases trying to see if (~ ) 13 we can glean an understanding of the failure mode that u-14 dominates this area so we can determine whether the test 15 frequencies are appropriate. 16 Other areas we're looking at are--as Ed mentioned '17 earlie r', BWRRPS Surveillance Requirements being changed and 18 the question was put to us, can we benchmark and then trend 19 the reliability of the RPS system using the nuclear plant 20 reliability data system and the INPO system. Bob Denning who is a Section Leader in Trends and 21 Patterns Section is in charge of that analysis and he's down ~ 22 ,.m.
1sj, 23 here to explain it later.
We also do studies of Tech Spec violations, trends 24 and patterns reports of violations of Tech Specs, what con-25 a s Acme Reporting Company $n %: l ? . m, n.... w I
yy \\ l
4{d ' 3 13g l
3 1_ stituted the majority, what's driving the violations from the h-root cost perspective and in overall plant perspective. Most 2 .3 . f them might be coming from fire protection. ,-x l '(,,) 4 When it'comes_to stimulating industry programs-- l well let me cover one more item that I missed on the 5 6 regulatory' requirements. f W ~ i . hat'we did in the reporting requirements, one of. 7 8 our trends and patters-reports for ESF actuations which is l ) 9 one of the requirements to report under the rule which will j i perform the basis for potential changes to the reporting re-30 quirements.- 11 i .12 With about 900 reports a. year'that we get on -( Q ctuations which, most safety significance, in our view,. 13 'd may not be necessary. 14 15 On the other hand, there are some things that we ig .would like to get, namely trend availability information and our branch coordinated the. development of that informa-17 3 1 tion with what the staff wanted. -18 Then we're proceeding along the lines of proposals 39 n how to give up some and give some additonal information. 20 On stimulating industry programs, an example of our 21 work there would be in the area of SCRAMS, participation in 22 !-) owners group meetings with the Westinghouse owners group and 23 thers and through interactions with them, I think we O 24 V stimulated their own tracking programs and they're tracking 25 l Acme Reporting Company 4202: 628 4898 w__.__-
. 0 7y; _ 0 132 j l 4 L 1 stituted the majority, what's driving the violations from the /m 1 () 2 root cost perspective and in overall plant perspective. Most i 3 of.them might be coming from fire protection. .sI,) 4' When it comes to stimulating industry programs-- s well let me cover one more item that I missed on the 5 6 regulatory requirements. 7 What we did in the reporting requirements,-one of -8 our trends and patters reports'for ESF actuations which is 9 ~one of the requirements to report under the rule which will g perform the basis for potential changes to the reporting re-quirements. 33 i 12 With about 900 reports a year that we get on ('] 33 actuations which,.most safety significance, in our view, V may not be necessary. 34 15 On the other hand, there are some things that we gg would like to'get, namely trend availability information and our. branch coordinated the development of that informa-37 tion with what the staff wanted. 18 Then we're proceeding along the lines of proposals 3g n how to give up some and give some additonal information. 20 On stimulating industry programs, an example of our 21 work there would be in the area of SCRAMS, participation in o2
r)
~ u) wners group meetings with the Westinghouse owners group and 23 thers and through interactions with them, I think we 24 ,V stimulated their own tracking programs and they're tracking 25 Acme Reporting Company I2021 628 4888
f1
^ 13 ?> j 1 the same information that we're tracking and this is kind of l C/ 2 a collegial effort where interaction with the NRC ir actually 3 used as a method to stimulate activity'in the industry. ,C/ 4 Generally they work along the same areas that we work on. 5 We also porticipar in the MPRDS users group at INPO 6 and there is partic.ipation in code groups and we just 7 designate some other members of our branch for that. ) 8 In applying operating experience to improve safety, j 9 generally.today, recently we have been looking at reducing i 10-event frequencies. ) 11 The example that we were discussing before was on l 12 threshholds and in our new plan experience, it was generally 13 seen that new plants had SCRAM rates about 5 times that of u 14 mature plants. 15 In looking at that, we decided to do a study and 16 we were also stimulated by the EDO program guidance in that 17 area and we produced a report on new plant operating experiences 18 which is on the table over here for anyone who would like it. 19 In processing that report, we have had utilities, 20 namely.Palo Verde and South Texas and others have specific 21 programs in place that went through all of the lessons in our 22 report and resolved each of the lessons. kJ 23 At South Texas, they actually had an inspection for about a week that actually inspected all of the areas in d 24. %). 25 report and went through them to see how well the utility was Acme Reporting Company 12OJe 628 4888 1
a .,7 134 dealing with each of the lessons in that effort. 1 () 2 So'that's an example of how we fed back information 3 back to industry fairly effectively for some new plants. In ,m.( )' 4 addition to that, INPO is going to be having a work shop on 5 this act'ivity. Actually Ed stimulated that work shop with 6 INPO subsequent to the publir uion of our report. 7 Other areas that we hs.ve actually applied operating 8 experience to improve safety are an area of SCRAM reports. 9 Another example would be where the SCRAM trend reports would 10 show that BLT is.the primary driver of all reactor SCRAMS 11 and within the BLT, the feed reg valves in the feed water 12 system would be the biggest signal component that you can ) 13 pin a large number of initiations on. n2 14 So, we have done studies of the feed reg valve, 15 by pass valve. We have studies underway of the turbine driven IG reed pumps and MSIV's using MPRDS and we're developing 17 lessons learned and good practices and the latest report on 18 feed reg valves will be transmitting to INPO and most likely several utilities wiii feed back: the lessons learned to 19 q 20 avoid initiation due to the feed reg valve. 21 And lastly,.in developing and improving operating i 22' experience data bases, there are several activities under y (s) 23 way. Internally the agency has a new data base that they're { >^N 24 developing called SINET and we're supporting the development N_). 25 of SINET. SINET is going to be kind of an integrated agency Acme Reporting Company 12021 628 4888
g,,>>
m. c
~ ~ 4 .. s.; f y m r 17iha g ~ j 4 + 1135: H ,e 1 m!
[
data base' for event!s information where you'll.be. able to igo 2 .and' pull out the-latest information.about events. N P 3L - We have also got a-- 1 V[ 4, 'MR.. WARD: Can you' give' 'an : example-of something 5-there that's--I don't. understand.what this SINET is.- q A ~G. . MR. WILLIAM: 'SINEf.righ't now,.think the only thing ] 7' that'is inJit Is L2R' abstracts.. [ '8 MR. WARD: -Okay. T.'l m' '9 MR '.' WILLIAM: But as.it developes,'it's, going toLbe. .{ 10 essentially, rightynow,.an event orientedcdata base'to capture j .d 11 'a lot,of eventsLinformation for the. agency, so11t's~a commonly j s, 12 a'ccessible data base. 13' . MR. WARD: So what sort of--who would you' expect to' 1 14 use?it'and how, just-a--I'm trying to. understand what'it does. 15: .MR. WILLIAM: It's for everybody, lit's for the-16. staff. Essentially it would be.our common agency da'ta; base n ,s .17. 'and I thinkJ that ifcpeople.would maximize the.use of it to. 18 the extent that they could, to meet their mi'ssion related r l'0 requirements, it might evolve to the point where we-could use '20 it for' events analysis.. .l b 1 21 MR. JORDAN: Maybe I could say tha't right now the' 'l l e x 22 . agency has, for instance, for a given event, we have the h. 23
5072 data file'which is on essentially a PC that gets dumped I
24 - into the NIH computer. We have preliminary notifications that 25 are paper copies about the same event. We have regional daily Acme Reporting Company , m,s,..... .i v ]
n..; . g 4, .1 130 1 reports. We have. staff notes that the NRR office, for 2 instance, developes. There have been the licensing event 3 reports on that issue and evaluations of those. They are
b
.CJ 4 all separate, they're not stoned together, so the proper 5 data network is--I'm sorry, the Sinet, son of proper data 6 network,'is an agency wide system that will sew all of us 7 together with software or hardware to make it accessible to 8 so that you can get the sum of all of the parts without an 9 event, including the inspection information without a particu-10 lar. event, so that it.is available to everyone in the agency, gi ' ' altogether. 12 MR. WARD: So if you identify an event and you're > (9 g3 interested in an event then you can go into this data base LJ 14 and get all these documents or find out what they are. 15 MR. JORDAN: That's right. 16 MR. WARD: That's what I was trying to find out. 17 What if there is some other parameter other than the name of 13 the event? Are there any other ways to go into the--can you-- MR. JORDAN: The data base is intended to go across 39 20 the agencies activities and we're starting with events type 21 data and then starting with unit data and so, for instance, / 22 the unit file, you would be able to get by searching on a ) particular unit the myriad of information related to unit ) 23 i and then the events file, the myriad of.information by event O 24 d 25 and then other files that are yet to be developed in the same i Acme Reporting Company (2026 628 46H8
> /s) - g, 1 kind of a fashion, but to integrate the whole agency's file, f (_f 2 MR. LEWIS:,Actually my question was along'these 3 lines because right--is this a commercial data base system /~j% - 4 you're applying to agency affairs or is this a home grown ( 5 data-base system? g MR. JORDAN: This is a contractor that will be 7 writing the software, put it together. -8 MR. LEWIS: A contractor who is a data base expert? D MR. JORDAN: Yes. It's IDM-SR which is a-- 10 MR. LEWIS: Which-- MR. JORDAN: IDM-SR. It's a color net product. 11 MR. LEWIS: Commercially available? L2 () 13 MR. JORDAN: Commercially available. Large main v g4 frame, state of the art data base. 15 MR. LEWIS: And these are real data base people 16 because-- MR. JORDAN: Yes. This activity is being managed 17 out of the office of administration. 18 MR. LEWIS: I wish you hadn't told me that. 19 MR. JORDAN: AEOD is not managing it. fe're pro- - 20 . 21 viding-- MR. LEWIS: I understand that. But you know the 22 s k ~) fit the custom needs writing of large data base soft ware to 23 24 of an agency is a non trivial business. V(^T 25 MR. JORDAN: Yes. I Acme Reporting Company (202) 628 4888
ft 13g L 1 DR. LEWIS: And the people who do it have to know -T 1 (_)L 2 more-than just what is available. They have to, in the best 3 'of all worlds,- they really have to sit around the agency and f^'y l (/ 4 interact with all the people and watch the way it sorts 5 before they even know how to beging writing a data base. 1 6 Did that'actually happen? 7 MR. JORDAN: The. program offices-- 8 DR. LEWIS: It sort of happened. 9 MR. JORDAN: This happened. The program offices, 10 at the Office Director level, are involved now in making sure 11 that, in fact, the putting of staff resources into the end .12 product will be totally useful to the entire staff. (p)x 13 So Vic Stello has our attention to make sure that ~- 14 we, in fact, are getting the service we need. 15 - DR. LEWIS: With all due respect to the Office of 16 Directors and not a bit more, they may not be the best possible 17' people to judge the-- 18 MR. JORDAN: You misunderstood me. The pressure was 3 10 Put on the office directors to put resources into it. q 20-DR. LEWIS: I see. 21' ~ MR. JORDAN: I am not a data base person and I 22 wouldn't pretend to be and so I got people that are extremely 7s Q). ?3' good at it and those are the ones that I am putting-- j ) 1'N 24 DR. LEWIS: The only reason I'm pressing the point, ) 1 25 is that I know some data base people that I have learned to Acme Reporting Company (2026 628 4888
f(( 139 l 1 appreciate. It's not just a matter of making sure that you ~ t,/ 2 can refer to things under 3 names. There's really much more 3 than that to making a good data base. 7S. (_) 4 MR. JORDAN: Yes. 5 DR. LEWIS: Okay. We will see, won't we. 6 MR. JORDAN: Yes, agreed. 7 MR. NOVAK: We have got to move on. I think basically g Mark's points have been covered. 9 DR. LEWIS: Now you'll have to talk twice as fast. 10 MR. NOVAK: Again, there are points that are really 11 important and I wish that we had more time. 12 The next issue is on human factor studies and Kathleen Black will talk about some of the studies that she x -) 13 14 has been responsible for regarding human performance. 15 You will see that the work that we have done 10 is not completed. If you've got some questions, do they examine drug related events and actions involving personnel 17 is injury. We'll answer those questions, but first let Kathleen kind of sum of what kind of work we have been doing in AEOD. 19 20 MS. BLACK: AEOD's divisional safety program begins by reviewing individual events and virtually all of the 21 organizations that Tom Novak supervises have some kind of 22 l em ) t l %J 23 input into the human factor studies. That is to say that I l or my people would look at types of human factor events (~T 21 N/ 25 that is to say, events involving people doing the wrong-- Acme Reporting Company (202s 629 esas
7@//;;g;i g e
n -
7

7

' 4. ' it ~ ggL l,., i 7 n, > t t-- ,4 Ibi ./- goingjto;ths wrong place?to perform;an action, eventsJin which 1 s. _ g r Lprocedures were deemed to b'e theEcause'or o'ne contributing 1 cause' f 62 tc g 3; jp ' ffthelevent whereas-other groups, Jack Rosenthals,.inaterms -3L ~.. L(( of his case study. reports,fwhere he.is looking-at_a smaller c p .e numberTof evetns'in^ detail',(would'identi'fy the contributors 5 9; 4 0.f-human < factors'to' air.. system problems, RHR problems and' 56: i. Lin'termsLof the case studies'or special. studies that are'done
7 by Mark, William's Trends and. Patiterns' people, _-they_ will also
. 8 be-identifying _ areas in which either human error or._ human-- p r <9 l': Jgood human practices turned out to have an' impact on1the 0 s perceived performance.!of'the-particular reactor and, indeed,' 11 4 12 :_ Lin yet another case of. performance indicators, also have some content of human performance and are tryingito developea 13 performance indicator that would speak!.to such human factor 14 15'. items'as training. ta' - I.could discuss, in..some more detail,_the' findings, r
17 but. effectively, what we try to;do is have? people look at it.~
from two-different view points.'. 18 u One-is just looking at th'e'peop1'e' type events'and 10: the other:is to look at the events ~and:the consequences of. -.~ 20 themLand' determine to what extent' human performance was
21 4
associated with the particular event. Dm I would be happy to talk to any specific questions, 23-121 if_there are any. I can. address also, to some extent, not necessarily what we're doing with drug related items because, 25 Acme Reporting Company L2026 620 4888 y- .__-u---_____nL_--..--._._._,_ .--_..-____---._,__._--l--
f7 141 1 indeed, we' haven't done very much and I'know that Tom will s (_,) 2
. speak to that later.
3 But,. MSS, in their Safeguards and Transportation ..p
(_.-
4 Group keeps a log or count on durg and alcohol reports that 5 come-into the Division. Most of these reports are in the 6 morning calls.- They.have collected data over a 10 year period ~ .7 and fijd out that the number of reports have gone from about s zero from year, let's say from-1980 up to maybe 35 or 40 t 9 or perhaps this year it is 60 reports. l 10 These are reports, they are not individuals, there l are.inor individuals involved than that. The bulk of these 11 12 individuals are not guards and some of the reports tell q () 13 about, as the result of off site drug abuse and essentially v we find no input in LER's that speak to the use of drugs or 14 the association of drugs with a particular event. 15 DR. LEWIS: I didnt realize that drug and alcohol 10 p7 abuse fell into N!!SS. That's news to me. gg MS. BLACK: Nell, I don ' t knme that it necessarily falls into them, but they publish an annual safeguard summary 39 20 event list. They're interested in theft of materials. .They may have gotten into this because perhaps guards were perceive 21 to be a likely problem in terms of alcohol abuse or drug abuse 22 [) %) on back shifts or that and since they have kept this log up 23 to date for so long, there seems to be no reason to have it 24 done by any other group. 25 Acme Reporting Company (2021 628 4688
yp} - 142 '1 1 DR. LEWIS: I see. It's an agency decision. I ,3 q) 2 just didn't know. 3 MR. NOVAK: I would like to, at this point, perhaps u,$ 4 Jack, you could tell us a little bit now about some of the i-5 work that you have been doing in the human factors and the 6 performance indicator area and we'll button it up. ] That is, for example, some of our ongoing develop-7 ment work so we can kind of--we'll try to give it, as best 8 we can,-the inference we're involved with which is going 9 to be in effect, 10 MR. ROSENTHAL: Okay, I'll take two minutes. 11 I wanted to make the point that in our AEOD studies, 12 we look at service water or running cooling water or air 33 g4 systems, et cetera, vertical cuts of the data amongst many 15 players. The PI Program is, in a sense perpendicular, to that 16 where you're cutting across many systems or areas at specific 17 p ints and my goal is to jusc to--it's the same data, is to 18 have these two ways working, I think it's comparable to each 19 ther. 20 We have a PI Procram which we briefed you on and we 21 sent you the latest PI report. We're monitoring right now 22 -d 23 7 PIs. That data is used as a tool. We don't write plans, but hopefully it gives us a--it's goal is to give us an early A 24 V 25 objective in the performance, but half of the PI_ sections Acme Reporting Company t2021 628 4886
a.3. a
s., *
.c ty 4:- c i .;w' p ; y,:., w pp qr i ,f w$ '
g.,
]M [1d 8_ w- .W D', ' 1: effortlis'taken'up writing 1those; reports,1 quarterly reports, K.M<. l 2_ - and thecotherl half of,the' activity is developmental. y/ 13 We are.having 6;pIs indicate.as related really to p- , d. g. 4' ihardware What we would like toldo is go to more progromatic ~ r-- 6 5 indicators?because,welthink the best programmatic indciators. ? '6 'would: lead the. actual hardware fault,.and we have been groping t + i b . t 5. 7" around. 8 .We wanted to look at training because we think that S e. 9-training is important iso-we---how'should.one 'do this? : 10'- We took the pass--fall exams on requal: exams.and 111-attempted to correlate them with.LER reports of operator' y> .. c ,D
12..
errors. 'It.'does not correlate. i ..W [i 13 DR.. LEWIS: The same thing is true of.that that ~ 't doesn't correlate. i f _14 .you tested, J 15 MR. WARD: Does that mean we should begin to thump A y. forjuselof this~non correlation? p 10< 17 DR. LEWIS: Well, we'once got a--yeah, we should 6 !18 use it. -F !'h [19 MR. ROSENTHAL: And maybe it.shouldn't surprise b> 20 .us'because one is measuring how he performs on theoretical i:n y i 21 ! test'.and'how:he performs in a. simulator while the LER's, even I22 ! presuming they're perfectly true, really report Harold, that ".O: P J23 personioperates'under low stress, normal. conditions, and l24 what we're really concerned with is how someone will--how an 25 operating organization is going to operate at 3:00 o' clock on Acme Reporting Company (2021 628 4B89 ~ n_.il ___ _____i ___1____-___.._______ ___.___._____m.-_
r 4 tfd 1 Idd l s 1 that shift when there are multiple mishaps taking place. L () 2 MR. WARD: Does that mean the individual operator exams aren't' worth while? 3 e'N MR. JORDAN: Say no. (y 4 MR. WARD: It would seem to indicate something 5 g approximately like that. DR. LEWIS: This is an example of a good dog not 7 8 barking. MR. ROSENTHAL: We're still interested.in a train-9 ing indicator or indicators and we're groping around for--it't 10 11 interesting. 12 We are attempting to and will be able to attribute (nv) 13 when a cause comes to events. Whatever decision we have, 14 whether it be trip or a safety system actuation or failure, 15 a signficant event, to try and say, was this due to an equip-16 ment failure, was it due to maintenance, personnel error of 17 'the operating crew, personal error of others, personnel errors 18 of others, or we have another bin that is design fabrication 19 and another bin of communications problems within the organiza-20 tion. We find that everybody is doing the_r own thing, fine, 21 but those are all, obviously, human oriented type attempts, 22 first level attempts, top level attempts to grapple with the ,3 V 23 problem. /~N 24 MR. NOVAK: Let me make a point. I was hoping-- \\_) 25 I really led Jack along. I didn't expect him to get into the Acme Reporting Company 62OJi 626 460s
6fp L 145 i i o r 1 whole discussion of performance indicators. I just wanted to ' __) 2 kind of pick that part which was human factors related and ( 3 which would be a discussion on human factors while .,m f (_/ 4 Kathleen had the floor. We'll do better the next time. '5 Before and I want him to catch his breath on l 6 performance indicators. That is an issue 7 I really feel that this is a challenge to us in' s AEOD. I know Carl and Jack have looked at the operation of 9 AEOD. We have looked ard at the equipment. We have looked to at what we have been doing. We've been looking at counselling 11 we've been looking at screening, a lot of the work which we l'r do is involved with machinery and the returns are not as good 13 as we would like, (wj In other words we are working as hard as we can, 14 15 but we're not coming up with as many nuggets as you might 16 have expected and maybe we're not doing it right, but we .' ind of feel that maybe that we just--maybe that cut that g7 .ig we-have.used to pick out machiner/ related events and so forth we've done a fairly good job but the other is going to be gg 20 tougher, but they take a lot more resources. The thing that 1 I am aiming at is we've got human factors--there is that humar 21 33 f actor part of the issue thathas got to be addressed and we \\ (.s4 ud 23 really haven't spent that much time, Kathleen has done a couple of studies. We did a gg 34 V wrong train, wrong unit study, but you got to be careful 25 Acme Reporting Company (2026 829 4868
3 L;/D-146 i 'because you can only look at what's reported, but we did and ,m. l ( ) 2 we saw a number of cases where people would go to the wrong L 3 unit, they were supposed to go to Unit 1 and they would go to (9 / 4 2. They would put Train A and they would go to Train B and-i 5 we try to learn from that. Should you be labeling? Should j t l 6 you be writing your procedures in a certain way? 7 _But the response to that, it went to NRR and they 8 are following up but we may have just seen the tip of the i { l 9 iceburg in'a sense because only after we-talk to them OER { 'l l did you know that the guy went to the wrong unit. If he 10 i 11 didn't trigger that level of the 40's, you don't know because i 12 you really don't know what fraction of the time he's making ()O a mistake of that kind. 33 J s. J You look at them and you say, they ought to be 14 J 15 presentable. We ought to be able to educate.the people to be careful and follow the right procedure and go to the right ] 16 l train and go to the right window. J 17 DR. LEWIS: If they cannot prevent it, but the 18 \\ probability can be reduced. gg MR. NOVAK: That's correct. And we ought to be 20 seeing how they're training their people, 21 DR. LEWIS: Quantitatively, 22 f ) fir. NOVAK: Similarly we looked at procedures, just 23 what have we learned, how do the procedures affect events and ,7 24 O we've got a study right now that I'm looking at. 25 Acme Reporting Company a e202 6284888
gl. jf7 .c r 1 We've been looking at events over the '84 - '85 time . O freme--xeth1een, why don't you suee take e few minuees. 2 3 MS. BLACK: Sure.
4 MR. WARD

Could I raise one point here though be- .5 cause'I think it's a gcod point. It's something that Glen 6 Reed was always raising, in a slightly different way though. But it seems that so many-in the agency's programs, 8 I guess 'the industry's programs are--related to human i> factors, are directed toward attempting to reduce the rate of g ' human error,. training, better procedures, exhortations to gg excellence, you know, that's one of the emptiest, I think, g; 12 but-~ 13 You know another apprach is to better define the 14 haidware and soft and administrative systems to accomodate 15 human neror'which is inevitable, of course. I mean, we ~ 16 argue about the rate, but I don't know that enough of the 17 programs'are being really directed toward that, which probably 18 has a higher pay off and the reason I bring up Glen Reed is 19 that he's, you-~know, continually says, well let's redesign 20 or design the system so that they're more forgiving and don't 21 put so much burden on the operator and attempt to drive the 22 operator toward perfection in performance which is impossible. zi I don't know--maybe it's not an appropriate comment l 21 for this meeting, but I think, you know, as you're looking 25 at experience, you know, it seems to me you have to keep that Acme Reporting Company I,9Ji 6 k' U 4 6 8 8 w--_----_________._______.-_-_.___-
V ids '1 sort of thougbt in mind, that humans are only so good and l (_,). 2 you probably aren't going to change them a whole lot. i I 3 MR. JORDAN: Maybe an example of that would be en l,S s 4 the difficulties that operators have with certain feed water 5 systems under certain plant conditions and that just tells you o that it's not the operator's fault really, the plant has a 7 sensitivity in that area,'it ought to be fixed and the .8 plants that have reduced their feed water trips during that 9 time frame have changed the valve or chance the control-10 system or have done something fairly dramatic to fix it and 11 that's the feed back process I'm talking about. 12 When you have a problem, you end up doing a fix () 13-and then you verify that the fix works and reduced the w-j4 rate of that'particular current, so we have a number of 15 plants that have done good jobs on fixing their feed water 16 systems, as an example. 37 DR. LEWIS: But Dave is making a very important 18 point, a deep point which is that the people that you get to do anything, even to serve on ACRS are a certain class 19 20 of people, you can't get a better job out of them than you 21 can get out of them and it is the job of the hardware designer to design something that can be operated by these people. 22 .?s) I always tell people that when I served in the Navy 23 in World War II, I was really impressed by the fact that r') 24 s-the ship was a complicated thing and the Navy had put together 25 l j Acme Reporting Company (202p 628 4888
l{fi 1 ~ 149 1 .a design, a set of training procedures that made it possible 'e'~; !%) 2 for a bunch of dunderheads like us to run it pretty effectively, 3 but that wasn't because we were~made any smarter, but it was ('3. (,) 4 because the ship was geared to a relatively low level of 5 human performance and there's a tendancy among, certainly a advisory committees, to always recommend several things. 7 Your research budget should be tripled and people should be 8 smarter than they really are and that will solve a lot of 9 problems and you can't do that, you know, you really have to 10 make an assessment of the capabilities available and make 11 the system work within that capability. That's a more 12 profitable route than trying to improve people. I think that ( -s) 13 is what you were saying. x_ > 14 MR. WARD: Yes. -15 MR. JORDAN: I have one comment with regards to that 10 Through the performance indicators, we have seen some objec-17 tive evidence that there is a gradiance in how well plans are 18 performed and so the high performers mect the goals and if 19 you get the median up so it's more like the high performers, 20 then you can improve safety. 21 DR. LEWIS: But you know there's that old legend, ,-s 22 half of them will always be below average. t G 23 MR. JORDAN: Yes. But I have to say that the i gracient across there is very large. r's 91 ~ (_) 25 DR. JORAN: Well, you know--I remember one of the-- Acme Reporting Company (70,i 628 4888
y-g %GW ', c N -w 'N< J 0 1 li O ' p9 ~ one'of the L,, 1-(around.this very' table--no, we had another. table, \\ k.. ' l 2'. ~early studiers of!the Three Mile accident said,-in'the' f [. L 31 s nuclear 1 business. we.can ' t ' af ford that, ' having handled: the ' ! f. 3 M I be'iow: average. O 5 MR. JORDAN: I agree, I understand. .n y MR. NOVAK:..I.;think our point was that thisLis an 6
7
-area.that we, as an office, have'not spent much time on an~. c, L 8.1 are'a which.should be directing-- 9. DR. LEWIS:. But~ Dave's point, just to repeat was N M 10 that.'that may be.true'but one should also direct some, 11-attention'to our work to make the system work because you ~ Lmentioned' people-going.to'the wrong train. The ground;is 121 i .' 13 ' full'fo pilots who have fettered the wrong _ engine and_they ~ ,14 'had'a problem. They're not-guilty twice, in general, but 15 that's.the sort of thing that is, simply going to happen 16 no matter how'much'you do'in. terms of increasing peoples 17 sensitivity, there.will.'be a rate at which it happens. . 18' MR.,NOVAK: Ildon't mean.to prolong this'or_even 19 . sound argumentative, but we see a good turn over and I think 20 what we're saying here is that we determine which goal we 21-think is achievable and I think now, when you look at the E-~- 22 rate and we'look ab the causes of events, we see a good 'l.h: 23 fraction of them are' human error related or there's a tie.in " 0' - 24-error. 1 25 MR. WYLIE: I think this is an area where you've got A;; Acme Reporting Company ) aos,saa.... = _1 -)
m, --- 9 .t 151 1 i a limited' amount of' work but you're accelerating our r"V 1 !%)' 2 efforts. 3 MR. NOVAK: My point being, you've got to pick 4 the right kind of a program. In other words you've got to l 5 define where to look. Too often I think people just start j l 6 out there and get into it and when you're done you don't' 7 really know what you're going to do with it, so I think we're 1 8 going to have to question what can we learn from this study, j 1 9 what are we going to try and answer,'see if we can construct a study'along that line intending to answer dif'ferent ques-10 l 11 .tions. With Kathleen and Jack and Mark, we kind of tried 1 12 to.put this together and I think it's one of the things that ( t-13 we will be looking at hard this coming year. 34 15 MR. EBERSOLE: When you're'doing that, do you study the burden in a relative sense, put on the operator, in the 10 i context of what feeds his brain,'what can he see which is right 1 17 versus what is wrong, what can he hear, can he take an action ig which is, in fact, reversible because he sees he did it wrong, 19 20 the equipment tells him that. j In'other words, you look at the burden put on the L 21 . operator in a differential context, plant to plant and try to 22 Q) develope a kind of a standard of excellence for operators 23 which ought not to be exceeded, you know, by the demands of /~N 24 'Q the designer. 25 Acme Reporting Company 1202s 626 4888 w
-- m-- r 4D ,1 152 i What you're going to see occurs after about 9 months. You o(,) 2 see a curve, but before you get, it will get in within a yeer, 3 but you don't see a lot of early reportings. - 1 (,) - 4 Completeness goes to whether or not those components f that we think should Se covered by NTRDS, did they. fail and are 5 6 they reported. 7 The way we check on that and it's a very limited check. We will look at a sample of a 100 LER's, each quarter 8 9 of the year so for over a year we look at 400 and we kind of pick out those to get a cross section of the utilities and to 10 it look for component failures and then we will look at the 12 LER's and then go to the NPRDS system and see were those (Jl 13 failures reported. L 14 Well, there's a question between ourselves and INPO 15 and the utilities as to whether or not that was a failure 16 reported under the rules of NPRDS. There may be a 10 percent 17 difference between them. I don't think that's too important. 18 The question is though, just how much of it is i l 19 getting in and again we see a certain response. I know in i 20-the-lst year or 2, we would say, judging around, two-thirds or 21 so were receiving those. 22 MR. f1ICHELSON: Is there a tendency to tendency to l (s-x skew what they do report to make sure at least they report 23 what's in LERS and kind of forget the rest? Is there any danc er (') 24 \\J 25 that's is what is occurring? Acme Reporting Company (2021 628 4888
a. f ,,, [] 9- ,o A # '; ( nl/ ' q] g g y_. g f y j 4 p..-., m j 4 , g,
p yi.[hx
[ i [ .k h d 1 .. o 1 4 -
u..-,
P MRbN'OVAKt. Well.the~ answer.is,fwe'8 re trying-to- ' '; L 4}ll [;!p 9 ' wet aveito 'get{out:in th~e. field. 1Infother>words, we"can'fth h p t
,2.
4 / e f3 yprovide?that kind of. understanding lwithlour' data.- g.g , :n 1)[ 4:
What we.do'is kind of structure the problem then

i. .e
q ':

u: g g 5 ;.
.more fieldJvisits,.'more hands-on experience watching people- ~ ,6 ? perform.those? activities..will'givet us some:finsight. .I dont:want:to mislead'you. ;AEOD^isi 7 MR.' JORDAN: I8 . n o t.. likely'to dojthat. You know,-this.is.a h'uman factor' ~ 's o g c m, n. 4 9 organization:and'they'are dealing:in'the field.of inspection, )examirling91'yo'ut's and the. arrangements, coloring.and those-
fl0 -
a .11 kinds.of_. issues and.' soli:- would. expect they" woul'd sbe' getting - ~ 4" the plant by3 plant' feed b'ck inJterms of those: kinds ofL a .12: '[' \\ 13 improvements, but,we would;be looking more at.the.' data: . #j; 34 - !with.enoughlin plant'look so that'we're-nottin an ivoryntower. D t M k So I wouldn't want your. expectations to be'beyond2 15 16 what we.'re'r'eally going.to deliver. We're really'looking 1 17. mostly at' the data. e t J j _MR..WYLIE: You, identify the problems? N 'MR. JORDAN: Yes. gg MR. NOVAK: Jack,'-I'm going to turn back'to you-20 c O'<' now in terms of, performance indicators. q 21; a. DR. LEWIS: llow are we doing on; time? >p y, -- 22 '\\ I [ Y 3 ?.: 23 - MR. -: NOVAK : It's now 3:00 o' clock. It's 4:20;'I. g think. The next area has to do with AEOD;. You have a ques-L s +3 tion, "To.what degree does AEOD track the NPRDS system? What 25. } x l b. Acme Reporting Company 7 - 120 J) 628 4868
r j !V i d{ {J;? f M, cfp g o qrit p y(t.rgx? t 'g ! 4 g .L g d >~ si o 1_ analysis is done?
~,
2 We're tracking it thd?way we have been doing it for n /,, 3 Lthe last... 5.or 6 years',and we're about to, in a sense, we think wb f 5 s g - g. h [4 we have reached a idir wherehehavetomakesomeverybasic s. 97 5,. deci.sions on just where is NPRDS. m, 2 6f^ Back in the early '80's NUPRON took it over from l _ y. 7 sortli.d f: a 3 group operation, one-third being Michigan and ' l,' y w8 ,,I;th; ink'one-third being the Southwest and there was 1 O. <c). T 9 committee that resulteds e 10 LINPO toUk it over arsi it wad a major commitment. 11 As I said,.after they took it over, they expanded and we \\ { 12 were advisee that we would be covered 6under NPRDS. J. (m) 13 For example, just to show you=the breath of it. 14 Hnen Hope Creek leas widened which WGs one o f the last 'BWR's, f ..s l\\ N. s' y,' 15 they submitted over 10,000 engines:#.)ng recqrds,,that. is they + ) g identifibd some 1(4000 components now that are in NPRDS 16 ,p b 17 and it,just gives you some feeling for the spectrum. You f so t.here 18 got some other people that are going to submit 2,000, is a wide variety of industry participation. 19 .f ,i ' But to get back, we heve beep, t racking timeliness, s r 20 1! vi 21 completeness and quality. Let me just go through these very g.: y,, k ('. ~ 22 quickly for you. ) fC.L).s -).' ctt 4t 4, Timeliness refers to, to how long after a component ,1 ' SV / 23 \\: ?/,;, DN IJ 2/, fails does it take before it shows up in N?RDS and it's been h p/ A. p 25 rather slow. I think we see now that it sort of levels out. Acme Reporting Company 12OJi 62H 4808
/ c 1i55 'l MR. NOVAK: My' understanding is-that INPO has y
i.,l :
2 directed ~the utilities to make sureiyour report those failures 3 Lthat you report so that-- .a ( J,. 4; ME..MICHELSON: Even those'there's kind of a printed 5 requirement or desire,'you're still only seeint about two-thirds 6 of the-- 7' MR. NOVAK: Something on that order. 8' MR. MICHELSON: And we don't know how bad'y they're 9 doing where there's no LER been:. written. You see, all of ur single valve failures don't necessarily require LERs and 10 11 yet we use the NPRDS, perhaps use the NPRDS data as an indica-tor of now many valves are failing, but you don't have a good 12 feel f r the population or the percent of population being I) 13 reported'and so on, you don't know. g4 15 .MR. NOVAK: The way INPO does it, when they do an ass'essment, they go to the maintenance work request and'they 16 wil-1 take out a certain set of variances and they will identify .g7 18 all the components that were replaced ~due to failure and then 19 they will test that against the NPRDS, so they go into that ~ 20 kindtof an evaluation. 21 MR. MICHELSON: Have they ever told you how they come out doing that kind of an evaluation? 22 7-
r/
MR. NOVAK: Yes, they have. 23 MR. MICHELSON: What kind of percentage did they /~N 24 i (w) 25 find? Acme ReporHng Company l a n, u.a...
g4g% g eq yi? c, f, 1 ~~ ' ~ ~ ,7 ; c < J' T-i Vi y + j ~ 7 ; g.p.. ~; 'l ~ ,c y x, .s _t. }m. ;:[ _. g i;- %fLallicomponentLfailures that could'.have.been reported?. 2' E r + _'j \\
TJ o, _
.MR. DENNING:- : It 's out L best indication ~,- based ton. wa -n 3 1 - I'tl :' f i -' 3 -: samplings, mas'to.Lwhat's-getting" int'o th'e' data base an'd'that 3: , ( f:: '4 .gets inlthere,las Ilmentionedfearlier,"over.a1 span-~of time m 5 'thatiranges!6;to 9,:monthc. 3 ,G A. lot of-cases', c it 's'~ not thel system where
30 days---
J 7; MR. MICHELSON': 75-percent'.- g, MR., DENNING: --iti's ' reported to 'you. .MR. MKCHELSON: 75 percent"'wcald.be very good. 9 That)would be more than adequate.- '10
MR.1EBERSOLE

I don't know about that.-.Let meLask .g1
12 lyou.. bout--you kow.the Salem case was: 3nevitable, it.did take;
~ 'a while'to: materialize, but thereThad been-a long' history: .13
f. single) track ~ accomplishments at that;critica1Efunction 14 J s

15 -
prior;to-the actual duplex faiure. n D.you think we're getting follow ups'on components ~ 16 q- ' of the SCRAM breakers now to preclude a repetion of Salem-17, 18'
with.perhaps more, spectacular results?
MR. JORDAN: It's clearly better than it was at -10 that time. 20. MR. EBERSOLE: It's sf'.11 pretty lopen. 21; t .MR.< JORDAN: I don't think we made up'our mind whetl-er ; ? j.p'.. -22 U it's really g d'or not in terms of the method. It comes back f - 23: '2C to Dr.. Lewis' comment, how much is acceptable. The greatest benefit we see now is that J25: p Acme Reporting Company p Ii 4 sacas are asas S:_ ___ ___ _ 2__
p.p_-- a 7,r M. - ~ --~~-~T--'--~~ 4 hh ' ,A y, l,, ~ 1 ^ 1 1 g/m. m, @l y ly " wy .y:. m' gv gQS:( a. 1 17 MR.... DENNING: Tom' Denning,:AEOD... ;Their(figures b J21 '(Iding[it thati.wayidon 't id'isacjrele..very[much Eitli ~our figures. ' . + The'.wouldisayL75.fto 80 percent.and weJsav'65'to'70: percent. a 3; y L pg. , SQ [4 ' So there -is -no. substarccial agreement. m ~ 7 5: 'MR.;MICHELSON: So you1mean'you think tuat:75-m' # 26
percent'of*all the.fallures'are gettingJinto the NPRDS,1is.
o a ,e '2 6, 7, thatiwhat.you're saying? l. , g: 8 ,MR.-DENNING: That's what was said. l -9' MR.' WILLIAM: We're runningsabout'60 to 70.. percent.- 1 =10 MR. MICHELSON: Well, it's inconceivable to me that-J x 11-some ofEthese utilities are only submitting.200. reports a 12
year are reporting 75. percent of what'sf happening to them 13' ;
.a'nd:another. utility reports 10,000 things to-- ~ 4
14'
.MR. NOVAK: No,ono. ~ L + .15 - 'MR.[MICHELSON: -- -NPRDS, is ' that also -75 percent? 16 MR.. 'NOVAK : - >Well",'.that's engineering records 17; MR. MICHELSON: Yes. 18' MR..NOVAK: Excuse-me, that's not component failures. 19" MR.:MICHELSON: I'm'sorry,.you did say engineering -j l l 20. records, I-had forgot,.yes. The component fai}ures is.where 3 .n; 21 'we are getting our statistical data from, of course? . j f I yt- .22 MR..NOVAK: Well that was part of the, original /d a 23 1 design ofiNPIsDS. There are roughly 50,000 component failures- ) i 4 i in th'e data base. ) K f21 t -t MICHELSON: But you don't think that 75 percent 1 ,.) 25 C M; Acme Reporting Company - l uan 6..... .-._-_n--._-_-.--___--.___-a
g4 y ~, 3,j k : L: a - + .u a.2 s yd ii;j k i ~ e L1. some of2 thes'elnow are going to have'a? system that they.can go. ,J.[ Y ,'~ g .W .NM 22 jintofmuchJmoreireadily;. There's.'ainew software being put out' i .y t hs 3' and,some'of usIhave not;been_able to really use;it as easily a ? . woi11d ' beldesirabl'e,. so: INPOLhas' reconfigure;thelthing' 3 - f51 and/we're testing itinow in' terms.of access. It's much more J .i 6L user-friendlytand'we;think.thatL willibe theinext stimulous. y 7' that.will c'ause an.L mprovement insinput.t :It's being able to -j i 'r ^8-get somethkngito. help..you'in.the plant. b 9: Softhere:is'enough. data in the files that utilities 10' .can benefit from.if they can get:into it readily andLhaving ~ 4 11 .the.-soft reference from,;you know, if I've got this' breaker 12: and these other-6' plants-have this breaker, I'm going to. call 'l 13-
them and find out what's happening with them.
I see a couple .v[ s 14: of 'f ailures here'. 15 I.believe the greatest benefit now is going to l 1 a 16; be in this utilization. There's a reasonable data bank i 17 there. There are areas in the' engineering.. reports and there 1 18 .are some things we're working with INPO to improve the quality 19 of that data, but there's certainly enough data compiled 20 already that the utilities would be interested in and would g q ~ et 21-substantial benefit out of it. 22 MR. EBERSOLE: Are you picking up the cases.where p St 'l 23 diversity saved the day for, for instance,-B&W and CE, where t 24 itLdoesn't exist at Westinghouse? I 25
MR. MICHELSON:
These are not event reports. L =MR. EBERSOLE: I know. I'm talking about NPxDS. L Acme Reporting Company l m....... =
%1, L 158 -q 1 MR. MICHELSON: Well NPRDS isn't-an event report. ,,,() l2 MR. EBERSOLE: I know that. I'm trying to pick up 31 the component finger data. ' (,! _ MR. EBERSOLE: You don't have it. You just have 4 individual data. You don't even know the challenge rates or 5 g anything like that. DR. LEWIS: That raises a question I remember back 7 8 from! scrap ferrier days and one of the rules on an LER now, not 9 MPRDS, if you have a system with dual trains and one fails, where you don't lose the function, is that an LER. 10 MR. JORDAN: No. 31 DR. LEWIS: It is not an LER. 12 /O MR. JORDAN: No, sir. ~ 13 \\,) DR. LEWIS Is there a monitoring system that will 14 tell you single component failures when you have a paired 15 system and the function is not lost? 16 MR. JORDAN: MPRDS should collect that component 17
failure, gg D R.: LEWIS:
Because somebody is monitoring the gg f ilure of the component which is backed up by another, that 20 21 is, are they separate component? 22 MR. JORDAN: Regardless of whether they are separate ~s %j 23 or not, the component failure is to be reported to MPRDS. r~s ~4 DR. LEWIS: Now do you find out that the component G 25 failed? Acme Reporting Company (202) 62B 4880
j/ 6.. o 't ,,s l 100 L i ~ 1 14R. ' NOVAK: Through the maintenance process. n /~N i (,)- 2 DR. LEWIS: Only through maintenance--only during i I 3 maintenance? ,m 1 ' _/ 4 MR. JORDAN: Surveillance, maintenance, whatever. ( 5 DR. LENIS: Not at the time that it failed? i 6 MR. JORDAN: Correct. 7 DR. LEWIS: Is that correct? 8 MR. DENNING: That's the basic way the system 9 works in-practice is that they're driven by the maintenance 10 work order system. 11 DR. LEWIS: So they have to do a separate testing 12 of the, I hate to use the werd, redundant because it's the ('~)T - 13 wrong word, but redundant components in order to determine w. 14 that one of them isn't working. 15 If it's trash and failure, then you wouldn't find it ? 16 MR. NOVAK: We would never know. 17. DR. LEWIS: I am very sensitive to this because if 18 have an airplane which has two fuel pumps for safety and ' 19 they are always turned on at the same time and the pressure 20 output is what is measured so one of them could be out of ] 21 service or operating intermittently for years and no one would 22 ever know. g !' J s .~s The first thing I did when I got the airplane was to i 23 J (; 24 make them separately test. To what extent, is this a general - v' 25 question, a general issue and what you're telling me is that 1 A, cme Reporting Company \\ ,e n sa...... 1
(? T 161 1 you won't pick up single failure in a redundant system except f .2 in' maintenance testing, so if it's intermittent, you may, in 3 fact, have it out of service for years. 4 MR. EBERSOLE: You could, with appropriate 5 instrumentation, but you just-- 6 MR. JORDAN: This is only a reporting system. 7 The LER rule would not have taken care of that either, because 3 you had to know that'the training was not operable in the 9 old rules to report it. DR. LEWIS: In my case, the train is operable. 10 11 I'm sorry, the function is there, but one train is inoperable. 12 MR. JORDAN: Right. When you know that one train lllh 13 is inoperable, under the old LER rule, you would report it. DR. LEWIS: Sure, I understand that. The question 14 15 is how do you know, and there is no requirement that one know. 16 MR. JORDAN: Right. DR.~ LEWIS: Even when the redundancy is extremely 17 33 important in insuring the safety of the plant, you're still not monitoring the effectiveness of the-- 39 MR. JORDAN: I have to say that there are sur-20 veillance tests and tech specs that test train by train. 21 DR. LEWIS: I understand. ~ 22 MR. JORDAN: And so that is normally the mechanism 23 by which we find that you have had a failure otherwise you (~ 24 ? l v w uld find the failure in the service and those train tests 25 Acme Reporting Company (20/p 626 4886 -nu-----.
.j .oy fa, 162-1 ' reasonably good'in determining whether or not t' hat train can-- s J' 2= DR. LEWIS: I have no doubt of that. MR. WILLIAM: The particular question that was 3 .,u -( ) to put to.us on BWR that Bob is answering'is, can you use 4 that data-base to come up'with reliability to validate that . r, 6 surveillance report. DR. LEWIS: There's a dicotomy here.
7, MR. WILLIAM

Right. 8 '9 DR. LEWIS: That's a good question. 10 MR. WILLIAM: Sure is. It's all wrapped up 11 .together and it depends on the reporting why it's, you know, 12 'whatLthey put into it, the information and time of the (m) 13 failure. '14 MR. MICHELSON: Well, they don't tell you how many 15 challenges there are per year or whatever either and you 16 can't calculate a failure rate. 17 MR. NOVAK: You can make an assumption. l 18 MR. MICHELSON: You have to make a guesstimate. 19 DR. LEWIS: So the rate question'can be improved. 20 MR. NOVAK: They are also looking for ways to-- 21 they had a pilot program on a number of safety systems 22 in which we and I think 10 or 12 utilities participated 7s. 1 %) where they chose a specific system and asked for very detailed. 23 information about the components of those systems and then 4 _fT 24 .L] 25 they tried to correlate how best they could correlate a lot Acme Reporting Company (2OJi 628 4888
gl'i ; ~ :, iv ' d ..4'. A.7 q 'f m qy- .,yg, );]ig n -l 3[ ' -163 sg..'^
1

ofL.infbrmationion'.just component' failures and'they'were.able
~
I L.
to come..lup. wit $ a~ fairly good model.so that'--a.nd it's a:wayfto' l 2 c, .+. 2 encourag'e MPRDS's,. reporting. 3 Q.n 4 .The ideacis=to use.that as-some performance indica-1 /,', 'L5- -tor:on,jI. guess,..onisomexavailability time,'as I understood.it. .g' A yway,.its.Ljust one more way of encouraging. utility J ' part:*.cipation. I think the;bo'ttom'line that I see'is,'again, y 8 ihbreEwelseeia?wideEspectrum.of utility. participation. ;It - g isn't necessarily eveyb'ody.l submitting two-thirds. ] Yo.u may h' ave somebody that's doing'it all and;you-10- ~ .it- 'mayIhave somebody<that's just not necessarily even.pssing J 1 mustardi I1think that's the other side of the coin. 12 I
if We?seef a wide spectrtnn of i participation, wider'.than g
[ l34 .the1 widest. 'Our desire 'is to bring =up the bottom; third of',
the-utilities and-we can so,.do better,'get that information j
~ ~ j.. '15. r6 - i
16 ;
in here and then everybody will begin to do it? jf MR'. WARD: Why?! . s MR. NOVAK': Why? y MR. WARD: When you're looking for statistics on .ig '20 equivalent performance, does it really matter whether--do 21 non r.jorting utilities have a different type of equipment? ] s> [ 22 MR. NOVAK: No. y 4 ~. d 23 MR. WARD: You think--you suspect they might have a i 24 'different typa of performance of equipment? '25: MR. NOVAK: I think their priorities are such that ] Acme Repo' ting Company r (20 0 628Af68 u 3;.. . +
~ 1 gD 164 1 they-just don't put the resources into MPRDS reporting. ~ s J 2 MR. JORDAN: MPRDS isn't an end in itself. I 1 3 MR. NOVAK: That's ; hat he's saying. l l 4i s 4 MR. MICHELSON: But you have got to know the total 5 population out there. 6 MR. WARD: Well,.how big a sample do you need t o 7 know? 8 MR. MICHELSON: No, you don't need a big sample 9 necessarily, but you got to-- 10 MR. WARD: You already got two-thirds. 11 MR. MICHELSON: --know though that if you see 12 50 failures that's 50 out of how many components--how many 13 utilities are reporting that component? A 100 utilities or lllh 14 So utilities. Is 50 reporting, you see 50 failures, then you know it's an average of one per utility or whatever, 15 16 but I don't know how many utilities are reporting that I don't know that they're not all reporting 17 component. 18 the same components. 19 MR. ROSENTHAL: If you have commonly used components 20 and a lot of then ar,d there's a failure, but the-- On the other hand, we are concerned with rare 21 events and let let's say an off speed water or electric 22 o s-) motor versus the explosive squint valve on the boilers 23 ladder stand by liquid control where even one failure of that 24 ("); i 25 velve--squint valve, would be very significant. Acme Reporting Company (2021 628 4868
d; y-* I hsM;- v wg ,7 .g N l' ~ 4 -6 165-tg % 'l s (;I..-
p. [
-s y W,..g n / $. k s Even..if you'didn't detectlit.\\ 0 ' e,, 3
1
' DR.::: LEWIS : / _,'l i2. .ROSENTHAL:' That's 'a point, 'thatiyou L perceive MR. a - -- "s, Ac w i 3; /dfal' teas:very reliablegwith very veryllittle'.datufon it"and~ 'soIthat oneor~twoirepor'sLmight change ycur perception =of M t e .n 51 .that? system,bu't~onceLyou'haveit-- o: -,4, .'6 - MR. WARD: - I'll appealfto our statistician-for an'
c a
/7: Authoritative., opinion. 'l ~ ~ . DR'.. LEWIS : -Appeal granted. g -9 -MR.' WARD: ButLI-:doubtLthat having 99 and a' half all Levents reported versus 67: percent will give .to; percent'ofc I! J l 1'1'. L.you1 statistically a'.very.different confidence level that 12 iyou!re going to"have. fi 5 DR. LEWIS: Let's'just say beyond reproach. 733 c' MR. JORDAN: That's my say. I; don't know what the ~ 143 l l .s .[151 optimum participation is, but.the' engineering data is our o ,f ig. biggest thrust right.now. p. N k 17 We know,.for instance, that you have the right .18' -set of breakers and a11 of them--and this is not 50 percent, </ ' S it's a much lower figure than that, so our focus is there 19 and'we believe that the system is really going to be worth 20 its salt and some.of these are going to--now that that they're i. 21 m r, [22 . going to have better access to be able to use it to a great h - 23 advantage, reliability and financially and all, so I really L 2d. aee us on the edge of substantial improvement. .%J; DR. LEWIS:.This problem of trying to collact all
25 s
Acme Reporting Company-aan su n.. [ _ = _; ., M j-
3 3 373-., .T g b 4 % 4,. y1 4m 9, 3 1 1, 1
y ac
,;g u ~ ,s il66 ,1 - 1 4 g,, s . 7.;, ~l l l" h, h; Y l1 i the ; data.is 'not!L uniqueltocyou.- q c sq.
1
] bd(j; /
2*
lMR.' JORDAN:: iNo. -In this case,~.we have.-had;sub - , ; ~ -. g ; ,h., .13 i stiantia'le d'iscussionsiwith INPO ' about our examinations - of' h M [4)
completeness and'.' timeliness.-Land.I'think our examinations.
,Ra 1 'S1 .for' quality and parti'cipation are going.,to.be:differe'ng in" ji ~ i c.,.. 1 6
the1 future, but,we're; going to stop__ harping lon'the time
.i i, andatheLnumbers and look more now',at the-quality engineering.-
7 "8
.' data.- k V* y, V: 1 9 'DR',. LEWIS: I'll wait'till I see:it. s 10 MR. JORDAN:. You'h' ave my! word on it. j -11. .DR.; LEWIS: .I know. l12 MR..JCRDAN:.LikelFrank Perdue. B 113-MR. : N OVAK : '.Unlesslthere are any more' questions on. ~ (
14 MPRDS, I.would movefon'.,
} 15 DR. LEWIS: What's the time now. . 16 '. iMR.LNOVAK: Ne're:right.on schedule. j 17, MR. EBERSOLE: Are you going t'o talk about the ,'w ' l'8 ; quality of LERs?. .c 19 MR. NOVAK: Yes. What is meant there really, was. 20_ when you -read the description,- our test is would a knowledgeablee 21 Person.be able to understand the cause'of the failure.and so 22 ~forth. .You're referring to the applicable quality of NPRDS, j 23-right? r g 24 - MR. EPERSOLE: I didn't know if you were done with f 25 Tthe slide. g1 Acme Reporting. Company a n, n.....
x;;; :. . n 7._ e f cyjk Nh, ' ' T ( L. x h /w y ; 4 .[ ) -f (
167 s
w ..t >, ,_ g r i R , ;r ' .1
MR.)NOVAK

.Okay,?I ' m., sorry..I'm notJdone:with'.Lthe y
..7 {c 7W. ~ dj.. 7 l2L Talides.sVITwas' going to-move to,abnormalloperating; occurrences ? \\ , ':i (butLsince I'talkedfa-'.little bit about quality? NPRD,mwelalso-6, ' " ' L3 a- '4 Ldo.look"atywhatlis theJg.uality,of.~the de'scription of-the v 3 !ca'use of(the event'and so forth.. l, r,l We. don't need to spendi l too..much; time on AO reports.'
6.
7. .We'havenbeen reporting AO'.s as:part of,the law. What we're' seeing'is'not that many'in the reactor end'of it. We'have 8 7 jg' seen anlincreaseiin medical mis administration andiso'forth. I. don't think;there's any real messages or inform'ation that 10 .'we can discuss in--terms of a normal occurrence.
17.;-
We havefa desire to improve the timeliness for you, 12 g; 13 - but-- 'gj DR. LEWIS: Why'do you concern yourselves with-medical misuse of radio active materials. :It's.true it's 15 an issue.with a problem-in hospitals. It Isn't, going to kill .16, '30,000 people on a given day.. 17 '18 MR. NOVAK: Kathleen do you'want~to try and: answer' ~ that. gg '20 - MS.. BLACK: Is the question why are there abnormal Dj 21 l ccurrences or-- DR. LEWIS: No, no, no. Why does NRC concern itself
O

i 1s. 92 ~ .23 with medical misuse of radio active materials? I know it's F.1 : part of the' law. y-? 2 A , 25 ; MS. BLACK: There was an event a long time ago l i Acme Reporting Company ) 1 c a n, u...... )
m y r;. 3, p
[$MM)F
j s
lm jy4; y ' ~ Q@' k A 11684 'l + y s c;' + ',. g ee q pg', ' n.!- J j and(:somshori~orJother ~ fitigot its regulation ~so; this mis-I m ,a i k.. ,e ,e. ? + ...e..j 2 cadm' ministration.had to..be reported. l f 4 n._ 4 ,1 r . DR.-LEWIS: I?see'.. J3;: M MS BLOCKi "And we found the areas.that'there were-q 5 4 per thousand:adniinistrations.- We recommendedLthat--I must'. -G tellhyou that!there;were another set.:of administrations' ~ s y 1., 7 'thatfjustTreferred;t'o:New York State. e y -i_. The; physicist who does all"the calculation is not l'
8
{ W licensed._ It'sinot certified, he's not. supervised by any 9' U, m 3 - 110 Lgovernment group.- ~ 11-X-ray 1 technologists, to some extent-are ir virtually 12' every state; because 'it's.an old ' profession' of the. commission. .13; MR..' WARD : ;Not'the oldest. 14. - MS. BLOCK 5 No, no. 1
/
~ 15. The' convention--we had recommended and this isLone H LIG of the places.whereLAE was disagreeable and we recommended q 17 that they,have:a quality assurance. They require licensing. y 18 for quality assurance programs for all rates and levels. 19 Through a bond hearing,'the commission' directed them q 1 20' to institute quality assurance and that effectively stemmed s ,d '21-from a very severe event in which people were killed at a
)
22-hospital in'the Middle West in 1974.
Oc

23 The therapy listed-in the administration had y
24 -apparentlyLshorted the lives of patients who had cancer, who 1 "5 could have-bee'n cured, whether that is true or conjecture Acme-Reporting Company , b. ' . 2o2: eae 4sse ( s' s ? [b _L L :-._
l f (vl ),; [ 169 'h g 1 is questionable and these are not hypothetical, 30,000 people h' (,) 2 who could be exposed in the vent of a melt down or whatever, 3 these are real people who are getting exposed and lives may es ' () 4 'be shortened. 5: DR.- LEWIS: I understand all of that, but there are 6 problems in the world with seat belts, there are problems l ' n the world with all sorts of other things, it seems to me i 7 8 'that'this,should be.beneath the NRC's contemot. 9 MR. MICHELSON: Theipotential-- l 10 DR. LEWIS: Not that it doesn't kill real people, 11 I understand that, but so cha plenty of things--never mind, '12 .let's move on. 4 /N 13 MR. MICHELSON: Maybe a better feel for it might. , \\_J g4 be, what kind of dose are we talking about for the public fror l 15 medical mis administrations collectively versus operating the 16 power plants collectively. 17 MS, BLACK: Admittedly, all the medical doctors 18 are going to argue that the radiation that you get from those 19 diagnostic administrations, not medical administrations, 20 diagnostic administrations, are not bad because they are 21 . preventing health problems be resolved, but it's up into the 22 millions of-- ,sC MR. MICHELSON: The potential for screwing up is 23 l i there. <] 24 \\_.)- DR. LEWIS: I know that, but that's true in auto-25 1 Acme Reporting Company (202: 628 48b8
-= 170 1 mobile' driver's and-- f 2 MR. JORDAN: You divide.that by 10 to the 4th-- 3 MR. MICHELSON: I just asked a simple question, how i: does that' compare with nuclear power plant operations in { 4 ) '5 terms of-- ) g DR. LEWIS: With normal operation--there wouldn't 7 be an NRC if it were just a matter of normal operation. 8 MR. MICHELSON: I think this is probably bigger i -9 doses to the public than nuclear pwoer plant operations, but { 10 it's in small quantities. DR. LEWIS: Medical administration of x-rays is gi 12 something like what, 20 milirems per person per year, somethin7 . {') like that. 13 +.s MR. MICHELSON: We 're talking about mis administra-14 15 tion, mal operation of equipment and whether or not-- DR. LEWIS: I carried a radiation monitor on my 16 17 airplane last night and measured what I got flying here, j l Believe me I got'more than I get from medical administration 18 doing that. We don't regulate that either. 19 20 You know, there are some things-- MR. MICHELSON: That's by your choice. ,1 'C R. LEWIS: No. I had to come here. My choice was 22 to stay in Santa Barbara where I live. Let's move on. 23 MR. NOVAK: We're going to move into one of the 24 V t final points, LER quality and Mark will spend a few minutes 25 Acme Reporting Company { 1202n 628 4688 I w___________
T ~ 1 h 1' L 171 l 1 briefly summarizing what we do in terms of reviewing the quality I 2 of LER's. 1 I 3 MR. WILLIAM: We have had a program in place for 4 a couple of years. We're the only place in the agency where 5 we systematically looked at the quality of the reports that th e 6 utilities were sending into us and for two years we had 7 a contractor in Idaho comeup with a bench mark analysis where j 8 they looked at the quality that should be had in each iten 9 that is required by the rule in that, did the licensee j 10 described in the past achieve shut down adequately? Could you l l l 11 really understand the root cause of the events? t 12 All of the requirements that were to be reported llll 13 and then the program was set up to feed back the results y of all that to the regions and the licensees and the utilities 15 and very briefly, I think we've seen an overall improvement 16 in the quality of the LERs from many many licensees who are g7 in the second round of these things and we're bringing the 13 program to a conclusion right now and-- 10 MR. MICHELSON: This is an actual grading process l l 20 you're going through, as near as I can see. Is there a documertt 21 that would describe this grading process? 22 MR. WILLIAM: I brought an example. It's in that ] ./ 23 stack right there. I can leave it with you. 3 34 MR. NOVAK: I think you're asking two different J 25 questions. You wanted a document that-- Acme Reporting Company (.2 D 2 ) 6 2 41 A 8 8 8
-= - e t i '7'-} ? -{ l g g g ;j. 's e\\ 3 A; m. .s. 3 'c 172 v. Y. '~ '4 j A k
1 IMR.;MICHELSON

- Have;you' brought a reference?'
]
2 MR.;NOVAK

'- You' wanted a' document that--
t' ,c[ (3, A. MR.;MICHELSON: A: document that described.the. ^ i %Q 4; process,-.how you go about-developing.a grade. It'ssa new q
o re'gfor

something.
m I6 0 .MR.; NOVAK :- I. don't think we have got'itl formalized i >/ + 7 .intoya.new. regulation'.- DL 8 'MR.'MICHELSON:.But it is something-- e, 9
MR. NOVAK:- We-have a 64cument in that stack.-

10 MR.~MICHELSON:- I would.like to get a copy of it,'
11 - ifJI may because'I' notice that.you'rel tending them out to
12
. alllthe regions. They're coming up for their shots now and ,p. fb) it looked.' kind;4of-interesting'and'I wondered.how you did.it.: L 13 9 .14 (. Ih's' not 'quite as obvious from :looking at the answers, so a 15 ' t d sp ald. J,. 16 ' MR.. WILLIAM: What we did was that we encouraged <t7 and.I think eventually we got-almbst everybody to send it 11'8-right'back to the people writing the LERs,-the aanager of' 19 - compliance or-whoever that is and through two years of that .go . process,,we can see the same utilities >have come up in just about all.the areas that are graded by ETMP, so we have seen 21 ,,7 22 :. real progress in that area. ~ ? 23; .Overall, I think we have seen an improvement and I b 7 .24 - ' think they-have got the most cost effective amount out of s :D .25: it-right now.- s q; Acme Reporting Company iaoar sae asse ct3
u ;., u ~ = =--- - - - - - - - l{y}0 E W.e
!h y, Y',
if ? jl,, jW -{p y I173J ps, s,s., ,b 3 f 7, "MR. 'MICHELSON:: ;One;of the things.I notice'fromL W r~.a. ' 9. l' /j' N 32" ' reading, allL the,LERs and If try to read.- them_ all, ; though I: g u.4 C '..n,
3;
.~cettainly skim'a lot andothatlis:that'there seems.to a~ dis-
4 Proportionate'.' number:of rathe/. useless: LBRs yet, e
5-
,c "Do: Vou1 have some kind of a.nrociram to sit down and-
~ - 7 g. r 4, s take?aE ake a.hard look-at whether we need'to.cet certai'n" hind
6' t
s-7 Lof' reports.'any.more because--there's some patterns to some-E8 'of'these that, you'know,.the tech' spec kind, if you'se going. j 1 g 9l Jtotdo a' tech. spec surveillance or youlgotLyourself a tech spec < --e i10 - nproblem,fyou.have.to report it,:but do.we reallyfneedLto see 11) those 'LERs in: the future? ?l2 They aren't helping you technicall, I; don't believe. l13 ; They; don't-help me at all.- Itzwould'evsn be nice if they-9 .g4 . kind-of gotl thrown away;before ILgot them. ' 15 MR'. WILLIAM: I-think a lot of those are being: g ~ 16 :. thrown away,~especially the'y're being removed in the area-of .17? fire. protection'and fire watches and th'e like and Ejeneric~ Ui I 18 ' Lac,tivities. .i ,19: MR. MICHELSON: Fire watch is another one where-20 the' guy--well, if he's falling asleep'that might be useful dat a, but-for the established fire watch an hour and a nalf after 21 n they knew they had a problem instead of an hour, that sort-of 22 -o' 23 thing,.that may be useful to record, but the LER system,'I 24; ' thought'was intended to be a higher level reporting system. 'O You've got a pretty large group of that sort of ( 25 l 9 'information which may be you ought to get out of the system Acme Reporting Company aan sae-seee f
c 174 1 with another rule change or something. 2 MR. JORDAN: We're looking at that and we have 3 clean up system isolation and control room ventilation system 4 actuation and those inds of things don't have any direct 5 connection with safety, there is a fairly large body, I don't 6 know what the number is, maybe Jack could hazard a guess, 7 but there are unnecessary reports in that group. 8 MR. MICHELSON: Unless they trigger something else 9 besides their own happenings, they're of minimal value. 10 MR. JORDAN: That's correct. 11 MR. MICHELSON: And if they're of minimal value 12 to you, what are they in the system for because I think the llh 13 system is for you. 14 MR. JORDAN: Right. We came up with about 900 15 that we would look at again out of the 3500 or so, but that was 16 just at first glance. 17 MR. MICHELSON: You mean 900 out of 3,000 or so a 18 year are reasonably useful to you, is that what you're say-19 ing? 20 MR. WILLIAM: No. They were in the class that we 21 reviewed and-- 22 MR. MICHELSON: That you want to get rid of? -) MR. WILLIAM: Yes. 23 MR. MICHELSON: Only less than a third then are E' 24 Y) 25 useless to you? Acme Reporting Company -, u. n o. l l
175 1; .MR. LEWIS: that's what he just said. 2 MR. MICliELSON: I would have said two-thirds my-1 3: self from what I have read. MR. NOVAK: And I think their future yields, you 4 5 know, there's data that is now.in the data system that may, 6 in fact, reinforce-itself at a later date. Okay, I think we have covered the questions that 7 we could, so we would turn the meeting back over to you. 3-9' MR. JORDAN: I would like to make one comment. Somewhere along the way, a question was asked 10 about whether we would--wheEher action we had taken had 11_ resulted in improvements and we said we couldn't think of 12 13 anything off hand. I have thought of several off hand. h DR. LEWIS: I see. 14 MR. JORDAN: So I thought we would sort of crank 15 .them back in. The Salem breakers problem is one that certainly is we have looked at, after the fact, and have seen now a sub-17 stantial' reduction with those items, 18' There were failures and now the failures have gone ig, and we have put an additional requirement on licensees through 20 a bulletin for further reporting. 21 The 'BWRBRD discharge is another one. There were o.n reported failures and those have gone away. 23 MR. MICHELSON: Those seem to be-- 24 MR. JORDAN: Clams. 25 Acme Repor..g Company 12026 628 48Se
176 1 MR. MICHELSON: Those haven't gone away. J^ q ( 2 MR. JORDAN: They have declined substantially. i e 1 3 MR. MICHELSON: I beg to differ. I have begun to se = ~j 4 some very disturbing LERs on Asiaric clams showing up and 0 5 I thought that problem had really been put to bed. 6 Some day we need a report on the bio fouling ) 7 situation. 8 DR. LEWIS: Do it in your subcommittee. 9 MR. JORDAN: We've got an ongoing case study on 10 that. 1 11 And then maybe more importantly, the review of 12 performance indicators INPO showed a substantial measureable h 13 real reductions in safety system actuations. An increase 14 in the previous plant availability over the last 2 or 3 15 years, so there are really trends--there really are reactions t 16 to industry and NRC pressure in follow up on particular issues 17 in the works. 18 DR. LEWIS: And they also are presumably hardly 19 due to just generally accumulated experience and familiarity 20 as the industry matures? That's bound to happen to some 21 extent. Whenever you use something for another year, 22 you know more about it than you did the year before, j 7--- ' ) 23 MR. JORDAN: Right. The concern, of course, was tha t f 24 every individual plant had learned and solved it's problems-- 4 / 23 DR. LEWIS: I understand. V \\ Acme Reporting Company l 12028 629 4H88
__y 3 f.' x .) JQ-F e n, 177 o W [' O - :1 .,i MR pTORDAN : : --and made gains, then the'ratefis very r.
5. /
/: t ~ ,,,h.-Qv 2; slow,1butlyou'can somehow select-that' experience. A L3 ' -DR. LEWIS: I~' understand. : g~ t l l f4 y gg,. EBERSOLE: ' A couple of-things. You mentioned' '*t, i 5 thedreaker problemsThave gone away, but you* notice that 'O we have got'a new: set of. problems in the area--one'is'from-l -) 7 equoia,. malpractice'in the'wel' ding of the shaft in the. ! 7. S !8-pump. .I really:: don't know quite.how to look'at that-yet, t 9 butfit-seems-that most of the failures occur'in the make up f. j b,"
>10 modeirather than the' breaker mode.
l11 'I don't know'whether you can part these out that'- i y 12 way;or not. [ 13 The full committee took a great deal of. interest.in the lastufew months in-all theNe funnyLthingsJhappening.with '141 '15 Lthe release of boron seeping out through gaskets and other- . cracks'on the.-syl stems an'd' leading-to a variety ofLdegrading; 10
17 effects.on the~ carbon steel and'the.more spectacular.one'was l'8 the recent~one St. Onifre where, within one cycle, it. corroded
.' 19 the bolts, you know, andiit spewed out in a leak.. 4 . 2'O Anyway, 'I was just going tio ask you if' you have a-- P '1 that set ~of events, the potential ill effects including the y 2 p.- 1 L i. 22' worse sort of--
gDq
.c 23. MR. ROSENTHAL: We'have an ongoing study on the-- { . 24. MR. EBERSOLE: I'm thinking about the one, for s 1 - 25: instance, on the bolts that secure the manifolds on the primary Acme Reporting Company g o [ [..' 12021 628 4888 I' l ,p a \\, j s _ l!/
i i 178 1 loops on this--this is a real LOCA if we're ever going to 2 have a LOCA, that's probably where it will come. 3 MR. ROSENTHAL: WE actually are doing some work in / 4 that area. We're not doing any work on the wire, the breaker 5 problem. and--that's it. 6 MR. EBERSOLE: That's all. 7 DR. LEWIS: I do encourage private conversation. 8 Thank you gentleman. 9 (Whereupon at 5:40 p.m., the hearing in the 10 above-entitled matter was concluded. 11 12 lllh 13 14 15 16 17 18 19 20 21 22 --.i 23 '3 24 ) 25 Acme Reporting Company 4202a 628 4888
1 CERTIFICATE I 2 3 This is to' certify that the attached proceedings before the 4 United States Nuclear Regulatory Commission in the matter of: 5 'Name ACRS SUBCOMMITTEE ON SYSTEMATIC ASSISSMENT OF EXPERICNCE MEETING - THE ROLE O AEOD 6 7 Docket Numbers 8 Places WASHINGTON DC 9 Date: NOVEMBER 3, 1987 10 were held as herein appears, and that this is the original 11 transcript thereof for the file of the United States Nuclear l 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 rec rd of the foregoing proceedings. e 16 /S/ Utt N 1 d44 17 (Signature typed): IRWIN L'. COFFE RRY 18 Official Reporter 19 Heritage Reporting Corporation 20 21. 22 23 24 25 4 Heritage Reporting Corporation (202) 628-4888 __}}

ML20236K162

Text

=

Navigation menu

Search